Initial import

505 files changed, 115996 insertions, 0 deletions

diff --git a/tools/perf/.gitignore b/tools/perf/.gitignore
new file mode 100644
index 000000000..812f90419
--- /dev/null
+++ b/tools/perf/.gitignore
@@ -0,0 +1,30 @@
+PERF-CFLAGS
+PERF-GUI-VARS
+PERF-VERSION-FILE
+FEATURE-DUMP
+perf
+perf-read-vdso32
+perf-read-vdsox32
+perf-help
+perf-record
+perf-report
+perf-stat
+perf-top
+perf*.1
+perf*.xml
+perf*.html
+common-cmds.h
+perf.data
+perf.data.old
+output.svg
+perf-archive
+perf-with-kcore
+tags
+TAGS
+cscope*
+config.mak
+config.mak.autogen
+*-bison.*
+*-flex.*
+*.pyc
+*.pyo
diff --git a/tools/perf/Build b/tools/perf/Build
new file mode 100644
index 000000000..b77370ef7
--- /dev/null
+++ b/tools/perf/Build
@@ -0,0 +1,44 @@
+perf-y += builtin-bench.o
+perf-y += builtin-annotate.o
+perf-y += builtin-diff.o
+perf-y += builtin-evlist.o
+perf-y += builtin-help.o
+perf-y += builtin-sched.o
+perf-y += builtin-buildid-list.o
+perf-y += builtin-buildid-cache.o
+perf-y += builtin-list.o
+perf-y += builtin-record.o
+perf-y += builtin-report.o
+perf-y += builtin-stat.o
+perf-y += builtin-timechart.o
+perf-y += builtin-top.o
+perf-y += builtin-script.o
+perf-y += builtin-kmem.o
+perf-y += builtin-lock.o
+perf-y += builtin-kvm.o
+perf-y += builtin-inject.o
+perf-y += builtin-mem.o
+perf-y += builtin-data.o
+
+perf-$(CONFIG_AUDIT) += builtin-trace.o
+perf-$(CONFIG_LIBELF) += builtin-probe.o
+
+perf-y += bench/
+perf-y += tests/
+
+perf-y += perf.o
+
+paths += -DPERF_HTML_PATH="BUILD_STR($(htmldir_SQ))"
+paths += -DPERF_INFO_PATH="BUILD_STR($(infodir_SQ))"
+paths += -DPERF_MAN_PATH="BUILD_STR($(mandir_SQ))"
+
+CFLAGS_builtin-help.o      += $(paths)
+CFLAGS_builtin-timechart.o += $(paths)
+CFLAGS_perf.o              += -DPERF_HTML_PATH="BUILD_STR($(htmldir_SQ))" -include $(OUTPUT)PERF-VERSION-FILE
+
+libperf-y += util/
+libperf-y += arch/
+libperf-y += ui/
+libperf-y += scripts/
+
+gtk-y += ui/gtk/
diff --git a/tools/perf/CREDITS b/tools/perf/CREDITS
new file mode 100644
index 000000000..c2ddcb3ac
--- /dev/null
+++ b/tools/perf/CREDITS
@@ -0,0 +1,30 @@
+Most of the infrastructure that 'perf' uses here has been reused
+from the Git project, as of version:
+
+    66996ec: Sync with 1.6.2.4
+
+Here is an (incomplete!) list of main contributors to those files
+in util/* and elsewhere:
+
+ Alex Riesen
+ Christian Couder
+ Dmitry Potapov
+ Jeff King
+ Johannes Schindelin
+ Johannes Sixt
+ Junio C Hamano
+ Linus Torvalds
+ Matthias Kestenholz
+ Michal Ostrowski
+ Miklos Vajna
+ Petr Baudis
+ Pierre Habouzit
+ René Scharfe
+ Samuel Tardieu
+ Shawn O. Pearce
+ Steffen Prohaska
+ Steve Haslam
+
+Thanks guys!
+
+The full history of the files can be found in the upstream Git commits.
diff --git a/tools/perf/Documentation/Build.txt b/tools/perf/Documentation/Build.txt
new file mode 100644
index 000000000..f6fc6507b
--- /dev/null
+++ b/tools/perf/Documentation/Build.txt
@@ -0,0 +1,49 @@
+
+1) perf build
+=============
+The perf build process consists of several separated building blocks,
+which are linked together to form the perf binary:
+  - libperf library (static)
+  - perf builtin commands
+  - traceevent library (static)
+  - GTK ui library
+
+Several makefiles govern the perf build:
+
+  - Makefile
+    top level Makefile working as a wrapper that calls the main
+    Makefile.perf with a -j option to do parallel builds.
+
+  - Makefile.perf
+    main makefile that triggers build of all perf objects including
+    installation and documentation processing.
+
+  - tools/build/Makefile.build
+    main makefile of the build framework
+
+  - tools/build/Build.include
+    build framework generic definitions
+
+  - Build makefiles
+    makefiles that defines build objects
+
+Please refer to tools/build/Documentation/Build.txt for more
+information about build framework.
+
+
+2) perf build
+=============
+The Makefile.perf triggers the build framework for build objects:
+   perf, libperf, gtk
+
+resulting in following objects:
+  $ ls  *-in.o
+  gtk-in.o  libperf-in.o  perf-in.o
+
+Those objects are then used in final linking:
+  libperf-gtk.so <- gtk-in.o  libperf-in.o
+  perf           <- perf-in.o libperf-in.o
+
+
+NOTE this description is omitting other libraries involved, only
+     focusing on build framework outcomes
diff --git a/tools/perf/Documentation/Makefile b/tools/perf/Documentation/Makefile
new file mode 100644
index 000000000..3ba1c0b09
--- /dev/null
+++ b/tools/perf/Documentation/Makefile
@@ -0,0 +1,343 @@
+include ../../scripts/Makefile.include
+include ../config/utilities.mak
+
+MAN1_TXT= \
+	$(filter-out $(addsuffix .txt, $(ARTICLES) $(SP_ARTICLES)), \
+		$(wildcard perf-*.txt)) \
+	perf.txt
+MAN5_TXT=
+MAN7_TXT=
+
+MAN_TXT = $(MAN1_TXT) $(MAN5_TXT) $(MAN7_TXT)
+_MAN_XML=$(patsubst %.txt,%.xml,$(MAN_TXT))
+_MAN_HTML=$(patsubst %.txt,%.html,$(MAN_TXT))
+
+MAN_XML=$(addprefix $(OUTPUT),$(_MAN_XML))
+MAN_HTML=$(addprefix $(OUTPUT),$(_MAN_HTML))
+
+ARTICLES =
+# with their own formatting rules.
+SP_ARTICLES =
+API_DOCS = $(patsubst %.txt,%,$(filter-out technical/api-index-skel.txt technical/api-index.txt, $(wildcard technical/api-*.txt)))
+SP_ARTICLES += $(API_DOCS)
+SP_ARTICLES += technical/api-index
+
+_DOC_HTML = $(_MAN_HTML)
+_DOC_HTML+=$(patsubst %,%.html,$(ARTICLES) $(SP_ARTICLES))
+DOC_HTML=$(addprefix $(OUTPUT),$(_DOC_HTML))
+
+_DOC_MAN1=$(patsubst %.txt,%.1,$(MAN1_TXT))
+_DOC_MAN5=$(patsubst %.txt,%.5,$(MAN5_TXT))
+_DOC_MAN7=$(patsubst %.txt,%.7,$(MAN7_TXT))
+
+DOC_MAN1=$(addprefix $(OUTPUT),$(_DOC_MAN1))
+DOC_MAN5=$(addprefix $(OUTPUT),$(_DOC_MAN5))
+DOC_MAN7=$(addprefix $(OUTPUT),$(_DOC_MAN7))
+
+# Make the path relative to DESTDIR, not prefix
+ifndef DESTDIR
+prefix?=$(HOME)
+endif
+bindir?=$(prefix)/bin
+htmldir?=$(prefix)/share/doc/perf-doc
+pdfdir?=$(prefix)/share/doc/perf-doc
+mandir?=$(prefix)/share/man
+man1dir=$(mandir)/man1
+man5dir=$(mandir)/man5
+man7dir=$(mandir)/man7
+
+ASCIIDOC=asciidoc
+ASCIIDOC_EXTRA = --unsafe
+MANPAGE_XSL = manpage-normal.xsl
+XMLTO_EXTRA =
+INSTALL?=install
+RM ?= rm -f
+DOC_REF = origin/man
+HTML_REF = origin/html
+
+infodir?=$(prefix)/share/info
+MAKEINFO=makeinfo
+INSTALL_INFO=install-info
+DOCBOOK2X_TEXI=docbook2x-texi
+DBLATEX=dblatex
+XMLTO=xmlto
+ifndef PERL_PATH
+	PERL_PATH = /usr/bin/perl
+endif
+
+-include ../config.mak.autogen
+-include ../config.mak
+
+_tmp_tool_path := $(call get-executable,$(ASCIIDOC))
+ifeq ($(_tmp_tool_path),)
+	missing_tools = $(ASCIIDOC)
+endif
+
+_tmp_tool_path := $(call get-executable,$(XMLTO))
+ifeq ($(_tmp_tool_path),)
+	missing_tools += $(XMLTO)
+endif
+
+#
+# For asciidoc ...
+#	-7.1.2,	no extra settings are needed.
+#	8.0-,	set ASCIIDOC8.
+#
+
+#
+# For docbook-xsl ...
+#	-1.68.1,	set ASCIIDOC_NO_ROFF? (based on changelog from 1.73.0)
+#	1.69.0,		no extra settings are needed?
+#	1.69.1-1.71.0,	set DOCBOOK_SUPPRESS_SP?
+#	1.71.1,		no extra settings are needed?
+#	1.72.0,		set DOCBOOK_XSL_172.
+#	1.73.0-,	set ASCIIDOC_NO_ROFF
+#
+
+#
+# If you had been using DOCBOOK_XSL_172 in an attempt to get rid
+# of 'the ".ft C" problem' in your generated manpages, and you
+# instead ended up with weird characters around callouts, try
+# using ASCIIDOC_NO_ROFF instead (it works fine with ASCIIDOC8).
+#
+
+ifdef ASCIIDOC8
+ASCIIDOC_EXTRA += -a asciidoc7compatible
+endif
+ifdef DOCBOOK_XSL_172
+ASCIIDOC_EXTRA += -a perf-asciidoc-no-roff
+MANPAGE_XSL = manpage-1.72.xsl
+else
+	ifdef ASCIIDOC_NO_ROFF
+	# docbook-xsl after 1.72 needs the regular XSL, but will not
+	# pass-thru raw roff codes from asciidoc.conf, so turn them off.
+	ASCIIDOC_EXTRA += -a perf-asciidoc-no-roff
+	endif
+endif
+ifdef MAN_BOLD_LITERAL
+XMLTO_EXTRA += -m manpage-bold-literal.xsl
+endif
+ifdef DOCBOOK_SUPPRESS_SP
+XMLTO_EXTRA += -m manpage-suppress-sp.xsl
+endif
+
+SHELL_PATH ?= $(SHELL)
+# Shell quote;
+SHELL_PATH_SQ = $(subst ','\'',$(SHELL_PATH))
+
+#
+# Please note that there is a minor bug in asciidoc.
+# The version after 6.0.3 _will_ include the patch found here:
+#   http://marc.theaimsgroup.com/?l=perf&m=111558757202243&w=2
+#
+# Until that version is released you may have to apply the patch
+# yourself - yes, all 6 characters of it!
+#
+
+QUIET_SUBDIR0  = +$(MAKE) -C # space to separate -C and subdir
+QUIET_SUBDIR1  =
+
+ifneq ($(findstring $(MAKEFLAGS),w),w)
+PRINT_DIR = --no-print-directory
+else # "make -w"
+NO_SUBDIR = :
+endif
+
+ifneq ($(findstring $(MAKEFLAGS),s),s)
+ifneq ($(V),1)
+	QUIET_ASCIIDOC	= @echo '  ASCIIDOC '$@;
+	QUIET_XMLTO	= @echo '  XMLTO    '$@;
+	QUIET_DB2TEXI	= @echo '  DB2TEXI  '$@;
+	QUIET_MAKEINFO	= @echo '  MAKEINFO '$@;
+	QUIET_DBLATEX	= @echo '  DBLATEX  '$@;
+	QUIET_XSLTPROC	= @echo '  XSLTPROC '$@;
+	QUIET_GEN	= @echo '  GEN      '$@;
+	QUIET_STDERR	= 2> /dev/null
+	QUIET_SUBDIR0	= +@subdir=
+	QUIET_SUBDIR1	= ;$(NO_SUBDIR) \
+			   echo '  SUBDIR   ' $$subdir; \
+			  $(MAKE) $(PRINT_DIR) -C $$subdir
+	export V
+endif
+endif
+
+all: html man
+
+html: $(DOC_HTML)
+
+$(DOC_HTML) $(DOC_MAN1) $(DOC_MAN5) $(DOC_MAN7): asciidoc.conf
+
+man: man1 man5 man7
+man1: $(DOC_MAN1)
+man5: $(DOC_MAN5)
+man7: $(DOC_MAN7)
+
+info: $(OUTPUT)perf.info $(OUTPUT)perfman.info
+
+pdf: $(OUTPUT)user-manual.pdf
+
+install: install-man
+
+check-man-tools:
+ifdef missing_tools
+	$(error "You need to install $(missing_tools) for man pages")
+endif
+
+do-install-man: man
+	$(call QUIET_INSTALL, Documentation-man) \
+		$(INSTALL) -d -m 755 $(DESTDIR)$(man1dir); \
+#		$(INSTALL) -d -m 755 $(DESTDIR)$(man5dir); \
+#		$(INSTALL) -d -m 755 $(DESTDIR)$(man7dir); \
+		$(INSTALL) -m 644 $(DOC_MAN1) $(DESTDIR)$(man1dir); \
+#		$(INSTALL) -m 644 $(DOC_MAN5) $(DESTDIR)$(man5dir); \
+#		$(INSTALL) -m 644 $(DOC_MAN7) $(DESTDIR)$(man7dir)
+
+install-man: check-man-tools man
+
+ifdef missing_tools
+  DO_INSTALL_MAN = $(warning Please install $(missing_tools) to have the man pages installed)
+else
+  DO_INSTALL_MAN = do-install-man
+endif
+
+try-install-man: $(DO_INSTALL_MAN)
+
+install-info: info
+	$(call QUIET_INSTALL, Documentation-info) \
+		$(INSTALL) -d -m 755 $(DESTDIR)$(infodir); \
+		$(INSTALL) -m 644 $(OUTPUT)perf.info $(OUTPUT)perfman.info $(DESTDIR)$(infodir); \
+	if test -r $(DESTDIR)$(infodir)/dir; then \
+		$(INSTALL_INFO) --info-dir=$(DESTDIR)$(infodir) perf.info ;\
+		$(INSTALL_INFO) --info-dir=$(DESTDIR)$(infodir) perfman.info ;\
+	else \
+	  echo "No directory found in $(DESTDIR)$(infodir)" >&2 ; \
+	fi
+
+install-pdf: pdf
+	$(call QUIET_INSTALL, Documentation-pdf) \
+		$(INSTALL) -d -m 755 $(DESTDIR)$(pdfdir); \
+		$(INSTALL) -m 644 $(OUTPUT)user-manual.pdf $(DESTDIR)$(pdfdir)
+
+#install-html: html
+#	'$(SHELL_PATH_SQ)' ./install-webdoc.sh $(DESTDIR)$(htmldir)
+
+
+#
+# Determine "include::" file references in asciidoc files.
+#
+$(OUTPUT)doc.dep : $(wildcard *.txt) build-docdep.perl
+	$(QUIET_GEN)$(RM) $@+ $@ && \
+	$(PERL_PATH) ./build-docdep.perl >$@+ $(QUIET_STDERR) && \
+	mv $@+ $@
+
+-include $(OUPTUT)doc.dep
+
+_cmds_txt = cmds-ancillaryinterrogators.txt \
+	cmds-ancillarymanipulators.txt \
+	cmds-mainporcelain.txt \
+	cmds-plumbinginterrogators.txt \
+	cmds-plumbingmanipulators.txt \
+	cmds-synchingrepositories.txt \
+	cmds-synchelpers.txt \
+	cmds-purehelpers.txt \
+	cmds-foreignscminterface.txt
+cmds_txt=$(addprefix $(OUTPUT),$(_cmds_txt))
+
+$(cmds_txt): $(OUTPUT)cmd-list.made
+
+$(OUTPUT)cmd-list.made: cmd-list.perl ../command-list.txt $(MAN1_TXT)
+	$(QUIET_GEN)$(RM) $@ && \
+	$(PERL_PATH) ./cmd-list.perl ../command-list.txt $(QUIET_STDERR) && \
+	date >$@
+
+CLEAN_FILES =									\
+	$(MAN_XML) $(addsuffix +,$(MAN_XML))					\
+	$(MAN_HTML) $(addsuffix +,$(MAN_HTML))					\
+	$(DOC_HTML) $(DOC_MAN1) $(DOC_MAN5) $(DOC_MAN7)				\
+	$(OUTPUT)*.texi $(OUTPUT)*.texi+ $(OUTPUT)*.texi++			\
+	$(OUTPUT)perf.info $(OUTPUT)perfman.info				\
+	$(OUTPUT)howto-index.txt $(OUTPUT)howto/*.html $(OUTPUT)doc.dep		\
+	$(OUTPUT)technical/api-*.html $(OUTPUT)technical/api-index.txt		\
+	$(cmds_txt) $(OUTPUT)*.made
+clean:
+	$(call QUIET_CLEAN, Documentation) $(RM) $(CLEAN_FILES)
+
+$(MAN_HTML): $(OUTPUT)%.html : %.txt
+	$(QUIET_ASCIIDOC)$(RM) $@+ $@ && \
+	$(ASCIIDOC) -b xhtml11 -d manpage -f asciidoc.conf \
+		$(ASCIIDOC_EXTRA) -aperf_version=$(PERF_VERSION) -o $@+ $< && \
+	mv $@+ $@
+
+$(OUTPUT)%.1 $(OUTPUT)%.5 $(OUTPUT)%.7 : $(OUTPUT)%.xml
+	$(QUIET_XMLTO)$(RM) $@ && \
+	$(XMLTO) -o $(OUTPUT). -m $(MANPAGE_XSL) $(XMLTO_EXTRA) man $<
+
+$(OUTPUT)%.xml : %.txt
+	$(QUIET_ASCIIDOC)$(RM) $@+ $@ && \
+	$(ASCIIDOC) -b docbook -d manpage -f asciidoc.conf \
+		$(ASCIIDOC_EXTRA) -aperf_version=$(PERF_VERSION) -o $@+ $< && \
+	mv $@+ $@
+
+XSLT = docbook.xsl
+XSLTOPTS = --xinclude --stringparam html.stylesheet docbook-xsl.css
+
+$(OUTPUT)user-manual.html: $(OUTPUT)user-manual.xml
+	$(QUIET_XSLTPROC)xsltproc $(XSLTOPTS) -o $@ $(XSLT) $<
+
+$(OUTPUT)perf.info: $(OUTPUT)user-manual.texi
+	$(QUIET_MAKEINFO)$(MAKEINFO) --no-split -o $@ $(OUTPUT)user-manual.texi
+
+$(OUTPUT)user-manual.texi: $(OUTPUT)user-manual.xml
+	$(QUIET_DB2TEXI)$(RM) $@+ $@ && \
+	$(DOCBOOK2X_TEXI) $(OUTPUT)user-manual.xml --encoding=UTF-8 --to-stdout >$@++ && \
+	$(PERL_PATH) fix-texi.perl <$@++ >$@+ && \
+	rm $@++ && \
+	mv $@+ $@
+
+$(OUTPUT)user-manual.pdf: $(OUTPUT)user-manual.xml
+	$(QUIET_DBLATEX)$(RM) $@+ $@ && \
+	$(DBLATEX) -o $@+ -p /etc/asciidoc/dblatex/asciidoc-dblatex.xsl -s /etc/asciidoc/dblatex/asciidoc-dblatex.sty $< && \
+	mv $@+ $@
+
+$(OUTPUT)perfman.texi: $(MAN_XML) cat-texi.perl
+	$(QUIET_DB2TEXI)$(RM) $@+ $@ && \
+	($(foreach xml,$(MAN_XML),$(DOCBOOK2X_TEXI) --encoding=UTF-8 \
+		--to-stdout $(xml) &&) true) > $@++ && \
+	$(PERL_PATH) cat-texi.perl $@ <$@++ >$@+ && \
+	rm $@++ && \
+	mv $@+ $@
+
+$(OUTPUT)perfman.info: $(OUTPUT)perfman.texi
+	$(QUIET_MAKEINFO)$(MAKEINFO) --no-split --no-validate $*.texi
+
+$(patsubst %.txt,%.texi,$(MAN_TXT)): %.texi : %.xml
+	$(QUIET_DB2TEXI)$(RM) $@+ $@ && \
+	$(DOCBOOK2X_TEXI) --to-stdout $*.xml >$@+ && \
+	mv $@+ $@
+
+howto-index.txt: howto-index.sh $(wildcard howto/*.txt)
+	$(QUIET_GEN)$(RM) $@+ $@ && \
+	'$(SHELL_PATH_SQ)' ./howto-index.sh $(wildcard howto/*.txt) >$@+ && \
+	mv $@+ $@
+
+$(patsubst %,%.html,$(ARTICLES)) : %.html : %.txt
+	$(QUIET_ASCIIDOC)$(ASCIIDOC) -b xhtml11 $*.txt
+
+WEBDOC_DEST = /pub/software/tools/perf/docs
+
+$(patsubst %.txt,%.html,$(wildcard howto/*.txt)): %.html : %.txt
+	$(QUIET_ASCIIDOC)$(RM) $@+ $@ && \
+	sed -e '1,/^$$/d' $< | $(ASCIIDOC) -b xhtml11 - >$@+ && \
+	mv $@+ $@
+
+# UNIMPLEMENTED
+#install-webdoc : html
+#	'$(SHELL_PATH_SQ)' ./install-webdoc.sh $(WEBDOC_DEST)
+
+# quick-install: quick-install-man
+
+# quick-install-man:
+#	'$(SHELL_PATH_SQ)' ./install-doc-quick.sh $(DOC_REF) $(DESTDIR)$(mandir)
+
+#quick-install-html:
+#	'$(SHELL_PATH_SQ)' ./install-doc-quick.sh $(HTML_REF) $(DESTDIR)$(htmldir)
diff --git a/tools/perf/Documentation/android.txt b/tools/perf/Documentation/android.txt
new file mode 100644
index 000000000..8484c3a04
--- /dev/null
+++ b/tools/perf/Documentation/android.txt
@@ -0,0 +1,78 @@
+How to compile perf for Android
+=========================================
+
+I. Set the Android NDK environment
+------------------------------------------------
+
+(a). Use the Android NDK
+------------------------------------------------
+1. You need to download and install the Android Native Development Kit (NDK).
+Set the NDK variable to point to the path where you installed the NDK:
+  export NDK=/path/to/android-ndk
+
+2. Set cross-compiling environment variables for NDK toolchain and sysroot.
+For arm:
+  export NDK_TOOLCHAIN=${NDK}/toolchains/arm-linux-androideabi-4.6/prebuilt/linux-x86/bin/arm-linux-androideabi-
+  export NDK_SYSROOT=${NDK}/platforms/android-9/arch-arm
+For x86:
+  export NDK_TOOLCHAIN=${NDK}/toolchains/x86-4.6/prebuilt/linux-x86/bin/i686-linux-android-
+  export NDK_SYSROOT=${NDK}/platforms/android-9/arch-x86
+
+This method is not working for Android NDK versions up to Revision 8b.
+perf uses some bionic enhancements that are not included in these NDK versions.
+You can use method (b) described below instead.
+
+(b). Use the Android source tree
+-----------------------------------------------
+1. Download the master branch of the Android source tree.
+Set the environment for the target you want using:
+  source build/envsetup.sh
+  lunch
+
+2. Build your own NDK sysroot to contain latest bionic changes and set the
+NDK sysroot environment variable.
+  cd ${ANDROID_BUILD_TOP}/ndk
+For arm:
+  ./build/tools/build-ndk-sysroot.sh --abi=arm
+  export NDK_SYSROOT=${ANDROID_BUILD_TOP}/ndk/build/platforms/android-3/arch-arm
+For x86:
+  ./build/tools/build-ndk-sysroot.sh --abi=x86
+  export NDK_SYSROOT=${ANDROID_BUILD_TOP}/ndk/build/platforms/android-3/arch-x86
+
+3. Set the NDK toolchain environment variable.
+For arm:
+  export NDK_TOOLCHAIN=${ANDROID_TOOLCHAIN}/arm-linux-androideabi-
+For x86:
+  export NDK_TOOLCHAIN=${ANDROID_TOOLCHAIN}/i686-linux-android-
+
+II. Compile perf for Android
+------------------------------------------------
+You need to run make with the NDK toolchain and sysroot defined above:
+For arm:
+  make ARCH=arm CROSS_COMPILE=${NDK_TOOLCHAIN} CFLAGS="--sysroot=${NDK_SYSROOT}"
+For x86:
+  make ARCH=x86 CROSS_COMPILE=${NDK_TOOLCHAIN} CFLAGS="--sysroot=${NDK_SYSROOT}"
+
+III. Install perf
+-----------------------------------------------
+You need to connect to your Android device/emulator using adb.
+Install perf using:
+  adb push perf /data/perf
+
+If you also want to use perf-archive you need busybox tools for Android.
+For installing perf-archive, you first need to replace #!/bin/bash with #!/system/bin/sh:
+  sed 's/#!\/bin\/bash/#!\/system\/bin\/sh/g' perf-archive >> /tmp/perf-archive
+  chmod +x /tmp/perf-archive
+  adb push /tmp/perf-archive /data/perf-archive
+
+IV. Environment settings for running perf
+------------------------------------------------
+Some perf features need environment variables to run properly.
+You need to set these before running perf on the target:
+  adb shell
+  # PERF_PAGER=cat
+
+IV. Run perf
+------------------------------------------------
+Run perf on your device/emulator to which you previously connected using adb:
+  # ./data/perf
diff --git a/tools/perf/Documentation/asciidoc.conf b/tools/perf/Documentation/asciidoc.conf
new file mode 100644
index 000000000..356b23a40
--- /dev/null
+++ b/tools/perf/Documentation/asciidoc.conf
@@ -0,0 +1,91 @@
+## linkperf: macro
+#
+# Usage: linkperf:command[manpage-section]
+#
+# Note, {0} is the manpage section, while {target} is the command.
+#
+# Show PERF link as: <command>(<section>); if section is defined, else just show
+# the command.
+
+[macros]
+(?su)[\\]?(?P<name>linkperf):(?P<target>\S*?)\[(?P<attrlist>.*?)\]=
+
+[attributes]
+asterisk=&#42;
+plus=&#43;
+caret=&#94;
+startsb=&#91;
+endsb=&#93;
+tilde=&#126;
+
+ifdef::backend-docbook[]
+[linkperf-inlinemacro]
+{0%{target}}
+{0#<citerefentry>}
+{0#<refentrytitle>{target}</refentrytitle><manvolnum>{0}</manvolnum>}
+{0#</citerefentry>}
+endif::backend-docbook[]
+
+ifdef::backend-docbook[]
+ifndef::perf-asciidoc-no-roff[]
+# "unbreak" docbook-xsl v1.68 for manpages. v1.69 works with or without this.
+# v1.72 breaks with this because it replaces dots not in roff requests.
+[listingblock]
+<example><title>{title}</title>
+<literallayout>
+ifdef::doctype-manpage[]
+&#10;.ft C&#10;
+endif::doctype-manpage[]
+|
+ifdef::doctype-manpage[]
+&#10;.ft&#10;
+endif::doctype-manpage[]
+</literallayout>
+{title#}</example>
+endif::perf-asciidoc-no-roff[]
+
+ifdef::perf-asciidoc-no-roff[]
+ifdef::doctype-manpage[]
+# The following two small workarounds insert a simple paragraph after screen
+[listingblock]
+<example><title>{title}</title>
+<literallayout>
+|
+</literallayout><simpara></simpara>
+{title#}</example>
+
+[verseblock]
+<formalpara{id? id="{id}"}><title>{title}</title><para>
+{title%}<literallayout{id? id="{id}"}>
+{title#}<literallayout>
+|
+</literallayout>
+{title#}</para></formalpara>
+{title%}<simpara></simpara>
+endif::doctype-manpage[]
+endif::perf-asciidoc-no-roff[]
+endif::backend-docbook[]
+
+ifdef::doctype-manpage[]
+ifdef::backend-docbook[]
+[header]
+template::[header-declarations]
+<refentry>
+<refmeta>
+<refentrytitle>{mantitle}</refentrytitle>
+<manvolnum>{manvolnum}</manvolnum>
+<refmiscinfo class="source">perf</refmiscinfo>
+<refmiscinfo class="version">{perf_version}</refmiscinfo>
+<refmiscinfo class="manual">perf Manual</refmiscinfo>
+</refmeta>
+<refnamediv>
+  <refname>{manname}</refname>
+  <refpurpose>{manpurpose}</refpurpose>
+</refnamediv>
+endif::backend-docbook[]
+endif::doctype-manpage[]
+
+ifdef::backend-xhtml11[]
+[linkperf-inlinemacro]
+<a href="{target}.html">{target}{0?({0})}</a>
+endif::backend-xhtml11[]
diff --git a/tools/perf/Documentation/examples.txt b/tools/perf/Documentation/examples.txt
new file mode 100644
index 000000000..a4e392156
--- /dev/null
+++ b/tools/perf/Documentation/examples.txt
@@ -0,0 +1,225 @@
+
+		------------------------------
+		****** perf by examples ******
+		------------------------------
+
+[ From an e-mail by Ingo Molnar, http://lkml.org/lkml/2009/8/4/346 ]
+
+
+First, discovery/enumeration of available counters can be done via
+'perf list':
+
+titan:~> perf list
+  [...]
+  kmem:kmalloc                             [Tracepoint event]
+  kmem:kmem_cache_alloc                    [Tracepoint event]
+  kmem:kmalloc_node                        [Tracepoint event]
+  kmem:kmem_cache_alloc_node               [Tracepoint event]
+  kmem:kfree                               [Tracepoint event]
+  kmem:kmem_cache_free                     [Tracepoint event]
+  kmem:mm_page_free                        [Tracepoint event]
+  kmem:mm_page_free_batched                [Tracepoint event]
+  kmem:mm_page_alloc                       [Tracepoint event]
+  kmem:mm_page_alloc_zone_locked           [Tracepoint event]
+  kmem:mm_page_pcpu_drain                  [Tracepoint event]
+  kmem:mm_page_alloc_extfrag               [Tracepoint event]
+
+Then any (or all) of the above event sources can be activated and
+measured. For example the page alloc/free properties of a 'hackbench
+run' are:
+
+ titan:~> perf stat -e kmem:mm_page_pcpu_drain -e kmem:mm_page_alloc
+ -e kmem:mm_page_free_batched -e kmem:mm_page_free ./hackbench 10
+ Time: 0.575
+
+ Performance counter stats for './hackbench 10':
+
+          13857  kmem:mm_page_pcpu_drain
+          27576  kmem:mm_page_alloc
+           6025  kmem:mm_page_free_batched
+          20934  kmem:mm_page_free
+
+    0.613972165  seconds time elapsed
+
+You can observe the statistical properties as well, by using the
+'repeat the workload N times' feature of perf stat:
+
+ titan:~> perf stat --repeat 5 -e kmem:mm_page_pcpu_drain -e
+   kmem:mm_page_alloc -e kmem:mm_page_free_batched -e
+   kmem:mm_page_free ./hackbench 10
+ Time: 0.627
+ Time: 0.644
+ Time: 0.564
+ Time: 0.559
+ Time: 0.626
+
+ Performance counter stats for './hackbench 10' (5 runs):
+
+          12920  kmem:mm_page_pcpu_drain    ( +-   3.359% )
+          25035  kmem:mm_page_alloc         ( +-   3.783% )
+           6104  kmem:mm_page_free_batched  ( +-   0.934% )
+          18376  kmem:mm_page_free	    ( +-   4.941% )
+
+    0.643954516  seconds time elapsed   ( +-   2.363% )
+
+Furthermore, these tracepoints can be used to sample the workload as
+well. For example the page allocations done by a 'git gc' can be
+captured the following way:
+
+ titan:~/git> perf record -e kmem:mm_page_alloc -c 1 ./git gc
+ Counting objects: 1148, done.
+ Delta compression using up to 2 threads.
+ Compressing objects: 100% (450/450), done.
+ Writing objects: 100% (1148/1148), done.
+ Total 1148 (delta 690), reused 1148 (delta 690)
+ [ perf record: Captured and wrote 0.267 MB perf.data (~11679 samples) ]
+
+To check which functions generated page allocations:
+
+ titan:~/git> perf report
+ # Samples: 10646
+ #
+ # Overhead          Command               Shared Object
+ # ........  ...............  ..........................
+ #
+    23.57%       git-repack  /lib64/libc-2.5.so
+    21.81%              git  /lib64/libc-2.5.so
+    14.59%              git  ./git
+    11.79%       git-repack  ./git
+     7.12%              git  /lib64/ld-2.5.so
+     3.16%       git-repack  /lib64/libpthread-2.5.so
+     2.09%       git-repack  /bin/bash
+     1.97%               rm  /lib64/libc-2.5.so
+     1.39%               mv  /lib64/ld-2.5.so
+     1.37%               mv  /lib64/libc-2.5.so
+     1.12%       git-repack  /lib64/ld-2.5.so
+     0.95%               rm  /lib64/ld-2.5.so
+     0.90%  git-update-serv  /lib64/libc-2.5.so
+     0.73%  git-update-serv  /lib64/ld-2.5.so
+     0.68%             perf  /lib64/libpthread-2.5.so
+     0.64%       git-repack  /usr/lib64/libz.so.1.2.3
+
+Or to see it on a more finegrained level:
+
+titan:~/git> perf report --sort comm,dso,symbol
+# Samples: 10646
+#
+# Overhead          Command               Shared Object  Symbol
+# ........  ...............  ..........................  ......
+#
+     9.35%       git-repack  ./git                       [.] insert_obj_hash
+     9.12%              git  ./git                       [.] insert_obj_hash
+     7.31%              git  /lib64/libc-2.5.so          [.] memcpy
+     6.34%       git-repack  /lib64/libc-2.5.so          [.] _int_malloc
+     6.24%       git-repack  /lib64/libc-2.5.so          [.] memcpy
+     5.82%       git-repack  /lib64/libc-2.5.so          [.] __GI___fork
+     5.47%              git  /lib64/libc-2.5.so          [.] _int_malloc
+     2.99%              git  /lib64/libc-2.5.so          [.] memset
+
+Furthermore, call-graph sampling can be done too, of page
+allocations - to see precisely what kind of page allocations there
+are:
+
+ titan:~/git> perf record -g -e kmem:mm_page_alloc -c 1 ./git gc
+ Counting objects: 1148, done.
+ Delta compression using up to 2 threads.
+ Compressing objects: 100% (450/450), done.
+ Writing objects: 100% (1148/1148), done.
+ Total 1148 (delta 690), reused 1148 (delta 690)
+ [ perf record: Captured and wrote 0.963 MB perf.data (~42069 samples) ]
+
+ titan:~/git> perf report -g
+ # Samples: 10686
+ #
+ # Overhead          Command               Shared Object
+ # ........  ...............  ..........................
+ #
+    23.25%       git-repack  /lib64/libc-2.5.so
+                |
+                |--50.00%-- _int_free
+                |
+                |--37.50%-- __GI___fork
+                |          make_child
+                |
+                |--12.50%-- ptmalloc_unlock_all2
+                |          make_child
+                |
+                 --6.25%-- __GI_strcpy
+    21.61%              git  /lib64/libc-2.5.so
+                |
+                |--30.00%-- __GI_read
+                |          |
+                |           --83.33%-- git_config_from_file
+                |                     git_config
+                |                     |
+   [...]
+
+Or you can observe the whole system's page allocations for 10
+seconds:
+
+titan:~/git> perf stat -a -e kmem:mm_page_pcpu_drain -e
+kmem:mm_page_alloc -e kmem:mm_page_free_batched -e
+kmem:mm_page_free sleep 10
+
+ Performance counter stats for 'sleep 10':
+
+         171585  kmem:mm_page_pcpu_drain
+         322114  kmem:mm_page_alloc
+          73623  kmem:mm_page_free_batched
+         254115  kmem:mm_page_free
+
+   10.000591410  seconds time elapsed
+
+Or observe how fluctuating the page allocations are, via statistical
+analysis done over ten 1-second intervals:
+
+ titan:~/git> perf stat --repeat 10 -a -e kmem:mm_page_pcpu_drain -e
+   kmem:mm_page_alloc -e kmem:mm_page_free_batched -e
+   kmem:mm_page_free sleep 1
+
+ Performance counter stats for 'sleep 1' (10 runs):
+
+          17254  kmem:mm_page_pcpu_drain    ( +-   3.709% )
+          34394  kmem:mm_page_alloc         ( +-   4.617% )
+           7509  kmem:mm_page_free_batched  ( +-   4.820% )
+          25653  kmem:mm_page_free	    ( +-   3.672% )
+
+    1.058135029  seconds time elapsed   ( +-   3.089% )
+
+Or you can annotate the recorded 'git gc' run on a per symbol basis
+and check which instructions/source-code generated page allocations:
+
+ titan:~/git> perf annotate __GI___fork
+ ------------------------------------------------
+  Percent |      Source code & Disassembly of libc-2.5.so
+ ------------------------------------------------
+          :
+          :
+          :      Disassembly of section .plt:
+          :      Disassembly of section .text:
+          :
+          :      00000031a2e95560 <__fork>:
+ [...]
+     0.00 :        31a2e95602:   b8 38 00 00 00          mov    $0x38,%eax
+     0.00 :        31a2e95607:   0f 05                   syscall
+    83.42 :        31a2e95609:   48 3d 00 f0 ff ff       cmp    $0xfffffffffffff000,%rax
+     0.00 :        31a2e9560f:   0f 87 4d 01 00 00       ja     31a2e95762 <__fork+0x202>
+     0.00 :        31a2e95615:   85 c0                   test   %eax,%eax
+
+( this shows that 83.42% of __GI___fork's page allocations come from
+  the 0x38 system call it performs. )
+
+etc. etc. - a lot more is possible. I could list a dozen of
+other different usecases straight away - neither of which is
+possible via /proc/vmstat.
+
+/proc/vmstat is not in the same league really, in terms of
+expressive power of system analysis and performance
+analysis.
+
+All that the above results needed were those new tracepoints
+in include/tracing/events/kmem.h.
+
+	Ingo
+
+
diff --git a/tools/perf/Documentation/jit-interface.txt b/tools/perf/Documentation/jit-interface.txt
new file mode 100644
index 000000000..a8656f564
--- /dev/null
+++ b/tools/perf/Documentation/jit-interface.txt
@@ -0,0 +1,15 @@
+perf supports a simple JIT interface to resolve symbols for dynamic code generated
+by a JIT.
+
+The JIT has to write a /tmp/perf-%d.map  (%d = pid of process) file
+
+This is a text file.
+
+Each line has the following format, fields separated with spaces:
+
+START SIZE symbolname
+
+START and SIZE are hex numbers without 0x.
+symbolname is the rest of the line, so it could contain special characters.
+
+The ownership of the file has to match the process.
diff --git a/tools/perf/Documentation/manpage-1.72.xsl b/tools/perf/Documentation/manpage-1.72.xsl
new file mode 100644
index 000000000..b4d315cb8
--- /dev/null
+++ b/tools/perf/Documentation/manpage-1.72.xsl
@@ -0,0 +1,14 @@
+<!-- manpage-1.72.xsl:
+     special settings for manpages rendered from asciidoc+docbook
+     handles peculiarities in docbook-xsl 1.72.0 -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<xsl:import href="manpage-base.xsl"/>
+
+<!-- these are the special values for the roff control characters
+     needed for docbook-xsl 1.72.0 -->
+<xsl:param name="git.docbook.backslash">&#x2593;</xsl:param>
+<xsl:param name="git.docbook.dot"      >&#x2302;</xsl:param>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/manpage-base.xsl b/tools/perf/Documentation/manpage-base.xsl
new file mode 100644
index 000000000..a264fa616
--- /dev/null
+++ b/tools/perf/Documentation/manpage-base.xsl
@@ -0,0 +1,35 @@
+<!-- manpage-base.xsl:
+     special formatting for manpages rendered from asciidoc+docbook -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<!-- these params silence some output from xmlto -->
+<xsl:param name="man.output.quietly" select="1"/>
+<xsl:param name="refentry.meta.get.quietly" select="1"/>
+
+<!-- convert asciidoc callouts to man page format;
+     git.docbook.backslash and git.docbook.dot params
+     must be supplied by another XSL file or other means -->
+<xsl:template match="co">
+	<xsl:value-of select="concat(
+			      $git.docbook.backslash,'fB(',
+			      substring-after(@id,'-'),')',
+			      $git.docbook.backslash,'fR')"/>
+</xsl:template>
+<xsl:template match="calloutlist">
+	<xsl:value-of select="$git.docbook.dot"/>
+	<xsl:text>sp&#10;</xsl:text>
+	<xsl:apply-templates/>
+	<xsl:text>&#10;</xsl:text>
+</xsl:template>
+<xsl:template match="callout">
+	<xsl:value-of select="concat(
+			      $git.docbook.backslash,'fB',
+			      substring-after(@arearefs,'-'),
+			      '. ',$git.docbook.backslash,'fR')"/>
+	<xsl:apply-templates/>
+	<xsl:value-of select="$git.docbook.dot"/>
+	<xsl:text>br&#10;</xsl:text>
+</xsl:template>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/manpage-bold-literal.xsl b/tools/perf/Documentation/manpage-bold-literal.xsl
new file mode 100644
index 000000000..608eb5df6
--- /dev/null
+++ b/tools/perf/Documentation/manpage-bold-literal.xsl
@@ -0,0 +1,17 @@
+<!-- manpage-bold-literal.xsl:
+     special formatting for manpages rendered from asciidoc+docbook -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<!-- render literal text as bold (instead of plain or monospace);
+     this makes literal text easier to distinguish in manpages
+     viewed on a tty -->
+<xsl:template match="literal">
+	<xsl:value-of select="$git.docbook.backslash"/>
+	<xsl:text>fB</xsl:text>
+	<xsl:apply-templates/>
+	<xsl:value-of select="$git.docbook.backslash"/>
+	<xsl:text>fR</xsl:text>
+</xsl:template>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/manpage-normal.xsl b/tools/perf/Documentation/manpage-normal.xsl
new file mode 100644
index 000000000..a48f5b11f
--- /dev/null
+++ b/tools/perf/Documentation/manpage-normal.xsl
@@ -0,0 +1,13 @@
+<!-- manpage-normal.xsl:
+     special settings for manpages rendered from asciidoc+docbook
+     handles anything we want to keep away from docbook-xsl 1.72.0 -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<xsl:import href="manpage-base.xsl"/>
+
+<!-- these are the normal values for the roff control characters -->
+<xsl:param name="git.docbook.backslash">\</xsl:param>
+<xsl:param name="git.docbook.dot"	>.</xsl:param>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/manpage-suppress-sp.xsl b/tools/perf/Documentation/manpage-suppress-sp.xsl
new file mode 100644
index 000000000..a63c7632a
--- /dev/null
+++ b/tools/perf/Documentation/manpage-suppress-sp.xsl
@@ -0,0 +1,21 @@
+<!-- manpage-suppress-sp.xsl:
+     special settings for manpages rendered from asciidoc+docbook
+     handles erroneous, inline .sp in manpage output of some
+     versions of docbook-xsl -->
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+		version="1.0">
+
+<!-- attempt to work around spurious .sp at the tail of the line
+     that some versions of docbook stylesheets seem to add -->
+<xsl:template match="simpara">
+  <xsl:variable name="content">
+    <xsl:apply-templates/>
+  </xsl:variable>
+  <xsl:value-of select="normalize-space($content)"/>
+  <xsl:if test="not(ancestor::authorblurb) and
+                not(ancestor::personblurb)">
+    <xsl:text>&#10;&#10;</xsl:text>
+  </xsl:if>
+</xsl:template>
+
+</xsl:stylesheet>
diff --git a/tools/perf/Documentation/perf-annotate.txt b/tools/perf/Documentation/perf-annotate.txt
new file mode 100644
index 000000000..e9cd39a92
--- /dev/null
+++ b/tools/perf/Documentation/perf-annotate.txt
@@ -0,0 +1,101 @@
+perf-annotate(1)
+================
+
+NAME
+----
+perf-annotate - Read perf.data (created by perf record) and display annotated code
+
+SYNOPSIS
+--------
+[verse]
+'perf annotate' [-i <file> | --input=file] [symbol_name]
+
+DESCRIPTION
+-----------
+This command reads the input file and displays an annotated version of the
+code. If the object file has debug symbols then the source code will be
+displayed alongside assembly code.
+
+If there is no debug info in the object, then annotated assembly is displayed.
+
+OPTIONS
+-------
+-i::
+--input=::
+        Input file name. (default: perf.data unless stdin is a fifo)
+
+-d::
+--dsos=<dso[,dso...]>::
+        Only consider symbols in these dsos.
+-s::
+--symbol=<symbol>::
+        Symbol to annotate.
+
+-f::
+--force::
+        Don't complain, do it.
+
+-v::
+--verbose::
+        Be more verbose. (Show symbol address, etc)
+
+-D::
+--dump-raw-trace::
+        Dump raw trace in ASCII.
+
+-k::
+--vmlinux=<file>::
+        vmlinux pathname.
+
+-m::
+--modules::
+        Load module symbols. WARNING: use only with -k and LIVE kernel.
+
+-l::
+--print-line::
+        Print matching source lines (may be slow).
+
+-P::
+--full-paths::
+        Don't shorten the displayed pathnames.
+
+--stdio:: Use the stdio interface.
+
+--tui:: Use the TUI interface. Use of --tui requires a tty, if one is not
+	present, as when piping to other commands, the stdio interface is
+	used. This interfaces starts by centering on the line with more
+	samples, TAB/UNTAB cycles through the lines with more samples.
+
+--gtk:: Use the GTK interface.
+
+-C::
+--cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can
+	be provided as a comma-separated list with no space: 0,1. Ranges of
+	CPUs are specified with -: 0-2. Default is to report samples on all
+	CPUs.
+
+--asm-raw::
+	Show raw instruction encoding of assembly instructions.
+
+--source::
+	Interleave source code with assembly code. Enabled by default,
+	disable with --no-source.
+
+--symfs=<directory>::
+        Look for files with symbols relative to this directory.
+
+-M::
+--disassembler-style=:: Set disassembler style for objdump.
+
+--objdump=<path>::
+        Path to objdump binary.
+
+--skip-missing::
+	Skip symbols that cannot be annotated.
+
+--group::
+	Show event group information together
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-archive.txt b/tools/perf/Documentation/perf-archive.txt
new file mode 100644
index 000000000..ac6ecbb3e
--- /dev/null
+++ b/tools/perf/Documentation/perf-archive.txt
@@ -0,0 +1,22 @@
+perf-archive(1)
+===============
+
+NAME
+----
+perf-archive - Create archive with object files with build-ids found in perf.data file
+
+SYNOPSIS
+--------
+[verse]
+'perf archive' [file]
+
+DESCRIPTION
+-----------
+This command runs perf-buildid-list --with-hits, and collects the files with the
+buildids found so that analysis of perf.data contents can be possible on another
+machine.
+
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-buildid-list[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-bench.txt b/tools/perf/Documentation/perf-bench.txt
new file mode 100644
index 000000000..f6480cbf3
--- /dev/null
+++ b/tools/perf/Documentation/perf-bench.txt
@@ -0,0 +1,218 @@
+perf-bench(1)
+=============
+
+NAME
+----
+perf-bench - General framework for benchmark suites
+
+SYNOPSIS
+--------
+[verse]
+'perf bench' [<common options>] <subsystem> <suite> [<options>]
+
+DESCRIPTION
+-----------
+This 'perf bench' command is a general framework for benchmark suites.
+
+COMMON OPTIONS
+--------------
+-r::
+--repeat=::
+Specify amount of times to repeat the run (default 10).
+
+-f::
+--format=::
+Specify format style.
+Current available format styles are:
+
+'default'::
+Default style. This is mainly for human reading.
+---------------------
+% perf bench sched pipe                      # with no style specified
+(executing 1000000 pipe operations between two tasks)
+        Total time:5.855 sec
+                5.855061 usecs/op
+		170792 ops/sec
+---------------------
+
+'simple'::
+This simple style is friendly for automated
+processing by scripts.
+---------------------
+% perf bench --format=simple sched pipe      # specified simple
+5.988
+---------------------
+
+SUBSYSTEM
+---------
+
+'sched'::
+	Scheduler and IPC mechanisms.
+
+'mem'::
+	Memory access performance.
+
+'numa'::
+	NUMA scheduling and MM benchmarks.
+
+'futex'::
+	Futex stressing benchmarks.
+
+'all'::
+	All benchmark subsystems.
+
+SUITES FOR 'sched'
+~~~~~~~~~~~~~~~~~~
+*messaging*::
+Suite for evaluating performance of scheduler and IPC mechanisms.
+Based on hackbench by Rusty Russell.
+
+Options of *messaging*
+^^^^^^^^^^^^^^^^^^^^^^
+-p::
+--pipe::
+Use pipe() instead of socketpair()
+
+-t::
+--thread::
+Be multi thread instead of multi process
+
+-g::
+--group=::
+Specify number of groups
+
+-l::
+--loop=::
+Specify number of loops
+
+Example of *messaging*
+^^^^^^^^^^^^^^^^^^^^^^
+
+---------------------
+% perf bench sched messaging                 # run with default
+options (20 sender and receiver processes per group)
+(10 groups == 400 processes run)
+
+      Total time:0.308 sec
+
+% perf bench sched messaging -t -g 20        # be multi-thread, with 20 groups
+(20 sender and receiver threads per group)
+(20 groups == 800 threads run)
+
+      Total time:0.582 sec
+---------------------
+
+*pipe*::
+Suite for pipe() system call.
+Based on pipe-test-1m.c by Ingo Molnar.
+
+Options of *pipe*
+^^^^^^^^^^^^^^^^^
+-l::
+--loop=::
+Specify number of loops.
+
+Example of *pipe*
+^^^^^^^^^^^^^^^^^
+
+---------------------
+% perf bench sched pipe
+(executing 1000000 pipe operations between two tasks)
+
+        Total time:8.091 sec
+                8.091833 usecs/op
+                123581 ops/sec
+
+% perf bench sched pipe -l 1000              # loop 1000
+(executing 1000 pipe operations between two tasks)
+
+        Total time:0.016 sec
+                16.948000 usecs/op
+                59004 ops/sec
+---------------------
+
+SUITES FOR 'mem'
+~~~~~~~~~~~~~~~~
+*memcpy*::
+Suite for evaluating performance of simple memory copy in various ways.
+
+Options of *memcpy*
+^^^^^^^^^^^^^^^^^^^
+-l::
+--length::
+Specify length of memory to copy (default: 1MB).
+Available units are B, KB, MB, GB and TB (case insensitive).
+
+-r::
+--routine::
+Specify routine to copy (default: default).
+Available routines are depend on the architecture.
+On x86-64, x86-64-unrolled, x86-64-movsq and x86-64-movsb are supported.
+
+-i::
+--iterations::
+Repeat memcpy invocation this number of times.
+
+-c::
+--cycle::
+Use perf's cpu-cycles event instead of gettimeofday syscall.
+
+-o::
+--only-prefault::
+Show only the result with page faults before memcpy.
+
+-n::
+--no-prefault::
+Show only the result without page faults before memcpy.
+
+*memset*::
+Suite for evaluating performance of simple memory set in various ways.
+
+Options of *memset*
+^^^^^^^^^^^^^^^^^^^
+-l::
+--length::
+Specify length of memory to set (default: 1MB).
+Available units are B, KB, MB, GB and TB (case insensitive).
+
+-r::
+--routine::
+Specify routine to set (default: default).
+Available routines are depend on the architecture.
+On x86-64, x86-64-unrolled, x86-64-stosq and x86-64-stosb are supported.
+
+-i::
+--iterations::
+Repeat memset invocation this number of times.
+
+-c::
+--cycle::
+Use perf's cpu-cycles event instead of gettimeofday syscall.
+
+-o::
+--only-prefault::
+Show only the result with page faults before memset.
+
+-n::
+--no-prefault::
+Show only the result without page faults before memset.
+
+SUITES FOR 'numa'
+~~~~~~~~~~~~~~~~~
+*mem*::
+Suite for evaluating NUMA workloads.
+
+SUITES FOR 'futex'
+~~~~~~~~~~~~~~~~~~
+*hash*::
+Suite for evaluating hash tables.
+
+*wake*::
+Suite for evaluating wake calls.
+
+*requeue*::
+Suite for evaluating requeue calls.
+
+SEE ALSO
+--------
+linkperf:perf[1]
diff --git a/tools/perf/Documentation/perf-buildid-cache.txt b/tools/perf/Documentation/perf-buildid-cache.txt
new file mode 100644
index 000000000..dd07b55f5
--- /dev/null
+++ b/tools/perf/Documentation/perf-buildid-cache.txt
@@ -0,0 +1,63 @@
+perf-buildid-cache(1)
+=====================
+
+NAME
+----
+perf-buildid-cache - Manage build-id cache.
+
+SYNOPSIS
+--------
+[verse]
+'perf buildid-cache <options>'
+
+DESCRIPTION
+-----------
+This command manages the build-id cache. It can add, remove, update and purge
+files to/from the cache. In the future it should as well set upper limits for
+the space used by the cache, etc.
+
+OPTIONS
+-------
+-a::
+--add=::
+        Add specified file to the cache.
+-k::
+--kcore::
+        Add specified kcore file to the cache. For the current host that is
+        /proc/kcore which requires root permissions to read. Be aware that
+        running 'perf buildid-cache' as root may update root's build-id cache
+        not the user's. Use the -v option to see where the file is created.
+        Note that the copied file contains only code sections not the whole core
+        image. Note also that files "kallsyms" and "modules" must also be in the
+        same directory and are also copied.  All 3 files are created with read
+        permissions for root only. kcore will not be added if there is already a
+        kcore in the cache (with the same build-id) that has the same modules at
+        the same addresses. Use the -v option to see if a copy of kcore is
+        actually made.
+-r::
+--remove=::
+        Remove a cached binary which has same build-id of specified file
+        from the cache.
+-p::
+--purge=::
+        Purge all cached binaries including older caches which have specified
+	path from the cache.
+-M::
+--missing=::
+	List missing build ids in the cache for the specified file.
+-u::
+--update=::
+	Update specified file of the cache. Note that this doesn't remove
+	older entires since those may be still needed for annotating old
+	(or remote) perf.data. Only if there is already a cache which has
+	exactly same build-id, that is replaced by new one. It can be used
+	to update kallsyms and kernel dso to vmlinux in order to support
+	annotation.
+
+-v::
+--verbose::
+	Be more verbose.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-report[1], linkperf:perf-buildid-list[1]
diff --git a/tools/perf/Documentation/perf-buildid-list.txt b/tools/perf/Documentation/perf-buildid-list.txt
new file mode 100644
index 000000000..25c52efcc
--- /dev/null
+++ b/tools/perf/Documentation/perf-buildid-list.txt
@@ -0,0 +1,43 @@
+perf-buildid-list(1)
+====================
+
+NAME
+----
+perf-buildid-list - List the buildids in a perf.data file
+
+SYNOPSIS
+--------
+[verse]
+'perf buildid-list <options>'
+
+DESCRIPTION
+-----------
+This command displays the buildids found in a perf.data file, so that other
+tools can be used to fetch packages with matching symbol tables for use by
+perf report.
+
+It can also be used to show the build id of the running kernel or in an ELF
+file using -i/--input.
+
+OPTIONS
+-------
+-H::
+--with-hits::
+        Show only DSOs with hits.
+-i::
+--input=::
+        Input file name. (default: perf.data unless stdin is a fifo)
+-f::
+--force::
+	Don't do ownership validation.
+-k::
+--kernel::
+	Show running kernel build id.
+-v::
+--verbose::
+	Be more verbose.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-top[1],
+linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-data.txt b/tools/perf/Documentation/perf-data.txt
new file mode 100644
index 000000000..be8fa1a0a
--- /dev/null
+++ b/tools/perf/Documentation/perf-data.txt
@@ -0,0 +1,40 @@
+perf-data(1)
+==============
+
+NAME
+----
+perf-data - Data file related processing
+
+SYNOPSIS
+--------
+[verse]
+'perf data' [<common options>] <command> [<options>]",
+
+DESCRIPTION
+-----------
+Data file related processing.
+
+COMMANDS
+--------
+convert::
+	Converts perf data file into another format (only CTF [1] format is support by now).
+	It's possible to set data-convert debug variable to get debug messages from conversion,
+	like:
+	  perf --debug data-convert data convert ...
+
+OPTIONS for 'convert'
+---------------------
+--to-ctf::
+	Triggers the CTF conversion, specify the path of CTF data directory.
+
+-i::
+	Specify input perf data file path.
+
+-v::
+--verbose::
+        Be more verbose (show counter open errors, etc).
+
+SEE ALSO
+--------
+linkperf:perf[1]
+[1] Common Trace Format - http://www.efficios.com/ctf
diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt
new file mode 100644
index 000000000..d1deb5738
--- /dev/null
+++ b/tools/perf/Documentation/perf-diff.txt
@@ -0,0 +1,214 @@
+perf-diff(1)
+============
+
+NAME
+----
+perf-diff - Read perf.data files and display the differential profile
+
+SYNOPSIS
+--------
+[verse]
+'perf diff' [baseline file] [data file1] [[data file2] ... ]
+
+DESCRIPTION
+-----------
+This command displays the performance difference amongst two or more perf.data
+files captured via perf record.
+
+If no parameters are passed it will assume perf.data.old and perf.data.
+
+The differential profile is displayed only for events matching both
+specified perf.data files.
+
+If no parameters are passed the samples will be sorted by dso and symbol.
+As the perf.data files could come from different binaries, the symbols addresses
+could vary. So perf diff is based on the comparison of the files and
+symbols name.
+
+OPTIONS
+-------
+-D::
+--dump-raw-trace::
+        Dump raw trace in ASCII.
+
+--kallsyms=<file>::
+        kallsyms pathname
+
+-m::
+--modules::
+        Load module symbols. WARNING: use only with -k and LIVE kernel
+
+-d::
+--dsos=::
+	Only consider symbols in these dsos. CSV that understands
+	file://filename entries.  This option will affect the percentage
+	of the Baseline/Delta column.  See --percentage for more info.
+
+-C::
+--comms=::
+	Only consider symbols in these comms. CSV that understands
+	file://filename entries.  This option will affect the percentage
+	of the Baseline/Delta column.  See --percentage for more info.
+
+-S::
+--symbols=::
+	Only consider these symbols. CSV that understands
+	file://filename entries.  This option will affect the percentage
+	of the Baseline/Delta column.  See --percentage for more info.
+
+-s::
+--sort=::
+	Sort by key(s): pid, comm, dso, symbol, cpu, parent, srcline.
+	Please see description of --sort in the perf-report man page.
+
+-t::
+--field-separator=::
+
+	Use a special separator character and don't pad with spaces, replacing
+	all occurrences of this separator in symbol names (and other output)
+	with a '.' character, that thus it's the only non valid separator.
+
+-v::
+--verbose::
+	Be verbose, for instance, show the raw counts in addition to the
+	diff.
+
+-f::
+--force::
+       Don't complain, do it.
+
+--symfs=<directory>::
+        Look for files with symbols relative to this directory.
+
+-b::
+--baseline-only::
+        Show only items with match in baseline.
+
+-c::
+--compute::
+        Differential computation selection - delta,ratio,wdiff (default is delta).
+        See COMPARISON METHODS section for more info.
+
+-p::
+--period::
+        Show period values for both compared hist entries.
+
+-F::
+--formula::
+        Show formula for given computation.
+
+-o::
+--order::
+       Specify compute sorting column number.
+
+--percentage::
+	Determine how to display the overhead percentage of filtered entries.
+	Filters can be applied by --comms, --dsos and/or --symbols options.
+
+	"relative" means it's relative to filtered entries only so that the
+	sum of shown entries will be always 100%.  "absolute" means it retains
+	the original value before and after the filter is applied.
+
+COMPARISON
+----------
+The comparison is governed by the baseline file. The baseline perf.data
+file is iterated for samples. All other perf.data files specified on
+the command line are searched for the baseline sample pair. If the pair
+is found, specified computation is made and result is displayed.
+
+All samples from non-baseline perf.data files, that do not match any
+baseline entry, are displayed with empty space within baseline column
+and possible computation results (delta) in their related column.
+
+Example files samples:
+- file A with samples f1, f2, f3, f4,    f6
+- file B with samples     f2,     f4, f5
+- file C with samples f1, f2,         f5
+
+Example output:
+  x - computation takes place for pair
+  b - baseline sample percentage
+
+- perf diff A B C
+
+  baseline/A compute/B compute/C  samples
+  ---------------------------------------
+  b                    x          f1
+  b          x         x          f2
+  b                               f3
+  b          x                    f4
+  b                               f6
+             x         x          f5
+
+- perf diff B A C
+
+  baseline/B compute/A compute/C  samples
+  ---------------------------------------
+  b          x         x          f2
+  b          x                    f4
+  b                    x          f5
+             x         x          f1
+             x                    f3
+             x                    f6
+
+- perf diff C B A
+
+  baseline/C compute/B compute/A  samples
+  ---------------------------------------
+  b                    x          f1
+  b          x         x          f2
+  b          x                    f5
+                       x          f3
+             x         x          f4
+                       x          f6
+
+COMPARISON METHODS
+------------------
+delta
+~~~~~
+If specified the 'Delta' column is displayed with value 'd' computed as:
+
+  d = A->period_percent - B->period_percent
+
+with:
+  - A/B being matching hist entry from data/baseline file specified
+    (or perf.data/perf.data.old) respectively.
+
+  - period_percent being the % of the hist entry period value within
+    single data file
+
+  - with filtering by -C, -d and/or -S, period_percent might be changed
+    relative to how entries are filtered.  Use --percentage=absolute to
+    prevent such fluctuation.
+
+ratio
+~~~~~
+If specified the 'Ratio' column is displayed with value 'r' computed as:
+
+  r = A->period / B->period
+
+with:
+  - A/B being matching hist entry from data/baseline file specified
+    (or perf.data/perf.data.old) respectively.
+
+  - period being the hist entry period value
+
+wdiff:WEIGHT-B,WEIGHT-A
+~~~~~~~~~~~~~~~~~~~~~~~
+If specified the 'Weighted diff' column is displayed with value 'd' computed as:
+
+   d = B->period * WEIGHT-A - A->period * WEIGHT-B
+
+  - A/B being matching hist entry from data/baseline file specified
+    (or perf.data/perf.data.old) respectively.
+
+  - period being the hist entry period value
+
+  - WEIGHT-A/WEIGHT-B being user supplied weights in the the '-c' option
+    behind ':' separator like '-c wdiff:1,2'.
+    - WEIGHT-A being the weight of the data file
+    - WEIGHT-B being the weight of the baseline data file
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-evlist.txt b/tools/perf/Documentation/perf-evlist.txt
new file mode 100644
index 000000000..1ceb3700f
--- /dev/null
+++ b/tools/perf/Documentation/perf-evlist.txt
@@ -0,0 +1,38 @@
+perf-evlist(1)
+==============
+
+NAME
+----
+perf-evlist - List the event names in a perf.data file
+
+SYNOPSIS
+--------
+[verse]
+'perf evlist <options>'
+
+DESCRIPTION
+-----------
+This command displays the names of events sampled in a perf.data file.
+
+OPTIONS
+-------
+-i::
+--input=::
+        Input file name. (default: perf.data unless stdin is a fifo)
+
+-F::
+--freq=::
+	Show just the sample frequency used for each event.
+
+-v::
+--verbose=::
+	Show all fields.
+
+-g::
+--group::
+	Show event group information.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-list[1],
+linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-help.txt b/tools/perf/Documentation/perf-help.txt
new file mode 100644
index 000000000..514391818
--- /dev/null
+++ b/tools/perf/Documentation/perf-help.txt
@@ -0,0 +1,38 @@
+perf-help(1)
+============
+
+NAME
+----
+perf-help - display help information about perf
+
+SYNOPSIS
+--------
+'perf help' [-a|--all] [COMMAND]
+
+DESCRIPTION
+-----------
+
+With no options and no COMMAND given, the synopsis of the 'perf'
+command and a list of the most commonly used perf commands are printed
+on the standard output.
+
+If the option '--all' or '-a' is given, then all available commands are
+printed on the standard output.
+
+If a perf command is named, a manual page for that command is brought
+up. The 'man' program is used by default for this purpose, but this
+can be overridden by other options or configuration variables.
+
+Note that `perf --help ...` is identical to `perf help ...` because the
+former is internally converted into the latter.
+
+OPTIONS
+-------
+-a::
+--all::
+	Prints all the available commands on the standard output. This
+	option supersedes any other option.
+
+PERF
+----
+Part of the linkperf:perf[1] suite
diff --git a/tools/perf/Documentation/perf-inject.txt b/tools/perf/Documentation/perf-inject.txt
new file mode 100644
index 000000000..dc7442cf3
--- /dev/null
+++ b/tools/perf/Documentation/perf-inject.txt
@@ -0,0 +1,49 @@
+perf-inject(1)
+==============
+
+NAME
+----
+perf-inject - Filter to augment the events stream with additional information
+
+SYNOPSIS
+--------
+[verse]
+'perf inject <options>'
+
+DESCRIPTION
+-----------
+perf-inject reads a perf-record event stream and repipes it to stdout.  At any
+point the processing code can inject other events into the event stream - in
+this case build-ids (-b option) are read and injected as needed into the event
+stream.
+
+Build-ids are just the first user of perf-inject - potentially anything that
+needs userspace processing to augment the events stream with additional
+information could make use of this facility.
+
+OPTIONS
+-------
+-b::
+--build-ids=::
+        Inject build-ids into the output stream
+-v::
+--verbose::
+	Be more verbose.
+-i::
+--input=::
+	Input file name. (default: stdin)
+-o::
+--output=::
+	Output file name. (default: stdout)
+-s::
+--sched-stat::
+	Merge sched_stat and sched_switch for getting events where and how long
+	tasks slept. sched_switch contains a callchain where a task slept and
+	sched_stat contains a timeslice how long a task slept.
+
+--kallsyms=<file>::
+	kallsyms pathname
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-report[1], linkperf:perf-archive[1]
diff --git a/tools/perf/Documentation/perf-kmem.txt b/tools/perf/Documentation/perf-kmem.txt
new file mode 100644
index 000000000..23219c65c
--- /dev/null
+++ b/tools/perf/Documentation/perf-kmem.txt
@@ -0,0 +1,57 @@
+perf-kmem(1)
+============
+
+NAME
+----
+perf-kmem - Tool to trace/measure kernel memory properties
+
+SYNOPSIS
+--------
+[verse]
+'perf kmem' {record|stat} [<options>]
+
+DESCRIPTION
+-----------
+There are two variants of perf kmem:
+
+  'perf kmem record <command>' to record the kmem events
+  of an arbitrary workload.
+
+  'perf kmem stat' to report kernel memory statistics.
+
+OPTIONS
+-------
+-i <file>::
+--input=<file>::
+	Select the input file (default: perf.data unless stdin is a fifo)
+
+-v::
+--verbose::
+        Be more verbose. (show symbol address, etc)
+
+--caller::
+	Show per-callsite statistics
+
+--alloc::
+	Show per-allocation statistics
+
+-s <key[,key2...]>::
+--sort=<key[,key2...]>::
+	Sort the output (default: frag,hit,bytes)
+
+-l <num>::
+--line=<num>::
+	Print n lines only
+
+--raw-ip::
+	Print raw ip instead of symbol
+
+--slab::
+	Analyze SLAB allocator events.
+
+--page::
+	Analyze page allocator events
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-kvm.txt b/tools/perf/Documentation/perf-kvm.txt
new file mode 100644
index 000000000..6252e7760
--- /dev/null
+++ b/tools/perf/Documentation/perf-kvm.txt
@@ -0,0 +1,158 @@
+perf-kvm(1)
+===========
+
+NAME
+----
+perf-kvm - Tool to trace/measure kvm guest os
+
+SYNOPSIS
+--------
+[verse]
+'perf kvm' [--host] [--guest] [--guestmount=<path>
+	[--guestkallsyms=<path> --guestmodules=<path> | --guestvmlinux=<path>]]
+	{top|record|report|diff|buildid-list} [<options>]
+'perf kvm' [--host] [--guest] [--guestkallsyms=<path> --guestmodules=<path>
+	| --guestvmlinux=<path>] {top|record|report|diff|buildid-list|stat} [<options>]
+'perf kvm stat [record|report|live] [<options>]
+
+DESCRIPTION
+-----------
+There are a couple of variants of perf kvm:
+
+  'perf kvm [options] top <command>' to generates and displays
+  a performance counter profile of guest os in realtime
+  of an arbitrary workload.
+
+  'perf kvm record <command>' to record the performance counter profile
+  of an arbitrary workload and save it into a perf data file. We set the
+  default behavior of perf kvm as --guest, so if neither --host nor --guest
+  is input, the perf data file name is perf.data.guest. If --host is input,
+  the perf data file name is perf.data.kvm. If you want to record data into
+  perf.data.host, please input --host --no-guest. The behaviors are shown as
+  following:
+    Default('')         ->  perf.data.guest
+    --host              ->  perf.data.kvm
+    --guest             ->  perf.data.guest
+    --host --guest      ->  perf.data.kvm
+    --host --no-guest   ->  perf.data.host
+
+  'perf kvm report' to display the performance counter profile information
+  recorded via perf kvm record.
+
+  'perf kvm diff' to displays the performance difference amongst two perf.data
+  files captured via perf record.
+
+  'perf kvm buildid-list' to  display the buildids found in a perf data file,
+  so that other tools can be used to fetch packages with matching symbol tables
+  for use by perf report. As buildid is read from /sys/kernel/notes in os, then
+  if you want to list the buildid for guest, please make sure your perf data file
+  was captured with --guestmount in perf kvm record.
+
+  'perf kvm stat <command>' to run a command and gather performance counter
+  statistics.
+  Especially, perf 'kvm stat record/report' generates a statistical analysis
+  of KVM events. Currently, vmexit, mmio (x86 only) and ioport (x86 only)
+  events are supported. 'perf kvm stat record <command>' records kvm events
+  and the events between start and end <command>.
+  And this command produces a file which contains tracing results of kvm
+  events.
+
+  'perf kvm stat report' reports statistical data which includes events
+  handled time, samples, and so on.
+
+  'perf kvm stat live' reports statistical data in a live mode (similar to
+  record + report but with statistical data updated live at a given display
+  rate).
+
+OPTIONS
+-------
+-i::
+--input=<path>::
+        Input file name.
+-o::
+--output=<path>::
+        Output file name.
+--host::
+        Collect host side performance profile.
+--guest::
+        Collect guest side performance profile.
+--guestmount=<path>::
+	Guest os root file system mount directory. Users mounts guest os
+        root directories under <path> by a specific filesystem access method,
+	typically, sshfs. For example, start 2 guest os. The one's pid is 8888
+	and the other's is 9999.
+        #mkdir ~/guestmount; cd ~/guestmount
+        #sshfs -o allow_other,direct_io -p 5551 localhost:/ 8888/
+        #sshfs -o allow_other,direct_io -p 5552 localhost:/ 9999/
+        #perf kvm --host --guest --guestmount=~/guestmount top
+--guestkallsyms=<path>::
+        Guest os /proc/kallsyms file copy. 'perf' kvm' reads it to get guest
+	kernel symbols. Users copy it out from guest os.
+--guestmodules=<path>::
+	Guest os /proc/modules file copy. 'perf' kvm' reads it to get guest
+	kernel module information. Users copy it out from guest os.
+--guestvmlinux=<path>::
+	Guest os kernel vmlinux.
+-v::
+--verbose::
+	Be more verbose (show counter open errors, etc).
+
+STAT REPORT OPTIONS
+-------------------
+--vcpu=<value>::
+       analyze events which occur on this vcpu. (default: all vcpus)
+
+--event=<value>::
+       event to be analyzed. Possible values: vmexit, mmio (x86 only),
+       ioport (x86 only). (default: vmexit)
+-k::
+--key=<value>::
+       Sorting key. Possible values: sample (default, sort by samples
+       number), time (sort by average time).
+-p::
+--pid=::
+    Analyze events only for given process ID(s) (comma separated list).
+
+STAT LIVE OPTIONS
+-----------------
+-d::
+--display::
+        Time in seconds between display updates
+
+-m::
+--mmap-pages=::
+    Number of mmap data pages (must be a power of two) or size
+    specification with appended unit character - B/K/M/G. The
+    size is rounded up to have nearest pages power of two value.
+
+-a::
+--all-cpus::
+        System-wide collection from all CPUs.
+
+-p::
+--pid=::
+    Analyze events only for given process ID(s) (comma separated list).
+
+--vcpu=<value>::
+       analyze events which occur on this vcpu. (default: all vcpus)
+
+
+--event=<value>::
+       event to be analyzed. Possible values: vmexit,
+       mmio (x86 only), ioport (x86 only).
+       (default: vmexit)
+
+-k::
+--key=<value>::
+       Sorting key. Possible values: sample (default, sort by samples
+       number), time (sort by average time).
+
+--duration=<value>::
+       Show events other than HLT (x86 only) or Wait state (s390 only)
+       that take longer than duration usecs.
+
+SEE ALSO
+--------
+linkperf:perf-top[1], linkperf:perf-record[1], linkperf:perf-report[1],
+linkperf:perf-diff[1], linkperf:perf-buildid-list[1],
+linkperf:perf-stat[1]
diff --git a/tools/perf/Documentation/perf-list.txt b/tools/perf/Documentation/perf-list.txt
new file mode 100644
index 000000000..bada8933f
--- /dev/null
+++ b/tools/perf/Documentation/perf-list.txt
@@ -0,0 +1,142 @@
+perf-list(1)
+============
+
+NAME
+----
+perf-list - List all symbolic event types
+
+SYNOPSIS
+--------
+[verse]
+'perf list' [hw|sw|cache|tracepoint|pmu|event_glob]
+
+DESCRIPTION
+-----------
+This command displays the symbolic event types which can be selected in the
+various perf commands with the -e option.
+
+[[EVENT_MODIFIERS]]
+EVENT MODIFIERS
+---------------
+
+Events can optionally have a modifier by appending a colon and one or
+more modifiers. Modifiers allow the user to restrict the events to be
+counted. The following modifiers exist:
+
+ u - user-space counting
+ k - kernel counting
+ h - hypervisor counting
+ I - non idle counting
+ G - guest counting (in KVM guests)
+ H - host counting (not in KVM guests)
+ p - precise level
+ S - read sample value (PERF_SAMPLE_READ)
+ D - pin the event to the PMU
+
+The 'p' modifier can be used for specifying how precise the instruction
+address should be. The 'p' modifier can be specified multiple times:
+
+ 0 - SAMPLE_IP can have arbitrary skid
+ 1 - SAMPLE_IP must have constant skid
+ 2 - SAMPLE_IP requested to have 0 skid
+ 3 - SAMPLE_IP must have 0 skid
+
+For Intel systems precise event sampling is implemented with PEBS
+which supports up to precise-level 2.
+
+On AMD systems it is implemented using IBS (up to precise-level 2).
+The precise modifier works with event types 0x76 (cpu-cycles, CPU
+clocks not halted) and 0xC1 (micro-ops retired). Both events map to
+IBS execution sampling (IBS op) with the IBS Op Counter Control bit
+(IbsOpCntCtl) set respectively (see AMD64 Architecture Programmer’s
+Manual Volume 2: System Programming, 13.3 Instruction-Based
+Sampling). Examples to use IBS:
+
+ perf record -a -e cpu-cycles:p ...    # use ibs op counting cycles
+ perf record -a -e r076:p ...          # same as -e cpu-cycles:p
+ perf record -a -e r0C1:p ...          # use ibs op counting micro-ops
+
+RAW HARDWARE EVENT DESCRIPTOR
+-----------------------------
+Even when an event is not available in a symbolic form within perf right now,
+it can be encoded in a per processor specific way.
+
+For instance For x86 CPUs NNN represents the raw register encoding with the
+layout of IA32_PERFEVTSELx MSRs (see [Intel® 64 and IA-32 Architectures Software Developer's Manual Volume 3B: System Programming Guide] Figure 30-1 Layout
+of IA32_PERFEVTSELx MSRs) or AMD's PerfEvtSeln (see [AMD64 Architecture Programmer’s Manual Volume 2: System Programming], Page 344,
+Figure 13-7 Performance Event-Select Register (PerfEvtSeln)).
+
+Note: Only the following bit fields can be set in x86 counter
+registers: event, umask, edge, inv, cmask. Esp. guest/host only and
+OS/user mode flags must be setup using <<EVENT_MODIFIERS, EVENT
+MODIFIERS>>.
+
+Example:
+
+If the Intel docs for a QM720 Core i7 describe an event as:
+
+  Event  Umask  Event Mask
+  Num.   Value  Mnemonic    Description                        Comment
+
+  A8H      01H  LSD.UOPS    Counts the number of micro-ops     Use cmask=1 and
+                            delivered by loop stream detector  invert to count
+                                                               cycles
+
+raw encoding of 0x1A8 can be used:
+
+ perf stat -e r1a8 -a sleep 1
+ perf record -e r1a8 ...
+
+You should refer to the processor specific documentation for getting these
+details. Some of them are referenced in the SEE ALSO section below.
+
+PARAMETERIZED EVENTS
+--------------------
+
+Some pmu events listed by 'perf-list' will be displayed with '?' in them. For
+example:
+
+  hv_gpci/dtbp_ptitc,phys_processor_idx=?/
+
+This means that when provided as an event, a value for '?' must
+also be supplied. For example:
+
+  perf stat -C 0 -e 'hv_gpci/dtbp_ptitc,phys_processor_idx=0x2/' ...
+
+OPTIONS
+-------
+
+Without options all known events will be listed.
+
+To limit the list use:
+
+. 'hw' or 'hardware' to list hardware events such as cache-misses, etc.
+
+. 'sw' or 'software' to list software events such as context switches, etc.
+
+. 'cache' or 'hwcache' to list hardware cache events such as L1-dcache-loads, etc.
+
+. 'tracepoint' to list all tracepoint events, alternatively use
+  'subsys_glob:event_glob' to filter by tracepoint subsystems such as sched,
+  block, etc.
+
+. 'pmu' to print the kernel supplied PMU events.
+
+. If none of the above is matched, it will apply the supplied glob to all
+  events, printing the ones that match.
+
+One or more types can be used at the same time, listing the events for the
+types specified.
+
+Support raw format:
+
+. '--raw-dump', shows the raw-dump of all the events.
+. '--raw-dump [hw|sw|cache|tracepoint|pmu|event_glob]', shows the raw-dump of
+  a certain kind of events.
+
+SEE ALSO
+--------
+linkperf:perf-stat[1], linkperf:perf-top[1],
+linkperf:perf-record[1],
+http://www.intel.com/Assets/PDF/manual/253669.pdf[Intel® 64 and IA-32 Architectures Software Developer's Manual Volume 3B: System Programming Guide],
+http://support.amd.com/us/Processor_TechDocs/24593_APM_v2.pdf[AMD64 Architecture Programmer’s Manual Volume 2: System Programming]
diff --git a/tools/perf/Documentation/perf-lock.txt b/tools/perf/Documentation/perf-lock.txt
new file mode 100644
index 000000000..ab25be28c
--- /dev/null
+++ b/tools/perf/Documentation/perf-lock.txt
@@ -0,0 +1,66 @@
+perf-lock(1)
+============
+
+NAME
+----
+perf-lock - Analyze lock events
+
+SYNOPSIS
+--------
+[verse]
+'perf lock' {record|report|script|info}
+
+DESCRIPTION
+-----------
+You can analyze various lock behaviours
+and statistics with this 'perf lock' command.
+
+  'perf lock record <command>' records lock events
+  between start and end <command>. And this command
+  produces the file "perf.data" which contains tracing
+  results of lock events.
+
+  'perf lock report' reports statistical data.
+
+  'perf lock script' shows raw lock events.
+
+  'perf lock info' shows metadata like threads or addresses
+  of lock instances.
+
+COMMON OPTIONS
+--------------
+
+-i::
+--input=<file>::
+        Input file name. (default: perf.data unless stdin is a fifo)
+
+-v::
+--verbose::
+        Be more verbose (show symbol address, etc).
+
+-D::
+--dump-raw-trace::
+        Dump raw trace in ASCII.
+
+REPORT OPTIONS
+--------------
+
+-k::
+--key=<value>::
+        Sorting key. Possible values: acquired (default), contended,
+	avg_wait, wait_total, wait_max, wait_min.
+
+INFO OPTIONS
+------------
+
+-t::
+--threads::
+	dump thread list in perf.data
+
+-m::
+--map::
+	dump map of lock instances (address:name table)
+
+SEE ALSO
+--------
+linkperf:perf[1]
diff --git a/tools/perf/Documentation/perf-mem.txt b/tools/perf/Documentation/perf-mem.txt
new file mode 100644
index 000000000..43310d866
--- /dev/null
+++ b/tools/perf/Documentation/perf-mem.txt
@@ -0,0 +1,53 @@
+perf-mem(1)
+===========
+
+NAME
+----
+perf-mem - Profile memory accesses
+
+SYNOPSIS
+--------
+[verse]
+'perf mem' [<options>] (record [<command>] | report)
+
+DESCRIPTION
+-----------
+"perf mem record" runs a command and gathers memory operation data
+from it, into perf.data. Perf record options are accepted and are passed through.
+
+"perf mem report" displays the result. It invokes perf report with the
+right set of options to display a memory access profile. By default, loads
+and stores are sampled. Use the -t option to limit to loads or stores.
+
+Note that on Intel systems the memory latency reported is the use-latency,
+not the pure load (or store latency). Use latency includes any pipeline
+queueing delays in addition to the memory subsystem latency.
+
+OPTIONS
+-------
+<command>...::
+	Any command you can specify in a shell.
+
+-t::
+--type=::
+	Select the memory operation type: load or store (default: load,store)
+
+-D::
+--dump-raw-samples=::
+	Dump the raw decoded samples on the screen in a format that is easy to parse with
+	one sample per line.
+
+-x::
+--field-separator::
+	Specify the field separator used when dump raw samples (-D option). By default,
+	The separator is the space character.
+
+-C::
+--cpu-list::
+	Restrict dump of raw samples to those provided via this option. Note that the same
+	option can be passed in record mode. It will be interpreted the same way as perf
+	record.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt
new file mode 100644
index 000000000..239609c09
--- /dev/null
+++ b/tools/perf/Documentation/perf-probe.txt
@@ -0,0 +1,222 @@
+perf-probe(1)
+=============
+
+NAME
+----
+perf-probe - Define new dynamic tracepoints
+
+SYNOPSIS
+--------
+[verse]
+'perf probe' [options] --add='PROBE' [...]
+or
+'perf probe' [options] PROBE
+or
+'perf probe' [options] --del='[GROUP:]EVENT' [...]
+or
+'perf probe' --list
+or
+'perf probe' [options] --line='LINE'
+or
+'perf probe' [options] --vars='PROBEPOINT'
+
+DESCRIPTION
+-----------
+This command defines dynamic tracepoint events, by symbol and registers
+without debuginfo, or by C expressions (C line numbers, C function names,
+and C local variables) with debuginfo.
+
+
+OPTIONS
+-------
+-k::
+--vmlinux=PATH::
+	Specify vmlinux path which has debuginfo (Dwarf binary).
+
+-m::
+--module=MODNAME|PATH::
+	Specify module name in which perf-probe searches probe points
+	or lines. If a path of module file is passed, perf-probe
+	treat it as an offline module (this means you can add a probe on
+        a module which has not been loaded yet).
+
+-s::
+--source=PATH::
+	Specify path to kernel source.
+
+-v::
+--verbose::
+        Be more verbose (show parsed arguments, etc).
+	Can not use with -q.
+
+-q::
+--quiet::
+	Be quiet (do not show any messages including errors).
+	Can not use with -v.
+
+-a::
+--add=::
+	Define a probe event (see PROBE SYNTAX for detail).
+
+-d::
+--del=::
+	Delete probe events. This accepts glob wildcards('*', '?') and character
+	classes(e.g. [a-z], [!A-Z]).
+
+-l::
+--list::
+	List up current probe events.
+
+-L::
+--line=::
+	Show source code lines which can be probed. This needs an argument
+	which specifies a range of the source code. (see LINE SYNTAX for detail)
+
+-V::
+--vars=::
+	Show available local variables at given probe point. The argument
+	syntax is same as PROBE SYNTAX, but NO ARGs.
+
+--externs::
+	(Only for --vars) Show external defined variables in addition to local
+	variables.
+
+-F::
+--funcs::
+	Show available functions in given module or kernel. With -x/--exec,
+	can also list functions in a user space executable / shared library.
+
+--filter=FILTER::
+	(Only for --vars and --funcs) Set filter. FILTER is a combination of glob
+	pattern, see FILTER PATTERN for detail.
+	Default FILTER is "!__k???tab_* & !__crc_*" for --vars, and "!_*"
+	for --funcs.
+	If several filters are specified, only the last filter is used.
+
+-f::
+--force::
+	Forcibly add events with existing name.
+
+-n::
+--dry-run::
+	Dry run. With this option, --add and --del doesn't execute actual
+	adding and removal operations.
+
+--max-probes=NUM::
+	Set the maximum number of probe points for an event. Default is 128.
+
+-x::
+--exec=PATH::
+	Specify path to the executable or shared library file for user
+	space tracing. Can also be used with --funcs option.
+
+--demangle::
+	Demangle application symbols. --no-demangle is also available
+	for disabling demangling.
+
+--demangle-kernel::
+	Demangle kernel symbols. --no-demangle-kernel is also available
+	for disabling kernel demangling.
+
+In absence of -m/-x options, perf probe checks if the first argument after
+the options is an absolute path name. If its an absolute path, perf probe
+uses it as a target module/target user space binary to probe.
+
+PROBE SYNTAX
+------------
+Probe points are defined by following syntax.
+
+    1) Define event based on function name
+     [EVENT=]FUNC[@SRC][:RLN|+OFFS|%return|;PTN] [ARG ...]
+
+    2) Define event based on source file with line number
+     [EVENT=]SRC:ALN [ARG ...]
+
+    3) Define event based on source file with lazy pattern
+     [EVENT=]SRC;PTN [ARG ...]
+
+
+'EVENT' specifies the name of new event, if omitted, it will be set the name of the probed function. Currently, event group name is set as 'probe'.
+'FUNC' specifies a probed function name, and it may have one of the following options; '+OFFS' is the offset from function entry address in bytes, ':RLN' is the relative-line number from function entry line, and '%return' means that it probes function return. And ';PTN' means lazy matching pattern (see LAZY MATCHING). Note that ';PTN' must be the end of the probe point definition.  In addition, '@SRC' specifies a source file which has that function.
+It is also possible to specify a probe point by the source line number or lazy matching by using 'SRC:ALN' or 'SRC;PTN' syntax, where 'SRC' is the source file path, ':ALN' is the line number and ';PTN' is the lazy matching pattern.
+'ARG' specifies the arguments of this probe point, (see PROBE ARGUMENT).
+
+PROBE ARGUMENT
+--------------
+Each probe argument follows below syntax.
+
+ [NAME=]LOCALVAR|$retval|%REG|@SYMBOL[:TYPE]
+
+'NAME' specifies the name of this argument (optional). You can use the name of local variable, local data structure member (e.g. var->field, var.field2), local array with fixed index (e.g. array[1], var->array[0], var->pointer[2]), or kprobe-tracer argument format (e.g. $retval, %ax, etc). Note that the name of this argument will be set as the last member name if you specify a local data structure member (e.g. field2 for 'var->field1.field2'.)
+'$vars' special argument is also available for NAME, it is expanded to the local variables which can access at given probe point.
+'TYPE' casts the type of this argument (optional). If omitted, perf probe automatically set the type based on debuginfo. You can specify 'string' type only for the local variable or structure member which is an array of or a pointer to 'char' or 'unsigned char' type.
+
+On x86 systems %REG is always the short form of the register: for example %AX. %RAX or %EAX is not valid.
+
+LINE SYNTAX
+-----------
+Line range is described by following syntax.
+
+ "FUNC[@SRC][:RLN[+NUM|-RLN2]]|SRC[:ALN[+NUM|-ALN2]]"
+
+FUNC specifies the function name of showing lines. 'RLN' is the start line
+number from function entry line, and 'RLN2' is the end line number. As same as
+probe syntax, 'SRC' means the source file path, 'ALN' is start line number,
+and 'ALN2' is end line number in the file. It is also possible to specify how
+many lines to show by using 'NUM'. Moreover, 'FUNC@SRC' combination is good
+for searching a specific function when several functions share same name.
+So, "source.c:100-120" shows lines between 100th to l20th in source.c file. And "func:10+20" shows 20 lines from 10th line of func function.
+
+LAZY MATCHING
+-------------
+ The lazy line matching is similar to glob matching but ignoring spaces in both of pattern and target. So this accepts wildcards('*', '?') and character classes(e.g. [a-z], [!A-Z]).
+
+e.g.
+ 'a=*' can matches 'a=b', 'a = b', 'a == b' and so on.
+
+This provides some sort of flexibility and robustness to probe point definitions against minor code changes. For example, actual 10th line of schedule() can be moved easily by modifying schedule(), but the same line matching 'rq=cpu_rq*' may still exist in the function.)
+
+FILTER PATTERN
+--------------
+ The filter pattern is a glob matching pattern(s) to filter variables.
+ In addition, you can use "!" for specifying filter-out rule. You also can give several rules combined with "&" or "|", and fold those rules as one rule by using "(" ")".
+
+e.g.
+ With --filter "foo* | bar*", perf probe -V shows variables which start with "foo" or "bar".
+ With --filter "!foo* & *bar", perf probe -V shows variables which don't start with "foo" and end with "bar", like "fizzbar". But "foobar" is filtered out.
+
+EXAMPLES
+--------
+Display which lines in schedule() can be probed:
+
+ ./perf probe --line schedule
+
+Add a probe on schedule() function 12th line with recording cpu local variable:
+
+ ./perf probe schedule:12 cpu
+ or
+ ./perf probe --add='schedule:12 cpu'
+
+ this will add one or more probes which has the name start with "schedule".
+
+ Add probes on lines in schedule() function which calls update_rq_clock().
+
+ ./perf probe 'schedule;update_rq_clock*'
+ or
+ ./perf probe --add='schedule;update_rq_clock*'
+
+Delete all probes on schedule().
+
+ ./perf probe --del='schedule*'
+
+Add probes at zfree() function on /bin/zsh
+
+ ./perf probe -x /bin/zsh zfree or ./perf probe /bin/zsh zfree
+
+Add probes at malloc() function on libc
+
+ ./perf probe -x /lib/libc.so.6 malloc or ./perf probe /lib/libc.so.6 malloc
+
+SEE ALSO
+--------
+linkperf:perf-trace[1], linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
new file mode 100644
index 000000000..4847a793d
--- /dev/null
+++ b/tools/perf/Documentation/perf-record.txt
@@ -0,0 +1,262 @@
+perf-record(1)
+==============
+
+NAME
+----
+perf-record - Run a command and record its profile into perf.data
+
+SYNOPSIS
+--------
+[verse]
+'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] <command>
+'perf record' [-e <EVENT> | --event=EVENT] [-l] [-a] -- <command> [<options>]
+
+DESCRIPTION
+-----------
+This command runs a command and gathers a performance counter profile
+from it, into perf.data - without displaying anything.
+
+This file can then be inspected later on, using 'perf report'.
+
+
+OPTIONS
+-------
+<command>...::
+	Any command you can specify in a shell.
+
+-e::
+--event=::
+	Select the PMU event. Selection can be:
+
+        - a symbolic event name	(use 'perf list' to list all events)
+
+        - a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a
+	  hexadecimal event descriptor.
+
+	- a symbolically formed PMU event like 'pmu/param1=0x3,param2/' where
+	  'param1', 'param2', etc are defined as formats for the PMU in
+	  /sys/bus/event_sources/devices/<pmu>/format/*.
+
+	- a symbolically formed event like 'pmu/config=M,config1=N,config3=K/'
+
+          where M, N, K are numbers (in decimal, hex, octal format). Acceptable
+          values for each of 'config', 'config1' and 'config2' are defined by
+          corresponding entries in /sys/bus/event_sources/devices/<pmu>/format/*
+          param1 and param2 are defined as formats for the PMU in:
+          /sys/bus/event_sources/devices/<pmu>/format/*
+
+        - a hardware breakpoint event in the form of '\mem:addr[/len][:access]'
+          where addr is the address in memory you want to break in.
+          Access is the memory access type (read, write, execute) it can
+          be passed as follows: '\mem:addr[:[r][w][x]]'. len is the range,
+          number of bytes from specified addr, which the breakpoint will cover.
+          If you want to profile read-write accesses in 0x1000, just set
+          'mem:0x1000:rw'.
+          If you want to profile write accesses in [0x1000~1008), just set
+          'mem:0x1000/8:w'.
+
+	- a group of events surrounded by a pair of brace ("{event1,event2,...}").
+	  Each event is separated by commas and the group should be quoted to
+	  prevent the shell interpretation.  You also need to use --group on
+	  "perf report" to view group events together.
+
+--filter=<filter>::
+        Event filter.
+
+-a::
+--all-cpus::
+        System-wide collection from all CPUs.
+
+-p::
+--pid=::
+	Record events on existing process ID (comma separated list).
+
+-t::
+--tid=::
+        Record events on existing thread ID (comma separated list).
+        This option also disables inheritance by default.  Enable it by adding
+        --inherit.
+
+-u::
+--uid=::
+        Record events in threads owned by uid. Name or number.
+
+-r::
+--realtime=::
+	Collect data with this RT SCHED_FIFO priority.
+
+--no-buffering::
+	Collect data without buffering.
+
+-c::
+--count=::
+	Event period to sample.
+
+-o::
+--output=::
+	Output file name.
+
+-i::
+--no-inherit::
+	Child tasks do not inherit counters.
+-F::
+--freq=::
+	Profile at this frequency.
+
+-m::
+--mmap-pages=::
+	Number of mmap data pages (must be a power of two) or size
+	specification with appended unit character - B/K/M/G. The
+	size is rounded up to have nearest pages power of two value.
+
+--group::
+	Put all events in a single event group.  This precedes the --event
+	option and remains only for backward compatibility.  See --event.
+
+-g::
+	Enables call-graph (stack chain/backtrace) recording.
+
+--call-graph::
+	Setup and enable call-graph (stack chain/backtrace) recording,
+	implies -g.
+
+	Allows specifying "fp" (frame pointer) or "dwarf"
+	(DWARF's CFI - Call Frame Information) or "lbr"
+	(Hardware Last Branch Record facility) as the method to collect
+	the information used to show the call graphs.
+
+	In some systems, where binaries are build with gcc
+	--fomit-frame-pointer, using the "fp" method will produce bogus
+	call graphs, using "dwarf", if available (perf tools linked to
+	the libunwind library) should be used instead.
+	Using the "lbr" method doesn't require any compiler options. It
+	will produce call graphs from the hardware LBR registers. The
+	main limition is that it is only available on new Intel
+	platforms, such as Haswell. It can only get user call chain. It
+	doesn't work with branch stack sampling at the same time.
+
+-q::
+--quiet::
+	Don't print any message, useful for scripting.
+
+-v::
+--verbose::
+	Be more verbose (show counter open errors, etc).
+
+-s::
+--stat::
+	Per thread counts.
+
+-d::
+--data::
+	Sample addresses.
+
+-T::
+--timestamp::
+	Sample timestamps. Use it with 'perf report -D' to see the timestamps,
+	for instance.
+
+-n::
+--no-samples::
+	Don't sample.
+
+-R::
+--raw-samples::
+Collect raw sample records from all opened counters (default for tracepoint counters).
+
+-C::
+--cpu::
+Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
+comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
+In per-thread mode with inheritance mode on (default), samples are captured only when
+the thread executes on the designated CPUs. Default is to monitor all CPUs.
+
+-N::
+--no-buildid-cache::
+Do not update the buildid cache. This saves some overhead in situations
+where the information in the perf.data file (which includes buildids)
+is sufficient.
+
+-G name,...::
+--cgroup name,...::
+monitor only in the container (cgroup) called "name". This option is available only
+in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
+container "name" are monitored when they run on the monitored CPUs. Multiple cgroups
+can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup
+to first event, second cgroup to second event and so on. It is possible to provide
+an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
+corresponding events, i.e., they always refer to events defined earlier on the command
+line.
+
+-b::
+--branch-any::
+Enable taken branch stack sampling. Any type of taken branch may be sampled.
+This is a shortcut for --branch-filter any. See --branch-filter for more infos.
+
+-j::
+--branch-filter::
+Enable taken branch stack sampling. Each sample captures a series of consecutive
+taken branches. The number of branches captured with each sample depends on the
+underlying hardware, the type of branches of interest, and the executed code.
+It is possible to select the types of branches captured by enabling filters. The
+following filters are defined:
+
+        - any:  any type of branches
+        - any_call: any function call or system call
+        - any_ret: any function return or system call return
+        - ind_call: any indirect branch
+        - u:  only when the branch target is at the user level
+        - k: only when the branch target is in the kernel
+        - hv: only when the target is at the hypervisor level
+	- in_tx: only when the target is in a hardware transaction
+	- no_tx: only when the target is not in a hardware transaction
+	- abort_tx: only when the target is a hardware transaction abort
+	- cond: conditional branches
+
++
+The option requires at least one branch type among any, any_call, any_ret, ind_call, cond.
+The privilege levels may be omitted, in which case, the privilege levels of the associated
+event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege
+levels are subject to permissions.  When sampling on multiple events, branch stack sampling
+is enabled for all the sampling events. The sampled branch type is the same for all events.
+The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k
+Note that this feature may not be available on all processors.
+
+--weight::
+Enable weightened sampling. An additional weight is recorded per sample and can be
+displayed with the weight and local_weight sort keys.  This currently works for TSX
+abort events and some memory events in precise mode on modern Intel CPUs.
+
+--transaction::
+Record transaction flags for transaction related events.
+
+--per-thread::
+Use per-thread mmaps.  By default per-cpu mmaps are created.  This option
+overrides that and uses per-thread mmaps.  A side-effect of that is that
+inheritance is automatically disabled.  --per-thread is ignored with a warning
+if combined with -a or -C options.
+
+-D::
+--delay=::
+After starting the program, wait msecs before measuring. This is useful to
+filter out the startup phase of the program, which is often very different.
+
+-I::
+--intr-regs::
+Capture machine state (registers) at interrupt, i.e., on counter overflows for
+each sample. List of captured registers depends on the architecture. This option
+is off by default.
+
+--running-time::
+Record running and enabled time for read events (:S)
+
+-k::
+--clockid::
+Sets the clock id to use for the various time fields in the perf_event_type
+records. See clock_gettime(). In particular CLOCK_MONOTONIC and
+CLOCK_MONOTONIC_RAW are supported, some events might also allow
+CLOCK_BOOTTIME, CLOCK_REALTIME and CLOCK_TAI.
+
+SEE ALSO
+--------
+linkperf:perf-stat[1], linkperf:perf-list[1]
diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
new file mode 100644
index 000000000..4879cf638
--- /dev/null
+++ b/tools/perf/Documentation/perf-report.txt
@@ -0,0 +1,328 @@
+perf-report(1)
+==============
+
+NAME
+----
+perf-report - Read perf.data (created by perf record) and display the profile
+
+SYNOPSIS
+--------
+[verse]
+'perf report' [-i <file> | --input=file]
+
+DESCRIPTION
+-----------
+This command displays the performance counter profile information recorded
+via perf record.
+
+OPTIONS
+-------
+-i::
+--input=::
+        Input file name. (default: perf.data unless stdin is a fifo)
+
+-v::
+--verbose::
+        Be more verbose. (show symbol address, etc)
+
+-n::
+--show-nr-samples::
+	Show the number of samples for each symbol
+
+--showcpuutilization::
+        Show sample percentage for different cpu modes.
+
+-T::
+--threads::
+	Show per-thread event counters
+-c::
+--comms=::
+	Only consider symbols in these comms. CSV that understands
+	file://filename entries.  This option will affect the percentage of
+	the overhead column.  See --percentage for more info.
+--pid=::
+        Only show events for given process ID (comma separated list).
+
+--tid=::
+        Only show events for given thread ID (comma separated list).
+-d::
+--dsos=::
+	Only consider symbols in these dsos. CSV that understands
+	file://filename entries.  This option will affect the percentage of
+	the overhead column.  See --percentage for more info.
+-S::
+--symbols=::
+	Only consider these symbols. CSV that understands
+	file://filename entries.  This option will affect the percentage of
+	the overhead column.  See --percentage for more info.
+
+--symbol-filter=::
+	Only show symbols that match (partially) with this filter.
+
+-U::
+--hide-unresolved::
+        Only display entries resolved to a symbol.
+
+-s::
+--sort=::
+	Sort histogram entries by given key(s) - multiple keys can be specified
+	in CSV format.  Following sort keys are available:
+	pid, comm, dso, symbol, parent, cpu, srcline, weight, local_weight.
+
+	Each key has following meaning:
+
+	- comm: command (name) of the task which can be read via /proc/<pid>/comm
+	- pid: command and tid of the task
+	- dso: name of library or module executed at the time of sample
+	- symbol: name of function executed at the time of sample
+	- parent: name of function matched to the parent regex filter. Unmatched
+	entries are displayed as "[other]".
+	- cpu: cpu number the task ran at the time of sample
+	- srcline: filename and line number executed at the time of sample.  The
+	DWARF debugging info must be provided.
+	- weight: Event specific weight, e.g. memory latency or transaction
+	abort cost. This is the global weight.
+	- local_weight: Local weight version of the weight above.
+	- transaction: Transaction abort flags.
+	- overhead: Overhead percentage of sample
+	- overhead_sys: Overhead percentage of sample running in system mode
+	- overhead_us: Overhead percentage of sample running in user mode
+	- overhead_guest_sys: Overhead percentage of sample running in system mode
+	on guest machine
+	- overhead_guest_us: Overhead percentage of sample running in user mode on
+	guest machine
+	- sample: Number of sample
+	- period: Raw number of event count of sample
+
+	By default, comm, dso and symbol keys are used.
+	(i.e. --sort comm,dso,symbol)
+
+	If --branch-stack option is used, following sort keys are also
+	available:
+	dso_from, dso_to, symbol_from, symbol_to, mispredict.
+
+	- dso_from: name of library or module branched from
+	- dso_to: name of library or module branched to
+	- symbol_from: name of function branched from
+	- symbol_to: name of function branched to
+	- mispredict: "N" for predicted branch, "Y" for mispredicted branch
+	- in_tx: branch in TSX transaction
+	- abort: TSX transaction abort.
+
+	And default sort keys are changed to comm, dso_from, symbol_from, dso_to
+	and symbol_to, see '--branch-stack'.
+
+-F::
+--fields=::
+	Specify output field - multiple keys can be specified in CSV format.
+	Following fields are available:
+	overhead, overhead_sys, overhead_us, overhead_children, sample and period.
+	Also it can contain any sort key(s).
+
+	By default, every sort keys not specified in -F will be appended
+	automatically.
+
+	If --mem-mode option is used, following sort keys are also available
+	(incompatible with --branch-stack):
+	symbol_daddr, dso_daddr, locked, tlb, mem, snoop, dcacheline.
+
+	- symbol_daddr: name of data symbol being executed on at the time of sample
+	- dso_daddr: name of library or module containing the data being executed
+	on at the time of sample
+	- locked: whether the bus was locked at the time of sample
+	- tlb: type of tlb access for the data at the time of sample
+	- mem: type of memory access for the data at the time of sample
+	- snoop: type of snoop (if any) for the data at the time of sample
+	- dcacheline: the cacheline the data address is on at the time of sample
+
+	And default sort keys are changed to local_weight, mem, sym, dso,
+	symbol_daddr, dso_daddr, snoop, tlb, locked, see '--mem-mode'.
+
+-p::
+--parent=<regex>::
+        A regex filter to identify parent. The parent is a caller of this
+	function and searched through the callchain, thus it requires callchain
+	information recorded. The pattern is in the exteneded regex format and
+	defaults to "\^sys_|^do_page_fault", see '--sort parent'.
+
+-x::
+--exclude-other::
+        Only display entries with parent-match.
+
+-w::
+--column-widths=<width[,width...]>::
+	Force each column width to the provided list, for large terminal
+	readability.  0 means no limit (default behavior).
+
+-t::
+--field-separator=::
+	Use a special separator character and don't pad with spaces, replacing
+	all occurrences of this separator in symbol names (and other output)
+	with a '.' character, that thus it's the only non valid separator.
+
+-D::
+--dump-raw-trace::
+        Dump raw trace in ASCII.
+
+-g [type,min[,limit],order[,key][,branch]]::
+--call-graph::
+        Display call chains using type, min percent threshold, optional print
+	limit and order.
+	type can be either:
+	- flat: single column, linear exposure of call chains.
+	- graph: use a graph tree, displaying absolute overhead rates.
+	- fractal: like graph, but displays relative rates. Each branch of
+		 the tree is considered as a new profiled object. +
+
+	order can be either:
+	- callee: callee based call graph.
+	- caller: inverted caller based call graph.
+
+	key can be:
+	- function: compare on functions
+	- address: compare on individual code addresses
+
+	branch can be:
+	- branch: include last branch information in callgraph
+	when available. Usually more convenient to use --branch-history
+	for this.
+
+	Default: fractal,0.5,callee,function.
+
+--children::
+	Accumulate callchain of children to parent entry so that then can
+	show up in the output.  The output will have a new "Children" column
+	and will be sorted on the data.  It requires callchains are recorded.
+
+--max-stack::
+	Set the stack depth limit when parsing the callchain, anything
+	beyond the specified depth will be ignored. This is a trade-off
+	between information loss and faster processing especially for
+	workloads that can have a very long callchain stack.
+
+	Default: 127
+
+-G::
+--inverted::
+        alias for inverted caller based call graph.
+
+--ignore-callees=<regex>::
+        Ignore callees of the function(s) matching the given regex.
+        This has the effect of collecting the callers of each such
+        function into one place in the call-graph tree.
+
+--pretty=<key>::
+        Pretty printing style.  key: normal, raw
+
+--stdio:: Use the stdio interface.
+
+--tui:: Use the TUI interface, that is integrated with annotate and allows
+        zooming into DSOs or threads, among other features. Use of --tui
+	requires a tty, if one is not present, as when piping to other
+	commands, the stdio interface is used.
+
+--gtk:: Use the GTK2 interface.
+
+-k::
+--vmlinux=<file>::
+        vmlinux pathname
+
+--kallsyms=<file>::
+        kallsyms pathname
+
+-m::
+--modules::
+        Load module symbols. WARNING: This should only be used with -k and
+        a LIVE kernel.
+
+-f::
+--force::
+        Don't complain, do it.
+
+--symfs=<directory>::
+        Look for files with symbols relative to this directory.
+
+-C::
+--cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can
+	be provided as a comma-separated list with no space: 0,1. Ranges of
+	CPUs are specified with -: 0-2. Default is to report samples on all
+	CPUs.
+
+-M::
+--disassembler-style=:: Set disassembler style for objdump.
+
+--source::
+	Interleave source code with assembly code. Enabled by default,
+	disable with --no-source.
+
+--asm-raw::
+	Show raw instruction encoding of assembly instructions.
+
+--show-total-period:: Show a column with the sum of periods.
+
+-I::
+--show-info::
+	Display extended information about the perf.data file. This adds
+	information which may be very large and thus may clutter the display.
+	It currently includes: cpu and numa topology of the host system.
+
+-b::
+--branch-stack::
+	Use the addresses of sampled taken branches instead of the instruction
+	address to build the histograms. To generate meaningful output, the
+	perf.data file must have been obtained using perf record -b or
+	perf record --branch-filter xxx where xxx is a branch filter option.
+	perf report is able to auto-detect whether a perf.data file contains
+	branch stacks and it will automatically switch to the branch view mode,
+	unless --no-branch-stack is used.
+
+--branch-history::
+	Add the addresses of sampled taken branches to the callstack.
+	This allows to examine the path the program took to each sample.
+	The data collection must have used -b (or -j) and -g.
+
+--objdump=<path>::
+        Path to objdump binary.
+
+--group::
+	Show event group information together.
+
+--demangle::
+	Demangle symbol names to human readable form. It's enabled by default,
+	disable with --no-demangle.
+
+--demangle-kernel::
+	Demangle kernel symbol names to human readable form (for C++ kernels).
+
+--mem-mode::
+	Use the data addresses of samples in addition to instruction addresses
+	to build the histograms.  To generate meaningful output, the perf.data
+	file must have been obtained using perf record -d -W and using a
+	special event -e cpu/mem-loads/ or -e cpu/mem-stores/. See
+	'perf mem' for simpler access.
+
+--percent-limit::
+	Do not show entries which have an overhead under that percent.
+	(Default: 0).
+
+--percentage::
+	Determine how to display the overhead percentage of filtered entries.
+	Filters can be applied by --comms, --dsos and/or --symbols options and
+	Zoom operations on the TUI (thread, dso, etc).
+
+	"relative" means it's relative to filtered entries only so that the
+	sum of shown entries will be always 100%.  "absolute" means it retains
+	the original value before and after the filter is applied.
+
+--header::
+	Show header information in the perf.data file.  This includes
+	various information like hostname, OS and perf version, cpu/mem
+	info, perf command line, event list and so on.  Currently only
+	--stdio output supports this feature.
+
+--header-only::
+	Show only perf.data header (forces --stdio).
+
+SEE ALSO
+--------
+linkperf:perf-stat[1], linkperf:perf-annotate[1]
diff --git a/tools/perf/Documentation/perf-sched.txt b/tools/perf/Documentation/perf-sched.txt
new file mode 100644
index 000000000..8ff4df956
--- /dev/null
+++ b/tools/perf/Documentation/perf-sched.txt
@@ -0,0 +1,55 @@
+perf-sched(1)
+==============
+
+NAME
+----
+perf-sched - Tool to trace/measure scheduler properties (latencies)
+
+SYNOPSIS
+--------
+[verse]
+'perf sched' {record|latency|map|replay|script}
+
+DESCRIPTION
+-----------
+There are five variants of perf sched:
+
+  'perf sched record <command>' to record the scheduling events
+  of an arbitrary workload.
+
+  'perf sched latency' to report the per task scheduling latencies
+  and other scheduling properties of the workload.
+
+  'perf sched script' to see a detailed trace of the workload that
+   was recorded (aliased to 'perf script' for now).
+
+  'perf sched replay' to simulate the workload that was recorded
+  via perf sched record. (this is done by starting up mockup threads
+  that mimic the workload based on the events in the trace. These
+  threads can then replay the timings (CPU runtime and sleep patterns)
+  of the workload as it occurred when it was recorded - and can repeat
+  it a number of times, measuring its performance.)
+
+  'perf sched map' to print a textual context-switching outline of
+  workload captured via perf sched record.  Columns stand for
+  individual CPUs, and the two-letter shortcuts stand for tasks that
+  are running on a CPU. A '*' denotes the CPU that had the event, and
+  a dot signals an idle CPU.
+
+OPTIONS
+-------
+-i::
+--input=<file>::
+        Input file name. (default: perf.data unless stdin is a fifo)
+
+-v::
+--verbose::
+        Be more verbose. (show symbol address, etc)
+
+-D::
+--dump-raw-trace=::
+        Display verbose dump of the sched data.
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-script-perl.txt b/tools/perf/Documentation/perf-script-perl.txt
new file mode 100644
index 000000000..dfbb506d2
--- /dev/null
+++ b/tools/perf/Documentation/perf-script-perl.txt
@@ -0,0 +1,216 @@
+perf-script-perl(1)
+==================
+
+NAME
+----
+perf-script-perl - Process trace data with a Perl script
+
+SYNOPSIS
+--------
+[verse]
+'perf script' [-s [Perl]:script[.pl] ]
+
+DESCRIPTION
+-----------
+
+This perf script option is used to process perf script data using perf's
+built-in Perl interpreter.  It reads and processes the input file and
+displays the results of the trace analysis implemented in the given
+Perl script, if any.
+
+STARTER SCRIPTS
+---------------
+
+You can avoid reading the rest of this document by running 'perf script
+-g perl' in the same directory as an existing perf.data trace file.
+That will generate a starter script containing a handler for each of
+the event types in the trace file; it simply prints every available
+field for each event in the trace file.
+
+You can also look at the existing scripts in
+~/libexec/perf-core/scripts/perl for typical examples showing how to
+do basic things like aggregate event data, print results, etc.  Also,
+the check-perf-script.pl script, while not interesting for its results,
+attempts to exercise all of the main scripting features.
+
+EVENT HANDLERS
+--------------
+
+When perf script is invoked using a trace script, a user-defined
+'handler function' is called for each event in the trace.  If there's
+no handler function defined for a given event type, the event is
+ignored (or passed to a 'trace_handled' function, see below) and the
+next event is processed.
+
+Most of the event's field values are passed as arguments to the
+handler function; some of the less common ones aren't - those are
+available as calls back into the perf executable (see below).
+
+As an example, the following perf record command can be used to record
+all sched_wakeup events in the system:
+
+ # perf record -a -e sched:sched_wakeup
+
+Traces meant to be processed using a script should be recorded with
+the above option: -a to enable system-wide collection.
+
+The format file for the sched_wakep event defines the following fields
+(see /sys/kernel/debug/tracing/events/sched/sched_wakeup/format):
+
+----
+ format:
+        field:unsigned short common_type;
+        field:unsigned char common_flags;
+        field:unsigned char common_preempt_count;
+        field:int common_pid;
+
+        field:char comm[TASK_COMM_LEN];
+        field:pid_t pid;
+        field:int prio;
+        field:int success;
+        field:int target_cpu;
+----
+
+The handler function for this event would be defined as:
+
+----
+sub sched::sched_wakeup
+{
+   my ($event_name, $context, $common_cpu, $common_secs,
+       $common_nsecs, $common_pid, $common_comm,
+       $comm, $pid, $prio, $success, $target_cpu) = @_;
+}
+----
+
+The handler function takes the form subsystem::event_name.
+
+The $common_* arguments in the handler's argument list are the set of
+arguments passed to all event handlers; some of the fields correspond
+to the common_* fields in the format file, but some are synthesized,
+and some of the common_* fields aren't common enough to to be passed
+to every event as arguments but are available as library functions.
+
+Here's a brief description of each of the invariant event args:
+
+ $event_name 	  	    the name of the event as text
+ $context		    an opaque 'cookie' used in calls back into perf
+ $common_cpu		    the cpu the event occurred on
+ $common_secs		    the secs portion of the event timestamp
+ $common_nsecs		    the nsecs portion of the event timestamp
+ $common_pid		    the pid of the current task
+ $common_comm		    the name of the current process
+
+All of the remaining fields in the event's format file have
+counterparts as handler function arguments of the same name, as can be
+seen in the example above.
+
+The above provides the basics needed to directly access every field of
+every event in a trace, which covers 90% of what you need to know to
+write a useful trace script.  The sections below cover the rest.
+
+SCRIPT LAYOUT
+-------------
+
+Every perf script Perl script should start by setting up a Perl module
+search path and 'use'ing a few support modules (see module
+descriptions below):
+
+----
+ use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+ use lib "./Perf-Trace-Util/lib";
+ use Perf::Trace::Core;
+ use Perf::Trace::Context;
+ use Perf::Trace::Util;
+----
+
+The rest of the script can contain handler functions and support
+functions in any order.
+
+Aside from the event handler functions discussed above, every script
+can implement a set of optional functions:
+
+*trace_begin*, if defined, is called before any event is processed and
+gives scripts a chance to do setup tasks:
+
+----
+ sub trace_begin
+ {
+ }
+----
+
+*trace_end*, if defined, is called after all events have been
+ processed and gives scripts a chance to do end-of-script tasks, such
+ as display results:
+
+----
+sub trace_end
+{
+}
+----
+
+*trace_unhandled*, if defined, is called after for any event that
+ doesn't have a handler explicitly defined for it.  The standard set
+ of common arguments are passed into it:
+
+----
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs,
+        $common_nsecs, $common_pid, $common_comm) = @_;
+}
+----
+
+The remaining sections provide descriptions of each of the available
+built-in perf script Perl modules and their associated functions.
+
+AVAILABLE MODULES AND FUNCTIONS
+-------------------------------
+
+The following sections describe the functions and variables available
+via the various Perf::Trace::* Perl modules.  To use the functions and
+variables from the given module, add the corresponding 'use
+Perf::Trace::XXX' line to your perf script script.
+
+Perf::Trace::Core Module
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+These functions provide some essential functions to user scripts.
+
+The *flag_str* and *symbol_str* functions provide human-readable
+strings for flag and symbolic fields.  These correspond to the strings
+and values parsed from the 'print fmt' fields of the event format
+files:
+
+  flag_str($event_name, $field_name, $field_value) - returns the string representation corresponding to $field_value for the flag field $field_name of event $event_name
+  symbol_str($event_name, $field_name, $field_value) - returns the string representation corresponding to $field_value for the symbolic field $field_name of event $event_name
+
+Perf::Trace::Context Module
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Some of the 'common' fields in the event format file aren't all that
+common, but need to be made accessible to user scripts nonetheless.
+
+Perf::Trace::Context defines a set of functions that can be used to
+access this data in the context of the current event.  Each of these
+functions expects a $context variable, which is the same as the
+$context variable passed into every event handler as the second
+argument.
+
+ common_pc($context) - returns common_preempt count for the current event
+ common_flags($context) - returns common_flags for the current event
+ common_lock_depth($context) - returns common_lock_depth for the current event
+
+Perf::Trace::Util Module
+~~~~~~~~~~~~~~~~~~~~~~~~
+
+Various utility functions for use with perf script:
+
+  nsecs($secs, $nsecs) - returns total nsecs given secs/nsecs pair
+  nsecs_secs($nsecs) - returns whole secs portion given nsecs
+  nsecs_nsecs($nsecs) - returns nsecs remainder given nsecs
+  nsecs_str($nsecs) - returns printable string in the form secs.nsecs
+  avg($total, $n) - returns average given a sum and a total number of values
+
+SEE ALSO
+--------
+linkperf:perf-script[1]
diff --git a/tools/perf/Documentation/perf-script-python.txt b/tools/perf/Documentation/perf-script-python.txt
new file mode 100644
index 000000000..54acba221
--- /dev/null
+++ b/tools/perf/Documentation/perf-script-python.txt
@@ -0,0 +1,620 @@
+perf-script-python(1)
+====================
+
+NAME
+----
+perf-script-python - Process trace data with a Python script
+
+SYNOPSIS
+--------
+[verse]
+'perf script' [-s [Python]:script[.py] ]
+
+DESCRIPTION
+-----------
+
+This perf script option is used to process perf script data using perf's
+built-in Python interpreter.  It reads and processes the input file and
+displays the results of the trace analysis implemented in the given
+Python script, if any.
+
+A QUICK EXAMPLE
+---------------
+
+This section shows the process, start to finish, of creating a working
+Python script that aggregates and extracts useful information from a
+raw perf script stream.  You can avoid reading the rest of this
+document if an example is enough for you; the rest of the document
+provides more details on each step and lists the library functions
+available to script writers.
+
+This example actually details the steps that were used to create the
+'syscall-counts' script you see when you list the available perf script
+scripts via 'perf script -l'.  As such, this script also shows how to
+integrate your script into the list of general-purpose 'perf script'
+scripts listed by that command.
+
+The syscall-counts script is a simple script, but demonstrates all the
+basic ideas necessary to create a useful script.  Here's an example
+of its output (syscall names are not yet supported, they will appear
+as numbers):
+
+----
+syscall events:
+
+event                                          count
+----------------------------------------  -----------
+sys_write                                     455067
+sys_getdents                                    4072
+sys_close                                       3037
+sys_swapoff                                     1769
+sys_read                                         923
+sys_sched_setparam                               826
+sys_open                                         331
+sys_newfstat                                     326
+sys_mmap                                         217
+sys_munmap                                       216
+sys_futex                                        141
+sys_select                                       102
+sys_poll                                          84
+sys_setitimer                                     12
+sys_writev                                         8
+15                                                 8
+sys_lseek                                          7
+sys_rt_sigprocmask                                 6
+sys_wait4                                          3
+sys_ioctl                                          3
+sys_set_robust_list                                1
+sys_exit                                           1
+56                                                 1
+sys_access                                         1
+----
+
+Basically our task is to keep a per-syscall tally that gets updated
+every time a system call occurs in the system.  Our script will do
+that, but first we need to record the data that will be processed by
+that script.  Theoretically, there are a couple of ways we could do
+that:
+
+- we could enable every event under the tracing/events/syscalls
+  directory, but this is over 600 syscalls, well beyond the number
+  allowable by perf.  These individual syscall events will however be
+  useful if we want to later use the guidance we get from the
+  general-purpose scripts to drill down and get more detail about
+  individual syscalls of interest.
+
+- we can enable the sys_enter and/or sys_exit syscalls found under
+  tracing/events/raw_syscalls.  These are called for all syscalls; the
+  'id' field can be used to distinguish between individual syscall
+  numbers.
+
+For this script, we only need to know that a syscall was entered; we
+don't care how it exited, so we'll use 'perf record' to record only
+the sys_enter events:
+
+----
+# perf record -a -e raw_syscalls:sys_enter
+
+^C[ perf record: Woken up 1 times to write data ]
+[ perf record: Captured and wrote 56.545 MB perf.data (~2470503 samples) ]
+----
+
+The options basically say to collect data for every syscall event
+system-wide and multiplex the per-cpu output into a single stream.
+That single stream will be recorded in a file in the current directory
+called perf.data.
+
+Once we have a perf.data file containing our data, we can use the -g
+'perf script' option to generate a Python script that will contain a
+callback handler for each event type found in the perf.data trace
+stream (for more details, see the STARTER SCRIPTS section).
+
+----
+# perf script -g python
+generated Python script: perf-script.py
+
+The output file created also in the current directory is named
+perf-script.py.  Here's the file in its entirety:
+
+# perf script event handlers, generated by perf script -g python
+# Licensed under the terms of the GNU GPL License version 2
+
+# The common_* event handler fields are the most useful fields common to
+# all events.  They don't necessarily correspond to the 'common_*' fields
+# in the format files.  Those fields not available as handler params can
+# be retrieved using Python functions of the form common_*(context).
+# See the perf-script-python Documentation for the list of available functions.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+
+def trace_begin():
+	print "in trace_begin"
+
+def trace_end():
+	print "in trace_end"
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, args):
+		print_header(event_name, common_cpu, common_secs, common_nsecs,
+			common_pid, common_comm)
+
+		print "id=%d, args=%s\n" % \
+		(id, args),
+
+def trace_unhandled(event_name, context, common_cpu, common_secs, common_nsecs,
+		common_pid, common_comm):
+		print_header(event_name, common_cpu, common_secs, common_nsecs,
+		common_pid, common_comm)
+
+def print_header(event_name, cpu, secs, nsecs, pid, comm):
+	print "%-20s %5u %05u.%09u %8u %-20s " % \
+	(event_name, cpu, secs, nsecs, pid, comm),
+----
+
+At the top is a comment block followed by some import statements and a
+path append which every perf script script should include.
+
+Following that are a couple generated functions, trace_begin() and
+trace_end(), which are called at the beginning and the end of the
+script respectively (for more details, see the SCRIPT_LAYOUT section
+below).
+
+Following those are the 'event handler' functions generated one for
+every event in the 'perf record' output.  The handler functions take
+the form subsystem__event_name, and contain named parameters, one for
+each field in the event; in this case, there's only one event,
+raw_syscalls__sys_enter().  (see the EVENT HANDLERS section below for
+more info on event handlers).
+
+The final couple of functions are, like the begin and end functions,
+generated for every script.  The first, trace_unhandled(), is called
+every time the script finds an event in the perf.data file that
+doesn't correspond to any event handler in the script.  This could
+mean either that the record step recorded event types that it wasn't
+really interested in, or the script was run against a trace file that
+doesn't correspond to the script.
+
+The script generated by -g option simply prints a line for each
+event found in the trace stream i.e. it basically just dumps the event
+and its parameter values to stdout.  The print_header() function is
+simply a utility function used for that purpose.  Let's rename the
+script and run it to see the default output:
+
+----
+# mv perf-script.py syscall-counts.py
+# perf script -s syscall-counts.py
+
+raw_syscalls__sys_enter     1 00840.847582083     7506 perf                  id=1, args=
+raw_syscalls__sys_enter     1 00840.847595764     7506 perf                  id=1, args=
+raw_syscalls__sys_enter     1 00840.847620860     7506 perf                  id=1, args=
+raw_syscalls__sys_enter     1 00840.847710478     6533 npviewer.bin          id=78, args=
+raw_syscalls__sys_enter     1 00840.847719204     6533 npviewer.bin          id=142, args=
+raw_syscalls__sys_enter     1 00840.847755445     6533 npviewer.bin          id=3, args=
+raw_syscalls__sys_enter     1 00840.847775601     6533 npviewer.bin          id=3, args=
+raw_syscalls__sys_enter     1 00840.847781820     6533 npviewer.bin          id=3, args=
+.
+.
+.
+----
+
+Of course, for this script, we're not interested in printing every
+trace event, but rather aggregating it in a useful way.  So we'll get
+rid of everything to do with printing as well as the trace_begin() and
+trace_unhandled() functions, which we won't be using.  That leaves us
+with this minimalistic skeleton:
+
+----
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+
+def trace_end():
+	print "in trace_end"
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, args):
+----
+
+In trace_end(), we'll simply print the results, but first we need to
+generate some results to print.  To do that we need to have our
+sys_enter() handler do the necessary tallying until all events have
+been counted.  A hash table indexed by syscall id is a good way to
+store that information; every time the sys_enter() handler is called,
+we simply increment a count associated with that hash entry indexed by
+that syscall id:
+
+----
+  syscalls = autodict()
+
+  try:
+    syscalls[id] += 1
+  except TypeError:
+    syscalls[id] = 1
+----
+
+The syscalls 'autodict' object is a special kind of Python dictionary
+(implemented in Core.py) that implements Perl's 'autovivifying' hashes
+in Python i.e. with autovivifying hashes, you can assign nested hash
+values without having to go to the trouble of creating intermediate
+levels if they don't exist e.g syscalls[comm][pid][id] = 1 will create
+the intermediate hash levels and finally assign the value 1 to the
+hash entry for 'id' (because the value being assigned isn't a hash
+object itself, the initial value is assigned in the TypeError
+exception.  Well, there may be a better way to do this in Python but
+that's what works for now).
+
+Putting that code into the raw_syscalls__sys_enter() handler, we
+effectively end up with a single-level dictionary keyed on syscall id
+and having the counts we've tallied as values.
+
+The print_syscall_totals() function iterates over the entries in the
+dictionary and displays a line for each entry containing the syscall
+name (the dictionary keys contain the syscall ids, which are passed to
+the Util function syscall_name(), which translates the raw syscall
+numbers to the corresponding syscall name strings).  The output is
+displayed after all the events in the trace have been processed, by
+calling the print_syscall_totals() function from the trace_end()
+handler called at the end of script processing.
+
+The final script producing the output shown above is shown in its
+entirety below (syscall_name() helper is not yet available, you can
+only deal with id's for now):
+
+----
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+syscalls = autodict()
+
+def trace_end():
+	print_syscall_totals()
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, args):
+	try:
+		syscalls[id] += 1
+	except TypeError:
+		syscalls[id] = 1
+
+def print_syscall_totals():
+    if for_comm is not None:
+	    print "\nsyscall events for %s:\n\n" % (for_comm),
+    else:
+	    print "\nsyscall events:\n\n",
+
+    print "%-40s  %10s\n" % ("event", "count"),
+    print "%-40s  %10s\n" % ("----------------------------------------", \
+                                 "-----------"),
+
+    for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \
+				  reverse = True):
+	    print "%-40s  %10d\n" % (syscall_name(id), val),
+----
+
+The script can be run just as before:
+
+  # perf script -s syscall-counts.py
+
+So those are the essential steps in writing and running a script.  The
+process can be generalized to any tracepoint or set of tracepoints
+you're interested in - basically find the tracepoint(s) you're
+interested in by looking at the list of available events shown by
+'perf list' and/or look in /sys/kernel/debug/tracing events for
+detailed event and field info, record the corresponding trace data
+using 'perf record', passing it the list of interesting events,
+generate a skeleton script using 'perf script -g python' and modify the
+code to aggregate and display it for your particular needs.
+
+After you've done that you may end up with a general-purpose script
+that you want to keep around and have available for future use.  By
+writing a couple of very simple shell scripts and putting them in the
+right place, you can have your script listed alongside the other
+scripts listed by the 'perf script -l' command e.g.:
+
+----
+root@tropicana:~# perf script -l
+List of available trace scripts:
+  wakeup-latency                       system-wide min/max/avg wakeup latency
+  rw-by-file <comm>                    r/w activity for a program, by file
+  rw-by-pid                            system-wide r/w activity
+----
+
+A nice side effect of doing this is that you also then capture the
+probably lengthy 'perf record' command needed to record the events for
+the script.
+
+To have the script appear as a 'built-in' script, you write two simple
+scripts, one for recording and one for 'reporting'.
+
+The 'record' script is a shell script with the same base name as your
+script, but with -record appended.  The shell script should be put
+into the perf/scripts/python/bin directory in the kernel source tree.
+In that script, you write the 'perf record' command-line needed for
+your script:
+
+----
+# cat kernel-source/tools/perf/scripts/python/bin/syscall-counts-record
+
+#!/bin/bash
+perf record -a -e raw_syscalls:sys_enter
+----
+
+The 'report' script is also a shell script with the same base name as
+your script, but with -report appended.  It should also be located in
+the perf/scripts/python/bin directory.  In that script, you write the
+'perf script -s' command-line needed for running your script:
+
+----
+# cat kernel-source/tools/perf/scripts/python/bin/syscall-counts-report
+
+#!/bin/bash
+# description: system-wide syscall counts
+perf script -s ~/libexec/perf-core/scripts/python/syscall-counts.py
+----
+
+Note that the location of the Python script given in the shell script
+is in the libexec/perf-core/scripts/python directory - this is where
+the script will be copied by 'make install' when you install perf.
+For the installation to install your script there, your script needs
+to be located in the perf/scripts/python directory in the kernel
+source tree:
+
+----
+# ls -al kernel-source/tools/perf/scripts/python
+
+root@tropicana:/home/trz/src/tip# ls -al tools/perf/scripts/python
+total 32
+drwxr-xr-x 4 trz trz 4096 2010-01-26 22:30 .
+drwxr-xr-x 4 trz trz 4096 2010-01-26 22:29 ..
+drwxr-xr-x 2 trz trz 4096 2010-01-26 22:29 bin
+-rw-r--r-- 1 trz trz 2548 2010-01-26 22:29 check-perf-script.py
+drwxr-xr-x 3 trz trz 4096 2010-01-26 22:49 Perf-Trace-Util
+-rw-r--r-- 1 trz trz 1462 2010-01-26 22:30 syscall-counts.py
+----
+
+Once you've done that (don't forget to do a new 'make install',
+otherwise your script won't show up at run-time), 'perf script -l'
+should show a new entry for your script:
+
+----
+root@tropicana:~# perf script -l
+List of available trace scripts:
+  wakeup-latency                       system-wide min/max/avg wakeup latency
+  rw-by-file <comm>                    r/w activity for a program, by file
+  rw-by-pid                            system-wide r/w activity
+  syscall-counts                       system-wide syscall counts
+----
+
+You can now perform the record step via 'perf script record':
+
+  # perf script record syscall-counts
+
+and display the output using 'perf script report':
+
+  # perf script report syscall-counts
+
+STARTER SCRIPTS
+---------------
+
+You can quickly get started writing a script for a particular set of
+trace data by generating a skeleton script using 'perf script -g
+python' in the same directory as an existing perf.data trace file.
+That will generate a starter script containing a handler for each of
+the event types in the trace file; it simply prints every available
+field for each event in the trace file.
+
+You can also look at the existing scripts in
+~/libexec/perf-core/scripts/python for typical examples showing how to
+do basic things like aggregate event data, print results, etc.  Also,
+the check-perf-script.py script, while not interesting for its results,
+attempts to exercise all of the main scripting features.
+
+EVENT HANDLERS
+--------------
+
+When perf script is invoked using a trace script, a user-defined
+'handler function' is called for each event in the trace.  If there's
+no handler function defined for a given event type, the event is
+ignored (or passed to a 'trace_handled' function, see below) and the
+next event is processed.
+
+Most of the event's field values are passed as arguments to the
+handler function; some of the less common ones aren't - those are
+available as calls back into the perf executable (see below).
+
+As an example, the following perf record command can be used to record
+all sched_wakeup events in the system:
+
+ # perf record -a -e sched:sched_wakeup
+
+Traces meant to be processed using a script should be recorded with
+the above option: -a to enable system-wide collection.
+
+The format file for the sched_wakep event defines the following fields
+(see /sys/kernel/debug/tracing/events/sched/sched_wakeup/format):
+
+----
+ format:
+        field:unsigned short common_type;
+        field:unsigned char common_flags;
+        field:unsigned char common_preempt_count;
+        field:int common_pid;
+
+        field:char comm[TASK_COMM_LEN];
+        field:pid_t pid;
+        field:int prio;
+        field:int success;
+        field:int target_cpu;
+----
+
+The handler function for this event would be defined as:
+
+----
+def sched__sched_wakeup(event_name, context, common_cpu, common_secs,
+       common_nsecs, common_pid, common_comm,
+       comm, pid, prio, success, target_cpu):
+       pass
+----
+
+The handler function takes the form subsystem__event_name.
+
+The common_* arguments in the handler's argument list are the set of
+arguments passed to all event handlers; some of the fields correspond
+to the common_* fields in the format file, but some are synthesized,
+and some of the common_* fields aren't common enough to to be passed
+to every event as arguments but are available as library functions.
+
+Here's a brief description of each of the invariant event args:
+
+ event_name 	  	    the name of the event as text
+ context		    an opaque 'cookie' used in calls back into perf
+ common_cpu		    the cpu the event occurred on
+ common_secs		    the secs portion of the event timestamp
+ common_nsecs		    the nsecs portion of the event timestamp
+ common_pid		    the pid of the current task
+ common_comm		    the name of the current process
+
+All of the remaining fields in the event's format file have
+counterparts as handler function arguments of the same name, as can be
+seen in the example above.
+
+The above provides the basics needed to directly access every field of
+every event in a trace, which covers 90% of what you need to know to
+write a useful trace script.  The sections below cover the rest.
+
+SCRIPT LAYOUT
+-------------
+
+Every perf script Python script should start by setting up a Python
+module search path and 'import'ing a few support modules (see module
+descriptions below):
+
+----
+ import os
+ import sys
+
+ sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	      '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+ from perf_trace_context import *
+ from Core import *
+----
+
+The rest of the script can contain handler functions and support
+functions in any order.
+
+Aside from the event handler functions discussed above, every script
+can implement a set of optional functions:
+
+*trace_begin*, if defined, is called before any event is processed and
+gives scripts a chance to do setup tasks:
+
+----
+def trace_begin:
+    pass
+----
+
+*trace_end*, if defined, is called after all events have been
+ processed and gives scripts a chance to do end-of-script tasks, such
+ as display results:
+
+----
+def trace_end:
+    pass
+----
+
+*trace_unhandled*, if defined, is called after for any event that
+ doesn't have a handler explicitly defined for it.  The standard set
+ of common arguments are passed into it:
+
+----
+def trace_unhandled(event_name, context, common_cpu, common_secs,
+        common_nsecs, common_pid, common_comm):
+    pass
+----
+
+The remaining sections provide descriptions of each of the available
+built-in perf script Python modules and their associated functions.
+
+AVAILABLE MODULES AND FUNCTIONS
+-------------------------------
+
+The following sections describe the functions and variables available
+via the various perf script Python modules.  To use the functions and
+variables from the given module, add the corresponding 'from XXXX
+import' line to your perf script script.
+
+Core.py Module
+~~~~~~~~~~~~~~
+
+These functions provide some essential functions to user scripts.
+
+The *flag_str* and *symbol_str* functions provide human-readable
+strings for flag and symbolic fields.  These correspond to the strings
+and values parsed from the 'print fmt' fields of the event format
+files:
+
+  flag_str(event_name, field_name, field_value) - returns the string representation corresponding to field_value for the flag field field_name of event event_name
+  symbol_str(event_name, field_name, field_value) - returns the string representation corresponding to field_value for the symbolic field field_name of event event_name
+
+The *autodict* function returns a special kind of Python
+dictionary that implements Perl's 'autovivifying' hashes in Python
+i.e. with autovivifying hashes, you can assign nested hash values
+without having to go to the trouble of creating intermediate levels if
+they don't exist.
+
+  autodict() - returns an autovivifying dictionary instance
+
+
+perf_trace_context Module
+~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Some of the 'common' fields in the event format file aren't all that
+common, but need to be made accessible to user scripts nonetheless.
+
+perf_trace_context defines a set of functions that can be used to
+access this data in the context of the current event.  Each of these
+functions expects a context variable, which is the same as the
+context variable passed into every event handler as the second
+argument.
+
+ common_pc(context) - returns common_preempt count for the current event
+ common_flags(context) - returns common_flags for the current event
+ common_lock_depth(context) - returns common_lock_depth for the current event
+
+Util.py Module
+~~~~~~~~~~~~~~
+
+Various utility functions for use with perf script:
+
+  nsecs(secs, nsecs) - returns total nsecs given secs/nsecs pair
+  nsecs_secs(nsecs) - returns whole secs portion given nsecs
+  nsecs_nsecs(nsecs) - returns nsecs remainder given nsecs
+  nsecs_str(nsecs) - returns printable string in the form secs.nsecs
+  avg(total, n) - returns average given a sum and a total number of values
+
+SEE ALSO
+--------
+linkperf:perf-script[1]
diff --git a/tools/perf/Documentation/perf-script.txt b/tools/perf/Documentation/perf-script.txt
new file mode 100644
index 000000000..79445750f
--- /dev/null
+++ b/tools/perf/Documentation/perf-script.txt
@@ -0,0 +1,227 @@
+perf-script(1)
+=============
+
+NAME
+----
+perf-script - Read perf.data (created by perf record) and display trace output
+
+SYNOPSIS
+--------
+[verse]
+'perf script' [<options>]
+'perf script' [<options>] record <script> [<record-options>] <command>
+'perf script' [<options>] report <script> [script-args]
+'perf script' [<options>] <script> <required-script-args> [<record-options>] <command>
+'perf script' [<options>] <top-script> [script-args]
+
+DESCRIPTION
+-----------
+This command reads the input file and displays the trace recorded.
+
+There are several variants of perf script:
+
+  'perf script' to see a detailed trace of the workload that was
+  recorded.
+
+  You can also run a set of pre-canned scripts that aggregate and
+  summarize the raw trace data in various ways (the list of scripts is
+  available via 'perf script -l').  The following variants allow you to
+  record and run those scripts:
+
+  'perf script record <script> <command>' to record the events required
+  for 'perf script report'.  <script> is the name displayed in the
+  output of 'perf script --list' i.e. the actual script name minus any
+  language extension.  If <command> is not specified, the events are
+  recorded using the -a (system-wide) 'perf record' option.
+
+  'perf script report <script> [args]' to run and display the results
+  of <script>.  <script> is the name displayed in the output of 'perf
+  trace --list' i.e. the actual script name minus any language
+  extension.  The perf.data output from a previous run of 'perf script
+  record <script>' is used and should be present for this command to
+  succeed.  [args] refers to the (mainly optional) args expected by
+  the script.
+
+  'perf script <script> <required-script-args> <command>' to both
+  record the events required for <script> and to run the <script>
+  using 'live-mode' i.e. without writing anything to disk.  <script>
+  is the name displayed in the output of 'perf script --list' i.e. the
+  actual script name minus any language extension.  If <command> is
+  not specified, the events are recorded using the -a (system-wide)
+  'perf record' option.  If <script> has any required args, they
+  should be specified before <command>.  This mode doesn't allow for
+  optional script args to be specified; if optional script args are
+  desired, they can be specified using separate 'perf script record'
+  and 'perf script report' commands, with the stdout of the record step
+  piped to the stdin of the report script, using the '-o -' and '-i -'
+  options of the corresponding commands.
+
+  'perf script <top-script>' to both record the events required for
+  <top-script> and to run the <top-script> using 'live-mode'
+  i.e. without writing anything to disk.  <top-script> is the name
+  displayed in the output of 'perf script --list' i.e. the actual
+  script name minus any language extension; a <top-script> is defined
+  as any script name ending with the string 'top'.
+
+  [<record-options>] can be passed to the record steps of 'perf script
+  record' and 'live-mode' variants; this isn't possible however for
+  <top-script> 'live-mode' or 'perf script report' variants.
+
+  See the 'SEE ALSO' section for links to language-specific
+  information on how to write and run your own trace scripts.
+
+OPTIONS
+-------
+<command>...::
+	Any command you can specify in a shell.
+
+-D::
+--dump-raw-script=::
+        Display verbose dump of the trace data.
+
+-L::
+--Latency=::
+        Show latency attributes (irqs/preemption disabled, etc).
+
+-l::
+--list=::
+        Display a list of available trace scripts.
+
+-s ['lang']::
+--script=::
+        Process trace data with the given script ([lang]:script[.ext]).
+	If the string 'lang' is specified in place of a script name, a
+        list of supported languages will be displayed instead.
+
+-g::
+--gen-script=::
+        Generate perf-script.[ext] starter script for given language,
+        using current perf.data.
+
+-a::
+        Force system-wide collection.  Scripts run without a <command>
+        normally use -a by default, while scripts run with a <command>
+        normally don't - this option allows the latter to be run in
+        system-wide mode.
+
+-i::
+--input=::
+        Input file name. (default: perf.data unless stdin is a fifo)
+
+-d::
+--debug-mode::
+        Do various checks like samples ordering and lost events.
+
+-f::
+--fields::
+        Comma separated list of fields to print. Options are:
+        comm, tid, pid, time, cpu, event, trace, ip, sym, dso, addr, symoff, srcline, period.
+        Field list can be prepended with the type, trace, sw or hw,
+        to indicate to which event type the field list applies.
+        e.g., -f sw:comm,tid,time,ip,sym  and -f trace:time,cpu,trace
+
+		perf script -f <fields>
+
+	is equivalent to:
+
+		perf script -f trace:<fields> -f sw:<fields> -f hw:<fields>
+
+	i.e., the specified fields apply to all event types if the type string
+	is not given.
+
+	The arguments are processed in the order received. A later usage can
+	reset a prior request. e.g.:
+
+		-f trace: -f comm,tid,time,ip,sym
+
+	The first -f suppresses trace events (field list is ""), but then the
+	second invocation sets the fields to comm,tid,time,ip,sym. In this case a
+	warning is given to the user:
+
+		"Overriding previous field request for all events."
+
+	Alternatively, consider the order:
+
+		-f comm,tid,time,ip,sym -f trace:
+
+	The first -f sets the fields for all events and the second -f
+	suppresses trace events. The user is given a warning message about
+	the override, and the result of the above is that only S/W and H/W
+	events are displayed with the given fields.
+
+	For the 'wildcard' option if a user selected field is invalid for an
+	event type, a message is displayed to the user that the option is
+	ignored for that type. For example:
+
+		$ perf script -f comm,tid,trace
+		'trace' not valid for hardware events. Ignoring.
+		'trace' not valid for software events. Ignoring.
+
+	Alternatively, if the type is given an invalid field is specified it
+	is an error. For example:
+
+        perf script -v -f sw:comm,tid,trace
+        'trace' not valid for software events.
+
+	At this point usage is displayed, and perf-script exits.
+
+	Finally, a user may not set fields to none for all event types.
+	i.e., -f "" is not allowed.
+
+-k::
+--vmlinux=<file>::
+        vmlinux pathname
+
+--kallsyms=<file>::
+        kallsyms pathname
+
+--symfs=<directory>::
+        Look for files with symbols relative to this directory.
+
+-G::
+--hide-call-graph::
+        When printing symbols do not display call chain.
+
+-C::
+--cpu:: Only report samples for the list of CPUs provided. Multiple CPUs can
+	be provided as a comma-separated list with no space: 0,1. Ranges of
+	CPUs are specified with -: 0-2. Default is to report samples on all
+	CPUs.
+
+-c::
+--comms=::
+	Only display events for these comms. CSV that understands
+	file://filename entries.
+
+--pid=::
+	Only show events for given process ID (comma separated list).
+
+--tid=::
+	Only show events for given thread ID (comma separated list).
+
+-I::
+--show-info::
+	Display extended information about the perf.data file. This adds
+	information which may be very large and thus may clutter the display.
+	It currently includes: cpu and numa topology of the host system.
+	It can only be used with the perf script report mode.
+
+--show-kernel-path::
+	Try to resolve the path of [kernel.kallsyms]
+
+--show-task-events
+	Display task related events (e.g. FORK, COMM, EXIT).
+
+--show-mmap-events
+	Display mmap related events (e.g. MMAP, MMAP2).
+
+--header
+	Show perf.data header.
+
+--header-only
+	Show only perf.data header.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-script-perl[1],
+linkperf:perf-script-python[1]
diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
new file mode 100644
index 000000000..04e150d83
--- /dev/null
+++ b/tools/perf/Documentation/perf-stat.txt
@@ -0,0 +1,177 @@
+perf-stat(1)
+============
+
+NAME
+----
+perf-stat - Run a command and gather performance counter statistics
+
+SYNOPSIS
+--------
+[verse]
+'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command>
+'perf stat' [-e <EVENT> | --event=EVENT] [-a] -- <command> [<options>]
+
+DESCRIPTION
+-----------
+This command runs a command and gathers performance counter statistics
+from it.
+
+
+OPTIONS
+-------
+<command>...::
+	Any command you can specify in a shell.
+
+
+-e::
+--event=::
+	Select the PMU event. Selection can be:
+
+	- a symbolic event name (use 'perf list' to list all events)
+
+	- a raw PMU event (eventsel+umask) in the form of rNNN where NNN is a
+	  hexadecimal event descriptor.
+
+	- a symbolically formed event like 'pmu/param1=0x3,param2/' where
+	  param1 and param2 are defined as formats for the PMU in
+	  /sys/bus/event_sources/devices/<pmu>/format/*
+
+	- a symbolically formed event like 'pmu/config=M,config1=N,config2=K/'
+	  where M, N, K are numbers (in decimal, hex, octal format).
+	  Acceptable values for each of 'config', 'config1' and 'config2'
+	  parameters are defined by corresponding entries in
+	  /sys/bus/event_sources/devices/<pmu>/format/*
+
+-i::
+--no-inherit::
+        child tasks do not inherit counters
+-p::
+--pid=<pid>::
+        stat events on existing process id (comma separated list)
+
+-t::
+--tid=<tid>::
+        stat events on existing thread id (comma separated list)
+
+
+-a::
+--all-cpus::
+        system-wide collection from all CPUs
+
+-c::
+--scale::
+	scale/normalize counter values
+
+-r::
+--repeat=<n>::
+	repeat command and print average + stddev (max: 100). 0 means forever.
+
+-B::
+--big-num::
+        print large numbers with thousands' separators according to locale
+
+-C::
+--cpu=::
+Count only on the list of CPUs provided. Multiple CPUs can be provided as a
+comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
+In per-thread mode, this option is ignored. The -a option is still necessary
+to activate system-wide monitoring. Default is to count on all CPUs.
+
+-A::
+--no-aggr::
+Do not aggregate counts across all monitored CPUs in system-wide mode (-a).
+This option is only valid in system-wide mode.
+
+-n::
+--null::
+        null run - don't start any counters
+
+-v::
+--verbose::
+        be more verbose (show counter open errors, etc)
+
+-x SEP::
+--field-separator SEP::
+print counts using a CSV-style output to make it easy to import directly into
+spreadsheets. Columns are separated by the string specified in SEP.
+
+-G name::
+--cgroup name::
+monitor only in the container (cgroup) called "name". This option is available only
+in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to
+container "name" are monitored when they run on the monitored CPUs. Multiple cgroups
+can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup
+to first event, second cgroup to second event and so on. It is possible to provide
+an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have
+corresponding events, i.e., they always refer to events defined earlier on the command
+line.
+
+-o file::
+--output file::
+Print the output into the designated file.
+
+--append::
+Append to the output file designated with the -o option. Ignored if -o is not specified.
+
+--log-fd::
+
+Log output to fd, instead of stderr.  Complementary to --output, and mutually exclusive
+with it.  --append may be used here.  Examples:
+     3>results  perf stat --log-fd 3          -- $cmd
+     3>>results perf stat --log-fd 3 --append -- $cmd
+
+--pre::
+--post::
+	Pre and post measurement hooks, e.g.:
+
+perf stat --repeat 10 --null --sync --pre 'make -s O=defconfig-build/clean' -- make -s -j64 O=defconfig-build/ bzImage
+
+-I msecs::
+--interval-print msecs::
+	Print count deltas every N milliseconds (minimum: 100ms)
+	example: perf stat -I 1000 -e cycles -a sleep 5
+
+--per-socket::
+Aggregate counts per processor socket for system-wide mode measurements.  This
+is a useful mode to detect imbalance between sockets.  To enable this mode,
+use --per-socket in addition to -a. (system-wide).  The output includes the
+socket number and the number of online processors on that socket. This is
+useful to gauge the amount of aggregation.
+
+--per-core::
+Aggregate counts per physical processor for system-wide mode measurements.  This
+is a useful mode to detect imbalance between physical cores.  To enable this mode,
+use --per-core in addition to -a. (system-wide).  The output includes the
+core number and the number of online logical processors on that physical processor.
+
+-D msecs::
+--delay msecs::
+After starting the program, wait msecs before measuring. This is useful to
+filter out the startup phase of the program, which is often very different.
+
+-T::
+--transaction::
+
+Print statistics of transactional execution if supported.
+
+EXAMPLES
+--------
+
+$ perf stat -- make -j
+
+ Performance counter stats for 'make -j':
+
+    8117.370256  task clock ticks     #      11.281 CPU utilization factor
+            678  context switches     #       0.000 M/sec
+            133  CPU migrations       #       0.000 M/sec
+         235724  pagefaults           #       0.029 M/sec
+    24821162526  CPU cycles           #    3057.784 M/sec
+    18687303457  instructions         #    2302.138 M/sec
+      172158895  cache references     #      21.209 M/sec
+       27075259  cache misses         #       3.335 M/sec
+
+ Wall-clock time elapsed:   719.554352 msecs
+
+SEE ALSO
+--------
+linkperf:perf-top[1], linkperf:perf-list[1]
diff --git a/tools/perf/Documentation/perf-test.txt b/tools/perf/Documentation/perf-test.txt
new file mode 100644
index 000000000..31a5c3ea7
--- /dev/null
+++ b/tools/perf/Documentation/perf-test.txt
@@ -0,0 +1,32 @@
+perf-test(1)
+============
+
+NAME
+----
+perf-test - Runs sanity tests.
+
+SYNOPSIS
+--------
+[verse]
+'perf test [<options>] [{list <test-name-fragment>|[<test-name-fragments>|<test-numbers>]}]'
+
+DESCRIPTION
+-----------
+This command does assorted sanity tests, initially through linked routines but
+also will look for a directory with more tests in the form of scripts.
+
+To get a list of available tests use 'perf test list', specifying a test name
+fragment will show all tests that have it.
+
+To run just specific tests, inform test name fragments or the numbers obtained
+from 'perf test list'.
+
+OPTIONS
+-------
+-s::
+--skip::
+	Tests to skip (comma separated numeric list).
+
+-v::
+--verbose::
+	Be more verbose.
diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt
new file mode 100644
index 000000000..df98d1c82
--- /dev/null
+++ b/tools/perf/Documentation/perf-timechart.txt
@@ -0,0 +1,126 @@
+perf-timechart(1)
+=================
+
+NAME
+----
+perf-timechart - Tool to visualize total system behavior during a workload
+
+SYNOPSIS
+--------
+[verse]
+'perf timechart' [<timechart options>] {record} [<record options>]
+
+DESCRIPTION
+-----------
+There are two variants of perf timechart:
+
+  'perf timechart record <command>' to record the system level events
+  of an arbitrary workload. By default timechart records only scheduler
+  and CPU events (task switches, running times, CPU power states, etc),
+  but it's possible to record IO (disk, network) activity using -I argument.
+
+  'perf timechart' to turn a trace into a Scalable Vector Graphics file,
+  that can be viewed with popular SVG viewers such as 'Inkscape'. Depending
+  on the events in the perf.data file, timechart will contain scheduler/cpu
+  events or IO events.
+
+  In IO mode, every bar has two charts: upper and lower.
+  Upper bar shows incoming events (disk reads, ingress network packets).
+  Lower bar shows outgoing events (disk writes, egress network packets).
+  There are also poll bars which show how much time application spent
+  in poll/epoll/select syscalls.
+
+TIMECHART OPTIONS
+-----------------
+-o::
+--output=::
+        Select the output file (default: output.svg)
+-i::
+--input=::
+        Select the input file (default: perf.data unless stdin is a fifo)
+-w::
+--width=::
+        Select the width of the SVG file (default: 1000)
+-P::
+--power-only::
+        Only output the CPU power section of the diagram
+-T::
+--tasks-only::
+        Don't output processor state transitions
+-p::
+--process::
+        Select the processes to display, by name or PID
+
+--symfs=<directory>::
+        Look for files with symbols relative to this directory.
+-n::
+--proc-num::
+        Print task info for at least given number of tasks.
+-t::
+--topology::
+        Sort CPUs according to topology.
+--highlight=<duration_nsecs|task_name>::
+	Highlight tasks (using different color) that run more than given
+	duration or tasks with given name. If number is given it's interpreted
+	as number of nanoseconds. If non-numeric string is given it's
+	interpreted as task name.
+--io-skip-eagain::
+	Don't draw EAGAIN IO events.
+--io-min-time=<nsecs>::
+	Draw small events as if they lasted min-time. Useful when you need
+	to see very small and fast IO. It's possible to specify ms or us
+	suffix to specify time in milliseconds or microseconds.
+	Default value is 1ms.
+--io-merge-dist=<nsecs>::
+	Merge events that are merge-dist nanoseconds apart.
+	Reduces number of figures on the SVG and makes it more render-friendly.
+	It's possible to specify ms or us suffix to specify time in
+	milliseconds or microseconds.
+	Default value is 1us.
+
+RECORD OPTIONS
+--------------
+-P::
+--power-only::
+        Record only power-related events
+-T::
+--tasks-only::
+        Record only tasks-related events
+-I::
+--io-only::
+        Record only io-related events
+-g::
+--callchain::
+        Do call-graph (stack chain/backtrace) recording
+
+EXAMPLES
+--------
+
+$ perf timechart record git pull
+
+  [ perf record: Woken up 13 times to write data ]
+  [ perf record: Captured and wrote 4.253 MB perf.data (~185801 samples) ]
+
+$ perf timechart
+
+  Written 10.2 seconds of trace to output.svg.
+
+Record system-wide timechart:
+
+  $ perf timechart record
+
+  then generate timechart and highlight 'gcc' tasks:
+
+  $ perf timechart --highlight gcc
+
+Record system-wide IO events:
+
+  $ perf timechart record -I
+
+  then generate timechart:
+
+  $ perf timechart
+
+SEE ALSO
+--------
+linkperf:perf-record[1]
diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt
new file mode 100644
index 000000000..3265b1070
--- /dev/null
+++ b/tools/perf/Documentation/perf-top.txt
@@ -0,0 +1,240 @@
+perf-top(1)
+===========
+
+NAME
+----
+perf-top - System profiling tool.
+
+SYNOPSIS
+--------
+[verse]
+'perf top' [-e <EVENT> | --event=EVENT] [<options>]
+
+DESCRIPTION
+-----------
+This command generates and displays a performance counter profile in real time.
+
+
+OPTIONS
+-------
+-a::
+--all-cpus::
+        System-wide collection.  (default)
+
+-c <count>::
+--count=<count>::
+	Event period to sample.
+
+-C <cpu-list>::
+--cpu=<cpu>::
+Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a
+comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
+Default is to monitor all CPUS.
+
+-d <seconds>::
+--delay=<seconds>::
+	Number of seconds to delay between refreshes.
+
+-e <event>::
+--event=<event>::
+	Select the PMU event. Selection can be a symbolic event name
+	(use 'perf list' to list all events) or a raw PMU
+	event (eventsel+umask) in the form of rNNN where NNN is a
+	hexadecimal event descriptor.
+
+-E <entries>::
+--entries=<entries>::
+	Display this many functions.
+
+-f <count>::
+--count-filter=<count>::
+	Only display functions with more events than this.
+
+--group::
+        Put the counters into a counter group.
+
+-F <freq>::
+--freq=<freq>::
+	Profile at this frequency.
+
+-i::
+--inherit::
+	Child tasks do not inherit counters.
+
+-k <path>::
+--vmlinux=<path>::
+	Path to vmlinux.  Required for annotation functionality.
+
+-m <pages>::
+--mmap-pages=<pages>::
+	Number of mmap data pages (must be a power of two) or size
+	specification with appended unit character - B/K/M/G. The
+	size is rounded up to have nearest pages power of two value.
+
+-p <pid>::
+--pid=<pid>::
+	Profile events on existing Process ID (comma separated list).
+
+-t <tid>::
+--tid=<tid>::
+        Profile events on existing thread ID (comma separated list).
+
+-u::
+--uid=::
+        Record events in threads owned by uid. Name or number.
+
+-r <priority>::
+--realtime=<priority>::
+	Collect data with this RT SCHED_FIFO priority.
+
+--sym-annotate=<symbol>::
+        Annotate this symbol.
+
+-K::
+--hide_kernel_symbols::
+        Hide kernel symbols.
+
+-U::
+--hide_user_symbols::
+        Hide user symbols.
+
+--demangle-kernel::
+        Demangle kernel symbols.
+
+-D::
+--dump-symtab::
+        Dump the symbol table used for profiling.
+
+-v::
+--verbose::
+	Be more verbose (show counter open errors, etc).
+
+-z::
+--zero::
+	Zero history across display updates.
+
+-s::
+--sort::
+	Sort by key(s): pid, comm, dso, symbol, parent, srcline, weight,
+	local_weight, abort, in_tx, transaction, overhead, sample, period.
+	Please see description of --sort in the perf-report man page.
+
+--fields=::
+	Specify output field - multiple keys can be specified in CSV format.
+	Following fields are available:
+	overhead, overhead_sys, overhead_us, overhead_children, sample and period.
+	Also it can contain any sort key(s).
+
+	By default, every sort keys not specified in --field will be appended
+	automatically.
+
+-n::
+--show-nr-samples::
+	Show a column with the number of samples.
+
+--show-total-period::
+	Show a column with the sum of periods.
+
+--dsos::
+	Only consider symbols in these dsos.  This option will affect the
+	percentage of the overhead column.  See --percentage for more info.
+
+--comms::
+	Only consider symbols in these comms.  This option will affect the
+	percentage of the overhead column.  See --percentage for more info.
+
+--symbols::
+	Only consider these symbols.  This option will affect the
+	percentage of the overhead column.  See --percentage for more info.
+
+-M::
+--disassembler-style=:: Set disassembler style for objdump.
+
+--source::
+	Interleave source code with assembly code. Enabled by default,
+	disable with --no-source.
+
+--asm-raw::
+	Show raw instruction encoding of assembly instructions.
+
+-g::
+	Enables call-graph (stack chain/backtrace) recording.
+
+--call-graph::
+	Setup and enable call-graph (stack chain/backtrace) recording,
+	implies -g.
+
+--children::
+	Accumulate callchain of children to parent entry so that then can
+	show up in the output.  The output will have a new "Children" column
+	and will be sorted on the data.  It requires -g/--call-graph option
+	enabled.
+
+--max-stack::
+	Set the stack depth limit when parsing the callchain, anything
+	beyond the specified depth will be ignored. This is a trade-off
+	between information loss and faster processing especially for
+	workloads that can have a very long callchain stack.
+
+	Default: 127
+
+--ignore-callees=<regex>::
+        Ignore callees of the function(s) matching the given regex.
+        This has the effect of collecting the callers of each such
+        function into one place in the call-graph tree.
+
+--percent-limit::
+	Do not show entries which have an overhead under that percent.
+	(Default: 0).
+
+--percentage::
+	Determine how to display the overhead percentage of filtered entries.
+	Filters can be applied by --comms, --dsos and/or --symbols options and
+	Zoom operations on the TUI (thread, dso, etc).
+
+	"relative" means it's relative to filtered entries only so that the
+	sum of shown entries will be always 100%. "absolute" means it retains
+	the original value before and after the filter is applied.
+
+-w::
+--column-widths=<width[,width...]>::
+	Force each column width to the provided list, for large terminal
+	readability.  0 means no limit (default behavior).
+
+
+INTERACTIVE PROMPTING KEYS
+--------------------------
+
+[d]::
+	Display refresh delay.
+
+[e]::
+	Number of entries to display.
+
+[E]::
+	Event to display when multiple counters are active.
+
+[f]::
+	Profile display filter (>= hit count).
+
+[F]::
+	Annotation display filter (>= % of total).
+
+[s]::
+	Annotate symbol.
+
+[S]::
+	Stop annotation, return to full profile display.
+
+[z]::
+	Toggle event count zeroing across display updates.
+
+[qQ]::
+	Quit.
+
+Pressing any unmapped key displays a menu, and prompts for input.
+
+
+SEE ALSO
+--------
+linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-trace.txt b/tools/perf/Documentation/perf-trace.txt
new file mode 100644
index 000000000..ba03fd5d1
--- /dev/null
+++ b/tools/perf/Documentation/perf-trace.txt
@@ -0,0 +1,164 @@
+perf-trace(1)
+=============
+
+NAME
+----
+perf-trace - strace inspired tool
+
+SYNOPSIS
+--------
+[verse]
+'perf trace'
+'perf trace record'
+
+DESCRIPTION
+-----------
+This command will show the events associated with the target, initially
+syscalls, but other system events like pagefaults, task lifetime events,
+scheduling events, etc.
+
+This is a live mode tool in addition to working with perf.data files like
+the other perf tools. Files can be generated using the 'perf record' command
+but the session needs to include the raw_syscalls events (-e 'raw_syscalls:*').
+Alternatively, 'perf trace record' can be used as a shortcut to
+automatically include the raw_syscalls events when writing events to a file.
+
+The following options apply to perf trace; options to perf trace record are
+found in the perf record man page.
+
+OPTIONS
+-------
+
+-a::
+--all-cpus::
+        System-wide collection from all CPUs.
+
+-e::
+--expr::
+	List of events to show, currently only syscall names.
+	Prefixing with ! shows all syscalls but the ones specified.  You may
+	need to escape it.
+
+-o::
+--output=::
+	Output file name.
+
+-p::
+--pid=::
+	Record events on existing process ID (comma separated list).
+
+-t::
+--tid=::
+        Record events on existing thread ID (comma separated list).
+
+-u::
+--uid=::
+        Record events in threads owned by uid. Name or number.
+
+--filter-pids=::
+	Filter out events for these pids and for 'trace' itself (comma separated list).
+
+-v::
+--verbose=::
+        Verbosity level.
+
+-i::
+--no-inherit::
+	Child tasks do not inherit counters.
+
+-m::
+--mmap-pages=::
+	Number of mmap data pages (must be a power of two) or size
+	specification with appended unit character - B/K/M/G. The
+	size is rounded up to have nearest pages power of two value.
+
+-C::
+--cpu::
+Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
+comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.
+In per-thread mode with inheritance mode on (default), Events are captured only when
+the thread executes on the designated CPUs. Default is to monitor all CPUs.
+
+--duration:
+	Show only events that had a duration greater than N.M ms.
+
+--sched:
+	Accrue thread runtime and provide a summary at the end of the session.
+
+-i
+--input
+	Process events from a given perf data file.
+
+-T
+--time
+	Print full timestamp rather time relative to first sample.
+
+--comm::
+        Show process COMM right beside its ID, on by default, disable with --no-comm.
+
+-s::
+--summary::
+	Show only a summary of syscalls by thread with min, max, and average times
+    (in msec) and relative stddev.
+
+-S::
+--with-summary::
+	Show all syscalls followed by a summary by thread with min, max, and
+    average times (in msec) and relative stddev.
+
+--tool_stats::
+	Show tool stats such as number of times fd->pathname was discovered thru
+	hooking the open syscall return + vfs_getname or via reading /proc/pid/fd, etc.
+
+-F=[all|min|maj]::
+--pf=[all|min|maj]::
+	Trace pagefaults. Optionally, you can specify whether you want minor,
+	major or all pagefaults. Default value is maj.
+
+--syscalls::
+	Trace system calls. This options is enabled by default.
+
+--event::
+	Trace other events, see 'perf list' for a complete list.
+
+PAGEFAULTS
+----------
+
+When tracing pagefaults, the format of the trace is as follows:
+
+<min|maj>fault [<ip.symbol>+<ip.offset>] => <addr.dso@addr.offset> (<map type><addr level>).
+
+- min/maj indicates whether fault event is minor or major;
+- ip.symbol shows symbol for instruction pointer (the code that generated the
+  fault); if no debug symbols available, perf trace will print raw IP;
+- addr.dso shows DSO for the faulted address;
+- map type is either 'd' for non-executable maps or 'x' for executable maps;
+- addr level is either 'k' for kernel dso or '.' for user dso.
+
+For symbols resolution you may need to install debugging symbols.
+
+Please be aware that duration is currently always 0 and doesn't reflect actual
+time it took for fault to be handled!
+
+When --verbose specified, perf trace tries to print all available information
+for both IP and fault address in the form of dso@symbol+offset.
+
+EXAMPLES
+--------
+
+Trace only major pagefaults:
+
+ $ perf trace --no-syscalls -F
+
+Trace syscalls, major and minor pagefaults:
+
+ $ perf trace -F all
+
+  1416.547 ( 0.000 ms): python/20235 majfault [CRYPTO_push_info_+0x0] => /lib/x86_64-linux-gnu/libcrypto.so.1.0.0@0x61be0 (x.)
+
+  As you can see, there was major pagefault in python process, from
+  CRYPTO_push_info_ routine which faulted somewhere in libcrypto.so.
+
+SEE ALSO
+--------
+linkperf:perf-record[1], linkperf:perf-script[1]
diff --git a/tools/perf/Documentation/perf.txt b/tools/perf/Documentation/perf.txt
new file mode 100644
index 000000000..2b1317763
--- /dev/null
+++ b/tools/perf/Documentation/perf.txt
@@ -0,0 +1,41 @@
+perf(1)
+=======
+
+NAME
+----
+perf - Performance analysis tools for Linux
+
+SYNOPSIS
+--------
+[verse]
+'perf' [--version] [--help] [OPTIONS] COMMAND [ARGS]
+
+OPTIONS
+-------
+--debug::
+	Setup debug variable (see list below) in value
+	range (0, 10). Use like:
+	  --debug verbose   # sets verbose = 1
+	  --debug verbose=2 # sets verbose = 2
+
+	List of debug variables allowed to set:
+	  verbose          - general debug messages
+	  ordered-events   - ordered events object debug messages
+	  data-convert     - data convert command debug messages
+
+--buildid-dir::
+	Setup buildid cache directory. It has higher priority than
+	buildid.dir config file option.
+
+DESCRIPTION
+-----------
+Performance counters for Linux are a new kernel-based subsystem
+that provide a framework for all things performance analysis. It
+covers hardware level (CPU/PMU, Performance Monitoring Unit) features
+and software features (software counters, tracepoints) as well.
+
+SEE ALSO
+--------
+linkperf:perf-stat[1], linkperf:perf-top[1],
+linkperf:perf-record[1], linkperf:perf-report[1],
+linkperf:perf-list[1]
diff --git a/tools/perf/Documentation/perfconfig.example b/tools/perf/Documentation/perfconfig.example
new file mode 100644
index 000000000..767ea2436
--- /dev/null
+++ b/tools/perf/Documentation/perfconfig.example
@@ -0,0 +1,29 @@
+[colors]
+
+	# These were the old defaults
+	top = red, lightgray
+	medium = green, lightgray
+	normal = black, lightgray
+	selected = lightgray, magenta
+	code = blue, lightgray
+	addr = magenta, lightgray
+
+[tui]
+
+	# Defaults if linked with libslang
+	report = on
+	annotate = on
+	top = on
+
+[buildid]
+
+	# Default, disable using /dev/null
+	dir = /root/.debug
+
+[annotate]
+
+	# Defaults
+	hide_src_code = false
+	use_offset = true
+	jump_arrows = true
+	show_nr_jumps = false
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
new file mode 100644
index 000000000..11ccbb22e
--- /dev/null
+++ b/tools/perf/MANIFEST
@@ -0,0 +1,63 @@
+tools/perf
+tools/scripts
+tools/build
+tools/lib/traceevent
+tools/lib/api
+tools/lib/symbol/kallsyms.c
+tools/lib/symbol/kallsyms.h
+tools/lib/util/find_next_bit.c
+tools/include/asm/bug.h
+tools/include/asm-generic/bitops/arch_hweight.h
+tools/include/asm-generic/bitops/atomic.h
+tools/include/asm-generic/bitops/const_hweight.h
+tools/include/asm-generic/bitops/__ffs.h
+tools/include/asm-generic/bitops/__fls.h
+tools/include/asm-generic/bitops/find.h
+tools/include/asm-generic/bitops/fls64.h
+tools/include/asm-generic/bitops/fls.h
+tools/include/asm-generic/bitops/hweight.h
+tools/include/asm-generic/bitops.h
+tools/include/linux/bitops.h
+tools/include/linux/compiler.h
+tools/include/linux/export.h
+tools/include/linux/hash.h
+tools/include/linux/log2.h
+tools/include/linux/types.h
+include/asm-generic/bitops/arch_hweight.h
+include/asm-generic/bitops/const_hweight.h
+include/asm-generic/bitops/fls64.h
+include/asm-generic/bitops/__fls.h
+include/asm-generic/bitops/fls.h
+include/linux/const.h
+include/linux/perf_event.h
+include/linux/rbtree.h
+include/linux/list.h
+include/linux/hash.h
+include/linux/stringify.h
+lib/find_next_bit.c
+lib/hweight.c
+lib/rbtree.c
+include/linux/swab.h
+arch/*/include/asm/unistd*.h
+arch/*/include/asm/perf_regs.h
+arch/*/include/uapi/asm/unistd*.h
+arch/*/include/uapi/asm/perf_regs.h
+arch/*/lib/memcpy*.S
+arch/*/lib/memset*.S
+include/linux/poison.h
+include/linux/magic.h
+include/linux/hw_breakpoint.h
+include/linux/rbtree_augmented.h
+include/uapi/linux/perf_event.h
+include/uapi/linux/const.h
+include/uapi/linux/swab.h
+include/uapi/linux/hw_breakpoint.h
+arch/x86/include/asm/svm.h
+arch/x86/include/asm/vmx.h
+arch/x86/include/asm/kvm_host.h
+arch/x86/include/uapi/asm/svm.h
+arch/x86/include/uapi/asm/vmx.h
+arch/x86/include/uapi/asm/kvm.h
+arch/x86/include/uapi/asm/kvm_perf.h
+arch/s390/include/uapi/asm/sie.h
+arch/s390/include/uapi/asm/kvm_perf.h
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
new file mode 100644
index 000000000..d31a7bbd7
--- /dev/null
+++ b/tools/perf/Makefile
@@ -0,0 +1,90 @@
+#
+# This is a simple wrapper Makefile that calls the main Makefile.perf
+# with a -j option to do parallel builds
+#
+# If you want to invoke the perf build in some non-standard way then
+# you can use the 'make -f Makefile.perf' method to invoke it.
+#
+
+#
+# Clear out the built-in rules GNU make defines by default (such as .o targets),
+# so that we pass through all targets to Makefile.perf:
+#
+.SUFFIXES:
+
+#
+# We don't want to pass along options like -j:
+#
+unexport MAKEFLAGS
+
+#
+# Do a parallel build with multiple jobs, based on the number of CPUs online
+# in this system: 'make -j8' on a 8-CPU system, etc.
+#
+# (To override it, run 'make JOBS=1' and similar.)
+#
+ifeq ($(JOBS),)
+  JOBS := $(shell (getconf _NPROCESSORS_ONLN || egrep -c '^processor|^CPU[0-9]' /proc/cpuinfo) 2>/dev/null)
+  ifeq ($(JOBS),0)
+    JOBS := 1
+  endif
+endif
+
+#
+# Only pass canonical directory names as the output directory:
+#
+ifneq ($(O),)
+  FULL_O := $(shell readlink -f $(O) || echo $(O))
+endif
+
+#
+# Only accept the 'DEBUG' variable from the command line:
+#
+ifeq ("$(origin DEBUG)", "command line")
+  ifeq ($(DEBUG),)
+    override DEBUG = 0
+  else
+    SET_DEBUG = "DEBUG=$(DEBUG)"
+  endif
+else
+  override DEBUG = 0
+endif
+
+define print_msg
+  @printf '  BUILD:   Doing '\''make \033[33m-j'$(JOBS)'\033[m'\'' parallel build\n'
+endef
+
+define make
+  @$(MAKE) -f Makefile.perf --no-print-directory -j$(JOBS) O=$(FULL_O) $(SET_DEBUG) $@
+endef
+
+#
+# Needed if no target specified:
+# (Except for tags and TAGS targets. The reason is that the
+# Makefile does not treat tags/TAGS as targets but as files
+# and thus won't rebuilt them once they are in place.)
+#
+all tags TAGS:
+	$(print_msg)
+	$(make)
+
+#
+# The clean target is not really parallel, don't print the jobs info:
+#
+clean:
+	$(make)
+
+#
+# The build-test target is not really parallel, don't print the jobs info:
+#
+build-test:
+	@$(MAKE) -f tests/make --no-print-directory
+
+#
+# All other targets get passed through:
+#
+%:
+	$(print_msg)
+	$(make)
+
+.PHONY: tags TAGS
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
new file mode 100644
index 000000000..c43a20517
--- /dev/null
+++ b/tools/perf/Makefile.perf
@@ -0,0 +1,552 @@
+include ../scripts/Makefile.include
+
+# The default target of this Makefile is...
+all:
+
+include config/utilities.mak
+
+# Define V to have a more verbose compile.
+#
+# Define VF to have a more verbose feature check output.
+#
+# Define O to save output files in a separate directory.
+#
+# Define ARCH as name of target architecture if you want cross-builds.
+#
+# Define CROSS_COMPILE as prefix name of compiler if you want cross-builds.
+#
+# Define NO_LIBPERL to disable perl script extension.
+#
+# Define NO_LIBPYTHON to disable python script extension.
+#
+# Define PYTHON to point to the python binary if the default
+# `python' is not correct; for example: PYTHON=python2
+#
+# Define PYTHON_CONFIG to point to the python-config binary if
+# the default `$(PYTHON)-config' is not correct.
+#
+# Define ASCIIDOC8 if you want to format documentation with AsciiDoc 8
+#
+# Define DOCBOOK_XSL_172 if you want to format man pages with DocBook XSL v1.72.
+#
+# Define LDFLAGS=-static to build a static binary.
+#
+# Define EXTRA_CFLAGS=-m64 or EXTRA_CFLAGS=-m32 as appropriate for cross-builds.
+#
+# Define NO_DWARF if you do not want debug-info analysis feature at all.
+#
+# Define WERROR=0 to disable treating any warnings as errors.
+#
+# Define NO_NEWT if you do not want TUI support. (deprecated)
+#
+# Define NO_SLANG if you do not want TUI support.
+#
+# Define NO_GTK2 if you do not want GTK+ GUI support.
+#
+# Define NO_DEMANGLE if you do not want C++ symbol demangling.
+#
+# Define NO_LIBELF if you do not want libelf dependency (e.g. cross-builds)
+#
+# Define NO_LIBUNWIND if you do not want libunwind dependency for dwarf
+# backtrace post unwind.
+#
+# Define NO_BACKTRACE if you do not want stack backtrace debug feature
+#
+# Define NO_LIBNUMA if you do not want numa perf benchmark
+#
+# Define NO_LIBAUDIT if you do not want libaudit support
+#
+# Define NO_LIBBIONIC if you do not want bionic support
+#
+# Define NO_LIBDW_DWARF_UNWIND if you do not want libdw support
+# for dwarf backtrace post unwind.
+#
+# Define NO_PERF_READ_VDSO32 if you do not want to build perf-read-vdso32
+# for reading the 32-bit compatibility VDSO in 64-bit mode
+#
+# Define NO_PERF_READ_VDSOX32 if you do not want to build perf-read-vdsox32
+# for reading the x32 mode 32-bit compatibility VDSO in 64-bit mode
+#
+# Define NO_ZLIB if you do not want to support compressed kernel modules
+#
+# Define LIBBABELTRACE if you DO want libbabeltrace support
+# for CTF data format.
+#
+# Define NO_LZMA if you do not want to support compressed (xz) kernel modules
+
+ifeq ($(srctree),)
+srctree := $(patsubst %/,%,$(dir $(shell pwd)))
+srctree := $(patsubst %/,%,$(dir $(srctree)))
+#$(info Determined 'srctree' to be $(srctree))
+endif
+
+ifneq ($(objtree),)
+#$(info Determined 'objtree' to be $(objtree))
+endif
+
+ifneq ($(OUTPUT),)
+#$(info Determined 'OUTPUT' to be $(OUTPUT))
+# Adding $(OUTPUT) as a directory to look for source files,
+# because use generated output files as sources dependency
+# for flex/bison parsers.
+VPATH += $(OUTPUT)
+export VPATH
+endif
+
+ifeq ($(V),1)
+  Q =
+else
+  Q = @
+endif
+
+# Do not use make's built-in rules
+# (this improves performance and avoids hard-to-debug behaviour);
+MAKEFLAGS += -r
+
+$(OUTPUT)PERF-VERSION-FILE: ../../.git/HEAD
+	$(Q)$(SHELL_PATH) util/PERF-VERSION-GEN $(OUTPUT)
+	$(Q)touch $(OUTPUT)PERF-VERSION-FILE
+
+CC = $(CROSS_COMPILE)gcc
+LD = $(CROSS_COMPILE)ld
+AR = $(CROSS_COMPILE)ar
+PKG_CONFIG = $(CROSS_COMPILE)pkg-config
+
+RM      = rm -f
+LN      = ln -f
+MKDIR   = mkdir
+FIND    = find
+INSTALL = install
+FLEX    = flex
+BISON   = bison
+STRIP   = strip
+
+LIB_DIR          = $(srctree)/tools/lib/api/
+TRACE_EVENT_DIR = $(srctree)/tools/lib/traceevent/
+
+# include config/Makefile by default and rule out
+# non-config cases
+config := 1
+
+NON_CONFIG_TARGETS := clean TAGS tags cscope help
+
+ifdef MAKECMDGOALS
+ifeq ($(filter-out $(NON_CONFIG_TARGETS),$(MAKECMDGOALS)),)
+  config := 0
+endif
+endif
+
+ifeq ($(config),1)
+include config/Makefile
+endif
+
+export prefix bindir sharedir sysconfdir DESTDIR
+
+# sparse is architecture-neutral, which means that we need to tell it
+# explicitly what architecture to check for. Fix this up for yours..
+SPARSE_FLAGS = -D__BIG_ENDIAN__ -D__powerpc__
+
+# Guard against environment variables
+PYRF_OBJS =
+SCRIPT_SH =
+
+SCRIPT_SH += perf-archive.sh
+SCRIPT_SH += perf-with-kcore.sh
+
+grep-libs = $(filter -l%,$(1))
+strip-libs = $(filter-out -l%,$(1))
+
+ifneq ($(OUTPUT),)
+  TE_PATH=$(OUTPUT)
+ifneq ($(subdir),)
+  LIB_PATH=$(OUTPUT)/../lib/api/
+else
+  LIB_PATH=$(OUTPUT)
+endif
+else
+  TE_PATH=$(TRACE_EVENT_DIR)
+  LIB_PATH=$(LIB_DIR)
+endif
+
+LIBTRACEEVENT = $(TE_PATH)libtraceevent.a
+export LIBTRACEEVENT
+
+LIBAPI = $(LIB_PATH)libapi.a
+export LIBAPI
+
+# python extension build directories
+PYTHON_EXTBUILD     := $(OUTPUT)python_ext_build/
+PYTHON_EXTBUILD_LIB := $(PYTHON_EXTBUILD)lib/
+PYTHON_EXTBUILD_TMP := $(PYTHON_EXTBUILD)tmp/
+export PYTHON_EXTBUILD_LIB PYTHON_EXTBUILD_TMP
+
+python-clean := $(call QUIET_CLEAN, python) $(RM) -r $(PYTHON_EXTBUILD) $(OUTPUT)python/perf.so
+
+PYTHON_EXT_SRCS := $(shell grep -v ^\# util/python-ext-sources)
+PYTHON_EXT_DEPS := util/python-ext-sources util/setup.py $(LIBTRACEEVENT) $(LIBAPI)
+
+$(OUTPUT)python/perf.so: $(PYTHON_EXT_SRCS) $(PYTHON_EXT_DEPS)
+	$(QUIET_GEN)CFLAGS='$(CFLAGS)' $(PYTHON_WORD) util/setup.py \
+	  --quiet build_ext; \
+	mkdir -p $(OUTPUT)python && \
+	cp $(PYTHON_EXTBUILD_LIB)perf.so $(OUTPUT)python/
+#
+# No Perl scripts right now:
+#
+
+SCRIPTS = $(patsubst %.sh,%,$(SCRIPT_SH))
+
+PROGRAMS += $(OUTPUT)perf
+
+ifndef NO_PERF_READ_VDSO32
+PROGRAMS += $(OUTPUT)perf-read-vdso32
+endif
+
+ifndef NO_PERF_READ_VDSOX32
+PROGRAMS += $(OUTPUT)perf-read-vdsox32
+endif
+
+# what 'all' will build and 'install' will install, in perfexecdir
+ALL_PROGRAMS = $(PROGRAMS) $(SCRIPTS)
+
+# what 'all' will build but not install in perfexecdir
+OTHER_PROGRAMS = $(OUTPUT)perf
+
+# Set paths to tools early so that they can be used for version tests.
+ifndef SHELL_PATH
+  SHELL_PATH = /bin/sh
+endif
+ifndef PERL_PATH
+  PERL_PATH = /usr/bin/perl
+endif
+
+export PERL_PATH
+
+LIB_FILE=$(OUTPUT)libperf.a
+
+PERFLIBS = $(LIB_FILE) $(LIBAPI) $(LIBTRACEEVENT)
+
+# We choose to avoid "if .. else if .. else .. endif endif"
+# because maintaining the nesting to match is a pain.  If
+# we had "elif" things would have been much nicer...
+
+-include arch/$(ARCH)/Makefile
+
+ifneq ($(OUTPUT),)
+  CFLAGS += -I$(OUTPUT)
+endif
+
+ifndef NO_GTK2
+  ALL_PROGRAMS += $(OUTPUT)libperf-gtk.so
+  GTK_IN := $(OUTPUT)gtk-in.o
+
+install-gtk: $(OUTPUT)libperf-gtk.so
+	$(call QUIET_INSTALL, 'GTK UI') \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(libdir_SQ)'; \
+		$(INSTALL) $(OUTPUT)libperf-gtk.so '$(DESTDIR_SQ)$(libdir_SQ)'
+endif
+
+ifdef ASCIIDOC8
+  export ASCIIDOC8
+endif
+
+LIBS = -Wl,--whole-archive $(PERFLIBS) -Wl,--no-whole-archive -Wl,--start-group $(EXTLIBS) -Wl,--end-group
+
+export INSTALL SHELL_PATH
+
+### Build rules
+
+SHELL = $(SHELL_PATH)
+
+all: shell_compatibility_test $(ALL_PROGRAMS) $(LANG_BINDINGS) $(OTHER_PROGRAMS)
+
+please_set_SHELL_PATH_to_a_more_modern_shell:
+	$(Q)$$(:)
+
+shell_compatibility_test: please_set_SHELL_PATH_to_a_more_modern_shell
+
+strip: $(PROGRAMS) $(OUTPUT)perf
+	$(STRIP) $(STRIP_OPTS) $(PROGRAMS) $(OUTPUT)perf
+
+PERF_IN := $(OUTPUT)perf-in.o
+
+export srctree OUTPUT RM CC LD AR CFLAGS V BISON FLEX
+build := -f $(srctree)/tools/build/Makefile.build dir=. obj
+
+$(PERF_IN): $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)common-cmds.h FORCE
+	$(Q)$(MAKE) $(build)=perf
+
+$(OUTPUT)perf: $(PERFLIBS) $(PERF_IN)
+	$(QUIET_LINK)$(CC) $(CFLAGS) $(LDFLAGS) $(PERF_IN) $(LIBS) -o $@
+
+$(GTK_IN): FORCE
+	$(Q)$(MAKE) $(build)=gtk
+
+$(OUTPUT)libperf-gtk.so: $(GTK_IN) $(PERFLIBS)
+	$(QUIET_LINK)$(CC) -o $@ -shared $(LDFLAGS) $(filter %.o,$^) $(GTK_LIBS)
+
+$(OUTPUT)common-cmds.h: util/generate-cmdlist.sh command-list.txt
+
+$(OUTPUT)common-cmds.h: $(wildcard Documentation/perf-*.txt)
+	$(QUIET_GEN). util/generate-cmdlist.sh > $@+ && mv $@+ $@
+
+$(SCRIPTS) : % : %.sh
+	$(QUIET_GEN)$(INSTALL) '$@.sh' '$(OUTPUT)$@'
+
+# These can record PERF_VERSION
+perf.spec $(SCRIPTS) \
+	: $(OUTPUT)PERF-VERSION-FILE
+
+.SUFFIXES:
+
+#
+# If a target does not match any of the later rules then prefix it by $(OUTPUT)
+# This makes targets like 'make O=/tmp/perf perf.o' work in a natural way.
+#
+ifneq ($(OUTPUT),)
+%.o: $(OUTPUT)%.o
+	@echo "    # Redirected target $@ => $(OUTPUT)$@"
+util/%.o: $(OUTPUT)util/%.o
+	@echo "    # Redirected target $@ => $(OUTPUT)$@"
+bench/%.o: $(OUTPUT)bench/%.o
+	@echo "    # Redirected target $@ => $(OUTPUT)$@"
+tests/%.o: $(OUTPUT)tests/%.o
+	@echo "    # Redirected target $@ => $(OUTPUT)$@"
+endif
+
+# These two need to be here so that when O= is not used they take precedence
+# over the general rule for .o
+
+# get relative building directory (to $(OUTPUT))
+# and '.' if it's $(OUTPUT) itself
+__build-dir = $(subst $(OUTPUT),,$(dir $@))
+build-dir   = $(if $(__build-dir),$(__build-dir),.)
+
+single_dep: $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)common-cmds.h
+
+$(OUTPUT)%.o: %.c single_dep FORCE
+	$(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=$(build-dir) $@
+
+$(OUTPUT)%.i: %.c single_dep FORCE
+	$(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=$(build-dir) $@
+
+$(OUTPUT)%.s: %.c single_dep FORCE
+	$(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=$(build-dir) $@
+
+$(OUTPUT)%-bison.o: %.c single_dep FORCE
+	$(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=$(build-dir) $@
+
+$(OUTPUT)%-flex.o: %.c single_dep FORCE
+	$(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=$(build-dir) $@
+
+$(OUTPUT)%.o: %.S single_dep FORCE
+	$(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=$(build-dir) $@
+
+$(OUTPUT)%.i: %.S single_dep FORCE
+	$(Q)$(MAKE) -f $(srctree)/tools/build/Makefile.build dir=$(build-dir) $@
+
+$(OUTPUT)perf-%: %.o $(PERFLIBS)
+	$(QUIET_LINK)$(CC) $(CFLAGS) -o $@ $(LDFLAGS) $(filter %.o,$^) $(LIBS)
+
+ifndef NO_PERF_READ_VDSO32
+$(OUTPUT)perf-read-vdso32: perf-read-vdso.c util/find-vdso-map.c
+	$(QUIET_CC)$(CC) -m32 $(filter -static,$(LDFLAGS)) -Wall -Werror -o $@ perf-read-vdso.c
+endif
+
+ifndef NO_PERF_READ_VDSOX32
+$(OUTPUT)perf-read-vdsox32: perf-read-vdso.c util/find-vdso-map.c
+	$(QUIET_CC)$(CC) -mx32 $(filter -static,$(LDFLAGS)) -Wall -Werror -o $@ perf-read-vdso.c
+endif
+
+$(patsubst perf-%,%.o,$(PROGRAMS)): $(wildcard */*.h)
+
+LIBPERF_IN := $(OUTPUT)libperf-in.o
+
+$(LIBPERF_IN): FORCE
+	$(Q)$(MAKE) $(build)=libperf
+
+$(LIB_FILE): $(LIBPERF_IN)
+	$(QUIET_AR)$(RM) $@ && $(AR) rcs $@ $(LIBPERF_IN) $(LIB_OBJS)
+
+LIBTRACEEVENT_FLAGS += plugin_dir=$(plugindir_SQ)
+
+$(LIBTRACEEVENT): FORCE
+	$(Q)$(MAKE) -C $(TRACE_EVENT_DIR) $(LIBTRACEEVENT_FLAGS) O=$(OUTPUT) $(OUTPUT)libtraceevent.a plugins
+
+$(LIBTRACEEVENT)-clean:
+	$(call QUIET_CLEAN, libtraceevent)
+	$(Q)$(MAKE) -C $(TRACE_EVENT_DIR) O=$(OUTPUT) clean >/dev/null
+
+install-traceevent-plugins: $(LIBTRACEEVENT)
+	$(Q)$(MAKE) -C $(TRACE_EVENT_DIR) $(LIBTRACEEVENT_FLAGS) O=$(OUTPUT) install_plugins
+
+$(LIBAPI): FORCE
+	$(Q)$(MAKE) -C $(LIB_DIR) O=$(OUTPUT) $(OUTPUT)libapi.a
+
+$(LIBAPI)-clean:
+	$(call QUIET_CLEAN, libapi)
+	$(Q)$(MAKE) -C $(LIB_DIR) O=$(OUTPUT) clean >/dev/null
+
+help:
+	@echo 'Perf make targets:'
+	@echo '  doc		- make *all* documentation (see below)'
+	@echo '  man		- make manpage documentation (access with man <foo>)'
+	@echo '  html		- make html documentation'
+	@echo '  info		- make GNU info documentation (access with info <foo>)'
+	@echo '  pdf		- make pdf documentation'
+	@echo '  TAGS		- use etags to make tag information for source browsing'
+	@echo '  tags		- use ctags to make tag information for source browsing'
+	@echo '  cscope	- use cscope to make interactive browsing database'
+	@echo ''
+	@echo 'Perf install targets:'
+	@echo '  NOTE: documentation build requires asciidoc, xmlto packages to be installed'
+	@echo '  HINT: use "prefix" or "DESTDIR" to install to a particular'
+	@echo '        path like "make prefix=/usr/local install install-doc"'
+	@echo '  install	- install compiled binaries'
+	@echo '  install-doc	- install *all* documentation'
+	@echo '  install-man	- install manpage documentation'
+	@echo '  install-html	- install html documentation'
+	@echo '  install-info	- install GNU info documentation'
+	@echo '  install-pdf	- install pdf documentation'
+	@echo ''
+	@echo '  quick-install-doc	- alias for quick-install-man'
+	@echo '  quick-install-man	- install the documentation quickly'
+	@echo '  quick-install-html	- install the html documentation quickly'
+	@echo ''
+	@echo 'Perf maintainer targets:'
+	@echo '  clean			- clean all binary objects and build output'
+
+
+DOC_TARGETS := doc man html info pdf
+
+INSTALL_DOC_TARGETS := $(patsubst %,install-%,$(DOC_TARGETS)) try-install-man
+INSTALL_DOC_TARGETS += quick-install-doc quick-install-man quick-install-html
+
+# 'make doc' should call 'make -C Documentation all'
+$(DOC_TARGETS):
+	$(QUIET_SUBDIR0)Documentation $(QUIET_SUBDIR1) $(@:doc=all)
+
+TAG_FOLDERS= . ../lib/traceevent ../lib/api ../lib/symbol
+TAG_FILES= ../../include/uapi/linux/perf_event.h
+
+TAGS:
+	$(QUIET_GEN)$(RM) TAGS; \
+	$(FIND) $(TAG_FOLDERS) -name '*.[hcS]' -print | xargs etags -a $(TAG_FILES)
+
+tags:
+	$(QUIET_GEN)$(RM) tags; \
+	$(FIND) $(TAG_FOLDERS) -name '*.[hcS]' -print | xargs ctags -a $(TAG_FILES)
+
+cscope:
+	$(QUIET_GEN)$(RM) cscope*; \
+	$(FIND) $(TAG_FOLDERS) -name '*.[hcS]' -print | xargs cscope -b $(TAG_FILES)
+
+### Testing rules
+
+# GNU make supports exporting all variables by "export" without parameters.
+# However, the environment gets quite big, and some programs have problems
+# with that.
+
+check: $(OUTPUT)common-cmds.h
+	if sparse; \
+	then \
+		for i in *.c */*.c; \
+		do \
+			sparse $(CFLAGS) $(SPARSE_FLAGS) $$i || exit; \
+		done; \
+	else \
+		exit 1; \
+	fi
+
+### Installation rules
+
+install-gtk:
+
+install-bin: all install-gtk
+	$(call QUIET_INSTALL, binaries) \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(bindir_SQ)'; \
+		$(INSTALL) $(OUTPUT)perf '$(DESTDIR_SQ)$(bindir_SQ)'; \
+		$(LN) '$(DESTDIR_SQ)$(bindir_SQ)/perf' '$(DESTDIR_SQ)$(bindir_SQ)/trace'
+ifndef NO_PERF_READ_VDSO32
+	$(call QUIET_INSTALL, perf-read-vdso32) \
+		$(INSTALL) $(OUTPUT)perf-read-vdso32 '$(DESTDIR_SQ)$(bindir_SQ)';
+endif
+ifndef NO_PERF_READ_VDSOX32
+	$(call QUIET_INSTALL, perf-read-vdsox32) \
+		$(INSTALL) $(OUTPUT)perf-read-vdsox32 '$(DESTDIR_SQ)$(bindir_SQ)';
+endif
+	$(call QUIET_INSTALL, libexec) \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
+	$(call QUIET_INSTALL, perf-archive) \
+		$(INSTALL) $(OUTPUT)perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
+	$(call QUIET_INSTALL, perf-with-kcore) \
+		$(INSTALL) $(OUTPUT)perf-with-kcore -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
+ifndef NO_LIBPERL
+	$(call QUIET_INSTALL, perl-scripts) \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'; \
+		$(INSTALL) scripts/perl/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'; \
+		$(INSTALL) scripts/perl/*.pl -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl'; \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'; \
+		$(INSTALL) scripts/perl/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'
+endif
+ifndef NO_LIBPYTHON
+	$(call QUIET_INSTALL, python-scripts) \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/Perf-Trace-Util/lib/Perf/Trace'; \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/bin'; \
+		$(INSTALL) scripts/python/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/Perf-Trace-Util/lib/Perf/Trace'; \
+		$(INSTALL) scripts/python/*.py -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python'; \
+		$(INSTALL) scripts/python/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/bin'
+endif
+	$(call QUIET_INSTALL, perf_completion-script) \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(sysconfdir_SQ)/bash_completion.d'; \
+		$(INSTALL) perf-completion.sh '$(DESTDIR_SQ)$(sysconfdir_SQ)/bash_completion.d/perf'
+	$(call QUIET_INSTALL, tests) \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests'; \
+		$(INSTALL) tests/attr.py '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests'; \
+		$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests/attr'; \
+		$(INSTALL) tests/attr/* '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests/attr'
+
+install: install-bin try-install-man install-traceevent-plugins
+
+install-python_ext:
+	$(PYTHON_WORD) util/setup.py --quiet install --root='/$(DESTDIR_SQ)'
+
+# 'make install-doc' should call 'make -C Documentation install'
+$(INSTALL_DOC_TARGETS):
+	$(QUIET_SUBDIR0)Documentation $(QUIET_SUBDIR1) $(@:-doc=)
+
+### Cleaning rules
+
+#
+# This is here, not in config/Makefile, because config/Makefile does
+# not get included for the clean target:
+#
+config-clean:
+	$(call QUIET_CLEAN, config)
+	$(Q)$(MAKE) -C $(srctree)/tools/build/feature/ clean >/dev/null
+
+clean: $(LIBTRACEEVENT)-clean $(LIBAPI)-clean config-clean
+	$(call QUIET_CLEAN, core-objs)  $(RM) $(LIB_FILE) $(OUTPUT)perf-archive $(OUTPUT)perf-with-kcore $(LANG_BINDINGS)
+	$(Q)find . -name '*.o' -delete -o -name '\.*.cmd' -delete -o -name '\.*.d' -delete
+	$(Q)$(RM) .config-detected
+	$(call QUIET_CLEAN, core-progs) $(RM) $(ALL_PROGRAMS) perf perf-read-vdso32 perf-read-vdsox32
+	$(call QUIET_CLEAN, core-gen)   $(RM)  *.spec *.pyc *.pyo */*.pyc */*.pyo $(OUTPUT)common-cmds.h TAGS tags cscope* $(OUTPUT)PERF-VERSION-FILE $(OUTPUT)FEATURE-DUMP $(OUTPUT)util/*-bison* $(OUTPUT)util/*-flex*
+	$(QUIET_SUBDIR0)Documentation $(QUIET_SUBDIR1) clean
+	$(python-clean)
+
+#
+# Trick: if ../../.git does not exist - we are building out of tree for example,
+# then force version regeneration:
+#
+ifeq ($(wildcard ../../.git/HEAD),)
+    GIT-HEAD-PHONY = ../../.git/HEAD
+else
+    GIT-HEAD-PHONY =
+endif
+
+FORCE:
+
+.PHONY: all install clean config-clean strip install-gtk
+.PHONY: shell_compatibility_test please_set_SHELL_PATH_to_a_more_modern_shell
+.PHONY: $(GIT-HEAD-PHONY) TAGS tags cscope FORCE single_dep
+
diff --git a/tools/perf/arch/Build b/tools/perf/arch/Build
new file mode 100644
index 000000000..109eb75cf
--- /dev/null
+++ b/tools/perf/arch/Build
@@ -0,0 +1,2 @@
+libperf-y += common.o
+libperf-y += $(ARCH)/
diff --git a/tools/perf/arch/arm/Build b/tools/perf/arch/arm/Build
new file mode 100644
index 000000000..41bf61da4
--- /dev/null
+++ b/tools/perf/arch/arm/Build
@@ -0,0 +1,2 @@
+libperf-y += util/
+libperf-$(CONFIG_DWARF_UNWIND) += tests/
diff --git a/tools/perf/arch/arm/Makefile b/tools/perf/arch/arm/Makefile
new file mode 100644
index 000000000..7fbca1750
--- /dev/null
+++ b/tools/perf/arch/arm/Makefile
@@ -0,0 +1,3 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+endif
diff --git a/tools/perf/arch/arm/include/perf_regs.h b/tools/perf/arch/arm/include/perf_regs.h
new file mode 100644
index 000000000..f619c9c5a
--- /dev/null
+++ b/tools/perf/arch/arm/include/perf_regs.h
@@ -0,0 +1,59 @@
+#ifndef ARCH_PERF_REGS_H
+#define ARCH_PERF_REGS_H
+
+#include <stdlib.h>
+#include <linux/types.h>
+#include <asm/perf_regs.h>
+
+void perf_regs_load(u64 *regs);
+
+#define PERF_REGS_MASK	((1ULL << PERF_REG_ARM_MAX) - 1)
+#define PERF_REGS_MAX	PERF_REG_ARM_MAX
+#define PERF_SAMPLE_REGS_ABI	PERF_SAMPLE_REGS_ABI_32
+
+#define PERF_REG_IP	PERF_REG_ARM_PC
+#define PERF_REG_SP	PERF_REG_ARM_SP
+
+static inline const char *perf_reg_name(int id)
+{
+	switch (id) {
+	case PERF_REG_ARM_R0:
+		return "r0";
+	case PERF_REG_ARM_R1:
+		return "r1";
+	case PERF_REG_ARM_R2:
+		return "r2";
+	case PERF_REG_ARM_R3:
+		return "r3";
+	case PERF_REG_ARM_R4:
+		return "r4";
+	case PERF_REG_ARM_R5:
+		return "r5";
+	case PERF_REG_ARM_R6:
+		return "r6";
+	case PERF_REG_ARM_R7:
+		return "r7";
+	case PERF_REG_ARM_R8:
+		return "r8";
+	case PERF_REG_ARM_R9:
+		return "r9";
+	case PERF_REG_ARM_R10:
+		return "r10";
+	case PERF_REG_ARM_FP:
+		return "fp";
+	case PERF_REG_ARM_IP:
+		return "ip";
+	case PERF_REG_ARM_SP:
+		return "sp";
+	case PERF_REG_ARM_LR:
+		return "lr";
+	case PERF_REG_ARM_PC:
+		return "pc";
+	default:
+		return NULL;
+	}
+
+	return NULL;
+}
+
+#endif /* ARCH_PERF_REGS_H */
diff --git a/tools/perf/arch/arm/tests/Build b/tools/perf/arch/arm/tests/Build
new file mode 100644
index 000000000..b30eff9bc
--- /dev/null
+++ b/tools/perf/arch/arm/tests/Build
@@ -0,0 +1,2 @@
+libperf-y += regs_load.o
+libperf-y += dwarf-unwind.o
diff --git a/tools/perf/arch/arm/tests/dwarf-unwind.c b/tools/perf/arch/arm/tests/dwarf-unwind.c
new file mode 100644
index 000000000..62eff847f
--- /dev/null
+++ b/tools/perf/arch/arm/tests/dwarf-unwind.c
@@ -0,0 +1,61 @@
+#include <string.h>
+#include "perf_regs.h"
+#include "thread.h"
+#include "map.h"
+#include "event.h"
+#include "debug.h"
+#include "tests/tests.h"
+
+#define STACK_SIZE 8192
+
+static int sample_ustack(struct perf_sample *sample,
+			 struct thread *thread, u64 *regs)
+{
+	struct stack_dump *stack = &sample->user_stack;
+	struct map *map;
+	unsigned long sp;
+	u64 stack_size, *buf;
+
+	buf = malloc(STACK_SIZE);
+	if (!buf) {
+		pr_debug("failed to allocate sample uregs data\n");
+		return -1;
+	}
+
+	sp = (unsigned long) regs[PERF_REG_ARM_SP];
+
+	map = map_groups__find(thread->mg, MAP__VARIABLE, (u64) sp);
+	if (!map) {
+		pr_debug("failed to get stack map\n");
+		free(buf);
+		return -1;
+	}
+
+	stack_size = map->end - sp;
+	stack_size = stack_size > STACK_SIZE ? STACK_SIZE : stack_size;
+
+	memcpy(buf, (void *) sp, stack_size);
+	stack->data = (char *) buf;
+	stack->size = stack_size;
+	return 0;
+}
+
+int test__arch_unwind_sample(struct perf_sample *sample,
+			     struct thread *thread)
+{
+	struct regs_dump *regs = &sample->user_regs;
+	u64 *buf;
+
+	buf = calloc(1, sizeof(u64) * PERF_REGS_MAX);
+	if (!buf) {
+		pr_debug("failed to allocate sample uregs data\n");
+		return -1;
+	}
+
+	perf_regs_load(buf);
+	regs->abi  = PERF_SAMPLE_REGS_ABI;
+	regs->regs = buf;
+	regs->mask = PERF_REGS_MASK;
+
+	return sample_ustack(sample, thread, buf);
+}
diff --git a/tools/perf/arch/arm/tests/regs_load.S b/tools/perf/arch/arm/tests/regs_load.S
new file mode 100644
index 000000000..e09e98394
--- /dev/null
+++ b/tools/perf/arch/arm/tests/regs_load.S
@@ -0,0 +1,58 @@
+#include <linux/linkage.h>
+
+#define R0 0x00
+#define R1 0x08
+#define R2 0x10
+#define R3 0x18
+#define R4 0x20
+#define R5 0x28
+#define R6 0x30
+#define R7 0x38
+#define R8 0x40
+#define R9 0x48
+#define SL 0x50
+#define FP 0x58
+#define IP 0x60
+#define SP 0x68
+#define LR 0x70
+#define PC 0x78
+
+/*
+ * Implementation of void perf_regs_load(u64 *regs);
+ *
+ * This functions fills in the 'regs' buffer from the actual registers values,
+ * in the way the perf built-in unwinding test expects them:
+ * - the PC at the time at the call to this function. Since this function
+ *   is called using a bl instruction, the PC value is taken from LR.
+ * The built-in unwinding test then unwinds the call stack from the dwarf
+ * information in unwind__get_entries.
+ *
+ * Notes:
+ * - the 8 bytes stride in the registers offsets comes from the fact
+ * that the registers are stored in an u64 array (u64 *regs),
+ * - the regs buffer needs to be zeroed before the call to this function,
+ * in this case using a calloc in dwarf-unwind.c.
+ */
+
+.text
+.type perf_regs_load,%function
+ENTRY(perf_regs_load)
+	str r0, [r0, #R0]
+	str r1, [r0, #R1]
+	str r2, [r0, #R2]
+	str r3, [r0, #R3]
+	str r4, [r0, #R4]
+	str r5, [r0, #R5]
+	str r6, [r0, #R6]
+	str r7, [r0, #R7]
+	str r8, [r0, #R8]
+	str r9, [r0, #R9]
+	str sl, [r0, #SL]
+	str fp, [r0, #FP]
+	str ip, [r0, #IP]
+	str sp, [r0, #SP]
+	str lr, [r0, #LR]
+	str lr, [r0, #PC]	// store pc as lr in order to skip the call
+	                        //  to this function
+	mov pc, lr
+ENDPROC(perf_regs_load)
diff --git a/tools/perf/arch/arm/util/Build b/tools/perf/arch/arm/util/Build
new file mode 100644
index 000000000..d22e3d07d
--- /dev/null
+++ b/tools/perf/arch/arm/util/Build
@@ -0,0 +1,4 @@
+libperf-$(CONFIG_DWARF) += dwarf-regs.o
+
+libperf-$(CONFIG_LIBUNWIND)          += unwind-libunwind.o
+libperf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o
diff --git a/tools/perf/arch/arm/util/dwarf-regs.c b/tools/perf/arch/arm/util/dwarf-regs.c
new file mode 100644
index 000000000..33ec5b339
--- /dev/null
+++ b/tools/perf/arch/arm/util/dwarf-regs.c
@@ -0,0 +1,64 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ * Copyright (C) 2010 Will Deacon, ARM Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <stddef.h>
+#include <dwarf-regs.h>
+
+struct pt_regs_dwarfnum {
+	const char *name;
+	unsigned int dwarfnum;
+};
+
+#define STR(s) #s
+#define REG_DWARFNUM_NAME(r, num) {.name = r, .dwarfnum = num}
+#define GPR_DWARFNUM_NAME(num) \
+	{.name = STR(%r##num), .dwarfnum = num}
+#define REG_DWARFNUM_END {.name = NULL, .dwarfnum = 0}
+
+/*
+ * Reference:
+ * http://infocenter.arm.com/help/topic/com.arm.doc.ihi0040a/IHI0040A_aadwarf.pdf
+ */
+static const struct pt_regs_dwarfnum regdwarfnum_table[] = {
+	GPR_DWARFNUM_NAME(0),
+	GPR_DWARFNUM_NAME(1),
+	GPR_DWARFNUM_NAME(2),
+	GPR_DWARFNUM_NAME(3),
+	GPR_DWARFNUM_NAME(4),
+	GPR_DWARFNUM_NAME(5),
+	GPR_DWARFNUM_NAME(6),
+	GPR_DWARFNUM_NAME(7),
+	GPR_DWARFNUM_NAME(8),
+	GPR_DWARFNUM_NAME(9),
+	GPR_DWARFNUM_NAME(10),
+	REG_DWARFNUM_NAME("%fp", 11),
+	REG_DWARFNUM_NAME("%ip", 12),
+	REG_DWARFNUM_NAME("%sp", 13),
+	REG_DWARFNUM_NAME("%lr", 14),
+	REG_DWARFNUM_NAME("%pc", 15),
+	REG_DWARFNUM_END,
+};
+
+/**
+ * get_arch_regstr() - lookup register name from it's DWARF register number
+ * @n:	the DWARF register number
+ *
+ * get_arch_regstr() returns the name of the register in struct
+ * regdwarfnum_table from it's DWARF register number. If the register is not
+ * found in the table, this returns NULL;
+ */
+const char *get_arch_regstr(unsigned int n)
+{
+	const struct pt_regs_dwarfnum *roff;
+	for (roff = regdwarfnum_table; roff->name != NULL; roff++)
+		if (roff->dwarfnum == n)
+			return roff->name;
+	return NULL;
+}
diff --git a/tools/perf/arch/arm/util/unwind-libdw.c b/tools/perf/arch/arm/util/unwind-libdw.c
new file mode 100644
index 000000000..b4176c601
--- /dev/null
+++ b/tools/perf/arch/arm/util/unwind-libdw.c
@@ -0,0 +1,36 @@
+#include <elfutils/libdwfl.h>
+#include "../../util/unwind-libdw.h"
+#include "../../util/perf_regs.h"
+
+bool libdw__arch_set_initial_registers(Dwfl_Thread *thread, void *arg)
+{
+	struct unwind_info *ui = arg;
+	struct regs_dump *user_regs = &ui->sample->user_regs;
+	Dwarf_Word dwarf_regs[PERF_REG_ARM_MAX];
+
+#define REG(r) ({						\
+	Dwarf_Word val = 0;					\
+	perf_reg_value(&val, user_regs, PERF_REG_ARM_##r);	\
+	val;							\
+})
+
+	dwarf_regs[0]  = REG(R0);
+	dwarf_regs[1]  = REG(R1);
+	dwarf_regs[2]  = REG(R2);
+	dwarf_regs[3]  = REG(R3);
+	dwarf_regs[4]  = REG(R4);
+	dwarf_regs[5]  = REG(R5);
+	dwarf_regs[6]  = REG(R6);
+	dwarf_regs[7]  = REG(R7);
+	dwarf_regs[8]  = REG(R8);
+	dwarf_regs[9]  = REG(R9);
+	dwarf_regs[10] = REG(R10);
+	dwarf_regs[11] = REG(FP);
+	dwarf_regs[12] = REG(IP);
+	dwarf_regs[13] = REG(SP);
+	dwarf_regs[14] = REG(LR);
+	dwarf_regs[15] = REG(PC);
+
+	return dwfl_thread_state_registers(thread, 0, PERF_REG_ARM_MAX,
+					   dwarf_regs);
+}
diff --git a/tools/perf/arch/arm/util/unwind-libunwind.c b/tools/perf/arch/arm/util/unwind-libunwind.c
new file mode 100644
index 000000000..62c397ed3
--- /dev/null
+++ b/tools/perf/arch/arm/util/unwind-libunwind.c
@@ -0,0 +1,49 @@
+
+#include <errno.h>
+#include <libunwind.h>
+#include "perf_regs.h"
+#include "../../util/unwind.h"
+#include "../../util/debug.h"
+
+int libunwind__arch_reg_id(int regnum)
+{
+	switch (regnum) {
+	case UNW_ARM_R0:
+		return PERF_REG_ARM_R0;
+	case UNW_ARM_R1:
+		return PERF_REG_ARM_R1;
+	case UNW_ARM_R2:
+		return PERF_REG_ARM_R2;
+	case UNW_ARM_R3:
+		return PERF_REG_ARM_R3;
+	case UNW_ARM_R4:
+		return PERF_REG_ARM_R4;
+	case UNW_ARM_R5:
+		return PERF_REG_ARM_R5;
+	case UNW_ARM_R6:
+		return PERF_REG_ARM_R6;
+	case UNW_ARM_R7:
+		return PERF_REG_ARM_R7;
+	case UNW_ARM_R8:
+		return PERF_REG_ARM_R8;
+	case UNW_ARM_R9:
+		return PERF_REG_ARM_R9;
+	case UNW_ARM_R10:
+		return PERF_REG_ARM_R10;
+	case UNW_ARM_R11:
+		return PERF_REG_ARM_FP;
+	case UNW_ARM_R12:
+		return PERF_REG_ARM_IP;
+	case UNW_ARM_R13:
+		return PERF_REG_ARM_SP;
+	case UNW_ARM_R14:
+		return PERF_REG_ARM_LR;
+	case UNW_ARM_R15:
+		return PERF_REG_ARM_PC;
+	default:
+		pr_err("unwind: invalid reg id %d\n", regnum);
+		return -EINVAL;
+	}
+
+	return -EINVAL;
+}
diff --git a/tools/perf/arch/arm64/Build b/tools/perf/arch/arm64/Build
new file mode 100644
index 000000000..54afe4a46
--- /dev/null
+++ b/tools/perf/arch/arm64/Build
@@ -0,0 +1 @@
+libperf-y += util/
diff --git a/tools/perf/arch/arm64/Makefile b/tools/perf/arch/arm64/Makefile
new file mode 100644
index 000000000..7fbca1750
--- /dev/null
+++ b/tools/perf/arch/arm64/Makefile
@@ -0,0 +1,3 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+endif
diff --git a/tools/perf/arch/arm64/include/perf_regs.h b/tools/perf/arch/arm64/include/perf_regs.h
new file mode 100644
index 000000000..1d3f39c3a
--- /dev/null
+++ b/tools/perf/arch/arm64/include/perf_regs.h
@@ -0,0 +1,90 @@
+#ifndef ARCH_PERF_REGS_H
+#define ARCH_PERF_REGS_H
+
+#include <stdlib.h>
+#include <linux/types.h>
+#include <asm/perf_regs.h>
+
+#define PERF_REGS_MASK	((1ULL << PERF_REG_ARM64_MAX) - 1)
+#define PERF_REGS_MAX	PERF_REG_ARM64_MAX
+
+#define PERF_REG_IP	PERF_REG_ARM64_PC
+#define PERF_REG_SP	PERF_REG_ARM64_SP
+
+static inline const char *perf_reg_name(int id)
+{
+	switch (id) {
+	case PERF_REG_ARM64_X0:
+		return "x0";
+	case PERF_REG_ARM64_X1:
+		return "x1";
+	case PERF_REG_ARM64_X2:
+		return "x2";
+	case PERF_REG_ARM64_X3:
+		return "x3";
+	case PERF_REG_ARM64_X4:
+		return "x4";
+	case PERF_REG_ARM64_X5:
+		return "x5";
+	case PERF_REG_ARM64_X6:
+		return "x6";
+	case PERF_REG_ARM64_X7:
+		return "x7";
+	case PERF_REG_ARM64_X8:
+		return "x8";
+	case PERF_REG_ARM64_X9:
+		return "x9";
+	case PERF_REG_ARM64_X10:
+		return "x10";
+	case PERF_REG_ARM64_X11:
+		return "x11";
+	case PERF_REG_ARM64_X12:
+		return "x12";
+	case PERF_REG_ARM64_X13:
+		return "x13";
+	case PERF_REG_ARM64_X14:
+		return "x14";
+	case PERF_REG_ARM64_X15:
+		return "x15";
+	case PERF_REG_ARM64_X16:
+		return "x16";
+	case PERF_REG_ARM64_X17:
+		return "x17";
+	case PERF_REG_ARM64_X18:
+		return "x18";
+	case PERF_REG_ARM64_X19:
+		return "x19";
+	case PERF_REG_ARM64_X20:
+		return "x20";
+	case PERF_REG_ARM64_X21:
+		return "x21";
+	case PERF_REG_ARM64_X22:
+		return "x22";
+	case PERF_REG_ARM64_X23:
+		return "x23";
+	case PERF_REG_ARM64_X24:
+		return "x24";
+	case PERF_REG_ARM64_X25:
+		return "x25";
+	case PERF_REG_ARM64_X26:
+		return "x26";
+	case PERF_REG_ARM64_X27:
+		return "x27";
+	case PERF_REG_ARM64_X28:
+		return "x28";
+	case PERF_REG_ARM64_X29:
+		return "x29";
+	case PERF_REG_ARM64_SP:
+		return "sp";
+	case PERF_REG_ARM64_LR:
+		return "lr";
+	case PERF_REG_ARM64_PC:
+		return "pc";
+	default:
+		return NULL;
+	}
+
+	return NULL;
+}
+
+#endif /* ARCH_PERF_REGS_H */
diff --git a/tools/perf/arch/arm64/util/Build b/tools/perf/arch/arm64/util/Build
new file mode 100644
index 000000000..e58123a89
--- /dev/null
+++ b/tools/perf/arch/arm64/util/Build
@@ -0,0 +1,2 @@
+libperf-$(CONFIG_DWARF)     += dwarf-regs.o
+libperf-$(CONFIG_LIBUNWIND) += unwind-libunwind.o
diff --git a/tools/perf/arch/arm64/util/dwarf-regs.c b/tools/perf/arch/arm64/util/dwarf-regs.c
new file mode 100644
index 000000000..d49efeb81
--- /dev/null
+++ b/tools/perf/arch/arm64/util/dwarf-regs.c
@@ -0,0 +1,80 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ * Copyright (C) 2010 Will Deacon, ARM Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <stddef.h>
+#include <dwarf-regs.h>
+
+struct pt_regs_dwarfnum {
+	const char *name;
+	unsigned int dwarfnum;
+};
+
+#define STR(s) #s
+#define REG_DWARFNUM_NAME(r, num) {.name = r, .dwarfnum = num}
+#define GPR_DWARFNUM_NAME(num) \
+	{.name = STR(%x##num), .dwarfnum = num}
+#define REG_DWARFNUM_END {.name = NULL, .dwarfnum = 0}
+
+/*
+ * Reference:
+ * http://infocenter.arm.com/help/topic/com.arm.doc.ihi0057b/IHI0057B_aadwarf64.pdf
+ */
+static const struct pt_regs_dwarfnum regdwarfnum_table[] = {
+	GPR_DWARFNUM_NAME(0),
+	GPR_DWARFNUM_NAME(1),
+	GPR_DWARFNUM_NAME(2),
+	GPR_DWARFNUM_NAME(3),
+	GPR_DWARFNUM_NAME(4),
+	GPR_DWARFNUM_NAME(5),
+	GPR_DWARFNUM_NAME(6),
+	GPR_DWARFNUM_NAME(7),
+	GPR_DWARFNUM_NAME(8),
+	GPR_DWARFNUM_NAME(9),
+	GPR_DWARFNUM_NAME(10),
+	GPR_DWARFNUM_NAME(11),
+	GPR_DWARFNUM_NAME(12),
+	GPR_DWARFNUM_NAME(13),
+	GPR_DWARFNUM_NAME(14),
+	GPR_DWARFNUM_NAME(15),
+	GPR_DWARFNUM_NAME(16),
+	GPR_DWARFNUM_NAME(17),
+	GPR_DWARFNUM_NAME(18),
+	GPR_DWARFNUM_NAME(19),
+	GPR_DWARFNUM_NAME(20),
+	GPR_DWARFNUM_NAME(21),
+	GPR_DWARFNUM_NAME(22),
+	GPR_DWARFNUM_NAME(23),
+	GPR_DWARFNUM_NAME(24),
+	GPR_DWARFNUM_NAME(25),
+	GPR_DWARFNUM_NAME(26),
+	GPR_DWARFNUM_NAME(27),
+	GPR_DWARFNUM_NAME(28),
+	GPR_DWARFNUM_NAME(29),
+	REG_DWARFNUM_NAME("%lr", 30),
+	REG_DWARFNUM_NAME("%sp", 31),
+	REG_DWARFNUM_END,
+};
+
+/**
+ * get_arch_regstr() - lookup register name from it's DWARF register number
+ * @n:	the DWARF register number
+ *
+ * get_arch_regstr() returns the name of the register in struct
+ * regdwarfnum_table from it's DWARF register number. If the register is not
+ * found in the table, this returns NULL;
+ */
+const char *get_arch_regstr(unsigned int n)
+{
+	const struct pt_regs_dwarfnum *roff;
+	for (roff = regdwarfnum_table; roff->name != NULL; roff++)
+		if (roff->dwarfnum == n)
+			return roff->name;
+	return NULL;
+}
diff --git a/tools/perf/arch/arm64/util/unwind-libunwind.c b/tools/perf/arch/arm64/util/unwind-libunwind.c
new file mode 100644
index 000000000..a87afa91a
--- /dev/null
+++ b/tools/perf/arch/arm64/util/unwind-libunwind.c
@@ -0,0 +1,83 @@
+
+#include <errno.h>
+#include <libunwind.h>
+#include "perf_regs.h"
+#include "../../util/unwind.h"
+#include "../../util/debug.h"
+
+int libunwind__arch_reg_id(int regnum)
+{
+	switch (regnum) {
+	case UNW_AARCH64_X0:
+		return PERF_REG_ARM64_X0;
+	case UNW_AARCH64_X1:
+		return PERF_REG_ARM64_X1;
+	case UNW_AARCH64_X2:
+		return PERF_REG_ARM64_X2;
+	case UNW_AARCH64_X3:
+		return PERF_REG_ARM64_X3;
+	case UNW_AARCH64_X4:
+		return PERF_REG_ARM64_X4;
+	case UNW_AARCH64_X5:
+		return PERF_REG_ARM64_X5;
+	case UNW_AARCH64_X6:
+		return PERF_REG_ARM64_X6;
+	case UNW_AARCH64_X7:
+		return PERF_REG_ARM64_X7;
+	case UNW_AARCH64_X8:
+		return PERF_REG_ARM64_X8;
+	case UNW_AARCH64_X9:
+		return PERF_REG_ARM64_X9;
+	case UNW_AARCH64_X10:
+		return PERF_REG_ARM64_X10;
+	case UNW_AARCH64_X11:
+		return PERF_REG_ARM64_X11;
+	case UNW_AARCH64_X12:
+		return PERF_REG_ARM64_X12;
+	case UNW_AARCH64_X13:
+		return PERF_REG_ARM64_X13;
+	case UNW_AARCH64_X14:
+		return PERF_REG_ARM64_X14;
+	case UNW_AARCH64_X15:
+		return PERF_REG_ARM64_X15;
+	case UNW_AARCH64_X16:
+		return PERF_REG_ARM64_X16;
+	case UNW_AARCH64_X17:
+		return PERF_REG_ARM64_X17;
+	case UNW_AARCH64_X18:
+		return PERF_REG_ARM64_X18;
+	case UNW_AARCH64_X19:
+		return PERF_REG_ARM64_X19;
+	case UNW_AARCH64_X20:
+		return PERF_REG_ARM64_X20;
+	case UNW_AARCH64_X21:
+		return PERF_REG_ARM64_X21;
+	case UNW_AARCH64_X22:
+		return PERF_REG_ARM64_X22;
+	case UNW_AARCH64_X23:
+		return PERF_REG_ARM64_X23;
+	case UNW_AARCH64_X24:
+		return PERF_REG_ARM64_X24;
+	case UNW_AARCH64_X25:
+		return PERF_REG_ARM64_X25;
+	case UNW_AARCH64_X26:
+		return PERF_REG_ARM64_X26;
+	case UNW_AARCH64_X27:
+		return PERF_REG_ARM64_X27;
+	case UNW_AARCH64_X28:
+		return PERF_REG_ARM64_X28;
+	case UNW_AARCH64_X29:
+		return PERF_REG_ARM64_X29;
+	case UNW_AARCH64_X30:
+		return PERF_REG_ARM64_LR;
+	case UNW_AARCH64_SP:
+		return PERF_REG_ARM64_SP;
+	case UNW_AARCH64_PC:
+		return PERF_REG_ARM64_PC;
+	default:
+		pr_err("unwind: invalid reg id %d\n", regnum);
+		return -EINVAL;
+	}
+
+	return -EINVAL;
+}
diff --git a/tools/perf/arch/common.c b/tools/perf/arch/common.c
new file mode 100644
index 000000000..49776f190
--- /dev/null
+++ b/tools/perf/arch/common.c
@@ -0,0 +1,220 @@
+#include <stdio.h>
+#include <sys/utsname.h>
+#include "common.h"
+#include "../util/debug.h"
+
+const char *const arm_triplets[] = {
+	"arm-eabi-",
+	"arm-linux-androideabi-",
+	"arm-unknown-linux-",
+	"arm-unknown-linux-gnu-",
+	"arm-unknown-linux-gnueabi-",
+	NULL
+};
+
+const char *const arm64_triplets[] = {
+	"aarch64-linux-android-",
+	NULL
+};
+
+const char *const powerpc_triplets[] = {
+	"powerpc-unknown-linux-gnu-",
+	"powerpc64-unknown-linux-gnu-",
+	NULL
+};
+
+const char *const s390_triplets[] = {
+	"s390-ibm-linux-",
+	NULL
+};
+
+const char *const sh_triplets[] = {
+	"sh-unknown-linux-gnu-",
+	"sh64-unknown-linux-gnu-",
+	NULL
+};
+
+const char *const sparc_triplets[] = {
+	"sparc-unknown-linux-gnu-",
+	"sparc64-unknown-linux-gnu-",
+	NULL
+};
+
+const char *const x86_triplets[] = {
+	"x86_64-pc-linux-gnu-",
+	"x86_64-unknown-linux-gnu-",
+	"i686-pc-linux-gnu-",
+	"i586-pc-linux-gnu-",
+	"i486-pc-linux-gnu-",
+	"i386-pc-linux-gnu-",
+	"i686-linux-android-",
+	"i686-android-linux-",
+	NULL
+};
+
+const char *const mips_triplets[] = {
+	"mips-unknown-linux-gnu-",
+	"mipsel-linux-android-",
+	NULL
+};
+
+static bool lookup_path(char *name)
+{
+	bool found = false;
+	char *path, *tmp;
+	char buf[PATH_MAX];
+	char *env = getenv("PATH");
+
+	if (!env)
+		return false;
+
+	env = strdup(env);
+	if (!env)
+		return false;
+
+	path = strtok_r(env, ":", &tmp);
+	while (path) {
+		scnprintf(buf, sizeof(buf), "%s/%s", path, name);
+		if (access(buf, F_OK) == 0) {
+			found = true;
+			break;
+		}
+		path = strtok_r(NULL, ":", &tmp);
+	}
+	free(env);
+	return found;
+}
+
+static int lookup_triplets(const char *const *triplets, const char *name)
+{
+	int i;
+	char buf[PATH_MAX];
+
+	for (i = 0; triplets[i] != NULL; i++) {
+		scnprintf(buf, sizeof(buf), "%s%s", triplets[i], name);
+		if (lookup_path(buf))
+			return i;
+	}
+	return -1;
+}
+
+/*
+ * Return architecture name in a normalized form.
+ * The conversion logic comes from the Makefile.
+ */
+static const char *normalize_arch(char *arch)
+{
+	if (!strcmp(arch, "x86_64"))
+		return "x86";
+	if (arch[0] == 'i' && arch[2] == '8' && arch[3] == '6')
+		return "x86";
+	if (!strcmp(arch, "sun4u") || !strncmp(arch, "sparc", 5))
+		return "sparc";
+	if (!strcmp(arch, "aarch64") || !strcmp(arch, "arm64"))
+		return "arm64";
+	if (!strncmp(arch, "arm", 3) || !strcmp(arch, "sa110"))
+		return "arm";
+	if (!strncmp(arch, "s390", 4))
+		return "s390";
+	if (!strncmp(arch, "parisc", 6))
+		return "parisc";
+	if (!strncmp(arch, "powerpc", 7) || !strncmp(arch, "ppc", 3))
+		return "powerpc";
+	if (!strncmp(arch, "mips", 4))
+		return "mips";
+	if (!strncmp(arch, "sh", 2) && isdigit(arch[2]))
+		return "sh";
+
+	return arch;
+}
+
+static int perf_session_env__lookup_binutils_path(struct perf_session_env *env,
+						  const char *name,
+						  const char **path)
+{
+	int idx;
+	const char *arch, *cross_env;
+	struct utsname uts;
+	const char *const *path_list;
+	char *buf = NULL;
+
+	arch = normalize_arch(env->arch);
+
+	if (uname(&uts) < 0)
+		goto out;
+
+	/*
+	 * We don't need to try to find objdump path for native system.
+	 * Just use default binutils path (e.g.: "objdump").
+	 */
+	if (!strcmp(normalize_arch(uts.machine), arch))
+		goto out;
+
+	cross_env = getenv("CROSS_COMPILE");
+	if (cross_env) {
+		if (asprintf(&buf, "%s%s", cross_env, name) < 0)
+			goto out_error;
+		if (buf[0] == '/') {
+			if (access(buf, F_OK) == 0)
+				goto out;
+			goto out_error;
+		}
+		if (lookup_path(buf))
+			goto out;
+		zfree(&buf);
+	}
+
+	if (!strcmp(arch, "arm"))
+		path_list = arm_triplets;
+	else if (!strcmp(arch, "arm64"))
+		path_list = arm64_triplets;
+	else if (!strcmp(arch, "powerpc"))
+		path_list = powerpc_triplets;
+	else if (!strcmp(arch, "sh"))
+		path_list = sh_triplets;
+	else if (!strcmp(arch, "s390"))
+		path_list = s390_triplets;
+	else if (!strcmp(arch, "sparc"))
+		path_list = sparc_triplets;
+	else if (!strcmp(arch, "x86"))
+		path_list = x86_triplets;
+	else if (!strcmp(arch, "mips"))
+		path_list = mips_triplets;
+	else {
+		ui__error("binutils for %s not supported.\n", arch);
+		goto out_error;
+	}
+
+	idx = lookup_triplets(path_list, name);
+	if (idx < 0) {
+		ui__error("Please install %s for %s.\n"
+			  "You can add it to PATH, set CROSS_COMPILE or "
+			  "override the default using --%s.\n",
+			  name, arch, name);
+		goto out_error;
+	}
+
+	if (asprintf(&buf, "%s%s", path_list[idx], name) < 0)
+		goto out_error;
+
+out:
+	*path = buf;
+	return 0;
+out_error:
+	free(buf);
+	*path = NULL;
+	return -1;
+}
+
+int perf_session_env__lookup_objdump(struct perf_session_env *env)
+{
+	/*
+	 * For live mode, env->arch will be NULL and we can use
+	 * the native objdump tool.
+	 */
+	if (env->arch == NULL)
+		return 0;
+
+	return perf_session_env__lookup_binutils_path(env, "objdump",
+						      &objdump_path);
+}
diff --git a/tools/perf/arch/common.h b/tools/perf/arch/common.h
new file mode 100644
index 000000000..ede246eda
--- /dev/null
+++ b/tools/perf/arch/common.h
@@ -0,0 +1,10 @@
+#ifndef ARCH_PERF_COMMON_H
+#define ARCH_PERF_COMMON_H
+
+#include "../util/session.h"
+
+extern const char *objdump_path;
+
+int perf_session_env__lookup_objdump(struct perf_session_env *env);
+
+#endif /* ARCH_PERF_COMMON_H */
diff --git a/tools/perf/arch/powerpc/Build b/tools/perf/arch/powerpc/Build
new file mode 100644
index 000000000..54afe4a46
--- /dev/null
+++ b/tools/perf/arch/powerpc/Build
@@ -0,0 +1 @@
+libperf-y += util/
diff --git a/tools/perf/arch/powerpc/Makefile b/tools/perf/arch/powerpc/Makefile
new file mode 100644
index 000000000..7fbca1750
--- /dev/null
+++ b/tools/perf/arch/powerpc/Makefile
@@ -0,0 +1,3 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+endif
diff --git a/tools/perf/arch/powerpc/util/Build b/tools/perf/arch/powerpc/util/Build
new file mode 100644
index 000000000..0af6e9b3f
--- /dev/null
+++ b/tools/perf/arch/powerpc/util/Build
@@ -0,0 +1,4 @@
+libperf-y += header.o
+
+libperf-$(CONFIG_DWARF) += dwarf-regs.o
+libperf-$(CONFIG_DWARF) += skip-callchain-idx.o
diff --git a/tools/perf/arch/powerpc/util/dwarf-regs.c b/tools/perf/arch/powerpc/util/dwarf-regs.c
new file mode 100644
index 000000000..733151cdf
--- /dev/null
+++ b/tools/perf/arch/powerpc/util/dwarf-regs.c
@@ -0,0 +1,88 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ * Copyright (C) 2010 Ian Munsie, IBM Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <stddef.h>
+#include <dwarf-regs.h>
+
+
+struct pt_regs_dwarfnum {
+	const char *name;
+	unsigned int dwarfnum;
+};
+
+#define STR(s) #s
+#define REG_DWARFNUM_NAME(r, num) {.name = r, .dwarfnum = num}
+#define GPR_DWARFNUM_NAME(num)	\
+	{.name = STR(%gpr##num), .dwarfnum = num}
+#define REG_DWARFNUM_END {.name = NULL, .dwarfnum = 0}
+
+/*
+ * Reference:
+ * http://refspecs.linuxfoundation.org/ELF/ppc64/PPC-elf64abi-1.9.html
+ */
+static const struct pt_regs_dwarfnum regdwarfnum_table[] = {
+	GPR_DWARFNUM_NAME(0),
+	GPR_DWARFNUM_NAME(1),
+	GPR_DWARFNUM_NAME(2),
+	GPR_DWARFNUM_NAME(3),
+	GPR_DWARFNUM_NAME(4),
+	GPR_DWARFNUM_NAME(5),
+	GPR_DWARFNUM_NAME(6),
+	GPR_DWARFNUM_NAME(7),
+	GPR_DWARFNUM_NAME(8),
+	GPR_DWARFNUM_NAME(9),
+	GPR_DWARFNUM_NAME(10),
+	GPR_DWARFNUM_NAME(11),
+	GPR_DWARFNUM_NAME(12),
+	GPR_DWARFNUM_NAME(13),
+	GPR_DWARFNUM_NAME(14),
+	GPR_DWARFNUM_NAME(15),
+	GPR_DWARFNUM_NAME(16),
+	GPR_DWARFNUM_NAME(17),
+	GPR_DWARFNUM_NAME(18),
+	GPR_DWARFNUM_NAME(19),
+	GPR_DWARFNUM_NAME(20),
+	GPR_DWARFNUM_NAME(21),
+	GPR_DWARFNUM_NAME(22),
+	GPR_DWARFNUM_NAME(23),
+	GPR_DWARFNUM_NAME(24),
+	GPR_DWARFNUM_NAME(25),
+	GPR_DWARFNUM_NAME(26),
+	GPR_DWARFNUM_NAME(27),
+	GPR_DWARFNUM_NAME(28),
+	GPR_DWARFNUM_NAME(29),
+	GPR_DWARFNUM_NAME(30),
+	GPR_DWARFNUM_NAME(31),
+	REG_DWARFNUM_NAME("%msr",   66),
+	REG_DWARFNUM_NAME("%ctr",   109),
+	REG_DWARFNUM_NAME("%link",  108),
+	REG_DWARFNUM_NAME("%xer",   101),
+	REG_DWARFNUM_NAME("%dar",   119),
+	REG_DWARFNUM_NAME("%dsisr", 118),
+	REG_DWARFNUM_END,
+};
+
+/**
+ * get_arch_regstr() - lookup register name from it's DWARF register number
+ * @n:	the DWARF register number
+ *
+ * get_arch_regstr() returns the name of the register in struct
+ * regdwarfnum_table from it's DWARF register number. If the register is not
+ * found in the table, this returns NULL;
+ */
+const char *get_arch_regstr(unsigned int n)
+{
+	const struct pt_regs_dwarfnum *roff;
+	for (roff = regdwarfnum_table; roff->name != NULL; roff++)
+		if (roff->dwarfnum == n)
+			return roff->name;
+	return NULL;
+}
diff --git a/tools/perf/arch/powerpc/util/header.c b/tools/perf/arch/powerpc/util/header.c
new file mode 100644
index 000000000..6c1b8a75d
--- /dev/null
+++ b/tools/perf/arch/powerpc/util/header.c
@@ -0,0 +1,34 @@
+#include <sys/types.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "../../util/header.h"
+#include "../../util/util.h"
+
+#define mfspr(rn)       ({unsigned long rval; \
+			 asm volatile("mfspr %0," __stringify(rn) \
+				      : "=r" (rval)); rval; })
+
+#define SPRN_PVR        0x11F	/* Processor Version Register */
+#define PVR_VER(pvr)    (((pvr) >>  16) & 0xFFFF) /* Version field */
+#define PVR_REV(pvr)    (((pvr) >>   0) & 0xFFFF) /* Revison field */
+
+int
+get_cpuid(char *buffer, size_t sz)
+{
+	unsigned long pvr;
+	int nb;
+
+	pvr = mfspr(SPRN_PVR);
+
+	nb = scnprintf(buffer, sz, "%lu,%lu$", PVR_VER(pvr), PVR_REV(pvr));
+
+	/* look for end marker to ensure the entire data fit */
+	if (strchr(buffer, '$')) {
+		buffer[nb-1] = '\0';
+		return 0;
+	}
+	return -1;
+}
diff --git a/tools/perf/arch/powerpc/util/skip-callchain-idx.c b/tools/perf/arch/powerpc/util/skip-callchain-idx.c
new file mode 100644
index 000000000..0c370f81e
--- /dev/null
+++ b/tools/perf/arch/powerpc/util/skip-callchain-idx.c
@@ -0,0 +1,280 @@
+/*
+ * Use DWARF Debug information to skip unnecessary callchain entries.
+ *
+ * Copyright (C) 2014 Sukadev Bhattiprolu, IBM Corporation.
+ * Copyright (C) 2014 Ulrich Weigand, IBM Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <inttypes.h>
+#include <dwarf.h>
+#include <elfutils/libdwfl.h>
+
+#include "util/thread.h"
+#include "util/callchain.h"
+#include "util/debug.h"
+
+/*
+ * When saving the callchain on Power, the kernel conservatively saves
+ * excess entries in the callchain. A few of these entries are needed
+ * in some cases but not others. If the unnecessary entries are not
+ * ignored, we end up with duplicate arcs in the call-graphs. Use
+ * DWARF debug information to skip over any unnecessary callchain
+ * entries.
+ *
+ * See function header for arch_adjust_callchain() below for more details.
+ *
+ * The libdwfl code in this file is based on code from elfutils
+ * (libdwfl/argp-std.c, libdwfl/tests/addrcfi.c, etc).
+ */
+static char *debuginfo_path;
+
+static const Dwfl_Callbacks offline_callbacks = {
+	.debuginfo_path = &debuginfo_path,
+	.find_debuginfo = dwfl_standard_find_debuginfo,
+	.section_address = dwfl_offline_section_address,
+};
+
+
+/*
+ * Use the DWARF expression for the Call-frame-address and determine
+ * if return address is in LR and if a new frame was allocated.
+ */
+static int check_return_reg(int ra_regno, Dwarf_Frame *frame)
+{
+	Dwarf_Op ops_mem[2];
+	Dwarf_Op dummy;
+	Dwarf_Op *ops = &dummy;
+	size_t nops;
+	int result;
+
+	result = dwarf_frame_register(frame, ra_regno, ops_mem, &ops, &nops);
+	if (result < 0) {
+		pr_debug("dwarf_frame_register() %s\n", dwarf_errmsg(-1));
+		return -1;
+	}
+
+	/*
+	 * Check if return address is on the stack.
+	 */
+	if (nops != 0 || ops != NULL)
+		return 0;
+
+	/*
+	 * Return address is in LR. Check if a frame was allocated
+	 * but not-yet used.
+	 */
+	result = dwarf_frame_cfa(frame, &ops, &nops);
+	if (result < 0) {
+		pr_debug("dwarf_frame_cfa() returns %d, %s\n", result,
+					dwarf_errmsg(-1));
+		return -1;
+	}
+
+	/*
+	 * If call frame address is in r1, no new frame was allocated.
+	 */
+	if (nops == 1 && ops[0].atom == DW_OP_bregx && ops[0].number == 1 &&
+				ops[0].number2 == 0)
+		return 1;
+
+	/*
+	 * A new frame was allocated but has not yet been used.
+	 */
+	return 2;
+}
+
+/*
+ * Get the DWARF frame from the .eh_frame section.
+ */
+static Dwarf_Frame *get_eh_frame(Dwfl_Module *mod, Dwarf_Addr pc)
+{
+	int		result;
+	Dwarf_Addr	bias;
+	Dwarf_CFI	*cfi;
+	Dwarf_Frame	*frame;
+
+	cfi = dwfl_module_eh_cfi(mod, &bias);
+	if (!cfi) {
+		pr_debug("%s(): no CFI - %s\n", __func__, dwfl_errmsg(-1));
+		return NULL;
+	}
+
+	result = dwarf_cfi_addrframe(cfi, pc-bias, &frame);
+	if (result) {
+		pr_debug("%s(): %s\n", __func__, dwfl_errmsg(-1));
+		return NULL;
+	}
+
+	return frame;
+}
+
+/*
+ * Get the DWARF frame from the .debug_frame section.
+ */
+static Dwarf_Frame *get_dwarf_frame(Dwfl_Module *mod, Dwarf_Addr pc)
+{
+	Dwarf_CFI       *cfi;
+	Dwarf_Addr      bias;
+	Dwarf_Frame     *frame;
+	int             result;
+
+	cfi = dwfl_module_dwarf_cfi(mod, &bias);
+	if (!cfi) {
+		pr_debug("%s(): no CFI - %s\n", __func__, dwfl_errmsg(-1));
+		return NULL;
+	}
+
+	result = dwarf_cfi_addrframe(cfi, pc-bias, &frame);
+	if (result) {
+		pr_debug("%s(): %s\n", __func__, dwfl_errmsg(-1));
+		return NULL;
+	}
+
+	return frame;
+}
+
+/*
+ * Return:
+ *	0 if return address for the program counter @pc is on stack
+ *	1 if return address is in LR and no new stack frame was allocated
+ *	2 if return address is in LR and a new frame was allocated (but not
+ *		yet used)
+ *	-1 in case of errors
+ */
+static int check_return_addr(struct dso *dso, u64 map_start, Dwarf_Addr pc)
+{
+	int		rc = -1;
+	Dwfl		*dwfl;
+	Dwfl_Module	*mod;
+	Dwarf_Frame	*frame;
+	int		ra_regno;
+	Dwarf_Addr	start = pc;
+	Dwarf_Addr	end = pc;
+	bool		signalp;
+	const char	*exec_file = dso->long_name;
+
+	dwfl = dso->dwfl;
+
+	if (!dwfl) {
+		dwfl = dwfl_begin(&offline_callbacks);
+		if (!dwfl) {
+			pr_debug("dwfl_begin() failed: %s\n", dwarf_errmsg(-1));
+			return -1;
+		}
+
+		mod = dwfl_report_elf(dwfl, exec_file, exec_file, -1,
+						map_start, false);
+		if (!mod) {
+			pr_debug("dwfl_report_elf() failed %s\n",
+						dwarf_errmsg(-1));
+			/*
+			 * We normally cache the DWARF debug info and never
+			 * call dwfl_end(). But to prevent fd leak, free in
+			 * case of error.
+			 */
+			dwfl_end(dwfl);
+			goto out;
+		}
+		dso->dwfl = dwfl;
+	}
+
+	mod = dwfl_addrmodule(dwfl, pc);
+	if (!mod) {
+		pr_debug("dwfl_addrmodule() failed, %s\n", dwarf_errmsg(-1));
+		goto out;
+	}
+
+	/*
+	 * To work with split debug info files (eg: glibc), check both
+	 * .eh_frame and .debug_frame sections of the ELF header.
+	 */
+	frame = get_eh_frame(mod, pc);
+	if (!frame) {
+		frame = get_dwarf_frame(mod, pc);
+		if (!frame)
+			goto out;
+	}
+
+	ra_regno = dwarf_frame_info(frame, &start, &end, &signalp);
+	if (ra_regno < 0) {
+		pr_debug("Return address register unavailable: %s\n",
+				dwarf_errmsg(-1));
+		goto out;
+	}
+
+	rc = check_return_reg(ra_regno, frame);
+
+out:
+	return rc;
+}
+
+/*
+ * The callchain saved by the kernel always includes the link register (LR).
+ *
+ *	0:	PERF_CONTEXT_USER
+ *	1:	Program counter (Next instruction pointer)
+ *	2:	LR value
+ *	3:	Caller's caller
+ *	4:	...
+ *
+ * The value in LR is only needed when it holds a return address. If the
+ * return address is on the stack, we should ignore the LR value.
+ *
+ * Further, when the return address is in the LR, if a new frame was just
+ * allocated but the LR was not saved into it, then the LR contains the
+ * caller, slot 4: contains the caller's caller and the contents of slot 3:
+ * (chain->ips[3]) is undefined and must be ignored.
+ *
+ * Use DWARF debug information to determine if any entries need to be skipped.
+ *
+ * Return:
+ *	index:	of callchain entry that needs to be ignored (if any)
+ *	-1	if no entry needs to be ignored or in case of errors
+ */
+int arch_skip_callchain_idx(struct thread *thread, struct ip_callchain *chain)
+{
+	struct addr_location al;
+	struct dso *dso = NULL;
+	int rc;
+	u64 ip;
+	u64 skip_slot = -1;
+
+	if (chain->nr < 3)
+		return skip_slot;
+
+	ip = chain->ips[2];
+
+	thread__find_addr_location(thread, PERF_RECORD_MISC_USER,
+			MAP__FUNCTION, ip, &al);
+
+	if (al.map)
+		dso = al.map->dso;
+
+	if (!dso) {
+		pr_debug("%" PRIx64 " dso is NULL\n", ip);
+		return skip_slot;
+	}
+
+	rc = check_return_addr(dso, al.map->start, ip);
+
+	pr_debug("[DSO %s, sym %s, ip 0x%" PRIx64 "] rc %d\n",
+				dso->long_name, al.sym->name, ip, rc);
+
+	if (rc == 0) {
+		/*
+		 * Return address on stack. Ignore LR value in callchain
+		 */
+		skip_slot = 2;
+	} else if (rc == 2) {
+		/*
+		 * New frame allocated but return address still in LR.
+		 * Ignore the caller's caller entry in callchain.
+		 */
+		skip_slot = 3;
+	}
+	return skip_slot;
+}
diff --git a/tools/perf/arch/s390/Build b/tools/perf/arch/s390/Build
new file mode 100644
index 000000000..54afe4a46
--- /dev/null
+++ b/tools/perf/arch/s390/Build
@@ -0,0 +1 @@
+libperf-y += util/
diff --git a/tools/perf/arch/s390/Makefile b/tools/perf/arch/s390/Makefile
new file mode 100644
index 000000000..21322e038
--- /dev/null
+++ b/tools/perf/arch/s390/Makefile
@@ -0,0 +1,4 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+endif
+HAVE_KVM_STAT_SUPPORT := 1
diff --git a/tools/perf/arch/s390/util/Build b/tools/perf/arch/s390/util/Build
new file mode 100644
index 000000000..8a61372bb
--- /dev/null
+++ b/tools/perf/arch/s390/util/Build
@@ -0,0 +1,4 @@
+libperf-y += header.o
+libperf-y += kvm-stat.o
+
+libperf-$(CONFIG_DWARF) += dwarf-regs.o
diff --git a/tools/perf/arch/s390/util/dwarf-regs.c b/tools/perf/arch/s390/util/dwarf-regs.c
new file mode 100644
index 000000000..0469df02e
--- /dev/null
+++ b/tools/perf/arch/s390/util/dwarf-regs.c
@@ -0,0 +1,22 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ *    Copyright IBM Corp. 2010
+ *    Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>,
+ *
+ */
+
+#include <stddef.h>
+#include <dwarf-regs.h>
+
+#define NUM_GPRS 16
+
+static const char *gpr_names[NUM_GPRS] = {
+	"%r0", "%r1",  "%r2",  "%r3",  "%r4",  "%r5",  "%r6",  "%r7",
+	"%r8", "%r9", "%r10", "%r11", "%r12", "%r13", "%r14", "%r15",
+};
+
+const char *get_arch_regstr(unsigned int n)
+{
+	return (n >= NUM_GPRS) ? NULL : gpr_names[n];
+}
diff --git a/tools/perf/arch/s390/util/header.c b/tools/perf/arch/s390/util/header.c
new file mode 100644
index 000000000..9fa6c3e57
--- /dev/null
+++ b/tools/perf/arch/s390/util/header.c
@@ -0,0 +1,28 @@
+/*
+ * Implementation of get_cpuid().
+ *
+ * Copyright 2014 IBM Corp.
+ * Author(s): Alexander Yarygin <yarygin@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ */
+
+#include <sys/types.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <string.h>
+
+#include "../../util/header.h"
+
+int get_cpuid(char *buffer, size_t sz)
+{
+	const char *cpuid = "IBM/S390";
+
+	if (strlen(cpuid) + 1 > sz)
+		return -1;
+
+	strcpy(buffer, cpuid);
+	return 0;
+}
diff --git a/tools/perf/arch/s390/util/kvm-stat.c b/tools/perf/arch/s390/util/kvm-stat.c
new file mode 100644
index 000000000..a5dbc07ec
--- /dev/null
+++ b/tools/perf/arch/s390/util/kvm-stat.c
@@ -0,0 +1,105 @@
+/*
+ * Arch specific functions for perf kvm stat.
+ *
+ * Copyright 2014 IBM Corp.
+ * Author(s): Alexander Yarygin <yarygin@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ */
+
+#include "../../util/kvm-stat.h"
+#include <asm/kvm_perf.h>
+
+define_exit_reasons_table(sie_exit_reasons, sie_intercept_code);
+define_exit_reasons_table(sie_icpt_insn_codes, icpt_insn_codes);
+define_exit_reasons_table(sie_sigp_order_codes, sigp_order_codes);
+define_exit_reasons_table(sie_diagnose_codes, diagnose_codes);
+define_exit_reasons_table(sie_icpt_prog_codes, icpt_prog_codes);
+
+static void event_icpt_insn_get_key(struct perf_evsel *evsel,
+				    struct perf_sample *sample,
+				    struct event_key *key)
+{
+	unsigned long insn;
+
+	insn = perf_evsel__intval(evsel, sample, "instruction");
+	key->key = icpt_insn_decoder(insn);
+	key->exit_reasons = sie_icpt_insn_codes;
+}
+
+static void event_sigp_get_key(struct perf_evsel *evsel,
+			       struct perf_sample *sample,
+			       struct event_key *key)
+{
+	key->key = perf_evsel__intval(evsel, sample, "order_code");
+	key->exit_reasons = sie_sigp_order_codes;
+}
+
+static void event_diag_get_key(struct perf_evsel *evsel,
+			       struct perf_sample *sample,
+			       struct event_key *key)
+{
+	key->key = perf_evsel__intval(evsel, sample, "code");
+	key->exit_reasons = sie_diagnose_codes;
+}
+
+static void event_icpt_prog_get_key(struct perf_evsel *evsel,
+				    struct perf_sample *sample,
+				    struct event_key *key)
+{
+	key->key = perf_evsel__intval(evsel, sample, "code");
+	key->exit_reasons = sie_icpt_prog_codes;
+}
+
+static struct child_event_ops child_events[] = {
+	{ .name = "kvm:kvm_s390_intercept_instruction",
+	  .get_key = event_icpt_insn_get_key },
+	{ .name = "kvm:kvm_s390_handle_sigp",
+	  .get_key = event_sigp_get_key },
+	{ .name = "kvm:kvm_s390_handle_diag",
+	  .get_key = event_diag_get_key },
+	{ .name = "kvm:kvm_s390_intercept_prog",
+	  .get_key = event_icpt_prog_get_key },
+	{ NULL, NULL },
+};
+
+static struct kvm_events_ops exit_events = {
+	.is_begin_event = exit_event_begin,
+	.is_end_event = exit_event_end,
+	.child_ops = child_events,
+	.decode_key = exit_event_decode_key,
+	.name = "VM-EXIT"
+};
+
+const char * const kvm_events_tp[] = {
+	"kvm:kvm_s390_sie_enter",
+	"kvm:kvm_s390_sie_exit",
+	"kvm:kvm_s390_intercept_instruction",
+	"kvm:kvm_s390_handle_sigp",
+	"kvm:kvm_s390_handle_diag",
+	"kvm:kvm_s390_intercept_prog",
+	NULL,
+};
+
+struct kvm_reg_events_ops kvm_reg_events_ops[] = {
+	{ .name = "vmexit", .ops = &exit_events },
+	{ NULL, NULL },
+};
+
+const char * const kvm_skip_events[] = {
+	"Wait state",
+	NULL,
+};
+
+int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid)
+{
+	if (strstr(cpuid, "IBM/S390")) {
+		kvm->exit_reasons = sie_exit_reasons;
+		kvm->exit_reasons_isa = "SIE";
+	} else
+		return -ENOTSUP;
+
+	return 0;
+}
diff --git a/tools/perf/arch/sh/Build b/tools/perf/arch/sh/Build
new file mode 100644
index 000000000..54afe4a46
--- /dev/null
+++ b/tools/perf/arch/sh/Build
@@ -0,0 +1 @@
+libperf-y += util/
diff --git a/tools/perf/arch/sh/Makefile b/tools/perf/arch/sh/Makefile
new file mode 100644
index 000000000..7fbca1750
--- /dev/null
+++ b/tools/perf/arch/sh/Makefile
@@ -0,0 +1,3 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+endif
diff --git a/tools/perf/arch/sh/util/Build b/tools/perf/arch/sh/util/Build
new file mode 100644
index 000000000..954e287bb
--- /dev/null
+++ b/tools/perf/arch/sh/util/Build
@@ -0,0 +1 @@
+libperf-$(CONFIG_DWARF) += dwarf-regs.o
diff --git a/tools/perf/arch/sh/util/dwarf-regs.c b/tools/perf/arch/sh/util/dwarf-regs.c
new file mode 100644
index 000000000..0d0897f57
--- /dev/null
+++ b/tools/perf/arch/sh/util/dwarf-regs.c
@@ -0,0 +1,55 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ * Copyright (C) 2010 Matt Fleming <matt@console-pimps.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <stddef.h>
+#include <dwarf-regs.h>
+
+/*
+ * Generic dwarf analysis helpers
+ */
+
+#define SH_MAX_REGS 18
+const char *sh_regs_table[SH_MAX_REGS] = {
+	"r0",
+	"r1",
+	"r2",
+	"r3",
+	"r4",
+	"r5",
+	"r6",
+	"r7",
+	"r8",
+	"r9",
+	"r10",
+	"r11",
+	"r12",
+	"r13",
+	"r14",
+	"r15",
+	"pc",
+	"pr",
+};
+
+/* Return architecture dependent register string (for kprobe-tracer) */
+const char *get_arch_regstr(unsigned int n)
+{
+	return (n <= SH_MAX_REGS) ? sh_regs_table[n] : NULL;
+}
diff --git a/tools/perf/arch/sparc/Build b/tools/perf/arch/sparc/Build
new file mode 100644
index 000000000..54afe4a46
--- /dev/null
+++ b/tools/perf/arch/sparc/Build
@@ -0,0 +1 @@
+libperf-y += util/
diff --git a/tools/perf/arch/sparc/Makefile b/tools/perf/arch/sparc/Makefile
new file mode 100644
index 000000000..7fbca1750
--- /dev/null
+++ b/tools/perf/arch/sparc/Makefile
@@ -0,0 +1,3 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+endif
diff --git a/tools/perf/arch/sparc/util/Build b/tools/perf/arch/sparc/util/Build
new file mode 100644
index 000000000..954e287bb
--- /dev/null
+++ b/tools/perf/arch/sparc/util/Build
@@ -0,0 +1 @@
+libperf-$(CONFIG_DWARF) += dwarf-regs.o
diff --git a/tools/perf/arch/sparc/util/dwarf-regs.c b/tools/perf/arch/sparc/util/dwarf-regs.c
new file mode 100644
index 000000000..92eda412f
--- /dev/null
+++ b/tools/perf/arch/sparc/util/dwarf-regs.c
@@ -0,0 +1,43 @@
+/*
+ * Mapping of DWARF debug register numbers into register names.
+ *
+ * Copyright (C) 2010 David S. Miller <davem@davemloft.net>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <stddef.h>
+#include <dwarf-regs.h>
+
+#define SPARC_MAX_REGS	96
+
+const char *sparc_regs_table[SPARC_MAX_REGS] = {
+	"%g0", "%g1", "%g2", "%g3", "%g4", "%g5", "%g6", "%g7",
+	"%o0", "%o1", "%o2", "%o3", "%o4", "%o5", "%sp", "%o7",
+	"%l0", "%l1", "%l2", "%l3", "%l4", "%l5", "%l6", "%l7",
+	"%i0", "%i1", "%i2", "%i3", "%i4", "%i5", "%fp", "%i7",
+	"%f0", "%f1", "%f2", "%f3", "%f4", "%f5", "%f6", "%f7",
+	"%f8", "%f9", "%f10", "%f11", "%f12", "%f13", "%f14", "%f15",
+	"%f16", "%f17", "%f18", "%f19", "%f20", "%f21", "%f22", "%f23",
+	"%f24", "%f25", "%f26", "%f27", "%f28", "%f29", "%f30", "%f31",
+	"%f32", "%f33", "%f34", "%f35", "%f36", "%f37", "%f38", "%f39",
+	"%f40", "%f41", "%f42", "%f43", "%f44", "%f45", "%f46", "%f47",
+	"%f48", "%f49", "%f50", "%f51", "%f52", "%f53", "%f54", "%f55",
+	"%f56", "%f57", "%f58", "%f59", "%f60", "%f61", "%f62", "%f63",
+};
+
+/**
+ * get_arch_regstr() - lookup register name from it's DWARF register number
+ * @n:	the DWARF register number
+ *
+ * get_arch_regstr() returns the name of the register in struct
+ * regdwarfnum_table from it's DWARF register number. If the register is not
+ * found in the table, this returns NULL;
+ */
+const char *get_arch_regstr(unsigned int n)
+{
+	return (n <= SPARC_MAX_REGS) ? sparc_regs_table[n] : NULL;
+}
diff --git a/tools/perf/arch/x86/Build b/tools/perf/arch/x86/Build
new file mode 100644
index 000000000..41bf61da4
--- /dev/null
+++ b/tools/perf/arch/x86/Build
@@ -0,0 +1,2 @@
+libperf-y += util/
+libperf-$(CONFIG_DWARF_UNWIND) += tests/
diff --git a/tools/perf/arch/x86/Makefile b/tools/perf/arch/x86/Makefile
new file mode 100644
index 000000000..21322e038
--- /dev/null
+++ b/tools/perf/arch/x86/Makefile
@@ -0,0 +1,4 @@
+ifndef NO_DWARF
+PERF_HAVE_DWARF_REGS := 1
+endif
+HAVE_KVM_STAT_SUPPORT := 1
diff --git a/tools/perf/arch/x86/include/perf_regs.h b/tools/perf/arch/x86/include/perf_regs.h
new file mode 100644
index 000000000..7df517acf
--- /dev/null
+++ b/tools/perf/arch/x86/include/perf_regs.h
@@ -0,0 +1,86 @@
+#ifndef ARCH_PERF_REGS_H
+#define ARCH_PERF_REGS_H
+
+#include <stdlib.h>
+#include <linux/types.h>
+#include <asm/perf_regs.h>
+
+void perf_regs_load(u64 *regs);
+
+#ifndef HAVE_ARCH_X86_64_SUPPORT
+#define PERF_REGS_MASK ((1ULL << PERF_REG_X86_32_MAX) - 1)
+#define PERF_REGS_MAX PERF_REG_X86_32_MAX
+#define PERF_SAMPLE_REGS_ABI PERF_SAMPLE_REGS_ABI_32
+#else
+#define REG_NOSUPPORT ((1ULL << PERF_REG_X86_DS) | \
+		       (1ULL << PERF_REG_X86_ES) | \
+		       (1ULL << PERF_REG_X86_FS) | \
+		       (1ULL << PERF_REG_X86_GS))
+#define PERF_REGS_MASK (((1ULL << PERF_REG_X86_64_MAX) - 1) & ~REG_NOSUPPORT)
+#define PERF_REGS_MAX PERF_REG_X86_64_MAX
+#define PERF_SAMPLE_REGS_ABI PERF_SAMPLE_REGS_ABI_64
+#endif
+#define PERF_REG_IP PERF_REG_X86_IP
+#define PERF_REG_SP PERF_REG_X86_SP
+
+static inline const char *perf_reg_name(int id)
+{
+	switch (id) {
+	case PERF_REG_X86_AX:
+		return "AX";
+	case PERF_REG_X86_BX:
+		return "BX";
+	case PERF_REG_X86_CX:
+		return "CX";
+	case PERF_REG_X86_DX:
+		return "DX";
+	case PERF_REG_X86_SI:
+		return "SI";
+	case PERF_REG_X86_DI:
+		return "DI";
+	case PERF_REG_X86_BP:
+		return "BP";
+	case PERF_REG_X86_SP:
+		return "SP";
+	case PERF_REG_X86_IP:
+		return "IP";
+	case PERF_REG_X86_FLAGS:
+		return "FLAGS";
+	case PERF_REG_X86_CS:
+		return "CS";
+	case PERF_REG_X86_SS:
+		return "SS";
+	case PERF_REG_X86_DS:
+		return "DS";
+	case PERF_REG_X86_ES:
+		return "ES";
+	case PERF_REG_X86_FS:
+		return "FS";
+	case PERF_REG_X86_GS:
+		return "GS";
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+	case PERF_REG_X86_R8:
+		return "R8";
+	case PERF_REG_X86_R9:
+		return "R9";
+	case PERF_REG_X86_R10:
+		return "R10";
+	case PERF_REG_X86_R11:
+		return "R11";
+	case PERF_REG_X86_R12:
+		return "R12";
+	case PERF_REG_X86_R13:
+		return "R13";
+	case PERF_REG_X86_R14:
+		return "R14";
+	case PERF_REG_X86_R15:
+		return "R15";
+#endif /* HAVE_ARCH_X86_64_SUPPORT */
+	default:
+		return NULL;
+	}
+
+	return NULL;
+}
+
+#endif /* ARCH_PERF_REGS_H */
diff --git a/tools/perf/arch/x86/tests/Build b/tools/perf/arch/x86/tests/Build
new file mode 100644
index 000000000..b30eff9bc
--- /dev/null
+++ b/tools/perf/arch/x86/tests/Build
@@ -0,0 +1,2 @@
+libperf-y += regs_load.o
+libperf-y += dwarf-unwind.o
diff --git a/tools/perf/arch/x86/tests/dwarf-unwind.c b/tools/perf/arch/x86/tests/dwarf-unwind.c
new file mode 100644
index 000000000..d8bbf7ad1
--- /dev/null
+++ b/tools/perf/arch/x86/tests/dwarf-unwind.c
@@ -0,0 +1,61 @@
+#include <string.h>
+#include "perf_regs.h"
+#include "thread.h"
+#include "map.h"
+#include "event.h"
+#include "debug.h"
+#include "tests/tests.h"
+
+#define STACK_SIZE 8192
+
+static int sample_ustack(struct perf_sample *sample,
+			 struct thread *thread, u64 *regs)
+{
+	struct stack_dump *stack = &sample->user_stack;
+	struct map *map;
+	unsigned long sp;
+	u64 stack_size, *buf;
+
+	buf = malloc(STACK_SIZE);
+	if (!buf) {
+		pr_debug("failed to allocate sample uregs data\n");
+		return -1;
+	}
+
+	sp = (unsigned long) regs[PERF_REG_X86_SP];
+
+	map = map_groups__find(thread->mg, MAP__VARIABLE, (u64) sp);
+	if (!map) {
+		pr_debug("failed to get stack map\n");
+		free(buf);
+		return -1;
+	}
+
+	stack_size = map->end - sp;
+	stack_size = stack_size > STACK_SIZE ? STACK_SIZE : stack_size;
+
+	memcpy(buf, (void *) sp, stack_size);
+	stack->data = (char *) buf;
+	stack->size = stack_size;
+	return 0;
+}
+
+int test__arch_unwind_sample(struct perf_sample *sample,
+			     struct thread *thread)
+{
+	struct regs_dump *regs = &sample->user_regs;
+	u64 *buf;
+
+	buf = malloc(sizeof(u64) * PERF_REGS_MAX);
+	if (!buf) {
+		pr_debug("failed to allocate sample uregs data\n");
+		return -1;
+	}
+
+	perf_regs_load(buf);
+	regs->abi  = PERF_SAMPLE_REGS_ABI;
+	regs->regs = buf;
+	regs->mask = PERF_REGS_MASK;
+
+	return sample_ustack(sample, thread, buf);
+}
diff --git a/tools/perf/arch/x86/tests/regs_load.S b/tools/perf/arch/x86/tests/regs_load.S
new file mode 100644
index 000000000..60875d5c5
--- /dev/null
+++ b/tools/perf/arch/x86/tests/regs_load.S
@@ -0,0 +1,98 @@
+#include <linux/linkage.h>
+
+#define AX	 0
+#define BX	 1 * 8
+#define CX	 2 * 8
+#define DX	 3 * 8
+#define SI	 4 * 8
+#define DI	 5 * 8
+#define BP	 6 * 8
+#define SP	 7 * 8
+#define IP	 8 * 8
+#define FLAGS	 9 * 8
+#define CS	10 * 8
+#define SS	11 * 8
+#define DS	12 * 8
+#define ES	13 * 8
+#define FS	14 * 8
+#define GS	15 * 8
+#define R8	16 * 8
+#define R9	17 * 8
+#define R10	18 * 8
+#define R11	19 * 8
+#define R12	20 * 8
+#define R13	21 * 8
+#define R14	22 * 8
+#define R15	23 * 8
+
+.text
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+ENTRY(perf_regs_load)
+	movq %rax, AX(%rdi)
+	movq %rbx, BX(%rdi)
+	movq %rcx, CX(%rdi)
+	movq %rdx, DX(%rdi)
+	movq %rsi, SI(%rdi)
+	movq %rdi, DI(%rdi)
+	movq %rbp, BP(%rdi)
+
+	leaq 8(%rsp), %rax /* exclude this call.  */
+	movq %rax, SP(%rdi)
+
+	movq 0(%rsp), %rax
+	movq %rax, IP(%rdi)
+
+	movq $0, FLAGS(%rdi)
+	movq $0, CS(%rdi)
+	movq $0, SS(%rdi)
+	movq $0, DS(%rdi)
+	movq $0, ES(%rdi)
+	movq $0, FS(%rdi)
+	movq $0, GS(%rdi)
+
+	movq %r8,  R8(%rdi)
+	movq %r9,  R9(%rdi)
+	movq %r10, R10(%rdi)
+	movq %r11, R11(%rdi)
+	movq %r12, R12(%rdi)
+	movq %r13, R13(%rdi)
+	movq %r14, R14(%rdi)
+	movq %r15, R15(%rdi)
+	ret
+ENDPROC(perf_regs_load)
+#else
+ENTRY(perf_regs_load)
+	push %edi
+	movl 8(%esp), %edi
+	movl %eax, AX(%edi)
+	movl %ebx, BX(%edi)
+	movl %ecx, CX(%edi)
+	movl %edx, DX(%edi)
+	movl %esi, SI(%edi)
+	pop %eax
+	movl %eax, DI(%edi)
+	movl %ebp, BP(%edi)
+
+	leal 4(%esp), %eax /* exclude this call.  */
+	movl %eax, SP(%edi)
+
+	movl 0(%esp), %eax
+	movl %eax, IP(%edi)
+
+	movl $0, FLAGS(%edi)
+	movl $0, CS(%edi)
+	movl $0, SS(%edi)
+	movl $0, DS(%edi)
+	movl $0, ES(%edi)
+	movl $0, FS(%edi)
+	movl $0, GS(%edi)
+	ret
+ENDPROC(perf_regs_load)
+#endif
+
+/*
+ * We need to provide note.GNU-stack section, saying that we want
+ * NOT executable stack. Otherwise the final linking will assume that
+ * the ELF stack should not be restricted at all and set it RWX.
+ */
+.section .note.GNU-stack,"",@progbits
diff --git a/tools/perf/arch/x86/util/Build b/tools/perf/arch/x86/util/Build
new file mode 100644
index 000000000..cfbccc4e3
--- /dev/null
+++ b/tools/perf/arch/x86/util/Build
@@ -0,0 +1,8 @@
+libperf-y += header.o
+libperf-y += tsc.o
+libperf-y += kvm-stat.o
+
+libperf-$(CONFIG_DWARF) += dwarf-regs.o
+
+libperf-$(CONFIG_LIBUNWIND)          += unwind-libunwind.o
+libperf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o
diff --git a/tools/perf/arch/x86/util/dwarf-regs.c b/tools/perf/arch/x86/util/dwarf-regs.c
new file mode 100644
index 000000000..be22dd463
--- /dev/null
+++ b/tools/perf/arch/x86/util/dwarf-regs.c
@@ -0,0 +1,75 @@
+/*
+ * dwarf-regs.c : Mapping of DWARF debug register numbers into register names.
+ * Extracted from probe-finder.c
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <stddef.h>
+#include <dwarf-regs.h>
+
+/*
+ * Generic dwarf analysis helpers
+ */
+
+#define X86_32_MAX_REGS 8
+const char *x86_32_regs_table[X86_32_MAX_REGS] = {
+	"%ax",
+	"%cx",
+	"%dx",
+	"%bx",
+	"$stack",	/* Stack address instead of %sp */
+	"%bp",
+	"%si",
+	"%di",
+};
+
+#define X86_64_MAX_REGS 16
+const char *x86_64_regs_table[X86_64_MAX_REGS] = {
+	"%ax",
+	"%dx",
+	"%cx",
+	"%bx",
+	"%si",
+	"%di",
+	"%bp",
+	"%sp",
+	"%r8",
+	"%r9",
+	"%r10",
+	"%r11",
+	"%r12",
+	"%r13",
+	"%r14",
+	"%r15",
+};
+
+/* TODO: switching by dwarf address size */
+#ifdef __x86_64__
+#define ARCH_MAX_REGS X86_64_MAX_REGS
+#define arch_regs_table x86_64_regs_table
+#else
+#define ARCH_MAX_REGS X86_32_MAX_REGS
+#define arch_regs_table x86_32_regs_table
+#endif
+
+/* Return architecture dependent register string (for kprobe-tracer) */
+const char *get_arch_regstr(unsigned int n)
+{
+	return (n <= ARCH_MAX_REGS) ? arch_regs_table[n] : NULL;
+}
diff --git a/tools/perf/arch/x86/util/header.c b/tools/perf/arch/x86/util/header.c
new file mode 100644
index 000000000..146d12a1c
--- /dev/null
+++ b/tools/perf/arch/x86/util/header.c
@@ -0,0 +1,59 @@
+#include <sys/types.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "../../util/header.h"
+
+static inline void
+cpuid(unsigned int op, unsigned int *a, unsigned int *b, unsigned int *c,
+      unsigned int *d)
+{
+	__asm__ __volatile__ (".byte 0x53\n\tcpuid\n\t"
+			      "movl %%ebx, %%esi\n\t.byte 0x5b"
+			: "=a" (*a),
+			"=S" (*b),
+			"=c" (*c),
+			"=d" (*d)
+			: "a" (op));
+}
+
+int
+get_cpuid(char *buffer, size_t sz)
+{
+	unsigned int a, b, c, d, lvl;
+	int family = -1, model = -1, step = -1;
+	int nb;
+	char vendor[16];
+
+	cpuid(0, &lvl, &b, &c, &d);
+	strncpy(&vendor[0], (char *)(&b), 4);
+	strncpy(&vendor[4], (char *)(&d), 4);
+	strncpy(&vendor[8], (char *)(&c), 4);
+	vendor[12] = '\0';
+
+	if (lvl >= 1) {
+		cpuid(1, &a, &b, &c, &d);
+
+		family = (a >> 8) & 0xf;  /* bits 11 - 8 */
+		model  = (a >> 4) & 0xf;  /* Bits  7 - 4 */
+		step   = a & 0xf;
+
+		/* extended family */
+		if (family == 0xf)
+			family += (a >> 20) & 0xff;
+
+		/* extended model */
+		if (family >= 0x6)
+			model += ((a >> 16) & 0xf) << 4;
+	}
+	nb = scnprintf(buffer, sz, "%s,%u,%u,%u$", vendor, family, model, step);
+
+	/* look for end marker to ensure the entire data fit */
+	if (strchr(buffer, '$')) {
+		buffer[nb-1] = '\0';
+		return 0;
+	}
+	return -1;
+}
diff --git a/tools/perf/arch/x86/util/kvm-stat.c b/tools/perf/arch/x86/util/kvm-stat.c
new file mode 100644
index 000000000..14e4e668f
--- /dev/null
+++ b/tools/perf/arch/x86/util/kvm-stat.c
@@ -0,0 +1,156 @@
+#include "../../util/kvm-stat.h"
+#include <asm/kvm_perf.h>
+
+define_exit_reasons_table(vmx_exit_reasons, VMX_EXIT_REASONS);
+define_exit_reasons_table(svm_exit_reasons, SVM_EXIT_REASONS);
+
+static struct kvm_events_ops exit_events = {
+	.is_begin_event = exit_event_begin,
+	.is_end_event = exit_event_end,
+	.decode_key = exit_event_decode_key,
+	.name = "VM-EXIT"
+};
+
+/*
+ * For the mmio events, we treat:
+ * the time of MMIO write: kvm_mmio(KVM_TRACE_MMIO_WRITE...) -> kvm_entry
+ * the time of MMIO read: kvm_exit -> kvm_mmio(KVM_TRACE_MMIO_READ...).
+ */
+static void mmio_event_get_key(struct perf_evsel *evsel, struct perf_sample *sample,
+			       struct event_key *key)
+{
+	key->key  = perf_evsel__intval(evsel, sample, "gpa");
+	key->info = perf_evsel__intval(evsel, sample, "type");
+}
+
+#define KVM_TRACE_MMIO_READ_UNSATISFIED 0
+#define KVM_TRACE_MMIO_READ 1
+#define KVM_TRACE_MMIO_WRITE 2
+
+static bool mmio_event_begin(struct perf_evsel *evsel,
+			     struct perf_sample *sample, struct event_key *key)
+{
+	/* MMIO read begin event in kernel. */
+	if (kvm_exit_event(evsel))
+		return true;
+
+	/* MMIO write begin event in kernel. */
+	if (!strcmp(evsel->name, "kvm:kvm_mmio") &&
+	    perf_evsel__intval(evsel, sample, "type") == KVM_TRACE_MMIO_WRITE) {
+		mmio_event_get_key(evsel, sample, key);
+		return true;
+	}
+
+	return false;
+}
+
+static bool mmio_event_end(struct perf_evsel *evsel, struct perf_sample *sample,
+			   struct event_key *key)
+{
+	/* MMIO write end event in kernel. */
+	if (kvm_entry_event(evsel))
+		return true;
+
+	/* MMIO read end event in kernel.*/
+	if (!strcmp(evsel->name, "kvm:kvm_mmio") &&
+	    perf_evsel__intval(evsel, sample, "type") == KVM_TRACE_MMIO_READ) {
+		mmio_event_get_key(evsel, sample, key);
+		return true;
+	}
+
+	return false;
+}
+
+static void mmio_event_decode_key(struct perf_kvm_stat *kvm __maybe_unused,
+				  struct event_key *key,
+				  char *decode)
+{
+	scnprintf(decode, DECODE_STR_LEN, "%#lx:%s",
+		  (unsigned long)key->key,
+		  key->info == KVM_TRACE_MMIO_WRITE ? "W" : "R");
+}
+
+static struct kvm_events_ops mmio_events = {
+	.is_begin_event = mmio_event_begin,
+	.is_end_event = mmio_event_end,
+	.decode_key = mmio_event_decode_key,
+	.name = "MMIO Access"
+};
+
+ /* The time of emulation pio access is from kvm_pio to kvm_entry. */
+static void ioport_event_get_key(struct perf_evsel *evsel,
+				 struct perf_sample *sample,
+				 struct event_key *key)
+{
+	key->key  = perf_evsel__intval(evsel, sample, "port");
+	key->info = perf_evsel__intval(evsel, sample, "rw");
+}
+
+static bool ioport_event_begin(struct perf_evsel *evsel,
+			       struct perf_sample *sample,
+			       struct event_key *key)
+{
+	if (!strcmp(evsel->name, "kvm:kvm_pio")) {
+		ioport_event_get_key(evsel, sample, key);
+		return true;
+	}
+
+	return false;
+}
+
+static bool ioport_event_end(struct perf_evsel *evsel,
+			     struct perf_sample *sample __maybe_unused,
+			     struct event_key *key __maybe_unused)
+{
+	return kvm_entry_event(evsel);
+}
+
+static void ioport_event_decode_key(struct perf_kvm_stat *kvm __maybe_unused,
+				    struct event_key *key,
+				    char *decode)
+{
+	scnprintf(decode, DECODE_STR_LEN, "%#llx:%s",
+		  (unsigned long long)key->key,
+		  key->info ? "POUT" : "PIN");
+}
+
+static struct kvm_events_ops ioport_events = {
+	.is_begin_event = ioport_event_begin,
+	.is_end_event = ioport_event_end,
+	.decode_key = ioport_event_decode_key,
+	.name = "IO Port Access"
+};
+
+const char * const kvm_events_tp[] = {
+	"kvm:kvm_entry",
+	"kvm:kvm_exit",
+	"kvm:kvm_mmio",
+	"kvm:kvm_pio",
+	NULL,
+};
+
+struct kvm_reg_events_ops kvm_reg_events_ops[] = {
+	{ .name = "vmexit", .ops = &exit_events },
+	{ .name = "mmio", .ops = &mmio_events },
+	{ .name = "ioport", .ops = &ioport_events },
+	{ NULL, NULL },
+};
+
+const char * const kvm_skip_events[] = {
+	"HLT",
+	NULL,
+};
+
+int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid)
+{
+	if (strstr(cpuid, "Intel")) {
+		kvm->exit_reasons = vmx_exit_reasons;
+		kvm->exit_reasons_isa = "VMX";
+	} else if (strstr(cpuid, "AMD")) {
+		kvm->exit_reasons = svm_exit_reasons;
+		kvm->exit_reasons_isa = "SVM";
+	} else
+		return -ENOTSUP;
+
+	return 0;
+}
diff --git a/tools/perf/arch/x86/util/tsc.c b/tools/perf/arch/x86/util/tsc.c
new file mode 100644
index 000000000..fd2868490
--- /dev/null
+++ b/tools/perf/arch/x86/util/tsc.c
@@ -0,0 +1,48 @@
+#include <stdbool.h>
+#include <errno.h>
+
+#include <linux/perf_event.h>
+
+#include "../../perf.h"
+#include <linux/types.h>
+#include "../../util/debug.h"
+#include "../../util/tsc.h"
+#include "tsc.h"
+
+int perf_read_tsc_conversion(const struct perf_event_mmap_page *pc,
+			     struct perf_tsc_conversion *tc)
+{
+	bool cap_user_time_zero;
+	u32 seq;
+	int i = 0;
+
+	while (1) {
+		seq = pc->lock;
+		rmb();
+		tc->time_mult = pc->time_mult;
+		tc->time_shift = pc->time_shift;
+		tc->time_zero = pc->time_zero;
+		cap_user_time_zero = pc->cap_user_time_zero;
+		rmb();
+		if (pc->lock == seq && !(seq & 1))
+			break;
+		if (++i > 10000) {
+			pr_debug("failed to get perf_event_mmap_page lock\n");
+			return -EINVAL;
+		}
+	}
+
+	if (!cap_user_time_zero)
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+u64 rdtsc(void)
+{
+	unsigned int low, high;
+
+	asm volatile("rdtsc" : "=a" (low), "=d" (high));
+
+	return low | ((u64)high) << 32;
+}
diff --git a/tools/perf/arch/x86/util/tsc.h b/tools/perf/arch/x86/util/tsc.h
new file mode 100644
index 000000000..2edc4d310
--- /dev/null
+++ b/tools/perf/arch/x86/util/tsc.h
@@ -0,0 +1,17 @@
+#ifndef TOOLS_PERF_ARCH_X86_UTIL_TSC_H__
+#define TOOLS_PERF_ARCH_X86_UTIL_TSC_H__
+
+#include <linux/types.h>
+
+struct perf_tsc_conversion {
+	u16 time_shift;
+	u32 time_mult;
+	u64 time_zero;
+};
+
+struct perf_event_mmap_page;
+
+int perf_read_tsc_conversion(const struct perf_event_mmap_page *pc,
+			     struct perf_tsc_conversion *tc);
+
+#endif /* TOOLS_PERF_ARCH_X86_UTIL_TSC_H__ */
diff --git a/tools/perf/arch/x86/util/unwind-libdw.c b/tools/perf/arch/x86/util/unwind-libdw.c
new file mode 100644
index 000000000..c4b72176c
--- /dev/null
+++ b/tools/perf/arch/x86/util/unwind-libdw.c
@@ -0,0 +1,51 @@
+#include <elfutils/libdwfl.h>
+#include "../../util/unwind-libdw.h"
+#include "../../util/perf_regs.h"
+
+bool libdw__arch_set_initial_registers(Dwfl_Thread *thread, void *arg)
+{
+	struct unwind_info *ui = arg;
+	struct regs_dump *user_regs = &ui->sample->user_regs;
+	Dwarf_Word dwarf_regs[17];
+	unsigned nregs;
+
+#define REG(r) ({						\
+	Dwarf_Word val = 0;					\
+	perf_reg_value(&val, user_regs, PERF_REG_X86_##r);	\
+	val;							\
+})
+
+	if (user_regs->abi == PERF_SAMPLE_REGS_ABI_32) {
+		dwarf_regs[0] = REG(AX);
+		dwarf_regs[1] = REG(CX);
+		dwarf_regs[2] = REG(DX);
+		dwarf_regs[3] = REG(BX);
+		dwarf_regs[4] = REG(SP);
+		dwarf_regs[5] = REG(BP);
+		dwarf_regs[6] = REG(SI);
+		dwarf_regs[7] = REG(DI);
+		dwarf_regs[8] = REG(IP);
+		nregs = 9;
+	} else {
+		dwarf_regs[0]  = REG(AX);
+		dwarf_regs[1]  = REG(DX);
+		dwarf_regs[2]  = REG(CX);
+		dwarf_regs[3]  = REG(BX);
+		dwarf_regs[4]  = REG(SI);
+		dwarf_regs[5]  = REG(DI);
+		dwarf_regs[6]  = REG(BP);
+		dwarf_regs[7]  = REG(SP);
+		dwarf_regs[8]  = REG(R8);
+		dwarf_regs[9]  = REG(R9);
+		dwarf_regs[10] = REG(R10);
+		dwarf_regs[11] = REG(R11);
+		dwarf_regs[12] = REG(R12);
+		dwarf_regs[13] = REG(R13);
+		dwarf_regs[14] = REG(R14);
+		dwarf_regs[15] = REG(R15);
+		dwarf_regs[16] = REG(IP);
+		nregs = 17;
+	}
+
+	return dwfl_thread_state_registers(thread, 0, nregs, dwarf_regs);
+}
diff --git a/tools/perf/arch/x86/util/unwind-libunwind.c b/tools/perf/arch/x86/util/unwind-libunwind.c
new file mode 100644
index 000000000..db25e93d9
--- /dev/null
+++ b/tools/perf/arch/x86/util/unwind-libunwind.c
@@ -0,0 +1,112 @@
+
+#include <errno.h>
+#include <libunwind.h>
+#include "perf_regs.h"
+#include "../../util/unwind.h"
+#include "../../util/debug.h"
+
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+int libunwind__arch_reg_id(int regnum)
+{
+	int id;
+
+	switch (regnum) {
+	case UNW_X86_64_RAX:
+		id = PERF_REG_X86_AX;
+		break;
+	case UNW_X86_64_RDX:
+		id = PERF_REG_X86_DX;
+		break;
+	case UNW_X86_64_RCX:
+		id = PERF_REG_X86_CX;
+		break;
+	case UNW_X86_64_RBX:
+		id = PERF_REG_X86_BX;
+		break;
+	case UNW_X86_64_RSI:
+		id = PERF_REG_X86_SI;
+		break;
+	case UNW_X86_64_RDI:
+		id = PERF_REG_X86_DI;
+		break;
+	case UNW_X86_64_RBP:
+		id = PERF_REG_X86_BP;
+		break;
+	case UNW_X86_64_RSP:
+		id = PERF_REG_X86_SP;
+		break;
+	case UNW_X86_64_R8:
+		id = PERF_REG_X86_R8;
+		break;
+	case UNW_X86_64_R9:
+		id = PERF_REG_X86_R9;
+		break;
+	case UNW_X86_64_R10:
+		id = PERF_REG_X86_R10;
+		break;
+	case UNW_X86_64_R11:
+		id = PERF_REG_X86_R11;
+		break;
+	case UNW_X86_64_R12:
+		id = PERF_REG_X86_R12;
+		break;
+	case UNW_X86_64_R13:
+		id = PERF_REG_X86_R13;
+		break;
+	case UNW_X86_64_R14:
+		id = PERF_REG_X86_R14;
+		break;
+	case UNW_X86_64_R15:
+		id = PERF_REG_X86_R15;
+		break;
+	case UNW_X86_64_RIP:
+		id = PERF_REG_X86_IP;
+		break;
+	default:
+		pr_err("unwind: invalid reg id %d\n", regnum);
+		return -EINVAL;
+	}
+
+	return id;
+}
+#else
+int libunwind__arch_reg_id(int regnum)
+{
+	int id;
+
+	switch (regnum) {
+	case UNW_X86_EAX:
+		id = PERF_REG_X86_AX;
+		break;
+	case UNW_X86_EDX:
+		id = PERF_REG_X86_DX;
+		break;
+	case UNW_X86_ECX:
+		id = PERF_REG_X86_CX;
+		break;
+	case UNW_X86_EBX:
+		id = PERF_REG_X86_BX;
+		break;
+	case UNW_X86_ESI:
+		id = PERF_REG_X86_SI;
+		break;
+	case UNW_X86_EDI:
+		id = PERF_REG_X86_DI;
+		break;
+	case UNW_X86_EBP:
+		id = PERF_REG_X86_BP;
+		break;
+	case UNW_X86_ESP:
+		id = PERF_REG_X86_SP;
+		break;
+	case UNW_X86_EIP:
+		id = PERF_REG_X86_IP;
+		break;
+	default:
+		pr_err("unwind: invalid reg id %d\n", regnum);
+		return -EINVAL;
+	}
+
+	return id;
+}
+#endif /* HAVE_ARCH_X86_64_SUPPORT */
diff --git a/tools/perf/bench/Build b/tools/perf/bench/Build
new file mode 100644
index 000000000..5ce98023d
--- /dev/null
+++ b/tools/perf/bench/Build
@@ -0,0 +1,11 @@
+perf-y += sched-messaging.o
+perf-y += sched-pipe.o
+perf-y += mem-memcpy.o
+perf-y += futex-hash.o
+perf-y += futex-wake.o
+perf-y += futex-requeue.o
+
+perf-$(CONFIG_X86_64) += mem-memcpy-x86-64-asm.o
+perf-$(CONFIG_X86_64) += mem-memset-x86-64-asm.o
+
+perf-$(CONFIG_NUMA) += numa.o
diff --git a/tools/perf/bench/bench.h b/tools/perf/bench/bench.h
new file mode 100644
index 000000000..3c4dd44d4
--- /dev/null
+++ b/tools/perf/bench/bench.h
@@ -0,0 +1,48 @@
+#ifndef BENCH_H
+#define BENCH_H
+
+/*
+ * The madvise transparent hugepage constants were added in glibc
+ * 2.13. For compatibility with older versions of glibc, define these
+ * tokens if they are not already defined.
+ *
+ * PA-RISC uses different madvise values from other architectures and
+ * needs to be special-cased.
+ */
+#ifdef __hppa__
+# ifndef MADV_HUGEPAGE
+#  define MADV_HUGEPAGE		67
+# endif
+# ifndef MADV_NOHUGEPAGE
+#  define MADV_NOHUGEPAGE	68
+# endif
+#else
+# ifndef MADV_HUGEPAGE
+#  define MADV_HUGEPAGE		14
+# endif
+# ifndef MADV_NOHUGEPAGE
+#  define MADV_NOHUGEPAGE	15
+# endif
+#endif
+
+extern int bench_numa(int argc, const char **argv, const char *prefix);
+extern int bench_sched_messaging(int argc, const char **argv, const char *prefix);
+extern int bench_sched_pipe(int argc, const char **argv, const char *prefix);
+extern int bench_mem_memcpy(int argc, const char **argv,
+			    const char *prefix __maybe_unused);
+extern int bench_mem_memset(int argc, const char **argv, const char *prefix);
+extern int bench_futex_hash(int argc, const char **argv, const char *prefix);
+extern int bench_futex_wake(int argc, const char **argv, const char *prefix);
+extern int bench_futex_requeue(int argc, const char **argv, const char *prefix);
+
+#define BENCH_FORMAT_DEFAULT_STR	"default"
+#define BENCH_FORMAT_DEFAULT		0
+#define BENCH_FORMAT_SIMPLE_STR		"simple"
+#define BENCH_FORMAT_SIMPLE		1
+
+#define BENCH_FORMAT_UNKNOWN		-1
+
+extern int bench_format;
+extern unsigned int bench_repeat;
+
+#endif
diff --git a/tools/perf/bench/futex-hash.c b/tools/perf/bench/futex-hash.c
new file mode 100644
index 000000000..fc9bebd2c
--- /dev/null
+++ b/tools/perf/bench/futex-hash.c
@@ -0,0 +1,215 @@
+/*
+ * Copyright (C) 2013  Davidlohr Bueso <davidlohr@hp.com>
+ *
+ * futex-hash: Stress the hell out of the Linux kernel futex uaddr hashing.
+ *
+ * This program is particularly useful for measuring the kernel's futex hash
+ * table/function implementation. In order for it to make sense, use with as
+ * many threads and futexes as possible.
+ */
+
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/stat.h"
+#include "../util/parse-options.h"
+#include "../util/header.h"
+#include "bench.h"
+#include "futex.h"
+
+#include <err.h>
+#include <stdlib.h>
+#include <sys/time.h>
+#include <pthread.h>
+
+static unsigned int nthreads = 0;
+static unsigned int nsecs    = 10;
+/* amount of futexes per thread */
+static unsigned int nfutexes = 1024;
+static bool fshared = false, done = false, silent = false;
+static int futex_flag = 0;
+
+struct timeval start, end, runtime;
+static pthread_mutex_t thread_lock;
+static unsigned int threads_starting;
+static struct stats throughput_stats;
+static pthread_cond_t thread_parent, thread_worker;
+
+struct worker {
+	int tid;
+	u_int32_t *futex;
+	pthread_t thread;
+	unsigned long ops;
+};
+
+static const struct option options[] = {
+	OPT_UINTEGER('t', "threads", &nthreads, "Specify amount of threads"),
+	OPT_UINTEGER('r', "runtime", &nsecs,    "Specify runtime (in seconds)"),
+	OPT_UINTEGER('f', "futexes", &nfutexes, "Specify amount of futexes per threads"),
+	OPT_BOOLEAN( 's', "silent",  &silent,   "Silent mode: do not display data/details"),
+	OPT_BOOLEAN( 'S', "shared",  &fshared,  "Use shared futexes instead of private ones"),
+	OPT_END()
+};
+
+static const char * const bench_futex_hash_usage[] = {
+	"perf bench futex hash <options>",
+	NULL
+};
+
+static void *workerfn(void *arg)
+{
+	int ret;
+	unsigned int i;
+	struct worker *w = (struct worker *) arg;
+
+	pthread_mutex_lock(&thread_lock);
+	threads_starting--;
+	if (!threads_starting)
+		pthread_cond_signal(&thread_parent);
+	pthread_cond_wait(&thread_worker, &thread_lock);
+	pthread_mutex_unlock(&thread_lock);
+
+	do {
+		for (i = 0; i < nfutexes; i++, w->ops++) {
+			/*
+			 * We want the futex calls to fail in order to stress
+			 * the hashing of uaddr and not measure other steps,
+			 * such as internal waitqueue handling, thus enlarging
+			 * the critical region protected by hb->lock.
+			 */
+			ret = futex_wait(&w->futex[i], 1234, NULL, futex_flag);
+			if (!silent &&
+			    (!ret || errno != EAGAIN || errno != EWOULDBLOCK))
+				warn("Non-expected futex return call");
+		}
+	}  while (!done);
+
+	return NULL;
+}
+
+static void toggle_done(int sig __maybe_unused,
+			siginfo_t *info __maybe_unused,
+			void *uc __maybe_unused)
+{
+	/* inform all threads that we're done for the day */
+	done = true;
+	gettimeofday(&end, NULL);
+	timersub(&end, &start, &runtime);
+}
+
+static void print_summary(void)
+{
+	unsigned long avg = avg_stats(&throughput_stats);
+	double stddev = stddev_stats(&throughput_stats);
+
+	printf("%sAveraged %ld operations/sec (+- %.2f%%), total secs = %d\n",
+	       !silent ? "\n" : "", avg, rel_stddev_stats(stddev, avg),
+	       (int) runtime.tv_sec);
+}
+
+int bench_futex_hash(int argc, const char **argv,
+		     const char *prefix __maybe_unused)
+{
+	int ret = 0;
+	cpu_set_t cpu;
+	struct sigaction act;
+	unsigned int i, ncpus;
+	pthread_attr_t thread_attr;
+	struct worker *worker = NULL;
+
+	argc = parse_options(argc, argv, options, bench_futex_hash_usage, 0);
+	if (argc) {
+		usage_with_options(bench_futex_hash_usage, options);
+		exit(EXIT_FAILURE);
+	}
+
+	ncpus = sysconf(_SC_NPROCESSORS_ONLN);
+
+	sigfillset(&act.sa_mask);
+	act.sa_sigaction = toggle_done;
+	sigaction(SIGINT, &act, NULL);
+
+	if (!nthreads) /* default to the number of CPUs */
+		nthreads = ncpus;
+
+	worker = calloc(nthreads, sizeof(*worker));
+	if (!worker)
+		goto errmem;
+
+	if (!fshared)
+		futex_flag = FUTEX_PRIVATE_FLAG;
+
+	printf("Run summary [PID %d]: %d threads, each operating on %d [%s] futexes for %d secs.\n\n",
+	       getpid(), nthreads, nfutexes, fshared ? "shared":"private", nsecs);
+
+	init_stats(&throughput_stats);
+	pthread_mutex_init(&thread_lock, NULL);
+	pthread_cond_init(&thread_parent, NULL);
+	pthread_cond_init(&thread_worker, NULL);
+
+	threads_starting = nthreads;
+	pthread_attr_init(&thread_attr);
+	gettimeofday(&start, NULL);
+	for (i = 0; i < nthreads; i++) {
+		worker[i].tid = i;
+		worker[i].futex = calloc(nfutexes, sizeof(*worker[i].futex));
+		if (!worker[i].futex)
+			goto errmem;
+
+		CPU_ZERO(&cpu);
+		CPU_SET(i % ncpus, &cpu);
+
+		ret = pthread_attr_setaffinity_np(&thread_attr, sizeof(cpu_set_t), &cpu);
+		if (ret)
+			err(EXIT_FAILURE, "pthread_attr_setaffinity_np");
+
+		ret = pthread_create(&worker[i].thread, &thread_attr, workerfn,
+				     (void *)(struct worker *) &worker[i]);
+		if (ret)
+			err(EXIT_FAILURE, "pthread_create");
+
+	}
+	pthread_attr_destroy(&thread_attr);
+
+	pthread_mutex_lock(&thread_lock);
+	while (threads_starting)
+		pthread_cond_wait(&thread_parent, &thread_lock);
+	pthread_cond_broadcast(&thread_worker);
+	pthread_mutex_unlock(&thread_lock);
+
+	sleep(nsecs);
+	toggle_done(0, NULL, NULL);
+
+	for (i = 0; i < nthreads; i++) {
+		ret = pthread_join(worker[i].thread, NULL);
+		if (ret)
+			err(EXIT_FAILURE, "pthread_join");
+	}
+
+	/* cleanup & report results */
+	pthread_cond_destroy(&thread_parent);
+	pthread_cond_destroy(&thread_worker);
+	pthread_mutex_destroy(&thread_lock);
+
+	for (i = 0; i < nthreads; i++) {
+		unsigned long t = worker[i].ops/runtime.tv_sec;
+		update_stats(&throughput_stats, t);
+		if (!silent) {
+			if (nfutexes == 1)
+				printf("[thread %2d] futex: %p [ %ld ops/sec ]\n",
+				       worker[i].tid, &worker[i].futex[0], t);
+			else
+				printf("[thread %2d] futexes: %p ... %p [ %ld ops/sec ]\n",
+				       worker[i].tid, &worker[i].futex[0],
+				       &worker[i].futex[nfutexes-1], t);
+		}
+
+		free(worker[i].futex);
+	}
+
+	print_summary();
+
+	free(worker);
+	return ret;
+errmem:
+	err(EXIT_FAILURE, "calloc");
+}
diff --git a/tools/perf/bench/futex-requeue.c b/tools/perf/bench/futex-requeue.c
new file mode 100644
index 000000000..ad0d9b534
--- /dev/null
+++ b/tools/perf/bench/futex-requeue.c
@@ -0,0 +1,212 @@
+/*
+ * Copyright (C) 2013  Davidlohr Bueso <davidlohr@hp.com>
+ *
+ * futex-requeue: Block a bunch of threads on futex1 and requeue them
+ *                on futex2, N at a time.
+ *
+ * This program is particularly useful to measure the latency of nthread
+ * requeues without waking up any tasks -- thus mimicking a regular futex_wait.
+ */
+
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/stat.h"
+#include "../util/parse-options.h"
+#include "../util/header.h"
+#include "bench.h"
+#include "futex.h"
+
+#include <err.h>
+#include <stdlib.h>
+#include <sys/time.h>
+#include <pthread.h>
+
+static u_int32_t futex1 = 0, futex2 = 0;
+
+/*
+ * How many tasks to requeue at a time.
+ * Default to 1 in order to make the kernel work more.
+ */
+static unsigned int nrequeue = 1;
+
+static pthread_t *worker;
+static bool done = false, silent = false, fshared = false;
+static pthread_mutex_t thread_lock;
+static pthread_cond_t thread_parent, thread_worker;
+static struct stats requeuetime_stats, requeued_stats;
+static unsigned int ncpus, threads_starting, nthreads = 0;
+static int futex_flag = 0;
+
+static const struct option options[] = {
+	OPT_UINTEGER('t', "threads",  &nthreads, "Specify amount of threads"),
+	OPT_UINTEGER('q', "nrequeue", &nrequeue, "Specify amount of threads to requeue at once"),
+	OPT_BOOLEAN( 's', "silent",   &silent,   "Silent mode: do not display data/details"),
+	OPT_BOOLEAN( 'S', "shared",   &fshared,  "Use shared futexes instead of private ones"),
+	OPT_END()
+};
+
+static const char * const bench_futex_requeue_usage[] = {
+	"perf bench futex requeue <options>",
+	NULL
+};
+
+static void print_summary(void)
+{
+	double requeuetime_avg = avg_stats(&requeuetime_stats);
+	double requeuetime_stddev = stddev_stats(&requeuetime_stats);
+	unsigned int requeued_avg = avg_stats(&requeued_stats);
+
+	printf("Requeued %d of %d threads in %.4f ms (+-%.2f%%)\n",
+	       requeued_avg,
+	       nthreads,
+	       requeuetime_avg/1e3,
+	       rel_stddev_stats(requeuetime_stddev, requeuetime_avg));
+}
+
+static void *workerfn(void *arg __maybe_unused)
+{
+	pthread_mutex_lock(&thread_lock);
+	threads_starting--;
+	if (!threads_starting)
+		pthread_cond_signal(&thread_parent);
+	pthread_cond_wait(&thread_worker, &thread_lock);
+	pthread_mutex_unlock(&thread_lock);
+
+	futex_wait(&futex1, 0, NULL, futex_flag);
+	return NULL;
+}
+
+static void block_threads(pthread_t *w,
+			  pthread_attr_t thread_attr)
+{
+	cpu_set_t cpu;
+	unsigned int i;
+
+	threads_starting = nthreads;
+
+	/* create and block all threads */
+	for (i = 0; i < nthreads; i++) {
+		CPU_ZERO(&cpu);
+		CPU_SET(i % ncpus, &cpu);
+
+		if (pthread_attr_setaffinity_np(&thread_attr, sizeof(cpu_set_t), &cpu))
+			err(EXIT_FAILURE, "pthread_attr_setaffinity_np");
+
+		if (pthread_create(&w[i], &thread_attr, workerfn, NULL))
+			err(EXIT_FAILURE, "pthread_create");
+	}
+}
+
+static void toggle_done(int sig __maybe_unused,
+			siginfo_t *info __maybe_unused,
+			void *uc __maybe_unused)
+{
+	done = true;
+}
+
+int bench_futex_requeue(int argc, const char **argv,
+			const char *prefix __maybe_unused)
+{
+	int ret = 0;
+	unsigned int i, j;
+	struct sigaction act;
+	pthread_attr_t thread_attr;
+
+	argc = parse_options(argc, argv, options, bench_futex_requeue_usage, 0);
+	if (argc)
+		goto err;
+
+	ncpus = sysconf(_SC_NPROCESSORS_ONLN);
+
+	sigfillset(&act.sa_mask);
+	act.sa_sigaction = toggle_done;
+	sigaction(SIGINT, &act, NULL);
+
+	if (!nthreads)
+		nthreads = ncpus;
+
+	worker = calloc(nthreads, sizeof(*worker));
+	if (!worker)
+		err(EXIT_FAILURE, "calloc");
+
+	if (!fshared)
+		futex_flag = FUTEX_PRIVATE_FLAG;
+
+	if (nrequeue > nthreads)
+		nrequeue = nthreads;
+
+	printf("Run summary [PID %d]: Requeuing %d threads (from [%s] %p to %p), "
+	       "%d at a time.\n\n",  getpid(), nthreads,
+	       fshared ? "shared":"private", &futex1, &futex2, nrequeue);
+
+	init_stats(&requeued_stats);
+	init_stats(&requeuetime_stats);
+	pthread_attr_init(&thread_attr);
+	pthread_mutex_init(&thread_lock, NULL);
+	pthread_cond_init(&thread_parent, NULL);
+	pthread_cond_init(&thread_worker, NULL);
+
+	for (j = 0; j < bench_repeat && !done; j++) {
+		unsigned int nrequeued = 0;
+		struct timeval start, end, runtime;
+
+		/* create, launch & block all threads */
+		block_threads(worker, thread_attr);
+
+		/* make sure all threads are already blocked */
+		pthread_mutex_lock(&thread_lock);
+		while (threads_starting)
+			pthread_cond_wait(&thread_parent, &thread_lock);
+		pthread_cond_broadcast(&thread_worker);
+		pthread_mutex_unlock(&thread_lock);
+
+		usleep(100000);
+
+		/* Ok, all threads are patiently blocked, start requeueing */
+		gettimeofday(&start, NULL);
+		while (nrequeued < nthreads) {
+			/*
+			 * Do not wakeup any tasks blocked on futex1, allowing
+			 * us to really measure futex_wait functionality.
+			 */
+			nrequeued += futex_cmp_requeue(&futex1, 0, &futex2, 0,
+						       nrequeue, futex_flag);
+		}
+
+		gettimeofday(&end, NULL);
+		timersub(&end, &start, &runtime);
+
+		update_stats(&requeued_stats, nrequeued);
+		update_stats(&requeuetime_stats, runtime.tv_usec);
+
+		if (!silent) {
+			printf("[Run %d]: Requeued %d of %d threads in %.4f ms\n",
+			       j + 1, nrequeued, nthreads, runtime.tv_usec/1e3);
+		}
+
+		/* everybody should be blocked on futex2, wake'em up */
+		nrequeued = futex_wake(&futex2, nrequeued, futex_flag);
+		if (nthreads != nrequeued)
+			warnx("couldn't wakeup all tasks (%d/%d)", nrequeued, nthreads);
+
+		for (i = 0; i < nthreads; i++) {
+			ret = pthread_join(worker[i], NULL);
+			if (ret)
+				err(EXIT_FAILURE, "pthread_join");
+		}
+	}
+
+	/* cleanup & report results */
+	pthread_cond_destroy(&thread_parent);
+	pthread_cond_destroy(&thread_worker);
+	pthread_mutex_destroy(&thread_lock);
+	pthread_attr_destroy(&thread_attr);
+
+	print_summary();
+
+	free(worker);
+	return ret;
+err:
+	usage_with_options(bench_futex_requeue_usage, options);
+	exit(EXIT_FAILURE);
+}
diff --git a/tools/perf/bench/futex-wake.c b/tools/perf/bench/futex-wake.c
new file mode 100644
index 000000000..929f762be
--- /dev/null
+++ b/tools/perf/bench/futex-wake.c
@@ -0,0 +1,198 @@
+/*
+ * Copyright (C) 2013  Davidlohr Bueso <davidlohr@hp.com>
+ *
+ * futex-wake: Block a bunch of threads on a futex and wake'em up, N at a time.
+ *
+ * This program is particularly useful to measure the latency of nthread wakeups
+ * in non-error situations:  all waiters are queued and all wake calls wakeup
+ * one or more tasks, and thus the waitqueue is never empty.
+ */
+
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/stat.h"
+#include "../util/parse-options.h"
+#include "../util/header.h"
+#include "bench.h"
+#include "futex.h"
+
+#include <err.h>
+#include <stdlib.h>
+#include <sys/time.h>
+#include <pthread.h>
+
+/* all threads will block on the same futex */
+static u_int32_t futex1 = 0;
+
+/*
+ * How many wakeups to do at a time.
+ * Default to 1 in order to make the kernel work more.
+ */
+static unsigned int nwakes = 1;
+
+pthread_t *worker;
+static bool done = false, silent = false, fshared = false;
+static pthread_mutex_t thread_lock;
+static pthread_cond_t thread_parent, thread_worker;
+static struct stats waketime_stats, wakeup_stats;
+static unsigned int ncpus, threads_starting, nthreads = 0;
+static int futex_flag = 0;
+
+static const struct option options[] = {
+	OPT_UINTEGER('t', "threads", &nthreads, "Specify amount of threads"),
+	OPT_UINTEGER('w', "nwakes",  &nwakes,   "Specify amount of threads to wake at once"),
+	OPT_BOOLEAN( 's', "silent",  &silent,   "Silent mode: do not display data/details"),
+	OPT_BOOLEAN( 'S', "shared",  &fshared,  "Use shared futexes instead of private ones"),
+	OPT_END()
+};
+
+static const char * const bench_futex_wake_usage[] = {
+	"perf bench futex wake <options>",
+	NULL
+};
+
+static void *workerfn(void *arg __maybe_unused)
+{
+	pthread_mutex_lock(&thread_lock);
+	threads_starting--;
+	if (!threads_starting)
+		pthread_cond_signal(&thread_parent);
+	pthread_cond_wait(&thread_worker, &thread_lock);
+	pthread_mutex_unlock(&thread_lock);
+
+	futex_wait(&futex1, 0, NULL, futex_flag);
+	return NULL;
+}
+
+static void print_summary(void)
+{
+	double waketime_avg = avg_stats(&waketime_stats);
+	double waketime_stddev = stddev_stats(&waketime_stats);
+	unsigned int wakeup_avg = avg_stats(&wakeup_stats);
+
+	printf("Wokeup %d of %d threads in %.4f ms (+-%.2f%%)\n",
+	       wakeup_avg,
+	       nthreads,
+	       waketime_avg/1e3,
+	       rel_stddev_stats(waketime_stddev, waketime_avg));
+}
+
+static void block_threads(pthread_t *w,
+			  pthread_attr_t thread_attr)
+{
+	cpu_set_t cpu;
+	unsigned int i;
+
+	threads_starting = nthreads;
+
+	/* create and block all threads */
+	for (i = 0; i < nthreads; i++) {
+		CPU_ZERO(&cpu);
+		CPU_SET(i % ncpus, &cpu);
+
+		if (pthread_attr_setaffinity_np(&thread_attr, sizeof(cpu_set_t), &cpu))
+			err(EXIT_FAILURE, "pthread_attr_setaffinity_np");
+
+		if (pthread_create(&w[i], &thread_attr, workerfn, NULL))
+			err(EXIT_FAILURE, "pthread_create");
+	}
+}
+
+static void toggle_done(int sig __maybe_unused,
+			siginfo_t *info __maybe_unused,
+			void *uc __maybe_unused)
+{
+	done = true;
+}
+
+int bench_futex_wake(int argc, const char **argv,
+		     const char *prefix __maybe_unused)
+{
+	int ret = 0;
+	unsigned int i, j;
+	struct sigaction act;
+	pthread_attr_t thread_attr;
+
+	argc = parse_options(argc, argv, options, bench_futex_wake_usage, 0);
+	if (argc) {
+		usage_with_options(bench_futex_wake_usage, options);
+		exit(EXIT_FAILURE);
+	}
+
+	ncpus = sysconf(_SC_NPROCESSORS_ONLN);
+
+	sigfillset(&act.sa_mask);
+	act.sa_sigaction = toggle_done;
+	sigaction(SIGINT, &act, NULL);
+
+	if (!nthreads)
+		nthreads = ncpus;
+
+	worker = calloc(nthreads, sizeof(*worker));
+	if (!worker)
+		err(EXIT_FAILURE, "calloc");
+
+	if (!fshared)
+		futex_flag = FUTEX_PRIVATE_FLAG;
+
+	printf("Run summary [PID %d]: blocking on %d threads (at [%s] futex %p), "
+	       "waking up %d at a time.\n\n",
+	       getpid(), nthreads, fshared ? "shared":"private",  &futex1, nwakes);
+
+	init_stats(&wakeup_stats);
+	init_stats(&waketime_stats);
+	pthread_attr_init(&thread_attr);
+	pthread_mutex_init(&thread_lock, NULL);
+	pthread_cond_init(&thread_parent, NULL);
+	pthread_cond_init(&thread_worker, NULL);
+
+	for (j = 0; j < bench_repeat && !done; j++) {
+		unsigned int nwoken = 0;
+		struct timeval start, end, runtime;
+
+		/* create, launch & block all threads */
+		block_threads(worker, thread_attr);
+
+		/* make sure all threads are already blocked */
+		pthread_mutex_lock(&thread_lock);
+		while (threads_starting)
+			pthread_cond_wait(&thread_parent, &thread_lock);
+		pthread_cond_broadcast(&thread_worker);
+		pthread_mutex_unlock(&thread_lock);
+
+		usleep(100000);
+
+		/* Ok, all threads are patiently blocked, start waking folks up */
+		gettimeofday(&start, NULL);
+		while (nwoken != nthreads)
+			nwoken += futex_wake(&futex1, nwakes, futex_flag);
+		gettimeofday(&end, NULL);
+		timersub(&end, &start, &runtime);
+
+		update_stats(&wakeup_stats, nwoken);
+		update_stats(&waketime_stats, runtime.tv_usec);
+
+		if (!silent) {
+			printf("[Run %d]: Wokeup %d of %d threads in %.4f ms\n",
+			       j + 1, nwoken, nthreads, runtime.tv_usec/1e3);
+		}
+
+		for (i = 0; i < nthreads; i++) {
+			ret = pthread_join(worker[i], NULL);
+			if (ret)
+				err(EXIT_FAILURE, "pthread_join");
+		}
+
+	}
+
+	/* cleanup & report results */
+	pthread_cond_destroy(&thread_parent);
+	pthread_cond_destroy(&thread_worker);
+	pthread_mutex_destroy(&thread_lock);
+	pthread_attr_destroy(&thread_attr);
+
+	print_summary();
+
+	free(worker);
+	return ret;
+}
diff --git a/tools/perf/bench/futex.h b/tools/perf/bench/futex.h
new file mode 100644
index 000000000..7ed22ff1e
--- /dev/null
+++ b/tools/perf/bench/futex.h
@@ -0,0 +1,84 @@
+/*
+ * Glibc independent futex library for testing kernel functionality.
+ * Shamelessly stolen from Darren Hart <dvhltc@us.ibm.com>
+ *    http://git.kernel.org/cgit/linux/kernel/git/dvhart/futextest.git/
+ */
+
+#ifndef _FUTEX_H
+#define _FUTEX_H
+
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <sys/types.h>
+#include <linux/futex.h>
+
+/**
+ * futex() - SYS_futex syscall wrapper
+ * @uaddr:	address of first futex
+ * @op:		futex op code
+ * @val:	typically expected value of uaddr, but varies by op
+ * @timeout:	typically an absolute struct timespec (except where noted
+ *		otherwise). Overloaded by some ops
+ * @uaddr2:	address of second futex for some ops\
+ * @val3:	varies by op
+ * @opflags:	flags to be bitwise OR'd with op, such as FUTEX_PRIVATE_FLAG
+ *
+ * futex() is used by all the following futex op wrappers. It can also be
+ * used for misuse and abuse testing. Generally, the specific op wrappers
+ * should be used instead. It is a macro instead of an static inline function as
+ * some of the types over overloaded (timeout is used for nr_requeue for
+ * example).
+ *
+ * These argument descriptions are the defaults for all
+ * like-named arguments in the following wrappers except where noted below.
+ */
+#define futex(uaddr, op, val, timeout, uaddr2, val3, opflags) \
+	syscall(SYS_futex, uaddr, op | opflags, val, timeout, uaddr2, val3)
+
+/**
+ * futex_wait() - block on uaddr with optional timeout
+ * @timeout:	relative timeout
+ */
+static inline int
+futex_wait(u_int32_t *uaddr, u_int32_t val, struct timespec *timeout, int opflags)
+{
+	return futex(uaddr, FUTEX_WAIT, val, timeout, NULL, 0, opflags);
+}
+
+/**
+ * futex_wake() - wake one or more tasks blocked on uaddr
+ * @nr_wake:	wake up to this many tasks
+ */
+static inline int
+futex_wake(u_int32_t *uaddr, int nr_wake, int opflags)
+{
+	return futex(uaddr, FUTEX_WAKE, nr_wake, NULL, NULL, 0, opflags);
+}
+
+/**
+* futex_cmp_requeue() - requeue tasks from uaddr to uaddr2
+* @nr_wake:        wake up to this many tasks
+* @nr_requeue:        requeue up to this many tasks
+*/
+static inline int
+futex_cmp_requeue(u_int32_t *uaddr, u_int32_t val, u_int32_t *uaddr2, int nr_wake,
+		 int nr_requeue, int opflags)
+{
+	return futex(uaddr, FUTEX_CMP_REQUEUE, nr_wake, nr_requeue, uaddr2,
+		 val, opflags);
+}
+
+#ifndef HAVE_PTHREAD_ATTR_SETAFFINITY_NP
+#include <pthread.h>
+static inline int pthread_attr_setaffinity_np(pthread_attr_t *attr,
+					      size_t cpusetsize,
+					      cpu_set_t *cpuset)
+{
+	attr = attr;
+	cpusetsize = cpusetsize;
+	cpuset = cpuset;
+	return 0;
+}
+#endif
+
+#endif /* _FUTEX_H */
diff --git a/tools/perf/bench/mem-memcpy-arch.h b/tools/perf/bench/mem-memcpy-arch.h
new file mode 100644
index 000000000..57b4ed871
--- /dev/null
+++ b/tools/perf/bench/mem-memcpy-arch.h
@@ -0,0 +1,12 @@
+
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+
+#define MEMCPY_FN(fn, name, desc)		\
+	extern void *fn(void *, const void *, size_t);
+
+#include "mem-memcpy-x86-64-asm-def.h"
+
+#undef MEMCPY_FN
+
+#endif
+
diff --git a/tools/perf/bench/mem-memcpy-x86-64-asm-def.h b/tools/perf/bench/mem-memcpy-x86-64-asm-def.h
new file mode 100644
index 000000000..8c0c1a277
--- /dev/null
+++ b/tools/perf/bench/mem-memcpy-x86-64-asm-def.h
@@ -0,0 +1,12 @@
+
+MEMCPY_FN(memcpy_orig,
+	"x86-64-unrolled",
+	"unrolled memcpy() in arch/x86/lib/memcpy_64.S")
+
+MEMCPY_FN(__memcpy,
+	"x86-64-movsq",
+	"movsq-based memcpy() in arch/x86/lib/memcpy_64.S")
+
+MEMCPY_FN(memcpy_erms,
+	"x86-64-movsb",
+	"movsb-based memcpy() in arch/x86/lib/memcpy_64.S")
diff --git a/tools/perf/bench/mem-memcpy-x86-64-asm.S b/tools/perf/bench/mem-memcpy-x86-64-asm.S
new file mode 100644
index 000000000..e4c2c3014
--- /dev/null
+++ b/tools/perf/bench/mem-memcpy-x86-64-asm.S
@@ -0,0 +1,10 @@
+#define memcpy MEMCPY /* don't hide glibc's memcpy() */
+#define altinstr_replacement text
+#define globl p2align 4; .globl
+#include "../../../arch/x86/lib/memcpy_64.S"
+/*
+ * We need to provide note.GNU-stack section, saying that we want
+ * NOT executable stack. Otherwise the final linking will assume that
+ * the ELF stack should not be restricted at all and set it RWX.
+ */
+.section .note.GNU-stack,"",@progbits
diff --git a/tools/perf/bench/mem-memcpy.c b/tools/perf/bench/mem-memcpy.c
new file mode 100644
index 000000000..d3dfb7936
--- /dev/null
+++ b/tools/perf/bench/mem-memcpy.c
@@ -0,0 +1,434 @@
+/*
+ * mem-memcpy.c
+ *
+ * memcpy: Simple memory copy in various ways
+ *
+ * Written by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
+ */
+
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/parse-options.h"
+#include "../util/header.h"
+#include "../util/cloexec.h"
+#include "bench.h"
+#include "mem-memcpy-arch.h"
+#include "mem-memset-arch.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/time.h>
+#include <errno.h>
+
+#define K 1024
+
+static const char	*length_str	= "1MB";
+static const char	*routine	= "default";
+static int		iterations	= 1;
+static bool		use_cycle;
+static int		cycle_fd;
+static bool		only_prefault;
+static bool		no_prefault;
+
+static const struct option options[] = {
+	OPT_STRING('l', "length", &length_str, "1MB",
+		    "Specify length of memory to copy. "
+		    "Available units: B, KB, MB, GB and TB (upper and lower)"),
+	OPT_STRING('r', "routine", &routine, "default",
+		    "Specify routine to copy, \"all\" runs all available routines"),
+	OPT_INTEGER('i', "iterations", &iterations,
+		    "repeat memcpy() invocation this number of times"),
+	OPT_BOOLEAN('c', "cycle", &use_cycle,
+		    "Use cycles event instead of gettimeofday() for measuring"),
+	OPT_BOOLEAN('o', "only-prefault", &only_prefault,
+		    "Show only the result with page faults before memcpy()"),
+	OPT_BOOLEAN('n', "no-prefault", &no_prefault,
+		    "Show only the result without page faults before memcpy()"),
+	OPT_END()
+};
+
+typedef void *(*memcpy_t)(void *, const void *, size_t);
+typedef void *(*memset_t)(void *, int, size_t);
+
+struct routine {
+	const char *name;
+	const char *desc;
+	union {
+		memcpy_t memcpy;
+		memset_t memset;
+	} fn;
+};
+
+struct routine memcpy_routines[] = {
+	{ .name = "default",
+	  .desc = "Default memcpy() provided by glibc",
+	  .fn.memcpy = memcpy },
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+
+#define MEMCPY_FN(_fn, _name, _desc) {.name = _name, .desc = _desc, .fn.memcpy = _fn},
+#include "mem-memcpy-x86-64-asm-def.h"
+#undef MEMCPY_FN
+
+#endif
+
+	{ NULL,
+	  NULL,
+	  {NULL}   }
+};
+
+static const char * const bench_mem_memcpy_usage[] = {
+	"perf bench mem memcpy <options>",
+	NULL
+};
+
+static struct perf_event_attr cycle_attr = {
+	.type		= PERF_TYPE_HARDWARE,
+	.config		= PERF_COUNT_HW_CPU_CYCLES
+};
+
+static void init_cycle(void)
+{
+	cycle_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1,
+				       perf_event_open_cloexec_flag());
+
+	if (cycle_fd < 0 && errno == ENOSYS)
+		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
+	else
+		BUG_ON(cycle_fd < 0);
+}
+
+static u64 get_cycle(void)
+{
+	int ret;
+	u64 clk;
+
+	ret = read(cycle_fd, &clk, sizeof(u64));
+	BUG_ON(ret != sizeof(u64));
+
+	return clk;
+}
+
+static double timeval2double(struct timeval *ts)
+{
+	return (double)ts->tv_sec +
+		(double)ts->tv_usec / (double)1000000;
+}
+
+#define pf (no_prefault ? 0 : 1)
+
+#define print_bps(x) do {					\
+		if (x < K)					\
+			printf(" %14lf B/Sec", x);		\
+		else if (x < K * K)				\
+			printf(" %14lfd KB/Sec", x / K);	\
+		else if (x < K * K * K)				\
+			printf(" %14lf MB/Sec", x / K / K);	\
+		else						\
+			printf(" %14lf GB/Sec", x / K / K / K); \
+	} while (0)
+
+struct bench_mem_info {
+	const struct routine *routines;
+	u64 (*do_cycle)(const struct routine *r, size_t len, bool prefault);
+	double (*do_gettimeofday)(const struct routine *r, size_t len, bool prefault);
+	const char *const *usage;
+};
+
+static void __bench_mem_routine(struct bench_mem_info *info, int r_idx, size_t len, double totallen)
+{
+	const struct routine *r = &info->routines[r_idx];
+	double result_bps[2];
+	u64 result_cycle[2];
+
+	result_cycle[0] = result_cycle[1] = 0ULL;
+	result_bps[0] = result_bps[1] = 0.0;
+
+	printf("Routine %s (%s)\n", r->name, r->desc);
+
+	if (bench_format == BENCH_FORMAT_DEFAULT)
+		printf("# Copying %s Bytes ...\n\n", length_str);
+
+	if (!only_prefault && !no_prefault) {
+		/* show both of results */
+		if (use_cycle) {
+			result_cycle[0] = info->do_cycle(r, len, false);
+			result_cycle[1] = info->do_cycle(r, len, true);
+		} else {
+			result_bps[0]   = info->do_gettimeofday(r, len, false);
+			result_bps[1]   = info->do_gettimeofday(r, len, true);
+		}
+	} else {
+		if (use_cycle)
+			result_cycle[pf] = info->do_cycle(r, len, only_prefault);
+		else
+			result_bps[pf] = info->do_gettimeofday(r, len, only_prefault);
+	}
+
+	switch (bench_format) {
+	case BENCH_FORMAT_DEFAULT:
+		if (!only_prefault && !no_prefault) {
+			if (use_cycle) {
+				printf(" %14lf Cycle/Byte\n",
+					(double)result_cycle[0]
+					/ totallen);
+				printf(" %14lf Cycle/Byte (with prefault)\n",
+					(double)result_cycle[1]
+					/ totallen);
+			} else {
+				print_bps(result_bps[0]);
+				printf("\n");
+				print_bps(result_bps[1]);
+				printf(" (with prefault)\n");
+			}
+		} else {
+			if (use_cycle) {
+				printf(" %14lf Cycle/Byte",
+					(double)result_cycle[pf]
+					/ totallen);
+			} else
+				print_bps(result_bps[pf]);
+
+			printf("%s\n", only_prefault ? " (with prefault)" : "");
+		}
+		break;
+	case BENCH_FORMAT_SIMPLE:
+		if (!only_prefault && !no_prefault) {
+			if (use_cycle) {
+				printf("%lf %lf\n",
+					(double)result_cycle[0] / totallen,
+					(double)result_cycle[1] / totallen);
+			} else {
+				printf("%lf %lf\n",
+					result_bps[0], result_bps[1]);
+			}
+		} else {
+			if (use_cycle) {
+				printf("%lf\n", (double)result_cycle[pf]
+					/ totallen);
+			} else
+				printf("%lf\n", result_bps[pf]);
+		}
+		break;
+	default:
+		/* reaching this means there's some disaster: */
+		die("unknown format: %d\n", bench_format);
+		break;
+	}
+}
+
+static int bench_mem_common(int argc, const char **argv,
+		     const char *prefix __maybe_unused,
+		     struct bench_mem_info *info)
+{
+	int i;
+	size_t len;
+	double totallen;
+
+	argc = parse_options(argc, argv, options,
+			     info->usage, 0);
+
+	if (no_prefault && only_prefault) {
+		fprintf(stderr, "Invalid options: -o and -n are mutually exclusive\n");
+		return 1;
+	}
+
+	if (use_cycle)
+		init_cycle();
+
+	len = (size_t)perf_atoll((char *)length_str);
+	totallen = (double)len * iterations;
+
+	if ((s64)len <= 0) {
+		fprintf(stderr, "Invalid length:%s\n", length_str);
+		return 1;
+	}
+
+	/* same to without specifying either of prefault and no-prefault */
+	if (only_prefault && no_prefault)
+		only_prefault = no_prefault = false;
+
+	if (!strncmp(routine, "all", 3)) {
+		for (i = 0; info->routines[i].name; i++)
+			__bench_mem_routine(info, i, len, totallen);
+		return 0;
+	}
+
+	for (i = 0; info->routines[i].name; i++) {
+		if (!strcmp(info->routines[i].name, routine))
+			break;
+	}
+	if (!info->routines[i].name) {
+		printf("Unknown routine:%s\n", routine);
+		printf("Available routines...\n");
+		for (i = 0; info->routines[i].name; i++) {
+			printf("\t%s ... %s\n",
+			       info->routines[i].name, info->routines[i].desc);
+		}
+		return 1;
+	}
+
+	__bench_mem_routine(info, i, len, totallen);
+
+	return 0;
+}
+
+static void memcpy_alloc_mem(void **dst, void **src, size_t length)
+{
+	*dst = zalloc(length);
+	if (!*dst)
+		die("memory allocation failed - maybe length is too large?\n");
+
+	*src = zalloc(length);
+	if (!*src)
+		die("memory allocation failed - maybe length is too large?\n");
+	/* Make sure to always replace the zero pages even if MMAP_THRESH is crossed */
+	memset(*src, 0, length);
+}
+
+static u64 do_memcpy_cycle(const struct routine *r, size_t len, bool prefault)
+{
+	u64 cycle_start = 0ULL, cycle_end = 0ULL;
+	void *src = NULL, *dst = NULL;
+	memcpy_t fn = r->fn.memcpy;
+	int i;
+
+	memcpy_alloc_mem(&dst, &src, len);
+
+	if (prefault)
+		fn(dst, src, len);
+
+	cycle_start = get_cycle();
+	for (i = 0; i < iterations; ++i)
+		fn(dst, src, len);
+	cycle_end = get_cycle();
+
+	free(src);
+	free(dst);
+	return cycle_end - cycle_start;
+}
+
+static double do_memcpy_gettimeofday(const struct routine *r, size_t len,
+				     bool prefault)
+{
+	struct timeval tv_start, tv_end, tv_diff;
+	memcpy_t fn = r->fn.memcpy;
+	void *src = NULL, *dst = NULL;
+	int i;
+
+	memcpy_alloc_mem(&dst, &src, len);
+
+	if (prefault)
+		fn(dst, src, len);
+
+	BUG_ON(gettimeofday(&tv_start, NULL));
+	for (i = 0; i < iterations; ++i)
+		fn(dst, src, len);
+	BUG_ON(gettimeofday(&tv_end, NULL));
+
+	timersub(&tv_end, &tv_start, &tv_diff);
+
+	free(src);
+	free(dst);
+	return (double)(((double)len * iterations) / timeval2double(&tv_diff));
+}
+
+int bench_mem_memcpy(int argc, const char **argv,
+		     const char *prefix __maybe_unused)
+{
+	struct bench_mem_info info = {
+		.routines = memcpy_routines,
+		.do_cycle = do_memcpy_cycle,
+		.do_gettimeofday = do_memcpy_gettimeofday,
+		.usage = bench_mem_memcpy_usage,
+	};
+
+	return bench_mem_common(argc, argv, prefix, &info);
+}
+
+static void memset_alloc_mem(void **dst, size_t length)
+{
+	*dst = zalloc(length);
+	if (!*dst)
+		die("memory allocation failed - maybe length is too large?\n");
+}
+
+static u64 do_memset_cycle(const struct routine *r, size_t len, bool prefault)
+{
+	u64 cycle_start = 0ULL, cycle_end = 0ULL;
+	memset_t fn = r->fn.memset;
+	void *dst = NULL;
+	int i;
+
+	memset_alloc_mem(&dst, len);
+
+	if (prefault)
+		fn(dst, -1, len);
+
+	cycle_start = get_cycle();
+	for (i = 0; i < iterations; ++i)
+		fn(dst, i, len);
+	cycle_end = get_cycle();
+
+	free(dst);
+	return cycle_end - cycle_start;
+}
+
+static double do_memset_gettimeofday(const struct routine *r, size_t len,
+				     bool prefault)
+{
+	struct timeval tv_start, tv_end, tv_diff;
+	memset_t fn = r->fn.memset;
+	void *dst = NULL;
+	int i;
+
+	memset_alloc_mem(&dst, len);
+
+	if (prefault)
+		fn(dst, -1, len);
+
+	BUG_ON(gettimeofday(&tv_start, NULL));
+	for (i = 0; i < iterations; ++i)
+		fn(dst, i, len);
+	BUG_ON(gettimeofday(&tv_end, NULL));
+
+	timersub(&tv_end, &tv_start, &tv_diff);
+
+	free(dst);
+	return (double)(((double)len * iterations) / timeval2double(&tv_diff));
+}
+
+static const char * const bench_mem_memset_usage[] = {
+	"perf bench mem memset <options>",
+	NULL
+};
+
+static const struct routine memset_routines[] = {
+	{ .name ="default",
+	  .desc = "Default memset() provided by glibc",
+	  .fn.memset = memset },
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+
+#define MEMSET_FN(_fn, _name, _desc) { .name = _name, .desc = _desc, .fn.memset = _fn },
+#include "mem-memset-x86-64-asm-def.h"
+#undef MEMSET_FN
+
+#endif
+
+	{ .name = NULL,
+	  .desc = NULL,
+	  .fn.memset = NULL   }
+};
+
+int bench_mem_memset(int argc, const char **argv,
+		     const char *prefix __maybe_unused)
+{
+	struct bench_mem_info info = {
+		.routines = memset_routines,
+		.do_cycle = do_memset_cycle,
+		.do_gettimeofday = do_memset_gettimeofday,
+		.usage = bench_mem_memset_usage,
+	};
+
+	return bench_mem_common(argc, argv, prefix, &info);
+}
diff --git a/tools/perf/bench/mem-memset-arch.h b/tools/perf/bench/mem-memset-arch.h
new file mode 100644
index 000000000..633800cb0
--- /dev/null
+++ b/tools/perf/bench/mem-memset-arch.h
@@ -0,0 +1,12 @@
+
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+
+#define MEMSET_FN(fn, name, desc)		\
+	extern void *fn(void *, int, size_t);
+
+#include "mem-memset-x86-64-asm-def.h"
+
+#undef MEMSET_FN
+
+#endif
+
diff --git a/tools/perf/bench/mem-memset-x86-64-asm-def.h b/tools/perf/bench/mem-memset-x86-64-asm-def.h
new file mode 100644
index 000000000..f02d02877
--- /dev/null
+++ b/tools/perf/bench/mem-memset-x86-64-asm-def.h
@@ -0,0 +1,12 @@
+
+MEMSET_FN(memset_orig,
+	"x86-64-unrolled",
+	"unrolled memset() in arch/x86/lib/memset_64.S")
+
+MEMSET_FN(__memset,
+	"x86-64-stosq",
+	"movsq-based memset() in arch/x86/lib/memset_64.S")
+
+MEMSET_FN(memset_erms,
+	"x86-64-stosb",
+	"movsb-based memset() in arch/x86/lib/memset_64.S")
diff --git a/tools/perf/bench/mem-memset-x86-64-asm.S b/tools/perf/bench/mem-memset-x86-64-asm.S
new file mode 100644
index 000000000..de278784c
--- /dev/null
+++ b/tools/perf/bench/mem-memset-x86-64-asm.S
@@ -0,0 +1,11 @@
+#define memset MEMSET /* don't hide glibc's memset() */
+#define altinstr_replacement text
+#define globl p2align 4; .globl
+#include "../../../arch/x86/lib/memset_64.S"
+
+/*
+ * We need to provide note.GNU-stack section, saying that we want
+ * NOT executable stack. Otherwise the final linking will assume that
+ * the ELF stack should not be restricted at all and set it RWX.
+ */
+.section .note.GNU-stack,"",@progbits
diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c
new file mode 100644
index 000000000..ba5efa471
--- /dev/null
+++ b/tools/perf/bench/numa.c
@@ -0,0 +1,1752 @@
+/*
+ * numa.c
+ *
+ * numa: Simulate NUMA-sensitive workload and measure their NUMA performance
+ */
+
+#include "../perf.h"
+#include "../builtin.h"
+#include "../util/util.h"
+#include "../util/parse-options.h"
+
+#include "bench.h"
+
+#include <errno.h>
+#include <sched.h>
+#include <stdio.h>
+#include <assert.h>
+#include <malloc.h>
+#include <signal.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <pthread.h>
+#include <sys/mman.h>
+#include <sys/time.h>
+#include <sys/wait.h>
+#include <sys/prctl.h>
+#include <sys/types.h>
+
+#include <numa.h>
+#include <numaif.h>
+
+/*
+ * Regular printout to the terminal, supressed if -q is specified:
+ */
+#define tprintf(x...) do { if (g && g->p.show_details >= 0) printf(x); } while (0)
+
+/*
+ * Debug printf:
+ */
+#define dprintf(x...) do { if (g && g->p.show_details >= 1) printf(x); } while (0)
+
+struct thread_data {
+	int			curr_cpu;
+	cpu_set_t		bind_cpumask;
+	int			bind_node;
+	u8			*process_data;
+	int			process_nr;
+	int			thread_nr;
+	int			task_nr;
+	unsigned int		loops_done;
+	u64			val;
+	u64			runtime_ns;
+	pthread_mutex_t		*process_lock;
+};
+
+/* Parameters set by options: */
+
+struct params {
+	/* Startup synchronization: */
+	bool			serialize_startup;
+
+	/* Task hierarchy: */
+	int			nr_proc;
+	int			nr_threads;
+
+	/* Working set sizes: */
+	const char		*mb_global_str;
+	const char		*mb_proc_str;
+	const char		*mb_proc_locked_str;
+	const char		*mb_thread_str;
+
+	double			mb_global;
+	double			mb_proc;
+	double			mb_proc_locked;
+	double			mb_thread;
+
+	/* Access patterns to the working set: */
+	bool			data_reads;
+	bool			data_writes;
+	bool			data_backwards;
+	bool			data_zero_memset;
+	bool			data_rand_walk;
+	u32			nr_loops;
+	u32			nr_secs;
+	u32			sleep_usecs;
+
+	/* Working set initialization: */
+	bool			init_zero;
+	bool			init_random;
+	bool			init_cpu0;
+
+	/* Misc options: */
+	int			show_details;
+	int			run_all;
+	int			thp;
+
+	long			bytes_global;
+	long			bytes_process;
+	long			bytes_process_locked;
+	long			bytes_thread;
+
+	int			nr_tasks;
+	bool			show_quiet;
+
+	bool			show_convergence;
+	bool			measure_convergence;
+
+	int			perturb_secs;
+	int			nr_cpus;
+	int			nr_nodes;
+
+	/* Affinity options -C and -N: */
+	char			*cpu_list_str;
+	char			*node_list_str;
+};
+
+
+/* Global, read-writable area, accessible to all processes and threads: */
+
+struct global_info {
+	u8			*data;
+
+	pthread_mutex_t		startup_mutex;
+	int			nr_tasks_started;
+
+	pthread_mutex_t		startup_done_mutex;
+
+	pthread_mutex_t		start_work_mutex;
+	int			nr_tasks_working;
+
+	pthread_mutex_t		stop_work_mutex;
+	u64			bytes_done;
+
+	struct thread_data	*threads;
+
+	/* Convergence latency measurement: */
+	bool			all_converged;
+	bool			stop_work;
+
+	int			print_once;
+
+	struct params		p;
+};
+
+static struct global_info	*g = NULL;
+
+static int parse_cpus_opt(const struct option *opt, const char *arg, int unset);
+static int parse_nodes_opt(const struct option *opt, const char *arg, int unset);
+
+struct params p0;
+
+static const struct option options[] = {
+	OPT_INTEGER('p', "nr_proc"	, &p0.nr_proc,		"number of processes"),
+	OPT_INTEGER('t', "nr_threads"	, &p0.nr_threads,	"number of threads per process"),
+
+	OPT_STRING('G', "mb_global"	, &p0.mb_global_str,	"MB", "global  memory (MBs)"),
+	OPT_STRING('P', "mb_proc"	, &p0.mb_proc_str,	"MB", "process memory (MBs)"),
+	OPT_STRING('L', "mb_proc_locked", &p0.mb_proc_locked_str,"MB", "process serialized/locked memory access (MBs), <= process_memory"),
+	OPT_STRING('T', "mb_thread"	, &p0.mb_thread_str,	"MB", "thread  memory (MBs)"),
+
+	OPT_UINTEGER('l', "nr_loops"	, &p0.nr_loops,		"max number of loops to run"),
+	OPT_UINTEGER('s', "nr_secs"	, &p0.nr_secs,		"max number of seconds to run"),
+	OPT_UINTEGER('u', "usleep"	, &p0.sleep_usecs,	"usecs to sleep per loop iteration"),
+
+	OPT_BOOLEAN('R', "data_reads"	, &p0.data_reads,	"access the data via writes (can be mixed with -W)"),
+	OPT_BOOLEAN('W', "data_writes"	, &p0.data_writes,	"access the data via writes (can be mixed with -R)"),
+	OPT_BOOLEAN('B', "data_backwards", &p0.data_backwards,	"access the data backwards as well"),
+	OPT_BOOLEAN('Z', "data_zero_memset", &p0.data_zero_memset,"access the data via glibc bzero only"),
+	OPT_BOOLEAN('r', "data_rand_walk", &p0.data_rand_walk,	"access the data with random (32bit LFSR) walk"),
+
+
+	OPT_BOOLEAN('z', "init_zero"	, &p0.init_zero,	"bzero the initial allocations"),
+	OPT_BOOLEAN('I', "init_random"	, &p0.init_random,	"randomize the contents of the initial allocations"),
+	OPT_BOOLEAN('0', "init_cpu0"	, &p0.init_cpu0,	"do the initial allocations on CPU#0"),
+	OPT_INTEGER('x', "perturb_secs", &p0.perturb_secs,	"perturb thread 0/0 every X secs, to test convergence stability"),
+
+	OPT_INCR   ('d', "show_details"	, &p0.show_details,	"Show details"),
+	OPT_INCR   ('a', "all"		, &p0.run_all,		"Run all tests in the suite"),
+	OPT_INTEGER('H', "thp"		, &p0.thp,		"MADV_NOHUGEPAGE < 0 < MADV_HUGEPAGE"),
+	OPT_BOOLEAN('c', "show_convergence", &p0.show_convergence, "show convergence details"),
+	OPT_BOOLEAN('m', "measure_convergence",	&p0.measure_convergence, "measure convergence latency"),
+	OPT_BOOLEAN('q', "quiet"	, &p0.show_quiet,	"quiet mode"),
+	OPT_BOOLEAN('S', "serialize-startup", &p0.serialize_startup,"serialize thread startup"),
+
+	/* Special option string parsing callbacks: */
+        OPT_CALLBACK('C', "cpus", NULL, "cpu[,cpu2,...cpuN]",
+			"bind the first N tasks to these specific cpus (the rest is unbound)",
+			parse_cpus_opt),
+        OPT_CALLBACK('M', "memnodes", NULL, "node[,node2,...nodeN]",
+			"bind the first N tasks to these specific memory nodes (the rest is unbound)",
+			parse_nodes_opt),
+	OPT_END()
+};
+
+static const char * const bench_numa_usage[] = {
+	"perf bench numa <options>",
+	NULL
+};
+
+static const char * const numa_usage[] = {
+	"perf bench numa mem [<options>]",
+	NULL
+};
+
+static cpu_set_t bind_to_cpu(int target_cpu)
+{
+	cpu_set_t orig_mask, mask;
+	int ret;
+
+	ret = sched_getaffinity(0, sizeof(orig_mask), &orig_mask);
+	BUG_ON(ret);
+
+	CPU_ZERO(&mask);
+
+	if (target_cpu == -1) {
+		int cpu;
+
+		for (cpu = 0; cpu < g->p.nr_cpus; cpu++)
+			CPU_SET(cpu, &mask);
+	} else {
+		BUG_ON(target_cpu < 0 || target_cpu >= g->p.nr_cpus);
+		CPU_SET(target_cpu, &mask);
+	}
+
+	ret = sched_setaffinity(0, sizeof(mask), &mask);
+	BUG_ON(ret);
+
+	return orig_mask;
+}
+
+static cpu_set_t bind_to_node(int target_node)
+{
+	int cpus_per_node = g->p.nr_cpus/g->p.nr_nodes;
+	cpu_set_t orig_mask, mask;
+	int cpu;
+	int ret;
+
+	BUG_ON(cpus_per_node*g->p.nr_nodes != g->p.nr_cpus);
+	BUG_ON(!cpus_per_node);
+
+	ret = sched_getaffinity(0, sizeof(orig_mask), &orig_mask);
+	BUG_ON(ret);
+
+	CPU_ZERO(&mask);
+
+	if (target_node == -1) {
+		for (cpu = 0; cpu < g->p.nr_cpus; cpu++)
+			CPU_SET(cpu, &mask);
+	} else {
+		int cpu_start = (target_node + 0) * cpus_per_node;
+		int cpu_stop  = (target_node + 1) * cpus_per_node;
+
+		BUG_ON(cpu_stop > g->p.nr_cpus);
+
+		for (cpu = cpu_start; cpu < cpu_stop; cpu++)
+			CPU_SET(cpu, &mask);
+	}
+
+	ret = sched_setaffinity(0, sizeof(mask), &mask);
+	BUG_ON(ret);
+
+	return orig_mask;
+}
+
+static void bind_to_cpumask(cpu_set_t mask)
+{
+	int ret;
+
+	ret = sched_setaffinity(0, sizeof(mask), &mask);
+	BUG_ON(ret);
+}
+
+static void mempol_restore(void)
+{
+	int ret;
+
+	ret = set_mempolicy(MPOL_DEFAULT, NULL, g->p.nr_nodes-1);
+
+	BUG_ON(ret);
+}
+
+static void bind_to_memnode(int node)
+{
+	unsigned long nodemask;
+	int ret;
+
+	if (node == -1)
+		return;
+
+	BUG_ON(g->p.nr_nodes > (int)sizeof(nodemask));
+	nodemask = 1L << node;
+
+	ret = set_mempolicy(MPOL_BIND, &nodemask, sizeof(nodemask)*8);
+	dprintf("binding to node %d, mask: %016lx => %d\n", node, nodemask, ret);
+
+	BUG_ON(ret);
+}
+
+#define HPSIZE (2*1024*1024)
+
+#define set_taskname(fmt...)				\
+do {							\
+	char name[20];					\
+							\
+	snprintf(name, 20, fmt);			\
+	prctl(PR_SET_NAME, name);			\
+} while (0)
+
+static u8 *alloc_data(ssize_t bytes0, int map_flags,
+		      int init_zero, int init_cpu0, int thp, int init_random)
+{
+	cpu_set_t orig_mask;
+	ssize_t bytes;
+	u8 *buf;
+	int ret;
+
+	if (!bytes0)
+		return NULL;
+
+	/* Allocate and initialize all memory on CPU#0: */
+	if (init_cpu0) {
+		orig_mask = bind_to_node(0);
+		bind_to_memnode(0);
+	}
+
+	bytes = bytes0 + HPSIZE;
+
+	buf = (void *)mmap(0, bytes, PROT_READ|PROT_WRITE, MAP_ANON|map_flags, -1, 0);
+	BUG_ON(buf == (void *)-1);
+
+	if (map_flags == MAP_PRIVATE) {
+		if (thp > 0) {
+			ret = madvise(buf, bytes, MADV_HUGEPAGE);
+			if (ret && !g->print_once) {
+				g->print_once = 1;
+				printf("WARNING: Could not enable THP - do: 'echo madvise > /sys/kernel/mm/transparent_hugepage/enabled'\n");
+			}
+		}
+		if (thp < 0) {
+			ret = madvise(buf, bytes, MADV_NOHUGEPAGE);
+			if (ret && !g->print_once) {
+				g->print_once = 1;
+				printf("WARNING: Could not disable THP: run a CONFIG_TRANSPARENT_HUGEPAGE kernel?\n");
+			}
+		}
+	}
+
+	if (init_zero) {
+		bzero(buf, bytes);
+	} else {
+		/* Initialize random contents, different in each word: */
+		if (init_random) {
+			u64 *wbuf = (void *)buf;
+			long off = rand();
+			long i;
+
+			for (i = 0; i < bytes/8; i++)
+				wbuf[i] = i + off;
+		}
+	}
+
+	/* Align to 2MB boundary: */
+	buf = (void *)(((unsigned long)buf + HPSIZE-1) & ~(HPSIZE-1));
+
+	/* Restore affinity: */
+	if (init_cpu0) {
+		bind_to_cpumask(orig_mask);
+		mempol_restore();
+	}
+
+	return buf;
+}
+
+static void free_data(void *data, ssize_t bytes)
+{
+	int ret;
+
+	if (!data)
+		return;
+
+	ret = munmap(data, bytes);
+	BUG_ON(ret);
+}
+
+/*
+ * Create a shared memory buffer that can be shared between processes, zeroed:
+ */
+static void * zalloc_shared_data(ssize_t bytes)
+{
+	return alloc_data(bytes, MAP_SHARED, 1, g->p.init_cpu0,  g->p.thp, g->p.init_random);
+}
+
+/*
+ * Create a shared memory buffer that can be shared between processes:
+ */
+static void * setup_shared_data(ssize_t bytes)
+{
+	return alloc_data(bytes, MAP_SHARED, 0, g->p.init_cpu0,  g->p.thp, g->p.init_random);
+}
+
+/*
+ * Allocate process-local memory - this will either be shared between
+ * threads of this process, or only be accessed by this thread:
+ */
+static void * setup_private_data(ssize_t bytes)
+{
+	return alloc_data(bytes, MAP_PRIVATE, 0, g->p.init_cpu0,  g->p.thp, g->p.init_random);
+}
+
+/*
+ * Return a process-shared (global) mutex:
+ */
+static void init_global_mutex(pthread_mutex_t *mutex)
+{
+	pthread_mutexattr_t attr;
+
+	pthread_mutexattr_init(&attr);
+	pthread_mutexattr_setpshared(&attr, PTHREAD_PROCESS_SHARED);
+	pthread_mutex_init(mutex, &attr);
+}
+
+static int parse_cpu_list(const char *arg)
+{
+	p0.cpu_list_str = strdup(arg);
+
+	dprintf("got CPU list: {%s}\n", p0.cpu_list_str);
+
+	return 0;
+}
+
+static int parse_setup_cpu_list(void)
+{
+	struct thread_data *td;
+	char *str0, *str;
+	int t;
+
+	if (!g->p.cpu_list_str)
+		return 0;
+
+	dprintf("g->p.nr_tasks: %d\n", g->p.nr_tasks);
+
+	str0 = str = strdup(g->p.cpu_list_str);
+	t = 0;
+
+	BUG_ON(!str);
+
+	tprintf("# binding tasks to CPUs:\n");
+	tprintf("#  ");
+
+	while (true) {
+		int bind_cpu, bind_cpu_0, bind_cpu_1;
+		char *tok, *tok_end, *tok_step, *tok_len, *tok_mul;
+		int bind_len;
+		int step;
+		int mul;
+
+		tok = strsep(&str, ",");
+		if (!tok)
+			break;
+
+		tok_end = strstr(tok, "-");
+
+		dprintf("\ntoken: {%s}, end: {%s}\n", tok, tok_end);
+		if (!tok_end) {
+			/* Single CPU specified: */
+			bind_cpu_0 = bind_cpu_1 = atol(tok);
+		} else {
+			/* CPU range specified (for example: "5-11"): */
+			bind_cpu_0 = atol(tok);
+			bind_cpu_1 = atol(tok_end + 1);
+		}
+
+		step = 1;
+		tok_step = strstr(tok, "#");
+		if (tok_step) {
+			step = atol(tok_step + 1);
+			BUG_ON(step <= 0 || step >= g->p.nr_cpus);
+		}
+
+		/*
+		 * Mask length.
+		 * Eg: "--cpus 8_4-16#4" means: '--cpus 8_4,12_4,16_4',
+		 * where the _4 means the next 4 CPUs are allowed.
+		 */
+		bind_len = 1;
+		tok_len = strstr(tok, "_");
+		if (tok_len) {
+			bind_len = atol(tok_len + 1);
+			BUG_ON(bind_len <= 0 || bind_len > g->p.nr_cpus);
+		}
+
+		/* Multiplicator shortcut, "0x8" is a shortcut for: "0,0,0,0,0,0,0,0" */
+		mul = 1;
+		tok_mul = strstr(tok, "x");
+		if (tok_mul) {
+			mul = atol(tok_mul + 1);
+			BUG_ON(mul <= 0);
+		}
+
+		dprintf("CPUs: %d_%d-%d#%dx%d\n", bind_cpu_0, bind_len, bind_cpu_1, step, mul);
+
+		if (bind_cpu_0 >= g->p.nr_cpus || bind_cpu_1 >= g->p.nr_cpus) {
+			printf("\nTest not applicable, system has only %d CPUs.\n", g->p.nr_cpus);
+			return -1;
+		}
+
+		BUG_ON(bind_cpu_0 < 0 || bind_cpu_1 < 0);
+		BUG_ON(bind_cpu_0 > bind_cpu_1);
+
+		for (bind_cpu = bind_cpu_0; bind_cpu <= bind_cpu_1; bind_cpu += step) {
+			int i;
+
+			for (i = 0; i < mul; i++) {
+				int cpu;
+
+				if (t >= g->p.nr_tasks) {
+					printf("\n# NOTE: ignoring bind CPUs starting at CPU#%d\n #", bind_cpu);
+					goto out;
+				}
+				td = g->threads + t;
+
+				if (t)
+					tprintf(",");
+				if (bind_len > 1) {
+					tprintf("%2d/%d", bind_cpu, bind_len);
+				} else {
+					tprintf("%2d", bind_cpu);
+				}
+
+				CPU_ZERO(&td->bind_cpumask);
+				for (cpu = bind_cpu; cpu < bind_cpu+bind_len; cpu++) {
+					BUG_ON(cpu < 0 || cpu >= g->p.nr_cpus);
+					CPU_SET(cpu, &td->bind_cpumask);
+				}
+				t++;
+			}
+		}
+	}
+out:
+
+	tprintf("\n");
+
+	if (t < g->p.nr_tasks)
+		printf("# NOTE: %d tasks bound, %d tasks unbound\n", t, g->p.nr_tasks - t);
+
+	free(str0);
+	return 0;
+}
+
+static int parse_cpus_opt(const struct option *opt __maybe_unused,
+			  const char *arg, int unset __maybe_unused)
+{
+	if (!arg)
+		return -1;
+
+	return parse_cpu_list(arg);
+}
+
+static int parse_node_list(const char *arg)
+{
+	p0.node_list_str = strdup(arg);
+
+	dprintf("got NODE list: {%s}\n", p0.node_list_str);
+
+	return 0;
+}
+
+static int parse_setup_node_list(void)
+{
+	struct thread_data *td;
+	char *str0, *str;
+	int t;
+
+	if (!g->p.node_list_str)
+		return 0;
+
+	dprintf("g->p.nr_tasks: %d\n", g->p.nr_tasks);
+
+	str0 = str = strdup(g->p.node_list_str);
+	t = 0;
+
+	BUG_ON(!str);
+
+	tprintf("# binding tasks to NODEs:\n");
+	tprintf("# ");
+
+	while (true) {
+		int bind_node, bind_node_0, bind_node_1;
+		char *tok, *tok_end, *tok_step, *tok_mul;
+		int step;
+		int mul;
+
+		tok = strsep(&str, ",");
+		if (!tok)
+			break;
+
+		tok_end = strstr(tok, "-");
+
+		dprintf("\ntoken: {%s}, end: {%s}\n", tok, tok_end);
+		if (!tok_end) {
+			/* Single NODE specified: */
+			bind_node_0 = bind_node_1 = atol(tok);
+		} else {
+			/* NODE range specified (for example: "5-11"): */
+			bind_node_0 = atol(tok);
+			bind_node_1 = atol(tok_end + 1);
+		}
+
+		step = 1;
+		tok_step = strstr(tok, "#");
+		if (tok_step) {
+			step = atol(tok_step + 1);
+			BUG_ON(step <= 0 || step >= g->p.nr_nodes);
+		}
+
+		/* Multiplicator shortcut, "0x8" is a shortcut for: "0,0,0,0,0,0,0,0" */
+		mul = 1;
+		tok_mul = strstr(tok, "x");
+		if (tok_mul) {
+			mul = atol(tok_mul + 1);
+			BUG_ON(mul <= 0);
+		}
+
+		dprintf("NODEs: %d-%d #%d\n", bind_node_0, bind_node_1, step);
+
+		if (bind_node_0 >= g->p.nr_nodes || bind_node_1 >= g->p.nr_nodes) {
+			printf("\nTest not applicable, system has only %d nodes.\n", g->p.nr_nodes);
+			return -1;
+		}
+
+		BUG_ON(bind_node_0 < 0 || bind_node_1 < 0);
+		BUG_ON(bind_node_0 > bind_node_1);
+
+		for (bind_node = bind_node_0; bind_node <= bind_node_1; bind_node += step) {
+			int i;
+
+			for (i = 0; i < mul; i++) {
+				if (t >= g->p.nr_tasks) {
+					printf("\n# NOTE: ignoring bind NODEs starting at NODE#%d\n", bind_node);
+					goto out;
+				}
+				td = g->threads + t;
+
+				if (!t)
+					tprintf(" %2d", bind_node);
+				else
+					tprintf(",%2d", bind_node);
+
+				td->bind_node = bind_node;
+				t++;
+			}
+		}
+	}
+out:
+
+	tprintf("\n");
+
+	if (t < g->p.nr_tasks)
+		printf("# NOTE: %d tasks mem-bound, %d tasks unbound\n", t, g->p.nr_tasks - t);
+
+	free(str0);
+	return 0;
+}
+
+static int parse_nodes_opt(const struct option *opt __maybe_unused,
+			  const char *arg, int unset __maybe_unused)
+{
+	if (!arg)
+		return -1;
+
+	return parse_node_list(arg);
+
+	return 0;
+}
+
+#define BIT(x) (1ul << x)
+
+static inline uint32_t lfsr_32(uint32_t lfsr)
+{
+	const uint32_t taps = BIT(1) | BIT(5) | BIT(6) | BIT(31);
+	return (lfsr>>1) ^ ((0x0u - (lfsr & 0x1u)) & taps);
+}
+
+/*
+ * Make sure there's real data dependency to RAM (when read
+ * accesses are enabled), so the compiler, the CPU and the
+ * kernel (KSM, zero page, etc.) cannot optimize away RAM
+ * accesses:
+ */
+static inline u64 access_data(u64 *data __attribute__((unused)), u64 val)
+{
+	if (g->p.data_reads)
+		val += *data;
+	if (g->p.data_writes)
+		*data = val + 1;
+	return val;
+}
+
+/*
+ * The worker process does two types of work, a forwards going
+ * loop and a backwards going loop.
+ *
+ * We do this so that on multiprocessor systems we do not create
+ * a 'train' of processing, with highly synchronized processes,
+ * skewing the whole benchmark.
+ */
+static u64 do_work(u8 *__data, long bytes, int nr, int nr_max, int loop, u64 val)
+{
+	long words = bytes/sizeof(u64);
+	u64 *data = (void *)__data;
+	long chunk_0, chunk_1;
+	u64 *d0, *d, *d1;
+	long off;
+	long i;
+
+	BUG_ON(!data && words);
+	BUG_ON(data && !words);
+
+	if (!data)
+		return val;
+
+	/* Very simple memset() work variant: */
+	if (g->p.data_zero_memset && !g->p.data_rand_walk) {
+		bzero(data, bytes);
+		return val;
+	}
+
+	/* Spread out by PID/TID nr and by loop nr: */
+	chunk_0 = words/nr_max;
+	chunk_1 = words/g->p.nr_loops;
+	off = nr*chunk_0 + loop*chunk_1;
+
+	while (off >= words)
+		off -= words;
+
+	if (g->p.data_rand_walk) {
+		u32 lfsr = nr + loop + val;
+		int j;
+
+		for (i = 0; i < words/1024; i++) {
+			long start, end;
+
+			lfsr = lfsr_32(lfsr);
+
+			start = lfsr % words;
+			end = min(start + 1024, words-1);
+
+			if (g->p.data_zero_memset) {
+				bzero(data + start, (end-start) * sizeof(u64));
+			} else {
+				for (j = start; j < end; j++)
+					val = access_data(data + j, val);
+			}
+		}
+	} else if (!g->p.data_backwards || (nr + loop) & 1) {
+
+		d0 = data + off;
+		d  = data + off + 1;
+		d1 = data + words;
+
+		/* Process data forwards: */
+		for (;;) {
+			if (unlikely(d >= d1))
+				d = data;
+			if (unlikely(d == d0))
+				break;
+
+			val = access_data(d, val);
+
+			d++;
+		}
+	} else {
+		/* Process data backwards: */
+
+		d0 = data + off;
+		d  = data + off - 1;
+		d1 = data + words;
+
+		/* Process data forwards: */
+		for (;;) {
+			if (unlikely(d < data))
+				d = data + words-1;
+			if (unlikely(d == d0))
+				break;
+
+			val = access_data(d, val);
+
+			d--;
+		}
+	}
+
+	return val;
+}
+
+static void update_curr_cpu(int task_nr, unsigned long bytes_worked)
+{
+	unsigned int cpu;
+
+	cpu = sched_getcpu();
+
+	g->threads[task_nr].curr_cpu = cpu;
+	prctl(0, bytes_worked);
+}
+
+#define MAX_NR_NODES	64
+
+/*
+ * Count the number of nodes a process's threads
+ * are spread out on.
+ *
+ * A count of 1 means that the process is compressed
+ * to a single node. A count of g->p.nr_nodes means it's
+ * spread out on the whole system.
+ */
+static int count_process_nodes(int process_nr)
+{
+	char node_present[MAX_NR_NODES] = { 0, };
+	int nodes;
+	int n, t;
+
+	for (t = 0; t < g->p.nr_threads; t++) {
+		struct thread_data *td;
+		int task_nr;
+		int node;
+
+		task_nr = process_nr*g->p.nr_threads + t;
+		td = g->threads + task_nr;
+
+		node = numa_node_of_cpu(td->curr_cpu);
+		if (node < 0) /* curr_cpu was likely still -1 */
+			return 0;
+
+		node_present[node] = 1;
+	}
+
+	nodes = 0;
+
+	for (n = 0; n < MAX_NR_NODES; n++)
+		nodes += node_present[n];
+
+	return nodes;
+}
+
+/*
+ * Count the number of distinct process-threads a node contains.
+ *
+ * A count of 1 means that the node contains only a single
+ * process. If all nodes on the system contain at most one
+ * process then we are well-converged.
+ */
+static int count_node_processes(int node)
+{
+	int processes = 0;
+	int t, p;
+
+	for (p = 0; p < g->p.nr_proc; p++) {
+		for (t = 0; t < g->p.nr_threads; t++) {
+			struct thread_data *td;
+			int task_nr;
+			int n;
+
+			task_nr = p*g->p.nr_threads + t;
+			td = g->threads + task_nr;
+
+			n = numa_node_of_cpu(td->curr_cpu);
+			if (n == node) {
+				processes++;
+				break;
+			}
+		}
+	}
+
+	return processes;
+}
+
+static void calc_convergence_compression(int *strong)
+{
+	unsigned int nodes_min, nodes_max;
+	int p;
+
+	nodes_min = -1;
+	nodes_max =  0;
+
+	for (p = 0; p < g->p.nr_proc; p++) {
+		unsigned int nodes = count_process_nodes(p);
+
+		if (!nodes) {
+			*strong = 0;
+			return;
+		}
+
+		nodes_min = min(nodes, nodes_min);
+		nodes_max = max(nodes, nodes_max);
+	}
+
+	/* Strong convergence: all threads compress on a single node: */
+	if (nodes_min == 1 && nodes_max == 1) {
+		*strong = 1;
+	} else {
+		*strong = 0;
+		tprintf(" {%d-%d}", nodes_min, nodes_max);
+	}
+}
+
+static void calc_convergence(double runtime_ns_max, double *convergence)
+{
+	unsigned int loops_done_min, loops_done_max;
+	int process_groups;
+	int nodes[MAX_NR_NODES];
+	int distance;
+	int nr_min;
+	int nr_max;
+	int strong;
+	int sum;
+	int nr;
+	int node;
+	int cpu;
+	int t;
+
+	if (!g->p.show_convergence && !g->p.measure_convergence)
+		return;
+
+	for (node = 0; node < g->p.nr_nodes; node++)
+		nodes[node] = 0;
+
+	loops_done_min = -1;
+	loops_done_max = 0;
+
+	for (t = 0; t < g->p.nr_tasks; t++) {
+		struct thread_data *td = g->threads + t;
+		unsigned int loops_done;
+
+		cpu = td->curr_cpu;
+
+		/* Not all threads have written it yet: */
+		if (cpu < 0)
+			continue;
+
+		node = numa_node_of_cpu(cpu);
+
+		nodes[node]++;
+
+		loops_done = td->loops_done;
+		loops_done_min = min(loops_done, loops_done_min);
+		loops_done_max = max(loops_done, loops_done_max);
+	}
+
+	nr_max = 0;
+	nr_min = g->p.nr_tasks;
+	sum = 0;
+
+	for (node = 0; node < g->p.nr_nodes; node++) {
+		nr = nodes[node];
+		nr_min = min(nr, nr_min);
+		nr_max = max(nr, nr_max);
+		sum += nr;
+	}
+	BUG_ON(nr_min > nr_max);
+
+	BUG_ON(sum > g->p.nr_tasks);
+
+	if (0 && (sum < g->p.nr_tasks))
+		return;
+
+	/*
+	 * Count the number of distinct process groups present
+	 * on nodes - when we are converged this will decrease
+	 * to g->p.nr_proc:
+	 */
+	process_groups = 0;
+
+	for (node = 0; node < g->p.nr_nodes; node++) {
+		int processes = count_node_processes(node);
+
+		nr = nodes[node];
+		tprintf(" %2d/%-2d", nr, processes);
+
+		process_groups += processes;
+	}
+
+	distance = nr_max - nr_min;
+
+	tprintf(" [%2d/%-2d]", distance, process_groups);
+
+	tprintf(" l:%3d-%-3d (%3d)",
+		loops_done_min, loops_done_max, loops_done_max-loops_done_min);
+
+	if (loops_done_min && loops_done_max) {
+		double skew = 1.0 - (double)loops_done_min/loops_done_max;
+
+		tprintf(" [%4.1f%%]", skew * 100.0);
+	}
+
+	calc_convergence_compression(&strong);
+
+	if (strong && process_groups == g->p.nr_proc) {
+		if (!*convergence) {
+			*convergence = runtime_ns_max;
+			tprintf(" (%6.1fs converged)\n", *convergence/1e9);
+			if (g->p.measure_convergence) {
+				g->all_converged = true;
+				g->stop_work = true;
+			}
+		}
+	} else {
+		if (*convergence) {
+			tprintf(" (%6.1fs de-converged)", runtime_ns_max/1e9);
+			*convergence = 0;
+		}
+		tprintf("\n");
+	}
+}
+
+static void show_summary(double runtime_ns_max, int l, double *convergence)
+{
+	tprintf("\r #  %5.1f%%  [%.1f mins]",
+		(double)(l+1)/g->p.nr_loops*100.0, runtime_ns_max/1e9 / 60.0);
+
+	calc_convergence(runtime_ns_max, convergence);
+
+	if (g->p.show_details >= 0)
+		fflush(stdout);
+}
+
+static void *worker_thread(void *__tdata)
+{
+	struct thread_data *td = __tdata;
+	struct timeval start0, start, stop, diff;
+	int process_nr = td->process_nr;
+	int thread_nr = td->thread_nr;
+	unsigned long last_perturbance;
+	int task_nr = td->task_nr;
+	int details = g->p.show_details;
+	int first_task, last_task;
+	double convergence = 0;
+	u64 val = td->val;
+	double runtime_ns_max;
+	u8 *global_data;
+	u8 *process_data;
+	u8 *thread_data;
+	u64 bytes_done;
+	long work_done;
+	u32 l;
+
+	bind_to_cpumask(td->bind_cpumask);
+	bind_to_memnode(td->bind_node);
+
+	set_taskname("thread %d/%d", process_nr, thread_nr);
+
+	global_data = g->data;
+	process_data = td->process_data;
+	thread_data = setup_private_data(g->p.bytes_thread);
+
+	bytes_done = 0;
+
+	last_task = 0;
+	if (process_nr == g->p.nr_proc-1 && thread_nr == g->p.nr_threads-1)
+		last_task = 1;
+
+	first_task = 0;
+	if (process_nr == 0 && thread_nr == 0)
+		first_task = 1;
+
+	if (details >= 2) {
+		printf("#  thread %2d / %2d global mem: %p, process mem: %p, thread mem: %p\n",
+			process_nr, thread_nr, global_data, process_data, thread_data);
+	}
+
+	if (g->p.serialize_startup) {
+		pthread_mutex_lock(&g->startup_mutex);
+		g->nr_tasks_started++;
+		pthread_mutex_unlock(&g->startup_mutex);
+
+		/* Here we will wait for the main process to start us all at once: */
+		pthread_mutex_lock(&g->start_work_mutex);
+		g->nr_tasks_working++;
+
+		/* Last one wake the main process: */
+		if (g->nr_tasks_working == g->p.nr_tasks)
+			pthread_mutex_unlock(&g->startup_done_mutex);
+
+		pthread_mutex_unlock(&g->start_work_mutex);
+	}
+
+	gettimeofday(&start0, NULL);
+
+	start = stop = start0;
+	last_perturbance = start.tv_sec;
+
+	for (l = 0; l < g->p.nr_loops; l++) {
+		start = stop;
+
+		if (g->stop_work)
+			break;
+
+		val += do_work(global_data,  g->p.bytes_global,  process_nr, g->p.nr_proc,	l, val);
+		val += do_work(process_data, g->p.bytes_process, thread_nr,  g->p.nr_threads,	l, val);
+		val += do_work(thread_data,  g->p.bytes_thread,  0,          1,		l, val);
+
+		if (g->p.sleep_usecs) {
+			pthread_mutex_lock(td->process_lock);
+			usleep(g->p.sleep_usecs);
+			pthread_mutex_unlock(td->process_lock);
+		}
+		/*
+		 * Amount of work to be done under a process-global lock:
+		 */
+		if (g->p.bytes_process_locked) {
+			pthread_mutex_lock(td->process_lock);
+			val += do_work(process_data, g->p.bytes_process_locked, thread_nr,  g->p.nr_threads,	l, val);
+			pthread_mutex_unlock(td->process_lock);
+		}
+
+		work_done = g->p.bytes_global + g->p.bytes_process +
+			    g->p.bytes_process_locked + g->p.bytes_thread;
+
+		update_curr_cpu(task_nr, work_done);
+		bytes_done += work_done;
+
+		if (details < 0 && !g->p.perturb_secs && !g->p.measure_convergence && !g->p.nr_secs)
+			continue;
+
+		td->loops_done = l;
+
+		gettimeofday(&stop, NULL);
+
+		/* Check whether our max runtime timed out: */
+		if (g->p.nr_secs) {
+			timersub(&stop, &start0, &diff);
+			if ((u32)diff.tv_sec >= g->p.nr_secs) {
+				g->stop_work = true;
+				break;
+			}
+		}
+
+		/* Update the summary at most once per second: */
+		if (start.tv_sec == stop.tv_sec)
+			continue;
+
+		/*
+		 * Perturb the first task's equilibrium every g->p.perturb_secs seconds,
+		 * by migrating to CPU#0:
+		 */
+		if (first_task && g->p.perturb_secs && (int)(stop.tv_sec - last_perturbance) >= g->p.perturb_secs) {
+			cpu_set_t orig_mask;
+			int target_cpu;
+			int this_cpu;
+
+			last_perturbance = stop.tv_sec;
+
+			/*
+			 * Depending on where we are running, move into
+			 * the other half of the system, to create some
+			 * real disturbance:
+			 */
+			this_cpu = g->threads[task_nr].curr_cpu;
+			if (this_cpu < g->p.nr_cpus/2)
+				target_cpu = g->p.nr_cpus-1;
+			else
+				target_cpu = 0;
+
+			orig_mask = bind_to_cpu(target_cpu);
+
+			/* Here we are running on the target CPU already */
+			if (details >= 1)
+				printf(" (injecting perturbalance, moved to CPU#%d)\n", target_cpu);
+
+			bind_to_cpumask(orig_mask);
+		}
+
+		if (details >= 3) {
+			timersub(&stop, &start, &diff);
+			runtime_ns_max = diff.tv_sec * 1000000000;
+			runtime_ns_max += diff.tv_usec * 1000;
+
+			if (details >= 0) {
+				printf(" #%2d / %2d: %14.2lf nsecs/op [val: %016"PRIx64"]\n",
+					process_nr, thread_nr, runtime_ns_max / bytes_done, val);
+			}
+			fflush(stdout);
+		}
+		if (!last_task)
+			continue;
+
+		timersub(&stop, &start0, &diff);
+		runtime_ns_max = diff.tv_sec * 1000000000ULL;
+		runtime_ns_max += diff.tv_usec * 1000ULL;
+
+		show_summary(runtime_ns_max, l, &convergence);
+	}
+
+	gettimeofday(&stop, NULL);
+	timersub(&stop, &start0, &diff);
+	td->runtime_ns = diff.tv_sec * 1000000000ULL;
+	td->runtime_ns += diff.tv_usec * 1000ULL;
+
+	free_data(thread_data, g->p.bytes_thread);
+
+	pthread_mutex_lock(&g->stop_work_mutex);
+	g->bytes_done += bytes_done;
+	pthread_mutex_unlock(&g->stop_work_mutex);
+
+	return NULL;
+}
+
+/*
+ * A worker process starts a couple of threads:
+ */
+static void worker_process(int process_nr)
+{
+	pthread_mutex_t process_lock;
+	struct thread_data *td;
+	pthread_t *pthreads;
+	u8 *process_data;
+	int task_nr;
+	int ret;
+	int t;
+
+	pthread_mutex_init(&process_lock, NULL);
+	set_taskname("process %d", process_nr);
+
+	/*
+	 * Pick up the memory policy and the CPU binding of our first thread,
+	 * so that we initialize memory accordingly:
+	 */
+	task_nr = process_nr*g->p.nr_threads;
+	td = g->threads + task_nr;
+
+	bind_to_memnode(td->bind_node);
+	bind_to_cpumask(td->bind_cpumask);
+
+	pthreads = zalloc(g->p.nr_threads * sizeof(pthread_t));
+	process_data = setup_private_data(g->p.bytes_process);
+
+	if (g->p.show_details >= 3) {
+		printf(" # process %2d global mem: %p, process mem: %p\n",
+			process_nr, g->data, process_data);
+	}
+
+	for (t = 0; t < g->p.nr_threads; t++) {
+		task_nr = process_nr*g->p.nr_threads + t;
+		td = g->threads + task_nr;
+
+		td->process_data = process_data;
+		td->process_nr   = process_nr;
+		td->thread_nr    = t;
+		td->task_nr	 = task_nr;
+		td->val          = rand();
+		td->curr_cpu	 = -1;
+		td->process_lock = &process_lock;
+
+		ret = pthread_create(pthreads + t, NULL, worker_thread, td);
+		BUG_ON(ret);
+	}
+
+	for (t = 0; t < g->p.nr_threads; t++) {
+                ret = pthread_join(pthreads[t], NULL);
+		BUG_ON(ret);
+	}
+
+	free_data(process_data, g->p.bytes_process);
+	free(pthreads);
+}
+
+static void print_summary(void)
+{
+	if (g->p.show_details < 0)
+		return;
+
+	printf("\n ###\n");
+	printf(" # %d %s will execute (on %d nodes, %d CPUs):\n",
+		g->p.nr_tasks, g->p.nr_tasks == 1 ? "task" : "tasks", g->p.nr_nodes, g->p.nr_cpus);
+	printf(" #      %5dx %5ldMB global  shared mem operations\n",
+			g->p.nr_loops, g->p.bytes_global/1024/1024);
+	printf(" #      %5dx %5ldMB process shared mem operations\n",
+			g->p.nr_loops, g->p.bytes_process/1024/1024);
+	printf(" #      %5dx %5ldMB thread  local  mem operations\n",
+			g->p.nr_loops, g->p.bytes_thread/1024/1024);
+
+	printf(" ###\n");
+
+	printf("\n ###\n"); fflush(stdout);
+}
+
+static void init_thread_data(void)
+{
+	ssize_t size = sizeof(*g->threads)*g->p.nr_tasks;
+	int t;
+
+	g->threads = zalloc_shared_data(size);
+
+	for (t = 0; t < g->p.nr_tasks; t++) {
+		struct thread_data *td = g->threads + t;
+		int cpu;
+
+		/* Allow all nodes by default: */
+		td->bind_node = -1;
+
+		/* Allow all CPUs by default: */
+		CPU_ZERO(&td->bind_cpumask);
+		for (cpu = 0; cpu < g->p.nr_cpus; cpu++)
+			CPU_SET(cpu, &td->bind_cpumask);
+	}
+}
+
+static void deinit_thread_data(void)
+{
+	ssize_t size = sizeof(*g->threads)*g->p.nr_tasks;
+
+	free_data(g->threads, size);
+}
+
+static int init(void)
+{
+	g = (void *)alloc_data(sizeof(*g), MAP_SHARED, 1, 0, 0 /* THP */, 0);
+
+	/* Copy over options: */
+	g->p = p0;
+
+	g->p.nr_cpus = numa_num_configured_cpus();
+
+	g->p.nr_nodes = numa_max_node() + 1;
+
+	/* char array in count_process_nodes(): */
+	BUG_ON(g->p.nr_nodes > MAX_NR_NODES || g->p.nr_nodes < 0);
+
+	if (g->p.show_quiet && !g->p.show_details)
+		g->p.show_details = -1;
+
+	/* Some memory should be specified: */
+	if (!g->p.mb_global_str && !g->p.mb_proc_str && !g->p.mb_thread_str)
+		return -1;
+
+	if (g->p.mb_global_str) {
+		g->p.mb_global = atof(g->p.mb_global_str);
+		BUG_ON(g->p.mb_global < 0);
+	}
+
+	if (g->p.mb_proc_str) {
+		g->p.mb_proc = atof(g->p.mb_proc_str);
+		BUG_ON(g->p.mb_proc < 0);
+	}
+
+	if (g->p.mb_proc_locked_str) {
+		g->p.mb_proc_locked = atof(g->p.mb_proc_locked_str);
+		BUG_ON(g->p.mb_proc_locked < 0);
+		BUG_ON(g->p.mb_proc_locked > g->p.mb_proc);
+	}
+
+	if (g->p.mb_thread_str) {
+		g->p.mb_thread = atof(g->p.mb_thread_str);
+		BUG_ON(g->p.mb_thread < 0);
+	}
+
+	BUG_ON(g->p.nr_threads <= 0);
+	BUG_ON(g->p.nr_proc <= 0);
+
+	g->p.nr_tasks = g->p.nr_proc*g->p.nr_threads;
+
+	g->p.bytes_global		= g->p.mb_global	*1024L*1024L;
+	g->p.bytes_process		= g->p.mb_proc		*1024L*1024L;
+	g->p.bytes_process_locked	= g->p.mb_proc_locked	*1024L*1024L;
+	g->p.bytes_thread		= g->p.mb_thread	*1024L*1024L;
+
+	g->data = setup_shared_data(g->p.bytes_global);
+
+	/* Startup serialization: */
+	init_global_mutex(&g->start_work_mutex);
+	init_global_mutex(&g->startup_mutex);
+	init_global_mutex(&g->startup_done_mutex);
+	init_global_mutex(&g->stop_work_mutex);
+
+	init_thread_data();
+
+	tprintf("#\n");
+	if (parse_setup_cpu_list() || parse_setup_node_list())
+		return -1;
+	tprintf("#\n");
+
+	print_summary();
+
+	return 0;
+}
+
+static void deinit(void)
+{
+	free_data(g->data, g->p.bytes_global);
+	g->data = NULL;
+
+	deinit_thread_data();
+
+	free_data(g, sizeof(*g));
+	g = NULL;
+}
+
+/*
+ * Print a short or long result, depending on the verbosity setting:
+ */
+static void print_res(const char *name, double val,
+		      const char *txt_unit, const char *txt_short, const char *txt_long)
+{
+	if (!name)
+		name = "main,";
+
+	if (!g->p.show_quiet)
+		printf(" %-30s %15.3f, %-15s %s\n", name, val, txt_unit, txt_short);
+	else
+		printf(" %14.3f %s\n", val, txt_long);
+}
+
+static int __bench_numa(const char *name)
+{
+	struct timeval start, stop, diff;
+	u64 runtime_ns_min, runtime_ns_sum;
+	pid_t *pids, pid, wpid;
+	double delta_runtime;
+	double runtime_avg;
+	double runtime_sec_max;
+	double runtime_sec_min;
+	int wait_stat;
+	double bytes;
+	int i, t;
+
+	if (init())
+		return -1;
+
+	pids = zalloc(g->p.nr_proc * sizeof(*pids));
+	pid = -1;
+
+	/* All threads try to acquire it, this way we can wait for them to start up: */
+	pthread_mutex_lock(&g->start_work_mutex);
+
+	if (g->p.serialize_startup) {
+		tprintf(" #\n");
+		tprintf(" # Startup synchronization: ..."); fflush(stdout);
+	}
+
+	gettimeofday(&start, NULL);
+
+	for (i = 0; i < g->p.nr_proc; i++) {
+		pid = fork();
+		dprintf(" # process %2d: PID %d\n", i, pid);
+
+		BUG_ON(pid < 0);
+		if (!pid) {
+			/* Child process: */
+			worker_process(i);
+
+			exit(0);
+		}
+		pids[i] = pid;
+
+	}
+	/* Wait for all the threads to start up: */
+	while (g->nr_tasks_started != g->p.nr_tasks)
+		usleep(1000);
+
+	BUG_ON(g->nr_tasks_started != g->p.nr_tasks);
+
+	if (g->p.serialize_startup) {
+		double startup_sec;
+
+		pthread_mutex_lock(&g->startup_done_mutex);
+
+		/* This will start all threads: */
+		pthread_mutex_unlock(&g->start_work_mutex);
+
+		/* This mutex is locked - the last started thread will wake us: */
+		pthread_mutex_lock(&g->startup_done_mutex);
+
+		gettimeofday(&stop, NULL);
+
+		timersub(&stop, &start, &diff);
+
+		startup_sec = diff.tv_sec * 1000000000.0;
+		startup_sec += diff.tv_usec * 1000.0;
+		startup_sec /= 1e9;
+
+		tprintf(" threads initialized in %.6f seconds.\n", startup_sec);
+		tprintf(" #\n");
+
+		start = stop;
+		pthread_mutex_unlock(&g->startup_done_mutex);
+	} else {
+		gettimeofday(&start, NULL);
+	}
+
+	/* Parent process: */
+
+
+	for (i = 0; i < g->p.nr_proc; i++) {
+		wpid = waitpid(pids[i], &wait_stat, 0);
+		BUG_ON(wpid < 0);
+		BUG_ON(!WIFEXITED(wait_stat));
+
+	}
+
+	runtime_ns_sum = 0;
+	runtime_ns_min = -1LL;
+
+	for (t = 0; t < g->p.nr_tasks; t++) {
+		u64 thread_runtime_ns = g->threads[t].runtime_ns;
+
+		runtime_ns_sum += thread_runtime_ns;
+		runtime_ns_min = min(thread_runtime_ns, runtime_ns_min);
+	}
+
+	gettimeofday(&stop, NULL);
+	timersub(&stop, &start, &diff);
+
+	BUG_ON(bench_format != BENCH_FORMAT_DEFAULT);
+
+	tprintf("\n ###\n");
+	tprintf("\n");
+
+	runtime_sec_max = diff.tv_sec * 1000000000.0;
+	runtime_sec_max += diff.tv_usec * 1000.0;
+	runtime_sec_max /= 1e9;
+
+	runtime_sec_min = runtime_ns_min/1e9;
+
+	bytes = g->bytes_done;
+	runtime_avg = (double)runtime_ns_sum / g->p.nr_tasks / 1e9;
+
+	if (g->p.measure_convergence) {
+		print_res(name, runtime_sec_max,
+			"secs,", "NUMA-convergence-latency", "secs latency to NUMA-converge");
+	}
+
+	print_res(name, runtime_sec_max,
+		"secs,", "runtime-max/thread",	"secs slowest (max) thread-runtime");
+
+	print_res(name, runtime_sec_min,
+		"secs,", "runtime-min/thread",	"secs fastest (min) thread-runtime");
+
+	print_res(name, runtime_avg,
+		"secs,", "runtime-avg/thread",	"secs average thread-runtime");
+
+	delta_runtime = (runtime_sec_max - runtime_sec_min)/2.0;
+	print_res(name, delta_runtime / runtime_sec_max * 100.0,
+		"%,", "spread-runtime/thread",	"% difference between max/avg runtime");
+
+	print_res(name, bytes / g->p.nr_tasks / 1e9,
+		"GB,", "data/thread",		"GB data processed, per thread");
+
+	print_res(name, bytes / 1e9,
+		"GB,", "data-total",		"GB data processed, total");
+
+	print_res(name, runtime_sec_max * 1e9 / (bytes / g->p.nr_tasks),
+		"nsecs,", "runtime/byte/thread","nsecs/byte/thread runtime");
+
+	print_res(name, bytes / g->p.nr_tasks / 1e9 / runtime_sec_max,
+		"GB/sec,", "thread-speed",	"GB/sec/thread speed");
+
+	print_res(name, bytes / runtime_sec_max / 1e9,
+		"GB/sec,", "total-speed",	"GB/sec total speed");
+
+	free(pids);
+
+	deinit();
+
+	return 0;
+}
+
+#define MAX_ARGS 50
+
+static int command_size(const char **argv)
+{
+	int size = 0;
+
+	while (*argv) {
+		size++;
+		argv++;
+	}
+
+	BUG_ON(size >= MAX_ARGS);
+
+	return size;
+}
+
+static void init_params(struct params *p, const char *name, int argc, const char **argv)
+{
+	int i;
+
+	printf("\n # Running %s \"perf bench numa", name);
+
+	for (i = 0; i < argc; i++)
+		printf(" %s", argv[i]);
+
+	printf("\"\n");
+
+	memset(p, 0, sizeof(*p));
+
+	/* Initialize nonzero defaults: */
+
+	p->serialize_startup		= 1;
+	p->data_reads			= true;
+	p->data_writes			= true;
+	p->data_backwards		= true;
+	p->data_rand_walk		= true;
+	p->nr_loops			= -1;
+	p->init_random			= true;
+	p->mb_global_str		= "1";
+	p->nr_proc			= 1;
+	p->nr_threads			= 1;
+	p->nr_secs			= 5;
+	p->run_all			= argc == 1;
+}
+
+static int run_bench_numa(const char *name, const char **argv)
+{
+	int argc = command_size(argv);
+
+	init_params(&p0, name, argc, argv);
+	argc = parse_options(argc, argv, options, bench_numa_usage, 0);
+	if (argc)
+		goto err;
+
+	if (__bench_numa(name))
+		goto err;
+
+	return 0;
+
+err:
+	return -1;
+}
+
+#define OPT_BW_RAM		"-s",  "20", "-zZq",    "--thp", " 1", "--no-data_rand_walk"
+#define OPT_BW_RAM_NOTHP	OPT_BW_RAM,		"--thp", "-1"
+
+#define OPT_CONV		"-s", "100", "-zZ0qcm", "--thp", " 1"
+#define OPT_CONV_NOTHP		OPT_CONV,		"--thp", "-1"
+
+#define OPT_BW			"-s",  "20", "-zZ0q",   "--thp", " 1"
+#define OPT_BW_NOTHP		OPT_BW,			"--thp", "-1"
+
+/*
+ * The built-in test-suite executed by "perf bench numa -a".
+ *
+ * (A minimum of 4 nodes and 16 GB of RAM is recommended.)
+ */
+static const char *tests[][MAX_ARGS] = {
+   /* Basic single-stream NUMA bandwidth measurements: */
+   { "RAM-bw-local,",	  "mem",  "-p",  "1",  "-t",  "1", "-P", "1024",
+			  "-C" ,   "0", "-M",   "0", OPT_BW_RAM },
+   { "RAM-bw-local-NOTHP,",
+			  "mem",  "-p",  "1",  "-t",  "1", "-P", "1024",
+			  "-C" ,   "0", "-M",   "0", OPT_BW_RAM_NOTHP },
+   { "RAM-bw-remote,",	  "mem",  "-p",  "1",  "-t",  "1", "-P", "1024",
+			  "-C" ,   "0", "-M",   "1", OPT_BW_RAM },
+
+   /* 2-stream NUMA bandwidth measurements: */
+   { "RAM-bw-local-2x,",  "mem",  "-p",  "2",  "-t",  "1", "-P", "1024",
+			   "-C", "0,2", "-M", "0x2", OPT_BW_RAM },
+   { "RAM-bw-remote-2x,", "mem",  "-p",  "2",  "-t",  "1", "-P", "1024",
+		 	   "-C", "0,2", "-M", "1x2", OPT_BW_RAM },
+
+   /* Cross-stream NUMA bandwidth measurement: */
+   { "RAM-bw-cross,",     "mem",  "-p",  "2",  "-t",  "1", "-P", "1024",
+		 	   "-C", "0,8", "-M", "1,0", OPT_BW_RAM },
+
+   /* Convergence latency measurements: */
+   { " 1x3-convergence,", "mem",  "-p",  "1", "-t",  "3", "-P",  "512", OPT_CONV },
+   { " 1x4-convergence,", "mem",  "-p",  "1", "-t",  "4", "-P",  "512", OPT_CONV },
+   { " 1x6-convergence,", "mem",  "-p",  "1", "-t",  "6", "-P", "1020", OPT_CONV },
+   { " 2x3-convergence,", "mem",  "-p",  "3", "-t",  "3", "-P", "1020", OPT_CONV },
+   { " 3x3-convergence,", "mem",  "-p",  "3", "-t",  "3", "-P", "1020", OPT_CONV },
+   { " 4x4-convergence,", "mem",  "-p",  "4", "-t",  "4", "-P",  "512", OPT_CONV },
+   { " 4x4-convergence-NOTHP,",
+			  "mem",  "-p",  "4", "-t",  "4", "-P",  "512", OPT_CONV_NOTHP },
+   { " 4x6-convergence,", "mem",  "-p",  "4", "-t",  "6", "-P", "1020", OPT_CONV },
+   { " 4x8-convergence,", "mem",  "-p",  "4", "-t",  "8", "-P",  "512", OPT_CONV },
+   { " 8x4-convergence,", "mem",  "-p",  "8", "-t",  "4", "-P",  "512", OPT_CONV },
+   { " 8x4-convergence-NOTHP,",
+			  "mem",  "-p",  "8", "-t",  "4", "-P",  "512", OPT_CONV_NOTHP },
+   { " 3x1-convergence,", "mem",  "-p",  "3", "-t",  "1", "-P",  "512", OPT_CONV },
+   { " 4x1-convergence,", "mem",  "-p",  "4", "-t",  "1", "-P",  "512", OPT_CONV },
+   { " 8x1-convergence,", "mem",  "-p",  "8", "-t",  "1", "-P",  "512", OPT_CONV },
+   { "16x1-convergence,", "mem",  "-p", "16", "-t",  "1", "-P",  "256", OPT_CONV },
+   { "32x1-convergence,", "mem",  "-p", "32", "-t",  "1", "-P",  "128", OPT_CONV },
+
+   /* Various NUMA process/thread layout bandwidth measurements: */
+   { " 2x1-bw-process,",  "mem",  "-p",  "2", "-t",  "1", "-P", "1024", OPT_BW },
+   { " 3x1-bw-process,",  "mem",  "-p",  "3", "-t",  "1", "-P", "1024", OPT_BW },
+   { " 4x1-bw-process,",  "mem",  "-p",  "4", "-t",  "1", "-P", "1024", OPT_BW },
+   { " 8x1-bw-process,",  "mem",  "-p",  "8", "-t",  "1", "-P", " 512", OPT_BW },
+   { " 8x1-bw-process-NOTHP,",
+			  "mem",  "-p",  "8", "-t",  "1", "-P", " 512", OPT_BW_NOTHP },
+   { "16x1-bw-process,",  "mem",  "-p", "16", "-t",  "1", "-P",  "256", OPT_BW },
+
+   { " 4x1-bw-thread,",	  "mem",  "-p",  "1", "-t",  "4", "-T",  "256", OPT_BW },
+   { " 8x1-bw-thread,",	  "mem",  "-p",  "1", "-t",  "8", "-T",  "256", OPT_BW },
+   { "16x1-bw-thread,",   "mem",  "-p",  "1", "-t", "16", "-T",  "128", OPT_BW },
+   { "32x1-bw-thread,",   "mem",  "-p",  "1", "-t", "32", "-T",   "64", OPT_BW },
+
+   { " 2x3-bw-thread,",	  "mem",  "-p",  "2", "-t",  "3", "-P",  "512", OPT_BW },
+   { " 4x4-bw-thread,",	  "mem",  "-p",  "4", "-t",  "4", "-P",  "512", OPT_BW },
+   { " 4x6-bw-thread,",	  "mem",  "-p",  "4", "-t",  "6", "-P",  "512", OPT_BW },
+   { " 4x8-bw-thread,",	  "mem",  "-p",  "4", "-t",  "8", "-P",  "512", OPT_BW },
+   { " 4x8-bw-thread-NOTHP,",
+			  "mem",  "-p",  "4", "-t",  "8", "-P",  "512", OPT_BW_NOTHP },
+   { " 3x3-bw-thread,",	  "mem",  "-p",  "3", "-t",  "3", "-P",  "512", OPT_BW },
+   { " 5x5-bw-thread,",	  "mem",  "-p",  "5", "-t",  "5", "-P",  "512", OPT_BW },
+
+   { "2x16-bw-thread,",   "mem",  "-p",  "2", "-t", "16", "-P",  "512", OPT_BW },
+   { "1x32-bw-thread,",   "mem",  "-p",  "1", "-t", "32", "-P", "2048", OPT_BW },
+
+   { "numa02-bw,",	  "mem",  "-p",  "1", "-t", "32", "-T",   "32", OPT_BW },
+   { "numa02-bw-NOTHP,",  "mem",  "-p",  "1", "-t", "32", "-T",   "32", OPT_BW_NOTHP },
+   { "numa01-bw-thread,", "mem",  "-p",  "2", "-t", "16", "-T",  "192", OPT_BW },
+   { "numa01-bw-thread-NOTHP,",
+			  "mem",  "-p",  "2", "-t", "16", "-T",  "192", OPT_BW_NOTHP },
+};
+
+static int bench_all(void)
+{
+	int nr = ARRAY_SIZE(tests);
+	int ret;
+	int i;
+
+	ret = system("echo ' #'; echo ' # Running test on: '$(uname -a); echo ' #'");
+	BUG_ON(ret < 0);
+
+	for (i = 0; i < nr; i++) {
+		run_bench_numa(tests[i][0], tests[i] + 1);
+	}
+
+	printf("\n");
+
+	return 0;
+}
+
+int bench_numa(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	init_params(&p0, "main,", argc, argv);
+	argc = parse_options(argc, argv, options, bench_numa_usage, 0);
+	if (argc)
+		goto err;
+
+	if (p0.run_all)
+		return bench_all();
+
+	if (__bench_numa(NULL))
+		goto err;
+
+	return 0;
+
+err:
+	usage_with_options(numa_usage, options);
+	return -1;
+}
diff --git a/tools/perf/bench/sched-messaging.c b/tools/perf/bench/sched-messaging.c
new file mode 100644
index 000000000..d7f281c28
--- /dev/null
+++ b/tools/perf/bench/sched-messaging.c
@@ -0,0 +1,331 @@
+/*
+ *
+ * sched-messaging.c
+ *
+ * messaging: Benchmark for scheduler and IPC mechanisms
+ *
+ * Based on hackbench by Rusty Russell <rusty@rustcorp.com.au>
+ * Ported to perf by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
+ *
+ */
+
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/parse-options.h"
+#include "../builtin.h"
+#include "bench.h"
+
+/* Test groups of 20 processes spraying to 20 receivers */
+#include <pthread.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <sys/wait.h>
+#include <sys/time.h>
+#include <poll.h>
+#include <limits.h>
+#include <err.h>
+
+#define DATASIZE 100
+
+static bool use_pipes = false;
+static unsigned int loops = 100;
+static bool thread_mode = false;
+static unsigned int num_groups = 10;
+
+struct sender_context {
+	unsigned int num_fds;
+	int ready_out;
+	int wakefd;
+	int out_fds[0];
+};
+
+struct receiver_context {
+	unsigned int num_packets;
+	int in_fds[2];
+	int ready_out;
+	int wakefd;
+};
+
+static void fdpair(int fds[2])
+{
+	if (use_pipes) {
+		if (pipe(fds) == 0)
+			return;
+	} else {
+		if (socketpair(AF_UNIX, SOCK_STREAM, 0, fds) == 0)
+			return;
+	}
+
+	err(EXIT_FAILURE, use_pipes ? "pipe()" : "socketpair()");
+}
+
+/* Block until we're ready to go */
+static void ready(int ready_out, int wakefd)
+{
+	char dummy;
+	struct pollfd pollfd = { .fd = wakefd, .events = POLLIN };
+
+	/* Tell them we're ready. */
+	if (write(ready_out, &dummy, 1) != 1)
+		err(EXIT_FAILURE, "CLIENT: ready write");
+
+	/* Wait for "GO" signal */
+	if (poll(&pollfd, 1, -1) != 1)
+		err(EXIT_FAILURE, "poll");
+}
+
+/* Sender sprays loops messages down each file descriptor */
+static void *sender(struct sender_context *ctx)
+{
+	char data[DATASIZE];
+	unsigned int i, j;
+
+	ready(ctx->ready_out, ctx->wakefd);
+
+	/* Now pump to every receiver. */
+	for (i = 0; i < loops; i++) {
+		for (j = 0; j < ctx->num_fds; j++) {
+			int ret, done = 0;
+
+again:
+			ret = write(ctx->out_fds[j], data + done,
+				    sizeof(data)-done);
+			if (ret < 0)
+				err(EXIT_FAILURE, "SENDER: write");
+			done += ret;
+			if (done < DATASIZE)
+				goto again;
+		}
+	}
+
+	return NULL;
+}
+
+
+/* One receiver per fd */
+static void *receiver(struct receiver_context* ctx)
+{
+	unsigned int i;
+
+	if (!thread_mode)
+		close(ctx->in_fds[1]);
+
+	/* Wait for start... */
+	ready(ctx->ready_out, ctx->wakefd);
+
+	/* Receive them all */
+	for (i = 0; i < ctx->num_packets; i++) {
+		char data[DATASIZE];
+		int ret, done = 0;
+
+again:
+		ret = read(ctx->in_fds[0], data + done, DATASIZE - done);
+		if (ret < 0)
+			err(EXIT_FAILURE, "SERVER: read");
+		done += ret;
+		if (done < DATASIZE)
+			goto again;
+	}
+
+	return NULL;
+}
+
+static pthread_t create_worker(void *ctx, void *(*func)(void *))
+{
+	pthread_attr_t attr;
+	pthread_t childid;
+	int ret;
+
+	if (!thread_mode) {
+		/* process mode */
+		/* Fork the receiver. */
+		switch (fork()) {
+		case -1:
+			err(EXIT_FAILURE, "fork()");
+			break;
+		case 0:
+			(*func) (ctx);
+			exit(0);
+			break;
+		default:
+			break;
+		}
+
+		return (pthread_t)0;
+	}
+
+	if (pthread_attr_init(&attr) != 0)
+		err(EXIT_FAILURE, "pthread_attr_init:");
+
+#ifndef __ia64__
+	if (pthread_attr_setstacksize(&attr, PTHREAD_STACK_MIN) != 0)
+		err(EXIT_FAILURE, "pthread_attr_setstacksize");
+#endif
+
+	ret = pthread_create(&childid, &attr, func, ctx);
+	if (ret != 0)
+		err(EXIT_FAILURE, "pthread_create failed");
+
+	return childid;
+}
+
+static void reap_worker(pthread_t id)
+{
+	int proc_status;
+	void *thread_status;
+
+	if (!thread_mode) {
+		/* process mode */
+		wait(&proc_status);
+		if (!WIFEXITED(proc_status))
+			exit(1);
+	} else {
+		pthread_join(id, &thread_status);
+	}
+}
+
+/* One group of senders and receivers */
+static unsigned int group(pthread_t *pth,
+		unsigned int num_fds,
+		int ready_out,
+		int wakefd)
+{
+	unsigned int i;
+	struct sender_context *snd_ctx = malloc(sizeof(struct sender_context)
+			+ num_fds * sizeof(int));
+
+	if (!snd_ctx)
+		err(EXIT_FAILURE, "malloc()");
+
+	for (i = 0; i < num_fds; i++) {
+		int fds[2];
+		struct receiver_context *ctx = malloc(sizeof(*ctx));
+
+		if (!ctx)
+			err(EXIT_FAILURE, "malloc()");
+
+
+		/* Create the pipe between client and server */
+		fdpair(fds);
+
+		ctx->num_packets = num_fds * loops;
+		ctx->in_fds[0] = fds[0];
+		ctx->in_fds[1] = fds[1];
+		ctx->ready_out = ready_out;
+		ctx->wakefd = wakefd;
+
+		pth[i] = create_worker(ctx, (void *)receiver);
+
+		snd_ctx->out_fds[i] = fds[1];
+		if (!thread_mode)
+			close(fds[0]);
+	}
+
+	/* Now we have all the fds, fork the senders */
+	for (i = 0; i < num_fds; i++) {
+		snd_ctx->ready_out = ready_out;
+		snd_ctx->wakefd = wakefd;
+		snd_ctx->num_fds = num_fds;
+
+		pth[num_fds+i] = create_worker(snd_ctx, (void *)sender);
+	}
+
+	/* Close the fds we have left */
+	if (!thread_mode)
+		for (i = 0; i < num_fds; i++)
+			close(snd_ctx->out_fds[i]);
+
+	/* Return number of children to reap */
+	return num_fds * 2;
+}
+
+static const struct option options[] = {
+	OPT_BOOLEAN('p', "pipe", &use_pipes,
+		    "Use pipe() instead of socketpair()"),
+	OPT_BOOLEAN('t', "thread", &thread_mode,
+		    "Be multi thread instead of multi process"),
+	OPT_UINTEGER('g', "group", &num_groups, "Specify number of groups"),
+	OPT_UINTEGER('l', "loop", &loops, "Specify number of loops"),
+	OPT_END()
+};
+
+static const char * const bench_sched_message_usage[] = {
+	"perf bench sched messaging <options>",
+	NULL
+};
+
+int bench_sched_messaging(int argc, const char **argv,
+		    const char *prefix __maybe_unused)
+{
+	unsigned int i, total_children;
+	struct timeval start, stop, diff;
+	unsigned int num_fds = 20;
+	int readyfds[2], wakefds[2];
+	char dummy;
+	pthread_t *pth_tab;
+
+	argc = parse_options(argc, argv, options,
+			     bench_sched_message_usage, 0);
+
+	pth_tab = malloc(num_fds * 2 * num_groups * sizeof(pthread_t));
+	if (!pth_tab)
+		err(EXIT_FAILURE, "main:malloc()");
+
+	fdpair(readyfds);
+	fdpair(wakefds);
+
+	total_children = 0;
+	for (i = 0; i < num_groups; i++)
+		total_children += group(pth_tab+total_children, num_fds,
+					readyfds[1], wakefds[0]);
+
+	/* Wait for everyone to be ready */
+	for (i = 0; i < total_children; i++)
+		if (read(readyfds[0], &dummy, 1) != 1)
+			err(EXIT_FAILURE, "Reading for readyfds");
+
+	gettimeofday(&start, NULL);
+
+	/* Kick them off */
+	if (write(wakefds[1], &dummy, 1) != 1)
+		err(EXIT_FAILURE, "Writing to start them");
+
+	/* Reap them all */
+	for (i = 0; i < total_children; i++)
+		reap_worker(pth_tab[i]);
+
+	gettimeofday(&stop, NULL);
+
+	timersub(&stop, &start, &diff);
+
+	switch (bench_format) {
+	case BENCH_FORMAT_DEFAULT:
+		printf("# %d sender and receiver %s per group\n",
+		       num_fds, thread_mode ? "threads" : "processes");
+		printf("# %d groups == %d %s run\n\n",
+		       num_groups, num_groups * 2 * num_fds,
+		       thread_mode ? "threads" : "processes");
+		printf(" %14s: %lu.%03lu [sec]\n", "Total time",
+		       diff.tv_sec,
+		       (unsigned long) (diff.tv_usec/1000));
+		break;
+	case BENCH_FORMAT_SIMPLE:
+		printf("%lu.%03lu\n", diff.tv_sec,
+		       (unsigned long) (diff.tv_usec/1000));
+		break;
+	default:
+		/* reaching here is something disaster */
+		fprintf(stderr, "Unknown format:%d\n", bench_format);
+		exit(1);
+		break;
+	}
+
+	free(pth_tab);
+
+	return 0;
+}
diff --git a/tools/perf/bench/sched-pipe.c b/tools/perf/bench/sched-pipe.c
new file mode 100644
index 000000000..005cc2837
--- /dev/null
+++ b/tools/perf/bench/sched-pipe.c
@@ -0,0 +1,184 @@
+/*
+ *
+ * sched-pipe.c
+ *
+ * pipe: Benchmark for pipe()
+ *
+ * Based on pipe-test-1m.c by Ingo Molnar <mingo@redhat.com>
+ *  http://people.redhat.com/mingo/cfs-scheduler/tools/pipe-test-1m.c
+ * Ported to perf by Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
+ */
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/parse-options.h"
+#include "../builtin.h"
+#include "bench.h"
+
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <signal.h>
+#include <sys/wait.h>
+#include <string.h>
+#include <errno.h>
+#include <assert.h>
+#include <sys/time.h>
+#include <sys/types.h>
+#include <sys/syscall.h>
+
+#include <pthread.h>
+
+struct thread_data {
+	int			nr;
+	int			pipe_read;
+	int			pipe_write;
+	pthread_t		pthread;
+};
+
+#define LOOPS_DEFAULT 1000000
+static	int			loops = LOOPS_DEFAULT;
+
+/* Use processes by default: */
+static bool			threaded;
+
+static const struct option options[] = {
+	OPT_INTEGER('l', "loop",	&loops,		"Specify number of loops"),
+	OPT_BOOLEAN('T', "threaded",	&threaded,	"Specify threads/process based task setup"),
+	OPT_END()
+};
+
+static const char * const bench_sched_pipe_usage[] = {
+	"perf bench sched pipe <options>",
+	NULL
+};
+
+static void *worker_thread(void *__tdata)
+{
+	struct thread_data *td = __tdata;
+	int m = 0, i;
+	int ret;
+
+	for (i = 0; i < loops; i++) {
+		if (!td->nr) {
+			ret = read(td->pipe_read, &m, sizeof(int));
+			BUG_ON(ret != sizeof(int));
+			ret = write(td->pipe_write, &m, sizeof(int));
+			BUG_ON(ret != sizeof(int));
+		} else {
+			ret = write(td->pipe_write, &m, sizeof(int));
+			BUG_ON(ret != sizeof(int));
+			ret = read(td->pipe_read, &m, sizeof(int));
+			BUG_ON(ret != sizeof(int));
+		}
+	}
+
+	return NULL;
+}
+
+int bench_sched_pipe(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	struct thread_data threads[2], *td;
+	int pipe_1[2], pipe_2[2];
+	struct timeval start, stop, diff;
+	unsigned long long result_usec = 0;
+	int nr_threads = 2;
+	int t;
+
+	/*
+	 * why does "ret" exist?
+	 * discarding returned value of read(), write()
+	 * causes error in building environment for perf
+	 */
+	int __maybe_unused ret, wait_stat;
+	pid_t pid, retpid __maybe_unused;
+
+	argc = parse_options(argc, argv, options, bench_sched_pipe_usage, 0);
+
+	BUG_ON(pipe(pipe_1));
+	BUG_ON(pipe(pipe_2));
+
+	gettimeofday(&start, NULL);
+
+	for (t = 0; t < nr_threads; t++) {
+		td = threads + t;
+
+		td->nr = t;
+
+		if (t == 0) {
+			td->pipe_read = pipe_1[0];
+			td->pipe_write = pipe_2[1];
+		} else {
+			td->pipe_write = pipe_1[1];
+			td->pipe_read = pipe_2[0];
+		}
+	}
+
+
+	if (threaded) {
+
+		for (t = 0; t < nr_threads; t++) {
+			td = threads + t;
+
+			ret = pthread_create(&td->pthread, NULL, worker_thread, td);
+			BUG_ON(ret);
+		}
+
+		for (t = 0; t < nr_threads; t++) {
+			td = threads + t;
+
+			ret = pthread_join(td->pthread, NULL);
+			BUG_ON(ret);
+		}
+
+	} else {
+		pid = fork();
+		assert(pid >= 0);
+
+		if (!pid) {
+			worker_thread(threads + 0);
+			exit(0);
+		} else {
+			worker_thread(threads + 1);
+		}
+
+		retpid = waitpid(pid, &wait_stat, 0);
+		assert((retpid == pid) && WIFEXITED(wait_stat));
+	}
+
+	gettimeofday(&stop, NULL);
+	timersub(&stop, &start, &diff);
+
+	switch (bench_format) {
+	case BENCH_FORMAT_DEFAULT:
+		printf("# Executed %d pipe operations between two %s\n\n",
+			loops, threaded ? "threads" : "processes");
+
+		result_usec = diff.tv_sec * 1000000;
+		result_usec += diff.tv_usec;
+
+		printf(" %14s: %lu.%03lu [sec]\n\n", "Total time",
+		       diff.tv_sec,
+		       (unsigned long) (diff.tv_usec/1000));
+
+		printf(" %14lf usecs/op\n",
+		       (double)result_usec / (double)loops);
+		printf(" %14d ops/sec\n",
+		       (int)((double)loops /
+			     ((double)result_usec / (double)1000000)));
+		break;
+
+	case BENCH_FORMAT_SIMPLE:
+		printf("%lu.%03lu\n",
+		       diff.tv_sec,
+		       (unsigned long) (diff.tv_usec / 1000));
+		break;
+
+	default:
+		/* reaching here is something disaster */
+		fprintf(stderr, "Unknown format:%d\n", bench_format);
+		exit(1);
+		break;
+	}
+
+	return 0;
+}
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
new file mode 100644
index 000000000..71bf7451c
--- /dev/null
+++ b/tools/perf/builtin-annotate.c
@@ -0,0 +1,386 @@
+/*
+ * builtin-annotate.c
+ *
+ * Builtin annotate command: Analyze the perf.data input file,
+ * look up and read DSOs and symbol information and display
+ * a histogram of results, along various sorting keys.
+ */
+#include "builtin.h"
+
+#include "util/util.h"
+#include "util/color.h"
+#include <linux/list.h>
+#include "util/cache.h"
+#include <linux/rbtree.h>
+#include "util/symbol.h"
+
+#include "perf.h"
+#include "util/debug.h"
+
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/annotate.h"
+#include "util/event.h"
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+#include "util/thread.h"
+#include "util/sort.h"
+#include "util/hist.h"
+#include "util/session.h"
+#include "util/tool.h"
+#include "util/data.h"
+#include "arch/common.h"
+
+#include <dlfcn.h>
+#include <linux/bitmap.h>
+
+struct perf_annotate {
+	struct perf_tool tool;
+	struct perf_session *session;
+	bool	   use_tui, use_stdio, use_gtk;
+	bool	   full_paths;
+	bool	   print_line;
+	bool	   skip_missing;
+	const char *sym_hist_filter;
+	const char *cpu_list;
+	DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
+};
+
+static int perf_evsel__add_sample(struct perf_evsel *evsel,
+				  struct perf_sample *sample __maybe_unused,
+				  struct addr_location *al,
+				  struct perf_annotate *ann)
+{
+	struct hists *hists = evsel__hists(evsel);
+	struct hist_entry *he;
+	int ret;
+
+	if (ann->sym_hist_filter != NULL &&
+	    (al->sym == NULL ||
+	     strcmp(ann->sym_hist_filter, al->sym->name) != 0)) {
+		/* We're only interested in a symbol named sym_hist_filter */
+		if (al->sym != NULL) {
+			rb_erase(&al->sym->rb_node,
+				 &al->map->dso->symbols[al->map->type]);
+			symbol__delete(al->sym);
+		}
+		return 0;
+	}
+
+	he = __hists__add_entry(hists, al, NULL, NULL, NULL, 1, 1, 0, true);
+	if (he == NULL)
+		return -ENOMEM;
+
+	ret = hist_entry__inc_addr_samples(he, evsel->idx, al->addr);
+	hists__inc_nr_samples(hists, true);
+	return ret;
+}
+
+static int process_sample_event(struct perf_tool *tool,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel,
+				struct machine *machine)
+{
+	struct perf_annotate *ann = container_of(tool, struct perf_annotate, tool);
+	struct addr_location al;
+
+	if (perf_event__preprocess_sample(event, machine, &al, sample) < 0) {
+		pr_warning("problem processing %d event, skipping it.\n",
+			   event->header.type);
+		return -1;
+	}
+
+	if (ann->cpu_list && !test_bit(sample->cpu, ann->cpu_bitmap))
+		return 0;
+
+	if (!al.filtered && perf_evsel__add_sample(evsel, sample, &al, ann)) {
+		pr_warning("problem incrementing symbol count, "
+			   "skipping event\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+static int hist_entry__tty_annotate(struct hist_entry *he,
+				    struct perf_evsel *evsel,
+				    struct perf_annotate *ann)
+{
+	return symbol__tty_annotate(he->ms.sym, he->ms.map, evsel,
+				    ann->print_line, ann->full_paths, 0, 0);
+}
+
+static void hists__find_annotations(struct hists *hists,
+				    struct perf_evsel *evsel,
+				    struct perf_annotate *ann)
+{
+	struct rb_node *nd = rb_first(&hists->entries), *next;
+	int key = K_RIGHT;
+
+	while (nd) {
+		struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node);
+		struct annotation *notes;
+
+		if (he->ms.sym == NULL || he->ms.map->dso->annotate_warned)
+			goto find_next;
+
+		notes = symbol__annotation(he->ms.sym);
+		if (notes->src == NULL) {
+find_next:
+			if (key == K_LEFT)
+				nd = rb_prev(nd);
+			else
+				nd = rb_next(nd);
+			continue;
+		}
+
+		if (use_browser == 2) {
+			int ret;
+			int (*annotate)(struct hist_entry *he,
+					struct perf_evsel *evsel,
+					struct hist_browser_timer *hbt);
+
+			annotate = dlsym(perf_gtk_handle,
+					 "hist_entry__gtk_annotate");
+			if (annotate == NULL) {
+				ui__error("GTK browser not found!\n");
+				return;
+			}
+
+			ret = annotate(he, evsel, NULL);
+			if (!ret || !ann->skip_missing)
+				return;
+
+			/* skip missing symbols */
+			nd = rb_next(nd);
+		} else if (use_browser == 1) {
+			key = hist_entry__tui_annotate(he, evsel, NULL);
+			switch (key) {
+			case -1:
+				if (!ann->skip_missing)
+					return;
+				/* fall through */
+			case K_RIGHT:
+				next = rb_next(nd);
+				break;
+			case K_LEFT:
+				next = rb_prev(nd);
+				break;
+			default:
+				return;
+			}
+
+			if (next != NULL)
+				nd = next;
+		} else {
+			hist_entry__tty_annotate(he, evsel, ann);
+			nd = rb_next(nd);
+			/*
+			 * Since we have a hist_entry per IP for the same
+			 * symbol, free he->ms.sym->src to signal we already
+			 * processed this symbol.
+			 */
+			zfree(&notes->src);
+		}
+	}
+}
+
+static int __cmd_annotate(struct perf_annotate *ann)
+{
+	int ret;
+	struct perf_session *session = ann->session;
+	struct perf_evsel *pos;
+	u64 total_nr_samples;
+
+	machines__set_symbol_filter(&session->machines, symbol__annotate_init);
+
+	if (ann->cpu_list) {
+		ret = perf_session__cpu_bitmap(session, ann->cpu_list,
+					       ann->cpu_bitmap);
+		if (ret)
+			goto out;
+	}
+
+	if (!objdump_path) {
+		ret = perf_session_env__lookup_objdump(&session->header.env);
+		if (ret)
+			goto out;
+	}
+
+	ret = perf_session__process_events(session);
+	if (ret)
+		goto out;
+
+	if (dump_trace) {
+		perf_session__fprintf_nr_events(session, stdout);
+		perf_evlist__fprintf_nr_events(session->evlist, stdout);
+		goto out;
+	}
+
+	if (verbose > 3)
+		perf_session__fprintf(session, stdout);
+
+	if (verbose > 2)
+		perf_session__fprintf_dsos(session, stdout);
+
+	total_nr_samples = 0;
+	evlist__for_each(session->evlist, pos) {
+		struct hists *hists = evsel__hists(pos);
+		u32 nr_samples = hists->stats.nr_events[PERF_RECORD_SAMPLE];
+
+		if (nr_samples > 0) {
+			total_nr_samples += nr_samples;
+			hists__collapse_resort(hists, NULL);
+			hists__output_resort(hists, NULL);
+
+			if (symbol_conf.event_group &&
+			    !perf_evsel__is_group_leader(pos))
+				continue;
+
+			hists__find_annotations(hists, pos, ann);
+		}
+	}
+
+	if (total_nr_samples == 0) {
+		ui__error("The %s file has no samples!\n", session->file->path);
+		goto out;
+	}
+
+	if (use_browser == 2) {
+		void (*show_annotations)(void);
+
+		show_annotations = dlsym(perf_gtk_handle,
+					 "perf_gtk__show_annotations");
+		if (show_annotations == NULL) {
+			ui__error("GTK browser not found!\n");
+			goto out;
+		}
+		show_annotations();
+	}
+
+out:
+	return ret;
+}
+
+static const char * const annotate_usage[] = {
+	"perf annotate [<options>]",
+	NULL
+};
+
+int cmd_annotate(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	struct perf_annotate annotate = {
+		.tool = {
+			.sample	= process_sample_event,
+			.mmap	= perf_event__process_mmap,
+			.mmap2	= perf_event__process_mmap2,
+			.comm	= perf_event__process_comm,
+			.exit	= perf_event__process_exit,
+			.fork	= perf_event__process_fork,
+			.ordered_events = true,
+			.ordering_requires_timestamps = true,
+		},
+	};
+	struct perf_data_file file = {
+		.path  = input_name,
+		.mode  = PERF_DATA_MODE_READ,
+	};
+	const struct option options[] = {
+	OPT_STRING('i', "input", &input_name, "file",
+		    "input file name"),
+	OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]",
+		   "only consider symbols in these dsos"),
+	OPT_STRING('s', "symbol", &annotate.sym_hist_filter, "symbol",
+		    "symbol to annotate"),
+	OPT_BOOLEAN('f', "force", &file.force, "don't complain, do it"),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_BOOLEAN(0, "gtk", &annotate.use_gtk, "Use the GTK interface"),
+	OPT_BOOLEAN(0, "tui", &annotate.use_tui, "Use the TUI interface"),
+	OPT_BOOLEAN(0, "stdio", &annotate.use_stdio, "Use the stdio interface"),
+	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+		   "file", "vmlinux pathname"),
+	OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules,
+		    "load module symbols - WARNING: use only with -k and LIVE kernel"),
+	OPT_BOOLEAN('l', "print-line", &annotate.print_line,
+		    "print matching source lines (may be slow)"),
+	OPT_BOOLEAN('P', "full-paths", &annotate.full_paths,
+		    "Don't shorten the displayed pathnames"),
+	OPT_BOOLEAN(0, "skip-missing", &annotate.skip_missing,
+		    "Skip symbols that cannot be annotated"),
+	OPT_STRING('C', "cpu", &annotate.cpu_list, "cpu", "list of cpus to profile"),
+	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
+		   "Look for files with symbols relative to this directory"),
+	OPT_BOOLEAN(0, "source", &symbol_conf.annotate_src,
+		    "Interleave source code with assembly code (default)"),
+	OPT_BOOLEAN(0, "asm-raw", &symbol_conf.annotate_asm_raw,
+		    "Display raw encoding of assembly instructions (default)"),
+	OPT_STRING('M', "disassembler-style", &disassembler_style, "disassembler style",
+		   "Specify disassembler style (e.g. -M intel for intel syntax)"),
+	OPT_STRING(0, "objdump", &objdump_path, "path",
+		   "objdump binary to use for disassembly and annotations"),
+	OPT_BOOLEAN(0, "group", &symbol_conf.event_group,
+		    "Show event group information together"),
+	OPT_END()
+	};
+	int ret = hists__init();
+
+	if (ret < 0)
+		return ret;
+
+	argc = parse_options(argc, argv, options, annotate_usage, 0);
+
+	if (annotate.use_stdio)
+		use_browser = 0;
+	else if (annotate.use_tui)
+		use_browser = 1;
+	else if (annotate.use_gtk)
+		use_browser = 2;
+
+	setup_browser(true);
+
+	annotate.session = perf_session__new(&file, false, &annotate.tool);
+	if (annotate.session == NULL)
+		return -1;
+
+	symbol_conf.priv_size = sizeof(struct annotation);
+	symbol_conf.try_vmlinux_path = true;
+
+	ret = symbol__init(&annotate.session->header.env);
+	if (ret < 0)
+		goto out_delete;
+
+	if (setup_sorting() < 0)
+		usage_with_options(annotate_usage, options);
+
+	if (argc) {
+		/*
+		 * Special case: if there's an argument left then assume that
+		 * it's a symbol filter:
+		 */
+		if (argc > 1)
+			usage_with_options(annotate_usage, options);
+
+		annotate.sym_hist_filter = argv[0];
+	}
+
+	ret = __cmd_annotate(&annotate);
+
+out_delete:
+	/*
+	 * Speed up the exit process, for large files this can
+	 * take quite a while.
+	 *
+	 * XXX Enable this when using valgrind or if we ever
+	 * librarize this command.
+	 *
+	 * Also experiment with obstacks to see how much speed
+	 * up we'll get here.
+	 *
+	 * perf_session__delete(session);
+	 */
+	return ret;
+}
diff --git a/tools/perf/builtin-bench.c b/tools/perf/builtin-bench.c
new file mode 100644
index 000000000..b9a56fa83
--- /dev/null
+++ b/tools/perf/builtin-bench.c
@@ -0,0 +1,289 @@
+/*
+ * builtin-bench.c
+ *
+ * General benchmarking collections provided by perf
+ *
+ * Copyright (C) 2009, Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
+ */
+
+/*
+ * Available benchmark collection list:
+ *
+ *  sched ... scheduler and IPC performance
+ *  mem   ... memory access performance
+ *  numa  ... NUMA scheduling and MM performance
+ *  futex ... Futex performance
+ */
+#include "perf.h"
+#include "util/util.h"
+#include "util/parse-options.h"
+#include "builtin.h"
+#include "bench/bench.h"
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/prctl.h>
+
+typedef int (*bench_fn_t)(int argc, const char **argv, const char *prefix);
+
+struct bench {
+	const char	*name;
+	const char	*summary;
+	bench_fn_t	fn;
+};
+
+#ifdef HAVE_LIBNUMA_SUPPORT
+static struct bench numa_benchmarks[] = {
+	{ "mem",	"Benchmark for NUMA workloads",			bench_numa		},
+	{ "all",	"Test all NUMA benchmarks",			NULL			},
+	{ NULL,		NULL,						NULL			}
+};
+#endif
+
+static struct bench sched_benchmarks[] = {
+	{ "messaging",	"Benchmark for scheduling and IPC",		bench_sched_messaging	},
+	{ "pipe",	"Benchmark for pipe() between two processes",	bench_sched_pipe	},
+	{ "all",	"Test all scheduler benchmarks",		NULL			},
+	{ NULL,		NULL,						NULL			}
+};
+
+static struct bench mem_benchmarks[] = {
+	{ "memcpy",	"Benchmark for memcpy()",			bench_mem_memcpy	},
+	{ "memset",	"Benchmark for memset() tests",			bench_mem_memset	},
+	{ "all",	"Test all memory benchmarks",			NULL			},
+	{ NULL,		NULL,						NULL			}
+};
+
+static struct bench futex_benchmarks[] = {
+	{ "hash",	"Benchmark for futex hash table",               bench_futex_hash	},
+	{ "wake",	"Benchmark for futex wake calls",               bench_futex_wake	},
+	{ "requeue",	"Benchmark for futex requeue calls",            bench_futex_requeue	},
+	{ "all",	"Test all futex benchmarks",			NULL			},
+	{ NULL,		NULL,						NULL			}
+};
+
+struct collection {
+	const char	*name;
+	const char	*summary;
+	struct bench	*benchmarks;
+};
+
+static struct collection collections[] = {
+	{ "sched",	"Scheduler and IPC benchmarks",			sched_benchmarks	},
+	{ "mem",	"Memory access benchmarks",			mem_benchmarks		},
+#ifdef HAVE_LIBNUMA_SUPPORT
+	{ "numa",	"NUMA scheduling and MM benchmarks",		numa_benchmarks		},
+#endif
+	{"futex",       "Futex stressing benchmarks",                   futex_benchmarks        },
+	{ "all",	"All benchmarks",				NULL			},
+	{ NULL,		NULL,						NULL			}
+};
+
+/* Iterate over all benchmark collections: */
+#define for_each_collection(coll) \
+	for (coll = collections; coll->name; coll++)
+
+/* Iterate over all benchmarks within a collection: */
+#define for_each_bench(coll, bench) \
+	for (bench = coll->benchmarks; bench && bench->name; bench++)
+
+static void dump_benchmarks(struct collection *coll)
+{
+	struct bench *bench;
+
+	printf("\n        # List of available benchmarks for collection '%s':\n\n", coll->name);
+
+	for_each_bench(coll, bench)
+		printf("%14s: %s\n", bench->name, bench->summary);
+
+	printf("\n");
+}
+
+static const char *bench_format_str;
+
+/* Output/formatting style, exported to benchmark modules: */
+int bench_format = BENCH_FORMAT_DEFAULT;
+unsigned int bench_repeat = 10; /* default number of times to repeat the run */
+
+static const struct option bench_options[] = {
+	OPT_STRING('f', "format", &bench_format_str, "default", "Specify format style"),
+	OPT_UINTEGER('r', "repeat",  &bench_repeat,   "Specify amount of times to repeat the run"),
+	OPT_END()
+};
+
+static const char * const bench_usage[] = {
+	"perf bench [<common options>] <collection> <benchmark> [<options>]",
+	NULL
+};
+
+static void print_usage(void)
+{
+	struct collection *coll;
+	int i;
+
+	printf("Usage: \n");
+	for (i = 0; bench_usage[i]; i++)
+		printf("\t%s\n", bench_usage[i]);
+	printf("\n");
+
+	printf("        # List of all available benchmark collections:\n\n");
+
+	for_each_collection(coll)
+		printf("%14s: %s\n", coll->name, coll->summary);
+	printf("\n");
+}
+
+static int bench_str2int(const char *str)
+{
+	if (!str)
+		return BENCH_FORMAT_DEFAULT;
+
+	if (!strcmp(str, BENCH_FORMAT_DEFAULT_STR))
+		return BENCH_FORMAT_DEFAULT;
+	else if (!strcmp(str, BENCH_FORMAT_SIMPLE_STR))
+		return BENCH_FORMAT_SIMPLE;
+
+	return BENCH_FORMAT_UNKNOWN;
+}
+
+/*
+ * Run a specific benchmark but first rename the running task's ->comm[]
+ * to something meaningful:
+ */
+static int run_bench(const char *coll_name, const char *bench_name, bench_fn_t fn,
+		     int argc, const char **argv, const char *prefix)
+{
+	int size;
+	char *name;
+	int ret;
+
+	size = strlen(coll_name) + 1 + strlen(bench_name) + 1;
+
+	name = zalloc(size);
+	BUG_ON(!name);
+
+	scnprintf(name, size, "%s-%s", coll_name, bench_name);
+
+	prctl(PR_SET_NAME, name);
+	argv[0] = name;
+
+	ret = fn(argc, argv, prefix);
+
+	free(name);
+
+	return ret;
+}
+
+static void run_collection(struct collection *coll)
+{
+	struct bench *bench;
+	const char *argv[2];
+
+	argv[1] = NULL;
+	/*
+	 * TODO:
+	 *
+	 * Preparing preset parameters for
+	 * embedded, ordinary PC, HPC, etc...
+	 * would be helpful.
+	 */
+	for_each_bench(coll, bench) {
+		if (!bench->fn)
+			break;
+		printf("# Running %s/%s benchmark...\n", coll->name, bench->name);
+		fflush(stdout);
+
+		argv[1] = bench->name;
+		run_bench(coll->name, bench->name, bench->fn, 1, argv, NULL);
+		printf("\n");
+	}
+}
+
+static void run_all_collections(void)
+{
+	struct collection *coll;
+
+	for_each_collection(coll)
+		run_collection(coll);
+}
+
+int cmd_bench(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	struct collection *coll;
+	int ret = 0;
+
+	if (argc < 2) {
+		/* No collection specified. */
+		print_usage();
+		goto end;
+	}
+
+	argc = parse_options(argc, argv, bench_options, bench_usage,
+			     PARSE_OPT_STOP_AT_NON_OPTION);
+
+	bench_format = bench_str2int(bench_format_str);
+	if (bench_format == BENCH_FORMAT_UNKNOWN) {
+		printf("Unknown format descriptor: '%s'\n", bench_format_str);
+		goto end;
+	}
+
+	if (bench_repeat == 0) {
+		printf("Invalid repeat option: Must specify a positive value\n");
+		goto end;
+	}
+
+	if (argc < 1) {
+		print_usage();
+		goto end;
+	}
+
+	if (!strcmp(argv[0], "all")) {
+		run_all_collections();
+		goto end;
+	}
+
+	for_each_collection(coll) {
+		struct bench *bench;
+
+		if (strcmp(coll->name, argv[0]))
+			continue;
+
+		if (argc < 2) {
+			/* No bench specified. */
+			dump_benchmarks(coll);
+			goto end;
+		}
+
+		if (!strcmp(argv[1], "all")) {
+			run_collection(coll);
+			goto end;
+		}
+
+		for_each_bench(coll, bench) {
+			if (strcmp(bench->name, argv[1]))
+				continue;
+
+			if (bench_format == BENCH_FORMAT_DEFAULT)
+				printf("# Running '%s/%s' benchmark:\n", coll->name, bench->name);
+			fflush(stdout);
+			ret = run_bench(coll->name, bench->name, bench->fn, argc-1, argv+1, prefix);
+			goto end;
+		}
+
+		if (!strcmp(argv[1], "-h") || !strcmp(argv[1], "--help")) {
+			dump_benchmarks(coll);
+			goto end;
+		}
+
+		printf("Unknown benchmark: '%s' for collection '%s'\n", argv[1], argv[0]);
+		ret = 1;
+		goto end;
+	}
+
+	printf("Unknown collection: '%s'\n", argv[0]);
+	ret = 1;
+
+end:
+	return ret;
+}
diff --git a/tools/perf/builtin-buildid-cache.c b/tools/perf/builtin-buildid-cache.c
new file mode 100644
index 000000000..d47a0cdc7
--- /dev/null
+++ b/tools/perf/builtin-buildid-cache.c
@@ -0,0 +1,448 @@
+/*
+ * builtin-buildid-cache.c
+ *
+ * Builtin buildid-cache command: Manages build-id cache
+ *
+ * Copyright (C) 2010, Red Hat Inc.
+ * Copyright (C) 2010, Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include <sys/types.h>
+#include <sys/time.h>
+#include <time.h>
+#include <dirent.h>
+#include <unistd.h>
+#include "builtin.h"
+#include "perf.h"
+#include "util/cache.h"
+#include "util/debug.h"
+#include "util/header.h"
+#include "util/parse-options.h"
+#include "util/strlist.h"
+#include "util/build-id.h"
+#include "util/session.h"
+#include "util/symbol.h"
+
+static int build_id_cache__kcore_buildid(const char *proc_dir, char *sbuildid)
+{
+	char root_dir[PATH_MAX];
+	char notes[PATH_MAX];
+	u8 build_id[BUILD_ID_SIZE];
+	char *p;
+
+	strlcpy(root_dir, proc_dir, sizeof(root_dir));
+
+	p = strrchr(root_dir, '/');
+	if (!p)
+		return -1;
+	*p = '\0';
+
+	scnprintf(notes, sizeof(notes), "%s/sys/kernel/notes", root_dir);
+
+	if (sysfs__read_build_id(notes, build_id, sizeof(build_id)))
+		return -1;
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuildid);
+
+	return 0;
+}
+
+static int build_id_cache__kcore_dir(char *dir, size_t sz)
+{
+	struct timeval tv;
+	struct tm tm;
+	char dt[32];
+
+	if (gettimeofday(&tv, NULL) || !localtime_r(&tv.tv_sec, &tm))
+		return -1;
+
+	if (!strftime(dt, sizeof(dt), "%Y%m%d%H%M%S", &tm))
+		return -1;
+
+	scnprintf(dir, sz, "%s%02u", dt, (unsigned)tv.tv_usec / 10000);
+
+	return 0;
+}
+
+static bool same_kallsyms_reloc(const char *from_dir, char *to_dir)
+{
+	char from[PATH_MAX];
+	char to[PATH_MAX];
+	const char *name;
+	u64 addr1 = 0, addr2 = 0;
+	int i;
+
+	scnprintf(from, sizeof(from), "%s/kallsyms", from_dir);
+	scnprintf(to, sizeof(to), "%s/kallsyms", to_dir);
+
+	for (i = 0; (name = ref_reloc_sym_names[i]) != NULL; i++) {
+		addr1 = kallsyms__get_function_start(from, name);
+		if (addr1)
+			break;
+	}
+
+	if (name)
+		addr2 = kallsyms__get_function_start(to, name);
+
+	return addr1 == addr2;
+}
+
+static int build_id_cache__kcore_existing(const char *from_dir, char *to_dir,
+					  size_t to_dir_sz)
+{
+	char from[PATH_MAX];
+	char to[PATH_MAX];
+	char to_subdir[PATH_MAX];
+	struct dirent *dent;
+	int ret = -1;
+	DIR *d;
+
+	d = opendir(to_dir);
+	if (!d)
+		return -1;
+
+	scnprintf(from, sizeof(from), "%s/modules", from_dir);
+
+	while (1) {
+		dent = readdir(d);
+		if (!dent)
+			break;
+		if (dent->d_type != DT_DIR)
+			continue;
+		scnprintf(to, sizeof(to), "%s/%s/modules", to_dir,
+			  dent->d_name);
+		scnprintf(to_subdir, sizeof(to_subdir), "%s/%s",
+			  to_dir, dent->d_name);
+		if (!compare_proc_modules(from, to) &&
+		    same_kallsyms_reloc(from_dir, to_subdir)) {
+			strlcpy(to_dir, to_subdir, to_dir_sz);
+			ret = 0;
+			break;
+		}
+	}
+
+	closedir(d);
+
+	return ret;
+}
+
+static int build_id_cache__add_kcore(const char *filename, bool force)
+{
+	char dir[32], sbuildid[BUILD_ID_SIZE * 2 + 1];
+	char from_dir[PATH_MAX], to_dir[PATH_MAX];
+	char *p;
+
+	strlcpy(from_dir, filename, sizeof(from_dir));
+
+	p = strrchr(from_dir, '/');
+	if (!p || strcmp(p + 1, "kcore"))
+		return -1;
+	*p = '\0';
+
+	if (build_id_cache__kcore_buildid(from_dir, sbuildid))
+		return -1;
+
+	scnprintf(to_dir, sizeof(to_dir), "%s/[kernel.kcore]/%s",
+		  buildid_dir, sbuildid);
+
+	if (!force &&
+	    !build_id_cache__kcore_existing(from_dir, to_dir, sizeof(to_dir))) {
+		pr_debug("same kcore found in %s\n", to_dir);
+		return 0;
+	}
+
+	if (build_id_cache__kcore_dir(dir, sizeof(dir)))
+		return -1;
+
+	scnprintf(to_dir, sizeof(to_dir), "%s/[kernel.kcore]/%s/%s",
+		  buildid_dir, sbuildid, dir);
+
+	if (mkdir_p(to_dir, 0755))
+		return -1;
+
+	if (kcore_copy(from_dir, to_dir)) {
+		/* Remove YYYYmmddHHMMSShh directory */
+		if (!rmdir(to_dir)) {
+			p = strrchr(to_dir, '/');
+			if (p)
+				*p = '\0';
+			/* Try to remove buildid directory */
+			if (!rmdir(to_dir)) {
+				p = strrchr(to_dir, '/');
+				if (p)
+					*p = '\0';
+				/* Try to remove [kernel.kcore] directory */
+				rmdir(to_dir);
+			}
+		}
+		return -1;
+	}
+
+	pr_debug("kcore added to build-id cache directory %s\n", to_dir);
+
+	return 0;
+}
+
+static int build_id_cache__add_file(const char *filename)
+{
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+	u8 build_id[BUILD_ID_SIZE];
+	int err;
+
+	if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) {
+		pr_debug("Couldn't read a build-id in %s\n", filename);
+		return -1;
+	}
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuild_id);
+	err = build_id_cache__add_s(sbuild_id, filename,
+				    false, false);
+	pr_debug("Adding %s %s: %s\n", sbuild_id, filename,
+		 err ? "FAIL" : "Ok");
+	return err;
+}
+
+static int build_id_cache__remove_file(const char *filename)
+{
+	u8 build_id[BUILD_ID_SIZE];
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	int err;
+
+	if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) {
+		pr_debug("Couldn't read a build-id in %s\n", filename);
+		return -1;
+	}
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuild_id);
+	err = build_id_cache__remove_s(sbuild_id);
+	pr_debug("Removing %s %s: %s\n", sbuild_id, filename,
+		 err ? "FAIL" : "Ok");
+
+	return err;
+}
+
+static int build_id_cache__purge_path(const char *pathname)
+{
+	struct strlist *list;
+	struct str_node *pos;
+	int err;
+
+	err = build_id_cache__list_build_ids(pathname, &list);
+	if (err)
+		goto out;
+
+	strlist__for_each(pos, list) {
+		err = build_id_cache__remove_s(pos->s);
+		pr_debug("Removing %s %s: %s\n", pos->s, pathname,
+			 err ? "FAIL" : "Ok");
+		if (err)
+			break;
+	}
+	strlist__delete(list);
+
+out:
+	pr_debug("Purging %s: %s\n", pathname, err ? "FAIL" : "Ok");
+
+	return err;
+}
+
+static bool dso__missing_buildid_cache(struct dso *dso, int parm __maybe_unused)
+{
+	char filename[PATH_MAX];
+	u8 build_id[BUILD_ID_SIZE];
+
+	if (dso__build_id_filename(dso, filename, sizeof(filename)) &&
+	    filename__read_build_id(filename, build_id,
+				    sizeof(build_id)) != sizeof(build_id)) {
+		if (errno == ENOENT)
+			return false;
+
+		pr_warning("Problems with %s file, consider removing it from the cache\n",
+			   filename);
+	} else if (memcmp(dso->build_id, build_id, sizeof(dso->build_id))) {
+		pr_warning("Problems with %s file, consider removing it from the cache\n",
+			   filename);
+	}
+
+	return true;
+}
+
+static int build_id_cache__fprintf_missing(struct perf_session *session, FILE *fp)
+{
+	perf_session__fprintf_dsos_buildid(session, fp, dso__missing_buildid_cache, 0);
+	return 0;
+}
+
+static int build_id_cache__update_file(const char *filename)
+{
+	u8 build_id[BUILD_ID_SIZE];
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	int err = 0;
+
+	if (filename__read_build_id(filename, &build_id, sizeof(build_id)) < 0) {
+		pr_debug("Couldn't read a build-id in %s\n", filename);
+		return -1;
+	}
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuild_id);
+	if (build_id_cache__cached(sbuild_id))
+		err = build_id_cache__remove_s(sbuild_id);
+
+	if (!err)
+		err = build_id_cache__add_s(sbuild_id, filename, false, false);
+
+	pr_debug("Updating %s %s: %s\n", sbuild_id, filename,
+		 err ? "FAIL" : "Ok");
+
+	return err;
+}
+
+int cmd_buildid_cache(int argc, const char **argv,
+		      const char *prefix __maybe_unused)
+{
+	struct strlist *list;
+	struct str_node *pos;
+	int ret = 0;
+	bool force = false;
+	char const *add_name_list_str = NULL,
+		   *remove_name_list_str = NULL,
+		   *purge_name_list_str = NULL,
+		   *missing_filename = NULL,
+		   *update_name_list_str = NULL,
+		   *kcore_filename = NULL;
+	char sbuf[STRERR_BUFSIZE];
+
+	struct perf_data_file file = {
+		.mode  = PERF_DATA_MODE_READ,
+	};
+	struct perf_session *session = NULL;
+
+	const struct option buildid_cache_options[] = {
+	OPT_STRING('a', "add", &add_name_list_str,
+		   "file list", "file(s) to add"),
+	OPT_STRING('k', "kcore", &kcore_filename,
+		   "file", "kcore file to add"),
+	OPT_STRING('r', "remove", &remove_name_list_str, "file list",
+		    "file(s) to remove"),
+	OPT_STRING('p', "purge", &purge_name_list_str, "path list",
+		    "path(s) to remove (remove old caches too)"),
+	OPT_STRING('M', "missing", &missing_filename, "file",
+		   "to find missing build ids in the cache"),
+	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
+	OPT_STRING('u', "update", &update_name_list_str, "file list",
+		    "file(s) to update"),
+	OPT_INCR('v', "verbose", &verbose, "be more verbose"),
+	OPT_END()
+	};
+	const char * const buildid_cache_usage[] = {
+		"perf buildid-cache [<options>]",
+		NULL
+	};
+
+	argc = parse_options(argc, argv, buildid_cache_options,
+			     buildid_cache_usage, 0);
+
+	if (argc || (!add_name_list_str && !kcore_filename &&
+		     !remove_name_list_str && !purge_name_list_str &&
+		     !missing_filename && !update_name_list_str))
+		usage_with_options(buildid_cache_usage, buildid_cache_options);
+
+	if (missing_filename) {
+		file.path = missing_filename;
+		file.force = force;
+
+		session = perf_session__new(&file, false, NULL);
+		if (session == NULL)
+			return -1;
+	}
+
+	if (symbol__init(session ? &session->header.env : NULL) < 0)
+		goto out;
+
+	setup_pager();
+
+	if (add_name_list_str) {
+		list = strlist__new(true, add_name_list_str);
+		if (list) {
+			strlist__for_each(pos, list)
+				if (build_id_cache__add_file(pos->s)) {
+					if (errno == EEXIST) {
+						pr_debug("%s already in the cache\n",
+							 pos->s);
+						continue;
+					}
+					pr_warning("Couldn't add %s: %s\n",
+						   pos->s, strerror_r(errno, sbuf, sizeof(sbuf)));
+				}
+
+			strlist__delete(list);
+		}
+	}
+
+	if (remove_name_list_str) {
+		list = strlist__new(true, remove_name_list_str);
+		if (list) {
+			strlist__for_each(pos, list)
+				if (build_id_cache__remove_file(pos->s)) {
+					if (errno == ENOENT) {
+						pr_debug("%s wasn't in the cache\n",
+							 pos->s);
+						continue;
+					}
+					pr_warning("Couldn't remove %s: %s\n",
+						   pos->s, strerror_r(errno, sbuf, sizeof(sbuf)));
+				}
+
+			strlist__delete(list);
+		}
+	}
+
+	if (purge_name_list_str) {
+		list = strlist__new(true, purge_name_list_str);
+		if (list) {
+			strlist__for_each(pos, list)
+				if (build_id_cache__purge_path(pos->s)) {
+					if (errno == ENOENT) {
+						pr_debug("%s wasn't in the cache\n",
+							 pos->s);
+						continue;
+					}
+					pr_warning("Couldn't remove %s: %s\n",
+						   pos->s, strerror_r(errno, sbuf, sizeof(sbuf)));
+				}
+
+			strlist__delete(list);
+		}
+	}
+
+	if (missing_filename)
+		ret = build_id_cache__fprintf_missing(session, stdout);
+
+	if (update_name_list_str) {
+		list = strlist__new(true, update_name_list_str);
+		if (list) {
+			strlist__for_each(pos, list)
+				if (build_id_cache__update_file(pos->s)) {
+					if (errno == ENOENT) {
+						pr_debug("%s wasn't in the cache\n",
+							 pos->s);
+						continue;
+					}
+					pr_warning("Couldn't update %s: %s\n",
+						   pos->s, strerror_r(errno, sbuf, sizeof(sbuf)));
+				}
+
+			strlist__delete(list);
+		}
+	}
+
+	if (kcore_filename && build_id_cache__add_kcore(kcore_filename, force))
+		pr_warning("Couldn't add %s\n", kcore_filename);
+
+out:
+	if (session)
+		perf_session__delete(session);
+
+	return ret;
+}
diff --git a/tools/perf/builtin-buildid-list.c b/tools/perf/builtin-buildid-list.c
new file mode 100644
index 000000000..feb420f74
--- /dev/null
+++ b/tools/perf/builtin-buildid-list.c
@@ -0,0 +1,111 @@
+/*
+ * builtin-buildid-list.c
+ *
+ * Builtin buildid-list command: list buildids in perf.data, in the running
+ * kernel and in ELF files.
+ *
+ * Copyright (C) 2009, Red Hat Inc.
+ * Copyright (C) 2009, Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include "builtin.h"
+#include "perf.h"
+#include "util/build-id.h"
+#include "util/cache.h"
+#include "util/debug.h"
+#include "util/parse-options.h"
+#include "util/session.h"
+#include "util/symbol.h"
+#include "util/data.h"
+
+static int sysfs__fprintf_build_id(FILE *fp)
+{
+	u8 kallsyms_build_id[BUILD_ID_SIZE];
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	if (sysfs__read_build_id("/sys/kernel/notes", kallsyms_build_id,
+				 sizeof(kallsyms_build_id)) != 0)
+		return -1;
+
+	build_id__sprintf(kallsyms_build_id, sizeof(kallsyms_build_id),
+			  sbuild_id);
+	fprintf(fp, "%s\n", sbuild_id);
+	return 0;
+}
+
+static int filename__fprintf_build_id(const char *name, FILE *fp)
+{
+	u8 build_id[BUILD_ID_SIZE];
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	if (filename__read_build_id(name, build_id,
+				    sizeof(build_id)) != sizeof(build_id))
+		return 0;
+
+	build_id__sprintf(build_id, sizeof(build_id), sbuild_id);
+	return fprintf(fp, "%s\n", sbuild_id);
+}
+
+static bool dso__skip_buildid(struct dso *dso, int with_hits)
+{
+	return with_hits && !dso->hit;
+}
+
+static int perf_session__list_build_ids(bool force, bool with_hits)
+{
+	struct perf_session *session;
+	struct perf_data_file file = {
+		.path  = input_name,
+		.mode  = PERF_DATA_MODE_READ,
+		.force = force,
+	};
+
+	symbol__elf_init();
+	/*
+	 * See if this is an ELF file first:
+	 */
+	if (filename__fprintf_build_id(input_name, stdout))
+		goto out;
+
+	session = perf_session__new(&file, false, &build_id__mark_dso_hit_ops);
+	if (session == NULL)
+		return -1;
+	/*
+	 * in pipe-mode, the only way to get the buildids is to parse
+	 * the record stream. Buildids are stored as RECORD_HEADER_BUILD_ID
+	 */
+	if (with_hits || perf_data_file__is_pipe(&file))
+		perf_session__process_events(session);
+
+	perf_session__fprintf_dsos_buildid(session, stdout, dso__skip_buildid, with_hits);
+	perf_session__delete(session);
+out:
+	return 0;
+}
+
+int cmd_buildid_list(int argc, const char **argv,
+		     const char *prefix __maybe_unused)
+{
+	bool show_kernel = false;
+	bool with_hits = false;
+	bool force = false;
+	const struct option options[] = {
+	OPT_BOOLEAN('H', "with-hits", &with_hits, "Show only DSOs with hits"),
+	OPT_STRING('i', "input", &input_name, "file", "input file name"),
+	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
+	OPT_BOOLEAN('k', "kernel", &show_kernel, "Show current kernel build id"),
+	OPT_INCR('v', "verbose", &verbose, "be more verbose"),
+	OPT_END()
+	};
+	const char * const buildid_list_usage[] = {
+		"perf buildid-list [<options>]",
+		NULL
+	};
+
+	argc = parse_options(argc, argv, options, buildid_list_usage, 0);
+	setup_pager();
+
+	if (show_kernel)
+		return sysfs__fprintf_build_id(stdout);
+
+	return perf_session__list_build_ids(force, with_hits);
+}
diff --git a/tools/perf/builtin-data.c b/tools/perf/builtin-data.c
new file mode 100644
index 000000000..d6525bc54
--- /dev/null
+++ b/tools/perf/builtin-data.c
@@ -0,0 +1,123 @@
+#include <linux/compiler.h>
+#include "builtin.h"
+#include "perf.h"
+#include "debug.h"
+#include "parse-options.h"
+#include "data-convert-bt.h"
+
+typedef int (*data_cmd_fn_t)(int argc, const char **argv, const char *prefix);
+
+struct data_cmd {
+	const char	*name;
+	const char	*summary;
+	data_cmd_fn_t	fn;
+};
+
+static struct data_cmd data_cmds[];
+
+#define for_each_cmd(cmd) \
+	for (cmd = data_cmds; cmd && cmd->name; cmd++)
+
+static const struct option data_options[] = {
+	OPT_END()
+};
+
+static const char * const data_subcommands[] = { "convert", NULL };
+
+static const char *data_usage[] = {
+	"perf data [<common options>] <command> [<options>]",
+	NULL
+};
+
+static void print_usage(void)
+{
+	struct data_cmd *cmd;
+
+	printf("Usage:\n");
+	printf("\t%s\n\n", data_usage[0]);
+	printf("\tAvailable commands:\n");
+
+	for_each_cmd(cmd) {
+		printf("\t %s\t- %s\n", cmd->name, cmd->summary);
+	}
+
+	printf("\n");
+}
+
+static const char * const data_convert_usage[] = {
+	"perf data convert [<options>]",
+	NULL
+};
+
+static int cmd_data_convert(int argc, const char **argv,
+			    const char *prefix __maybe_unused)
+{
+	const char *to_ctf     = NULL;
+	bool force = false;
+	const struct option options[] = {
+		OPT_INCR('v', "verbose", &verbose, "be more verbose"),
+		OPT_STRING('i', "input", &input_name, "file", "input file name"),
+#ifdef HAVE_LIBBABELTRACE_SUPPORT
+		OPT_STRING(0, "to-ctf", &to_ctf, NULL, "Convert to CTF format"),
+#endif
+		OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
+		OPT_END()
+	};
+
+#ifndef HAVE_LIBBABELTRACE_SUPPORT
+	pr_err("No conversion support compiled in.\n");
+	return -1;
+#endif
+
+	argc = parse_options(argc, argv, options,
+			     data_convert_usage, 0);
+	if (argc) {
+		usage_with_options(data_convert_usage, options);
+		return -1;
+	}
+
+	if (to_ctf) {
+#ifdef HAVE_LIBBABELTRACE_SUPPORT
+		return bt_convert__perf2ctf(input_name, to_ctf, force);
+#else
+		pr_err("The libbabeltrace support is not compiled in.\n");
+		return -1;
+#endif
+	}
+
+	return 0;
+}
+
+static struct data_cmd data_cmds[] = {
+	{ "convert", "converts data file between formats", cmd_data_convert },
+	{ .name = NULL, },
+};
+
+int cmd_data(int argc, const char **argv, const char *prefix)
+{
+	struct data_cmd *cmd;
+	const char *cmdstr;
+
+	/* No command specified. */
+	if (argc < 2)
+		goto usage;
+
+	argc = parse_options_subcommand(argc, argv, data_options, data_subcommands, data_usage,
+			     PARSE_OPT_STOP_AT_NON_OPTION);
+	if (argc < 1)
+		goto usage;
+
+	cmdstr = argv[0];
+
+	for_each_cmd(cmd) {
+		if (strcmp(cmd->name, cmdstr))
+			continue;
+
+		return cmd->fn(argc, argv, prefix);
+	}
+
+	pr_err("Unknown command: %s\n", cmdstr);
+usage:
+	print_usage();
+	return -1;
+}
diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c
new file mode 100644
index 000000000..df6307b40
--- /dev/null
+++ b/tools/perf/builtin-diff.c
@@ -0,0 +1,1285 @@
+/*
+ * builtin-diff.c
+ *
+ * Builtin diff command: Analyze two perf.data input files, look up and read
+ * DSOs and symbol information, sort them and produce a diff.
+ */
+#include "builtin.h"
+
+#include "util/debug.h"
+#include "util/event.h"
+#include "util/hist.h"
+#include "util/evsel.h"
+#include "util/evlist.h"
+#include "util/session.h"
+#include "util/tool.h"
+#include "util/sort.h"
+#include "util/symbol.h"
+#include "util/util.h"
+#include "util/data.h"
+
+#include <stdlib.h>
+#include <math.h>
+
+/* Diff command specific HPP columns. */
+enum {
+	PERF_HPP_DIFF__BASELINE,
+	PERF_HPP_DIFF__PERIOD,
+	PERF_HPP_DIFF__PERIOD_BASELINE,
+	PERF_HPP_DIFF__DELTA,
+	PERF_HPP_DIFF__RATIO,
+	PERF_HPP_DIFF__WEIGHTED_DIFF,
+	PERF_HPP_DIFF__FORMULA,
+
+	PERF_HPP_DIFF__MAX_INDEX
+};
+
+struct diff_hpp_fmt {
+	struct perf_hpp_fmt	 fmt;
+	int			 idx;
+	char			*header;
+	int			 header_width;
+};
+
+struct data__file {
+	struct perf_session	*session;
+	struct perf_data_file	file;
+	int			 idx;
+	struct hists		*hists;
+	struct diff_hpp_fmt	 fmt[PERF_HPP_DIFF__MAX_INDEX];
+};
+
+static struct data__file *data__files;
+static int data__files_cnt;
+
+#define data__for_each_file_start(i, d, s)	\
+	for (i = s, d = &data__files[s];	\
+	     i < data__files_cnt;		\
+	     i++, d = &data__files[i])
+
+#define data__for_each_file(i, d) data__for_each_file_start(i, d, 0)
+#define data__for_each_file_new(i, d) data__for_each_file_start(i, d, 1)
+
+static bool force;
+static bool show_period;
+static bool show_formula;
+static bool show_baseline_only;
+static unsigned int sort_compute;
+
+static s64 compute_wdiff_w1;
+static s64 compute_wdiff_w2;
+
+enum {
+	COMPUTE_DELTA,
+	COMPUTE_RATIO,
+	COMPUTE_WEIGHTED_DIFF,
+	COMPUTE_MAX,
+};
+
+const char *compute_names[COMPUTE_MAX] = {
+	[COMPUTE_DELTA] = "delta",
+	[COMPUTE_RATIO] = "ratio",
+	[COMPUTE_WEIGHTED_DIFF] = "wdiff",
+};
+
+static int compute;
+
+static int compute_2_hpp[COMPUTE_MAX] = {
+	[COMPUTE_DELTA]		= PERF_HPP_DIFF__DELTA,
+	[COMPUTE_RATIO]		= PERF_HPP_DIFF__RATIO,
+	[COMPUTE_WEIGHTED_DIFF]	= PERF_HPP_DIFF__WEIGHTED_DIFF,
+};
+
+#define MAX_COL_WIDTH 70
+
+static struct header_column {
+	const char *name;
+	int width;
+} columns[PERF_HPP_DIFF__MAX_INDEX] = {
+	[PERF_HPP_DIFF__BASELINE] = {
+		.name  = "Baseline",
+	},
+	[PERF_HPP_DIFF__PERIOD] = {
+		.name  = "Period",
+		.width = 14,
+	},
+	[PERF_HPP_DIFF__PERIOD_BASELINE] = {
+		.name  = "Base period",
+		.width = 14,
+	},
+	[PERF_HPP_DIFF__DELTA] = {
+		.name  = "Delta",
+		.width = 7,
+	},
+	[PERF_HPP_DIFF__RATIO] = {
+		.name  = "Ratio",
+		.width = 14,
+	},
+	[PERF_HPP_DIFF__WEIGHTED_DIFF] = {
+		.name  = "Weighted diff",
+		.width = 14,
+	},
+	[PERF_HPP_DIFF__FORMULA] = {
+		.name  = "Formula",
+		.width = MAX_COL_WIDTH,
+	}
+};
+
+static int setup_compute_opt_wdiff(char *opt)
+{
+	char *w1_str = opt;
+	char *w2_str;
+
+	int ret = -EINVAL;
+
+	if (!opt)
+		goto out;
+
+	w2_str = strchr(opt, ',');
+	if (!w2_str)
+		goto out;
+
+	*w2_str++ = 0x0;
+	if (!*w2_str)
+		goto out;
+
+	compute_wdiff_w1 = strtol(w1_str, NULL, 10);
+	compute_wdiff_w2 = strtol(w2_str, NULL, 10);
+
+	if (!compute_wdiff_w1 || !compute_wdiff_w2)
+		goto out;
+
+	pr_debug("compute wdiff w1(%" PRId64 ") w2(%" PRId64 ")\n",
+		  compute_wdiff_w1, compute_wdiff_w2);
+
+	ret = 0;
+
+ out:
+	if (ret)
+		pr_err("Failed: wrong weight data, use 'wdiff:w1,w2'\n");
+
+	return ret;
+}
+
+static int setup_compute_opt(char *opt)
+{
+	if (compute == COMPUTE_WEIGHTED_DIFF)
+		return setup_compute_opt_wdiff(opt);
+
+	if (opt) {
+		pr_err("Failed: extra option specified '%s'", opt);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int setup_compute(const struct option *opt, const char *str,
+			 int unset __maybe_unused)
+{
+	int *cp = (int *) opt->value;
+	char *cstr = (char *) str;
+	char buf[50];
+	unsigned i;
+	char *option;
+
+	if (!str) {
+		*cp = COMPUTE_DELTA;
+		return 0;
+	}
+
+	option = strchr(str, ':');
+	if (option) {
+		unsigned len = option++ - str;
+
+		/*
+		 * The str data are not writeable, so we need
+		 * to use another buffer.
+		 */
+
+		/* No option value is longer. */
+		if (len >= sizeof(buf))
+			return -EINVAL;
+
+		strncpy(buf, str, len);
+		buf[len] = 0x0;
+		cstr = buf;
+	}
+
+	for (i = 0; i < COMPUTE_MAX; i++)
+		if (!strcmp(cstr, compute_names[i])) {
+			*cp = i;
+			return setup_compute_opt(option);
+		}
+
+	pr_err("Failed: '%s' is not computation method "
+	       "(use 'delta','ratio' or 'wdiff')\n", str);
+	return -EINVAL;
+}
+
+static double period_percent(struct hist_entry *he, u64 period)
+{
+	u64 total = hists__total_period(he->hists);
+
+	return (period * 100.0) / total;
+}
+
+static double compute_delta(struct hist_entry *he, struct hist_entry *pair)
+{
+	double old_percent = period_percent(he, he->stat.period);
+	double new_percent = period_percent(pair, pair->stat.period);
+
+	pair->diff.period_ratio_delta = new_percent - old_percent;
+	pair->diff.computed = true;
+	return pair->diff.period_ratio_delta;
+}
+
+static double compute_ratio(struct hist_entry *he, struct hist_entry *pair)
+{
+	double old_period = he->stat.period ?: 1;
+	double new_period = pair->stat.period;
+
+	pair->diff.computed = true;
+	pair->diff.period_ratio = new_period / old_period;
+	return pair->diff.period_ratio;
+}
+
+static s64 compute_wdiff(struct hist_entry *he, struct hist_entry *pair)
+{
+	u64 old_period = he->stat.period;
+	u64 new_period = pair->stat.period;
+
+	pair->diff.computed = true;
+	pair->diff.wdiff = new_period * compute_wdiff_w2 -
+			   old_period * compute_wdiff_w1;
+
+	return pair->diff.wdiff;
+}
+
+static int formula_delta(struct hist_entry *he, struct hist_entry *pair,
+			 char *buf, size_t size)
+{
+	u64 he_total = he->hists->stats.total_period;
+	u64 pair_total = pair->hists->stats.total_period;
+
+	if (symbol_conf.filter_relative) {
+		he_total = he->hists->stats.total_non_filtered_period;
+		pair_total = pair->hists->stats.total_non_filtered_period;
+	}
+	return scnprintf(buf, size,
+			 "(%" PRIu64 " * 100 / %" PRIu64 ") - "
+			 "(%" PRIu64 " * 100 / %" PRIu64 ")",
+			 pair->stat.period, pair_total,
+			 he->stat.period, he_total);
+}
+
+static int formula_ratio(struct hist_entry *he, struct hist_entry *pair,
+			 char *buf, size_t size)
+{
+	double old_period = he->stat.period;
+	double new_period = pair->stat.period;
+
+	return scnprintf(buf, size, "%.0F / %.0F", new_period, old_period);
+}
+
+static int formula_wdiff(struct hist_entry *he, struct hist_entry *pair,
+			 char *buf, size_t size)
+{
+	u64 old_period = he->stat.period;
+	u64 new_period = pair->stat.period;
+
+	return scnprintf(buf, size,
+		  "(%" PRIu64 " * " "%" PRId64 ") - (%" PRIu64 " * " "%" PRId64 ")",
+		  new_period, compute_wdiff_w2, old_period, compute_wdiff_w1);
+}
+
+static int formula_fprintf(struct hist_entry *he, struct hist_entry *pair,
+			   char *buf, size_t size)
+{
+	switch (compute) {
+	case COMPUTE_DELTA:
+		return formula_delta(he, pair, buf, size);
+	case COMPUTE_RATIO:
+		return formula_ratio(he, pair, buf, size);
+	case COMPUTE_WEIGHTED_DIFF:
+		return formula_wdiff(he, pair, buf, size);
+	default:
+		BUG_ON(1);
+	}
+
+	return -1;
+}
+
+static int hists__add_entry(struct hists *hists,
+			    struct addr_location *al, u64 period,
+			    u64 weight, u64 transaction)
+{
+	if (__hists__add_entry(hists, al, NULL, NULL, NULL, period, weight,
+			       transaction, true) != NULL)
+		return 0;
+	return -ENOMEM;
+}
+
+static int diff__process_sample_event(struct perf_tool *tool __maybe_unused,
+				      union perf_event *event,
+				      struct perf_sample *sample,
+				      struct perf_evsel *evsel,
+				      struct machine *machine)
+{
+	struct addr_location al;
+	struct hists *hists = evsel__hists(evsel);
+
+	if (perf_event__preprocess_sample(event, machine, &al, sample) < 0) {
+		pr_warning("problem processing %d event, skipping it.\n",
+			   event->header.type);
+		return -1;
+	}
+
+	if (hists__add_entry(hists, &al, sample->period,
+			     sample->weight, sample->transaction)) {
+		pr_warning("problem incrementing symbol period, skipping event\n");
+		return -1;
+	}
+
+	/*
+	 * The total_period is updated here before going to the output
+	 * tree since normally only the baseline hists will call
+	 * hists__output_resort() and precompute needs the total
+	 * period in order to sort entries by percentage delta.
+	 */
+	hists->stats.total_period += sample->period;
+	if (!al.filtered)
+		hists->stats.total_non_filtered_period += sample->period;
+
+	return 0;
+}
+
+static struct perf_tool tool = {
+	.sample	= diff__process_sample_event,
+	.mmap	= perf_event__process_mmap,
+	.mmap2	= perf_event__process_mmap2,
+	.comm	= perf_event__process_comm,
+	.exit	= perf_event__process_exit,
+	.fork	= perf_event__process_fork,
+	.lost	= perf_event__process_lost,
+	.ordered_events = true,
+	.ordering_requires_timestamps = true,
+};
+
+static struct perf_evsel *evsel_match(struct perf_evsel *evsel,
+				      struct perf_evlist *evlist)
+{
+	struct perf_evsel *e;
+
+	evlist__for_each(evlist, e) {
+		if (perf_evsel__match2(evsel, e))
+			return e;
+	}
+
+	return NULL;
+}
+
+static void perf_evlist__collapse_resort(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel) {
+		struct hists *hists = evsel__hists(evsel);
+
+		hists__collapse_resort(hists, NULL);
+	}
+}
+
+static struct data__file *fmt_to_data_file(struct perf_hpp_fmt *fmt)
+{
+	struct diff_hpp_fmt *dfmt = container_of(fmt, struct diff_hpp_fmt, fmt);
+	void *ptr = dfmt - dfmt->idx;
+	struct data__file *d = container_of(ptr, struct data__file, fmt);
+
+	return d;
+}
+
+static struct hist_entry*
+get_pair_data(struct hist_entry *he, struct data__file *d)
+{
+	if (hist_entry__has_pairs(he)) {
+		struct hist_entry *pair;
+
+		list_for_each_entry(pair, &he->pairs.head, pairs.node)
+			if (pair->hists == d->hists)
+				return pair;
+	}
+
+	return NULL;
+}
+
+static struct hist_entry*
+get_pair_fmt(struct hist_entry *he, struct diff_hpp_fmt *dfmt)
+{
+	struct data__file *d = fmt_to_data_file(&dfmt->fmt);
+
+	return get_pair_data(he, d);
+}
+
+static void hists__baseline_only(struct hists *hists)
+{
+	struct rb_root *root;
+	struct rb_node *next;
+
+	if (sort__need_collapse)
+		root = &hists->entries_collapsed;
+	else
+		root = hists->entries_in;
+
+	next = rb_first(root);
+	while (next != NULL) {
+		struct hist_entry *he = rb_entry(next, struct hist_entry, rb_node_in);
+
+		next = rb_next(&he->rb_node_in);
+		if (!hist_entry__next_pair(he)) {
+			rb_erase(&he->rb_node_in, root);
+			hist_entry__delete(he);
+		}
+	}
+}
+
+static void hists__precompute(struct hists *hists)
+{
+	struct rb_root *root;
+	struct rb_node *next;
+
+	if (sort__need_collapse)
+		root = &hists->entries_collapsed;
+	else
+		root = hists->entries_in;
+
+	next = rb_first(root);
+	while (next != NULL) {
+		struct hist_entry *he, *pair;
+		struct data__file *d;
+		int i;
+
+		he   = rb_entry(next, struct hist_entry, rb_node_in);
+		next = rb_next(&he->rb_node_in);
+
+		data__for_each_file_new(i, d) {
+			pair = get_pair_data(he, d);
+			if (!pair)
+				continue;
+
+			switch (compute) {
+			case COMPUTE_DELTA:
+				compute_delta(he, pair);
+				break;
+			case COMPUTE_RATIO:
+				compute_ratio(he, pair);
+				break;
+			case COMPUTE_WEIGHTED_DIFF:
+				compute_wdiff(he, pair);
+				break;
+			default:
+				BUG_ON(1);
+			}
+		}
+	}
+}
+
+static int64_t cmp_doubles(double l, double r)
+{
+	if (l > r)
+		return -1;
+	else if (l < r)
+		return 1;
+	else
+		return 0;
+}
+
+static int64_t
+__hist_entry__cmp_compute(struct hist_entry *left, struct hist_entry *right,
+			int c)
+{
+	switch (c) {
+	case COMPUTE_DELTA:
+	{
+		double l = left->diff.period_ratio_delta;
+		double r = right->diff.period_ratio_delta;
+
+		return cmp_doubles(l, r);
+	}
+	case COMPUTE_RATIO:
+	{
+		double l = left->diff.period_ratio;
+		double r = right->diff.period_ratio;
+
+		return cmp_doubles(l, r);
+	}
+	case COMPUTE_WEIGHTED_DIFF:
+	{
+		s64 l = left->diff.wdiff;
+		s64 r = right->diff.wdiff;
+
+		return r - l;
+	}
+	default:
+		BUG_ON(1);
+	}
+
+	return 0;
+}
+
+static int64_t
+hist_entry__cmp_compute(struct hist_entry *left, struct hist_entry *right,
+			int c, int sort_idx)
+{
+	bool pairs_left  = hist_entry__has_pairs(left);
+	bool pairs_right = hist_entry__has_pairs(right);
+	struct hist_entry *p_right, *p_left;
+
+	if (!pairs_left && !pairs_right)
+		return 0;
+
+	if (!pairs_left || !pairs_right)
+		return pairs_left ? -1 : 1;
+
+	p_left  = get_pair_data(left,  &data__files[sort_idx]);
+	p_right = get_pair_data(right, &data__files[sort_idx]);
+
+	if (!p_left && !p_right)
+		return 0;
+
+	if (!p_left || !p_right)
+		return p_left ? -1 : 1;
+
+	/*
+	 * We have 2 entries of same kind, let's
+	 * make the data comparison.
+	 */
+	return __hist_entry__cmp_compute(p_left, p_right, c);
+}
+
+static int64_t
+hist_entry__cmp_compute_idx(struct hist_entry *left, struct hist_entry *right,
+			    int c, int sort_idx)
+{
+	struct hist_entry *p_right, *p_left;
+
+	p_left  = get_pair_data(left,  &data__files[sort_idx]);
+	p_right = get_pair_data(right, &data__files[sort_idx]);
+
+	if (!p_left && !p_right)
+		return 0;
+
+	if (!p_left || !p_right)
+		return p_left ? -1 : 1;
+
+	if (c != COMPUTE_DELTA) {
+		/*
+		 * The delta can be computed without the baseline, but
+		 * others are not.  Put those entries which have no
+		 * values below.
+		 */
+		if (left->dummy && right->dummy)
+			return 0;
+
+		if (left->dummy || right->dummy)
+			return left->dummy ? 1 : -1;
+	}
+
+	return __hist_entry__cmp_compute(p_left, p_right, c);
+}
+
+static int64_t
+hist_entry__cmp_nop(struct perf_hpp_fmt *fmt __maybe_unused,
+		    struct hist_entry *left __maybe_unused,
+		    struct hist_entry *right __maybe_unused)
+{
+	return 0;
+}
+
+static int64_t
+hist_entry__cmp_baseline(struct perf_hpp_fmt *fmt __maybe_unused,
+			 struct hist_entry *left, struct hist_entry *right)
+{
+	if (left->stat.period == right->stat.period)
+		return 0;
+	return left->stat.period > right->stat.period ? 1 : -1;
+}
+
+static int64_t
+hist_entry__cmp_delta(struct perf_hpp_fmt *fmt,
+		      struct hist_entry *left, struct hist_entry *right)
+{
+	struct data__file *d = fmt_to_data_file(fmt);
+
+	return hist_entry__cmp_compute(right, left, COMPUTE_DELTA, d->idx);
+}
+
+static int64_t
+hist_entry__cmp_ratio(struct perf_hpp_fmt *fmt,
+		      struct hist_entry *left, struct hist_entry *right)
+{
+	struct data__file *d = fmt_to_data_file(fmt);
+
+	return hist_entry__cmp_compute(right, left, COMPUTE_RATIO, d->idx);
+}
+
+static int64_t
+hist_entry__cmp_wdiff(struct perf_hpp_fmt *fmt,
+		      struct hist_entry *left, struct hist_entry *right)
+{
+	struct data__file *d = fmt_to_data_file(fmt);
+
+	return hist_entry__cmp_compute(right, left, COMPUTE_WEIGHTED_DIFF, d->idx);
+}
+
+static int64_t
+hist_entry__cmp_delta_idx(struct perf_hpp_fmt *fmt __maybe_unused,
+			  struct hist_entry *left, struct hist_entry *right)
+{
+	return hist_entry__cmp_compute_idx(right, left, COMPUTE_DELTA,
+					   sort_compute);
+}
+
+static int64_t
+hist_entry__cmp_ratio_idx(struct perf_hpp_fmt *fmt __maybe_unused,
+			  struct hist_entry *left, struct hist_entry *right)
+{
+	return hist_entry__cmp_compute_idx(right, left, COMPUTE_RATIO,
+					   sort_compute);
+}
+
+static int64_t
+hist_entry__cmp_wdiff_idx(struct perf_hpp_fmt *fmt __maybe_unused,
+			  struct hist_entry *left, struct hist_entry *right)
+{
+	return hist_entry__cmp_compute_idx(right, left, COMPUTE_WEIGHTED_DIFF,
+					   sort_compute);
+}
+
+static void hists__process(struct hists *hists)
+{
+	if (show_baseline_only)
+		hists__baseline_only(hists);
+
+	hists__precompute(hists);
+	hists__output_resort(hists, NULL);
+
+	hists__fprintf(hists, true, 0, 0, 0, stdout);
+}
+
+static void data__fprintf(void)
+{
+	struct data__file *d;
+	int i;
+
+	fprintf(stdout, "# Data files:\n");
+
+	data__for_each_file(i, d)
+		fprintf(stdout, "#  [%d] %s %s\n",
+			d->idx, d->file.path,
+			!d->idx ? "(Baseline)" : "");
+
+	fprintf(stdout, "#\n");
+}
+
+static void data_process(void)
+{
+	struct perf_evlist *evlist_base = data__files[0].session->evlist;
+	struct perf_evsel *evsel_base;
+	bool first = true;
+
+	evlist__for_each(evlist_base, evsel_base) {
+		struct hists *hists_base = evsel__hists(evsel_base);
+		struct data__file *d;
+		int i;
+
+		data__for_each_file_new(i, d) {
+			struct perf_evlist *evlist = d->session->evlist;
+			struct perf_evsel *evsel;
+			struct hists *hists;
+
+			evsel = evsel_match(evsel_base, evlist);
+			if (!evsel)
+				continue;
+
+			hists = evsel__hists(evsel);
+			d->hists = hists;
+
+			hists__match(hists_base, hists);
+
+			if (!show_baseline_only)
+				hists__link(hists_base, hists);
+		}
+
+		fprintf(stdout, "%s# Event '%s'\n#\n", first ? "" : "\n",
+			perf_evsel__name(evsel_base));
+
+		first = false;
+
+		if (verbose || data__files_cnt > 2)
+			data__fprintf();
+
+		hists__process(hists_base);
+	}
+}
+
+static void data__free(struct data__file *d)
+{
+	int col;
+
+	for (col = 0; col < PERF_HPP_DIFF__MAX_INDEX; col++) {
+		struct diff_hpp_fmt *fmt = &d->fmt[col];
+
+		zfree(&fmt->header);
+	}
+}
+
+static int __cmd_diff(void)
+{
+	struct data__file *d;
+	int ret = -EINVAL, i;
+
+	data__for_each_file(i, d) {
+		d->session = perf_session__new(&d->file, false, &tool);
+		if (!d->session) {
+			pr_err("Failed to open %s\n", d->file.path);
+			ret = -1;
+			goto out_delete;
+		}
+
+		ret = perf_session__process_events(d->session);
+		if (ret) {
+			pr_err("Failed to process %s\n", d->file.path);
+			goto out_delete;
+		}
+
+		perf_evlist__collapse_resort(d->session->evlist);
+	}
+
+	data_process();
+
+ out_delete:
+	data__for_each_file(i, d) {
+		if (d->session)
+			perf_session__delete(d->session);
+
+		data__free(d);
+	}
+
+	free(data__files);
+	return ret;
+}
+
+static const char * const diff_usage[] = {
+	"perf diff [<options>] [old_file] [new_file]",
+	NULL,
+};
+
+static const struct option options[] = {
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('b', "baseline-only", &show_baseline_only,
+		    "Show only items with match in baseline"),
+	OPT_CALLBACK('c', "compute", &compute,
+		     "delta,ratio,wdiff:w1,w2 (default delta)",
+		     "Entries differential computation selection",
+		     setup_compute),
+	OPT_BOOLEAN('p', "period", &show_period,
+		    "Show period values."),
+	OPT_BOOLEAN('F', "formula", &show_formula,
+		    "Show formula."),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
+	OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name,
+		   "file", "kallsyms pathname"),
+	OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules,
+		    "load module symbols - WARNING: use only with -k and LIVE kernel"),
+	OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]",
+		   "only consider symbols in these dsos"),
+	OPT_STRING('C', "comms", &symbol_conf.comm_list_str, "comm[,comm...]",
+		   "only consider symbols in these comms"),
+	OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]",
+		   "only consider these symbols"),
+	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
+		   "sort by key(s): pid, comm, dso, symbol, parent, cpu, srcline, ..."
+		   " Please refer the man page for the complete list."),
+	OPT_STRING_NOEMPTY('t', "field-separator", &symbol_conf.field_sep, "separator",
+		   "separator for columns, no spaces will be added between "
+		   "columns '.' is reserved."),
+	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
+		    "Look for files with symbols relative to this directory"),
+	OPT_UINTEGER('o', "order", &sort_compute, "Specify compute sorting."),
+	OPT_CALLBACK(0, "percentage", NULL, "relative|absolute",
+		     "How to display percentage of filtered entries", parse_filter_percentage),
+	OPT_END()
+};
+
+static double baseline_percent(struct hist_entry *he)
+{
+	u64 total = hists__total_period(he->hists);
+
+	return 100.0 * he->stat.period / total;
+}
+
+static int hpp__color_baseline(struct perf_hpp_fmt *fmt,
+			       struct perf_hpp *hpp, struct hist_entry *he)
+{
+	struct diff_hpp_fmt *dfmt =
+		container_of(fmt, struct diff_hpp_fmt, fmt);
+	double percent = baseline_percent(he);
+	char pfmt[20] = " ";
+
+	if (!he->dummy) {
+		scnprintf(pfmt, 20, "%%%d.2f%%%%", dfmt->header_width - 1);
+		return percent_color_snprintf(hpp->buf, hpp->size,
+					      pfmt, percent);
+	} else
+		return scnprintf(hpp->buf, hpp->size, "%*s",
+				 dfmt->header_width, pfmt);
+}
+
+static int hpp__entry_baseline(struct hist_entry *he, char *buf, size_t size)
+{
+	double percent = baseline_percent(he);
+	const char *fmt = symbol_conf.field_sep ? "%.2f" : "%6.2f%%";
+	int ret = 0;
+
+	if (!he->dummy)
+		ret = scnprintf(buf, size, fmt, percent);
+
+	return ret;
+}
+
+static int __hpp__color_compare(struct perf_hpp_fmt *fmt,
+				struct perf_hpp *hpp, struct hist_entry *he,
+				int comparison_method)
+{
+	struct diff_hpp_fmt *dfmt =
+		container_of(fmt, struct diff_hpp_fmt, fmt);
+	struct hist_entry *pair = get_pair_fmt(he, dfmt);
+	double diff;
+	s64 wdiff;
+	char pfmt[20] = " ";
+
+	if (!pair)
+		goto no_print;
+
+	switch (comparison_method) {
+	case COMPUTE_DELTA:
+		if (pair->diff.computed)
+			diff = pair->diff.period_ratio_delta;
+		else
+			diff = compute_delta(he, pair);
+
+		scnprintf(pfmt, 20, "%%%+d.2f%%%%", dfmt->header_width - 1);
+		return percent_color_snprintf(hpp->buf, hpp->size,
+					pfmt, diff);
+	case COMPUTE_RATIO:
+		if (he->dummy)
+			goto dummy_print;
+		if (pair->diff.computed)
+			diff = pair->diff.period_ratio;
+		else
+			diff = compute_ratio(he, pair);
+
+		scnprintf(pfmt, 20, "%%%d.6f", dfmt->header_width);
+		return value_color_snprintf(hpp->buf, hpp->size,
+					pfmt, diff);
+	case COMPUTE_WEIGHTED_DIFF:
+		if (he->dummy)
+			goto dummy_print;
+		if (pair->diff.computed)
+			wdiff = pair->diff.wdiff;
+		else
+			wdiff = compute_wdiff(he, pair);
+
+		scnprintf(pfmt, 20, "%%14ld", dfmt->header_width);
+		return color_snprintf(hpp->buf, hpp->size,
+				get_percent_color(wdiff),
+				pfmt, wdiff);
+	default:
+		BUG_ON(1);
+	}
+dummy_print:
+	return scnprintf(hpp->buf, hpp->size, "%*s",
+			dfmt->header_width, "N/A");
+no_print:
+	return scnprintf(hpp->buf, hpp->size, "%*s",
+			dfmt->header_width, pfmt);
+}
+
+static int hpp__color_delta(struct perf_hpp_fmt *fmt,
+			struct perf_hpp *hpp, struct hist_entry *he)
+{
+	return __hpp__color_compare(fmt, hpp, he, COMPUTE_DELTA);
+}
+
+static int hpp__color_ratio(struct perf_hpp_fmt *fmt,
+			struct perf_hpp *hpp, struct hist_entry *he)
+{
+	return __hpp__color_compare(fmt, hpp, he, COMPUTE_RATIO);
+}
+
+static int hpp__color_wdiff(struct perf_hpp_fmt *fmt,
+			struct perf_hpp *hpp, struct hist_entry *he)
+{
+	return __hpp__color_compare(fmt, hpp, he, COMPUTE_WEIGHTED_DIFF);
+}
+
+static void
+hpp__entry_unpair(struct hist_entry *he, int idx, char *buf, size_t size)
+{
+	switch (idx) {
+	case PERF_HPP_DIFF__PERIOD_BASELINE:
+		scnprintf(buf, size, "%" PRIu64, he->stat.period);
+		break;
+
+	default:
+		break;
+	}
+}
+
+static void
+hpp__entry_pair(struct hist_entry *he, struct hist_entry *pair,
+		int idx, char *buf, size_t size)
+{
+	double diff;
+	double ratio;
+	s64 wdiff;
+
+	switch (idx) {
+	case PERF_HPP_DIFF__DELTA:
+		if (pair->diff.computed)
+			diff = pair->diff.period_ratio_delta;
+		else
+			diff = compute_delta(he, pair);
+
+		scnprintf(buf, size, "%+4.2F%%", diff);
+		break;
+
+	case PERF_HPP_DIFF__RATIO:
+		/* No point for ratio number if we are dummy.. */
+		if (he->dummy) {
+			scnprintf(buf, size, "N/A");
+			break;
+		}
+
+		if (pair->diff.computed)
+			ratio = pair->diff.period_ratio;
+		else
+			ratio = compute_ratio(he, pair);
+
+		if (ratio > 0.0)
+			scnprintf(buf, size, "%14.6F", ratio);
+		break;
+
+	case PERF_HPP_DIFF__WEIGHTED_DIFF:
+		/* No point for wdiff number if we are dummy.. */
+		if (he->dummy) {
+			scnprintf(buf, size, "N/A");
+			break;
+		}
+
+		if (pair->diff.computed)
+			wdiff = pair->diff.wdiff;
+		else
+			wdiff = compute_wdiff(he, pair);
+
+		if (wdiff != 0)
+			scnprintf(buf, size, "%14ld", wdiff);
+		break;
+
+	case PERF_HPP_DIFF__FORMULA:
+		formula_fprintf(he, pair, buf, size);
+		break;
+
+	case PERF_HPP_DIFF__PERIOD:
+		scnprintf(buf, size, "%" PRIu64, pair->stat.period);
+		break;
+
+	default:
+		BUG_ON(1);
+	};
+}
+
+static void
+__hpp__entry_global(struct hist_entry *he, struct diff_hpp_fmt *dfmt,
+		    char *buf, size_t size)
+{
+	struct hist_entry *pair = get_pair_fmt(he, dfmt);
+	int idx = dfmt->idx;
+
+	/* baseline is special */
+	if (idx == PERF_HPP_DIFF__BASELINE)
+		hpp__entry_baseline(he, buf, size);
+	else {
+		if (pair)
+			hpp__entry_pair(he, pair, idx, buf, size);
+		else
+			hpp__entry_unpair(he, idx, buf, size);
+	}
+}
+
+static int hpp__entry_global(struct perf_hpp_fmt *_fmt, struct perf_hpp *hpp,
+			     struct hist_entry *he)
+{
+	struct diff_hpp_fmt *dfmt =
+		container_of(_fmt, struct diff_hpp_fmt, fmt);
+	char buf[MAX_COL_WIDTH] = " ";
+
+	__hpp__entry_global(he, dfmt, buf, MAX_COL_WIDTH);
+
+	if (symbol_conf.field_sep)
+		return scnprintf(hpp->buf, hpp->size, "%s", buf);
+	else
+		return scnprintf(hpp->buf, hpp->size, "%*s",
+				 dfmt->header_width, buf);
+}
+
+static int hpp__header(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+		       struct perf_evsel *evsel __maybe_unused)
+{
+	struct diff_hpp_fmt *dfmt =
+		container_of(fmt, struct diff_hpp_fmt, fmt);
+
+	BUG_ON(!dfmt->header);
+	return scnprintf(hpp->buf, hpp->size, dfmt->header);
+}
+
+static int hpp__width(struct perf_hpp_fmt *fmt,
+		      struct perf_hpp *hpp __maybe_unused,
+		      struct perf_evsel *evsel __maybe_unused)
+{
+	struct diff_hpp_fmt *dfmt =
+		container_of(fmt, struct diff_hpp_fmt, fmt);
+
+	BUG_ON(dfmt->header_width <= 0);
+	return dfmt->header_width;
+}
+
+static void init_header(struct data__file *d, struct diff_hpp_fmt *dfmt)
+{
+#define MAX_HEADER_NAME 100
+	char buf_indent[MAX_HEADER_NAME];
+	char buf[MAX_HEADER_NAME];
+	const char *header = NULL;
+	int width = 0;
+
+	BUG_ON(dfmt->idx >= PERF_HPP_DIFF__MAX_INDEX);
+	header = columns[dfmt->idx].name;
+	width  = columns[dfmt->idx].width;
+
+	/* Only our defined HPP fmts should appear here. */
+	BUG_ON(!header);
+
+	if (data__files_cnt > 2)
+		scnprintf(buf, MAX_HEADER_NAME, "%s/%d", header, d->idx);
+
+#define NAME (data__files_cnt > 2 ? buf : header)
+	dfmt->header_width = width;
+	width = (int) strlen(NAME);
+	if (dfmt->header_width < width)
+		dfmt->header_width = width;
+
+	scnprintf(buf_indent, MAX_HEADER_NAME, "%*s",
+		  dfmt->header_width, NAME);
+
+	dfmt->header = strdup(buf_indent);
+#undef MAX_HEADER_NAME
+#undef NAME
+}
+
+static void data__hpp_register(struct data__file *d, int idx)
+{
+	struct diff_hpp_fmt *dfmt = &d->fmt[idx];
+	struct perf_hpp_fmt *fmt = &dfmt->fmt;
+
+	dfmt->idx = idx;
+
+	fmt->header = hpp__header;
+	fmt->width  = hpp__width;
+	fmt->entry  = hpp__entry_global;
+	fmt->cmp    = hist_entry__cmp_nop;
+	fmt->collapse = hist_entry__cmp_nop;
+
+	/* TODO more colors */
+	switch (idx) {
+	case PERF_HPP_DIFF__BASELINE:
+		fmt->color = hpp__color_baseline;
+		fmt->sort  = hist_entry__cmp_baseline;
+		break;
+	case PERF_HPP_DIFF__DELTA:
+		fmt->color = hpp__color_delta;
+		fmt->sort  = hist_entry__cmp_delta;
+		break;
+	case PERF_HPP_DIFF__RATIO:
+		fmt->color = hpp__color_ratio;
+		fmt->sort  = hist_entry__cmp_ratio;
+		break;
+	case PERF_HPP_DIFF__WEIGHTED_DIFF:
+		fmt->color = hpp__color_wdiff;
+		fmt->sort  = hist_entry__cmp_wdiff;
+		break;
+	default:
+		fmt->sort  = hist_entry__cmp_nop;
+		break;
+	}
+
+	init_header(d, dfmt);
+	perf_hpp__column_register(fmt);
+	perf_hpp__register_sort_field(fmt);
+}
+
+static int ui_init(void)
+{
+	struct data__file *d;
+	struct perf_hpp_fmt *fmt;
+	int i;
+
+	data__for_each_file(i, d) {
+
+		/*
+		 * Baseline or compute realted columns:
+		 *
+		 *   PERF_HPP_DIFF__BASELINE
+		 *   PERF_HPP_DIFF__DELTA
+		 *   PERF_HPP_DIFF__RATIO
+		 *   PERF_HPP_DIFF__WEIGHTED_DIFF
+		 */
+		data__hpp_register(d, i ? compute_2_hpp[compute] :
+					  PERF_HPP_DIFF__BASELINE);
+
+		/*
+		 * And the rest:
+		 *
+		 * PERF_HPP_DIFF__FORMULA
+		 * PERF_HPP_DIFF__PERIOD
+		 * PERF_HPP_DIFF__PERIOD_BASELINE
+		 */
+		if (show_formula && i)
+			data__hpp_register(d, PERF_HPP_DIFF__FORMULA);
+
+		if (show_period)
+			data__hpp_register(d, i ? PERF_HPP_DIFF__PERIOD :
+						  PERF_HPP_DIFF__PERIOD_BASELINE);
+	}
+
+	if (!sort_compute)
+		return 0;
+
+	/*
+	 * Prepend an fmt to sort on columns at 'sort_compute' first.
+	 * This fmt is added only to the sort list but not to the
+	 * output fields list.
+	 *
+	 * Note that this column (data) can be compared twice - one
+	 * for this 'sort_compute' fmt and another for the normal
+	 * diff_hpp_fmt.  But it shouldn't a problem as most entries
+	 * will be sorted out by first try or baseline and comparing
+	 * is not a costly operation.
+	 */
+	fmt = zalloc(sizeof(*fmt));
+	if (fmt == NULL) {
+		pr_err("Memory allocation failed\n");
+		return -1;
+	}
+
+	fmt->cmp      = hist_entry__cmp_nop;
+	fmt->collapse = hist_entry__cmp_nop;
+
+	switch (compute) {
+	case COMPUTE_DELTA:
+		fmt->sort = hist_entry__cmp_delta_idx;
+		break;
+	case COMPUTE_RATIO:
+		fmt->sort = hist_entry__cmp_ratio_idx;
+		break;
+	case COMPUTE_WEIGHTED_DIFF:
+		fmt->sort = hist_entry__cmp_wdiff_idx;
+		break;
+	default:
+		BUG_ON(1);
+	}
+
+	list_add(&fmt->sort_list, &perf_hpp__sort_list);
+	return 0;
+}
+
+static int data_init(int argc, const char **argv)
+{
+	struct data__file *d;
+	static const char *defaults[] = {
+		"perf.data.old",
+		"perf.data",
+	};
+	bool use_default = true;
+	int i;
+
+	data__files_cnt = 2;
+
+	if (argc) {
+		if (argc == 1)
+			defaults[1] = argv[0];
+		else {
+			data__files_cnt = argc;
+			use_default = false;
+		}
+	} else if (perf_guest) {
+		defaults[0] = "perf.data.host";
+		defaults[1] = "perf.data.guest";
+	}
+
+	if (sort_compute >= (unsigned int) data__files_cnt) {
+		pr_err("Order option out of limit.\n");
+		return -EINVAL;
+	}
+
+	data__files = zalloc(sizeof(*data__files) * data__files_cnt);
+	if (!data__files)
+		return -ENOMEM;
+
+	data__for_each_file(i, d) {
+		struct perf_data_file *file = &d->file;
+
+		file->path  = use_default ? defaults[i] : argv[i];
+		file->mode  = PERF_DATA_MODE_READ,
+		file->force = force,
+
+		d->idx  = i;
+	}
+
+	return 0;
+}
+
+int cmd_diff(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	int ret = hists__init();
+
+	if (ret < 0)
+		return ret;
+
+	perf_config(perf_default_config, NULL);
+
+	argc = parse_options(argc, argv, options, diff_usage, 0);
+
+	if (symbol__init(NULL) < 0)
+		return -1;
+
+	if (data_init(argc, argv) < 0)
+		return -1;
+
+	if (ui_init() < 0)
+		return -1;
+
+	sort__mode = SORT_MODE__DIFF;
+
+	if (setup_sorting() < 0)
+		usage_with_options(diff_usage, options);
+
+	setup_pager();
+
+	sort__setup_elide(NULL);
+
+	return __cmd_diff();
+}
diff --git a/tools/perf/builtin-evlist.c b/tools/perf/builtin-evlist.c
new file mode 100644
index 000000000..695ec5a50
--- /dev/null
+++ b/tools/perf/builtin-evlist.c
@@ -0,0 +1,69 @@
+/*
+ * Builtin evlist command: Show the list of event selectors present
+ * in a perf.data file.
+ */
+#include "builtin.h"
+
+#include "util/util.h"
+
+#include <linux/list.h>
+
+#include "perf.h"
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/parse-events.h"
+#include "util/parse-options.h"
+#include "util/session.h"
+#include "util/data.h"
+#include "util/debug.h"
+
+static int __cmd_evlist(const char *file_name, struct perf_attr_details *details)
+{
+	struct perf_session *session;
+	struct perf_evsel *pos;
+	struct perf_data_file file = {
+		.path = file_name,
+		.mode = PERF_DATA_MODE_READ,
+		.force = details->force,
+	};
+
+	session = perf_session__new(&file, 0, NULL);
+	if (session == NULL)
+		return -1;
+
+	evlist__for_each(session->evlist, pos)
+		perf_evsel__fprintf(pos, details, stdout);
+
+	perf_session__delete(session);
+	return 0;
+}
+
+int cmd_evlist(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	struct perf_attr_details details = { .verbose = false, };
+	const struct option options[] = {
+	OPT_STRING('i', "input", &input_name, "file", "Input file name"),
+	OPT_BOOLEAN('F', "freq", &details.freq, "Show the sample frequency"),
+	OPT_BOOLEAN('v', "verbose", &details.verbose,
+		    "Show all event attr details"),
+	OPT_BOOLEAN('g', "group", &details.event_group,
+		    "Show event group information"),
+	OPT_BOOLEAN('f', "force", &details.force, "don't complain, do it"),
+	OPT_END()
+	};
+	const char * const evlist_usage[] = {
+		"perf evlist [<options>]",
+		NULL
+	};
+
+	argc = parse_options(argc, argv, options, evlist_usage, 0);
+	if (argc)
+		usage_with_options(evlist_usage, options);
+
+	if (details.event_group && (details.verbose || details.freq)) {
+		pr_err("--group option is not compatible with other options\n");
+		usage_with_options(evlist_usage, options);
+	}
+
+	return __cmd_evlist(input_name, &details);
+}
diff --git a/tools/perf/builtin-help.c b/tools/perf/builtin-help.c
new file mode 100644
index 000000000..36486eade
--- /dev/null
+++ b/tools/perf/builtin-help.c
@@ -0,0 +1,503 @@
+/*
+ * builtin-help.c
+ *
+ * Builtin help command
+ */
+#include "perf.h"
+#include "util/cache.h"
+#include "builtin.h"
+#include "util/exec_cmd.h"
+#include "common-cmds.h"
+#include "util/parse-options.h"
+#include "util/run-command.h"
+#include "util/help.h"
+#include "util/debug.h"
+
+static struct man_viewer_list {
+	struct man_viewer_list *next;
+	char name[FLEX_ARRAY];
+} *man_viewer_list;
+
+static struct man_viewer_info_list {
+	struct man_viewer_info_list *next;
+	const char *info;
+	char name[FLEX_ARRAY];
+} *man_viewer_info_list;
+
+enum help_format {
+	HELP_FORMAT_NONE,
+	HELP_FORMAT_MAN,
+	HELP_FORMAT_INFO,
+	HELP_FORMAT_WEB,
+};
+
+static enum help_format parse_help_format(const char *format)
+{
+	if (!strcmp(format, "man"))
+		return HELP_FORMAT_MAN;
+	if (!strcmp(format, "info"))
+		return HELP_FORMAT_INFO;
+	if (!strcmp(format, "web") || !strcmp(format, "html"))
+		return HELP_FORMAT_WEB;
+
+	pr_err("unrecognized help format '%s'", format);
+	return HELP_FORMAT_NONE;
+}
+
+static const char *get_man_viewer_info(const char *name)
+{
+	struct man_viewer_info_list *viewer;
+
+	for (viewer = man_viewer_info_list; viewer; viewer = viewer->next) {
+		if (!strcasecmp(name, viewer->name))
+			return viewer->info;
+	}
+	return NULL;
+}
+
+static int check_emacsclient_version(void)
+{
+	struct strbuf buffer = STRBUF_INIT;
+	struct child_process ec_process;
+	const char *argv_ec[] = { "emacsclient", "--version", NULL };
+	int version;
+
+	/* emacsclient prints its version number on stderr */
+	memset(&ec_process, 0, sizeof(ec_process));
+	ec_process.argv = argv_ec;
+	ec_process.err = -1;
+	ec_process.stdout_to_stderr = 1;
+	if (start_command(&ec_process)) {
+		fprintf(stderr, "Failed to start emacsclient.\n");
+		return -1;
+	}
+	strbuf_read(&buffer, ec_process.err, 20);
+	close(ec_process.err);
+
+	/*
+	 * Don't bother checking return value, because "emacsclient --version"
+	 * seems to always exits with code 1.
+	 */
+	finish_command(&ec_process);
+
+	if (prefixcmp(buffer.buf, "emacsclient")) {
+		fprintf(stderr, "Failed to parse emacsclient version.\n");
+		strbuf_release(&buffer);
+		return -1;
+	}
+
+	strbuf_remove(&buffer, 0, strlen("emacsclient"));
+	version = atoi(buffer.buf);
+
+	if (version < 22) {
+		fprintf(stderr,
+			"emacsclient version '%d' too old (< 22).\n",
+			version);
+		strbuf_release(&buffer);
+		return -1;
+	}
+
+	strbuf_release(&buffer);
+	return 0;
+}
+
+static void exec_woman_emacs(const char *path, const char *page)
+{
+	char sbuf[STRERR_BUFSIZE];
+
+	if (!check_emacsclient_version()) {
+		/* This works only with emacsclient version >= 22. */
+		struct strbuf man_page = STRBUF_INIT;
+
+		if (!path)
+			path = "emacsclient";
+		strbuf_addf(&man_page, "(woman \"%s\")", page);
+		execlp(path, "emacsclient", "-e", man_page.buf, NULL);
+		warning("failed to exec '%s': %s", path,
+			strerror_r(errno, sbuf, sizeof(sbuf)));
+	}
+}
+
+static void exec_man_konqueror(const char *path, const char *page)
+{
+	const char *display = getenv("DISPLAY");
+
+	if (display && *display) {
+		struct strbuf man_page = STRBUF_INIT;
+		const char *filename = "kfmclient";
+		char sbuf[STRERR_BUFSIZE];
+
+		/* It's simpler to launch konqueror using kfmclient. */
+		if (path) {
+			const char *file = strrchr(path, '/');
+			if (file && !strcmp(file + 1, "konqueror")) {
+				char *new = strdup(path);
+				char *dest = strrchr(new, '/');
+
+				/* strlen("konqueror") == strlen("kfmclient") */
+				strcpy(dest + 1, "kfmclient");
+				path = new;
+			}
+			if (file)
+				filename = file;
+		} else
+			path = "kfmclient";
+		strbuf_addf(&man_page, "man:%s(1)", page);
+		execlp(path, filename, "newTab", man_page.buf, NULL);
+		warning("failed to exec '%s': %s", path,
+			strerror_r(errno, sbuf, sizeof(sbuf)));
+	}
+}
+
+static void exec_man_man(const char *path, const char *page)
+{
+	char sbuf[STRERR_BUFSIZE];
+
+	if (!path)
+		path = "man";
+	execlp(path, "man", page, NULL);
+	warning("failed to exec '%s': %s", path,
+		strerror_r(errno, sbuf, sizeof(sbuf)));
+}
+
+static void exec_man_cmd(const char *cmd, const char *page)
+{
+	struct strbuf shell_cmd = STRBUF_INIT;
+	char sbuf[STRERR_BUFSIZE];
+
+	strbuf_addf(&shell_cmd, "%s %s", cmd, page);
+	execl("/bin/sh", "sh", "-c", shell_cmd.buf, NULL);
+	warning("failed to exec '%s': %s", cmd,
+		strerror_r(errno, sbuf, sizeof(sbuf)));
+}
+
+static void add_man_viewer(const char *name)
+{
+	struct man_viewer_list **p = &man_viewer_list;
+	size_t len = strlen(name);
+
+	while (*p)
+		p = &((*p)->next);
+	*p = zalloc(sizeof(**p) + len + 1);
+	strncpy((*p)->name, name, len);
+}
+
+static int supported_man_viewer(const char *name, size_t len)
+{
+	return (!strncasecmp("man", name, len) ||
+		!strncasecmp("woman", name, len) ||
+		!strncasecmp("konqueror", name, len));
+}
+
+static void do_add_man_viewer_info(const char *name,
+				   size_t len,
+				   const char *value)
+{
+	struct man_viewer_info_list *new = zalloc(sizeof(*new) + len + 1);
+
+	strncpy(new->name, name, len);
+	new->info = strdup(value);
+	new->next = man_viewer_info_list;
+	man_viewer_info_list = new;
+}
+
+static int add_man_viewer_path(const char *name,
+			       size_t len,
+			       const char *value)
+{
+	if (supported_man_viewer(name, len))
+		do_add_man_viewer_info(name, len, value);
+	else
+		warning("'%s': path for unsupported man viewer.\n"
+			"Please consider using 'man.<tool>.cmd' instead.",
+			name);
+
+	return 0;
+}
+
+static int add_man_viewer_cmd(const char *name,
+			      size_t len,
+			      const char *value)
+{
+	if (supported_man_viewer(name, len))
+		warning("'%s': cmd for supported man viewer.\n"
+			"Please consider using 'man.<tool>.path' instead.",
+			name);
+	else
+		do_add_man_viewer_info(name, len, value);
+
+	return 0;
+}
+
+static int add_man_viewer_info(const char *var, const char *value)
+{
+	const char *name = var + 4;
+	const char *subkey = strrchr(name, '.');
+
+	if (!subkey)
+		return error("Config with no key for man viewer: %s", name);
+
+	if (!strcmp(subkey, ".path")) {
+		if (!value)
+			return config_error_nonbool(var);
+		return add_man_viewer_path(name, subkey - name, value);
+	}
+	if (!strcmp(subkey, ".cmd")) {
+		if (!value)
+			return config_error_nonbool(var);
+		return add_man_viewer_cmd(name, subkey - name, value);
+	}
+
+	warning("'%s': unsupported man viewer sub key.", subkey);
+	return 0;
+}
+
+static int perf_help_config(const char *var, const char *value, void *cb)
+{
+	enum help_format *help_formatp = cb;
+
+	if (!strcmp(var, "help.format")) {
+		if (!value)
+			return config_error_nonbool(var);
+		*help_formatp = parse_help_format(value);
+		if (*help_formatp == HELP_FORMAT_NONE)
+			return -1;
+		return 0;
+	}
+	if (!strcmp(var, "man.viewer")) {
+		if (!value)
+			return config_error_nonbool(var);
+		add_man_viewer(value);
+		return 0;
+	}
+	if (!prefixcmp(var, "man."))
+		return add_man_viewer_info(var, value);
+
+	return perf_default_config(var, value, cb);
+}
+
+static struct cmdnames main_cmds, other_cmds;
+
+void list_common_cmds_help(void)
+{
+	unsigned int i, longest = 0;
+
+	for (i = 0; i < ARRAY_SIZE(common_cmds); i++) {
+		if (longest < strlen(common_cmds[i].name))
+			longest = strlen(common_cmds[i].name);
+	}
+
+	puts(" The most commonly used perf commands are:");
+	for (i = 0; i < ARRAY_SIZE(common_cmds); i++) {
+		printf("   %-*s   ", longest, common_cmds[i].name);
+		puts(common_cmds[i].help);
+	}
+}
+
+static int is_perf_command(const char *s)
+{
+	return is_in_cmdlist(&main_cmds, s) ||
+		is_in_cmdlist(&other_cmds, s);
+}
+
+static const char *prepend(const char *prefix, const char *cmd)
+{
+	size_t pre_len = strlen(prefix);
+	size_t cmd_len = strlen(cmd);
+	char *p = malloc(pre_len + cmd_len + 1);
+	memcpy(p, prefix, pre_len);
+	strcpy(p + pre_len, cmd);
+	return p;
+}
+
+static const char *cmd_to_page(const char *perf_cmd)
+{
+	if (!perf_cmd)
+		return "perf";
+	else if (!prefixcmp(perf_cmd, "perf"))
+		return perf_cmd;
+	else
+		return prepend("perf-", perf_cmd);
+}
+
+static void setup_man_path(void)
+{
+	struct strbuf new_path = STRBUF_INIT;
+	const char *old_path = getenv("MANPATH");
+
+	/* We should always put ':' after our path. If there is no
+	 * old_path, the ':' at the end will let 'man' to try
+	 * system-wide paths after ours to find the manual page. If
+	 * there is old_path, we need ':' as delimiter. */
+	strbuf_addstr(&new_path, system_path(PERF_MAN_PATH));
+	strbuf_addch(&new_path, ':');
+	if (old_path)
+		strbuf_addstr(&new_path, old_path);
+
+	setenv("MANPATH", new_path.buf, 1);
+
+	strbuf_release(&new_path);
+}
+
+static void exec_viewer(const char *name, const char *page)
+{
+	const char *info = get_man_viewer_info(name);
+
+	if (!strcasecmp(name, "man"))
+		exec_man_man(info, page);
+	else if (!strcasecmp(name, "woman"))
+		exec_woman_emacs(info, page);
+	else if (!strcasecmp(name, "konqueror"))
+		exec_man_konqueror(info, page);
+	else if (info)
+		exec_man_cmd(info, page);
+	else
+		warning("'%s': unknown man viewer.", name);
+}
+
+static int show_man_page(const char *perf_cmd)
+{
+	struct man_viewer_list *viewer;
+	const char *page = cmd_to_page(perf_cmd);
+	const char *fallback = getenv("PERF_MAN_VIEWER");
+
+	setup_man_path();
+	for (viewer = man_viewer_list; viewer; viewer = viewer->next)
+		exec_viewer(viewer->name, page); /* will return when unable */
+
+	if (fallback)
+		exec_viewer(fallback, page);
+	exec_viewer("man", page);
+
+	pr_err("no man viewer handled the request");
+	return -1;
+}
+
+static int show_info_page(const char *perf_cmd)
+{
+	const char *page = cmd_to_page(perf_cmd);
+	setenv("INFOPATH", system_path(PERF_INFO_PATH), 1);
+	execlp("info", "info", "perfman", page, NULL);
+	return -1;
+}
+
+static int get_html_page_path(struct strbuf *page_path, const char *page)
+{
+	struct stat st;
+	const char *html_path = system_path(PERF_HTML_PATH);
+
+	/* Check that we have a perf documentation directory. */
+	if (stat(mkpath("%s/perf.html", html_path), &st)
+	    || !S_ISREG(st.st_mode)) {
+		pr_err("'%s': not a documentation directory.", html_path);
+		return -1;
+	}
+
+	strbuf_init(page_path, 0);
+	strbuf_addf(page_path, "%s/%s.html", html_path, page);
+
+	return 0;
+}
+
+/*
+ * If open_html is not defined in a platform-specific way (see for
+ * example compat/mingw.h), we use the script web--browse to display
+ * HTML.
+ */
+#ifndef open_html
+static void open_html(const char *path)
+{
+	execl_perf_cmd("web--browse", "-c", "help.browser", path, NULL);
+}
+#endif
+
+static int show_html_page(const char *perf_cmd)
+{
+	const char *page = cmd_to_page(perf_cmd);
+	struct strbuf page_path; /* it leaks but we exec bellow */
+
+	if (get_html_page_path(&page_path, page) != 0)
+		return -1;
+
+	open_html(page_path.buf);
+
+	return 0;
+}
+
+int cmd_help(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	bool show_all = false;
+	enum help_format help_format = HELP_FORMAT_MAN;
+	struct option builtin_help_options[] = {
+	OPT_BOOLEAN('a', "all", &show_all, "print all available commands"),
+	OPT_SET_UINT('m', "man", &help_format, "show man page", HELP_FORMAT_MAN),
+	OPT_SET_UINT('w', "web", &help_format, "show manual in web browser",
+			HELP_FORMAT_WEB),
+	OPT_SET_UINT('i', "info", &help_format, "show info page",
+			HELP_FORMAT_INFO),
+	OPT_END(),
+	};
+	const char * const builtin_help_subcommands[] = {
+		"buildid-cache", "buildid-list", "diff", "evlist", "help", "list",
+		"record", "report", "bench", "stat", "timechart", "top", "annotate",
+		"script", "sched", "kmem", "lock", "kvm", "test", "inject", "mem", "data",
+#ifdef HAVE_LIBELF_SUPPORT
+		"probe",
+#endif
+#ifdef HAVE_LIBAUDIT_SUPPORT
+		"trace",
+#endif
+	NULL };
+	const char *builtin_help_usage[] = {
+		"perf help [--all] [--man|--web|--info] [command]",
+		NULL
+	};
+	const char *alias;
+	int rc = 0;
+
+	load_command_list("perf-", &main_cmds, &other_cmds);
+
+	perf_config(perf_help_config, &help_format);
+
+	argc = parse_options_subcommand(argc, argv, builtin_help_options,
+			builtin_help_subcommands, builtin_help_usage, 0);
+
+	if (show_all) {
+		printf("\n usage: %s\n\n", perf_usage_string);
+		list_commands("perf commands", &main_cmds, &other_cmds);
+		printf(" %s\n\n", perf_more_info_string);
+		return 0;
+	}
+
+	if (!argv[0]) {
+		printf("\n usage: %s\n\n", perf_usage_string);
+		list_common_cmds_help();
+		printf("\n %s\n\n", perf_more_info_string);
+		return 0;
+	}
+
+	alias = alias_lookup(argv[0]);
+	if (alias && !is_perf_command(argv[0])) {
+		printf("`perf %s' is aliased to `%s'\n", argv[0], alias);
+		return 0;
+	}
+
+	switch (help_format) {
+	case HELP_FORMAT_MAN:
+		rc = show_man_page(argv[0]);
+		break;
+	case HELP_FORMAT_INFO:
+		rc = show_info_page(argv[0]);
+		break;
+	case HELP_FORMAT_WEB:
+		rc = show_html_page(argv[0]);
+		break;
+	case HELP_FORMAT_NONE:
+		/* fall-through */
+	default:
+		rc = -1;
+		break;
+	}
+
+	return rc;
+}
diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c
new file mode 100644
index 000000000..40a33d733
--- /dev/null
+++ b/tools/perf/builtin-inject.c
@@ -0,0 +1,482 @@
+/*
+ * builtin-inject.c
+ *
+ * Builtin inject command: Examine the live mode (stdin) event stream
+ * and repipe it to stdout while optionally injecting additional
+ * events into it.
+ */
+#include "builtin.h"
+
+#include "perf.h"
+#include "util/color.h"
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/session.h"
+#include "util/tool.h"
+#include "util/debug.h"
+#include "util/build-id.h"
+#include "util/data.h"
+
+#include "util/parse-options.h"
+
+#include <linux/list.h>
+
+struct perf_inject {
+	struct perf_tool	tool;
+	struct perf_session	*session;
+	bool			build_ids;
+	bool			sched_stat;
+	const char		*input_name;
+	struct perf_data_file	output;
+	u64			bytes_written;
+	struct list_head	samples;
+};
+
+struct event_entry {
+	struct list_head node;
+	u32		 tid;
+	union perf_event event[0];
+};
+
+static int perf_event__repipe_synth(struct perf_tool *tool,
+				    union perf_event *event)
+{
+	struct perf_inject *inject = container_of(tool, struct perf_inject, tool);
+	ssize_t size;
+
+	size = perf_data_file__write(&inject->output, event,
+				     event->header.size);
+	if (size < 0)
+		return -errno;
+
+	inject->bytes_written += size;
+	return 0;
+}
+
+static int perf_event__repipe_oe_synth(struct perf_tool *tool,
+				       union perf_event *event,
+				       struct ordered_events *oe __maybe_unused)
+{
+	return perf_event__repipe_synth(tool, event);
+}
+
+static int perf_event__repipe_op2_synth(struct perf_tool *tool,
+					union perf_event *event,
+					struct perf_session *session
+					__maybe_unused)
+{
+	return perf_event__repipe_synth(tool, event);
+}
+
+static int perf_event__repipe_attr(struct perf_tool *tool,
+				   union perf_event *event,
+				   struct perf_evlist **pevlist)
+{
+	struct perf_inject *inject = container_of(tool, struct perf_inject,
+						  tool);
+	int ret;
+
+	ret = perf_event__process_attr(tool, event, pevlist);
+	if (ret)
+		return ret;
+
+	if (!inject->output.is_pipe)
+		return 0;
+
+	return perf_event__repipe_synth(tool, event);
+}
+
+static int perf_event__repipe(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample __maybe_unused,
+			      struct machine *machine __maybe_unused)
+{
+	return perf_event__repipe_synth(tool, event);
+}
+
+typedef int (*inject_handler)(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct perf_evsel *evsel,
+			      struct machine *machine);
+
+static int perf_event__repipe_sample(struct perf_tool *tool,
+				     union perf_event *event,
+				     struct perf_sample *sample,
+				     struct perf_evsel *evsel,
+				     struct machine *machine)
+{
+	if (evsel->handler) {
+		inject_handler f = evsel->handler;
+		return f(tool, event, sample, evsel, machine);
+	}
+
+	build_id__mark_dso_hit(tool, event, sample, evsel, machine);
+
+	return perf_event__repipe_synth(tool, event);
+}
+
+static int perf_event__repipe_mmap(struct perf_tool *tool,
+				   union perf_event *event,
+				   struct perf_sample *sample,
+				   struct machine *machine)
+{
+	int err;
+
+	err = perf_event__process_mmap(tool, event, sample, machine);
+	perf_event__repipe(tool, event, sample, machine);
+
+	return err;
+}
+
+static int perf_event__repipe_mmap2(struct perf_tool *tool,
+				   union perf_event *event,
+				   struct perf_sample *sample,
+				   struct machine *machine)
+{
+	int err;
+
+	err = perf_event__process_mmap2(tool, event, sample, machine);
+	perf_event__repipe(tool, event, sample, machine);
+
+	return err;
+}
+
+static int perf_event__repipe_fork(struct perf_tool *tool,
+				   union perf_event *event,
+				   struct perf_sample *sample,
+				   struct machine *machine)
+{
+	int err;
+
+	err = perf_event__process_fork(tool, event, sample, machine);
+	perf_event__repipe(tool, event, sample, machine);
+
+	return err;
+}
+
+static int perf_event__repipe_tracing_data(struct perf_tool *tool,
+					   union perf_event *event,
+					   struct perf_session *session)
+{
+	int err;
+
+	perf_event__repipe_synth(tool, event);
+	err = perf_event__process_tracing_data(tool, event, session);
+
+	return err;
+}
+
+static int dso__read_build_id(struct dso *dso)
+{
+	if (dso->has_build_id)
+		return 0;
+
+	if (filename__read_build_id(dso->long_name, dso->build_id,
+				    sizeof(dso->build_id)) > 0) {
+		dso->has_build_id = true;
+		return 0;
+	}
+
+	return -1;
+}
+
+static int dso__inject_build_id(struct dso *dso, struct perf_tool *tool,
+				struct machine *machine)
+{
+	u16 misc = PERF_RECORD_MISC_USER;
+	int err;
+
+	if (dso__read_build_id(dso) < 0) {
+		pr_debug("no build_id found for %s\n", dso->long_name);
+		return -1;
+	}
+
+	if (dso->kernel)
+		misc = PERF_RECORD_MISC_KERNEL;
+
+	err = perf_event__synthesize_build_id(tool, dso, misc, perf_event__repipe,
+					      machine);
+	if (err) {
+		pr_err("Can't synthesize build_id event for %s\n", dso->long_name);
+		return -1;
+	}
+
+	return 0;
+}
+
+static int perf_event__inject_buildid(struct perf_tool *tool,
+				      union perf_event *event,
+				      struct perf_sample *sample,
+				      struct perf_evsel *evsel __maybe_unused,
+				      struct machine *machine)
+{
+	struct addr_location al;
+	struct thread *thread;
+	u8 cpumode;
+
+	cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+
+	thread = machine__findnew_thread(machine, sample->pid, sample->tid);
+	if (thread == NULL) {
+		pr_err("problem processing %d event, skipping it.\n",
+		       event->header.type);
+		goto repipe;
+	}
+
+	thread__find_addr_map(thread, cpumode, MAP__FUNCTION, sample->ip, &al);
+
+	if (al.map != NULL) {
+		if (!al.map->dso->hit) {
+			al.map->dso->hit = 1;
+			if (map__load(al.map, NULL) >= 0) {
+				dso__inject_build_id(al.map->dso, tool, machine);
+				/*
+				 * If this fails, too bad, let the other side
+				 * account this as unresolved.
+				 */
+			} else {
+#ifdef HAVE_LIBELF_SUPPORT
+				pr_warning("no symbols found in %s, maybe "
+					   "install a debug package?\n",
+					   al.map->dso->long_name);
+#endif
+			}
+		}
+	}
+
+repipe:
+	perf_event__repipe(tool, event, sample, machine);
+	return 0;
+}
+
+static int perf_inject__sched_process_exit(struct perf_tool *tool,
+					   union perf_event *event __maybe_unused,
+					   struct perf_sample *sample,
+					   struct perf_evsel *evsel __maybe_unused,
+					   struct machine *machine __maybe_unused)
+{
+	struct perf_inject *inject = container_of(tool, struct perf_inject, tool);
+	struct event_entry *ent;
+
+	list_for_each_entry(ent, &inject->samples, node) {
+		if (sample->tid == ent->tid) {
+			list_del_init(&ent->node);
+			free(ent);
+			break;
+		}
+	}
+
+	return 0;
+}
+
+static int perf_inject__sched_switch(struct perf_tool *tool,
+				     union perf_event *event,
+				     struct perf_sample *sample,
+				     struct perf_evsel *evsel,
+				     struct machine *machine)
+{
+	struct perf_inject *inject = container_of(tool, struct perf_inject, tool);
+	struct event_entry *ent;
+
+	perf_inject__sched_process_exit(tool, event, sample, evsel, machine);
+
+	ent = malloc(event->header.size + sizeof(struct event_entry));
+	if (ent == NULL) {
+		color_fprintf(stderr, PERF_COLOR_RED,
+			     "Not enough memory to process sched switch event!");
+		return -1;
+	}
+
+	ent->tid = sample->tid;
+	memcpy(&ent->event, event, event->header.size);
+	list_add(&ent->node, &inject->samples);
+	return 0;
+}
+
+static int perf_inject__sched_stat(struct perf_tool *tool,
+				   union perf_event *event __maybe_unused,
+				   struct perf_sample *sample,
+				   struct perf_evsel *evsel,
+				   struct machine *machine)
+{
+	struct event_entry *ent;
+	union perf_event *event_sw;
+	struct perf_sample sample_sw;
+	struct perf_inject *inject = container_of(tool, struct perf_inject, tool);
+	u32 pid = perf_evsel__intval(evsel, sample, "pid");
+
+	list_for_each_entry(ent, &inject->samples, node) {
+		if (pid == ent->tid)
+			goto found;
+	}
+
+	return 0;
+found:
+	event_sw = &ent->event[0];
+	perf_evsel__parse_sample(evsel, event_sw, &sample_sw);
+
+	sample_sw.period = sample->period;
+	sample_sw.time	 = sample->time;
+	perf_event__synthesize_sample(event_sw, evsel->attr.sample_type,
+				      evsel->attr.read_format, &sample_sw,
+				      false);
+	build_id__mark_dso_hit(tool, event_sw, &sample_sw, evsel, machine);
+	return perf_event__repipe(tool, event_sw, &sample_sw, machine);
+}
+
+static void sig_handler(int sig __maybe_unused)
+{
+	session_done = 1;
+}
+
+static int perf_evsel__check_stype(struct perf_evsel *evsel,
+				   u64 sample_type, const char *sample_msg)
+{
+	struct perf_event_attr *attr = &evsel->attr;
+	const char *name = perf_evsel__name(evsel);
+
+	if (!(attr->sample_type & sample_type)) {
+		pr_err("Samples for %s event do not have %s attribute set.",
+			name, sample_msg);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int __cmd_inject(struct perf_inject *inject)
+{
+	int ret = -EINVAL;
+	struct perf_session *session = inject->session;
+	struct perf_data_file *file_out = &inject->output;
+	int fd = perf_data_file__fd(file_out);
+
+	signal(SIGINT, sig_handler);
+
+	if (inject->build_ids || inject->sched_stat) {
+		inject->tool.mmap	  = perf_event__repipe_mmap;
+		inject->tool.mmap2	  = perf_event__repipe_mmap2;
+		inject->tool.fork	  = perf_event__repipe_fork;
+		inject->tool.tracing_data = perf_event__repipe_tracing_data;
+	}
+
+	if (inject->build_ids) {
+		inject->tool.sample = perf_event__inject_buildid;
+	} else if (inject->sched_stat) {
+		struct perf_evsel *evsel;
+
+		evlist__for_each(session->evlist, evsel) {
+			const char *name = perf_evsel__name(evsel);
+
+			if (!strcmp(name, "sched:sched_switch")) {
+				if (perf_evsel__check_stype(evsel, PERF_SAMPLE_TID, "TID"))
+					return -EINVAL;
+
+				evsel->handler = perf_inject__sched_switch;
+			} else if (!strcmp(name, "sched:sched_process_exit"))
+				evsel->handler = perf_inject__sched_process_exit;
+			else if (!strncmp(name, "sched:sched_stat_", 17))
+				evsel->handler = perf_inject__sched_stat;
+		}
+	}
+
+	if (!file_out->is_pipe)
+		lseek(fd, session->header.data_offset, SEEK_SET);
+
+	ret = perf_session__process_events(session);
+
+	if (!file_out->is_pipe) {
+		if (inject->build_ids)
+			perf_header__set_feat(&session->header,
+					      HEADER_BUILD_ID);
+		session->header.data_size = inject->bytes_written;
+		perf_session__write_header(session, session->evlist, fd, true);
+	}
+
+	return ret;
+}
+
+int cmd_inject(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	struct perf_inject inject = {
+		.tool = {
+			.sample		= perf_event__repipe_sample,
+			.mmap		= perf_event__repipe,
+			.mmap2		= perf_event__repipe,
+			.comm		= perf_event__repipe,
+			.fork		= perf_event__repipe,
+			.exit		= perf_event__repipe,
+			.lost		= perf_event__repipe,
+			.read		= perf_event__repipe_sample,
+			.throttle	= perf_event__repipe,
+			.unthrottle	= perf_event__repipe,
+			.attr		= perf_event__repipe_attr,
+			.tracing_data	= perf_event__repipe_op2_synth,
+			.finished_round	= perf_event__repipe_oe_synth,
+			.build_id	= perf_event__repipe_op2_synth,
+			.id_index	= perf_event__repipe_op2_synth,
+		},
+		.input_name  = "-",
+		.samples = LIST_HEAD_INIT(inject.samples),
+		.output = {
+			.path = "-",
+			.mode = PERF_DATA_MODE_WRITE,
+		},
+	};
+	struct perf_data_file file = {
+		.mode = PERF_DATA_MODE_READ,
+	};
+	int ret;
+
+	const struct option options[] = {
+		OPT_BOOLEAN('b', "build-ids", &inject.build_ids,
+			    "Inject build-ids into the output stream"),
+		OPT_STRING('i', "input", &inject.input_name, "file",
+			   "input file name"),
+		OPT_STRING('o', "output", &inject.output.path, "file",
+			   "output file name"),
+		OPT_BOOLEAN('s', "sched-stat", &inject.sched_stat,
+			    "Merge sched-stat and sched-switch for getting events "
+			    "where and how long tasks slept"),
+		OPT_INCR('v', "verbose", &verbose,
+			 "be more verbose (show build ids, etc)"),
+		OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name, "file",
+			   "kallsyms pathname"),
+		OPT_BOOLEAN('f', "force", &file.force, "don't complain, do it"),
+		OPT_END()
+	};
+	const char * const inject_usage[] = {
+		"perf inject [<options>]",
+		NULL
+	};
+
+	argc = parse_options(argc, argv, options, inject_usage, 0);
+
+	/*
+	 * Any (unrecognized) arguments left?
+	 */
+	if (argc)
+		usage_with_options(inject_usage, options);
+
+	if (perf_data_file__open(&inject.output)) {
+		perror("failed to create output file");
+		return -1;
+	}
+
+	inject.tool.ordered_events = inject.sched_stat;
+
+	file.path = inject.input_name;
+	inject.session = perf_session__new(&file, true, &inject.tool);
+	if (inject.session == NULL)
+		return -1;
+
+	if (symbol__init(&inject.session->header.env) < 0)
+		return -1;
+
+	ret = __cmd_inject(&inject);
+
+	perf_session__delete(inject.session);
+
+	return ret;
+}
diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c
new file mode 100644
index 000000000..1634186d5
--- /dev/null
+++ b/tools/perf/builtin-kmem.c
@@ -0,0 +1,1199 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+#include "util/session.h"
+#include "util/tool.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+#include "util/data.h"
+#include "util/cpumap.h"
+
+#include "util/debug.h"
+
+#include <linux/rbtree.h>
+#include <linux/string.h>
+#include <locale.h>
+
+static int	kmem_slab;
+static int	kmem_page;
+
+static long	kmem_page_size;
+
+struct alloc_stat;
+typedef int (*sort_fn_t)(struct alloc_stat *, struct alloc_stat *);
+
+static int			alloc_flag;
+static int			caller_flag;
+
+static int			alloc_lines = -1;
+static int			caller_lines = -1;
+
+static bool			raw_ip;
+
+struct alloc_stat {
+	u64	call_site;
+	u64	ptr;
+	u64	bytes_req;
+	u64	bytes_alloc;
+	u32	hit;
+	u32	pingpong;
+
+	short	alloc_cpu;
+
+	struct rb_node node;
+};
+
+static struct rb_root root_alloc_stat;
+static struct rb_root root_alloc_sorted;
+static struct rb_root root_caller_stat;
+static struct rb_root root_caller_sorted;
+
+static unsigned long total_requested, total_allocated;
+static unsigned long nr_allocs, nr_cross_allocs;
+
+static int insert_alloc_stat(unsigned long call_site, unsigned long ptr,
+			     int bytes_req, int bytes_alloc, int cpu)
+{
+	struct rb_node **node = &root_alloc_stat.rb_node;
+	struct rb_node *parent = NULL;
+	struct alloc_stat *data = NULL;
+
+	while (*node) {
+		parent = *node;
+		data = rb_entry(*node, struct alloc_stat, node);
+
+		if (ptr > data->ptr)
+			node = &(*node)->rb_right;
+		else if (ptr < data->ptr)
+			node = &(*node)->rb_left;
+		else
+			break;
+	}
+
+	if (data && data->ptr == ptr) {
+		data->hit++;
+		data->bytes_req += bytes_req;
+		data->bytes_alloc += bytes_alloc;
+	} else {
+		data = malloc(sizeof(*data));
+		if (!data) {
+			pr_err("%s: malloc failed\n", __func__);
+			return -1;
+		}
+		data->ptr = ptr;
+		data->pingpong = 0;
+		data->hit = 1;
+		data->bytes_req = bytes_req;
+		data->bytes_alloc = bytes_alloc;
+
+		rb_link_node(&data->node, parent, node);
+		rb_insert_color(&data->node, &root_alloc_stat);
+	}
+	data->call_site = call_site;
+	data->alloc_cpu = cpu;
+	return 0;
+}
+
+static int insert_caller_stat(unsigned long call_site,
+			      int bytes_req, int bytes_alloc)
+{
+	struct rb_node **node = &root_caller_stat.rb_node;
+	struct rb_node *parent = NULL;
+	struct alloc_stat *data = NULL;
+
+	while (*node) {
+		parent = *node;
+		data = rb_entry(*node, struct alloc_stat, node);
+
+		if (call_site > data->call_site)
+			node = &(*node)->rb_right;
+		else if (call_site < data->call_site)
+			node = &(*node)->rb_left;
+		else
+			break;
+	}
+
+	if (data && data->call_site == call_site) {
+		data->hit++;
+		data->bytes_req += bytes_req;
+		data->bytes_alloc += bytes_alloc;
+	} else {
+		data = malloc(sizeof(*data));
+		if (!data) {
+			pr_err("%s: malloc failed\n", __func__);
+			return -1;
+		}
+		data->call_site = call_site;
+		data->pingpong = 0;
+		data->hit = 1;
+		data->bytes_req = bytes_req;
+		data->bytes_alloc = bytes_alloc;
+
+		rb_link_node(&data->node, parent, node);
+		rb_insert_color(&data->node, &root_caller_stat);
+	}
+
+	return 0;
+}
+
+static int perf_evsel__process_alloc_event(struct perf_evsel *evsel,
+					   struct perf_sample *sample)
+{
+	unsigned long ptr = perf_evsel__intval(evsel, sample, "ptr"),
+		      call_site = perf_evsel__intval(evsel, sample, "call_site");
+	int bytes_req = perf_evsel__intval(evsel, sample, "bytes_req"),
+	    bytes_alloc = perf_evsel__intval(evsel, sample, "bytes_alloc");
+
+	if (insert_alloc_stat(call_site, ptr, bytes_req, bytes_alloc, sample->cpu) ||
+	    insert_caller_stat(call_site, bytes_req, bytes_alloc))
+		return -1;
+
+	total_requested += bytes_req;
+	total_allocated += bytes_alloc;
+
+	nr_allocs++;
+	return 0;
+}
+
+static int perf_evsel__process_alloc_node_event(struct perf_evsel *evsel,
+						struct perf_sample *sample)
+{
+	int ret = perf_evsel__process_alloc_event(evsel, sample);
+
+	if (!ret) {
+		int node1 = cpu__get_node(sample->cpu),
+		    node2 = perf_evsel__intval(evsel, sample, "node");
+
+		if (node1 != node2)
+			nr_cross_allocs++;
+	}
+
+	return ret;
+}
+
+static int ptr_cmp(struct alloc_stat *, struct alloc_stat *);
+static int callsite_cmp(struct alloc_stat *, struct alloc_stat *);
+
+static struct alloc_stat *search_alloc_stat(unsigned long ptr,
+					    unsigned long call_site,
+					    struct rb_root *root,
+					    sort_fn_t sort_fn)
+{
+	struct rb_node *node = root->rb_node;
+	struct alloc_stat key = { .ptr = ptr, .call_site = call_site };
+
+	while (node) {
+		struct alloc_stat *data;
+		int cmp;
+
+		data = rb_entry(node, struct alloc_stat, node);
+
+		cmp = sort_fn(&key, data);
+		if (cmp < 0)
+			node = node->rb_left;
+		else if (cmp > 0)
+			node = node->rb_right;
+		else
+			return data;
+	}
+	return NULL;
+}
+
+static int perf_evsel__process_free_event(struct perf_evsel *evsel,
+					  struct perf_sample *sample)
+{
+	unsigned long ptr = perf_evsel__intval(evsel, sample, "ptr");
+	struct alloc_stat *s_alloc, *s_caller;
+
+	s_alloc = search_alloc_stat(ptr, 0, &root_alloc_stat, ptr_cmp);
+	if (!s_alloc)
+		return 0;
+
+	if ((short)sample->cpu != s_alloc->alloc_cpu) {
+		s_alloc->pingpong++;
+
+		s_caller = search_alloc_stat(0, s_alloc->call_site,
+					     &root_caller_stat, callsite_cmp);
+		if (!s_caller)
+			return -1;
+		s_caller->pingpong++;
+	}
+	s_alloc->alloc_cpu = -1;
+
+	return 0;
+}
+
+static u64 total_page_alloc_bytes;
+static u64 total_page_free_bytes;
+static u64 total_page_nomatch_bytes;
+static u64 total_page_fail_bytes;
+static unsigned long nr_page_allocs;
+static unsigned long nr_page_frees;
+static unsigned long nr_page_fails;
+static unsigned long nr_page_nomatch;
+
+static bool use_pfn;
+
+#define MAX_MIGRATE_TYPES  6
+#define MAX_PAGE_ORDER     11
+
+static int order_stats[MAX_PAGE_ORDER][MAX_MIGRATE_TYPES];
+
+struct page_stat {
+	struct rb_node 	node;
+	u64 		page;
+	int 		order;
+	unsigned 	gfp_flags;
+	unsigned 	migrate_type;
+	u64		alloc_bytes;
+	u64 		free_bytes;
+	int 		nr_alloc;
+	int 		nr_free;
+};
+
+static struct rb_root page_tree;
+static struct rb_root page_alloc_tree;
+static struct rb_root page_alloc_sorted;
+
+static struct page_stat *search_page(unsigned long page, bool create)
+{
+	struct rb_node **node = &page_tree.rb_node;
+	struct rb_node *parent = NULL;
+	struct page_stat *data;
+
+	while (*node) {
+		s64 cmp;
+
+		parent = *node;
+		data = rb_entry(*node, struct page_stat, node);
+
+		cmp = data->page - page;
+		if (cmp < 0)
+			node = &parent->rb_left;
+		else if (cmp > 0)
+			node = &parent->rb_right;
+		else
+			return data;
+	}
+
+	if (!create)
+		return NULL;
+
+	data = zalloc(sizeof(*data));
+	if (data != NULL) {
+		data->page = page;
+
+		rb_link_node(&data->node, parent, node);
+		rb_insert_color(&data->node, &page_tree);
+	}
+
+	return data;
+}
+
+static int page_stat_cmp(struct page_stat *a, struct page_stat *b)
+{
+	if (a->page > b->page)
+		return -1;
+	if (a->page < b->page)
+		return 1;
+	if (a->order > b->order)
+		return -1;
+	if (a->order < b->order)
+		return 1;
+	if (a->migrate_type > b->migrate_type)
+		return -1;
+	if (a->migrate_type < b->migrate_type)
+		return 1;
+	if (a->gfp_flags > b->gfp_flags)
+		return -1;
+	if (a->gfp_flags < b->gfp_flags)
+		return 1;
+	return 0;
+}
+
+static struct page_stat *search_page_alloc_stat(struct page_stat *pstat, bool create)
+{
+	struct rb_node **node = &page_alloc_tree.rb_node;
+	struct rb_node *parent = NULL;
+	struct page_stat *data;
+
+	while (*node) {
+		s64 cmp;
+
+		parent = *node;
+		data = rb_entry(*node, struct page_stat, node);
+
+		cmp = page_stat_cmp(data, pstat);
+		if (cmp < 0)
+			node = &parent->rb_left;
+		else if (cmp > 0)
+			node = &parent->rb_right;
+		else
+			return data;
+	}
+
+	if (!create)
+		return NULL;
+
+	data = zalloc(sizeof(*data));
+	if (data != NULL) {
+		data->page = pstat->page;
+		data->order = pstat->order;
+		data->gfp_flags = pstat->gfp_flags;
+		data->migrate_type = pstat->migrate_type;
+
+		rb_link_node(&data->node, parent, node);
+		rb_insert_color(&data->node, &page_alloc_tree);
+	}
+
+	return data;
+}
+
+static bool valid_page(u64 pfn_or_page)
+{
+	if (use_pfn && pfn_or_page == -1UL)
+		return false;
+	if (!use_pfn && pfn_or_page == 0)
+		return false;
+	return true;
+}
+
+static int perf_evsel__process_page_alloc_event(struct perf_evsel *evsel,
+						struct perf_sample *sample)
+{
+	u64 page;
+	unsigned int order = perf_evsel__intval(evsel, sample, "order");
+	unsigned int gfp_flags = perf_evsel__intval(evsel, sample, "gfp_flags");
+	unsigned int migrate_type = perf_evsel__intval(evsel, sample,
+						       "migratetype");
+	u64 bytes = kmem_page_size << order;
+	struct page_stat *pstat;
+	struct page_stat this = {
+		.order = order,
+		.gfp_flags = gfp_flags,
+		.migrate_type = migrate_type,
+	};
+
+	if (use_pfn)
+		page = perf_evsel__intval(evsel, sample, "pfn");
+	else
+		page = perf_evsel__intval(evsel, sample, "page");
+
+	nr_page_allocs++;
+	total_page_alloc_bytes += bytes;
+
+	if (!valid_page(page)) {
+		nr_page_fails++;
+		total_page_fail_bytes += bytes;
+
+		return 0;
+	}
+
+	/*
+	 * This is to find the current page (with correct gfp flags and
+	 * migrate type) at free event.
+	 */
+	pstat = search_page(page, true);
+	if (pstat == NULL)
+		return -ENOMEM;
+
+	pstat->order = order;
+	pstat->gfp_flags = gfp_flags;
+	pstat->migrate_type = migrate_type;
+
+	this.page = page;
+	pstat = search_page_alloc_stat(&this, true);
+	if (pstat == NULL)
+		return -ENOMEM;
+
+	pstat->nr_alloc++;
+	pstat->alloc_bytes += bytes;
+
+	order_stats[order][migrate_type]++;
+
+	return 0;
+}
+
+static int perf_evsel__process_page_free_event(struct perf_evsel *evsel,
+						struct perf_sample *sample)
+{
+	u64 page;
+	unsigned int order = perf_evsel__intval(evsel, sample, "order");
+	u64 bytes = kmem_page_size << order;
+	struct page_stat *pstat;
+	struct page_stat this = {
+		.order = order,
+	};
+
+	if (use_pfn)
+		page = perf_evsel__intval(evsel, sample, "pfn");
+	else
+		page = perf_evsel__intval(evsel, sample, "page");
+
+	nr_page_frees++;
+	total_page_free_bytes += bytes;
+
+	pstat = search_page(page, false);
+	if (pstat == NULL) {
+		pr_debug2("missing free at page %"PRIx64" (order: %d)\n",
+			  page, order);
+
+		nr_page_nomatch++;
+		total_page_nomatch_bytes += bytes;
+
+		return 0;
+	}
+
+	this.page = page;
+	this.gfp_flags = pstat->gfp_flags;
+	this.migrate_type = pstat->migrate_type;
+
+	rb_erase(&pstat->node, &page_tree);
+	free(pstat);
+
+	pstat = search_page_alloc_stat(&this, false);
+	if (pstat == NULL)
+		return -ENOENT;
+
+	pstat->nr_free++;
+	pstat->free_bytes += bytes;
+
+	return 0;
+}
+
+typedef int (*tracepoint_handler)(struct perf_evsel *evsel,
+				  struct perf_sample *sample);
+
+static int process_sample_event(struct perf_tool *tool __maybe_unused,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel,
+				struct machine *machine)
+{
+	struct thread *thread = machine__findnew_thread(machine, sample->pid,
+							sample->tid);
+
+	if (thread == NULL) {
+		pr_debug("problem processing %d event, skipping it.\n",
+			 event->header.type);
+		return -1;
+	}
+
+	dump_printf(" ... thread: %s:%d\n", thread__comm_str(thread), thread->tid);
+
+	if (evsel->handler != NULL) {
+		tracepoint_handler f = evsel->handler;
+		return f(evsel, sample);
+	}
+
+	return 0;
+}
+
+static struct perf_tool perf_kmem = {
+	.sample		 = process_sample_event,
+	.comm		 = perf_event__process_comm,
+	.mmap		 = perf_event__process_mmap,
+	.mmap2		 = perf_event__process_mmap2,
+	.ordered_events	 = true,
+};
+
+static double fragmentation(unsigned long n_req, unsigned long n_alloc)
+{
+	if (n_alloc == 0)
+		return 0.0;
+	else
+		return 100.0 - (100.0 * n_req / n_alloc);
+}
+
+static void __print_slab_result(struct rb_root *root,
+				struct perf_session *session,
+				int n_lines, int is_caller)
+{
+	struct rb_node *next;
+	struct machine *machine = &session->machines.host;
+
+	printf("%.105s\n", graph_dotted_line);
+	printf(" %-34s |",  is_caller ? "Callsite": "Alloc Ptr");
+	printf(" Total_alloc/Per | Total_req/Per   | Hit      | Ping-pong | Frag\n");
+	printf("%.105s\n", graph_dotted_line);
+
+	next = rb_first(root);
+
+	while (next && n_lines--) {
+		struct alloc_stat *data = rb_entry(next, struct alloc_stat,
+						   node);
+		struct symbol *sym = NULL;
+		struct map *map;
+		char buf[BUFSIZ];
+		u64 addr;
+
+		if (is_caller) {
+			addr = data->call_site;
+			if (!raw_ip)
+				sym = machine__find_kernel_function(machine, addr, &map, NULL);
+		} else
+			addr = data->ptr;
+
+		if (sym != NULL)
+			snprintf(buf, sizeof(buf), "%s+%" PRIx64 "", sym->name,
+				 addr - map->unmap_ip(map, sym->start));
+		else
+			snprintf(buf, sizeof(buf), "%#" PRIx64 "", addr);
+		printf(" %-34s |", buf);
+
+		printf(" %9llu/%-5lu | %9llu/%-5lu | %8lu | %9lu | %6.3f%%\n",
+		       (unsigned long long)data->bytes_alloc,
+		       (unsigned long)data->bytes_alloc / data->hit,
+		       (unsigned long long)data->bytes_req,
+		       (unsigned long)data->bytes_req / data->hit,
+		       (unsigned long)data->hit,
+		       (unsigned long)data->pingpong,
+		       fragmentation(data->bytes_req, data->bytes_alloc));
+
+		next = rb_next(next);
+	}
+
+	if (n_lines == -1)
+		printf(" ...                                | ...             | ...             | ...      | ...       | ...   \n");
+
+	printf("%.105s\n", graph_dotted_line);
+}
+
+static const char * const migrate_type_str[] = {
+	"UNMOVABL",
+	"RECLAIM",
+	"MOVABLE",
+	"RESERVED",
+	"CMA/ISLT",
+	"UNKNOWN",
+};
+
+static void __print_page_result(struct rb_root *root,
+				struct perf_session *session __maybe_unused,
+				int n_lines)
+{
+	struct rb_node *next = rb_first(root);
+	const char *format;
+
+	printf("\n%.80s\n", graph_dotted_line);
+	printf(" %-16s | Total alloc (KB) | Hits      | Order | Mig.type | GFP flags\n",
+	       use_pfn ? "PFN" : "Page");
+	printf("%.80s\n", graph_dotted_line);
+
+	if (use_pfn)
+		format = " %16llu | %'16llu | %'9d | %5d | %8s |  %08lx\n";
+	else
+		format = " %016llx | %'16llu | %'9d | %5d | %8s |  %08lx\n";
+
+	while (next && n_lines--) {
+		struct page_stat *data;
+
+		data = rb_entry(next, struct page_stat, node);
+
+		printf(format, (unsigned long long)data->page,
+		       (unsigned long long)data->alloc_bytes / 1024,
+		       data->nr_alloc, data->order,
+		       migrate_type_str[data->migrate_type],
+		       (unsigned long)data->gfp_flags);
+
+		next = rb_next(next);
+	}
+
+	if (n_lines == -1)
+		printf(" ...              | ...              | ...       | ...   | ...      | ...     \n");
+
+	printf("%.80s\n", graph_dotted_line);
+}
+
+static void print_slab_summary(void)
+{
+	printf("\nSUMMARY (SLAB allocator)");
+	printf("\n========================\n");
+	printf("Total bytes requested: %'lu\n", total_requested);
+	printf("Total bytes allocated: %'lu\n", total_allocated);
+	printf("Total bytes wasted on internal fragmentation: %'lu\n",
+	       total_allocated - total_requested);
+	printf("Internal fragmentation: %f%%\n",
+	       fragmentation(total_requested, total_allocated));
+	printf("Cross CPU allocations: %'lu/%'lu\n", nr_cross_allocs, nr_allocs);
+}
+
+static void print_page_summary(void)
+{
+	int o, m;
+	u64 nr_alloc_freed = nr_page_frees - nr_page_nomatch;
+	u64 total_alloc_freed_bytes = total_page_free_bytes - total_page_nomatch_bytes;
+
+	printf("\nSUMMARY (page allocator)");
+	printf("\n========================\n");
+	printf("%-30s: %'16lu   [ %'16"PRIu64" KB ]\n", "Total allocation requests",
+	       nr_page_allocs, total_page_alloc_bytes / 1024);
+	printf("%-30s: %'16lu   [ %'16"PRIu64" KB ]\n", "Total free requests",
+	       nr_page_frees, total_page_free_bytes / 1024);
+	printf("\n");
+
+	printf("%-30s: %'16"PRIu64"   [ %'16"PRIu64" KB ]\n", "Total alloc+freed requests",
+	       nr_alloc_freed, (total_alloc_freed_bytes) / 1024);
+	printf("%-30s: %'16"PRIu64"   [ %'16"PRIu64" KB ]\n", "Total alloc-only requests",
+	       nr_page_allocs - nr_alloc_freed,
+	       (total_page_alloc_bytes - total_alloc_freed_bytes) / 1024);
+	printf("%-30s: %'16lu   [ %'16"PRIu64" KB ]\n", "Total free-only requests",
+	       nr_page_nomatch, total_page_nomatch_bytes / 1024);
+	printf("\n");
+
+	printf("%-30s: %'16lu   [ %'16"PRIu64" KB ]\n", "Total allocation failures",
+	       nr_page_fails, total_page_fail_bytes / 1024);
+	printf("\n");
+
+	printf("%5s  %12s  %12s  %12s  %12s  %12s\n", "Order",  "Unmovable",
+	       "Reclaimable", "Movable", "Reserved", "CMA/Isolated");
+	printf("%.5s  %.12s  %.12s  %.12s  %.12s  %.12s\n", graph_dotted_line,
+	       graph_dotted_line, graph_dotted_line, graph_dotted_line,
+	       graph_dotted_line, graph_dotted_line);
+
+	for (o = 0; o < MAX_PAGE_ORDER; o++) {
+		printf("%5d", o);
+		for (m = 0; m < MAX_MIGRATE_TYPES - 1; m++) {
+			if (order_stats[o][m])
+				printf("  %'12d", order_stats[o][m]);
+			else
+				printf("  %12c", '.');
+		}
+		printf("\n");
+	}
+}
+
+static void print_slab_result(struct perf_session *session)
+{
+	if (caller_flag)
+		__print_slab_result(&root_caller_sorted, session, caller_lines, 1);
+	if (alloc_flag)
+		__print_slab_result(&root_alloc_sorted, session, alloc_lines, 0);
+	print_slab_summary();
+}
+
+static void print_page_result(struct perf_session *session)
+{
+	if (alloc_flag)
+		__print_page_result(&page_alloc_sorted, session, alloc_lines);
+	print_page_summary();
+}
+
+static void print_result(struct perf_session *session)
+{
+	if (kmem_slab)
+		print_slab_result(session);
+	if (kmem_page)
+		print_page_result(session);
+}
+
+struct sort_dimension {
+	const char		name[20];
+	sort_fn_t		cmp;
+	struct list_head	list;
+};
+
+static LIST_HEAD(caller_sort);
+static LIST_HEAD(alloc_sort);
+
+static void sort_slab_insert(struct rb_root *root, struct alloc_stat *data,
+			     struct list_head *sort_list)
+{
+	struct rb_node **new = &(root->rb_node);
+	struct rb_node *parent = NULL;
+	struct sort_dimension *sort;
+
+	while (*new) {
+		struct alloc_stat *this;
+		int cmp = 0;
+
+		this = rb_entry(*new, struct alloc_stat, node);
+		parent = *new;
+
+		list_for_each_entry(sort, sort_list, list) {
+			cmp = sort->cmp(data, this);
+			if (cmp)
+				break;
+		}
+
+		if (cmp > 0)
+			new = &((*new)->rb_left);
+		else
+			new = &((*new)->rb_right);
+	}
+
+	rb_link_node(&data->node, parent, new);
+	rb_insert_color(&data->node, root);
+}
+
+static void __sort_slab_result(struct rb_root *root, struct rb_root *root_sorted,
+			       struct list_head *sort_list)
+{
+	struct rb_node *node;
+	struct alloc_stat *data;
+
+	for (;;) {
+		node = rb_first(root);
+		if (!node)
+			break;
+
+		rb_erase(node, root);
+		data = rb_entry(node, struct alloc_stat, node);
+		sort_slab_insert(root_sorted, data, sort_list);
+	}
+}
+
+static void sort_page_insert(struct rb_root *root, struct page_stat *data)
+{
+	struct rb_node **new = &root->rb_node;
+	struct rb_node *parent = NULL;
+
+	while (*new) {
+		struct page_stat *this;
+		int cmp = 0;
+
+		this = rb_entry(*new, struct page_stat, node);
+		parent = *new;
+
+		/* TODO: support more sort key */
+		cmp = data->alloc_bytes - this->alloc_bytes;
+
+		if (cmp > 0)
+			new = &parent->rb_left;
+		else
+			new = &parent->rb_right;
+	}
+
+	rb_link_node(&data->node, parent, new);
+	rb_insert_color(&data->node, root);
+}
+
+static void __sort_page_result(struct rb_root *root, struct rb_root *root_sorted)
+{
+	struct rb_node *node;
+	struct page_stat *data;
+
+	for (;;) {
+		node = rb_first(root);
+		if (!node)
+			break;
+
+		rb_erase(node, root);
+		data = rb_entry(node, struct page_stat, node);
+		sort_page_insert(root_sorted, data);
+	}
+}
+
+static void sort_result(void)
+{
+	if (kmem_slab) {
+		__sort_slab_result(&root_alloc_stat, &root_alloc_sorted,
+				   &alloc_sort);
+		__sort_slab_result(&root_caller_stat, &root_caller_sorted,
+				   &caller_sort);
+	}
+	if (kmem_page) {
+		__sort_page_result(&page_alloc_tree, &page_alloc_sorted);
+	}
+}
+
+static int __cmd_kmem(struct perf_session *session)
+{
+	int err = -EINVAL;
+	struct perf_evsel *evsel;
+	const struct perf_evsel_str_handler kmem_tracepoints[] = {
+		/* slab allocator */
+		{ "kmem:kmalloc",		perf_evsel__process_alloc_event, },
+    		{ "kmem:kmem_cache_alloc",	perf_evsel__process_alloc_event, },
+		{ "kmem:kmalloc_node",		perf_evsel__process_alloc_node_event, },
+    		{ "kmem:kmem_cache_alloc_node", perf_evsel__process_alloc_node_event, },
+		{ "kmem:kfree",			perf_evsel__process_free_event, },
+    		{ "kmem:kmem_cache_free",	perf_evsel__process_free_event, },
+		/* page allocator */
+		{ "kmem:mm_page_alloc",		perf_evsel__process_page_alloc_event, },
+		{ "kmem:mm_page_free",		perf_evsel__process_page_free_event, },
+	};
+
+	if (!perf_session__has_traces(session, "kmem record"))
+		goto out;
+
+	if (perf_session__set_tracepoints_handlers(session, kmem_tracepoints)) {
+		pr_err("Initializing perf session tracepoint handlers failed\n");
+		goto out;
+	}
+
+	evlist__for_each(session->evlist, evsel) {
+		if (!strcmp(perf_evsel__name(evsel), "kmem:mm_page_alloc") &&
+		    perf_evsel__field(evsel, "pfn")) {
+			use_pfn = true;
+			break;
+		}
+	}
+
+	setup_pager();
+	err = perf_session__process_events(session);
+	if (err != 0) {
+		pr_err("error during process events: %d\n", err);
+		goto out;
+	}
+	sort_result();
+	print_result(session);
+out:
+	return err;
+}
+
+static int ptr_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+	if (l->ptr < r->ptr)
+		return -1;
+	else if (l->ptr > r->ptr)
+		return 1;
+	return 0;
+}
+
+static struct sort_dimension ptr_sort_dimension = {
+	.name	= "ptr",
+	.cmp	= ptr_cmp,
+};
+
+static int callsite_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+	if (l->call_site < r->call_site)
+		return -1;
+	else if (l->call_site > r->call_site)
+		return 1;
+	return 0;
+}
+
+static struct sort_dimension callsite_sort_dimension = {
+	.name	= "callsite",
+	.cmp	= callsite_cmp,
+};
+
+static int hit_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+	if (l->hit < r->hit)
+		return -1;
+	else if (l->hit > r->hit)
+		return 1;
+	return 0;
+}
+
+static struct sort_dimension hit_sort_dimension = {
+	.name	= "hit",
+	.cmp	= hit_cmp,
+};
+
+static int bytes_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+	if (l->bytes_alloc < r->bytes_alloc)
+		return -1;
+	else if (l->bytes_alloc > r->bytes_alloc)
+		return 1;
+	return 0;
+}
+
+static struct sort_dimension bytes_sort_dimension = {
+	.name	= "bytes",
+	.cmp	= bytes_cmp,
+};
+
+static int frag_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+	double x, y;
+
+	x = fragmentation(l->bytes_req, l->bytes_alloc);
+	y = fragmentation(r->bytes_req, r->bytes_alloc);
+
+	if (x < y)
+		return -1;
+	else if (x > y)
+		return 1;
+	return 0;
+}
+
+static struct sort_dimension frag_sort_dimension = {
+	.name	= "frag",
+	.cmp	= frag_cmp,
+};
+
+static int pingpong_cmp(struct alloc_stat *l, struct alloc_stat *r)
+{
+	if (l->pingpong < r->pingpong)
+		return -1;
+	else if (l->pingpong > r->pingpong)
+		return 1;
+	return 0;
+}
+
+static struct sort_dimension pingpong_sort_dimension = {
+	.name	= "pingpong",
+	.cmp	= pingpong_cmp,
+};
+
+static struct sort_dimension *avail_sorts[] = {
+	&ptr_sort_dimension,
+	&callsite_sort_dimension,
+	&hit_sort_dimension,
+	&bytes_sort_dimension,
+	&frag_sort_dimension,
+	&pingpong_sort_dimension,
+};
+
+#define NUM_AVAIL_SORTS	((int)ARRAY_SIZE(avail_sorts))
+
+static int sort_dimension__add(const char *tok, struct list_head *list)
+{
+	struct sort_dimension *sort;
+	int i;
+
+	for (i = 0; i < NUM_AVAIL_SORTS; i++) {
+		if (!strcmp(avail_sorts[i]->name, tok)) {
+			sort = memdup(avail_sorts[i], sizeof(*avail_sorts[i]));
+			if (!sort) {
+				pr_err("%s: memdup failed\n", __func__);
+				return -1;
+			}
+			list_add_tail(&sort->list, list);
+			return 0;
+		}
+	}
+
+	return -1;
+}
+
+static int setup_sorting(struct list_head *sort_list, const char *arg)
+{
+	char *tok;
+	char *str = strdup(arg);
+	char *pos = str;
+
+	if (!str) {
+		pr_err("%s: strdup failed\n", __func__);
+		return -1;
+	}
+
+	while (true) {
+		tok = strsep(&pos, ",");
+		if (!tok)
+			break;
+		if (sort_dimension__add(tok, sort_list) < 0) {
+			error("Unknown --sort key: '%s'", tok);
+			free(str);
+			return -1;
+		}
+	}
+
+	free(str);
+	return 0;
+}
+
+static int parse_sort_opt(const struct option *opt __maybe_unused,
+			  const char *arg, int unset __maybe_unused)
+{
+	if (!arg)
+		return -1;
+
+	if (caller_flag > alloc_flag)
+		return setup_sorting(&caller_sort, arg);
+	else
+		return setup_sorting(&alloc_sort, arg);
+
+	return 0;
+}
+
+static int parse_caller_opt(const struct option *opt __maybe_unused,
+			    const char *arg __maybe_unused,
+			    int unset __maybe_unused)
+{
+	caller_flag = (alloc_flag + 1);
+	return 0;
+}
+
+static int parse_alloc_opt(const struct option *opt __maybe_unused,
+			   const char *arg __maybe_unused,
+			   int unset __maybe_unused)
+{
+	alloc_flag = (caller_flag + 1);
+	return 0;
+}
+
+static int parse_slab_opt(const struct option *opt __maybe_unused,
+			  const char *arg __maybe_unused,
+			  int unset __maybe_unused)
+{
+	kmem_slab = (kmem_page + 1);
+	return 0;
+}
+
+static int parse_page_opt(const struct option *opt __maybe_unused,
+			  const char *arg __maybe_unused,
+			  int unset __maybe_unused)
+{
+	kmem_page = (kmem_slab + 1);
+	return 0;
+}
+
+static int parse_line_opt(const struct option *opt __maybe_unused,
+			  const char *arg, int unset __maybe_unused)
+{
+	int lines;
+
+	if (!arg)
+		return -1;
+
+	lines = strtoul(arg, NULL, 10);
+
+	if (caller_flag > alloc_flag)
+		caller_lines = lines;
+	else
+		alloc_lines = lines;
+
+	return 0;
+}
+
+static int __cmd_record(int argc, const char **argv)
+{
+	const char * const record_args[] = {
+	"record", "-a", "-R", "-c", "1",
+	};
+	const char * const slab_events[] = {
+	"-e", "kmem:kmalloc",
+	"-e", "kmem:kmalloc_node",
+	"-e", "kmem:kfree",
+	"-e", "kmem:kmem_cache_alloc",
+	"-e", "kmem:kmem_cache_alloc_node",
+	"-e", "kmem:kmem_cache_free",
+	};
+	const char * const page_events[] = {
+	"-e", "kmem:mm_page_alloc",
+	"-e", "kmem:mm_page_free",
+	};
+	unsigned int rec_argc, i, j;
+	const char **rec_argv;
+
+	rec_argc = ARRAY_SIZE(record_args) + argc - 1;
+	if (kmem_slab)
+		rec_argc += ARRAY_SIZE(slab_events);
+	if (kmem_page)
+		rec_argc += ARRAY_SIZE(page_events);
+
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+	if (rec_argv == NULL)
+		return -ENOMEM;
+
+	for (i = 0; i < ARRAY_SIZE(record_args); i++)
+		rec_argv[i] = strdup(record_args[i]);
+
+	if (kmem_slab) {
+		for (j = 0; j < ARRAY_SIZE(slab_events); j++, i++)
+			rec_argv[i] = strdup(slab_events[j]);
+	}
+	if (kmem_page) {
+		for (j = 0; j < ARRAY_SIZE(page_events); j++, i++)
+			rec_argv[i] = strdup(page_events[j]);
+	}
+
+	for (j = 1; j < (unsigned int)argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	return cmd_record(i, rec_argv, NULL);
+}
+
+int cmd_kmem(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	const char * const default_sort_order = "frag,hit,bytes";
+	struct perf_data_file file = {
+		.mode = PERF_DATA_MODE_READ,
+	};
+	const struct option kmem_options[] = {
+	OPT_STRING('i', "input", &input_name, "file", "input file name"),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_CALLBACK_NOOPT(0, "caller", NULL, NULL,
+			   "show per-callsite statistics", parse_caller_opt),
+	OPT_CALLBACK_NOOPT(0, "alloc", NULL, NULL,
+			   "show per-allocation statistics", parse_alloc_opt),
+	OPT_CALLBACK('s', "sort", NULL, "key[,key2...]",
+		     "sort by keys: ptr, call_site, bytes, hit, pingpong, frag",
+		     parse_sort_opt),
+	OPT_CALLBACK('l', "line", NULL, "num", "show n lines", parse_line_opt),
+	OPT_BOOLEAN(0, "raw-ip", &raw_ip, "show raw ip instead of symbol"),
+	OPT_BOOLEAN('f', "force", &file.force, "don't complain, do it"),
+	OPT_CALLBACK_NOOPT(0, "slab", NULL, NULL, "Analyze slab allocator",
+			   parse_slab_opt),
+	OPT_CALLBACK_NOOPT(0, "page", NULL, NULL, "Analyze page allocator",
+			   parse_page_opt),
+	OPT_END()
+	};
+	const char *const kmem_subcommands[] = { "record", "stat", NULL };
+	const char *kmem_usage[] = {
+		NULL,
+		NULL
+	};
+	struct perf_session *session;
+	int ret = -1;
+
+	argc = parse_options_subcommand(argc, argv, kmem_options,
+					kmem_subcommands, kmem_usage, 0);
+
+	if (!argc)
+		usage_with_options(kmem_usage, kmem_options);
+
+	if (kmem_slab == 0 && kmem_page == 0)
+		kmem_slab = 1;  /* for backward compatibility */
+
+	if (!strncmp(argv[0], "rec", 3)) {
+		symbol__init(NULL);
+		return __cmd_record(argc, argv);
+	}
+
+	file.path = input_name;
+
+	session = perf_session__new(&file, false, &perf_kmem);
+	if (session == NULL)
+		return -1;
+
+	if (kmem_page) {
+		struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+
+		if (evsel == NULL || evsel->tp_format == NULL) {
+			pr_err("invalid event found.. aborting\n");
+			return -1;
+		}
+
+		kmem_page_size = pevent_get_page_size(evsel->tp_format->pevent);
+	}
+
+	symbol__init(&session->header.env);
+
+	if (!strcmp(argv[0], "stat")) {
+		setlocale(LC_ALL, "");
+
+		if (cpu__setup_cpunode_map())
+			goto out_delete;
+
+		if (list_empty(&caller_sort))
+			setup_sorting(&caller_sort, default_sort_order);
+		if (list_empty(&alloc_sort))
+			setup_sorting(&alloc_sort, default_sort_order);
+
+		ret = __cmd_kmem(session);
+	} else
+		usage_with_options(kmem_usage, kmem_options);
+
+out_delete:
+	perf_session__delete(session);
+
+	return ret;
+}
+
diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c
new file mode 100644
index 000000000..1f9338f61
--- /dev/null
+++ b/tools/perf/builtin-kvm.c
@@ -0,0 +1,1581 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/evsel.h"
+#include "util/evlist.h"
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+#include "util/session.h"
+#include "util/intlist.h"
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+#include "util/debug.h"
+#include <api/fs/debugfs.h>
+#include "util/tool.h"
+#include "util/stat.h"
+#include "util/top.h"
+#include "util/data.h"
+#include "util/ordered-events.h"
+
+#include <sys/prctl.h>
+#ifdef HAVE_TIMERFD_SUPPORT
+#include <sys/timerfd.h>
+#endif
+
+#include <termios.h>
+#include <semaphore.h>
+#include <pthread.h>
+#include <math.h>
+
+#ifdef HAVE_KVM_STAT_SUPPORT
+#include <asm/kvm_perf.h>
+#include "util/kvm-stat.h"
+
+void exit_event_get_key(struct perf_evsel *evsel,
+			struct perf_sample *sample,
+			struct event_key *key)
+{
+	key->info = 0;
+	key->key = perf_evsel__intval(evsel, sample, KVM_EXIT_REASON);
+}
+
+bool kvm_exit_event(struct perf_evsel *evsel)
+{
+	return !strcmp(evsel->name, KVM_EXIT_TRACE);
+}
+
+bool exit_event_begin(struct perf_evsel *evsel,
+		      struct perf_sample *sample, struct event_key *key)
+{
+	if (kvm_exit_event(evsel)) {
+		exit_event_get_key(evsel, sample, key);
+		return true;
+	}
+
+	return false;
+}
+
+bool kvm_entry_event(struct perf_evsel *evsel)
+{
+	return !strcmp(evsel->name, KVM_ENTRY_TRACE);
+}
+
+bool exit_event_end(struct perf_evsel *evsel,
+		    struct perf_sample *sample __maybe_unused,
+		    struct event_key *key __maybe_unused)
+{
+	return kvm_entry_event(evsel);
+}
+
+static const char *get_exit_reason(struct perf_kvm_stat *kvm,
+				   struct exit_reasons_table *tbl,
+				   u64 exit_code)
+{
+	while (tbl->reason != NULL) {
+		if (tbl->exit_code == exit_code)
+			return tbl->reason;
+		tbl++;
+	}
+
+	pr_err("unknown kvm exit code:%lld on %s\n",
+		(unsigned long long)exit_code, kvm->exit_reasons_isa);
+	return "UNKNOWN";
+}
+
+void exit_event_decode_key(struct perf_kvm_stat *kvm,
+			   struct event_key *key,
+			   char *decode)
+{
+	const char *exit_reason = get_exit_reason(kvm, key->exit_reasons,
+						  key->key);
+
+	scnprintf(decode, DECODE_STR_LEN, "%s", exit_reason);
+}
+
+static bool register_kvm_events_ops(struct perf_kvm_stat *kvm)
+{
+	struct kvm_reg_events_ops *events_ops = kvm_reg_events_ops;
+
+	for (events_ops = kvm_reg_events_ops; events_ops->name; events_ops++) {
+		if (!strcmp(events_ops->name, kvm->report_event)) {
+			kvm->events_ops = events_ops->ops;
+			return true;
+		}
+	}
+
+	return false;
+}
+
+struct vcpu_event_record {
+	int vcpu_id;
+	u64 start_time;
+	struct kvm_event *last_event;
+};
+
+
+static void init_kvm_event_record(struct perf_kvm_stat *kvm)
+{
+	unsigned int i;
+
+	for (i = 0; i < EVENTS_CACHE_SIZE; i++)
+		INIT_LIST_HEAD(&kvm->kvm_events_cache[i]);
+}
+
+#ifdef HAVE_TIMERFD_SUPPORT
+static void clear_events_cache_stats(struct list_head *kvm_events_cache)
+{
+	struct list_head *head;
+	struct kvm_event *event;
+	unsigned int i;
+	int j;
+
+	for (i = 0; i < EVENTS_CACHE_SIZE; i++) {
+		head = &kvm_events_cache[i];
+		list_for_each_entry(event, head, hash_entry) {
+			/* reset stats for event */
+			event->total.time = 0;
+			init_stats(&event->total.stats);
+
+			for (j = 0; j < event->max_vcpu; ++j) {
+				event->vcpu[j].time = 0;
+				init_stats(&event->vcpu[j].stats);
+			}
+		}
+	}
+}
+#endif
+
+static int kvm_events_hash_fn(u64 key)
+{
+	return key & (EVENTS_CACHE_SIZE - 1);
+}
+
+static bool kvm_event_expand(struct kvm_event *event, int vcpu_id)
+{
+	int old_max_vcpu = event->max_vcpu;
+	void *prev;
+
+	if (vcpu_id < event->max_vcpu)
+		return true;
+
+	while (event->max_vcpu <= vcpu_id)
+		event->max_vcpu += DEFAULT_VCPU_NUM;
+
+	prev = event->vcpu;
+	event->vcpu = realloc(event->vcpu,
+			      event->max_vcpu * sizeof(*event->vcpu));
+	if (!event->vcpu) {
+		free(prev);
+		pr_err("Not enough memory\n");
+		return false;
+	}
+
+	memset(event->vcpu + old_max_vcpu, 0,
+	       (event->max_vcpu - old_max_vcpu) * sizeof(*event->vcpu));
+	return true;
+}
+
+static struct kvm_event *kvm_alloc_init_event(struct event_key *key)
+{
+	struct kvm_event *event;
+
+	event = zalloc(sizeof(*event));
+	if (!event) {
+		pr_err("Not enough memory\n");
+		return NULL;
+	}
+
+	event->key = *key;
+	init_stats(&event->total.stats);
+	return event;
+}
+
+static struct kvm_event *find_create_kvm_event(struct perf_kvm_stat *kvm,
+					       struct event_key *key)
+{
+	struct kvm_event *event;
+	struct list_head *head;
+
+	BUG_ON(key->key == INVALID_KEY);
+
+	head = &kvm->kvm_events_cache[kvm_events_hash_fn(key->key)];
+	list_for_each_entry(event, head, hash_entry) {
+		if (event->key.key == key->key && event->key.info == key->info)
+			return event;
+	}
+
+	event = kvm_alloc_init_event(key);
+	if (!event)
+		return NULL;
+
+	list_add(&event->hash_entry, head);
+	return event;
+}
+
+static bool handle_begin_event(struct perf_kvm_stat *kvm,
+			       struct vcpu_event_record *vcpu_record,
+			       struct event_key *key, u64 timestamp)
+{
+	struct kvm_event *event = NULL;
+
+	if (key->key != INVALID_KEY)
+		event = find_create_kvm_event(kvm, key);
+
+	vcpu_record->last_event = event;
+	vcpu_record->start_time = timestamp;
+	return true;
+}
+
+static void
+kvm_update_event_stats(struct kvm_event_stats *kvm_stats, u64 time_diff)
+{
+	kvm_stats->time += time_diff;
+	update_stats(&kvm_stats->stats, time_diff);
+}
+
+static double kvm_event_rel_stddev(int vcpu_id, struct kvm_event *event)
+{
+	struct kvm_event_stats *kvm_stats = &event->total;
+
+	if (vcpu_id != -1)
+		kvm_stats = &event->vcpu[vcpu_id];
+
+	return rel_stddev_stats(stddev_stats(&kvm_stats->stats),
+				avg_stats(&kvm_stats->stats));
+}
+
+static bool update_kvm_event(struct kvm_event *event, int vcpu_id,
+			     u64 time_diff)
+{
+	if (vcpu_id == -1) {
+		kvm_update_event_stats(&event->total, time_diff);
+		return true;
+	}
+
+	if (!kvm_event_expand(event, vcpu_id))
+		return false;
+
+	kvm_update_event_stats(&event->vcpu[vcpu_id], time_diff);
+	return true;
+}
+
+static bool is_child_event(struct perf_kvm_stat *kvm,
+			   struct perf_evsel *evsel,
+			   struct perf_sample *sample,
+			   struct event_key *key)
+{
+	struct child_event_ops *child_ops;
+
+	child_ops = kvm->events_ops->child_ops;
+
+	if (!child_ops)
+		return false;
+
+	for (; child_ops->name; child_ops++) {
+		if (!strcmp(evsel->name, child_ops->name)) {
+			child_ops->get_key(evsel, sample, key);
+			return true;
+		}
+	}
+
+	return false;
+}
+
+static bool handle_child_event(struct perf_kvm_stat *kvm,
+			       struct vcpu_event_record *vcpu_record,
+			       struct event_key *key,
+			       struct perf_sample *sample __maybe_unused)
+{
+	struct kvm_event *event = NULL;
+
+	if (key->key != INVALID_KEY)
+		event = find_create_kvm_event(kvm, key);
+
+	vcpu_record->last_event = event;
+
+	return true;
+}
+
+static bool skip_event(const char *event)
+{
+	const char * const *skip_events;
+
+	for (skip_events = kvm_skip_events; *skip_events; skip_events++)
+		if (!strcmp(event, *skip_events))
+			return true;
+
+	return false;
+}
+
+static bool handle_end_event(struct perf_kvm_stat *kvm,
+			     struct vcpu_event_record *vcpu_record,
+			     struct event_key *key,
+			     struct perf_sample *sample)
+{
+	struct kvm_event *event;
+	u64 time_begin, time_diff;
+	int vcpu;
+
+	if (kvm->trace_vcpu == -1)
+		vcpu = -1;
+	else
+		vcpu = vcpu_record->vcpu_id;
+
+	event = vcpu_record->last_event;
+	time_begin = vcpu_record->start_time;
+
+	/* The begin event is not caught. */
+	if (!time_begin)
+		return true;
+
+	/*
+	 * In some case, the 'begin event' only records the start timestamp,
+	 * the actual event is recognized in the 'end event' (e.g. mmio-event).
+	 */
+
+	/* Both begin and end events did not get the key. */
+	if (!event && key->key == INVALID_KEY)
+		return true;
+
+	if (!event)
+		event = find_create_kvm_event(kvm, key);
+
+	if (!event)
+		return false;
+
+	vcpu_record->last_event = NULL;
+	vcpu_record->start_time = 0;
+
+	/* seems to happen once in a while during live mode */
+	if (sample->time < time_begin) {
+		pr_debug("End time before begin time; skipping event.\n");
+		return true;
+	}
+
+	time_diff = sample->time - time_begin;
+
+	if (kvm->duration && time_diff > kvm->duration) {
+		char decode[DECODE_STR_LEN];
+
+		kvm->events_ops->decode_key(kvm, &event->key, decode);
+		if (!skip_event(decode)) {
+			pr_info("%" PRIu64 " VM %d, vcpu %d: %s event took %" PRIu64 "usec\n",
+				 sample->time, sample->pid, vcpu_record->vcpu_id,
+				 decode, time_diff/1000);
+		}
+	}
+
+	return update_kvm_event(event, vcpu, time_diff);
+}
+
+static
+struct vcpu_event_record *per_vcpu_record(struct thread *thread,
+					  struct perf_evsel *evsel,
+					  struct perf_sample *sample)
+{
+	/* Only kvm_entry records vcpu id. */
+	if (!thread__priv(thread) && kvm_entry_event(evsel)) {
+		struct vcpu_event_record *vcpu_record;
+
+		vcpu_record = zalloc(sizeof(*vcpu_record));
+		if (!vcpu_record) {
+			pr_err("%s: Not enough memory\n", __func__);
+			return NULL;
+		}
+
+		vcpu_record->vcpu_id = perf_evsel__intval(evsel, sample, VCPU_ID);
+		thread__set_priv(thread, vcpu_record);
+	}
+
+	return thread__priv(thread);
+}
+
+static bool handle_kvm_event(struct perf_kvm_stat *kvm,
+			     struct thread *thread,
+			     struct perf_evsel *evsel,
+			     struct perf_sample *sample)
+{
+	struct vcpu_event_record *vcpu_record;
+	struct event_key key = { .key = INVALID_KEY,
+				 .exit_reasons = kvm->exit_reasons };
+
+	vcpu_record = per_vcpu_record(thread, evsel, sample);
+	if (!vcpu_record)
+		return true;
+
+	/* only process events for vcpus user cares about */
+	if ((kvm->trace_vcpu != -1) &&
+	    (kvm->trace_vcpu != vcpu_record->vcpu_id))
+		return true;
+
+	if (kvm->events_ops->is_begin_event(evsel, sample, &key))
+		return handle_begin_event(kvm, vcpu_record, &key, sample->time);
+
+	if (is_child_event(kvm, evsel, sample, &key))
+		return handle_child_event(kvm, vcpu_record, &key, sample);
+
+	if (kvm->events_ops->is_end_event(evsel, sample, &key))
+		return handle_end_event(kvm, vcpu_record, &key, sample);
+
+	return true;
+}
+
+#define GET_EVENT_KEY(func, field)					\
+static u64 get_event_ ##func(struct kvm_event *event, int vcpu)		\
+{									\
+	if (vcpu == -1)							\
+		return event->total.field;				\
+									\
+	if (vcpu >= event->max_vcpu)					\
+		return 0;						\
+									\
+	return event->vcpu[vcpu].field;					\
+}
+
+#define COMPARE_EVENT_KEY(func, field)					\
+GET_EVENT_KEY(func, field)						\
+static int compare_kvm_event_ ## func(struct kvm_event *one,		\
+					struct kvm_event *two, int vcpu)\
+{									\
+	return get_event_ ##func(one, vcpu) >				\
+				get_event_ ##func(two, vcpu);		\
+}
+
+GET_EVENT_KEY(time, time);
+COMPARE_EVENT_KEY(count, stats.n);
+COMPARE_EVENT_KEY(mean, stats.mean);
+GET_EVENT_KEY(max, stats.max);
+GET_EVENT_KEY(min, stats.min);
+
+#define DEF_SORT_NAME_KEY(name, compare_key)				\
+	{ #name, compare_kvm_event_ ## compare_key }
+
+static struct kvm_event_key keys[] = {
+	DEF_SORT_NAME_KEY(sample, count),
+	DEF_SORT_NAME_KEY(time, mean),
+	{ NULL, NULL }
+};
+
+static bool select_key(struct perf_kvm_stat *kvm)
+{
+	int i;
+
+	for (i = 0; keys[i].name; i++) {
+		if (!strcmp(keys[i].name, kvm->sort_key)) {
+			kvm->compare = keys[i].key;
+			return true;
+		}
+	}
+
+	pr_err("Unknown compare key:%s\n", kvm->sort_key);
+	return false;
+}
+
+static void insert_to_result(struct rb_root *result, struct kvm_event *event,
+			     key_cmp_fun bigger, int vcpu)
+{
+	struct rb_node **rb = &result->rb_node;
+	struct rb_node *parent = NULL;
+	struct kvm_event *p;
+
+	while (*rb) {
+		p = container_of(*rb, struct kvm_event, rb);
+		parent = *rb;
+
+		if (bigger(event, p, vcpu))
+			rb = &(*rb)->rb_left;
+		else
+			rb = &(*rb)->rb_right;
+	}
+
+	rb_link_node(&event->rb, parent, rb);
+	rb_insert_color(&event->rb, result);
+}
+
+static void
+update_total_count(struct perf_kvm_stat *kvm, struct kvm_event *event)
+{
+	int vcpu = kvm->trace_vcpu;
+
+	kvm->total_count += get_event_count(event, vcpu);
+	kvm->total_time += get_event_time(event, vcpu);
+}
+
+static bool event_is_valid(struct kvm_event *event, int vcpu)
+{
+	return !!get_event_count(event, vcpu);
+}
+
+static void sort_result(struct perf_kvm_stat *kvm)
+{
+	unsigned int i;
+	int vcpu = kvm->trace_vcpu;
+	struct kvm_event *event;
+
+	for (i = 0; i < EVENTS_CACHE_SIZE; i++) {
+		list_for_each_entry(event, &kvm->kvm_events_cache[i], hash_entry) {
+			if (event_is_valid(event, vcpu)) {
+				update_total_count(kvm, event);
+				insert_to_result(&kvm->result, event,
+						 kvm->compare, vcpu);
+			}
+		}
+	}
+}
+
+/* returns left most element of result, and erase it */
+static struct kvm_event *pop_from_result(struct rb_root *result)
+{
+	struct rb_node *node = rb_first(result);
+
+	if (!node)
+		return NULL;
+
+	rb_erase(node, result);
+	return container_of(node, struct kvm_event, rb);
+}
+
+static void print_vcpu_info(struct perf_kvm_stat *kvm)
+{
+	int vcpu = kvm->trace_vcpu;
+
+	pr_info("Analyze events for ");
+
+	if (kvm->opts.target.system_wide)
+		pr_info("all VMs, ");
+	else if (kvm->opts.target.pid)
+		pr_info("pid(s) %s, ", kvm->opts.target.pid);
+	else
+		pr_info("dazed and confused on what is monitored, ");
+
+	if (vcpu == -1)
+		pr_info("all VCPUs:\n\n");
+	else
+		pr_info("VCPU %d:\n\n", vcpu);
+}
+
+static void show_timeofday(void)
+{
+	char date[64];
+	struct timeval tv;
+	struct tm ltime;
+
+	gettimeofday(&tv, NULL);
+	if (localtime_r(&tv.tv_sec, &ltime)) {
+		strftime(date, sizeof(date), "%H:%M:%S", &ltime);
+		pr_info("%s.%06ld", date, tv.tv_usec);
+	} else
+		pr_info("00:00:00.000000");
+
+	return;
+}
+
+static void print_result(struct perf_kvm_stat *kvm)
+{
+	char decode[DECODE_STR_LEN];
+	struct kvm_event *event;
+	int vcpu = kvm->trace_vcpu;
+
+	if (kvm->live) {
+		puts(CONSOLE_CLEAR);
+		show_timeofday();
+	}
+
+	pr_info("\n\n");
+	print_vcpu_info(kvm);
+	pr_info("%*s ", DECODE_STR_LEN, kvm->events_ops->name);
+	pr_info("%10s ", "Samples");
+	pr_info("%9s ", "Samples%");
+
+	pr_info("%9s ", "Time%");
+	pr_info("%11s ", "Min Time");
+	pr_info("%11s ", "Max Time");
+	pr_info("%16s ", "Avg time");
+	pr_info("\n\n");
+
+	while ((event = pop_from_result(&kvm->result))) {
+		u64 ecount, etime, max, min;
+
+		ecount = get_event_count(event, vcpu);
+		etime = get_event_time(event, vcpu);
+		max = get_event_max(event, vcpu);
+		min = get_event_min(event, vcpu);
+
+		kvm->events_ops->decode_key(kvm, &event->key, decode);
+		pr_info("%*s ", DECODE_STR_LEN, decode);
+		pr_info("%10llu ", (unsigned long long)ecount);
+		pr_info("%8.2f%% ", (double)ecount / kvm->total_count * 100);
+		pr_info("%8.2f%% ", (double)etime / kvm->total_time * 100);
+		pr_info("%9.2fus ", (double)min / 1e3);
+		pr_info("%9.2fus ", (double)max / 1e3);
+		pr_info("%9.2fus ( +-%7.2f%% )", (double)etime / ecount/1e3,
+			kvm_event_rel_stddev(vcpu, event));
+		pr_info("\n");
+	}
+
+	pr_info("\nTotal Samples:%" PRIu64 ", Total events handled time:%.2fus.\n\n",
+		kvm->total_count, kvm->total_time / 1e3);
+
+	if (kvm->lost_events)
+		pr_info("\nLost events: %" PRIu64 "\n\n", kvm->lost_events);
+}
+
+#ifdef HAVE_TIMERFD_SUPPORT
+static int process_lost_event(struct perf_tool *tool,
+			      union perf_event *event __maybe_unused,
+			      struct perf_sample *sample __maybe_unused,
+			      struct machine *machine __maybe_unused)
+{
+	struct perf_kvm_stat *kvm = container_of(tool, struct perf_kvm_stat, tool);
+
+	kvm->lost_events++;
+	return 0;
+}
+#endif
+
+static bool skip_sample(struct perf_kvm_stat *kvm,
+			struct perf_sample *sample)
+{
+	if (kvm->pid_list && intlist__find(kvm->pid_list, sample->pid) == NULL)
+		return true;
+
+	return false;
+}
+
+static int process_sample_event(struct perf_tool *tool,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel,
+				struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_kvm_stat *kvm = container_of(tool, struct perf_kvm_stat,
+						 tool);
+
+	if (skip_sample(kvm, sample))
+		return 0;
+
+	thread = machine__findnew_thread(machine, sample->pid, sample->tid);
+	if (thread == NULL) {
+		pr_debug("problem processing %d event, skipping it.\n",
+			event->header.type);
+		return -1;
+	}
+
+	if (!handle_kvm_event(kvm, thread, evsel, sample))
+		return -1;
+
+	return 0;
+}
+
+static int cpu_isa_config(struct perf_kvm_stat *kvm)
+{
+	char buf[64], *cpuid;
+	int err;
+
+	if (kvm->live) {
+		err = get_cpuid(buf, sizeof(buf));
+		if (err != 0) {
+			pr_err("Failed to look up CPU type\n");
+			return err;
+		}
+		cpuid = buf;
+	} else
+		cpuid = kvm->session->header.env.cpuid;
+
+	if (!cpuid) {
+		pr_err("Failed to look up CPU type\n");
+		return -EINVAL;
+	}
+
+	err = cpu_isa_init(kvm, cpuid);
+	if (err == -ENOTSUP)
+		pr_err("CPU %s is not supported.\n", cpuid);
+
+	return err;
+}
+
+static bool verify_vcpu(int vcpu)
+{
+	if (vcpu != -1 && vcpu < 0) {
+		pr_err("Invalid vcpu:%d.\n", vcpu);
+		return false;
+	}
+
+	return true;
+}
+
+#ifdef HAVE_TIMERFD_SUPPORT
+/* keeping the max events to a modest level to keep
+ * the processing of samples per mmap smooth.
+ */
+#define PERF_KVM__MAX_EVENTS_PER_MMAP  25
+
+static s64 perf_kvm__mmap_read_idx(struct perf_kvm_stat *kvm, int idx,
+				   u64 *mmap_time)
+{
+	union perf_event *event;
+	struct perf_sample sample;
+	s64 n = 0;
+	int err;
+
+	*mmap_time = ULLONG_MAX;
+	while ((event = perf_evlist__mmap_read(kvm->evlist, idx)) != NULL) {
+		err = perf_evlist__parse_sample(kvm->evlist, event, &sample);
+		if (err) {
+			perf_evlist__mmap_consume(kvm->evlist, idx);
+			pr_err("Failed to parse sample\n");
+			return -1;
+		}
+
+		err = perf_session__queue_event(kvm->session, event, &sample, 0);
+		/*
+		 * FIXME: Here we can't consume the event, as perf_session__queue_event will
+		 *        point to it, and it'll get possibly overwritten by the kernel.
+		 */
+		perf_evlist__mmap_consume(kvm->evlist, idx);
+
+		if (err) {
+			pr_err("Failed to enqueue sample: %d\n", err);
+			return -1;
+		}
+
+		/* save time stamp of our first sample for this mmap */
+		if (n == 0)
+			*mmap_time = sample.time;
+
+		/* limit events per mmap handled all at once */
+		n++;
+		if (n == PERF_KVM__MAX_EVENTS_PER_MMAP)
+			break;
+	}
+
+	return n;
+}
+
+static int perf_kvm__mmap_read(struct perf_kvm_stat *kvm)
+{
+	int i, err, throttled = 0;
+	s64 n, ntotal = 0;
+	u64 flush_time = ULLONG_MAX, mmap_time;
+
+	for (i = 0; i < kvm->evlist->nr_mmaps; i++) {
+		n = perf_kvm__mmap_read_idx(kvm, i, &mmap_time);
+		if (n < 0)
+			return -1;
+
+		/* flush time is going to be the minimum of all the individual
+		 * mmap times. Essentially, we flush all the samples queued up
+		 * from the last pass under our minimal start time -- that leaves
+		 * a very small race for samples to come in with a lower timestamp.
+		 * The ioctl to return the perf_clock timestamp should close the
+		 * race entirely.
+		 */
+		if (mmap_time < flush_time)
+			flush_time = mmap_time;
+
+		ntotal += n;
+		if (n == PERF_KVM__MAX_EVENTS_PER_MMAP)
+			throttled = 1;
+	}
+
+	/* flush queue after each round in which we processed events */
+	if (ntotal) {
+		struct ordered_events *oe = &kvm->session->ordered_events;
+
+		oe->next_flush = flush_time;
+		err = ordered_events__flush(oe, OE_FLUSH__ROUND);
+		if (err) {
+			if (kvm->lost_events)
+				pr_info("\nLost events: %" PRIu64 "\n\n",
+					kvm->lost_events);
+			return err;
+		}
+	}
+
+	return throttled;
+}
+
+static volatile int done;
+
+static void sig_handler(int sig __maybe_unused)
+{
+	done = 1;
+}
+
+static int perf_kvm__timerfd_create(struct perf_kvm_stat *kvm)
+{
+	struct itimerspec new_value;
+	int rc = -1;
+
+	kvm->timerfd = timerfd_create(CLOCK_MONOTONIC, TFD_NONBLOCK);
+	if (kvm->timerfd < 0) {
+		pr_err("timerfd_create failed\n");
+		goto out;
+	}
+
+	new_value.it_value.tv_sec = kvm->display_time;
+	new_value.it_value.tv_nsec = 0;
+	new_value.it_interval.tv_sec = kvm->display_time;
+	new_value.it_interval.tv_nsec = 0;
+
+	if (timerfd_settime(kvm->timerfd, 0, &new_value, NULL) != 0) {
+		pr_err("timerfd_settime failed: %d\n", errno);
+		close(kvm->timerfd);
+		goto out;
+	}
+
+	rc = 0;
+out:
+	return rc;
+}
+
+static int perf_kvm__handle_timerfd(struct perf_kvm_stat *kvm)
+{
+	uint64_t c;
+	int rc;
+
+	rc = read(kvm->timerfd, &c, sizeof(uint64_t));
+	if (rc < 0) {
+		if (errno == EAGAIN)
+			return 0;
+
+		pr_err("Failed to read timer fd: %d\n", errno);
+		return -1;
+	}
+
+	if (rc != sizeof(uint64_t)) {
+		pr_err("Error reading timer fd - invalid size returned\n");
+		return -1;
+	}
+
+	if (c != 1)
+		pr_debug("Missed timer beats: %" PRIu64 "\n", c-1);
+
+	/* update display */
+	sort_result(kvm);
+	print_result(kvm);
+
+	/* reset counts */
+	clear_events_cache_stats(kvm->kvm_events_cache);
+	kvm->total_count = 0;
+	kvm->total_time = 0;
+	kvm->lost_events = 0;
+
+	return 0;
+}
+
+static int fd_set_nonblock(int fd)
+{
+	long arg = 0;
+
+	arg = fcntl(fd, F_GETFL);
+	if (arg < 0) {
+		pr_err("Failed to get current flags for fd %d\n", fd);
+		return -1;
+	}
+
+	if (fcntl(fd, F_SETFL, arg | O_NONBLOCK) < 0) {
+		pr_err("Failed to set non-block option on fd %d\n", fd);
+		return -1;
+	}
+
+	return 0;
+}
+
+static int perf_kvm__handle_stdin(void)
+{
+	int c;
+
+	c = getc(stdin);
+	if (c == 'q')
+		return 1;
+
+	return 0;
+}
+
+static int kvm_events_live_report(struct perf_kvm_stat *kvm)
+{
+	int nr_stdin, ret, err = -EINVAL;
+	struct termios save;
+
+	/* live flag must be set first */
+	kvm->live = true;
+
+	ret = cpu_isa_config(kvm);
+	if (ret < 0)
+		return ret;
+
+	if (!verify_vcpu(kvm->trace_vcpu) ||
+	    !select_key(kvm) ||
+	    !register_kvm_events_ops(kvm)) {
+		goto out;
+	}
+
+	set_term_quiet_input(&save);
+	init_kvm_event_record(kvm);
+
+	signal(SIGINT, sig_handler);
+	signal(SIGTERM, sig_handler);
+
+	/* add timer fd */
+	if (perf_kvm__timerfd_create(kvm) < 0) {
+		err = -1;
+		goto out;
+	}
+
+	if (perf_evlist__add_pollfd(kvm->evlist, kvm->timerfd) < 0)
+		goto out;
+
+	nr_stdin = perf_evlist__add_pollfd(kvm->evlist, fileno(stdin));
+	if (nr_stdin < 0)
+		goto out;
+
+	if (fd_set_nonblock(fileno(stdin)) != 0)
+		goto out;
+
+	/* everything is good - enable the events and process */
+	perf_evlist__enable(kvm->evlist);
+
+	while (!done) {
+		struct fdarray *fda = &kvm->evlist->pollfd;
+		int rc;
+
+		rc = perf_kvm__mmap_read(kvm);
+		if (rc < 0)
+			break;
+
+		err = perf_kvm__handle_timerfd(kvm);
+		if (err)
+			goto out;
+
+		if (fda->entries[nr_stdin].revents & POLLIN)
+			done = perf_kvm__handle_stdin();
+
+		if (!rc && !done)
+			err = fdarray__poll(fda, 100);
+	}
+
+	perf_evlist__disable(kvm->evlist);
+
+	if (err == 0) {
+		sort_result(kvm);
+		print_result(kvm);
+	}
+
+out:
+	if (kvm->timerfd >= 0)
+		close(kvm->timerfd);
+
+	tcsetattr(0, TCSAFLUSH, &save);
+	return err;
+}
+
+static int kvm_live_open_events(struct perf_kvm_stat *kvm)
+{
+	int err, rc = -1;
+	struct perf_evsel *pos;
+	struct perf_evlist *evlist = kvm->evlist;
+	char sbuf[STRERR_BUFSIZE];
+
+	perf_evlist__config(evlist, &kvm->opts);
+
+	/*
+	 * Note: exclude_{guest,host} do not apply here.
+	 *       This command processes KVM tracepoints from host only
+	 */
+	evlist__for_each(evlist, pos) {
+		struct perf_event_attr *attr = &pos->attr;
+
+		/* make sure these *are* set */
+		perf_evsel__set_sample_bit(pos, TID);
+		perf_evsel__set_sample_bit(pos, TIME);
+		perf_evsel__set_sample_bit(pos, CPU);
+		perf_evsel__set_sample_bit(pos, RAW);
+		/* make sure these are *not*; want as small a sample as possible */
+		perf_evsel__reset_sample_bit(pos, PERIOD);
+		perf_evsel__reset_sample_bit(pos, IP);
+		perf_evsel__reset_sample_bit(pos, CALLCHAIN);
+		perf_evsel__reset_sample_bit(pos, ADDR);
+		perf_evsel__reset_sample_bit(pos, READ);
+		attr->mmap = 0;
+		attr->comm = 0;
+		attr->task = 0;
+
+		attr->sample_period = 1;
+
+		attr->watermark = 0;
+		attr->wakeup_events = 1000;
+
+		/* will enable all once we are ready */
+		attr->disabled = 1;
+	}
+
+	err = perf_evlist__open(evlist);
+	if (err < 0) {
+		printf("Couldn't create the events: %s\n",
+		       strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out;
+	}
+
+	if (perf_evlist__mmap(evlist, kvm->opts.mmap_pages, false) < 0) {
+		ui__error("Failed to mmap the events: %s\n",
+			  strerror_r(errno, sbuf, sizeof(sbuf)));
+		perf_evlist__close(evlist);
+		goto out;
+	}
+
+	rc = 0;
+
+out:
+	return rc;
+}
+#endif
+
+static int read_events(struct perf_kvm_stat *kvm)
+{
+	int ret;
+
+	struct perf_tool eops = {
+		.sample			= process_sample_event,
+		.comm			= perf_event__process_comm,
+		.ordered_events		= true,
+	};
+	struct perf_data_file file = {
+		.path = kvm->file_name,
+		.mode = PERF_DATA_MODE_READ,
+		.force = kvm->force,
+	};
+
+	kvm->tool = eops;
+	kvm->session = perf_session__new(&file, false, &kvm->tool);
+	if (!kvm->session) {
+		pr_err("Initializing perf session failed\n");
+		return -1;
+	}
+
+	symbol__init(&kvm->session->header.env);
+
+	if (!perf_session__has_traces(kvm->session, "kvm record"))
+		return -EINVAL;
+
+	/*
+	 * Do not use 'isa' recorded in kvm_exit tracepoint since it is not
+	 * traced in the old kernel.
+	 */
+	ret = cpu_isa_config(kvm);
+	if (ret < 0)
+		return ret;
+
+	return perf_session__process_events(kvm->session);
+}
+
+static int parse_target_str(struct perf_kvm_stat *kvm)
+{
+	if (kvm->opts.target.pid) {
+		kvm->pid_list = intlist__new(kvm->opts.target.pid);
+		if (kvm->pid_list == NULL) {
+			pr_err("Error parsing process id string\n");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int kvm_events_report_vcpu(struct perf_kvm_stat *kvm)
+{
+	int ret = -EINVAL;
+	int vcpu = kvm->trace_vcpu;
+
+	if (parse_target_str(kvm) != 0)
+		goto exit;
+
+	if (!verify_vcpu(vcpu))
+		goto exit;
+
+	if (!select_key(kvm))
+		goto exit;
+
+	if (!register_kvm_events_ops(kvm))
+		goto exit;
+
+	init_kvm_event_record(kvm);
+	setup_pager();
+
+	ret = read_events(kvm);
+	if (ret)
+		goto exit;
+
+	sort_result(kvm);
+	print_result(kvm);
+
+exit:
+	return ret;
+}
+
+#define STRDUP_FAIL_EXIT(s)		\
+	({	char *_p;		\
+	_p = strdup(s);		\
+		if (!_p)		\
+			return -ENOMEM;	\
+		_p;			\
+	})
+
+static int
+kvm_events_record(struct perf_kvm_stat *kvm, int argc, const char **argv)
+{
+	unsigned int rec_argc, i, j, events_tp_size;
+	const char **rec_argv;
+	const char * const record_args[] = {
+		"record",
+		"-R",
+		"-m", "1024",
+		"-c", "1",
+	};
+	const char * const kvm_stat_record_usage[] = {
+		"perf kvm stat record [<options>]",
+		NULL
+	};
+	const char * const *events_tp;
+	events_tp_size = 0;
+
+	for (events_tp = kvm_events_tp; *events_tp; events_tp++)
+		events_tp_size++;
+
+	rec_argc = ARRAY_SIZE(record_args) + argc + 2 +
+		   2 * events_tp_size;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+	if (rec_argv == NULL)
+		return -ENOMEM;
+
+	for (i = 0; i < ARRAY_SIZE(record_args); i++)
+		rec_argv[i] = STRDUP_FAIL_EXIT(record_args[i]);
+
+	for (j = 0; j < events_tp_size; j++) {
+		rec_argv[i++] = "-e";
+		rec_argv[i++] = STRDUP_FAIL_EXIT(kvm_events_tp[j]);
+	}
+
+	rec_argv[i++] = STRDUP_FAIL_EXIT("-o");
+	rec_argv[i++] = STRDUP_FAIL_EXIT(kvm->file_name);
+
+	for (j = 1; j < (unsigned int)argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	set_option_flag(record_options, 'e', "event", PARSE_OPT_HIDDEN);
+	set_option_flag(record_options, 0, "filter", PARSE_OPT_HIDDEN);
+	set_option_flag(record_options, 'R', "raw-samples", PARSE_OPT_HIDDEN);
+
+	set_option_flag(record_options, 'F', "freq", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 0, "group", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'g', NULL, PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 0, "call-graph", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'd', "data", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'T', "timestamp", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'P', "period", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'n', "no-samples", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'N', "no-buildid-cache", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'B', "no-buildid", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'G', "cgroup", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'b', "branch-any", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'j', "branch-filter", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 'W', "weight", PARSE_OPT_DISABLED);
+	set_option_flag(record_options, 0, "transaction", PARSE_OPT_DISABLED);
+
+	record_usage = kvm_stat_record_usage;
+	return cmd_record(i, rec_argv, NULL);
+}
+
+static int
+kvm_events_report(struct perf_kvm_stat *kvm, int argc, const char **argv)
+{
+	const struct option kvm_events_report_options[] = {
+		OPT_STRING(0, "event", &kvm->report_event, "report event",
+			   "event for reporting: vmexit, "
+			   "mmio (x86 only), ioport (x86 only)"),
+		OPT_INTEGER(0, "vcpu", &kvm->trace_vcpu,
+			    "vcpu id to report"),
+		OPT_STRING('k', "key", &kvm->sort_key, "sort-key",
+			    "key for sorting: sample(sort by samples number)"
+			    " time (sort by avg time)"),
+		OPT_STRING('p', "pid", &kvm->opts.target.pid, "pid",
+			   "analyze events only for given process id(s)"),
+		OPT_BOOLEAN('f', "force", &kvm->force, "don't complain, do it"),
+		OPT_END()
+	};
+
+	const char * const kvm_events_report_usage[] = {
+		"perf kvm stat report [<options>]",
+		NULL
+	};
+
+	if (argc) {
+		argc = parse_options(argc, argv,
+				     kvm_events_report_options,
+				     kvm_events_report_usage, 0);
+		if (argc)
+			usage_with_options(kvm_events_report_usage,
+					   kvm_events_report_options);
+	}
+
+	if (!kvm->opts.target.pid)
+		kvm->opts.target.system_wide = true;
+
+	return kvm_events_report_vcpu(kvm);
+}
+
+#ifdef HAVE_TIMERFD_SUPPORT
+static struct perf_evlist *kvm_live_event_list(void)
+{
+	struct perf_evlist *evlist;
+	char *tp, *name, *sys;
+	int err = -1;
+	const char * const *events_tp;
+
+	evlist = perf_evlist__new();
+	if (evlist == NULL)
+		return NULL;
+
+	for (events_tp = kvm_events_tp; *events_tp; events_tp++) {
+
+		tp = strdup(*events_tp);
+		if (tp == NULL)
+			goto out;
+
+		/* split tracepoint into subsystem and name */
+		sys = tp;
+		name = strchr(tp, ':');
+		if (name == NULL) {
+			pr_err("Error parsing %s tracepoint: subsystem delimiter not found\n",
+			       *events_tp);
+			free(tp);
+			goto out;
+		}
+		*name = '\0';
+		name++;
+
+		if (perf_evlist__add_newtp(evlist, sys, name, NULL)) {
+			pr_err("Failed to add %s tracepoint to the list\n", *events_tp);
+			free(tp);
+			goto out;
+		}
+
+		free(tp);
+	}
+
+	err = 0;
+
+out:
+	if (err) {
+		perf_evlist__delete(evlist);
+		evlist = NULL;
+	}
+
+	return evlist;
+}
+
+static int kvm_events_live(struct perf_kvm_stat *kvm,
+			   int argc, const char **argv)
+{
+	char errbuf[BUFSIZ];
+	int err;
+
+	const struct option live_options[] = {
+		OPT_STRING('p', "pid", &kvm->opts.target.pid, "pid",
+			"record events on existing process id"),
+		OPT_CALLBACK('m', "mmap-pages", &kvm->opts.mmap_pages, "pages",
+			"number of mmap data pages",
+			perf_evlist__parse_mmap_pages),
+		OPT_INCR('v', "verbose", &verbose,
+			"be more verbose (show counter open errors, etc)"),
+		OPT_BOOLEAN('a', "all-cpus", &kvm->opts.target.system_wide,
+			"system-wide collection from all CPUs"),
+		OPT_UINTEGER('d', "display", &kvm->display_time,
+			"time in seconds between display updates"),
+		OPT_STRING(0, "event", &kvm->report_event, "report event",
+			"event for reporting: "
+			"vmexit, mmio (x86 only), ioport (x86 only)"),
+		OPT_INTEGER(0, "vcpu", &kvm->trace_vcpu,
+			"vcpu id to report"),
+		OPT_STRING('k', "key", &kvm->sort_key, "sort-key",
+			"key for sorting: sample(sort by samples number)"
+			" time (sort by avg time)"),
+		OPT_U64(0, "duration", &kvm->duration,
+			"show events other than"
+			" HLT (x86 only) or Wait state (s390 only)"
+			" that take longer than duration usecs"),
+		OPT_END()
+	};
+	const char * const live_usage[] = {
+		"perf kvm stat live [<options>]",
+		NULL
+	};
+	struct perf_data_file file = {
+		.mode = PERF_DATA_MODE_WRITE,
+	};
+
+
+	/* event handling */
+	kvm->tool.sample = process_sample_event;
+	kvm->tool.comm   = perf_event__process_comm;
+	kvm->tool.exit   = perf_event__process_exit;
+	kvm->tool.fork   = perf_event__process_fork;
+	kvm->tool.lost   = process_lost_event;
+	kvm->tool.ordered_events = true;
+	perf_tool__fill_defaults(&kvm->tool);
+
+	/* set defaults */
+	kvm->display_time = 1;
+	kvm->opts.user_interval = 1;
+	kvm->opts.mmap_pages = 512;
+	kvm->opts.target.uses_mmap = false;
+	kvm->opts.target.uid_str = NULL;
+	kvm->opts.target.uid = UINT_MAX;
+
+	symbol__init(NULL);
+	disable_buildid_cache();
+
+	use_browser = 0;
+	setup_browser(false);
+
+	if (argc) {
+		argc = parse_options(argc, argv, live_options,
+				     live_usage, 0);
+		if (argc)
+			usage_with_options(live_usage, live_options);
+	}
+
+	kvm->duration *= NSEC_PER_USEC;   /* convert usec to nsec */
+
+	/*
+	 * target related setups
+	 */
+	err = target__validate(&kvm->opts.target);
+	if (err) {
+		target__strerror(&kvm->opts.target, err, errbuf, BUFSIZ);
+		ui__warning("%s", errbuf);
+	}
+
+	if (target__none(&kvm->opts.target))
+		kvm->opts.target.system_wide = true;
+
+
+	/*
+	 * generate the event list
+	 */
+	kvm->evlist = kvm_live_event_list();
+	if (kvm->evlist == NULL) {
+		err = -1;
+		goto out;
+	}
+
+	symbol_conf.nr_events = kvm->evlist->nr_entries;
+
+	if (perf_evlist__create_maps(kvm->evlist, &kvm->opts.target) < 0)
+		usage_with_options(live_usage, live_options);
+
+	/*
+	 * perf session
+	 */
+	kvm->session = perf_session__new(&file, false, &kvm->tool);
+	if (kvm->session == NULL) {
+		err = -1;
+		goto out;
+	}
+	kvm->session->evlist = kvm->evlist;
+	perf_session__set_id_hdr_size(kvm->session);
+	ordered_events__set_copy_on_queue(&kvm->session->ordered_events, true);
+	machine__synthesize_threads(&kvm->session->machines.host, &kvm->opts.target,
+				    kvm->evlist->threads, false);
+	err = kvm_live_open_events(kvm);
+	if (err)
+		goto out;
+
+	err = kvm_events_live_report(kvm);
+
+out:
+	exit_browser(0);
+
+	if (kvm->session)
+		perf_session__delete(kvm->session);
+	kvm->session = NULL;
+	if (kvm->evlist)
+		perf_evlist__delete(kvm->evlist);
+
+	return err;
+}
+#endif
+
+static void print_kvm_stat_usage(void)
+{
+	printf("Usage: perf kvm stat <command>\n\n");
+
+	printf("# Available commands:\n");
+	printf("\trecord: record kvm events\n");
+	printf("\treport: report statistical data of kvm events\n");
+	printf("\tlive:   live reporting of statistical data of kvm events\n");
+
+	printf("\nOtherwise, it is the alias of 'perf stat':\n");
+}
+
+static int kvm_cmd_stat(const char *file_name, int argc, const char **argv)
+{
+	struct perf_kvm_stat kvm = {
+		.file_name = file_name,
+
+		.trace_vcpu	= -1,
+		.report_event	= "vmexit",
+		.sort_key	= "sample",
+
+	};
+
+	if (argc == 1) {
+		print_kvm_stat_usage();
+		goto perf_stat;
+	}
+
+	if (!strncmp(argv[1], "rec", 3))
+		return kvm_events_record(&kvm, argc - 1, argv + 1);
+
+	if (!strncmp(argv[1], "rep", 3))
+		return kvm_events_report(&kvm, argc - 1 , argv + 1);
+
+#ifdef HAVE_TIMERFD_SUPPORT
+	if (!strncmp(argv[1], "live", 4))
+		return kvm_events_live(&kvm, argc - 1 , argv + 1);
+#endif
+
+perf_stat:
+	return cmd_stat(argc, argv, NULL);
+}
+#endif /* HAVE_KVM_STAT_SUPPORT */
+
+static int __cmd_record(const char *file_name, int argc, const char **argv)
+{
+	int rec_argc, i = 0, j;
+	const char **rec_argv;
+
+	rec_argc = argc + 2;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+	rec_argv[i++] = strdup("record");
+	rec_argv[i++] = strdup("-o");
+	rec_argv[i++] = strdup(file_name);
+	for (j = 1; j < argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	return cmd_record(i, rec_argv, NULL);
+}
+
+static int __cmd_report(const char *file_name, int argc, const char **argv)
+{
+	int rec_argc, i = 0, j;
+	const char **rec_argv;
+
+	rec_argc = argc + 2;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+	rec_argv[i++] = strdup("report");
+	rec_argv[i++] = strdup("-i");
+	rec_argv[i++] = strdup(file_name);
+	for (j = 1; j < argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	return cmd_report(i, rec_argv, NULL);
+}
+
+static int
+__cmd_buildid_list(const char *file_name, int argc, const char **argv)
+{
+	int rec_argc, i = 0, j;
+	const char **rec_argv;
+
+	rec_argc = argc + 2;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+	rec_argv[i++] = strdup("buildid-list");
+	rec_argv[i++] = strdup("-i");
+	rec_argv[i++] = strdup(file_name);
+	for (j = 1; j < argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	return cmd_buildid_list(i, rec_argv, NULL);
+}
+
+int cmd_kvm(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	const char *file_name = NULL;
+	const struct option kvm_options[] = {
+		OPT_STRING('i', "input", &file_name, "file",
+			   "Input file name"),
+		OPT_STRING('o', "output", &file_name, "file",
+			   "Output file name"),
+		OPT_BOOLEAN(0, "guest", &perf_guest,
+			    "Collect guest os data"),
+		OPT_BOOLEAN(0, "host", &perf_host,
+			    "Collect host os data"),
+		OPT_STRING(0, "guestmount", &symbol_conf.guestmount, "directory",
+			   "guest mount directory under which every guest os"
+			   " instance has a subdir"),
+		OPT_STRING(0, "guestvmlinux", &symbol_conf.default_guest_vmlinux_name,
+			   "file", "file saving guest os vmlinux"),
+		OPT_STRING(0, "guestkallsyms", &symbol_conf.default_guest_kallsyms,
+			   "file", "file saving guest os /proc/kallsyms"),
+		OPT_STRING(0, "guestmodules", &symbol_conf.default_guest_modules,
+			   "file", "file saving guest os /proc/modules"),
+		OPT_INCR('v', "verbose", &verbose,
+			    "be more verbose (show counter open errors, etc)"),
+		OPT_END()
+	};
+
+	const char *const kvm_subcommands[] = { "top", "record", "report", "diff",
+						"buildid-list", "stat", NULL };
+	const char *kvm_usage[] = { NULL, NULL };
+
+	perf_host  = 0;
+	perf_guest = 1;
+
+	argc = parse_options_subcommand(argc, argv, kvm_options, kvm_subcommands, kvm_usage,
+					PARSE_OPT_STOP_AT_NON_OPTION);
+	if (!argc)
+		usage_with_options(kvm_usage, kvm_options);
+
+	if (!perf_host)
+		perf_guest = 1;
+
+	if (!file_name) {
+		file_name = get_filename_for_perf_kvm();
+
+		if (!file_name) {
+			pr_err("Failed to allocate memory for filename\n");
+			return -ENOMEM;
+		}
+	}
+
+	if (!strncmp(argv[0], "rec", 3))
+		return __cmd_record(file_name, argc, argv);
+	else if (!strncmp(argv[0], "rep", 3))
+		return __cmd_report(file_name, argc, argv);
+	else if (!strncmp(argv[0], "diff", 4))
+		return cmd_diff(argc, argv, NULL);
+	else if (!strncmp(argv[0], "top", 3))
+		return cmd_top(argc, argv, NULL);
+	else if (!strncmp(argv[0], "buildid-list", 12))
+		return __cmd_buildid_list(file_name, argc, argv);
+#ifdef HAVE_KVM_STAT_SUPPORT
+	else if (!strncmp(argv[0], "stat", 4))
+		return kvm_cmd_stat(file_name, argc, argv);
+#endif
+	else
+		usage_with_options(kvm_usage, kvm_options);
+
+	return 0;
+}
diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c
new file mode 100644
index 000000000..af5bd0514
--- /dev/null
+++ b/tools/perf/builtin-list.c
@@ -0,0 +1,82 @@
+/*
+ * builtin-list.c
+ *
+ * Builtin list command: list all event types
+ *
+ * Copyright (C) 2009, Thomas Gleixner <tglx@linutronix.de>
+ * Copyright (C) 2008-2009, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include "builtin.h"
+
+#include "perf.h"
+
+#include "util/parse-events.h"
+#include "util/cache.h"
+#include "util/pmu.h"
+#include "util/parse-options.h"
+
+int cmd_list(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	int i;
+	bool raw_dump = false;
+	struct option list_options[] = {
+		OPT_BOOLEAN(0, "raw-dump", &raw_dump, "Dump raw events"),
+		OPT_END()
+	};
+	const char * const list_usage[] = {
+		"perf list [hw|sw|cache|tracepoint|pmu|event_glob]",
+		NULL
+	};
+
+	set_option_flag(list_options, 0, "raw-dump", PARSE_OPT_HIDDEN);
+
+	argc = parse_options(argc, argv, list_options, list_usage,
+			     PARSE_OPT_STOP_AT_NON_OPTION);
+
+	setup_pager();
+
+	if (!raw_dump)
+		printf("\nList of pre-defined events (to be used in -e):\n\n");
+
+	if (argc == 0) {
+		print_events(NULL, raw_dump);
+		return 0;
+	}
+
+	for (i = 0; i < argc; ++i) {
+		if (strcmp(argv[i], "tracepoint") == 0)
+			print_tracepoint_events(NULL, NULL, raw_dump);
+		else if (strcmp(argv[i], "hw") == 0 ||
+			 strcmp(argv[i], "hardware") == 0)
+			print_symbol_events(NULL, PERF_TYPE_HARDWARE,
+					event_symbols_hw, PERF_COUNT_HW_MAX, raw_dump);
+		else if (strcmp(argv[i], "sw") == 0 ||
+			 strcmp(argv[i], "software") == 0)
+			print_symbol_events(NULL, PERF_TYPE_SOFTWARE,
+					event_symbols_sw, PERF_COUNT_SW_MAX, raw_dump);
+		else if (strcmp(argv[i], "cache") == 0 ||
+			 strcmp(argv[i], "hwcache") == 0)
+			print_hwcache_events(NULL, raw_dump);
+		else if (strcmp(argv[i], "pmu") == 0)
+			print_pmu_events(NULL, raw_dump);
+		else {
+			char *sep = strchr(argv[i], ':'), *s;
+			int sep_idx;
+
+			if (sep == NULL) {
+				print_events(argv[i], raw_dump);
+				continue;
+			}
+			sep_idx = sep - argv[i];
+			s = strdup(argv[i]);
+			if (s == NULL)
+				return -1;
+
+			s[sep_idx] = '\0';
+			print_tracepoint_events(s, s + sep_idx + 1, raw_dump);
+			free(s);
+		}
+	}
+	return 0;
+}
diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c
new file mode 100644
index 000000000..d49c2ab85
--- /dev/null
+++ b/tools/perf/builtin-lock.c
@@ -0,0 +1,1021 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+
+#include "util/debug.h"
+#include "util/session.h"
+#include "util/tool.h"
+#include "util/data.h"
+
+#include <sys/types.h>
+#include <sys/prctl.h>
+#include <semaphore.h>
+#include <pthread.h>
+#include <math.h>
+#include <limits.h>
+
+#include <linux/list.h>
+#include <linux/hash.h>
+
+static struct perf_session *session;
+
+/* based on kernel/lockdep.c */
+#define LOCKHASH_BITS		12
+#define LOCKHASH_SIZE		(1UL << LOCKHASH_BITS)
+
+static struct list_head lockhash_table[LOCKHASH_SIZE];
+
+#define __lockhashfn(key)	hash_long((unsigned long)key, LOCKHASH_BITS)
+#define lockhashentry(key)	(lockhash_table + __lockhashfn((key)))
+
+struct lock_stat {
+	struct list_head	hash_entry;
+	struct rb_node		rb;		/* used for sorting */
+
+	/*
+	 * FIXME: perf_evsel__intval() returns u64,
+	 * so address of lockdep_map should be dealed as 64bit.
+	 * Is there more better solution?
+	 */
+	void			*addr;		/* address of lockdep_map, used as ID */
+	char			*name;		/* for strcpy(), we cannot use const */
+
+	unsigned int		nr_acquire;
+	unsigned int		nr_acquired;
+	unsigned int		nr_contended;
+	unsigned int		nr_release;
+
+	unsigned int		nr_readlock;
+	unsigned int		nr_trylock;
+
+	/* these times are in nano sec. */
+	u64                     avg_wait_time;
+	u64			wait_time_total;
+	u64			wait_time_min;
+	u64			wait_time_max;
+
+	int			discard; /* flag of blacklist */
+};
+
+/*
+ * States of lock_seq_stat
+ *
+ * UNINITIALIZED is required for detecting first event of acquire.
+ * As the nature of lock events, there is no guarantee
+ * that the first event for the locks are acquire,
+ * it can be acquired, contended or release.
+ */
+#define SEQ_STATE_UNINITIALIZED      0	       /* initial state */
+#define SEQ_STATE_RELEASED	1
+#define SEQ_STATE_ACQUIRING	2
+#define SEQ_STATE_ACQUIRED	3
+#define SEQ_STATE_READ_ACQUIRED	4
+#define SEQ_STATE_CONTENDED	5
+
+/*
+ * MAX_LOCK_DEPTH
+ * Imported from include/linux/sched.h.
+ * Should this be synchronized?
+ */
+#define MAX_LOCK_DEPTH 48
+
+/*
+ * struct lock_seq_stat:
+ * Place to put on state of one lock sequence
+ * 1) acquire -> acquired -> release
+ * 2) acquire -> contended -> acquired -> release
+ * 3) acquire (with read or try) -> release
+ * 4) Are there other patterns?
+ */
+struct lock_seq_stat {
+	struct list_head        list;
+	int			state;
+	u64			prev_event_time;
+	void                    *addr;
+
+	int                     read_count;
+};
+
+struct thread_stat {
+	struct rb_node		rb;
+
+	u32                     tid;
+	struct list_head        seq_list;
+};
+
+static struct rb_root		thread_stats;
+
+static struct thread_stat *thread_stat_find(u32 tid)
+{
+	struct rb_node *node;
+	struct thread_stat *st;
+
+	node = thread_stats.rb_node;
+	while (node) {
+		st = container_of(node, struct thread_stat, rb);
+		if (st->tid == tid)
+			return st;
+		else if (tid < st->tid)
+			node = node->rb_left;
+		else
+			node = node->rb_right;
+	}
+
+	return NULL;
+}
+
+static void thread_stat_insert(struct thread_stat *new)
+{
+	struct rb_node **rb = &thread_stats.rb_node;
+	struct rb_node *parent = NULL;
+	struct thread_stat *p;
+
+	while (*rb) {
+		p = container_of(*rb, struct thread_stat, rb);
+		parent = *rb;
+
+		if (new->tid < p->tid)
+			rb = &(*rb)->rb_left;
+		else if (new->tid > p->tid)
+			rb = &(*rb)->rb_right;
+		else
+			BUG_ON("inserting invalid thread_stat\n");
+	}
+
+	rb_link_node(&new->rb, parent, rb);
+	rb_insert_color(&new->rb, &thread_stats);
+}
+
+static struct thread_stat *thread_stat_findnew_after_first(u32 tid)
+{
+	struct thread_stat *st;
+
+	st = thread_stat_find(tid);
+	if (st)
+		return st;
+
+	st = zalloc(sizeof(struct thread_stat));
+	if (!st) {
+		pr_err("memory allocation failed\n");
+		return NULL;
+	}
+
+	st->tid = tid;
+	INIT_LIST_HEAD(&st->seq_list);
+
+	thread_stat_insert(st);
+
+	return st;
+}
+
+static struct thread_stat *thread_stat_findnew_first(u32 tid);
+static struct thread_stat *(*thread_stat_findnew)(u32 tid) =
+	thread_stat_findnew_first;
+
+static struct thread_stat *thread_stat_findnew_first(u32 tid)
+{
+	struct thread_stat *st;
+
+	st = zalloc(sizeof(struct thread_stat));
+	if (!st) {
+		pr_err("memory allocation failed\n");
+		return NULL;
+	}
+	st->tid = tid;
+	INIT_LIST_HEAD(&st->seq_list);
+
+	rb_link_node(&st->rb, NULL, &thread_stats.rb_node);
+	rb_insert_color(&st->rb, &thread_stats);
+
+	thread_stat_findnew = thread_stat_findnew_after_first;
+	return st;
+}
+
+/* build simple key function one is bigger than two */
+#define SINGLE_KEY(member)						\
+	static int lock_stat_key_ ## member(struct lock_stat *one,	\
+					 struct lock_stat *two)		\
+	{								\
+		return one->member > two->member;			\
+	}
+
+SINGLE_KEY(nr_acquired)
+SINGLE_KEY(nr_contended)
+SINGLE_KEY(avg_wait_time)
+SINGLE_KEY(wait_time_total)
+SINGLE_KEY(wait_time_max)
+
+static int lock_stat_key_wait_time_min(struct lock_stat *one,
+					struct lock_stat *two)
+{
+	u64 s1 = one->wait_time_min;
+	u64 s2 = two->wait_time_min;
+	if (s1 == ULLONG_MAX)
+		s1 = 0;
+	if (s2 == ULLONG_MAX)
+		s2 = 0;
+	return s1 > s2;
+}
+
+struct lock_key {
+	/*
+	 * name: the value for specify by user
+	 * this should be simpler than raw name of member
+	 * e.g. nr_acquired -> acquired, wait_time_total -> wait_total
+	 */
+	const char		*name;
+	int			(*key)(struct lock_stat*, struct lock_stat*);
+};
+
+static const char		*sort_key = "acquired";
+
+static int			(*compare)(struct lock_stat *, struct lock_stat *);
+
+static struct rb_root		result;	/* place to store sorted data */
+
+#define DEF_KEY_LOCK(name, fn_suffix)	\
+	{ #name, lock_stat_key_ ## fn_suffix }
+struct lock_key keys[] = {
+	DEF_KEY_LOCK(acquired, nr_acquired),
+	DEF_KEY_LOCK(contended, nr_contended),
+	DEF_KEY_LOCK(avg_wait, avg_wait_time),
+	DEF_KEY_LOCK(wait_total, wait_time_total),
+	DEF_KEY_LOCK(wait_min, wait_time_min),
+	DEF_KEY_LOCK(wait_max, wait_time_max),
+
+	/* extra comparisons much complicated should be here */
+
+	{ NULL, NULL }
+};
+
+static int select_key(void)
+{
+	int i;
+
+	for (i = 0; keys[i].name; i++) {
+		if (!strcmp(keys[i].name, sort_key)) {
+			compare = keys[i].key;
+			return 0;
+		}
+	}
+
+	pr_err("Unknown compare key: %s\n", sort_key);
+
+	return -1;
+}
+
+static void insert_to_result(struct lock_stat *st,
+			     int (*bigger)(struct lock_stat *, struct lock_stat *))
+{
+	struct rb_node **rb = &result.rb_node;
+	struct rb_node *parent = NULL;
+	struct lock_stat *p;
+
+	while (*rb) {
+		p = container_of(*rb, struct lock_stat, rb);
+		parent = *rb;
+
+		if (bigger(st, p))
+			rb = &(*rb)->rb_left;
+		else
+			rb = &(*rb)->rb_right;
+	}
+
+	rb_link_node(&st->rb, parent, rb);
+	rb_insert_color(&st->rb, &result);
+}
+
+/* returns left most element of result, and erase it */
+static struct lock_stat *pop_from_result(void)
+{
+	struct rb_node *node = result.rb_node;
+
+	if (!node)
+		return NULL;
+
+	while (node->rb_left)
+		node = node->rb_left;
+
+	rb_erase(node, &result);
+	return container_of(node, struct lock_stat, rb);
+}
+
+static struct lock_stat *lock_stat_findnew(void *addr, const char *name)
+{
+	struct list_head *entry = lockhashentry(addr);
+	struct lock_stat *ret, *new;
+
+	list_for_each_entry(ret, entry, hash_entry) {
+		if (ret->addr == addr)
+			return ret;
+	}
+
+	new = zalloc(sizeof(struct lock_stat));
+	if (!new)
+		goto alloc_failed;
+
+	new->addr = addr;
+	new->name = zalloc(sizeof(char) * strlen(name) + 1);
+	if (!new->name) {
+		free(new);
+		goto alloc_failed;
+	}
+
+	strcpy(new->name, name);
+	new->wait_time_min = ULLONG_MAX;
+
+	list_add(&new->hash_entry, entry);
+	return new;
+
+alloc_failed:
+	pr_err("memory allocation failed\n");
+	return NULL;
+}
+
+struct trace_lock_handler {
+	int (*acquire_event)(struct perf_evsel *evsel,
+			     struct perf_sample *sample);
+
+	int (*acquired_event)(struct perf_evsel *evsel,
+			      struct perf_sample *sample);
+
+	int (*contended_event)(struct perf_evsel *evsel,
+			       struct perf_sample *sample);
+
+	int (*release_event)(struct perf_evsel *evsel,
+			     struct perf_sample *sample);
+};
+
+static struct lock_seq_stat *get_seq(struct thread_stat *ts, void *addr)
+{
+	struct lock_seq_stat *seq;
+
+	list_for_each_entry(seq, &ts->seq_list, list) {
+		if (seq->addr == addr)
+			return seq;
+	}
+
+	seq = zalloc(sizeof(struct lock_seq_stat));
+	if (!seq) {
+		pr_err("memory allocation failed\n");
+		return NULL;
+	}
+	seq->state = SEQ_STATE_UNINITIALIZED;
+	seq->addr = addr;
+
+	list_add(&seq->list, &ts->seq_list);
+	return seq;
+}
+
+enum broken_state {
+	BROKEN_ACQUIRE,
+	BROKEN_ACQUIRED,
+	BROKEN_CONTENDED,
+	BROKEN_RELEASE,
+	BROKEN_MAX,
+};
+
+static int bad_hist[BROKEN_MAX];
+
+enum acquire_flags {
+	TRY_LOCK = 1,
+	READ_LOCK = 2,
+};
+
+static int report_lock_acquire_event(struct perf_evsel *evsel,
+				     struct perf_sample *sample)
+{
+	void *addr;
+	struct lock_stat *ls;
+	struct thread_stat *ts;
+	struct lock_seq_stat *seq;
+	const char *name = perf_evsel__strval(evsel, sample, "name");
+	u64 tmp = perf_evsel__intval(evsel, sample, "lockdep_addr");
+	int flag = perf_evsel__intval(evsel, sample, "flag");
+
+	memcpy(&addr, &tmp, sizeof(void *));
+
+	ls = lock_stat_findnew(addr, name);
+	if (!ls)
+		return -ENOMEM;
+	if (ls->discard)
+		return 0;
+
+	ts = thread_stat_findnew(sample->tid);
+	if (!ts)
+		return -ENOMEM;
+
+	seq = get_seq(ts, addr);
+	if (!seq)
+		return -ENOMEM;
+
+	switch (seq->state) {
+	case SEQ_STATE_UNINITIALIZED:
+	case SEQ_STATE_RELEASED:
+		if (!flag) {
+			seq->state = SEQ_STATE_ACQUIRING;
+		} else {
+			if (flag & TRY_LOCK)
+				ls->nr_trylock++;
+			if (flag & READ_LOCK)
+				ls->nr_readlock++;
+			seq->state = SEQ_STATE_READ_ACQUIRED;
+			seq->read_count = 1;
+			ls->nr_acquired++;
+		}
+		break;
+	case SEQ_STATE_READ_ACQUIRED:
+		if (flag & READ_LOCK) {
+			seq->read_count++;
+			ls->nr_acquired++;
+			goto end;
+		} else {
+			goto broken;
+		}
+		break;
+	case SEQ_STATE_ACQUIRED:
+	case SEQ_STATE_ACQUIRING:
+	case SEQ_STATE_CONTENDED:
+broken:
+		/* broken lock sequence, discard it */
+		ls->discard = 1;
+		bad_hist[BROKEN_ACQUIRE]++;
+		list_del(&seq->list);
+		free(seq);
+		goto end;
+	default:
+		BUG_ON("Unknown state of lock sequence found!\n");
+		break;
+	}
+
+	ls->nr_acquire++;
+	seq->prev_event_time = sample->time;
+end:
+	return 0;
+}
+
+static int report_lock_acquired_event(struct perf_evsel *evsel,
+				      struct perf_sample *sample)
+{
+	void *addr;
+	struct lock_stat *ls;
+	struct thread_stat *ts;
+	struct lock_seq_stat *seq;
+	u64 contended_term;
+	const char *name = perf_evsel__strval(evsel, sample, "name");
+	u64 tmp = perf_evsel__intval(evsel, sample, "lockdep_addr");
+
+	memcpy(&addr, &tmp, sizeof(void *));
+
+	ls = lock_stat_findnew(addr, name);
+	if (!ls)
+		return -ENOMEM;
+	if (ls->discard)
+		return 0;
+
+	ts = thread_stat_findnew(sample->tid);
+	if (!ts)
+		return -ENOMEM;
+
+	seq = get_seq(ts, addr);
+	if (!seq)
+		return -ENOMEM;
+
+	switch (seq->state) {
+	case SEQ_STATE_UNINITIALIZED:
+		/* orphan event, do nothing */
+		return 0;
+	case SEQ_STATE_ACQUIRING:
+		break;
+	case SEQ_STATE_CONTENDED:
+		contended_term = sample->time - seq->prev_event_time;
+		ls->wait_time_total += contended_term;
+		if (contended_term < ls->wait_time_min)
+			ls->wait_time_min = contended_term;
+		if (ls->wait_time_max < contended_term)
+			ls->wait_time_max = contended_term;
+		break;
+	case SEQ_STATE_RELEASED:
+	case SEQ_STATE_ACQUIRED:
+	case SEQ_STATE_READ_ACQUIRED:
+		/* broken lock sequence, discard it */
+		ls->discard = 1;
+		bad_hist[BROKEN_ACQUIRED]++;
+		list_del(&seq->list);
+		free(seq);
+		goto end;
+	default:
+		BUG_ON("Unknown state of lock sequence found!\n");
+		break;
+	}
+
+	seq->state = SEQ_STATE_ACQUIRED;
+	ls->nr_acquired++;
+	ls->avg_wait_time = ls->nr_contended ? ls->wait_time_total/ls->nr_contended : 0;
+	seq->prev_event_time = sample->time;
+end:
+	return 0;
+}
+
+static int report_lock_contended_event(struct perf_evsel *evsel,
+				       struct perf_sample *sample)
+{
+	void *addr;
+	struct lock_stat *ls;
+	struct thread_stat *ts;
+	struct lock_seq_stat *seq;
+	const char *name = perf_evsel__strval(evsel, sample, "name");
+	u64 tmp = perf_evsel__intval(evsel, sample, "lockdep_addr");
+
+	memcpy(&addr, &tmp, sizeof(void *));
+
+	ls = lock_stat_findnew(addr, name);
+	if (!ls)
+		return -ENOMEM;
+	if (ls->discard)
+		return 0;
+
+	ts = thread_stat_findnew(sample->tid);
+	if (!ts)
+		return -ENOMEM;
+
+	seq = get_seq(ts, addr);
+	if (!seq)
+		return -ENOMEM;
+
+	switch (seq->state) {
+	case SEQ_STATE_UNINITIALIZED:
+		/* orphan event, do nothing */
+		return 0;
+	case SEQ_STATE_ACQUIRING:
+		break;
+	case SEQ_STATE_RELEASED:
+	case SEQ_STATE_ACQUIRED:
+	case SEQ_STATE_READ_ACQUIRED:
+	case SEQ_STATE_CONTENDED:
+		/* broken lock sequence, discard it */
+		ls->discard = 1;
+		bad_hist[BROKEN_CONTENDED]++;
+		list_del(&seq->list);
+		free(seq);
+		goto end;
+	default:
+		BUG_ON("Unknown state of lock sequence found!\n");
+		break;
+	}
+
+	seq->state = SEQ_STATE_CONTENDED;
+	ls->nr_contended++;
+	ls->avg_wait_time = ls->wait_time_total/ls->nr_contended;
+	seq->prev_event_time = sample->time;
+end:
+	return 0;
+}
+
+static int report_lock_release_event(struct perf_evsel *evsel,
+				     struct perf_sample *sample)
+{
+	void *addr;
+	struct lock_stat *ls;
+	struct thread_stat *ts;
+	struct lock_seq_stat *seq;
+	const char *name = perf_evsel__strval(evsel, sample, "name");
+	u64 tmp = perf_evsel__intval(evsel, sample, "lockdep_addr");
+
+	memcpy(&addr, &tmp, sizeof(void *));
+
+	ls = lock_stat_findnew(addr, name);
+	if (!ls)
+		return -ENOMEM;
+	if (ls->discard)
+		return 0;
+
+	ts = thread_stat_findnew(sample->tid);
+	if (!ts)
+		return -ENOMEM;
+
+	seq = get_seq(ts, addr);
+	if (!seq)
+		return -ENOMEM;
+
+	switch (seq->state) {
+	case SEQ_STATE_UNINITIALIZED:
+		goto end;
+	case SEQ_STATE_ACQUIRED:
+		break;
+	case SEQ_STATE_READ_ACQUIRED:
+		seq->read_count--;
+		BUG_ON(seq->read_count < 0);
+		if (!seq->read_count) {
+			ls->nr_release++;
+			goto end;
+		}
+		break;
+	case SEQ_STATE_ACQUIRING:
+	case SEQ_STATE_CONTENDED:
+	case SEQ_STATE_RELEASED:
+		/* broken lock sequence, discard it */
+		ls->discard = 1;
+		bad_hist[BROKEN_RELEASE]++;
+		goto free_seq;
+	default:
+		BUG_ON("Unknown state of lock sequence found!\n");
+		break;
+	}
+
+	ls->nr_release++;
+free_seq:
+	list_del(&seq->list);
+	free(seq);
+end:
+	return 0;
+}
+
+/* lock oriented handlers */
+/* TODO: handlers for CPU oriented, thread oriented */
+static struct trace_lock_handler report_lock_ops  = {
+	.acquire_event		= report_lock_acquire_event,
+	.acquired_event		= report_lock_acquired_event,
+	.contended_event	= report_lock_contended_event,
+	.release_event		= report_lock_release_event,
+};
+
+static struct trace_lock_handler *trace_handler;
+
+static int perf_evsel__process_lock_acquire(struct perf_evsel *evsel,
+					     struct perf_sample *sample)
+{
+	if (trace_handler->acquire_event)
+		return trace_handler->acquire_event(evsel, sample);
+	return 0;
+}
+
+static int perf_evsel__process_lock_acquired(struct perf_evsel *evsel,
+					      struct perf_sample *sample)
+{
+	if (trace_handler->acquired_event)
+		return trace_handler->acquired_event(evsel, sample);
+	return 0;
+}
+
+static int perf_evsel__process_lock_contended(struct perf_evsel *evsel,
+					      struct perf_sample *sample)
+{
+	if (trace_handler->contended_event)
+		return trace_handler->contended_event(evsel, sample);
+	return 0;
+}
+
+static int perf_evsel__process_lock_release(struct perf_evsel *evsel,
+					    struct perf_sample *sample)
+{
+	if (trace_handler->release_event)
+		return trace_handler->release_event(evsel, sample);
+	return 0;
+}
+
+static void print_bad_events(int bad, int total)
+{
+	/* Output for debug, this have to be removed */
+	int i;
+	const char *name[4] =
+		{ "acquire", "acquired", "contended", "release" };
+
+	pr_info("\n=== output for debug===\n\n");
+	pr_info("bad: %d, total: %d\n", bad, total);
+	pr_info("bad rate: %.2f %%\n", (double)bad / (double)total * 100);
+	pr_info("histogram of events caused bad sequence\n");
+	for (i = 0; i < BROKEN_MAX; i++)
+		pr_info(" %10s: %d\n", name[i], bad_hist[i]);
+}
+
+/* TODO: various way to print, coloring, nano or milli sec */
+static void print_result(void)
+{
+	struct lock_stat *st;
+	char cut_name[20];
+	int bad, total;
+
+	pr_info("%20s ", "Name");
+	pr_info("%10s ", "acquired");
+	pr_info("%10s ", "contended");
+
+	pr_info("%15s ", "avg wait (ns)");
+	pr_info("%15s ", "total wait (ns)");
+	pr_info("%15s ", "max wait (ns)");
+	pr_info("%15s ", "min wait (ns)");
+
+	pr_info("\n\n");
+
+	bad = total = 0;
+	while ((st = pop_from_result())) {
+		total++;
+		if (st->discard) {
+			bad++;
+			continue;
+		}
+		bzero(cut_name, 20);
+
+		if (strlen(st->name) < 16) {
+			/* output raw name */
+			pr_info("%20s ", st->name);
+		} else {
+			strncpy(cut_name, st->name, 16);
+			cut_name[16] = '.';
+			cut_name[17] = '.';
+			cut_name[18] = '.';
+			cut_name[19] = '\0';
+			/* cut off name for saving output style */
+			pr_info("%20s ", cut_name);
+		}
+
+		pr_info("%10u ", st->nr_acquired);
+		pr_info("%10u ", st->nr_contended);
+
+		pr_info("%15" PRIu64 " ", st->avg_wait_time);
+		pr_info("%15" PRIu64 " ", st->wait_time_total);
+		pr_info("%15" PRIu64 " ", st->wait_time_max);
+		pr_info("%15" PRIu64 " ", st->wait_time_min == ULLONG_MAX ?
+		       0 : st->wait_time_min);
+		pr_info("\n");
+	}
+
+	print_bad_events(bad, total);
+}
+
+static bool info_threads, info_map;
+
+static void dump_threads(void)
+{
+	struct thread_stat *st;
+	struct rb_node *node;
+	struct thread *t;
+
+	pr_info("%10s: comm\n", "Thread ID");
+
+	node = rb_first(&thread_stats);
+	while (node) {
+		st = container_of(node, struct thread_stat, rb);
+		t = perf_session__findnew(session, st->tid);
+		pr_info("%10d: %s\n", st->tid, thread__comm_str(t));
+		node = rb_next(node);
+	};
+}
+
+static void dump_map(void)
+{
+	unsigned int i;
+	struct lock_stat *st;
+
+	pr_info("Address of instance: name of class\n");
+	for (i = 0; i < LOCKHASH_SIZE; i++) {
+		list_for_each_entry(st, &lockhash_table[i], hash_entry) {
+			pr_info(" %p: %s\n", st->addr, st->name);
+		}
+	}
+}
+
+static int dump_info(void)
+{
+	int rc = 0;
+
+	if (info_threads)
+		dump_threads();
+	else if (info_map)
+		dump_map();
+	else {
+		rc = -1;
+		pr_err("Unknown type of information\n");
+	}
+
+	return rc;
+}
+
+typedef int (*tracepoint_handler)(struct perf_evsel *evsel,
+				  struct perf_sample *sample);
+
+static int process_sample_event(struct perf_tool *tool __maybe_unused,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel,
+				struct machine *machine)
+{
+	struct thread *thread = machine__findnew_thread(machine, sample->pid,
+							sample->tid);
+
+	if (thread == NULL) {
+		pr_debug("problem processing %d event, skipping it.\n",
+			event->header.type);
+		return -1;
+	}
+
+	if (evsel->handler != NULL) {
+		tracepoint_handler f = evsel->handler;
+		return f(evsel, sample);
+	}
+
+	return 0;
+}
+
+static void sort_result(void)
+{
+	unsigned int i;
+	struct lock_stat *st;
+
+	for (i = 0; i < LOCKHASH_SIZE; i++) {
+		list_for_each_entry(st, &lockhash_table[i], hash_entry) {
+			insert_to_result(st, compare);
+		}
+	}
+}
+
+static const struct perf_evsel_str_handler lock_tracepoints[] = {
+	{ "lock:lock_acquire",	 perf_evsel__process_lock_acquire,   }, /* CONFIG_LOCKDEP */
+	{ "lock:lock_acquired",	 perf_evsel__process_lock_acquired,  }, /* CONFIG_LOCKDEP, CONFIG_LOCK_STAT */
+	{ "lock:lock_contended", perf_evsel__process_lock_contended, }, /* CONFIG_LOCKDEP, CONFIG_LOCK_STAT */
+	{ "lock:lock_release",	 perf_evsel__process_lock_release,   }, /* CONFIG_LOCKDEP */
+};
+
+static bool force;
+
+static int __cmd_report(bool display_info)
+{
+	int err = -EINVAL;
+	struct perf_tool eops = {
+		.sample		 = process_sample_event,
+		.comm		 = perf_event__process_comm,
+		.ordered_events	 = true,
+	};
+	struct perf_data_file file = {
+		.path = input_name,
+		.mode = PERF_DATA_MODE_READ,
+		.force = force,
+	};
+
+	session = perf_session__new(&file, false, &eops);
+	if (!session) {
+		pr_err("Initializing perf session failed\n");
+		return -1;
+	}
+
+	symbol__init(&session->header.env);
+
+	if (!perf_session__has_traces(session, "lock record"))
+		goto out_delete;
+
+	if (perf_session__set_tracepoints_handlers(session, lock_tracepoints)) {
+		pr_err("Initializing perf session tracepoint handlers failed\n");
+		goto out_delete;
+	}
+
+	if (select_key())
+		goto out_delete;
+
+	err = perf_session__process_events(session);
+	if (err)
+		goto out_delete;
+
+	setup_pager();
+	if (display_info) /* used for info subcommand */
+		err = dump_info();
+	else {
+		sort_result();
+		print_result();
+	}
+
+out_delete:
+	perf_session__delete(session);
+	return err;
+}
+
+static int __cmd_record(int argc, const char **argv)
+{
+	const char *record_args[] = {
+		"record", "-R", "-m", "1024", "-c", "1",
+	};
+	unsigned int rec_argc, i, j, ret;
+	const char **rec_argv;
+
+	for (i = 0; i < ARRAY_SIZE(lock_tracepoints); i++) {
+		if (!is_valid_tracepoint(lock_tracepoints[i].name)) {
+				pr_err("tracepoint %s is not enabled. "
+				       "Are CONFIG_LOCKDEP and CONFIG_LOCK_STAT enabled?\n",
+				       lock_tracepoints[i].name);
+				return 1;
+		}
+	}
+
+	rec_argc = ARRAY_SIZE(record_args) + argc - 1;
+	/* factor of 2 is for -e in front of each tracepoint */
+	rec_argc += 2 * ARRAY_SIZE(lock_tracepoints);
+
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+	if (!rec_argv)
+		return -ENOMEM;
+
+	for (i = 0; i < ARRAY_SIZE(record_args); i++)
+		rec_argv[i] = strdup(record_args[i]);
+
+	for (j = 0; j < ARRAY_SIZE(lock_tracepoints); j++) {
+		rec_argv[i++] = "-e";
+		rec_argv[i++] = strdup(lock_tracepoints[j].name);
+	}
+
+	for (j = 1; j < (unsigned int)argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	ret = cmd_record(i, rec_argv, NULL);
+	free(rec_argv);
+	return ret;
+}
+
+int cmd_lock(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	const struct option info_options[] = {
+	OPT_BOOLEAN('t', "threads", &info_threads,
+		    "dump thread list in perf.data"),
+	OPT_BOOLEAN('m', "map", &info_map,
+		    "map of lock instances (address:name table)"),
+	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
+	OPT_END()
+	};
+	const struct option lock_options[] = {
+	OPT_STRING('i', "input", &input_name, "file", "input file name"),
+	OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"),
+	OPT_END()
+	};
+	const struct option report_options[] = {
+	OPT_STRING('k', "key", &sort_key, "acquired",
+		    "key for sorting (acquired / contended / avg_wait / wait_total / wait_max / wait_min)"),
+	OPT_BOOLEAN('f', "force", &force, "don't complain, do it"),
+	/* TODO: type */
+	OPT_END()
+	};
+	const char * const info_usage[] = {
+		"perf lock info [<options>]",
+		NULL
+	};
+	const char *const lock_subcommands[] = { "record", "report", "script",
+						 "info", NULL };
+	const char *lock_usage[] = {
+		NULL,
+		NULL
+	};
+	const char * const report_usage[] = {
+		"perf lock report [<options>]",
+		NULL
+	};
+	unsigned int i;
+	int rc = 0;
+
+	for (i = 0; i < LOCKHASH_SIZE; i++)
+		INIT_LIST_HEAD(lockhash_table + i);
+
+	argc = parse_options_subcommand(argc, argv, lock_options, lock_subcommands,
+					lock_usage, PARSE_OPT_STOP_AT_NON_OPTION);
+	if (!argc)
+		usage_with_options(lock_usage, lock_options);
+
+	if (!strncmp(argv[0], "rec", 3)) {
+		return __cmd_record(argc, argv);
+	} else if (!strncmp(argv[0], "report", 6)) {
+		trace_handler = &report_lock_ops;
+		if (argc) {
+			argc = parse_options(argc, argv,
+					     report_options, report_usage, 0);
+			if (argc)
+				usage_with_options(report_usage, report_options);
+		}
+		rc = __cmd_report(false);
+	} else if (!strcmp(argv[0], "script")) {
+		/* Aliased to 'perf script' */
+		return cmd_script(argc, argv, prefix);
+	} else if (!strcmp(argv[0], "info")) {
+		if (argc) {
+			argc = parse_options(argc, argv,
+					     info_options, info_usage, 0);
+			if (argc)
+				usage_with_options(info_usage, info_options);
+		}
+		/* recycling report_lock_ops */
+		trace_handler = &report_lock_ops;
+		rc = __cmd_report(true);
+	} else {
+		usage_with_options(lock_usage, lock_options);
+	}
+
+	return rc;
+}
diff --git a/tools/perf/builtin-mem.c b/tools/perf/builtin-mem.c
new file mode 100644
index 000000000..675216e08
--- /dev/null
+++ b/tools/perf/builtin-mem.c
@@ -0,0 +1,326 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+#include "util/tool.h"
+#include "util/session.h"
+#include "util/data.h"
+
+#define MEM_OPERATION_LOAD	0x1
+#define MEM_OPERATION_STORE	0x2
+
+struct perf_mem {
+	struct perf_tool	tool;
+	char const		*input_name;
+	bool			hide_unresolved;
+	bool			dump_raw;
+	bool			force;
+	int			operation;
+	const char		*cpu_list;
+	DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
+};
+
+static int __cmd_record(int argc, const char **argv, struct perf_mem *mem)
+{
+	int rec_argc, i = 0, j;
+	const char **rec_argv;
+	int ret;
+
+	rec_argc = argc + 7; /* max number of arguments */
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+	if (!rec_argv)
+		return -1;
+
+	rec_argv[i++] = "record";
+
+	if (mem->operation & MEM_OPERATION_LOAD)
+		rec_argv[i++] = "-W";
+
+	rec_argv[i++] = "-d";
+
+	if (mem->operation & MEM_OPERATION_LOAD) {
+		rec_argv[i++] = "-e";
+		rec_argv[i++] = "cpu/mem-loads/pp";
+	}
+
+	if (mem->operation & MEM_OPERATION_STORE) {
+		rec_argv[i++] = "-e";
+		rec_argv[i++] = "cpu/mem-stores/pp";
+	}
+
+	for (j = 1; j < argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	ret = cmd_record(i, rec_argv, NULL);
+	free(rec_argv);
+	return ret;
+}
+
+static int
+dump_raw_samples(struct perf_tool *tool,
+		 union perf_event *event,
+		 struct perf_sample *sample,
+		 struct machine *machine)
+{
+	struct perf_mem *mem = container_of(tool, struct perf_mem, tool);
+	struct addr_location al;
+	const char *fmt;
+
+	if (perf_event__preprocess_sample(event, machine, &al, sample) < 0) {
+		fprintf(stderr, "problem processing %d event, skipping it.\n",
+				event->header.type);
+		return -1;
+	}
+
+	if (al.filtered || (mem->hide_unresolved && al.sym == NULL))
+		return 0;
+
+	if (al.map != NULL)
+		al.map->dso->hit = 1;
+
+	if (symbol_conf.field_sep) {
+		fmt = "%d%s%d%s0x%"PRIx64"%s0x%"PRIx64"%s%"PRIu64
+		      "%s0x%"PRIx64"%s%s:%s\n";
+	} else {
+		fmt = "%5d%s%5d%s0x%016"PRIx64"%s0x016%"PRIx64
+		      "%s%5"PRIu64"%s0x%06"PRIx64"%s%s:%s\n";
+		symbol_conf.field_sep = " ";
+	}
+
+	printf(fmt,
+		sample->pid,
+		symbol_conf.field_sep,
+		sample->tid,
+		symbol_conf.field_sep,
+		sample->ip,
+		symbol_conf.field_sep,
+		sample->addr,
+		symbol_conf.field_sep,
+		sample->weight,
+		symbol_conf.field_sep,
+		sample->data_src,
+		symbol_conf.field_sep,
+		al.map ? (al.map->dso ? al.map->dso->long_name : "???") : "???",
+		al.sym ? al.sym->name : "???");
+
+	return 0;
+}
+
+static int process_sample_event(struct perf_tool *tool,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel __maybe_unused,
+				struct machine *machine)
+{
+	return dump_raw_samples(tool, event, sample, machine);
+}
+
+static int report_raw_events(struct perf_mem *mem)
+{
+	struct perf_data_file file = {
+		.path = input_name,
+		.mode = PERF_DATA_MODE_READ,
+		.force = mem->force,
+	};
+	int err = -EINVAL;
+	int ret;
+	struct perf_session *session = perf_session__new(&file, false,
+							 &mem->tool);
+
+	if (session == NULL)
+		return -1;
+
+	if (mem->cpu_list) {
+		ret = perf_session__cpu_bitmap(session, mem->cpu_list,
+					       mem->cpu_bitmap);
+		if (ret)
+			goto out_delete;
+	}
+
+	if (symbol__init(&session->header.env) < 0)
+		return -1;
+
+	printf("# PID, TID, IP, ADDR, LOCAL WEIGHT, DSRC, SYMBOL\n");
+
+	err = perf_session__process_events(session);
+	if (err)
+		return err;
+
+	return 0;
+
+out_delete:
+	perf_session__delete(session);
+	return err;
+}
+
+static int report_events(int argc, const char **argv, struct perf_mem *mem)
+{
+	const char **rep_argv;
+	int ret, i = 0, j, rep_argc;
+
+	if (mem->dump_raw)
+		return report_raw_events(mem);
+
+	rep_argc = argc + 3;
+	rep_argv = calloc(rep_argc + 1, sizeof(char *));
+	if (!rep_argv)
+		return -1;
+
+	rep_argv[i++] = "report";
+	rep_argv[i++] = "--mem-mode";
+	rep_argv[i++] = "-n"; /* display number of samples */
+
+	/*
+	 * there is no weight (cost) associated with stores, so don't print
+	 * the column
+	 */
+	if (!(mem->operation & MEM_OPERATION_LOAD))
+		rep_argv[i++] = "--sort=mem,sym,dso,symbol_daddr,"
+				"dso_daddr,tlb,locked";
+
+	for (j = 1; j < argc; j++, i++)
+		rep_argv[i] = argv[j];
+
+	ret = cmd_report(i, rep_argv, NULL);
+	free(rep_argv);
+	return ret;
+}
+
+struct mem_mode {
+	const char *name;
+	int mode;
+};
+
+#define MEM_OPT(n, m) \
+	{ .name = n, .mode = (m) }
+
+#define MEM_END { .name = NULL }
+
+static const struct mem_mode mem_modes[]={
+	MEM_OPT("load", MEM_OPERATION_LOAD),
+	MEM_OPT("store", MEM_OPERATION_STORE),
+	MEM_END
+};
+
+static int
+parse_mem_ops(const struct option *opt, const char *str, int unset)
+{
+	int *mode = (int *)opt->value;
+	const struct mem_mode *m;
+	char *s, *os = NULL, *p;
+	int ret = -1;
+
+	if (unset)
+		return 0;
+
+	/* str may be NULL in case no arg is passed to -t */
+	if (str) {
+		/* because str is read-only */
+		s = os = strdup(str);
+		if (!s)
+			return -1;
+
+		/* reset mode */
+		*mode = 0;
+
+		for (;;) {
+			p = strchr(s, ',');
+			if (p)
+				*p = '\0';
+
+			for (m = mem_modes; m->name; m++) {
+				if (!strcasecmp(s, m->name))
+					break;
+			}
+			if (!m->name) {
+				fprintf(stderr, "unknown sampling op %s,"
+					    " check man page\n", s);
+				goto error;
+			}
+
+			*mode |= m->mode;
+
+			if (!p)
+				break;
+
+			s = p + 1;
+		}
+	}
+	ret = 0;
+
+	if (*mode == 0)
+		*mode = MEM_OPERATION_LOAD;
+error:
+	free(os);
+	return ret;
+}
+
+int cmd_mem(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	struct stat st;
+	struct perf_mem mem = {
+		.tool = {
+			.sample		= process_sample_event,
+			.mmap		= perf_event__process_mmap,
+			.mmap2		= perf_event__process_mmap2,
+			.comm		= perf_event__process_comm,
+			.lost		= perf_event__process_lost,
+			.fork		= perf_event__process_fork,
+			.build_id	= perf_event__process_build_id,
+			.ordered_events	= true,
+		},
+		.input_name		 = "perf.data",
+		/*
+		 * default to both load an store sampling
+		 */
+		.operation		 = MEM_OPERATION_LOAD | MEM_OPERATION_STORE,
+	};
+	const struct option mem_options[] = {
+	OPT_CALLBACK('t', "type", &mem.operation,
+		   "type", "memory operations(load,store) Default load,store",
+		    parse_mem_ops),
+	OPT_BOOLEAN('D', "dump-raw-samples", &mem.dump_raw,
+		    "dump raw samples in ASCII"),
+	OPT_BOOLEAN('U', "hide-unresolved", &mem.hide_unresolved,
+		    "Only display entries resolved to a symbol"),
+	OPT_STRING('i', "input", &input_name, "file",
+		   "input file name"),
+	OPT_STRING('C', "cpu", &mem.cpu_list, "cpu",
+		   "list of cpus to profile"),
+	OPT_STRING_NOEMPTY('x', "field-separator", &symbol_conf.field_sep,
+		   "separator",
+		   "separator for columns, no spaces will be added"
+		   " between columns '.' is reserved."),
+	OPT_BOOLEAN('f', "force", &mem.force, "don't complain, do it"),
+	OPT_END()
+	};
+	const char *const mem_subcommands[] = { "record", "report", NULL };
+	const char *mem_usage[] = {
+		NULL,
+		NULL
+	};
+
+
+	argc = parse_options_subcommand(argc, argv, mem_options, mem_subcommands,
+					mem_usage, PARSE_OPT_STOP_AT_NON_OPTION);
+
+	if (!argc || !(strncmp(argv[0], "rec", 3) || mem.operation))
+		usage_with_options(mem_usage, mem_options);
+
+	if (!mem.input_name || !strlen(mem.input_name)) {
+		if (!fstat(STDIN_FILENO, &st) && S_ISFIFO(st.st_mode))
+			mem.input_name = "-";
+		else
+			mem.input_name = "perf.data";
+	}
+
+	if (!strncmp(argv[0], "rec", 3))
+		return __cmd_record(argc, argv, &mem);
+	else if (!strncmp(argv[0], "rep", 3))
+		return report_events(argc, argv, &mem);
+	else
+		usage_with_options(mem_usage, mem_options);
+
+	return 0;
+}
diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c
new file mode 100644
index 000000000..f7b1af67e
--- /dev/null
+++ b/tools/perf/builtin-probe.c
@@ -0,0 +1,526 @@
+/*
+ * builtin-probe.c
+ *
+ * Builtin probe command: Set up probe events by C expression
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+#include <sys/utsname.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "perf.h"
+#include "builtin.h"
+#include "util/util.h"
+#include "util/strlist.h"
+#include "util/strfilter.h"
+#include "util/symbol.h"
+#include "util/debug.h"
+#include <api/fs/debugfs.h>
+#include "util/parse-options.h"
+#include "util/probe-finder.h"
+#include "util/probe-event.h"
+
+#define DEFAULT_VAR_FILTER "!__k???tab_* & !__crc_*"
+#define DEFAULT_FUNC_FILTER "!_*"
+
+/* Session management structure */
+static struct {
+	bool list_events;
+	bool force_add;
+	bool show_lines;
+	bool show_vars;
+	bool show_ext_vars;
+	bool show_funcs;
+	bool mod_events;
+	bool uprobes;
+	bool quiet;
+	bool target_used;
+	int nevents;
+	struct perf_probe_event events[MAX_PROBES];
+	struct strlist *dellist;
+	struct line_range line_range;
+	char *target;
+	int max_probe_points;
+	struct strfilter *filter;
+} params;
+
+/* Parse an event definition. Note that any error must die. */
+static int parse_probe_event(const char *str)
+{
+	struct perf_probe_event *pev = &params.events[params.nevents];
+	int ret;
+
+	pr_debug("probe-definition(%d): %s\n", params.nevents, str);
+	if (++params.nevents == MAX_PROBES) {
+		pr_err("Too many probes (> %d) were specified.", MAX_PROBES);
+		return -1;
+	}
+
+	pev->uprobes = params.uprobes;
+	if (params.target) {
+		pev->target = strdup(params.target);
+		if (!pev->target)
+			return -ENOMEM;
+		params.target_used = true;
+	}
+
+	/* Parse a perf-probe command into event */
+	ret = parse_perf_probe_command(str, pev);
+	pr_debug("%d arguments\n", pev->nargs);
+
+	return ret;
+}
+
+static int set_target(const char *ptr)
+{
+	int found = 0;
+	const char *buf;
+
+	/*
+	 * The first argument after options can be an absolute path
+	 * to an executable / library or kernel module.
+	 *
+	 * TODO: Support relative path, and $PATH, $LD_LIBRARY_PATH,
+	 * short module name.
+	 */
+	if (!params.target && ptr && *ptr == '/') {
+		params.target = strdup(ptr);
+		if (!params.target)
+			return -ENOMEM;
+		params.target_used = false;
+
+		found = 1;
+		buf = ptr + (strlen(ptr) - 3);
+
+		if (strcmp(buf, ".ko"))
+			params.uprobes = true;
+
+	}
+
+	return found;
+}
+
+static int parse_probe_event_argv(int argc, const char **argv)
+{
+	int i, len, ret, found_target;
+	char *buf;
+
+	found_target = set_target(argv[0]);
+	if (found_target < 0)
+		return found_target;
+
+	if (found_target && argc == 1)
+		return 0;
+
+	/* Bind up rest arguments */
+	len = 0;
+	for (i = 0; i < argc; i++) {
+		if (i == 0 && found_target)
+			continue;
+
+		len += strlen(argv[i]) + 1;
+	}
+	buf = zalloc(len + 1);
+	if (buf == NULL)
+		return -ENOMEM;
+	len = 0;
+	for (i = 0; i < argc; i++) {
+		if (i == 0 && found_target)
+			continue;
+
+		len += sprintf(&buf[len], "%s ", argv[i]);
+	}
+	params.mod_events = true;
+	ret = parse_probe_event(buf);
+	free(buf);
+	return ret;
+}
+
+static int opt_add_probe_event(const struct option *opt __maybe_unused,
+			      const char *str, int unset __maybe_unused)
+{
+	if (str) {
+		params.mod_events = true;
+		return parse_probe_event(str);
+	} else
+		return 0;
+}
+
+static int opt_del_probe_event(const struct option *opt __maybe_unused,
+			       const char *str, int unset __maybe_unused)
+{
+	if (str) {
+		params.mod_events = true;
+		if (!params.dellist)
+			params.dellist = strlist__new(true, NULL);
+		strlist__add(params.dellist, str);
+	}
+	return 0;
+}
+
+static int opt_set_target(const struct option *opt, const char *str,
+			int unset __maybe_unused)
+{
+	int ret = -ENOENT;
+	char *tmp;
+
+	if  (str) {
+		if (!strcmp(opt->long_name, "exec"))
+			params.uprobes = true;
+#ifdef HAVE_DWARF_SUPPORT
+		else if (!strcmp(opt->long_name, "module"))
+			params.uprobes = false;
+#endif
+		else
+			return ret;
+
+		/* Expand given path to absolute path, except for modulename */
+		if (params.uprobes || strchr(str, '/')) {
+			tmp = realpath(str, NULL);
+			if (!tmp) {
+				pr_warning("Failed to get the absolute path of %s: %m\n", str);
+				return ret;
+			}
+		} else {
+			tmp = strdup(str);
+			if (!tmp)
+				return -ENOMEM;
+		}
+		free(params.target);
+		params.target = tmp;
+		params.target_used = false;
+		ret = 0;
+	}
+
+	return ret;
+}
+
+#ifdef HAVE_DWARF_SUPPORT
+static int opt_show_lines(const struct option *opt __maybe_unused,
+			  const char *str, int unset __maybe_unused)
+{
+	int ret = 0;
+
+	if (!str)
+		return 0;
+
+	if (params.show_lines) {
+		pr_warning("Warning: more than one --line options are"
+			   " detected. Only the first one is valid.\n");
+		return 0;
+	}
+
+	params.show_lines = true;
+	ret = parse_line_range_desc(str, &params.line_range);
+
+	return ret;
+}
+
+static int opt_show_vars(const struct option *opt __maybe_unused,
+			 const char *str, int unset __maybe_unused)
+{
+	struct perf_probe_event *pev = &params.events[params.nevents];
+	int ret;
+
+	if (!str)
+		return 0;
+
+	ret = parse_probe_event(str);
+	if (!ret && pev->nargs != 0) {
+		pr_err("  Error: '--vars' doesn't accept arguments.\n");
+		return -EINVAL;
+	}
+	params.show_vars = true;
+
+	return ret;
+}
+#endif
+
+static int opt_set_filter(const struct option *opt __maybe_unused,
+			  const char *str, int unset __maybe_unused)
+{
+	const char *err;
+
+	if (str) {
+		pr_debug2("Set filter: %s\n", str);
+		if (params.filter)
+			strfilter__delete(params.filter);
+		params.filter = strfilter__new(str, &err);
+		if (!params.filter) {
+			pr_err("Filter parse error at %td.\n", err - str + 1);
+			pr_err("Source: \"%s\"\n", str);
+			pr_err("         %*c\n", (int)(err - str + 1), '^');
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int init_params(void)
+{
+	return line_range__init(&params.line_range);
+}
+
+static void cleanup_params(void)
+{
+	int i;
+
+	for (i = 0; i < params.nevents; i++)
+		clear_perf_probe_event(params.events + i);
+	if (params.dellist)
+		strlist__delete(params.dellist);
+	line_range__clear(&params.line_range);
+	free(params.target);
+	if (params.filter)
+		strfilter__delete(params.filter);
+	memset(&params, 0, sizeof(params));
+}
+
+static void pr_err_with_code(const char *msg, int err)
+{
+	char sbuf[STRERR_BUFSIZE];
+
+	pr_err("%s", msg);
+	pr_debug(" Reason: %s (Code: %d)",
+		 strerror_r(-err, sbuf, sizeof(sbuf)), err);
+	pr_err("\n");
+}
+
+static int
+__cmd_probe(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	const char * const probe_usage[] = {
+		"perf probe [<options>] 'PROBEDEF' ['PROBEDEF' ...]",
+		"perf probe [<options>] --add 'PROBEDEF' [--add 'PROBEDEF' ...]",
+		"perf probe [<options>] --del '[GROUP:]EVENT' ...",
+		"perf probe --list",
+#ifdef HAVE_DWARF_SUPPORT
+		"perf probe [<options>] --line 'LINEDESC'",
+		"perf probe [<options>] --vars 'PROBEPOINT'",
+#endif
+		NULL
+};
+	struct option options[] = {
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show parsed arguments, etc)"),
+	OPT_BOOLEAN('q', "quiet", &params.quiet,
+		    "be quiet (do not show any mesages)"),
+	OPT_BOOLEAN('l', "list", &params.list_events,
+		    "list up current probe events"),
+	OPT_CALLBACK('d', "del", NULL, "[GROUP:]EVENT", "delete a probe event.",
+		opt_del_probe_event),
+	OPT_CALLBACK('a', "add", NULL,
+#ifdef HAVE_DWARF_SUPPORT
+		"[EVENT=]FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT"
+		" [[NAME=]ARG ...]",
+#else
+		"[EVENT=]FUNC[+OFF|%return] [[NAME=]ARG ...]",
+#endif
+		"probe point definition, where\n"
+		"\t\tGROUP:\tGroup name (optional)\n"
+		"\t\tEVENT:\tEvent name\n"
+		"\t\tFUNC:\tFunction name\n"
+		"\t\tOFF:\tOffset from function entry (in byte)\n"
+		"\t\t%return:\tPut the probe at function return\n"
+#ifdef HAVE_DWARF_SUPPORT
+		"\t\tSRC:\tSource code path\n"
+		"\t\tRL:\tRelative line number from function entry.\n"
+		"\t\tAL:\tAbsolute line number in file.\n"
+		"\t\tPT:\tLazy expression of line code.\n"
+		"\t\tARG:\tProbe argument (local variable name or\n"
+		"\t\t\tkprobe-tracer argument format.)\n",
+#else
+		"\t\tARG:\tProbe argument (kprobe-tracer argument format.)\n",
+#endif
+		opt_add_probe_event),
+	OPT_BOOLEAN('f', "force", &params.force_add, "forcibly add events"
+		    " with existing name"),
+#ifdef HAVE_DWARF_SUPPORT
+	OPT_CALLBACK('L', "line", NULL,
+		     "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]",
+		     "Show source code lines.", opt_show_lines),
+	OPT_CALLBACK('V', "vars", NULL,
+		     "FUNC[@SRC][+OFF|%return|:RL|;PT]|SRC:AL|SRC;PT",
+		     "Show accessible variables on PROBEDEF", opt_show_vars),
+	OPT_BOOLEAN('\0', "externs", &params.show_ext_vars,
+		    "Show external variables too (with --vars only)"),
+	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+		   "file", "vmlinux pathname"),
+	OPT_STRING('s', "source", &symbol_conf.source_prefix,
+		   "directory", "path to kernel source"),
+	OPT_CALLBACK('m', "module", NULL, "modname|path",
+		"target module name (for online) or path (for offline)",
+		opt_set_target),
+#endif
+	OPT__DRY_RUN(&probe_event_dry_run),
+	OPT_INTEGER('\0', "max-probes", &params.max_probe_points,
+		 "Set how many probe points can be found for a probe."),
+	OPT_BOOLEAN('F', "funcs", &params.show_funcs,
+		    "Show potential probe-able functions."),
+	OPT_CALLBACK('\0', "filter", NULL,
+		     "[!]FILTER", "Set a filter (with --vars/funcs only)\n"
+		     "\t\t\t(default: \"" DEFAULT_VAR_FILTER "\" for --vars,\n"
+		     "\t\t\t \"" DEFAULT_FUNC_FILTER "\" for --funcs)",
+		     opt_set_filter),
+	OPT_CALLBACK('x', "exec", NULL, "executable|path",
+			"target executable name or path", opt_set_target),
+	OPT_BOOLEAN(0, "demangle", &symbol_conf.demangle,
+		    "Enable symbol demangling"),
+	OPT_BOOLEAN(0, "demangle-kernel", &symbol_conf.demangle_kernel,
+		    "Enable kernel symbol demangling"),
+	OPT_END()
+	};
+	int ret;
+
+	set_option_flag(options, 'a', "add", PARSE_OPT_EXCLUSIVE);
+	set_option_flag(options, 'd', "del", PARSE_OPT_EXCLUSIVE);
+	set_option_flag(options, 'l', "list", PARSE_OPT_EXCLUSIVE);
+#ifdef HAVE_DWARF_SUPPORT
+	set_option_flag(options, 'L', "line", PARSE_OPT_EXCLUSIVE);
+	set_option_flag(options, 'V', "vars", PARSE_OPT_EXCLUSIVE);
+#endif
+
+	argc = parse_options(argc, argv, options, probe_usage,
+			     PARSE_OPT_STOP_AT_NON_OPTION);
+	if (argc > 0) {
+		if (strcmp(argv[0], "-") == 0) {
+			pr_warning("  Error: '-' is not supported.\n");
+			usage_with_options(probe_usage, options);
+		}
+		ret = parse_probe_event_argv(argc, argv);
+		if (ret < 0) {
+			pr_err_with_code("  Error: Command Parse Error.", ret);
+			return ret;
+		}
+	}
+
+	if (params.quiet) {
+		if (verbose != 0) {
+			pr_err("  Error: -v and -q are exclusive.\n");
+			return -EINVAL;
+		}
+		verbose = -1;
+	}
+
+	if (params.max_probe_points == 0)
+		params.max_probe_points = MAX_PROBES;
+
+	if ((!params.nevents && !params.dellist && !params.list_events &&
+	     !params.show_lines && !params.show_funcs))
+		usage_with_options(probe_usage, options);
+
+	/*
+	 * Only consider the user's kernel image path if given.
+	 */
+	symbol_conf.try_vmlinux_path = (symbol_conf.vmlinux_name == NULL);
+
+	if (params.list_events) {
+		if (params.uprobes) {
+			pr_warning("  Error: Don't use --list with --exec.\n");
+			usage_with_options(probe_usage, options);
+		}
+		ret = show_perf_probe_events();
+		if (ret < 0)
+			pr_err_with_code("  Error: Failed to show event list.", ret);
+		return ret;
+	}
+	if (params.show_funcs) {
+		if (!params.filter)
+			params.filter = strfilter__new(DEFAULT_FUNC_FILTER,
+						       NULL);
+		ret = show_available_funcs(params.target, params.filter,
+					params.uprobes);
+		strfilter__delete(params.filter);
+		params.filter = NULL;
+		if (ret < 0)
+			pr_err_with_code("  Error: Failed to show functions.", ret);
+		return ret;
+	}
+
+#ifdef HAVE_DWARF_SUPPORT
+	if (params.show_lines) {
+		ret = show_line_range(&params.line_range, params.target,
+				      params.uprobes);
+		if (ret < 0)
+			pr_err_with_code("  Error: Failed to show lines.", ret);
+		return ret;
+	}
+	if (params.show_vars) {
+		if (!params.filter)
+			params.filter = strfilter__new(DEFAULT_VAR_FILTER,
+						       NULL);
+
+		ret = show_available_vars(params.events, params.nevents,
+					  params.max_probe_points,
+					  params.target,
+					  params.filter,
+					  params.show_ext_vars);
+		strfilter__delete(params.filter);
+		params.filter = NULL;
+		if (ret < 0)
+			pr_err_with_code("  Error: Failed to show vars.", ret);
+		return ret;
+	}
+#endif
+
+	if (params.dellist) {
+		ret = del_perf_probe_events(params.dellist);
+		if (ret < 0) {
+			pr_err_with_code("  Error: Failed to delete events.", ret);
+			return ret;
+		}
+	}
+
+	if (params.nevents) {
+		/* Ensure the last given target is used */
+		if (params.target && !params.target_used) {
+			pr_warning("  Error: -x/-m must follow the probe definitions.\n");
+			usage_with_options(probe_usage, options);
+		}
+
+		ret = add_perf_probe_events(params.events, params.nevents,
+					    params.max_probe_points,
+					    params.force_add);
+		if (ret < 0) {
+			pr_err_with_code("  Error: Failed to add events.", ret);
+			return ret;
+		}
+	}
+	return 0;
+}
+
+int cmd_probe(int argc, const char **argv, const char *prefix)
+{
+	int ret;
+
+	ret = init_params();
+	if (!ret) {
+		ret = __cmd_probe(argc, argv, prefix);
+		cleanup_params();
+	}
+
+	return ret;
+}
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
new file mode 100644
index 000000000..c3efdfb63
--- /dev/null
+++ b/tools/perf/builtin-record.c
@@ -0,0 +1,1015 @@
+/*
+ * builtin-record.c
+ *
+ * Builtin record command: Record the profile of a workload
+ * (or a CPU, or a PID) into the perf.data output file - for
+ * later analysis via perf report.
+ */
+#include "builtin.h"
+
+#include "perf.h"
+
+#include "util/build-id.h"
+#include "util/util.h"
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+
+#include "util/callchain.h"
+#include "util/cgroup.h"
+#include "util/header.h"
+#include "util/event.h"
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/debug.h"
+#include "util/session.h"
+#include "util/tool.h"
+#include "util/symbol.h"
+#include "util/cpumap.h"
+#include "util/thread_map.h"
+#include "util/data.h"
+
+#include <unistd.h>
+#include <sched.h>
+#include <sys/mman.h>
+
+
+struct record {
+	struct perf_tool	tool;
+	struct record_opts	opts;
+	u64			bytes_written;
+	struct perf_data_file	file;
+	struct perf_evlist	*evlist;
+	struct perf_session	*session;
+	const char		*progname;
+	int			realtime_prio;
+	bool			no_buildid;
+	bool			no_buildid_cache;
+	long			samples;
+};
+
+static int record__write(struct record *rec, void *bf, size_t size)
+{
+	if (perf_data_file__write(rec->session->file, bf, size) < 0) {
+		pr_err("failed to write perf data, error: %m\n");
+		return -1;
+	}
+
+	rec->bytes_written += size;
+	return 0;
+}
+
+static int process_synthesized_event(struct perf_tool *tool,
+				     union perf_event *event,
+				     struct perf_sample *sample __maybe_unused,
+				     struct machine *machine __maybe_unused)
+{
+	struct record *rec = container_of(tool, struct record, tool);
+	return record__write(rec, event, event->header.size);
+}
+
+static int record__mmap_read(struct record *rec, int idx)
+{
+	struct perf_mmap *md = &rec->evlist->mmap[idx];
+	u64 head = perf_mmap__read_head(md);
+	u64 old = md->prev;
+	unsigned char *data = md->base + page_size;
+	unsigned long size;
+	void *buf;
+	int rc = 0;
+
+	if (old == head)
+		return 0;
+
+	rec->samples++;
+
+	size = head - old;
+
+	if ((old & md->mask) + size != (head & md->mask)) {
+		buf = &data[old & md->mask];
+		size = md->mask + 1 - (old & md->mask);
+		old += size;
+
+		if (record__write(rec, buf, size) < 0) {
+			rc = -1;
+			goto out;
+		}
+	}
+
+	buf = &data[old & md->mask];
+	size = head - old;
+	old += size;
+
+	if (record__write(rec, buf, size) < 0) {
+		rc = -1;
+		goto out;
+	}
+
+	md->prev = old;
+	perf_evlist__mmap_consume(rec->evlist, idx);
+out:
+	return rc;
+}
+
+static volatile int done = 0;
+static volatile int signr = -1;
+static volatile int child_finished = 0;
+
+static void sig_handler(int sig)
+{
+	if (sig == SIGCHLD)
+		child_finished = 1;
+	else
+		signr = sig;
+
+	done = 1;
+}
+
+static void record__sig_exit(void)
+{
+	if (signr == -1)
+		return;
+
+	signal(signr, SIG_DFL);
+	raise(signr);
+}
+
+static int record__open(struct record *rec)
+{
+	char msg[512];
+	struct perf_evsel *pos;
+	struct perf_evlist *evlist = rec->evlist;
+	struct perf_session *session = rec->session;
+	struct record_opts *opts = &rec->opts;
+	int rc = 0;
+
+	perf_evlist__config(evlist, opts);
+
+	evlist__for_each(evlist, pos) {
+try_again:
+		if (perf_evsel__open(pos, evlist->cpus, evlist->threads) < 0) {
+			if (perf_evsel__fallback(pos, errno, msg, sizeof(msg))) {
+				if (verbose)
+					ui__warning("%s\n", msg);
+				goto try_again;
+			}
+
+			rc = -errno;
+			perf_evsel__open_strerror(pos, &opts->target,
+						  errno, msg, sizeof(msg));
+			ui__error("%s\n", msg);
+			goto out;
+		}
+	}
+
+	if (perf_evlist__apply_filters(evlist, &pos)) {
+		error("failed to set filter \"%s\" on event %s with %d (%s)\n",
+			pos->filter, perf_evsel__name(pos), errno,
+			strerror_r(errno, msg, sizeof(msg)));
+		rc = -1;
+		goto out;
+	}
+
+	if (perf_evlist__mmap(evlist, opts->mmap_pages, false) < 0) {
+		if (errno == EPERM) {
+			pr_err("Permission error mapping pages.\n"
+			       "Consider increasing "
+			       "/proc/sys/kernel/perf_event_mlock_kb,\n"
+			       "or try again with a smaller value of -m/--mmap_pages.\n"
+			       "(current value: %u)\n", opts->mmap_pages);
+			rc = -errno;
+		} else {
+			pr_err("failed to mmap with %d (%s)\n", errno,
+				strerror_r(errno, msg, sizeof(msg)));
+			rc = -errno;
+		}
+		goto out;
+	}
+
+	session->evlist = evlist;
+	perf_session__set_id_hdr_size(session);
+out:
+	return rc;
+}
+
+static int process_sample_event(struct perf_tool *tool,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel,
+				struct machine *machine)
+{
+	struct record *rec = container_of(tool, struct record, tool);
+
+	rec->samples++;
+
+	return build_id__mark_dso_hit(tool, event, sample, evsel, machine);
+}
+
+static int process_buildids(struct record *rec)
+{
+	struct perf_data_file *file  = &rec->file;
+	struct perf_session *session = rec->session;
+
+	u64 size = lseek(perf_data_file__fd(file), 0, SEEK_CUR);
+	if (size == 0)
+		return 0;
+
+	file->size = size;
+
+	/*
+	 * During this process, it'll load kernel map and replace the
+	 * dso->long_name to a real pathname it found.  In this case
+	 * we prefer the vmlinux path like
+	 *   /lib/modules/3.16.4/build/vmlinux
+	 *
+	 * rather than build-id path (in debug directory).
+	 *   $HOME/.debug/.build-id/f0/6e17aa50adf4d00b88925e03775de107611551
+	 */
+	symbol_conf.ignore_vmlinux_buildid = true;
+
+	return perf_session__process_events(session);
+}
+
+static void perf_event__synthesize_guest_os(struct machine *machine, void *data)
+{
+	int err;
+	struct perf_tool *tool = data;
+	/*
+	 *As for guest kernel when processing subcommand record&report,
+	 *we arrange module mmap prior to guest kernel mmap and trigger
+	 *a preload dso because default guest module symbols are loaded
+	 *from guest kallsyms instead of /lib/modules/XXX/XXX. This
+	 *method is used to avoid symbol missing when the first addr is
+	 *in module instead of in guest kernel.
+	 */
+	err = perf_event__synthesize_modules(tool, process_synthesized_event,
+					     machine);
+	if (err < 0)
+		pr_err("Couldn't record guest kernel [%d]'s reference"
+		       " relocation symbol.\n", machine->pid);
+
+	/*
+	 * We use _stext for guest kernel because guest kernel's /proc/kallsyms
+	 * have no _text sometimes.
+	 */
+	err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event,
+						 machine);
+	if (err < 0)
+		pr_err("Couldn't record guest kernel [%d]'s reference"
+		       " relocation symbol.\n", machine->pid);
+}
+
+static struct perf_event_header finished_round_event = {
+	.size = sizeof(struct perf_event_header),
+	.type = PERF_RECORD_FINISHED_ROUND,
+};
+
+static int record__mmap_read_all(struct record *rec)
+{
+	u64 bytes_written = rec->bytes_written;
+	int i;
+	int rc = 0;
+
+	for (i = 0; i < rec->evlist->nr_mmaps; i++) {
+		if (rec->evlist->mmap[i].base) {
+			if (record__mmap_read(rec, i) != 0) {
+				rc = -1;
+				goto out;
+			}
+		}
+	}
+
+	/*
+	 * Mark the round finished in case we wrote
+	 * at least one event.
+	 */
+	if (bytes_written != rec->bytes_written)
+		rc = record__write(rec, &finished_round_event, sizeof(finished_round_event));
+
+out:
+	return rc;
+}
+
+static void record__init_features(struct record *rec)
+{
+	struct perf_session *session = rec->session;
+	int feat;
+
+	for (feat = HEADER_FIRST_FEATURE; feat < HEADER_LAST_FEATURE; feat++)
+		perf_header__set_feat(&session->header, feat);
+
+	if (rec->no_buildid)
+		perf_header__clear_feat(&session->header, HEADER_BUILD_ID);
+
+	if (!have_tracepoints(&rec->evlist->entries))
+		perf_header__clear_feat(&session->header, HEADER_TRACING_DATA);
+
+	if (!rec->opts.branch_stack)
+		perf_header__clear_feat(&session->header, HEADER_BRANCH_STACK);
+}
+
+static volatile int workload_exec_errno;
+
+/*
+ * perf_evlist__prepare_workload will send a SIGUSR1
+ * if the fork fails, since we asked by setting its
+ * want_signal to true.
+ */
+static void workload_exec_failed_signal(int signo __maybe_unused,
+					siginfo_t *info,
+					void *ucontext __maybe_unused)
+{
+	workload_exec_errno = info->si_value.sival_int;
+	done = 1;
+	child_finished = 1;
+}
+
+static int __cmd_record(struct record *rec, int argc, const char **argv)
+{
+	int err;
+	int status = 0;
+	unsigned long waking = 0;
+	const bool forks = argc > 0;
+	struct machine *machine;
+	struct perf_tool *tool = &rec->tool;
+	struct record_opts *opts = &rec->opts;
+	struct perf_data_file *file = &rec->file;
+	struct perf_session *session;
+	bool disabled = false, draining = false;
+	int fd;
+
+	rec->progname = argv[0];
+
+	atexit(record__sig_exit);
+	signal(SIGCHLD, sig_handler);
+	signal(SIGINT, sig_handler);
+	signal(SIGTERM, sig_handler);
+
+	session = perf_session__new(file, false, tool);
+	if (session == NULL) {
+		pr_err("Perf session creation failed.\n");
+		return -1;
+	}
+
+	fd = perf_data_file__fd(file);
+	rec->session = session;
+
+	record__init_features(rec);
+
+	if (forks) {
+		err = perf_evlist__prepare_workload(rec->evlist, &opts->target,
+						    argv, file->is_pipe,
+						    workload_exec_failed_signal);
+		if (err < 0) {
+			pr_err("Couldn't run the workload!\n");
+			status = err;
+			goto out_delete_session;
+		}
+	}
+
+	if (record__open(rec) != 0) {
+		err = -1;
+		goto out_child;
+	}
+
+	if (!rec->evlist->nr_groups)
+		perf_header__clear_feat(&session->header, HEADER_GROUP_DESC);
+
+	if (file->is_pipe) {
+		err = perf_header__write_pipe(fd);
+		if (err < 0)
+			goto out_child;
+	} else {
+		err = perf_session__write_header(session, rec->evlist, fd, false);
+		if (err < 0)
+			goto out_child;
+	}
+
+	if (!rec->no_buildid
+	    && !perf_header__has_feat(&session->header, HEADER_BUILD_ID)) {
+		pr_err("Couldn't generate buildids. "
+		       "Use --no-buildid to profile anyway.\n");
+		err = -1;
+		goto out_child;
+	}
+
+	machine = &session->machines.host;
+
+	if (file->is_pipe) {
+		err = perf_event__synthesize_attrs(tool, session,
+						   process_synthesized_event);
+		if (err < 0) {
+			pr_err("Couldn't synthesize attrs.\n");
+			goto out_child;
+		}
+
+		if (have_tracepoints(&rec->evlist->entries)) {
+			/*
+			 * FIXME err <= 0 here actually means that
+			 * there were no tracepoints so its not really
+			 * an error, just that we don't need to
+			 * synthesize anything.  We really have to
+			 * return this more properly and also
+			 * propagate errors that now are calling die()
+			 */
+			err = perf_event__synthesize_tracing_data(tool,	fd, rec->evlist,
+								  process_synthesized_event);
+			if (err <= 0) {
+				pr_err("Couldn't record tracing data.\n");
+				goto out_child;
+			}
+			rec->bytes_written += err;
+		}
+	}
+
+	err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event,
+						 machine);
+	if (err < 0)
+		pr_err("Couldn't record kernel reference relocation symbol\n"
+		       "Symbol resolution may be skewed if relocation was used (e.g. kexec).\n"
+		       "Check /proc/kallsyms permission or run as root.\n");
+
+	err = perf_event__synthesize_modules(tool, process_synthesized_event,
+					     machine);
+	if (err < 0)
+		pr_err("Couldn't record kernel module information.\n"
+		       "Symbol resolution may be skewed if relocation was used (e.g. kexec).\n"
+		       "Check /proc/modules permission or run as root.\n");
+
+	if (perf_guest) {
+		machines__process_guests(&session->machines,
+					 perf_event__synthesize_guest_os, tool);
+	}
+
+	err = __machine__synthesize_threads(machine, tool, &opts->target, rec->evlist->threads,
+					    process_synthesized_event, opts->sample_address);
+	if (err != 0)
+		goto out_child;
+
+	if (rec->realtime_prio) {
+		struct sched_param param;
+
+		param.sched_priority = rec->realtime_prio;
+		if (sched_setscheduler(0, SCHED_FIFO, &param)) {
+			pr_err("Could not set realtime priority.\n");
+			err = -1;
+			goto out_child;
+		}
+	}
+
+	/*
+	 * When perf is starting the traced process, all the events
+	 * (apart from group members) have enable_on_exec=1 set,
+	 * so don't spoil it by prematurely enabling them.
+	 */
+	if (!target__none(&opts->target) && !opts->initial_delay)
+		perf_evlist__enable(rec->evlist);
+
+	/*
+	 * Let the child rip
+	 */
+	if (forks)
+		perf_evlist__start_workload(rec->evlist);
+
+	if (opts->initial_delay) {
+		usleep(opts->initial_delay * 1000);
+		perf_evlist__enable(rec->evlist);
+	}
+
+	for (;;) {
+		int hits = rec->samples;
+
+		if (record__mmap_read_all(rec) < 0) {
+			err = -1;
+			goto out_child;
+		}
+
+		if (hits == rec->samples) {
+			if (done || draining)
+				break;
+			err = perf_evlist__poll(rec->evlist, -1);
+			/*
+			 * Propagate error, only if there's any. Ignore positive
+			 * number of returned events and interrupt error.
+			 */
+			if (err > 0 || (err < 0 && errno == EINTR))
+				err = 0;
+			waking++;
+
+			if (perf_evlist__filter_pollfd(rec->evlist, POLLERR | POLLHUP) == 0)
+				draining = true;
+		}
+
+		/*
+		 * When perf is starting the traced process, at the end events
+		 * die with the process and we wait for that. Thus no need to
+		 * disable events in this case.
+		 */
+		if (done && !disabled && !target__none(&opts->target)) {
+			perf_evlist__disable(rec->evlist);
+			disabled = true;
+		}
+	}
+
+	if (forks && workload_exec_errno) {
+		char msg[STRERR_BUFSIZE];
+		const char *emsg = strerror_r(workload_exec_errno, msg, sizeof(msg));
+		pr_err("Workload failed: %s\n", emsg);
+		err = -1;
+		goto out_child;
+	}
+
+	if (!quiet)
+		fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n", waking);
+
+out_child:
+	if (forks) {
+		int exit_status;
+
+		if (!child_finished)
+			kill(rec->evlist->workload.pid, SIGTERM);
+
+		wait(&exit_status);
+
+		if (err < 0)
+			status = err;
+		else if (WIFEXITED(exit_status))
+			status = WEXITSTATUS(exit_status);
+		else if (WIFSIGNALED(exit_status))
+			signr = WTERMSIG(exit_status);
+	} else
+		status = err;
+
+	/* this will be recalculated during process_buildids() */
+	rec->samples = 0;
+
+	if (!err && !file->is_pipe) {
+		rec->session->header.data_size += rec->bytes_written;
+
+		if (!rec->no_buildid)
+			process_buildids(rec);
+		perf_session__write_header(rec->session, rec->evlist, fd, true);
+	}
+
+	if (!err && !quiet) {
+		char samples[128];
+
+		if (rec->samples)
+			scnprintf(samples, sizeof(samples),
+				  " (%" PRIu64 " samples)", rec->samples);
+		else
+			samples[0] = '\0';
+
+		fprintf(stderr,	"[ perf record: Captured and wrote %.3f MB %s%s ]\n",
+			perf_data_file__size(file) / 1024.0 / 1024.0,
+			file->path, samples);
+	}
+
+out_delete_session:
+	perf_session__delete(session);
+	return status;
+}
+
+#define BRANCH_OPT(n, m) \
+	{ .name = n, .mode = (m) }
+
+#define BRANCH_END { .name = NULL }
+
+struct branch_mode {
+	const char *name;
+	int mode;
+};
+
+static const struct branch_mode branch_modes[] = {
+	BRANCH_OPT("u", PERF_SAMPLE_BRANCH_USER),
+	BRANCH_OPT("k", PERF_SAMPLE_BRANCH_KERNEL),
+	BRANCH_OPT("hv", PERF_SAMPLE_BRANCH_HV),
+	BRANCH_OPT("any", PERF_SAMPLE_BRANCH_ANY),
+	BRANCH_OPT("any_call", PERF_SAMPLE_BRANCH_ANY_CALL),
+	BRANCH_OPT("any_ret", PERF_SAMPLE_BRANCH_ANY_RETURN),
+	BRANCH_OPT("ind_call", PERF_SAMPLE_BRANCH_IND_CALL),
+	BRANCH_OPT("abort_tx", PERF_SAMPLE_BRANCH_ABORT_TX),
+	BRANCH_OPT("in_tx", PERF_SAMPLE_BRANCH_IN_TX),
+	BRANCH_OPT("no_tx", PERF_SAMPLE_BRANCH_NO_TX),
+	BRANCH_OPT("cond", PERF_SAMPLE_BRANCH_COND),
+	BRANCH_END
+};
+
+static int
+parse_branch_stack(const struct option *opt, const char *str, int unset)
+{
+#define ONLY_PLM \
+	(PERF_SAMPLE_BRANCH_USER	|\
+	 PERF_SAMPLE_BRANCH_KERNEL	|\
+	 PERF_SAMPLE_BRANCH_HV)
+
+	uint64_t *mode = (uint64_t *)opt->value;
+	const struct branch_mode *br;
+	char *s, *os = NULL, *p;
+	int ret = -1;
+
+	if (unset)
+		return 0;
+
+	/*
+	 * cannot set it twice, -b + --branch-filter for instance
+	 */
+	if (*mode)
+		return -1;
+
+	/* str may be NULL in case no arg is passed to -b */
+	if (str) {
+		/* because str is read-only */
+		s = os = strdup(str);
+		if (!s)
+			return -1;
+
+		for (;;) {
+			p = strchr(s, ',');
+			if (p)
+				*p = '\0';
+
+			for (br = branch_modes; br->name; br++) {
+				if (!strcasecmp(s, br->name))
+					break;
+			}
+			if (!br->name) {
+				ui__warning("unknown branch filter %s,"
+					    " check man page\n", s);
+				goto error;
+			}
+
+			*mode |= br->mode;
+
+			if (!p)
+				break;
+
+			s = p + 1;
+		}
+	}
+	ret = 0;
+
+	/* default to any branch */
+	if ((*mode & ~ONLY_PLM) == 0) {
+		*mode = PERF_SAMPLE_BRANCH_ANY;
+	}
+error:
+	free(os);
+	return ret;
+}
+
+static void callchain_debug(void)
+{
+	static const char *str[CALLCHAIN_MAX] = { "NONE", "FP", "DWARF", "LBR" };
+
+	pr_debug("callchain: type %s\n", str[callchain_param.record_mode]);
+
+	if (callchain_param.record_mode == CALLCHAIN_DWARF)
+		pr_debug("callchain: stack dump size %d\n",
+			 callchain_param.dump_size);
+}
+
+int record_parse_callchain_opt(const struct option *opt __maybe_unused,
+			       const char *arg,
+			       int unset)
+{
+	int ret;
+
+	callchain_param.enabled = !unset;
+
+	/* --no-call-graph */
+	if (unset) {
+		callchain_param.record_mode = CALLCHAIN_NONE;
+		pr_debug("callchain: disabled\n");
+		return 0;
+	}
+
+	ret = parse_callchain_record_opt(arg);
+	if (!ret)
+		callchain_debug();
+
+	return ret;
+}
+
+int record_callchain_opt(const struct option *opt __maybe_unused,
+			 const char *arg __maybe_unused,
+			 int unset __maybe_unused)
+{
+	callchain_param.enabled = true;
+
+	if (callchain_param.record_mode == CALLCHAIN_NONE)
+		callchain_param.record_mode = CALLCHAIN_FP;
+
+	callchain_debug();
+	return 0;
+}
+
+static int perf_record_config(const char *var, const char *value, void *cb)
+{
+	if (!strcmp(var, "record.call-graph"))
+		var = "call-graph.record-mode"; /* fall-through */
+
+	return perf_default_config(var, value, cb);
+}
+
+struct clockid_map {
+	const char *name;
+	int clockid;
+};
+
+#define CLOCKID_MAP(n, c)	\
+	{ .name = n, .clockid = (c), }
+
+#define CLOCKID_END	{ .name = NULL, }
+
+
+/*
+ * Add the missing ones, we need to build on many distros...
+ */
+#ifndef CLOCK_MONOTONIC_RAW
+#define CLOCK_MONOTONIC_RAW 4
+#endif
+#ifndef CLOCK_BOOTTIME
+#define CLOCK_BOOTTIME 7
+#endif
+#ifndef CLOCK_TAI
+#define CLOCK_TAI 11
+#endif
+
+static const struct clockid_map clockids[] = {
+	/* available for all events, NMI safe */
+	CLOCKID_MAP("monotonic", CLOCK_MONOTONIC),
+	CLOCKID_MAP("monotonic_raw", CLOCK_MONOTONIC_RAW),
+
+	/* available for some events */
+	CLOCKID_MAP("realtime", CLOCK_REALTIME),
+	CLOCKID_MAP("boottime", CLOCK_BOOTTIME),
+	CLOCKID_MAP("tai", CLOCK_TAI),
+
+	/* available for the lazy */
+	CLOCKID_MAP("mono", CLOCK_MONOTONIC),
+	CLOCKID_MAP("raw", CLOCK_MONOTONIC_RAW),
+	CLOCKID_MAP("real", CLOCK_REALTIME),
+	CLOCKID_MAP("boot", CLOCK_BOOTTIME),
+
+	CLOCKID_END,
+};
+
+static int parse_clockid(const struct option *opt, const char *str, int unset)
+{
+	struct record_opts *opts = (struct record_opts *)opt->value;
+	const struct clockid_map *cm;
+	const char *ostr = str;
+
+	if (unset) {
+		opts->use_clockid = 0;
+		return 0;
+	}
+
+	/* no arg passed */
+	if (!str)
+		return 0;
+
+	/* no setting it twice */
+	if (opts->use_clockid)
+		return -1;
+
+	opts->use_clockid = true;
+
+	/* if its a number, we're done */
+	if (sscanf(str, "%d", &opts->clockid) == 1)
+		return 0;
+
+	/* allow a "CLOCK_" prefix to the name */
+	if (!strncasecmp(str, "CLOCK_", 6))
+		str += 6;
+
+	for (cm = clockids; cm->name; cm++) {
+		if (!strcasecmp(str, cm->name)) {
+			opts->clockid = cm->clockid;
+			return 0;
+		}
+	}
+
+	opts->use_clockid = false;
+	ui__warning("unknown clockid %s, check man page\n", ostr);
+	return -1;
+}
+
+static const char * const __record_usage[] = {
+	"perf record [<options>] [<command>]",
+	"perf record [<options>] -- <command> [<options>]",
+	NULL
+};
+const char * const *record_usage = __record_usage;
+
+/*
+ * XXX Ideally would be local to cmd_record() and passed to a record__new
+ * because we need to have access to it in record__exit, that is called
+ * after cmd_record() exits, but since record_options need to be accessible to
+ * builtin-script, leave it here.
+ *
+ * At least we don't ouch it in all the other functions here directly.
+ *
+ * Just say no to tons of global variables, sigh.
+ */
+static struct record record = {
+	.opts = {
+		.sample_time	     = true,
+		.mmap_pages	     = UINT_MAX,
+		.user_freq	     = UINT_MAX,
+		.user_interval	     = ULLONG_MAX,
+		.freq		     = 4000,
+		.target		     = {
+			.uses_mmap   = true,
+			.default_per_cpu = true,
+		},
+	},
+	.tool = {
+		.sample		= process_sample_event,
+		.fork		= perf_event__process_fork,
+		.comm		= perf_event__process_comm,
+		.mmap		= perf_event__process_mmap,
+		.mmap2		= perf_event__process_mmap2,
+	},
+};
+
+#define CALLCHAIN_HELP "setup and enables call-graph (stack chain/backtrace) recording: "
+
+#ifdef HAVE_DWARF_UNWIND_SUPPORT
+const char record_callchain_help[] = CALLCHAIN_HELP "fp dwarf lbr";
+#else
+const char record_callchain_help[] = CALLCHAIN_HELP "fp lbr";
+#endif
+
+/*
+ * XXX Will stay a global variable till we fix builtin-script.c to stop messing
+ * with it and switch to use the library functions in perf_evlist that came
+ * from builtin-record.c, i.e. use record_opts,
+ * perf_evlist__prepare_workload, etc instead of fork+exec'in 'perf record',
+ * using pipes, etc.
+ */
+struct option __record_options[] = {
+	OPT_CALLBACK('e', "event", &record.evlist, "event",
+		     "event selector. use 'perf list' to list available events",
+		     parse_events_option),
+	OPT_CALLBACK(0, "filter", &record.evlist, "filter",
+		     "event filter", parse_filter),
+	OPT_STRING('p', "pid", &record.opts.target.pid, "pid",
+		    "record events on existing process id"),
+	OPT_STRING('t', "tid", &record.opts.target.tid, "tid",
+		    "record events on existing thread id"),
+	OPT_INTEGER('r', "realtime", &record.realtime_prio,
+		    "collect data with this RT SCHED_FIFO priority"),
+	OPT_BOOLEAN(0, "no-buffering", &record.opts.no_buffering,
+		    "collect data without buffering"),
+	OPT_BOOLEAN('R', "raw-samples", &record.opts.raw_samples,
+		    "collect raw sample records from all opened counters"),
+	OPT_BOOLEAN('a', "all-cpus", &record.opts.target.system_wide,
+			    "system-wide collection from all CPUs"),
+	OPT_STRING('C', "cpu", &record.opts.target.cpu_list, "cpu",
+		    "list of cpus to monitor"),
+	OPT_U64('c', "count", &record.opts.user_interval, "event period to sample"),
+	OPT_STRING('o', "output", &record.file.path, "file",
+		    "output file name"),
+	OPT_BOOLEAN_SET('i', "no-inherit", &record.opts.no_inherit,
+			&record.opts.no_inherit_set,
+			"child tasks do not inherit counters"),
+	OPT_UINTEGER('F', "freq", &record.opts.user_freq, "profile at this frequency"),
+	OPT_CALLBACK('m', "mmap-pages", &record.opts.mmap_pages, "pages",
+		     "number of mmap data pages",
+		     perf_evlist__parse_mmap_pages),
+	OPT_BOOLEAN(0, "group", &record.opts.group,
+		    "put the counters into a counter group"),
+	OPT_CALLBACK_NOOPT('g', NULL, &record.opts,
+			   NULL, "enables call-graph recording" ,
+			   &record_callchain_opt),
+	OPT_CALLBACK(0, "call-graph", &record.opts,
+		     "mode[,dump_size]", record_callchain_help,
+		     &record_parse_callchain_opt),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show counter open errors, etc)"),
+	OPT_BOOLEAN('q', "quiet", &quiet, "don't print any message"),
+	OPT_BOOLEAN('s', "stat", &record.opts.inherit_stat,
+		    "per thread counts"),
+	OPT_BOOLEAN('d', "data", &record.opts.sample_address,
+		    "Sample addresses"),
+	OPT_BOOLEAN('T', "timestamp", &record.opts.sample_time, "Sample timestamps"),
+	OPT_BOOLEAN('P', "period", &record.opts.period, "Sample period"),
+	OPT_BOOLEAN('n', "no-samples", &record.opts.no_samples,
+		    "don't sample"),
+	OPT_BOOLEAN('N', "no-buildid-cache", &record.no_buildid_cache,
+		    "do not update the buildid cache"),
+	OPT_BOOLEAN('B', "no-buildid", &record.no_buildid,
+		    "do not collect buildids in perf.data"),
+	OPT_CALLBACK('G', "cgroup", &record.evlist, "name",
+		     "monitor event in cgroup name only",
+		     parse_cgroups),
+	OPT_UINTEGER('D', "delay", &record.opts.initial_delay,
+		  "ms to wait before starting measurement after program start"),
+	OPT_STRING('u', "uid", &record.opts.target.uid_str, "user",
+		   "user to profile"),
+
+	OPT_CALLBACK_NOOPT('b', "branch-any", &record.opts.branch_stack,
+		     "branch any", "sample any taken branches",
+		     parse_branch_stack),
+
+	OPT_CALLBACK('j', "branch-filter", &record.opts.branch_stack,
+		     "branch filter mask", "branch stack filter modes",
+		     parse_branch_stack),
+	OPT_BOOLEAN('W', "weight", &record.opts.sample_weight,
+		    "sample by weight (on special events only)"),
+	OPT_BOOLEAN(0, "transaction", &record.opts.sample_transaction,
+		    "sample transaction flags (special events only)"),
+	OPT_BOOLEAN(0, "per-thread", &record.opts.target.per_thread,
+		    "use per-thread mmaps"),
+	OPT_BOOLEAN('I', "intr-regs", &record.opts.sample_intr_regs,
+		    "Sample machine registers on interrupt"),
+	OPT_BOOLEAN(0, "running-time", &record.opts.running_time,
+		    "Record running/enabled time of read (:S) events"),
+	OPT_CALLBACK('k', "clockid", &record.opts,
+	"clockid", "clockid to use for events, see clock_gettime()",
+	parse_clockid),
+	OPT_END()
+};
+
+struct option *record_options = __record_options;
+
+int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	int err = -ENOMEM;
+	struct record *rec = &record;
+	char errbuf[BUFSIZ];
+
+	rec->evlist = perf_evlist__new();
+	if (rec->evlist == NULL)
+		return -ENOMEM;
+
+	perf_config(perf_record_config, rec);
+
+	argc = parse_options(argc, argv, record_options, record_usage,
+			    PARSE_OPT_STOP_AT_NON_OPTION);
+	if (!argc && target__none(&rec->opts.target))
+		usage_with_options(record_usage, record_options);
+
+	if (nr_cgroups && !rec->opts.target.system_wide) {
+		ui__error("cgroup monitoring only available in"
+			  " system-wide mode\n");
+		usage_with_options(record_usage, record_options);
+	}
+
+	symbol__init(NULL);
+
+	if (symbol_conf.kptr_restrict)
+		pr_warning(
+"WARNING: Kernel address maps (/proc/{kallsyms,modules}) are restricted,\n"
+"check /proc/sys/kernel/kptr_restrict.\n\n"
+"Samples in kernel functions may not be resolved if a suitable vmlinux\n"
+"file is not found in the buildid cache or in the vmlinux path.\n\n"
+"Samples in kernel modules won't be resolved at all.\n\n"
+"If some relocation was applied (e.g. kexec) symbols may be misresolved\n"
+"even with a suitable vmlinux or kallsyms file.\n\n");
+
+	if (rec->no_buildid_cache || rec->no_buildid)
+		disable_buildid_cache();
+
+	if (rec->evlist->nr_entries == 0 &&
+	    perf_evlist__add_default(rec->evlist) < 0) {
+		pr_err("Not enough memory for event selector list\n");
+		goto out_symbol_exit;
+	}
+
+	if (rec->opts.target.tid && !rec->opts.no_inherit_set)
+		rec->opts.no_inherit = true;
+
+	err = target__validate(&rec->opts.target);
+	if (err) {
+		target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
+		ui__warning("%s", errbuf);
+	}
+
+	err = target__parse_uid(&rec->opts.target);
+	if (err) {
+		int saved_errno = errno;
+
+		target__strerror(&rec->opts.target, err, errbuf, BUFSIZ);
+		ui__error("%s", errbuf);
+
+		err = -saved_errno;
+		goto out_symbol_exit;
+	}
+
+	err = -ENOMEM;
+	if (perf_evlist__create_maps(rec->evlist, &rec->opts.target) < 0)
+		usage_with_options(record_usage, record_options);
+
+	if (record_opts__config(&rec->opts)) {
+		err = -EINVAL;
+		goto out_symbol_exit;
+	}
+
+	err = __cmd_record(&record, argc, argv);
+out_symbol_exit:
+	perf_evlist__delete(rec->evlist);
+	symbol__exit();
+	return err;
+}
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
new file mode 100644
index 000000000..b63aeda71
--- /dev/null
+++ b/tools/perf/builtin-report.c
@@ -0,0 +1,877 @@
+/*
+ * builtin-report.c
+ *
+ * Builtin report command: Analyze the perf.data input file,
+ * look up and read DSOs and symbol information and display
+ * a histogram of results, along various sorting keys.
+ */
+#include "builtin.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+
+#include "util/annotate.h"
+#include "util/color.h"
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include "util/symbol.h"
+#include "util/callchain.h"
+#include "util/strlist.h"
+#include "util/values.h"
+
+#include "perf.h"
+#include "util/debug.h"
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/header.h"
+#include "util/session.h"
+#include "util/tool.h"
+
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+
+#include "util/thread.h"
+#include "util/sort.h"
+#include "util/hist.h"
+#include "util/data.h"
+#include "arch/common.h"
+
+#include <dlfcn.h>
+#include <linux/bitmap.h>
+
+struct report {
+	struct perf_tool	tool;
+	struct perf_session	*session;
+	bool			force, use_tui, use_gtk, use_stdio;
+	bool			hide_unresolved;
+	bool			dont_use_callchains;
+	bool			show_full_info;
+	bool			show_threads;
+	bool			inverted_callchain;
+	bool			mem_mode;
+	bool			header;
+	bool			header_only;
+	int			max_stack;
+	struct perf_read_values	show_threads_values;
+	const char		*pretty_printing_style;
+	const char		*cpu_list;
+	const char		*symbol_filter_str;
+	float			min_percent;
+	u64			nr_entries;
+	u64			queue_size;
+	DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
+};
+
+static int report__config(const char *var, const char *value, void *cb)
+{
+	struct report *rep = cb;
+
+	if (!strcmp(var, "report.group")) {
+		symbol_conf.event_group = perf_config_bool(var, value);
+		return 0;
+	}
+	if (!strcmp(var, "report.percent-limit")) {
+		rep->min_percent = strtof(value, NULL);
+		return 0;
+	}
+	if (!strcmp(var, "report.children")) {
+		symbol_conf.cumulate_callchain = perf_config_bool(var, value);
+		return 0;
+	}
+	if (!strcmp(var, "report.queue-size")) {
+		rep->queue_size = perf_config_u64(var, value);
+		return 0;
+	}
+
+	return perf_default_config(var, value, cb);
+}
+
+static int hist_iter__report_callback(struct hist_entry_iter *iter,
+				      struct addr_location *al, bool single,
+				      void *arg)
+{
+	int err = 0;
+	struct report *rep = arg;
+	struct hist_entry *he = iter->he;
+	struct perf_evsel *evsel = iter->evsel;
+	struct mem_info *mi;
+	struct branch_info *bi;
+
+	if (!ui__has_annotation())
+		return 0;
+
+	if (sort__mode == SORT_MODE__BRANCH) {
+		bi = he->branch_info;
+		err = addr_map_symbol__inc_samples(&bi->from, evsel->idx);
+		if (err)
+			goto out;
+
+		err = addr_map_symbol__inc_samples(&bi->to, evsel->idx);
+
+	} else if (rep->mem_mode) {
+		mi = he->mem_info;
+		err = addr_map_symbol__inc_samples(&mi->daddr, evsel->idx);
+		if (err)
+			goto out;
+
+		err = hist_entry__inc_addr_samples(he, evsel->idx, al->addr);
+
+	} else if (symbol_conf.cumulate_callchain) {
+		if (single)
+			err = hist_entry__inc_addr_samples(he, evsel->idx,
+							   al->addr);
+	} else {
+		err = hist_entry__inc_addr_samples(he, evsel->idx, al->addr);
+	}
+
+out:
+	return err;
+}
+
+static int process_sample_event(struct perf_tool *tool,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel,
+				struct machine *machine)
+{
+	struct report *rep = container_of(tool, struct report, tool);
+	struct addr_location al;
+	struct hist_entry_iter iter = {
+		.hide_unresolved = rep->hide_unresolved,
+		.add_entry_cb = hist_iter__report_callback,
+	};
+	int ret;
+
+	if (perf_event__preprocess_sample(event, machine, &al, sample) < 0) {
+		pr_debug("problem processing %d event, skipping it.\n",
+			 event->header.type);
+		return -1;
+	}
+
+	if (rep->hide_unresolved && al.sym == NULL)
+		return 0;
+
+	if (rep->cpu_list && !test_bit(sample->cpu, rep->cpu_bitmap))
+		return 0;
+
+	if (sort__mode == SORT_MODE__BRANCH)
+		iter.ops = &hist_iter_branch;
+	else if (rep->mem_mode)
+		iter.ops = &hist_iter_mem;
+	else if (symbol_conf.cumulate_callchain)
+		iter.ops = &hist_iter_cumulative;
+	else
+		iter.ops = &hist_iter_normal;
+
+	if (al.map != NULL)
+		al.map->dso->hit = 1;
+
+	ret = hist_entry_iter__add(&iter, &al, evsel, sample, rep->max_stack,
+				   rep);
+	if (ret < 0)
+		pr_debug("problem adding hist entry, skipping event\n");
+
+	return ret;
+}
+
+static int process_read_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample __maybe_unused,
+			      struct perf_evsel *evsel,
+			      struct machine *machine __maybe_unused)
+{
+	struct report *rep = container_of(tool, struct report, tool);
+
+	if (rep->show_threads) {
+		const char *name = evsel ? perf_evsel__name(evsel) : "unknown";
+		perf_read_values_add_value(&rep->show_threads_values,
+					   event->read.pid, event->read.tid,
+					   event->read.id,
+					   name,
+					   event->read.value);
+	}
+
+	dump_printf(": %d %d %s %" PRIu64 "\n", event->read.pid, event->read.tid,
+		    evsel ? perf_evsel__name(evsel) : "FAIL",
+		    event->read.value);
+
+	return 0;
+}
+
+/* For pipe mode, sample_type is not currently set */
+static int report__setup_sample_type(struct report *rep)
+{
+	struct perf_session *session = rep->session;
+	u64 sample_type = perf_evlist__combined_sample_type(session->evlist);
+	bool is_pipe = perf_data_file__is_pipe(session->file);
+
+	if (!is_pipe && !(sample_type & PERF_SAMPLE_CALLCHAIN)) {
+		if (sort__has_parent) {
+			ui__error("Selected --sort parent, but no "
+				    "callchain data. Did you call "
+				    "'perf record' without -g?\n");
+			return -EINVAL;
+		}
+		if (symbol_conf.use_callchain) {
+			ui__error("Selected -g or --branch-history but no "
+				  "callchain data. Did\n"
+				  "you call 'perf record' without -g?\n");
+			return -1;
+		}
+	} else if (!rep->dont_use_callchains &&
+		   callchain_param.mode != CHAIN_NONE &&
+		   !symbol_conf.use_callchain) {
+			symbol_conf.use_callchain = true;
+			if (callchain_register_param(&callchain_param) < 0) {
+				ui__error("Can't register callchain params.\n");
+				return -EINVAL;
+			}
+	}
+
+	if (symbol_conf.cumulate_callchain) {
+		/* Silently ignore if callchain is missing */
+		if (!(sample_type & PERF_SAMPLE_CALLCHAIN)) {
+			symbol_conf.cumulate_callchain = false;
+			perf_hpp__cancel_cumulate();
+		}
+	}
+
+	if (sort__mode == SORT_MODE__BRANCH) {
+		if (!is_pipe &&
+		    !(sample_type & PERF_SAMPLE_BRANCH_STACK)) {
+			ui__error("Selected -b but no branch data. "
+				  "Did you call perf record without -b?\n");
+			return -1;
+		}
+	}
+
+	if (symbol_conf.use_callchain || symbol_conf.cumulate_callchain) {
+		if ((sample_type & PERF_SAMPLE_REGS_USER) &&
+		    (sample_type & PERF_SAMPLE_STACK_USER))
+			callchain_param.record_mode = CALLCHAIN_DWARF;
+		else if (sample_type & PERF_SAMPLE_BRANCH_STACK)
+			callchain_param.record_mode = CALLCHAIN_LBR;
+		else
+			callchain_param.record_mode = CALLCHAIN_FP;
+	}
+	return 0;
+}
+
+static void sig_handler(int sig __maybe_unused)
+{
+	session_done = 1;
+}
+
+static size_t hists__fprintf_nr_sample_events(struct hists *hists, struct report *rep,
+					      const char *evname, FILE *fp)
+{
+	size_t ret;
+	char unit;
+	unsigned long nr_samples = hists->stats.nr_events[PERF_RECORD_SAMPLE];
+	u64 nr_events = hists->stats.total_period;
+	struct perf_evsel *evsel = hists_to_evsel(hists);
+	char buf[512];
+	size_t size = sizeof(buf);
+
+	if (symbol_conf.filter_relative) {
+		nr_samples = hists->stats.nr_non_filtered_samples;
+		nr_events = hists->stats.total_non_filtered_period;
+	}
+
+	if (perf_evsel__is_group_event(evsel)) {
+		struct perf_evsel *pos;
+
+		perf_evsel__group_desc(evsel, buf, size);
+		evname = buf;
+
+		for_each_group_member(pos, evsel) {
+			const struct hists *pos_hists = evsel__hists(pos);
+
+			if (symbol_conf.filter_relative) {
+				nr_samples += pos_hists->stats.nr_non_filtered_samples;
+				nr_events += pos_hists->stats.total_non_filtered_period;
+			} else {
+				nr_samples += pos_hists->stats.nr_events[PERF_RECORD_SAMPLE];
+				nr_events += pos_hists->stats.total_period;
+			}
+		}
+	}
+
+	nr_samples = convert_unit(nr_samples, &unit);
+	ret = fprintf(fp, "# Samples: %lu%c", nr_samples, unit);
+	if (evname != NULL)
+		ret += fprintf(fp, " of event '%s'", evname);
+
+	if (rep->mem_mode) {
+		ret += fprintf(fp, "\n# Total weight : %" PRIu64, nr_events);
+		ret += fprintf(fp, "\n# Sort order   : %s", sort_order ? : default_mem_sort_order);
+	} else
+		ret += fprintf(fp, "\n# Event count (approx.): %" PRIu64, nr_events);
+	return ret + fprintf(fp, "\n#\n");
+}
+
+static int perf_evlist__tty_browse_hists(struct perf_evlist *evlist,
+					 struct report *rep,
+					 const char *help)
+{
+	struct perf_evsel *pos;
+
+	evlist__for_each(evlist, pos) {
+		struct hists *hists = evsel__hists(pos);
+		const char *evname = perf_evsel__name(pos);
+
+		if (symbol_conf.event_group &&
+		    !perf_evsel__is_group_leader(pos))
+			continue;
+
+		hists__fprintf_nr_sample_events(hists, rep, evname, stdout);
+		hists__fprintf(hists, true, 0, 0, rep->min_percent, stdout);
+		fprintf(stdout, "\n\n");
+	}
+
+	if (sort_order == NULL &&
+	    parent_pattern == default_parent_pattern) {
+		fprintf(stdout, "#\n# (%s)\n#\n", help);
+
+		if (rep->show_threads) {
+			bool style = !strcmp(rep->pretty_printing_style, "raw");
+			perf_read_values_display(stdout, &rep->show_threads_values,
+						 style);
+			perf_read_values_destroy(&rep->show_threads_values);
+		}
+	}
+
+	return 0;
+}
+
+static void report__warn_kptr_restrict(const struct report *rep)
+{
+	struct map *kernel_map = rep->session->machines.host.vmlinux_maps[MAP__FUNCTION];
+	struct kmap *kernel_kmap = kernel_map ? map__kmap(kernel_map) : NULL;
+
+	if (kernel_map == NULL ||
+	    (kernel_map->dso->hit &&
+	     (kernel_kmap->ref_reloc_sym == NULL ||
+	      kernel_kmap->ref_reloc_sym->addr == 0))) {
+		const char *desc =
+		    "As no suitable kallsyms nor vmlinux was found, kernel samples\n"
+		    "can't be resolved.";
+
+		if (kernel_map) {
+			const struct dso *kdso = kernel_map->dso;
+			if (!RB_EMPTY_ROOT(&kdso->symbols[MAP__FUNCTION])) {
+				desc = "If some relocation was applied (e.g. "
+				       "kexec) symbols may be misresolved.";
+			}
+		}
+
+		ui__warning(
+"Kernel address maps (/proc/{kallsyms,modules}) were restricted.\n\n"
+"Check /proc/sys/kernel/kptr_restrict before running 'perf record'.\n\n%s\n\n"
+"Samples in kernel modules can't be resolved as well.\n\n",
+		desc);
+	}
+}
+
+static int report__gtk_browse_hists(struct report *rep, const char *help)
+{
+	int (*hist_browser)(struct perf_evlist *evlist, const char *help,
+			    struct hist_browser_timer *timer, float min_pcnt);
+
+	hist_browser = dlsym(perf_gtk_handle, "perf_evlist__gtk_browse_hists");
+
+	if (hist_browser == NULL) {
+		ui__error("GTK browser not found!\n");
+		return -1;
+	}
+
+	return hist_browser(rep->session->evlist, help, NULL, rep->min_percent);
+}
+
+static int report__browse_hists(struct report *rep)
+{
+	int ret;
+	struct perf_session *session = rep->session;
+	struct perf_evlist *evlist = session->evlist;
+	const char *help = "For a higher level overview, try: perf report --sort comm,dso";
+
+	switch (use_browser) {
+	case 1:
+		ret = perf_evlist__tui_browse_hists(evlist, help, NULL,
+						    rep->min_percent,
+						    &session->header.env);
+		/*
+		 * Usually "ret" is the last pressed key, and we only
+		 * care if the key notifies us to switch data file.
+		 */
+		if (ret != K_SWITCH_INPUT_DATA)
+			ret = 0;
+		break;
+	case 2:
+		ret = report__gtk_browse_hists(rep, help);
+		break;
+	default:
+		ret = perf_evlist__tty_browse_hists(evlist, rep, help);
+		break;
+	}
+
+	return ret;
+}
+
+static void report__collapse_hists(struct report *rep)
+{
+	struct ui_progress prog;
+	struct perf_evsel *pos;
+
+	ui_progress__init(&prog, rep->nr_entries, "Merging related events...");
+
+	evlist__for_each(rep->session->evlist, pos) {
+		struct hists *hists = evsel__hists(pos);
+
+		if (pos->idx == 0)
+			hists->symbol_filter_str = rep->symbol_filter_str;
+
+		hists__collapse_resort(hists, &prog);
+
+		/* Non-group events are considered as leader */
+		if (symbol_conf.event_group &&
+		    !perf_evsel__is_group_leader(pos)) {
+			struct hists *leader_hists = evsel__hists(pos->leader);
+
+			hists__match(leader_hists, hists);
+			hists__link(leader_hists, hists);
+		}
+	}
+
+	ui_progress__finish();
+}
+
+static void report__output_resort(struct report *rep)
+{
+	struct ui_progress prog;
+	struct perf_evsel *pos;
+
+	ui_progress__init(&prog, rep->nr_entries, "Sorting events for output...");
+
+	evlist__for_each(rep->session->evlist, pos)
+		hists__output_resort(evsel__hists(pos), &prog);
+
+	ui_progress__finish();
+}
+
+static int __cmd_report(struct report *rep)
+{
+	int ret;
+	struct perf_session *session = rep->session;
+	struct perf_evsel *pos;
+	struct perf_data_file *file = session->file;
+
+	signal(SIGINT, sig_handler);
+
+	if (rep->cpu_list) {
+		ret = perf_session__cpu_bitmap(session, rep->cpu_list,
+					       rep->cpu_bitmap);
+		if (ret)
+			return ret;
+	}
+
+	if (rep->show_threads)
+		perf_read_values_init(&rep->show_threads_values);
+
+	ret = report__setup_sample_type(rep);
+	if (ret)
+		return ret;
+
+	ret = perf_session__process_events(session);
+	if (ret)
+		return ret;
+
+	report__warn_kptr_restrict(rep);
+
+	evlist__for_each(session->evlist, pos)
+		rep->nr_entries += evsel__hists(pos)->nr_entries;
+
+	if (use_browser == 0) {
+		if (verbose > 3)
+			perf_session__fprintf(session, stdout);
+
+		if (verbose > 2)
+			perf_session__fprintf_dsos(session, stdout);
+
+		if (dump_trace) {
+			perf_session__fprintf_nr_events(session, stdout);
+			perf_evlist__fprintf_nr_events(session->evlist, stdout);
+			return 0;
+		}
+	}
+
+	report__collapse_hists(rep);
+
+	if (session_done())
+		return 0;
+
+	/*
+	 * recalculate number of entries after collapsing since it
+	 * might be changed during the collapse phase.
+	 */
+	rep->nr_entries = 0;
+	evlist__for_each(session->evlist, pos)
+		rep->nr_entries += evsel__hists(pos)->nr_entries;
+
+	if (rep->nr_entries == 0) {
+		ui__error("The %s file has no samples!\n", file->path);
+		return 0;
+	}
+
+	report__output_resort(rep);
+
+	return report__browse_hists(rep);
+}
+
+static int
+report_parse_callchain_opt(const struct option *opt, const char *arg, int unset)
+{
+	struct report *rep = (struct report *)opt->value;
+
+	/*
+	 * --no-call-graph
+	 */
+	if (unset) {
+		rep->dont_use_callchains = true;
+		return 0;
+	}
+
+	return parse_callchain_report_opt(arg);
+}
+
+int
+report_parse_ignore_callees_opt(const struct option *opt __maybe_unused,
+				const char *arg, int unset __maybe_unused)
+{
+	if (arg) {
+		int err = regcomp(&ignore_callees_regex, arg, REG_EXTENDED);
+		if (err) {
+			char buf[BUFSIZ];
+			regerror(err, &ignore_callees_regex, buf, sizeof(buf));
+			pr_err("Invalid --ignore-callees regex: %s\n%s", arg, buf);
+			return -1;
+		}
+		have_ignore_callees = 1;
+	}
+
+	return 0;
+}
+
+static int
+parse_branch_mode(const struct option *opt __maybe_unused,
+		  const char *str __maybe_unused, int unset)
+{
+	int *branch_mode = opt->value;
+
+	*branch_mode = !unset;
+	return 0;
+}
+
+static int
+parse_percent_limit(const struct option *opt, const char *str,
+		    int unset __maybe_unused)
+{
+	struct report *rep = opt->value;
+
+	rep->min_percent = strtof(str, NULL);
+	return 0;
+}
+
+int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	struct perf_session *session;
+	struct stat st;
+	bool has_br_stack = false;
+	int branch_mode = -1;
+	bool branch_call_mode = false;
+	char callchain_default_opt[] = "fractal,0.5,callee";
+	const char * const report_usage[] = {
+		"perf report [<options>]",
+		NULL
+	};
+	struct report report = {
+		.tool = {
+			.sample		 = process_sample_event,
+			.mmap		 = perf_event__process_mmap,
+			.mmap2		 = perf_event__process_mmap2,
+			.comm		 = perf_event__process_comm,
+			.exit		 = perf_event__process_exit,
+			.fork		 = perf_event__process_fork,
+			.lost		 = perf_event__process_lost,
+			.read		 = process_read_event,
+			.attr		 = perf_event__process_attr,
+			.tracing_data	 = perf_event__process_tracing_data,
+			.build_id	 = perf_event__process_build_id,
+			.ordered_events	 = true,
+			.ordering_requires_timestamps = true,
+		},
+		.max_stack		 = PERF_MAX_STACK_DEPTH,
+		.pretty_printing_style	 = "normal",
+	};
+	const struct option options[] = {
+	OPT_STRING('i', "input", &input_name, "file",
+		    "input file name"),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+		   "file", "vmlinux pathname"),
+	OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name,
+		   "file", "kallsyms pathname"),
+	OPT_BOOLEAN('f', "force", &report.force, "don't complain, do it"),
+	OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules,
+		    "load module symbols - WARNING: use only with -k and LIVE kernel"),
+	OPT_BOOLEAN('n', "show-nr-samples", &symbol_conf.show_nr_samples,
+		    "Show a column with the number of samples"),
+	OPT_BOOLEAN('T', "threads", &report.show_threads,
+		    "Show per-thread event counters"),
+	OPT_STRING(0, "pretty", &report.pretty_printing_style, "key",
+		   "pretty printing style key: normal raw"),
+	OPT_BOOLEAN(0, "tui", &report.use_tui, "Use the TUI interface"),
+	OPT_BOOLEAN(0, "gtk", &report.use_gtk, "Use the GTK2 interface"),
+	OPT_BOOLEAN(0, "stdio", &report.use_stdio,
+		    "Use the stdio interface"),
+	OPT_BOOLEAN(0, "header", &report.header, "Show data header."),
+	OPT_BOOLEAN(0, "header-only", &report.header_only,
+		    "Show only data header."),
+	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
+		   "sort by key(s): pid, comm, dso, symbol, parent, cpu, srcline, ..."
+		   " Please refer the man page for the complete list."),
+	OPT_STRING('F', "fields", &field_order, "key[,keys...]",
+		   "output field(s): overhead, period, sample plus all of sort keys"),
+	OPT_BOOLEAN(0, "showcpuutilization", &symbol_conf.show_cpu_utilization,
+		    "Show sample percentage for different cpu modes"),
+	OPT_STRING('p', "parent", &parent_pattern, "regex",
+		   "regex filter to identify parent, see: '--sort parent'"),
+	OPT_BOOLEAN('x', "exclude-other", &symbol_conf.exclude_other,
+		    "Only display entries with parent-match"),
+	OPT_CALLBACK_DEFAULT('g', "call-graph", &report, "output_type,min_percent[,print_limit],call_order[,branch]",
+		     "Display callchains using output_type (graph, flat, fractal, or none) , min percent threshold, optional print limit, callchain order, key (function or address), add branches. "
+		     "Default: fractal,0.5,callee,function", &report_parse_callchain_opt, callchain_default_opt),
+	OPT_BOOLEAN(0, "children", &symbol_conf.cumulate_callchain,
+		    "Accumulate callchains of children and show total overhead as well"),
+	OPT_INTEGER(0, "max-stack", &report.max_stack,
+		    "Set the maximum stack depth when parsing the callchain, "
+		    "anything beyond the specified depth will be ignored. "
+		    "Default: " __stringify(PERF_MAX_STACK_DEPTH)),
+	OPT_BOOLEAN('G', "inverted", &report.inverted_callchain,
+		    "alias for inverted call graph"),
+	OPT_CALLBACK(0, "ignore-callees", NULL, "regex",
+		   "ignore callees of these functions in call graphs",
+		   report_parse_ignore_callees_opt),
+	OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]",
+		   "only consider symbols in these dsos"),
+	OPT_STRING('c', "comms", &symbol_conf.comm_list_str, "comm[,comm...]",
+		   "only consider symbols in these comms"),
+	OPT_STRING(0, "pid", &symbol_conf.pid_list_str, "pid[,pid...]",
+		   "only consider symbols in these pids"),
+	OPT_STRING(0, "tid", &symbol_conf.tid_list_str, "tid[,tid...]",
+		   "only consider symbols in these tids"),
+	OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]",
+		   "only consider these symbols"),
+	OPT_STRING(0, "symbol-filter", &report.symbol_filter_str, "filter",
+		   "only show symbols that (partially) match with this filter"),
+	OPT_STRING('w', "column-widths", &symbol_conf.col_width_list_str,
+		   "width[,width...]",
+		   "don't try to adjust column width, use these fixed values"),
+	OPT_STRING_NOEMPTY('t', "field-separator", &symbol_conf.field_sep, "separator",
+		   "separator for columns, no spaces will be added between "
+		   "columns '.' is reserved."),
+	OPT_BOOLEAN('U', "hide-unresolved", &report.hide_unresolved,
+		    "Only display entries resolved to a symbol"),
+	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
+		    "Look for files with symbols relative to this directory"),
+	OPT_STRING('C', "cpu", &report.cpu_list, "cpu",
+		   "list of cpus to profile"),
+	OPT_BOOLEAN('I', "show-info", &report.show_full_info,
+		    "Display extended information about perf.data file"),
+	OPT_BOOLEAN(0, "source", &symbol_conf.annotate_src,
+		    "Interleave source code with assembly code (default)"),
+	OPT_BOOLEAN(0, "asm-raw", &symbol_conf.annotate_asm_raw,
+		    "Display raw encoding of assembly instructions (default)"),
+	OPT_STRING('M', "disassembler-style", &disassembler_style, "disassembler style",
+		   "Specify disassembler style (e.g. -M intel for intel syntax)"),
+	OPT_BOOLEAN(0, "show-total-period", &symbol_conf.show_total_period,
+		    "Show a column with the sum of periods"),
+	OPT_BOOLEAN(0, "group", &symbol_conf.event_group,
+		    "Show event group information together"),
+	OPT_CALLBACK_NOOPT('b', "branch-stack", &branch_mode, "",
+		    "use branch records for per branch histogram filling",
+		    parse_branch_mode),
+	OPT_BOOLEAN(0, "branch-history", &branch_call_mode,
+		    "add last branch records to call history"),
+	OPT_STRING(0, "objdump", &objdump_path, "path",
+		   "objdump binary to use for disassembly and annotations"),
+	OPT_BOOLEAN(0, "demangle", &symbol_conf.demangle,
+		    "Disable symbol demangling"),
+	OPT_BOOLEAN(0, "demangle-kernel", &symbol_conf.demangle_kernel,
+		    "Enable kernel symbol demangling"),
+	OPT_BOOLEAN(0, "mem-mode", &report.mem_mode, "mem access profile"),
+	OPT_CALLBACK(0, "percent-limit", &report, "percent",
+		     "Don't show entries under that percent", parse_percent_limit),
+	OPT_CALLBACK(0, "percentage", NULL, "relative|absolute",
+		     "how to display percentage of filtered entries", parse_filter_percentage),
+	OPT_END()
+	};
+	struct perf_data_file file = {
+		.mode  = PERF_DATA_MODE_READ,
+	};
+	int ret = hists__init();
+
+	if (ret < 0)
+		return ret;
+
+	perf_config(report__config, &report);
+
+	argc = parse_options(argc, argv, options, report_usage, 0);
+
+	if (report.use_stdio)
+		use_browser = 0;
+	else if (report.use_tui)
+		use_browser = 1;
+	else if (report.use_gtk)
+		use_browser = 2;
+
+	if (report.inverted_callchain)
+		callchain_param.order = ORDER_CALLER;
+
+	if (!input_name || !strlen(input_name)) {
+		if (!fstat(STDIN_FILENO, &st) && S_ISFIFO(st.st_mode))
+			input_name = "-";
+		else
+			input_name = "perf.data";
+	}
+
+	file.path  = input_name;
+	file.force = report.force;
+
+repeat:
+	session = perf_session__new(&file, false, &report.tool);
+	if (session == NULL)
+		return -1;
+
+	if (report.queue_size) {
+		ordered_events__set_alloc_size(&session->ordered_events,
+					       report.queue_size);
+	}
+
+	report.session = session;
+
+	has_br_stack = perf_header__has_feat(&session->header,
+					     HEADER_BRANCH_STACK);
+
+	/*
+	 * Branch mode is a tristate:
+	 * -1 means default, so decide based on the file having branch data.
+	 * 0/1 means the user chose a mode.
+	 */
+	if (((branch_mode == -1 && has_br_stack) || branch_mode == 1) &&
+	    !branch_call_mode) {
+		sort__mode = SORT_MODE__BRANCH;
+		symbol_conf.cumulate_callchain = false;
+	}
+	if (branch_call_mode) {
+		callchain_param.key = CCKEY_ADDRESS;
+		callchain_param.branch_callstack = 1;
+		symbol_conf.use_callchain = true;
+		callchain_register_param(&callchain_param);
+		if (sort_order == NULL)
+			sort_order = "srcline,symbol,dso";
+	}
+
+	if (report.mem_mode) {
+		if (sort__mode == SORT_MODE__BRANCH) {
+			pr_err("branch and mem mode incompatible\n");
+			goto error;
+		}
+		sort__mode = SORT_MODE__MEMORY;
+		symbol_conf.cumulate_callchain = false;
+	}
+
+	if (setup_sorting() < 0) {
+		if (sort_order)
+			parse_options_usage(report_usage, options, "s", 1);
+		if (field_order)
+			parse_options_usage(sort_order ? NULL : report_usage,
+					    options, "F", 1);
+		goto error;
+	}
+
+	/* Force tty output for header output. */
+	if (report.header || report.header_only)
+		use_browser = 0;
+
+	if (strcmp(input_name, "-") != 0)
+		setup_browser(true);
+	else
+		use_browser = 0;
+
+	if (report.header || report.header_only) {
+		perf_session__fprintf_info(session, stdout,
+					   report.show_full_info);
+		if (report.header_only)
+			return 0;
+	} else if (use_browser == 0) {
+		fputs("# To display the perf.data header info, please use --header/--header-only options.\n#\n",
+		      stdout);
+	}
+
+	/*
+	 * Only in the TUI browser we are doing integrated annotation,
+	 * so don't allocate extra space that won't be used in the stdio
+	 * implementation.
+	 */
+	if (ui__has_annotation()) {
+		symbol_conf.priv_size = sizeof(struct annotation);
+		machines__set_symbol_filter(&session->machines,
+					    symbol__annotate_init);
+		/*
+ 		 * For searching by name on the "Browse map details".
+ 		 * providing it only in verbose mode not to bloat too
+ 		 * much struct symbol.
+ 		 */
+		if (verbose) {
+			/*
+			 * XXX: Need to provide a less kludgy way to ask for
+			 * more space per symbol, the u32 is for the index on
+			 * the ui browser.
+			 * See symbol__browser_index.
+			 */
+			symbol_conf.priv_size += sizeof(u32);
+			symbol_conf.sort_by_name = true;
+		}
+	}
+
+	if (symbol__init(&session->header.env) < 0)
+		goto error;
+
+	if (argc) {
+		/*
+		 * Special case: if there's an argument left then assume that
+		 * it's a symbol filter:
+		 */
+		if (argc > 1)
+			usage_with_options(report_usage, options);
+
+		report.symbol_filter_str = argv[0];
+	}
+
+	sort__setup_elide(stdout);
+
+	ret = __cmd_report(&report);
+	if (ret == K_SWITCH_INPUT_DATA) {
+		perf_session__delete(session);
+		goto repeat;
+	} else
+		ret = 0;
+
+error:
+	perf_session__delete(session);
+	return ret;
+}
diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
new file mode 100644
index 000000000..5275bab70
--- /dev/null
+++ b/tools/perf/builtin-sched.c
@@ -0,0 +1,1808 @@
+#include "builtin.h"
+#include "perf.h"
+
+#include "util/util.h"
+#include "util/evlist.h"
+#include "util/cache.h"
+#include "util/evsel.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+#include "util/session.h"
+#include "util/tool.h"
+#include "util/cloexec.h"
+
+#include "util/parse-options.h"
+#include "util/trace-event.h"
+
+#include "util/debug.h"
+
+#include <sys/prctl.h>
+#include <sys/resource.h>
+
+#include <semaphore.h>
+#include <pthread.h>
+#include <math.h>
+#include <api/fs/fs.h>
+
+#define PR_SET_NAME		15               /* Set process name */
+#define MAX_CPUS		4096
+#define COMM_LEN		20
+#define SYM_LEN			129
+#define MAX_PID			1024000
+
+struct sched_atom;
+
+struct task_desc {
+	unsigned long		nr;
+	unsigned long		pid;
+	char			comm[COMM_LEN];
+
+	unsigned long		nr_events;
+	unsigned long		curr_event;
+	struct sched_atom	**atoms;
+
+	pthread_t		thread;
+	sem_t			sleep_sem;
+
+	sem_t			ready_for_work;
+	sem_t			work_done_sem;
+
+	u64			cpu_usage;
+};
+
+enum sched_event_type {
+	SCHED_EVENT_RUN,
+	SCHED_EVENT_SLEEP,
+	SCHED_EVENT_WAKEUP,
+	SCHED_EVENT_MIGRATION,
+};
+
+struct sched_atom {
+	enum sched_event_type	type;
+	int			specific_wait;
+	u64			timestamp;
+	u64			duration;
+	unsigned long		nr;
+	sem_t			*wait_sem;
+	struct task_desc	*wakee;
+};
+
+#define TASK_STATE_TO_CHAR_STR "RSDTtZXxKWP"
+
+enum thread_state {
+	THREAD_SLEEPING = 0,
+	THREAD_WAIT_CPU,
+	THREAD_SCHED_IN,
+	THREAD_IGNORE
+};
+
+struct work_atom {
+	struct list_head	list;
+	enum thread_state	state;
+	u64			sched_out_time;
+	u64			wake_up_time;
+	u64			sched_in_time;
+	u64			runtime;
+};
+
+struct work_atoms {
+	struct list_head	work_list;
+	struct thread		*thread;
+	struct rb_node		node;
+	u64			max_lat;
+	u64			max_lat_at;
+	u64			total_lat;
+	u64			nb_atoms;
+	u64			total_runtime;
+};
+
+typedef int (*sort_fn_t)(struct work_atoms *, struct work_atoms *);
+
+struct perf_sched;
+
+struct trace_sched_handler {
+	int (*switch_event)(struct perf_sched *sched, struct perf_evsel *evsel,
+			    struct perf_sample *sample, struct machine *machine);
+
+	int (*runtime_event)(struct perf_sched *sched, struct perf_evsel *evsel,
+			     struct perf_sample *sample, struct machine *machine);
+
+	int (*wakeup_event)(struct perf_sched *sched, struct perf_evsel *evsel,
+			    struct perf_sample *sample, struct machine *machine);
+
+	/* PERF_RECORD_FORK event, not sched_process_fork tracepoint */
+	int (*fork_event)(struct perf_sched *sched, union perf_event *event,
+			  struct machine *machine);
+
+	int (*migrate_task_event)(struct perf_sched *sched,
+				  struct perf_evsel *evsel,
+				  struct perf_sample *sample,
+				  struct machine *machine);
+};
+
+struct perf_sched {
+	struct perf_tool tool;
+	const char	 *sort_order;
+	unsigned long	 nr_tasks;
+	struct task_desc **pid_to_task;
+	struct task_desc **tasks;
+	const struct trace_sched_handler *tp_handler;
+	pthread_mutex_t	 start_work_mutex;
+	pthread_mutex_t	 work_done_wait_mutex;
+	int		 profile_cpu;
+/*
+ * Track the current task - that way we can know whether there's any
+ * weird events, such as a task being switched away that is not current.
+ */
+	int		 max_cpu;
+	u32		 curr_pid[MAX_CPUS];
+	struct thread	 *curr_thread[MAX_CPUS];
+	char		 next_shortname1;
+	char		 next_shortname2;
+	unsigned int	 replay_repeat;
+	unsigned long	 nr_run_events;
+	unsigned long	 nr_sleep_events;
+	unsigned long	 nr_wakeup_events;
+	unsigned long	 nr_sleep_corrections;
+	unsigned long	 nr_run_events_optimized;
+	unsigned long	 targetless_wakeups;
+	unsigned long	 multitarget_wakeups;
+	unsigned long	 nr_runs;
+	unsigned long	 nr_timestamps;
+	unsigned long	 nr_unordered_timestamps;
+	unsigned long	 nr_context_switch_bugs;
+	unsigned long	 nr_events;
+	unsigned long	 nr_lost_chunks;
+	unsigned long	 nr_lost_events;
+	u64		 run_measurement_overhead;
+	u64		 sleep_measurement_overhead;
+	u64		 start_time;
+	u64		 cpu_usage;
+	u64		 runavg_cpu_usage;
+	u64		 parent_cpu_usage;
+	u64		 runavg_parent_cpu_usage;
+	u64		 sum_runtime;
+	u64		 sum_fluct;
+	u64		 run_avg;
+	u64		 all_runtime;
+	u64		 all_count;
+	u64		 cpu_last_switched[MAX_CPUS];
+	struct rb_root	 atom_root, sorted_atom_root;
+	struct list_head sort_list, cmp_pid;
+	bool force;
+};
+
+static u64 get_nsecs(void)
+{
+	struct timespec ts;
+
+	clock_gettime(CLOCK_MONOTONIC, &ts);
+
+	return ts.tv_sec * 1000000000ULL + ts.tv_nsec;
+}
+
+static void burn_nsecs(struct perf_sched *sched, u64 nsecs)
+{
+	u64 T0 = get_nsecs(), T1;
+
+	do {
+		T1 = get_nsecs();
+	} while (T1 + sched->run_measurement_overhead < T0 + nsecs);
+}
+
+static void sleep_nsecs(u64 nsecs)
+{
+	struct timespec ts;
+
+	ts.tv_nsec = nsecs % 999999999;
+	ts.tv_sec = nsecs / 999999999;
+
+	nanosleep(&ts, NULL);
+}
+
+static void calibrate_run_measurement_overhead(struct perf_sched *sched)
+{
+	u64 T0, T1, delta, min_delta = 1000000000ULL;
+	int i;
+
+	for (i = 0; i < 10; i++) {
+		T0 = get_nsecs();
+		burn_nsecs(sched, 0);
+		T1 = get_nsecs();
+		delta = T1-T0;
+		min_delta = min(min_delta, delta);
+	}
+	sched->run_measurement_overhead = min_delta;
+
+	printf("run measurement overhead: %" PRIu64 " nsecs\n", min_delta);
+}
+
+static void calibrate_sleep_measurement_overhead(struct perf_sched *sched)
+{
+	u64 T0, T1, delta, min_delta = 1000000000ULL;
+	int i;
+
+	for (i = 0; i < 10; i++) {
+		T0 = get_nsecs();
+		sleep_nsecs(10000);
+		T1 = get_nsecs();
+		delta = T1-T0;
+		min_delta = min(min_delta, delta);
+	}
+	min_delta -= 10000;
+	sched->sleep_measurement_overhead = min_delta;
+
+	printf("sleep measurement overhead: %" PRIu64 " nsecs\n", min_delta);
+}
+
+static struct sched_atom *
+get_new_event(struct task_desc *task, u64 timestamp)
+{
+	struct sched_atom *event = zalloc(sizeof(*event));
+	unsigned long idx = task->nr_events;
+	size_t size;
+
+	event->timestamp = timestamp;
+	event->nr = idx;
+
+	task->nr_events++;
+	size = sizeof(struct sched_atom *) * task->nr_events;
+	task->atoms = realloc(task->atoms, size);
+	BUG_ON(!task->atoms);
+
+	task->atoms[idx] = event;
+
+	return event;
+}
+
+static struct sched_atom *last_event(struct task_desc *task)
+{
+	if (!task->nr_events)
+		return NULL;
+
+	return task->atoms[task->nr_events - 1];
+}
+
+static void add_sched_event_run(struct perf_sched *sched, struct task_desc *task,
+				u64 timestamp, u64 duration)
+{
+	struct sched_atom *event, *curr_event = last_event(task);
+
+	/*
+	 * optimize an existing RUN event by merging this one
+	 * to it:
+	 */
+	if (curr_event && curr_event->type == SCHED_EVENT_RUN) {
+		sched->nr_run_events_optimized++;
+		curr_event->duration += duration;
+		return;
+	}
+
+	event = get_new_event(task, timestamp);
+
+	event->type = SCHED_EVENT_RUN;
+	event->duration = duration;
+
+	sched->nr_run_events++;
+}
+
+static void add_sched_event_wakeup(struct perf_sched *sched, struct task_desc *task,
+				   u64 timestamp, struct task_desc *wakee)
+{
+	struct sched_atom *event, *wakee_event;
+
+	event = get_new_event(task, timestamp);
+	event->type = SCHED_EVENT_WAKEUP;
+	event->wakee = wakee;
+
+	wakee_event = last_event(wakee);
+	if (!wakee_event || wakee_event->type != SCHED_EVENT_SLEEP) {
+		sched->targetless_wakeups++;
+		return;
+	}
+	if (wakee_event->wait_sem) {
+		sched->multitarget_wakeups++;
+		return;
+	}
+
+	wakee_event->wait_sem = zalloc(sizeof(*wakee_event->wait_sem));
+	sem_init(wakee_event->wait_sem, 0, 0);
+	wakee_event->specific_wait = 1;
+	event->wait_sem = wakee_event->wait_sem;
+
+	sched->nr_wakeup_events++;
+}
+
+static void add_sched_event_sleep(struct perf_sched *sched, struct task_desc *task,
+				  u64 timestamp, u64 task_state __maybe_unused)
+{
+	struct sched_atom *event = get_new_event(task, timestamp);
+
+	event->type = SCHED_EVENT_SLEEP;
+
+	sched->nr_sleep_events++;
+}
+
+static struct task_desc *register_pid(struct perf_sched *sched,
+				      unsigned long pid, const char *comm)
+{
+	struct task_desc *task;
+	static int pid_max;
+
+	if (sched->pid_to_task == NULL) {
+		if (sysctl__read_int("kernel/pid_max", &pid_max) < 0)
+			pid_max = MAX_PID;
+		BUG_ON((sched->pid_to_task = calloc(pid_max, sizeof(struct task_desc *))) == NULL);
+	}
+	if (pid >= (unsigned long)pid_max) {
+		BUG_ON((sched->pid_to_task = realloc(sched->pid_to_task, (pid + 1) *
+			sizeof(struct task_desc *))) == NULL);
+		while (pid >= (unsigned long)pid_max)
+			sched->pid_to_task[pid_max++] = NULL;
+	}
+
+	task = sched->pid_to_task[pid];
+
+	if (task)
+		return task;
+
+	task = zalloc(sizeof(*task));
+	task->pid = pid;
+	task->nr = sched->nr_tasks;
+	strcpy(task->comm, comm);
+	/*
+	 * every task starts in sleeping state - this gets ignored
+	 * if there's no wakeup pointing to this sleep state:
+	 */
+	add_sched_event_sleep(sched, task, 0, 0);
+
+	sched->pid_to_task[pid] = task;
+	sched->nr_tasks++;
+	sched->tasks = realloc(sched->tasks, sched->nr_tasks * sizeof(struct task_desc *));
+	BUG_ON(!sched->tasks);
+	sched->tasks[task->nr] = task;
+
+	if (verbose)
+		printf("registered task #%ld, PID %ld (%s)\n", sched->nr_tasks, pid, comm);
+
+	return task;
+}
+
+
+static void print_task_traces(struct perf_sched *sched)
+{
+	struct task_desc *task;
+	unsigned long i;
+
+	for (i = 0; i < sched->nr_tasks; i++) {
+		task = sched->tasks[i];
+		printf("task %6ld (%20s:%10ld), nr_events: %ld\n",
+			task->nr, task->comm, task->pid, task->nr_events);
+	}
+}
+
+static void add_cross_task_wakeups(struct perf_sched *sched)
+{
+	struct task_desc *task1, *task2;
+	unsigned long i, j;
+
+	for (i = 0; i < sched->nr_tasks; i++) {
+		task1 = sched->tasks[i];
+		j = i + 1;
+		if (j == sched->nr_tasks)
+			j = 0;
+		task2 = sched->tasks[j];
+		add_sched_event_wakeup(sched, task1, 0, task2);
+	}
+}
+
+static void perf_sched__process_event(struct perf_sched *sched,
+				      struct sched_atom *atom)
+{
+	int ret = 0;
+
+	switch (atom->type) {
+		case SCHED_EVENT_RUN:
+			burn_nsecs(sched, atom->duration);
+			break;
+		case SCHED_EVENT_SLEEP:
+			if (atom->wait_sem)
+				ret = sem_wait(atom->wait_sem);
+			BUG_ON(ret);
+			break;
+		case SCHED_EVENT_WAKEUP:
+			if (atom->wait_sem)
+				ret = sem_post(atom->wait_sem);
+			BUG_ON(ret);
+			break;
+		case SCHED_EVENT_MIGRATION:
+			break;
+		default:
+			BUG_ON(1);
+	}
+}
+
+static u64 get_cpu_usage_nsec_parent(void)
+{
+	struct rusage ru;
+	u64 sum;
+	int err;
+
+	err = getrusage(RUSAGE_SELF, &ru);
+	BUG_ON(err);
+
+	sum =  ru.ru_utime.tv_sec*1e9 + ru.ru_utime.tv_usec*1e3;
+	sum += ru.ru_stime.tv_sec*1e9 + ru.ru_stime.tv_usec*1e3;
+
+	return sum;
+}
+
+static int self_open_counters(struct perf_sched *sched, unsigned long cur_task)
+{
+	struct perf_event_attr attr;
+	char sbuf[STRERR_BUFSIZE], info[STRERR_BUFSIZE];
+	int fd;
+	struct rlimit limit;
+	bool need_privilege = false;
+
+	memset(&attr, 0, sizeof(attr));
+
+	attr.type = PERF_TYPE_SOFTWARE;
+	attr.config = PERF_COUNT_SW_TASK_CLOCK;
+
+force_again:
+	fd = sys_perf_event_open(&attr, 0, -1, -1,
+				 perf_event_open_cloexec_flag());
+
+	if (fd < 0) {
+		if (errno == EMFILE) {
+			if (sched->force) {
+				BUG_ON(getrlimit(RLIMIT_NOFILE, &limit) == -1);
+				limit.rlim_cur += sched->nr_tasks - cur_task;
+				if (limit.rlim_cur > limit.rlim_max) {
+					limit.rlim_max = limit.rlim_cur;
+					need_privilege = true;
+				}
+				if (setrlimit(RLIMIT_NOFILE, &limit) == -1) {
+					if (need_privilege && errno == EPERM)
+						strcpy(info, "Need privilege\n");
+				} else
+					goto force_again;
+			} else
+				strcpy(info, "Have a try with -f option\n");
+		}
+		pr_err("Error: sys_perf_event_open() syscall returned "
+		       "with %d (%s)\n%s", fd,
+		       strerror_r(errno, sbuf, sizeof(sbuf)), info);
+		exit(EXIT_FAILURE);
+	}
+	return fd;
+}
+
+static u64 get_cpu_usage_nsec_self(int fd)
+{
+	u64 runtime;
+	int ret;
+
+	ret = read(fd, &runtime, sizeof(runtime));
+	BUG_ON(ret != sizeof(runtime));
+
+	return runtime;
+}
+
+struct sched_thread_parms {
+	struct task_desc  *task;
+	struct perf_sched *sched;
+	int fd;
+};
+
+static void *thread_func(void *ctx)
+{
+	struct sched_thread_parms *parms = ctx;
+	struct task_desc *this_task = parms->task;
+	struct perf_sched *sched = parms->sched;
+	u64 cpu_usage_0, cpu_usage_1;
+	unsigned long i, ret;
+	char comm2[22];
+	int fd = parms->fd;
+
+	zfree(&parms);
+
+	sprintf(comm2, ":%s", this_task->comm);
+	prctl(PR_SET_NAME, comm2);
+	if (fd < 0)
+		return NULL;
+again:
+	ret = sem_post(&this_task->ready_for_work);
+	BUG_ON(ret);
+	ret = pthread_mutex_lock(&sched->start_work_mutex);
+	BUG_ON(ret);
+	ret = pthread_mutex_unlock(&sched->start_work_mutex);
+	BUG_ON(ret);
+
+	cpu_usage_0 = get_cpu_usage_nsec_self(fd);
+
+	for (i = 0; i < this_task->nr_events; i++) {
+		this_task->curr_event = i;
+		perf_sched__process_event(sched, this_task->atoms[i]);
+	}
+
+	cpu_usage_1 = get_cpu_usage_nsec_self(fd);
+	this_task->cpu_usage = cpu_usage_1 - cpu_usage_0;
+	ret = sem_post(&this_task->work_done_sem);
+	BUG_ON(ret);
+
+	ret = pthread_mutex_lock(&sched->work_done_wait_mutex);
+	BUG_ON(ret);
+	ret = pthread_mutex_unlock(&sched->work_done_wait_mutex);
+	BUG_ON(ret);
+
+	goto again;
+}
+
+static void create_tasks(struct perf_sched *sched)
+{
+	struct task_desc *task;
+	pthread_attr_t attr;
+	unsigned long i;
+	int err;
+
+	err = pthread_attr_init(&attr);
+	BUG_ON(err);
+	err = pthread_attr_setstacksize(&attr,
+			(size_t) max(16 * 1024, PTHREAD_STACK_MIN));
+	BUG_ON(err);
+	err = pthread_mutex_lock(&sched->start_work_mutex);
+	BUG_ON(err);
+	err = pthread_mutex_lock(&sched->work_done_wait_mutex);
+	BUG_ON(err);
+	for (i = 0; i < sched->nr_tasks; i++) {
+		struct sched_thread_parms *parms = malloc(sizeof(*parms));
+		BUG_ON(parms == NULL);
+		parms->task = task = sched->tasks[i];
+		parms->sched = sched;
+		parms->fd = self_open_counters(sched, i);
+		sem_init(&task->sleep_sem, 0, 0);
+		sem_init(&task->ready_for_work, 0, 0);
+		sem_init(&task->work_done_sem, 0, 0);
+		task->curr_event = 0;
+		err = pthread_create(&task->thread, &attr, thread_func, parms);
+		BUG_ON(err);
+	}
+}
+
+static void wait_for_tasks(struct perf_sched *sched)
+{
+	u64 cpu_usage_0, cpu_usage_1;
+	struct task_desc *task;
+	unsigned long i, ret;
+
+	sched->start_time = get_nsecs();
+	sched->cpu_usage = 0;
+	pthread_mutex_unlock(&sched->work_done_wait_mutex);
+
+	for (i = 0; i < sched->nr_tasks; i++) {
+		task = sched->tasks[i];
+		ret = sem_wait(&task->ready_for_work);
+		BUG_ON(ret);
+		sem_init(&task->ready_for_work, 0, 0);
+	}
+	ret = pthread_mutex_lock(&sched->work_done_wait_mutex);
+	BUG_ON(ret);
+
+	cpu_usage_0 = get_cpu_usage_nsec_parent();
+
+	pthread_mutex_unlock(&sched->start_work_mutex);
+
+	for (i = 0; i < sched->nr_tasks; i++) {
+		task = sched->tasks[i];
+		ret = sem_wait(&task->work_done_sem);
+		BUG_ON(ret);
+		sem_init(&task->work_done_sem, 0, 0);
+		sched->cpu_usage += task->cpu_usage;
+		task->cpu_usage = 0;
+	}
+
+	cpu_usage_1 = get_cpu_usage_nsec_parent();
+	if (!sched->runavg_cpu_usage)
+		sched->runavg_cpu_usage = sched->cpu_usage;
+	sched->runavg_cpu_usage = (sched->runavg_cpu_usage * (sched->replay_repeat - 1) + sched->cpu_usage) / sched->replay_repeat;
+
+	sched->parent_cpu_usage = cpu_usage_1 - cpu_usage_0;
+	if (!sched->runavg_parent_cpu_usage)
+		sched->runavg_parent_cpu_usage = sched->parent_cpu_usage;
+	sched->runavg_parent_cpu_usage = (sched->runavg_parent_cpu_usage * (sched->replay_repeat - 1) +
+					 sched->parent_cpu_usage)/sched->replay_repeat;
+
+	ret = pthread_mutex_lock(&sched->start_work_mutex);
+	BUG_ON(ret);
+
+	for (i = 0; i < sched->nr_tasks; i++) {
+		task = sched->tasks[i];
+		sem_init(&task->sleep_sem, 0, 0);
+		task->curr_event = 0;
+	}
+}
+
+static void run_one_test(struct perf_sched *sched)
+{
+	u64 T0, T1, delta, avg_delta, fluct;
+
+	T0 = get_nsecs();
+	wait_for_tasks(sched);
+	T1 = get_nsecs();
+
+	delta = T1 - T0;
+	sched->sum_runtime += delta;
+	sched->nr_runs++;
+
+	avg_delta = sched->sum_runtime / sched->nr_runs;
+	if (delta < avg_delta)
+		fluct = avg_delta - delta;
+	else
+		fluct = delta - avg_delta;
+	sched->sum_fluct += fluct;
+	if (!sched->run_avg)
+		sched->run_avg = delta;
+	sched->run_avg = (sched->run_avg * (sched->replay_repeat - 1) + delta) / sched->replay_repeat;
+
+	printf("#%-3ld: %0.3f, ", sched->nr_runs, (double)delta / 1000000.0);
+
+	printf("ravg: %0.2f, ", (double)sched->run_avg / 1e6);
+
+	printf("cpu: %0.2f / %0.2f",
+		(double)sched->cpu_usage / 1e6, (double)sched->runavg_cpu_usage / 1e6);
+
+#if 0
+	/*
+	 * rusage statistics done by the parent, these are less
+	 * accurate than the sched->sum_exec_runtime based statistics:
+	 */
+	printf(" [%0.2f / %0.2f]",
+		(double)sched->parent_cpu_usage/1e6,
+		(double)sched->runavg_parent_cpu_usage/1e6);
+#endif
+
+	printf("\n");
+
+	if (sched->nr_sleep_corrections)
+		printf(" (%ld sleep corrections)\n", sched->nr_sleep_corrections);
+	sched->nr_sleep_corrections = 0;
+}
+
+static void test_calibrations(struct perf_sched *sched)
+{
+	u64 T0, T1;
+
+	T0 = get_nsecs();
+	burn_nsecs(sched, 1e6);
+	T1 = get_nsecs();
+
+	printf("the run test took %" PRIu64 " nsecs\n", T1 - T0);
+
+	T0 = get_nsecs();
+	sleep_nsecs(1e6);
+	T1 = get_nsecs();
+
+	printf("the sleep test took %" PRIu64 " nsecs\n", T1 - T0);
+}
+
+static int
+replay_wakeup_event(struct perf_sched *sched,
+		    struct perf_evsel *evsel, struct perf_sample *sample,
+		    struct machine *machine __maybe_unused)
+{
+	const char *comm = perf_evsel__strval(evsel, sample, "comm");
+	const u32 pid	 = perf_evsel__intval(evsel, sample, "pid");
+	struct task_desc *waker, *wakee;
+
+	if (verbose) {
+		printf("sched_wakeup event %p\n", evsel);
+
+		printf(" ... pid %d woke up %s/%d\n", sample->tid, comm, pid);
+	}
+
+	waker = register_pid(sched, sample->tid, "<unknown>");
+	wakee = register_pid(sched, pid, comm);
+
+	add_sched_event_wakeup(sched, waker, sample->time, wakee);
+	return 0;
+}
+
+static int replay_switch_event(struct perf_sched *sched,
+			       struct perf_evsel *evsel,
+			       struct perf_sample *sample,
+			       struct machine *machine __maybe_unused)
+{
+	const char *prev_comm  = perf_evsel__strval(evsel, sample, "prev_comm"),
+		   *next_comm  = perf_evsel__strval(evsel, sample, "next_comm");
+	const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
+		  next_pid = perf_evsel__intval(evsel, sample, "next_pid");
+	const u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state");
+	struct task_desc *prev, __maybe_unused *next;
+	u64 timestamp0, timestamp = sample->time;
+	int cpu = sample->cpu;
+	s64 delta;
+
+	if (verbose)
+		printf("sched_switch event %p\n", evsel);
+
+	if (cpu >= MAX_CPUS || cpu < 0)
+		return 0;
+
+	timestamp0 = sched->cpu_last_switched[cpu];
+	if (timestamp0)
+		delta = timestamp - timestamp0;
+	else
+		delta = 0;
+
+	if (delta < 0) {
+		pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta);
+		return -1;
+	}
+
+	pr_debug(" ... switch from %s/%d to %s/%d [ran %" PRIu64 " nsecs]\n",
+		 prev_comm, prev_pid, next_comm, next_pid, delta);
+
+	prev = register_pid(sched, prev_pid, prev_comm);
+	next = register_pid(sched, next_pid, next_comm);
+
+	sched->cpu_last_switched[cpu] = timestamp;
+
+	add_sched_event_run(sched, prev, timestamp, delta);
+	add_sched_event_sleep(sched, prev, timestamp, prev_state);
+
+	return 0;
+}
+
+static int replay_fork_event(struct perf_sched *sched,
+			     union perf_event *event,
+			     struct machine *machine)
+{
+	struct thread *child, *parent;
+
+	child = machine__findnew_thread(machine, event->fork.pid,
+					event->fork.tid);
+	parent = machine__findnew_thread(machine, event->fork.ppid,
+					 event->fork.ptid);
+
+	if (child == NULL || parent == NULL) {
+		pr_debug("thread does not exist on fork event: child %p, parent %p\n",
+				 child, parent);
+		return 0;
+	}
+
+	if (verbose) {
+		printf("fork event\n");
+		printf("... parent: %s/%d\n", thread__comm_str(parent), parent->tid);
+		printf("...  child: %s/%d\n", thread__comm_str(child), child->tid);
+	}
+
+	register_pid(sched, parent->tid, thread__comm_str(parent));
+	register_pid(sched, child->tid, thread__comm_str(child));
+	return 0;
+}
+
+struct sort_dimension {
+	const char		*name;
+	sort_fn_t		cmp;
+	struct list_head	list;
+};
+
+static int
+thread_lat_cmp(struct list_head *list, struct work_atoms *l, struct work_atoms *r)
+{
+	struct sort_dimension *sort;
+	int ret = 0;
+
+	BUG_ON(list_empty(list));
+
+	list_for_each_entry(sort, list, list) {
+		ret = sort->cmp(l, r);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
+static struct work_atoms *
+thread_atoms_search(struct rb_root *root, struct thread *thread,
+			 struct list_head *sort_list)
+{
+	struct rb_node *node = root->rb_node;
+	struct work_atoms key = { .thread = thread };
+
+	while (node) {
+		struct work_atoms *atoms;
+		int cmp;
+
+		atoms = container_of(node, struct work_atoms, node);
+
+		cmp = thread_lat_cmp(sort_list, &key, atoms);
+		if (cmp > 0)
+			node = node->rb_left;
+		else if (cmp < 0)
+			node = node->rb_right;
+		else {
+			BUG_ON(thread != atoms->thread);
+			return atoms;
+		}
+	}
+	return NULL;
+}
+
+static void
+__thread_latency_insert(struct rb_root *root, struct work_atoms *data,
+			 struct list_head *sort_list)
+{
+	struct rb_node **new = &(root->rb_node), *parent = NULL;
+
+	while (*new) {
+		struct work_atoms *this;
+		int cmp;
+
+		this = container_of(*new, struct work_atoms, node);
+		parent = *new;
+
+		cmp = thread_lat_cmp(sort_list, data, this);
+
+		if (cmp > 0)
+			new = &((*new)->rb_left);
+		else
+			new = &((*new)->rb_right);
+	}
+
+	rb_link_node(&data->node, parent, new);
+	rb_insert_color(&data->node, root);
+}
+
+static int thread_atoms_insert(struct perf_sched *sched, struct thread *thread)
+{
+	struct work_atoms *atoms = zalloc(sizeof(*atoms));
+	if (!atoms) {
+		pr_err("No memory at %s\n", __func__);
+		return -1;
+	}
+
+	atoms->thread = thread__get(thread);
+	INIT_LIST_HEAD(&atoms->work_list);
+	__thread_latency_insert(&sched->atom_root, atoms, &sched->cmp_pid);
+	return 0;
+}
+
+static char sched_out_state(u64 prev_state)
+{
+	const char *str = TASK_STATE_TO_CHAR_STR;
+
+	return str[prev_state];
+}
+
+static int
+add_sched_out_event(struct work_atoms *atoms,
+		    char run_state,
+		    u64 timestamp)
+{
+	struct work_atom *atom = zalloc(sizeof(*atom));
+	if (!atom) {
+		pr_err("Non memory at %s", __func__);
+		return -1;
+	}
+
+	atom->sched_out_time = timestamp;
+
+	if (run_state == 'R') {
+		atom->state = THREAD_WAIT_CPU;
+		atom->wake_up_time = atom->sched_out_time;
+	}
+
+	list_add_tail(&atom->list, &atoms->work_list);
+	return 0;
+}
+
+static void
+add_runtime_event(struct work_atoms *atoms, u64 delta,
+		  u64 timestamp __maybe_unused)
+{
+	struct work_atom *atom;
+
+	BUG_ON(list_empty(&atoms->work_list));
+
+	atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+
+	atom->runtime += delta;
+	atoms->total_runtime += delta;
+}
+
+static void
+add_sched_in_event(struct work_atoms *atoms, u64 timestamp)
+{
+	struct work_atom *atom;
+	u64 delta;
+
+	if (list_empty(&atoms->work_list))
+		return;
+
+	atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+
+	if (atom->state != THREAD_WAIT_CPU)
+		return;
+
+	if (timestamp < atom->wake_up_time) {
+		atom->state = THREAD_IGNORE;
+		return;
+	}
+
+	atom->state = THREAD_SCHED_IN;
+	atom->sched_in_time = timestamp;
+
+	delta = atom->sched_in_time - atom->wake_up_time;
+	atoms->total_lat += delta;
+	if (delta > atoms->max_lat) {
+		atoms->max_lat = delta;
+		atoms->max_lat_at = timestamp;
+	}
+	atoms->nb_atoms++;
+}
+
+static int latency_switch_event(struct perf_sched *sched,
+				struct perf_evsel *evsel,
+				struct perf_sample *sample,
+				struct machine *machine)
+{
+	const u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
+		  next_pid = perf_evsel__intval(evsel, sample, "next_pid");
+	const u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state");
+	struct work_atoms *out_events, *in_events;
+	struct thread *sched_out, *sched_in;
+	u64 timestamp0, timestamp = sample->time;
+	int cpu = sample->cpu;
+	s64 delta;
+
+	BUG_ON(cpu >= MAX_CPUS || cpu < 0);
+
+	timestamp0 = sched->cpu_last_switched[cpu];
+	sched->cpu_last_switched[cpu] = timestamp;
+	if (timestamp0)
+		delta = timestamp - timestamp0;
+	else
+		delta = 0;
+
+	if (delta < 0) {
+		pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta);
+		return -1;
+	}
+
+	sched_out = machine__findnew_thread(machine, -1, prev_pid);
+	sched_in = machine__findnew_thread(machine, -1, next_pid);
+
+	out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid);
+	if (!out_events) {
+		if (thread_atoms_insert(sched, sched_out))
+			return -1;
+		out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid);
+		if (!out_events) {
+			pr_err("out-event: Internal tree error");
+			return -1;
+		}
+	}
+	if (add_sched_out_event(out_events, sched_out_state(prev_state), timestamp))
+		return -1;
+
+	in_events = thread_atoms_search(&sched->atom_root, sched_in, &sched->cmp_pid);
+	if (!in_events) {
+		if (thread_atoms_insert(sched, sched_in))
+			return -1;
+		in_events = thread_atoms_search(&sched->atom_root, sched_in, &sched->cmp_pid);
+		if (!in_events) {
+			pr_err("in-event: Internal tree error");
+			return -1;
+		}
+		/*
+		 * Take came in we have not heard about yet,
+		 * add in an initial atom in runnable state:
+		 */
+		if (add_sched_out_event(in_events, 'R', timestamp))
+			return -1;
+	}
+	add_sched_in_event(in_events, timestamp);
+
+	return 0;
+}
+
+static int latency_runtime_event(struct perf_sched *sched,
+				 struct perf_evsel *evsel,
+				 struct perf_sample *sample,
+				 struct machine *machine)
+{
+	const u32 pid	   = perf_evsel__intval(evsel, sample, "pid");
+	const u64 runtime  = perf_evsel__intval(evsel, sample, "runtime");
+	struct thread *thread = machine__findnew_thread(machine, -1, pid);
+	struct work_atoms *atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid);
+	u64 timestamp = sample->time;
+	int cpu = sample->cpu;
+
+	BUG_ON(cpu >= MAX_CPUS || cpu < 0);
+	if (!atoms) {
+		if (thread_atoms_insert(sched, thread))
+			return -1;
+		atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid);
+		if (!atoms) {
+			pr_err("in-event: Internal tree error");
+			return -1;
+		}
+		if (add_sched_out_event(atoms, 'R', timestamp))
+			return -1;
+	}
+
+	add_runtime_event(atoms, runtime, timestamp);
+	return 0;
+}
+
+static int latency_wakeup_event(struct perf_sched *sched,
+				struct perf_evsel *evsel,
+				struct perf_sample *sample,
+				struct machine *machine)
+{
+	const u32 pid	  = perf_evsel__intval(evsel, sample, "pid");
+	struct work_atoms *atoms;
+	struct work_atom *atom;
+	struct thread *wakee;
+	u64 timestamp = sample->time;
+
+	wakee = machine__findnew_thread(machine, -1, pid);
+	atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid);
+	if (!atoms) {
+		if (thread_atoms_insert(sched, wakee))
+			return -1;
+		atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid);
+		if (!atoms) {
+			pr_err("wakeup-event: Internal tree error");
+			return -1;
+		}
+		if (add_sched_out_event(atoms, 'S', timestamp))
+			return -1;
+	}
+
+	BUG_ON(list_empty(&atoms->work_list));
+
+	atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+
+	/*
+	 * As we do not guarantee the wakeup event happens when
+	 * task is out of run queue, also may happen when task is
+	 * on run queue and wakeup only change ->state to TASK_RUNNING,
+	 * then we should not set the ->wake_up_time when wake up a
+	 * task which is on run queue.
+	 *
+	 * You WILL be missing events if you've recorded only
+	 * one CPU, or are only looking at only one, so don't
+	 * skip in this case.
+	 */
+	if (sched->profile_cpu == -1 && atom->state != THREAD_SLEEPING)
+		return 0;
+
+	sched->nr_timestamps++;
+	if (atom->sched_out_time > timestamp) {
+		sched->nr_unordered_timestamps++;
+		return 0;
+	}
+
+	atom->state = THREAD_WAIT_CPU;
+	atom->wake_up_time = timestamp;
+	return 0;
+}
+
+static int latency_migrate_task_event(struct perf_sched *sched,
+				      struct perf_evsel *evsel,
+				      struct perf_sample *sample,
+				      struct machine *machine)
+{
+	const u32 pid = perf_evsel__intval(evsel, sample, "pid");
+	u64 timestamp = sample->time;
+	struct work_atoms *atoms;
+	struct work_atom *atom;
+	struct thread *migrant;
+
+	/*
+	 * Only need to worry about migration when profiling one CPU.
+	 */
+	if (sched->profile_cpu == -1)
+		return 0;
+
+	migrant = machine__findnew_thread(machine, -1, pid);
+	atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid);
+	if (!atoms) {
+		if (thread_atoms_insert(sched, migrant))
+			return -1;
+		register_pid(sched, migrant->tid, thread__comm_str(migrant));
+		atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid);
+		if (!atoms) {
+			pr_err("migration-event: Internal tree error");
+			return -1;
+		}
+		if (add_sched_out_event(atoms, 'R', timestamp))
+			return -1;
+	}
+
+	BUG_ON(list_empty(&atoms->work_list));
+
+	atom = list_entry(atoms->work_list.prev, struct work_atom, list);
+	atom->sched_in_time = atom->sched_out_time = atom->wake_up_time = timestamp;
+
+	sched->nr_timestamps++;
+
+	if (atom->sched_out_time > timestamp)
+		sched->nr_unordered_timestamps++;
+
+	return 0;
+}
+
+static void output_lat_thread(struct perf_sched *sched, struct work_atoms *work_list)
+{
+	int i;
+	int ret;
+	u64 avg;
+
+	if (!work_list->nb_atoms)
+		return;
+	/*
+	 * Ignore idle threads:
+	 */
+	if (!strcmp(thread__comm_str(work_list->thread), "swapper"))
+		return;
+
+	sched->all_runtime += work_list->total_runtime;
+	sched->all_count   += work_list->nb_atoms;
+
+	ret = printf("  %s:%d ", thread__comm_str(work_list->thread), work_list->thread->tid);
+
+	for (i = 0; i < 24 - ret; i++)
+		printf(" ");
+
+	avg = work_list->total_lat / work_list->nb_atoms;
+
+	printf("|%11.3f ms |%9" PRIu64 " | avg:%9.3f ms | max:%9.3f ms | max at: %13.6f s\n",
+	      (double)work_list->total_runtime / 1e6,
+		 work_list->nb_atoms, (double)avg / 1e6,
+		 (double)work_list->max_lat / 1e6,
+		 (double)work_list->max_lat_at / 1e9);
+}
+
+static int pid_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	if (l->thread->tid < r->thread->tid)
+		return -1;
+	if (l->thread->tid > r->thread->tid)
+		return 1;
+
+	return 0;
+}
+
+static int avg_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	u64 avgl, avgr;
+
+	if (!l->nb_atoms)
+		return -1;
+
+	if (!r->nb_atoms)
+		return 1;
+
+	avgl = l->total_lat / l->nb_atoms;
+	avgr = r->total_lat / r->nb_atoms;
+
+	if (avgl < avgr)
+		return -1;
+	if (avgl > avgr)
+		return 1;
+
+	return 0;
+}
+
+static int max_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	if (l->max_lat < r->max_lat)
+		return -1;
+	if (l->max_lat > r->max_lat)
+		return 1;
+
+	return 0;
+}
+
+static int switch_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	if (l->nb_atoms < r->nb_atoms)
+		return -1;
+	if (l->nb_atoms > r->nb_atoms)
+		return 1;
+
+	return 0;
+}
+
+static int runtime_cmp(struct work_atoms *l, struct work_atoms *r)
+{
+	if (l->total_runtime < r->total_runtime)
+		return -1;
+	if (l->total_runtime > r->total_runtime)
+		return 1;
+
+	return 0;
+}
+
+static int sort_dimension__add(const char *tok, struct list_head *list)
+{
+	size_t i;
+	static struct sort_dimension avg_sort_dimension = {
+		.name = "avg",
+		.cmp  = avg_cmp,
+	};
+	static struct sort_dimension max_sort_dimension = {
+		.name = "max",
+		.cmp  = max_cmp,
+	};
+	static struct sort_dimension pid_sort_dimension = {
+		.name = "pid",
+		.cmp  = pid_cmp,
+	};
+	static struct sort_dimension runtime_sort_dimension = {
+		.name = "runtime",
+		.cmp  = runtime_cmp,
+	};
+	static struct sort_dimension switch_sort_dimension = {
+		.name = "switch",
+		.cmp  = switch_cmp,
+	};
+	struct sort_dimension *available_sorts[] = {
+		&pid_sort_dimension,
+		&avg_sort_dimension,
+		&max_sort_dimension,
+		&switch_sort_dimension,
+		&runtime_sort_dimension,
+	};
+
+	for (i = 0; i < ARRAY_SIZE(available_sorts); i++) {
+		if (!strcmp(available_sorts[i]->name, tok)) {
+			list_add_tail(&available_sorts[i]->list, list);
+
+			return 0;
+		}
+	}
+
+	return -1;
+}
+
+static void perf_sched__sort_lat(struct perf_sched *sched)
+{
+	struct rb_node *node;
+
+	for (;;) {
+		struct work_atoms *data;
+		node = rb_first(&sched->atom_root);
+		if (!node)
+			break;
+
+		rb_erase(node, &sched->atom_root);
+		data = rb_entry(node, struct work_atoms, node);
+		__thread_latency_insert(&sched->sorted_atom_root, data, &sched->sort_list);
+	}
+}
+
+static int process_sched_wakeup_event(struct perf_tool *tool,
+				      struct perf_evsel *evsel,
+				      struct perf_sample *sample,
+				      struct machine *machine)
+{
+	struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
+
+	if (sched->tp_handler->wakeup_event)
+		return sched->tp_handler->wakeup_event(sched, evsel, sample, machine);
+
+	return 0;
+}
+
+static int map_switch_event(struct perf_sched *sched, struct perf_evsel *evsel,
+			    struct perf_sample *sample, struct machine *machine)
+{
+	const u32 next_pid = perf_evsel__intval(evsel, sample, "next_pid");
+	struct thread *sched_in;
+	int new_shortname;
+	u64 timestamp0, timestamp = sample->time;
+	s64 delta;
+	int cpu, this_cpu = sample->cpu;
+
+	BUG_ON(this_cpu >= MAX_CPUS || this_cpu < 0);
+
+	if (this_cpu > sched->max_cpu)
+		sched->max_cpu = this_cpu;
+
+	timestamp0 = sched->cpu_last_switched[this_cpu];
+	sched->cpu_last_switched[this_cpu] = timestamp;
+	if (timestamp0)
+		delta = timestamp - timestamp0;
+	else
+		delta = 0;
+
+	if (delta < 0) {
+		pr_err("hm, delta: %" PRIu64 " < 0 ?\n", delta);
+		return -1;
+	}
+
+	sched_in = machine__findnew_thread(machine, -1, next_pid);
+
+	sched->curr_thread[this_cpu] = sched_in;
+
+	printf("  ");
+
+	new_shortname = 0;
+	if (!sched_in->shortname[0]) {
+		if (!strcmp(thread__comm_str(sched_in), "swapper")) {
+			/*
+			 * Don't allocate a letter-number for swapper:0
+			 * as a shortname. Instead, we use '.' for it.
+			 */
+			sched_in->shortname[0] = '.';
+			sched_in->shortname[1] = ' ';
+		} else {
+			sched_in->shortname[0] = sched->next_shortname1;
+			sched_in->shortname[1] = sched->next_shortname2;
+
+			if (sched->next_shortname1 < 'Z') {
+				sched->next_shortname1++;
+			} else {
+				sched->next_shortname1 = 'A';
+				if (sched->next_shortname2 < '9')
+					sched->next_shortname2++;
+				else
+					sched->next_shortname2 = '0';
+			}
+		}
+		new_shortname = 1;
+	}
+
+	for (cpu = 0; cpu <= sched->max_cpu; cpu++) {
+		if (cpu != this_cpu)
+			printf(" ");
+		else
+			printf("*");
+
+		if (sched->curr_thread[cpu])
+			printf("%2s ", sched->curr_thread[cpu]->shortname);
+		else
+			printf("   ");
+	}
+
+	printf("  %12.6f secs ", (double)timestamp/1e9);
+	if (new_shortname) {
+		printf("%s => %s:%d\n",
+		       sched_in->shortname, thread__comm_str(sched_in), sched_in->tid);
+	} else {
+		printf("\n");
+	}
+
+	return 0;
+}
+
+static int process_sched_switch_event(struct perf_tool *tool,
+				      struct perf_evsel *evsel,
+				      struct perf_sample *sample,
+				      struct machine *machine)
+{
+	struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
+	int this_cpu = sample->cpu, err = 0;
+	u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"),
+	    next_pid = perf_evsel__intval(evsel, sample, "next_pid");
+
+	if (sched->curr_pid[this_cpu] != (u32)-1) {
+		/*
+		 * Are we trying to switch away a PID that is
+		 * not current?
+		 */
+		if (sched->curr_pid[this_cpu] != prev_pid)
+			sched->nr_context_switch_bugs++;
+	}
+
+	if (sched->tp_handler->switch_event)
+		err = sched->tp_handler->switch_event(sched, evsel, sample, machine);
+
+	sched->curr_pid[this_cpu] = next_pid;
+	return err;
+}
+
+static int process_sched_runtime_event(struct perf_tool *tool,
+				       struct perf_evsel *evsel,
+				       struct perf_sample *sample,
+				       struct machine *machine)
+{
+	struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
+
+	if (sched->tp_handler->runtime_event)
+		return sched->tp_handler->runtime_event(sched, evsel, sample, machine);
+
+	return 0;
+}
+
+static int perf_sched__process_fork_event(struct perf_tool *tool,
+					  union perf_event *event,
+					  struct perf_sample *sample,
+					  struct machine *machine)
+{
+	struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
+
+	/* run the fork event through the perf machineruy */
+	perf_event__process_fork(tool, event, sample, machine);
+
+	/* and then run additional processing needed for this command */
+	if (sched->tp_handler->fork_event)
+		return sched->tp_handler->fork_event(sched, event, machine);
+
+	return 0;
+}
+
+static int process_sched_migrate_task_event(struct perf_tool *tool,
+					    struct perf_evsel *evsel,
+					    struct perf_sample *sample,
+					    struct machine *machine)
+{
+	struct perf_sched *sched = container_of(tool, struct perf_sched, tool);
+
+	if (sched->tp_handler->migrate_task_event)
+		return sched->tp_handler->migrate_task_event(sched, evsel, sample, machine);
+
+	return 0;
+}
+
+typedef int (*tracepoint_handler)(struct perf_tool *tool,
+				  struct perf_evsel *evsel,
+				  struct perf_sample *sample,
+				  struct machine *machine);
+
+static int perf_sched__process_tracepoint_sample(struct perf_tool *tool __maybe_unused,
+						 union perf_event *event __maybe_unused,
+						 struct perf_sample *sample,
+						 struct perf_evsel *evsel,
+						 struct machine *machine)
+{
+	int err = 0;
+
+	if (evsel->handler != NULL) {
+		tracepoint_handler f = evsel->handler;
+		err = f(tool, evsel, sample, machine);
+	}
+
+	return err;
+}
+
+static int perf_sched__read_events(struct perf_sched *sched)
+{
+	const struct perf_evsel_str_handler handlers[] = {
+		{ "sched:sched_switch",	      process_sched_switch_event, },
+		{ "sched:sched_stat_runtime", process_sched_runtime_event, },
+		{ "sched:sched_wakeup",	      process_sched_wakeup_event, },
+		{ "sched:sched_wakeup_new",   process_sched_wakeup_event, },
+		{ "sched:sched_migrate_task", process_sched_migrate_task_event, },
+	};
+	struct perf_session *session;
+	struct perf_data_file file = {
+		.path = input_name,
+		.mode = PERF_DATA_MODE_READ,
+		.force = sched->force,
+	};
+	int rc = -1;
+
+	session = perf_session__new(&file, false, &sched->tool);
+	if (session == NULL) {
+		pr_debug("No Memory for session\n");
+		return -1;
+	}
+
+	symbol__init(&session->header.env);
+
+	if (perf_session__set_tracepoints_handlers(session, handlers))
+		goto out_delete;
+
+	if (perf_session__has_traces(session, "record -R")) {
+		int err = perf_session__process_events(session);
+		if (err) {
+			pr_err("Failed to process events, error %d", err);
+			goto out_delete;
+		}
+
+		sched->nr_events      = session->evlist->stats.nr_events[0];
+		sched->nr_lost_events = session->evlist->stats.total_lost;
+		sched->nr_lost_chunks = session->evlist->stats.nr_events[PERF_RECORD_LOST];
+	}
+
+	rc = 0;
+out_delete:
+	perf_session__delete(session);
+	return rc;
+}
+
+static void print_bad_events(struct perf_sched *sched)
+{
+	if (sched->nr_unordered_timestamps && sched->nr_timestamps) {
+		printf("  INFO: %.3f%% unordered timestamps (%ld out of %ld)\n",
+			(double)sched->nr_unordered_timestamps/(double)sched->nr_timestamps*100.0,
+			sched->nr_unordered_timestamps, sched->nr_timestamps);
+	}
+	if (sched->nr_lost_events && sched->nr_events) {
+		printf("  INFO: %.3f%% lost events (%ld out of %ld, in %ld chunks)\n",
+			(double)sched->nr_lost_events/(double)sched->nr_events * 100.0,
+			sched->nr_lost_events, sched->nr_events, sched->nr_lost_chunks);
+	}
+	if (sched->nr_context_switch_bugs && sched->nr_timestamps) {
+		printf("  INFO: %.3f%% context switch bugs (%ld out of %ld)",
+			(double)sched->nr_context_switch_bugs/(double)sched->nr_timestamps*100.0,
+			sched->nr_context_switch_bugs, sched->nr_timestamps);
+		if (sched->nr_lost_events)
+			printf(" (due to lost events?)");
+		printf("\n");
+	}
+}
+
+static int perf_sched__lat(struct perf_sched *sched)
+{
+	struct rb_node *next;
+
+	setup_pager();
+
+	if (perf_sched__read_events(sched))
+		return -1;
+
+	perf_sched__sort_lat(sched);
+
+	printf("\n -----------------------------------------------------------------------------------------------------------------\n");
+	printf("  Task                  |   Runtime ms  | Switches | Average delay ms | Maximum delay ms | Maximum delay at       |\n");
+	printf(" -----------------------------------------------------------------------------------------------------------------\n");
+
+	next = rb_first(&sched->sorted_atom_root);
+
+	while (next) {
+		struct work_atoms *work_list;
+
+		work_list = rb_entry(next, struct work_atoms, node);
+		output_lat_thread(sched, work_list);
+		next = rb_next(next);
+		thread__zput(work_list->thread);
+	}
+
+	printf(" -----------------------------------------------------------------------------------------------------------------\n");
+	printf("  TOTAL:                |%11.3f ms |%9" PRIu64 " |\n",
+		(double)sched->all_runtime / 1e6, sched->all_count);
+
+	printf(" ---------------------------------------------------\n");
+
+	print_bad_events(sched);
+	printf("\n");
+
+	return 0;
+}
+
+static int perf_sched__map(struct perf_sched *sched)
+{
+	sched->max_cpu = sysconf(_SC_NPROCESSORS_CONF);
+
+	setup_pager();
+	if (perf_sched__read_events(sched))
+		return -1;
+	print_bad_events(sched);
+	return 0;
+}
+
+static int perf_sched__replay(struct perf_sched *sched)
+{
+	unsigned long i;
+
+	calibrate_run_measurement_overhead(sched);
+	calibrate_sleep_measurement_overhead(sched);
+
+	test_calibrations(sched);
+
+	if (perf_sched__read_events(sched))
+		return -1;
+
+	printf("nr_run_events:        %ld\n", sched->nr_run_events);
+	printf("nr_sleep_events:      %ld\n", sched->nr_sleep_events);
+	printf("nr_wakeup_events:     %ld\n", sched->nr_wakeup_events);
+
+	if (sched->targetless_wakeups)
+		printf("target-less wakeups:  %ld\n", sched->targetless_wakeups);
+	if (sched->multitarget_wakeups)
+		printf("multi-target wakeups: %ld\n", sched->multitarget_wakeups);
+	if (sched->nr_run_events_optimized)
+		printf("run atoms optimized: %ld\n",
+			sched->nr_run_events_optimized);
+
+	print_task_traces(sched);
+	add_cross_task_wakeups(sched);
+
+	create_tasks(sched);
+	printf("------------------------------------------------------------\n");
+	for (i = 0; i < sched->replay_repeat; i++)
+		run_one_test(sched);
+
+	return 0;
+}
+
+static void setup_sorting(struct perf_sched *sched, const struct option *options,
+			  const char * const usage_msg[])
+{
+	char *tmp, *tok, *str = strdup(sched->sort_order);
+
+	for (tok = strtok_r(str, ", ", &tmp);
+			tok; tok = strtok_r(NULL, ", ", &tmp)) {
+		if (sort_dimension__add(tok, &sched->sort_list) < 0) {
+			error("Unknown --sort key: `%s'", tok);
+			usage_with_options(usage_msg, options);
+		}
+	}
+
+	free(str);
+
+	sort_dimension__add("pid", &sched->cmp_pid);
+}
+
+static int __cmd_record(int argc, const char **argv)
+{
+	unsigned int rec_argc, i, j;
+	const char **rec_argv;
+	const char * const record_args[] = {
+		"record",
+		"-a",
+		"-R",
+		"-m", "1024",
+		"-c", "1",
+		"-e", "sched:sched_switch",
+		"-e", "sched:sched_stat_wait",
+		"-e", "sched:sched_stat_sleep",
+		"-e", "sched:sched_stat_iowait",
+		"-e", "sched:sched_stat_runtime",
+		"-e", "sched:sched_process_fork",
+		"-e", "sched:sched_wakeup",
+		"-e", "sched:sched_wakeup_new",
+		"-e", "sched:sched_migrate_task",
+	};
+
+	rec_argc = ARRAY_SIZE(record_args) + argc - 1;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+	if (rec_argv == NULL)
+		return -ENOMEM;
+
+	for (i = 0; i < ARRAY_SIZE(record_args); i++)
+		rec_argv[i] = strdup(record_args[i]);
+
+	for (j = 1; j < (unsigned int)argc; j++, i++)
+		rec_argv[i] = argv[j];
+
+	BUG_ON(i != rec_argc);
+
+	return cmd_record(i, rec_argv, NULL);
+}
+
+int cmd_sched(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	const char default_sort_order[] = "avg, max, switch, runtime";
+	struct perf_sched sched = {
+		.tool = {
+			.sample		 = perf_sched__process_tracepoint_sample,
+			.comm		 = perf_event__process_comm,
+			.lost		 = perf_event__process_lost,
+			.fork		 = perf_sched__process_fork_event,
+			.ordered_events = true,
+		},
+		.cmp_pid	      = LIST_HEAD_INIT(sched.cmp_pid),
+		.sort_list	      = LIST_HEAD_INIT(sched.sort_list),
+		.start_work_mutex     = PTHREAD_MUTEX_INITIALIZER,
+		.work_done_wait_mutex = PTHREAD_MUTEX_INITIALIZER,
+		.sort_order	      = default_sort_order,
+		.replay_repeat	      = 10,
+		.profile_cpu	      = -1,
+		.next_shortname1      = 'A',
+		.next_shortname2      = '0',
+	};
+	const struct option latency_options[] = {
+	OPT_STRING('s', "sort", &sched.sort_order, "key[,key2...]",
+		   "sort by key(s): runtime, switch, avg, max"),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_INTEGER('C', "CPU", &sched.profile_cpu,
+		    "CPU to profile on"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_END()
+	};
+	const struct option replay_options[] = {
+	OPT_UINTEGER('r', "repeat", &sched.replay_repeat,
+		     "repeat the workload replay N times (-1: infinite)"),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_BOOLEAN('f', "force", &sched.force, "don't complain, do it"),
+	OPT_END()
+	};
+	const struct option sched_options[] = {
+	OPT_STRING('i', "input", &input_name, "file",
+		    "input file name"),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_END()
+	};
+	const char * const latency_usage[] = {
+		"perf sched latency [<options>]",
+		NULL
+	};
+	const char * const replay_usage[] = {
+		"perf sched replay [<options>]",
+		NULL
+	};
+	const char *const sched_subcommands[] = { "record", "latency", "map",
+						  "replay", "script", NULL };
+	const char *sched_usage[] = {
+		NULL,
+		NULL
+	};
+	struct trace_sched_handler lat_ops  = {
+		.wakeup_event	    = latency_wakeup_event,
+		.switch_event	    = latency_switch_event,
+		.runtime_event	    = latency_runtime_event,
+		.migrate_task_event = latency_migrate_task_event,
+	};
+	struct trace_sched_handler map_ops  = {
+		.switch_event	    = map_switch_event,
+	};
+	struct trace_sched_handler replay_ops  = {
+		.wakeup_event	    = replay_wakeup_event,
+		.switch_event	    = replay_switch_event,
+		.fork_event	    = replay_fork_event,
+	};
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(sched.curr_pid); i++)
+		sched.curr_pid[i] = -1;
+
+	argc = parse_options_subcommand(argc, argv, sched_options, sched_subcommands,
+					sched_usage, PARSE_OPT_STOP_AT_NON_OPTION);
+	if (!argc)
+		usage_with_options(sched_usage, sched_options);
+
+	/*
+	 * Aliased to 'perf script' for now:
+	 */
+	if (!strcmp(argv[0], "script"))
+		return cmd_script(argc, argv, prefix);
+
+	if (!strncmp(argv[0], "rec", 3)) {
+		return __cmd_record(argc, argv);
+	} else if (!strncmp(argv[0], "lat", 3)) {
+		sched.tp_handler = &lat_ops;
+		if (argc > 1) {
+			argc = parse_options(argc, argv, latency_options, latency_usage, 0);
+			if (argc)
+				usage_with_options(latency_usage, latency_options);
+		}
+		setup_sorting(&sched, latency_options, latency_usage);
+		return perf_sched__lat(&sched);
+	} else if (!strcmp(argv[0], "map")) {
+		sched.tp_handler = &map_ops;
+		setup_sorting(&sched, latency_options, latency_usage);
+		return perf_sched__map(&sched);
+	} else if (!strncmp(argv[0], "rep", 3)) {
+		sched.tp_handler = &replay_ops;
+		if (argc) {
+			argc = parse_options(argc, argv, replay_options, replay_usage, 0);
+			if (argc)
+				usage_with_options(replay_usage, replay_options);
+		}
+		return perf_sched__replay(&sched);
+	} else {
+		usage_with_options(sched_usage, sched_options);
+	}
+
+	return 0;
+}
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
new file mode 100644
index 000000000..58f10b8e6
--- /dev/null
+++ b/tools/perf/builtin-script.c
@@ -0,0 +1,1844 @@
+#include "builtin.h"
+
+#include "perf.h"
+#include "util/cache.h"
+#include "util/debug.h"
+#include "util/exec_cmd.h"
+#include "util/header.h"
+#include "util/parse-options.h"
+#include "util/session.h"
+#include "util/tool.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/trace-event.h"
+#include "util/util.h"
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/sort.h"
+#include "util/data.h"
+#include <linux/bitmap.h>
+
+static char const		*script_name;
+static char const		*generate_script_lang;
+static bool			debug_mode;
+static u64			last_timestamp;
+static u64			nr_unordered;
+static bool			no_callchain;
+static bool			latency_format;
+static bool			system_wide;
+static const char		*cpu_list;
+static DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
+
+enum perf_output_field {
+	PERF_OUTPUT_COMM            = 1U << 0,
+	PERF_OUTPUT_TID             = 1U << 1,
+	PERF_OUTPUT_PID             = 1U << 2,
+	PERF_OUTPUT_TIME            = 1U << 3,
+	PERF_OUTPUT_CPU             = 1U << 4,
+	PERF_OUTPUT_EVNAME          = 1U << 5,
+	PERF_OUTPUT_TRACE           = 1U << 6,
+	PERF_OUTPUT_IP              = 1U << 7,
+	PERF_OUTPUT_SYM             = 1U << 8,
+	PERF_OUTPUT_DSO             = 1U << 9,
+	PERF_OUTPUT_ADDR            = 1U << 10,
+	PERF_OUTPUT_SYMOFFSET       = 1U << 11,
+	PERF_OUTPUT_SRCLINE         = 1U << 12,
+	PERF_OUTPUT_PERIOD          = 1U << 13,
+};
+
+struct output_option {
+	const char *str;
+	enum perf_output_field field;
+} all_output_options[] = {
+	{.str = "comm",  .field = PERF_OUTPUT_COMM},
+	{.str = "tid",   .field = PERF_OUTPUT_TID},
+	{.str = "pid",   .field = PERF_OUTPUT_PID},
+	{.str = "time",  .field = PERF_OUTPUT_TIME},
+	{.str = "cpu",   .field = PERF_OUTPUT_CPU},
+	{.str = "event", .field = PERF_OUTPUT_EVNAME},
+	{.str = "trace", .field = PERF_OUTPUT_TRACE},
+	{.str = "ip",    .field = PERF_OUTPUT_IP},
+	{.str = "sym",   .field = PERF_OUTPUT_SYM},
+	{.str = "dso",   .field = PERF_OUTPUT_DSO},
+	{.str = "addr",  .field = PERF_OUTPUT_ADDR},
+	{.str = "symoff", .field = PERF_OUTPUT_SYMOFFSET},
+	{.str = "srcline", .field = PERF_OUTPUT_SRCLINE},
+	{.str = "period", .field = PERF_OUTPUT_PERIOD},
+};
+
+/* default set to maintain compatibility with current format */
+static struct {
+	bool user_set;
+	bool wildcard_set;
+	unsigned int print_ip_opts;
+	u64 fields;
+	u64 invalid_fields;
+} output[PERF_TYPE_MAX] = {
+
+	[PERF_TYPE_HARDWARE] = {
+		.user_set = false,
+
+		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
+			      PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
+			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
+			      PERF_OUTPUT_SYM | PERF_OUTPUT_DSO |
+			      PERF_OUTPUT_PERIOD,
+
+		.invalid_fields = PERF_OUTPUT_TRACE,
+	},
+
+	[PERF_TYPE_SOFTWARE] = {
+		.user_set = false,
+
+		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
+			      PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
+			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
+			      PERF_OUTPUT_SYM | PERF_OUTPUT_DSO |
+			      PERF_OUTPUT_PERIOD,
+
+		.invalid_fields = PERF_OUTPUT_TRACE,
+	},
+
+	[PERF_TYPE_TRACEPOINT] = {
+		.user_set = false,
+
+		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
+				  PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
+				  PERF_OUTPUT_EVNAME | PERF_OUTPUT_TRACE,
+	},
+
+	[PERF_TYPE_RAW] = {
+		.user_set = false,
+
+		.fields = PERF_OUTPUT_COMM | PERF_OUTPUT_TID |
+			      PERF_OUTPUT_CPU | PERF_OUTPUT_TIME |
+			      PERF_OUTPUT_EVNAME | PERF_OUTPUT_IP |
+			      PERF_OUTPUT_SYM | PERF_OUTPUT_DSO |
+			      PERF_OUTPUT_PERIOD,
+
+		.invalid_fields = PERF_OUTPUT_TRACE,
+	},
+};
+
+static bool output_set_by_user(void)
+{
+	int j;
+	for (j = 0; j < PERF_TYPE_MAX; ++j) {
+		if (output[j].user_set)
+			return true;
+	}
+	return false;
+}
+
+static const char *output_field2str(enum perf_output_field field)
+{
+	int i, imax = ARRAY_SIZE(all_output_options);
+	const char *str = "";
+
+	for (i = 0; i < imax; ++i) {
+		if (all_output_options[i].field == field) {
+			str = all_output_options[i].str;
+			break;
+		}
+	}
+	return str;
+}
+
+#define PRINT_FIELD(x)  (output[attr->type].fields & PERF_OUTPUT_##x)
+
+static int perf_evsel__check_stype(struct perf_evsel *evsel,
+				   u64 sample_type, const char *sample_msg,
+				   enum perf_output_field field)
+{
+	struct perf_event_attr *attr = &evsel->attr;
+	int type = attr->type;
+	const char *evname;
+
+	if (attr->sample_type & sample_type)
+		return 0;
+
+	if (output[type].user_set) {
+		evname = perf_evsel__name(evsel);
+		pr_err("Samples for '%s' event do not have %s attribute set. "
+		       "Cannot print '%s' field.\n",
+		       evname, sample_msg, output_field2str(field));
+		return -1;
+	}
+
+	/* user did not ask for it explicitly so remove from the default list */
+	output[type].fields &= ~field;
+	evname = perf_evsel__name(evsel);
+	pr_debug("Samples for '%s' event do not have %s attribute set. "
+		 "Skipping '%s' field.\n",
+		 evname, sample_msg, output_field2str(field));
+
+	return 0;
+}
+
+static int perf_evsel__check_attr(struct perf_evsel *evsel,
+				  struct perf_session *session)
+{
+	struct perf_event_attr *attr = &evsel->attr;
+
+	if (PRINT_FIELD(TRACE) &&
+		!perf_session__has_traces(session, "record -R"))
+		return -EINVAL;
+
+	if (PRINT_FIELD(IP)) {
+		if (perf_evsel__check_stype(evsel, PERF_SAMPLE_IP, "IP",
+					    PERF_OUTPUT_IP))
+			return -EINVAL;
+	}
+
+	if (PRINT_FIELD(ADDR) &&
+		perf_evsel__check_stype(evsel, PERF_SAMPLE_ADDR, "ADDR",
+					PERF_OUTPUT_ADDR))
+		return -EINVAL;
+
+	if (PRINT_FIELD(SYM) && !PRINT_FIELD(IP) && !PRINT_FIELD(ADDR)) {
+		pr_err("Display of symbols requested but neither sample IP nor "
+			   "sample address\nis selected. Hence, no addresses to convert "
+		       "to symbols.\n");
+		return -EINVAL;
+	}
+	if (PRINT_FIELD(SYMOFFSET) && !PRINT_FIELD(SYM)) {
+		pr_err("Display of offsets requested but symbol is not"
+		       "selected.\n");
+		return -EINVAL;
+	}
+	if (PRINT_FIELD(DSO) && !PRINT_FIELD(IP) && !PRINT_FIELD(ADDR)) {
+		pr_err("Display of DSO requested but neither sample IP nor "
+			   "sample address\nis selected. Hence, no addresses to convert "
+		       "to DSO.\n");
+		return -EINVAL;
+	}
+	if (PRINT_FIELD(SRCLINE) && !PRINT_FIELD(IP)) {
+		pr_err("Display of source line number requested but sample IP is not\n"
+		       "selected. Hence, no address to lookup the source line number.\n");
+		return -EINVAL;
+	}
+
+	if ((PRINT_FIELD(PID) || PRINT_FIELD(TID)) &&
+		perf_evsel__check_stype(evsel, PERF_SAMPLE_TID, "TID",
+					PERF_OUTPUT_TID|PERF_OUTPUT_PID))
+		return -EINVAL;
+
+	if (PRINT_FIELD(TIME) &&
+		perf_evsel__check_stype(evsel, PERF_SAMPLE_TIME, "TIME",
+					PERF_OUTPUT_TIME))
+		return -EINVAL;
+
+	if (PRINT_FIELD(CPU) &&
+		perf_evsel__check_stype(evsel, PERF_SAMPLE_CPU, "CPU",
+					PERF_OUTPUT_CPU))
+		return -EINVAL;
+
+	if (PRINT_FIELD(PERIOD) &&
+		perf_evsel__check_stype(evsel, PERF_SAMPLE_PERIOD, "PERIOD",
+					PERF_OUTPUT_PERIOD))
+		return -EINVAL;
+
+	return 0;
+}
+
+static void set_print_ip_opts(struct perf_event_attr *attr)
+{
+	unsigned int type = attr->type;
+
+	output[type].print_ip_opts = 0;
+	if (PRINT_FIELD(IP))
+		output[type].print_ip_opts |= PRINT_IP_OPT_IP;
+
+	if (PRINT_FIELD(SYM))
+		output[type].print_ip_opts |= PRINT_IP_OPT_SYM;
+
+	if (PRINT_FIELD(DSO))
+		output[type].print_ip_opts |= PRINT_IP_OPT_DSO;
+
+	if (PRINT_FIELD(SYMOFFSET))
+		output[type].print_ip_opts |= PRINT_IP_OPT_SYMOFFSET;
+
+	if (PRINT_FIELD(SRCLINE))
+		output[type].print_ip_opts |= PRINT_IP_OPT_SRCLINE;
+}
+
+/*
+ * verify all user requested events exist and the samples
+ * have the expected data
+ */
+static int perf_session__check_output_opt(struct perf_session *session)
+{
+	int j;
+	struct perf_evsel *evsel;
+
+	for (j = 0; j < PERF_TYPE_MAX; ++j) {
+		evsel = perf_session__find_first_evtype(session, j);
+
+		/*
+		 * even if fields is set to 0 (ie., show nothing) event must
+		 * exist if user explicitly includes it on the command line
+		 */
+		if (!evsel && output[j].user_set && !output[j].wildcard_set) {
+			pr_err("%s events do not exist. "
+			       "Remove corresponding -f option to proceed.\n",
+			       event_type(j));
+			return -1;
+		}
+
+		if (evsel && output[j].fields &&
+			perf_evsel__check_attr(evsel, session))
+			return -1;
+
+		if (evsel == NULL)
+			continue;
+
+		set_print_ip_opts(&evsel->attr);
+	}
+
+	if (!no_callchain) {
+		bool use_callchain = false;
+
+		evlist__for_each(session->evlist, evsel) {
+			if (evsel->attr.sample_type & PERF_SAMPLE_CALLCHAIN) {
+				use_callchain = true;
+				break;
+			}
+		}
+		if (!use_callchain)
+			symbol_conf.use_callchain = false;
+	}
+
+	/*
+	 * set default for tracepoints to print symbols only
+	 * if callchains are present
+	 */
+	if (symbol_conf.use_callchain &&
+	    !output[PERF_TYPE_TRACEPOINT].user_set) {
+		struct perf_event_attr *attr;
+
+		j = PERF_TYPE_TRACEPOINT;
+		evsel = perf_session__find_first_evtype(session, j);
+		if (evsel == NULL)
+			goto out;
+
+		attr = &evsel->attr;
+
+		if (attr->sample_type & PERF_SAMPLE_CALLCHAIN) {
+			output[j].fields |= PERF_OUTPUT_IP;
+			output[j].fields |= PERF_OUTPUT_SYM;
+			output[j].fields |= PERF_OUTPUT_DSO;
+			set_print_ip_opts(attr);
+		}
+	}
+
+out:
+	return 0;
+}
+
+static void print_sample_start(struct perf_sample *sample,
+			       struct thread *thread,
+			       struct perf_evsel *evsel)
+{
+	struct perf_event_attr *attr = &evsel->attr;
+	unsigned long secs;
+	unsigned long usecs;
+	unsigned long long nsecs;
+
+	if (PRINT_FIELD(COMM)) {
+		if (latency_format)
+			printf("%8.8s ", thread__comm_str(thread));
+		else if (PRINT_FIELD(IP) && symbol_conf.use_callchain)
+			printf("%s ", thread__comm_str(thread));
+		else
+			printf("%16s ", thread__comm_str(thread));
+	}
+
+	if (PRINT_FIELD(PID) && PRINT_FIELD(TID))
+		printf("%5d/%-5d ", sample->pid, sample->tid);
+	else if (PRINT_FIELD(PID))
+		printf("%5d ", sample->pid);
+	else if (PRINT_FIELD(TID))
+		printf("%5d ", sample->tid);
+
+	if (PRINT_FIELD(CPU)) {
+		if (latency_format)
+			printf("%3d ", sample->cpu);
+		else
+			printf("[%03d] ", sample->cpu);
+	}
+
+	if (PRINT_FIELD(TIME)) {
+		nsecs = sample->time;
+		secs = nsecs / NSECS_PER_SEC;
+		nsecs -= secs * NSECS_PER_SEC;
+		usecs = nsecs / NSECS_PER_USEC;
+		printf("%5lu.%06lu: ", secs, usecs);
+	}
+}
+
+static void print_sample_addr(union perf_event *event,
+			  struct perf_sample *sample,
+			  struct thread *thread,
+			  struct perf_event_attr *attr)
+{
+	struct addr_location al;
+
+	printf("%16" PRIx64, sample->addr);
+
+	if (!sample_addr_correlates_sym(attr))
+		return;
+
+	perf_event__preprocess_sample_addr(event, sample, thread, &al);
+
+	if (PRINT_FIELD(SYM)) {
+		printf(" ");
+		if (PRINT_FIELD(SYMOFFSET))
+			symbol__fprintf_symname_offs(al.sym, &al, stdout);
+		else
+			symbol__fprintf_symname(al.sym, stdout);
+	}
+
+	if (PRINT_FIELD(DSO)) {
+		printf(" (");
+		map__fprintf_dsoname(al.map, stdout);
+		printf(")");
+	}
+}
+
+static void print_sample_bts(union perf_event *event,
+			     struct perf_sample *sample,
+			     struct perf_evsel *evsel,
+			     struct thread *thread,
+			     struct addr_location *al)
+{
+	struct perf_event_attr *attr = &evsel->attr;
+	bool print_srcline_last = false;
+
+	/* print branch_from information */
+	if (PRINT_FIELD(IP)) {
+		unsigned int print_opts = output[attr->type].print_ip_opts;
+
+		if (symbol_conf.use_callchain && sample->callchain) {
+			printf("\n");
+		} else {
+			printf(" ");
+			if (print_opts & PRINT_IP_OPT_SRCLINE) {
+				print_srcline_last = true;
+				print_opts &= ~PRINT_IP_OPT_SRCLINE;
+			}
+		}
+		perf_evsel__print_ip(evsel, sample, al, print_opts,
+				     PERF_MAX_STACK_DEPTH);
+	}
+
+	/* print branch_to information */
+	if (PRINT_FIELD(ADDR) ||
+	    ((evsel->attr.sample_type & PERF_SAMPLE_ADDR) &&
+	     !output[attr->type].user_set)) {
+		printf(" => ");
+		print_sample_addr(event, sample, thread, attr);
+	}
+
+	if (print_srcline_last)
+		map__fprintf_srcline(al->map, al->addr, "\n  ", stdout);
+
+	printf("\n");
+}
+
+static void process_event(union perf_event *event, struct perf_sample *sample,
+			  struct perf_evsel *evsel, struct addr_location *al)
+{
+	struct thread *thread = al->thread;
+	struct perf_event_attr *attr = &evsel->attr;
+
+	if (output[attr->type].fields == 0)
+		return;
+
+	print_sample_start(sample, thread, evsel);
+
+	if (PRINT_FIELD(PERIOD))
+		printf("%10" PRIu64 " ", sample->period);
+
+	if (PRINT_FIELD(EVNAME)) {
+		const char *evname = perf_evsel__name(evsel);
+		printf("%s: ", evname ? evname : "[unknown]");
+	}
+
+	if (is_bts_event(attr)) {
+		print_sample_bts(event, sample, evsel, thread, al);
+		return;
+	}
+
+	if (PRINT_FIELD(TRACE))
+		event_format__print(evsel->tp_format, sample->cpu,
+				    sample->raw_data, sample->raw_size);
+	if (PRINT_FIELD(ADDR))
+		print_sample_addr(event, sample, thread, attr);
+
+	if (PRINT_FIELD(IP)) {
+		if (!symbol_conf.use_callchain)
+			printf(" ");
+		else
+			printf("\n");
+
+		perf_evsel__print_ip(evsel, sample, al,
+				     output[attr->type].print_ip_opts,
+				     PERF_MAX_STACK_DEPTH);
+	}
+
+	printf("\n");
+}
+
+static int default_start_script(const char *script __maybe_unused,
+				int argc __maybe_unused,
+				const char **argv __maybe_unused)
+{
+	return 0;
+}
+
+static int default_flush_script(void)
+{
+	return 0;
+}
+
+static int default_stop_script(void)
+{
+	return 0;
+}
+
+static int default_generate_script(struct pevent *pevent __maybe_unused,
+				   const char *outfile __maybe_unused)
+{
+	return 0;
+}
+
+static struct scripting_ops default_scripting_ops = {
+	.start_script		= default_start_script,
+	.flush_script		= default_flush_script,
+	.stop_script		= default_stop_script,
+	.process_event		= process_event,
+	.generate_script	= default_generate_script,
+};
+
+static struct scripting_ops	*scripting_ops;
+
+static void setup_scripting(void)
+{
+	setup_perl_scripting();
+	setup_python_scripting();
+
+	scripting_ops = &default_scripting_ops;
+}
+
+static int flush_scripting(void)
+{
+	return scripting_ops->flush_script();
+}
+
+static int cleanup_scripting(void)
+{
+	pr_debug("\nperf script stopped\n");
+
+	return scripting_ops->stop_script();
+}
+
+static int process_sample_event(struct perf_tool *tool __maybe_unused,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel,
+				struct machine *machine)
+{
+	struct addr_location al;
+
+	if (debug_mode) {
+		if (sample->time < last_timestamp) {
+			pr_err("Samples misordered, previous: %" PRIu64
+				" this: %" PRIu64 "\n", last_timestamp,
+				sample->time);
+			nr_unordered++;
+		}
+		last_timestamp = sample->time;
+		return 0;
+	}
+
+	if (perf_event__preprocess_sample(event, machine, &al, sample) < 0) {
+		pr_err("problem processing %d event, skipping it.\n",
+		       event->header.type);
+		return -1;
+	}
+
+	if (al.filtered)
+		return 0;
+
+	if (cpu_list && !test_bit(sample->cpu, cpu_bitmap))
+		return 0;
+
+	scripting_ops->process_event(event, sample, evsel, &al);
+
+	return 0;
+}
+
+struct perf_script {
+	struct perf_tool	tool;
+	struct perf_session	*session;
+	bool			show_task_events;
+	bool			show_mmap_events;
+};
+
+static int process_attr(struct perf_tool *tool, union perf_event *event,
+			struct perf_evlist **pevlist)
+{
+	struct perf_script *scr = container_of(tool, struct perf_script, tool);
+	struct perf_evlist *evlist;
+	struct perf_evsel *evsel, *pos;
+	int err;
+
+	err = perf_event__process_attr(tool, event, pevlist);
+	if (err)
+		return err;
+
+	evlist = *pevlist;
+	evsel = perf_evlist__last(*pevlist);
+
+	if (evsel->attr.type >= PERF_TYPE_MAX)
+		return 0;
+
+	evlist__for_each(evlist, pos) {
+		if (pos->attr.type == evsel->attr.type && pos != evsel)
+			return 0;
+	}
+
+	set_print_ip_opts(&evsel->attr);
+
+	return perf_evsel__check_attr(evsel, scr->session);
+}
+
+static int process_comm_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+	int ret = -1;
+
+	thread = machine__findnew_thread(machine, event->comm.pid, event->comm.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing COMM event, skipping it.\n");
+		return -1;
+	}
+
+	if (perf_event__process_comm(tool, event, sample, machine) < 0)
+		goto out;
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = 0;
+		sample->tid = event->comm.tid;
+		sample->pid = event->comm.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+	ret = 0;
+
+out:
+	return ret;
+}
+
+static int process_fork_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+
+	if (perf_event__process_fork(tool, event, sample, machine) < 0)
+		return -1;
+
+	thread = machine__findnew_thread(machine, event->fork.pid, event->fork.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing FORK event, skipping it.\n");
+		return -1;
+	}
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = event->fork.time;
+		sample->tid = event->fork.tid;
+		sample->pid = event->fork.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+
+	return 0;
+}
+static int process_exit_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+
+	thread = machine__findnew_thread(machine, event->fork.pid, event->fork.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing EXIT event, skipping it.\n");
+		return -1;
+	}
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = 0;
+		sample->tid = event->comm.tid;
+		sample->pid = event->comm.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+
+	if (perf_event__process_exit(tool, event, sample, machine) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int process_mmap_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+
+	if (perf_event__process_mmap(tool, event, sample, machine) < 0)
+		return -1;
+
+	thread = machine__findnew_thread(machine, event->mmap.pid, event->mmap.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing MMAP event, skipping it.\n");
+		return -1;
+	}
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = 0;
+		sample->tid = event->mmap.tid;
+		sample->pid = event->mmap.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+
+	return 0;
+}
+
+static int process_mmap2_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine)
+{
+	struct thread *thread;
+	struct perf_script *script = container_of(tool, struct perf_script, tool);
+	struct perf_session *session = script->session;
+	struct perf_evsel *evsel = perf_evlist__first(session->evlist);
+
+	if (perf_event__process_mmap2(tool, event, sample, machine) < 0)
+		return -1;
+
+	thread = machine__findnew_thread(machine, event->mmap2.pid, event->mmap2.tid);
+	if (thread == NULL) {
+		pr_debug("problem processing MMAP2 event, skipping it.\n");
+		return -1;
+	}
+
+	if (!evsel->attr.sample_id_all) {
+		sample->cpu = 0;
+		sample->time = 0;
+		sample->tid = event->mmap2.tid;
+		sample->pid = event->mmap2.pid;
+	}
+	print_sample_start(sample, thread, evsel);
+	perf_event__fprintf(event, stdout);
+
+	return 0;
+}
+
+static void sig_handler(int sig __maybe_unused)
+{
+	session_done = 1;
+}
+
+static int __cmd_script(struct perf_script *script)
+{
+	int ret;
+
+	signal(SIGINT, sig_handler);
+
+	/* override event processing functions */
+	if (script->show_task_events) {
+		script->tool.comm = process_comm_event;
+		script->tool.fork = process_fork_event;
+		script->tool.exit = process_exit_event;
+	}
+	if (script->show_mmap_events) {
+		script->tool.mmap = process_mmap_event;
+		script->tool.mmap2 = process_mmap2_event;
+	}
+
+	ret = perf_session__process_events(script->session);
+
+	if (debug_mode)
+		pr_err("Misordered timestamps: %" PRIu64 "\n", nr_unordered);
+
+	return ret;
+}
+
+struct script_spec {
+	struct list_head	node;
+	struct scripting_ops	*ops;
+	char			spec[0];
+};
+
+static LIST_HEAD(script_specs);
+
+static struct script_spec *script_spec__new(const char *spec,
+					    struct scripting_ops *ops)
+{
+	struct script_spec *s = malloc(sizeof(*s) + strlen(spec) + 1);
+
+	if (s != NULL) {
+		strcpy(s->spec, spec);
+		s->ops = ops;
+	}
+
+	return s;
+}
+
+static void script_spec__add(struct script_spec *s)
+{
+	list_add_tail(&s->node, &script_specs);
+}
+
+static struct script_spec *script_spec__find(const char *spec)
+{
+	struct script_spec *s;
+
+	list_for_each_entry(s, &script_specs, node)
+		if (strcasecmp(s->spec, spec) == 0)
+			return s;
+	return NULL;
+}
+
+static struct script_spec *script_spec__findnew(const char *spec,
+						struct scripting_ops *ops)
+{
+	struct script_spec *s = script_spec__find(spec);
+
+	if (s)
+		return s;
+
+	s = script_spec__new(spec, ops);
+	if (!s)
+		return NULL;
+
+	script_spec__add(s);
+
+	return s;
+}
+
+int script_spec_register(const char *spec, struct scripting_ops *ops)
+{
+	struct script_spec *s;
+
+	s = script_spec__find(spec);
+	if (s)
+		return -1;
+
+	s = script_spec__findnew(spec, ops);
+	if (!s)
+		return -1;
+
+	return 0;
+}
+
+static struct scripting_ops *script_spec__lookup(const char *spec)
+{
+	struct script_spec *s = script_spec__find(spec);
+	if (!s)
+		return NULL;
+
+	return s->ops;
+}
+
+static void list_available_languages(void)
+{
+	struct script_spec *s;
+
+	fprintf(stderr, "\n");
+	fprintf(stderr, "Scripting language extensions (used in "
+		"perf script -s [spec:]script.[spec]):\n\n");
+
+	list_for_each_entry(s, &script_specs, node)
+		fprintf(stderr, "  %-42s [%s]\n", s->spec, s->ops->name);
+
+	fprintf(stderr, "\n");
+}
+
+static int parse_scriptname(const struct option *opt __maybe_unused,
+			    const char *str, int unset __maybe_unused)
+{
+	char spec[PATH_MAX];
+	const char *script, *ext;
+	int len;
+
+	if (strcmp(str, "lang") == 0) {
+		list_available_languages();
+		exit(0);
+	}
+
+	script = strchr(str, ':');
+	if (script) {
+		len = script - str;
+		if (len >= PATH_MAX) {
+			fprintf(stderr, "invalid language specifier");
+			return -1;
+		}
+		strncpy(spec, str, len);
+		spec[len] = '\0';
+		scripting_ops = script_spec__lookup(spec);
+		if (!scripting_ops) {
+			fprintf(stderr, "invalid language specifier");
+			return -1;
+		}
+		script++;
+	} else {
+		script = str;
+		ext = strrchr(script, '.');
+		if (!ext) {
+			fprintf(stderr, "invalid script extension");
+			return -1;
+		}
+		scripting_ops = script_spec__lookup(++ext);
+		if (!scripting_ops) {
+			fprintf(stderr, "invalid script extension");
+			return -1;
+		}
+	}
+
+	script_name = strdup(script);
+
+	return 0;
+}
+
+static int parse_output_fields(const struct option *opt __maybe_unused,
+			    const char *arg, int unset __maybe_unused)
+{
+	char *tok;
+	int i, imax = ARRAY_SIZE(all_output_options);
+	int j;
+	int rc = 0;
+	char *str = strdup(arg);
+	int type = -1;
+
+	if (!str)
+		return -ENOMEM;
+
+	/* first word can state for which event type the user is specifying
+	 * the fields. If no type exists, the specified fields apply to all
+	 * event types found in the file minus the invalid fields for a type.
+	 */
+	tok = strchr(str, ':');
+	if (tok) {
+		*tok = '\0';
+		tok++;
+		if (!strcmp(str, "hw"))
+			type = PERF_TYPE_HARDWARE;
+		else if (!strcmp(str, "sw"))
+			type = PERF_TYPE_SOFTWARE;
+		else if (!strcmp(str, "trace"))
+			type = PERF_TYPE_TRACEPOINT;
+		else if (!strcmp(str, "raw"))
+			type = PERF_TYPE_RAW;
+		else {
+			fprintf(stderr, "Invalid event type in field string.\n");
+			rc = -EINVAL;
+			goto out;
+		}
+
+		if (output[type].user_set)
+			pr_warning("Overriding previous field request for %s events.\n",
+				   event_type(type));
+
+		output[type].fields = 0;
+		output[type].user_set = true;
+		output[type].wildcard_set = false;
+
+	} else {
+		tok = str;
+		if (strlen(str) == 0) {
+			fprintf(stderr,
+				"Cannot set fields to 'none' for all event types.\n");
+			rc = -EINVAL;
+			goto out;
+		}
+
+		if (output_set_by_user())
+			pr_warning("Overriding previous field request for all events.\n");
+
+		for (j = 0; j < PERF_TYPE_MAX; ++j) {
+			output[j].fields = 0;
+			output[j].user_set = true;
+			output[j].wildcard_set = true;
+		}
+	}
+
+	tok = strtok(tok, ",");
+	while (tok) {
+		for (i = 0; i < imax; ++i) {
+			if (strcmp(tok, all_output_options[i].str) == 0)
+				break;
+		}
+		if (i == imax) {
+			fprintf(stderr, "Invalid field requested.\n");
+			rc = -EINVAL;
+			goto out;
+		}
+
+		if (type == -1) {
+			/* add user option to all events types for
+			 * which it is valid
+			 */
+			for (j = 0; j < PERF_TYPE_MAX; ++j) {
+				if (output[j].invalid_fields & all_output_options[i].field) {
+					pr_warning("\'%s\' not valid for %s events. Ignoring.\n",
+						   all_output_options[i].str, event_type(j));
+				} else
+					output[j].fields |= all_output_options[i].field;
+			}
+		} else {
+			if (output[type].invalid_fields & all_output_options[i].field) {
+				fprintf(stderr, "\'%s\' not valid for %s events.\n",
+					 all_output_options[i].str, event_type(type));
+
+				rc = -EINVAL;
+				goto out;
+			}
+			output[type].fields |= all_output_options[i].field;
+		}
+
+		tok = strtok(NULL, ",");
+	}
+
+	if (type >= 0) {
+		if (output[type].fields == 0) {
+			pr_debug("No fields requested for %s type. "
+				 "Events will not be displayed.\n", event_type(type));
+		}
+	}
+
+out:
+	free(str);
+	return rc;
+}
+
+/* Helper function for filesystems that return a dent->d_type DT_UNKNOWN */
+static int is_directory(const char *base_path, const struct dirent *dent)
+{
+	char path[PATH_MAX];
+	struct stat st;
+
+	sprintf(path, "%s/%s", base_path, dent->d_name);
+	if (stat(path, &st))
+		return 0;
+
+	return S_ISDIR(st.st_mode);
+}
+
+#define for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next)\
+	while (!readdir_r(scripts_dir, &lang_dirent, &lang_next) &&	\
+	       lang_next)						\
+		if ((lang_dirent.d_type == DT_DIR ||			\
+		     (lang_dirent.d_type == DT_UNKNOWN &&		\
+		      is_directory(scripts_path, &lang_dirent))) &&	\
+		    (strcmp(lang_dirent.d_name, ".")) &&		\
+		    (strcmp(lang_dirent.d_name, "..")))
+
+#define for_each_script(lang_path, lang_dir, script_dirent, script_next)\
+	while (!readdir_r(lang_dir, &script_dirent, &script_next) &&	\
+	       script_next)						\
+		if (script_dirent.d_type != DT_DIR &&			\
+		    (script_dirent.d_type != DT_UNKNOWN ||		\
+		     !is_directory(lang_path, &script_dirent)))
+
+
+#define RECORD_SUFFIX			"-record"
+#define REPORT_SUFFIX			"-report"
+
+struct script_desc {
+	struct list_head	node;
+	char			*name;
+	char			*half_liner;
+	char			*args;
+};
+
+static LIST_HEAD(script_descs);
+
+static struct script_desc *script_desc__new(const char *name)
+{
+	struct script_desc *s = zalloc(sizeof(*s));
+
+	if (s != NULL && name)
+		s->name = strdup(name);
+
+	return s;
+}
+
+static void script_desc__delete(struct script_desc *s)
+{
+	zfree(&s->name);
+	zfree(&s->half_liner);
+	zfree(&s->args);
+	free(s);
+}
+
+static void script_desc__add(struct script_desc *s)
+{
+	list_add_tail(&s->node, &script_descs);
+}
+
+static struct script_desc *script_desc__find(const char *name)
+{
+	struct script_desc *s;
+
+	list_for_each_entry(s, &script_descs, node)
+		if (strcasecmp(s->name, name) == 0)
+			return s;
+	return NULL;
+}
+
+static struct script_desc *script_desc__findnew(const char *name)
+{
+	struct script_desc *s = script_desc__find(name);
+
+	if (s)
+		return s;
+
+	s = script_desc__new(name);
+	if (!s)
+		goto out_delete_desc;
+
+	script_desc__add(s);
+
+	return s;
+
+out_delete_desc:
+	script_desc__delete(s);
+
+	return NULL;
+}
+
+static const char *ends_with(const char *str, const char *suffix)
+{
+	size_t suffix_len = strlen(suffix);
+	const char *p = str;
+
+	if (strlen(str) > suffix_len) {
+		p = str + strlen(str) - suffix_len;
+		if (!strncmp(p, suffix, suffix_len))
+			return p;
+	}
+
+	return NULL;
+}
+
+static int read_script_info(struct script_desc *desc, const char *filename)
+{
+	char line[BUFSIZ], *p;
+	FILE *fp;
+
+	fp = fopen(filename, "r");
+	if (!fp)
+		return -1;
+
+	while (fgets(line, sizeof(line), fp)) {
+		p = ltrim(line);
+		if (strlen(p) == 0)
+			continue;
+		if (*p != '#')
+			continue;
+		p++;
+		if (strlen(p) && *p == '!')
+			continue;
+
+		p = ltrim(p);
+		if (strlen(p) && p[strlen(p) - 1] == '\n')
+			p[strlen(p) - 1] = '\0';
+
+		if (!strncmp(p, "description:", strlen("description:"))) {
+			p += strlen("description:");
+			desc->half_liner = strdup(ltrim(p));
+			continue;
+		}
+
+		if (!strncmp(p, "args:", strlen("args:"))) {
+			p += strlen("args:");
+			desc->args = strdup(ltrim(p));
+			continue;
+		}
+	}
+
+	fclose(fp);
+
+	return 0;
+}
+
+static char *get_script_root(struct dirent *script_dirent, const char *suffix)
+{
+	char *script_root, *str;
+
+	script_root = strdup(script_dirent->d_name);
+	if (!script_root)
+		return NULL;
+
+	str = (char *)ends_with(script_root, suffix);
+	if (!str) {
+		free(script_root);
+		return NULL;
+	}
+
+	*str = '\0';
+	return script_root;
+}
+
+static int list_available_scripts(const struct option *opt __maybe_unused,
+				  const char *s __maybe_unused,
+				  int unset __maybe_unused)
+{
+	struct dirent *script_next, *lang_next, script_dirent, lang_dirent;
+	char scripts_path[MAXPATHLEN];
+	DIR *scripts_dir, *lang_dir;
+	char script_path[MAXPATHLEN];
+	char lang_path[MAXPATHLEN];
+	struct script_desc *desc;
+	char first_half[BUFSIZ];
+	char *script_root;
+
+	snprintf(scripts_path, MAXPATHLEN, "%s/scripts", perf_exec_path());
+
+	scripts_dir = opendir(scripts_path);
+	if (!scripts_dir)
+		return -1;
+
+	for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next) {
+		snprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path,
+			 lang_dirent.d_name);
+		lang_dir = opendir(lang_path);
+		if (!lang_dir)
+			continue;
+
+		for_each_script(lang_path, lang_dir, script_dirent, script_next) {
+			script_root = get_script_root(&script_dirent, REPORT_SUFFIX);
+			if (script_root) {
+				desc = script_desc__findnew(script_root);
+				snprintf(script_path, MAXPATHLEN, "%s/%s",
+					 lang_path, script_dirent.d_name);
+				read_script_info(desc, script_path);
+				free(script_root);
+			}
+		}
+	}
+
+	fprintf(stdout, "List of available trace scripts:\n");
+	list_for_each_entry(desc, &script_descs, node) {
+		sprintf(first_half, "%s %s", desc->name,
+			desc->args ? desc->args : "");
+		fprintf(stdout, "  %-36s %s\n", first_half,
+			desc->half_liner ? desc->half_liner : "");
+	}
+
+	exit(0);
+}
+
+/*
+ * Some scripts specify the required events in their "xxx-record" file,
+ * this function will check if the events in perf.data match those
+ * mentioned in the "xxx-record".
+ *
+ * Fixme: All existing "xxx-record" are all in good formats "-e event ",
+ * which is covered well now. And new parsing code should be added to
+ * cover the future complexing formats like event groups etc.
+ */
+static int check_ev_match(char *dir_name, char *scriptname,
+			struct perf_session *session)
+{
+	char filename[MAXPATHLEN], evname[128];
+	char line[BUFSIZ], *p;
+	struct perf_evsel *pos;
+	int match, len;
+	FILE *fp;
+
+	sprintf(filename, "%s/bin/%s-record", dir_name, scriptname);
+
+	fp = fopen(filename, "r");
+	if (!fp)
+		return -1;
+
+	while (fgets(line, sizeof(line), fp)) {
+		p = ltrim(line);
+		if (*p == '#')
+			continue;
+
+		while (strlen(p)) {
+			p = strstr(p, "-e");
+			if (!p)
+				break;
+
+			p += 2;
+			p = ltrim(p);
+			len = strcspn(p, " \t");
+			if (!len)
+				break;
+
+			snprintf(evname, len + 1, "%s", p);
+
+			match = 0;
+			evlist__for_each(session->evlist, pos) {
+				if (!strcmp(perf_evsel__name(pos), evname)) {
+					match = 1;
+					break;
+				}
+			}
+
+			if (!match) {
+				fclose(fp);
+				return -1;
+			}
+		}
+	}
+
+	fclose(fp);
+	return 0;
+}
+
+/*
+ * Return -1 if none is found, otherwise the actual scripts number.
+ *
+ * Currently the only user of this function is the script browser, which
+ * will list all statically runnable scripts, select one, execute it and
+ * show the output in a perf browser.
+ */
+int find_scripts(char **scripts_array, char **scripts_path_array)
+{
+	struct dirent *script_next, *lang_next, script_dirent, lang_dirent;
+	char scripts_path[MAXPATHLEN], lang_path[MAXPATHLEN];
+	DIR *scripts_dir, *lang_dir;
+	struct perf_session *session;
+	struct perf_data_file file = {
+		.path = input_name,
+		.mode = PERF_DATA_MODE_READ,
+	};
+	char *temp;
+	int i = 0;
+
+	session = perf_session__new(&file, false, NULL);
+	if (!session)
+		return -1;
+
+	snprintf(scripts_path, MAXPATHLEN, "%s/scripts", perf_exec_path());
+
+	scripts_dir = opendir(scripts_path);
+	if (!scripts_dir) {
+		perf_session__delete(session);
+		return -1;
+	}
+
+	for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next) {
+		snprintf(lang_path, MAXPATHLEN, "%s/%s", scripts_path,
+			 lang_dirent.d_name);
+#ifdef NO_LIBPERL
+		if (strstr(lang_path, "perl"))
+			continue;
+#endif
+#ifdef NO_LIBPYTHON
+		if (strstr(lang_path, "python"))
+			continue;
+#endif
+
+		lang_dir = opendir(lang_path);
+		if (!lang_dir)
+			continue;
+
+		for_each_script(lang_path, lang_dir, script_dirent, script_next) {
+			/* Skip those real time scripts: xxxtop.p[yl] */
+			if (strstr(script_dirent.d_name, "top."))
+				continue;
+			sprintf(scripts_path_array[i], "%s/%s", lang_path,
+				script_dirent.d_name);
+			temp = strchr(script_dirent.d_name, '.');
+			snprintf(scripts_array[i],
+				(temp - script_dirent.d_name) + 1,
+				"%s", script_dirent.d_name);
+
+			if (check_ev_match(lang_path,
+					scripts_array[i], session))
+				continue;
+
+			i++;
+		}
+		closedir(lang_dir);
+	}
+
+	closedir(scripts_dir);
+	perf_session__delete(session);
+	return i;
+}
+
+static char *get_script_path(const char *script_root, const char *suffix)
+{
+	struct dirent *script_next, *lang_next, script_dirent, lang_dirent;
+	char scripts_path[MAXPATHLEN];
+	char script_path[MAXPATHLEN];
+	DIR *scripts_dir, *lang_dir;
+	char lang_path[MAXPATHLEN];
+	char *__script_root;
+
+	snprintf(scripts_path, MAXPATHLEN, "%s/scripts", perf_exec_path());
+
+	scripts_dir = opendir(scripts_path);
+	if (!scripts_dir)
+		return NULL;
+
+	for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next) {
+		snprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path,
+			 lang_dirent.d_name);
+		lang_dir = opendir(lang_path);
+		if (!lang_dir)
+			continue;
+
+		for_each_script(lang_path, lang_dir, script_dirent, script_next) {
+			__script_root = get_script_root(&script_dirent, suffix);
+			if (__script_root && !strcmp(script_root, __script_root)) {
+				free(__script_root);
+				closedir(lang_dir);
+				closedir(scripts_dir);
+				snprintf(script_path, MAXPATHLEN, "%s/%s",
+					 lang_path, script_dirent.d_name);
+				return strdup(script_path);
+			}
+			free(__script_root);
+		}
+		closedir(lang_dir);
+	}
+	closedir(scripts_dir);
+
+	return NULL;
+}
+
+static bool is_top_script(const char *script_path)
+{
+	return ends_with(script_path, "top") == NULL ? false : true;
+}
+
+static int has_required_arg(char *script_path)
+{
+	struct script_desc *desc;
+	int n_args = 0;
+	char *p;
+
+	desc = script_desc__new(NULL);
+
+	if (read_script_info(desc, script_path))
+		goto out;
+
+	if (!desc->args)
+		goto out;
+
+	for (p = desc->args; *p; p++)
+		if (*p == '<')
+			n_args++;
+out:
+	script_desc__delete(desc);
+
+	return n_args;
+}
+
+static int have_cmd(int argc, const char **argv)
+{
+	char **__argv = malloc(sizeof(const char *) * argc);
+
+	if (!__argv) {
+		pr_err("malloc failed\n");
+		return -1;
+	}
+
+	memcpy(__argv, argv, sizeof(const char *) * argc);
+	argc = parse_options(argc, (const char **)__argv, record_options,
+			     NULL, PARSE_OPT_STOP_AT_NON_OPTION);
+	free(__argv);
+
+	system_wide = (argc == 0);
+
+	return 0;
+}
+
+int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	bool show_full_info = false;
+	bool header = false;
+	bool header_only = false;
+	bool script_started = false;
+	char *rec_script_path = NULL;
+	char *rep_script_path = NULL;
+	struct perf_session *session;
+	char *script_path = NULL;
+	const char **__argv;
+	int i, j, err = 0;
+	struct perf_script script = {
+		.tool = {
+			.sample		 = process_sample_event,
+			.mmap		 = perf_event__process_mmap,
+			.mmap2		 = perf_event__process_mmap2,
+			.comm		 = perf_event__process_comm,
+			.exit		 = perf_event__process_exit,
+			.fork		 = perf_event__process_fork,
+			.attr		 = process_attr,
+			.tracing_data	 = perf_event__process_tracing_data,
+			.build_id	 = perf_event__process_build_id,
+			.ordered_events	 = true,
+			.ordering_requires_timestamps = true,
+		},
+	};
+	struct perf_data_file file = {
+		.mode = PERF_DATA_MODE_READ,
+	};
+	const struct option options[] = {
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_INCR('v', "verbose", &verbose,
+		 "be more verbose (show symbol address, etc)"),
+	OPT_BOOLEAN('L', "Latency", &latency_format,
+		    "show latency attributes (irqs/preemption disabled, etc)"),
+	OPT_CALLBACK_NOOPT('l', "list", NULL, NULL, "list available scripts",
+			   list_available_scripts),
+	OPT_CALLBACK('s', "script", NULL, "name",
+		     "script file name (lang:script name, script name, or *)",
+		     parse_scriptname),
+	OPT_STRING('g', "gen-script", &generate_script_lang, "lang",
+		   "generate perf-script.xx script in specified language"),
+	OPT_STRING('i', "input", &input_name, "file", "input file name"),
+	OPT_BOOLEAN('d', "debug-mode", &debug_mode,
+		   "do various checks like samples ordering and lost events"),
+	OPT_BOOLEAN(0, "header", &header, "Show data header."),
+	OPT_BOOLEAN(0, "header-only", &header_only, "Show only data header."),
+	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+		   "file", "vmlinux pathname"),
+	OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name,
+		   "file", "kallsyms pathname"),
+	OPT_BOOLEAN('G', "hide-call-graph", &no_callchain,
+		    "When printing symbols do not display call chain"),
+	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
+		    "Look for files with symbols relative to this directory"),
+	OPT_CALLBACK('F', "fields", NULL, "str",
+		     "comma separated output fields prepend with 'type:'. "
+		     "Valid types: hw,sw,trace,raw. "
+		     "Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso,"
+		     "addr,symoff,period", parse_output_fields),
+	OPT_BOOLEAN('a', "all-cpus", &system_wide,
+		    "system-wide collection from all CPUs"),
+	OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]",
+		   "only consider these symbols"),
+	OPT_STRING('C', "cpu", &cpu_list, "cpu", "list of cpus to profile"),
+	OPT_STRING('c', "comms", &symbol_conf.comm_list_str, "comm[,comm...]",
+		   "only display events for these comms"),
+	OPT_STRING(0, "pid", &symbol_conf.pid_list_str, "pid[,pid...]",
+		   "only consider symbols in these pids"),
+	OPT_STRING(0, "tid", &symbol_conf.tid_list_str, "tid[,tid...]",
+		   "only consider symbols in these tids"),
+	OPT_BOOLEAN('I', "show-info", &show_full_info,
+		    "display extended information from perf.data file"),
+	OPT_BOOLEAN('\0', "show-kernel-path", &symbol_conf.show_kernel_path,
+		    "Show the path of [kernel.kallsyms]"),
+	OPT_BOOLEAN('\0', "show-task-events", &script.show_task_events,
+		    "Show the fork/comm/exit events"),
+	OPT_BOOLEAN('\0', "show-mmap-events", &script.show_mmap_events,
+		    "Show the mmap events"),
+	OPT_BOOLEAN('f', "force", &file.force, "don't complain, do it"),
+	OPT_END()
+	};
+	const char * const script_subcommands[] = { "record", "report", NULL };
+	const char *script_usage[] = {
+		"perf script [<options>]",
+		"perf script [<options>] record <script> [<record-options>] <command>",
+		"perf script [<options>] report <script> [script-args]",
+		"perf script [<options>] <script> [<record-options>] <command>",
+		"perf script [<options>] <top-script> [script-args]",
+		NULL
+	};
+
+	setup_scripting();
+
+	argc = parse_options_subcommand(argc, argv, options, script_subcommands, script_usage,
+			     PARSE_OPT_STOP_AT_NON_OPTION);
+
+	file.path = input_name;
+
+	if (argc > 1 && !strncmp(argv[0], "rec", strlen("rec"))) {
+		rec_script_path = get_script_path(argv[1], RECORD_SUFFIX);
+		if (!rec_script_path)
+			return cmd_record(argc, argv, NULL);
+	}
+
+	if (argc > 1 && !strncmp(argv[0], "rep", strlen("rep"))) {
+		rep_script_path = get_script_path(argv[1], REPORT_SUFFIX);
+		if (!rep_script_path) {
+			fprintf(stderr,
+				"Please specify a valid report script"
+				"(see 'perf script -l' for listing)\n");
+			return -1;
+		}
+	}
+
+	/* make sure PERF_EXEC_PATH is set for scripts */
+	perf_set_argv_exec_path(perf_exec_path());
+
+	if (argc && !script_name && !rec_script_path && !rep_script_path) {
+		int live_pipe[2];
+		int rep_args;
+		pid_t pid;
+
+		rec_script_path = get_script_path(argv[0], RECORD_SUFFIX);
+		rep_script_path = get_script_path(argv[0], REPORT_SUFFIX);
+
+		if (!rec_script_path && !rep_script_path) {
+			fprintf(stderr, " Couldn't find script %s\n\n See perf"
+				" script -l for available scripts.\n", argv[0]);
+			usage_with_options(script_usage, options);
+		}
+
+		if (is_top_script(argv[0])) {
+			rep_args = argc - 1;
+		} else {
+			int rec_args;
+
+			rep_args = has_required_arg(rep_script_path);
+			rec_args = (argc - 1) - rep_args;
+			if (rec_args < 0) {
+				fprintf(stderr, " %s script requires options."
+					"\n\n See perf script -l for available "
+					"scripts and options.\n", argv[0]);
+				usage_with_options(script_usage, options);
+			}
+		}
+
+		if (pipe(live_pipe) < 0) {
+			perror("failed to create pipe");
+			return -1;
+		}
+
+		pid = fork();
+		if (pid < 0) {
+			perror("failed to fork");
+			return -1;
+		}
+
+		if (!pid) {
+			j = 0;
+
+			dup2(live_pipe[1], 1);
+			close(live_pipe[0]);
+
+			if (is_top_script(argv[0])) {
+				system_wide = true;
+			} else if (!system_wide) {
+				if (have_cmd(argc - rep_args, &argv[rep_args]) != 0) {
+					err = -1;
+					goto out;
+				}
+			}
+
+			__argv = malloc((argc + 6) * sizeof(const char *));
+			if (!__argv) {
+				pr_err("malloc failed\n");
+				err = -ENOMEM;
+				goto out;
+			}
+
+			__argv[j++] = "/bin/sh";
+			__argv[j++] = rec_script_path;
+			if (system_wide)
+				__argv[j++] = "-a";
+			__argv[j++] = "-q";
+			__argv[j++] = "-o";
+			__argv[j++] = "-";
+			for (i = rep_args + 1; i < argc; i++)
+				__argv[j++] = argv[i];
+			__argv[j++] = NULL;
+
+			execvp("/bin/sh", (char **)__argv);
+			free(__argv);
+			exit(-1);
+		}
+
+		dup2(live_pipe[0], 0);
+		close(live_pipe[1]);
+
+		__argv = malloc((argc + 4) * sizeof(const char *));
+		if (!__argv) {
+			pr_err("malloc failed\n");
+			err = -ENOMEM;
+			goto out;
+		}
+
+		j = 0;
+		__argv[j++] = "/bin/sh";
+		__argv[j++] = rep_script_path;
+		for (i = 1; i < rep_args + 1; i++)
+			__argv[j++] = argv[i];
+		__argv[j++] = "-i";
+		__argv[j++] = "-";
+		__argv[j++] = NULL;
+
+		execvp("/bin/sh", (char **)__argv);
+		free(__argv);
+		exit(-1);
+	}
+
+	if (rec_script_path)
+		script_path = rec_script_path;
+	if (rep_script_path)
+		script_path = rep_script_path;
+
+	if (script_path) {
+		j = 0;
+
+		if (!rec_script_path)
+			system_wide = false;
+		else if (!system_wide) {
+			if (have_cmd(argc - 1, &argv[1]) != 0) {
+				err = -1;
+				goto out;
+			}
+		}
+
+		__argv = malloc((argc + 2) * sizeof(const char *));
+		if (!__argv) {
+			pr_err("malloc failed\n");
+			err = -ENOMEM;
+			goto out;
+		}
+
+		__argv[j++] = "/bin/sh";
+		__argv[j++] = script_path;
+		if (system_wide)
+			__argv[j++] = "-a";
+		for (i = 2; i < argc; i++)
+			__argv[j++] = argv[i];
+		__argv[j++] = NULL;
+
+		execvp("/bin/sh", (char **)__argv);
+		free(__argv);
+		exit(-1);
+	}
+
+	if (!script_name)
+		setup_pager();
+
+	session = perf_session__new(&file, false, &script.tool);
+	if (session == NULL)
+		return -1;
+
+	if (header || header_only) {
+		perf_session__fprintf_info(session, stdout, show_full_info);
+		if (header_only)
+			goto out_delete;
+	}
+
+	if (symbol__init(&session->header.env) < 0)
+		goto out_delete;
+
+	script.session = session;
+
+	if (cpu_list) {
+		err = perf_session__cpu_bitmap(session, cpu_list, cpu_bitmap);
+		if (err < 0)
+			goto out_delete;
+	}
+
+	if (!no_callchain)
+		symbol_conf.use_callchain = true;
+	else
+		symbol_conf.use_callchain = false;
+
+	if (generate_script_lang) {
+		struct stat perf_stat;
+		int input;
+
+		if (output_set_by_user()) {
+			fprintf(stderr,
+				"custom fields not supported for generated scripts");
+			err = -EINVAL;
+			goto out_delete;
+		}
+
+		input = open(file.path, O_RDONLY);	/* input_name */
+		if (input < 0) {
+			err = -errno;
+			perror("failed to open file");
+			goto out_delete;
+		}
+
+		err = fstat(input, &perf_stat);
+		if (err < 0) {
+			perror("failed to stat file");
+			goto out_delete;
+		}
+
+		if (!perf_stat.st_size) {
+			fprintf(stderr, "zero-sized file, nothing to do!\n");
+			goto out_delete;
+		}
+
+		scripting_ops = script_spec__lookup(generate_script_lang);
+		if (!scripting_ops) {
+			fprintf(stderr, "invalid language specifier");
+			err = -ENOENT;
+			goto out_delete;
+		}
+
+		err = scripting_ops->generate_script(session->tevent.pevent,
+						     "perf-script");
+		goto out_delete;
+	}
+
+	if (script_name) {
+		err = scripting_ops->start_script(script_name, argc, argv);
+		if (err)
+			goto out_delete;
+		pr_debug("perf script started with script %s\n\n", script_name);
+		script_started = true;
+	}
+
+
+	err = perf_session__check_output_opt(session);
+	if (err < 0)
+		goto out_delete;
+
+	err = __cmd_script(&script);
+
+	flush_scripting();
+
+out_delete:
+	perf_session__delete(session);
+
+	if (script_started)
+		cleanup_scripting();
+out:
+	return err;
+}
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
new file mode 100644
index 000000000..f7b821878
--- /dev/null
+++ b/tools/perf/builtin-stat.c
@@ -0,0 +1,1935 @@
+/*
+ * builtin-stat.c
+ *
+ * Builtin stat command: Give a precise performance counters summary
+ * overview about any workload, CPU or specific PID.
+ *
+ * Sample output:
+
+   $ perf stat ./hackbench 10
+
+  Time: 0.118
+
+  Performance counter stats for './hackbench 10':
+
+       1708.761321 task-clock                #   11.037 CPUs utilized
+            41,190 context-switches          #    0.024 M/sec
+             6,735 CPU-migrations            #    0.004 M/sec
+            17,318 page-faults               #    0.010 M/sec
+     5,205,202,243 cycles                    #    3.046 GHz
+     3,856,436,920 stalled-cycles-frontend   #   74.09% frontend cycles idle
+     1,600,790,871 stalled-cycles-backend    #   30.75% backend  cycles idle
+     2,603,501,247 instructions              #    0.50  insns per cycle
+                                             #    1.48  stalled cycles per insn
+       484,357,498 branches                  #  283.455 M/sec
+         6,388,934 branch-misses             #    1.32% of all branches
+
+        0.154822978  seconds time elapsed
+
+ *
+ * Copyright (C) 2008-2011, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
+ *
+ * Improvements and fixes by:
+ *
+ *   Arjan van de Ven <arjan@linux.intel.com>
+ *   Yanmin Zhang <yanmin.zhang@intel.com>
+ *   Wu Fengguang <fengguang.wu@intel.com>
+ *   Mike Galbraith <efault@gmx.de>
+ *   Paul Mackerras <paulus@samba.org>
+ *   Jaswinder Singh Rajput <jaswinder@kernel.org>
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+
+#include "perf.h"
+#include "builtin.h"
+#include "util/cgroup.h"
+#include "util/util.h"
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+#include "util/pmu.h"
+#include "util/event.h"
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/debug.h"
+#include "util/color.h"
+#include "util/stat.h"
+#include "util/header.h"
+#include "util/cpumap.h"
+#include "util/thread.h"
+#include "util/thread_map.h"
+
+#include <stdlib.h>
+#include <sys/prctl.h>
+#include <locale.h>
+
+#define DEFAULT_SEPARATOR	" "
+#define CNTR_NOT_SUPPORTED	"<not supported>"
+#define CNTR_NOT_COUNTED	"<not counted>"
+
+static void print_stat(int argc, const char **argv);
+static void print_counter_aggr(struct perf_evsel *counter, char *prefix);
+static void print_counter(struct perf_evsel *counter, char *prefix);
+static void print_aggr(char *prefix);
+
+/* Default events used for perf stat -T */
+static const char * const transaction_attrs[] = {
+	"task-clock",
+	"{"
+	"instructions,"
+	"cycles,"
+	"cpu/cycles-t/,"
+	"cpu/tx-start/,"
+	"cpu/el-start/,"
+	"cpu/cycles-ct/"
+	"}"
+};
+
+/* More limited version when the CPU does not have all events. */
+static const char * const transaction_limited_attrs[] = {
+	"task-clock",
+	"{"
+	"instructions,"
+	"cycles,"
+	"cpu/cycles-t/,"
+	"cpu/tx-start/"
+	"}"
+};
+
+/* must match transaction_attrs and the beginning limited_attrs */
+enum {
+	T_TASK_CLOCK,
+	T_INSTRUCTIONS,
+	T_CYCLES,
+	T_CYCLES_IN_TX,
+	T_TRANSACTION_START,
+	T_ELISION_START,
+	T_CYCLES_IN_TX_CP,
+};
+
+static struct perf_evlist	*evsel_list;
+
+static struct target target = {
+	.uid	= UINT_MAX,
+};
+
+enum aggr_mode {
+	AGGR_NONE,
+	AGGR_GLOBAL,
+	AGGR_SOCKET,
+	AGGR_CORE,
+};
+
+static int			run_count			=  1;
+static bool			no_inherit			= false;
+static bool			scale				=  true;
+static enum aggr_mode		aggr_mode			= AGGR_GLOBAL;
+static volatile pid_t		child_pid			= -1;
+static bool			null_run			=  false;
+static int			detailed_run			=  0;
+static bool			transaction_run;
+static bool			big_num				=  true;
+static int			big_num_opt			=  -1;
+static const char		*csv_sep			= NULL;
+static bool			csv_output			= false;
+static bool			group				= false;
+static FILE			*output				= NULL;
+static const char		*pre_cmd			= NULL;
+static const char		*post_cmd			= NULL;
+static bool			sync_run			= false;
+static unsigned int		interval			= 0;
+static unsigned int		initial_delay			= 0;
+static unsigned int		unit_width			= 4; /* strlen("unit") */
+static bool			forever				= false;
+static struct timespec		ref_time;
+static struct cpu_map		*aggr_map;
+static int			(*aggr_get_id)(struct cpu_map *m, int cpu);
+
+static volatile int done = 0;
+
+struct perf_stat {
+	struct stats	  res_stats[3];
+};
+
+static inline void diff_timespec(struct timespec *r, struct timespec *a,
+				 struct timespec *b)
+{
+	r->tv_sec = a->tv_sec - b->tv_sec;
+	if (a->tv_nsec < b->tv_nsec) {
+		r->tv_nsec = a->tv_nsec + 1000000000L - b->tv_nsec;
+		r->tv_sec--;
+	} else {
+		r->tv_nsec = a->tv_nsec - b->tv_nsec ;
+	}
+}
+
+static inline struct cpu_map *perf_evsel__cpus(struct perf_evsel *evsel)
+{
+	return (evsel->cpus && !target.cpu_list) ? evsel->cpus : evsel_list->cpus;
+}
+
+static inline int perf_evsel__nr_cpus(struct perf_evsel *evsel)
+{
+	return perf_evsel__cpus(evsel)->nr;
+}
+
+static void perf_evsel__reset_stat_priv(struct perf_evsel *evsel)
+{
+	int i;
+	struct perf_stat *ps = evsel->priv;
+
+	for (i = 0; i < 3; i++)
+		init_stats(&ps->res_stats[i]);
+}
+
+static int perf_evsel__alloc_stat_priv(struct perf_evsel *evsel)
+{
+	evsel->priv = zalloc(sizeof(struct perf_stat));
+	if (evsel->priv == NULL)
+		return -ENOMEM;
+	perf_evsel__reset_stat_priv(evsel);
+	return 0;
+}
+
+static void perf_evsel__free_stat_priv(struct perf_evsel *evsel)
+{
+	zfree(&evsel->priv);
+}
+
+static int perf_evsel__alloc_prev_raw_counts(struct perf_evsel *evsel)
+{
+	void *addr;
+	size_t sz;
+
+	sz = sizeof(*evsel->counts) +
+	     (perf_evsel__nr_cpus(evsel) * sizeof(struct perf_counts_values));
+
+	addr = zalloc(sz);
+	if (!addr)
+		return -ENOMEM;
+
+	evsel->prev_raw_counts =  addr;
+
+	return 0;
+}
+
+static void perf_evsel__free_prev_raw_counts(struct perf_evsel *evsel)
+{
+	zfree(&evsel->prev_raw_counts);
+}
+
+static void perf_evlist__free_stats(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel) {
+		perf_evsel__free_stat_priv(evsel);
+		perf_evsel__free_counts(evsel);
+		perf_evsel__free_prev_raw_counts(evsel);
+	}
+}
+
+static int perf_evlist__alloc_stats(struct perf_evlist *evlist, bool alloc_raw)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel) {
+		if (perf_evsel__alloc_stat_priv(evsel) < 0 ||
+		    perf_evsel__alloc_counts(evsel, perf_evsel__nr_cpus(evsel)) < 0 ||
+		    (alloc_raw && perf_evsel__alloc_prev_raw_counts(evsel) < 0))
+			goto out_free;
+	}
+
+	return 0;
+
+out_free:
+	perf_evlist__free_stats(evlist);
+	return -1;
+}
+
+static struct stats runtime_nsecs_stats[MAX_NR_CPUS];
+static struct stats runtime_cycles_stats[MAX_NR_CPUS];
+static struct stats runtime_stalled_cycles_front_stats[MAX_NR_CPUS];
+static struct stats runtime_stalled_cycles_back_stats[MAX_NR_CPUS];
+static struct stats runtime_branches_stats[MAX_NR_CPUS];
+static struct stats runtime_cacherefs_stats[MAX_NR_CPUS];
+static struct stats runtime_l1_dcache_stats[MAX_NR_CPUS];
+static struct stats runtime_l1_icache_stats[MAX_NR_CPUS];
+static struct stats runtime_ll_cache_stats[MAX_NR_CPUS];
+static struct stats runtime_itlb_cache_stats[MAX_NR_CPUS];
+static struct stats runtime_dtlb_cache_stats[MAX_NR_CPUS];
+static struct stats runtime_cycles_in_tx_stats[MAX_NR_CPUS];
+static struct stats walltime_nsecs_stats;
+static struct stats runtime_transaction_stats[MAX_NR_CPUS];
+static struct stats runtime_elision_stats[MAX_NR_CPUS];
+
+static void perf_stat__reset_stats(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel) {
+		perf_evsel__reset_stat_priv(evsel);
+		perf_evsel__reset_counts(evsel, perf_evsel__nr_cpus(evsel));
+	}
+
+	memset(runtime_nsecs_stats, 0, sizeof(runtime_nsecs_stats));
+	memset(runtime_cycles_stats, 0, sizeof(runtime_cycles_stats));
+	memset(runtime_stalled_cycles_front_stats, 0, sizeof(runtime_stalled_cycles_front_stats));
+	memset(runtime_stalled_cycles_back_stats, 0, sizeof(runtime_stalled_cycles_back_stats));
+	memset(runtime_branches_stats, 0, sizeof(runtime_branches_stats));
+	memset(runtime_cacherefs_stats, 0, sizeof(runtime_cacherefs_stats));
+	memset(runtime_l1_dcache_stats, 0, sizeof(runtime_l1_dcache_stats));
+	memset(runtime_l1_icache_stats, 0, sizeof(runtime_l1_icache_stats));
+	memset(runtime_ll_cache_stats, 0, sizeof(runtime_ll_cache_stats));
+	memset(runtime_itlb_cache_stats, 0, sizeof(runtime_itlb_cache_stats));
+	memset(runtime_dtlb_cache_stats, 0, sizeof(runtime_dtlb_cache_stats));
+	memset(runtime_cycles_in_tx_stats, 0,
+			sizeof(runtime_cycles_in_tx_stats));
+	memset(runtime_transaction_stats, 0,
+		sizeof(runtime_transaction_stats));
+	memset(runtime_elision_stats, 0, sizeof(runtime_elision_stats));
+	memset(&walltime_nsecs_stats, 0, sizeof(walltime_nsecs_stats));
+}
+
+static int create_perf_stat_counter(struct perf_evsel *evsel)
+{
+	struct perf_event_attr *attr = &evsel->attr;
+
+	if (scale)
+		attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED |
+				    PERF_FORMAT_TOTAL_TIME_RUNNING;
+
+	attr->inherit = !no_inherit;
+
+	if (target__has_cpu(&target))
+		return perf_evsel__open_per_cpu(evsel, perf_evsel__cpus(evsel));
+
+	if (!target__has_task(&target) && perf_evsel__is_group_leader(evsel)) {
+		attr->disabled = 1;
+		if (!initial_delay)
+			attr->enable_on_exec = 1;
+	}
+
+	return perf_evsel__open_per_thread(evsel, evsel_list->threads);
+}
+
+/*
+ * Does the counter have nsecs as a unit?
+ */
+static inline int nsec_counter(struct perf_evsel *evsel)
+{
+	if (perf_evsel__match(evsel, SOFTWARE, SW_CPU_CLOCK) ||
+	    perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK))
+		return 1;
+
+	return 0;
+}
+
+static struct perf_evsel *nth_evsel(int n)
+{
+	static struct perf_evsel **array;
+	static int array_len;
+	struct perf_evsel *ev;
+	int j;
+
+	/* Assumes this only called when evsel_list does not change anymore. */
+	if (!array) {
+		evlist__for_each(evsel_list, ev)
+			array_len++;
+		array = malloc(array_len * sizeof(void *));
+		if (!array)
+			exit(ENOMEM);
+		j = 0;
+		evlist__for_each(evsel_list, ev)
+			array[j++] = ev;
+	}
+	if (n < array_len)
+		return array[n];
+	return NULL;
+}
+
+/*
+ * Update various tracking values we maintain to print
+ * more semantic information such as miss/hit ratios,
+ * instruction rates, etc:
+ */
+static void update_shadow_stats(struct perf_evsel *counter, u64 *count,
+				int cpu)
+{
+	if (perf_evsel__match(counter, SOFTWARE, SW_TASK_CLOCK))
+		update_stats(&runtime_nsecs_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HARDWARE, HW_CPU_CYCLES))
+		update_stats(&runtime_cycles_stats[cpu], count[0]);
+	else if (transaction_run &&
+		 perf_evsel__cmp(counter, nth_evsel(T_CYCLES_IN_TX)))
+		update_stats(&runtime_cycles_in_tx_stats[cpu], count[0]);
+	else if (transaction_run &&
+		 perf_evsel__cmp(counter, nth_evsel(T_TRANSACTION_START)))
+		update_stats(&runtime_transaction_stats[cpu], count[0]);
+	else if (transaction_run &&
+		 perf_evsel__cmp(counter, nth_evsel(T_ELISION_START)))
+		update_stats(&runtime_elision_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HARDWARE, HW_STALLED_CYCLES_FRONTEND))
+		update_stats(&runtime_stalled_cycles_front_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HARDWARE, HW_STALLED_CYCLES_BACKEND))
+		update_stats(&runtime_stalled_cycles_back_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HARDWARE, HW_BRANCH_INSTRUCTIONS))
+		update_stats(&runtime_branches_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HARDWARE, HW_CACHE_REFERENCES))
+		update_stats(&runtime_cacherefs_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_L1D))
+		update_stats(&runtime_l1_dcache_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_L1I))
+		update_stats(&runtime_l1_icache_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_LL))
+		update_stats(&runtime_ll_cache_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_DTLB))
+		update_stats(&runtime_dtlb_cache_stats[cpu], count[0]);
+	else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_ITLB))
+		update_stats(&runtime_itlb_cache_stats[cpu], count[0]);
+}
+
+static void zero_per_pkg(struct perf_evsel *counter)
+{
+	if (counter->per_pkg_mask)
+		memset(counter->per_pkg_mask, 0, MAX_NR_CPUS);
+}
+
+static int check_per_pkg(struct perf_evsel *counter, int cpu, bool *skip)
+{
+	unsigned long *mask = counter->per_pkg_mask;
+	struct cpu_map *cpus = perf_evsel__cpus(counter);
+	int s;
+
+	*skip = false;
+
+	if (!counter->per_pkg)
+		return 0;
+
+	if (cpu_map__empty(cpus))
+		return 0;
+
+	if (!mask) {
+		mask = zalloc(MAX_NR_CPUS);
+		if (!mask)
+			return -ENOMEM;
+
+		counter->per_pkg_mask = mask;
+	}
+
+	s = cpu_map__get_socket(cpus, cpu);
+	if (s < 0)
+		return -1;
+
+	*skip = test_and_set_bit(s, mask) == 1;
+	return 0;
+}
+
+static int read_cb(struct perf_evsel *evsel, int cpu, int thread __maybe_unused,
+		   struct perf_counts_values *count)
+{
+	struct perf_counts_values *aggr = &evsel->counts->aggr;
+	static struct perf_counts_values zero;
+	bool skip = false;
+
+	if (check_per_pkg(evsel, cpu, &skip)) {
+		pr_err("failed to read per-pkg counter\n");
+		return -1;
+	}
+
+	if (skip)
+		count = &zero;
+
+	switch (aggr_mode) {
+	case AGGR_CORE:
+	case AGGR_SOCKET:
+	case AGGR_NONE:
+		if (!evsel->snapshot)
+			perf_evsel__compute_deltas(evsel, cpu, count);
+		perf_counts_values__scale(count, scale, NULL);
+		evsel->counts->cpu[cpu] = *count;
+		if (aggr_mode == AGGR_NONE)
+			update_shadow_stats(evsel, count->values, cpu);
+		break;
+	case AGGR_GLOBAL:
+		aggr->val += count->val;
+		if (scale) {
+			aggr->ena += count->ena;
+			aggr->run += count->run;
+		}
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int read_counter(struct perf_evsel *counter);
+
+/*
+ * Read out the results of a single counter:
+ * aggregate counts across CPUs in system-wide mode
+ */
+static int read_counter_aggr(struct perf_evsel *counter)
+{
+	struct perf_counts_values *aggr = &counter->counts->aggr;
+	struct perf_stat *ps = counter->priv;
+	u64 *count = counter->counts->aggr.values;
+	int i;
+
+	aggr->val = aggr->ena = aggr->run = 0;
+
+	if (read_counter(counter))
+		return -1;
+
+	if (!counter->snapshot)
+		perf_evsel__compute_deltas(counter, -1, aggr);
+	perf_counts_values__scale(aggr, scale, &counter->counts->scaled);
+
+	for (i = 0; i < 3; i++)
+		update_stats(&ps->res_stats[i], count[i]);
+
+	if (verbose) {
+		fprintf(output, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
+			perf_evsel__name(counter), count[0], count[1], count[2]);
+	}
+
+	/*
+	 * Save the full runtime - to allow normalization during printout:
+	 */
+	update_shadow_stats(counter, count, 0);
+
+	return 0;
+}
+
+/*
+ * Read out the results of a single counter:
+ * do not aggregate counts across CPUs in system-wide mode
+ */
+static int read_counter(struct perf_evsel *counter)
+{
+	int nthreads = thread_map__nr(evsel_list->threads);
+	int ncpus = perf_evsel__nr_cpus(counter);
+	int cpu, thread;
+
+	if (!counter->supported)
+		return -ENOENT;
+
+	if (counter->system_wide)
+		nthreads = 1;
+
+	if (counter->per_pkg)
+		zero_per_pkg(counter);
+
+	for (thread = 0; thread < nthreads; thread++) {
+		for (cpu = 0; cpu < ncpus; cpu++) {
+			if (perf_evsel__read_cb(counter, cpu, thread, read_cb))
+				return -1;
+		}
+	}
+
+	return 0;
+}
+
+static void print_interval(void)
+{
+	static int num_print_interval;
+	struct perf_evsel *counter;
+	struct perf_stat *ps;
+	struct timespec ts, rs;
+	char prefix[64];
+
+	if (aggr_mode == AGGR_GLOBAL) {
+		evlist__for_each(evsel_list, counter) {
+			ps = counter->priv;
+			memset(ps->res_stats, 0, sizeof(ps->res_stats));
+			read_counter_aggr(counter);
+		}
+	} else	{
+		evlist__for_each(evsel_list, counter) {
+			ps = counter->priv;
+			memset(ps->res_stats, 0, sizeof(ps->res_stats));
+			read_counter(counter);
+		}
+	}
+
+	clock_gettime(CLOCK_MONOTONIC, &ts);
+	diff_timespec(&rs, &ts, &ref_time);
+	sprintf(prefix, "%6lu.%09lu%s", rs.tv_sec, rs.tv_nsec, csv_sep);
+
+	if (num_print_interval == 0 && !csv_output) {
+		switch (aggr_mode) {
+		case AGGR_SOCKET:
+			fprintf(output, "#           time socket cpus             counts %*s events\n", unit_width, "unit");
+			break;
+		case AGGR_CORE:
+			fprintf(output, "#           time core         cpus             counts %*s events\n", unit_width, "unit");
+			break;
+		case AGGR_NONE:
+			fprintf(output, "#           time CPU                counts %*s events\n", unit_width, "unit");
+			break;
+		case AGGR_GLOBAL:
+		default:
+			fprintf(output, "#           time             counts %*s events\n", unit_width, "unit");
+		}
+	}
+
+	if (++num_print_interval == 25)
+		num_print_interval = 0;
+
+	switch (aggr_mode) {
+	case AGGR_CORE:
+	case AGGR_SOCKET:
+		print_aggr(prefix);
+		break;
+	case AGGR_NONE:
+		evlist__for_each(evsel_list, counter)
+			print_counter(counter, prefix);
+		break;
+	case AGGR_GLOBAL:
+	default:
+		evlist__for_each(evsel_list, counter)
+			print_counter_aggr(counter, prefix);
+	}
+
+	fflush(output);
+}
+
+static void handle_initial_delay(void)
+{
+	struct perf_evsel *counter;
+
+	if (initial_delay) {
+		const int ncpus = cpu_map__nr(evsel_list->cpus),
+			nthreads = thread_map__nr(evsel_list->threads);
+
+		usleep(initial_delay * 1000);
+		evlist__for_each(evsel_list, counter)
+			perf_evsel__enable(counter, ncpus, nthreads);
+	}
+}
+
+static volatile int workload_exec_errno;
+
+/*
+ * perf_evlist__prepare_workload will send a SIGUSR1
+ * if the fork fails, since we asked by setting its
+ * want_signal to true.
+ */
+static void workload_exec_failed_signal(int signo __maybe_unused, siginfo_t *info,
+					void *ucontext __maybe_unused)
+{
+	workload_exec_errno = info->si_value.sival_int;
+}
+
+static int __run_perf_stat(int argc, const char **argv)
+{
+	char msg[512];
+	unsigned long long t0, t1;
+	struct perf_evsel *counter;
+	struct timespec ts;
+	size_t l;
+	int status = 0;
+	const bool forks = (argc > 0);
+
+	if (interval) {
+		ts.tv_sec  = interval / 1000;
+		ts.tv_nsec = (interval % 1000) * 1000000;
+	} else {
+		ts.tv_sec  = 1;
+		ts.tv_nsec = 0;
+	}
+
+	if (forks) {
+		if (perf_evlist__prepare_workload(evsel_list, &target, argv, false,
+						  workload_exec_failed_signal) < 0) {
+			perror("failed to prepare workload");
+			return -1;
+		}
+		child_pid = evsel_list->workload.pid;
+	}
+
+	if (group)
+		perf_evlist__set_leader(evsel_list);
+
+	evlist__for_each(evsel_list, counter) {
+		if (create_perf_stat_counter(counter) < 0) {
+			/*
+			 * PPC returns ENXIO for HW counters until 2.6.37
+			 * (behavior changed with commit b0a873e).
+			 */
+			if (errno == EINVAL || errno == ENOSYS ||
+			    errno == ENOENT || errno == EOPNOTSUPP ||
+			    errno == ENXIO) {
+				if (verbose)
+					ui__warning("%s event is not supported by the kernel.\n",
+						    perf_evsel__name(counter));
+				counter->supported = false;
+				continue;
+			}
+
+			perf_evsel__open_strerror(counter, &target,
+						  errno, msg, sizeof(msg));
+			ui__error("%s\n", msg);
+
+			if (child_pid != -1)
+				kill(child_pid, SIGTERM);
+
+			return -1;
+		}
+		counter->supported = true;
+
+		l = strlen(counter->unit);
+		if (l > unit_width)
+			unit_width = l;
+	}
+
+	if (perf_evlist__apply_filters(evsel_list, &counter)) {
+		error("failed to set filter \"%s\" on event %s with %d (%s)\n",
+			counter->filter, perf_evsel__name(counter), errno,
+			strerror_r(errno, msg, sizeof(msg)));
+		return -1;
+	}
+
+	/*
+	 * Enable counters and exec the command:
+	 */
+	t0 = rdclock();
+	clock_gettime(CLOCK_MONOTONIC, &ref_time);
+
+	if (forks) {
+		perf_evlist__start_workload(evsel_list);
+		handle_initial_delay();
+
+		if (interval) {
+			while (!waitpid(child_pid, &status, WNOHANG)) {
+				nanosleep(&ts, NULL);
+				print_interval();
+			}
+		}
+		wait(&status);
+
+		if (workload_exec_errno) {
+			const char *emsg = strerror_r(workload_exec_errno, msg, sizeof(msg));
+			pr_err("Workload failed: %s\n", emsg);
+			return -1;
+		}
+
+		if (WIFSIGNALED(status))
+			psignal(WTERMSIG(status), argv[0]);
+	} else {
+		handle_initial_delay();
+		while (!done) {
+			nanosleep(&ts, NULL);
+			if (interval)
+				print_interval();
+		}
+	}
+
+	t1 = rdclock();
+
+	update_stats(&walltime_nsecs_stats, t1 - t0);
+
+	if (aggr_mode == AGGR_GLOBAL) {
+		evlist__for_each(evsel_list, counter) {
+			read_counter_aggr(counter);
+			perf_evsel__close_fd(counter, perf_evsel__nr_cpus(counter),
+					     thread_map__nr(evsel_list->threads));
+		}
+	} else {
+		evlist__for_each(evsel_list, counter) {
+			read_counter(counter);
+			perf_evsel__close_fd(counter, perf_evsel__nr_cpus(counter), 1);
+		}
+	}
+
+	return WEXITSTATUS(status);
+}
+
+static int run_perf_stat(int argc, const char **argv)
+{
+	int ret;
+
+	if (pre_cmd) {
+		ret = system(pre_cmd);
+		if (ret)
+			return ret;
+	}
+
+	if (sync_run)
+		sync();
+
+	ret = __run_perf_stat(argc, argv);
+	if (ret)
+		return ret;
+
+	if (post_cmd) {
+		ret = system(post_cmd);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
+static void print_running(u64 run, u64 ena)
+{
+	if (csv_output) {
+		fprintf(output, "%s%" PRIu64 "%s%.2f",
+					csv_sep,
+					run,
+					csv_sep,
+					ena ? 100.0 * run / ena : 100.0);
+	} else if (run != ena) {
+		fprintf(output, "  (%.2f%%)", 100.0 * run / ena);
+	}
+}
+
+static void print_noise_pct(double total, double avg)
+{
+	double pct = rel_stddev_stats(total, avg);
+
+	if (csv_output)
+		fprintf(output, "%s%.2f%%", csv_sep, pct);
+	else if (pct)
+		fprintf(output, "  ( +-%6.2f%% )", pct);
+}
+
+static void print_noise(struct perf_evsel *evsel, double avg)
+{
+	struct perf_stat *ps;
+
+	if (run_count == 1)
+		return;
+
+	ps = evsel->priv;
+	print_noise_pct(stddev_stats(&ps->res_stats[0]), avg);
+}
+
+static void aggr_printout(struct perf_evsel *evsel, int id, int nr)
+{
+	switch (aggr_mode) {
+	case AGGR_CORE:
+		fprintf(output, "S%d-C%*d%s%*d%s",
+			cpu_map__id_to_socket(id),
+			csv_output ? 0 : -8,
+			cpu_map__id_to_cpu(id),
+			csv_sep,
+			csv_output ? 0 : 4,
+			nr,
+			csv_sep);
+		break;
+	case AGGR_SOCKET:
+		fprintf(output, "S%*d%s%*d%s",
+			csv_output ? 0 : -5,
+			id,
+			csv_sep,
+			csv_output ? 0 : 4,
+			nr,
+			csv_sep);
+			break;
+	case AGGR_NONE:
+		fprintf(output, "CPU%*d%s",
+			csv_output ? 0 : -4,
+			perf_evsel__cpus(evsel)->map[id], csv_sep);
+		break;
+	case AGGR_GLOBAL:
+	default:
+		break;
+	}
+}
+
+static void nsec_printout(int id, int nr, struct perf_evsel *evsel, double avg)
+{
+	double msecs = avg / 1e6;
+	const char *fmt_v, *fmt_n;
+	char name[25];
+
+	fmt_v = csv_output ? "%.6f%s" : "%18.6f%s";
+	fmt_n = csv_output ? "%s" : "%-25s";
+
+	aggr_printout(evsel, id, nr);
+
+	scnprintf(name, sizeof(name), "%s%s",
+		  perf_evsel__name(evsel), csv_output ? "" : " (msec)");
+
+	fprintf(output, fmt_v, msecs, csv_sep);
+
+	if (csv_output)
+		fprintf(output, "%s%s", evsel->unit, csv_sep);
+	else
+		fprintf(output, "%-*s%s", unit_width, evsel->unit, csv_sep);
+
+	fprintf(output, fmt_n, name);
+
+	if (evsel->cgrp)
+		fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
+
+	if (csv_output || interval)
+		return;
+
+	if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK))
+		fprintf(output, " # %8.3f CPUs utilized          ",
+			avg / avg_stats(&walltime_nsecs_stats));
+	else
+		fprintf(output, "                                   ");
+}
+
+/* used for get_ratio_color() */
+enum grc_type {
+	GRC_STALLED_CYCLES_FE,
+	GRC_STALLED_CYCLES_BE,
+	GRC_CACHE_MISSES,
+	GRC_MAX_NR
+};
+
+static const char *get_ratio_color(enum grc_type type, double ratio)
+{
+	static const double grc_table[GRC_MAX_NR][3] = {
+		[GRC_STALLED_CYCLES_FE] = { 50.0, 30.0, 10.0 },
+		[GRC_STALLED_CYCLES_BE] = { 75.0, 50.0, 20.0 },
+		[GRC_CACHE_MISSES] 	= { 20.0, 10.0, 5.0 },
+	};
+	const char *color = PERF_COLOR_NORMAL;
+
+	if (ratio > grc_table[type][0])
+		color = PERF_COLOR_RED;
+	else if (ratio > grc_table[type][1])
+		color = PERF_COLOR_MAGENTA;
+	else if (ratio > grc_table[type][2])
+		color = PERF_COLOR_YELLOW;
+
+	return color;
+}
+
+static void print_stalled_cycles_frontend(int cpu,
+					  struct perf_evsel *evsel
+					  __maybe_unused, double avg)
+{
+	double total, ratio = 0.0;
+	const char *color;
+
+	total = avg_stats(&runtime_cycles_stats[cpu]);
+
+	if (total)
+		ratio = avg / total * 100.0;
+
+	color = get_ratio_color(GRC_STALLED_CYCLES_FE, ratio);
+
+	fprintf(output, " #  ");
+	color_fprintf(output, color, "%6.2f%%", ratio);
+	fprintf(output, " frontend cycles idle   ");
+}
+
+static void print_stalled_cycles_backend(int cpu,
+					 struct perf_evsel *evsel
+					 __maybe_unused, double avg)
+{
+	double total, ratio = 0.0;
+	const char *color;
+
+	total = avg_stats(&runtime_cycles_stats[cpu]);
+
+	if (total)
+		ratio = avg / total * 100.0;
+
+	color = get_ratio_color(GRC_STALLED_CYCLES_BE, ratio);
+
+	fprintf(output, " #  ");
+	color_fprintf(output, color, "%6.2f%%", ratio);
+	fprintf(output, " backend  cycles idle   ");
+}
+
+static void print_branch_misses(int cpu,
+				struct perf_evsel *evsel __maybe_unused,
+				double avg)
+{
+	double total, ratio = 0.0;
+	const char *color;
+
+	total = avg_stats(&runtime_branches_stats[cpu]);
+
+	if (total)
+		ratio = avg / total * 100.0;
+
+	color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+	fprintf(output, " #  ");
+	color_fprintf(output, color, "%6.2f%%", ratio);
+	fprintf(output, " of all branches        ");
+}
+
+static void print_l1_dcache_misses(int cpu,
+				   struct perf_evsel *evsel __maybe_unused,
+				   double avg)
+{
+	double total, ratio = 0.0;
+	const char *color;
+
+	total = avg_stats(&runtime_l1_dcache_stats[cpu]);
+
+	if (total)
+		ratio = avg / total * 100.0;
+
+	color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+	fprintf(output, " #  ");
+	color_fprintf(output, color, "%6.2f%%", ratio);
+	fprintf(output, " of all L1-dcache hits  ");
+}
+
+static void print_l1_icache_misses(int cpu,
+				   struct perf_evsel *evsel __maybe_unused,
+				   double avg)
+{
+	double total, ratio = 0.0;
+	const char *color;
+
+	total = avg_stats(&runtime_l1_icache_stats[cpu]);
+
+	if (total)
+		ratio = avg / total * 100.0;
+
+	color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+	fprintf(output, " #  ");
+	color_fprintf(output, color, "%6.2f%%", ratio);
+	fprintf(output, " of all L1-icache hits  ");
+}
+
+static void print_dtlb_cache_misses(int cpu,
+				    struct perf_evsel *evsel __maybe_unused,
+				    double avg)
+{
+	double total, ratio = 0.0;
+	const char *color;
+
+	total = avg_stats(&runtime_dtlb_cache_stats[cpu]);
+
+	if (total)
+		ratio = avg / total * 100.0;
+
+	color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+	fprintf(output, " #  ");
+	color_fprintf(output, color, "%6.2f%%", ratio);
+	fprintf(output, " of all dTLB cache hits ");
+}
+
+static void print_itlb_cache_misses(int cpu,
+				    struct perf_evsel *evsel __maybe_unused,
+				    double avg)
+{
+	double total, ratio = 0.0;
+	const char *color;
+
+	total = avg_stats(&runtime_itlb_cache_stats[cpu]);
+
+	if (total)
+		ratio = avg / total * 100.0;
+
+	color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+	fprintf(output, " #  ");
+	color_fprintf(output, color, "%6.2f%%", ratio);
+	fprintf(output, " of all iTLB cache hits ");
+}
+
+static void print_ll_cache_misses(int cpu,
+				  struct perf_evsel *evsel __maybe_unused,
+				  double avg)
+{
+	double total, ratio = 0.0;
+	const char *color;
+
+	total = avg_stats(&runtime_ll_cache_stats[cpu]);
+
+	if (total)
+		ratio = avg / total * 100.0;
+
+	color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+	fprintf(output, " #  ");
+	color_fprintf(output, color, "%6.2f%%", ratio);
+	fprintf(output, " of all LL-cache hits   ");
+}
+
+static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg)
+{
+	double total, ratio = 0.0, total2;
+	double sc =  evsel->scale;
+	const char *fmt;
+	int cpu = cpu_map__id_to_cpu(id);
+
+	if (csv_output) {
+		fmt = sc != 1.0 ?  "%.2f%s" : "%.0f%s";
+	} else {
+		if (big_num)
+			fmt = sc != 1.0 ? "%'18.2f%s" : "%'18.0f%s";
+		else
+			fmt = sc != 1.0 ? "%18.2f%s" : "%18.0f%s";
+	}
+
+	aggr_printout(evsel, id, nr);
+
+	if (aggr_mode == AGGR_GLOBAL)
+		cpu = 0;
+
+	fprintf(output, fmt, avg, csv_sep);
+
+	if (evsel->unit)
+		fprintf(output, "%-*s%s",
+			csv_output ? 0 : unit_width,
+			evsel->unit, csv_sep);
+
+	fprintf(output, "%-*s", csv_output ? 0 : 25, perf_evsel__name(evsel));
+
+	if (evsel->cgrp)
+		fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
+
+	if (csv_output || interval)
+		return;
+
+	if (perf_evsel__match(evsel, HARDWARE, HW_INSTRUCTIONS)) {
+		total = avg_stats(&runtime_cycles_stats[cpu]);
+		if (total) {
+			ratio = avg / total;
+			fprintf(output, " #   %5.2f  insns per cycle        ", ratio);
+		} else {
+			fprintf(output, "                                   ");
+		}
+		total = avg_stats(&runtime_stalled_cycles_front_stats[cpu]);
+		total = max(total, avg_stats(&runtime_stalled_cycles_back_stats[cpu]));
+
+		if (total && avg) {
+			ratio = total / avg;
+			fprintf(output, "\n");
+			if (aggr_mode == AGGR_NONE)
+				fprintf(output, "        ");
+			fprintf(output, "                                                  #   %5.2f  stalled cycles per insn", ratio);
+		}
+
+	} else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) &&
+			runtime_branches_stats[cpu].n != 0) {
+		print_branch_misses(cpu, evsel, avg);
+	} else if (
+		evsel->attr.type == PERF_TYPE_HW_CACHE &&
+		evsel->attr.config ==  ( PERF_COUNT_HW_CACHE_L1D |
+					((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+					((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+			runtime_l1_dcache_stats[cpu].n != 0) {
+		print_l1_dcache_misses(cpu, evsel, avg);
+	} else if (
+		evsel->attr.type == PERF_TYPE_HW_CACHE &&
+		evsel->attr.config ==  ( PERF_COUNT_HW_CACHE_L1I |
+					((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+					((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+			runtime_l1_icache_stats[cpu].n != 0) {
+		print_l1_icache_misses(cpu, evsel, avg);
+	} else if (
+		evsel->attr.type == PERF_TYPE_HW_CACHE &&
+		evsel->attr.config ==  ( PERF_COUNT_HW_CACHE_DTLB |
+					((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+					((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+			runtime_dtlb_cache_stats[cpu].n != 0) {
+		print_dtlb_cache_misses(cpu, evsel, avg);
+	} else if (
+		evsel->attr.type == PERF_TYPE_HW_CACHE &&
+		evsel->attr.config ==  ( PERF_COUNT_HW_CACHE_ITLB |
+					((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+					((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+			runtime_itlb_cache_stats[cpu].n != 0) {
+		print_itlb_cache_misses(cpu, evsel, avg);
+	} else if (
+		evsel->attr.type == PERF_TYPE_HW_CACHE &&
+		evsel->attr.config ==  ( PERF_COUNT_HW_CACHE_LL |
+					((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+					((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+			runtime_ll_cache_stats[cpu].n != 0) {
+		print_ll_cache_misses(cpu, evsel, avg);
+	} else if (perf_evsel__match(evsel, HARDWARE, HW_CACHE_MISSES) &&
+			runtime_cacherefs_stats[cpu].n != 0) {
+		total = avg_stats(&runtime_cacherefs_stats[cpu]);
+
+		if (total)
+			ratio = avg * 100 / total;
+
+		fprintf(output, " # %8.3f %% of all cache refs    ", ratio);
+
+	} else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) {
+		print_stalled_cycles_frontend(cpu, evsel, avg);
+	} else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_BACKEND)) {
+		print_stalled_cycles_backend(cpu, evsel, avg);
+	} else if (perf_evsel__match(evsel, HARDWARE, HW_CPU_CYCLES)) {
+		total = avg_stats(&runtime_nsecs_stats[cpu]);
+
+		if (total) {
+			ratio = avg / total;
+			fprintf(output, " # %8.3f GHz                    ", ratio);
+		} else {
+			fprintf(output, "                                   ");
+		}
+	} else if (transaction_run &&
+		   perf_evsel__cmp(evsel, nth_evsel(T_CYCLES_IN_TX))) {
+		total = avg_stats(&runtime_cycles_stats[cpu]);
+		if (total)
+			fprintf(output,
+				" #   %5.2f%% transactional cycles   ",
+				100.0 * (avg / total));
+	} else if (transaction_run &&
+		   perf_evsel__cmp(evsel, nth_evsel(T_CYCLES_IN_TX_CP))) {
+		total = avg_stats(&runtime_cycles_stats[cpu]);
+		total2 = avg_stats(&runtime_cycles_in_tx_stats[cpu]);
+		if (total2 < avg)
+			total2 = avg;
+		if (total)
+			fprintf(output,
+				" #   %5.2f%% aborted cycles         ",
+				100.0 * ((total2-avg) / total));
+	} else if (transaction_run &&
+		   perf_evsel__cmp(evsel, nth_evsel(T_TRANSACTION_START)) &&
+		   avg > 0 &&
+		   runtime_cycles_in_tx_stats[cpu].n != 0) {
+		total = avg_stats(&runtime_cycles_in_tx_stats[cpu]);
+
+		if (total)
+			ratio = total / avg;
+
+		fprintf(output, " # %8.0f cycles / transaction   ", ratio);
+	} else if (transaction_run &&
+		   perf_evsel__cmp(evsel, nth_evsel(T_ELISION_START)) &&
+		   avg > 0 &&
+		   runtime_cycles_in_tx_stats[cpu].n != 0) {
+		total = avg_stats(&runtime_cycles_in_tx_stats[cpu]);
+
+		if (total)
+			ratio = total / avg;
+
+		fprintf(output, " # %8.0f cycles / elision       ", ratio);
+	} else if (runtime_nsecs_stats[cpu].n != 0) {
+		char unit = 'M';
+
+		total = avg_stats(&runtime_nsecs_stats[cpu]);
+
+		if (total)
+			ratio = 1000.0 * avg / total;
+		if (ratio < 0.001) {
+			ratio *= 1000;
+			unit = 'K';
+		}
+
+		fprintf(output, " # %8.3f %c/sec                  ", ratio, unit);
+	} else {
+		fprintf(output, "                                   ");
+	}
+}
+
+static void print_aggr(char *prefix)
+{
+	struct perf_evsel *counter;
+	int cpu, cpu2, s, s2, id, nr;
+	double uval;
+	u64 ena, run, val;
+
+	if (!(aggr_map || aggr_get_id))
+		return;
+
+	for (s = 0; s < aggr_map->nr; s++) {
+		id = aggr_map->map[s];
+		evlist__for_each(evsel_list, counter) {
+			val = ena = run = 0;
+			nr = 0;
+			for (cpu = 0; cpu < perf_evsel__nr_cpus(counter); cpu++) {
+				cpu2 = perf_evsel__cpus(counter)->map[cpu];
+				s2 = aggr_get_id(evsel_list->cpus, cpu2);
+				if (s2 != id)
+					continue;
+				val += counter->counts->cpu[cpu].val;
+				ena += counter->counts->cpu[cpu].ena;
+				run += counter->counts->cpu[cpu].run;
+				nr++;
+			}
+			if (prefix)
+				fprintf(output, "%s", prefix);
+
+			if (run == 0 || ena == 0) {
+				aggr_printout(counter, id, nr);
+
+				fprintf(output, "%*s%s",
+					csv_output ? 0 : 18,
+					counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
+					csv_sep);
+
+				fprintf(output, "%-*s%s",
+					csv_output ? 0 : unit_width,
+					counter->unit, csv_sep);
+
+				fprintf(output, "%*s",
+					csv_output ? 0 : -25,
+					perf_evsel__name(counter));
+
+				if (counter->cgrp)
+					fprintf(output, "%s%s",
+						csv_sep, counter->cgrp->name);
+
+				print_running(run, ena);
+				fputc('\n', output);
+				continue;
+			}
+			uval = val * counter->scale;
+
+			if (nsec_counter(counter))
+				nsec_printout(id, nr, counter, uval);
+			else
+				abs_printout(id, nr, counter, uval);
+
+			if (!csv_output)
+				print_noise(counter, 1.0);
+
+			print_running(run, ena);
+			fputc('\n', output);
+		}
+	}
+}
+
+/*
+ * Print out the results of a single counter:
+ * aggregated counts in system-wide mode
+ */
+static void print_counter_aggr(struct perf_evsel *counter, char *prefix)
+{
+	struct perf_stat *ps = counter->priv;
+	double avg = avg_stats(&ps->res_stats[0]);
+	int scaled = counter->counts->scaled;
+	double uval;
+	double avg_enabled, avg_running;
+
+	avg_enabled = avg_stats(&ps->res_stats[1]);
+	avg_running = avg_stats(&ps->res_stats[2]);
+
+	if (prefix)
+		fprintf(output, "%s", prefix);
+
+	if (scaled == -1 || !counter->supported) {
+		fprintf(output, "%*s%s",
+			csv_output ? 0 : 18,
+			counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
+			csv_sep);
+		fprintf(output, "%-*s%s",
+			csv_output ? 0 : unit_width,
+			counter->unit, csv_sep);
+		fprintf(output, "%*s",
+			csv_output ? 0 : -25,
+			perf_evsel__name(counter));
+
+		if (counter->cgrp)
+			fprintf(output, "%s%s", csv_sep, counter->cgrp->name);
+
+		print_running(avg_running, avg_enabled);
+		fputc('\n', output);
+		return;
+	}
+
+	uval = avg * counter->scale;
+
+	if (nsec_counter(counter))
+		nsec_printout(-1, 0, counter, uval);
+	else
+		abs_printout(-1, 0, counter, uval);
+
+	print_noise(counter, avg);
+
+	print_running(avg_running, avg_enabled);
+	fprintf(output, "\n");
+}
+
+/*
+ * Print out the results of a single counter:
+ * does not use aggregated count in system-wide
+ */
+static void print_counter(struct perf_evsel *counter, char *prefix)
+{
+	u64 ena, run, val;
+	double uval;
+	int cpu;
+
+	for (cpu = 0; cpu < perf_evsel__nr_cpus(counter); cpu++) {
+		val = counter->counts->cpu[cpu].val;
+		ena = counter->counts->cpu[cpu].ena;
+		run = counter->counts->cpu[cpu].run;
+
+		if (prefix)
+			fprintf(output, "%s", prefix);
+
+		if (run == 0 || ena == 0) {
+			fprintf(output, "CPU%*d%s%*s%s",
+				csv_output ? 0 : -4,
+				perf_evsel__cpus(counter)->map[cpu], csv_sep,
+				csv_output ? 0 : 18,
+				counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
+				csv_sep);
+
+				fprintf(output, "%-*s%s",
+					csv_output ? 0 : unit_width,
+					counter->unit, csv_sep);
+
+				fprintf(output, "%*s",
+					csv_output ? 0 : -25,
+					perf_evsel__name(counter));
+
+			if (counter->cgrp)
+				fprintf(output, "%s%s",
+					csv_sep, counter->cgrp->name);
+
+			print_running(run, ena);
+			fputc('\n', output);
+			continue;
+		}
+
+		uval = val * counter->scale;
+
+		if (nsec_counter(counter))
+			nsec_printout(cpu, 0, counter, uval);
+		else
+			abs_printout(cpu, 0, counter, uval);
+
+		if (!csv_output)
+			print_noise(counter, 1.0);
+		print_running(run, ena);
+
+		fputc('\n', output);
+	}
+}
+
+static void print_stat(int argc, const char **argv)
+{
+	struct perf_evsel *counter;
+	int i;
+
+	fflush(stdout);
+
+	if (!csv_output) {
+		fprintf(output, "\n");
+		fprintf(output, " Performance counter stats for ");
+		if (target.system_wide)
+			fprintf(output, "\'system wide");
+		else if (target.cpu_list)
+			fprintf(output, "\'CPU(s) %s", target.cpu_list);
+		else if (!target__has_task(&target)) {
+			fprintf(output, "\'%s", argv[0]);
+			for (i = 1; i < argc; i++)
+				fprintf(output, " %s", argv[i]);
+		} else if (target.pid)
+			fprintf(output, "process id \'%s", target.pid);
+		else
+			fprintf(output, "thread id \'%s", target.tid);
+
+		fprintf(output, "\'");
+		if (run_count > 1)
+			fprintf(output, " (%d runs)", run_count);
+		fprintf(output, ":\n\n");
+	}
+
+	switch (aggr_mode) {
+	case AGGR_CORE:
+	case AGGR_SOCKET:
+		print_aggr(NULL);
+		break;
+	case AGGR_GLOBAL:
+		evlist__for_each(evsel_list, counter)
+			print_counter_aggr(counter, NULL);
+		break;
+	case AGGR_NONE:
+		evlist__for_each(evsel_list, counter)
+			print_counter(counter, NULL);
+		break;
+	default:
+		break;
+	}
+
+	if (!csv_output) {
+		if (!null_run)
+			fprintf(output, "\n");
+		fprintf(output, " %17.9f seconds time elapsed",
+				avg_stats(&walltime_nsecs_stats)/1e9);
+		if (run_count > 1) {
+			fprintf(output, "                                        ");
+			print_noise_pct(stddev_stats(&walltime_nsecs_stats),
+					avg_stats(&walltime_nsecs_stats));
+		}
+		fprintf(output, "\n\n");
+	}
+}
+
+static volatile int signr = -1;
+
+static void skip_signal(int signo)
+{
+	if ((child_pid == -1) || interval)
+		done = 1;
+
+	signr = signo;
+	/*
+	 * render child_pid harmless
+	 * won't send SIGTERM to a random
+	 * process in case of race condition
+	 * and fast PID recycling
+	 */
+	child_pid = -1;
+}
+
+static void sig_atexit(void)
+{
+	sigset_t set, oset;
+
+	/*
+	 * avoid race condition with SIGCHLD handler
+	 * in skip_signal() which is modifying child_pid
+	 * goal is to avoid send SIGTERM to a random
+	 * process
+	 */
+	sigemptyset(&set);
+	sigaddset(&set, SIGCHLD);
+	sigprocmask(SIG_BLOCK, &set, &oset);
+
+	if (child_pid != -1)
+		kill(child_pid, SIGTERM);
+
+	sigprocmask(SIG_SETMASK, &oset, NULL);
+
+	if (signr == -1)
+		return;
+
+	signal(signr, SIG_DFL);
+	kill(getpid(), signr);
+}
+
+static int stat__set_big_num(const struct option *opt __maybe_unused,
+			     const char *s __maybe_unused, int unset)
+{
+	big_num_opt = unset ? 0 : 1;
+	return 0;
+}
+
+static int perf_stat_init_aggr_mode(void)
+{
+	switch (aggr_mode) {
+	case AGGR_SOCKET:
+		if (cpu_map__build_socket_map(evsel_list->cpus, &aggr_map)) {
+			perror("cannot build socket map");
+			return -1;
+		}
+		aggr_get_id = cpu_map__get_socket;
+		break;
+	case AGGR_CORE:
+		if (cpu_map__build_core_map(evsel_list->cpus, &aggr_map)) {
+			perror("cannot build core map");
+			return -1;
+		}
+		aggr_get_id = cpu_map__get_core;
+		break;
+	case AGGR_NONE:
+	case AGGR_GLOBAL:
+	default:
+		break;
+	}
+	return 0;
+}
+
+static int setup_events(const char * const *attrs, unsigned len)
+{
+	unsigned i;
+
+	for (i = 0; i < len; i++) {
+		if (parse_events(evsel_list, attrs[i]))
+			return -1;
+	}
+	return 0;
+}
+
+/*
+ * Add default attributes, if there were no attributes specified or
+ * if -d/--detailed, -d -d or -d -d -d is used:
+ */
+static int add_default_attributes(void)
+{
+	struct perf_event_attr default_attrs[] = {
+
+  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK		},
+  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES	},
+  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CPU_MIGRATIONS		},
+  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_PAGE_FAULTS		},
+
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CPU_CYCLES		},
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_STALLED_CYCLES_FRONTEND	},
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_STALLED_CYCLES_BACKEND	},
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS		},
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS	},
+  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES		},
+
+};
+
+/*
+ * Detailed stats (-d), covering the L1 and last level data caches:
+ */
+	struct perf_event_attr detailed_attrs[] = {
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_L1D		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_ACCESS	<< 16)				},
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_L1D		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_MISS	<< 16)				},
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_LL			<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_ACCESS	<< 16)				},
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_LL			<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_MISS	<< 16)				},
+};
+
+/*
+ * Very detailed stats (-d -d), covering the instruction cache and the TLB caches:
+ */
+	struct perf_event_attr very_detailed_attrs[] = {
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_L1I		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_ACCESS	<< 16)				},
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_L1I		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_MISS	<< 16)				},
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_DTLB		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_ACCESS	<< 16)				},
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_DTLB		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_MISS	<< 16)				},
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_ITLB		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_ACCESS	<< 16)				},
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_ITLB		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_READ		<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_MISS	<< 16)				},
+
+};
+
+/*
+ * Very, very detailed stats (-d -d -d), adding prefetch events:
+ */
+	struct perf_event_attr very_very_detailed_attrs[] = {
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_L1D		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_PREFETCH	<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_ACCESS	<< 16)				},
+
+  { .type = PERF_TYPE_HW_CACHE,
+    .config =
+	 PERF_COUNT_HW_CACHE_L1D		<<  0  |
+	(PERF_COUNT_HW_CACHE_OP_PREFETCH	<<  8) |
+	(PERF_COUNT_HW_CACHE_RESULT_MISS	<< 16)				},
+};
+
+	/* Set attrs if no event is selected and !null_run: */
+	if (null_run)
+		return 0;
+
+	if (transaction_run) {
+		int err;
+		if (pmu_have_event("cpu", "cycles-ct") &&
+		    pmu_have_event("cpu", "el-start"))
+			err = setup_events(transaction_attrs,
+					ARRAY_SIZE(transaction_attrs));
+		else
+			err = setup_events(transaction_limited_attrs,
+				 ARRAY_SIZE(transaction_limited_attrs));
+		if (err < 0) {
+			fprintf(stderr, "Cannot set up transaction events\n");
+			return -1;
+		}
+		return 0;
+	}
+
+	if (!evsel_list->nr_entries) {
+		if (perf_evlist__add_default_attrs(evsel_list, default_attrs) < 0)
+			return -1;
+	}
+
+	/* Detailed events get appended to the event list: */
+
+	if (detailed_run <  1)
+		return 0;
+
+	/* Append detailed run extra attributes: */
+	if (perf_evlist__add_default_attrs(evsel_list, detailed_attrs) < 0)
+		return -1;
+
+	if (detailed_run < 2)
+		return 0;
+
+	/* Append very detailed run extra attributes: */
+	if (perf_evlist__add_default_attrs(evsel_list, very_detailed_attrs) < 0)
+		return -1;
+
+	if (detailed_run < 3)
+		return 0;
+
+	/* Append very, very detailed run extra attributes: */
+	return perf_evlist__add_default_attrs(evsel_list, very_very_detailed_attrs);
+}
+
+int cmd_stat(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	bool append_file = false;
+	int output_fd = 0;
+	const char *output_name	= NULL;
+	const struct option options[] = {
+	OPT_BOOLEAN('T', "transaction", &transaction_run,
+		    "hardware transaction statistics"),
+	OPT_CALLBACK('e', "event", &evsel_list, "event",
+		     "event selector. use 'perf list' to list available events",
+		     parse_events_option),
+	OPT_CALLBACK(0, "filter", &evsel_list, "filter",
+		     "event filter", parse_filter),
+	OPT_BOOLEAN('i', "no-inherit", &no_inherit,
+		    "child tasks do not inherit counters"),
+	OPT_STRING('p', "pid", &target.pid, "pid",
+		   "stat events on existing process id"),
+	OPT_STRING('t', "tid", &target.tid, "tid",
+		   "stat events on existing thread id"),
+	OPT_BOOLEAN('a', "all-cpus", &target.system_wide,
+		    "system-wide collection from all CPUs"),
+	OPT_BOOLEAN('g', "group", &group,
+		    "put the counters into a counter group"),
+	OPT_BOOLEAN('c', "scale", &scale, "scale/normalize counters"),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show counter open errors, etc)"),
+	OPT_INTEGER('r', "repeat", &run_count,
+		    "repeat command and print average + stddev (max: 100, forever: 0)"),
+	OPT_BOOLEAN('n', "null", &null_run,
+		    "null run - dont start any counters"),
+	OPT_INCR('d', "detailed", &detailed_run,
+		    "detailed run - start a lot of events"),
+	OPT_BOOLEAN('S', "sync", &sync_run,
+		    "call sync() before starting a run"),
+	OPT_CALLBACK_NOOPT('B', "big-num", NULL, NULL,
+			   "print large numbers with thousands\' separators",
+			   stat__set_big_num),
+	OPT_STRING('C', "cpu", &target.cpu_list, "cpu",
+		    "list of cpus to monitor in system-wide"),
+	OPT_SET_UINT('A', "no-aggr", &aggr_mode,
+		    "disable CPU count aggregation", AGGR_NONE),
+	OPT_STRING('x', "field-separator", &csv_sep, "separator",
+		   "print counts with custom separator"),
+	OPT_CALLBACK('G', "cgroup", &evsel_list, "name",
+		     "monitor event in cgroup name only", parse_cgroups),
+	OPT_STRING('o', "output", &output_name, "file", "output file name"),
+	OPT_BOOLEAN(0, "append", &append_file, "append to the output file"),
+	OPT_INTEGER(0, "log-fd", &output_fd,
+		    "log output to fd, instead of stderr"),
+	OPT_STRING(0, "pre", &pre_cmd, "command",
+			"command to run prior to the measured command"),
+	OPT_STRING(0, "post", &post_cmd, "command",
+			"command to run after to the measured command"),
+	OPT_UINTEGER('I', "interval-print", &interval,
+		    "print counts at regular interval in ms (>= 100)"),
+	OPT_SET_UINT(0, "per-socket", &aggr_mode,
+		     "aggregate counts per processor socket", AGGR_SOCKET),
+	OPT_SET_UINT(0, "per-core", &aggr_mode,
+		     "aggregate counts per physical processor core", AGGR_CORE),
+	OPT_UINTEGER('D', "delay", &initial_delay,
+		     "ms to wait before starting measurement after program start"),
+	OPT_END()
+	};
+	const char * const stat_usage[] = {
+		"perf stat [<options>] [<command>]",
+		NULL
+	};
+	int status = -EINVAL, run_idx;
+	const char *mode;
+
+	setlocale(LC_ALL, "");
+
+	evsel_list = perf_evlist__new();
+	if (evsel_list == NULL)
+		return -ENOMEM;
+
+	argc = parse_options(argc, argv, options, stat_usage,
+		PARSE_OPT_STOP_AT_NON_OPTION);
+
+	output = stderr;
+	if (output_name && strcmp(output_name, "-"))
+		output = NULL;
+
+	if (output_name && output_fd) {
+		fprintf(stderr, "cannot use both --output and --log-fd\n");
+		parse_options_usage(stat_usage, options, "o", 1);
+		parse_options_usage(NULL, options, "log-fd", 0);
+		goto out;
+	}
+
+	if (output_fd < 0) {
+		fprintf(stderr, "argument to --log-fd must be a > 0\n");
+		parse_options_usage(stat_usage, options, "log-fd", 0);
+		goto out;
+	}
+
+	if (!output) {
+		struct timespec tm;
+		mode = append_file ? "a" : "w";
+
+		output = fopen(output_name, mode);
+		if (!output) {
+			perror("failed to create output file");
+			return -1;
+		}
+		clock_gettime(CLOCK_REALTIME, &tm);
+		fprintf(output, "# started on %s\n", ctime(&tm.tv_sec));
+	} else if (output_fd > 0) {
+		mode = append_file ? "a" : "w";
+		output = fdopen(output_fd, mode);
+		if (!output) {
+			perror("Failed opening logfd");
+			return -errno;
+		}
+	}
+
+	if (csv_sep) {
+		csv_output = true;
+		if (!strcmp(csv_sep, "\\t"))
+			csv_sep = "\t";
+	} else
+		csv_sep = DEFAULT_SEPARATOR;
+
+	/*
+	 * let the spreadsheet do the pretty-printing
+	 */
+	if (csv_output) {
+		/* User explicitly passed -B? */
+		if (big_num_opt == 1) {
+			fprintf(stderr, "-B option not supported with -x\n");
+			parse_options_usage(stat_usage, options, "B", 1);
+			parse_options_usage(NULL, options, "x", 1);
+			goto out;
+		} else /* Nope, so disable big number formatting */
+			big_num = false;
+	} else if (big_num_opt == 0) /* User passed --no-big-num */
+		big_num = false;
+
+	if (!argc && target__none(&target))
+		usage_with_options(stat_usage, options);
+
+	if (run_count < 0) {
+		pr_err("Run count must be a positive number\n");
+		parse_options_usage(stat_usage, options, "r", 1);
+		goto out;
+	} else if (run_count == 0) {
+		forever = true;
+		run_count = 1;
+	}
+
+	/* no_aggr, cgroup are for system-wide only */
+	if ((aggr_mode != AGGR_GLOBAL || nr_cgroups) &&
+	    !target__has_cpu(&target)) {
+		fprintf(stderr, "both cgroup and no-aggregation "
+			"modes only available in system-wide mode\n");
+
+		parse_options_usage(stat_usage, options, "G", 1);
+		parse_options_usage(NULL, options, "A", 1);
+		parse_options_usage(NULL, options, "a", 1);
+		goto out;
+	}
+
+	if (add_default_attributes())
+		goto out;
+
+	target__validate(&target);
+
+	if (perf_evlist__create_maps(evsel_list, &target) < 0) {
+		if (target__has_task(&target)) {
+			pr_err("Problems finding threads of monitor\n");
+			parse_options_usage(stat_usage, options, "p", 1);
+			parse_options_usage(NULL, options, "t", 1);
+		} else if (target__has_cpu(&target)) {
+			perror("failed to parse CPUs map");
+			parse_options_usage(stat_usage, options, "C", 1);
+			parse_options_usage(NULL, options, "a", 1);
+		}
+		goto out;
+	}
+	if (interval && interval < 100) {
+		pr_err("print interval must be >= 100ms\n");
+		parse_options_usage(stat_usage, options, "I", 1);
+		goto out;
+	}
+
+	if (perf_evlist__alloc_stats(evsel_list, interval))
+		goto out;
+
+	if (perf_stat_init_aggr_mode())
+		goto out;
+
+	/*
+	 * We dont want to block the signals - that would cause
+	 * child tasks to inherit that and Ctrl-C would not work.
+	 * What we want is for Ctrl-C to work in the exec()-ed
+	 * task, but being ignored by perf stat itself:
+	 */
+	atexit(sig_atexit);
+	if (!forever)
+		signal(SIGINT,  skip_signal);
+	signal(SIGCHLD, skip_signal);
+	signal(SIGALRM, skip_signal);
+	signal(SIGABRT, skip_signal);
+
+	status = 0;
+	for (run_idx = 0; forever || run_idx < run_count; run_idx++) {
+		if (run_count != 1 && verbose)
+			fprintf(output, "[ perf stat: executing run #%d ... ]\n",
+				run_idx + 1);
+
+		status = run_perf_stat(argc, argv);
+		if (forever && status != -1) {
+			print_stat(argc, argv);
+			perf_stat__reset_stats(evsel_list);
+		}
+	}
+
+	if (!forever && status != -1 && !interval)
+		print_stat(argc, argv);
+
+	perf_evlist__free_stats(evsel_list);
+out:
+	perf_evlist__delete(evsel_list);
+	return status;
+}
diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c
new file mode 100644
index 000000000..e50fe1187
--- /dev/null
+++ b/tools/perf/builtin-timechart.c
@@ -0,0 +1,2011 @@
+/*
+ * builtin-timechart.c - make an svg timechart of system activity
+ *
+ * (C) Copyright 2009 Intel Corporation
+ *
+ * Authors:
+ *     Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <traceevent/event-parse.h>
+
+#include "builtin.h"
+
+#include "util/util.h"
+
+#include "util/color.h"
+#include <linux/list.h>
+#include "util/cache.h"
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include <linux/rbtree.h>
+#include "util/symbol.h"
+#include "util/callchain.h"
+#include "util/strlist.h"
+
+#include "perf.h"
+#include "util/header.h"
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+#include "util/event.h"
+#include "util/session.h"
+#include "util/svghelper.h"
+#include "util/tool.h"
+#include "util/data.h"
+#include "util/debug.h"
+
+#define SUPPORT_OLD_POWER_EVENTS 1
+#define PWR_EVENT_EXIT -1
+
+struct per_pid;
+struct power_event;
+struct wake_event;
+
+struct timechart {
+	struct perf_tool	tool;
+	struct per_pid		*all_data;
+	struct power_event	*power_events;
+	struct wake_event	*wake_events;
+	int			proc_num;
+	unsigned int		numcpus;
+	u64			min_freq,	/* Lowest CPU frequency seen */
+				max_freq,	/* Highest CPU frequency seen */
+				turbo_frequency,
+				first_time, last_time;
+	bool			power_only,
+				tasks_only,
+				with_backtrace,
+				topology;
+	/* IO related settings */
+	u64			io_events;
+	bool			io_only,
+				skip_eagain;
+	u64			min_time,
+				merge_dist;
+	bool			force;
+};
+
+struct per_pidcomm;
+struct cpu_sample;
+struct io_sample;
+
+/*
+ * Datastructure layout:
+ * We keep an list of "pid"s, matching the kernels notion of a task struct.
+ * Each "pid" entry, has a list of "comm"s.
+ *	this is because we want to track different programs different, while
+ *	exec will reuse the original pid (by design).
+ * Each comm has a list of samples that will be used to draw
+ * final graph.
+ */
+
+struct per_pid {
+	struct per_pid *next;
+
+	int		pid;
+	int		ppid;
+
+	u64		start_time;
+	u64		end_time;
+	u64		total_time;
+	u64		total_bytes;
+	int		display;
+
+	struct per_pidcomm *all;
+	struct per_pidcomm *current;
+};
+
+
+struct per_pidcomm {
+	struct per_pidcomm *next;
+
+	u64		start_time;
+	u64		end_time;
+	u64		total_time;
+	u64		max_bytes;
+	u64		total_bytes;
+
+	int		Y;
+	int		display;
+
+	long		state;
+	u64		state_since;
+
+	char		*comm;
+
+	struct cpu_sample *samples;
+	struct io_sample  *io_samples;
+};
+
+struct sample_wrapper {
+	struct sample_wrapper *next;
+
+	u64		timestamp;
+	unsigned char	data[0];
+};
+
+#define TYPE_NONE	0
+#define TYPE_RUNNING	1
+#define TYPE_WAITING	2
+#define TYPE_BLOCKED	3
+
+struct cpu_sample {
+	struct cpu_sample *next;
+
+	u64 start_time;
+	u64 end_time;
+	int type;
+	int cpu;
+	const char *backtrace;
+};
+
+enum {
+	IOTYPE_READ,
+	IOTYPE_WRITE,
+	IOTYPE_SYNC,
+	IOTYPE_TX,
+	IOTYPE_RX,
+	IOTYPE_POLL,
+};
+
+struct io_sample {
+	struct io_sample *next;
+
+	u64 start_time;
+	u64 end_time;
+	u64 bytes;
+	int type;
+	int fd;
+	int err;
+	int merges;
+};
+
+#define CSTATE 1
+#define PSTATE 2
+
+struct power_event {
+	struct power_event *next;
+	int type;
+	int state;
+	u64 start_time;
+	u64 end_time;
+	int cpu;
+};
+
+struct wake_event {
+	struct wake_event *next;
+	int waker;
+	int wakee;
+	u64 time;
+	const char *backtrace;
+};
+
+struct process_filter {
+	char			*name;
+	int			pid;
+	struct process_filter	*next;
+};
+
+static struct process_filter *process_filter;
+
+
+static struct per_pid *find_create_pid(struct timechart *tchart, int pid)
+{
+	struct per_pid *cursor = tchart->all_data;
+
+	while (cursor) {
+		if (cursor->pid == pid)
+			return cursor;
+		cursor = cursor->next;
+	}
+	cursor = zalloc(sizeof(*cursor));
+	assert(cursor != NULL);
+	cursor->pid = pid;
+	cursor->next = tchart->all_data;
+	tchart->all_data = cursor;
+	return cursor;
+}
+
+static void pid_set_comm(struct timechart *tchart, int pid, char *comm)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	p = find_create_pid(tchart, pid);
+	c = p->all;
+	while (c) {
+		if (c->comm && strcmp(c->comm, comm) == 0) {
+			p->current = c;
+			return;
+		}
+		if (!c->comm) {
+			c->comm = strdup(comm);
+			p->current = c;
+			return;
+		}
+		c = c->next;
+	}
+	c = zalloc(sizeof(*c));
+	assert(c != NULL);
+	c->comm = strdup(comm);
+	p->current = c;
+	c->next = p->all;
+	p->all = c;
+}
+
+static void pid_fork(struct timechart *tchart, int pid, int ppid, u64 timestamp)
+{
+	struct per_pid *p, *pp;
+	p = find_create_pid(tchart, pid);
+	pp = find_create_pid(tchart, ppid);
+	p->ppid = ppid;
+	if (pp->current && pp->current->comm && !p->current)
+		pid_set_comm(tchart, pid, pp->current->comm);
+
+	p->start_time = timestamp;
+	if (p->current && !p->current->start_time) {
+		p->current->start_time = timestamp;
+		p->current->state_since = timestamp;
+	}
+}
+
+static void pid_exit(struct timechart *tchart, int pid, u64 timestamp)
+{
+	struct per_pid *p;
+	p = find_create_pid(tchart, pid);
+	p->end_time = timestamp;
+	if (p->current)
+		p->current->end_time = timestamp;
+}
+
+static void pid_put_sample(struct timechart *tchart, int pid, int type,
+			   unsigned int cpu, u64 start, u64 end,
+			   const char *backtrace)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	struct cpu_sample *sample;
+
+	p = find_create_pid(tchart, pid);
+	c = p->current;
+	if (!c) {
+		c = zalloc(sizeof(*c));
+		assert(c != NULL);
+		p->current = c;
+		c->next = p->all;
+		p->all = c;
+	}
+
+	sample = zalloc(sizeof(*sample));
+	assert(sample != NULL);
+	sample->start_time = start;
+	sample->end_time = end;
+	sample->type = type;
+	sample->next = c->samples;
+	sample->cpu = cpu;
+	sample->backtrace = backtrace;
+	c->samples = sample;
+
+	if (sample->type == TYPE_RUNNING && end > start && start > 0) {
+		c->total_time += (end-start);
+		p->total_time += (end-start);
+	}
+
+	if (c->start_time == 0 || c->start_time > start)
+		c->start_time = start;
+	if (p->start_time == 0 || p->start_time > start)
+		p->start_time = start;
+}
+
+#define MAX_CPUS 4096
+
+static u64 cpus_cstate_start_times[MAX_CPUS];
+static int cpus_cstate_state[MAX_CPUS];
+static u64 cpus_pstate_start_times[MAX_CPUS];
+static u64 cpus_pstate_state[MAX_CPUS];
+
+static int process_comm_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample __maybe_unused,
+			      struct machine *machine __maybe_unused)
+{
+	struct timechart *tchart = container_of(tool, struct timechart, tool);
+	pid_set_comm(tchart, event->comm.tid, event->comm.comm);
+	return 0;
+}
+
+static int process_fork_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample __maybe_unused,
+			      struct machine *machine __maybe_unused)
+{
+	struct timechart *tchart = container_of(tool, struct timechart, tool);
+	pid_fork(tchart, event->fork.pid, event->fork.ppid, event->fork.time);
+	return 0;
+}
+
+static int process_exit_event(struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample __maybe_unused,
+			      struct machine *machine __maybe_unused)
+{
+	struct timechart *tchart = container_of(tool, struct timechart, tool);
+	pid_exit(tchart, event->fork.pid, event->fork.time);
+	return 0;
+}
+
+#ifdef SUPPORT_OLD_POWER_EVENTS
+static int use_old_power_events;
+#endif
+
+static void c_state_start(int cpu, u64 timestamp, int state)
+{
+	cpus_cstate_start_times[cpu] = timestamp;
+	cpus_cstate_state[cpu] = state;
+}
+
+static void c_state_end(struct timechart *tchart, int cpu, u64 timestamp)
+{
+	struct power_event *pwr = zalloc(sizeof(*pwr));
+
+	if (!pwr)
+		return;
+
+	pwr->state = cpus_cstate_state[cpu];
+	pwr->start_time = cpus_cstate_start_times[cpu];
+	pwr->end_time = timestamp;
+	pwr->cpu = cpu;
+	pwr->type = CSTATE;
+	pwr->next = tchart->power_events;
+
+	tchart->power_events = pwr;
+}
+
+static void p_state_change(struct timechart *tchart, int cpu, u64 timestamp, u64 new_freq)
+{
+	struct power_event *pwr;
+
+	if (new_freq > 8000000) /* detect invalid data */
+		return;
+
+	pwr = zalloc(sizeof(*pwr));
+	if (!pwr)
+		return;
+
+	pwr->state = cpus_pstate_state[cpu];
+	pwr->start_time = cpus_pstate_start_times[cpu];
+	pwr->end_time = timestamp;
+	pwr->cpu = cpu;
+	pwr->type = PSTATE;
+	pwr->next = tchart->power_events;
+
+	if (!pwr->start_time)
+		pwr->start_time = tchart->first_time;
+
+	tchart->power_events = pwr;
+
+	cpus_pstate_state[cpu] = new_freq;
+	cpus_pstate_start_times[cpu] = timestamp;
+
+	if ((u64)new_freq > tchart->max_freq)
+		tchart->max_freq = new_freq;
+
+	if (new_freq < tchart->min_freq || tchart->min_freq == 0)
+		tchart->min_freq = new_freq;
+
+	if (new_freq == tchart->max_freq - 1000)
+		tchart->turbo_frequency = tchart->max_freq;
+}
+
+static void sched_wakeup(struct timechart *tchart, int cpu, u64 timestamp,
+			 int waker, int wakee, u8 flags, const char *backtrace)
+{
+	struct per_pid *p;
+	struct wake_event *we = zalloc(sizeof(*we));
+
+	if (!we)
+		return;
+
+	we->time = timestamp;
+	we->waker = waker;
+	we->backtrace = backtrace;
+
+	if ((flags & TRACE_FLAG_HARDIRQ) || (flags & TRACE_FLAG_SOFTIRQ))
+		we->waker = -1;
+
+	we->wakee = wakee;
+	we->next = tchart->wake_events;
+	tchart->wake_events = we;
+	p = find_create_pid(tchart, we->wakee);
+
+	if (p && p->current && p->current->state == TYPE_NONE) {
+		p->current->state_since = timestamp;
+		p->current->state = TYPE_WAITING;
+	}
+	if (p && p->current && p->current->state == TYPE_BLOCKED) {
+		pid_put_sample(tchart, p->pid, p->current->state, cpu,
+			       p->current->state_since, timestamp, NULL);
+		p->current->state_since = timestamp;
+		p->current->state = TYPE_WAITING;
+	}
+}
+
+static void sched_switch(struct timechart *tchart, int cpu, u64 timestamp,
+			 int prev_pid, int next_pid, u64 prev_state,
+			 const char *backtrace)
+{
+	struct per_pid *p = NULL, *prev_p;
+
+	prev_p = find_create_pid(tchart, prev_pid);
+
+	p = find_create_pid(tchart, next_pid);
+
+	if (prev_p->current && prev_p->current->state != TYPE_NONE)
+		pid_put_sample(tchart, prev_pid, TYPE_RUNNING, cpu,
+			       prev_p->current->state_since, timestamp,
+			       backtrace);
+	if (p && p->current) {
+		if (p->current->state != TYPE_NONE)
+			pid_put_sample(tchart, next_pid, p->current->state, cpu,
+				       p->current->state_since, timestamp,
+				       backtrace);
+
+		p->current->state_since = timestamp;
+		p->current->state = TYPE_RUNNING;
+	}
+
+	if (prev_p->current) {
+		prev_p->current->state = TYPE_NONE;
+		prev_p->current->state_since = timestamp;
+		if (prev_state & 2)
+			prev_p->current->state = TYPE_BLOCKED;
+		if (prev_state == 0)
+			prev_p->current->state = TYPE_WAITING;
+	}
+}
+
+static const char *cat_backtrace(union perf_event *event,
+				 struct perf_sample *sample,
+				 struct machine *machine)
+{
+	struct addr_location al;
+	unsigned int i;
+	char *p = NULL;
+	size_t p_len;
+	u8 cpumode = PERF_RECORD_MISC_USER;
+	struct addr_location tal;
+	struct ip_callchain *chain = sample->callchain;
+	FILE *f = open_memstream(&p, &p_len);
+
+	if (!f) {
+		perror("open_memstream error");
+		return NULL;
+	}
+
+	if (!chain)
+		goto exit;
+
+	if (perf_event__preprocess_sample(event, machine, &al, sample) < 0) {
+		fprintf(stderr, "problem processing %d event, skipping it.\n",
+			event->header.type);
+		goto exit;
+	}
+
+	for (i = 0; i < chain->nr; i++) {
+		u64 ip;
+
+		if (callchain_param.order == ORDER_CALLEE)
+			ip = chain->ips[i];
+		else
+			ip = chain->ips[chain->nr - i - 1];
+
+		if (ip >= PERF_CONTEXT_MAX) {
+			switch (ip) {
+			case PERF_CONTEXT_HV:
+				cpumode = PERF_RECORD_MISC_HYPERVISOR;
+				break;
+			case PERF_CONTEXT_KERNEL:
+				cpumode = PERF_RECORD_MISC_KERNEL;
+				break;
+			case PERF_CONTEXT_USER:
+				cpumode = PERF_RECORD_MISC_USER;
+				break;
+			default:
+				pr_debug("invalid callchain context: "
+					 "%"PRId64"\n", (s64) ip);
+
+				/*
+				 * It seems the callchain is corrupted.
+				 * Discard all.
+				 */
+				zfree(&p);
+				goto exit;
+			}
+			continue;
+		}
+
+		tal.filtered = 0;
+		thread__find_addr_location(al.thread, cpumode,
+					   MAP__FUNCTION, ip, &tal);
+
+		if (tal.sym)
+			fprintf(f, "..... %016" PRIx64 " %s\n", ip,
+				tal.sym->name);
+		else
+			fprintf(f, "..... %016" PRIx64 "\n", ip);
+	}
+
+exit:
+	fclose(f);
+
+	return p;
+}
+
+typedef int (*tracepoint_handler)(struct timechart *tchart,
+				  struct perf_evsel *evsel,
+				  struct perf_sample *sample,
+				  const char *backtrace);
+
+static int process_sample_event(struct perf_tool *tool,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel,
+				struct machine *machine)
+{
+	struct timechart *tchart = container_of(tool, struct timechart, tool);
+
+	if (evsel->attr.sample_type & PERF_SAMPLE_TIME) {
+		if (!tchart->first_time || tchart->first_time > sample->time)
+			tchart->first_time = sample->time;
+		if (tchart->last_time < sample->time)
+			tchart->last_time = sample->time;
+	}
+
+	if (evsel->handler != NULL) {
+		tracepoint_handler f = evsel->handler;
+		return f(tchart, evsel, sample,
+			 cat_backtrace(event, sample, machine));
+	}
+
+	return 0;
+}
+
+static int
+process_sample_cpu_idle(struct timechart *tchart __maybe_unused,
+			struct perf_evsel *evsel,
+			struct perf_sample *sample,
+			const char *backtrace __maybe_unused)
+{
+	u32 state = perf_evsel__intval(evsel, sample, "state");
+	u32 cpu_id = perf_evsel__intval(evsel, sample, "cpu_id");
+
+	if (state == (u32)PWR_EVENT_EXIT)
+		c_state_end(tchart, cpu_id, sample->time);
+	else
+		c_state_start(cpu_id, sample->time, state);
+	return 0;
+}
+
+static int
+process_sample_cpu_frequency(struct timechart *tchart,
+			     struct perf_evsel *evsel,
+			     struct perf_sample *sample,
+			     const char *backtrace __maybe_unused)
+{
+	u32 state = perf_evsel__intval(evsel, sample, "state");
+	u32 cpu_id = perf_evsel__intval(evsel, sample, "cpu_id");
+
+	p_state_change(tchart, cpu_id, sample->time, state);
+	return 0;
+}
+
+static int
+process_sample_sched_wakeup(struct timechart *tchart,
+			    struct perf_evsel *evsel,
+			    struct perf_sample *sample,
+			    const char *backtrace)
+{
+	u8 flags = perf_evsel__intval(evsel, sample, "common_flags");
+	int waker = perf_evsel__intval(evsel, sample, "common_pid");
+	int wakee = perf_evsel__intval(evsel, sample, "pid");
+
+	sched_wakeup(tchart, sample->cpu, sample->time, waker, wakee, flags, backtrace);
+	return 0;
+}
+
+static int
+process_sample_sched_switch(struct timechart *tchart,
+			    struct perf_evsel *evsel,
+			    struct perf_sample *sample,
+			    const char *backtrace)
+{
+	int prev_pid = perf_evsel__intval(evsel, sample, "prev_pid");
+	int next_pid = perf_evsel__intval(evsel, sample, "next_pid");
+	u64 prev_state = perf_evsel__intval(evsel, sample, "prev_state");
+
+	sched_switch(tchart, sample->cpu, sample->time, prev_pid, next_pid,
+		     prev_state, backtrace);
+	return 0;
+}
+
+#ifdef SUPPORT_OLD_POWER_EVENTS
+static int
+process_sample_power_start(struct timechart *tchart __maybe_unused,
+			   struct perf_evsel *evsel,
+			   struct perf_sample *sample,
+			   const char *backtrace __maybe_unused)
+{
+	u64 cpu_id = perf_evsel__intval(evsel, sample, "cpu_id");
+	u64 value = perf_evsel__intval(evsel, sample, "value");
+
+	c_state_start(cpu_id, sample->time, value);
+	return 0;
+}
+
+static int
+process_sample_power_end(struct timechart *tchart,
+			 struct perf_evsel *evsel __maybe_unused,
+			 struct perf_sample *sample,
+			 const char *backtrace __maybe_unused)
+{
+	c_state_end(tchart, sample->cpu, sample->time);
+	return 0;
+}
+
+static int
+process_sample_power_frequency(struct timechart *tchart,
+			       struct perf_evsel *evsel,
+			       struct perf_sample *sample,
+			       const char *backtrace __maybe_unused)
+{
+	u64 cpu_id = perf_evsel__intval(evsel, sample, "cpu_id");
+	u64 value = perf_evsel__intval(evsel, sample, "value");
+
+	p_state_change(tchart, cpu_id, sample->time, value);
+	return 0;
+}
+#endif /* SUPPORT_OLD_POWER_EVENTS */
+
+/*
+ * After the last sample we need to wrap up the current C/P state
+ * and close out each CPU for these.
+ */
+static void end_sample_processing(struct timechart *tchart)
+{
+	u64 cpu;
+	struct power_event *pwr;
+
+	for (cpu = 0; cpu <= tchart->numcpus; cpu++) {
+		/* C state */
+#if 0
+		pwr = zalloc(sizeof(*pwr));
+		if (!pwr)
+			return;
+
+		pwr->state = cpus_cstate_state[cpu];
+		pwr->start_time = cpus_cstate_start_times[cpu];
+		pwr->end_time = tchart->last_time;
+		pwr->cpu = cpu;
+		pwr->type = CSTATE;
+		pwr->next = tchart->power_events;
+
+		tchart->power_events = pwr;
+#endif
+		/* P state */
+
+		pwr = zalloc(sizeof(*pwr));
+		if (!pwr)
+			return;
+
+		pwr->state = cpus_pstate_state[cpu];
+		pwr->start_time = cpus_pstate_start_times[cpu];
+		pwr->end_time = tchart->last_time;
+		pwr->cpu = cpu;
+		pwr->type = PSTATE;
+		pwr->next = tchart->power_events;
+
+		if (!pwr->start_time)
+			pwr->start_time = tchart->first_time;
+		if (!pwr->state)
+			pwr->state = tchart->min_freq;
+		tchart->power_events = pwr;
+	}
+}
+
+static int pid_begin_io_sample(struct timechart *tchart, int pid, int type,
+			       u64 start, int fd)
+{
+	struct per_pid *p = find_create_pid(tchart, pid);
+	struct per_pidcomm *c = p->current;
+	struct io_sample *sample;
+	struct io_sample *prev;
+
+	if (!c) {
+		c = zalloc(sizeof(*c));
+		if (!c)
+			return -ENOMEM;
+		p->current = c;
+		c->next = p->all;
+		p->all = c;
+	}
+
+	prev = c->io_samples;
+
+	if (prev && prev->start_time && !prev->end_time) {
+		pr_warning("Skip invalid start event: "
+			   "previous event already started!\n");
+
+		/* remove previous event that has been started,
+		 * we are not sure we will ever get an end for it */
+		c->io_samples = prev->next;
+		free(prev);
+		return 0;
+	}
+
+	sample = zalloc(sizeof(*sample));
+	if (!sample)
+		return -ENOMEM;
+	sample->start_time = start;
+	sample->type = type;
+	sample->fd = fd;
+	sample->next = c->io_samples;
+	c->io_samples = sample;
+
+	if (c->start_time == 0 || c->start_time > start)
+		c->start_time = start;
+
+	return 0;
+}
+
+static int pid_end_io_sample(struct timechart *tchart, int pid, int type,
+			     u64 end, long ret)
+{
+	struct per_pid *p = find_create_pid(tchart, pid);
+	struct per_pidcomm *c = p->current;
+	struct io_sample *sample, *prev;
+
+	if (!c) {
+		pr_warning("Invalid pidcomm!\n");
+		return -1;
+	}
+
+	sample = c->io_samples;
+
+	if (!sample) /* skip partially captured events */
+		return 0;
+
+	if (sample->end_time) {
+		pr_warning("Skip invalid end event: "
+			   "previous event already ended!\n");
+		return 0;
+	}
+
+	if (sample->type != type) {
+		pr_warning("Skip invalid end event: invalid event type!\n");
+		return 0;
+	}
+
+	sample->end_time = end;
+	prev = sample->next;
+
+	/* we want to be able to see small and fast transfers, so make them
+	 * at least min_time long, but don't overlap them */
+	if (sample->end_time - sample->start_time < tchart->min_time)
+		sample->end_time = sample->start_time + tchart->min_time;
+	if (prev && sample->start_time < prev->end_time) {
+		if (prev->err) /* try to make errors more visible */
+			sample->start_time = prev->end_time;
+		else
+			prev->end_time = sample->start_time;
+	}
+
+	if (ret < 0) {
+		sample->err = ret;
+	} else if (type == IOTYPE_READ || type == IOTYPE_WRITE ||
+		   type == IOTYPE_TX || type == IOTYPE_RX) {
+
+		if ((u64)ret > c->max_bytes)
+			c->max_bytes = ret;
+
+		c->total_bytes += ret;
+		p->total_bytes += ret;
+		sample->bytes = ret;
+	}
+
+	/* merge two requests to make svg smaller and render-friendly */
+	if (prev &&
+	    prev->type == sample->type &&
+	    prev->err == sample->err &&
+	    prev->fd == sample->fd &&
+	    prev->end_time + tchart->merge_dist >= sample->start_time) {
+
+		sample->bytes += prev->bytes;
+		sample->merges += prev->merges + 1;
+
+		sample->start_time = prev->start_time;
+		sample->next = prev->next;
+		free(prev);
+
+		if (!sample->err && sample->bytes > c->max_bytes)
+			c->max_bytes = sample->bytes;
+	}
+
+	tchart->io_events++;
+
+	return 0;
+}
+
+static int
+process_enter_read(struct timechart *tchart,
+		   struct perf_evsel *evsel,
+		   struct perf_sample *sample)
+{
+	long fd = perf_evsel__intval(evsel, sample, "fd");
+	return pid_begin_io_sample(tchart, sample->tid, IOTYPE_READ,
+				   sample->time, fd);
+}
+
+static int
+process_exit_read(struct timechart *tchart,
+		  struct perf_evsel *evsel,
+		  struct perf_sample *sample)
+{
+	long ret = perf_evsel__intval(evsel, sample, "ret");
+	return pid_end_io_sample(tchart, sample->tid, IOTYPE_READ,
+				 sample->time, ret);
+}
+
+static int
+process_enter_write(struct timechart *tchart,
+		    struct perf_evsel *evsel,
+		    struct perf_sample *sample)
+{
+	long fd = perf_evsel__intval(evsel, sample, "fd");
+	return pid_begin_io_sample(tchart, sample->tid, IOTYPE_WRITE,
+				   sample->time, fd);
+}
+
+static int
+process_exit_write(struct timechart *tchart,
+		   struct perf_evsel *evsel,
+		   struct perf_sample *sample)
+{
+	long ret = perf_evsel__intval(evsel, sample, "ret");
+	return pid_end_io_sample(tchart, sample->tid, IOTYPE_WRITE,
+				 sample->time, ret);
+}
+
+static int
+process_enter_sync(struct timechart *tchart,
+		   struct perf_evsel *evsel,
+		   struct perf_sample *sample)
+{
+	long fd = perf_evsel__intval(evsel, sample, "fd");
+	return pid_begin_io_sample(tchart, sample->tid, IOTYPE_SYNC,
+				   sample->time, fd);
+}
+
+static int
+process_exit_sync(struct timechart *tchart,
+		  struct perf_evsel *evsel,
+		  struct perf_sample *sample)
+{
+	long ret = perf_evsel__intval(evsel, sample, "ret");
+	return pid_end_io_sample(tchart, sample->tid, IOTYPE_SYNC,
+				 sample->time, ret);
+}
+
+static int
+process_enter_tx(struct timechart *tchart,
+		 struct perf_evsel *evsel,
+		 struct perf_sample *sample)
+{
+	long fd = perf_evsel__intval(evsel, sample, "fd");
+	return pid_begin_io_sample(tchart, sample->tid, IOTYPE_TX,
+				   sample->time, fd);
+}
+
+static int
+process_exit_tx(struct timechart *tchart,
+		struct perf_evsel *evsel,
+		struct perf_sample *sample)
+{
+	long ret = perf_evsel__intval(evsel, sample, "ret");
+	return pid_end_io_sample(tchart, sample->tid, IOTYPE_TX,
+				 sample->time, ret);
+}
+
+static int
+process_enter_rx(struct timechart *tchart,
+		 struct perf_evsel *evsel,
+		 struct perf_sample *sample)
+{
+	long fd = perf_evsel__intval(evsel, sample, "fd");
+	return pid_begin_io_sample(tchart, sample->tid, IOTYPE_RX,
+				   sample->time, fd);
+}
+
+static int
+process_exit_rx(struct timechart *tchart,
+		struct perf_evsel *evsel,
+		struct perf_sample *sample)
+{
+	long ret = perf_evsel__intval(evsel, sample, "ret");
+	return pid_end_io_sample(tchart, sample->tid, IOTYPE_RX,
+				 sample->time, ret);
+}
+
+static int
+process_enter_poll(struct timechart *tchart,
+		   struct perf_evsel *evsel,
+		   struct perf_sample *sample)
+{
+	long fd = perf_evsel__intval(evsel, sample, "fd");
+	return pid_begin_io_sample(tchart, sample->tid, IOTYPE_POLL,
+				   sample->time, fd);
+}
+
+static int
+process_exit_poll(struct timechart *tchart,
+		  struct perf_evsel *evsel,
+		  struct perf_sample *sample)
+{
+	long ret = perf_evsel__intval(evsel, sample, "ret");
+	return pid_end_io_sample(tchart, sample->tid, IOTYPE_POLL,
+				 sample->time, ret);
+}
+
+/*
+ * Sort the pid datastructure
+ */
+static void sort_pids(struct timechart *tchart)
+{
+	struct per_pid *new_list, *p, *cursor, *prev;
+	/* sort by ppid first, then by pid, lowest to highest */
+
+	new_list = NULL;
+
+	while (tchart->all_data) {
+		p = tchart->all_data;
+		tchart->all_data = p->next;
+		p->next = NULL;
+
+		if (new_list == NULL) {
+			new_list = p;
+			p->next = NULL;
+			continue;
+		}
+		prev = NULL;
+		cursor = new_list;
+		while (cursor) {
+			if (cursor->ppid > p->ppid ||
+				(cursor->ppid == p->ppid && cursor->pid > p->pid)) {
+				/* must insert before */
+				if (prev) {
+					p->next = prev->next;
+					prev->next = p;
+					cursor = NULL;
+					continue;
+				} else {
+					p->next = new_list;
+					new_list = p;
+					cursor = NULL;
+					continue;
+				}
+			}
+
+			prev = cursor;
+			cursor = cursor->next;
+			if (!cursor)
+				prev->next = p;
+		}
+	}
+	tchart->all_data = new_list;
+}
+
+
+static void draw_c_p_states(struct timechart *tchart)
+{
+	struct power_event *pwr;
+	pwr = tchart->power_events;
+
+	/*
+	 * two pass drawing so that the P state bars are on top of the C state blocks
+	 */
+	while (pwr) {
+		if (pwr->type == CSTATE)
+			svg_cstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state);
+		pwr = pwr->next;
+	}
+
+	pwr = tchart->power_events;
+	while (pwr) {
+		if (pwr->type == PSTATE) {
+			if (!pwr->state)
+				pwr->state = tchart->min_freq;
+			svg_pstate(pwr->cpu, pwr->start_time, pwr->end_time, pwr->state);
+		}
+		pwr = pwr->next;
+	}
+}
+
+static void draw_wakeups(struct timechart *tchart)
+{
+	struct wake_event *we;
+	struct per_pid *p;
+	struct per_pidcomm *c;
+
+	we = tchart->wake_events;
+	while (we) {
+		int from = 0, to = 0;
+		char *task_from = NULL, *task_to = NULL;
+
+		/* locate the column of the waker and wakee */
+		p = tchart->all_data;
+		while (p) {
+			if (p->pid == we->waker || p->pid == we->wakee) {
+				c = p->all;
+				while (c) {
+					if (c->Y && c->start_time <= we->time && c->end_time >= we->time) {
+						if (p->pid == we->waker && !from) {
+							from = c->Y;
+							task_from = strdup(c->comm);
+						}
+						if (p->pid == we->wakee && !to) {
+							to = c->Y;
+							task_to = strdup(c->comm);
+						}
+					}
+					c = c->next;
+				}
+				c = p->all;
+				while (c) {
+					if (p->pid == we->waker && !from) {
+						from = c->Y;
+						task_from = strdup(c->comm);
+					}
+					if (p->pid == we->wakee && !to) {
+						to = c->Y;
+						task_to = strdup(c->comm);
+					}
+					c = c->next;
+				}
+			}
+			p = p->next;
+		}
+
+		if (!task_from) {
+			task_from = malloc(40);
+			sprintf(task_from, "[%i]", we->waker);
+		}
+		if (!task_to) {
+			task_to = malloc(40);
+			sprintf(task_to, "[%i]", we->wakee);
+		}
+
+		if (we->waker == -1)
+			svg_interrupt(we->time, to, we->backtrace);
+		else if (from && to && abs(from - to) == 1)
+			svg_wakeline(we->time, from, to, we->backtrace);
+		else
+			svg_partial_wakeline(we->time, from, task_from, to,
+					     task_to, we->backtrace);
+		we = we->next;
+
+		free(task_from);
+		free(task_to);
+	}
+}
+
+static void draw_cpu_usage(struct timechart *tchart)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	struct cpu_sample *sample;
+	p = tchart->all_data;
+	while (p) {
+		c = p->all;
+		while (c) {
+			sample = c->samples;
+			while (sample) {
+				if (sample->type == TYPE_RUNNING) {
+					svg_process(sample->cpu,
+						    sample->start_time,
+						    sample->end_time,
+						    p->pid,
+						    c->comm,
+						    sample->backtrace);
+				}
+
+				sample = sample->next;
+			}
+			c = c->next;
+		}
+		p = p->next;
+	}
+}
+
+static void draw_io_bars(struct timechart *tchart)
+{
+	const char *suf;
+	double bytes;
+	char comm[256];
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	struct io_sample *sample;
+	int Y = 1;
+
+	p = tchart->all_data;
+	while (p) {
+		c = p->all;
+		while (c) {
+			if (!c->display) {
+				c->Y = 0;
+				c = c->next;
+				continue;
+			}
+
+			svg_box(Y, c->start_time, c->end_time, "process3");
+			sample = c->io_samples;
+			for (sample = c->io_samples; sample; sample = sample->next) {
+				double h = (double)sample->bytes / c->max_bytes;
+
+				if (tchart->skip_eagain &&
+				    sample->err == -EAGAIN)
+					continue;
+
+				if (sample->err)
+					h = 1;
+
+				if (sample->type == IOTYPE_SYNC)
+					svg_fbox(Y,
+						sample->start_time,
+						sample->end_time,
+						1,
+						sample->err ? "error" : "sync",
+						sample->fd,
+						sample->err,
+						sample->merges);
+				else if (sample->type == IOTYPE_POLL)
+					svg_fbox(Y,
+						sample->start_time,
+						sample->end_time,
+						1,
+						sample->err ? "error" : "poll",
+						sample->fd,
+						sample->err,
+						sample->merges);
+				else if (sample->type == IOTYPE_READ)
+					svg_ubox(Y,
+						sample->start_time,
+						sample->end_time,
+						h,
+						sample->err ? "error" : "disk",
+						sample->fd,
+						sample->err,
+						sample->merges);
+				else if (sample->type == IOTYPE_WRITE)
+					svg_lbox(Y,
+						sample->start_time,
+						sample->end_time,
+						h,
+						sample->err ? "error" : "disk",
+						sample->fd,
+						sample->err,
+						sample->merges);
+				else if (sample->type == IOTYPE_RX)
+					svg_ubox(Y,
+						sample->start_time,
+						sample->end_time,
+						h,
+						sample->err ? "error" : "net",
+						sample->fd,
+						sample->err,
+						sample->merges);
+				else if (sample->type == IOTYPE_TX)
+					svg_lbox(Y,
+						sample->start_time,
+						sample->end_time,
+						h,
+						sample->err ? "error" : "net",
+						sample->fd,
+						sample->err,
+						sample->merges);
+			}
+
+			suf = "";
+			bytes = c->total_bytes;
+			if (bytes > 1024) {
+				bytes = bytes / 1024;
+				suf = "K";
+			}
+			if (bytes > 1024) {
+				bytes = bytes / 1024;
+				suf = "M";
+			}
+			if (bytes > 1024) {
+				bytes = bytes / 1024;
+				suf = "G";
+			}
+
+
+			sprintf(comm, "%s:%i (%3.1f %sbytes)", c->comm ?: "", p->pid, bytes, suf);
+			svg_text(Y, c->start_time, comm);
+
+			c->Y = Y;
+			Y++;
+			c = c->next;
+		}
+		p = p->next;
+	}
+}
+
+static void draw_process_bars(struct timechart *tchart)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	struct cpu_sample *sample;
+	int Y = 0;
+
+	Y = 2 * tchart->numcpus + 2;
+
+	p = tchart->all_data;
+	while (p) {
+		c = p->all;
+		while (c) {
+			if (!c->display) {
+				c->Y = 0;
+				c = c->next;
+				continue;
+			}
+
+			svg_box(Y, c->start_time, c->end_time, "process");
+			sample = c->samples;
+			while (sample) {
+				if (sample->type == TYPE_RUNNING)
+					svg_running(Y, sample->cpu,
+						    sample->start_time,
+						    sample->end_time,
+						    sample->backtrace);
+				if (sample->type == TYPE_BLOCKED)
+					svg_blocked(Y, sample->cpu,
+						    sample->start_time,
+						    sample->end_time,
+						    sample->backtrace);
+				if (sample->type == TYPE_WAITING)
+					svg_waiting(Y, sample->cpu,
+						    sample->start_time,
+						    sample->end_time,
+						    sample->backtrace);
+				sample = sample->next;
+			}
+
+			if (c->comm) {
+				char comm[256];
+				if (c->total_time > 5000000000) /* 5 seconds */
+					sprintf(comm, "%s:%i (%2.2fs)", c->comm, p->pid, c->total_time / 1000000000.0);
+				else
+					sprintf(comm, "%s:%i (%3.1fms)", c->comm, p->pid, c->total_time / 1000000.0);
+
+				svg_text(Y, c->start_time, comm);
+			}
+			c->Y = Y;
+			Y++;
+			c = c->next;
+		}
+		p = p->next;
+	}
+}
+
+static void add_process_filter(const char *string)
+{
+	int pid = strtoull(string, NULL, 10);
+	struct process_filter *filt = malloc(sizeof(*filt));
+
+	if (!filt)
+		return;
+
+	filt->name = strdup(string);
+	filt->pid  = pid;
+	filt->next = process_filter;
+
+	process_filter = filt;
+}
+
+static int passes_filter(struct per_pid *p, struct per_pidcomm *c)
+{
+	struct process_filter *filt;
+	if (!process_filter)
+		return 1;
+
+	filt = process_filter;
+	while (filt) {
+		if (filt->pid && p->pid == filt->pid)
+			return 1;
+		if (strcmp(filt->name, c->comm) == 0)
+			return 1;
+		filt = filt->next;
+	}
+	return 0;
+}
+
+static int determine_display_tasks_filtered(struct timechart *tchart)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	int count = 0;
+
+	p = tchart->all_data;
+	while (p) {
+		p->display = 0;
+		if (p->start_time == 1)
+			p->start_time = tchart->first_time;
+
+		/* no exit marker, task kept running to the end */
+		if (p->end_time == 0)
+			p->end_time = tchart->last_time;
+
+		c = p->all;
+
+		while (c) {
+			c->display = 0;
+
+			if (c->start_time == 1)
+				c->start_time = tchart->first_time;
+
+			if (passes_filter(p, c)) {
+				c->display = 1;
+				p->display = 1;
+				count++;
+			}
+
+			if (c->end_time == 0)
+				c->end_time = tchart->last_time;
+
+			c = c->next;
+		}
+		p = p->next;
+	}
+	return count;
+}
+
+static int determine_display_tasks(struct timechart *tchart, u64 threshold)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	int count = 0;
+
+	p = tchart->all_data;
+	while (p) {
+		p->display = 0;
+		if (p->start_time == 1)
+			p->start_time = tchart->first_time;
+
+		/* no exit marker, task kept running to the end */
+		if (p->end_time == 0)
+			p->end_time = tchart->last_time;
+		if (p->total_time >= threshold)
+			p->display = 1;
+
+		c = p->all;
+
+		while (c) {
+			c->display = 0;
+
+			if (c->start_time == 1)
+				c->start_time = tchart->first_time;
+
+			if (c->total_time >= threshold) {
+				c->display = 1;
+				count++;
+			}
+
+			if (c->end_time == 0)
+				c->end_time = tchart->last_time;
+
+			c = c->next;
+		}
+		p = p->next;
+	}
+	return count;
+}
+
+static int determine_display_io_tasks(struct timechart *timechart, u64 threshold)
+{
+	struct per_pid *p;
+	struct per_pidcomm *c;
+	int count = 0;
+
+	p = timechart->all_data;
+	while (p) {
+		/* no exit marker, task kept running to the end */
+		if (p->end_time == 0)
+			p->end_time = timechart->last_time;
+
+		c = p->all;
+
+		while (c) {
+			c->display = 0;
+
+			if (c->total_bytes >= threshold) {
+				c->display = 1;
+				count++;
+			}
+
+			if (c->end_time == 0)
+				c->end_time = timechart->last_time;
+
+			c = c->next;
+		}
+		p = p->next;
+	}
+	return count;
+}
+
+#define BYTES_THRESH (1 * 1024 * 1024)
+#define TIME_THRESH 10000000
+
+static void write_svg_file(struct timechart *tchart, const char *filename)
+{
+	u64 i;
+	int count;
+	int thresh = tchart->io_events ? BYTES_THRESH : TIME_THRESH;
+
+	if (tchart->power_only)
+		tchart->proc_num = 0;
+
+	/* We'd like to show at least proc_num tasks;
+	 * be less picky if we have fewer */
+	do {
+		if (process_filter)
+			count = determine_display_tasks_filtered(tchart);
+		else if (tchart->io_events)
+			count = determine_display_io_tasks(tchart, thresh);
+		else
+			count = determine_display_tasks(tchart, thresh);
+		thresh /= 10;
+	} while (!process_filter && thresh && count < tchart->proc_num);
+
+	if (!tchart->proc_num)
+		count = 0;
+
+	if (tchart->io_events) {
+		open_svg(filename, 0, count, tchart->first_time, tchart->last_time);
+
+		svg_time_grid(0.5);
+		svg_io_legenda();
+
+		draw_io_bars(tchart);
+	} else {
+		open_svg(filename, tchart->numcpus, count, tchart->first_time, tchart->last_time);
+
+		svg_time_grid(0);
+
+		svg_legenda();
+
+		for (i = 0; i < tchart->numcpus; i++)
+			svg_cpu_box(i, tchart->max_freq, tchart->turbo_frequency);
+
+		draw_cpu_usage(tchart);
+		if (tchart->proc_num)
+			draw_process_bars(tchart);
+		if (!tchart->tasks_only)
+			draw_c_p_states(tchart);
+		if (tchart->proc_num)
+			draw_wakeups(tchart);
+	}
+
+	svg_close();
+}
+
+static int process_header(struct perf_file_section *section __maybe_unused,
+			  struct perf_header *ph,
+			  int feat,
+			  int fd __maybe_unused,
+			  void *data)
+{
+	struct timechart *tchart = data;
+
+	switch (feat) {
+	case HEADER_NRCPUS:
+		tchart->numcpus = ph->env.nr_cpus_avail;
+		break;
+
+	case HEADER_CPU_TOPOLOGY:
+		if (!tchart->topology)
+			break;
+
+		if (svg_build_topology_map(ph->env.sibling_cores,
+					   ph->env.nr_sibling_cores,
+					   ph->env.sibling_threads,
+					   ph->env.nr_sibling_threads))
+			fprintf(stderr, "problem building topology\n");
+		break;
+
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static int __cmd_timechart(struct timechart *tchart, const char *output_name)
+{
+	const struct perf_evsel_str_handler power_tracepoints[] = {
+		{ "power:cpu_idle",		process_sample_cpu_idle },
+		{ "power:cpu_frequency",	process_sample_cpu_frequency },
+		{ "sched:sched_wakeup",		process_sample_sched_wakeup },
+		{ "sched:sched_switch",		process_sample_sched_switch },
+#ifdef SUPPORT_OLD_POWER_EVENTS
+		{ "power:power_start",		process_sample_power_start },
+		{ "power:power_end",		process_sample_power_end },
+		{ "power:power_frequency",	process_sample_power_frequency },
+#endif
+
+		{ "syscalls:sys_enter_read",		process_enter_read },
+		{ "syscalls:sys_enter_pread64",		process_enter_read },
+		{ "syscalls:sys_enter_readv",		process_enter_read },
+		{ "syscalls:sys_enter_preadv",		process_enter_read },
+		{ "syscalls:sys_enter_write",		process_enter_write },
+		{ "syscalls:sys_enter_pwrite64",	process_enter_write },
+		{ "syscalls:sys_enter_writev",		process_enter_write },
+		{ "syscalls:sys_enter_pwritev",		process_enter_write },
+		{ "syscalls:sys_enter_sync",		process_enter_sync },
+		{ "syscalls:sys_enter_sync_file_range",	process_enter_sync },
+		{ "syscalls:sys_enter_fsync",		process_enter_sync },
+		{ "syscalls:sys_enter_msync",		process_enter_sync },
+		{ "syscalls:sys_enter_recvfrom",	process_enter_rx },
+		{ "syscalls:sys_enter_recvmmsg",	process_enter_rx },
+		{ "syscalls:sys_enter_recvmsg",		process_enter_rx },
+		{ "syscalls:sys_enter_sendto",		process_enter_tx },
+		{ "syscalls:sys_enter_sendmsg",		process_enter_tx },
+		{ "syscalls:sys_enter_sendmmsg",	process_enter_tx },
+		{ "syscalls:sys_enter_epoll_pwait",	process_enter_poll },
+		{ "syscalls:sys_enter_epoll_wait",	process_enter_poll },
+		{ "syscalls:sys_enter_poll",		process_enter_poll },
+		{ "syscalls:sys_enter_ppoll",		process_enter_poll },
+		{ "syscalls:sys_enter_pselect6",	process_enter_poll },
+		{ "syscalls:sys_enter_select",		process_enter_poll },
+
+		{ "syscalls:sys_exit_read",		process_exit_read },
+		{ "syscalls:sys_exit_pread64",		process_exit_read },
+		{ "syscalls:sys_exit_readv",		process_exit_read },
+		{ "syscalls:sys_exit_preadv",		process_exit_read },
+		{ "syscalls:sys_exit_write",		process_exit_write },
+		{ "syscalls:sys_exit_pwrite64",		process_exit_write },
+		{ "syscalls:sys_exit_writev",		process_exit_write },
+		{ "syscalls:sys_exit_pwritev",		process_exit_write },
+		{ "syscalls:sys_exit_sync",		process_exit_sync },
+		{ "syscalls:sys_exit_sync_file_range",	process_exit_sync },
+		{ "syscalls:sys_exit_fsync",		process_exit_sync },
+		{ "syscalls:sys_exit_msync",		process_exit_sync },
+		{ "syscalls:sys_exit_recvfrom",		process_exit_rx },
+		{ "syscalls:sys_exit_recvmmsg",		process_exit_rx },
+		{ "syscalls:sys_exit_recvmsg",		process_exit_rx },
+		{ "syscalls:sys_exit_sendto",		process_exit_tx },
+		{ "syscalls:sys_exit_sendmsg",		process_exit_tx },
+		{ "syscalls:sys_exit_sendmmsg",		process_exit_tx },
+		{ "syscalls:sys_exit_epoll_pwait",	process_exit_poll },
+		{ "syscalls:sys_exit_epoll_wait",	process_exit_poll },
+		{ "syscalls:sys_exit_poll",		process_exit_poll },
+		{ "syscalls:sys_exit_ppoll",		process_exit_poll },
+		{ "syscalls:sys_exit_pselect6",		process_exit_poll },
+		{ "syscalls:sys_exit_select",		process_exit_poll },
+	};
+	struct perf_data_file file = {
+		.path = input_name,
+		.mode = PERF_DATA_MODE_READ,
+		.force = tchart->force,
+	};
+
+	struct perf_session *session = perf_session__new(&file, false,
+							 &tchart->tool);
+	int ret = -EINVAL;
+
+	if (session == NULL)
+		return -1;
+
+	symbol__init(&session->header.env);
+
+	(void)perf_header__process_sections(&session->header,
+					    perf_data_file__fd(session->file),
+					    tchart,
+					    process_header);
+
+	if (!perf_session__has_traces(session, "timechart record"))
+		goto out_delete;
+
+	if (perf_session__set_tracepoints_handlers(session,
+						   power_tracepoints)) {
+		pr_err("Initializing session tracepoint handlers failed\n");
+		goto out_delete;
+	}
+
+	ret = perf_session__process_events(session);
+	if (ret)
+		goto out_delete;
+
+	end_sample_processing(tchart);
+
+	sort_pids(tchart);
+
+	write_svg_file(tchart, output_name);
+
+	pr_info("Written %2.1f seconds of trace to %s.\n",
+		(tchart->last_time - tchart->first_time) / 1000000000.0, output_name);
+out_delete:
+	perf_session__delete(session);
+	return ret;
+}
+
+static int timechart__io_record(int argc, const char **argv)
+{
+	unsigned int rec_argc, i;
+	const char **rec_argv;
+	const char **p;
+	char *filter = NULL;
+
+	const char * const common_args[] = {
+		"record", "-a", "-R", "-c", "1",
+	};
+	unsigned int common_args_nr = ARRAY_SIZE(common_args);
+
+	const char * const disk_events[] = {
+		"syscalls:sys_enter_read",
+		"syscalls:sys_enter_pread64",
+		"syscalls:sys_enter_readv",
+		"syscalls:sys_enter_preadv",
+		"syscalls:sys_enter_write",
+		"syscalls:sys_enter_pwrite64",
+		"syscalls:sys_enter_writev",
+		"syscalls:sys_enter_pwritev",
+		"syscalls:sys_enter_sync",
+		"syscalls:sys_enter_sync_file_range",
+		"syscalls:sys_enter_fsync",
+		"syscalls:sys_enter_msync",
+
+		"syscalls:sys_exit_read",
+		"syscalls:sys_exit_pread64",
+		"syscalls:sys_exit_readv",
+		"syscalls:sys_exit_preadv",
+		"syscalls:sys_exit_write",
+		"syscalls:sys_exit_pwrite64",
+		"syscalls:sys_exit_writev",
+		"syscalls:sys_exit_pwritev",
+		"syscalls:sys_exit_sync",
+		"syscalls:sys_exit_sync_file_range",
+		"syscalls:sys_exit_fsync",
+		"syscalls:sys_exit_msync",
+	};
+	unsigned int disk_events_nr = ARRAY_SIZE(disk_events);
+
+	const char * const net_events[] = {
+		"syscalls:sys_enter_recvfrom",
+		"syscalls:sys_enter_recvmmsg",
+		"syscalls:sys_enter_recvmsg",
+		"syscalls:sys_enter_sendto",
+		"syscalls:sys_enter_sendmsg",
+		"syscalls:sys_enter_sendmmsg",
+
+		"syscalls:sys_exit_recvfrom",
+		"syscalls:sys_exit_recvmmsg",
+		"syscalls:sys_exit_recvmsg",
+		"syscalls:sys_exit_sendto",
+		"syscalls:sys_exit_sendmsg",
+		"syscalls:sys_exit_sendmmsg",
+	};
+	unsigned int net_events_nr = ARRAY_SIZE(net_events);
+
+	const char * const poll_events[] = {
+		"syscalls:sys_enter_epoll_pwait",
+		"syscalls:sys_enter_epoll_wait",
+		"syscalls:sys_enter_poll",
+		"syscalls:sys_enter_ppoll",
+		"syscalls:sys_enter_pselect6",
+		"syscalls:sys_enter_select",
+
+		"syscalls:sys_exit_epoll_pwait",
+		"syscalls:sys_exit_epoll_wait",
+		"syscalls:sys_exit_poll",
+		"syscalls:sys_exit_ppoll",
+		"syscalls:sys_exit_pselect6",
+		"syscalls:sys_exit_select",
+	};
+	unsigned int poll_events_nr = ARRAY_SIZE(poll_events);
+
+	rec_argc = common_args_nr +
+		disk_events_nr * 4 +
+		net_events_nr * 4 +
+		poll_events_nr * 4 +
+		argc;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+	if (rec_argv == NULL)
+		return -ENOMEM;
+
+	if (asprintf(&filter, "common_pid != %d", getpid()) < 0)
+		return -ENOMEM;
+
+	p = rec_argv;
+	for (i = 0; i < common_args_nr; i++)
+		*p++ = strdup(common_args[i]);
+
+	for (i = 0; i < disk_events_nr; i++) {
+		if (!is_valid_tracepoint(disk_events[i])) {
+			rec_argc -= 4;
+			continue;
+		}
+
+		*p++ = "-e";
+		*p++ = strdup(disk_events[i]);
+		*p++ = "--filter";
+		*p++ = filter;
+	}
+	for (i = 0; i < net_events_nr; i++) {
+		if (!is_valid_tracepoint(net_events[i])) {
+			rec_argc -= 4;
+			continue;
+		}
+
+		*p++ = "-e";
+		*p++ = strdup(net_events[i]);
+		*p++ = "--filter";
+		*p++ = filter;
+	}
+	for (i = 0; i < poll_events_nr; i++) {
+		if (!is_valid_tracepoint(poll_events[i])) {
+			rec_argc -= 4;
+			continue;
+		}
+
+		*p++ = "-e";
+		*p++ = strdup(poll_events[i]);
+		*p++ = "--filter";
+		*p++ = filter;
+	}
+
+	for (i = 0; i < (unsigned int)argc; i++)
+		*p++ = argv[i];
+
+	return cmd_record(rec_argc, rec_argv, NULL);
+}
+
+
+static int timechart__record(struct timechart *tchart, int argc, const char **argv)
+{
+	unsigned int rec_argc, i, j;
+	const char **rec_argv;
+	const char **p;
+	unsigned int record_elems;
+
+	const char * const common_args[] = {
+		"record", "-a", "-R", "-c", "1",
+	};
+	unsigned int common_args_nr = ARRAY_SIZE(common_args);
+
+	const char * const backtrace_args[] = {
+		"-g",
+	};
+	unsigned int backtrace_args_no = ARRAY_SIZE(backtrace_args);
+
+	const char * const power_args[] = {
+		"-e", "power:cpu_frequency",
+		"-e", "power:cpu_idle",
+	};
+	unsigned int power_args_nr = ARRAY_SIZE(power_args);
+
+	const char * const old_power_args[] = {
+#ifdef SUPPORT_OLD_POWER_EVENTS
+		"-e", "power:power_start",
+		"-e", "power:power_end",
+		"-e", "power:power_frequency",
+#endif
+	};
+	unsigned int old_power_args_nr = ARRAY_SIZE(old_power_args);
+
+	const char * const tasks_args[] = {
+		"-e", "sched:sched_wakeup",
+		"-e", "sched:sched_switch",
+	};
+	unsigned int tasks_args_nr = ARRAY_SIZE(tasks_args);
+
+#ifdef SUPPORT_OLD_POWER_EVENTS
+	if (!is_valid_tracepoint("power:cpu_idle") &&
+	    is_valid_tracepoint("power:power_start")) {
+		use_old_power_events = 1;
+		power_args_nr = 0;
+	} else {
+		old_power_args_nr = 0;
+	}
+#endif
+
+	if (tchart->power_only)
+		tasks_args_nr = 0;
+
+	if (tchart->tasks_only) {
+		power_args_nr = 0;
+		old_power_args_nr = 0;
+	}
+
+	if (!tchart->with_backtrace)
+		backtrace_args_no = 0;
+
+	record_elems = common_args_nr + tasks_args_nr +
+		power_args_nr + old_power_args_nr + backtrace_args_no;
+
+	rec_argc = record_elems + argc;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+	if (rec_argv == NULL)
+		return -ENOMEM;
+
+	p = rec_argv;
+	for (i = 0; i < common_args_nr; i++)
+		*p++ = strdup(common_args[i]);
+
+	for (i = 0; i < backtrace_args_no; i++)
+		*p++ = strdup(backtrace_args[i]);
+
+	for (i = 0; i < tasks_args_nr; i++)
+		*p++ = strdup(tasks_args[i]);
+
+	for (i = 0; i < power_args_nr; i++)
+		*p++ = strdup(power_args[i]);
+
+	for (i = 0; i < old_power_args_nr; i++)
+		*p++ = strdup(old_power_args[i]);
+
+	for (j = 0; j < (unsigned int)argc; j++)
+		*p++ = argv[j];
+
+	return cmd_record(rec_argc, rec_argv, NULL);
+}
+
+static int
+parse_process(const struct option *opt __maybe_unused, const char *arg,
+	      int __maybe_unused unset)
+{
+	if (arg)
+		add_process_filter(arg);
+	return 0;
+}
+
+static int
+parse_highlight(const struct option *opt __maybe_unused, const char *arg,
+		int __maybe_unused unset)
+{
+	unsigned long duration = strtoul(arg, NULL, 0);
+
+	if (svg_highlight || svg_highlight_name)
+		return -1;
+
+	if (duration)
+		svg_highlight = duration;
+	else
+		svg_highlight_name = strdup(arg);
+
+	return 0;
+}
+
+static int
+parse_time(const struct option *opt, const char *arg, int __maybe_unused unset)
+{
+	char unit = 'n';
+	u64 *value = opt->value;
+
+	if (sscanf(arg, "%" PRIu64 "%cs", value, &unit) > 0) {
+		switch (unit) {
+		case 'm':
+			*value *= 1000000;
+			break;
+		case 'u':
+			*value *= 1000;
+			break;
+		case 'n':
+			break;
+		default:
+			return -1;
+		}
+	}
+
+	return 0;
+}
+
+int cmd_timechart(int argc, const char **argv,
+		  const char *prefix __maybe_unused)
+{
+	struct timechart tchart = {
+		.tool = {
+			.comm		 = process_comm_event,
+			.fork		 = process_fork_event,
+			.exit		 = process_exit_event,
+			.sample		 = process_sample_event,
+			.ordered_events	 = true,
+		},
+		.proc_num = 15,
+		.min_time = 1000000,
+		.merge_dist = 1000,
+	};
+	const char *output_name = "output.svg";
+	const struct option timechart_options[] = {
+	OPT_STRING('i', "input", &input_name, "file", "input file name"),
+	OPT_STRING('o', "output", &output_name, "file", "output file name"),
+	OPT_INTEGER('w', "width", &svg_page_width, "page width"),
+	OPT_CALLBACK(0, "highlight", NULL, "duration or task name",
+		      "highlight tasks. Pass duration in ns or process name.",
+		       parse_highlight),
+	OPT_BOOLEAN('P', "power-only", &tchart.power_only, "output power data only"),
+	OPT_BOOLEAN('T', "tasks-only", &tchart.tasks_only,
+		    "output processes data only"),
+	OPT_CALLBACK('p', "process", NULL, "process",
+		      "process selector. Pass a pid or process name.",
+		       parse_process),
+	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
+		    "Look for files with symbols relative to this directory"),
+	OPT_INTEGER('n', "proc-num", &tchart.proc_num,
+		    "min. number of tasks to print"),
+	OPT_BOOLEAN('t', "topology", &tchart.topology,
+		    "sort CPUs according to topology"),
+	OPT_BOOLEAN(0, "io-skip-eagain", &tchart.skip_eagain,
+		    "skip EAGAIN errors"),
+	OPT_CALLBACK(0, "io-min-time", &tchart.min_time, "time",
+		     "all IO faster than min-time will visually appear longer",
+		     parse_time),
+	OPT_CALLBACK(0, "io-merge-dist", &tchart.merge_dist, "time",
+		     "merge events that are merge-dist us apart",
+		     parse_time),
+	OPT_BOOLEAN('f', "force", &tchart.force, "don't complain, do it"),
+	OPT_END()
+	};
+	const char * const timechart_subcommands[] = { "record", NULL };
+	const char *timechart_usage[] = {
+		"perf timechart [<options>] {record}",
+		NULL
+	};
+
+	const struct option timechart_record_options[] = {
+	OPT_BOOLEAN('P', "power-only", &tchart.power_only, "output power data only"),
+	OPT_BOOLEAN('T', "tasks-only", &tchart.tasks_only,
+		    "output processes data only"),
+	OPT_BOOLEAN('I', "io-only", &tchart.io_only,
+		    "record only IO data"),
+	OPT_BOOLEAN('g', "callchain", &tchart.with_backtrace, "record callchain"),
+	OPT_END()
+	};
+	const char * const timechart_record_usage[] = {
+		"perf timechart record [<options>]",
+		NULL
+	};
+	argc = parse_options_subcommand(argc, argv, timechart_options, timechart_subcommands,
+			timechart_usage, PARSE_OPT_STOP_AT_NON_OPTION);
+
+	if (tchart.power_only && tchart.tasks_only) {
+		pr_err("-P and -T options cannot be used at the same time.\n");
+		return -1;
+	}
+
+	if (argc && !strncmp(argv[0], "rec", 3)) {
+		argc = parse_options(argc, argv, timechart_record_options,
+				     timechart_record_usage,
+				     PARSE_OPT_STOP_AT_NON_OPTION);
+
+		if (tchart.power_only && tchart.tasks_only) {
+			pr_err("-P and -T options cannot be used at the same time.\n");
+			return -1;
+		}
+
+		if (tchart.io_only)
+			return timechart__io_record(argc, argv);
+		else
+			return timechart__record(&tchart, argc, argv);
+	} else if (argc)
+		usage_with_options(timechart_usage, timechart_options);
+
+	setup_pager();
+
+	return __cmd_timechart(&tchart, output_name);
+}
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
new file mode 100644
index 000000000..6a4d5d41c
--- /dev/null
+++ b/tools/perf/builtin-top.c
@@ -0,0 +1,1273 @@
+/*
+ * builtin-top.c
+ *
+ * Builtin top command: Display a continuously updated profile of
+ * any workload, CPU or specific PID.
+ *
+ * Copyright (C) 2008, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
+ *		 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Improvements and fixes by:
+ *
+ *   Arjan van de Ven <arjan@linux.intel.com>
+ *   Yanmin Zhang <yanmin.zhang@intel.com>
+ *   Wu Fengguang <fengguang.wu@intel.com>
+ *   Mike Galbraith <efault@gmx.de>
+ *   Paul Mackerras <paulus@samba.org>
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+#include "builtin.h"
+
+#include "perf.h"
+
+#include "util/annotate.h"
+#include "util/cache.h"
+#include "util/color.h"
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/machine.h"
+#include "util/session.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/thread_map.h"
+#include "util/top.h"
+#include "util/util.h"
+#include <linux/rbtree.h>
+#include "util/parse-options.h"
+#include "util/parse-events.h"
+#include "util/cpumap.h"
+#include "util/xyarray.h"
+#include "util/sort.h"
+#include "util/intlist.h"
+#include "arch/common.h"
+
+#include "util/debug.h"
+
+#include <assert.h>
+#include <elf.h>
+#include <fcntl.h>
+
+#include <stdio.h>
+#include <termios.h>
+#include <unistd.h>
+#include <inttypes.h>
+
+#include <errno.h>
+#include <time.h>
+#include <sched.h>
+
+#include <sys/syscall.h>
+#include <sys/ioctl.h>
+#include <poll.h>
+#include <sys/prctl.h>
+#include <sys/wait.h>
+#include <sys/uio.h>
+#include <sys/utsname.h>
+#include <sys/mman.h>
+
+#include <linux/types.h>
+
+static volatile int done;
+
+#define HEADER_LINE_NR  5
+
+static void perf_top__update_print_entries(struct perf_top *top)
+{
+	top->print_entries = top->winsize.ws_row - HEADER_LINE_NR;
+}
+
+static void perf_top__sig_winch(int sig __maybe_unused,
+				siginfo_t *info __maybe_unused, void *arg)
+{
+	struct perf_top *top = arg;
+
+	get_term_dimensions(&top->winsize);
+	perf_top__update_print_entries(top);
+}
+
+static int perf_top__parse_source(struct perf_top *top, struct hist_entry *he)
+{
+	struct symbol *sym;
+	struct annotation *notes;
+	struct map *map;
+	int err = -1;
+
+	if (!he || !he->ms.sym)
+		return -1;
+
+	sym = he->ms.sym;
+	map = he->ms.map;
+
+	/*
+	 * We can't annotate with just /proc/kallsyms
+	 */
+	if (map->dso->symtab_type == DSO_BINARY_TYPE__KALLSYMS &&
+	    !dso__is_kcore(map->dso)) {
+		pr_err("Can't annotate %s: No vmlinux file was found in the "
+		       "path\n", sym->name);
+		sleep(1);
+		return -1;
+	}
+
+	notes = symbol__annotation(sym);
+	if (notes->src != NULL) {
+		pthread_mutex_lock(&notes->lock);
+		goto out_assign;
+	}
+
+	pthread_mutex_lock(&notes->lock);
+
+	if (symbol__alloc_hist(sym) < 0) {
+		pthread_mutex_unlock(&notes->lock);
+		pr_err("Not enough memory for annotating '%s' symbol!\n",
+		       sym->name);
+		sleep(1);
+		return err;
+	}
+
+	err = symbol__annotate(sym, map, 0);
+	if (err == 0) {
+out_assign:
+		top->sym_filter_entry = he;
+	}
+
+	pthread_mutex_unlock(&notes->lock);
+	return err;
+}
+
+static void __zero_source_counters(struct hist_entry *he)
+{
+	struct symbol *sym = he->ms.sym;
+	symbol__annotate_zero_histograms(sym);
+}
+
+static void ui__warn_map_erange(struct map *map, struct symbol *sym, u64 ip)
+{
+	struct utsname uts;
+	int err = uname(&uts);
+
+	ui__warning("Out of bounds address found:\n\n"
+		    "Addr:   %" PRIx64 "\n"
+		    "DSO:    %s %c\n"
+		    "Map:    %" PRIx64 "-%" PRIx64 "\n"
+		    "Symbol: %" PRIx64 "-%" PRIx64 " %c %s\n"
+		    "Arch:   %s\n"
+		    "Kernel: %s\n"
+		    "Tools:  %s\n\n"
+		    "Not all samples will be on the annotation output.\n\n"
+		    "Please report to linux-kernel@vger.kernel.org\n",
+		    ip, map->dso->long_name, dso__symtab_origin(map->dso),
+		    map->start, map->end, sym->start, sym->end,
+		    sym->binding == STB_GLOBAL ? 'g' :
+		    sym->binding == STB_LOCAL  ? 'l' : 'w', sym->name,
+		    err ? "[unknown]" : uts.machine,
+		    err ? "[unknown]" : uts.release, perf_version_string);
+	if (use_browser <= 0)
+		sleep(5);
+
+	map->erange_warned = true;
+}
+
+static void perf_top__record_precise_ip(struct perf_top *top,
+					struct hist_entry *he,
+					int counter, u64 ip)
+{
+	struct annotation *notes;
+	struct symbol *sym;
+	int err = 0;
+
+	if (he == NULL || he->ms.sym == NULL ||
+	    ((top->sym_filter_entry == NULL ||
+	      top->sym_filter_entry->ms.sym != he->ms.sym) && use_browser != 1))
+		return;
+
+	sym = he->ms.sym;
+	notes = symbol__annotation(sym);
+
+	if (pthread_mutex_trylock(&notes->lock))
+		return;
+
+	ip = he->ms.map->map_ip(he->ms.map, ip);
+
+	if (ui__has_annotation())
+		err = hist_entry__inc_addr_samples(he, counter, ip);
+
+	pthread_mutex_unlock(&notes->lock);
+
+	/*
+	 * This function is now called with he->hists->lock held.
+	 * Release it before going to sleep.
+	 */
+	pthread_mutex_unlock(&he->hists->lock);
+
+	if (err == -ERANGE && !he->ms.map->erange_warned)
+		ui__warn_map_erange(he->ms.map, sym, ip);
+	else if (err == -ENOMEM) {
+		pr_err("Not enough memory for annotating '%s' symbol!\n",
+		       sym->name);
+		sleep(1);
+	}
+
+	pthread_mutex_lock(&he->hists->lock);
+}
+
+static void perf_top__show_details(struct perf_top *top)
+{
+	struct hist_entry *he = top->sym_filter_entry;
+	struct annotation *notes;
+	struct symbol *symbol;
+	int more;
+
+	if (!he)
+		return;
+
+	symbol = he->ms.sym;
+	notes = symbol__annotation(symbol);
+
+	pthread_mutex_lock(&notes->lock);
+
+	if (notes->src == NULL)
+		goto out_unlock;
+
+	printf("Showing %s for %s\n", perf_evsel__name(top->sym_evsel), symbol->name);
+	printf("  Events  Pcnt (>=%d%%)\n", top->sym_pcnt_filter);
+
+	more = symbol__annotate_printf(symbol, he->ms.map, top->sym_evsel,
+				       0, top->sym_pcnt_filter, top->print_entries, 4);
+	if (top->zero)
+		symbol__annotate_zero_histogram(symbol, top->sym_evsel->idx);
+	else
+		symbol__annotate_decay_histogram(symbol, top->sym_evsel->idx);
+	if (more != 0)
+		printf("%d lines not displayed, maybe increase display entries [e]\n", more);
+out_unlock:
+	pthread_mutex_unlock(&notes->lock);
+}
+
+static void perf_top__print_sym_table(struct perf_top *top)
+{
+	char bf[160];
+	int printed = 0;
+	const int win_width = top->winsize.ws_col - 1;
+	struct hists *hists = evsel__hists(top->sym_evsel);
+
+	puts(CONSOLE_CLEAR);
+
+	perf_top__header_snprintf(top, bf, sizeof(bf));
+	printf("%s\n", bf);
+
+	perf_top__reset_sample_counters(top);
+
+	printf("%-*.*s\n", win_width, win_width, graph_dotted_line);
+
+	if (hists->stats.nr_lost_warned !=
+	    hists->stats.nr_events[PERF_RECORD_LOST]) {
+		hists->stats.nr_lost_warned =
+			      hists->stats.nr_events[PERF_RECORD_LOST];
+		color_fprintf(stdout, PERF_COLOR_RED,
+			      "WARNING: LOST %d chunks, Check IO/CPU overload",
+			      hists->stats.nr_lost_warned);
+		++printed;
+	}
+
+	if (top->sym_filter_entry) {
+		perf_top__show_details(top);
+		return;
+	}
+
+	if (top->zero) {
+		hists__delete_entries(hists);
+	} else {
+		hists__decay_entries(hists, top->hide_user_symbols,
+				     top->hide_kernel_symbols);
+	}
+
+	hists__collapse_resort(hists, NULL);
+	hists__output_resort(hists, NULL);
+
+	hists__output_recalc_col_len(hists, top->print_entries - printed);
+	putchar('\n');
+	hists__fprintf(hists, false, top->print_entries - printed, win_width,
+		       top->min_percent, stdout);
+}
+
+static void prompt_integer(int *target, const char *msg)
+{
+	char *buf = malloc(0), *p;
+	size_t dummy = 0;
+	int tmp;
+
+	fprintf(stdout, "\n%s: ", msg);
+	if (getline(&buf, &dummy, stdin) < 0)
+		return;
+
+	p = strchr(buf, '\n');
+	if (p)
+		*p = 0;
+
+	p = buf;
+	while(*p) {
+		if (!isdigit(*p))
+			goto out_free;
+		p++;
+	}
+	tmp = strtoul(buf, NULL, 10);
+	*target = tmp;
+out_free:
+	free(buf);
+}
+
+static void prompt_percent(int *target, const char *msg)
+{
+	int tmp = 0;
+
+	prompt_integer(&tmp, msg);
+	if (tmp >= 0 && tmp <= 100)
+		*target = tmp;
+}
+
+static void perf_top__prompt_symbol(struct perf_top *top, const char *msg)
+{
+	char *buf = malloc(0), *p;
+	struct hist_entry *syme = top->sym_filter_entry, *n, *found = NULL;
+	struct hists *hists = evsel__hists(top->sym_evsel);
+	struct rb_node *next;
+	size_t dummy = 0;
+
+	/* zero counters of active symbol */
+	if (syme) {
+		__zero_source_counters(syme);
+		top->sym_filter_entry = NULL;
+	}
+
+	fprintf(stdout, "\n%s: ", msg);
+	if (getline(&buf, &dummy, stdin) < 0)
+		goto out_free;
+
+	p = strchr(buf, '\n');
+	if (p)
+		*p = 0;
+
+	next = rb_first(&hists->entries);
+	while (next) {
+		n = rb_entry(next, struct hist_entry, rb_node);
+		if (n->ms.sym && !strcmp(buf, n->ms.sym->name)) {
+			found = n;
+			break;
+		}
+		next = rb_next(&n->rb_node);
+	}
+
+	if (!found) {
+		fprintf(stderr, "Sorry, %s is not active.\n", buf);
+		sleep(1);
+	} else
+		perf_top__parse_source(top, found);
+
+out_free:
+	free(buf);
+}
+
+static void perf_top__print_mapped_keys(struct perf_top *top)
+{
+	char *name = NULL;
+
+	if (top->sym_filter_entry) {
+		struct symbol *sym = top->sym_filter_entry->ms.sym;
+		name = sym->name;
+	}
+
+	fprintf(stdout, "\nMapped keys:\n");
+	fprintf(stdout, "\t[d]     display refresh delay.             \t(%d)\n", top->delay_secs);
+	fprintf(stdout, "\t[e]     display entries (lines).           \t(%d)\n", top->print_entries);
+
+	if (top->evlist->nr_entries > 1)
+		fprintf(stdout, "\t[E]     active event counter.              \t(%s)\n", perf_evsel__name(top->sym_evsel));
+
+	fprintf(stdout, "\t[f]     profile display filter (count).    \t(%d)\n", top->count_filter);
+
+	fprintf(stdout, "\t[F]     annotate display filter (percent). \t(%d%%)\n", top->sym_pcnt_filter);
+	fprintf(stdout, "\t[s]     annotate symbol.                   \t(%s)\n", name?: "NULL");
+	fprintf(stdout, "\t[S]     stop annotation.\n");
+
+	fprintf(stdout,
+		"\t[K]     hide kernel_symbols symbols.     \t(%s)\n",
+		top->hide_kernel_symbols ? "yes" : "no");
+	fprintf(stdout,
+		"\t[U]     hide user symbols.               \t(%s)\n",
+		top->hide_user_symbols ? "yes" : "no");
+	fprintf(stdout, "\t[z]     toggle sample zeroing.             \t(%d)\n", top->zero ? 1 : 0);
+	fprintf(stdout, "\t[qQ]    quit.\n");
+}
+
+static int perf_top__key_mapped(struct perf_top *top, int c)
+{
+	switch (c) {
+		case 'd':
+		case 'e':
+		case 'f':
+		case 'z':
+		case 'q':
+		case 'Q':
+		case 'K':
+		case 'U':
+		case 'F':
+		case 's':
+		case 'S':
+			return 1;
+		case 'E':
+			return top->evlist->nr_entries > 1 ? 1 : 0;
+		default:
+			break;
+	}
+
+	return 0;
+}
+
+static bool perf_top__handle_keypress(struct perf_top *top, int c)
+{
+	bool ret = true;
+
+	if (!perf_top__key_mapped(top, c)) {
+		struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
+		struct termios save;
+
+		perf_top__print_mapped_keys(top);
+		fprintf(stdout, "\nEnter selection, or unmapped key to continue: ");
+		fflush(stdout);
+
+		set_term_quiet_input(&save);
+
+		poll(&stdin_poll, 1, -1);
+		c = getc(stdin);
+
+		tcsetattr(0, TCSAFLUSH, &save);
+		if (!perf_top__key_mapped(top, c))
+			return ret;
+	}
+
+	switch (c) {
+		case 'd':
+			prompt_integer(&top->delay_secs, "Enter display delay");
+			if (top->delay_secs < 1)
+				top->delay_secs = 1;
+			break;
+		case 'e':
+			prompt_integer(&top->print_entries, "Enter display entries (lines)");
+			if (top->print_entries == 0) {
+				struct sigaction act = {
+					.sa_sigaction = perf_top__sig_winch,
+					.sa_flags     = SA_SIGINFO,
+				};
+				perf_top__sig_winch(SIGWINCH, NULL, top);
+				sigaction(SIGWINCH, &act, NULL);
+			} else {
+				signal(SIGWINCH, SIG_DFL);
+			}
+			break;
+		case 'E':
+			if (top->evlist->nr_entries > 1) {
+				/* Select 0 as the default event: */
+				int counter = 0;
+
+				fprintf(stderr, "\nAvailable events:");
+
+				evlist__for_each(top->evlist, top->sym_evsel)
+					fprintf(stderr, "\n\t%d %s", top->sym_evsel->idx, perf_evsel__name(top->sym_evsel));
+
+				prompt_integer(&counter, "Enter details event counter");
+
+				if (counter >= top->evlist->nr_entries) {
+					top->sym_evsel = perf_evlist__first(top->evlist);
+					fprintf(stderr, "Sorry, no such event, using %s.\n", perf_evsel__name(top->sym_evsel));
+					sleep(1);
+					break;
+				}
+				evlist__for_each(top->evlist, top->sym_evsel)
+					if (top->sym_evsel->idx == counter)
+						break;
+			} else
+				top->sym_evsel = perf_evlist__first(top->evlist);
+			break;
+		case 'f':
+			prompt_integer(&top->count_filter, "Enter display event count filter");
+			break;
+		case 'F':
+			prompt_percent(&top->sym_pcnt_filter,
+				       "Enter details display event filter (percent)");
+			break;
+		case 'K':
+			top->hide_kernel_symbols = !top->hide_kernel_symbols;
+			break;
+		case 'q':
+		case 'Q':
+			printf("exiting.\n");
+			if (top->dump_symtab)
+				perf_session__fprintf_dsos(top->session, stderr);
+			ret = false;
+			break;
+		case 's':
+			perf_top__prompt_symbol(top, "Enter details symbol");
+			break;
+		case 'S':
+			if (!top->sym_filter_entry)
+				break;
+			else {
+				struct hist_entry *syme = top->sym_filter_entry;
+
+				top->sym_filter_entry = NULL;
+				__zero_source_counters(syme);
+			}
+			break;
+		case 'U':
+			top->hide_user_symbols = !top->hide_user_symbols;
+			break;
+		case 'z':
+			top->zero = !top->zero;
+			break;
+		default:
+			break;
+	}
+
+	return ret;
+}
+
+static void perf_top__sort_new_samples(void *arg)
+{
+	struct perf_top *t = arg;
+	struct hists *hists;
+
+	perf_top__reset_sample_counters(t);
+
+	if (t->evlist->selected != NULL)
+		t->sym_evsel = t->evlist->selected;
+
+	hists = evsel__hists(t->sym_evsel);
+
+	if (t->zero) {
+		hists__delete_entries(hists);
+	} else {
+		hists__decay_entries(hists, t->hide_user_symbols,
+				     t->hide_kernel_symbols);
+	}
+
+	hists__collapse_resort(hists, NULL);
+	hists__output_resort(hists, NULL);
+}
+
+static void *display_thread_tui(void *arg)
+{
+	struct perf_evsel *pos;
+	struct perf_top *top = arg;
+	const char *help = "For a higher level overview, try: perf top --sort comm,dso";
+	struct hist_browser_timer hbt = {
+		.timer		= perf_top__sort_new_samples,
+		.arg		= top,
+		.refresh	= top->delay_secs,
+	};
+
+	perf_top__sort_new_samples(top);
+
+	/*
+	 * Initialize the uid_filter_str, in the future the TUI will allow
+	 * Zooming in/out UIDs. For now juse use whatever the user passed
+	 * via --uid.
+	 */
+	evlist__for_each(top->evlist, pos) {
+		struct hists *hists = evsel__hists(pos);
+		hists->uid_filter_str = top->record_opts.target.uid_str;
+	}
+
+	perf_evlist__tui_browse_hists(top->evlist, help, &hbt, top->min_percent,
+				      &top->session->header.env);
+
+	done = 1;
+	return NULL;
+}
+
+static void display_sig(int sig __maybe_unused)
+{
+	done = 1;
+}
+
+static void display_setup_sig(void)
+{
+	signal(SIGSEGV, display_sig);
+	signal(SIGFPE,  display_sig);
+	signal(SIGINT,  display_sig);
+	signal(SIGQUIT, display_sig);
+	signal(SIGTERM, display_sig);
+}
+
+static void *display_thread(void *arg)
+{
+	struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
+	struct termios save;
+	struct perf_top *top = arg;
+	int delay_msecs, c;
+
+	display_setup_sig();
+	pthread__unblock_sigwinch();
+repeat:
+	delay_msecs = top->delay_secs * 1000;
+	set_term_quiet_input(&save);
+	/* trash return*/
+	getc(stdin);
+
+	while (!done) {
+		perf_top__print_sym_table(top);
+		/*
+		 * Either timeout expired or we got an EINTR due to SIGWINCH,
+		 * refresh screen in both cases.
+		 */
+		switch (poll(&stdin_poll, 1, delay_msecs)) {
+		case 0:
+			continue;
+		case -1:
+			if (errno == EINTR)
+				continue;
+			/* Fall trhu */
+		default:
+			c = getc(stdin);
+			tcsetattr(0, TCSAFLUSH, &save);
+
+			if (perf_top__handle_keypress(top, c))
+				goto repeat;
+			done = 1;
+		}
+	}
+
+	tcsetattr(0, TCSAFLUSH, &save);
+	return NULL;
+}
+
+static int symbol_filter(struct map *map, struct symbol *sym)
+{
+	const char *name = sym->name;
+
+	if (!map->dso->kernel)
+		return 0;
+	/*
+	 * ppc64 uses function descriptors and appends a '.' to the
+	 * start of every instruction address. Remove it.
+	 */
+	if (name[0] == '.')
+		name++;
+
+	if (!strcmp(name, "_text") ||
+	    !strcmp(name, "_etext") ||
+	    !strcmp(name, "_sinittext") ||
+	    !strncmp("init_module", name, 11) ||
+	    !strncmp("cleanup_module", name, 14) ||
+	    strstr(name, "_text_start") ||
+	    strstr(name, "_text_end"))
+		return 1;
+
+	if (symbol__is_idle(sym))
+		sym->ignore = true;
+
+	return 0;
+}
+
+static int hist_iter__top_callback(struct hist_entry_iter *iter,
+				   struct addr_location *al, bool single,
+				   void *arg)
+{
+	struct perf_top *top = arg;
+	struct hist_entry *he = iter->he;
+	struct perf_evsel *evsel = iter->evsel;
+
+	if (sort__has_sym && single) {
+		u64 ip = al->addr;
+
+		if (al->map)
+			ip = al->map->unmap_ip(al->map, ip);
+
+		perf_top__record_precise_ip(top, he, evsel->idx, ip);
+	}
+
+	return 0;
+}
+
+static void perf_event__process_sample(struct perf_tool *tool,
+				       const union perf_event *event,
+				       struct perf_evsel *evsel,
+				       struct perf_sample *sample,
+				       struct machine *machine)
+{
+	struct perf_top *top = container_of(tool, struct perf_top, tool);
+	struct addr_location al;
+	int err;
+
+	if (!machine && perf_guest) {
+		static struct intlist *seen;
+
+		if (!seen)
+			seen = intlist__new(NULL);
+
+		if (!intlist__has_entry(seen, sample->pid)) {
+			pr_err("Can't find guest [%d]'s kernel information\n",
+				sample->pid);
+			intlist__add(seen, sample->pid);
+		}
+		return;
+	}
+
+	if (!machine) {
+		pr_err("%u unprocessable samples recorded.\r",
+		       top->session->evlist->stats.nr_unprocessable_samples++);
+		return;
+	}
+
+	if (event->header.misc & PERF_RECORD_MISC_EXACT_IP)
+		top->exact_samples++;
+
+	if (perf_event__preprocess_sample(event, machine, &al, sample) < 0)
+		return;
+
+	if (!top->kptr_restrict_warned &&
+	    symbol_conf.kptr_restrict &&
+	    al.cpumode == PERF_RECORD_MISC_KERNEL) {
+		ui__warning(
+"Kernel address maps (/proc/{kallsyms,modules}) are restricted.\n\n"
+"Check /proc/sys/kernel/kptr_restrict.\n\n"
+"Kernel%s samples will not be resolved.\n",
+			  al.map && !RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION]) ?
+			  " modules" : "");
+		if (use_browser <= 0)
+			sleep(5);
+		top->kptr_restrict_warned = true;
+	}
+
+	if (al.sym == NULL) {
+		const char *msg = "Kernel samples will not be resolved.\n";
+		/*
+		 * As we do lazy loading of symtabs we only will know if the
+		 * specified vmlinux file is invalid when we actually have a
+		 * hit in kernel space and then try to load it. So if we get
+		 * here and there are _no_ symbols in the DSO backing the
+		 * kernel map, bail out.
+		 *
+		 * We may never get here, for instance, if we use -K/
+		 * --hide-kernel-symbols, even if the user specifies an
+		 * invalid --vmlinux ;-)
+		 */
+		if (!top->kptr_restrict_warned && !top->vmlinux_warned &&
+		    al.map == machine->vmlinux_maps[MAP__FUNCTION] &&
+		    RB_EMPTY_ROOT(&al.map->dso->symbols[MAP__FUNCTION])) {
+			if (symbol_conf.vmlinux_name) {
+				char serr[256];
+				dso__strerror_load(al.map->dso, serr, sizeof(serr));
+				ui__warning("The %s file can't be used: %s\n%s",
+					    symbol_conf.vmlinux_name, serr, msg);
+			} else {
+				ui__warning("A vmlinux file was not found.\n%s",
+					    msg);
+			}
+
+			if (use_browser <= 0)
+				sleep(5);
+			top->vmlinux_warned = true;
+		}
+	}
+
+	if (al.sym == NULL || !al.sym->ignore) {
+		struct hists *hists = evsel__hists(evsel);
+		struct hist_entry_iter iter = {
+			.add_entry_cb = hist_iter__top_callback,
+		};
+
+		if (symbol_conf.cumulate_callchain)
+			iter.ops = &hist_iter_cumulative;
+		else
+			iter.ops = &hist_iter_normal;
+
+		pthread_mutex_lock(&hists->lock);
+
+		err = hist_entry_iter__add(&iter, &al, evsel, sample,
+					   top->max_stack, top);
+		if (err < 0)
+			pr_err("Problem incrementing symbol period, skipping event\n");
+
+		pthread_mutex_unlock(&hists->lock);
+	}
+
+	return;
+}
+
+static void perf_top__mmap_read_idx(struct perf_top *top, int idx)
+{
+	struct perf_sample sample;
+	struct perf_evsel *evsel;
+	struct perf_session *session = top->session;
+	union perf_event *event;
+	struct machine *machine;
+	u8 origin;
+	int ret;
+
+	while ((event = perf_evlist__mmap_read(top->evlist, idx)) != NULL) {
+		ret = perf_evlist__parse_sample(top->evlist, event, &sample);
+		if (ret) {
+			pr_err("Can't parse sample, err = %d\n", ret);
+			goto next_event;
+		}
+
+		evsel = perf_evlist__id2evsel(session->evlist, sample.id);
+		assert(evsel != NULL);
+
+		origin = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+
+		if (event->header.type == PERF_RECORD_SAMPLE)
+			++top->samples;
+
+		switch (origin) {
+		case PERF_RECORD_MISC_USER:
+			++top->us_samples;
+			if (top->hide_user_symbols)
+				goto next_event;
+			machine = &session->machines.host;
+			break;
+		case PERF_RECORD_MISC_KERNEL:
+			++top->kernel_samples;
+			if (top->hide_kernel_symbols)
+				goto next_event;
+			machine = &session->machines.host;
+			break;
+		case PERF_RECORD_MISC_GUEST_KERNEL:
+			++top->guest_kernel_samples;
+			machine = perf_session__find_machine(session,
+							     sample.pid);
+			break;
+		case PERF_RECORD_MISC_GUEST_USER:
+			++top->guest_us_samples;
+			/*
+			 * TODO: we don't process guest user from host side
+			 * except simple counting.
+			 */
+			/* Fall thru */
+		default:
+			goto next_event;
+		}
+
+
+		if (event->header.type == PERF_RECORD_SAMPLE) {
+			perf_event__process_sample(&top->tool, event, evsel,
+						   &sample, machine);
+		} else if (event->header.type < PERF_RECORD_MAX) {
+			hists__inc_nr_events(evsel__hists(evsel), event->header.type);
+			machine__process_event(machine, event, &sample);
+		} else
+			++session->evlist->stats.nr_unknown_events;
+next_event:
+		perf_evlist__mmap_consume(top->evlist, idx);
+	}
+}
+
+static void perf_top__mmap_read(struct perf_top *top)
+{
+	int i;
+
+	for (i = 0; i < top->evlist->nr_mmaps; i++)
+		perf_top__mmap_read_idx(top, i);
+}
+
+static int perf_top__start_counters(struct perf_top *top)
+{
+	char msg[512];
+	struct perf_evsel *counter;
+	struct perf_evlist *evlist = top->evlist;
+	struct record_opts *opts = &top->record_opts;
+
+	perf_evlist__config(evlist, opts);
+
+	evlist__for_each(evlist, counter) {
+try_again:
+		if (perf_evsel__open(counter, top->evlist->cpus,
+				     top->evlist->threads) < 0) {
+			if (perf_evsel__fallback(counter, errno, msg, sizeof(msg))) {
+				if (verbose)
+					ui__warning("%s\n", msg);
+				goto try_again;
+			}
+
+			perf_evsel__open_strerror(counter, &opts->target,
+						  errno, msg, sizeof(msg));
+			ui__error("%s\n", msg);
+			goto out_err;
+		}
+	}
+
+	if (perf_evlist__mmap(evlist, opts->mmap_pages, false) < 0) {
+		ui__error("Failed to mmap with %d (%s)\n",
+			    errno, strerror_r(errno, msg, sizeof(msg)));
+		goto out_err;
+	}
+
+	return 0;
+
+out_err:
+	return -1;
+}
+
+static int perf_top__setup_sample_type(struct perf_top *top __maybe_unused)
+{
+	if (!sort__has_sym) {
+		if (symbol_conf.use_callchain) {
+			ui__error("Selected -g but \"sym\" not present in --sort/-s.");
+			return -EINVAL;
+		}
+	} else if (callchain_param.mode != CHAIN_NONE) {
+		if (callchain_register_param(&callchain_param) < 0) {
+			ui__error("Can't register callchain params.\n");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int __cmd_top(struct perf_top *top)
+{
+	struct record_opts *opts = &top->record_opts;
+	pthread_t thread;
+	int ret;
+
+	top->session = perf_session__new(NULL, false, NULL);
+	if (top->session == NULL)
+		return -1;
+
+	machines__set_symbol_filter(&top->session->machines, symbol_filter);
+
+	if (!objdump_path) {
+		ret = perf_session_env__lookup_objdump(&top->session->header.env);
+		if (ret)
+			goto out_delete;
+	}
+
+	ret = perf_top__setup_sample_type(top);
+	if (ret)
+		goto out_delete;
+
+	machine__synthesize_threads(&top->session->machines.host, &opts->target,
+				    top->evlist->threads, false);
+	ret = perf_top__start_counters(top);
+	if (ret)
+		goto out_delete;
+
+	top->session->evlist = top->evlist;
+	perf_session__set_id_hdr_size(top->session);
+
+	/*
+	 * When perf is starting the traced process, all the events (apart from
+	 * group members) have enable_on_exec=1 set, so don't spoil it by
+	 * prematurely enabling them.
+	 *
+	 * XXX 'top' still doesn't start workloads like record, trace, but should,
+	 * so leave the check here.
+	 */
+        if (!target__none(&opts->target))
+                perf_evlist__enable(top->evlist);
+
+	/* Wait for a minimal set of events before starting the snapshot */
+	perf_evlist__poll(top->evlist, 100);
+
+	perf_top__mmap_read(top);
+
+	ret = -1;
+	if (pthread_create(&thread, NULL, (use_browser > 0 ? display_thread_tui :
+							    display_thread), top)) {
+		ui__error("Could not create display thread.\n");
+		goto out_delete;
+	}
+
+	if (top->realtime_prio) {
+		struct sched_param param;
+
+		param.sched_priority = top->realtime_prio;
+		if (sched_setscheduler(0, SCHED_FIFO, &param)) {
+			ui__error("Could not set realtime priority.\n");
+			goto out_join;
+		}
+	}
+
+	while (!done) {
+		u64 hits = top->samples;
+
+		perf_top__mmap_read(top);
+
+		if (hits == top->samples)
+			ret = perf_evlist__poll(top->evlist, 100);
+	}
+
+	ret = 0;
+out_join:
+	pthread_join(thread, NULL);
+out_delete:
+	perf_session__delete(top->session);
+	top->session = NULL;
+
+	return ret;
+}
+
+static int
+callchain_opt(const struct option *opt, const char *arg, int unset)
+{
+	symbol_conf.use_callchain = true;
+	return record_callchain_opt(opt, arg, unset);
+}
+
+static int
+parse_callchain_opt(const struct option *opt, const char *arg, int unset)
+{
+	symbol_conf.use_callchain = true;
+	return record_parse_callchain_opt(opt, arg, unset);
+}
+
+static int perf_top_config(const char *var, const char *value, void *cb)
+{
+	if (!strcmp(var, "top.call-graph"))
+		var = "call-graph.record-mode"; /* fall-through */
+	if (!strcmp(var, "top.children")) {
+		symbol_conf.cumulate_callchain = perf_config_bool(var, value);
+		return 0;
+	}
+
+	return perf_default_config(var, value, cb);
+}
+
+static int
+parse_percent_limit(const struct option *opt, const char *arg,
+		    int unset __maybe_unused)
+{
+	struct perf_top *top = opt->value;
+
+	top->min_percent = strtof(arg, NULL);
+	return 0;
+}
+
+int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	char errbuf[BUFSIZ];
+	struct perf_top top = {
+		.count_filter	     = 5,
+		.delay_secs	     = 2,
+		.record_opts = {
+			.mmap_pages	= UINT_MAX,
+			.user_freq	= UINT_MAX,
+			.user_interval	= ULLONG_MAX,
+			.freq		= 4000, /* 4 KHz */
+			.target		= {
+				.uses_mmap   = true,
+			},
+		},
+		.max_stack	     = PERF_MAX_STACK_DEPTH,
+		.sym_pcnt_filter     = 5,
+	};
+	struct record_opts *opts = &top.record_opts;
+	struct target *target = &opts->target;
+	const struct option options[] = {
+	OPT_CALLBACK('e', "event", &top.evlist, "event",
+		     "event selector. use 'perf list' to list available events",
+		     parse_events_option),
+	OPT_U64('c', "count", &opts->user_interval, "event period to sample"),
+	OPT_STRING('p', "pid", &target->pid, "pid",
+		    "profile events on existing process id"),
+	OPT_STRING('t', "tid", &target->tid, "tid",
+		    "profile events on existing thread id"),
+	OPT_BOOLEAN('a', "all-cpus", &target->system_wide,
+			    "system-wide collection from all CPUs"),
+	OPT_STRING('C', "cpu", &target->cpu_list, "cpu",
+		    "list of cpus to monitor"),
+	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
+		   "file", "vmlinux pathname"),
+	OPT_BOOLEAN(0, "ignore-vmlinux", &symbol_conf.ignore_vmlinux,
+		    "don't load vmlinux even if found"),
+	OPT_BOOLEAN('K', "hide_kernel_symbols", &top.hide_kernel_symbols,
+		    "hide kernel symbols"),
+	OPT_CALLBACK('m', "mmap-pages", &opts->mmap_pages, "pages",
+		     "number of mmap data pages",
+		     perf_evlist__parse_mmap_pages),
+	OPT_INTEGER('r', "realtime", &top.realtime_prio,
+		    "collect data with this RT SCHED_FIFO priority"),
+	OPT_INTEGER('d', "delay", &top.delay_secs,
+		    "number of seconds to delay between refreshes"),
+	OPT_BOOLEAN('D', "dump-symtab", &top.dump_symtab,
+			    "dump the symbol table used for profiling"),
+	OPT_INTEGER('f', "count-filter", &top.count_filter,
+		    "only display functions with more events than this"),
+	OPT_BOOLEAN(0, "group", &opts->group,
+			    "put the counters into a counter group"),
+	OPT_BOOLEAN('i', "no-inherit", &opts->no_inherit,
+		    "child tasks do not inherit counters"),
+	OPT_STRING(0, "sym-annotate", &top.sym_filter, "symbol name",
+		    "symbol to annotate"),
+	OPT_BOOLEAN('z', "zero", &top.zero, "zero history across updates"),
+	OPT_UINTEGER('F', "freq", &opts->user_freq, "profile at this frequency"),
+	OPT_INTEGER('E', "entries", &top.print_entries,
+		    "display this many functions"),
+	OPT_BOOLEAN('U', "hide_user_symbols", &top.hide_user_symbols,
+		    "hide user symbols"),
+	OPT_BOOLEAN(0, "tui", &top.use_tui, "Use the TUI interface"),
+	OPT_BOOLEAN(0, "stdio", &top.use_stdio, "Use the stdio interface"),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show counter open errors, etc)"),
+	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
+		   "sort by key(s): pid, comm, dso, symbol, parent, cpu, srcline, ..."
+		   " Please refer the man page for the complete list."),
+	OPT_STRING(0, "fields", &field_order, "key[,keys...]",
+		   "output field(s): overhead, period, sample plus all of sort keys"),
+	OPT_BOOLEAN('n', "show-nr-samples", &symbol_conf.show_nr_samples,
+		    "Show a column with the number of samples"),
+	OPT_CALLBACK_NOOPT('g', NULL, &top.record_opts,
+			   NULL, "enables call-graph recording",
+			   &callchain_opt),
+	OPT_CALLBACK(0, "call-graph", &top.record_opts,
+		     "mode[,dump_size]", record_callchain_help,
+		     &parse_callchain_opt),
+	OPT_BOOLEAN(0, "children", &symbol_conf.cumulate_callchain,
+		    "Accumulate callchains of children and show total overhead as well"),
+	OPT_INTEGER(0, "max-stack", &top.max_stack,
+		    "Set the maximum stack depth when parsing the callchain. "
+		    "Default: " __stringify(PERF_MAX_STACK_DEPTH)),
+	OPT_CALLBACK(0, "ignore-callees", NULL, "regex",
+		   "ignore callees of these functions in call graphs",
+		   report_parse_ignore_callees_opt),
+	OPT_BOOLEAN(0, "show-total-period", &symbol_conf.show_total_period,
+		    "Show a column with the sum of periods"),
+	OPT_STRING(0, "dsos", &symbol_conf.dso_list_str, "dso[,dso...]",
+		   "only consider symbols in these dsos"),
+	OPT_STRING(0, "comms", &symbol_conf.comm_list_str, "comm[,comm...]",
+		   "only consider symbols in these comms"),
+	OPT_STRING(0, "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]",
+		   "only consider these symbols"),
+	OPT_BOOLEAN(0, "source", &symbol_conf.annotate_src,
+		    "Interleave source code with assembly code (default)"),
+	OPT_BOOLEAN(0, "asm-raw", &symbol_conf.annotate_asm_raw,
+		    "Display raw encoding of assembly instructions (default)"),
+	OPT_BOOLEAN(0, "demangle-kernel", &symbol_conf.demangle_kernel,
+		    "Enable kernel symbol demangling"),
+	OPT_STRING(0, "objdump", &objdump_path, "path",
+		    "objdump binary to use for disassembly and annotations"),
+	OPT_STRING('M', "disassembler-style", &disassembler_style, "disassembler style",
+		   "Specify disassembler style (e.g. -M intel for intel syntax)"),
+	OPT_STRING('u', "uid", &target->uid_str, "user", "user to profile"),
+	OPT_CALLBACK(0, "percent-limit", &top, "percent",
+		     "Don't show entries under that percent", parse_percent_limit),
+	OPT_CALLBACK(0, "percentage", NULL, "relative|absolute",
+		     "How to display percentage of filtered entries", parse_filter_percentage),
+	OPT_STRING('w', "column-widths", &symbol_conf.col_width_list_str,
+		   "width[,width...]",
+		   "don't try to adjust column width, use these fixed values"),
+	OPT_END()
+	};
+	const char * const top_usage[] = {
+		"perf top [<options>]",
+		NULL
+	};
+	int status = hists__init();
+
+	if (status < 0)
+		return status;
+
+	top.evlist = perf_evlist__new();
+	if (top.evlist == NULL)
+		return -ENOMEM;
+
+	perf_config(perf_top_config, &top);
+
+	argc = parse_options(argc, argv, options, top_usage, 0);
+	if (argc)
+		usage_with_options(top_usage, options);
+
+	sort__mode = SORT_MODE__TOP;
+	/* display thread wants entries to be collapsed in a different tree */
+	sort__need_collapse = 1;
+
+	if (setup_sorting() < 0) {
+		if (sort_order)
+			parse_options_usage(top_usage, options, "s", 1);
+		if (field_order)
+			parse_options_usage(sort_order ? NULL : top_usage,
+					    options, "fields", 0);
+		goto out_delete_evlist;
+	}
+
+	if (top.use_stdio)
+		use_browser = 0;
+	else if (top.use_tui)
+		use_browser = 1;
+
+	setup_browser(false);
+
+	status = target__validate(target);
+	if (status) {
+		target__strerror(target, status, errbuf, BUFSIZ);
+		ui__warning("%s\n", errbuf);
+	}
+
+	status = target__parse_uid(target);
+	if (status) {
+		int saved_errno = errno;
+
+		target__strerror(target, status, errbuf, BUFSIZ);
+		ui__error("%s\n", errbuf);
+
+		status = -saved_errno;
+		goto out_delete_evlist;
+	}
+
+	if (target__none(target))
+		target->system_wide = true;
+
+	if (perf_evlist__create_maps(top.evlist, target) < 0)
+		usage_with_options(top_usage, options);
+
+	if (!top.evlist->nr_entries &&
+	    perf_evlist__add_default(top.evlist) < 0) {
+		ui__error("Not enough memory for event selector list\n");
+		goto out_delete_evlist;
+	}
+
+	symbol_conf.nr_events = top.evlist->nr_entries;
+
+	if (top.delay_secs < 1)
+		top.delay_secs = 1;
+
+	if (record_opts__config(opts)) {
+		status = -EINVAL;
+		goto out_delete_evlist;
+	}
+
+	top.sym_evsel = perf_evlist__first(top.evlist);
+
+	if (!symbol_conf.use_callchain) {
+		symbol_conf.cumulate_callchain = false;
+		perf_hpp__cancel_cumulate();
+	}
+
+	symbol_conf.priv_size = sizeof(struct annotation);
+
+	symbol_conf.try_vmlinux_path = (symbol_conf.vmlinux_name == NULL);
+	if (symbol__init(NULL) < 0)
+		return -1;
+
+	sort__setup_elide(stdout);
+
+	get_term_dimensions(&top.winsize);
+	if (top.print_entries == 0) {
+		struct sigaction act = {
+			.sa_sigaction = perf_top__sig_winch,
+			.sa_flags     = SA_SIGINFO,
+		};
+		perf_top__update_print_entries(&top);
+		sigaction(SIGWINCH, &act, NULL);
+	}
+
+	status = __cmd_top(&top);
+
+out_delete_evlist:
+	perf_evlist__delete(top.evlist);
+
+	return status;
+}
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
new file mode 100644
index 000000000..e12297036
--- /dev/null
+++ b/tools/perf/builtin-trace.c
@@ -0,0 +1,2797 @@
+#include <traceevent/event-parse.h>
+#include "builtin.h"
+#include "util/color.h"
+#include "util/debug.h"
+#include "util/evlist.h"
+#include "util/machine.h"
+#include "util/session.h"
+#include "util/thread.h"
+#include "util/parse-options.h"
+#include "util/strlist.h"
+#include "util/intlist.h"
+#include "util/thread_map.h"
+#include "util/stat.h"
+#include "trace-event.h"
+#include "util/parse-events.h"
+
+#include <libaudit.h>
+#include <stdlib.h>
+#include <sys/eventfd.h>
+#include <sys/mman.h>
+#include <linux/futex.h>
+
+/* For older distros: */
+#ifndef MAP_STACK
+# define MAP_STACK		0x20000
+#endif
+
+#ifndef MADV_HWPOISON
+# define MADV_HWPOISON		100
+#endif
+
+#ifndef MADV_MERGEABLE
+# define MADV_MERGEABLE		12
+#endif
+
+#ifndef MADV_UNMERGEABLE
+# define MADV_UNMERGEABLE	13
+#endif
+
+#ifndef EFD_SEMAPHORE
+# define EFD_SEMAPHORE		1
+#endif
+
+struct tp_field {
+	int offset;
+	union {
+		u64 (*integer)(struct tp_field *field, struct perf_sample *sample);
+		void *(*pointer)(struct tp_field *field, struct perf_sample *sample);
+	};
+};
+
+#define TP_UINT_FIELD(bits) \
+static u64 tp_field__u##bits(struct tp_field *field, struct perf_sample *sample) \
+{ \
+	u##bits value; \
+	memcpy(&value, sample->raw_data + field->offset, sizeof(value)); \
+	return value;  \
+}
+
+TP_UINT_FIELD(8);
+TP_UINT_FIELD(16);
+TP_UINT_FIELD(32);
+TP_UINT_FIELD(64);
+
+#define TP_UINT_FIELD__SWAPPED(bits) \
+static u64 tp_field__swapped_u##bits(struct tp_field *field, struct perf_sample *sample) \
+{ \
+	u##bits value; \
+	memcpy(&value, sample->raw_data + field->offset, sizeof(value)); \
+	return bswap_##bits(value);\
+}
+
+TP_UINT_FIELD__SWAPPED(16);
+TP_UINT_FIELD__SWAPPED(32);
+TP_UINT_FIELD__SWAPPED(64);
+
+static int tp_field__init_uint(struct tp_field *field,
+			       struct format_field *format_field,
+			       bool needs_swap)
+{
+	field->offset = format_field->offset;
+
+	switch (format_field->size) {
+	case 1:
+		field->integer = tp_field__u8;
+		break;
+	case 2:
+		field->integer = needs_swap ? tp_field__swapped_u16 : tp_field__u16;
+		break;
+	case 4:
+		field->integer = needs_swap ? tp_field__swapped_u32 : tp_field__u32;
+		break;
+	case 8:
+		field->integer = needs_swap ? tp_field__swapped_u64 : tp_field__u64;
+		break;
+	default:
+		return -1;
+	}
+
+	return 0;
+}
+
+static void *tp_field__ptr(struct tp_field *field, struct perf_sample *sample)
+{
+	return sample->raw_data + field->offset;
+}
+
+static int tp_field__init_ptr(struct tp_field *field, struct format_field *format_field)
+{
+	field->offset = format_field->offset;
+	field->pointer = tp_field__ptr;
+	return 0;
+}
+
+struct syscall_tp {
+	struct tp_field id;
+	union {
+		struct tp_field args, ret;
+	};
+};
+
+static int perf_evsel__init_tp_uint_field(struct perf_evsel *evsel,
+					  struct tp_field *field,
+					  const char *name)
+{
+	struct format_field *format_field = perf_evsel__field(evsel, name);
+
+	if (format_field == NULL)
+		return -1;
+
+	return tp_field__init_uint(field, format_field, evsel->needs_swap);
+}
+
+#define perf_evsel__init_sc_tp_uint_field(evsel, name) \
+	({ struct syscall_tp *sc = evsel->priv;\
+	   perf_evsel__init_tp_uint_field(evsel, &sc->name, #name); })
+
+static int perf_evsel__init_tp_ptr_field(struct perf_evsel *evsel,
+					 struct tp_field *field,
+					 const char *name)
+{
+	struct format_field *format_field = perf_evsel__field(evsel, name);
+
+	if (format_field == NULL)
+		return -1;
+
+	return tp_field__init_ptr(field, format_field);
+}
+
+#define perf_evsel__init_sc_tp_ptr_field(evsel, name) \
+	({ struct syscall_tp *sc = evsel->priv;\
+	   perf_evsel__init_tp_ptr_field(evsel, &sc->name, #name); })
+
+static void perf_evsel__delete_priv(struct perf_evsel *evsel)
+{
+	zfree(&evsel->priv);
+	perf_evsel__delete(evsel);
+}
+
+static int perf_evsel__init_syscall_tp(struct perf_evsel *evsel, void *handler)
+{
+	evsel->priv = malloc(sizeof(struct syscall_tp));
+	if (evsel->priv != NULL) {
+		if (perf_evsel__init_sc_tp_uint_field(evsel, id))
+			goto out_delete;
+
+		evsel->handler = handler;
+		return 0;
+	}
+
+	return -ENOMEM;
+
+out_delete:
+	zfree(&evsel->priv);
+	return -ENOENT;
+}
+
+static struct perf_evsel *perf_evsel__syscall_newtp(const char *direction, void *handler)
+{
+	struct perf_evsel *evsel = perf_evsel__newtp("raw_syscalls", direction);
+
+	/* older kernel (e.g., RHEL6) use syscalls:{enter,exit} */
+	if (evsel == NULL)
+		evsel = perf_evsel__newtp("syscalls", direction);
+
+	if (evsel) {
+		if (perf_evsel__init_syscall_tp(evsel, handler))
+			goto out_delete;
+	}
+
+	return evsel;
+
+out_delete:
+	perf_evsel__delete_priv(evsel);
+	return NULL;
+}
+
+#define perf_evsel__sc_tp_uint(evsel, name, sample) \
+	({ struct syscall_tp *fields = evsel->priv; \
+	   fields->name.integer(&fields->name, sample); })
+
+#define perf_evsel__sc_tp_ptr(evsel, name, sample) \
+	({ struct syscall_tp *fields = evsel->priv; \
+	   fields->name.pointer(&fields->name, sample); })
+
+static int perf_evlist__add_syscall_newtp(struct perf_evlist *evlist,
+					  void *sys_enter_handler,
+					  void *sys_exit_handler)
+{
+	int ret = -1;
+	struct perf_evsel *sys_enter, *sys_exit;
+
+	sys_enter = perf_evsel__syscall_newtp("sys_enter", sys_enter_handler);
+	if (sys_enter == NULL)
+		goto out;
+
+	if (perf_evsel__init_sc_tp_ptr_field(sys_enter, args))
+		goto out_delete_sys_enter;
+
+	sys_exit = perf_evsel__syscall_newtp("sys_exit", sys_exit_handler);
+	if (sys_exit == NULL)
+		goto out_delete_sys_enter;
+
+	if (perf_evsel__init_sc_tp_uint_field(sys_exit, ret))
+		goto out_delete_sys_exit;
+
+	perf_evlist__add(evlist, sys_enter);
+	perf_evlist__add(evlist, sys_exit);
+
+	ret = 0;
+out:
+	return ret;
+
+out_delete_sys_exit:
+	perf_evsel__delete_priv(sys_exit);
+out_delete_sys_enter:
+	perf_evsel__delete_priv(sys_enter);
+	goto out;
+}
+
+
+struct syscall_arg {
+	unsigned long val;
+	struct thread *thread;
+	struct trace  *trace;
+	void	      *parm;
+	u8	      idx;
+	u8	      mask;
+};
+
+struct strarray {
+	int	    offset;
+	int	    nr_entries;
+	const char **entries;
+};
+
+#define DEFINE_STRARRAY(array) struct strarray strarray__##array = { \
+	.nr_entries = ARRAY_SIZE(array), \
+	.entries = array, \
+}
+
+#define DEFINE_STRARRAY_OFFSET(array, off) struct strarray strarray__##array = { \
+	.offset	    = off, \
+	.nr_entries = ARRAY_SIZE(array), \
+	.entries = array, \
+}
+
+static size_t __syscall_arg__scnprintf_strarray(char *bf, size_t size,
+						const char *intfmt,
+					        struct syscall_arg *arg)
+{
+	struct strarray *sa = arg->parm;
+	int idx = arg->val - sa->offset;
+
+	if (idx < 0 || idx >= sa->nr_entries)
+		return scnprintf(bf, size, intfmt, arg->val);
+
+	return scnprintf(bf, size, "%s", sa->entries[idx]);
+}
+
+static size_t syscall_arg__scnprintf_strarray(char *bf, size_t size,
+					      struct syscall_arg *arg)
+{
+	return __syscall_arg__scnprintf_strarray(bf, size, "%d", arg);
+}
+
+#define SCA_STRARRAY syscall_arg__scnprintf_strarray
+
+#if defined(__i386__) || defined(__x86_64__)
+/*
+ * FIXME: Make this available to all arches as soon as the ioctl beautifier
+ * 	  gets rewritten to support all arches.
+ */
+static size_t syscall_arg__scnprintf_strhexarray(char *bf, size_t size,
+						 struct syscall_arg *arg)
+{
+	return __syscall_arg__scnprintf_strarray(bf, size, "%#x", arg);
+}
+
+#define SCA_STRHEXARRAY syscall_arg__scnprintf_strhexarray
+#endif /* defined(__i386__) || defined(__x86_64__) */
+
+static size_t syscall_arg__scnprintf_fd(char *bf, size_t size,
+					struct syscall_arg *arg);
+
+#define SCA_FD syscall_arg__scnprintf_fd
+
+static size_t syscall_arg__scnprintf_fd_at(char *bf, size_t size,
+					   struct syscall_arg *arg)
+{
+	int fd = arg->val;
+
+	if (fd == AT_FDCWD)
+		return scnprintf(bf, size, "CWD");
+
+	return syscall_arg__scnprintf_fd(bf, size, arg);
+}
+
+#define SCA_FDAT syscall_arg__scnprintf_fd_at
+
+static size_t syscall_arg__scnprintf_close_fd(char *bf, size_t size,
+					      struct syscall_arg *arg);
+
+#define SCA_CLOSE_FD syscall_arg__scnprintf_close_fd
+
+static size_t syscall_arg__scnprintf_hex(char *bf, size_t size,
+					 struct syscall_arg *arg)
+{
+	return scnprintf(bf, size, "%#lx", arg->val);
+}
+
+#define SCA_HEX syscall_arg__scnprintf_hex
+
+static size_t syscall_arg__scnprintf_mmap_prot(char *bf, size_t size,
+					       struct syscall_arg *arg)
+{
+	int printed = 0, prot = arg->val;
+
+	if (prot == PROT_NONE)
+		return scnprintf(bf, size, "NONE");
+#define	P_MMAP_PROT(n) \
+	if (prot & PROT_##n) { \
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
+		prot &= ~PROT_##n; \
+	}
+
+	P_MMAP_PROT(EXEC);
+	P_MMAP_PROT(READ);
+	P_MMAP_PROT(WRITE);
+#ifdef PROT_SEM
+	P_MMAP_PROT(SEM);
+#endif
+	P_MMAP_PROT(GROWSDOWN);
+	P_MMAP_PROT(GROWSUP);
+#undef P_MMAP_PROT
+
+	if (prot)
+		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", prot);
+
+	return printed;
+}
+
+#define SCA_MMAP_PROT syscall_arg__scnprintf_mmap_prot
+
+static size_t syscall_arg__scnprintf_mmap_flags(char *bf, size_t size,
+						struct syscall_arg *arg)
+{
+	int printed = 0, flags = arg->val;
+
+#define	P_MMAP_FLAG(n) \
+	if (flags & MAP_##n) { \
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
+		flags &= ~MAP_##n; \
+	}
+
+	P_MMAP_FLAG(SHARED);
+	P_MMAP_FLAG(PRIVATE);
+#ifdef MAP_32BIT
+	P_MMAP_FLAG(32BIT);
+#endif
+	P_MMAP_FLAG(ANONYMOUS);
+	P_MMAP_FLAG(DENYWRITE);
+	P_MMAP_FLAG(EXECUTABLE);
+	P_MMAP_FLAG(FILE);
+	P_MMAP_FLAG(FIXED);
+	P_MMAP_FLAG(GROWSDOWN);
+#ifdef MAP_HUGETLB
+	P_MMAP_FLAG(HUGETLB);
+#endif
+	P_MMAP_FLAG(LOCKED);
+	P_MMAP_FLAG(NONBLOCK);
+	P_MMAP_FLAG(NORESERVE);
+	P_MMAP_FLAG(POPULATE);
+	P_MMAP_FLAG(STACK);
+#ifdef MAP_UNINITIALIZED
+	P_MMAP_FLAG(UNINITIALIZED);
+#endif
+#undef P_MMAP_FLAG
+
+	if (flags)
+		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);
+
+	return printed;
+}
+
+#define SCA_MMAP_FLAGS syscall_arg__scnprintf_mmap_flags
+
+static size_t syscall_arg__scnprintf_mremap_flags(char *bf, size_t size,
+						  struct syscall_arg *arg)
+{
+	int printed = 0, flags = arg->val;
+
+#define P_MREMAP_FLAG(n) \
+	if (flags & MREMAP_##n) { \
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
+		flags &= ~MREMAP_##n; \
+	}
+
+	P_MREMAP_FLAG(MAYMOVE);
+#ifdef MREMAP_FIXED
+	P_MREMAP_FLAG(FIXED);
+#endif
+#undef P_MREMAP_FLAG
+
+	if (flags)
+		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);
+
+	return printed;
+}
+
+#define SCA_MREMAP_FLAGS syscall_arg__scnprintf_mremap_flags
+
+static size_t syscall_arg__scnprintf_madvise_behavior(char *bf, size_t size,
+						      struct syscall_arg *arg)
+{
+	int behavior = arg->val;
+
+	switch (behavior) {
+#define	P_MADV_BHV(n) case MADV_##n: return scnprintf(bf, size, #n)
+	P_MADV_BHV(NORMAL);
+	P_MADV_BHV(RANDOM);
+	P_MADV_BHV(SEQUENTIAL);
+	P_MADV_BHV(WILLNEED);
+	P_MADV_BHV(DONTNEED);
+	P_MADV_BHV(REMOVE);
+	P_MADV_BHV(DONTFORK);
+	P_MADV_BHV(DOFORK);
+	P_MADV_BHV(HWPOISON);
+#ifdef MADV_SOFT_OFFLINE
+	P_MADV_BHV(SOFT_OFFLINE);
+#endif
+	P_MADV_BHV(MERGEABLE);
+	P_MADV_BHV(UNMERGEABLE);
+#ifdef MADV_HUGEPAGE
+	P_MADV_BHV(HUGEPAGE);
+#endif
+#ifdef MADV_NOHUGEPAGE
+	P_MADV_BHV(NOHUGEPAGE);
+#endif
+#ifdef MADV_DONTDUMP
+	P_MADV_BHV(DONTDUMP);
+#endif
+#ifdef MADV_DODUMP
+	P_MADV_BHV(DODUMP);
+#endif
+#undef P_MADV_PHV
+	default: break;
+	}
+
+	return scnprintf(bf, size, "%#x", behavior);
+}
+
+#define SCA_MADV_BHV syscall_arg__scnprintf_madvise_behavior
+
+static size_t syscall_arg__scnprintf_flock(char *bf, size_t size,
+					   struct syscall_arg *arg)
+{
+	int printed = 0, op = arg->val;
+
+	if (op == 0)
+		return scnprintf(bf, size, "NONE");
+#define	P_CMD(cmd) \
+	if ((op & LOCK_##cmd) == LOCK_##cmd) { \
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #cmd); \
+		op &= ~LOCK_##cmd; \
+	}
+
+	P_CMD(SH);
+	P_CMD(EX);
+	P_CMD(NB);
+	P_CMD(UN);
+	P_CMD(MAND);
+	P_CMD(RW);
+	P_CMD(READ);
+	P_CMD(WRITE);
+#undef P_OP
+
+	if (op)
+		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", op);
+
+	return printed;
+}
+
+#define SCA_FLOCK syscall_arg__scnprintf_flock
+
+static size_t syscall_arg__scnprintf_futex_op(char *bf, size_t size, struct syscall_arg *arg)
+{
+	enum syscall_futex_args {
+		SCF_UADDR   = (1 << 0),
+		SCF_OP	    = (1 << 1),
+		SCF_VAL	    = (1 << 2),
+		SCF_TIMEOUT = (1 << 3),
+		SCF_UADDR2  = (1 << 4),
+		SCF_VAL3    = (1 << 5),
+	};
+	int op = arg->val;
+	int cmd = op & FUTEX_CMD_MASK;
+	size_t printed = 0;
+
+	switch (cmd) {
+#define	P_FUTEX_OP(n) case FUTEX_##n: printed = scnprintf(bf, size, #n);
+	P_FUTEX_OP(WAIT);	    arg->mask |= SCF_VAL3|SCF_UADDR2;		  break;
+	P_FUTEX_OP(WAKE);	    arg->mask |= SCF_VAL3|SCF_UADDR2|SCF_TIMEOUT; break;
+	P_FUTEX_OP(FD);		    arg->mask |= SCF_VAL3|SCF_UADDR2|SCF_TIMEOUT; break;
+	P_FUTEX_OP(REQUEUE);	    arg->mask |= SCF_VAL3|SCF_TIMEOUT;	          break;
+	P_FUTEX_OP(CMP_REQUEUE);    arg->mask |= SCF_TIMEOUT;			  break;
+	P_FUTEX_OP(CMP_REQUEUE_PI); arg->mask |= SCF_TIMEOUT;			  break;
+	P_FUTEX_OP(WAKE_OP);							  break;
+	P_FUTEX_OP(LOCK_PI);	    arg->mask |= SCF_VAL3|SCF_UADDR2|SCF_TIMEOUT; break;
+	P_FUTEX_OP(UNLOCK_PI);	    arg->mask |= SCF_VAL3|SCF_UADDR2|SCF_TIMEOUT; break;
+	P_FUTEX_OP(TRYLOCK_PI);	    arg->mask |= SCF_VAL3|SCF_UADDR2;		  break;
+	P_FUTEX_OP(WAIT_BITSET);    arg->mask |= SCF_UADDR2;			  break;
+	P_FUTEX_OP(WAKE_BITSET);    arg->mask |= SCF_UADDR2;			  break;
+	P_FUTEX_OP(WAIT_REQUEUE_PI);						  break;
+	default: printed = scnprintf(bf, size, "%#x", cmd);			  break;
+	}
+
+	if (op & FUTEX_PRIVATE_FLAG)
+		printed += scnprintf(bf + printed, size - printed, "|PRIV");
+
+	if (op & FUTEX_CLOCK_REALTIME)
+		printed += scnprintf(bf + printed, size - printed, "|CLKRT");
+
+	return printed;
+}
+
+#define SCA_FUTEX_OP  syscall_arg__scnprintf_futex_op
+
+static const char *epoll_ctl_ops[] = { "ADD", "DEL", "MOD", };
+static DEFINE_STRARRAY_OFFSET(epoll_ctl_ops, 1);
+
+static const char *itimers[] = { "REAL", "VIRTUAL", "PROF", };
+static DEFINE_STRARRAY(itimers);
+
+static const char *whences[] = { "SET", "CUR", "END",
+#ifdef SEEK_DATA
+"DATA",
+#endif
+#ifdef SEEK_HOLE
+"HOLE",
+#endif
+};
+static DEFINE_STRARRAY(whences);
+
+static const char *fcntl_cmds[] = {
+	"DUPFD", "GETFD", "SETFD", "GETFL", "SETFL", "GETLK", "SETLK",
+	"SETLKW", "SETOWN", "GETOWN", "SETSIG", "GETSIG", "F_GETLK64",
+	"F_SETLK64", "F_SETLKW64", "F_SETOWN_EX", "F_GETOWN_EX",
+	"F_GETOWNER_UIDS",
+};
+static DEFINE_STRARRAY(fcntl_cmds);
+
+static const char *rlimit_resources[] = {
+	"CPU", "FSIZE", "DATA", "STACK", "CORE", "RSS", "NPROC", "NOFILE",
+	"MEMLOCK", "AS", "LOCKS", "SIGPENDING", "MSGQUEUE", "NICE", "RTPRIO",
+	"RTTIME",
+};
+static DEFINE_STRARRAY(rlimit_resources);
+
+static const char *sighow[] = { "BLOCK", "UNBLOCK", "SETMASK", };
+static DEFINE_STRARRAY(sighow);
+
+static const char *clockid[] = {
+	"REALTIME", "MONOTONIC", "PROCESS_CPUTIME_ID", "THREAD_CPUTIME_ID",
+	"MONOTONIC_RAW", "REALTIME_COARSE", "MONOTONIC_COARSE",
+};
+static DEFINE_STRARRAY(clockid);
+
+static const char *socket_families[] = {
+	"UNSPEC", "LOCAL", "INET", "AX25", "IPX", "APPLETALK", "NETROM",
+	"BRIDGE", "ATMPVC", "X25", "INET6", "ROSE", "DECnet", "NETBEUI",
+	"SECURITY", "KEY", "NETLINK", "PACKET", "ASH", "ECONET", "ATMSVC",
+	"RDS", "SNA", "IRDA", "PPPOX", "WANPIPE", "LLC", "IB", "CAN", "TIPC",
+	"BLUETOOTH", "IUCV", "RXRPC", "ISDN", "PHONET", "IEEE802154", "CAIF",
+	"ALG", "NFC", "VSOCK",
+};
+static DEFINE_STRARRAY(socket_families);
+
+#ifndef SOCK_TYPE_MASK
+#define SOCK_TYPE_MASK 0xf
+#endif
+
+static size_t syscall_arg__scnprintf_socket_type(char *bf, size_t size,
+						      struct syscall_arg *arg)
+{
+	size_t printed;
+	int type = arg->val,
+	    flags = type & ~SOCK_TYPE_MASK;
+
+	type &= SOCK_TYPE_MASK;
+	/*
+ 	 * Can't use a strarray, MIPS may override for ABI reasons.
+ 	 */
+	switch (type) {
+#define	P_SK_TYPE(n) case SOCK_##n: printed = scnprintf(bf, size, #n); break;
+	P_SK_TYPE(STREAM);
+	P_SK_TYPE(DGRAM);
+	P_SK_TYPE(RAW);
+	P_SK_TYPE(RDM);
+	P_SK_TYPE(SEQPACKET);
+	P_SK_TYPE(DCCP);
+	P_SK_TYPE(PACKET);
+#undef P_SK_TYPE
+	default:
+		printed = scnprintf(bf, size, "%#x", type);
+	}
+
+#define	P_SK_FLAG(n) \
+	if (flags & SOCK_##n) { \
+		printed += scnprintf(bf + printed, size - printed, "|%s", #n); \
+		flags &= ~SOCK_##n; \
+	}
+
+	P_SK_FLAG(CLOEXEC);
+	P_SK_FLAG(NONBLOCK);
+#undef P_SK_FLAG
+
+	if (flags)
+		printed += scnprintf(bf + printed, size - printed, "|%#x", flags);
+
+	return printed;
+}
+
+#define SCA_SK_TYPE syscall_arg__scnprintf_socket_type
+
+#ifndef MSG_PROBE
+#define MSG_PROBE	     0x10
+#endif
+#ifndef MSG_WAITFORONE
+#define MSG_WAITFORONE	0x10000
+#endif
+#ifndef MSG_SENDPAGE_NOTLAST
+#define MSG_SENDPAGE_NOTLAST 0x20000
+#endif
+#ifndef MSG_FASTOPEN
+#define MSG_FASTOPEN	     0x20000000
+#endif
+
+static size_t syscall_arg__scnprintf_msg_flags(char *bf, size_t size,
+					       struct syscall_arg *arg)
+{
+	int printed = 0, flags = arg->val;
+
+	if (flags == 0)
+		return scnprintf(bf, size, "NONE");
+#define	P_MSG_FLAG(n) \
+	if (flags & MSG_##n) { \
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
+		flags &= ~MSG_##n; \
+	}
+
+	P_MSG_FLAG(OOB);
+	P_MSG_FLAG(PEEK);
+	P_MSG_FLAG(DONTROUTE);
+	P_MSG_FLAG(TRYHARD);
+	P_MSG_FLAG(CTRUNC);
+	P_MSG_FLAG(PROBE);
+	P_MSG_FLAG(TRUNC);
+	P_MSG_FLAG(DONTWAIT);
+	P_MSG_FLAG(EOR);
+	P_MSG_FLAG(WAITALL);
+	P_MSG_FLAG(FIN);
+	P_MSG_FLAG(SYN);
+	P_MSG_FLAG(CONFIRM);
+	P_MSG_FLAG(RST);
+	P_MSG_FLAG(ERRQUEUE);
+	P_MSG_FLAG(NOSIGNAL);
+	P_MSG_FLAG(MORE);
+	P_MSG_FLAG(WAITFORONE);
+	P_MSG_FLAG(SENDPAGE_NOTLAST);
+	P_MSG_FLAG(FASTOPEN);
+	P_MSG_FLAG(CMSG_CLOEXEC);
+#undef P_MSG_FLAG
+
+	if (flags)
+		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);
+
+	return printed;
+}
+
+#define SCA_MSG_FLAGS syscall_arg__scnprintf_msg_flags
+
+static size_t syscall_arg__scnprintf_access_mode(char *bf, size_t size,
+						 struct syscall_arg *arg)
+{
+	size_t printed = 0;
+	int mode = arg->val;
+
+	if (mode == F_OK) /* 0 */
+		return scnprintf(bf, size, "F");
+#define	P_MODE(n) \
+	if (mode & n##_OK) { \
+		printed += scnprintf(bf + printed, size - printed, "%s", #n); \
+		mode &= ~n##_OK; \
+	}
+
+	P_MODE(R);
+	P_MODE(W);
+	P_MODE(X);
+#undef P_MODE
+
+	if (mode)
+		printed += scnprintf(bf + printed, size - printed, "|%#x", mode);
+
+	return printed;
+}
+
+#define SCA_ACCMODE syscall_arg__scnprintf_access_mode
+
+static size_t syscall_arg__scnprintf_open_flags(char *bf, size_t size,
+					       struct syscall_arg *arg)
+{
+	int printed = 0, flags = arg->val;
+
+	if (!(flags & O_CREAT))
+		arg->mask |= 1 << (arg->idx + 1); /* Mask the mode parm */
+
+	if (flags == 0)
+		return scnprintf(bf, size, "RDONLY");
+#define	P_FLAG(n) \
+	if (flags & O_##n) { \
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
+		flags &= ~O_##n; \
+	}
+
+	P_FLAG(APPEND);
+	P_FLAG(ASYNC);
+	P_FLAG(CLOEXEC);
+	P_FLAG(CREAT);
+	P_FLAG(DIRECT);
+	P_FLAG(DIRECTORY);
+	P_FLAG(EXCL);
+	P_FLAG(LARGEFILE);
+	P_FLAG(NOATIME);
+	P_FLAG(NOCTTY);
+#ifdef O_NONBLOCK
+	P_FLAG(NONBLOCK);
+#elif O_NDELAY
+	P_FLAG(NDELAY);
+#endif
+#ifdef O_PATH
+	P_FLAG(PATH);
+#endif
+	P_FLAG(RDWR);
+#ifdef O_DSYNC
+	if ((flags & O_SYNC) == O_SYNC)
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", "SYNC");
+	else {
+		P_FLAG(DSYNC);
+	}
+#else
+	P_FLAG(SYNC);
+#endif
+	P_FLAG(TRUNC);
+	P_FLAG(WRONLY);
+#undef P_FLAG
+
+	if (flags)
+		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);
+
+	return printed;
+}
+
+#define SCA_OPEN_FLAGS syscall_arg__scnprintf_open_flags
+
+static size_t syscall_arg__scnprintf_eventfd_flags(char *bf, size_t size,
+						   struct syscall_arg *arg)
+{
+	int printed = 0, flags = arg->val;
+
+	if (flags == 0)
+		return scnprintf(bf, size, "NONE");
+#define	P_FLAG(n) \
+	if (flags & EFD_##n) { \
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
+		flags &= ~EFD_##n; \
+	}
+
+	P_FLAG(SEMAPHORE);
+	P_FLAG(CLOEXEC);
+	P_FLAG(NONBLOCK);
+#undef P_FLAG
+
+	if (flags)
+		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);
+
+	return printed;
+}
+
+#define SCA_EFD_FLAGS syscall_arg__scnprintf_eventfd_flags
+
+static size_t syscall_arg__scnprintf_pipe_flags(char *bf, size_t size,
+						struct syscall_arg *arg)
+{
+	int printed = 0, flags = arg->val;
+
+#define	P_FLAG(n) \
+	if (flags & O_##n) { \
+		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
+		flags &= ~O_##n; \
+	}
+
+	P_FLAG(CLOEXEC);
+	P_FLAG(NONBLOCK);
+#undef P_FLAG
+
+	if (flags)
+		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);
+
+	return printed;
+}
+
+#define SCA_PIPE_FLAGS syscall_arg__scnprintf_pipe_flags
+
+static size_t syscall_arg__scnprintf_signum(char *bf, size_t size, struct syscall_arg *arg)
+{
+	int sig = arg->val;
+
+	switch (sig) {
+#define	P_SIGNUM(n) case SIG##n: return scnprintf(bf, size, #n)
+	P_SIGNUM(HUP);
+	P_SIGNUM(INT);
+	P_SIGNUM(QUIT);
+	P_SIGNUM(ILL);
+	P_SIGNUM(TRAP);
+	P_SIGNUM(ABRT);
+	P_SIGNUM(BUS);
+	P_SIGNUM(FPE);
+	P_SIGNUM(KILL);
+	P_SIGNUM(USR1);
+	P_SIGNUM(SEGV);
+	P_SIGNUM(USR2);
+	P_SIGNUM(PIPE);
+	P_SIGNUM(ALRM);
+	P_SIGNUM(TERM);
+	P_SIGNUM(CHLD);
+	P_SIGNUM(CONT);
+	P_SIGNUM(STOP);
+	P_SIGNUM(TSTP);
+	P_SIGNUM(TTIN);
+	P_SIGNUM(TTOU);
+	P_SIGNUM(URG);
+	P_SIGNUM(XCPU);
+	P_SIGNUM(XFSZ);
+	P_SIGNUM(VTALRM);
+	P_SIGNUM(PROF);
+	P_SIGNUM(WINCH);
+	P_SIGNUM(IO);
+	P_SIGNUM(PWR);
+	P_SIGNUM(SYS);
+#ifdef SIGEMT
+	P_SIGNUM(EMT);
+#endif
+#ifdef SIGSTKFLT
+	P_SIGNUM(STKFLT);
+#endif
+#ifdef SIGSWI
+	P_SIGNUM(SWI);
+#endif
+	default: break;
+	}
+
+	return scnprintf(bf, size, "%#x", sig);
+}
+
+#define SCA_SIGNUM syscall_arg__scnprintf_signum
+
+#if defined(__i386__) || defined(__x86_64__)
+/*
+ * FIXME: Make this available to all arches.
+ */
+#define TCGETS		0x5401
+
+static const char *tioctls[] = {
+	"TCGETS", "TCSETS", "TCSETSW", "TCSETSF", "TCGETA", "TCSETA", "TCSETAW",
+	"TCSETAF", "TCSBRK", "TCXONC", "TCFLSH", "TIOCEXCL", "TIOCNXCL",
+	"TIOCSCTTY", "TIOCGPGRP", "TIOCSPGRP", "TIOCOUTQ", "TIOCSTI",
+	"TIOCGWINSZ", "TIOCSWINSZ", "TIOCMGET", "TIOCMBIS", "TIOCMBIC",
+	"TIOCMSET", "TIOCGSOFTCAR", "TIOCSSOFTCAR", "FIONREAD", "TIOCLINUX",
+	"TIOCCONS", "TIOCGSERIAL", "TIOCSSERIAL", "TIOCPKT", "FIONBIO",
+	"TIOCNOTTY", "TIOCSETD", "TIOCGETD", "TCSBRKP", [0x27] = "TIOCSBRK",
+	"TIOCCBRK", "TIOCGSID", "TCGETS2", "TCSETS2", "TCSETSW2", "TCSETSF2",
+	"TIOCGRS485", "TIOCSRS485", "TIOCGPTN", "TIOCSPTLCK",
+	"TIOCGDEV||TCGETX", "TCSETX", "TCSETXF", "TCSETXW", "TIOCSIG",
+	"TIOCVHANGUP", "TIOCGPKT", "TIOCGPTLCK", "TIOCGEXCL",
+	[0x50] = "FIONCLEX", "FIOCLEX", "FIOASYNC", "TIOCSERCONFIG",
+	"TIOCSERGWILD", "TIOCSERSWILD", "TIOCGLCKTRMIOS", "TIOCSLCKTRMIOS",
+	"TIOCSERGSTRUCT", "TIOCSERGETLSR", "TIOCSERGETMULTI", "TIOCSERSETMULTI",
+	"TIOCMIWAIT", "TIOCGICOUNT", [0x60] = "FIOQSIZE",
+};
+
+static DEFINE_STRARRAY_OFFSET(tioctls, 0x5401);
+#endif /* defined(__i386__) || defined(__x86_64__) */
+
+#define STRARRAY(arg, name, array) \
+	  .arg_scnprintf = { [arg] = SCA_STRARRAY, }, \
+	  .arg_parm	 = { [arg] = &strarray__##array, }
+
+static struct syscall_fmt {
+	const char *name;
+	const char *alias;
+	size_t	   (*arg_scnprintf[6])(char *bf, size_t size, struct syscall_arg *arg);
+	void	   *arg_parm[6];
+	bool	   errmsg;
+	bool	   timeout;
+	bool	   hexret;
+} syscall_fmts[] = {
+	{ .name	    = "access",	    .errmsg = true,
+	  .arg_scnprintf = { [1] = SCA_ACCMODE, /* mode */ }, },
+	{ .name	    = "arch_prctl", .errmsg = true, .alias = "prctl", },
+	{ .name	    = "brk",	    .hexret = true,
+	  .arg_scnprintf = { [0] = SCA_HEX, /* brk */ }, },
+	{ .name     = "clock_gettime",  .errmsg = true, STRARRAY(0, clk_id, clockid), },
+	{ .name	    = "close",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_CLOSE_FD, /* fd */ }, },
+	{ .name	    = "connect",    .errmsg = true, },
+	{ .name	    = "dup",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "dup2",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "dup3",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "epoll_ctl",  .errmsg = true, STRARRAY(1, op, epoll_ctl_ops), },
+	{ .name	    = "eventfd2",   .errmsg = true,
+	  .arg_scnprintf = { [1] = SCA_EFD_FLAGS, /* flags */ }, },
+	{ .name	    = "faccessat",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	{ .name	    = "fadvise64",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "fallocate",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "fchdir",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "fchmod",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "fchmodat",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
+	{ .name	    = "fchown",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "fchownat",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
+	{ .name	    = "fcntl",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
+			     [1] = SCA_STRARRAY, /* cmd */ },
+	  .arg_parm	 = { [1] = &strarray__fcntl_cmds, /* cmd */ }, },
+	{ .name	    = "fdatasync",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "flock",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
+			     [1] = SCA_FLOCK, /* cmd */ }, },
+	{ .name	    = "fsetxattr",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "fstat",	    .errmsg = true, .alias = "newfstat",
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "fstatat",    .errmsg = true, .alias = "newfstatat",
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	{ .name	    = "fstatfs",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "fsync",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "ftruncate", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "futex",	    .errmsg = true,
+	  .arg_scnprintf = { [1] = SCA_FUTEX_OP, /* op */ }, },
+	{ .name	    = "futimesat", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
+	{ .name	    = "getdents",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "getdents64", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "getitimer",  .errmsg = true, STRARRAY(0, which, itimers), },
+	{ .name	    = "getrlimit",  .errmsg = true, STRARRAY(0, resource, rlimit_resources), },
+	{ .name	    = "ioctl",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
+#if defined(__i386__) || defined(__x86_64__)
+/*
+ * FIXME: Make this available to all arches.
+ */
+			     [1] = SCA_STRHEXARRAY, /* cmd */
+			     [2] = SCA_HEX, /* arg */ },
+	  .arg_parm	 = { [1] = &strarray__tioctls, /* cmd */ }, },
+#else
+			     [2] = SCA_HEX, /* arg */ }, },
+#endif
+	{ .name	    = "kill",	    .errmsg = true,
+	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
+	{ .name	    = "linkat",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
+	{ .name	    = "lseek",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
+			     [2] = SCA_STRARRAY, /* whence */ },
+	  .arg_parm	 = { [2] = &strarray__whences, /* whence */ }, },
+	{ .name	    = "lstat",	    .errmsg = true, .alias = "newlstat", },
+	{ .name     = "madvise",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_HEX,	 /* start */
+			     [2] = SCA_MADV_BHV, /* behavior */ }, },
+	{ .name	    = "mkdirat",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
+	{ .name	    = "mknodat",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
+	{ .name	    = "mlock",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
+	{ .name	    = "mlockall",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
+	{ .name	    = "mmap",	    .hexret = true,
+	  .arg_scnprintf = { [0] = SCA_HEX,	  /* addr */
+			     [2] = SCA_MMAP_PROT, /* prot */
+			     [3] = SCA_MMAP_FLAGS, /* flags */
+			     [4] = SCA_FD, 	  /* fd */ }, },
+	{ .name	    = "mprotect",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_HEX, /* start */
+			     [2] = SCA_MMAP_PROT, /* prot */ }, },
+	{ .name	    = "mremap",	    .hexret = true,
+	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */
+			     [3] = SCA_MREMAP_FLAGS, /* flags */
+			     [4] = SCA_HEX, /* new_addr */ }, },
+	{ .name	    = "munlock",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
+	{ .name	    = "munmap",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
+	{ .name	    = "name_to_handle_at", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	{ .name	    = "newfstatat", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	{ .name	    = "open",	    .errmsg = true,
+	  .arg_scnprintf = { [1] = SCA_OPEN_FLAGS, /* flags */ }, },
+	{ .name	    = "open_by_handle_at", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
+			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
+	{ .name	    = "openat",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
+			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
+	{ .name	    = "pipe2",	    .errmsg = true,
+	  .arg_scnprintf = { [1] = SCA_PIPE_FLAGS, /* flags */ }, },
+	{ .name	    = "poll",	    .errmsg = true, .timeout = true, },
+	{ .name	    = "ppoll",	    .errmsg = true, .timeout = true, },
+	{ .name	    = "pread",	    .errmsg = true, .alias = "pread64",
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "preadv",	    .errmsg = true, .alias = "pread",
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "prlimit64",  .errmsg = true, STRARRAY(1, resource, rlimit_resources), },
+	{ .name	    = "pwrite",	    .errmsg = true, .alias = "pwrite64",
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "pwritev",    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "read",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "readlinkat", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	{ .name	    = "readv",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "recvfrom",   .errmsg = true,
+	  .arg_scnprintf = { [3] = SCA_MSG_FLAGS, /* flags */ }, },
+	{ .name	    = "recvmmsg",   .errmsg = true,
+	  .arg_scnprintf = { [3] = SCA_MSG_FLAGS, /* flags */ }, },
+	{ .name	    = "recvmsg",    .errmsg = true,
+	  .arg_scnprintf = { [2] = SCA_MSG_FLAGS, /* flags */ }, },
+	{ .name	    = "renameat",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	{ .name	    = "rt_sigaction", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_SIGNUM, /* sig */ }, },
+	{ .name	    = "rt_sigprocmask",  .errmsg = true, STRARRAY(0, how, sighow), },
+	{ .name	    = "rt_sigqueueinfo", .errmsg = true,
+	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
+	{ .name	    = "rt_tgsigqueueinfo", .errmsg = true,
+	  .arg_scnprintf = { [2] = SCA_SIGNUM, /* sig */ }, },
+	{ .name	    = "select",	    .errmsg = true, .timeout = true, },
+	{ .name	    = "sendmmsg",    .errmsg = true,
+	  .arg_scnprintf = { [3] = SCA_MSG_FLAGS, /* flags */ }, },
+	{ .name	    = "sendmsg",    .errmsg = true,
+	  .arg_scnprintf = { [2] = SCA_MSG_FLAGS, /* flags */ }, },
+	{ .name	    = "sendto",	    .errmsg = true,
+	  .arg_scnprintf = { [3] = SCA_MSG_FLAGS, /* flags */ }, },
+	{ .name	    = "setitimer",  .errmsg = true, STRARRAY(0, which, itimers), },
+	{ .name	    = "setrlimit",  .errmsg = true, STRARRAY(0, resource, rlimit_resources), },
+	{ .name	    = "shutdown",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "socket",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_STRARRAY, /* family */
+			     [1] = SCA_SK_TYPE, /* type */ },
+	  .arg_parm	 = { [0] = &strarray__socket_families, /* family */ }, },
+	{ .name	    = "socketpair", .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_STRARRAY, /* family */
+			     [1] = SCA_SK_TYPE, /* type */ },
+	  .arg_parm	 = { [0] = &strarray__socket_families, /* family */ }, },
+	{ .name	    = "stat",	    .errmsg = true, .alias = "newstat", },
+	{ .name	    = "symlinkat",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	{ .name	    = "tgkill",	    .errmsg = true,
+	  .arg_scnprintf = { [2] = SCA_SIGNUM, /* sig */ }, },
+	{ .name	    = "tkill",	    .errmsg = true,
+	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
+	{ .name	    = "uname",	    .errmsg = true, .alias = "newuname", },
+	{ .name	    = "unlinkat",   .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
+	{ .name	    = "utimensat",  .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FDAT, /* dirfd */ }, },
+	{ .name	    = "write",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+	{ .name	    = "writev",	    .errmsg = true,
+	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
+};
+
+static int syscall_fmt__cmp(const void *name, const void *fmtp)
+{
+	const struct syscall_fmt *fmt = fmtp;
+	return strcmp(name, fmt->name);
+}
+
+static struct syscall_fmt *syscall_fmt__find(const char *name)
+{
+	const int nmemb = ARRAY_SIZE(syscall_fmts);
+	return bsearch(name, syscall_fmts, nmemb, sizeof(struct syscall_fmt), syscall_fmt__cmp);
+}
+
+struct syscall {
+	struct event_format *tp_format;
+	int		    nr_args;
+	struct format_field *args;
+	const char	    *name;
+	bool		    filtered;
+	bool		    is_exit;
+	struct syscall_fmt  *fmt;
+	size_t		    (**arg_scnprintf)(char *bf, size_t size, struct syscall_arg *arg);
+	void		    **arg_parm;
+};
+
+static size_t fprintf_duration(unsigned long t, FILE *fp)
+{
+	double duration = (double)t / NSEC_PER_MSEC;
+	size_t printed = fprintf(fp, "(");
+
+	if (duration >= 1.0)
+		printed += color_fprintf(fp, PERF_COLOR_RED, "%6.3f ms", duration);
+	else if (duration >= 0.01)
+		printed += color_fprintf(fp, PERF_COLOR_YELLOW, "%6.3f ms", duration);
+	else
+		printed += color_fprintf(fp, PERF_COLOR_NORMAL, "%6.3f ms", duration);
+	return printed + fprintf(fp, "): ");
+}
+
+struct thread_trace {
+	u64		  entry_time;
+	u64		  exit_time;
+	bool		  entry_pending;
+	unsigned long	  nr_events;
+	unsigned long	  pfmaj, pfmin;
+	char		  *entry_str;
+	double		  runtime_ms;
+	struct {
+		int	  max;
+		char	  **table;
+	} paths;
+
+	struct intlist *syscall_stats;
+};
+
+static struct thread_trace *thread_trace__new(void)
+{
+	struct thread_trace *ttrace =  zalloc(sizeof(struct thread_trace));
+
+	if (ttrace)
+		ttrace->paths.max = -1;
+
+	ttrace->syscall_stats = intlist__new(NULL);
+
+	return ttrace;
+}
+
+static struct thread_trace *thread__trace(struct thread *thread, FILE *fp)
+{
+	struct thread_trace *ttrace;
+
+	if (thread == NULL)
+		goto fail;
+
+	if (thread__priv(thread) == NULL)
+		thread__set_priv(thread, thread_trace__new());
+
+	if (thread__priv(thread) == NULL)
+		goto fail;
+
+	ttrace = thread__priv(thread);
+	++ttrace->nr_events;
+
+	return ttrace;
+fail:
+	color_fprintf(fp, PERF_COLOR_RED,
+		      "WARNING: not enough memory, dropping samples!\n");
+	return NULL;
+}
+
+#define TRACE_PFMAJ		(1 << 0)
+#define TRACE_PFMIN		(1 << 1)
+
+struct trace {
+	struct perf_tool	tool;
+	struct {
+		int		machine;
+		int		open_id;
+	}			audit;
+	struct {
+		int		max;
+		struct syscall  *table;
+	} syscalls;
+	struct record_opts	opts;
+	struct perf_evlist	*evlist;
+	struct machine		*host;
+	struct thread		*current;
+	u64			base_time;
+	FILE			*output;
+	unsigned long		nr_events;
+	struct strlist		*ev_qualifier;
+	const char 		*last_vfs_getname;
+	struct intlist		*tid_list;
+	struct intlist		*pid_list;
+	struct {
+		size_t		nr;
+		pid_t		*entries;
+	}			filter_pids;
+	double			duration_filter;
+	double			runtime_ms;
+	struct {
+		u64		vfs_getname,
+				proc_getname;
+	} stats;
+	bool			not_ev_qualifier;
+	bool			live;
+	bool			full_time;
+	bool			sched;
+	bool			multiple_threads;
+	bool			summary;
+	bool			summary_only;
+	bool			show_comm;
+	bool			show_tool_stats;
+	bool			trace_syscalls;
+	bool			force;
+	int			trace_pgfaults;
+};
+
+static int trace__set_fd_pathname(struct thread *thread, int fd, const char *pathname)
+{
+	struct thread_trace *ttrace = thread__priv(thread);
+
+	if (fd > ttrace->paths.max) {
+		char **npath = realloc(ttrace->paths.table, (fd + 1) * sizeof(char *));
+
+		if (npath == NULL)
+			return -1;
+
+		if (ttrace->paths.max != -1) {
+			memset(npath + ttrace->paths.max + 1, 0,
+			       (fd - ttrace->paths.max) * sizeof(char *));
+		} else {
+			memset(npath, 0, (fd + 1) * sizeof(char *));
+		}
+
+		ttrace->paths.table = npath;
+		ttrace->paths.max   = fd;
+	}
+
+	ttrace->paths.table[fd] = strdup(pathname);
+
+	return ttrace->paths.table[fd] != NULL ? 0 : -1;
+}
+
+static int thread__read_fd_path(struct thread *thread, int fd)
+{
+	char linkname[PATH_MAX], pathname[PATH_MAX];
+	struct stat st;
+	int ret;
+
+	if (thread->pid_ == thread->tid) {
+		scnprintf(linkname, sizeof(linkname),
+			  "/proc/%d/fd/%d", thread->pid_, fd);
+	} else {
+		scnprintf(linkname, sizeof(linkname),
+			  "/proc/%d/task/%d/fd/%d", thread->pid_, thread->tid, fd);
+	}
+
+	if (lstat(linkname, &st) < 0 || st.st_size + 1 > (off_t)sizeof(pathname))
+		return -1;
+
+	ret = readlink(linkname, pathname, sizeof(pathname));
+
+	if (ret < 0 || ret > st.st_size)
+		return -1;
+
+	pathname[ret] = '\0';
+	return trace__set_fd_pathname(thread, fd, pathname);
+}
+
+static const char *thread__fd_path(struct thread *thread, int fd,
+				   struct trace *trace)
+{
+	struct thread_trace *ttrace = thread__priv(thread);
+
+	if (ttrace == NULL)
+		return NULL;
+
+	if (fd < 0)
+		return NULL;
+
+	if ((fd > ttrace->paths.max || ttrace->paths.table[fd] == NULL)) {
+		if (!trace->live)
+			return NULL;
+		++trace->stats.proc_getname;
+		if (thread__read_fd_path(thread, fd))
+			return NULL;
+	}
+
+	return ttrace->paths.table[fd];
+}
+
+static size_t syscall_arg__scnprintf_fd(char *bf, size_t size,
+					struct syscall_arg *arg)
+{
+	int fd = arg->val;
+	size_t printed = scnprintf(bf, size, "%d", fd);
+	const char *path = thread__fd_path(arg->thread, fd, arg->trace);
+
+	if (path)
+		printed += scnprintf(bf + printed, size - printed, "<%s>", path);
+
+	return printed;
+}
+
+static size_t syscall_arg__scnprintf_close_fd(char *bf, size_t size,
+					      struct syscall_arg *arg)
+{
+	int fd = arg->val;
+	size_t printed = syscall_arg__scnprintf_fd(bf, size, arg);
+	struct thread_trace *ttrace = thread__priv(arg->thread);
+
+	if (ttrace && fd >= 0 && fd <= ttrace->paths.max)
+		zfree(&ttrace->paths.table[fd]);
+
+	return printed;
+}
+
+static bool trace__filter_duration(struct trace *trace, double t)
+{
+	return t < (trace->duration_filter * NSEC_PER_MSEC);
+}
+
+static size_t trace__fprintf_tstamp(struct trace *trace, u64 tstamp, FILE *fp)
+{
+	double ts = (double)(tstamp - trace->base_time) / NSEC_PER_MSEC;
+
+	return fprintf(fp, "%10.3f ", ts);
+}
+
+static bool done = false;
+static bool interrupted = false;
+
+static void sig_handler(int sig)
+{
+	done = true;
+	interrupted = sig == SIGINT;
+}
+
+static size_t trace__fprintf_entry_head(struct trace *trace, struct thread *thread,
+					u64 duration, u64 tstamp, FILE *fp)
+{
+	size_t printed = trace__fprintf_tstamp(trace, tstamp, fp);
+	printed += fprintf_duration(duration, fp);
+
+	if (trace->multiple_threads) {
+		if (trace->show_comm)
+			printed += fprintf(fp, "%.14s/", thread__comm_str(thread));
+		printed += fprintf(fp, "%d ", thread->tid);
+	}
+
+	return printed;
+}
+
+static int trace__process_event(struct trace *trace, struct machine *machine,
+				union perf_event *event, struct perf_sample *sample)
+{
+	int ret = 0;
+
+	switch (event->header.type) {
+	case PERF_RECORD_LOST:
+		color_fprintf(trace->output, PERF_COLOR_RED,
+			      "LOST %" PRIu64 " events!\n", event->lost.lost);
+		ret = machine__process_lost_event(machine, event, sample);
+	default:
+		ret = machine__process_event(machine, event, sample);
+		break;
+	}
+
+	return ret;
+}
+
+static int trace__tool_process(struct perf_tool *tool,
+			       union perf_event *event,
+			       struct perf_sample *sample,
+			       struct machine *machine)
+{
+	struct trace *trace = container_of(tool, struct trace, tool);
+	return trace__process_event(trace, machine, event, sample);
+}
+
+static int trace__symbols_init(struct trace *trace, struct perf_evlist *evlist)
+{
+	int err = symbol__init(NULL);
+
+	if (err)
+		return err;
+
+	trace->host = machine__new_host();
+	if (trace->host == NULL)
+		return -ENOMEM;
+
+	err = __machine__synthesize_threads(trace->host, &trace->tool, &trace->opts.target,
+					    evlist->threads, trace__tool_process, false);
+	if (err)
+		symbol__exit();
+
+	return err;
+}
+
+static int syscall__set_arg_fmts(struct syscall *sc)
+{
+	struct format_field *field;
+	int idx = 0;
+
+	sc->arg_scnprintf = calloc(sc->nr_args, sizeof(void *));
+	if (sc->arg_scnprintf == NULL)
+		return -1;
+
+	if (sc->fmt)
+		sc->arg_parm = sc->fmt->arg_parm;
+
+	for (field = sc->args; field; field = field->next) {
+		if (sc->fmt && sc->fmt->arg_scnprintf[idx])
+			sc->arg_scnprintf[idx] = sc->fmt->arg_scnprintf[idx];
+		else if (field->flags & FIELD_IS_POINTER)
+			sc->arg_scnprintf[idx] = syscall_arg__scnprintf_hex;
+		++idx;
+	}
+
+	return 0;
+}
+
+static int trace__read_syscall_info(struct trace *trace, int id)
+{
+	char tp_name[128];
+	struct syscall *sc;
+	const char *name = audit_syscall_to_name(id, trace->audit.machine);
+
+	if (name == NULL)
+		return -1;
+
+	if (id > trace->syscalls.max) {
+		struct syscall *nsyscalls = realloc(trace->syscalls.table, (id + 1) * sizeof(*sc));
+
+		if (nsyscalls == NULL)
+			return -1;
+
+		if (trace->syscalls.max != -1) {
+			memset(nsyscalls + trace->syscalls.max + 1, 0,
+			       (id - trace->syscalls.max) * sizeof(*sc));
+		} else {
+			memset(nsyscalls, 0, (id + 1) * sizeof(*sc));
+		}
+
+		trace->syscalls.table = nsyscalls;
+		trace->syscalls.max   = id;
+	}
+
+	sc = trace->syscalls.table + id;
+	sc->name = name;
+
+	if (trace->ev_qualifier) {
+		bool in = strlist__find(trace->ev_qualifier, name) != NULL;
+
+		if (!(in ^ trace->not_ev_qualifier)) {
+			sc->filtered = true;
+			/*
+			 * No need to do read tracepoint information since this will be
+			 * filtered out.
+			 */
+			return 0;
+		}
+	}
+
+	sc->fmt  = syscall_fmt__find(sc->name);
+
+	snprintf(tp_name, sizeof(tp_name), "sys_enter_%s", sc->name);
+	sc->tp_format = trace_event__tp_format("syscalls", tp_name);
+
+	if (sc->tp_format == NULL && sc->fmt && sc->fmt->alias) {
+		snprintf(tp_name, sizeof(tp_name), "sys_enter_%s", sc->fmt->alias);
+		sc->tp_format = trace_event__tp_format("syscalls", tp_name);
+	}
+
+	if (sc->tp_format == NULL)
+		return -1;
+
+	sc->args = sc->tp_format->format.fields;
+	sc->nr_args = sc->tp_format->format.nr_fields;
+	/* drop nr field - not relevant here; does not exist on older kernels */
+	if (sc->args && strcmp(sc->args->name, "nr") == 0) {
+		sc->args = sc->args->next;
+		--sc->nr_args;
+	}
+
+	sc->is_exit = !strcmp(name, "exit_group") || !strcmp(name, "exit");
+
+	return syscall__set_arg_fmts(sc);
+}
+
+/*
+ * args is to be interpreted as a series of longs but we need to handle
+ * 8-byte unaligned accesses. args points to raw_data within the event
+ * and raw_data is guaranteed to be 8-byte unaligned because it is
+ * preceded by raw_size which is a u32. So we need to copy args to a temp
+ * variable to read it. Most notably this avoids extended load instructions
+ * on unaligned addresses
+ */
+
+static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size,
+				      unsigned char *args, struct trace *trace,
+				      struct thread *thread)
+{
+	size_t printed = 0;
+	unsigned char *p;
+	unsigned long val;
+
+	if (sc->args != NULL) {
+		struct format_field *field;
+		u8 bit = 1;
+		struct syscall_arg arg = {
+			.idx	= 0,
+			.mask	= 0,
+			.trace  = trace,
+			.thread = thread,
+		};
+
+		for (field = sc->args; field;
+		     field = field->next, ++arg.idx, bit <<= 1) {
+			if (arg.mask & bit)
+				continue;
+
+			/* special care for unaligned accesses */
+			p = args + sizeof(unsigned long) * arg.idx;
+			memcpy(&val, p, sizeof(val));
+
+			/*
+ 			 * Suppress this argument if its value is zero and
+ 			 * and we don't have a string associated in an
+ 			 * strarray for it.
+ 			 */
+			if (val == 0 &&
+			    !(sc->arg_scnprintf &&
+			      sc->arg_scnprintf[arg.idx] == SCA_STRARRAY &&
+			      sc->arg_parm[arg.idx]))
+				continue;
+
+			printed += scnprintf(bf + printed, size - printed,
+					     "%s%s: ", printed ? ", " : "", field->name);
+			if (sc->arg_scnprintf && sc->arg_scnprintf[arg.idx]) {
+				arg.val = val;
+				if (sc->arg_parm)
+					arg.parm = sc->arg_parm[arg.idx];
+				printed += sc->arg_scnprintf[arg.idx](bf + printed,
+								      size - printed, &arg);
+			} else {
+				printed += scnprintf(bf + printed, size - printed,
+						     "%ld", val);
+			}
+		}
+	} else {
+		int i = 0;
+
+		while (i < 6) {
+			/* special care for unaligned accesses */
+			p = args + sizeof(unsigned long) * i;
+			memcpy(&val, p, sizeof(val));
+			printed += scnprintf(bf + printed, size - printed,
+					     "%sarg%d: %ld",
+					     printed ? ", " : "", i, val);
+			++i;
+		}
+	}
+
+	return printed;
+}
+
+typedef int (*tracepoint_handler)(struct trace *trace, struct perf_evsel *evsel,
+				  union perf_event *event,
+				  struct perf_sample *sample);
+
+static struct syscall *trace__syscall_info(struct trace *trace,
+					   struct perf_evsel *evsel, int id)
+{
+
+	if (id < 0) {
+
+		/*
+		 * XXX: Noticed on x86_64, reproduced as far back as 3.0.36, haven't tried
+		 * before that, leaving at a higher verbosity level till that is
+		 * explained. Reproduced with plain ftrace with:
+		 *
+		 * echo 1 > /t/events/raw_syscalls/sys_exit/enable
+		 * grep "NR -1 " /t/trace_pipe
+		 *
+		 * After generating some load on the machine.
+ 		 */
+		if (verbose > 1) {
+			static u64 n;
+			fprintf(trace->output, "Invalid syscall %d id, skipping (%s, %" PRIu64 ") ...\n",
+				id, perf_evsel__name(evsel), ++n);
+		}
+		return NULL;
+	}
+
+	if ((id > trace->syscalls.max || trace->syscalls.table[id].name == NULL) &&
+	    trace__read_syscall_info(trace, id))
+		goto out_cant_read;
+
+	if ((id > trace->syscalls.max || trace->syscalls.table[id].name == NULL))
+		goto out_cant_read;
+
+	return &trace->syscalls.table[id];
+
+out_cant_read:
+	if (verbose) {
+		fprintf(trace->output, "Problems reading syscall %d", id);
+		if (id <= trace->syscalls.max && trace->syscalls.table[id].name != NULL)
+			fprintf(trace->output, "(%s)", trace->syscalls.table[id].name);
+		fputs(" information\n", trace->output);
+	}
+	return NULL;
+}
+
+static void thread__update_stats(struct thread_trace *ttrace,
+				 int id, struct perf_sample *sample)
+{
+	struct int_node *inode;
+	struct stats *stats;
+	u64 duration = 0;
+
+	inode = intlist__findnew(ttrace->syscall_stats, id);
+	if (inode == NULL)
+		return;
+
+	stats = inode->priv;
+	if (stats == NULL) {
+		stats = malloc(sizeof(struct stats));
+		if (stats == NULL)
+			return;
+		init_stats(stats);
+		inode->priv = stats;
+	}
+
+	if (ttrace->entry_time && sample->time > ttrace->entry_time)
+		duration = sample->time - ttrace->entry_time;
+
+	update_stats(stats, duration);
+}
+
+static int trace__printf_interrupted_entry(struct trace *trace, struct perf_sample *sample)
+{
+	struct thread_trace *ttrace;
+	u64 duration;
+	size_t printed;
+
+	if (trace->current == NULL)
+		return 0;
+
+	ttrace = thread__priv(trace->current);
+
+	if (!ttrace->entry_pending)
+		return 0;
+
+	duration = sample->time - ttrace->entry_time;
+
+	printed  = trace__fprintf_entry_head(trace, trace->current, duration, sample->time, trace->output);
+	printed += fprintf(trace->output, "%-70s) ...\n", ttrace->entry_str);
+	ttrace->entry_pending = false;
+
+	return printed;
+}
+
+static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
+			    union perf_event *event __maybe_unused,
+			    struct perf_sample *sample)
+{
+	char *msg;
+	void *args;
+	size_t printed = 0;
+	struct thread *thread;
+	int id = perf_evsel__sc_tp_uint(evsel, id, sample);
+	struct syscall *sc = trace__syscall_info(trace, evsel, id);
+	struct thread_trace *ttrace;
+
+	if (sc == NULL)
+		return -1;
+
+	if (sc->filtered)
+		return 0;
+
+	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
+	ttrace = thread__trace(thread, trace->output);
+	if (ttrace == NULL)
+		return -1;
+
+	args = perf_evsel__sc_tp_ptr(evsel, args, sample);
+
+	if (ttrace->entry_str == NULL) {
+		ttrace->entry_str = malloc(1024);
+		if (!ttrace->entry_str)
+			return -1;
+	}
+
+	if (!trace->summary_only)
+		trace__printf_interrupted_entry(trace, sample);
+
+	ttrace->entry_time = sample->time;
+	msg = ttrace->entry_str;
+	printed += scnprintf(msg + printed, 1024 - printed, "%s(", sc->name);
+
+	printed += syscall__scnprintf_args(sc, msg + printed, 1024 - printed,
+					   args, trace, thread);
+
+	if (sc->is_exit) {
+		if (!trace->duration_filter && !trace->summary_only) {
+			trace__fprintf_entry_head(trace, thread, 1, sample->time, trace->output);
+			fprintf(trace->output, "%-70s\n", ttrace->entry_str);
+		}
+	} else
+		ttrace->entry_pending = true;
+
+	if (trace->current != thread) {
+		thread__put(trace->current);
+		trace->current = thread__get(thread);
+	}
+
+	return 0;
+}
+
+static int trace__sys_exit(struct trace *trace, struct perf_evsel *evsel,
+			   union perf_event *event __maybe_unused,
+			   struct perf_sample *sample)
+{
+	long ret;
+	u64 duration = 0;
+	struct thread *thread;
+	int id = perf_evsel__sc_tp_uint(evsel, id, sample);
+	struct syscall *sc = trace__syscall_info(trace, evsel, id);
+	struct thread_trace *ttrace;
+
+	if (sc == NULL)
+		return -1;
+
+	if (sc->filtered)
+		return 0;
+
+	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
+	ttrace = thread__trace(thread, trace->output);
+	if (ttrace == NULL)
+		return -1;
+
+	if (trace->summary)
+		thread__update_stats(ttrace, id, sample);
+
+	ret = perf_evsel__sc_tp_uint(evsel, ret, sample);
+
+	if (id == trace->audit.open_id && ret >= 0 && trace->last_vfs_getname) {
+		trace__set_fd_pathname(thread, ret, trace->last_vfs_getname);
+		trace->last_vfs_getname = NULL;
+		++trace->stats.vfs_getname;
+	}
+
+	ttrace->exit_time = sample->time;
+
+	if (ttrace->entry_time) {
+		duration = sample->time - ttrace->entry_time;
+		if (trace__filter_duration(trace, duration))
+			goto out;
+	} else if (trace->duration_filter)
+		goto out;
+
+	if (trace->summary_only)
+		goto out;
+
+	trace__fprintf_entry_head(trace, thread, duration, sample->time, trace->output);
+
+	if (ttrace->entry_pending) {
+		fprintf(trace->output, "%-70s", ttrace->entry_str);
+	} else {
+		fprintf(trace->output, " ... [");
+		color_fprintf(trace->output, PERF_COLOR_YELLOW, "continued");
+		fprintf(trace->output, "]: %s()", sc->name);
+	}
+
+	if (sc->fmt == NULL) {
+signed_print:
+		fprintf(trace->output, ") = %ld", ret);
+	} else if (ret < 0 && sc->fmt->errmsg) {
+		char bf[STRERR_BUFSIZE];
+		const char *emsg = strerror_r(-ret, bf, sizeof(bf)),
+			   *e = audit_errno_to_name(-ret);
+
+		fprintf(trace->output, ") = -1 %s %s", e, emsg);
+	} else if (ret == 0 && sc->fmt->timeout)
+		fprintf(trace->output, ") = 0 Timeout");
+	else if (sc->fmt->hexret)
+		fprintf(trace->output, ") = %#lx", ret);
+	else
+		goto signed_print;
+
+	fputc('\n', trace->output);
+out:
+	ttrace->entry_pending = false;
+
+	return 0;
+}
+
+static int trace__vfs_getname(struct trace *trace, struct perf_evsel *evsel,
+			      union perf_event *event __maybe_unused,
+			      struct perf_sample *sample)
+{
+	trace->last_vfs_getname = perf_evsel__rawptr(evsel, sample, "pathname");
+	return 0;
+}
+
+static int trace__sched_stat_runtime(struct trace *trace, struct perf_evsel *evsel,
+				     union perf_event *event __maybe_unused,
+				     struct perf_sample *sample)
+{
+        u64 runtime = perf_evsel__intval(evsel, sample, "runtime");
+	double runtime_ms = (double)runtime / NSEC_PER_MSEC;
+	struct thread *thread = machine__findnew_thread(trace->host,
+							sample->pid,
+							sample->tid);
+	struct thread_trace *ttrace = thread__trace(thread, trace->output);
+
+	if (ttrace == NULL)
+		goto out_dump;
+
+	ttrace->runtime_ms += runtime_ms;
+	trace->runtime_ms += runtime_ms;
+	return 0;
+
+out_dump:
+	fprintf(trace->output, "%s: comm=%s,pid=%u,runtime=%" PRIu64 ",vruntime=%" PRIu64 ")\n",
+	       evsel->name,
+	       perf_evsel__strval(evsel, sample, "comm"),
+	       (pid_t)perf_evsel__intval(evsel, sample, "pid"),
+	       runtime,
+	       perf_evsel__intval(evsel, sample, "vruntime"));
+	return 0;
+}
+
+static int trace__event_handler(struct trace *trace, struct perf_evsel *evsel,
+				union perf_event *event __maybe_unused,
+				struct perf_sample *sample)
+{
+	trace__printf_interrupted_entry(trace, sample);
+	trace__fprintf_tstamp(trace, sample->time, trace->output);
+
+	if (trace->trace_syscalls)
+		fprintf(trace->output, "(         ): ");
+
+	fprintf(trace->output, "%s:", evsel->name);
+
+	if (evsel->tp_format) {
+		event_format__fprintf(evsel->tp_format, sample->cpu,
+				      sample->raw_data, sample->raw_size,
+				      trace->output);
+	}
+
+	fprintf(trace->output, ")\n");
+	return 0;
+}
+
+static void print_location(FILE *f, struct perf_sample *sample,
+			   struct addr_location *al,
+			   bool print_dso, bool print_sym)
+{
+
+	if ((verbose || print_dso) && al->map)
+		fprintf(f, "%s@", al->map->dso->long_name);
+
+	if ((verbose || print_sym) && al->sym)
+		fprintf(f, "%s+0x%" PRIx64, al->sym->name,
+			al->addr - al->sym->start);
+	else if (al->map)
+		fprintf(f, "0x%" PRIx64, al->addr);
+	else
+		fprintf(f, "0x%" PRIx64, sample->addr);
+}
+
+static int trace__pgfault(struct trace *trace,
+			  struct perf_evsel *evsel,
+			  union perf_event *event,
+			  struct perf_sample *sample)
+{
+	struct thread *thread;
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct addr_location al;
+	char map_type = 'd';
+	struct thread_trace *ttrace;
+
+	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
+	ttrace = thread__trace(thread, trace->output);
+	if (ttrace == NULL)
+		return -1;
+
+	if (evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS_MAJ)
+		ttrace->pfmaj++;
+	else
+		ttrace->pfmin++;
+
+	if (trace->summary_only)
+		return 0;
+
+	thread__find_addr_location(thread, cpumode, MAP__FUNCTION,
+			      sample->ip, &al);
+
+	trace__fprintf_entry_head(trace, thread, 0, sample->time, trace->output);
+
+	fprintf(trace->output, "%sfault [",
+		evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS_MAJ ?
+		"maj" : "min");
+
+	print_location(trace->output, sample, &al, false, true);
+
+	fprintf(trace->output, "] => ");
+
+	thread__find_addr_location(thread, cpumode, MAP__VARIABLE,
+				   sample->addr, &al);
+
+	if (!al.map) {
+		thread__find_addr_location(thread, cpumode,
+					   MAP__FUNCTION, sample->addr, &al);
+
+		if (al.map)
+			map_type = 'x';
+		else
+			map_type = '?';
+	}
+
+	print_location(trace->output, sample, &al, true, false);
+
+	fprintf(trace->output, " (%c%c)\n", map_type, al.level);
+
+	return 0;
+}
+
+static bool skip_sample(struct trace *trace, struct perf_sample *sample)
+{
+	if ((trace->pid_list && intlist__find(trace->pid_list, sample->pid)) ||
+	    (trace->tid_list && intlist__find(trace->tid_list, sample->tid)))
+		return false;
+
+	if (trace->pid_list || trace->tid_list)
+		return true;
+
+	return false;
+}
+
+static int trace__process_sample(struct perf_tool *tool,
+				 union perf_event *event,
+				 struct perf_sample *sample,
+				 struct perf_evsel *evsel,
+				 struct machine *machine __maybe_unused)
+{
+	struct trace *trace = container_of(tool, struct trace, tool);
+	int err = 0;
+
+	tracepoint_handler handler = evsel->handler;
+
+	if (skip_sample(trace, sample))
+		return 0;
+
+	if (!trace->full_time && trace->base_time == 0)
+		trace->base_time = sample->time;
+
+	if (handler) {
+		++trace->nr_events;
+		handler(trace, evsel, event, sample);
+	}
+
+	return err;
+}
+
+static int parse_target_str(struct trace *trace)
+{
+	if (trace->opts.target.pid) {
+		trace->pid_list = intlist__new(trace->opts.target.pid);
+		if (trace->pid_list == NULL) {
+			pr_err("Error parsing process id string\n");
+			return -EINVAL;
+		}
+	}
+
+	if (trace->opts.target.tid) {
+		trace->tid_list = intlist__new(trace->opts.target.tid);
+		if (trace->tid_list == NULL) {
+			pr_err("Error parsing thread id string\n");
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int trace__record(struct trace *trace, int argc, const char **argv)
+{
+	unsigned int rec_argc, i, j;
+	const char **rec_argv;
+	const char * const record_args[] = {
+		"record",
+		"-R",
+		"-m", "1024",
+		"-c", "1",
+	};
+
+	const char * const sc_args[] = { "-e", };
+	unsigned int sc_args_nr = ARRAY_SIZE(sc_args);
+	const char * const majpf_args[] = { "-e", "major-faults" };
+	unsigned int majpf_args_nr = ARRAY_SIZE(majpf_args);
+	const char * const minpf_args[] = { "-e", "minor-faults" };
+	unsigned int minpf_args_nr = ARRAY_SIZE(minpf_args);
+
+	/* +1 is for the event string below */
+	rec_argc = ARRAY_SIZE(record_args) + sc_args_nr + 1 +
+		majpf_args_nr + minpf_args_nr + argc;
+	rec_argv = calloc(rec_argc + 1, sizeof(char *));
+
+	if (rec_argv == NULL)
+		return -ENOMEM;
+
+	j = 0;
+	for (i = 0; i < ARRAY_SIZE(record_args); i++)
+		rec_argv[j++] = record_args[i];
+
+	if (trace->trace_syscalls) {
+		for (i = 0; i < sc_args_nr; i++)
+			rec_argv[j++] = sc_args[i];
+
+		/* event string may be different for older kernels - e.g., RHEL6 */
+		if (is_valid_tracepoint("raw_syscalls:sys_enter"))
+			rec_argv[j++] = "raw_syscalls:sys_enter,raw_syscalls:sys_exit";
+		else if (is_valid_tracepoint("syscalls:sys_enter"))
+			rec_argv[j++] = "syscalls:sys_enter,syscalls:sys_exit";
+		else {
+			pr_err("Neither raw_syscalls nor syscalls events exist.\n");
+			return -1;
+		}
+	}
+
+	if (trace->trace_pgfaults & TRACE_PFMAJ)
+		for (i = 0; i < majpf_args_nr; i++)
+			rec_argv[j++] = majpf_args[i];
+
+	if (trace->trace_pgfaults & TRACE_PFMIN)
+		for (i = 0; i < minpf_args_nr; i++)
+			rec_argv[j++] = minpf_args[i];
+
+	for (i = 0; i < (unsigned int)argc; i++)
+		rec_argv[j++] = argv[i];
+
+	return cmd_record(j, rec_argv, NULL);
+}
+
+static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp);
+
+static void perf_evlist__add_vfs_getname(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evsel__newtp("probe", "vfs_getname");
+	if (evsel == NULL)
+		return;
+
+	if (perf_evsel__field(evsel, "pathname") == NULL) {
+		perf_evsel__delete(evsel);
+		return;
+	}
+
+	evsel->handler = trace__vfs_getname;
+	perf_evlist__add(evlist, evsel);
+}
+
+static int perf_evlist__add_pgfault(struct perf_evlist *evlist,
+				    u64 config)
+{
+	struct perf_evsel *evsel;
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_SOFTWARE,
+		.mmap_data = 1,
+	};
+
+	attr.config = config;
+	attr.sample_period = 1;
+
+	event_attr_init(&attr);
+
+	evsel = perf_evsel__new(&attr);
+	if (!evsel)
+		return -ENOMEM;
+
+	evsel->handler = trace__pgfault;
+	perf_evlist__add(evlist, evsel);
+
+	return 0;
+}
+
+static void trace__handle_event(struct trace *trace, union perf_event *event, struct perf_sample *sample)
+{
+	const u32 type = event->header.type;
+	struct perf_evsel *evsel;
+
+	if (!trace->full_time && trace->base_time == 0)
+		trace->base_time = sample->time;
+
+	if (type != PERF_RECORD_SAMPLE) {
+		trace__process_event(trace, trace->host, event, sample);
+		return;
+	}
+
+	evsel = perf_evlist__id2evsel(trace->evlist, sample->id);
+	if (evsel == NULL) {
+		fprintf(trace->output, "Unknown tp ID %" PRIu64 ", skipping...\n", sample->id);
+		return;
+	}
+
+	if (evsel->attr.type == PERF_TYPE_TRACEPOINT &&
+	    sample->raw_data == NULL) {
+		fprintf(trace->output, "%s sample with no payload for tid: %d, cpu %d, raw_size=%d, skipping...\n",
+		       perf_evsel__name(evsel), sample->tid,
+		       sample->cpu, sample->raw_size);
+	} else {
+		tracepoint_handler handler = evsel->handler;
+		handler(trace, evsel, event, sample);
+	}
+}
+
+static int trace__run(struct trace *trace, int argc, const char **argv)
+{
+	struct perf_evlist *evlist = trace->evlist;
+	int err = -1, i;
+	unsigned long before;
+	const bool forks = argc > 0;
+	bool draining = false;
+
+	trace->live = true;
+
+	if (trace->trace_syscalls &&
+	    perf_evlist__add_syscall_newtp(evlist, trace__sys_enter,
+					   trace__sys_exit))
+		goto out_error_raw_syscalls;
+
+	if (trace->trace_syscalls)
+		perf_evlist__add_vfs_getname(evlist);
+
+	if ((trace->trace_pgfaults & TRACE_PFMAJ) &&
+	    perf_evlist__add_pgfault(evlist, PERF_COUNT_SW_PAGE_FAULTS_MAJ)) {
+		goto out_error_mem;
+	}
+
+	if ((trace->trace_pgfaults & TRACE_PFMIN) &&
+	    perf_evlist__add_pgfault(evlist, PERF_COUNT_SW_PAGE_FAULTS_MIN))
+		goto out_error_mem;
+
+	if (trace->sched &&
+	    perf_evlist__add_newtp(evlist, "sched", "sched_stat_runtime",
+				   trace__sched_stat_runtime))
+		goto out_error_sched_stat_runtime;
+
+	err = perf_evlist__create_maps(evlist, &trace->opts.target);
+	if (err < 0) {
+		fprintf(trace->output, "Problems parsing the target to trace, check your options!\n");
+		goto out_delete_evlist;
+	}
+
+	err = trace__symbols_init(trace, evlist);
+	if (err < 0) {
+		fprintf(trace->output, "Problems initializing symbol libraries!\n");
+		goto out_delete_evlist;
+	}
+
+	perf_evlist__config(evlist, &trace->opts);
+
+	signal(SIGCHLD, sig_handler);
+	signal(SIGINT, sig_handler);
+
+	if (forks) {
+		err = perf_evlist__prepare_workload(evlist, &trace->opts.target,
+						    argv, false, NULL);
+		if (err < 0) {
+			fprintf(trace->output, "Couldn't run the workload!\n");
+			goto out_delete_evlist;
+		}
+	}
+
+	err = perf_evlist__open(evlist);
+	if (err < 0)
+		goto out_error_open;
+
+	/*
+	 * Better not use !target__has_task() here because we need to cover the
+	 * case where no threads were specified in the command line, but a
+	 * workload was, and in that case we will fill in the thread_map when
+	 * we fork the workload in perf_evlist__prepare_workload.
+	 */
+	if (trace->filter_pids.nr > 0)
+		err = perf_evlist__set_filter_pids(evlist, trace->filter_pids.nr, trace->filter_pids.entries);
+	else if (evlist->threads->map[0] == -1)
+		err = perf_evlist__set_filter_pid(evlist, getpid());
+
+	if (err < 0) {
+		printf("err=%d,%s\n", -err, strerror(-err));
+		exit(1);
+	}
+
+	err = perf_evlist__mmap(evlist, trace->opts.mmap_pages, false);
+	if (err < 0)
+		goto out_error_mmap;
+
+	if (!target__none(&trace->opts.target))
+		perf_evlist__enable(evlist);
+
+	if (forks)
+		perf_evlist__start_workload(evlist);
+
+	trace->multiple_threads = evlist->threads->map[0] == -1 ||
+				  evlist->threads->nr > 1 ||
+				  perf_evlist__first(evlist)->attr.inherit;
+again:
+	before = trace->nr_events;
+
+	for (i = 0; i < evlist->nr_mmaps; i++) {
+		union perf_event *event;
+
+		while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
+			struct perf_sample sample;
+
+			++trace->nr_events;
+
+			err = perf_evlist__parse_sample(evlist, event, &sample);
+			if (err) {
+				fprintf(trace->output, "Can't parse sample, err = %d, skipping...\n", err);
+				goto next_event;
+			}
+
+			trace__handle_event(trace, event, &sample);
+next_event:
+			perf_evlist__mmap_consume(evlist, i);
+
+			if (interrupted)
+				goto out_disable;
+
+			if (done && !draining) {
+				perf_evlist__disable(evlist);
+				draining = true;
+			}
+		}
+	}
+
+	if (trace->nr_events == before) {
+		int timeout = done ? 100 : -1;
+
+		if (!draining && perf_evlist__poll(evlist, timeout) > 0) {
+			if (perf_evlist__filter_pollfd(evlist, POLLERR | POLLHUP) == 0)
+				draining = true;
+
+			goto again;
+		}
+	} else {
+		goto again;
+	}
+
+out_disable:
+	thread__zput(trace->current);
+
+	perf_evlist__disable(evlist);
+
+	if (!err) {
+		if (trace->summary)
+			trace__fprintf_thread_summary(trace, trace->output);
+
+		if (trace->show_tool_stats) {
+			fprintf(trace->output, "Stats:\n "
+					       " vfs_getname : %" PRIu64 "\n"
+					       " proc_getname: %" PRIu64 "\n",
+				trace->stats.vfs_getname,
+				trace->stats.proc_getname);
+		}
+	}
+
+out_delete_evlist:
+	perf_evlist__delete(evlist);
+	trace->evlist = NULL;
+	trace->live = false;
+	return err;
+{
+	char errbuf[BUFSIZ];
+
+out_error_sched_stat_runtime:
+	debugfs__strerror_open_tp(errno, errbuf, sizeof(errbuf), "sched", "sched_stat_runtime");
+	goto out_error;
+
+out_error_raw_syscalls:
+	debugfs__strerror_open_tp(errno, errbuf, sizeof(errbuf), "raw_syscalls", "sys_(enter|exit)");
+	goto out_error;
+
+out_error_mmap:
+	perf_evlist__strerror_mmap(evlist, errno, errbuf, sizeof(errbuf));
+	goto out_error;
+
+out_error_open:
+	perf_evlist__strerror_open(evlist, errno, errbuf, sizeof(errbuf));
+
+out_error:
+	fprintf(trace->output, "%s\n", errbuf);
+	goto out_delete_evlist;
+}
+out_error_mem:
+	fprintf(trace->output, "Not enough memory to run!\n");
+	goto out_delete_evlist;
+}
+
+static int trace__replay(struct trace *trace)
+{
+	const struct perf_evsel_str_handler handlers[] = {
+		{ "probe:vfs_getname",	     trace__vfs_getname, },
+	};
+	struct perf_data_file file = {
+		.path  = input_name,
+		.mode  = PERF_DATA_MODE_READ,
+		.force = trace->force,
+	};
+	struct perf_session *session;
+	struct perf_evsel *evsel;
+	int err = -1;
+
+	trace->tool.sample	  = trace__process_sample;
+	trace->tool.mmap	  = perf_event__process_mmap;
+	trace->tool.mmap2	  = perf_event__process_mmap2;
+	trace->tool.comm	  = perf_event__process_comm;
+	trace->tool.exit	  = perf_event__process_exit;
+	trace->tool.fork	  = perf_event__process_fork;
+	trace->tool.attr	  = perf_event__process_attr;
+	trace->tool.tracing_data = perf_event__process_tracing_data;
+	trace->tool.build_id	  = perf_event__process_build_id;
+
+	trace->tool.ordered_events = true;
+	trace->tool.ordering_requires_timestamps = true;
+
+	/* add tid to output */
+	trace->multiple_threads = true;
+
+	session = perf_session__new(&file, false, &trace->tool);
+	if (session == NULL)
+		return -1;
+
+	if (symbol__init(&session->header.env) < 0)
+		goto out;
+
+	trace->host = &session->machines.host;
+
+	err = perf_session__set_tracepoints_handlers(session, handlers);
+	if (err)
+		goto out;
+
+	evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
+						     "raw_syscalls:sys_enter");
+	/* older kernels have syscalls tp versus raw_syscalls */
+	if (evsel == NULL)
+		evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
+							     "syscalls:sys_enter");
+
+	if (evsel &&
+	    (perf_evsel__init_syscall_tp(evsel, trace__sys_enter) < 0 ||
+	    perf_evsel__init_sc_tp_ptr_field(evsel, args))) {
+		pr_err("Error during initialize raw_syscalls:sys_enter event\n");
+		goto out;
+	}
+
+	evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
+						     "raw_syscalls:sys_exit");
+	if (evsel == NULL)
+		evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
+							     "syscalls:sys_exit");
+	if (evsel &&
+	    (perf_evsel__init_syscall_tp(evsel, trace__sys_exit) < 0 ||
+	    perf_evsel__init_sc_tp_uint_field(evsel, ret))) {
+		pr_err("Error during initialize raw_syscalls:sys_exit event\n");
+		goto out;
+	}
+
+	evlist__for_each(session->evlist, evsel) {
+		if (evsel->attr.type == PERF_TYPE_SOFTWARE &&
+		    (evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS_MAJ ||
+		     evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS_MIN ||
+		     evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS))
+			evsel->handler = trace__pgfault;
+	}
+
+	err = parse_target_str(trace);
+	if (err != 0)
+		goto out;
+
+	setup_pager();
+
+	err = perf_session__process_events(session);
+	if (err)
+		pr_err("Failed to process events, error %d", err);
+
+	else if (trace->summary)
+		trace__fprintf_thread_summary(trace, trace->output);
+
+out:
+	perf_session__delete(session);
+
+	return err;
+}
+
+static size_t trace__fprintf_threads_header(FILE *fp)
+{
+	size_t printed;
+
+	printed  = fprintf(fp, "\n Summary of events:\n\n");
+
+	return printed;
+}
+
+static size_t thread__dump_stats(struct thread_trace *ttrace,
+				 struct trace *trace, FILE *fp)
+{
+	struct stats *stats;
+	size_t printed = 0;
+	struct syscall *sc;
+	struct int_node *inode = intlist__first(ttrace->syscall_stats);
+
+	if (inode == NULL)
+		return 0;
+
+	printed += fprintf(fp, "\n");
+
+	printed += fprintf(fp, "   syscall            calls      min       avg       max      stddev\n");
+	printed += fprintf(fp, "                               (msec)    (msec)    (msec)        (%%)\n");
+	printed += fprintf(fp, "   --------------- -------- --------- --------- ---------     ------\n");
+
+	/* each int_node is a syscall */
+	while (inode) {
+		stats = inode->priv;
+		if (stats) {
+			double min = (double)(stats->min) / NSEC_PER_MSEC;
+			double max = (double)(stats->max) / NSEC_PER_MSEC;
+			double avg = avg_stats(stats);
+			double pct;
+			u64 n = (u64) stats->n;
+
+			pct = avg ? 100.0 * stddev_stats(stats)/avg : 0.0;
+			avg /= NSEC_PER_MSEC;
+
+			sc = &trace->syscalls.table[inode->i];
+			printed += fprintf(fp, "   %-15s", sc->name);
+			printed += fprintf(fp, " %8" PRIu64 " %9.3f %9.3f",
+					   n, min, avg);
+			printed += fprintf(fp, " %9.3f %9.2f%%\n", max, pct);
+		}
+
+		inode = intlist__next(inode);
+	}
+
+	printed += fprintf(fp, "\n\n");
+
+	return printed;
+}
+
+/* struct used to pass data to per-thread function */
+struct summary_data {
+	FILE *fp;
+	struct trace *trace;
+	size_t printed;
+};
+
+static int trace__fprintf_one_thread(struct thread *thread, void *priv)
+{
+	struct summary_data *data = priv;
+	FILE *fp = data->fp;
+	size_t printed = data->printed;
+	struct trace *trace = data->trace;
+	struct thread_trace *ttrace = thread__priv(thread);
+	double ratio;
+
+	if (ttrace == NULL)
+		return 0;
+
+	ratio = (double)ttrace->nr_events / trace->nr_events * 100.0;
+
+	printed += fprintf(fp, " %s (%d), ", thread__comm_str(thread), thread->tid);
+	printed += fprintf(fp, "%lu events, ", ttrace->nr_events);
+	printed += fprintf(fp, "%.1f%%", ratio);
+	if (ttrace->pfmaj)
+		printed += fprintf(fp, ", %lu majfaults", ttrace->pfmaj);
+	if (ttrace->pfmin)
+		printed += fprintf(fp, ", %lu minfaults", ttrace->pfmin);
+	printed += fprintf(fp, ", %.3f msec\n", ttrace->runtime_ms);
+	printed += thread__dump_stats(ttrace, trace, fp);
+
+	data->printed += printed;
+
+	return 0;
+}
+
+static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp)
+{
+	struct summary_data data = {
+		.fp = fp,
+		.trace = trace
+	};
+	data.printed = trace__fprintf_threads_header(fp);
+
+	machine__for_each_thread(trace->host, trace__fprintf_one_thread, &data);
+
+	return data.printed;
+}
+
+static int trace__set_duration(const struct option *opt, const char *str,
+			       int unset __maybe_unused)
+{
+	struct trace *trace = opt->value;
+
+	trace->duration_filter = atof(str);
+	return 0;
+}
+
+static int trace__set_filter_pids(const struct option *opt, const char *str,
+				  int unset __maybe_unused)
+{
+	int ret = -1;
+	size_t i;
+	struct trace *trace = opt->value;
+	/*
+	 * FIXME: introduce a intarray class, plain parse csv and create a
+	 * { int nr, int entries[] } struct...
+	 */
+	struct intlist *list = intlist__new(str);
+
+	if (list == NULL)
+		return -1;
+
+	i = trace->filter_pids.nr = intlist__nr_entries(list) + 1;
+	trace->filter_pids.entries = calloc(i, sizeof(pid_t));
+
+	if (trace->filter_pids.entries == NULL)
+		goto out;
+
+	trace->filter_pids.entries[0] = getpid();
+
+	for (i = 1; i < trace->filter_pids.nr; ++i)
+		trace->filter_pids.entries[i] = intlist__entry(list, i - 1)->i;
+
+	intlist__delete(list);
+	ret = 0;
+out:
+	return ret;
+}
+
+static int trace__open_output(struct trace *trace, const char *filename)
+{
+	struct stat st;
+
+	if (!stat(filename, &st) && st.st_size) {
+		char oldname[PATH_MAX];
+
+		scnprintf(oldname, sizeof(oldname), "%s.old", filename);
+		unlink(oldname);
+		rename(filename, oldname);
+	}
+
+	trace->output = fopen(filename, "w");
+
+	return trace->output == NULL ? -errno : 0;
+}
+
+static int parse_pagefaults(const struct option *opt, const char *str,
+			    int unset __maybe_unused)
+{
+	int *trace_pgfaults = opt->value;
+
+	if (strcmp(str, "all") == 0)
+		*trace_pgfaults |= TRACE_PFMAJ | TRACE_PFMIN;
+	else if (strcmp(str, "maj") == 0)
+		*trace_pgfaults |= TRACE_PFMAJ;
+	else if (strcmp(str, "min") == 0)
+		*trace_pgfaults |= TRACE_PFMIN;
+	else
+		return -1;
+
+	return 0;
+}
+
+static void evlist__set_evsel_handler(struct perf_evlist *evlist, void *handler)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel)
+		evsel->handler = handler;
+}
+
+int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	const char *trace_usage[] = {
+		"perf trace [<options>] [<command>]",
+		"perf trace [<options>] -- <command> [<options>]",
+		"perf trace record [<options>] [<command>]",
+		"perf trace record [<options>] -- <command> [<options>]",
+		NULL
+	};
+	struct trace trace = {
+		.audit = {
+			.machine = audit_detect_machine(),
+			.open_id = audit_name_to_syscall("open", trace.audit.machine),
+		},
+		.syscalls = {
+			. max = -1,
+		},
+		.opts = {
+			.target = {
+				.uid	   = UINT_MAX,
+				.uses_mmap = true,
+			},
+			.user_freq     = UINT_MAX,
+			.user_interval = ULLONG_MAX,
+			.no_buffering  = true,
+			.mmap_pages    = UINT_MAX,
+		},
+		.output = stdout,
+		.show_comm = true,
+		.trace_syscalls = true,
+	};
+	const char *output_name = NULL;
+	const char *ev_qualifier_str = NULL;
+	const struct option trace_options[] = {
+	OPT_CALLBACK(0, "event", &trace.evlist, "event",
+		     "event selector. use 'perf list' to list available events",
+		     parse_events_option),
+	OPT_BOOLEAN(0, "comm", &trace.show_comm,
+		    "show the thread COMM next to its id"),
+	OPT_BOOLEAN(0, "tool_stats", &trace.show_tool_stats, "show tool stats"),
+	OPT_STRING('e', "expr", &ev_qualifier_str, "expr",
+		    "list of events to trace"),
+	OPT_STRING('o', "output", &output_name, "file", "output file name"),
+	OPT_STRING('i', "input", &input_name, "file", "Analyze events in file"),
+	OPT_STRING('p', "pid", &trace.opts.target.pid, "pid",
+		    "trace events on existing process id"),
+	OPT_STRING('t', "tid", &trace.opts.target.tid, "tid",
+		    "trace events on existing thread id"),
+	OPT_CALLBACK(0, "filter-pids", &trace, "float",
+		     "show only events with duration > N.M ms", trace__set_filter_pids),
+	OPT_BOOLEAN('a', "all-cpus", &trace.opts.target.system_wide,
+		    "system-wide collection from all CPUs"),
+	OPT_STRING('C', "cpu", &trace.opts.target.cpu_list, "cpu",
+		    "list of cpus to monitor"),
+	OPT_BOOLEAN(0, "no-inherit", &trace.opts.no_inherit,
+		    "child tasks do not inherit counters"),
+	OPT_CALLBACK('m', "mmap-pages", &trace.opts.mmap_pages, "pages",
+		     "number of mmap data pages",
+		     perf_evlist__parse_mmap_pages),
+	OPT_STRING('u', "uid", &trace.opts.target.uid_str, "user",
+		   "user to profile"),
+	OPT_CALLBACK(0, "duration", &trace, "float",
+		     "show only events with duration > N.M ms",
+		     trace__set_duration),
+	OPT_BOOLEAN(0, "sched", &trace.sched, "show blocking scheduler events"),
+	OPT_INCR('v', "verbose", &verbose, "be more verbose"),
+	OPT_BOOLEAN('T', "time", &trace.full_time,
+		    "Show full timestamp, not time relative to first start"),
+	OPT_BOOLEAN('s', "summary", &trace.summary_only,
+		    "Show only syscall summary with statistics"),
+	OPT_BOOLEAN('S', "with-summary", &trace.summary,
+		    "Show all syscalls and summary with statistics"),
+	OPT_CALLBACK_DEFAULT('F', "pf", &trace.trace_pgfaults, "all|maj|min",
+		     "Trace pagefaults", parse_pagefaults, "maj"),
+	OPT_BOOLEAN(0, "syscalls", &trace.trace_syscalls, "Trace syscalls"),
+	OPT_BOOLEAN('f', "force", &trace.force, "don't complain, do it"),
+	OPT_END()
+	};
+	const char * const trace_subcommands[] = { "record", NULL };
+	int err;
+	char bf[BUFSIZ];
+
+	signal(SIGSEGV, sighandler_dump_stack);
+	signal(SIGFPE, sighandler_dump_stack);
+
+	trace.evlist = perf_evlist__new();
+	if (trace.evlist == NULL)
+		return -ENOMEM;
+
+	if (trace.evlist == NULL) {
+		pr_err("Not enough memory to run!\n");
+		goto out;
+	}
+
+	argc = parse_options_subcommand(argc, argv, trace_options, trace_subcommands,
+				 trace_usage, PARSE_OPT_STOP_AT_NON_OPTION);
+
+	if (trace.trace_pgfaults) {
+		trace.opts.sample_address = true;
+		trace.opts.sample_time = true;
+	}
+
+	if (trace.evlist->nr_entries > 0)
+		evlist__set_evsel_handler(trace.evlist, trace__event_handler);
+
+	if ((argc >= 1) && (strcmp(argv[0], "record") == 0))
+		return trace__record(&trace, argc-1, &argv[1]);
+
+	/* summary_only implies summary option, but don't overwrite summary if set */
+	if (trace.summary_only)
+		trace.summary = trace.summary_only;
+
+	if (!trace.trace_syscalls && !trace.trace_pgfaults &&
+	    trace.evlist->nr_entries == 0 /* Was --events used? */) {
+		pr_err("Please specify something to trace.\n");
+		return -1;
+	}
+
+	if (output_name != NULL) {
+		err = trace__open_output(&trace, output_name);
+		if (err < 0) {
+			perror("failed to create output file");
+			goto out;
+		}
+	}
+
+	if (ev_qualifier_str != NULL) {
+		const char *s = ev_qualifier_str;
+
+		trace.not_ev_qualifier = *s == '!';
+		if (trace.not_ev_qualifier)
+			++s;
+		trace.ev_qualifier = strlist__new(true, s);
+		if (trace.ev_qualifier == NULL) {
+			fputs("Not enough memory to parse event qualifier",
+			      trace.output);
+			err = -ENOMEM;
+			goto out_close;
+		}
+	}
+
+	err = target__validate(&trace.opts.target);
+	if (err) {
+		target__strerror(&trace.opts.target, err, bf, sizeof(bf));
+		fprintf(trace.output, "%s", bf);
+		goto out_close;
+	}
+
+	err = target__parse_uid(&trace.opts.target);
+	if (err) {
+		target__strerror(&trace.opts.target, err, bf, sizeof(bf));
+		fprintf(trace.output, "%s", bf);
+		goto out_close;
+	}
+
+	if (!argc && target__none(&trace.opts.target))
+		trace.opts.target.system_wide = true;
+
+	if (input_name)
+		err = trace__replay(&trace);
+	else
+		err = trace__run(&trace, argc, argv);
+
+out_close:
+	if (output_name != NULL)
+		fclose(trace.output);
+out:
+	return err;
+}
diff --git a/tools/perf/builtin.h b/tools/perf/builtin.h
new file mode 100644
index 000000000..3688ad290
--- /dev/null
+++ b/tools/perf/builtin.h
@@ -0,0 +1,43 @@
+#ifndef BUILTIN_H
+#define BUILTIN_H
+
+#include "util/util.h"
+#include "util/strbuf.h"
+
+extern const char perf_usage_string[];
+extern const char perf_more_info_string[];
+
+extern void list_common_cmds_help(void);
+extern const char *help_unknown_cmd(const char *cmd);
+extern void prune_packed_objects(int);
+extern int read_line_with_nul(char *buf, int size, FILE *file);
+extern int check_pager_config(const char *cmd);
+
+extern int cmd_annotate(int argc, const char **argv, const char *prefix);
+extern int cmd_bench(int argc, const char **argv, const char *prefix);
+extern int cmd_buildid_cache(int argc, const char **argv, const char *prefix);
+extern int cmd_buildid_list(int argc, const char **argv, const char *prefix);
+extern int cmd_diff(int argc, const char **argv, const char *prefix);
+extern int cmd_evlist(int argc, const char **argv, const char *prefix);
+extern int cmd_help(int argc, const char **argv, const char *prefix);
+extern int cmd_sched(int argc, const char **argv, const char *prefix);
+extern int cmd_list(int argc, const char **argv, const char *prefix);
+extern int cmd_record(int argc, const char **argv, const char *prefix);
+extern int cmd_report(int argc, const char **argv, const char *prefix);
+extern int cmd_stat(int argc, const char **argv, const char *prefix);
+extern int cmd_timechart(int argc, const char **argv, const char *prefix);
+extern int cmd_top(int argc, const char **argv, const char *prefix);
+extern int cmd_script(int argc, const char **argv, const char *prefix);
+extern int cmd_version(int argc, const char **argv, const char *prefix);
+extern int cmd_probe(int argc, const char **argv, const char *prefix);
+extern int cmd_kmem(int argc, const char **argv, const char *prefix);
+extern int cmd_lock(int argc, const char **argv, const char *prefix);
+extern int cmd_kvm(int argc, const char **argv, const char *prefix);
+extern int cmd_test(int argc, const char **argv, const char *prefix);
+extern int cmd_trace(int argc, const char **argv, const char *prefix);
+extern int cmd_inject(int argc, const char **argv, const char *prefix);
+extern int cmd_mem(int argc, const char **argv, const char *prefix);
+extern int cmd_data(int argc, const char **argv, const char *prefix);
+
+extern int find_scripts(char **scripts_array, char **scripts_path_array);
+#endif
diff --git a/tools/perf/command-list.txt b/tools/perf/command-list.txt
new file mode 100644
index 000000000..00fcaf8a5
--- /dev/null
+++ b/tools/perf/command-list.txt
@@ -0,0 +1,28 @@
+#
+# List of known perf commands.
+# command name			category [deprecated] [common]
+#
+perf-annotate			mainporcelain common
+perf-archive			mainporcelain common
+perf-bench			mainporcelain common
+perf-buildid-cache		mainporcelain common
+perf-buildid-list		mainporcelain common
+perf-data			mainporcelain common
+perf-diff			mainporcelain common
+perf-evlist			mainporcelain common
+perf-inject			mainporcelain common
+perf-kmem			mainporcelain common
+perf-kvm			mainporcelain common
+perf-list			mainporcelain common
+perf-lock			mainporcelain common
+perf-mem			mainporcelain common
+perf-probe			mainporcelain full
+perf-record			mainporcelain common
+perf-report			mainporcelain common
+perf-sched			mainporcelain common
+perf-script			mainporcelain common
+perf-stat			mainporcelain common
+perf-test			mainporcelain common
+perf-timechart			mainporcelain common
+perf-top			mainporcelain common
+perf-trace			mainporcelain common
diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
new file mode 100644
index 000000000..59a98c643
--- /dev/null
+++ b/tools/perf/config/Makefile
@@ -0,0 +1,710 @@
+
+ifeq ($(src-perf),)
+src-perf := $(srctree)/tools/perf
+endif
+
+ifeq ($(obj-perf),)
+obj-perf := $(OUTPUT)
+endif
+
+ifneq ($(obj-perf),)
+obj-perf := $(abspath $(obj-perf))/
+endif
+
+$(shell echo -n > .config-detected)
+detected     = $(shell echo "$(1)=y"       >> .config-detected)
+detected_var = $(shell echo "$(1)=$($(1))" >> .config-detected)
+
+CFLAGS := $(EXTRA_CFLAGS) $(EXTRA_WARNINGS)
+
+include $(src-perf)/config/Makefile.arch
+
+$(call detected_var,ARCH)
+
+NO_PERF_REGS := 1
+
+# Additional ARCH settings for x86
+ifeq ($(ARCH),x86)
+  $(call detected,CONFIG_X86)
+  ifeq (${IS_64_BIT}, 1)
+    CFLAGS += -DHAVE_ARCH_X86_64_SUPPORT
+    ARCH_INCLUDE = ../../arch/x86/lib/memcpy_64.S ../../arch/x86/lib/memset_64.S
+    LIBUNWIND_LIBS = -lunwind -lunwind-x86_64
+    $(call detected,CONFIG_X86_64)
+  else
+    LIBUNWIND_LIBS = -lunwind -lunwind-x86
+  endif
+  NO_PERF_REGS := 0
+endif
+
+ifeq ($(ARCH),arm)
+  NO_PERF_REGS := 0
+  LIBUNWIND_LIBS = -lunwind -lunwind-arm
+endif
+
+ifeq ($(ARCH),arm64)
+  NO_PERF_REGS := 0
+  LIBUNWIND_LIBS = -lunwind -lunwind-aarch64
+endif
+
+ifeq ($(NO_PERF_REGS),0)
+  $(call detected,CONFIG_PERF_REGS)
+endif
+
+# So far there's only x86 and arm libdw unwind support merged in perf.
+# Disable it on all other architectures in case libdw unwind
+# support is detected in system. Add supported architectures
+# to the check.
+ifneq ($(ARCH),$(filter $(ARCH),x86 arm))
+  NO_LIBDW_DWARF_UNWIND := 1
+endif
+
+ifeq ($(LIBUNWIND_LIBS),)
+  NO_LIBUNWIND := 1
+else
+  #
+  # For linking with debug library, run like:
+  #
+  #   make DEBUG=1 LIBUNWIND_DIR=/opt/libunwind/
+  #
+  ifdef LIBUNWIND_DIR
+    LIBUNWIND_CFLAGS  = -I$(LIBUNWIND_DIR)/include
+    LIBUNWIND_LDFLAGS = -L$(LIBUNWIND_DIR)/lib
+  endif
+  LIBUNWIND_LDFLAGS += $(LIBUNWIND_LIBS)
+
+  # Set per-feature check compilation flags
+  FEATURE_CHECK_CFLAGS-libunwind = $(LIBUNWIND_CFLAGS)
+  FEATURE_CHECK_LDFLAGS-libunwind = $(LIBUNWIND_LDFLAGS)
+  FEATURE_CHECK_CFLAGS-libunwind-debug-frame = $(LIBUNWIND_CFLAGS)
+  FEATURE_CHECK_LDFLAGS-libunwind-debug-frame = $(LIBUNWIND_LDFLAGS)
+endif
+
+ifeq ($(NO_PERF_REGS),0)
+  CFLAGS += -DHAVE_PERF_REGS_SUPPORT
+endif
+
+ifndef NO_LIBELF
+  # for linking with debug library, run like:
+  # make DEBUG=1 LIBDW_DIR=/opt/libdw/
+  ifdef LIBDW_DIR
+    LIBDW_CFLAGS  := -I$(LIBDW_DIR)/include
+    LIBDW_LDFLAGS := -L$(LIBDW_DIR)/lib
+  endif
+  FEATURE_CHECK_CFLAGS-libdw-dwarf-unwind := $(LIBDW_CFLAGS)
+  FEATURE_CHECK_LDFLAGS-libdw-dwarf-unwind := $(LIBDW_LDFLAGS) -ldw
+endif
+
+ifdef LIBBABELTRACE
+  # for linking with debug library, run like:
+  # make DEBUG=1 LIBBABELTRACE_DIR=/opt/libbabeltrace/
+  ifdef LIBBABELTRACE_DIR
+    LIBBABELTRACE_CFLAGS  := -I$(LIBBABELTRACE_DIR)/include
+    LIBBABELTRACE_LDFLAGS := -L$(LIBBABELTRACE_DIR)/lib
+  endif
+  FEATURE_CHECK_CFLAGS-libbabeltrace := $(LIBBABELTRACE_CFLAGS)
+  FEATURE_CHECK_LDFLAGS-libbabeltrace := $(LIBBABELTRACE_LDFLAGS) -lbabeltrace-ctf
+endif
+
+# include ARCH specific config
+-include $(src-perf)/arch/$(ARCH)/Makefile
+
+include $(src-perf)/config/utilities.mak
+
+ifeq ($(call get-executable,$(FLEX)),)
+  dummy := $(error Error: $(FLEX) is missing on this system, please install it)
+endif
+
+ifeq ($(call get-executable,$(BISON)),)
+  dummy := $(error Error: $(BISON) is missing on this system, please install it)
+endif
+
+# Treat warnings as errors unless directed not to
+ifneq ($(WERROR),0)
+  CFLAGS += -Werror
+endif
+
+ifndef DEBUG
+  DEBUG := 0
+endif
+
+ifeq ($(DEBUG),0)
+  CFLAGS += -O6
+endif
+
+ifdef PARSER_DEBUG
+  PARSER_DEBUG_BISON := -t
+  PARSER_DEBUG_FLEX  := -d
+  CFLAGS             += -DPARSER_DEBUG
+  $(call detected_var,PARSER_DEBUG_BISON)
+  $(call detected_var,PARSER_DEBUG_FLEX)
+endif
+
+ifndef NO_LIBPYTHON
+  # Try different combinations to accommodate systems that only have
+  # python[2][-config] in weird combinations but always preferring
+  # python2 and python2-config as per pep-0394. If we catch a
+  # python[-config] in version 3, the version check will kill it.
+  PYTHON2 := $(if $(call get-executable,python2),python2,python)
+  override PYTHON := $(call get-executable-or-default,PYTHON,$(PYTHON2))
+  PYTHON2_CONFIG := \
+    $(if $(call get-executable,$(PYTHON)-config),$(PYTHON)-config,python-config)
+  override PYTHON_CONFIG := \
+    $(call get-executable-or-default,PYTHON_CONFIG,$(PYTHON2_CONFIG))
+
+  PYTHON_CONFIG_SQ := $(call shell-sq,$(PYTHON_CONFIG))
+
+  PYTHON_EMBED_LDOPTS := $(shell $(PYTHON_CONFIG_SQ) --ldflags 2>/dev/null)
+  PYTHON_EMBED_CCOPTS := $(shell $(PYTHON_CONFIG_SQ) --cflags 2>/dev/null)
+
+  FEATURE_CHECK_CFLAGS-libpython := $(PYTHON_EMBED_CCOPTS)
+  FEATURE_CHECK_LDFLAGS-libpython := $(PYTHON_EMBED_LDOPTS)
+  FEATURE_CHECK_CFLAGS-libpython-version := $(PYTHON_EMBED_CCOPTS)
+  FEATURE_CHECK_LDFLAGS-libpython-version := $(PYTHON_EMBED_LDOPTS)
+endif
+
+CFLAGS += -fno-omit-frame-pointer
+CFLAGS += -ggdb3
+CFLAGS += -funwind-tables
+CFLAGS += -Wall
+CFLAGS += -Wextra
+CFLAGS += -std=gnu99
+
+# Enforce a non-executable stack, as we may regress (again) in the future by
+# adding assembler files missing the .GNU-stack linker note.
+LDFLAGS += -Wl,-z,noexecstack
+
+EXTLIBS = -lpthread -lrt -lm -ldl
+
+include $(srctree)/tools/build/Makefile.feature
+
+ifeq ($(feature-stackprotector-all), 1)
+  CFLAGS += -fstack-protector-all
+endif
+
+ifeq ($(DEBUG),0)
+  ifeq ($(feature-fortify-source), 1)
+    CFLAGS += -D_FORTIFY_SOURCE=2
+  endif
+endif
+
+CFLAGS += -I$(src-perf)/util/include
+CFLAGS += -I$(src-perf)/arch/$(ARCH)/include
+CFLAGS += -I$(srctree)/tools/include/
+CFLAGS += -I$(srctree)/arch/$(ARCH)/include/uapi
+CFLAGS += -I$(srctree)/arch/$(ARCH)/include
+CFLAGS += -I$(srctree)/include/uapi
+CFLAGS += -I$(srctree)/include
+
+# $(obj-perf)      for generated common-cmds.h
+# $(obj-perf)/util for generated bison/flex headers
+ifneq ($(OUTPUT),)
+CFLAGS += -I$(obj-perf)/util
+CFLAGS += -I$(obj-perf)
+endif
+
+CFLAGS += -I$(src-perf)/util
+CFLAGS += -I$(src-perf)
+CFLAGS += -I$(srctree)/tools/lib/
+
+CFLAGS += -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE
+
+ifeq ($(feature-sync-compare-and-swap), 1)
+  CFLAGS += -DHAVE_SYNC_COMPARE_AND_SWAP_SUPPORT
+endif
+
+ifeq ($(feature-pthread-attr-setaffinity-np), 1)
+  CFLAGS += -DHAVE_PTHREAD_ATTR_SETAFFINITY_NP
+endif
+
+ifndef NO_BIONIC
+  $(call feature_check,bionic)
+  ifeq ($(feature-bionic), 1)
+    BIONIC := 1
+    EXTLIBS := $(filter-out -lrt,$(EXTLIBS))
+    EXTLIBS := $(filter-out -lpthread,$(EXTLIBS))
+  endif
+endif
+
+ifdef NO_LIBELF
+  NO_DWARF := 1
+  NO_DEMANGLE := 1
+  NO_LIBUNWIND := 1
+  NO_LIBDW_DWARF_UNWIND := 1
+else
+  ifeq ($(feature-libelf), 0)
+    ifeq ($(feature-glibc), 1)
+      LIBC_SUPPORT := 1
+    endif
+    ifeq ($(BIONIC),1)
+      LIBC_SUPPORT := 1
+    endif
+    ifeq ($(LIBC_SUPPORT),1)
+      msg := $(warning No libelf found, disables 'probe' tool, please install elfutils-libelf-devel/libelf-dev);
+
+      NO_LIBELF := 1
+      NO_DWARF := 1
+      NO_DEMANGLE := 1
+      NO_LIBUNWIND := 1
+      NO_LIBDW_DWARF_UNWIND := 1
+    else
+      ifneq ($(filter s% -static%,$(LDFLAGS),),)
+        msg := $(error No static glibc found, please install glibc-static);
+      else
+        msg := $(error No gnu/libc-version.h found, please install glibc-dev[el]);
+      endif
+    endif
+  else
+    ifndef NO_LIBDW_DWARF_UNWIND
+      ifneq ($(feature-libdw-dwarf-unwind),1)
+        NO_LIBDW_DWARF_UNWIND := 1
+        msg := $(warning No libdw DWARF unwind found, Please install elfutils-devel/libdw-dev >= 0.158 and/or set LIBDW_DIR);
+      endif
+    endif
+    ifneq ($(feature-dwarf), 1)
+      msg := $(warning No libdw.h found or old libdw.h found or elfutils is older than 0.138, disables dwarf support. Please install new elfutils-devel/libdw-dev);
+      NO_DWARF := 1
+    endif # Dwarf support
+  endif # libelf support
+endif # NO_LIBELF
+
+ifndef NO_LIBELF
+  CFLAGS += -DHAVE_LIBELF_SUPPORT
+  EXTLIBS += -lelf
+  $(call detected,CONFIG_LIBELF)
+
+  ifeq ($(feature-libelf-mmap), 1)
+    CFLAGS += -DHAVE_LIBELF_MMAP_SUPPORT
+  endif
+
+  ifeq ($(feature-libelf-getphdrnum), 1)
+    CFLAGS += -DHAVE_ELF_GETPHDRNUM_SUPPORT
+  endif
+
+  # include ARCH specific config
+  -include $(src-perf)/arch/$(ARCH)/Makefile
+
+  ifndef NO_DWARF
+    ifeq ($(origin PERF_HAVE_DWARF_REGS), undefined)
+      msg := $(warning DWARF register mappings have not been defined for architecture $(ARCH), DWARF support disabled);
+      NO_DWARF := 1
+    else
+      CFLAGS += -DHAVE_DWARF_SUPPORT $(LIBDW_CFLAGS)
+      LDFLAGS += $(LIBDW_LDFLAGS)
+      EXTLIBS += -ldw
+      $(call detected,CONFIG_DWARF)
+    endif # PERF_HAVE_DWARF_REGS
+  endif # NO_DWARF
+endif # NO_LIBELF
+
+ifeq ($(ARCH),powerpc)
+  ifndef NO_DWARF
+    CFLAGS += -DHAVE_SKIP_CALLCHAIN_IDX
+  endif
+endif
+
+ifndef NO_LIBUNWIND
+  ifneq ($(feature-libunwind), 1)
+    msg := $(warning No libunwind found. Please install libunwind-dev[el] >= 1.1 and/or set LIBUNWIND_DIR);
+    NO_LIBUNWIND := 1
+  endif
+endif
+
+dwarf-post-unwind := 1
+dwarf-post-unwind-text := BUG
+
+# setup DWARF post unwinder
+ifdef NO_LIBUNWIND
+  ifdef NO_LIBDW_DWARF_UNWIND
+    msg := $(warning Disabling post unwind, no support found.);
+    dwarf-post-unwind := 0
+  else
+    dwarf-post-unwind-text := libdw
+    $(call detected,CONFIG_LIBDW_DWARF_UNWIND)
+  endif
+else
+  dwarf-post-unwind-text := libunwind
+  $(call detected,CONFIG_LIBUNWIND)
+  # Enable libunwind support by default.
+  ifndef NO_LIBDW_DWARF_UNWIND
+    NO_LIBDW_DWARF_UNWIND := 1
+  endif
+endif
+
+ifeq ($(dwarf-post-unwind),1)
+  CFLAGS += -DHAVE_DWARF_UNWIND_SUPPORT
+  $(call detected,CONFIG_DWARF_UNWIND)
+else
+  NO_DWARF_UNWIND := 1
+endif
+
+ifndef NO_LIBUNWIND
+  ifeq ($(ARCH),$(filter $(ARCH),arm arm64))
+    $(call feature_check,libunwind-debug-frame)
+    ifneq ($(feature-libunwind-debug-frame), 1)
+      msg := $(warning No debug_frame support found in libunwind);
+      CFLAGS += -DNO_LIBUNWIND_DEBUG_FRAME
+    endif
+  else
+    # non-ARM has no dwarf_find_debug_frame() function:
+    CFLAGS += -DNO_LIBUNWIND_DEBUG_FRAME
+  endif
+  CFLAGS  += -DHAVE_LIBUNWIND_SUPPORT
+  EXTLIBS += $(LIBUNWIND_LIBS)
+  CFLAGS  += $(LIBUNWIND_CFLAGS)
+  LDFLAGS += $(LIBUNWIND_LDFLAGS)
+endif
+
+ifndef NO_LIBAUDIT
+  ifneq ($(feature-libaudit), 1)
+    msg := $(warning No libaudit.h found, disables 'trace' tool, please install audit-libs-devel or libaudit-dev);
+    NO_LIBAUDIT := 1
+  else
+    CFLAGS += -DHAVE_LIBAUDIT_SUPPORT
+    EXTLIBS += -laudit
+    $(call detected,CONFIG_AUDIT)
+  endif
+endif
+
+ifdef NO_NEWT
+  NO_SLANG=1
+endif
+
+ifndef NO_SLANG
+  ifneq ($(feature-libslang), 1)
+    msg := $(warning slang not found, disables TUI support. Please install slang-devel or libslang-dev);
+    NO_SLANG := 1
+  else
+    # Fedora has /usr/include/slang/slang.h, but ubuntu /usr/include/slang.h
+    CFLAGS += -I/usr/include/slang
+    CFLAGS += -DHAVE_SLANG_SUPPORT
+    EXTLIBS += -lslang
+    $(call detected,CONFIG_SLANG)
+  endif
+endif
+
+ifndef NO_GTK2
+  FLAGS_GTK2=$(CFLAGS) $(LDFLAGS) $(EXTLIBS) $(shell $(PKG_CONFIG) --libs --cflags gtk+-2.0 2>/dev/null)
+  ifneq ($(feature-gtk2), 1)
+    msg := $(warning GTK2 not found, disables GTK2 support. Please install gtk2-devel or libgtk2.0-dev);
+    NO_GTK2 := 1
+  else
+    ifeq ($(feature-gtk2-infobar), 1)
+      GTK_CFLAGS := -DHAVE_GTK_INFO_BAR_SUPPORT
+    endif
+    CFLAGS += -DHAVE_GTK2_SUPPORT
+    GTK_CFLAGS += $(shell $(PKG_CONFIG) --cflags gtk+-2.0 2>/dev/null)
+    GTK_LIBS := $(shell $(PKG_CONFIG) --libs gtk+-2.0 2>/dev/null)
+    EXTLIBS += -ldl
+  endif
+endif
+
+grep-libs  = $(filter -l%,$(1))
+strip-libs = $(filter-out -l%,$(1))
+
+ifdef NO_LIBPERL
+  CFLAGS += -DNO_LIBPERL
+else
+  PERL_EMBED_LDOPTS = $(shell perl -MExtUtils::Embed -e ldopts 2>/dev/null)
+  PERL_EMBED_LDFLAGS = $(call strip-libs,$(PERL_EMBED_LDOPTS))
+  PERL_EMBED_LIBADD = $(call grep-libs,$(PERL_EMBED_LDOPTS))
+  PERL_EMBED_CCOPTS = `perl -MExtUtils::Embed -e ccopts 2>/dev/null`
+  FLAGS_PERL_EMBED=$(PERL_EMBED_CCOPTS) $(PERL_EMBED_LDOPTS)
+
+  ifneq ($(feature-libperl), 1)
+    CFLAGS += -DNO_LIBPERL
+    NO_LIBPERL := 1
+    msg := $(warning Missing perl devel files. Disabling perl scripting support, please install perl-ExtUtils-Embed/libperl-dev);
+  else
+    LDFLAGS += $(PERL_EMBED_LDFLAGS)
+    EXTLIBS += $(PERL_EMBED_LIBADD)
+    $(call detected,CONFIG_LIBPERL)
+  endif
+endif
+
+ifeq ($(feature-timerfd), 1)
+  CFLAGS += -DHAVE_TIMERFD_SUPPORT
+else
+  msg := $(warning No timerfd support. Disables 'perf kvm stat live');
+endif
+
+disable-python = $(eval $(disable-python_code))
+define disable-python_code
+  CFLAGS += -DNO_LIBPYTHON
+  $(warning $1)
+  NO_LIBPYTHON := 1
+endef
+
+ifdef NO_LIBPYTHON
+  $(call disable-python,Python support disabled by user)
+else
+
+  ifndef PYTHON
+    $(call disable-python,No python interpreter was found: disables Python support - please install python-devel/python-dev)
+  else
+    PYTHON_WORD := $(call shell-wordify,$(PYTHON))
+
+    ifndef PYTHON_CONFIG
+      $(call disable-python,No 'python-config' tool was found: disables Python support - please install python-devel/python-dev)
+    else
+
+      PYTHON_CONFIG_SQ := $(call shell-sq,$(PYTHON_CONFIG))
+
+      PYTHON_EMBED_LDOPTS := $(shell $(PYTHON_CONFIG_SQ) --ldflags 2>/dev/null)
+      PYTHON_EMBED_LDFLAGS := $(call strip-libs,$(PYTHON_EMBED_LDOPTS))
+      PYTHON_EMBED_LIBADD := $(call grep-libs,$(PYTHON_EMBED_LDOPTS))
+      PYTHON_EMBED_CCOPTS := $(shell $(PYTHON_CONFIG_SQ) --cflags 2>/dev/null)
+      FLAGS_PYTHON_EMBED := $(PYTHON_EMBED_CCOPTS) $(PYTHON_EMBED_LDOPTS)
+
+      ifneq ($(feature-libpython), 1)
+        $(call disable-python,No 'Python.h' (for Python 2.x support) was found: disables Python support - please install python-devel/python-dev)
+      else
+
+        ifneq ($(feature-libpython-version), 1)
+          $(warning Python 3 is not yet supported; please set)
+          $(warning PYTHON and/or PYTHON_CONFIG appropriately.)
+          $(warning If you also have Python 2 installed, then)
+          $(warning try something like:)
+          $(warning $(and ,))
+          $(warning $(and ,)  make PYTHON=python2)
+          $(warning $(and ,))
+          $(warning Otherwise, disable Python support entirely:)
+          $(warning $(and ,))
+          $(warning $(and ,)  make NO_LIBPYTHON=1)
+          $(warning $(and ,))
+          $(error   $(and ,))
+        else
+          LDFLAGS += $(PYTHON_EMBED_LDFLAGS)
+          EXTLIBS += $(PYTHON_EMBED_LIBADD)
+          LANG_BINDINGS += $(obj-perf)python/perf.so
+          $(call detected,CONFIG_LIBPYTHON)
+        endif
+      endif
+    endif
+  endif
+endif
+
+ifeq ($(feature-libbfd), 1)
+  EXTLIBS += -lbfd
+
+  # call all detections now so we get correct
+  # status in VF output
+  $(call feature_check,liberty)
+  $(call feature_check,liberty-z)
+  $(call feature_check,cplus-demangle)
+
+  ifeq ($(feature-liberty), 1)
+    EXTLIBS += -liberty
+  else
+    ifeq ($(feature-liberty-z), 1)
+      EXTLIBS += -liberty -lz
+    endif
+  endif
+endif
+
+ifdef NO_DEMANGLE
+  CFLAGS += -DNO_DEMANGLE
+else
+  ifdef HAVE_CPLUS_DEMANGLE_SUPPORT
+    EXTLIBS += -liberty
+    CFLAGS += -DHAVE_CPLUS_DEMANGLE_SUPPORT
+  else
+    ifneq ($(feature-libbfd), 1)
+      ifneq ($(feature-liberty), 1)
+        ifneq ($(feature-liberty-z), 1)
+          # we dont have neither HAVE_CPLUS_DEMANGLE_SUPPORT
+          # or any of 'bfd iberty z' trinity
+          ifeq ($(feature-cplus-demangle), 1)
+            EXTLIBS += -liberty
+            CFLAGS += -DHAVE_CPLUS_DEMANGLE_SUPPORT
+          else
+            msg := $(warning No bfd.h/libbfd found, please install binutils-dev[el]/zlib-static/libiberty-dev to gain symbol demangling)
+            CFLAGS += -DNO_DEMANGLE
+          endif
+        endif
+      endif
+    endif
+  endif
+endif
+
+ifneq ($(filter -lbfd,$(EXTLIBS)),)
+  CFLAGS += -DHAVE_LIBBFD_SUPPORT
+endif
+
+ifndef NO_ZLIB
+  ifeq ($(feature-zlib), 1)
+    CFLAGS += -DHAVE_ZLIB_SUPPORT
+    EXTLIBS += -lz
+    $(call detected,CONFIG_ZLIB)
+  else
+    NO_ZLIB := 1
+  endif
+endif
+
+ifndef NO_LZMA
+  ifeq ($(feature-lzma), 1)
+    CFLAGS += -DHAVE_LZMA_SUPPORT
+    EXTLIBS += -llzma
+    $(call detected,CONFIG_LZMA)
+  else
+    msg := $(warning No liblzma found, disables xz kernel module decompression, please install xz-devel/liblzma-dev);
+    NO_LZMA := 1
+  endif
+endif
+
+ifndef NO_BACKTRACE
+  ifeq ($(feature-backtrace), 1)
+    CFLAGS += -DHAVE_BACKTRACE_SUPPORT
+  endif
+endif
+
+ifndef NO_LIBNUMA
+  ifeq ($(feature-libnuma), 0)
+    msg := $(warning No numa.h found, disables 'perf bench numa mem' benchmark, please install numactl-devel/libnuma-devel/libnuma-dev);
+    NO_LIBNUMA := 1
+  else
+    CFLAGS += -DHAVE_LIBNUMA_SUPPORT
+    EXTLIBS += -lnuma
+    $(call detected,CONFIG_NUMA)
+  endif
+endif
+
+ifdef HAVE_KVM_STAT_SUPPORT
+    CFLAGS += -DHAVE_KVM_STAT_SUPPORT
+endif
+
+ifeq (${IS_64_BIT}, 1)
+  ifndef NO_PERF_READ_VDSO32
+    $(call feature_check,compile-32)
+    ifeq ($(feature-compile-32), 1)
+      CFLAGS += -DHAVE_PERF_READ_VDSO32
+    else
+      NO_PERF_READ_VDSO32 := 1
+    endif
+  endif
+  ifneq ($(ARCH), x86)
+    NO_PERF_READ_VDSOX32 := 1
+  endif
+  ifndef NO_PERF_READ_VDSOX32
+    $(call feature_check,compile-x32)
+    ifeq ($(feature-compile-x32), 1)
+      CFLAGS += -DHAVE_PERF_READ_VDSOX32
+    else
+      NO_PERF_READ_VDSOX32 := 1
+    endif
+  endif
+else
+  NO_PERF_READ_VDSO32 := 1
+  NO_PERF_READ_VDSOX32 := 1
+endif
+
+ifdef LIBBABELTRACE
+  $(call feature_check,libbabeltrace)
+  ifeq ($(feature-libbabeltrace), 1)
+    CFLAGS += -DHAVE_LIBBABELTRACE_SUPPORT $(LIBBABELTRACE_CFLAGS)
+    LDFLAGS += $(LIBBABELTRACE_LDFLAGS)
+    EXTLIBS += -lbabeltrace-ctf
+    $(call detected,CONFIG_LIBBABELTRACE)
+  else
+    msg := $(warning No libbabeltrace found, disables 'perf data' CTF format support, please install libbabeltrace-dev[el]/libbabeltrace-ctf-dev);
+  endif
+endif
+
+# Among the variables below, these:
+#   perfexecdir
+#   template_dir
+#   mandir
+#   infodir
+#   htmldir
+#   ETC_PERFCONFIG (but not sysconfdir)
+# can be specified as a relative path some/where/else;
+# this is interpreted as relative to $(prefix) and "perf" at
+# runtime figures out where they are based on the path to the executable.
+# This can help installing the suite in a relocatable way.
+
+# Make the path relative to DESTDIR, not to prefix
+ifndef DESTDIR
+prefix ?= $(HOME)
+endif
+bindir_relative = bin
+bindir = $(prefix)/$(bindir_relative)
+mandir = share/man
+infodir = share/info
+perfexecdir = libexec/perf-core
+sharedir = $(prefix)/share
+template_dir = share/perf-core/templates
+htmldir = share/doc/perf-doc
+ifeq ($(prefix),/usr)
+sysconfdir = /etc
+ETC_PERFCONFIG = $(sysconfdir)/perfconfig
+else
+sysconfdir = $(prefix)/etc
+ETC_PERFCONFIG = etc/perfconfig
+endif
+ifndef lib
+ifeq ($(ARCH)$(IS_64_BIT), x861)
+lib = lib64
+else
+lib = lib
+endif
+endif # lib
+libdir = $(prefix)/$(lib)
+
+# Shell quote (do not use $(call) to accommodate ancient setups);
+ETC_PERFCONFIG_SQ = $(subst ','\'',$(ETC_PERFCONFIG))
+DESTDIR_SQ = $(subst ','\'',$(DESTDIR))
+bindir_SQ = $(subst ','\'',$(bindir))
+mandir_SQ = $(subst ','\'',$(mandir))
+infodir_SQ = $(subst ','\'',$(infodir))
+perfexecdir_SQ = $(subst ','\'',$(perfexecdir))
+template_dir_SQ = $(subst ','\'',$(template_dir))
+htmldir_SQ = $(subst ','\'',$(htmldir))
+prefix_SQ = $(subst ','\'',$(prefix))
+sysconfdir_SQ = $(subst ','\'',$(sysconfdir))
+libdir_SQ = $(subst ','\'',$(libdir))
+
+ifneq ($(filter /%,$(firstword $(perfexecdir))),)
+perfexec_instdir = $(perfexecdir)
+else
+perfexec_instdir = $(prefix)/$(perfexecdir)
+endif
+perfexec_instdir_SQ = $(subst ','\'',$(perfexec_instdir))
+
+# If we install to $(HOME) we keep the traceevent default:
+# $(HOME)/.traceevent/plugins
+# Otherwise we install plugins into the global $(libdir).
+ifdef DESTDIR
+plugindir=$(libdir)/traceevent/plugins
+plugindir_SQ= $(subst ','\'',$(plugindir))
+endif
+
+print_var = $(eval $(print_var_code)) $(info $(MSG))
+define print_var_code
+    MSG = $(shell printf '...%30s: %s' $(1) $($(1)))
+endef
+
+ifeq ($(VF),1)
+  $(call print_var,prefix)
+  $(call print_var,bindir)
+  $(call print_var,libdir)
+  $(call print_var,sysconfdir)
+  $(call print_var,LIBUNWIND_DIR)
+  $(call print_var,LIBDW_DIR)
+  $(info )
+endif
+
+$(call detected_var,bindir_SQ)
+$(call detected_var,PYTHON_WORD)
+ifneq ($(OUTPUT),)
+$(call detected_var,OUTPUT)
+endif
+$(call detected_var,htmldir_SQ)
+$(call detected_var,infodir_SQ)
+$(call detected_var,mandir_SQ)
+$(call detected_var,ETC_PERFCONFIG_SQ)
+$(call detected_var,prefix_SQ)
+$(call detected_var,perfexecdir_SQ)
+$(call detected_var,LIBDIR)
+$(call detected_var,GTK_CFLAGS)
+$(call detected_var,PERL_EMBED_CCOPTS)
+$(call detected_var,PYTHON_EMBED_CCOPTS)
diff --git a/tools/perf/config/Makefile.arch b/tools/perf/config/Makefile.arch
new file mode 100644
index 000000000..e11fbd6fa
--- /dev/null
+++ b/tools/perf/config/Makefile.arch
@@ -0,0 +1,18 @@
+ifndef ARCH
+ARCH := $(shell uname -m 2>/dev/null || echo not)
+endif
+
+ARCH := $(shell echo $(ARCH) | sed -e s/i.86/x86/ -e s/x86_64/x86/ \
+                                  -e s/sun4u/sparc/ -e s/sparc64/sparc/ \
+                                  -e /arm64/!s/arm.*/arm/ -e s/sa110/arm/ \
+                                  -e s/s390x/s390/ -e s/parisc64/parisc/ \
+                                  -e s/ppc.*/powerpc/ -e s/mips.*/mips/ \
+                                  -e s/sh[234].*/sh/ -e s/aarch64.*/arm64/ \
+                                  -e s/tile.*/tile/ )
+
+LP64 := $(shell echo __LP64__ | ${CC} ${CFLAGS} -E -x c - | tail -n 1)
+ifeq ($(LP64), 1)
+  IS_64_BIT := 1
+else
+  IS_64_BIT := 0
+endif
diff --git a/tools/perf/config/utilities.mak b/tools/perf/config/utilities.mak
new file mode 100644
index 000000000..c16ce8330
--- /dev/null
+++ b/tools/perf/config/utilities.mak
@@ -0,0 +1,179 @@
+# This allows us to work with the newline character:
+define newline
+
+
+endef
+newline := $(newline)
+
+# nl-escape
+#
+# Usage: escape = $(call nl-escape[,escape])
+#
+# This is used as the common way to specify
+# what should replace a newline when escaping
+# newlines; the default is a bizarre string.
+#
+nl-escape = $(if $(1),$(1),m822df3020w6a44id34bt574ctac44eb9f4n)
+
+# escape-nl
+#
+# Usage: escaped-text = $(call escape-nl,text[,escape])
+#
+# GNU make's $(shell ...) function converts to a
+# single space each newline character in the output
+# produced during the expansion; this may not be
+# desirable.
+#
+# The only solution is to change each newline into
+# something that won't be converted, so that the
+# information can be recovered later with
+# $(call unescape-nl...)
+#
+escape-nl = $(subst $(newline),$(call nl-escape,$(2)),$(1))
+
+# unescape-nl
+#
+# Usage: text = $(call unescape-nl,escaped-text[,escape])
+#
+# See escape-nl.
+#
+unescape-nl = $(subst $(call nl-escape,$(2)),$(newline),$(1))
+
+# shell-escape-nl
+#
+# Usage: $(shell some-command | $(call shell-escape-nl[,escape]))
+#
+# Use this to escape newlines from within a shell call;
+# the default escape is a bizarre string.
+#
+# NOTE: The escape is used directly as a string constant
+#       in an `awk' program that is delimited by shell
+#       single-quotes, so be wary of the characters
+#       that are chosen.
+#
+define shell-escape-nl
+awk 'NR==1 {t=$$0} NR>1 {t=t "$(nl-escape)" $$0} END {printf t}'
+endef
+
+# shell-unescape-nl
+#
+# Usage: $(shell some-command | $(call shell-unescape-nl[,escape]))
+#
+# Use this to unescape newlines from within a shell call;
+# the default escape is a bizarre string.
+#
+# NOTE: The escape is used directly as an extended regular
+#       expression constant in an `awk' program that is
+#       delimited by shell single-quotes, so be wary
+#       of the characters that are chosen.
+#
+# (The bash shell has a bug where `{gsub(...),...}' is
+#  misinterpreted as a brace expansion; this can be
+#  overcome by putting a space between `{' and `gsub').
+#
+define shell-unescape-nl
+awk 'NR==1 {t=$$0} NR>1 {t=t "\n" $$0} END { gsub(/$(nl-escape)/,"\n",t); printf t }'
+endef
+
+# escape-for-shell-sq
+#
+# Usage: embeddable-text = $(call escape-for-shell-sq,text)
+#
+# This function produces text that is suitable for
+# embedding in a shell string that is delimited by
+# single-quotes.
+#
+escape-for-shell-sq =  $(subst ','\'',$(1))
+
+# shell-sq
+#
+# Usage: single-quoted-and-escaped-text = $(call shell-sq,text)
+#
+shell-sq = '$(escape-for-shell-sq)'
+
+# shell-wordify
+#
+# Usage: wordified-text = $(call shell-wordify,text)
+#
+# For instance:
+#
+#  |define text
+#  |hello
+#  |world
+#  |endef
+#  |
+#  |target:
+#  |	echo $(call shell-wordify,$(text))
+#
+# At least GNU make gets confused by expanding a newline
+# within the context of a command line of a makefile rule
+# (this is in constrast to a `$(shell ...)' function call,
+# which can handle it just fine).
+#
+# This function avoids the problem by producing a string
+# that works as a shell word, regardless of whether or
+# not it contains a newline.
+#
+# If the text to be wordified contains a newline, then
+# an intrictate shell command substitution is constructed
+# to render the text as a single line; when the shell
+# processes the resulting escaped text, it transforms
+# it into the original unescaped text.
+#
+# If the text does not contain a newline, then this function
+# produces the same results as the `$(shell-sq)' function.
+#
+shell-wordify = $(if $(findstring $(newline),$(1)),$(_sw-esc-nl),$(shell-sq))
+define _sw-esc-nl
+"$$(echo $(call escape-nl,$(shell-sq),$(2)) | $(call shell-unescape-nl,$(2)))"
+endef
+
+# is-absolute
+#
+# Usage: bool-value = $(call is-absolute,path)
+#
+is-absolute = $(shell echo $(shell-sq) | grep -q ^/ && echo y)
+
+# lookup
+#
+# Usage: absolute-executable-path-or-empty = $(call lookup,path)
+#
+# (It's necessary to use `sh -c' because GNU make messes up by
+#  trying too hard and getting things wrong).
+#
+lookup = $(call unescape-nl,$(shell sh -c $(_l-sh)))
+_l-sh = $(call shell-sq,command -v $(shell-sq) | $(call shell-escape-nl,))
+
+# is-executable
+#
+# Usage: bool-value = $(call is-executable,path)
+#
+# (It's necessary to use `sh -c' because GNU make messes up by
+#  trying too hard and getting things wrong).
+#
+is-executable = $(call _is-executable-helper,$(shell-sq))
+_is-executable-helper = $(shell sh -c $(_is-executable-sh))
+_is-executable-sh = $(call shell-sq,test -f $(1) -a -x $(1) && echo y)
+
+# get-executable
+#
+# Usage: absolute-executable-path-or-empty = $(call get-executable,path)
+#
+# The goal is to get an absolute path for an executable;
+# the `command -v' is defined by POSIX, but it's not
+# necessarily very portable, so it's only used if
+# relative path resolution is requested, as determined
+# by the presence of a leading `/'.
+#
+get-executable = $(if $(1),$(if $(is-absolute),$(_ge-abspath),$(lookup)))
+_ge-abspath = $(if $(is-executable),$(1))
+
+# get-supplied-or-default-executable
+#
+# Usage: absolute-executable-path-or-empty = $(call get-executable-or-default,variable,default)
+#
+define get-executable-or-default
+$(if $($(1)),$(call _ge_attempt,$($(1)),$(1)),$(call _ge_attempt,$(2)))
+endef
+_ge_attempt = $(if $(get-executable),$(get-executable),$(call _gea_err,$(2)))
+_gea_err  = $(if $(1),$(error Please set '$(1)' appropriately))
diff --git a/tools/perf/design.txt b/tools/perf/design.txt
new file mode 100644
index 000000000..a28dca258
--- /dev/null
+++ b/tools/perf/design.txt
@@ -0,0 +1,462 @@
+
+Performance Counters for Linux
+------------------------------
+
+Performance counters are special hardware registers available on most modern
+CPUs. These registers count the number of certain types of hw events: such
+as instructions executed, cachemisses suffered, or branches mis-predicted -
+without slowing down the kernel or applications. These registers can also
+trigger interrupts when a threshold number of events have passed - and can
+thus be used to profile the code that runs on that CPU.
+
+The Linux Performance Counter subsystem provides an abstraction of these
+hardware capabilities. It provides per task and per CPU counters, counter
+groups, and it provides event capabilities on top of those.  It
+provides "virtual" 64-bit counters, regardless of the width of the
+underlying hardware counters.
+
+Performance counters are accessed via special file descriptors.
+There's one file descriptor per virtual counter used.
+
+The special file descriptor is opened via the sys_perf_event_open()
+system call:
+
+   int sys_perf_event_open(struct perf_event_attr *hw_event_uptr,
+			     pid_t pid, int cpu, int group_fd,
+			     unsigned long flags);
+
+The syscall returns the new fd. The fd can be used via the normal
+VFS system calls: read() can be used to read the counter, fcntl()
+can be used to set the blocking mode, etc.
+
+Multiple counters can be kept open at a time, and the counters
+can be poll()ed.
+
+When creating a new counter fd, 'perf_event_attr' is:
+
+struct perf_event_attr {
+        /*
+         * The MSB of the config word signifies if the rest contains cpu
+         * specific (raw) counter configuration data, if unset, the next
+         * 7 bits are an event type and the rest of the bits are the event
+         * identifier.
+         */
+        __u64                   config;
+
+        __u64                   irq_period;
+        __u32                   record_type;
+        __u32                   read_format;
+
+        __u64                   disabled       :  1, /* off by default        */
+                                inherit        :  1, /* children inherit it   */
+                                pinned         :  1, /* must always be on PMU */
+                                exclusive      :  1, /* only group on PMU     */
+                                exclude_user   :  1, /* don't count user      */
+                                exclude_kernel :  1, /* ditto kernel          */
+                                exclude_hv     :  1, /* ditto hypervisor      */
+                                exclude_idle   :  1, /* don't count when idle */
+                                mmap           :  1, /* include mmap data     */
+                                munmap         :  1, /* include munmap data   */
+                                comm           :  1, /* include comm data     */
+
+                                __reserved_1   : 52;
+
+        __u32                   extra_config_len;
+        __u32                   wakeup_events;  /* wakeup every n events */
+
+        __u64                   __reserved_2;
+        __u64                   __reserved_3;
+};
+
+The 'config' field specifies what the counter should count.  It
+is divided into 3 bit-fields:
+
+raw_type: 1 bit   (most significant bit)	0x8000_0000_0000_0000
+type:	  7 bits  (next most significant)	0x7f00_0000_0000_0000
+event_id: 56 bits (least significant)		0x00ff_ffff_ffff_ffff
+
+If 'raw_type' is 1, then the counter will count a hardware event
+specified by the remaining 63 bits of event_config.  The encoding is
+machine-specific.
+
+If 'raw_type' is 0, then the 'type' field says what kind of counter
+this is, with the following encoding:
+
+enum perf_type_id {
+	PERF_TYPE_HARDWARE		= 0,
+	PERF_TYPE_SOFTWARE		= 1,
+	PERF_TYPE_TRACEPOINT		= 2,
+};
+
+A counter of PERF_TYPE_HARDWARE will count the hardware event
+specified by 'event_id':
+
+/*
+ * Generalized performance counter event types, used by the hw_event.event_id
+ * parameter of the sys_perf_event_open() syscall:
+ */
+enum perf_hw_id {
+	/*
+	 * Common hardware events, generalized by the kernel:
+	 */
+	PERF_COUNT_HW_CPU_CYCLES		= 0,
+	PERF_COUNT_HW_INSTRUCTIONS		= 1,
+	PERF_COUNT_HW_CACHE_REFERENCES		= 2,
+	PERF_COUNT_HW_CACHE_MISSES		= 3,
+	PERF_COUNT_HW_BRANCH_INSTRUCTIONS	= 4,
+	PERF_COUNT_HW_BRANCH_MISSES		= 5,
+	PERF_COUNT_HW_BUS_CYCLES		= 6,
+};
+
+These are standardized types of events that work relatively uniformly
+on all CPUs that implement Performance Counters support under Linux,
+although there may be variations (e.g., different CPUs might count
+cache references and misses at different levels of the cache hierarchy).
+If a CPU is not able to count the selected event, then the system call
+will return -EINVAL.
+
+More hw_event_types are supported as well, but they are CPU-specific
+and accessed as raw events.  For example, to count "External bus
+cycles while bus lock signal asserted" events on Intel Core CPUs, pass
+in a 0x4064 event_id value and set hw_event.raw_type to 1.
+
+A counter of type PERF_TYPE_SOFTWARE will count one of the available
+software events, selected by 'event_id':
+
+/*
+ * Special "software" counters provided by the kernel, even if the hardware
+ * does not support performance counters. These counters measure various
+ * physical and sw events of the kernel (and allow the profiling of them as
+ * well):
+ */
+enum perf_sw_ids {
+	PERF_COUNT_SW_CPU_CLOCK		= 0,
+	PERF_COUNT_SW_TASK_CLOCK	= 1,
+	PERF_COUNT_SW_PAGE_FAULTS	= 2,
+	PERF_COUNT_SW_CONTEXT_SWITCHES	= 3,
+	PERF_COUNT_SW_CPU_MIGRATIONS	= 4,
+	PERF_COUNT_SW_PAGE_FAULTS_MIN	= 5,
+	PERF_COUNT_SW_PAGE_FAULTS_MAJ	= 6,
+	PERF_COUNT_SW_ALIGNMENT_FAULTS	= 7,
+	PERF_COUNT_SW_EMULATION_FAULTS	= 8,
+};
+
+Counters of the type PERF_TYPE_TRACEPOINT are available when the ftrace event
+tracer is available, and event_id values can be obtained from
+/debug/tracing/events/*/*/id
+
+
+Counters come in two flavours: counting counters and sampling
+counters.  A "counting" counter is one that is used for counting the
+number of events that occur, and is characterised by having
+irq_period = 0.
+
+
+A read() on a counter returns the current value of the counter and possible
+additional values as specified by 'read_format', each value is a u64 (8 bytes)
+in size.
+
+/*
+ * Bits that can be set in hw_event.read_format to request that
+ * reads on the counter should return the indicated quantities,
+ * in increasing order of bit value, after the counter value.
+ */
+enum perf_event_read_format {
+        PERF_FORMAT_TOTAL_TIME_ENABLED  =  1,
+        PERF_FORMAT_TOTAL_TIME_RUNNING  =  2,
+};
+
+Using these additional values one can establish the overcommit ratio for a
+particular counter allowing one to take the round-robin scheduling effect
+into account.
+
+
+A "sampling" counter is one that is set up to generate an interrupt
+every N events, where N is given by 'irq_period'.  A sampling counter
+has irq_period > 0. The record_type controls what data is recorded on each
+interrupt:
+
+/*
+ * Bits that can be set in hw_event.record_type to request information
+ * in the overflow packets.
+ */
+enum perf_event_record_format {
+        PERF_RECORD_IP          = 1U << 0,
+        PERF_RECORD_TID         = 1U << 1,
+        PERF_RECORD_TIME        = 1U << 2,
+        PERF_RECORD_ADDR        = 1U << 3,
+        PERF_RECORD_GROUP       = 1U << 4,
+        PERF_RECORD_CALLCHAIN   = 1U << 5,
+};
+
+Such (and other) events will be recorded in a ring-buffer, which is
+available to user-space using mmap() (see below).
+
+The 'disabled' bit specifies whether the counter starts out disabled
+or enabled.  If it is initially disabled, it can be enabled by ioctl
+or prctl (see below).
+
+The 'inherit' bit, if set, specifies that this counter should count
+events on descendant tasks as well as the task specified.  This only
+applies to new descendents, not to any existing descendents at the
+time the counter is created (nor to any new descendents of existing
+descendents).
+
+The 'pinned' bit, if set, specifies that the counter should always be
+on the CPU if at all possible.  It only applies to hardware counters
+and only to group leaders.  If a pinned counter cannot be put onto the
+CPU (e.g. because there are not enough hardware counters or because of
+a conflict with some other event), then the counter goes into an
+'error' state, where reads return end-of-file (i.e. read() returns 0)
+until the counter is subsequently enabled or disabled.
+
+The 'exclusive' bit, if set, specifies that when this counter's group
+is on the CPU, it should be the only group using the CPU's counters.
+In future, this will allow sophisticated monitoring programs to supply
+extra configuration information via 'extra_config_len' to exploit
+advanced features of the CPU's Performance Monitor Unit (PMU) that are
+not otherwise accessible and that might disrupt other hardware
+counters.
+
+The 'exclude_user', 'exclude_kernel' and 'exclude_hv' bits provide a
+way to request that counting of events be restricted to times when the
+CPU is in user, kernel and/or hypervisor mode.
+
+The 'mmap' and 'munmap' bits allow recording of PROT_EXEC mmap/munmap
+operations, these can be used to relate userspace IP addresses to actual
+code, even after the mapping (or even the whole process) is gone,
+these events are recorded in the ring-buffer (see below).
+
+The 'comm' bit allows tracking of process comm data on process creation.
+This too is recorded in the ring-buffer (see below).
+
+The 'pid' parameter to the sys_perf_event_open() system call allows the
+counter to be specific to a task:
+
+ pid == 0: if the pid parameter is zero, the counter is attached to the
+ current task.
+
+ pid > 0: the counter is attached to a specific task (if the current task
+ has sufficient privilege to do so)
+
+ pid < 0: all tasks are counted (per cpu counters)
+
+The 'cpu' parameter allows a counter to be made specific to a CPU:
+
+ cpu >= 0: the counter is restricted to a specific CPU
+ cpu == -1: the counter counts on all CPUs
+
+(Note: the combination of 'pid == -1' and 'cpu == -1' is not valid.)
+
+A 'pid > 0' and 'cpu == -1' counter is a per task counter that counts
+events of that task and 'follows' that task to whatever CPU the task
+gets schedule to. Per task counters can be created by any user, for
+their own tasks.
+
+A 'pid == -1' and 'cpu == x' counter is a per CPU counter that counts
+all events on CPU-x. Per CPU counters need CAP_SYS_ADMIN privilege.
+
+The 'flags' parameter is currently unused and must be zero.
+
+The 'group_fd' parameter allows counter "groups" to be set up.  A
+counter group has one counter which is the group "leader".  The leader
+is created first, with group_fd = -1 in the sys_perf_event_open call
+that creates it.  The rest of the group members are created
+subsequently, with group_fd giving the fd of the group leader.
+(A single counter on its own is created with group_fd = -1 and is
+considered to be a group with only 1 member.)
+
+A counter group is scheduled onto the CPU as a unit, that is, it will
+only be put onto the CPU if all of the counters in the group can be
+put onto the CPU.  This means that the values of the member counters
+can be meaningfully compared, added, divided (to get ratios), etc.,
+with each other, since they have counted events for the same set of
+executed instructions.
+
+
+Like stated, asynchronous events, like counter overflow or PROT_EXEC mmap
+tracking are logged into a ring-buffer. This ring-buffer is created and
+accessed through mmap().
+
+The mmap size should be 1+2^n pages, where the first page is a meta-data page
+(struct perf_event_mmap_page) that contains various bits of information such
+as where the ring-buffer head is.
+
+/*
+ * Structure of the page that can be mapped via mmap
+ */
+struct perf_event_mmap_page {
+        __u32   version;                /* version number of this structure */
+        __u32   compat_version;         /* lowest version this is compat with */
+
+        /*
+         * Bits needed to read the hw counters in user-space.
+         *
+         *   u32 seq;
+         *   s64 count;
+         *
+         *   do {
+         *     seq = pc->lock;
+         *
+         *     barrier()
+         *     if (pc->index) {
+         *       count = pmc_read(pc->index - 1);
+         *       count += pc->offset;
+         *     } else
+         *       goto regular_read;
+         *
+         *     barrier();
+         *   } while (pc->lock != seq);
+         *
+         * NOTE: for obvious reason this only works on self-monitoring
+         *       processes.
+         */
+        __u32   lock;                   /* seqlock for synchronization */
+        __u32   index;                  /* hardware counter identifier */
+        __s64   offset;                 /* add to hardware counter value */
+
+        /*
+         * Control data for the mmap() data buffer.
+         *
+         * User-space reading this value should issue an rmb(), on SMP capable
+         * platforms, after reading this value -- see perf_event_wakeup().
+         */
+        __u32   data_head;              /* head in the data section */
+};
+
+NOTE: the hw-counter userspace bits are arch specific and are currently only
+      implemented on powerpc.
+
+The following 2^n pages are the ring-buffer which contains events of the form:
+
+#define PERF_RECORD_MISC_KERNEL          (1 << 0)
+#define PERF_RECORD_MISC_USER            (1 << 1)
+#define PERF_RECORD_MISC_OVERFLOW        (1 << 2)
+
+struct perf_event_header {
+        __u32   type;
+        __u16   misc;
+        __u16   size;
+};
+
+enum perf_event_type {
+
+        /*
+         * The MMAP events record the PROT_EXEC mappings so that we can
+         * correlate userspace IPs to code. They have the following structure:
+         *
+         * struct {
+         *      struct perf_event_header        header;
+         *
+         *      u32                             pid, tid;
+         *      u64                             addr;
+         *      u64                             len;
+         *      u64                             pgoff;
+         *      char                            filename[];
+         * };
+         */
+        PERF_RECORD_MMAP                 = 1,
+        PERF_RECORD_MUNMAP               = 2,
+
+        /*
+         * struct {
+         *      struct perf_event_header        header;
+         *
+         *      u32                             pid, tid;
+         *      char                            comm[];
+         * };
+         */
+        PERF_RECORD_COMM                 = 3,
+
+        /*
+         * When header.misc & PERF_RECORD_MISC_OVERFLOW the event_type field
+         * will be PERF_RECORD_*
+         *
+         * struct {
+         *      struct perf_event_header        header;
+         *
+         *      { u64                   ip;       } && PERF_RECORD_IP
+         *      { u32                   pid, tid; } && PERF_RECORD_TID
+         *      { u64                   time;     } && PERF_RECORD_TIME
+         *      { u64                   addr;     } && PERF_RECORD_ADDR
+         *
+         *      { u64                   nr;
+         *        { u64 event, val; }   cnt[nr];  } && PERF_RECORD_GROUP
+         *
+         *      { u16                   nr,
+         *                              hv,
+         *                              kernel,
+         *                              user;
+         *        u64                   ips[nr];  } && PERF_RECORD_CALLCHAIN
+         * };
+         */
+};
+
+NOTE: PERF_RECORD_CALLCHAIN is arch specific and currently only implemented
+      on x86.
+
+Notification of new events is possible through poll()/select()/epoll() and
+fcntl() managing signals.
+
+Normally a notification is generated for every page filled, however one can
+additionally set perf_event_attr.wakeup_events to generate one every
+so many counter overflow events.
+
+Future work will include a splice() interface to the ring-buffer.
+
+
+Counters can be enabled and disabled in two ways: via ioctl and via
+prctl.  When a counter is disabled, it doesn't count or generate
+events but does continue to exist and maintain its count value.
+
+An individual counter can be enabled with
+
+	ioctl(fd, PERF_EVENT_IOC_ENABLE, 0);
+
+or disabled with
+
+	ioctl(fd, PERF_EVENT_IOC_DISABLE, 0);
+
+For a counter group, pass PERF_IOC_FLAG_GROUP as the third argument.
+Enabling or disabling the leader of a group enables or disables the
+whole group; that is, while the group leader is disabled, none of the
+counters in the group will count.  Enabling or disabling a member of a
+group other than the leader only affects that counter - disabling an
+non-leader stops that counter from counting but doesn't affect any
+other counter.
+
+Additionally, non-inherited overflow counters can use
+
+	ioctl(fd, PERF_EVENT_IOC_REFRESH, nr);
+
+to enable a counter for 'nr' events, after which it gets disabled again.
+
+A process can enable or disable all the counter groups that are
+attached to it, using prctl:
+
+	prctl(PR_TASK_PERF_EVENTS_ENABLE);
+
+	prctl(PR_TASK_PERF_EVENTS_DISABLE);
+
+This applies to all counters on the current process, whether created
+by this process or by another, and doesn't affect any counters that
+this process has created on other processes.  It only enables or
+disables the group leaders, not any other members in the groups.
+
+
+Arch requirements
+-----------------
+
+If your architecture does not have hardware performance metrics, you can
+still use the generic software counters based on hrtimers for sampling.
+
+So to start with, in order to add HAVE_PERF_EVENTS to your Kconfig, you
+will need at least this:
+	- asm/perf_event.h - a basic stub will suffice at first
+	- support for atomic64 types (and associated helper functions)
+
+If your architecture does have hardware capabilities, you can override the
+weak stub hw_perf_event_init() to register hardware counters.
+
+Architectures that have d-cache aliassing issues, such as Sparc and ARM,
+should select PERF_USE_VMALLOC in order to avoid these for perf mmap().
diff --git a/tools/perf/perf-archive.sh b/tools/perf/perf-archive.sh
new file mode 100644
index 000000000..e91930620
--- /dev/null
+++ b/tools/perf/perf-archive.sh
@@ -0,0 +1,47 @@
+#!/bin/bash
+# perf archive
+# Arnaldo Carvalho de Melo <acme@redhat.com>
+
+PERF_DATA=perf.data
+if [ $# -ne 0 ] ; then
+	PERF_DATA=$1
+fi
+
+#
+# PERF_BUILDID_DIR environment variable set by perf
+# path to buildid directory, default to $HOME/.debug
+#
+if [ -z $PERF_BUILDID_DIR ]; then
+	PERF_BUILDID_DIR=~/.debug/
+else
+        # append / to make substitutions work
+        PERF_BUILDID_DIR=$PERF_BUILDID_DIR/
+fi
+
+BUILDIDS=$(mktemp /tmp/perf-archive-buildids.XXXXXX)
+NOBUILDID=0000000000000000000000000000000000000000
+
+perf buildid-list -i $PERF_DATA --with-hits | grep -v "^$NOBUILDID " > $BUILDIDS
+if [ ! -s $BUILDIDS ] ; then
+	echo "perf archive: no build-ids found"
+	rm $BUILDIDS || true
+	exit 1
+fi
+
+MANIFEST=$(mktemp /tmp/perf-archive-manifest.XXXXXX)
+PERF_BUILDID_LINKDIR=$(readlink -f $PERF_BUILDID_DIR)/
+
+cut -d ' ' -f 1 $BUILDIDS | \
+while read build_id ; do
+	linkname=$PERF_BUILDID_DIR.build-id/${build_id:0:2}/${build_id:2}
+	filename=$(readlink -f $linkname)
+	echo ${linkname#$PERF_BUILDID_DIR} >> $MANIFEST
+	echo ${filename#$PERF_BUILDID_LINKDIR} >> $MANIFEST
+done
+
+tar cjf $PERF_DATA.tar.bz2 -C $PERF_BUILDID_DIR -T $MANIFEST
+rm $MANIFEST $BUILDIDS || true
+echo -e "Now please run:\n"
+echo -e "$ tar xvf $PERF_DATA.tar.bz2 -C ~/.debug\n"
+echo "wherever you need to run 'perf report' on."
+exit 0
diff --git a/tools/perf/perf-completion.sh b/tools/perf/perf-completion.sh
new file mode 100644
index 000000000..3ba80b235
--- /dev/null
+++ b/tools/perf/perf-completion.sh
@@ -0,0 +1,258 @@
+# perf bash and zsh completion
+
+# Taken from git.git's completion script.
+__my_reassemble_comp_words_by_ref()
+{
+	local exclude i j first
+	# Which word separators to exclude?
+	exclude="${1//[^$COMP_WORDBREAKS]}"
+	cword_=$COMP_CWORD
+	if [ -z "$exclude" ]; then
+		words_=("${COMP_WORDS[@]}")
+		return
+	fi
+	# List of word completion separators has shrunk;
+	# re-assemble words to complete.
+	for ((i=0, j=0; i < ${#COMP_WORDS[@]}; i++, j++)); do
+		# Append each nonempty word consisting of just
+		# word separator characters to the current word.
+		first=t
+		while
+			[ $i -gt 0 ] &&
+			[ -n "${COMP_WORDS[$i]}" ] &&
+			# word consists of excluded word separators
+			[ "${COMP_WORDS[$i]//[^$exclude]}" = "${COMP_WORDS[$i]}" ]
+		do
+			# Attach to the previous token,
+			# unless the previous token is the command name.
+			if [ $j -ge 2 ] && [ -n "$first" ]; then
+				((j--))
+			fi
+			first=
+			words_[$j]=${words_[j]}${COMP_WORDS[i]}
+			if [ $i = $COMP_CWORD ]; then
+				cword_=$j
+			fi
+			if (($i < ${#COMP_WORDS[@]} - 1)); then
+				((i++))
+			else
+				# Done.
+				return
+			fi
+		done
+		words_[$j]=${words_[j]}${COMP_WORDS[i]}
+		if [ $i = $COMP_CWORD ]; then
+			cword_=$j
+		fi
+	done
+}
+
+# Define preload_get_comp_words_by_ref="false", if the function
+# __perf_get_comp_words_by_ref() is required instead.
+preload_get_comp_words_by_ref="true"
+
+if [ $preload_get_comp_words_by_ref = "true" ]; then
+	type _get_comp_words_by_ref &>/dev/null ||
+	preload_get_comp_words_by_ref="false"
+fi
+[ $preload_get_comp_words_by_ref = "true" ] ||
+__perf_get_comp_words_by_ref()
+{
+	local exclude cur_ words_ cword_
+	if [ "$1" = "-n" ]; then
+		exclude=$2
+		shift 2
+	fi
+	__my_reassemble_comp_words_by_ref "$exclude"
+	cur_=${words_[cword_]}
+	while [ $# -gt 0 ]; do
+		case "$1" in
+		cur)
+			cur=$cur_
+			;;
+		prev)
+			prev=${words_[$cword_-1]}
+			;;
+		words)
+			words=("${words_[@]}")
+			;;
+		cword)
+			cword=$cword_
+			;;
+		esac
+		shift
+	done
+}
+
+# Define preload__ltrim_colon_completions="false", if the function
+# __perf__ltrim_colon_completions() is required instead.
+preload__ltrim_colon_completions="true"
+
+if [ $preload__ltrim_colon_completions = "true" ]; then
+	type __ltrim_colon_completions &>/dev/null ||
+	preload__ltrim_colon_completions="false"
+fi
+[ $preload__ltrim_colon_completions = "true" ] ||
+__perf__ltrim_colon_completions()
+{
+	if [[ "$1" == *:* && "$COMP_WORDBREAKS" == *:* ]]; then
+		# Remove colon-word prefix from COMPREPLY items
+		local colon_word=${1%"${1##*:}"}
+		local i=${#COMPREPLY[*]}
+		while [[ $((--i)) -ge 0 ]]; do
+			COMPREPLY[$i]=${COMPREPLY[$i]#"$colon_word"}
+		done
+	fi
+}
+
+__perfcomp ()
+{
+	COMPREPLY=( $( compgen -W "$1" -- "$2" ) )
+}
+
+__perfcomp_colon ()
+{
+	__perfcomp "$1" "$2"
+	if [ $preload__ltrim_colon_completions = "true" ]; then
+		__ltrim_colon_completions $cur
+	else
+		__perf__ltrim_colon_completions $cur
+	fi
+}
+
+__perf_prev_skip_opts ()
+{
+	local i cmd_ cmds_
+
+	let i=cword-1
+	cmds_=$($cmd $1 --list-cmds)
+	prev_skip_opts=()
+	while [ $i -ge 0 ]; do
+		if [[ ${words[i]} == $1 ]]; then
+			return
+		fi
+		for cmd_ in $cmds_; do
+			if [[ ${words[i]} == $cmd_ ]]; then
+				prev_skip_opts=${words[i]}
+				return
+			fi
+		done
+		((i--))
+	done
+}
+
+__perf_main ()
+{
+	local cmd
+
+	cmd=${words[0]}
+	COMPREPLY=()
+
+	# Skip options backward and find the last perf command
+	__perf_prev_skip_opts
+	# List perf subcommands or long options
+	if [ -z $prev_skip_opts ]; then
+		if [[ $cur == --* ]]; then
+			cmds=$($cmd --list-opts)
+		else
+			cmds=$($cmd --list-cmds)
+		fi
+		__perfcomp "$cmds" "$cur"
+	# List possible events for -e option
+	elif [[ $prev == @("-e"|"--event") &&
+		$prev_skip_opts == @(record|stat|top) ]]; then
+		evts=$($cmd list --raw-dump)
+		__perfcomp_colon "$evts" "$cur"
+	else
+		# List subcommands for perf commands
+		if [[ $prev_skip_opts == @(kvm|kmem|mem|lock|sched|
+			|data|help|script|test|timechart|trace) ]]; then
+			subcmds=$($cmd $prev_skip_opts --list-cmds)
+			__perfcomp_colon "$subcmds" "$cur"
+		fi
+		# List long option names
+		if [[ $cur == --* ]];  then
+			subcmd=$prev_skip_opts
+			__perf_prev_skip_opts $subcmd
+			subcmd=$subcmd" "$prev_skip_opts
+			opts=$($cmd $subcmd --list-opts)
+			__perfcomp "$opts" "$cur"
+		fi
+	fi
+}
+
+if [[ -n ${ZSH_VERSION-} ]]; then
+	autoload -U +X compinit && compinit
+
+	__perfcomp ()
+	{
+		emulate -L zsh
+
+		local c IFS=$' \t\n'
+		local -a array
+
+		for c in ${=1}; do
+			case $c in
+			--*=*|*.) ;;
+			*) c="$c " ;;
+			esac
+			array[${#array[@]}+1]="$c"
+		done
+
+		compset -P '*[=:]'
+		compadd -Q -S '' -a -- array && _ret=0
+	}
+
+	__perfcomp_colon ()
+	{
+		emulate -L zsh
+
+		local cur_="${2-$cur}"
+		local c IFS=$' \t\n'
+		local -a array
+
+		if [[ "$cur_" == *:* ]]; then
+			local colon_word=${cur_%"${cur_##*:}"}
+		fi
+
+		for c in ${=1}; do
+			case $c in
+			--*=*|*.) ;;
+			*) c="$c " ;;
+			esac
+			array[$#array+1]=${c#"$colon_word"}
+		done
+
+		compset -P '*[=:]'
+		compadd -Q -S '' -a -- array && _ret=0
+	}
+
+	_perf ()
+	{
+		local _ret=1 cur cword prev
+		cur=${words[CURRENT]}
+		prev=${words[CURRENT-1]}
+		let cword=CURRENT-1
+		emulate ksh -c __perf_main
+		let _ret && _default && _ret=0
+		return _ret
+	}
+
+	compdef _perf perf
+	return
+fi
+
+type perf &>/dev/null &&
+_perf()
+{
+	local cur words cword prev
+	if [ $preload_get_comp_words_by_ref = "true" ]; then
+		_get_comp_words_by_ref -n =: cur words cword prev
+	else
+		__perf_get_comp_words_by_ref -n =: cur words cword prev
+	fi
+	__perf_main
+} &&
+
+complete -o bashdefault -o default -o nospace -F _perf perf 2>/dev/null \
+	|| complete -o default -o nospace -F _perf perf
diff --git a/tools/perf/perf-read-vdso.c b/tools/perf/perf-read-vdso.c
new file mode 100644
index 000000000..764e2547c
--- /dev/null
+++ b/tools/perf/perf-read-vdso.c
@@ -0,0 +1,34 @@
+#include <stdio.h>
+#include <string.h>
+
+#define VDSO__MAP_NAME "[vdso]"
+
+/*
+ * Include definition of find_vdso_map() also used in util/vdso.c for
+ * building perf.
+ */
+#include "util/find-vdso-map.c"
+
+int main(void)
+{
+	void *start, *end;
+	size_t size, written;
+
+	if (find_vdso_map(&start, &end))
+		return 1;
+
+	size = end - start;
+
+	while (size) {
+		written = fwrite(start, 1, size, stdout);
+		if (!written)
+			return 1;
+		start += written;
+		size -= written;
+	}
+
+	if (fflush(stdout))
+		return 1;
+
+	return 0;
+}
diff --git a/tools/perf/perf-sys.h b/tools/perf/perf-sys.h
new file mode 100644
index 000000000..6ef68165c
--- /dev/null
+++ b/tools/perf/perf-sys.h
@@ -0,0 +1,190 @@
+#ifndef _PERF_SYS_H
+#define _PERF_SYS_H
+
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/syscall.h>
+#include <linux/types.h>
+#include <linux/perf_event.h>
+
+#if defined(__i386__)
+#define mb()		asm volatile("lock; addl $0,0(%%esp)" ::: "memory")
+#define wmb()		asm volatile("lock; addl $0,0(%%esp)" ::: "memory")
+#define rmb()		asm volatile("lock; addl $0,0(%%esp)" ::: "memory")
+#define cpu_relax()	asm volatile("rep; nop" ::: "memory");
+#define CPUINFO_PROC	{"model name"}
+#ifndef __NR_perf_event_open
+# define __NR_perf_event_open 336
+#endif
+#ifndef __NR_futex
+# define __NR_futex 240
+#endif
+#ifndef __NR_gettid
+# define __NR_gettid 224
+#endif
+#endif
+
+#if defined(__x86_64__)
+#define mb()		asm volatile("mfence" ::: "memory")
+#define wmb()		asm volatile("sfence" ::: "memory")
+#define rmb()		asm volatile("lfence" ::: "memory")
+#define cpu_relax()	asm volatile("rep; nop" ::: "memory");
+#define CPUINFO_PROC	{"model name"}
+#ifndef __NR_perf_event_open
+# define __NR_perf_event_open 298
+#endif
+#ifndef __NR_futex
+# define __NR_futex 202
+#endif
+#ifndef __NR_gettid
+# define __NR_gettid 186
+#endif
+#endif
+
+#ifdef __powerpc__
+#include "../../arch/powerpc/include/uapi/asm/unistd.h"
+#define mb()		asm volatile ("sync" ::: "memory")
+#define wmb()		asm volatile ("sync" ::: "memory")
+#define rmb()		asm volatile ("sync" ::: "memory")
+#define CPUINFO_PROC	{"cpu"}
+#endif
+
+#ifdef __s390__
+#define mb()		asm volatile("bcr 15,0" ::: "memory")
+#define wmb()		asm volatile("bcr 15,0" ::: "memory")
+#define rmb()		asm volatile("bcr 15,0" ::: "memory")
+#define CPUINFO_PROC	{"vendor_id"}
+#endif
+
+#ifdef __sh__
+#if defined(__SH4A__) || defined(__SH5__)
+# define mb()		asm volatile("synco" ::: "memory")
+# define wmb()		asm volatile("synco" ::: "memory")
+# define rmb()		asm volatile("synco" ::: "memory")
+#else
+# define mb()		asm volatile("" ::: "memory")
+# define wmb()		asm volatile("" ::: "memory")
+# define rmb()		asm volatile("" ::: "memory")
+#endif
+#define CPUINFO_PROC	{"cpu type"}
+#endif
+
+#ifdef __hppa__
+#define mb()		asm volatile("" ::: "memory")
+#define wmb()		asm volatile("" ::: "memory")
+#define rmb()		asm volatile("" ::: "memory")
+#define CPUINFO_PROC	{"cpu"}
+#endif
+
+#ifdef __sparc__
+#ifdef __LP64__
+#define mb()		asm volatile("ba,pt %%xcc, 1f\n"	\
+				     "membar #StoreLoad\n"	\
+				     "1:\n":::"memory")
+#else
+#define mb()		asm volatile("":::"memory")
+#endif
+#define wmb()		asm volatile("":::"memory")
+#define rmb()		asm volatile("":::"memory")
+#define CPUINFO_PROC	{"cpu"}
+#endif
+
+#ifdef __alpha__
+#define mb()		asm volatile("mb" ::: "memory")
+#define wmb()		asm volatile("wmb" ::: "memory")
+#define rmb()		asm volatile("mb" ::: "memory")
+#define CPUINFO_PROC	{"cpu model"}
+#endif
+
+#ifdef __ia64__
+#define mb()		asm volatile ("mf" ::: "memory")
+#define wmb()		asm volatile ("mf" ::: "memory")
+#define rmb()		asm volatile ("mf" ::: "memory")
+#define cpu_relax()	asm volatile ("hint @pause" ::: "memory")
+#define CPUINFO_PROC	{"model name"}
+#endif
+
+#ifdef __arm__
+/*
+ * Use the __kuser_memory_barrier helper in the CPU helper page. See
+ * arch/arm/kernel/entry-armv.S in the kernel source for details.
+ */
+#define mb()		((void(*)(void))0xffff0fa0)()
+#define wmb()		((void(*)(void))0xffff0fa0)()
+#define rmb()		((void(*)(void))0xffff0fa0)()
+#define CPUINFO_PROC	{"model name", "Processor"}
+#endif
+
+#ifdef __aarch64__
+#define mb()		asm volatile("dmb ish" ::: "memory")
+#define wmb()		asm volatile("dmb ishst" ::: "memory")
+#define rmb()		asm volatile("dmb ishld" ::: "memory")
+#define cpu_relax()	asm volatile("yield" ::: "memory")
+#endif
+
+#ifdef __mips__
+#define mb()		asm volatile(					\
+				".set	mips2\n\t"			\
+				"sync\n\t"				\
+				".set	mips0"				\
+				: /* no output */			\
+				: /* no input */			\
+				: "memory")
+#define wmb()	mb()
+#define rmb()	mb()
+#define CPUINFO_PROC	{"cpu model"}
+#endif
+
+#ifdef __arc__
+#define mb()		asm volatile("" ::: "memory")
+#define wmb()		asm volatile("" ::: "memory")
+#define rmb()		asm volatile("" ::: "memory")
+#define CPUINFO_PROC	{"Processor"}
+#endif
+
+#ifdef __metag__
+#define mb()		asm volatile("" ::: "memory")
+#define wmb()		asm volatile("" ::: "memory")
+#define rmb()		asm volatile("" ::: "memory")
+#define CPUINFO_PROC	{"CPU"}
+#endif
+
+#ifdef __xtensa__
+#define mb()		asm volatile("memw" ::: "memory")
+#define wmb()		asm volatile("memw" ::: "memory")
+#define rmb()		asm volatile("" ::: "memory")
+#define CPUINFO_PROC	{"core ID"}
+#endif
+
+#ifdef __tile__
+#define mb()		asm volatile ("mf" ::: "memory")
+#define wmb()		asm volatile ("mf" ::: "memory")
+#define rmb()		asm volatile ("mf" ::: "memory")
+#define cpu_relax()	asm volatile ("mfspr zero, PASS" ::: "memory")
+#define CPUINFO_PROC    {"model name"}
+#endif
+
+#define barrier() asm volatile ("" ::: "memory")
+
+#ifndef cpu_relax
+#define cpu_relax() barrier()
+#endif
+
+static inline int
+sys_perf_event_open(struct perf_event_attr *attr,
+		      pid_t pid, int cpu, int group_fd,
+		      unsigned long flags)
+{
+	int fd;
+
+	fd = syscall(__NR_perf_event_open, attr, pid, cpu,
+		     group_fd, flags);
+
+#ifdef HAVE_ATTR_TEST
+	if (unlikely(test_attr__enabled))
+		test_attr__open(attr, pid, cpu, fd, group_fd, flags);
+#endif
+	return fd;
+}
+
+#endif /* _PERF_SYS_H */
diff --git a/tools/perf/perf-with-kcore.sh b/tools/perf/perf-with-kcore.sh
new file mode 100644
index 000000000..c7ff90a90
--- /dev/null
+++ b/tools/perf/perf-with-kcore.sh
@@ -0,0 +1,259 @@
+#!/bin/bash
+# perf-with-kcore: use perf with a copy of kcore
+# Copyright (c) 2014, Intel Corporation.
+#
+# This program is free software; you can redistribute it and/or modify it
+# under the terms and conditions of the GNU General Public License,
+# version 2, as published by the Free Software Foundation.
+#
+# This program is distributed in the hope it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+# more details.
+
+set -e
+
+usage()
+{
+        echo "Usage: perf-with-kcore <perf sub-command> <perf.data directory> [<sub-command options> [ -- <workload>]]" >&2
+        echo "       <perf sub-command> can be record, script, report or inject" >&2
+        echo "   or: perf-with-kcore fix_buildid_cache_permissions" >&2
+        exit 1
+}
+
+find_perf()
+{
+	if [ -n "$PERF" ] ; then
+		return
+	fi
+	PERF=`which perf || true`
+	if [ -z "$PERF" ] ; then
+		echo "Failed to find perf" >&2
+	        exit 1
+	fi
+	if [ ! -x "$PERF" ] ; then
+		echo "Failed to find perf" >&2
+	        exit 1
+	fi
+	echo "Using $PERF"
+	"$PERF" version
+}
+
+copy_kcore()
+{
+	echo "Copying kcore"
+
+	if [ $EUID -eq 0 ] ; then
+		SUDO=""
+	else
+		SUDO="sudo"
+	fi
+
+	rm -f perf.data.junk
+	("$PERF" record -o perf.data.junk $PERF_OPTIONS -- sleep 60) >/dev/null 2>/dev/null &
+	PERF_PID=$!
+
+	# Need to make sure that perf has started
+	sleep 1
+
+	KCORE=$(($SUDO "$PERF" buildid-cache -v -f -k /proc/kcore >/dev/null) 2>&1)
+	case "$KCORE" in
+	"kcore added to build-id cache directory "*)
+		KCORE_DIR=${KCORE#"kcore added to build-id cache directory "}
+	;;
+	*)
+		kill $PERF_PID
+		wait >/dev/null 2>/dev/null || true
+		rm perf.data.junk
+		echo "$KCORE"
+		echo "Failed to find kcore" >&2
+		exit 1
+	;;
+	esac
+
+	kill $PERF_PID
+	wait >/dev/null 2>/dev/null || true
+	rm perf.data.junk
+
+	$SUDO cp -a "$KCORE_DIR" "$(pwd)/$PERF_DATA_DIR"
+	$SUDO rm -f "$KCORE_DIR/kcore"
+	$SUDO rm -f "$KCORE_DIR/kallsyms"
+	$SUDO rm -f "$KCORE_DIR/modules"
+	$SUDO rmdir "$KCORE_DIR"
+
+	KCORE_DIR_BASENAME=$(basename "$KCORE_DIR")
+	KCORE_DIR="$(pwd)/$PERF_DATA_DIR/$KCORE_DIR_BASENAME"
+
+	$SUDO chown $UID "$KCORE_DIR"
+	$SUDO chown $UID "$KCORE_DIR/kcore"
+	$SUDO chown $UID "$KCORE_DIR/kallsyms"
+	$SUDO chown $UID "$KCORE_DIR/modules"
+
+	$SUDO chgrp $GROUPS "$KCORE_DIR"
+	$SUDO chgrp $GROUPS "$KCORE_DIR/kcore"
+	$SUDO chgrp $GROUPS "$KCORE_DIR/kallsyms"
+	$SUDO chgrp $GROUPS "$KCORE_DIR/modules"
+
+	ln -s "$KCORE_DIR_BASENAME" "$PERF_DATA_DIR/kcore_dir"
+}
+
+fix_buildid_cache_permissions()
+{
+	if [ $EUID -ne 0 ] ; then
+		echo "This script must be run as root via sudo " >&2
+		exit 1
+	fi
+
+	if [ -z "$SUDO_USER" ] ; then
+		echo "This script must be run via sudo" >&2
+		exit 1
+	fi
+
+	USER_HOME=$(bash <<< "echo ~$SUDO_USER")
+
+	if [ "$HOME" != "$USER_HOME" ] ; then
+		echo "Fix unnecessary because root has a home: $HOME" >&2
+		exit 1
+	fi
+
+	echo "Fixing buildid cache permissions"
+
+	find "$USER_HOME/.debug" -xdev -type d          ! -user "$SUDO_USER" -ls -exec chown    "$SUDO_USER" \{\} \;
+	find "$USER_HOME/.debug" -xdev -type f -links 1 ! -user "$SUDO_USER" -ls -exec chown    "$SUDO_USER" \{\} \;
+	find "$USER_HOME/.debug" -xdev -type l          ! -user "$SUDO_USER" -ls -exec chown -h "$SUDO_USER" \{\} \;
+
+	if [ -n "$SUDO_GID" ] ; then
+		find "$USER_HOME/.debug" -xdev -type d          ! -group "$SUDO_GID" -ls -exec chgrp    "$SUDO_GID" \{\} \;
+		find "$USER_HOME/.debug" -xdev -type f -links 1 ! -group "$SUDO_GID" -ls -exec chgrp    "$SUDO_GID" \{\} \;
+		find "$USER_HOME/.debug" -xdev -type l          ! -group "$SUDO_GID" -ls -exec chgrp -h "$SUDO_GID" \{\} \;
+	fi
+
+	echo "Done"
+}
+
+check_buildid_cache_permissions()
+{
+	if [ $EUID -eq 0 ] ; then
+		return
+	fi
+
+	PERMISSIONS_OK+=$(find "$HOME/.debug" -xdev -type d          ! -user "$USER" -print -quit)
+	PERMISSIONS_OK+=$(find "$HOME/.debug" -xdev -type f -links 1 ! -user "$USER" -print -quit)
+	PERMISSIONS_OK+=$(find "$HOME/.debug" -xdev -type l          ! -user "$USER" -print -quit)
+
+	PERMISSIONS_OK+=$(find "$HOME/.debug" -xdev -type d          ! -group "$GROUPS" -print -quit)
+	PERMISSIONS_OK+=$(find "$HOME/.debug" -xdev -type f -links 1 ! -group "$GROUPS" -print -quit)
+	PERMISSIONS_OK+=$(find "$HOME/.debug" -xdev -type l          ! -group "$GROUPS" -print -quit)
+
+	if [ -n "$PERMISSIONS_OK" ] ; then
+		echo "*** WARNING *** buildid cache permissions may need fixing" >&2
+	fi
+}
+
+record()
+{
+	echo "Recording"
+
+	if [ $EUID -ne 0 ] ; then
+
+		if [ "$(cat /proc/sys/kernel/kptr_restrict)" -ne 0 ] ; then
+			echo "*** WARNING *** /proc/sys/kernel/kptr_restrict prevents access to kernel addresses" >&2
+		fi
+
+		if echo "$PERF_OPTIONS" | grep -q ' -a \|^-a \| -a$\|^-a$\| --all-cpus \|^--all-cpus \| --all-cpus$\|^--all-cpus$' ; then
+			echo "*** WARNING *** system-wide tracing without root access will not be able to read all necessary information from /proc" >&2
+		fi
+
+		if echo "$PERF_OPTIONS" | grep -q 'intel_pt\|intel_bts\| -I\|^-I' ; then
+			if [ "$(cat /proc/sys/kernel/perf_event_paranoid)" -gt -1 ] ; then
+				echo "*** WARNING *** /proc/sys/kernel/perf_event_paranoid restricts buffer size and tracepoint (sched_switch) use" >&2
+			fi
+
+			if echo "$PERF_OPTIONS" | grep -q ' --per-thread \|^--per-thread \| --per-thread$\|^--per-thread$' ; then
+				true
+			elif echo "$PERF_OPTIONS" | grep -q ' -t \|^-t \| -t$\|^-t$' ; then
+				true
+			elif [ ! -r /sys/kernel/debug -o ! -x /sys/kernel/debug ] ; then
+				echo "*** WARNING *** /sys/kernel/debug permissions prevent tracepoint (sched_switch) use" >&2
+			fi
+		fi
+	fi
+
+	if [ -z "$1" ] ; then
+		echo "Workload is required for recording" >&2
+		usage
+	fi
+
+	if [ -e "$PERF_DATA_DIR" ] ; then
+		echo "'$PERF_DATA_DIR' exists" >&2
+		exit 1
+	fi
+
+	find_perf
+
+	mkdir "$PERF_DATA_DIR"
+
+	echo "$PERF record -o $PERF_DATA_DIR/perf.data $PERF_OPTIONS -- $*"
+	"$PERF" record -o "$PERF_DATA_DIR/perf.data" $PERF_OPTIONS -- $* || true
+
+	if rmdir "$PERF_DATA_DIR" > /dev/null 2>/dev/null ; then
+		exit 1
+	fi
+
+	copy_kcore
+
+	echo "Done"
+}
+
+subcommand()
+{
+	find_perf
+	check_buildid_cache_permissions
+	echo "$PERF $PERF_SUB_COMMAND -i $PERF_DATA_DIR/perf.data --kallsyms=$PERF_DATA_DIR/kcore_dir/kallsyms $*"
+	"$PERF" $PERF_SUB_COMMAND -i "$PERF_DATA_DIR/perf.data" "--kallsyms=$PERF_DATA_DIR/kcore_dir/kallsyms" $*
+}
+
+if [ "$1" = "fix_buildid_cache_permissions" ] ; then
+	fix_buildid_cache_permissions
+	exit 0
+fi
+
+PERF_SUB_COMMAND=$1
+PERF_DATA_DIR=$2
+shift || true
+shift || true
+
+if [ -z "$PERF_SUB_COMMAND" ] ; then
+	usage
+fi
+
+if [ -z "$PERF_DATA_DIR" ] ; then
+	usage
+fi
+
+case "$PERF_SUB_COMMAND" in
+"record")
+	while [ "$1" != "--" ] ; do
+		PERF_OPTIONS+="$1 "
+		shift || break
+	done
+	if [ "$1" != "--" ] ; then
+		echo "Options and workload are required for recording" >&2
+		usage
+	fi
+	shift
+	record $*
+;;
+"script")
+	subcommand $*
+;;
+"report")
+	subcommand $*
+;;
+"inject")
+	subcommand $*
+;;
+*)
+	usage
+;;
+esac
diff --git a/tools/perf/perf.c b/tools/perf/perf.c
new file mode 100644
index 000000000..b857fcbd0
--- /dev/null
+++ b/tools/perf/perf.c
@@ -0,0 +1,610 @@
+/*
+ * perf.c
+ *
+ * Performance analysis utility.
+ *
+ * This is the main hub from which the sub-commands (perf stat,
+ * perf top, perf record, perf report, etc.) are started.
+ */
+#include "builtin.h"
+
+#include "util/exec_cmd.h"
+#include "util/cache.h"
+#include "util/quote.h"
+#include "util/run-command.h"
+#include "util/parse-events.h"
+#include "util/parse-options.h"
+#include "util/debug.h"
+#include <api/fs/debugfs.h>
+#include <pthread.h>
+
+const char perf_usage_string[] =
+	"perf [--version] [--help] [OPTIONS] COMMAND [ARGS]";
+
+const char perf_more_info_string[] =
+	"See 'perf help COMMAND' for more information on a specific command.";
+
+int use_browser = -1;
+static int use_pager = -1;
+const char *input_name;
+
+struct cmd_struct {
+	const char *cmd;
+	int (*fn)(int, const char **, const char *);
+	int option;
+};
+
+static struct cmd_struct commands[] = {
+	{ "buildid-cache", cmd_buildid_cache, 0 },
+	{ "buildid-list", cmd_buildid_list, 0 },
+	{ "diff",	cmd_diff,	0 },
+	{ "evlist",	cmd_evlist,	0 },
+	{ "help",	cmd_help,	0 },
+	{ "list",	cmd_list,	0 },
+	{ "record",	cmd_record,	0 },
+	{ "report",	cmd_report,	0 },
+	{ "bench",	cmd_bench,	0 },
+	{ "stat",	cmd_stat,	0 },
+	{ "timechart",	cmd_timechart,	0 },
+	{ "top",	cmd_top,	0 },
+	{ "annotate",	cmd_annotate,	0 },
+	{ "version",	cmd_version,	0 },
+	{ "script",	cmd_script,	0 },
+	{ "sched",	cmd_sched,	0 },
+#ifdef HAVE_LIBELF_SUPPORT
+	{ "probe",	cmd_probe,	0 },
+#endif
+	{ "kmem",	cmd_kmem,	0 },
+	{ "lock",	cmd_lock,	0 },
+	{ "kvm",	cmd_kvm,	0 },
+	{ "test",	cmd_test,	0 },
+#ifdef HAVE_LIBAUDIT_SUPPORT
+	{ "trace",	cmd_trace,	0 },
+#endif
+	{ "inject",	cmd_inject,	0 },
+	{ "mem",	cmd_mem,	0 },
+	{ "data",	cmd_data,	0 },
+};
+
+struct pager_config {
+	const char *cmd;
+	int val;
+};
+
+static int pager_command_config(const char *var, const char *value, void *data)
+{
+	struct pager_config *c = data;
+	if (!prefixcmp(var, "pager.") && !strcmp(var + 6, c->cmd))
+		c->val = perf_config_bool(var, value);
+	return 0;
+}
+
+/* returns 0 for "no pager", 1 for "use pager", and -1 for "not specified" */
+int check_pager_config(const char *cmd)
+{
+	struct pager_config c;
+	c.cmd = cmd;
+	c.val = -1;
+	perf_config(pager_command_config, &c);
+	return c.val;
+}
+
+static int browser_command_config(const char *var, const char *value, void *data)
+{
+	struct pager_config *c = data;
+	if (!prefixcmp(var, "tui.") && !strcmp(var + 4, c->cmd))
+		c->val = perf_config_bool(var, value);
+	if (!prefixcmp(var, "gtk.") && !strcmp(var + 4, c->cmd))
+		c->val = perf_config_bool(var, value) ? 2 : 0;
+	return 0;
+}
+
+/*
+ * returns 0 for "no tui", 1 for "use tui", 2 for "use gtk",
+ * and -1 for "not specified"
+ */
+static int check_browser_config(const char *cmd)
+{
+	struct pager_config c;
+	c.cmd = cmd;
+	c.val = -1;
+	perf_config(browser_command_config, &c);
+	return c.val;
+}
+
+static void commit_pager_choice(void)
+{
+	switch (use_pager) {
+	case 0:
+		setenv("PERF_PAGER", "cat", 1);
+		break;
+	case 1:
+		/* setup_pager(); */
+		break;
+	default:
+		break;
+	}
+}
+
+struct option options[] = {
+	OPT_ARGUMENT("help", "help"),
+	OPT_ARGUMENT("version", "version"),
+	OPT_ARGUMENT("exec-path", "exec-path"),
+	OPT_ARGUMENT("html-path", "html-path"),
+	OPT_ARGUMENT("paginate", "paginate"),
+	OPT_ARGUMENT("no-pager", "no-pager"),
+	OPT_ARGUMENT("perf-dir", "perf-dir"),
+	OPT_ARGUMENT("work-tree", "work-tree"),
+	OPT_ARGUMENT("debugfs-dir", "debugfs-dir"),
+	OPT_ARGUMENT("buildid-dir", "buildid-dir"),
+	OPT_ARGUMENT("list-cmds", "list-cmds"),
+	OPT_ARGUMENT("list-opts", "list-opts"),
+	OPT_ARGUMENT("debug", "debug"),
+	OPT_END()
+};
+
+static int handle_options(const char ***argv, int *argc, int *envchanged)
+{
+	int handled = 0;
+
+	while (*argc > 0) {
+		const char *cmd = (*argv)[0];
+		if (cmd[0] != '-')
+			break;
+
+		/*
+		 * For legacy reasons, the "version" and "help"
+		 * commands can be written with "--" prepended
+		 * to make them look like flags.
+		 */
+		if (!strcmp(cmd, "--help") || !strcmp(cmd, "--version"))
+			break;
+
+		/*
+		 * Check remaining flags.
+		 */
+		if (!prefixcmp(cmd, CMD_EXEC_PATH)) {
+			cmd += strlen(CMD_EXEC_PATH);
+			if (*cmd == '=')
+				perf_set_argv_exec_path(cmd + 1);
+			else {
+				puts(perf_exec_path());
+				exit(0);
+			}
+		} else if (!strcmp(cmd, "--html-path")) {
+			puts(system_path(PERF_HTML_PATH));
+			exit(0);
+		} else if (!strcmp(cmd, "-p") || !strcmp(cmd, "--paginate")) {
+			use_pager = 1;
+		} else if (!strcmp(cmd, "--no-pager")) {
+			use_pager = 0;
+			if (envchanged)
+				*envchanged = 1;
+		} else if (!strcmp(cmd, "--perf-dir")) {
+			if (*argc < 2) {
+				fprintf(stderr, "No directory given for --perf-dir.\n");
+				usage(perf_usage_string);
+			}
+			setenv(PERF_DIR_ENVIRONMENT, (*argv)[1], 1);
+			if (envchanged)
+				*envchanged = 1;
+			(*argv)++;
+			(*argc)--;
+			handled++;
+		} else if (!prefixcmp(cmd, CMD_PERF_DIR)) {
+			setenv(PERF_DIR_ENVIRONMENT, cmd + strlen(CMD_PERF_DIR), 1);
+			if (envchanged)
+				*envchanged = 1;
+		} else if (!strcmp(cmd, "--work-tree")) {
+			if (*argc < 2) {
+				fprintf(stderr, "No directory given for --work-tree.\n");
+				usage(perf_usage_string);
+			}
+			setenv(PERF_WORK_TREE_ENVIRONMENT, (*argv)[1], 1);
+			if (envchanged)
+				*envchanged = 1;
+			(*argv)++;
+			(*argc)--;
+		} else if (!prefixcmp(cmd, CMD_WORK_TREE)) {
+			setenv(PERF_WORK_TREE_ENVIRONMENT, cmd + strlen(CMD_WORK_TREE), 1);
+			if (envchanged)
+				*envchanged = 1;
+		} else if (!strcmp(cmd, "--debugfs-dir")) {
+			if (*argc < 2) {
+				fprintf(stderr, "No directory given for --debugfs-dir.\n");
+				usage(perf_usage_string);
+			}
+			perf_debugfs_set_path((*argv)[1]);
+			if (envchanged)
+				*envchanged = 1;
+			(*argv)++;
+			(*argc)--;
+		} else if (!strcmp(cmd, "--buildid-dir")) {
+			if (*argc < 2) {
+				fprintf(stderr, "No directory given for --buildid-dir.\n");
+				usage(perf_usage_string);
+			}
+			set_buildid_dir((*argv)[1]);
+			if (envchanged)
+				*envchanged = 1;
+			(*argv)++;
+			(*argc)--;
+		} else if (!prefixcmp(cmd, CMD_DEBUGFS_DIR)) {
+			perf_debugfs_set_path(cmd + strlen(CMD_DEBUGFS_DIR));
+			fprintf(stderr, "dir: %s\n", debugfs_mountpoint);
+			if (envchanged)
+				*envchanged = 1;
+		} else if (!strcmp(cmd, "--list-cmds")) {
+			unsigned int i;
+
+			for (i = 0; i < ARRAY_SIZE(commands); i++) {
+				struct cmd_struct *p = commands+i;
+				printf("%s ", p->cmd);
+			}
+			putchar('\n');
+			exit(0);
+		} else if (!strcmp(cmd, "--list-opts")) {
+			unsigned int i;
+
+			for (i = 0; i < ARRAY_SIZE(options)-1; i++) {
+				struct option *p = options+i;
+				printf("--%s ", p->long_name);
+			}
+			putchar('\n');
+			exit(0);
+		} else if (!strcmp(cmd, "--debug")) {
+			if (*argc < 2) {
+				fprintf(stderr, "No variable specified for --debug.\n");
+				usage(perf_usage_string);
+			}
+			if (perf_debug_option((*argv)[1]))
+				usage(perf_usage_string);
+
+			(*argv)++;
+			(*argc)--;
+		} else {
+			fprintf(stderr, "Unknown option: %s\n", cmd);
+			usage(perf_usage_string);
+		}
+
+		(*argv)++;
+		(*argc)--;
+		handled++;
+	}
+	return handled;
+}
+
+static int handle_alias(int *argcp, const char ***argv)
+{
+	int envchanged = 0, ret = 0, saved_errno = errno;
+	int count, option_count;
+	const char **new_argv;
+	const char *alias_command;
+	char *alias_string;
+
+	alias_command = (*argv)[0];
+	alias_string = alias_lookup(alias_command);
+	if (alias_string) {
+		if (alias_string[0] == '!') {
+			if (*argcp > 1) {
+				struct strbuf buf;
+
+				strbuf_init(&buf, PATH_MAX);
+				strbuf_addstr(&buf, alias_string);
+				sq_quote_argv(&buf, (*argv) + 1, PATH_MAX);
+				free(alias_string);
+				alias_string = buf.buf;
+			}
+			ret = system(alias_string + 1);
+			if (ret >= 0 && WIFEXITED(ret) &&
+			    WEXITSTATUS(ret) != 127)
+				exit(WEXITSTATUS(ret));
+			die("Failed to run '%s' when expanding alias '%s'",
+			    alias_string + 1, alias_command);
+		}
+		count = split_cmdline(alias_string, &new_argv);
+		if (count < 0)
+			die("Bad alias.%s string", alias_command);
+		option_count = handle_options(&new_argv, &count, &envchanged);
+		if (envchanged)
+			die("alias '%s' changes environment variables\n"
+				 "You can use '!perf' in the alias to do this.",
+				 alias_command);
+		memmove(new_argv - option_count, new_argv,
+				count * sizeof(char *));
+		new_argv -= option_count;
+
+		if (count < 1)
+			die("empty alias for %s", alias_command);
+
+		if (!strcmp(alias_command, new_argv[0]))
+			die("recursive alias: %s", alias_command);
+
+		new_argv = realloc(new_argv, sizeof(char *) *
+				    (count + *argcp + 1));
+		/* insert after command name */
+		memcpy(new_argv + count, *argv + 1, sizeof(char *) * *argcp);
+		new_argv[count + *argcp] = NULL;
+
+		*argv = new_argv;
+		*argcp += count - 1;
+
+		ret = 1;
+	}
+
+	errno = saved_errno;
+
+	return ret;
+}
+
+const char perf_version_string[] = PERF_VERSION;
+
+#define RUN_SETUP	(1<<0)
+#define USE_PAGER	(1<<1)
+/*
+ * require working tree to be present -- anything uses this needs
+ * RUN_SETUP for reading from the configuration file.
+ */
+#define NEED_WORK_TREE	(1<<2)
+
+static int run_builtin(struct cmd_struct *p, int argc, const char **argv)
+{
+	int status;
+	struct stat st;
+	const char *prefix;
+	char sbuf[STRERR_BUFSIZE];
+
+	prefix = NULL;
+	if (p->option & RUN_SETUP)
+		prefix = NULL; /* setup_perf_directory(); */
+
+	if (use_browser == -1)
+		use_browser = check_browser_config(p->cmd);
+
+	if (use_pager == -1 && p->option & RUN_SETUP)
+		use_pager = check_pager_config(p->cmd);
+	if (use_pager == -1 && p->option & USE_PAGER)
+		use_pager = 1;
+	commit_pager_choice();
+
+	status = p->fn(argc, argv, prefix);
+	exit_browser(status);
+
+	if (status)
+		return status & 0xff;
+
+	/* Somebody closed stdout? */
+	if (fstat(fileno(stdout), &st))
+		return 0;
+	/* Ignore write errors for pipes and sockets.. */
+	if (S_ISFIFO(st.st_mode) || S_ISSOCK(st.st_mode))
+		return 0;
+
+	status = 1;
+	/* Check for ENOSPC and EIO errors.. */
+	if (fflush(stdout)) {
+		fprintf(stderr, "write failure on standard output: %s",
+			strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out;
+	}
+	if (ferror(stdout)) {
+		fprintf(stderr, "unknown write failure on standard output");
+		goto out;
+	}
+	if (fclose(stdout)) {
+		fprintf(stderr, "close failed on standard output: %s",
+			strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out;
+	}
+	status = 0;
+out:
+	return status;
+}
+
+static void handle_internal_command(int argc, const char **argv)
+{
+	const char *cmd = argv[0];
+	unsigned int i;
+	static const char ext[] = STRIP_EXTENSION;
+
+	if (sizeof(ext) > 1) {
+		i = strlen(argv[0]) - strlen(ext);
+		if (i > 0 && !strcmp(argv[0] + i, ext)) {
+			char *argv0 = strdup(argv[0]);
+			argv[0] = cmd = argv0;
+			argv0[i] = '\0';
+		}
+	}
+
+	/* Turn "perf cmd --help" into "perf help cmd" */
+	if (argc > 1 && !strcmp(argv[1], "--help")) {
+		argv[1] = argv[0];
+		argv[0] = cmd = "help";
+	}
+
+	for (i = 0; i < ARRAY_SIZE(commands); i++) {
+		struct cmd_struct *p = commands+i;
+		if (strcmp(p->cmd, cmd))
+			continue;
+		exit(run_builtin(p, argc, argv));
+	}
+}
+
+static void execv_dashed_external(const char **argv)
+{
+	struct strbuf cmd = STRBUF_INIT;
+	const char *tmp;
+	int status;
+
+	strbuf_addf(&cmd, "perf-%s", argv[0]);
+
+	/*
+	 * argv[0] must be the perf command, but the argv array
+	 * belongs to the caller, and may be reused in
+	 * subsequent loop iterations. Save argv[0] and
+	 * restore it on error.
+	 */
+	tmp = argv[0];
+	argv[0] = cmd.buf;
+
+	/*
+	 * if we fail because the command is not found, it is
+	 * OK to return. Otherwise, we just pass along the status code.
+	 */
+	status = run_command_v_opt(argv, 0);
+	if (status != -ERR_RUN_COMMAND_EXEC) {
+		if (IS_RUN_COMMAND_ERR(status))
+			die("unable to run '%s'", argv[0]);
+		exit(-status);
+	}
+	errno = ENOENT; /* as if we called execvp */
+
+	argv[0] = tmp;
+
+	strbuf_release(&cmd);
+}
+
+static int run_argv(int *argcp, const char ***argv)
+{
+	int done_alias = 0;
+
+	while (1) {
+		/* See if it's an internal command */
+		handle_internal_command(*argcp, *argv);
+
+		/* .. then try the external ones */
+		execv_dashed_external(*argv);
+
+		/* It could be an alias -- this works around the insanity
+		 * of overriding "perf log" with "perf show" by having
+		 * alias.log = show
+		 */
+		if (done_alias || !handle_alias(argcp, argv))
+			break;
+		done_alias = 1;
+	}
+
+	return done_alias;
+}
+
+static void pthread__block_sigwinch(void)
+{
+	sigset_t set;
+
+	sigemptyset(&set);
+	sigaddset(&set, SIGWINCH);
+	pthread_sigmask(SIG_BLOCK, &set, NULL);
+}
+
+void pthread__unblock_sigwinch(void)
+{
+	sigset_t set;
+
+	sigemptyset(&set);
+	sigaddset(&set, SIGWINCH);
+	pthread_sigmask(SIG_UNBLOCK, &set, NULL);
+}
+
+int main(int argc, const char **argv)
+{
+	const char *cmd;
+	char sbuf[STRERR_BUFSIZE];
+
+	/* The page_size is placed in util object. */
+	page_size = sysconf(_SC_PAGE_SIZE);
+	cacheline_size = sysconf(_SC_LEVEL1_DCACHE_LINESIZE);
+
+	cmd = perf_extract_argv0_path(argv[0]);
+	if (!cmd)
+		cmd = "perf-help";
+	/* get debugfs mount point from /proc/mounts */
+	perf_debugfs_mount(NULL);
+	/*
+	 * "perf-xxxx" is the same as "perf xxxx", but we obviously:
+	 *
+	 *  - cannot take flags in between the "perf" and the "xxxx".
+	 *  - cannot execute it externally (since it would just do
+	 *    the same thing over again)
+	 *
+	 * So we just directly call the internal command handler, and
+	 * die if that one cannot handle it.
+	 */
+	if (!prefixcmp(cmd, "perf-")) {
+		cmd += 5;
+		argv[0] = cmd;
+		handle_internal_command(argc, argv);
+		fprintf(stderr, "cannot handle %s internally", cmd);
+		goto out;
+	}
+	if (!prefixcmp(cmd, "trace")) {
+#ifdef HAVE_LIBAUDIT_SUPPORT
+		set_buildid_dir(NULL);
+		setup_path();
+		argv[0] = "trace";
+		return cmd_trace(argc, argv, NULL);
+#else
+		fprintf(stderr,
+			"trace command not available: missing audit-libs devel package at build time.\n");
+		goto out;
+#endif
+	}
+	/* Look for flags.. */
+	argv++;
+	argc--;
+	handle_options(&argv, &argc, NULL);
+	commit_pager_choice();
+	set_buildid_dir(NULL);
+
+	if (argc > 0) {
+		if (!prefixcmp(argv[0], "--"))
+			argv[0] += 2;
+	} else {
+		/* The user didn't specify a command; give them help */
+		printf("\n usage: %s\n\n", perf_usage_string);
+		list_common_cmds_help();
+		printf("\n %s\n\n", perf_more_info_string);
+		goto out;
+	}
+	cmd = argv[0];
+
+	test_attr__init();
+
+	/*
+	 * We use PATH to find perf commands, but we prepend some higher
+	 * precedence paths: the "--exec-path" option, the PERF_EXEC_PATH
+	 * environment, and the $(perfexecdir) from the Makefile at build
+	 * time.
+	 */
+	setup_path();
+	/*
+	 * Block SIGWINCH notifications so that the thread that wants it can
+	 * unblock and get syscalls like select interrupted instead of waiting
+	 * forever while the signal goes to some other non interested thread.
+	 */
+	pthread__block_sigwinch();
+
+	while (1) {
+		static int done_help;
+		int was_alias = run_argv(&argc, &argv);
+
+		if (errno != ENOENT)
+			break;
+
+		if (was_alias) {
+			fprintf(stderr, "Expansion of alias '%s' failed; "
+				"'%s' is not a perf-command\n",
+				cmd, argv[0]);
+			goto out;
+		}
+		if (!done_help) {
+			cmd = argv[0] = help_unknown_cmd(cmd);
+			done_help = 1;
+		} else
+			break;
+	}
+
+	fprintf(stderr, "Failed to run command '%s': %s\n",
+		cmd, strerror_r(errno, sbuf, sizeof(sbuf)));
+out:
+	return 1;
+}
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
new file mode 100644
index 000000000..e14bb6372
--- /dev/null
+++ b/tools/perf/perf.h
@@ -0,0 +1,72 @@
+#ifndef _PERF_PERF_H
+#define _PERF_PERF_H
+
+#include <time.h>
+#include <stdbool.h>
+#include <linux/types.h>
+#include <linux/perf_event.h>
+
+extern bool test_attr__enabled;
+void test_attr__init(void);
+void test_attr__open(struct perf_event_attr *attr, pid_t pid, int cpu,
+		     int fd, int group_fd, unsigned long flags);
+
+#define HAVE_ATTR_TEST
+#include "perf-sys.h"
+
+#ifndef NSEC_PER_SEC
+# define NSEC_PER_SEC			1000000000ULL
+#endif
+#ifndef NSEC_PER_USEC
+# define NSEC_PER_USEC			1000ULL
+#endif
+
+static inline unsigned long long rdclock(void)
+{
+	struct timespec ts;
+
+	clock_gettime(CLOCK_MONOTONIC, &ts);
+	return ts.tv_sec * 1000000000ULL + ts.tv_nsec;
+}
+
+#define MAX_NR_CPUS			1024
+
+extern const char *input_name;
+extern bool perf_host, perf_guest;
+extern const char perf_version_string[];
+
+void pthread__unblock_sigwinch(void);
+
+#include "util/target.h"
+
+struct record_opts {
+	struct target target;
+	bool	     group;
+	bool	     inherit_stat;
+	bool	     no_buffering;
+	bool	     no_inherit;
+	bool	     no_inherit_set;
+	bool	     no_samples;
+	bool	     raw_samples;
+	bool	     sample_address;
+	bool	     sample_weight;
+	bool	     sample_time;
+	bool	     period;
+	bool	     sample_intr_regs;
+	bool	     running_time;
+	unsigned int freq;
+	unsigned int mmap_pages;
+	unsigned int user_freq;
+	u64          branch_stack;
+	u64	     default_interval;
+	u64	     user_interval;
+	bool	     sample_transaction;
+	unsigned     initial_delay;
+	bool         use_clockid;
+	clockid_t    clockid;
+};
+
+struct option;
+extern const char * const *record_usage;
+extern struct option *record_options;
+#endif
diff --git a/tools/perf/python/twatch.py b/tools/perf/python/twatch.py
new file mode 100755
index 000000000..2225162ee
--- /dev/null
+++ b/tools/perf/python/twatch.py
@@ -0,0 +1,41 @@
+#! /usr/bin/python
+# -*- python -*-
+# -*- coding: utf-8 -*-
+#   twatch - Experimental use of the perf python interface
+#   Copyright (C) 2011 Arnaldo Carvalho de Melo <acme@redhat.com>
+#
+#   This application is free software; you can redistribute it and/or
+#   modify it under the terms of the GNU General Public License
+#   as published by the Free Software Foundation; version 2.
+#
+#   This application is distributed in the hope that it will be useful,
+#   but WITHOUT ANY WARRANTY; without even the implied warranty of
+#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+#   General Public License for more details.
+
+import perf
+
+def main():
+	cpus = perf.cpu_map()
+	threads = perf.thread_map()
+	evsel = perf.evsel(task = 1, comm = 1, mmap = 0,
+			   wakeup_events = 1, watermark = 1,
+			   sample_id_all = 1,
+			   sample_type = perf.SAMPLE_PERIOD | perf.SAMPLE_TID | perf.SAMPLE_CPU)
+	evsel.open(cpus = cpus, threads = threads);
+	evlist = perf.evlist(cpus, threads)
+	evlist.add(evsel)
+	evlist.mmap()
+	while True:
+		evlist.poll(timeout = -1)
+		for cpu in cpus:
+			event = evlist.read_on_cpu(cpu)
+			if not event:
+				continue
+			print "cpu: %2d, pid: %4d, tid: %4d" % (event.sample_cpu,
+								event.sample_pid,
+								event.sample_tid),
+			print event
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/perf/scripts/Build b/tools/perf/scripts/Build
new file mode 100644
index 000000000..41efd7e36
--- /dev/null
+++ b/tools/perf/scripts/Build
@@ -0,0 +1,2 @@
+libperf-$(CONFIG_LIBPERL)   += perl/Perf-Trace-Util/
+libperf-$(CONFIG_LIBPYTHON) += python/Perf-Trace-Util/
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Build b/tools/perf/scripts/perl/Perf-Trace-Util/Build
new file mode 100644
index 000000000..928e11017
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/Build
@@ -0,0 +1,3 @@
+libperf-y += Context.o
+
+CFLAGS_Context.o += $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs -Wno-undef -Wno-switch-default
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Context.c b/tools/perf/scripts/perl/Perf-Trace-Util/Context.c
new file mode 100644
index 000000000..28431d1bb
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/Context.c
@@ -0,0 +1,138 @@
+/*
+ * This file was generated automatically by ExtUtils::ParseXS version 2.18_02 from the
+ * contents of Context.xs. Do not edit this file, edit Context.xs instead.
+ *
+ *	ANY CHANGES MADE HERE WILL BE LOST! 
+ *
+ */
+#include <stdbool.h>
+#ifndef HAS_BOOL
+# define HAS_BOOL 1
+#endif
+#line 1 "Context.xs"
+/*
+ * Context.xs.  XS interfaces for perf script.
+ *
+ * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include "EXTERN.h"
+#include "perl.h"
+#include "XSUB.h"
+#include "../../../perf.h"
+#include "../../../util/trace-event.h"
+
+#ifndef PERL_UNUSED_VAR
+#  define PERL_UNUSED_VAR(var) if (0) var = var
+#endif
+
+#line 42 "Context.c"
+
+XS(XS_Perf__Trace__Context_common_pc); /* prototype to pass -Wmissing-prototypes */
+XS(XS_Perf__Trace__Context_common_pc)
+{
+#ifdef dVAR
+    dVAR; dXSARGS;
+#else
+    dXSARGS;
+#endif
+    if (items != 1)
+       Perl_croak(aTHX_ "Usage: %s(%s)", "Perf::Trace::Context::common_pc", "context");
+    PERL_UNUSED_VAR(cv); /* -W */
+    {
+	struct scripting_context *	context = INT2PTR(struct scripting_context *,SvIV(ST(0)));
+	int	RETVAL;
+	dXSTARG;
+
+	RETVAL = common_pc(context);
+	XSprePUSH; PUSHi((IV)RETVAL);
+    }
+    XSRETURN(1);
+}
+
+
+XS(XS_Perf__Trace__Context_common_flags); /* prototype to pass -Wmissing-prototypes */
+XS(XS_Perf__Trace__Context_common_flags)
+{
+#ifdef dVAR
+    dVAR; dXSARGS;
+#else
+    dXSARGS;
+#endif
+    if (items != 1)
+       Perl_croak(aTHX_ "Usage: %s(%s)", "Perf::Trace::Context::common_flags", "context");
+    PERL_UNUSED_VAR(cv); /* -W */
+    {
+	struct scripting_context *	context = INT2PTR(struct scripting_context *,SvIV(ST(0)));
+	int	RETVAL;
+	dXSTARG;
+
+	RETVAL = common_flags(context);
+	XSprePUSH; PUSHi((IV)RETVAL);
+    }
+    XSRETURN(1);
+}
+
+
+XS(XS_Perf__Trace__Context_common_lock_depth); /* prototype to pass -Wmissing-prototypes */
+XS(XS_Perf__Trace__Context_common_lock_depth)
+{
+#ifdef dVAR
+    dVAR; dXSARGS;
+#else
+    dXSARGS;
+#endif
+    if (items != 1)
+       Perl_croak(aTHX_ "Usage: %s(%s)", "Perf::Trace::Context::common_lock_depth", "context");
+    PERL_UNUSED_VAR(cv); /* -W */
+    {
+	struct scripting_context *	context = INT2PTR(struct scripting_context *,SvIV(ST(0)));
+	int	RETVAL;
+	dXSTARG;
+
+	RETVAL = common_lock_depth(context);
+	XSprePUSH; PUSHi((IV)RETVAL);
+    }
+    XSRETURN(1);
+}
+
+#ifdef __cplusplus
+extern "C"
+#endif
+XS(boot_Perf__Trace__Context); /* prototype to pass -Wmissing-prototypes */
+XS(boot_Perf__Trace__Context)
+{
+#ifdef dVAR
+    dVAR; dXSARGS;
+#else
+    dXSARGS;
+#endif
+    const char* file = __FILE__;
+
+    PERL_UNUSED_VAR(cv); /* -W */
+    PERL_UNUSED_VAR(items); /* -W */
+    XS_VERSION_BOOTCHECK ;
+
+        newXSproto("Perf::Trace::Context::common_pc", XS_Perf__Trace__Context_common_pc, file, "$");
+        newXSproto("Perf::Trace::Context::common_flags", XS_Perf__Trace__Context_common_flags, file, "$");
+        newXSproto("Perf::Trace::Context::common_lock_depth", XS_Perf__Trace__Context_common_lock_depth, file, "$");
+    if (PL_unitcheckav)
+         call_list(PL_scopestack_ix, PL_unitcheckav);
+    XSRETURN_YES;
+}
+
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs b/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs
new file mode 100644
index 000000000..8c7ea4244
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs
@@ -0,0 +1,42 @@
+/*
+ * Context.xs.  XS interfaces for perf script.
+ *
+ * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include "EXTERN.h"
+#include "perl.h"
+#include "XSUB.h"
+#include "../../../perf.h"
+#include "../../../util/trace-event.h"
+
+MODULE = Perf::Trace::Context		PACKAGE = Perf::Trace::Context
+PROTOTYPES: ENABLE
+
+int
+common_pc(context)
+	struct scripting_context * context
+
+int
+common_flags(context)
+	struct scripting_context * context
+
+int
+common_lock_depth(context)
+	struct scripting_context * context
+
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Makefile.PL b/tools/perf/scripts/perl/Perf-Trace-Util/Makefile.PL
new file mode 100644
index 000000000..decdeb0f6
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/Makefile.PL
@@ -0,0 +1,17 @@
+use 5.010000;
+use ExtUtils::MakeMaker;
+# See lib/ExtUtils/MakeMaker.pm for details of how to influence
+# the contents of the Makefile that is written.
+WriteMakefile(
+    NAME              => 'Perf::Trace::Context',
+    VERSION_FROM      => 'lib/Perf/Trace/Context.pm', # finds $VERSION
+    PREREQ_PM         => {}, # e.g., Module::Name => 1.1
+    ($] >= 5.005 ?     ## Add these new keywords supported since 5.005
+      (ABSTRACT_FROM  => 'lib/Perf/Trace/Context.pm', # retrieve abstract from module
+       AUTHOR         => 'Tom Zanussi <tzanussi@gmail.com>') : ()),
+    LIBS              => [''], # e.g., '-lm'
+    DEFINE            => '-I ../..', # e.g., '-DHAVE_SOMETHING'
+    INC               => '-I.', # e.g., '-I. -I/usr/include/other'
+	# Un-comment this if you add C files to link with later:
+    OBJECT            => 'Context.o', # link all the C files too
+);
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/README b/tools/perf/scripts/perl/Perf-Trace-Util/README
new file mode 100644
index 000000000..2f0c7f304
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/README
@@ -0,0 +1,59 @@
+Perf-Trace-Util version 0.01
+============================
+
+This module contains utility functions for use with perf script.
+
+Core.pm and Util.pm are pure Perl modules; Core.pm contains routines
+that the core perf support for Perl calls on and should always be
+'used', while Util.pm contains useful but optional utility functions
+that scripts may want to use.  Context.pm contains the Perl->C
+interface that allows scripts to access data in the embedding perf
+executable; scripts wishing to do that should 'use Context.pm'.
+
+The Perl->C perf interface is completely driven by Context.xs.  If you
+want to add new Perl functions that end up accessing C data in the
+perf executable, you add desciptions of the new functions here.
+scripting_context is a pointer to the perf data in the perf executable
+that you want to access - it's passed as the second parameter,
+$context, to all handler functions.
+
+After you do that:
+
+  perl Makefile.PL   # to create a Makefile for the next step
+  make               # to create Context.c
+
+  edit Context.c to add const to the char* file = __FILE__ line in
+  XS(boot_Perf__Trace__Context) to silence a warning/error.
+
+  You can delete the Makefile, object files and anything else that was
+  generated e.g. blib and shared library, etc, except for of course
+  Context.c
+
+  You should then be able to run the normal perf make as usual.
+
+INSTALLATION
+
+Building perf with perf script Perl scripting should install this
+module in the right place.
+
+You should make sure libperl and ExtUtils/Embed.pm are installed first
+e.g. apt-get install libperl-dev or yum install perl-ExtUtils-Embed.
+
+DEPENDENCIES
+
+This module requires these other modules and libraries:
+
+  None
+
+COPYRIGHT AND LICENCE
+
+Copyright (C) 2009 by Tom Zanussi <tzanussi@gmail.com>
+
+This library is free software; you can redistribute it and/or modify
+it under the same terms as Perl itself, either Perl version 5.10.0 or,
+at your option, any later version of Perl 5 you may have available.
+
+Alternatively, this software may be distributed under the terms of the
+GNU General Public License ("GPL") version 2 as published by the Free
+Software Foundation.
+
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm
new file mode 100644
index 000000000..4e2f6039a
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm
@@ -0,0 +1,55 @@
+package Perf::Trace::Context;
+
+use 5.010000;
+use strict;
+use warnings;
+
+require Exporter;
+
+our @ISA = qw(Exporter);
+
+our %EXPORT_TAGS = ( 'all' => [ qw(
+) ] );
+
+our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } );
+
+our @EXPORT = qw(
+	common_pc common_flags common_lock_depth
+);
+
+our $VERSION = '0.01';
+
+require XSLoader;
+XSLoader::load('Perf::Trace::Context', $VERSION);
+
+1;
+__END__
+=head1 NAME
+
+Perf::Trace::Context - Perl extension for accessing functions in perf.
+
+=head1 SYNOPSIS
+
+  use Perf::Trace::Context;
+
+=head1 SEE ALSO
+
+Perf (script) documentation
+
+=head1 AUTHOR
+
+Tom Zanussi, E<lt>tzanussi@gmail.com<gt>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (C) 2009 by Tom Zanussi
+
+This library is free software; you can redistribute it and/or modify
+it under the same terms as Perl itself, either Perl version 5.10.0 or,
+at your option, any later version of Perl 5 you may have available.
+
+Alternatively, this software may be distributed under the terms of the
+GNU General Public License ("GPL") version 2 as published by the Free
+Software Foundation.
+
+=cut
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm
new file mode 100644
index 000000000..9158458d3
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm
@@ -0,0 +1,192 @@
+package Perf::Trace::Core;
+
+use 5.010000;
+use strict;
+use warnings;
+
+require Exporter;
+
+our @ISA = qw(Exporter);
+
+our %EXPORT_TAGS = ( 'all' => [ qw(
+) ] );
+
+our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } );
+
+our @EXPORT = qw(
+define_flag_field define_flag_value flag_str dump_flag_fields
+define_symbolic_field define_symbolic_value symbol_str dump_symbolic_fields
+trace_flag_str
+);
+
+our $VERSION = '0.01';
+
+my %trace_flags = (0x00 => "NONE",
+		   0x01 => "IRQS_OFF",
+		   0x02 => "IRQS_NOSUPPORT",
+		   0x04 => "NEED_RESCHED",
+		   0x08 => "HARDIRQ",
+		   0x10 => "SOFTIRQ");
+
+sub trace_flag_str
+{
+    my ($value) = @_;
+
+    my $string;
+
+    my $print_delim = 0;
+
+    foreach my $idx (sort {$a <=> $b} keys %trace_flags) {
+	if (!$value && !$idx) {
+	    $string .= "NONE";
+	    last;
+	}
+
+	if ($idx && ($value & $idx) == $idx) {
+	    if ($print_delim) {
+		$string .= " | ";
+	    }
+	    $string .= "$trace_flags{$idx}";
+	    $print_delim = 1;
+	    $value &= ~$idx;
+	}
+    }
+
+    return $string;
+}
+
+my %flag_fields;
+my %symbolic_fields;
+
+sub flag_str
+{
+    my ($event_name, $field_name, $value) = @_;
+
+    my $string;
+
+    if ($flag_fields{$event_name}{$field_name}) {
+	my $print_delim = 0;
+	foreach my $idx (sort {$a <=> $b} keys %{$flag_fields{$event_name}{$field_name}{"values"}}) {
+	    if (!$value && !$idx) {
+		$string .= "$flag_fields{$event_name}{$field_name}{'values'}{$idx}";
+		last;
+	    }
+	    if ($idx && ($value & $idx) == $idx) {
+		if ($print_delim && $flag_fields{$event_name}{$field_name}{'delim'}) {
+		    $string .= " $flag_fields{$event_name}{$field_name}{'delim'} ";
+		}
+		$string .= "$flag_fields{$event_name}{$field_name}{'values'}{$idx}";
+		$print_delim = 1;
+		$value &= ~$idx;
+	    }
+	}
+    }
+
+    return $string;
+}
+
+sub define_flag_field
+{
+    my ($event_name, $field_name, $delim) = @_;
+
+    $flag_fields{$event_name}{$field_name}{"delim"} = $delim;
+}
+
+sub define_flag_value
+{
+    my ($event_name, $field_name, $value, $field_str) = @_;
+
+    $flag_fields{$event_name}{$field_name}{"values"}{$value} = $field_str;
+}
+
+sub dump_flag_fields
+{
+    for my $event (keys %flag_fields) {
+	print "event $event:\n";
+	for my $field (keys %{$flag_fields{$event}}) {
+	    print "    field: $field:\n";
+	    print "        delim: $flag_fields{$event}{$field}{'delim'}\n";
+	    foreach my $idx (sort {$a <=> $b} keys %{$flag_fields{$event}{$field}{"values"}}) {
+		print "        value $idx: $flag_fields{$event}{$field}{'values'}{$idx}\n";
+	    }
+	}
+    }
+}
+
+sub symbol_str
+{
+    my ($event_name, $field_name, $value) = @_;
+
+    if ($symbolic_fields{$event_name}{$field_name}) {
+	foreach my $idx (sort {$a <=> $b} keys %{$symbolic_fields{$event_name}{$field_name}{"values"}}) {
+	    if (!$value && !$idx) {
+		return "$symbolic_fields{$event_name}{$field_name}{'values'}{$idx}";
+		last;
+	    }
+	    if ($value == $idx) {
+		return "$symbolic_fields{$event_name}{$field_name}{'values'}{$idx}";
+	    }
+	}
+    }
+
+    return undef;
+}
+
+sub define_symbolic_field
+{
+    my ($event_name, $field_name) = @_;
+
+    # nothing to do, really
+}
+
+sub define_symbolic_value
+{
+    my ($event_name, $field_name, $value, $field_str) = @_;
+
+    $symbolic_fields{$event_name}{$field_name}{"values"}{$value} = $field_str;
+}
+
+sub dump_symbolic_fields
+{
+    for my $event (keys %symbolic_fields) {
+	print "event $event:\n";
+	for my $field (keys %{$symbolic_fields{$event}}) {
+	    print "    field: $field:\n";
+	    foreach my $idx (sort {$a <=> $b} keys %{$symbolic_fields{$event}{$field}{"values"}}) {
+		print "        value $idx: $symbolic_fields{$event}{$field}{'values'}{$idx}\n";
+	    }
+	}
+    }
+}
+
+1;
+__END__
+=head1 NAME
+
+Perf::Trace::Core - Perl extension for perf script
+
+=head1 SYNOPSIS
+
+  use Perf::Trace::Core
+
+=head1 SEE ALSO
+
+Perf (script) documentation
+
+=head1 AUTHOR
+
+Tom Zanussi, E<lt>tzanussi@gmail.com<gt>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (C) 2009 by Tom Zanussi
+
+This library is free software; you can redistribute it and/or modify
+it under the same terms as Perl itself, either Perl version 5.10.0 or,
+at your option, any later version of Perl 5 you may have available.
+
+Alternatively, this software may be distributed under the terms of the
+GNU General Public License ("GPL") version 2 as published by the Free
+Software Foundation.
+
+=cut
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm
new file mode 100644
index 000000000..053500114
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm
@@ -0,0 +1,94 @@
+package Perf::Trace::Util;
+
+use 5.010000;
+use strict;
+use warnings;
+
+require Exporter;
+
+our @ISA = qw(Exporter);
+
+our %EXPORT_TAGS = ( 'all' => [ qw(
+) ] );
+
+our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } );
+
+our @EXPORT = qw(
+avg nsecs nsecs_secs nsecs_nsecs nsecs_usecs print_nsecs
+clear_term
+);
+
+our $VERSION = '0.01';
+
+sub avg
+{
+    my ($total, $n) = @_;
+
+    return $total / $n;
+}
+
+my $NSECS_PER_SEC    = 1000000000;
+
+sub nsecs
+{
+    my ($secs, $nsecs) = @_;
+
+    return $secs * $NSECS_PER_SEC + $nsecs;
+}
+
+sub nsecs_secs {
+    my ($nsecs) = @_;
+
+    return $nsecs / $NSECS_PER_SEC;
+}
+
+sub nsecs_nsecs {
+    my ($nsecs) = @_;
+
+    return $nsecs % $NSECS_PER_SEC;
+}
+
+sub nsecs_str {
+    my ($nsecs) = @_;
+
+    my $str = sprintf("%5u.%09u", nsecs_secs($nsecs), nsecs_nsecs($nsecs));
+
+    return $str;
+}
+
+sub clear_term
+{
+    print "\x1b[H\x1b[2J";
+}
+
+1;
+__END__
+=head1 NAME
+
+Perf::Trace::Util - Perl extension for perf script
+
+=head1 SYNOPSIS
+
+  use Perf::Trace::Util;
+
+=head1 SEE ALSO
+
+Perf (script) documentation
+
+=head1 AUTHOR
+
+Tom Zanussi, E<lt>tzanussi@gmail.com<gt>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (C) 2009 by Tom Zanussi
+
+This library is free software; you can redistribute it and/or modify
+it under the same terms as Perl itself, either Perl version 5.10.0 or,
+at your option, any later version of Perl 5 you may have available.
+
+Alternatively, this software may be distributed under the terms of the
+GNU General Public License ("GPL") version 2 as published by the Free
+Software Foundation.
+
+=cut
diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/typemap b/tools/perf/scripts/perl/Perf-Trace-Util/typemap
new file mode 100644
index 000000000..840836804
--- /dev/null
+++ b/tools/perf/scripts/perl/Perf-Trace-Util/typemap
@@ -0,0 +1 @@
+struct scripting_context * T_PTR
diff --git a/tools/perf/scripts/perl/bin/check-perf-trace-record b/tools/perf/scripts/perl/bin/check-perf-trace-record
new file mode 100644
index 000000000..423ad6aed
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/check-perf-trace-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -a -e kmem:kmalloc -e irq:softirq_entry -e kmem:kfree
diff --git a/tools/perf/scripts/perl/bin/failed-syscalls-record b/tools/perf/scripts/perl/bin/failed-syscalls-record
new file mode 100644
index 000000000..74685f318
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/failed-syscalls-record
@@ -0,0 +1,3 @@
+#!/bin/bash
+(perf record -e raw_syscalls:sys_exit $@ || \
+ perf record -e syscalls:sys_exit $@) 2> /dev/null
diff --git a/tools/perf/scripts/perl/bin/failed-syscalls-report b/tools/perf/scripts/perl/bin/failed-syscalls-report
new file mode 100644
index 000000000..9f83cc1ad
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/failed-syscalls-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: system-wide failed syscalls
+# args: [comm]
+if [ $# -gt 0 ] ; then
+    if ! expr match "$1" "-" > /dev/null ; then
+	comm=$1
+	shift
+    fi
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/failed-syscalls.pl $comm
diff --git a/tools/perf/scripts/perl/bin/rw-by-file-record b/tools/perf/scripts/perl/bin/rw-by-file-record
new file mode 100644
index 000000000..33efc8673
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rw-by-file-record
@@ -0,0 +1,3 @@
+#!/bin/bash
+perf record -e syscalls:sys_enter_read -e syscalls:sys_enter_write $@
+
diff --git a/tools/perf/scripts/perl/bin/rw-by-file-report b/tools/perf/scripts/perl/bin/rw-by-file-report
new file mode 100644
index 000000000..77200b3f3
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rw-by-file-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: r/w activity for a program, by file
+# args: <comm>
+if [ $# -lt 1 ] ; then
+    echo "usage: rw-by-file <comm>"
+    exit
+fi
+comm=$1
+shift
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-file.pl $comm
diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-record b/tools/perf/scripts/perl/bin/rw-by-pid-record
new file mode 100644
index 000000000..7cb9db230
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rw-by-pid-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@
diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-report b/tools/perf/scripts/perl/bin/rw-by-pid-report
new file mode 100644
index 000000000..a27b9f311
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rw-by-pid-report
@@ -0,0 +1,3 @@
+#!/bin/bash
+# description: system-wide r/w activity
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-pid.pl
diff --git a/tools/perf/scripts/perl/bin/rwtop-record b/tools/perf/scripts/perl/bin/rwtop-record
new file mode 100644
index 000000000..7cb9db230
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rwtop-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e syscalls:sys_enter_read -e syscalls:sys_exit_read -e syscalls:sys_enter_write -e syscalls:sys_exit_write $@
diff --git a/tools/perf/scripts/perl/bin/rwtop-report b/tools/perf/scripts/perl/bin/rwtop-report
new file mode 100644
index 000000000..83e11ec2e
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/rwtop-report
@@ -0,0 +1,20 @@
+#!/bin/bash
+# description: system-wide r/w top
+# args: [interval]
+n_args=0
+for i in "$@"
+do
+    if expr match "$i" "-" > /dev/null ; then
+	break
+    fi
+    n_args=$(( $n_args + 1 ))
+done
+if [ "$n_args" -gt 1 ] ; then
+    echo "usage: rwtop-report [interval]"
+    exit
+fi
+if [ "$n_args" -gt 0 ] ; then
+    interval=$1
+    shift
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rwtop.pl $interval
diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-record b/tools/perf/scripts/perl/bin/wakeup-latency-record
new file mode 100644
index 000000000..464251a1b
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/wakeup-latency-record
@@ -0,0 +1,6 @@
+#!/bin/bash
+perf record -e sched:sched_switch -e sched:sched_wakeup $@
+
+
+
+
diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-report b/tools/perf/scripts/perl/bin/wakeup-latency-report
new file mode 100644
index 000000000..889e8130c
--- /dev/null
+++ b/tools/perf/scripts/perl/bin/wakeup-latency-report
@@ -0,0 +1,3 @@
+#!/bin/bash
+# description: system-wide min/max/avg wakeup latency
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/wakeup-latency.pl
diff --git a/tools/perf/scripts/perl/check-perf-trace.pl b/tools/perf/scripts/perl/check-perf-trace.pl
new file mode 100644
index 000000000..4e7076c20
--- /dev/null
+++ b/tools/perf/scripts/perl/check-perf-trace.pl
@@ -0,0 +1,106 @@
+# perf script event handlers, generated by perf script -g perl
+# (c) 2009, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# This script tests basic functionality such as flag and symbol
+# strings, common_xxx() calls back into perf, begin, end, unhandled
+# events, etc.  Basically, if this script runs successfully and
+# displays expected results, perl scripting support should be ok.
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Context;
+use Perf::Trace::Util;
+
+sub trace_begin
+{
+    print "trace_begin\n";
+}
+
+sub trace_end
+{
+    print "trace_end\n";
+
+    print_unhandled();
+}
+
+sub irq::softirq_entry
+{
+	my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	    $common_pid, $common_comm,
+	    $vec) = @_;
+
+	print_header($event_name, $common_cpu, $common_secs, $common_nsecs,
+		     $common_pid, $common_comm);
+
+	print_uncommon($context);
+
+	printf("vec=%s\n",
+	       symbol_str("irq::softirq_entry", "vec", $vec));
+}
+
+sub kmem::kmalloc
+{
+	my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	    $common_pid, $common_comm,
+	    $call_site, $ptr, $bytes_req, $bytes_alloc,
+	    $gfp_flags) = @_;
+
+	print_header($event_name, $common_cpu, $common_secs, $common_nsecs,
+		     $common_pid, $common_comm);
+
+	print_uncommon($context);
+
+	printf("call_site=%p, ptr=%p, bytes_req=%u, bytes_alloc=%u, ".
+	       "gfp_flags=%s\n",
+	       $call_site, $ptr, $bytes_req, $bytes_alloc,
+
+	       flag_str("kmem::kmalloc", "gfp_flags", $gfp_flags));
+}
+
+# print trace fields not included in handler args
+sub print_uncommon
+{
+    my ($context) = @_;
+
+    printf("common_preempt_count=%d, common_flags=%s, common_lock_depth=%d, ",
+	   common_pc($context), trace_flag_str(common_flags($context)),
+	   common_lock_depth($context));
+
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+	return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+	   "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+	printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
+
+sub print_header
+{
+	my ($event_name, $cpu, $secs, $nsecs, $pid, $comm) = @_;
+
+	printf("%-20s %5u %05u.%09u %8u %-20s ",
+	       $event_name, $cpu, $secs, $nsecs, $pid, $comm);
+}
diff --git a/tools/perf/scripts/perl/failed-syscalls.pl b/tools/perf/scripts/perl/failed-syscalls.pl
new file mode 100644
index 000000000..55e7ae4c5
--- /dev/null
+++ b/tools/perf/scripts/perl/failed-syscalls.pl
@@ -0,0 +1,47 @@
+# failed system call counts
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Displays system-wide failed system call totals
+# If a [comm] arg is specified, only syscalls called by [comm] are displayed.
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Context;
+use Perf::Trace::Util;
+
+my $for_comm = shift;
+
+my %failed_syscalls;
+
+sub raw_syscalls::sys_exit
+{
+	my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	    $common_pid, $common_comm,
+	    $id, $ret) = @_;
+
+	if ($ret < 0) {
+	    $failed_syscalls{$common_comm}++;
+	}
+}
+
+sub syscalls::sys_exit
+{
+	raw_syscalls::sys_exit(@_)
+}
+
+sub trace_end
+{
+    printf("\nfailed syscalls by comm:\n\n");
+
+    printf("%-20s  %10s\n", "comm", "# errors");
+    printf("%-20s  %6s  %10s\n", "--------------------", "----------");
+
+    foreach my $comm (sort {$failed_syscalls{$b} <=> $failed_syscalls{$a}}
+		      keys %failed_syscalls) {
+	next if ($for_comm && $comm ne $for_comm);
+
+	printf("%-20s  %10s\n", $comm, $failed_syscalls{$comm});
+    }
+}
diff --git a/tools/perf/scripts/perl/rw-by-file.pl b/tools/perf/scripts/perl/rw-by-file.pl
new file mode 100644
index 000000000..74844ee2b
--- /dev/null
+++ b/tools/perf/scripts/perl/rw-by-file.pl
@@ -0,0 +1,106 @@
+#!/usr/bin/perl -w
+# (c) 2009, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# Display r/w activity for files read/written to for a given program
+
+# The common_* event handler fields are the most useful fields common to
+# all events.  They don't necessarily correspond to the 'common_*' fields
+# in the status files.  Those fields not available as handler params can
+# be retrieved via script functions of the form get_common_*().
+
+use 5.010000;
+use strict;
+use warnings;
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Util;
+
+my $usage = "perf script -s rw-by-file.pl <comm>\n";
+
+my $for_comm = shift or die $usage;
+
+my %reads;
+my %writes;
+
+sub syscalls::sys_enter_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm, $nr, $fd, $buf, $count) = @_;
+
+    if ($common_comm eq $for_comm) {
+	$reads{$fd}{bytes_requested} += $count;
+	$reads{$fd}{total_reads}++;
+    }
+}
+
+sub syscalls::sys_enter_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm, $nr, $fd, $buf, $count) = @_;
+
+    if ($common_comm eq $for_comm) {
+	$writes{$fd}{bytes_written} += $count;
+	$writes{$fd}{total_writes}++;
+    }
+}
+
+sub trace_end
+{
+    printf("file read counts for $for_comm:\n\n");
+
+    printf("%6s  %10s  %10s\n", "fd", "# reads", "bytes_requested");
+    printf("%6s  %10s  %10s\n", "------", "----------", "-----------");
+
+    foreach my $fd (sort {$reads{$b}{bytes_requested} <=>
+			      $reads{$a}{bytes_requested}} keys %reads) {
+	my $total_reads = $reads{$fd}{total_reads};
+	my $bytes_requested = $reads{$fd}{bytes_requested};
+	printf("%6u  %10u  %10u\n", $fd, $total_reads, $bytes_requested);
+    }
+
+    printf("\nfile write counts for $for_comm:\n\n");
+
+    printf("%6s  %10s  %10s\n", "fd", "# writes", "bytes_written");
+    printf("%6s  %10s  %10s\n", "------", "----------", "-----------");
+
+    foreach my $fd (sort {$writes{$b}{bytes_written} <=>
+			      $writes{$a}{bytes_written}} keys %writes) {
+	my $total_writes = $writes{$fd}{total_writes};
+	my $bytes_written = $writes{$fd}{bytes_written};
+	printf("%6u  %10u  %10u\n", $fd, $total_writes, $bytes_written);
+    }
+
+    print_unhandled();
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+	return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+	   "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+	printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
+
+
diff --git a/tools/perf/scripts/perl/rw-by-pid.pl b/tools/perf/scripts/perl/rw-by-pid.pl
new file mode 100644
index 000000000..9db23c9da
--- /dev/null
+++ b/tools/perf/scripts/perl/rw-by-pid.pl
@@ -0,0 +1,184 @@
+#!/usr/bin/perl -w
+# (c) 2009, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# Display r/w activity for all processes
+
+# The common_* event handler fields are the most useful fields common to
+# all events.  They don't necessarily correspond to the 'common_*' fields
+# in the status files.  Those fields not available as handler params can
+# be retrieved via script functions of the form get_common_*().
+
+use 5.010000;
+use strict;
+use warnings;
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Util;
+
+my %reads;
+my %writes;
+
+sub syscalls::sys_exit_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$nr, $ret) = @_;
+
+    if ($ret > 0) {
+	$reads{$common_pid}{bytes_read} += $ret;
+    } else {
+	if (!defined ($reads{$common_pid}{bytes_read})) {
+	    $reads{$common_pid}{bytes_read} = 0;
+	}
+	$reads{$common_pid}{errors}{$ret}++;
+    }
+}
+
+sub syscalls::sys_enter_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$nr, $fd, $buf, $count) = @_;
+
+    $reads{$common_pid}{bytes_requested} += $count;
+    $reads{$common_pid}{total_reads}++;
+    $reads{$common_pid}{comm} = $common_comm;
+}
+
+sub syscalls::sys_exit_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$nr, $ret) = @_;
+
+    if ($ret <= 0) {
+	$writes{$common_pid}{errors}{$ret}++;
+    }
+}
+
+sub syscalls::sys_enter_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$nr, $fd, $buf, $count) = @_;
+
+    $writes{$common_pid}{bytes_written} += $count;
+    $writes{$common_pid}{total_writes}++;
+    $writes{$common_pid}{comm} = $common_comm;
+}
+
+sub trace_end
+{
+    printf("read counts by pid:\n\n");
+
+    printf("%6s  %20s  %10s  %10s  %10s\n", "pid", "comm",
+	   "# reads", "bytes_requested", "bytes_read");
+    printf("%6s  %-20s  %10s  %10s  %10s\n", "------", "--------------------",
+	   "-----------", "----------", "----------");
+
+    foreach my $pid (sort { ($reads{$b}{bytes_read} || 0) <=>
+				($reads{$a}{bytes_read} || 0) } keys %reads) {
+	my $comm = $reads{$pid}{comm} || "";
+	my $total_reads = $reads{$pid}{total_reads} || 0;
+	my $bytes_requested = $reads{$pid}{bytes_requested} || 0;
+	my $bytes_read = $reads{$pid}{bytes_read} || 0;
+
+	printf("%6s  %-20s  %10s  %10s  %10s\n", $pid, $comm,
+	       $total_reads, $bytes_requested, $bytes_read);
+    }
+
+    printf("\nfailed reads by pid:\n\n");
+
+    printf("%6s  %20s  %6s  %10s\n", "pid", "comm", "error #", "# errors");
+    printf("%6s  %20s  %6s  %10s\n", "------", "--------------------",
+	   "------", "----------");
+
+    my @errcounts = ();
+
+    foreach my $pid (keys %reads) {
+	foreach my $error (keys %{$reads{$pid}{errors}}) {
+	    my $comm = $reads{$pid}{comm} || "";
+	    my $errcount = $reads{$pid}{errors}{$error} || 0;
+	    push @errcounts, [$pid, $comm, $error, $errcount];
+	}
+    }
+
+    @errcounts = sort { $b->[3] <=> $a->[3] } @errcounts;
+
+    for my $i (0 .. $#errcounts) {
+	printf("%6d  %-20s  %6d  %10s\n", $errcounts[$i][0],
+	       $errcounts[$i][1], $errcounts[$i][2], $errcounts[$i][3]);
+    }
+
+    printf("\nwrite counts by pid:\n\n");
+
+    printf("%6s  %20s  %10s  %10s\n", "pid", "comm",
+	   "# writes", "bytes_written");
+    printf("%6s  %-20s  %10s  %10s\n", "------", "--------------------",
+	   "-----------", "----------");
+
+    foreach my $pid (sort { ($writes{$b}{bytes_written} || 0) <=>
+			($writes{$a}{bytes_written} || 0)} keys %writes) {
+	my $comm = $writes{$pid}{comm} || "";
+	my $total_writes = $writes{$pid}{total_writes} || 0;
+	my $bytes_written = $writes{$pid}{bytes_written} || 0;
+
+	printf("%6s  %-20s  %10s  %10s\n", $pid, $comm,
+	       $total_writes, $bytes_written);
+    }
+
+    printf("\nfailed writes by pid:\n\n");
+
+    printf("%6s  %20s  %6s  %10s\n", "pid", "comm", "error #", "# errors");
+    printf("%6s  %20s  %6s  %10s\n", "------", "--------------------",
+	   "------", "----------");
+
+    @errcounts = ();
+
+    foreach my $pid (keys %writes) {
+	foreach my $error (keys %{$writes{$pid}{errors}}) {
+	    my $comm = $writes{$pid}{comm} || "";
+	    my $errcount = $writes{$pid}{errors}{$error} || 0;
+	    push @errcounts, [$pid, $comm, $error, $errcount];
+	}
+    }
+
+    @errcounts = sort { $b->[3] <=> $a->[3] } @errcounts;
+
+    for my $i (0 .. $#errcounts) {
+	printf("%6d  %-20s  %6d  %10s\n", $errcounts[$i][0],
+	       $errcounts[$i][1], $errcounts[$i][2], $errcounts[$i][3]);
+    }
+
+    print_unhandled();
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+	return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+	   "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+	printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
diff --git a/tools/perf/scripts/perl/rwtop.pl b/tools/perf/scripts/perl/rwtop.pl
new file mode 100644
index 000000000..8b2078702
--- /dev/null
+++ b/tools/perf/scripts/perl/rwtop.pl
@@ -0,0 +1,203 @@
+#!/usr/bin/perl -w
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# read/write top
+#
+# Periodically displays system-wide r/w call activity, broken down by
+# pid.  If an [interval] arg is specified, the display will be
+# refreshed every [interval] seconds.  The default interval is 3
+# seconds.
+
+use 5.010000;
+use strict;
+use warnings;
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Util;
+use POSIX qw/SIGALRM SA_RESTART/;
+
+my $default_interval = 3;
+my $nlines = 20;
+my $print_thread;
+my $print_pending = 0;
+
+my %reads;
+my %writes;
+
+my $interval = shift;
+if (!$interval) {
+    $interval = $default_interval;
+}
+
+sub syscalls::sys_exit_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$nr, $ret) = @_;
+
+    print_check();
+
+    if ($ret > 0) {
+	$reads{$common_pid}{bytes_read} += $ret;
+    } else {
+	if (!defined ($reads{$common_pid}{bytes_read})) {
+	    $reads{$common_pid}{bytes_read} = 0;
+	}
+	$reads{$common_pid}{errors}{$ret}++;
+    }
+}
+
+sub syscalls::sys_enter_read
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$nr, $fd, $buf, $count) = @_;
+
+    print_check();
+
+    $reads{$common_pid}{bytes_requested} += $count;
+    $reads{$common_pid}{total_reads}++;
+    $reads{$common_pid}{comm} = $common_comm;
+}
+
+sub syscalls::sys_exit_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$nr, $ret) = @_;
+
+    print_check();
+
+    if ($ret <= 0) {
+	$writes{$common_pid}{errors}{$ret}++;
+    }
+}
+
+sub syscalls::sys_enter_write
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$nr, $fd, $buf, $count) = @_;
+
+    print_check();
+
+    $writes{$common_pid}{bytes_written} += $count;
+    $writes{$common_pid}{total_writes}++;
+    $writes{$common_pid}{comm} = $common_comm;
+}
+
+sub trace_begin
+{
+    my $sa = POSIX::SigAction->new(\&set_print_pending);
+    $sa->flags(SA_RESTART);
+    $sa->safe(1);
+    POSIX::sigaction(SIGALRM, $sa) or die "Can't set SIGALRM handler: $!\n";
+    alarm 1;
+}
+
+sub trace_end
+{
+    print_unhandled();
+    print_totals();
+}
+
+sub print_check()
+{
+    if ($print_pending == 1) {
+	$print_pending = 0;
+	print_totals();
+    }
+}
+
+sub set_print_pending()
+{
+    $print_pending = 1;
+    alarm $interval;
+}
+
+sub print_totals
+{
+    my $count;
+
+    $count = 0;
+
+    clear_term();
+
+    printf("\nread counts by pid:\n\n");
+
+    printf("%6s  %20s  %10s  %10s  %10s\n", "pid", "comm",
+	   "# reads", "bytes_req", "bytes_read");
+    printf("%6s  %-20s  %10s  %10s  %10s\n", "------", "--------------------",
+	   "----------", "----------", "----------");
+
+    foreach my $pid (sort { ($reads{$b}{bytes_read} || 0) <=>
+			       ($reads{$a}{bytes_read} || 0) } keys %reads) {
+	my $comm = $reads{$pid}{comm} || "";
+	my $total_reads = $reads{$pid}{total_reads} || 0;
+	my $bytes_requested = $reads{$pid}{bytes_requested} || 0;
+	my $bytes_read = $reads{$pid}{bytes_read} || 0;
+
+	printf("%6s  %-20s  %10s  %10s  %10s\n", $pid, $comm,
+	       $total_reads, $bytes_requested, $bytes_read);
+
+	if (++$count == $nlines) {
+	    last;
+	}
+    }
+
+    $count = 0;
+
+    printf("\nwrite counts by pid:\n\n");
+
+    printf("%6s  %20s  %10s  %13s\n", "pid", "comm",
+	   "# writes", "bytes_written");
+    printf("%6s  %-20s  %10s  %13s\n", "------", "--------------------",
+	   "----------", "-------------");
+
+    foreach my $pid (sort { ($writes{$b}{bytes_written} || 0) <=>
+			($writes{$a}{bytes_written} || 0)} keys %writes) {
+	my $comm = $writes{$pid}{comm} || "";
+	my $total_writes = $writes{$pid}{total_writes} || 0;
+	my $bytes_written = $writes{$pid}{bytes_written} || 0;
+
+	printf("%6s  %-20s  %10s  %13s\n", $pid, $comm,
+	       $total_writes, $bytes_written);
+
+	if (++$count == $nlines) {
+	    last;
+	}
+    }
+
+    %reads = ();
+    %writes = ();
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+	return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+	   "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+	printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
diff --git a/tools/perf/scripts/perl/wakeup-latency.pl b/tools/perf/scripts/perl/wakeup-latency.pl
new file mode 100644
index 000000000..d9143dcec
--- /dev/null
+++ b/tools/perf/scripts/perl/wakeup-latency.pl
@@ -0,0 +1,107 @@
+#!/usr/bin/perl -w
+# (c) 2009, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+
+# Display avg/min/max wakeup latency
+
+# The common_* event handler fields are the most useful fields common to
+# all events.  They don't necessarily correspond to the 'common_*' fields
+# in the status files.  Those fields not available as handler params can
+# be retrieved via script functions of the form get_common_*().
+
+use 5.010000;
+use strict;
+use warnings;
+
+use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib";
+use lib "./Perf-Trace-Util/lib";
+use Perf::Trace::Core;
+use Perf::Trace::Util;
+
+my %last_wakeup;
+
+my $max_wakeup_latency;
+my $min_wakeup_latency;
+my $total_wakeup_latency = 0;
+my $total_wakeups = 0;
+
+sub sched::sched_switch
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$prev_comm, $prev_pid, $prev_prio, $prev_state, $next_comm, $next_pid,
+	$next_prio) = @_;
+
+    my $wakeup_ts = $last_wakeup{$common_cpu}{ts};
+    if ($wakeup_ts) {
+	my $switch_ts = nsecs($common_secs, $common_nsecs);
+	my $wakeup_latency = $switch_ts - $wakeup_ts;
+	if ($wakeup_latency > $max_wakeup_latency) {
+	    $max_wakeup_latency = $wakeup_latency;
+	}
+	if ($wakeup_latency < $min_wakeup_latency) {
+	    $min_wakeup_latency = $wakeup_latency;
+	}
+	$total_wakeup_latency += $wakeup_latency;
+	$total_wakeups++;
+    }
+    $last_wakeup{$common_cpu}{ts} = 0;
+}
+
+sub sched::sched_wakeup
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm,
+	$comm, $pid, $prio, $success, $target_cpu) = @_;
+
+    $last_wakeup{$target_cpu}{ts} = nsecs($common_secs, $common_nsecs);
+}
+
+sub trace_begin
+{
+    $min_wakeup_latency = 1000000000;
+    $max_wakeup_latency = 0;
+}
+
+sub trace_end
+{
+    printf("wakeup_latency stats:\n\n");
+    print "total_wakeups: $total_wakeups\n";
+    if ($total_wakeups) {
+	printf("avg_wakeup_latency (ns): %u\n",
+	       avg($total_wakeup_latency, $total_wakeups));
+    } else {
+	printf("avg_wakeup_latency (ns): N/A\n");
+    }
+    printf("min_wakeup_latency (ns): %u\n", $min_wakeup_latency);
+    printf("max_wakeup_latency (ns): %u\n", $max_wakeup_latency);
+
+    print_unhandled();
+}
+
+my %unhandled;
+
+sub print_unhandled
+{
+    if ((scalar keys %unhandled) == 0) {
+	return;
+    }
+
+    print "\nunhandled events:\n\n";
+
+    printf("%-40s  %10s\n", "event", "count");
+    printf("%-40s  %10s\n", "----------------------------------------",
+	   "-----------");
+
+    foreach my $event_name (keys %unhandled) {
+	printf("%-40s  %10d\n", $event_name, $unhandled{$event_name});
+    }
+}
+
+sub trace_unhandled
+{
+    my ($event_name, $context, $common_cpu, $common_secs, $common_nsecs,
+	$common_pid, $common_comm) = @_;
+
+    $unhandled{$event_name}++;
+}
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/Build b/tools/perf/scripts/python/Perf-Trace-Util/Build
new file mode 100644
index 000000000..aefc15c94
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/Build
@@ -0,0 +1,3 @@
+libperf-y += Context.o
+
+CFLAGS_Context.o += $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-nested-externs
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/Context.c b/tools/perf/scripts/python/Perf-Trace-Util/Context.c
new file mode 100644
index 000000000..fcd1dd667
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/Context.c
@@ -0,0 +1,88 @@
+/*
+ * Context.c.  Python interfaces for perf script.
+ *
+ * Copyright (C) 2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <Python.h>
+#include "../../../perf.h"
+#include "../../../util/trace-event.h"
+
+PyMODINIT_FUNC initperf_trace_context(void);
+
+static PyObject *perf_trace_context_common_pc(PyObject *obj, PyObject *args)
+{
+	static struct scripting_context *scripting_context;
+	PyObject *context;
+	int retval;
+
+	if (!PyArg_ParseTuple(args, "O", &context))
+		return NULL;
+
+	scripting_context = PyCObject_AsVoidPtr(context);
+	retval = common_pc(scripting_context);
+
+	return Py_BuildValue("i", retval);
+}
+
+static PyObject *perf_trace_context_common_flags(PyObject *obj,
+						 PyObject *args)
+{
+	static struct scripting_context *scripting_context;
+	PyObject *context;
+	int retval;
+
+	if (!PyArg_ParseTuple(args, "O", &context))
+		return NULL;
+
+	scripting_context = PyCObject_AsVoidPtr(context);
+	retval = common_flags(scripting_context);
+
+	return Py_BuildValue("i", retval);
+}
+
+static PyObject *perf_trace_context_common_lock_depth(PyObject *obj,
+						      PyObject *args)
+{
+	static struct scripting_context *scripting_context;
+	PyObject *context;
+	int retval;
+
+	if (!PyArg_ParseTuple(args, "O", &context))
+		return NULL;
+
+	scripting_context = PyCObject_AsVoidPtr(context);
+	retval = common_lock_depth(scripting_context);
+
+	return Py_BuildValue("i", retval);
+}
+
+static PyMethodDef ContextMethods[] = {
+	{ "common_pc", perf_trace_context_common_pc, METH_VARARGS,
+	  "Get the common preempt count event field value."},
+	{ "common_flags", perf_trace_context_common_flags, METH_VARARGS,
+	  "Get the common flags event field value."},
+	{ "common_lock_depth", perf_trace_context_common_lock_depth,
+	  METH_VARARGS,	"Get the common lock depth event field value."},
+	{ NULL, NULL, 0, NULL}
+};
+
+PyMODINIT_FUNC initperf_trace_context(void)
+{
+	(void) Py_InitModule("perf_trace_context", ContextMethods);
+}
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py
new file mode 100644
index 000000000..38dfb720f
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py
@@ -0,0 +1,122 @@
+# Core.py - Python extension for perf script, core functions
+#
+# Copyright (C) 2010 by Tom Zanussi <tzanussi@gmail.com>
+#
+# This software may be distributed under the terms of the GNU General
+# Public License ("GPL") version 2 as published by the Free Software
+# Foundation.
+
+from collections import defaultdict
+
+def autodict():
+    return defaultdict(autodict)
+
+flag_fields = autodict()
+symbolic_fields = autodict()
+
+def define_flag_field(event_name, field_name, delim):
+    flag_fields[event_name][field_name]['delim'] = delim
+
+def define_flag_value(event_name, field_name, value, field_str):
+    flag_fields[event_name][field_name]['values'][value] = field_str
+
+def define_symbolic_field(event_name, field_name):
+    # nothing to do, really
+    pass
+
+def define_symbolic_value(event_name, field_name, value, field_str):
+    symbolic_fields[event_name][field_name]['values'][value] = field_str
+
+def flag_str(event_name, field_name, value):
+    string = ""
+
+    if flag_fields[event_name][field_name]:
+	print_delim = 0
+        keys = flag_fields[event_name][field_name]['values'].keys()
+        keys.sort()
+        for idx in keys:
+            if not value and not idx:
+                string += flag_fields[event_name][field_name]['values'][idx]
+                break
+            if idx and (value & idx) == idx:
+                if print_delim and flag_fields[event_name][field_name]['delim']:
+                    string += " " + flag_fields[event_name][field_name]['delim'] + " "
+                string += flag_fields[event_name][field_name]['values'][idx]
+                print_delim = 1
+                value &= ~idx
+
+    return string
+
+def symbol_str(event_name, field_name, value):
+    string = ""
+
+    if symbolic_fields[event_name][field_name]:
+        keys = symbolic_fields[event_name][field_name]['values'].keys()
+        keys.sort()
+        for idx in keys:
+            if not value and not idx:
+		string = symbolic_fields[event_name][field_name]['values'][idx]
+                break
+	    if (value == idx):
+		string = symbolic_fields[event_name][field_name]['values'][idx]
+                break
+
+    return string
+
+trace_flags = { 0x00: "NONE", \
+                    0x01: "IRQS_OFF", \
+                    0x02: "IRQS_NOSUPPORT", \
+                    0x04: "NEED_RESCHED", \
+                    0x08: "HARDIRQ", \
+                    0x10: "SOFTIRQ" }
+
+def trace_flag_str(value):
+    string = ""
+    print_delim = 0
+
+    keys = trace_flags.keys()
+
+    for idx in keys:
+	if not value and not idx:
+	    string += "NONE"
+	    break
+
+	if idx and (value & idx) == idx:
+	    if print_delim:
+		string += " | ";
+	    string += trace_flags[idx]
+	    print_delim = 1
+	    value &= ~idx
+
+    return string
+
+
+def taskState(state):
+	states = {
+		0 : "R",
+		1 : "S",
+		2 : "D",
+		64: "DEAD"
+	}
+
+	if state not in states:
+		return "Unknown"
+
+	return states[state]
+
+
+class EventHeaders:
+	def __init__(self, common_cpu, common_secs, common_nsecs,
+		     common_pid, common_comm, common_callchain):
+		self.cpu = common_cpu
+		self.secs = common_secs
+		self.nsecs = common_nsecs
+		self.pid = common_pid
+		self.comm = common_comm
+		self.callchain = common_callchain
+
+	def ts(self):
+		return (self.secs * (10 ** 9)) + self.nsecs
+
+	def ts_format(self):
+		return "%d.%d" % (self.secs, int(self.nsecs / 1000))
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/EventClass.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/EventClass.py
new file mode 100755
index 000000000..9e0985794
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/EventClass.py
@@ -0,0 +1,94 @@
+# EventClass.py
+#
+# This is a library defining some events types classes, which could
+# be used by other scripts to analyzing the perf samples.
+#
+# Currently there are just a few classes defined for examples,
+# PerfEvent is the base class for all perf event sample, PebsEvent
+# is a HW base Intel x86 PEBS event, and user could add more SW/HW
+# event classes based on requirements.
+
+import struct
+
+# Event types, user could add more here
+EVTYPE_GENERIC  = 0
+EVTYPE_PEBS     = 1     # Basic PEBS event
+EVTYPE_PEBS_LL  = 2     # PEBS event with load latency info
+EVTYPE_IBS      = 3
+
+#
+# Currently we don't have good way to tell the event type, but by
+# the size of raw buffer, raw PEBS event with load latency data's
+# size is 176 bytes, while the pure PEBS event's size is 144 bytes.
+#
+def create_event(name, comm, dso, symbol, raw_buf):
+        if (len(raw_buf) == 144):
+                event = PebsEvent(name, comm, dso, symbol, raw_buf)
+        elif (len(raw_buf) == 176):
+                event = PebsNHM(name, comm, dso, symbol, raw_buf)
+        else:
+                event = PerfEvent(name, comm, dso, symbol, raw_buf)
+
+        return event
+
+class PerfEvent(object):
+        event_num = 0
+        def __init__(self, name, comm, dso, symbol, raw_buf, ev_type=EVTYPE_GENERIC):
+                self.name       = name
+                self.comm       = comm
+                self.dso        = dso
+                self.symbol     = symbol
+                self.raw_buf    = raw_buf
+                self.ev_type    = ev_type
+                PerfEvent.event_num += 1
+
+        def show(self):
+                print "PMU event: name=%12s, symbol=%24s, comm=%8s, dso=%12s" % (self.name, self.symbol, self.comm, self.dso)
+
+#
+# Basic Intel PEBS (Precise Event-based Sampling) event, whose raw buffer
+# contains the context info when that event happened: the EFLAGS and
+# linear IP info, as well as all the registers.
+#
+class PebsEvent(PerfEvent):
+        pebs_num = 0
+        def __init__(self, name, comm, dso, symbol, raw_buf, ev_type=EVTYPE_PEBS):
+                tmp_buf=raw_buf[0:80]
+                flags, ip, ax, bx, cx, dx, si, di, bp, sp = struct.unpack('QQQQQQQQQQ', tmp_buf)
+                self.flags = flags
+                self.ip    = ip
+                self.ax    = ax
+                self.bx    = bx
+                self.cx    = cx
+                self.dx    = dx
+                self.si    = si
+                self.di    = di
+                self.bp    = bp
+                self.sp    = sp
+
+                PerfEvent.__init__(self, name, comm, dso, symbol, raw_buf, ev_type)
+                PebsEvent.pebs_num += 1
+                del tmp_buf
+
+#
+# Intel Nehalem and Westmere support PEBS plus Load Latency info which lie
+# in the four 64 bit words write after the PEBS data:
+#       Status: records the IA32_PERF_GLOBAL_STATUS register value
+#       DLA:    Data Linear Address (EIP)
+#       DSE:    Data Source Encoding, where the latency happens, hit or miss
+#               in L1/L2/L3 or IO operations
+#       LAT:    the actual latency in cycles
+#
+class PebsNHM(PebsEvent):
+        pebs_nhm_num = 0
+        def __init__(self, name, comm, dso, symbol, raw_buf, ev_type=EVTYPE_PEBS_LL):
+                tmp_buf=raw_buf[144:176]
+                status, dla, dse, lat = struct.unpack('QQQQ', tmp_buf)
+                self.status = status
+                self.dla = dla
+                self.dse = dse
+                self.lat = lat
+
+                PebsEvent.__init__(self, name, comm, dso, symbol, raw_buf, ev_type)
+                PebsNHM.pebs_nhm_num += 1
+                del tmp_buf
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py
new file mode 100644
index 000000000..fdd92f699
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py
@@ -0,0 +1,184 @@
+# SchedGui.py - Python extension for perf script, basic GUI code for
+#		traces drawing and overview.
+#
+# Copyright (C) 2010 by Frederic Weisbecker <fweisbec@gmail.com>
+#
+# This software is distributed under the terms of the GNU General
+# Public License ("GPL") version 2 as published by the Free Software
+# Foundation.
+
+
+try:
+	import wx
+except ImportError:
+	raise ImportError, "You need to install the wxpython lib for this script"
+
+
+class RootFrame(wx.Frame):
+	Y_OFFSET = 100
+	RECT_HEIGHT = 100
+	RECT_SPACE = 50
+	EVENT_MARKING_WIDTH = 5
+
+	def __init__(self, sched_tracer, title, parent = None, id = -1):
+		wx.Frame.__init__(self, parent, id, title)
+
+		(self.screen_width, self.screen_height) = wx.GetDisplaySize()
+		self.screen_width -= 10
+		self.screen_height -= 10
+		self.zoom = 0.5
+		self.scroll_scale = 20
+		self.sched_tracer = sched_tracer
+		self.sched_tracer.set_root_win(self)
+		(self.ts_start, self.ts_end) = sched_tracer.interval()
+		self.update_width_virtual()
+		self.nr_rects = sched_tracer.nr_rectangles() + 1
+		self.height_virtual = RootFrame.Y_OFFSET + (self.nr_rects * (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE))
+
+		# whole window panel
+		self.panel = wx.Panel(self, size=(self.screen_width, self.screen_height))
+
+		# scrollable container
+		self.scroll = wx.ScrolledWindow(self.panel)
+		self.scroll.SetScrollbars(self.scroll_scale, self.scroll_scale, self.width_virtual / self.scroll_scale, self.height_virtual / self.scroll_scale)
+		self.scroll.EnableScrolling(True, True)
+		self.scroll.SetFocus()
+
+		# scrollable drawing area
+		self.scroll_panel = wx.Panel(self.scroll, size=(self.screen_width - 15, self.screen_height / 2))
+		self.scroll_panel.Bind(wx.EVT_PAINT, self.on_paint)
+		self.scroll_panel.Bind(wx.EVT_KEY_DOWN, self.on_key_press)
+		self.scroll_panel.Bind(wx.EVT_LEFT_DOWN, self.on_mouse_down)
+		self.scroll.Bind(wx.EVT_PAINT, self.on_paint)
+		self.scroll.Bind(wx.EVT_KEY_DOWN, self.on_key_press)
+		self.scroll.Bind(wx.EVT_LEFT_DOWN, self.on_mouse_down)
+
+		self.scroll.Fit()
+		self.Fit()
+
+		self.scroll_panel.SetDimensions(-1, -1, self.width_virtual, self.height_virtual, wx.SIZE_USE_EXISTING)
+
+		self.txt = None
+
+		self.Show(True)
+
+	def us_to_px(self, val):
+		return val / (10 ** 3) * self.zoom
+
+	def px_to_us(self, val):
+		return (val / self.zoom) * (10 ** 3)
+
+	def scroll_start(self):
+		(x, y) = self.scroll.GetViewStart()
+		return (x * self.scroll_scale, y * self.scroll_scale)
+
+	def scroll_start_us(self):
+		(x, y) = self.scroll_start()
+		return self.px_to_us(x)
+
+	def paint_rectangle_zone(self, nr, color, top_color, start, end):
+		offset_px = self.us_to_px(start - self.ts_start)
+		width_px = self.us_to_px(end - self.ts_start)
+
+		offset_py = RootFrame.Y_OFFSET + (nr * (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE))
+		width_py = RootFrame.RECT_HEIGHT
+
+		dc = self.dc
+
+		if top_color is not None:
+			(r, g, b) = top_color
+			top_color = wx.Colour(r, g, b)
+			brush = wx.Brush(top_color, wx.SOLID)
+			dc.SetBrush(brush)
+			dc.DrawRectangle(offset_px, offset_py, width_px, RootFrame.EVENT_MARKING_WIDTH)
+			width_py -= RootFrame.EVENT_MARKING_WIDTH
+			offset_py += RootFrame.EVENT_MARKING_WIDTH
+
+		(r ,g, b) = color
+		color = wx.Colour(r, g, b)
+		brush = wx.Brush(color, wx.SOLID)
+		dc.SetBrush(brush)
+		dc.DrawRectangle(offset_px, offset_py, width_px, width_py)
+
+	def update_rectangles(self, dc, start, end):
+		start += self.ts_start
+		end += self.ts_start
+		self.sched_tracer.fill_zone(start, end)
+
+	def on_paint(self, event):
+		dc = wx.PaintDC(self.scroll_panel)
+		self.dc = dc
+
+		width = min(self.width_virtual, self.screen_width)
+		(x, y) = self.scroll_start()
+		start = self.px_to_us(x)
+		end = self.px_to_us(x + width)
+		self.update_rectangles(dc, start, end)
+
+	def rect_from_ypixel(self, y):
+		y -= RootFrame.Y_OFFSET
+		rect = y / (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE)
+		height = y % (RootFrame.RECT_HEIGHT + RootFrame.RECT_SPACE)
+
+		if rect < 0 or rect > self.nr_rects - 1 or height > RootFrame.RECT_HEIGHT:
+			return -1
+
+		return rect
+
+	def update_summary(self, txt):
+		if self.txt:
+			self.txt.Destroy()
+		self.txt = wx.StaticText(self.panel, -1, txt, (0, (self.screen_height / 2) + 50))
+
+
+	def on_mouse_down(self, event):
+		(x, y) = event.GetPositionTuple()
+		rect = self.rect_from_ypixel(y)
+		if rect == -1:
+			return
+
+		t = self.px_to_us(x) + self.ts_start
+
+		self.sched_tracer.mouse_down(rect, t)
+
+
+	def update_width_virtual(self):
+		self.width_virtual = self.us_to_px(self.ts_end - self.ts_start)
+
+	def __zoom(self, x):
+		self.update_width_virtual()
+		(xpos, ypos) = self.scroll.GetViewStart()
+		xpos = self.us_to_px(x) / self.scroll_scale
+		self.scroll.SetScrollbars(self.scroll_scale, self.scroll_scale, self.width_virtual / self.scroll_scale, self.height_virtual / self.scroll_scale, xpos, ypos)
+		self.Refresh()
+
+	def zoom_in(self):
+		x = self.scroll_start_us()
+		self.zoom *= 2
+		self.__zoom(x)
+
+	def zoom_out(self):
+		x = self.scroll_start_us()
+		self.zoom /= 2
+		self.__zoom(x)
+
+
+	def on_key_press(self, event):
+		key = event.GetRawKeyCode()
+		if key == ord("+"):
+			self.zoom_in()
+			return
+		if key == ord("-"):
+			self.zoom_out()
+			return
+
+		key = event.GetKeyCode()
+		(x, y) = self.scroll.GetViewStart()
+		if key == wx.WXK_RIGHT:
+			self.scroll.Scroll(x + 1, y)
+		elif key == wx.WXK_LEFT:
+			self.scroll.Scroll(x - 1, y)
+		elif key == wx.WXK_DOWN:
+			self.scroll.Scroll(x, y + 1)
+		elif key == wx.WXK_UP:
+			self.scroll.Scroll(x, y - 1)
diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py
new file mode 100644
index 000000000..15c840024
--- /dev/null
+++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py
@@ -0,0 +1,86 @@
+# Util.py - Python extension for perf script, miscellaneous utility code
+#
+# Copyright (C) 2010 by Tom Zanussi <tzanussi@gmail.com>
+#
+# This software may be distributed under the terms of the GNU General
+# Public License ("GPL") version 2 as published by the Free Software
+# Foundation.
+
+import errno, os
+
+FUTEX_WAIT = 0
+FUTEX_WAKE = 1
+FUTEX_PRIVATE_FLAG = 128
+FUTEX_CLOCK_REALTIME = 256
+FUTEX_CMD_MASK = ~(FUTEX_PRIVATE_FLAG | FUTEX_CLOCK_REALTIME)
+
+NSECS_PER_SEC    = 1000000000
+
+def avg(total, n):
+    return total / n
+
+def nsecs(secs, nsecs):
+    return secs * NSECS_PER_SEC + nsecs
+
+def nsecs_secs(nsecs):
+    return nsecs / NSECS_PER_SEC
+
+def nsecs_nsecs(nsecs):
+    return nsecs % NSECS_PER_SEC
+
+def nsecs_str(nsecs):
+    str = "%5u.%09u" % (nsecs_secs(nsecs), nsecs_nsecs(nsecs)),
+    return str
+
+def add_stats(dict, key, value):
+	if not dict.has_key(key):
+		dict[key] = (value, value, value, 1)
+	else:
+		min, max, avg, count = dict[key]
+		if value < min:
+			min = value
+		if value > max:
+			max = value
+		avg = (avg + value) / 2
+		dict[key] = (min, max, avg, count + 1)
+
+def clear_term():
+    print("\x1b[H\x1b[2J")
+
+audit_package_warned = False
+
+try:
+	import audit
+	machine_to_id = {
+		'x86_64': audit.MACH_86_64,
+		'alpha'	: audit.MACH_ALPHA,
+		'ia64'	: audit.MACH_IA64,
+		'ppc'	: audit.MACH_PPC,
+		'ppc64'	: audit.MACH_PPC64,
+		's390'	: audit.MACH_S390,
+		's390x'	: audit.MACH_S390X,
+		'i386'	: audit.MACH_X86,
+		'i586'	: audit.MACH_X86,
+		'i686'	: audit.MACH_X86,
+	}
+	try:
+		machine_to_id['armeb'] = audit.MACH_ARMEB
+	except:
+		pass
+	machine_id = machine_to_id[os.uname()[4]]
+except:
+	if not audit_package_warned:
+		audit_package_warned = True
+		print "Install the audit-libs-python package to get syscall names"
+
+def syscall_name(id):
+	try:
+		return audit.audit_syscall_to_name(id, machine_id)
+	except:
+		return str(id)
+
+def strerror(nr):
+	try:
+		return errno.errorcode[abs(nr)]
+	except:
+		return "Unknown %d errno" % nr
diff --git a/tools/perf/scripts/python/bin/event_analyzing_sample-record b/tools/perf/scripts/python/bin/event_analyzing_sample-record
new file mode 100644
index 000000000..5ce652dab
--- /dev/null
+++ b/tools/perf/scripts/python/bin/event_analyzing_sample-record
@@ -0,0 +1,8 @@
+#!/bin/bash
+
+#
+# event_analyzing_sample.py can cover all type of perf samples including
+# the tracepoints, so no special record requirements, just record what
+# you want to analyze.
+#
+perf record $@
diff --git a/tools/perf/scripts/python/bin/event_analyzing_sample-report b/tools/perf/scripts/python/bin/event_analyzing_sample-report
new file mode 100644
index 000000000..0941fc94e
--- /dev/null
+++ b/tools/perf/scripts/python/bin/event_analyzing_sample-report
@@ -0,0 +1,3 @@
+#!/bin/bash
+# description: analyze all perf samples
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/event_analyzing_sample.py
diff --git a/tools/perf/scripts/python/bin/export-to-postgresql-record b/tools/perf/scripts/python/bin/export-to-postgresql-record
new file mode 100644
index 000000000..221d66e05
--- /dev/null
+++ b/tools/perf/scripts/python/bin/export-to-postgresql-record
@@ -0,0 +1,8 @@
+#!/bin/bash
+
+#
+# export perf data to a postgresql database. Can cover
+# perf ip samples (excluding the tracepoints). No special
+# record requirements, just record what you want to export.
+#
+perf record $@
diff --git a/tools/perf/scripts/python/bin/export-to-postgresql-report b/tools/perf/scripts/python/bin/export-to-postgresql-report
new file mode 100644
index 000000000..cd335b6e2
--- /dev/null
+++ b/tools/perf/scripts/python/bin/export-to-postgresql-report
@@ -0,0 +1,29 @@
+#!/bin/bash
+# description: export perf data to a postgresql database
+# args: [database name] [columns] [calls]
+n_args=0
+for i in "$@"
+do
+    if expr match "$i" "-" > /dev/null ; then
+	break
+    fi
+    n_args=$(( $n_args + 1 ))
+done
+if [ "$n_args" -gt 3 ] ; then
+    echo "usage: export-to-postgresql-report [database name] [columns] [calls]"
+    exit
+fi
+if [ "$n_args" -gt 2 ] ; then
+    dbname=$1
+    columns=$2
+    calls=$3
+    shift 3
+elif [ "$n_args" -gt 1 ] ; then
+    dbname=$1
+    columns=$2
+    shift 2
+elif [ "$n_args" -gt 0 ] ; then
+    dbname=$1
+    shift
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/export-to-postgresql.py $dbname $columns $calls
diff --git a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record
new file mode 100644
index 000000000..74685f318
--- /dev/null
+++ b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-record
@@ -0,0 +1,3 @@
+#!/bin/bash
+(perf record -e raw_syscalls:sys_exit $@ || \
+ perf record -e syscalls:sys_exit $@) 2> /dev/null
diff --git a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report
new file mode 100644
index 000000000..fda5096d0
--- /dev/null
+++ b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: system-wide failed syscalls, by pid
+# args: [comm]
+if [ $# -gt 0 ] ; then
+    if ! expr match "$1" "-" > /dev/null ; then
+	comm=$1
+	shift
+    fi
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/failed-syscalls-by-pid.py $comm
diff --git a/tools/perf/scripts/python/bin/futex-contention-record b/tools/perf/scripts/python/bin/futex-contention-record
new file mode 100644
index 000000000..b1495c9a9
--- /dev/null
+++ b/tools/perf/scripts/python/bin/futex-contention-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e syscalls:sys_enter_futex -e syscalls:sys_exit_futex $@
diff --git a/tools/perf/scripts/python/bin/futex-contention-report b/tools/perf/scripts/python/bin/futex-contention-report
new file mode 100644
index 000000000..6c4427109
--- /dev/null
+++ b/tools/perf/scripts/python/bin/futex-contention-report
@@ -0,0 +1,4 @@
+#!/bin/bash
+# description: futext contention measurement
+
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/futex-contention.py
diff --git a/tools/perf/scripts/python/bin/net_dropmonitor-record b/tools/perf/scripts/python/bin/net_dropmonitor-record
new file mode 100755
index 000000000..423fb81da
--- /dev/null
+++ b/tools/perf/scripts/python/bin/net_dropmonitor-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -e skb:kfree_skb $@
diff --git a/tools/perf/scripts/python/bin/net_dropmonitor-report b/tools/perf/scripts/python/bin/net_dropmonitor-report
new file mode 100755
index 000000000..8d698f5a0
--- /dev/null
+++ b/tools/perf/scripts/python/bin/net_dropmonitor-report
@@ -0,0 +1,4 @@
+#!/bin/bash
+# description: display a table of dropped frames
+
+perf script -s "$PERF_EXEC_PATH"/scripts/python/net_dropmonitor.py $@
diff --git a/tools/perf/scripts/python/bin/netdev-times-record b/tools/perf/scripts/python/bin/netdev-times-record
new file mode 100644
index 000000000..558754b84
--- /dev/null
+++ b/tools/perf/scripts/python/bin/netdev-times-record
@@ -0,0 +1,8 @@
+#!/bin/bash
+perf record -e net:net_dev_xmit -e net:net_dev_queue		\
+		-e net:netif_receive_skb -e net:netif_rx		\
+		-e skb:consume_skb -e skb:kfree_skb			\
+		-e skb:skb_copy_datagram_iovec -e napi:napi_poll	\
+		-e irq:irq_handler_entry -e irq:irq_handler_exit	\
+		-e irq:softirq_entry -e irq:softirq_exit		\
+		-e irq:softirq_raise $@
diff --git a/tools/perf/scripts/python/bin/netdev-times-report b/tools/perf/scripts/python/bin/netdev-times-report
new file mode 100644
index 000000000..8f759291d
--- /dev/null
+++ b/tools/perf/scripts/python/bin/netdev-times-report
@@ -0,0 +1,5 @@
+#!/bin/bash
+# description: display a process of packet and processing time
+# args: [tx] [rx] [dev=] [debug]
+
+perf script -s "$PERF_EXEC_PATH"/scripts/python/netdev-times.py $@
diff --git a/tools/perf/scripts/python/bin/sched-migration-record b/tools/perf/scripts/python/bin/sched-migration-record
new file mode 100644
index 000000000..7493fddbe
--- /dev/null
+++ b/tools/perf/scripts/python/bin/sched-migration-record
@@ -0,0 +1,2 @@
+#!/bin/bash
+perf record -m 16384 -e sched:sched_wakeup -e sched:sched_wakeup_new -e sched:sched_switch -e sched:sched_migrate_task $@
diff --git a/tools/perf/scripts/python/bin/sched-migration-report b/tools/perf/scripts/python/bin/sched-migration-report
new file mode 100644
index 000000000..68b037a18
--- /dev/null
+++ b/tools/perf/scripts/python/bin/sched-migration-report
@@ -0,0 +1,3 @@
+#!/bin/bash
+# description: sched migration overview
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/sched-migration.py
diff --git a/tools/perf/scripts/python/bin/sctop-record b/tools/perf/scripts/python/bin/sctop-record
new file mode 100644
index 000000000..d6940841e
--- /dev/null
+++ b/tools/perf/scripts/python/bin/sctop-record
@@ -0,0 +1,3 @@
+#!/bin/bash
+(perf record -e raw_syscalls:sys_enter $@ || \
+ perf record -e syscalls:sys_enter $@) 2> /dev/null
diff --git a/tools/perf/scripts/python/bin/sctop-report b/tools/perf/scripts/python/bin/sctop-report
new file mode 100644
index 000000000..c32db2941
--- /dev/null
+++ b/tools/perf/scripts/python/bin/sctop-report
@@ -0,0 +1,24 @@
+#!/bin/bash
+# description: syscall top
+# args: [comm] [interval]
+n_args=0
+for i in "$@"
+do
+    if expr match "$i" "-" > /dev/null ; then
+	break
+    fi
+    n_args=$(( $n_args + 1 ))
+done
+if [ "$n_args" -gt 2 ] ; then
+    echo "usage: sctop-report [comm] [interval]"
+    exit
+fi
+if [ "$n_args" -gt 1 ] ; then
+    comm=$1
+    interval=$2
+    shift 2
+elif [ "$n_args" -gt 0 ] ; then
+    interval=$1
+    shift
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/sctop.py $comm $interval
diff --git a/tools/perf/scripts/python/bin/syscall-counts-by-pid-record b/tools/perf/scripts/python/bin/syscall-counts-by-pid-record
new file mode 100644
index 000000000..d6940841e
--- /dev/null
+++ b/tools/perf/scripts/python/bin/syscall-counts-by-pid-record
@@ -0,0 +1,3 @@
+#!/bin/bash
+(perf record -e raw_syscalls:sys_enter $@ || \
+ perf record -e syscalls:sys_enter $@) 2> /dev/null
diff --git a/tools/perf/scripts/python/bin/syscall-counts-by-pid-report b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report
new file mode 100644
index 000000000..16eb8d65c
--- /dev/null
+++ b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: system-wide syscall counts, by pid
+# args: [comm]
+if [ $# -gt 0 ] ; then
+    if ! expr match "$1" "-" > /dev/null ; then
+	comm=$1
+	shift
+    fi
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts-by-pid.py $comm
diff --git a/tools/perf/scripts/python/bin/syscall-counts-record b/tools/perf/scripts/python/bin/syscall-counts-record
new file mode 100644
index 000000000..d6940841e
--- /dev/null
+++ b/tools/perf/scripts/python/bin/syscall-counts-record
@@ -0,0 +1,3 @@
+#!/bin/bash
+(perf record -e raw_syscalls:sys_enter $@ || \
+ perf record -e syscalls:sys_enter $@) 2> /dev/null
diff --git a/tools/perf/scripts/python/bin/syscall-counts-report b/tools/perf/scripts/python/bin/syscall-counts-report
new file mode 100644
index 000000000..0f0e9d453
--- /dev/null
+++ b/tools/perf/scripts/python/bin/syscall-counts-report
@@ -0,0 +1,10 @@
+#!/bin/bash
+# description: system-wide syscall counts
+# args: [comm]
+if [ $# -gt 0 ] ; then
+    if ! expr match "$1" "-" > /dev/null ; then
+	comm=$1
+	shift
+    fi
+fi
+perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts.py $comm
diff --git a/tools/perf/scripts/python/check-perf-trace.py b/tools/perf/scripts/python/check-perf-trace.py
new file mode 100644
index 000000000..334599c60
--- /dev/null
+++ b/tools/perf/scripts/python/check-perf-trace.py
@@ -0,0 +1,82 @@
+# perf script event handlers, generated by perf script -g python
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# This script tests basic functionality such as flag and symbol
+# strings, common_xxx() calls back into perf, begin, end, unhandled
+# events, etc.  Basically, if this script runs successfully and
+# displays expected results, Python scripting support should be ok.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from Core import *
+from perf_trace_context import *
+
+unhandled = autodict()
+
+def trace_begin():
+	print "trace_begin"
+	pass
+
+def trace_end():
+        print_unhandled()
+
+def irq__softirq_entry(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, vec):
+		print_header(event_name, common_cpu, common_secs, common_nsecs,
+			common_pid, common_comm)
+
+                print_uncommon(context)
+
+		print "vec=%s\n" % \
+		(symbol_str("irq__softirq_entry", "vec", vec)),
+
+def kmem__kmalloc(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, call_site, ptr, bytes_req, bytes_alloc,
+	gfp_flags):
+		print_header(event_name, common_cpu, common_secs, common_nsecs,
+			common_pid, common_comm)
+
+                print_uncommon(context)
+
+		print "call_site=%u, ptr=%u, bytes_req=%u, " \
+		"bytes_alloc=%u, gfp_flags=%s\n" % \
+		(call_site, ptr, bytes_req, bytes_alloc,
+
+		flag_str("kmem__kmalloc", "gfp_flags", gfp_flags)),
+
+def trace_unhandled(event_name, context, event_fields_dict):
+    try:
+        unhandled[event_name] += 1
+    except TypeError:
+        unhandled[event_name] = 1
+
+def print_header(event_name, cpu, secs, nsecs, pid, comm):
+	print "%-20s %5u %05u.%09u %8u %-20s " % \
+	(event_name, cpu, secs, nsecs, pid, comm),
+
+# print trace fields not included in handler args
+def print_uncommon(context):
+    print "common_preempt_count=%d, common_flags=%s, common_lock_depth=%d, " \
+        % (common_pc(context), trace_flag_str(common_flags(context)), \
+               common_lock_depth(context))
+
+def print_unhandled():
+    keys = unhandled.keys()
+    if not keys:
+        return
+
+    print "\nunhandled events:\n\n",
+
+    print "%-40s  %10s\n" % ("event", "count"),
+    print "%-40s  %10s\n" % ("----------------------------------------", \
+                                 "-----------"),
+
+    for event_name in keys:
+	print "%-40s  %10d\n" % (event_name, unhandled[event_name])
diff --git a/tools/perf/scripts/python/event_analyzing_sample.py b/tools/perf/scripts/python/event_analyzing_sample.py
new file mode 100644
index 000000000..163c39fa1
--- /dev/null
+++ b/tools/perf/scripts/python/event_analyzing_sample.py
@@ -0,0 +1,189 @@
+# event_analyzing_sample.py: general event handler in python
+#
+# Current perf report is already very powerful with the annotation integrated,
+# and this script is not trying to be as powerful as perf report, but
+# providing end user/developer a flexible way to analyze the events other
+# than trace points.
+#
+# The 2 database related functions in this script just show how to gather
+# the basic information, and users can modify and write their own functions
+# according to their specific requirement.
+#
+# The first function "show_general_events" just does a basic grouping for all
+# generic events with the help of sqlite, and the 2nd one "show_pebs_ll" is
+# for a x86 HW PMU event: PEBS with load latency data.
+#
+
+import os
+import sys
+import math
+import struct
+import sqlite3
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+        '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from EventClass import *
+
+#
+# If the perf.data has a big number of samples, then the insert operation
+# will be very time consuming (about 10+ minutes for 10000 samples) if the
+# .db database is on disk. Move the .db file to RAM based FS to speedup
+# the handling, which will cut the time down to several seconds.
+#
+con = sqlite3.connect("/dev/shm/perf.db")
+con.isolation_level = None
+
+def trace_begin():
+	print "In trace_begin:\n"
+
+        #
+        # Will create several tables at the start, pebs_ll is for PEBS data with
+        # load latency info, while gen_events is for general event.
+        #
+        con.execute("""
+                create table if not exists gen_events (
+                        name text,
+                        symbol text,
+                        comm text,
+                        dso text
+                );""")
+        con.execute("""
+                create table if not exists pebs_ll (
+                        name text,
+                        symbol text,
+                        comm text,
+                        dso text,
+                        flags integer,
+                        ip integer,
+                        status integer,
+                        dse integer,
+                        dla integer,
+                        lat integer
+                );""")
+
+#
+# Create and insert event object to a database so that user could
+# do more analysis with simple database commands.
+#
+def process_event(param_dict):
+        event_attr = param_dict["attr"]
+        sample     = param_dict["sample"]
+        raw_buf    = param_dict["raw_buf"]
+        comm       = param_dict["comm"]
+        name       = param_dict["ev_name"]
+
+        # Symbol and dso info are not always resolved
+        if (param_dict.has_key("dso")):
+                dso = param_dict["dso"]
+        else:
+                dso = "Unknown_dso"
+
+        if (param_dict.has_key("symbol")):
+                symbol = param_dict["symbol"]
+        else:
+                symbol = "Unknown_symbol"
+
+        # Create the event object and insert it to the right table in database
+        event = create_event(name, comm, dso, symbol, raw_buf)
+        insert_db(event)
+
+def insert_db(event):
+        if event.ev_type == EVTYPE_GENERIC:
+                con.execute("insert into gen_events values(?, ?, ?, ?)",
+                                (event.name, event.symbol, event.comm, event.dso))
+        elif event.ev_type == EVTYPE_PEBS_LL:
+                event.ip &= 0x7fffffffffffffff
+                event.dla &= 0x7fffffffffffffff
+                con.execute("insert into pebs_ll values (?, ?, ?, ?, ?, ?, ?, ?, ?, ?)",
+                        (event.name, event.symbol, event.comm, event.dso, event.flags,
+                                event.ip, event.status, event.dse, event.dla, event.lat))
+
+def trace_end():
+	print "In trace_end:\n"
+        # We show the basic info for the 2 type of event classes
+        show_general_events()
+        show_pebs_ll()
+        con.close()
+
+#
+# As the event number may be very big, so we can't use linear way
+# to show the histogram in real number, but use a log2 algorithm.
+#
+
+def num2sym(num):
+        # Each number will have at least one '#'
+        snum = '#' * (int)(math.log(num, 2) + 1)
+        return snum
+
+def show_general_events():
+
+        # Check the total record number in the table
+        count = con.execute("select count(*) from gen_events")
+        for t in count:
+                print "There is %d records in gen_events table" % t[0]
+                if t[0] == 0:
+                        return
+
+        print "Statistics about the general events grouped by thread/symbol/dso: \n"
+
+         # Group by thread
+        commq = con.execute("select comm, count(comm) from gen_events group by comm order by -count(comm)")
+        print "\n%16s %8s %16s\n%s" % ("comm", "number", "histogram", "="*42)
+        for row in commq:
+             print "%16s %8d     %s" % (row[0], row[1], num2sym(row[1]))
+
+        # Group by symbol
+        print "\n%32s %8s %16s\n%s" % ("symbol", "number", "histogram", "="*58)
+        symbolq = con.execute("select symbol, count(symbol) from gen_events group by symbol order by -count(symbol)")
+        for row in symbolq:
+             print "%32s %8d     %s" % (row[0], row[1], num2sym(row[1]))
+
+        # Group by dso
+        print "\n%40s %8s %16s\n%s" % ("dso", "number", "histogram", "="*74)
+        dsoq = con.execute("select dso, count(dso) from gen_events group by dso order by -count(dso)")
+        for row in dsoq:
+             print "%40s %8d     %s" % (row[0], row[1], num2sym(row[1]))
+
+#
+# This function just shows the basic info, and we could do more with the
+# data in the tables, like checking the function parameters when some
+# big latency events happen.
+#
+def show_pebs_ll():
+
+        count = con.execute("select count(*) from pebs_ll")
+        for t in count:
+                print "There is %d records in pebs_ll table" % t[0]
+                if t[0] == 0:
+                        return
+
+        print "Statistics about the PEBS Load Latency events grouped by thread/symbol/dse/latency: \n"
+
+        # Group by thread
+        commq = con.execute("select comm, count(comm) from pebs_ll group by comm order by -count(comm)")
+        print "\n%16s %8s %16s\n%s" % ("comm", "number", "histogram", "="*42)
+        for row in commq:
+             print "%16s %8d     %s" % (row[0], row[1], num2sym(row[1]))
+
+        # Group by symbol
+        print "\n%32s %8s %16s\n%s" % ("symbol", "number", "histogram", "="*58)
+        symbolq = con.execute("select symbol, count(symbol) from pebs_ll group by symbol order by -count(symbol)")
+        for row in symbolq:
+             print "%32s %8d     %s" % (row[0], row[1], num2sym(row[1]))
+
+        # Group by dse
+        dseq = con.execute("select dse, count(dse) from pebs_ll group by dse order by -count(dse)")
+        print "\n%32s %8s %16s\n%s" % ("dse", "number", "histogram", "="*58)
+        for row in dseq:
+             print "%32s %8d     %s" % (row[0], row[1], num2sym(row[1]))
+
+        # Group by latency
+        latq = con.execute("select lat, count(lat) from pebs_ll group by lat order by lat")
+        print "\n%32s %8s %16s\n%s" % ("latency", "number", "histogram", "="*58)
+        for row in latq:
+             print "%32s %8d     %s" % (row[0], row[1], num2sym(row[1]))
+
+def trace_unhandled(event_name, context, event_fields_dict):
+		print ' '.join(['%s=%s'%(k,str(v))for k,v in sorted(event_fields_dict.items())])
diff --git a/tools/perf/scripts/python/export-to-postgresql.py b/tools/perf/scripts/python/export-to-postgresql.py
new file mode 100644
index 000000000..4cdafd880
--- /dev/null
+++ b/tools/perf/scripts/python/export-to-postgresql.py
@@ -0,0 +1,444 @@
+# export-to-postgresql.py: export perf data to a postgresql database
+# Copyright (c) 2014, Intel Corporation.
+#
+# This program is free software; you can redistribute it and/or modify it
+# under the terms and conditions of the GNU General Public License,
+# version 2, as published by the Free Software Foundation.
+#
+# This program is distributed in the hope it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+# more details.
+
+import os
+import sys
+import struct
+import datetime
+
+from PySide.QtSql import *
+
+# Need to access PostgreSQL C library directly to use COPY FROM STDIN
+from ctypes import *
+libpq = CDLL("libpq.so.5")
+PQconnectdb = libpq.PQconnectdb
+PQconnectdb.restype = c_void_p
+PQfinish = libpq.PQfinish
+PQstatus = libpq.PQstatus
+PQexec = libpq.PQexec
+PQexec.restype = c_void_p
+PQresultStatus = libpq.PQresultStatus
+PQputCopyData = libpq.PQputCopyData
+PQputCopyData.argtypes = [ c_void_p, c_void_p, c_int ]
+PQputCopyEnd = libpq.PQputCopyEnd
+PQputCopyEnd.argtypes = [ c_void_p, c_void_p ]
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+# These perf imports are not used at present
+#from perf_trace_context import *
+#from Core import *
+
+perf_db_export_mode = True
+perf_db_export_calls = False
+
+def usage():
+	print >> sys.stderr, "Usage is: export-to-postgresql.py <database name> [<columns>] [<calls>]"
+	print >> sys.stderr, "where:	columns		'all' or 'branches'"
+	print >> sys.stderr, "		calls		'calls' => create calls table"
+	raise Exception("Too few arguments")
+
+if (len(sys.argv) < 2):
+	usage()
+
+dbname = sys.argv[1]
+
+if (len(sys.argv) >= 3):
+	columns = sys.argv[2]
+else:
+	columns = "all"
+
+if columns not in ("all", "branches"):
+	usage()
+
+branches = (columns == "branches")
+
+if (len(sys.argv) >= 4):
+	if (sys.argv[3] == "calls"):
+		perf_db_export_calls = True
+	else:
+		usage()
+
+output_dir_name = os.getcwd() + "/" + dbname + "-perf-data"
+os.mkdir(output_dir_name)
+
+def do_query(q, s):
+	if (q.exec_(s)):
+		return
+	raise Exception("Query failed: " + q.lastError().text())
+
+print datetime.datetime.today(), "Creating database..."
+
+db = QSqlDatabase.addDatabase('QPSQL')
+query = QSqlQuery(db)
+db.setDatabaseName('postgres')
+db.open()
+try:
+	do_query(query, 'CREATE DATABASE ' + dbname)
+except:
+	os.rmdir(output_dir_name)
+	raise
+query.finish()
+query.clear()
+db.close()
+
+db.setDatabaseName(dbname)
+db.open()
+
+query = QSqlQuery(db)
+do_query(query, 'SET client_min_messages TO WARNING')
+
+do_query(query, 'CREATE TABLE selected_events ('
+		'id		bigint		NOT NULL,'
+		'name		varchar(80))')
+do_query(query, 'CREATE TABLE machines ('
+		'id		bigint		NOT NULL,'
+		'pid		integer,'
+		'root_dir 	varchar(4096))')
+do_query(query, 'CREATE TABLE threads ('
+		'id		bigint		NOT NULL,'
+		'machine_id	bigint,'
+		'process_id	bigint,'
+		'pid		integer,'
+		'tid		integer)')
+do_query(query, 'CREATE TABLE comms ('
+		'id		bigint		NOT NULL,'
+		'comm		varchar(16))')
+do_query(query, 'CREATE TABLE comm_threads ('
+		'id		bigint		NOT NULL,'
+		'comm_id	bigint,'
+		'thread_id	bigint)')
+do_query(query, 'CREATE TABLE dsos ('
+		'id		bigint		NOT NULL,'
+		'machine_id	bigint,'
+		'short_name	varchar(256),'
+		'long_name	varchar(4096),'
+		'build_id	varchar(64))')
+do_query(query, 'CREATE TABLE symbols ('
+		'id		bigint		NOT NULL,'
+		'dso_id		bigint,'
+		'sym_start	bigint,'
+		'sym_end	bigint,'
+		'binding	integer,'
+		'name		varchar(2048))')
+do_query(query, 'CREATE TABLE branch_types ('
+		'id		integer		NOT NULL,'
+		'name		varchar(80))')
+
+if branches:
+	do_query(query, 'CREATE TABLE samples ('
+		'id		bigint		NOT NULL,'
+		'evsel_id	bigint,'
+		'machine_id	bigint,'
+		'thread_id	bigint,'
+		'comm_id	bigint,'
+		'dso_id		bigint,'
+		'symbol_id	bigint,'
+		'sym_offset	bigint,'
+		'ip		bigint,'
+		'time		bigint,'
+		'cpu		integer,'
+		'to_dso_id	bigint,'
+		'to_symbol_id	bigint,'
+		'to_sym_offset	bigint,'
+		'to_ip		bigint,'
+		'branch_type	integer,'
+		'in_tx		boolean)')
+else:
+	do_query(query, 'CREATE TABLE samples ('
+		'id		bigint		NOT NULL,'
+		'evsel_id	bigint,'
+		'machine_id	bigint,'
+		'thread_id	bigint,'
+		'comm_id	bigint,'
+		'dso_id		bigint,'
+		'symbol_id	bigint,'
+		'sym_offset	bigint,'
+		'ip		bigint,'
+		'time		bigint,'
+		'cpu		integer,'
+		'to_dso_id	bigint,'
+		'to_symbol_id	bigint,'
+		'to_sym_offset	bigint,'
+		'to_ip		bigint,'
+		'period		bigint,'
+		'weight		bigint,'
+		'transaction	bigint,'
+		'data_src	bigint,'
+		'branch_type	integer,'
+		'in_tx		boolean)')
+
+if perf_db_export_calls:
+	do_query(query, 'CREATE TABLE call_paths ('
+		'id		bigint		NOT NULL,'
+		'parent_id	bigint,'
+		'symbol_id	bigint,'
+		'ip		bigint)')
+	do_query(query, 'CREATE TABLE calls ('
+		'id		bigint		NOT NULL,'
+		'thread_id	bigint,'
+		'comm_id	bigint,'
+		'call_path_id	bigint,'
+		'call_time	bigint,'
+		'return_time	bigint,'
+		'branch_count	bigint,'
+		'call_id	bigint,'
+		'return_id	bigint,'
+		'parent_call_path_id	bigint,'
+		'flags		integer)')
+
+do_query(query, 'CREATE VIEW samples_view AS '
+	'SELECT '
+		'id,'
+		'time,'
+		'cpu,'
+		'(SELECT pid FROM threads WHERE id = thread_id) AS pid,'
+		'(SELECT tid FROM threads WHERE id = thread_id) AS tid,'
+		'(SELECT comm FROM comms WHERE id = comm_id) AS command,'
+		'(SELECT name FROM selected_events WHERE id = evsel_id) AS event,'
+		'to_hex(ip) AS ip_hex,'
+		'(SELECT name FROM symbols WHERE id = symbol_id) AS symbol,'
+		'sym_offset,'
+		'(SELECT short_name FROM dsos WHERE id = dso_id) AS dso_short_name,'
+		'to_hex(to_ip) AS to_ip_hex,'
+		'(SELECT name FROM symbols WHERE id = to_symbol_id) AS to_symbol,'
+		'to_sym_offset,'
+		'(SELECT short_name FROM dsos WHERE id = to_dso_id) AS to_dso_short_name,'
+		'(SELECT name FROM branch_types WHERE id = branch_type) AS branch_type_name,'
+		'in_tx'
+	' FROM samples')
+
+
+file_header = struct.pack("!11sii", "PGCOPY\n\377\r\n\0", 0, 0)
+file_trailer = "\377\377"
+
+def open_output_file(file_name):
+	path_name = output_dir_name + "/" + file_name
+	file = open(path_name, "w+")
+	file.write(file_header)
+	return file
+
+def close_output_file(file):
+	file.write(file_trailer)
+	file.close()
+
+def copy_output_file_direct(file, table_name):
+	close_output_file(file)
+	sql = "COPY " + table_name + " FROM '" + file.name + "' (FORMAT 'binary')"
+	do_query(query, sql)
+
+# Use COPY FROM STDIN because security may prevent postgres from accessing the files directly
+def copy_output_file(file, table_name):
+	conn = PQconnectdb("dbname = " + dbname)
+	if (PQstatus(conn)):
+		raise Exception("COPY FROM STDIN PQconnectdb failed")
+	file.write(file_trailer)
+	file.seek(0)
+	sql = "COPY " + table_name + " FROM STDIN (FORMAT 'binary')"
+	res = PQexec(conn, sql)
+	if (PQresultStatus(res) != 4):
+		raise Exception("COPY FROM STDIN PQexec failed")
+	data = file.read(65536)
+	while (len(data)):
+		ret = PQputCopyData(conn, data, len(data))
+		if (ret != 1):
+			raise Exception("COPY FROM STDIN PQputCopyData failed, error " + str(ret))
+		data = file.read(65536)
+	ret = PQputCopyEnd(conn, None)
+	if (ret != 1):
+		raise Exception("COPY FROM STDIN PQputCopyEnd failed, error " + str(ret))
+	PQfinish(conn)
+
+def remove_output_file(file):
+	name = file.name
+	file.close()
+	os.unlink(name)
+
+evsel_file		= open_output_file("evsel_table.bin")
+machine_file		= open_output_file("machine_table.bin")
+thread_file		= open_output_file("thread_table.bin")
+comm_file		= open_output_file("comm_table.bin")
+comm_thread_file	= open_output_file("comm_thread_table.bin")
+dso_file		= open_output_file("dso_table.bin")
+symbol_file		= open_output_file("symbol_table.bin")
+branch_type_file	= open_output_file("branch_type_table.bin")
+sample_file		= open_output_file("sample_table.bin")
+if perf_db_export_calls:
+	call_path_file		= open_output_file("call_path_table.bin")
+	call_file		= open_output_file("call_table.bin")
+
+def trace_begin():
+	print datetime.datetime.today(), "Writing to intermediate files..."
+	# id == 0 means unknown.  It is easier to create records for them than replace the zeroes with NULLs
+	evsel_table(0, "unknown")
+	machine_table(0, 0, "unknown")
+	thread_table(0, 0, 0, -1, -1)
+	comm_table(0, "unknown")
+	dso_table(0, 0, "unknown", "unknown", "")
+	symbol_table(0, 0, 0, 0, 0, "unknown")
+	sample_table(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)
+	if perf_db_export_calls:
+		call_path_table(0, 0, 0, 0)
+
+unhandled_count = 0
+
+def trace_end():
+	print datetime.datetime.today(), "Copying to database..."
+	copy_output_file(evsel_file,		"selected_events")
+	copy_output_file(machine_file,		"machines")
+	copy_output_file(thread_file,		"threads")
+	copy_output_file(comm_file,		"comms")
+	copy_output_file(comm_thread_file,	"comm_threads")
+	copy_output_file(dso_file,		"dsos")
+	copy_output_file(symbol_file,		"symbols")
+	copy_output_file(branch_type_file,	"branch_types")
+	copy_output_file(sample_file,		"samples")
+	if perf_db_export_calls:
+		copy_output_file(call_path_file,	"call_paths")
+		copy_output_file(call_file,		"calls")
+
+	print datetime.datetime.today(), "Removing intermediate files..."
+	remove_output_file(evsel_file)
+	remove_output_file(machine_file)
+	remove_output_file(thread_file)
+	remove_output_file(comm_file)
+	remove_output_file(comm_thread_file)
+	remove_output_file(dso_file)
+	remove_output_file(symbol_file)
+	remove_output_file(branch_type_file)
+	remove_output_file(sample_file)
+	if perf_db_export_calls:
+		remove_output_file(call_path_file)
+		remove_output_file(call_file)
+	os.rmdir(output_dir_name)
+	print datetime.datetime.today(), "Adding primary keys"
+	do_query(query, 'ALTER TABLE selected_events ADD PRIMARY KEY (id)')
+	do_query(query, 'ALTER TABLE machines        ADD PRIMARY KEY (id)')
+	do_query(query, 'ALTER TABLE threads         ADD PRIMARY KEY (id)')
+	do_query(query, 'ALTER TABLE comms           ADD PRIMARY KEY (id)')
+	do_query(query, 'ALTER TABLE comm_threads    ADD PRIMARY KEY (id)')
+	do_query(query, 'ALTER TABLE dsos            ADD PRIMARY KEY (id)')
+	do_query(query, 'ALTER TABLE symbols         ADD PRIMARY KEY (id)')
+	do_query(query, 'ALTER TABLE branch_types    ADD PRIMARY KEY (id)')
+	do_query(query, 'ALTER TABLE samples         ADD PRIMARY KEY (id)')
+	if perf_db_export_calls:
+		do_query(query, 'ALTER TABLE call_paths      ADD PRIMARY KEY (id)')
+		do_query(query, 'ALTER TABLE calls           ADD PRIMARY KEY (id)')
+
+	print datetime.datetime.today(), "Adding foreign keys"
+	do_query(query, 'ALTER TABLE threads '
+					'ADD CONSTRAINT machinefk  FOREIGN KEY (machine_id)   REFERENCES machines   (id),'
+					'ADD CONSTRAINT processfk  FOREIGN KEY (process_id)   REFERENCES threads    (id)')
+	do_query(query, 'ALTER TABLE comm_threads '
+					'ADD CONSTRAINT commfk     FOREIGN KEY (comm_id)      REFERENCES comms      (id),'
+					'ADD CONSTRAINT threadfk   FOREIGN KEY (thread_id)    REFERENCES threads    (id)')
+	do_query(query, 'ALTER TABLE dsos '
+					'ADD CONSTRAINT machinefk  FOREIGN KEY (machine_id)   REFERENCES machines   (id)')
+	do_query(query, 'ALTER TABLE symbols '
+					'ADD CONSTRAINT dsofk      FOREIGN KEY (dso_id)       REFERENCES dsos       (id)')
+	do_query(query, 'ALTER TABLE samples '
+					'ADD CONSTRAINT evselfk    FOREIGN KEY (evsel_id)     REFERENCES selected_events (id),'
+					'ADD CONSTRAINT machinefk  FOREIGN KEY (machine_id)   REFERENCES machines   (id),'
+					'ADD CONSTRAINT threadfk   FOREIGN KEY (thread_id)    REFERENCES threads    (id),'
+					'ADD CONSTRAINT commfk     FOREIGN KEY (comm_id)      REFERENCES comms      (id),'
+					'ADD CONSTRAINT dsofk      FOREIGN KEY (dso_id)       REFERENCES dsos       (id),'
+					'ADD CONSTRAINT symbolfk   FOREIGN KEY (symbol_id)    REFERENCES symbols    (id),'
+					'ADD CONSTRAINT todsofk    FOREIGN KEY (to_dso_id)    REFERENCES dsos       (id),'
+					'ADD CONSTRAINT tosymbolfk FOREIGN KEY (to_symbol_id) REFERENCES symbols    (id)')
+	if perf_db_export_calls:
+		do_query(query, 'ALTER TABLE call_paths '
+					'ADD CONSTRAINT parentfk    FOREIGN KEY (parent_id)    REFERENCES call_paths (id),'
+					'ADD CONSTRAINT symbolfk    FOREIGN KEY (symbol_id)    REFERENCES symbols    (id)')
+		do_query(query, 'ALTER TABLE calls '
+					'ADD CONSTRAINT threadfk    FOREIGN KEY (thread_id)    REFERENCES threads    (id),'
+					'ADD CONSTRAINT commfk      FOREIGN KEY (comm_id)      REFERENCES comms      (id),'
+					'ADD CONSTRAINT call_pathfk FOREIGN KEY (call_path_id) REFERENCES call_paths (id),'
+					'ADD CONSTRAINT callfk      FOREIGN KEY (call_id)      REFERENCES samples    (id),'
+					'ADD CONSTRAINT returnfk    FOREIGN KEY (return_id)    REFERENCES samples    (id),'
+					'ADD CONSTRAINT parent_call_pathfk FOREIGN KEY (parent_call_path_id) REFERENCES call_paths (id)')
+		do_query(query, 'CREATE INDEX pcpid_idx ON calls (parent_call_path_id)')
+
+	if (unhandled_count):
+		print datetime.datetime.today(), "Warning: ", unhandled_count, " unhandled events"
+	print datetime.datetime.today(), "Done"
+
+def trace_unhandled(event_name, context, event_fields_dict):
+	global unhandled_count
+	unhandled_count += 1
+
+def sched__sched_switch(*x):
+	pass
+
+def evsel_table(evsel_id, evsel_name, *x):
+	n = len(evsel_name)
+	fmt = "!hiqi" + str(n) + "s"
+	value = struct.pack(fmt, 2, 8, evsel_id, n, evsel_name)
+	evsel_file.write(value)
+
+def machine_table(machine_id, pid, root_dir, *x):
+	n = len(root_dir)
+	fmt = "!hiqiii" + str(n) + "s"
+	value = struct.pack(fmt, 3, 8, machine_id, 4, pid, n, root_dir)
+	machine_file.write(value)
+
+def thread_table(thread_id, machine_id, process_id, pid, tid, *x):
+	value = struct.pack("!hiqiqiqiiii", 5, 8, thread_id, 8, machine_id, 8, process_id, 4, pid, 4, tid)
+	thread_file.write(value)
+
+def comm_table(comm_id, comm_str, *x):
+	n = len(comm_str)
+	fmt = "!hiqi" + str(n) + "s"
+	value = struct.pack(fmt, 2, 8, comm_id, n, comm_str)
+	comm_file.write(value)
+
+def comm_thread_table(comm_thread_id, comm_id, thread_id, *x):
+	fmt = "!hiqiqiq"
+	value = struct.pack(fmt, 3, 8, comm_thread_id, 8, comm_id, 8, thread_id)
+	comm_thread_file.write(value)
+
+def dso_table(dso_id, machine_id, short_name, long_name, build_id, *x):
+	n1 = len(short_name)
+	n2 = len(long_name)
+	n3 = len(build_id)
+	fmt = "!hiqiqi" + str(n1) + "si"  + str(n2) + "si" + str(n3) + "s"
+	value = struct.pack(fmt, 5, 8, dso_id, 8, machine_id, n1, short_name, n2, long_name, n3, build_id)
+	dso_file.write(value)
+
+def symbol_table(symbol_id, dso_id, sym_start, sym_end, binding, symbol_name, *x):
+	n = len(symbol_name)
+	fmt = "!hiqiqiqiqiii" + str(n) + "s"
+	value = struct.pack(fmt, 6, 8, symbol_id, 8, dso_id, 8, sym_start, 8, sym_end, 4, binding, n, symbol_name)
+	symbol_file.write(value)
+
+def branch_type_table(branch_type, name, *x):
+	n = len(name)
+	fmt = "!hiii" + str(n) + "s"
+	value = struct.pack(fmt, 2, 4, branch_type, n, name)
+	branch_type_file.write(value)
+
+def sample_table(sample_id, evsel_id, machine_id, thread_id, comm_id, dso_id, symbol_id, sym_offset, ip, time, cpu, to_dso_id, to_symbol_id, to_sym_offset, to_ip, period, weight, transaction, data_src, branch_type, in_tx, *x):
+	if branches:
+		value = struct.pack("!hiqiqiqiqiqiqiqiqiqiqiiiqiqiqiqiiiB", 17, 8, sample_id, 8, evsel_id, 8, machine_id, 8, thread_id, 8, comm_id, 8, dso_id, 8, symbol_id, 8, sym_offset, 8, ip, 8, time, 4, cpu, 8, to_dso_id, 8, to_symbol_id, 8, to_sym_offset, 8, to_ip, 4, branch_type, 1, in_tx)
+	else:
+		value = struct.pack("!hiqiqiqiqiqiqiqiqiqiqiiiqiqiqiqiqiqiqiqiiiB", 21, 8, sample_id, 8, evsel_id, 8, machine_id, 8, thread_id, 8, comm_id, 8, dso_id, 8, symbol_id, 8, sym_offset, 8, ip, 8, time, 4, cpu, 8, to_dso_id, 8, to_symbol_id, 8, to_sym_offset, 8, to_ip, 8, period, 8, weight, 8, transaction, 8, data_src, 4, branch_type, 1, in_tx)
+	sample_file.write(value)
+
+def call_path_table(cp_id, parent_id, symbol_id, ip, *x):
+	fmt = "!hiqiqiqiq"
+	value = struct.pack(fmt, 4, 8, cp_id, 8, parent_id, 8, symbol_id, 8, ip)
+	call_path_file.write(value)
+
+def call_return_table(cr_id, thread_id, comm_id, call_path_id, call_time, return_time, branch_count, call_id, return_id, parent_call_path_id, flags, *x):
+	fmt = "!hiqiqiqiqiqiqiqiqiqiqii"
+	value = struct.pack(fmt, 11, 8, cr_id, 8, thread_id, 8, comm_id, 8, call_path_id, 8, call_time, 8, return_time, 8, branch_count, 8, call_id, 8, return_id, 8, parent_call_path_id, 4, flags)
+	call_file.write(value)
diff --git a/tools/perf/scripts/python/failed-syscalls-by-pid.py b/tools/perf/scripts/python/failed-syscalls-by-pid.py
new file mode 100644
index 000000000..cafeff3d7
--- /dev/null
+++ b/tools/perf/scripts/python/failed-syscalls-by-pid.py
@@ -0,0 +1,78 @@
+# failed system call counts, by pid
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Displays system-wide failed system call totals, broken down by pid.
+# If a [comm] arg is specified, only syscalls called by [comm] are displayed.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+usage = "perf script -s syscall-counts-by-pid.py [comm|pid]\n";
+
+for_comm = None
+for_pid = None
+
+if len(sys.argv) > 2:
+	sys.exit(usage)
+
+if len(sys.argv) > 1:
+	try:
+		for_pid = int(sys.argv[1])
+	except:
+		for_comm = sys.argv[1]
+
+syscalls = autodict()
+
+def trace_begin():
+	print "Press control+C to stop and show the summary"
+
+def trace_end():
+	print_error_totals()
+
+def raw_syscalls__sys_exit(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, id, ret):
+	if (for_comm and common_comm != for_comm) or \
+	   (for_pid  and common_pid  != for_pid ):
+		return
+
+	if ret < 0:
+		try:
+			syscalls[common_comm][common_pid][id][ret] += 1
+		except TypeError:
+			syscalls[common_comm][common_pid][id][ret] = 1
+
+def syscalls__sys_exit(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, ret):
+	raw_syscalls__sys_exit(**locals())
+
+def print_error_totals():
+    if for_comm is not None:
+	    print "\nsyscall errors for %s:\n\n" % (for_comm),
+    else:
+	    print "\nsyscall errors:\n\n",
+
+    print "%-30s  %10s\n" % ("comm [pid]", "count"),
+    print "%-30s  %10s\n" % ("------------------------------", \
+                                 "----------"),
+
+    comm_keys = syscalls.keys()
+    for comm in comm_keys:
+	    pid_keys = syscalls[comm].keys()
+	    for pid in pid_keys:
+		    print "\n%s [%d]\n" % (comm, pid),
+		    id_keys = syscalls[comm][pid].keys()
+		    for id in id_keys:
+			    print "  syscall: %-16s\n" % syscall_name(id),
+			    ret_keys = syscalls[comm][pid][id].keys()
+			    for ret, val in sorted(syscalls[comm][pid][id].iteritems(), key = lambda(k, v): (v, k),  reverse = True):
+				    print "    err = %-20s  %10d\n" % (strerror(ret), val),
diff --git a/tools/perf/scripts/python/futex-contention.py b/tools/perf/scripts/python/futex-contention.py
new file mode 100644
index 000000000..0f5cf437b
--- /dev/null
+++ b/tools/perf/scripts/python/futex-contention.py
@@ -0,0 +1,50 @@
+# futex contention
+# (c) 2010, Arnaldo Carvalho de Melo <acme@redhat.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Translation of:
+#
+# http://sourceware.org/systemtap/wiki/WSFutexContention
+#
+# to perf python scripting.
+#
+# Measures futex contention
+
+import os, sys
+sys.path.append(os.environ['PERF_EXEC_PATH'] + '/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+from Util import *
+
+process_names = {}
+thread_thislock = {}
+thread_blocktime = {}
+
+lock_waits = {} # long-lived stats on (tid,lock) blockage elapsed time
+process_names = {} # long-lived pid-to-execname mapping
+
+def syscalls__sys_enter_futex(event, ctxt, cpu, s, ns, tid, comm, callchain,
+			      nr, uaddr, op, val, utime, uaddr2, val3):
+	cmd = op & FUTEX_CMD_MASK
+	if cmd != FUTEX_WAIT:
+		return # we don't care about originators of WAKE events
+
+	process_names[tid] = comm
+	thread_thislock[tid] = uaddr
+	thread_blocktime[tid] = nsecs(s, ns)
+
+def syscalls__sys_exit_futex(event, ctxt, cpu, s, ns, tid, comm, callchain,
+			     nr, ret):
+	if thread_blocktime.has_key(tid):
+		elapsed = nsecs(s, ns) - thread_blocktime[tid]
+		add_stats(lock_waits, (tid, thread_thislock[tid]), elapsed)
+		del thread_blocktime[tid]
+		del thread_thislock[tid]
+
+def trace_begin():
+	print "Press control+C to stop and show the summary"
+
+def trace_end():
+	for (tid, lock) in lock_waits:
+		min, max, avg, count = lock_waits[tid, lock]
+		print "%s[%d] lock %x contended %d times, %d avg ns" % \
+		      (process_names[tid], tid, lock, count, avg)
+
diff --git a/tools/perf/scripts/python/net_dropmonitor.py b/tools/perf/scripts/python/net_dropmonitor.py
new file mode 100755
index 000000000..0b6ce8c25
--- /dev/null
+++ b/tools/perf/scripts/python/net_dropmonitor.py
@@ -0,0 +1,75 @@
+# Monitor the system for dropped packets and proudce a report of drop locations and counts
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+		'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+drop_log = {}
+kallsyms = []
+
+def get_kallsyms_table():
+	global kallsyms
+
+	try:
+		f = open("/proc/kallsyms", "r")
+	except:
+		return
+
+	for line in f:
+		loc = int(line.split()[0], 16)
+		name = line.split()[2]
+		kallsyms.append((loc, name))
+	kallsyms.sort()
+
+def get_sym(sloc):
+	loc = int(sloc)
+
+	# Invariant: kallsyms[i][0] <= loc for all 0 <= i <= start
+	#            kallsyms[i][0] > loc for all end <= i < len(kallsyms)
+	start, end = -1, len(kallsyms)
+	while end != start + 1:
+		pivot = (start + end) // 2
+		if loc < kallsyms[pivot][0]:
+			end = pivot
+		else:
+			start = pivot
+
+	# Now (start == -1 or kallsyms[start][0] <= loc)
+	# and (start == len(kallsyms) - 1 or loc < kallsyms[start + 1][0])
+	if start >= 0:
+		symloc, name = kallsyms[start]
+		return (name, loc - symloc)
+	else:
+		return (None, 0)
+
+def print_drop_table():
+	print "%25s %25s %25s" % ("LOCATION", "OFFSET", "COUNT")
+	for i in drop_log.keys():
+		(sym, off) = get_sym(i)
+		if sym == None:
+			sym = i
+		print "%25s %25s %25s" % (sym, off, drop_log[i])
+
+
+def trace_begin():
+	print "Starting trace (Ctrl-C to dump results)"
+
+def trace_end():
+	print "Gathering kallsyms data"
+	get_kallsyms_table()
+	print_drop_table()
+
+# called from perf, when it finds a correspoinding event
+def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm, callchain,
+		   skbaddr, location, protocol):
+	slocation = str(location)
+	try:
+		drop_log[slocation] = drop_log[slocation] + 1
+	except:
+		drop_log[slocation] = 1
diff --git a/tools/perf/scripts/python/netdev-times.py b/tools/perf/scripts/python/netdev-times.py
new file mode 100644
index 000000000..4d21ef2d6
--- /dev/null
+++ b/tools/perf/scripts/python/netdev-times.py
@@ -0,0 +1,464 @@
+# Display a process of packets and processed time.
+# It helps us to investigate networking or network device.
+#
+# options
+# tx: show only tx chart
+# rx: show only rx chart
+# dev=: show only thing related to specified device
+# debug: work with debug mode. It shows buffer status.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+all_event_list = []; # insert all tracepoint event related with this script
+irq_dic = {}; # key is cpu and value is a list which stacks irqs
+              # which raise NET_RX softirq
+net_rx_dic = {}; # key is cpu and value include time of NET_RX softirq-entry
+		 # and a list which stacks receive
+receive_hunk_list = []; # a list which include a sequence of receive events
+rx_skb_list = []; # received packet list for matching
+		       # skb_copy_datagram_iovec
+
+buffer_budget = 65536; # the budget of rx_skb_list, tx_queue_list and
+		       # tx_xmit_list
+of_count_rx_skb_list = 0; # overflow count
+
+tx_queue_list = []; # list of packets which pass through dev_queue_xmit
+of_count_tx_queue_list = 0; # overflow count
+
+tx_xmit_list = [];  # list of packets which pass through dev_hard_start_xmit
+of_count_tx_xmit_list = 0; # overflow count
+
+tx_free_list = [];  # list of packets which is freed
+
+# options
+show_tx = 0;
+show_rx = 0;
+dev = 0; # store a name of device specified by option "dev="
+debug = 0;
+
+# indices of event_info tuple
+EINFO_IDX_NAME=   0
+EINFO_IDX_CONTEXT=1
+EINFO_IDX_CPU=    2
+EINFO_IDX_TIME=   3
+EINFO_IDX_PID=    4
+EINFO_IDX_COMM=   5
+
+# Calculate a time interval(msec) from src(nsec) to dst(nsec)
+def diff_msec(src, dst):
+	return (dst - src) / 1000000.0
+
+# Display a process of transmitting a packet
+def print_transmit(hunk):
+	if dev != 0 and hunk['dev'].find(dev) < 0:
+		return
+	print "%7s %5d %6d.%06dsec %12.3fmsec      %12.3fmsec" % \
+		(hunk['dev'], hunk['len'],
+		nsecs_secs(hunk['queue_t']),
+		nsecs_nsecs(hunk['queue_t'])/1000,
+		diff_msec(hunk['queue_t'], hunk['xmit_t']),
+		diff_msec(hunk['xmit_t'], hunk['free_t']))
+
+# Format for displaying rx packet processing
+PF_IRQ_ENTRY= "  irq_entry(+%.3fmsec irq=%d:%s)"
+PF_SOFT_ENTRY="  softirq_entry(+%.3fmsec)"
+PF_NAPI_POLL= "  napi_poll_exit(+%.3fmsec %s)"
+PF_JOINT=     "         |"
+PF_WJOINT=    "         |            |"
+PF_NET_RECV=  "         |---netif_receive_skb(+%.3fmsec skb=%x len=%d)"
+PF_NET_RX=    "         |---netif_rx(+%.3fmsec skb=%x)"
+PF_CPY_DGRAM= "         |      skb_copy_datagram_iovec(+%.3fmsec %d:%s)"
+PF_KFREE_SKB= "         |      kfree_skb(+%.3fmsec location=%x)"
+PF_CONS_SKB=  "         |      consume_skb(+%.3fmsec)"
+
+# Display a process of received packets and interrputs associated with
+# a NET_RX softirq
+def print_receive(hunk):
+	show_hunk = 0
+	irq_list = hunk['irq_list']
+	cpu = irq_list[0]['cpu']
+	base_t = irq_list[0]['irq_ent_t']
+	# check if this hunk should be showed
+	if dev != 0:
+		for i in range(len(irq_list)):
+			if irq_list[i]['name'].find(dev) >= 0:
+				show_hunk = 1
+				break
+	else:
+		show_hunk = 1
+	if show_hunk == 0:
+		return
+
+	print "%d.%06dsec cpu=%d" % \
+		(nsecs_secs(base_t), nsecs_nsecs(base_t)/1000, cpu)
+	for i in range(len(irq_list)):
+		print PF_IRQ_ENTRY % \
+			(diff_msec(base_t, irq_list[i]['irq_ent_t']),
+			irq_list[i]['irq'], irq_list[i]['name'])
+		print PF_JOINT
+		irq_event_list = irq_list[i]['event_list']
+		for j in range(len(irq_event_list)):
+			irq_event = irq_event_list[j]
+			if irq_event['event'] == 'netif_rx':
+				print PF_NET_RX % \
+					(diff_msec(base_t, irq_event['time']),
+					irq_event['skbaddr'])
+				print PF_JOINT
+	print PF_SOFT_ENTRY % \
+		diff_msec(base_t, hunk['sirq_ent_t'])
+	print PF_JOINT
+	event_list = hunk['event_list']
+	for i in range(len(event_list)):
+		event = event_list[i]
+		if event['event_name'] == 'napi_poll':
+			print PF_NAPI_POLL % \
+			    (diff_msec(base_t, event['event_t']), event['dev'])
+			if i == len(event_list) - 1:
+				print ""
+			else:
+				print PF_JOINT
+		else:
+			print PF_NET_RECV % \
+			    (diff_msec(base_t, event['event_t']), event['skbaddr'],
+				event['len'])
+			if 'comm' in event.keys():
+				print PF_WJOINT
+				print PF_CPY_DGRAM % \
+					(diff_msec(base_t, event['comm_t']),
+					event['pid'], event['comm'])
+			elif 'handle' in event.keys():
+				print PF_WJOINT
+				if event['handle'] == "kfree_skb":
+					print PF_KFREE_SKB % \
+						(diff_msec(base_t,
+						event['comm_t']),
+						event['location'])
+				elif event['handle'] == "consume_skb":
+					print PF_CONS_SKB % \
+						diff_msec(base_t,
+							event['comm_t'])
+			print PF_JOINT
+
+def trace_begin():
+	global show_tx
+	global show_rx
+	global dev
+	global debug
+
+	for i in range(len(sys.argv)):
+		if i == 0:
+			continue
+		arg = sys.argv[i]
+		if arg == 'tx':
+			show_tx = 1
+		elif arg =='rx':
+			show_rx = 1
+		elif arg.find('dev=',0, 4) >= 0:
+			dev = arg[4:]
+		elif arg == 'debug':
+			debug = 1
+	if show_tx == 0  and show_rx == 0:
+		show_tx = 1
+		show_rx = 1
+
+def trace_end():
+	# order all events in time
+	all_event_list.sort(lambda a,b :cmp(a[EINFO_IDX_TIME],
+					    b[EINFO_IDX_TIME]))
+	# process all events
+	for i in range(len(all_event_list)):
+		event_info = all_event_list[i]
+		name = event_info[EINFO_IDX_NAME]
+		if name == 'irq__softirq_exit':
+			handle_irq_softirq_exit(event_info)
+		elif name == 'irq__softirq_entry':
+			handle_irq_softirq_entry(event_info)
+		elif name == 'irq__softirq_raise':
+			handle_irq_softirq_raise(event_info)
+		elif name == 'irq__irq_handler_entry':
+			handle_irq_handler_entry(event_info)
+		elif name == 'irq__irq_handler_exit':
+			handle_irq_handler_exit(event_info)
+		elif name == 'napi__napi_poll':
+			handle_napi_poll(event_info)
+		elif name == 'net__netif_receive_skb':
+			handle_netif_receive_skb(event_info)
+		elif name == 'net__netif_rx':
+			handle_netif_rx(event_info)
+		elif name == 'skb__skb_copy_datagram_iovec':
+			handle_skb_copy_datagram_iovec(event_info)
+		elif name == 'net__net_dev_queue':
+			handle_net_dev_queue(event_info)
+		elif name == 'net__net_dev_xmit':
+			handle_net_dev_xmit(event_info)
+		elif name == 'skb__kfree_skb':
+			handle_kfree_skb(event_info)
+		elif name == 'skb__consume_skb':
+			handle_consume_skb(event_info)
+	# display receive hunks
+	if show_rx:
+		for i in range(len(receive_hunk_list)):
+			print_receive(receive_hunk_list[i])
+	# display transmit hunks
+	if show_tx:
+		print "   dev    len      Qdisc        " \
+			"       netdevice             free"
+		for i in range(len(tx_free_list)):
+			print_transmit(tx_free_list[i])
+	if debug:
+		print "debug buffer status"
+		print "----------------------------"
+		print "xmit Qdisc:remain:%d overflow:%d" % \
+			(len(tx_queue_list), of_count_tx_queue_list)
+		print "xmit netdevice:remain:%d overflow:%d" % \
+			(len(tx_xmit_list), of_count_tx_xmit_list)
+		print "receive:remain:%d overflow:%d" % \
+			(len(rx_skb_list), of_count_rx_skb_list)
+
+# called from perf, when it finds a correspoinding event
+def irq__softirq_entry(name, context, cpu, sec, nsec, pid, comm, callchain, vec):
+	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+		return
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+	all_event_list.append(event_info)
+
+def irq__softirq_exit(name, context, cpu, sec, nsec, pid, comm, callchain, vec):
+	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+		return
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+	all_event_list.append(event_info)
+
+def irq__softirq_raise(name, context, cpu, sec, nsec, pid, comm, callchain, vec):
+	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+		return
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+	all_event_list.append(event_info)
+
+def irq__irq_handler_entry(name, context, cpu, sec, nsec, pid, comm,
+			callchain, irq, irq_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			irq, irq_name)
+	all_event_list.append(event_info)
+
+def irq__irq_handler_exit(name, context, cpu, sec, nsec, pid, comm, callchain, irq, ret):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, irq, ret)
+	all_event_list.append(event_info)
+
+def napi__napi_poll(name, context, cpu, sec, nsec, pid, comm, callchain, napi, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			napi, dev_name)
+	all_event_list.append(event_info)
+
+def net__netif_receive_skb(name, context, cpu, sec, nsec, pid, comm, callchain, skbaddr,
+			skblen, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, dev_name)
+	all_event_list.append(event_info)
+
+def net__netif_rx(name, context, cpu, sec, nsec, pid, comm, callchain, skbaddr,
+			skblen, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, dev_name)
+	all_event_list.append(event_info)
+
+def net__net_dev_queue(name, context, cpu, sec, nsec, pid, comm, callchain,
+			skbaddr, skblen, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, dev_name)
+	all_event_list.append(event_info)
+
+def net__net_dev_xmit(name, context, cpu, sec, nsec, pid, comm, callchain,
+			skbaddr, skblen, rc, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, rc ,dev_name)
+	all_event_list.append(event_info)
+
+def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm, callchain,
+			skbaddr, protocol, location):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, protocol, location)
+	all_event_list.append(event_info)
+
+def skb__consume_skb(name, context, cpu, sec, nsec, pid, comm, callchain, skbaddr):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr)
+	all_event_list.append(event_info)
+
+def skb__skb_copy_datagram_iovec(name, context, cpu, sec, nsec, pid, comm, callchain,
+	skbaddr, skblen):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen)
+	all_event_list.append(event_info)
+
+def handle_irq_handler_entry(event_info):
+	(name, context, cpu, time, pid, comm, irq, irq_name) = event_info
+	if cpu not in irq_dic.keys():
+		irq_dic[cpu] = []
+	irq_record = {'irq':irq, 'name':irq_name, 'cpu':cpu, 'irq_ent_t':time}
+	irq_dic[cpu].append(irq_record)
+
+def handle_irq_handler_exit(event_info):
+	(name, context, cpu, time, pid, comm, irq, ret) = event_info
+	if cpu not in irq_dic.keys():
+		return
+	irq_record = irq_dic[cpu].pop()
+	if irq != irq_record['irq']:
+		return
+	irq_record.update({'irq_ext_t':time})
+	# if an irq doesn't include NET_RX softirq, drop.
+	if 'event_list' in irq_record.keys():
+		irq_dic[cpu].append(irq_record)
+
+def handle_irq_softirq_raise(event_info):
+	(name, context, cpu, time, pid, comm, vec) = event_info
+	if cpu not in irq_dic.keys() \
+	or len(irq_dic[cpu]) == 0:
+		return
+	irq_record = irq_dic[cpu].pop()
+	if 'event_list' in irq_record.keys():
+		irq_event_list = irq_record['event_list']
+	else:
+		irq_event_list = []
+	irq_event_list.append({'time':time, 'event':'sirq_raise'})
+	irq_record.update({'event_list':irq_event_list})
+	irq_dic[cpu].append(irq_record)
+
+def handle_irq_softirq_entry(event_info):
+	(name, context, cpu, time, pid, comm, vec) = event_info
+	net_rx_dic[cpu] = {'sirq_ent_t':time, 'event_list':[]}
+
+def handle_irq_softirq_exit(event_info):
+	(name, context, cpu, time, pid, comm, vec) = event_info
+	irq_list = []
+	event_list = 0
+	if cpu in irq_dic.keys():
+		irq_list = irq_dic[cpu]
+		del irq_dic[cpu]
+	if cpu in net_rx_dic.keys():
+		sirq_ent_t = net_rx_dic[cpu]['sirq_ent_t']
+		event_list = net_rx_dic[cpu]['event_list']
+		del net_rx_dic[cpu]
+	if irq_list == [] or event_list == 0:
+		return
+	rec_data = {'sirq_ent_t':sirq_ent_t, 'sirq_ext_t':time,
+		    'irq_list':irq_list, 'event_list':event_list}
+	# merge information realted to a NET_RX softirq
+	receive_hunk_list.append(rec_data)
+
+def handle_napi_poll(event_info):
+	(name, context, cpu, time, pid, comm, napi, dev_name) = event_info
+	if cpu in net_rx_dic.keys():
+		event_list = net_rx_dic[cpu]['event_list']
+		rec_data = {'event_name':'napi_poll',
+				'dev':dev_name, 'event_t':time}
+		event_list.append(rec_data)
+
+def handle_netif_rx(event_info):
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, dev_name) = event_info
+	if cpu not in irq_dic.keys() \
+	or len(irq_dic[cpu]) == 0:
+		return
+	irq_record = irq_dic[cpu].pop()
+	if 'event_list' in irq_record.keys():
+		irq_event_list = irq_record['event_list']
+	else:
+		irq_event_list = []
+	irq_event_list.append({'time':time, 'event':'netif_rx',
+		'skbaddr':skbaddr, 'skblen':skblen, 'dev_name':dev_name})
+	irq_record.update({'event_list':irq_event_list})
+	irq_dic[cpu].append(irq_record)
+
+def handle_netif_receive_skb(event_info):
+	global of_count_rx_skb_list
+
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, dev_name) = event_info
+	if cpu in net_rx_dic.keys():
+		rec_data = {'event_name':'netif_receive_skb',
+			    'event_t':time, 'skbaddr':skbaddr, 'len':skblen}
+		event_list = net_rx_dic[cpu]['event_list']
+		event_list.append(rec_data)
+		rx_skb_list.insert(0, rec_data)
+		if len(rx_skb_list) > buffer_budget:
+			rx_skb_list.pop()
+			of_count_rx_skb_list += 1
+
+def handle_net_dev_queue(event_info):
+	global of_count_tx_queue_list
+
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, dev_name) = event_info
+	skb = {'dev':dev_name, 'skbaddr':skbaddr, 'len':skblen, 'queue_t':time}
+	tx_queue_list.insert(0, skb)
+	if len(tx_queue_list) > buffer_budget:
+		tx_queue_list.pop()
+		of_count_tx_queue_list += 1
+
+def handle_net_dev_xmit(event_info):
+	global of_count_tx_xmit_list
+
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, rc, dev_name) = event_info
+	if rc == 0: # NETDEV_TX_OK
+		for i in range(len(tx_queue_list)):
+			skb = tx_queue_list[i]
+			if skb['skbaddr'] == skbaddr:
+				skb['xmit_t'] = time
+				tx_xmit_list.insert(0, skb)
+				del tx_queue_list[i]
+				if len(tx_xmit_list) > buffer_budget:
+					tx_xmit_list.pop()
+					of_count_tx_xmit_list += 1
+				return
+
+def handle_kfree_skb(event_info):
+	(name, context, cpu, time, pid, comm,
+		skbaddr, protocol, location) = event_info
+	for i in range(len(tx_queue_list)):
+		skb = tx_queue_list[i]
+		if skb['skbaddr'] == skbaddr:
+			del tx_queue_list[i]
+			return
+	for i in range(len(tx_xmit_list)):
+		skb = tx_xmit_list[i]
+		if skb['skbaddr'] == skbaddr:
+			skb['free_t'] = time
+			tx_free_list.append(skb)
+			del tx_xmit_list[i]
+			return
+	for i in range(len(rx_skb_list)):
+		rec_data = rx_skb_list[i]
+		if rec_data['skbaddr'] == skbaddr:
+			rec_data.update({'handle':"kfree_skb",
+					'comm':comm, 'pid':pid, 'comm_t':time})
+			del rx_skb_list[i]
+			return
+
+def handle_consume_skb(event_info):
+	(name, context, cpu, time, pid, comm, skbaddr) = event_info
+	for i in range(len(tx_xmit_list)):
+		skb = tx_xmit_list[i]
+		if skb['skbaddr'] == skbaddr:
+			skb['free_t'] = time
+			tx_free_list.append(skb)
+			del tx_xmit_list[i]
+			return
+
+def handle_skb_copy_datagram_iovec(event_info):
+	(name, context, cpu, time, pid, comm, skbaddr, skblen) = event_info
+	for i in range(len(rx_skb_list)):
+		rec_data = rx_skb_list[i]
+		if skbaddr == rec_data['skbaddr']:
+			rec_data.update({'handle':"skb_copy_datagram_iovec",
+					'comm':comm, 'pid':pid, 'comm_t':time})
+			del rx_skb_list[i]
+			return
diff --git a/tools/perf/scripts/python/sched-migration.py b/tools/perf/scripts/python/sched-migration.py
new file mode 100644
index 000000000..de66cb3b7
--- /dev/null
+++ b/tools/perf/scripts/python/sched-migration.py
@@ -0,0 +1,460 @@
+#!/usr/bin/python
+#
+# Cpu task migration overview toy
+#
+# Copyright (C) 2010 Frederic Weisbecker <fweisbec@gmail.com>
+#
+# perf script event handlers have been generated by perf script -g python
+#
+# This software is distributed under the terms of the GNU General
+# Public License ("GPL") version 2 as published by the Free Software
+# Foundation.
+
+
+import os
+import sys
+
+from collections import defaultdict
+from UserList import UserList
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+sys.path.append('scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from SchedGui import *
+
+
+threads = { 0 : "idle"}
+
+def thread_name(pid):
+	return "%s:%d" % (threads[pid], pid)
+
+class RunqueueEventUnknown:
+	@staticmethod
+	def color():
+		return None
+
+	def __repr__(self):
+		return "unknown"
+
+class RunqueueEventSleep:
+	@staticmethod
+	def color():
+		return (0, 0, 0xff)
+
+	def __init__(self, sleeper):
+		self.sleeper = sleeper
+
+	def __repr__(self):
+		return "%s gone to sleep" % thread_name(self.sleeper)
+
+class RunqueueEventWakeup:
+	@staticmethod
+	def color():
+		return (0xff, 0xff, 0)
+
+	def __init__(self, wakee):
+		self.wakee = wakee
+
+	def __repr__(self):
+		return "%s woke up" % thread_name(self.wakee)
+
+class RunqueueEventFork:
+	@staticmethod
+	def color():
+		return (0, 0xff, 0)
+
+	def __init__(self, child):
+		self.child = child
+
+	def __repr__(self):
+		return "new forked task %s" % thread_name(self.child)
+
+class RunqueueMigrateIn:
+	@staticmethod
+	def color():
+		return (0, 0xf0, 0xff)
+
+	def __init__(self, new):
+		self.new = new
+
+	def __repr__(self):
+		return "task migrated in %s" % thread_name(self.new)
+
+class RunqueueMigrateOut:
+	@staticmethod
+	def color():
+		return (0xff, 0, 0xff)
+
+	def __init__(self, old):
+		self.old = old
+
+	def __repr__(self):
+		return "task migrated out %s" % thread_name(self.old)
+
+class RunqueueSnapshot:
+	def __init__(self, tasks = [0], event = RunqueueEventUnknown()):
+		self.tasks = tuple(tasks)
+		self.event = event
+
+	def sched_switch(self, prev, prev_state, next):
+		event = RunqueueEventUnknown()
+
+		if taskState(prev_state) == "R" and next in self.tasks \
+			and prev in self.tasks:
+			return self
+
+		if taskState(prev_state) != "R":
+			event = RunqueueEventSleep(prev)
+
+		next_tasks = list(self.tasks[:])
+		if prev in self.tasks:
+			if taskState(prev_state) != "R":
+				next_tasks.remove(prev)
+		elif taskState(prev_state) == "R":
+			next_tasks.append(prev)
+
+		if next not in next_tasks:
+			next_tasks.append(next)
+
+		return RunqueueSnapshot(next_tasks, event)
+
+	def migrate_out(self, old):
+		if old not in self.tasks:
+			return self
+		next_tasks = [task for task in self.tasks if task != old]
+
+		return RunqueueSnapshot(next_tasks, RunqueueMigrateOut(old))
+
+	def __migrate_in(self, new, event):
+		if new in self.tasks:
+			self.event = event
+			return self
+		next_tasks = self.tasks[:] + tuple([new])
+
+		return RunqueueSnapshot(next_tasks, event)
+
+	def migrate_in(self, new):
+		return self.__migrate_in(new, RunqueueMigrateIn(new))
+
+	def wake_up(self, new):
+		return self.__migrate_in(new, RunqueueEventWakeup(new))
+
+	def wake_up_new(self, new):
+		return self.__migrate_in(new, RunqueueEventFork(new))
+
+	def load(self):
+		""" Provide the number of tasks on the runqueue.
+		    Don't count idle"""
+		return len(self.tasks) - 1
+
+	def __repr__(self):
+		ret = self.tasks.__repr__()
+		ret += self.origin_tostring()
+
+		return ret
+
+class TimeSlice:
+	def __init__(self, start, prev):
+		self.start = start
+		self.prev = prev
+		self.end = start
+		# cpus that triggered the event
+		self.event_cpus = []
+		if prev is not None:
+			self.total_load = prev.total_load
+			self.rqs = prev.rqs.copy()
+		else:
+			self.rqs = defaultdict(RunqueueSnapshot)
+			self.total_load = 0
+
+	def __update_total_load(self, old_rq, new_rq):
+		diff = new_rq.load() - old_rq.load()
+		self.total_load += diff
+
+	def sched_switch(self, ts_list, prev, prev_state, next, cpu):
+		old_rq = self.prev.rqs[cpu]
+		new_rq = old_rq.sched_switch(prev, prev_state, next)
+
+		if old_rq is new_rq:
+			return
+
+		self.rqs[cpu] = new_rq
+		self.__update_total_load(old_rq, new_rq)
+		ts_list.append(self)
+		self.event_cpus = [cpu]
+
+	def migrate(self, ts_list, new, old_cpu, new_cpu):
+		if old_cpu == new_cpu:
+			return
+		old_rq = self.prev.rqs[old_cpu]
+		out_rq = old_rq.migrate_out(new)
+		self.rqs[old_cpu] = out_rq
+		self.__update_total_load(old_rq, out_rq)
+
+		new_rq = self.prev.rqs[new_cpu]
+		in_rq = new_rq.migrate_in(new)
+		self.rqs[new_cpu] = in_rq
+		self.__update_total_load(new_rq, in_rq)
+
+		ts_list.append(self)
+
+		if old_rq is not out_rq:
+			self.event_cpus.append(old_cpu)
+		self.event_cpus.append(new_cpu)
+
+	def wake_up(self, ts_list, pid, cpu, fork):
+		old_rq = self.prev.rqs[cpu]
+		if fork:
+			new_rq = old_rq.wake_up_new(pid)
+		else:
+			new_rq = old_rq.wake_up(pid)
+
+		if new_rq is old_rq:
+			return
+		self.rqs[cpu] = new_rq
+		self.__update_total_load(old_rq, new_rq)
+		ts_list.append(self)
+		self.event_cpus = [cpu]
+
+	def next(self, t):
+		self.end = t
+		return TimeSlice(t, self)
+
+class TimeSliceList(UserList):
+	def __init__(self, arg = []):
+		self.data = arg
+
+	def get_time_slice(self, ts):
+		if len(self.data) == 0:
+			slice = TimeSlice(ts, TimeSlice(-1, None))
+		else:
+			slice = self.data[-1].next(ts)
+		return slice
+
+	def find_time_slice(self, ts):
+		start = 0
+		end = len(self.data)
+		found = -1
+		searching = True
+		while searching:
+			if start == end or start == end - 1:
+				searching = False
+
+			i = (end + start) / 2
+			if self.data[i].start <= ts and self.data[i].end >= ts:
+				found = i
+				end = i
+				continue
+
+			if self.data[i].end < ts:
+				start = i
+
+			elif self.data[i].start > ts:
+				end = i
+
+		return found
+
+	def set_root_win(self, win):
+		self.root_win = win
+
+	def mouse_down(self, cpu, t):
+		idx = self.find_time_slice(t)
+		if idx == -1:
+			return
+
+		ts = self[idx]
+		rq = ts.rqs[cpu]
+		raw = "CPU: %d\n" % cpu
+		raw += "Last event : %s\n" % rq.event.__repr__()
+		raw += "Timestamp : %d.%06d\n" % (ts.start / (10 ** 9), (ts.start % (10 ** 9)) / 1000)
+		raw += "Duration : %6d us\n" % ((ts.end - ts.start) / (10 ** 6))
+		raw += "Load = %d\n" % rq.load()
+		for t in rq.tasks:
+			raw += "%s \n" % thread_name(t)
+
+		self.root_win.update_summary(raw)
+
+	def update_rectangle_cpu(self, slice, cpu):
+		rq = slice.rqs[cpu]
+
+		if slice.total_load != 0:
+			load_rate = rq.load() / float(slice.total_load)
+		else:
+			load_rate = 0
+
+		red_power = int(0xff - (0xff * load_rate))
+		color = (0xff, red_power, red_power)
+
+		top_color = None
+
+		if cpu in slice.event_cpus:
+			top_color = rq.event.color()
+
+		self.root_win.paint_rectangle_zone(cpu, color, top_color, slice.start, slice.end)
+
+	def fill_zone(self, start, end):
+		i = self.find_time_slice(start)
+		if i == -1:
+			return
+
+		for i in xrange(i, len(self.data)):
+			timeslice = self.data[i]
+			if timeslice.start > end:
+				return
+
+			for cpu in timeslice.rqs:
+				self.update_rectangle_cpu(timeslice, cpu)
+
+	def interval(self):
+		if len(self.data) == 0:
+			return (0, 0)
+
+		return (self.data[0].start, self.data[-1].end)
+
+	def nr_rectangles(self):
+		last_ts = self.data[-1]
+		max_cpu = 0
+		for cpu in last_ts.rqs:
+			if cpu > max_cpu:
+				max_cpu = cpu
+		return max_cpu
+
+
+class SchedEventProxy:
+	def __init__(self):
+		self.current_tsk = defaultdict(lambda : -1)
+		self.timeslices = TimeSliceList()
+
+	def sched_switch(self, headers, prev_comm, prev_pid, prev_prio, prev_state,
+			 next_comm, next_pid, next_prio):
+		""" Ensure the task we sched out this cpu is really the one
+		    we logged. Otherwise we may have missed traces """
+
+		on_cpu_task = self.current_tsk[headers.cpu]
+
+		if on_cpu_task != -1 and on_cpu_task != prev_pid:
+			print "Sched switch event rejected ts: %s cpu: %d prev: %s(%d) next: %s(%d)" % \
+				(headers.ts_format(), headers.cpu, prev_comm, prev_pid, next_comm, next_pid)
+
+		threads[prev_pid] = prev_comm
+		threads[next_pid] = next_comm
+		self.current_tsk[headers.cpu] = next_pid
+
+		ts = self.timeslices.get_time_slice(headers.ts())
+		ts.sched_switch(self.timeslices, prev_pid, prev_state, next_pid, headers.cpu)
+
+	def migrate(self, headers, pid, prio, orig_cpu, dest_cpu):
+		ts = self.timeslices.get_time_slice(headers.ts())
+		ts.migrate(self.timeslices, pid, orig_cpu, dest_cpu)
+
+	def wake_up(self, headers, comm, pid, success, target_cpu, fork):
+		if success == 0:
+			return
+		ts = self.timeslices.get_time_slice(headers.ts())
+		ts.wake_up(self.timeslices, pid, target_cpu, fork)
+
+
+def trace_begin():
+	global parser
+	parser = SchedEventProxy()
+
+def trace_end():
+	app = wx.App(False)
+	timeslices = parser.timeslices
+	frame = RootFrame(timeslices, "Migration")
+	app.MainLoop()
+
+def sched__sched_stat_runtime(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, runtime, vruntime):
+	pass
+
+def sched__sched_stat_iowait(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, delay):
+	pass
+
+def sched__sched_stat_sleep(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, delay):
+	pass
+
+def sched__sched_stat_wait(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, delay):
+	pass
+
+def sched__sched_process_fork(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, parent_comm, parent_pid, child_comm, child_pid):
+	pass
+
+def sched__sched_process_wait(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, prio):
+	pass
+
+def sched__sched_process_exit(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, prio):
+	pass
+
+def sched__sched_process_free(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, prio):
+	pass
+
+def sched__sched_migrate_task(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, prio, orig_cpu,
+	dest_cpu):
+	headers = EventHeaders(common_cpu, common_secs, common_nsecs,
+				common_pid, common_comm, common_callchain)
+	parser.migrate(headers, pid, prio, orig_cpu, dest_cpu)
+
+def sched__sched_switch(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm, common_callchain,
+	prev_comm, prev_pid, prev_prio, prev_state,
+	next_comm, next_pid, next_prio):
+
+	headers = EventHeaders(common_cpu, common_secs, common_nsecs,
+				common_pid, common_comm, common_callchain)
+	parser.sched_switch(headers, prev_comm, prev_pid, prev_prio, prev_state,
+			 next_comm, next_pid, next_prio)
+
+def sched__sched_wakeup_new(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, prio, success,
+	target_cpu):
+	headers = EventHeaders(common_cpu, common_secs, common_nsecs,
+				common_pid, common_comm, common_callchain)
+	parser.wake_up(headers, comm, pid, success, target_cpu, 1)
+
+def sched__sched_wakeup(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, prio, success,
+	target_cpu):
+	headers = EventHeaders(common_cpu, common_secs, common_nsecs,
+				common_pid, common_comm, common_callchain)
+	parser.wake_up(headers, comm, pid, success, target_cpu, 0)
+
+def sched__sched_wait_task(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid, prio):
+	pass
+
+def sched__sched_kthread_stop_ret(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, ret):
+	pass
+
+def sched__sched_kthread_stop(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, comm, pid):
+	pass
+
+def trace_unhandled(event_name, context, event_fields_dict):
+	pass
diff --git a/tools/perf/scripts/python/sctop.py b/tools/perf/scripts/python/sctop.py
new file mode 100644
index 000000000..61621b93a
--- /dev/null
+++ b/tools/perf/scripts/python/sctop.py
@@ -0,0 +1,80 @@
+# system call top
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Periodically displays system-wide system call totals, broken down by
+# syscall.  If a [comm] arg is specified, only syscalls called by
+# [comm] are displayed. If an [interval] arg is specified, the display
+# will be refreshed every [interval] seconds.  The default interval is
+# 3 seconds.
+
+import os, sys, thread, time
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+usage = "perf script -s sctop.py [comm] [interval]\n";
+
+for_comm = None
+default_interval = 3
+interval = default_interval
+
+if len(sys.argv) > 3:
+	sys.exit(usage)
+
+if len(sys.argv) > 2:
+	for_comm = sys.argv[1]
+	interval = int(sys.argv[2])
+elif len(sys.argv) > 1:
+	try:
+		interval = int(sys.argv[1])
+	except ValueError:
+		for_comm = sys.argv[1]
+		interval = default_interval
+
+syscalls = autodict()
+
+def trace_begin():
+	thread.start_new_thread(print_syscall_totals, (interval,))
+	pass
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, id, args):
+	if for_comm is not None:
+		if common_comm != for_comm:
+			return
+	try:
+		syscalls[id] += 1
+	except TypeError:
+		syscalls[id] = 1
+
+def syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, args):
+	raw_syscalls__sys_enter(**locals())
+
+def print_syscall_totals(interval):
+	while 1:
+		clear_term()
+		if for_comm is not None:
+			print "\nsyscall events for %s:\n\n" % (for_comm),
+		else:
+			print "\nsyscall events:\n\n",
+
+		print "%-40s  %10s\n" % ("event", "count"),
+		print "%-40s  %10s\n" % ("----------------------------------------", \
+						 "----------"),
+
+		for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \
+					      reverse = True):
+			try:
+				print "%-40s  %10d\n" % (syscall_name(id), val),
+			except TypeError:
+				pass
+		syscalls.clear()
+		time.sleep(interval)
diff --git a/tools/perf/scripts/python/syscall-counts-by-pid.py b/tools/perf/scripts/python/syscall-counts-by-pid.py
new file mode 100644
index 000000000..daf314cc5
--- /dev/null
+++ b/tools/perf/scripts/python/syscall-counts-by-pid.py
@@ -0,0 +1,74 @@
+# system call counts, by pid
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Displays system-wide system call totals, broken down by syscall.
+# If a [comm] arg is specified, only syscalls called by [comm] are displayed.
+
+import os, sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import syscall_name
+
+usage = "perf script -s syscall-counts-by-pid.py [comm]\n";
+
+for_comm = None
+for_pid = None
+
+if len(sys.argv) > 2:
+	sys.exit(usage)
+
+if len(sys.argv) > 1:
+	try:
+		for_pid = int(sys.argv[1])
+	except:
+		for_comm = sys.argv[1]
+
+syscalls = autodict()
+
+def trace_begin():
+	print "Press control+C to stop and show the summary"
+
+def trace_end():
+	print_syscall_totals()
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, id, args):
+
+	if (for_comm and common_comm != for_comm) or \
+	   (for_pid  and common_pid  != for_pid ):
+		return
+	try:
+		syscalls[common_comm][common_pid][id] += 1
+	except TypeError:
+		syscalls[common_comm][common_pid][id] = 1
+
+def syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, args):
+	raw_syscalls__sys_enter(**locals())
+
+def print_syscall_totals():
+    if for_comm is not None:
+	    print "\nsyscall events for %s:\n\n" % (for_comm),
+    else:
+	    print "\nsyscall events by comm/pid:\n\n",
+
+    print "%-40s  %10s\n" % ("comm [pid]/syscalls", "count"),
+    print "%-40s  %10s\n" % ("----------------------------------------", \
+                                 "----------"),
+
+    comm_keys = syscalls.keys()
+    for comm in comm_keys:
+	    pid_keys = syscalls[comm].keys()
+	    for pid in pid_keys:
+		    print "\n%s [%d]\n" % (comm, pid),
+		    id_keys = syscalls[comm][pid].keys()
+		    for id, val in sorted(syscalls[comm][pid].iteritems(), \
+				  key = lambda(k, v): (v, k),  reverse = True):
+			    print "  %-38s  %10d\n" % (syscall_name(id), val),
diff --git a/tools/perf/scripts/python/syscall-counts.py b/tools/perf/scripts/python/syscall-counts.py
new file mode 100644
index 000000000..e66a7730a
--- /dev/null
+++ b/tools/perf/scripts/python/syscall-counts.py
@@ -0,0 +1,64 @@
+# system call counts
+# (c) 2010, Tom Zanussi <tzanussi@gmail.com>
+# Licensed under the terms of the GNU GPL License version 2
+#
+# Displays system-wide system call totals, broken down by syscall.
+# If a [comm] arg is specified, only syscalls called by [comm] are displayed.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import syscall_name
+
+usage = "perf script -s syscall-counts.py [comm]\n";
+
+for_comm = None
+
+if len(sys.argv) > 2:
+	sys.exit(usage)
+
+if len(sys.argv) > 1:
+	for_comm = sys.argv[1]
+
+syscalls = autodict()
+
+def trace_begin():
+	print "Press control+C to stop and show the summary"
+
+def trace_end():
+	print_syscall_totals()
+
+def raw_syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	common_callchain, id, args):
+	if for_comm is not None:
+		if common_comm != for_comm:
+			return
+	try:
+		syscalls[id] += 1
+	except TypeError:
+		syscalls[id] = 1
+
+def syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, args):
+	raw_syscalls__sys_enter(**locals())
+
+def print_syscall_totals():
+    if for_comm is not None:
+	    print "\nsyscall events for %s:\n\n" % (for_comm),
+    else:
+	    print "\nsyscall events:\n\n",
+
+    print "%-40s  %10s\n" % ("event", "count"),
+    print "%-40s  %10s\n" % ("----------------------------------------", \
+                                 "-----------"),
+
+    for id, val in sorted(syscalls.iteritems(), key = lambda(k, v): (v, k), \
+				  reverse = True):
+	    print "%-40s  %10d\n" % (syscall_name(id), val),
diff --git a/tools/perf/tests/Build b/tools/perf/tests/Build
new file mode 100644
index 000000000..6a8801b32
--- /dev/null
+++ b/tools/perf/tests/Build
@@ -0,0 +1,43 @@
+perf-y += builtin-test.o
+perf-y += parse-events.o
+perf-y += dso-data.o
+perf-y += attr.o
+perf-y += vmlinux-kallsyms.o
+perf-y += open-syscall.o
+perf-y += open-syscall-all-cpus.o
+perf-y += open-syscall-tp-fields.o
+perf-y += mmap-basic.o
+perf-y += perf-record.o
+perf-y += rdpmc.o
+perf-y += evsel-roundtrip-name.o
+perf-y += evsel-tp-sched.o
+perf-y += fdarray.o
+perf-y += pmu.o
+perf-y += hists_common.o
+perf-y += hists_link.o
+perf-y += hists_filter.o
+perf-y += hists_output.o
+perf-y += hists_cumulate.o
+perf-y += python-use.o
+perf-y += bp_signal.o
+perf-y += bp_signal_overflow.o
+perf-y += task-exit.o
+perf-y += sw-clock.o
+perf-y += mmap-thread-lookup.o
+perf-y += thread-mg-share.o
+perf-y += switch-tracking.o
+perf-y += keep-tracking.o
+perf-y += code-reading.o
+perf-y += sample-parsing.o
+perf-y += parse-no-sample-id-all.o
+perf-y += kmod-path.o
+
+perf-$(CONFIG_X86) += perf-time-to-tsc.o
+
+ifeq ($(ARCH),$(filter $(ARCH),x86 arm))
+perf-$(CONFIG_DWARF_UNWIND) += dwarf-unwind.o
+endif
+
+CFLAGS_attr.o         += -DBINDIR="BUILD_STR($(bindir_SQ))" -DPYTHON="BUILD_STR($(PYTHON_WORD))"
+CFLAGS_python-use.o   += -DPYTHONPATH="BUILD_STR($(OUTPUT)python)" -DPYTHON="BUILD_STR($(PYTHON_WORD))"
+CFLAGS_dwarf-unwind.o += -fno-optimize-sibling-calls
diff --git a/tools/perf/tests/attr.c b/tools/perf/tests/attr.c
new file mode 100644
index 000000000..2dfc9ad0e
--- /dev/null
+++ b/tools/perf/tests/attr.c
@@ -0,0 +1,176 @@
+/*
+ * The struct perf_event_attr test support.
+ *
+ * This test is embedded inside into perf directly and is governed
+ * by the PERF_TEST_ATTR environment variable and hook inside
+ * sys_perf_event_open function.
+ *
+ * The general idea is to store 'struct perf_event_attr' details for
+ * each event created within single perf command. Each event details
+ * are stored into separate text file. Once perf command is finished
+ * these files can be checked for values we expect for command.
+ *
+ * Besides 'struct perf_event_attr' values we also store 'fd' and
+ * 'group_fd' values to allow checking for groups created.
+ *
+ * This all is triggered by setting PERF_TEST_ATTR environment variable.
+ * It must contain name of existing directory with access and write
+ * permissions. All the event text files are stored there.
+ */
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include "../perf.h"
+#include "util.h"
+#include "exec_cmd.h"
+#include "tests.h"
+
+#define ENV "PERF_TEST_ATTR"
+
+extern int verbose;
+
+static char *dir;
+
+void test_attr__init(void)
+{
+	dir = getenv(ENV);
+	test_attr__enabled = (dir != NULL);
+}
+
+#define BUFSIZE 1024
+
+#define __WRITE_ASS(str, fmt, data)					\
+do {									\
+	char buf[BUFSIZE];						\
+	size_t size;							\
+									\
+	size = snprintf(buf, BUFSIZE, #str "=%"fmt "\n", data);		\
+	if (1 != fwrite(buf, size, 1, file)) {				\
+		perror("test attr - failed to write event file");	\
+		fclose(file);						\
+		return -1;						\
+	}								\
+									\
+} while (0)
+
+#define WRITE_ASS(field, fmt) __WRITE_ASS(field, fmt, attr->field)
+
+static int store_event(struct perf_event_attr *attr, pid_t pid, int cpu,
+		       int fd, int group_fd, unsigned long flags)
+{
+	FILE *file;
+	char path[PATH_MAX];
+
+	snprintf(path, PATH_MAX, "%s/event-%d-%llu-%d", dir,
+		 attr->type, attr->config, fd);
+
+	file = fopen(path, "w+");
+	if (!file) {
+		perror("test attr - failed to open event file");
+		return -1;
+	}
+
+	if (fprintf(file, "[event-%d-%llu-%d]\n",
+		    attr->type, attr->config, fd) < 0) {
+		perror("test attr - failed to write event file");
+		fclose(file);
+		return -1;
+	}
+
+	/* syscall arguments */
+	__WRITE_ASS(fd,       "d", fd);
+	__WRITE_ASS(group_fd, "d", group_fd);
+	__WRITE_ASS(cpu,      "d", cpu);
+	__WRITE_ASS(pid,      "d", pid);
+	__WRITE_ASS(flags,   "lu", flags);
+
+	/* struct perf_event_attr */
+	WRITE_ASS(type,   PRIu32);
+	WRITE_ASS(size,   PRIu32);
+	WRITE_ASS(config,  "llu");
+	WRITE_ASS(sample_period, "llu");
+	WRITE_ASS(sample_type,   "llu");
+	WRITE_ASS(read_format,   "llu");
+	WRITE_ASS(disabled,       "d");
+	WRITE_ASS(inherit,        "d");
+	WRITE_ASS(pinned,         "d");
+	WRITE_ASS(exclusive,      "d");
+	WRITE_ASS(exclude_user,   "d");
+	WRITE_ASS(exclude_kernel, "d");
+	WRITE_ASS(exclude_hv,     "d");
+	WRITE_ASS(exclude_idle,   "d");
+	WRITE_ASS(mmap,           "d");
+	WRITE_ASS(comm,           "d");
+	WRITE_ASS(freq,           "d");
+	WRITE_ASS(inherit_stat,   "d");
+	WRITE_ASS(enable_on_exec, "d");
+	WRITE_ASS(task,           "d");
+	WRITE_ASS(watermark,      "d");
+	WRITE_ASS(precise_ip,     "d");
+	WRITE_ASS(mmap_data,      "d");
+	WRITE_ASS(sample_id_all,  "d");
+	WRITE_ASS(exclude_host,   "d");
+	WRITE_ASS(exclude_guest,  "d");
+	WRITE_ASS(exclude_callchain_kernel, "d");
+	WRITE_ASS(exclude_callchain_user, "d");
+	WRITE_ASS(wakeup_events, PRIu32);
+	WRITE_ASS(bp_type, PRIu32);
+	WRITE_ASS(config1, "llu");
+	WRITE_ASS(config2, "llu");
+	WRITE_ASS(branch_sample_type, "llu");
+	WRITE_ASS(sample_regs_user,   "llu");
+	WRITE_ASS(sample_stack_user,  PRIu32);
+
+	fclose(file);
+	return 0;
+}
+
+void test_attr__open(struct perf_event_attr *attr, pid_t pid, int cpu,
+		     int fd, int group_fd, unsigned long flags)
+{
+	int errno_saved = errno;
+
+	if (store_event(attr, pid, cpu, fd, group_fd, flags))
+		die("test attr FAILED");
+
+	errno = errno_saved;
+}
+
+static int run_dir(const char *d, const char *perf)
+{
+	char v[] = "-vvvvv";
+	int vcnt = min(verbose, (int) sizeof(v) - 1);
+	char cmd[3*PATH_MAX];
+
+	if (verbose)
+		vcnt++;
+
+	snprintf(cmd, 3*PATH_MAX, PYTHON " %s/attr.py -d %s/attr/ -p %s %.*s",
+		 d, d, perf, vcnt, v);
+
+	return system(cmd);
+}
+
+int test__attr(void)
+{
+	struct stat st;
+	char path_perf[PATH_MAX];
+	char path_dir[PATH_MAX];
+
+	/* First try developement tree tests. */
+	if (!lstat("./tests", &st))
+		return run_dir("./tests", "./perf");
+
+	/* Then installed path. */
+	snprintf(path_dir,  PATH_MAX, "%s/tests", perf_exec_path());
+	snprintf(path_perf, PATH_MAX, "%s/perf", BINDIR);
+
+	if (!lstat(path_dir, &st) &&
+	    !lstat(path_perf, &st))
+		return run_dir(path_dir, path_perf);
+
+	fprintf(stderr, " (omitted)");
+	return 0;
+}
diff --git a/tools/perf/tests/attr.py b/tools/perf/tests/attr.py
new file mode 100644
index 000000000..1091bd47a
--- /dev/null
+++ b/tools/perf/tests/attr.py
@@ -0,0 +1,331 @@
+#! /usr/bin/python
+
+import os
+import sys
+import glob
+import optparse
+import tempfile
+import logging
+import shutil
+import ConfigParser
+
+class Fail(Exception):
+    def __init__(self, test, msg):
+        self.msg = msg
+        self.test = test
+    def getMsg(self):
+        return '\'%s\' - %s' % (self.test.path, self.msg)
+
+class Unsup(Exception):
+    def __init__(self, test):
+        self.test = test
+    def getMsg(self):
+        return '\'%s\'' % self.test.path
+
+class Event(dict):
+    terms = [
+        'cpu',
+        'flags',
+        'type',
+        'size',
+        'config',
+        'sample_period',
+        'sample_type',
+        'read_format',
+        'disabled',
+        'inherit',
+        'pinned',
+        'exclusive',
+        'exclude_user',
+        'exclude_kernel',
+        'exclude_hv',
+        'exclude_idle',
+        'mmap',
+        'comm',
+        'freq',
+        'inherit_stat',
+        'enable_on_exec',
+        'task',
+        'watermark',
+        'precise_ip',
+        'mmap_data',
+        'sample_id_all',
+        'exclude_host',
+        'exclude_guest',
+        'exclude_callchain_kernel',
+        'exclude_callchain_user',
+        'wakeup_events',
+        'bp_type',
+        'config1',
+        'config2',
+        'branch_sample_type',
+        'sample_regs_user',
+        'sample_stack_user',
+    ]
+
+    def add(self, data):
+        for key, val in data:
+            log.debug("      %s = %s" % (key, val))
+            self[key] = val
+
+    def __init__(self, name, data, base):
+        log.debug("    Event %s" % name);
+        self.name  = name;
+        self.group = ''
+        self.add(base)
+        self.add(data)
+
+    def compare_data(self, a, b):
+        # Allow multiple values in assignment separated by '|'
+        a_list = a.split('|')
+        b_list = b.split('|')
+
+        for a_item in a_list:
+            for b_item in b_list:
+                if (a_item == b_item):
+                    return True
+                elif (a_item == '*') or (b_item == '*'):
+                    return True
+
+        return False
+
+    def equal(self, other):
+        for t in Event.terms:
+            log.debug("      [%s] %s %s" % (t, self[t], other[t]));
+            if not self.has_key(t) or not other.has_key(t):
+                return False
+            if not self.compare_data(self[t], other[t]):
+                return False
+        return True
+
+    def diff(self, other):
+        for t in Event.terms:
+            if not self.has_key(t) or not other.has_key(t):
+                continue
+            if not self.compare_data(self[t], other[t]):
+		log.warning("expected %s=%s, got %s" % (t, self[t], other[t]))
+
+# Test file description needs to have following sections:
+# [config]
+#   - just single instance in file
+#   - needs to specify:
+#     'command' - perf command name
+#     'args'    - special command arguments
+#     'ret'     - expected command return value (0 by default)
+#
+# [eventX:base]
+#   - one or multiple instances in file
+#   - expected values assignments
+class Test(object):
+    def __init__(self, path, options):
+        parser = ConfigParser.SafeConfigParser()
+        parser.read(path)
+
+        log.warning("running '%s'" % path)
+
+        self.path     = path
+        self.test_dir = options.test_dir
+        self.perf     = options.perf
+        self.command  = parser.get('config', 'command')
+        self.args     = parser.get('config', 'args')
+
+        try:
+            self.ret  = parser.get('config', 'ret')
+        except:
+            self.ret  = 0
+
+        self.expect   = {}
+        self.result   = {}
+        log.debug("  loading expected events");
+        self.load_events(path, self.expect)
+
+    def is_event(self, name):
+        if name.find("event") == -1:
+            return False
+        else:
+            return True
+
+    def load_events(self, path, events):
+        parser_event = ConfigParser.SafeConfigParser()
+        parser_event.read(path)
+
+        # The event record section header contains 'event' word,
+        # optionaly followed by ':' allowing to load 'parent
+        # event' first as a base
+        for section in filter(self.is_event, parser_event.sections()):
+
+            parser_items = parser_event.items(section);
+            base_items   = {}
+
+            # Read parent event if there's any
+            if (':' in section):
+                base = section[section.index(':') + 1:]
+                parser_base = ConfigParser.SafeConfigParser()
+                parser_base.read(self.test_dir + '/' + base)
+                base_items = parser_base.items('event')
+
+            e = Event(section, parser_items, base_items)
+            events[section] = e
+
+    def run_cmd(self, tempdir):
+        cmd = "PERF_TEST_ATTR=%s %s %s -o %s/perf.data %s" % (tempdir,
+              self.perf, self.command, tempdir, self.args)
+        ret = os.WEXITSTATUS(os.system(cmd))
+
+        log.info("  '%s' ret %d " % (cmd, ret))
+
+        if ret != int(self.ret):
+            raise Unsup(self)
+
+    def compare(self, expect, result):
+        match = {}
+
+        log.debug("  compare");
+
+        # For each expected event find all matching
+        # events in result. Fail if there's not any.
+        for exp_name, exp_event in expect.items():
+            exp_list = []
+            log.debug("    matching [%s]" % exp_name)
+            for res_name, res_event in result.items():
+                log.debug("      to [%s]" % res_name)
+                if (exp_event.equal(res_event)):
+                    exp_list.append(res_name)
+                    log.debug("    ->OK")
+                else:
+                    log.debug("    ->FAIL");
+
+            log.debug("    match: [%s] matches %s" % (exp_name, str(exp_list)))
+
+            # we did not any matching event - fail
+            if (not exp_list):
+		exp_event.diff(res_event)
+                raise Fail(self, 'match failure');
+
+            match[exp_name] = exp_list
+
+        # For each defined group in the expected events
+        # check we match the same group in the result.
+        for exp_name, exp_event in expect.items():
+            group = exp_event.group
+
+            if (group == ''):
+                continue
+
+            for res_name in match[exp_name]:
+                res_group = result[res_name].group
+                if res_group not in match[group]:
+                    raise Fail(self, 'group failure')
+
+                log.debug("    group: [%s] matches group leader %s" %
+                         (exp_name, str(match[group])))
+
+        log.debug("  matched")
+
+    def resolve_groups(self, events):
+        for name, event in events.items():
+            group_fd = event['group_fd'];
+            if group_fd == '-1':
+                continue;
+
+            for iname, ievent in events.items():
+                if (ievent['fd'] == group_fd):
+                    event.group = iname
+                    log.debug('[%s] has group leader [%s]' % (name, iname))
+                    break;
+
+    def run(self):
+        tempdir = tempfile.mkdtemp();
+
+        try:
+            # run the test script
+            self.run_cmd(tempdir);
+
+            # load events expectation for the test
+            log.debug("  loading result events");
+            for f in glob.glob(tempdir + '/event*'):
+                self.load_events(f, self.result);
+
+            # resolve group_fd to event names
+            self.resolve_groups(self.expect);
+            self.resolve_groups(self.result);
+
+            # do the expectation - results matching - both ways
+            self.compare(self.expect, self.result)
+            self.compare(self.result, self.expect)
+
+        finally:
+            # cleanup
+            shutil.rmtree(tempdir)
+
+
+def run_tests(options):
+    for f in glob.glob(options.test_dir + '/' + options.test):
+        try:
+            Test(f, options).run()
+        except Unsup, obj:
+            log.warning("unsupp  %s" % obj.getMsg())
+
+def setup_log(verbose):
+    global log
+    level = logging.CRITICAL
+
+    if verbose == 1:
+        level = logging.WARNING
+    if verbose == 2:
+        level = logging.INFO
+    if verbose >= 3:
+        level = logging.DEBUG
+
+    log = logging.getLogger('test')
+    log.setLevel(level)
+    ch  = logging.StreamHandler()
+    ch.setLevel(level)
+    formatter = logging.Formatter('%(message)s')
+    ch.setFormatter(formatter)
+    log.addHandler(ch)
+
+USAGE = '''%s [OPTIONS]
+  -d dir  # tests dir
+  -p path # perf binary
+  -t test # single test
+  -v      # verbose level
+''' % sys.argv[0]
+
+def main():
+    parser = optparse.OptionParser(usage=USAGE)
+
+    parser.add_option("-t", "--test",
+                      action="store", type="string", dest="test")
+    parser.add_option("-d", "--test-dir",
+                      action="store", type="string", dest="test_dir")
+    parser.add_option("-p", "--perf",
+                      action="store", type="string", dest="perf")
+    parser.add_option("-v", "--verbose",
+                      action="count", dest="verbose")
+
+    options, args = parser.parse_args()
+    if args:
+        parser.error('FAILED wrong arguments %s' %  ' '.join(args))
+        return -1
+
+    setup_log(options.verbose)
+
+    if not options.test_dir:
+        print 'FAILED no -d option specified'
+        sys.exit(-1)
+
+    if not options.test:
+        options.test = 'test*'
+
+    try:
+        run_tests(options)
+
+    except Fail, obj:
+        print "FAILED %s" % obj.getMsg();
+        sys.exit(-1)
+
+    sys.exit(0)
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/perf/tests/attr/README b/tools/perf/tests/attr/README
new file mode 100644
index 000000000..430024f61
--- /dev/null
+++ b/tools/perf/tests/attr/README
@@ -0,0 +1,64 @@
+The struct perf_event_attr test (attr tests) support
+====================================================
+This testing support is embedded into perf directly and is governed
+by the PERF_TEST_ATTR environment variable and hook inside the
+sys_perf_event_open function.
+
+The general idea is to store 'struct perf_event_attr' details for
+each event created within single perf command. Each event details
+are stored into separate text file. Once perf command is finished
+these files are checked for values we expect for command.
+
+The attr tests consist of following parts:
+
+tests/attr.c
+------------
+This is the sys_perf_event_open hook implementation. The hook
+is triggered when the PERF_TEST_ATTR environment variable is
+defined. It must contain name of existing directory with access
+and write permissions.
+
+For each sys_perf_event_open call event details are stored in
+separate file. Besides 'struct perf_event_attr' values we also
+store 'fd' and 'group_fd' values to allow checking for groups.
+
+tests/attr.py
+-------------
+This is the python script that does all the hard work. It reads
+the test definition, executes it and checks results.
+
+tests/attr/
+-----------
+Directory containing all attr test definitions.
+Following tests are defined (with perf commands):
+
+  perf record kill                              (test-record-basic)
+  perf record -b kill                           (test-record-branch-any)
+  perf record -j any kill                       (test-record-branch-filter-any)
+  perf record -j any_call kill                  (test-record-branch-filter-any_call)
+  perf record -j any_ret kill                   (test-record-branch-filter-any_ret)
+  perf record -j hv kill                        (test-record-branch-filter-hv)
+  perf record -j ind_call kill                  (test-record-branch-filter-ind_call)
+  perf record -j k kill                         (test-record-branch-filter-k)
+  perf record -j u kill                         (test-record-branch-filter-u)
+  perf record -c 123 kill                       (test-record-count)
+  perf record -d kill                           (test-record-data)
+  perf record -F 100 kill                       (test-record-freq)
+  perf record -g kill                           (test-record-graph-default)
+  perf record --call-graph dwarf kill		(test-record-graph-dwarf)
+  perf record --call-graph fp kill              (test-record-graph-fp)
+  perf record --group -e cycles,instructions kill (test-record-group)
+  perf record -e '{cycles,instructions}' kill   (test-record-group1)
+  perf record -D kill                           (test-record-no-delay)
+  perf record -i kill                           (test-record-no-inherit)
+  perf record -n kill                           (test-record-no-samples)
+  perf record -c 100 -P kill                    (test-record-period)
+  perf record -R kill                           (test-record-raw)
+  perf stat -e cycles kill                      (test-stat-basic)
+  perf stat kill                                (test-stat-default)
+  perf stat -d kill                             (test-stat-detailed-1)
+  perf stat -dd kill                            (test-stat-detailed-2)
+  perf stat -ddd kill                           (test-stat-detailed-3)
+  perf stat --group -e cycles,instructions kill (test-stat-group)
+  perf stat -e '{cycles,instructions}' kill     (test-stat-group1)
+  perf stat -i -e cycles kill                   (test-stat-no-inherit)
diff --git a/tools/perf/tests/attr/base-record b/tools/perf/tests/attr/base-record
new file mode 100644
index 000000000..7e6d74946
--- /dev/null
+++ b/tools/perf/tests/attr/base-record
@@ -0,0 +1,41 @@
+[event]
+fd=1
+group_fd=-1
+# 0 or PERF_FLAG_FD_CLOEXEC flag
+flags=0|8
+cpu=*
+type=0|1
+size=112
+config=0
+sample_period=4000
+sample_type=263
+read_format=0
+disabled=1
+inherit=1
+pinned=0
+exclusive=0
+exclude_user=0
+exclude_kernel=0
+exclude_hv=0
+exclude_idle=0
+mmap=1
+comm=1
+freq=1
+inherit_stat=0
+enable_on_exec=1
+task=0
+watermark=0
+precise_ip=0
+mmap_data=0
+sample_id_all=1
+exclude_host=0|1
+exclude_guest=0|1
+exclude_callchain_kernel=0
+exclude_callchain_user=0
+wakeup_events=0
+bp_type=0
+config1=0
+config2=0
+branch_sample_type=0
+sample_regs_user=0
+sample_stack_user=0
diff --git a/tools/perf/tests/attr/base-stat b/tools/perf/tests/attr/base-stat
new file mode 100644
index 000000000..f4cf148f1
--- /dev/null
+++ b/tools/perf/tests/attr/base-stat
@@ -0,0 +1,41 @@
+[event]
+fd=1
+group_fd=-1
+# 0 or PERF_FLAG_FD_CLOEXEC flag
+flags=0|8
+cpu=*
+type=0
+size=112
+config=0
+sample_period=0
+sample_type=0
+read_format=3
+disabled=1
+inherit=1
+pinned=0
+exclusive=0
+exclude_user=0
+exclude_kernel=0
+exclude_hv=0
+exclude_idle=0
+mmap=0
+comm=0
+freq=0
+inherit_stat=0
+enable_on_exec=1
+task=0
+watermark=0
+precise_ip=0
+mmap_data=0
+sample_id_all=0
+exclude_host=0|1
+exclude_guest=0|1
+exclude_callchain_kernel=0
+exclude_callchain_user=0
+wakeup_events=0
+bp_type=0
+config1=0
+config2=0
+branch_sample_type=0
+sample_regs_user=0
+sample_stack_user=0
diff --git a/tools/perf/tests/attr/test-record-C0 b/tools/perf/tests/attr/test-record-C0
new file mode 100644
index 000000000..d6a7e43f6
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-C0
@@ -0,0 +1,13 @@
+[config]
+command = record
+args    = -C 0 kill >/dev/null 2>&1
+
+[event:base-record]
+cpu=0
+
+# no enable on exec for CPU attached
+enable_on_exec=0
+
+# PERF_SAMPLE_IP | PERF_SAMPLE_TID PERF_SAMPLE_TIME | # PERF_SAMPLE_PERIOD
+# + PERF_SAMPLE_CPU added by -C 0
+sample_type=391
diff --git a/tools/perf/tests/attr/test-record-basic b/tools/perf/tests/attr/test-record-basic
new file mode 100644
index 000000000..55c042837
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-basic
@@ -0,0 +1,5 @@
+[config]
+command = record
+args    = kill >/dev/null 2>&1
+
+[event:base-record]
diff --git a/tools/perf/tests/attr/test-record-branch-any b/tools/perf/tests/attr/test-record-branch-any
new file mode 100644
index 000000000..1421960ed
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-branch-any
@@ -0,0 +1,8 @@
+[config]
+command = record
+args    = -b kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=2311
+branch_sample_type=8
diff --git a/tools/perf/tests/attr/test-record-branch-filter-any b/tools/perf/tests/attr/test-record-branch-filter-any
new file mode 100644
index 000000000..915c4df0e
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-branch-filter-any
@@ -0,0 +1,8 @@
+[config]
+command = record
+args    = -j any kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=2311
+branch_sample_type=8
diff --git a/tools/perf/tests/attr/test-record-branch-filter-any_call b/tools/perf/tests/attr/test-record-branch-filter-any_call
new file mode 100644
index 000000000..8708dbd4f
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-branch-filter-any_call
@@ -0,0 +1,8 @@
+[config]
+command = record
+args    = -j any_call kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=2311
+branch_sample_type=16
diff --git a/tools/perf/tests/attr/test-record-branch-filter-any_ret b/tools/perf/tests/attr/test-record-branch-filter-any_ret
new file mode 100644
index 000000000..0d3607a6d
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-branch-filter-any_ret
@@ -0,0 +1,8 @@
+[config]
+command = record
+args    = -j any_ret kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=2311
+branch_sample_type=32
diff --git a/tools/perf/tests/attr/test-record-branch-filter-hv b/tools/perf/tests/attr/test-record-branch-filter-hv
new file mode 100644
index 000000000..f25526740
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-branch-filter-hv
@@ -0,0 +1,8 @@
+[config]
+command = record
+args    = -j hv kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=2311
+branch_sample_type=8
diff --git a/tools/perf/tests/attr/test-record-branch-filter-ind_call b/tools/perf/tests/attr/test-record-branch-filter-ind_call
new file mode 100644
index 000000000..e862dd179
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-branch-filter-ind_call
@@ -0,0 +1,8 @@
+[config]
+command = record
+args    = -j ind_call kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=2311
+branch_sample_type=64
diff --git a/tools/perf/tests/attr/test-record-branch-filter-k b/tools/perf/tests/attr/test-record-branch-filter-k
new file mode 100644
index 000000000..182971e89
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-branch-filter-k
@@ -0,0 +1,8 @@
+[config]
+command = record
+args    = -j k kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=2311
+branch_sample_type=8
diff --git a/tools/perf/tests/attr/test-record-branch-filter-u b/tools/perf/tests/attr/test-record-branch-filter-u
new file mode 100644
index 000000000..83449ef9e
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-branch-filter-u
@@ -0,0 +1,8 @@
+[config]
+command = record
+args    = -j u kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=2311
+branch_sample_type=8
diff --git a/tools/perf/tests/attr/test-record-count b/tools/perf/tests/attr/test-record-count
new file mode 100644
index 000000000..2f841de56
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-count
@@ -0,0 +1,8 @@
+[config]
+command = record
+args    = -c 123 kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=123
+sample_type=7
+freq=0
diff --git a/tools/perf/tests/attr/test-record-data b/tools/perf/tests/attr/test-record-data
new file mode 100644
index 000000000..716e143b5
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-data
@@ -0,0 +1,11 @@
+[config]
+command = record
+args    = -d kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+
+# sample_type = PERF_SAMPLE_IP | PERF_SAMPLE_TID | PERF_SAMPLE_TIME |
+# PERF_SAMPLE_ADDR | PERF_SAMPLE_PERIOD | PERF_SAMPLE_DATA_SRC
+sample_type=33039
+mmap_data=1
diff --git a/tools/perf/tests/attr/test-record-freq b/tools/perf/tests/attr/test-record-freq
new file mode 100644
index 000000000..600d0f8f2
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-freq
@@ -0,0 +1,6 @@
+[config]
+command = record
+args    = -F 100 kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=100
diff --git a/tools/perf/tests/attr/test-record-graph-default b/tools/perf/tests/attr/test-record-graph-default
new file mode 100644
index 000000000..853597a9a
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-graph-default
@@ -0,0 +1,6 @@
+[config]
+command = record
+args    = -g kill >/dev/null 2>&1
+
+[event:base-record]
+sample_type=295
diff --git a/tools/perf/tests/attr/test-record-graph-dwarf b/tools/perf/tests/attr/test-record-graph-dwarf
new file mode 100644
index 000000000..d6f324ea5
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-graph-dwarf
@@ -0,0 +1,10 @@
+[config]
+command = record
+args    = --call-graph dwarf -- kill >/dev/null 2>&1
+
+[event:base-record]
+sample_type=12583
+exclude_callchain_user=1
+sample_stack_user=8192
+# TODO different for each arch, no support for that now
+sample_regs_user=*
diff --git a/tools/perf/tests/attr/test-record-graph-fp b/tools/perf/tests/attr/test-record-graph-fp
new file mode 100644
index 000000000..055e3bee7
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-graph-fp
@@ -0,0 +1,6 @@
+[config]
+command = record
+args    = --call-graph fp kill >/dev/null 2>&1
+
+[event:base-record]
+sample_type=295
diff --git a/tools/perf/tests/attr/test-record-group b/tools/perf/tests/attr/test-record-group
new file mode 100644
index 000000000..57739cacd
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-group
@@ -0,0 +1,20 @@
+[config]
+command = record
+args    = --group -e cycles,instructions kill >/dev/null 2>&1
+
+[event-1:base-record]
+fd=1
+group_fd=-1
+sample_type=327
+read_format=4
+
+[event-2:base-record]
+fd=2
+group_fd=1
+config=1
+sample_type=327
+read_format=4
+mmap=0
+comm=0
+enable_on_exec=0
+disabled=0
diff --git a/tools/perf/tests/attr/test-record-group-sampling b/tools/perf/tests/attr/test-record-group-sampling
new file mode 100644
index 000000000..658f5d60c
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-group-sampling
@@ -0,0 +1,36 @@
+[config]
+command = record
+args    = -e '{cycles,cache-misses}:S' kill >/dev/null 2>&1
+
+[event-1:base-record]
+fd=1
+group_fd=-1
+sample_type=343
+read_format=12
+inherit=0
+
+[event-2:base-record]
+fd=2
+group_fd=1
+
+# cache-misses
+type=0
+config=3
+
+# default | PERF_SAMPLE_READ
+sample_type=343
+
+# PERF_FORMAT_ID | PERF_FORMAT_GROUP
+read_format=12
+
+mmap=0
+comm=0
+enable_on_exec=0
+disabled=0
+
+# inherit is disabled for group sampling
+inherit=0
+
+# sampling disabled
+sample_freq=0
+sample_period=0
diff --git a/tools/perf/tests/attr/test-record-group1 b/tools/perf/tests/attr/test-record-group1
new file mode 100644
index 000000000..c5548d054
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-group1
@@ -0,0 +1,21 @@
+[config]
+command = record
+args    = -e '{cycles,instructions}' kill >/dev/null 2>&1
+
+[event-1:base-record]
+fd=1
+group_fd=-1
+sample_type=327
+read_format=4
+
+[event-2:base-record]
+fd=2
+group_fd=1
+type=0
+config=1
+sample_type=327
+read_format=4
+mmap=0
+comm=0
+enable_on_exec=0
+disabled=0
diff --git a/tools/perf/tests/attr/test-record-no-delay b/tools/perf/tests/attr/test-record-no-delay
new file mode 100644
index 000000000..f253b78cd
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-no-delay
@@ -0,0 +1,9 @@
+[config]
+command = record
+args    = -D kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=263
+watermark=0
+wakeup_events=1
diff --git a/tools/perf/tests/attr/test-record-no-inherit b/tools/perf/tests/attr/test-record-no-inherit
new file mode 100644
index 000000000..44edcb2ed
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-no-inherit
@@ -0,0 +1,7 @@
+[config]
+command = record
+args    = -i kill >/dev/null 2>&1
+
+[event:base-record]
+sample_type=263
+inherit=0
diff --git a/tools/perf/tests/attr/test-record-no-samples b/tools/perf/tests/attr/test-record-no-samples
new file mode 100644
index 000000000..d0141b241
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-no-samples
@@ -0,0 +1,6 @@
+[config]
+command = record
+args    = -n kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=0
diff --git a/tools/perf/tests/attr/test-record-period b/tools/perf/tests/attr/test-record-period
new file mode 100644
index 000000000..8abc5314f
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-period
@@ -0,0 +1,7 @@
+[config]
+command = record
+args    = -c 100 -P kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=100
+freq=0
diff --git a/tools/perf/tests/attr/test-record-raw b/tools/perf/tests/attr/test-record-raw
new file mode 100644
index 000000000..4a8ef25b5
--- /dev/null
+++ b/tools/perf/tests/attr/test-record-raw
@@ -0,0 +1,7 @@
+[config]
+command = record
+args    = -R kill >/dev/null 2>&1
+
+[event:base-record]
+sample_period=4000
+sample_type=1415
diff --git a/tools/perf/tests/attr/test-stat-C0 b/tools/perf/tests/attr/test-stat-C0
new file mode 100644
index 000000000..aa8359507
--- /dev/null
+++ b/tools/perf/tests/attr/test-stat-C0
@@ -0,0 +1,9 @@
+[config]
+command = stat
+args    = -e cycles -C 0 kill >/dev/null 2>&1
+ret     = 1
+
+[event:base-stat]
+# events are enabled by default when attached to cpu
+disabled=0
+enable_on_exec=0
diff --git a/tools/perf/tests/attr/test-stat-basic b/tools/perf/tests/attr/test-stat-basic
new file mode 100644
index 000000000..74e17881f
--- /dev/null
+++ b/tools/perf/tests/attr/test-stat-basic
@@ -0,0 +1,6 @@
+[config]
+command = stat
+args    = -e cycles kill >/dev/null 2>&1
+ret     = 1
+
+[event:base-stat]
diff --git a/tools/perf/tests/attr/test-stat-default b/tools/perf/tests/attr/test-stat-default
new file mode 100644
index 000000000..19270f54c
--- /dev/null
+++ b/tools/perf/tests/attr/test-stat-default
@@ -0,0 +1,64 @@
+[config]
+command = stat
+args    = kill >/dev/null 2>&1
+ret     = 1
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_TASK_CLOCK
+[event1:base-stat]
+fd=1
+type=1
+config=1
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CONTEXT_SWITCHES
+[event2:base-stat]
+fd=2
+type=1
+config=3
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CPU_MIGRATIONS
+[event3:base-stat]
+fd=3
+type=1
+config=4
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_PAGE_FAULTS
+[event4:base-stat]
+fd=4
+type=1
+config=2
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_CPU_CYCLES
+[event5:base-stat]
+fd=5
+type=0
+config=0
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_FRONTEND
+[event6:base-stat]
+fd=6
+type=0
+config=7
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_BACKEND
+[event7:base-stat]
+fd=7
+type=0
+config=8
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_INSTRUCTIONS
+[event8:base-stat]
+fd=8
+type=0
+config=1
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS
+[event9:base-stat]
+fd=9
+type=0
+config=4
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES
+[event10:base-stat]
+fd=10
+type=0
+config=5
diff --git a/tools/perf/tests/attr/test-stat-detailed-1 b/tools/perf/tests/attr/test-stat-detailed-1
new file mode 100644
index 000000000..51426b871
--- /dev/null
+++ b/tools/perf/tests/attr/test-stat-detailed-1
@@ -0,0 +1,101 @@
+[config]
+command = stat
+args    = -d kill >/dev/null 2>&1
+ret     = 1
+
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_TASK_CLOCK
+[event1:base-stat]
+fd=1
+type=1
+config=1
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CONTEXT_SWITCHES
+[event2:base-stat]
+fd=2
+type=1
+config=3
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CPU_MIGRATIONS
+[event3:base-stat]
+fd=3
+type=1
+config=4
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_PAGE_FAULTS
+[event4:base-stat]
+fd=4
+type=1
+config=2
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_CPU_CYCLES
+[event5:base-stat]
+fd=5
+type=0
+config=0
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_FRONTEND
+[event6:base-stat]
+fd=6
+type=0
+config=7
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_BACKEND
+[event7:base-stat]
+fd=7
+type=0
+config=8
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_INSTRUCTIONS
+[event8:base-stat]
+fd=8
+type=0
+config=1
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS
+[event9:base-stat]
+fd=9
+type=0
+config=4
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES
+[event10:base-stat]
+fd=10
+type=0
+config=5
+
+# PERF_TYPE_HW_CACHE /
+#  PERF_COUNT_HW_CACHE_L1D                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event11:base-stat]
+fd=11
+type=3
+config=0
+
+# PERF_TYPE_HW_CACHE /
+#  PERF_COUNT_HW_CACHE_L1D                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event12:base-stat]
+fd=12
+type=3
+config=65536
+
+# PERF_TYPE_HW_CACHE /
+#  PERF_COUNT_HW_CACHE_LL                 <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event13:base-stat]
+fd=13
+type=3
+config=2
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_LL                 <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event14:base-stat]
+fd=14
+type=3
+config=65538
diff --git a/tools/perf/tests/attr/test-stat-detailed-2 b/tools/perf/tests/attr/test-stat-detailed-2
new file mode 100644
index 000000000..8de5acc31
--- /dev/null
+++ b/tools/perf/tests/attr/test-stat-detailed-2
@@ -0,0 +1,155 @@
+[config]
+command = stat
+args    = -dd kill >/dev/null 2>&1
+ret     = 1
+
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_TASK_CLOCK
+[event1:base-stat]
+fd=1
+type=1
+config=1
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CONTEXT_SWITCHES
+[event2:base-stat]
+fd=2
+type=1
+config=3
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CPU_MIGRATIONS
+[event3:base-stat]
+fd=3
+type=1
+config=4
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_PAGE_FAULTS
+[event4:base-stat]
+fd=4
+type=1
+config=2
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_CPU_CYCLES
+[event5:base-stat]
+fd=5
+type=0
+config=0
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_FRONTEND
+[event6:base-stat]
+fd=6
+type=0
+config=7
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_BACKEND
+[event7:base-stat]
+fd=7
+type=0
+config=8
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_INSTRUCTIONS
+[event8:base-stat]
+fd=8
+type=0
+config=1
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS
+[event9:base-stat]
+fd=9
+type=0
+config=4
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES
+[event10:base-stat]
+fd=10
+type=0
+config=5
+
+# PERF_TYPE_HW_CACHE /
+#  PERF_COUNT_HW_CACHE_L1D                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event11:base-stat]
+fd=11
+type=3
+config=0
+
+# PERF_TYPE_HW_CACHE /
+#  PERF_COUNT_HW_CACHE_L1D                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event12:base-stat]
+fd=12
+type=3
+config=65536
+
+# PERF_TYPE_HW_CACHE /
+#  PERF_COUNT_HW_CACHE_LL                 <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event13:base-stat]
+fd=13
+type=3
+config=2
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_LL                 <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event14:base-stat]
+fd=14
+type=3
+config=65538
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_L1I                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event15:base-stat]
+fd=15
+type=3
+config=1
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_L1I                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event16:base-stat]
+fd=16
+type=3
+config=65537
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_DTLB               <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event17:base-stat]
+fd=17
+type=3
+config=3
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_DTLB               <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event18:base-stat]
+fd=18
+type=3
+config=65539
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_ITLB               <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event19:base-stat]
+fd=19
+type=3
+config=4
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_ITLB               <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event20:base-stat]
+fd=20
+type=3
+config=65540
diff --git a/tools/perf/tests/attr/test-stat-detailed-3 b/tools/perf/tests/attr/test-stat-detailed-3
new file mode 100644
index 000000000..0a1f45bf7
--- /dev/null
+++ b/tools/perf/tests/attr/test-stat-detailed-3
@@ -0,0 +1,173 @@
+[config]
+command = stat
+args    = -ddd kill >/dev/null 2>&1
+ret     = 1
+
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_TASK_CLOCK
+[event1:base-stat]
+fd=1
+type=1
+config=1
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CONTEXT_SWITCHES
+[event2:base-stat]
+fd=2
+type=1
+config=3
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_CPU_MIGRATIONS
+[event3:base-stat]
+fd=3
+type=1
+config=4
+
+# PERF_TYPE_SOFTWARE / PERF_COUNT_SW_PAGE_FAULTS
+[event4:base-stat]
+fd=4
+type=1
+config=2
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_CPU_CYCLES
+[event5:base-stat]
+fd=5
+type=0
+config=0
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_FRONTEND
+[event6:base-stat]
+fd=6
+type=0
+config=7
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_STALLED_CYCLES_BACKEND
+[event7:base-stat]
+fd=7
+type=0
+config=8
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_INSTRUCTIONS
+[event8:base-stat]
+fd=8
+type=0
+config=1
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_INSTRUCTIONS
+[event9:base-stat]
+fd=9
+type=0
+config=4
+
+# PERF_TYPE_HARDWARE / PERF_COUNT_HW_BRANCH_MISSES
+[event10:base-stat]
+fd=10
+type=0
+config=5
+
+# PERF_TYPE_HW_CACHE /
+#  PERF_COUNT_HW_CACHE_L1D                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event11:base-stat]
+fd=11
+type=3
+config=0
+
+# PERF_TYPE_HW_CACHE /
+#  PERF_COUNT_HW_CACHE_L1D                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event12:base-stat]
+fd=12
+type=3
+config=65536
+
+# PERF_TYPE_HW_CACHE /
+#  PERF_COUNT_HW_CACHE_LL                 <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event13:base-stat]
+fd=13
+type=3
+config=2
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_LL                 <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event14:base-stat]
+fd=14
+type=3
+config=65538
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_L1I                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event15:base-stat]
+fd=15
+type=3
+config=1
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_L1I                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event16:base-stat]
+fd=16
+type=3
+config=65537
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_DTLB               <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event17:base-stat]
+fd=17
+type=3
+config=3
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_DTLB               <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event18:base-stat]
+fd=18
+type=3
+config=65539
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_ITLB               <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event19:base-stat]
+fd=19
+type=3
+config=4
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_ITLB               <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_READ            <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event20:base-stat]
+fd=20
+type=3
+config=65540
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_L1D                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_PREFETCH        <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_ACCESS      << 16)
+[event21:base-stat]
+fd=21
+type=3
+config=512
+
+# PERF_TYPE_HW_CACHE,
+#  PERF_COUNT_HW_CACHE_L1D                <<  0  |
+# (PERF_COUNT_HW_CACHE_OP_PREFETCH        <<  8) |
+# (PERF_COUNT_HW_CACHE_RESULT_MISS        << 16)
+[event22:base-stat]
+fd=22
+type=3
+config=66048
diff --git a/tools/perf/tests/attr/test-stat-group b/tools/perf/tests/attr/test-stat-group
new file mode 100644
index 000000000..fdc1596a8
--- /dev/null
+++ b/tools/perf/tests/attr/test-stat-group
@@ -0,0 +1,15 @@
+[config]
+command = stat
+args    = --group -e cycles,instructions kill >/dev/null 2>&1
+ret     = 1
+
+[event-1:base-stat]
+fd=1
+group_fd=-1
+
+[event-2:base-stat]
+fd=2
+group_fd=1
+config=1
+disabled=0
+enable_on_exec=0
diff --git a/tools/perf/tests/attr/test-stat-group1 b/tools/perf/tests/attr/test-stat-group1
new file mode 100644
index 000000000..2a1f86e4a
--- /dev/null
+++ b/tools/perf/tests/attr/test-stat-group1
@@ -0,0 +1,15 @@
+[config]
+command = stat
+args    = -e '{cycles,instructions}' kill >/dev/null 2>&1
+ret     = 1
+
+[event-1:base-stat]
+fd=1
+group_fd=-1
+
+[event-2:base-stat]
+fd=2
+group_fd=1
+config=1
+disabled=0
+enable_on_exec=0
diff --git a/tools/perf/tests/attr/test-stat-no-inherit b/tools/perf/tests/attr/test-stat-no-inherit
new file mode 100644
index 000000000..d54b2a1e3
--- /dev/null
+++ b/tools/perf/tests/attr/test-stat-no-inherit
@@ -0,0 +1,7 @@
+[config]
+command = stat
+args    = -i -e cycles kill >/dev/null 2>&1
+ret     = 1
+
+[event:base-stat]
+inherit=0
diff --git a/tools/perf/tests/bp_signal.c b/tools/perf/tests/bp_signal.c
new file mode 100644
index 000000000..a02b035fd
--- /dev/null
+++ b/tools/perf/tests/bp_signal.c
@@ -0,0 +1,194 @@
+/*
+ * Inspired by breakpoint overflow test done by
+ * Vince Weaver <vincent.weaver@maine.edu> for perf_event_tests
+ * (git://github.com/deater/perf_event_tests)
+ */
+
+/*
+ * Powerpc needs __SANE_USERSPACE_TYPES__ before <linux/types.h> to select
+ * 'int-ll64.h' and avoid compile warnings when printing __u64 with %llu.
+ */
+#define __SANE_USERSPACE_TYPES__
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <time.h>
+#include <fcntl.h>
+#include <signal.h>
+#include <sys/mman.h>
+#include <linux/compiler.h>
+#include <linux/hw_breakpoint.h>
+
+#include "tests.h"
+#include "debug.h"
+#include "perf.h"
+#include "cloexec.h"
+
+static int fd1;
+static int fd2;
+static int overflows;
+
+__attribute__ ((noinline))
+static int test_function(void)
+{
+	return time(NULL);
+}
+
+static void sig_handler(int signum __maybe_unused,
+			siginfo_t *oh __maybe_unused,
+			void *uc __maybe_unused)
+{
+	overflows++;
+
+	if (overflows > 10) {
+		/*
+		 * This should be executed only once during
+		 * this test, if we are here for the 10th
+		 * time, consider this the recursive issue.
+		 *
+		 * We can get out of here by disable events,
+		 * so no new SIGIO is delivered.
+		 */
+		ioctl(fd1, PERF_EVENT_IOC_DISABLE, 0);
+		ioctl(fd2, PERF_EVENT_IOC_DISABLE, 0);
+	}
+}
+
+static int bp_event(void *fn, int setup_signal)
+{
+	struct perf_event_attr pe;
+	int fd;
+
+	memset(&pe, 0, sizeof(struct perf_event_attr));
+	pe.type = PERF_TYPE_BREAKPOINT;
+	pe.size = sizeof(struct perf_event_attr);
+
+	pe.config = 0;
+	pe.bp_type = HW_BREAKPOINT_X;
+	pe.bp_addr = (unsigned long) fn;
+	pe.bp_len = sizeof(long);
+
+	pe.sample_period = 1;
+	pe.sample_type = PERF_SAMPLE_IP;
+	pe.wakeup_events = 1;
+
+	pe.disabled = 1;
+	pe.exclude_kernel = 1;
+	pe.exclude_hv = 1;
+
+	fd = sys_perf_event_open(&pe, 0, -1, -1,
+				 perf_event_open_cloexec_flag());
+	if (fd < 0) {
+		pr_debug("failed opening event %llx\n", pe.config);
+		return TEST_FAIL;
+	}
+
+	if (setup_signal) {
+		fcntl(fd, F_SETFL, O_RDWR|O_NONBLOCK|O_ASYNC);
+		fcntl(fd, F_SETSIG, SIGIO);
+		fcntl(fd, F_SETOWN, getpid());
+	}
+
+	ioctl(fd, PERF_EVENT_IOC_RESET, 0);
+
+	return fd;
+}
+
+static long long bp_count(int fd)
+{
+	long long count;
+	int ret;
+
+	ret = read(fd, &count, sizeof(long long));
+	if (ret != sizeof(long long)) {
+		pr_debug("failed to read: %d\n", ret);
+		return TEST_FAIL;
+	}
+
+	return count;
+}
+
+int test__bp_signal(void)
+{
+	struct sigaction sa;
+	long long count1, count2;
+
+	/* setup SIGIO signal handler */
+	memset(&sa, 0, sizeof(struct sigaction));
+	sa.sa_sigaction = (void *) sig_handler;
+	sa.sa_flags = SA_SIGINFO;
+
+	if (sigaction(SIGIO, &sa, NULL) < 0) {
+		pr_debug("failed setting up signal handler\n");
+		return TEST_FAIL;
+	}
+
+	/*
+	 * We create following events:
+	 *
+	 * fd1 - breakpoint event on test_function with SIGIO
+	 *       signal configured. We should get signal
+	 *       notification each time the breakpoint is hit
+	 *
+	 * fd2 - breakpoint event on sig_handler without SIGIO
+	 *       configured.
+	 *
+	 * Following processing should happen:
+	 *   - execute test_function
+	 *   - fd1 event breakpoint hit -> count1 == 1
+	 *   - SIGIO is delivered       -> overflows == 1
+	 *   - fd2 event breakpoint hit -> count2 == 1
+	 *
+	 * The test case check following error conditions:
+	 * - we get stuck in signal handler because of debug
+	 *   exception being triggered receursively due to
+	 *   the wrong RF EFLAG management
+	 *
+	 * - we never trigger the sig_handler breakpoint due
+	 *   to the rong RF EFLAG management
+	 *
+	 */
+
+	fd1 = bp_event(test_function, 1);
+	fd2 = bp_event(sig_handler, 0);
+
+	ioctl(fd1, PERF_EVENT_IOC_ENABLE, 0);
+	ioctl(fd2, PERF_EVENT_IOC_ENABLE, 0);
+
+	/*
+	 * Kick off the test by trigering 'fd1'
+	 * breakpoint.
+	 */
+	test_function();
+
+	ioctl(fd1, PERF_EVENT_IOC_DISABLE, 0);
+	ioctl(fd2, PERF_EVENT_IOC_DISABLE, 0);
+
+	count1 = bp_count(fd1);
+	count2 = bp_count(fd2);
+
+	close(fd1);
+	close(fd2);
+
+	pr_debug("count1 %lld, count2 %lld, overflow %d\n",
+		 count1, count2, overflows);
+
+	if (count1 != 1) {
+		if (count1 == 11)
+			pr_debug("failed: RF EFLAG recursion issue detected\n");
+		else
+			pr_debug("failed: wrong count for bp1%lld\n", count1);
+	}
+
+	if (overflows != 1)
+		pr_debug("failed: wrong overflow hit\n");
+
+	if (count2 != 1)
+		pr_debug("failed: wrong count for bp2\n");
+
+	return count1 == 1 && overflows == 1 && count2 == 1 ?
+		TEST_OK : TEST_FAIL;
+}
diff --git a/tools/perf/tests/bp_signal_overflow.c b/tools/perf/tests/bp_signal_overflow.c
new file mode 100644
index 000000000..e76537724
--- /dev/null
+++ b/tools/perf/tests/bp_signal_overflow.c
@@ -0,0 +1,134 @@
+/*
+ * Originally done by Vince Weaver <vincent.weaver@maine.edu> for
+ * perf_event_tests (git://github.com/deater/perf_event_tests)
+ */
+
+/*
+ * Powerpc needs __SANE_USERSPACE_TYPES__ before <linux/types.h> to select
+ * 'int-ll64.h' and avoid compile warnings when printing __u64 with %llu.
+ */
+#define __SANE_USERSPACE_TYPES__
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <time.h>
+#include <fcntl.h>
+#include <signal.h>
+#include <sys/mman.h>
+#include <linux/compiler.h>
+#include <linux/hw_breakpoint.h>
+
+#include "tests.h"
+#include "debug.h"
+#include "perf.h"
+#include "cloexec.h"
+
+static int overflows;
+
+__attribute__ ((noinline))
+static int test_function(void)
+{
+	return time(NULL);
+}
+
+static void sig_handler(int signum __maybe_unused,
+			siginfo_t *oh __maybe_unused,
+			void *uc __maybe_unused)
+{
+	overflows++;
+}
+
+static long long bp_count(int fd)
+{
+	long long count;
+	int ret;
+
+	ret = read(fd, &count, sizeof(long long));
+	if (ret != sizeof(long long)) {
+		pr_debug("failed to read: %d\n", ret);
+		return TEST_FAIL;
+	}
+
+	return count;
+}
+
+#define EXECUTIONS 10000
+#define THRESHOLD  100
+
+int test__bp_signal_overflow(void)
+{
+	struct perf_event_attr pe;
+	struct sigaction sa;
+	long long count;
+	int fd, i, fails = 0;
+
+	/* setup SIGIO signal handler */
+	memset(&sa, 0, sizeof(struct sigaction));
+	sa.sa_sigaction = (void *) sig_handler;
+	sa.sa_flags = SA_SIGINFO;
+
+	if (sigaction(SIGIO, &sa, NULL) < 0) {
+		pr_debug("failed setting up signal handler\n");
+		return TEST_FAIL;
+	}
+
+	memset(&pe, 0, sizeof(struct perf_event_attr));
+	pe.type = PERF_TYPE_BREAKPOINT;
+	pe.size = sizeof(struct perf_event_attr);
+
+	pe.config = 0;
+	pe.bp_type = HW_BREAKPOINT_X;
+	pe.bp_addr = (unsigned long) test_function;
+	pe.bp_len = sizeof(long);
+
+	pe.sample_period = THRESHOLD;
+	pe.sample_type = PERF_SAMPLE_IP;
+	pe.wakeup_events = 1;
+
+	pe.disabled = 1;
+	pe.exclude_kernel = 1;
+	pe.exclude_hv = 1;
+
+	fd = sys_perf_event_open(&pe, 0, -1, -1,
+				 perf_event_open_cloexec_flag());
+	if (fd < 0) {
+		pr_debug("failed opening event %llx\n", pe.config);
+		return TEST_FAIL;
+	}
+
+	fcntl(fd, F_SETFL, O_RDWR|O_NONBLOCK|O_ASYNC);
+	fcntl(fd, F_SETSIG, SIGIO);
+	fcntl(fd, F_SETOWN, getpid());
+
+	ioctl(fd, PERF_EVENT_IOC_RESET, 0);
+	ioctl(fd, PERF_EVENT_IOC_ENABLE, 0);
+
+	for (i = 0; i < EXECUTIONS; i++)
+		test_function();
+
+	ioctl(fd, PERF_EVENT_IOC_DISABLE, 0);
+
+	count = bp_count(fd);
+
+	close(fd);
+
+	pr_debug("count %lld, overflow %d\n",
+		 count, overflows);
+
+	if (count != EXECUTIONS) {
+		pr_debug("\tWrong number of executions %lld != %d\n",
+		count, EXECUTIONS);
+		fails++;
+	}
+
+	if (overflows != EXECUTIONS / THRESHOLD) {
+		pr_debug("\tWrong number of overflows %d != %d\n",
+		overflows, EXECUTIONS / THRESHOLD);
+		fails++;
+	}
+
+	return fails ? TEST_FAIL : TEST_OK;
+}
diff --git a/tools/perf/tests/builtin-test.c b/tools/perf/tests/builtin-test.c
new file mode 100644
index 000000000..4f4098167
--- /dev/null
+++ b/tools/perf/tests/builtin-test.c
@@ -0,0 +1,331 @@
+/*
+ * builtin-test.c
+ *
+ * Builtin regression testing command: ever growing number of sanity tests
+ */
+#include <unistd.h>
+#include <string.h>
+#include "builtin.h"
+#include "hist.h"
+#include "intlist.h"
+#include "tests.h"
+#include "debug.h"
+#include "color.h"
+#include "parse-options.h"
+#include "symbol.h"
+
+static struct test {
+	const char *desc;
+	int (*func)(void);
+} tests[] = {
+	{
+		.desc = "vmlinux symtab matches kallsyms",
+		.func = test__vmlinux_matches_kallsyms,
+	},
+	{
+		.desc = "detect open syscall event",
+		.func = test__open_syscall_event,
+	},
+	{
+		.desc = "detect open syscall event on all cpus",
+		.func = test__open_syscall_event_on_all_cpus,
+	},
+	{
+		.desc = "read samples using the mmap interface",
+		.func = test__basic_mmap,
+	},
+	{
+		.desc = "parse events tests",
+		.func = test__parse_events,
+	},
+#if defined(__x86_64__) || defined(__i386__)
+	{
+		.desc = "x86 rdpmc test",
+		.func = test__rdpmc,
+	},
+#endif
+	{
+		.desc = "Validate PERF_RECORD_* events & perf_sample fields",
+		.func = test__PERF_RECORD,
+	},
+	{
+		.desc = "Test perf pmu format parsing",
+		.func = test__pmu,
+	},
+	{
+		.desc = "Test dso data read",
+		.func = test__dso_data,
+	},
+	{
+		.desc = "Test dso data cache",
+		.func = test__dso_data_cache,
+	},
+	{
+		.desc = "Test dso data reopen",
+		.func = test__dso_data_reopen,
+	},
+	{
+		.desc = "roundtrip evsel->name check",
+		.func = test__perf_evsel__roundtrip_name_test,
+	},
+	{
+		.desc = "Check parsing of sched tracepoints fields",
+		.func = test__perf_evsel__tp_sched_test,
+	},
+	{
+		.desc = "Generate and check syscalls:sys_enter_open event fields",
+		.func = test__syscall_open_tp_fields,
+	},
+	{
+		.desc = "struct perf_event_attr setup",
+		.func = test__attr,
+	},
+	{
+		.desc = "Test matching and linking multiple hists",
+		.func = test__hists_link,
+	},
+	{
+		.desc = "Try 'import perf' in python, checking link problems",
+		.func = test__python_use,
+	},
+	{
+		.desc = "Test breakpoint overflow signal handler",
+		.func = test__bp_signal,
+	},
+	{
+		.desc = "Test breakpoint overflow sampling",
+		.func = test__bp_signal_overflow,
+	},
+	{
+		.desc = "Test number of exit event of a simple workload",
+		.func = test__task_exit,
+	},
+	{
+		.desc = "Test software clock events have valid period values",
+		.func = test__sw_clock_freq,
+	},
+#if defined(__x86_64__) || defined(__i386__)
+	{
+		.desc = "Test converting perf time to TSC",
+		.func = test__perf_time_to_tsc,
+	},
+#endif
+	{
+		.desc = "Test object code reading",
+		.func = test__code_reading,
+	},
+	{
+		.desc = "Test sample parsing",
+		.func = test__sample_parsing,
+	},
+	{
+		.desc = "Test using a dummy software event to keep tracking",
+		.func = test__keep_tracking,
+	},
+	{
+		.desc = "Test parsing with no sample_id_all bit set",
+		.func = test__parse_no_sample_id_all,
+	},
+#if defined(__x86_64__) || defined(__i386__) || defined(__arm__)
+#ifdef HAVE_DWARF_UNWIND_SUPPORT
+	{
+		.desc = "Test dwarf unwind",
+		.func = test__dwarf_unwind,
+	},
+#endif
+#endif
+	{
+		.desc = "Test filtering hist entries",
+		.func = test__hists_filter,
+	},
+	{
+		.desc = "Test mmap thread lookup",
+		.func = test__mmap_thread_lookup,
+	},
+	{
+		.desc = "Test thread mg sharing",
+		.func = test__thread_mg_share,
+	},
+	{
+		.desc = "Test output sorting of hist entries",
+		.func = test__hists_output,
+	},
+	{
+		.desc = "Test cumulation of child hist entries",
+		.func = test__hists_cumulate,
+	},
+	{
+		.desc = "Test tracking with sched_switch",
+		.func = test__switch_tracking,
+	},
+	{
+		.desc = "Filter fds with revents mask in a fdarray",
+		.func = test__fdarray__filter,
+	},
+	{
+		.desc = "Add fd to a fdarray, making it autogrow",
+		.func = test__fdarray__add,
+	},
+	{
+		.desc = "Test kmod_path__parse function",
+		.func = test__kmod_path__parse,
+	},
+	{
+		.func = NULL,
+	},
+};
+
+static bool perf_test__matches(int curr, int argc, const char *argv[])
+{
+	int i;
+
+	if (argc == 0)
+		return true;
+
+	for (i = 0; i < argc; ++i) {
+		char *end;
+		long nr = strtoul(argv[i], &end, 10);
+
+		if (*end == '\0') {
+			if (nr == curr + 1)
+				return true;
+			continue;
+		}
+
+		if (strstr(tests[curr].desc, argv[i]))
+			return true;
+	}
+
+	return false;
+}
+
+static int run_test(struct test *test)
+{
+	int status, err = -1, child = fork();
+	char sbuf[STRERR_BUFSIZE];
+
+	if (child < 0) {
+		pr_err("failed to fork test: %s\n",
+			strerror_r(errno, sbuf, sizeof(sbuf)));
+		return -1;
+	}
+
+	if (!child) {
+		pr_debug("test child forked, pid %d\n", getpid());
+		err = test->func();
+		exit(err);
+	}
+
+	wait(&status);
+
+	if (WIFEXITED(status)) {
+		err = WEXITSTATUS(status);
+		pr_debug("test child finished with %d\n", err);
+	} else if (WIFSIGNALED(status)) {
+		err = -1;
+		pr_debug("test child interrupted\n");
+	}
+
+	return err;
+}
+
+static int __cmd_test(int argc, const char *argv[], struct intlist *skiplist)
+{
+	int i = 0;
+	int width = 0;
+
+	while (tests[i].func) {
+		int len = strlen(tests[i].desc);
+
+		if (width < len)
+			width = len;
+		++i;
+	}
+
+	i = 0;
+	while (tests[i].func) {
+		int curr = i++, err;
+
+		if (!perf_test__matches(curr, argc, argv))
+			continue;
+
+		pr_info("%2d: %-*s:", i, width, tests[curr].desc);
+
+		if (intlist__find(skiplist, i)) {
+			color_fprintf(stderr, PERF_COLOR_YELLOW, " Skip (user override)\n");
+			continue;
+		}
+
+		pr_debug("\n--- start ---\n");
+		err = run_test(&tests[curr]);
+		pr_debug("---- end ----\n%s:", tests[curr].desc);
+
+		switch (err) {
+		case TEST_OK:
+			pr_info(" Ok\n");
+			break;
+		case TEST_SKIP:
+			color_fprintf(stderr, PERF_COLOR_YELLOW, " Skip\n");
+			break;
+		case TEST_FAIL:
+		default:
+			color_fprintf(stderr, PERF_COLOR_RED, " FAILED!\n");
+			break;
+		}
+	}
+
+	return 0;
+}
+
+static int perf_test__list(int argc, const char **argv)
+{
+	int i = 0;
+
+	while (tests[i].func) {
+		int curr = i++;
+
+		if (argc > 1 && !strstr(tests[curr].desc, argv[1]))
+			continue;
+
+		pr_info("%2d: %s\n", i, tests[curr].desc);
+	}
+
+	return 0;
+}
+
+int cmd_test(int argc, const char **argv, const char *prefix __maybe_unused)
+{
+	const char *test_usage[] = {
+	"perf test [<options>] [{list <test-name-fragment>|[<test-name-fragments>|<test-numbers>]}]",
+	NULL,
+	};
+	const char *skip = NULL;
+	const struct option test_options[] = {
+	OPT_STRING('s', "skip", &skip, "tests", "tests to skip"),
+	OPT_INCR('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_END()
+	};
+	const char * const test_subcommands[] = { "list", NULL };
+	struct intlist *skiplist = NULL;
+        int ret = hists__init();
+
+        if (ret < 0)
+                return ret;
+
+	argc = parse_options_subcommand(argc, argv, test_options, test_subcommands, test_usage, 0);
+	if (argc >= 1 && !strcmp(argv[0], "list"))
+		return perf_test__list(argc, argv);
+
+	symbol_conf.priv_size = sizeof(int);
+	symbol_conf.sort_by_name = true;
+	symbol_conf.try_vmlinux_path = true;
+
+	if (symbol__init(NULL) < 0)
+		return -1;
+
+	if (skip != NULL)
+		skiplist = intlist__new(skip);
+
+	return __cmd_test(argc, argv, skiplist);
+}
diff --git a/tools/perf/tests/code-reading.c b/tools/perf/tests/code-reading.c
new file mode 100644
index 000000000..f671ec37a
--- /dev/null
+++ b/tools/perf/tests/code-reading.c
@@ -0,0 +1,578 @@
+#include <linux/types.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <ctype.h>
+#include <string.h>
+
+#include "parse-events.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "thread_map.h"
+#include "cpumap.h"
+#include "machine.h"
+#include "event.h"
+#include "thread.h"
+
+#include "tests.h"
+
+#define BUFSZ	1024
+#define READLEN	128
+
+struct state {
+	u64 done[1024];
+	size_t done_cnt;
+};
+
+static unsigned int hex(char c)
+{
+	if (c >= '0' && c <= '9')
+		return c - '0';
+	if (c >= 'a' && c <= 'f')
+		return c - 'a' + 10;
+	return c - 'A' + 10;
+}
+
+static void read_objdump_line(const char *line, size_t line_len, void **buf,
+			      size_t *len)
+{
+	const char *p;
+	size_t i;
+
+	/* Skip to a colon */
+	p = strchr(line, ':');
+	if (!p)
+		return;
+	i = p + 1 - line;
+
+	/* Read bytes */
+	while (*len) {
+		char c1, c2;
+
+		/* Skip spaces */
+		for (; i < line_len; i++) {
+			if (!isspace(line[i]))
+				break;
+		}
+		/* Get 2 hex digits */
+		if (i >= line_len || !isxdigit(line[i]))
+			break;
+		c1 = line[i++];
+		if (i >= line_len || !isxdigit(line[i]))
+			break;
+		c2 = line[i++];
+		/* Followed by a space */
+		if (i < line_len && line[i] && !isspace(line[i]))
+			break;
+		/* Store byte */
+		*(unsigned char *)*buf = (hex(c1) << 4) | hex(c2);
+		*buf += 1;
+		*len -= 1;
+	}
+}
+
+static int read_objdump_output(FILE *f, void **buf, size_t *len)
+{
+	char *line = NULL;
+	size_t line_len;
+	ssize_t ret;
+	int err = 0;
+
+	while (1) {
+		ret = getline(&line, &line_len, f);
+		if (feof(f))
+			break;
+		if (ret < 0) {
+			pr_debug("getline failed\n");
+			err = -1;
+			break;
+		}
+		read_objdump_line(line, ret, buf, len);
+	}
+
+	free(line);
+
+	return err;
+}
+
+static int read_via_objdump(const char *filename, u64 addr, void *buf,
+			    size_t len)
+{
+	char cmd[PATH_MAX * 2];
+	const char *fmt;
+	FILE *f;
+	int ret;
+
+	fmt = "%s -d --start-address=0x%"PRIx64" --stop-address=0x%"PRIx64" %s";
+	ret = snprintf(cmd, sizeof(cmd), fmt, "objdump", addr, addr + len,
+		       filename);
+	if (ret <= 0 || (size_t)ret >= sizeof(cmd))
+		return -1;
+
+	pr_debug("Objdump command is: %s\n", cmd);
+
+	/* Ignore objdump errors */
+	strcat(cmd, " 2>/dev/null");
+
+	f = popen(cmd, "r");
+	if (!f) {
+		pr_debug("popen failed\n");
+		return -1;
+	}
+
+	ret = read_objdump_output(f, &buf, &len);
+	if (len) {
+		pr_debug("objdump read too few bytes\n");
+		if (!ret)
+			ret = len;
+	}
+
+	pclose(f);
+
+	return ret;
+}
+
+static int read_object_code(u64 addr, size_t len, u8 cpumode,
+			    struct thread *thread, struct state *state)
+{
+	struct addr_location al;
+	unsigned char buf1[BUFSZ];
+	unsigned char buf2[BUFSZ];
+	size_t ret_len;
+	u64 objdump_addr;
+	int ret;
+
+	pr_debug("Reading object code for memory address: %#"PRIx64"\n", addr);
+
+	thread__find_addr_map(thread, cpumode, MAP__FUNCTION, addr, &al);
+	if (!al.map || !al.map->dso) {
+		pr_debug("thread__find_addr_map failed\n");
+		return -1;
+	}
+
+	pr_debug("File is: %s\n", al.map->dso->long_name);
+
+	if (al.map->dso->symtab_type == DSO_BINARY_TYPE__KALLSYMS &&
+	    !dso__is_kcore(al.map->dso)) {
+		pr_debug("Unexpected kernel address - skipping\n");
+		return 0;
+	}
+
+	pr_debug("On file address is: %#"PRIx64"\n", al.addr);
+
+	if (len > BUFSZ)
+		len = BUFSZ;
+
+	/* Do not go off the map */
+	if (addr + len > al.map->end)
+		len = al.map->end - addr;
+
+	/* Read the object code using perf */
+	ret_len = dso__data_read_offset(al.map->dso, thread->mg->machine,
+					al.addr, buf1, len);
+	if (ret_len != len) {
+		pr_debug("dso__data_read_offset failed\n");
+		return -1;
+	}
+
+	/*
+	 * Converting addresses for use by objdump requires more information.
+	 * map__load() does that.  See map__rip_2objdump() for details.
+	 */
+	if (map__load(al.map, NULL))
+		return -1;
+
+	/* objdump struggles with kcore - try each map only once */
+	if (dso__is_kcore(al.map->dso)) {
+		size_t d;
+
+		for (d = 0; d < state->done_cnt; d++) {
+			if (state->done[d] == al.map->start) {
+				pr_debug("kcore map tested already");
+				pr_debug(" - skipping\n");
+				return 0;
+			}
+		}
+		if (state->done_cnt >= ARRAY_SIZE(state->done)) {
+			pr_debug("Too many kcore maps - skipping\n");
+			return 0;
+		}
+		state->done[state->done_cnt++] = al.map->start;
+	}
+
+	/* Read the object code using objdump */
+	objdump_addr = map__rip_2objdump(al.map, al.addr);
+	ret = read_via_objdump(al.map->dso->long_name, objdump_addr, buf2, len);
+	if (ret > 0) {
+		/*
+		 * The kernel maps are inaccurate - assume objdump is right in
+		 * that case.
+		 */
+		if (cpumode == PERF_RECORD_MISC_KERNEL ||
+		    cpumode == PERF_RECORD_MISC_GUEST_KERNEL) {
+			len -= ret;
+			if (len) {
+				pr_debug("Reducing len to %zu\n", len);
+			} else if (dso__is_kcore(al.map->dso)) {
+				/*
+				 * objdump cannot handle very large segments
+				 * that may be found in kcore.
+				 */
+				pr_debug("objdump failed for kcore");
+				pr_debug(" - skipping\n");
+				return 0;
+			} else {
+				return -1;
+			}
+		}
+	}
+	if (ret < 0) {
+		pr_debug("read_via_objdump failed\n");
+		return -1;
+	}
+
+	/* The results should be identical */
+	if (memcmp(buf1, buf2, len)) {
+		pr_debug("Bytes read differ from those read by objdump\n");
+		return -1;
+	}
+	pr_debug("Bytes read match those read by objdump\n");
+
+	return 0;
+}
+
+static int process_sample_event(struct machine *machine,
+				struct perf_evlist *evlist,
+				union perf_event *event, struct state *state)
+{
+	struct perf_sample sample;
+	struct thread *thread;
+	u8 cpumode;
+
+	if (perf_evlist__parse_sample(evlist, event, &sample)) {
+		pr_debug("perf_evlist__parse_sample failed\n");
+		return -1;
+	}
+
+	thread = machine__findnew_thread(machine, sample.pid, sample.tid);
+	if (!thread) {
+		pr_debug("machine__findnew_thread failed\n");
+		return -1;
+	}
+
+	cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+
+	return read_object_code(sample.ip, READLEN, cpumode, thread, state);
+}
+
+static int process_event(struct machine *machine, struct perf_evlist *evlist,
+			 union perf_event *event, struct state *state)
+{
+	if (event->header.type == PERF_RECORD_SAMPLE)
+		return process_sample_event(machine, evlist, event, state);
+
+	if (event->header.type == PERF_RECORD_THROTTLE ||
+	    event->header.type == PERF_RECORD_UNTHROTTLE)
+		return 0;
+
+	if (event->header.type < PERF_RECORD_MAX) {
+		int ret;
+
+		ret = machine__process_event(machine, event, NULL);
+		if (ret < 0)
+			pr_debug("machine__process_event failed, event type %u\n",
+				 event->header.type);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int process_events(struct machine *machine, struct perf_evlist *evlist,
+			  struct state *state)
+{
+	union perf_event *event;
+	int i, ret;
+
+	for (i = 0; i < evlist->nr_mmaps; i++) {
+		while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
+			ret = process_event(machine, evlist, event, state);
+			perf_evlist__mmap_consume(evlist, i);
+			if (ret < 0)
+				return ret;
+		}
+	}
+	return 0;
+}
+
+static int comp(const void *a, const void *b)
+{
+	return *(int *)a - *(int *)b;
+}
+
+static void do_sort_something(void)
+{
+	int buf[40960], i;
+
+	for (i = 0; i < (int)ARRAY_SIZE(buf); i++)
+		buf[i] = ARRAY_SIZE(buf) - i - 1;
+
+	qsort(buf, ARRAY_SIZE(buf), sizeof(int), comp);
+
+	for (i = 0; i < (int)ARRAY_SIZE(buf); i++) {
+		if (buf[i] != i) {
+			pr_debug("qsort failed\n");
+			break;
+		}
+	}
+}
+
+static void sort_something(void)
+{
+	int i;
+
+	for (i = 0; i < 10; i++)
+		do_sort_something();
+}
+
+static void syscall_something(void)
+{
+	int pipefd[2];
+	int i;
+
+	for (i = 0; i < 1000; i++) {
+		if (pipe(pipefd) < 0) {
+			pr_debug("pipe failed\n");
+			break;
+		}
+		close(pipefd[1]);
+		close(pipefd[0]);
+	}
+}
+
+static void fs_something(void)
+{
+	const char *test_file_name = "temp-perf-code-reading-test-file--";
+	FILE *f;
+	int i;
+
+	for (i = 0; i < 1000; i++) {
+		f = fopen(test_file_name, "w+");
+		if (f) {
+			fclose(f);
+			unlink(test_file_name);
+		}
+	}
+}
+
+static void do_something(void)
+{
+	fs_something();
+
+	sort_something();
+
+	syscall_something();
+}
+
+enum {
+	TEST_CODE_READING_OK,
+	TEST_CODE_READING_NO_VMLINUX,
+	TEST_CODE_READING_NO_KCORE,
+	TEST_CODE_READING_NO_ACCESS,
+	TEST_CODE_READING_NO_KERNEL_OBJ,
+};
+
+static int do_test_code_reading(bool try_kcore)
+{
+	struct machines machines;
+	struct machine *machine;
+	struct thread *thread;
+	struct record_opts opts = {
+		.mmap_pages	     = UINT_MAX,
+		.user_freq	     = UINT_MAX,
+		.user_interval	     = ULLONG_MAX,
+		.freq		     = 4000,
+		.target		     = {
+			.uses_mmap   = true,
+		},
+	};
+	struct state state = {
+		.done_cnt = 0,
+	};
+	struct thread_map *threads = NULL;
+	struct cpu_map *cpus = NULL;
+	struct perf_evlist *evlist = NULL;
+	struct perf_evsel *evsel = NULL;
+	int err = -1, ret;
+	pid_t pid;
+	struct map *map;
+	bool have_vmlinux, have_kcore, excl_kernel = false;
+
+	pid = getpid();
+
+	machines__init(&machines);
+	machine = &machines.host;
+
+	ret = machine__create_kernel_maps(machine);
+	if (ret < 0) {
+		pr_debug("machine__create_kernel_maps failed\n");
+		goto out_err;
+	}
+
+	/* Force the use of kallsyms instead of vmlinux to try kcore */
+	if (try_kcore)
+		symbol_conf.kallsyms_name = "/proc/kallsyms";
+
+	/* Load kernel map */
+	map = machine->vmlinux_maps[MAP__FUNCTION];
+	ret = map__load(map, NULL);
+	if (ret < 0) {
+		pr_debug("map__load failed\n");
+		goto out_err;
+	}
+	have_vmlinux = dso__is_vmlinux(map->dso);
+	have_kcore = dso__is_kcore(map->dso);
+
+	/* 2nd time through we just try kcore */
+	if (try_kcore && !have_kcore)
+		return TEST_CODE_READING_NO_KCORE;
+
+	/* No point getting kernel events if there is no kernel object */
+	if (!have_vmlinux && !have_kcore)
+		excl_kernel = true;
+
+	threads = thread_map__new_by_tid(pid);
+	if (!threads) {
+		pr_debug("thread_map__new_by_tid failed\n");
+		goto out_err;
+	}
+
+	ret = perf_event__synthesize_thread_map(NULL, threads,
+						perf_event__process, machine, false);
+	if (ret < 0) {
+		pr_debug("perf_event__synthesize_thread_map failed\n");
+		goto out_err;
+	}
+
+	thread = machine__findnew_thread(machine, pid, pid);
+	if (!thread) {
+		pr_debug("machine__findnew_thread failed\n");
+		goto out_err;
+	}
+
+	cpus = cpu_map__new(NULL);
+	if (!cpus) {
+		pr_debug("cpu_map__new failed\n");
+		goto out_err;
+	}
+
+	while (1) {
+		const char *str;
+
+		evlist = perf_evlist__new();
+		if (!evlist) {
+			pr_debug("perf_evlist__new failed\n");
+			goto out_err;
+		}
+
+		perf_evlist__set_maps(evlist, cpus, threads);
+
+		if (excl_kernel)
+			str = "cycles:u";
+		else
+			str = "cycles";
+		pr_debug("Parsing event '%s'\n", str);
+		ret = parse_events(evlist, str);
+		if (ret < 0) {
+			pr_debug("parse_events failed\n");
+			goto out_err;
+		}
+
+		perf_evlist__config(evlist, &opts);
+
+		evsel = perf_evlist__first(evlist);
+
+		evsel->attr.comm = 1;
+		evsel->attr.disabled = 1;
+		evsel->attr.enable_on_exec = 0;
+
+		ret = perf_evlist__open(evlist);
+		if (ret < 0) {
+			if (!excl_kernel) {
+				excl_kernel = true;
+				perf_evlist__set_maps(evlist, NULL, NULL);
+				perf_evlist__delete(evlist);
+				evlist = NULL;
+				continue;
+			}
+			pr_debug("perf_evlist__open failed\n");
+			goto out_err;
+		}
+		break;
+	}
+
+	ret = perf_evlist__mmap(evlist, UINT_MAX, false);
+	if (ret < 0) {
+		pr_debug("perf_evlist__mmap failed\n");
+		goto out_err;
+	}
+
+	perf_evlist__enable(evlist);
+
+	do_something();
+
+	perf_evlist__disable(evlist);
+
+	ret = process_events(machine, evlist, &state);
+	if (ret < 0)
+		goto out_err;
+
+	if (!have_vmlinux && !have_kcore && !try_kcore)
+		err = TEST_CODE_READING_NO_KERNEL_OBJ;
+	else if (!have_vmlinux && !try_kcore)
+		err = TEST_CODE_READING_NO_VMLINUX;
+	else if (excl_kernel)
+		err = TEST_CODE_READING_NO_ACCESS;
+	else
+		err = TEST_CODE_READING_OK;
+out_err:
+	if (evlist) {
+		perf_evlist__delete(evlist);
+	} else {
+		cpu_map__delete(cpus);
+		thread_map__delete(threads);
+	}
+	machines__destroy_kernel_maps(&machines);
+	machine__delete_threads(machine);
+	machines__exit(&machines);
+
+	return err;
+}
+
+int test__code_reading(void)
+{
+	int ret;
+
+	ret = do_test_code_reading(false);
+	if (!ret)
+		ret = do_test_code_reading(true);
+
+	switch (ret) {
+	case TEST_CODE_READING_OK:
+		return 0;
+	case TEST_CODE_READING_NO_VMLINUX:
+		fprintf(stderr, " (no vmlinux)");
+		return 0;
+	case TEST_CODE_READING_NO_KCORE:
+		fprintf(stderr, " (no kcore)");
+		return 0;
+	case TEST_CODE_READING_NO_ACCESS:
+		fprintf(stderr, " (no access)");
+		return 0;
+	case TEST_CODE_READING_NO_KERNEL_OBJ:
+		fprintf(stderr, " (no kernel obj)");
+		return 0;
+	default:
+		return -1;
+	};
+}
diff --git a/tools/perf/tests/dso-data.c b/tools/perf/tests/dso-data.c
new file mode 100644
index 000000000..513e5febb
--- /dev/null
+++ b/tools/perf/tests/dso-data.c
@@ -0,0 +1,365 @@
+#include <stdlib.h>
+#include <linux/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <string.h>
+#include <sys/time.h>
+#include <sys/resource.h>
+#include <api/fs/fs.h>
+#include "util.h"
+#include "machine.h"
+#include "symbol.h"
+#include "tests.h"
+#include "debug.h"
+
+static char *test_file(int size)
+{
+#define TEMPL "/tmp/perf-test-XXXXXX"
+	static char buf_templ[sizeof(TEMPL)];
+	char *templ = buf_templ;
+	int fd, i;
+	unsigned char *buf;
+
+	strcpy(buf_templ, TEMPL);
+#undef TEMPL
+
+	fd = mkstemp(templ);
+	if (fd < 0) {
+		perror("mkstemp failed");
+		return NULL;
+	}
+
+	buf = malloc(size);
+	if (!buf) {
+		close(fd);
+		return NULL;
+	}
+
+	for (i = 0; i < size; i++)
+		buf[i] = (unsigned char) ((int) i % 10);
+
+	if (size != write(fd, buf, size))
+		templ = NULL;
+
+	free(buf);
+	close(fd);
+	return templ;
+}
+
+#define TEST_FILE_SIZE (DSO__DATA_CACHE_SIZE * 20)
+
+struct test_data_offset {
+	off_t offset;
+	u8 data[10];
+	int size;
+};
+
+struct test_data_offset offsets[] = {
+	/* Fill first cache page. */
+	{
+		.offset = 10,
+		.data   = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 },
+		.size   = 10,
+	},
+	/* Read first cache page. */
+	{
+		.offset = 10,
+		.data   = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 },
+		.size   = 10,
+	},
+	/* Fill cache boundary pages. */
+	{
+		.offset = DSO__DATA_CACHE_SIZE - DSO__DATA_CACHE_SIZE % 10,
+		.data   = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 },
+		.size   = 10,
+	},
+	/* Read cache boundary pages. */
+	{
+		.offset = DSO__DATA_CACHE_SIZE - DSO__DATA_CACHE_SIZE % 10,
+		.data   = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 },
+		.size   = 10,
+	},
+	/* Fill final cache page. */
+	{
+		.offset = TEST_FILE_SIZE - 10,
+		.data   = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 },
+		.size   = 10,
+	},
+	/* Read final cache page. */
+	{
+		.offset = TEST_FILE_SIZE - 10,
+		.data   = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 },
+		.size   = 10,
+	},
+	/* Read final cache page. */
+	{
+		.offset = TEST_FILE_SIZE - 3,
+		.data   = { 7, 8, 9, 0, 0, 0, 0, 0, 0, 0 },
+		.size   = 3,
+	},
+};
+
+int test__dso_data(void)
+{
+	struct machine machine;
+	struct dso *dso;
+	char *file = test_file(TEST_FILE_SIZE);
+	size_t i;
+
+	TEST_ASSERT_VAL("No test file", file);
+
+	memset(&machine, 0, sizeof(machine));
+
+	dso = dso__new((const char *)file);
+
+	TEST_ASSERT_VAL("Failed to access to dso",
+			dso__data_fd(dso, &machine) >= 0);
+
+	/* Basic 10 bytes tests. */
+	for (i = 0; i < ARRAY_SIZE(offsets); i++) {
+		struct test_data_offset *data = &offsets[i];
+		ssize_t size;
+		u8 buf[10];
+
+		memset(buf, 0, 10);
+		size = dso__data_read_offset(dso, &machine, data->offset,
+				     buf, 10);
+
+		TEST_ASSERT_VAL("Wrong size", size == data->size);
+		TEST_ASSERT_VAL("Wrong data", !memcmp(buf, data->data, 10));
+	}
+
+	/* Read cross multiple cache pages. */
+	{
+		ssize_t size;
+		int c;
+		u8 *buf;
+
+		buf = malloc(TEST_FILE_SIZE);
+		TEST_ASSERT_VAL("ENOMEM\n", buf);
+
+		/* First iteration to fill caches, second one to read them. */
+		for (c = 0; c < 2; c++) {
+			memset(buf, 0, TEST_FILE_SIZE);
+			size = dso__data_read_offset(dso, &machine, 10,
+						     buf, TEST_FILE_SIZE);
+
+			TEST_ASSERT_VAL("Wrong size",
+				size == (TEST_FILE_SIZE - 10));
+
+			for (i = 0; i < (size_t)size; i++)
+				TEST_ASSERT_VAL("Wrong data",
+					buf[i] == (i % 10));
+		}
+
+		free(buf);
+	}
+
+	dso__delete(dso);
+	unlink(file);
+	return 0;
+}
+
+static long open_files_cnt(void)
+{
+	char path[PATH_MAX];
+	struct dirent *dent;
+	DIR *dir;
+	long nr = 0;
+
+	scnprintf(path, PATH_MAX, "%s/self/fd", procfs__mountpoint());
+	pr_debug("fd path: %s\n", path);
+
+	dir = opendir(path);
+	TEST_ASSERT_VAL("failed to open fd directory", dir);
+
+	while ((dent = readdir(dir)) != NULL) {
+		if (!strcmp(dent->d_name, ".") ||
+		    !strcmp(dent->d_name, ".."))
+			continue;
+
+		nr++;
+	}
+
+	closedir(dir);
+	return nr - 1;
+}
+
+static struct dso **dsos;
+
+static int dsos__create(int cnt, int size)
+{
+	int i;
+
+	dsos = malloc(sizeof(dsos) * cnt);
+	TEST_ASSERT_VAL("failed to alloc dsos array", dsos);
+
+	for (i = 0; i < cnt; i++) {
+		char *file;
+
+		file = test_file(size);
+		TEST_ASSERT_VAL("failed to get dso file", file);
+
+		dsos[i] = dso__new(file);
+		TEST_ASSERT_VAL("failed to get dso", dsos[i]);
+	}
+
+	return 0;
+}
+
+static void dsos__delete(int cnt)
+{
+	int i;
+
+	for (i = 0; i < cnt; i++) {
+		struct dso *dso = dsos[i];
+
+		unlink(dso->name);
+		dso__delete(dso);
+	}
+
+	free(dsos);
+}
+
+static int set_fd_limit(int n)
+{
+	struct rlimit rlim;
+
+	if (getrlimit(RLIMIT_NOFILE, &rlim))
+		return -1;
+
+	pr_debug("file limit %ld, new %d\n", (long) rlim.rlim_cur, n);
+
+	rlim.rlim_cur = n;
+	return setrlimit(RLIMIT_NOFILE, &rlim);
+}
+
+int test__dso_data_cache(void)
+{
+	struct machine machine;
+	long nr_end, nr = open_files_cnt();
+	int dso_cnt, limit, i, fd;
+
+	memset(&machine, 0, sizeof(machine));
+
+	/* set as system limit */
+	limit = nr * 4;
+	TEST_ASSERT_VAL("failed to set file limit", !set_fd_limit(limit));
+
+	/* and this is now our dso open FDs limit */
+	dso_cnt = limit / 2;
+	TEST_ASSERT_VAL("failed to create dsos\n",
+		!dsos__create(dso_cnt, TEST_FILE_SIZE));
+
+	for (i = 0; i < (dso_cnt - 1); i++) {
+		struct dso *dso = dsos[i];
+
+		/*
+		 * Open dsos via dso__data_fd(), it opens the data
+		 * file and keep it open (unless open file limit).
+		 */
+		fd = dso__data_fd(dso, &machine);
+		TEST_ASSERT_VAL("failed to get fd", fd > 0);
+
+		if (i % 2) {
+			#define BUFSIZE 10
+			u8 buf[BUFSIZE];
+			ssize_t n;
+
+			n = dso__data_read_offset(dso, &machine, 0, buf, BUFSIZE);
+			TEST_ASSERT_VAL("failed to read dso", n == BUFSIZE);
+		}
+	}
+
+	/* verify the first one is already open */
+	TEST_ASSERT_VAL("dsos[0] is not open", dsos[0]->data.fd != -1);
+
+	/* open +1 dso to reach the allowed limit */
+	fd = dso__data_fd(dsos[i], &machine);
+	TEST_ASSERT_VAL("failed to get fd", fd > 0);
+
+	/* should force the first one to be closed */
+	TEST_ASSERT_VAL("failed to close dsos[0]", dsos[0]->data.fd == -1);
+
+	/* cleanup everything */
+	dsos__delete(dso_cnt);
+
+	/* Make sure we did not leak any file descriptor. */
+	nr_end = open_files_cnt();
+	pr_debug("nr start %ld, nr stop %ld\n", nr, nr_end);
+	TEST_ASSERT_VAL("failed leadking files", nr == nr_end);
+	return 0;
+}
+
+int test__dso_data_reopen(void)
+{
+	struct machine machine;
+	long nr_end, nr = open_files_cnt();
+	int fd, fd_extra;
+
+#define dso_0 (dsos[0])
+#define dso_1 (dsos[1])
+#define dso_2 (dsos[2])
+
+	memset(&machine, 0, sizeof(machine));
+
+	/*
+	 * Test scenario:
+	 * - create 3 dso objects
+	 * - set process file descriptor limit to current
+	 *   files count + 3
+	 * - test that the first dso gets closed when we
+	 *   reach the files count limit
+	 */
+
+	/* Make sure we are able to open 3 fds anyway */
+	TEST_ASSERT_VAL("failed to set file limit",
+			!set_fd_limit((nr + 3)));
+
+	TEST_ASSERT_VAL("failed to create dsos\n", !dsos__create(3, TEST_FILE_SIZE));
+
+	/* open dso_0 */
+	fd = dso__data_fd(dso_0, &machine);
+	TEST_ASSERT_VAL("failed to get fd", fd > 0);
+
+	/* open dso_1 */
+	fd = dso__data_fd(dso_1, &machine);
+	TEST_ASSERT_VAL("failed to get fd", fd > 0);
+
+	/*
+	 * open extra file descriptor and we just
+	 * reached the files count limit
+	 */
+	fd_extra = open("/dev/null", O_RDONLY);
+	TEST_ASSERT_VAL("failed to open extra fd", fd_extra > 0);
+
+	/* open dso_2 */
+	fd = dso__data_fd(dso_2, &machine);
+	TEST_ASSERT_VAL("failed to get fd", fd > 0);
+
+	/*
+	 * dso_0 should get closed, because we reached
+	 * the file descriptor limit
+	 */
+	TEST_ASSERT_VAL("failed to close dso_0", dso_0->data.fd == -1);
+
+	/* open dso_0 */
+	fd = dso__data_fd(dso_0, &machine);
+	TEST_ASSERT_VAL("failed to get fd", fd > 0);
+
+	/*
+	 * dso_1 should get closed, because we reached
+	 * the file descriptor limit
+	 */
+	TEST_ASSERT_VAL("failed to close dso_1", dso_1->data.fd == -1);
+
+	/* cleanup everything */
+	close(fd_extra);
+	dsos__delete(3);
+
+	/* Make sure we did not leak any file descriptor. */
+	nr_end = open_files_cnt();
+	pr_debug("nr start %ld, nr stop %ld\n", nr, nr_end);
+	TEST_ASSERT_VAL("failed leadking files", nr == nr_end);
+	return 0;
+}
diff --git a/tools/perf/tests/dwarf-unwind.c b/tools/perf/tests/dwarf-unwind.c
new file mode 100644
index 000000000..0bf06bec6
--- /dev/null
+++ b/tools/perf/tests/dwarf-unwind.c
@@ -0,0 +1,179 @@
+#include <linux/compiler.h>
+#include <linux/types.h>
+#include <unistd.h>
+#include "tests.h"
+#include "debug.h"
+#include "machine.h"
+#include "event.h"
+#include "unwind.h"
+#include "perf_regs.h"
+#include "map.h"
+#include "thread.h"
+#include "callchain.h"
+
+/* For bsearch. We try to unwind functions in shared object. */
+#include <stdlib.h>
+
+static int mmap_handler(struct perf_tool *tool __maybe_unused,
+			union perf_event *event,
+			struct perf_sample *sample __maybe_unused,
+			struct machine *machine)
+{
+	return machine__process_mmap2_event(machine, event, NULL);
+}
+
+static int init_live_machine(struct machine *machine)
+{
+	union perf_event event;
+	pid_t pid = getpid();
+
+	return perf_event__synthesize_mmap_events(NULL, &event, pid, pid,
+						  mmap_handler, machine, true);
+}
+
+#define MAX_STACK 8
+
+static int unwind_entry(struct unwind_entry *entry, void *arg)
+{
+	unsigned long *cnt = (unsigned long *) arg;
+	char *symbol = entry->sym ? entry->sym->name : NULL;
+	static const char *funcs[MAX_STACK] = {
+		"test__arch_unwind_sample",
+		"unwind_thread",
+		"compare",
+		"bsearch",
+		"krava_3",
+		"krava_2",
+		"krava_1",
+		"test__dwarf_unwind"
+	};
+
+	if (*cnt >= MAX_STACK) {
+		pr_debug("failed: crossed the max stack value %d\n", MAX_STACK);
+		return -1;
+	}
+
+	if (!symbol) {
+		pr_debug("failed: got unresolved address 0x%" PRIx64 "\n",
+			 entry->ip);
+		return -1;
+	}
+
+	pr_debug("got: %s 0x%" PRIx64 "\n", symbol, entry->ip);
+	return strcmp((const char *) symbol, funcs[(*cnt)++]);
+}
+
+__attribute__ ((noinline))
+static int unwind_thread(struct thread *thread)
+{
+	struct perf_sample sample;
+	unsigned long cnt = 0;
+	int err = -1;
+
+	memset(&sample, 0, sizeof(sample));
+
+	if (test__arch_unwind_sample(&sample, thread)) {
+		pr_debug("failed to get unwind sample\n");
+		goto out;
+	}
+
+	err = unwind__get_entries(unwind_entry, &cnt, thread,
+				  &sample, MAX_STACK);
+	if (err)
+		pr_debug("unwind failed\n");
+	else if (cnt != MAX_STACK) {
+		pr_debug("got wrong number of stack entries %lu != %d\n",
+			 cnt, MAX_STACK);
+		err = -1;
+	}
+
+ out:
+	free(sample.user_stack.data);
+	free(sample.user_regs.regs);
+	return err;
+}
+
+static int global_unwind_retval = -INT_MAX;
+
+__attribute__ ((noinline))
+static int compare(void *p1, void *p2)
+{
+	/* Any possible value should be 'thread' */
+	struct thread *thread = *(struct thread **)p1;
+
+	if (global_unwind_retval == -INT_MAX)
+		global_unwind_retval = unwind_thread(thread);
+
+	return p1 - p2;
+}
+
+__attribute__ ((noinline))
+static int krava_3(struct thread *thread)
+{
+	struct thread *array[2] = {thread, thread};
+	void *fp = &bsearch;
+	/*
+	 * make _bsearch a volatile function pointer to
+	 * prevent potential optimization, which may expand
+	 * bsearch and call compare directly from this function,
+	 * instead of libc shared object.
+	 */
+	void *(*volatile _bsearch)(void *, void *, size_t,
+			size_t, int (*)(void *, void *));
+
+	_bsearch = fp;
+	_bsearch(array, &thread, 2, sizeof(struct thread **), compare);
+	return global_unwind_retval;
+}
+
+__attribute__ ((noinline))
+static int krava_2(struct thread *thread)
+{
+	return krava_3(thread);
+}
+
+__attribute__ ((noinline))
+static int krava_1(struct thread *thread)
+{
+	return krava_2(thread);
+}
+
+int test__dwarf_unwind(void)
+{
+	struct machines machines;
+	struct machine *machine;
+	struct thread *thread;
+	int err = -1;
+
+	machines__init(&machines);
+
+	machine = machines__find(&machines, HOST_KERNEL_ID);
+	if (!machine) {
+		pr_err("Could not get machine\n");
+		return -1;
+	}
+
+	callchain_param.record_mode = CALLCHAIN_DWARF;
+
+	if (init_live_machine(machine)) {
+		pr_err("Could not init machine\n");
+		goto out;
+	}
+
+	if (verbose > 1)
+		machine__fprintf(machine, stderr);
+
+	thread = machine__find_thread(machine, getpid(), getpid());
+	if (!thread) {
+		pr_err("Could not get thread\n");
+		goto out;
+	}
+
+	err = krava_1(thread);
+
+ out:
+	machine__delete_threads(machine);
+	machine__exit(machine);
+	machines__exit(&machines);
+	return err;
+}
diff --git a/tools/perf/tests/evsel-roundtrip-name.c b/tools/perf/tests/evsel-roundtrip-name.c
new file mode 100644
index 000000000..b8d8341b3
--- /dev/null
+++ b/tools/perf/tests/evsel-roundtrip-name.c
@@ -0,0 +1,115 @@
+#include "evlist.h"
+#include "evsel.h"
+#include "parse-events.h"
+#include "tests.h"
+#include "debug.h"
+
+static int perf_evsel__roundtrip_cache_name_test(void)
+{
+	char name[128];
+	int type, op, err = 0, ret = 0, i, idx;
+	struct perf_evsel *evsel;
+	struct perf_evlist *evlist = perf_evlist__new();
+
+        if (evlist == NULL)
+                return -ENOMEM;
+
+	for (type = 0; type < PERF_COUNT_HW_CACHE_MAX; type++) {
+		for (op = 0; op < PERF_COUNT_HW_CACHE_OP_MAX; op++) {
+			/* skip invalid cache type */
+			if (!perf_evsel__is_cache_op_valid(type, op))
+				continue;
+
+			for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) {
+				__perf_evsel__hw_cache_type_op_res_name(type, op, i,
+									name, sizeof(name));
+				err = parse_events(evlist, name);
+				if (err)
+					ret = err;
+			}
+		}
+	}
+
+	idx = 0;
+	evsel = perf_evlist__first(evlist);
+
+	for (type = 0; type < PERF_COUNT_HW_CACHE_MAX; type++) {
+		for (op = 0; op < PERF_COUNT_HW_CACHE_OP_MAX; op++) {
+			/* skip invalid cache type */
+			if (!perf_evsel__is_cache_op_valid(type, op))
+				continue;
+
+			for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) {
+				__perf_evsel__hw_cache_type_op_res_name(type, op, i,
+									name, sizeof(name));
+				if (evsel->idx != idx)
+					continue;
+
+				++idx;
+
+				if (strcmp(perf_evsel__name(evsel), name)) {
+					pr_debug("%s != %s\n", perf_evsel__name(evsel), name);
+					ret = -1;
+				}
+
+				evsel = perf_evsel__next(evsel);
+			}
+		}
+	}
+
+	perf_evlist__delete(evlist);
+	return ret;
+}
+
+static int __perf_evsel__name_array_test(const char *names[], int nr_names)
+{
+	int i, err;
+	struct perf_evsel *evsel;
+	struct perf_evlist *evlist = perf_evlist__new();
+
+        if (evlist == NULL)
+                return -ENOMEM;
+
+	for (i = 0; i < nr_names; ++i) {
+		err = parse_events(evlist, names[i]);
+		if (err) {
+			pr_debug("failed to parse event '%s', err %d\n",
+				 names[i], err);
+			goto out_delete_evlist;
+		}
+	}
+
+	err = 0;
+	evlist__for_each(evlist, evsel) {
+		if (strcmp(perf_evsel__name(evsel), names[evsel->idx])) {
+			--err;
+			pr_debug("%s != %s\n", perf_evsel__name(evsel), names[evsel->idx]);
+		}
+	}
+
+out_delete_evlist:
+	perf_evlist__delete(evlist);
+	return err;
+}
+
+#define perf_evsel__name_array_test(names) \
+	__perf_evsel__name_array_test(names, ARRAY_SIZE(names))
+
+int test__perf_evsel__roundtrip_name_test(void)
+{
+	int err = 0, ret = 0;
+
+	err = perf_evsel__name_array_test(perf_evsel__hw_names);
+	if (err)
+		ret = err;
+
+	err = perf_evsel__name_array_test(perf_evsel__sw_names);
+	if (err)
+		ret = err;
+
+	err = perf_evsel__roundtrip_cache_name_test();
+	if (err)
+		ret = err;
+
+	return ret;
+}
diff --git a/tools/perf/tests/evsel-tp-sched.c b/tools/perf/tests/evsel-tp-sched.c
new file mode 100644
index 000000000..52162425c
--- /dev/null
+++ b/tools/perf/tests/evsel-tp-sched.c
@@ -0,0 +1,82 @@
+#include <traceevent/event-parse.h>
+#include "evsel.h"
+#include "tests.h"
+#include "debug.h"
+
+static int perf_evsel__test_field(struct perf_evsel *evsel, const char *name,
+				  int size, bool should_be_signed)
+{
+	struct format_field *field = perf_evsel__field(evsel, name);
+	int is_signed;
+	int ret = 0;
+
+	if (field == NULL) {
+		pr_debug("%s: \"%s\" field not found!\n", evsel->name, name);
+		return -1;
+	}
+
+	is_signed = !!(field->flags | FIELD_IS_SIGNED);
+	if (should_be_signed && !is_signed) {
+		pr_debug("%s: \"%s\" signedness(%d) is wrong, should be %d\n",
+			 evsel->name, name, is_signed, should_be_signed);
+		ret = -1;
+	}
+
+	if (field->size != size) {
+		pr_debug("%s: \"%s\" size (%d) should be %d!\n",
+			 evsel->name, name, field->size, size);
+		ret = -1;
+	}
+
+	return ret;
+}
+
+int test__perf_evsel__tp_sched_test(void)
+{
+	struct perf_evsel *evsel = perf_evsel__newtp("sched", "sched_switch");
+	int ret = 0;
+
+	if (evsel == NULL) {
+		pr_debug("perf_evsel__new\n");
+		return -1;
+	}
+
+	if (perf_evsel__test_field(evsel, "prev_comm", 16, true))
+		ret = -1;
+
+	if (perf_evsel__test_field(evsel, "prev_pid", 4, true))
+		ret = -1;
+
+	if (perf_evsel__test_field(evsel, "prev_prio", 4, true))
+		ret = -1;
+
+	if (perf_evsel__test_field(evsel, "prev_state", sizeof(long), true))
+		ret = -1;
+
+	if (perf_evsel__test_field(evsel, "next_comm", 16, true))
+		ret = -1;
+
+	if (perf_evsel__test_field(evsel, "next_pid", 4, true))
+		ret = -1;
+
+	if (perf_evsel__test_field(evsel, "next_prio", 4, true))
+		ret = -1;
+
+	perf_evsel__delete(evsel);
+
+	evsel = perf_evsel__newtp("sched", "sched_wakeup");
+
+	if (perf_evsel__test_field(evsel, "comm", 16, true))
+		ret = -1;
+
+	if (perf_evsel__test_field(evsel, "pid", 4, true))
+		ret = -1;
+
+	if (perf_evsel__test_field(evsel, "prio", 4, true))
+		ret = -1;
+
+	if (perf_evsel__test_field(evsel, "target_cpu", 4, true))
+		ret = -1;
+
+	return ret;
+}
diff --git a/tools/perf/tests/fdarray.c b/tools/perf/tests/fdarray.c
new file mode 100644
index 000000000..d24b83795
--- /dev/null
+++ b/tools/perf/tests/fdarray.c
@@ -0,0 +1,174 @@
+#include <api/fd/array.h>
+#include "util/debug.h"
+#include "tests/tests.h"
+
+static void fdarray__init_revents(struct fdarray *fda, short revents)
+{
+	int fd;
+
+	fda->nr = fda->nr_alloc;
+
+	for (fd = 0; fd < fda->nr; ++fd) {
+		fda->entries[fd].fd	 = fda->nr - fd;
+		fda->entries[fd].revents = revents;
+	}
+}
+
+static int fdarray__fprintf_prefix(struct fdarray *fda, const char *prefix, FILE *fp)
+{
+	int printed = 0;
+
+	if (!verbose)
+		return 0;
+
+	printed += fprintf(fp, "\n%s: ", prefix);
+	return printed + fdarray__fprintf(fda, fp);
+}
+
+int test__fdarray__filter(void)
+{
+	int nr_fds, expected_fd[2], fd, err = TEST_FAIL;
+	struct fdarray *fda = fdarray__new(5, 5);
+
+	if (fda == NULL) {
+		pr_debug("\nfdarray__new() failed!");
+		goto out;
+	}
+
+	fdarray__init_revents(fda, POLLIN);
+	nr_fds = fdarray__filter(fda, POLLHUP, NULL);
+	if (nr_fds != fda->nr_alloc) {
+		pr_debug("\nfdarray__filter()=%d != %d shouldn't have filtered anything",
+			 nr_fds, fda->nr_alloc);
+		goto out_delete;
+	}
+
+	fdarray__init_revents(fda, POLLHUP);
+	nr_fds = fdarray__filter(fda, POLLHUP, NULL);
+	if (nr_fds != 0) {
+		pr_debug("\nfdarray__filter()=%d != %d, should have filtered all fds",
+			 nr_fds, fda->nr_alloc);
+		goto out_delete;
+	}
+
+	fdarray__init_revents(fda, POLLHUP);
+	fda->entries[2].revents = POLLIN;
+	expected_fd[0] = fda->entries[2].fd;
+
+	pr_debug("\nfiltering all but fda->entries[2]:");
+	fdarray__fprintf_prefix(fda, "before", stderr);
+	nr_fds = fdarray__filter(fda, POLLHUP, NULL);
+	fdarray__fprintf_prefix(fda, " after", stderr);
+	if (nr_fds != 1) {
+		pr_debug("\nfdarray__filter()=%d != 1, should have left just one event", nr_fds);
+		goto out_delete;
+	}
+
+	if (fda->entries[0].fd != expected_fd[0]) {
+		pr_debug("\nfda->entries[0].fd=%d != %d\n",
+			 fda->entries[0].fd, expected_fd[0]);
+		goto out_delete;
+	}
+
+	fdarray__init_revents(fda, POLLHUP);
+	fda->entries[0].revents = POLLIN;
+	expected_fd[0] = fda->entries[0].fd;
+	fda->entries[3].revents = POLLIN;
+	expected_fd[1] = fda->entries[3].fd;
+
+	pr_debug("\nfiltering all but (fda->entries[0], fda->entries[3]):");
+	fdarray__fprintf_prefix(fda, "before", stderr);
+	nr_fds = fdarray__filter(fda, POLLHUP, NULL);
+	fdarray__fprintf_prefix(fda, " after", stderr);
+	if (nr_fds != 2) {
+		pr_debug("\nfdarray__filter()=%d != 2, should have left just two events",
+			 nr_fds);
+		goto out_delete;
+	}
+
+	for (fd = 0; fd < 2; ++fd) {
+		if (fda->entries[fd].fd != expected_fd[fd]) {
+			pr_debug("\nfda->entries[%d].fd=%d != %d\n", fd,
+				 fda->entries[fd].fd, expected_fd[fd]);
+			goto out_delete;
+		}
+	}
+
+	pr_debug("\n");
+
+	err = 0;
+out_delete:
+	fdarray__delete(fda);
+out:
+	return err;
+}
+
+int test__fdarray__add(void)
+{
+	int err = TEST_FAIL;
+	struct fdarray *fda = fdarray__new(2, 2);
+
+	if (fda == NULL) {
+		pr_debug("\nfdarray__new() failed!");
+		goto out;
+	}
+
+#define FDA_CHECK(_idx, _fd, _revents)					   \
+	if (fda->entries[_idx].fd != _fd) {				   \
+		pr_debug("\n%d: fda->entries[%d](%d) != %d!",		   \
+			 __LINE__, _idx, fda->entries[1].fd, _fd);	   \
+		goto out_delete;					   \
+	}								   \
+	if (fda->entries[_idx].events != (_revents)) {			   \
+		pr_debug("\n%d: fda->entries[%d].revents(%d) != %d!",	   \
+			 __LINE__, _idx, fda->entries[_idx].fd, _revents); \
+		goto out_delete;					   \
+	}
+
+#define FDA_ADD(_idx, _fd, _revents, _nr)				   \
+	if (fdarray__add(fda, _fd, _revents) < 0) {			   \
+		pr_debug("\n%d: fdarray__add(fda, %d, %d) failed!",	   \
+			 __LINE__,_fd, _revents);			   \
+		goto out_delete;					   \
+	}								   \
+	if (fda->nr != _nr) {						   \
+		pr_debug("\n%d: fdarray__add(fda, %d, %d)=%d != %d",	   \
+			 __LINE__,_fd, _revents, fda->nr, _nr);		   \
+		goto out_delete;					   \
+	}								   \
+	FDA_CHECK(_idx, _fd, _revents)
+
+	FDA_ADD(0, 1, POLLIN, 1);
+	FDA_ADD(1, 2, POLLERR, 2);
+
+	fdarray__fprintf_prefix(fda, "before growing array", stderr);
+
+	FDA_ADD(2, 35, POLLHUP, 3);
+
+	if (fda->entries == NULL) {
+		pr_debug("\nfdarray__add(fda, 35, POLLHUP) should have allocated fda->pollfd!");
+		goto out_delete;
+	}
+
+	fdarray__fprintf_prefix(fda, "after 3rd add", stderr);
+
+	FDA_ADD(3, 88, POLLIN | POLLOUT, 4);
+
+	fdarray__fprintf_prefix(fda, "after 4th add", stderr);
+
+	FDA_CHECK(0, 1, POLLIN);
+	FDA_CHECK(1, 2, POLLERR);
+	FDA_CHECK(2, 35, POLLHUP);
+	FDA_CHECK(3, 88, POLLIN | POLLOUT);
+
+#undef FDA_ADD
+#undef FDA_CHECK
+
+	pr_debug("\n");
+
+	err = 0;
+out_delete:
+	fdarray__delete(fda);
+out:
+	return err;
+}
diff --git a/tools/perf/tests/hists_common.c b/tools/perf/tests/hists_common.c
new file mode 100644
index 000000000..a62c09134
--- /dev/null
+++ b/tools/perf/tests/hists_common.c
@@ -0,0 +1,209 @@
+#include "perf.h"
+#include "util/debug.h"
+#include "util/symbol.h"
+#include "util/sort.h"
+#include "util/evsel.h"
+#include "util/evlist.h"
+#include "util/machine.h"
+#include "util/thread.h"
+#include "tests/hists_common.h"
+
+static struct {
+	u32 pid;
+	const char *comm;
+} fake_threads[] = {
+	{ FAKE_PID_PERF1, "perf" },
+	{ FAKE_PID_PERF2, "perf" },
+	{ FAKE_PID_BASH,  "bash" },
+};
+
+static struct {
+	u32 pid;
+	u64 start;
+	const char *filename;
+} fake_mmap_info[] = {
+	{ FAKE_PID_PERF1, FAKE_MAP_PERF,   "perf" },
+	{ FAKE_PID_PERF1, FAKE_MAP_LIBC,   "libc" },
+	{ FAKE_PID_PERF1, FAKE_MAP_KERNEL, "[kernel]" },
+	{ FAKE_PID_PERF2, FAKE_MAP_PERF,   "perf" },
+	{ FAKE_PID_PERF2, FAKE_MAP_LIBC,   "libc" },
+	{ FAKE_PID_PERF2, FAKE_MAP_KERNEL, "[kernel]" },
+	{ FAKE_PID_BASH,  FAKE_MAP_BASH,   "bash" },
+	{ FAKE_PID_BASH,  FAKE_MAP_LIBC,   "libc" },
+	{ FAKE_PID_BASH,  FAKE_MAP_KERNEL, "[kernel]" },
+};
+
+struct fake_sym {
+	u64 start;
+	u64 length;
+	const char *name;
+};
+
+static struct fake_sym perf_syms[] = {
+	{ FAKE_SYM_OFFSET1, FAKE_SYM_LENGTH, "main" },
+	{ FAKE_SYM_OFFSET2, FAKE_SYM_LENGTH, "run_command" },
+	{ FAKE_SYM_OFFSET3, FAKE_SYM_LENGTH, "cmd_record" },
+};
+
+static struct fake_sym bash_syms[] = {
+	{ FAKE_SYM_OFFSET1, FAKE_SYM_LENGTH, "main" },
+	{ FAKE_SYM_OFFSET2, FAKE_SYM_LENGTH, "xmalloc" },
+	{ FAKE_SYM_OFFSET3, FAKE_SYM_LENGTH, "xfree" },
+};
+
+static struct fake_sym libc_syms[] = {
+	{ 700, 100, "malloc" },
+	{ 800, 100, "free" },
+	{ 900, 100, "realloc" },
+	{ FAKE_SYM_OFFSET1, FAKE_SYM_LENGTH, "malloc" },
+	{ FAKE_SYM_OFFSET2, FAKE_SYM_LENGTH, "free" },
+	{ FAKE_SYM_OFFSET3, FAKE_SYM_LENGTH, "realloc" },
+};
+
+static struct fake_sym kernel_syms[] = {
+	{ FAKE_SYM_OFFSET1, FAKE_SYM_LENGTH, "schedule" },
+	{ FAKE_SYM_OFFSET2, FAKE_SYM_LENGTH, "page_fault" },
+	{ FAKE_SYM_OFFSET3, FAKE_SYM_LENGTH, "sys_perf_event_open" },
+};
+
+static struct {
+	const char *dso_name;
+	struct fake_sym *syms;
+	size_t nr_syms;
+} fake_symbols[] = {
+	{ "perf", perf_syms, ARRAY_SIZE(perf_syms) },
+	{ "bash", bash_syms, ARRAY_SIZE(bash_syms) },
+	{ "libc", libc_syms, ARRAY_SIZE(libc_syms) },
+	{ "[kernel]", kernel_syms, ARRAY_SIZE(kernel_syms) },
+};
+
+struct machine *setup_fake_machine(struct machines *machines)
+{
+	struct machine *machine = machines__find(machines, HOST_KERNEL_ID);
+	size_t i;
+
+	if (machine == NULL) {
+		pr_debug("Not enough memory for machine setup\n");
+		return NULL;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(fake_threads); i++) {
+		struct thread *thread;
+
+		thread = machine__findnew_thread(machine, fake_threads[i].pid,
+						 fake_threads[i].pid);
+		if (thread == NULL)
+			goto out;
+
+		thread__set_comm(thread, fake_threads[i].comm, 0);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(fake_mmap_info); i++) {
+		union perf_event fake_mmap_event = {
+			.mmap = {
+				.header = { .misc = PERF_RECORD_MISC_USER, },
+				.pid = fake_mmap_info[i].pid,
+				.tid = fake_mmap_info[i].pid,
+				.start = fake_mmap_info[i].start,
+				.len = FAKE_MAP_LENGTH,
+				.pgoff = 0ULL,
+			},
+		};
+
+		strcpy(fake_mmap_event.mmap.filename,
+		       fake_mmap_info[i].filename);
+
+		machine__process_mmap_event(machine, &fake_mmap_event, NULL);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(fake_symbols); i++) {
+		size_t k;
+		struct dso *dso;
+
+		dso = __dsos__findnew(&machine->user_dsos,
+				      fake_symbols[i].dso_name);
+		if (dso == NULL)
+			goto out;
+
+		/* emulate dso__load() */
+		dso__set_loaded(dso, MAP__FUNCTION);
+
+		for (k = 0; k < fake_symbols[i].nr_syms; k++) {
+			struct symbol *sym;
+			struct fake_sym *fsym = &fake_symbols[i].syms[k];
+
+			sym = symbol__new(fsym->start, fsym->length,
+					  STB_GLOBAL, fsym->name);
+			if (sym == NULL)
+				goto out;
+
+			symbols__insert(&dso->symbols[MAP__FUNCTION], sym);
+		}
+	}
+
+	return machine;
+
+out:
+	pr_debug("Not enough memory for machine setup\n");
+	machine__delete_threads(machine);
+	machine__delete(machine);
+	return NULL;
+}
+
+void print_hists_in(struct hists *hists)
+{
+	int i = 0;
+	struct rb_root *root;
+	struct rb_node *node;
+
+	if (sort__need_collapse)
+		root = &hists->entries_collapsed;
+	else
+		root = hists->entries_in;
+
+	pr_info("----- %s --------\n", __func__);
+	node = rb_first(root);
+	while (node) {
+		struct hist_entry *he;
+
+		he = rb_entry(node, struct hist_entry, rb_node_in);
+
+		if (!he->filtered) {
+			pr_info("%2d: entry: %-8s [%-8s] %20s: period = %"PRIu64"\n",
+				i, thread__comm_str(he->thread),
+				he->ms.map->dso->short_name,
+				he->ms.sym->name, he->stat.period);
+		}
+
+		i++;
+		node = rb_next(node);
+	}
+}
+
+void print_hists_out(struct hists *hists)
+{
+	int i = 0;
+	struct rb_root *root;
+	struct rb_node *node;
+
+	root = &hists->entries;
+
+	pr_info("----- %s --------\n", __func__);
+	node = rb_first(root);
+	while (node) {
+		struct hist_entry *he;
+
+		he = rb_entry(node, struct hist_entry, rb_node);
+
+		if (!he->filtered) {
+			pr_info("%2d: entry: %8s:%5d [%-8s] %20s: period = %"PRIu64"/%"PRIu64"\n",
+				i, thread__comm_str(he->thread), he->thread->tid,
+				he->ms.map->dso->short_name,
+				he->ms.sym->name, he->stat.period,
+				he->stat_acc ? he->stat_acc->period : 0);
+		}
+
+		i++;
+		node = rb_next(node);
+	}
+}
diff --git a/tools/perf/tests/hists_common.h b/tools/perf/tests/hists_common.h
new file mode 100644
index 000000000..888254e86
--- /dev/null
+++ b/tools/perf/tests/hists_common.h
@@ -0,0 +1,75 @@
+#ifndef __PERF_TESTS__HISTS_COMMON_H__
+#define __PERF_TESTS__HISTS_COMMON_H__
+
+struct machine;
+struct machines;
+
+#define FAKE_PID_PERF1  100
+#define FAKE_PID_PERF2  200
+#define FAKE_PID_BASH   300
+
+#define FAKE_MAP_PERF    0x400000
+#define FAKE_MAP_BASH    0x400000
+#define FAKE_MAP_LIBC    0x500000
+#define FAKE_MAP_KERNEL  0xf00000
+#define FAKE_MAP_LENGTH  0x100000
+
+#define FAKE_SYM_OFFSET1  700
+#define FAKE_SYM_OFFSET2  800
+#define FAKE_SYM_OFFSET3  900
+#define FAKE_SYM_LENGTH   100
+
+#define FAKE_IP_PERF_MAIN  FAKE_MAP_PERF + FAKE_SYM_OFFSET1
+#define FAKE_IP_PERF_RUN_COMMAND  FAKE_MAP_PERF + FAKE_SYM_OFFSET2
+#define FAKE_IP_PERF_CMD_RECORD  FAKE_MAP_PERF + FAKE_SYM_OFFSET3
+#define FAKE_IP_BASH_MAIN  FAKE_MAP_BASH + FAKE_SYM_OFFSET1
+#define FAKE_IP_BASH_XMALLOC  FAKE_MAP_BASH + FAKE_SYM_OFFSET2
+#define FAKE_IP_BASH_XFREE  FAKE_MAP_BASH + FAKE_SYM_OFFSET3
+#define FAKE_IP_LIBC_MALLOC  FAKE_MAP_LIBC + FAKE_SYM_OFFSET1
+#define FAKE_IP_LIBC_FREE  FAKE_MAP_LIBC + FAKE_SYM_OFFSET2
+#define FAKE_IP_LIBC_REALLOC  FAKE_MAP_LIBC + FAKE_SYM_OFFSET3
+#define FAKE_IP_KERNEL_SCHEDULE  FAKE_MAP_KERNEL + FAKE_SYM_OFFSET1
+#define FAKE_IP_KERNEL_PAGE_FAULT  FAKE_MAP_KERNEL + FAKE_SYM_OFFSET2
+#define FAKE_IP_KERNEL_SYS_PERF_EVENT_OPEN  FAKE_MAP_KERNEL + FAKE_SYM_OFFSET3
+
+/*
+ * The setup_fake_machine() provides a test environment which consists
+ * of 3 processes that have 3 mappings and in turn, have 3 symbols
+ * respectively.  See below table:
+ *
+ * Command:  Pid  Shared Object               Symbol
+ * .............  .............  ...................
+ *    perf:  100           perf  main
+ *    perf:  100           perf  run_command
+ *    perf:  100           perf  cmd_record
+ *    perf:  100           libc  malloc
+ *    perf:  100           libc  free
+ *    perf:  100           libc  realloc
+ *    perf:  100       [kernel]  schedule
+ *    perf:  100       [kernel]  page_fault
+ *    perf:  100       [kernel]  sys_perf_event_open
+ *    perf:  200           perf  main
+ *    perf:  200           perf  run_command
+ *    perf:  200           perf  cmd_record
+ *    perf:  200           libc  malloc
+ *    perf:  200           libc  free
+ *    perf:  200           libc  realloc
+ *    perf:  200       [kernel]  schedule
+ *    perf:  200       [kernel]  page_fault
+ *    perf:  200       [kernel]  sys_perf_event_open
+ *    bash:  300           bash  main
+ *    bash:  300           bash  xmalloc
+ *    bash:  300           bash  xfree
+ *    bash:  300           libc  malloc
+ *    bash:  300           libc  free
+ *    bash:  300           libc  realloc
+ *    bash:  300       [kernel]  schedule
+ *    bash:  300       [kernel]  page_fault
+ *    bash:  300       [kernel]  sys_perf_event_open
+ */
+struct machine *setup_fake_machine(struct machines *machines);
+
+void print_hists_in(struct hists *hists);
+void print_hists_out(struct hists *hists);
+
+#endif /* __PERF_TESTS__HISTS_COMMON_H__ */
diff --git a/tools/perf/tests/hists_cumulate.c b/tools/perf/tests/hists_cumulate.c
new file mode 100644
index 000000000..186199664
--- /dev/null
+++ b/tools/perf/tests/hists_cumulate.c
@@ -0,0 +1,726 @@
+#include "perf.h"
+#include "util/debug.h"
+#include "util/symbol.h"
+#include "util/sort.h"
+#include "util/evsel.h"
+#include "util/evlist.h"
+#include "util/machine.h"
+#include "util/thread.h"
+#include "util/parse-events.h"
+#include "tests/tests.h"
+#include "tests/hists_common.h"
+
+struct sample {
+	u32 pid;
+	u64 ip;
+	struct thread *thread;
+	struct map *map;
+	struct symbol *sym;
+};
+
+/* For the numbers, see hists_common.c */
+static struct sample fake_samples[] = {
+	/* perf [kernel] schedule() */
+	{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_KERNEL_SCHEDULE, },
+	/* perf [perf]   main() */
+	{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_PERF_MAIN, },
+	/* perf [perf]   cmd_record() */
+	{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_PERF_CMD_RECORD, },
+	/* perf [libc]   malloc() */
+	{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_LIBC_MALLOC, },
+	/* perf [libc]   free() */
+	{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_LIBC_FREE, },
+	/* perf [perf]   main() */
+	{ .pid = FAKE_PID_PERF2, .ip = FAKE_IP_PERF_MAIN, },
+	/* perf [kernel] page_fault() */
+	{ .pid = FAKE_PID_PERF2, .ip = FAKE_IP_KERNEL_PAGE_FAULT, },
+	/* bash [bash]   main() */
+	{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_BASH_MAIN, },
+	/* bash [bash]   xmalloc() */
+	{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_BASH_XMALLOC, },
+	/* bash [kernel] page_fault() */
+	{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_KERNEL_PAGE_FAULT, },
+};
+
+/*
+ * Will be casted to struct ip_callchain which has all 64 bit entries
+ * of nr and ips[].
+ */
+static u64 fake_callchains[][10] = {
+	/*   schedule => run_command => main */
+	{ 3, FAKE_IP_KERNEL_SCHEDULE, FAKE_IP_PERF_RUN_COMMAND, FAKE_IP_PERF_MAIN, },
+	/*   main  */
+	{ 1, FAKE_IP_PERF_MAIN, },
+	/*   cmd_record => run_command => main */
+	{ 3, FAKE_IP_PERF_CMD_RECORD, FAKE_IP_PERF_RUN_COMMAND, FAKE_IP_PERF_MAIN, },
+	/*   malloc => cmd_record => run_command => main */
+	{ 4, FAKE_IP_LIBC_MALLOC, FAKE_IP_PERF_CMD_RECORD, FAKE_IP_PERF_RUN_COMMAND,
+	     FAKE_IP_PERF_MAIN, },
+	/*   free => cmd_record => run_command => main */
+	{ 4, FAKE_IP_LIBC_FREE, FAKE_IP_PERF_CMD_RECORD, FAKE_IP_PERF_RUN_COMMAND,
+	     FAKE_IP_PERF_MAIN, },
+	/*   main */
+	{ 1, FAKE_IP_PERF_MAIN, },
+	/*   page_fault => sys_perf_event_open => run_command => main */
+	{ 4, FAKE_IP_KERNEL_PAGE_FAULT, FAKE_IP_KERNEL_SYS_PERF_EVENT_OPEN,
+	     FAKE_IP_PERF_RUN_COMMAND, FAKE_IP_PERF_MAIN, },
+	/*   main */
+	{ 1, FAKE_IP_BASH_MAIN, },
+	/*   xmalloc => malloc => xmalloc => malloc => xmalloc => main */
+	{ 6, FAKE_IP_BASH_XMALLOC, FAKE_IP_LIBC_MALLOC, FAKE_IP_BASH_XMALLOC,
+	     FAKE_IP_LIBC_MALLOC, FAKE_IP_BASH_XMALLOC, FAKE_IP_BASH_MAIN, },
+	/*   page_fault => malloc => main */
+	{ 3, FAKE_IP_KERNEL_PAGE_FAULT, FAKE_IP_LIBC_MALLOC, FAKE_IP_BASH_MAIN, },
+};
+
+static int add_hist_entries(struct hists *hists, struct machine *machine)
+{
+	struct addr_location al;
+	struct perf_evsel *evsel = hists_to_evsel(hists);
+	struct perf_sample sample = { .period = 1000, };
+	size_t i;
+
+	for (i = 0; i < ARRAY_SIZE(fake_samples); i++) {
+		const union perf_event event = {
+			.header = {
+				.misc = PERF_RECORD_MISC_USER,
+			},
+		};
+		struct hist_entry_iter iter = {
+			.hide_unresolved = false,
+		};
+
+		if (symbol_conf.cumulate_callchain)
+			iter.ops = &hist_iter_cumulative;
+		else
+			iter.ops = &hist_iter_normal;
+
+		sample.pid = fake_samples[i].pid;
+		sample.tid = fake_samples[i].pid;
+		sample.ip = fake_samples[i].ip;
+		sample.callchain = (struct ip_callchain *)fake_callchains[i];
+
+		if (perf_event__preprocess_sample(&event, machine, &al,
+						  &sample) < 0)
+			goto out;
+
+		if (hist_entry_iter__add(&iter, &al, evsel, &sample,
+					 PERF_MAX_STACK_DEPTH, NULL) < 0)
+			goto out;
+
+		fake_samples[i].thread = al.thread;
+		fake_samples[i].map = al.map;
+		fake_samples[i].sym = al.sym;
+	}
+
+	return TEST_OK;
+
+out:
+	pr_debug("Not enough memory for adding a hist entry\n");
+	return TEST_FAIL;
+}
+
+static void del_hist_entries(struct hists *hists)
+{
+	struct hist_entry *he;
+	struct rb_root *root_in;
+	struct rb_root *root_out;
+	struct rb_node *node;
+
+	if (sort__need_collapse)
+		root_in = &hists->entries_collapsed;
+	else
+		root_in = hists->entries_in;
+
+	root_out = &hists->entries;
+
+	while (!RB_EMPTY_ROOT(root_out)) {
+		node = rb_first(root_out);
+
+		he = rb_entry(node, struct hist_entry, rb_node);
+		rb_erase(node, root_out);
+		rb_erase(&he->rb_node_in, root_in);
+		hist_entry__delete(he);
+	}
+}
+
+typedef int (*test_fn_t)(struct perf_evsel *, struct machine *);
+
+#define COMM(he)  (thread__comm_str(he->thread))
+#define DSO(he)   (he->ms.map->dso->short_name)
+#define SYM(he)   (he->ms.sym->name)
+#define CPU(he)   (he->cpu)
+#define PID(he)   (he->thread->tid)
+#define DEPTH(he) (he->callchain->max_depth)
+#define CDSO(cl)  (cl->ms.map->dso->short_name)
+#define CSYM(cl)  (cl->ms.sym->name)
+
+struct result {
+	u64 children;
+	u64 self;
+	const char *comm;
+	const char *dso;
+	const char *sym;
+};
+
+struct callchain_result {
+	u64 nr;
+	struct {
+		const char *dso;
+		const char *sym;
+	} node[10];
+};
+
+static int do_test(struct hists *hists, struct result *expected, size_t nr_expected,
+		   struct callchain_result *expected_callchain, size_t nr_callchain)
+{
+	char buf[32];
+	size_t i, c;
+	struct hist_entry *he;
+	struct rb_root *root;
+	struct rb_node *node;
+	struct callchain_node *cnode;
+	struct callchain_list *clist;
+
+	/*
+	 * adding and deleting hist entries must be done outside of this
+	 * function since TEST_ASSERT_VAL() returns in case of failure.
+	 */
+	hists__collapse_resort(hists, NULL);
+	hists__output_resort(hists, NULL);
+
+	if (verbose > 2) {
+		pr_info("use callchain: %d, cumulate callchain: %d\n",
+			symbol_conf.use_callchain,
+			symbol_conf.cumulate_callchain);
+		print_hists_out(hists);
+	}
+
+	root = &hists->entries;
+	for (node = rb_first(root), i = 0;
+	     node && (he = rb_entry(node, struct hist_entry, rb_node));
+	     node = rb_next(node), i++) {
+		scnprintf(buf, sizeof(buf), "Invalid hist entry #%zd", i);
+
+		TEST_ASSERT_VAL("Incorrect number of hist entry",
+				i < nr_expected);
+		TEST_ASSERT_VAL(buf, he->stat.period == expected[i].self &&
+				!strcmp(COMM(he), expected[i].comm) &&
+				!strcmp(DSO(he), expected[i].dso) &&
+				!strcmp(SYM(he), expected[i].sym));
+
+		if (symbol_conf.cumulate_callchain)
+			TEST_ASSERT_VAL(buf, he->stat_acc->period == expected[i].children);
+
+		if (!symbol_conf.use_callchain)
+			continue;
+
+		/* check callchain entries */
+		root = &he->callchain->node.rb_root;
+		cnode = rb_entry(rb_first(root), struct callchain_node, rb_node);
+
+		c = 0;
+		list_for_each_entry(clist, &cnode->val, list) {
+			scnprintf(buf, sizeof(buf), "Invalid callchain entry #%zd/%zd", i, c);
+
+			TEST_ASSERT_VAL("Incorrect number of callchain entry",
+					c < expected_callchain[i].nr);
+			TEST_ASSERT_VAL(buf,
+				!strcmp(CDSO(clist), expected_callchain[i].node[c].dso) &&
+				!strcmp(CSYM(clist), expected_callchain[i].node[c].sym));
+			c++;
+		}
+		/* TODO: handle multiple child nodes properly */
+		TEST_ASSERT_VAL("Incorrect number of callchain entry",
+				c <= expected_callchain[i].nr);
+	}
+	TEST_ASSERT_VAL("Incorrect number of hist entry",
+			i == nr_expected);
+	TEST_ASSERT_VAL("Incorrect number of callchain entry",
+			!symbol_conf.use_callchain || nr_expected == nr_callchain);
+	return 0;
+}
+
+/* NO callchain + NO children */
+static int test1(struct perf_evsel *evsel, struct machine *machine)
+{
+	int err;
+	struct hists *hists = evsel__hists(evsel);
+	/*
+	 * expected output:
+	 *
+	 * Overhead  Command  Shared Object          Symbol
+	 * ========  =======  =============  ==============
+	 *   20.00%     perf  perf           [.] main
+	 *   10.00%     bash  [kernel]       [k] page_fault
+	 *   10.00%     bash  bash           [.] main
+	 *   10.00%     bash  bash           [.] xmalloc
+	 *   10.00%     perf  [kernel]       [k] page_fault
+	 *   10.00%     perf  [kernel]       [k] schedule
+	 *   10.00%     perf  libc           [.] free
+	 *   10.00%     perf  libc           [.] malloc
+	 *   10.00%     perf  perf           [.] cmd_record
+	 */
+	struct result expected[] = {
+		{ 0, 2000, "perf", "perf",     "main" },
+		{ 0, 1000, "bash", "[kernel]", "page_fault" },
+		{ 0, 1000, "bash", "bash",     "main" },
+		{ 0, 1000, "bash", "bash",     "xmalloc" },
+		{ 0, 1000, "perf", "[kernel]", "page_fault" },
+		{ 0, 1000, "perf", "[kernel]", "schedule" },
+		{ 0, 1000, "perf", "libc",     "free" },
+		{ 0, 1000, "perf", "libc",     "malloc" },
+		{ 0, 1000, "perf", "perf",     "cmd_record" },
+	};
+
+	symbol_conf.use_callchain = false;
+	symbol_conf.cumulate_callchain = false;
+
+	setup_sorting();
+	callchain_register_param(&callchain_param);
+
+	err = add_hist_entries(hists, machine);
+	if (err < 0)
+		goto out;
+
+	err = do_test(hists, expected, ARRAY_SIZE(expected), NULL, 0);
+
+out:
+	del_hist_entries(hists);
+	reset_output_field();
+	return err;
+}
+
+/* callcain + NO children */
+static int test2(struct perf_evsel *evsel, struct machine *machine)
+{
+	int err;
+	struct hists *hists = evsel__hists(evsel);
+	/*
+	 * expected output:
+	 *
+	 * Overhead  Command  Shared Object          Symbol
+	 * ========  =======  =============  ==============
+	 *   20.00%     perf  perf           [.] main
+	 *              |
+	 *              --- main
+	 *
+	 *   10.00%     bash  [kernel]       [k] page_fault
+	 *              |
+	 *              --- page_fault
+	 *                  malloc
+	 *                  main
+	 *
+	 *   10.00%     bash  bash           [.] main
+	 *              |
+	 *              --- main
+	 *
+	 *   10.00%     bash  bash           [.] xmalloc
+	 *              |
+	 *              --- xmalloc
+	 *                  malloc
+	 *                  xmalloc     <--- NOTE: there's a cycle
+	 *                  malloc
+	 *                  xmalloc
+	 *                  main
+	 *
+	 *   10.00%     perf  [kernel]       [k] page_fault
+	 *              |
+	 *              --- page_fault
+	 *                  sys_perf_event_open
+	 *                  run_command
+	 *                  main
+	 *
+	 *   10.00%     perf  [kernel]       [k] schedule
+	 *              |
+	 *              --- schedule
+	 *                  run_command
+	 *                  main
+	 *
+	 *   10.00%     perf  libc           [.] free
+	 *              |
+	 *              --- free
+	 *                  cmd_record
+	 *                  run_command
+	 *                  main
+	 *
+	 *   10.00%     perf  libc           [.] malloc
+	 *              |
+	 *              --- malloc
+	 *                  cmd_record
+	 *                  run_command
+	 *                  main
+	 *
+	 *   10.00%     perf  perf           [.] cmd_record
+	 *              |
+	 *              --- cmd_record
+	 *                  run_command
+	 *                  main
+	 *
+	 */
+	struct result expected[] = {
+		{ 0, 2000, "perf", "perf",     "main" },
+		{ 0, 1000, "bash", "[kernel]", "page_fault" },
+		{ 0, 1000, "bash", "bash",     "main" },
+		{ 0, 1000, "bash", "bash",     "xmalloc" },
+		{ 0, 1000, "perf", "[kernel]", "page_fault" },
+		{ 0, 1000, "perf", "[kernel]", "schedule" },
+		{ 0, 1000, "perf", "libc",     "free" },
+		{ 0, 1000, "perf", "libc",     "malloc" },
+		{ 0, 1000, "perf", "perf",     "cmd_record" },
+	};
+	struct callchain_result expected_callchain[] = {
+		{
+			1, {	{ "perf",     "main" }, },
+		},
+		{
+			3, {	{ "[kernel]", "page_fault" },
+				{ "libc",     "malloc" },
+				{ "bash",     "main" }, },
+		},
+		{
+			1, {	{ "bash",     "main" }, },
+		},
+		{
+			6, {	{ "bash",     "xmalloc" },
+				{ "libc",     "malloc" },
+				{ "bash",     "xmalloc" },
+				{ "libc",     "malloc" },
+				{ "bash",     "xmalloc" },
+				{ "bash",     "main" }, },
+		},
+		{
+			4, {	{ "[kernel]", "page_fault" },
+				{ "[kernel]", "sys_perf_event_open" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			3, {	{ "[kernel]", "schedule" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			4, {	{ "libc",     "free" },
+				{ "perf",     "cmd_record" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			4, {	{ "libc",     "malloc" },
+				{ "perf",     "cmd_record" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			3, {	{ "perf",     "cmd_record" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+	};
+
+	symbol_conf.use_callchain = true;
+	symbol_conf.cumulate_callchain = false;
+
+	setup_sorting();
+	callchain_register_param(&callchain_param);
+
+	err = add_hist_entries(hists, machine);
+	if (err < 0)
+		goto out;
+
+	err = do_test(hists, expected, ARRAY_SIZE(expected),
+		      expected_callchain, ARRAY_SIZE(expected_callchain));
+
+out:
+	del_hist_entries(hists);
+	reset_output_field();
+	return err;
+}
+
+/* NO callchain + children */
+static int test3(struct perf_evsel *evsel, struct machine *machine)
+{
+	int err;
+	struct hists *hists = evsel__hists(evsel);
+	/*
+	 * expected output:
+	 *
+	 * Children      Self  Command  Shared Object                   Symbol
+	 * ========  ========  =======  =============  =======================
+	 *   70.00%    20.00%     perf  perf           [.] main
+	 *   50.00%     0.00%     perf  perf           [.] run_command
+	 *   30.00%    10.00%     bash  bash           [.] main
+	 *   30.00%    10.00%     perf  perf           [.] cmd_record
+	 *   20.00%     0.00%     bash  libc           [.] malloc
+	 *   10.00%    10.00%     bash  [kernel]       [k] page_fault
+	 *   10.00%    10.00%     bash  bash           [.] xmalloc
+	 *   10.00%    10.00%     perf  [kernel]       [k] page_fault
+	 *   10.00%    10.00%     perf  libc           [.] malloc
+	 *   10.00%    10.00%     perf  [kernel]       [k] schedule
+	 *   10.00%    10.00%     perf  libc           [.] free
+	 *   10.00%     0.00%     perf  [kernel]       [k] sys_perf_event_open
+	 */
+	struct result expected[] = {
+		{ 7000, 2000, "perf", "perf",     "main" },
+		{ 5000,    0, "perf", "perf",     "run_command" },
+		{ 3000, 1000, "bash", "bash",     "main" },
+		{ 3000, 1000, "perf", "perf",     "cmd_record" },
+		{ 2000,    0, "bash", "libc",     "malloc" },
+		{ 1000, 1000, "bash", "[kernel]", "page_fault" },
+		{ 1000, 1000, "bash", "bash",     "xmalloc" },
+		{ 1000, 1000, "perf", "[kernel]", "page_fault" },
+		{ 1000, 1000, "perf", "[kernel]", "schedule" },
+		{ 1000, 1000, "perf", "libc",     "free" },
+		{ 1000, 1000, "perf", "libc",     "malloc" },
+		{ 1000,    0, "perf", "[kernel]", "sys_perf_event_open" },
+	};
+
+	symbol_conf.use_callchain = false;
+	symbol_conf.cumulate_callchain = true;
+
+	setup_sorting();
+	callchain_register_param(&callchain_param);
+
+	err = add_hist_entries(hists, machine);
+	if (err < 0)
+		goto out;
+
+	err = do_test(hists, expected, ARRAY_SIZE(expected), NULL, 0);
+
+out:
+	del_hist_entries(hists);
+	reset_output_field();
+	return err;
+}
+
+/* callchain + children */
+static int test4(struct perf_evsel *evsel, struct machine *machine)
+{
+	int err;
+	struct hists *hists = evsel__hists(evsel);
+	/*
+	 * expected output:
+	 *
+	 * Children      Self  Command  Shared Object                   Symbol
+	 * ========  ========  =======  =============  =======================
+	 *   70.00%    20.00%     perf  perf           [.] main
+	 *              |
+	 *              --- main
+	 *
+	 *   50.00%     0.00%     perf  perf           [.] run_command
+	 *              |
+	 *              --- run_command
+	 *                  main
+	 *
+	 *   30.00%    10.00%     bash  bash           [.] main
+	 *              |
+	 *              --- main
+	 *
+	 *   30.00%    10.00%     perf  perf           [.] cmd_record
+	 *              |
+	 *              --- cmd_record
+	 *                  run_command
+	 *                  main
+	 *
+	 *   20.00%     0.00%     bash  libc           [.] malloc
+	 *              |
+	 *              --- malloc
+	 *                 |
+	 *                 |--50.00%-- xmalloc
+	 *                 |           main
+	 *                  --50.00%-- main
+	 *
+	 *   10.00%    10.00%     bash  [kernel]       [k] page_fault
+	 *              |
+	 *              --- page_fault
+	 *                  malloc
+	 *                  main
+	 *
+	 *   10.00%    10.00%     bash  bash           [.] xmalloc
+	 *              |
+	 *              --- xmalloc
+	 *                  malloc
+	 *                  xmalloc     <--- NOTE: there's a cycle
+	 *                  malloc
+	 *                  xmalloc
+	 *                  main
+	 *
+	 *   10.00%     0.00%     perf  [kernel]       [k] sys_perf_event_open
+	 *              |
+	 *              --- sys_perf_event_open
+	 *                  run_command
+	 *                  main
+	 *
+	 *   10.00%    10.00%     perf  [kernel]       [k] page_fault
+	 *              |
+	 *              --- page_fault
+	 *                  sys_perf_event_open
+	 *                  run_command
+	 *                  main
+	 *
+	 *   10.00%    10.00%     perf  [kernel]       [k] schedule
+	 *              |
+	 *              --- schedule
+	 *                  run_command
+	 *                  main
+	 *
+	 *   10.00%    10.00%     perf  libc           [.] free
+	 *              |
+	 *              --- free
+	 *                  cmd_record
+	 *                  run_command
+	 *                  main
+	 *
+	 *   10.00%    10.00%     perf  libc           [.] malloc
+	 *              |
+	 *              --- malloc
+	 *                  cmd_record
+	 *                  run_command
+	 *                  main
+	 *
+	 */
+	struct result expected[] = {
+		{ 7000, 2000, "perf", "perf",     "main" },
+		{ 5000,    0, "perf", "perf",     "run_command" },
+		{ 3000, 1000, "bash", "bash",     "main" },
+		{ 3000, 1000, "perf", "perf",     "cmd_record" },
+		{ 2000,    0, "bash", "libc",     "malloc" },
+		{ 1000, 1000, "bash", "[kernel]", "page_fault" },
+		{ 1000, 1000, "bash", "bash",     "xmalloc" },
+		{ 1000,    0, "perf", "[kernel]", "sys_perf_event_open" },
+		{ 1000, 1000, "perf", "[kernel]", "page_fault" },
+		{ 1000, 1000, "perf", "[kernel]", "schedule" },
+		{ 1000, 1000, "perf", "libc",     "free" },
+		{ 1000, 1000, "perf", "libc",     "malloc" },
+	};
+	struct callchain_result expected_callchain[] = {
+		{
+			1, {	{ "perf",     "main" }, },
+		},
+		{
+			2, {	{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			1, {	{ "bash",     "main" }, },
+		},
+		{
+			3, {	{ "perf",     "cmd_record" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			4, {	{ "libc",     "malloc" },
+				{ "bash",     "xmalloc" },
+				{ "bash",     "main" },
+				{ "bash",     "main" }, },
+		},
+		{
+			3, {	{ "[kernel]", "page_fault" },
+				{ "libc",     "malloc" },
+				{ "bash",     "main" }, },
+		},
+		{
+			6, {	{ "bash",     "xmalloc" },
+				{ "libc",     "malloc" },
+				{ "bash",     "xmalloc" },
+				{ "libc",     "malloc" },
+				{ "bash",     "xmalloc" },
+				{ "bash",     "main" }, },
+		},
+		{
+			3, {	{ "[kernel]", "sys_perf_event_open" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			4, {	{ "[kernel]", "page_fault" },
+				{ "[kernel]", "sys_perf_event_open" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			3, {	{ "[kernel]", "schedule" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			4, {	{ "libc",     "free" },
+				{ "perf",     "cmd_record" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+		{
+			4, {	{ "libc",     "malloc" },
+				{ "perf",     "cmd_record" },
+				{ "perf",     "run_command" },
+				{ "perf",     "main" }, },
+		},
+	};
+
+	symbol_conf.use_callchain = true;
+	symbol_conf.cumulate_callchain = true;
+
+	setup_sorting();
+	callchain_register_param(&callchain_param);
+
+	err = add_hist_entries(hists, machine);
+	if (err < 0)
+		goto out;
+
+	err = do_test(hists, expected, ARRAY_SIZE(expected),
+		      expected_callchain, ARRAY_SIZE(expected_callchain));
+
+out:
+	del_hist_entries(hists);
+	reset_output_field();
+	return err;
+}
+
+int test__hists_cumulate(void)
+{
+	int err = TEST_FAIL;
+	struct machines machines;
+	struct machine *machine;
+	struct perf_evsel *evsel;
+	struct perf_evlist *evlist = perf_evlist__new();
+	size_t i;
+	test_fn_t testcases[] = {
+		test1,
+		test2,
+		test3,
+		test4,
+	};
+
+	TEST_ASSERT_VAL("No memory", evlist);
+
+	err = parse_events(evlist, "cpu-clock");
+	if (err)
+		goto out;
+
+	machines__init(&machines);
+
+	/* setup threads/dso/map/symbols also */
+	machine = setup_fake_machine(&machines);
+	if (!machine)
+		goto out;
+
+	if (verbose > 1)
+		machine__fprintf(machine, stderr);
+
+	evsel = perf_evlist__first(evlist);
+
+	for (i = 0; i < ARRAY_SIZE(testcases); i++) {
+		err = testcases[i](evsel, machine);
+		if (err < 0)
+			break;
+	}
+
+out:
+	/* tear down everything */
+	perf_evlist__delete(evlist);
+	machines__exit(&machines);
+
+	return err;
+}
diff --git a/tools/perf/tests/hists_filter.c b/tools/perf/tests/hists_filter.c
new file mode 100644
index 000000000..59e53db79
--- /dev/null
+++ b/tools/perf/tests/hists_filter.c
@@ -0,0 +1,290 @@
+#include "perf.h"
+#include "util/debug.h"
+#include "util/symbol.h"
+#include "util/sort.h"
+#include "util/evsel.h"
+#include "util/evlist.h"
+#include "util/machine.h"
+#include "util/thread.h"
+#include "util/parse-events.h"
+#include "tests/tests.h"
+#include "tests/hists_common.h"
+
+struct sample {
+	u32 pid;
+	u64 ip;
+	struct thread *thread;
+	struct map *map;
+	struct symbol *sym;
+};
+
+/* For the numbers, see hists_common.c */
+static struct sample fake_samples[] = {
+	/* perf [kernel] schedule() */
+	{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_KERNEL_SCHEDULE, },
+	/* perf [perf]   main() */
+	{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_PERF_MAIN, },
+	/* perf [libc]   malloc() */
+	{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_LIBC_MALLOC, },
+	/* perf [perf]   main() */
+	{ .pid = FAKE_PID_PERF2, .ip = FAKE_IP_PERF_MAIN, }, /* will be merged */
+	/* perf [perf]   cmd_record() */
+	{ .pid = FAKE_PID_PERF2, .ip = FAKE_IP_PERF_CMD_RECORD, },
+	/* perf [kernel] page_fault() */
+	{ .pid = FAKE_PID_PERF2, .ip = FAKE_IP_KERNEL_PAGE_FAULT, },
+	/* bash [bash]   main() */
+	{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_BASH_MAIN, },
+	/* bash [bash]   xmalloc() */
+	{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_BASH_XMALLOC, },
+	/* bash [libc]   malloc() */
+	{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_LIBC_MALLOC, },
+	/* bash [kernel] page_fault() */
+	{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_KERNEL_PAGE_FAULT, },
+};
+
+static int add_hist_entries(struct perf_evlist *evlist,
+			    struct machine *machine)
+{
+	struct perf_evsel *evsel;
+	struct addr_location al;
+	struct perf_sample sample = { .period = 100, };
+	size_t i;
+
+	/*
+	 * each evsel will have 10 samples but the 4th sample
+	 * (perf [perf] main) will be collapsed to an existing entry
+	 * so total 9 entries will be in the tree.
+	 */
+	evlist__for_each(evlist, evsel) {
+		for (i = 0; i < ARRAY_SIZE(fake_samples); i++) {
+			const union perf_event event = {
+				.header = {
+					.misc = PERF_RECORD_MISC_USER,
+				},
+			};
+			struct hist_entry_iter iter = {
+				.ops = &hist_iter_normal,
+				.hide_unresolved = false,
+			};
+			struct hists *hists = evsel__hists(evsel);
+
+			/* make sure it has no filter at first */
+			hists->thread_filter = NULL;
+			hists->dso_filter = NULL;
+			hists->symbol_filter_str = NULL;
+
+			sample.pid = fake_samples[i].pid;
+			sample.tid = fake_samples[i].pid;
+			sample.ip = fake_samples[i].ip;
+
+			if (perf_event__preprocess_sample(&event, machine, &al,
+							  &sample) < 0)
+				goto out;
+
+			if (hist_entry_iter__add(&iter, &al, evsel, &sample,
+						 PERF_MAX_STACK_DEPTH, NULL) < 0)
+				goto out;
+
+			fake_samples[i].thread = al.thread;
+			fake_samples[i].map = al.map;
+			fake_samples[i].sym = al.sym;
+		}
+	}
+
+	return 0;
+
+out:
+	pr_debug("Not enough memory for adding a hist entry\n");
+	return TEST_FAIL;
+}
+
+int test__hists_filter(void)
+{
+	int err = TEST_FAIL;
+	struct machines machines;
+	struct machine *machine;
+	struct perf_evsel *evsel;
+	struct perf_evlist *evlist = perf_evlist__new();
+
+	TEST_ASSERT_VAL("No memory", evlist);
+
+	err = parse_events(evlist, "cpu-clock");
+	if (err)
+		goto out;
+	err = parse_events(evlist, "task-clock");
+	if (err)
+		goto out;
+
+	/* default sort order (comm,dso,sym) will be used */
+	if (setup_sorting() < 0)
+		goto out;
+
+	machines__init(&machines);
+
+	/* setup threads/dso/map/symbols also */
+	machine = setup_fake_machine(&machines);
+	if (!machine)
+		goto out;
+
+	if (verbose > 1)
+		machine__fprintf(machine, stderr);
+
+	/* process sample events */
+	err = add_hist_entries(evlist, machine);
+	if (err < 0)
+		goto out;
+
+	evlist__for_each(evlist, evsel) {
+		struct hists *hists = evsel__hists(evsel);
+
+		hists__collapse_resort(hists, NULL);
+		hists__output_resort(hists, NULL);
+
+		if (verbose > 2) {
+			pr_info("Normal histogram\n");
+			print_hists_out(hists);
+		}
+
+		TEST_ASSERT_VAL("Invalid nr samples",
+				hists->stats.nr_events[PERF_RECORD_SAMPLE] == 10);
+		TEST_ASSERT_VAL("Invalid nr hist entries",
+				hists->nr_entries == 9);
+		TEST_ASSERT_VAL("Invalid total period",
+				hists->stats.total_period == 1000);
+		TEST_ASSERT_VAL("Unmatched nr samples",
+				hists->stats.nr_events[PERF_RECORD_SAMPLE] ==
+				hists->stats.nr_non_filtered_samples);
+		TEST_ASSERT_VAL("Unmatched nr hist entries",
+				hists->nr_entries == hists->nr_non_filtered_entries);
+		TEST_ASSERT_VAL("Unmatched total period",
+				hists->stats.total_period ==
+				hists->stats.total_non_filtered_period);
+
+		/* now applying thread filter for 'bash' */
+		hists->thread_filter = fake_samples[9].thread;
+		hists__filter_by_thread(hists);
+
+		if (verbose > 2) {
+			pr_info("Histogram for thread filter\n");
+			print_hists_out(hists);
+		}
+
+		/* normal stats should be invariant */
+		TEST_ASSERT_VAL("Invalid nr samples",
+				hists->stats.nr_events[PERF_RECORD_SAMPLE] == 10);
+		TEST_ASSERT_VAL("Invalid nr hist entries",
+				hists->nr_entries == 9);
+		TEST_ASSERT_VAL("Invalid total period",
+				hists->stats.total_period == 1000);
+
+		/* but filter stats are changed */
+		TEST_ASSERT_VAL("Unmatched nr samples for thread filter",
+				hists->stats.nr_non_filtered_samples == 4);
+		TEST_ASSERT_VAL("Unmatched nr hist entries for thread filter",
+				hists->nr_non_filtered_entries == 4);
+		TEST_ASSERT_VAL("Unmatched total period for thread filter",
+				hists->stats.total_non_filtered_period == 400);
+
+		/* remove thread filter first */
+		hists->thread_filter = NULL;
+		hists__filter_by_thread(hists);
+
+		/* now applying dso filter for 'kernel' */
+		hists->dso_filter = fake_samples[0].map->dso;
+		hists__filter_by_dso(hists);
+
+		if (verbose > 2) {
+			pr_info("Histogram for dso filter\n");
+			print_hists_out(hists);
+		}
+
+		/* normal stats should be invariant */
+		TEST_ASSERT_VAL("Invalid nr samples",
+				hists->stats.nr_events[PERF_RECORD_SAMPLE] == 10);
+		TEST_ASSERT_VAL("Invalid nr hist entries",
+				hists->nr_entries == 9);
+		TEST_ASSERT_VAL("Invalid total period",
+				hists->stats.total_period == 1000);
+
+		/* but filter stats are changed */
+		TEST_ASSERT_VAL("Unmatched nr samples for dso filter",
+				hists->stats.nr_non_filtered_samples == 3);
+		TEST_ASSERT_VAL("Unmatched nr hist entries for dso filter",
+				hists->nr_non_filtered_entries == 3);
+		TEST_ASSERT_VAL("Unmatched total period for dso filter",
+				hists->stats.total_non_filtered_period == 300);
+
+		/* remove dso filter first */
+		hists->dso_filter = NULL;
+		hists__filter_by_dso(hists);
+
+		/*
+		 * now applying symbol filter for 'main'.  Also note that
+		 * there's 3 samples that have 'main' symbol but the 4th
+		 * entry of fake_samples was collapsed already so it won't
+		 * be counted as a separate entry but the sample count and
+		 * total period will be remained.
+		 */
+		hists->symbol_filter_str = "main";
+		hists__filter_by_symbol(hists);
+
+		if (verbose > 2) {
+			pr_info("Histogram for symbol filter\n");
+			print_hists_out(hists);
+		}
+
+		/* normal stats should be invariant */
+		TEST_ASSERT_VAL("Invalid nr samples",
+				hists->stats.nr_events[PERF_RECORD_SAMPLE] == 10);
+		TEST_ASSERT_VAL("Invalid nr hist entries",
+				hists->nr_entries == 9);
+		TEST_ASSERT_VAL("Invalid total period",
+				hists->stats.total_period == 1000);
+
+		/* but filter stats are changed */
+		TEST_ASSERT_VAL("Unmatched nr samples for symbol filter",
+				hists->stats.nr_non_filtered_samples == 3);
+		TEST_ASSERT_VAL("Unmatched nr hist entries for symbol filter",
+				hists->nr_non_filtered_entries == 2);
+		TEST_ASSERT_VAL("Unmatched total period for symbol filter",
+				hists->stats.total_non_filtered_period == 300);
+
+		/* now applying all filters at once. */
+		hists->thread_filter = fake_samples[1].thread;
+		hists->dso_filter = fake_samples[1].map->dso;
+		hists__filter_by_thread(hists);
+		hists__filter_by_dso(hists);
+
+		if (verbose > 2) {
+			pr_info("Histogram for all filters\n");
+			print_hists_out(hists);
+		}
+
+		/* normal stats should be invariant */
+		TEST_ASSERT_VAL("Invalid nr samples",
+				hists->stats.nr_events[PERF_RECORD_SAMPLE] == 10);
+		TEST_ASSERT_VAL("Invalid nr hist entries",
+				hists->nr_entries == 9);
+		TEST_ASSERT_VAL("Invalid total period",
+				hists->stats.total_period == 1000);
+
+		/* but filter stats are changed */
+		TEST_ASSERT_VAL("Unmatched nr samples for all filter",
+				hists->stats.nr_non_filtered_samples == 2);
+		TEST_ASSERT_VAL("Unmatched nr hist entries for all filter",
+				hists->nr_non_filtered_entries == 1);
+		TEST_ASSERT_VAL("Unmatched total period for all filter",
+				hists->stats.total_non_filtered_period == 200);
+	}
+
+
+	err = TEST_OK;
+
+out:
+	/* tear down everything */
+	perf_evlist__delete(evlist);
+	reset_output_field();
+	machines__exit(&machines);
+
+	return err;
+}
diff --git a/tools/perf/tests/hists_link.c b/tools/perf/tests/hists_link.c
new file mode 100644
index 000000000..278ba8344
--- /dev/null
+++ b/tools/perf/tests/hists_link.c
@@ -0,0 +1,346 @@
+#include "perf.h"
+#include "tests.h"
+#include "debug.h"
+#include "symbol.h"
+#include "sort.h"
+#include "evsel.h"
+#include "evlist.h"
+#include "machine.h"
+#include "thread.h"
+#include "parse-events.h"
+#include "hists_common.h"
+
+struct sample {
+	u32 pid;
+	u64 ip;
+	struct thread *thread;
+	struct map *map;
+	struct symbol *sym;
+};
+
+/* For the numbers, see hists_common.c */
+static struct sample fake_common_samples[] = {
+	/* perf [kernel] schedule() */
+	{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_KERNEL_SCHEDULE, },
+	/* perf [perf]   main() */
+	{ .pid = FAKE_PID_PERF2, .ip = FAKE_IP_PERF_MAIN, },
+	/* perf [perf]   cmd_record() */
+	{ .pid = FAKE_PID_PERF2, .ip = FAKE_IP_PERF_CMD_RECORD, },
+	/* bash [bash]   xmalloc() */
+	{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_BASH_XMALLOC, },
+	/* bash [libc]   malloc() */
+	{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_LIBC_MALLOC, },
+};
+
+static struct sample fake_samples[][5] = {
+	{
+		/* perf [perf]   run_command() */
+		{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_PERF_RUN_COMMAND, },
+		/* perf [libc]   malloc() */
+		{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_LIBC_MALLOC, },
+		/* perf [kernel] page_fault() */
+		{ .pid = FAKE_PID_PERF1, .ip = FAKE_IP_KERNEL_PAGE_FAULT, },
+		/* perf [kernel] sys_perf_event_open() */
+		{ .pid = FAKE_PID_PERF2, .ip = FAKE_IP_KERNEL_SYS_PERF_EVENT_OPEN, },
+		/* bash [libc]   free() */
+		{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_LIBC_FREE, },
+	},
+	{
+		/* perf [libc]   free() */
+		{ .pid = FAKE_PID_PERF2, .ip = FAKE_IP_LIBC_FREE, },
+		/* bash [libc]   malloc() */
+		{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_LIBC_MALLOC, }, /* will be merged */
+		/* bash [bash]   xfee() */
+		{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_BASH_XFREE, },
+		/* bash [libc]   realloc() */
+		{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_LIBC_REALLOC, },
+		/* bash [kernel] page_fault() */
+		{ .pid = FAKE_PID_BASH,  .ip = FAKE_IP_KERNEL_PAGE_FAULT, },
+	},
+};
+
+static int add_hist_entries(struct perf_evlist *evlist, struct machine *machine)
+{
+	struct perf_evsel *evsel;
+	struct addr_location al;
+	struct hist_entry *he;
+	struct perf_sample sample = { .period = 1, };
+	size_t i = 0, k;
+
+	/*
+	 * each evsel will have 10 samples - 5 common and 5 distinct.
+	 * However the second evsel also has a collapsed entry for
+	 * "bash [libc] malloc" so total 9 entries will be in the tree.
+	 */
+	evlist__for_each(evlist, evsel) {
+		struct hists *hists = evsel__hists(evsel);
+
+		for (k = 0; k < ARRAY_SIZE(fake_common_samples); k++) {
+			const union perf_event event = {
+				.header = {
+					.misc = PERF_RECORD_MISC_USER,
+				},
+			};
+
+			sample.pid = fake_common_samples[k].pid;
+			sample.tid = fake_common_samples[k].pid;
+			sample.ip = fake_common_samples[k].ip;
+			if (perf_event__preprocess_sample(&event, machine, &al,
+							  &sample) < 0)
+				goto out;
+
+			he = __hists__add_entry(hists, &al, NULL,
+						NULL, NULL, 1, 1, 0, true);
+			if (he == NULL)
+				goto out;
+
+			fake_common_samples[k].thread = al.thread;
+			fake_common_samples[k].map = al.map;
+			fake_common_samples[k].sym = al.sym;
+		}
+
+		for (k = 0; k < ARRAY_SIZE(fake_samples[i]); k++) {
+			const union perf_event event = {
+				.header = {
+					.misc = PERF_RECORD_MISC_USER,
+				},
+			};
+
+			sample.pid = fake_samples[i][k].pid;
+			sample.tid = fake_samples[i][k].pid;
+			sample.ip = fake_samples[i][k].ip;
+			if (perf_event__preprocess_sample(&event, machine, &al,
+							  &sample) < 0)
+				goto out;
+
+			he = __hists__add_entry(hists, &al, NULL,
+						NULL, NULL, 1, 1, 0, true);
+			if (he == NULL)
+				goto out;
+
+			fake_samples[i][k].thread = al.thread;
+			fake_samples[i][k].map = al.map;
+			fake_samples[i][k].sym = al.sym;
+		}
+		i++;
+	}
+
+	return 0;
+
+out:
+	pr_debug("Not enough memory for adding a hist entry\n");
+	return -1;
+}
+
+static int find_sample(struct sample *samples, size_t nr_samples,
+		       struct thread *t, struct map *m, struct symbol *s)
+{
+	while (nr_samples--) {
+		if (samples->thread == t && samples->map == m &&
+		    samples->sym == s)
+			return 1;
+		samples++;
+	}
+	return 0;
+}
+
+static int __validate_match(struct hists *hists)
+{
+	size_t count = 0;
+	struct rb_root *root;
+	struct rb_node *node;
+
+	/*
+	 * Only entries from fake_common_samples should have a pair.
+	 */
+	if (sort__need_collapse)
+		root = &hists->entries_collapsed;
+	else
+		root = hists->entries_in;
+
+	node = rb_first(root);
+	while (node) {
+		struct hist_entry *he;
+
+		he = rb_entry(node, struct hist_entry, rb_node_in);
+
+		if (hist_entry__has_pairs(he)) {
+			if (find_sample(fake_common_samples,
+					ARRAY_SIZE(fake_common_samples),
+					he->thread, he->ms.map, he->ms.sym)) {
+				count++;
+			} else {
+				pr_debug("Can't find the matched entry\n");
+				return -1;
+			}
+		}
+
+		node = rb_next(node);
+	}
+
+	if (count != ARRAY_SIZE(fake_common_samples)) {
+		pr_debug("Invalid count for matched entries: %zd of %zd\n",
+			 count, ARRAY_SIZE(fake_common_samples));
+		return -1;
+	}
+
+	return 0;
+}
+
+static int validate_match(struct hists *leader, struct hists *other)
+{
+	return __validate_match(leader) || __validate_match(other);
+}
+
+static int __validate_link(struct hists *hists, int idx)
+{
+	size_t count = 0;
+	size_t count_pair = 0;
+	size_t count_dummy = 0;
+	struct rb_root *root;
+	struct rb_node *node;
+
+	/*
+	 * Leader hists (idx = 0) will have dummy entries from other,
+	 * and some entries will have no pair.  However every entry
+	 * in other hists should have (dummy) pair.
+	 */
+	if (sort__need_collapse)
+		root = &hists->entries_collapsed;
+	else
+		root = hists->entries_in;
+
+	node = rb_first(root);
+	while (node) {
+		struct hist_entry *he;
+
+		he = rb_entry(node, struct hist_entry, rb_node_in);
+
+		if (hist_entry__has_pairs(he)) {
+			if (!find_sample(fake_common_samples,
+					 ARRAY_SIZE(fake_common_samples),
+					 he->thread, he->ms.map, he->ms.sym) &&
+			    !find_sample(fake_samples[idx],
+					 ARRAY_SIZE(fake_samples[idx]),
+					 he->thread, he->ms.map, he->ms.sym)) {
+				count_dummy++;
+			}
+			count_pair++;
+		} else if (idx) {
+			pr_debug("A entry from the other hists should have pair\n");
+			return -1;
+		}
+
+		count++;
+		node = rb_next(node);
+	}
+
+	/*
+	 * Note that we have a entry collapsed in the other (idx = 1) hists.
+	 */
+	if (idx == 0) {
+		if (count_dummy != ARRAY_SIZE(fake_samples[1]) - 1) {
+			pr_debug("Invalid count of dummy entries: %zd of %zd\n",
+				 count_dummy, ARRAY_SIZE(fake_samples[1]) - 1);
+			return -1;
+		}
+		if (count != count_pair + ARRAY_SIZE(fake_samples[0])) {
+			pr_debug("Invalid count of total leader entries: %zd of %zd\n",
+				 count, count_pair + ARRAY_SIZE(fake_samples[0]));
+			return -1;
+		}
+	} else {
+		if (count != count_pair) {
+			pr_debug("Invalid count of total other entries: %zd of %zd\n",
+				 count, count_pair);
+			return -1;
+		}
+		if (count_dummy > 0) {
+			pr_debug("Other hists should not have dummy entries: %zd\n",
+				 count_dummy);
+			return -1;
+		}
+	}
+
+	return 0;
+}
+
+static int validate_link(struct hists *leader, struct hists *other)
+{
+	return __validate_link(leader, 0) || __validate_link(other, 1);
+}
+
+int test__hists_link(void)
+{
+	int err = -1;
+	struct hists *hists, *first_hists;
+	struct machines machines;
+	struct machine *machine = NULL;
+	struct perf_evsel *evsel, *first;
+	struct perf_evlist *evlist = perf_evlist__new();
+
+	if (evlist == NULL)
+                return -ENOMEM;
+
+	err = parse_events(evlist, "cpu-clock");
+	if (err)
+		goto out;
+	err = parse_events(evlist, "task-clock");
+	if (err)
+		goto out;
+
+	/* default sort order (comm,dso,sym) will be used */
+	if (setup_sorting() < 0)
+		goto out;
+
+	machines__init(&machines);
+
+	/* setup threads/dso/map/symbols also */
+	machine = setup_fake_machine(&machines);
+	if (!machine)
+		goto out;
+
+	if (verbose > 1)
+		machine__fprintf(machine, stderr);
+
+	/* process sample events */
+	err = add_hist_entries(evlist, machine);
+	if (err < 0)
+		goto out;
+
+	evlist__for_each(evlist, evsel) {
+		hists = evsel__hists(evsel);
+		hists__collapse_resort(hists, NULL);
+
+		if (verbose > 2)
+			print_hists_in(hists);
+	}
+
+	first = perf_evlist__first(evlist);
+	evsel = perf_evlist__last(evlist);
+
+	first_hists = evsel__hists(first);
+	hists = evsel__hists(evsel);
+
+	/* match common entries */
+	hists__match(first_hists, hists);
+	err = validate_match(first_hists, hists);
+	if (err)
+		goto out;
+
+	/* link common and/or dummy entries */
+	hists__link(first_hists, hists);
+	err = validate_link(first_hists, hists);
+	if (err)
+		goto out;
+
+	err = 0;
+
+out:
+	/* tear down everything */
+	perf_evlist__delete(evlist);
+	reset_output_field();
+	machines__exit(&machines);
+
+	return err;
+}
diff --git a/tools/perf/tests/hists_output.c b/tools/perf/tests/hists_output.c
new file mode 100644
index 000000000..b52c9faea
--- /dev/null
+++ b/tools/perf/tests/hists_output.c
@@ -0,0 +1,621 @@
+#include "perf.h"
+#include "util/debug.h"
+#include "util/symbol.h"
+#include "util/sort.h"
+#include "util/evsel.h"
+#include "util/evlist.h"
+#include "util/machine.h"
+#include "util/thread.h"
+#include "util/parse-events.h"
+#include "tests/tests.h"
+#include "tests/hists_common.h"
+
+struct sample {
+	u32 cpu;
+	u32 pid;
+	u64 ip;
+	struct thread *thread;
+	struct map *map;
+	struct symbol *sym;
+};
+
+/* For the numbers, see hists_common.c */
+static struct sample fake_samples[] = {
+	/* perf [kernel] schedule() */
+	{ .cpu = 0, .pid = FAKE_PID_PERF1, .ip = FAKE_IP_KERNEL_SCHEDULE, },
+	/* perf [perf]   main() */
+	{ .cpu = 1, .pid = FAKE_PID_PERF1, .ip = FAKE_IP_PERF_MAIN, },
+	/* perf [perf]   cmd_record() */
+	{ .cpu = 1, .pid = FAKE_PID_PERF1, .ip = FAKE_IP_PERF_CMD_RECORD, },
+	/* perf [libc]   malloc() */
+	{ .cpu = 1, .pid = FAKE_PID_PERF1, .ip = FAKE_IP_LIBC_MALLOC, },
+	/* perf [libc]   free() */
+	{ .cpu = 2, .pid = FAKE_PID_PERF1, .ip = FAKE_IP_LIBC_FREE, },
+	/* perf [perf]   main() */
+	{ .cpu = 2, .pid = FAKE_PID_PERF2, .ip = FAKE_IP_PERF_MAIN, },
+	/* perf [kernel] page_fault() */
+	{ .cpu = 2, .pid = FAKE_PID_PERF2, .ip = FAKE_IP_KERNEL_PAGE_FAULT, },
+	/* bash [bash]   main() */
+	{ .cpu = 3, .pid = FAKE_PID_BASH,  .ip = FAKE_IP_BASH_MAIN, },
+	/* bash [bash]   xmalloc() */
+	{ .cpu = 0, .pid = FAKE_PID_BASH,  .ip = FAKE_IP_BASH_XMALLOC, },
+	/* bash [kernel] page_fault() */
+	{ .cpu = 1, .pid = FAKE_PID_BASH,  .ip = FAKE_IP_KERNEL_PAGE_FAULT, },
+};
+
+static int add_hist_entries(struct hists *hists, struct machine *machine)
+{
+	struct addr_location al;
+	struct perf_evsel *evsel = hists_to_evsel(hists);
+	struct perf_sample sample = { .period = 100, };
+	size_t i;
+
+	for (i = 0; i < ARRAY_SIZE(fake_samples); i++) {
+		const union perf_event event = {
+			.header = {
+				.misc = PERF_RECORD_MISC_USER,
+			},
+		};
+		struct hist_entry_iter iter = {
+			.ops = &hist_iter_normal,
+			.hide_unresolved = false,
+		};
+
+		sample.cpu = fake_samples[i].cpu;
+		sample.pid = fake_samples[i].pid;
+		sample.tid = fake_samples[i].pid;
+		sample.ip = fake_samples[i].ip;
+
+		if (perf_event__preprocess_sample(&event, machine, &al,
+						  &sample) < 0)
+			goto out;
+
+		if (hist_entry_iter__add(&iter, &al, evsel, &sample,
+					 PERF_MAX_STACK_DEPTH, NULL) < 0)
+			goto out;
+
+		fake_samples[i].thread = al.thread;
+		fake_samples[i].map = al.map;
+		fake_samples[i].sym = al.sym;
+	}
+
+	return TEST_OK;
+
+out:
+	pr_debug("Not enough memory for adding a hist entry\n");
+	return TEST_FAIL;
+}
+
+static void del_hist_entries(struct hists *hists)
+{
+	struct hist_entry *he;
+	struct rb_root *root_in;
+	struct rb_root *root_out;
+	struct rb_node *node;
+
+	if (sort__need_collapse)
+		root_in = &hists->entries_collapsed;
+	else
+		root_in = hists->entries_in;
+
+	root_out = &hists->entries;
+
+	while (!RB_EMPTY_ROOT(root_out)) {
+		node = rb_first(root_out);
+
+		he = rb_entry(node, struct hist_entry, rb_node);
+		rb_erase(node, root_out);
+		rb_erase(&he->rb_node_in, root_in);
+		hist_entry__delete(he);
+	}
+}
+
+typedef int (*test_fn_t)(struct perf_evsel *, struct machine *);
+
+#define COMM(he)  (thread__comm_str(he->thread))
+#define DSO(he)   (he->ms.map->dso->short_name)
+#define SYM(he)   (he->ms.sym->name)
+#define CPU(he)   (he->cpu)
+#define PID(he)   (he->thread->tid)
+
+/* default sort keys (no field) */
+static int test1(struct perf_evsel *evsel, struct machine *machine)
+{
+	int err;
+	struct hists *hists = evsel__hists(evsel);
+	struct hist_entry *he;
+	struct rb_root *root;
+	struct rb_node *node;
+
+	field_order = NULL;
+	sort_order = NULL; /* equivalent to sort_order = "comm,dso,sym" */
+
+	setup_sorting();
+
+	/*
+	 * expected output:
+	 *
+	 * Overhead  Command  Shared Object          Symbol
+	 * ========  =======  =============  ==============
+	 *   20.00%     perf  perf           [.] main
+	 *   10.00%     bash  [kernel]       [k] page_fault
+	 *   10.00%     bash  bash           [.] main
+	 *   10.00%     bash  bash           [.] xmalloc
+	 *   10.00%     perf  [kernel]       [k] page_fault
+	 *   10.00%     perf  [kernel]       [k] schedule
+	 *   10.00%     perf  libc           [.] free
+	 *   10.00%     perf  libc           [.] malloc
+	 *   10.00%     perf  perf           [.] cmd_record
+	 */
+	err = add_hist_entries(hists, machine);
+	if (err < 0)
+		goto out;
+
+	hists__collapse_resort(hists, NULL);
+	hists__output_resort(hists, NULL);
+
+	if (verbose > 2) {
+		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
+		print_hists_out(hists);
+	}
+
+	root = &hists->entries;
+	node = rb_first(root);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "perf") &&
+			!strcmp(SYM(he), "main") && he->stat.period == 200);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "bash") && !strcmp(DSO(he), "[kernel]") &&
+			!strcmp(SYM(he), "page_fault") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "bash") && !strcmp(DSO(he), "bash") &&
+			!strcmp(SYM(he), "main") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "bash") && !strcmp(DSO(he), "bash") &&
+			!strcmp(SYM(he), "xmalloc") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "[kernel]") &&
+			!strcmp(SYM(he), "page_fault") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "[kernel]") &&
+			!strcmp(SYM(he), "schedule") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "libc") &&
+			!strcmp(SYM(he), "free") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "libc") &&
+			!strcmp(SYM(he), "malloc") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "perf") &&
+			!strcmp(SYM(he), "cmd_record") && he->stat.period == 100);
+
+out:
+	del_hist_entries(hists);
+	reset_output_field();
+	return err;
+}
+
+/* mixed fields and sort keys */
+static int test2(struct perf_evsel *evsel, struct machine *machine)
+{
+	int err;
+	struct hists *hists = evsel__hists(evsel);
+	struct hist_entry *he;
+	struct rb_root *root;
+	struct rb_node *node;
+
+	field_order = "overhead,cpu";
+	sort_order = "pid";
+
+	setup_sorting();
+
+	/*
+	 * expected output:
+	 *
+	 * Overhead  CPU  Command:  Pid
+	 * ========  ===  =============
+	 *   30.00%    1  perf   :  100
+	 *   10.00%    0  perf   :  100
+	 *   10.00%    2  perf   :  100
+	 *   20.00%    2  perf   :  200
+	 *   10.00%    0  bash   :  300
+	 *   10.00%    1  bash   :  300
+	 *   10.00%    3  bash   :  300
+	 */
+	err = add_hist_entries(hists, machine);
+	if (err < 0)
+		goto out;
+
+	hists__collapse_resort(hists, NULL);
+	hists__output_resort(hists, NULL);
+
+	if (verbose > 2) {
+		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
+		print_hists_out(hists);
+	}
+
+	root = &hists->entries;
+	node = rb_first(root);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 1 && PID(he) == 100 && he->stat.period == 300);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 0 && PID(he) == 100 && he->stat.period == 100);
+
+out:
+	del_hist_entries(hists);
+	reset_output_field();
+	return err;
+}
+
+/* fields only (no sort key) */
+static int test3(struct perf_evsel *evsel, struct machine *machine)
+{
+	int err;
+	struct hists *hists = evsel__hists(evsel);
+	struct hist_entry *he;
+	struct rb_root *root;
+	struct rb_node *node;
+
+	field_order = "comm,overhead,dso";
+	sort_order = NULL;
+
+	setup_sorting();
+
+	/*
+	 * expected output:
+	 *
+	 * Command  Overhead  Shared Object
+	 * =======  ========  =============
+	 *    bash    20.00%  bash
+	 *    bash    10.00%  [kernel]
+	 *    perf    30.00%  perf
+	 *    perf    20.00%  [kernel]
+	 *    perf    20.00%  libc
+	 */
+	err = add_hist_entries(hists, machine);
+	if (err < 0)
+		goto out;
+
+	hists__collapse_resort(hists, NULL);
+	hists__output_resort(hists, NULL);
+
+	if (verbose > 2) {
+		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
+		print_hists_out(hists);
+	}
+
+	root = &hists->entries;
+	node = rb_first(root);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "bash") && !strcmp(DSO(he), "bash") &&
+			he->stat.period == 200);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "bash") && !strcmp(DSO(he), "[kernel]") &&
+			he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "perf") &&
+			he->stat.period == 300);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "[kernel]") &&
+			he->stat.period == 200);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "libc") &&
+			he->stat.period == 200);
+
+out:
+	del_hist_entries(hists);
+	reset_output_field();
+	return err;
+}
+
+/* handle duplicate 'dso' field */
+static int test4(struct perf_evsel *evsel, struct machine *machine)
+{
+	int err;
+	struct hists *hists = evsel__hists(evsel);
+	struct hist_entry *he;
+	struct rb_root *root;
+	struct rb_node *node;
+
+	field_order = "dso,sym,comm,overhead,dso";
+	sort_order = "sym";
+
+	setup_sorting();
+
+	/*
+	 * expected output:
+	 *
+	 * Shared Object          Symbol  Command  Overhead
+	 * =============  ==============  =======  ========
+	 *          perf  [.] cmd_record     perf    10.00%
+	 *          libc  [.] free           perf    10.00%
+	 *          bash  [.] main           bash    10.00%
+	 *          perf  [.] main           perf    20.00%
+	 *          libc  [.] malloc         perf    10.00%
+	 *      [kernel]  [k] page_fault     bash    10.00%
+	 *      [kernel]  [k] page_fault     perf    10.00%
+	 *      [kernel]  [k] schedule       perf    10.00%
+	 *          bash  [.] xmalloc        bash    10.00%
+	 */
+	err = add_hist_entries(hists, machine);
+	if (err < 0)
+		goto out;
+
+	hists__collapse_resort(hists, NULL);
+	hists__output_resort(hists, NULL);
+
+	if (verbose > 2) {
+		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
+		print_hists_out(hists);
+	}
+
+	root = &hists->entries;
+	node = rb_first(root);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(DSO(he), "perf") && !strcmp(SYM(he), "cmd_record") &&
+			!strcmp(COMM(he), "perf") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(DSO(he), "libc") && !strcmp(SYM(he), "free") &&
+			!strcmp(COMM(he), "perf") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(DSO(he), "bash") && !strcmp(SYM(he), "main") &&
+			!strcmp(COMM(he), "bash") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(DSO(he), "perf") && !strcmp(SYM(he), "main") &&
+			!strcmp(COMM(he), "perf") && he->stat.period == 200);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(DSO(he), "libc") && !strcmp(SYM(he), "malloc") &&
+			!strcmp(COMM(he), "perf") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(DSO(he), "[kernel]") && !strcmp(SYM(he), "page_fault") &&
+			!strcmp(COMM(he), "bash") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(DSO(he), "[kernel]") && !strcmp(SYM(he), "page_fault") &&
+			!strcmp(COMM(he), "perf") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(DSO(he), "[kernel]") && !strcmp(SYM(he), "schedule") &&
+			!strcmp(COMM(he), "perf") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			!strcmp(DSO(he), "bash") && !strcmp(SYM(he), "xmalloc") &&
+			!strcmp(COMM(he), "bash") && he->stat.period == 100);
+
+out:
+	del_hist_entries(hists);
+	reset_output_field();
+	return err;
+}
+
+/* full sort keys w/o overhead field */
+static int test5(struct perf_evsel *evsel, struct machine *machine)
+{
+	int err;
+	struct hists *hists = evsel__hists(evsel);
+	struct hist_entry *he;
+	struct rb_root *root;
+	struct rb_node *node;
+
+	field_order = "cpu,pid,comm,dso,sym";
+	sort_order = "dso,pid";
+
+	setup_sorting();
+
+	/*
+	 * expected output:
+	 *
+	 * CPU  Command:  Pid  Command  Shared Object          Symbol
+	 * ===  =============  =======  =============  ==============
+	 *   0     perf:  100     perf       [kernel]  [k] schedule
+	 *   2     perf:  200     perf       [kernel]  [k] page_fault
+	 *   1     bash:  300     bash       [kernel]  [k] page_fault
+	 *   0     bash:  300     bash           bash  [.] xmalloc
+	 *   3     bash:  300     bash           bash  [.] main
+	 *   1     perf:  100     perf           libc  [.] malloc
+	 *   2     perf:  100     perf           libc  [.] free
+	 *   1     perf:  100     perf           perf  [.] cmd_record
+	 *   1     perf:  100     perf           perf  [.] main
+	 *   2     perf:  200     perf           perf  [.] main
+	 */
+	err = add_hist_entries(hists, machine);
+	if (err < 0)
+		goto out;
+
+	hists__collapse_resort(hists, NULL);
+	hists__output_resort(hists, NULL);
+
+	if (verbose > 2) {
+		pr_info("[fields = %s, sort = %s]\n", field_order, sort_order);
+		print_hists_out(hists);
+	}
+
+	root = &hists->entries;
+	node = rb_first(root);
+	he = rb_entry(node, struct hist_entry, rb_node);
+
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 0 && PID(he) == 100 &&
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "[kernel]") &&
+			!strcmp(SYM(he), "schedule") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 2 && PID(he) == 200 &&
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "[kernel]") &&
+			!strcmp(SYM(he), "page_fault") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 1 && PID(he) == 300 &&
+			!strcmp(COMM(he), "bash") && !strcmp(DSO(he), "[kernel]") &&
+			!strcmp(SYM(he), "page_fault") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 0 && PID(he) == 300 &&
+			!strcmp(COMM(he), "bash") && !strcmp(DSO(he), "bash") &&
+			!strcmp(SYM(he), "xmalloc") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 3 && PID(he) == 300 &&
+			!strcmp(COMM(he), "bash") && !strcmp(DSO(he), "bash") &&
+			!strcmp(SYM(he), "main") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 1 && PID(he) == 100 &&
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "libc") &&
+			!strcmp(SYM(he), "malloc") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 2 && PID(he) == 100 &&
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "libc") &&
+			!strcmp(SYM(he), "free") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 1 && PID(he) == 100 &&
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "perf") &&
+			!strcmp(SYM(he), "cmd_record") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 1 && PID(he) == 100 &&
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "perf") &&
+			!strcmp(SYM(he), "main") && he->stat.period == 100);
+
+	node = rb_next(node);
+	he = rb_entry(node, struct hist_entry, rb_node);
+	TEST_ASSERT_VAL("Invalid hist entry",
+			CPU(he) == 2 && PID(he) == 200 &&
+			!strcmp(COMM(he), "perf") && !strcmp(DSO(he), "perf") &&
+			!strcmp(SYM(he), "main") && he->stat.period == 100);
+
+out:
+	del_hist_entries(hists);
+	reset_output_field();
+	return err;
+}
+
+int test__hists_output(void)
+{
+	int err = TEST_FAIL;
+	struct machines machines;
+	struct machine *machine;
+	struct perf_evsel *evsel;
+	struct perf_evlist *evlist = perf_evlist__new();
+	size_t i;
+	test_fn_t testcases[] = {
+		test1,
+		test2,
+		test3,
+		test4,
+		test5,
+	};
+
+	TEST_ASSERT_VAL("No memory", evlist);
+
+	err = parse_events(evlist, "cpu-clock");
+	if (err)
+		goto out;
+
+	machines__init(&machines);
+
+	/* setup threads/dso/map/symbols also */
+	machine = setup_fake_machine(&machines);
+	if (!machine)
+		goto out;
+
+	if (verbose > 1)
+		machine__fprintf(machine, stderr);
+
+	evsel = perf_evlist__first(evlist);
+
+	for (i = 0; i < ARRAY_SIZE(testcases); i++) {
+		err = testcases[i](evsel, machine);
+		if (err < 0)
+			break;
+	}
+
+out:
+	/* tear down everything */
+	perf_evlist__delete(evlist);
+	machines__exit(&machines);
+
+	return err;
+}
diff --git a/tools/perf/tests/keep-tracking.c b/tools/perf/tests/keep-tracking.c
new file mode 100644
index 000000000..7a5ab7b0b
--- /dev/null
+++ b/tools/perf/tests/keep-tracking.c
@@ -0,0 +1,152 @@
+#include <linux/types.h>
+#include <unistd.h>
+#include <sys/prctl.h>
+
+#include "parse-events.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "thread_map.h"
+#include "cpumap.h"
+#include "tests.h"
+
+#define CHECK__(x) {				\
+	while ((x) < 0) {			\
+		pr_debug(#x " failed!\n");	\
+		goto out_err;			\
+	}					\
+}
+
+#define CHECK_NOT_NULL__(x) {			\
+	while ((x) == NULL) {			\
+		pr_debug(#x " failed!\n");	\
+		goto out_err;			\
+	}					\
+}
+
+static int find_comm(struct perf_evlist *evlist, const char *comm)
+{
+	union perf_event *event;
+	int i, found;
+
+	found = 0;
+	for (i = 0; i < evlist->nr_mmaps; i++) {
+		while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
+			if (event->header.type == PERF_RECORD_COMM &&
+			    (pid_t)event->comm.pid == getpid() &&
+			    (pid_t)event->comm.tid == getpid() &&
+			    strcmp(event->comm.comm, comm) == 0)
+				found += 1;
+			perf_evlist__mmap_consume(evlist, i);
+		}
+	}
+	return found;
+}
+
+/**
+ * test__keep_tracking - test using a dummy software event to keep tracking.
+ *
+ * This function implements a test that checks that tracking events continue
+ * when an event is disabled but a dummy software event is not disabled.  If the
+ * test passes %0 is returned, otherwise %-1 is returned.
+ */
+int test__keep_tracking(void)
+{
+	struct record_opts opts = {
+		.mmap_pages	     = UINT_MAX,
+		.user_freq	     = UINT_MAX,
+		.user_interval	     = ULLONG_MAX,
+		.freq		     = 4000,
+		.target		     = {
+			.uses_mmap   = true,
+		},
+	};
+	struct thread_map *threads = NULL;
+	struct cpu_map *cpus = NULL;
+	struct perf_evlist *evlist = NULL;
+	struct perf_evsel *evsel = NULL;
+	int found, err = -1;
+	const char *comm;
+
+	threads = thread_map__new(-1, getpid(), UINT_MAX);
+	CHECK_NOT_NULL__(threads);
+
+	cpus = cpu_map__new(NULL);
+	CHECK_NOT_NULL__(cpus);
+
+	evlist = perf_evlist__new();
+	CHECK_NOT_NULL__(evlist);
+
+	perf_evlist__set_maps(evlist, cpus, threads);
+
+	CHECK__(parse_events(evlist, "dummy:u"));
+	CHECK__(parse_events(evlist, "cycles:u"));
+
+	perf_evlist__config(evlist, &opts);
+
+	evsel = perf_evlist__first(evlist);
+
+	evsel->attr.comm = 1;
+	evsel->attr.disabled = 1;
+	evsel->attr.enable_on_exec = 0;
+
+	if (perf_evlist__open(evlist) < 0) {
+		fprintf(stderr, " (not supported)");
+		err = 0;
+		goto out_err;
+	}
+
+	CHECK__(perf_evlist__mmap(evlist, UINT_MAX, false));
+
+	/*
+	 * First, test that a 'comm' event can be found when the event is
+	 * enabled.
+	 */
+
+	perf_evlist__enable(evlist);
+
+	comm = "Test COMM 1";
+	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0));
+
+	perf_evlist__disable(evlist);
+
+	found = find_comm(evlist, comm);
+	if (found != 1) {
+		pr_debug("First time, failed to find tracking event.\n");
+		goto out_err;
+	}
+
+	/*
+	 * Secondly, test that a 'comm' event can be found when the event is
+	 * disabled with the dummy event still enabled.
+	 */
+
+	perf_evlist__enable(evlist);
+
+	evsel = perf_evlist__last(evlist);
+
+	CHECK__(perf_evlist__disable_event(evlist, evsel));
+
+	comm = "Test COMM 2";
+	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0));
+
+	perf_evlist__disable(evlist);
+
+	found = find_comm(evlist, comm);
+	if (found != 1) {
+		pr_debug("Seconf time, failed to find tracking event.\n");
+		goto out_err;
+	}
+
+	err = 0;
+
+out_err:
+	if (evlist) {
+		perf_evlist__disable(evlist);
+		perf_evlist__delete(evlist);
+	} else {
+		cpu_map__delete(cpus);
+		thread_map__delete(threads);
+	}
+
+	return err;
+}
diff --git a/tools/perf/tests/kmod-path.c b/tools/perf/tests/kmod-path.c
new file mode 100644
index 000000000..e8d7cbb93
--- /dev/null
+++ b/tools/perf/tests/kmod-path.c
@@ -0,0 +1,73 @@
+#include <stdbool.h>
+#include "tests.h"
+#include "dso.h"
+#include "debug.h"
+
+static int test(const char *path, bool alloc_name, bool alloc_ext,
+		bool kmod, bool comp, const char *name, const char *ext)
+{
+	struct kmod_path m;
+
+	memset(&m, 0x0, sizeof(m));
+
+	TEST_ASSERT_VAL("kmod_path__parse",
+			!__kmod_path__parse(&m, path, alloc_name, alloc_ext));
+
+	pr_debug("%s - alloc name %d, alloc ext %d, kmod %d, comp %d, name '%s', ext '%s'\n",
+		 path, alloc_name, alloc_ext, m.kmod, m.comp, m.name, m.ext);
+
+	TEST_ASSERT_VAL("wrong kmod", m.kmod == kmod);
+	TEST_ASSERT_VAL("wrong comp", m.comp == comp);
+
+	if (ext)
+		TEST_ASSERT_VAL("wrong ext", m.ext && !strcmp(ext, m.ext));
+	else
+		TEST_ASSERT_VAL("wrong ext", !m.ext);
+
+	if (name)
+		TEST_ASSERT_VAL("wrong name", m.name && !strcmp(name, m.name));
+	else
+		TEST_ASSERT_VAL("wrong name", !m.name);
+
+	free(m.name);
+	free(m.ext);
+	return 0;
+}
+
+#define T(path, an, ae, k, c, n, e) \
+	TEST_ASSERT_VAL("failed", !test(path, an, ae, k, c, n, e))
+
+int test__kmod_path__parse(void)
+{
+	/* path                alloc_name  alloc_ext   kmod  comp   name     ext */
+	T("/xxxx/xxxx/x-x.ko", true      , true      , true, false, "[x_x]", NULL);
+	T("/xxxx/xxxx/x-x.ko", false     , true      , true, false, NULL   , NULL);
+	T("/xxxx/xxxx/x-x.ko", true      , false     , true, false, "[x_x]", NULL);
+	T("/xxxx/xxxx/x-x.ko", false     , false     , true, false, NULL   , NULL);
+
+	/* path                alloc_name  alloc_ext   kmod  comp  name   ext */
+	T("/xxxx/xxxx/x.ko.gz", true     , true      , true, true, "[x]", "gz");
+	T("/xxxx/xxxx/x.ko.gz", false    , true      , true, true, NULL , "gz");
+	T("/xxxx/xxxx/x.ko.gz", true     , false     , true, true, "[x]", NULL);
+	T("/xxxx/xxxx/x.ko.gz", false    , false     , true, true, NULL , NULL);
+
+	/* path              alloc_name  alloc_ext  kmod   comp  name    ext */
+	T("/xxxx/xxxx/x.gz", true      , true     , false, true, "x.gz" ,"gz");
+	T("/xxxx/xxxx/x.gz", false     , true     , false, true, NULL   ,"gz");
+	T("/xxxx/xxxx/x.gz", true      , false    , false, true, "x.gz" , NULL);
+	T("/xxxx/xxxx/x.gz", false     , false    , false, true, NULL   , NULL);
+
+	/* path   alloc_name  alloc_ext  kmod   comp  name     ext */
+	T("x.gz", true      , true     , false, true, "x.gz", "gz");
+	T("x.gz", false     , true     , false, true, NULL  , "gz");
+	T("x.gz", true      , false    , false, true, "x.gz", NULL);
+	T("x.gz", false     , false    , false, true, NULL  , NULL);
+
+	/* path      alloc_name  alloc_ext  kmod  comp  name  ext */
+	T("x.ko.gz", true      , true     , true, true, "[x]", "gz");
+	T("x.ko.gz", false     , true     , true, true, NULL , "gz");
+	T("x.ko.gz", true      , false    , true, true, "[x]", NULL);
+	T("x.ko.gz", false     , false    , true, true, NULL , NULL);
+
+	return 0;
+}
diff --git a/tools/perf/tests/make b/tools/perf/tests/make
new file mode 100644
index 000000000..bff85324f
--- /dev/null
+++ b/tools/perf/tests/make
@@ -0,0 +1,232 @@
+PERF := .
+MK   := Makefile
+
+include config/Makefile.arch
+
+# FIXME looks like x86 is the only arch running tests ;-)
+# we need some IS_(32/64) flag to make this generic
+ifeq ($(ARCH)$(IS_64_BIT), x861)
+lib = lib64
+else
+lib = lib
+endif
+
+has = $(shell which $1 2>/dev/null)
+
+# standard single make variable specified
+make_clean_all      := clean all
+make_python_perf_so := python/perf.so
+make_debug          := DEBUG=1
+make_no_libperl     := NO_LIBPERL=1
+make_no_libpython   := NO_LIBPYTHON=1
+make_no_scripts     := NO_LIBPYTHON=1 NO_LIBPERL=1
+make_no_newt        := NO_NEWT=1
+make_no_slang       := NO_SLANG=1
+make_no_gtk2        := NO_GTK2=1
+make_no_ui          := NO_NEWT=1 NO_SLANG=1 NO_GTK2=1
+make_no_demangle    := NO_DEMANGLE=1
+make_no_libelf      := NO_LIBELF=1
+make_no_libunwind   := NO_LIBUNWIND=1
+make_no_libdw_dwarf_unwind := NO_LIBDW_DWARF_UNWIND=1
+make_no_backtrace   := NO_BACKTRACE=1
+make_no_libnuma     := NO_LIBNUMA=1
+make_no_libaudit    := NO_LIBAUDIT=1
+make_no_libbionic   := NO_LIBBIONIC=1
+make_tags           := tags
+make_cscope         := cscope
+make_help           := help
+make_doc            := doc
+make_perf_o           := perf.o
+make_util_map_o       := util/map.o
+make_util_pmu_bison_o := util/pmu-bison.o
+make_install        := install
+make_install_bin    := install-bin
+make_install_doc    := install-doc
+make_install_man    := install-man
+make_install_html   := install-html
+make_install_info   := install-info
+make_install_pdf    := install-pdf
+make_static         := LDFLAGS=-static
+
+# all the NO_* variable combined
+make_minimal        := NO_LIBPERL=1 NO_LIBPYTHON=1 NO_NEWT=1 NO_GTK2=1
+make_minimal        += NO_DEMANGLE=1 NO_LIBELF=1 NO_LIBUNWIND=1 NO_BACKTRACE=1
+make_minimal        += NO_LIBNUMA=1 NO_LIBAUDIT=1 NO_LIBBIONIC=1
+make_minimal        += NO_LIBDW_DWARF_UNWIND=1
+
+# $(run) contains all available tests
+run := make_pure
+run += make_clean_all
+run += make_python_perf_so
+run += make_debug
+run += make_no_libperl
+run += make_no_libpython
+run += make_no_scripts
+run += make_no_newt
+run += make_no_slang
+run += make_no_gtk2
+run += make_no_ui
+run += make_no_demangle
+run += make_no_libelf
+run += make_no_libunwind
+run += make_no_libdw_dwarf_unwind
+run += make_no_backtrace
+run += make_no_libnuma
+run += make_no_libaudit
+run += make_no_libbionic
+run += make_help
+run += make_doc
+run += make_perf_o
+run += make_util_map_o
+run += make_util_pmu_bison_o
+run += make_install
+run += make_install_bin
+# FIXME 'install-*' commented out till they're fixed
+# run += make_install_doc
+# run += make_install_man
+# run += make_install_html
+# run += make_install_info
+# run += make_install_pdf
+run += make_minimal
+run += make_static
+
+ifneq ($(call has,ctags),)
+run += make_tags
+endif
+ifneq ($(call has,cscope),)
+run += make_cscope
+endif
+
+# $(run_O) contains same portion of $(run) tests with '_O' attached
+# to distinguish O=... tests
+run_O := $(addsuffix _O,$(run))
+
+# disable some tests for O=...
+run_O := $(filter-out make_python_perf_so_O,$(run_O))
+
+# define test for each compile as 'test_NAME' variable
+# with the test itself as a value
+test_make_tags   = test -f tags
+test_make_cscope = test -f cscope.out
+
+test_make_tags_O   := $(test_make_tags)
+test_make_cscope_O := $(test_make_cscope)
+
+test_ok          := true
+test_make_help   := $(test_ok)
+test_make_doc    := $(test_ok)
+test_make_help_O := $(test_ok)
+test_make_doc_O  := $(test_ok)
+
+test_make_python_perf_so := test -f $(PERF)/python/perf.so
+
+test_make_perf_o           := test -f $(PERF)/perf.o
+test_make_util_map_o       := test -f $(PERF)/util/map.o
+test_make_util_pmu_bison_o := test -f $(PERF)/util/pmu-bison.o
+
+define test_dest_files
+  for file in $(1); do				\
+    if [ ! -x $$TMP_DEST/$$file ]; then		\
+      echo "  failed to find: $$file";		\
+    fi						\
+  done
+endef
+
+installed_files_bin := bin/perf
+installed_files_bin += etc/bash_completion.d/perf
+installed_files_bin += libexec/perf-core/perf-archive
+
+installed_files_plugins := $(lib)/traceevent/plugins/plugin_cfg80211.so
+installed_files_plugins += $(lib)/traceevent/plugins/plugin_scsi.so
+installed_files_plugins += $(lib)/traceevent/plugins/plugin_xen.so
+installed_files_plugins += $(lib)/traceevent/plugins/plugin_function.so
+installed_files_plugins += $(lib)/traceevent/plugins/plugin_sched_switch.so
+installed_files_plugins += $(lib)/traceevent/plugins/plugin_mac80211.so
+installed_files_plugins += $(lib)/traceevent/plugins/plugin_kvm.so
+installed_files_plugins += $(lib)/traceevent/plugins/plugin_kmem.so
+installed_files_plugins += $(lib)/traceevent/plugins/plugin_hrtimer.so
+installed_files_plugins += $(lib)/traceevent/plugins/plugin_jbd2.so
+
+installed_files_all := $(installed_files_bin)
+installed_files_all += $(installed_files_plugins)
+
+test_make_install       := $(call test_dest_files,$(installed_files_all))
+test_make_install_O     := $(call test_dest_files,$(installed_files_all))
+test_make_install_bin   := $(call test_dest_files,$(installed_files_bin))
+test_make_install_bin_O := $(call test_dest_files,$(installed_files_bin))
+
+# FIXME nothing gets installed
+test_make_install_man    := test -f $$TMP_DEST/share/man/man1/perf.1
+test_make_install_man_O  := $(test_make_install_man)
+
+# FIXME nothing gets installed
+test_make_install_doc    := $(test_ok)
+test_make_install_doc_O  := $(test_ok)
+
+# FIXME nothing gets installed
+test_make_install_html   := $(test_ok)
+test_make_install_html_O := $(test_ok)
+
+# FIXME nothing gets installed
+test_make_install_info   := $(test_ok)
+test_make_install_info_O := $(test_ok)
+
+# FIXME nothing gets installed
+test_make_install_pdf    := $(test_ok)
+test_make_install_pdf_O  := $(test_ok)
+
+test_make_python_perf_so_O    := test -f $$TMP_O/python/perf.so
+test_make_perf_o_O            := test -f $$TMP_O/perf.o
+test_make_util_map_o_O        := test -f $$TMP_O/util/map.o
+test_make_util_pmu_bison_o_O := test -f $$TMP_O/util/pmu-bison.o
+
+test_default = test -x $(PERF)/perf
+test = $(if $(test_$1),$(test_$1),$(test_default))
+
+test_default_O = test -x $$TMP_O/perf
+test_O = $(if $(test_$1),$(test_$1),$(test_default_O))
+
+all:
+
+ifdef DEBUG
+d := $(info run   $(run))
+d := $(info run_O $(run_O))
+endif
+
+MAKEFLAGS := --no-print-directory
+
+clean := @(cd $(PERF); make -s -f $(MK) clean >/dev/null)
+
+$(run):
+	$(call clean)
+	@TMP_DEST=$$(mktemp -d); \
+	cmd="cd $(PERF) && make -f $(MK) DESTDIR=$$TMP_DEST $($@)"; \
+	echo "- $@: $$cmd" && echo $$cmd > $@ && \
+	( eval $$cmd ) >> $@ 2>&1; \
+	echo "  test: $(call test,$@)" >> $@ 2>&1; \
+	$(call test,$@) && \
+	rm -rf $@ $$TMP_DEST || (cat $@ ; false)
+
+$(run_O):
+	$(call clean)
+	@TMP_O=$$(mktemp -d); \
+	TMP_DEST=$$(mktemp -d); \
+	cmd="cd $(PERF) && make -f $(MK) O=$$TMP_O DESTDIR=$$TMP_DEST $($(patsubst %_O,%,$@))"; \
+	echo "- $@: $$cmd" && echo $$cmd > $@ && \
+	( eval $$cmd ) >> $@ 2>&1 && \
+	echo "  test: $(call test_O,$@)" >> $@ 2>&1; \
+	$(call test_O,$@) && \
+	rm -rf $@ $$TMP_O $$TMP_DEST || (cat $@ ; false)
+
+tarpkg:
+	@cmd="$(PERF)/tests/perf-targz-src-pkg $(PERF)"; \
+	echo "- $@: $$cmd" && echo $$cmd > $@ && \
+	( eval $$cmd ) >> $@ 2>&1
+
+all: $(run) $(run_O) tarpkg
+	@echo OK
+
+out: $(run_O)
+	@echo OK
+
+.PHONY: all $(run) $(run_O) tarpkg clean
diff --git a/tools/perf/tests/mmap-basic.c b/tools/perf/tests/mmap-basic.c
new file mode 100644
index 000000000..9b9622a33
--- /dev/null
+++ b/tools/perf/tests/mmap-basic.c
@@ -0,0 +1,149 @@
+#include "evlist.h"
+#include "evsel.h"
+#include "thread_map.h"
+#include "cpumap.h"
+#include "tests.h"
+
+/*
+ * This test will generate random numbers of calls to some getpid syscalls,
+ * then establish an mmap for a group of events that are created to monitor
+ * the syscalls.
+ *
+ * It will receive the events, using mmap, use its PERF_SAMPLE_ID generated
+ * sample.id field to map back to its respective perf_evsel instance.
+ *
+ * Then it checks if the number of syscalls reported as perf events by
+ * the kernel corresponds to the number of syscalls made.
+ */
+int test__basic_mmap(void)
+{
+	int err = -1;
+	union perf_event *event;
+	struct thread_map *threads;
+	struct cpu_map *cpus;
+	struct perf_evlist *evlist;
+	cpu_set_t cpu_set;
+	const char *syscall_names[] = { "getsid", "getppid", "getpgrp",
+					"getpgid", };
+	pid_t (*syscalls[])(void) = { (void *)getsid, getppid, getpgrp,
+				      (void*)getpgid };
+#define nsyscalls ARRAY_SIZE(syscall_names)
+	unsigned int nr_events[nsyscalls],
+		     expected_nr_events[nsyscalls], i, j;
+	struct perf_evsel *evsels[nsyscalls], *evsel;
+	char sbuf[STRERR_BUFSIZE];
+
+	threads = thread_map__new(-1, getpid(), UINT_MAX);
+	if (threads == NULL) {
+		pr_debug("thread_map__new\n");
+		return -1;
+	}
+
+	cpus = cpu_map__new(NULL);
+	if (cpus == NULL) {
+		pr_debug("cpu_map__new\n");
+		goto out_free_threads;
+	}
+
+	CPU_ZERO(&cpu_set);
+	CPU_SET(cpus->map[0], &cpu_set);
+	sched_setaffinity(0, sizeof(cpu_set), &cpu_set);
+	if (sched_setaffinity(0, sizeof(cpu_set), &cpu_set) < 0) {
+		pr_debug("sched_setaffinity() failed on CPU %d: %s ",
+			 cpus->map[0], strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_free_cpus;
+	}
+
+	evlist = perf_evlist__new();
+	if (evlist == NULL) {
+		pr_debug("perf_evlist__new\n");
+		goto out_free_cpus;
+	}
+
+	perf_evlist__set_maps(evlist, cpus, threads);
+
+	for (i = 0; i < nsyscalls; ++i) {
+		char name[64];
+
+		snprintf(name, sizeof(name), "sys_enter_%s", syscall_names[i]);
+		evsels[i] = perf_evsel__newtp("syscalls", name);
+		if (evsels[i] == NULL) {
+			pr_debug("perf_evsel__new\n");
+			goto out_delete_evlist;
+		}
+
+		evsels[i]->attr.wakeup_events = 1;
+		perf_evsel__set_sample_id(evsels[i], false);
+
+		perf_evlist__add(evlist, evsels[i]);
+
+		if (perf_evsel__open(evsels[i], cpus, threads) < 0) {
+			pr_debug("failed to open counter: %s, "
+				 "tweak /proc/sys/kernel/perf_event_paranoid?\n",
+				 strerror_r(errno, sbuf, sizeof(sbuf)));
+			goto out_delete_evlist;
+		}
+
+		nr_events[i] = 0;
+		expected_nr_events[i] = 1 + rand() % 127;
+	}
+
+	if (perf_evlist__mmap(evlist, 128, true) < 0) {
+		pr_debug("failed to mmap events: %d (%s)\n", errno,
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	for (i = 0; i < nsyscalls; ++i)
+		for (j = 0; j < expected_nr_events[i]; ++j) {
+			int foo = syscalls[i]();
+			++foo;
+		}
+
+	while ((event = perf_evlist__mmap_read(evlist, 0)) != NULL) {
+		struct perf_sample sample;
+
+		if (event->header.type != PERF_RECORD_SAMPLE) {
+			pr_debug("unexpected %s event\n",
+				 perf_event__name(event->header.type));
+			goto out_delete_evlist;
+		}
+
+		err = perf_evlist__parse_sample(evlist, event, &sample);
+		if (err) {
+			pr_err("Can't parse sample, err = %d\n", err);
+			goto out_delete_evlist;
+		}
+
+		err = -1;
+		evsel = perf_evlist__id2evsel(evlist, sample.id);
+		if (evsel == NULL) {
+			pr_debug("event with id %" PRIu64
+				 " doesn't map to an evsel\n", sample.id);
+			goto out_delete_evlist;
+		}
+		nr_events[evsel->idx]++;
+		perf_evlist__mmap_consume(evlist, 0);
+	}
+
+	err = 0;
+	evlist__for_each(evlist, evsel) {
+		if (nr_events[evsel->idx] != expected_nr_events[evsel->idx]) {
+			pr_debug("expected %d %s events, got %d\n",
+				 expected_nr_events[evsel->idx],
+				 perf_evsel__name(evsel), nr_events[evsel->idx]);
+			err = -1;
+			goto out_delete_evlist;
+		}
+	}
+
+out_delete_evlist:
+	perf_evlist__delete(evlist);
+	cpus	= NULL;
+	threads = NULL;
+out_free_cpus:
+	cpu_map__delete(cpus);
+out_free_threads:
+	thread_map__delete(threads);
+	return err;
+}
diff --git a/tools/perf/tests/mmap-thread-lookup.c b/tools/perf/tests/mmap-thread-lookup.c
new file mode 100644
index 000000000..2113f1c86
--- /dev/null
+++ b/tools/perf/tests/mmap-thread-lookup.c
@@ -0,0 +1,233 @@
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <sys/types.h>
+#include <sys/mman.h>
+#include <pthread.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include "debug.h"
+#include "tests.h"
+#include "machine.h"
+#include "thread_map.h"
+#include "symbol.h"
+#include "thread.h"
+
+#define THREADS 4
+
+static int go_away;
+
+struct thread_data {
+	pthread_t	pt;
+	pid_t		tid;
+	void		*map;
+	int		ready[2];
+};
+
+static struct thread_data threads[THREADS];
+
+static int thread_init(struct thread_data *td)
+{
+	void *map;
+
+	map = mmap(NULL, page_size,
+		   PROT_READ|PROT_WRITE|PROT_EXEC,
+		   MAP_SHARED|MAP_ANONYMOUS, -1, 0);
+
+	if (map == MAP_FAILED) {
+		perror("mmap failed");
+		return -1;
+	}
+
+	td->map = map;
+	td->tid = syscall(SYS_gettid);
+
+	pr_debug("tid = %d, map = %p\n", td->tid, map);
+	return 0;
+}
+
+static void *thread_fn(void *arg)
+{
+	struct thread_data *td = arg;
+	ssize_t ret;
+	int go;
+
+	if (thread_init(td))
+		return NULL;
+
+	/* Signal thread_create thread is initialized. */
+	ret = write(td->ready[1], &go, sizeof(int));
+	if (ret != sizeof(int)) {
+		pr_err("failed to notify\n");
+		return NULL;
+	}
+
+	while (!go_away) {
+		/* Waiting for main thread to kill us. */
+		usleep(100);
+	}
+
+	munmap(td->map, page_size);
+	return NULL;
+}
+
+static int thread_create(int i)
+{
+	struct thread_data *td = &threads[i];
+	int err, go;
+
+	if (pipe(td->ready))
+		return -1;
+
+	err = pthread_create(&td->pt, NULL, thread_fn, td);
+	if (!err) {
+		/* Wait for thread initialization. */
+		ssize_t ret = read(td->ready[0], &go, sizeof(int));
+		err = ret != sizeof(int);
+	}
+
+	close(td->ready[0]);
+	close(td->ready[1]);
+	return err;
+}
+
+static int threads_create(void)
+{
+	struct thread_data *td0 = &threads[0];
+	int i, err = 0;
+
+	go_away = 0;
+
+	/* 0 is main thread */
+	if (thread_init(td0))
+		return -1;
+
+	for (i = 1; !err && i < THREADS; i++)
+		err = thread_create(i);
+
+	return err;
+}
+
+static int threads_destroy(void)
+{
+	struct thread_data *td0 = &threads[0];
+	int i, err = 0;
+
+	/* cleanup the main thread */
+	munmap(td0->map, page_size);
+
+	go_away = 1;
+
+	for (i = 1; !err && i < THREADS; i++)
+		err = pthread_join(threads[i].pt, NULL);
+
+	return err;
+}
+
+typedef int (*synth_cb)(struct machine *machine);
+
+static int synth_all(struct machine *machine)
+{
+	return perf_event__synthesize_threads(NULL,
+					      perf_event__process,
+					      machine, 0);
+}
+
+static int synth_process(struct machine *machine)
+{
+	struct thread_map *map;
+	int err;
+
+	map = thread_map__new_by_pid(getpid());
+
+	err = perf_event__synthesize_thread_map(NULL, map,
+						perf_event__process,
+						machine, 0);
+
+	thread_map__delete(map);
+	return err;
+}
+
+static int mmap_events(synth_cb synth)
+{
+	struct machines machines;
+	struct machine *machine;
+	int err, i;
+
+	/*
+	 * The threads_create will not return before all threads
+	 * are spawned and all created memory map.
+	 *
+	 * They will loop until threads_destroy is called, so we
+	 * can safely run synthesizing function.
+	 */
+	TEST_ASSERT_VAL("failed to create threads", !threads_create());
+
+	machines__init(&machines);
+	machine = &machines.host;
+
+	dump_trace = verbose > 1 ? 1 : 0;
+
+	err = synth(machine);
+
+	dump_trace = 0;
+
+	TEST_ASSERT_VAL("failed to destroy threads", !threads_destroy());
+	TEST_ASSERT_VAL("failed to synthesize maps", !err);
+
+	/*
+	 * All data is synthesized, try to find map for each
+	 * thread object.
+	 */
+	for (i = 0; i < THREADS; i++) {
+		struct thread_data *td = &threads[i];
+		struct addr_location al;
+		struct thread *thread;
+
+		thread = machine__findnew_thread(machine, getpid(), td->tid);
+
+		pr_debug("looking for map %p\n", td->map);
+
+		thread__find_addr_map(thread,
+				      PERF_RECORD_MISC_USER, MAP__FUNCTION,
+				      (unsigned long) (td->map + 1), &al);
+
+		if (!al.map) {
+			pr_debug("failed, couldn't find map\n");
+			err = -1;
+			break;
+		}
+
+		pr_debug("map %p, addr %" PRIx64 "\n", al.map, al.map->start);
+	}
+
+	machine__delete_threads(machine);
+	machines__exit(&machines);
+	return err;
+}
+
+/*
+ * This test creates 'THREADS' number of threads (including
+ * main thread) and each thread creates memory map.
+ *
+ * When threads are created, we synthesize them with both
+ * (separate tests):
+ *   perf_event__synthesize_thread_map (process based)
+ *   perf_event__synthesize_threads    (global)
+ *
+ * We test we can find all memory maps via:
+ *   thread__find_addr_map
+ *
+ * by using all thread objects.
+ */
+int test__mmap_thread_lookup(void)
+{
+	/* perf_event__synthesize_threads synthesize */
+	TEST_ASSERT_VAL("failed with sythesizing all",
+			!mmap_events(synth_all));
+
+	/* perf_event__synthesize_thread_map synthesize */
+	TEST_ASSERT_VAL("failed with sythesizing process",
+			!mmap_events(synth_process));
+
+	return 0;
+}
diff --git a/tools/perf/tests/open-syscall-all-cpus.c b/tools/perf/tests/open-syscall-all-cpus.c
new file mode 100644
index 000000000..3ec885c48
--- /dev/null
+++ b/tools/perf/tests/open-syscall-all-cpus.c
@@ -0,0 +1,115 @@
+#include "evsel.h"
+#include "tests.h"
+#include "thread_map.h"
+#include "cpumap.h"
+#include "debug.h"
+
+int test__open_syscall_event_on_all_cpus(void)
+{
+	int err = -1, fd, cpu;
+	struct cpu_map *cpus;
+	struct perf_evsel *evsel;
+	unsigned int nr_open_calls = 111, i;
+	cpu_set_t cpu_set;
+	struct thread_map *threads = thread_map__new(-1, getpid(), UINT_MAX);
+	char sbuf[STRERR_BUFSIZE];
+
+	if (threads == NULL) {
+		pr_debug("thread_map__new\n");
+		return -1;
+	}
+
+	cpus = cpu_map__new(NULL);
+	if (cpus == NULL) {
+		pr_debug("cpu_map__new\n");
+		goto out_thread_map_delete;
+	}
+
+	CPU_ZERO(&cpu_set);
+
+	evsel = perf_evsel__newtp("syscalls", "sys_enter_open");
+	if (evsel == NULL) {
+		if (tracefs_configured())
+			pr_debug("is tracefs mounted on /sys/kernel/tracing?\n");
+		else if (debugfs_configured())
+			pr_debug("is debugfs mounted on /sys/kernel/debug?\n");
+		else
+			pr_debug("Neither tracefs or debugfs is enabled in this kernel\n");
+		goto out_thread_map_delete;
+	}
+
+	if (perf_evsel__open(evsel, cpus, threads) < 0) {
+		pr_debug("failed to open counter: %s, "
+			 "tweak /proc/sys/kernel/perf_event_paranoid?\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_evsel_delete;
+	}
+
+	for (cpu = 0; cpu < cpus->nr; ++cpu) {
+		unsigned int ncalls = nr_open_calls + cpu;
+		/*
+		 * XXX eventually lift this restriction in a way that
+		 * keeps perf building on older glibc installations
+		 * without CPU_ALLOC. 1024 cpus in 2010 still seems
+		 * a reasonable upper limit tho :-)
+		 */
+		if (cpus->map[cpu] >= CPU_SETSIZE) {
+			pr_debug("Ignoring CPU %d\n", cpus->map[cpu]);
+			continue;
+		}
+
+		CPU_SET(cpus->map[cpu], &cpu_set);
+		if (sched_setaffinity(0, sizeof(cpu_set), &cpu_set) < 0) {
+			pr_debug("sched_setaffinity() failed on CPU %d: %s ",
+				 cpus->map[cpu],
+				 strerror_r(errno, sbuf, sizeof(sbuf)));
+			goto out_close_fd;
+		}
+		for (i = 0; i < ncalls; ++i) {
+			fd = open("/etc/passwd", O_RDONLY);
+			close(fd);
+		}
+		CPU_CLR(cpus->map[cpu], &cpu_set);
+	}
+
+	/*
+	 * Here we need to explicitely preallocate the counts, as if
+	 * we use the auto allocation it will allocate just for 1 cpu,
+	 * as we start by cpu 0.
+	 */
+	if (perf_evsel__alloc_counts(evsel, cpus->nr) < 0) {
+		pr_debug("perf_evsel__alloc_counts(ncpus=%d)\n", cpus->nr);
+		goto out_close_fd;
+	}
+
+	err = 0;
+
+	for (cpu = 0; cpu < cpus->nr; ++cpu) {
+		unsigned int expected;
+
+		if (cpus->map[cpu] >= CPU_SETSIZE)
+			continue;
+
+		if (perf_evsel__read_on_cpu(evsel, cpu, 0) < 0) {
+			pr_debug("perf_evsel__read_on_cpu\n");
+			err = -1;
+			break;
+		}
+
+		expected = nr_open_calls + cpu;
+		if (evsel->counts->cpu[cpu].val != expected) {
+			pr_debug("perf_evsel__read_on_cpu: expected to intercept %d calls on cpu %d, got %" PRIu64 "\n",
+				 expected, cpus->map[cpu], evsel->counts->cpu[cpu].val);
+			err = -1;
+		}
+	}
+
+	perf_evsel__free_counts(evsel);
+out_close_fd:
+	perf_evsel__close_fd(evsel, 1, threads->nr);
+out_evsel_delete:
+	perf_evsel__delete(evsel);
+out_thread_map_delete:
+	thread_map__delete(threads);
+	return err;
+}
diff --git a/tools/perf/tests/open-syscall-tp-fields.c b/tools/perf/tests/open-syscall-tp-fields.c
new file mode 100644
index 000000000..127dcae0b
--- /dev/null
+++ b/tools/perf/tests/open-syscall-tp-fields.c
@@ -0,0 +1,121 @@
+#include "perf.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "thread_map.h"
+#include "tests.h"
+#include "debug.h"
+
+int test__syscall_open_tp_fields(void)
+{
+	struct record_opts opts = {
+		.target = {
+			.uid = UINT_MAX,
+			.uses_mmap = true,
+		},
+		.no_buffering = true,
+		.freq	      = 1,
+		.mmap_pages   = 256,
+		.raw_samples  = true,
+	};
+	const char *filename = "/etc/passwd";
+	int flags = O_RDONLY | O_DIRECTORY;
+	struct perf_evlist *evlist = perf_evlist__new();
+	struct perf_evsel *evsel;
+	int err = -1, i, nr_events = 0, nr_polls = 0;
+	char sbuf[STRERR_BUFSIZE];
+
+	if (evlist == NULL) {
+		pr_debug("%s: perf_evlist__new\n", __func__);
+		goto out;
+	}
+
+	evsel = perf_evsel__newtp("syscalls", "sys_enter_open");
+	if (evsel == NULL) {
+		pr_debug("%s: perf_evsel__newtp\n", __func__);
+		goto out_delete_evlist;
+	}
+
+	perf_evlist__add(evlist, evsel);
+
+	err = perf_evlist__create_maps(evlist, &opts.target);
+	if (err < 0) {
+		pr_debug("%s: perf_evlist__create_maps\n", __func__);
+		goto out_delete_evlist;
+	}
+
+	perf_evsel__config(evsel, &opts);
+
+	evlist->threads->map[0] = getpid();
+
+	err = perf_evlist__open(evlist);
+	if (err < 0) {
+		pr_debug("perf_evlist__open: %s\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	err = perf_evlist__mmap(evlist, UINT_MAX, false);
+	if (err < 0) {
+		pr_debug("perf_evlist__mmap: %s\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	perf_evlist__enable(evlist);
+
+	/*
+	 * Generate the event:
+	 */
+	open(filename, flags);
+
+	while (1) {
+		int before = nr_events;
+
+		for (i = 0; i < evlist->nr_mmaps; i++) {
+			union perf_event *event;
+
+			while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
+				const u32 type = event->header.type;
+				int tp_flags;
+				struct perf_sample sample;
+
+				++nr_events;
+
+				if (type != PERF_RECORD_SAMPLE) {
+					perf_evlist__mmap_consume(evlist, i);
+					continue;
+				}
+
+				err = perf_evsel__parse_sample(evsel, event, &sample);
+				if (err) {
+					pr_err("Can't parse sample, err = %d\n", err);
+					goto out_delete_evlist;
+				}
+
+				tp_flags = perf_evsel__intval(evsel, &sample, "flags");
+
+				if (flags != tp_flags) {
+					pr_debug("%s: Expected flags=%#x, got %#x\n",
+						 __func__, flags, tp_flags);
+					goto out_delete_evlist;
+				}
+
+				goto out_ok;
+			}
+		}
+
+		if (nr_events == before)
+			perf_evlist__poll(evlist, 10);
+
+		if (++nr_polls > 5) {
+			pr_debug("%s: no events!\n", __func__);
+			goto out_delete_evlist;
+		}
+	}
+out_ok:
+	err = 0;
+out_delete_evlist:
+	perf_evlist__delete(evlist);
+out:
+	return err;
+}
diff --git a/tools/perf/tests/open-syscall.c b/tools/perf/tests/open-syscall.c
new file mode 100644
index 000000000..07aa319bf
--- /dev/null
+++ b/tools/perf/tests/open-syscall.c
@@ -0,0 +1,61 @@
+#include "thread_map.h"
+#include "evsel.h"
+#include "debug.h"
+#include "tests.h"
+
+int test__open_syscall_event(void)
+{
+	int err = -1, fd;
+	struct perf_evsel *evsel;
+	unsigned int nr_open_calls = 111, i;
+	struct thread_map *threads = thread_map__new(-1, getpid(), UINT_MAX);
+	char sbuf[STRERR_BUFSIZE];
+
+	if (threads == NULL) {
+		pr_debug("thread_map__new\n");
+		return -1;
+	}
+
+	evsel = perf_evsel__newtp("syscalls", "sys_enter_open");
+	if (evsel == NULL) {
+		if (tracefs_configured())
+			pr_debug("is tracefs mounted on /sys/kernel/tracing?\n");
+		else if (debugfs_configured())
+			pr_debug("is debugfs mounted on /sys/kernel/debug?\n");
+		else
+			pr_debug("Neither tracefs or debugfs is enabled in this kernel\n");
+		goto out_thread_map_delete;
+	}
+
+	if (perf_evsel__open_per_thread(evsel, threads) < 0) {
+		pr_debug("failed to open counter: %s, "
+			 "tweak /proc/sys/kernel/perf_event_paranoid?\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_evsel_delete;
+	}
+
+	for (i = 0; i < nr_open_calls; ++i) {
+		fd = open("/etc/passwd", O_RDONLY);
+		close(fd);
+	}
+
+	if (perf_evsel__read_on_cpu(evsel, 0, 0) < 0) {
+		pr_debug("perf_evsel__read_on_cpu\n");
+		goto out_close_fd;
+	}
+
+	if (evsel->counts->cpu[0].val != nr_open_calls) {
+		pr_debug("perf_evsel__read_on_cpu: expected to intercept %d calls, got %" PRIu64 "\n",
+			 nr_open_calls, evsel->counts->cpu[0].val);
+		goto out_close_fd;
+	}
+
+	err = 0;
+out_close_fd:
+	perf_evsel__close_fd(evsel, 1, threads->nr);
+out_evsel_delete:
+	perf_evsel__delete(evsel);
+out_thread_map_delete:
+	thread_map__delete(threads);
+	return err;
+}
diff --git a/tools/perf/tests/parse-events.c b/tools/perf/tests/parse-events.c
new file mode 100644
index 000000000..3de744961
--- /dev/null
+++ b/tools/perf/tests/parse-events.c
@@ -0,0 +1,1734 @@
+
+#include "parse-events.h"
+#include "evsel.h"
+#include "evlist.h"
+#include <api/fs/fs.h>
+#include <api/fs/tracefs.h>
+#include <api/fs/debugfs.h>
+#include "tests.h"
+#include "debug.h"
+#include <linux/hw_breakpoint.h>
+
+#define PERF_TP_SAMPLE_TYPE (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | \
+			     PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD)
+
+static int test__checkevent_tracepoint(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 0 == evlist->nr_groups);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_TRACEPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong sample_type",
+		PERF_TP_SAMPLE_TYPE == evsel->attr.sample_type);
+	TEST_ASSERT_VAL("wrong sample_period", 1 == evsel->attr.sample_period);
+	return 0;
+}
+
+static int test__checkevent_tracepoint_multi(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+
+	TEST_ASSERT_VAL("wrong number of entries", evlist->nr_entries > 1);
+	TEST_ASSERT_VAL("wrong number of groups", 0 == evlist->nr_groups);
+
+	evlist__for_each(evlist, evsel) {
+		TEST_ASSERT_VAL("wrong type",
+			PERF_TYPE_TRACEPOINT == evsel->attr.type);
+		TEST_ASSERT_VAL("wrong sample_type",
+			PERF_TP_SAMPLE_TYPE == evsel->attr.sample_type);
+		TEST_ASSERT_VAL("wrong sample_period",
+			1 == evsel->attr.sample_period);
+	}
+	return 0;
+}
+
+static int test__checkevent_raw(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 0x1a == evsel->attr.config);
+	return 0;
+}
+
+static int test__checkevent_numeric(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", 1 == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 1 == evsel->attr.config);
+	return 0;
+}
+
+static int test__checkevent_symbolic_name(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config);
+	return 0;
+}
+
+static int test__checkevent_symbolic_name_config(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong period",
+			100000 == evsel->attr.sample_period);
+	TEST_ASSERT_VAL("wrong config1",
+			0 == evsel->attr.config1);
+	TEST_ASSERT_VAL("wrong config2",
+			1 == evsel->attr.config2);
+	return 0;
+}
+
+static int test__checkevent_symbolic_alias(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_SOFTWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_SW_PAGE_FAULTS == evsel->attr.config);
+	return 0;
+}
+
+static int test__checkevent_genhw(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HW_CACHE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", (1 << 16) == evsel->attr.config);
+	return 0;
+}
+
+static int test__checkevent_breakpoint(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_BREAKPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong bp_type", (HW_BREAKPOINT_R | HW_BREAKPOINT_W) ==
+					 evsel->attr.bp_type);
+	TEST_ASSERT_VAL("wrong bp_len", HW_BREAKPOINT_LEN_4 ==
+					evsel->attr.bp_len);
+	return 0;
+}
+
+static int test__checkevent_breakpoint_x(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_BREAKPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong bp_type",
+			HW_BREAKPOINT_X == evsel->attr.bp_type);
+	TEST_ASSERT_VAL("wrong bp_len", sizeof(long) == evsel->attr.bp_len);
+	return 0;
+}
+
+static int test__checkevent_breakpoint_r(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type",
+			PERF_TYPE_BREAKPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong bp_type",
+			HW_BREAKPOINT_R == evsel->attr.bp_type);
+	TEST_ASSERT_VAL("wrong bp_len",
+			HW_BREAKPOINT_LEN_4 == evsel->attr.bp_len);
+	return 0;
+}
+
+static int test__checkevent_breakpoint_w(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type",
+			PERF_TYPE_BREAKPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong bp_type",
+			HW_BREAKPOINT_W == evsel->attr.bp_type);
+	TEST_ASSERT_VAL("wrong bp_len",
+			HW_BREAKPOINT_LEN_4 == evsel->attr.bp_len);
+	return 0;
+}
+
+static int test__checkevent_breakpoint_rw(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type",
+			PERF_TYPE_BREAKPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong bp_type",
+		(HW_BREAKPOINT_R|HW_BREAKPOINT_W) == evsel->attr.bp_type);
+	TEST_ASSERT_VAL("wrong bp_len",
+			HW_BREAKPOINT_LEN_4 == evsel->attr.bp_len);
+	return 0;
+}
+
+static int test__checkevent_tracepoint_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+
+	return test__checkevent_tracepoint(evlist);
+}
+
+static int
+test__checkevent_tracepoint_multi_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+
+	TEST_ASSERT_VAL("wrong number of entries", evlist->nr_entries > 1);
+
+	evlist__for_each(evlist, evsel) {
+		TEST_ASSERT_VAL("wrong exclude_user",
+				!evsel->attr.exclude_user);
+		TEST_ASSERT_VAL("wrong exclude_kernel",
+				evsel->attr.exclude_kernel);
+		TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+		TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	}
+
+	return test__checkevent_tracepoint_multi(evlist);
+}
+
+static int test__checkevent_raw_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip);
+
+	return test__checkevent_raw(evlist);
+}
+
+static int test__checkevent_numeric_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip);
+
+	return test__checkevent_numeric(evlist);
+}
+
+static int test__checkevent_symbolic_name_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+
+	return test__checkevent_symbolic_name(evlist);
+}
+
+static int test__checkevent_exclude_host_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+
+	return test__checkevent_symbolic_name(evlist);
+}
+
+static int test__checkevent_exclude_guest_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+
+	return test__checkevent_symbolic_name(evlist);
+}
+
+static int test__checkevent_symbolic_alias_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+
+	return test__checkevent_symbolic_alias(evlist);
+}
+
+static int test__checkevent_genhw_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip);
+
+	return test__checkevent_genhw(evlist);
+}
+
+static int test__checkevent_exclude_idle_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude idle", evsel->attr.exclude_idle);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+
+	return test__checkevent_symbolic_name(evlist);
+}
+
+static int test__checkevent_exclude_idle_modifier_1(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude idle", evsel->attr.exclude_idle);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+
+	return test__checkevent_symbolic_name(evlist);
+}
+
+static int test__checkevent_breakpoint_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong name",
+			!strcmp(perf_evsel__name(evsel), "mem:0:u"));
+
+	return test__checkevent_breakpoint(evlist);
+}
+
+static int test__checkevent_breakpoint_x_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong name",
+			!strcmp(perf_evsel__name(evsel), "mem:0:x:k"));
+
+	return test__checkevent_breakpoint_x(evlist);
+}
+
+static int test__checkevent_breakpoint_r_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong name",
+			!strcmp(perf_evsel__name(evsel), "mem:0:r:hp"));
+
+	return test__checkevent_breakpoint_r(evlist);
+}
+
+static int test__checkevent_breakpoint_w_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong name",
+			!strcmp(perf_evsel__name(evsel), "mem:0:w:up"));
+
+	return test__checkevent_breakpoint_w(evlist);
+}
+
+static int test__checkevent_breakpoint_rw_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong name",
+			!strcmp(perf_evsel__name(evsel), "mem:0:rw:kp"));
+
+	return test__checkevent_breakpoint_rw(evlist);
+}
+
+static int test__checkevent_pmu(struct perf_evlist *evlist)
+{
+
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",    10 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong config1",    1 == evsel->attr.config1);
+	TEST_ASSERT_VAL("wrong config2",    3 == evsel->attr.config2);
+	TEST_ASSERT_VAL("wrong period",  1000 == evsel->attr.sample_period);
+
+	return 0;
+}
+
+static int test__checkevent_list(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 3 == evlist->nr_entries);
+
+	/* r1 */
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 1 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong config1", 0 == evsel->attr.config1);
+	TEST_ASSERT_VAL("wrong config2", 0 == evsel->attr.config2);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+
+	/* syscalls:sys_enter_open:k */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_TRACEPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong sample_type",
+		PERF_TP_SAMPLE_TYPE == evsel->attr.sample_type);
+	TEST_ASSERT_VAL("wrong sample_period", 1 == evsel->attr.sample_period);
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+
+	/* 1:1:hp */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", 1 == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 1 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip);
+
+	return 0;
+}
+
+static int test__checkevent_pmu_name(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	/* cpu/config=1,name=krava/u */
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",  1 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong name", !strcmp(perf_evsel__name(evsel), "krava"));
+
+	/* cpu/config=2/u" */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",  2 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong name",
+			!strcmp(perf_evsel__name(evsel), "cpu/config=2/u"));
+
+	return 0;
+}
+
+static int test__checkevent_pmu_events(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong exclude_user",
+			!evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel",
+			evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong pinned", !evsel->attr.pinned);
+
+	return 0;
+}
+
+
+static int test__checkevent_pmu_events_mix(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	/* pmu-event:u */
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong exclude_user",
+			!evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel",
+			evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong pinned", !evsel->attr.pinned);
+
+	/* cpu/pmu-event/u*/
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_RAW == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong exclude_user",
+			!evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel",
+			evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong pinned", !evsel->attr.pinned);
+
+	return 0;
+}
+
+static int test__checkterms_simple(struct list_head *terms)
+{
+	struct parse_events_term *term;
+
+	/* config=10 */
+	term = list_entry(terms->next, struct parse_events_term, list);
+	TEST_ASSERT_VAL("wrong type term",
+			term->type_term == PARSE_EVENTS__TERM_TYPE_CONFIG);
+	TEST_ASSERT_VAL("wrong type val",
+			term->type_val == PARSE_EVENTS__TERM_TYPE_NUM);
+	TEST_ASSERT_VAL("wrong val", term->val.num == 10);
+	TEST_ASSERT_VAL("wrong config", !term->config);
+
+	/* config1 */
+	term = list_entry(term->list.next, struct parse_events_term, list);
+	TEST_ASSERT_VAL("wrong type term",
+			term->type_term == PARSE_EVENTS__TERM_TYPE_CONFIG1);
+	TEST_ASSERT_VAL("wrong type val",
+			term->type_val == PARSE_EVENTS__TERM_TYPE_NUM);
+	TEST_ASSERT_VAL("wrong val", term->val.num == 1);
+	TEST_ASSERT_VAL("wrong config", !term->config);
+
+	/* config2=3 */
+	term = list_entry(term->list.next, struct parse_events_term, list);
+	TEST_ASSERT_VAL("wrong type term",
+			term->type_term == PARSE_EVENTS__TERM_TYPE_CONFIG2);
+	TEST_ASSERT_VAL("wrong type val",
+			term->type_val == PARSE_EVENTS__TERM_TYPE_NUM);
+	TEST_ASSERT_VAL("wrong val", term->val.num == 3);
+	TEST_ASSERT_VAL("wrong config", !term->config);
+
+	/* umask=1*/
+	term = list_entry(term->list.next, struct parse_events_term, list);
+	TEST_ASSERT_VAL("wrong type term",
+			term->type_term == PARSE_EVENTS__TERM_TYPE_USER);
+	TEST_ASSERT_VAL("wrong type val",
+			term->type_val == PARSE_EVENTS__TERM_TYPE_NUM);
+	TEST_ASSERT_VAL("wrong val", term->val.num == 1);
+	TEST_ASSERT_VAL("wrong config", !strcmp(term->config, "umask"));
+
+	return 0;
+}
+
+static int test__group1(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups);
+
+	/* instructions:k */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* cycles:upp */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	/* use of precise requires exclude_guest */
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip == 2);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	return 0;
+}
+
+static int test__group2(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 3 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups);
+
+	/* faults + :ku modifier */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_SOFTWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_SW_PAGE_FAULTS == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* cache-references + :u modifier */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CACHE_REFERENCES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* cycles:k */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	return 0;
+}
+
+static int test__group3(struct perf_evlist *evlist __maybe_unused)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 5 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 2 == evlist->nr_groups);
+
+	/* group1 syscalls:sys_enter_open:H */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_TRACEPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong sample_type",
+		PERF_TP_SAMPLE_TYPE == evsel->attr.sample_type);
+	TEST_ASSERT_VAL("wrong sample_period", 1 == evsel->attr.sample_period);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong group name",
+		!strcmp(leader->group_name, "group1"));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* group1 cycles:kppp */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	/* use of precise requires exclude_guest */
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip == 3);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* group2 cycles + G modifier */
+	evsel = leader = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong group name",
+		!strcmp(leader->group_name, "group2"));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* group2 1:3 + G modifier */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", 1 == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 3 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* instructions:u */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	return 0;
+}
+
+static int test__group4(struct perf_evlist *evlist __maybe_unused)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups);
+
+	/* cycles:u + p */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	/* use of precise requires exclude_guest */
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip == 1);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* instructions:kp + p */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	/* use of precise requires exclude_guest */
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip == 2);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	return 0;
+}
+
+static int test__group5(struct perf_evlist *evlist __maybe_unused)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 5 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 2 == evlist->nr_groups);
+
+	/* cycles + G */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* instructions + G */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* cycles:G */
+	evsel = leader = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+	TEST_ASSERT_VAL("wrong sample_read", !evsel->sample_read);
+
+	/* instructions:G */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+
+	/* cycles */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+
+	return 0;
+}
+
+static int test__group_gh1(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups);
+
+	/* cycles + :H group modifier */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+
+	/* cache-misses:G + :H group modifier */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+
+	return 0;
+}
+
+static int test__group_gh2(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups);
+
+	/* cycles + :G group modifier */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+
+	/* cache-misses:H + :G group modifier */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+
+	return 0;
+}
+
+static int test__group_gh3(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups);
+
+	/* cycles:G + :u group modifier */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+
+	/* cache-misses:H + :u group modifier */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+
+	return 0;
+}
+
+static int test__group_gh4(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong number of groups", 1 == evlist->nr_groups);
+
+	/* cycles:G + :uG group modifier */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", perf_evsel__is_group_leader(evsel));
+	TEST_ASSERT_VAL("wrong nr_members", evsel->nr_members == 2);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 0);
+
+	/* cache-misses:H + :uG group modifier */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", !evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong group_idx", perf_evsel__group_idx(evsel) == 1);
+
+	return 0;
+}
+
+static int test__leader_sample1(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 3 == evlist->nr_entries);
+
+	/* cycles - sampling group leader */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong sample_read", evsel->sample_read);
+
+	/* cache-misses - not sampling */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong sample_read", evsel->sample_read);
+
+	/* branch-misses - not sampling */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_BRANCH_MISSES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", !evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", !evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong sample_read", evsel->sample_read);
+
+	return 0;
+}
+
+static int test__leader_sample2(struct perf_evlist *evlist __maybe_unused)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 2 == evlist->nr_entries);
+
+	/* instructions - sampling group leader */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_INSTRUCTIONS == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong sample_read", evsel->sample_read);
+
+	/* branch-misses - not sampling */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_BRANCH_MISSES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong exclude guest", evsel->attr.exclude_guest);
+	TEST_ASSERT_VAL("wrong exclude host", !evsel->attr.exclude_host);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong sample_read", evsel->sample_read);
+
+	return 0;
+}
+
+static int test__checkevent_pinned_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", evsel->attr.precise_ip);
+	TEST_ASSERT_VAL("wrong pinned", evsel->attr.pinned);
+
+	return test__checkevent_symbolic_name(evlist);
+}
+
+static int test__pinned_group(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel, *leader;
+
+	TEST_ASSERT_VAL("wrong number of entries", 3 == evlist->nr_entries);
+
+	/* cycles - group leader */
+	evsel = leader = perf_evlist__first(evlist);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CPU_CYCLES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong group name", !evsel->group_name);
+	TEST_ASSERT_VAL("wrong leader", evsel->leader == leader);
+	TEST_ASSERT_VAL("wrong pinned", evsel->attr.pinned);
+
+	/* cache-misses - can not be pinned, but will go on with the leader */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_HARDWARE == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_CACHE_MISSES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong pinned", !evsel->attr.pinned);
+
+	/* branch-misses - ditto */
+	evsel = perf_evsel__next(evsel);
+	TEST_ASSERT_VAL("wrong config",
+			PERF_COUNT_HW_BRANCH_MISSES == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong pinned", !evsel->attr.pinned);
+
+	return 0;
+}
+
+static int test__checkevent_breakpoint_len(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_BREAKPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong bp_type", (HW_BREAKPOINT_R | HW_BREAKPOINT_W) ==
+					 evsel->attr.bp_type);
+	TEST_ASSERT_VAL("wrong bp_len", HW_BREAKPOINT_LEN_1 ==
+					evsel->attr.bp_len);
+
+	return 0;
+}
+
+static int test__checkevent_breakpoint_len_w(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong number of entries", 1 == evlist->nr_entries);
+	TEST_ASSERT_VAL("wrong type", PERF_TYPE_BREAKPOINT == evsel->attr.type);
+	TEST_ASSERT_VAL("wrong config", 0 == evsel->attr.config);
+	TEST_ASSERT_VAL("wrong bp_type", HW_BREAKPOINT_W ==
+					 evsel->attr.bp_type);
+	TEST_ASSERT_VAL("wrong bp_len", HW_BREAKPOINT_LEN_2 ==
+					evsel->attr.bp_len);
+
+	return 0;
+}
+
+static int
+test__checkevent_breakpoint_len_rw_modifier(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel = perf_evlist__first(evlist);
+
+	TEST_ASSERT_VAL("wrong exclude_user", !evsel->attr.exclude_user);
+	TEST_ASSERT_VAL("wrong exclude_kernel", evsel->attr.exclude_kernel);
+	TEST_ASSERT_VAL("wrong exclude_hv", evsel->attr.exclude_hv);
+	TEST_ASSERT_VAL("wrong precise_ip", !evsel->attr.precise_ip);
+
+	return test__checkevent_breakpoint_rw(evlist);
+}
+
+static int count_tracepoints(void)
+{
+	char events_path[PATH_MAX];
+	struct dirent *events_ent;
+	const char *mountpoint;
+	DIR *events_dir;
+	int cnt = 0;
+
+	mountpoint = tracefs_find_mountpoint();
+	if (mountpoint) {
+		scnprintf(events_path, PATH_MAX, "%s/events",
+			  mountpoint);
+	} else {
+		mountpoint = debugfs_find_mountpoint();
+		scnprintf(events_path, PATH_MAX, "%s/tracing/events",
+			  mountpoint);
+	}
+
+	events_dir = opendir(events_path);
+
+	TEST_ASSERT_VAL("Can't open events dir", events_dir);
+
+	while ((events_ent = readdir(events_dir))) {
+		char sys_path[PATH_MAX];
+		struct dirent *sys_ent;
+		DIR *sys_dir;
+
+		if (!strcmp(events_ent->d_name, ".")
+		    || !strcmp(events_ent->d_name, "..")
+		    || !strcmp(events_ent->d_name, "enable")
+		    || !strcmp(events_ent->d_name, "header_event")
+		    || !strcmp(events_ent->d_name, "header_page"))
+			continue;
+
+		scnprintf(sys_path, PATH_MAX, "%s/%s",
+			  events_path, events_ent->d_name);
+
+		sys_dir = opendir(sys_path);
+		TEST_ASSERT_VAL("Can't open sys dir", sys_dir);
+
+		while ((sys_ent = readdir(sys_dir))) {
+			if (!strcmp(sys_ent->d_name, ".")
+			    || !strcmp(sys_ent->d_name, "..")
+			    || !strcmp(sys_ent->d_name, "enable")
+			    || !strcmp(sys_ent->d_name, "filter"))
+				continue;
+
+			cnt++;
+		}
+
+		closedir(sys_dir);
+	}
+
+	closedir(events_dir);
+	return cnt;
+}
+
+static int test__all_tracepoints(struct perf_evlist *evlist)
+{
+	TEST_ASSERT_VAL("wrong events count",
+			count_tracepoints() == evlist->nr_entries);
+
+	return test__checkevent_tracepoint_multi(evlist);
+}
+
+struct evlist_test {
+	const char *name;
+	__u32 type;
+	const int id;
+	int (*check)(struct perf_evlist *evlist);
+};
+
+static struct evlist_test test__events[] = {
+	{
+		.name  = "syscalls:sys_enter_open",
+		.check = test__checkevent_tracepoint,
+		.id    = 0,
+	},
+	{
+		.name  = "syscalls:*",
+		.check = test__checkevent_tracepoint_multi,
+		.id    = 1,
+	},
+	{
+		.name  = "r1a",
+		.check = test__checkevent_raw,
+		.id    = 2,
+	},
+	{
+		.name  = "1:1",
+		.check = test__checkevent_numeric,
+		.id    = 3,
+	},
+	{
+		.name  = "instructions",
+		.check = test__checkevent_symbolic_name,
+		.id    = 4,
+	},
+	{
+		.name  = "cycles/period=100000,config2/",
+		.check = test__checkevent_symbolic_name_config,
+		.id    = 5,
+	},
+	{
+		.name  = "faults",
+		.check = test__checkevent_symbolic_alias,
+		.id    = 6,
+	},
+	{
+		.name  = "L1-dcache-load-miss",
+		.check = test__checkevent_genhw,
+		.id    = 7,
+	},
+	{
+		.name  = "mem:0",
+		.check = test__checkevent_breakpoint,
+		.id    = 8,
+	},
+	{
+		.name  = "mem:0:x",
+		.check = test__checkevent_breakpoint_x,
+		.id    = 9,
+	},
+	{
+		.name  = "mem:0:r",
+		.check = test__checkevent_breakpoint_r,
+		.id    = 10,
+	},
+	{
+		.name  = "mem:0:w",
+		.check = test__checkevent_breakpoint_w,
+		.id    = 11,
+	},
+	{
+		.name  = "syscalls:sys_enter_open:k",
+		.check = test__checkevent_tracepoint_modifier,
+		.id    = 12,
+	},
+	{
+		.name  = "syscalls:*:u",
+		.check = test__checkevent_tracepoint_multi_modifier,
+		.id    = 13,
+	},
+	{
+		.name  = "r1a:kp",
+		.check = test__checkevent_raw_modifier,
+		.id    = 14,
+	},
+	{
+		.name  = "1:1:hp",
+		.check = test__checkevent_numeric_modifier,
+		.id    = 15,
+	},
+	{
+		.name  = "instructions:h",
+		.check = test__checkevent_symbolic_name_modifier,
+		.id    = 16,
+	},
+	{
+		.name  = "faults:u",
+		.check = test__checkevent_symbolic_alias_modifier,
+		.id    = 17,
+	},
+	{
+		.name  = "L1-dcache-load-miss:kp",
+		.check = test__checkevent_genhw_modifier,
+		.id    = 18,
+	},
+	{
+		.name  = "mem:0:u",
+		.check = test__checkevent_breakpoint_modifier,
+		.id    = 19,
+	},
+	{
+		.name  = "mem:0:x:k",
+		.check = test__checkevent_breakpoint_x_modifier,
+		.id    = 20,
+	},
+	{
+		.name  = "mem:0:r:hp",
+		.check = test__checkevent_breakpoint_r_modifier,
+		.id    = 21,
+	},
+	{
+		.name  = "mem:0:w:up",
+		.check = test__checkevent_breakpoint_w_modifier,
+		.id    = 22,
+	},
+	{
+		.name  = "r1,syscalls:sys_enter_open:k,1:1:hp",
+		.check = test__checkevent_list,
+		.id    = 23,
+	},
+	{
+		.name  = "instructions:G",
+		.check = test__checkevent_exclude_host_modifier,
+		.id    = 24,
+	},
+	{
+		.name  = "instructions:H",
+		.check = test__checkevent_exclude_guest_modifier,
+		.id    = 25,
+	},
+	{
+		.name  = "mem:0:rw",
+		.check = test__checkevent_breakpoint_rw,
+		.id    = 26,
+	},
+	{
+		.name  = "mem:0:rw:kp",
+		.check = test__checkevent_breakpoint_rw_modifier,
+		.id    = 27,
+	},
+	{
+		.name  = "{instructions:k,cycles:upp}",
+		.check = test__group1,
+		.id    = 28,
+	},
+	{
+		.name  = "{faults:k,cache-references}:u,cycles:k",
+		.check = test__group2,
+		.id    = 29,
+	},
+	{
+		.name  = "group1{syscalls:sys_enter_open:H,cycles:kppp},group2{cycles,1:3}:G,instructions:u",
+		.check = test__group3,
+		.id    = 30,
+	},
+	{
+		.name  = "{cycles:u,instructions:kp}:p",
+		.check = test__group4,
+		.id    = 31,
+	},
+	{
+		.name  = "{cycles,instructions}:G,{cycles:G,instructions:G},cycles",
+		.check = test__group5,
+		.id    = 32,
+	},
+	{
+		.name  = "*:*",
+		.check = test__all_tracepoints,
+		.id    = 33,
+	},
+	{
+		.name  = "{cycles,cache-misses:G}:H",
+		.check = test__group_gh1,
+		.id    = 34,
+	},
+	{
+		.name  = "{cycles,cache-misses:H}:G",
+		.check = test__group_gh2,
+		.id    = 35,
+	},
+	{
+		.name  = "{cycles:G,cache-misses:H}:u",
+		.check = test__group_gh3,
+		.id    = 36,
+	},
+	{
+		.name  = "{cycles:G,cache-misses:H}:uG",
+		.check = test__group_gh4,
+		.id    = 37,
+	},
+	{
+		.name  = "{cycles,cache-misses,branch-misses}:S",
+		.check = test__leader_sample1,
+		.id    = 38,
+	},
+	{
+		.name  = "{instructions,branch-misses}:Su",
+		.check = test__leader_sample2,
+		.id    = 39,
+	},
+	{
+		.name  = "instructions:uDp",
+		.check = test__checkevent_pinned_modifier,
+		.id    = 40,
+	},
+	{
+		.name  = "{cycles,cache-misses,branch-misses}:D",
+		.check = test__pinned_group,
+		.id    = 41,
+	},
+	{
+		.name  = "mem:0/1",
+		.check = test__checkevent_breakpoint_len,
+		.id    = 42,
+	},
+	{
+		.name  = "mem:0/2:w",
+		.check = test__checkevent_breakpoint_len_w,
+		.id    = 43,
+	},
+	{
+		.name  = "mem:0/4:rw:u",
+		.check = test__checkevent_breakpoint_len_rw_modifier,
+		.id    = 44
+	},
+#if defined(__s390x__)
+	{
+		.name  = "kvm-s390:kvm_s390_create_vm",
+		.check = test__checkevent_tracepoint,
+		.id    = 100,
+	},
+#endif
+	{
+		.name  = "instructions:I",
+		.check = test__checkevent_exclude_idle_modifier,
+		.id    = 45,
+	},
+	{
+		.name  = "instructions:kIG",
+		.check = test__checkevent_exclude_idle_modifier_1,
+		.id    = 46,
+	},
+};
+
+static struct evlist_test test__events_pmu[] = {
+	{
+		.name  = "cpu/config=10,config1,config2=3,period=1000/u",
+		.check = test__checkevent_pmu,
+		.id    = 0,
+	},
+	{
+		.name  = "cpu/config=1,name=krava/u,cpu/config=2/u",
+		.check = test__checkevent_pmu_name,
+		.id    = 1,
+	},
+};
+
+struct terms_test {
+	const char *str;
+	__u32 type;
+	int (*check)(struct list_head *terms);
+};
+
+static struct terms_test test__terms[] = {
+	[0] = {
+		.str   = "config=10,config1,config2=3,umask=1",
+		.check = test__checkterms_simple,
+	},
+};
+
+static int test_event(struct evlist_test *e)
+{
+	struct perf_evlist *evlist;
+	int ret;
+
+	evlist = perf_evlist__new();
+	if (evlist == NULL)
+		return -ENOMEM;
+
+	ret = parse_events(evlist, e->name);
+	if (ret) {
+		pr_debug("failed to parse event '%s', err %d\n",
+			 e->name, ret);
+	} else {
+		ret = e->check(evlist);
+	}
+
+	perf_evlist__delete(evlist);
+
+	return ret;
+}
+
+static int test_events(struct evlist_test *events, unsigned cnt)
+{
+	int ret1, ret2 = 0;
+	unsigned i;
+
+	for (i = 0; i < cnt; i++) {
+		struct evlist_test *e = &events[i];
+
+		pr_debug("running test %d '%s'\n", e->id, e->name);
+		ret1 = test_event(e);
+		if (ret1)
+			ret2 = ret1;
+	}
+
+	return ret2;
+}
+
+static int test_term(struct terms_test *t)
+{
+	struct list_head terms;
+	int ret;
+
+	INIT_LIST_HEAD(&terms);
+
+	ret = parse_events_terms(&terms, t->str);
+	if (ret) {
+		pr_debug("failed to parse terms '%s', err %d\n",
+			 t->str , ret);
+		return ret;
+	}
+
+	ret = t->check(&terms);
+	parse_events__free_terms(&terms);
+
+	return ret;
+}
+
+static int test_terms(struct terms_test *terms, unsigned cnt)
+{
+	int ret = 0;
+	unsigned i;
+
+	for (i = 0; i < cnt; i++) {
+		struct terms_test *t = &terms[i];
+
+		pr_debug("running test %d '%s'\n", i, t->str);
+		ret = test_term(t);
+		if (ret)
+			break;
+	}
+
+	return ret;
+}
+
+static int test_pmu(void)
+{
+	struct stat st;
+	char path[PATH_MAX];
+	int ret;
+
+	snprintf(path, PATH_MAX, "%s/bus/event_source/devices/cpu/format/",
+		 sysfs__mountpoint());
+
+	ret = stat(path, &st);
+	if (ret)
+		pr_debug("omitting PMU cpu tests\n");
+	return !ret;
+}
+
+static int test_pmu_events(void)
+{
+	struct stat st;
+	char path[PATH_MAX];
+	struct dirent *ent;
+	DIR *dir;
+	int ret;
+
+	snprintf(path, PATH_MAX, "%s/bus/event_source/devices/cpu/events/",
+		 sysfs__mountpoint());
+
+	ret = stat(path, &st);
+	if (ret) {
+		pr_debug("omitting PMU cpu events tests\n");
+		return 0;
+	}
+
+	dir = opendir(path);
+	if (!dir) {
+		pr_debug("can't open pmu event dir");
+		return -1;
+	}
+
+	while (!ret && (ent = readdir(dir))) {
+#define MAX_NAME 100
+		struct evlist_test e;
+		char name[MAX_NAME];
+
+		if (!strcmp(ent->d_name, ".") ||
+		    !strcmp(ent->d_name, ".."))
+			continue;
+
+		snprintf(name, MAX_NAME, "cpu/event=%s/u", ent->d_name);
+
+		e.name  = name;
+		e.check = test__checkevent_pmu_events;
+
+		ret = test_event(&e);
+		if (ret)
+			break;
+		snprintf(name, MAX_NAME, "%s:u,cpu/event=%s/u", ent->d_name, ent->d_name);
+		e.name  = name;
+		e.check = test__checkevent_pmu_events_mix;
+		ret = test_event(&e);
+#undef MAX_NAME
+	}
+
+	closedir(dir);
+	return ret;
+}
+
+int test__parse_events(void)
+{
+	int ret1, ret2 = 0;
+
+#define TEST_EVENTS(tests)				\
+do {							\
+	ret1 = test_events(tests, ARRAY_SIZE(tests));	\
+	if (!ret2)					\
+		ret2 = ret1;				\
+} while (0)
+
+	TEST_EVENTS(test__events);
+
+	if (test_pmu())
+		TEST_EVENTS(test__events_pmu);
+
+	if (test_pmu()) {
+		int ret = test_pmu_events();
+		if (ret)
+			return ret;
+	}
+
+	ret1 = test_terms(test__terms, ARRAY_SIZE(test__terms));
+	if (!ret2)
+		ret2 = ret1;
+
+	return ret2;
+}
diff --git a/tools/perf/tests/parse-no-sample-id-all.c b/tools/perf/tests/parse-no-sample-id-all.c
new file mode 100644
index 000000000..2c63ea658
--- /dev/null
+++ b/tools/perf/tests/parse-no-sample-id-all.c
@@ -0,0 +1,109 @@
+#include <linux/types.h>
+#include <stddef.h>
+
+#include "tests.h"
+
+#include "event.h"
+#include "evlist.h"
+#include "header.h"
+#include "util.h"
+#include "debug.h"
+
+static int process_event(struct perf_evlist **pevlist, union perf_event *event)
+{
+	struct perf_sample sample;
+
+	if (event->header.type == PERF_RECORD_HEADER_ATTR) {
+		if (perf_event__process_attr(NULL, event, pevlist)) {
+			pr_debug("perf_event__process_attr failed\n");
+			return -1;
+		}
+		return 0;
+	}
+
+	if (event->header.type >= PERF_RECORD_USER_TYPE_START)
+		return -1;
+
+	if (!*pevlist)
+		return -1;
+
+	if (perf_evlist__parse_sample(*pevlist, event, &sample)) {
+		pr_debug("perf_evlist__parse_sample failed\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+static int process_events(union perf_event **events, size_t count)
+{
+	struct perf_evlist *evlist = NULL;
+	int err = 0;
+	size_t i;
+
+	for (i = 0; i < count && !err; i++)
+		err = process_event(&evlist, events[i]);
+
+	if (evlist)
+		perf_evlist__delete(evlist);
+
+	return err;
+}
+
+struct test_attr_event {
+	struct attr_event attr;
+	u64 id;
+};
+
+/**
+ * test__parse_no_sample_id_all - test parsing with no sample_id_all bit set.
+ *
+ * This function tests parsing data produced on kernel's that do not support the
+ * sample_id_all bit.  Without the sample_id_all bit, non-sample events (such as
+ * mmap events) do not have an id sample appended, and consequently logic
+ * designed to determine the id will not work.  That case happens when there is
+ * more than one selected event, so this test processes three events: 2
+ * attributes representing the selected events and one mmap event.
+ *
+ * Return: %0 on success, %-1 if the test fails.
+ */
+int test__parse_no_sample_id_all(void)
+{
+	int err;
+
+	struct test_attr_event event1 = {
+		.attr = {
+			.header = {
+				.type = PERF_RECORD_HEADER_ATTR,
+				.size = sizeof(struct test_attr_event),
+			},
+		},
+		.id = 1,
+	};
+	struct test_attr_event event2 = {
+		.attr = {
+			.header = {
+				.type = PERF_RECORD_HEADER_ATTR,
+				.size = sizeof(struct test_attr_event),
+			},
+		},
+		.id = 2,
+	};
+	struct mmap_event event3 = {
+		.header = {
+			.type = PERF_RECORD_MMAP,
+			.size = sizeof(struct mmap_event),
+		},
+	};
+	union perf_event *events[] = {
+		(union perf_event *)&event1,
+		(union perf_event *)&event2,
+		(union perf_event *)&event3,
+	};
+
+	err = process_events(events, ARRAY_SIZE(events));
+	if (err)
+		return -1;
+
+	return 0;
+}
diff --git a/tools/perf/tests/perf-record.c b/tools/perf/tests/perf-record.c
new file mode 100644
index 000000000..7a228a2a0
--- /dev/null
+++ b/tools/perf/tests/perf-record.c
@@ -0,0 +1,314 @@
+#include <sched.h>
+#include "evlist.h"
+#include "evsel.h"
+#include "perf.h"
+#include "debug.h"
+#include "tests.h"
+
+static int sched__get_first_possible_cpu(pid_t pid, cpu_set_t *maskp)
+{
+	int i, cpu = -1, nrcpus = 1024;
+realloc:
+	CPU_ZERO(maskp);
+
+	if (sched_getaffinity(pid, sizeof(*maskp), maskp) == -1) {
+		if (errno == EINVAL && nrcpus < (1024 << 8)) {
+			nrcpus = nrcpus << 2;
+			goto realloc;
+		}
+		perror("sched_getaffinity");
+			return -1;
+	}
+
+	for (i = 0; i < nrcpus; i++) {
+		if (CPU_ISSET(i, maskp)) {
+			if (cpu == -1)
+				cpu = i;
+			else
+				CPU_CLR(i, maskp);
+		}
+	}
+
+	return cpu;
+}
+
+int test__PERF_RECORD(void)
+{
+	struct record_opts opts = {
+		.target = {
+			.uid = UINT_MAX,
+			.uses_mmap = true,
+		},
+		.no_buffering = true,
+		.freq	      = 10,
+		.mmap_pages   = 256,
+	};
+	cpu_set_t cpu_mask;
+	size_t cpu_mask_size = sizeof(cpu_mask);
+	struct perf_evlist *evlist = perf_evlist__new_default();
+	struct perf_evsel *evsel;
+	struct perf_sample sample;
+	const char *cmd = "sleep";
+	const char *argv[] = { cmd, "1", NULL, };
+	char *bname, *mmap_filename;
+	u64 prev_time = 0;
+	bool found_cmd_mmap = false,
+	     found_libc_mmap = false,
+	     found_vdso_mmap = false,
+	     found_ld_mmap = false;
+	int err = -1, errs = 0, i, wakeups = 0;
+	u32 cpu;
+	int total_events = 0, nr_events[PERF_RECORD_MAX] = { 0, };
+	char sbuf[STRERR_BUFSIZE];
+
+	if (evlist == NULL || argv == NULL) {
+		pr_debug("Not enough memory to create evlist\n");
+		goto out;
+	}
+
+	/*
+	 * Create maps of threads and cpus to monitor. In this case
+	 * we start with all threads and cpus (-1, -1) but then in
+	 * perf_evlist__prepare_workload we'll fill in the only thread
+	 * we're monitoring, the one forked there.
+	 */
+	err = perf_evlist__create_maps(evlist, &opts.target);
+	if (err < 0) {
+		pr_debug("Not enough memory to create thread/cpu maps\n");
+		goto out_delete_evlist;
+	}
+
+	/*
+	 * Prepare the workload in argv[] to run, it'll fork it, and then wait
+	 * for perf_evlist__start_workload() to exec it. This is done this way
+	 * so that we have time to open the evlist (calling sys_perf_event_open
+	 * on all the fds) and then mmap them.
+	 */
+	err = perf_evlist__prepare_workload(evlist, &opts.target, argv, false, NULL);
+	if (err < 0) {
+		pr_debug("Couldn't run the workload!\n");
+		goto out_delete_evlist;
+	}
+
+	/*
+	 * Config the evsels, setting attr->comm on the first one, etc.
+	 */
+	evsel = perf_evlist__first(evlist);
+	perf_evsel__set_sample_bit(evsel, CPU);
+	perf_evsel__set_sample_bit(evsel, TID);
+	perf_evsel__set_sample_bit(evsel, TIME);
+	perf_evlist__config(evlist, &opts);
+
+	err = sched__get_first_possible_cpu(evlist->workload.pid, &cpu_mask);
+	if (err < 0) {
+		pr_debug("sched__get_first_possible_cpu: %s\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	cpu = err;
+
+	/*
+	 * So that we can check perf_sample.cpu on all the samples.
+	 */
+	if (sched_setaffinity(evlist->workload.pid, cpu_mask_size, &cpu_mask) < 0) {
+		pr_debug("sched_setaffinity: %s\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	/*
+	 * Call sys_perf_event_open on all the fds on all the evsels,
+	 * grouping them if asked to.
+	 */
+	err = perf_evlist__open(evlist);
+	if (err < 0) {
+		pr_debug("perf_evlist__open: %s\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	/*
+	 * mmap the first fd on a given CPU and ask for events for the other
+	 * fds in the same CPU to be injected in the same mmap ring buffer
+	 * (using ioctl(PERF_EVENT_IOC_SET_OUTPUT)).
+	 */
+	err = perf_evlist__mmap(evlist, opts.mmap_pages, false);
+	if (err < 0) {
+		pr_debug("perf_evlist__mmap: %s\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	/*
+	 * Now that all is properly set up, enable the events, they will
+	 * count just on workload.pid, which will start...
+	 */
+	perf_evlist__enable(evlist);
+
+	/*
+	 * Now!
+	 */
+	perf_evlist__start_workload(evlist);
+
+	while (1) {
+		int before = total_events;
+
+		for (i = 0; i < evlist->nr_mmaps; i++) {
+			union perf_event *event;
+
+			while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
+				const u32 type = event->header.type;
+				const char *name = perf_event__name(type);
+
+				++total_events;
+				if (type < PERF_RECORD_MAX)
+					nr_events[type]++;
+
+				err = perf_evlist__parse_sample(evlist, event, &sample);
+				if (err < 0) {
+					if (verbose)
+						perf_event__fprintf(event, stderr);
+					pr_debug("Couldn't parse sample\n");
+					goto out_delete_evlist;
+				}
+
+				if (verbose) {
+					pr_info("%" PRIu64" %d ", sample.time, sample.cpu);
+					perf_event__fprintf(event, stderr);
+				}
+
+				if (prev_time > sample.time) {
+					pr_debug("%s going backwards in time, prev=%" PRIu64 ", curr=%" PRIu64 "\n",
+						 name, prev_time, sample.time);
+					++errs;
+				}
+
+				prev_time = sample.time;
+
+				if (sample.cpu != cpu) {
+					pr_debug("%s with unexpected cpu, expected %d, got %d\n",
+						 name, cpu, sample.cpu);
+					++errs;
+				}
+
+				if ((pid_t)sample.pid != evlist->workload.pid) {
+					pr_debug("%s with unexpected pid, expected %d, got %d\n",
+						 name, evlist->workload.pid, sample.pid);
+					++errs;
+				}
+
+				if ((pid_t)sample.tid != evlist->workload.pid) {
+					pr_debug("%s with unexpected tid, expected %d, got %d\n",
+						 name, evlist->workload.pid, sample.tid);
+					++errs;
+				}
+
+				if ((type == PERF_RECORD_COMM ||
+				     type == PERF_RECORD_MMAP ||
+				     type == PERF_RECORD_MMAP2 ||
+				     type == PERF_RECORD_FORK ||
+				     type == PERF_RECORD_EXIT) &&
+				     (pid_t)event->comm.pid != evlist->workload.pid) {
+					pr_debug("%s with unexpected pid/tid\n", name);
+					++errs;
+				}
+
+				if ((type == PERF_RECORD_COMM ||
+				     type == PERF_RECORD_MMAP ||
+				     type == PERF_RECORD_MMAP2) &&
+				     event->comm.pid != event->comm.tid) {
+					pr_debug("%s with different pid/tid!\n", name);
+					++errs;
+				}
+
+				switch (type) {
+				case PERF_RECORD_COMM:
+					if (strcmp(event->comm.comm, cmd)) {
+						pr_debug("%s with unexpected comm!\n", name);
+						++errs;
+					}
+					break;
+				case PERF_RECORD_EXIT:
+					goto found_exit;
+				case PERF_RECORD_MMAP:
+					mmap_filename = event->mmap.filename;
+					goto check_bname;
+				case PERF_RECORD_MMAP2:
+					mmap_filename = event->mmap2.filename;
+				check_bname:
+					bname = strrchr(mmap_filename, '/');
+					if (bname != NULL) {
+						if (!found_cmd_mmap)
+							found_cmd_mmap = !strcmp(bname + 1, cmd);
+						if (!found_libc_mmap)
+							found_libc_mmap = !strncmp(bname + 1, "libc", 4);
+						if (!found_ld_mmap)
+							found_ld_mmap = !strncmp(bname + 1, "ld", 2);
+					} else if (!found_vdso_mmap)
+						found_vdso_mmap = !strcmp(mmap_filename, "[vdso]");
+					break;
+
+				case PERF_RECORD_SAMPLE:
+					/* Just ignore samples for now */
+					break;
+				default:
+					pr_debug("Unexpected perf_event->header.type %d!\n",
+						 type);
+					++errs;
+				}
+
+				perf_evlist__mmap_consume(evlist, i);
+			}
+		}
+
+		/*
+		 * We don't use poll here because at least at 3.1 times the
+		 * PERF_RECORD_{!SAMPLE} events don't honour
+		 * perf_event_attr.wakeup_events, just PERF_EVENT_SAMPLE does.
+		 */
+		if (total_events == before && false)
+			perf_evlist__poll(evlist, -1);
+
+		sleep(1);
+		if (++wakeups > 5) {
+			pr_debug("No PERF_RECORD_EXIT event!\n");
+			break;
+		}
+	}
+
+found_exit:
+	if (nr_events[PERF_RECORD_COMM] > 1) {
+		pr_debug("Excessive number of PERF_RECORD_COMM events!\n");
+		++errs;
+	}
+
+	if (nr_events[PERF_RECORD_COMM] == 0) {
+		pr_debug("Missing PERF_RECORD_COMM for %s!\n", cmd);
+		++errs;
+	}
+
+	if (!found_cmd_mmap) {
+		pr_debug("PERF_RECORD_MMAP for %s missing!\n", cmd);
+		++errs;
+	}
+
+	if (!found_libc_mmap) {
+		pr_debug("PERF_RECORD_MMAP for %s missing!\n", "libc");
+		++errs;
+	}
+
+	if (!found_ld_mmap) {
+		pr_debug("PERF_RECORD_MMAP for %s missing!\n", "ld");
+		++errs;
+	}
+
+	if (!found_vdso_mmap) {
+		pr_debug("PERF_RECORD_MMAP for %s missing!\n", "[vdso]");
+		++errs;
+	}
+out_delete_evlist:
+	perf_evlist__delete(evlist);
+out:
+	return (err < 0 || errs > 0) ? -1 : 0;
+}
diff --git a/tools/perf/tests/perf-targz-src-pkg b/tools/perf/tests/perf-targz-src-pkg
new file mode 100755
index 000000000..238aa3927
--- /dev/null
+++ b/tools/perf/tests/perf-targz-src-pkg
@@ -0,0 +1,21 @@
+#!/bin/sh
+# Test one of the main kernel Makefile targets to generate a perf sources tarball
+# suitable for build outside the full kernel sources.
+#
+# This is to test that the tools/perf/MANIFEST file lists all the files needed to
+# be in such tarball, which sometimes gets broken when we move files around,
+# like when we made some files that were in tools/perf/ available to other tools/
+# codebases by moving it to tools/include/, etc.
+
+PERF=$1
+cd ${PERF}/../..
+make perf-targz-src-pkg > /dev/null
+TARBALL=$(ls -rt perf-*.tar.gz)
+TMP_DEST=$(mktemp -d)
+tar xf ${TARBALL} -C $TMP_DEST
+rm -f ${TARBALL}
+cd - > /dev/null
+make -C $TMP_DEST/perf*/tools/perf > /dev/null 2>&1
+RC=$?
+rm -rf ${TMP_DEST}
+exit $RC
diff --git a/tools/perf/tests/perf-time-to-tsc.c b/tools/perf/tests/perf-time-to-tsc.c
new file mode 100644
index 000000000..f238442b2
--- /dev/null
+++ b/tools/perf/tests/perf-time-to-tsc.c
@@ -0,0 +1,162 @@
+#include <stdio.h>
+#include <unistd.h>
+#include <linux/types.h>
+#include <sys/prctl.h>
+
+#include "parse-events.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "thread_map.h"
+#include "cpumap.h"
+#include "tsc.h"
+#include "tests.h"
+
+#define CHECK__(x) {				\
+	while ((x) < 0) {			\
+		pr_debug(#x " failed!\n");	\
+		goto out_err;			\
+	}					\
+}
+
+#define CHECK_NOT_NULL__(x) {			\
+	while ((x) == NULL) {			\
+		pr_debug(#x " failed!\n");	\
+		goto out_err;			\
+	}					\
+}
+
+/**
+ * test__perf_time_to_tsc - test converting perf time to TSC.
+ *
+ * This function implements a test that checks that the conversion of perf time
+ * to and from TSC is consistent with the order of events.  If the test passes
+ * %0 is returned, otherwise %-1 is returned.  If TSC conversion is not
+ * supported then then the test passes but " (not supported)" is printed.
+ */
+int test__perf_time_to_tsc(void)
+{
+	struct record_opts opts = {
+		.mmap_pages	     = UINT_MAX,
+		.user_freq	     = UINT_MAX,
+		.user_interval	     = ULLONG_MAX,
+		.freq		     = 4000,
+		.target		     = {
+			.uses_mmap   = true,
+		},
+		.sample_time	     = true,
+	};
+	struct thread_map *threads = NULL;
+	struct cpu_map *cpus = NULL;
+	struct perf_evlist *evlist = NULL;
+	struct perf_evsel *evsel = NULL;
+	int err = -1, ret, i;
+	const char *comm1, *comm2;
+	struct perf_tsc_conversion tc;
+	struct perf_event_mmap_page *pc;
+	union perf_event *event;
+	u64 test_tsc, comm1_tsc, comm2_tsc;
+	u64 test_time, comm1_time = 0, comm2_time = 0;
+
+	threads = thread_map__new(-1, getpid(), UINT_MAX);
+	CHECK_NOT_NULL__(threads);
+
+	cpus = cpu_map__new(NULL);
+	CHECK_NOT_NULL__(cpus);
+
+	evlist = perf_evlist__new();
+	CHECK_NOT_NULL__(evlist);
+
+	perf_evlist__set_maps(evlist, cpus, threads);
+
+	CHECK__(parse_events(evlist, "cycles:u"));
+
+	perf_evlist__config(evlist, &opts);
+
+	evsel = perf_evlist__first(evlist);
+
+	evsel->attr.comm = 1;
+	evsel->attr.disabled = 1;
+	evsel->attr.enable_on_exec = 0;
+
+	CHECK__(perf_evlist__open(evlist));
+
+	CHECK__(perf_evlist__mmap(evlist, UINT_MAX, false));
+
+	pc = evlist->mmap[0].base;
+	ret = perf_read_tsc_conversion(pc, &tc);
+	if (ret) {
+		if (ret == -EOPNOTSUPP) {
+			fprintf(stderr, " (not supported)");
+			return 0;
+		}
+		goto out_err;
+	}
+
+	perf_evlist__enable(evlist);
+
+	comm1 = "Test COMM 1";
+	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm1, 0, 0, 0));
+
+	test_tsc = rdtsc();
+
+	comm2 = "Test COMM 2";
+	CHECK__(prctl(PR_SET_NAME, (unsigned long)comm2, 0, 0, 0));
+
+	perf_evlist__disable(evlist);
+
+	for (i = 0; i < evlist->nr_mmaps; i++) {
+		while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
+			struct perf_sample sample;
+
+			if (event->header.type != PERF_RECORD_COMM ||
+			    (pid_t)event->comm.pid != getpid() ||
+			    (pid_t)event->comm.tid != getpid())
+				goto next_event;
+
+			if (strcmp(event->comm.comm, comm1) == 0) {
+				CHECK__(perf_evsel__parse_sample(evsel, event,
+								 &sample));
+				comm1_time = sample.time;
+			}
+			if (strcmp(event->comm.comm, comm2) == 0) {
+				CHECK__(perf_evsel__parse_sample(evsel, event,
+								 &sample));
+				comm2_time = sample.time;
+			}
+next_event:
+			perf_evlist__mmap_consume(evlist, i);
+		}
+	}
+
+	if (!comm1_time || !comm2_time)
+		goto out_err;
+
+	test_time = tsc_to_perf_time(test_tsc, &tc);
+	comm1_tsc = perf_time_to_tsc(comm1_time, &tc);
+	comm2_tsc = perf_time_to_tsc(comm2_time, &tc);
+
+	pr_debug("1st event perf time %"PRIu64" tsc %"PRIu64"\n",
+		 comm1_time, comm1_tsc);
+	pr_debug("rdtsc          time %"PRIu64" tsc %"PRIu64"\n",
+		 test_time, test_tsc);
+	pr_debug("2nd event perf time %"PRIu64" tsc %"PRIu64"\n",
+		 comm2_time, comm2_tsc);
+
+	if (test_time <= comm1_time ||
+	    test_time >= comm2_time)
+		goto out_err;
+
+	if (test_tsc <= comm1_tsc ||
+	    test_tsc >= comm2_tsc)
+		goto out_err;
+
+	err = 0;
+
+out_err:
+	if (evlist) {
+		perf_evlist__disable(evlist);
+		perf_evlist__delete(evlist);
+	}
+
+	return err;
+}
diff --git a/tools/perf/tests/pmu.c b/tools/perf/tests/pmu.c
new file mode 100644
index 000000000..eeb68bb19
--- /dev/null
+++ b/tools/perf/tests/pmu.c
@@ -0,0 +1,173 @@
+#include "parse-events.h"
+#include "pmu.h"
+#include "util.h"
+#include "tests.h"
+
+/* Simulated format definitions. */
+static struct test_format {
+	const char *name;
+	const char *value;
+} test_formats[] = {
+	{ "krava01", "config:0-1,62-63\n", },
+	{ "krava02", "config:10-17\n", },
+	{ "krava03", "config:5\n", },
+	{ "krava11", "config1:0,2,4,6,8,20-28\n", },
+	{ "krava12", "config1:63\n", },
+	{ "krava13", "config1:45-47\n", },
+	{ "krava21", "config2:0-3,10-13,20-23,30-33,40-43,50-53,60-63\n", },
+	{ "krava22", "config2:8,18,48,58\n", },
+	{ "krava23", "config2:28-29,38\n", },
+};
+
+/* Simulated users input. */
+static struct parse_events_term test_terms[] = {
+	{
+		.config    = (char *) "krava01",
+		.val.num   = 15,
+		.type_val  = PARSE_EVENTS__TERM_TYPE_NUM,
+		.type_term = PARSE_EVENTS__TERM_TYPE_USER,
+	},
+	{
+		.config    = (char *) "krava02",
+		.val.num   = 170,
+		.type_val  = PARSE_EVENTS__TERM_TYPE_NUM,
+		.type_term = PARSE_EVENTS__TERM_TYPE_USER,
+	},
+	{
+		.config    = (char *) "krava03",
+		.val.num   = 1,
+		.type_val  = PARSE_EVENTS__TERM_TYPE_NUM,
+		.type_term = PARSE_EVENTS__TERM_TYPE_USER,
+	},
+	{
+		.config    = (char *) "krava11",
+		.val.num   = 27,
+		.type_val  = PARSE_EVENTS__TERM_TYPE_NUM,
+		.type_term = PARSE_EVENTS__TERM_TYPE_USER,
+	},
+	{
+		.config    = (char *) "krava12",
+		.val.num   = 1,
+		.type_val  = PARSE_EVENTS__TERM_TYPE_NUM,
+		.type_term = PARSE_EVENTS__TERM_TYPE_USER,
+	},
+	{
+		.config    = (char *) "krava13",
+		.val.num   = 2,
+		.type_val  = PARSE_EVENTS__TERM_TYPE_NUM,
+		.type_term = PARSE_EVENTS__TERM_TYPE_USER,
+	},
+	{
+		.config    = (char *) "krava21",
+		.val.num   = 119,
+		.type_val  = PARSE_EVENTS__TERM_TYPE_NUM,
+		.type_term = PARSE_EVENTS__TERM_TYPE_USER,
+	},
+	{
+		.config    = (char *) "krava22",
+		.val.num   = 11,
+		.type_val  = PARSE_EVENTS__TERM_TYPE_NUM,
+		.type_term = PARSE_EVENTS__TERM_TYPE_USER,
+	},
+	{
+		.config    = (char *) "krava23",
+		.val.num   = 2,
+		.type_val  = PARSE_EVENTS__TERM_TYPE_NUM,
+		.type_term = PARSE_EVENTS__TERM_TYPE_USER,
+	},
+};
+
+/*
+ * Prepare format directory data, exported by kernel
+ * at /sys/bus/event_source/devices/<dev>/format.
+ */
+static char *test_format_dir_get(void)
+{
+	static char dir[PATH_MAX];
+	unsigned int i;
+
+	snprintf(dir, PATH_MAX, "/tmp/perf-pmu-test-format-XXXXXX");
+	if (!mkdtemp(dir))
+		return NULL;
+
+	for (i = 0; i < ARRAY_SIZE(test_formats); i++) {
+		static char name[PATH_MAX];
+		struct test_format *format = &test_formats[i];
+		FILE *file;
+
+		snprintf(name, PATH_MAX, "%s/%s", dir, format->name);
+
+		file = fopen(name, "w");
+		if (!file)
+			return NULL;
+
+		if (1 != fwrite(format->value, strlen(format->value), 1, file))
+			break;
+
+		fclose(file);
+	}
+
+	return dir;
+}
+
+/* Cleanup format directory. */
+static int test_format_dir_put(char *dir)
+{
+	char buf[PATH_MAX];
+	snprintf(buf, PATH_MAX, "rm -f %s/*\n", dir);
+	if (system(buf))
+		return -1;
+
+	snprintf(buf, PATH_MAX, "rmdir %s\n", dir);
+	return system(buf);
+}
+
+static struct list_head *test_terms_list(void)
+{
+	static LIST_HEAD(terms);
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(test_terms); i++)
+		list_add_tail(&test_terms[i].list, &terms);
+
+	return &terms;
+}
+
+int test__pmu(void)
+{
+	char *format = test_format_dir_get();
+	LIST_HEAD(formats);
+	struct list_head *terms = test_terms_list();
+	int ret;
+
+	if (!format)
+		return -EINVAL;
+
+	do {
+		struct perf_event_attr attr;
+
+		memset(&attr, 0, sizeof(attr));
+
+		ret = perf_pmu__format_parse(format, &formats);
+		if (ret)
+			break;
+
+		ret = perf_pmu__config_terms(&formats, &attr, terms, false);
+		if (ret)
+			break;
+
+		ret = -EINVAL;
+
+		if (attr.config  != 0xc00000000002a823)
+			break;
+		if (attr.config1 != 0x8000400000000145)
+			break;
+		if (attr.config2 != 0x0400000020041d07)
+			break;
+
+		ret = 0;
+	} while (0);
+
+	test_format_dir_put(format);
+	return ret;
+}
diff --git a/tools/perf/tests/python-use.c b/tools/perf/tests/python-use.c
new file mode 100644
index 000000000..7760277c6
--- /dev/null
+++ b/tools/perf/tests/python-use.c
@@ -0,0 +1,23 @@
+/*
+ * Just test if we can load the python binding.
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include "tests.h"
+
+extern int verbose;
+
+int test__python_use(void)
+{
+	char *cmd;
+	int ret;
+
+	if (asprintf(&cmd, "echo \"import sys ; sys.path.append('%s'); import perf\" | %s %s",
+		     PYTHONPATH, PYTHON, verbose ? "" : "2> /dev/null") < 0)
+		return -1;
+
+	ret = system(cmd) ? -1 : 0;
+	free(cmd);
+	return ret;
+}
diff --git a/tools/perf/tests/rdpmc.c b/tools/perf/tests/rdpmc.c
new file mode 100644
index 000000000..d31f2c4d9
--- /dev/null
+++ b/tools/perf/tests/rdpmc.c
@@ -0,0 +1,177 @@
+#include <unistd.h>
+#include <stdlib.h>
+#include <signal.h>
+#include <sys/mman.h>
+#include <linux/types.h>
+#include "perf.h"
+#include "debug.h"
+#include "tests.h"
+#include "cloexec.h"
+
+#if defined(__x86_64__) || defined(__i386__)
+
+static u64 rdpmc(unsigned int counter)
+{
+	unsigned int low, high;
+
+	asm volatile("rdpmc" : "=a" (low), "=d" (high) : "c" (counter));
+
+	return low | ((u64)high) << 32;
+}
+
+static u64 rdtsc(void)
+{
+	unsigned int low, high;
+
+	asm volatile("rdtsc" : "=a" (low), "=d" (high));
+
+	return low | ((u64)high) << 32;
+}
+
+static u64 mmap_read_self(void *addr)
+{
+	struct perf_event_mmap_page *pc = addr;
+	u32 seq, idx, time_mult = 0, time_shift = 0;
+	u64 count, cyc = 0, time_offset = 0, enabled, running, delta;
+
+	do {
+		seq = pc->lock;
+		barrier();
+
+		enabled = pc->time_enabled;
+		running = pc->time_running;
+
+		if (enabled != running) {
+			cyc = rdtsc();
+			time_mult = pc->time_mult;
+			time_shift = pc->time_shift;
+			time_offset = pc->time_offset;
+		}
+
+		idx = pc->index;
+		count = pc->offset;
+		if (idx)
+			count += rdpmc(idx - 1);
+
+		barrier();
+	} while (pc->lock != seq);
+
+	if (enabled != running) {
+		u64 quot, rem;
+
+		quot = (cyc >> time_shift);
+		rem = cyc & ((1 << time_shift) - 1);
+		delta = time_offset + quot * time_mult +
+			((rem * time_mult) >> time_shift);
+
+		enabled += delta;
+		if (idx)
+			running += delta;
+
+		quot = count / running;
+		rem = count % running;
+		count = quot * enabled + (rem * enabled) / running;
+	}
+
+	return count;
+}
+
+/*
+ * If the RDPMC instruction faults then signal this back to the test parent task:
+ */
+static void segfault_handler(int sig __maybe_unused,
+			     siginfo_t *info __maybe_unused,
+			     void *uc __maybe_unused)
+{
+	exit(-1);
+}
+
+static int __test__rdpmc(void)
+{
+	volatile int tmp = 0;
+	u64 i, loops = 1000;
+	int n;
+	int fd;
+	void *addr;
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_HARDWARE,
+		.config = PERF_COUNT_HW_INSTRUCTIONS,
+		.exclude_kernel = 1,
+	};
+	u64 delta_sum = 0;
+        struct sigaction sa;
+	char sbuf[STRERR_BUFSIZE];
+
+	sigfillset(&sa.sa_mask);
+	sa.sa_sigaction = segfault_handler;
+	sigaction(SIGSEGV, &sa, NULL);
+
+	fd = sys_perf_event_open(&attr, 0, -1, -1,
+				 perf_event_open_cloexec_flag());
+	if (fd < 0) {
+		pr_err("Error: sys_perf_event_open() syscall returned "
+		       "with %d (%s)\n", fd,
+		       strerror_r(errno, sbuf, sizeof(sbuf)));
+		return -1;
+	}
+
+	addr = mmap(NULL, page_size, PROT_READ, MAP_SHARED, fd, 0);
+	if (addr == (void *)(-1)) {
+		pr_err("Error: mmap() syscall returned with (%s)\n",
+		       strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_close;
+	}
+
+	for (n = 0; n < 6; n++) {
+		u64 stamp, now, delta;
+
+		stamp = mmap_read_self(addr);
+
+		for (i = 0; i < loops; i++)
+			tmp++;
+
+		now = mmap_read_self(addr);
+		loops *= 10;
+
+		delta = now - stamp;
+		pr_debug("%14d: %14Lu\n", n, (long long)delta);
+
+		delta_sum += delta;
+	}
+
+	munmap(addr, page_size);
+	pr_debug("   ");
+out_close:
+	close(fd);
+
+	if (!delta_sum)
+		return -1;
+
+	return 0;
+}
+
+int test__rdpmc(void)
+{
+	int status = 0;
+	int wret = 0;
+	int ret;
+	int pid;
+
+	pid = fork();
+	if (pid < 0)
+		return -1;
+
+	if (!pid) {
+		ret = __test__rdpmc();
+
+		exit(ret);
+	}
+
+	wret = waitpid(pid, &status, 0);
+	if (wret < 0 || status)
+		return -1;
+
+	return 0;
+}
+
+#endif
diff --git a/tools/perf/tests/sample-parsing.c b/tools/perf/tests/sample-parsing.c
new file mode 100644
index 000000000..30c02181e
--- /dev/null
+++ b/tools/perf/tests/sample-parsing.c
@@ -0,0 +1,346 @@
+#include <stdbool.h>
+#include <linux/types.h>
+
+#include "util.h"
+#include "event.h"
+#include "evsel.h"
+#include "debug.h"
+
+#include "tests.h"
+
+#define COMP(m) do {					\
+	if (s1->m != s2->m) {				\
+		pr_debug("Samples differ at '"#m"'\n");	\
+		return false;				\
+	}						\
+} while (0)
+
+#define MCOMP(m) do {					\
+	if (memcmp(&s1->m, &s2->m, sizeof(s1->m))) {	\
+		pr_debug("Samples differ at '"#m"'\n");	\
+		return false;				\
+	}						\
+} while (0)
+
+static bool samples_same(const struct perf_sample *s1,
+			 const struct perf_sample *s2,
+			 u64 type, u64 read_format)
+{
+	size_t i;
+
+	if (type & PERF_SAMPLE_IDENTIFIER)
+		COMP(id);
+
+	if (type & PERF_SAMPLE_IP)
+		COMP(ip);
+
+	if (type & PERF_SAMPLE_TID) {
+		COMP(pid);
+		COMP(tid);
+	}
+
+	if (type & PERF_SAMPLE_TIME)
+		COMP(time);
+
+	if (type & PERF_SAMPLE_ADDR)
+		COMP(addr);
+
+	if (type & PERF_SAMPLE_ID)
+		COMP(id);
+
+	if (type & PERF_SAMPLE_STREAM_ID)
+		COMP(stream_id);
+
+	if (type & PERF_SAMPLE_CPU)
+		COMP(cpu);
+
+	if (type & PERF_SAMPLE_PERIOD)
+		COMP(period);
+
+	if (type & PERF_SAMPLE_READ) {
+		if (read_format & PERF_FORMAT_GROUP)
+			COMP(read.group.nr);
+		else
+			COMP(read.one.value);
+		if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+			COMP(read.time_enabled);
+		if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+			COMP(read.time_running);
+		/* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */
+		if (read_format & PERF_FORMAT_GROUP) {
+			for (i = 0; i < s1->read.group.nr; i++)
+				MCOMP(read.group.values[i]);
+		} else {
+			COMP(read.one.id);
+		}
+	}
+
+	if (type & PERF_SAMPLE_CALLCHAIN) {
+		COMP(callchain->nr);
+		for (i = 0; i < s1->callchain->nr; i++)
+			COMP(callchain->ips[i]);
+	}
+
+	if (type & PERF_SAMPLE_RAW) {
+		COMP(raw_size);
+		if (memcmp(s1->raw_data, s2->raw_data, s1->raw_size)) {
+			pr_debug("Samples differ at 'raw_data'\n");
+			return false;
+		}
+	}
+
+	if (type & PERF_SAMPLE_BRANCH_STACK) {
+		COMP(branch_stack->nr);
+		for (i = 0; i < s1->branch_stack->nr; i++)
+			MCOMP(branch_stack->entries[i]);
+	}
+
+	if (type & PERF_SAMPLE_REGS_USER) {
+		size_t sz = hweight_long(s1->user_regs.mask) * sizeof(u64);
+
+		COMP(user_regs.mask);
+		COMP(user_regs.abi);
+		if (s1->user_regs.abi &&
+		    (!s1->user_regs.regs || !s2->user_regs.regs ||
+		     memcmp(s1->user_regs.regs, s2->user_regs.regs, sz))) {
+			pr_debug("Samples differ at 'user_regs'\n");
+			return false;
+		}
+	}
+
+	if (type & PERF_SAMPLE_STACK_USER) {
+		COMP(user_stack.size);
+		if (memcmp(s1->user_stack.data, s2->user_stack.data,
+			   s1->user_stack.size)) {
+			pr_debug("Samples differ at 'user_stack'\n");
+			return false;
+		}
+	}
+
+	if (type & PERF_SAMPLE_WEIGHT)
+		COMP(weight);
+
+	if (type & PERF_SAMPLE_DATA_SRC)
+		COMP(data_src);
+
+	if (type & PERF_SAMPLE_TRANSACTION)
+		COMP(transaction);
+
+	if (type & PERF_SAMPLE_REGS_INTR) {
+		size_t sz = hweight_long(s1->intr_regs.mask) * sizeof(u64);
+
+		COMP(intr_regs.mask);
+		COMP(intr_regs.abi);
+		if (s1->intr_regs.abi &&
+		    (!s1->intr_regs.regs || !s2->intr_regs.regs ||
+		     memcmp(s1->intr_regs.regs, s2->intr_regs.regs, sz))) {
+			pr_debug("Samples differ at 'intr_regs'\n");
+			return false;
+		}
+	}
+
+	return true;
+}
+
+static int do_test(u64 sample_type, u64 sample_regs, u64 read_format)
+{
+	struct perf_evsel evsel = {
+		.needs_swap = false,
+		.attr = {
+			.sample_type = sample_type,
+			.read_format = read_format,
+		},
+	};
+	union perf_event *event;
+	union {
+		struct ip_callchain callchain;
+		u64 data[64];
+	} callchain = {
+		/* 3 ips */
+		.data = {3, 201, 202, 203},
+	};
+	union {
+		struct branch_stack branch_stack;
+		u64 data[64];
+	} branch_stack = {
+		/* 1 branch_entry */
+		.data = {1, 211, 212, 213},
+	};
+	u64 regs[64];
+	const u64 raw_data[] = {0x123456780a0b0c0dULL, 0x1102030405060708ULL};
+	const u64 data[] = {0x2211443366558877ULL, 0, 0xaabbccddeeff4321ULL};
+	struct perf_sample sample = {
+		.ip		= 101,
+		.pid		= 102,
+		.tid		= 103,
+		.time		= 104,
+		.addr		= 105,
+		.id		= 106,
+		.stream_id	= 107,
+		.period		= 108,
+		.weight		= 109,
+		.cpu		= 110,
+		.raw_size	= sizeof(raw_data),
+		.data_src	= 111,
+		.transaction	= 112,
+		.raw_data	= (void *)raw_data,
+		.callchain	= &callchain.callchain,
+		.branch_stack	= &branch_stack.branch_stack,
+		.user_regs	= {
+			.abi	= PERF_SAMPLE_REGS_ABI_64,
+			.mask	= sample_regs,
+			.regs	= regs,
+		},
+		.user_stack	= {
+			.size	= sizeof(data),
+			.data	= (void *)data,
+		},
+		.read		= {
+			.time_enabled = 0x030a59d664fca7deULL,
+			.time_running = 0x011b6ae553eb98edULL,
+		},
+		.intr_regs	= {
+			.abi	= PERF_SAMPLE_REGS_ABI_64,
+			.mask	= sample_regs,
+			.regs	= regs,
+		},
+	};
+	struct sample_read_value values[] = {{1, 5}, {9, 3}, {2, 7}, {6, 4},};
+	struct perf_sample sample_out;
+	size_t i, sz, bufsz;
+	int err, ret = -1;
+
+	if (sample_type & PERF_SAMPLE_REGS_USER)
+		evsel.attr.sample_regs_user = sample_regs;
+
+	if (sample_type & PERF_SAMPLE_REGS_INTR)
+		evsel.attr.sample_regs_intr = sample_regs;
+
+	for (i = 0; i < sizeof(regs); i++)
+		*(i + (u8 *)regs) = i & 0xfe;
+
+	if (read_format & PERF_FORMAT_GROUP) {
+		sample.read.group.nr     = 4;
+		sample.read.group.values = values;
+	} else {
+		sample.read.one.value = 0x08789faeb786aa87ULL;
+		sample.read.one.id    = 99;
+	}
+
+	sz = perf_event__sample_event_size(&sample, sample_type, read_format);
+	bufsz = sz + 4096; /* Add a bit for overrun checking */
+	event = malloc(bufsz);
+	if (!event) {
+		pr_debug("malloc failed\n");
+		return -1;
+	}
+
+	memset(event, 0xff, bufsz);
+	event->header.type = PERF_RECORD_SAMPLE;
+	event->header.misc = 0;
+	event->header.size = sz;
+
+	err = perf_event__synthesize_sample(event, sample_type, read_format,
+					    &sample, false);
+	if (err) {
+		pr_debug("%s failed for sample_type %#"PRIx64", error %d\n",
+			 "perf_event__synthesize_sample", sample_type, err);
+		goto out_free;
+	}
+
+	/* The data does not contain 0xff so we use that to check the size */
+	for (i = bufsz; i > 0; i--) {
+		if (*(i - 1 + (u8 *)event) != 0xff)
+			break;
+	}
+	if (i != sz) {
+		pr_debug("Event size mismatch: actual %zu vs expected %zu\n",
+			 i, sz);
+		goto out_free;
+	}
+
+	evsel.sample_size = __perf_evsel__sample_size(sample_type);
+
+	err = perf_evsel__parse_sample(&evsel, event, &sample_out);
+	if (err) {
+		pr_debug("%s failed for sample_type %#"PRIx64", error %d\n",
+			 "perf_evsel__parse_sample", sample_type, err);
+		goto out_free;
+	}
+
+	if (!samples_same(&sample, &sample_out, sample_type, read_format)) {
+		pr_debug("parsing failed for sample_type %#"PRIx64"\n",
+			 sample_type);
+		goto out_free;
+	}
+
+	ret = 0;
+out_free:
+	free(event);
+	if (ret && read_format)
+		pr_debug("read_format %#"PRIx64"\n", read_format);
+	return ret;
+}
+
+/**
+ * test__sample_parsing - test sample parsing.
+ *
+ * This function implements a test that synthesizes a sample event, parses it
+ * and then checks that the parsed sample matches the original sample.  The test
+ * checks sample format bits separately and together.  If the test passes %0 is
+ * returned, otherwise %-1 is returned.
+ */
+int test__sample_parsing(void)
+{
+	const u64 rf[] = {4, 5, 6, 7, 12, 13, 14, 15};
+	u64 sample_type;
+	u64 sample_regs;
+	size_t i;
+	int err;
+
+	/*
+	 * Fail the test if it has not been updated when new sample format bits
+	 * were added.  Please actually update the test rather than just change
+	 * the condition below.
+	 */
+	if (PERF_SAMPLE_MAX > PERF_SAMPLE_REGS_INTR << 1) {
+		pr_debug("sample format has changed, some new PERF_SAMPLE_ bit was introduced - test needs updating\n");
+		return -1;
+	}
+
+	/* Test each sample format bit separately */
+	for (sample_type = 1; sample_type != PERF_SAMPLE_MAX;
+	     sample_type <<= 1) {
+		/* Test read_format variations */
+		if (sample_type == PERF_SAMPLE_READ) {
+			for (i = 0; i < ARRAY_SIZE(rf); i++) {
+				err = do_test(sample_type, 0, rf[i]);
+				if (err)
+					return err;
+			}
+			continue;
+		}
+		sample_regs = 0;
+
+		if (sample_type == PERF_SAMPLE_REGS_USER)
+			sample_regs = 0x3fff;
+
+		if (sample_type == PERF_SAMPLE_REGS_INTR)
+			sample_regs = 0xff0fff;
+
+		err = do_test(sample_type, sample_regs, 0);
+		if (err)
+			return err;
+	}
+
+	/* Test all sample format bits together */
+	sample_type = PERF_SAMPLE_MAX - 1;
+	sample_regs = 0x3fff; /* shared yb intr and user regs */
+	for (i = 0; i < ARRAY_SIZE(rf); i++) {
+		err = do_test(sample_type, sample_regs, rf[i]);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
diff --git a/tools/perf/tests/sw-clock.c b/tools/perf/tests/sw-clock.c
new file mode 100644
index 000000000..1aa21c907
--- /dev/null
+++ b/tools/perf/tests/sw-clock.c
@@ -0,0 +1,124 @@
+#include <unistd.h>
+#include <stdlib.h>
+#include <signal.h>
+#include <sys/mman.h>
+
+#include "tests.h"
+#include "util/evsel.h"
+#include "util/evlist.h"
+#include "util/cpumap.h"
+#include "util/thread_map.h"
+
+#define NR_LOOPS  10000000
+
+/*
+ * This test will open software clock events (cpu-clock, task-clock)
+ * then check their frequency -> period conversion has no artifact of
+ * setting period to 1 forcefully.
+ */
+static int __test__sw_clock_freq(enum perf_sw_ids clock_id)
+{
+	int i, err = -1;
+	volatile int tmp = 0;
+	u64 total_periods = 0;
+	int nr_samples = 0;
+	char sbuf[STRERR_BUFSIZE];
+	union perf_event *event;
+	struct perf_evsel *evsel;
+	struct perf_evlist *evlist;
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_SOFTWARE,
+		.config = clock_id,
+		.sample_type = PERF_SAMPLE_PERIOD,
+		.exclude_kernel = 1,
+		.disabled = 1,
+		.freq = 1,
+	};
+
+	attr.sample_freq = 500;
+
+	evlist = perf_evlist__new();
+	if (evlist == NULL) {
+		pr_debug("perf_evlist__new\n");
+		return -1;
+	}
+
+	evsel = perf_evsel__new(&attr);
+	if (evsel == NULL) {
+		pr_debug("perf_evsel__new\n");
+		goto out_delete_evlist;
+	}
+	perf_evlist__add(evlist, evsel);
+
+	evlist->cpus = cpu_map__dummy_new();
+	evlist->threads = thread_map__new_by_tid(getpid());
+	if (!evlist->cpus || !evlist->threads) {
+		err = -ENOMEM;
+		pr_debug("Not enough memory to create thread/cpu maps\n");
+		goto out_delete_evlist;
+	}
+
+	if (perf_evlist__open(evlist)) {
+		const char *knob = "/proc/sys/kernel/perf_event_max_sample_rate";
+
+		err = -errno;
+		pr_debug("Couldn't open evlist: %s\nHint: check %s, using %" PRIu64 " in this test.\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)),
+			 knob, (u64)attr.sample_freq);
+		goto out_delete_evlist;
+	}
+
+	err = perf_evlist__mmap(evlist, 128, true);
+	if (err < 0) {
+		pr_debug("failed to mmap event: %d (%s)\n", errno,
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	perf_evlist__enable(evlist);
+
+	/* collect samples */
+	for (i = 0; i < NR_LOOPS; i++)
+		tmp++;
+
+	perf_evlist__disable(evlist);
+
+	while ((event = perf_evlist__mmap_read(evlist, 0)) != NULL) {
+		struct perf_sample sample;
+
+		if (event->header.type != PERF_RECORD_SAMPLE)
+			goto next_event;
+
+		err = perf_evlist__parse_sample(evlist, event, &sample);
+		if (err < 0) {
+			pr_debug("Error during parse sample\n");
+			goto out_delete_evlist;
+		}
+
+		total_periods += sample.period;
+		nr_samples++;
+next_event:
+		perf_evlist__mmap_consume(evlist, 0);
+	}
+
+	if ((u64) nr_samples == total_periods) {
+		pr_debug("All (%d) samples have period value of 1!\n",
+			 nr_samples);
+		err = -1;
+	}
+
+out_delete_evlist:
+	perf_evlist__delete(evlist);
+	return err;
+}
+
+int test__sw_clock_freq(void)
+{
+	int ret;
+
+	ret = __test__sw_clock_freq(PERF_COUNT_SW_CPU_CLOCK);
+	if (!ret)
+		ret = __test__sw_clock_freq(PERF_COUNT_SW_TASK_CLOCK);
+
+	return ret;
+}
diff --git a/tools/perf/tests/switch-tracking.c b/tools/perf/tests/switch-tracking.c
new file mode 100644
index 000000000..cc68648c7
--- /dev/null
+++ b/tools/perf/tests/switch-tracking.c
@@ -0,0 +1,572 @@
+#include <sys/time.h>
+#include <sys/prctl.h>
+#include <time.h>
+#include <stdlib.h>
+
+#include "parse-events.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "thread_map.h"
+#include "cpumap.h"
+#include "tests.h"
+
+static int spin_sleep(void)
+{
+	struct timeval start, now, diff, maxtime;
+	struct timespec ts;
+	int err, i;
+
+	maxtime.tv_sec = 0;
+	maxtime.tv_usec = 50000;
+
+	err = gettimeofday(&start, NULL);
+	if (err)
+		return err;
+
+	/* Spin for 50ms */
+	while (1) {
+		for (i = 0; i < 1000; i++)
+			barrier();
+
+		err = gettimeofday(&now, NULL);
+		if (err)
+			return err;
+
+		timersub(&now, &start, &diff);
+		if (timercmp(&diff, &maxtime, > /* For checkpatch */))
+			break;
+	}
+
+	ts.tv_nsec = 50 * 1000 * 1000;
+	ts.tv_sec = 0;
+
+	/* Sleep for 50ms */
+	err = nanosleep(&ts, NULL);
+	if (err == EINTR)
+		err = 0;
+
+	return err;
+}
+
+struct switch_tracking {
+	struct perf_evsel *switch_evsel;
+	struct perf_evsel *cycles_evsel;
+	pid_t *tids;
+	int nr_tids;
+	int comm_seen[4];
+	int cycles_before_comm_1;
+	int cycles_between_comm_2_and_comm_3;
+	int cycles_after_comm_4;
+};
+
+static int check_comm(struct switch_tracking *switch_tracking,
+		      union perf_event *event, const char *comm, int nr)
+{
+	if (event->header.type == PERF_RECORD_COMM &&
+	    (pid_t)event->comm.pid == getpid() &&
+	    (pid_t)event->comm.tid == getpid() &&
+	    strcmp(event->comm.comm, comm) == 0) {
+		if (switch_tracking->comm_seen[nr]) {
+			pr_debug("Duplicate comm event\n");
+			return -1;
+		}
+		switch_tracking->comm_seen[nr] = 1;
+		pr_debug3("comm event: %s nr: %d\n", event->comm.comm, nr);
+		return 1;
+	}
+	return 0;
+}
+
+static int check_cpu(struct switch_tracking *switch_tracking, int cpu)
+{
+	int i, nr = cpu + 1;
+
+	if (cpu < 0)
+		return -1;
+
+	if (!switch_tracking->tids) {
+		switch_tracking->tids = calloc(nr, sizeof(pid_t));
+		if (!switch_tracking->tids)
+			return -1;
+		for (i = 0; i < nr; i++)
+			switch_tracking->tids[i] = -1;
+		switch_tracking->nr_tids = nr;
+		return 0;
+	}
+
+	if (cpu >= switch_tracking->nr_tids) {
+		void *addr;
+
+		addr = realloc(switch_tracking->tids, nr * sizeof(pid_t));
+		if (!addr)
+			return -1;
+		switch_tracking->tids = addr;
+		for (i = switch_tracking->nr_tids; i < nr; i++)
+			switch_tracking->tids[i] = -1;
+		switch_tracking->nr_tids = nr;
+		return 0;
+	}
+
+	return 0;
+}
+
+static int process_sample_event(struct perf_evlist *evlist,
+				union perf_event *event,
+				struct switch_tracking *switch_tracking)
+{
+	struct perf_sample sample;
+	struct perf_evsel *evsel;
+	pid_t next_tid, prev_tid;
+	int cpu, err;
+
+	if (perf_evlist__parse_sample(evlist, event, &sample)) {
+		pr_debug("perf_evlist__parse_sample failed\n");
+		return -1;
+	}
+
+	evsel = perf_evlist__id2evsel(evlist, sample.id);
+	if (evsel == switch_tracking->switch_evsel) {
+		next_tid = perf_evsel__intval(evsel, &sample, "next_pid");
+		prev_tid = perf_evsel__intval(evsel, &sample, "prev_pid");
+		cpu = sample.cpu;
+		pr_debug3("sched_switch: cpu: %d prev_tid %d next_tid %d\n",
+			  cpu, prev_tid, next_tid);
+		err = check_cpu(switch_tracking, cpu);
+		if (err)
+			return err;
+		/*
+		 * Check for no missing sched_switch events i.e. that the
+		 * evsel->system_wide flag has worked.
+		 */
+		if (switch_tracking->tids[cpu] != -1 &&
+		    switch_tracking->tids[cpu] != prev_tid) {
+			pr_debug("Missing sched_switch events\n");
+			return -1;
+		}
+		switch_tracking->tids[cpu] = next_tid;
+	}
+
+	if (evsel == switch_tracking->cycles_evsel) {
+		pr_debug3("cycles event\n");
+		if (!switch_tracking->comm_seen[0])
+			switch_tracking->cycles_before_comm_1 = 1;
+		if (switch_tracking->comm_seen[1] &&
+		    !switch_tracking->comm_seen[2])
+			switch_tracking->cycles_between_comm_2_and_comm_3 = 1;
+		if (switch_tracking->comm_seen[3])
+			switch_tracking->cycles_after_comm_4 = 1;
+	}
+
+	return 0;
+}
+
+static int process_event(struct perf_evlist *evlist, union perf_event *event,
+			 struct switch_tracking *switch_tracking)
+{
+	if (event->header.type == PERF_RECORD_SAMPLE)
+		return process_sample_event(evlist, event, switch_tracking);
+
+	if (event->header.type == PERF_RECORD_COMM) {
+		int err, done = 0;
+
+		err = check_comm(switch_tracking, event, "Test COMM 1", 0);
+		if (err < 0)
+			return -1;
+		done += err;
+		err = check_comm(switch_tracking, event, "Test COMM 2", 1);
+		if (err < 0)
+			return -1;
+		done += err;
+		err = check_comm(switch_tracking, event, "Test COMM 3", 2);
+		if (err < 0)
+			return -1;
+		done += err;
+		err = check_comm(switch_tracking, event, "Test COMM 4", 3);
+		if (err < 0)
+			return -1;
+		done += err;
+		if (done != 1) {
+			pr_debug("Unexpected comm event\n");
+			return -1;
+		}
+	}
+
+	return 0;
+}
+
+struct event_node {
+	struct list_head list;
+	union perf_event *event;
+	u64 event_time;
+};
+
+static int add_event(struct perf_evlist *evlist, struct list_head *events,
+		     union perf_event *event)
+{
+	struct perf_sample sample;
+	struct event_node *node;
+
+	node = malloc(sizeof(struct event_node));
+	if (!node) {
+		pr_debug("malloc failed\n");
+		return -1;
+	}
+	node->event = event;
+	list_add(&node->list, events);
+
+	if (perf_evlist__parse_sample(evlist, event, &sample)) {
+		pr_debug("perf_evlist__parse_sample failed\n");
+		return -1;
+	}
+
+	if (!sample.time) {
+		pr_debug("event with no time\n");
+		return -1;
+	}
+
+	node->event_time = sample.time;
+
+	return 0;
+}
+
+static void free_event_nodes(struct list_head *events)
+{
+	struct event_node *node;
+
+	while (!list_empty(events)) {
+		node = list_entry(events->next, struct event_node, list);
+		list_del(&node->list);
+		free(node);
+	}
+}
+
+static int compar(const void *a, const void *b)
+{
+	const struct event_node *nodea = a;
+	const struct event_node *nodeb = b;
+	s64 cmp = nodea->event_time - nodeb->event_time;
+
+	return cmp;
+}
+
+static int process_events(struct perf_evlist *evlist,
+			  struct switch_tracking *switch_tracking)
+{
+	union perf_event *event;
+	unsigned pos, cnt = 0;
+	LIST_HEAD(events);
+	struct event_node *events_array, *node;
+	int i, ret;
+
+	for (i = 0; i < evlist->nr_mmaps; i++) {
+		while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
+			cnt += 1;
+			ret = add_event(evlist, &events, event);
+			perf_evlist__mmap_consume(evlist, i);
+			if (ret < 0)
+				goto out_free_nodes;
+		}
+	}
+
+	events_array = calloc(cnt, sizeof(struct event_node));
+	if (!events_array) {
+		pr_debug("calloc failed\n");
+		ret = -1;
+		goto out_free_nodes;
+	}
+
+	pos = 0;
+	list_for_each_entry(node, &events, list)
+		events_array[pos++] = *node;
+
+	qsort(events_array, cnt, sizeof(struct event_node), compar);
+
+	for (pos = 0; pos < cnt; pos++) {
+		ret = process_event(evlist, events_array[pos].event,
+				    switch_tracking);
+		if (ret < 0)
+			goto out_free;
+	}
+
+	ret = 0;
+out_free:
+	pr_debug("%u events recorded\n", cnt);
+	free(events_array);
+out_free_nodes:
+	free_event_nodes(&events);
+	return ret;
+}
+
+/**
+ * test__switch_tracking - test using sched_switch and tracking events.
+ *
+ * This function implements a test that checks that sched_switch events and
+ * tracking events can be recorded for a workload (current process) using the
+ * evsel->system_wide and evsel->tracking flags (respectively) with other events
+ * sometimes enabled or disabled.
+ */
+int test__switch_tracking(void)
+{
+	const char *sched_switch = "sched:sched_switch";
+	struct switch_tracking switch_tracking = { .tids = NULL, };
+	struct record_opts opts = {
+		.mmap_pages	     = UINT_MAX,
+		.user_freq	     = UINT_MAX,
+		.user_interval	     = ULLONG_MAX,
+		.freq		     = 4000,
+		.target		     = {
+			.uses_mmap   = true,
+		},
+	};
+	struct thread_map *threads = NULL;
+	struct cpu_map *cpus = NULL;
+	struct perf_evlist *evlist = NULL;
+	struct perf_evsel *evsel, *cpu_clocks_evsel, *cycles_evsel;
+	struct perf_evsel *switch_evsel, *tracking_evsel;
+	const char *comm;
+	int err = -1;
+
+	threads = thread_map__new(-1, getpid(), UINT_MAX);
+	if (!threads) {
+		pr_debug("thread_map__new failed!\n");
+		goto out_err;
+	}
+
+	cpus = cpu_map__new(NULL);
+	if (!cpus) {
+		pr_debug("cpu_map__new failed!\n");
+		goto out_err;
+	}
+
+	evlist = perf_evlist__new();
+	if (!evlist) {
+		pr_debug("perf_evlist__new failed!\n");
+		goto out_err;
+	}
+
+	perf_evlist__set_maps(evlist, cpus, threads);
+
+	/* First event */
+	err = parse_events(evlist, "cpu-clock:u");
+	if (err) {
+		pr_debug("Failed to parse event dummy:u\n");
+		goto out_err;
+	}
+
+	cpu_clocks_evsel = perf_evlist__last(evlist);
+
+	/* Second event */
+	err = parse_events(evlist, "cycles:u");
+	if (err) {
+		pr_debug("Failed to parse event cycles:u\n");
+		goto out_err;
+	}
+
+	cycles_evsel = perf_evlist__last(evlist);
+
+	/* Third event */
+	if (!perf_evlist__can_select_event(evlist, sched_switch)) {
+		fprintf(stderr, " (no sched_switch)");
+		err = 0;
+		goto out;
+	}
+
+	err = parse_events(evlist, sched_switch);
+	if (err) {
+		pr_debug("Failed to parse event %s\n", sched_switch);
+		goto out_err;
+	}
+
+	switch_evsel = perf_evlist__last(evlist);
+
+	perf_evsel__set_sample_bit(switch_evsel, CPU);
+	perf_evsel__set_sample_bit(switch_evsel, TIME);
+
+	switch_evsel->system_wide = true;
+	switch_evsel->no_aux_samples = true;
+	switch_evsel->immediate = true;
+
+	/* Test moving an event to the front */
+	if (cycles_evsel == perf_evlist__first(evlist)) {
+		pr_debug("cycles event already at front");
+		goto out_err;
+	}
+	perf_evlist__to_front(evlist, cycles_evsel);
+	if (cycles_evsel != perf_evlist__first(evlist)) {
+		pr_debug("Failed to move cycles event to front");
+		goto out_err;
+	}
+
+	perf_evsel__set_sample_bit(cycles_evsel, CPU);
+	perf_evsel__set_sample_bit(cycles_evsel, TIME);
+
+	/* Fourth event */
+	err = parse_events(evlist, "dummy:u");
+	if (err) {
+		pr_debug("Failed to parse event dummy:u\n");
+		goto out_err;
+	}
+
+	tracking_evsel = perf_evlist__last(evlist);
+
+	perf_evlist__set_tracking_event(evlist, tracking_evsel);
+
+	tracking_evsel->attr.freq = 0;
+	tracking_evsel->attr.sample_period = 1;
+
+	perf_evsel__set_sample_bit(tracking_evsel, TIME);
+
+	/* Config events */
+	perf_evlist__config(evlist, &opts);
+
+	/* Check moved event is still at the front */
+	if (cycles_evsel != perf_evlist__first(evlist)) {
+		pr_debug("Front event no longer at front");
+		goto out_err;
+	}
+
+	/* Check tracking event is tracking */
+	if (!tracking_evsel->attr.mmap || !tracking_evsel->attr.comm) {
+		pr_debug("Tracking event not tracking\n");
+		goto out_err;
+	}
+
+	/* Check non-tracking events are not tracking */
+	evlist__for_each(evlist, evsel) {
+		if (evsel != tracking_evsel) {
+			if (evsel->attr.mmap || evsel->attr.comm) {
+				pr_debug("Non-tracking event is tracking\n");
+				goto out_err;
+			}
+		}
+	}
+
+	if (perf_evlist__open(evlist) < 0) {
+		fprintf(stderr, " (not supported)");
+		err = 0;
+		goto out;
+	}
+
+	err = perf_evlist__mmap(evlist, UINT_MAX, false);
+	if (err) {
+		pr_debug("perf_evlist__mmap failed!\n");
+		goto out_err;
+	}
+
+	perf_evlist__enable(evlist);
+
+	err = perf_evlist__disable_event(evlist, cpu_clocks_evsel);
+	if (err) {
+		pr_debug("perf_evlist__disable_event failed!\n");
+		goto out_err;
+	}
+
+	err = spin_sleep();
+	if (err) {
+		pr_debug("spin_sleep failed!\n");
+		goto out_err;
+	}
+
+	comm = "Test COMM 1";
+	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
+	if (err) {
+		pr_debug("PR_SET_NAME failed!\n");
+		goto out_err;
+	}
+
+	err = perf_evlist__disable_event(evlist, cycles_evsel);
+	if (err) {
+		pr_debug("perf_evlist__disable_event failed!\n");
+		goto out_err;
+	}
+
+	comm = "Test COMM 2";
+	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
+	if (err) {
+		pr_debug("PR_SET_NAME failed!\n");
+		goto out_err;
+	}
+
+	err = spin_sleep();
+	if (err) {
+		pr_debug("spin_sleep failed!\n");
+		goto out_err;
+	}
+
+	comm = "Test COMM 3";
+	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
+	if (err) {
+		pr_debug("PR_SET_NAME failed!\n");
+		goto out_err;
+	}
+
+	err = perf_evlist__enable_event(evlist, cycles_evsel);
+	if (err) {
+		pr_debug("perf_evlist__disable_event failed!\n");
+		goto out_err;
+	}
+
+	comm = "Test COMM 4";
+	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
+	if (err) {
+		pr_debug("PR_SET_NAME failed!\n");
+		goto out_err;
+	}
+
+	err = spin_sleep();
+	if (err) {
+		pr_debug("spin_sleep failed!\n");
+		goto out_err;
+	}
+
+	perf_evlist__disable(evlist);
+
+	switch_tracking.switch_evsel = switch_evsel;
+	switch_tracking.cycles_evsel = cycles_evsel;
+
+	err = process_events(evlist, &switch_tracking);
+
+	zfree(&switch_tracking.tids);
+
+	if (err)
+		goto out_err;
+
+	/* Check all 4 comm events were seen i.e. that evsel->tracking works */
+	if (!switch_tracking.comm_seen[0] || !switch_tracking.comm_seen[1] ||
+	    !switch_tracking.comm_seen[2] || !switch_tracking.comm_seen[3]) {
+		pr_debug("Missing comm events\n");
+		goto out_err;
+	}
+
+	/* Check cycles event got enabled */
+	if (!switch_tracking.cycles_before_comm_1) {
+		pr_debug("Missing cycles events\n");
+		goto out_err;
+	}
+
+	/* Check cycles event got disabled */
+	if (switch_tracking.cycles_between_comm_2_and_comm_3) {
+		pr_debug("cycles events even though event was disabled\n");
+		goto out_err;
+	}
+
+	/* Check cycles event got enabled again */
+	if (!switch_tracking.cycles_after_comm_4) {
+		pr_debug("Missing cycles events\n");
+		goto out_err;
+	}
+out:
+	if (evlist) {
+		perf_evlist__disable(evlist);
+		perf_evlist__delete(evlist);
+	} else {
+		cpu_map__delete(cpus);
+		thread_map__delete(threads);
+	}
+
+	return err;
+
+out_err:
+	err = -1;
+	goto out;
+}
diff --git a/tools/perf/tests/task-exit.c b/tools/perf/tests/task-exit.c
new file mode 100644
index 000000000..3a8fedef8
--- /dev/null
+++ b/tools/perf/tests/task-exit.c
@@ -0,0 +1,120 @@
+#include "evlist.h"
+#include "evsel.h"
+#include "thread_map.h"
+#include "cpumap.h"
+#include "tests.h"
+
+#include <signal.h>
+
+static int exited;
+static int nr_exit;
+
+static void sig_handler(int sig __maybe_unused)
+{
+	exited = 1;
+}
+
+/*
+ * perf_evlist__prepare_workload will send a SIGUSR1 if the fork fails, since
+ * we asked by setting its exec_error to this handler.
+ */
+static void workload_exec_failed_signal(int signo __maybe_unused,
+					siginfo_t *info __maybe_unused,
+					void *ucontext __maybe_unused)
+{
+	exited	= 1;
+	nr_exit = -1;
+}
+
+/*
+ * This test will start a workload that does nothing then it checks
+ * if the number of exit event reported by the kernel is 1 or not
+ * in order to check the kernel returns correct number of event.
+ */
+int test__task_exit(void)
+{
+	int err = -1;
+	union perf_event *event;
+	struct perf_evsel *evsel;
+	struct perf_evlist *evlist;
+	struct target target = {
+		.uid		= UINT_MAX,
+		.uses_mmap	= true,
+	};
+	const char *argv[] = { "true", NULL };
+	char sbuf[STRERR_BUFSIZE];
+
+	signal(SIGCHLD, sig_handler);
+
+	evlist = perf_evlist__new_default();
+	if (evlist == NULL) {
+		pr_debug("perf_evlist__new_default\n");
+		return -1;
+	}
+
+	/*
+	 * Create maps of threads and cpus to monitor. In this case
+	 * we start with all threads and cpus (-1, -1) but then in
+	 * perf_evlist__prepare_workload we'll fill in the only thread
+	 * we're monitoring, the one forked there.
+	 */
+	evlist->cpus = cpu_map__dummy_new();
+	evlist->threads = thread_map__new_by_tid(-1);
+	if (!evlist->cpus || !evlist->threads) {
+		err = -ENOMEM;
+		pr_debug("Not enough memory to create thread/cpu maps\n");
+		goto out_delete_evlist;
+	}
+
+	err = perf_evlist__prepare_workload(evlist, &target, argv, false,
+					    workload_exec_failed_signal);
+	if (err < 0) {
+		pr_debug("Couldn't run the workload!\n");
+		goto out_delete_evlist;
+	}
+
+	evsel = perf_evlist__first(evlist);
+	evsel->attr.task = 1;
+	evsel->attr.sample_freq = 0;
+	evsel->attr.inherit = 0;
+	evsel->attr.watermark = 0;
+	evsel->attr.wakeup_events = 1;
+	evsel->attr.exclude_kernel = 1;
+
+	err = perf_evlist__open(evlist);
+	if (err < 0) {
+		pr_debug("Couldn't open the evlist: %s\n",
+			 strerror_r(-err, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	if (perf_evlist__mmap(evlist, 128, true) < 0) {
+		pr_debug("failed to mmap events: %d (%s)\n", errno,
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		goto out_delete_evlist;
+	}
+
+	perf_evlist__start_workload(evlist);
+
+retry:
+	while ((event = perf_evlist__mmap_read(evlist, 0)) != NULL) {
+		if (event->header.type == PERF_RECORD_EXIT)
+			nr_exit++;
+
+		perf_evlist__mmap_consume(evlist, 0);
+	}
+
+	if (!exited || !nr_exit) {
+		perf_evlist__poll(evlist, -1);
+		goto retry;
+	}
+
+	if (nr_exit != 1) {
+		pr_debug("received %d EXIT records\n", nr_exit);
+		err = -1;
+	}
+
+out_delete_evlist:
+	perf_evlist__delete(evlist);
+	return err;
+}
diff --git a/tools/perf/tests/tests.h b/tools/perf/tests/tests.h
new file mode 100644
index 000000000..52758a33f
--- /dev/null
+++ b/tools/perf/tests/tests.h
@@ -0,0 +1,64 @@
+#ifndef TESTS_H
+#define TESTS_H
+
+#define TEST_ASSERT_VAL(text, cond)					 \
+do {									 \
+	if (!(cond)) {							 \
+		pr_debug("FAILED %s:%d %s\n", __FILE__, __LINE__, text); \
+		return -1;						 \
+	}								 \
+} while (0)
+
+enum {
+	TEST_OK   =  0,
+	TEST_FAIL = -1,
+	TEST_SKIP = -2,
+};
+
+/* Tests */
+int test__vmlinux_matches_kallsyms(void);
+int test__open_syscall_event(void);
+int test__open_syscall_event_on_all_cpus(void);
+int test__basic_mmap(void);
+int test__PERF_RECORD(void);
+int test__rdpmc(void);
+int test__perf_evsel__roundtrip_name_test(void);
+int test__perf_evsel__tp_sched_test(void);
+int test__syscall_open_tp_fields(void);
+int test__pmu(void);
+int test__attr(void);
+int test__dso_data(void);
+int test__dso_data_cache(void);
+int test__dso_data_reopen(void);
+int test__parse_events(void);
+int test__hists_link(void);
+int test__python_use(void);
+int test__bp_signal(void);
+int test__bp_signal_overflow(void);
+int test__task_exit(void);
+int test__sw_clock_freq(void);
+int test__perf_time_to_tsc(void);
+int test__code_reading(void);
+int test__sample_parsing(void);
+int test__keep_tracking(void);
+int test__parse_no_sample_id_all(void);
+int test__dwarf_unwind(void);
+int test__hists_filter(void);
+int test__mmap_thread_lookup(void);
+int test__thread_mg_share(void);
+int test__hists_output(void);
+int test__hists_cumulate(void);
+int test__switch_tracking(void);
+int test__fdarray__filter(void);
+int test__fdarray__add(void);
+int test__kmod_path__parse(void);
+
+#if defined(__x86_64__) || defined(__i386__) || defined(__arm__)
+#ifdef HAVE_DWARF_UNWIND_SUPPORT
+struct thread;
+struct perf_sample;
+int test__arch_unwind_sample(struct perf_sample *sample,
+			     struct thread *thread);
+#endif
+#endif
+#endif /* TESTS_H */
diff --git a/tools/perf/tests/thread-mg-share.c b/tools/perf/tests/thread-mg-share.c
new file mode 100644
index 000000000..b028499dd
--- /dev/null
+++ b/tools/perf/tests/thread-mg-share.c
@@ -0,0 +1,91 @@
+#include "tests.h"
+#include "machine.h"
+#include "thread.h"
+#include "map.h"
+#include "debug.h"
+
+int test__thread_mg_share(void)
+{
+	struct machines machines;
+	struct machine *machine;
+
+	/* thread group */
+	struct thread *leader;
+	struct thread *t1, *t2, *t3;
+	struct map_groups *mg;
+
+	/* other process */
+	struct thread *other, *other_leader;
+	struct map_groups *other_mg;
+
+	/*
+	 * This test create 2 processes abstractions (struct thread)
+	 * with several threads and checks they properly share and
+	 * maintain map groups info (struct map_groups).
+	 *
+	 * thread group (pid: 0, tids: 0, 1, 2, 3)
+	 * other  group (pid: 4, tids: 4, 5)
+	*/
+
+	machines__init(&machines);
+	machine = &machines.host;
+
+	/* create process with 4 threads */
+	leader = machine__findnew_thread(machine, 0, 0);
+	t1     = machine__findnew_thread(machine, 0, 1);
+	t2     = machine__findnew_thread(machine, 0, 2);
+	t3     = machine__findnew_thread(machine, 0, 3);
+
+	/* and create 1 separated process, without thread leader */
+	other  = machine__findnew_thread(machine, 4, 5);
+
+	TEST_ASSERT_VAL("failed to create threads",
+			leader && t1 && t2 && t3 && other);
+
+	mg = leader->mg;
+	TEST_ASSERT_VAL("wrong refcnt", mg->refcnt == 4);
+
+	/* test the map groups pointer is shared */
+	TEST_ASSERT_VAL("map groups don't match", mg == t1->mg);
+	TEST_ASSERT_VAL("map groups don't match", mg == t2->mg);
+	TEST_ASSERT_VAL("map groups don't match", mg == t3->mg);
+
+	/*
+	 * Verify the other leader was created by previous call.
+	 * It should have shared map groups with no change in
+	 * refcnt.
+	 */
+	other_leader = machine__find_thread(machine, 4, 4);
+	TEST_ASSERT_VAL("failed to find other leader", other_leader);
+
+	other_mg = other->mg;
+	TEST_ASSERT_VAL("wrong refcnt", other_mg->refcnt == 2);
+
+	TEST_ASSERT_VAL("map groups don't match", other_mg == other_leader->mg);
+
+	/* release thread group */
+	thread__delete(leader);
+	TEST_ASSERT_VAL("wrong refcnt", mg->refcnt == 3);
+
+	thread__delete(t1);
+	TEST_ASSERT_VAL("wrong refcnt", mg->refcnt == 2);
+
+	thread__delete(t2);
+	TEST_ASSERT_VAL("wrong refcnt", mg->refcnt == 1);
+
+	thread__delete(t3);
+
+	/* release other group  */
+	thread__delete(other_leader);
+	TEST_ASSERT_VAL("wrong refcnt", other_mg->refcnt == 1);
+
+	thread__delete(other);
+
+	/*
+	 * Cannot call machine__delete_threads(machine) now,
+	 * because we've already released all the threads.
+	 */
+
+	machines__exit(&machines);
+	return 0;
+}
diff --git a/tools/perf/tests/vmlinux-kallsyms.c b/tools/perf/tests/vmlinux-kallsyms.c
new file mode 100644
index 000000000..3d9088003
--- /dev/null
+++ b/tools/perf/tests/vmlinux-kallsyms.c
@@ -0,0 +1,242 @@
+#include <linux/compiler.h>
+#include <linux/rbtree.h>
+#include <string.h>
+#include "map.h"
+#include "symbol.h"
+#include "util.h"
+#include "tests.h"
+#include "debug.h"
+#include "machine.h"
+
+static int vmlinux_matches_kallsyms_filter(struct map *map __maybe_unused,
+					   struct symbol *sym)
+{
+	bool *visited = symbol__priv(sym);
+	*visited = true;
+	return 0;
+}
+
+#define UM(x) kallsyms_map->unmap_ip(kallsyms_map, (x))
+
+int test__vmlinux_matches_kallsyms(void)
+{
+	int err = -1;
+	struct rb_node *nd;
+	struct symbol *sym;
+	struct map *kallsyms_map, *vmlinux_map;
+	struct machine kallsyms, vmlinux;
+	enum map_type type = MAP__FUNCTION;
+	u64 mem_start, mem_end;
+
+	/*
+	 * Step 1:
+	 *
+	 * Init the machines that will hold kernel, modules obtained from
+	 * both vmlinux + .ko files and from /proc/kallsyms split by modules.
+	 */
+	machine__init(&kallsyms, "", HOST_KERNEL_ID);
+	machine__init(&vmlinux, "", HOST_KERNEL_ID);
+
+	/*
+	 * Step 2:
+	 *
+	 * Create the kernel maps for kallsyms and the DSO where we will then
+	 * load /proc/kallsyms. Also create the modules maps from /proc/modules
+	 * and find the .ko files that match them in /lib/modules/`uname -r`/.
+	 */
+	if (machine__create_kernel_maps(&kallsyms) < 0) {
+		pr_debug("machine__create_kernel_maps ");
+		goto out;
+	}
+
+	/*
+	 * Step 3:
+	 *
+	 * Load and split /proc/kallsyms into multiple maps, one per module.
+	 */
+	if (machine__load_kallsyms(&kallsyms, "/proc/kallsyms", type, NULL) <= 0) {
+		pr_debug("dso__load_kallsyms ");
+		goto out;
+	}
+
+	/*
+	 * Step 4:
+	 *
+	 * kallsyms will be internally on demand sorted by name so that we can
+	 * find the reference relocation * symbol, i.e. the symbol we will use
+	 * to see if the running kernel was relocated by checking if it has the
+	 * same value in the vmlinux file we load.
+	 */
+	kallsyms_map = machine__kernel_map(&kallsyms, type);
+
+	/*
+	 * Step 5:
+	 *
+	 * Now repeat step 2, this time for the vmlinux file we'll auto-locate.
+	 */
+	if (machine__create_kernel_maps(&vmlinux) < 0) {
+		pr_debug("machine__create_kernel_maps ");
+		goto out;
+	}
+
+	vmlinux_map = machine__kernel_map(&vmlinux, type);
+
+	/*
+	 * Step 6:
+	 *
+	 * Locate a vmlinux file in the vmlinux path that has a buildid that
+	 * matches the one of the running kernel.
+	 *
+	 * While doing that look if we find the ref reloc symbol, if we find it
+	 * we'll have its ref_reloc_symbol.unrelocated_addr and then
+	 * maps__reloc_vmlinux will notice and set proper ->[un]map_ip routines
+	 * to fixup the symbols.
+	 */
+	if (machine__load_vmlinux_path(&vmlinux, type,
+				       vmlinux_matches_kallsyms_filter) <= 0) {
+		pr_debug("Couldn't find a vmlinux that matches the kernel running on this machine, skipping test\n");
+		err = TEST_SKIP;
+		goto out;
+	}
+
+	err = 0;
+	/*
+	 * Step 7:
+	 *
+	 * Now look at the symbols in the vmlinux DSO and check if we find all of them
+	 * in the kallsyms dso. For the ones that are in both, check its names and
+	 * end addresses too.
+	 */
+	for (nd = rb_first(&vmlinux_map->dso->symbols[type]); nd; nd = rb_next(nd)) {
+		struct symbol *pair, *first_pair;
+		bool backwards = true;
+
+		sym  = rb_entry(nd, struct symbol, rb_node);
+
+		if (sym->start == sym->end)
+			continue;
+
+		mem_start = vmlinux_map->unmap_ip(vmlinux_map, sym->start);
+		mem_end = vmlinux_map->unmap_ip(vmlinux_map, sym->end);
+
+		first_pair = machine__find_kernel_symbol(&kallsyms, type,
+							 mem_start, NULL, NULL);
+		pair = first_pair;
+
+		if (pair && UM(pair->start) == mem_start) {
+next_pair:
+			if (strcmp(sym->name, pair->name) == 0) {
+				/*
+				 * kallsyms don't have the symbol end, so we
+				 * set that by using the next symbol start - 1,
+				 * in some cases we get this up to a page
+				 * wrong, trace_kmalloc when I was developing
+				 * this code was one such example, 2106 bytes
+				 * off the real size. More than that and we
+				 * _really_ have a problem.
+				 */
+				s64 skew = mem_end - UM(pair->end);
+				if (llabs(skew) >= page_size)
+					pr_debug("%#" PRIx64 ": diff end addr for %s v: %#" PRIx64 " k: %#" PRIx64 "\n",
+						 mem_start, sym->name, mem_end,
+						 UM(pair->end));
+
+				/*
+				 * Do not count this as a failure, because we
+				 * could really find a case where it's not
+				 * possible to get proper function end from
+				 * kallsyms.
+				 */
+				continue;
+
+			} else {
+				struct rb_node *nnd;
+detour:
+				nnd = backwards ? rb_prev(&pair->rb_node) :
+						  rb_next(&pair->rb_node);
+				if (nnd) {
+					struct symbol *next = rb_entry(nnd, struct symbol, rb_node);
+
+					if (UM(next->start) == mem_start) {
+						pair = next;
+						goto next_pair;
+					}
+				}
+
+				if (backwards) {
+					backwards = false;
+					pair = first_pair;
+					goto detour;
+				}
+
+				pr_debug("%#" PRIx64 ": diff name v: %s k: %s\n",
+					 mem_start, sym->name, pair->name);
+			}
+		} else
+			pr_debug("%#" PRIx64 ": %s not on kallsyms\n",
+				 mem_start, sym->name);
+
+		err = -1;
+	}
+
+	if (!verbose)
+		goto out;
+
+	pr_info("Maps only in vmlinux:\n");
+
+	for (nd = rb_first(&vmlinux.kmaps.maps[type]); nd; nd = rb_next(nd)) {
+		struct map *pos = rb_entry(nd, struct map, rb_node), *pair;
+		/*
+		 * If it is the kernel, kallsyms is always "[kernel.kallsyms]", while
+		 * the kernel will have the path for the vmlinux file being used,
+		 * so use the short name, less descriptive but the same ("[kernel]" in
+		 * both cases.
+		 */
+		pair = map_groups__find_by_name(&kallsyms.kmaps, type,
+						(pos->dso->kernel ?
+							pos->dso->short_name :
+							pos->dso->name));
+		if (pair)
+			pair->priv = 1;
+		else
+			map__fprintf(pos, stderr);
+	}
+
+	pr_info("Maps in vmlinux with a different name in kallsyms:\n");
+
+	for (nd = rb_first(&vmlinux.kmaps.maps[type]); nd; nd = rb_next(nd)) {
+		struct map *pos = rb_entry(nd, struct map, rb_node), *pair;
+
+		mem_start = vmlinux_map->unmap_ip(vmlinux_map, pos->start);
+		mem_end = vmlinux_map->unmap_ip(vmlinux_map, pos->end);
+
+		pair = map_groups__find(&kallsyms.kmaps, type, mem_start);
+		if (pair == NULL || pair->priv)
+			continue;
+
+		if (pair->start == mem_start) {
+			pair->priv = 1;
+			pr_info(" %" PRIx64 "-%" PRIx64 " %" PRIx64 " %s in kallsyms as",
+				pos->start, pos->end, pos->pgoff, pos->dso->name);
+			if (mem_end != pair->end)
+				pr_info(":\n*%" PRIx64 "-%" PRIx64 " %" PRIx64,
+					pair->start, pair->end, pair->pgoff);
+			pr_info(" %s\n", pair->dso->name);
+			pair->priv = 1;
+		}
+	}
+
+	pr_info("Maps only in kallsyms:\n");
+
+	for (nd = rb_first(&kallsyms.kmaps.maps[type]);
+	     nd; nd = rb_next(nd)) {
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+
+		if (!pos->priv)
+			map__fprintf(pos, stderr);
+	}
+out:
+	machine__exit(&kallsyms);
+	machine__exit(&vmlinux);
+	return err;
+}
diff --git a/tools/perf/ui/Build b/tools/perf/ui/Build
new file mode 100644
index 000000000..0a73538c0
--- /dev/null
+++ b/tools/perf/ui/Build
@@ -0,0 +1,14 @@
+libperf-y += setup.o
+libperf-y += helpline.o
+libperf-y += progress.o
+libperf-y += util.o
+libperf-y += hist.o
+libperf-y += stdio/hist.o
+
+CFLAGS_setup.o += -DLIBDIR="BUILD_STR($(LIBDIR))"
+
+libperf-$(CONFIG_SLANG) += browser.o
+libperf-$(CONFIG_SLANG) += browsers/
+libperf-$(CONFIG_SLANG) += tui/
+
+CFLAGS_browser.o += -DENABLE_SLFUTURE_CONST
diff --git a/tools/perf/ui/browser.c b/tools/perf/ui/browser.c
new file mode 100644
index 000000000..6680fa5cb
--- /dev/null
+++ b/tools/perf/ui/browser.c
@@ -0,0 +1,722 @@
+#include "../util.h"
+#include "../cache.h"
+#include "../../perf.h"
+#include "libslang.h"
+#include "ui.h"
+#include "util.h"
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <stdlib.h>
+#include <sys/ttydefaults.h>
+#include "browser.h"
+#include "helpline.h"
+#include "keysyms.h"
+#include "../color.h"
+
+static int ui_browser__percent_color(struct ui_browser *browser,
+				     double percent, bool current)
+{
+	if (current && (!browser->use_navkeypressed || browser->navkeypressed))
+		return HE_COLORSET_SELECTED;
+	if (percent >= MIN_RED)
+		return HE_COLORSET_TOP;
+	if (percent >= MIN_GREEN)
+		return HE_COLORSET_MEDIUM;
+	return HE_COLORSET_NORMAL;
+}
+
+int ui_browser__set_color(struct ui_browser *browser, int color)
+{
+	int ret = browser->current_color;
+	browser->current_color = color;
+	SLsmg_set_color(color);
+	return ret;
+}
+
+void ui_browser__set_percent_color(struct ui_browser *browser,
+				   double percent, bool current)
+{
+	 int color = ui_browser__percent_color(browser, percent, current);
+	 ui_browser__set_color(browser, color);
+}
+
+void ui_browser__gotorc(struct ui_browser *browser, int y, int x)
+{
+	SLsmg_gotorc(browser->y + y, browser->x + x);
+}
+
+static struct list_head *
+ui_browser__list_head_filter_entries(struct ui_browser *browser,
+				     struct list_head *pos)
+{
+	do {
+		if (!browser->filter || !browser->filter(browser, pos))
+			return pos;
+		pos = pos->next;
+	} while (pos != browser->entries);
+
+	return NULL;
+}
+
+static struct list_head *
+ui_browser__list_head_filter_prev_entries(struct ui_browser *browser,
+					  struct list_head *pos)
+{
+	do {
+		if (!browser->filter || !browser->filter(browser, pos))
+			return pos;
+		pos = pos->prev;
+	} while (pos != browser->entries);
+
+	return NULL;
+}
+
+void ui_browser__list_head_seek(struct ui_browser *browser, off_t offset, int whence)
+{
+	struct list_head *head = browser->entries;
+	struct list_head *pos;
+
+	if (browser->nr_entries == 0)
+		return;
+
+	switch (whence) {
+	case SEEK_SET:
+		pos = ui_browser__list_head_filter_entries(browser, head->next);
+		break;
+	case SEEK_CUR:
+		pos = browser->top;
+		break;
+	case SEEK_END:
+		pos = ui_browser__list_head_filter_prev_entries(browser, head->prev);
+		break;
+	default:
+		return;
+	}
+
+	assert(pos != NULL);
+
+	if (offset > 0) {
+		while (offset-- != 0)
+			pos = ui_browser__list_head_filter_entries(browser, pos->next);
+	} else {
+		while (offset++ != 0)
+			pos = ui_browser__list_head_filter_prev_entries(browser, pos->prev);
+	}
+
+	browser->top = pos;
+}
+
+void ui_browser__rb_tree_seek(struct ui_browser *browser, off_t offset, int whence)
+{
+	struct rb_root *root = browser->entries;
+	struct rb_node *nd;
+
+	switch (whence) {
+	case SEEK_SET:
+		nd = rb_first(root);
+		break;
+	case SEEK_CUR:
+		nd = browser->top;
+		break;
+	case SEEK_END:
+		nd = rb_last(root);
+		break;
+	default:
+		return;
+	}
+
+	if (offset > 0) {
+		while (offset-- != 0)
+			nd = rb_next(nd);
+	} else {
+		while (offset++ != 0)
+			nd = rb_prev(nd);
+	}
+
+	browser->top = nd;
+}
+
+unsigned int ui_browser__rb_tree_refresh(struct ui_browser *browser)
+{
+	struct rb_node *nd;
+	int row = 0;
+
+	if (browser->top == NULL)
+                browser->top = rb_first(browser->entries);
+
+	nd = browser->top;
+
+	while (nd != NULL) {
+		ui_browser__gotorc(browser, row, 0);
+		browser->write(browser, nd, row);
+		if (++row == browser->rows)
+			break;
+		nd = rb_next(nd);
+	}
+
+	return row;
+}
+
+bool ui_browser__is_current_entry(struct ui_browser *browser, unsigned row)
+{
+	return browser->top_idx + row == browser->index;
+}
+
+void ui_browser__refresh_dimensions(struct ui_browser *browser)
+{
+	browser->width = SLtt_Screen_Cols - 1;
+	browser->height = browser->rows = SLtt_Screen_Rows - 2;
+	browser->y = 1;
+	browser->x = 0;
+}
+
+void ui_browser__handle_resize(struct ui_browser *browser)
+{
+	ui__refresh_dimensions(false);
+	ui_browser__show(browser, browser->title, ui_helpline__current);
+	ui_browser__refresh(browser);
+}
+
+int ui_browser__warning(struct ui_browser *browser, int timeout,
+			const char *format, ...)
+{
+	va_list args;
+	char *text;
+	int key = 0, err;
+
+	va_start(args, format);
+	err = vasprintf(&text, format, args);
+	va_end(args);
+
+	if (err < 0) {
+		va_start(args, format);
+		ui_helpline__vpush(format, args);
+		va_end(args);
+	} else {
+		while ((key = ui__question_window("Warning!", text,
+						   "Press any key...",
+						   timeout)) == K_RESIZE)
+			ui_browser__handle_resize(browser);
+		free(text);
+	}
+
+	return key;
+}
+
+int ui_browser__help_window(struct ui_browser *browser, const char *text)
+{
+	int key;
+
+	while ((key = ui__help_window(text)) == K_RESIZE)
+		ui_browser__handle_resize(browser);
+
+	return key;
+}
+
+bool ui_browser__dialog_yesno(struct ui_browser *browser, const char *text)
+{
+	int key;
+
+	while ((key = ui__dialog_yesno(text)) == K_RESIZE)
+		ui_browser__handle_resize(browser);
+
+	return key == K_ENTER || toupper(key) == 'Y';
+}
+
+void ui_browser__reset_index(struct ui_browser *browser)
+{
+	browser->index = browser->top_idx = 0;
+	browser->seek(browser, 0, SEEK_SET);
+}
+
+void __ui_browser__show_title(struct ui_browser *browser, const char *title)
+{
+	SLsmg_gotorc(0, 0);
+	ui_browser__set_color(browser, HE_COLORSET_ROOT);
+	slsmg_write_nstring(title, browser->width + 1);
+}
+
+void ui_browser__show_title(struct ui_browser *browser, const char *title)
+{
+	pthread_mutex_lock(&ui__lock);
+	__ui_browser__show_title(browser, title);
+	pthread_mutex_unlock(&ui__lock);
+}
+
+int ui_browser__show(struct ui_browser *browser, const char *title,
+		     const char *helpline, ...)
+{
+	int err;
+	va_list ap;
+
+	if (browser->refresh_dimensions == NULL)
+		browser->refresh_dimensions = ui_browser__refresh_dimensions;
+
+	browser->refresh_dimensions(browser);
+
+	pthread_mutex_lock(&ui__lock);
+	__ui_browser__show_title(browser, title);
+
+	browser->title = title;
+	zfree(&browser->helpline);
+
+	va_start(ap, helpline);
+	err = vasprintf(&browser->helpline, helpline, ap);
+	va_end(ap);
+	if (err > 0)
+		ui_helpline__push(browser->helpline);
+	pthread_mutex_unlock(&ui__lock);
+	return err ? 0 : -1;
+}
+
+void ui_browser__hide(struct ui_browser *browser)
+{
+	pthread_mutex_lock(&ui__lock);
+	ui_helpline__pop();
+	zfree(&browser->helpline);
+	pthread_mutex_unlock(&ui__lock);
+}
+
+static void ui_browser__scrollbar_set(struct ui_browser *browser)
+{
+	int height = browser->height, h = 0, pct = 0,
+	    col = browser->width,
+	    row = 0;
+
+	if (browser->nr_entries > 1) {
+		pct = ((browser->index * (browser->height - 1)) /
+		       (browser->nr_entries - 1));
+	}
+
+	SLsmg_set_char_set(1);
+
+	while (h < height) {
+	        ui_browser__gotorc(browser, row++, col);
+		SLsmg_write_char(h == pct ? SLSMG_DIAMOND_CHAR : SLSMG_CKBRD_CHAR);
+		++h;
+	}
+
+	SLsmg_set_char_set(0);
+}
+
+static int __ui_browser__refresh(struct ui_browser *browser)
+{
+	int row;
+	int width = browser->width;
+
+	row = browser->refresh(browser);
+	ui_browser__set_color(browser, HE_COLORSET_NORMAL);
+
+	if (!browser->use_navkeypressed || browser->navkeypressed)
+		ui_browser__scrollbar_set(browser);
+	else
+		width += 1;
+
+	SLsmg_fill_region(browser->y + row, browser->x,
+			  browser->height - row, width, ' ');
+
+	return 0;
+}
+
+int ui_browser__refresh(struct ui_browser *browser)
+{
+	pthread_mutex_lock(&ui__lock);
+	__ui_browser__refresh(browser);
+	pthread_mutex_unlock(&ui__lock);
+
+	return 0;
+}
+
+/*
+ * Here we're updating nr_entries _after_ we started browsing, i.e.  we have to
+ * forget about any reference to any entry in the underlying data structure,
+ * that is why we do a SEEK_SET. Think about 'perf top' in the hists browser
+ * after an output_resort and hist decay.
+ */
+void ui_browser__update_nr_entries(struct ui_browser *browser, u32 nr_entries)
+{
+	off_t offset = nr_entries - browser->nr_entries;
+
+	browser->nr_entries = nr_entries;
+
+	if (offset < 0) {
+		if (browser->top_idx < (u64)-offset)
+			offset = -browser->top_idx;
+
+		browser->index += offset;
+		browser->top_idx += offset;
+	}
+
+	browser->top = NULL;
+	browser->seek(browser, browser->top_idx, SEEK_SET);
+}
+
+int ui_browser__run(struct ui_browser *browser, int delay_secs)
+{
+	int err, key;
+
+	while (1) {
+		off_t offset;
+
+		pthread_mutex_lock(&ui__lock);
+		err = __ui_browser__refresh(browser);
+		SLsmg_refresh();
+		pthread_mutex_unlock(&ui__lock);
+		if (err < 0)
+			break;
+
+		key = ui__getch(delay_secs);
+
+		if (key == K_RESIZE) {
+			ui__refresh_dimensions(false);
+			browser->refresh_dimensions(browser);
+			__ui_browser__show_title(browser, browser->title);
+			ui_helpline__puts(browser->helpline);
+			continue;
+		}
+
+		if (browser->use_navkeypressed && !browser->navkeypressed) {
+			if (key == K_DOWN || key == K_UP ||
+			    key == K_PGDN || key == K_PGUP ||
+			    key == K_HOME || key == K_END ||
+			    key == ' ') {
+				browser->navkeypressed = true;
+				continue;
+			} else
+				return key;
+		}
+
+		switch (key) {
+		case K_DOWN:
+			if (browser->index == browser->nr_entries - 1)
+				break;
+			++browser->index;
+			if (browser->index == browser->top_idx + browser->rows) {
+				++browser->top_idx;
+				browser->seek(browser, +1, SEEK_CUR);
+			}
+			break;
+		case K_UP:
+			if (browser->index == 0)
+				break;
+			--browser->index;
+			if (browser->index < browser->top_idx) {
+				--browser->top_idx;
+				browser->seek(browser, -1, SEEK_CUR);
+			}
+			break;
+		case K_PGDN:
+		case ' ':
+			if (browser->top_idx + browser->rows > browser->nr_entries - 1)
+				break;
+
+			offset = browser->rows;
+			if (browser->index + offset > browser->nr_entries - 1)
+				offset = browser->nr_entries - 1 - browser->index;
+			browser->index += offset;
+			browser->top_idx += offset;
+			browser->seek(browser, +offset, SEEK_CUR);
+			break;
+		case K_PGUP:
+			if (browser->top_idx == 0)
+				break;
+
+			if (browser->top_idx < browser->rows)
+				offset = browser->top_idx;
+			else
+				offset = browser->rows;
+
+			browser->index -= offset;
+			browser->top_idx -= offset;
+			browser->seek(browser, -offset, SEEK_CUR);
+			break;
+		case K_HOME:
+			ui_browser__reset_index(browser);
+			break;
+		case K_END:
+			offset = browser->rows - 1;
+			if (offset >= browser->nr_entries)
+				offset = browser->nr_entries - 1;
+
+			browser->index = browser->nr_entries - 1;
+			browser->top_idx = browser->index - offset;
+			browser->seek(browser, -offset, SEEK_END);
+			break;
+		default:
+			return key;
+		}
+	}
+	return -1;
+}
+
+unsigned int ui_browser__list_head_refresh(struct ui_browser *browser)
+{
+	struct list_head *pos;
+	struct list_head *head = browser->entries;
+	int row = 0;
+
+	if (browser->top == NULL || browser->top == browser->entries)
+                browser->top = ui_browser__list_head_filter_entries(browser, head->next);
+
+	pos = browser->top;
+
+	list_for_each_from(pos, head) {
+		if (!browser->filter || !browser->filter(browser, pos)) {
+			ui_browser__gotorc(browser, row, 0);
+			browser->write(browser, pos, row);
+			if (++row == browser->rows)
+				break;
+		}
+	}
+
+	return row;
+}
+
+static struct ui_browser_colorset {
+	const char *name, *fg, *bg;
+	int colorset;
+} ui_browser__colorsets[] = {
+	{
+		.colorset = HE_COLORSET_TOP,
+		.name	  = "top",
+		.fg	  = "red",
+		.bg	  = "default",
+	},
+	{
+		.colorset = HE_COLORSET_MEDIUM,
+		.name	  = "medium",
+		.fg	  = "green",
+		.bg	  = "default",
+	},
+	{
+		.colorset = HE_COLORSET_NORMAL,
+		.name	  = "normal",
+		.fg	  = "default",
+		.bg	  = "default",
+	},
+	{
+		.colorset = HE_COLORSET_SELECTED,
+		.name	  = "selected",
+		.fg	  = "black",
+		.bg	  = "lightgray",
+	},
+	{
+		.colorset = HE_COLORSET_CODE,
+		.name	  = "code",
+		.fg	  = "blue",
+		.bg	  = "default",
+	},
+	{
+		.colorset = HE_COLORSET_ADDR,
+		.name	  = "addr",
+		.fg	  = "magenta",
+		.bg	  = "default",
+	},
+	{
+		.colorset = HE_COLORSET_ROOT,
+		.name	  = "root",
+		.fg	  = "white",
+		.bg	  = "blue",
+	},
+	{
+		.name = NULL,
+	}
+};
+
+
+static int ui_browser__color_config(const char *var, const char *value,
+				    void *data __maybe_unused)
+{
+	char *fg = NULL, *bg;
+	int i;
+
+	/* same dir for all commands */
+	if (prefixcmp(var, "colors.") != 0)
+		return 0;
+
+	for (i = 0; ui_browser__colorsets[i].name != NULL; ++i) {
+		const char *name = var + 7;
+
+		if (strcmp(ui_browser__colorsets[i].name, name) != 0)
+			continue;
+
+		fg = strdup(value);
+		if (fg == NULL)
+			break;
+
+		bg = strchr(fg, ',');
+		if (bg == NULL)
+			break;
+
+		*bg = '\0';
+		while (isspace(*++bg));
+		ui_browser__colorsets[i].bg = bg;
+		ui_browser__colorsets[i].fg = fg;
+		return 0;
+	}
+
+	free(fg);
+	return -1;
+}
+
+void ui_browser__argv_seek(struct ui_browser *browser, off_t offset, int whence)
+{
+	switch (whence) {
+	case SEEK_SET:
+		browser->top = browser->entries;
+		break;
+	case SEEK_CUR:
+		browser->top = browser->top + browser->top_idx + offset;
+		break;
+	case SEEK_END:
+		browser->top = browser->top + browser->nr_entries - 1 + offset;
+		break;
+	default:
+		return;
+	}
+}
+
+unsigned int ui_browser__argv_refresh(struct ui_browser *browser)
+{
+	unsigned int row = 0, idx = browser->top_idx;
+	char **pos;
+
+	if (browser->top == NULL)
+		browser->top = browser->entries;
+
+	pos = (char **)browser->top;
+	while (idx < browser->nr_entries) {
+		if (!browser->filter || !browser->filter(browser, *pos)) {
+			ui_browser__gotorc(browser, row, 0);
+			browser->write(browser, pos, row);
+			if (++row == browser->rows)
+				break;
+		}
+
+		++idx;
+		++pos;
+	}
+
+	return row;
+}
+
+void __ui_browser__vline(struct ui_browser *browser, unsigned int column,
+			 u16 start, u16 end)
+{
+	SLsmg_set_char_set(1);
+	ui_browser__gotorc(browser, start, column);
+	SLsmg_draw_vline(end - start + 1);
+	SLsmg_set_char_set(0);
+}
+
+void ui_browser__write_graph(struct ui_browser *browser __maybe_unused,
+			     int graph)
+{
+	SLsmg_set_char_set(1);
+	SLsmg_write_char(graph);
+	SLsmg_set_char_set(0);
+}
+
+static void __ui_browser__line_arrow_up(struct ui_browser *browser,
+					unsigned int column,
+					u64 start, u64 end)
+{
+	unsigned int row, end_row;
+
+	SLsmg_set_char_set(1);
+
+	if (start < browser->top_idx + browser->rows) {
+		row = start - browser->top_idx;
+		ui_browser__gotorc(browser, row, column);
+		SLsmg_write_char(SLSMG_LLCORN_CHAR);
+		ui_browser__gotorc(browser, row, column + 1);
+		SLsmg_draw_hline(2);
+
+		if (row-- == 0)
+			goto out;
+	} else
+		row = browser->rows - 1;
+
+	if (end > browser->top_idx)
+		end_row = end - browser->top_idx;
+	else
+		end_row = 0;
+
+	ui_browser__gotorc(browser, end_row, column);
+	SLsmg_draw_vline(row - end_row + 1);
+
+	ui_browser__gotorc(browser, end_row, column);
+	if (end >= browser->top_idx) {
+		SLsmg_write_char(SLSMG_ULCORN_CHAR);
+		ui_browser__gotorc(browser, end_row, column + 1);
+		SLsmg_write_char(SLSMG_HLINE_CHAR);
+		ui_browser__gotorc(browser, end_row, column + 2);
+		SLsmg_write_char(SLSMG_RARROW_CHAR);
+	}
+out:
+	SLsmg_set_char_set(0);
+}
+
+static void __ui_browser__line_arrow_down(struct ui_browser *browser,
+					  unsigned int column,
+					  u64 start, u64 end)
+{
+	unsigned int row, end_row;
+
+	SLsmg_set_char_set(1);
+
+	if (start >= browser->top_idx) {
+		row = start - browser->top_idx;
+		ui_browser__gotorc(browser, row, column);
+		SLsmg_write_char(SLSMG_ULCORN_CHAR);
+		ui_browser__gotorc(browser, row, column + 1);
+		SLsmg_draw_hline(2);
+
+		if (row++ == 0)
+			goto out;
+	} else
+		row = 0;
+
+	if (end >= browser->top_idx + browser->rows)
+		end_row = browser->rows - 1;
+	else
+		end_row = end - browser->top_idx;
+
+	ui_browser__gotorc(browser, row, column);
+	SLsmg_draw_vline(end_row - row + 1);
+
+	ui_browser__gotorc(browser, end_row, column);
+	if (end < browser->top_idx + browser->rows) {
+		SLsmg_write_char(SLSMG_LLCORN_CHAR);
+		ui_browser__gotorc(browser, end_row, column + 1);
+		SLsmg_write_char(SLSMG_HLINE_CHAR);
+		ui_browser__gotorc(browser, end_row, column + 2);
+		SLsmg_write_char(SLSMG_RARROW_CHAR);
+	}
+out:
+	SLsmg_set_char_set(0);
+}
+
+void __ui_browser__line_arrow(struct ui_browser *browser, unsigned int column,
+			      u64 start, u64 end)
+{
+	if (start > end)
+		__ui_browser__line_arrow_up(browser, column, start, end);
+	else
+		__ui_browser__line_arrow_down(browser, column, start, end);
+}
+
+void ui_browser__init(void)
+{
+	int i = 0;
+
+	perf_config(ui_browser__color_config, NULL);
+
+	while (ui_browser__colorsets[i].name) {
+		struct ui_browser_colorset *c = &ui_browser__colorsets[i++];
+		sltt_set_color(c->colorset, c->name, c->fg, c->bg);
+	}
+
+	annotate_browser__init();
+}
diff --git a/tools/perf/ui/browser.h b/tools/perf/ui/browser.h
new file mode 100644
index 000000000..92ae72113
--- /dev/null
+++ b/tools/perf/ui/browser.h
@@ -0,0 +1,75 @@
+#ifndef _PERF_UI_BROWSER_H_
+#define _PERF_UI_BROWSER_H_ 1
+
+#include <linux/types.h>
+
+#define HE_COLORSET_TOP		50
+#define HE_COLORSET_MEDIUM	51
+#define HE_COLORSET_NORMAL	52
+#define HE_COLORSET_SELECTED	53
+#define HE_COLORSET_CODE	54
+#define HE_COLORSET_ADDR	55
+#define HE_COLORSET_ROOT	56
+
+struct ui_browser {
+	u64	      index, top_idx;
+	void	      *top, *entries;
+	u16	      y, x, width, height, rows;
+	int	      current_color;
+	void	      *priv;
+	const char    *title;
+	char	      *helpline;
+	void 	      (*refresh_dimensions)(struct ui_browser *browser);
+	unsigned int  (*refresh)(struct ui_browser *browser);
+	void	      (*write)(struct ui_browser *browser, void *entry, int row);
+	void	      (*seek)(struct ui_browser *browser, off_t offset, int whence);
+	bool	      (*filter)(struct ui_browser *browser, void *entry);
+	u32	      nr_entries;
+	bool	      navkeypressed;
+	bool	      use_navkeypressed;
+};
+
+int  ui_browser__set_color(struct ui_browser *browser, int color);
+void ui_browser__set_percent_color(struct ui_browser *browser,
+				   double percent, bool current);
+bool ui_browser__is_current_entry(struct ui_browser *browser, unsigned row);
+void ui_browser__refresh_dimensions(struct ui_browser *browser);
+void ui_browser__reset_index(struct ui_browser *browser);
+
+void ui_browser__gotorc(struct ui_browser *browser, int y, int x);
+void ui_browser__write_graph(struct ui_browser *browser, int graph);
+void __ui_browser__line_arrow(struct ui_browser *browser, unsigned int column,
+			      u64 start, u64 end);
+void __ui_browser__show_title(struct ui_browser *browser, const char *title);
+void ui_browser__show_title(struct ui_browser *browser, const char *title);
+int ui_browser__show(struct ui_browser *browser, const char *title,
+		     const char *helpline, ...);
+void ui_browser__hide(struct ui_browser *browser);
+int ui_browser__refresh(struct ui_browser *browser);
+int ui_browser__run(struct ui_browser *browser, int delay_secs);
+void ui_browser__update_nr_entries(struct ui_browser *browser, u32 nr_entries);
+void ui_browser__handle_resize(struct ui_browser *browser);
+void __ui_browser__vline(struct ui_browser *browser, unsigned int column,
+			 u16 start, u16 end);
+
+int ui_browser__warning(struct ui_browser *browser, int timeout,
+			const char *format, ...);
+int ui_browser__help_window(struct ui_browser *browser, const char *text);
+bool ui_browser__dialog_yesno(struct ui_browser *browser, const char *text);
+int ui_browser__input_window(const char *title, const char *text, char *input,
+			     const char *exit_msg, int delay_sec);
+struct perf_session_env;
+int tui__header_window(struct perf_session_env *env);
+
+void ui_browser__argv_seek(struct ui_browser *browser, off_t offset, int whence);
+unsigned int ui_browser__argv_refresh(struct ui_browser *browser);
+
+void ui_browser__rb_tree_seek(struct ui_browser *browser, off_t offset, int whence);
+unsigned int ui_browser__rb_tree_refresh(struct ui_browser *browser);
+
+void ui_browser__list_head_seek(struct ui_browser *browser, off_t offset, int whence);
+unsigned int ui_browser__list_head_refresh(struct ui_browser *browser);
+
+void ui_browser__init(void);
+void annotate_browser__init(void);
+#endif /* _PERF_UI_BROWSER_H_ */
diff --git a/tools/perf/ui/browsers/Build b/tools/perf/ui/browsers/Build
new file mode 100644
index 000000000..de223f5be
--- /dev/null
+++ b/tools/perf/ui/browsers/Build
@@ -0,0 +1,10 @@
+libperf-y += annotate.o
+libperf-y += hists.o
+libperf-y += map.o
+libperf-y += scripts.o
+libperf-y += header.o
+
+CFLAGS_annotate.o += -DENABLE_SLFUTURE_CONST
+CFLAGS_hists.o    += -DENABLE_SLFUTURE_CONST
+CFLAGS_map.o      += -DENABLE_SLFUTURE_CONST
+CFLAGS_scripts.o  += -DENABLE_SLFUTURE_CONST
diff --git a/tools/perf/ui/browsers/annotate.c b/tools/perf/ui/browsers/annotate.c
new file mode 100644
index 000000000..e5250eb2d
--- /dev/null
+++ b/tools/perf/ui/browsers/annotate.c
@@ -0,0 +1,1039 @@
+#include "../../util/util.h"
+#include "../browser.h"
+#include "../helpline.h"
+#include "../libslang.h"
+#include "../ui.h"
+#include "../util.h"
+#include "../../util/annotate.h"
+#include "../../util/hist.h"
+#include "../../util/sort.h"
+#include "../../util/symbol.h"
+#include "../../util/evsel.h"
+#include <pthread.h>
+
+struct browser_disasm_line {
+	struct rb_node	rb_node;
+	u32		idx;
+	int		idx_asm;
+	int		jump_sources;
+	/*
+	 * actual length of this array is saved on the nr_events field
+	 * of the struct annotate_browser
+	 */
+	double		percent[1];
+};
+
+static struct annotate_browser_opt {
+	bool hide_src_code,
+	     use_offset,
+	     jump_arrows,
+	     show_linenr,
+	     show_nr_jumps;
+} annotate_browser__opts = {
+	.use_offset	= true,
+	.jump_arrows	= true,
+};
+
+struct annotate_browser {
+	struct ui_browser b;
+	struct rb_root	  entries;
+	struct rb_node	  *curr_hot;
+	struct disasm_line  *selection;
+	struct disasm_line  **offsets;
+	int		    nr_events;
+	u64		    start;
+	int		    nr_asm_entries;
+	int		    nr_entries;
+	int		    max_jump_sources;
+	int		    nr_jumps;
+	bool		    searching_backwards;
+	u8		    addr_width;
+	u8		    jumps_width;
+	u8		    target_width;
+	u8		    min_addr_width;
+	u8		    max_addr_width;
+	char		    search_bf[128];
+};
+
+static inline struct browser_disasm_line *disasm_line__browser(struct disasm_line *dl)
+{
+	return (struct browser_disasm_line *)(dl + 1);
+}
+
+static bool disasm_line__filter(struct ui_browser *browser __maybe_unused,
+				void *entry)
+{
+	if (annotate_browser__opts.hide_src_code) {
+		struct disasm_line *dl = list_entry(entry, struct disasm_line, node);
+		return dl->offset == -1;
+	}
+
+	return false;
+}
+
+static int annotate_browser__jumps_percent_color(struct annotate_browser *browser,
+						 int nr, bool current)
+{
+	if (current && (!browser->b.use_navkeypressed || browser->b.navkeypressed))
+		return HE_COLORSET_SELECTED;
+	if (nr == browser->max_jump_sources)
+		return HE_COLORSET_TOP;
+	if (nr > 1)
+		return HE_COLORSET_MEDIUM;
+	return HE_COLORSET_NORMAL;
+}
+
+static int annotate_browser__set_jumps_percent_color(struct annotate_browser *browser,
+						     int nr, bool current)
+{
+	 int color = annotate_browser__jumps_percent_color(browser, nr, current);
+	 return ui_browser__set_color(&browser->b, color);
+}
+
+static void annotate_browser__write(struct ui_browser *browser, void *entry, int row)
+{
+	struct annotate_browser *ab = container_of(browser, struct annotate_browser, b);
+	struct disasm_line *dl = list_entry(entry, struct disasm_line, node);
+	struct browser_disasm_line *bdl = disasm_line__browser(dl);
+	bool current_entry = ui_browser__is_current_entry(browser, row);
+	bool change_color = (!annotate_browser__opts.hide_src_code &&
+			     (!current_entry || (browser->use_navkeypressed &&
+					         !browser->navkeypressed)));
+	int width = browser->width, printed;
+	int i, pcnt_width = 7 * ab->nr_events;
+	double percent_max = 0.0;
+	char bf[256];
+
+	for (i = 0; i < ab->nr_events; i++) {
+		if (bdl->percent[i] > percent_max)
+			percent_max = bdl->percent[i];
+	}
+
+	if (dl->offset != -1 && percent_max != 0.0) {
+		for (i = 0; i < ab->nr_events; i++) {
+			ui_browser__set_percent_color(browser, bdl->percent[i],
+						      current_entry);
+			slsmg_printf("%6.2f ", bdl->percent[i]);
+		}
+	} else {
+		ui_browser__set_percent_color(browser, 0, current_entry);
+		slsmg_write_nstring(" ", pcnt_width);
+	}
+
+	SLsmg_write_char(' ');
+
+	/* The scroll bar isn't being used */
+	if (!browser->navkeypressed)
+		width += 1;
+
+	if (!*dl->line)
+		slsmg_write_nstring(" ", width - pcnt_width);
+	else if (dl->offset == -1) {
+		if (dl->line_nr && annotate_browser__opts.show_linenr)
+			printed = scnprintf(bf, sizeof(bf), "%-*d ",
+					ab->addr_width + 1, dl->line_nr);
+		else
+			printed = scnprintf(bf, sizeof(bf), "%*s  ",
+				    ab->addr_width, " ");
+		slsmg_write_nstring(bf, printed);
+		slsmg_write_nstring(dl->line, width - printed - pcnt_width + 1);
+	} else {
+		u64 addr = dl->offset;
+		int color = -1;
+
+		if (!annotate_browser__opts.use_offset)
+			addr += ab->start;
+
+		if (!annotate_browser__opts.use_offset) {
+			printed = scnprintf(bf, sizeof(bf), "%" PRIx64 ": ", addr);
+		} else {
+			if (bdl->jump_sources) {
+				if (annotate_browser__opts.show_nr_jumps) {
+					int prev;
+					printed = scnprintf(bf, sizeof(bf), "%*d ",
+							    ab->jumps_width,
+							    bdl->jump_sources);
+					prev = annotate_browser__set_jumps_percent_color(ab, bdl->jump_sources,
+											 current_entry);
+					slsmg_write_nstring(bf, printed);
+					ui_browser__set_color(browser, prev);
+				}
+
+				printed = scnprintf(bf, sizeof(bf), "%*" PRIx64 ": ",
+						    ab->target_width, addr);
+			} else {
+				printed = scnprintf(bf, sizeof(bf), "%*s  ",
+						    ab->addr_width, " ");
+			}
+		}
+
+		if (change_color)
+			color = ui_browser__set_color(browser, HE_COLORSET_ADDR);
+		slsmg_write_nstring(bf, printed);
+		if (change_color)
+			ui_browser__set_color(browser, color);
+		if (dl->ins && dl->ins->ops->scnprintf) {
+			if (ins__is_jump(dl->ins)) {
+				bool fwd = dl->ops.target.offset > (u64)dl->offset;
+
+				ui_browser__write_graph(browser, fwd ? SLSMG_DARROW_CHAR :
+								    SLSMG_UARROW_CHAR);
+				SLsmg_write_char(' ');
+			} else if (ins__is_call(dl->ins)) {
+				ui_browser__write_graph(browser, SLSMG_RARROW_CHAR);
+				SLsmg_write_char(' ');
+			} else {
+				slsmg_write_nstring(" ", 2);
+			}
+		} else {
+			if (strcmp(dl->name, "retq")) {
+				slsmg_write_nstring(" ", 2);
+			} else {
+				ui_browser__write_graph(browser, SLSMG_LARROW_CHAR);
+				SLsmg_write_char(' ');
+			}
+		}
+
+		disasm_line__scnprintf(dl, bf, sizeof(bf), !annotate_browser__opts.use_offset);
+		slsmg_write_nstring(bf, width - pcnt_width - 3 - printed);
+	}
+
+	if (current_entry)
+		ab->selection = dl;
+}
+
+static bool disasm_line__is_valid_jump(struct disasm_line *dl, struct symbol *sym)
+{
+	if (!dl || !dl->ins || !ins__is_jump(dl->ins)
+	    || !disasm_line__has_offset(dl)
+	    || dl->ops.target.offset >= symbol__size(sym))
+		return false;
+
+	return true;
+}
+
+static void annotate_browser__draw_current_jump(struct ui_browser *browser)
+{
+	struct annotate_browser *ab = container_of(browser, struct annotate_browser, b);
+	struct disasm_line *cursor = ab->selection, *target;
+	struct browser_disasm_line *btarget, *bcursor;
+	unsigned int from, to;
+	struct map_symbol *ms = ab->b.priv;
+	struct symbol *sym = ms->sym;
+	u8 pcnt_width = 7;
+
+	/* PLT symbols contain external offsets */
+	if (strstr(sym->name, "@plt"))
+		return;
+
+	if (!disasm_line__is_valid_jump(cursor, sym))
+		return;
+
+	target = ab->offsets[cursor->ops.target.offset];
+	if (!target)
+		return;
+
+	bcursor = disasm_line__browser(cursor);
+	btarget = disasm_line__browser(target);
+
+	if (annotate_browser__opts.hide_src_code) {
+		from = bcursor->idx_asm;
+		to = btarget->idx_asm;
+	} else {
+		from = (u64)bcursor->idx;
+		to = (u64)btarget->idx;
+	}
+
+	pcnt_width *= ab->nr_events;
+
+	ui_browser__set_color(browser, HE_COLORSET_CODE);
+	__ui_browser__line_arrow(browser, pcnt_width + 2 + ab->addr_width,
+				 from, to);
+}
+
+static unsigned int annotate_browser__refresh(struct ui_browser *browser)
+{
+	struct annotate_browser *ab = container_of(browser, struct annotate_browser, b);
+	int ret = ui_browser__list_head_refresh(browser);
+	int pcnt_width;
+
+	pcnt_width = 7 * ab->nr_events;
+
+	if (annotate_browser__opts.jump_arrows)
+		annotate_browser__draw_current_jump(browser);
+
+	ui_browser__set_color(browser, HE_COLORSET_NORMAL);
+	__ui_browser__vline(browser, pcnt_width, 0, browser->height - 1);
+	return ret;
+}
+
+static int disasm__cmp(struct browser_disasm_line *a,
+		       struct browser_disasm_line *b, int nr_pcnt)
+{
+	int i;
+
+	for (i = 0; i < nr_pcnt; i++) {
+		if (a->percent[i] == b->percent[i])
+			continue;
+		return a->percent[i] < b->percent[i];
+	}
+	return 0;
+}
+
+static void disasm_rb_tree__insert(struct rb_root *root, struct browser_disasm_line *bdl,
+				   int nr_events)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct browser_disasm_line *l;
+
+	while (*p != NULL) {
+		parent = *p;
+		l = rb_entry(parent, struct browser_disasm_line, rb_node);
+
+		if (disasm__cmp(bdl, l, nr_events))
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&bdl->rb_node, parent, p);
+	rb_insert_color(&bdl->rb_node, root);
+}
+
+static void annotate_browser__set_top(struct annotate_browser *browser,
+				      struct disasm_line *pos, u32 idx)
+{
+	unsigned back;
+
+	ui_browser__refresh_dimensions(&browser->b);
+	back = browser->b.height / 2;
+	browser->b.top_idx = browser->b.index = idx;
+
+	while (browser->b.top_idx != 0 && back != 0) {
+		pos = list_entry(pos->node.prev, struct disasm_line, node);
+
+		if (disasm_line__filter(&browser->b, &pos->node))
+			continue;
+
+		--browser->b.top_idx;
+		--back;
+	}
+
+	browser->b.top = pos;
+	browser->b.navkeypressed = true;
+}
+
+static void annotate_browser__set_rb_top(struct annotate_browser *browser,
+					 struct rb_node *nd)
+{
+	struct browser_disasm_line *bpos;
+	struct disasm_line *pos;
+	u32 idx;
+
+	bpos = rb_entry(nd, struct browser_disasm_line, rb_node);
+	pos = ((struct disasm_line *)bpos) - 1;
+	idx = bpos->idx;
+	if (annotate_browser__opts.hide_src_code)
+		idx = bpos->idx_asm;
+	annotate_browser__set_top(browser, pos, idx);
+	browser->curr_hot = nd;
+}
+
+static void annotate_browser__calc_percent(struct annotate_browser *browser,
+					   struct perf_evsel *evsel)
+{
+	struct map_symbol *ms = browser->b.priv;
+	struct symbol *sym = ms->sym;
+	struct annotation *notes = symbol__annotation(sym);
+	struct disasm_line *pos, *next;
+	s64 len = symbol__size(sym);
+
+	browser->entries = RB_ROOT;
+
+	pthread_mutex_lock(&notes->lock);
+
+	list_for_each_entry(pos, &notes->src->source, node) {
+		struct browser_disasm_line *bpos = disasm_line__browser(pos);
+		const char *path = NULL;
+		double max_percent = 0.0;
+		int i;
+
+		if (pos->offset == -1) {
+			RB_CLEAR_NODE(&bpos->rb_node);
+			continue;
+		}
+
+		next = disasm__get_next_ip_line(&notes->src->source, pos);
+
+		for (i = 0; i < browser->nr_events; i++) {
+			bpos->percent[i] = disasm__calc_percent(notes,
+						evsel->idx + i,
+						pos->offset,
+						next ? next->offset : len,
+					        &path);
+
+			if (max_percent < bpos->percent[i])
+				max_percent = bpos->percent[i];
+		}
+
+		if (max_percent < 0.01) {
+			RB_CLEAR_NODE(&bpos->rb_node);
+			continue;
+		}
+		disasm_rb_tree__insert(&browser->entries, bpos,
+				       browser->nr_events);
+	}
+	pthread_mutex_unlock(&notes->lock);
+
+	browser->curr_hot = rb_last(&browser->entries);
+}
+
+static bool annotate_browser__toggle_source(struct annotate_browser *browser)
+{
+	struct disasm_line *dl;
+	struct browser_disasm_line *bdl;
+	off_t offset = browser->b.index - browser->b.top_idx;
+
+	browser->b.seek(&browser->b, offset, SEEK_CUR);
+	dl = list_entry(browser->b.top, struct disasm_line, node);
+	bdl = disasm_line__browser(dl);
+
+	if (annotate_browser__opts.hide_src_code) {
+		if (bdl->idx_asm < offset)
+			offset = bdl->idx;
+
+		browser->b.nr_entries = browser->nr_entries;
+		annotate_browser__opts.hide_src_code = false;
+		browser->b.seek(&browser->b, -offset, SEEK_CUR);
+		browser->b.top_idx = bdl->idx - offset;
+		browser->b.index = bdl->idx;
+	} else {
+		if (bdl->idx_asm < 0) {
+			ui_helpline__puts("Only available for assembly lines.");
+			browser->b.seek(&browser->b, -offset, SEEK_CUR);
+			return false;
+		}
+
+		if (bdl->idx_asm < offset)
+			offset = bdl->idx_asm;
+
+		browser->b.nr_entries = browser->nr_asm_entries;
+		annotate_browser__opts.hide_src_code = true;
+		browser->b.seek(&browser->b, -offset, SEEK_CUR);
+		browser->b.top_idx = bdl->idx_asm - offset;
+		browser->b.index = bdl->idx_asm;
+	}
+
+	return true;
+}
+
+static void annotate_browser__init_asm_mode(struct annotate_browser *browser)
+{
+	ui_browser__reset_index(&browser->b);
+	browser->b.nr_entries = browser->nr_asm_entries;
+}
+
+#define SYM_TITLE_MAX_SIZE (PATH_MAX + 64)
+
+static int sym_title(struct symbol *sym, struct map *map, char *title,
+		     size_t sz)
+{
+	return snprintf(title, sz, "%s  %s", sym->name, map->dso->long_name);
+}
+
+static bool annotate_browser__callq(struct annotate_browser *browser,
+				    struct perf_evsel *evsel,
+				    struct hist_browser_timer *hbt)
+{
+	struct map_symbol *ms = browser->b.priv;
+	struct disasm_line *dl = browser->selection;
+	struct annotation *notes;
+	struct addr_map_symbol target = {
+		.map = ms->map,
+		.addr = map__objdump_2mem(ms->map, dl->ops.target.addr),
+	};
+	char title[SYM_TITLE_MAX_SIZE];
+
+	if (!ins__is_call(dl->ins))
+		return false;
+
+	if (map_groups__find_ams(&target, NULL) ||
+	    map__rip_2objdump(target.map, target.map->map_ip(target.map,
+							     target.addr)) !=
+	    dl->ops.target.addr) {
+		ui_helpline__puts("The called function was not found.");
+		return true;
+	}
+
+	notes = symbol__annotation(target.sym);
+	pthread_mutex_lock(&notes->lock);
+
+	if (notes->src == NULL && symbol__alloc_hist(target.sym) < 0) {
+		pthread_mutex_unlock(&notes->lock);
+		ui__warning("Not enough memory for annotating '%s' symbol!\n",
+			    target.sym->name);
+		return true;
+	}
+
+	pthread_mutex_unlock(&notes->lock);
+	symbol__tui_annotate(target.sym, target.map, evsel, hbt);
+	sym_title(ms->sym, ms->map, title, sizeof(title));
+	ui_browser__show_title(&browser->b, title);
+	return true;
+}
+
+static
+struct disasm_line *annotate_browser__find_offset(struct annotate_browser *browser,
+					  s64 offset, s64 *idx)
+{
+	struct map_symbol *ms = browser->b.priv;
+	struct symbol *sym = ms->sym;
+	struct annotation *notes = symbol__annotation(sym);
+	struct disasm_line *pos;
+
+	*idx = 0;
+	list_for_each_entry(pos, &notes->src->source, node) {
+		if (pos->offset == offset)
+			return pos;
+		if (!disasm_line__filter(&browser->b, &pos->node))
+			++*idx;
+	}
+
+	return NULL;
+}
+
+static bool annotate_browser__jump(struct annotate_browser *browser)
+{
+	struct disasm_line *dl = browser->selection;
+	s64 idx;
+
+	if (!ins__is_jump(dl->ins))
+		return false;
+
+	dl = annotate_browser__find_offset(browser, dl->ops.target.offset, &idx);
+	if (dl == NULL) {
+		ui_helpline__puts("Invalid jump offset");
+		return true;
+	}
+
+	annotate_browser__set_top(browser, dl, idx);
+
+	return true;
+}
+
+static
+struct disasm_line *annotate_browser__find_string(struct annotate_browser *browser,
+					  char *s, s64 *idx)
+{
+	struct map_symbol *ms = browser->b.priv;
+	struct symbol *sym = ms->sym;
+	struct annotation *notes = symbol__annotation(sym);
+	struct disasm_line *pos = browser->selection;
+
+	*idx = browser->b.index;
+	list_for_each_entry_continue(pos, &notes->src->source, node) {
+		if (disasm_line__filter(&browser->b, &pos->node))
+			continue;
+
+		++*idx;
+
+		if (pos->line && strstr(pos->line, s) != NULL)
+			return pos;
+	}
+
+	return NULL;
+}
+
+static bool __annotate_browser__search(struct annotate_browser *browser)
+{
+	struct disasm_line *dl;
+	s64 idx;
+
+	dl = annotate_browser__find_string(browser, browser->search_bf, &idx);
+	if (dl == NULL) {
+		ui_helpline__puts("String not found!");
+		return false;
+	}
+
+	annotate_browser__set_top(browser, dl, idx);
+	browser->searching_backwards = false;
+	return true;
+}
+
+static
+struct disasm_line *annotate_browser__find_string_reverse(struct annotate_browser *browser,
+						  char *s, s64 *idx)
+{
+	struct map_symbol *ms = browser->b.priv;
+	struct symbol *sym = ms->sym;
+	struct annotation *notes = symbol__annotation(sym);
+	struct disasm_line *pos = browser->selection;
+
+	*idx = browser->b.index;
+	list_for_each_entry_continue_reverse(pos, &notes->src->source, node) {
+		if (disasm_line__filter(&browser->b, &pos->node))
+			continue;
+
+		--*idx;
+
+		if (pos->line && strstr(pos->line, s) != NULL)
+			return pos;
+	}
+
+	return NULL;
+}
+
+static bool __annotate_browser__search_reverse(struct annotate_browser *browser)
+{
+	struct disasm_line *dl;
+	s64 idx;
+
+	dl = annotate_browser__find_string_reverse(browser, browser->search_bf, &idx);
+	if (dl == NULL) {
+		ui_helpline__puts("String not found!");
+		return false;
+	}
+
+	annotate_browser__set_top(browser, dl, idx);
+	browser->searching_backwards = true;
+	return true;
+}
+
+static bool annotate_browser__search_window(struct annotate_browser *browser,
+					    int delay_secs)
+{
+	if (ui_browser__input_window("Search", "String: ", browser->search_bf,
+				     "ENTER: OK, ESC: Cancel",
+				     delay_secs * 2) != K_ENTER ||
+	    !*browser->search_bf)
+		return false;
+
+	return true;
+}
+
+static bool annotate_browser__search(struct annotate_browser *browser, int delay_secs)
+{
+	if (annotate_browser__search_window(browser, delay_secs))
+		return __annotate_browser__search(browser);
+
+	return false;
+}
+
+static bool annotate_browser__continue_search(struct annotate_browser *browser,
+					      int delay_secs)
+{
+	if (!*browser->search_bf)
+		return annotate_browser__search(browser, delay_secs);
+
+	return __annotate_browser__search(browser);
+}
+
+static bool annotate_browser__search_reverse(struct annotate_browser *browser,
+					   int delay_secs)
+{
+	if (annotate_browser__search_window(browser, delay_secs))
+		return __annotate_browser__search_reverse(browser);
+
+	return false;
+}
+
+static
+bool annotate_browser__continue_search_reverse(struct annotate_browser *browser,
+					       int delay_secs)
+{
+	if (!*browser->search_bf)
+		return annotate_browser__search_reverse(browser, delay_secs);
+
+	return __annotate_browser__search_reverse(browser);
+}
+
+static void annotate_browser__update_addr_width(struct annotate_browser *browser)
+{
+	if (annotate_browser__opts.use_offset)
+		browser->target_width = browser->min_addr_width;
+	else
+		browser->target_width = browser->max_addr_width;
+
+	browser->addr_width = browser->target_width;
+
+	if (annotate_browser__opts.show_nr_jumps)
+		browser->addr_width += browser->jumps_width + 1;
+}
+
+static int annotate_browser__run(struct annotate_browser *browser,
+				 struct perf_evsel *evsel,
+				 struct hist_browser_timer *hbt)
+{
+	struct rb_node *nd = NULL;
+	struct map_symbol *ms = browser->b.priv;
+	struct symbol *sym = ms->sym;
+	const char *help = "Press 'h' for help on key bindings";
+	int delay_secs = hbt ? hbt->refresh : 0;
+	int key;
+	char title[SYM_TITLE_MAX_SIZE];
+
+	sym_title(sym, ms->map, title, sizeof(title));
+	if (ui_browser__show(&browser->b, title, help) < 0)
+		return -1;
+
+	annotate_browser__calc_percent(browser, evsel);
+
+	if (browser->curr_hot) {
+		annotate_browser__set_rb_top(browser, browser->curr_hot);
+		browser->b.navkeypressed = false;
+	}
+
+	nd = browser->curr_hot;
+
+	while (1) {
+		key = ui_browser__run(&browser->b, delay_secs);
+
+		if (delay_secs != 0) {
+			annotate_browser__calc_percent(browser, evsel);
+			/*
+			 * Current line focus got out of the list of most active
+			 * lines, NULL it so that if TAB|UNTAB is pressed, we
+			 * move to curr_hot (current hottest line).
+			 */
+			if (nd != NULL && RB_EMPTY_NODE(nd))
+				nd = NULL;
+		}
+
+		switch (key) {
+		case K_TIMER:
+			if (hbt)
+				hbt->timer(hbt->arg);
+
+			if (delay_secs != 0)
+				symbol__annotate_decay_histogram(sym, evsel->idx);
+			continue;
+		case K_TAB:
+			if (nd != NULL) {
+				nd = rb_prev(nd);
+				if (nd == NULL)
+					nd = rb_last(&browser->entries);
+			} else
+				nd = browser->curr_hot;
+			break;
+		case K_UNTAB:
+			if (nd != NULL)
+				nd = rb_next(nd);
+				if (nd == NULL)
+					nd = rb_first(&browser->entries);
+			else
+				nd = browser->curr_hot;
+			break;
+		case K_F1:
+		case 'h':
+			ui_browser__help_window(&browser->b,
+		"UP/DOWN/PGUP\n"
+		"PGDN/SPACE    Navigate\n"
+		"q/ESC/CTRL+C  Exit\n\n"
+		"->            Go to target\n"
+		"<-            Exit\n"
+		"H             Cycle thru hottest instructions\n"
+		"j             Toggle showing jump to target arrows\n"
+		"J             Toggle showing number of jump sources on targets\n"
+		"n             Search next string\n"
+		"o             Toggle disassembler output/simplified view\n"
+		"s             Toggle source code view\n"
+		"/             Search string\n"
+		"k             Toggle line numbers\n"
+		"r             Run available scripts\n"
+		"?             Search string backwards\n");
+			continue;
+		case 'r':
+			{
+				script_browse(NULL);
+				continue;
+			}
+		case 'k':
+			annotate_browser__opts.show_linenr =
+				!annotate_browser__opts.show_linenr;
+			break;
+		case 'H':
+			nd = browser->curr_hot;
+			break;
+		case 's':
+			if (annotate_browser__toggle_source(browser))
+				ui_helpline__puts(help);
+			continue;
+		case 'o':
+			annotate_browser__opts.use_offset = !annotate_browser__opts.use_offset;
+			annotate_browser__update_addr_width(browser);
+			continue;
+		case 'j':
+			annotate_browser__opts.jump_arrows = !annotate_browser__opts.jump_arrows;
+			continue;
+		case 'J':
+			annotate_browser__opts.show_nr_jumps = !annotate_browser__opts.show_nr_jumps;
+			annotate_browser__update_addr_width(browser);
+			continue;
+		case '/':
+			if (annotate_browser__search(browser, delay_secs)) {
+show_help:
+				ui_helpline__puts(help);
+			}
+			continue;
+		case 'n':
+			if (browser->searching_backwards ?
+			    annotate_browser__continue_search_reverse(browser, delay_secs) :
+			    annotate_browser__continue_search(browser, delay_secs))
+				goto show_help;
+			continue;
+		case '?':
+			if (annotate_browser__search_reverse(browser, delay_secs))
+				goto show_help;
+			continue;
+		case 'D': {
+			static int seq;
+			ui_helpline__pop();
+			ui_helpline__fpush("%d: nr_ent=%d, height=%d, idx=%d, top_idx=%d, nr_asm_entries=%d",
+					   seq++, browser->b.nr_entries,
+					   browser->b.height,
+					   browser->b.index,
+					   browser->b.top_idx,
+					   browser->nr_asm_entries);
+		}
+			continue;
+		case K_ENTER:
+		case K_RIGHT:
+			if (browser->selection == NULL)
+				ui_helpline__puts("Huh? No selection. Report to linux-kernel@vger.kernel.org");
+			else if (browser->selection->offset == -1)
+				ui_helpline__puts("Actions are only available for assembly lines.");
+			else if (!browser->selection->ins) {
+				if (strcmp(browser->selection->name, "retq"))
+					goto show_sup_ins;
+				goto out;
+			} else if (!(annotate_browser__jump(browser) ||
+				     annotate_browser__callq(browser, evsel, hbt))) {
+show_sup_ins:
+				ui_helpline__puts("Actions are only available for 'callq', 'retq' & jump instructions.");
+			}
+			continue;
+		case K_LEFT:
+		case K_ESC:
+		case 'q':
+		case CTRL('c'):
+			goto out;
+		default:
+			continue;
+		}
+
+		if (nd != NULL)
+			annotate_browser__set_rb_top(browser, nd);
+	}
+out:
+	ui_browser__hide(&browser->b);
+	return key;
+}
+
+int map_symbol__tui_annotate(struct map_symbol *ms, struct perf_evsel *evsel,
+			     struct hist_browser_timer *hbt)
+{
+	return symbol__tui_annotate(ms->sym, ms->map, evsel, hbt);
+}
+
+int hist_entry__tui_annotate(struct hist_entry *he, struct perf_evsel *evsel,
+			     struct hist_browser_timer *hbt)
+{
+	return map_symbol__tui_annotate(&he->ms, evsel, hbt);
+}
+
+static void annotate_browser__mark_jump_targets(struct annotate_browser *browser,
+						size_t size)
+{
+	u64 offset;
+	struct map_symbol *ms = browser->b.priv;
+	struct symbol *sym = ms->sym;
+
+	/* PLT symbols contain external offsets */
+	if (strstr(sym->name, "@plt"))
+		return;
+
+	for (offset = 0; offset < size; ++offset) {
+		struct disasm_line *dl = browser->offsets[offset], *dlt;
+		struct browser_disasm_line *bdlt;
+
+		if (!disasm_line__is_valid_jump(dl, sym))
+			continue;
+
+		dlt = browser->offsets[dl->ops.target.offset];
+		/*
+ 		 * FIXME: Oops, no jump target? Buggy disassembler? Or do we
+ 		 * have to adjust to the previous offset?
+ 		 */
+		if (dlt == NULL)
+			continue;
+
+		bdlt = disasm_line__browser(dlt);
+		if (++bdlt->jump_sources > browser->max_jump_sources)
+			browser->max_jump_sources = bdlt->jump_sources;
+
+		++browser->nr_jumps;
+	}
+}
+
+static inline int width_jumps(int n)
+{
+	if (n >= 100)
+		return 5;
+	if (n / 10)
+		return 2;
+	return 1;
+}
+
+int symbol__tui_annotate(struct symbol *sym, struct map *map,
+			 struct perf_evsel *evsel,
+			 struct hist_browser_timer *hbt)
+{
+	struct disasm_line *pos, *n;
+	struct annotation *notes;
+	size_t size;
+	struct map_symbol ms = {
+		.map = map,
+		.sym = sym,
+	};
+	struct annotate_browser browser = {
+		.b = {
+			.refresh = annotate_browser__refresh,
+			.seek	 = ui_browser__list_head_seek,
+			.write	 = annotate_browser__write,
+			.filter  = disasm_line__filter,
+			.priv	 = &ms,
+			.use_navkeypressed = true,
+		},
+	};
+	int ret = -1;
+	int nr_pcnt = 1;
+	size_t sizeof_bdl = sizeof(struct browser_disasm_line);
+
+	if (sym == NULL)
+		return -1;
+
+	size = symbol__size(sym);
+
+	if (map->dso->annotate_warned)
+		return -1;
+
+	browser.offsets = zalloc(size * sizeof(struct disasm_line *));
+	if (browser.offsets == NULL) {
+		ui__error("Not enough memory!");
+		return -1;
+	}
+
+	if (perf_evsel__is_group_event(evsel)) {
+		nr_pcnt = evsel->nr_members;
+		sizeof_bdl += sizeof(double) * (nr_pcnt - 1);
+	}
+
+	if (symbol__annotate(sym, map, sizeof_bdl) < 0) {
+		ui__error("%s", ui_helpline__last_msg);
+		goto out_free_offsets;
+	}
+
+	ui_helpline__push("Press <- or ESC to exit");
+
+	notes = symbol__annotation(sym);
+	browser.start = map__rip_2objdump(map, sym->start);
+
+	list_for_each_entry(pos, &notes->src->source, node) {
+		struct browser_disasm_line *bpos;
+		size_t line_len = strlen(pos->line);
+
+		if (browser.b.width < line_len)
+			browser.b.width = line_len;
+		bpos = disasm_line__browser(pos);
+		bpos->idx = browser.nr_entries++;
+		if (pos->offset != -1) {
+			bpos->idx_asm = browser.nr_asm_entries++;
+			/*
+			 * FIXME: short term bandaid to cope with assembly
+			 * routines that comes with labels in the same column
+			 * as the address in objdump, sigh.
+			 *
+			 * E.g. copy_user_generic_unrolled
+ 			 */
+			if (pos->offset < (s64)size)
+				browser.offsets[pos->offset] = pos;
+		} else
+			bpos->idx_asm = -1;
+	}
+
+	annotate_browser__mark_jump_targets(&browser, size);
+
+	browser.addr_width = browser.target_width = browser.min_addr_width = hex_width(size);
+	browser.max_addr_width = hex_width(sym->end);
+	browser.jumps_width = width_jumps(browser.max_jump_sources);
+	browser.nr_events = nr_pcnt;
+	browser.b.nr_entries = browser.nr_entries;
+	browser.b.entries = &notes->src->source,
+	browser.b.width += 18; /* Percentage */
+
+	if (annotate_browser__opts.hide_src_code)
+		annotate_browser__init_asm_mode(&browser);
+
+	annotate_browser__update_addr_width(&browser);
+
+	ret = annotate_browser__run(&browser, evsel, hbt);
+	list_for_each_entry_safe(pos, n, &notes->src->source, node) {
+		list_del(&pos->node);
+		disasm_line__free(pos);
+	}
+
+out_free_offsets:
+	free(browser.offsets);
+	return ret;
+}
+
+#define ANNOTATE_CFG(n) \
+	{ .name = #n, .value = &annotate_browser__opts.n, }
+
+/*
+ * Keep the entries sorted, they are bsearch'ed
+ */
+static struct annotate_config {
+	const char *name;
+	bool *value;
+} annotate__configs[] = {
+	ANNOTATE_CFG(hide_src_code),
+	ANNOTATE_CFG(jump_arrows),
+	ANNOTATE_CFG(show_linenr),
+	ANNOTATE_CFG(show_nr_jumps),
+	ANNOTATE_CFG(use_offset),
+};
+
+#undef ANNOTATE_CFG
+
+static int annotate_config__cmp(const void *name, const void *cfgp)
+{
+	const struct annotate_config *cfg = cfgp;
+
+	return strcmp(name, cfg->name);
+}
+
+static int annotate__config(const char *var, const char *value,
+			    void *data __maybe_unused)
+{
+	struct annotate_config *cfg;
+	const char *name;
+
+	if (prefixcmp(var, "annotate.") != 0)
+		return 0;
+
+	name = var + 9;
+	cfg = bsearch(name, annotate__configs, ARRAY_SIZE(annotate__configs),
+		      sizeof(struct annotate_config), annotate_config__cmp);
+
+	if (cfg == NULL)
+		return -1;
+
+	*cfg->value = perf_config_bool(name, value);
+	return 0;
+}
+
+void annotate_browser__init(void)
+{
+	perf_config(annotate__config, NULL);
+}
diff --git a/tools/perf/ui/browsers/header.c b/tools/perf/ui/browsers/header.c
new file mode 100644
index 000000000..e8278c558
--- /dev/null
+++ b/tools/perf/ui/browsers/header.c
@@ -0,0 +1,128 @@
+#include "util/cache.h"
+#include "util/debug.h"
+#include "ui/browser.h"
+#include "ui/keysyms.h"
+#include "ui/ui.h"
+#include "ui/util.h"
+#include "ui/libslang.h"
+#include "util/header.h"
+#include "util/session.h"
+
+static void ui_browser__argv_write(struct ui_browser *browser,
+				   void *entry, int row)
+{
+	char **arg = entry;
+	char *str = *arg;
+	char empty[] = " ";
+	bool current_entry = ui_browser__is_current_entry(browser, row);
+	unsigned long offset = (unsigned long)browser->priv;
+
+	if (offset >= strlen(str))
+		str = empty;
+	else
+		str = str + offset;
+
+	ui_browser__set_color(browser, current_entry ? HE_COLORSET_SELECTED :
+						       HE_COLORSET_NORMAL);
+
+	slsmg_write_nstring(str, browser->width);
+}
+
+static int list_menu__run(struct ui_browser *menu)
+{
+	int key;
+	unsigned long offset;
+	const char help[] =
+	"h/?/F1        Show this window\n"
+	"UP/DOWN/PGUP\n"
+	"PGDN/SPACE\n"
+	"LEFT/RIGHT    Navigate\n"
+	"q/ESC/CTRL+C  Exit browser";
+
+	if (ui_browser__show(menu, "Header information", "Press 'q' to exit") < 0)
+		return -1;
+
+	while (1) {
+		key = ui_browser__run(menu, 0);
+
+		switch (key) {
+		case K_RIGHT:
+			offset = (unsigned long)menu->priv;
+			offset += 10;
+			menu->priv = (void *)offset;
+			continue;
+		case K_LEFT:
+			offset = (unsigned long)menu->priv;
+			if (offset >= 10)
+				offset -= 10;
+			menu->priv = (void *)offset;
+			continue;
+		case K_F1:
+		case 'h':
+		case '?':
+			ui_browser__help_window(menu, help);
+			continue;
+		case K_ESC:
+		case 'q':
+		case CTRL('c'):
+			key = -1;
+			break;
+		default:
+			continue;
+		}
+
+		break;
+	}
+
+	ui_browser__hide(menu);
+	return key;
+}
+
+static int ui__list_menu(int argc, char * const argv[])
+{
+	struct ui_browser menu = {
+		.entries    = (void *)argv,
+		.refresh    = ui_browser__argv_refresh,
+		.seek	    = ui_browser__argv_seek,
+		.write	    = ui_browser__argv_write,
+		.nr_entries = argc,
+	};
+
+	return list_menu__run(&menu);
+}
+
+int tui__header_window(struct perf_session_env *env)
+{
+	int i, argc = 0;
+	char **argv;
+	struct perf_session *session;
+	char *ptr, *pos;
+	size_t size;
+	FILE *fp = open_memstream(&ptr, &size);
+
+	session = container_of(env, struct perf_session, header.env);
+	perf_header__fprintf_info(session, fp, true);
+	fclose(fp);
+
+	for (pos = ptr, argc = 0; (pos = strchr(pos, '\n')) != NULL; pos++)
+		argc++;
+
+	argv = calloc(argc + 1, sizeof(*argv));
+	if (argv == NULL)
+		goto out;
+
+	argv[0] = pos = ptr;
+	for (i = 1; (pos = strchr(pos, '\n')) != NULL; i++) {
+		*pos++ = '\0';
+		argv[i] = pos;
+	}
+
+	BUG_ON(i != argc + 1);
+
+	ui__list_menu(argc, argv);
+
+out:
+	free(argv);
+	free(ptr);
+	return 0;
+}
diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c
new file mode 100644
index 000000000..995b7a859
--- /dev/null
+++ b/tools/perf/ui/browsers/hists.c
@@ -0,0 +1,2042 @@
+#include <stdio.h>
+#include "../libslang.h"
+#include <stdlib.h>
+#include <string.h>
+#include <linux/rbtree.h>
+
+#include "../../util/evsel.h"
+#include "../../util/evlist.h"
+#include "../../util/hist.h"
+#include "../../util/pstack.h"
+#include "../../util/sort.h"
+#include "../../util/util.h"
+#include "../../util/top.h"
+#include "../../arch/common.h"
+
+#include "../browser.h"
+#include "../helpline.h"
+#include "../util.h"
+#include "../ui.h"
+#include "map.h"
+#include "annotate.h"
+
+struct hist_browser {
+	struct ui_browser   b;
+	struct hists	    *hists;
+	struct hist_entry   *he_selection;
+	struct map_symbol   *selection;
+	int		     print_seq;
+	bool		     show_dso;
+	bool		     show_headers;
+	float		     min_pcnt;
+	u64		     nr_non_filtered_entries;
+	u64		     nr_callchain_rows;
+};
+
+extern void hist_browser__init_hpp(void);
+
+static int hists__browser_title(struct hists *hists,
+				struct hist_browser_timer *hbt,
+				char *bf, size_t size);
+static void hist_browser__update_nr_entries(struct hist_browser *hb);
+
+static struct rb_node *hists__filter_entries(struct rb_node *nd,
+					     float min_pcnt);
+
+static bool hist_browser__has_filter(struct hist_browser *hb)
+{
+	return hists__has_filter(hb->hists) || hb->min_pcnt;
+}
+
+static int hist_browser__get_folding(struct hist_browser *browser)
+{
+	struct rb_node *nd;
+	struct hists *hists = browser->hists;
+	int unfolded_rows = 0;
+
+	for (nd = rb_first(&hists->entries);
+	     (nd = hists__filter_entries(nd, browser->min_pcnt)) != NULL;
+	     nd = rb_next(nd)) {
+		struct hist_entry *he =
+			rb_entry(nd, struct hist_entry, rb_node);
+
+		if (he->ms.unfolded)
+			unfolded_rows += he->nr_rows;
+	}
+	return unfolded_rows;
+}
+
+static u32 hist_browser__nr_entries(struct hist_browser *hb)
+{
+	u32 nr_entries;
+
+	if (hist_browser__has_filter(hb))
+		nr_entries = hb->nr_non_filtered_entries;
+	else
+		nr_entries = hb->hists->nr_entries;
+
+	hb->nr_callchain_rows = hist_browser__get_folding(hb);
+	return nr_entries + hb->nr_callchain_rows;
+}
+
+static void hist_browser__update_rows(struct hist_browser *hb)
+{
+	struct ui_browser *browser = &hb->b;
+	u16 header_offset = hb->show_headers ? 1 : 0, index_row;
+
+	browser->rows = browser->height - header_offset;
+	/*
+	 * Verify if we were at the last line and that line isn't
+	 * visibe because we now show the header line(s).
+	 */
+	index_row = browser->index - browser->top_idx;
+	if (index_row >= browser->rows)
+		browser->index -= index_row - browser->rows + 1;
+}
+
+static void hist_browser__refresh_dimensions(struct ui_browser *browser)
+{
+	struct hist_browser *hb = container_of(browser, struct hist_browser, b);
+
+	/* 3 == +/- toggle symbol before actual hist_entry rendering */
+	browser->width = 3 + (hists__sort_list_width(hb->hists) + sizeof("[k]"));
+	/*
+ 	 * FIXME: Just keeping existing behaviour, but this really should be
+ 	 *	  before updating browser->width, as it will invalidate the
+ 	 *	  calculation above. Fix this and the fallout in another
+ 	 *	  changeset.
+ 	 */
+	ui_browser__refresh_dimensions(browser);
+	hist_browser__update_rows(hb);
+}
+
+static void hist_browser__gotorc(struct hist_browser *browser, int row, int column)
+{
+	u16 header_offset = browser->show_headers ? 1 : 0;
+
+	ui_browser__gotorc(&browser->b, row + header_offset, column);
+}
+
+static void hist_browser__reset(struct hist_browser *browser)
+{
+	/*
+	 * The hists__remove_entry_filter() already folds non-filtered
+	 * entries so we can assume it has 0 callchain rows.
+	 */
+	browser->nr_callchain_rows = 0;
+
+	hist_browser__update_nr_entries(browser);
+	browser->b.nr_entries = hist_browser__nr_entries(browser);
+	hist_browser__refresh_dimensions(&browser->b);
+	ui_browser__reset_index(&browser->b);
+}
+
+static char tree__folded_sign(bool unfolded)
+{
+	return unfolded ? '-' : '+';
+}
+
+static char map_symbol__folded(const struct map_symbol *ms)
+{
+	return ms->has_children ? tree__folded_sign(ms->unfolded) : ' ';
+}
+
+static char hist_entry__folded(const struct hist_entry *he)
+{
+	return map_symbol__folded(&he->ms);
+}
+
+static char callchain_list__folded(const struct callchain_list *cl)
+{
+	return map_symbol__folded(&cl->ms);
+}
+
+static void map_symbol__set_folding(struct map_symbol *ms, bool unfold)
+{
+	ms->unfolded = unfold ? ms->has_children : false;
+}
+
+static int callchain_node__count_rows_rb_tree(struct callchain_node *node)
+{
+	int n = 0;
+	struct rb_node *nd;
+
+	for (nd = rb_first(&node->rb_root); nd; nd = rb_next(nd)) {
+		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+		struct callchain_list *chain;
+		char folded_sign = ' '; /* No children */
+
+		list_for_each_entry(chain, &child->val, list) {
+			++n;
+			/* We need this because we may not have children */
+			folded_sign = callchain_list__folded(chain);
+			if (folded_sign == '+')
+				break;
+		}
+
+		if (folded_sign == '-') /* Have children and they're unfolded */
+			n += callchain_node__count_rows_rb_tree(child);
+	}
+
+	return n;
+}
+
+static int callchain_node__count_rows(struct callchain_node *node)
+{
+	struct callchain_list *chain;
+	bool unfolded = false;
+	int n = 0;
+
+	list_for_each_entry(chain, &node->val, list) {
+		++n;
+		unfolded = chain->ms.unfolded;
+	}
+
+	if (unfolded)
+		n += callchain_node__count_rows_rb_tree(node);
+
+	return n;
+}
+
+static int callchain__count_rows(struct rb_root *chain)
+{
+	struct rb_node *nd;
+	int n = 0;
+
+	for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+		struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+		n += callchain_node__count_rows(node);
+	}
+
+	return n;
+}
+
+static bool map_symbol__toggle_fold(struct map_symbol *ms)
+{
+	if (!ms)
+		return false;
+
+	if (!ms->has_children)
+		return false;
+
+	ms->unfolded = !ms->unfolded;
+	return true;
+}
+
+static void callchain_node__init_have_children_rb_tree(struct callchain_node *node)
+{
+	struct rb_node *nd = rb_first(&node->rb_root);
+
+	for (nd = rb_first(&node->rb_root); nd; nd = rb_next(nd)) {
+		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+		struct callchain_list *chain;
+		bool first = true;
+
+		list_for_each_entry(chain, &child->val, list) {
+			if (first) {
+				first = false;
+				chain->ms.has_children = chain->list.next != &child->val ||
+							 !RB_EMPTY_ROOT(&child->rb_root);
+			} else
+				chain->ms.has_children = chain->list.next == &child->val &&
+							 !RB_EMPTY_ROOT(&child->rb_root);
+		}
+
+		callchain_node__init_have_children_rb_tree(child);
+	}
+}
+
+static void callchain_node__init_have_children(struct callchain_node *node,
+					       bool has_sibling)
+{
+	struct callchain_list *chain;
+
+	chain = list_entry(node->val.next, struct callchain_list, list);
+	chain->ms.has_children = has_sibling;
+
+	if (!list_empty(&node->val)) {
+		chain = list_entry(node->val.prev, struct callchain_list, list);
+		chain->ms.has_children = !RB_EMPTY_ROOT(&node->rb_root);
+	}
+
+	callchain_node__init_have_children_rb_tree(node);
+}
+
+static void callchain__init_have_children(struct rb_root *root)
+{
+	struct rb_node *nd = rb_first(root);
+	bool has_sibling = nd && rb_next(nd);
+
+	for (nd = rb_first(root); nd; nd = rb_next(nd)) {
+		struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+		callchain_node__init_have_children(node, has_sibling);
+	}
+}
+
+static void hist_entry__init_have_children(struct hist_entry *he)
+{
+	if (!he->init_have_children) {
+		he->ms.has_children = !RB_EMPTY_ROOT(&he->sorted_chain);
+		callchain__init_have_children(&he->sorted_chain);
+		he->init_have_children = true;
+	}
+}
+
+static bool hist_browser__toggle_fold(struct hist_browser *browser)
+{
+	if (map_symbol__toggle_fold(browser->selection)) {
+		struct hist_entry *he = browser->he_selection;
+
+		hist_entry__init_have_children(he);
+		browser->b.nr_entries -= he->nr_rows;
+		browser->nr_callchain_rows -= he->nr_rows;
+
+		if (he->ms.unfolded)
+			he->nr_rows = callchain__count_rows(&he->sorted_chain);
+		else
+			he->nr_rows = 0;
+
+		browser->b.nr_entries += he->nr_rows;
+		browser->nr_callchain_rows += he->nr_rows;
+
+		return true;
+	}
+
+	/* If it doesn't have children, no toggling performed */
+	return false;
+}
+
+static int callchain_node__set_folding_rb_tree(struct callchain_node *node, bool unfold)
+{
+	int n = 0;
+	struct rb_node *nd;
+
+	for (nd = rb_first(&node->rb_root); nd; nd = rb_next(nd)) {
+		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+		struct callchain_list *chain;
+		bool has_children = false;
+
+		list_for_each_entry(chain, &child->val, list) {
+			++n;
+			map_symbol__set_folding(&chain->ms, unfold);
+			has_children = chain->ms.has_children;
+		}
+
+		if (has_children)
+			n += callchain_node__set_folding_rb_tree(child, unfold);
+	}
+
+	return n;
+}
+
+static int callchain_node__set_folding(struct callchain_node *node, bool unfold)
+{
+	struct callchain_list *chain;
+	bool has_children = false;
+	int n = 0;
+
+	list_for_each_entry(chain, &node->val, list) {
+		++n;
+		map_symbol__set_folding(&chain->ms, unfold);
+		has_children = chain->ms.has_children;
+	}
+
+	if (has_children)
+		n += callchain_node__set_folding_rb_tree(node, unfold);
+
+	return n;
+}
+
+static int callchain__set_folding(struct rb_root *chain, bool unfold)
+{
+	struct rb_node *nd;
+	int n = 0;
+
+	for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+		struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+		n += callchain_node__set_folding(node, unfold);
+	}
+
+	return n;
+}
+
+static void hist_entry__set_folding(struct hist_entry *he, bool unfold)
+{
+	hist_entry__init_have_children(he);
+	map_symbol__set_folding(&he->ms, unfold);
+
+	if (he->ms.has_children) {
+		int n = callchain__set_folding(&he->sorted_chain, unfold);
+		he->nr_rows = unfold ? n : 0;
+	} else
+		he->nr_rows = 0;
+}
+
+static void
+__hist_browser__set_folding(struct hist_browser *browser, bool unfold)
+{
+	struct rb_node *nd;
+	struct hists *hists = browser->hists;
+
+	for (nd = rb_first(&hists->entries);
+	     (nd = hists__filter_entries(nd, browser->min_pcnt)) != NULL;
+	     nd = rb_next(nd)) {
+		struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node);
+		hist_entry__set_folding(he, unfold);
+		browser->nr_callchain_rows += he->nr_rows;
+	}
+}
+
+static void hist_browser__set_folding(struct hist_browser *browser, bool unfold)
+{
+	browser->nr_callchain_rows = 0;
+	__hist_browser__set_folding(browser, unfold);
+
+	browser->b.nr_entries = hist_browser__nr_entries(browser);
+	/* Go to the start, we may be way after valid entries after a collapse */
+	ui_browser__reset_index(&browser->b);
+}
+
+static void ui_browser__warn_lost_events(struct ui_browser *browser)
+{
+	ui_browser__warning(browser, 4,
+		"Events are being lost, check IO/CPU overload!\n\n"
+		"You may want to run 'perf' using a RT scheduler policy:\n\n"
+		" perf top -r 80\n\n"
+		"Or reduce the sampling frequency.");
+}
+
+static int hist_browser__run(struct hist_browser *browser,
+			     struct hist_browser_timer *hbt)
+{
+	int key;
+	char title[160];
+	int delay_secs = hbt ? hbt->refresh : 0;
+
+	browser->b.entries = &browser->hists->entries;
+	browser->b.nr_entries = hist_browser__nr_entries(browser);
+
+	hists__browser_title(browser->hists, hbt, title, sizeof(title));
+
+	if (ui_browser__show(&browser->b, title,
+			     "Press '?' for help on key bindings") < 0)
+		return -1;
+
+	while (1) {
+		key = ui_browser__run(&browser->b, delay_secs);
+
+		switch (key) {
+		case K_TIMER: {
+			u64 nr_entries;
+			hbt->timer(hbt->arg);
+
+			if (hist_browser__has_filter(browser))
+				hist_browser__update_nr_entries(browser);
+
+			nr_entries = hist_browser__nr_entries(browser);
+			ui_browser__update_nr_entries(&browser->b, nr_entries);
+
+			if (browser->hists->stats.nr_lost_warned !=
+			    browser->hists->stats.nr_events[PERF_RECORD_LOST]) {
+				browser->hists->stats.nr_lost_warned =
+					browser->hists->stats.nr_events[PERF_RECORD_LOST];
+				ui_browser__warn_lost_events(&browser->b);
+			}
+
+			hists__browser_title(browser->hists,
+					     hbt, title, sizeof(title));
+			ui_browser__show_title(&browser->b, title);
+			continue;
+		}
+		case 'D': { /* Debug */
+			static int seq;
+			struct hist_entry *h = rb_entry(browser->b.top,
+							struct hist_entry, rb_node);
+			ui_helpline__pop();
+			ui_helpline__fpush("%d: nr_ent=(%d,%d), rows=%d, idx=%d, fve: idx=%d, row_off=%d, nrows=%d",
+					   seq++, browser->b.nr_entries,
+					   browser->hists->nr_entries,
+					   browser->b.rows,
+					   browser->b.index,
+					   browser->b.top_idx,
+					   h->row_offset, h->nr_rows);
+		}
+			break;
+		case 'C':
+			/* Collapse the whole world. */
+			hist_browser__set_folding(browser, false);
+			break;
+		case 'E':
+			/* Expand the whole world. */
+			hist_browser__set_folding(browser, true);
+			break;
+		case 'H':
+			browser->show_headers = !browser->show_headers;
+			hist_browser__update_rows(browser);
+			break;
+		case K_ENTER:
+			if (hist_browser__toggle_fold(browser))
+				break;
+			/* fall thru */
+		default:
+			goto out;
+		}
+	}
+out:
+	ui_browser__hide(&browser->b);
+	return key;
+}
+
+struct callchain_print_arg {
+	/* for hists browser */
+	off_t	row_offset;
+	bool	is_current_entry;
+
+	/* for file dump */
+	FILE	*fp;
+	int	printed;
+};
+
+typedef void (*print_callchain_entry_fn)(struct hist_browser *browser,
+					 struct callchain_list *chain,
+					 const char *str, int offset,
+					 unsigned short row,
+					 struct callchain_print_arg *arg);
+
+static void hist_browser__show_callchain_entry(struct hist_browser *browser,
+					       struct callchain_list *chain,
+					       const char *str, int offset,
+					       unsigned short row,
+					       struct callchain_print_arg *arg)
+{
+	int color, width;
+	char folded_sign = callchain_list__folded(chain);
+	bool show_annotated = browser->show_dso && chain->ms.sym && symbol__annotation(chain->ms.sym)->src;
+
+	color = HE_COLORSET_NORMAL;
+	width = browser->b.width - (offset + 2);
+	if (ui_browser__is_current_entry(&browser->b, row)) {
+		browser->selection = &chain->ms;
+		color = HE_COLORSET_SELECTED;
+		arg->is_current_entry = true;
+	}
+
+	ui_browser__set_color(&browser->b, color);
+	hist_browser__gotorc(browser, row, 0);
+	slsmg_write_nstring(" ", offset);
+	slsmg_printf("%c", folded_sign);
+	ui_browser__write_graph(&browser->b, show_annotated ? SLSMG_RARROW_CHAR : ' ');
+	slsmg_write_nstring(str, width);
+}
+
+static void hist_browser__fprintf_callchain_entry(struct hist_browser *b __maybe_unused,
+						  struct callchain_list *chain,
+						  const char *str, int offset,
+						  unsigned short row __maybe_unused,
+						  struct callchain_print_arg *arg)
+{
+	char folded_sign = callchain_list__folded(chain);
+
+	arg->printed += fprintf(arg->fp, "%*s%c %s\n", offset, " ",
+				folded_sign, str);
+}
+
+typedef bool (*check_output_full_fn)(struct hist_browser *browser,
+				     unsigned short row);
+
+static bool hist_browser__check_output_full(struct hist_browser *browser,
+					    unsigned short row)
+{
+	return browser->b.rows == row;
+}
+
+static bool hist_browser__check_dump_full(struct hist_browser *browser __maybe_unused,
+					  unsigned short row __maybe_unused)
+{
+	return false;
+}
+
+#define LEVEL_OFFSET_STEP 3
+
+static int hist_browser__show_callchain(struct hist_browser *browser,
+					struct rb_root *root, int level,
+					unsigned short row, u64 total,
+					print_callchain_entry_fn print,
+					struct callchain_print_arg *arg,
+					check_output_full_fn is_output_full)
+{
+	struct rb_node *node;
+	int first_row = row, offset = level * LEVEL_OFFSET_STEP;
+	u64 new_total;
+	bool need_percent;
+
+	node = rb_first(root);
+	need_percent = node && rb_next(node);
+
+	while (node) {
+		struct callchain_node *child = rb_entry(node, struct callchain_node, rb_node);
+		struct rb_node *next = rb_next(node);
+		u64 cumul = callchain_cumul_hits(child);
+		struct callchain_list *chain;
+		char folded_sign = ' ';
+		int first = true;
+		int extra_offset = 0;
+
+		list_for_each_entry(chain, &child->val, list) {
+			char bf[1024], *alloc_str;
+			const char *str;
+			bool was_first = first;
+
+			if (first)
+				first = false;
+			else if (need_percent)
+				extra_offset = LEVEL_OFFSET_STEP;
+
+			folded_sign = callchain_list__folded(chain);
+			if (arg->row_offset != 0) {
+				arg->row_offset--;
+				goto do_next;
+			}
+
+			alloc_str = NULL;
+			str = callchain_list__sym_name(chain, bf, sizeof(bf),
+						       browser->show_dso);
+
+			if (was_first && need_percent) {
+				double percent = cumul * 100.0 / total;
+
+				if (asprintf(&alloc_str, "%2.2f%% %s", percent, str) < 0)
+					str = "Not enough memory!";
+				else
+					str = alloc_str;
+			}
+
+			print(browser, chain, str, offset + extra_offset, row, arg);
+
+			free(alloc_str);
+
+			if (is_output_full(browser, ++row))
+				goto out;
+do_next:
+			if (folded_sign == '+')
+				break;
+		}
+
+		if (folded_sign == '-') {
+			const int new_level = level + (extra_offset ? 2 : 1);
+
+			if (callchain_param.mode == CHAIN_GRAPH_REL)
+				new_total = child->children_hit;
+			else
+				new_total = total;
+
+			row += hist_browser__show_callchain(browser, &child->rb_root,
+							    new_level, row, new_total,
+							    print, arg, is_output_full);
+		}
+		if (is_output_full(browser, row))
+			break;
+		node = next;
+	}
+out:
+	return row - first_row;
+}
+
+struct hpp_arg {
+	struct ui_browser *b;
+	char folded_sign;
+	bool current_entry;
+};
+
+static int __hpp__slsmg_color_printf(struct perf_hpp *hpp, const char *fmt, ...)
+{
+	struct hpp_arg *arg = hpp->ptr;
+	int ret, len;
+	va_list args;
+	double percent;
+
+	va_start(args, fmt);
+	len = va_arg(args, int);
+	percent = va_arg(args, double);
+	va_end(args);
+
+	ui_browser__set_percent_color(arg->b, percent, arg->current_entry);
+
+	ret = scnprintf(hpp->buf, hpp->size, fmt, len, percent);
+	slsmg_printf("%s", hpp->buf);
+
+	advance_hpp(hpp, ret);
+	return ret;
+}
+
+#define __HPP_COLOR_PERCENT_FN(_type, _field)				\
+static u64 __hpp_get_##_field(struct hist_entry *he)			\
+{									\
+	return he->stat._field;						\
+}									\
+									\
+static int								\
+hist_browser__hpp_color_##_type(struct perf_hpp_fmt *fmt,		\
+				struct perf_hpp *hpp,			\
+				struct hist_entry *he)			\
+{									\
+	return hpp__fmt(fmt, hpp, he, __hpp_get_##_field, " %*.2f%%",	\
+			__hpp__slsmg_color_printf, true);		\
+}
+
+#define __HPP_COLOR_ACC_PERCENT_FN(_type, _field)			\
+static u64 __hpp_get_acc_##_field(struct hist_entry *he)		\
+{									\
+	return he->stat_acc->_field;					\
+}									\
+									\
+static int								\
+hist_browser__hpp_color_##_type(struct perf_hpp_fmt *fmt,		\
+				struct perf_hpp *hpp,			\
+				struct hist_entry *he)			\
+{									\
+	if (!symbol_conf.cumulate_callchain) {				\
+		int len = fmt->user_len ?: fmt->len;			\
+		int ret = scnprintf(hpp->buf, hpp->size,		\
+				    "%*s", len, "N/A");			\
+		slsmg_printf("%s", hpp->buf);				\
+									\
+		return ret;						\
+	}								\
+	return hpp__fmt(fmt, hpp, he, __hpp_get_acc_##_field,		\
+			" %*.2f%%", __hpp__slsmg_color_printf, true);	\
+}
+
+__HPP_COLOR_PERCENT_FN(overhead, period)
+__HPP_COLOR_PERCENT_FN(overhead_sys, period_sys)
+__HPP_COLOR_PERCENT_FN(overhead_us, period_us)
+__HPP_COLOR_PERCENT_FN(overhead_guest_sys, period_guest_sys)
+__HPP_COLOR_PERCENT_FN(overhead_guest_us, period_guest_us)
+__HPP_COLOR_ACC_PERCENT_FN(overhead_acc, period)
+
+#undef __HPP_COLOR_PERCENT_FN
+#undef __HPP_COLOR_ACC_PERCENT_FN
+
+void hist_browser__init_hpp(void)
+{
+	perf_hpp__format[PERF_HPP__OVERHEAD].color =
+				hist_browser__hpp_color_overhead;
+	perf_hpp__format[PERF_HPP__OVERHEAD_SYS].color =
+				hist_browser__hpp_color_overhead_sys;
+	perf_hpp__format[PERF_HPP__OVERHEAD_US].color =
+				hist_browser__hpp_color_overhead_us;
+	perf_hpp__format[PERF_HPP__OVERHEAD_GUEST_SYS].color =
+				hist_browser__hpp_color_overhead_guest_sys;
+	perf_hpp__format[PERF_HPP__OVERHEAD_GUEST_US].color =
+				hist_browser__hpp_color_overhead_guest_us;
+	perf_hpp__format[PERF_HPP__OVERHEAD_ACC].color =
+				hist_browser__hpp_color_overhead_acc;
+}
+
+static int hist_browser__show_entry(struct hist_browser *browser,
+				    struct hist_entry *entry,
+				    unsigned short row)
+{
+	char s[256];
+	int printed = 0;
+	int width = browser->b.width;
+	char folded_sign = ' ';
+	bool current_entry = ui_browser__is_current_entry(&browser->b, row);
+	off_t row_offset = entry->row_offset;
+	bool first = true;
+	struct perf_hpp_fmt *fmt;
+
+	if (current_entry) {
+		browser->he_selection = entry;
+		browser->selection = &entry->ms;
+	}
+
+	if (symbol_conf.use_callchain) {
+		hist_entry__init_have_children(entry);
+		folded_sign = hist_entry__folded(entry);
+	}
+
+	if (row_offset == 0) {
+		struct hpp_arg arg = {
+			.b		= &browser->b,
+			.folded_sign	= folded_sign,
+			.current_entry	= current_entry,
+		};
+		struct perf_hpp hpp = {
+			.buf		= s,
+			.size		= sizeof(s),
+			.ptr		= &arg,
+		};
+
+		hist_browser__gotorc(browser, row, 0);
+
+		perf_hpp__for_each_format(fmt) {
+			if (perf_hpp__should_skip(fmt))
+				continue;
+
+			if (current_entry && browser->b.navkeypressed) {
+				ui_browser__set_color(&browser->b,
+						      HE_COLORSET_SELECTED);
+			} else {
+				ui_browser__set_color(&browser->b,
+						      HE_COLORSET_NORMAL);
+			}
+
+			if (first) {
+				if (symbol_conf.use_callchain) {
+					slsmg_printf("%c ", folded_sign);
+					width -= 2;
+				}
+				first = false;
+			} else {
+				slsmg_printf("  ");
+				width -= 2;
+			}
+
+			if (fmt->color) {
+				width -= fmt->color(fmt, &hpp, entry);
+			} else {
+				width -= fmt->entry(fmt, &hpp, entry);
+				slsmg_printf("%s", s);
+			}
+		}
+
+		/* The scroll bar isn't being used */
+		if (!browser->b.navkeypressed)
+			width += 1;
+
+		slsmg_write_nstring("", width);
+
+		++row;
+		++printed;
+	} else
+		--row_offset;
+
+	if (folded_sign == '-' && row != browser->b.rows) {
+		u64 total = hists__total_period(entry->hists);
+		struct callchain_print_arg arg = {
+			.row_offset = row_offset,
+			.is_current_entry = current_entry,
+		};
+
+		if (callchain_param.mode == CHAIN_GRAPH_REL) {
+			if (symbol_conf.cumulate_callchain)
+				total = entry->stat_acc->period;
+			else
+				total = entry->stat.period;
+		}
+
+		printed += hist_browser__show_callchain(browser,
+					&entry->sorted_chain, 1, row, total,
+					hist_browser__show_callchain_entry, &arg,
+					hist_browser__check_output_full);
+
+		if (arg.is_current_entry)
+			browser->he_selection = entry;
+	}
+
+	return printed;
+}
+
+static int advance_hpp_check(struct perf_hpp *hpp, int inc)
+{
+	advance_hpp(hpp, inc);
+	return hpp->size <= 0;
+}
+
+static int hists__scnprintf_headers(char *buf, size_t size, struct hists *hists)
+{
+	struct perf_hpp dummy_hpp = {
+		.buf    = buf,
+		.size   = size,
+	};
+	struct perf_hpp_fmt *fmt;
+	size_t ret = 0;
+
+	if (symbol_conf.use_callchain) {
+		ret = scnprintf(buf, size, "  ");
+		if (advance_hpp_check(&dummy_hpp, ret))
+			return ret;
+	}
+
+	perf_hpp__for_each_format(fmt) {
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		ret = fmt->header(fmt, &dummy_hpp, hists_to_evsel(hists));
+		if (advance_hpp_check(&dummy_hpp, ret))
+			break;
+
+		ret = scnprintf(dummy_hpp.buf, dummy_hpp.size, "  ");
+		if (advance_hpp_check(&dummy_hpp, ret))
+			break;
+	}
+
+	return ret;
+}
+
+static void hist_browser__show_headers(struct hist_browser *browser)
+{
+	char headers[1024];
+
+	hists__scnprintf_headers(headers, sizeof(headers), browser->hists);
+	ui_browser__gotorc(&browser->b, 0, 0);
+	ui_browser__set_color(&browser->b, HE_COLORSET_ROOT);
+	slsmg_write_nstring(headers, browser->b.width + 1);
+}
+
+static void ui_browser__hists_init_top(struct ui_browser *browser)
+{
+	if (browser->top == NULL) {
+		struct hist_browser *hb;
+
+		hb = container_of(browser, struct hist_browser, b);
+		browser->top = rb_first(&hb->hists->entries);
+	}
+}
+
+static unsigned int hist_browser__refresh(struct ui_browser *browser)
+{
+	unsigned row = 0;
+	u16 header_offset = 0;
+	struct rb_node *nd;
+	struct hist_browser *hb = container_of(browser, struct hist_browser, b);
+
+	if (hb->show_headers) {
+		hist_browser__show_headers(hb);
+		header_offset = 1;
+	}
+
+	ui_browser__hists_init_top(browser);
+
+	for (nd = browser->top; nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+		float percent;
+
+		if (h->filtered)
+			continue;
+
+		percent = hist_entry__get_percent_limit(h);
+		if (percent < hb->min_pcnt)
+			continue;
+
+		row += hist_browser__show_entry(hb, h, row);
+		if (row == browser->rows)
+			break;
+	}
+
+	return row + header_offset;
+}
+
+static struct rb_node *hists__filter_entries(struct rb_node *nd,
+					     float min_pcnt)
+{
+	while (nd != NULL) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+		float percent = hist_entry__get_percent_limit(h);
+
+		if (!h->filtered && percent >= min_pcnt)
+			return nd;
+
+		nd = rb_next(nd);
+	}
+
+	return NULL;
+}
+
+static struct rb_node *hists__filter_prev_entries(struct rb_node *nd,
+						  float min_pcnt)
+{
+	while (nd != NULL) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+		float percent = hist_entry__get_percent_limit(h);
+
+		if (!h->filtered && percent >= min_pcnt)
+			return nd;
+
+		nd = rb_prev(nd);
+	}
+
+	return NULL;
+}
+
+static void ui_browser__hists_seek(struct ui_browser *browser,
+				   off_t offset, int whence)
+{
+	struct hist_entry *h;
+	struct rb_node *nd;
+	bool first = true;
+	struct hist_browser *hb;
+
+	hb = container_of(browser, struct hist_browser, b);
+
+	if (browser->nr_entries == 0)
+		return;
+
+	ui_browser__hists_init_top(browser);
+
+	switch (whence) {
+	case SEEK_SET:
+		nd = hists__filter_entries(rb_first(browser->entries),
+					   hb->min_pcnt);
+		break;
+	case SEEK_CUR:
+		nd = browser->top;
+		goto do_offset;
+	case SEEK_END:
+		nd = hists__filter_prev_entries(rb_last(browser->entries),
+						hb->min_pcnt);
+		first = false;
+		break;
+	default:
+		return;
+	}
+
+	/*
+	 * Moves not relative to the first visible entry invalidates its
+	 * row_offset:
+	 */
+	h = rb_entry(browser->top, struct hist_entry, rb_node);
+	h->row_offset = 0;
+
+	/*
+	 * Here we have to check if nd is expanded (+), if it is we can't go
+	 * the next top level hist_entry, instead we must compute an offset of
+	 * what _not_ to show and not change the first visible entry.
+	 *
+	 * This offset increments when we are going from top to bottom and
+	 * decreases when we're going from bottom to top.
+	 *
+	 * As we don't have backpointers to the top level in the callchains
+	 * structure, we need to always print the whole hist_entry callchain,
+	 * skipping the first ones that are before the first visible entry
+	 * and stop when we printed enough lines to fill the screen.
+	 */
+do_offset:
+	if (offset > 0) {
+		do {
+			h = rb_entry(nd, struct hist_entry, rb_node);
+			if (h->ms.unfolded) {
+				u16 remaining = h->nr_rows - h->row_offset;
+				if (offset > remaining) {
+					offset -= remaining;
+					h->row_offset = 0;
+				} else {
+					h->row_offset += offset;
+					offset = 0;
+					browser->top = nd;
+					break;
+				}
+			}
+			nd = hists__filter_entries(rb_next(nd), hb->min_pcnt);
+			if (nd == NULL)
+				break;
+			--offset;
+			browser->top = nd;
+		} while (offset != 0);
+	} else if (offset < 0) {
+		while (1) {
+			h = rb_entry(nd, struct hist_entry, rb_node);
+			if (h->ms.unfolded) {
+				if (first) {
+					if (-offset > h->row_offset) {
+						offset += h->row_offset;
+						h->row_offset = 0;
+					} else {
+						h->row_offset += offset;
+						offset = 0;
+						browser->top = nd;
+						break;
+					}
+				} else {
+					if (-offset > h->nr_rows) {
+						offset += h->nr_rows;
+						h->row_offset = 0;
+					} else {
+						h->row_offset = h->nr_rows + offset;
+						offset = 0;
+						browser->top = nd;
+						break;
+					}
+				}
+			}
+
+			nd = hists__filter_prev_entries(rb_prev(nd),
+							hb->min_pcnt);
+			if (nd == NULL)
+				break;
+			++offset;
+			browser->top = nd;
+			if (offset == 0) {
+				/*
+				 * Last unfiltered hist_entry, check if it is
+				 * unfolded, if it is then we should have
+				 * row_offset at its last entry.
+				 */
+				h = rb_entry(nd, struct hist_entry, rb_node);
+				if (h->ms.unfolded)
+					h->row_offset = h->nr_rows;
+				break;
+			}
+			first = false;
+		}
+	} else {
+		browser->top = nd;
+		h = rb_entry(nd, struct hist_entry, rb_node);
+		h->row_offset = 0;
+	}
+}
+
+static int hist_browser__fprintf_callchain(struct hist_browser *browser,
+					   struct hist_entry *he, FILE *fp)
+{
+	u64 total = hists__total_period(he->hists);
+	struct callchain_print_arg arg  = {
+		.fp = fp,
+	};
+
+	if (symbol_conf.cumulate_callchain)
+		total = he->stat_acc->period;
+
+	hist_browser__show_callchain(browser, &he->sorted_chain, 1, 0, total,
+				     hist_browser__fprintf_callchain_entry, &arg,
+				     hist_browser__check_dump_full);
+	return arg.printed;
+}
+
+static int hist_browser__fprintf_entry(struct hist_browser *browser,
+				       struct hist_entry *he, FILE *fp)
+{
+	char s[8192];
+	int printed = 0;
+	char folded_sign = ' ';
+	struct perf_hpp hpp = {
+		.buf = s,
+		.size = sizeof(s),
+	};
+	struct perf_hpp_fmt *fmt;
+	bool first = true;
+	int ret;
+
+	if (symbol_conf.use_callchain)
+		folded_sign = hist_entry__folded(he);
+
+	if (symbol_conf.use_callchain)
+		printed += fprintf(fp, "%c ", folded_sign);
+
+	perf_hpp__for_each_format(fmt) {
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		if (!first) {
+			ret = scnprintf(hpp.buf, hpp.size, "  ");
+			advance_hpp(&hpp, ret);
+		} else
+			first = false;
+
+		ret = fmt->entry(fmt, &hpp, he);
+		advance_hpp(&hpp, ret);
+	}
+	printed += fprintf(fp, "%s\n", rtrim(s));
+
+	if (folded_sign == '-')
+		printed += hist_browser__fprintf_callchain(browser, he, fp);
+
+	return printed;
+}
+
+static int hist_browser__fprintf(struct hist_browser *browser, FILE *fp)
+{
+	struct rb_node *nd = hists__filter_entries(rb_first(browser->b.entries),
+						   browser->min_pcnt);
+	int printed = 0;
+
+	while (nd) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+		printed += hist_browser__fprintf_entry(browser, h, fp);
+		nd = hists__filter_entries(rb_next(nd), browser->min_pcnt);
+	}
+
+	return printed;
+}
+
+static int hist_browser__dump(struct hist_browser *browser)
+{
+	char filename[64];
+	FILE *fp;
+
+	while (1) {
+		scnprintf(filename, sizeof(filename), "perf.hist.%d", browser->print_seq);
+		if (access(filename, F_OK))
+			break;
+		/*
+ 		 * XXX: Just an arbitrary lazy upper limit
+ 		 */
+		if (++browser->print_seq == 8192) {
+			ui_helpline__fpush("Too many perf.hist.N files, nothing written!");
+			return -1;
+		}
+	}
+
+	fp = fopen(filename, "w");
+	if (fp == NULL) {
+		char bf[64];
+		const char *err = strerror_r(errno, bf, sizeof(bf));
+		ui_helpline__fpush("Couldn't write to %s: %s", filename, err);
+		return -1;
+	}
+
+	++browser->print_seq;
+	hist_browser__fprintf(browser, fp);
+	fclose(fp);
+	ui_helpline__fpush("%s written!", filename);
+
+	return 0;
+}
+
+static struct hist_browser *hist_browser__new(struct hists *hists)
+{
+	struct hist_browser *browser = zalloc(sizeof(*browser));
+
+	if (browser) {
+		browser->hists = hists;
+		browser->b.refresh = hist_browser__refresh;
+		browser->b.refresh_dimensions = hist_browser__refresh_dimensions;
+		browser->b.seek = ui_browser__hists_seek;
+		browser->b.use_navkeypressed = true;
+		browser->show_headers = symbol_conf.show_hist_headers;
+	}
+
+	return browser;
+}
+
+static void hist_browser__delete(struct hist_browser *browser)
+{
+	free(browser);
+}
+
+static struct hist_entry *hist_browser__selected_entry(struct hist_browser *browser)
+{
+	return browser->he_selection;
+}
+
+static struct thread *hist_browser__selected_thread(struct hist_browser *browser)
+{
+	return browser->he_selection->thread;
+}
+
+/* Check whether the browser is for 'top' or 'report' */
+static inline bool is_report_browser(void *timer)
+{
+	return timer == NULL;
+}
+
+static int hists__browser_title(struct hists *hists,
+				struct hist_browser_timer *hbt,
+				char *bf, size_t size)
+{
+	char unit;
+	int printed;
+	const struct dso *dso = hists->dso_filter;
+	const struct thread *thread = hists->thread_filter;
+	unsigned long nr_samples = hists->stats.nr_events[PERF_RECORD_SAMPLE];
+	u64 nr_events = hists->stats.total_period;
+	struct perf_evsel *evsel = hists_to_evsel(hists);
+	const char *ev_name = perf_evsel__name(evsel);
+	char buf[512];
+	size_t buflen = sizeof(buf);
+
+	if (symbol_conf.filter_relative) {
+		nr_samples = hists->stats.nr_non_filtered_samples;
+		nr_events = hists->stats.total_non_filtered_period;
+	}
+
+	if (perf_evsel__is_group_event(evsel)) {
+		struct perf_evsel *pos;
+
+		perf_evsel__group_desc(evsel, buf, buflen);
+		ev_name = buf;
+
+		for_each_group_member(pos, evsel) {
+			struct hists *pos_hists = evsel__hists(pos);
+
+			if (symbol_conf.filter_relative) {
+				nr_samples += pos_hists->stats.nr_non_filtered_samples;
+				nr_events += pos_hists->stats.total_non_filtered_period;
+			} else {
+				nr_samples += pos_hists->stats.nr_events[PERF_RECORD_SAMPLE];
+				nr_events += pos_hists->stats.total_period;
+			}
+		}
+	}
+
+	nr_samples = convert_unit(nr_samples, &unit);
+	printed = scnprintf(bf, size,
+			   "Samples: %lu%c of event '%s', Event count (approx.): %" PRIu64,
+			   nr_samples, unit, ev_name, nr_events);
+
+
+	if (hists->uid_filter_str)
+		printed += snprintf(bf + printed, size - printed,
+				    ", UID: %s", hists->uid_filter_str);
+	if (thread)
+		printed += scnprintf(bf + printed, size - printed,
+				    ", Thread: %s(%d)",
+				     (thread->comm_set ? thread__comm_str(thread) : ""),
+				    thread->tid);
+	if (dso)
+		printed += scnprintf(bf + printed, size - printed,
+				    ", DSO: %s", dso->short_name);
+	if (!is_report_browser(hbt)) {
+		struct perf_top *top = hbt->arg;
+
+		if (top->zero)
+			printed += scnprintf(bf + printed, size - printed, " [z]");
+	}
+
+	return printed;
+}
+
+static inline void free_popup_options(char **options, int n)
+{
+	int i;
+
+	for (i = 0; i < n; ++i)
+		zfree(&options[i]);
+}
+
+/*
+ * Only runtime switching of perf data file will make "input_name" point
+ * to a malloced buffer. So add "is_input_name_malloced" flag to decide
+ * whether we need to call free() for current "input_name" during the switch.
+ */
+static bool is_input_name_malloced = false;
+
+static int switch_data_file(void)
+{
+	char *pwd, *options[32], *abs_path[32], *tmp;
+	DIR *pwd_dir;
+	int nr_options = 0, choice = -1, ret = -1;
+	struct dirent *dent;
+
+	pwd = getenv("PWD");
+	if (!pwd)
+		return ret;
+
+	pwd_dir = opendir(pwd);
+	if (!pwd_dir)
+		return ret;
+
+	memset(options, 0, sizeof(options));
+	memset(options, 0, sizeof(abs_path));
+
+	while ((dent = readdir(pwd_dir))) {
+		char path[PATH_MAX];
+		u64 magic;
+		char *name = dent->d_name;
+		FILE *file;
+
+		if (!(dent->d_type == DT_REG))
+			continue;
+
+		snprintf(path, sizeof(path), "%s/%s", pwd, name);
+
+		file = fopen(path, "r");
+		if (!file)
+			continue;
+
+		if (fread(&magic, 1, 8, file) < 8)
+			goto close_file_and_continue;
+
+		if (is_perf_magic(magic)) {
+			options[nr_options] = strdup(name);
+			if (!options[nr_options])
+				goto close_file_and_continue;
+
+			abs_path[nr_options] = strdup(path);
+			if (!abs_path[nr_options]) {
+				zfree(&options[nr_options]);
+				ui__warning("Can't search all data files due to memory shortage.\n");
+				fclose(file);
+				break;
+			}
+
+			nr_options++;
+		}
+
+close_file_and_continue:
+		fclose(file);
+		if (nr_options >= 32) {
+			ui__warning("Too many perf data files in PWD!\n"
+				    "Only the first 32 files will be listed.\n");
+			break;
+		}
+	}
+	closedir(pwd_dir);
+
+	if (nr_options) {
+		choice = ui__popup_menu(nr_options, options);
+		if (choice < nr_options && choice >= 0) {
+			tmp = strdup(abs_path[choice]);
+			if (tmp) {
+				if (is_input_name_malloced)
+					free((void *)input_name);
+				input_name = tmp;
+				is_input_name_malloced = true;
+				ret = 0;
+			} else
+				ui__warning("Data switch failed due to memory shortage!\n");
+		}
+	}
+
+	free_popup_options(options, nr_options);
+	free_popup_options(abs_path, nr_options);
+	return ret;
+}
+
+static void hist_browser__update_nr_entries(struct hist_browser *hb)
+{
+	u64 nr_entries = 0;
+	struct rb_node *nd = rb_first(&hb->hists->entries);
+
+	if (hb->min_pcnt == 0) {
+		hb->nr_non_filtered_entries = hb->hists->nr_non_filtered_entries;
+		return;
+	}
+
+	while ((nd = hists__filter_entries(nd, hb->min_pcnt)) != NULL) {
+		nr_entries++;
+		nd = rb_next(nd);
+	}
+
+	hb->nr_non_filtered_entries = nr_entries;
+}
+
+static int perf_evsel__hists_browse(struct perf_evsel *evsel, int nr_events,
+				    const char *helpline,
+				    bool left_exits,
+				    struct hist_browser_timer *hbt,
+				    float min_pcnt,
+				    struct perf_session_env *env)
+{
+	struct hists *hists = evsel__hists(evsel);
+	struct hist_browser *browser = hist_browser__new(hists);
+	struct branch_info *bi;
+	struct pstack *fstack;
+	char *options[16];
+	int nr_options = 0;
+	int key = -1;
+	char buf[64];
+	char script_opt[64];
+	int delay_secs = hbt ? hbt->refresh : 0;
+	struct perf_hpp_fmt *fmt;
+
+#define HIST_BROWSER_HELP_COMMON					\
+	"h/?/F1        Show this window\n"				\
+	"UP/DOWN/PGUP\n"						\
+	"PGDN/SPACE    Navigate\n"					\
+	"q/ESC/CTRL+C  Exit browser\n\n"				\
+	"For multiple event sessions:\n\n"				\
+	"TAB/UNTAB     Switch events\n\n"				\
+	"For symbolic views (--sort has sym):\n\n"			\
+	"->            Zoom into DSO/Threads & Annotate current symbol\n" \
+	"<-            Zoom out\n"					\
+	"a             Annotate current symbol\n"			\
+	"C             Collapse all callchains\n"			\
+	"d             Zoom into current DSO\n"				\
+	"E             Expand all callchains\n"				\
+	"F             Toggle percentage of filtered entries\n"		\
+	"H             Display column headers\n"			\
+
+	/* help messages are sorted by lexical order of the hotkey */
+	const char report_help[] = HIST_BROWSER_HELP_COMMON
+	"i             Show header information\n"
+	"P             Print histograms to perf.hist.N\n"
+	"r             Run available scripts\n"
+	"s             Switch to another data file in PWD\n"
+	"t             Zoom into current Thread\n"
+	"V             Verbose (DSO names in callchains, etc)\n"
+	"/             Filter symbol by name";
+	const char top_help[] = HIST_BROWSER_HELP_COMMON
+	"P             Print histograms to perf.hist.N\n"
+	"t             Zoom into current Thread\n"
+	"V             Verbose (DSO names in callchains, etc)\n"
+	"z             Toggle zeroing of samples\n"
+	"/             Filter symbol by name";
+
+	if (browser == NULL)
+		return -1;
+
+	if (min_pcnt) {
+		browser->min_pcnt = min_pcnt;
+		hist_browser__update_nr_entries(browser);
+	}
+
+	fstack = pstack__new(2);
+	if (fstack == NULL)
+		goto out;
+
+	ui_helpline__push(helpline);
+
+	memset(options, 0, sizeof(options));
+
+	perf_hpp__for_each_format(fmt)
+		perf_hpp__reset_width(fmt, hists);
+
+	if (symbol_conf.col_width_list_str)
+		perf_hpp__set_user_width(symbol_conf.col_width_list_str);
+
+	while (1) {
+		struct thread *thread = NULL;
+		const struct dso *dso = NULL;
+		int choice = 0,
+		    annotate = -2, zoom_dso = -2, zoom_thread = -2,
+		    annotate_f = -2, annotate_t = -2, browse_map = -2;
+		int scripts_comm = -2, scripts_symbol = -2,
+		    scripts_all = -2, switch_data = -2;
+
+		nr_options = 0;
+
+		key = hist_browser__run(browser, hbt);
+
+		if (browser->he_selection != NULL) {
+			thread = hist_browser__selected_thread(browser);
+			dso = browser->selection->map ? browser->selection->map->dso : NULL;
+		}
+		switch (key) {
+		case K_TAB:
+		case K_UNTAB:
+			if (nr_events == 1)
+				continue;
+			/*
+			 * Exit the browser, let hists__browser_tree
+			 * go to the next or previous
+			 */
+			goto out_free_stack;
+		case 'a':
+			if (!sort__has_sym) {
+				ui_browser__warning(&browser->b, delay_secs * 2,
+			"Annotation is only available for symbolic views, "
+			"include \"sym*\" in --sort to use it.");
+				continue;
+			}
+
+			if (browser->selection == NULL ||
+			    browser->selection->sym == NULL ||
+			    browser->selection->map->dso->annotate_warned)
+				continue;
+			goto do_annotate;
+		case 'P':
+			hist_browser__dump(browser);
+			continue;
+		case 'd':
+			goto zoom_dso;
+		case 'V':
+			browser->show_dso = !browser->show_dso;
+			continue;
+		case 't':
+			goto zoom_thread;
+		case '/':
+			if (ui_browser__input_window("Symbol to show",
+					"Please enter the name of symbol you want to see",
+					buf, "ENTER: OK, ESC: Cancel",
+					delay_secs * 2) == K_ENTER) {
+				hists->symbol_filter_str = *buf ? buf : NULL;
+				hists__filter_by_symbol(hists);
+				hist_browser__reset(browser);
+			}
+			continue;
+		case 'r':
+			if (is_report_browser(hbt))
+				goto do_scripts;
+			continue;
+		case 's':
+			if (is_report_browser(hbt))
+				goto do_data_switch;
+			continue;
+		case 'i':
+			/* env->arch is NULL for live-mode (i.e. perf top) */
+			if (env->arch)
+				tui__header_window(env);
+			continue;
+		case 'F':
+			symbol_conf.filter_relative ^= 1;
+			continue;
+		case 'z':
+			if (!is_report_browser(hbt)) {
+				struct perf_top *top = hbt->arg;
+
+				top->zero = !top->zero;
+			}
+			continue;
+		case K_F1:
+		case 'h':
+		case '?':
+			ui_browser__help_window(&browser->b,
+				is_report_browser(hbt) ? report_help : top_help);
+			continue;
+		case K_ENTER:
+		case K_RIGHT:
+			/* menu */
+			break;
+		case K_LEFT: {
+			const void *top;
+
+			if (pstack__empty(fstack)) {
+				/*
+				 * Go back to the perf_evsel_menu__run or other user
+				 */
+				if (left_exits)
+					goto out_free_stack;
+				continue;
+			}
+			top = pstack__pop(fstack);
+			if (top == &browser->hists->dso_filter)
+				goto zoom_out_dso;
+			if (top == &browser->hists->thread_filter)
+				goto zoom_out_thread;
+			continue;
+		}
+		case K_ESC:
+			if (!left_exits &&
+			    !ui_browser__dialog_yesno(&browser->b,
+					       "Do you really want to exit?"))
+				continue;
+			/* Fall thru */
+		case 'q':
+		case CTRL('c'):
+			goto out_free_stack;
+		default:
+			continue;
+		}
+
+		if (!sort__has_sym)
+			goto add_exit_option;
+
+		if (browser->selection == NULL)
+			goto skip_annotation;
+
+		if (sort__mode == SORT_MODE__BRANCH) {
+			bi = browser->he_selection->branch_info;
+
+			if (bi == NULL)
+				goto skip_annotation;
+
+			if (bi->from.sym != NULL &&
+			    !bi->from.map->dso->annotate_warned &&
+			    asprintf(&options[nr_options], "Annotate %s", bi->from.sym->name) > 0) {
+				annotate_f = nr_options++;
+			}
+
+			if (bi->to.sym != NULL &&
+			    !bi->to.map->dso->annotate_warned &&
+			    (bi->to.sym != bi->from.sym ||
+			     bi->to.map->dso != bi->from.map->dso) &&
+			    asprintf(&options[nr_options], "Annotate %s", bi->to.sym->name) > 0) {
+				annotate_t = nr_options++;
+			}
+		} else {
+			if (browser->selection->sym != NULL &&
+			    !browser->selection->map->dso->annotate_warned) {
+				struct annotation *notes;
+
+				notes = symbol__annotation(browser->selection->sym);
+
+				if (notes->src &&
+				    asprintf(&options[nr_options], "Annotate %s",
+						 browser->selection->sym->name) > 0) {
+					annotate = nr_options++;
+				}
+			}
+		}
+skip_annotation:
+		if (thread != NULL &&
+		    asprintf(&options[nr_options], "Zoom %s %s(%d) thread",
+			     (browser->hists->thread_filter ? "out of" : "into"),
+			     (thread->comm_set ? thread__comm_str(thread) : ""),
+			     thread->tid) > 0)
+			zoom_thread = nr_options++;
+
+		if (dso != NULL &&
+		    asprintf(&options[nr_options], "Zoom %s %s DSO",
+			     (browser->hists->dso_filter ? "out of" : "into"),
+			     (dso->kernel ? "the Kernel" : dso->short_name)) > 0)
+			zoom_dso = nr_options++;
+
+		if (browser->selection != NULL &&
+		    browser->selection->map != NULL &&
+		    asprintf(&options[nr_options], "Browse map details") > 0)
+			browse_map = nr_options++;
+
+		/* perf script support */
+		if (browser->he_selection) {
+			struct symbol *sym;
+
+			if (asprintf(&options[nr_options], "Run scripts for samples of thread [%s]",
+				     thread__comm_str(browser->he_selection->thread)) > 0)
+				scripts_comm = nr_options++;
+
+			sym = browser->he_selection->ms.sym;
+			if (sym && sym->namelen &&
+				asprintf(&options[nr_options], "Run scripts for samples of symbol [%s]",
+						sym->name) > 0)
+				scripts_symbol = nr_options++;
+		}
+
+		if (asprintf(&options[nr_options], "Run scripts for all samples") > 0)
+			scripts_all = nr_options++;
+
+		if (is_report_browser(hbt) && asprintf(&options[nr_options],
+				"Switch to another data file in PWD") > 0)
+			switch_data = nr_options++;
+add_exit_option:
+		options[nr_options++] = (char *)"Exit";
+retry_popup_menu:
+		choice = ui__popup_menu(nr_options, options);
+
+		if (choice == nr_options - 1)
+			break;
+
+		if (choice == -1) {
+			free_popup_options(options, nr_options - 1);
+			continue;
+		}
+
+		if (choice == annotate || choice == annotate_t || choice == annotate_f) {
+			struct hist_entry *he;
+			struct annotation *notes;
+			struct map_symbol ms;
+			int err;
+do_annotate:
+			if (!objdump_path && perf_session_env__lookup_objdump(env))
+				continue;
+
+			he = hist_browser__selected_entry(browser);
+			if (he == NULL)
+				continue;
+
+			if (choice == annotate_f) {
+				ms.map = he->branch_info->from.map;
+				ms.sym = he->branch_info->from.sym;
+			} else if (choice == annotate_t) {
+				ms.map = he->branch_info->to.map;
+				ms.sym = he->branch_info->to.sym;
+			} else {
+				ms = *browser->selection;
+			}
+
+			notes = symbol__annotation(ms.sym);
+			if (!notes->src)
+				continue;
+
+			err = map_symbol__tui_annotate(&ms, evsel, hbt);
+			/*
+			 * offer option to annotate the other branch source or target
+			 * (if they exists) when returning from annotate
+			 */
+			if ((err == 'q' || err == CTRL('c'))
+			    && annotate_t != -2 && annotate_f != -2)
+				goto retry_popup_menu;
+
+			ui_browser__update_nr_entries(&browser->b, browser->hists->nr_entries);
+			if (err)
+				ui_browser__handle_resize(&browser->b);
+
+		} else if (choice == browse_map)
+			map__browse(browser->selection->map);
+		else if (choice == zoom_dso) {
+zoom_dso:
+			if (browser->hists->dso_filter) {
+				pstack__remove(fstack, &browser->hists->dso_filter);
+zoom_out_dso:
+				ui_helpline__pop();
+				browser->hists->dso_filter = NULL;
+				perf_hpp__set_elide(HISTC_DSO, false);
+			} else {
+				if (dso == NULL)
+					continue;
+				ui_helpline__fpush("To zoom out press <- or -> + \"Zoom out of %s DSO\"",
+						   dso->kernel ? "the Kernel" : dso->short_name);
+				browser->hists->dso_filter = dso;
+				perf_hpp__set_elide(HISTC_DSO, true);
+				pstack__push(fstack, &browser->hists->dso_filter);
+			}
+			hists__filter_by_dso(hists);
+			hist_browser__reset(browser);
+		} else if (choice == zoom_thread) {
+zoom_thread:
+			if (browser->hists->thread_filter) {
+				pstack__remove(fstack, &browser->hists->thread_filter);
+zoom_out_thread:
+				ui_helpline__pop();
+				thread__zput(browser->hists->thread_filter);
+				perf_hpp__set_elide(HISTC_THREAD, false);
+			} else {
+				ui_helpline__fpush("To zoom out press <- or -> + \"Zoom out of %s(%d) thread\"",
+						   thread->comm_set ? thread__comm_str(thread) : "",
+						   thread->tid);
+				browser->hists->thread_filter = thread__get(thread);
+				perf_hpp__set_elide(HISTC_THREAD, false);
+				pstack__push(fstack, &browser->hists->thread_filter);
+			}
+			hists__filter_by_thread(hists);
+			hist_browser__reset(browser);
+		}
+		/* perf scripts support */
+		else if (choice == scripts_all || choice == scripts_comm ||
+				choice == scripts_symbol) {
+do_scripts:
+			memset(script_opt, 0, 64);
+
+			if (choice == scripts_comm)
+				sprintf(script_opt, " -c %s ", thread__comm_str(browser->he_selection->thread));
+
+			if (choice == scripts_symbol)
+				sprintf(script_opt, " -S %s ", browser->he_selection->ms.sym->name);
+
+			script_browse(script_opt);
+		}
+		/* Switch to another data file */
+		else if (choice == switch_data) {
+do_data_switch:
+			if (!switch_data_file()) {
+				key = K_SWITCH_INPUT_DATA;
+				break;
+			} else
+				ui__warning("Won't switch the data files due to\n"
+					"no valid data file get selected!\n");
+		}
+	}
+out_free_stack:
+	pstack__delete(fstack);
+out:
+	hist_browser__delete(browser);
+	free_popup_options(options, nr_options - 1);
+	return key;
+}
+
+struct perf_evsel_menu {
+	struct ui_browser b;
+	struct perf_evsel *selection;
+	bool lost_events, lost_events_warned;
+	float min_pcnt;
+	struct perf_session_env *env;
+};
+
+static void perf_evsel_menu__write(struct ui_browser *browser,
+				   void *entry, int row)
+{
+	struct perf_evsel_menu *menu = container_of(browser,
+						    struct perf_evsel_menu, b);
+	struct perf_evsel *evsel = list_entry(entry, struct perf_evsel, node);
+	struct hists *hists = evsel__hists(evsel);
+	bool current_entry = ui_browser__is_current_entry(browser, row);
+	unsigned long nr_events = hists->stats.nr_events[PERF_RECORD_SAMPLE];
+	const char *ev_name = perf_evsel__name(evsel);
+	char bf[256], unit;
+	const char *warn = " ";
+	size_t printed;
+
+	ui_browser__set_color(browser, current_entry ? HE_COLORSET_SELECTED :
+						       HE_COLORSET_NORMAL);
+
+	if (perf_evsel__is_group_event(evsel)) {
+		struct perf_evsel *pos;
+
+		ev_name = perf_evsel__group_name(evsel);
+
+		for_each_group_member(pos, evsel) {
+			struct hists *pos_hists = evsel__hists(pos);
+			nr_events += pos_hists->stats.nr_events[PERF_RECORD_SAMPLE];
+		}
+	}
+
+	nr_events = convert_unit(nr_events, &unit);
+	printed = scnprintf(bf, sizeof(bf), "%lu%c%s%s", nr_events,
+			   unit, unit == ' ' ? "" : " ", ev_name);
+	slsmg_printf("%s", bf);
+
+	nr_events = hists->stats.nr_events[PERF_RECORD_LOST];
+	if (nr_events != 0) {
+		menu->lost_events = true;
+		if (!current_entry)
+			ui_browser__set_color(browser, HE_COLORSET_TOP);
+		nr_events = convert_unit(nr_events, &unit);
+		printed += scnprintf(bf, sizeof(bf), ": %ld%c%schunks LOST!",
+				     nr_events, unit, unit == ' ' ? "" : " ");
+		warn = bf;
+	}
+
+	slsmg_write_nstring(warn, browser->width - printed);
+
+	if (current_entry)
+		menu->selection = evsel;
+}
+
+static int perf_evsel_menu__run(struct perf_evsel_menu *menu,
+				int nr_events, const char *help,
+				struct hist_browser_timer *hbt)
+{
+	struct perf_evlist *evlist = menu->b.priv;
+	struct perf_evsel *pos;
+	const char *title = "Available samples";
+	int delay_secs = hbt ? hbt->refresh : 0;
+	int key;
+
+	if (ui_browser__show(&menu->b, title,
+			     "ESC: exit, ENTER|->: Browse histograms") < 0)
+		return -1;
+
+	while (1) {
+		key = ui_browser__run(&menu->b, delay_secs);
+
+		switch (key) {
+		case K_TIMER:
+			hbt->timer(hbt->arg);
+
+			if (!menu->lost_events_warned && menu->lost_events) {
+				ui_browser__warn_lost_events(&menu->b);
+				menu->lost_events_warned = true;
+			}
+			continue;
+		case K_RIGHT:
+		case K_ENTER:
+			if (!menu->selection)
+				continue;
+			pos = menu->selection;
+browse_hists:
+			perf_evlist__set_selected(evlist, pos);
+			/*
+			 * Give the calling tool a chance to populate the non
+			 * default evsel resorted hists tree.
+			 */
+			if (hbt)
+				hbt->timer(hbt->arg);
+			key = perf_evsel__hists_browse(pos, nr_events, help,
+						       true, hbt,
+						       menu->min_pcnt,
+						       menu->env);
+			ui_browser__show_title(&menu->b, title);
+			switch (key) {
+			case K_TAB:
+				if (pos->node.next == &evlist->entries)
+					pos = perf_evlist__first(evlist);
+				else
+					pos = perf_evsel__next(pos);
+				goto browse_hists;
+			case K_UNTAB:
+				if (pos->node.prev == &evlist->entries)
+					pos = perf_evlist__last(evlist);
+				else
+					pos = perf_evsel__prev(pos);
+				goto browse_hists;
+			case K_ESC:
+				if (!ui_browser__dialog_yesno(&menu->b,
+						"Do you really want to exit?"))
+					continue;
+				/* Fall thru */
+			case K_SWITCH_INPUT_DATA:
+			case 'q':
+			case CTRL('c'):
+				goto out;
+			default:
+				continue;
+			}
+		case K_LEFT:
+			continue;
+		case K_ESC:
+			if (!ui_browser__dialog_yesno(&menu->b,
+					       "Do you really want to exit?"))
+				continue;
+			/* Fall thru */
+		case 'q':
+		case CTRL('c'):
+			goto out;
+		default:
+			continue;
+		}
+	}
+
+out:
+	ui_browser__hide(&menu->b);
+	return key;
+}
+
+static bool filter_group_entries(struct ui_browser *browser __maybe_unused,
+				 void *entry)
+{
+	struct perf_evsel *evsel = list_entry(entry, struct perf_evsel, node);
+
+	if (symbol_conf.event_group && !perf_evsel__is_group_leader(evsel))
+		return true;
+
+	return false;
+}
+
+static int __perf_evlist__tui_browse_hists(struct perf_evlist *evlist,
+					   int nr_entries, const char *help,
+					   struct hist_browser_timer *hbt,
+					   float min_pcnt,
+					   struct perf_session_env *env)
+{
+	struct perf_evsel *pos;
+	struct perf_evsel_menu menu = {
+		.b = {
+			.entries    = &evlist->entries,
+			.refresh    = ui_browser__list_head_refresh,
+			.seek	    = ui_browser__list_head_seek,
+			.write	    = perf_evsel_menu__write,
+			.filter	    = filter_group_entries,
+			.nr_entries = nr_entries,
+			.priv	    = evlist,
+		},
+		.min_pcnt = min_pcnt,
+		.env = env,
+	};
+
+	ui_helpline__push("Press ESC to exit");
+
+	evlist__for_each(evlist, pos) {
+		const char *ev_name = perf_evsel__name(pos);
+		size_t line_len = strlen(ev_name) + 7;
+
+		if (menu.b.width < line_len)
+			menu.b.width = line_len;
+	}
+
+	return perf_evsel_menu__run(&menu, nr_entries, help, hbt);
+}
+
+int perf_evlist__tui_browse_hists(struct perf_evlist *evlist, const char *help,
+				  struct hist_browser_timer *hbt,
+				  float min_pcnt,
+				  struct perf_session_env *env)
+{
+	int nr_entries = evlist->nr_entries;
+
+single_entry:
+	if (nr_entries == 1) {
+		struct perf_evsel *first = perf_evlist__first(evlist);
+
+		return perf_evsel__hists_browse(first, nr_entries, help,
+						false, hbt, min_pcnt,
+						env);
+	}
+
+	if (symbol_conf.event_group) {
+		struct perf_evsel *pos;
+
+		nr_entries = 0;
+		evlist__for_each(evlist, pos) {
+			if (perf_evsel__is_group_leader(pos))
+				nr_entries++;
+		}
+
+		if (nr_entries == 1)
+			goto single_entry;
+	}
+
+	return __perf_evlist__tui_browse_hists(evlist, nr_entries, help,
+					       hbt, min_pcnt, env);
+}
diff --git a/tools/perf/ui/browsers/map.c b/tools/perf/ui/browsers/map.c
new file mode 100644
index 000000000..b11639f33
--- /dev/null
+++ b/tools/perf/ui/browsers/map.c
@@ -0,0 +1,130 @@
+#include "../libslang.h"
+#include <elf.h>
+#include <inttypes.h>
+#include <sys/ttydefaults.h>
+#include <string.h>
+#include <linux/bitops.h>
+#include "../../util/util.h"
+#include "../../util/debug.h"
+#include "../../util/symbol.h"
+#include "../browser.h"
+#include "../helpline.h"
+#include "../keysyms.h"
+#include "map.h"
+
+struct map_browser {
+	struct ui_browser b;
+	struct map	  *map;
+	u8		  addrlen;
+};
+
+static void map_browser__write(struct ui_browser *browser, void *nd, int row)
+{
+	struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+	struct map_browser *mb = container_of(browser, struct map_browser, b);
+	bool current_entry = ui_browser__is_current_entry(browser, row);
+	int width;
+
+	ui_browser__set_percent_color(browser, 0, current_entry);
+	slsmg_printf("%*" PRIx64 " %*" PRIx64 " %c ",
+		     mb->addrlen, sym->start, mb->addrlen, sym->end,
+		     sym->binding == STB_GLOBAL ? 'g' :
+		     sym->binding == STB_LOCAL  ? 'l' : 'w');
+	width = browser->width - ((mb->addrlen * 2) + 4);
+	if (width > 0)
+		slsmg_write_nstring(sym->name, width);
+}
+
+/* FIXME uber-kludgy, see comment on cmd_report... */
+static u32 *symbol__browser_index(struct symbol *browser)
+{
+	return ((void *)browser) - sizeof(struct rb_node) - sizeof(u32);
+}
+
+static int map_browser__search(struct map_browser *browser)
+{
+	char target[512];
+	struct symbol *sym;
+	int err = ui_browser__input_window("Search by name/addr",
+					   "Prefix with 0x to search by address",
+					   target, "ENTER: OK, ESC: Cancel", 0);
+	if (err != K_ENTER)
+		return -1;
+
+	if (target[0] == '0' && tolower(target[1]) == 'x') {
+		u64 addr = strtoull(target, NULL, 16);
+		sym = map__find_symbol(browser->map, addr, NULL);
+	} else
+		sym = map__find_symbol_by_name(browser->map, target, NULL);
+
+	if (sym != NULL) {
+		u32 *idx = symbol__browser_index(sym);
+
+		browser->b.top = &sym->rb_node;
+		browser->b.index = browser->b.top_idx = *idx;
+	} else
+		ui_helpline__fpush("%s not found!", target);
+
+	return 0;
+}
+
+static int map_browser__run(struct map_browser *browser)
+{
+	int key;
+
+	if (ui_browser__show(&browser->b, browser->map->dso->long_name,
+			     "Press <- or ESC to exit, %s / to search",
+			     verbose ? "" : "restart with -v to use") < 0)
+		return -1;
+
+	while (1) {
+		key = ui_browser__run(&browser->b, 0);
+
+		switch (key) {
+		case '/':
+			if (verbose)
+				map_browser__search(browser);
+		default:
+			break;
+                case K_LEFT:
+                case K_ESC:
+                case 'q':
+                case CTRL('c'):
+                        goto out;
+		}
+	}
+out:
+	ui_browser__hide(&browser->b);
+	return key;
+}
+
+int map__browse(struct map *map)
+{
+	struct map_browser mb = {
+		.b = {
+			.entries = &map->dso->symbols[map->type],
+			.refresh = ui_browser__rb_tree_refresh,
+			.seek	 = ui_browser__rb_tree_seek,
+			.write	 = map_browser__write,
+		},
+		.map = map,
+	};
+	struct rb_node *nd;
+	char tmp[BITS_PER_LONG / 4];
+	u64 maxaddr = 0;
+
+	for (nd = rb_first(mb.b.entries); nd; nd = rb_next(nd)) {
+		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+
+		if (maxaddr < pos->end)
+			maxaddr = pos->end;
+		if (verbose) {
+			u32 *idx = symbol__browser_index(pos);
+			*idx = mb.b.nr_entries;
+		}
+		++mb.b.nr_entries;
+	}
+
+	mb.addrlen = snprintf(tmp, sizeof(tmp), "%" PRIx64, maxaddr);
+	return map_browser__run(&mb);
+}
diff --git a/tools/perf/ui/browsers/map.h b/tools/perf/ui/browsers/map.h
new file mode 100644
index 000000000..2d58e4b3e
--- /dev/null
+++ b/tools/perf/ui/browsers/map.h
@@ -0,0 +1,6 @@
+#ifndef _PERF_UI_MAP_BROWSER_H_
+#define _PERF_UI_MAP_BROWSER_H_ 1
+struct map;
+
+int map__browse(struct map *map);
+#endif /* _PERF_UI_MAP_BROWSER_H_ */
diff --git a/tools/perf/ui/browsers/scripts.c b/tools/perf/ui/browsers/scripts.c
new file mode 100644
index 000000000..402d2bd30
--- /dev/null
+++ b/tools/perf/ui/browsers/scripts.c
@@ -0,0 +1,187 @@
+#include <elf.h>
+#include <inttypes.h>
+#include <sys/ttydefaults.h>
+#include <string.h>
+#include "../../util/sort.h"
+#include "../../util/util.h"
+#include "../../util/hist.h"
+#include "../../util/debug.h"
+#include "../../util/symbol.h"
+#include "../browser.h"
+#include "../helpline.h"
+#include "../libslang.h"
+
+/* 2048 lines should be enough for a script output */
+#define MAX_LINES		2048
+
+/* 160 bytes for one output line */
+#define AVERAGE_LINE_LEN	160
+
+struct script_line {
+	struct list_head node;
+	char line[AVERAGE_LINE_LEN];
+};
+
+struct perf_script_browser {
+	struct ui_browser b;
+	struct list_head entries;
+	const char *script_name;
+	int nr_lines;
+};
+
+#define SCRIPT_NAMELEN	128
+#define SCRIPT_MAX_NO	64
+/*
+ * Usually the full path for a script is:
+ *	/home/username/libexec/perf-core/scripts/python/xxx.py
+ *	/home/username/libexec/perf-core/scripts/perl/xxx.pl
+ * So 256 should be long enough to contain the full path.
+ */
+#define SCRIPT_FULLPATH_LEN	256
+
+/*
+ * When success, will copy the full path of the selected script
+ * into  the buffer pointed by script_name, and return 0.
+ * Return -1 on failure.
+ */
+static int list_scripts(char *script_name)
+{
+	char *buf, *names[SCRIPT_MAX_NO], *paths[SCRIPT_MAX_NO];
+	int i, num, choice, ret = -1;
+
+	/* Preset the script name to SCRIPT_NAMELEN */
+	buf = malloc(SCRIPT_MAX_NO * (SCRIPT_NAMELEN + SCRIPT_FULLPATH_LEN));
+	if (!buf)
+		return ret;
+
+	for (i = 0; i < SCRIPT_MAX_NO; i++) {
+		names[i] = buf + i * (SCRIPT_NAMELEN + SCRIPT_FULLPATH_LEN);
+		paths[i] = names[i] + SCRIPT_NAMELEN;
+	}
+
+	num = find_scripts(names, paths);
+	if (num > 0) {
+		choice = ui__popup_menu(num, names);
+		if (choice < num && choice >= 0) {
+			strcpy(script_name, paths[choice]);
+			ret = 0;
+		}
+	}
+
+	free(buf);
+	return ret;
+}
+
+static void script_browser__write(struct ui_browser *browser,
+				   void *entry, int row)
+{
+	struct script_line *sline = list_entry(entry, struct script_line, node);
+	bool current_entry = ui_browser__is_current_entry(browser, row);
+
+	ui_browser__set_color(browser, current_entry ? HE_COLORSET_SELECTED :
+						       HE_COLORSET_NORMAL);
+
+	slsmg_write_nstring(sline->line, browser->width);
+}
+
+static int script_browser__run(struct perf_script_browser *browser)
+{
+	int key;
+
+	if (ui_browser__show(&browser->b, browser->script_name,
+			     "Press <- or ESC to exit") < 0)
+		return -1;
+
+	while (1) {
+		key = ui_browser__run(&browser->b, 0);
+
+		/* We can add some special key handling here if needed */
+		break;
+	}
+
+	ui_browser__hide(&browser->b);
+	return key;
+}
+
+
+int script_browse(const char *script_opt)
+{
+	char cmd[SCRIPT_FULLPATH_LEN*2], script_name[SCRIPT_FULLPATH_LEN];
+	char *line = NULL;
+	size_t len = 0;
+	ssize_t retlen;
+	int ret = -1, nr_entries = 0;
+	FILE *fp;
+	void *buf;
+	struct script_line *sline;
+
+	struct perf_script_browser script = {
+		.b = {
+			.refresh    = ui_browser__list_head_refresh,
+			.seek	    = ui_browser__list_head_seek,
+			.write	    = script_browser__write,
+		},
+		.script_name = script_name,
+	};
+
+	INIT_LIST_HEAD(&script.entries);
+
+	/* Save each line of the output in one struct script_line object. */
+	buf = zalloc((sizeof(*sline)) * MAX_LINES);
+	if (!buf)
+		return -1;
+	sline = buf;
+
+	memset(script_name, 0, SCRIPT_FULLPATH_LEN);
+	if (list_scripts(script_name))
+		goto exit;
+
+	sprintf(cmd, "perf script -s %s ", script_name);
+
+	if (script_opt)
+		strcat(cmd, script_opt);
+
+	if (input_name) {
+		strcat(cmd, " -i ");
+		strcat(cmd, input_name);
+	}
+
+	strcat(cmd, " 2>&1");
+
+	fp = popen(cmd, "r");
+	if (!fp)
+		goto exit;
+
+	while ((retlen = getline(&line, &len, fp)) != -1) {
+		strncpy(sline->line, line, AVERAGE_LINE_LEN);
+
+		/* If one output line is very large, just cut it short */
+		if (retlen >= AVERAGE_LINE_LEN) {
+			sline->line[AVERAGE_LINE_LEN - 1] = '\0';
+			sline->line[AVERAGE_LINE_LEN - 2] = '\n';
+		}
+		list_add_tail(&sline->node, &script.entries);
+
+		if (script.b.width < retlen)
+			script.b.width = retlen;
+
+		if (nr_entries++ >= MAX_LINES - 1)
+			break;
+		sline++;
+	}
+
+	if (script.b.width > AVERAGE_LINE_LEN)
+		script.b.width = AVERAGE_LINE_LEN;
+
+	free(line);
+	pclose(fp);
+
+	script.nr_lines = nr_entries;
+	script.b.nr_entries = nr_entries;
+	script.b.entries = &script.entries;
+
+	ret = script_browser__run(&script);
+exit:
+	free(buf);
+	return ret;
+}
diff --git a/tools/perf/ui/gtk/Build b/tools/perf/ui/gtk/Build
new file mode 100644
index 000000000..ec22e899a
--- /dev/null
+++ b/tools/perf/ui/gtk/Build
@@ -0,0 +1,9 @@
+CFLAGS_gtk += -fPIC $(GTK_CFLAGS)
+
+gtk-y += browser.o
+gtk-y += hists.o
+gtk-y += setup.o
+gtk-y += util.o
+gtk-y += helpline.o
+gtk-y += progress.o
+gtk-y += annotate.o
diff --git a/tools/perf/ui/gtk/annotate.c b/tools/perf/ui/gtk/annotate.c
new file mode 100644
index 000000000..9c7ff8d31
--- /dev/null
+++ b/tools/perf/ui/gtk/annotate.c
@@ -0,0 +1,252 @@
+#include "gtk.h"
+#include "util/debug.h"
+#include "util/annotate.h"
+#include "util/evsel.h"
+#include "ui/helpline.h"
+
+
+enum {
+	ANN_COL__PERCENT,
+	ANN_COL__OFFSET,
+	ANN_COL__LINE,
+
+	MAX_ANN_COLS
+};
+
+static const char *const col_names[] = {
+	"Overhead",
+	"Offset",
+	"Line"
+};
+
+static int perf_gtk__get_percent(char *buf, size_t size, struct symbol *sym,
+				 struct disasm_line *dl, int evidx)
+{
+	struct sym_hist *symhist;
+	double percent = 0.0;
+	const char *markup;
+	int ret = 0;
+
+	strcpy(buf, "");
+
+	if (dl->offset == (s64) -1)
+		return 0;
+
+	symhist = annotation__histogram(symbol__annotation(sym), evidx);
+	if (!symbol_conf.event_group && !symhist->addr[dl->offset])
+		return 0;
+
+	percent = 100.0 * symhist->addr[dl->offset] / symhist->sum;
+
+	markup = perf_gtk__get_percent_color(percent);
+	if (markup)
+		ret += scnprintf(buf, size, "%s", markup);
+	ret += scnprintf(buf + ret, size - ret, "%6.2f%%", percent);
+	if (markup)
+		ret += scnprintf(buf + ret, size - ret, "</span>");
+
+	return ret;
+}
+
+static int perf_gtk__get_offset(char *buf, size_t size, struct symbol *sym,
+				struct map *map, struct disasm_line *dl)
+{
+	u64 start = map__rip_2objdump(map, sym->start);
+
+	strcpy(buf, "");
+
+	if (dl->offset == (s64) -1)
+		return 0;
+
+	return scnprintf(buf, size, "%"PRIx64, start + dl->offset);
+}
+
+static int perf_gtk__get_line(char *buf, size_t size, struct disasm_line *dl)
+{
+	int ret = 0;
+	char *line = g_markup_escape_text(dl->line, -1);
+	const char *markup = "<span fgcolor='gray'>";
+
+	strcpy(buf, "");
+
+	if (!line)
+		return 0;
+
+	if (dl->offset != (s64) -1)
+		markup = NULL;
+
+	if (markup)
+		ret += scnprintf(buf, size, "%s", markup);
+	ret += scnprintf(buf + ret, size - ret, "%s", line);
+	if (markup)
+		ret += scnprintf(buf + ret, size - ret, "</span>");
+
+	g_free(line);
+	return ret;
+}
+
+static int perf_gtk__annotate_symbol(GtkWidget *window, struct symbol *sym,
+				struct map *map, struct perf_evsel *evsel,
+				struct hist_browser_timer *hbt __maybe_unused)
+{
+	struct disasm_line *pos, *n;
+	struct annotation *notes;
+	GType col_types[MAX_ANN_COLS];
+	GtkCellRenderer *renderer;
+	GtkListStore *store;
+	GtkWidget *view;
+	int i;
+	char s[512];
+
+	notes = symbol__annotation(sym);
+
+	for (i = 0; i < MAX_ANN_COLS; i++) {
+		col_types[i] = G_TYPE_STRING;
+	}
+	store = gtk_list_store_newv(MAX_ANN_COLS, col_types);
+
+	view = gtk_tree_view_new();
+	renderer = gtk_cell_renderer_text_new();
+
+	for (i = 0; i < MAX_ANN_COLS; i++) {
+		gtk_tree_view_insert_column_with_attributes(GTK_TREE_VIEW(view),
+					-1, col_names[i], renderer, "markup",
+					i, NULL);
+	}
+
+	gtk_tree_view_set_model(GTK_TREE_VIEW(view), GTK_TREE_MODEL(store));
+	g_object_unref(GTK_TREE_MODEL(store));
+
+	list_for_each_entry(pos, &notes->src->source, node) {
+		GtkTreeIter iter;
+		int ret = 0;
+
+		gtk_list_store_append(store, &iter);
+
+		if (perf_evsel__is_group_event(evsel)) {
+			for (i = 0; i < evsel->nr_members; i++) {
+				ret += perf_gtk__get_percent(s + ret,
+							     sizeof(s) - ret,
+							     sym, pos,
+							     evsel->idx + i);
+				ret += scnprintf(s + ret, sizeof(s) - ret, " ");
+			}
+		} else {
+			ret = perf_gtk__get_percent(s, sizeof(s), sym, pos,
+						    evsel->idx);
+		}
+
+		if (ret)
+			gtk_list_store_set(store, &iter, ANN_COL__PERCENT, s, -1);
+		if (perf_gtk__get_offset(s, sizeof(s), sym, map, pos))
+			gtk_list_store_set(store, &iter, ANN_COL__OFFSET, s, -1);
+		if (perf_gtk__get_line(s, sizeof(s), pos))
+			gtk_list_store_set(store, &iter, ANN_COL__LINE, s, -1);
+	}
+
+	gtk_container_add(GTK_CONTAINER(window), view);
+
+	list_for_each_entry_safe(pos, n, &notes->src->source, node) {
+		list_del(&pos->node);
+		disasm_line__free(pos);
+	}
+
+	return 0;
+}
+
+static int symbol__gtk_annotate(struct symbol *sym, struct map *map,
+				struct perf_evsel *evsel,
+				struct hist_browser_timer *hbt)
+{
+	GtkWidget *window;
+	GtkWidget *notebook;
+	GtkWidget *scrolled_window;
+	GtkWidget *tab_label;
+
+	if (map->dso->annotate_warned)
+		return -1;
+
+	if (symbol__annotate(sym, map, 0) < 0) {
+		ui__error("%s", ui_helpline__current);
+		return -1;
+	}
+
+	if (perf_gtk__is_active_context(pgctx)) {
+		window = pgctx->main_window;
+		notebook = pgctx->notebook;
+	} else {
+		GtkWidget *vbox;
+		GtkWidget *infobar;
+		GtkWidget *statbar;
+
+		signal(SIGSEGV, perf_gtk__signal);
+		signal(SIGFPE,  perf_gtk__signal);
+		signal(SIGINT,  perf_gtk__signal);
+		signal(SIGQUIT, perf_gtk__signal);
+		signal(SIGTERM, perf_gtk__signal);
+
+		window = gtk_window_new(GTK_WINDOW_TOPLEVEL);
+		gtk_window_set_title(GTK_WINDOW(window), "perf annotate");
+
+		g_signal_connect(window, "delete_event", gtk_main_quit, NULL);
+
+		pgctx = perf_gtk__activate_context(window);
+		if (!pgctx)
+			return -1;
+
+		vbox = gtk_vbox_new(FALSE, 0);
+		notebook = gtk_notebook_new();
+		pgctx->notebook = notebook;
+
+		gtk_box_pack_start(GTK_BOX(vbox), notebook, TRUE, TRUE, 0);
+
+		infobar = perf_gtk__setup_info_bar();
+		if (infobar) {
+			gtk_box_pack_start(GTK_BOX(vbox), infobar,
+					   FALSE, FALSE, 0);
+		}
+
+		statbar = perf_gtk__setup_statusbar();
+		gtk_box_pack_start(GTK_BOX(vbox), statbar, FALSE, FALSE, 0);
+
+		gtk_container_add(GTK_CONTAINER(window), vbox);
+	}
+
+	scrolled_window = gtk_scrolled_window_new(NULL, NULL);
+	tab_label = gtk_label_new(sym->name);
+
+	gtk_scrolled_window_set_policy(GTK_SCROLLED_WINDOW(scrolled_window),
+				       GTK_POLICY_AUTOMATIC,
+				       GTK_POLICY_AUTOMATIC);
+
+	gtk_notebook_append_page(GTK_NOTEBOOK(notebook), scrolled_window,
+				 tab_label);
+
+	perf_gtk__annotate_symbol(scrolled_window, sym, map, evsel, hbt);
+	return 0;
+}
+
+int hist_entry__gtk_annotate(struct hist_entry *he,
+			     struct perf_evsel *evsel,
+			     struct hist_browser_timer *hbt)
+{
+	return symbol__gtk_annotate(he->ms.sym, he->ms.map, evsel, hbt);
+}
+
+void perf_gtk__show_annotations(void)
+{
+	GtkWidget *window;
+
+	if (!perf_gtk__is_active_context(pgctx))
+		return;
+
+	window = pgctx->main_window;
+	gtk_widget_show_all(window);
+
+	perf_gtk__resize_window(window);
+	gtk_window_set_position(GTK_WINDOW(window), GTK_WIN_POS_CENTER);
+
+	gtk_main();
+
+	perf_gtk__deactivate_context(&pgctx);
+}
diff --git a/tools/perf/ui/gtk/browser.c b/tools/perf/ui/gtk/browser.c
new file mode 100644
index 000000000..c24d91221
--- /dev/null
+++ b/tools/perf/ui/gtk/browser.c
@@ -0,0 +1,87 @@
+#include "../evlist.h"
+#include "../cache.h"
+#include "../evsel.h"
+#include "../sort.h"
+#include "../hist.h"
+#include "../helpline.h"
+#include "gtk.h"
+
+#include <signal.h>
+
+void perf_gtk__signal(int sig)
+{
+	perf_gtk__exit(false);
+	psignal(sig, "perf");
+}
+
+void perf_gtk__resize_window(GtkWidget *window)
+{
+	GdkRectangle rect;
+	GdkScreen *screen;
+	int monitor;
+	int height;
+	int width;
+
+	screen = gtk_widget_get_screen(window);
+
+	monitor = gdk_screen_get_monitor_at_window(screen, window->window);
+
+	gdk_screen_get_monitor_geometry(screen, monitor, &rect);
+
+	width	= rect.width * 3 / 4;
+	height	= rect.height * 3 / 4;
+
+	gtk_window_resize(GTK_WINDOW(window), width, height);
+}
+
+const char *perf_gtk__get_percent_color(double percent)
+{
+	if (percent >= MIN_RED)
+		return "<span fgcolor='red'>";
+	if (percent >= MIN_GREEN)
+		return "<span fgcolor='dark green'>";
+	return NULL;
+}
+
+#ifdef HAVE_GTK_INFO_BAR_SUPPORT
+GtkWidget *perf_gtk__setup_info_bar(void)
+{
+	GtkWidget *info_bar;
+	GtkWidget *label;
+	GtkWidget *content_area;
+
+	info_bar = gtk_info_bar_new();
+	gtk_widget_set_no_show_all(info_bar, TRUE);
+
+	label = gtk_label_new("");
+	gtk_widget_show(label);
+
+	content_area = gtk_info_bar_get_content_area(GTK_INFO_BAR(info_bar));
+	gtk_container_add(GTK_CONTAINER(content_area), label);
+
+	gtk_info_bar_add_button(GTK_INFO_BAR(info_bar), GTK_STOCK_OK,
+				GTK_RESPONSE_OK);
+	g_signal_connect(info_bar, "response",
+			 G_CALLBACK(gtk_widget_hide), NULL);
+
+	pgctx->info_bar = info_bar;
+	pgctx->message_label = label;
+
+	return info_bar;
+}
+#endif
+
+GtkWidget *perf_gtk__setup_statusbar(void)
+{
+	GtkWidget *stbar;
+	unsigned ctxid;
+
+	stbar = gtk_statusbar_new();
+
+	ctxid = gtk_statusbar_get_context_id(GTK_STATUSBAR(stbar),
+					     "perf report");
+	pgctx->statbar = stbar;
+	pgctx->statbar_ctx_id = ctxid;
+
+	return stbar;
+}
diff --git a/tools/perf/ui/gtk/gtk.h b/tools/perf/ui/gtk/gtk.h
new file mode 100644
index 000000000..0a9173ff9
--- /dev/null
+++ b/tools/perf/ui/gtk/gtk.h
@@ -0,0 +1,67 @@
+#ifndef _PERF_GTK_H_
+#define _PERF_GTK_H_ 1
+
+#include <stdbool.h>
+
+#pragma GCC diagnostic ignored "-Wstrict-prototypes"
+#include <gtk/gtk.h>
+#pragma GCC diagnostic error "-Wstrict-prototypes"
+
+
+struct perf_gtk_context {
+	GtkWidget *main_window;
+	GtkWidget *notebook;
+
+#ifdef HAVE_GTK_INFO_BAR_SUPPORT
+	GtkWidget *info_bar;
+	GtkWidget *message_label;
+#endif
+	GtkWidget *statbar;
+	guint statbar_ctx_id;
+};
+
+int perf_gtk__init(void);
+void perf_gtk__exit(bool wait_for_ok);
+
+extern struct perf_gtk_context *pgctx;
+
+static inline bool perf_gtk__is_active_context(struct perf_gtk_context *ctx)
+{
+	return ctx && ctx->main_window;
+}
+
+struct perf_gtk_context *perf_gtk__activate_context(GtkWidget *window);
+int perf_gtk__deactivate_context(struct perf_gtk_context **ctx);
+
+void perf_gtk__init_helpline(void);
+void gtk_ui_progress__init(void);
+void perf_gtk__init_hpp(void);
+
+void perf_gtk__signal(int sig);
+void perf_gtk__resize_window(GtkWidget *window);
+const char *perf_gtk__get_percent_color(double percent);
+GtkWidget *perf_gtk__setup_statusbar(void);
+
+#ifdef HAVE_GTK_INFO_BAR_SUPPORT
+GtkWidget *perf_gtk__setup_info_bar(void);
+#else
+static inline GtkWidget *perf_gtk__setup_info_bar(void)
+{
+	return NULL;
+}
+#endif
+
+struct perf_evsel;
+struct perf_evlist;
+struct hist_entry;
+struct hist_browser_timer;
+
+int perf_evlist__gtk_browse_hists(struct perf_evlist *evlist, const char *help,
+				  struct hist_browser_timer *hbt,
+				  float min_pcnt);
+int hist_entry__gtk_annotate(struct hist_entry *he,
+			     struct perf_evsel *evsel,
+			     struct hist_browser_timer *hbt);
+void perf_gtk__show_annotations(void);
+
+#endif /* _PERF_GTK_H_ */
diff --git a/tools/perf/ui/gtk/helpline.c b/tools/perf/ui/gtk/helpline.c
new file mode 100644
index 000000000..3388cbd12
--- /dev/null
+++ b/tools/perf/ui/gtk/helpline.c
@@ -0,0 +1,57 @@
+#include <stdio.h>
+#include <string.h>
+
+#include "gtk.h"
+#include "../ui.h"
+#include "../helpline.h"
+#include "../../util/debug.h"
+
+static void gtk_helpline_pop(void)
+{
+	if (!perf_gtk__is_active_context(pgctx))
+		return;
+
+	gtk_statusbar_pop(GTK_STATUSBAR(pgctx->statbar),
+			  pgctx->statbar_ctx_id);
+}
+
+static void gtk_helpline_push(const char *msg)
+{
+	if (!perf_gtk__is_active_context(pgctx))
+		return;
+
+	gtk_statusbar_push(GTK_STATUSBAR(pgctx->statbar),
+			   pgctx->statbar_ctx_id, msg);
+}
+
+static int gtk_helpline_show(const char *fmt, va_list ap)
+{
+	int ret;
+	char *ptr;
+	static int backlog;
+
+	ret = vscnprintf(ui_helpline__current + backlog,
+			 sizeof(ui_helpline__current) - backlog, fmt, ap);
+	backlog += ret;
+
+	/* only first line can be displayed */
+	ptr = strchr(ui_helpline__current, '\n');
+	if (ptr && (ptr - ui_helpline__current) <= backlog) {
+		*ptr = '\0';
+		ui_helpline__puts(ui_helpline__current);
+		backlog = 0;
+	}
+
+	return ret;
+}
+
+static struct ui_helpline gtk_helpline_fns = {
+	.pop	= gtk_helpline_pop,
+	.push	= gtk_helpline_push,
+	.show	= gtk_helpline_show,
+};
+
+void perf_gtk__init_helpline(void)
+{
+	helpline_fns = &gtk_helpline_fns;
+}
diff --git a/tools/perf/ui/gtk/hists.c b/tools/perf/ui/gtk/hists.c
new file mode 100644
index 000000000..4b3585eed
--- /dev/null
+++ b/tools/perf/ui/gtk/hists.c
@@ -0,0 +1,356 @@
+#include "../evlist.h"
+#include "../cache.h"
+#include "../evsel.h"
+#include "../sort.h"
+#include "../hist.h"
+#include "../helpline.h"
+#include "gtk.h"
+
+#define MAX_COLUMNS			32
+
+static int __percent_color_snprintf(struct perf_hpp *hpp, const char *fmt, ...)
+{
+	int ret = 0;
+	int len;
+	va_list args;
+	double percent;
+	const char *markup;
+	char *buf = hpp->buf;
+	size_t size = hpp->size;
+
+	va_start(args, fmt);
+	len = va_arg(args, int);
+	percent = va_arg(args, double);
+	va_end(args);
+
+	markup = perf_gtk__get_percent_color(percent);
+	if (markup)
+		ret += scnprintf(buf, size, markup);
+
+	ret += scnprintf(buf + ret, size - ret, fmt, len, percent);
+
+	if (markup)
+		ret += scnprintf(buf + ret, size - ret, "</span>");
+
+	return ret;
+}
+
+#define __HPP_COLOR_PERCENT_FN(_type, _field)					\
+static u64 he_get_##_field(struct hist_entry *he)				\
+{										\
+	return he->stat._field;							\
+}										\
+										\
+static int perf_gtk__hpp_color_##_type(struct perf_hpp_fmt *fmt,		\
+				       struct perf_hpp *hpp,			\
+				       struct hist_entry *he)			\
+{										\
+	return hpp__fmt(fmt, hpp, he, he_get_##_field, " %*.2f%%",		\
+			__percent_color_snprintf, true);			\
+}
+
+#define __HPP_COLOR_ACC_PERCENT_FN(_type, _field)				\
+static u64 he_get_acc_##_field(struct hist_entry *he)				\
+{										\
+	return he->stat_acc->_field;						\
+}										\
+										\
+static int perf_gtk__hpp_color_##_type(struct perf_hpp_fmt *fmt __maybe_unused,	\
+				       struct perf_hpp *hpp,			\
+				       struct hist_entry *he)			\
+{										\
+	return hpp__fmt_acc(fmt, hpp, he, he_get_acc_##_field, " %*.2f%%", 	\
+			    __percent_color_snprintf, true);			\
+}
+
+__HPP_COLOR_PERCENT_FN(overhead, period)
+__HPP_COLOR_PERCENT_FN(overhead_sys, period_sys)
+__HPP_COLOR_PERCENT_FN(overhead_us, period_us)
+__HPP_COLOR_PERCENT_FN(overhead_guest_sys, period_guest_sys)
+__HPP_COLOR_PERCENT_FN(overhead_guest_us, period_guest_us)
+__HPP_COLOR_ACC_PERCENT_FN(overhead_acc, period)
+
+#undef __HPP_COLOR_PERCENT_FN
+
+
+void perf_gtk__init_hpp(void)
+{
+	perf_hpp__format[PERF_HPP__OVERHEAD].color =
+				perf_gtk__hpp_color_overhead;
+	perf_hpp__format[PERF_HPP__OVERHEAD_SYS].color =
+				perf_gtk__hpp_color_overhead_sys;
+	perf_hpp__format[PERF_HPP__OVERHEAD_US].color =
+				perf_gtk__hpp_color_overhead_us;
+	perf_hpp__format[PERF_HPP__OVERHEAD_GUEST_SYS].color =
+				perf_gtk__hpp_color_overhead_guest_sys;
+	perf_hpp__format[PERF_HPP__OVERHEAD_GUEST_US].color =
+				perf_gtk__hpp_color_overhead_guest_us;
+	perf_hpp__format[PERF_HPP__OVERHEAD_ACC].color =
+				perf_gtk__hpp_color_overhead_acc;
+}
+
+static void perf_gtk__add_callchain(struct rb_root *root, GtkTreeStore *store,
+				    GtkTreeIter *parent, int col, u64 total)
+{
+	struct rb_node *nd;
+	bool has_single_node = (rb_first(root) == rb_last(root));
+
+	for (nd = rb_first(root); nd; nd = rb_next(nd)) {
+		struct callchain_node *node;
+		struct callchain_list *chain;
+		GtkTreeIter iter, new_parent;
+		bool need_new_parent;
+		double percent;
+		u64 hits, child_total;
+
+		node = rb_entry(nd, struct callchain_node, rb_node);
+
+		hits = callchain_cumul_hits(node);
+		percent = 100.0 * hits / total;
+
+		new_parent = *parent;
+		need_new_parent = !has_single_node && (node->val_nr > 1);
+
+		list_for_each_entry(chain, &node->val, list) {
+			char buf[128];
+
+			gtk_tree_store_append(store, &iter, &new_parent);
+
+			scnprintf(buf, sizeof(buf), "%5.2f%%", percent);
+			gtk_tree_store_set(store, &iter, 0, buf, -1);
+
+			callchain_list__sym_name(chain, buf, sizeof(buf), false);
+			gtk_tree_store_set(store, &iter, col, buf, -1);
+
+			if (need_new_parent) {
+				/*
+				 * Only show the top-most symbol in a callchain
+				 * if it's not the only callchain.
+				 */
+				new_parent = iter;
+				need_new_parent = false;
+			}
+		}
+
+		if (callchain_param.mode == CHAIN_GRAPH_REL)
+			child_total = node->children_hit;
+		else
+			child_total = total;
+
+		/* Now 'iter' contains info of the last callchain_list */
+		perf_gtk__add_callchain(&node->rb_root, store, &iter, col,
+					child_total);
+	}
+}
+
+static void on_row_activated(GtkTreeView *view, GtkTreePath *path,
+			     GtkTreeViewColumn *col __maybe_unused,
+			     gpointer user_data __maybe_unused)
+{
+	bool expanded = gtk_tree_view_row_expanded(view, path);
+
+	if (expanded)
+		gtk_tree_view_collapse_row(view, path);
+	else
+		gtk_tree_view_expand_row(view, path, FALSE);
+}
+
+static void perf_gtk__show_hists(GtkWidget *window, struct hists *hists,
+				 float min_pcnt)
+{
+	struct perf_hpp_fmt *fmt;
+	GType col_types[MAX_COLUMNS];
+	GtkCellRenderer *renderer;
+	GtkTreeStore *store;
+	struct rb_node *nd;
+	GtkWidget *view;
+	int col_idx;
+	int sym_col = -1;
+	int nr_cols;
+	char s[512];
+
+	struct perf_hpp hpp = {
+		.buf		= s,
+		.size		= sizeof(s),
+	};
+
+	nr_cols = 0;
+
+	perf_hpp__for_each_format(fmt)
+		col_types[nr_cols++] = G_TYPE_STRING;
+
+	store = gtk_tree_store_newv(nr_cols, col_types);
+
+	view = gtk_tree_view_new();
+
+	renderer = gtk_cell_renderer_text_new();
+
+	col_idx = 0;
+
+	perf_hpp__for_each_format(fmt) {
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		/*
+		 * XXX no way to determine where symcol column is..
+		 *     Just use last column for now.
+		 */
+		if (perf_hpp__is_sort_entry(fmt))
+			sym_col = col_idx;
+
+		gtk_tree_view_insert_column_with_attributes(GTK_TREE_VIEW(view),
+							    -1, fmt->name,
+							    renderer, "markup",
+							    col_idx++, NULL);
+	}
+
+	for (col_idx = 0; col_idx < nr_cols; col_idx++) {
+		GtkTreeViewColumn *column;
+
+		column = gtk_tree_view_get_column(GTK_TREE_VIEW(view), col_idx);
+		gtk_tree_view_column_set_resizable(column, TRUE);
+
+		if (col_idx == sym_col) {
+			gtk_tree_view_set_expander_column(GTK_TREE_VIEW(view),
+							  column);
+		}
+	}
+
+	gtk_tree_view_set_model(GTK_TREE_VIEW(view), GTK_TREE_MODEL(store));
+
+	g_object_unref(GTK_TREE_MODEL(store));
+
+	for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+		GtkTreeIter iter;
+		u64 total = hists__total_period(h->hists);
+		float percent;
+
+		if (h->filtered)
+			continue;
+
+		percent = hist_entry__get_percent_limit(h);
+		if (percent < min_pcnt)
+			continue;
+
+		gtk_tree_store_append(store, &iter, NULL);
+
+		col_idx = 0;
+
+		perf_hpp__for_each_format(fmt) {
+			if (perf_hpp__should_skip(fmt))
+				continue;
+
+			if (fmt->color)
+				fmt->color(fmt, &hpp, h);
+			else
+				fmt->entry(fmt, &hpp, h);
+
+			gtk_tree_store_set(store, &iter, col_idx++, s, -1);
+		}
+
+		if (symbol_conf.use_callchain && sort__has_sym) {
+			if (callchain_param.mode == CHAIN_GRAPH_REL)
+				total = symbol_conf.cumulate_callchain ?
+					h->stat_acc->period : h->stat.period;
+
+			perf_gtk__add_callchain(&h->sorted_chain, store, &iter,
+						sym_col, total);
+		}
+	}
+
+	gtk_tree_view_set_rules_hint(GTK_TREE_VIEW(view), TRUE);
+
+	g_signal_connect(view, "row-activated",
+			 G_CALLBACK(on_row_activated), NULL);
+	gtk_container_add(GTK_CONTAINER(window), view);
+}
+
+int perf_evlist__gtk_browse_hists(struct perf_evlist *evlist,
+				  const char *help,
+				  struct hist_browser_timer *hbt __maybe_unused,
+				  float min_pcnt)
+{
+	struct perf_evsel *pos;
+	GtkWidget *vbox;
+	GtkWidget *notebook;
+	GtkWidget *info_bar;
+	GtkWidget *statbar;
+	GtkWidget *window;
+
+	signal(SIGSEGV, perf_gtk__signal);
+	signal(SIGFPE,  perf_gtk__signal);
+	signal(SIGINT,  perf_gtk__signal);
+	signal(SIGQUIT, perf_gtk__signal);
+	signal(SIGTERM, perf_gtk__signal);
+
+	window = gtk_window_new(GTK_WINDOW_TOPLEVEL);
+
+	gtk_window_set_title(GTK_WINDOW(window), "perf report");
+
+	g_signal_connect(window, "delete_event", gtk_main_quit, NULL);
+
+	pgctx = perf_gtk__activate_context(window);
+	if (!pgctx)
+		return -1;
+
+	vbox = gtk_vbox_new(FALSE, 0);
+
+	notebook = gtk_notebook_new();
+
+	gtk_box_pack_start(GTK_BOX(vbox), notebook, TRUE, TRUE, 0);
+
+	info_bar = perf_gtk__setup_info_bar();
+	if (info_bar)
+		gtk_box_pack_start(GTK_BOX(vbox), info_bar, FALSE, FALSE, 0);
+
+	statbar = perf_gtk__setup_statusbar();
+	gtk_box_pack_start(GTK_BOX(vbox), statbar, FALSE, FALSE, 0);
+
+	gtk_container_add(GTK_CONTAINER(window), vbox);
+
+	evlist__for_each(evlist, pos) {
+		struct hists *hists = evsel__hists(pos);
+		const char *evname = perf_evsel__name(pos);
+		GtkWidget *scrolled_window;
+		GtkWidget *tab_label;
+		char buf[512];
+		size_t size = sizeof(buf);
+
+		if (symbol_conf.event_group) {
+			if (!perf_evsel__is_group_leader(pos))
+				continue;
+
+			if (pos->nr_members > 1) {
+				perf_evsel__group_desc(pos, buf, size);
+				evname = buf;
+			}
+		}
+
+		scrolled_window = gtk_scrolled_window_new(NULL, NULL);
+
+		gtk_scrolled_window_set_policy(GTK_SCROLLED_WINDOW(scrolled_window),
+							GTK_POLICY_AUTOMATIC,
+							GTK_POLICY_AUTOMATIC);
+
+		perf_gtk__show_hists(scrolled_window, hists, min_pcnt);
+
+		tab_label = gtk_label_new(evname);
+
+		gtk_notebook_append_page(GTK_NOTEBOOK(notebook), scrolled_window, tab_label);
+	}
+
+	gtk_widget_show_all(window);
+
+	perf_gtk__resize_window(window);
+
+	gtk_window_set_position(GTK_WINDOW(window), GTK_WIN_POS_CENTER);
+
+	ui_helpline__push(help);
+
+	gtk_main();
+
+	perf_gtk__deactivate_context(&pgctx);
+
+	return 0;
+}
diff --git a/tools/perf/ui/gtk/progress.c b/tools/perf/ui/gtk/progress.c
new file mode 100644
index 000000000..b656655fb
--- /dev/null
+++ b/tools/perf/ui/gtk/progress.c
@@ -0,0 +1,59 @@
+#include <inttypes.h>
+
+#include "gtk.h"
+#include "../progress.h"
+#include "util.h"
+
+static GtkWidget *dialog;
+static GtkWidget *progress;
+
+static void gtk_ui_progress__update(struct ui_progress *p)
+{
+	double fraction = p->total ? 1.0 * p->curr / p->total : 0.0;
+	char buf[1024];
+
+	if (dialog == NULL) {
+		GtkWidget *vbox = gtk_vbox_new(TRUE, 5);
+		GtkWidget *label = gtk_label_new(p->title);
+
+		dialog = gtk_window_new(GTK_WINDOW_TOPLEVEL);
+		progress = gtk_progress_bar_new();
+
+		gtk_box_pack_start(GTK_BOX(vbox), label, TRUE, FALSE, 3);
+		gtk_box_pack_start(GTK_BOX(vbox), progress, TRUE, TRUE, 3);
+
+		gtk_container_add(GTK_CONTAINER(dialog), vbox);
+
+		gtk_window_set_title(GTK_WINDOW(dialog), "perf");
+		gtk_window_resize(GTK_WINDOW(dialog), 300, 80);
+		gtk_window_set_position(GTK_WINDOW(dialog), GTK_WIN_POS_CENTER);
+
+		gtk_widget_show_all(dialog);
+	}
+
+	gtk_progress_bar_set_fraction(GTK_PROGRESS_BAR(progress), fraction);
+	snprintf(buf, sizeof(buf), "%"PRIu64" / %"PRIu64, p->curr, p->total);
+	gtk_progress_bar_set_text(GTK_PROGRESS_BAR(progress), buf);
+
+	/* we didn't call gtk_main yet, so do it manually */
+	while (gtk_events_pending())
+		gtk_main_iteration();
+}
+
+static void gtk_ui_progress__finish(void)
+{
+	/* this will also destroy all of its children */
+	gtk_widget_destroy(dialog);
+
+	dialog = NULL;
+}
+
+static struct ui_progress_ops gtk_ui_progress__ops = {
+	.update		= gtk_ui_progress__update,
+	.finish		= gtk_ui_progress__finish,
+};
+
+void gtk_ui_progress__init(void)
+{
+	ui_progress__ops = &gtk_ui_progress__ops;
+}
diff --git a/tools/perf/ui/gtk/setup.c b/tools/perf/ui/gtk/setup.c
new file mode 100644
index 000000000..1d57676f8
--- /dev/null
+++ b/tools/perf/ui/gtk/setup.c
@@ -0,0 +1,23 @@
+#include "gtk.h"
+#include "../../util/cache.h"
+#include "../../util/debug.h"
+
+extern struct perf_error_ops perf_gtk_eops;
+
+int perf_gtk__init(void)
+{
+	perf_error__register(&perf_gtk_eops);
+	perf_gtk__init_helpline();
+	gtk_ui_progress__init();
+	perf_gtk__init_hpp();
+
+	return gtk_init_check(NULL, NULL) ? 0 : -1;
+}
+
+void perf_gtk__exit(bool wait_for_ok __maybe_unused)
+{
+	if (!perf_gtk__is_active_context(pgctx))
+		return;
+	perf_error__unregister(&perf_gtk_eops);
+	gtk_main_quit();
+}
diff --git a/tools/perf/ui/gtk/util.c b/tools/perf/ui/gtk/util.c
new file mode 100644
index 000000000..52e7fc48a
--- /dev/null
+++ b/tools/perf/ui/gtk/util.c
@@ -0,0 +1,112 @@
+#include "../util.h"
+#include "../../util/debug.h"
+#include "gtk.h"
+
+#include <string.h>
+
+
+struct perf_gtk_context *pgctx;
+
+struct perf_gtk_context *perf_gtk__activate_context(GtkWidget *window)
+{
+	struct perf_gtk_context *ctx;
+
+	ctx = malloc(sizeof(*pgctx));
+	if (ctx)
+		ctx->main_window = window;
+
+	return ctx;
+}
+
+int perf_gtk__deactivate_context(struct perf_gtk_context **ctx)
+{
+	if (!perf_gtk__is_active_context(*ctx))
+		return -1;
+
+	zfree(ctx);
+	return 0;
+}
+
+static int perf_gtk__error(const char *format, va_list args)
+{
+	char *msg;
+	GtkWidget *dialog;
+
+	if (!perf_gtk__is_active_context(pgctx) ||
+	    vasprintf(&msg, format, args) < 0) {
+		fprintf(stderr, "Error:\n");
+		vfprintf(stderr, format, args);
+		fprintf(stderr, "\n");
+		return -1;
+	}
+
+	dialog = gtk_message_dialog_new_with_markup(GTK_WINDOW(pgctx->main_window),
+					GTK_DIALOG_DESTROY_WITH_PARENT,
+					GTK_MESSAGE_ERROR,
+					GTK_BUTTONS_CLOSE,
+					"<b>Error</b>\n\n%s", msg);
+	gtk_dialog_run(GTK_DIALOG(dialog));
+
+	gtk_widget_destroy(dialog);
+	free(msg);
+	return 0;
+}
+
+#ifdef HAVE_GTK_INFO_BAR_SUPPORT
+static int perf_gtk__warning_info_bar(const char *format, va_list args)
+{
+	char *msg;
+
+	if (!perf_gtk__is_active_context(pgctx) ||
+	    vasprintf(&msg, format, args) < 0) {
+		fprintf(stderr, "Warning:\n");
+		vfprintf(stderr, format, args);
+		fprintf(stderr, "\n");
+		return -1;
+	}
+
+	gtk_label_set_text(GTK_LABEL(pgctx->message_label), msg);
+	gtk_info_bar_set_message_type(GTK_INFO_BAR(pgctx->info_bar),
+				      GTK_MESSAGE_WARNING);
+	gtk_widget_show(pgctx->info_bar);
+
+	free(msg);
+	return 0;
+}
+#else
+static int perf_gtk__warning_statusbar(const char *format, va_list args)
+{
+	char *msg, *p;
+
+	if (!perf_gtk__is_active_context(pgctx) ||
+	    vasprintf(&msg, format, args) < 0) {
+		fprintf(stderr, "Warning:\n");
+		vfprintf(stderr, format, args);
+		fprintf(stderr, "\n");
+		return -1;
+	}
+
+	gtk_statusbar_pop(GTK_STATUSBAR(pgctx->statbar),
+			  pgctx->statbar_ctx_id);
+
+	/* Only first line can be displayed */
+	p = strchr(msg, '\n');
+	if (p)
+		*p = '\0';
+
+	gtk_statusbar_push(GTK_STATUSBAR(pgctx->statbar),
+			   pgctx->statbar_ctx_id, msg);
+
+	free(msg);
+	return 0;
+}
+#endif
+
+struct perf_error_ops perf_gtk_eops = {
+	.error		= perf_gtk__error,
+#ifdef HAVE_GTK_INFO_BAR_SUPPORT
+	.warning	= perf_gtk__warning_info_bar,
+#else
+	.warning	= perf_gtk__warning_statusbar,
+#endif
+};
diff --git a/tools/perf/ui/helpline.c b/tools/perf/ui/helpline.c
new file mode 100644
index 000000000..700fb3cfa
--- /dev/null
+++ b/tools/perf/ui/helpline.c
@@ -0,0 +1,73 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "../debug.h"
+#include "helpline.h"
+#include "ui.h"
+
+char ui_helpline__current[512];
+
+static void nop_helpline__pop(void)
+{
+}
+
+static void nop_helpline__push(const char *msg __maybe_unused)
+{
+}
+
+static int nop_helpline__show(const char *fmt __maybe_unused,
+			       va_list ap __maybe_unused)
+{
+	return 0;
+}
+
+static struct ui_helpline default_helpline_fns = {
+	.pop	= nop_helpline__pop,
+	.push	= nop_helpline__push,
+	.show	= nop_helpline__show,
+};
+
+struct ui_helpline *helpline_fns = &default_helpline_fns;
+
+void ui_helpline__pop(void)
+{
+	helpline_fns->pop();
+}
+
+void ui_helpline__push(const char *msg)
+{
+	helpline_fns->push(msg);
+}
+
+void ui_helpline__vpush(const char *fmt, va_list ap)
+{
+	char *s;
+
+	if (vasprintf(&s, fmt, ap) < 0)
+		vfprintf(stderr, fmt, ap);
+	else {
+		ui_helpline__push(s);
+		free(s);
+	}
+}
+
+void ui_helpline__fpush(const char *fmt, ...)
+{
+	va_list ap;
+
+	va_start(ap, fmt);
+	ui_helpline__vpush(fmt, ap);
+	va_end(ap);
+}
+
+void ui_helpline__puts(const char *msg)
+{
+	ui_helpline__pop();
+	ui_helpline__push(msg);
+}
+
+int ui_helpline__vshow(const char *fmt, va_list ap)
+{
+	return helpline_fns->show(fmt, ap);
+}
diff --git a/tools/perf/ui/helpline.h b/tools/perf/ui/helpline.h
new file mode 100644
index 000000000..46181f4fc
--- /dev/null
+++ b/tools/perf/ui/helpline.h
@@ -0,0 +1,29 @@
+#ifndef _PERF_UI_HELPLINE_H_
+#define _PERF_UI_HELPLINE_H_ 1
+
+#include <stdio.h>
+#include <stdarg.h>
+
+#include "../util/cache.h"
+
+struct ui_helpline {
+	void (*pop)(void);
+	void (*push)(const char *msg);
+	int  (*show)(const char *fmt, va_list ap);
+};
+
+extern struct ui_helpline *helpline_fns;
+
+void ui_helpline__init(void);
+
+void ui_helpline__pop(void);
+void ui_helpline__push(const char *msg);
+void ui_helpline__vpush(const char *fmt, va_list ap);
+void ui_helpline__fpush(const char *fmt, ...);
+void ui_helpline__puts(const char *msg);
+int  ui_helpline__vshow(const char *fmt, va_list ap);
+
+extern char ui_helpline__current[512];
+extern char ui_helpline__last_msg[];
+
+#endif /* _PERF_UI_HELPLINE_H_ */
diff --git a/tools/perf/ui/hist.c b/tools/perf/ui/hist.c
new file mode 100644
index 000000000..25d608394
--- /dev/null
+++ b/tools/perf/ui/hist.c
@@ -0,0 +1,693 @@
+#include <math.h>
+#include <linux/compiler.h>
+
+#include "../util/hist.h"
+#include "../util/util.h"
+#include "../util/sort.h"
+#include "../util/evsel.h"
+
+/* hist period print (hpp) functions */
+
+#define hpp__call_print_fn(hpp, fn, fmt, ...)			\
+({								\
+	int __ret = fn(hpp, fmt, ##__VA_ARGS__);		\
+	advance_hpp(hpp, __ret);				\
+	__ret;							\
+})
+
+static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
+		      hpp_field_fn get_field, const char *fmt, int len,
+		      hpp_snprint_fn print_fn, bool fmt_percent)
+{
+	int ret;
+	struct hists *hists = he->hists;
+	struct perf_evsel *evsel = hists_to_evsel(hists);
+	char *buf = hpp->buf;
+	size_t size = hpp->size;
+
+	if (fmt_percent) {
+		double percent = 0.0;
+		u64 total = hists__total_period(hists);
+
+		if (total)
+			percent = 100.0 * get_field(he) / total;
+
+		ret = hpp__call_print_fn(hpp, print_fn, fmt, len, percent);
+	} else
+		ret = hpp__call_print_fn(hpp, print_fn, fmt, len, get_field(he));
+
+	if (perf_evsel__is_group_event(evsel)) {
+		int prev_idx, idx_delta;
+		struct hist_entry *pair;
+		int nr_members = evsel->nr_members;
+
+		prev_idx = perf_evsel__group_idx(evsel);
+
+		list_for_each_entry(pair, &he->pairs.head, pairs.node) {
+			u64 period = get_field(pair);
+			u64 total = hists__total_period(pair->hists);
+
+			if (!total)
+				continue;
+
+			evsel = hists_to_evsel(pair->hists);
+			idx_delta = perf_evsel__group_idx(evsel) - prev_idx - 1;
+
+			while (idx_delta--) {
+				/*
+				 * zero-fill group members in the middle which
+				 * have no sample
+				 */
+				if (fmt_percent) {
+					ret += hpp__call_print_fn(hpp, print_fn,
+								  fmt, len, 0.0);
+				} else {
+					ret += hpp__call_print_fn(hpp, print_fn,
+								  fmt, len, 0ULL);
+				}
+			}
+
+			if (fmt_percent) {
+				ret += hpp__call_print_fn(hpp, print_fn, fmt, len,
+							  100.0 * period / total);
+			} else {
+				ret += hpp__call_print_fn(hpp, print_fn, fmt,
+							  len, period);
+			}
+
+			prev_idx = perf_evsel__group_idx(evsel);
+		}
+
+		idx_delta = nr_members - prev_idx - 1;
+
+		while (idx_delta--) {
+			/*
+			 * zero-fill group members at last which have no sample
+			 */
+			if (fmt_percent) {
+				ret += hpp__call_print_fn(hpp, print_fn,
+							  fmt, len, 0.0);
+			} else {
+				ret += hpp__call_print_fn(hpp, print_fn,
+							  fmt, len, 0ULL);
+			}
+		}
+	}
+
+	/*
+	 * Restore original buf and size as it's where caller expects
+	 * the result will be saved.
+	 */
+	hpp->buf = buf;
+	hpp->size = size;
+
+	return ret;
+}
+
+int hpp__fmt(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+	     struct hist_entry *he, hpp_field_fn get_field,
+	     const char *fmtstr, hpp_snprint_fn print_fn, bool fmt_percent)
+{
+	int len = fmt->user_len ?: fmt->len;
+
+	if (symbol_conf.field_sep) {
+		return __hpp__fmt(hpp, he, get_field, fmtstr, 1,
+				  print_fn, fmt_percent);
+	}
+
+	if (fmt_percent)
+		len -= 2; /* 2 for a space and a % sign */
+	else
+		len -= 1;
+
+	return  __hpp__fmt(hpp, he, get_field, fmtstr, len, print_fn, fmt_percent);
+}
+
+int hpp__fmt_acc(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+		 struct hist_entry *he, hpp_field_fn get_field,
+		 const char *fmtstr, hpp_snprint_fn print_fn, bool fmt_percent)
+{
+	if (!symbol_conf.cumulate_callchain) {
+		int len = fmt->user_len ?: fmt->len;
+		return snprintf(hpp->buf, hpp->size, " %*s", len - 1, "N/A");
+	}
+
+	return hpp__fmt(fmt, hpp, he, get_field, fmtstr, print_fn, fmt_percent);
+}
+
+static int field_cmp(u64 field_a, u64 field_b)
+{
+	if (field_a > field_b)
+		return 1;
+	if (field_a < field_b)
+		return -1;
+	return 0;
+}
+
+static int __hpp__sort(struct hist_entry *a, struct hist_entry *b,
+		       hpp_field_fn get_field)
+{
+	s64 ret;
+	int i, nr_members;
+	struct perf_evsel *evsel;
+	struct hist_entry *pair;
+	u64 *fields_a, *fields_b;
+
+	ret = field_cmp(get_field(a), get_field(b));
+	if (ret || !symbol_conf.event_group)
+		return ret;
+
+	evsel = hists_to_evsel(a->hists);
+	if (!perf_evsel__is_group_event(evsel))
+		return ret;
+
+	nr_members = evsel->nr_members;
+	fields_a = calloc(nr_members, sizeof(*fields_a));
+	fields_b = calloc(nr_members, sizeof(*fields_b));
+
+	if (!fields_a || !fields_b)
+		goto out;
+
+	list_for_each_entry(pair, &a->pairs.head, pairs.node) {
+		evsel = hists_to_evsel(pair->hists);
+		fields_a[perf_evsel__group_idx(evsel)] = get_field(pair);
+	}
+
+	list_for_each_entry(pair, &b->pairs.head, pairs.node) {
+		evsel = hists_to_evsel(pair->hists);
+		fields_b[perf_evsel__group_idx(evsel)] = get_field(pair);
+	}
+
+	for (i = 1; i < nr_members; i++) {
+		ret = field_cmp(fields_a[i], fields_b[i]);
+		if (ret)
+			break;
+	}
+
+out:
+	free(fields_a);
+	free(fields_b);
+
+	return ret;
+}
+
+static int __hpp__sort_acc(struct hist_entry *a, struct hist_entry *b,
+			   hpp_field_fn get_field)
+{
+	s64 ret = 0;
+
+	if (symbol_conf.cumulate_callchain) {
+		/*
+		 * Put caller above callee when they have equal period.
+		 */
+		ret = field_cmp(get_field(a), get_field(b));
+		if (ret)
+			return ret;
+
+		if (a->thread != b->thread || !symbol_conf.use_callchain)
+			return 0;
+
+		ret = b->callchain->max_depth - a->callchain->max_depth;
+	}
+	return ret;
+}
+
+static int hpp__width_fn(struct perf_hpp_fmt *fmt,
+			 struct perf_hpp *hpp __maybe_unused,
+			 struct perf_evsel *evsel)
+{
+	int len = fmt->user_len ?: fmt->len;
+
+	if (symbol_conf.event_group)
+		len = max(len, evsel->nr_members * fmt->len);
+
+	if (len < (int)strlen(fmt->name))
+		len = strlen(fmt->name);
+
+	return len;
+}
+
+static int hpp__header_fn(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+			  struct perf_evsel *evsel)
+{
+	int len = hpp__width_fn(fmt, hpp, evsel);
+	return scnprintf(hpp->buf, hpp->size, "%*s", len, fmt->name);
+}
+
+static int hpp_color_scnprintf(struct perf_hpp *hpp, const char *fmt, ...)
+{
+	va_list args;
+	ssize_t ssize = hpp->size;
+	double percent;
+	int ret, len;
+
+	va_start(args, fmt);
+	len = va_arg(args, int);
+	percent = va_arg(args, double);
+	ret = percent_color_len_snprintf(hpp->buf, hpp->size, fmt, len, percent);
+	va_end(args);
+
+	return (ret >= ssize) ? (ssize - 1) : ret;
+}
+
+static int hpp_entry_scnprintf(struct perf_hpp *hpp, const char *fmt, ...)
+{
+	va_list args;
+	ssize_t ssize = hpp->size;
+	int ret;
+
+	va_start(args, fmt);
+	ret = vsnprintf(hpp->buf, hpp->size, fmt, args);
+	va_end(args);
+
+	return (ret >= ssize) ? (ssize - 1) : ret;
+}
+
+#define __HPP_COLOR_PERCENT_FN(_type, _field)					\
+static u64 he_get_##_field(struct hist_entry *he)				\
+{										\
+	return he->stat._field;							\
+}										\
+										\
+static int hpp__color_##_type(struct perf_hpp_fmt *fmt,				\
+			      struct perf_hpp *hpp, struct hist_entry *he) 	\
+{										\
+	return hpp__fmt(fmt, hpp, he, he_get_##_field, " %*.2f%%",		\
+			hpp_color_scnprintf, true);				\
+}
+
+#define __HPP_ENTRY_PERCENT_FN(_type, _field)					\
+static int hpp__entry_##_type(struct perf_hpp_fmt *fmt,				\
+			      struct perf_hpp *hpp, struct hist_entry *he) 	\
+{										\
+	return hpp__fmt(fmt, hpp, he, he_get_##_field, " %*.2f%%",		\
+			hpp_entry_scnprintf, true);				\
+}
+
+#define __HPP_SORT_FN(_type, _field)						\
+static int64_t hpp__sort_##_type(struct perf_hpp_fmt *fmt __maybe_unused, 	\
+				 struct hist_entry *a, struct hist_entry *b) 	\
+{										\
+	return __hpp__sort(a, b, he_get_##_field);				\
+}
+
+#define __HPP_COLOR_ACC_PERCENT_FN(_type, _field)				\
+static u64 he_get_acc_##_field(struct hist_entry *he)				\
+{										\
+	return he->stat_acc->_field;						\
+}										\
+										\
+static int hpp__color_##_type(struct perf_hpp_fmt *fmt,				\
+			      struct perf_hpp *hpp, struct hist_entry *he) 	\
+{										\
+	return hpp__fmt_acc(fmt, hpp, he, he_get_acc_##_field, " %*.2f%%", 	\
+			    hpp_color_scnprintf, true);				\
+}
+
+#define __HPP_ENTRY_ACC_PERCENT_FN(_type, _field)				\
+static int hpp__entry_##_type(struct perf_hpp_fmt *fmt,				\
+			      struct perf_hpp *hpp, struct hist_entry *he) 	\
+{										\
+	return hpp__fmt_acc(fmt, hpp, he, he_get_acc_##_field, " %*.2f%%",	\
+			    hpp_entry_scnprintf, true);				\
+}
+
+#define __HPP_SORT_ACC_FN(_type, _field)					\
+static int64_t hpp__sort_##_type(struct perf_hpp_fmt *fmt __maybe_unused, 	\
+				 struct hist_entry *a, struct hist_entry *b) 	\
+{										\
+	return __hpp__sort_acc(a, b, he_get_acc_##_field);			\
+}
+
+#define __HPP_ENTRY_RAW_FN(_type, _field)					\
+static u64 he_get_raw_##_field(struct hist_entry *he)				\
+{										\
+	return he->stat._field;							\
+}										\
+										\
+static int hpp__entry_##_type(struct perf_hpp_fmt *fmt,				\
+			      struct perf_hpp *hpp, struct hist_entry *he) 	\
+{										\
+	return hpp__fmt(fmt, hpp, he, he_get_raw_##_field, " %*"PRIu64, 	\
+			hpp_entry_scnprintf, false);				\
+}
+
+#define __HPP_SORT_RAW_FN(_type, _field)					\
+static int64_t hpp__sort_##_type(struct perf_hpp_fmt *fmt __maybe_unused, 	\
+				 struct hist_entry *a, struct hist_entry *b) 	\
+{										\
+	return __hpp__sort(a, b, he_get_raw_##_field);				\
+}
+
+
+#define HPP_PERCENT_FNS(_type, _field)					\
+__HPP_COLOR_PERCENT_FN(_type, _field)					\
+__HPP_ENTRY_PERCENT_FN(_type, _field)					\
+__HPP_SORT_FN(_type, _field)
+
+#define HPP_PERCENT_ACC_FNS(_type, _field)				\
+__HPP_COLOR_ACC_PERCENT_FN(_type, _field)				\
+__HPP_ENTRY_ACC_PERCENT_FN(_type, _field)				\
+__HPP_SORT_ACC_FN(_type, _field)
+
+#define HPP_RAW_FNS(_type, _field)					\
+__HPP_ENTRY_RAW_FN(_type, _field)					\
+__HPP_SORT_RAW_FN(_type, _field)
+
+HPP_PERCENT_FNS(overhead, period)
+HPP_PERCENT_FNS(overhead_sys, period_sys)
+HPP_PERCENT_FNS(overhead_us, period_us)
+HPP_PERCENT_FNS(overhead_guest_sys, period_guest_sys)
+HPP_PERCENT_FNS(overhead_guest_us, period_guest_us)
+HPP_PERCENT_ACC_FNS(overhead_acc, period)
+
+HPP_RAW_FNS(samples, nr_events)
+HPP_RAW_FNS(period, period)
+
+static int64_t hpp__nop_cmp(struct perf_hpp_fmt *fmt __maybe_unused,
+			    struct hist_entry *a __maybe_unused,
+			    struct hist_entry *b __maybe_unused)
+{
+	return 0;
+}
+
+#define HPP__COLOR_PRINT_FNS(_name, _fn)		\
+	{						\
+		.name   = _name,			\
+		.header	= hpp__header_fn,		\
+		.width	= hpp__width_fn,		\
+		.color	= hpp__color_ ## _fn,		\
+		.entry	= hpp__entry_ ## _fn,		\
+		.cmp	= hpp__nop_cmp,			\
+		.collapse = hpp__nop_cmp,		\
+		.sort	= hpp__sort_ ## _fn,		\
+	}
+
+#define HPP__COLOR_ACC_PRINT_FNS(_name, _fn)		\
+	{						\
+		.name   = _name,			\
+		.header	= hpp__header_fn,		\
+		.width	= hpp__width_fn,		\
+		.color	= hpp__color_ ## _fn,		\
+		.entry	= hpp__entry_ ## _fn,		\
+		.cmp	= hpp__nop_cmp,			\
+		.collapse = hpp__nop_cmp,		\
+		.sort	= hpp__sort_ ## _fn,		\
+	}
+
+#define HPP__PRINT_FNS(_name, _fn)			\
+	{						\
+		.name   = _name,			\
+		.header	= hpp__header_fn,		\
+		.width	= hpp__width_fn,		\
+		.entry	= hpp__entry_ ## _fn,		\
+		.cmp	= hpp__nop_cmp,			\
+		.collapse = hpp__nop_cmp,		\
+		.sort	= hpp__sort_ ## _fn,		\
+	}
+
+struct perf_hpp_fmt perf_hpp__format[] = {
+	HPP__COLOR_PRINT_FNS("Overhead", overhead),
+	HPP__COLOR_PRINT_FNS("sys", overhead_sys),
+	HPP__COLOR_PRINT_FNS("usr", overhead_us),
+	HPP__COLOR_PRINT_FNS("guest sys", overhead_guest_sys),
+	HPP__COLOR_PRINT_FNS("guest usr", overhead_guest_us),
+	HPP__COLOR_ACC_PRINT_FNS("Children", overhead_acc),
+	HPP__PRINT_FNS("Samples", samples),
+	HPP__PRINT_FNS("Period", period)
+};
+
+LIST_HEAD(perf_hpp__list);
+LIST_HEAD(perf_hpp__sort_list);
+
+
+#undef HPP__COLOR_PRINT_FNS
+#undef HPP__COLOR_ACC_PRINT_FNS
+#undef HPP__PRINT_FNS
+
+#undef HPP_PERCENT_FNS
+#undef HPP_PERCENT_ACC_FNS
+#undef HPP_RAW_FNS
+
+#undef __HPP_HEADER_FN
+#undef __HPP_WIDTH_FN
+#undef __HPP_COLOR_PERCENT_FN
+#undef __HPP_ENTRY_PERCENT_FN
+#undef __HPP_COLOR_ACC_PERCENT_FN
+#undef __HPP_ENTRY_ACC_PERCENT_FN
+#undef __HPP_ENTRY_RAW_FN
+#undef __HPP_SORT_FN
+#undef __HPP_SORT_ACC_FN
+#undef __HPP_SORT_RAW_FN
+
+
+void perf_hpp__init(void)
+{
+	struct list_head *list;
+	int i;
+
+	for (i = 0; i < PERF_HPP__MAX_INDEX; i++) {
+		struct perf_hpp_fmt *fmt = &perf_hpp__format[i];
+
+		INIT_LIST_HEAD(&fmt->list);
+
+		/* sort_list may be linked by setup_sorting() */
+		if (fmt->sort_list.next == NULL)
+			INIT_LIST_HEAD(&fmt->sort_list);
+	}
+
+	/*
+	 * If user specified field order, no need to setup default fields.
+	 */
+	if (is_strict_order(field_order))
+		return;
+
+	if (symbol_conf.cumulate_callchain) {
+		perf_hpp__column_enable(PERF_HPP__OVERHEAD_ACC);
+		perf_hpp__format[PERF_HPP__OVERHEAD].name = "Self";
+	}
+
+	perf_hpp__column_enable(PERF_HPP__OVERHEAD);
+
+	if (symbol_conf.show_cpu_utilization) {
+		perf_hpp__column_enable(PERF_HPP__OVERHEAD_SYS);
+		perf_hpp__column_enable(PERF_HPP__OVERHEAD_US);
+
+		if (perf_guest) {
+			perf_hpp__column_enable(PERF_HPP__OVERHEAD_GUEST_SYS);
+			perf_hpp__column_enable(PERF_HPP__OVERHEAD_GUEST_US);
+		}
+	}
+
+	if (symbol_conf.show_nr_samples)
+		perf_hpp__column_enable(PERF_HPP__SAMPLES);
+
+	if (symbol_conf.show_total_period)
+		perf_hpp__column_enable(PERF_HPP__PERIOD);
+
+	/* prepend overhead field for backward compatiblity.  */
+	list = &perf_hpp__format[PERF_HPP__OVERHEAD].sort_list;
+	if (list_empty(list))
+		list_add(list, &perf_hpp__sort_list);
+
+	if (symbol_conf.cumulate_callchain) {
+		list = &perf_hpp__format[PERF_HPP__OVERHEAD_ACC].sort_list;
+		if (list_empty(list))
+			list_add(list, &perf_hpp__sort_list);
+	}
+}
+
+void perf_hpp__column_register(struct perf_hpp_fmt *format)
+{
+	list_add_tail(&format->list, &perf_hpp__list);
+}
+
+void perf_hpp__column_unregister(struct perf_hpp_fmt *format)
+{
+	list_del(&format->list);
+}
+
+void perf_hpp__register_sort_field(struct perf_hpp_fmt *format)
+{
+	list_add_tail(&format->sort_list, &perf_hpp__sort_list);
+}
+
+void perf_hpp__column_enable(unsigned col)
+{
+	BUG_ON(col >= PERF_HPP__MAX_INDEX);
+	perf_hpp__column_register(&perf_hpp__format[col]);
+}
+
+void perf_hpp__column_disable(unsigned col)
+{
+	BUG_ON(col >= PERF_HPP__MAX_INDEX);
+	perf_hpp__column_unregister(&perf_hpp__format[col]);
+}
+
+void perf_hpp__cancel_cumulate(void)
+{
+	if (is_strict_order(field_order))
+		return;
+
+	perf_hpp__column_disable(PERF_HPP__OVERHEAD_ACC);
+	perf_hpp__format[PERF_HPP__OVERHEAD].name = "Overhead";
+}
+
+void perf_hpp__setup_output_field(void)
+{
+	struct perf_hpp_fmt *fmt;
+
+	/* append sort keys to output field */
+	perf_hpp__for_each_sort_list(fmt) {
+		if (!list_empty(&fmt->list))
+			continue;
+
+		/*
+		 * sort entry fields are dynamically created,
+		 * so they can share a same sort key even though
+		 * the list is empty.
+		 */
+		if (perf_hpp__is_sort_entry(fmt)) {
+			struct perf_hpp_fmt *pos;
+
+			perf_hpp__for_each_format(pos) {
+				if (perf_hpp__same_sort_entry(pos, fmt))
+					goto next;
+			}
+		}
+
+		perf_hpp__column_register(fmt);
+next:
+		continue;
+	}
+}
+
+void perf_hpp__append_sort_keys(void)
+{
+	struct perf_hpp_fmt *fmt;
+
+	/* append output fields to sort keys */
+	perf_hpp__for_each_format(fmt) {
+		if (!list_empty(&fmt->sort_list))
+			continue;
+
+		/*
+		 * sort entry fields are dynamically created,
+		 * so they can share a same sort key even though
+		 * the list is empty.
+		 */
+		if (perf_hpp__is_sort_entry(fmt)) {
+			struct perf_hpp_fmt *pos;
+
+			perf_hpp__for_each_sort_list(pos) {
+				if (perf_hpp__same_sort_entry(pos, fmt))
+					goto next;
+			}
+		}
+
+		perf_hpp__register_sort_field(fmt);
+next:
+		continue;
+	}
+}
+
+void perf_hpp__reset_output_field(void)
+{
+	struct perf_hpp_fmt *fmt, *tmp;
+
+	/* reset output fields */
+	perf_hpp__for_each_format_safe(fmt, tmp) {
+		list_del_init(&fmt->list);
+		list_del_init(&fmt->sort_list);
+	}
+
+	/* reset sort keys */
+	perf_hpp__for_each_sort_list_safe(fmt, tmp) {
+		list_del_init(&fmt->list);
+		list_del_init(&fmt->sort_list);
+	}
+}
+
+/*
+ * See hists__fprintf to match the column widths
+ */
+unsigned int hists__sort_list_width(struct hists *hists)
+{
+	struct perf_hpp_fmt *fmt;
+	int ret = 0;
+	bool first = true;
+	struct perf_hpp dummy_hpp;
+
+	perf_hpp__for_each_format(fmt) {
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		if (first)
+			first = false;
+		else
+			ret += 2;
+
+		ret += fmt->width(fmt, &dummy_hpp, hists_to_evsel(hists));
+	}
+
+	if (verbose && sort__has_sym) /* Addr + origin */
+		ret += 3 + BITS_PER_LONG / 4;
+
+	return ret;
+}
+
+void perf_hpp__reset_width(struct perf_hpp_fmt *fmt, struct hists *hists)
+{
+	int idx;
+
+	if (perf_hpp__is_sort_entry(fmt))
+		return perf_hpp__reset_sort_width(fmt, hists);
+
+	for (idx = 0; idx < PERF_HPP__MAX_INDEX; idx++) {
+		if (fmt == &perf_hpp__format[idx])
+			break;
+	}
+
+	if (idx == PERF_HPP__MAX_INDEX)
+		return;
+
+	switch (idx) {
+	case PERF_HPP__OVERHEAD:
+	case PERF_HPP__OVERHEAD_SYS:
+	case PERF_HPP__OVERHEAD_US:
+	case PERF_HPP__OVERHEAD_ACC:
+		fmt->len = 8;
+		break;
+
+	case PERF_HPP__OVERHEAD_GUEST_SYS:
+	case PERF_HPP__OVERHEAD_GUEST_US:
+		fmt->len = 9;
+		break;
+
+	case PERF_HPP__SAMPLES:
+	case PERF_HPP__PERIOD:
+		fmt->len = 12;
+		break;
+
+	default:
+		break;
+	}
+}
+
+void perf_hpp__set_user_width(const char *width_list_str)
+{
+	struct perf_hpp_fmt *fmt;
+	const char *ptr = width_list_str;
+
+	perf_hpp__for_each_format(fmt) {
+		char *p;
+
+		int len = strtol(ptr, &p, 10);
+		fmt->user_len = len;
+
+		if (*p == ',')
+			ptr = p + 1;
+		else
+			break;
+	}
+}
diff --git a/tools/perf/ui/keysyms.h b/tools/perf/ui/keysyms.h
new file mode 100644
index 000000000..65092d576
--- /dev/null
+++ b/tools/perf/ui/keysyms.h
@@ -0,0 +1,28 @@
+#ifndef _PERF_KEYSYMS_H_
+#define _PERF_KEYSYMS_H_ 1
+
+#include "libslang.h"
+
+#define K_DOWN	SL_KEY_DOWN
+#define K_END	SL_KEY_END
+#define K_ENTER	'\r'
+#define K_ESC	033
+#define K_F1	SL_KEY_F(1)
+#define K_HOME	SL_KEY_HOME
+#define K_LEFT	SL_KEY_LEFT
+#define K_PGDN	SL_KEY_NPAGE
+#define K_PGUP	SL_KEY_PPAGE
+#define K_RIGHT	SL_KEY_RIGHT
+#define K_TAB	'\t'
+#define K_UNTAB	SL_KEY_UNTAB
+#define K_UP	SL_KEY_UP
+#define K_BKSPC 0x7f
+#define K_DEL	SL_KEY_DELETE
+
+/* Not really keys */
+#define K_TIMER	 -1
+#define K_ERROR	 -2
+#define K_RESIZE -3
+#define K_SWITCH_INPUT_DATA -4
+
+#endif /* _PERF_KEYSYMS_H_ */
diff --git a/tools/perf/ui/libslang.h b/tools/perf/ui/libslang.h
new file mode 100644
index 000000000..4d54b6450
--- /dev/null
+++ b/tools/perf/ui/libslang.h
@@ -0,0 +1,29 @@
+#ifndef _PERF_UI_SLANG_H_
+#define _PERF_UI_SLANG_H_ 1
+/*
+ * slang versions <= 2.0.6 have a "#if HAVE_LONG_LONG" that breaks
+ * the build if it isn't defined. Use the equivalent one that glibc
+ * has on features.h.
+ */
+#include <features.h>
+#ifndef HAVE_LONG_LONG
+#define HAVE_LONG_LONG __GLIBC_HAVE_LONG_LONG
+#endif
+#include <slang.h>
+
+#if SLANG_VERSION < 20104
+#define slsmg_printf(msg, args...) \
+	SLsmg_printf((char *)(msg), ##args)
+#define slsmg_write_nstring(msg, len) \
+	SLsmg_write_nstring((char *)(msg), len)
+#define sltt_set_color(obj, name, fg, bg) \
+	SLtt_set_color(obj,(char *)(name), (char *)(fg), (char *)(bg))
+#else
+#define slsmg_printf SLsmg_printf
+#define slsmg_write_nstring SLsmg_write_nstring
+#define sltt_set_color SLtt_set_color
+#endif
+
+#define SL_KEY_UNTAB 0x1000
+
+#endif /* _PERF_UI_SLANG_H_ */
diff --git a/tools/perf/ui/progress.c b/tools/perf/ui/progress.c
new file mode 100644
index 000000000..a0f24c711
--- /dev/null
+++ b/tools/perf/ui/progress.c
@@ -0,0 +1,38 @@
+#include "../cache.h"
+#include "progress.h"
+
+static void null_progress__update(struct ui_progress *p __maybe_unused)
+{
+}
+
+static struct ui_progress_ops null_progress__ops =
+{
+	.update = null_progress__update,
+};
+
+struct ui_progress_ops *ui_progress__ops = &null_progress__ops;
+
+void ui_progress__update(struct ui_progress *p, u64 adv)
+{
+	p->curr += adv;
+
+	if (p->curr >= p->next) {
+		p->next += p->step;
+		ui_progress__ops->update(p);
+	}
+}
+
+void ui_progress__init(struct ui_progress *p, u64 total, const char *title)
+{
+	p->curr = 0;
+	p->next = p->step = total / 16;
+	p->total = total;
+	p->title = title;
+
+}
+
+void ui_progress__finish(void)
+{
+	if (ui_progress__ops->finish)
+		ui_progress__ops->finish();
+}
diff --git a/tools/perf/ui/progress.h b/tools/perf/ui/progress.h
new file mode 100644
index 000000000..717d39d30
--- /dev/null
+++ b/tools/perf/ui/progress.h
@@ -0,0 +1,23 @@
+#ifndef _PERF_UI_PROGRESS_H_
+#define _PERF_UI_PROGRESS_H_ 1
+
+#include <linux/types.h>
+
+void ui_progress__finish(void);
+
+struct ui_progress {
+	const char *title;
+	u64 curr, next, step, total;
+};
+
+void ui_progress__init(struct ui_progress *p, u64 total, const char *title);
+void ui_progress__update(struct ui_progress *p, u64 adv);
+
+struct ui_progress_ops {
+	void (*update)(struct ui_progress *p);
+	void (*finish)(void);
+};
+
+extern struct ui_progress_ops *ui_progress__ops;
+
+#endif
diff --git a/tools/perf/ui/setup.c b/tools/perf/ui/setup.c
new file mode 100644
index 000000000..ba51fa8a1
--- /dev/null
+++ b/tools/perf/ui/setup.c
@@ -0,0 +1,107 @@
+#include <pthread.h>
+#include <dlfcn.h>
+
+#include "../util/cache.h"
+#include "../util/debug.h"
+#include "../util/hist.h"
+
+pthread_mutex_t ui__lock = PTHREAD_MUTEX_INITIALIZER;
+void *perf_gtk_handle;
+
+#ifdef HAVE_GTK2_SUPPORT
+static int setup_gtk_browser(void)
+{
+	int (*perf_ui_init)(void);
+
+	if (perf_gtk_handle)
+		return 0;
+
+	perf_gtk_handle = dlopen(PERF_GTK_DSO, RTLD_LAZY);
+	if (perf_gtk_handle == NULL) {
+		char buf[PATH_MAX];
+		scnprintf(buf, sizeof(buf), "%s/%s", LIBDIR, PERF_GTK_DSO);
+		perf_gtk_handle = dlopen(buf, RTLD_LAZY);
+	}
+	if (perf_gtk_handle == NULL)
+		return -1;
+
+	perf_ui_init = dlsym(perf_gtk_handle, "perf_gtk__init");
+	if (perf_ui_init == NULL)
+		goto out_close;
+
+	if (perf_ui_init() == 0)
+		return 0;
+
+out_close:
+	dlclose(perf_gtk_handle);
+	return -1;
+}
+
+static void exit_gtk_browser(bool wait_for_ok)
+{
+	void (*perf_ui_exit)(bool);
+
+	if (perf_gtk_handle == NULL)
+		return;
+
+	perf_ui_exit = dlsym(perf_gtk_handle, "perf_gtk__exit");
+	if (perf_ui_exit == NULL)
+		goto out_close;
+
+	perf_ui_exit(wait_for_ok);
+
+out_close:
+	dlclose(perf_gtk_handle);
+
+	perf_gtk_handle = NULL;
+}
+#else
+static inline int setup_gtk_browser(void) { return -1; }
+static inline void exit_gtk_browser(bool wait_for_ok __maybe_unused) {}
+#endif
+
+void setup_browser(bool fallback_to_pager)
+{
+	if (use_browser < 2 && (!isatty(1) || dump_trace))
+		use_browser = 0;
+
+	/* default to TUI */
+	if (use_browser < 0)
+		use_browser = 1;
+
+	switch (use_browser) {
+	case 2:
+		if (setup_gtk_browser() == 0)
+			break;
+		printf("GTK browser requested but could not find %s\n",
+		       PERF_GTK_DSO);
+		sleep(1);
+		/* fall through */
+	case 1:
+		use_browser = 1;
+		if (ui__init() == 0)
+			break;
+		/* fall through */
+	default:
+		use_browser = 0;
+		if (fallback_to_pager)
+			setup_pager();
+		break;
+	}
+}
+
+void exit_browser(bool wait_for_ok)
+{
+	switch (use_browser) {
+	case 2:
+		exit_gtk_browser(wait_for_ok);
+		break;
+
+	case 1:
+		ui__exit(wait_for_ok);
+		break;
+
+	default:
+		break;
+	}
+}
diff --git a/tools/perf/ui/stdio/hist.c b/tools/perf/ui/stdio/hist.c
new file mode 100644
index 000000000..dfcbc9014
--- /dev/null
+++ b/tools/perf/ui/stdio/hist.c
@@ -0,0 +1,511 @@
+#include <stdio.h>
+
+#include "../../util/util.h"
+#include "../../util/hist.h"
+#include "../../util/sort.h"
+#include "../../util/evsel.h"
+
+
+static size_t callchain__fprintf_left_margin(FILE *fp, int left_margin)
+{
+	int i;
+	int ret = fprintf(fp, "            ");
+
+	for (i = 0; i < left_margin; i++)
+		ret += fprintf(fp, " ");
+
+	return ret;
+}
+
+static size_t ipchain__fprintf_graph_line(FILE *fp, int depth, int depth_mask,
+					  int left_margin)
+{
+	int i;
+	size_t ret = callchain__fprintf_left_margin(fp, left_margin);
+
+	for (i = 0; i < depth; i++)
+		if (depth_mask & (1 << i))
+			ret += fprintf(fp, "|          ");
+		else
+			ret += fprintf(fp, "           ");
+
+	ret += fprintf(fp, "\n");
+
+	return ret;
+}
+
+static size_t ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain,
+				     int depth, int depth_mask, int period,
+				     u64 total_samples, u64 hits,
+				     int left_margin)
+{
+	int i;
+	size_t ret = 0;
+	char bf[1024];
+
+	ret += callchain__fprintf_left_margin(fp, left_margin);
+	for (i = 0; i < depth; i++) {
+		if (depth_mask & (1 << i))
+			ret += fprintf(fp, "|");
+		else
+			ret += fprintf(fp, " ");
+		if (!period && i == depth - 1) {
+			double percent;
+
+			percent = hits * 100.0 / total_samples;
+			ret += percent_color_fprintf(fp, "--%2.2f%%-- ", percent);
+		} else
+			ret += fprintf(fp, "%s", "          ");
+	}
+	fputs(callchain_list__sym_name(chain, bf, sizeof(bf), false), fp);
+	fputc('\n', fp);
+	return ret;
+}
+
+static struct symbol *rem_sq_bracket;
+static struct callchain_list rem_hits;
+
+static void init_rem_hits(void)
+{
+	rem_sq_bracket = malloc(sizeof(*rem_sq_bracket) + 6);
+	if (!rem_sq_bracket) {
+		fprintf(stderr, "Not enough memory to display remaining hits\n");
+		return;
+	}
+
+	strcpy(rem_sq_bracket->name, "[...]");
+	rem_hits.ms.sym = rem_sq_bracket;
+}
+
+static size_t __callchain__fprintf_graph(FILE *fp, struct rb_root *root,
+					 u64 total_samples, int depth,
+					 int depth_mask, int left_margin)
+{
+	struct rb_node *node, *next;
+	struct callchain_node *child;
+	struct callchain_list *chain;
+	int new_depth_mask = depth_mask;
+	u64 remaining;
+	size_t ret = 0;
+	int i;
+	uint entries_printed = 0;
+
+	remaining = total_samples;
+
+	node = rb_first(root);
+	while (node) {
+		u64 new_total;
+		u64 cumul;
+
+		child = rb_entry(node, struct callchain_node, rb_node);
+		cumul = callchain_cumul_hits(child);
+		remaining -= cumul;
+
+		/*
+		 * The depth mask manages the output of pipes that show
+		 * the depth. We don't want to keep the pipes of the current
+		 * level for the last child of this depth.
+		 * Except if we have remaining filtered hits. They will
+		 * supersede the last child
+		 */
+		next = rb_next(node);
+		if (!next && (callchain_param.mode != CHAIN_GRAPH_REL || !remaining))
+			new_depth_mask &= ~(1 << (depth - 1));
+
+		/*
+		 * But we keep the older depth mask for the line separator
+		 * to keep the level link until we reach the last child
+		 */
+		ret += ipchain__fprintf_graph_line(fp, depth, depth_mask,
+						   left_margin);
+		i = 0;
+		list_for_each_entry(chain, &child->val, list) {
+			ret += ipchain__fprintf_graph(fp, chain, depth,
+						      new_depth_mask, i++,
+						      total_samples,
+						      cumul,
+						      left_margin);
+		}
+
+		if (callchain_param.mode == CHAIN_GRAPH_REL)
+			new_total = child->children_hit;
+		else
+			new_total = total_samples;
+
+		ret += __callchain__fprintf_graph(fp, &child->rb_root, new_total,
+						  depth + 1,
+						  new_depth_mask | (1 << depth),
+						  left_margin);
+		node = next;
+		if (++entries_printed == callchain_param.print_limit)
+			break;
+	}
+
+	if (callchain_param.mode == CHAIN_GRAPH_REL &&
+		remaining && remaining != total_samples) {
+
+		if (!rem_sq_bracket)
+			return ret;
+
+		new_depth_mask &= ~(1 << (depth - 1));
+		ret += ipchain__fprintf_graph(fp, &rem_hits, depth,
+					      new_depth_mask, 0, total_samples,
+					      remaining, left_margin);
+	}
+
+	return ret;
+}
+
+static size_t callchain__fprintf_graph(FILE *fp, struct rb_root *root,
+				       u64 total_samples, int left_margin)
+{
+	struct callchain_node *cnode;
+	struct callchain_list *chain;
+	u32 entries_printed = 0;
+	bool printed = false;
+	struct rb_node *node;
+	int i = 0;
+	int ret = 0;
+	char bf[1024];
+
+	/*
+	 * If have one single callchain root, don't bother printing
+	 * its percentage (100 % in fractal mode and the same percentage
+	 * than the hist in graph mode). This also avoid one level of column.
+	 */
+	node = rb_first(root);
+	if (node && !rb_next(node)) {
+		cnode = rb_entry(node, struct callchain_node, rb_node);
+		list_for_each_entry(chain, &cnode->val, list) {
+			/*
+			 * If we sort by symbol, the first entry is the same than
+			 * the symbol. No need to print it otherwise it appears as
+			 * displayed twice.
+			 */
+			if (!i++ && field_order == NULL &&
+			    sort_order && !prefixcmp(sort_order, "sym"))
+				continue;
+			if (!printed) {
+				ret += callchain__fprintf_left_margin(fp, left_margin);
+				ret += fprintf(fp, "|\n");
+				ret += callchain__fprintf_left_margin(fp, left_margin);
+				ret += fprintf(fp, "---");
+				left_margin += 3;
+				printed = true;
+			} else
+				ret += callchain__fprintf_left_margin(fp, left_margin);
+
+			ret += fprintf(fp, "%s\n", callchain_list__sym_name(chain, bf, sizeof(bf),
+							false));
+
+			if (++entries_printed == callchain_param.print_limit)
+				break;
+		}
+		root = &cnode->rb_root;
+	}
+
+	ret += __callchain__fprintf_graph(fp, root, total_samples,
+					  1, 1, left_margin);
+	ret += fprintf(fp, "\n");
+
+	return ret;
+}
+
+static size_t __callchain__fprintf_flat(FILE *fp, struct callchain_node *node,
+					u64 total_samples)
+{
+	struct callchain_list *chain;
+	size_t ret = 0;
+	char bf[1024];
+
+	if (!node)
+		return 0;
+
+	ret += __callchain__fprintf_flat(fp, node->parent, total_samples);
+
+
+	list_for_each_entry(chain, &node->val, list) {
+		if (chain->ip >= PERF_CONTEXT_MAX)
+			continue;
+		ret += fprintf(fp, "                %s\n", callchain_list__sym_name(chain,
+					bf, sizeof(bf), false));
+	}
+
+	return ret;
+}
+
+static size_t callchain__fprintf_flat(FILE *fp, struct rb_root *tree,
+				      u64 total_samples)
+{
+	size_t ret = 0;
+	u32 entries_printed = 0;
+	struct callchain_node *chain;
+	struct rb_node *rb_node = rb_first(tree);
+
+	while (rb_node) {
+		double percent;
+
+		chain = rb_entry(rb_node, struct callchain_node, rb_node);
+		percent = chain->hit * 100.0 / total_samples;
+
+		ret = percent_color_fprintf(fp, "           %6.2f%%\n", percent);
+		ret += __callchain__fprintf_flat(fp, chain, total_samples);
+		ret += fprintf(fp, "\n");
+		if (++entries_printed == callchain_param.print_limit)
+			break;
+
+		rb_node = rb_next(rb_node);
+	}
+
+	return ret;
+}
+
+static size_t hist_entry_callchain__fprintf(struct hist_entry *he,
+					    u64 total_samples, int left_margin,
+					    FILE *fp)
+{
+	switch (callchain_param.mode) {
+	case CHAIN_GRAPH_REL:
+		return callchain__fprintf_graph(fp, &he->sorted_chain,
+						symbol_conf.cumulate_callchain ?
+						he->stat_acc->period : he->stat.period,
+						left_margin);
+		break;
+	case CHAIN_GRAPH_ABS:
+		return callchain__fprintf_graph(fp, &he->sorted_chain, total_samples,
+						left_margin);
+		break;
+	case CHAIN_FLAT:
+		return callchain__fprintf_flat(fp, &he->sorted_chain, total_samples);
+		break;
+	case CHAIN_NONE:
+		break;
+	default:
+		pr_err("Bad callchain mode\n");
+	}
+
+	return 0;
+}
+
+static size_t hist_entry__callchain_fprintf(struct hist_entry *he,
+					    struct hists *hists,
+					    FILE *fp)
+{
+	int left_margin = 0;
+	u64 total_period = hists->stats.total_period;
+
+	if (field_order == NULL && (sort_order == NULL ||
+				    !prefixcmp(sort_order, "comm"))) {
+		struct perf_hpp_fmt *fmt;
+
+		perf_hpp__for_each_format(fmt) {
+			if (!perf_hpp__is_sort_entry(fmt))
+				continue;
+
+			/* must be 'comm' sort entry */
+			left_margin = fmt->width(fmt, NULL, hists_to_evsel(hists));
+			left_margin -= thread__comm_len(he->thread);
+			break;
+		}
+	}
+	return hist_entry_callchain__fprintf(he, total_period, left_margin, fp);
+}
+
+static int hist_entry__snprintf(struct hist_entry *he, struct perf_hpp *hpp)
+{
+	const char *sep = symbol_conf.field_sep;
+	struct perf_hpp_fmt *fmt;
+	char *start = hpp->buf;
+	int ret;
+	bool first = true;
+
+	if (symbol_conf.exclude_other && !he->parent)
+		return 0;
+
+	perf_hpp__for_each_format(fmt) {
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		/*
+		 * If there's no field_sep, we still need
+		 * to display initial '  '.
+		 */
+		if (!sep || !first) {
+			ret = scnprintf(hpp->buf, hpp->size, "%s", sep ?: "  ");
+			advance_hpp(hpp, ret);
+		} else
+			first = false;
+
+		if (perf_hpp__use_color() && fmt->color)
+			ret = fmt->color(fmt, hpp, he);
+		else
+			ret = fmt->entry(fmt, hpp, he);
+
+		advance_hpp(hpp, ret);
+	}
+
+	return hpp->buf - start;
+}
+
+static int hist_entry__fprintf(struct hist_entry *he, size_t size,
+			       struct hists *hists,
+			       char *bf, size_t bfsz, FILE *fp)
+{
+	int ret;
+	struct perf_hpp hpp = {
+		.buf		= bf,
+		.size		= size,
+	};
+
+	if (size == 0 || size > bfsz)
+		size = hpp.size = bfsz;
+
+	hist_entry__snprintf(he, &hpp);
+
+	ret = fprintf(fp, "%s\n", bf);
+
+	if (symbol_conf.use_callchain)
+		ret += hist_entry__callchain_fprintf(he, hists, fp);
+
+	return ret;
+}
+
+size_t hists__fprintf(struct hists *hists, bool show_header, int max_rows,
+		      int max_cols, float min_pcnt, FILE *fp)
+{
+	struct perf_hpp_fmt *fmt;
+	struct rb_node *nd;
+	size_t ret = 0;
+	unsigned int width;
+	const char *sep = symbol_conf.field_sep;
+	int nr_rows = 0;
+	char bf[96];
+	struct perf_hpp dummy_hpp = {
+		.buf	= bf,
+		.size	= sizeof(bf),
+	};
+	bool first = true;
+	size_t linesz;
+	char *line = NULL;
+
+	init_rem_hits();
+
+	perf_hpp__for_each_format(fmt)
+		perf_hpp__reset_width(fmt, hists);
+
+	if (symbol_conf.col_width_list_str)
+		perf_hpp__set_user_width(symbol_conf.col_width_list_str);
+
+	if (!show_header)
+		goto print_entries;
+
+	fprintf(fp, "# ");
+
+	perf_hpp__for_each_format(fmt) {
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		if (!first)
+			fprintf(fp, "%s", sep ?: "  ");
+		else
+			first = false;
+
+		fmt->header(fmt, &dummy_hpp, hists_to_evsel(hists));
+		fprintf(fp, "%s", bf);
+	}
+
+	fprintf(fp, "\n");
+	if (max_rows && ++nr_rows >= max_rows)
+		goto out;
+
+	if (sep)
+		goto print_entries;
+
+	first = true;
+
+	fprintf(fp, "# ");
+
+	perf_hpp__for_each_format(fmt) {
+		unsigned int i;
+
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		if (!first)
+			fprintf(fp, "%s", sep ?: "  ");
+		else
+			first = false;
+
+		width = fmt->width(fmt, &dummy_hpp, hists_to_evsel(hists));
+		for (i = 0; i < width; i++)
+			fprintf(fp, ".");
+	}
+
+	fprintf(fp, "\n");
+	if (max_rows && ++nr_rows >= max_rows)
+		goto out;
+
+	fprintf(fp, "#\n");
+	if (max_rows && ++nr_rows >= max_rows)
+		goto out;
+
+print_entries:
+	linesz = hists__sort_list_width(hists) + 3 + 1;
+	linesz += perf_hpp__color_overhead();
+	line = malloc(linesz);
+	if (line == NULL) {
+		ret = -1;
+		goto out;
+	}
+
+	for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+		float percent;
+
+		if (h->filtered)
+			continue;
+
+		percent = hist_entry__get_percent_limit(h);
+		if (percent < min_pcnt)
+			continue;
+
+		ret += hist_entry__fprintf(h, max_cols, hists, line, linesz, fp);
+
+		if (max_rows && ++nr_rows >= max_rows)
+			break;
+
+		if (h->ms.map == NULL && verbose > 1) {
+			__map_groups__fprintf_maps(h->thread->mg,
+						   MAP__FUNCTION, fp);
+			fprintf(fp, "%.10s end\n", graph_dotted_line);
+		}
+	}
+
+	free(line);
+out:
+	zfree(&rem_sq_bracket);
+
+	return ret;
+}
+
+size_t events_stats__fprintf(struct events_stats *stats, FILE *fp)
+{
+	int i;
+	size_t ret = 0;
+
+	for (i = 0; i < PERF_RECORD_HEADER_MAX; ++i) {
+		const char *name;
+
+		if (stats->nr_events[i] == 0)
+			continue;
+
+		name = perf_event__name(i);
+		if (!strcmp(name, "UNKNOWN"))
+			continue;
+
+		ret += fprintf(fp, "%16s events: %10d\n", name,
+			       stats->nr_events[i]);
+	}
+
+	return ret;
+}
diff --git a/tools/perf/ui/tui/Build b/tools/perf/ui/tui/Build
new file mode 100644
index 000000000..9e4c6ca41
--- /dev/null
+++ b/tools/perf/ui/tui/Build
@@ -0,0 +1,4 @@
+libperf-y += setup.o
+libperf-y += util.o
+libperf-y += helpline.o
+libperf-y += progress.o
diff --git a/tools/perf/ui/tui/helpline.c b/tools/perf/ui/tui/helpline.c
new file mode 100644
index 000000000..88f5143a5
--- /dev/null
+++ b/tools/perf/ui/tui/helpline.c
@@ -0,0 +1,61 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <pthread.h>
+
+#include "../../util/debug.h"
+#include "../helpline.h"
+#include "../ui.h"
+#include "../libslang.h"
+
+char ui_helpline__last_msg[1024];
+bool tui_helpline__set;
+
+static void tui_helpline__pop(void)
+{
+}
+
+static void tui_helpline__push(const char *msg)
+{
+	const size_t sz = sizeof(ui_helpline__current);
+
+	SLsmg_gotorc(SLtt_Screen_Rows - 1, 0);
+	SLsmg_set_color(0);
+	SLsmg_write_nstring((char *)msg, SLtt_Screen_Cols);
+	SLsmg_refresh();
+	strncpy(ui_helpline__current, msg, sz)[sz - 1] = '\0';
+}
+
+static int tui_helpline__show(const char *format, va_list ap)
+{
+	int ret;
+	static int backlog;
+
+	pthread_mutex_lock(&ui__lock);
+	ret = vscnprintf(ui_helpline__last_msg + backlog,
+			sizeof(ui_helpline__last_msg) - backlog, format, ap);
+	backlog += ret;
+
+	tui_helpline__set = true;
+
+	if (ui_helpline__last_msg[backlog - 1] == '\n') {
+		ui_helpline__puts(ui_helpline__last_msg);
+		SLsmg_refresh();
+		backlog = 0;
+	}
+	pthread_mutex_unlock(&ui__lock);
+
+	return ret;
+}
+
+struct ui_helpline tui_helpline_fns = {
+	.pop	= tui_helpline__pop,
+	.push	= tui_helpline__push,
+	.show	= tui_helpline__show,
+};
+
+void ui_helpline__init(void)
+{
+	helpline_fns = &tui_helpline_fns;
+	ui_helpline__puts(" ");
+}
diff --git a/tools/perf/ui/tui/progress.c b/tools/perf/ui/tui/progress.c
new file mode 100644
index 000000000..c61d14b10
--- /dev/null
+++ b/tools/perf/ui/tui/progress.c
@@ -0,0 +1,44 @@
+#include "../cache.h"
+#include "../progress.h"
+#include "../libslang.h"
+#include "../ui.h"
+#include "tui.h"
+#include "../browser.h"
+
+static void tui_progress__update(struct ui_progress *p)
+{
+	int bar, y;
+	/*
+	 * FIXME: We should have a per UI backend way of showing progress,
+	 * stdio will just show a percentage as NN%, etc.
+	 */
+	if (use_browser <= 0)
+		return;
+
+	if (p->total == 0)
+		return;
+
+	ui__refresh_dimensions(false);
+	pthread_mutex_lock(&ui__lock);
+	y = SLtt_Screen_Rows / 2 - 2;
+	SLsmg_set_color(0);
+	SLsmg_draw_box(y, 0, 3, SLtt_Screen_Cols);
+	SLsmg_gotorc(y++, 1);
+	SLsmg_write_string((char *)p->title);
+	SLsmg_fill_region(y, 1, 1, SLtt_Screen_Cols - 2, ' ');
+	SLsmg_set_color(HE_COLORSET_SELECTED);
+	bar = ((SLtt_Screen_Cols - 2) * p->curr) / p->total;
+	SLsmg_fill_region(y, 1, 1, bar, ' ');
+	SLsmg_refresh();
+	pthread_mutex_unlock(&ui__lock);
+}
+
+static struct ui_progress_ops tui_progress__ops =
+{
+	.update		= tui_progress__update,
+};
+
+void tui_progress__init(void)
+{
+	ui_progress__ops = &tui_progress__ops;
+}
diff --git a/tools/perf/ui/tui/setup.c b/tools/perf/ui/tui/setup.c
new file mode 100644
index 000000000..b77e1d771
--- /dev/null
+++ b/tools/perf/ui/tui/setup.c
@@ -0,0 +1,174 @@
+#include <signal.h>
+#include <stdbool.h>
+#ifdef HAVE_BACKTRACE_SUPPORT
+#include <execinfo.h>
+#endif
+
+#include "../../util/cache.h"
+#include "../../util/debug.h"
+#include "../browser.h"
+#include "../helpline.h"
+#include "../ui.h"
+#include "../util.h"
+#include "../libslang.h"
+#include "../keysyms.h"
+#include "tui.h"
+
+static volatile int ui__need_resize;
+
+extern struct perf_error_ops perf_tui_eops;
+extern bool tui_helpline__set;
+
+extern void hist_browser__init_hpp(void);
+
+void ui__refresh_dimensions(bool force)
+{
+	if (force || ui__need_resize) {
+		ui__need_resize = 0;
+		pthread_mutex_lock(&ui__lock);
+		SLtt_get_screen_size();
+		SLsmg_reinit_smg();
+		pthread_mutex_unlock(&ui__lock);
+	}
+}
+
+static void ui__sigwinch(int sig __maybe_unused)
+{
+	ui__need_resize = 1;
+}
+
+static void ui__setup_sigwinch(void)
+{
+	static bool done;
+
+	if (done)
+		return;
+
+	done = true;
+	pthread__unblock_sigwinch();
+	signal(SIGWINCH, ui__sigwinch);
+}
+
+int ui__getch(int delay_secs)
+{
+	struct timeval timeout, *ptimeout = delay_secs ? &timeout : NULL;
+	fd_set read_set;
+	int err, key;
+
+	ui__setup_sigwinch();
+
+	FD_ZERO(&read_set);
+	FD_SET(0, &read_set);
+
+	if (delay_secs) {
+		timeout.tv_sec = delay_secs;
+		timeout.tv_usec = 0;
+	}
+
+        err = select(1, &read_set, NULL, NULL, ptimeout);
+
+	if (err == 0)
+		return K_TIMER;
+
+	if (err == -1) {
+		if (errno == EINTR)
+			return K_RESIZE;
+		return K_ERROR;
+	}
+
+	key = SLang_getkey();
+	if (key != K_ESC)
+		return key;
+
+	FD_ZERO(&read_set);
+	FD_SET(0, &read_set);
+	timeout.tv_sec = 0;
+	timeout.tv_usec = 20;
+        err = select(1, &read_set, NULL, NULL, &timeout);
+	if (err == 0)
+		return K_ESC;
+
+	SLang_ungetkey(key);
+	return SLkp_getkey();
+}
+
+#ifdef HAVE_BACKTRACE_SUPPORT
+static void ui__signal_backtrace(int sig)
+{
+	void *stackdump[32];
+	size_t size;
+
+	ui__exit(false);
+	psignal(sig, "perf");
+
+	printf("-------- backtrace --------\n");
+	size = backtrace(stackdump, ARRAY_SIZE(stackdump));
+	backtrace_symbols_fd(stackdump, size, STDOUT_FILENO);
+
+	exit(0);
+}
+#else
+# define ui__signal_backtrace  ui__signal
+#endif
+
+static void ui__signal(int sig)
+{
+	ui__exit(false);
+	psignal(sig, "perf");
+	exit(0);
+}
+
+int ui__init(void)
+{
+	int err;
+
+	SLutf8_enable(-1);
+	SLtt_get_terminfo();
+	SLtt_get_screen_size();
+
+	err = SLsmg_init_smg();
+	if (err < 0)
+		goto out;
+	err = SLang_init_tty(0, 0, 0);
+	if (err < 0)
+		goto out;
+
+	err = SLkp_init();
+	if (err < 0) {
+		pr_err("TUI initialization failed.\n");
+		goto out;
+	}
+
+	SLkp_define_keysym((char *)"^(kB)", SL_KEY_UNTAB);
+
+	ui_helpline__init();
+	ui_browser__init();
+	tui_progress__init();
+
+	signal(SIGSEGV, ui__signal_backtrace);
+	signal(SIGFPE, ui__signal_backtrace);
+	signal(SIGINT, ui__signal);
+	signal(SIGQUIT, ui__signal);
+	signal(SIGTERM, ui__signal);
+
+	perf_error__register(&perf_tui_eops);
+
+	hist_browser__init_hpp();
+out:
+	return err;
+}
+
+void ui__exit(bool wait_for_ok)
+{
+	if (wait_for_ok && tui_helpline__set)
+		ui__question_window("Fatal Error",
+				    ui_helpline__last_msg,
+				    "Press any key...", 0);
+
+	SLtt_set_cursor_visibility(1);
+	SLsmg_refresh();
+	SLsmg_reset_smg();
+	SLang_reset_tty();
+
+	perf_error__unregister(&perf_tui_eops);
+}
diff --git a/tools/perf/ui/tui/tui.h b/tools/perf/ui/tui/tui.h
new file mode 100644
index 000000000..18961c7b6
--- /dev/null
+++ b/tools/perf/ui/tui/tui.h
@@ -0,0 +1,6 @@
+#ifndef _PERF_TUI_H_
+#define _PERF_TUI_H_ 1
+
+void tui_progress__init(void);
+
+#endif /* _PERF_TUI_H_ */
diff --git a/tools/perf/ui/tui/util.c b/tools/perf/ui/tui/util.c
new file mode 100644
index 000000000..bf890f72f
--- /dev/null
+++ b/tools/perf/ui/tui/util.c
@@ -0,0 +1,256 @@
+#include "../../util/util.h"
+#include <signal.h>
+#include <stdbool.h>
+#include <string.h>
+#include <sys/ttydefaults.h>
+
+#include "../../util/cache.h"
+#include "../../util/debug.h"
+#include "../browser.h"
+#include "../keysyms.h"
+#include "../helpline.h"
+#include "../ui.h"
+#include "../util.h"
+#include "../libslang.h"
+
+static void ui_browser__argv_write(struct ui_browser *browser,
+				   void *entry, int row)
+{
+	char **arg = entry;
+	bool current_entry = ui_browser__is_current_entry(browser, row);
+
+	ui_browser__set_color(browser, current_entry ? HE_COLORSET_SELECTED :
+						       HE_COLORSET_NORMAL);
+	slsmg_write_nstring(*arg, browser->width);
+}
+
+static int popup_menu__run(struct ui_browser *menu)
+{
+	int key;
+
+	if (ui_browser__show(menu, " ", "ESC: exit, ENTER|->: Select option") < 0)
+		return -1;
+
+	while (1) {
+		key = ui_browser__run(menu, 0);
+
+		switch (key) {
+		case K_RIGHT:
+		case K_ENTER:
+			key = menu->index;
+			break;
+		case K_LEFT:
+		case K_ESC:
+		case 'q':
+		case CTRL('c'):
+			key = -1;
+			break;
+		default:
+			continue;
+		}
+
+		break;
+	}
+
+	ui_browser__hide(menu);
+	return key;
+}
+
+int ui__popup_menu(int argc, char * const argv[])
+{
+	struct ui_browser menu = {
+		.entries    = (void *)argv,
+		.refresh    = ui_browser__argv_refresh,
+		.seek	    = ui_browser__argv_seek,
+		.write	    = ui_browser__argv_write,
+		.nr_entries = argc,
+	};
+
+	return popup_menu__run(&menu);
+}
+
+int ui_browser__input_window(const char *title, const char *text, char *input,
+			     const char *exit_msg, int delay_secs)
+{
+	int x, y, len, key;
+	int max_len = 60, nr_lines = 0;
+	static char buf[50];
+	const char *t;
+
+	t = text;
+	while (1) {
+		const char *sep = strchr(t, '\n');
+
+		if (sep == NULL)
+			sep = strchr(t, '\0');
+		len = sep - t;
+		if (max_len < len)
+			max_len = len;
+		++nr_lines;
+		if (*sep == '\0')
+			break;
+		t = sep + 1;
+	}
+
+	pthread_mutex_lock(&ui__lock);
+
+	max_len += 2;
+	nr_lines += 8;
+	y = SLtt_Screen_Rows / 2 - nr_lines / 2;
+	x = SLtt_Screen_Cols / 2 - max_len / 2;
+
+	SLsmg_set_color(0);
+	SLsmg_draw_box(y, x++, nr_lines, max_len);
+	if (title) {
+		SLsmg_gotorc(y, x + 1);
+		SLsmg_write_string((char *)title);
+	}
+	SLsmg_gotorc(++y, x);
+	nr_lines -= 7;
+	max_len -= 2;
+	SLsmg_write_wrapped_string((unsigned char *)text, y, x,
+				   nr_lines, max_len, 1);
+	y += nr_lines;
+	len = 5;
+	while (len--) {
+		SLsmg_gotorc(y + len - 1, x);
+		SLsmg_write_nstring((char *)" ", max_len);
+	}
+	SLsmg_draw_box(y++, x + 1, 3, max_len - 2);
+
+	SLsmg_gotorc(y + 3, x);
+	SLsmg_write_nstring((char *)exit_msg, max_len);
+	SLsmg_refresh();
+
+	pthread_mutex_unlock(&ui__lock);
+
+	x += 2;
+	len = 0;
+	key = ui__getch(delay_secs);
+	while (key != K_TIMER && key != K_ENTER && key != K_ESC) {
+		pthread_mutex_lock(&ui__lock);
+
+		if (key == K_BKSPC) {
+			if (len == 0) {
+				pthread_mutex_unlock(&ui__lock);
+				goto next_key;
+			}
+			SLsmg_gotorc(y, x + --len);
+			SLsmg_write_char(' ');
+		} else {
+			buf[len] = key;
+			SLsmg_gotorc(y, x + len++);
+			SLsmg_write_char(key);
+		}
+		SLsmg_refresh();
+
+		pthread_mutex_unlock(&ui__lock);
+
+		/* XXX more graceful overflow handling needed */
+		if (len == sizeof(buf) - 1) {
+			ui_helpline__push("maximum size of symbol name reached!");
+			key = K_ENTER;
+			break;
+		}
+next_key:
+		key = ui__getch(delay_secs);
+	}
+
+	buf[len] = '\0';
+	strncpy(input, buf, len+1);
+	return key;
+}
+
+int ui__question_window(const char *title, const char *text,
+			const char *exit_msg, int delay_secs)
+{
+	int x, y;
+	int max_len = 0, nr_lines = 0;
+	const char *t;
+
+	t = text;
+	while (1) {
+		const char *sep = strchr(t, '\n');
+		int len;
+
+		if (sep == NULL)
+			sep = strchr(t, '\0');
+		len = sep - t;
+		if (max_len < len)
+			max_len = len;
+		++nr_lines;
+		if (*sep == '\0')
+			break;
+		t = sep + 1;
+	}
+
+	pthread_mutex_lock(&ui__lock);
+
+	max_len += 2;
+	nr_lines += 4;
+	y = SLtt_Screen_Rows / 2 - nr_lines / 2,
+	x = SLtt_Screen_Cols / 2 - max_len / 2;
+
+	SLsmg_set_color(0);
+	SLsmg_draw_box(y, x++, nr_lines, max_len);
+	if (title) {
+		SLsmg_gotorc(y, x + 1);
+		SLsmg_write_string((char *)title);
+	}
+	SLsmg_gotorc(++y, x);
+	nr_lines -= 2;
+	max_len -= 2;
+	SLsmg_write_wrapped_string((unsigned char *)text, y, x,
+				   nr_lines, max_len, 1);
+	SLsmg_gotorc(y + nr_lines - 2, x);
+	SLsmg_write_nstring((char *)" ", max_len);
+	SLsmg_gotorc(y + nr_lines - 1, x);
+	SLsmg_write_nstring((char *)exit_msg, max_len);
+	SLsmg_refresh();
+
+	pthread_mutex_unlock(&ui__lock);
+
+	return ui__getch(delay_secs);
+}
+
+int ui__help_window(const char *text)
+{
+	return ui__question_window("Help", text, "Press any key...", 0);
+}
+
+int ui__dialog_yesno(const char *msg)
+{
+	return ui__question_window(NULL, msg, "Enter: Yes, ESC: No", 0);
+}
+
+static int __ui__warning(const char *title, const char *format, va_list args)
+{
+	char *s;
+
+	if (vasprintf(&s, format, args) > 0) {
+		int key;
+
+		key = ui__question_window(title, s, "Press any key...", 0);
+		free(s);
+		return key;
+	}
+
+	fprintf(stderr, "%s\n", title);
+	vfprintf(stderr, format, args);
+	return K_ESC;
+}
+
+static int perf_tui__error(const char *format, va_list args)
+{
+	return __ui__warning("Error:", format, args);
+}
+
+static int perf_tui__warning(const char *format, va_list args)
+{
+	return __ui__warning("Warning:", format, args);
+}
+
+struct perf_error_ops perf_tui_eops = {
+	.error		= perf_tui__error,
+	.warning	= perf_tui__warning,
+};
diff --git a/tools/perf/ui/ui.h b/tools/perf/ui/ui.h
new file mode 100644
index 000000000..ab88383f8
--- /dev/null
+++ b/tools/perf/ui/ui.h
@@ -0,0 +1,29 @@
+#ifndef _PERF_UI_H_
+#define _PERF_UI_H_ 1
+
+#include <pthread.h>
+#include <stdbool.h>
+#include <linux/compiler.h>
+
+extern pthread_mutex_t ui__lock;
+extern void *perf_gtk_handle;
+
+extern int use_browser;
+
+void setup_browser(bool fallback_to_pager);
+void exit_browser(bool wait_for_ok);
+
+#ifdef HAVE_SLANG_SUPPORT
+int ui__init(void);
+void ui__exit(bool wait_for_ok);
+#else
+static inline int ui__init(void)
+{
+	return -1;
+}
+static inline void ui__exit(bool wait_for_ok __maybe_unused) {}
+#endif
+
+void ui__refresh_dimensions(bool force);
+
+#endif /* _PERF_UI_H_ */
diff --git a/tools/perf/ui/util.c b/tools/perf/ui/util.c
new file mode 100644
index 000000000..e3e0a963d
--- /dev/null
+++ b/tools/perf/ui/util.c
@@ -0,0 +1,84 @@
+#include "util.h"
+#include "../debug.h"
+
+
+/*
+ * Default error logging functions
+ */
+static int perf_stdio__error(const char *format, va_list args)
+{
+	fprintf(stderr, "Error:\n");
+	vfprintf(stderr, format, args);
+	return 0;
+}
+
+static int perf_stdio__warning(const char *format, va_list args)
+{
+	fprintf(stderr, "Warning:\n");
+	vfprintf(stderr, format, args);
+	return 0;
+}
+
+static struct perf_error_ops default_eops =
+{
+	.error		= perf_stdio__error,
+	.warning	= perf_stdio__warning,
+};
+
+static struct perf_error_ops *perf_eops = &default_eops;
+
+
+int ui__error(const char *format, ...)
+{
+	int ret;
+	va_list args;
+
+	va_start(args, format);
+	ret = perf_eops->error(format, args);
+	va_end(args);
+
+	return ret;
+}
+
+int ui__warning(const char *format, ...)
+{
+	int ret;
+	va_list args;
+
+	va_start(args, format);
+	ret = perf_eops->warning(format, args);
+	va_end(args);
+
+	return ret;
+}
+
+/**
+ * perf_error__register - Register error logging functions
+ * @eops: The pointer to error logging function struct
+ *
+ * Register UI-specific error logging functions. Before calling this,
+ * other logging functions should be unregistered, if any.
+ */
+int perf_error__register(struct perf_error_ops *eops)
+{
+	if (perf_eops != &default_eops)
+		return -1;
+
+	perf_eops = eops;
+	return 0;
+}
+
+/**
+ * perf_error__unregister - Unregister error logging functions
+ * @eops: The pointer to error logging function struct
+ *
+ * Unregister already registered error logging functions.
+ */
+int perf_error__unregister(struct perf_error_ops *eops)
+{
+	if (perf_eops != eops)
+		return -1;
+
+	perf_eops = &default_eops;
+	return 0;
+}
diff --git a/tools/perf/ui/util.h b/tools/perf/ui/util.h
new file mode 100644
index 000000000..361f08c52
--- /dev/null
+++ b/tools/perf/ui/util.h
@@ -0,0 +1,21 @@
+#ifndef _PERF_UI_UTIL_H_
+#define _PERF_UI_UTIL_H_ 1
+
+#include <stdarg.h>
+
+int ui__getch(int delay_secs);
+int ui__popup_menu(int argc, char * const argv[]);
+int ui__help_window(const char *text);
+int ui__dialog_yesno(const char *msg);
+int ui__question_window(const char *title, const char *text,
+			const char *exit_msg, int delay_secs);
+
+struct perf_error_ops {
+	int (*error)(const char *format, va_list args);
+	int (*warning)(const char *format, va_list args);
+};
+
+int perf_error__register(struct perf_error_ops *eops);
+int perf_error__unregister(struct perf_error_ops *eops);
+
+#endif /* _PERF_UI_UTIL_H_ */
diff --git a/tools/perf/util/Build b/tools/perf/util/Build
new file mode 100644
index 000000000..797490a40
--- /dev/null
+++ b/tools/perf/util/Build
@@ -0,0 +1,145 @@
+libperf-y += abspath.o
+libperf-y += alias.o
+libperf-y += annotate.o
+libperf-y += build-id.o
+libperf-y += config.o
+libperf-y += ctype.o
+libperf-y += db-export.o
+libperf-y += environment.o
+libperf-y += event.o
+libperf-y += evlist.o
+libperf-y += evsel.o
+libperf-y += exec_cmd.o
+libperf-y += find_next_bit.o
+libperf-y += help.o
+libperf-y += kallsyms.o
+libperf-y += levenshtein.o
+libperf-y += parse-options.o
+libperf-y += parse-events.o
+libperf-y += path.o
+libperf-y += rbtree.o
+libperf-y += bitmap.o
+libperf-y += hweight.o
+libperf-y += run-command.o
+libperf-y += quote.o
+libperf-y += strbuf.o
+libperf-y += string.o
+libperf-y += strlist.o
+libperf-y += strfilter.o
+libperf-y += top.o
+libperf-y += usage.o
+libperf-y += wrapper.o
+libperf-y += sigchain.o
+libperf-y += dso.o
+libperf-y += symbol.o
+libperf-y += color.o
+libperf-y += pager.o
+libperf-y += header.o
+libperf-y += callchain.o
+libperf-y += values.o
+libperf-y += debug.o
+libperf-y += machine.o
+libperf-y += map.o
+libperf-y += pstack.o
+libperf-y += session.o
+libperf-y += ordered-events.o
+libperf-y += comm.o
+libperf-y += thread.o
+libperf-y += thread_map.o
+libperf-y += trace-event-parse.o
+libperf-y += parse-events-flex.o
+libperf-y += parse-events-bison.o
+libperf-y += pmu.o
+libperf-y += pmu-flex.o
+libperf-y += pmu-bison.o
+libperf-y += trace-event-read.o
+libperf-y += trace-event-info.o
+libperf-y += trace-event-scripting.o
+libperf-y += trace-event.o
+libperf-y += svghelper.o
+libperf-y += sort.o
+libperf-y += hist.o
+libperf-y += util.o
+libperf-y += xyarray.o
+libperf-y += cpumap.o
+libperf-y += cgroup.o
+libperf-y += target.o
+libperf-y += rblist.o
+libperf-y += intlist.o
+libperf-y += vdso.o
+libperf-y += stat.o
+libperf-y += record.o
+libperf-y += srcline.o
+libperf-y += data.o
+libperf-$(CONFIG_X86) += tsc.o
+libperf-y += cloexec.o
+libperf-y += thread-stack.o
+
+libperf-$(CONFIG_LIBELF) += symbol-elf.o
+libperf-$(CONFIG_LIBELF) += probe-event.o
+
+ifndef CONFIG_LIBELF
+libperf-y += symbol-minimal.o
+endif
+
+libperf-$(CONFIG_DWARF) += probe-finder.o
+libperf-$(CONFIG_DWARF) += dwarf-aux.o
+
+libperf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o
+libperf-$(CONFIG_LIBUNWIND)          += unwind-libunwind.o
+
+libperf-$(CONFIG_LIBBABELTRACE) += data-convert-bt.o
+
+libperf-y += scripting-engines/
+
+libperf-$(CONFIG_PERF_REGS) += perf_regs.o
+libperf-$(CONFIG_ZLIB) += zlib.o
+libperf-$(CONFIG_LZMA) += lzma.o
+
+CFLAGS_config.o   += -DETC_PERFCONFIG="BUILD_STR($(ETC_PERFCONFIG_SQ))"
+CFLAGS_exec_cmd.o += -DPERF_EXEC_PATH="BUILD_STR($(perfexecdir_SQ))" -DPREFIX="BUILD_STR($(prefix_SQ))"
+
+$(OUTPUT)util/parse-events-flex.c: util/parse-events.l $(OUTPUT)util/parse-events-bison.c
+	$(call rule_mkdir)
+	@$(call echo-cmd,flex)$(FLEX) -o $@ --header-file=$(OUTPUT)util/parse-events-flex.h $(PARSER_DEBUG_FLEX) util/parse-events.l
+
+$(OUTPUT)util/parse-events-bison.c: util/parse-events.y
+	$(call rule_mkdir)
+	@$(call echo-cmd,bison)$(BISON) -v util/parse-events.y -d $(PARSER_DEBUG_BISON) -o $@ -p parse_events_
+
+$(OUTPUT)util/pmu-flex.c: util/pmu.l $(OUTPUT)util/pmu-bison.c
+	$(call rule_mkdir)
+	@$(call echo-cmd,flex)$(FLEX) -o $@ --header-file=$(OUTPUT)util/pmu-flex.h util/pmu.l
+
+$(OUTPUT)util/pmu-bison.c: util/pmu.y
+	$(call rule_mkdir)
+	@$(call echo-cmd,bison)$(BISON) -v util/pmu.y -d -o $@ -p perf_pmu_
+
+CFLAGS_parse-events-flex.o  += -w
+CFLAGS_pmu-flex.o           += -w
+CFLAGS_parse-events-bison.o += -DYYENABLE_NLS=0 -DYYLTYPE_IS_TRIVIAL=0 -w
+CFLAGS_pmu-bison.o          += -DYYENABLE_NLS=0 -DYYLTYPE_IS_TRIVIAL=0 -w
+
+$(OUTPUT)util/parse-events.o: $(OUTPUT)util/parse-events-flex.c $(OUTPUT)util/parse-events-bison.c
+$(OUTPUT)util/pmu.o: $(OUTPUT)util/pmu-flex.c $(OUTPUT)util/pmu-bison.c
+
+CFLAGS_find_next_bit.o += -Wno-unused-parameter -DETC_PERFCONFIG="BUILD_STR($(ETC_PERFCONFIG_SQ))"
+CFLAGS_rbtree.o        += -Wno-unused-parameter -DETC_PERFCONFIG="BUILD_STR($(ETC_PERFCONFIG_SQ))"
+CFLAGS_hweight.o       += -Wno-unused-parameter -DETC_PERFCONFIG="BUILD_STR($(ETC_PERFCONFIG_SQ))"
+CFLAGS_parse-events.o  += -Wno-redundant-decls
+
+$(OUTPUT)util/kallsyms.o: ../lib/symbol/kallsyms.c FORCE
+	$(call rule_mkdir)
+	$(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)util/find_next_bit.o: ../lib/util/find_next_bit.c FORCE
+	$(call rule_mkdir)
+	$(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)util/rbtree.o: ../../lib/rbtree.c FORCE
+	$(call rule_mkdir)
+	$(call if_changed_dep,cc_o_c)
+
+$(OUTPUT)util/hweight.o: ../../lib/hweight.c FORCE
+	$(call rule_mkdir)
+	$(call if_changed_dep,cc_o_c)
diff --git a/tools/perf/util/PERF-VERSION-GEN b/tools/perf/util/PERF-VERSION-GEN
new file mode 100755
index 000000000..39f175075
--- /dev/null
+++ b/tools/perf/util/PERF-VERSION-GEN
@@ -0,0 +1,50 @@
+#!/bin/sh
+
+if [ $# -eq 1 ]  ; then
+	OUTPUT=$1
+fi
+
+GVF=${OUTPUT}PERF-VERSION-FILE
+
+LF='
+'
+
+#
+# First check if there is a .git to get the version from git describe
+# otherwise try to get the version from the kernel Makefile
+#
+CID=
+TAG=
+if test -d ../../.git -o -f ../../.git
+then
+	TAG=$(git describe --abbrev=0 --match "v[0-9].[0-9]*" 2>/dev/null )
+	CID=$(git log -1 --abbrev=4 --pretty=format:"%h" 2>/dev/null) && CID="-g$CID"
+elif test -f ../../PERF-VERSION-FILE
+then
+	TAG=$(cut -d' ' -f3 ../../PERF-VERSION-FILE | sed -e 's/\"//g')
+fi
+if test -z "$TAG"
+then
+	TAG=$(MAKEFLAGS= make -sC ../.. kernelversion)
+fi
+VN="$TAG$CID"
+if test -n "$CID"
+then
+	# format version string, strip trailing zero of sublevel:
+	VN=$(echo "$VN" | sed -e 's/-/./g;s/\([0-9]*[.][0-9]*\)[.]0/\1/')
+fi
+
+VN=$(expr "$VN" : v*'\(.*\)')
+
+if test -r $GVF
+then
+	VC=$(sed -e 's/^#define PERF_VERSION "\(.*\)"/\1/' <$GVF)
+else
+	VC=unset
+fi
+test "$VN" = "$VC" || {
+	echo >&2 "  PERF_VERSION = $VN"
+	echo "#define PERF_VERSION \"$VN\"" >$GVF
+}
+
+
diff --git a/tools/perf/util/abspath.c b/tools/perf/util/abspath.c
new file mode 100644
index 000000000..0e76affe9
--- /dev/null
+++ b/tools/perf/util/abspath.c
@@ -0,0 +1,37 @@
+#include "cache.h"
+
+static const char *get_pwd_cwd(void)
+{
+	static char cwd[PATH_MAX + 1];
+	char *pwd;
+	struct stat cwd_stat, pwd_stat;
+	if (getcwd(cwd, PATH_MAX) == NULL)
+		return NULL;
+	pwd = getenv("PWD");
+	if (pwd && strcmp(pwd, cwd)) {
+		stat(cwd, &cwd_stat);
+		if (!stat(pwd, &pwd_stat) &&
+		    pwd_stat.st_dev == cwd_stat.st_dev &&
+		    pwd_stat.st_ino == cwd_stat.st_ino) {
+			strlcpy(cwd, pwd, PATH_MAX);
+		}
+	}
+	return cwd;
+}
+
+const char *make_nonrelative_path(const char *path)
+{
+	static char buf[PATH_MAX + 1];
+
+	if (is_absolute_path(path)) {
+		if (strlcpy(buf, path, PATH_MAX) >= PATH_MAX)
+			die("Too long path: %.*s", 60, path);
+	} else {
+		const char *cwd = get_pwd_cwd();
+		if (!cwd)
+			die("Cannot determine the current working directory");
+		if (snprintf(buf, PATH_MAX, "%s/%s", cwd, path) >= PATH_MAX)
+			die("Too long path: %.*s", 60, path);
+	}
+	return buf;
+}
diff --git a/tools/perf/util/alias.c b/tools/perf/util/alias.c
new file mode 100644
index 000000000..c0b43ee40
--- /dev/null
+++ b/tools/perf/util/alias.c
@@ -0,0 +1,76 @@
+#include "cache.h"
+
+static const char *alias_key;
+static char *alias_val;
+
+static int alias_lookup_cb(const char *k, const char *v,
+			   void *cb __maybe_unused)
+{
+	if (!prefixcmp(k, "alias.") && !strcmp(k+6, alias_key)) {
+		if (!v)
+			return config_error_nonbool(k);
+		alias_val = strdup(v);
+		return 0;
+	}
+	return 0;
+}
+
+char *alias_lookup(const char *alias)
+{
+	alias_key = alias;
+	alias_val = NULL;
+	perf_config(alias_lookup_cb, NULL);
+	return alias_val;
+}
+
+int split_cmdline(char *cmdline, const char ***argv)
+{
+	int src, dst, count = 0, size = 16;
+	char quoted = 0;
+
+	*argv = malloc(sizeof(char*) * size);
+
+	/* split alias_string */
+	(*argv)[count++] = cmdline;
+	for (src = dst = 0; cmdline[src];) {
+		char c = cmdline[src];
+		if (!quoted && isspace(c)) {
+			cmdline[dst++] = 0;
+			while (cmdline[++src]
+					&& isspace(cmdline[src]))
+				; /* skip */
+			if (count >= size) {
+				size += 16;
+				*argv = realloc(*argv, sizeof(char*) * size);
+			}
+			(*argv)[count++] = cmdline + dst;
+		} else if (!quoted && (c == '\'' || c == '"')) {
+			quoted = c;
+			src++;
+		} else if (c == quoted) {
+			quoted = 0;
+			src++;
+		} else {
+			if (c == '\\' && quoted != '\'') {
+				src++;
+				c = cmdline[src];
+				if (!c) {
+					zfree(argv);
+					return error("cmdline ends with \\");
+				}
+			}
+			cmdline[dst++] = c;
+			src++;
+		}
+	}
+
+	cmdline[dst] = 0;
+
+	if (quoted) {
+		zfree(argv);
+		return error("unclosed quote");
+	}
+
+	return count;
+}
+
diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
new file mode 100644
index 000000000..7f5bdfc9b
--- /dev/null
+++ b/tools/perf/util/annotate.c
@@ -0,0 +1,1479 @@
+/*
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Parts came from builtin-annotate.c, see those files for further
+ * copyright notes.
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+
+#include "util.h"
+#include "ui/ui.h"
+#include "sort.h"
+#include "build-id.h"
+#include "color.h"
+#include "cache.h"
+#include "symbol.h"
+#include "debug.h"
+#include "annotate.h"
+#include "evsel.h"
+#include <regex.h>
+#include <pthread.h>
+#include <linux/bitops.h>
+
+const char 	*disassembler_style;
+const char	*objdump_path;
+static regex_t	 file_lineno;
+
+static struct ins *ins__find(const char *name);
+static int disasm_line__parse(char *line, char **namep, char **rawp);
+
+static void ins__delete(struct ins_operands *ops)
+{
+	if (ops == NULL)
+		return;
+	zfree(&ops->source.raw);
+	zfree(&ops->source.name);
+	zfree(&ops->target.raw);
+	zfree(&ops->target.name);
+}
+
+static int ins__raw_scnprintf(struct ins *ins, char *bf, size_t size,
+			      struct ins_operands *ops)
+{
+	return scnprintf(bf, size, "%-6.6s %s", ins->name, ops->raw);
+}
+
+int ins__scnprintf(struct ins *ins, char *bf, size_t size,
+		  struct ins_operands *ops)
+{
+	if (ins->ops->scnprintf)
+		return ins->ops->scnprintf(ins, bf, size, ops);
+
+	return ins__raw_scnprintf(ins, bf, size, ops);
+}
+
+static int call__parse(struct ins_operands *ops)
+{
+	char *endptr, *tok, *name;
+
+	ops->target.addr = strtoull(ops->raw, &endptr, 16);
+
+	name = strchr(endptr, '<');
+	if (name == NULL)
+		goto indirect_call;
+
+	name++;
+
+	tok = strchr(name, '>');
+	if (tok == NULL)
+		return -1;
+
+	*tok = '\0';
+	ops->target.name = strdup(name);
+	*tok = '>';
+
+	return ops->target.name == NULL ? -1 : 0;
+
+indirect_call:
+	tok = strchr(endptr, '(');
+	if (tok != NULL) {
+		ops->target.addr = 0;
+		return 0;
+	}
+
+	tok = strchr(endptr, '*');
+	if (tok == NULL)
+		return -1;
+
+	ops->target.addr = strtoull(tok + 1, NULL, 16);
+	return 0;
+}
+
+static int call__scnprintf(struct ins *ins, char *bf, size_t size,
+			   struct ins_operands *ops)
+{
+	if (ops->target.name)
+		return scnprintf(bf, size, "%-6.6s %s", ins->name, ops->target.name);
+
+	if (ops->target.addr == 0)
+		return ins__raw_scnprintf(ins, bf, size, ops);
+
+	return scnprintf(bf, size, "%-6.6s *%" PRIx64, ins->name, ops->target.addr);
+}
+
+static struct ins_ops call_ops = {
+	.parse	   = call__parse,
+	.scnprintf = call__scnprintf,
+};
+
+bool ins__is_call(const struct ins *ins)
+{
+	return ins->ops == &call_ops;
+}
+
+static int jump__parse(struct ins_operands *ops)
+{
+	const char *s = strchr(ops->raw, '+');
+
+	ops->target.addr = strtoull(ops->raw, NULL, 16);
+
+	if (s++ != NULL)
+		ops->target.offset = strtoull(s, NULL, 16);
+	else
+		ops->target.offset = UINT64_MAX;
+
+	return 0;
+}
+
+static int jump__scnprintf(struct ins *ins, char *bf, size_t size,
+			   struct ins_operands *ops)
+{
+	return scnprintf(bf, size, "%-6.6s %" PRIx64, ins->name, ops->target.offset);
+}
+
+static struct ins_ops jump_ops = {
+	.parse	   = jump__parse,
+	.scnprintf = jump__scnprintf,
+};
+
+bool ins__is_jump(const struct ins *ins)
+{
+	return ins->ops == &jump_ops;
+}
+
+static int comment__symbol(char *raw, char *comment, u64 *addrp, char **namep)
+{
+	char *endptr, *name, *t;
+
+	if (strstr(raw, "(%rip)") == NULL)
+		return 0;
+
+	*addrp = strtoull(comment, &endptr, 16);
+	name = strchr(endptr, '<');
+	if (name == NULL)
+		return -1;
+
+	name++;
+
+	t = strchr(name, '>');
+	if (t == NULL)
+		return 0;
+
+	*t = '\0';
+	*namep = strdup(name);
+	*t = '>';
+
+	return 0;
+}
+
+static int lock__parse(struct ins_operands *ops)
+{
+	char *name;
+
+	ops->locked.ops = zalloc(sizeof(*ops->locked.ops));
+	if (ops->locked.ops == NULL)
+		return 0;
+
+	if (disasm_line__parse(ops->raw, &name, &ops->locked.ops->raw) < 0)
+		goto out_free_ops;
+
+	ops->locked.ins = ins__find(name);
+	free(name);
+
+	if (ops->locked.ins == NULL)
+		goto out_free_ops;
+
+	if (!ops->locked.ins->ops)
+		return 0;
+
+	if (ops->locked.ins->ops->parse &&
+	    ops->locked.ins->ops->parse(ops->locked.ops) < 0)
+		goto out_free_ops;
+
+	return 0;
+
+out_free_ops:
+	zfree(&ops->locked.ops);
+	return 0;
+}
+
+static int lock__scnprintf(struct ins *ins, char *bf, size_t size,
+			   struct ins_operands *ops)
+{
+	int printed;
+
+	if (ops->locked.ins == NULL)
+		return ins__raw_scnprintf(ins, bf, size, ops);
+
+	printed = scnprintf(bf, size, "%-6.6s ", ins->name);
+	return printed + ins__scnprintf(ops->locked.ins, bf + printed,
+					size - printed, ops->locked.ops);
+}
+
+static void lock__delete(struct ins_operands *ops)
+{
+	struct ins *ins = ops->locked.ins;
+
+	if (ins && ins->ops->free)
+		ins->ops->free(ops->locked.ops);
+	else
+		ins__delete(ops->locked.ops);
+
+	zfree(&ops->locked.ops);
+	zfree(&ops->target.raw);
+	zfree(&ops->target.name);
+}
+
+static struct ins_ops lock_ops = {
+	.free	   = lock__delete,
+	.parse	   = lock__parse,
+	.scnprintf = lock__scnprintf,
+};
+
+static int mov__parse(struct ins_operands *ops)
+{
+	char *s = strchr(ops->raw, ','), *target, *comment, prev;
+
+	if (s == NULL)
+		return -1;
+
+	*s = '\0';
+	ops->source.raw = strdup(ops->raw);
+	*s = ',';
+
+	if (ops->source.raw == NULL)
+		return -1;
+
+	target = ++s;
+	comment = strchr(s, '#');
+
+	if (comment != NULL)
+		s = comment - 1;
+	else
+		s = strchr(s, '\0') - 1;
+
+	while (s > target && isspace(s[0]))
+		--s;
+	s++;
+	prev = *s;
+	*s = '\0';
+
+	ops->target.raw = strdup(target);
+	*s = prev;
+
+	if (ops->target.raw == NULL)
+		goto out_free_source;
+
+	if (comment == NULL)
+		return 0;
+
+	while (comment[0] != '\0' && isspace(comment[0]))
+		++comment;
+
+	comment__symbol(ops->source.raw, comment, &ops->source.addr, &ops->source.name);
+	comment__symbol(ops->target.raw, comment, &ops->target.addr, &ops->target.name);
+
+	return 0;
+
+out_free_source:
+	zfree(&ops->source.raw);
+	return -1;
+}
+
+static int mov__scnprintf(struct ins *ins, char *bf, size_t size,
+			   struct ins_operands *ops)
+{
+	return scnprintf(bf, size, "%-6.6s %s,%s", ins->name,
+			 ops->source.name ?: ops->source.raw,
+			 ops->target.name ?: ops->target.raw);
+}
+
+static struct ins_ops mov_ops = {
+	.parse	   = mov__parse,
+	.scnprintf = mov__scnprintf,
+};
+
+static int dec__parse(struct ins_operands *ops)
+{
+	char *target, *comment, *s, prev;
+
+	target = s = ops->raw;
+
+	while (s[0] != '\0' && !isspace(s[0]))
+		++s;
+	prev = *s;
+	*s = '\0';
+
+	ops->target.raw = strdup(target);
+	*s = prev;
+
+	if (ops->target.raw == NULL)
+		return -1;
+
+	comment = strchr(s, '#');
+	if (comment == NULL)
+		return 0;
+
+	while (comment[0] != '\0' && isspace(comment[0]))
+		++comment;
+
+	comment__symbol(ops->target.raw, comment, &ops->target.addr, &ops->target.name);
+
+	return 0;
+}
+
+static int dec__scnprintf(struct ins *ins, char *bf, size_t size,
+			   struct ins_operands *ops)
+{
+	return scnprintf(bf, size, "%-6.6s %s", ins->name,
+			 ops->target.name ?: ops->target.raw);
+}
+
+static struct ins_ops dec_ops = {
+	.parse	   = dec__parse,
+	.scnprintf = dec__scnprintf,
+};
+
+static int nop__scnprintf(struct ins *ins __maybe_unused, char *bf, size_t size,
+			  struct ins_operands *ops __maybe_unused)
+{
+	return scnprintf(bf, size, "%-6.6s", "nop");
+}
+
+static struct ins_ops nop_ops = {
+	.scnprintf = nop__scnprintf,
+};
+
+/*
+ * Must be sorted by name!
+ */
+static struct ins instructions[] = {
+	{ .name = "add",   .ops  = &mov_ops, },
+	{ .name = "addl",  .ops  = &mov_ops, },
+	{ .name = "addq",  .ops  = &mov_ops, },
+	{ .name = "addw",  .ops  = &mov_ops, },
+	{ .name = "and",   .ops  = &mov_ops, },
+	{ .name = "bts",   .ops  = &mov_ops, },
+	{ .name = "call",  .ops  = &call_ops, },
+	{ .name = "callq", .ops  = &call_ops, },
+	{ .name = "cmp",   .ops  = &mov_ops, },
+	{ .name = "cmpb",  .ops  = &mov_ops, },
+	{ .name = "cmpl",  .ops  = &mov_ops, },
+	{ .name = "cmpq",  .ops  = &mov_ops, },
+	{ .name = "cmpw",  .ops  = &mov_ops, },
+	{ .name = "cmpxch", .ops  = &mov_ops, },
+	{ .name = "dec",   .ops  = &dec_ops, },
+	{ .name = "decl",  .ops  = &dec_ops, },
+	{ .name = "imul",  .ops  = &mov_ops, },
+	{ .name = "inc",   .ops  = &dec_ops, },
+	{ .name = "incl",  .ops  = &dec_ops, },
+	{ .name = "ja",	   .ops  = &jump_ops, },
+	{ .name = "jae",   .ops  = &jump_ops, },
+	{ .name = "jb",	   .ops  = &jump_ops, },
+	{ .name = "jbe",   .ops  = &jump_ops, },
+	{ .name = "jc",	   .ops  = &jump_ops, },
+	{ .name = "jcxz",  .ops  = &jump_ops, },
+	{ .name = "je",	   .ops  = &jump_ops, },
+	{ .name = "jecxz", .ops  = &jump_ops, },
+	{ .name = "jg",	   .ops  = &jump_ops, },
+	{ .name = "jge",   .ops  = &jump_ops, },
+	{ .name = "jl",    .ops  = &jump_ops, },
+	{ .name = "jle",   .ops  = &jump_ops, },
+	{ .name = "jmp",   .ops  = &jump_ops, },
+	{ .name = "jmpq",  .ops  = &jump_ops, },
+	{ .name = "jna",   .ops  = &jump_ops, },
+	{ .name = "jnae",  .ops  = &jump_ops, },
+	{ .name = "jnb",   .ops  = &jump_ops, },
+	{ .name = "jnbe",  .ops  = &jump_ops, },
+	{ .name = "jnc",   .ops  = &jump_ops, },
+	{ .name = "jne",   .ops  = &jump_ops, },
+	{ .name = "jng",   .ops  = &jump_ops, },
+	{ .name = "jnge",  .ops  = &jump_ops, },
+	{ .name = "jnl",   .ops  = &jump_ops, },
+	{ .name = "jnle",  .ops  = &jump_ops, },
+	{ .name = "jno",   .ops  = &jump_ops, },
+	{ .name = "jnp",   .ops  = &jump_ops, },
+	{ .name = "jns",   .ops  = &jump_ops, },
+	{ .name = "jnz",   .ops  = &jump_ops, },
+	{ .name = "jo",	   .ops  = &jump_ops, },
+	{ .name = "jp",	   .ops  = &jump_ops, },
+	{ .name = "jpe",   .ops  = &jump_ops, },
+	{ .name = "jpo",   .ops  = &jump_ops, },
+	{ .name = "jrcxz", .ops  = &jump_ops, },
+	{ .name = "js",	   .ops  = &jump_ops, },
+	{ .name = "jz",	   .ops  = &jump_ops, },
+	{ .name = "lea",   .ops  = &mov_ops, },
+	{ .name = "lock",  .ops  = &lock_ops, },
+	{ .name = "mov",   .ops  = &mov_ops, },
+	{ .name = "movb",  .ops  = &mov_ops, },
+	{ .name = "movdqa",.ops  = &mov_ops, },
+	{ .name = "movl",  .ops  = &mov_ops, },
+	{ .name = "movq",  .ops  = &mov_ops, },
+	{ .name = "movslq", .ops  = &mov_ops, },
+	{ .name = "movzbl", .ops  = &mov_ops, },
+	{ .name = "movzwl", .ops  = &mov_ops, },
+	{ .name = "nop",   .ops  = &nop_ops, },
+	{ .name = "nopl",  .ops  = &nop_ops, },
+	{ .name = "nopw",  .ops  = &nop_ops, },
+	{ .name = "or",    .ops  = &mov_ops, },
+	{ .name = "orl",   .ops  = &mov_ops, },
+	{ .name = "test",  .ops  = &mov_ops, },
+	{ .name = "testb", .ops  = &mov_ops, },
+	{ .name = "testl", .ops  = &mov_ops, },
+	{ .name = "xadd",  .ops  = &mov_ops, },
+	{ .name = "xbeginl", .ops  = &jump_ops, },
+	{ .name = "xbeginq", .ops  = &jump_ops, },
+};
+
+static int ins__cmp(const void *name, const void *insp)
+{
+	const struct ins *ins = insp;
+
+	return strcmp(name, ins->name);
+}
+
+static struct ins *ins__find(const char *name)
+{
+	const int nmemb = ARRAY_SIZE(instructions);
+
+	return bsearch(name, instructions, nmemb, sizeof(struct ins), ins__cmp);
+}
+
+int symbol__annotate_init(struct map *map __maybe_unused, struct symbol *sym)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	pthread_mutex_init(&notes->lock, NULL);
+	return 0;
+}
+
+int symbol__alloc_hist(struct symbol *sym)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	const size_t size = symbol__size(sym);
+	size_t sizeof_sym_hist;
+
+	/* Check for overflow when calculating sizeof_sym_hist */
+	if (size > (SIZE_MAX - sizeof(struct sym_hist)) / sizeof(u64))
+		return -1;
+
+	sizeof_sym_hist = (sizeof(struct sym_hist) + size * sizeof(u64));
+
+	/* Check for overflow in zalloc argument */
+	if (sizeof_sym_hist > (SIZE_MAX - sizeof(*notes->src))
+				/ symbol_conf.nr_events)
+		return -1;
+
+	notes->src = zalloc(sizeof(*notes->src) + symbol_conf.nr_events * sizeof_sym_hist);
+	if (notes->src == NULL)
+		return -1;
+	notes->src->sizeof_sym_hist = sizeof_sym_hist;
+	notes->src->nr_histograms   = symbol_conf.nr_events;
+	INIT_LIST_HEAD(&notes->src->source);
+	return 0;
+}
+
+void symbol__annotate_zero_histograms(struct symbol *sym)
+{
+	struct annotation *notes = symbol__annotation(sym);
+
+	pthread_mutex_lock(&notes->lock);
+	if (notes->src != NULL)
+		memset(notes->src->histograms, 0,
+		       notes->src->nr_histograms * notes->src->sizeof_sym_hist);
+	pthread_mutex_unlock(&notes->lock);
+}
+
+static int __symbol__inc_addr_samples(struct symbol *sym, struct map *map,
+				      struct annotation *notes, int evidx, u64 addr)
+{
+	unsigned offset;
+	struct sym_hist *h;
+
+	pr_debug3("%s: addr=%#" PRIx64 "\n", __func__, map->unmap_ip(map, addr));
+
+	if (addr < sym->start || addr >= sym->end)
+		return -ERANGE;
+
+	offset = addr - sym->start;
+	h = annotation__histogram(notes, evidx);
+	h->sum++;
+	h->addr[offset]++;
+
+	pr_debug3("%#" PRIx64 " %s: period++ [addr: %#" PRIx64 ", %#" PRIx64
+		  ", evidx=%d] => %" PRIu64 "\n", sym->start, sym->name,
+		  addr, addr - sym->start, evidx, h->addr[offset]);
+	return 0;
+}
+
+static int symbol__inc_addr_samples(struct symbol *sym, struct map *map,
+				    int evidx, u64 addr)
+{
+	struct annotation *notes;
+
+	if (sym == NULL)
+		return 0;
+
+	notes = symbol__annotation(sym);
+	if (notes->src == NULL) {
+		if (symbol__alloc_hist(sym) < 0)
+			return -ENOMEM;
+	}
+
+	return __symbol__inc_addr_samples(sym, map, notes, evidx, addr);
+}
+
+int addr_map_symbol__inc_samples(struct addr_map_symbol *ams, int evidx)
+{
+	return symbol__inc_addr_samples(ams->sym, ams->map, evidx, ams->al_addr);
+}
+
+int hist_entry__inc_addr_samples(struct hist_entry *he, int evidx, u64 ip)
+{
+	return symbol__inc_addr_samples(he->ms.sym, he->ms.map, evidx, ip);
+}
+
+static void disasm_line__init_ins(struct disasm_line *dl)
+{
+	dl->ins = ins__find(dl->name);
+
+	if (dl->ins == NULL)
+		return;
+
+	if (!dl->ins->ops)
+		return;
+
+	if (dl->ins->ops->parse && dl->ins->ops->parse(&dl->ops) < 0)
+		dl->ins = NULL;
+}
+
+static int disasm_line__parse(char *line, char **namep, char **rawp)
+{
+	char *name = line, tmp;
+
+	while (isspace(name[0]))
+		++name;
+
+	if (name[0] == '\0')
+		return -1;
+
+	*rawp = name + 1;
+
+	while ((*rawp)[0] != '\0' && !isspace((*rawp)[0]))
+		++*rawp;
+
+	tmp = (*rawp)[0];
+	(*rawp)[0] = '\0';
+	*namep = strdup(name);
+
+	if (*namep == NULL)
+		goto out_free_name;
+
+	(*rawp)[0] = tmp;
+
+	if ((*rawp)[0] != '\0') {
+		(*rawp)++;
+		while (isspace((*rawp)[0]))
+			++(*rawp);
+	}
+
+	return 0;
+
+out_free_name:
+	zfree(namep);
+	return -1;
+}
+
+static struct disasm_line *disasm_line__new(s64 offset, char *line,
+					size_t privsize, int line_nr)
+{
+	struct disasm_line *dl = zalloc(sizeof(*dl) + privsize);
+
+	if (dl != NULL) {
+		dl->offset = offset;
+		dl->line = strdup(line);
+		dl->line_nr = line_nr;
+		if (dl->line == NULL)
+			goto out_delete;
+
+		if (offset != -1) {
+			if (disasm_line__parse(dl->line, &dl->name, &dl->ops.raw) < 0)
+				goto out_free_line;
+
+			disasm_line__init_ins(dl);
+		}
+	}
+
+	return dl;
+
+out_free_line:
+	zfree(&dl->line);
+out_delete:
+	free(dl);
+	return NULL;
+}
+
+void disasm_line__free(struct disasm_line *dl)
+{
+	zfree(&dl->line);
+	zfree(&dl->name);
+	if (dl->ins && dl->ins->ops->free)
+		dl->ins->ops->free(&dl->ops);
+	else
+		ins__delete(&dl->ops);
+	free(dl);
+}
+
+int disasm_line__scnprintf(struct disasm_line *dl, char *bf, size_t size, bool raw)
+{
+	if (raw || !dl->ins)
+		return scnprintf(bf, size, "%-6.6s %s", dl->name, dl->ops.raw);
+
+	return ins__scnprintf(dl->ins, bf, size, &dl->ops);
+}
+
+static void disasm__add(struct list_head *head, struct disasm_line *line)
+{
+	list_add_tail(&line->node, head);
+}
+
+struct disasm_line *disasm__get_next_ip_line(struct list_head *head, struct disasm_line *pos)
+{
+	list_for_each_entry_continue(pos, head, node)
+		if (pos->offset >= 0)
+			return pos;
+
+	return NULL;
+}
+
+double disasm__calc_percent(struct annotation *notes, int evidx, s64 offset,
+			    s64 end, const char **path)
+{
+	struct source_line *src_line = notes->src->lines;
+	double percent = 0.0;
+
+	if (src_line) {
+		size_t sizeof_src_line = sizeof(*src_line) +
+				sizeof(src_line->p) * (src_line->nr_pcnt - 1);
+
+		while (offset < end) {
+			src_line = (void *)notes->src->lines +
+					(sizeof_src_line * offset);
+
+			if (*path == NULL)
+				*path = src_line->path;
+
+			percent += src_line->p[evidx].percent;
+			offset++;
+		}
+	} else {
+		struct sym_hist *h = annotation__histogram(notes, evidx);
+		unsigned int hits = 0;
+
+		while (offset < end)
+			hits += h->addr[offset++];
+
+		if (h->sum)
+			percent = 100.0 * hits / h->sum;
+	}
+
+	return percent;
+}
+
+static int disasm_line__print(struct disasm_line *dl, struct symbol *sym, u64 start,
+		      struct perf_evsel *evsel, u64 len, int min_pcnt, int printed,
+		      int max_lines, struct disasm_line *queue)
+{
+	static const char *prev_line;
+	static const char *prev_color;
+
+	if (dl->offset != -1) {
+		const char *path = NULL;
+		double percent, max_percent = 0.0;
+		double *ppercents = &percent;
+		int i, nr_percent = 1;
+		const char *color;
+		struct annotation *notes = symbol__annotation(sym);
+		s64 offset = dl->offset;
+		const u64 addr = start + offset;
+		struct disasm_line *next;
+
+		next = disasm__get_next_ip_line(&notes->src->source, dl);
+
+		if (perf_evsel__is_group_event(evsel)) {
+			nr_percent = evsel->nr_members;
+			ppercents = calloc(nr_percent, sizeof(double));
+			if (ppercents == NULL)
+				return -1;
+		}
+
+		for (i = 0; i < nr_percent; i++) {
+			percent = disasm__calc_percent(notes,
+					notes->src->lines ? i : evsel->idx + i,
+					offset,
+					next ? next->offset : (s64) len,
+					&path);
+
+			ppercents[i] = percent;
+			if (percent > max_percent)
+				max_percent = percent;
+		}
+
+		if (max_percent < min_pcnt)
+			return -1;
+
+		if (max_lines && printed >= max_lines)
+			return 1;
+
+		if (queue != NULL) {
+			list_for_each_entry_from(queue, &notes->src->source, node) {
+				if (queue == dl)
+					break;
+				disasm_line__print(queue, sym, start, evsel, len,
+						    0, 0, 1, NULL);
+			}
+		}
+
+		color = get_percent_color(max_percent);
+
+		/*
+		 * Also color the filename and line if needed, with
+		 * the same color than the percentage. Don't print it
+		 * twice for close colored addr with the same filename:line
+		 */
+		if (path) {
+			if (!prev_line || strcmp(prev_line, path)
+				       || color != prev_color) {
+				color_fprintf(stdout, color, " %s", path);
+				prev_line = path;
+				prev_color = color;
+			}
+		}
+
+		for (i = 0; i < nr_percent; i++) {
+			percent = ppercents[i];
+			color = get_percent_color(percent);
+			color_fprintf(stdout, color, " %7.2f", percent);
+		}
+
+		printf(" :	");
+		color_fprintf(stdout, PERF_COLOR_MAGENTA, "  %" PRIx64 ":", addr);
+		color_fprintf(stdout, PERF_COLOR_BLUE, "%s\n", dl->line);
+
+		if (ppercents != &percent)
+			free(ppercents);
+
+	} else if (max_lines && printed >= max_lines)
+		return 1;
+	else {
+		int width = 8;
+
+		if (queue)
+			return -1;
+
+		if (perf_evsel__is_group_event(evsel))
+			width *= evsel->nr_members;
+
+		if (!*dl->line)
+			printf(" %*s:\n", width, " ");
+		else
+			printf(" %*s:	%s\n", width, " ", dl->line);
+	}
+
+	return 0;
+}
+
+/*
+ * symbol__parse_objdump_line() parses objdump output (with -d --no-show-raw)
+ * which looks like following
+ *
+ *  0000000000415500 <_init>:
+ *    415500:       sub    $0x8,%rsp
+ *    415504:       mov    0x2f5ad5(%rip),%rax        # 70afe0 <_DYNAMIC+0x2f8>
+ *    41550b:       test   %rax,%rax
+ *    41550e:       je     415515 <_init+0x15>
+ *    415510:       callq  416e70 <__gmon_start__@plt>
+ *    415515:       add    $0x8,%rsp
+ *    415519:       retq
+ *
+ * it will be parsed and saved into struct disasm_line as
+ *  <offset>       <name>  <ops.raw>
+ *
+ * The offset will be a relative offset from the start of the symbol and -1
+ * means that it's not a disassembly line so should be treated differently.
+ * The ops.raw part will be parsed further according to type of the instruction.
+ */
+static int symbol__parse_objdump_line(struct symbol *sym, struct map *map,
+				      FILE *file, size_t privsize,
+				      int *line_nr)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct disasm_line *dl;
+	char *line = NULL, *parsed_line, *tmp, *tmp2, *c;
+	size_t line_len;
+	s64 line_ip, offset = -1;
+	regmatch_t match[2];
+
+	if (getline(&line, &line_len, file) < 0)
+		return -1;
+
+	if (!line)
+		return -1;
+
+	while (line_len != 0 && isspace(line[line_len - 1]))
+		line[--line_len] = '\0';
+
+	c = strchr(line, '\n');
+	if (c)
+		*c = 0;
+
+	line_ip = -1;
+	parsed_line = line;
+
+	/* /filename:linenr ? Save line number and ignore. */
+	if (regexec(&file_lineno, line, 2, match, 0) == 0) {
+		*line_nr = atoi(line + match[1].rm_so);
+		return 0;
+	}
+
+	/*
+	 * Strip leading spaces:
+	 */
+	tmp = line;
+	while (*tmp) {
+		if (*tmp != ' ')
+			break;
+		tmp++;
+	}
+
+	if (*tmp) {
+		/*
+		 * Parse hexa addresses followed by ':'
+		 */
+		line_ip = strtoull(tmp, &tmp2, 16);
+		if (*tmp2 != ':' || tmp == tmp2 || tmp2[1] == '\0')
+			line_ip = -1;
+	}
+
+	if (line_ip != -1) {
+		u64 start = map__rip_2objdump(map, sym->start),
+		    end = map__rip_2objdump(map, sym->end);
+
+		offset = line_ip - start;
+		if ((u64)line_ip < start || (u64)line_ip >= end)
+			offset = -1;
+		else
+			parsed_line = tmp2 + 1;
+	}
+
+	dl = disasm_line__new(offset, parsed_line, privsize, *line_nr);
+	free(line);
+	(*line_nr)++;
+
+	if (dl == NULL)
+		return -1;
+
+	if (dl->ops.target.offset == UINT64_MAX)
+		dl->ops.target.offset = dl->ops.target.addr -
+					map__rip_2objdump(map, sym->start);
+
+	/* kcore has no symbols, so add the call target name */
+	if (dl->ins && ins__is_call(dl->ins) && !dl->ops.target.name) {
+		struct addr_map_symbol target = {
+			.map = map,
+			.addr = dl->ops.target.addr,
+		};
+
+		if (!map_groups__find_ams(&target, NULL) &&
+		    target.sym->start == target.al_addr)
+			dl->ops.target.name = strdup(target.sym->name);
+	}
+
+	disasm__add(&notes->src->source, dl);
+
+	return 0;
+}
+
+static __attribute__((constructor)) void symbol__init_regexpr(void)
+{
+	regcomp(&file_lineno, "^/[^:]+:([0-9]+)", REG_EXTENDED);
+}
+
+static void delete_last_nop(struct symbol *sym)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct list_head *list = &notes->src->source;
+	struct disasm_line *dl;
+
+	while (!list_empty(list)) {
+		dl = list_entry(list->prev, struct disasm_line, node);
+
+		if (dl->ins && dl->ins->ops) {
+			if (dl->ins->ops != &nop_ops)
+				return;
+		} else {
+			if (!strstr(dl->line, " nop ") &&
+			    !strstr(dl->line, " nopl ") &&
+			    !strstr(dl->line, " nopw "))
+				return;
+		}
+
+		list_del(&dl->node);
+		disasm_line__free(dl);
+	}
+}
+
+int symbol__annotate(struct symbol *sym, struct map *map, size_t privsize)
+{
+	struct dso *dso = map->dso;
+	char *filename = dso__build_id_filename(dso, NULL, 0);
+	bool free_filename = true;
+	char command[PATH_MAX * 2];
+	FILE *file;
+	int err = 0;
+	char symfs_filename[PATH_MAX];
+	struct kcore_extract kce;
+	bool delete_extract = false;
+	int lineno = 0;
+
+	if (filename)
+		symbol__join_symfs(symfs_filename, filename);
+
+	if (filename == NULL) {
+		if (dso->has_build_id) {
+			pr_err("Can't annotate %s: not enough memory\n",
+			       sym->name);
+			return -ENOMEM;
+		}
+		goto fallback;
+	} else if (dso__is_kcore(dso)) {
+		goto fallback;
+	} else if (readlink(symfs_filename, command, sizeof(command)) < 0 ||
+		   strstr(command, "[kernel.kallsyms]") ||
+		   access(symfs_filename, R_OK)) {
+		free(filename);
+fallback:
+		/*
+		 * If we don't have build-ids or the build-id file isn't in the
+		 * cache, or is just a kallsyms file, well, lets hope that this
+		 * DSO is the same as when 'perf record' ran.
+		 */
+		filename = (char *)dso->long_name;
+		symbol__join_symfs(symfs_filename, filename);
+		free_filename = false;
+	}
+
+	if (dso->symtab_type == DSO_BINARY_TYPE__KALLSYMS &&
+	    !dso__is_kcore(dso)) {
+		char bf[BUILD_ID_SIZE * 2 + 16] = " with build id ";
+		char *build_id_msg = NULL;
+
+		if (dso->annotate_warned)
+			goto out_free_filename;
+
+		if (dso->has_build_id) {
+			build_id__sprintf(dso->build_id,
+					  sizeof(dso->build_id), bf + 15);
+			build_id_msg = bf;
+		}
+		err = -ENOENT;
+		dso->annotate_warned = 1;
+		pr_err("Can't annotate %s:\n\n"
+		       "No vmlinux file%s\nwas found in the path.\n\n"
+		       "Please use:\n\n"
+		       "  perf buildid-cache -vu vmlinux\n\n"
+		       "or:\n\n"
+		       "  --vmlinux vmlinux\n",
+		       sym->name, build_id_msg ?: "");
+		goto out_free_filename;
+	}
+
+	pr_debug("%s: filename=%s, sym=%s, start=%#" PRIx64 ", end=%#" PRIx64 "\n", __func__,
+		 filename, sym->name, map->unmap_ip(map, sym->start),
+		 map->unmap_ip(map, sym->end));
+
+	pr_debug("annotating [%p] %30s : [%p] %30s\n",
+		 dso, dso->long_name, sym, sym->name);
+
+	if (dso__is_kcore(dso)) {
+		kce.kcore_filename = symfs_filename;
+		kce.addr = map__rip_2objdump(map, sym->start);
+		kce.offs = sym->start;
+		kce.len = sym->end - sym->start;
+		if (!kcore_extract__create(&kce)) {
+			delete_extract = true;
+			strlcpy(symfs_filename, kce.extract_filename,
+				sizeof(symfs_filename));
+			if (free_filename) {
+				free(filename);
+				free_filename = false;
+			}
+			filename = symfs_filename;
+		}
+	} else if (dso__needs_decompress(dso)) {
+		char tmp[PATH_MAX];
+		struct kmod_path m;
+		int fd;
+		bool ret;
+
+		if (kmod_path__parse_ext(&m, symfs_filename))
+			goto out_free_filename;
+
+		snprintf(tmp, PATH_MAX, "/tmp/perf-kmod-XXXXXX");
+
+		fd = mkstemp(tmp);
+		if (fd < 0) {
+			free(m.ext);
+			goto out_free_filename;
+		}
+
+		ret = decompress_to_file(m.ext, symfs_filename, fd);
+
+		free(m.ext);
+		close(fd);
+
+		if (!ret)
+			goto out_free_filename;
+
+		strcpy(symfs_filename, tmp);
+	}
+
+	snprintf(command, sizeof(command),
+		 "%s %s%s --start-address=0x%016" PRIx64
+		 " --stop-address=0x%016" PRIx64
+		 " -l -d %s %s -C %s 2>/dev/null|grep -v %s|expand",
+		 objdump_path ? objdump_path : "objdump",
+		 disassembler_style ? "-M " : "",
+		 disassembler_style ? disassembler_style : "",
+		 map__rip_2objdump(map, sym->start),
+		 map__rip_2objdump(map, sym->end),
+		 symbol_conf.annotate_asm_raw ? "" : "--no-show-raw",
+		 symbol_conf.annotate_src ? "-S" : "",
+		 symfs_filename, filename);
+
+	pr_debug("Executing: %s\n", command);
+
+	file = popen(command, "r");
+	if (!file)
+		goto out_remove_tmp;
+
+	while (!feof(file))
+		if (symbol__parse_objdump_line(sym, map, file, privsize,
+			    &lineno) < 0)
+			break;
+
+	/*
+	 * kallsyms does not have symbol sizes so there may a nop at the end.
+	 * Remove it.
+	 */
+	if (dso__is_kcore(dso))
+		delete_last_nop(sym);
+
+	pclose(file);
+
+out_remove_tmp:
+	if (dso__needs_decompress(dso))
+		unlink(symfs_filename);
+out_free_filename:
+	if (delete_extract)
+		kcore_extract__delete(&kce);
+	if (free_filename)
+		free(filename);
+	return err;
+}
+
+static void insert_source_line(struct rb_root *root, struct source_line *src_line)
+{
+	struct source_line *iter;
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	int i, ret;
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct source_line, node);
+
+		ret = strcmp(iter->path, src_line->path);
+		if (ret == 0) {
+			for (i = 0; i < src_line->nr_pcnt; i++)
+				iter->p[i].percent_sum += src_line->p[i].percent;
+			return;
+		}
+
+		if (ret < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	for (i = 0; i < src_line->nr_pcnt; i++)
+		src_line->p[i].percent_sum = src_line->p[i].percent;
+
+	rb_link_node(&src_line->node, parent, p);
+	rb_insert_color(&src_line->node, root);
+}
+
+static int cmp_source_line(struct source_line *a, struct source_line *b)
+{
+	int i;
+
+	for (i = 0; i < a->nr_pcnt; i++) {
+		if (a->p[i].percent_sum == b->p[i].percent_sum)
+			continue;
+		return a->p[i].percent_sum > b->p[i].percent_sum;
+	}
+
+	return 0;
+}
+
+static void __resort_source_line(struct rb_root *root, struct source_line *src_line)
+{
+	struct source_line *iter;
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct source_line, node);
+
+		if (cmp_source_line(src_line, iter))
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&src_line->node, parent, p);
+	rb_insert_color(&src_line->node, root);
+}
+
+static void resort_source_line(struct rb_root *dest_root, struct rb_root *src_root)
+{
+	struct source_line *src_line;
+	struct rb_node *node;
+
+	node = rb_first(src_root);
+	while (node) {
+		struct rb_node *next;
+
+		src_line = rb_entry(node, struct source_line, node);
+		next = rb_next(node);
+		rb_erase(node, src_root);
+
+		__resort_source_line(dest_root, src_line);
+		node = next;
+	}
+}
+
+static void symbol__free_source_line(struct symbol *sym, int len)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct source_line *src_line = notes->src->lines;
+	size_t sizeof_src_line;
+	int i;
+
+	sizeof_src_line = sizeof(*src_line) +
+			  (sizeof(src_line->p) * (src_line->nr_pcnt - 1));
+
+	for (i = 0; i < len; i++) {
+		free_srcline(src_line->path);
+		src_line = (void *)src_line + sizeof_src_line;
+	}
+
+	zfree(&notes->src->lines);
+}
+
+/* Get the filename:line for the colored entries */
+static int symbol__get_source_line(struct symbol *sym, struct map *map,
+				   struct perf_evsel *evsel,
+				   struct rb_root *root, int len)
+{
+	u64 start;
+	int i, k;
+	int evidx = evsel->idx;
+	struct source_line *src_line;
+	struct annotation *notes = symbol__annotation(sym);
+	struct sym_hist *h = annotation__histogram(notes, evidx);
+	struct rb_root tmp_root = RB_ROOT;
+	int nr_pcnt = 1;
+	u64 h_sum = h->sum;
+	size_t sizeof_src_line = sizeof(struct source_line);
+
+	if (perf_evsel__is_group_event(evsel)) {
+		for (i = 1; i < evsel->nr_members; i++) {
+			h = annotation__histogram(notes, evidx + i);
+			h_sum += h->sum;
+		}
+		nr_pcnt = evsel->nr_members;
+		sizeof_src_line += (nr_pcnt - 1) * sizeof(src_line->p);
+	}
+
+	if (!h_sum)
+		return 0;
+
+	src_line = notes->src->lines = calloc(len, sizeof_src_line);
+	if (!notes->src->lines)
+		return -1;
+
+	start = map__rip_2objdump(map, sym->start);
+
+	for (i = 0; i < len; i++) {
+		u64 offset;
+		double percent_max = 0.0;
+
+		src_line->nr_pcnt = nr_pcnt;
+
+		for (k = 0; k < nr_pcnt; k++) {
+			h = annotation__histogram(notes, evidx + k);
+			src_line->p[k].percent = 100.0 * h->addr[i] / h->sum;
+
+			if (src_line->p[k].percent > percent_max)
+				percent_max = src_line->p[k].percent;
+		}
+
+		if (percent_max <= 0.5)
+			goto next;
+
+		offset = start + i;
+		src_line->path = get_srcline(map->dso, offset, NULL, false);
+		insert_source_line(&tmp_root, src_line);
+
+	next:
+		src_line = (void *)src_line + sizeof_src_line;
+	}
+
+	resort_source_line(root, &tmp_root);
+	return 0;
+}
+
+static void print_summary(struct rb_root *root, const char *filename)
+{
+	struct source_line *src_line;
+	struct rb_node *node;
+
+	printf("\nSorted summary for file %s\n", filename);
+	printf("----------------------------------------------\n\n");
+
+	if (RB_EMPTY_ROOT(root)) {
+		printf(" Nothing higher than %1.1f%%\n", MIN_GREEN);
+		return;
+	}
+
+	node = rb_first(root);
+	while (node) {
+		double percent, percent_max = 0.0;
+		const char *color;
+		char *path;
+		int i;
+
+		src_line = rb_entry(node, struct source_line, node);
+		for (i = 0; i < src_line->nr_pcnt; i++) {
+			percent = src_line->p[i].percent_sum;
+			color = get_percent_color(percent);
+			color_fprintf(stdout, color, " %7.2f", percent);
+
+			if (percent > percent_max)
+				percent_max = percent;
+		}
+
+		path = src_line->path;
+		color = get_percent_color(percent_max);
+		color_fprintf(stdout, color, " %s\n", path);
+
+		node = rb_next(node);
+	}
+}
+
+static void symbol__annotate_hits(struct symbol *sym, struct perf_evsel *evsel)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct sym_hist *h = annotation__histogram(notes, evsel->idx);
+	u64 len = symbol__size(sym), offset;
+
+	for (offset = 0; offset < len; ++offset)
+		if (h->addr[offset] != 0)
+			printf("%*" PRIx64 ": %" PRIu64 "\n", BITS_PER_LONG / 2,
+			       sym->start + offset, h->addr[offset]);
+	printf("%*s: %" PRIu64 "\n", BITS_PER_LONG / 2, "h->sum", h->sum);
+}
+
+int symbol__annotate_printf(struct symbol *sym, struct map *map,
+			    struct perf_evsel *evsel, bool full_paths,
+			    int min_pcnt, int max_lines, int context)
+{
+	struct dso *dso = map->dso;
+	char *filename;
+	const char *d_filename;
+	const char *evsel_name = perf_evsel__name(evsel);
+	struct annotation *notes = symbol__annotation(sym);
+	struct disasm_line *pos, *queue = NULL;
+	u64 start = map__rip_2objdump(map, sym->start);
+	int printed = 2, queue_len = 0;
+	int more = 0;
+	u64 len;
+	int width = 8;
+	int namelen, evsel_name_len, graph_dotted_len;
+
+	filename = strdup(dso->long_name);
+	if (!filename)
+		return -ENOMEM;
+
+	if (full_paths)
+		d_filename = filename;
+	else
+		d_filename = basename(filename);
+
+	len = symbol__size(sym);
+	namelen = strlen(d_filename);
+	evsel_name_len = strlen(evsel_name);
+
+	if (perf_evsel__is_group_event(evsel))
+		width *= evsel->nr_members;
+
+	printf(" %-*.*s|	Source code & Disassembly of %s for %s\n",
+	       width, width, "Percent", d_filename, evsel_name);
+
+	graph_dotted_len = width + namelen + evsel_name_len;
+	printf("-%-*.*s-----------------------------------------\n",
+	       graph_dotted_len, graph_dotted_len, graph_dotted_line);
+
+	if (verbose)
+		symbol__annotate_hits(sym, evsel);
+
+	list_for_each_entry(pos, &notes->src->source, node) {
+		if (context && queue == NULL) {
+			queue = pos;
+			queue_len = 0;
+		}
+
+		switch (disasm_line__print(pos, sym, start, evsel, len,
+					    min_pcnt, printed, max_lines,
+					    queue)) {
+		case 0:
+			++printed;
+			if (context) {
+				printed += queue_len;
+				queue = NULL;
+				queue_len = 0;
+			}
+			break;
+		case 1:
+			/* filtered by max_lines */
+			++more;
+			break;
+		case -1:
+		default:
+			/*
+			 * Filtered by min_pcnt or non IP lines when
+			 * context != 0
+			 */
+			if (!context)
+				break;
+			if (queue_len == context)
+				queue = list_entry(queue->node.next, typeof(*queue), node);
+			else
+				++queue_len;
+			break;
+		}
+	}
+
+	free(filename);
+
+	return more;
+}
+
+void symbol__annotate_zero_histogram(struct symbol *sym, int evidx)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct sym_hist *h = annotation__histogram(notes, evidx);
+
+	memset(h, 0, notes->src->sizeof_sym_hist);
+}
+
+void symbol__annotate_decay_histogram(struct symbol *sym, int evidx)
+{
+	struct annotation *notes = symbol__annotation(sym);
+	struct sym_hist *h = annotation__histogram(notes, evidx);
+	int len = symbol__size(sym), offset;
+
+	h->sum = 0;
+	for (offset = 0; offset < len; ++offset) {
+		h->addr[offset] = h->addr[offset] * 7 / 8;
+		h->sum += h->addr[offset];
+	}
+}
+
+void disasm__purge(struct list_head *head)
+{
+	struct disasm_line *pos, *n;
+
+	list_for_each_entry_safe(pos, n, head, node) {
+		list_del(&pos->node);
+		disasm_line__free(pos);
+	}
+}
+
+static size_t disasm_line__fprintf(struct disasm_line *dl, FILE *fp)
+{
+	size_t printed;
+
+	if (dl->offset == -1)
+		return fprintf(fp, "%s\n", dl->line);
+
+	printed = fprintf(fp, "%#" PRIx64 " %s", dl->offset, dl->name);
+
+	if (dl->ops.raw[0] != '\0') {
+		printed += fprintf(fp, "%.*s %s\n", 6 - (int)printed, " ",
+				   dl->ops.raw);
+	}
+
+	return printed + fprintf(fp, "\n");
+}
+
+size_t disasm__fprintf(struct list_head *head, FILE *fp)
+{
+	struct disasm_line *pos;
+	size_t printed = 0;
+
+	list_for_each_entry(pos, head, node)
+		printed += disasm_line__fprintf(pos, fp);
+
+	return printed;
+}
+
+int symbol__tty_annotate(struct symbol *sym, struct map *map,
+			 struct perf_evsel *evsel, bool print_lines,
+			 bool full_paths, int min_pcnt, int max_lines)
+{
+	struct dso *dso = map->dso;
+	struct rb_root source_line = RB_ROOT;
+	u64 len;
+
+	if (symbol__annotate(sym, map, 0) < 0)
+		return -1;
+
+	len = symbol__size(sym);
+
+	if (print_lines) {
+		symbol__get_source_line(sym, map, evsel, &source_line, len);
+		print_summary(&source_line, dso->long_name);
+	}
+
+	symbol__annotate_printf(sym, map, evsel, full_paths,
+				min_pcnt, max_lines, 0);
+	if (print_lines)
+		symbol__free_source_line(sym, len);
+
+	disasm__purge(&symbol__annotation(sym)->src->source);
+
+	return 0;
+}
+
+int hist_entry__annotate(struct hist_entry *he, size_t privsize)
+{
+	return symbol__annotate(he->ms.sym, he->ms.map, privsize);
+}
+
+bool ui__has_annotation(void)
+{
+	return use_browser == 1 && sort__has_sym;
+}
diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h
new file mode 100644
index 000000000..cadbdc90a
--- /dev/null
+++ b/tools/perf/util/annotate.h
@@ -0,0 +1,172 @@
+#ifndef __PERF_ANNOTATE_H
+#define __PERF_ANNOTATE_H
+
+#include <stdbool.h>
+#include <stdint.h>
+#include <linux/types.h>
+#include "symbol.h"
+#include "hist.h"
+#include "sort.h"
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <pthread.h>
+
+struct ins;
+
+struct ins_operands {
+	char	*raw;
+	struct {
+		char	*raw;
+		char	*name;
+		u64	addr;
+		u64	offset;
+	} target;
+	union {
+		struct {
+			char	*raw;
+			char	*name;
+			u64	addr;
+		} source;
+		struct {
+			struct ins *ins;
+			struct ins_operands *ops;
+		} locked;
+	};
+};
+
+struct ins_ops {
+	void (*free)(struct ins_operands *ops);
+	int (*parse)(struct ins_operands *ops);
+	int (*scnprintf)(struct ins *ins, char *bf, size_t size,
+			 struct ins_operands *ops);
+};
+
+struct ins {
+	const char     *name;
+	struct ins_ops *ops;
+};
+
+bool ins__is_jump(const struct ins *ins);
+bool ins__is_call(const struct ins *ins);
+int ins__scnprintf(struct ins *ins, char *bf, size_t size, struct ins_operands *ops);
+
+struct annotation;
+
+struct disasm_line {
+	struct list_head    node;
+	s64		    offset;
+	char		    *line;
+	char		    *name;
+	struct ins	    *ins;
+	int		    line_nr;
+	struct ins_operands ops;
+};
+
+static inline bool disasm_line__has_offset(const struct disasm_line *dl)
+{
+	return dl->ops.target.offset != UINT64_MAX;
+}
+
+void disasm_line__free(struct disasm_line *dl);
+struct disasm_line *disasm__get_next_ip_line(struct list_head *head, struct disasm_line *pos);
+int disasm_line__scnprintf(struct disasm_line *dl, char *bf, size_t size, bool raw);
+size_t disasm__fprintf(struct list_head *head, FILE *fp);
+double disasm__calc_percent(struct annotation *notes, int evidx, s64 offset,
+			    s64 end, const char **path);
+
+struct sym_hist {
+	u64		sum;
+	u64		addr[0];
+};
+
+struct source_line_percent {
+	double		percent;
+	double		percent_sum;
+};
+
+struct source_line {
+	struct rb_node	node;
+	char		*path;
+	int		nr_pcnt;
+	struct source_line_percent p[1];
+};
+
+/** struct annotated_source - symbols with hits have this attached as in sannotation
+ *
+ * @histogram: Array of addr hit histograms per event being monitored
+ * @lines: If 'print_lines' is specified, per source code line percentages
+ * @source: source parsed from a disassembler like objdump -dS
+ *
+ * lines is allocated, percentages calculated and all sorted by percentage
+ * when the annotation is about to be presented, so the percentages are for
+ * one of the entries in the histogram array, i.e. for the event/counter being
+ * presented. It is deallocated right after symbol__{tui,tty,etc}_annotate
+ * returns.
+ */
+struct annotated_source {
+	struct list_head   source;
+	struct source_line *lines;
+	int    		   nr_histograms;
+	int    		   sizeof_sym_hist;
+	struct sym_hist	   histograms[0];
+};
+
+struct annotation {
+	pthread_mutex_t		lock;
+	struct annotated_source *src;
+};
+
+static inline struct sym_hist *annotation__histogram(struct annotation *notes, int idx)
+{
+	return (((void *)&notes->src->histograms) +
+	 	(notes->src->sizeof_sym_hist * idx));
+}
+
+static inline struct annotation *symbol__annotation(struct symbol *sym)
+{
+	return (void *)sym - symbol_conf.priv_size;
+}
+
+int addr_map_symbol__inc_samples(struct addr_map_symbol *ams, int evidx);
+
+int hist_entry__inc_addr_samples(struct hist_entry *he, int evidx, u64 addr);
+
+int symbol__alloc_hist(struct symbol *sym);
+void symbol__annotate_zero_histograms(struct symbol *sym);
+
+int symbol__annotate(struct symbol *sym, struct map *map, size_t privsize);
+
+int hist_entry__annotate(struct hist_entry *he, size_t privsize);
+
+int symbol__annotate_init(struct map *map __maybe_unused, struct symbol *sym);
+int symbol__annotate_printf(struct symbol *sym, struct map *map,
+			    struct perf_evsel *evsel, bool full_paths,
+			    int min_pcnt, int max_lines, int context);
+void symbol__annotate_zero_histogram(struct symbol *sym, int evidx);
+void symbol__annotate_decay_histogram(struct symbol *sym, int evidx);
+void disasm__purge(struct list_head *head);
+
+bool ui__has_annotation(void);
+
+int symbol__tty_annotate(struct symbol *sym, struct map *map,
+			 struct perf_evsel *evsel, bool print_lines,
+			 bool full_paths, int min_pcnt, int max_lines);
+
+#ifdef HAVE_SLANG_SUPPORT
+int symbol__tui_annotate(struct symbol *sym, struct map *map,
+			 struct perf_evsel *evsel,
+			 struct hist_browser_timer *hbt);
+#else
+static inline int symbol__tui_annotate(struct symbol *sym __maybe_unused,
+				struct map *map __maybe_unused,
+				struct perf_evsel *evsel  __maybe_unused,
+				struct hist_browser_timer *hbt
+				__maybe_unused)
+{
+	return 0;
+}
+#endif
+
+extern const char	*disassembler_style;
+
+#endif	/* __PERF_ANNOTATE_H */
diff --git a/tools/perf/util/bitmap.c b/tools/perf/util/bitmap.c
new file mode 100644
index 000000000..0a1adc111
--- /dev/null
+++ b/tools/perf/util/bitmap.c
@@ -0,0 +1,31 @@
+/*
+ * From lib/bitmap.c
+ * Helper functions for bitmap.h.
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+#include <linux/bitmap.h>
+
+int __bitmap_weight(const unsigned long *bitmap, int bits)
+{
+	int k, w = 0, lim = bits/BITS_PER_LONG;
+
+	for (k = 0; k < lim; k++)
+		w += hweight_long(bitmap[k]);
+
+	if (bits % BITS_PER_LONG)
+		w += hweight_long(bitmap[k] & BITMAP_LAST_WORD_MASK(bits));
+
+	return w;
+}
+
+void __bitmap_or(unsigned long *dst, const unsigned long *bitmap1,
+		 const unsigned long *bitmap2, int bits)
+{
+	int k;
+	int nr = BITS_TO_LONGS(bits);
+
+	for (k = 0; k < nr; k++)
+		dst[k] = bitmap1[k] | bitmap2[k];
+}
diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c
new file mode 100644
index 000000000..61867dff5
--- /dev/null
+++ b/tools/perf/util/build-id.c
@@ -0,0 +1,538 @@
+/*
+ * build-id.c
+ *
+ * build-id support
+ *
+ * Copyright (C) 2009, 2010 Red Hat Inc.
+ * Copyright (C) 2009, 2010 Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+#include "util.h"
+#include <stdio.h>
+#include "build-id.h"
+#include "event.h"
+#include "symbol.h"
+#include <linux/kernel.h>
+#include "debug.h"
+#include "session.h"
+#include "tool.h"
+#include "header.h"
+#include "vdso.h"
+
+
+static bool no_buildid_cache;
+
+int build_id__mark_dso_hit(struct perf_tool *tool __maybe_unused,
+			   union perf_event *event,
+			   struct perf_sample *sample,
+			   struct perf_evsel *evsel __maybe_unused,
+			   struct machine *machine)
+{
+	struct addr_location al;
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct thread *thread = machine__findnew_thread(machine, sample->pid,
+							sample->tid);
+
+	if (thread == NULL) {
+		pr_err("problem processing %d event, skipping it.\n",
+			event->header.type);
+		return -1;
+	}
+
+	thread__find_addr_map(thread, cpumode, MAP__FUNCTION, sample->ip, &al);
+
+	if (al.map != NULL)
+		al.map->dso->hit = 1;
+
+	return 0;
+}
+
+static int perf_event__exit_del_thread(struct perf_tool *tool __maybe_unused,
+				       union perf_event *event,
+				       struct perf_sample *sample
+				       __maybe_unused,
+				       struct machine *machine)
+{
+	struct thread *thread = machine__findnew_thread(machine,
+							event->fork.pid,
+							event->fork.tid);
+
+	dump_printf("(%d:%d):(%d:%d)\n", event->fork.pid, event->fork.tid,
+		    event->fork.ppid, event->fork.ptid);
+
+	if (thread)
+		machine__remove_thread(machine, thread);
+
+	return 0;
+}
+
+struct perf_tool build_id__mark_dso_hit_ops = {
+	.sample	= build_id__mark_dso_hit,
+	.mmap	= perf_event__process_mmap,
+	.mmap2	= perf_event__process_mmap2,
+	.fork	= perf_event__process_fork,
+	.exit	= perf_event__exit_del_thread,
+	.attr		 = perf_event__process_attr,
+	.build_id	 = perf_event__process_build_id,
+};
+
+int build_id__sprintf(const u8 *build_id, int len, char *bf)
+{
+	char *bid = bf;
+	const u8 *raw = build_id;
+	int i;
+
+	for (i = 0; i < len; ++i) {
+		sprintf(bid, "%02x", *raw);
+		++raw;
+		bid += 2;
+	}
+
+	return raw - build_id;
+}
+
+/* asnprintf consolidates asprintf and snprintf */
+static int asnprintf(char **strp, size_t size, const char *fmt, ...)
+{
+	va_list ap;
+	int ret;
+
+	if (!strp)
+		return -EINVAL;
+
+	va_start(ap, fmt);
+	if (*strp)
+		ret = vsnprintf(*strp, size, fmt, ap);
+	else
+		ret = vasprintf(strp, fmt, ap);
+	va_end(ap);
+
+	return ret;
+}
+
+static char *build_id__filename(const char *sbuild_id, char *bf, size_t size)
+{
+	char *tmp = bf;
+	int ret = asnprintf(&bf, size, "%s/.build-id/%.2s/%s", buildid_dir,
+			    sbuild_id, sbuild_id + 2);
+	if (ret < 0 || (tmp && size < (unsigned int)ret))
+		return NULL;
+	return bf;
+}
+
+char *dso__build_id_filename(const struct dso *dso, char *bf, size_t size)
+{
+	char build_id_hex[BUILD_ID_SIZE * 2 + 1];
+
+	if (!dso->has_build_id)
+		return NULL;
+
+	build_id__sprintf(dso->build_id, sizeof(dso->build_id), build_id_hex);
+	return build_id__filename(build_id_hex, bf, size);
+}
+
+#define dsos__for_each_with_build_id(pos, head)	\
+	list_for_each_entry(pos, head, node)	\
+		if (!pos->has_build_id)		\
+			continue;		\
+		else
+
+static int write_buildid(const char *name, size_t name_len, u8 *build_id,
+			 pid_t pid, u16 misc, int fd)
+{
+	int err;
+	struct build_id_event b;
+	size_t len;
+
+	len = name_len + 1;
+	len = PERF_ALIGN(len, NAME_ALIGN);
+
+	memset(&b, 0, sizeof(b));
+	memcpy(&b.build_id, build_id, BUILD_ID_SIZE);
+	b.pid = pid;
+	b.header.misc = misc;
+	b.header.size = sizeof(b) + len;
+
+	err = writen(fd, &b, sizeof(b));
+	if (err < 0)
+		return err;
+
+	return write_padded(fd, name, name_len + 1, len);
+}
+
+static int __dsos__write_buildid_table(struct list_head *head,
+				       struct machine *machine,
+				       pid_t pid, u16 misc, int fd)
+{
+	char nm[PATH_MAX];
+	struct dso *pos;
+
+	dsos__for_each_with_build_id(pos, head) {
+		int err;
+		const char *name;
+		size_t name_len;
+
+		if (!pos->hit)
+			continue;
+
+		if (dso__is_vdso(pos)) {
+			name = pos->short_name;
+			name_len = pos->short_name_len + 1;
+		} else if (dso__is_kcore(pos)) {
+			machine__mmap_name(machine, nm, sizeof(nm));
+			name = nm;
+			name_len = strlen(nm) + 1;
+		} else {
+			name = pos->long_name;
+			name_len = pos->long_name_len + 1;
+		}
+
+		err = write_buildid(name, name_len, pos->build_id,
+				    pid, misc, fd);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int machine__write_buildid_table(struct machine *machine, int fd)
+{
+	int err;
+	u16 kmisc = PERF_RECORD_MISC_KERNEL,
+	    umisc = PERF_RECORD_MISC_USER;
+
+	if (!machine__is_host(machine)) {
+		kmisc = PERF_RECORD_MISC_GUEST_KERNEL;
+		umisc = PERF_RECORD_MISC_GUEST_USER;
+	}
+
+	err = __dsos__write_buildid_table(&machine->kernel_dsos.head, machine,
+					  machine->pid, kmisc, fd);
+	if (err == 0)
+		err = __dsos__write_buildid_table(&machine->user_dsos.head,
+						  machine, machine->pid, umisc,
+						  fd);
+	return err;
+}
+
+int perf_session__write_buildid_table(struct perf_session *session, int fd)
+{
+	struct rb_node *nd;
+	int err = machine__write_buildid_table(&session->machines.host, fd);
+
+	if (err)
+		return err;
+
+	for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		err = machine__write_buildid_table(pos, fd);
+		if (err)
+			break;
+	}
+	return err;
+}
+
+static int __dsos__hit_all(struct list_head *head)
+{
+	struct dso *pos;
+
+	list_for_each_entry(pos, head, node)
+		pos->hit = true;
+
+	return 0;
+}
+
+static int machine__hit_all_dsos(struct machine *machine)
+{
+	int err;
+
+	err = __dsos__hit_all(&machine->kernel_dsos.head);
+	if (err)
+		return err;
+
+	return __dsos__hit_all(&machine->user_dsos.head);
+}
+
+int dsos__hit_all(struct perf_session *session)
+{
+	struct rb_node *nd;
+	int err;
+
+	err = machine__hit_all_dsos(&session->machines.host);
+	if (err)
+		return err;
+
+	for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+
+		err = machine__hit_all_dsos(pos);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+void disable_buildid_cache(void)
+{
+	no_buildid_cache = true;
+}
+
+static char *build_id_cache__dirname_from_path(const char *name,
+					       bool is_kallsyms, bool is_vdso)
+{
+	char *realname = (char *)name, *filename;
+	bool slash = is_kallsyms || is_vdso;
+
+	if (!slash) {
+		realname = realpath(name, NULL);
+		if (!realname)
+			return NULL;
+	}
+
+	if (asprintf(&filename, "%s%s%s", buildid_dir, slash ? "/" : "",
+		     is_vdso ? DSO__NAME_VDSO : realname) < 0)
+		filename = NULL;
+
+	if (!slash)
+		free(realname);
+
+	return filename;
+}
+
+int build_id_cache__list_build_ids(const char *pathname,
+				   struct strlist **result)
+{
+	struct strlist *list;
+	char *dir_name;
+	DIR *dir;
+	struct dirent *d;
+	int ret = 0;
+
+	list = strlist__new(true, NULL);
+	dir_name = build_id_cache__dirname_from_path(pathname, false, false);
+	if (!list || !dir_name) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	/* List up all dirents */
+	dir = opendir(dir_name);
+	if (!dir) {
+		ret = -errno;
+		goto out;
+	}
+
+	while ((d = readdir(dir)) != NULL) {
+		if (!strcmp(d->d_name, ".") || !strcmp(d->d_name, ".."))
+			continue;
+		strlist__add(list, d->d_name);
+	}
+	closedir(dir);
+
+out:
+	free(dir_name);
+	if (ret)
+		strlist__delete(list);
+	else
+		*result = list;
+
+	return ret;
+}
+
+int build_id_cache__add_s(const char *sbuild_id, const char *name,
+			  bool is_kallsyms, bool is_vdso)
+{
+	const size_t size = PATH_MAX;
+	char *realname = NULL, *filename = NULL, *dir_name = NULL,
+	     *linkname = zalloc(size), *targetname, *tmp;
+	int err = -1;
+
+	if (!is_kallsyms) {
+		realname = realpath(name, NULL);
+		if (!realname)
+			goto out_free;
+	}
+
+	dir_name = build_id_cache__dirname_from_path(name, is_kallsyms, is_vdso);
+	if (!dir_name)
+		goto out_free;
+
+	if (mkdir_p(dir_name, 0755))
+		goto out_free;
+
+	if (asprintf(&filename, "%s/%s", dir_name, sbuild_id) < 0) {
+		filename = NULL;
+		goto out_free;
+	}
+
+	if (access(filename, F_OK)) {
+		if (is_kallsyms) {
+			 if (copyfile("/proc/kallsyms", filename))
+				goto out_free;
+		} else if (link(realname, filename) && errno != EEXIST &&
+				copyfile(name, filename))
+			goto out_free;
+	}
+
+	if (!build_id__filename(sbuild_id, linkname, size))
+		goto out_free;
+	tmp = strrchr(linkname, '/');
+	*tmp = '\0';
+
+	if (access(linkname, X_OK) && mkdir_p(linkname, 0755))
+		goto out_free;
+
+	*tmp = '/';
+	targetname = filename + strlen(buildid_dir) - 5;
+	memcpy(targetname, "../..", 5);
+
+	if (symlink(targetname, linkname) == 0)
+		err = 0;
+out_free:
+	if (!is_kallsyms)
+		free(realname);
+	free(filename);
+	free(dir_name);
+	free(linkname);
+	return err;
+}
+
+static int build_id_cache__add_b(const u8 *build_id, size_t build_id_size,
+				 const char *name, bool is_kallsyms,
+				 bool is_vdso)
+{
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	build_id__sprintf(build_id, build_id_size, sbuild_id);
+
+	return build_id_cache__add_s(sbuild_id, name, is_kallsyms, is_vdso);
+}
+
+bool build_id_cache__cached(const char *sbuild_id)
+{
+	bool ret = false;
+	char *filename = build_id__filename(sbuild_id, NULL, 0);
+
+	if (filename && !access(filename, F_OK))
+		ret = true;
+	free(filename);
+
+	return ret;
+}
+
+int build_id_cache__remove_s(const char *sbuild_id)
+{
+	const size_t size = PATH_MAX;
+	char *filename = zalloc(size),
+	     *linkname = zalloc(size), *tmp;
+	int err = -1;
+
+	if (filename == NULL || linkname == NULL)
+		goto out_free;
+
+	if (!build_id__filename(sbuild_id, linkname, size))
+		goto out_free;
+
+	if (access(linkname, F_OK))
+		goto out_free;
+
+	if (readlink(linkname, filename, size - 1) < 0)
+		goto out_free;
+
+	if (unlink(linkname))
+		goto out_free;
+
+	/*
+	 * Since the link is relative, we must make it absolute:
+	 */
+	tmp = strrchr(linkname, '/') + 1;
+	snprintf(tmp, size - (tmp - linkname), "%s", filename);
+
+	if (unlink(linkname))
+		goto out_free;
+
+	err = 0;
+out_free:
+	free(filename);
+	free(linkname);
+	return err;
+}
+
+static int dso__cache_build_id(struct dso *dso, struct machine *machine)
+{
+	bool is_kallsyms = dso->kernel && dso->long_name[0] != '/';
+	bool is_vdso = dso__is_vdso(dso);
+	const char *name = dso->long_name;
+	char nm[PATH_MAX];
+
+	if (dso__is_kcore(dso)) {
+		is_kallsyms = true;
+		machine__mmap_name(machine, nm, sizeof(nm));
+		name = nm;
+	}
+	return build_id_cache__add_b(dso->build_id, sizeof(dso->build_id), name,
+				     is_kallsyms, is_vdso);
+}
+
+static int __dsos__cache_build_ids(struct list_head *head,
+				   struct machine *machine)
+{
+	struct dso *pos;
+	int err = 0;
+
+	dsos__for_each_with_build_id(pos, head)
+		if (dso__cache_build_id(pos, machine))
+			err = -1;
+
+	return err;
+}
+
+static int machine__cache_build_ids(struct machine *machine)
+{
+	int ret = __dsos__cache_build_ids(&machine->kernel_dsos.head, machine);
+	ret |= __dsos__cache_build_ids(&machine->user_dsos.head, machine);
+	return ret;
+}
+
+int perf_session__cache_build_ids(struct perf_session *session)
+{
+	struct rb_node *nd;
+	int ret;
+
+	if (no_buildid_cache)
+		return 0;
+
+	if (mkdir(buildid_dir, 0755) != 0 && errno != EEXIST)
+		return -1;
+
+	ret = machine__cache_build_ids(&session->machines.host);
+
+	for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		ret |= machine__cache_build_ids(pos);
+	}
+	return ret ? -1 : 0;
+}
+
+static bool machine__read_build_ids(struct machine *machine, bool with_hits)
+{
+	bool ret;
+
+	ret  = __dsos__read_build_ids(&machine->kernel_dsos.head, with_hits);
+	ret |= __dsos__read_build_ids(&machine->user_dsos.head, with_hits);
+	return ret;
+}
+
+bool perf_session__read_build_ids(struct perf_session *session, bool with_hits)
+{
+	struct rb_node *nd;
+	bool ret = machine__read_build_ids(&session->machines.host, with_hits);
+
+	for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		ret |= machine__read_build_ids(pos, with_hits);
+	}
+
+	return ret;
+}
diff --git a/tools/perf/util/build-id.h b/tools/perf/util/build-id.h
new file mode 100644
index 000000000..85011222c
--- /dev/null
+++ b/tools/perf/util/build-id.h
@@ -0,0 +1,34 @@
+#ifndef PERF_BUILD_ID_H_
+#define PERF_BUILD_ID_H_ 1
+
+#define BUILD_ID_SIZE 20
+
+#include "tool.h"
+#include "strlist.h"
+#include <linux/types.h>
+
+extern struct perf_tool build_id__mark_dso_hit_ops;
+struct dso;
+
+int build_id__sprintf(const u8 *build_id, int len, char *bf);
+char *dso__build_id_filename(const struct dso *dso, char *bf, size_t size);
+
+int build_id__mark_dso_hit(struct perf_tool *tool, union perf_event *event,
+			   struct perf_sample *sample, struct perf_evsel *evsel,
+			   struct machine *machine);
+
+int dsos__hit_all(struct perf_session *session);
+
+bool perf_session__read_build_ids(struct perf_session *session, bool with_hits);
+int perf_session__write_buildid_table(struct perf_session *session, int fd);
+int perf_session__cache_build_ids(struct perf_session *session);
+
+int build_id_cache__list_build_ids(const char *pathname,
+				   struct strlist **result);
+bool build_id_cache__cached(const char *sbuild_id);
+int build_id_cache__add_s(const char *sbuild_id,
+			  const char *name, bool is_kallsyms, bool is_vdso);
+int build_id_cache__remove_s(const char *sbuild_id);
+void disable_buildid_cache(void);
+
+#endif
diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h
new file mode 100644
index 000000000..fbcca21d6
--- /dev/null
+++ b/tools/perf/util/cache.h
@@ -0,0 +1,80 @@
+#ifndef __PERF_CACHE_H
+#define __PERF_CACHE_H
+
+#include <stdbool.h>
+#include "util.h"
+#include "strbuf.h"
+#include "../perf.h"
+#include "../ui/ui.h"
+
+#define CMD_EXEC_PATH "--exec-path"
+#define CMD_PERF_DIR "--perf-dir="
+#define CMD_WORK_TREE "--work-tree="
+#define CMD_DEBUGFS_DIR "--debugfs-dir="
+
+#define PERF_DIR_ENVIRONMENT "PERF_DIR"
+#define PERF_WORK_TREE_ENVIRONMENT "PERF_WORK_TREE"
+#define EXEC_PATH_ENVIRONMENT "PERF_EXEC_PATH"
+#define DEFAULT_PERF_DIR_ENVIRONMENT ".perf"
+#define PERF_DEBUGFS_ENVIRONMENT "PERF_DEBUGFS_DIR"
+#define PERF_TRACEFS_ENVIRONMENT "PERF_TRACEFS_DIR"
+
+typedef int (*config_fn_t)(const char *, const char *, void *);
+extern int perf_default_config(const char *, const char *, void *);
+extern int perf_config(config_fn_t fn, void *);
+extern int perf_config_int(const char *, const char *);
+extern u64 perf_config_u64(const char *, const char *);
+extern int perf_config_bool(const char *, const char *);
+extern int config_error_nonbool(const char *);
+extern const char *perf_config_dirname(const char *, const char *);
+
+/* pager.c */
+extern void setup_pager(void);
+extern const char *pager_program;
+extern int pager_in_use(void);
+extern int pager_use_color;
+
+char *alias_lookup(const char *alias);
+int split_cmdline(char *cmdline, const char ***argv);
+
+#define alloc_nr(x) (((x)+16)*3/2)
+
+/*
+ * Realloc the buffer pointed at by variable 'x' so that it can hold
+ * at least 'nr' entries; the number of entries currently allocated
+ * is 'alloc', using the standard growing factor alloc_nr() macro.
+ *
+ * DO NOT USE any expression with side-effect for 'x' or 'alloc'.
+ */
+#define ALLOC_GROW(x, nr, alloc) \
+	do { \
+		if ((nr) > alloc) { \
+			if (alloc_nr(alloc) < (nr)) \
+				alloc = (nr); \
+			else \
+				alloc = alloc_nr(alloc); \
+			x = xrealloc((x), alloc * sizeof(*(x))); \
+		} \
+	} while(0)
+
+
+static inline int is_absolute_path(const char *path)
+{
+	return path[0] == '/';
+}
+
+const char *make_nonrelative_path(const char *path);
+char *strip_path_suffix(const char *path, const char *suffix);
+
+extern char *mkpath(const char *fmt, ...) __attribute__((format (printf, 1, 2)));
+extern char *perf_path(const char *fmt, ...) __attribute__((format (printf, 1, 2)));
+
+extern char *perf_pathdup(const char *fmt, ...)
+	__attribute__((format (printf, 1, 2)));
+
+#ifndef __UCLIBC__
+/* Matches the libc/libbsd function attribute so we declare this unconditionally: */
+extern size_t strlcpy(char *dest, const char *src, size_t size);
+#endif
+
+#endif /* __PERF_CACHE_H */
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
new file mode 100644
index 000000000..9f643ee77
--- /dev/null
+++ b/tools/perf/util/callchain.c
@@ -0,0 +1,881 @@
+/*
+ * Copyright (C) 2009-2011, Frederic Weisbecker <fweisbec@gmail.com>
+ *
+ * Handle the callchains from the stream in an ad-hoc radix tree and then
+ * sort them in an rbtree.
+ *
+ * Using a radix for code path provides a fast retrieval and factorizes
+ * memory use. Also that lets us use the paths in a hierarchical graph view.
+ *
+ */
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <errno.h>
+#include <math.h>
+
+#include "asm/bug.h"
+
+#include "hist.h"
+#include "util.h"
+#include "sort.h"
+#include "machine.h"
+#include "callchain.h"
+
+__thread struct callchain_cursor callchain_cursor;
+
+#ifdef HAVE_DWARF_UNWIND_SUPPORT
+static int get_stack_size(const char *str, unsigned long *_size)
+{
+	char *endptr;
+	unsigned long size;
+	unsigned long max_size = round_down(USHRT_MAX, sizeof(u64));
+
+	size = strtoul(str, &endptr, 0);
+
+	do {
+		if (*endptr)
+			break;
+
+		size = round_up(size, sizeof(u64));
+		if (!size || size > max_size)
+			break;
+
+		*_size = size;
+		return 0;
+
+	} while (0);
+
+	pr_err("callchain: Incorrect stack dump size (max %ld): %s\n",
+	       max_size, str);
+	return -1;
+}
+#endif /* HAVE_DWARF_UNWIND_SUPPORT */
+
+int parse_callchain_record_opt(const char *arg)
+{
+	char *tok, *name, *saveptr = NULL;
+	char *buf;
+	int ret = -1;
+
+	/* We need buffer that we know we can write to. */
+	buf = malloc(strlen(arg) + 1);
+	if (!buf)
+		return -ENOMEM;
+
+	strcpy(buf, arg);
+
+	tok = strtok_r((char *)buf, ",", &saveptr);
+	name = tok ? : (char *)buf;
+
+	do {
+		/* Framepointer style */
+		if (!strncmp(name, "fp", sizeof("fp"))) {
+			if (!strtok_r(NULL, ",", &saveptr)) {
+				callchain_param.record_mode = CALLCHAIN_FP;
+				ret = 0;
+			} else
+				pr_err("callchain: No more arguments "
+				       "needed for --call-graph fp\n");
+			break;
+
+#ifdef HAVE_DWARF_UNWIND_SUPPORT
+		/* Dwarf style */
+		} else if (!strncmp(name, "dwarf", sizeof("dwarf"))) {
+			const unsigned long default_stack_dump_size = 8192;
+
+			ret = 0;
+			callchain_param.record_mode = CALLCHAIN_DWARF;
+			callchain_param.dump_size = default_stack_dump_size;
+
+			tok = strtok_r(NULL, ",", &saveptr);
+			if (tok) {
+				unsigned long size = 0;
+
+				ret = get_stack_size(tok, &size);
+				callchain_param.dump_size = size;
+			}
+#endif /* HAVE_DWARF_UNWIND_SUPPORT */
+		} else if (!strncmp(name, "lbr", sizeof("lbr"))) {
+			if (!strtok_r(NULL, ",", &saveptr)) {
+				callchain_param.record_mode = CALLCHAIN_LBR;
+				ret = 0;
+			} else
+				pr_err("callchain: No more arguments "
+					"needed for --call-graph lbr\n");
+			break;
+		} else {
+			pr_err("callchain: Unknown --call-graph option "
+			       "value: %s\n", arg);
+			break;
+		}
+
+	} while (0);
+
+	free(buf);
+	return ret;
+}
+
+static int parse_callchain_mode(const char *value)
+{
+	if (!strncmp(value, "graph", strlen(value))) {
+		callchain_param.mode = CHAIN_GRAPH_ABS;
+		return 0;
+	}
+	if (!strncmp(value, "flat", strlen(value))) {
+		callchain_param.mode = CHAIN_FLAT;
+		return 0;
+	}
+	if (!strncmp(value, "fractal", strlen(value))) {
+		callchain_param.mode = CHAIN_GRAPH_REL;
+		return 0;
+	}
+	return -1;
+}
+
+static int parse_callchain_order(const char *value)
+{
+	if (!strncmp(value, "caller", strlen(value))) {
+		callchain_param.order = ORDER_CALLER;
+		return 0;
+	}
+	if (!strncmp(value, "callee", strlen(value))) {
+		callchain_param.order = ORDER_CALLEE;
+		return 0;
+	}
+	return -1;
+}
+
+static int parse_callchain_sort_key(const char *value)
+{
+	if (!strncmp(value, "function", strlen(value))) {
+		callchain_param.key = CCKEY_FUNCTION;
+		return 0;
+	}
+	if (!strncmp(value, "address", strlen(value))) {
+		callchain_param.key = CCKEY_ADDRESS;
+		return 0;
+	}
+	if (!strncmp(value, "branch", strlen(value))) {
+		callchain_param.branch_callstack = 1;
+		return 0;
+	}
+	return -1;
+}
+
+int
+parse_callchain_report_opt(const char *arg)
+{
+	char *tok;
+	char *endptr;
+	bool minpcnt_set = false;
+
+	symbol_conf.use_callchain = true;
+
+	if (!arg)
+		return 0;
+
+	while ((tok = strtok((char *)arg, ",")) != NULL) {
+		if (!strncmp(tok, "none", strlen(tok))) {
+			callchain_param.mode = CHAIN_NONE;
+			symbol_conf.use_callchain = false;
+			return 0;
+		}
+
+		if (!parse_callchain_mode(tok) ||
+		    !parse_callchain_order(tok) ||
+		    !parse_callchain_sort_key(tok)) {
+			/* parsing ok - move on to the next */
+		} else if (!minpcnt_set) {
+			/* try to get the min percent */
+			callchain_param.min_percent = strtod(tok, &endptr);
+			if (tok == endptr)
+				return -1;
+			minpcnt_set = true;
+		} else {
+			/* try print limit at last */
+			callchain_param.print_limit = strtoul(tok, &endptr, 0);
+			if (tok == endptr)
+				return -1;
+		}
+
+		arg = NULL;
+	}
+
+	if (callchain_register_param(&callchain_param) < 0) {
+		pr_err("Can't register callchain params\n");
+		return -1;
+	}
+	return 0;
+}
+
+int perf_callchain_config(const char *var, const char *value)
+{
+	char *endptr;
+
+	if (prefixcmp(var, "call-graph."))
+		return 0;
+	var += sizeof("call-graph.") - 1;
+
+	if (!strcmp(var, "record-mode"))
+		return parse_callchain_record_opt(value);
+#ifdef HAVE_DWARF_UNWIND_SUPPORT
+	if (!strcmp(var, "dump-size")) {
+		unsigned long size = 0;
+		int ret;
+
+		ret = get_stack_size(value, &size);
+		callchain_param.dump_size = size;
+
+		return ret;
+	}
+#endif
+	if (!strcmp(var, "print-type"))
+		return parse_callchain_mode(value);
+	if (!strcmp(var, "order"))
+		return parse_callchain_order(value);
+	if (!strcmp(var, "sort-key"))
+		return parse_callchain_sort_key(value);
+	if (!strcmp(var, "threshold")) {
+		callchain_param.min_percent = strtod(value, &endptr);
+		if (value == endptr)
+			return -1;
+	}
+	if (!strcmp(var, "print-limit")) {
+		callchain_param.print_limit = strtod(value, &endptr);
+		if (value == endptr)
+			return -1;
+	}
+
+	return 0;
+}
+
+static void
+rb_insert_callchain(struct rb_root *root, struct callchain_node *chain,
+		    enum chain_mode mode)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct callchain_node *rnode;
+	u64 chain_cumul = callchain_cumul_hits(chain);
+
+	while (*p) {
+		u64 rnode_cumul;
+
+		parent = *p;
+		rnode = rb_entry(parent, struct callchain_node, rb_node);
+		rnode_cumul = callchain_cumul_hits(rnode);
+
+		switch (mode) {
+		case CHAIN_FLAT:
+			if (rnode->hit < chain->hit)
+				p = &(*p)->rb_left;
+			else
+				p = &(*p)->rb_right;
+			break;
+		case CHAIN_GRAPH_ABS: /* Falldown */
+		case CHAIN_GRAPH_REL:
+			if (rnode_cumul < chain_cumul)
+				p = &(*p)->rb_left;
+			else
+				p = &(*p)->rb_right;
+			break;
+		case CHAIN_NONE:
+		default:
+			break;
+		}
+	}
+
+	rb_link_node(&chain->rb_node, parent, p);
+	rb_insert_color(&chain->rb_node, root);
+}
+
+static void
+__sort_chain_flat(struct rb_root *rb_root, struct callchain_node *node,
+		  u64 min_hit)
+{
+	struct rb_node *n;
+	struct callchain_node *child;
+
+	n = rb_first(&node->rb_root_in);
+	while (n) {
+		child = rb_entry(n, struct callchain_node, rb_node_in);
+		n = rb_next(n);
+
+		__sort_chain_flat(rb_root, child, min_hit);
+	}
+
+	if (node->hit && node->hit >= min_hit)
+		rb_insert_callchain(rb_root, node, CHAIN_FLAT);
+}
+
+/*
+ * Once we get every callchains from the stream, we can now
+ * sort them by hit
+ */
+static void
+sort_chain_flat(struct rb_root *rb_root, struct callchain_root *root,
+		u64 min_hit, struct callchain_param *param __maybe_unused)
+{
+	__sort_chain_flat(rb_root, &root->node, min_hit);
+}
+
+static void __sort_chain_graph_abs(struct callchain_node *node,
+				   u64 min_hit)
+{
+	struct rb_node *n;
+	struct callchain_node *child;
+
+	node->rb_root = RB_ROOT;
+	n = rb_first(&node->rb_root_in);
+
+	while (n) {
+		child = rb_entry(n, struct callchain_node, rb_node_in);
+		n = rb_next(n);
+
+		__sort_chain_graph_abs(child, min_hit);
+		if (callchain_cumul_hits(child) >= min_hit)
+			rb_insert_callchain(&node->rb_root, child,
+					    CHAIN_GRAPH_ABS);
+	}
+}
+
+static void
+sort_chain_graph_abs(struct rb_root *rb_root, struct callchain_root *chain_root,
+		     u64 min_hit, struct callchain_param *param __maybe_unused)
+{
+	__sort_chain_graph_abs(&chain_root->node, min_hit);
+	rb_root->rb_node = chain_root->node.rb_root.rb_node;
+}
+
+static void __sort_chain_graph_rel(struct callchain_node *node,
+				   double min_percent)
+{
+	struct rb_node *n;
+	struct callchain_node *child;
+	u64 min_hit;
+
+	node->rb_root = RB_ROOT;
+	min_hit = ceil(node->children_hit * min_percent);
+
+	n = rb_first(&node->rb_root_in);
+	while (n) {
+		child = rb_entry(n, struct callchain_node, rb_node_in);
+		n = rb_next(n);
+
+		__sort_chain_graph_rel(child, min_percent);
+		if (callchain_cumul_hits(child) >= min_hit)
+			rb_insert_callchain(&node->rb_root, child,
+					    CHAIN_GRAPH_REL);
+	}
+}
+
+static void
+sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_root *chain_root,
+		     u64 min_hit __maybe_unused, struct callchain_param *param)
+{
+	__sort_chain_graph_rel(&chain_root->node, param->min_percent / 100.0);
+	rb_root->rb_node = chain_root->node.rb_root.rb_node;
+}
+
+int callchain_register_param(struct callchain_param *param)
+{
+	switch (param->mode) {
+	case CHAIN_GRAPH_ABS:
+		param->sort = sort_chain_graph_abs;
+		break;
+	case CHAIN_GRAPH_REL:
+		param->sort = sort_chain_graph_rel;
+		break;
+	case CHAIN_FLAT:
+		param->sort = sort_chain_flat;
+		break;
+	case CHAIN_NONE:
+	default:
+		return -1;
+	}
+	return 0;
+}
+
+/*
+ * Create a child for a parent. If inherit_children, then the new child
+ * will become the new parent of it's parent children
+ */
+static struct callchain_node *
+create_child(struct callchain_node *parent, bool inherit_children)
+{
+	struct callchain_node *new;
+
+	new = zalloc(sizeof(*new));
+	if (!new) {
+		perror("not enough memory to create child for code path tree");
+		return NULL;
+	}
+	new->parent = parent;
+	INIT_LIST_HEAD(&new->val);
+
+	if (inherit_children) {
+		struct rb_node *n;
+		struct callchain_node *child;
+
+		new->rb_root_in = parent->rb_root_in;
+		parent->rb_root_in = RB_ROOT;
+
+		n = rb_first(&new->rb_root_in);
+		while (n) {
+			child = rb_entry(n, struct callchain_node, rb_node_in);
+			child->parent = new;
+			n = rb_next(n);
+		}
+
+		/* make it the first child */
+		rb_link_node(&new->rb_node_in, NULL, &parent->rb_root_in.rb_node);
+		rb_insert_color(&new->rb_node_in, &parent->rb_root_in);
+	}
+
+	return new;
+}
+
+
+/*
+ * Fill the node with callchain values
+ */
+static void
+fill_node(struct callchain_node *node, struct callchain_cursor *cursor)
+{
+	struct callchain_cursor_node *cursor_node;
+
+	node->val_nr = cursor->nr - cursor->pos;
+	if (!node->val_nr)
+		pr_warning("Warning: empty node in callchain tree\n");
+
+	cursor_node = callchain_cursor_current(cursor);
+
+	while (cursor_node) {
+		struct callchain_list *call;
+
+		call = zalloc(sizeof(*call));
+		if (!call) {
+			perror("not enough memory for the code path tree");
+			return;
+		}
+		call->ip = cursor_node->ip;
+		call->ms.sym = cursor_node->sym;
+		call->ms.map = cursor_node->map;
+		list_add_tail(&call->list, &node->val);
+
+		callchain_cursor_advance(cursor);
+		cursor_node = callchain_cursor_current(cursor);
+	}
+}
+
+static struct callchain_node *
+add_child(struct callchain_node *parent,
+	  struct callchain_cursor *cursor,
+	  u64 period)
+{
+	struct callchain_node *new;
+
+	new = create_child(parent, false);
+	fill_node(new, cursor);
+
+	new->children_hit = 0;
+	new->hit = period;
+	return new;
+}
+
+static s64 match_chain(struct callchain_cursor_node *node,
+		      struct callchain_list *cnode)
+{
+	struct symbol *sym = node->sym;
+
+	if (cnode->ms.sym && sym &&
+	    callchain_param.key == CCKEY_FUNCTION)
+		return cnode->ms.sym->start - sym->start;
+	else
+		return cnode->ip - node->ip;
+}
+
+/*
+ * Split the parent in two parts (a new child is created) and
+ * give a part of its callchain to the created child.
+ * Then create another child to host the given callchain of new branch
+ */
+static void
+split_add_child(struct callchain_node *parent,
+		struct callchain_cursor *cursor,
+		struct callchain_list *to_split,
+		u64 idx_parents, u64 idx_local, u64 period)
+{
+	struct callchain_node *new;
+	struct list_head *old_tail;
+	unsigned int idx_total = idx_parents + idx_local;
+
+	/* split */
+	new = create_child(parent, true);
+
+	/* split the callchain and move a part to the new child */
+	old_tail = parent->val.prev;
+	list_del_range(&to_split->list, old_tail);
+	new->val.next = &to_split->list;
+	new->val.prev = old_tail;
+	to_split->list.prev = &new->val;
+	old_tail->next = &new->val;
+
+	/* split the hits */
+	new->hit = parent->hit;
+	new->children_hit = parent->children_hit;
+	parent->children_hit = callchain_cumul_hits(new);
+	new->val_nr = parent->val_nr - idx_local;
+	parent->val_nr = idx_local;
+
+	/* create a new child for the new branch if any */
+	if (idx_total < cursor->nr) {
+		struct callchain_node *first;
+		struct callchain_list *cnode;
+		struct callchain_cursor_node *node;
+		struct rb_node *p, **pp;
+
+		parent->hit = 0;
+		parent->children_hit += period;
+
+		node = callchain_cursor_current(cursor);
+		new = add_child(parent, cursor, period);
+
+		/*
+		 * This is second child since we moved parent's children
+		 * to new (first) child above.
+		 */
+		p = parent->rb_root_in.rb_node;
+		first = rb_entry(p, struct callchain_node, rb_node_in);
+		cnode = list_first_entry(&first->val, struct callchain_list,
+					 list);
+
+		if (match_chain(node, cnode) < 0)
+			pp = &p->rb_left;
+		else
+			pp = &p->rb_right;
+
+		rb_link_node(&new->rb_node_in, p, pp);
+		rb_insert_color(&new->rb_node_in, &parent->rb_root_in);
+	} else {
+		parent->hit = period;
+	}
+}
+
+static int
+append_chain(struct callchain_node *root,
+	     struct callchain_cursor *cursor,
+	     u64 period);
+
+static void
+append_chain_children(struct callchain_node *root,
+		      struct callchain_cursor *cursor,
+		      u64 period)
+{
+	struct callchain_node *rnode;
+	struct callchain_cursor_node *node;
+	struct rb_node **p = &root->rb_root_in.rb_node;
+	struct rb_node *parent = NULL;
+
+	node = callchain_cursor_current(cursor);
+	if (!node)
+		return;
+
+	/* lookup in childrens */
+	while (*p) {
+		s64 ret;
+
+		parent = *p;
+		rnode = rb_entry(parent, struct callchain_node, rb_node_in);
+
+		/* If at least first entry matches, rely to children */
+		ret = append_chain(rnode, cursor, period);
+		if (ret == 0)
+			goto inc_children_hit;
+
+		if (ret < 0)
+			p = &parent->rb_left;
+		else
+			p = &parent->rb_right;
+	}
+	/* nothing in children, add to the current node */
+	rnode = add_child(root, cursor, period);
+	rb_link_node(&rnode->rb_node_in, parent, p);
+	rb_insert_color(&rnode->rb_node_in, &root->rb_root_in);
+
+inc_children_hit:
+	root->children_hit += period;
+}
+
+static int
+append_chain(struct callchain_node *root,
+	     struct callchain_cursor *cursor,
+	     u64 period)
+{
+	struct callchain_list *cnode;
+	u64 start = cursor->pos;
+	bool found = false;
+	u64 matches;
+	int cmp = 0;
+
+	/*
+	 * Lookup in the current node
+	 * If we have a symbol, then compare the start to match
+	 * anywhere inside a function, unless function
+	 * mode is disabled.
+	 */
+	list_for_each_entry(cnode, &root->val, list) {
+		struct callchain_cursor_node *node;
+
+		node = callchain_cursor_current(cursor);
+		if (!node)
+			break;
+
+		cmp = match_chain(node, cnode);
+		if (cmp)
+			break;
+
+		found = true;
+
+		callchain_cursor_advance(cursor);
+	}
+
+	/* matches not, relay no the parent */
+	if (!found) {
+		WARN_ONCE(!cmp, "Chain comparison error\n");
+		return cmp;
+	}
+
+	matches = cursor->pos - start;
+
+	/* we match only a part of the node. Split it and add the new chain */
+	if (matches < root->val_nr) {
+		split_add_child(root, cursor, cnode, start, matches, period);
+		return 0;
+	}
+
+	/* we match 100% of the path, increment the hit */
+	if (matches == root->val_nr && cursor->pos == cursor->nr) {
+		root->hit += period;
+		return 0;
+	}
+
+	/* We match the node and still have a part remaining */
+	append_chain_children(root, cursor, period);
+
+	return 0;
+}
+
+int callchain_append(struct callchain_root *root,
+		     struct callchain_cursor *cursor,
+		     u64 period)
+{
+	if (!cursor->nr)
+		return 0;
+
+	callchain_cursor_commit(cursor);
+
+	append_chain_children(&root->node, cursor, period);
+
+	if (cursor->nr > root->max_depth)
+		root->max_depth = cursor->nr;
+
+	return 0;
+}
+
+static int
+merge_chain_branch(struct callchain_cursor *cursor,
+		   struct callchain_node *dst, struct callchain_node *src)
+{
+	struct callchain_cursor_node **old_last = cursor->last;
+	struct callchain_node *child;
+	struct callchain_list *list, *next_list;
+	struct rb_node *n;
+	int old_pos = cursor->nr;
+	int err = 0;
+
+	list_for_each_entry_safe(list, next_list, &src->val, list) {
+		callchain_cursor_append(cursor, list->ip,
+					list->ms.map, list->ms.sym);
+		list_del(&list->list);
+		free(list);
+	}
+
+	if (src->hit) {
+		callchain_cursor_commit(cursor);
+		append_chain_children(dst, cursor, src->hit);
+	}
+
+	n = rb_first(&src->rb_root_in);
+	while (n) {
+		child = container_of(n, struct callchain_node, rb_node_in);
+		n = rb_next(n);
+		rb_erase(&child->rb_node_in, &src->rb_root_in);
+
+		err = merge_chain_branch(cursor, dst, child);
+		if (err)
+			break;
+
+		free(child);
+	}
+
+	cursor->nr = old_pos;
+	cursor->last = old_last;
+
+	return err;
+}
+
+int callchain_merge(struct callchain_cursor *cursor,
+		    struct callchain_root *dst, struct callchain_root *src)
+{
+	return merge_chain_branch(cursor, &dst->node, &src->node);
+}
+
+int callchain_cursor_append(struct callchain_cursor *cursor,
+			    u64 ip, struct map *map, struct symbol *sym)
+{
+	struct callchain_cursor_node *node = *cursor->last;
+
+	if (!node) {
+		node = calloc(1, sizeof(*node));
+		if (!node)
+			return -ENOMEM;
+
+		*cursor->last = node;
+	}
+
+	node->ip = ip;
+	node->map = map;
+	node->sym = sym;
+
+	cursor->nr++;
+
+	cursor->last = &node->next;
+
+	return 0;
+}
+
+int sample__resolve_callchain(struct perf_sample *sample, struct symbol **parent,
+			      struct perf_evsel *evsel, struct addr_location *al,
+			      int max_stack)
+{
+	if (sample->callchain == NULL)
+		return 0;
+
+	if (symbol_conf.use_callchain || symbol_conf.cumulate_callchain ||
+	    sort__has_parent) {
+		return thread__resolve_callchain(al->thread, evsel, sample,
+						 parent, al, max_stack);
+	}
+	return 0;
+}
+
+int hist_entry__append_callchain(struct hist_entry *he, struct perf_sample *sample)
+{
+	if (!symbol_conf.use_callchain || sample->callchain == NULL)
+		return 0;
+	return callchain_append(he->callchain, &callchain_cursor, sample->period);
+}
+
+int fill_callchain_info(struct addr_location *al, struct callchain_cursor_node *node,
+			bool hide_unresolved)
+{
+	al->map = node->map;
+	al->sym = node->sym;
+	if (node->map)
+		al->addr = node->map->map_ip(node->map, node->ip);
+	else
+		al->addr = node->ip;
+
+	if (al->sym == NULL) {
+		if (hide_unresolved)
+			return 0;
+		if (al->map == NULL)
+			goto out;
+	}
+
+	if (al->map->groups == &al->machine->kmaps) {
+		if (machine__is_host(al->machine)) {
+			al->cpumode = PERF_RECORD_MISC_KERNEL;
+			al->level = 'k';
+		} else {
+			al->cpumode = PERF_RECORD_MISC_GUEST_KERNEL;
+			al->level = 'g';
+		}
+	} else {
+		if (machine__is_host(al->machine)) {
+			al->cpumode = PERF_RECORD_MISC_USER;
+			al->level = '.';
+		} else if (perf_guest) {
+			al->cpumode = PERF_RECORD_MISC_GUEST_USER;
+			al->level = 'u';
+		} else {
+			al->cpumode = PERF_RECORD_MISC_HYPERVISOR;
+			al->level = 'H';
+		}
+	}
+
+out:
+	return 1;
+}
+
+char *callchain_list__sym_name(struct callchain_list *cl,
+			       char *bf, size_t bfsize, bool show_dso)
+{
+	int printed;
+
+	if (cl->ms.sym) {
+		if (callchain_param.key == CCKEY_ADDRESS &&
+		    cl->ms.map && !cl->srcline)
+			cl->srcline = get_srcline(cl->ms.map->dso,
+						  map__rip_2objdump(cl->ms.map,
+								    cl->ip),
+						  cl->ms.sym, false);
+		if (cl->srcline)
+			printed = scnprintf(bf, bfsize, "%s %s",
+					cl->ms.sym->name, cl->srcline);
+		else
+			printed = scnprintf(bf, bfsize, "%s", cl->ms.sym->name);
+	} else
+		printed = scnprintf(bf, bfsize, "%#" PRIx64, cl->ip);
+
+	if (show_dso)
+		scnprintf(bf + printed, bfsize - printed, " %s",
+			  cl->ms.map ?
+			  cl->ms.map->dso->short_name :
+			  "unknown");
+
+	return bf;
+}
+
+static void free_callchain_node(struct callchain_node *node)
+{
+	struct callchain_list *list, *tmp;
+	struct callchain_node *child;
+	struct rb_node *n;
+
+	list_for_each_entry_safe(list, tmp, &node->val, list) {
+		list_del(&list->list);
+		free(list);
+	}
+
+	n = rb_first(&node->rb_root_in);
+	while (n) {
+		child = container_of(n, struct callchain_node, rb_node_in);
+		n = rb_next(n);
+		rb_erase(&child->rb_node_in, &node->rb_root_in);
+
+		free_callchain_node(child);
+		free(child);
+	}
+}
+
+void free_callchain(struct callchain_root *root)
+{
+	if (!symbol_conf.use_callchain)
+		return;
+
+	free_callchain_node(&root->node);
+}
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
new file mode 100644
index 000000000..6033a0a21
--- /dev/null
+++ b/tools/perf/util/callchain.h
@@ -0,0 +1,204 @@
+#ifndef __PERF_CALLCHAIN_H
+#define __PERF_CALLCHAIN_H
+
+#include "../perf.h"
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include "event.h"
+#include "symbol.h"
+
+enum perf_call_graph_mode {
+	CALLCHAIN_NONE,
+	CALLCHAIN_FP,
+	CALLCHAIN_DWARF,
+	CALLCHAIN_LBR,
+	CALLCHAIN_MAX
+};
+
+enum chain_mode {
+	CHAIN_NONE,
+	CHAIN_FLAT,
+	CHAIN_GRAPH_ABS,
+	CHAIN_GRAPH_REL
+};
+
+enum chain_order {
+	ORDER_CALLER,
+	ORDER_CALLEE
+};
+
+struct callchain_node {
+	struct callchain_node	*parent;
+	struct list_head	val;
+	struct rb_node		rb_node_in; /* to insert nodes in an rbtree */
+	struct rb_node		rb_node;    /* to sort nodes in an output tree */
+	struct rb_root		rb_root_in; /* input tree of children */
+	struct rb_root		rb_root;    /* sorted output tree of children */
+	unsigned int		val_nr;
+	u64			hit;
+	u64			children_hit;
+};
+
+struct callchain_root {
+	u64			max_depth;
+	struct callchain_node	node;
+};
+
+struct callchain_param;
+
+typedef void (*sort_chain_func_t)(struct rb_root *, struct callchain_root *,
+				 u64, struct callchain_param *);
+
+enum chain_key {
+	CCKEY_FUNCTION,
+	CCKEY_ADDRESS
+};
+
+struct callchain_param {
+	bool			enabled;
+	enum perf_call_graph_mode record_mode;
+	u32			dump_size;
+	enum chain_mode 	mode;
+	u32			print_limit;
+	double			min_percent;
+	sort_chain_func_t	sort;
+	enum chain_order	order;
+	enum chain_key		key;
+	bool			branch_callstack;
+};
+
+extern struct callchain_param callchain_param;
+
+struct callchain_list {
+	u64			ip;
+	struct map_symbol	ms;
+	char		       *srcline;
+	struct list_head	list;
+};
+
+/*
+ * A callchain cursor is a single linked list that
+ * let one feed a callchain progressively.
+ * It keeps persistent allocated entries to minimize
+ * allocations.
+ */
+struct callchain_cursor_node {
+	u64				ip;
+	struct map			*map;
+	struct symbol			*sym;
+	struct callchain_cursor_node	*next;
+};
+
+struct callchain_cursor {
+	u64				nr;
+	struct callchain_cursor_node	*first;
+	struct callchain_cursor_node	**last;
+	u64				pos;
+	struct callchain_cursor_node	*curr;
+};
+
+extern __thread struct callchain_cursor callchain_cursor;
+
+static inline void callchain_init(struct callchain_root *root)
+{
+	INIT_LIST_HEAD(&root->node.val);
+
+	root->node.parent = NULL;
+	root->node.hit = 0;
+	root->node.children_hit = 0;
+	root->node.rb_root_in = RB_ROOT;
+	root->max_depth = 0;
+}
+
+static inline u64 callchain_cumul_hits(struct callchain_node *node)
+{
+	return node->hit + node->children_hit;
+}
+
+int callchain_register_param(struct callchain_param *param);
+int callchain_append(struct callchain_root *root,
+		     struct callchain_cursor *cursor,
+		     u64 period);
+
+int callchain_merge(struct callchain_cursor *cursor,
+		    struct callchain_root *dst, struct callchain_root *src);
+
+/*
+ * Initialize a cursor before adding entries inside, but keep
+ * the previously allocated entries as a cache.
+ */
+static inline void callchain_cursor_reset(struct callchain_cursor *cursor)
+{
+	cursor->nr = 0;
+	cursor->last = &cursor->first;
+}
+
+int callchain_cursor_append(struct callchain_cursor *cursor, u64 ip,
+			    struct map *map, struct symbol *sym);
+
+/* Close a cursor writing session. Initialize for the reader */
+static inline void callchain_cursor_commit(struct callchain_cursor *cursor)
+{
+	cursor->curr = cursor->first;
+	cursor->pos = 0;
+}
+
+/* Cursor reading iteration helpers */
+static inline struct callchain_cursor_node *
+callchain_cursor_current(struct callchain_cursor *cursor)
+{
+	if (cursor->pos == cursor->nr)
+		return NULL;
+
+	return cursor->curr;
+}
+
+static inline void callchain_cursor_advance(struct callchain_cursor *cursor)
+{
+	cursor->curr = cursor->curr->next;
+	cursor->pos++;
+}
+
+struct option;
+struct hist_entry;
+
+int record_parse_callchain_opt(const struct option *opt, const char *arg, int unset);
+int record_callchain_opt(const struct option *opt, const char *arg, int unset);
+
+int sample__resolve_callchain(struct perf_sample *sample, struct symbol **parent,
+			      struct perf_evsel *evsel, struct addr_location *al,
+			      int max_stack);
+int hist_entry__append_callchain(struct hist_entry *he, struct perf_sample *sample);
+int fill_callchain_info(struct addr_location *al, struct callchain_cursor_node *node,
+			bool hide_unresolved);
+
+extern const char record_callchain_help[];
+int parse_callchain_record_opt(const char *arg);
+int parse_callchain_report_opt(const char *arg);
+int perf_callchain_config(const char *var, const char *value);
+
+static inline void callchain_cursor_snapshot(struct callchain_cursor *dest,
+					     struct callchain_cursor *src)
+{
+	*dest = *src;
+
+	dest->first = src->curr;
+	dest->nr -= src->pos;
+}
+
+#ifdef HAVE_SKIP_CALLCHAIN_IDX
+extern int arch_skip_callchain_idx(struct thread *thread, struct ip_callchain *chain);
+#else
+static inline int arch_skip_callchain_idx(struct thread *thread __maybe_unused,
+			struct ip_callchain *chain __maybe_unused)
+{
+	return -1;
+}
+#endif
+
+char *callchain_list__sym_name(struct callchain_list *cl,
+			       char *bf, size_t bfsize, bool show_dso);
+
+void free_callchain(struct callchain_root *root);
+
+#endif	/* __PERF_CALLCHAIN_H */
diff --git a/tools/perf/util/cgroup.c b/tools/perf/util/cgroup.c
new file mode 100644
index 000000000..88f7be399
--- /dev/null
+++ b/tools/perf/util/cgroup.c
@@ -0,0 +1,177 @@
+#include "util.h"
+#include "../perf.h"
+#include "parse-options.h"
+#include "evsel.h"
+#include "cgroup.h"
+#include "evlist.h"
+
+int nr_cgroups;
+
+static int
+cgroupfs_find_mountpoint(char *buf, size_t maxlen)
+{
+	FILE *fp;
+	char mountpoint[PATH_MAX + 1], tokens[PATH_MAX + 1], type[PATH_MAX + 1];
+	char *token, *saved_ptr = NULL;
+	int found = 0;
+
+	fp = fopen("/proc/mounts", "r");
+	if (!fp)
+		return -1;
+
+	/*
+	 * in order to handle split hierarchy, we need to scan /proc/mounts
+	 * and inspect every cgroupfs mount point to find one that has
+	 * perf_event subsystem
+	 */
+	while (fscanf(fp, "%*s %"STR(PATH_MAX)"s %"STR(PATH_MAX)"s %"
+				STR(PATH_MAX)"s %*d %*d\n",
+				mountpoint, type, tokens) == 3) {
+
+		if (!strcmp(type, "cgroup")) {
+
+			token = strtok_r(tokens, ",", &saved_ptr);
+
+			while (token != NULL) {
+				if (!strcmp(token, "perf_event")) {
+					found = 1;
+					break;
+				}
+				token = strtok_r(NULL, ",", &saved_ptr);
+			}
+		}
+		if (found)
+			break;
+	}
+	fclose(fp);
+	if (!found)
+		return -1;
+
+	if (strlen(mountpoint) < maxlen) {
+		strcpy(buf, mountpoint);
+		return 0;
+	}
+	return -1;
+}
+
+static int open_cgroup(char *name)
+{
+	char path[PATH_MAX + 1];
+	char mnt[PATH_MAX + 1];
+	int fd;
+
+
+	if (cgroupfs_find_mountpoint(mnt, PATH_MAX + 1))
+		return -1;
+
+	snprintf(path, PATH_MAX, "%s/%s", mnt, name);
+
+	fd = open(path, O_RDONLY);
+	if (fd == -1)
+		fprintf(stderr, "no access to cgroup %s\n", path);
+
+	return fd;
+}
+
+static int add_cgroup(struct perf_evlist *evlist, char *str)
+{
+	struct perf_evsel *counter;
+	struct cgroup_sel *cgrp = NULL;
+	int n;
+	/*
+	 * check if cgrp is already defined, if so we reuse it
+	 */
+	evlist__for_each(evlist, counter) {
+		cgrp = counter->cgrp;
+		if (!cgrp)
+			continue;
+		if (!strcmp(cgrp->name, str))
+			break;
+
+		cgrp = NULL;
+	}
+
+	if (!cgrp) {
+		cgrp = zalloc(sizeof(*cgrp));
+		if (!cgrp)
+			return -1;
+
+		cgrp->name = str;
+
+		cgrp->fd = open_cgroup(str);
+		if (cgrp->fd == -1) {
+			free(cgrp);
+			return -1;
+		}
+	}
+
+	/*
+	 * find corresponding event
+	 * if add cgroup N, then need to find event N
+	 */
+	n = 0;
+	evlist__for_each(evlist, counter) {
+		if (n == nr_cgroups)
+			goto found;
+		n++;
+	}
+	if (cgrp->refcnt == 0)
+		free(cgrp);
+
+	return -1;
+found:
+	cgrp->refcnt++;
+	counter->cgrp = cgrp;
+	return 0;
+}
+
+void close_cgroup(struct cgroup_sel *cgrp)
+{
+	if (!cgrp)
+		return;
+
+	/* XXX: not reentrant */
+	if (--cgrp->refcnt == 0) {
+		close(cgrp->fd);
+		zfree(&cgrp->name);
+		free(cgrp);
+	}
+}
+
+int parse_cgroups(const struct option *opt __maybe_unused, const char *str,
+		  int unset __maybe_unused)
+{
+	struct perf_evlist *evlist = *(struct perf_evlist **)opt->value;
+	const char *p, *e, *eos = str + strlen(str);
+	char *s;
+	int ret;
+
+	if (list_empty(&evlist->entries)) {
+		fprintf(stderr, "must define events before cgroups\n");
+		return -1;
+	}
+
+	for (;;) {
+		p = strchr(str, ',');
+		e = p ? p : eos;
+
+		/* allow empty cgroups, i.e., skip */
+		if (e - str) {
+			/* termination added */
+			s = strndup(str, e - str);
+			if (!s)
+				return -1;
+			ret = add_cgroup(evlist, s);
+			if (ret) {
+				free(s);
+				return -1;
+			}
+		}
+		/* nr_cgroups is increased een for empty cgroups */
+		nr_cgroups++;
+		if (!p)
+			break;
+		str = p+1;
+	}
+	return 0;
+}
diff --git a/tools/perf/util/cgroup.h b/tools/perf/util/cgroup.h
new file mode 100644
index 000000000..89acd6deb
--- /dev/null
+++ b/tools/perf/util/cgroup.h
@@ -0,0 +1,17 @@
+#ifndef __CGROUP_H__
+#define __CGROUP_H__
+
+struct option;
+
+struct cgroup_sel {
+	char *name;
+	int fd;
+	int refcnt;
+};
+
+
+extern int nr_cgroups; /* number of explicit cgroups defined */
+extern void close_cgroup(struct cgroup_sel *cgrp);
+extern int parse_cgroups(const struct option *opt, const char *str, int unset);
+
+#endif /* __CGROUP_H__ */
diff --git a/tools/perf/util/cloexec.c b/tools/perf/util/cloexec.c
new file mode 100644
index 000000000..2babddaa2
--- /dev/null
+++ b/tools/perf/util/cloexec.c
@@ -0,0 +1,96 @@
+#include <sched.h>
+#include "util.h"
+#include "../perf.h"
+#include "cloexec.h"
+#include "asm/bug.h"
+#include "debug.h"
+
+static unsigned long flag = PERF_FLAG_FD_CLOEXEC;
+
+#ifdef __GLIBC_PREREQ
+#if !__GLIBC_PREREQ(2, 6)
+int __weak sched_getcpu(void)
+{
+	errno = ENOSYS;
+	return -1;
+}
+#endif
+#endif
+
+static int perf_flag_probe(void)
+{
+	/* use 'safest' configuration as used in perf_evsel__fallback() */
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_SOFTWARE,
+		.config = PERF_COUNT_SW_CPU_CLOCK,
+		.exclude_kernel = 1,
+	};
+	int fd;
+	int err;
+	int cpu;
+	pid_t pid = -1;
+	char sbuf[STRERR_BUFSIZE];
+
+	cpu = sched_getcpu();
+	if (cpu < 0)
+		cpu = 0;
+
+	/*
+	 * Using -1 for the pid is a workaround to avoid gratuitous jump label
+	 * changes.
+	 */
+	while (1) {
+		/* check cloexec flag */
+		fd = sys_perf_event_open(&attr, pid, cpu, -1,
+					 PERF_FLAG_FD_CLOEXEC);
+		if (fd < 0 && pid == -1 && errno == EACCES) {
+			pid = 0;
+			continue;
+		}
+		break;
+	}
+	err = errno;
+
+	if (fd >= 0) {
+		close(fd);
+		return 1;
+	}
+
+	WARN_ONCE(err != EINVAL && err != EBUSY,
+		  "perf_event_open(..., PERF_FLAG_FD_CLOEXEC) failed with unexpected error %d (%s)\n",
+		  err, strerror_r(err, sbuf, sizeof(sbuf)));
+
+	/* not supported, confirm error related to PERF_FLAG_FD_CLOEXEC */
+	while (1) {
+		fd = sys_perf_event_open(&attr, pid, cpu, -1, 0);
+		if (fd < 0 && pid == -1 && errno == EACCES) {
+			pid = 0;
+			continue;
+		}
+		break;
+	}
+	err = errno;
+
+	if (fd >= 0)
+		close(fd);
+
+	if (WARN_ONCE(fd < 0 && err != EBUSY,
+		      "perf_event_open(..., 0) failed unexpectedly with error %d (%s)\n",
+		      err, strerror_r(err, sbuf, sizeof(sbuf))))
+		return -1;
+
+	return 0;
+}
+
+unsigned long perf_event_open_cloexec_flag(void)
+{
+	static bool probed;
+
+	if (!probed) {
+		if (perf_flag_probe() <= 0)
+			flag = 0;
+		probed = true;
+	}
+
+	return flag;
+}
diff --git a/tools/perf/util/cloexec.h b/tools/perf/util/cloexec.h
new file mode 100644
index 000000000..68888c29b
--- /dev/null
+++ b/tools/perf/util/cloexec.h
@@ -0,0 +1,12 @@
+#ifndef __PERF_CLOEXEC_H
+#define __PERF_CLOEXEC_H
+
+unsigned long perf_event_open_cloexec_flag(void);
+
+#ifdef __GLIBC_PREREQ
+#if !__GLIBC_PREREQ(2, 6)
+extern int sched_getcpu(void) __THROW;
+#endif
+#endif
+
+#endif /* __PERF_CLOEXEC_H */
diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
new file mode 100644
index 000000000..55355b3d4
--- /dev/null
+++ b/tools/perf/util/color.c
@@ -0,0 +1,227 @@
+#include <linux/kernel.h>
+#include "cache.h"
+#include "color.h"
+#include <math.h>
+
+int perf_use_color_default = -1;
+
+int perf_config_colorbool(const char *var, const char *value, int stdout_is_tty)
+{
+	if (value) {
+		if (!strcasecmp(value, "never"))
+			return 0;
+		if (!strcasecmp(value, "always"))
+			return 1;
+		if (!strcasecmp(value, "auto"))
+			goto auto_color;
+	}
+
+	/* Missing or explicit false to turn off colorization */
+	if (!perf_config_bool(var, value))
+		return 0;
+
+	/* any normal truth value defaults to 'auto' */
+ auto_color:
+	if (stdout_is_tty < 0)
+		stdout_is_tty = isatty(1);
+	if (stdout_is_tty || (pager_in_use() && pager_use_color)) {
+		char *term = getenv("TERM");
+		if (term && strcmp(term, "dumb"))
+			return 1;
+	}
+	return 0;
+}
+
+int perf_color_default_config(const char *var, const char *value, void *cb)
+{
+	if (!strcmp(var, "color.ui")) {
+		perf_use_color_default = perf_config_colorbool(var, value, -1);
+		return 0;
+	}
+
+	return perf_default_config(var, value, cb);
+}
+
+static int __color_vsnprintf(char *bf, size_t size, const char *color,
+			     const char *fmt, va_list args, const char *trail)
+{
+	int r = 0;
+
+	/*
+	 * Auto-detect:
+	 */
+	if (perf_use_color_default < 0) {
+		if (isatty(1) || pager_in_use())
+			perf_use_color_default = 1;
+		else
+			perf_use_color_default = 0;
+	}
+
+	if (perf_use_color_default && *color)
+		r += scnprintf(bf, size, "%s", color);
+	r += vscnprintf(bf + r, size - r, fmt, args);
+	if (perf_use_color_default && *color)
+		r += scnprintf(bf + r, size - r, "%s", PERF_COLOR_RESET);
+	if (trail)
+		r += scnprintf(bf + r, size - r, "%s", trail);
+	return r;
+}
+
+static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
+		va_list args, const char *trail)
+{
+	int r = 0;
+
+	/*
+	 * Auto-detect:
+	 */
+	if (perf_use_color_default < 0) {
+		if (isatty(fileno(fp)) || pager_in_use())
+			perf_use_color_default = 1;
+		else
+			perf_use_color_default = 0;
+	}
+
+	if (perf_use_color_default && *color)
+		r += fprintf(fp, "%s", color);
+	r += vfprintf(fp, fmt, args);
+	if (perf_use_color_default && *color)
+		r += fprintf(fp, "%s", PERF_COLOR_RESET);
+	if (trail)
+		r += fprintf(fp, "%s", trail);
+	return r;
+}
+
+int color_vsnprintf(char *bf, size_t size, const char *color,
+		    const char *fmt, va_list args)
+{
+	return __color_vsnprintf(bf, size, color, fmt, args, NULL);
+}
+
+int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args)
+{
+	return __color_vfprintf(fp, color, fmt, args, NULL);
+}
+
+int color_snprintf(char *bf, size_t size, const char *color,
+		   const char *fmt, ...)
+{
+	va_list args;
+	int r;
+
+	va_start(args, fmt);
+	r = color_vsnprintf(bf, size, color, fmt, args);
+	va_end(args);
+	return r;
+}
+
+int color_fprintf(FILE *fp, const char *color, const char *fmt, ...)
+{
+	va_list args;
+	int r;
+
+	va_start(args, fmt);
+	r = color_vfprintf(fp, color, fmt, args);
+	va_end(args);
+	return r;
+}
+
+int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...)
+{
+	va_list args;
+	int r;
+	va_start(args, fmt);
+	r = __color_vfprintf(fp, color, fmt, args, "\n");
+	va_end(args);
+	return r;
+}
+
+/*
+ * This function splits the buffer by newlines and colors the lines individually.
+ *
+ * Returns 0 on success.
+ */
+int color_fwrite_lines(FILE *fp, const char *color,
+		size_t count, const char *buf)
+{
+	if (!*color)
+		return fwrite(buf, count, 1, fp) != 1;
+
+	while (count) {
+		char *p = memchr(buf, '\n', count);
+
+		if (p != buf && (fputs(color, fp) < 0 ||
+				fwrite(buf, p ? (size_t)(p - buf) : count, 1, fp) != 1 ||
+				fputs(PERF_COLOR_RESET, fp) < 0))
+			return -1;
+		if (!p)
+			return 0;
+		if (fputc('\n', fp) < 0)
+			return -1;
+		count -= p + 1 - buf;
+		buf = p + 1;
+	}
+	return 0;
+}
+
+const char *get_percent_color(double percent)
+{
+	const char *color = PERF_COLOR_NORMAL;
+
+	/*
+	 * We color high-overhead entries in red, mid-overhead
+	 * entries in green - and keep the low overhead places
+	 * normal:
+	 */
+	if (fabs(percent) >= MIN_RED)
+		color = PERF_COLOR_RED;
+	else {
+		if (fabs(percent) > MIN_GREEN)
+			color = PERF_COLOR_GREEN;
+	}
+	return color;
+}
+
+int percent_color_fprintf(FILE *fp, const char *fmt, double percent)
+{
+	int r;
+	const char *color;
+
+	color = get_percent_color(percent);
+	r = color_fprintf(fp, color, fmt, percent);
+
+	return r;
+}
+
+int value_color_snprintf(char *bf, size_t size, const char *fmt, double value)
+{
+	const char *color = get_percent_color(value);
+	return color_snprintf(bf, size, color, fmt, value);
+}
+
+int percent_color_snprintf(char *bf, size_t size, const char *fmt, ...)
+{
+	va_list args;
+	double percent;
+
+	va_start(args, fmt);
+	percent = va_arg(args, double);
+	va_end(args);
+	return value_color_snprintf(bf, size, fmt, percent);
+}
+
+int percent_color_len_snprintf(char *bf, size_t size, const char *fmt, ...)
+{
+	va_list args;
+	int len;
+	double percent;
+	const char *color;
+
+	va_start(args, fmt);
+	len = va_arg(args, int);
+	percent = va_arg(args, double);
+	va_end(args);
+
+	color = get_percent_color(percent);
+	return color_snprintf(bf, size, color, fmt, len, percent);
+}
diff --git a/tools/perf/util/color.h b/tools/perf/util/color.h
new file mode 100644
index 000000000..38146f922
--- /dev/null
+++ b/tools/perf/util/color.h
@@ -0,0 +1,46 @@
+#ifndef __PERF_COLOR_H
+#define __PERF_COLOR_H
+
+/* "\033[1;38;5;2xx;48;5;2xxm\0" is 23 bytes */
+#define COLOR_MAXLEN 24
+
+#define PERF_COLOR_NORMAL	""
+#define PERF_COLOR_RESET	"\033[m"
+#define PERF_COLOR_BOLD		"\033[1m"
+#define PERF_COLOR_RED		"\033[31m"
+#define PERF_COLOR_GREEN	"\033[32m"
+#define PERF_COLOR_YELLOW	"\033[33m"
+#define PERF_COLOR_BLUE		"\033[34m"
+#define PERF_COLOR_MAGENTA	"\033[35m"
+#define PERF_COLOR_CYAN		"\033[36m"
+#define PERF_COLOR_BG_RED	"\033[41m"
+
+#define MIN_GREEN	0.5
+#define MIN_RED		5.0
+
+/*
+ * This variable stores the value of color.ui
+ */
+extern int perf_use_color_default;
+
+
+/*
+ * Use this instead of perf_default_config if you need the value of color.ui.
+ */
+int perf_color_default_config(const char *var, const char *value, void *cb);
+
+int perf_config_colorbool(const char *var, const char *value, int stdout_is_tty);
+int color_vsnprintf(char *bf, size_t size, const char *color,
+		    const char *fmt, va_list args);
+int color_vfprintf(FILE *fp, const char *color, const char *fmt, va_list args);
+int color_fprintf(FILE *fp, const char *color, const char *fmt, ...);
+int color_snprintf(char *bf, size_t size, const char *color, const char *fmt, ...);
+int color_fprintf_ln(FILE *fp, const char *color, const char *fmt, ...);
+int color_fwrite_lines(FILE *fp, const char *color, size_t count, const char *buf);
+int value_color_snprintf(char *bf, size_t size, const char *fmt, double value);
+int percent_color_snprintf(char *bf, size_t size, const char *fmt, ...);
+int percent_color_len_snprintf(char *bf, size_t size, const char *fmt, ...);
+int percent_color_fprintf(FILE *fp, const char *fmt, double percent);
+const char *get_percent_color(double percent);
+
+#endif /* __PERF_COLOR_H */
diff --git a/tools/perf/util/comm.c b/tools/perf/util/comm.c
new file mode 100644
index 000000000..b2bb59df6
--- /dev/null
+++ b/tools/perf/util/comm.c
@@ -0,0 +1,125 @@
+#include "comm.h"
+#include "util.h"
+#include <stdlib.h>
+#include <stdio.h>
+
+struct comm_str {
+	char *str;
+	struct rb_node rb_node;
+	int ref;
+};
+
+/* Should perhaps be moved to struct machine */
+static struct rb_root comm_str_root;
+
+static void comm_str__get(struct comm_str *cs)
+{
+	cs->ref++;
+}
+
+static void comm_str__put(struct comm_str *cs)
+{
+	if (!--cs->ref) {
+		rb_erase(&cs->rb_node, &comm_str_root);
+		zfree(&cs->str);
+		free(cs);
+	}
+}
+
+static struct comm_str *comm_str__alloc(const char *str)
+{
+	struct comm_str *cs;
+
+	cs = zalloc(sizeof(*cs));
+	if (!cs)
+		return NULL;
+
+	cs->str = strdup(str);
+	if (!cs->str) {
+		free(cs);
+		return NULL;
+	}
+
+	return cs;
+}
+
+static struct comm_str *comm_str__findnew(const char *str, struct rb_root *root)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct comm_str *iter, *new;
+	int cmp;
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct comm_str, rb_node);
+
+		cmp = strcmp(str, iter->str);
+		if (!cmp)
+			return iter;
+
+		if (cmp < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	new = comm_str__alloc(str);
+	if (!new)
+		return NULL;
+
+	rb_link_node(&new->rb_node, parent, p);
+	rb_insert_color(&new->rb_node, root);
+
+	return new;
+}
+
+struct comm *comm__new(const char *str, u64 timestamp, bool exec)
+{
+	struct comm *comm = zalloc(sizeof(*comm));
+
+	if (!comm)
+		return NULL;
+
+	comm->start = timestamp;
+	comm->exec = exec;
+
+	comm->comm_str = comm_str__findnew(str, &comm_str_root);
+	if (!comm->comm_str) {
+		free(comm);
+		return NULL;
+	}
+
+	comm_str__get(comm->comm_str);
+
+	return comm;
+}
+
+int comm__override(struct comm *comm, const char *str, u64 timestamp, bool exec)
+{
+	struct comm_str *new, *old = comm->comm_str;
+
+	new = comm_str__findnew(str, &comm_str_root);
+	if (!new)
+		return -ENOMEM;
+
+	comm_str__get(new);
+	comm_str__put(old);
+	comm->comm_str = new;
+	comm->start = timestamp;
+	if (exec)
+		comm->exec = true;
+
+	return 0;
+}
+
+void comm__free(struct comm *comm)
+{
+	comm_str__put(comm->comm_str);
+	free(comm);
+}
+
+const char *comm__str(const struct comm *comm)
+{
+	return comm->comm_str->str;
+}
diff --git a/tools/perf/util/comm.h b/tools/perf/util/comm.h
new file mode 100644
index 000000000..71c9c3934
--- /dev/null
+++ b/tools/perf/util/comm.h
@@ -0,0 +1,27 @@
+#ifndef __PERF_COMM_H
+#define __PERF_COMM_H
+
+#include "../perf.h"
+#include <linux/rbtree.h>
+#include <linux/list.h>
+
+struct comm_str;
+
+struct comm {
+	struct comm_str *comm_str;
+	u64 start;
+	struct list_head list;
+	bool exec;
+	union { /* Tool specific area */
+		void	*priv;
+		u64	db_id;
+	};
+};
+
+void comm__free(struct comm *comm);
+struct comm *comm__new(const char *str, u64 timestamp, bool exec);
+const char *comm__str(const struct comm *comm);
+int comm__override(struct comm *comm, const char *str, u64 timestamp,
+		   bool exec);
+
+#endif  /* __PERF_COMM_H */
diff --git a/tools/perf/util/config.c b/tools/perf/util/config.c
new file mode 100644
index 000000000..e18f653cd
--- /dev/null
+++ b/tools/perf/util/config.c
@@ -0,0 +1,564 @@
+/*
+ * config.c
+ *
+ * Helper functions for parsing config items.
+ * Originally copied from GIT source.
+ *
+ * Copyright (C) Linus Torvalds, 2005
+ * Copyright (C) Johannes Schindelin, 2005
+ *
+ */
+#include "util.h"
+#include "cache.h"
+#include "exec_cmd.h"
+#include "util/hist.h"  /* perf_hist_config */
+
+#define MAXNAME (256)
+
+#define DEBUG_CACHE_DIR ".debug"
+
+
+char buildid_dir[MAXPATHLEN]; /* root dir for buildid, binary cache */
+
+static FILE *config_file;
+static const char *config_file_name;
+static int config_linenr;
+static int config_file_eof;
+
+static const char *config_exclusive_filename;
+
+static int get_next_char(void)
+{
+	int c;
+	FILE *f;
+
+	c = '\n';
+	if ((f = config_file) != NULL) {
+		c = fgetc(f);
+		if (c == '\r') {
+			/* DOS like systems */
+			c = fgetc(f);
+			if (c != '\n') {
+				ungetc(c, f);
+				c = '\r';
+			}
+		}
+		if (c == '\n')
+			config_linenr++;
+		if (c == EOF) {
+			config_file_eof = 1;
+			c = '\n';
+		}
+	}
+	return c;
+}
+
+static char *parse_value(void)
+{
+	static char value[1024];
+	int quote = 0, comment = 0, space = 0;
+	size_t len = 0;
+
+	for (;;) {
+		int c = get_next_char();
+
+		if (len >= sizeof(value) - 1)
+			return NULL;
+		if (c == '\n') {
+			if (quote)
+				return NULL;
+			value[len] = 0;
+			return value;
+		}
+		if (comment)
+			continue;
+		if (isspace(c) && !quote) {
+			space = 1;
+			continue;
+		}
+		if (!quote) {
+			if (c == ';' || c == '#') {
+				comment = 1;
+				continue;
+			}
+		}
+		if (space) {
+			if (len)
+				value[len++] = ' ';
+			space = 0;
+		}
+		if (c == '\\') {
+			c = get_next_char();
+			switch (c) {
+			case '\n':
+				continue;
+			case 't':
+				c = '\t';
+				break;
+			case 'b':
+				c = '\b';
+				break;
+			case 'n':
+				c = '\n';
+				break;
+			/* Some characters escape as themselves */
+			case '\\': case '"':
+				break;
+			/* Reject unknown escape sequences */
+			default:
+				return NULL;
+			}
+			value[len++] = c;
+			continue;
+		}
+		if (c == '"') {
+			quote = 1-quote;
+			continue;
+		}
+		value[len++] = c;
+	}
+}
+
+static inline int iskeychar(int c)
+{
+	return isalnum(c) || c == '-' || c == '_';
+}
+
+static int get_value(config_fn_t fn, void *data, char *name, unsigned int len)
+{
+	int c;
+	char *value;
+
+	/* Get the full name */
+	for (;;) {
+		c = get_next_char();
+		if (config_file_eof)
+			break;
+		if (!iskeychar(c))
+			break;
+		name[len++] = c;
+		if (len >= MAXNAME)
+			return -1;
+	}
+	name[len] = 0;
+	while (c == ' ' || c == '\t')
+		c = get_next_char();
+
+	value = NULL;
+	if (c != '\n') {
+		if (c != '=')
+			return -1;
+		value = parse_value();
+		if (!value)
+			return -1;
+	}
+	return fn(name, value, data);
+}
+
+static int get_extended_base_var(char *name, int baselen, int c)
+{
+	do {
+		if (c == '\n')
+			return -1;
+		c = get_next_char();
+	} while (isspace(c));
+
+	/* We require the format to be '[base "extension"]' */
+	if (c != '"')
+		return -1;
+	name[baselen++] = '.';
+
+	for (;;) {
+		int ch = get_next_char();
+
+		if (ch == '\n')
+			return -1;
+		if (ch == '"')
+			break;
+		if (ch == '\\') {
+			ch = get_next_char();
+			if (ch == '\n')
+				return -1;
+		}
+		name[baselen++] = ch;
+		if (baselen > MAXNAME / 2)
+			return -1;
+	}
+
+	/* Final ']' */
+	if (get_next_char() != ']')
+		return -1;
+	return baselen;
+}
+
+static int get_base_var(char *name)
+{
+	int baselen = 0;
+
+	for (;;) {
+		int c = get_next_char();
+		if (config_file_eof)
+			return -1;
+		if (c == ']')
+			return baselen;
+		if (isspace(c))
+			return get_extended_base_var(name, baselen, c);
+		if (!iskeychar(c) && c != '.')
+			return -1;
+		if (baselen > MAXNAME / 2)
+			return -1;
+		name[baselen++] = tolower(c);
+	}
+}
+
+static int perf_parse_file(config_fn_t fn, void *data)
+{
+	int comment = 0;
+	int baselen = 0;
+	static char var[MAXNAME];
+
+	/* U+FEFF Byte Order Mark in UTF8 */
+	static const unsigned char *utf8_bom = (unsigned char *) "\xef\xbb\xbf";
+	const unsigned char *bomptr = utf8_bom;
+
+	for (;;) {
+		int line, c = get_next_char();
+
+		if (bomptr && *bomptr) {
+			/* We are at the file beginning; skip UTF8-encoded BOM
+			 * if present. Sane editors won't put this in on their
+			 * own, but e.g. Windows Notepad will do it happily. */
+			if ((unsigned char) c == *bomptr) {
+				bomptr++;
+				continue;
+			} else {
+				/* Do not tolerate partial BOM. */
+				if (bomptr != utf8_bom)
+					break;
+				/* No BOM at file beginning. Cool. */
+				bomptr = NULL;
+			}
+		}
+		if (c == '\n') {
+			if (config_file_eof)
+				return 0;
+			comment = 0;
+			continue;
+		}
+		if (comment || isspace(c))
+			continue;
+		if (c == '#' || c == ';') {
+			comment = 1;
+			continue;
+		}
+		if (c == '[') {
+			baselen = get_base_var(var);
+			if (baselen <= 0)
+				break;
+			var[baselen++] = '.';
+			var[baselen] = 0;
+			continue;
+		}
+		if (!isalpha(c))
+			break;
+		var[baselen] = tolower(c);
+
+		/*
+		 * The get_value function might or might not reach the '\n',
+		 * so saving the current line number for error reporting.
+		 */
+		line = config_linenr;
+		if (get_value(fn, data, var, baselen+1) < 0) {
+			config_linenr = line;
+			break;
+		}
+	}
+	die("bad config file line %d in %s", config_linenr, config_file_name);
+}
+
+static int parse_unit_factor(const char *end, unsigned long *val)
+{
+	if (!*end)
+		return 1;
+	else if (!strcasecmp(end, "k")) {
+		*val *= 1024;
+		return 1;
+	}
+	else if (!strcasecmp(end, "m")) {
+		*val *= 1024 * 1024;
+		return 1;
+	}
+	else if (!strcasecmp(end, "g")) {
+		*val *= 1024 * 1024 * 1024;
+		return 1;
+	}
+	return 0;
+}
+
+static int perf_parse_llong(const char *value, long long *ret)
+{
+	if (value && *value) {
+		char *end;
+		long long val = strtoll(value, &end, 0);
+		unsigned long factor = 1;
+
+		if (!parse_unit_factor(end, &factor))
+			return 0;
+		*ret = val * factor;
+		return 1;
+	}
+	return 0;
+}
+
+static int perf_parse_long(const char *value, long *ret)
+{
+	if (value && *value) {
+		char *end;
+		long val = strtol(value, &end, 0);
+		unsigned long factor = 1;
+		if (!parse_unit_factor(end, &factor))
+			return 0;
+		*ret = val * factor;
+		return 1;
+	}
+	return 0;
+}
+
+static void die_bad_config(const char *name)
+{
+	if (config_file_name)
+		die("bad config value for '%s' in %s", name, config_file_name);
+	die("bad config value for '%s'", name);
+}
+
+u64 perf_config_u64(const char *name, const char *value)
+{
+	long long ret = 0;
+
+	if (!perf_parse_llong(value, &ret))
+		die_bad_config(name);
+	return (u64) ret;
+}
+
+int perf_config_int(const char *name, const char *value)
+{
+	long ret = 0;
+	if (!perf_parse_long(value, &ret))
+		die_bad_config(name);
+	return ret;
+}
+
+static int perf_config_bool_or_int(const char *name, const char *value, int *is_bool)
+{
+	*is_bool = 1;
+	if (!value)
+		return 1;
+	if (!*value)
+		return 0;
+	if (!strcasecmp(value, "true") || !strcasecmp(value, "yes") || !strcasecmp(value, "on"))
+		return 1;
+	if (!strcasecmp(value, "false") || !strcasecmp(value, "no") || !strcasecmp(value, "off"))
+		return 0;
+	*is_bool = 0;
+	return perf_config_int(name, value);
+}
+
+int perf_config_bool(const char *name, const char *value)
+{
+	int discard;
+	return !!perf_config_bool_or_int(name, value, &discard);
+}
+
+const char *perf_config_dirname(const char *name, const char *value)
+{
+	if (!name)
+		return NULL;
+	return value;
+}
+
+static int perf_default_core_config(const char *var __maybe_unused,
+				    const char *value __maybe_unused)
+{
+	/* Add other config variables here. */
+	return 0;
+}
+
+static int perf_ui_config(const char *var, const char *value)
+{
+	/* Add other config variables here. */
+	if (!strcmp(var, "ui.show-headers")) {
+		symbol_conf.show_hist_headers = perf_config_bool(var, value);
+		return 0;
+	}
+	return 0;
+}
+
+int perf_default_config(const char *var, const char *value,
+			void *dummy __maybe_unused)
+{
+	if (!prefixcmp(var, "core."))
+		return perf_default_core_config(var, value);
+
+	if (!prefixcmp(var, "hist."))
+		return perf_hist_config(var, value);
+
+	if (!prefixcmp(var, "ui."))
+		return perf_ui_config(var, value);
+
+	if (!prefixcmp(var, "call-graph."))
+		return perf_callchain_config(var, value);
+
+	/* Add other config variables here. */
+	return 0;
+}
+
+static int perf_config_from_file(config_fn_t fn, const char *filename, void *data)
+{
+	int ret;
+	FILE *f = fopen(filename, "r");
+
+	ret = -1;
+	if (f) {
+		config_file = f;
+		config_file_name = filename;
+		config_linenr = 1;
+		config_file_eof = 0;
+		ret = perf_parse_file(fn, data);
+		fclose(f);
+		config_file_name = NULL;
+	}
+	return ret;
+}
+
+static const char *perf_etc_perfconfig(void)
+{
+	static const char *system_wide;
+	if (!system_wide)
+		system_wide = system_path(ETC_PERFCONFIG);
+	return system_wide;
+}
+
+static int perf_env_bool(const char *k, int def)
+{
+	const char *v = getenv(k);
+	return v ? perf_config_bool(k, v) : def;
+}
+
+static int perf_config_system(void)
+{
+	return !perf_env_bool("PERF_CONFIG_NOSYSTEM", 0);
+}
+
+static int perf_config_global(void)
+{
+	return !perf_env_bool("PERF_CONFIG_NOGLOBAL", 0);
+}
+
+int perf_config(config_fn_t fn, void *data)
+{
+	int ret = 0, found = 0;
+	const char *home = NULL;
+
+	/* Setting $PERF_CONFIG makes perf read _only_ the given config file. */
+	if (config_exclusive_filename)
+		return perf_config_from_file(fn, config_exclusive_filename, data);
+	if (perf_config_system() && !access(perf_etc_perfconfig(), R_OK)) {
+		ret += perf_config_from_file(fn, perf_etc_perfconfig(),
+					    data);
+		found += 1;
+	}
+
+	home = getenv("HOME");
+	if (perf_config_global() && home) {
+		char *user_config = strdup(mkpath("%s/.perfconfig", home));
+		struct stat st;
+
+		if (user_config == NULL) {
+			warning("Not enough memory to process %s/.perfconfig, "
+				"ignoring it.", home);
+			goto out;
+		}
+
+		if (stat(user_config, &st) < 0)
+			goto out_free;
+
+		if (st.st_uid && (st.st_uid != geteuid())) {
+			warning("File %s not owned by current user or root, "
+				"ignoring it.", user_config);
+			goto out_free;
+		}
+
+		if (!st.st_size)
+			goto out_free;
+
+		ret += perf_config_from_file(fn, user_config, data);
+		found += 1;
+out_free:
+		free(user_config);
+	}
+out:
+	if (found == 0)
+		return -1;
+	return ret;
+}
+
+/*
+ * Call this to report error for your variable that should not
+ * get a boolean value (i.e. "[my] var" means "true").
+ */
+int config_error_nonbool(const char *var)
+{
+	return error("Missing value for '%s'", var);
+}
+
+struct buildid_dir_config {
+	char *dir;
+};
+
+static int buildid_dir_command_config(const char *var, const char *value,
+				      void *data)
+{
+	struct buildid_dir_config *c = data;
+	const char *v;
+
+	/* same dir for all commands */
+	if (!strcmp(var, "buildid.dir")) {
+		v = perf_config_dirname(var, value);
+		if (!v)
+			return -1;
+		strncpy(c->dir, v, MAXPATHLEN-1);
+		c->dir[MAXPATHLEN-1] = '\0';
+	}
+	return 0;
+}
+
+static void check_buildid_dir_config(void)
+{
+	struct buildid_dir_config c;
+	c.dir = buildid_dir;
+	perf_config(buildid_dir_command_config, &c);
+}
+
+void set_buildid_dir(const char *dir)
+{
+	if (dir)
+		scnprintf(buildid_dir, MAXPATHLEN-1, "%s", dir);
+
+	/* try config file */
+	if (buildid_dir[0] == '\0')
+		check_buildid_dir_config();
+
+	/* default to $HOME/.debug */
+	if (buildid_dir[0] == '\0') {
+		char *v = getenv("HOME");
+		if (v) {
+			snprintf(buildid_dir, MAXPATHLEN-1, "%s/%s",
+				 v, DEBUG_CACHE_DIR);
+		} else {
+			strncpy(buildid_dir, DEBUG_CACHE_DIR, MAXPATHLEN-1);
+		}
+		buildid_dir[MAXPATHLEN-1] = '\0';
+	}
+	/* for communicating with external commands */
+	setenv("PERF_BUILDID_DIR", buildid_dir, 1);
+}
diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c
new file mode 100644
index 000000000..c4e55b710
--- /dev/null
+++ b/tools/perf/util/cpumap.c
@@ -0,0 +1,479 @@
+#include "util.h"
+#include <api/fs/fs.h>
+#include "../perf.h"
+#include "cpumap.h"
+#include <assert.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+static struct cpu_map *cpu_map__default_new(void)
+{
+	struct cpu_map *cpus;
+	int nr_cpus;
+
+	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
+	if (nr_cpus < 0)
+		return NULL;
+
+	cpus = malloc(sizeof(*cpus) + nr_cpus * sizeof(int));
+	if (cpus != NULL) {
+		int i;
+		for (i = 0; i < nr_cpus; ++i)
+			cpus->map[i] = i;
+
+		cpus->nr = nr_cpus;
+	}
+
+	return cpus;
+}
+
+static struct cpu_map *cpu_map__trim_new(int nr_cpus, int *tmp_cpus)
+{
+	size_t payload_size = nr_cpus * sizeof(int);
+	struct cpu_map *cpus = malloc(sizeof(*cpus) + payload_size);
+
+	if (cpus != NULL) {
+		cpus->nr = nr_cpus;
+		memcpy(cpus->map, tmp_cpus, payload_size);
+	}
+
+	return cpus;
+}
+
+struct cpu_map *cpu_map__read(FILE *file)
+{
+	struct cpu_map *cpus = NULL;
+	int nr_cpus = 0;
+	int *tmp_cpus = NULL, *tmp;
+	int max_entries = 0;
+	int n, cpu, prev;
+	char sep;
+
+	sep = 0;
+	prev = -1;
+	for (;;) {
+		n = fscanf(file, "%u%c", &cpu, &sep);
+		if (n <= 0)
+			break;
+		if (prev >= 0) {
+			int new_max = nr_cpus + cpu - prev - 1;
+
+			if (new_max >= max_entries) {
+				max_entries = new_max + MAX_NR_CPUS / 2;
+				tmp = realloc(tmp_cpus, max_entries * sizeof(int));
+				if (tmp == NULL)
+					goto out_free_tmp;
+				tmp_cpus = tmp;
+			}
+
+			while (++prev < cpu)
+				tmp_cpus[nr_cpus++] = prev;
+		}
+		if (nr_cpus == max_entries) {
+			max_entries += MAX_NR_CPUS;
+			tmp = realloc(tmp_cpus, max_entries * sizeof(int));
+			if (tmp == NULL)
+				goto out_free_tmp;
+			tmp_cpus = tmp;
+		}
+
+		tmp_cpus[nr_cpus++] = cpu;
+		if (n == 2 && sep == '-')
+			prev = cpu;
+		else
+			prev = -1;
+		if (n == 1 || sep == '\n')
+			break;
+	}
+
+	if (nr_cpus > 0)
+		cpus = cpu_map__trim_new(nr_cpus, tmp_cpus);
+	else
+		cpus = cpu_map__default_new();
+out_free_tmp:
+	free(tmp_cpus);
+	return cpus;
+}
+
+static struct cpu_map *cpu_map__read_all_cpu_map(void)
+{
+	struct cpu_map *cpus = NULL;
+	FILE *onlnf;
+
+	onlnf = fopen("/sys/devices/system/cpu/online", "r");
+	if (!onlnf)
+		return cpu_map__default_new();
+
+	cpus = cpu_map__read(onlnf);
+	fclose(onlnf);
+	return cpus;
+}
+
+struct cpu_map *cpu_map__new(const char *cpu_list)
+{
+	struct cpu_map *cpus = NULL;
+	unsigned long start_cpu, end_cpu = 0;
+	char *p = NULL;
+	int i, nr_cpus = 0;
+	int *tmp_cpus = NULL, *tmp;
+	int max_entries = 0;
+
+	if (!cpu_list)
+		return cpu_map__read_all_cpu_map();
+
+	if (!isdigit(*cpu_list))
+		goto out;
+
+	while (isdigit(*cpu_list)) {
+		p = NULL;
+		start_cpu = strtoul(cpu_list, &p, 0);
+		if (start_cpu >= INT_MAX
+		    || (*p != '\0' && *p != ',' && *p != '-'))
+			goto invalid;
+
+		if (*p == '-') {
+			cpu_list = ++p;
+			p = NULL;
+			end_cpu = strtoul(cpu_list, &p, 0);
+
+			if (end_cpu >= INT_MAX || (*p != '\0' && *p != ','))
+				goto invalid;
+
+			if (end_cpu < start_cpu)
+				goto invalid;
+		} else {
+			end_cpu = start_cpu;
+		}
+
+		for (; start_cpu <= end_cpu; start_cpu++) {
+			/* check for duplicates */
+			for (i = 0; i < nr_cpus; i++)
+				if (tmp_cpus[i] == (int)start_cpu)
+					goto invalid;
+
+			if (nr_cpus == max_entries) {
+				max_entries += MAX_NR_CPUS;
+				tmp = realloc(tmp_cpus, max_entries * sizeof(int));
+				if (tmp == NULL)
+					goto invalid;
+				tmp_cpus = tmp;
+			}
+			tmp_cpus[nr_cpus++] = (int)start_cpu;
+		}
+		if (*p)
+			++p;
+
+		cpu_list = p;
+	}
+
+	if (nr_cpus > 0)
+		cpus = cpu_map__trim_new(nr_cpus, tmp_cpus);
+	else
+		cpus = cpu_map__default_new();
+invalid:
+	free(tmp_cpus);
+out:
+	return cpus;
+}
+
+size_t cpu_map__fprintf(struct cpu_map *map, FILE *fp)
+{
+	int i;
+	size_t printed = fprintf(fp, "%d cpu%s: ",
+				 map->nr, map->nr > 1 ? "s" : "");
+	for (i = 0; i < map->nr; ++i)
+		printed += fprintf(fp, "%s%d", i ? ", " : "", map->map[i]);
+
+	return printed + fprintf(fp, "\n");
+}
+
+struct cpu_map *cpu_map__dummy_new(void)
+{
+	struct cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int));
+
+	if (cpus != NULL) {
+		cpus->nr = 1;
+		cpus->map[0] = -1;
+	}
+
+	return cpus;
+}
+
+void cpu_map__delete(struct cpu_map *map)
+{
+	free(map);
+}
+
+int cpu_map__get_socket(struct cpu_map *map, int idx)
+{
+	FILE *fp;
+	const char *mnt;
+	char path[PATH_MAX];
+	int cpu, ret;
+
+	if (idx > map->nr)
+		return -1;
+
+	cpu = map->map[idx];
+
+	mnt = sysfs__mountpoint();
+	if (!mnt)
+		return -1;
+
+	snprintf(path, PATH_MAX,
+		"%s/devices/system/cpu/cpu%d/topology/physical_package_id",
+		mnt, cpu);
+
+	fp = fopen(path, "r");
+	if (!fp)
+		return -1;
+	ret = fscanf(fp, "%d", &cpu);
+	fclose(fp);
+	return ret == 1 ? cpu : -1;
+}
+
+static int cmp_ids(const void *a, const void *b)
+{
+	return *(int *)a - *(int *)b;
+}
+
+static int cpu_map__build_map(struct cpu_map *cpus, struct cpu_map **res,
+			      int (*f)(struct cpu_map *map, int cpu))
+{
+	struct cpu_map *c;
+	int nr = cpus->nr;
+	int cpu, s1, s2;
+
+	/* allocate as much as possible */
+	c = calloc(1, sizeof(*c) + nr * sizeof(int));
+	if (!c)
+		return -1;
+
+	for (cpu = 0; cpu < nr; cpu++) {
+		s1 = f(cpus, cpu);
+		for (s2 = 0; s2 < c->nr; s2++) {
+			if (s1 == c->map[s2])
+				break;
+		}
+		if (s2 == c->nr) {
+			c->map[c->nr] = s1;
+			c->nr++;
+		}
+	}
+	/* ensure we process id in increasing order */
+	qsort(c->map, c->nr, sizeof(int), cmp_ids);
+
+	*res = c;
+	return 0;
+}
+
+int cpu_map__get_core(struct cpu_map *map, int idx)
+{
+	FILE *fp;
+	const char *mnt;
+	char path[PATH_MAX];
+	int cpu, ret, s;
+
+	if (idx > map->nr)
+		return -1;
+
+	cpu = map->map[idx];
+
+	mnt = sysfs__mountpoint();
+	if (!mnt)
+		return -1;
+
+	snprintf(path, PATH_MAX,
+		"%s/devices/system/cpu/cpu%d/topology/core_id",
+		mnt, cpu);
+
+	fp = fopen(path, "r");
+	if (!fp)
+		return -1;
+	ret = fscanf(fp, "%d", &cpu);
+	fclose(fp);
+	if (ret != 1)
+		return -1;
+
+	s = cpu_map__get_socket(map, idx);
+	if (s == -1)
+		return -1;
+
+	/*
+	 * encode socket in upper 16 bits
+	 * core_id is relative to socket, and
+	 * we need a global id. So we combine
+	 * socket+ core id
+	 */
+	return (s << 16) | (cpu & 0xffff);
+}
+
+int cpu_map__build_socket_map(struct cpu_map *cpus, struct cpu_map **sockp)
+{
+	return cpu_map__build_map(cpus, sockp, cpu_map__get_socket);
+}
+
+int cpu_map__build_core_map(struct cpu_map *cpus, struct cpu_map **corep)
+{
+	return cpu_map__build_map(cpus, corep, cpu_map__get_core);
+}
+
+/* setup simple routines to easily access node numbers given a cpu number */
+static int get_max_num(char *path, int *max)
+{
+	size_t num;
+	char *buf;
+	int err = 0;
+
+	if (filename__read_str(path, &buf, &num))
+		return -1;
+
+	buf[num] = '\0';
+
+	/* start on the right, to find highest node num */
+	while (--num) {
+		if ((buf[num] == ',') || (buf[num] == '-')) {
+			num++;
+			break;
+		}
+	}
+	if (sscanf(&buf[num], "%d", max) < 1) {
+		err = -1;
+		goto out;
+	}
+
+	/* convert from 0-based to 1-based */
+	(*max)++;
+
+out:
+	free(buf);
+	return err;
+}
+
+/* Determine highest possible cpu in the system for sparse allocation */
+static void set_max_cpu_num(void)
+{
+	const char *mnt;
+	char path[PATH_MAX];
+	int ret = -1;
+
+	/* set up default */
+	max_cpu_num = 4096;
+
+	mnt = sysfs__mountpoint();
+	if (!mnt)
+		goto out;
+
+	/* get the highest possible cpu number for a sparse allocation */
+	ret = snprintf(path, PATH_MAX, "%s/devices/system/cpu/possible", mnt);
+	if (ret == PATH_MAX) {
+		pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
+		goto out;
+	}
+
+	ret = get_max_num(path, &max_cpu_num);
+
+out:
+	if (ret)
+		pr_err("Failed to read max cpus, using default of %d\n", max_cpu_num);
+}
+
+/* Determine highest possible node in the system for sparse allocation */
+static void set_max_node_num(void)
+{
+	const char *mnt;
+	char path[PATH_MAX];
+	int ret = -1;
+
+	/* set up default */
+	max_node_num = 8;
+
+	mnt = sysfs__mountpoint();
+	if (!mnt)
+		goto out;
+
+	/* get the highest possible cpu number for a sparse allocation */
+	ret = snprintf(path, PATH_MAX, "%s/devices/system/node/possible", mnt);
+	if (ret == PATH_MAX) {
+		pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
+		goto out;
+	}
+
+	ret = get_max_num(path, &max_node_num);
+
+out:
+	if (ret)
+		pr_err("Failed to read max nodes, using default of %d\n", max_node_num);
+}
+
+static int init_cpunode_map(void)
+{
+	int i;
+
+	set_max_cpu_num();
+	set_max_node_num();
+
+	cpunode_map = calloc(max_cpu_num, sizeof(int));
+	if (!cpunode_map) {
+		pr_err("%s: calloc failed\n", __func__);
+		return -1;
+	}
+
+	for (i = 0; i < max_cpu_num; i++)
+		cpunode_map[i] = -1;
+
+	return 0;
+}
+
+int cpu__setup_cpunode_map(void)
+{
+	struct dirent *dent1, *dent2;
+	DIR *dir1, *dir2;
+	unsigned int cpu, mem;
+	char buf[PATH_MAX];
+	char path[PATH_MAX];
+	const char *mnt;
+	int n;
+
+	/* initialize globals */
+	if (init_cpunode_map())
+		return -1;
+
+	mnt = sysfs__mountpoint();
+	if (!mnt)
+		return 0;
+
+	n = snprintf(path, PATH_MAX, "%s/devices/system/node", mnt);
+	if (n == PATH_MAX) {
+		pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
+		return -1;
+	}
+
+	dir1 = opendir(path);
+	if (!dir1)
+		return 0;
+
+	/* walk tree and setup map */
+	while ((dent1 = readdir(dir1)) != NULL) {
+		if (dent1->d_type != DT_DIR || sscanf(dent1->d_name, "node%u", &mem) < 1)
+			continue;
+
+		n = snprintf(buf, PATH_MAX, "%s/%s", path, dent1->d_name);
+		if (n == PATH_MAX) {
+			pr_err("sysfs path crossed PATH_MAX(%d) size\n", PATH_MAX);
+			continue;
+		}
+
+		dir2 = opendir(buf);
+		if (!dir2)
+			continue;
+		while ((dent2 = readdir(dir2)) != NULL) {
+			if (dent2->d_type != DT_LNK || sscanf(dent2->d_name, "cpu%u", &cpu) < 1)
+				continue;
+			cpunode_map[cpu] = mem;
+		}
+		closedir(dir2);
+	}
+	closedir(dir1);
+	return 0;
+}
diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h
new file mode 100644
index 000000000..61a654849
--- /dev/null
+++ b/tools/perf/util/cpumap.h
@@ -0,0 +1,84 @@
+#ifndef __PERF_CPUMAP_H
+#define __PERF_CPUMAP_H
+
+#include <stdio.h>
+#include <stdbool.h>
+
+#include "perf.h"
+#include "util/debug.h"
+
+struct cpu_map {
+	int nr;
+	int map[];
+};
+
+struct cpu_map *cpu_map__new(const char *cpu_list);
+struct cpu_map *cpu_map__dummy_new(void);
+void cpu_map__delete(struct cpu_map *map);
+struct cpu_map *cpu_map__read(FILE *file);
+size_t cpu_map__fprintf(struct cpu_map *map, FILE *fp);
+int cpu_map__get_socket(struct cpu_map *map, int idx);
+int cpu_map__get_core(struct cpu_map *map, int idx);
+int cpu_map__build_socket_map(struct cpu_map *cpus, struct cpu_map **sockp);
+int cpu_map__build_core_map(struct cpu_map *cpus, struct cpu_map **corep);
+
+static inline int cpu_map__socket(struct cpu_map *sock, int s)
+{
+	if (!sock || s > sock->nr || s < 0)
+		return 0;
+	return sock->map[s];
+}
+
+static inline int cpu_map__id_to_socket(int id)
+{
+	return id >> 16;
+}
+
+static inline int cpu_map__id_to_cpu(int id)
+{
+	return id & 0xffff;
+}
+
+static inline int cpu_map__nr(const struct cpu_map *map)
+{
+	return map ? map->nr : 1;
+}
+
+static inline bool cpu_map__empty(const struct cpu_map *map)
+{
+	return map ? map->map[0] == -1 : true;
+}
+
+int max_cpu_num;
+int max_node_num;
+int *cpunode_map;
+
+int cpu__setup_cpunode_map(void);
+
+static inline int cpu__max_node(void)
+{
+	if (unlikely(!max_node_num))
+		pr_debug("cpu_map not initialized\n");
+
+	return max_node_num;
+}
+
+static inline int cpu__max_cpu(void)
+{
+	if (unlikely(!max_cpu_num))
+		pr_debug("cpu_map not initialized\n");
+
+	return max_cpu_num;
+}
+
+static inline int cpu__get_node(int cpu)
+{
+	if (unlikely(cpunode_map == NULL)) {
+		pr_debug("cpu_map not initialized\n");
+		return -1;
+	}
+
+	return cpunode_map[cpu];
+}
+
+#endif /* __PERF_CPUMAP_H */
diff --git a/tools/perf/util/ctype.c b/tools/perf/util/ctype.c
new file mode 100644
index 000000000..aada3ac5e
--- /dev/null
+++ b/tools/perf/util/ctype.c
@@ -0,0 +1,39 @@
+/*
+ * Sane locale-independent, ASCII ctype.
+ *
+ * No surprises, and works with signed and unsigned chars.
+ */
+#include "util.h"
+
+enum {
+	S = GIT_SPACE,
+	A = GIT_ALPHA,
+	D = GIT_DIGIT,
+	G = GIT_GLOB_SPECIAL,	/* *, ?, [, \\ */
+	R = GIT_REGEX_SPECIAL,	/* $, (, ), +, ., ^, {, | * */
+	P = GIT_PRINT_EXTRA,	/* printable - alpha - digit - glob - regex */
+
+	PS = GIT_SPACE | GIT_PRINT_EXTRA,
+};
+
+unsigned char sane_ctype[256] = {
+/*	0  1  2  3  4  5  6  7  8  9  A  B  C  D  E  F			    */
+
+	0, 0, 0, 0, 0, 0, 0, 0, 0, S, S, 0, 0, S, 0, 0,		/*   0.. 15 */
+	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,		/*  16.. 31 */
+	PS,P, P, P, R, P, P, P, R, R, G, R, P, P, R, P,		/*  32.. 47 */
+	D, D, D, D, D, D, D, D, D, D, P, P, P, P, P, G,		/*  48.. 63 */
+	P, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,		/*  64.. 79 */
+	A, A, A, A, A, A, A, A, A, A, A, G, G, P, R, P,		/*  80.. 95 */
+	P, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,		/*  96..111 */
+	A, A, A, A, A, A, A, A, A, A, A, R, R, P, P, 0,		/* 112..127 */
+	/* Nothing in the 128.. range */
+};
+
+const char *graph_line =
+	"_____________________________________________________________________"
+	"_____________________________________________________________________";
+const char *graph_dotted_line =
+	"---------------------------------------------------------------------"
+	"---------------------------------------------------------------------"
+	"---------------------------------------------------------------------";
diff --git a/tools/perf/util/data-convert-bt.c b/tools/perf/util/data-convert-bt.c
new file mode 100644
index 000000000..dd17c9a32
--- /dev/null
+++ b/tools/perf/util/data-convert-bt.c
@@ -0,0 +1,857 @@
+/*
+ * CTF writing support via babeltrace.
+ *
+ * Copyright (C) 2014, Jiri Olsa <jolsa@redhat.com>
+ * Copyright (C) 2014, Sebastian Andrzej Siewior <bigeasy@linutronix.de>
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+
+#include <linux/compiler.h>
+#include <babeltrace/ctf-writer/writer.h>
+#include <babeltrace/ctf-writer/clock.h>
+#include <babeltrace/ctf-writer/stream.h>
+#include <babeltrace/ctf-writer/event.h>
+#include <babeltrace/ctf-writer/event-types.h>
+#include <babeltrace/ctf-writer/event-fields.h>
+#include <babeltrace/ctf/events.h>
+#include <traceevent/event-parse.h>
+#include "asm/bug.h"
+#include "data-convert-bt.h"
+#include "session.h"
+#include "util.h"
+#include "debug.h"
+#include "tool.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "machine.h"
+
+#define pr_N(n, fmt, ...) \
+	eprintf(n, debug_data_convert, fmt, ##__VA_ARGS__)
+
+#define pr(fmt, ...)  pr_N(1, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr2(fmt, ...) pr_N(2, pr_fmt(fmt), ##__VA_ARGS__)
+
+#define pr_time2(t, fmt, ...) pr_time_N(2, debug_data_convert, t, pr_fmt(fmt), ##__VA_ARGS__)
+
+struct evsel_priv {
+	struct bt_ctf_event_class *event_class;
+};
+
+struct ctf_writer {
+	/* writer primitives */
+	struct bt_ctf_writer		*writer;
+	struct bt_ctf_stream		*stream;
+	struct bt_ctf_stream_class	*stream_class;
+	struct bt_ctf_clock		*clock;
+
+	/* data types */
+	union {
+		struct {
+			struct bt_ctf_field_type	*s64;
+			struct bt_ctf_field_type	*u64;
+			struct bt_ctf_field_type	*s32;
+			struct bt_ctf_field_type	*u32;
+			struct bt_ctf_field_type	*string;
+			struct bt_ctf_field_type	*u64_hex;
+		};
+		struct bt_ctf_field_type *array[6];
+	} data;
+};
+
+struct convert {
+	struct perf_tool	tool;
+	struct ctf_writer	writer;
+
+	u64			events_size;
+	u64			events_count;
+};
+
+static int value_set(struct bt_ctf_field_type *type,
+		     struct bt_ctf_event *event,
+		     const char *name, u64 val)
+{
+	struct bt_ctf_field *field;
+	bool sign = bt_ctf_field_type_integer_get_signed(type);
+	int ret;
+
+	field = bt_ctf_field_create(type);
+	if (!field) {
+		pr_err("failed to create a field %s\n", name);
+		return -1;
+	}
+
+	if (sign) {
+		ret = bt_ctf_field_signed_integer_set_value(field, val);
+		if (ret) {
+			pr_err("failed to set field value %s\n", name);
+			goto err;
+		}
+	} else {
+		ret = bt_ctf_field_unsigned_integer_set_value(field, val);
+		if (ret) {
+			pr_err("failed to set field value %s\n", name);
+			goto err;
+		}
+	}
+
+	ret = bt_ctf_event_set_payload(event, name, field);
+	if (ret) {
+		pr_err("failed to set payload %s\n", name);
+		goto err;
+	}
+
+	pr2("  SET [%s = %" PRIu64 "]\n", name, val);
+
+err:
+	bt_ctf_field_put(field);
+	return ret;
+}
+
+#define __FUNC_VALUE_SET(_name, _val_type)				\
+static __maybe_unused int value_set_##_name(struct ctf_writer *cw,	\
+			     struct bt_ctf_event *event,		\
+			     const char *name,				\
+			     _val_type val)				\
+{									\
+	struct bt_ctf_field_type *type = cw->data._name;		\
+	return value_set(type, event, name, (u64) val);			\
+}
+
+#define FUNC_VALUE_SET(_name) __FUNC_VALUE_SET(_name, _name)
+
+FUNC_VALUE_SET(s32)
+FUNC_VALUE_SET(u32)
+FUNC_VALUE_SET(s64)
+FUNC_VALUE_SET(u64)
+__FUNC_VALUE_SET(u64_hex, u64)
+
+static struct bt_ctf_field_type*
+get_tracepoint_field_type(struct ctf_writer *cw, struct format_field *field)
+{
+	unsigned long flags = field->flags;
+
+	if (flags & FIELD_IS_STRING)
+		return cw->data.string;
+
+	if (!(flags & FIELD_IS_SIGNED)) {
+		/* unsigned long are mostly pointers */
+		if (flags & FIELD_IS_LONG || flags & FIELD_IS_POINTER)
+			return cw->data.u64_hex;
+	}
+
+	if (flags & FIELD_IS_SIGNED) {
+		if (field->size == 8)
+			return cw->data.s64;
+		else
+			return cw->data.s32;
+	}
+
+	if (field->size == 8)
+		return cw->data.u64;
+	else
+		return cw->data.u32;
+}
+
+static int add_tracepoint_field_value(struct ctf_writer *cw,
+				      struct bt_ctf_event_class *event_class,
+				      struct bt_ctf_event *event,
+				      struct perf_sample *sample,
+				      struct format_field *fmtf)
+{
+	struct bt_ctf_field_type *type;
+	struct bt_ctf_field *array_field;
+	struct bt_ctf_field *field;
+	const char *name = fmtf->name;
+	void *data = sample->raw_data;
+	unsigned long long value_int;
+	unsigned long flags = fmtf->flags;
+	unsigned int n_items;
+	unsigned int i;
+	unsigned int offset;
+	unsigned int len;
+	int ret;
+
+	offset = fmtf->offset;
+	len = fmtf->size;
+	if (flags & FIELD_IS_STRING)
+		flags &= ~FIELD_IS_ARRAY;
+
+	if (flags & FIELD_IS_DYNAMIC) {
+		unsigned long long tmp_val;
+
+		tmp_val = pevent_read_number(fmtf->event->pevent,
+				data + offset, len);
+		offset = tmp_val;
+		len = offset >> 16;
+		offset &= 0xffff;
+	}
+
+	if (flags & FIELD_IS_ARRAY) {
+
+		type = bt_ctf_event_class_get_field_by_name(
+				event_class, name);
+		array_field = bt_ctf_field_create(type);
+		bt_ctf_field_type_put(type);
+		if (!array_field) {
+			pr_err("Failed to create array type %s\n", name);
+			return -1;
+		}
+
+		len = fmtf->size / fmtf->arraylen;
+		n_items = fmtf->arraylen;
+	} else {
+		n_items = 1;
+		array_field = NULL;
+	}
+
+	type = get_tracepoint_field_type(cw, fmtf);
+
+	for (i = 0; i < n_items; i++) {
+		if (!(flags & FIELD_IS_STRING))
+			value_int = pevent_read_number(
+					fmtf->event->pevent,
+					data + offset + i * len, len);
+
+		if (flags & FIELD_IS_ARRAY)
+			field = bt_ctf_field_array_get_field(array_field, i);
+		else
+			field = bt_ctf_field_create(type);
+
+		if (!field) {
+			pr_err("failed to create a field %s\n", name);
+			return -1;
+		}
+
+		if (flags & FIELD_IS_STRING)
+			ret = bt_ctf_field_string_set_value(field,
+					data + offset + i * len);
+		else if (!(flags & FIELD_IS_SIGNED))
+			ret = bt_ctf_field_unsigned_integer_set_value(
+					field, value_int);
+		else
+			ret = bt_ctf_field_signed_integer_set_value(
+					field, value_int);
+		if (ret) {
+			pr_err("failed to set file value %s\n", name);
+			goto err_put_field;
+		}
+		if (!(flags & FIELD_IS_ARRAY)) {
+			ret = bt_ctf_event_set_payload(event, name, field);
+			if (ret) {
+				pr_err("failed to set payload %s\n", name);
+				goto err_put_field;
+			}
+		}
+		bt_ctf_field_put(field);
+	}
+	if (flags & FIELD_IS_ARRAY) {
+		ret = bt_ctf_event_set_payload(event, name, array_field);
+		if (ret) {
+			pr_err("Failed add payload array %s\n", name);
+			return -1;
+		}
+		bt_ctf_field_put(array_field);
+	}
+	return 0;
+
+err_put_field:
+	bt_ctf_field_put(field);
+	return -1;
+}
+
+static int add_tracepoint_fields_values(struct ctf_writer *cw,
+					struct bt_ctf_event_class *event_class,
+					struct bt_ctf_event *event,
+					struct format_field *fields,
+					struct perf_sample *sample)
+{
+	struct format_field *field;
+	int ret;
+
+	for (field = fields; field; field = field->next) {
+		ret = add_tracepoint_field_value(cw, event_class, event, sample,
+				field);
+		if (ret)
+			return -1;
+	}
+	return 0;
+}
+
+static int add_tracepoint_values(struct ctf_writer *cw,
+				 struct bt_ctf_event_class *event_class,
+				 struct bt_ctf_event *event,
+				 struct perf_evsel *evsel,
+				 struct perf_sample *sample)
+{
+	struct format_field *common_fields = evsel->tp_format->format.common_fields;
+	struct format_field *fields        = evsel->tp_format->format.fields;
+	int ret;
+
+	ret = add_tracepoint_fields_values(cw, event_class, event,
+					   common_fields, sample);
+	if (!ret)
+		ret = add_tracepoint_fields_values(cw, event_class, event,
+						   fields, sample);
+
+	return ret;
+}
+
+static int add_generic_values(struct ctf_writer *cw,
+			      struct bt_ctf_event *event,
+			      struct perf_evsel *evsel,
+			      struct perf_sample *sample)
+{
+	u64 type = evsel->attr.sample_type;
+	int ret;
+
+	/*
+	 * missing:
+	 *   PERF_SAMPLE_TIME         - not needed as we have it in
+	 *                              ctf event header
+	 *   PERF_SAMPLE_READ         - TODO
+	 *   PERF_SAMPLE_CALLCHAIN    - TODO
+	 *   PERF_SAMPLE_RAW          - tracepoint fields are handled separately
+	 *   PERF_SAMPLE_BRANCH_STACK - TODO
+	 *   PERF_SAMPLE_REGS_USER    - TODO
+	 *   PERF_SAMPLE_STACK_USER   - TODO
+	 */
+
+	if (type & PERF_SAMPLE_IP) {
+		ret = value_set_u64_hex(cw, event, "perf_ip", sample->ip);
+		if (ret)
+			return -1;
+	}
+
+	if (type & PERF_SAMPLE_TID) {
+		ret = value_set_s32(cw, event, "perf_tid", sample->tid);
+		if (ret)
+			return -1;
+
+		ret = value_set_s32(cw, event, "perf_pid", sample->pid);
+		if (ret)
+			return -1;
+	}
+
+	if ((type & PERF_SAMPLE_ID) ||
+	    (type & PERF_SAMPLE_IDENTIFIER)) {
+		ret = value_set_u64(cw, event, "perf_id", sample->id);
+		if (ret)
+			return -1;
+	}
+
+	if (type & PERF_SAMPLE_STREAM_ID) {
+		ret = value_set_u64(cw, event, "perf_stream_id", sample->stream_id);
+		if (ret)
+			return -1;
+	}
+
+	if (type & PERF_SAMPLE_CPU) {
+		ret = value_set_u32(cw, event, "perf_cpu", sample->cpu);
+		if (ret)
+			return -1;
+	}
+
+	if (type & PERF_SAMPLE_PERIOD) {
+		ret = value_set_u64(cw, event, "perf_period", sample->period);
+		if (ret)
+			return -1;
+	}
+
+	if (type & PERF_SAMPLE_WEIGHT) {
+		ret = value_set_u64(cw, event, "perf_weight", sample->weight);
+		if (ret)
+			return -1;
+	}
+
+	if (type & PERF_SAMPLE_DATA_SRC) {
+		ret = value_set_u64(cw, event, "perf_data_src",
+				sample->data_src);
+		if (ret)
+			return -1;
+	}
+
+	if (type & PERF_SAMPLE_TRANSACTION) {
+		ret = value_set_u64(cw, event, "perf_transaction",
+				sample->transaction);
+		if (ret)
+			return -1;
+	}
+
+	return 0;
+}
+
+static int process_sample_event(struct perf_tool *tool,
+				union perf_event *_event __maybe_unused,
+				struct perf_sample *sample,
+				struct perf_evsel *evsel,
+				struct machine *machine __maybe_unused)
+{
+	struct convert *c = container_of(tool, struct convert, tool);
+	struct evsel_priv *priv = evsel->priv;
+	struct ctf_writer *cw = &c->writer;
+	struct bt_ctf_event_class *event_class;
+	struct bt_ctf_event *event;
+	int ret;
+
+	if (WARN_ONCE(!priv, "Failed to setup all events.\n"))
+		return 0;
+
+	event_class = priv->event_class;
+
+	/* update stats */
+	c->events_count++;
+	c->events_size += _event->header.size;
+
+	pr_time2(sample->time, "sample %" PRIu64 "\n", c->events_count);
+
+	event = bt_ctf_event_create(event_class);
+	if (!event) {
+		pr_err("Failed to create an CTF event\n");
+		return -1;
+	}
+
+	bt_ctf_clock_set_time(cw->clock, sample->time);
+
+	ret = add_generic_values(cw, event, evsel, sample);
+	if (ret)
+		return -1;
+
+	if (evsel->attr.type == PERF_TYPE_TRACEPOINT) {
+		ret = add_tracepoint_values(cw, event_class, event,
+					    evsel, sample);
+		if (ret)
+			return -1;
+	}
+
+	bt_ctf_stream_append_event(cw->stream, event);
+	bt_ctf_event_put(event);
+	return 0;
+}
+
+static int add_tracepoint_fields_types(struct ctf_writer *cw,
+				       struct format_field *fields,
+				       struct bt_ctf_event_class *event_class)
+{
+	struct format_field *field;
+	int ret;
+
+	for (field = fields; field; field = field->next) {
+		struct bt_ctf_field_type *type;
+		unsigned long flags = field->flags;
+
+		pr2("  field '%s'\n", field->name);
+
+		type = get_tracepoint_field_type(cw, field);
+		if (!type)
+			return -1;
+
+		/*
+		 * A string is an array of chars. For this we use the string
+		 * type and don't care that it is an array. What we don't
+		 * support is an array of strings.
+		 */
+		if (flags & FIELD_IS_STRING)
+			flags &= ~FIELD_IS_ARRAY;
+
+		if (flags & FIELD_IS_ARRAY)
+			type = bt_ctf_field_type_array_create(type, field->arraylen);
+
+		ret = bt_ctf_event_class_add_field(event_class, type,
+				field->name);
+
+		if (flags & FIELD_IS_ARRAY)
+			bt_ctf_field_type_put(type);
+
+		if (ret) {
+			pr_err("Failed to add field '%s\n", field->name);
+			return -1;
+		}
+	}
+
+	return 0;
+}
+
+static int add_tracepoint_types(struct ctf_writer *cw,
+				struct perf_evsel *evsel,
+				struct bt_ctf_event_class *class)
+{
+	struct format_field *common_fields = evsel->tp_format->format.common_fields;
+	struct format_field *fields        = evsel->tp_format->format.fields;
+	int ret;
+
+	ret = add_tracepoint_fields_types(cw, common_fields, class);
+	if (!ret)
+		ret = add_tracepoint_fields_types(cw, fields, class);
+
+	return ret;
+}
+
+static int add_generic_types(struct ctf_writer *cw, struct perf_evsel *evsel,
+			     struct bt_ctf_event_class *event_class)
+{
+	u64 type = evsel->attr.sample_type;
+
+	/*
+	 * missing:
+	 *   PERF_SAMPLE_TIME         - not needed as we have it in
+	 *                              ctf event header
+	 *   PERF_SAMPLE_READ         - TODO
+	 *   PERF_SAMPLE_CALLCHAIN    - TODO
+	 *   PERF_SAMPLE_RAW          - tracepoint fields are handled separately
+	 *   PERF_SAMPLE_BRANCH_STACK - TODO
+	 *   PERF_SAMPLE_REGS_USER    - TODO
+	 *   PERF_SAMPLE_STACK_USER   - TODO
+	 */
+
+#define ADD_FIELD(cl, t, n)						\
+	do {								\
+		pr2("  field '%s'\n", n);				\
+		if (bt_ctf_event_class_add_field(cl, t, n)) {		\
+			pr_err("Failed to add field '%s;\n", n);	\
+			return -1;					\
+		}							\
+	} while (0)
+
+	if (type & PERF_SAMPLE_IP)
+		ADD_FIELD(event_class, cw->data.u64_hex, "perf_ip");
+
+	if (type & PERF_SAMPLE_TID) {
+		ADD_FIELD(event_class, cw->data.s32, "perf_tid");
+		ADD_FIELD(event_class, cw->data.s32, "perf_pid");
+	}
+
+	if ((type & PERF_SAMPLE_ID) ||
+	    (type & PERF_SAMPLE_IDENTIFIER))
+		ADD_FIELD(event_class, cw->data.u64, "perf_id");
+
+	if (type & PERF_SAMPLE_STREAM_ID)
+		ADD_FIELD(event_class, cw->data.u64, "perf_stream_id");
+
+	if (type & PERF_SAMPLE_CPU)
+		ADD_FIELD(event_class, cw->data.u32, "perf_cpu");
+
+	if (type & PERF_SAMPLE_PERIOD)
+		ADD_FIELD(event_class, cw->data.u64, "perf_period");
+
+	if (type & PERF_SAMPLE_WEIGHT)
+		ADD_FIELD(event_class, cw->data.u64, "perf_weight");
+
+	if (type & PERF_SAMPLE_DATA_SRC)
+		ADD_FIELD(event_class, cw->data.u64, "perf_data_src");
+
+	if (type & PERF_SAMPLE_TRANSACTION)
+		ADD_FIELD(event_class, cw->data.u64, "perf_transaction");
+
+#undef ADD_FIELD
+	return 0;
+}
+
+static int add_event(struct ctf_writer *cw, struct perf_evsel *evsel)
+{
+	struct bt_ctf_event_class *event_class;
+	struct evsel_priv *priv;
+	const char *name = perf_evsel__name(evsel);
+	int ret;
+
+	pr("Adding event '%s' (type %d)\n", name, evsel->attr.type);
+
+	event_class = bt_ctf_event_class_create(name);
+	if (!event_class)
+		return -1;
+
+	ret = add_generic_types(cw, evsel, event_class);
+	if (ret)
+		goto err;
+
+	if (evsel->attr.type == PERF_TYPE_TRACEPOINT) {
+		ret = add_tracepoint_types(cw, evsel, event_class);
+		if (ret)
+			goto err;
+	}
+
+	ret = bt_ctf_stream_class_add_event_class(cw->stream_class, event_class);
+	if (ret) {
+		pr("Failed to add event class into stream.\n");
+		goto err;
+	}
+
+	priv = malloc(sizeof(*priv));
+	if (!priv)
+		goto err;
+
+	priv->event_class = event_class;
+	evsel->priv       = priv;
+	return 0;
+
+err:
+	bt_ctf_event_class_put(event_class);
+	pr_err("Failed to add event '%s'.\n", name);
+	return -1;
+}
+
+static int setup_events(struct ctf_writer *cw, struct perf_session *session)
+{
+	struct perf_evlist *evlist = session->evlist;
+	struct perf_evsel *evsel;
+	int ret;
+
+	evlist__for_each(evlist, evsel) {
+		ret = add_event(cw, evsel);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+static int ctf_writer__setup_env(struct ctf_writer *cw,
+				 struct perf_session *session)
+{
+	struct perf_header *header = &session->header;
+	struct bt_ctf_writer *writer = cw->writer;
+
+#define ADD(__n, __v)							\
+do {									\
+	if (bt_ctf_writer_add_environment_field(writer, __n, __v))	\
+		return -1;						\
+} while (0)
+
+	ADD("host",    header->env.hostname);
+	ADD("sysname", "Linux");
+	ADD("release", header->env.os_release);
+	ADD("version", header->env.version);
+	ADD("machine", header->env.arch);
+	ADD("domain", "kernel");
+	ADD("tracer_name", "perf");
+
+#undef ADD
+	return 0;
+}
+
+static int ctf_writer__setup_clock(struct ctf_writer *cw)
+{
+	struct bt_ctf_clock *clock = cw->clock;
+
+	bt_ctf_clock_set_description(clock, "perf clock");
+
+#define SET(__n, __v)				\
+do {						\
+	if (bt_ctf_clock_set_##__n(clock, __v))	\
+		return -1;			\
+} while (0)
+
+	SET(frequency,   1000000000);
+	SET(offset_s,    0);
+	SET(offset,      0);
+	SET(precision,   10);
+	SET(is_absolute, 0);
+
+#undef SET
+	return 0;
+}
+
+static struct bt_ctf_field_type *create_int_type(int size, bool sign, bool hex)
+{
+	struct bt_ctf_field_type *type;
+
+	type = bt_ctf_field_type_integer_create(size);
+	if (!type)
+		return NULL;
+
+	if (sign &&
+	    bt_ctf_field_type_integer_set_signed(type, 1))
+		goto err;
+
+	if (hex &&
+	    bt_ctf_field_type_integer_set_base(type, BT_CTF_INTEGER_BASE_HEXADECIMAL))
+		goto err;
+
+	pr2("Created type: INTEGER %d-bit %ssigned %s\n",
+	    size, sign ? "un" : "", hex ? "hex" : "");
+	return type;
+
+err:
+	bt_ctf_field_type_put(type);
+	return NULL;
+}
+
+static void ctf_writer__cleanup_data(struct ctf_writer *cw)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(cw->data.array); i++)
+		bt_ctf_field_type_put(cw->data.array[i]);
+}
+
+static int ctf_writer__init_data(struct ctf_writer *cw)
+{
+#define CREATE_INT_TYPE(type, size, sign, hex)		\
+do {							\
+	(type) = create_int_type(size, sign, hex);	\
+	if (!(type))					\
+		goto err;				\
+} while (0)
+
+	CREATE_INT_TYPE(cw->data.s64, 64, true,  false);
+	CREATE_INT_TYPE(cw->data.u64, 64, false, false);
+	CREATE_INT_TYPE(cw->data.s32, 32, true,  false);
+	CREATE_INT_TYPE(cw->data.u32, 32, false, false);
+	CREATE_INT_TYPE(cw->data.u64_hex, 64, false, true);
+
+	cw->data.string  = bt_ctf_field_type_string_create();
+	if (cw->data.string)
+		return 0;
+
+err:
+	ctf_writer__cleanup_data(cw);
+	pr_err("Failed to create data types.\n");
+	return -1;
+}
+
+static void ctf_writer__cleanup(struct ctf_writer *cw)
+{
+	ctf_writer__cleanup_data(cw);
+
+	bt_ctf_clock_put(cw->clock);
+	bt_ctf_stream_put(cw->stream);
+	bt_ctf_stream_class_put(cw->stream_class);
+	bt_ctf_writer_put(cw->writer);
+
+	/* and NULL all the pointers */
+	memset(cw, 0, sizeof(*cw));
+}
+
+static int ctf_writer__init(struct ctf_writer *cw, const char *path)
+{
+	struct bt_ctf_writer		*writer;
+	struct bt_ctf_stream_class	*stream_class;
+	struct bt_ctf_stream		*stream;
+	struct bt_ctf_clock		*clock;
+
+	/* CTF writer */
+	writer = bt_ctf_writer_create(path);
+	if (!writer)
+		goto err;
+
+	cw->writer = writer;
+
+	/* CTF clock */
+	clock = bt_ctf_clock_create("perf_clock");
+	if (!clock) {
+		pr("Failed to create CTF clock.\n");
+		goto err_cleanup;
+	}
+
+	cw->clock = clock;
+
+	if (ctf_writer__setup_clock(cw)) {
+		pr("Failed to setup CTF clock.\n");
+		goto err_cleanup;
+	}
+
+	/* CTF stream class */
+	stream_class = bt_ctf_stream_class_create("perf_stream");
+	if (!stream_class) {
+		pr("Failed to create CTF stream class.\n");
+		goto err_cleanup;
+	}
+
+	cw->stream_class = stream_class;
+
+	/* CTF clock stream setup */
+	if (bt_ctf_stream_class_set_clock(stream_class, clock)) {
+		pr("Failed to assign CTF clock to stream class.\n");
+		goto err_cleanup;
+	}
+
+	if (ctf_writer__init_data(cw))
+		goto err_cleanup;
+
+	/* CTF stream instance */
+	stream = bt_ctf_writer_create_stream(writer, stream_class);
+	if (!stream) {
+		pr("Failed to create CTF stream.\n");
+		goto err_cleanup;
+	}
+
+	cw->stream = stream;
+
+	/* CTF clock writer setup */
+	if (bt_ctf_writer_add_clock(writer, clock)) {
+		pr("Failed to assign CTF clock to writer.\n");
+		goto err_cleanup;
+	}
+
+	return 0;
+
+err_cleanup:
+	ctf_writer__cleanup(cw);
+err:
+	pr_err("Failed to setup CTF writer.\n");
+	return -1;
+}
+
+int bt_convert__perf2ctf(const char *input, const char *path, bool force)
+{
+	struct perf_session *session;
+	struct perf_data_file file = {
+		.path = input,
+		.mode = PERF_DATA_MODE_READ,
+		.force = force,
+	};
+	struct convert c = {
+		.tool = {
+			.sample          = process_sample_event,
+			.mmap            = perf_event__process_mmap,
+			.mmap2           = perf_event__process_mmap2,
+			.comm            = perf_event__process_comm,
+			.exit            = perf_event__process_exit,
+			.fork            = perf_event__process_fork,
+			.lost            = perf_event__process_lost,
+			.tracing_data    = perf_event__process_tracing_data,
+			.build_id        = perf_event__process_build_id,
+			.ordered_events  = true,
+			.ordering_requires_timestamps = true,
+		},
+	};
+	struct ctf_writer *cw = &c.writer;
+	int err = -1;
+
+	/* CTF writer */
+	if (ctf_writer__init(cw, path))
+		return -1;
+
+	/* perf.data session */
+	session = perf_session__new(&file, 0, &c.tool);
+	if (!session)
+		goto free_writer;
+
+	/* CTF writer env/clock setup  */
+	if (ctf_writer__setup_env(cw, session))
+		goto free_session;
+
+	/* CTF events setup */
+	if (setup_events(cw, session))
+		goto free_session;
+
+	err = perf_session__process_events(session);
+	if (!err)
+		err = bt_ctf_stream_flush(cw->stream);
+
+	fprintf(stderr,
+		"[ perf data convert: Converted '%s' into CTF data '%s' ]\n",
+		file.path, path);
+
+	fprintf(stderr,
+		"[ perf data convert: Converted and wrote %.3f MB (%" PRIu64 " samples) ]\n",
+		(double) c.events_size / 1024.0 / 1024.0,
+		c.events_count);
+
+	/* its all good */
+free_session:
+	perf_session__delete(session);
+
+free_writer:
+	ctf_writer__cleanup(cw);
+	return err;
+}
diff --git a/tools/perf/util/data-convert-bt.h b/tools/perf/util/data-convert-bt.h
new file mode 100644
index 000000000..4c204342a
--- /dev/null
+++ b/tools/perf/util/data-convert-bt.h
@@ -0,0 +1,8 @@
+#ifndef __DATA_CONVERT_BT_H
+#define __DATA_CONVERT_BT_H
+#ifdef HAVE_LIBBABELTRACE_SUPPORT
+
+int bt_convert__perf2ctf(const char *input_name, const char *to_ctf, bool force);
+
+#endif /* HAVE_LIBBABELTRACE_SUPPORT */
+#endif /* __DATA_CONVERT_BT_H */
diff --git a/tools/perf/util/data.c b/tools/perf/util/data.c
new file mode 100644
index 000000000..1921942fc
--- /dev/null
+++ b/tools/perf/util/data.c
@@ -0,0 +1,138 @@
+#include <linux/compiler.h>
+#include <linux/kernel.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+#include <string.h>
+
+#include "data.h"
+#include "util.h"
+#include "debug.h"
+
+static bool check_pipe(struct perf_data_file *file)
+{
+	struct stat st;
+	bool is_pipe = false;
+	int fd = perf_data_file__is_read(file) ?
+		 STDIN_FILENO : STDOUT_FILENO;
+
+	if (!file->path) {
+		if (!fstat(fd, &st) && S_ISFIFO(st.st_mode))
+			is_pipe = true;
+	} else {
+		if (!strcmp(file->path, "-"))
+			is_pipe = true;
+	}
+
+	if (is_pipe)
+		file->fd = fd;
+
+	return file->is_pipe = is_pipe;
+}
+
+static int check_backup(struct perf_data_file *file)
+{
+	struct stat st;
+
+	if (!stat(file->path, &st) && st.st_size) {
+		/* TODO check errors properly */
+		char oldname[PATH_MAX];
+		snprintf(oldname, sizeof(oldname), "%s.old",
+			 file->path);
+		unlink(oldname);
+		rename(file->path, oldname);
+	}
+
+	return 0;
+}
+
+static int open_file_read(struct perf_data_file *file)
+{
+	struct stat st;
+	int fd;
+	char sbuf[STRERR_BUFSIZE];
+
+	fd = open(file->path, O_RDONLY);
+	if (fd < 0) {
+		int err = errno;
+
+		pr_err("failed to open %s: %s", file->path,
+			strerror_r(err, sbuf, sizeof(sbuf)));
+		if (err == ENOENT && !strcmp(file->path, "perf.data"))
+			pr_err("  (try 'perf record' first)");
+		pr_err("\n");
+		return -err;
+	}
+
+	if (fstat(fd, &st) < 0)
+		goto out_close;
+
+	if (!file->force && st.st_uid && (st.st_uid != geteuid())) {
+		pr_err("File %s not owned by current user or root (use -f to override)\n",
+		       file->path);
+		goto out_close;
+	}
+
+	if (!st.st_size) {
+		pr_info("zero-sized file (%s), nothing to do!\n",
+			file->path);
+		goto out_close;
+	}
+
+	file->size = st.st_size;
+	return fd;
+
+ out_close:
+	close(fd);
+	return -1;
+}
+
+static int open_file_write(struct perf_data_file *file)
+{
+	int fd;
+	char sbuf[STRERR_BUFSIZE];
+
+	if (check_backup(file))
+		return -1;
+
+	fd = open(file->path, O_CREAT|O_RDWR|O_TRUNC, S_IRUSR|S_IWUSR);
+
+	if (fd < 0)
+		pr_err("failed to open %s : %s\n", file->path,
+			strerror_r(errno, sbuf, sizeof(sbuf)));
+
+	return fd;
+}
+
+static int open_file(struct perf_data_file *file)
+{
+	int fd;
+
+	fd = perf_data_file__is_read(file) ?
+	     open_file_read(file) : open_file_write(file);
+
+	file->fd = fd;
+	return fd < 0 ? -1 : 0;
+}
+
+int perf_data_file__open(struct perf_data_file *file)
+{
+	if (check_pipe(file))
+		return 0;
+
+	if (!file->path)
+		file->path = "perf.data";
+
+	return open_file(file);
+}
+
+void perf_data_file__close(struct perf_data_file *file)
+{
+	close(file->fd);
+}
+
+ssize_t perf_data_file__write(struct perf_data_file *file,
+			      void *buf, size_t size)
+{
+	return writen(file->fd, buf, size);
+}
diff --git a/tools/perf/util/data.h b/tools/perf/util/data.h
new file mode 100644
index 000000000..2b15d0c95
--- /dev/null
+++ b/tools/perf/util/data.h
@@ -0,0 +1,50 @@
+#ifndef __PERF_DATA_H
+#define __PERF_DATA_H
+
+#include <stdbool.h>
+
+enum perf_data_mode {
+	PERF_DATA_MODE_WRITE,
+	PERF_DATA_MODE_READ,
+};
+
+struct perf_data_file {
+	const char		*path;
+	int			 fd;
+	bool			 is_pipe;
+	bool			 force;
+	unsigned long		 size;
+	enum perf_data_mode	 mode;
+};
+
+static inline bool perf_data_file__is_read(struct perf_data_file *file)
+{
+	return file->mode == PERF_DATA_MODE_READ;
+}
+
+static inline bool perf_data_file__is_write(struct perf_data_file *file)
+{
+	return file->mode == PERF_DATA_MODE_WRITE;
+}
+
+static inline int perf_data_file__is_pipe(struct perf_data_file *file)
+{
+	return file->is_pipe;
+}
+
+static inline int perf_data_file__fd(struct perf_data_file *file)
+{
+	return file->fd;
+}
+
+static inline unsigned long perf_data_file__size(struct perf_data_file *file)
+{
+	return file->size;
+}
+
+int perf_data_file__open(struct perf_data_file *file);
+void perf_data_file__close(struct perf_data_file *file);
+ssize_t perf_data_file__write(struct perf_data_file *file,
+			      void *buf, size_t size);
+
+#endif /* __PERF_DATA_H */
diff --git a/tools/perf/util/db-export.c b/tools/perf/util/db-export.c
new file mode 100644
index 000000000..bb39a3ffc
--- /dev/null
+++ b/tools/perf/util/db-export.c
@@ -0,0 +1,428 @@
+/*
+ * db-export.c: Support for exporting data suitable for import to a database
+ * Copyright (c) 2014, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ */
+
+#include <errno.h>
+
+#include "evsel.h"
+#include "machine.h"
+#include "thread.h"
+#include "comm.h"
+#include "symbol.h"
+#include "event.h"
+#include "util.h"
+#include "thread-stack.h"
+#include "db-export.h"
+
+struct deferred_export {
+	struct list_head node;
+	struct comm *comm;
+};
+
+static int db_export__deferred(struct db_export *dbe)
+{
+	struct deferred_export *de;
+	int err;
+
+	while (!list_empty(&dbe->deferred)) {
+		de = list_entry(dbe->deferred.next, struct deferred_export,
+				node);
+		err = dbe->export_comm(dbe, de->comm);
+		list_del(&de->node);
+		free(de);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static void db_export__free_deferred(struct db_export *dbe)
+{
+	struct deferred_export *de;
+
+	while (!list_empty(&dbe->deferred)) {
+		de = list_entry(dbe->deferred.next, struct deferred_export,
+				node);
+		list_del(&de->node);
+		free(de);
+	}
+}
+
+static int db_export__defer_comm(struct db_export *dbe, struct comm *comm)
+{
+	struct deferred_export *de;
+
+	de = zalloc(sizeof(struct deferred_export));
+	if (!de)
+		return -ENOMEM;
+
+	de->comm = comm;
+	list_add_tail(&de->node, &dbe->deferred);
+
+	return 0;
+}
+
+int db_export__init(struct db_export *dbe)
+{
+	memset(dbe, 0, sizeof(struct db_export));
+	INIT_LIST_HEAD(&dbe->deferred);
+	return 0;
+}
+
+int db_export__flush(struct db_export *dbe)
+{
+	return db_export__deferred(dbe);
+}
+
+void db_export__exit(struct db_export *dbe)
+{
+	db_export__free_deferred(dbe);
+	call_return_processor__free(dbe->crp);
+	dbe->crp = NULL;
+}
+
+int db_export__evsel(struct db_export *dbe, struct perf_evsel *evsel)
+{
+	if (evsel->db_id)
+		return 0;
+
+	evsel->db_id = ++dbe->evsel_last_db_id;
+
+	if (dbe->export_evsel)
+		return dbe->export_evsel(dbe, evsel);
+
+	return 0;
+}
+
+int db_export__machine(struct db_export *dbe, struct machine *machine)
+{
+	if (machine->db_id)
+		return 0;
+
+	machine->db_id = ++dbe->machine_last_db_id;
+
+	if (dbe->export_machine)
+		return dbe->export_machine(dbe, machine);
+
+	return 0;
+}
+
+int db_export__thread(struct db_export *dbe, struct thread *thread,
+		      struct machine *machine, struct comm *comm)
+{
+	u64 main_thread_db_id = 0;
+	int err;
+
+	if (thread->db_id)
+		return 0;
+
+	thread->db_id = ++dbe->thread_last_db_id;
+
+	if (thread->pid_ != -1) {
+		struct thread *main_thread;
+
+		if (thread->pid_ == thread->tid) {
+			main_thread = thread;
+		} else {
+			main_thread = machine__findnew_thread(machine,
+							      thread->pid_,
+							      thread->pid_);
+			if (!main_thread)
+				return -ENOMEM;
+			err = db_export__thread(dbe, main_thread, machine,
+						comm);
+			if (err)
+				return err;
+			if (comm) {
+				err = db_export__comm_thread(dbe, comm, thread);
+				if (err)
+					return err;
+			}
+		}
+		main_thread_db_id = main_thread->db_id;
+	}
+
+	if (dbe->export_thread)
+		return dbe->export_thread(dbe, thread, main_thread_db_id,
+					  machine);
+
+	return 0;
+}
+
+int db_export__comm(struct db_export *dbe, struct comm *comm,
+		    struct thread *main_thread)
+{
+	int err;
+
+	if (comm->db_id)
+		return 0;
+
+	comm->db_id = ++dbe->comm_last_db_id;
+
+	if (dbe->export_comm) {
+		if (main_thread->comm_set)
+			err = dbe->export_comm(dbe, comm);
+		else
+			err = db_export__defer_comm(dbe, comm);
+		if (err)
+			return err;
+	}
+
+	return db_export__comm_thread(dbe, comm, main_thread);
+}
+
+int db_export__comm_thread(struct db_export *dbe, struct comm *comm,
+			   struct thread *thread)
+{
+	u64 db_id;
+
+	db_id = ++dbe->comm_thread_last_db_id;
+
+	if (dbe->export_comm_thread)
+		return dbe->export_comm_thread(dbe, db_id, comm, thread);
+
+	return 0;
+}
+
+int db_export__dso(struct db_export *dbe, struct dso *dso,
+		   struct machine *machine)
+{
+	if (dso->db_id)
+		return 0;
+
+	dso->db_id = ++dbe->dso_last_db_id;
+
+	if (dbe->export_dso)
+		return dbe->export_dso(dbe, dso, machine);
+
+	return 0;
+}
+
+int db_export__symbol(struct db_export *dbe, struct symbol *sym,
+		      struct dso *dso)
+{
+	u64 *sym_db_id = symbol__priv(sym);
+
+	if (*sym_db_id)
+		return 0;
+
+	*sym_db_id = ++dbe->symbol_last_db_id;
+
+	if (dbe->export_symbol)
+		return dbe->export_symbol(dbe, sym, dso);
+
+	return 0;
+}
+
+static struct thread *get_main_thread(struct machine *machine, struct thread *thread)
+{
+	if (thread->pid_ == thread->tid)
+		return thread;
+
+	if (thread->pid_ == -1)
+		return NULL;
+
+	return machine__find_thread(machine, thread->pid_, thread->pid_);
+}
+
+static int db_ids_from_al(struct db_export *dbe, struct addr_location *al,
+			  u64 *dso_db_id, u64 *sym_db_id, u64 *offset)
+{
+	int err;
+
+	if (al->map) {
+		struct dso *dso = al->map->dso;
+
+		err = db_export__dso(dbe, dso, al->machine);
+		if (err)
+			return err;
+		*dso_db_id = dso->db_id;
+
+		if (!al->sym) {
+			al->sym = symbol__new(al->addr, 0, 0, "unknown");
+			if (al->sym)
+				symbols__insert(&dso->symbols[al->map->type],
+						al->sym);
+		}
+
+		if (al->sym) {
+			u64 *db_id = symbol__priv(al->sym);
+
+			err = db_export__symbol(dbe, al->sym, dso);
+			if (err)
+				return err;
+			*sym_db_id = *db_id;
+			*offset = al->addr - al->sym->start;
+		}
+	}
+
+	return 0;
+}
+
+int db_export__branch_type(struct db_export *dbe, u32 branch_type,
+			   const char *name)
+{
+	if (dbe->export_branch_type)
+		return dbe->export_branch_type(dbe, branch_type, name);
+
+	return 0;
+}
+
+int db_export__sample(struct db_export *dbe, union perf_event *event,
+		      struct perf_sample *sample, struct perf_evsel *evsel,
+		      struct addr_location *al)
+{
+	struct thread* thread = al->thread;
+	struct export_sample es = {
+		.event = event,
+		.sample = sample,
+		.evsel = evsel,
+		.al = al,
+	};
+	struct thread *main_thread;
+	struct comm *comm = NULL;
+	int err;
+
+	err = db_export__evsel(dbe, evsel);
+	if (err)
+		return err;
+
+	err = db_export__machine(dbe, al->machine);
+	if (err)
+		return err;
+
+	main_thread = get_main_thread(al->machine, thread);
+	if (main_thread)
+		comm = machine__thread_exec_comm(al->machine, main_thread);
+
+	err = db_export__thread(dbe, thread, al->machine, comm);
+	if (err)
+		return err;
+
+	if (comm) {
+		err = db_export__comm(dbe, comm, main_thread);
+		if (err)
+			return err;
+		es.comm_db_id = comm->db_id;
+	}
+
+	es.db_id = ++dbe->sample_last_db_id;
+
+	err = db_ids_from_al(dbe, al, &es.dso_db_id, &es.sym_db_id, &es.offset);
+	if (err)
+		return err;
+
+	if ((evsel->attr.sample_type & PERF_SAMPLE_ADDR) &&
+	    sample_addr_correlates_sym(&evsel->attr)) {
+		struct addr_location addr_al;
+
+		perf_event__preprocess_sample_addr(event, sample, thread, &addr_al);
+		err = db_ids_from_al(dbe, &addr_al, &es.addr_dso_db_id,
+				     &es.addr_sym_db_id, &es.addr_offset);
+		if (err)
+			return err;
+		if (dbe->crp) {
+			err = thread_stack__process(thread, comm, sample, al,
+						    &addr_al, es.db_id,
+						    dbe->crp);
+			if (err)
+				return err;
+		}
+	}
+
+	if (dbe->export_sample)
+		return dbe->export_sample(dbe, &es);
+
+	return 0;
+}
+
+static struct {
+	u32 branch_type;
+	const char *name;
+} branch_types[] = {
+	{0, "no branch"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL, "call"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_RETURN, "return"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CONDITIONAL, "conditional jump"},
+	{PERF_IP_FLAG_BRANCH, "unconditional jump"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL | PERF_IP_FLAG_INTERRUPT,
+	 "software interrupt"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_RETURN | PERF_IP_FLAG_INTERRUPT,
+	 "return from interrupt"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL | PERF_IP_FLAG_SYSCALLRET,
+	 "system call"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_RETURN | PERF_IP_FLAG_SYSCALLRET,
+	 "return from system call"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_ASYNC, "asynchronous branch"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CALL | PERF_IP_FLAG_ASYNC |
+	 PERF_IP_FLAG_INTERRUPT, "hardware interrupt"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_TX_ABORT, "transaction abort"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_TRACE_BEGIN, "trace begin"},
+	{PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_TRACE_END, "trace end"},
+	{0, NULL}
+};
+
+int db_export__branch_types(struct db_export *dbe)
+{
+	int i, err = 0;
+
+	for (i = 0; branch_types[i].name ; i++) {
+		err = db_export__branch_type(dbe, branch_types[i].branch_type,
+					     branch_types[i].name);
+		if (err)
+			break;
+	}
+	return err;
+}
+
+int db_export__call_path(struct db_export *dbe, struct call_path *cp)
+{
+	int err;
+
+	if (cp->db_id)
+		return 0;
+
+	if (cp->parent) {
+		err = db_export__call_path(dbe, cp->parent);
+		if (err)
+			return err;
+	}
+
+	cp->db_id = ++dbe->call_path_last_db_id;
+
+	if (dbe->export_call_path)
+		return dbe->export_call_path(dbe, cp);
+
+	return 0;
+}
+
+int db_export__call_return(struct db_export *dbe, struct call_return *cr)
+{
+	int err;
+
+	if (cr->db_id)
+		return 0;
+
+	err = db_export__call_path(dbe, cr->cp);
+	if (err)
+		return err;
+
+	cr->db_id = ++dbe->call_return_last_db_id;
+
+	if (dbe->export_call_return)
+		return dbe->export_call_return(dbe, cr);
+
+	return 0;
+}
diff --git a/tools/perf/util/db-export.h b/tools/perf/util/db-export.h
new file mode 100644
index 000000000..25e22fd76
--- /dev/null
+++ b/tools/perf/util/db-export.h
@@ -0,0 +1,106 @@
+/*
+ * db-export.h: Support for exporting data suitable for import to a database
+ * Copyright (c) 2014, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ */
+
+#ifndef __PERF_DB_EXPORT_H
+#define __PERF_DB_EXPORT_H
+
+#include <linux/types.h>
+#include <linux/list.h>
+
+struct perf_evsel;
+struct machine;
+struct thread;
+struct comm;
+struct dso;
+struct perf_sample;
+struct addr_location;
+struct call_return_processor;
+struct call_path;
+struct call_return;
+
+struct export_sample {
+	union perf_event	*event;
+	struct perf_sample	*sample;
+	struct perf_evsel	*evsel;
+	struct addr_location	*al;
+	u64			db_id;
+	u64			comm_db_id;
+	u64			dso_db_id;
+	u64			sym_db_id;
+	u64			offset; /* ip offset from symbol start */
+	u64			addr_dso_db_id;
+	u64			addr_sym_db_id;
+	u64			addr_offset; /* addr offset from symbol start */
+};
+
+struct db_export {
+	int (*export_evsel)(struct db_export *dbe, struct perf_evsel *evsel);
+	int (*export_machine)(struct db_export *dbe, struct machine *machine);
+	int (*export_thread)(struct db_export *dbe, struct thread *thread,
+			     u64 main_thread_db_id, struct machine *machine);
+	int (*export_comm)(struct db_export *dbe, struct comm *comm);
+	int (*export_comm_thread)(struct db_export *dbe, u64 db_id,
+				  struct comm *comm, struct thread *thread);
+	int (*export_dso)(struct db_export *dbe, struct dso *dso,
+			  struct machine *machine);
+	int (*export_symbol)(struct db_export *dbe, struct symbol *sym,
+			     struct dso *dso);
+	int (*export_branch_type)(struct db_export *dbe, u32 branch_type,
+				  const char *name);
+	int (*export_sample)(struct db_export *dbe, struct export_sample *es);
+	int (*export_call_path)(struct db_export *dbe, struct call_path *cp);
+	int (*export_call_return)(struct db_export *dbe,
+				  struct call_return *cr);
+	struct call_return_processor *crp;
+	u64 evsel_last_db_id;
+	u64 machine_last_db_id;
+	u64 thread_last_db_id;
+	u64 comm_last_db_id;
+	u64 comm_thread_last_db_id;
+	u64 dso_last_db_id;
+	u64 symbol_last_db_id;
+	u64 sample_last_db_id;
+	u64 call_path_last_db_id;
+	u64 call_return_last_db_id;
+	struct list_head deferred;
+};
+
+int db_export__init(struct db_export *dbe);
+int db_export__flush(struct db_export *dbe);
+void db_export__exit(struct db_export *dbe);
+int db_export__evsel(struct db_export *dbe, struct perf_evsel *evsel);
+int db_export__machine(struct db_export *dbe, struct machine *machine);
+int db_export__thread(struct db_export *dbe, struct thread *thread,
+		      struct machine *machine, struct comm *comm);
+int db_export__comm(struct db_export *dbe, struct comm *comm,
+		    struct thread *main_thread);
+int db_export__comm_thread(struct db_export *dbe, struct comm *comm,
+			   struct thread *thread);
+int db_export__dso(struct db_export *dbe, struct dso *dso,
+		   struct machine *machine);
+int db_export__symbol(struct db_export *dbe, struct symbol *sym,
+		      struct dso *dso);
+int db_export__branch_type(struct db_export *dbe, u32 branch_type,
+			   const char *name);
+int db_export__sample(struct db_export *dbe, union perf_event *event,
+		      struct perf_sample *sample, struct perf_evsel *evsel,
+		      struct addr_location *al);
+
+int db_export__branch_types(struct db_export *dbe);
+
+int db_export__call_path(struct db_export *dbe, struct call_path *cp);
+int db_export__call_return(struct db_export *dbe, struct call_return *cr);
+
+#endif
diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c
new file mode 100644
index 000000000..2da5581ec
--- /dev/null
+++ b/tools/perf/util/debug.c
@@ -0,0 +1,189 @@
+/* For general debugging purposes */
+
+#include "../perf.h"
+
+#include <string.h>
+#include <stdarg.h>
+#include <stdio.h>
+
+#include "cache.h"
+#include "color.h"
+#include "event.h"
+#include "debug.h"
+#include "util.h"
+#include "target.h"
+
+#define NSECS_PER_SEC  1000000000ULL
+#define NSECS_PER_USEC 1000ULL
+
+int verbose;
+bool dump_trace = false, quiet = false;
+int debug_ordered_events;
+static int redirect_to_stderr;
+int debug_data_convert;
+
+static int _eprintf(int level, int var, const char *fmt, va_list args)
+{
+	int ret = 0;
+
+	if (var >= level) {
+		if (use_browser >= 1 && !redirect_to_stderr)
+			ui_helpline__vshow(fmt, args);
+		else
+			ret = vfprintf(stderr, fmt, args);
+	}
+
+	return ret;
+}
+
+int eprintf(int level, int var, const char *fmt, ...)
+{
+	va_list args;
+	int ret;
+
+	va_start(args, fmt);
+	ret = _eprintf(level, var, fmt, args);
+	va_end(args);
+
+	return ret;
+}
+
+static int __eprintf_time(u64 t, const char *fmt, va_list args)
+{
+	int ret = 0;
+	u64 secs, usecs, nsecs = t;
+
+	secs   = nsecs / NSECS_PER_SEC;
+	nsecs -= secs  * NSECS_PER_SEC;
+	usecs  = nsecs / NSECS_PER_USEC;
+
+	ret = fprintf(stderr, "[%13" PRIu64 ".%06" PRIu64 "] ",
+		      secs, usecs);
+	ret += vfprintf(stderr, fmt, args);
+	return ret;
+}
+
+int eprintf_time(int level, int var, u64 t, const char *fmt, ...)
+{
+	int ret = 0;
+	va_list args;
+
+	if (var >= level) {
+		va_start(args, fmt);
+		ret = __eprintf_time(t, fmt, args);
+		va_end(args);
+	}
+
+	return ret;
+}
+
+/*
+ * Overloading libtraceevent standard info print
+ * function, display with -v in perf.
+ */
+void pr_stat(const char *fmt, ...)
+{
+	va_list args;
+
+	va_start(args, fmt);
+	_eprintf(1, verbose, fmt, args);
+	va_end(args);
+	eprintf(1, verbose, "\n");
+}
+
+int dump_printf(const char *fmt, ...)
+{
+	va_list args;
+	int ret = 0;
+
+	if (dump_trace) {
+		va_start(args, fmt);
+		ret = vprintf(fmt, args);
+		va_end(args);
+	}
+
+	return ret;
+}
+
+void trace_event(union perf_event *event)
+{
+	unsigned char *raw_event = (void *)event;
+	const char *color = PERF_COLOR_BLUE;
+	int i, j;
+
+	if (!dump_trace)
+		return;
+
+	printf(".");
+	color_fprintf(stdout, color, "\n. ... raw event: size %d bytes\n",
+		      event->header.size);
+
+	for (i = 0; i < event->header.size; i++) {
+		if ((i & 15) == 0) {
+			printf(".");
+			color_fprintf(stdout, color, "  %04x: ", i);
+		}
+
+		color_fprintf(stdout, color, " %02x", raw_event[i]);
+
+		if (((i & 15) == 15) || i == event->header.size-1) {
+			color_fprintf(stdout, color, "  ");
+			for (j = 0; j < 15-(i & 15); j++)
+				color_fprintf(stdout, color, "   ");
+			for (j = i & ~15; j <= i; j++) {
+				color_fprintf(stdout, color, "%c",
+					      isprint(raw_event[j]) ?
+					      raw_event[j] : '.');
+			}
+			color_fprintf(stdout, color, "\n");
+		}
+	}
+	printf(".\n");
+}
+
+static struct debug_variable {
+	const char *name;
+	int *ptr;
+} debug_variables[] = {
+	{ .name = "verbose",		.ptr = &verbose },
+	{ .name = "ordered-events",	.ptr = &debug_ordered_events},
+	{ .name = "stderr",		.ptr = &redirect_to_stderr},
+	{ .name = "data-convert",	.ptr = &debug_data_convert },
+	{ .name = NULL, }
+};
+
+int perf_debug_option(const char *str)
+{
+	struct debug_variable *var = &debug_variables[0];
+	char *vstr, *s = strdup(str);
+	int v = 1;
+
+	vstr = strchr(s, '=');
+	if (vstr)
+		*vstr++ = 0;
+
+	while (var->name) {
+		if (!strcmp(s, var->name))
+			break;
+		var++;
+	}
+
+	if (!var->name) {
+		pr_err("Unknown debug variable name '%s'\n", s);
+		free(s);
+		return -1;
+	}
+
+	if (vstr) {
+		v = atoi(vstr);
+		/*
+		 * Allow only values in range (0, 10),
+		 * otherwise set 0.
+		 */
+		v = (v < 0) || (v > 10) ? 0 : v;
+	}
+
+	*var->ptr = v;
+	free(s);
+	return 0;
+}
diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h
new file mode 100644
index 000000000..caac2fdc6
--- /dev/null
+++ b/tools/perf/util/debug.h
@@ -0,0 +1,56 @@
+/* For debugging general purposes */
+#ifndef __PERF_DEBUG_H
+#define __PERF_DEBUG_H
+
+#include <stdbool.h>
+#include <string.h>
+#include "event.h"
+#include "../ui/helpline.h"
+#include "../ui/progress.h"
+#include "../ui/util.h"
+
+extern int verbose;
+extern bool quiet, dump_trace;
+extern int debug_ordered_events;
+extern int debug_data_convert;
+
+#ifndef pr_fmt
+#define pr_fmt(fmt) fmt
+#endif
+
+#define pr_err(fmt, ...) \
+	eprintf(0, verbose, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_warning(fmt, ...) \
+	eprintf(0, verbose, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_info(fmt, ...) \
+	eprintf(0, verbose, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug(fmt, ...) \
+	eprintf(1, verbose, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debugN(n, fmt, ...) \
+	eprintf(n, verbose, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug2(fmt, ...) pr_debugN(2, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug3(fmt, ...) pr_debugN(3, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_debug4(fmt, ...) pr_debugN(4, pr_fmt(fmt), ##__VA_ARGS__)
+
+#define pr_time_N(n, var, t, fmt, ...) \
+	eprintf_time(n, var, t, fmt, ##__VA_ARGS__)
+
+#define pr_oe_time(t, fmt, ...)  pr_time_N(1, debug_ordered_events, t, pr_fmt(fmt), ##__VA_ARGS__)
+#define pr_oe_time2(t, fmt, ...) pr_time_N(2, debug_ordered_events, t, pr_fmt(fmt), ##__VA_ARGS__)
+
+#define STRERR_BUFSIZE	128	/* For the buffer size of strerror_r */
+
+int dump_printf(const char *fmt, ...) __attribute__((format(printf, 1, 2)));
+void trace_event(union perf_event *event);
+
+int ui__error(const char *format, ...) __attribute__((format(printf, 1, 2)));
+int ui__warning(const char *format, ...) __attribute__((format(printf, 1, 2)));
+
+void pr_stat(const char *fmt, ...);
+
+int eprintf(int level, int var, const char *fmt, ...) __attribute__((format(printf, 3, 4)));
+int eprintf_time(int level, int var, u64 t, const char *fmt, ...) __attribute__((format(printf, 4, 5)));
+
+int perf_debug_option(const char *str);
+
+#endif	/* __PERF_DEBUG_H */
diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c
new file mode 100644
index 000000000..fc0ddd579
--- /dev/null
+++ b/tools/perf/util/dso.c
@@ -0,0 +1,1172 @@
+#include <asm/bug.h>
+#include <sys/time.h>
+#include <sys/resource.h>
+#include "symbol.h"
+#include "dso.h"
+#include "machine.h"
+#include "util.h"
+#include "debug.h"
+
+char dso__symtab_origin(const struct dso *dso)
+{
+	static const char origin[] = {
+		[DSO_BINARY_TYPE__KALLSYMS]			= 'k',
+		[DSO_BINARY_TYPE__VMLINUX]			= 'v',
+		[DSO_BINARY_TYPE__JAVA_JIT]			= 'j',
+		[DSO_BINARY_TYPE__DEBUGLINK]			= 'l',
+		[DSO_BINARY_TYPE__BUILD_ID_CACHE]		= 'B',
+		[DSO_BINARY_TYPE__FEDORA_DEBUGINFO]		= 'f',
+		[DSO_BINARY_TYPE__UBUNTU_DEBUGINFO]		= 'u',
+		[DSO_BINARY_TYPE__OPENEMBEDDED_DEBUGINFO]	= 'o',
+		[DSO_BINARY_TYPE__BUILDID_DEBUGINFO]		= 'b',
+		[DSO_BINARY_TYPE__SYSTEM_PATH_DSO]		= 'd',
+		[DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE]		= 'K',
+		[DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE_COMP]	= 'm',
+		[DSO_BINARY_TYPE__GUEST_KALLSYMS]		= 'g',
+		[DSO_BINARY_TYPE__GUEST_KMODULE]		= 'G',
+		[DSO_BINARY_TYPE__GUEST_KMODULE_COMP]		= 'M',
+		[DSO_BINARY_TYPE__GUEST_VMLINUX]		= 'V',
+	};
+
+	if (dso == NULL || dso->symtab_type == DSO_BINARY_TYPE__NOT_FOUND)
+		return '!';
+	return origin[dso->symtab_type];
+}
+
+int dso__read_binary_type_filename(const struct dso *dso,
+				   enum dso_binary_type type,
+				   char *root_dir, char *filename, size_t size)
+{
+	char build_id_hex[BUILD_ID_SIZE * 2 + 1];
+	int ret = 0;
+	size_t len;
+
+	switch (type) {
+	case DSO_BINARY_TYPE__DEBUGLINK: {
+		char *debuglink;
+
+		len = __symbol__join_symfs(filename, size, dso->long_name);
+		debuglink = filename + len;
+		while (debuglink != filename && *debuglink != '/')
+			debuglink--;
+		if (*debuglink == '/')
+			debuglink++;
+		ret = filename__read_debuglink(filename, debuglink,
+					       size - (debuglink - filename));
+		}
+		break;
+	case DSO_BINARY_TYPE__BUILD_ID_CACHE:
+		/* skip the locally configured cache if a symfs is given */
+		if (symbol_conf.symfs[0] ||
+		    (dso__build_id_filename(dso, filename, size) == NULL))
+			ret = -1;
+		break;
+
+	case DSO_BINARY_TYPE__FEDORA_DEBUGINFO:
+		len = __symbol__join_symfs(filename, size, "/usr/lib/debug");
+		snprintf(filename + len, size - len, "%s.debug", dso->long_name);
+		break;
+
+	case DSO_BINARY_TYPE__UBUNTU_DEBUGINFO:
+		len = __symbol__join_symfs(filename, size, "/usr/lib/debug");
+		snprintf(filename + len, size - len, "%s", dso->long_name);
+		break;
+
+	case DSO_BINARY_TYPE__OPENEMBEDDED_DEBUGINFO:
+	{
+		const char *last_slash;
+		size_t dir_size;
+
+		last_slash = dso->long_name + dso->long_name_len;
+		while (last_slash != dso->long_name && *last_slash != '/')
+			last_slash--;
+
+		len = __symbol__join_symfs(filename, size, "");
+		dir_size = last_slash - dso->long_name + 2;
+		if (dir_size > (size - len)) {
+			ret = -1;
+			break;
+		}
+		len += scnprintf(filename + len, dir_size, "%s",  dso->long_name);
+		len += scnprintf(filename + len , size - len, ".debug%s",
+								last_slash);
+		break;
+	}
+
+	case DSO_BINARY_TYPE__BUILDID_DEBUGINFO:
+		if (!dso->has_build_id) {
+			ret = -1;
+			break;
+		}
+
+		build_id__sprintf(dso->build_id,
+				  sizeof(dso->build_id),
+				  build_id_hex);
+		len = __symbol__join_symfs(filename, size, "/usr/lib/debug/.build-id/");
+		snprintf(filename + len, size - len, "%.2s/%s.debug",
+			 build_id_hex, build_id_hex + 2);
+		break;
+
+	case DSO_BINARY_TYPE__VMLINUX:
+	case DSO_BINARY_TYPE__GUEST_VMLINUX:
+	case DSO_BINARY_TYPE__SYSTEM_PATH_DSO:
+		__symbol__join_symfs(filename, size, dso->long_name);
+		break;
+
+	case DSO_BINARY_TYPE__GUEST_KMODULE:
+	case DSO_BINARY_TYPE__GUEST_KMODULE_COMP:
+		path__join3(filename, size, symbol_conf.symfs,
+			    root_dir, dso->long_name);
+		break;
+
+	case DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE:
+	case DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE_COMP:
+		__symbol__join_symfs(filename, size, dso->long_name);
+		break;
+
+	case DSO_BINARY_TYPE__KCORE:
+	case DSO_BINARY_TYPE__GUEST_KCORE:
+		snprintf(filename, size, "%s", dso->long_name);
+		break;
+
+	default:
+	case DSO_BINARY_TYPE__KALLSYMS:
+	case DSO_BINARY_TYPE__GUEST_KALLSYMS:
+	case DSO_BINARY_TYPE__JAVA_JIT:
+	case DSO_BINARY_TYPE__NOT_FOUND:
+		ret = -1;
+		break;
+	}
+
+	return ret;
+}
+
+static const struct {
+	const char *fmt;
+	int (*decompress)(const char *input, int output);
+} compressions[] = {
+#ifdef HAVE_ZLIB_SUPPORT
+	{ "gz", gzip_decompress_to_file },
+#endif
+#ifdef HAVE_LZMA_SUPPORT
+	{ "xz", lzma_decompress_to_file },
+#endif
+	{ NULL, NULL },
+};
+
+bool is_supported_compression(const char *ext)
+{
+	unsigned i;
+
+	for (i = 0; compressions[i].fmt; i++) {
+		if (!strcmp(ext, compressions[i].fmt))
+			return true;
+	}
+	return false;
+}
+
+bool is_kernel_module(const char *pathname)
+{
+	struct kmod_path m;
+
+	if (kmod_path__parse(&m, pathname))
+		return NULL;
+
+	return m.kmod;
+}
+
+bool decompress_to_file(const char *ext, const char *filename, int output_fd)
+{
+	unsigned i;
+
+	for (i = 0; compressions[i].fmt; i++) {
+		if (!strcmp(ext, compressions[i].fmt))
+			return !compressions[i].decompress(filename,
+							   output_fd);
+	}
+	return false;
+}
+
+bool dso__needs_decompress(struct dso *dso)
+{
+	return dso->symtab_type == DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE_COMP ||
+		dso->symtab_type == DSO_BINARY_TYPE__GUEST_KMODULE_COMP;
+}
+
+/*
+ * Parses kernel module specified in @path and updates
+ * @m argument like:
+ *
+ *    @comp - true if @path contains supported compression suffix,
+ *            false otherwise
+ *    @kmod - true if @path contains '.ko' suffix in right position,
+ *            false otherwise
+ *    @name - if (@alloc_name && @kmod) is true, it contains strdup-ed base name
+ *            of the kernel module without suffixes, otherwise strudup-ed
+ *            base name of @path
+ *    @ext  - if (@alloc_ext && @comp) is true, it contains strdup-ed string
+ *            the compression suffix
+ *
+ * Returns 0 if there's no strdup error, -ENOMEM otherwise.
+ */
+int __kmod_path__parse(struct kmod_path *m, const char *path,
+		       bool alloc_name, bool alloc_ext)
+{
+	const char *name = strrchr(path, '/');
+	const char *ext  = strrchr(path, '.');
+
+	memset(m, 0x0, sizeof(*m));
+	name = name ? name + 1 : path;
+
+	/* No extension, just return name. */
+	if (ext == NULL) {
+		if (alloc_name) {
+			m->name = strdup(name);
+			return m->name ? 0 : -ENOMEM;
+		}
+		return 0;
+	}
+
+	if (is_supported_compression(ext + 1)) {
+		m->comp = true;
+		ext -= 3;
+	}
+
+	/* Check .ko extension only if there's enough name left. */
+	if (ext > name)
+		m->kmod = !strncmp(ext, ".ko", 3);
+
+	if (alloc_name) {
+		if (m->kmod) {
+			if (asprintf(&m->name, "[%.*s]", (int) (ext - name), name) == -1)
+				return -ENOMEM;
+		} else {
+			if (asprintf(&m->name, "%s", name) == -1)
+				return -ENOMEM;
+		}
+
+		strxfrchar(m->name, '-', '_');
+	}
+
+	if (alloc_ext && m->comp) {
+		m->ext = strdup(ext + 4);
+		if (!m->ext) {
+			free((void *) m->name);
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * Global list of open DSOs and the counter.
+ */
+static LIST_HEAD(dso__data_open);
+static long dso__data_open_cnt;
+
+static void dso__list_add(struct dso *dso)
+{
+	list_add_tail(&dso->data.open_entry, &dso__data_open);
+	dso__data_open_cnt++;
+}
+
+static void dso__list_del(struct dso *dso)
+{
+	list_del(&dso->data.open_entry);
+	WARN_ONCE(dso__data_open_cnt <= 0,
+		  "DSO data fd counter out of bounds.");
+	dso__data_open_cnt--;
+}
+
+static void close_first_dso(void);
+
+static int do_open(char *name)
+{
+	int fd;
+	char sbuf[STRERR_BUFSIZE];
+
+	do {
+		fd = open(name, O_RDONLY);
+		if (fd >= 0)
+			return fd;
+
+		pr_debug("dso open failed: %s\n",
+			 strerror_r(errno, sbuf, sizeof(sbuf)));
+		if (!dso__data_open_cnt || errno != EMFILE)
+			break;
+
+		close_first_dso();
+	} while (1);
+
+	return -1;
+}
+
+static int __open_dso(struct dso *dso, struct machine *machine)
+{
+	int fd;
+	char *root_dir = (char *)"";
+	char *name = malloc(PATH_MAX);
+
+	if (!name)
+		return -ENOMEM;
+
+	if (machine)
+		root_dir = machine->root_dir;
+
+	if (dso__read_binary_type_filename(dso, dso->binary_type,
+					    root_dir, name, PATH_MAX)) {
+		free(name);
+		return -EINVAL;
+	}
+
+	fd = do_open(name);
+	free(name);
+	return fd;
+}
+
+static void check_data_close(void);
+
+/**
+ * dso_close - Open DSO data file
+ * @dso: dso object
+ *
+ * Open @dso's data file descriptor and updates
+ * list/count of open DSO objects.
+ */
+static int open_dso(struct dso *dso, struct machine *machine)
+{
+	int fd = __open_dso(dso, machine);
+
+	if (fd >= 0) {
+		dso__list_add(dso);
+		/*
+		 * Check if we crossed the allowed number
+		 * of opened DSOs and close one if needed.
+		 */
+		check_data_close();
+	}
+
+	return fd;
+}
+
+static void close_data_fd(struct dso *dso)
+{
+	if (dso->data.fd >= 0) {
+		close(dso->data.fd);
+		dso->data.fd = -1;
+		dso->data.file_size = 0;
+		dso__list_del(dso);
+	}
+}
+
+/**
+ * dso_close - Close DSO data file
+ * @dso: dso object
+ *
+ * Close @dso's data file descriptor and updates
+ * list/count of open DSO objects.
+ */
+static void close_dso(struct dso *dso)
+{
+	close_data_fd(dso);
+}
+
+static void close_first_dso(void)
+{
+	struct dso *dso;
+
+	dso = list_first_entry(&dso__data_open, struct dso, data.open_entry);
+	close_dso(dso);
+}
+
+static rlim_t get_fd_limit(void)
+{
+	struct rlimit l;
+	rlim_t limit = 0;
+
+	/* Allow half of the current open fd limit. */
+	if (getrlimit(RLIMIT_NOFILE, &l) == 0) {
+		if (l.rlim_cur == RLIM_INFINITY)
+			limit = l.rlim_cur;
+		else
+			limit = l.rlim_cur / 2;
+	} else {
+		pr_err("failed to get fd limit\n");
+		limit = 1;
+	}
+
+	return limit;
+}
+
+static bool may_cache_fd(void)
+{
+	static rlim_t limit;
+
+	if (!limit)
+		limit = get_fd_limit();
+
+	if (limit == RLIM_INFINITY)
+		return true;
+
+	return limit > (rlim_t) dso__data_open_cnt;
+}
+
+/*
+ * Check and close LRU dso if we crossed allowed limit
+ * for opened dso file descriptors. The limit is half
+ * of the RLIMIT_NOFILE files opened.
+*/
+static void check_data_close(void)
+{
+	bool cache_fd = may_cache_fd();
+
+	if (!cache_fd)
+		close_first_dso();
+}
+
+/**
+ * dso__data_close - Close DSO data file
+ * @dso: dso object
+ *
+ * External interface to close @dso's data file descriptor.
+ */
+void dso__data_close(struct dso *dso)
+{
+	close_dso(dso);
+}
+
+/**
+ * dso__data_fd - Get dso's data file descriptor
+ * @dso: dso object
+ * @machine: machine object
+ *
+ * External interface to find dso's file, open it and
+ * returns file descriptor.
+ */
+int dso__data_fd(struct dso *dso, struct machine *machine)
+{
+	enum dso_binary_type binary_type_data[] = {
+		DSO_BINARY_TYPE__BUILD_ID_CACHE,
+		DSO_BINARY_TYPE__SYSTEM_PATH_DSO,
+		DSO_BINARY_TYPE__NOT_FOUND,
+	};
+	int i = 0;
+
+	if (dso->data.status == DSO_DATA_STATUS_ERROR)
+		return -1;
+
+	if (dso->data.fd >= 0)
+		goto out;
+
+	if (dso->binary_type != DSO_BINARY_TYPE__NOT_FOUND) {
+		dso->data.fd = open_dso(dso, machine);
+		goto out;
+	}
+
+	do {
+		dso->binary_type = binary_type_data[i++];
+
+		dso->data.fd = open_dso(dso, machine);
+		if (dso->data.fd >= 0)
+			goto out;
+
+	} while (dso->binary_type != DSO_BINARY_TYPE__NOT_FOUND);
+out:
+	if (dso->data.fd >= 0)
+		dso->data.status = DSO_DATA_STATUS_OK;
+	else
+		dso->data.status = DSO_DATA_STATUS_ERROR;
+
+	return dso->data.fd;
+}
+
+bool dso__data_status_seen(struct dso *dso, enum dso_data_status_seen by)
+{
+	u32 flag = 1 << by;
+
+	if (dso->data.status_seen & flag)
+		return true;
+
+	dso->data.status_seen |= flag;
+
+	return false;
+}
+
+static void
+dso_cache__free(struct rb_root *root)
+{
+	struct rb_node *next = rb_first(root);
+
+	while (next) {
+		struct dso_cache *cache;
+
+		cache = rb_entry(next, struct dso_cache, rb_node);
+		next = rb_next(&cache->rb_node);
+		rb_erase(&cache->rb_node, root);
+		free(cache);
+	}
+}
+
+static struct dso_cache *dso_cache__find(const struct rb_root *root, u64 offset)
+{
+	struct rb_node * const *p = &root->rb_node;
+	const struct rb_node *parent = NULL;
+	struct dso_cache *cache;
+
+	while (*p != NULL) {
+		u64 end;
+
+		parent = *p;
+		cache = rb_entry(parent, struct dso_cache, rb_node);
+		end = cache->offset + DSO__DATA_CACHE_SIZE;
+
+		if (offset < cache->offset)
+			p = &(*p)->rb_left;
+		else if (offset >= end)
+			p = &(*p)->rb_right;
+		else
+			return cache;
+	}
+	return NULL;
+}
+
+static void
+dso_cache__insert(struct rb_root *root, struct dso_cache *new)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct dso_cache *cache;
+	u64 offset = new->offset;
+
+	while (*p != NULL) {
+		u64 end;
+
+		parent = *p;
+		cache = rb_entry(parent, struct dso_cache, rb_node);
+		end = cache->offset + DSO__DATA_CACHE_SIZE;
+
+		if (offset < cache->offset)
+			p = &(*p)->rb_left;
+		else if (offset >= end)
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&new->rb_node, parent, p);
+	rb_insert_color(&new->rb_node, root);
+}
+
+static ssize_t
+dso_cache__memcpy(struct dso_cache *cache, u64 offset,
+		  u8 *data, u64 size)
+{
+	u64 cache_offset = offset - cache->offset;
+	u64 cache_size   = min(cache->size - cache_offset, size);
+
+	memcpy(data, cache->data + cache_offset, cache_size);
+	return cache_size;
+}
+
+static ssize_t
+dso_cache__read(struct dso *dso, u64 offset, u8 *data, ssize_t size)
+{
+	struct dso_cache *cache;
+	ssize_t ret;
+
+	do {
+		u64 cache_offset;
+
+		ret = -ENOMEM;
+
+		cache = zalloc(sizeof(*cache) + DSO__DATA_CACHE_SIZE);
+		if (!cache)
+			break;
+
+		cache_offset = offset & DSO__DATA_CACHE_MASK;
+
+		ret = pread(dso->data.fd, cache->data, DSO__DATA_CACHE_SIZE, cache_offset);
+		if (ret <= 0)
+			break;
+
+		cache->offset = cache_offset;
+		cache->size   = ret;
+		dso_cache__insert(&dso->data.cache, cache);
+
+		ret = dso_cache__memcpy(cache, offset, data, size);
+
+	} while (0);
+
+	if (ret <= 0)
+		free(cache);
+
+	return ret;
+}
+
+static ssize_t dso_cache_read(struct dso *dso, u64 offset,
+			      u8 *data, ssize_t size)
+{
+	struct dso_cache *cache;
+
+	cache = dso_cache__find(&dso->data.cache, offset);
+	if (cache)
+		return dso_cache__memcpy(cache, offset, data, size);
+	else
+		return dso_cache__read(dso, offset, data, size);
+}
+
+/*
+ * Reads and caches dso data DSO__DATA_CACHE_SIZE size chunks
+ * in the rb_tree. Any read to already cached data is served
+ * by cached data.
+ */
+static ssize_t cached_read(struct dso *dso, u64 offset, u8 *data, ssize_t size)
+{
+	ssize_t r = 0;
+	u8 *p = data;
+
+	do {
+		ssize_t ret;
+
+		ret = dso_cache_read(dso, offset, p, size);
+		if (ret < 0)
+			return ret;
+
+		/* Reached EOF, return what we have. */
+		if (!ret)
+			break;
+
+		BUG_ON(ret > size);
+
+		r      += ret;
+		p      += ret;
+		offset += ret;
+		size   -= ret;
+
+	} while (size);
+
+	return r;
+}
+
+static int data_file_size(struct dso *dso)
+{
+	struct stat st;
+	char sbuf[STRERR_BUFSIZE];
+
+	if (!dso->data.file_size) {
+		if (fstat(dso->data.fd, &st)) {
+			pr_err("dso mmap failed, fstat: %s\n",
+				strerror_r(errno, sbuf, sizeof(sbuf)));
+			return -1;
+		}
+		dso->data.file_size = st.st_size;
+	}
+
+	return 0;
+}
+
+/**
+ * dso__data_size - Return dso data size
+ * @dso: dso object
+ * @machine: machine object
+ *
+ * Return: dso data size
+ */
+off_t dso__data_size(struct dso *dso, struct machine *machine)
+{
+	int fd;
+
+	fd = dso__data_fd(dso, machine);
+	if (fd < 0)
+		return fd;
+
+	if (data_file_size(dso))
+		return -1;
+
+	/* For now just estimate dso data size is close to file size */
+	return dso->data.file_size;
+}
+
+static ssize_t data_read_offset(struct dso *dso, u64 offset,
+				u8 *data, ssize_t size)
+{
+	if (data_file_size(dso))
+		return -1;
+
+	/* Check the offset sanity. */
+	if (offset > dso->data.file_size)
+		return -1;
+
+	if (offset + size < offset)
+		return -1;
+
+	return cached_read(dso, offset, data, size);
+}
+
+/**
+ * dso__data_read_offset - Read data from dso file offset
+ * @dso: dso object
+ * @machine: machine object
+ * @offset: file offset
+ * @data: buffer to store data
+ * @size: size of the @data buffer
+ *
+ * External interface to read data from dso file offset. Open
+ * dso data file and use cached_read to get the data.
+ */
+ssize_t dso__data_read_offset(struct dso *dso, struct machine *machine,
+			      u64 offset, u8 *data, ssize_t size)
+{
+	if (dso__data_fd(dso, machine) < 0)
+		return -1;
+
+	return data_read_offset(dso, offset, data, size);
+}
+
+/**
+ * dso__data_read_addr - Read data from dso address
+ * @dso: dso object
+ * @machine: machine object
+ * @add: virtual memory address
+ * @data: buffer to store data
+ * @size: size of the @data buffer
+ *
+ * External interface to read data from dso address.
+ */
+ssize_t dso__data_read_addr(struct dso *dso, struct map *map,
+			    struct machine *machine, u64 addr,
+			    u8 *data, ssize_t size)
+{
+	u64 offset = map->map_ip(map, addr);
+	return dso__data_read_offset(dso, machine, offset, data, size);
+}
+
+struct map *dso__new_map(const char *name)
+{
+	struct map *map = NULL;
+	struct dso *dso = dso__new(name);
+
+	if (dso)
+		map = map__new2(0, dso, MAP__FUNCTION);
+
+	return map;
+}
+
+struct dso *dso__kernel_findnew(struct machine *machine, const char *name,
+		    const char *short_name, int dso_type)
+{
+	/*
+	 * The kernel dso could be created by build_id processing.
+	 */
+	struct dso *dso = __dsos__findnew(&machine->kernel_dsos, name);
+
+	/*
+	 * We need to run this in all cases, since during the build_id
+	 * processing we had no idea this was the kernel dso.
+	 */
+	if (dso != NULL) {
+		dso__set_short_name(dso, short_name, false);
+		dso->kernel = dso_type;
+	}
+
+	return dso;
+}
+
+/*
+ * Find a matching entry and/or link current entry to RB tree.
+ * Either one of the dso or name parameter must be non-NULL or the
+ * function will not work.
+ */
+static struct dso *dso__findlink_by_longname(struct rb_root *root,
+					     struct dso *dso, const char *name)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node  *parent = NULL;
+
+	if (!name)
+		name = dso->long_name;
+	/*
+	 * Find node with the matching name
+	 */
+	while (*p) {
+		struct dso *this = rb_entry(*p, struct dso, rb_node);
+		int rc = strcmp(name, this->long_name);
+
+		parent = *p;
+		if (rc == 0) {
+			/*
+			 * In case the new DSO is a duplicate of an existing
+			 * one, print an one-time warning & put the new entry
+			 * at the end of the list of duplicates.
+			 */
+			if (!dso || (dso == this))
+				return this;	/* Find matching dso */
+			/*
+			 * The core kernel DSOs may have duplicated long name.
+			 * In this case, the short name should be different.
+			 * Comparing the short names to differentiate the DSOs.
+			 */
+			rc = strcmp(dso->short_name, this->short_name);
+			if (rc == 0) {
+				pr_err("Duplicated dso name: %s\n", name);
+				return NULL;
+			}
+		}
+		if (rc < 0)
+			p = &parent->rb_left;
+		else
+			p = &parent->rb_right;
+	}
+	if (dso) {
+		/* Add new node and rebalance tree */
+		rb_link_node(&dso->rb_node, parent, p);
+		rb_insert_color(&dso->rb_node, root);
+	}
+	return NULL;
+}
+
+static inline struct dso *
+dso__find_by_longname(const struct rb_root *root, const char *name)
+{
+	return dso__findlink_by_longname((struct rb_root *)root, NULL, name);
+}
+
+void dso__set_long_name(struct dso *dso, const char *name, bool name_allocated)
+{
+	if (name == NULL)
+		return;
+
+	if (dso->long_name_allocated)
+		free((char *)dso->long_name);
+
+	dso->long_name		 = name;
+	dso->long_name_len	 = strlen(name);
+	dso->long_name_allocated = name_allocated;
+}
+
+void dso__set_short_name(struct dso *dso, const char *name, bool name_allocated)
+{
+	if (name == NULL)
+		return;
+
+	if (dso->short_name_allocated)
+		free((char *)dso->short_name);
+
+	dso->short_name		  = name;
+	dso->short_name_len	  = strlen(name);
+	dso->short_name_allocated = name_allocated;
+}
+
+static void dso__set_basename(struct dso *dso)
+{
+       /*
+        * basename() may modify path buffer, so we must pass
+        * a copy.
+        */
+       char *base, *lname = strdup(dso->long_name);
+
+       if (!lname)
+               return;
+
+       /*
+        * basename() may return a pointer to internal
+        * storage which is reused in subsequent calls
+        * so copy the result.
+        */
+       base = strdup(basename(lname));
+
+       free(lname);
+
+       if (!base)
+               return;
+
+       dso__set_short_name(dso, base, true);
+}
+
+int dso__name_len(const struct dso *dso)
+{
+	if (!dso)
+		return strlen("[unknown]");
+	if (verbose)
+		return dso->long_name_len;
+
+	return dso->short_name_len;
+}
+
+bool dso__loaded(const struct dso *dso, enum map_type type)
+{
+	return dso->loaded & (1 << type);
+}
+
+bool dso__sorted_by_name(const struct dso *dso, enum map_type type)
+{
+	return dso->sorted_by_name & (1 << type);
+}
+
+void dso__set_sorted_by_name(struct dso *dso, enum map_type type)
+{
+	dso->sorted_by_name |= (1 << type);
+}
+
+struct dso *dso__new(const char *name)
+{
+	struct dso *dso = calloc(1, sizeof(*dso) + strlen(name) + 1);
+
+	if (dso != NULL) {
+		int i;
+		strcpy(dso->name, name);
+		dso__set_long_name(dso, dso->name, false);
+		dso__set_short_name(dso, dso->name, false);
+		for (i = 0; i < MAP__NR_TYPES; ++i)
+			dso->symbols[i] = dso->symbol_names[i] = RB_ROOT;
+		dso->data.cache = RB_ROOT;
+		dso->data.fd = -1;
+		dso->data.status = DSO_DATA_STATUS_UNKNOWN;
+		dso->symtab_type = DSO_BINARY_TYPE__NOT_FOUND;
+		dso->binary_type = DSO_BINARY_TYPE__NOT_FOUND;
+		dso->is_64_bit = (sizeof(void *) == 8);
+		dso->loaded = 0;
+		dso->rel = 0;
+		dso->sorted_by_name = 0;
+		dso->has_build_id = 0;
+		dso->has_srcline = 1;
+		dso->a2l_fails = 1;
+		dso->kernel = DSO_TYPE_USER;
+		dso->needs_swap = DSO_SWAP__UNSET;
+		RB_CLEAR_NODE(&dso->rb_node);
+		INIT_LIST_HEAD(&dso->node);
+		INIT_LIST_HEAD(&dso->data.open_entry);
+	}
+
+	return dso;
+}
+
+void dso__delete(struct dso *dso)
+{
+	int i;
+
+	if (!RB_EMPTY_NODE(&dso->rb_node))
+		pr_err("DSO %s is still in rbtree when being deleted!\n",
+		       dso->long_name);
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		symbols__delete(&dso->symbols[i]);
+
+	if (dso->short_name_allocated) {
+		zfree((char **)&dso->short_name);
+		dso->short_name_allocated = false;
+	}
+
+	if (dso->long_name_allocated) {
+		zfree((char **)&dso->long_name);
+		dso->long_name_allocated = false;
+	}
+
+	dso__data_close(dso);
+	dso_cache__free(&dso->data.cache);
+	dso__free_a2l(dso);
+	zfree(&dso->symsrc_filename);
+	free(dso);
+}
+
+void dso__set_build_id(struct dso *dso, void *build_id)
+{
+	memcpy(dso->build_id, build_id, sizeof(dso->build_id));
+	dso->has_build_id = 1;
+}
+
+bool dso__build_id_equal(const struct dso *dso, u8 *build_id)
+{
+	return memcmp(dso->build_id, build_id, sizeof(dso->build_id)) == 0;
+}
+
+void dso__read_running_kernel_build_id(struct dso *dso, struct machine *machine)
+{
+	char path[PATH_MAX];
+
+	if (machine__is_default_guest(machine))
+		return;
+	sprintf(path, "%s/sys/kernel/notes", machine->root_dir);
+	if (sysfs__read_build_id(path, dso->build_id,
+				 sizeof(dso->build_id)) == 0)
+		dso->has_build_id = true;
+}
+
+int dso__kernel_module_get_build_id(struct dso *dso,
+				    const char *root_dir)
+{
+	char filename[PATH_MAX];
+	/*
+	 * kernel module short names are of the form "[module]" and
+	 * we need just "module" here.
+	 */
+	const char *name = dso->short_name + 1;
+
+	snprintf(filename, sizeof(filename),
+		 "%s/sys/module/%.*s/notes/.note.gnu.build-id",
+		 root_dir, (int)strlen(name) - 1, name);
+
+	if (sysfs__read_build_id(filename, dso->build_id,
+				 sizeof(dso->build_id)) == 0)
+		dso->has_build_id = true;
+
+	return 0;
+}
+
+bool __dsos__read_build_ids(struct list_head *head, bool with_hits)
+{
+	bool have_build_id = false;
+	struct dso *pos;
+
+	list_for_each_entry(pos, head, node) {
+		if (with_hits && !pos->hit)
+			continue;
+		if (pos->has_build_id) {
+			have_build_id = true;
+			continue;
+		}
+		if (filename__read_build_id(pos->long_name, pos->build_id,
+					    sizeof(pos->build_id)) > 0) {
+			have_build_id	  = true;
+			pos->has_build_id = true;
+		}
+	}
+
+	return have_build_id;
+}
+
+void dsos__add(struct dsos *dsos, struct dso *dso)
+{
+	list_add_tail(&dso->node, &dsos->head);
+	dso__findlink_by_longname(&dsos->root, dso, NULL);
+}
+
+struct dso *dsos__find(const struct dsos *dsos, const char *name,
+		       bool cmp_short)
+{
+	struct dso *pos;
+
+	if (cmp_short) {
+		list_for_each_entry(pos, &dsos->head, node)
+			if (strcmp(pos->short_name, name) == 0)
+				return pos;
+		return NULL;
+	}
+	return dso__find_by_longname(&dsos->root, name);
+}
+
+struct dso *dsos__addnew(struct dsos *dsos, const char *name)
+{
+	struct dso *dso = dso__new(name);
+
+	if (dso != NULL) {
+		dsos__add(dsos, dso);
+		dso__set_basename(dso);
+	}
+	return dso;
+}
+
+struct dso *__dsos__findnew(struct dsos *dsos, const char *name)
+{
+	struct dso *dso = dsos__find(dsos, name, false);
+
+	return dso ? dso : dsos__addnew(dsos, name);
+}
+
+size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp,
+			       bool (skip)(struct dso *dso, int parm), int parm)
+{
+	struct dso *pos;
+	size_t ret = 0;
+
+	list_for_each_entry(pos, head, node) {
+		if (skip && skip(pos, parm))
+			continue;
+		ret += dso__fprintf_buildid(pos, fp);
+		ret += fprintf(fp, " %s\n", pos->long_name);
+	}
+	return ret;
+}
+
+size_t __dsos__fprintf(struct list_head *head, FILE *fp)
+{
+	struct dso *pos;
+	size_t ret = 0;
+
+	list_for_each_entry(pos, head, node) {
+		int i;
+		for (i = 0; i < MAP__NR_TYPES; ++i)
+			ret += dso__fprintf(pos, i, fp);
+	}
+
+	return ret;
+}
+
+size_t dso__fprintf_buildid(struct dso *dso, FILE *fp)
+{
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+	build_id__sprintf(dso->build_id, sizeof(dso->build_id), sbuild_id);
+	return fprintf(fp, "%s", sbuild_id);
+}
+
+size_t dso__fprintf(struct dso *dso, enum map_type type, FILE *fp)
+{
+	struct rb_node *nd;
+	size_t ret = fprintf(fp, "dso: %s (", dso->short_name);
+
+	if (dso->short_name != dso->long_name)
+		ret += fprintf(fp, "%s, ", dso->long_name);
+	ret += fprintf(fp, "%s, %sloaded, ", map_type__name[type],
+		       dso__loaded(dso, type) ? "" : "NOT ");
+	ret += dso__fprintf_buildid(dso, fp);
+	ret += fprintf(fp, ")\n");
+	for (nd = rb_first(&dso->symbols[type]); nd; nd = rb_next(nd)) {
+		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+		ret += symbol__fprintf(pos, fp);
+	}
+
+	return ret;
+}
+
+enum dso_type dso__type(struct dso *dso, struct machine *machine)
+{
+	int fd;
+
+	fd = dso__data_fd(dso, machine);
+	if (fd < 0)
+		return DSO__TYPE_UNKNOWN;
+
+	return dso__type_fd(fd);
+}
+
+int dso__strerror_load(struct dso *dso, char *buf, size_t buflen)
+{
+	int idx, errnum = dso->load_errno;
+	/*
+	 * This must have a same ordering as the enum dso_load_errno.
+	 */
+	static const char *dso_load__error_str[] = {
+	"Internal tools/perf/ library error",
+	"Invalid ELF file",
+	"Can not read build id",
+	"Mismatching build id",
+	"Decompression failure",
+	};
+
+	BUG_ON(buflen == 0);
+
+	if (errnum >= 0) {
+		const char *err = strerror_r(errnum, buf, buflen);
+
+		if (err != buf)
+			scnprintf(buf, buflen, "%s", err);
+
+		return 0;
+	}
+
+	if (errnum <  __DSO_LOAD_ERRNO__START || errnum >= __DSO_LOAD_ERRNO__END)
+		return -1;
+
+	idx = errnum - __DSO_LOAD_ERRNO__START;
+	scnprintf(buf, buflen, "%s", dso_load__error_str[idx]);
+	return 0;
+}
diff --git a/tools/perf/util/dso.h b/tools/perf/util/dso.h
new file mode 100644
index 000000000..e0901b4ed
--- /dev/null
+++ b/tools/perf/util/dso.h
@@ -0,0 +1,325 @@
+#ifndef __PERF_DSO
+#define __PERF_DSO
+
+#include <linux/types.h>
+#include <linux/rbtree.h>
+#include <stdbool.h>
+#include <linux/types.h>
+#include <linux/bitops.h>
+#include "map.h"
+#include "build-id.h"
+
+enum dso_binary_type {
+	DSO_BINARY_TYPE__KALLSYMS = 0,
+	DSO_BINARY_TYPE__GUEST_KALLSYMS,
+	DSO_BINARY_TYPE__VMLINUX,
+	DSO_BINARY_TYPE__GUEST_VMLINUX,
+	DSO_BINARY_TYPE__JAVA_JIT,
+	DSO_BINARY_TYPE__DEBUGLINK,
+	DSO_BINARY_TYPE__BUILD_ID_CACHE,
+	DSO_BINARY_TYPE__FEDORA_DEBUGINFO,
+	DSO_BINARY_TYPE__UBUNTU_DEBUGINFO,
+	DSO_BINARY_TYPE__BUILDID_DEBUGINFO,
+	DSO_BINARY_TYPE__SYSTEM_PATH_DSO,
+	DSO_BINARY_TYPE__GUEST_KMODULE,
+	DSO_BINARY_TYPE__GUEST_KMODULE_COMP,
+	DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE,
+	DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE_COMP,
+	DSO_BINARY_TYPE__KCORE,
+	DSO_BINARY_TYPE__GUEST_KCORE,
+	DSO_BINARY_TYPE__OPENEMBEDDED_DEBUGINFO,
+	DSO_BINARY_TYPE__NOT_FOUND,
+};
+
+enum dso_kernel_type {
+	DSO_TYPE_USER = 0,
+	DSO_TYPE_KERNEL,
+	DSO_TYPE_GUEST_KERNEL
+};
+
+enum dso_swap_type {
+	DSO_SWAP__UNSET,
+	DSO_SWAP__NO,
+	DSO_SWAP__YES,
+};
+
+enum dso_data_status {
+	DSO_DATA_STATUS_ERROR	= -1,
+	DSO_DATA_STATUS_UNKNOWN	= 0,
+	DSO_DATA_STATUS_OK	= 1,
+};
+
+enum dso_data_status_seen {
+	DSO_DATA_STATUS_SEEN_ITRACE,
+};
+
+enum dso_type {
+	DSO__TYPE_UNKNOWN,
+	DSO__TYPE_64BIT,
+	DSO__TYPE_32BIT,
+	DSO__TYPE_X32BIT,
+};
+
+enum dso_load_errno {
+	DSO_LOAD_ERRNO__SUCCESS		= 0,
+
+	/*
+	 * Choose an arbitrary negative big number not to clash with standard
+	 * errno since SUS requires the errno has distinct positive values.
+	 * See 'Issue 6' in the link below.
+	 *
+	 * http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/errno.h.html
+	 */
+	__DSO_LOAD_ERRNO__START		= -10000,
+
+	DSO_LOAD_ERRNO__INTERNAL_ERROR	= __DSO_LOAD_ERRNO__START,
+
+	/* for symsrc__init() */
+	DSO_LOAD_ERRNO__INVALID_ELF,
+	DSO_LOAD_ERRNO__CANNOT_READ_BUILDID,
+	DSO_LOAD_ERRNO__MISMATCHING_BUILDID,
+
+	/* for decompress_kmodule */
+	DSO_LOAD_ERRNO__DECOMPRESSION_FAILURE,
+
+	__DSO_LOAD_ERRNO__END,
+};
+
+#define DSO__SWAP(dso, type, val)			\
+({							\
+	type ____r = val;				\
+	BUG_ON(dso->needs_swap == DSO_SWAP__UNSET);	\
+	if (dso->needs_swap == DSO_SWAP__YES) {		\
+		switch (sizeof(____r)) {		\
+		case 2:					\
+			____r = bswap_16(val);		\
+			break;				\
+		case 4:					\
+			____r = bswap_32(val);		\
+			break;				\
+		case 8:					\
+			____r = bswap_64(val);		\
+			break;				\
+		default:				\
+			BUG_ON(1);			\
+		}					\
+	}						\
+	____r;						\
+})
+
+#define DSO__DATA_CACHE_SIZE 4096
+#define DSO__DATA_CACHE_MASK ~(DSO__DATA_CACHE_SIZE - 1)
+
+struct dso_cache {
+	struct rb_node	rb_node;
+	u64 offset;
+	u64 size;
+	char data[0];
+};
+
+/*
+ * DSOs are put into both a list for fast iteration and rbtree for fast
+ * long name lookup.
+ */
+struct dsos {
+	struct list_head head;
+	struct rb_root	 root;	/* rbtree root sorted by long name */
+};
+
+struct dso {
+	struct list_head node;
+	struct rb_node	 rb_node;	/* rbtree node sorted by long name */
+	struct rb_root	 symbols[MAP__NR_TYPES];
+	struct rb_root	 symbol_names[MAP__NR_TYPES];
+	void		 *a2l;
+	char		 *symsrc_filename;
+	unsigned int	 a2l_fails;
+	enum dso_kernel_type	kernel;
+	enum dso_swap_type	needs_swap;
+	enum dso_binary_type	symtab_type;
+	enum dso_binary_type	binary_type;
+	enum dso_load_errno	load_errno;
+	u8		 adjust_symbols:1;
+	u8		 has_build_id:1;
+	u8		 has_srcline:1;
+	u8		 hit:1;
+	u8		 annotate_warned:1;
+	u8		 short_name_allocated:1;
+	u8		 long_name_allocated:1;
+	u8		 is_64_bit:1;
+	u8		 sorted_by_name;
+	u8		 loaded;
+	u8		 rel;
+	u8		 build_id[BUILD_ID_SIZE];
+	const char	 *short_name;
+	const char	 *long_name;
+	u16		 long_name_len;
+	u16		 short_name_len;
+	void		*dwfl;			/* DWARF debug info */
+
+	/* dso data file */
+	struct {
+		struct rb_root	 cache;
+		int		 fd;
+		int		 status;
+		u32		 status_seen;
+		size_t		 file_size;
+		struct list_head open_entry;
+		u64		 debug_frame_offset;
+		u64		 eh_frame_hdr_offset;
+	} data;
+
+	union { /* Tool specific area */
+		void	 *priv;
+		u64	 db_id;
+	};
+
+	char		 name[0];
+};
+
+/* dso__for_each_symbol - iterate over the symbols of given type
+ *
+ * @dso: the 'struct dso *' in which symbols itereated
+ * @pos: the 'struct symbol *' to use as a loop cursor
+ * @n: the 'struct rb_node *' to use as a temporary storage
+ * @type: the 'enum map_type' type of symbols
+ */
+#define dso__for_each_symbol(dso, pos, n, type)	\
+	symbols__for_each_entry(&(dso)->symbols[(type)], pos, n)
+
+static inline void dso__set_loaded(struct dso *dso, enum map_type type)
+{
+	dso->loaded |= (1 << type);
+}
+
+struct dso *dso__new(const char *name);
+void dso__delete(struct dso *dso);
+
+void dso__set_short_name(struct dso *dso, const char *name, bool name_allocated);
+void dso__set_long_name(struct dso *dso, const char *name, bool name_allocated);
+
+int dso__name_len(const struct dso *dso);
+
+bool dso__loaded(const struct dso *dso, enum map_type type);
+
+bool dso__sorted_by_name(const struct dso *dso, enum map_type type);
+void dso__set_sorted_by_name(struct dso *dso, enum map_type type);
+void dso__sort_by_name(struct dso *dso, enum map_type type);
+
+void dso__set_build_id(struct dso *dso, void *build_id);
+bool dso__build_id_equal(const struct dso *dso, u8 *build_id);
+void dso__read_running_kernel_build_id(struct dso *dso,
+				       struct machine *machine);
+int dso__kernel_module_get_build_id(struct dso *dso, const char *root_dir);
+
+char dso__symtab_origin(const struct dso *dso);
+int dso__read_binary_type_filename(const struct dso *dso, enum dso_binary_type type,
+				   char *root_dir, char *filename, size_t size);
+bool is_supported_compression(const char *ext);
+bool is_kernel_module(const char *pathname);
+bool decompress_to_file(const char *ext, const char *filename, int output_fd);
+bool dso__needs_decompress(struct dso *dso);
+
+struct kmod_path {
+	char *name;
+	char *ext;
+	bool  comp;
+	bool  kmod;
+};
+
+int __kmod_path__parse(struct kmod_path *m, const char *path,
+		     bool alloc_name, bool alloc_ext);
+
+#define kmod_path__parse(__m, __p)      __kmod_path__parse(__m, __p, false, false)
+#define kmod_path__parse_name(__m, __p) __kmod_path__parse(__m, __p, true , false)
+#define kmod_path__parse_ext(__m, __p)  __kmod_path__parse(__m, __p, false, true)
+
+/*
+ * The dso__data_* external interface provides following functions:
+ *   dso__data_fd
+ *   dso__data_close
+ *   dso__data_size
+ *   dso__data_read_offset
+ *   dso__data_read_addr
+ *
+ * Please refer to the dso.c object code for each function and
+ * arguments documentation. Following text tries to explain the
+ * dso file descriptor caching.
+ *
+ * The dso__data* interface allows caching of opened file descriptors
+ * to speed up the dso data accesses. The idea is to leave the file
+ * descriptor opened ideally for the whole life of the dso object.
+ *
+ * The current usage of the dso__data_* interface is as follows:
+ *
+ * Get DSO's fd:
+ *   int fd = dso__data_fd(dso, machine);
+ *   USE 'fd' SOMEHOW
+ *
+ * Read DSO's data:
+ *   n = dso__data_read_offset(dso_0, &machine, 0, buf, BUFSIZE);
+ *   n = dso__data_read_addr(dso_0, &machine, 0, buf, BUFSIZE);
+ *
+ * Eventually close DSO's fd:
+ *   dso__data_close(dso);
+ *
+ * It is not necessary to close the DSO object data file. Each time new
+ * DSO data file is opened, the limit (RLIMIT_NOFILE/2) is checked. Once
+ * it is crossed, the oldest opened DSO object is closed.
+ *
+ * The dso__delete function calls close_dso function to ensure the
+ * data file descriptor gets closed/unmapped before the dso object
+ * is freed.
+ *
+ * TODO
+*/
+int dso__data_fd(struct dso *dso, struct machine *machine);
+void dso__data_close(struct dso *dso);
+
+off_t dso__data_size(struct dso *dso, struct machine *machine);
+ssize_t dso__data_read_offset(struct dso *dso, struct machine *machine,
+			      u64 offset, u8 *data, ssize_t size);
+ssize_t dso__data_read_addr(struct dso *dso, struct map *map,
+			    struct machine *machine, u64 addr,
+			    u8 *data, ssize_t size);
+bool dso__data_status_seen(struct dso *dso, enum dso_data_status_seen by);
+
+struct map *dso__new_map(const char *name);
+struct dso *dso__kernel_findnew(struct machine *machine, const char *name,
+				const char *short_name, int dso_type);
+
+void dsos__add(struct dsos *dsos, struct dso *dso);
+struct dso *dsos__addnew(struct dsos *dsos, const char *name);
+struct dso *dsos__find(const struct dsos *dsos, const char *name,
+		       bool cmp_short);
+struct dso *__dsos__findnew(struct dsos *dsos, const char *name);
+bool __dsos__read_build_ids(struct list_head *head, bool with_hits);
+
+size_t __dsos__fprintf_buildid(struct list_head *head, FILE *fp,
+			       bool (skip)(struct dso *dso, int parm), int parm);
+size_t __dsos__fprintf(struct list_head *head, FILE *fp);
+
+size_t dso__fprintf_buildid(struct dso *dso, FILE *fp);
+size_t dso__fprintf_symbols_by_name(struct dso *dso,
+				    enum map_type type, FILE *fp);
+size_t dso__fprintf(struct dso *dso, enum map_type type, FILE *fp);
+
+static inline bool dso__is_vmlinux(struct dso *dso)
+{
+	return dso->binary_type == DSO_BINARY_TYPE__VMLINUX ||
+	       dso->binary_type == DSO_BINARY_TYPE__GUEST_VMLINUX;
+}
+
+static inline bool dso__is_kcore(struct dso *dso)
+{
+	return dso->binary_type == DSO_BINARY_TYPE__KCORE ||
+	       dso->binary_type == DSO_BINARY_TYPE__GUEST_KCORE;
+}
+
+void dso__free_a2l(struct dso *dso);
+
+enum dso_type dso__type(struct dso *dso, struct machine *machine);
+
+int dso__strerror_load(struct dso *dso, char *buf, size_t buflen);
+
+#endif /* __PERF_DSO */
diff --git a/tools/perf/util/dwarf-aux.c b/tools/perf/util/dwarf-aux.c
new file mode 100644
index 000000000..c34e02402
--- /dev/null
+++ b/tools/perf/util/dwarf-aux.c
@@ -0,0 +1,905 @@
+/*
+ * dwarf-aux.c : libdw auxiliary interfaces
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <stdbool.h>
+#include "util.h"
+#include "debug.h"
+#include "dwarf-aux.h"
+
+/**
+ * cu_find_realpath - Find the realpath of the target file
+ * @cu_die: A DIE(dwarf information entry) of CU(compilation Unit)
+ * @fname:  The tail filename of the target file
+ *
+ * Find the real(long) path of @fname in @cu_die.
+ */
+const char *cu_find_realpath(Dwarf_Die *cu_die, const char *fname)
+{
+	Dwarf_Files *files;
+	size_t nfiles, i;
+	const char *src = NULL;
+	int ret;
+
+	if (!fname)
+		return NULL;
+
+	ret = dwarf_getsrcfiles(cu_die, &files, &nfiles);
+	if (ret != 0)
+		return NULL;
+
+	for (i = 0; i < nfiles; i++) {
+		src = dwarf_filesrc(files, i, NULL, NULL);
+		if (strtailcmp(src, fname) == 0)
+			break;
+	}
+	if (i == nfiles)
+		return NULL;
+	return src;
+}
+
+/**
+ * cu_get_comp_dir - Get the path of compilation directory
+ * @cu_die: a CU DIE
+ *
+ * Get the path of compilation directory of given @cu_die.
+ * Since this depends on DW_AT_comp_dir, older gcc will not
+ * embedded it. In that case, this returns NULL.
+ */
+const char *cu_get_comp_dir(Dwarf_Die *cu_die)
+{
+	Dwarf_Attribute attr;
+	if (dwarf_attr(cu_die, DW_AT_comp_dir, &attr) == NULL)
+		return NULL;
+	return dwarf_formstring(&attr);
+}
+
+/**
+ * cu_find_lineinfo - Get a line number and file name for given address
+ * @cu_die: a CU DIE
+ * @addr: An address
+ * @fname: a pointer which returns the file name string
+ * @lineno: a pointer which returns the line number
+ *
+ * Find a line number and file name for @addr in @cu_die.
+ */
+int cu_find_lineinfo(Dwarf_Die *cu_die, unsigned long addr,
+		    const char **fname, int *lineno)
+{
+	Dwarf_Line *line;
+	Dwarf_Addr laddr;
+
+	line = dwarf_getsrc_die(cu_die, (Dwarf_Addr)addr);
+	if (line && dwarf_lineaddr(line, &laddr) == 0 &&
+	    addr == (unsigned long)laddr && dwarf_lineno(line, lineno) == 0) {
+		*fname = dwarf_linesrc(line, NULL, NULL);
+		if (!*fname)
+			/* line number is useless without filename */
+			*lineno = 0;
+	}
+
+	return *lineno ?: -ENOENT;
+}
+
+static int __die_find_inline_cb(Dwarf_Die *die_mem, void *data);
+
+/**
+ * cu_walk_functions_at - Walk on function DIEs at given address
+ * @cu_die: A CU DIE
+ * @addr: An address
+ * @callback: A callback which called with found DIEs
+ * @data: A user data
+ *
+ * Walk on function DIEs at given @addr in @cu_die. Passed DIEs
+ * should be subprogram or inlined-subroutines.
+ */
+int cu_walk_functions_at(Dwarf_Die *cu_die, Dwarf_Addr addr,
+		    int (*callback)(Dwarf_Die *, void *), void *data)
+{
+	Dwarf_Die die_mem;
+	Dwarf_Die *sc_die;
+	int ret = -ENOENT;
+
+	/* Inlined function could be recursive. Trace it until fail */
+	for (sc_die = die_find_realfunc(cu_die, addr, &die_mem);
+	     sc_die != NULL;
+	     sc_die = die_find_child(sc_die, __die_find_inline_cb, &addr,
+				     &die_mem)) {
+		ret = callback(sc_die, data);
+		if (ret)
+			break;
+	}
+
+	return ret;
+
+}
+
+/**
+ * die_compare_name - Compare diename and tname
+ * @dw_die: a DIE
+ * @tname: a string of target name
+ *
+ * Compare the name of @dw_die and @tname. Return false if @dw_die has no name.
+ */
+bool die_compare_name(Dwarf_Die *dw_die, const char *tname)
+{
+	const char *name;
+	name = dwarf_diename(dw_die);
+	return name ? (strcmp(tname, name) == 0) : false;
+}
+
+/**
+ * die_get_call_lineno - Get callsite line number of inline-function instance
+ * @in_die: a DIE of an inlined function instance
+ *
+ * Get call-site line number of @in_die. This means from where the inline
+ * function is called.
+ */
+int die_get_call_lineno(Dwarf_Die *in_die)
+{
+	Dwarf_Attribute attr;
+	Dwarf_Word ret;
+
+	if (!dwarf_attr(in_die, DW_AT_call_line, &attr))
+		return -ENOENT;
+
+	dwarf_formudata(&attr, &ret);
+	return (int)ret;
+}
+
+/**
+ * die_get_type - Get type DIE
+ * @vr_die: a DIE of a variable
+ * @die_mem: where to store a type DIE
+ *
+ * Get a DIE of the type of given variable (@vr_die), and store
+ * it to die_mem. Return NULL if fails to get a type DIE.
+ */
+Dwarf_Die *die_get_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
+{
+	Dwarf_Attribute attr;
+
+	if (dwarf_attr_integrate(vr_die, DW_AT_type, &attr) &&
+	    dwarf_formref_die(&attr, die_mem))
+		return die_mem;
+	else
+		return NULL;
+}
+
+/* Get a type die, but skip qualifiers */
+static Dwarf_Die *__die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
+{
+	int tag;
+
+	do {
+		vr_die = die_get_type(vr_die, die_mem);
+		if (!vr_die)
+			break;
+		tag = dwarf_tag(vr_die);
+	} while (tag == DW_TAG_const_type ||
+		 tag == DW_TAG_restrict_type ||
+		 tag == DW_TAG_volatile_type ||
+		 tag == DW_TAG_shared_type);
+
+	return vr_die;
+}
+
+/**
+ * die_get_real_type - Get a type die, but skip qualifiers and typedef
+ * @vr_die: a DIE of a variable
+ * @die_mem: where to store a type DIE
+ *
+ * Get a DIE of the type of given variable (@vr_die), and store
+ * it to die_mem. Return NULL if fails to get a type DIE.
+ * If the type is qualifiers (e.g. const) or typedef, this skips it
+ * and tries to find real type (structure or basic types, e.g. int).
+ */
+Dwarf_Die *die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
+{
+	do {
+		vr_die = __die_get_real_type(vr_die, die_mem);
+	} while (vr_die && dwarf_tag(vr_die) == DW_TAG_typedef);
+
+	return vr_die;
+}
+
+/* Get attribute and translate it as a udata */
+static int die_get_attr_udata(Dwarf_Die *tp_die, unsigned int attr_name,
+			      Dwarf_Word *result)
+{
+	Dwarf_Attribute attr;
+
+	if (dwarf_attr(tp_die, attr_name, &attr) == NULL ||
+	    dwarf_formudata(&attr, result) != 0)
+		return -ENOENT;
+
+	return 0;
+}
+
+/* Get attribute and translate it as a sdata */
+static int die_get_attr_sdata(Dwarf_Die *tp_die, unsigned int attr_name,
+			      Dwarf_Sword *result)
+{
+	Dwarf_Attribute attr;
+
+	if (dwarf_attr(tp_die, attr_name, &attr) == NULL ||
+	    dwarf_formsdata(&attr, result) != 0)
+		return -ENOENT;
+
+	return 0;
+}
+
+/**
+ * die_is_signed_type - Check whether a type DIE is signed or not
+ * @tp_die: a DIE of a type
+ *
+ * Get the encoding of @tp_die and return true if the encoding
+ * is signed.
+ */
+bool die_is_signed_type(Dwarf_Die *tp_die)
+{
+	Dwarf_Word ret;
+
+	if (die_get_attr_udata(tp_die, DW_AT_encoding, &ret))
+		return false;
+
+	return (ret == DW_ATE_signed_char || ret == DW_ATE_signed ||
+		ret == DW_ATE_signed_fixed);
+}
+
+/**
+ * die_is_func_def - Ensure that this DIE is a subprogram and definition
+ * @dw_die: a DIE
+ *
+ * Ensure that this DIE is a subprogram and NOT a declaration. This
+ * returns true if @dw_die is a function definition.
+ **/
+bool die_is_func_def(Dwarf_Die *dw_die)
+{
+	Dwarf_Attribute attr;
+
+	return (dwarf_tag(dw_die) == DW_TAG_subprogram &&
+		dwarf_attr(dw_die, DW_AT_declaration, &attr) == NULL);
+}
+
+/**
+ * die_is_func_instance - Ensure that this DIE is an instance of a subprogram
+ * @dw_die: a DIE
+ *
+ * Ensure that this DIE is an instance (which has an entry address).
+ * This returns true if @dw_die is a function instance. If not, you need to
+ * call die_walk_instances() to find actual instances.
+ **/
+bool die_is_func_instance(Dwarf_Die *dw_die)
+{
+	Dwarf_Addr tmp;
+
+	/* Actually gcc optimizes non-inline as like as inlined */
+	return !dwarf_func_inline(dw_die) && dwarf_entrypc(dw_die, &tmp) == 0;
+}
+/**
+ * die_get_data_member_location - Get the data-member offset
+ * @mb_die: a DIE of a member of a data structure
+ * @offs: The offset of the member in the data structure
+ *
+ * Get the offset of @mb_die in the data structure including @mb_die, and
+ * stores result offset to @offs. If any error occurs this returns errno.
+ */
+int die_get_data_member_location(Dwarf_Die *mb_die, Dwarf_Word *offs)
+{
+	Dwarf_Attribute attr;
+	Dwarf_Op *expr;
+	size_t nexpr;
+	int ret;
+
+	if (dwarf_attr(mb_die, DW_AT_data_member_location, &attr) == NULL)
+		return -ENOENT;
+
+	if (dwarf_formudata(&attr, offs) != 0) {
+		/* DW_AT_data_member_location should be DW_OP_plus_uconst */
+		ret = dwarf_getlocation(&attr, &expr, &nexpr);
+		if (ret < 0 || nexpr == 0)
+			return -ENOENT;
+
+		if (expr[0].atom != DW_OP_plus_uconst || nexpr != 1) {
+			pr_debug("Unable to get offset:Unexpected OP %x (%zd)\n",
+				 expr[0].atom, nexpr);
+			return -ENOTSUP;
+		}
+		*offs = (Dwarf_Word)expr[0].number;
+	}
+	return 0;
+}
+
+/* Get the call file index number in CU DIE */
+static int die_get_call_fileno(Dwarf_Die *in_die)
+{
+	Dwarf_Sword idx;
+
+	if (die_get_attr_sdata(in_die, DW_AT_call_file, &idx) == 0)
+		return (int)idx;
+	else
+		return -ENOENT;
+}
+
+/* Get the declared file index number in CU DIE */
+static int die_get_decl_fileno(Dwarf_Die *pdie)
+{
+	Dwarf_Sword idx;
+
+	if (die_get_attr_sdata(pdie, DW_AT_decl_file, &idx) == 0)
+		return (int)idx;
+	else
+		return -ENOENT;
+}
+
+/**
+ * die_get_call_file - Get callsite file name of inlined function instance
+ * @in_die: a DIE of an inlined function instance
+ *
+ * Get call-site file name of @in_die. This means from which file the inline
+ * function is called.
+ */
+const char *die_get_call_file(Dwarf_Die *in_die)
+{
+	Dwarf_Die cu_die;
+	Dwarf_Files *files;
+	int idx;
+
+	idx = die_get_call_fileno(in_die);
+	if (idx < 0 || !dwarf_diecu(in_die, &cu_die, NULL, NULL) ||
+	    dwarf_getsrcfiles(&cu_die, &files, NULL) != 0)
+		return NULL;
+
+	return dwarf_filesrc(files, idx, NULL, NULL);
+}
+
+
+/**
+ * die_find_child - Generic DIE search function in DIE tree
+ * @rt_die: a root DIE
+ * @callback: a callback function
+ * @data: a user data passed to the callback function
+ * @die_mem: a buffer for result DIE
+ *
+ * Trace DIE tree from @rt_die and call @callback for each child DIE.
+ * If @callback returns DIE_FIND_CB_END, this stores the DIE into
+ * @die_mem and returns it. If @callback returns DIE_FIND_CB_CONTINUE,
+ * this continues to trace the tree. Optionally, @callback can return
+ * DIE_FIND_CB_CHILD and DIE_FIND_CB_SIBLING, those means trace only
+ * the children and trace only the siblings respectively.
+ * Returns NULL if @callback can't find any appropriate DIE.
+ */
+Dwarf_Die *die_find_child(Dwarf_Die *rt_die,
+			  int (*callback)(Dwarf_Die *, void *),
+			  void *data, Dwarf_Die *die_mem)
+{
+	Dwarf_Die child_die;
+	int ret;
+
+	ret = dwarf_child(rt_die, die_mem);
+	if (ret != 0)
+		return NULL;
+
+	do {
+		ret = callback(die_mem, data);
+		if (ret == DIE_FIND_CB_END)
+			return die_mem;
+
+		if ((ret & DIE_FIND_CB_CHILD) &&
+		    die_find_child(die_mem, callback, data, &child_die)) {
+			memcpy(die_mem, &child_die, sizeof(Dwarf_Die));
+			return die_mem;
+		}
+	} while ((ret & DIE_FIND_CB_SIBLING) &&
+		 dwarf_siblingof(die_mem, die_mem) == 0);
+
+	return NULL;
+}
+
+struct __addr_die_search_param {
+	Dwarf_Addr	addr;
+	Dwarf_Die	*die_mem;
+};
+
+/* die_find callback for non-inlined function search */
+static int __die_search_func_cb(Dwarf_Die *fn_die, void *data)
+{
+	struct __addr_die_search_param *ad = data;
+
+	/*
+	 * Since a declaration entry doesn't has given pc, this always returns
+	 * function definition entry.
+	 */
+	if (dwarf_tag(fn_die) == DW_TAG_subprogram &&
+	    dwarf_haspc(fn_die, ad->addr)) {
+		memcpy(ad->die_mem, fn_die, sizeof(Dwarf_Die));
+		return DWARF_CB_ABORT;
+	}
+	return DWARF_CB_OK;
+}
+
+/**
+ * die_find_realfunc - Search a non-inlined function at given address
+ * @cu_die: a CU DIE which including @addr
+ * @addr: target address
+ * @die_mem: a buffer for result DIE
+ *
+ * Search a non-inlined function DIE which includes @addr. Stores the
+ * DIE to @die_mem and returns it if found. Returns NULL if failed.
+ */
+Dwarf_Die *die_find_realfunc(Dwarf_Die *cu_die, Dwarf_Addr addr,
+				    Dwarf_Die *die_mem)
+{
+	struct __addr_die_search_param ad;
+	ad.addr = addr;
+	ad.die_mem = die_mem;
+	/* dwarf_getscopes can't find subprogram. */
+	if (!dwarf_getfuncs(cu_die, __die_search_func_cb, &ad, 0))
+		return NULL;
+	else
+		return die_mem;
+}
+
+/* die_find callback for inline function search */
+static int __die_find_inline_cb(Dwarf_Die *die_mem, void *data)
+{
+	Dwarf_Addr *addr = data;
+
+	if (dwarf_tag(die_mem) == DW_TAG_inlined_subroutine &&
+	    dwarf_haspc(die_mem, *addr))
+		return DIE_FIND_CB_END;
+
+	return DIE_FIND_CB_CONTINUE;
+}
+
+/**
+ * die_find_top_inlinefunc - Search the top inlined function at given address
+ * @sp_die: a subprogram DIE which including @addr
+ * @addr: target address
+ * @die_mem: a buffer for result DIE
+ *
+ * Search an inlined function DIE which includes @addr. Stores the
+ * DIE to @die_mem and returns it if found. Returns NULL if failed.
+ * Even if several inlined functions are expanded recursively, this
+ * doesn't trace it down, and returns the topmost one.
+ */
+Dwarf_Die *die_find_top_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr,
+				   Dwarf_Die *die_mem)
+{
+	return die_find_child(sp_die, __die_find_inline_cb, &addr, die_mem);
+}
+
+/**
+ * die_find_inlinefunc - Search an inlined function at given address
+ * @sp_die: a subprogram DIE which including @addr
+ * @addr: target address
+ * @die_mem: a buffer for result DIE
+ *
+ * Search an inlined function DIE which includes @addr. Stores the
+ * DIE to @die_mem and returns it if found. Returns NULL if failed.
+ * If several inlined functions are expanded recursively, this trace
+ * it down and returns deepest one.
+ */
+Dwarf_Die *die_find_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr,
+			       Dwarf_Die *die_mem)
+{
+	Dwarf_Die tmp_die;
+
+	sp_die = die_find_child(sp_die, __die_find_inline_cb, &addr, &tmp_die);
+	if (!sp_die)
+		return NULL;
+
+	/* Inlined function could be recursive. Trace it until fail */
+	while (sp_die) {
+		memcpy(die_mem, sp_die, sizeof(Dwarf_Die));
+		sp_die = die_find_child(sp_die, __die_find_inline_cb, &addr,
+					&tmp_die);
+	}
+
+	return die_mem;
+}
+
+struct __instance_walk_param {
+	void    *addr;
+	int	(*callback)(Dwarf_Die *, void *);
+	void    *data;
+	int	retval;
+};
+
+static int __die_walk_instances_cb(Dwarf_Die *inst, void *data)
+{
+	struct __instance_walk_param *iwp = data;
+	Dwarf_Attribute attr_mem;
+	Dwarf_Die origin_mem;
+	Dwarf_Attribute *attr;
+	Dwarf_Die *origin;
+	int tmp;
+
+	attr = dwarf_attr(inst, DW_AT_abstract_origin, &attr_mem);
+	if (attr == NULL)
+		return DIE_FIND_CB_CONTINUE;
+
+	origin = dwarf_formref_die(attr, &origin_mem);
+	if (origin == NULL || origin->addr != iwp->addr)
+		return DIE_FIND_CB_CONTINUE;
+
+	/* Ignore redundant instances */
+	if (dwarf_tag(inst) == DW_TAG_inlined_subroutine) {
+		dwarf_decl_line(origin, &tmp);
+		if (die_get_call_lineno(inst) == tmp) {
+			tmp = die_get_decl_fileno(origin);
+			if (die_get_call_fileno(inst) == tmp)
+				return DIE_FIND_CB_CONTINUE;
+		}
+	}
+
+	iwp->retval = iwp->callback(inst, iwp->data);
+
+	return (iwp->retval) ? DIE_FIND_CB_END : DIE_FIND_CB_CONTINUE;
+}
+
+/**
+ * die_walk_instances - Walk on instances of given DIE
+ * @or_die: an abstract original DIE
+ * @callback: a callback function which is called with instance DIE
+ * @data: user data
+ *
+ * Walk on the instances of give @in_die. @in_die must be an inlined function
+ * declartion. This returns the return value of @callback if it returns
+ * non-zero value, or -ENOENT if there is no instance.
+ */
+int die_walk_instances(Dwarf_Die *or_die, int (*callback)(Dwarf_Die *, void *),
+		       void *data)
+{
+	Dwarf_Die cu_die;
+	Dwarf_Die die_mem;
+	struct __instance_walk_param iwp = {
+		.addr = or_die->addr,
+		.callback = callback,
+		.data = data,
+		.retval = -ENOENT,
+	};
+
+	if (dwarf_diecu(or_die, &cu_die, NULL, NULL) == NULL)
+		return -ENOENT;
+
+	die_find_child(&cu_die, __die_walk_instances_cb, &iwp, &die_mem);
+
+	return iwp.retval;
+}
+
+/* Line walker internal parameters */
+struct __line_walk_param {
+	bool recursive;
+	line_walk_callback_t callback;
+	void *data;
+	int retval;
+};
+
+static int __die_walk_funclines_cb(Dwarf_Die *in_die, void *data)
+{
+	struct __line_walk_param *lw = data;
+	Dwarf_Addr addr = 0;
+	const char *fname;
+	int lineno;
+
+	if (dwarf_tag(in_die) == DW_TAG_inlined_subroutine) {
+		fname = die_get_call_file(in_die);
+		lineno = die_get_call_lineno(in_die);
+		if (fname && lineno > 0 && dwarf_entrypc(in_die, &addr) == 0) {
+			lw->retval = lw->callback(fname, lineno, addr, lw->data);
+			if (lw->retval != 0)
+				return DIE_FIND_CB_END;
+		}
+	}
+	if (!lw->recursive)
+		/* Don't need to search recursively */
+		return DIE_FIND_CB_SIBLING;
+
+	if (addr) {
+		fname = dwarf_decl_file(in_die);
+		if (fname && dwarf_decl_line(in_die, &lineno) == 0) {
+			lw->retval = lw->callback(fname, lineno, addr, lw->data);
+			if (lw->retval != 0)
+				return DIE_FIND_CB_END;
+		}
+	}
+
+	/* Continue to search nested inlined function call-sites */
+	return DIE_FIND_CB_CONTINUE;
+}
+
+/* Walk on lines of blocks included in given DIE */
+static int __die_walk_funclines(Dwarf_Die *sp_die, bool recursive,
+				line_walk_callback_t callback, void *data)
+{
+	struct __line_walk_param lw = {
+		.recursive = recursive,
+		.callback = callback,
+		.data = data,
+		.retval = 0,
+	};
+	Dwarf_Die die_mem;
+	Dwarf_Addr addr;
+	const char *fname;
+	int lineno;
+
+	/* Handle function declaration line */
+	fname = dwarf_decl_file(sp_die);
+	if (fname && dwarf_decl_line(sp_die, &lineno) == 0 &&
+	    dwarf_entrypc(sp_die, &addr) == 0) {
+		lw.retval = callback(fname, lineno, addr, data);
+		if (lw.retval != 0)
+			goto done;
+	}
+	die_find_child(sp_die, __die_walk_funclines_cb, &lw, &die_mem);
+done:
+	return lw.retval;
+}
+
+static int __die_walk_culines_cb(Dwarf_Die *sp_die, void *data)
+{
+	struct __line_walk_param *lw = data;
+
+	lw->retval = __die_walk_funclines(sp_die, true, lw->callback, lw->data);
+	if (lw->retval != 0)
+		return DWARF_CB_ABORT;
+
+	return DWARF_CB_OK;
+}
+
+/**
+ * die_walk_lines - Walk on lines inside given DIE
+ * @rt_die: a root DIE (CU, subprogram or inlined_subroutine)
+ * @callback: callback routine
+ * @data: user data
+ *
+ * Walk on all lines inside given @rt_die and call @callback on each line.
+ * If the @rt_die is a function, walk only on the lines inside the function,
+ * otherwise @rt_die must be a CU DIE.
+ * Note that this walks not only dwarf line list, but also function entries
+ * and inline call-site.
+ */
+int die_walk_lines(Dwarf_Die *rt_die, line_walk_callback_t callback, void *data)
+{
+	Dwarf_Lines *lines;
+	Dwarf_Line *line;
+	Dwarf_Addr addr;
+	const char *fname;
+	int lineno, ret = 0;
+	Dwarf_Die die_mem, *cu_die;
+	size_t nlines, i;
+
+	/* Get the CU die */
+	if (dwarf_tag(rt_die) != DW_TAG_compile_unit)
+		cu_die = dwarf_diecu(rt_die, &die_mem, NULL, NULL);
+	else
+		cu_die = rt_die;
+	if (!cu_die) {
+		pr_debug2("Failed to get CU from given DIE.\n");
+		return -EINVAL;
+	}
+
+	/* Get lines list in the CU */
+	if (dwarf_getsrclines(cu_die, &lines, &nlines) != 0) {
+		pr_debug2("Failed to get source lines on this CU.\n");
+		return -ENOENT;
+	}
+	pr_debug2("Get %zd lines from this CU\n", nlines);
+
+	/* Walk on the lines on lines list */
+	for (i = 0; i < nlines; i++) {
+		line = dwarf_onesrcline(lines, i);
+		if (line == NULL ||
+		    dwarf_lineno(line, &lineno) != 0 ||
+		    dwarf_lineaddr(line, &addr) != 0) {
+			pr_debug2("Failed to get line info. "
+				  "Possible error in debuginfo.\n");
+			continue;
+		}
+		/* Filter lines based on address */
+		if (rt_die != cu_die)
+			/*
+			 * Address filtering
+			 * The line is included in given function, and
+			 * no inline block includes it.
+			 */
+			if (!dwarf_haspc(rt_die, addr) ||
+			    die_find_inlinefunc(rt_die, addr, &die_mem))
+				continue;
+		/* Get source line */
+		fname = dwarf_linesrc(line, NULL, NULL);
+
+		ret = callback(fname, lineno, addr, data);
+		if (ret != 0)
+			return ret;
+	}
+
+	/*
+	 * Dwarf lines doesn't include function declarations and inlined
+	 * subroutines. We have to check functions list or given function.
+	 */
+	if (rt_die != cu_die)
+		/*
+		 * Don't need walk functions recursively, because nested
+		 * inlined functions don't have lines of the specified DIE.
+		 */
+		ret = __die_walk_funclines(rt_die, false, callback, data);
+	else {
+		struct __line_walk_param param = {
+			.callback = callback,
+			.data = data,
+			.retval = 0,
+		};
+		dwarf_getfuncs(cu_die, __die_walk_culines_cb, &param, 0);
+		ret = param.retval;
+	}
+
+	return ret;
+}
+
+struct __find_variable_param {
+	const char *name;
+	Dwarf_Addr addr;
+};
+
+static int __die_find_variable_cb(Dwarf_Die *die_mem, void *data)
+{
+	struct __find_variable_param *fvp = data;
+	Dwarf_Attribute attr;
+	int tag;
+
+	tag = dwarf_tag(die_mem);
+	if ((tag == DW_TAG_formal_parameter ||
+	     tag == DW_TAG_variable) &&
+	    die_compare_name(die_mem, fvp->name) &&
+	/* Does the DIE have location information or external instance? */
+	    (dwarf_attr(die_mem, DW_AT_external, &attr) ||
+	     dwarf_attr(die_mem, DW_AT_location, &attr)))
+		return DIE_FIND_CB_END;
+	if (dwarf_haspc(die_mem, fvp->addr))
+		return DIE_FIND_CB_CONTINUE;
+	else
+		return DIE_FIND_CB_SIBLING;
+}
+
+/**
+ * die_find_variable_at - Find a given name variable at given address
+ * @sp_die: a function DIE
+ * @name: variable name
+ * @addr: address
+ * @die_mem: a buffer for result DIE
+ *
+ * Find a variable DIE called @name at @addr in @sp_die.
+ */
+Dwarf_Die *die_find_variable_at(Dwarf_Die *sp_die, const char *name,
+				Dwarf_Addr addr, Dwarf_Die *die_mem)
+{
+	struct __find_variable_param fvp = { .name = name, .addr = addr};
+
+	return die_find_child(sp_die, __die_find_variable_cb, (void *)&fvp,
+			      die_mem);
+}
+
+static int __die_find_member_cb(Dwarf_Die *die_mem, void *data)
+{
+	const char *name = data;
+
+	if (dwarf_tag(die_mem) == DW_TAG_member) {
+		if (die_compare_name(die_mem, name))
+			return DIE_FIND_CB_END;
+		else if (!dwarf_diename(die_mem)) {	/* Unnamed structure */
+			Dwarf_Die type_die, tmp_die;
+			if (die_get_type(die_mem, &type_die) &&
+			    die_find_member(&type_die, name, &tmp_die))
+				return DIE_FIND_CB_END;
+		}
+	}
+	return DIE_FIND_CB_SIBLING;
+}
+
+/**
+ * die_find_member - Find a given name member in a data structure
+ * @st_die: a data structure type DIE
+ * @name: member name
+ * @die_mem: a buffer for result DIE
+ *
+ * Find a member DIE called @name in @st_die.
+ */
+Dwarf_Die *die_find_member(Dwarf_Die *st_die, const char *name,
+			   Dwarf_Die *die_mem)
+{
+	return die_find_child(st_die, __die_find_member_cb, (void *)name,
+			      die_mem);
+}
+
+/**
+ * die_get_typename - Get the name of given variable DIE
+ * @vr_die: a variable DIE
+ * @buf: a buffer for result type name
+ * @len: a max-length of @buf
+ *
+ * Get the name of @vr_die and stores it to @buf. Return the actual length
+ * of type name if succeeded. Return -E2BIG if @len is not enough long, and
+ * Return -ENOENT if failed to find type name.
+ * Note that the result will stores typedef name if possible, and stores
+ * "*(function_type)" if the type is a function pointer.
+ */
+int die_get_typename(Dwarf_Die *vr_die, char *buf, int len)
+{
+	Dwarf_Die type;
+	int tag, ret, ret2;
+	const char *tmp = "";
+
+	if (__die_get_real_type(vr_die, &type) == NULL)
+		return -ENOENT;
+
+	tag = dwarf_tag(&type);
+	if (tag == DW_TAG_array_type || tag == DW_TAG_pointer_type)
+		tmp = "*";
+	else if (tag == DW_TAG_subroutine_type) {
+		/* Function pointer */
+		ret = snprintf(buf, len, "(function_type)");
+		return (ret >= len) ? -E2BIG : ret;
+	} else {
+		if (!dwarf_diename(&type))
+			return -ENOENT;
+		if (tag == DW_TAG_union_type)
+			tmp = "union ";
+		else if (tag == DW_TAG_structure_type)
+			tmp = "struct ";
+		else if (tag == DW_TAG_enumeration_type)
+			tmp = "enum ";
+		/* Write a base name */
+		ret = snprintf(buf, len, "%s%s", tmp, dwarf_diename(&type));
+		return (ret >= len) ? -E2BIG : ret;
+	}
+	ret = die_get_typename(&type, buf, len);
+	if (ret > 0) {
+		ret2 = snprintf(buf + ret, len - ret, "%s", tmp);
+		ret = (ret2 >= len - ret) ? -E2BIG : ret2 + ret;
+	}
+	return ret;
+}
+
+/**
+ * die_get_varname - Get the name and type of given variable DIE
+ * @vr_die: a variable DIE
+ * @buf: a buffer for type and variable name
+ * @len: the max-length of @buf
+ *
+ * Get the name and type of @vr_die and stores it in @buf as "type\tname".
+ */
+int die_get_varname(Dwarf_Die *vr_die, char *buf, int len)
+{
+	int ret, ret2;
+
+	ret = die_get_typename(vr_die, buf, len);
+	if (ret < 0) {
+		pr_debug("Failed to get type, make it unknown.\n");
+		ret = snprintf(buf, len, "(unknown_type)");
+	}
+	if (ret > 0) {
+		ret2 = snprintf(buf + ret, len - ret, "\t%s",
+				dwarf_diename(vr_die));
+		ret = (ret2 >= len - ret) ? -E2BIG : ret2 + ret;
+	}
+	return ret;
+}
+
diff --git a/tools/perf/util/dwarf-aux.h b/tools/perf/util/dwarf-aux.h
new file mode 100644
index 000000000..af7dbcd5f
--- /dev/null
+++ b/tools/perf/util/dwarf-aux.h
@@ -0,0 +1,121 @@
+#ifndef _DWARF_AUX_H
+#define _DWARF_AUX_H
+/*
+ * dwarf-aux.h : libdw auxiliary interfaces
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <dwarf.h>
+#include <elfutils/libdw.h>
+#include <elfutils/libdwfl.h>
+#include <elfutils/version.h>
+
+/* Find the realpath of the target file */
+extern const char *cu_find_realpath(Dwarf_Die *cu_die, const char *fname);
+
+/* Get DW_AT_comp_dir (should be NULL with older gcc) */
+extern const char *cu_get_comp_dir(Dwarf_Die *cu_die);
+
+/* Get a line number and file name for given address */
+extern int cu_find_lineinfo(Dwarf_Die *cudie, unsigned long addr,
+			    const char **fname, int *lineno);
+
+/* Walk on funcitons at given address */
+extern int cu_walk_functions_at(Dwarf_Die *cu_die, Dwarf_Addr addr,
+			int (*callback)(Dwarf_Die *, void *), void *data);
+
+/* Ensure that this DIE is a subprogram and definition (not declaration) */
+extern bool die_is_func_def(Dwarf_Die *dw_die);
+
+/* Ensure that this DIE is an instance of a subprogram */
+extern bool die_is_func_instance(Dwarf_Die *dw_die);
+
+/* Compare diename and tname */
+extern bool die_compare_name(Dwarf_Die *dw_die, const char *tname);
+
+/* Get callsite line number of inline-function instance */
+extern int die_get_call_lineno(Dwarf_Die *in_die);
+
+/* Get callsite file name of inlined function instance */
+extern const char *die_get_call_file(Dwarf_Die *in_die);
+
+/* Get type die */
+extern Dwarf_Die *die_get_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem);
+
+/* Get a type die, but skip qualifiers and typedef */
+extern Dwarf_Die *die_get_real_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem);
+
+/* Check whether the DIE is signed or not */
+extern bool die_is_signed_type(Dwarf_Die *tp_die);
+
+/* Get data_member_location offset */
+extern int die_get_data_member_location(Dwarf_Die *mb_die, Dwarf_Word *offs);
+
+/* Return values for die_find_child() callbacks */
+enum {
+	DIE_FIND_CB_END = 0,		/* End of Search */
+	DIE_FIND_CB_CHILD = 1,		/* Search only children */
+	DIE_FIND_CB_SIBLING = 2,	/* Search only siblings */
+	DIE_FIND_CB_CONTINUE = 3,	/* Search children and siblings */
+};
+
+/* Search child DIEs */
+extern Dwarf_Die *die_find_child(Dwarf_Die *rt_die,
+				 int (*callback)(Dwarf_Die *, void *),
+				 void *data, Dwarf_Die *die_mem);
+
+/* Search a non-inlined function including given address */
+extern Dwarf_Die *die_find_realfunc(Dwarf_Die *cu_die, Dwarf_Addr addr,
+				    Dwarf_Die *die_mem);
+
+/* Search the top inlined function including given address */
+extern Dwarf_Die *die_find_top_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr,
+					  Dwarf_Die *die_mem);
+
+/* Search the deepest inlined function including given address */
+extern Dwarf_Die *die_find_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr,
+				      Dwarf_Die *die_mem);
+
+/* Walk on the instances of given DIE */
+extern int die_walk_instances(Dwarf_Die *in_die,
+			      int (*callback)(Dwarf_Die *, void *), void *data);
+
+/* Walker on lines (Note: line number will not be sorted) */
+typedef int (* line_walk_callback_t) (const char *fname, int lineno,
+				      Dwarf_Addr addr, void *data);
+
+/*
+ * Walk on lines inside given DIE. If the DIE is a subprogram, walk only on
+ * the lines inside the subprogram, otherwise the DIE must be a CU DIE.
+ */
+extern int die_walk_lines(Dwarf_Die *rt_die, line_walk_callback_t callback,
+			  void *data);
+
+/* Find a variable called 'name' at given address */
+extern Dwarf_Die *die_find_variable_at(Dwarf_Die *sp_die, const char *name,
+				       Dwarf_Addr addr, Dwarf_Die *die_mem);
+
+/* Find a member called 'name' */
+extern Dwarf_Die *die_find_member(Dwarf_Die *st_die, const char *name,
+				  Dwarf_Die *die_mem);
+
+/* Get the name of given variable DIE */
+extern int die_get_typename(Dwarf_Die *vr_die, char *buf, int len);
+
+/* Get the name and type of given variable DIE, stored as "type\tname" */
+extern int die_get_varname(Dwarf_Die *vr_die, char *buf, int len);
+#endif
diff --git a/tools/perf/util/environment.c b/tools/perf/util/environment.c
new file mode 100644
index 000000000..275b0ee34
--- /dev/null
+++ b/tools/perf/util/environment.c
@@ -0,0 +1,9 @@
+/*
+ * We put all the perf config variables in this same object
+ * file, so that programs can link against the config parser
+ * without having to link against all the rest of perf.
+ */
+#include "cache.h"
+
+const char *pager_program;
+int pager_use_color = 1;
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
new file mode 100644
index 000000000..ff866c4d2
--- /dev/null
+++ b/tools/perf/util/event.c
@@ -0,0 +1,978 @@
+#include <linux/types.h>
+#include <sys/mman.h>
+#include "event.h"
+#include "debug.h"
+#include "hist.h"
+#include "machine.h"
+#include "sort.h"
+#include "string.h"
+#include "strlist.h"
+#include "thread.h"
+#include "thread_map.h"
+#include "symbol/kallsyms.h"
+
+static const char *perf_event__names[] = {
+	[0]					= "TOTAL",
+	[PERF_RECORD_MMAP]			= "MMAP",
+	[PERF_RECORD_MMAP2]			= "MMAP2",
+	[PERF_RECORD_LOST]			= "LOST",
+	[PERF_RECORD_COMM]			= "COMM",
+	[PERF_RECORD_EXIT]			= "EXIT",
+	[PERF_RECORD_THROTTLE]			= "THROTTLE",
+	[PERF_RECORD_UNTHROTTLE]		= "UNTHROTTLE",
+	[PERF_RECORD_FORK]			= "FORK",
+	[PERF_RECORD_READ]			= "READ",
+	[PERF_RECORD_SAMPLE]			= "SAMPLE",
+	[PERF_RECORD_HEADER_ATTR]		= "ATTR",
+	[PERF_RECORD_HEADER_EVENT_TYPE]		= "EVENT_TYPE",
+	[PERF_RECORD_HEADER_TRACING_DATA]	= "TRACING_DATA",
+	[PERF_RECORD_HEADER_BUILD_ID]		= "BUILD_ID",
+	[PERF_RECORD_FINISHED_ROUND]		= "FINISHED_ROUND",
+	[PERF_RECORD_ID_INDEX]			= "ID_INDEX",
+};
+
+const char *perf_event__name(unsigned int id)
+{
+	if (id >= ARRAY_SIZE(perf_event__names))
+		return "INVALID";
+	if (!perf_event__names[id])
+		return "UNKNOWN";
+	return perf_event__names[id];
+}
+
+static struct perf_sample synth_sample = {
+	.pid	   = -1,
+	.tid	   = -1,
+	.time	   = -1,
+	.stream_id = -1,
+	.cpu	   = -1,
+	.period	   = 1,
+};
+
+/*
+ * Assumes that the first 4095 bytes of /proc/pid/stat contains
+ * the comm, tgid and ppid.
+ */
+static int perf_event__get_comm_ids(pid_t pid, char *comm, size_t len,
+				    pid_t *tgid, pid_t *ppid)
+{
+	char filename[PATH_MAX];
+	char bf[4096];
+	int fd;
+	size_t size = 0, n;
+	char *nl, *name, *tgids, *ppids;
+
+	*tgid = -1;
+	*ppid = -1;
+
+	snprintf(filename, sizeof(filename), "/proc/%d/status", pid);
+
+	fd = open(filename, O_RDONLY);
+	if (fd < 0) {
+		pr_debug("couldn't open %s\n", filename);
+		return -1;
+	}
+
+	n = read(fd, bf, sizeof(bf) - 1);
+	close(fd);
+	if (n <= 0) {
+		pr_warning("Couldn't get COMM, tigd and ppid for pid %d\n",
+			   pid);
+		return -1;
+	}
+	bf[n] = '\0';
+
+	name = strstr(bf, "Name:");
+	tgids = strstr(bf, "Tgid:");
+	ppids = strstr(bf, "PPid:");
+
+	if (name) {
+		name += 5;  /* strlen("Name:") */
+
+		while (*name && isspace(*name))
+			++name;
+
+		nl = strchr(name, '\n');
+		if (nl)
+			*nl = '\0';
+
+		size = strlen(name);
+		if (size >= len)
+			size = len - 1;
+		memcpy(comm, name, size);
+		comm[size] = '\0';
+	} else {
+		pr_debug("Name: string not found for pid %d\n", pid);
+	}
+
+	if (tgids) {
+		tgids += 5;  /* strlen("Tgid:") */
+		*tgid = atoi(tgids);
+	} else {
+		pr_debug("Tgid: string not found for pid %d\n", pid);
+	}
+
+	if (ppids) {
+		ppids += 5;  /* strlen("PPid:") */
+		*ppid = atoi(ppids);
+	} else {
+		pr_debug("PPid: string not found for pid %d\n", pid);
+	}
+
+	return 0;
+}
+
+static int perf_event__prepare_comm(union perf_event *event, pid_t pid,
+				    struct machine *machine,
+				    pid_t *tgid, pid_t *ppid)
+{
+	size_t size;
+
+	*ppid = -1;
+
+	memset(&event->comm, 0, sizeof(event->comm));
+
+	if (machine__is_host(machine)) {
+		if (perf_event__get_comm_ids(pid, event->comm.comm,
+					     sizeof(event->comm.comm),
+					     tgid, ppid) != 0) {
+			return -1;
+		}
+	} else {
+		*tgid = machine->pid;
+	}
+
+	if (*tgid < 0)
+		return -1;
+
+	event->comm.pid = *tgid;
+	event->comm.header.type = PERF_RECORD_COMM;
+
+	size = strlen(event->comm.comm) + 1;
+	size = PERF_ALIGN(size, sizeof(u64));
+	memset(event->comm.comm + size, 0, machine->id_hdr_size);
+	event->comm.header.size = (sizeof(event->comm) -
+				(sizeof(event->comm.comm) - size) +
+				machine->id_hdr_size);
+	event->comm.tid = pid;
+
+	return 0;
+}
+
+static pid_t perf_event__synthesize_comm(struct perf_tool *tool,
+					 union perf_event *event, pid_t pid,
+					 perf_event__handler_t process,
+					 struct machine *machine)
+{
+	pid_t tgid, ppid;
+
+	if (perf_event__prepare_comm(event, pid, machine, &tgid, &ppid) != 0)
+		return -1;
+
+	if (process(tool, event, &synth_sample, machine) != 0)
+		return -1;
+
+	return tgid;
+}
+
+static int perf_event__synthesize_fork(struct perf_tool *tool,
+				       union perf_event *event,
+				       pid_t pid, pid_t tgid, pid_t ppid,
+				       perf_event__handler_t process,
+				       struct machine *machine)
+{
+	memset(&event->fork, 0, sizeof(event->fork) + machine->id_hdr_size);
+
+	/*
+	 * for main thread set parent to ppid from status file. For other
+	 * threads set parent pid to main thread. ie., assume main thread
+	 * spawns all threads in a process
+	*/
+	if (tgid == pid) {
+		event->fork.ppid = ppid;
+		event->fork.ptid = ppid;
+	} else {
+		event->fork.ppid = tgid;
+		event->fork.ptid = tgid;
+	}
+	event->fork.pid  = tgid;
+	event->fork.tid  = pid;
+	event->fork.header.type = PERF_RECORD_FORK;
+
+	event->fork.header.size = (sizeof(event->fork) + machine->id_hdr_size);
+
+	if (process(tool, event, &synth_sample, machine) != 0)
+		return -1;
+
+	return 0;
+}
+
+int perf_event__synthesize_mmap_events(struct perf_tool *tool,
+				       union perf_event *event,
+				       pid_t pid, pid_t tgid,
+				       perf_event__handler_t process,
+				       struct machine *machine,
+				       bool mmap_data)
+{
+	char filename[PATH_MAX];
+	FILE *fp;
+	int rc = 0;
+
+	if (machine__is_default_guest(machine))
+		return 0;
+
+	snprintf(filename, sizeof(filename), "%s/proc/%d/maps",
+		 machine->root_dir, pid);
+
+	fp = fopen(filename, "r");
+	if (fp == NULL) {
+		/*
+		 * We raced with a task exiting - just return:
+		 */
+		pr_debug("couldn't open %s\n", filename);
+		return -1;
+	}
+
+	event->header.type = PERF_RECORD_MMAP2;
+
+	while (1) {
+		char bf[BUFSIZ];
+		char prot[5];
+		char execname[PATH_MAX];
+		char anonstr[] = "//anon";
+		unsigned int ino;
+		size_t size;
+		ssize_t n;
+
+		if (fgets(bf, sizeof(bf), fp) == NULL)
+			break;
+
+		/* ensure null termination since stack will be reused. */
+		strcpy(execname, "");
+
+		/* 00400000-0040c000 r-xp 00000000 fd:01 41038  /bin/cat */
+		n = sscanf(bf, "%"PRIx64"-%"PRIx64" %s %"PRIx64" %x:%x %u %s\n",
+		       &event->mmap2.start, &event->mmap2.len, prot,
+		       &event->mmap2.pgoff, &event->mmap2.maj,
+		       &event->mmap2.min,
+		       &ino, execname);
+
+		/*
+ 		 * Anon maps don't have the execname.
+ 		 */
+		if (n < 7)
+			continue;
+
+		event->mmap2.ino = (u64)ino;
+
+		/*
+		 * Just like the kernel, see __perf_event_mmap in kernel/perf_event.c
+		 */
+		if (machine__is_host(machine))
+			event->header.misc = PERF_RECORD_MISC_USER;
+		else
+			event->header.misc = PERF_RECORD_MISC_GUEST_USER;
+
+		/* map protection and flags bits */
+		event->mmap2.prot = 0;
+		event->mmap2.flags = 0;
+		if (prot[0] == 'r')
+			event->mmap2.prot |= PROT_READ;
+		if (prot[1] == 'w')
+			event->mmap2.prot |= PROT_WRITE;
+		if (prot[2] == 'x')
+			event->mmap2.prot |= PROT_EXEC;
+
+		if (prot[3] == 's')
+			event->mmap2.flags |= MAP_SHARED;
+		else
+			event->mmap2.flags |= MAP_PRIVATE;
+
+		if (prot[2] != 'x') {
+			if (!mmap_data || prot[0] != 'r')
+				continue;
+
+			event->header.misc |= PERF_RECORD_MISC_MMAP_DATA;
+		}
+
+		if (!strcmp(execname, ""))
+			strcpy(execname, anonstr);
+
+		size = strlen(execname) + 1;
+		memcpy(event->mmap2.filename, execname, size);
+		size = PERF_ALIGN(size, sizeof(u64));
+		event->mmap2.len -= event->mmap.start;
+		event->mmap2.header.size = (sizeof(event->mmap2) -
+					(sizeof(event->mmap2.filename) - size));
+		memset(event->mmap2.filename + size, 0, machine->id_hdr_size);
+		event->mmap2.header.size += machine->id_hdr_size;
+		event->mmap2.pid = tgid;
+		event->mmap2.tid = pid;
+
+		if (process(tool, event, &synth_sample, machine) != 0) {
+			rc = -1;
+			break;
+		}
+	}
+
+	fclose(fp);
+	return rc;
+}
+
+int perf_event__synthesize_modules(struct perf_tool *tool,
+				   perf_event__handler_t process,
+				   struct machine *machine)
+{
+	int rc = 0;
+	struct rb_node *nd;
+	struct map_groups *kmaps = &machine->kmaps;
+	union perf_event *event = zalloc((sizeof(event->mmap) +
+					  machine->id_hdr_size));
+	if (event == NULL) {
+		pr_debug("Not enough memory synthesizing mmap event "
+			 "for kernel modules\n");
+		return -1;
+	}
+
+	event->header.type = PERF_RECORD_MMAP;
+
+	/*
+	 * kernel uses 0 for user space maps, see kernel/perf_event.c
+	 * __perf_event_mmap
+	 */
+	if (machine__is_host(machine))
+		event->header.misc = PERF_RECORD_MISC_KERNEL;
+	else
+		event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL;
+
+	for (nd = rb_first(&kmaps->maps[MAP__FUNCTION]);
+	     nd; nd = rb_next(nd)) {
+		size_t size;
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+
+		if (pos->dso->kernel)
+			continue;
+
+		size = PERF_ALIGN(pos->dso->long_name_len + 1, sizeof(u64));
+		event->mmap.header.type = PERF_RECORD_MMAP;
+		event->mmap.header.size = (sizeof(event->mmap) -
+				        (sizeof(event->mmap.filename) - size));
+		memset(event->mmap.filename + size, 0, machine->id_hdr_size);
+		event->mmap.header.size += machine->id_hdr_size;
+		event->mmap.start = pos->start;
+		event->mmap.len   = pos->end - pos->start;
+		event->mmap.pid   = machine->pid;
+
+		memcpy(event->mmap.filename, pos->dso->long_name,
+		       pos->dso->long_name_len + 1);
+		if (process(tool, event, &synth_sample, machine) != 0) {
+			rc = -1;
+			break;
+		}
+	}
+
+	free(event);
+	return rc;
+}
+
+static int __event__synthesize_thread(union perf_event *comm_event,
+				      union perf_event *mmap_event,
+				      union perf_event *fork_event,
+				      pid_t pid, int full,
+					  perf_event__handler_t process,
+				      struct perf_tool *tool,
+				      struct machine *machine, bool mmap_data)
+{
+	char filename[PATH_MAX];
+	DIR *tasks;
+	struct dirent dirent, *next;
+	pid_t tgid, ppid;
+	int rc = 0;
+
+	/* special case: only send one comm event using passed in pid */
+	if (!full) {
+		tgid = perf_event__synthesize_comm(tool, comm_event, pid,
+						   process, machine);
+
+		if (tgid == -1)
+			return -1;
+
+		return perf_event__synthesize_mmap_events(tool, mmap_event, pid, tgid,
+							  process, machine, mmap_data);
+	}
+
+	if (machine__is_default_guest(machine))
+		return 0;
+
+	snprintf(filename, sizeof(filename), "%s/proc/%d/task",
+		 machine->root_dir, pid);
+
+	tasks = opendir(filename);
+	if (tasks == NULL) {
+		pr_debug("couldn't open %s\n", filename);
+		return 0;
+	}
+
+	while (!readdir_r(tasks, &dirent, &next) && next) {
+		char *end;
+		pid_t _pid;
+
+		_pid = strtol(dirent.d_name, &end, 10);
+		if (*end)
+			continue;
+
+		rc = -1;
+		if (perf_event__prepare_comm(comm_event, _pid, machine,
+					     &tgid, &ppid) != 0)
+			break;
+
+		if (perf_event__synthesize_fork(tool, fork_event, _pid, tgid,
+						ppid, process, machine) < 0)
+			break;
+		/*
+		 * Send the prepared comm event
+		 */
+		if (process(tool, comm_event, &synth_sample, machine) != 0)
+			break;
+
+		rc = 0;
+		if (_pid == pid) {
+			/* process the parent's maps too */
+			rc = perf_event__synthesize_mmap_events(tool, mmap_event, pid, tgid,
+						process, machine, mmap_data);
+			if (rc)
+				break;
+		}
+	}
+
+	closedir(tasks);
+	return rc;
+}
+
+int perf_event__synthesize_thread_map(struct perf_tool *tool,
+				      struct thread_map *threads,
+				      perf_event__handler_t process,
+				      struct machine *machine,
+				      bool mmap_data)
+{
+	union perf_event *comm_event, *mmap_event, *fork_event;
+	int err = -1, thread, j;
+
+	comm_event = malloc(sizeof(comm_event->comm) + machine->id_hdr_size);
+	if (comm_event == NULL)
+		goto out;
+
+	mmap_event = malloc(sizeof(mmap_event->mmap) + machine->id_hdr_size);
+	if (mmap_event == NULL)
+		goto out_free_comm;
+
+	fork_event = malloc(sizeof(fork_event->fork) + machine->id_hdr_size);
+	if (fork_event == NULL)
+		goto out_free_mmap;
+
+	err = 0;
+	for (thread = 0; thread < threads->nr; ++thread) {
+		if (__event__synthesize_thread(comm_event, mmap_event,
+					       fork_event,
+					       threads->map[thread], 0,
+					       process, tool, machine,
+					       mmap_data)) {
+			err = -1;
+			break;
+		}
+
+		/*
+		 * comm.pid is set to thread group id by
+		 * perf_event__synthesize_comm
+		 */
+		if ((int) comm_event->comm.pid != threads->map[thread]) {
+			bool need_leader = true;
+
+			/* is thread group leader in thread_map? */
+			for (j = 0; j < threads->nr; ++j) {
+				if ((int) comm_event->comm.pid == threads->map[j]) {
+					need_leader = false;
+					break;
+				}
+			}
+
+			/* if not, generate events for it */
+			if (need_leader &&
+			    __event__synthesize_thread(comm_event, mmap_event,
+						       fork_event,
+						       comm_event->comm.pid, 0,
+						       process, tool, machine,
+						       mmap_data)) {
+				err = -1;
+				break;
+			}
+		}
+	}
+	free(fork_event);
+out_free_mmap:
+	free(mmap_event);
+out_free_comm:
+	free(comm_event);
+out:
+	return err;
+}
+
+int perf_event__synthesize_threads(struct perf_tool *tool,
+				   perf_event__handler_t process,
+				   struct machine *machine, bool mmap_data)
+{
+	DIR *proc;
+	char proc_path[PATH_MAX];
+	struct dirent dirent, *next;
+	union perf_event *comm_event, *mmap_event, *fork_event;
+	int err = -1;
+
+	if (machine__is_default_guest(machine))
+		return 0;
+
+	comm_event = malloc(sizeof(comm_event->comm) + machine->id_hdr_size);
+	if (comm_event == NULL)
+		goto out;
+
+	mmap_event = malloc(sizeof(mmap_event->mmap) + machine->id_hdr_size);
+	if (mmap_event == NULL)
+		goto out_free_comm;
+
+	fork_event = malloc(sizeof(fork_event->fork) + machine->id_hdr_size);
+	if (fork_event == NULL)
+		goto out_free_mmap;
+
+	snprintf(proc_path, sizeof(proc_path), "%s/proc", machine->root_dir);
+	proc = opendir(proc_path);
+
+	if (proc == NULL)
+		goto out_free_fork;
+
+	while (!readdir_r(proc, &dirent, &next) && next) {
+		char *end;
+		pid_t pid = strtol(dirent.d_name, &end, 10);
+
+		if (*end) /* only interested in proper numerical dirents */
+			continue;
+		/*
+ 		 * We may race with exiting thread, so don't stop just because
+ 		 * one thread couldn't be synthesized.
+ 		 */
+		__event__synthesize_thread(comm_event, mmap_event, fork_event, pid,
+					   1, process, tool, machine, mmap_data);
+	}
+
+	err = 0;
+	closedir(proc);
+out_free_fork:
+	free(fork_event);
+out_free_mmap:
+	free(mmap_event);
+out_free_comm:
+	free(comm_event);
+out:
+	return err;
+}
+
+struct process_symbol_args {
+	const char *name;
+	u64	   start;
+};
+
+static int find_symbol_cb(void *arg, const char *name, char type,
+			  u64 start)
+{
+	struct process_symbol_args *args = arg;
+
+	/*
+	 * Must be a function or at least an alias, as in PARISC64, where "_text" is
+	 * an 'A' to the same address as "_stext".
+	 */
+	if (!(symbol_type__is_a(type, MAP__FUNCTION) ||
+	      type == 'A') || strcmp(name, args->name))
+		return 0;
+
+	args->start = start;
+	return 1;
+}
+
+u64 kallsyms__get_function_start(const char *kallsyms_filename,
+				 const char *symbol_name)
+{
+	struct process_symbol_args args = { .name = symbol_name, };
+
+	if (kallsyms__parse(kallsyms_filename, &args, find_symbol_cb) <= 0)
+		return 0;
+
+	return args.start;
+}
+
+int perf_event__synthesize_kernel_mmap(struct perf_tool *tool,
+				       perf_event__handler_t process,
+				       struct machine *machine)
+{
+	size_t size;
+	const char *mmap_name;
+	char name_buff[PATH_MAX];
+	struct map *map;
+	struct kmap *kmap;
+	int err;
+	union perf_event *event;
+
+	if (machine->vmlinux_maps[0] == NULL)
+		return -1;
+
+	/*
+	 * We should get this from /sys/kernel/sections/.text, but till that is
+	 * available use this, and after it is use this as a fallback for older
+	 * kernels.
+	 */
+	event = zalloc((sizeof(event->mmap) + machine->id_hdr_size));
+	if (event == NULL) {
+		pr_debug("Not enough memory synthesizing mmap event "
+			 "for kernel modules\n");
+		return -1;
+	}
+
+	mmap_name = machine__mmap_name(machine, name_buff, sizeof(name_buff));
+	if (machine__is_host(machine)) {
+		/*
+		 * kernel uses PERF_RECORD_MISC_USER for user space maps,
+		 * see kernel/perf_event.c __perf_event_mmap
+		 */
+		event->header.misc = PERF_RECORD_MISC_KERNEL;
+	} else {
+		event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL;
+	}
+
+	map = machine->vmlinux_maps[MAP__FUNCTION];
+	kmap = map__kmap(map);
+	size = snprintf(event->mmap.filename, sizeof(event->mmap.filename),
+			"%s%s", mmap_name, kmap->ref_reloc_sym->name) + 1;
+	size = PERF_ALIGN(size, sizeof(u64));
+	event->mmap.header.type = PERF_RECORD_MMAP;
+	event->mmap.header.size = (sizeof(event->mmap) -
+			(sizeof(event->mmap.filename) - size) + machine->id_hdr_size);
+	event->mmap.pgoff = kmap->ref_reloc_sym->addr;
+	event->mmap.start = map->start;
+	event->mmap.len   = map->end - event->mmap.start;
+	event->mmap.pid   = machine->pid;
+
+	err = process(tool, event, &synth_sample, machine);
+	free(event);
+
+	return err;
+}
+
+size_t perf_event__fprintf_comm(union perf_event *event, FILE *fp)
+{
+	const char *s;
+
+	if (event->header.misc & PERF_RECORD_MISC_COMM_EXEC)
+		s = " exec";
+	else
+		s = "";
+
+	return fprintf(fp, "%s: %s:%d/%d\n", s, event->comm.comm, event->comm.pid, event->comm.tid);
+}
+
+int perf_event__process_comm(struct perf_tool *tool __maybe_unused,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine)
+{
+	return machine__process_comm_event(machine, event, sample);
+}
+
+int perf_event__process_lost(struct perf_tool *tool __maybe_unused,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine)
+{
+	return machine__process_lost_event(machine, event, sample);
+}
+
+size_t perf_event__fprintf_mmap(union perf_event *event, FILE *fp)
+{
+	return fprintf(fp, " %d/%d: [%#" PRIx64 "(%#" PRIx64 ") @ %#" PRIx64 "]: %c %s\n",
+		       event->mmap.pid, event->mmap.tid, event->mmap.start,
+		       event->mmap.len, event->mmap.pgoff,
+		       (event->header.misc & PERF_RECORD_MISC_MMAP_DATA) ? 'r' : 'x',
+		       event->mmap.filename);
+}
+
+size_t perf_event__fprintf_mmap2(union perf_event *event, FILE *fp)
+{
+	return fprintf(fp, " %d/%d: [%#" PRIx64 "(%#" PRIx64 ") @ %#" PRIx64
+			   " %02x:%02x %"PRIu64" %"PRIu64"]: %c%c%c%c %s\n",
+		       event->mmap2.pid, event->mmap2.tid, event->mmap2.start,
+		       event->mmap2.len, event->mmap2.pgoff, event->mmap2.maj,
+		       event->mmap2.min, event->mmap2.ino,
+		       event->mmap2.ino_generation,
+		       (event->mmap2.prot & PROT_READ) ? 'r' : '-',
+		       (event->mmap2.prot & PROT_WRITE) ? 'w' : '-',
+		       (event->mmap2.prot & PROT_EXEC) ? 'x' : '-',
+		       (event->mmap2.flags & MAP_SHARED) ? 's' : 'p',
+		       event->mmap2.filename);
+}
+
+int perf_event__process_mmap(struct perf_tool *tool __maybe_unused,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine)
+{
+	return machine__process_mmap_event(machine, event, sample);
+}
+
+int perf_event__process_mmap2(struct perf_tool *tool __maybe_unused,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine)
+{
+	return machine__process_mmap2_event(machine, event, sample);
+}
+
+size_t perf_event__fprintf_task(union perf_event *event, FILE *fp)
+{
+	return fprintf(fp, "(%d:%d):(%d:%d)\n",
+		       event->fork.pid, event->fork.tid,
+		       event->fork.ppid, event->fork.ptid);
+}
+
+int perf_event__process_fork(struct perf_tool *tool __maybe_unused,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine)
+{
+	return machine__process_fork_event(machine, event, sample);
+}
+
+int perf_event__process_exit(struct perf_tool *tool __maybe_unused,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine)
+{
+	return machine__process_exit_event(machine, event, sample);
+}
+
+size_t perf_event__fprintf(union perf_event *event, FILE *fp)
+{
+	size_t ret = fprintf(fp, "PERF_RECORD_%s",
+			     perf_event__name(event->header.type));
+
+	switch (event->header.type) {
+	case PERF_RECORD_COMM:
+		ret += perf_event__fprintf_comm(event, fp);
+		break;
+	case PERF_RECORD_FORK:
+	case PERF_RECORD_EXIT:
+		ret += perf_event__fprintf_task(event, fp);
+		break;
+	case PERF_RECORD_MMAP:
+		ret += perf_event__fprintf_mmap(event, fp);
+		break;
+	case PERF_RECORD_MMAP2:
+		ret += perf_event__fprintf_mmap2(event, fp);
+		break;
+	default:
+		ret += fprintf(fp, "\n");
+	}
+
+	return ret;
+}
+
+int perf_event__process(struct perf_tool *tool __maybe_unused,
+			union perf_event *event,
+			struct perf_sample *sample,
+			struct machine *machine)
+{
+	return machine__process_event(machine, event, sample);
+}
+
+void thread__find_addr_map(struct thread *thread, u8 cpumode,
+			   enum map_type type, u64 addr,
+			   struct addr_location *al)
+{
+	struct map_groups *mg = thread->mg;
+	struct machine *machine = mg->machine;
+	bool load_map = false;
+
+	al->machine = machine;
+	al->thread = thread;
+	al->addr = addr;
+	al->cpumode = cpumode;
+	al->filtered = 0;
+
+	if (machine == NULL) {
+		al->map = NULL;
+		return;
+	}
+
+	if (cpumode == PERF_RECORD_MISC_KERNEL && perf_host) {
+		al->level = 'k';
+		mg = &machine->kmaps;
+		load_map = true;
+	} else if (cpumode == PERF_RECORD_MISC_USER && perf_host) {
+		al->level = '.';
+	} else if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL && perf_guest) {
+		al->level = 'g';
+		mg = &machine->kmaps;
+		load_map = true;
+	} else if (cpumode == PERF_RECORD_MISC_GUEST_USER && perf_guest) {
+		al->level = 'u';
+	} else {
+		al->level = 'H';
+		al->map = NULL;
+
+		if ((cpumode == PERF_RECORD_MISC_GUEST_USER ||
+			cpumode == PERF_RECORD_MISC_GUEST_KERNEL) &&
+			!perf_guest)
+			al->filtered |= (1 << HIST_FILTER__GUEST);
+		if ((cpumode == PERF_RECORD_MISC_USER ||
+			cpumode == PERF_RECORD_MISC_KERNEL) &&
+			!perf_host)
+			al->filtered |= (1 << HIST_FILTER__HOST);
+
+		return;
+	}
+try_again:
+	al->map = map_groups__find(mg, type, al->addr);
+	if (al->map == NULL) {
+		/*
+		 * If this is outside of all known maps, and is a negative
+		 * address, try to look it up in the kernel dso, as it might be
+		 * a vsyscall or vdso (which executes in user-mode).
+		 *
+		 * XXX This is nasty, we should have a symbol list in the
+		 * "[vdso]" dso, but for now lets use the old trick of looking
+		 * in the whole kernel symbol list.
+		 */
+		if (cpumode == PERF_RECORD_MISC_USER && machine &&
+		    mg != &machine->kmaps &&
+		    machine__kernel_ip(machine, al->addr)) {
+			mg = &machine->kmaps;
+			load_map = true;
+			goto try_again;
+		}
+	} else {
+		/*
+		 * Kernel maps might be changed when loading symbols so loading
+		 * must be done prior to using kernel maps.
+		 */
+		if (load_map)
+			map__load(al->map, machine->symbol_filter);
+		al->addr = al->map->map_ip(al->map, al->addr);
+	}
+}
+
+void thread__find_addr_location(struct thread *thread,
+				u8 cpumode, enum map_type type, u64 addr,
+				struct addr_location *al)
+{
+	thread__find_addr_map(thread, cpumode, type, addr, al);
+	if (al->map != NULL)
+		al->sym = map__find_symbol(al->map, al->addr,
+					   thread->mg->machine->symbol_filter);
+	else
+		al->sym = NULL;
+}
+
+int perf_event__preprocess_sample(const union perf_event *event,
+				  struct machine *machine,
+				  struct addr_location *al,
+				  struct perf_sample *sample)
+{
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct thread *thread = machine__findnew_thread(machine, sample->pid,
+							sample->tid);
+
+	if (thread == NULL)
+		return -1;
+
+	dump_printf(" ... thread: %s:%d\n", thread__comm_str(thread), thread->tid);
+	/*
+	 * Have we already created the kernel maps for this machine?
+	 *
+	 * This should have happened earlier, when we processed the kernel MMAP
+	 * events, but for older perf.data files there was no such thing, so do
+	 * it now.
+	 */
+	if (cpumode == PERF_RECORD_MISC_KERNEL &&
+	    machine->vmlinux_maps[MAP__FUNCTION] == NULL)
+		machine__create_kernel_maps(machine);
+
+	thread__find_addr_map(thread, cpumode, MAP__FUNCTION, sample->ip, al);
+	dump_printf(" ...... dso: %s\n",
+		    al->map ? al->map->dso->long_name :
+			al->level == 'H' ? "[hypervisor]" : "<not found>");
+
+	if (thread__is_filtered(thread))
+		al->filtered |= (1 << HIST_FILTER__THREAD);
+
+	al->sym = NULL;
+	al->cpu = sample->cpu;
+
+	if (al->map) {
+		struct dso *dso = al->map->dso;
+
+		if (symbol_conf.dso_list &&
+		    (!dso || !(strlist__has_entry(symbol_conf.dso_list,
+						  dso->short_name) ||
+			       (dso->short_name != dso->long_name &&
+				strlist__has_entry(symbol_conf.dso_list,
+						   dso->long_name))))) {
+			al->filtered |= (1 << HIST_FILTER__DSO);
+		}
+
+		al->sym = map__find_symbol(al->map, al->addr,
+					   machine->symbol_filter);
+	}
+
+	if (symbol_conf.sym_list &&
+		(!al->sym || !strlist__has_entry(symbol_conf.sym_list,
+						al->sym->name))) {
+		al->filtered |= (1 << HIST_FILTER__SYMBOL);
+	}
+
+	return 0;
+}
+
+bool is_bts_event(struct perf_event_attr *attr)
+{
+	return attr->type == PERF_TYPE_HARDWARE &&
+	       (attr->config & PERF_COUNT_HW_BRANCH_INSTRUCTIONS) &&
+	       attr->sample_period == 1;
+}
+
+bool sample_addr_correlates_sym(struct perf_event_attr *attr)
+{
+	if (attr->type == PERF_TYPE_SOFTWARE &&
+	    (attr->config == PERF_COUNT_SW_PAGE_FAULTS ||
+	     attr->config == PERF_COUNT_SW_PAGE_FAULTS_MIN ||
+	     attr->config == PERF_COUNT_SW_PAGE_FAULTS_MAJ))
+		return true;
+
+	if (is_bts_event(attr))
+		return true;
+
+	return false;
+}
+
+void perf_event__preprocess_sample_addr(union perf_event *event,
+					struct perf_sample *sample,
+					struct thread *thread,
+					struct addr_location *al)
+{
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+
+	thread__find_addr_map(thread, cpumode, MAP__FUNCTION, sample->addr, al);
+	if (!al->map)
+		thread__find_addr_map(thread, cpumode, MAP__VARIABLE,
+				      sample->addr, al);
+
+	al->cpu = sample->cpu;
+	al->sym = NULL;
+
+	if (al->map)
+		al->sym = map__find_symbol(al->map, al->addr, NULL);
+}
diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h
new file mode 100644
index 000000000..09b9e8d3f
--- /dev/null
+++ b/tools/perf/util/event.h
@@ -0,0 +1,395 @@
+#ifndef __PERF_RECORD_H
+#define __PERF_RECORD_H
+
+#include <limits.h>
+#include <stdio.h>
+
+#include "../perf.h"
+#include "map.h"
+#include "build-id.h"
+#include "perf_regs.h"
+
+struct mmap_event {
+	struct perf_event_header header;
+	u32 pid, tid;
+	u64 start;
+	u64 len;
+	u64 pgoff;
+	char filename[PATH_MAX];
+};
+
+struct mmap2_event {
+	struct perf_event_header header;
+	u32 pid, tid;
+	u64 start;
+	u64 len;
+	u64 pgoff;
+	u32 maj;
+	u32 min;
+	u64 ino;
+	u64 ino_generation;
+	u32 prot;
+	u32 flags;
+	char filename[PATH_MAX];
+};
+
+struct comm_event {
+	struct perf_event_header header;
+	u32 pid, tid;
+	char comm[16];
+};
+
+struct fork_event {
+	struct perf_event_header header;
+	u32 pid, ppid;
+	u32 tid, ptid;
+	u64 time;
+};
+
+struct lost_event {
+	struct perf_event_header header;
+	u64 id;
+	u64 lost;
+};
+
+/*
+ * PERF_FORMAT_ENABLED | PERF_FORMAT_RUNNING | PERF_FORMAT_ID
+ */
+struct read_event {
+	struct perf_event_header header;
+	u32 pid, tid;
+	u64 value;
+	u64 time_enabled;
+	u64 time_running;
+	u64 id;
+};
+
+struct throttle_event {
+	struct perf_event_header header;
+	u64 time;
+	u64 id;
+	u64 stream_id;
+};
+
+#define PERF_SAMPLE_MASK				\
+	(PERF_SAMPLE_IP | PERF_SAMPLE_TID |		\
+	 PERF_SAMPLE_TIME | PERF_SAMPLE_ADDR |		\
+	PERF_SAMPLE_ID | PERF_SAMPLE_STREAM_ID |	\
+	 PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD |		\
+	 PERF_SAMPLE_IDENTIFIER)
+
+/* perf sample has 16 bits size limit */
+#define PERF_SAMPLE_MAX_SIZE (1 << 16)
+
+struct sample_event {
+	struct perf_event_header        header;
+	u64 array[];
+};
+
+struct regs_dump {
+	u64 abi;
+	u64 mask;
+	u64 *regs;
+
+	/* Cached values/mask filled by first register access. */
+	u64 cache_regs[PERF_REGS_MAX];
+	u64 cache_mask;
+};
+
+struct stack_dump {
+	u16 offset;
+	u64 size;
+	char *data;
+};
+
+struct sample_read_value {
+	u64 value;
+	u64 id;
+};
+
+struct sample_read {
+	u64 time_enabled;
+	u64 time_running;
+	union {
+		struct {
+			u64 nr;
+			struct sample_read_value *values;
+		} group;
+		struct sample_read_value one;
+	};
+};
+
+struct ip_callchain {
+	u64 nr;
+	u64 ips[0];
+};
+
+struct branch_flags {
+	u64 mispred:1;
+	u64 predicted:1;
+	u64 in_tx:1;
+	u64 abort:1;
+	u64 reserved:60;
+};
+
+struct branch_entry {
+	u64			from;
+	u64			to;
+	struct branch_flags	flags;
+};
+
+struct branch_stack {
+	u64			nr;
+	struct branch_entry	entries[0];
+};
+
+enum {
+	PERF_IP_FLAG_BRANCH		= 1ULL << 0,
+	PERF_IP_FLAG_CALL		= 1ULL << 1,
+	PERF_IP_FLAG_RETURN		= 1ULL << 2,
+	PERF_IP_FLAG_CONDITIONAL	= 1ULL << 3,
+	PERF_IP_FLAG_SYSCALLRET		= 1ULL << 4,
+	PERF_IP_FLAG_ASYNC		= 1ULL << 5,
+	PERF_IP_FLAG_INTERRUPT		= 1ULL << 6,
+	PERF_IP_FLAG_TX_ABORT		= 1ULL << 7,
+	PERF_IP_FLAG_TRACE_BEGIN	= 1ULL << 8,
+	PERF_IP_FLAG_TRACE_END		= 1ULL << 9,
+	PERF_IP_FLAG_IN_TX		= 1ULL << 10,
+};
+
+#define PERF_BRANCH_MASK		(\
+	PERF_IP_FLAG_BRANCH		|\
+	PERF_IP_FLAG_CALL		|\
+	PERF_IP_FLAG_RETURN		|\
+	PERF_IP_FLAG_CONDITIONAL	|\
+	PERF_IP_FLAG_SYSCALLRET		|\
+	PERF_IP_FLAG_ASYNC		|\
+	PERF_IP_FLAG_INTERRUPT		|\
+	PERF_IP_FLAG_TX_ABORT		|\
+	PERF_IP_FLAG_TRACE_BEGIN	|\
+	PERF_IP_FLAG_TRACE_END)
+
+struct perf_sample {
+	u64 ip;
+	u32 pid, tid;
+	u64 time;
+	u64 addr;
+	u64 id;
+	u64 stream_id;
+	u64 period;
+	u64 weight;
+	u64 transaction;
+	u32 cpu;
+	u32 raw_size;
+	u64 data_src;
+	u32 flags;
+	u16 insn_len;
+	void *raw_data;
+	struct ip_callchain *callchain;
+	struct branch_stack *branch_stack;
+	struct regs_dump  user_regs;
+	struct regs_dump  intr_regs;
+	struct stack_dump user_stack;
+	struct sample_read read;
+};
+
+#define PERF_MEM_DATA_SRC_NONE \
+	(PERF_MEM_S(OP, NA) |\
+	 PERF_MEM_S(LVL, NA) |\
+	 PERF_MEM_S(SNOOP, NA) |\
+	 PERF_MEM_S(LOCK, NA) |\
+	 PERF_MEM_S(TLB, NA))
+
+struct build_id_event {
+	struct perf_event_header header;
+	pid_t			 pid;
+	u8			 build_id[PERF_ALIGN(BUILD_ID_SIZE, sizeof(u64))];
+	char			 filename[];
+};
+
+enum perf_user_event_type { /* above any possible kernel type */
+	PERF_RECORD_USER_TYPE_START		= 64,
+	PERF_RECORD_HEADER_ATTR			= 64,
+	PERF_RECORD_HEADER_EVENT_TYPE		= 65, /* depreceated */
+	PERF_RECORD_HEADER_TRACING_DATA		= 66,
+	PERF_RECORD_HEADER_BUILD_ID		= 67,
+	PERF_RECORD_FINISHED_ROUND		= 68,
+	PERF_RECORD_ID_INDEX			= 69,
+	PERF_RECORD_HEADER_MAX
+};
+
+/*
+ * The kernel collects the number of events it couldn't send in a stretch and
+ * when possible sends this number in a PERF_RECORD_LOST event. The number of
+ * such "chunks" of lost events is stored in .nr_events[PERF_EVENT_LOST] while
+ * total_lost tells exactly how many events the kernel in fact lost, i.e. it is
+ * the sum of all struct lost_event.lost fields reported.
+ *
+ * The total_period is needed because by default auto-freq is used, so
+ * multipling nr_events[PERF_EVENT_SAMPLE] by a frequency isn't possible to get
+ * the total number of low level events, it is necessary to to sum all struct
+ * sample_event.period and stash the result in total_period.
+ */
+struct events_stats {
+	u64 total_period;
+	u64 total_non_filtered_period;
+	u64 total_lost;
+	u64 total_invalid_chains;
+	u32 nr_events[PERF_RECORD_HEADER_MAX];
+	u32 nr_non_filtered_samples;
+	u32 nr_lost_warned;
+	u32 nr_unknown_events;
+	u32 nr_invalid_chains;
+	u32 nr_unknown_id;
+	u32 nr_unprocessable_samples;
+};
+
+struct attr_event {
+	struct perf_event_header header;
+	struct perf_event_attr attr;
+	u64 id[];
+};
+
+#define MAX_EVENT_NAME 64
+
+struct perf_trace_event_type {
+	u64	event_id;
+	char	name[MAX_EVENT_NAME];
+};
+
+struct event_type_event {
+	struct perf_event_header header;
+	struct perf_trace_event_type event_type;
+};
+
+struct tracing_data_event {
+	struct perf_event_header header;
+	u32 size;
+};
+
+struct id_index_entry {
+	u64 id;
+	u64 idx;
+	u64 cpu;
+	u64 tid;
+};
+
+struct id_index_event {
+	struct perf_event_header header;
+	u64 nr;
+	struct id_index_entry entries[0];
+};
+
+union perf_event {
+	struct perf_event_header	header;
+	struct mmap_event		mmap;
+	struct mmap2_event		mmap2;
+	struct comm_event		comm;
+	struct fork_event		fork;
+	struct lost_event		lost;
+	struct read_event		read;
+	struct throttle_event		throttle;
+	struct sample_event		sample;
+	struct attr_event		attr;
+	struct event_type_event		event_type;
+	struct tracing_data_event	tracing_data;
+	struct build_id_event		build_id;
+	struct id_index_event		id_index;
+};
+
+void perf_event__print_totals(void);
+
+struct perf_tool;
+struct thread_map;
+
+typedef int (*perf_event__handler_t)(struct perf_tool *tool,
+				     union perf_event *event,
+				     struct perf_sample *sample,
+				     struct machine *machine);
+
+int perf_event__synthesize_thread_map(struct perf_tool *tool,
+				      struct thread_map *threads,
+				      perf_event__handler_t process,
+				      struct machine *machine, bool mmap_data);
+int perf_event__synthesize_threads(struct perf_tool *tool,
+				   perf_event__handler_t process,
+				   struct machine *machine, bool mmap_data);
+int perf_event__synthesize_kernel_mmap(struct perf_tool *tool,
+				       perf_event__handler_t process,
+				       struct machine *machine);
+
+int perf_event__synthesize_modules(struct perf_tool *tool,
+				   perf_event__handler_t process,
+				   struct machine *machine);
+
+int perf_event__process_comm(struct perf_tool *tool,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine);
+int perf_event__process_lost(struct perf_tool *tool,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine);
+int perf_event__process_mmap(struct perf_tool *tool,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine);
+int perf_event__process_mmap2(struct perf_tool *tool,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine);
+int perf_event__process_fork(struct perf_tool *tool,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine);
+int perf_event__process_exit(struct perf_tool *tool,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct machine *machine);
+int perf_event__process(struct perf_tool *tool,
+			union perf_event *event,
+			struct perf_sample *sample,
+			struct machine *machine);
+
+struct addr_location;
+
+int perf_event__preprocess_sample(const union perf_event *event,
+				  struct machine *machine,
+				  struct addr_location *al,
+				  struct perf_sample *sample);
+
+struct thread;
+
+bool is_bts_event(struct perf_event_attr *attr);
+bool sample_addr_correlates_sym(struct perf_event_attr *attr);
+void perf_event__preprocess_sample_addr(union perf_event *event,
+					struct perf_sample *sample,
+					struct thread *thread,
+					struct addr_location *al);
+
+const char *perf_event__name(unsigned int id);
+
+size_t perf_event__sample_event_size(const struct perf_sample *sample, u64 type,
+				     u64 read_format);
+int perf_event__synthesize_sample(union perf_event *event, u64 type,
+				  u64 read_format,
+				  const struct perf_sample *sample,
+				  bool swapped);
+
+int perf_event__synthesize_mmap_events(struct perf_tool *tool,
+				       union perf_event *event,
+				       pid_t pid, pid_t tgid,
+				       perf_event__handler_t process,
+				       struct machine *machine,
+				       bool mmap_data);
+
+size_t perf_event__fprintf_comm(union perf_event *event, FILE *fp);
+size_t perf_event__fprintf_mmap(union perf_event *event, FILE *fp);
+size_t perf_event__fprintf_mmap2(union perf_event *event, FILE *fp);
+size_t perf_event__fprintf_task(union perf_event *event, FILE *fp);
+size_t perf_event__fprintf(union perf_event *event, FILE *fp);
+
+u64 kallsyms__get_function_start(const char *kallsyms_filename,
+				 const char *symbol_name);
+
+#endif /* __PERF_RECORD_H */
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
new file mode 100644
index 000000000..080be93ee
--- /dev/null
+++ b/tools/perf/util/evlist.c
@@ -0,0 +1,1570 @@
+/*
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Parts came from builtin-{top,stat,record}.c, see those files for further
+ * copyright notes.
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+#include "util.h"
+#include <api/fs/fs.h>
+#include <poll.h>
+#include "cpumap.h"
+#include "thread_map.h"
+#include "target.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "debug.h"
+#include <unistd.h>
+
+#include "parse-events.h"
+#include "parse-options.h"
+
+#include <sys/mman.h>
+
+#include <linux/bitops.h>
+#include <linux/hash.h>
+#include <linux/log2.h>
+
+static void perf_evlist__mmap_put(struct perf_evlist *evlist, int idx);
+static void __perf_evlist__munmap(struct perf_evlist *evlist, int idx);
+
+#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
+#define SID(e, x, y) xyarray__entry(e->sample_id, x, y)
+
+void perf_evlist__init(struct perf_evlist *evlist, struct cpu_map *cpus,
+		       struct thread_map *threads)
+{
+	int i;
+
+	for (i = 0; i < PERF_EVLIST__HLIST_SIZE; ++i)
+		INIT_HLIST_HEAD(&evlist->heads[i]);
+	INIT_LIST_HEAD(&evlist->entries);
+	perf_evlist__set_maps(evlist, cpus, threads);
+	fdarray__init(&evlist->pollfd, 64);
+	evlist->workload.pid = -1;
+}
+
+struct perf_evlist *perf_evlist__new(void)
+{
+	struct perf_evlist *evlist = zalloc(sizeof(*evlist));
+
+	if (evlist != NULL)
+		perf_evlist__init(evlist, NULL, NULL);
+
+	return evlist;
+}
+
+struct perf_evlist *perf_evlist__new_default(void)
+{
+	struct perf_evlist *evlist = perf_evlist__new();
+
+	if (evlist && perf_evlist__add_default(evlist)) {
+		perf_evlist__delete(evlist);
+		evlist = NULL;
+	}
+
+	return evlist;
+}
+
+/**
+ * perf_evlist__set_id_pos - set the positions of event ids.
+ * @evlist: selected event list
+ *
+ * Events with compatible sample types all have the same id_pos
+ * and is_pos.  For convenience, put a copy on evlist.
+ */
+void perf_evlist__set_id_pos(struct perf_evlist *evlist)
+{
+	struct perf_evsel *first = perf_evlist__first(evlist);
+
+	evlist->id_pos = first->id_pos;
+	evlist->is_pos = first->is_pos;
+}
+
+static void perf_evlist__update_id_pos(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel)
+		perf_evsel__calc_id_pos(evsel);
+
+	perf_evlist__set_id_pos(evlist);
+}
+
+static void perf_evlist__purge(struct perf_evlist *evlist)
+{
+	struct perf_evsel *pos, *n;
+
+	evlist__for_each_safe(evlist, n, pos) {
+		list_del_init(&pos->node);
+		perf_evsel__delete(pos);
+	}
+
+	evlist->nr_entries = 0;
+}
+
+void perf_evlist__exit(struct perf_evlist *evlist)
+{
+	zfree(&evlist->mmap);
+	fdarray__exit(&evlist->pollfd);
+}
+
+void perf_evlist__delete(struct perf_evlist *evlist)
+{
+	perf_evlist__munmap(evlist);
+	perf_evlist__close(evlist);
+	cpu_map__delete(evlist->cpus);
+	thread_map__delete(evlist->threads);
+	evlist->cpus = NULL;
+	evlist->threads = NULL;
+	perf_evlist__purge(evlist);
+	perf_evlist__exit(evlist);
+	free(evlist);
+}
+
+void perf_evlist__add(struct perf_evlist *evlist, struct perf_evsel *entry)
+{
+	list_add_tail(&entry->node, &evlist->entries);
+	entry->idx = evlist->nr_entries;
+	entry->tracking = !entry->idx;
+
+	if (!evlist->nr_entries++)
+		perf_evlist__set_id_pos(evlist);
+}
+
+void perf_evlist__splice_list_tail(struct perf_evlist *evlist,
+				   struct list_head *list,
+				   int nr_entries)
+{
+	bool set_id_pos = !evlist->nr_entries;
+
+	list_splice_tail(list, &evlist->entries);
+	evlist->nr_entries += nr_entries;
+	if (set_id_pos)
+		perf_evlist__set_id_pos(evlist);
+}
+
+void __perf_evlist__set_leader(struct list_head *list)
+{
+	struct perf_evsel *evsel, *leader;
+
+	leader = list_entry(list->next, struct perf_evsel, node);
+	evsel = list_entry(list->prev, struct perf_evsel, node);
+
+	leader->nr_members = evsel->idx - leader->idx + 1;
+
+	__evlist__for_each(list, evsel) {
+		evsel->leader = leader;
+	}
+}
+
+void perf_evlist__set_leader(struct perf_evlist *evlist)
+{
+	if (evlist->nr_entries) {
+		evlist->nr_groups = evlist->nr_entries > 1 ? 1 : 0;
+		__perf_evlist__set_leader(&evlist->entries);
+	}
+}
+
+int perf_evlist__add_default(struct perf_evlist *evlist)
+{
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_HARDWARE,
+		.config = PERF_COUNT_HW_CPU_CYCLES,
+	};
+	struct perf_evsel *evsel;
+
+	event_attr_init(&attr);
+
+	evsel = perf_evsel__new(&attr);
+	if (evsel == NULL)
+		goto error;
+
+	/* use strdup() because free(evsel) assumes name is allocated */
+	evsel->name = strdup("cycles");
+	if (!evsel->name)
+		goto error_free;
+
+	perf_evlist__add(evlist, evsel);
+	return 0;
+error_free:
+	perf_evsel__delete(evsel);
+error:
+	return -ENOMEM;
+}
+
+static int perf_evlist__add_attrs(struct perf_evlist *evlist,
+				  struct perf_event_attr *attrs, size_t nr_attrs)
+{
+	struct perf_evsel *evsel, *n;
+	LIST_HEAD(head);
+	size_t i;
+
+	for (i = 0; i < nr_attrs; i++) {
+		evsel = perf_evsel__new_idx(attrs + i, evlist->nr_entries + i);
+		if (evsel == NULL)
+			goto out_delete_partial_list;
+		list_add_tail(&evsel->node, &head);
+	}
+
+	perf_evlist__splice_list_tail(evlist, &head, nr_attrs);
+
+	return 0;
+
+out_delete_partial_list:
+	__evlist__for_each_safe(&head, n, evsel)
+		perf_evsel__delete(evsel);
+	return -1;
+}
+
+int __perf_evlist__add_default_attrs(struct perf_evlist *evlist,
+				     struct perf_event_attr *attrs, size_t nr_attrs)
+{
+	size_t i;
+
+	for (i = 0; i < nr_attrs; i++)
+		event_attr_init(attrs + i);
+
+	return perf_evlist__add_attrs(evlist, attrs, nr_attrs);
+}
+
+struct perf_evsel *
+perf_evlist__find_tracepoint_by_id(struct perf_evlist *evlist, int id)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel) {
+		if (evsel->attr.type   == PERF_TYPE_TRACEPOINT &&
+		    (int)evsel->attr.config == id)
+			return evsel;
+	}
+
+	return NULL;
+}
+
+struct perf_evsel *
+perf_evlist__find_tracepoint_by_name(struct perf_evlist *evlist,
+				     const char *name)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel) {
+		if ((evsel->attr.type == PERF_TYPE_TRACEPOINT) &&
+		    (strcmp(evsel->name, name) == 0))
+			return evsel;
+	}
+
+	return NULL;
+}
+
+int perf_evlist__add_newtp(struct perf_evlist *evlist,
+			   const char *sys, const char *name, void *handler)
+{
+	struct perf_evsel *evsel = perf_evsel__newtp(sys, name);
+
+	if (evsel == NULL)
+		return -1;
+
+	evsel->handler = handler;
+	perf_evlist__add(evlist, evsel);
+	return 0;
+}
+
+static int perf_evlist__nr_threads(struct perf_evlist *evlist,
+				   struct perf_evsel *evsel)
+{
+	if (evsel->system_wide)
+		return 1;
+	else
+		return thread_map__nr(evlist->threads);
+}
+
+void perf_evlist__disable(struct perf_evlist *evlist)
+{
+	int cpu, thread;
+	struct perf_evsel *pos;
+	int nr_cpus = cpu_map__nr(evlist->cpus);
+	int nr_threads;
+
+	for (cpu = 0; cpu < nr_cpus; cpu++) {
+		evlist__for_each(evlist, pos) {
+			if (!perf_evsel__is_group_leader(pos) || !pos->fd)
+				continue;
+			nr_threads = perf_evlist__nr_threads(evlist, pos);
+			for (thread = 0; thread < nr_threads; thread++)
+				ioctl(FD(pos, cpu, thread),
+				      PERF_EVENT_IOC_DISABLE, 0);
+		}
+	}
+}
+
+void perf_evlist__enable(struct perf_evlist *evlist)
+{
+	int cpu, thread;
+	struct perf_evsel *pos;
+	int nr_cpus = cpu_map__nr(evlist->cpus);
+	int nr_threads;
+
+	for (cpu = 0; cpu < nr_cpus; cpu++) {
+		evlist__for_each(evlist, pos) {
+			if (!perf_evsel__is_group_leader(pos) || !pos->fd)
+				continue;
+			nr_threads = perf_evlist__nr_threads(evlist, pos);
+			for (thread = 0; thread < nr_threads; thread++)
+				ioctl(FD(pos, cpu, thread),
+				      PERF_EVENT_IOC_ENABLE, 0);
+		}
+	}
+}
+
+int perf_evlist__disable_event(struct perf_evlist *evlist,
+			       struct perf_evsel *evsel)
+{
+	int cpu, thread, err;
+	int nr_cpus = cpu_map__nr(evlist->cpus);
+	int nr_threads = perf_evlist__nr_threads(evlist, evsel);
+
+	if (!evsel->fd)
+		return 0;
+
+	for (cpu = 0; cpu < nr_cpus; cpu++) {
+		for (thread = 0; thread < nr_threads; thread++) {
+			err = ioctl(FD(evsel, cpu, thread),
+				    PERF_EVENT_IOC_DISABLE, 0);
+			if (err)
+				return err;
+		}
+	}
+	return 0;
+}
+
+int perf_evlist__enable_event(struct perf_evlist *evlist,
+			      struct perf_evsel *evsel)
+{
+	int cpu, thread, err;
+	int nr_cpus = cpu_map__nr(evlist->cpus);
+	int nr_threads = perf_evlist__nr_threads(evlist, evsel);
+
+	if (!evsel->fd)
+		return -EINVAL;
+
+	for (cpu = 0; cpu < nr_cpus; cpu++) {
+		for (thread = 0; thread < nr_threads; thread++) {
+			err = ioctl(FD(evsel, cpu, thread),
+				    PERF_EVENT_IOC_ENABLE, 0);
+			if (err)
+				return err;
+		}
+	}
+	return 0;
+}
+
+static int perf_evlist__enable_event_cpu(struct perf_evlist *evlist,
+					 struct perf_evsel *evsel, int cpu)
+{
+	int thread, err;
+	int nr_threads = perf_evlist__nr_threads(evlist, evsel);
+
+	if (!evsel->fd)
+		return -EINVAL;
+
+	for (thread = 0; thread < nr_threads; thread++) {
+		err = ioctl(FD(evsel, cpu, thread),
+			    PERF_EVENT_IOC_ENABLE, 0);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+static int perf_evlist__enable_event_thread(struct perf_evlist *evlist,
+					    struct perf_evsel *evsel,
+					    int thread)
+{
+	int cpu, err;
+	int nr_cpus = cpu_map__nr(evlist->cpus);
+
+	if (!evsel->fd)
+		return -EINVAL;
+
+	for (cpu = 0; cpu < nr_cpus; cpu++) {
+		err = ioctl(FD(evsel, cpu, thread), PERF_EVENT_IOC_ENABLE, 0);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+int perf_evlist__enable_event_idx(struct perf_evlist *evlist,
+				  struct perf_evsel *evsel, int idx)
+{
+	bool per_cpu_mmaps = !cpu_map__empty(evlist->cpus);
+
+	if (per_cpu_mmaps)
+		return perf_evlist__enable_event_cpu(evlist, evsel, idx);
+	else
+		return perf_evlist__enable_event_thread(evlist, evsel, idx);
+}
+
+int perf_evlist__alloc_pollfd(struct perf_evlist *evlist)
+{
+	int nr_cpus = cpu_map__nr(evlist->cpus);
+	int nr_threads = thread_map__nr(evlist->threads);
+	int nfds = 0;
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel) {
+		if (evsel->system_wide)
+			nfds += nr_cpus;
+		else
+			nfds += nr_cpus * nr_threads;
+	}
+
+	if (fdarray__available_entries(&evlist->pollfd) < nfds &&
+	    fdarray__grow(&evlist->pollfd, nfds) < 0)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int __perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd, int idx)
+{
+	int pos = fdarray__add(&evlist->pollfd, fd, POLLIN | POLLERR | POLLHUP);
+	/*
+	 * Save the idx so that when we filter out fds POLLHUP'ed we can
+	 * close the associated evlist->mmap[] entry.
+	 */
+	if (pos >= 0) {
+		evlist->pollfd.priv[pos].idx = idx;
+
+		fcntl(fd, F_SETFL, O_NONBLOCK);
+	}
+
+	return pos;
+}
+
+int perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd)
+{
+	return __perf_evlist__add_pollfd(evlist, fd, -1);
+}
+
+static void perf_evlist__munmap_filtered(struct fdarray *fda, int fd)
+{
+	struct perf_evlist *evlist = container_of(fda, struct perf_evlist, pollfd);
+
+	perf_evlist__mmap_put(evlist, fda->priv[fd].idx);
+}
+
+int perf_evlist__filter_pollfd(struct perf_evlist *evlist, short revents_and_mask)
+{
+	return fdarray__filter(&evlist->pollfd, revents_and_mask,
+			       perf_evlist__munmap_filtered);
+}
+
+int perf_evlist__poll(struct perf_evlist *evlist, int timeout)
+{
+	return fdarray__poll(&evlist->pollfd, timeout);
+}
+
+static void perf_evlist__id_hash(struct perf_evlist *evlist,
+				 struct perf_evsel *evsel,
+				 int cpu, int thread, u64 id)
+{
+	int hash;
+	struct perf_sample_id *sid = SID(evsel, cpu, thread);
+
+	sid->id = id;
+	sid->evsel = evsel;
+	hash = hash_64(sid->id, PERF_EVLIST__HLIST_BITS);
+	hlist_add_head(&sid->node, &evlist->heads[hash]);
+}
+
+void perf_evlist__id_add(struct perf_evlist *evlist, struct perf_evsel *evsel,
+			 int cpu, int thread, u64 id)
+{
+	perf_evlist__id_hash(evlist, evsel, cpu, thread, id);
+	evsel->id[evsel->ids++] = id;
+}
+
+static int perf_evlist__id_add_fd(struct perf_evlist *evlist,
+				  struct perf_evsel *evsel,
+				  int cpu, int thread, int fd)
+{
+	u64 read_data[4] = { 0, };
+	int id_idx = 1; /* The first entry is the counter value */
+	u64 id;
+	int ret;
+
+	ret = ioctl(fd, PERF_EVENT_IOC_ID, &id);
+	if (!ret)
+		goto add;
+
+	if (errno != ENOTTY)
+		return -1;
+
+	/* Legacy way to get event id.. All hail to old kernels! */
+
+	/*
+	 * This way does not work with group format read, so bail
+	 * out in that case.
+	 */
+	if (perf_evlist__read_format(evlist) & PERF_FORMAT_GROUP)
+		return -1;
+
+	if (!(evsel->attr.read_format & PERF_FORMAT_ID) ||
+	    read(fd, &read_data, sizeof(read_data)) == -1)
+		return -1;
+
+	if (evsel->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		++id_idx;
+	if (evsel->attr.read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		++id_idx;
+
+	id = read_data[id_idx];
+
+ add:
+	perf_evlist__id_add(evlist, evsel, cpu, thread, id);
+	return 0;
+}
+
+static void perf_evlist__set_sid_idx(struct perf_evlist *evlist,
+				     struct perf_evsel *evsel, int idx, int cpu,
+				     int thread)
+{
+	struct perf_sample_id *sid = SID(evsel, cpu, thread);
+	sid->idx = idx;
+	if (evlist->cpus && cpu >= 0)
+		sid->cpu = evlist->cpus->map[cpu];
+	else
+		sid->cpu = -1;
+	if (!evsel->system_wide && evlist->threads && thread >= 0)
+		sid->tid = evlist->threads->map[thread];
+	else
+		sid->tid = -1;
+}
+
+struct perf_sample_id *perf_evlist__id2sid(struct perf_evlist *evlist, u64 id)
+{
+	struct hlist_head *head;
+	struct perf_sample_id *sid;
+	int hash;
+
+	hash = hash_64(id, PERF_EVLIST__HLIST_BITS);
+	head = &evlist->heads[hash];
+
+	hlist_for_each_entry(sid, head, node)
+		if (sid->id == id)
+			return sid;
+
+	return NULL;
+}
+
+struct perf_evsel *perf_evlist__id2evsel(struct perf_evlist *evlist, u64 id)
+{
+	struct perf_sample_id *sid;
+
+	if (evlist->nr_entries == 1)
+		return perf_evlist__first(evlist);
+
+	sid = perf_evlist__id2sid(evlist, id);
+	if (sid)
+		return sid->evsel;
+
+	if (!perf_evlist__sample_id_all(evlist))
+		return perf_evlist__first(evlist);
+
+	return NULL;
+}
+
+static int perf_evlist__event2id(struct perf_evlist *evlist,
+				 union perf_event *event, u64 *id)
+{
+	const u64 *array = event->sample.array;
+	ssize_t n;
+
+	n = (event->header.size - sizeof(event->header)) >> 3;
+
+	if (event->header.type == PERF_RECORD_SAMPLE) {
+		if (evlist->id_pos >= n)
+			return -1;
+		*id = array[evlist->id_pos];
+	} else {
+		if (evlist->is_pos > n)
+			return -1;
+		n -= evlist->is_pos;
+		*id = array[n];
+	}
+	return 0;
+}
+
+static struct perf_evsel *perf_evlist__event2evsel(struct perf_evlist *evlist,
+						   union perf_event *event)
+{
+	struct perf_evsel *first = perf_evlist__first(evlist);
+	struct hlist_head *head;
+	struct perf_sample_id *sid;
+	int hash;
+	u64 id;
+
+	if (evlist->nr_entries == 1)
+		return first;
+
+	if (!first->attr.sample_id_all &&
+	    event->header.type != PERF_RECORD_SAMPLE)
+		return first;
+
+	if (perf_evlist__event2id(evlist, event, &id))
+		return NULL;
+
+	/* Synthesized events have an id of zero */
+	if (!id)
+		return first;
+
+	hash = hash_64(id, PERF_EVLIST__HLIST_BITS);
+	head = &evlist->heads[hash];
+
+	hlist_for_each_entry(sid, head, node) {
+		if (sid->id == id)
+			return sid->evsel;
+	}
+	return NULL;
+}
+
+union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx)
+{
+	struct perf_mmap *md = &evlist->mmap[idx];
+	u64 head = perf_mmap__read_head(md);
+	u64 old = md->prev;
+	unsigned char *data = md->base + page_size;
+	union perf_event *event = NULL;
+
+	if (evlist->overwrite) {
+		/*
+		 * If we're further behind than half the buffer, there's a chance
+		 * the writer will bite our tail and mess up the samples under us.
+		 *
+		 * If we somehow ended up ahead of the head, we got messed up.
+		 *
+		 * In either case, truncate and restart at head.
+		 */
+		int diff = head - old;
+		if (diff > md->mask / 2 || diff < 0) {
+			fprintf(stderr, "WARNING: failed to keep up with mmap data.\n");
+
+			/*
+			 * head points to a known good entry, start there.
+			 */
+			old = head;
+		}
+	}
+
+	if (old != head) {
+		size_t size;
+
+		event = (union perf_event *)&data[old & md->mask];
+		size = event->header.size;
+
+		/*
+		 * Event straddles the mmap boundary -- header should always
+		 * be inside due to u64 alignment of output.
+		 */
+		if ((old & md->mask) + size != ((old + size) & md->mask)) {
+			unsigned int offset = old;
+			unsigned int len = min(sizeof(*event), size), cpy;
+			void *dst = md->event_copy;
+
+			do {
+				cpy = min(md->mask + 1 - (offset & md->mask), len);
+				memcpy(dst, &data[offset & md->mask], cpy);
+				offset += cpy;
+				dst += cpy;
+				len -= cpy;
+			} while (len);
+
+			event = (union perf_event *) md->event_copy;
+		}
+
+		old += size;
+	}
+
+	md->prev = old;
+
+	return event;
+}
+
+static bool perf_mmap__empty(struct perf_mmap *md)
+{
+	return perf_mmap__read_head(md) == md->prev;
+}
+
+static void perf_evlist__mmap_get(struct perf_evlist *evlist, int idx)
+{
+	++evlist->mmap[idx].refcnt;
+}
+
+static void perf_evlist__mmap_put(struct perf_evlist *evlist, int idx)
+{
+	BUG_ON(evlist->mmap[idx].refcnt == 0);
+
+	if (--evlist->mmap[idx].refcnt == 0)
+		__perf_evlist__munmap(evlist, idx);
+}
+
+void perf_evlist__mmap_consume(struct perf_evlist *evlist, int idx)
+{
+	struct perf_mmap *md = &evlist->mmap[idx];
+
+	if (!evlist->overwrite) {
+		u64 old = md->prev;
+
+		perf_mmap__write_tail(md, old);
+	}
+
+	if (md->refcnt == 1 && perf_mmap__empty(md))
+		perf_evlist__mmap_put(evlist, idx);
+}
+
+static void __perf_evlist__munmap(struct perf_evlist *evlist, int idx)
+{
+	if (evlist->mmap[idx].base != NULL) {
+		munmap(evlist->mmap[idx].base, evlist->mmap_len);
+		evlist->mmap[idx].base = NULL;
+		evlist->mmap[idx].refcnt = 0;
+	}
+}
+
+void perf_evlist__munmap(struct perf_evlist *evlist)
+{
+	int i;
+
+	if (evlist->mmap == NULL)
+		return;
+
+	for (i = 0; i < evlist->nr_mmaps; i++)
+		__perf_evlist__munmap(evlist, i);
+
+	zfree(&evlist->mmap);
+}
+
+static int perf_evlist__alloc_mmap(struct perf_evlist *evlist)
+{
+	evlist->nr_mmaps = cpu_map__nr(evlist->cpus);
+	if (cpu_map__empty(evlist->cpus))
+		evlist->nr_mmaps = thread_map__nr(evlist->threads);
+	evlist->mmap = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap));
+	return evlist->mmap != NULL ? 0 : -ENOMEM;
+}
+
+struct mmap_params {
+	int prot;
+	int mask;
+};
+
+static int __perf_evlist__mmap(struct perf_evlist *evlist, int idx,
+			       struct mmap_params *mp, int fd)
+{
+	/*
+	 * The last one will be done at perf_evlist__mmap_consume(), so that we
+	 * make sure we don't prevent tools from consuming every last event in
+	 * the ring buffer.
+	 *
+	 * I.e. we can get the POLLHUP meaning that the fd doesn't exist
+	 * anymore, but the last events for it are still in the ring buffer,
+	 * waiting to be consumed.
+	 *
+	 * Tools can chose to ignore this at their own discretion, but the
+	 * evlist layer can't just drop it when filtering events in
+	 * perf_evlist__filter_pollfd().
+	 */
+	evlist->mmap[idx].refcnt = 2;
+	evlist->mmap[idx].prev = 0;
+	evlist->mmap[idx].mask = mp->mask;
+	evlist->mmap[idx].base = mmap(NULL, evlist->mmap_len, mp->prot,
+				      MAP_SHARED, fd, 0);
+	if (evlist->mmap[idx].base == MAP_FAILED) {
+		pr_debug2("failed to mmap perf event ring buffer, error %d\n",
+			  errno);
+		evlist->mmap[idx].base = NULL;
+		return -1;
+	}
+
+	return 0;
+}
+
+static int perf_evlist__mmap_per_evsel(struct perf_evlist *evlist, int idx,
+				       struct mmap_params *mp, int cpu,
+				       int thread, int *output)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel) {
+		int fd;
+
+		if (evsel->system_wide && thread)
+			continue;
+
+		fd = FD(evsel, cpu, thread);
+
+		if (*output == -1) {
+			*output = fd;
+			if (__perf_evlist__mmap(evlist, idx, mp, *output) < 0)
+				return -1;
+		} else {
+			if (ioctl(fd, PERF_EVENT_IOC_SET_OUTPUT, *output) != 0)
+				return -1;
+
+			perf_evlist__mmap_get(evlist, idx);
+		}
+
+		/*
+		 * The system_wide flag causes a selected event to be opened
+		 * always without a pid.  Consequently it will never get a
+		 * POLLHUP, but it is used for tracking in combination with
+		 * other events, so it should not need to be polled anyway.
+		 * Therefore don't add it for polling.
+		 */
+		if (!evsel->system_wide &&
+		    __perf_evlist__add_pollfd(evlist, fd, idx) < 0) {
+			perf_evlist__mmap_put(evlist, idx);
+			return -1;
+		}
+
+		if (evsel->attr.read_format & PERF_FORMAT_ID) {
+			if (perf_evlist__id_add_fd(evlist, evsel, cpu, thread,
+						   fd) < 0)
+				return -1;
+			perf_evlist__set_sid_idx(evlist, evsel, idx, cpu,
+						 thread);
+		}
+	}
+
+	return 0;
+}
+
+static int perf_evlist__mmap_per_cpu(struct perf_evlist *evlist,
+				     struct mmap_params *mp)
+{
+	int cpu, thread;
+	int nr_cpus = cpu_map__nr(evlist->cpus);
+	int nr_threads = thread_map__nr(evlist->threads);
+
+	pr_debug2("perf event ring buffer mmapped per cpu\n");
+	for (cpu = 0; cpu < nr_cpus; cpu++) {
+		int output = -1;
+
+		for (thread = 0; thread < nr_threads; thread++) {
+			if (perf_evlist__mmap_per_evsel(evlist, cpu, mp, cpu,
+							thread, &output))
+				goto out_unmap;
+		}
+	}
+
+	return 0;
+
+out_unmap:
+	for (cpu = 0; cpu < nr_cpus; cpu++)
+		__perf_evlist__munmap(evlist, cpu);
+	return -1;
+}
+
+static int perf_evlist__mmap_per_thread(struct perf_evlist *evlist,
+					struct mmap_params *mp)
+{
+	int thread;
+	int nr_threads = thread_map__nr(evlist->threads);
+
+	pr_debug2("perf event ring buffer mmapped per thread\n");
+	for (thread = 0; thread < nr_threads; thread++) {
+		int output = -1;
+
+		if (perf_evlist__mmap_per_evsel(evlist, thread, mp, 0, thread,
+						&output))
+			goto out_unmap;
+	}
+
+	return 0;
+
+out_unmap:
+	for (thread = 0; thread < nr_threads; thread++)
+		__perf_evlist__munmap(evlist, thread);
+	return -1;
+}
+
+static size_t perf_evlist__mmap_size(unsigned long pages)
+{
+	if (pages == UINT_MAX) {
+		int max;
+
+		if (sysctl__read_int("kernel/perf_event_mlock_kb", &max) < 0) {
+			/*
+			 * Pick a once upon a time good value, i.e. things look
+			 * strange since we can't read a sysctl value, but lets not
+			 * die yet...
+			 */
+			max = 512;
+		} else {
+			max -= (page_size / 1024);
+		}
+
+		pages = (max * 1024) / page_size;
+		if (!is_power_of_2(pages))
+			pages = rounddown_pow_of_two(pages);
+	} else if (!is_power_of_2(pages))
+		return 0;
+
+	return (pages + 1) * page_size;
+}
+
+static long parse_pages_arg(const char *str, unsigned long min,
+			    unsigned long max)
+{
+	unsigned long pages, val;
+	static struct parse_tag tags[] = {
+		{ .tag  = 'B', .mult = 1       },
+		{ .tag  = 'K', .mult = 1 << 10 },
+		{ .tag  = 'M', .mult = 1 << 20 },
+		{ .tag  = 'G', .mult = 1 << 30 },
+		{ .tag  = 0 },
+	};
+
+	if (str == NULL)
+		return -EINVAL;
+
+	val = parse_tag_value(str, tags);
+	if (val != (unsigned long) -1) {
+		/* we got file size value */
+		pages = PERF_ALIGN(val, page_size) / page_size;
+	} else {
+		/* we got pages count value */
+		char *eptr;
+		pages = strtoul(str, &eptr, 10);
+		if (*eptr != '\0')
+			return -EINVAL;
+	}
+
+	if (pages == 0 && min == 0) {
+		/* leave number of pages at 0 */
+	} else if (!is_power_of_2(pages)) {
+		/* round pages up to next power of 2 */
+		pages = roundup_pow_of_two(pages);
+		if (!pages)
+			return -EINVAL;
+		pr_info("rounding mmap pages size to %lu bytes (%lu pages)\n",
+			pages * page_size, pages);
+	}
+
+	if (pages > max)
+		return -EINVAL;
+
+	return pages;
+}
+
+int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,
+				  int unset __maybe_unused)
+{
+	unsigned int *mmap_pages = opt->value;
+	unsigned long max = UINT_MAX;
+	long pages;
+
+	if (max > SIZE_MAX / page_size)
+		max = SIZE_MAX / page_size;
+
+	pages = parse_pages_arg(str, 1, max);
+	if (pages < 0) {
+		pr_err("Invalid argument for --mmap_pages/-m\n");
+		return -1;
+	}
+
+	*mmap_pages = pages;
+	return 0;
+}
+
+/**
+ * perf_evlist__mmap - Create mmaps to receive events.
+ * @evlist: list of events
+ * @pages: map length in pages
+ * @overwrite: overwrite older events?
+ *
+ * If @overwrite is %false the user needs to signal event consumption using
+ * perf_mmap__write_tail().  Using perf_evlist__mmap_read() does this
+ * automatically.
+ *
+ * Return: %0 on success, negative error code otherwise.
+ */
+int perf_evlist__mmap(struct perf_evlist *evlist, unsigned int pages,
+		      bool overwrite)
+{
+	struct perf_evsel *evsel;
+	const struct cpu_map *cpus = evlist->cpus;
+	const struct thread_map *threads = evlist->threads;
+	struct mmap_params mp = {
+		.prot = PROT_READ | (overwrite ? 0 : PROT_WRITE),
+	};
+
+	if (evlist->mmap == NULL && perf_evlist__alloc_mmap(evlist) < 0)
+		return -ENOMEM;
+
+	if (evlist->pollfd.entries == NULL && perf_evlist__alloc_pollfd(evlist) < 0)
+		return -ENOMEM;
+
+	evlist->overwrite = overwrite;
+	evlist->mmap_len = perf_evlist__mmap_size(pages);
+	pr_debug("mmap size %zuB\n", evlist->mmap_len);
+	mp.mask = evlist->mmap_len - page_size - 1;
+
+	evlist__for_each(evlist, evsel) {
+		if ((evsel->attr.read_format & PERF_FORMAT_ID) &&
+		    evsel->sample_id == NULL &&
+		    perf_evsel__alloc_id(evsel, cpu_map__nr(cpus), threads->nr) < 0)
+			return -ENOMEM;
+	}
+
+	if (cpu_map__empty(cpus))
+		return perf_evlist__mmap_per_thread(evlist, &mp);
+
+	return perf_evlist__mmap_per_cpu(evlist, &mp);
+}
+
+int perf_evlist__create_maps(struct perf_evlist *evlist, struct target *target)
+{
+	evlist->threads = thread_map__new_str(target->pid, target->tid,
+					      target->uid);
+
+	if (evlist->threads == NULL)
+		return -1;
+
+	if (target__uses_dummy_map(target))
+		evlist->cpus = cpu_map__dummy_new();
+	else
+		evlist->cpus = cpu_map__new(target->cpu_list);
+
+	if (evlist->cpus == NULL)
+		goto out_delete_threads;
+
+	return 0;
+
+out_delete_threads:
+	thread_map__delete(evlist->threads);
+	evlist->threads = NULL;
+	return -1;
+}
+
+int perf_evlist__apply_filters(struct perf_evlist *evlist, struct perf_evsel **err_evsel)
+{
+	struct perf_evsel *evsel;
+	int err = 0;
+	const int ncpus = cpu_map__nr(evlist->cpus),
+		  nthreads = thread_map__nr(evlist->threads);
+
+	evlist__for_each(evlist, evsel) {
+		if (evsel->filter == NULL)
+			continue;
+
+		err = perf_evsel__set_filter(evsel, ncpus, nthreads, evsel->filter);
+		if (err) {
+			*err_evsel = evsel;
+			break;
+		}
+	}
+
+	return err;
+}
+
+int perf_evlist__set_filter(struct perf_evlist *evlist, const char *filter)
+{
+	struct perf_evsel *evsel;
+	int err = 0;
+	const int ncpus = cpu_map__nr(evlist->cpus),
+		  nthreads = thread_map__nr(evlist->threads);
+
+	evlist__for_each(evlist, evsel) {
+		err = perf_evsel__set_filter(evsel, ncpus, nthreads, filter);
+		if (err)
+			break;
+	}
+
+	return err;
+}
+
+int perf_evlist__set_filter_pids(struct perf_evlist *evlist, size_t npids, pid_t *pids)
+{
+	char *filter;
+	int ret = -1;
+	size_t i;
+
+	for (i = 0; i < npids; ++i) {
+		if (i == 0) {
+			if (asprintf(&filter, "common_pid != %d", pids[i]) < 0)
+				return -1;
+		} else {
+			char *tmp;
+
+			if (asprintf(&tmp, "%s && common_pid != %d", filter, pids[i]) < 0)
+				goto out_free;
+
+			free(filter);
+			filter = tmp;
+		}
+	}
+
+	ret = perf_evlist__set_filter(evlist, filter);
+out_free:
+	free(filter);
+	return ret;
+}
+
+int perf_evlist__set_filter_pid(struct perf_evlist *evlist, pid_t pid)
+{
+	return perf_evlist__set_filter_pids(evlist, 1, &pid);
+}
+
+bool perf_evlist__valid_sample_type(struct perf_evlist *evlist)
+{
+	struct perf_evsel *pos;
+
+	if (evlist->nr_entries == 1)
+		return true;
+
+	if (evlist->id_pos < 0 || evlist->is_pos < 0)
+		return false;
+
+	evlist__for_each(evlist, pos) {
+		if (pos->id_pos != evlist->id_pos ||
+		    pos->is_pos != evlist->is_pos)
+			return false;
+	}
+
+	return true;
+}
+
+u64 __perf_evlist__combined_sample_type(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+
+	if (evlist->combined_sample_type)
+		return evlist->combined_sample_type;
+
+	evlist__for_each(evlist, evsel)
+		evlist->combined_sample_type |= evsel->attr.sample_type;
+
+	return evlist->combined_sample_type;
+}
+
+u64 perf_evlist__combined_sample_type(struct perf_evlist *evlist)
+{
+	evlist->combined_sample_type = 0;
+	return __perf_evlist__combined_sample_type(evlist);
+}
+
+bool perf_evlist__valid_read_format(struct perf_evlist *evlist)
+{
+	struct perf_evsel *first = perf_evlist__first(evlist), *pos = first;
+	u64 read_format = first->attr.read_format;
+	u64 sample_type = first->attr.sample_type;
+
+	evlist__for_each(evlist, pos) {
+		if (read_format != pos->attr.read_format)
+			return false;
+	}
+
+	/* PERF_SAMPLE_READ imples PERF_FORMAT_ID. */
+	if ((sample_type & PERF_SAMPLE_READ) &&
+	    !(read_format & PERF_FORMAT_ID)) {
+		return false;
+	}
+
+	return true;
+}
+
+u64 perf_evlist__read_format(struct perf_evlist *evlist)
+{
+	struct perf_evsel *first = perf_evlist__first(evlist);
+	return first->attr.read_format;
+}
+
+u16 perf_evlist__id_hdr_size(struct perf_evlist *evlist)
+{
+	struct perf_evsel *first = perf_evlist__first(evlist);
+	struct perf_sample *data;
+	u64 sample_type;
+	u16 size = 0;
+
+	if (!first->attr.sample_id_all)
+		goto out;
+
+	sample_type = first->attr.sample_type;
+
+	if (sample_type & PERF_SAMPLE_TID)
+		size += sizeof(data->tid) * 2;
+
+       if (sample_type & PERF_SAMPLE_TIME)
+		size += sizeof(data->time);
+
+	if (sample_type & PERF_SAMPLE_ID)
+		size += sizeof(data->id);
+
+	if (sample_type & PERF_SAMPLE_STREAM_ID)
+		size += sizeof(data->stream_id);
+
+	if (sample_type & PERF_SAMPLE_CPU)
+		size += sizeof(data->cpu) * 2;
+
+	if (sample_type & PERF_SAMPLE_IDENTIFIER)
+		size += sizeof(data->id);
+out:
+	return size;
+}
+
+bool perf_evlist__valid_sample_id_all(struct perf_evlist *evlist)
+{
+	struct perf_evsel *first = perf_evlist__first(evlist), *pos = first;
+
+	evlist__for_each_continue(evlist, pos) {
+		if (first->attr.sample_id_all != pos->attr.sample_id_all)
+			return false;
+	}
+
+	return true;
+}
+
+bool perf_evlist__sample_id_all(struct perf_evlist *evlist)
+{
+	struct perf_evsel *first = perf_evlist__first(evlist);
+	return first->attr.sample_id_all;
+}
+
+void perf_evlist__set_selected(struct perf_evlist *evlist,
+			       struct perf_evsel *evsel)
+{
+	evlist->selected = evsel;
+}
+
+void perf_evlist__close(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+	int ncpus = cpu_map__nr(evlist->cpus);
+	int nthreads = thread_map__nr(evlist->threads);
+	int n;
+
+	evlist__for_each_reverse(evlist, evsel) {
+		n = evsel->cpus ? evsel->cpus->nr : ncpus;
+		perf_evsel__close(evsel, n, nthreads);
+	}
+}
+
+static int perf_evlist__create_syswide_maps(struct perf_evlist *evlist)
+{
+	int err = -ENOMEM;
+
+	/*
+	 * Try reading /sys/devices/system/cpu/online to get
+	 * an all cpus map.
+	 *
+	 * FIXME: -ENOMEM is the best we can do here, the cpu_map
+	 * code needs an overhaul to properly forward the
+	 * error, and we may not want to do that fallback to a
+	 * default cpu identity map :-\
+	 */
+	evlist->cpus = cpu_map__new(NULL);
+	if (evlist->cpus == NULL)
+		goto out;
+
+	evlist->threads = thread_map__new_dummy();
+	if (evlist->threads == NULL)
+		goto out_free_cpus;
+
+	err = 0;
+out:
+	return err;
+out_free_cpus:
+	cpu_map__delete(evlist->cpus);
+	evlist->cpus = NULL;
+	goto out;
+}
+
+int perf_evlist__open(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+	int err;
+
+	/*
+	 * Default: one fd per CPU, all threads, aka systemwide
+	 * as sys_perf_event_open(cpu = -1, thread = -1) is EINVAL
+	 */
+	if (evlist->threads == NULL && evlist->cpus == NULL) {
+		err = perf_evlist__create_syswide_maps(evlist);
+		if (err < 0)
+			goto out_err;
+	}
+
+	perf_evlist__update_id_pos(evlist);
+
+	evlist__for_each(evlist, evsel) {
+		err = perf_evsel__open(evsel, evlist->cpus, evlist->threads);
+		if (err < 0)
+			goto out_err;
+	}
+
+	return 0;
+out_err:
+	perf_evlist__close(evlist);
+	errno = -err;
+	return err;
+}
+
+int perf_evlist__prepare_workload(struct perf_evlist *evlist, struct target *target,
+				  const char *argv[], bool pipe_output,
+				  void (*exec_error)(int signo, siginfo_t *info, void *ucontext))
+{
+	int child_ready_pipe[2], go_pipe[2];
+	char bf;
+
+	if (pipe(child_ready_pipe) < 0) {
+		perror("failed to create 'ready' pipe");
+		return -1;
+	}
+
+	if (pipe(go_pipe) < 0) {
+		perror("failed to create 'go' pipe");
+		goto out_close_ready_pipe;
+	}
+
+	evlist->workload.pid = fork();
+	if (evlist->workload.pid < 0) {
+		perror("failed to fork");
+		goto out_close_pipes;
+	}
+
+	if (!evlist->workload.pid) {
+		int ret;
+
+		if (pipe_output)
+			dup2(2, 1);
+
+		signal(SIGTERM, SIG_DFL);
+
+		close(child_ready_pipe[0]);
+		close(go_pipe[1]);
+		fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
+
+		/*
+		 * Tell the parent we're ready to go
+		 */
+		close(child_ready_pipe[1]);
+
+		/*
+		 * Wait until the parent tells us to go.
+		 */
+		ret = read(go_pipe[0], &bf, 1);
+		/*
+		 * The parent will ask for the execvp() to be performed by
+		 * writing exactly one byte, in workload.cork_fd, usually via
+		 * perf_evlist__start_workload().
+		 *
+		 * For cancelling the workload without actually running it,
+		 * the parent will just close workload.cork_fd, without writing
+		 * anything, i.e. read will return zero and we just exit()
+		 * here.
+		 */
+		if (ret != 1) {
+			if (ret == -1)
+				perror("unable to read pipe");
+			exit(ret);
+		}
+
+		execvp(argv[0], (char **)argv);
+
+		if (exec_error) {
+			union sigval val;
+
+			val.sival_int = errno;
+			if (sigqueue(getppid(), SIGUSR1, val))
+				perror(argv[0]);
+		} else
+			perror(argv[0]);
+		exit(-1);
+	}
+
+	if (exec_error) {
+		struct sigaction act = {
+			.sa_flags     = SA_SIGINFO,
+			.sa_sigaction = exec_error,
+		};
+		sigaction(SIGUSR1, &act, NULL);
+	}
+
+	if (target__none(target)) {
+		if (evlist->threads == NULL) {
+			fprintf(stderr, "FATAL: evlist->threads need to be set at this point (%s:%d).\n",
+				__func__, __LINE__);
+			goto out_close_pipes;
+		}
+		evlist->threads->map[0] = evlist->workload.pid;
+	}
+
+	close(child_ready_pipe[1]);
+	close(go_pipe[0]);
+	/*
+	 * wait for child to settle
+	 */
+	if (read(child_ready_pipe[0], &bf, 1) == -1) {
+		perror("unable to read pipe");
+		goto out_close_pipes;
+	}
+
+	fcntl(go_pipe[1], F_SETFD, FD_CLOEXEC);
+	evlist->workload.cork_fd = go_pipe[1];
+	close(child_ready_pipe[0]);
+	return 0;
+
+out_close_pipes:
+	close(go_pipe[0]);
+	close(go_pipe[1]);
+out_close_ready_pipe:
+	close(child_ready_pipe[0]);
+	close(child_ready_pipe[1]);
+	return -1;
+}
+
+int perf_evlist__start_workload(struct perf_evlist *evlist)
+{
+	if (evlist->workload.cork_fd > 0) {
+		char bf = 0;
+		int ret;
+		/*
+		 * Remove the cork, let it rip!
+		 */
+		ret = write(evlist->workload.cork_fd, &bf, 1);
+		if (ret < 0)
+			perror("enable to write to pipe");
+
+		close(evlist->workload.cork_fd);
+		return ret;
+	}
+
+	return 0;
+}
+
+int perf_evlist__parse_sample(struct perf_evlist *evlist, union perf_event *event,
+			      struct perf_sample *sample)
+{
+	struct perf_evsel *evsel = perf_evlist__event2evsel(evlist, event);
+
+	if (!evsel)
+		return -EFAULT;
+	return perf_evsel__parse_sample(evsel, event, sample);
+}
+
+size_t perf_evlist__fprintf(struct perf_evlist *evlist, FILE *fp)
+{
+	struct perf_evsel *evsel;
+	size_t printed = 0;
+
+	evlist__for_each(evlist, evsel) {
+		printed += fprintf(fp, "%s%s", evsel->idx ? ", " : "",
+				   perf_evsel__name(evsel));
+	}
+
+	return printed + fprintf(fp, "\n");
+}
+
+int perf_evlist__strerror_open(struct perf_evlist *evlist __maybe_unused,
+			       int err, char *buf, size_t size)
+{
+	int printed, value;
+	char sbuf[STRERR_BUFSIZE], *emsg = strerror_r(err, sbuf, sizeof(sbuf));
+
+	switch (err) {
+	case EACCES:
+	case EPERM:
+		printed = scnprintf(buf, size,
+				    "Error:\t%s.\n"
+				    "Hint:\tCheck /proc/sys/kernel/perf_event_paranoid setting.", emsg);
+
+		value = perf_event_paranoid();
+
+		printed += scnprintf(buf + printed, size - printed, "\nHint:\t");
+
+		if (value >= 2) {
+			printed += scnprintf(buf + printed, size - printed,
+					     "For your workloads it needs to be <= 1\nHint:\t");
+		}
+		printed += scnprintf(buf + printed, size - printed,
+				     "For system wide tracing it needs to be set to -1.\n");
+
+		printed += scnprintf(buf + printed, size - printed,
+				    "Hint:\tTry: 'sudo sh -c \"echo -1 > /proc/sys/kernel/perf_event_paranoid\"'\n"
+				    "Hint:\tThe current value is %d.", value);
+		break;
+	default:
+		scnprintf(buf, size, "%s", emsg);
+		break;
+	}
+
+	return 0;
+}
+
+int perf_evlist__strerror_mmap(struct perf_evlist *evlist, int err, char *buf, size_t size)
+{
+	char sbuf[STRERR_BUFSIZE], *emsg = strerror_r(err, sbuf, sizeof(sbuf));
+	int pages_attempted = evlist->mmap_len / 1024, pages_max_per_user, printed = 0;
+
+	switch (err) {
+	case EPERM:
+		sysctl__read_int("kernel/perf_event_mlock_kb", &pages_max_per_user);
+		printed += scnprintf(buf + printed, size - printed,
+				     "Error:\t%s.\n"
+				     "Hint:\tCheck /proc/sys/kernel/perf_event_mlock_kb (%d kB) setting.\n"
+				     "Hint:\tTried using %zd kB.\n",
+				     emsg, pages_max_per_user, pages_attempted);
+
+		if (pages_attempted >= pages_max_per_user) {
+			printed += scnprintf(buf + printed, size - printed,
+					     "Hint:\tTry 'sudo sh -c \"echo %d > /proc/sys/kernel/perf_event_mlock_kb\"', or\n",
+					     pages_max_per_user + pages_attempted);
+		}
+
+		printed += scnprintf(buf + printed, size - printed,
+				     "Hint:\tTry using a smaller -m/--mmap-pages value.");
+		break;
+	default:
+		scnprintf(buf, size, "%s", emsg);
+		break;
+	}
+
+	return 0;
+}
+
+void perf_evlist__to_front(struct perf_evlist *evlist,
+			   struct perf_evsel *move_evsel)
+{
+	struct perf_evsel *evsel, *n;
+	LIST_HEAD(move);
+
+	if (move_evsel == perf_evlist__first(evlist))
+		return;
+
+	evlist__for_each_safe(evlist, n, evsel) {
+		if (evsel->leader == move_evsel->leader)
+			list_move_tail(&evsel->node, &move);
+	}
+
+	list_splice(&move, &evlist->entries);
+}
+
+void perf_evlist__set_tracking_event(struct perf_evlist *evlist,
+				     struct perf_evsel *tracking_evsel)
+{
+	struct perf_evsel *evsel;
+
+	if (tracking_evsel->tracking)
+		return;
+
+	evlist__for_each(evlist, evsel) {
+		if (evsel != tracking_evsel)
+			evsel->tracking = false;
+	}
+
+	tracking_evsel->tracking = true;
+}
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
new file mode 100644
index 000000000..b5cce95d6
--- /dev/null
+++ b/tools/perf/util/evlist.h
@@ -0,0 +1,284 @@
+#ifndef __PERF_EVLIST_H
+#define __PERF_EVLIST_H 1
+
+#include <linux/list.h>
+#include <api/fd/array.h>
+#include <stdio.h>
+#include "../perf.h"
+#include "event.h"
+#include "evsel.h"
+#include "util.h"
+#include <unistd.h>
+
+struct pollfd;
+struct thread_map;
+struct cpu_map;
+struct record_opts;
+
+#define PERF_EVLIST__HLIST_BITS 8
+#define PERF_EVLIST__HLIST_SIZE (1 << PERF_EVLIST__HLIST_BITS)
+
+/**
+ * struct perf_mmap - perf's ring buffer mmap details
+ *
+ * @refcnt - e.g. code using PERF_EVENT_IOC_SET_OUTPUT to share this
+ */
+struct perf_mmap {
+	void		 *base;
+	int		 mask;
+	int		 refcnt;
+	u64		 prev;
+	char		 event_copy[PERF_SAMPLE_MAX_SIZE] __attribute__((aligned(8)));
+};
+
+struct perf_evlist {
+	struct list_head entries;
+	struct hlist_head heads[PERF_EVLIST__HLIST_SIZE];
+	int		 nr_entries;
+	int		 nr_groups;
+	int		 nr_mmaps;
+	size_t		 mmap_len;
+	int		 id_pos;
+	int		 is_pos;
+	u64		 combined_sample_type;
+	struct {
+		int	cork_fd;
+		pid_t	pid;
+	} workload;
+	bool		 overwrite;
+	struct fdarray	 pollfd;
+	struct perf_mmap *mmap;
+	struct thread_map *threads;
+	struct cpu_map	  *cpus;
+	struct perf_evsel *selected;
+	struct events_stats stats;
+};
+
+struct perf_evsel_str_handler {
+	const char *name;
+	void	   *handler;
+};
+
+struct perf_evlist *perf_evlist__new(void);
+struct perf_evlist *perf_evlist__new_default(void);
+void perf_evlist__init(struct perf_evlist *evlist, struct cpu_map *cpus,
+		       struct thread_map *threads);
+void perf_evlist__exit(struct perf_evlist *evlist);
+void perf_evlist__delete(struct perf_evlist *evlist);
+
+void perf_evlist__add(struct perf_evlist *evlist, struct perf_evsel *entry);
+int perf_evlist__add_default(struct perf_evlist *evlist);
+int __perf_evlist__add_default_attrs(struct perf_evlist *evlist,
+				     struct perf_event_attr *attrs, size_t nr_attrs);
+
+#define perf_evlist__add_default_attrs(evlist, array) \
+	__perf_evlist__add_default_attrs(evlist, array, ARRAY_SIZE(array))
+
+int perf_evlist__add_newtp(struct perf_evlist *evlist,
+			   const char *sys, const char *name, void *handler);
+
+int perf_evlist__set_filter(struct perf_evlist *evlist, const char *filter);
+int perf_evlist__set_filter_pid(struct perf_evlist *evlist, pid_t pid);
+int perf_evlist__set_filter_pids(struct perf_evlist *evlist, size_t npids, pid_t *pids);
+
+struct perf_evsel *
+perf_evlist__find_tracepoint_by_id(struct perf_evlist *evlist, int id);
+
+struct perf_evsel *
+perf_evlist__find_tracepoint_by_name(struct perf_evlist *evlist,
+				     const char *name);
+
+void perf_evlist__id_add(struct perf_evlist *evlist, struct perf_evsel *evsel,
+			 int cpu, int thread, u64 id);
+
+int perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd);
+int perf_evlist__alloc_pollfd(struct perf_evlist *evlist);
+int perf_evlist__filter_pollfd(struct perf_evlist *evlist, short revents_and_mask);
+
+int perf_evlist__poll(struct perf_evlist *evlist, int timeout);
+
+struct perf_evsel *perf_evlist__id2evsel(struct perf_evlist *evlist, u64 id);
+
+struct perf_sample_id *perf_evlist__id2sid(struct perf_evlist *evlist, u64 id);
+
+union perf_event *perf_evlist__mmap_read(struct perf_evlist *evlist, int idx);
+
+void perf_evlist__mmap_consume(struct perf_evlist *evlist, int idx);
+
+int perf_evlist__open(struct perf_evlist *evlist);
+void perf_evlist__close(struct perf_evlist *evlist);
+
+void perf_evlist__set_id_pos(struct perf_evlist *evlist);
+bool perf_can_sample_identifier(void);
+void perf_evlist__config(struct perf_evlist *evlist, struct record_opts *opts);
+int record_opts__config(struct record_opts *opts);
+
+int perf_evlist__prepare_workload(struct perf_evlist *evlist,
+				  struct target *target,
+				  const char *argv[], bool pipe_output,
+				  void (*exec_error)(int signo, siginfo_t *info,
+						     void *ucontext));
+int perf_evlist__start_workload(struct perf_evlist *evlist);
+
+struct option;
+
+int perf_evlist__parse_mmap_pages(const struct option *opt,
+				  const char *str,
+				  int unset);
+
+int perf_evlist__mmap(struct perf_evlist *evlist, unsigned int pages,
+		      bool overwrite);
+void perf_evlist__munmap(struct perf_evlist *evlist);
+
+void perf_evlist__disable(struct perf_evlist *evlist);
+void perf_evlist__enable(struct perf_evlist *evlist);
+
+int perf_evlist__disable_event(struct perf_evlist *evlist,
+			       struct perf_evsel *evsel);
+int perf_evlist__enable_event(struct perf_evlist *evlist,
+			      struct perf_evsel *evsel);
+int perf_evlist__enable_event_idx(struct perf_evlist *evlist,
+				  struct perf_evsel *evsel, int idx);
+
+void perf_evlist__set_selected(struct perf_evlist *evlist,
+			       struct perf_evsel *evsel);
+
+static inline void perf_evlist__set_maps(struct perf_evlist *evlist,
+					 struct cpu_map *cpus,
+					 struct thread_map *threads)
+{
+	evlist->cpus	= cpus;
+	evlist->threads	= threads;
+}
+
+int perf_evlist__create_maps(struct perf_evlist *evlist, struct target *target);
+int perf_evlist__apply_filters(struct perf_evlist *evlist, struct perf_evsel **err_evsel);
+
+void __perf_evlist__set_leader(struct list_head *list);
+void perf_evlist__set_leader(struct perf_evlist *evlist);
+
+u64 perf_evlist__read_format(struct perf_evlist *evlist);
+u64 __perf_evlist__combined_sample_type(struct perf_evlist *evlist);
+u64 perf_evlist__combined_sample_type(struct perf_evlist *evlist);
+bool perf_evlist__sample_id_all(struct perf_evlist *evlist);
+u16 perf_evlist__id_hdr_size(struct perf_evlist *evlist);
+
+int perf_evlist__parse_sample(struct perf_evlist *evlist, union perf_event *event,
+			      struct perf_sample *sample);
+
+bool perf_evlist__valid_sample_type(struct perf_evlist *evlist);
+bool perf_evlist__valid_sample_id_all(struct perf_evlist *evlist);
+bool perf_evlist__valid_read_format(struct perf_evlist *evlist);
+
+void perf_evlist__splice_list_tail(struct perf_evlist *evlist,
+				   struct list_head *list,
+				   int nr_entries);
+
+static inline struct perf_evsel *perf_evlist__first(struct perf_evlist *evlist)
+{
+	return list_entry(evlist->entries.next, struct perf_evsel, node);
+}
+
+static inline struct perf_evsel *perf_evlist__last(struct perf_evlist *evlist)
+{
+	return list_entry(evlist->entries.prev, struct perf_evsel, node);
+}
+
+size_t perf_evlist__fprintf(struct perf_evlist *evlist, FILE *fp);
+
+int perf_evlist__strerror_open(struct perf_evlist *evlist, int err, char *buf, size_t size);
+int perf_evlist__strerror_mmap(struct perf_evlist *evlist, int err, char *buf, size_t size);
+
+static inline u64 perf_mmap__read_head(struct perf_mmap *mm)
+{
+	struct perf_event_mmap_page *pc = mm->base;
+	u64 head = ACCESS_ONCE(pc->data_head);
+	rmb();
+	return head;
+}
+
+static inline void perf_mmap__write_tail(struct perf_mmap *md, u64 tail)
+{
+	struct perf_event_mmap_page *pc = md->base;
+
+	/*
+	 * ensure all reads are done before we write the tail out.
+	 */
+	mb();
+	pc->data_tail = tail;
+}
+
+bool perf_evlist__can_select_event(struct perf_evlist *evlist, const char *str);
+void perf_evlist__to_front(struct perf_evlist *evlist,
+			   struct perf_evsel *move_evsel);
+
+/**
+ * __evlist__for_each - iterate thru all the evsels
+ * @list: list_head instance to iterate
+ * @evsel: struct evsel iterator
+ */
+#define __evlist__for_each(list, evsel) \
+        list_for_each_entry(evsel, list, node)
+
+/**
+ * evlist__for_each - iterate thru all the evsels
+ * @evlist: evlist instance to iterate
+ * @evsel: struct evsel iterator
+ */
+#define evlist__for_each(evlist, evsel) \
+	__evlist__for_each(&(evlist)->entries, evsel)
+
+/**
+ * __evlist__for_each_continue - continue iteration thru all the evsels
+ * @list: list_head instance to iterate
+ * @evsel: struct evsel iterator
+ */
+#define __evlist__for_each_continue(list, evsel) \
+        list_for_each_entry_continue(evsel, list, node)
+
+/**
+ * evlist__for_each_continue - continue iteration thru all the evsels
+ * @evlist: evlist instance to iterate
+ * @evsel: struct evsel iterator
+ */
+#define evlist__for_each_continue(evlist, evsel) \
+	__evlist__for_each_continue(&(evlist)->entries, evsel)
+
+/**
+ * __evlist__for_each_reverse - iterate thru all the evsels in reverse order
+ * @list: list_head instance to iterate
+ * @evsel: struct evsel iterator
+ */
+#define __evlist__for_each_reverse(list, evsel) \
+        list_for_each_entry_reverse(evsel, list, node)
+
+/**
+ * evlist__for_each_reverse - iterate thru all the evsels in reverse order
+ * @evlist: evlist instance to iterate
+ * @evsel: struct evsel iterator
+ */
+#define evlist__for_each_reverse(evlist, evsel) \
+	__evlist__for_each_reverse(&(evlist)->entries, evsel)
+
+/**
+ * __evlist__for_each_safe - safely iterate thru all the evsels
+ * @list: list_head instance to iterate
+ * @tmp: struct evsel temp iterator
+ * @evsel: struct evsel iterator
+ */
+#define __evlist__for_each_safe(list, tmp, evsel) \
+        list_for_each_entry_safe(evsel, tmp, list, node)
+
+/**
+ * evlist__for_each_safe - safely iterate thru all the evsels
+ * @evlist: evlist instance to iterate
+ * @evsel: struct evsel iterator
+ * @tmp: struct evsel temp iterator
+ */
+#define evlist__for_each_safe(evlist, tmp, evsel) \
+	__evlist__for_each_safe(&(evlist)->entries, tmp, evsel)
+
+void perf_evlist__set_tracking_event(struct perf_evlist *evlist,
+				     struct perf_evsel *tracking_evsel);
+
+#endif /* __PERF_EVLIST_H */
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
new file mode 100644
index 000000000..33e3fd8c2
--- /dev/null
+++ b/tools/perf/util/evsel.c
@@ -0,0 +1,2190 @@
+/*
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Parts came from builtin-{top,stat,record}.c, see those files for further
+ * copyright notes.
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+
+#include <byteswap.h>
+#include <linux/bitops.h>
+#include <api/fs/debugfs.h>
+#include <traceevent/event-parse.h>
+#include <linux/hw_breakpoint.h>
+#include <linux/perf_event.h>
+#include <sys/resource.h>
+#include "asm/bug.h"
+#include "callchain.h"
+#include "cgroup.h"
+#include "evsel.h"
+#include "evlist.h"
+#include "util.h"
+#include "cpumap.h"
+#include "thread_map.h"
+#include "target.h"
+#include "perf_regs.h"
+#include "debug.h"
+#include "trace-event.h"
+
+static struct {
+	bool sample_id_all;
+	bool exclude_guest;
+	bool mmap2;
+	bool cloexec;
+	bool clockid;
+	bool clockid_wrong;
+} perf_missing_features;
+
+static clockid_t clockid;
+
+static int perf_evsel__no_extra_init(struct perf_evsel *evsel __maybe_unused)
+{
+	return 0;
+}
+
+static void perf_evsel__no_extra_fini(struct perf_evsel *evsel __maybe_unused)
+{
+}
+
+static struct {
+	size_t	size;
+	int	(*init)(struct perf_evsel *evsel);
+	void	(*fini)(struct perf_evsel *evsel);
+} perf_evsel__object = {
+	.size = sizeof(struct perf_evsel),
+	.init = perf_evsel__no_extra_init,
+	.fini = perf_evsel__no_extra_fini,
+};
+
+int perf_evsel__object_config(size_t object_size,
+			      int (*init)(struct perf_evsel *evsel),
+			      void (*fini)(struct perf_evsel *evsel))
+{
+
+	if (object_size == 0)
+		goto set_methods;
+
+	if (perf_evsel__object.size > object_size)
+		return -EINVAL;
+
+	perf_evsel__object.size = object_size;
+
+set_methods:
+	if (init != NULL)
+		perf_evsel__object.init = init;
+
+	if (fini != NULL)
+		perf_evsel__object.fini = fini;
+
+	return 0;
+}
+
+#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
+
+int __perf_evsel__sample_size(u64 sample_type)
+{
+	u64 mask = sample_type & PERF_SAMPLE_MASK;
+	int size = 0;
+	int i;
+
+	for (i = 0; i < 64; i++) {
+		if (mask & (1ULL << i))
+			size++;
+	}
+
+	size *= sizeof(u64);
+
+	return size;
+}
+
+/**
+ * __perf_evsel__calc_id_pos - calculate id_pos.
+ * @sample_type: sample type
+ *
+ * This function returns the position of the event id (PERF_SAMPLE_ID or
+ * PERF_SAMPLE_IDENTIFIER) in a sample event i.e. in the array of struct
+ * sample_event.
+ */
+static int __perf_evsel__calc_id_pos(u64 sample_type)
+{
+	int idx = 0;
+
+	if (sample_type & PERF_SAMPLE_IDENTIFIER)
+		return 0;
+
+	if (!(sample_type & PERF_SAMPLE_ID))
+		return -1;
+
+	if (sample_type & PERF_SAMPLE_IP)
+		idx += 1;
+
+	if (sample_type & PERF_SAMPLE_TID)
+		idx += 1;
+
+	if (sample_type & PERF_SAMPLE_TIME)
+		idx += 1;
+
+	if (sample_type & PERF_SAMPLE_ADDR)
+		idx += 1;
+
+	return idx;
+}
+
+/**
+ * __perf_evsel__calc_is_pos - calculate is_pos.
+ * @sample_type: sample type
+ *
+ * This function returns the position (counting backwards) of the event id
+ * (PERF_SAMPLE_ID or PERF_SAMPLE_IDENTIFIER) in a non-sample event i.e. if
+ * sample_id_all is used there is an id sample appended to non-sample events.
+ */
+static int __perf_evsel__calc_is_pos(u64 sample_type)
+{
+	int idx = 1;
+
+	if (sample_type & PERF_SAMPLE_IDENTIFIER)
+		return 1;
+
+	if (!(sample_type & PERF_SAMPLE_ID))
+		return -1;
+
+	if (sample_type & PERF_SAMPLE_CPU)
+		idx += 1;
+
+	if (sample_type & PERF_SAMPLE_STREAM_ID)
+		idx += 1;
+
+	return idx;
+}
+
+void perf_evsel__calc_id_pos(struct perf_evsel *evsel)
+{
+	evsel->id_pos = __perf_evsel__calc_id_pos(evsel->attr.sample_type);
+	evsel->is_pos = __perf_evsel__calc_is_pos(evsel->attr.sample_type);
+}
+
+void __perf_evsel__set_sample_bit(struct perf_evsel *evsel,
+				  enum perf_event_sample_format bit)
+{
+	if (!(evsel->attr.sample_type & bit)) {
+		evsel->attr.sample_type |= bit;
+		evsel->sample_size += sizeof(u64);
+		perf_evsel__calc_id_pos(evsel);
+	}
+}
+
+void __perf_evsel__reset_sample_bit(struct perf_evsel *evsel,
+				    enum perf_event_sample_format bit)
+{
+	if (evsel->attr.sample_type & bit) {
+		evsel->attr.sample_type &= ~bit;
+		evsel->sample_size -= sizeof(u64);
+		perf_evsel__calc_id_pos(evsel);
+	}
+}
+
+void perf_evsel__set_sample_id(struct perf_evsel *evsel,
+			       bool can_sample_identifier)
+{
+	if (can_sample_identifier) {
+		perf_evsel__reset_sample_bit(evsel, ID);
+		perf_evsel__set_sample_bit(evsel, IDENTIFIER);
+	} else {
+		perf_evsel__set_sample_bit(evsel, ID);
+	}
+	evsel->attr.read_format |= PERF_FORMAT_ID;
+}
+
+void perf_evsel__init(struct perf_evsel *evsel,
+		      struct perf_event_attr *attr, int idx)
+{
+	evsel->idx	   = idx;
+	evsel->tracking	   = !idx;
+	evsel->attr	   = *attr;
+	evsel->leader	   = evsel;
+	evsel->unit	   = "";
+	evsel->scale	   = 1.0;
+	INIT_LIST_HEAD(&evsel->node);
+	perf_evsel__object.init(evsel);
+	evsel->sample_size = __perf_evsel__sample_size(attr->sample_type);
+	perf_evsel__calc_id_pos(evsel);
+}
+
+struct perf_evsel *perf_evsel__new_idx(struct perf_event_attr *attr, int idx)
+{
+	struct perf_evsel *evsel = zalloc(perf_evsel__object.size);
+
+	if (evsel != NULL)
+		perf_evsel__init(evsel, attr, idx);
+
+	return evsel;
+}
+
+struct perf_evsel *perf_evsel__newtp_idx(const char *sys, const char *name, int idx)
+{
+	struct perf_evsel *evsel = zalloc(perf_evsel__object.size);
+
+	if (evsel != NULL) {
+		struct perf_event_attr attr = {
+			.type	       = PERF_TYPE_TRACEPOINT,
+			.sample_type   = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME |
+					  PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD),
+		};
+
+		if (asprintf(&evsel->name, "%s:%s", sys, name) < 0)
+			goto out_free;
+
+		evsel->tp_format = trace_event__tp_format(sys, name);
+		if (evsel->tp_format == NULL)
+			goto out_free;
+
+		event_attr_init(&attr);
+		attr.config = evsel->tp_format->id;
+		attr.sample_period = 1;
+		perf_evsel__init(evsel, &attr, idx);
+	}
+
+	return evsel;
+
+out_free:
+	zfree(&evsel->name);
+	free(evsel);
+	return NULL;
+}
+
+const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX] = {
+	"cycles",
+	"instructions",
+	"cache-references",
+	"cache-misses",
+	"branches",
+	"branch-misses",
+	"bus-cycles",
+	"stalled-cycles-frontend",
+	"stalled-cycles-backend",
+	"ref-cycles",
+};
+
+static const char *__perf_evsel__hw_name(u64 config)
+{
+	if (config < PERF_COUNT_HW_MAX && perf_evsel__hw_names[config])
+		return perf_evsel__hw_names[config];
+
+	return "unknown-hardware";
+}
+
+static int perf_evsel__add_modifiers(struct perf_evsel *evsel, char *bf, size_t size)
+{
+	int colon = 0, r = 0;
+	struct perf_event_attr *attr = &evsel->attr;
+	bool exclude_guest_default = false;
+
+#define MOD_PRINT(context, mod)	do {					\
+		if (!attr->exclude_##context) {				\
+			if (!colon) colon = ++r;			\
+			r += scnprintf(bf + r, size - r, "%c", mod);	\
+		} } while(0)
+
+	if (attr->exclude_kernel || attr->exclude_user || attr->exclude_hv) {
+		MOD_PRINT(kernel, 'k');
+		MOD_PRINT(user, 'u');
+		MOD_PRINT(hv, 'h');
+		exclude_guest_default = true;
+	}
+
+	if (attr->precise_ip) {
+		if (!colon)
+			colon = ++r;
+		r += scnprintf(bf + r, size - r, "%.*s", attr->precise_ip, "ppp");
+		exclude_guest_default = true;
+	}
+
+	if (attr->exclude_host || attr->exclude_guest == exclude_guest_default) {
+		MOD_PRINT(host, 'H');
+		MOD_PRINT(guest, 'G');
+	}
+#undef MOD_PRINT
+	if (colon)
+		bf[colon - 1] = ':';
+	return r;
+}
+
+static int perf_evsel__hw_name(struct perf_evsel *evsel, char *bf, size_t size)
+{
+	int r = scnprintf(bf, size, "%s", __perf_evsel__hw_name(evsel->attr.config));
+	return r + perf_evsel__add_modifiers(evsel, bf + r, size - r);
+}
+
+const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX] = {
+	"cpu-clock",
+	"task-clock",
+	"page-faults",
+	"context-switches",
+	"cpu-migrations",
+	"minor-faults",
+	"major-faults",
+	"alignment-faults",
+	"emulation-faults",
+	"dummy",
+};
+
+static const char *__perf_evsel__sw_name(u64 config)
+{
+	if (config < PERF_COUNT_SW_MAX && perf_evsel__sw_names[config])
+		return perf_evsel__sw_names[config];
+	return "unknown-software";
+}
+
+static int perf_evsel__sw_name(struct perf_evsel *evsel, char *bf, size_t size)
+{
+	int r = scnprintf(bf, size, "%s", __perf_evsel__sw_name(evsel->attr.config));
+	return r + perf_evsel__add_modifiers(evsel, bf + r, size - r);
+}
+
+static int __perf_evsel__bp_name(char *bf, size_t size, u64 addr, u64 type)
+{
+	int r;
+
+	r = scnprintf(bf, size, "mem:0x%" PRIx64 ":", addr);
+
+	if (type & HW_BREAKPOINT_R)
+		r += scnprintf(bf + r, size - r, "r");
+
+	if (type & HW_BREAKPOINT_W)
+		r += scnprintf(bf + r, size - r, "w");
+
+	if (type & HW_BREAKPOINT_X)
+		r += scnprintf(bf + r, size - r, "x");
+
+	return r;
+}
+
+static int perf_evsel__bp_name(struct perf_evsel *evsel, char *bf, size_t size)
+{
+	struct perf_event_attr *attr = &evsel->attr;
+	int r = __perf_evsel__bp_name(bf, size, attr->bp_addr, attr->bp_type);
+	return r + perf_evsel__add_modifiers(evsel, bf + r, size - r);
+}
+
+const char *perf_evsel__hw_cache[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_EVSEL__MAX_ALIASES] = {
+ { "L1-dcache",	"l1-d",		"l1d",		"L1-data",		},
+ { "L1-icache",	"l1-i",		"l1i",		"L1-instruction",	},
+ { "LLC",	"L2",							},
+ { "dTLB",	"d-tlb",	"Data-TLB",				},
+ { "iTLB",	"i-tlb",	"Instruction-TLB",			},
+ { "branch",	"branches",	"bpu",		"btb",		"bpc",	},
+ { "node",								},
+};
+
+const char *perf_evsel__hw_cache_op[PERF_COUNT_HW_CACHE_OP_MAX]
+				   [PERF_EVSEL__MAX_ALIASES] = {
+ { "load",	"loads",	"read",					},
+ { "store",	"stores",	"write",				},
+ { "prefetch",	"prefetches",	"speculative-read", "speculative-load",	},
+};
+
+const char *perf_evsel__hw_cache_result[PERF_COUNT_HW_CACHE_RESULT_MAX]
+				       [PERF_EVSEL__MAX_ALIASES] = {
+ { "refs",	"Reference",	"ops",		"access",		},
+ { "misses",	"miss",							},
+};
+
+#define C(x)		PERF_COUNT_HW_CACHE_##x
+#define CACHE_READ	(1 << C(OP_READ))
+#define CACHE_WRITE	(1 << C(OP_WRITE))
+#define CACHE_PREFETCH	(1 << C(OP_PREFETCH))
+#define COP(x)		(1 << x)
+
+/*
+ * cache operartion stat
+ * L1I : Read and prefetch only
+ * ITLB and BPU : Read-only
+ */
+static unsigned long perf_evsel__hw_cache_stat[C(MAX)] = {
+ [C(L1D)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(L1I)]	= (CACHE_READ | CACHE_PREFETCH),
+ [C(LL)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(DTLB)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(ITLB)]	= (CACHE_READ),
+ [C(BPU)]	= (CACHE_READ),
+ [C(NODE)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+};
+
+bool perf_evsel__is_cache_op_valid(u8 type, u8 op)
+{
+	if (perf_evsel__hw_cache_stat[type] & COP(op))
+		return true;	/* valid */
+	else
+		return false;	/* invalid */
+}
+
+int __perf_evsel__hw_cache_type_op_res_name(u8 type, u8 op, u8 result,
+					    char *bf, size_t size)
+{
+	if (result) {
+		return scnprintf(bf, size, "%s-%s-%s", perf_evsel__hw_cache[type][0],
+				 perf_evsel__hw_cache_op[op][0],
+				 perf_evsel__hw_cache_result[result][0]);
+	}
+
+	return scnprintf(bf, size, "%s-%s", perf_evsel__hw_cache[type][0],
+			 perf_evsel__hw_cache_op[op][1]);
+}
+
+static int __perf_evsel__hw_cache_name(u64 config, char *bf, size_t size)
+{
+	u8 op, result, type = (config >>  0) & 0xff;
+	const char *err = "unknown-ext-hardware-cache-type";
+
+	if (type > PERF_COUNT_HW_CACHE_MAX)
+		goto out_err;
+
+	op = (config >>  8) & 0xff;
+	err = "unknown-ext-hardware-cache-op";
+	if (op > PERF_COUNT_HW_CACHE_OP_MAX)
+		goto out_err;
+
+	result = (config >> 16) & 0xff;
+	err = "unknown-ext-hardware-cache-result";
+	if (result > PERF_COUNT_HW_CACHE_RESULT_MAX)
+		goto out_err;
+
+	err = "invalid-cache";
+	if (!perf_evsel__is_cache_op_valid(type, op))
+		goto out_err;
+
+	return __perf_evsel__hw_cache_type_op_res_name(type, op, result, bf, size);
+out_err:
+	return scnprintf(bf, size, "%s", err);
+}
+
+static int perf_evsel__hw_cache_name(struct perf_evsel *evsel, char *bf, size_t size)
+{
+	int ret = __perf_evsel__hw_cache_name(evsel->attr.config, bf, size);
+	return ret + perf_evsel__add_modifiers(evsel, bf + ret, size - ret);
+}
+
+static int perf_evsel__raw_name(struct perf_evsel *evsel, char *bf, size_t size)
+{
+	int ret = scnprintf(bf, size, "raw 0x%" PRIx64, evsel->attr.config);
+	return ret + perf_evsel__add_modifiers(evsel, bf + ret, size - ret);
+}
+
+const char *perf_evsel__name(struct perf_evsel *evsel)
+{
+	char bf[128];
+
+	if (evsel->name)
+		return evsel->name;
+
+	switch (evsel->attr.type) {
+	case PERF_TYPE_RAW:
+		perf_evsel__raw_name(evsel, bf, sizeof(bf));
+		break;
+
+	case PERF_TYPE_HARDWARE:
+		perf_evsel__hw_name(evsel, bf, sizeof(bf));
+		break;
+
+	case PERF_TYPE_HW_CACHE:
+		perf_evsel__hw_cache_name(evsel, bf, sizeof(bf));
+		break;
+
+	case PERF_TYPE_SOFTWARE:
+		perf_evsel__sw_name(evsel, bf, sizeof(bf));
+		break;
+
+	case PERF_TYPE_TRACEPOINT:
+		scnprintf(bf, sizeof(bf), "%s", "unknown tracepoint");
+		break;
+
+	case PERF_TYPE_BREAKPOINT:
+		perf_evsel__bp_name(evsel, bf, sizeof(bf));
+		break;
+
+	default:
+		scnprintf(bf, sizeof(bf), "unknown attr type: %d",
+			  evsel->attr.type);
+		break;
+	}
+
+	evsel->name = strdup(bf);
+
+	return evsel->name ?: "unknown";
+}
+
+const char *perf_evsel__group_name(struct perf_evsel *evsel)
+{
+	return evsel->group_name ?: "anon group";
+}
+
+int perf_evsel__group_desc(struct perf_evsel *evsel, char *buf, size_t size)
+{
+	int ret;
+	struct perf_evsel *pos;
+	const char *group_name = perf_evsel__group_name(evsel);
+
+	ret = scnprintf(buf, size, "%s", group_name);
+
+	ret += scnprintf(buf + ret, size - ret, " { %s",
+			 perf_evsel__name(evsel));
+
+	for_each_group_member(pos, evsel)
+		ret += scnprintf(buf + ret, size - ret, ", %s",
+				 perf_evsel__name(pos));
+
+	ret += scnprintf(buf + ret, size - ret, " }");
+
+	return ret;
+}
+
+static void
+perf_evsel__config_callgraph(struct perf_evsel *evsel,
+			     struct record_opts *opts)
+{
+	bool function = perf_evsel__is_function_event(evsel);
+	struct perf_event_attr *attr = &evsel->attr;
+
+	perf_evsel__set_sample_bit(evsel, CALLCHAIN);
+
+	if (callchain_param.record_mode == CALLCHAIN_LBR) {
+		if (!opts->branch_stack) {
+			if (attr->exclude_user) {
+				pr_warning("LBR callstack option is only available "
+					   "to get user callchain information. "
+					   "Falling back to framepointers.\n");
+			} else {
+				perf_evsel__set_sample_bit(evsel, BRANCH_STACK);
+				attr->branch_sample_type = PERF_SAMPLE_BRANCH_USER |
+							PERF_SAMPLE_BRANCH_CALL_STACK;
+			}
+		} else
+			 pr_warning("Cannot use LBR callstack with branch stack. "
+				    "Falling back to framepointers.\n");
+	}
+
+	if (callchain_param.record_mode == CALLCHAIN_DWARF) {
+		if (!function) {
+			perf_evsel__set_sample_bit(evsel, REGS_USER);
+			perf_evsel__set_sample_bit(evsel, STACK_USER);
+			attr->sample_regs_user = PERF_REGS_MASK;
+			attr->sample_stack_user = callchain_param.dump_size;
+			attr->exclude_callchain_user = 1;
+		} else {
+			pr_info("Cannot use DWARF unwind for function trace event,"
+				" falling back to framepointers.\n");
+		}
+	}
+
+	if (function) {
+		pr_info("Disabling user space callchains for function trace event.\n");
+		attr->exclude_callchain_user = 1;
+	}
+}
+
+/*
+ * The enable_on_exec/disabled value strategy:
+ *
+ *  1) For any type of traced program:
+ *    - all independent events and group leaders are disabled
+ *    - all group members are enabled
+ *
+ *     Group members are ruled by group leaders. They need to
+ *     be enabled, because the group scheduling relies on that.
+ *
+ *  2) For traced programs executed by perf:
+ *     - all independent events and group leaders have
+ *       enable_on_exec set
+ *     - we don't specifically enable or disable any event during
+ *       the record command
+ *
+ *     Independent events and group leaders are initially disabled
+ *     and get enabled by exec. Group members are ruled by group
+ *     leaders as stated in 1).
+ *
+ *  3) For traced programs attached by perf (pid/tid):
+ *     - we specifically enable or disable all events during
+ *       the record command
+ *
+ *     When attaching events to already running traced we
+ *     enable/disable events specifically, as there's no
+ *     initial traced exec call.
+ */
+void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)
+{
+	struct perf_evsel *leader = evsel->leader;
+	struct perf_event_attr *attr = &evsel->attr;
+	int track = evsel->tracking;
+	bool per_cpu = opts->target.default_per_cpu && !opts->target.per_thread;
+
+	attr->sample_id_all = perf_missing_features.sample_id_all ? 0 : 1;
+	attr->inherit	    = !opts->no_inherit;
+
+	perf_evsel__set_sample_bit(evsel, IP);
+	perf_evsel__set_sample_bit(evsel, TID);
+
+	if (evsel->sample_read) {
+		perf_evsel__set_sample_bit(evsel, READ);
+
+		/*
+		 * We need ID even in case of single event, because
+		 * PERF_SAMPLE_READ process ID specific data.
+		 */
+		perf_evsel__set_sample_id(evsel, false);
+
+		/*
+		 * Apply group format only if we belong to group
+		 * with more than one members.
+		 */
+		if (leader->nr_members > 1) {
+			attr->read_format |= PERF_FORMAT_GROUP;
+			attr->inherit = 0;
+		}
+	}
+
+	/*
+	 * We default some events to have a default interval. But keep
+	 * it a weak assumption overridable by the user.
+	 */
+	if (!attr->sample_period || (opts->user_freq != UINT_MAX ||
+				     opts->user_interval != ULLONG_MAX)) {
+		if (opts->freq) {
+			perf_evsel__set_sample_bit(evsel, PERIOD);
+			attr->freq		= 1;
+			attr->sample_freq	= opts->freq;
+		} else {
+			attr->sample_period = opts->default_interval;
+		}
+	}
+
+	/*
+	 * Disable sampling for all group members other
+	 * than leader in case leader 'leads' the sampling.
+	 */
+	if ((leader != evsel) && leader->sample_read) {
+		attr->sample_freq   = 0;
+		attr->sample_period = 0;
+	}
+
+	if (opts->no_samples)
+		attr->sample_freq = 0;
+
+	if (opts->inherit_stat)
+		attr->inherit_stat = 1;
+
+	if (opts->sample_address) {
+		perf_evsel__set_sample_bit(evsel, ADDR);
+		attr->mmap_data = track;
+	}
+
+	/*
+	 * We don't allow user space callchains for  function trace
+	 * event, due to issues with page faults while tracing page
+	 * fault handler and its overall trickiness nature.
+	 */
+	if (perf_evsel__is_function_event(evsel))
+		evsel->attr.exclude_callchain_user = 1;
+
+	if (callchain_param.enabled && !evsel->no_aux_samples)
+		perf_evsel__config_callgraph(evsel, opts);
+
+	if (opts->sample_intr_regs) {
+		attr->sample_regs_intr = PERF_REGS_MASK;
+		perf_evsel__set_sample_bit(evsel, REGS_INTR);
+	}
+
+	if (target__has_cpu(&opts->target))
+		perf_evsel__set_sample_bit(evsel, CPU);
+
+	if (opts->period)
+		perf_evsel__set_sample_bit(evsel, PERIOD);
+
+	/*
+	 * When the user explicitely disabled time don't force it here.
+	 */
+	if (opts->sample_time &&
+	    (!perf_missing_features.sample_id_all &&
+	    (!opts->no_inherit || target__has_cpu(&opts->target) || per_cpu)))
+		perf_evsel__set_sample_bit(evsel, TIME);
+
+	if (opts->raw_samples && !evsel->no_aux_samples) {
+		perf_evsel__set_sample_bit(evsel, TIME);
+		perf_evsel__set_sample_bit(evsel, RAW);
+		perf_evsel__set_sample_bit(evsel, CPU);
+	}
+
+	if (opts->sample_address)
+		perf_evsel__set_sample_bit(evsel, DATA_SRC);
+
+	if (opts->no_buffering) {
+		attr->watermark = 0;
+		attr->wakeup_events = 1;
+	}
+	if (opts->branch_stack && !evsel->no_aux_samples) {
+		perf_evsel__set_sample_bit(evsel, BRANCH_STACK);
+		attr->branch_sample_type = opts->branch_stack;
+	}
+
+	if (opts->sample_weight)
+		perf_evsel__set_sample_bit(evsel, WEIGHT);
+
+	attr->task  = track;
+	attr->mmap  = track;
+	attr->mmap2 = track && !perf_missing_features.mmap2;
+	attr->comm  = track;
+
+	if (opts->sample_transaction)
+		perf_evsel__set_sample_bit(evsel, TRANSACTION);
+
+	if (opts->running_time) {
+		evsel->attr.read_format |=
+			PERF_FORMAT_TOTAL_TIME_ENABLED |
+			PERF_FORMAT_TOTAL_TIME_RUNNING;
+	}
+
+	/*
+	 * XXX see the function comment above
+	 *
+	 * Disabling only independent events or group leaders,
+	 * keeping group members enabled.
+	 */
+	if (perf_evsel__is_group_leader(evsel))
+		attr->disabled = 1;
+
+	/*
+	 * Setting enable_on_exec for independent events and
+	 * group leaders for traced executed by perf.
+	 */
+	if (target__none(&opts->target) && perf_evsel__is_group_leader(evsel) &&
+		!opts->initial_delay)
+		attr->enable_on_exec = 1;
+
+	if (evsel->immediate) {
+		attr->disabled = 0;
+		attr->enable_on_exec = 0;
+	}
+
+	clockid = opts->clockid;
+	if (opts->use_clockid) {
+		attr->use_clockid = 1;
+		attr->clockid = opts->clockid;
+	}
+}
+
+static int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+	int cpu, thread;
+
+	if (evsel->system_wide)
+		nthreads = 1;
+
+	evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
+
+	if (evsel->fd) {
+		for (cpu = 0; cpu < ncpus; cpu++) {
+			for (thread = 0; thread < nthreads; thread++) {
+				FD(evsel, cpu, thread) = -1;
+			}
+		}
+	}
+
+	return evsel->fd != NULL ? 0 : -ENOMEM;
+}
+
+static int perf_evsel__run_ioctl(struct perf_evsel *evsel, int ncpus, int nthreads,
+			  int ioc,  void *arg)
+{
+	int cpu, thread;
+
+	if (evsel->system_wide)
+		nthreads = 1;
+
+	for (cpu = 0; cpu < ncpus; cpu++) {
+		for (thread = 0; thread < nthreads; thread++) {
+			int fd = FD(evsel, cpu, thread),
+			    err = ioctl(fd, ioc, arg);
+
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
+
+int perf_evsel__set_filter(struct perf_evsel *evsel, int ncpus, int nthreads,
+			   const char *filter)
+{
+	return perf_evsel__run_ioctl(evsel, ncpus, nthreads,
+				     PERF_EVENT_IOC_SET_FILTER,
+				     (void *)filter);
+}
+
+int perf_evsel__enable(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+	return perf_evsel__run_ioctl(evsel, ncpus, nthreads,
+				     PERF_EVENT_IOC_ENABLE,
+				     0);
+}
+
+int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+	if (ncpus == 0 || nthreads == 0)
+		return 0;
+
+	if (evsel->system_wide)
+		nthreads = 1;
+
+	evsel->sample_id = xyarray__new(ncpus, nthreads, sizeof(struct perf_sample_id));
+	if (evsel->sample_id == NULL)
+		return -ENOMEM;
+
+	evsel->id = zalloc(ncpus * nthreads * sizeof(u64));
+	if (evsel->id == NULL) {
+		xyarray__delete(evsel->sample_id);
+		evsel->sample_id = NULL;
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+void perf_evsel__reset_counts(struct perf_evsel *evsel, int ncpus)
+{
+	memset(evsel->counts, 0, (sizeof(*evsel->counts) +
+				 (ncpus * sizeof(struct perf_counts_values))));
+}
+
+int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus)
+{
+	evsel->counts = zalloc((sizeof(*evsel->counts) +
+				(ncpus * sizeof(struct perf_counts_values))));
+	return evsel->counts != NULL ? 0 : -ENOMEM;
+}
+
+static void perf_evsel__free_fd(struct perf_evsel *evsel)
+{
+	xyarray__delete(evsel->fd);
+	evsel->fd = NULL;
+}
+
+static void perf_evsel__free_id(struct perf_evsel *evsel)
+{
+	xyarray__delete(evsel->sample_id);
+	evsel->sample_id = NULL;
+	zfree(&evsel->id);
+}
+
+void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+	int cpu, thread;
+
+	if (evsel->system_wide)
+		nthreads = 1;
+
+	for (cpu = 0; cpu < ncpus; cpu++)
+		for (thread = 0; thread < nthreads; ++thread) {
+			close(FD(evsel, cpu, thread));
+			FD(evsel, cpu, thread) = -1;
+		}
+}
+
+void perf_evsel__free_counts(struct perf_evsel *evsel)
+{
+	zfree(&evsel->counts);
+}
+
+void perf_evsel__exit(struct perf_evsel *evsel)
+{
+	assert(list_empty(&evsel->node));
+	perf_evsel__free_fd(evsel);
+	perf_evsel__free_id(evsel);
+	close_cgroup(evsel->cgrp);
+	zfree(&evsel->group_name);
+	zfree(&evsel->name);
+	perf_evsel__object.fini(evsel);
+}
+
+void perf_evsel__delete(struct perf_evsel *evsel)
+{
+	perf_evsel__exit(evsel);
+	free(evsel);
+}
+
+void perf_evsel__compute_deltas(struct perf_evsel *evsel, int cpu,
+				struct perf_counts_values *count)
+{
+	struct perf_counts_values tmp;
+
+	if (!evsel->prev_raw_counts)
+		return;
+
+	if (cpu == -1) {
+		tmp = evsel->prev_raw_counts->aggr;
+		evsel->prev_raw_counts->aggr = *count;
+	} else {
+		tmp = evsel->prev_raw_counts->cpu[cpu];
+		evsel->prev_raw_counts->cpu[cpu] = *count;
+	}
+
+	count->val = count->val - tmp.val;
+	count->ena = count->ena - tmp.ena;
+	count->run = count->run - tmp.run;
+}
+
+void perf_counts_values__scale(struct perf_counts_values *count,
+			       bool scale, s8 *pscaled)
+{
+	s8 scaled = 0;
+
+	if (scale) {
+		if (count->run == 0) {
+			scaled = -1;
+			count->val = 0;
+		} else if (count->run < count->ena) {
+			scaled = 1;
+			count->val = (u64)((double) count->val * count->ena / count->run + 0.5);
+		}
+	} else
+		count->ena = count->run = 0;
+
+	if (pscaled)
+		*pscaled = scaled;
+}
+
+int perf_evsel__read_cb(struct perf_evsel *evsel, int cpu, int thread,
+			perf_evsel__read_cb_t cb)
+{
+	struct perf_counts_values count;
+
+	memset(&count, 0, sizeof(count));
+
+	if (FD(evsel, cpu, thread) < 0)
+		return -EINVAL;
+
+	if (readn(FD(evsel, cpu, thread), &count, sizeof(count)) < 0)
+		return -errno;
+
+	return cb(evsel, cpu, thread, &count);
+}
+
+int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
+			      int cpu, int thread, bool scale)
+{
+	struct perf_counts_values count;
+	size_t nv = scale ? 3 : 1;
+
+	if (FD(evsel, cpu, thread) < 0)
+		return -EINVAL;
+
+	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0)
+		return -ENOMEM;
+
+	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0)
+		return -errno;
+
+	perf_evsel__compute_deltas(evsel, cpu, &count);
+	perf_counts_values__scale(&count, scale, NULL);
+	evsel->counts->cpu[cpu] = count;
+	return 0;
+}
+
+static int get_group_fd(struct perf_evsel *evsel, int cpu, int thread)
+{
+	struct perf_evsel *leader = evsel->leader;
+	int fd;
+
+	if (perf_evsel__is_group_leader(evsel))
+		return -1;
+
+	/*
+	 * Leader must be already processed/open,
+	 * if not it's a bug.
+	 */
+	BUG_ON(!leader->fd);
+
+	fd = FD(leader, cpu, thread);
+	BUG_ON(fd == -1);
+
+	return fd;
+}
+
+struct bit_names {
+	int bit;
+	const char *name;
+};
+
+static void __p_bits(char *buf, size_t size, u64 value, struct bit_names *bits)
+{
+	bool first_bit = true;
+	int i = 0;
+
+	do {
+		if (value & bits[i].bit) {
+			buf += scnprintf(buf, size, "%s%s", first_bit ? "" : "|", bits[i].name);
+			first_bit = false;
+		}
+	} while (bits[++i].name != NULL);
+}
+
+static void __p_sample_type(char *buf, size_t size, u64 value)
+{
+#define bit_name(n) { PERF_SAMPLE_##n, #n }
+	struct bit_names bits[] = {
+		bit_name(IP), bit_name(TID), bit_name(TIME), bit_name(ADDR),
+		bit_name(READ), bit_name(CALLCHAIN), bit_name(ID), bit_name(CPU),
+		bit_name(PERIOD), bit_name(STREAM_ID), bit_name(RAW),
+		bit_name(BRANCH_STACK), bit_name(REGS_USER), bit_name(STACK_USER),
+		bit_name(IDENTIFIER), bit_name(REGS_INTR),
+		{ .name = NULL, }
+	};
+#undef bit_name
+	__p_bits(buf, size, value, bits);
+}
+
+static void __p_read_format(char *buf, size_t size, u64 value)
+{
+#define bit_name(n) { PERF_FORMAT_##n, #n }
+	struct bit_names bits[] = {
+		bit_name(TOTAL_TIME_ENABLED), bit_name(TOTAL_TIME_RUNNING),
+		bit_name(ID), bit_name(GROUP),
+		{ .name = NULL, }
+	};
+#undef bit_name
+	__p_bits(buf, size, value, bits);
+}
+
+#define BUF_SIZE		1024
+
+#define p_hex(val)		snprintf(buf, BUF_SIZE, "%"PRIx64, (uint64_t)(val))
+#define p_unsigned(val)		snprintf(buf, BUF_SIZE, "%"PRIu64, (uint64_t)(val))
+#define p_signed(val)		snprintf(buf, BUF_SIZE, "%"PRId64, (int64_t)(val))
+#define p_sample_type(val)	__p_sample_type(buf, BUF_SIZE, val)
+#define p_read_format(val)	__p_read_format(buf, BUF_SIZE, val)
+
+#define PRINT_ATTRn(_n, _f, _p)				\
+do {							\
+	if (attr->_f) {					\
+		_p(attr->_f);				\
+		ret += attr__fprintf(fp, _n, buf, priv);\
+	}						\
+} while (0)
+
+#define PRINT_ATTRf(_f, _p)	PRINT_ATTRn(#_f, _f, _p)
+
+int perf_event_attr__fprintf(FILE *fp, struct perf_event_attr *attr,
+			     attr__fprintf_f attr__fprintf, void *priv)
+{
+	char buf[BUF_SIZE];
+	int ret = 0;
+
+	PRINT_ATTRf(type, p_unsigned);
+	PRINT_ATTRf(size, p_unsigned);
+	PRINT_ATTRf(config, p_hex);
+	PRINT_ATTRn("{ sample_period, sample_freq }", sample_period, p_unsigned);
+	PRINT_ATTRf(sample_type, p_sample_type);
+	PRINT_ATTRf(read_format, p_read_format);
+
+	PRINT_ATTRf(disabled, p_unsigned);
+	PRINT_ATTRf(inherit, p_unsigned);
+	PRINT_ATTRf(pinned, p_unsigned);
+	PRINT_ATTRf(exclusive, p_unsigned);
+	PRINT_ATTRf(exclude_user, p_unsigned);
+	PRINT_ATTRf(exclude_kernel, p_unsigned);
+	PRINT_ATTRf(exclude_hv, p_unsigned);
+	PRINT_ATTRf(exclude_idle, p_unsigned);
+	PRINT_ATTRf(mmap, p_unsigned);
+	PRINT_ATTRf(comm, p_unsigned);
+	PRINT_ATTRf(freq, p_unsigned);
+	PRINT_ATTRf(inherit_stat, p_unsigned);
+	PRINT_ATTRf(enable_on_exec, p_unsigned);
+	PRINT_ATTRf(task, p_unsigned);
+	PRINT_ATTRf(watermark, p_unsigned);
+	PRINT_ATTRf(precise_ip, p_unsigned);
+	PRINT_ATTRf(mmap_data, p_unsigned);
+	PRINT_ATTRf(sample_id_all, p_unsigned);
+	PRINT_ATTRf(exclude_host, p_unsigned);
+	PRINT_ATTRf(exclude_guest, p_unsigned);
+	PRINT_ATTRf(exclude_callchain_kernel, p_unsigned);
+	PRINT_ATTRf(exclude_callchain_user, p_unsigned);
+	PRINT_ATTRf(mmap2, p_unsigned);
+	PRINT_ATTRf(comm_exec, p_unsigned);
+	PRINT_ATTRf(use_clockid, p_unsigned);
+
+	PRINT_ATTRn("{ wakeup_events, wakeup_watermark }", wakeup_events, p_unsigned);
+	PRINT_ATTRf(bp_type, p_unsigned);
+	PRINT_ATTRn("{ bp_addr, config1 }", bp_addr, p_hex);
+	PRINT_ATTRn("{ bp_len, config2 }", bp_len, p_hex);
+	PRINT_ATTRf(sample_regs_user, p_hex);
+	PRINT_ATTRf(sample_stack_user, p_unsigned);
+	PRINT_ATTRf(clockid, p_signed);
+	PRINT_ATTRf(sample_regs_intr, p_hex);
+
+	return ret;
+}
+
+static int __open_attr__fprintf(FILE *fp, const char *name, const char *val,
+				void *priv __attribute__((unused)))
+{
+	return fprintf(fp, "  %-32s %s\n", name, val);
+}
+
+static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
+			      struct thread_map *threads)
+{
+	int cpu, thread, nthreads;
+	unsigned long flags = PERF_FLAG_FD_CLOEXEC;
+	int pid = -1, err;
+	enum { NO_CHANGE, SET_TO_MAX, INCREASED_MAX } set_rlimit = NO_CHANGE;
+
+	if (evsel->system_wide)
+		nthreads = 1;
+	else
+		nthreads = threads->nr;
+
+	if (evsel->fd == NULL &&
+	    perf_evsel__alloc_fd(evsel, cpus->nr, nthreads) < 0)
+		return -ENOMEM;
+
+	if (evsel->cgrp) {
+		flags |= PERF_FLAG_PID_CGROUP;
+		pid = evsel->cgrp->fd;
+	}
+
+fallback_missing_features:
+	if (perf_missing_features.clockid_wrong)
+		evsel->attr.clockid = CLOCK_MONOTONIC; /* should always work */
+	if (perf_missing_features.clockid) {
+		evsel->attr.use_clockid = 0;
+		evsel->attr.clockid = 0;
+	}
+	if (perf_missing_features.cloexec)
+		flags &= ~(unsigned long)PERF_FLAG_FD_CLOEXEC;
+	if (perf_missing_features.mmap2)
+		evsel->attr.mmap2 = 0;
+	if (perf_missing_features.exclude_guest)
+		evsel->attr.exclude_guest = evsel->attr.exclude_host = 0;
+retry_sample_id:
+	if (perf_missing_features.sample_id_all)
+		evsel->attr.sample_id_all = 0;
+
+	if (verbose >= 2) {
+		fprintf(stderr, "%.60s\n", graph_dotted_line);
+		fprintf(stderr, "perf_event_attr:\n");
+		perf_event_attr__fprintf(stderr, &evsel->attr, __open_attr__fprintf, NULL);
+		fprintf(stderr, "%.60s\n", graph_dotted_line);
+	}
+
+	for (cpu = 0; cpu < cpus->nr; cpu++) {
+
+		for (thread = 0; thread < nthreads; thread++) {
+			int group_fd;
+
+			if (!evsel->cgrp && !evsel->system_wide)
+				pid = threads->map[thread];
+
+			group_fd = get_group_fd(evsel, cpu, thread);
+retry_open:
+			pr_debug2("sys_perf_event_open: pid %d  cpu %d  group_fd %d  flags %#lx\n",
+				  pid, cpus->map[cpu], group_fd, flags);
+
+			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
+								     pid,
+								     cpus->map[cpu],
+								     group_fd, flags);
+			if (FD(evsel, cpu, thread) < 0) {
+				err = -errno;
+				pr_debug2("sys_perf_event_open failed, error %d\n",
+					  err);
+				goto try_fallback;
+			}
+			set_rlimit = NO_CHANGE;
+
+			/*
+			 * If we succeeded but had to kill clockid, fail and
+			 * have perf_evsel__open_strerror() print us a nice
+			 * error.
+			 */
+			if (perf_missing_features.clockid ||
+			    perf_missing_features.clockid_wrong) {
+				err = -EINVAL;
+				goto out_close;
+			}
+		}
+	}
+
+	return 0;
+
+try_fallback:
+	/*
+	 * perf stat needs between 5 and 22 fds per CPU. When we run out
+	 * of them try to increase the limits.
+	 */
+	if (err == -EMFILE && set_rlimit < INCREASED_MAX) {
+		struct rlimit l;
+		int old_errno = errno;
+
+		if (getrlimit(RLIMIT_NOFILE, &l) == 0) {
+			if (set_rlimit == NO_CHANGE)
+				l.rlim_cur = l.rlim_max;
+			else {
+				l.rlim_cur = l.rlim_max + 1000;
+				l.rlim_max = l.rlim_cur;
+			}
+			if (setrlimit(RLIMIT_NOFILE, &l) == 0) {
+				set_rlimit++;
+				errno = old_errno;
+				goto retry_open;
+			}
+		}
+		errno = old_errno;
+	}
+
+	if (err != -EINVAL || cpu > 0 || thread > 0)
+		goto out_close;
+
+	/*
+	 * Must probe features in the order they were added to the
+	 * perf_event_attr interface.
+	 */
+	if (!perf_missing_features.clockid_wrong && evsel->attr.use_clockid) {
+		perf_missing_features.clockid_wrong = true;
+		goto fallback_missing_features;
+	} else if (!perf_missing_features.clockid && evsel->attr.use_clockid) {
+		perf_missing_features.clockid = true;
+		goto fallback_missing_features;
+	} else if (!perf_missing_features.cloexec && (flags & PERF_FLAG_FD_CLOEXEC)) {
+		perf_missing_features.cloexec = true;
+		goto fallback_missing_features;
+	} else if (!perf_missing_features.mmap2 && evsel->attr.mmap2) {
+		perf_missing_features.mmap2 = true;
+		goto fallback_missing_features;
+	} else if (!perf_missing_features.exclude_guest &&
+		   (evsel->attr.exclude_guest || evsel->attr.exclude_host)) {
+		perf_missing_features.exclude_guest = true;
+		goto fallback_missing_features;
+	} else if (!perf_missing_features.sample_id_all) {
+		perf_missing_features.sample_id_all = true;
+		goto retry_sample_id;
+	}
+
+out_close:
+	do {
+		while (--thread >= 0) {
+			close(FD(evsel, cpu, thread));
+			FD(evsel, cpu, thread) = -1;
+		}
+		thread = nthreads;
+	} while (--cpu >= 0);
+	return err;
+}
+
+void perf_evsel__close(struct perf_evsel *evsel, int ncpus, int nthreads)
+{
+	if (evsel->fd == NULL)
+		return;
+
+	perf_evsel__close_fd(evsel, ncpus, nthreads);
+	perf_evsel__free_fd(evsel);
+}
+
+static struct {
+	struct cpu_map map;
+	int cpus[1];
+} empty_cpu_map = {
+	.map.nr	= 1,
+	.cpus	= { -1, },
+};
+
+static struct {
+	struct thread_map map;
+	int threads[1];
+} empty_thread_map = {
+	.map.nr	 = 1,
+	.threads = { -1, },
+};
+
+int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
+		     struct thread_map *threads)
+{
+	if (cpus == NULL) {
+		/* Work around old compiler warnings about strict aliasing */
+		cpus = &empty_cpu_map.map;
+	}
+
+	if (threads == NULL)
+		threads = &empty_thread_map.map;
+
+	return __perf_evsel__open(evsel, cpus, threads);
+}
+
+int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
+			     struct cpu_map *cpus)
+{
+	return __perf_evsel__open(evsel, cpus, &empty_thread_map.map);
+}
+
+int perf_evsel__open_per_thread(struct perf_evsel *evsel,
+				struct thread_map *threads)
+{
+	return __perf_evsel__open(evsel, &empty_cpu_map.map, threads);
+}
+
+static int perf_evsel__parse_id_sample(const struct perf_evsel *evsel,
+				       const union perf_event *event,
+				       struct perf_sample *sample)
+{
+	u64 type = evsel->attr.sample_type;
+	const u64 *array = event->sample.array;
+	bool swapped = evsel->needs_swap;
+	union u64_swap u;
+
+	array += ((event->header.size -
+		   sizeof(event->header)) / sizeof(u64)) - 1;
+
+	if (type & PERF_SAMPLE_IDENTIFIER) {
+		sample->id = *array;
+		array--;
+	}
+
+	if (type & PERF_SAMPLE_CPU) {
+		u.val64 = *array;
+		if (swapped) {
+			/* undo swap of u64, then swap on individual u32s */
+			u.val64 = bswap_64(u.val64);
+			u.val32[0] = bswap_32(u.val32[0]);
+		}
+
+		sample->cpu = u.val32[0];
+		array--;
+	}
+
+	if (type & PERF_SAMPLE_STREAM_ID) {
+		sample->stream_id = *array;
+		array--;
+	}
+
+	if (type & PERF_SAMPLE_ID) {
+		sample->id = *array;
+		array--;
+	}
+
+	if (type & PERF_SAMPLE_TIME) {
+		sample->time = *array;
+		array--;
+	}
+
+	if (type & PERF_SAMPLE_TID) {
+		u.val64 = *array;
+		if (swapped) {
+			/* undo swap of u64, then swap on individual u32s */
+			u.val64 = bswap_64(u.val64);
+			u.val32[0] = bswap_32(u.val32[0]);
+			u.val32[1] = bswap_32(u.val32[1]);
+		}
+
+		sample->pid = u.val32[0];
+		sample->tid = u.val32[1];
+		array--;
+	}
+
+	return 0;
+}
+
+static inline bool overflow(const void *endp, u16 max_size, const void *offset,
+			    u64 size)
+{
+	return size > max_size || offset + size > endp;
+}
+
+#define OVERFLOW_CHECK(offset, size, max_size)				\
+	do {								\
+		if (overflow(endp, (max_size), (offset), (size)))	\
+			return -EFAULT;					\
+	} while (0)
+
+#define OVERFLOW_CHECK_u64(offset) \
+	OVERFLOW_CHECK(offset, sizeof(u64), sizeof(u64))
+
+int perf_evsel__parse_sample(struct perf_evsel *evsel, union perf_event *event,
+			     struct perf_sample *data)
+{
+	u64 type = evsel->attr.sample_type;
+	bool swapped = evsel->needs_swap;
+	const u64 *array;
+	u16 max_size = event->header.size;
+	const void *endp = (void *)event + max_size;
+	u64 sz;
+
+	/*
+	 * used for cross-endian analysis. See git commit 65014ab3
+	 * for why this goofiness is needed.
+	 */
+	union u64_swap u;
+
+	memset(data, 0, sizeof(*data));
+	data->cpu = data->pid = data->tid = -1;
+	data->stream_id = data->id = data->time = -1ULL;
+	data->period = evsel->attr.sample_period;
+	data->weight = 0;
+
+	if (event->header.type != PERF_RECORD_SAMPLE) {
+		if (!evsel->attr.sample_id_all)
+			return 0;
+		return perf_evsel__parse_id_sample(evsel, event, data);
+	}
+
+	array = event->sample.array;
+
+	/*
+	 * The evsel's sample_size is based on PERF_SAMPLE_MASK which includes
+	 * up to PERF_SAMPLE_PERIOD.  After that overflow() must be used to
+	 * check the format does not go past the end of the event.
+	 */
+	if (evsel->sample_size + sizeof(event->header) > event->header.size)
+		return -EFAULT;
+
+	data->id = -1ULL;
+	if (type & PERF_SAMPLE_IDENTIFIER) {
+		data->id = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_IP) {
+		data->ip = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_TID) {
+		u.val64 = *array;
+		if (swapped) {
+			/* undo swap of u64, then swap on individual u32s */
+			u.val64 = bswap_64(u.val64);
+			u.val32[0] = bswap_32(u.val32[0]);
+			u.val32[1] = bswap_32(u.val32[1]);
+		}
+
+		data->pid = u.val32[0];
+		data->tid = u.val32[1];
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_TIME) {
+		data->time = *array;
+		array++;
+	}
+
+	data->addr = 0;
+	if (type & PERF_SAMPLE_ADDR) {
+		data->addr = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_ID) {
+		data->id = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_STREAM_ID) {
+		data->stream_id = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_CPU) {
+
+		u.val64 = *array;
+		if (swapped) {
+			/* undo swap of u64, then swap on individual u32s */
+			u.val64 = bswap_64(u.val64);
+			u.val32[0] = bswap_32(u.val32[0]);
+		}
+
+		data->cpu = u.val32[0];
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_PERIOD) {
+		data->period = *array;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_READ) {
+		u64 read_format = evsel->attr.read_format;
+
+		OVERFLOW_CHECK_u64(array);
+		if (read_format & PERF_FORMAT_GROUP)
+			data->read.group.nr = *array;
+		else
+			data->read.one.value = *array;
+
+		array++;
+
+		if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) {
+			OVERFLOW_CHECK_u64(array);
+			data->read.time_enabled = *array;
+			array++;
+		}
+
+		if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) {
+			OVERFLOW_CHECK_u64(array);
+			data->read.time_running = *array;
+			array++;
+		}
+
+		/* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */
+		if (read_format & PERF_FORMAT_GROUP) {
+			const u64 max_group_nr = UINT64_MAX /
+					sizeof(struct sample_read_value);
+
+			if (data->read.group.nr > max_group_nr)
+				return -EFAULT;
+			sz = data->read.group.nr *
+			     sizeof(struct sample_read_value);
+			OVERFLOW_CHECK(array, sz, max_size);
+			data->read.group.values =
+					(struct sample_read_value *)array;
+			array = (void *)array + sz;
+		} else {
+			OVERFLOW_CHECK_u64(array);
+			data->read.one.id = *array;
+			array++;
+		}
+	}
+
+	if (type & PERF_SAMPLE_CALLCHAIN) {
+		const u64 max_callchain_nr = UINT64_MAX / sizeof(u64);
+
+		OVERFLOW_CHECK_u64(array);
+		data->callchain = (struct ip_callchain *)array++;
+		if (data->callchain->nr > max_callchain_nr)
+			return -EFAULT;
+		sz = data->callchain->nr * sizeof(u64);
+		OVERFLOW_CHECK(array, sz, max_size);
+		array = (void *)array + sz;
+	}
+
+	if (type & PERF_SAMPLE_RAW) {
+		OVERFLOW_CHECK_u64(array);
+		u.val64 = *array;
+		if (WARN_ONCE(swapped,
+			      "Endianness of raw data not corrected!\n")) {
+			/* undo swap of u64, then swap on individual u32s */
+			u.val64 = bswap_64(u.val64);
+			u.val32[0] = bswap_32(u.val32[0]);
+			u.val32[1] = bswap_32(u.val32[1]);
+		}
+		data->raw_size = u.val32[0];
+		array = (void *)array + sizeof(u32);
+
+		OVERFLOW_CHECK(array, data->raw_size, max_size);
+		data->raw_data = (void *)array;
+		array = (void *)array + data->raw_size;
+	}
+
+	if (type & PERF_SAMPLE_BRANCH_STACK) {
+		const u64 max_branch_nr = UINT64_MAX /
+					  sizeof(struct branch_entry);
+
+		OVERFLOW_CHECK_u64(array);
+		data->branch_stack = (struct branch_stack *)array++;
+
+		if (data->branch_stack->nr > max_branch_nr)
+			return -EFAULT;
+		sz = data->branch_stack->nr * sizeof(struct branch_entry);
+		OVERFLOW_CHECK(array, sz, max_size);
+		array = (void *)array + sz;
+	}
+
+	if (type & PERF_SAMPLE_REGS_USER) {
+		OVERFLOW_CHECK_u64(array);
+		data->user_regs.abi = *array;
+		array++;
+
+		if (data->user_regs.abi) {
+			u64 mask = evsel->attr.sample_regs_user;
+
+			sz = hweight_long(mask) * sizeof(u64);
+			OVERFLOW_CHECK(array, sz, max_size);
+			data->user_regs.mask = mask;
+			data->user_regs.regs = (u64 *)array;
+			array = (void *)array + sz;
+		}
+	}
+
+	if (type & PERF_SAMPLE_STACK_USER) {
+		OVERFLOW_CHECK_u64(array);
+		sz = *array++;
+
+		data->user_stack.offset = ((char *)(array - 1)
+					  - (char *) event);
+
+		if (!sz) {
+			data->user_stack.size = 0;
+		} else {
+			OVERFLOW_CHECK(array, sz, max_size);
+			data->user_stack.data = (char *)array;
+			array = (void *)array + sz;
+			OVERFLOW_CHECK_u64(array);
+			data->user_stack.size = *array++;
+			if (WARN_ONCE(data->user_stack.size > sz,
+				      "user stack dump failure\n"))
+				return -EFAULT;
+		}
+	}
+
+	data->weight = 0;
+	if (type & PERF_SAMPLE_WEIGHT) {
+		OVERFLOW_CHECK_u64(array);
+		data->weight = *array;
+		array++;
+	}
+
+	data->data_src = PERF_MEM_DATA_SRC_NONE;
+	if (type & PERF_SAMPLE_DATA_SRC) {
+		OVERFLOW_CHECK_u64(array);
+		data->data_src = *array;
+		array++;
+	}
+
+	data->transaction = 0;
+	if (type & PERF_SAMPLE_TRANSACTION) {
+		OVERFLOW_CHECK_u64(array);
+		data->transaction = *array;
+		array++;
+	}
+
+	data->intr_regs.abi = PERF_SAMPLE_REGS_ABI_NONE;
+	if (type & PERF_SAMPLE_REGS_INTR) {
+		OVERFLOW_CHECK_u64(array);
+		data->intr_regs.abi = *array;
+		array++;
+
+		if (data->intr_regs.abi != PERF_SAMPLE_REGS_ABI_NONE) {
+			u64 mask = evsel->attr.sample_regs_intr;
+
+			sz = hweight_long(mask) * sizeof(u64);
+			OVERFLOW_CHECK(array, sz, max_size);
+			data->intr_regs.mask = mask;
+			data->intr_regs.regs = (u64 *)array;
+			array = (void *)array + sz;
+		}
+	}
+
+	return 0;
+}
+
+size_t perf_event__sample_event_size(const struct perf_sample *sample, u64 type,
+				     u64 read_format)
+{
+	size_t sz, result = sizeof(struct sample_event);
+
+	if (type & PERF_SAMPLE_IDENTIFIER)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_IP)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_TID)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_TIME)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_ADDR)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_ID)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_STREAM_ID)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_CPU)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_PERIOD)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_READ) {
+		result += sizeof(u64);
+		if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+			result += sizeof(u64);
+		if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+			result += sizeof(u64);
+		/* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */
+		if (read_format & PERF_FORMAT_GROUP) {
+			sz = sample->read.group.nr *
+			     sizeof(struct sample_read_value);
+			result += sz;
+		} else {
+			result += sizeof(u64);
+		}
+	}
+
+	if (type & PERF_SAMPLE_CALLCHAIN) {
+		sz = (sample->callchain->nr + 1) * sizeof(u64);
+		result += sz;
+	}
+
+	if (type & PERF_SAMPLE_RAW) {
+		result += sizeof(u32);
+		result += sample->raw_size;
+	}
+
+	if (type & PERF_SAMPLE_BRANCH_STACK) {
+		sz = sample->branch_stack->nr * sizeof(struct branch_entry);
+		sz += sizeof(u64);
+		result += sz;
+	}
+
+	if (type & PERF_SAMPLE_REGS_USER) {
+		if (sample->user_regs.abi) {
+			result += sizeof(u64);
+			sz = hweight_long(sample->user_regs.mask) * sizeof(u64);
+			result += sz;
+		} else {
+			result += sizeof(u64);
+		}
+	}
+
+	if (type & PERF_SAMPLE_STACK_USER) {
+		sz = sample->user_stack.size;
+		result += sizeof(u64);
+		if (sz) {
+			result += sz;
+			result += sizeof(u64);
+		}
+	}
+
+	if (type & PERF_SAMPLE_WEIGHT)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_DATA_SRC)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_TRANSACTION)
+		result += sizeof(u64);
+
+	if (type & PERF_SAMPLE_REGS_INTR) {
+		if (sample->intr_regs.abi) {
+			result += sizeof(u64);
+			sz = hweight_long(sample->intr_regs.mask) * sizeof(u64);
+			result += sz;
+		} else {
+			result += sizeof(u64);
+		}
+	}
+
+	return result;
+}
+
+int perf_event__synthesize_sample(union perf_event *event, u64 type,
+				  u64 read_format,
+				  const struct perf_sample *sample,
+				  bool swapped)
+{
+	u64 *array;
+	size_t sz;
+	/*
+	 * used for cross-endian analysis. See git commit 65014ab3
+	 * for why this goofiness is needed.
+	 */
+	union u64_swap u;
+
+	array = event->sample.array;
+
+	if (type & PERF_SAMPLE_IDENTIFIER) {
+		*array = sample->id;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_IP) {
+		*array = sample->ip;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_TID) {
+		u.val32[0] = sample->pid;
+		u.val32[1] = sample->tid;
+		if (swapped) {
+			/*
+			 * Inverse of what is done in perf_evsel__parse_sample
+			 */
+			u.val32[0] = bswap_32(u.val32[0]);
+			u.val32[1] = bswap_32(u.val32[1]);
+			u.val64 = bswap_64(u.val64);
+		}
+
+		*array = u.val64;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_TIME) {
+		*array = sample->time;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_ADDR) {
+		*array = sample->addr;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_ID) {
+		*array = sample->id;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_STREAM_ID) {
+		*array = sample->stream_id;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_CPU) {
+		u.val32[0] = sample->cpu;
+		if (swapped) {
+			/*
+			 * Inverse of what is done in perf_evsel__parse_sample
+			 */
+			u.val32[0] = bswap_32(u.val32[0]);
+			u.val64 = bswap_64(u.val64);
+		}
+		*array = u.val64;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_PERIOD) {
+		*array = sample->period;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_READ) {
+		if (read_format & PERF_FORMAT_GROUP)
+			*array = sample->read.group.nr;
+		else
+			*array = sample->read.one.value;
+		array++;
+
+		if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) {
+			*array = sample->read.time_enabled;
+			array++;
+		}
+
+		if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) {
+			*array = sample->read.time_running;
+			array++;
+		}
+
+		/* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */
+		if (read_format & PERF_FORMAT_GROUP) {
+			sz = sample->read.group.nr *
+			     sizeof(struct sample_read_value);
+			memcpy(array, sample->read.group.values, sz);
+			array = (void *)array + sz;
+		} else {
+			*array = sample->read.one.id;
+			array++;
+		}
+	}
+
+	if (type & PERF_SAMPLE_CALLCHAIN) {
+		sz = (sample->callchain->nr + 1) * sizeof(u64);
+		memcpy(array, sample->callchain, sz);
+		array = (void *)array + sz;
+	}
+
+	if (type & PERF_SAMPLE_RAW) {
+		u.val32[0] = sample->raw_size;
+		if (WARN_ONCE(swapped,
+			      "Endianness of raw data not corrected!\n")) {
+			/*
+			 * Inverse of what is done in perf_evsel__parse_sample
+			 */
+			u.val32[0] = bswap_32(u.val32[0]);
+			u.val32[1] = bswap_32(u.val32[1]);
+			u.val64 = bswap_64(u.val64);
+		}
+		*array = u.val64;
+		array = (void *)array + sizeof(u32);
+
+		memcpy(array, sample->raw_data, sample->raw_size);
+		array = (void *)array + sample->raw_size;
+	}
+
+	if (type & PERF_SAMPLE_BRANCH_STACK) {
+		sz = sample->branch_stack->nr * sizeof(struct branch_entry);
+		sz += sizeof(u64);
+		memcpy(array, sample->branch_stack, sz);
+		array = (void *)array + sz;
+	}
+
+	if (type & PERF_SAMPLE_REGS_USER) {
+		if (sample->user_regs.abi) {
+			*array++ = sample->user_regs.abi;
+			sz = hweight_long(sample->user_regs.mask) * sizeof(u64);
+			memcpy(array, sample->user_regs.regs, sz);
+			array = (void *)array + sz;
+		} else {
+			*array++ = 0;
+		}
+	}
+
+	if (type & PERF_SAMPLE_STACK_USER) {
+		sz = sample->user_stack.size;
+		*array++ = sz;
+		if (sz) {
+			memcpy(array, sample->user_stack.data, sz);
+			array = (void *)array + sz;
+			*array++ = sz;
+		}
+	}
+
+	if (type & PERF_SAMPLE_WEIGHT) {
+		*array = sample->weight;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_DATA_SRC) {
+		*array = sample->data_src;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_TRANSACTION) {
+		*array = sample->transaction;
+		array++;
+	}
+
+	if (type & PERF_SAMPLE_REGS_INTR) {
+		if (sample->intr_regs.abi) {
+			*array++ = sample->intr_regs.abi;
+			sz = hweight_long(sample->intr_regs.mask) * sizeof(u64);
+			memcpy(array, sample->intr_regs.regs, sz);
+			array = (void *)array + sz;
+		} else {
+			*array++ = 0;
+		}
+	}
+
+	return 0;
+}
+
+struct format_field *perf_evsel__field(struct perf_evsel *evsel, const char *name)
+{
+	return pevent_find_field(evsel->tp_format, name);
+}
+
+void *perf_evsel__rawptr(struct perf_evsel *evsel, struct perf_sample *sample,
+			 const char *name)
+{
+	struct format_field *field = perf_evsel__field(evsel, name);
+	int offset;
+
+	if (!field)
+		return NULL;
+
+	offset = field->offset;
+
+	if (field->flags & FIELD_IS_DYNAMIC) {
+		offset = *(int *)(sample->raw_data + field->offset);
+		offset &= 0xffff;
+	}
+
+	return sample->raw_data + offset;
+}
+
+u64 perf_evsel__intval(struct perf_evsel *evsel, struct perf_sample *sample,
+		       const char *name)
+{
+	struct format_field *field = perf_evsel__field(evsel, name);
+	void *ptr;
+	u64 value;
+
+	if (!field)
+		return 0;
+
+	ptr = sample->raw_data + field->offset;
+
+	switch (field->size) {
+	case 1:
+		return *(u8 *)ptr;
+	case 2:
+		value = *(u16 *)ptr;
+		break;
+	case 4:
+		value = *(u32 *)ptr;
+		break;
+	case 8:
+		memcpy(&value, ptr, sizeof(u64));
+		break;
+	default:
+		return 0;
+	}
+
+	if (!evsel->needs_swap)
+		return value;
+
+	switch (field->size) {
+	case 2:
+		return bswap_16(value);
+	case 4:
+		return bswap_32(value);
+	case 8:
+		return bswap_64(value);
+	default:
+		return 0;
+	}
+
+	return 0;
+}
+
+static int comma_fprintf(FILE *fp, bool *first, const char *fmt, ...)
+{
+	va_list args;
+	int ret = 0;
+
+	if (!*first) {
+		ret += fprintf(fp, ",");
+	} else {
+		ret += fprintf(fp, ":");
+		*first = false;
+	}
+
+	va_start(args, fmt);
+	ret += vfprintf(fp, fmt, args);
+	va_end(args);
+	return ret;
+}
+
+static int __print_attr__fprintf(FILE *fp, const char *name, const char *val, void *priv)
+{
+	return comma_fprintf(fp, (bool *)priv, " %s: %s", name, val);
+}
+
+int perf_evsel__fprintf(struct perf_evsel *evsel,
+			struct perf_attr_details *details, FILE *fp)
+{
+	bool first = true;
+	int printed = 0;
+
+	if (details->event_group) {
+		struct perf_evsel *pos;
+
+		if (!perf_evsel__is_group_leader(evsel))
+			return 0;
+
+		if (evsel->nr_members > 1)
+			printed += fprintf(fp, "%s{", evsel->group_name ?: "");
+
+		printed += fprintf(fp, "%s", perf_evsel__name(evsel));
+		for_each_group_member(pos, evsel)
+			printed += fprintf(fp, ",%s", perf_evsel__name(pos));
+
+		if (evsel->nr_members > 1)
+			printed += fprintf(fp, "}");
+		goto out;
+	}
+
+	printed += fprintf(fp, "%s", perf_evsel__name(evsel));
+
+	if (details->verbose) {
+		printed += perf_event_attr__fprintf(fp, &evsel->attr,
+						    __print_attr__fprintf, &first);
+	} else if (details->freq) {
+		printed += comma_fprintf(fp, &first, " sample_freq=%" PRIu64,
+					 (u64)evsel->attr.sample_freq);
+	}
+out:
+	fputc('\n', fp);
+	return ++printed;
+}
+
+bool perf_evsel__fallback(struct perf_evsel *evsel, int err,
+			  char *msg, size_t msgsize)
+{
+	if ((err == ENOENT || err == ENXIO || err == ENODEV) &&
+	    evsel->attr.type   == PERF_TYPE_HARDWARE &&
+	    evsel->attr.config == PERF_COUNT_HW_CPU_CYCLES) {
+		/*
+		 * If it's cycles then fall back to hrtimer based
+		 * cpu-clock-tick sw counter, which is always available even if
+		 * no PMU support.
+		 *
+		 * PPC returns ENXIO until 2.6.37 (behavior changed with commit
+		 * b0a873e).
+		 */
+		scnprintf(msg, msgsize, "%s",
+"The cycles event is not supported, trying to fall back to cpu-clock-ticks");
+
+		evsel->attr.type   = PERF_TYPE_SOFTWARE;
+		evsel->attr.config = PERF_COUNT_SW_CPU_CLOCK;
+
+		zfree(&evsel->name);
+		return true;
+	}
+
+	return false;
+}
+
+int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
+			      int err, char *msg, size_t size)
+{
+	char sbuf[STRERR_BUFSIZE];
+
+	switch (err) {
+	case EPERM:
+	case EACCES:
+		return scnprintf(msg, size,
+		 "You may not have permission to collect %sstats.\n"
+		 "Consider tweaking /proc/sys/kernel/perf_event_paranoid:\n"
+		 " -1 - Not paranoid at all\n"
+		 "  0 - Disallow raw tracepoint access for unpriv\n"
+		 "  1 - Disallow cpu events for unpriv\n"
+		 "  2 - Disallow kernel profiling for unpriv",
+				 target->system_wide ? "system-wide " : "");
+	case ENOENT:
+		return scnprintf(msg, size, "The %s event is not supported.",
+				 perf_evsel__name(evsel));
+	case EMFILE:
+		return scnprintf(msg, size, "%s",
+			 "Too many events are opened.\n"
+			 "Try again after reducing the number of events.");
+	case ENODEV:
+		if (target->cpu_list)
+			return scnprintf(msg, size, "%s",
+	 "No such device - did you specify an out-of-range profile CPU?\n");
+		break;
+	case EOPNOTSUPP:
+		if (evsel->attr.precise_ip)
+			return scnprintf(msg, size, "%s",
+	"\'precise\' request may not be supported. Try removing 'p' modifier.");
+#if defined(__i386__) || defined(__x86_64__)
+		if (evsel->attr.type == PERF_TYPE_HARDWARE)
+			return scnprintf(msg, size, "%s",
+	"No hardware sampling interrupt available.\n"
+	"No APIC? If so then you can boot the kernel with the \"lapic\" boot parameter to force-enable it.");
+#endif
+		break;
+	case EBUSY:
+		if (find_process("oprofiled"))
+			return scnprintf(msg, size,
+	"The PMU counters are busy/taken by another profiler.\n"
+	"We found oprofile daemon running, please stop it and try again.");
+		break;
+	case EINVAL:
+		if (perf_missing_features.clockid)
+			return scnprintf(msg, size, "clockid feature not supported.");
+		if (perf_missing_features.clockid_wrong)
+			return scnprintf(msg, size, "wrong clockid (%d).", clockid);
+		break;
+	default:
+		break;
+	}
+
+	return scnprintf(msg, size,
+	"The sys_perf_event_open() syscall returned with %d (%s) for event (%s).\n"
+	"/bin/dmesg may provide additional information.\n"
+	"No CONFIG_PERF_EVENTS=y kernel support configured?\n",
+			 err, strerror_r(err, sbuf, sizeof(sbuf)),
+			 perf_evsel__name(evsel));
+}
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
new file mode 100644
index 000000000..e486151b0
--- /dev/null
+++ b/tools/perf/util/evsel.h
@@ -0,0 +1,369 @@
+#ifndef __PERF_EVSEL_H
+#define __PERF_EVSEL_H 1
+
+#include <linux/list.h>
+#include <stdbool.h>
+#include <stddef.h>
+#include <linux/perf_event.h>
+#include <linux/types.h>
+#include "xyarray.h"
+#include "symbol.h"
+
+struct perf_counts_values {
+	union {
+		struct {
+			u64 val;
+			u64 ena;
+			u64 run;
+		};
+		u64 values[3];
+	};
+};
+
+struct perf_counts {
+	s8		   	  scaled;
+	struct perf_counts_values aggr;
+	struct perf_counts_values cpu[];
+};
+
+struct perf_evsel;
+
+/*
+ * Per fd, to map back from PERF_SAMPLE_ID to evsel, only used when there are
+ * more than one entry in the evlist.
+ */
+struct perf_sample_id {
+	struct hlist_node 	node;
+	u64		 	id;
+	struct perf_evsel	*evsel;
+	int			idx;
+	int			cpu;
+	pid_t			tid;
+
+	/* Holds total ID period value for PERF_SAMPLE_READ processing. */
+	u64			period;
+};
+
+struct cgroup_sel;
+
+/** struct perf_evsel - event selector
+ *
+ * @name - Can be set to retain the original event name passed by the user,
+ *         so that when showing results in tools such as 'perf stat', we
+ *         show the name used, not some alias.
+ * @id_pos: the position of the event id (PERF_SAMPLE_ID or
+ *          PERF_SAMPLE_IDENTIFIER) in a sample event i.e. in the array of
+ *          struct sample_event
+ * @is_pos: the position (counting backwards) of the event id (PERF_SAMPLE_ID or
+ *          PERF_SAMPLE_IDENTIFIER) in a non-sample event i.e. if sample_id_all
+ *          is used there is an id sample appended to non-sample events
+ * @priv:   And what is in its containing unnamed union are tool specific
+ */
+struct perf_evsel {
+	struct list_head	node;
+	struct perf_event_attr	attr;
+	char			*filter;
+	struct xyarray		*fd;
+	struct xyarray		*sample_id;
+	u64			*id;
+	struct perf_counts	*counts;
+	struct perf_counts	*prev_raw_counts;
+	int			idx;
+	u32			ids;
+	char			*name;
+	double			scale;
+	const char		*unit;
+	bool			snapshot;
+	struct event_format	*tp_format;
+	union {
+		void		*priv;
+		off_t		id_offset;
+		u64		db_id;
+	};
+	struct cgroup_sel	*cgrp;
+	void			*handler;
+	struct cpu_map		*cpus;
+	unsigned int		sample_size;
+	int			id_pos;
+	int			is_pos;
+	bool 			supported;
+	bool 			needs_swap;
+	bool			no_aux_samples;
+	bool			immediate;
+	bool			system_wide;
+	bool			tracking;
+	bool			per_pkg;
+	unsigned long		*per_pkg_mask;
+	/* parse modifier helper */
+	int			exclude_GH;
+	int			nr_members;
+	int			sample_read;
+	struct perf_evsel	*leader;
+	char			*group_name;
+};
+
+union u64_swap {
+	u64 val64;
+	u32 val32[2];
+};
+
+struct cpu_map;
+struct target;
+struct thread_map;
+struct perf_evlist;
+struct record_opts;
+
+void perf_counts_values__scale(struct perf_counts_values *count,
+			       bool scale, s8 *pscaled);
+
+void perf_evsel__compute_deltas(struct perf_evsel *evsel, int cpu,
+				struct perf_counts_values *count);
+
+int perf_evsel__object_config(size_t object_size,
+			      int (*init)(struct perf_evsel *evsel),
+			      void (*fini)(struct perf_evsel *evsel));
+
+struct perf_evsel *perf_evsel__new_idx(struct perf_event_attr *attr, int idx);
+
+static inline struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr)
+{
+	return perf_evsel__new_idx(attr, 0);
+}
+
+struct perf_evsel *perf_evsel__newtp_idx(const char *sys, const char *name, int idx);
+
+static inline struct perf_evsel *perf_evsel__newtp(const char *sys, const char *name)
+{
+	return perf_evsel__newtp_idx(sys, name, 0);
+}
+
+struct event_format *event_format__new(const char *sys, const char *name);
+
+void perf_evsel__init(struct perf_evsel *evsel,
+		      struct perf_event_attr *attr, int idx);
+void perf_evsel__exit(struct perf_evsel *evsel);
+void perf_evsel__delete(struct perf_evsel *evsel);
+
+void perf_evsel__config(struct perf_evsel *evsel,
+			struct record_opts *opts);
+
+int __perf_evsel__sample_size(u64 sample_type);
+void perf_evsel__calc_id_pos(struct perf_evsel *evsel);
+
+bool perf_evsel__is_cache_op_valid(u8 type, u8 op);
+
+#define PERF_EVSEL__MAX_ALIASES 8
+
+extern const char *perf_evsel__hw_cache[PERF_COUNT_HW_CACHE_MAX]
+				       [PERF_EVSEL__MAX_ALIASES];
+extern const char *perf_evsel__hw_cache_op[PERF_COUNT_HW_CACHE_OP_MAX]
+					  [PERF_EVSEL__MAX_ALIASES];
+extern const char *perf_evsel__hw_cache_result[PERF_COUNT_HW_CACHE_RESULT_MAX]
+					      [PERF_EVSEL__MAX_ALIASES];
+extern const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX];
+extern const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX];
+int __perf_evsel__hw_cache_type_op_res_name(u8 type, u8 op, u8 result,
+					    char *bf, size_t size);
+const char *perf_evsel__name(struct perf_evsel *evsel);
+
+const char *perf_evsel__group_name(struct perf_evsel *evsel);
+int perf_evsel__group_desc(struct perf_evsel *evsel, char *buf, size_t size);
+
+int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads);
+int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus);
+void perf_evsel__reset_counts(struct perf_evsel *evsel, int ncpus);
+void perf_evsel__free_counts(struct perf_evsel *evsel);
+void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads);
+
+void __perf_evsel__set_sample_bit(struct perf_evsel *evsel,
+				  enum perf_event_sample_format bit);
+void __perf_evsel__reset_sample_bit(struct perf_evsel *evsel,
+				    enum perf_event_sample_format bit);
+
+#define perf_evsel__set_sample_bit(evsel, bit) \
+	__perf_evsel__set_sample_bit(evsel, PERF_SAMPLE_##bit)
+
+#define perf_evsel__reset_sample_bit(evsel, bit) \
+	__perf_evsel__reset_sample_bit(evsel, PERF_SAMPLE_##bit)
+
+void perf_evsel__set_sample_id(struct perf_evsel *evsel,
+			       bool use_sample_identifier);
+
+int perf_evsel__set_filter(struct perf_evsel *evsel, int ncpus, int nthreads,
+			   const char *filter);
+int perf_evsel__enable(struct perf_evsel *evsel, int ncpus, int nthreads);
+
+int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
+			     struct cpu_map *cpus);
+int perf_evsel__open_per_thread(struct perf_evsel *evsel,
+				struct thread_map *threads);
+int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
+		     struct thread_map *threads);
+void perf_evsel__close(struct perf_evsel *evsel, int ncpus, int nthreads);
+
+struct perf_sample;
+
+void *perf_evsel__rawptr(struct perf_evsel *evsel, struct perf_sample *sample,
+			 const char *name);
+u64 perf_evsel__intval(struct perf_evsel *evsel, struct perf_sample *sample,
+		       const char *name);
+
+static inline char *perf_evsel__strval(struct perf_evsel *evsel,
+				       struct perf_sample *sample,
+				       const char *name)
+{
+	return perf_evsel__rawptr(evsel, sample, name);
+}
+
+struct format_field;
+
+struct format_field *perf_evsel__field(struct perf_evsel *evsel, const char *name);
+
+#define perf_evsel__match(evsel, t, c)		\
+	(evsel->attr.type == PERF_TYPE_##t &&	\
+	 evsel->attr.config == PERF_COUNT_##c)
+
+static inline bool perf_evsel__match2(struct perf_evsel *e1,
+				      struct perf_evsel *e2)
+{
+	return (e1->attr.type == e2->attr.type) &&
+	       (e1->attr.config == e2->attr.config);
+}
+
+#define perf_evsel__cmp(a, b)			\
+	((a) &&					\
+	 (b) &&					\
+	 (a)->attr.type == (b)->attr.type &&	\
+	 (a)->attr.config == (b)->attr.config)
+
+typedef int (perf_evsel__read_cb_t)(struct perf_evsel *evsel,
+				    int cpu, int thread,
+				    struct perf_counts_values *count);
+
+int perf_evsel__read_cb(struct perf_evsel *evsel, int cpu, int thread,
+			perf_evsel__read_cb_t cb);
+
+int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
+			      int cpu, int thread, bool scale);
+
+/**
+ * perf_evsel__read_on_cpu - Read out the results on a CPU and thread
+ *
+ * @evsel - event selector to read value
+ * @cpu - CPU of interest
+ * @thread - thread of interest
+ */
+static inline int perf_evsel__read_on_cpu(struct perf_evsel *evsel,
+					  int cpu, int thread)
+{
+	return __perf_evsel__read_on_cpu(evsel, cpu, thread, false);
+}
+
+/**
+ * perf_evsel__read_on_cpu_scaled - Read out the results on a CPU and thread, scaled
+ *
+ * @evsel - event selector to read value
+ * @cpu - CPU of interest
+ * @thread - thread of interest
+ */
+static inline int perf_evsel__read_on_cpu_scaled(struct perf_evsel *evsel,
+						 int cpu, int thread)
+{
+	return __perf_evsel__read_on_cpu(evsel, cpu, thread, true);
+}
+
+int perf_evsel__parse_sample(struct perf_evsel *evsel, union perf_event *event,
+			     struct perf_sample *sample);
+
+static inline struct perf_evsel *perf_evsel__next(struct perf_evsel *evsel)
+{
+	return list_entry(evsel->node.next, struct perf_evsel, node);
+}
+
+static inline struct perf_evsel *perf_evsel__prev(struct perf_evsel *evsel)
+{
+	return list_entry(evsel->node.prev, struct perf_evsel, node);
+}
+
+/**
+ * perf_evsel__is_group_leader - Return whether given evsel is a leader event
+ *
+ * @evsel - evsel selector to be tested
+ *
+ * Return %true if @evsel is a group leader or a stand-alone event
+ */
+static inline bool perf_evsel__is_group_leader(const struct perf_evsel *evsel)
+{
+	return evsel->leader == evsel;
+}
+
+/**
+ * perf_evsel__is_group_event - Return whether given evsel is a group event
+ *
+ * @evsel - evsel selector to be tested
+ *
+ * Return %true iff event group view is enabled and @evsel is a actual group
+ * leader which has other members in the group
+ */
+static inline bool perf_evsel__is_group_event(struct perf_evsel *evsel)
+{
+	if (!symbol_conf.event_group)
+		return false;
+
+	return perf_evsel__is_group_leader(evsel) && evsel->nr_members > 1;
+}
+
+/**
+ * perf_evsel__is_function_event - Return whether given evsel is a function
+ * trace event
+ *
+ * @evsel - evsel selector to be tested
+ *
+ * Return %true if event is function trace event
+ */
+static inline bool perf_evsel__is_function_event(struct perf_evsel *evsel)
+{
+#define FUNCTION_EVENT "ftrace:function"
+
+	return evsel->name &&
+	       !strncmp(FUNCTION_EVENT, evsel->name, sizeof(FUNCTION_EVENT));
+
+#undef FUNCTION_EVENT
+}
+
+struct perf_attr_details {
+	bool freq;
+	bool verbose;
+	bool event_group;
+	bool force;
+};
+
+int perf_evsel__fprintf(struct perf_evsel *evsel,
+			struct perf_attr_details *details, FILE *fp);
+
+bool perf_evsel__fallback(struct perf_evsel *evsel, int err,
+			  char *msg, size_t msgsize);
+int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
+			      int err, char *msg, size_t size);
+
+static inline int perf_evsel__group_idx(struct perf_evsel *evsel)
+{
+	return evsel->idx - evsel->leader->idx;
+}
+
+#define for_each_group_member(_evsel, _leader) 					\
+for ((_evsel) = list_entry((_leader)->node.next, struct perf_evsel, node); 	\
+     (_evsel) && (_evsel)->leader == (_leader);					\
+     (_evsel) = list_entry((_evsel)->node.next, struct perf_evsel, node))
+
+static inline bool has_branch_callstack(struct perf_evsel *evsel)
+{
+	return evsel->attr.branch_sample_type & PERF_SAMPLE_BRANCH_CALL_STACK;
+}
+
+typedef int (*attr__fprintf_f)(FILE *, const char *, const char *, void *);
+
+int perf_event_attr__fprintf(FILE *fp, struct perf_event_attr *attr,
+			     attr__fprintf_f attr__fprintf, void *priv);
+
+#endif /* __PERF_EVSEL_H */
diff --git a/tools/perf/util/exec_cmd.c b/tools/perf/util/exec_cmd.c
new file mode 100644
index 000000000..7adf4ad15
--- /dev/null
+++ b/tools/perf/util/exec_cmd.c
@@ -0,0 +1,148 @@
+#include "cache.h"
+#include "exec_cmd.h"
+#include "quote.h"
+
+#include <string.h>
+
+#define MAX_ARGS	32
+
+static const char *argv_exec_path;
+static const char *argv0_path;
+
+const char *system_path(const char *path)
+{
+	static const char *prefix = PREFIX;
+	struct strbuf d = STRBUF_INIT;
+
+	if (is_absolute_path(path))
+		return path;
+
+	strbuf_addf(&d, "%s/%s", prefix, path);
+	path = strbuf_detach(&d, NULL);
+	return path;
+}
+
+const char *perf_extract_argv0_path(const char *argv0)
+{
+	const char *slash;
+
+	if (!argv0 || !*argv0)
+		return NULL;
+	slash = argv0 + strlen(argv0);
+
+	while (argv0 <= slash && !is_dir_sep(*slash))
+		slash--;
+
+	if (slash >= argv0) {
+		argv0_path = strndup(argv0, slash - argv0);
+		return argv0_path ? slash + 1 : NULL;
+	}
+
+	return argv0;
+}
+
+void perf_set_argv_exec_path(const char *exec_path)
+{
+	argv_exec_path = exec_path;
+	/*
+	 * Propagate this setting to external programs.
+	 */
+	setenv(EXEC_PATH_ENVIRONMENT, exec_path, 1);
+}
+
+
+/* Returns the highest-priority, location to look for perf programs. */
+const char *perf_exec_path(void)
+{
+	const char *env;
+
+	if (argv_exec_path)
+		return argv_exec_path;
+
+	env = getenv(EXEC_PATH_ENVIRONMENT);
+	if (env && *env) {
+		return env;
+	}
+
+	return system_path(PERF_EXEC_PATH);
+}
+
+static void add_path(struct strbuf *out, const char *path)
+{
+	if (path && *path) {
+		if (is_absolute_path(path))
+			strbuf_addstr(out, path);
+		else
+			strbuf_addstr(out, make_nonrelative_path(path));
+
+		strbuf_addch(out, PATH_SEP);
+	}
+}
+
+void setup_path(void)
+{
+	const char *old_path = getenv("PATH");
+	struct strbuf new_path = STRBUF_INIT;
+
+	add_path(&new_path, perf_exec_path());
+	add_path(&new_path, argv0_path);
+
+	if (old_path)
+		strbuf_addstr(&new_path, old_path);
+	else
+		strbuf_addstr(&new_path, "/usr/local/bin:/usr/bin:/bin");
+
+	setenv("PATH", new_path.buf, 1);
+
+	strbuf_release(&new_path);
+}
+
+static const char **prepare_perf_cmd(const char **argv)
+{
+	int argc;
+	const char **nargv;
+
+	for (argc = 0; argv[argc]; argc++)
+		; /* just counting */
+	nargv = malloc(sizeof(*nargv) * (argc + 2));
+
+	nargv[0] = "perf";
+	for (argc = 0; argv[argc]; argc++)
+		nargv[argc + 1] = argv[argc];
+	nargv[argc + 1] = NULL;
+	return nargv;
+}
+
+int execv_perf_cmd(const char **argv) {
+	const char **nargv = prepare_perf_cmd(argv);
+
+	/* execvp() can only ever return if it fails */
+	execvp("perf", (char **)nargv);
+
+	free(nargv);
+	return -1;
+}
+
+
+int execl_perf_cmd(const char *cmd,...)
+{
+	int argc;
+	const char *argv[MAX_ARGS + 1];
+	const char *arg;
+	va_list param;
+
+	va_start(param, cmd);
+	argv[0] = cmd;
+	argc = 1;
+	while (argc < MAX_ARGS) {
+		arg = argv[argc++] = va_arg(param, char *);
+		if (!arg)
+			break;
+	}
+	va_end(param);
+	if (MAX_ARGS <= argc)
+		return error("too many args to run %s", cmd);
+
+	argv[argc] = NULL;
+	return execv_perf_cmd(argv);
+}
diff --git a/tools/perf/util/exec_cmd.h b/tools/perf/util/exec_cmd.h
new file mode 100644
index 000000000..bc4b91596
--- /dev/null
+++ b/tools/perf/util/exec_cmd.h
@@ -0,0 +1,12 @@
+#ifndef __PERF_EXEC_CMD_H
+#define __PERF_EXEC_CMD_H
+
+extern void perf_set_argv_exec_path(const char *exec_path);
+extern const char *perf_extract_argv0_path(const char *path);
+extern const char *perf_exec_path(void);
+extern void setup_path(void);
+extern int execv_perf_cmd(const char **argv); /* NULL terminated */
+extern int execl_perf_cmd(const char *cmd, ...);
+extern const char *system_path(const char *path);
+
+#endif /* __PERF_EXEC_CMD_H */
diff --git a/tools/perf/util/find-vdso-map.c b/tools/perf/util/find-vdso-map.c
new file mode 100644
index 000000000..95ef1cffc
--- /dev/null
+++ b/tools/perf/util/find-vdso-map.c
@@ -0,0 +1,30 @@
+static int find_vdso_map(void **start, void **end)
+{
+	FILE *maps;
+	char line[128];
+	int found = 0;
+
+	maps = fopen("/proc/self/maps", "r");
+	if (!maps) {
+		fprintf(stderr, "vdso: cannot open maps\n");
+		return -1;
+	}
+
+	while (!found && fgets(line, sizeof(line), maps)) {
+		int m = -1;
+
+		/* We care only about private r-x mappings. */
+		if (2 != sscanf(line, "%p-%p r-xp %*x %*x:%*x %*u %n",
+				start, end, &m))
+			continue;
+		if (m < 0)
+			continue;
+
+		if (!strncmp(&line[m], VDSO__MAP_NAME,
+			     sizeof(VDSO__MAP_NAME) - 1))
+			found = 1;
+	}
+
+	fclose(maps);
+	return !found;
+}
diff --git a/tools/perf/util/generate-cmdlist.sh b/tools/perf/util/generate-cmdlist.sh
new file mode 100755
index 000000000..36a885d2c
--- /dev/null
+++ b/tools/perf/util/generate-cmdlist.sh
@@ -0,0 +1,39 @@
+#!/bin/sh
+
+echo "/* Automatically generated by $0 */
+struct cmdname_help
+{
+    char name[16];
+    char help[80];
+};
+
+static struct cmdname_help common_cmds[] = {"
+
+sed -n -e 's/^perf-\([^ 	]*\)[ 	].* common.*/\1/p' command-list.txt |
+sort |
+while read cmd
+do
+     sed -n '
+     /^NAME/,/perf-'"$cmd"'/H
+     ${
+            x
+            s/.*perf-'"$cmd"' - \(.*\)/  {"'"$cmd"'", "\1"},/
+	    p
+     }' "Documentation/perf-$cmd.txt"
+done
+
+echo "#ifdef HAVE_LIBELF_SUPPORT"
+sed -n -e 's/^perf-\([^ 	]*\)[ 	].* full.*/\1/p' command-list.txt |
+sort |
+while read cmd
+do
+     sed -n '
+     /^NAME/,/perf-'"$cmd"'/H
+     ${
+            x
+            s/.*perf-'"$cmd"' - \(.*\)/  {"'"$cmd"'", "\1"},/
+	    p
+     }' "Documentation/perf-$cmd.txt"
+done
+echo "#endif /* HAVE_LIBELF_SUPPORT */"
+echo "};"
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
new file mode 100644
index 000000000..918fd8ae2
--- /dev/null
+++ b/tools/perf/util/header.c
@@ -0,0 +1,2781 @@
+#include "util.h"
+#include <sys/types.h>
+#include <byteswap.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+#include <linux/bitops.h>
+#include <sys/utsname.h>
+
+#include "evlist.h"
+#include "evsel.h"
+#include "header.h"
+#include "../perf.h"
+#include "trace-event.h"
+#include "session.h"
+#include "symbol.h"
+#include "debug.h"
+#include "cpumap.h"
+#include "pmu.h"
+#include "vdso.h"
+#include "strbuf.h"
+#include "build-id.h"
+#include "data.h"
+
+static u32 header_argc;
+static const char **header_argv;
+
+/*
+ * magic2 = "PERFILE2"
+ * must be a numerical value to let the endianness
+ * determine the memory layout. That way we are able
+ * to detect endianness when reading the perf.data file
+ * back.
+ *
+ * we check for legacy (PERFFILE) format.
+ */
+static const char *__perf_magic1 = "PERFFILE";
+static const u64 __perf_magic2    = 0x32454c4946524550ULL;
+static const u64 __perf_magic2_sw = 0x50455246494c4532ULL;
+
+#define PERF_MAGIC	__perf_magic2
+
+struct perf_file_attr {
+	struct perf_event_attr	attr;
+	struct perf_file_section	ids;
+};
+
+void perf_header__set_feat(struct perf_header *header, int feat)
+{
+	set_bit(feat, header->adds_features);
+}
+
+void perf_header__clear_feat(struct perf_header *header, int feat)
+{
+	clear_bit(feat, header->adds_features);
+}
+
+bool perf_header__has_feat(const struct perf_header *header, int feat)
+{
+	return test_bit(feat, header->adds_features);
+}
+
+static int do_write(int fd, const void *buf, size_t size)
+{
+	while (size) {
+		int ret = write(fd, buf, size);
+
+		if (ret < 0)
+			return -errno;
+
+		size -= ret;
+		buf += ret;
+	}
+
+	return 0;
+}
+
+int write_padded(int fd, const void *bf, size_t count, size_t count_aligned)
+{
+	static const char zero_buf[NAME_ALIGN];
+	int err = do_write(fd, bf, count);
+
+	if (!err)
+		err = do_write(fd, zero_buf, count_aligned - count);
+
+	return err;
+}
+
+static int do_write_string(int fd, const char *str)
+{
+	u32 len, olen;
+	int ret;
+
+	olen = strlen(str) + 1;
+	len = PERF_ALIGN(olen, NAME_ALIGN);
+
+	/* write len, incl. \0 */
+	ret = do_write(fd, &len, sizeof(len));
+	if (ret < 0)
+		return ret;
+
+	return write_padded(fd, str, olen, len);
+}
+
+static char *do_read_string(int fd, struct perf_header *ph)
+{
+	ssize_t sz, ret;
+	u32 len;
+	char *buf;
+
+	sz = readn(fd, &len, sizeof(len));
+	if (sz < (ssize_t)sizeof(len))
+		return NULL;
+
+	if (ph->needs_swap)
+		len = bswap_32(len);
+
+	buf = malloc(len);
+	if (!buf)
+		return NULL;
+
+	ret = readn(fd, buf, len);
+	if (ret == (ssize_t)len) {
+		/*
+		 * strings are padded by zeroes
+		 * thus the actual strlen of buf
+		 * may be less than len
+		 */
+		return buf;
+	}
+
+	free(buf);
+	return NULL;
+}
+
+int
+perf_header__set_cmdline(int argc, const char **argv)
+{
+	int i;
+
+	/*
+	 * If header_argv has already been set, do not override it.
+	 * This allows a command to set the cmdline, parse args and
+	 * then call another builtin function that implements a
+	 * command -- e.g, cmd_kvm calling cmd_record.
+	 */
+	if (header_argv)
+		return 0;
+
+	header_argc = (u32)argc;
+
+	/* do not include NULL termination */
+	header_argv = calloc(argc, sizeof(char *));
+	if (!header_argv)
+		return -ENOMEM;
+
+	/*
+	 * must copy argv contents because it gets moved
+	 * around during option parsing
+	 */
+	for (i = 0; i < argc ; i++)
+		header_argv[i] = argv[i];
+
+	return 0;
+}
+
+static int write_tracing_data(int fd, struct perf_header *h __maybe_unused,
+			    struct perf_evlist *evlist)
+{
+	return read_tracing_data(fd, &evlist->entries);
+}
+
+
+static int write_build_id(int fd, struct perf_header *h,
+			  struct perf_evlist *evlist __maybe_unused)
+{
+	struct perf_session *session;
+	int err;
+
+	session = container_of(h, struct perf_session, header);
+
+	if (!perf_session__read_build_ids(session, true))
+		return -1;
+
+	err = perf_session__write_buildid_table(session, fd);
+	if (err < 0) {
+		pr_debug("failed to write buildid table\n");
+		return err;
+	}
+	perf_session__cache_build_ids(session);
+
+	return 0;
+}
+
+static int write_hostname(int fd, struct perf_header *h __maybe_unused,
+			  struct perf_evlist *evlist __maybe_unused)
+{
+	struct utsname uts;
+	int ret;
+
+	ret = uname(&uts);
+	if (ret < 0)
+		return -1;
+
+	return do_write_string(fd, uts.nodename);
+}
+
+static int write_osrelease(int fd, struct perf_header *h __maybe_unused,
+			   struct perf_evlist *evlist __maybe_unused)
+{
+	struct utsname uts;
+	int ret;
+
+	ret = uname(&uts);
+	if (ret < 0)
+		return -1;
+
+	return do_write_string(fd, uts.release);
+}
+
+static int write_arch(int fd, struct perf_header *h __maybe_unused,
+		      struct perf_evlist *evlist __maybe_unused)
+{
+	struct utsname uts;
+	int ret;
+
+	ret = uname(&uts);
+	if (ret < 0)
+		return -1;
+
+	return do_write_string(fd, uts.machine);
+}
+
+static int write_version(int fd, struct perf_header *h __maybe_unused,
+			 struct perf_evlist *evlist __maybe_unused)
+{
+	return do_write_string(fd, perf_version_string);
+}
+
+static int __write_cpudesc(int fd, const char *cpuinfo_proc)
+{
+	FILE *file;
+	char *buf = NULL;
+	char *s, *p;
+	const char *search = cpuinfo_proc;
+	size_t len = 0;
+	int ret = -1;
+
+	if (!search)
+		return -1;
+
+	file = fopen("/proc/cpuinfo", "r");
+	if (!file)
+		return -1;
+
+	while (getline(&buf, &len, file) > 0) {
+		ret = strncmp(buf, search, strlen(search));
+		if (!ret)
+			break;
+	}
+
+	if (ret) {
+		ret = -1;
+		goto done;
+	}
+
+	s = buf;
+
+	p = strchr(buf, ':');
+	if (p && *(p+1) == ' ' && *(p+2))
+		s = p + 2;
+	p = strchr(s, '\n');
+	if (p)
+		*p = '\0';
+
+	/* squash extra space characters (branding string) */
+	p = s;
+	while (*p) {
+		if (isspace(*p)) {
+			char *r = p + 1;
+			char *q = r;
+			*p = ' ';
+			while (*q && isspace(*q))
+				q++;
+			if (q != (p+1))
+				while ((*r++ = *q++));
+		}
+		p++;
+	}
+	ret = do_write_string(fd, s);
+done:
+	free(buf);
+	fclose(file);
+	return ret;
+}
+
+static int write_cpudesc(int fd, struct perf_header *h __maybe_unused,
+		       struct perf_evlist *evlist __maybe_unused)
+{
+#ifndef CPUINFO_PROC
+#define CPUINFO_PROC {"model name", }
+#endif
+	const char *cpuinfo_procs[] = CPUINFO_PROC;
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(cpuinfo_procs); i++) {
+		int ret;
+		ret = __write_cpudesc(fd, cpuinfo_procs[i]);
+		if (ret >= 0)
+			return ret;
+	}
+	return -1;
+}
+
+
+static int write_nrcpus(int fd, struct perf_header *h __maybe_unused,
+			struct perf_evlist *evlist __maybe_unused)
+{
+	long nr;
+	u32 nrc, nra;
+	int ret;
+
+	nr = sysconf(_SC_NPROCESSORS_CONF);
+	if (nr < 0)
+		return -1;
+
+	nrc = (u32)(nr & UINT_MAX);
+
+	nr = sysconf(_SC_NPROCESSORS_ONLN);
+	if (nr < 0)
+		return -1;
+
+	nra = (u32)(nr & UINT_MAX);
+
+	ret = do_write(fd, &nrc, sizeof(nrc));
+	if (ret < 0)
+		return ret;
+
+	return do_write(fd, &nra, sizeof(nra));
+}
+
+static int write_event_desc(int fd, struct perf_header *h __maybe_unused,
+			    struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+	u32 nre, nri, sz;
+	int ret;
+
+	nre = evlist->nr_entries;
+
+	/*
+	 * write number of events
+	 */
+	ret = do_write(fd, &nre, sizeof(nre));
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * size of perf_event_attr struct
+	 */
+	sz = (u32)sizeof(evsel->attr);
+	ret = do_write(fd, &sz, sizeof(sz));
+	if (ret < 0)
+		return ret;
+
+	evlist__for_each(evlist, evsel) {
+		ret = do_write(fd, &evsel->attr, sz);
+		if (ret < 0)
+			return ret;
+		/*
+		 * write number of unique id per event
+		 * there is one id per instance of an event
+		 *
+		 * copy into an nri to be independent of the
+		 * type of ids,
+		 */
+		nri = evsel->ids;
+		ret = do_write(fd, &nri, sizeof(nri));
+		if (ret < 0)
+			return ret;
+
+		/*
+		 * write event string as passed on cmdline
+		 */
+		ret = do_write_string(fd, perf_evsel__name(evsel));
+		if (ret < 0)
+			return ret;
+		/*
+		 * write unique ids for this event
+		 */
+		ret = do_write(fd, evsel->id, evsel->ids * sizeof(u64));
+		if (ret < 0)
+			return ret;
+	}
+	return 0;
+}
+
+static int write_cmdline(int fd, struct perf_header *h __maybe_unused,
+			 struct perf_evlist *evlist __maybe_unused)
+{
+	char buf[MAXPATHLEN];
+	char proc[32];
+	u32 i, n;
+	int ret;
+
+	/*
+	 * actual atual path to perf binary
+	 */
+	sprintf(proc, "/proc/%d/exe", getpid());
+	ret = readlink(proc, buf, sizeof(buf));
+	if (ret <= 0)
+		return -1;
+
+	/* readlink() does not add null termination */
+	buf[ret] = '\0';
+
+	/* account for binary path */
+	n = header_argc + 1;
+
+	ret = do_write(fd, &n, sizeof(n));
+	if (ret < 0)
+		return ret;
+
+	ret = do_write_string(fd, buf);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0 ; i < header_argc; i++) {
+		ret = do_write_string(fd, header_argv[i]);
+		if (ret < 0)
+			return ret;
+	}
+	return 0;
+}
+
+#define CORE_SIB_FMT \
+	"/sys/devices/system/cpu/cpu%d/topology/core_siblings_list"
+#define THRD_SIB_FMT \
+	"/sys/devices/system/cpu/cpu%d/topology/thread_siblings_list"
+
+struct cpu_topo {
+	u32 core_sib;
+	u32 thread_sib;
+	char **core_siblings;
+	char **thread_siblings;
+};
+
+static int build_cpu_topo(struct cpu_topo *tp, int cpu)
+{
+	FILE *fp;
+	char filename[MAXPATHLEN];
+	char *buf = NULL, *p;
+	size_t len = 0;
+	ssize_t sret;
+	u32 i = 0;
+	int ret = -1;
+
+	sprintf(filename, CORE_SIB_FMT, cpu);
+	fp = fopen(filename, "r");
+	if (!fp)
+		goto try_threads;
+
+	sret = getline(&buf, &len, fp);
+	fclose(fp);
+	if (sret <= 0)
+		goto try_threads;
+
+	p = strchr(buf, '\n');
+	if (p)
+		*p = '\0';
+
+	for (i = 0; i < tp->core_sib; i++) {
+		if (!strcmp(buf, tp->core_siblings[i]))
+			break;
+	}
+	if (i == tp->core_sib) {
+		tp->core_siblings[i] = buf;
+		tp->core_sib++;
+		buf = NULL;
+		len = 0;
+	}
+	ret = 0;
+
+try_threads:
+	sprintf(filename, THRD_SIB_FMT, cpu);
+	fp = fopen(filename, "r");
+	if (!fp)
+		goto done;
+
+	if (getline(&buf, &len, fp) <= 0)
+		goto done;
+
+	p = strchr(buf, '\n');
+	if (p)
+		*p = '\0';
+
+	for (i = 0; i < tp->thread_sib; i++) {
+		if (!strcmp(buf, tp->thread_siblings[i]))
+			break;
+	}
+	if (i == tp->thread_sib) {
+		tp->thread_siblings[i] = buf;
+		tp->thread_sib++;
+		buf = NULL;
+	}
+	ret = 0;
+done:
+	if(fp)
+		fclose(fp);
+	free(buf);
+	return ret;
+}
+
+static void free_cpu_topo(struct cpu_topo *tp)
+{
+	u32 i;
+
+	if (!tp)
+		return;
+
+	for (i = 0 ; i < tp->core_sib; i++)
+		zfree(&tp->core_siblings[i]);
+
+	for (i = 0 ; i < tp->thread_sib; i++)
+		zfree(&tp->thread_siblings[i]);
+
+	free(tp);
+}
+
+static struct cpu_topo *build_cpu_topology(void)
+{
+	struct cpu_topo *tp;
+	void *addr;
+	u32 nr, i;
+	size_t sz;
+	long ncpus;
+	int ret = -1;
+
+	ncpus = sysconf(_SC_NPROCESSORS_CONF);
+	if (ncpus < 0)
+		return NULL;
+
+	nr = (u32)(ncpus & UINT_MAX);
+
+	sz = nr * sizeof(char *);
+
+	addr = calloc(1, sizeof(*tp) + 2 * sz);
+	if (!addr)
+		return NULL;
+
+	tp = addr;
+
+	addr += sizeof(*tp);
+	tp->core_siblings = addr;
+	addr += sz;
+	tp->thread_siblings = addr;
+
+	for (i = 0; i < nr; i++) {
+		ret = build_cpu_topo(tp, i);
+		if (ret < 0)
+			break;
+	}
+	if (ret) {
+		free_cpu_topo(tp);
+		tp = NULL;
+	}
+	return tp;
+}
+
+static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
+			  struct perf_evlist *evlist __maybe_unused)
+{
+	struct cpu_topo *tp;
+	u32 i;
+	int ret;
+
+	tp = build_cpu_topology();
+	if (!tp)
+		return -1;
+
+	ret = do_write(fd, &tp->core_sib, sizeof(tp->core_sib));
+	if (ret < 0)
+		goto done;
+
+	for (i = 0; i < tp->core_sib; i++) {
+		ret = do_write_string(fd, tp->core_siblings[i]);
+		if (ret < 0)
+			goto done;
+	}
+	ret = do_write(fd, &tp->thread_sib, sizeof(tp->thread_sib));
+	if (ret < 0)
+		goto done;
+
+	for (i = 0; i < tp->thread_sib; i++) {
+		ret = do_write_string(fd, tp->thread_siblings[i]);
+		if (ret < 0)
+			break;
+	}
+done:
+	free_cpu_topo(tp);
+	return ret;
+}
+
+
+
+static int write_total_mem(int fd, struct perf_header *h __maybe_unused,
+			  struct perf_evlist *evlist __maybe_unused)
+{
+	char *buf = NULL;
+	FILE *fp;
+	size_t len = 0;
+	int ret = -1, n;
+	uint64_t mem;
+
+	fp = fopen("/proc/meminfo", "r");
+	if (!fp)
+		return -1;
+
+	while (getline(&buf, &len, fp) > 0) {
+		ret = strncmp(buf, "MemTotal:", 9);
+		if (!ret)
+			break;
+	}
+	if (!ret) {
+		n = sscanf(buf, "%*s %"PRIu64, &mem);
+		if (n == 1)
+			ret = do_write(fd, &mem, sizeof(mem));
+	} else
+		ret = -1;
+	free(buf);
+	fclose(fp);
+	return ret;
+}
+
+static int write_topo_node(int fd, int node)
+{
+	char str[MAXPATHLEN];
+	char field[32];
+	char *buf = NULL, *p;
+	size_t len = 0;
+	FILE *fp;
+	u64 mem_total, mem_free, mem;
+	int ret = -1;
+
+	sprintf(str, "/sys/devices/system/node/node%d/meminfo", node);
+	fp = fopen(str, "r");
+	if (!fp)
+		return -1;
+
+	while (getline(&buf, &len, fp) > 0) {
+		/* skip over invalid lines */
+		if (!strchr(buf, ':'))
+			continue;
+		if (sscanf(buf, "%*s %*d %31s %"PRIu64, field, &mem) != 2)
+			goto done;
+		if (!strcmp(field, "MemTotal:"))
+			mem_total = mem;
+		if (!strcmp(field, "MemFree:"))
+			mem_free = mem;
+	}
+
+	fclose(fp);
+	fp = NULL;
+
+	ret = do_write(fd, &mem_total, sizeof(u64));
+	if (ret)
+		goto done;
+
+	ret = do_write(fd, &mem_free, sizeof(u64));
+	if (ret)
+		goto done;
+
+	ret = -1;
+	sprintf(str, "/sys/devices/system/node/node%d/cpulist", node);
+
+	fp = fopen(str, "r");
+	if (!fp)
+		goto done;
+
+	if (getline(&buf, &len, fp) <= 0)
+		goto done;
+
+	p = strchr(buf, '\n');
+	if (p)
+		*p = '\0';
+
+	ret = do_write_string(fd, buf);
+done:
+	free(buf);
+	if (fp)
+		fclose(fp);
+	return ret;
+}
+
+static int write_numa_topology(int fd, struct perf_header *h __maybe_unused,
+			  struct perf_evlist *evlist __maybe_unused)
+{
+	char *buf = NULL;
+	size_t len = 0;
+	FILE *fp;
+	struct cpu_map *node_map = NULL;
+	char *c;
+	u32 nr, i, j;
+	int ret = -1;
+
+	fp = fopen("/sys/devices/system/node/online", "r");
+	if (!fp)
+		return -1;
+
+	if (getline(&buf, &len, fp) <= 0)
+		goto done;
+
+	c = strchr(buf, '\n');
+	if (c)
+		*c = '\0';
+
+	node_map = cpu_map__new(buf);
+	if (!node_map)
+		goto done;
+
+	nr = (u32)node_map->nr;
+
+	ret = do_write(fd, &nr, sizeof(nr));
+	if (ret < 0)
+		goto done;
+
+	for (i = 0; i < nr; i++) {
+		j = (u32)node_map->map[i];
+		ret = do_write(fd, &j, sizeof(j));
+		if (ret < 0)
+			break;
+
+		ret = write_topo_node(fd, i);
+		if (ret < 0)
+			break;
+	}
+done:
+	free(buf);
+	fclose(fp);
+	free(node_map);
+	return ret;
+}
+
+/*
+ * File format:
+ *
+ * struct pmu_mappings {
+ *	u32	pmu_num;
+ *	struct pmu_map {
+ *		u32	type;
+ *		char	name[];
+ *	}[pmu_num];
+ * };
+ */
+
+static int write_pmu_mappings(int fd, struct perf_header *h __maybe_unused,
+			      struct perf_evlist *evlist __maybe_unused)
+{
+	struct perf_pmu *pmu = NULL;
+	off_t offset = lseek(fd, 0, SEEK_CUR);
+	__u32 pmu_num = 0;
+	int ret;
+
+	/* write real pmu_num later */
+	ret = do_write(fd, &pmu_num, sizeof(pmu_num));
+	if (ret < 0)
+		return ret;
+
+	while ((pmu = perf_pmu__scan(pmu))) {
+		if (!pmu->name)
+			continue;
+		pmu_num++;
+
+		ret = do_write(fd, &pmu->type, sizeof(pmu->type));
+		if (ret < 0)
+			return ret;
+
+		ret = do_write_string(fd, pmu->name);
+		if (ret < 0)
+			return ret;
+	}
+
+	if (pwrite(fd, &pmu_num, sizeof(pmu_num), offset) != sizeof(pmu_num)) {
+		/* discard all */
+		lseek(fd, offset, SEEK_SET);
+		return -1;
+	}
+
+	return 0;
+}
+
+/*
+ * File format:
+ *
+ * struct group_descs {
+ *	u32	nr_groups;
+ *	struct group_desc {
+ *		char	name[];
+ *		u32	leader_idx;
+ *		u32	nr_members;
+ *	}[nr_groups];
+ * };
+ */
+static int write_group_desc(int fd, struct perf_header *h __maybe_unused,
+			    struct perf_evlist *evlist)
+{
+	u32 nr_groups = evlist->nr_groups;
+	struct perf_evsel *evsel;
+	int ret;
+
+	ret = do_write(fd, &nr_groups, sizeof(nr_groups));
+	if (ret < 0)
+		return ret;
+
+	evlist__for_each(evlist, evsel) {
+		if (perf_evsel__is_group_leader(evsel) &&
+		    evsel->nr_members > 1) {
+			const char *name = evsel->group_name ?: "{anon_group}";
+			u32 leader_idx = evsel->idx;
+			u32 nr_members = evsel->nr_members;
+
+			ret = do_write_string(fd, name);
+			if (ret < 0)
+				return ret;
+
+			ret = do_write(fd, &leader_idx, sizeof(leader_idx));
+			if (ret < 0)
+				return ret;
+
+			ret = do_write(fd, &nr_members, sizeof(nr_members));
+			if (ret < 0)
+				return ret;
+		}
+	}
+	return 0;
+}
+
+/*
+ * default get_cpuid(): nothing gets recorded
+ * actual implementation must be in arch/$(ARCH)/util/header.c
+ */
+int __attribute__ ((weak)) get_cpuid(char *buffer __maybe_unused,
+				     size_t sz __maybe_unused)
+{
+	return -1;
+}
+
+static int write_cpuid(int fd, struct perf_header *h __maybe_unused,
+		       struct perf_evlist *evlist __maybe_unused)
+{
+	char buffer[64];
+	int ret;
+
+	ret = get_cpuid(buffer, sizeof(buffer));
+	if (!ret)
+		goto write_it;
+
+	return -1;
+write_it:
+	return do_write_string(fd, buffer);
+}
+
+static int write_branch_stack(int fd __maybe_unused,
+			      struct perf_header *h __maybe_unused,
+		       struct perf_evlist *evlist __maybe_unused)
+{
+	return 0;
+}
+
+static void print_hostname(struct perf_header *ph, int fd __maybe_unused,
+			   FILE *fp)
+{
+	fprintf(fp, "# hostname : %s\n", ph->env.hostname);
+}
+
+static void print_osrelease(struct perf_header *ph, int fd __maybe_unused,
+			    FILE *fp)
+{
+	fprintf(fp, "# os release : %s\n", ph->env.os_release);
+}
+
+static void print_arch(struct perf_header *ph, int fd __maybe_unused, FILE *fp)
+{
+	fprintf(fp, "# arch : %s\n", ph->env.arch);
+}
+
+static void print_cpudesc(struct perf_header *ph, int fd __maybe_unused,
+			  FILE *fp)
+{
+	fprintf(fp, "# cpudesc : %s\n", ph->env.cpu_desc);
+}
+
+static void print_nrcpus(struct perf_header *ph, int fd __maybe_unused,
+			 FILE *fp)
+{
+	fprintf(fp, "# nrcpus online : %u\n", ph->env.nr_cpus_online);
+	fprintf(fp, "# nrcpus avail : %u\n", ph->env.nr_cpus_avail);
+}
+
+static void print_version(struct perf_header *ph, int fd __maybe_unused,
+			  FILE *fp)
+{
+	fprintf(fp, "# perf version : %s\n", ph->env.version);
+}
+
+static void print_cmdline(struct perf_header *ph, int fd __maybe_unused,
+			  FILE *fp)
+{
+	int nr, i;
+	char *str;
+
+	nr = ph->env.nr_cmdline;
+	str = ph->env.cmdline;
+
+	fprintf(fp, "# cmdline : ");
+
+	for (i = 0; i < nr; i++) {
+		fprintf(fp, "%s ", str);
+		str += strlen(str) + 1;
+	}
+	fputc('\n', fp);
+}
+
+static void print_cpu_topology(struct perf_header *ph, int fd __maybe_unused,
+			       FILE *fp)
+{
+	int nr, i;
+	char *str;
+
+	nr = ph->env.nr_sibling_cores;
+	str = ph->env.sibling_cores;
+
+	for (i = 0; i < nr; i++) {
+		fprintf(fp, "# sibling cores   : %s\n", str);
+		str += strlen(str) + 1;
+	}
+
+	nr = ph->env.nr_sibling_threads;
+	str = ph->env.sibling_threads;
+
+	for (i = 0; i < nr; i++) {
+		fprintf(fp, "# sibling threads : %s\n", str);
+		str += strlen(str) + 1;
+	}
+}
+
+static void free_event_desc(struct perf_evsel *events)
+{
+	struct perf_evsel *evsel;
+
+	if (!events)
+		return;
+
+	for (evsel = events; evsel->attr.size; evsel++) {
+		zfree(&evsel->name);
+		zfree(&evsel->id);
+	}
+
+	free(events);
+}
+
+static struct perf_evsel *
+read_event_desc(struct perf_header *ph, int fd)
+{
+	struct perf_evsel *evsel, *events = NULL;
+	u64 *id;
+	void *buf = NULL;
+	u32 nre, sz, nr, i, j;
+	ssize_t ret;
+	size_t msz;
+
+	/* number of events */
+	ret = readn(fd, &nre, sizeof(nre));
+	if (ret != (ssize_t)sizeof(nre))
+		goto error;
+
+	if (ph->needs_swap)
+		nre = bswap_32(nre);
+
+	ret = readn(fd, &sz, sizeof(sz));
+	if (ret != (ssize_t)sizeof(sz))
+		goto error;
+
+	if (ph->needs_swap)
+		sz = bswap_32(sz);
+
+	/* buffer to hold on file attr struct */
+	buf = malloc(sz);
+	if (!buf)
+		goto error;
+
+	/* the last event terminates with evsel->attr.size == 0: */
+	events = calloc(nre + 1, sizeof(*events));
+	if (!events)
+		goto error;
+
+	msz = sizeof(evsel->attr);
+	if (sz < msz)
+		msz = sz;
+
+	for (i = 0, evsel = events; i < nre; evsel++, i++) {
+		evsel->idx = i;
+
+		/*
+		 * must read entire on-file attr struct to
+		 * sync up with layout.
+		 */
+		ret = readn(fd, buf, sz);
+		if (ret != (ssize_t)sz)
+			goto error;
+
+		if (ph->needs_swap)
+			perf_event__attr_swap(buf);
+
+		memcpy(&evsel->attr, buf, msz);
+
+		ret = readn(fd, &nr, sizeof(nr));
+		if (ret != (ssize_t)sizeof(nr))
+			goto error;
+
+		if (ph->needs_swap) {
+			nr = bswap_32(nr);
+			evsel->needs_swap = true;
+		}
+
+		evsel->name = do_read_string(fd, ph);
+
+		if (!nr)
+			continue;
+
+		id = calloc(nr, sizeof(*id));
+		if (!id)
+			goto error;
+		evsel->ids = nr;
+		evsel->id = id;
+
+		for (j = 0 ; j < nr; j++) {
+			ret = readn(fd, id, sizeof(*id));
+			if (ret != (ssize_t)sizeof(*id))
+				goto error;
+			if (ph->needs_swap)
+				*id = bswap_64(*id);
+			id++;
+		}
+	}
+out:
+	free(buf);
+	return events;
+error:
+	if (events)
+		free_event_desc(events);
+	events = NULL;
+	goto out;
+}
+
+static int __desc_attr__fprintf(FILE *fp, const char *name, const char *val,
+				void *priv __attribute__((unused)))
+{
+	return fprintf(fp, ", %s = %s", name, val);
+}
+
+static void print_event_desc(struct perf_header *ph, int fd, FILE *fp)
+{
+	struct perf_evsel *evsel, *events = read_event_desc(ph, fd);
+	u32 j;
+	u64 *id;
+
+	if (!events) {
+		fprintf(fp, "# event desc: not available or unable to read\n");
+		return;
+	}
+
+	for (evsel = events; evsel->attr.size; evsel++) {
+		fprintf(fp, "# event : name = %s, ", evsel->name);
+
+		if (evsel->ids) {
+			fprintf(fp, ", id = {");
+			for (j = 0, id = evsel->id; j < evsel->ids; j++, id++) {
+				if (j)
+					fputc(',', fp);
+				fprintf(fp, " %"PRIu64, *id);
+			}
+			fprintf(fp, " }");
+		}
+
+		perf_event_attr__fprintf(fp, &evsel->attr, __desc_attr__fprintf, NULL);
+
+		fputc('\n', fp);
+	}
+
+	free_event_desc(events);
+}
+
+static void print_total_mem(struct perf_header *ph, int fd __maybe_unused,
+			    FILE *fp)
+{
+	fprintf(fp, "# total memory : %Lu kB\n", ph->env.total_mem);
+}
+
+static void print_numa_topology(struct perf_header *ph, int fd __maybe_unused,
+				FILE *fp)
+{
+	u32 nr, c, i;
+	char *str, *tmp;
+	uint64_t mem_total, mem_free;
+
+	/* nr nodes */
+	nr = ph->env.nr_numa_nodes;
+	str = ph->env.numa_nodes;
+
+	for (i = 0; i < nr; i++) {
+		/* node number */
+		c = strtoul(str, &tmp, 0);
+		if (*tmp != ':')
+			goto error;
+
+		str = tmp + 1;
+		mem_total = strtoull(str, &tmp, 0);
+		if (*tmp != ':')
+			goto error;
+
+		str = tmp + 1;
+		mem_free = strtoull(str, &tmp, 0);
+		if (*tmp != ':')
+			goto error;
+
+		fprintf(fp, "# node%u meminfo  : total = %"PRIu64" kB,"
+			    " free = %"PRIu64" kB\n",
+			c, mem_total, mem_free);
+
+		str = tmp + 1;
+		fprintf(fp, "# node%u cpu list : %s\n", c, str);
+
+		str += strlen(str) + 1;
+	}
+	return;
+error:
+	fprintf(fp, "# numa topology : not available\n");
+}
+
+static void print_cpuid(struct perf_header *ph, int fd __maybe_unused, FILE *fp)
+{
+	fprintf(fp, "# cpuid : %s\n", ph->env.cpuid);
+}
+
+static void print_branch_stack(struct perf_header *ph __maybe_unused,
+			       int fd __maybe_unused, FILE *fp)
+{
+	fprintf(fp, "# contains samples with branch stack\n");
+}
+
+static void print_pmu_mappings(struct perf_header *ph, int fd __maybe_unused,
+			       FILE *fp)
+{
+	const char *delimiter = "# pmu mappings: ";
+	char *str, *tmp;
+	u32 pmu_num;
+	u32 type;
+
+	pmu_num = ph->env.nr_pmu_mappings;
+	if (!pmu_num) {
+		fprintf(fp, "# pmu mappings: not available\n");
+		return;
+	}
+
+	str = ph->env.pmu_mappings;
+
+	while (pmu_num) {
+		type = strtoul(str, &tmp, 0);
+		if (*tmp != ':')
+			goto error;
+
+		str = tmp + 1;
+		fprintf(fp, "%s%s = %" PRIu32, delimiter, str, type);
+
+		delimiter = ", ";
+		str += strlen(str) + 1;
+		pmu_num--;
+	}
+
+	fprintf(fp, "\n");
+
+	if (!pmu_num)
+		return;
+error:
+	fprintf(fp, "# pmu mappings: unable to read\n");
+}
+
+static void print_group_desc(struct perf_header *ph, int fd __maybe_unused,
+			     FILE *fp)
+{
+	struct perf_session *session;
+	struct perf_evsel *evsel;
+	u32 nr = 0;
+
+	session = container_of(ph, struct perf_session, header);
+
+	evlist__for_each(session->evlist, evsel) {
+		if (perf_evsel__is_group_leader(evsel) &&
+		    evsel->nr_members > 1) {
+			fprintf(fp, "# group: %s{%s", evsel->group_name ?: "",
+				perf_evsel__name(evsel));
+
+			nr = evsel->nr_members - 1;
+		} else if (nr) {
+			fprintf(fp, ",%s", perf_evsel__name(evsel));
+
+			if (--nr == 0)
+				fprintf(fp, "}\n");
+		}
+	}
+}
+
+static int __event_process_build_id(struct build_id_event *bev,
+				    char *filename,
+				    struct perf_session *session)
+{
+	int err = -1;
+	struct dsos *dsos;
+	struct machine *machine;
+	u16 misc;
+	struct dso *dso;
+	enum dso_kernel_type dso_type;
+
+	machine = perf_session__findnew_machine(session, bev->pid);
+	if (!machine)
+		goto out;
+
+	misc = bev->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+
+	switch (misc) {
+	case PERF_RECORD_MISC_KERNEL:
+		dso_type = DSO_TYPE_KERNEL;
+		dsos = &machine->kernel_dsos;
+		break;
+	case PERF_RECORD_MISC_GUEST_KERNEL:
+		dso_type = DSO_TYPE_GUEST_KERNEL;
+		dsos = &machine->kernel_dsos;
+		break;
+	case PERF_RECORD_MISC_USER:
+	case PERF_RECORD_MISC_GUEST_USER:
+		dso_type = DSO_TYPE_USER;
+		dsos = &machine->user_dsos;
+		break;
+	default:
+		goto out;
+	}
+
+	dso = __dsos__findnew(dsos, filename);
+	if (dso != NULL) {
+		char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+		dso__set_build_id(dso, &bev->build_id);
+
+		if (!is_kernel_module(filename))
+			dso->kernel = dso_type;
+
+		build_id__sprintf(dso->build_id, sizeof(dso->build_id),
+				  sbuild_id);
+		pr_debug("build id event received for %s: %s\n",
+			 dso->long_name, sbuild_id);
+	}
+
+	err = 0;
+out:
+	return err;
+}
+
+static int perf_header__read_build_ids_abi_quirk(struct perf_header *header,
+						 int input, u64 offset, u64 size)
+{
+	struct perf_session *session = container_of(header, struct perf_session, header);
+	struct {
+		struct perf_event_header   header;
+		u8			   build_id[PERF_ALIGN(BUILD_ID_SIZE, sizeof(u64))];
+		char			   filename[0];
+	} old_bev;
+	struct build_id_event bev;
+	char filename[PATH_MAX];
+	u64 limit = offset + size;
+
+	while (offset < limit) {
+		ssize_t len;
+
+		if (readn(input, &old_bev, sizeof(old_bev)) != sizeof(old_bev))
+			return -1;
+
+		if (header->needs_swap)
+			perf_event_header__bswap(&old_bev.header);
+
+		len = old_bev.header.size - sizeof(old_bev);
+		if (readn(input, filename, len) != len)
+			return -1;
+
+		bev.header = old_bev.header;
+
+		/*
+		 * As the pid is the missing value, we need to fill
+		 * it properly. The header.misc value give us nice hint.
+		 */
+		bev.pid	= HOST_KERNEL_ID;
+		if (bev.header.misc == PERF_RECORD_MISC_GUEST_USER ||
+		    bev.header.misc == PERF_RECORD_MISC_GUEST_KERNEL)
+			bev.pid	= DEFAULT_GUEST_KERNEL_ID;
+
+		memcpy(bev.build_id, old_bev.build_id, sizeof(bev.build_id));
+		__event_process_build_id(&bev, filename, session);
+
+		offset += bev.header.size;
+	}
+
+	return 0;
+}
+
+static int perf_header__read_build_ids(struct perf_header *header,
+				       int input, u64 offset, u64 size)
+{
+	struct perf_session *session = container_of(header, struct perf_session, header);
+	struct build_id_event bev;
+	char filename[PATH_MAX];
+	u64 limit = offset + size, orig_offset = offset;
+	int err = -1;
+
+	while (offset < limit) {
+		ssize_t len;
+
+		if (readn(input, &bev, sizeof(bev)) != sizeof(bev))
+			goto out;
+
+		if (header->needs_swap)
+			perf_event_header__bswap(&bev.header);
+
+		len = bev.header.size - sizeof(bev);
+		if (readn(input, filename, len) != len)
+			goto out;
+		/*
+		 * The a1645ce1 changeset:
+		 *
+		 * "perf: 'perf kvm' tool for monitoring guest performance from host"
+		 *
+		 * Added a field to struct build_id_event that broke the file
+		 * format.
+		 *
+		 * Since the kernel build-id is the first entry, process the
+		 * table using the old format if the well known
+		 * '[kernel.kallsyms]' string for the kernel build-id has the
+		 * first 4 characters chopped off (where the pid_t sits).
+		 */
+		if (memcmp(filename, "nel.kallsyms]", 13) == 0) {
+			if (lseek(input, orig_offset, SEEK_SET) == (off_t)-1)
+				return -1;
+			return perf_header__read_build_ids_abi_quirk(header, input, offset, size);
+		}
+
+		__event_process_build_id(&bev, filename, session);
+
+		offset += bev.header.size;
+	}
+	err = 0;
+out:
+	return err;
+}
+
+static int process_tracing_data(struct perf_file_section *section __maybe_unused,
+				struct perf_header *ph __maybe_unused,
+				int fd, void *data)
+{
+	ssize_t ret = trace_report(fd, data, false);
+	return ret < 0 ? -1 : 0;
+}
+
+static int process_build_id(struct perf_file_section *section,
+			    struct perf_header *ph, int fd,
+			    void *data __maybe_unused)
+{
+	if (perf_header__read_build_ids(ph, fd, section->offset, section->size))
+		pr_debug("Failed to read buildids, continuing...\n");
+	return 0;
+}
+
+static int process_hostname(struct perf_file_section *section __maybe_unused,
+			    struct perf_header *ph, int fd,
+			    void *data __maybe_unused)
+{
+	ph->env.hostname = do_read_string(fd, ph);
+	return ph->env.hostname ? 0 : -ENOMEM;
+}
+
+static int process_osrelease(struct perf_file_section *section __maybe_unused,
+			     struct perf_header *ph, int fd,
+			     void *data __maybe_unused)
+{
+	ph->env.os_release = do_read_string(fd, ph);
+	return ph->env.os_release ? 0 : -ENOMEM;
+}
+
+static int process_version(struct perf_file_section *section __maybe_unused,
+			   struct perf_header *ph, int fd,
+			   void *data __maybe_unused)
+{
+	ph->env.version = do_read_string(fd, ph);
+	return ph->env.version ? 0 : -ENOMEM;
+}
+
+static int process_arch(struct perf_file_section *section __maybe_unused,
+			struct perf_header *ph,	int fd,
+			void *data __maybe_unused)
+{
+	ph->env.arch = do_read_string(fd, ph);
+	return ph->env.arch ? 0 : -ENOMEM;
+}
+
+static int process_nrcpus(struct perf_file_section *section __maybe_unused,
+			  struct perf_header *ph, int fd,
+			  void *data __maybe_unused)
+{
+	ssize_t ret;
+	u32 nr;
+
+	ret = readn(fd, &nr, sizeof(nr));
+	if (ret != sizeof(nr))
+		return -1;
+
+	if (ph->needs_swap)
+		nr = bswap_32(nr);
+
+	ph->env.nr_cpus_online = nr;
+
+	ret = readn(fd, &nr, sizeof(nr));
+	if (ret != sizeof(nr))
+		return -1;
+
+	if (ph->needs_swap)
+		nr = bswap_32(nr);
+
+	ph->env.nr_cpus_avail = nr;
+	return 0;
+}
+
+static int process_cpudesc(struct perf_file_section *section __maybe_unused,
+			   struct perf_header *ph, int fd,
+			   void *data __maybe_unused)
+{
+	ph->env.cpu_desc = do_read_string(fd, ph);
+	return ph->env.cpu_desc ? 0 : -ENOMEM;
+}
+
+static int process_cpuid(struct perf_file_section *section __maybe_unused,
+			 struct perf_header *ph,  int fd,
+			 void *data __maybe_unused)
+{
+	ph->env.cpuid = do_read_string(fd, ph);
+	return ph->env.cpuid ? 0 : -ENOMEM;
+}
+
+static int process_total_mem(struct perf_file_section *section __maybe_unused,
+			     struct perf_header *ph, int fd,
+			     void *data __maybe_unused)
+{
+	uint64_t mem;
+	ssize_t ret;
+
+	ret = readn(fd, &mem, sizeof(mem));
+	if (ret != sizeof(mem))
+		return -1;
+
+	if (ph->needs_swap)
+		mem = bswap_64(mem);
+
+	ph->env.total_mem = mem;
+	return 0;
+}
+
+static struct perf_evsel *
+perf_evlist__find_by_index(struct perf_evlist *evlist, int idx)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(evlist, evsel) {
+		if (evsel->idx == idx)
+			return evsel;
+	}
+
+	return NULL;
+}
+
+static void
+perf_evlist__set_event_name(struct perf_evlist *evlist,
+			    struct perf_evsel *event)
+{
+	struct perf_evsel *evsel;
+
+	if (!event->name)
+		return;
+
+	evsel = perf_evlist__find_by_index(evlist, event->idx);
+	if (!evsel)
+		return;
+
+	if (evsel->name)
+		return;
+
+	evsel->name = strdup(event->name);
+}
+
+static int
+process_event_desc(struct perf_file_section *section __maybe_unused,
+		   struct perf_header *header, int fd,
+		   void *data __maybe_unused)
+{
+	struct perf_session *session;
+	struct perf_evsel *evsel, *events = read_event_desc(header, fd);
+
+	if (!events)
+		return 0;
+
+	session = container_of(header, struct perf_session, header);
+	for (evsel = events; evsel->attr.size; evsel++)
+		perf_evlist__set_event_name(session->evlist, evsel);
+
+	free_event_desc(events);
+
+	return 0;
+}
+
+static int process_cmdline(struct perf_file_section *section __maybe_unused,
+			   struct perf_header *ph, int fd,
+			   void *data __maybe_unused)
+{
+	ssize_t ret;
+	char *str;
+	u32 nr, i;
+	struct strbuf sb;
+
+	ret = readn(fd, &nr, sizeof(nr));
+	if (ret != sizeof(nr))
+		return -1;
+
+	if (ph->needs_swap)
+		nr = bswap_32(nr);
+
+	ph->env.nr_cmdline = nr;
+	strbuf_init(&sb, 128);
+
+	for (i = 0; i < nr; i++) {
+		str = do_read_string(fd, ph);
+		if (!str)
+			goto error;
+
+		/* include a NULL character at the end */
+		strbuf_add(&sb, str, strlen(str) + 1);
+		free(str);
+	}
+	ph->env.cmdline = strbuf_detach(&sb, NULL);
+	return 0;
+
+error:
+	strbuf_release(&sb);
+	return -1;
+}
+
+static int process_cpu_topology(struct perf_file_section *section __maybe_unused,
+				struct perf_header *ph, int fd,
+				void *data __maybe_unused)
+{
+	ssize_t ret;
+	u32 nr, i;
+	char *str;
+	struct strbuf sb;
+
+	ret = readn(fd, &nr, sizeof(nr));
+	if (ret != sizeof(nr))
+		return -1;
+
+	if (ph->needs_swap)
+		nr = bswap_32(nr);
+
+	ph->env.nr_sibling_cores = nr;
+	strbuf_init(&sb, 128);
+
+	for (i = 0; i < nr; i++) {
+		str = do_read_string(fd, ph);
+		if (!str)
+			goto error;
+
+		/* include a NULL character at the end */
+		strbuf_add(&sb, str, strlen(str) + 1);
+		free(str);
+	}
+	ph->env.sibling_cores = strbuf_detach(&sb, NULL);
+
+	ret = readn(fd, &nr, sizeof(nr));
+	if (ret != sizeof(nr))
+		return -1;
+
+	if (ph->needs_swap)
+		nr = bswap_32(nr);
+
+	ph->env.nr_sibling_threads = nr;
+
+	for (i = 0; i < nr; i++) {
+		str = do_read_string(fd, ph);
+		if (!str)
+			goto error;
+
+		/* include a NULL character at the end */
+		strbuf_add(&sb, str, strlen(str) + 1);
+		free(str);
+	}
+	ph->env.sibling_threads = strbuf_detach(&sb, NULL);
+	return 0;
+
+error:
+	strbuf_release(&sb);
+	return -1;
+}
+
+static int process_numa_topology(struct perf_file_section *section __maybe_unused,
+				 struct perf_header *ph, int fd,
+				 void *data __maybe_unused)
+{
+	ssize_t ret;
+	u32 nr, node, i;
+	char *str;
+	uint64_t mem_total, mem_free;
+	struct strbuf sb;
+
+	/* nr nodes */
+	ret = readn(fd, &nr, sizeof(nr));
+	if (ret != sizeof(nr))
+		goto error;
+
+	if (ph->needs_swap)
+		nr = bswap_32(nr);
+
+	ph->env.nr_numa_nodes = nr;
+	strbuf_init(&sb, 256);
+
+	for (i = 0; i < nr; i++) {
+		/* node number */
+		ret = readn(fd, &node, sizeof(node));
+		if (ret != sizeof(node))
+			goto error;
+
+		ret = readn(fd, &mem_total, sizeof(u64));
+		if (ret != sizeof(u64))
+			goto error;
+
+		ret = readn(fd, &mem_free, sizeof(u64));
+		if (ret != sizeof(u64))
+			goto error;
+
+		if (ph->needs_swap) {
+			node = bswap_32(node);
+			mem_total = bswap_64(mem_total);
+			mem_free = bswap_64(mem_free);
+		}
+
+		strbuf_addf(&sb, "%u:%"PRIu64":%"PRIu64":",
+			    node, mem_total, mem_free);
+
+		str = do_read_string(fd, ph);
+		if (!str)
+			goto error;
+
+		/* include a NULL character at the end */
+		strbuf_add(&sb, str, strlen(str) + 1);
+		free(str);
+	}
+	ph->env.numa_nodes = strbuf_detach(&sb, NULL);
+	return 0;
+
+error:
+	strbuf_release(&sb);
+	return -1;
+}
+
+static int process_pmu_mappings(struct perf_file_section *section __maybe_unused,
+				struct perf_header *ph, int fd,
+				void *data __maybe_unused)
+{
+	ssize_t ret;
+	char *name;
+	u32 pmu_num;
+	u32 type;
+	struct strbuf sb;
+
+	ret = readn(fd, &pmu_num, sizeof(pmu_num));
+	if (ret != sizeof(pmu_num))
+		return -1;
+
+	if (ph->needs_swap)
+		pmu_num = bswap_32(pmu_num);
+
+	if (!pmu_num) {
+		pr_debug("pmu mappings not available\n");
+		return 0;
+	}
+
+	ph->env.nr_pmu_mappings = pmu_num;
+	strbuf_init(&sb, 128);
+
+	while (pmu_num) {
+		if (readn(fd, &type, sizeof(type)) != sizeof(type))
+			goto error;
+		if (ph->needs_swap)
+			type = bswap_32(type);
+
+		name = do_read_string(fd, ph);
+		if (!name)
+			goto error;
+
+		strbuf_addf(&sb, "%u:%s", type, name);
+		/* include a NULL character at the end */
+		strbuf_add(&sb, "", 1);
+
+		free(name);
+		pmu_num--;
+	}
+	ph->env.pmu_mappings = strbuf_detach(&sb, NULL);
+	return 0;
+
+error:
+	strbuf_release(&sb);
+	return -1;
+}
+
+static int process_group_desc(struct perf_file_section *section __maybe_unused,
+			      struct perf_header *ph, int fd,
+			      void *data __maybe_unused)
+{
+	size_t ret = -1;
+	u32 i, nr, nr_groups;
+	struct perf_session *session;
+	struct perf_evsel *evsel, *leader = NULL;
+	struct group_desc {
+		char *name;
+		u32 leader_idx;
+		u32 nr_members;
+	} *desc;
+
+	if (readn(fd, &nr_groups, sizeof(nr_groups)) != sizeof(nr_groups))
+		return -1;
+
+	if (ph->needs_swap)
+		nr_groups = bswap_32(nr_groups);
+
+	ph->env.nr_groups = nr_groups;
+	if (!nr_groups) {
+		pr_debug("group desc not available\n");
+		return 0;
+	}
+
+	desc = calloc(nr_groups, sizeof(*desc));
+	if (!desc)
+		return -1;
+
+	for (i = 0; i < nr_groups; i++) {
+		desc[i].name = do_read_string(fd, ph);
+		if (!desc[i].name)
+			goto out_free;
+
+		if (readn(fd, &desc[i].leader_idx, sizeof(u32)) != sizeof(u32))
+			goto out_free;
+
+		if (readn(fd, &desc[i].nr_members, sizeof(u32)) != sizeof(u32))
+			goto out_free;
+
+		if (ph->needs_swap) {
+			desc[i].leader_idx = bswap_32(desc[i].leader_idx);
+			desc[i].nr_members = bswap_32(desc[i].nr_members);
+		}
+	}
+
+	/*
+	 * Rebuild group relationship based on the group_desc
+	 */
+	session = container_of(ph, struct perf_session, header);
+	session->evlist->nr_groups = nr_groups;
+
+	i = nr = 0;
+	evlist__for_each(session->evlist, evsel) {
+		if (evsel->idx == (int) desc[i].leader_idx) {
+			evsel->leader = evsel;
+			/* {anon_group} is a dummy name */
+			if (strcmp(desc[i].name, "{anon_group}")) {
+				evsel->group_name = desc[i].name;
+				desc[i].name = NULL;
+			}
+			evsel->nr_members = desc[i].nr_members;
+
+			if (i >= nr_groups || nr > 0) {
+				pr_debug("invalid group desc\n");
+				goto out_free;
+			}
+
+			leader = evsel;
+			nr = evsel->nr_members - 1;
+			i++;
+		} else if (nr) {
+			/* This is a group member */
+			evsel->leader = leader;
+
+			nr--;
+		}
+	}
+
+	if (i != nr_groups || nr != 0) {
+		pr_debug("invalid group desc\n");
+		goto out_free;
+	}
+
+	ret = 0;
+out_free:
+	for (i = 0; i < nr_groups; i++)
+		zfree(&desc[i].name);
+	free(desc);
+
+	return ret;
+}
+
+struct feature_ops {
+	int (*write)(int fd, struct perf_header *h, struct perf_evlist *evlist);
+	void (*print)(struct perf_header *h, int fd, FILE *fp);
+	int (*process)(struct perf_file_section *section,
+		       struct perf_header *h, int fd, void *data);
+	const char *name;
+	bool full_only;
+};
+
+#define FEAT_OPA(n, func) \
+	[n] = { .name = #n, .write = write_##func, .print = print_##func }
+#define FEAT_OPP(n, func) \
+	[n] = { .name = #n, .write = write_##func, .print = print_##func, \
+		.process = process_##func }
+#define FEAT_OPF(n, func) \
+	[n] = { .name = #n, .write = write_##func, .print = print_##func, \
+		.process = process_##func, .full_only = true }
+
+/* feature_ops not implemented: */
+#define print_tracing_data	NULL
+#define print_build_id		NULL
+
+static const struct feature_ops feat_ops[HEADER_LAST_FEATURE] = {
+	FEAT_OPP(HEADER_TRACING_DATA,	tracing_data),
+	FEAT_OPP(HEADER_BUILD_ID,	build_id),
+	FEAT_OPP(HEADER_HOSTNAME,	hostname),
+	FEAT_OPP(HEADER_OSRELEASE,	osrelease),
+	FEAT_OPP(HEADER_VERSION,	version),
+	FEAT_OPP(HEADER_ARCH,		arch),
+	FEAT_OPP(HEADER_NRCPUS,		nrcpus),
+	FEAT_OPP(HEADER_CPUDESC,	cpudesc),
+	FEAT_OPP(HEADER_CPUID,		cpuid),
+	FEAT_OPP(HEADER_TOTAL_MEM,	total_mem),
+	FEAT_OPP(HEADER_EVENT_DESC,	event_desc),
+	FEAT_OPP(HEADER_CMDLINE,	cmdline),
+	FEAT_OPF(HEADER_CPU_TOPOLOGY,	cpu_topology),
+	FEAT_OPF(HEADER_NUMA_TOPOLOGY,	numa_topology),
+	FEAT_OPA(HEADER_BRANCH_STACK,	branch_stack),
+	FEAT_OPP(HEADER_PMU_MAPPINGS,	pmu_mappings),
+	FEAT_OPP(HEADER_GROUP_DESC,	group_desc),
+};
+
+struct header_print_data {
+	FILE *fp;
+	bool full; /* extended list of headers */
+};
+
+static int perf_file_section__fprintf_info(struct perf_file_section *section,
+					   struct perf_header *ph,
+					   int feat, int fd, void *data)
+{
+	struct header_print_data *hd = data;
+
+	if (lseek(fd, section->offset, SEEK_SET) == (off_t)-1) {
+		pr_debug("Failed to lseek to %" PRIu64 " offset for feature "
+				"%d, continuing...\n", section->offset, feat);
+		return 0;
+	}
+	if (feat >= HEADER_LAST_FEATURE) {
+		pr_warning("unknown feature %d\n", feat);
+		return 0;
+	}
+	if (!feat_ops[feat].print)
+		return 0;
+
+	if (!feat_ops[feat].full_only || hd->full)
+		feat_ops[feat].print(ph, fd, hd->fp);
+	else
+		fprintf(hd->fp, "# %s info available, use -I to display\n",
+			feat_ops[feat].name);
+
+	return 0;
+}
+
+int perf_header__fprintf_info(struct perf_session *session, FILE *fp, bool full)
+{
+	struct header_print_data hd;
+	struct perf_header *header = &session->header;
+	int fd = perf_data_file__fd(session->file);
+	hd.fp = fp;
+	hd.full = full;
+
+	perf_header__process_sections(header, fd, &hd,
+				      perf_file_section__fprintf_info);
+	return 0;
+}
+
+static int do_write_feat(int fd, struct perf_header *h, int type,
+			 struct perf_file_section **p,
+			 struct perf_evlist *evlist)
+{
+	int err;
+	int ret = 0;
+
+	if (perf_header__has_feat(h, type)) {
+		if (!feat_ops[type].write)
+			return -1;
+
+		(*p)->offset = lseek(fd, 0, SEEK_CUR);
+
+		err = feat_ops[type].write(fd, h, evlist);
+		if (err < 0) {
+			pr_debug("failed to write feature %d\n", type);
+
+			/* undo anything written */
+			lseek(fd, (*p)->offset, SEEK_SET);
+
+			return -1;
+		}
+		(*p)->size = lseek(fd, 0, SEEK_CUR) - (*p)->offset;
+		(*p)++;
+	}
+	return ret;
+}
+
+static int perf_header__adds_write(struct perf_header *header,
+				   struct perf_evlist *evlist, int fd)
+{
+	int nr_sections;
+	struct perf_file_section *feat_sec, *p;
+	int sec_size;
+	u64 sec_start;
+	int feat;
+	int err;
+
+	nr_sections = bitmap_weight(header->adds_features, HEADER_FEAT_BITS);
+	if (!nr_sections)
+		return 0;
+
+	feat_sec = p = calloc(nr_sections, sizeof(*feat_sec));
+	if (feat_sec == NULL)
+		return -ENOMEM;
+
+	sec_size = sizeof(*feat_sec) * nr_sections;
+
+	sec_start = header->feat_offset;
+	lseek(fd, sec_start + sec_size, SEEK_SET);
+
+	for_each_set_bit(feat, header->adds_features, HEADER_FEAT_BITS) {
+		if (do_write_feat(fd, header, feat, &p, evlist))
+			perf_header__clear_feat(header, feat);
+	}
+
+	lseek(fd, sec_start, SEEK_SET);
+	/*
+	 * may write more than needed due to dropped feature, but
+	 * this is okay, reader will skip the mising entries
+	 */
+	err = do_write(fd, feat_sec, sec_size);
+	if (err < 0)
+		pr_debug("failed to write feature section\n");
+	free(feat_sec);
+	return err;
+}
+
+int perf_header__write_pipe(int fd)
+{
+	struct perf_pipe_file_header f_header;
+	int err;
+
+	f_header = (struct perf_pipe_file_header){
+		.magic	   = PERF_MAGIC,
+		.size	   = sizeof(f_header),
+	};
+
+	err = do_write(fd, &f_header, sizeof(f_header));
+	if (err < 0) {
+		pr_debug("failed to write perf pipe header\n");
+		return err;
+	}
+
+	return 0;
+}
+
+int perf_session__write_header(struct perf_session *session,
+			       struct perf_evlist *evlist,
+			       int fd, bool at_exit)
+{
+	struct perf_file_header f_header;
+	struct perf_file_attr   f_attr;
+	struct perf_header *header = &session->header;
+	struct perf_evsel *evsel;
+	u64 attr_offset;
+	int err;
+
+	lseek(fd, sizeof(f_header), SEEK_SET);
+
+	evlist__for_each(session->evlist, evsel) {
+		evsel->id_offset = lseek(fd, 0, SEEK_CUR);
+		err = do_write(fd, evsel->id, evsel->ids * sizeof(u64));
+		if (err < 0) {
+			pr_debug("failed to write perf header\n");
+			return err;
+		}
+	}
+
+	attr_offset = lseek(fd, 0, SEEK_CUR);
+
+	evlist__for_each(evlist, evsel) {
+		f_attr = (struct perf_file_attr){
+			.attr = evsel->attr,
+			.ids  = {
+				.offset = evsel->id_offset,
+				.size   = evsel->ids * sizeof(u64),
+			}
+		};
+		err = do_write(fd, &f_attr, sizeof(f_attr));
+		if (err < 0) {
+			pr_debug("failed to write perf header attribute\n");
+			return err;
+		}
+	}
+
+	if (!header->data_offset)
+		header->data_offset = lseek(fd, 0, SEEK_CUR);
+	header->feat_offset = header->data_offset + header->data_size;
+
+	if (at_exit) {
+		err = perf_header__adds_write(header, evlist, fd);
+		if (err < 0)
+			return err;
+	}
+
+	f_header = (struct perf_file_header){
+		.magic	   = PERF_MAGIC,
+		.size	   = sizeof(f_header),
+		.attr_size = sizeof(f_attr),
+		.attrs = {
+			.offset = attr_offset,
+			.size   = evlist->nr_entries * sizeof(f_attr),
+		},
+		.data = {
+			.offset = header->data_offset,
+			.size	= header->data_size,
+		},
+		/* event_types is ignored, store zeros */
+	};
+
+	memcpy(&f_header.adds_features, &header->adds_features, sizeof(header->adds_features));
+
+	lseek(fd, 0, SEEK_SET);
+	err = do_write(fd, &f_header, sizeof(f_header));
+	if (err < 0) {
+		pr_debug("failed to write perf header\n");
+		return err;
+	}
+	lseek(fd, header->data_offset + header->data_size, SEEK_SET);
+
+	return 0;
+}
+
+static int perf_header__getbuffer64(struct perf_header *header,
+				    int fd, void *buf, size_t size)
+{
+	if (readn(fd, buf, size) <= 0)
+		return -1;
+
+	if (header->needs_swap)
+		mem_bswap_64(buf, size);
+
+	return 0;
+}
+
+int perf_header__process_sections(struct perf_header *header, int fd,
+				  void *data,
+				  int (*process)(struct perf_file_section *section,
+						 struct perf_header *ph,
+						 int feat, int fd, void *data))
+{
+	struct perf_file_section *feat_sec, *sec;
+	int nr_sections;
+	int sec_size;
+	int feat;
+	int err;
+
+	nr_sections = bitmap_weight(header->adds_features, HEADER_FEAT_BITS);
+	if (!nr_sections)
+		return 0;
+
+	feat_sec = sec = calloc(nr_sections, sizeof(*feat_sec));
+	if (!feat_sec)
+		return -1;
+
+	sec_size = sizeof(*feat_sec) * nr_sections;
+
+	lseek(fd, header->feat_offset, SEEK_SET);
+
+	err = perf_header__getbuffer64(header, fd, feat_sec, sec_size);
+	if (err < 0)
+		goto out_free;
+
+	for_each_set_bit(feat, header->adds_features, HEADER_LAST_FEATURE) {
+		err = process(sec++, header, feat, fd, data);
+		if (err < 0)
+			goto out_free;
+	}
+	err = 0;
+out_free:
+	free(feat_sec);
+	return err;
+}
+
+static const int attr_file_abi_sizes[] = {
+	[0] = PERF_ATTR_SIZE_VER0,
+	[1] = PERF_ATTR_SIZE_VER1,
+	[2] = PERF_ATTR_SIZE_VER2,
+	[3] = PERF_ATTR_SIZE_VER3,
+	[4] = PERF_ATTR_SIZE_VER4,
+	0,
+};
+
+/*
+ * In the legacy file format, the magic number is not used to encode endianness.
+ * hdr_sz was used to encode endianness. But given that hdr_sz can vary based
+ * on ABI revisions, we need to try all combinations for all endianness to
+ * detect the endianness.
+ */
+static int try_all_file_abis(uint64_t hdr_sz, struct perf_header *ph)
+{
+	uint64_t ref_size, attr_size;
+	int i;
+
+	for (i = 0 ; attr_file_abi_sizes[i]; i++) {
+		ref_size = attr_file_abi_sizes[i]
+			 + sizeof(struct perf_file_section);
+		if (hdr_sz != ref_size) {
+			attr_size = bswap_64(hdr_sz);
+			if (attr_size != ref_size)
+				continue;
+
+			ph->needs_swap = true;
+		}
+		pr_debug("ABI%d perf.data file detected, need_swap=%d\n",
+			 i,
+			 ph->needs_swap);
+		return 0;
+	}
+	/* could not determine endianness */
+	return -1;
+}
+
+#define PERF_PIPE_HDR_VER0	16
+
+static const size_t attr_pipe_abi_sizes[] = {
+	[0] = PERF_PIPE_HDR_VER0,
+	0,
+};
+
+/*
+ * In the legacy pipe format, there is an implicit assumption that endiannesss
+ * between host recording the samples, and host parsing the samples is the
+ * same. This is not always the case given that the pipe output may always be
+ * redirected into a file and analyzed on a different machine with possibly a
+ * different endianness and perf_event ABI revsions in the perf tool itself.
+ */
+static int try_all_pipe_abis(uint64_t hdr_sz, struct perf_header *ph)
+{
+	u64 attr_size;
+	int i;
+
+	for (i = 0 ; attr_pipe_abi_sizes[i]; i++) {
+		if (hdr_sz != attr_pipe_abi_sizes[i]) {
+			attr_size = bswap_64(hdr_sz);
+			if (attr_size != hdr_sz)
+				continue;
+
+			ph->needs_swap = true;
+		}
+		pr_debug("Pipe ABI%d perf.data file detected\n", i);
+		return 0;
+	}
+	return -1;
+}
+
+bool is_perf_magic(u64 magic)
+{
+	if (!memcmp(&magic, __perf_magic1, sizeof(magic))
+		|| magic == __perf_magic2
+		|| magic == __perf_magic2_sw)
+		return true;
+
+	return false;
+}
+
+static int check_magic_endian(u64 magic, uint64_t hdr_sz,
+			      bool is_pipe, struct perf_header *ph)
+{
+	int ret;
+
+	/* check for legacy format */
+	ret = memcmp(&magic, __perf_magic1, sizeof(magic));
+	if (ret == 0) {
+		ph->version = PERF_HEADER_VERSION_1;
+		pr_debug("legacy perf.data format\n");
+		if (is_pipe)
+			return try_all_pipe_abis(hdr_sz, ph);
+
+		return try_all_file_abis(hdr_sz, ph);
+	}
+	/*
+	 * the new magic number serves two purposes:
+	 * - unique number to identify actual perf.data files
+	 * - encode endianness of file
+	 */
+	ph->version = PERF_HEADER_VERSION_2;
+
+	/* check magic number with one endianness */
+	if (magic == __perf_magic2)
+		return 0;
+
+	/* check magic number with opposite endianness */
+	if (magic != __perf_magic2_sw)
+		return -1;
+
+	ph->needs_swap = true;
+
+	return 0;
+}
+
+int perf_file_header__read(struct perf_file_header *header,
+			   struct perf_header *ph, int fd)
+{
+	ssize_t ret;
+
+	lseek(fd, 0, SEEK_SET);
+
+	ret = readn(fd, header, sizeof(*header));
+	if (ret <= 0)
+		return -1;
+
+	if (check_magic_endian(header->magic,
+			       header->attr_size, false, ph) < 0) {
+		pr_debug("magic/endian check failed\n");
+		return -1;
+	}
+
+	if (ph->needs_swap) {
+		mem_bswap_64(header, offsetof(struct perf_file_header,
+			     adds_features));
+	}
+
+	if (header->size != sizeof(*header)) {
+		/* Support the previous format */
+		if (header->size == offsetof(typeof(*header), adds_features))
+			bitmap_zero(header->adds_features, HEADER_FEAT_BITS);
+		else
+			return -1;
+	} else if (ph->needs_swap) {
+		/*
+		 * feature bitmap is declared as an array of unsigned longs --
+		 * not good since its size can differ between the host that
+		 * generated the data file and the host analyzing the file.
+		 *
+		 * We need to handle endianness, but we don't know the size of
+		 * the unsigned long where the file was generated. Take a best
+		 * guess at determining it: try 64-bit swap first (ie., file
+		 * created on a 64-bit host), and check if the hostname feature
+		 * bit is set (this feature bit is forced on as of fbe96f2).
+		 * If the bit is not, undo the 64-bit swap and try a 32-bit
+		 * swap. If the hostname bit is still not set (e.g., older data
+		 * file), punt and fallback to the original behavior --
+		 * clearing all feature bits and setting buildid.
+		 */
+		mem_bswap_64(&header->adds_features,
+			    BITS_TO_U64(HEADER_FEAT_BITS));
+
+		if (!test_bit(HEADER_HOSTNAME, header->adds_features)) {
+			/* unswap as u64 */
+			mem_bswap_64(&header->adds_features,
+				    BITS_TO_U64(HEADER_FEAT_BITS));
+
+			/* unswap as u32 */
+			mem_bswap_32(&header->adds_features,
+				    BITS_TO_U32(HEADER_FEAT_BITS));
+		}
+
+		if (!test_bit(HEADER_HOSTNAME, header->adds_features)) {
+			bitmap_zero(header->adds_features, HEADER_FEAT_BITS);
+			set_bit(HEADER_BUILD_ID, header->adds_features);
+		}
+	}
+
+	memcpy(&ph->adds_features, &header->adds_features,
+	       sizeof(ph->adds_features));
+
+	ph->data_offset  = header->data.offset;
+	ph->data_size	 = header->data.size;
+	ph->feat_offset  = header->data.offset + header->data.size;
+	return 0;
+}
+
+static int perf_file_section__process(struct perf_file_section *section,
+				      struct perf_header *ph,
+				      int feat, int fd, void *data)
+{
+	if (lseek(fd, section->offset, SEEK_SET) == (off_t)-1) {
+		pr_debug("Failed to lseek to %" PRIu64 " offset for feature "
+			  "%d, continuing...\n", section->offset, feat);
+		return 0;
+	}
+
+	if (feat >= HEADER_LAST_FEATURE) {
+		pr_debug("unknown feature %d, continuing...\n", feat);
+		return 0;
+	}
+
+	if (!feat_ops[feat].process)
+		return 0;
+
+	return feat_ops[feat].process(section, ph, fd, data);
+}
+
+static int perf_file_header__read_pipe(struct perf_pipe_file_header *header,
+				       struct perf_header *ph, int fd,
+				       bool repipe)
+{
+	ssize_t ret;
+
+	ret = readn(fd, header, sizeof(*header));
+	if (ret <= 0)
+		return -1;
+
+	if (check_magic_endian(header->magic, header->size, true, ph) < 0) {
+		pr_debug("endian/magic failed\n");
+		return -1;
+	}
+
+	if (ph->needs_swap)
+		header->size = bswap_64(header->size);
+
+	if (repipe && do_write(STDOUT_FILENO, header, sizeof(*header)) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int perf_header__read_pipe(struct perf_session *session)
+{
+	struct perf_header *header = &session->header;
+	struct perf_pipe_file_header f_header;
+
+	if (perf_file_header__read_pipe(&f_header, header,
+					perf_data_file__fd(session->file),
+					session->repipe) < 0) {
+		pr_debug("incompatible file format\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int read_attr(int fd, struct perf_header *ph,
+		     struct perf_file_attr *f_attr)
+{
+	struct perf_event_attr *attr = &f_attr->attr;
+	size_t sz, left;
+	size_t our_sz = sizeof(f_attr->attr);
+	ssize_t ret;
+
+	memset(f_attr, 0, sizeof(*f_attr));
+
+	/* read minimal guaranteed structure */
+	ret = readn(fd, attr, PERF_ATTR_SIZE_VER0);
+	if (ret <= 0) {
+		pr_debug("cannot read %d bytes of header attr\n",
+			 PERF_ATTR_SIZE_VER0);
+		return -1;
+	}
+
+	/* on file perf_event_attr size */
+	sz = attr->size;
+
+	if (ph->needs_swap)
+		sz = bswap_32(sz);
+
+	if (sz == 0) {
+		/* assume ABI0 */
+		sz =  PERF_ATTR_SIZE_VER0;
+	} else if (sz > our_sz) {
+		pr_debug("file uses a more recent and unsupported ABI"
+			 " (%zu bytes extra)\n", sz - our_sz);
+		return -1;
+	}
+	/* what we have not yet read and that we know about */
+	left = sz - PERF_ATTR_SIZE_VER0;
+	if (left) {
+		void *ptr = attr;
+		ptr += PERF_ATTR_SIZE_VER0;
+
+		ret = readn(fd, ptr, left);
+	}
+	/* read perf_file_section, ids are read in caller */
+	ret = readn(fd, &f_attr->ids, sizeof(f_attr->ids));
+
+	return ret <= 0 ? -1 : 0;
+}
+
+static int perf_evsel__prepare_tracepoint_event(struct perf_evsel *evsel,
+						struct pevent *pevent)
+{
+	struct event_format *event;
+	char bf[128];
+
+	/* already prepared */
+	if (evsel->tp_format)
+		return 0;
+
+	if (pevent == NULL) {
+		pr_debug("broken or missing trace data\n");
+		return -1;
+	}
+
+	event = pevent_find_event(pevent, evsel->attr.config);
+	if (event == NULL)
+		return -1;
+
+	if (!evsel->name) {
+		snprintf(bf, sizeof(bf), "%s:%s", event->system, event->name);
+		evsel->name = strdup(bf);
+		if (evsel->name == NULL)
+			return -1;
+	}
+
+	evsel->tp_format = event;
+	return 0;
+}
+
+static int perf_evlist__prepare_tracepoint_events(struct perf_evlist *evlist,
+						  struct pevent *pevent)
+{
+	struct perf_evsel *pos;
+
+	evlist__for_each(evlist, pos) {
+		if (pos->attr.type == PERF_TYPE_TRACEPOINT &&
+		    perf_evsel__prepare_tracepoint_event(pos, pevent))
+			return -1;
+	}
+
+	return 0;
+}
+
+int perf_session__read_header(struct perf_session *session)
+{
+	struct perf_data_file *file = session->file;
+	struct perf_header *header = &session->header;
+	struct perf_file_header	f_header;
+	struct perf_file_attr	f_attr;
+	u64			f_id;
+	int nr_attrs, nr_ids, i, j;
+	int fd = perf_data_file__fd(file);
+
+	session->evlist = perf_evlist__new();
+	if (session->evlist == NULL)
+		return -ENOMEM;
+
+	if (perf_data_file__is_pipe(file))
+		return perf_header__read_pipe(session);
+
+	if (perf_file_header__read(&f_header, header, fd) < 0)
+		return -EINVAL;
+
+	/*
+	 * Sanity check that perf.data was written cleanly; data size is
+	 * initialized to 0 and updated only if the on_exit function is run.
+	 * If data size is still 0 then the file contains only partial
+	 * information.  Just warn user and process it as much as it can.
+	 */
+	if (f_header.data.size == 0) {
+		pr_warning("WARNING: The %s file's data size field is 0 which is unexpected.\n"
+			   "Was the 'perf record' command properly terminated?\n",
+			   file->path);
+	}
+
+	nr_attrs = f_header.attrs.size / f_header.attr_size;
+	lseek(fd, f_header.attrs.offset, SEEK_SET);
+
+	for (i = 0; i < nr_attrs; i++) {
+		struct perf_evsel *evsel;
+		off_t tmp;
+
+		if (read_attr(fd, header, &f_attr) < 0)
+			goto out_errno;
+
+		if (header->needs_swap) {
+			f_attr.ids.size   = bswap_64(f_attr.ids.size);
+			f_attr.ids.offset = bswap_64(f_attr.ids.offset);
+			perf_event__attr_swap(&f_attr.attr);
+		}
+
+		tmp = lseek(fd, 0, SEEK_CUR);
+		evsel = perf_evsel__new(&f_attr.attr);
+
+		if (evsel == NULL)
+			goto out_delete_evlist;
+
+		evsel->needs_swap = header->needs_swap;
+		/*
+		 * Do it before so that if perf_evsel__alloc_id fails, this
+		 * entry gets purged too at perf_evlist__delete().
+		 */
+		perf_evlist__add(session->evlist, evsel);
+
+		nr_ids = f_attr.ids.size / sizeof(u64);
+		/*
+		 * We don't have the cpu and thread maps on the header, so
+		 * for allocating the perf_sample_id table we fake 1 cpu and
+		 * hattr->ids threads.
+		 */
+		if (perf_evsel__alloc_id(evsel, 1, nr_ids))
+			goto out_delete_evlist;
+
+		lseek(fd, f_attr.ids.offset, SEEK_SET);
+
+		for (j = 0; j < nr_ids; j++) {
+			if (perf_header__getbuffer64(header, fd, &f_id, sizeof(f_id)))
+				goto out_errno;
+
+			perf_evlist__id_add(session->evlist, evsel, 0, j, f_id);
+		}
+
+		lseek(fd, tmp, SEEK_SET);
+	}
+
+	symbol_conf.nr_events = nr_attrs;
+
+	perf_header__process_sections(header, fd, &session->tevent,
+				      perf_file_section__process);
+
+	if (perf_evlist__prepare_tracepoint_events(session->evlist,
+						   session->tevent.pevent))
+		goto out_delete_evlist;
+
+	return 0;
+out_errno:
+	return -errno;
+
+out_delete_evlist:
+	perf_evlist__delete(session->evlist);
+	session->evlist = NULL;
+	return -ENOMEM;
+}
+
+int perf_event__synthesize_attr(struct perf_tool *tool,
+				struct perf_event_attr *attr, u32 ids, u64 *id,
+				perf_event__handler_t process)
+{
+	union perf_event *ev;
+	size_t size;
+	int err;
+
+	size = sizeof(struct perf_event_attr);
+	size = PERF_ALIGN(size, sizeof(u64));
+	size += sizeof(struct perf_event_header);
+	size += ids * sizeof(u64);
+
+	ev = malloc(size);
+
+	if (ev == NULL)
+		return -ENOMEM;
+
+	ev->attr.attr = *attr;
+	memcpy(ev->attr.id, id, ids * sizeof(u64));
+
+	ev->attr.header.type = PERF_RECORD_HEADER_ATTR;
+	ev->attr.header.size = (u16)size;
+
+	if (ev->attr.header.size == size)
+		err = process(tool, ev, NULL, NULL);
+	else
+		err = -E2BIG;
+
+	free(ev);
+
+	return err;
+}
+
+int perf_event__synthesize_attrs(struct perf_tool *tool,
+				   struct perf_session *session,
+				   perf_event__handler_t process)
+{
+	struct perf_evsel *evsel;
+	int err = 0;
+
+	evlist__for_each(session->evlist, evsel) {
+		err = perf_event__synthesize_attr(tool, &evsel->attr, evsel->ids,
+						  evsel->id, process);
+		if (err) {
+			pr_debug("failed to create perf header attribute\n");
+			return err;
+		}
+	}
+
+	return err;
+}
+
+int perf_event__process_attr(struct perf_tool *tool __maybe_unused,
+			     union perf_event *event,
+			     struct perf_evlist **pevlist)
+{
+	u32 i, ids, n_ids;
+	struct perf_evsel *evsel;
+	struct perf_evlist *evlist = *pevlist;
+
+	if (evlist == NULL) {
+		*pevlist = evlist = perf_evlist__new();
+		if (evlist == NULL)
+			return -ENOMEM;
+	}
+
+	evsel = perf_evsel__new(&event->attr.attr);
+	if (evsel == NULL)
+		return -ENOMEM;
+
+	perf_evlist__add(evlist, evsel);
+
+	ids = event->header.size;
+	ids -= (void *)&event->attr.id - (void *)event;
+	n_ids = ids / sizeof(u64);
+	/*
+	 * We don't have the cpu and thread maps on the header, so
+	 * for allocating the perf_sample_id table we fake 1 cpu and
+	 * hattr->ids threads.
+	 */
+	if (perf_evsel__alloc_id(evsel, 1, n_ids))
+		return -ENOMEM;
+
+	for (i = 0; i < n_ids; i++) {
+		perf_evlist__id_add(evlist, evsel, 0, i, event->attr.id[i]);
+	}
+
+	symbol_conf.nr_events = evlist->nr_entries;
+
+	return 0;
+}
+
+int perf_event__synthesize_tracing_data(struct perf_tool *tool, int fd,
+					struct perf_evlist *evlist,
+					perf_event__handler_t process)
+{
+	union perf_event ev;
+	struct tracing_data *tdata;
+	ssize_t size = 0, aligned_size = 0, padding;
+	int err __maybe_unused = 0;
+
+	/*
+	 * We are going to store the size of the data followed
+	 * by the data contents. Since the fd descriptor is a pipe,
+	 * we cannot seek back to store the size of the data once
+	 * we know it. Instead we:
+	 *
+	 * - write the tracing data to the temp file
+	 * - get/write the data size to pipe
+	 * - write the tracing data from the temp file
+	 *   to the pipe
+	 */
+	tdata = tracing_data_get(&evlist->entries, fd, true);
+	if (!tdata)
+		return -1;
+
+	memset(&ev, 0, sizeof(ev));
+
+	ev.tracing_data.header.type = PERF_RECORD_HEADER_TRACING_DATA;
+	size = tdata->size;
+	aligned_size = PERF_ALIGN(size, sizeof(u64));
+	padding = aligned_size - size;
+	ev.tracing_data.header.size = sizeof(ev.tracing_data);
+	ev.tracing_data.size = aligned_size;
+
+	process(tool, &ev, NULL, NULL);
+
+	/*
+	 * The put function will copy all the tracing data
+	 * stored in temp file to the pipe.
+	 */
+	tracing_data_put(tdata);
+
+	write_padded(fd, NULL, 0, padding);
+
+	return aligned_size;
+}
+
+int perf_event__process_tracing_data(struct perf_tool *tool __maybe_unused,
+				     union perf_event *event,
+				     struct perf_session *session)
+{
+	ssize_t size_read, padding, size = event->tracing_data.size;
+	int fd = perf_data_file__fd(session->file);
+	off_t offset = lseek(fd, 0, SEEK_CUR);
+	char buf[BUFSIZ];
+
+	/* setup for reading amidst mmap */
+	lseek(fd, offset + sizeof(struct tracing_data_event),
+	      SEEK_SET);
+
+	size_read = trace_report(fd, &session->tevent,
+				 session->repipe);
+	padding = PERF_ALIGN(size_read, sizeof(u64)) - size_read;
+
+	if (readn(fd, buf, padding) < 0) {
+		pr_err("%s: reading input file", __func__);
+		return -1;
+	}
+	if (session->repipe) {
+		int retw = write(STDOUT_FILENO, buf, padding);
+		if (retw <= 0 || retw != padding) {
+			pr_err("%s: repiping tracing data padding", __func__);
+			return -1;
+		}
+	}
+
+	if (size_read + padding != size) {
+		pr_err("%s: tracing data size mismatch", __func__);
+		return -1;
+	}
+
+	perf_evlist__prepare_tracepoint_events(session->evlist,
+					       session->tevent.pevent);
+
+	return size_read + padding;
+}
+
+int perf_event__synthesize_build_id(struct perf_tool *tool,
+				    struct dso *pos, u16 misc,
+				    perf_event__handler_t process,
+				    struct machine *machine)
+{
+	union perf_event ev;
+	size_t len;
+	int err = 0;
+
+	if (!pos->hit)
+		return err;
+
+	memset(&ev, 0, sizeof(ev));
+
+	len = pos->long_name_len + 1;
+	len = PERF_ALIGN(len, NAME_ALIGN);
+	memcpy(&ev.build_id.build_id, pos->build_id, sizeof(pos->build_id));
+	ev.build_id.header.type = PERF_RECORD_HEADER_BUILD_ID;
+	ev.build_id.header.misc = misc;
+	ev.build_id.pid = machine->pid;
+	ev.build_id.header.size = sizeof(ev.build_id) + len;
+	memcpy(&ev.build_id.filename, pos->long_name, pos->long_name_len);
+
+	err = process(tool, &ev, NULL, machine);
+
+	return err;
+}
+
+int perf_event__process_build_id(struct perf_tool *tool __maybe_unused,
+				 union perf_event *event,
+				 struct perf_session *session)
+{
+	__event_process_build_id(&event->build_id,
+				 event->build_id.filename,
+				 session);
+	return 0;
+}
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
new file mode 100644
index 000000000..3bb90ac17
--- /dev/null
+++ b/tools/perf/util/header.h
@@ -0,0 +1,159 @@
+#ifndef __PERF_HEADER_H
+#define __PERF_HEADER_H
+
+#include <linux/perf_event.h>
+#include <sys/types.h>
+#include <stdbool.h>
+#include <linux/bitmap.h>
+#include <linux/types.h>
+#include "event.h"
+
+
+enum {
+	HEADER_RESERVED		= 0,	/* always cleared */
+	HEADER_FIRST_FEATURE	= 1,
+	HEADER_TRACING_DATA	= 1,
+	HEADER_BUILD_ID,
+
+	HEADER_HOSTNAME,
+	HEADER_OSRELEASE,
+	HEADER_VERSION,
+	HEADER_ARCH,
+	HEADER_NRCPUS,
+	HEADER_CPUDESC,
+	HEADER_CPUID,
+	HEADER_TOTAL_MEM,
+	HEADER_CMDLINE,
+	HEADER_EVENT_DESC,
+	HEADER_CPU_TOPOLOGY,
+	HEADER_NUMA_TOPOLOGY,
+	HEADER_BRANCH_STACK,
+	HEADER_PMU_MAPPINGS,
+	HEADER_GROUP_DESC,
+	HEADER_LAST_FEATURE,
+	HEADER_FEAT_BITS	= 256,
+};
+
+enum perf_header_version {
+	PERF_HEADER_VERSION_1,
+	PERF_HEADER_VERSION_2,
+};
+
+struct perf_file_section {
+	u64 offset;
+	u64 size;
+};
+
+struct perf_file_header {
+	u64				magic;
+	u64				size;
+	u64				attr_size;
+	struct perf_file_section	attrs;
+	struct perf_file_section	data;
+	/* event_types is ignored */
+	struct perf_file_section	event_types;
+	DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS);
+};
+
+struct perf_pipe_file_header {
+	u64				magic;
+	u64				size;
+};
+
+struct perf_header;
+
+int perf_file_header__read(struct perf_file_header *header,
+			   struct perf_header *ph, int fd);
+
+struct perf_session_env {
+	char			*hostname;
+	char			*os_release;
+	char			*version;
+	char			*arch;
+	int			nr_cpus_online;
+	int			nr_cpus_avail;
+	char			*cpu_desc;
+	char			*cpuid;
+	unsigned long long	total_mem;
+
+	int			nr_cmdline;
+	int			nr_sibling_cores;
+	int			nr_sibling_threads;
+	int			nr_numa_nodes;
+	int			nr_pmu_mappings;
+	int			nr_groups;
+	char			*cmdline;
+	char			*sibling_cores;
+	char			*sibling_threads;
+	char			*numa_nodes;
+	char			*pmu_mappings;
+};
+
+struct perf_header {
+	enum perf_header_version	version;
+	bool				needs_swap;
+	u64				data_offset;
+	u64				data_size;
+	u64				feat_offset;
+	DECLARE_BITMAP(adds_features, HEADER_FEAT_BITS);
+	struct perf_session_env 	env;
+};
+
+struct perf_evlist;
+struct perf_session;
+
+int perf_session__read_header(struct perf_session *session);
+int perf_session__write_header(struct perf_session *session,
+			       struct perf_evlist *evlist,
+			       int fd, bool at_exit);
+int perf_header__write_pipe(int fd);
+
+void perf_header__set_feat(struct perf_header *header, int feat);
+void perf_header__clear_feat(struct perf_header *header, int feat);
+bool perf_header__has_feat(const struct perf_header *header, int feat);
+
+int perf_header__set_cmdline(int argc, const char **argv);
+
+int perf_header__process_sections(struct perf_header *header, int fd,
+				  void *data,
+				  int (*process)(struct perf_file_section *section,
+				  struct perf_header *ph,
+				  int feat, int fd, void *data));
+
+int perf_header__fprintf_info(struct perf_session *s, FILE *fp, bool full);
+
+int perf_event__synthesize_attr(struct perf_tool *tool,
+				struct perf_event_attr *attr, u32 ids, u64 *id,
+				perf_event__handler_t process);
+int perf_event__synthesize_attrs(struct perf_tool *tool,
+				 struct perf_session *session,
+				 perf_event__handler_t process);
+int perf_event__process_attr(struct perf_tool *tool, union perf_event *event,
+			     struct perf_evlist **pevlist);
+
+int perf_event__synthesize_tracing_data(struct perf_tool *tool,
+					int fd, struct perf_evlist *evlist,
+					perf_event__handler_t process);
+int perf_event__process_tracing_data(struct perf_tool *tool,
+				     union perf_event *event,
+				     struct perf_session *session);
+
+int perf_event__synthesize_build_id(struct perf_tool *tool,
+				    struct dso *pos, u16 misc,
+				    perf_event__handler_t process,
+				    struct machine *machine);
+int perf_event__process_build_id(struct perf_tool *tool,
+				 union perf_event *event,
+				 struct perf_session *session);
+bool is_perf_magic(u64 magic);
+
+#define NAME_ALIGN 64
+
+int write_padded(int fd, const void *bf, size_t count, size_t count_aligned);
+
+/*
+ * arch specific callback
+ */
+int get_cpuid(char *buffer, size_t sz);
+
+#endif /* __PERF_HEADER_H */
diff --git a/tools/perf/util/help.c b/tools/perf/util/help.c
new file mode 100644
index 000000000..86c37c472
--- /dev/null
+++ b/tools/perf/util/help.c
@@ -0,0 +1,339 @@
+#include "cache.h"
+#include "../builtin.h"
+#include "exec_cmd.h"
+#include "levenshtein.h"
+#include "help.h"
+#include <termios.h>
+
+void add_cmdname(struct cmdnames *cmds, const char *name, size_t len)
+{
+	struct cmdname *ent = malloc(sizeof(*ent) + len + 1);
+
+	ent->len = len;
+	memcpy(ent->name, name, len);
+	ent->name[len] = 0;
+
+	ALLOC_GROW(cmds->names, cmds->cnt + 1, cmds->alloc);
+	cmds->names[cmds->cnt++] = ent;
+}
+
+static void clean_cmdnames(struct cmdnames *cmds)
+{
+	unsigned int i;
+
+	for (i = 0; i < cmds->cnt; ++i)
+		zfree(&cmds->names[i]);
+	zfree(&cmds->names);
+	cmds->cnt = 0;
+	cmds->alloc = 0;
+}
+
+static int cmdname_compare(const void *a_, const void *b_)
+{
+	struct cmdname *a = *(struct cmdname **)a_;
+	struct cmdname *b = *(struct cmdname **)b_;
+	return strcmp(a->name, b->name);
+}
+
+static void uniq(struct cmdnames *cmds)
+{
+	unsigned int i, j;
+
+	if (!cmds->cnt)
+		return;
+
+	for (i = j = 1; i < cmds->cnt; i++)
+		if (strcmp(cmds->names[i]->name, cmds->names[i-1]->name))
+			cmds->names[j++] = cmds->names[i];
+
+	cmds->cnt = j;
+}
+
+void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes)
+{
+	size_t ci, cj, ei;
+	int cmp;
+
+	ci = cj = ei = 0;
+	while (ci < cmds->cnt && ei < excludes->cnt) {
+		cmp = strcmp(cmds->names[ci]->name, excludes->names[ei]->name);
+		if (cmp < 0)
+			cmds->names[cj++] = cmds->names[ci++];
+		else if (cmp == 0)
+			ci++, ei++;
+		else if (cmp > 0)
+			ei++;
+	}
+
+	while (ci < cmds->cnt)
+		cmds->names[cj++] = cmds->names[ci++];
+
+	cmds->cnt = cj;
+}
+
+static void pretty_print_string_list(struct cmdnames *cmds, int longest)
+{
+	int cols = 1, rows;
+	int space = longest + 1; /* min 1 SP between words */
+	struct winsize win;
+	int max_cols;
+	int i, j;
+
+	get_term_dimensions(&win);
+	max_cols = win.ws_col - 1; /* don't print *on* the edge */
+
+	if (space < max_cols)
+		cols = max_cols / space;
+	rows = (cmds->cnt + cols - 1) / cols;
+
+	for (i = 0; i < rows; i++) {
+		printf("  ");
+
+		for (j = 0; j < cols; j++) {
+			unsigned int n = j * rows + i;
+			unsigned int size = space;
+
+			if (n >= cmds->cnt)
+				break;
+			if (j == cols-1 || n + rows >= cmds->cnt)
+				size = 1;
+			printf("%-*s", size, cmds->names[n]->name);
+		}
+		putchar('\n');
+	}
+}
+
+static int is_executable(const char *name)
+{
+	struct stat st;
+
+	if (stat(name, &st) || /* stat, not lstat */
+	    !S_ISREG(st.st_mode))
+		return 0;
+
+	return st.st_mode & S_IXUSR;
+}
+
+static void list_commands_in_dir(struct cmdnames *cmds,
+					 const char *path,
+					 const char *prefix)
+{
+	int prefix_len;
+	DIR *dir = opendir(path);
+	struct dirent *de;
+	struct strbuf buf = STRBUF_INIT;
+	int len;
+
+	if (!dir)
+		return;
+	if (!prefix)
+		prefix = "perf-";
+	prefix_len = strlen(prefix);
+
+	strbuf_addf(&buf, "%s/", path);
+	len = buf.len;
+
+	while ((de = readdir(dir)) != NULL) {
+		int entlen;
+
+		if (prefixcmp(de->d_name, prefix))
+			continue;
+
+		strbuf_setlen(&buf, len);
+		strbuf_addstr(&buf, de->d_name);
+		if (!is_executable(buf.buf))
+			continue;
+
+		entlen = strlen(de->d_name) - prefix_len;
+		if (has_extension(de->d_name, ".exe"))
+			entlen -= 4;
+
+		add_cmdname(cmds, de->d_name + prefix_len, entlen);
+	}
+	closedir(dir);
+	strbuf_release(&buf);
+}
+
+void load_command_list(const char *prefix,
+		struct cmdnames *main_cmds,
+		struct cmdnames *other_cmds)
+{
+	const char *env_path = getenv("PATH");
+	const char *exec_path = perf_exec_path();
+
+	if (exec_path) {
+		list_commands_in_dir(main_cmds, exec_path, prefix);
+		qsort(main_cmds->names, main_cmds->cnt,
+		      sizeof(*main_cmds->names), cmdname_compare);
+		uniq(main_cmds);
+	}
+
+	if (env_path) {
+		char *paths, *path, *colon;
+		path = paths = strdup(env_path);
+		while (1) {
+			if ((colon = strchr(path, PATH_SEP)))
+				*colon = 0;
+			if (!exec_path || strcmp(path, exec_path))
+				list_commands_in_dir(other_cmds, path, prefix);
+
+			if (!colon)
+				break;
+			path = colon + 1;
+		}
+		free(paths);
+
+		qsort(other_cmds->names, other_cmds->cnt,
+		      sizeof(*other_cmds->names), cmdname_compare);
+		uniq(other_cmds);
+	}
+	exclude_cmds(other_cmds, main_cmds);
+}
+
+void list_commands(const char *title, struct cmdnames *main_cmds,
+		   struct cmdnames *other_cmds)
+{
+	unsigned int i, longest = 0;
+
+	for (i = 0; i < main_cmds->cnt; i++)
+		if (longest < main_cmds->names[i]->len)
+			longest = main_cmds->names[i]->len;
+	for (i = 0; i < other_cmds->cnt; i++)
+		if (longest < other_cmds->names[i]->len)
+			longest = other_cmds->names[i]->len;
+
+	if (main_cmds->cnt) {
+		const char *exec_path = perf_exec_path();
+		printf("available %s in '%s'\n", title, exec_path);
+		printf("----------------");
+		mput_char('-', strlen(title) + strlen(exec_path));
+		putchar('\n');
+		pretty_print_string_list(main_cmds, longest);
+		putchar('\n');
+	}
+
+	if (other_cmds->cnt) {
+		printf("%s available from elsewhere on your $PATH\n", title);
+		printf("---------------------------------------");
+		mput_char('-', strlen(title));
+		putchar('\n');
+		pretty_print_string_list(other_cmds, longest);
+		putchar('\n');
+	}
+}
+
+int is_in_cmdlist(struct cmdnames *c, const char *s)
+{
+	unsigned int i;
+
+	for (i = 0; i < c->cnt; i++)
+		if (!strcmp(s, c->names[i]->name))
+			return 1;
+	return 0;
+}
+
+static int autocorrect;
+static struct cmdnames aliases;
+
+static int perf_unknown_cmd_config(const char *var, const char *value, void *cb)
+{
+	if (!strcmp(var, "help.autocorrect"))
+		autocorrect = perf_config_int(var,value);
+	/* Also use aliases for command lookup */
+	if (!prefixcmp(var, "alias."))
+		add_cmdname(&aliases, var + 6, strlen(var + 6));
+
+	return perf_default_config(var, value, cb);
+}
+
+static int levenshtein_compare(const void *p1, const void *p2)
+{
+	const struct cmdname *const *c1 = p1, *const *c2 = p2;
+	const char *s1 = (*c1)->name, *s2 = (*c2)->name;
+	int l1 = (*c1)->len;
+	int l2 = (*c2)->len;
+	return l1 != l2 ? l1 - l2 : strcmp(s1, s2);
+}
+
+static void add_cmd_list(struct cmdnames *cmds, struct cmdnames *old)
+{
+	unsigned int i;
+
+	ALLOC_GROW(cmds->names, cmds->cnt + old->cnt, cmds->alloc);
+
+	for (i = 0; i < old->cnt; i++)
+		cmds->names[cmds->cnt++] = old->names[i];
+	zfree(&old->names);
+	old->cnt = 0;
+}
+
+const char *help_unknown_cmd(const char *cmd)
+{
+	unsigned int i, n = 0, best_similarity = 0;
+	struct cmdnames main_cmds, other_cmds;
+
+	memset(&main_cmds, 0, sizeof(main_cmds));
+	memset(&other_cmds, 0, sizeof(main_cmds));
+	memset(&aliases, 0, sizeof(aliases));
+
+	perf_config(perf_unknown_cmd_config, NULL);
+
+	load_command_list("perf-", &main_cmds, &other_cmds);
+
+	add_cmd_list(&main_cmds, &aliases);
+	add_cmd_list(&main_cmds, &other_cmds);
+	qsort(main_cmds.names, main_cmds.cnt,
+	      sizeof(main_cmds.names), cmdname_compare);
+	uniq(&main_cmds);
+
+	if (main_cmds.cnt) {
+		/* This reuses cmdname->len for similarity index */
+		for (i = 0; i < main_cmds.cnt; ++i)
+			main_cmds.names[i]->len =
+				levenshtein(cmd, main_cmds.names[i]->name, 0, 2, 1, 4);
+
+		qsort(main_cmds.names, main_cmds.cnt,
+		      sizeof(*main_cmds.names), levenshtein_compare);
+
+		best_similarity = main_cmds.names[0]->len;
+		n = 1;
+		while (n < main_cmds.cnt && best_similarity == main_cmds.names[n]->len)
+			++n;
+	}
+
+	if (autocorrect && n == 1) {
+		const char *assumed = main_cmds.names[0]->name;
+
+		main_cmds.names[0] = NULL;
+		clean_cmdnames(&main_cmds);
+		fprintf(stderr, "WARNING: You called a perf program named '%s', "
+			"which does not exist.\n"
+			"Continuing under the assumption that you meant '%s'\n",
+			cmd, assumed);
+		if (autocorrect > 0) {
+			fprintf(stderr, "in %0.1f seconds automatically...\n",
+				(float)autocorrect/10.0);
+			poll(NULL, 0, autocorrect * 100);
+		}
+		return assumed;
+	}
+
+	fprintf(stderr, "perf: '%s' is not a perf-command. See 'perf --help'.\n", cmd);
+
+	if (main_cmds.cnt && best_similarity < 6) {
+		fprintf(stderr, "\nDid you mean %s?\n",
+			n < 2 ? "this": "one of these");
+
+		for (i = 0; i < n; i++)
+			fprintf(stderr, "\t%s\n", main_cmds.names[i]->name);
+	}
+
+	exit(1);
+}
+
+int cmd_version(int argc __maybe_unused, const char **argv __maybe_unused,
+		const char *prefix __maybe_unused)
+{
+	printf("perf version %s\n", perf_version_string);
+	return 0;
+}
diff --git a/tools/perf/util/help.h b/tools/perf/util/help.h
new file mode 100644
index 000000000..7f5c6dedd
--- /dev/null
+++ b/tools/perf/util/help.h
@@ -0,0 +1,29 @@
+#ifndef __PERF_HELP_H
+#define __PERF_HELP_H
+
+struct cmdnames {
+	size_t alloc;
+	size_t cnt;
+	struct cmdname {
+		size_t len; /* also used for similarity index in help.c */
+		char name[FLEX_ARRAY];
+	} **names;
+};
+
+static inline void mput_char(char c, unsigned int num)
+{
+	while(num--)
+		putchar(c);
+}
+
+void load_command_list(const char *prefix,
+		struct cmdnames *main_cmds,
+		struct cmdnames *other_cmds);
+void add_cmdname(struct cmdnames *cmds, const char *name, size_t len);
+/* Here we require that excludes is a sorted list. */
+void exclude_cmds(struct cmdnames *cmds, struct cmdnames *excludes);
+int is_in_cmdlist(struct cmdnames *c, const char *s);
+void list_commands(const char *title, struct cmdnames *main_cmds,
+		   struct cmdnames *other_cmds);
+
+#endif /* __PERF_HELP_H */
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
new file mode 100644
index 000000000..cc22b9158
--- /dev/null
+++ b/tools/perf/util/hist.c
@@ -0,0 +1,1482 @@
+#include "util.h"
+#include "build-id.h"
+#include "hist.h"
+#include "session.h"
+#include "sort.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "annotate.h"
+#include "ui/progress.h"
+#include <math.h>
+
+static bool hists__filter_entry_by_dso(struct hists *hists,
+				       struct hist_entry *he);
+static bool hists__filter_entry_by_thread(struct hists *hists,
+					  struct hist_entry *he);
+static bool hists__filter_entry_by_symbol(struct hists *hists,
+					  struct hist_entry *he);
+
+u16 hists__col_len(struct hists *hists, enum hist_column col)
+{
+	return hists->col_len[col];
+}
+
+void hists__set_col_len(struct hists *hists, enum hist_column col, u16 len)
+{
+	hists->col_len[col] = len;
+}
+
+bool hists__new_col_len(struct hists *hists, enum hist_column col, u16 len)
+{
+	if (len > hists__col_len(hists, col)) {
+		hists__set_col_len(hists, col, len);
+		return true;
+	}
+	return false;
+}
+
+void hists__reset_col_len(struct hists *hists)
+{
+	enum hist_column col;
+
+	for (col = 0; col < HISTC_NR_COLS; ++col)
+		hists__set_col_len(hists, col, 0);
+}
+
+static void hists__set_unres_dso_col_len(struct hists *hists, int dso)
+{
+	const unsigned int unresolved_col_width = BITS_PER_LONG / 4;
+
+	if (hists__col_len(hists, dso) < unresolved_col_width &&
+	    !symbol_conf.col_width_list_str && !symbol_conf.field_sep &&
+	    !symbol_conf.dso_list)
+		hists__set_col_len(hists, dso, unresolved_col_width);
+}
+
+void hists__calc_col_len(struct hists *hists, struct hist_entry *h)
+{
+	const unsigned int unresolved_col_width = BITS_PER_LONG / 4;
+	int symlen;
+	u16 len;
+
+	/*
+	 * +4 accounts for '[x] ' priv level info
+	 * +2 accounts for 0x prefix on raw addresses
+	 * +3 accounts for ' y ' symtab origin info
+	 */
+	if (h->ms.sym) {
+		symlen = h->ms.sym->namelen + 4;
+		if (verbose)
+			symlen += BITS_PER_LONG / 4 + 2 + 3;
+		hists__new_col_len(hists, HISTC_SYMBOL, symlen);
+	} else {
+		symlen = unresolved_col_width + 4 + 2;
+		hists__new_col_len(hists, HISTC_SYMBOL, symlen);
+		hists__set_unres_dso_col_len(hists, HISTC_DSO);
+	}
+
+	len = thread__comm_len(h->thread);
+	if (hists__new_col_len(hists, HISTC_COMM, len))
+		hists__set_col_len(hists, HISTC_THREAD, len + 6);
+
+	if (h->ms.map) {
+		len = dso__name_len(h->ms.map->dso);
+		hists__new_col_len(hists, HISTC_DSO, len);
+	}
+
+	if (h->parent)
+		hists__new_col_len(hists, HISTC_PARENT, h->parent->namelen);
+
+	if (h->branch_info) {
+		if (h->branch_info->from.sym) {
+			symlen = (int)h->branch_info->from.sym->namelen + 4;
+			if (verbose)
+				symlen += BITS_PER_LONG / 4 + 2 + 3;
+			hists__new_col_len(hists, HISTC_SYMBOL_FROM, symlen);
+
+			symlen = dso__name_len(h->branch_info->from.map->dso);
+			hists__new_col_len(hists, HISTC_DSO_FROM, symlen);
+		} else {
+			symlen = unresolved_col_width + 4 + 2;
+			hists__new_col_len(hists, HISTC_SYMBOL_FROM, symlen);
+			hists__set_unres_dso_col_len(hists, HISTC_DSO_FROM);
+		}
+
+		if (h->branch_info->to.sym) {
+			symlen = (int)h->branch_info->to.sym->namelen + 4;
+			if (verbose)
+				symlen += BITS_PER_LONG / 4 + 2 + 3;
+			hists__new_col_len(hists, HISTC_SYMBOL_TO, symlen);
+
+			symlen = dso__name_len(h->branch_info->to.map->dso);
+			hists__new_col_len(hists, HISTC_DSO_TO, symlen);
+		} else {
+			symlen = unresolved_col_width + 4 + 2;
+			hists__new_col_len(hists, HISTC_SYMBOL_TO, symlen);
+			hists__set_unres_dso_col_len(hists, HISTC_DSO_TO);
+		}
+	}
+
+	if (h->mem_info) {
+		if (h->mem_info->daddr.sym) {
+			symlen = (int)h->mem_info->daddr.sym->namelen + 4
+			       + unresolved_col_width + 2;
+			hists__new_col_len(hists, HISTC_MEM_DADDR_SYMBOL,
+					   symlen);
+			hists__new_col_len(hists, HISTC_MEM_DCACHELINE,
+					   symlen + 1);
+		} else {
+			symlen = unresolved_col_width + 4 + 2;
+			hists__new_col_len(hists, HISTC_MEM_DADDR_SYMBOL,
+					   symlen);
+		}
+		if (h->mem_info->daddr.map) {
+			symlen = dso__name_len(h->mem_info->daddr.map->dso);
+			hists__new_col_len(hists, HISTC_MEM_DADDR_DSO,
+					   symlen);
+		} else {
+			symlen = unresolved_col_width + 4 + 2;
+			hists__set_unres_dso_col_len(hists, HISTC_MEM_DADDR_DSO);
+		}
+	} else {
+		symlen = unresolved_col_width + 4 + 2;
+		hists__new_col_len(hists, HISTC_MEM_DADDR_SYMBOL, symlen);
+		hists__set_unres_dso_col_len(hists, HISTC_MEM_DADDR_DSO);
+	}
+
+	hists__new_col_len(hists, HISTC_MEM_LOCKED, 6);
+	hists__new_col_len(hists, HISTC_MEM_TLB, 22);
+	hists__new_col_len(hists, HISTC_MEM_SNOOP, 12);
+	hists__new_col_len(hists, HISTC_MEM_LVL, 21 + 3);
+	hists__new_col_len(hists, HISTC_LOCAL_WEIGHT, 12);
+	hists__new_col_len(hists, HISTC_GLOBAL_WEIGHT, 12);
+
+	if (h->transaction)
+		hists__new_col_len(hists, HISTC_TRANSACTION,
+				   hist_entry__transaction_len());
+}
+
+void hists__output_recalc_col_len(struct hists *hists, int max_rows)
+{
+	struct rb_node *next = rb_first(&hists->entries);
+	struct hist_entry *n;
+	int row = 0;
+
+	hists__reset_col_len(hists);
+
+	while (next && row++ < max_rows) {
+		n = rb_entry(next, struct hist_entry, rb_node);
+		if (!n->filtered)
+			hists__calc_col_len(hists, n);
+		next = rb_next(&n->rb_node);
+	}
+}
+
+static void he_stat__add_cpumode_period(struct he_stat *he_stat,
+					unsigned int cpumode, u64 period)
+{
+	switch (cpumode) {
+	case PERF_RECORD_MISC_KERNEL:
+		he_stat->period_sys += period;
+		break;
+	case PERF_RECORD_MISC_USER:
+		he_stat->period_us += period;
+		break;
+	case PERF_RECORD_MISC_GUEST_KERNEL:
+		he_stat->period_guest_sys += period;
+		break;
+	case PERF_RECORD_MISC_GUEST_USER:
+		he_stat->period_guest_us += period;
+		break;
+	default:
+		break;
+	}
+}
+
+static void he_stat__add_period(struct he_stat *he_stat, u64 period,
+				u64 weight)
+{
+
+	he_stat->period		+= period;
+	he_stat->weight		+= weight;
+	he_stat->nr_events	+= 1;
+}
+
+static void he_stat__add_stat(struct he_stat *dest, struct he_stat *src)
+{
+	dest->period		+= src->period;
+	dest->period_sys	+= src->period_sys;
+	dest->period_us		+= src->period_us;
+	dest->period_guest_sys	+= src->period_guest_sys;
+	dest->period_guest_us	+= src->period_guest_us;
+	dest->nr_events		+= src->nr_events;
+	dest->weight		+= src->weight;
+}
+
+static void he_stat__decay(struct he_stat *he_stat)
+{
+	he_stat->period = (he_stat->period * 7) / 8;
+	he_stat->nr_events = (he_stat->nr_events * 7) / 8;
+	/* XXX need decay for weight too? */
+}
+
+static bool hists__decay_entry(struct hists *hists, struct hist_entry *he)
+{
+	u64 prev_period = he->stat.period;
+	u64 diff;
+
+	if (prev_period == 0)
+		return true;
+
+	he_stat__decay(&he->stat);
+	if (symbol_conf.cumulate_callchain)
+		he_stat__decay(he->stat_acc);
+
+	diff = prev_period - he->stat.period;
+
+	hists->stats.total_period -= diff;
+	if (!he->filtered)
+		hists->stats.total_non_filtered_period -= diff;
+
+	return he->stat.period == 0;
+}
+
+static void hists__delete_entry(struct hists *hists, struct hist_entry *he)
+{
+	rb_erase(&he->rb_node, &hists->entries);
+
+	if (sort__need_collapse)
+		rb_erase(&he->rb_node_in, &hists->entries_collapsed);
+
+	--hists->nr_entries;
+	if (!he->filtered)
+		--hists->nr_non_filtered_entries;
+
+	hist_entry__delete(he);
+}
+
+void hists__decay_entries(struct hists *hists, bool zap_user, bool zap_kernel)
+{
+	struct rb_node *next = rb_first(&hists->entries);
+	struct hist_entry *n;
+
+	while (next) {
+		n = rb_entry(next, struct hist_entry, rb_node);
+		next = rb_next(&n->rb_node);
+		if (((zap_user && n->level == '.') ||
+		     (zap_kernel && n->level != '.') ||
+		     hists__decay_entry(hists, n))) {
+			hists__delete_entry(hists, n);
+		}
+	}
+}
+
+void hists__delete_entries(struct hists *hists)
+{
+	struct rb_node *next = rb_first(&hists->entries);
+	struct hist_entry *n;
+
+	while (next) {
+		n = rb_entry(next, struct hist_entry, rb_node);
+		next = rb_next(&n->rb_node);
+
+		hists__delete_entry(hists, n);
+	}
+}
+
+/*
+ * histogram, sorted on item, collects periods
+ */
+
+static struct hist_entry *hist_entry__new(struct hist_entry *template,
+					  bool sample_self)
+{
+	size_t callchain_size = 0;
+	struct hist_entry *he;
+
+	if (symbol_conf.use_callchain)
+		callchain_size = sizeof(struct callchain_root);
+
+	he = zalloc(sizeof(*he) + callchain_size);
+
+	if (he != NULL) {
+		*he = *template;
+
+		if (symbol_conf.cumulate_callchain) {
+			he->stat_acc = malloc(sizeof(he->stat));
+			if (he->stat_acc == NULL) {
+				free(he);
+				return NULL;
+			}
+			memcpy(he->stat_acc, &he->stat, sizeof(he->stat));
+			if (!sample_self)
+				memset(&he->stat, 0, sizeof(he->stat));
+		}
+
+		if (he->ms.map)
+			he->ms.map->referenced = true;
+
+		if (he->branch_info) {
+			/*
+			 * This branch info is (a part of) allocated from
+			 * sample__resolve_bstack() and will be freed after
+			 * adding new entries.  So we need to save a copy.
+			 */
+			he->branch_info = malloc(sizeof(*he->branch_info));
+			if (he->branch_info == NULL) {
+				free(he->stat_acc);
+				free(he);
+				return NULL;
+			}
+
+			memcpy(he->branch_info, template->branch_info,
+			       sizeof(*he->branch_info));
+
+			if (he->branch_info->from.map)
+				he->branch_info->from.map->referenced = true;
+			if (he->branch_info->to.map)
+				he->branch_info->to.map->referenced = true;
+		}
+
+		if (he->mem_info) {
+			if (he->mem_info->iaddr.map)
+				he->mem_info->iaddr.map->referenced = true;
+			if (he->mem_info->daddr.map)
+				he->mem_info->daddr.map->referenced = true;
+		}
+
+		if (symbol_conf.use_callchain)
+			callchain_init(he->callchain);
+
+		INIT_LIST_HEAD(&he->pairs.node);
+		thread__get(he->thread);
+	}
+
+	return he;
+}
+
+static u8 symbol__parent_filter(const struct symbol *parent)
+{
+	if (symbol_conf.exclude_other && parent == NULL)
+		return 1 << HIST_FILTER__PARENT;
+	return 0;
+}
+
+static struct hist_entry *add_hist_entry(struct hists *hists,
+					 struct hist_entry *entry,
+					 struct addr_location *al,
+					 bool sample_self)
+{
+	struct rb_node **p;
+	struct rb_node *parent = NULL;
+	struct hist_entry *he;
+	int64_t cmp;
+	u64 period = entry->stat.period;
+	u64 weight = entry->stat.weight;
+
+	p = &hists->entries_in->rb_node;
+
+	while (*p != NULL) {
+		parent = *p;
+		he = rb_entry(parent, struct hist_entry, rb_node_in);
+
+		/*
+		 * Make sure that it receives arguments in a same order as
+		 * hist_entry__collapse() so that we can use an appropriate
+		 * function when searching an entry regardless which sort
+		 * keys were used.
+		 */
+		cmp = hist_entry__cmp(he, entry);
+
+		if (!cmp) {
+			if (sample_self)
+				he_stat__add_period(&he->stat, period, weight);
+			if (symbol_conf.cumulate_callchain)
+				he_stat__add_period(he->stat_acc, period, weight);
+
+			/*
+			 * This mem info was allocated from sample__resolve_mem
+			 * and will not be used anymore.
+			 */
+			zfree(&entry->mem_info);
+
+			/* If the map of an existing hist_entry has
+			 * become out-of-date due to an exec() or
+			 * similar, update it.  Otherwise we will
+			 * mis-adjust symbol addresses when computing
+			 * the history counter to increment.
+			 */
+			if (he->ms.map != entry->ms.map) {
+				he->ms.map = entry->ms.map;
+				if (he->ms.map)
+					he->ms.map->referenced = true;
+			}
+			goto out;
+		}
+
+		if (cmp < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	he = hist_entry__new(entry, sample_self);
+	if (!he)
+		return NULL;
+
+	hists->nr_entries++;
+
+	rb_link_node(&he->rb_node_in, parent, p);
+	rb_insert_color(&he->rb_node_in, hists->entries_in);
+out:
+	if (sample_self)
+		he_stat__add_cpumode_period(&he->stat, al->cpumode, period);
+	if (symbol_conf.cumulate_callchain)
+		he_stat__add_cpumode_period(he->stat_acc, al->cpumode, period);
+	return he;
+}
+
+struct hist_entry *__hists__add_entry(struct hists *hists,
+				      struct addr_location *al,
+				      struct symbol *sym_parent,
+				      struct branch_info *bi,
+				      struct mem_info *mi,
+				      u64 period, u64 weight, u64 transaction,
+				      bool sample_self)
+{
+	struct hist_entry entry = {
+		.thread	= al->thread,
+		.comm = thread__comm(al->thread),
+		.ms = {
+			.map	= al->map,
+			.sym	= al->sym,
+		},
+		.cpu	 = al->cpu,
+		.cpumode = al->cpumode,
+		.ip	 = al->addr,
+		.level	 = al->level,
+		.stat = {
+			.nr_events = 1,
+			.period	= period,
+			.weight = weight,
+		},
+		.parent = sym_parent,
+		.filtered = symbol__parent_filter(sym_parent) | al->filtered,
+		.hists	= hists,
+		.branch_info = bi,
+		.mem_info = mi,
+		.transaction = transaction,
+	};
+
+	return add_hist_entry(hists, &entry, al, sample_self);
+}
+
+static int
+iter_next_nop_entry(struct hist_entry_iter *iter __maybe_unused,
+		    struct addr_location *al __maybe_unused)
+{
+	return 0;
+}
+
+static int
+iter_add_next_nop_entry(struct hist_entry_iter *iter __maybe_unused,
+			struct addr_location *al __maybe_unused)
+{
+	return 0;
+}
+
+static int
+iter_prepare_mem_entry(struct hist_entry_iter *iter, struct addr_location *al)
+{
+	struct perf_sample *sample = iter->sample;
+	struct mem_info *mi;
+
+	mi = sample__resolve_mem(sample, al);
+	if (mi == NULL)
+		return -ENOMEM;
+
+	iter->priv = mi;
+	return 0;
+}
+
+static int
+iter_add_single_mem_entry(struct hist_entry_iter *iter, struct addr_location *al)
+{
+	u64 cost;
+	struct mem_info *mi = iter->priv;
+	struct hists *hists = evsel__hists(iter->evsel);
+	struct hist_entry *he;
+
+	if (mi == NULL)
+		return -EINVAL;
+
+	cost = iter->sample->weight;
+	if (!cost)
+		cost = 1;
+
+	/*
+	 * must pass period=weight in order to get the correct
+	 * sorting from hists__collapse_resort() which is solely
+	 * based on periods. We want sorting be done on nr_events * weight
+	 * and this is indirectly achieved by passing period=weight here
+	 * and the he_stat__add_period() function.
+	 */
+	he = __hists__add_entry(hists, al, iter->parent, NULL, mi,
+				cost, cost, 0, true);
+	if (!he)
+		return -ENOMEM;
+
+	iter->he = he;
+	return 0;
+}
+
+static int
+iter_finish_mem_entry(struct hist_entry_iter *iter,
+		      struct addr_location *al __maybe_unused)
+{
+	struct perf_evsel *evsel = iter->evsel;
+	struct hists *hists = evsel__hists(evsel);
+	struct hist_entry *he = iter->he;
+	int err = -EINVAL;
+
+	if (he == NULL)
+		goto out;
+
+	hists__inc_nr_samples(hists, he->filtered);
+
+	err = hist_entry__append_callchain(he, iter->sample);
+
+out:
+	/*
+	 * We don't need to free iter->priv (mem_info) here since
+	 * the mem info was either already freed in add_hist_entry() or
+	 * passed to a new hist entry by hist_entry__new().
+	 */
+	iter->priv = NULL;
+
+	iter->he = NULL;
+	return err;
+}
+
+static int
+iter_prepare_branch_entry(struct hist_entry_iter *iter, struct addr_location *al)
+{
+	struct branch_info *bi;
+	struct perf_sample *sample = iter->sample;
+
+	bi = sample__resolve_bstack(sample, al);
+	if (!bi)
+		return -ENOMEM;
+
+	iter->curr = 0;
+	iter->total = sample->branch_stack->nr;
+
+	iter->priv = bi;
+	return 0;
+}
+
+static int
+iter_add_single_branch_entry(struct hist_entry_iter *iter __maybe_unused,
+			     struct addr_location *al __maybe_unused)
+{
+	/* to avoid calling callback function */
+	iter->he = NULL;
+
+	return 0;
+}
+
+static int
+iter_next_branch_entry(struct hist_entry_iter *iter, struct addr_location *al)
+{
+	struct branch_info *bi = iter->priv;
+	int i = iter->curr;
+
+	if (bi == NULL)
+		return 0;
+
+	if (iter->curr >= iter->total)
+		return 0;
+
+	al->map = bi[i].to.map;
+	al->sym = bi[i].to.sym;
+	al->addr = bi[i].to.addr;
+	return 1;
+}
+
+static int
+iter_add_next_branch_entry(struct hist_entry_iter *iter, struct addr_location *al)
+{
+	struct branch_info *bi;
+	struct perf_evsel *evsel = iter->evsel;
+	struct hists *hists = evsel__hists(evsel);
+	struct hist_entry *he = NULL;
+	int i = iter->curr;
+	int err = 0;
+
+	bi = iter->priv;
+
+	if (iter->hide_unresolved && !(bi[i].from.sym && bi[i].to.sym))
+		goto out;
+
+	/*
+	 * The report shows the percentage of total branches captured
+	 * and not events sampled. Thus we use a pseudo period of 1.
+	 */
+	he = __hists__add_entry(hists, al, iter->parent, &bi[i], NULL,
+				1, 1, 0, true);
+	if (he == NULL)
+		return -ENOMEM;
+
+	hists__inc_nr_samples(hists, he->filtered);
+
+out:
+	iter->he = he;
+	iter->curr++;
+	return err;
+}
+
+static int
+iter_finish_branch_entry(struct hist_entry_iter *iter,
+			 struct addr_location *al __maybe_unused)
+{
+	zfree(&iter->priv);
+	iter->he = NULL;
+
+	return iter->curr >= iter->total ? 0 : -1;
+}
+
+static int
+iter_prepare_normal_entry(struct hist_entry_iter *iter __maybe_unused,
+			  struct addr_location *al __maybe_unused)
+{
+	return 0;
+}
+
+static int
+iter_add_single_normal_entry(struct hist_entry_iter *iter, struct addr_location *al)
+{
+	struct perf_evsel *evsel = iter->evsel;
+	struct perf_sample *sample = iter->sample;
+	struct hist_entry *he;
+
+	he = __hists__add_entry(evsel__hists(evsel), al, iter->parent, NULL, NULL,
+				sample->period, sample->weight,
+				sample->transaction, true);
+	if (he == NULL)
+		return -ENOMEM;
+
+	iter->he = he;
+	return 0;
+}
+
+static int
+iter_finish_normal_entry(struct hist_entry_iter *iter,
+			 struct addr_location *al __maybe_unused)
+{
+	struct hist_entry *he = iter->he;
+	struct perf_evsel *evsel = iter->evsel;
+	struct perf_sample *sample = iter->sample;
+
+	if (he == NULL)
+		return 0;
+
+	iter->he = NULL;
+
+	hists__inc_nr_samples(evsel__hists(evsel), he->filtered);
+
+	return hist_entry__append_callchain(he, sample);
+}
+
+static int
+iter_prepare_cumulative_entry(struct hist_entry_iter *iter __maybe_unused,
+			      struct addr_location *al __maybe_unused)
+{
+	struct hist_entry **he_cache;
+
+	callchain_cursor_commit(&callchain_cursor);
+
+	/*
+	 * This is for detecting cycles or recursions so that they're
+	 * cumulated only one time to prevent entries more than 100%
+	 * overhead.
+	 */
+	he_cache = malloc(sizeof(*he_cache) * (PERF_MAX_STACK_DEPTH + 1));
+	if (he_cache == NULL)
+		return -ENOMEM;
+
+	iter->priv = he_cache;
+	iter->curr = 0;
+
+	return 0;
+}
+
+static int
+iter_add_single_cumulative_entry(struct hist_entry_iter *iter,
+				 struct addr_location *al)
+{
+	struct perf_evsel *evsel = iter->evsel;
+	struct hists *hists = evsel__hists(evsel);
+	struct perf_sample *sample = iter->sample;
+	struct hist_entry **he_cache = iter->priv;
+	struct hist_entry *he;
+	int err = 0;
+
+	he = __hists__add_entry(hists, al, iter->parent, NULL, NULL,
+				sample->period, sample->weight,
+				sample->transaction, true);
+	if (he == NULL)
+		return -ENOMEM;
+
+	iter->he = he;
+	he_cache[iter->curr++] = he;
+
+	hist_entry__append_callchain(he, sample);
+
+	/*
+	 * We need to re-initialize the cursor since callchain_append()
+	 * advanced the cursor to the end.
+	 */
+	callchain_cursor_commit(&callchain_cursor);
+
+	hists__inc_nr_samples(hists, he->filtered);
+
+	return err;
+}
+
+static int
+iter_next_cumulative_entry(struct hist_entry_iter *iter,
+			   struct addr_location *al)
+{
+	struct callchain_cursor_node *node;
+
+	node = callchain_cursor_current(&callchain_cursor);
+	if (node == NULL)
+		return 0;
+
+	return fill_callchain_info(al, node, iter->hide_unresolved);
+}
+
+static int
+iter_add_next_cumulative_entry(struct hist_entry_iter *iter,
+			       struct addr_location *al)
+{
+	struct perf_evsel *evsel = iter->evsel;
+	struct perf_sample *sample = iter->sample;
+	struct hist_entry **he_cache = iter->priv;
+	struct hist_entry *he;
+	struct hist_entry he_tmp = {
+		.cpu = al->cpu,
+		.thread = al->thread,
+		.comm = thread__comm(al->thread),
+		.ip = al->addr,
+		.ms = {
+			.map = al->map,
+			.sym = al->sym,
+		},
+		.parent = iter->parent,
+	};
+	int i;
+	struct callchain_cursor cursor;
+
+	callchain_cursor_snapshot(&cursor, &callchain_cursor);
+
+	callchain_cursor_advance(&callchain_cursor);
+
+	/*
+	 * Check if there's duplicate entries in the callchain.
+	 * It's possible that it has cycles or recursive calls.
+	 */
+	for (i = 0; i < iter->curr; i++) {
+		if (hist_entry__cmp(he_cache[i], &he_tmp) == 0) {
+			/* to avoid calling callback function */
+			iter->he = NULL;
+			return 0;
+		}
+	}
+
+	he = __hists__add_entry(evsel__hists(evsel), al, iter->parent, NULL, NULL,
+				sample->period, sample->weight,
+				sample->transaction, false);
+	if (he == NULL)
+		return -ENOMEM;
+
+	iter->he = he;
+	he_cache[iter->curr++] = he;
+
+	if (symbol_conf.use_callchain)
+		callchain_append(he->callchain, &cursor, sample->period);
+	return 0;
+}
+
+static int
+iter_finish_cumulative_entry(struct hist_entry_iter *iter,
+			     struct addr_location *al __maybe_unused)
+{
+	zfree(&iter->priv);
+	iter->he = NULL;
+
+	return 0;
+}
+
+const struct hist_iter_ops hist_iter_mem = {
+	.prepare_entry 		= iter_prepare_mem_entry,
+	.add_single_entry 	= iter_add_single_mem_entry,
+	.next_entry 		= iter_next_nop_entry,
+	.add_next_entry 	= iter_add_next_nop_entry,
+	.finish_entry 		= iter_finish_mem_entry,
+};
+
+const struct hist_iter_ops hist_iter_branch = {
+	.prepare_entry 		= iter_prepare_branch_entry,
+	.add_single_entry 	= iter_add_single_branch_entry,
+	.next_entry 		= iter_next_branch_entry,
+	.add_next_entry 	= iter_add_next_branch_entry,
+	.finish_entry 		= iter_finish_branch_entry,
+};
+
+const struct hist_iter_ops hist_iter_normal = {
+	.prepare_entry 		= iter_prepare_normal_entry,
+	.add_single_entry 	= iter_add_single_normal_entry,
+	.next_entry 		= iter_next_nop_entry,
+	.add_next_entry 	= iter_add_next_nop_entry,
+	.finish_entry 		= iter_finish_normal_entry,
+};
+
+const struct hist_iter_ops hist_iter_cumulative = {
+	.prepare_entry 		= iter_prepare_cumulative_entry,
+	.add_single_entry 	= iter_add_single_cumulative_entry,
+	.next_entry 		= iter_next_cumulative_entry,
+	.add_next_entry 	= iter_add_next_cumulative_entry,
+	.finish_entry 		= iter_finish_cumulative_entry,
+};
+
+int hist_entry_iter__add(struct hist_entry_iter *iter, struct addr_location *al,
+			 struct perf_evsel *evsel, struct perf_sample *sample,
+			 int max_stack_depth, void *arg)
+{
+	int err, err2;
+
+	err = sample__resolve_callchain(sample, &iter->parent, evsel, al,
+					max_stack_depth);
+	if (err)
+		return err;
+
+	iter->evsel = evsel;
+	iter->sample = sample;
+
+	err = iter->ops->prepare_entry(iter, al);
+	if (err)
+		goto out;
+
+	err = iter->ops->add_single_entry(iter, al);
+	if (err)
+		goto out;
+
+	if (iter->he && iter->add_entry_cb) {
+		err = iter->add_entry_cb(iter, al, true, arg);
+		if (err)
+			goto out;
+	}
+
+	while (iter->ops->next_entry(iter, al)) {
+		err = iter->ops->add_next_entry(iter, al);
+		if (err)
+			break;
+
+		if (iter->he && iter->add_entry_cb) {
+			err = iter->add_entry_cb(iter, al, false, arg);
+			if (err)
+				goto out;
+		}
+	}
+
+out:
+	err2 = iter->ops->finish_entry(iter, al);
+	if (!err)
+		err = err2;
+
+	return err;
+}
+
+int64_t
+hist_entry__cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct perf_hpp_fmt *fmt;
+	int64_t cmp = 0;
+
+	perf_hpp__for_each_sort_list(fmt) {
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		cmp = fmt->cmp(fmt, left, right);
+		if (cmp)
+			break;
+	}
+
+	return cmp;
+}
+
+int64_t
+hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
+{
+	struct perf_hpp_fmt *fmt;
+	int64_t cmp = 0;
+
+	perf_hpp__for_each_sort_list(fmt) {
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		cmp = fmt->collapse(fmt, left, right);
+		if (cmp)
+			break;
+	}
+
+	return cmp;
+}
+
+void hist_entry__delete(struct hist_entry *he)
+{
+	thread__zput(he->thread);
+	zfree(&he->branch_info);
+	zfree(&he->mem_info);
+	zfree(&he->stat_acc);
+	free_srcline(he->srcline);
+	free_callchain(he->callchain);
+	free(he);
+}
+
+/*
+ * collapse the histogram
+ */
+
+static bool hists__collapse_insert_entry(struct hists *hists __maybe_unused,
+					 struct rb_root *root,
+					 struct hist_entry *he)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct hist_entry *iter;
+	int64_t cmp;
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct hist_entry, rb_node_in);
+
+		cmp = hist_entry__collapse(iter, he);
+
+		if (!cmp) {
+			he_stat__add_stat(&iter->stat, &he->stat);
+			if (symbol_conf.cumulate_callchain)
+				he_stat__add_stat(iter->stat_acc, he->stat_acc);
+
+			if (symbol_conf.use_callchain) {
+				callchain_cursor_reset(&callchain_cursor);
+				callchain_merge(&callchain_cursor,
+						iter->callchain,
+						he->callchain);
+			}
+			hist_entry__delete(he);
+			return false;
+		}
+
+		if (cmp < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	hists->nr_entries++;
+
+	rb_link_node(&he->rb_node_in, parent, p);
+	rb_insert_color(&he->rb_node_in, root);
+	return true;
+}
+
+static struct rb_root *hists__get_rotate_entries_in(struct hists *hists)
+{
+	struct rb_root *root;
+
+	pthread_mutex_lock(&hists->lock);
+
+	root = hists->entries_in;
+	if (++hists->entries_in > &hists->entries_in_array[1])
+		hists->entries_in = &hists->entries_in_array[0];
+
+	pthread_mutex_unlock(&hists->lock);
+
+	return root;
+}
+
+static void hists__apply_filters(struct hists *hists, struct hist_entry *he)
+{
+	hists__filter_entry_by_dso(hists, he);
+	hists__filter_entry_by_thread(hists, he);
+	hists__filter_entry_by_symbol(hists, he);
+}
+
+void hists__collapse_resort(struct hists *hists, struct ui_progress *prog)
+{
+	struct rb_root *root;
+	struct rb_node *next;
+	struct hist_entry *n;
+
+	if (!sort__need_collapse)
+		return;
+
+	hists->nr_entries = 0;
+
+	root = hists__get_rotate_entries_in(hists);
+
+	next = rb_first(root);
+
+	while (next) {
+		if (session_done())
+			break;
+		n = rb_entry(next, struct hist_entry, rb_node_in);
+		next = rb_next(&n->rb_node_in);
+
+		rb_erase(&n->rb_node_in, root);
+		if (hists__collapse_insert_entry(hists, &hists->entries_collapsed, n)) {
+			/*
+			 * If it wasn't combined with one of the entries already
+			 * collapsed, we need to apply the filters that may have
+			 * been set by, say, the hist_browser.
+			 */
+			hists__apply_filters(hists, n);
+		}
+		if (prog)
+			ui_progress__update(prog, 1);
+	}
+}
+
+static int hist_entry__sort(struct hist_entry *a, struct hist_entry *b)
+{
+	struct perf_hpp_fmt *fmt;
+	int64_t cmp = 0;
+
+	perf_hpp__for_each_sort_list(fmt) {
+		if (perf_hpp__should_skip(fmt))
+			continue;
+
+		cmp = fmt->sort(fmt, a, b);
+		if (cmp)
+			break;
+	}
+
+	return cmp;
+}
+
+static void hists__reset_filter_stats(struct hists *hists)
+{
+	hists->nr_non_filtered_entries = 0;
+	hists->stats.total_non_filtered_period = 0;
+}
+
+void hists__reset_stats(struct hists *hists)
+{
+	hists->nr_entries = 0;
+	hists->stats.total_period = 0;
+
+	hists__reset_filter_stats(hists);
+}
+
+static void hists__inc_filter_stats(struct hists *hists, struct hist_entry *h)
+{
+	hists->nr_non_filtered_entries++;
+	hists->stats.total_non_filtered_period += h->stat.period;
+}
+
+void hists__inc_stats(struct hists *hists, struct hist_entry *h)
+{
+	if (!h->filtered)
+		hists__inc_filter_stats(hists, h);
+
+	hists->nr_entries++;
+	hists->stats.total_period += h->stat.period;
+}
+
+static void __hists__insert_output_entry(struct rb_root *entries,
+					 struct hist_entry *he,
+					 u64 min_callchain_hits)
+{
+	struct rb_node **p = &entries->rb_node;
+	struct rb_node *parent = NULL;
+	struct hist_entry *iter;
+
+	if (symbol_conf.use_callchain)
+		callchain_param.sort(&he->sorted_chain, he->callchain,
+				      min_callchain_hits, &callchain_param);
+
+	while (*p != NULL) {
+		parent = *p;
+		iter = rb_entry(parent, struct hist_entry, rb_node);
+
+		if (hist_entry__sort(he, iter) > 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&he->rb_node, parent, p);
+	rb_insert_color(&he->rb_node, entries);
+}
+
+void hists__output_resort(struct hists *hists, struct ui_progress *prog)
+{
+	struct rb_root *root;
+	struct rb_node *next;
+	struct hist_entry *n;
+	u64 min_callchain_hits;
+
+	min_callchain_hits = hists->stats.total_period * (callchain_param.min_percent / 100);
+
+	if (sort__need_collapse)
+		root = &hists->entries_collapsed;
+	else
+		root = hists->entries_in;
+
+	next = rb_first(root);
+	hists->entries = RB_ROOT;
+
+	hists__reset_stats(hists);
+	hists__reset_col_len(hists);
+
+	while (next) {
+		n = rb_entry(next, struct hist_entry, rb_node_in);
+		next = rb_next(&n->rb_node_in);
+
+		__hists__insert_output_entry(&hists->entries, n, min_callchain_hits);
+		hists__inc_stats(hists, n);
+
+		if (!n->filtered)
+			hists__calc_col_len(hists, n);
+
+		if (prog)
+			ui_progress__update(prog, 1);
+	}
+}
+
+static void hists__remove_entry_filter(struct hists *hists, struct hist_entry *h,
+				       enum hist_filter filter)
+{
+	h->filtered &= ~(1 << filter);
+	if (h->filtered)
+		return;
+
+	/* force fold unfiltered entry for simplicity */
+	h->ms.unfolded = false;
+	h->row_offset = 0;
+	h->nr_rows = 0;
+
+	hists->stats.nr_non_filtered_samples += h->stat.nr_events;
+
+	hists__inc_filter_stats(hists, h);
+	hists__calc_col_len(hists, h);
+}
+
+
+static bool hists__filter_entry_by_dso(struct hists *hists,
+				       struct hist_entry *he)
+{
+	if (hists->dso_filter != NULL &&
+	    (he->ms.map == NULL || he->ms.map->dso != hists->dso_filter)) {
+		he->filtered |= (1 << HIST_FILTER__DSO);
+		return true;
+	}
+
+	return false;
+}
+
+void hists__filter_by_dso(struct hists *hists)
+{
+	struct rb_node *nd;
+
+	hists->stats.nr_non_filtered_samples = 0;
+
+	hists__reset_filter_stats(hists);
+	hists__reset_col_len(hists);
+
+	for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+		if (symbol_conf.exclude_other && !h->parent)
+			continue;
+
+		if (hists__filter_entry_by_dso(hists, h))
+			continue;
+
+		hists__remove_entry_filter(hists, h, HIST_FILTER__DSO);
+	}
+}
+
+static bool hists__filter_entry_by_thread(struct hists *hists,
+					  struct hist_entry *he)
+{
+	if (hists->thread_filter != NULL &&
+	    he->thread != hists->thread_filter) {
+		he->filtered |= (1 << HIST_FILTER__THREAD);
+		return true;
+	}
+
+	return false;
+}
+
+void hists__filter_by_thread(struct hists *hists)
+{
+	struct rb_node *nd;
+
+	hists->stats.nr_non_filtered_samples = 0;
+
+	hists__reset_filter_stats(hists);
+	hists__reset_col_len(hists);
+
+	for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+		if (hists__filter_entry_by_thread(hists, h))
+			continue;
+
+		hists__remove_entry_filter(hists, h, HIST_FILTER__THREAD);
+	}
+}
+
+static bool hists__filter_entry_by_symbol(struct hists *hists,
+					  struct hist_entry *he)
+{
+	if (hists->symbol_filter_str != NULL &&
+	    (!he->ms.sym || strstr(he->ms.sym->name,
+				   hists->symbol_filter_str) == NULL)) {
+		he->filtered |= (1 << HIST_FILTER__SYMBOL);
+		return true;
+	}
+
+	return false;
+}
+
+void hists__filter_by_symbol(struct hists *hists)
+{
+	struct rb_node *nd;
+
+	hists->stats.nr_non_filtered_samples = 0;
+
+	hists__reset_filter_stats(hists);
+	hists__reset_col_len(hists);
+
+	for (nd = rb_first(&hists->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *h = rb_entry(nd, struct hist_entry, rb_node);
+
+		if (hists__filter_entry_by_symbol(hists, h))
+			continue;
+
+		hists__remove_entry_filter(hists, h, HIST_FILTER__SYMBOL);
+	}
+}
+
+void events_stats__inc(struct events_stats *stats, u32 type)
+{
+	++stats->nr_events[0];
+	++stats->nr_events[type];
+}
+
+void hists__inc_nr_events(struct hists *hists, u32 type)
+{
+	events_stats__inc(&hists->stats, type);
+}
+
+void hists__inc_nr_samples(struct hists *hists, bool filtered)
+{
+	events_stats__inc(&hists->stats, PERF_RECORD_SAMPLE);
+	if (!filtered)
+		hists->stats.nr_non_filtered_samples++;
+}
+
+static struct hist_entry *hists__add_dummy_entry(struct hists *hists,
+						 struct hist_entry *pair)
+{
+	struct rb_root *root;
+	struct rb_node **p;
+	struct rb_node *parent = NULL;
+	struct hist_entry *he;
+	int64_t cmp;
+
+	if (sort__need_collapse)
+		root = &hists->entries_collapsed;
+	else
+		root = hists->entries_in;
+
+	p = &root->rb_node;
+
+	while (*p != NULL) {
+		parent = *p;
+		he = rb_entry(parent, struct hist_entry, rb_node_in);
+
+		cmp = hist_entry__collapse(he, pair);
+
+		if (!cmp)
+			goto out;
+
+		if (cmp < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	he = hist_entry__new(pair, true);
+	if (he) {
+		memset(&he->stat, 0, sizeof(he->stat));
+		he->hists = hists;
+		rb_link_node(&he->rb_node_in, parent, p);
+		rb_insert_color(&he->rb_node_in, root);
+		hists__inc_stats(hists, he);
+		he->dummy = true;
+	}
+out:
+	return he;
+}
+
+static struct hist_entry *hists__find_entry(struct hists *hists,
+					    struct hist_entry *he)
+{
+	struct rb_node *n;
+
+	if (sort__need_collapse)
+		n = hists->entries_collapsed.rb_node;
+	else
+		n = hists->entries_in->rb_node;
+
+	while (n) {
+		struct hist_entry *iter = rb_entry(n, struct hist_entry, rb_node_in);
+		int64_t cmp = hist_entry__collapse(iter, he);
+
+		if (cmp < 0)
+			n = n->rb_left;
+		else if (cmp > 0)
+			n = n->rb_right;
+		else
+			return iter;
+	}
+
+	return NULL;
+}
+
+/*
+ * Look for pairs to link to the leader buckets (hist_entries):
+ */
+void hists__match(struct hists *leader, struct hists *other)
+{
+	struct rb_root *root;
+	struct rb_node *nd;
+	struct hist_entry *pos, *pair;
+
+	if (sort__need_collapse)
+		root = &leader->entries_collapsed;
+	else
+		root = leader->entries_in;
+
+	for (nd = rb_first(root); nd; nd = rb_next(nd)) {
+		pos  = rb_entry(nd, struct hist_entry, rb_node_in);
+		pair = hists__find_entry(other, pos);
+
+		if (pair)
+			hist_entry__add_pair(pair, pos);
+	}
+}
+
+/*
+ * Look for entries in the other hists that are not present in the leader, if
+ * we find them, just add a dummy entry on the leader hists, with period=0,
+ * nr_events=0, to serve as the list header.
+ */
+int hists__link(struct hists *leader, struct hists *other)
+{
+	struct rb_root *root;
+	struct rb_node *nd;
+	struct hist_entry *pos, *pair;
+
+	if (sort__need_collapse)
+		root = &other->entries_collapsed;
+	else
+		root = other->entries_in;
+
+	for (nd = rb_first(root); nd; nd = rb_next(nd)) {
+		pos = rb_entry(nd, struct hist_entry, rb_node_in);
+
+		if (!hist_entry__has_pairs(pos)) {
+			pair = hists__add_dummy_entry(leader, pos);
+			if (pair == NULL)
+				return -1;
+			hist_entry__add_pair(pos, pair);
+		}
+	}
+
+	return 0;
+}
+
+
+size_t perf_evlist__fprintf_nr_events(struct perf_evlist *evlist, FILE *fp)
+{
+	struct perf_evsel *pos;
+	size_t ret = 0;
+
+	evlist__for_each(evlist, pos) {
+		ret += fprintf(fp, "%s stats:\n", perf_evsel__name(pos));
+		ret += events_stats__fprintf(&evsel__hists(pos)->stats, fp);
+	}
+
+	return ret;
+}
+
+
+u64 hists__total_period(struct hists *hists)
+{
+	return symbol_conf.filter_relative ? hists->stats.total_non_filtered_period :
+		hists->stats.total_period;
+}
+
+int parse_filter_percentage(const struct option *opt __maybe_unused,
+			    const char *arg, int unset __maybe_unused)
+{
+	if (!strcmp(arg, "relative"))
+		symbol_conf.filter_relative = true;
+	else if (!strcmp(arg, "absolute"))
+		symbol_conf.filter_relative = false;
+	else
+		return -1;
+
+	return 0;
+}
+
+int perf_hist_config(const char *var, const char *value)
+{
+	if (!strcmp(var, "hist.percentage"))
+		return parse_filter_percentage(NULL, value, 0);
+
+	return 0;
+}
+
+static int hists_evsel__init(struct perf_evsel *evsel)
+{
+	struct hists *hists = evsel__hists(evsel);
+
+	memset(hists, 0, sizeof(*hists));
+	hists->entries_in_array[0] = hists->entries_in_array[1] = RB_ROOT;
+	hists->entries_in = &hists->entries_in_array[0];
+	hists->entries_collapsed = RB_ROOT;
+	hists->entries = RB_ROOT;
+	pthread_mutex_init(&hists->lock, NULL);
+	return 0;
+}
+
+/*
+ * XXX We probably need a hists_evsel__exit() to free the hist_entries
+ * stored in the rbtree...
+ */
+
+int hists__init(void)
+{
+	int err = perf_evsel__object_config(sizeof(struct hists_evsel),
+					    hists_evsel__init, NULL);
+	if (err)
+		fputs("FATAL ERROR: Couldn't setup hists class\n", stderr);
+
+	return err;
+}
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
new file mode 100644
index 000000000..9f31b89a5
--- /dev/null
+++ b/tools/perf/util/hist.h
@@ -0,0 +1,358 @@
+#ifndef __PERF_HIST_H
+#define __PERF_HIST_H
+
+#include <linux/types.h>
+#include <pthread.h>
+#include "callchain.h"
+#include "evsel.h"
+#include "header.h"
+#include "color.h"
+#include "ui/progress.h"
+
+struct hist_entry;
+struct addr_location;
+struct symbol;
+
+enum hist_filter {
+	HIST_FILTER__DSO,
+	HIST_FILTER__THREAD,
+	HIST_FILTER__PARENT,
+	HIST_FILTER__SYMBOL,
+	HIST_FILTER__GUEST,
+	HIST_FILTER__HOST,
+};
+
+enum hist_column {
+	HISTC_SYMBOL,
+	HISTC_DSO,
+	HISTC_THREAD,
+	HISTC_COMM,
+	HISTC_PARENT,
+	HISTC_CPU,
+	HISTC_SRCLINE,
+	HISTC_MISPREDICT,
+	HISTC_IN_TX,
+	HISTC_ABORT,
+	HISTC_SYMBOL_FROM,
+	HISTC_SYMBOL_TO,
+	HISTC_DSO_FROM,
+	HISTC_DSO_TO,
+	HISTC_LOCAL_WEIGHT,
+	HISTC_GLOBAL_WEIGHT,
+	HISTC_MEM_DADDR_SYMBOL,
+	HISTC_MEM_DADDR_DSO,
+	HISTC_MEM_LOCKED,
+	HISTC_MEM_TLB,
+	HISTC_MEM_LVL,
+	HISTC_MEM_SNOOP,
+	HISTC_MEM_DCACHELINE,
+	HISTC_TRANSACTION,
+	HISTC_NR_COLS, /* Last entry */
+};
+
+struct thread;
+struct dso;
+
+struct hists {
+	struct rb_root		entries_in_array[2];
+	struct rb_root		*entries_in;
+	struct rb_root		entries;
+	struct rb_root		entries_collapsed;
+	u64			nr_entries;
+	u64			nr_non_filtered_entries;
+	struct thread		*thread_filter;
+	const struct dso	*dso_filter;
+	const char		*uid_filter_str;
+	const char		*symbol_filter_str;
+	pthread_mutex_t		lock;
+	struct events_stats	stats;
+	u64			event_stream;
+	u16			col_len[HISTC_NR_COLS];
+};
+
+struct hist_entry_iter;
+
+struct hist_iter_ops {
+	int (*prepare_entry)(struct hist_entry_iter *, struct addr_location *);
+	int (*add_single_entry)(struct hist_entry_iter *, struct addr_location *);
+	int (*next_entry)(struct hist_entry_iter *, struct addr_location *);
+	int (*add_next_entry)(struct hist_entry_iter *, struct addr_location *);
+	int (*finish_entry)(struct hist_entry_iter *, struct addr_location *);
+};
+
+struct hist_entry_iter {
+	int total;
+	int curr;
+
+	bool hide_unresolved;
+
+	struct perf_evsel *evsel;
+	struct perf_sample *sample;
+	struct hist_entry *he;
+	struct symbol *parent;
+	void *priv;
+
+	const struct hist_iter_ops *ops;
+	/* user-defined callback function (optional) */
+	int (*add_entry_cb)(struct hist_entry_iter *iter,
+			    struct addr_location *al, bool single, void *arg);
+};
+
+extern const struct hist_iter_ops hist_iter_normal;
+extern const struct hist_iter_ops hist_iter_branch;
+extern const struct hist_iter_ops hist_iter_mem;
+extern const struct hist_iter_ops hist_iter_cumulative;
+
+struct hist_entry *__hists__add_entry(struct hists *hists,
+				      struct addr_location *al,
+				      struct symbol *parent,
+				      struct branch_info *bi,
+				      struct mem_info *mi, u64 period,
+				      u64 weight, u64 transaction,
+				      bool sample_self);
+int hist_entry_iter__add(struct hist_entry_iter *iter, struct addr_location *al,
+			 struct perf_evsel *evsel, struct perf_sample *sample,
+			 int max_stack_depth, void *arg);
+
+int64_t hist_entry__cmp(struct hist_entry *left, struct hist_entry *right);
+int64_t hist_entry__collapse(struct hist_entry *left, struct hist_entry *right);
+int hist_entry__transaction_len(void);
+int hist_entry__sort_snprintf(struct hist_entry *he, char *bf, size_t size,
+			      struct hists *hists);
+void hist_entry__delete(struct hist_entry *he);
+
+void hists__output_resort(struct hists *hists, struct ui_progress *prog);
+void hists__collapse_resort(struct hists *hists, struct ui_progress *prog);
+
+void hists__decay_entries(struct hists *hists, bool zap_user, bool zap_kernel);
+void hists__delete_entries(struct hists *hists);
+void hists__output_recalc_col_len(struct hists *hists, int max_rows);
+
+u64 hists__total_period(struct hists *hists);
+void hists__reset_stats(struct hists *hists);
+void hists__inc_stats(struct hists *hists, struct hist_entry *h);
+void hists__inc_nr_events(struct hists *hists, u32 type);
+void hists__inc_nr_samples(struct hists *hists, bool filtered);
+void events_stats__inc(struct events_stats *stats, u32 type);
+size_t events_stats__fprintf(struct events_stats *stats, FILE *fp);
+
+size_t hists__fprintf(struct hists *hists, bool show_header, int max_rows,
+		      int max_cols, float min_pcnt, FILE *fp);
+size_t perf_evlist__fprintf_nr_events(struct perf_evlist *evlist, FILE *fp);
+
+void hists__filter_by_dso(struct hists *hists);
+void hists__filter_by_thread(struct hists *hists);
+void hists__filter_by_symbol(struct hists *hists);
+
+static inline bool hists__has_filter(struct hists *hists)
+{
+	return hists->thread_filter || hists->dso_filter ||
+		hists->symbol_filter_str;
+}
+
+u16 hists__col_len(struct hists *hists, enum hist_column col);
+void hists__set_col_len(struct hists *hists, enum hist_column col, u16 len);
+bool hists__new_col_len(struct hists *hists, enum hist_column col, u16 len);
+void hists__reset_col_len(struct hists *hists);
+void hists__calc_col_len(struct hists *hists, struct hist_entry *he);
+
+void hists__match(struct hists *leader, struct hists *other);
+int hists__link(struct hists *leader, struct hists *other);
+
+struct hists_evsel {
+	struct perf_evsel evsel;
+	struct hists	  hists;
+};
+
+static inline struct perf_evsel *hists_to_evsel(struct hists *hists)
+{
+	struct hists_evsel *hevsel = container_of(hists, struct hists_evsel, hists);
+	return &hevsel->evsel;
+}
+
+static inline struct hists *evsel__hists(struct perf_evsel *evsel)
+{
+	struct hists_evsel *hevsel = (struct hists_evsel *)evsel;
+	return &hevsel->hists;
+}
+
+int hists__init(void);
+
+struct perf_hpp {
+	char *buf;
+	size_t size;
+	const char *sep;
+	void *ptr;
+};
+
+struct perf_hpp_fmt {
+	const char *name;
+	int (*header)(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+		      struct perf_evsel *evsel);
+	int (*width)(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+		     struct perf_evsel *evsel);
+	int (*color)(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+		     struct hist_entry *he);
+	int (*entry)(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+		     struct hist_entry *he);
+	int64_t (*cmp)(struct perf_hpp_fmt *fmt,
+		       struct hist_entry *a, struct hist_entry *b);
+	int64_t (*collapse)(struct perf_hpp_fmt *fmt,
+			    struct hist_entry *a, struct hist_entry *b);
+	int64_t (*sort)(struct perf_hpp_fmt *fmt,
+			struct hist_entry *a, struct hist_entry *b);
+
+	struct list_head list;
+	struct list_head sort_list;
+	bool elide;
+	int len;
+	int user_len;
+};
+
+extern struct list_head perf_hpp__list;
+extern struct list_head perf_hpp__sort_list;
+
+#define perf_hpp__for_each_format(format) \
+	list_for_each_entry(format, &perf_hpp__list, list)
+
+#define perf_hpp__for_each_format_safe(format, tmp)	\
+	list_for_each_entry_safe(format, tmp, &perf_hpp__list, list)
+
+#define perf_hpp__for_each_sort_list(format) \
+	list_for_each_entry(format, &perf_hpp__sort_list, sort_list)
+
+#define perf_hpp__for_each_sort_list_safe(format, tmp)	\
+	list_for_each_entry_safe(format, tmp, &perf_hpp__sort_list, sort_list)
+
+extern struct perf_hpp_fmt perf_hpp__format[];
+
+enum {
+	/* Matches perf_hpp__format array. */
+	PERF_HPP__OVERHEAD,
+	PERF_HPP__OVERHEAD_SYS,
+	PERF_HPP__OVERHEAD_US,
+	PERF_HPP__OVERHEAD_GUEST_SYS,
+	PERF_HPP__OVERHEAD_GUEST_US,
+	PERF_HPP__OVERHEAD_ACC,
+	PERF_HPP__SAMPLES,
+	PERF_HPP__PERIOD,
+
+	PERF_HPP__MAX_INDEX
+};
+
+void perf_hpp__init(void);
+void perf_hpp__column_register(struct perf_hpp_fmt *format);
+void perf_hpp__column_unregister(struct perf_hpp_fmt *format);
+void perf_hpp__column_enable(unsigned col);
+void perf_hpp__column_disable(unsigned col);
+void perf_hpp__cancel_cumulate(void);
+
+void perf_hpp__register_sort_field(struct perf_hpp_fmt *format);
+void perf_hpp__setup_output_field(void);
+void perf_hpp__reset_output_field(void);
+void perf_hpp__append_sort_keys(void);
+
+bool perf_hpp__is_sort_entry(struct perf_hpp_fmt *format);
+bool perf_hpp__same_sort_entry(struct perf_hpp_fmt *a, struct perf_hpp_fmt *b);
+
+static inline bool perf_hpp__should_skip(struct perf_hpp_fmt *format)
+{
+	return format->elide;
+}
+
+void perf_hpp__reset_width(struct perf_hpp_fmt *fmt, struct hists *hists);
+void perf_hpp__reset_sort_width(struct perf_hpp_fmt *fmt, struct hists *hists);
+void perf_hpp__set_user_width(const char *width_list_str);
+
+typedef u64 (*hpp_field_fn)(struct hist_entry *he);
+typedef int (*hpp_callback_fn)(struct perf_hpp *hpp, bool front);
+typedef int (*hpp_snprint_fn)(struct perf_hpp *hpp, const char *fmt, ...);
+
+int hpp__fmt(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+	     struct hist_entry *he, hpp_field_fn get_field,
+	     const char *fmtstr, hpp_snprint_fn print_fn, bool fmt_percent);
+int hpp__fmt_acc(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+		 struct hist_entry *he, hpp_field_fn get_field,
+		 const char *fmtstr, hpp_snprint_fn print_fn, bool fmt_percent);
+
+static inline void advance_hpp(struct perf_hpp *hpp, int inc)
+{
+	hpp->buf  += inc;
+	hpp->size -= inc;
+}
+
+static inline size_t perf_hpp__use_color(void)
+{
+	return !symbol_conf.field_sep;
+}
+
+static inline size_t perf_hpp__color_overhead(void)
+{
+	return perf_hpp__use_color() ?
+	       (COLOR_MAXLEN + sizeof(PERF_COLOR_RESET)) * PERF_HPP__MAX_INDEX
+	       : 0;
+}
+
+struct perf_evlist;
+
+struct hist_browser_timer {
+	void (*timer)(void *arg);
+	void *arg;
+	int refresh;
+};
+
+#ifdef HAVE_SLANG_SUPPORT
+#include "../ui/keysyms.h"
+int map_symbol__tui_annotate(struct map_symbol *ms, struct perf_evsel *evsel,
+			     struct hist_browser_timer *hbt);
+
+int hist_entry__tui_annotate(struct hist_entry *he, struct perf_evsel *evsel,
+			     struct hist_browser_timer *hbt);
+
+int perf_evlist__tui_browse_hists(struct perf_evlist *evlist, const char *help,
+				  struct hist_browser_timer *hbt,
+				  float min_pcnt,
+				  struct perf_session_env *env);
+int script_browse(const char *script_opt);
+#else
+static inline
+int perf_evlist__tui_browse_hists(struct perf_evlist *evlist __maybe_unused,
+				  const char *help __maybe_unused,
+				  struct hist_browser_timer *hbt __maybe_unused,
+				  float min_pcnt __maybe_unused,
+				  struct perf_session_env *env __maybe_unused)
+{
+	return 0;
+}
+static inline int map_symbol__tui_annotate(struct map_symbol *ms __maybe_unused,
+					   struct perf_evsel *evsel __maybe_unused,
+					   struct hist_browser_timer *hbt __maybe_unused)
+{
+	return 0;
+}
+
+static inline int hist_entry__tui_annotate(struct hist_entry *he __maybe_unused,
+					   struct perf_evsel *evsel __maybe_unused,
+					   struct hist_browser_timer *hbt __maybe_unused)
+{
+	return 0;
+}
+
+static inline int script_browse(const char *script_opt __maybe_unused)
+{
+	return 0;
+}
+
+#define K_LEFT  -1000
+#define K_RIGHT -2000
+#define K_SWITCH_INPUT_DATA -3000
+#endif
+
+unsigned int hists__sort_list_width(struct hists *hists);
+
+struct option;
+int parse_filter_percentage(const struct option *opt __maybe_unused,
+			    const char *arg, int unset __maybe_unused);
+int perf_hist_config(const char *var, const char *value);
+
+#endif	/* __PERF_HIST_H */
diff --git a/tools/perf/util/include/asm/alternative-asm.h b/tools/perf/util/include/asm/alternative-asm.h
new file mode 100644
index 000000000..3a3a0f164
--- /dev/null
+++ b/tools/perf/util/include/asm/alternative-asm.h
@@ -0,0 +1,9 @@
+#ifndef _PERF_ASM_ALTERNATIVE_ASM_H
+#define _PERF_ASM_ALTERNATIVE_ASM_H
+
+/* Just disable it so we can build arch/x86/lib/memcpy_64.S for perf bench: */
+
+#define altinstruction_entry #
+#define ALTERNATIVE_2 #
+
+#endif
diff --git a/tools/perf/util/include/asm/asm-offsets.h b/tools/perf/util/include/asm/asm-offsets.h
new file mode 100644
index 000000000..ed5389425
--- /dev/null
+++ b/tools/perf/util/include/asm/asm-offsets.h
@@ -0,0 +1 @@
+/* stub */
diff --git a/tools/perf/util/include/asm/byteorder.h b/tools/perf/util/include/asm/byteorder.h
new file mode 100644
index 000000000..2a9bdc066
--- /dev/null
+++ b/tools/perf/util/include/asm/byteorder.h
@@ -0,0 +1,2 @@
+#include <asm/types.h>
+#include "../../../../include/uapi/linux/swab.h"
diff --git a/tools/perf/util/include/asm/cpufeature.h b/tools/perf/util/include/asm/cpufeature.h
new file mode 100644
index 000000000..acffd5e4d
--- /dev/null
+++ b/tools/perf/util/include/asm/cpufeature.h
@@ -0,0 +1,9 @@
+
+#ifndef PERF_CPUFEATURE_H
+#define PERF_CPUFEATURE_H
+
+/* cpufeature.h ... dummy header file for including arch/x86/lib/memcpy_64.S */
+
+#define X86_FEATURE_REP_GOOD 0
+
+#endif	/* PERF_CPUFEATURE_H */
diff --git a/tools/perf/util/include/asm/dwarf2.h b/tools/perf/util/include/asm/dwarf2.h
new file mode 100644
index 000000000..afe38199e
--- /dev/null
+++ b/tools/perf/util/include/asm/dwarf2.h
@@ -0,0 +1,13 @@
+
+#ifndef PERF_DWARF2_H
+#define PERF_DWARF2_H
+
+/* dwarf2.h ... dummy header file for including arch/x86/lib/mem{cpy,set}_64.S */
+
+#define CFI_STARTPROC
+#define CFI_ENDPROC
+#define CFI_REMEMBER_STATE
+#define CFI_RESTORE_STATE
+
+#endif	/* PERF_DWARF2_H */
+
diff --git a/tools/perf/util/include/asm/swab.h b/tools/perf/util/include/asm/swab.h
new file mode 100644
index 000000000..ed5389425
--- /dev/null
+++ b/tools/perf/util/include/asm/swab.h
@@ -0,0 +1 @@
+/* stub */
diff --git a/tools/perf/util/include/asm/system.h b/tools/perf/util/include/asm/system.h
new file mode 100644
index 000000000..710cecca9
--- /dev/null
+++ b/tools/perf/util/include/asm/system.h
@@ -0,0 +1 @@
+/* Empty */
diff --git a/tools/perf/util/include/asm/uaccess.h b/tools/perf/util/include/asm/uaccess.h
new file mode 100644
index 000000000..d0f72b8fc
--- /dev/null
+++ b/tools/perf/util/include/asm/uaccess.h
@@ -0,0 +1,14 @@
+#ifndef _PERF_ASM_UACCESS_H_
+#define _PERF_ASM_UACCESS_H_
+
+#define __get_user(src, dest)						\
+({									\
+	(src) = *dest;							\
+	0;								\
+})
+
+#define get_user	__get_user
+
+#define access_ok(type, addr, size)	1
+
+#endif
diff --git a/tools/perf/util/include/asm/unistd_32.h b/tools/perf/util/include/asm/unistd_32.h
new file mode 100644
index 000000000..8b1378917
--- /dev/null
+++ b/tools/perf/util/include/asm/unistd_32.h
@@ -0,0 +1 @@
+
diff --git a/tools/perf/util/include/asm/unistd_64.h b/tools/perf/util/include/asm/unistd_64.h
new file mode 100644
index 000000000..8b1378917
--- /dev/null
+++ b/tools/perf/util/include/asm/unistd_64.h
@@ -0,0 +1 @@
+
diff --git a/tools/perf/util/include/dwarf-regs.h b/tools/perf/util/include/dwarf-regs.h
new file mode 100644
index 000000000..8f149655f
--- /dev/null
+++ b/tools/perf/util/include/dwarf-regs.h
@@ -0,0 +1,8 @@
+#ifndef _PERF_DWARF_REGS_H_
+#define _PERF_DWARF_REGS_H_
+
+#ifdef HAVE_DWARF_SUPPORT
+const char *get_arch_regstr(unsigned int n);
+#endif
+
+#endif
diff --git a/tools/perf/util/include/linux/bitmap.h b/tools/perf/util/include/linux/bitmap.h
new file mode 100644
index 000000000..40bd21488
--- /dev/null
+++ b/tools/perf/util/include/linux/bitmap.h
@@ -0,0 +1,66 @@
+#ifndef _PERF_BITOPS_H
+#define _PERF_BITOPS_H
+
+#include <string.h>
+#include <linux/bitops.h>
+
+#define DECLARE_BITMAP(name,bits) \
+	unsigned long name[BITS_TO_LONGS(bits)]
+
+int __bitmap_weight(const unsigned long *bitmap, int bits);
+void __bitmap_or(unsigned long *dst, const unsigned long *bitmap1,
+		 const unsigned long *bitmap2, int bits);
+
+#define BITMAP_LAST_WORD_MASK(nbits)					\
+(									\
+	((nbits) % BITS_PER_LONG) ?					\
+		(1UL<<((nbits) % BITS_PER_LONG))-1 : ~0UL		\
+)
+
+#define small_const_nbits(nbits) \
+	(__builtin_constant_p(nbits) && (nbits) <= BITS_PER_LONG)
+
+static inline void bitmap_zero(unsigned long *dst, int nbits)
+{
+	if (small_const_nbits(nbits))
+		*dst = 0UL;
+	else {
+		int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
+		memset(dst, 0, len);
+	}
+}
+
+static inline int bitmap_weight(const unsigned long *src, int nbits)
+{
+	if (small_const_nbits(nbits))
+		return hweight_long(*src & BITMAP_LAST_WORD_MASK(nbits));
+	return __bitmap_weight(src, nbits);
+}
+
+static inline void bitmap_or(unsigned long *dst, const unsigned long *src1,
+			     const unsigned long *src2, int nbits)
+{
+	if (small_const_nbits(nbits))
+		*dst = *src1 | *src2;
+	else
+		__bitmap_or(dst, src1, src2, nbits);
+}
+
+/**
+ * test_and_set_bit - Set a bit and return its old value
+ * @nr: Bit to set
+ * @addr: Address to count from
+ */
+static inline int test_and_set_bit(int nr, unsigned long *addr)
+{
+	unsigned long mask = BIT_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BIT_WORD(nr);
+	unsigned long old;
+
+	old = *p;
+	*p = old | mask;
+
+	return (old & mask) != 0;
+}
+
+#endif /* _PERF_BITOPS_H */
diff --git a/tools/perf/util/include/linux/const.h b/tools/perf/util/include/linux/const.h
new file mode 100644
index 000000000..c10a35e1a
--- /dev/null
+++ b/tools/perf/util/include/linux/const.h
@@ -0,0 +1 @@
+#include "../../../../include/uapi/linux/const.h"
diff --git a/tools/perf/util/include/linux/ctype.h b/tools/perf/util/include/linux/ctype.h
new file mode 100644
index 000000000..a53d4ee1e
--- /dev/null
+++ b/tools/perf/util/include/linux/ctype.h
@@ -0,0 +1 @@
+#include "../util.h"
diff --git a/tools/perf/util/include/linux/kernel.h b/tools/perf/util/include/linux/kernel.h
new file mode 100644
index 000000000..09e8e7aea
--- /dev/null
+++ b/tools/perf/util/include/linux/kernel.h
@@ -0,0 +1,107 @@
+#ifndef PERF_LINUX_KERNEL_H_
+#define PERF_LINUX_KERNEL_H_
+
+#include <stdarg.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <assert.h>
+
+#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
+
+#define PERF_ALIGN(x, a)	__PERF_ALIGN_MASK(x, (typeof(x))(a)-1)
+#define __PERF_ALIGN_MASK(x, mask)	(((x)+(mask))&~(mask))
+
+#ifndef offsetof
+#define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER)
+#endif
+
+#ifndef container_of
+/**
+ * container_of - cast a member of a structure out to the containing structure
+ * @ptr:	the pointer to the member.
+ * @type:	the type of the container struct this is embedded in.
+ * @member:	the name of the member within the struct.
+ *
+ */
+#define container_of(ptr, type, member) ({			\
+	const typeof(((type *)0)->member) * __mptr = (ptr);	\
+	(type *)((char *)__mptr - offsetof(type, member)); })
+#endif
+
+#define BUILD_BUG_ON_ZERO(e) (sizeof(struct { int:-!!(e); }))
+
+#ifndef max
+#define max(x, y) ({				\
+	typeof(x) _max1 = (x);			\
+	typeof(y) _max2 = (y);			\
+	(void) (&_max1 == &_max2);		\
+	_max1 > _max2 ? _max1 : _max2; })
+#endif
+
+#ifndef min
+#define min(x, y) ({				\
+	typeof(x) _min1 = (x);			\
+	typeof(y) _min2 = (y);			\
+	(void) (&_min1 == &_min2);		\
+	_min1 < _min2 ? _min1 : _min2; })
+#endif
+
+#ifndef roundup
+#define roundup(x, y) (                                \
+{                                                      \
+	const typeof(y) __y = y;		       \
+	(((x) + (__y - 1)) / __y) * __y;	       \
+}                                                      \
+)
+#endif
+
+#ifndef BUG_ON
+#ifdef NDEBUG
+#define BUG_ON(cond) do { if (cond) {} } while (0)
+#else
+#define BUG_ON(cond) assert(!(cond))
+#endif
+#endif
+
+/*
+ * Both need more care to handle endianness
+ * (Don't use bitmap_copy_le() for now)
+ */
+#define cpu_to_le64(x)	(x)
+#define cpu_to_le32(x)	(x)
+
+static inline int
+vscnprintf(char *buf, size_t size, const char *fmt, va_list args)
+{
+	int i;
+	ssize_t ssize = size;
+
+	i = vsnprintf(buf, size, fmt, args);
+
+	return (i >= ssize) ? (ssize - 1) : i;
+}
+
+static inline int scnprintf(char * buf, size_t size, const char * fmt, ...)
+{
+	va_list args;
+	ssize_t ssize = size;
+	int i;
+
+	va_start(args, fmt);
+	i = vsnprintf(buf, size, fmt, args);
+	va_end(args);
+
+	return (i >= ssize) ? (ssize - 1) : i;
+}
+
+/*
+ * This looks more complex than it should be. But we need to
+ * get the type for the ~ right in round_down (it needs to be
+ * as wide as the result!), and we want to evaluate the macro
+ * arguments just once each.
+ */
+#define __round_mask(x, y) ((__typeof__(x))((y)-1))
+#define round_up(x, y) ((((x)-1) | __round_mask(x, y))+1)
+#define round_down(x, y) ((x) & ~__round_mask(x, y))
+
+#endif
diff --git a/tools/perf/util/include/linux/linkage.h b/tools/perf/util/include/linux/linkage.h
new file mode 100644
index 000000000..06387cffe
--- /dev/null
+++ b/tools/perf/util/include/linux/linkage.h
@@ -0,0 +1,13 @@
+
+#ifndef PERF_LINUX_LINKAGE_H_
+#define PERF_LINUX_LINKAGE_H_
+
+/* linkage.h ... for including arch/x86/lib/memcpy_64.S */
+
+#define ENTRY(name)				\
+	.globl name;				\
+	name:
+
+#define ENDPROC(name)
+
+#endif	/* PERF_LINUX_LINKAGE_H_ */
diff --git a/tools/perf/util/include/linux/list.h b/tools/perf/util/include/linux/list.h
new file mode 100644
index 000000000..76ddbc726
--- /dev/null
+++ b/tools/perf/util/include/linux/list.h
@@ -0,0 +1,29 @@
+#include <linux/kernel.h>
+#include <linux/types.h>
+
+#include "../../../../include/linux/list.h"
+
+#ifndef PERF_LIST_H
+#define PERF_LIST_H
+/**
+ * list_del_range - deletes range of entries from list.
+ * @begin: first element in the range to delete from the list.
+ * @end: last element in the range to delete from the list.
+ * Note: list_empty on the range of entries does not return true after this,
+ * the entries is in an undefined state.
+ */
+static inline void list_del_range(struct list_head *begin,
+				  struct list_head *end)
+{
+	begin->prev->next = end->next;
+	end->next->prev = begin->prev;
+}
+
+/**
+ * list_for_each_from	-	iterate over a list from one of its nodes
+ * @pos:  the &struct list_head to use as a loop cursor, from where to start
+ * @head: the head for your list.
+ */
+#define list_for_each_from(pos, head) \
+	for (; pos != (head); pos = pos->next)
+#endif
diff --git a/tools/perf/util/include/linux/poison.h b/tools/perf/util/include/linux/poison.h
new file mode 100644
index 000000000..fef6dbc9c
--- /dev/null
+++ b/tools/perf/util/include/linux/poison.h
@@ -0,0 +1 @@
+#include "../../../../include/linux/poison.h"
diff --git a/tools/perf/util/include/linux/rbtree.h b/tools/perf/util/include/linux/rbtree.h
new file mode 100644
index 000000000..2a030c5af
--- /dev/null
+++ b/tools/perf/util/include/linux/rbtree.h
@@ -0,0 +1,2 @@
+#include <stdbool.h>
+#include "../../../../include/linux/rbtree.h"
diff --git a/tools/perf/util/include/linux/rbtree_augmented.h b/tools/perf/util/include/linux/rbtree_augmented.h
new file mode 100644
index 000000000..9d6fcdf17
--- /dev/null
+++ b/tools/perf/util/include/linux/rbtree_augmented.h
@@ -0,0 +1,2 @@
+#include <stdbool.h>
+#include "../../../../include/linux/rbtree_augmented.h"
diff --git a/tools/perf/util/include/linux/string.h b/tools/perf/util/include/linux/string.h
new file mode 100644
index 000000000..6f19c548e
--- /dev/null
+++ b/tools/perf/util/include/linux/string.h
@@ -0,0 +1,3 @@
+#include <string.h>
+
+void *memdup(const void *src, size_t len);
diff --git a/tools/perf/util/intlist.c b/tools/perf/util/intlist.c
new file mode 100644
index 000000000..89715b64a
--- /dev/null
+++ b/tools/perf/util/intlist.c
@@ -0,0 +1,146 @@
+/*
+ * Based on intlist.c by:
+ * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Licensed under the GPLv2.
+ */
+
+#include <errno.h>
+#include <stdlib.h>
+#include <linux/compiler.h>
+
+#include "intlist.h"
+
+static struct rb_node *intlist__node_new(struct rblist *rblist __maybe_unused,
+					 const void *entry)
+{
+	int i = (int)((long)entry);
+	struct rb_node *rc = NULL;
+	struct int_node *node = malloc(sizeof(*node));
+
+	if (node != NULL) {
+		node->i = i;
+		node->priv = NULL;
+		rc = &node->rb_node;
+	}
+
+	return rc;
+}
+
+static void int_node__delete(struct int_node *ilist)
+{
+	free(ilist);
+}
+
+static void intlist__node_delete(struct rblist *rblist __maybe_unused,
+				 struct rb_node *rb_node)
+{
+	struct int_node *node = container_of(rb_node, struct int_node, rb_node);
+
+	int_node__delete(node);
+}
+
+static int intlist__node_cmp(struct rb_node *rb_node, const void *entry)
+{
+	int i = (int)((long)entry);
+	struct int_node *node = container_of(rb_node, struct int_node, rb_node);
+
+	return node->i - i;
+}
+
+int intlist__add(struct intlist *ilist, int i)
+{
+	return rblist__add_node(&ilist->rblist, (void *)((long)i));
+}
+
+void intlist__remove(struct intlist *ilist, struct int_node *node)
+{
+	rblist__remove_node(&ilist->rblist, &node->rb_node);
+}
+
+static struct int_node *__intlist__findnew(struct intlist *ilist,
+					   int i, bool create)
+{
+	struct int_node *node = NULL;
+	struct rb_node *rb_node;
+
+	if (ilist == NULL)
+		return NULL;
+
+	if (create)
+		rb_node = rblist__findnew(&ilist->rblist, (void *)((long)i));
+	else
+		rb_node = rblist__find(&ilist->rblist, (void *)((long)i));
+
+	if (rb_node)
+		node = container_of(rb_node, struct int_node, rb_node);
+
+	return node;
+}
+
+struct int_node *intlist__find(struct intlist *ilist, int i)
+{
+	return __intlist__findnew(ilist, i, false);
+}
+
+struct int_node *intlist__findnew(struct intlist *ilist, int i)
+{
+	return __intlist__findnew(ilist, i, true);
+}
+
+static int intlist__parse_list(struct intlist *ilist, const char *s)
+{
+	char *sep;
+	int err;
+
+	do {
+		long value = strtol(s, &sep, 10);
+		err = -EINVAL;
+		if (*sep != ',' && *sep != '\0')
+			break;
+		err = intlist__add(ilist, value);
+		if (err)
+			break;
+		s = sep + 1;
+	} while (*sep != '\0');
+
+	return err;
+}
+
+struct intlist *intlist__new(const char *slist)
+{
+	struct intlist *ilist = malloc(sizeof(*ilist));
+
+	if (ilist != NULL) {
+		rblist__init(&ilist->rblist);
+		ilist->rblist.node_cmp    = intlist__node_cmp;
+		ilist->rblist.node_new    = intlist__node_new;
+		ilist->rblist.node_delete = intlist__node_delete;
+
+		if (slist && intlist__parse_list(ilist, slist))
+			goto out_delete;
+	}
+
+	return ilist;
+out_delete:
+	intlist__delete(ilist);
+	return NULL;
+}
+
+void intlist__delete(struct intlist *ilist)
+{
+	if (ilist != NULL)
+		rblist__delete(&ilist->rblist);
+}
+
+struct int_node *intlist__entry(const struct intlist *ilist, unsigned int idx)
+{
+	struct int_node *node = NULL;
+	struct rb_node *rb_node;
+
+	rb_node = rblist__entry(&ilist->rblist, idx);
+	if (rb_node)
+		node = container_of(rb_node, struct int_node, rb_node);
+
+	return node;
+}
diff --git a/tools/perf/util/intlist.h b/tools/perf/util/intlist.h
new file mode 100644
index 000000000..aa6877d36
--- /dev/null
+++ b/tools/perf/util/intlist.h
@@ -0,0 +1,77 @@
+#ifndef __PERF_INTLIST_H
+#define __PERF_INTLIST_H
+
+#include <linux/rbtree.h>
+#include <stdbool.h>
+
+#include "rblist.h"
+
+struct int_node {
+	struct rb_node rb_node;
+	int i;
+	void *priv;
+};
+
+struct intlist {
+	struct rblist rblist;
+};
+
+struct intlist *intlist__new(const char *slist);
+void intlist__delete(struct intlist *ilist);
+
+void intlist__remove(struct intlist *ilist, struct int_node *in);
+int intlist__add(struct intlist *ilist, int i);
+
+struct int_node *intlist__entry(const struct intlist *ilist, unsigned int idx);
+struct int_node *intlist__find(struct intlist *ilist, int i);
+struct int_node *intlist__findnew(struct intlist *ilist, int i);
+
+static inline bool intlist__has_entry(struct intlist *ilist, int i)
+{
+	return intlist__find(ilist, i) != NULL;
+}
+
+static inline bool intlist__empty(const struct intlist *ilist)
+{
+	return rblist__empty(&ilist->rblist);
+}
+
+static inline unsigned int intlist__nr_entries(const struct intlist *ilist)
+{
+	return rblist__nr_entries(&ilist->rblist);
+}
+
+/* For intlist iteration */
+static inline struct int_node *intlist__first(struct intlist *ilist)
+{
+	struct rb_node *rn = rb_first(&ilist->rblist.entries);
+	return rn ? rb_entry(rn, struct int_node, rb_node) : NULL;
+}
+static inline struct int_node *intlist__next(struct int_node *in)
+{
+	struct rb_node *rn;
+	if (!in)
+		return NULL;
+	rn = rb_next(&in->rb_node);
+	return rn ? rb_entry(rn, struct int_node, rb_node) : NULL;
+}
+
+/**
+ * intlist_for_each      - iterate over a intlist
+ * @pos:	the &struct int_node to use as a loop cursor.
+ * @ilist:	the &struct intlist for loop.
+ */
+#define intlist__for_each(pos, ilist)	\
+	for (pos = intlist__first(ilist); pos; pos = intlist__next(pos))
+
+/**
+ * intlist_for_each_safe - iterate over a intlist safe against removal of
+ *                         int_node
+ * @pos:	the &struct int_node to use as a loop cursor.
+ * @n:		another &struct int_node to use as temporary storage.
+ * @ilist:	the &struct intlist for loop.
+ */
+#define intlist__for_each_safe(pos, n, ilist)	\
+	for (pos = intlist__first(ilist), n = intlist__next(pos); pos;\
+	     pos = n, n = intlist__next(n))
+#endif /* __PERF_INTLIST_H */
diff --git a/tools/perf/util/kvm-stat.h b/tools/perf/util/kvm-stat.h
new file mode 100644
index 000000000..ae825d4ec
--- /dev/null
+++ b/tools/perf/util/kvm-stat.h
@@ -0,0 +1,140 @@
+#ifndef __PERF_KVM_STAT_H
+#define __PERF_KVM_STAT_H
+
+#include "../perf.h"
+#include "evsel.h"
+#include "evlist.h"
+#include "session.h"
+#include "tool.h"
+#include "stat.h"
+
+struct event_key {
+	#define INVALID_KEY     (~0ULL)
+	u64 key;
+	int info;
+	struct exit_reasons_table *exit_reasons;
+};
+
+struct kvm_event_stats {
+	u64 time;
+	struct stats stats;
+};
+
+struct kvm_event {
+	struct list_head hash_entry;
+	struct rb_node rb;
+
+	struct event_key key;
+
+	struct kvm_event_stats total;
+
+	#define DEFAULT_VCPU_NUM 8
+	int max_vcpu;
+	struct kvm_event_stats *vcpu;
+};
+
+typedef int (*key_cmp_fun)(struct kvm_event*, struct kvm_event*, int);
+
+struct kvm_event_key {
+	const char *name;
+	key_cmp_fun key;
+};
+
+struct perf_kvm_stat;
+
+struct child_event_ops {
+	void (*get_key)(struct perf_evsel *evsel,
+			struct perf_sample *sample,
+			struct event_key *key);
+	const char *name;
+};
+
+struct kvm_events_ops {
+	bool (*is_begin_event)(struct perf_evsel *evsel,
+			       struct perf_sample *sample,
+			       struct event_key *key);
+	bool (*is_end_event)(struct perf_evsel *evsel,
+			     struct perf_sample *sample, struct event_key *key);
+	struct child_event_ops *child_ops;
+	void (*decode_key)(struct perf_kvm_stat *kvm, struct event_key *key,
+			   char *decode);
+	const char *name;
+};
+
+struct exit_reasons_table {
+	unsigned long exit_code;
+	const char *reason;
+};
+
+#define EVENTS_BITS		12
+#define EVENTS_CACHE_SIZE	(1UL << EVENTS_BITS)
+
+struct perf_kvm_stat {
+	struct perf_tool    tool;
+	struct record_opts  opts;
+	struct perf_evlist  *evlist;
+	struct perf_session *session;
+
+	const char *file_name;
+	const char *report_event;
+	const char *sort_key;
+	int trace_vcpu;
+
+	struct exit_reasons_table *exit_reasons;
+	const char *exit_reasons_isa;
+
+	struct kvm_events_ops *events_ops;
+	key_cmp_fun compare;
+	struct list_head kvm_events_cache[EVENTS_CACHE_SIZE];
+
+	u64 total_time;
+	u64 total_count;
+	u64 lost_events;
+	u64 duration;
+
+	struct intlist *pid_list;
+
+	struct rb_root result;
+
+	int timerfd;
+	unsigned int display_time;
+	bool live;
+	bool force;
+};
+
+struct kvm_reg_events_ops {
+	const char *name;
+	struct kvm_events_ops *ops;
+};
+
+void exit_event_get_key(struct perf_evsel *evsel,
+			struct perf_sample *sample,
+			struct event_key *key);
+bool exit_event_begin(struct perf_evsel *evsel,
+		      struct perf_sample *sample,
+		      struct event_key *key);
+bool exit_event_end(struct perf_evsel *evsel,
+		    struct perf_sample *sample,
+		    struct event_key *key);
+void exit_event_decode_key(struct perf_kvm_stat *kvm,
+			   struct event_key *key,
+			   char *decode);
+
+bool kvm_exit_event(struct perf_evsel *evsel);
+bool kvm_entry_event(struct perf_evsel *evsel);
+
+#define define_exit_reasons_table(name, symbols)	\
+	static struct exit_reasons_table name[] = {	\
+		symbols, { -1, NULL }			\
+	}
+
+/*
+ * arch specific callbacks and data structures
+ */
+int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid);
+
+extern const char * const kvm_events_tp[];
+extern struct kvm_reg_events_ops kvm_reg_events_ops[];
+extern const char * const kvm_skip_events[];
+
+#endif /* __PERF_KVM_STAT_H */
diff --git a/tools/perf/util/levenshtein.c b/tools/perf/util/levenshtein.c
new file mode 100644
index 000000000..e521d1516
--- /dev/null
+++ b/tools/perf/util/levenshtein.c
@@ -0,0 +1,84 @@
+#include "cache.h"
+#include "levenshtein.h"
+
+/*
+ * This function implements the Damerau-Levenshtein algorithm to
+ * calculate a distance between strings.
+ *
+ * Basically, it says how many letters need to be swapped, substituted,
+ * deleted from, or added to string1, at least, to get string2.
+ *
+ * The idea is to build a distance matrix for the substrings of both
+ * strings.  To avoid a large space complexity, only the last three rows
+ * are kept in memory (if swaps had the same or higher cost as one deletion
+ * plus one insertion, only two rows would be needed).
+ *
+ * At any stage, "i + 1" denotes the length of the current substring of
+ * string1 that the distance is calculated for.
+ *
+ * row2 holds the current row, row1 the previous row (i.e. for the substring
+ * of string1 of length "i"), and row0 the row before that.
+ *
+ * In other words, at the start of the big loop, row2[j + 1] contains the
+ * Damerau-Levenshtein distance between the substring of string1 of length
+ * "i" and the substring of string2 of length "j + 1".
+ *
+ * All the big loop does is determine the partial minimum-cost paths.
+ *
+ * It does so by calculating the costs of the path ending in characters
+ * i (in string1) and j (in string2), respectively, given that the last
+ * operation is a substition, a swap, a deletion, or an insertion.
+ *
+ * This implementation allows the costs to be weighted:
+ *
+ * - w (as in "sWap")
+ * - s (as in "Substitution")
+ * - a (for insertion, AKA "Add")
+ * - d (as in "Deletion")
+ *
+ * Note that this algorithm calculates a distance _iff_ d == a.
+ */
+int levenshtein(const char *string1, const char *string2,
+		int w, int s, int a, int d)
+{
+	int len1 = strlen(string1), len2 = strlen(string2);
+	int *row0 = malloc(sizeof(int) * (len2 + 1));
+	int *row1 = malloc(sizeof(int) * (len2 + 1));
+	int *row2 = malloc(sizeof(int) * (len2 + 1));
+	int i, j;
+
+	for (j = 0; j <= len2; j++)
+		row1[j] = j * a;
+	for (i = 0; i < len1; i++) {
+		int *dummy;
+
+		row2[0] = (i + 1) * d;
+		for (j = 0; j < len2; j++) {
+			/* substitution */
+			row2[j + 1] = row1[j] + s * (string1[i] != string2[j]);
+			/* swap */
+			if (i > 0 && j > 0 && string1[i - 1] == string2[j] &&
+					string1[i] == string2[j - 1] &&
+					row2[j + 1] > row0[j - 1] + w)
+				row2[j + 1] = row0[j - 1] + w;
+			/* deletion */
+			if (row2[j + 1] > row1[j + 1] + d)
+				row2[j + 1] = row1[j + 1] + d;
+			/* insertion */
+			if (row2[j + 1] > row2[j] + a)
+				row2[j + 1] = row2[j] + a;
+		}
+
+		dummy = row0;
+		row0 = row1;
+		row1 = row2;
+		row2 = dummy;
+	}
+
+	i = row1[len2];
+	free(row0);
+	free(row1);
+	free(row2);
+
+	return i;
+}
diff --git a/tools/perf/util/levenshtein.h b/tools/perf/util/levenshtein.h
new file mode 100644
index 000000000..b0fcb6d8a
--- /dev/null
+++ b/tools/perf/util/levenshtein.h
@@ -0,0 +1,8 @@
+#ifndef __PERF_LEVENSHTEIN_H
+#define __PERF_LEVENSHTEIN_H
+
+int levenshtein(const char *string1, const char *string2,
+	int swap_penalty, int substition_penalty,
+	int insertion_penalty, int deletion_penalty);
+
+#endif /* __PERF_LEVENSHTEIN_H */
diff --git a/tools/perf/util/lzma.c b/tools/perf/util/lzma.c
new file mode 100644
index 000000000..95a1acb61
--- /dev/null
+++ b/tools/perf/util/lzma.c
@@ -0,0 +1,95 @@
+#include <lzma.h>
+#include <stdio.h>
+#include <linux/compiler.h>
+#include "util.h"
+#include "debug.h"
+
+#define BUFSIZE 8192
+
+static const char *lzma_strerror(lzma_ret ret)
+{
+	switch ((int) ret) {
+	case LZMA_MEM_ERROR:
+		return "Memory allocation failed";
+	case LZMA_OPTIONS_ERROR:
+		return "Unsupported decompressor flags";
+	case LZMA_FORMAT_ERROR:
+		return "The input is not in the .xz format";
+	case LZMA_DATA_ERROR:
+		return "Compressed file is corrupt";
+	case LZMA_BUF_ERROR:
+		return "Compressed file is truncated or otherwise corrupt";
+	default:
+		return "Unknown error, possibly a bug";
+	}
+}
+
+int lzma_decompress_to_file(const char *input, int output_fd)
+{
+	lzma_action action = LZMA_RUN;
+	lzma_stream strm   = LZMA_STREAM_INIT;
+	lzma_ret ret;
+
+	u8 buf_in[BUFSIZE];
+	u8 buf_out[BUFSIZE];
+	FILE *infile;
+
+	infile = fopen(input, "rb");
+	if (!infile) {
+		pr_err("lzma: fopen failed on %s: '%s'\n",
+		       input, strerror(errno));
+		return -1;
+	}
+
+	ret = lzma_stream_decoder(&strm, UINT64_MAX, LZMA_CONCATENATED);
+	if (ret != LZMA_OK) {
+		pr_err("lzma: lzma_stream_decoder failed %s (%d)\n",
+			lzma_strerror(ret), ret);
+		return -1;
+	}
+
+	strm.next_in   = NULL;
+	strm.avail_in  = 0;
+	strm.next_out  = buf_out;
+	strm.avail_out = sizeof(buf_out);
+
+	while (1) {
+		if (strm.avail_in == 0 && !feof(infile)) {
+			strm.next_in  = buf_in;
+			strm.avail_in = fread(buf_in, 1, sizeof(buf_in), infile);
+
+			if (ferror(infile)) {
+				pr_err("lzma: read error: %s\n", strerror(errno));
+				return -1;
+			}
+
+			if (feof(infile))
+				action = LZMA_FINISH;
+		}
+
+		ret = lzma_code(&strm, action);
+
+		if (strm.avail_out == 0 || ret == LZMA_STREAM_END) {
+			ssize_t write_size = sizeof(buf_out) - strm.avail_out;
+
+			if (writen(output_fd, buf_out, write_size) != write_size) {
+				pr_err("lzma: write error: %s\n", strerror(errno));
+				return -1;
+			}
+
+			strm.next_out  = buf_out;
+			strm.avail_out = sizeof(buf_out);
+		}
+
+		if (ret != LZMA_OK) {
+			if (ret == LZMA_STREAM_END)
+				return 0;
+
+			pr_err("lzma: failed %s\n", lzma_strerror(ret));
+			return -1;
+		}
+	}
+
+	fclose(infile);
+	return 0;
+}
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
new file mode 100644
index 000000000..527e032e2
--- /dev/null
+++ b/tools/perf/util/machine.c
@@ -0,0 +1,1847 @@
+#include "callchain.h"
+#include "debug.h"
+#include "event.h"
+#include "evsel.h"
+#include "hist.h"
+#include "machine.h"
+#include "map.h"
+#include "sort.h"
+#include "strlist.h"
+#include "thread.h"
+#include "vdso.h"
+#include <stdbool.h>
+#include <symbol/kallsyms.h>
+#include "unwind.h"
+#include "linux/hash.h"
+
+static void dsos__init(struct dsos *dsos)
+{
+	INIT_LIST_HEAD(&dsos->head);
+	dsos->root = RB_ROOT;
+}
+
+int machine__init(struct machine *machine, const char *root_dir, pid_t pid)
+{
+	map_groups__init(&machine->kmaps, machine);
+	RB_CLEAR_NODE(&machine->rb_node);
+	dsos__init(&machine->user_dsos);
+	dsos__init(&machine->kernel_dsos);
+
+	machine->threads = RB_ROOT;
+	INIT_LIST_HEAD(&machine->dead_threads);
+	machine->last_match = NULL;
+
+	machine->vdso_info = NULL;
+
+	machine->pid = pid;
+
+	machine->symbol_filter = NULL;
+	machine->id_hdr_size = 0;
+	machine->comm_exec = false;
+	machine->kernel_start = 0;
+
+	machine->root_dir = strdup(root_dir);
+	if (machine->root_dir == NULL)
+		return -ENOMEM;
+
+	if (pid != HOST_KERNEL_ID) {
+		struct thread *thread = machine__findnew_thread(machine, -1,
+								pid);
+		char comm[64];
+
+		if (thread == NULL)
+			return -ENOMEM;
+
+		snprintf(comm, sizeof(comm), "[guest/%d]", pid);
+		thread__set_comm(thread, comm, 0);
+	}
+
+	machine->current_tid = NULL;
+
+	return 0;
+}
+
+struct machine *machine__new_host(void)
+{
+	struct machine *machine = malloc(sizeof(*machine));
+
+	if (machine != NULL) {
+		machine__init(machine, "", HOST_KERNEL_ID);
+
+		if (machine__create_kernel_maps(machine) < 0)
+			goto out_delete;
+	}
+
+	return machine;
+out_delete:
+	free(machine);
+	return NULL;
+}
+
+static void dsos__delete(struct dsos *dsos)
+{
+	struct dso *pos, *n;
+
+	list_for_each_entry_safe(pos, n, &dsos->head, node) {
+		RB_CLEAR_NODE(&pos->rb_node);
+		list_del(&pos->node);
+		dso__delete(pos);
+	}
+}
+
+void machine__delete_threads(struct machine *machine)
+{
+	struct rb_node *nd = rb_first(&machine->threads);
+
+	while (nd) {
+		struct thread *t = rb_entry(nd, struct thread, rb_node);
+
+		nd = rb_next(nd);
+		machine__remove_thread(machine, t);
+	}
+}
+
+void machine__exit(struct machine *machine)
+{
+	map_groups__exit(&machine->kmaps);
+	dsos__delete(&machine->user_dsos);
+	dsos__delete(&machine->kernel_dsos);
+	vdso__exit(machine);
+	zfree(&machine->root_dir);
+	zfree(&machine->current_tid);
+}
+
+void machine__delete(struct machine *machine)
+{
+	machine__exit(machine);
+	free(machine);
+}
+
+void machines__init(struct machines *machines)
+{
+	machine__init(&machines->host, "", HOST_KERNEL_ID);
+	machines->guests = RB_ROOT;
+	machines->symbol_filter = NULL;
+}
+
+void machines__exit(struct machines *machines)
+{
+	machine__exit(&machines->host);
+	/* XXX exit guest */
+}
+
+struct machine *machines__add(struct machines *machines, pid_t pid,
+			      const char *root_dir)
+{
+	struct rb_node **p = &machines->guests.rb_node;
+	struct rb_node *parent = NULL;
+	struct machine *pos, *machine = malloc(sizeof(*machine));
+
+	if (machine == NULL)
+		return NULL;
+
+	if (machine__init(machine, root_dir, pid) != 0) {
+		free(machine);
+		return NULL;
+	}
+
+	machine->symbol_filter = machines->symbol_filter;
+
+	while (*p != NULL) {
+		parent = *p;
+		pos = rb_entry(parent, struct machine, rb_node);
+		if (pid < pos->pid)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&machine->rb_node, parent, p);
+	rb_insert_color(&machine->rb_node, &machines->guests);
+
+	return machine;
+}
+
+void machines__set_symbol_filter(struct machines *machines,
+				 symbol_filter_t symbol_filter)
+{
+	struct rb_node *nd;
+
+	machines->symbol_filter = symbol_filter;
+	machines->host.symbol_filter = symbol_filter;
+
+	for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) {
+		struct machine *machine = rb_entry(nd, struct machine, rb_node);
+
+		machine->symbol_filter = symbol_filter;
+	}
+}
+
+void machines__set_comm_exec(struct machines *machines, bool comm_exec)
+{
+	struct rb_node *nd;
+
+	machines->host.comm_exec = comm_exec;
+
+	for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) {
+		struct machine *machine = rb_entry(nd, struct machine, rb_node);
+
+		machine->comm_exec = comm_exec;
+	}
+}
+
+struct machine *machines__find(struct machines *machines, pid_t pid)
+{
+	struct rb_node **p = &machines->guests.rb_node;
+	struct rb_node *parent = NULL;
+	struct machine *machine;
+	struct machine *default_machine = NULL;
+
+	if (pid == HOST_KERNEL_ID)
+		return &machines->host;
+
+	while (*p != NULL) {
+		parent = *p;
+		machine = rb_entry(parent, struct machine, rb_node);
+		if (pid < machine->pid)
+			p = &(*p)->rb_left;
+		else if (pid > machine->pid)
+			p = &(*p)->rb_right;
+		else
+			return machine;
+		if (!machine->pid)
+			default_machine = machine;
+	}
+
+	return default_machine;
+}
+
+struct machine *machines__findnew(struct machines *machines, pid_t pid)
+{
+	char path[PATH_MAX];
+	const char *root_dir = "";
+	struct machine *machine = machines__find(machines, pid);
+
+	if (machine && (machine->pid == pid))
+		goto out;
+
+	if ((pid != HOST_KERNEL_ID) &&
+	    (pid != DEFAULT_GUEST_KERNEL_ID) &&
+	    (symbol_conf.guestmount)) {
+		sprintf(path, "%s/%d", symbol_conf.guestmount, pid);
+		if (access(path, R_OK)) {
+			static struct strlist *seen;
+
+			if (!seen)
+				seen = strlist__new(true, NULL);
+
+			if (!strlist__has_entry(seen, path)) {
+				pr_err("Can't access file %s\n", path);
+				strlist__add(seen, path);
+			}
+			machine = NULL;
+			goto out;
+		}
+		root_dir = path;
+	}
+
+	machine = machines__add(machines, pid, root_dir);
+out:
+	return machine;
+}
+
+void machines__process_guests(struct machines *machines,
+			      machine__process_t process, void *data)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		process(pos, data);
+	}
+}
+
+char *machine__mmap_name(struct machine *machine, char *bf, size_t size)
+{
+	if (machine__is_host(machine))
+		snprintf(bf, size, "[%s]", "kernel.kallsyms");
+	else if (machine__is_default_guest(machine))
+		snprintf(bf, size, "[%s]", "guest.kernel.kallsyms");
+	else {
+		snprintf(bf, size, "[%s.%d]", "guest.kernel.kallsyms",
+			 machine->pid);
+	}
+
+	return bf;
+}
+
+void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size)
+{
+	struct rb_node *node;
+	struct machine *machine;
+
+	machines->host.id_hdr_size = id_hdr_size;
+
+	for (node = rb_first(&machines->guests); node; node = rb_next(node)) {
+		machine = rb_entry(node, struct machine, rb_node);
+		machine->id_hdr_size = id_hdr_size;
+	}
+
+	return;
+}
+
+static void machine__update_thread_pid(struct machine *machine,
+				       struct thread *th, pid_t pid)
+{
+	struct thread *leader;
+
+	if (pid == th->pid_ || pid == -1 || th->pid_ != -1)
+		return;
+
+	th->pid_ = pid;
+
+	if (th->pid_ == th->tid)
+		return;
+
+	leader = machine__findnew_thread(machine, th->pid_, th->pid_);
+	if (!leader)
+		goto out_err;
+
+	if (!leader->mg)
+		leader->mg = map_groups__new(machine);
+
+	if (!leader->mg)
+		goto out_err;
+
+	if (th->mg == leader->mg)
+		return;
+
+	if (th->mg) {
+		/*
+		 * Maps are created from MMAP events which provide the pid and
+		 * tid.  Consequently there never should be any maps on a thread
+		 * with an unknown pid.  Just print an error if there are.
+		 */
+		if (!map_groups__empty(th->mg))
+			pr_err("Discarding thread maps for %d:%d\n",
+			       th->pid_, th->tid);
+		map_groups__delete(th->mg);
+	}
+
+	th->mg = map_groups__get(leader->mg);
+
+	return;
+
+out_err:
+	pr_err("Failed to join map groups for %d:%d\n", th->pid_, th->tid);
+}
+
+static struct thread *__machine__findnew_thread(struct machine *machine,
+						pid_t pid, pid_t tid,
+						bool create)
+{
+	struct rb_node **p = &machine->threads.rb_node;
+	struct rb_node *parent = NULL;
+	struct thread *th;
+
+	/*
+	 * Front-end cache - TID lookups come in blocks,
+	 * so most of the time we dont have to look up
+	 * the full rbtree:
+	 */
+	th = machine->last_match;
+	if (th != NULL) {
+		if (th->tid == tid) {
+			machine__update_thread_pid(machine, th, pid);
+			return th;
+		}
+
+		thread__zput(machine->last_match);
+	}
+
+	while (*p != NULL) {
+		parent = *p;
+		th = rb_entry(parent, struct thread, rb_node);
+
+		if (th->tid == tid) {
+			machine->last_match = thread__get(th);
+			machine__update_thread_pid(machine, th, pid);
+			return th;
+		}
+
+		if (tid < th->tid)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	if (!create)
+		return NULL;
+
+	th = thread__new(pid, tid);
+	if (th != NULL) {
+		rb_link_node(&th->rb_node, parent, p);
+		rb_insert_color(&th->rb_node, &machine->threads);
+
+		/*
+		 * We have to initialize map_groups separately
+		 * after rb tree is updated.
+		 *
+		 * The reason is that we call machine__findnew_thread
+		 * within thread__init_map_groups to find the thread
+		 * leader and that would screwed the rb tree.
+		 */
+		if (thread__init_map_groups(th, machine)) {
+			rb_erase(&th->rb_node, &machine->threads);
+			thread__delete(th);
+			return NULL;
+		}
+		/*
+		 * It is now in the rbtree, get a ref
+		 */
+		thread__get(th);
+		machine->last_match = thread__get(th);
+	}
+
+	return th;
+}
+
+struct thread *machine__findnew_thread(struct machine *machine, pid_t pid,
+				       pid_t tid)
+{
+	return __machine__findnew_thread(machine, pid, tid, true);
+}
+
+struct thread *machine__find_thread(struct machine *machine, pid_t pid,
+				    pid_t tid)
+{
+	return __machine__findnew_thread(machine, pid, tid, false);
+}
+
+struct comm *machine__thread_exec_comm(struct machine *machine,
+				       struct thread *thread)
+{
+	if (machine->comm_exec)
+		return thread__exec_comm(thread);
+	else
+		return thread__comm(thread);
+}
+
+int machine__process_comm_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample)
+{
+	struct thread *thread = machine__findnew_thread(machine,
+							event->comm.pid,
+							event->comm.tid);
+	bool exec = event->header.misc & PERF_RECORD_MISC_COMM_EXEC;
+
+	if (exec)
+		machine->comm_exec = true;
+
+	if (dump_trace)
+		perf_event__fprintf_comm(event, stdout);
+
+	if (thread == NULL ||
+	    __thread__set_comm(thread, event->comm.comm, sample->time, exec)) {
+		dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+int machine__process_lost_event(struct machine *machine __maybe_unused,
+				union perf_event *event, struct perf_sample *sample __maybe_unused)
+{
+	dump_printf(": id:%" PRIu64 ": lost:%" PRIu64 "\n",
+		    event->lost.id, event->lost.lost);
+	return 0;
+}
+
+static struct dso*
+machine__module_dso(struct machine *machine, struct kmod_path *m,
+		    const char *filename)
+{
+	struct dso *dso;
+
+	dso = dsos__find(&machine->kernel_dsos, m->name, true);
+	if (!dso) {
+		dso = dsos__addnew(&machine->kernel_dsos, m->name);
+		if (dso == NULL)
+			return NULL;
+
+		if (machine__is_host(machine))
+			dso->symtab_type = DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE;
+		else
+			dso->symtab_type = DSO_BINARY_TYPE__GUEST_KMODULE;
+
+		/* _KMODULE_COMP should be next to _KMODULE */
+		if (m->kmod && m->comp)
+			dso->symtab_type++;
+
+		dso__set_short_name(dso, strdup(m->name), true);
+		dso__set_long_name(dso, strdup(filename), true);
+	}
+
+	return dso;
+}
+
+struct map *machine__new_module(struct machine *machine, u64 start,
+				const char *filename)
+{
+	struct map *map = NULL;
+	struct dso *dso;
+	struct kmod_path m;
+
+	if (kmod_path__parse_name(&m, filename))
+		return NULL;
+
+	map = map_groups__find_by_name(&machine->kmaps, MAP__FUNCTION,
+				       m.name);
+	if (map)
+		goto out;
+
+	dso = machine__module_dso(machine, &m, filename);
+	if (dso == NULL)
+		goto out;
+
+	map = map__new2(start, dso, MAP__FUNCTION);
+	if (map == NULL)
+		goto out;
+
+	map_groups__insert(&machine->kmaps, map);
+
+out:
+	free(m.name);
+	return map;
+}
+
+size_t machines__fprintf_dsos(struct machines *machines, FILE *fp)
+{
+	struct rb_node *nd;
+	size_t ret = __dsos__fprintf(&machines->host.kernel_dsos.head, fp) +
+		     __dsos__fprintf(&machines->host.user_dsos.head, fp);
+
+	for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		ret += __dsos__fprintf(&pos->kernel_dsos.head, fp);
+		ret += __dsos__fprintf(&pos->user_dsos.head, fp);
+	}
+
+	return ret;
+}
+
+size_t machine__fprintf_dsos_buildid(struct machine *m, FILE *fp,
+				     bool (skip)(struct dso *dso, int parm), int parm)
+{
+	return __dsos__fprintf_buildid(&m->kernel_dsos.head, fp, skip, parm) +
+	       __dsos__fprintf_buildid(&m->user_dsos.head, fp, skip, parm);
+}
+
+size_t machines__fprintf_dsos_buildid(struct machines *machines, FILE *fp,
+				     bool (skip)(struct dso *dso, int parm), int parm)
+{
+	struct rb_node *nd;
+	size_t ret = machine__fprintf_dsos_buildid(&machines->host, fp, skip, parm);
+
+	for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) {
+		struct machine *pos = rb_entry(nd, struct machine, rb_node);
+		ret += machine__fprintf_dsos_buildid(pos, fp, skip, parm);
+	}
+	return ret;
+}
+
+size_t machine__fprintf_vmlinux_path(struct machine *machine, FILE *fp)
+{
+	int i;
+	size_t printed = 0;
+	struct dso *kdso = machine->vmlinux_maps[MAP__FUNCTION]->dso;
+
+	if (kdso->has_build_id) {
+		char filename[PATH_MAX];
+		if (dso__build_id_filename(kdso, filename, sizeof(filename)))
+			printed += fprintf(fp, "[0] %s\n", filename);
+	}
+
+	for (i = 0; i < vmlinux_path__nr_entries; ++i)
+		printed += fprintf(fp, "[%d] %s\n",
+				   i + kdso->has_build_id, vmlinux_path[i]);
+
+	return printed;
+}
+
+size_t machine__fprintf(struct machine *machine, FILE *fp)
+{
+	size_t ret = 0;
+	struct rb_node *nd;
+
+	for (nd = rb_first(&machine->threads); nd; nd = rb_next(nd)) {
+		struct thread *pos = rb_entry(nd, struct thread, rb_node);
+
+		ret += thread__fprintf(pos, fp);
+	}
+
+	return ret;
+}
+
+static struct dso *machine__get_kernel(struct machine *machine)
+{
+	const char *vmlinux_name = NULL;
+	struct dso *kernel;
+
+	if (machine__is_host(machine)) {
+		vmlinux_name = symbol_conf.vmlinux_name;
+		if (!vmlinux_name)
+			vmlinux_name = "[kernel.kallsyms]";
+
+		kernel = dso__kernel_findnew(machine, vmlinux_name,
+					     "[kernel]",
+					     DSO_TYPE_KERNEL);
+	} else {
+		char bf[PATH_MAX];
+
+		if (machine__is_default_guest(machine))
+			vmlinux_name = symbol_conf.default_guest_vmlinux_name;
+		if (!vmlinux_name)
+			vmlinux_name = machine__mmap_name(machine, bf,
+							  sizeof(bf));
+
+		kernel = dso__kernel_findnew(machine, vmlinux_name,
+					     "[guest.kernel]",
+					     DSO_TYPE_GUEST_KERNEL);
+	}
+
+	if (kernel != NULL && (!kernel->has_build_id))
+		dso__read_running_kernel_build_id(kernel, machine);
+
+	return kernel;
+}
+
+struct process_args {
+	u64 start;
+};
+
+static void machine__get_kallsyms_filename(struct machine *machine, char *buf,
+					   size_t bufsz)
+{
+	if (machine__is_default_guest(machine))
+		scnprintf(buf, bufsz, "%s", symbol_conf.default_guest_kallsyms);
+	else
+		scnprintf(buf, bufsz, "%s/proc/kallsyms", machine->root_dir);
+}
+
+const char *ref_reloc_sym_names[] = {"_text", "_stext", NULL};
+
+/* Figure out the start address of kernel map from /proc/kallsyms.
+ * Returns the name of the start symbol in *symbol_name. Pass in NULL as
+ * symbol_name if it's not that important.
+ */
+static u64 machine__get_running_kernel_start(struct machine *machine,
+					     const char **symbol_name)
+{
+	char filename[PATH_MAX];
+	int i;
+	const char *name;
+	u64 addr = 0;
+
+	machine__get_kallsyms_filename(machine, filename, PATH_MAX);
+
+	if (symbol__restricted_filename(filename, "/proc/kallsyms"))
+		return 0;
+
+	for (i = 0; (name = ref_reloc_sym_names[i]) != NULL; i++) {
+		addr = kallsyms__get_function_start(filename, name);
+		if (addr)
+			break;
+	}
+
+	if (symbol_name)
+		*symbol_name = name;
+
+	return addr;
+}
+
+int __machine__create_kernel_maps(struct machine *machine, struct dso *kernel)
+{
+	enum map_type type;
+	u64 start = machine__get_running_kernel_start(machine, NULL);
+
+	for (type = 0; type < MAP__NR_TYPES; ++type) {
+		struct kmap *kmap;
+
+		machine->vmlinux_maps[type] = map__new2(start, kernel, type);
+		if (machine->vmlinux_maps[type] == NULL)
+			return -1;
+
+		machine->vmlinux_maps[type]->map_ip =
+			machine->vmlinux_maps[type]->unmap_ip =
+				identity__map_ip;
+		kmap = map__kmap(machine->vmlinux_maps[type]);
+		if (!kmap)
+			return -1;
+
+		kmap->kmaps = &machine->kmaps;
+		map_groups__insert(&machine->kmaps,
+				   machine->vmlinux_maps[type]);
+	}
+
+	return 0;
+}
+
+void machine__destroy_kernel_maps(struct machine *machine)
+{
+	enum map_type type;
+
+	for (type = 0; type < MAP__NR_TYPES; ++type) {
+		struct kmap *kmap;
+
+		if (machine->vmlinux_maps[type] == NULL)
+			continue;
+
+		kmap = map__kmap(machine->vmlinux_maps[type]);
+		map_groups__remove(&machine->kmaps,
+				   machine->vmlinux_maps[type]);
+		if (kmap && kmap->ref_reloc_sym) {
+			/*
+			 * ref_reloc_sym is shared among all maps, so free just
+			 * on one of them.
+			 */
+			if (type == MAP__FUNCTION) {
+				zfree((char **)&kmap->ref_reloc_sym->name);
+				zfree(&kmap->ref_reloc_sym);
+			} else
+				kmap->ref_reloc_sym = NULL;
+		}
+
+		map__delete(machine->vmlinux_maps[type]);
+		machine->vmlinux_maps[type] = NULL;
+	}
+}
+
+int machines__create_guest_kernel_maps(struct machines *machines)
+{
+	int ret = 0;
+	struct dirent **namelist = NULL;
+	int i, items = 0;
+	char path[PATH_MAX];
+	pid_t pid;
+	char *endp;
+
+	if (symbol_conf.default_guest_vmlinux_name ||
+	    symbol_conf.default_guest_modules ||
+	    symbol_conf.default_guest_kallsyms) {
+		machines__create_kernel_maps(machines, DEFAULT_GUEST_KERNEL_ID);
+	}
+
+	if (symbol_conf.guestmount) {
+		items = scandir(symbol_conf.guestmount, &namelist, NULL, NULL);
+		if (items <= 0)
+			return -ENOENT;
+		for (i = 0; i < items; i++) {
+			if (!isdigit(namelist[i]->d_name[0])) {
+				/* Filter out . and .. */
+				continue;
+			}
+			pid = (pid_t)strtol(namelist[i]->d_name, &endp, 10);
+			if ((*endp != '\0') ||
+			    (endp == namelist[i]->d_name) ||
+			    (errno == ERANGE)) {
+				pr_debug("invalid directory (%s). Skipping.\n",
+					 namelist[i]->d_name);
+				continue;
+			}
+			sprintf(path, "%s/%s/proc/kallsyms",
+				symbol_conf.guestmount,
+				namelist[i]->d_name);
+			ret = access(path, R_OK);
+			if (ret) {
+				pr_debug("Can't access file %s\n", path);
+				goto failure;
+			}
+			machines__create_kernel_maps(machines, pid);
+		}
+failure:
+		free(namelist);
+	}
+
+	return ret;
+}
+
+void machines__destroy_kernel_maps(struct machines *machines)
+{
+	struct rb_node *next = rb_first(&machines->guests);
+
+	machine__destroy_kernel_maps(&machines->host);
+
+	while (next) {
+		struct machine *pos = rb_entry(next, struct machine, rb_node);
+
+		next = rb_next(&pos->rb_node);
+		rb_erase(&pos->rb_node, &machines->guests);
+		machine__delete(pos);
+	}
+}
+
+int machines__create_kernel_maps(struct machines *machines, pid_t pid)
+{
+	struct machine *machine = machines__findnew(machines, pid);
+
+	if (machine == NULL)
+		return -1;
+
+	return machine__create_kernel_maps(machine);
+}
+
+int machine__load_kallsyms(struct machine *machine, const char *filename,
+			   enum map_type type, symbol_filter_t filter)
+{
+	struct map *map = machine->vmlinux_maps[type];
+	int ret = dso__load_kallsyms(map->dso, filename, map, filter);
+
+	if (ret > 0) {
+		dso__set_loaded(map->dso, type);
+		/*
+		 * Since /proc/kallsyms will have multiple sessions for the
+		 * kernel, with modules between them, fixup the end of all
+		 * sections.
+		 */
+		__map_groups__fixup_end(&machine->kmaps, type);
+	}
+
+	return ret;
+}
+
+int machine__load_vmlinux_path(struct machine *machine, enum map_type type,
+			       symbol_filter_t filter)
+{
+	struct map *map = machine->vmlinux_maps[type];
+	int ret = dso__load_vmlinux_path(map->dso, map, filter);
+
+	if (ret > 0)
+		dso__set_loaded(map->dso, type);
+
+	return ret;
+}
+
+static void map_groups__fixup_end(struct map_groups *mg)
+{
+	int i;
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		__map_groups__fixup_end(mg, i);
+}
+
+static char *get_kernel_version(const char *root_dir)
+{
+	char version[PATH_MAX];
+	FILE *file;
+	char *name, *tmp;
+	const char *prefix = "Linux version ";
+
+	sprintf(version, "%s/proc/version", root_dir);
+	file = fopen(version, "r");
+	if (!file)
+		return NULL;
+
+	version[0] = '\0';
+	tmp = fgets(version, sizeof(version), file);
+	fclose(file);
+
+	name = strstr(version, prefix);
+	if (!name)
+		return NULL;
+	name += strlen(prefix);
+	tmp = strchr(name, ' ');
+	if (tmp)
+		*tmp = '\0';
+
+	return strdup(name);
+}
+
+static bool is_kmod_dso(struct dso *dso)
+{
+	return dso->symtab_type == DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE ||
+	       dso->symtab_type == DSO_BINARY_TYPE__GUEST_KMODULE;
+}
+
+static int map_groups__set_module_path(struct map_groups *mg, const char *path,
+				       struct kmod_path *m)
+{
+	struct map *map;
+	char *long_name;
+
+	map = map_groups__find_by_name(mg, MAP__FUNCTION, m->name);
+	if (map == NULL)
+		return 0;
+
+	long_name = strdup(path);
+	if (long_name == NULL)
+		return -ENOMEM;
+
+	dso__set_long_name(map->dso, long_name, true);
+	dso__kernel_module_get_build_id(map->dso, "");
+
+	/*
+	 * Full name could reveal us kmod compression, so
+	 * we need to update the symtab_type if needed.
+	 */
+	if (m->comp && is_kmod_dso(map->dso))
+		map->dso->symtab_type++;
+
+	return 0;
+}
+
+static int map_groups__set_modules_path_dir(struct map_groups *mg,
+				const char *dir_name, int depth)
+{
+	struct dirent *dent;
+	DIR *dir = opendir(dir_name);
+	int ret = 0;
+
+	if (!dir) {
+		pr_debug("%s: cannot open %s dir\n", __func__, dir_name);
+		return -1;
+	}
+
+	while ((dent = readdir(dir)) != NULL) {
+		char path[PATH_MAX];
+		struct stat st;
+
+		/*sshfs might return bad dent->d_type, so we have to stat*/
+		snprintf(path, sizeof(path), "%s/%s", dir_name, dent->d_name);
+		if (stat(path, &st))
+			continue;
+
+		if (S_ISDIR(st.st_mode)) {
+			if (!strcmp(dent->d_name, ".") ||
+			    !strcmp(dent->d_name, ".."))
+				continue;
+
+			/* Do not follow top-level source and build symlinks */
+			if (depth == 0) {
+				if (!strcmp(dent->d_name, "source") ||
+				    !strcmp(dent->d_name, "build"))
+					continue;
+			}
+
+			ret = map_groups__set_modules_path_dir(mg, path,
+							       depth + 1);
+			if (ret < 0)
+				goto out;
+		} else {
+			struct kmod_path m;
+
+			ret = kmod_path__parse_name(&m, dent->d_name);
+			if (ret)
+				goto out;
+
+			if (m.kmod)
+				ret = map_groups__set_module_path(mg, path, &m);
+
+			free(m.name);
+
+			if (ret)
+				goto out;
+		}
+	}
+
+out:
+	closedir(dir);
+	return ret;
+}
+
+static int machine__set_modules_path(struct machine *machine)
+{
+	char *version;
+	char modules_path[PATH_MAX];
+
+	version = get_kernel_version(machine->root_dir);
+	if (!version)
+		return -1;
+
+	snprintf(modules_path, sizeof(modules_path), "%s/lib/modules/%s",
+		 machine->root_dir, version);
+	free(version);
+
+	return map_groups__set_modules_path_dir(&machine->kmaps, modules_path, 0);
+}
+
+static int machine__create_module(void *arg, const char *name, u64 start)
+{
+	struct machine *machine = arg;
+	struct map *map;
+
+	map = machine__new_module(machine, start, name);
+	if (map == NULL)
+		return -1;
+
+	dso__kernel_module_get_build_id(map->dso, machine->root_dir);
+
+	return 0;
+}
+
+static int machine__create_modules(struct machine *machine)
+{
+	const char *modules;
+	char path[PATH_MAX];
+
+	if (machine__is_default_guest(machine)) {
+		modules = symbol_conf.default_guest_modules;
+	} else {
+		snprintf(path, PATH_MAX, "%s/proc/modules", machine->root_dir);
+		modules = path;
+	}
+
+	if (symbol__restricted_filename(modules, "/proc/modules"))
+		return -1;
+
+	if (modules__parse(modules, machine, machine__create_module))
+		return -1;
+
+	if (!machine__set_modules_path(machine))
+		return 0;
+
+	pr_debug("Problems setting modules path maps, continuing anyway...\n");
+
+	return 0;
+}
+
+int machine__create_kernel_maps(struct machine *machine)
+{
+	struct dso *kernel = machine__get_kernel(machine);
+	const char *name;
+	u64 addr = machine__get_running_kernel_start(machine, &name);
+	if (!addr)
+		return -1;
+
+	if (kernel == NULL ||
+	    __machine__create_kernel_maps(machine, kernel) < 0)
+		return -1;
+
+	if (symbol_conf.use_modules && machine__create_modules(machine) < 0) {
+		if (machine__is_host(machine))
+			pr_debug("Problems creating module maps, "
+				 "continuing anyway...\n");
+		else
+			pr_debug("Problems creating module maps for guest %d, "
+				 "continuing anyway...\n", machine->pid);
+	}
+
+	/*
+	 * Now that we have all the maps created, just set the ->end of them:
+	 */
+	map_groups__fixup_end(&machine->kmaps);
+
+	if (maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps, name,
+					     addr)) {
+		machine__destroy_kernel_maps(machine);
+		return -1;
+	}
+
+	return 0;
+}
+
+static void machine__set_kernel_mmap_len(struct machine *machine,
+					 union perf_event *event)
+{
+	int i;
+
+	for (i = 0; i < MAP__NR_TYPES; i++) {
+		machine->vmlinux_maps[i]->start = event->mmap.start;
+		machine->vmlinux_maps[i]->end   = (event->mmap.start +
+						   event->mmap.len);
+		/*
+		 * Be a bit paranoid here, some perf.data file came with
+		 * a zero sized synthesized MMAP event for the kernel.
+		 */
+		if (machine->vmlinux_maps[i]->end == 0)
+			machine->vmlinux_maps[i]->end = ~0ULL;
+	}
+}
+
+static bool machine__uses_kcore(struct machine *machine)
+{
+	struct dso *dso;
+
+	list_for_each_entry(dso, &machine->kernel_dsos.head, node) {
+		if (dso__is_kcore(dso))
+			return true;
+	}
+
+	return false;
+}
+
+static int machine__process_kernel_mmap_event(struct machine *machine,
+					      union perf_event *event)
+{
+	struct map *map;
+	char kmmap_prefix[PATH_MAX];
+	enum dso_kernel_type kernel_type;
+	bool is_kernel_mmap;
+
+	/* If we have maps from kcore then we do not need or want any others */
+	if (machine__uses_kcore(machine))
+		return 0;
+
+	machine__mmap_name(machine, kmmap_prefix, sizeof(kmmap_prefix));
+	if (machine__is_host(machine))
+		kernel_type = DSO_TYPE_KERNEL;
+	else
+		kernel_type = DSO_TYPE_GUEST_KERNEL;
+
+	is_kernel_mmap = memcmp(event->mmap.filename,
+				kmmap_prefix,
+				strlen(kmmap_prefix) - 1) == 0;
+	if (event->mmap.filename[0] == '/' ||
+	    (!is_kernel_mmap && event->mmap.filename[0] == '[')) {
+		map = machine__new_module(machine, event->mmap.start,
+					  event->mmap.filename);
+		if (map == NULL)
+			goto out_problem;
+
+		map->end = map->start + event->mmap.len;
+	} else if (is_kernel_mmap) {
+		const char *symbol_name = (event->mmap.filename +
+				strlen(kmmap_prefix));
+		/*
+		 * Should be there already, from the build-id table in
+		 * the header.
+		 */
+		struct dso *kernel = NULL;
+		struct dso *dso;
+
+		list_for_each_entry(dso, &machine->kernel_dsos.head, node) {
+			if (is_kernel_module(dso->long_name))
+				continue;
+
+			kernel = dso;
+			break;
+		}
+
+		if (kernel == NULL)
+			kernel = __dsos__findnew(&machine->kernel_dsos,
+						 kmmap_prefix);
+		if (kernel == NULL)
+			goto out_problem;
+
+		kernel->kernel = kernel_type;
+		if (__machine__create_kernel_maps(machine, kernel) < 0)
+			goto out_problem;
+
+		if (strstr(kernel->long_name, "vmlinux"))
+			dso__set_short_name(kernel, "[kernel.vmlinux]", false);
+
+		machine__set_kernel_mmap_len(machine, event);
+
+		/*
+		 * Avoid using a zero address (kptr_restrict) for the ref reloc
+		 * symbol. Effectively having zero here means that at record
+		 * time /proc/sys/kernel/kptr_restrict was non zero.
+		 */
+		if (event->mmap.pgoff != 0) {
+			maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps,
+							 symbol_name,
+							 event->mmap.pgoff);
+		}
+
+		if (machine__is_default_guest(machine)) {
+			/*
+			 * preload dso of guest kernel and modules
+			 */
+			dso__load(kernel, machine->vmlinux_maps[MAP__FUNCTION],
+				  NULL);
+		}
+	}
+	return 0;
+out_problem:
+	return -1;
+}
+
+int machine__process_mmap2_event(struct machine *machine,
+				 union perf_event *event,
+				 struct perf_sample *sample __maybe_unused)
+{
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct thread *thread;
+	struct map *map;
+	enum map_type type;
+	int ret = 0;
+
+	if (dump_trace)
+		perf_event__fprintf_mmap2(event, stdout);
+
+	if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL ||
+	    cpumode == PERF_RECORD_MISC_KERNEL) {
+		ret = machine__process_kernel_mmap_event(machine, event);
+		if (ret < 0)
+			goto out_problem;
+		return 0;
+	}
+
+	thread = machine__findnew_thread(machine, event->mmap2.pid,
+					event->mmap2.tid);
+	if (thread == NULL)
+		goto out_problem;
+
+	if (event->header.misc & PERF_RECORD_MISC_MMAP_DATA)
+		type = MAP__VARIABLE;
+	else
+		type = MAP__FUNCTION;
+
+	map = map__new(machine, event->mmap2.start,
+			event->mmap2.len, event->mmap2.pgoff,
+			event->mmap2.pid, event->mmap2.maj,
+			event->mmap2.min, event->mmap2.ino,
+			event->mmap2.ino_generation,
+			event->mmap2.prot,
+			event->mmap2.flags,
+			event->mmap2.filename, type, thread);
+
+	if (map == NULL)
+		goto out_problem;
+
+	thread__insert_map(thread, map);
+	return 0;
+
+out_problem:
+	dump_printf("problem processing PERF_RECORD_MMAP2, skipping event.\n");
+	return 0;
+}
+
+int machine__process_mmap_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample __maybe_unused)
+{
+	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct thread *thread;
+	struct map *map;
+	enum map_type type;
+	int ret = 0;
+
+	if (dump_trace)
+		perf_event__fprintf_mmap(event, stdout);
+
+	if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL ||
+	    cpumode == PERF_RECORD_MISC_KERNEL) {
+		ret = machine__process_kernel_mmap_event(machine, event);
+		if (ret < 0)
+			goto out_problem;
+		return 0;
+	}
+
+	thread = machine__findnew_thread(machine, event->mmap.pid,
+					 event->mmap.tid);
+	if (thread == NULL)
+		goto out_problem;
+
+	if (event->header.misc & PERF_RECORD_MISC_MMAP_DATA)
+		type = MAP__VARIABLE;
+	else
+		type = MAP__FUNCTION;
+
+	map = map__new(machine, event->mmap.start,
+			event->mmap.len, event->mmap.pgoff,
+			event->mmap.pid, 0, 0, 0, 0, 0, 0,
+			event->mmap.filename,
+			type, thread);
+
+	if (map == NULL)
+		goto out_problem;
+
+	thread__insert_map(thread, map);
+	return 0;
+
+out_problem:
+	dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
+	return 0;
+}
+
+void machine__remove_thread(struct machine *machine, struct thread *th)
+{
+	if (machine->last_match == th)
+		thread__zput(machine->last_match);
+
+	rb_erase(&th->rb_node, &machine->threads);
+	/*
+	 * Move it first to the dead_threads list, then drop the reference,
+	 * if this is the last reference, then the thread__delete destructor
+	 * will be called and we will remove it from the dead_threads list.
+	 */
+	list_add_tail(&th->node, &machine->dead_threads);
+	thread__put(th);
+}
+
+int machine__process_fork_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample)
+{
+	struct thread *thread = machine__find_thread(machine,
+						     event->fork.pid,
+						     event->fork.tid);
+	struct thread *parent = machine__findnew_thread(machine,
+							event->fork.ppid,
+							event->fork.ptid);
+
+	/* if a thread currently exists for the thread id remove it */
+	if (thread != NULL)
+		machine__remove_thread(machine, thread);
+
+	thread = machine__findnew_thread(machine, event->fork.pid,
+					 event->fork.tid);
+	if (dump_trace)
+		perf_event__fprintf_task(event, stdout);
+
+	if (thread == NULL || parent == NULL ||
+	    thread__fork(thread, parent, sample->time) < 0) {
+		dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+int machine__process_exit_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample __maybe_unused)
+{
+	struct thread *thread = machine__find_thread(machine,
+						     event->fork.pid,
+						     event->fork.tid);
+
+	if (dump_trace)
+		perf_event__fprintf_task(event, stdout);
+
+	if (thread != NULL)
+		thread__exited(thread);
+
+	return 0;
+}
+
+int machine__process_event(struct machine *machine, union perf_event *event,
+			   struct perf_sample *sample)
+{
+	int ret;
+
+	switch (event->header.type) {
+	case PERF_RECORD_COMM:
+		ret = machine__process_comm_event(machine, event, sample); break;
+	case PERF_RECORD_MMAP:
+		ret = machine__process_mmap_event(machine, event, sample); break;
+	case PERF_RECORD_MMAP2:
+		ret = machine__process_mmap2_event(machine, event, sample); break;
+	case PERF_RECORD_FORK:
+		ret = machine__process_fork_event(machine, event, sample); break;
+	case PERF_RECORD_EXIT:
+		ret = machine__process_exit_event(machine, event, sample); break;
+	case PERF_RECORD_LOST:
+		ret = machine__process_lost_event(machine, event, sample); break;
+	default:
+		ret = -1;
+		break;
+	}
+
+	return ret;
+}
+
+static bool symbol__match_regex(struct symbol *sym, regex_t *regex)
+{
+	if (sym->name && !regexec(regex, sym->name, 0, NULL, 0))
+		return 1;
+	return 0;
+}
+
+static void ip__resolve_ams(struct thread *thread,
+			    struct addr_map_symbol *ams,
+			    u64 ip)
+{
+	struct addr_location al;
+
+	memset(&al, 0, sizeof(al));
+	/*
+	 * We cannot use the header.misc hint to determine whether a
+	 * branch stack address is user, kernel, guest, hypervisor.
+	 * Branches may straddle the kernel/user/hypervisor boundaries.
+	 * Thus, we have to try consecutively until we find a match
+	 * or else, the symbol is unknown
+	 */
+	thread__find_cpumode_addr_location(thread, MAP__FUNCTION, ip, &al);
+
+	ams->addr = ip;
+	ams->al_addr = al.addr;
+	ams->sym = al.sym;
+	ams->map = al.map;
+}
+
+static void ip__resolve_data(struct thread *thread,
+			     u8 m, struct addr_map_symbol *ams, u64 addr)
+{
+	struct addr_location al;
+
+	memset(&al, 0, sizeof(al));
+
+	thread__find_addr_location(thread, m, MAP__VARIABLE, addr, &al);
+	if (al.map == NULL) {
+		/*
+		 * some shared data regions have execute bit set which puts
+		 * their mapping in the MAP__FUNCTION type array.
+		 * Check there as a fallback option before dropping the sample.
+		 */
+		thread__find_addr_location(thread, m, MAP__FUNCTION, addr, &al);
+	}
+
+	ams->addr = addr;
+	ams->al_addr = al.addr;
+	ams->sym = al.sym;
+	ams->map = al.map;
+}
+
+struct mem_info *sample__resolve_mem(struct perf_sample *sample,
+				     struct addr_location *al)
+{
+	struct mem_info *mi = zalloc(sizeof(*mi));
+
+	if (!mi)
+		return NULL;
+
+	ip__resolve_ams(al->thread, &mi->iaddr, sample->ip);
+	ip__resolve_data(al->thread, al->cpumode, &mi->daddr, sample->addr);
+	mi->data_src.val = sample->data_src;
+
+	return mi;
+}
+
+static int add_callchain_ip(struct thread *thread,
+			    struct symbol **parent,
+			    struct addr_location *root_al,
+			    u8 *cpumode,
+			    u64 ip)
+{
+	struct addr_location al;
+
+	al.filtered = 0;
+	al.sym = NULL;
+	if (!cpumode) {
+		thread__find_cpumode_addr_location(thread, MAP__FUNCTION,
+						   ip, &al);
+	} else {
+		if (ip >= PERF_CONTEXT_MAX) {
+			switch (ip) {
+			case PERF_CONTEXT_HV:
+				*cpumode = PERF_RECORD_MISC_HYPERVISOR;
+				break;
+			case PERF_CONTEXT_KERNEL:
+				*cpumode = PERF_RECORD_MISC_KERNEL;
+				break;
+			case PERF_CONTEXT_USER:
+				*cpumode = PERF_RECORD_MISC_USER;
+				break;
+			default:
+				pr_debug("invalid callchain context: "
+					 "%"PRId64"\n", (s64) ip);
+				/*
+				 * It seems the callchain is corrupted.
+				 * Discard all.
+				 */
+				callchain_cursor_reset(&callchain_cursor);
+				return 1;
+			}
+			return 0;
+		}
+		thread__find_addr_location(thread, *cpumode, MAP__FUNCTION,
+					   ip, &al);
+	}
+
+	if (al.sym != NULL) {
+		if (sort__has_parent && !*parent &&
+		    symbol__match_regex(al.sym, &parent_regex))
+			*parent = al.sym;
+		else if (have_ignore_callees && root_al &&
+		  symbol__match_regex(al.sym, &ignore_callees_regex)) {
+			/* Treat this symbol as the root,
+			   forgetting its callees. */
+			*root_al = al;
+			callchain_cursor_reset(&callchain_cursor);
+		}
+	}
+
+	return callchain_cursor_append(&callchain_cursor, al.addr, al.map, al.sym);
+}
+
+struct branch_info *sample__resolve_bstack(struct perf_sample *sample,
+					   struct addr_location *al)
+{
+	unsigned int i;
+	const struct branch_stack *bs = sample->branch_stack;
+	struct branch_info *bi = calloc(bs->nr, sizeof(struct branch_info));
+
+	if (!bi)
+		return NULL;
+
+	for (i = 0; i < bs->nr; i++) {
+		ip__resolve_ams(al->thread, &bi[i].to, bs->entries[i].to);
+		ip__resolve_ams(al->thread, &bi[i].from, bs->entries[i].from);
+		bi[i].flags = bs->entries[i].flags;
+	}
+	return bi;
+}
+
+#define CHASHSZ 127
+#define CHASHBITS 7
+#define NO_ENTRY 0xff
+
+#define PERF_MAX_BRANCH_DEPTH 127
+
+/* Remove loops. */
+static int remove_loops(struct branch_entry *l, int nr)
+{
+	int i, j, off;
+	unsigned char chash[CHASHSZ];
+
+	memset(chash, NO_ENTRY, sizeof(chash));
+
+	BUG_ON(PERF_MAX_BRANCH_DEPTH > 255);
+
+	for (i = 0; i < nr; i++) {
+		int h = hash_64(l[i].from, CHASHBITS) % CHASHSZ;
+
+		/* no collision handling for now */
+		if (chash[h] == NO_ENTRY) {
+			chash[h] = i;
+		} else if (l[chash[h]].from == l[i].from) {
+			bool is_loop = true;
+			/* check if it is a real loop */
+			off = 0;
+			for (j = chash[h]; j < i && i + off < nr; j++, off++)
+				if (l[j].from != l[i + off].from) {
+					is_loop = false;
+					break;
+				}
+			if (is_loop) {
+				memmove(l + i, l + i + off,
+					(nr - (i + off)) * sizeof(*l));
+				nr -= off;
+			}
+		}
+	}
+	return nr;
+}
+
+/*
+ * Recolve LBR callstack chain sample
+ * Return:
+ * 1 on success get LBR callchain information
+ * 0 no available LBR callchain information, should try fp
+ * negative error code on other errors.
+ */
+static int resolve_lbr_callchain_sample(struct thread *thread,
+					struct perf_sample *sample,
+					struct symbol **parent,
+					struct addr_location *root_al,
+					int max_stack)
+{
+	struct ip_callchain *chain = sample->callchain;
+	int chain_nr = min(max_stack, (int)chain->nr);
+	u8 cpumode = PERF_RECORD_MISC_USER;
+	int i, j, err;
+	u64 ip;
+
+	for (i = 0; i < chain_nr; i++) {
+		if (chain->ips[i] == PERF_CONTEXT_USER)
+			break;
+	}
+
+	/* LBR only affects the user callchain */
+	if (i != chain_nr) {
+		struct branch_stack *lbr_stack = sample->branch_stack;
+		int lbr_nr = lbr_stack->nr;
+		/*
+		 * LBR callstack can only get user call chain.
+		 * The mix_chain_nr is kernel call chain
+		 * number plus LBR user call chain number.
+		 * i is kernel call chain number,
+		 * 1 is PERF_CONTEXT_USER,
+		 * lbr_nr + 1 is the user call chain number.
+		 * For details, please refer to the comments
+		 * in callchain__printf
+		 */
+		int mix_chain_nr = i + 1 + lbr_nr + 1;
+
+		if (mix_chain_nr > PERF_MAX_STACK_DEPTH + PERF_MAX_BRANCH_DEPTH) {
+			pr_warning("corrupted callchain. skipping...\n");
+			return 0;
+		}
+
+		for (j = 0; j < mix_chain_nr; j++) {
+			if (callchain_param.order == ORDER_CALLEE) {
+				if (j < i + 1)
+					ip = chain->ips[j];
+				else if (j > i + 1)
+					ip = lbr_stack->entries[j - i - 2].from;
+				else
+					ip = lbr_stack->entries[0].to;
+			} else {
+				if (j < lbr_nr)
+					ip = lbr_stack->entries[lbr_nr - j - 1].from;
+				else if (j > lbr_nr)
+					ip = chain->ips[i + 1 - (j - lbr_nr)];
+				else
+					ip = lbr_stack->entries[0].to;
+			}
+
+			err = add_callchain_ip(thread, parent, root_al, &cpumode, ip);
+			if (err)
+				return (err < 0) ? err : 0;
+		}
+		return 1;
+	}
+
+	return 0;
+}
+
+static int thread__resolve_callchain_sample(struct thread *thread,
+					    struct perf_evsel *evsel,
+					    struct perf_sample *sample,
+					    struct symbol **parent,
+					    struct addr_location *root_al,
+					    int max_stack)
+{
+	struct branch_stack *branch = sample->branch_stack;
+	struct ip_callchain *chain = sample->callchain;
+	int chain_nr = min(max_stack, (int)chain->nr);
+	u8 cpumode = PERF_RECORD_MISC_USER;
+	int i, j, err;
+	int skip_idx = -1;
+	int first_call = 0;
+
+	callchain_cursor_reset(&callchain_cursor);
+
+	if (has_branch_callstack(evsel)) {
+		err = resolve_lbr_callchain_sample(thread, sample, parent,
+						   root_al, max_stack);
+		if (err)
+			return (err < 0) ? err : 0;
+	}
+
+	/*
+	 * Based on DWARF debug information, some architectures skip
+	 * a callchain entry saved by the kernel.
+	 */
+	if (chain->nr < PERF_MAX_STACK_DEPTH)
+		skip_idx = arch_skip_callchain_idx(thread, chain);
+
+	/*
+	 * Add branches to call stack for easier browsing. This gives
+	 * more context for a sample than just the callers.
+	 *
+	 * This uses individual histograms of paths compared to the
+	 * aggregated histograms the normal LBR mode uses.
+	 *
+	 * Limitations for now:
+	 * - No extra filters
+	 * - No annotations (should annotate somehow)
+	 */
+
+	if (branch && callchain_param.branch_callstack) {
+		int nr = min(max_stack, (int)branch->nr);
+		struct branch_entry be[nr];
+
+		if (branch->nr > PERF_MAX_BRANCH_DEPTH) {
+			pr_warning("corrupted branch chain. skipping...\n");
+			goto check_calls;
+		}
+
+		for (i = 0; i < nr; i++) {
+			if (callchain_param.order == ORDER_CALLEE) {
+				be[i] = branch->entries[i];
+				/*
+				 * Check for overlap into the callchain.
+				 * The return address is one off compared to
+				 * the branch entry. To adjust for this
+				 * assume the calling instruction is not longer
+				 * than 8 bytes.
+				 */
+				if (i == skip_idx ||
+				    chain->ips[first_call] >= PERF_CONTEXT_MAX)
+					first_call++;
+				else if (be[i].from < chain->ips[first_call] &&
+				    be[i].from >= chain->ips[first_call] - 8)
+					first_call++;
+			} else
+				be[i] = branch->entries[branch->nr - i - 1];
+		}
+
+		nr = remove_loops(be, nr);
+
+		for (i = 0; i < nr; i++) {
+			err = add_callchain_ip(thread, parent, root_al,
+					       NULL, be[i].to);
+			if (!err)
+				err = add_callchain_ip(thread, parent, root_al,
+						       NULL, be[i].from);
+			if (err == -EINVAL)
+				break;
+			if (err)
+				return err;
+		}
+		chain_nr -= nr;
+	}
+
+check_calls:
+	if (chain->nr > PERF_MAX_STACK_DEPTH) {
+		pr_warning("corrupted callchain. skipping...\n");
+		return 0;
+	}
+
+	for (i = first_call; i < chain_nr; i++) {
+		u64 ip;
+
+		if (callchain_param.order == ORDER_CALLEE)
+			j = i;
+		else
+			j = chain->nr - i - 1;
+
+#ifdef HAVE_SKIP_CALLCHAIN_IDX
+		if (j == skip_idx)
+			continue;
+#endif
+		ip = chain->ips[j];
+
+		err = add_callchain_ip(thread, parent, root_al, &cpumode, ip);
+
+		if (err)
+			return (err < 0) ? err : 0;
+	}
+
+	return 0;
+}
+
+static int unwind_entry(struct unwind_entry *entry, void *arg)
+{
+	struct callchain_cursor *cursor = arg;
+	return callchain_cursor_append(cursor, entry->ip,
+				       entry->map, entry->sym);
+}
+
+int thread__resolve_callchain(struct thread *thread,
+			      struct perf_evsel *evsel,
+			      struct perf_sample *sample,
+			      struct symbol **parent,
+			      struct addr_location *root_al,
+			      int max_stack)
+{
+	int ret = thread__resolve_callchain_sample(thread, evsel,
+						   sample, parent,
+						   root_al, max_stack);
+	if (ret)
+		return ret;
+
+	/* Can we do dwarf post unwind? */
+	if (!((evsel->attr.sample_type & PERF_SAMPLE_REGS_USER) &&
+	      (evsel->attr.sample_type & PERF_SAMPLE_STACK_USER)))
+		return 0;
+
+	/* Bail out if nothing was captured. */
+	if ((!sample->user_regs.regs) ||
+	    (!sample->user_stack.size))
+		return 0;
+
+	return unwind__get_entries(unwind_entry, &callchain_cursor,
+				   thread, sample, max_stack);
+
+}
+
+int machine__for_each_thread(struct machine *machine,
+			     int (*fn)(struct thread *thread, void *p),
+			     void *priv)
+{
+	struct rb_node *nd;
+	struct thread *thread;
+	int rc = 0;
+
+	for (nd = rb_first(&machine->threads); nd; nd = rb_next(nd)) {
+		thread = rb_entry(nd, struct thread, rb_node);
+		rc = fn(thread, priv);
+		if (rc != 0)
+			return rc;
+	}
+
+	list_for_each_entry(thread, &machine->dead_threads, node) {
+		rc = fn(thread, priv);
+		if (rc != 0)
+			return rc;
+	}
+	return rc;
+}
+
+int __machine__synthesize_threads(struct machine *machine, struct perf_tool *tool,
+				  struct target *target, struct thread_map *threads,
+				  perf_event__handler_t process, bool data_mmap)
+{
+	if (target__has_task(target))
+		return perf_event__synthesize_thread_map(tool, threads, process, machine, data_mmap);
+	else if (target__has_cpu(target))
+		return perf_event__synthesize_threads(tool, process, machine, data_mmap);
+	/* command specified */
+	return 0;
+}
+
+pid_t machine__get_current_tid(struct machine *machine, int cpu)
+{
+	if (cpu < 0 || cpu >= MAX_NR_CPUS || !machine->current_tid)
+		return -1;
+
+	return machine->current_tid[cpu];
+}
+
+int machine__set_current_tid(struct machine *machine, int cpu, pid_t pid,
+			     pid_t tid)
+{
+	struct thread *thread;
+
+	if (cpu < 0)
+		return -EINVAL;
+
+	if (!machine->current_tid) {
+		int i;
+
+		machine->current_tid = calloc(MAX_NR_CPUS, sizeof(pid_t));
+		if (!machine->current_tid)
+			return -ENOMEM;
+		for (i = 0; i < MAX_NR_CPUS; i++)
+			machine->current_tid[i] = -1;
+	}
+
+	if (cpu >= MAX_NR_CPUS) {
+		pr_err("Requested CPU %d too large. ", cpu);
+		pr_err("Consider raising MAX_NR_CPUS\n");
+		return -EINVAL;
+	}
+
+	machine->current_tid[cpu] = tid;
+
+	thread = machine__findnew_thread(machine, pid, tid);
+	if (!thread)
+		return -ENOMEM;
+
+	thread->cpu = cpu;
+
+	return 0;
+}
+
+int machine__get_kernel_start(struct machine *machine)
+{
+	struct map *map = machine__kernel_map(machine, MAP__FUNCTION);
+	int err = 0;
+
+	/*
+	 * The only addresses above 2^63 are kernel addresses of a 64-bit
+	 * kernel.  Note that addresses are unsigned so that on a 32-bit system
+	 * all addresses including kernel addresses are less than 2^32.  In
+	 * that case (32-bit system), if the kernel mapping is unknown, all
+	 * addresses will be assumed to be in user space - see
+	 * machine__kernel_ip().
+	 */
+	machine->kernel_start = 1ULL << 63;
+	if (map) {
+		err = map__load(map, machine->symbol_filter);
+		if (map->start)
+			machine->kernel_start = map->start;
+	}
+	return err;
+}
diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h
new file mode 100644
index 000000000..6d64cedb9
--- /dev/null
+++ b/tools/perf/util/machine.h
@@ -0,0 +1,227 @@
+#ifndef __PERF_MACHINE_H
+#define __PERF_MACHINE_H
+
+#include <sys/types.h>
+#include <linux/rbtree.h>
+#include "map.h"
+#include "dso.h"
+#include "event.h"
+
+struct addr_location;
+struct branch_stack;
+struct perf_evsel;
+struct perf_sample;
+struct symbol;
+struct thread;
+union perf_event;
+
+/* Native host kernel uses -1 as pid index in machine */
+#define	HOST_KERNEL_ID			(-1)
+#define	DEFAULT_GUEST_KERNEL_ID		(0)
+
+extern const char *ref_reloc_sym_names[];
+
+struct vdso_info;
+
+struct machine {
+	struct rb_node	  rb_node;
+	pid_t		  pid;
+	u16		  id_hdr_size;
+	bool		  comm_exec;
+	char		  *root_dir;
+	struct rb_root	  threads;
+	struct list_head  dead_threads;
+	struct thread	  *last_match;
+	struct vdso_info  *vdso_info;
+	struct dsos	  user_dsos;
+	struct dsos	  kernel_dsos;
+	struct map_groups kmaps;
+	struct map	  *vmlinux_maps[MAP__NR_TYPES];
+	u64		  kernel_start;
+	symbol_filter_t	  symbol_filter;
+	pid_t		  *current_tid;
+	union { /* Tool specific area */
+		void	  *priv;
+		u64	  db_id;
+	};
+};
+
+static inline
+struct map *machine__kernel_map(struct machine *machine, enum map_type type)
+{
+	return machine->vmlinux_maps[type];
+}
+
+int machine__get_kernel_start(struct machine *machine);
+
+static inline u64 machine__kernel_start(struct machine *machine)
+{
+	if (!machine->kernel_start)
+		machine__get_kernel_start(machine);
+	return machine->kernel_start;
+}
+
+static inline bool machine__kernel_ip(struct machine *machine, u64 ip)
+{
+	u64 kernel_start = machine__kernel_start(machine);
+
+	return ip >= kernel_start;
+}
+
+struct thread *machine__find_thread(struct machine *machine, pid_t pid,
+				    pid_t tid);
+struct comm *machine__thread_exec_comm(struct machine *machine,
+				       struct thread *thread);
+
+int machine__process_comm_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample);
+int machine__process_exit_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample);
+int machine__process_fork_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample);
+int machine__process_lost_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample);
+int machine__process_mmap_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample);
+int machine__process_mmap2_event(struct machine *machine, union perf_event *event,
+				 struct perf_sample *sample);
+int machine__process_event(struct machine *machine, union perf_event *event,
+				struct perf_sample *sample);
+
+typedef void (*machine__process_t)(struct machine *machine, void *data);
+
+struct machines {
+	struct machine host;
+	struct rb_root guests;
+	symbol_filter_t symbol_filter;
+};
+
+void machines__init(struct machines *machines);
+void machines__exit(struct machines *machines);
+
+void machines__process_guests(struct machines *machines,
+			      machine__process_t process, void *data);
+
+struct machine *machines__add(struct machines *machines, pid_t pid,
+			      const char *root_dir);
+struct machine *machines__find_host(struct machines *machines);
+struct machine *machines__find(struct machines *machines, pid_t pid);
+struct machine *machines__findnew(struct machines *machines, pid_t pid);
+
+void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size);
+char *machine__mmap_name(struct machine *machine, char *bf, size_t size);
+
+void machines__set_symbol_filter(struct machines *machines,
+				 symbol_filter_t symbol_filter);
+void machines__set_comm_exec(struct machines *machines, bool comm_exec);
+
+struct machine *machine__new_host(void);
+int machine__init(struct machine *machine, const char *root_dir, pid_t pid);
+void machine__exit(struct machine *machine);
+void machine__delete_threads(struct machine *machine);
+void machine__delete(struct machine *machine);
+void machine__remove_thread(struct machine *machine, struct thread *th);
+
+struct branch_info *sample__resolve_bstack(struct perf_sample *sample,
+					   struct addr_location *al);
+struct mem_info *sample__resolve_mem(struct perf_sample *sample,
+				     struct addr_location *al);
+int thread__resolve_callchain(struct thread *thread,
+			      struct perf_evsel *evsel,
+			      struct perf_sample *sample,
+			      struct symbol **parent,
+			      struct addr_location *root_al,
+			      int max_stack);
+
+/*
+ * Default guest kernel is defined by parameter --guestkallsyms
+ * and --guestmodules
+ */
+static inline bool machine__is_default_guest(struct machine *machine)
+{
+	return machine ? machine->pid == DEFAULT_GUEST_KERNEL_ID : false;
+}
+
+static inline bool machine__is_host(struct machine *machine)
+{
+	return machine ? machine->pid == HOST_KERNEL_ID : false;
+}
+
+struct thread *machine__findnew_thread(struct machine *machine, pid_t pid,
+				       pid_t tid);
+
+size_t machine__fprintf(struct machine *machine, FILE *fp);
+
+static inline
+struct symbol *machine__find_kernel_symbol(struct machine *machine,
+					   enum map_type type, u64 addr,
+					   struct map **mapp,
+					   symbol_filter_t filter)
+{
+	return map_groups__find_symbol(&machine->kmaps, type, addr,
+				       mapp, filter);
+}
+
+static inline
+struct symbol *machine__find_kernel_function(struct machine *machine, u64 addr,
+					     struct map **mapp,
+					     symbol_filter_t filter)
+{
+	return machine__find_kernel_symbol(machine, MAP__FUNCTION, addr,
+					   mapp, filter);
+}
+
+static inline
+struct symbol *machine__find_kernel_function_by_name(struct machine *machine,
+						     const char *name,
+						     struct map **mapp,
+						     symbol_filter_t filter)
+{
+	return map_groups__find_function_by_name(&machine->kmaps, name, mapp,
+						 filter);
+}
+
+struct map *machine__new_module(struct machine *machine, u64 start,
+				const char *filename);
+
+int machine__load_kallsyms(struct machine *machine, const char *filename,
+			   enum map_type type, symbol_filter_t filter);
+int machine__load_vmlinux_path(struct machine *machine, enum map_type type,
+			       symbol_filter_t filter);
+
+size_t machine__fprintf_dsos_buildid(struct machine *machine, FILE *fp,
+				     bool (skip)(struct dso *dso, int parm), int parm);
+size_t machines__fprintf_dsos(struct machines *machines, FILE *fp);
+size_t machines__fprintf_dsos_buildid(struct machines *machines, FILE *fp,
+				     bool (skip)(struct dso *dso, int parm), int parm);
+
+void machine__destroy_kernel_maps(struct machine *machine);
+int __machine__create_kernel_maps(struct machine *machine, struct dso *kernel);
+int machine__create_kernel_maps(struct machine *machine);
+
+int machines__create_kernel_maps(struct machines *machines, pid_t pid);
+int machines__create_guest_kernel_maps(struct machines *machines);
+void machines__destroy_kernel_maps(struct machines *machines);
+
+size_t machine__fprintf_vmlinux_path(struct machine *machine, FILE *fp);
+
+int machine__for_each_thread(struct machine *machine,
+			     int (*fn)(struct thread *thread, void *p),
+			     void *priv);
+
+int __machine__synthesize_threads(struct machine *machine, struct perf_tool *tool,
+				  struct target *target, struct thread_map *threads,
+				  perf_event__handler_t process, bool data_mmap);
+static inline
+int machine__synthesize_threads(struct machine *machine, struct target *target,
+				struct thread_map *threads, bool data_mmap)
+{
+	return __machine__synthesize_threads(machine, NULL, target, threads,
+					     perf_event__process, data_mmap);
+}
+
+pid_t machine__get_current_tid(struct machine *machine, int cpu);
+int machine__set_current_tid(struct machine *machine, int cpu, pid_t pid,
+			     pid_t tid);
+
+#endif /* __PERF_MACHINE_H */
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
new file mode 100644
index 000000000..a14f08f41
--- /dev/null
+++ b/tools/perf/util/map.c
@@ -0,0 +1,800 @@
+#include "symbol.h"
+#include <errno.h>
+#include <inttypes.h>
+#include <limits.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdio.h>
+#include <unistd.h>
+#include "map.h"
+#include "thread.h"
+#include "strlist.h"
+#include "vdso.h"
+#include "build-id.h"
+#include "util.h"
+#include "debug.h"
+#include "machine.h"
+#include <linux/string.h>
+
+const char *map_type__name[MAP__NR_TYPES] = {
+	[MAP__FUNCTION] = "Functions",
+	[MAP__VARIABLE] = "Variables",
+};
+
+static inline int is_anon_memory(const char *filename)
+{
+	return !strcmp(filename, "//anon") ||
+	       !strcmp(filename, "/dev/zero (deleted)") ||
+	       !strcmp(filename, "/anon_hugepage (deleted)");
+}
+
+static inline int is_no_dso_memory(const char *filename)
+{
+	return !strncmp(filename, "[stack", 6) ||
+	       !strncmp(filename, "/SYSV",5)   ||
+	       !strcmp(filename, "[heap]");
+}
+
+static inline int is_android_lib(const char *filename)
+{
+	return !strncmp(filename, "/data/app-lib", 13) ||
+	       !strncmp(filename, "/system/lib", 11);
+}
+
+static inline bool replace_android_lib(const char *filename, char *newfilename)
+{
+	const char *libname;
+	char *app_abi;
+	size_t app_abi_length, new_length;
+	size_t lib_length = 0;
+
+	libname  = strrchr(filename, '/');
+	if (libname)
+		lib_length = strlen(libname);
+
+	app_abi = getenv("APP_ABI");
+	if (!app_abi)
+		return false;
+
+	app_abi_length = strlen(app_abi);
+
+	if (!strncmp(filename, "/data/app-lib", 13)) {
+		char *apk_path;
+
+		if (!app_abi_length)
+			return false;
+
+		new_length = 7 + app_abi_length + lib_length;
+
+		apk_path = getenv("APK_PATH");
+		if (apk_path) {
+			new_length += strlen(apk_path) + 1;
+			if (new_length > PATH_MAX)
+				return false;
+			snprintf(newfilename, new_length,
+				 "%s/libs/%s/%s", apk_path, app_abi, libname);
+		} else {
+			if (new_length > PATH_MAX)
+				return false;
+			snprintf(newfilename, new_length,
+				 "libs/%s/%s", app_abi, libname);
+		}
+		return true;
+	}
+
+	if (!strncmp(filename, "/system/lib/", 11)) {
+		char *ndk, *app;
+		const char *arch;
+		size_t ndk_length;
+		size_t app_length;
+
+		ndk = getenv("NDK_ROOT");
+		app = getenv("APP_PLATFORM");
+
+		if (!(ndk && app))
+			return false;
+
+		ndk_length = strlen(ndk);
+		app_length = strlen(app);
+
+		if (!(ndk_length && app_length && app_abi_length))
+			return false;
+
+		arch = !strncmp(app_abi, "arm", 3) ? "arm" :
+		       !strncmp(app_abi, "mips", 4) ? "mips" :
+		       !strncmp(app_abi, "x86", 3) ? "x86" : NULL;
+
+		if (!arch)
+			return false;
+
+		new_length = 27 + ndk_length +
+			     app_length + lib_length
+			   + strlen(arch);
+
+		if (new_length > PATH_MAX)
+			return false;
+		snprintf(newfilename, new_length,
+			"%s/platforms/%s/arch-%s/usr/lib/%s",
+			ndk, app, arch, libname);
+
+		return true;
+	}
+	return false;
+}
+
+void map__init(struct map *map, enum map_type type,
+	       u64 start, u64 end, u64 pgoff, struct dso *dso)
+{
+	map->type     = type;
+	map->start    = start;
+	map->end      = end;
+	map->pgoff    = pgoff;
+	map->reloc    = 0;
+	map->dso      = dso;
+	map->map_ip   = map__map_ip;
+	map->unmap_ip = map__unmap_ip;
+	RB_CLEAR_NODE(&map->rb_node);
+	map->groups   = NULL;
+	map->referenced = false;
+	map->erange_warned = false;
+}
+
+struct map *map__new(struct machine *machine, u64 start, u64 len,
+		     u64 pgoff, u32 pid, u32 d_maj, u32 d_min, u64 ino,
+		     u64 ino_gen, u32 prot, u32 flags, char *filename,
+		     enum map_type type, struct thread *thread)
+{
+	struct map *map = malloc(sizeof(*map));
+
+	if (map != NULL) {
+		char newfilename[PATH_MAX];
+		struct dso *dso;
+		int anon, no_dso, vdso, android;
+
+		android = is_android_lib(filename);
+		anon = is_anon_memory(filename);
+		vdso = is_vdso_map(filename);
+		no_dso = is_no_dso_memory(filename);
+
+		map->maj = d_maj;
+		map->min = d_min;
+		map->ino = ino;
+		map->ino_generation = ino_gen;
+		map->prot = prot;
+		map->flags = flags;
+
+		if ((anon || no_dso) && type == MAP__FUNCTION) {
+			snprintf(newfilename, sizeof(newfilename), "/tmp/perf-%d.map", pid);
+			filename = newfilename;
+		}
+
+		if (android) {
+			if (replace_android_lib(filename, newfilename))
+				filename = newfilename;
+		}
+
+		if (vdso) {
+			pgoff = 0;
+			dso = vdso__dso_findnew(machine, thread);
+		} else
+			dso = __dsos__findnew(&machine->user_dsos, filename);
+
+		if (dso == NULL)
+			goto out_delete;
+
+		map__init(map, type, start, start + len, pgoff, dso);
+
+		if (anon || no_dso) {
+			map->map_ip = map->unmap_ip = identity__map_ip;
+
+			/*
+			 * Set memory without DSO as loaded. All map__find_*
+			 * functions still return NULL, and we avoid the
+			 * unnecessary map__load warning.
+			 */
+			if (type != MAP__FUNCTION)
+				dso__set_loaded(dso, map->type);
+		}
+	}
+	return map;
+out_delete:
+	free(map);
+	return NULL;
+}
+
+/*
+ * Constructor variant for modules (where we know from /proc/modules where
+ * they are loaded) and for vmlinux, where only after we load all the
+ * symbols we'll know where it starts and ends.
+ */
+struct map *map__new2(u64 start, struct dso *dso, enum map_type type)
+{
+	struct map *map = calloc(1, (sizeof(*map) +
+				     (dso->kernel ? sizeof(struct kmap) : 0)));
+	if (map != NULL) {
+		/*
+		 * ->end will be filled after we load all the symbols
+		 */
+		map__init(map, type, start, 0, 0, dso);
+	}
+
+	return map;
+}
+
+void map__delete(struct map *map)
+{
+	free(map);
+}
+
+void map__fixup_start(struct map *map)
+{
+	struct rb_root *symbols = &map->dso->symbols[map->type];
+	struct rb_node *nd = rb_first(symbols);
+	if (nd != NULL) {
+		struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+		map->start = sym->start;
+	}
+}
+
+void map__fixup_end(struct map *map)
+{
+	struct rb_root *symbols = &map->dso->symbols[map->type];
+	struct rb_node *nd = rb_last(symbols);
+	if (nd != NULL) {
+		struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
+		map->end = sym->end;
+	}
+}
+
+#define DSO__DELETED "(deleted)"
+
+int map__load(struct map *map, symbol_filter_t filter)
+{
+	const char *name = map->dso->long_name;
+	int nr;
+
+	if (dso__loaded(map->dso, map->type))
+		return 0;
+
+	nr = dso__load(map->dso, map, filter);
+	if (nr < 0) {
+		if (map->dso->has_build_id) {
+			char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+
+			build_id__sprintf(map->dso->build_id,
+					  sizeof(map->dso->build_id),
+					  sbuild_id);
+			pr_warning("%s with build id %s not found",
+				   name, sbuild_id);
+		} else
+			pr_warning("Failed to open %s", name);
+
+		pr_warning(", continuing without symbols\n");
+		return -1;
+	} else if (nr == 0) {
+#ifdef HAVE_LIBELF_SUPPORT
+		const size_t len = strlen(name);
+		const size_t real_len = len - sizeof(DSO__DELETED);
+
+		if (len > sizeof(DSO__DELETED) &&
+		    strcmp(name + real_len + 1, DSO__DELETED) == 0) {
+			pr_warning("%.*s was updated (is prelink enabled?). "
+				"Restart the long running apps that use it!\n",
+				   (int)real_len, name);
+		} else {
+			pr_warning("no symbols found in %s, maybe install "
+				   "a debug package?\n", name);
+		}
+#endif
+		return -1;
+	}
+
+	return 0;
+}
+
+struct symbol *map__find_symbol(struct map *map, u64 addr,
+				symbol_filter_t filter)
+{
+	if (map__load(map, filter) < 0)
+		return NULL;
+
+	return dso__find_symbol(map->dso, map->type, addr);
+}
+
+struct symbol *map__find_symbol_by_name(struct map *map, const char *name,
+					symbol_filter_t filter)
+{
+	if (map__load(map, filter) < 0)
+		return NULL;
+
+	if (!dso__sorted_by_name(map->dso, map->type))
+		dso__sort_by_name(map->dso, map->type);
+
+	return dso__find_symbol_by_name(map->dso, map->type, name);
+}
+
+struct map *map__clone(struct map *map)
+{
+	return memdup(map, sizeof(*map));
+}
+
+int map__overlap(struct map *l, struct map *r)
+{
+	if (l->start > r->start) {
+		struct map *t = l;
+		l = r;
+		r = t;
+	}
+
+	if (l->end > r->start)
+		return 1;
+
+	return 0;
+}
+
+size_t map__fprintf(struct map *map, FILE *fp)
+{
+	return fprintf(fp, " %" PRIx64 "-%" PRIx64 " %" PRIx64 " %s\n",
+		       map->start, map->end, map->pgoff, map->dso->name);
+}
+
+size_t map__fprintf_dsoname(struct map *map, FILE *fp)
+{
+	const char *dsoname = "[unknown]";
+
+	if (map && map->dso && (map->dso->name || map->dso->long_name)) {
+		if (symbol_conf.show_kernel_path && map->dso->long_name)
+			dsoname = map->dso->long_name;
+		else if (map->dso->name)
+			dsoname = map->dso->name;
+	}
+
+	return fprintf(fp, "%s", dsoname);
+}
+
+int map__fprintf_srcline(struct map *map, u64 addr, const char *prefix,
+			 FILE *fp)
+{
+	char *srcline;
+	int ret = 0;
+
+	if (map && map->dso) {
+		srcline = get_srcline(map->dso,
+				      map__rip_2objdump(map, addr), NULL, true);
+		if (srcline != SRCLINE_UNKNOWN)
+			ret = fprintf(fp, "%s%s", prefix, srcline);
+		free_srcline(srcline);
+	}
+	return ret;
+}
+
+/**
+ * map__rip_2objdump - convert symbol start address to objdump address.
+ * @map: memory map
+ * @rip: symbol start address
+ *
+ * objdump wants/reports absolute IPs for ET_EXEC, and RIPs for ET_DYN.
+ * map->dso->adjust_symbols==1 for ET_EXEC-like cases except ET_REL which is
+ * relative to section start.
+ *
+ * Return: Address suitable for passing to "objdump --start-address="
+ */
+u64 map__rip_2objdump(struct map *map, u64 rip)
+{
+	if (!map->dso->adjust_symbols)
+		return rip;
+
+	if (map->dso->rel)
+		return rip - map->pgoff;
+
+	return map->unmap_ip(map, rip) - map->reloc;
+}
+
+/**
+ * map__objdump_2mem - convert objdump address to a memory address.
+ * @map: memory map
+ * @ip: objdump address
+ *
+ * Closely related to map__rip_2objdump(), this function takes an address from
+ * objdump and converts it to a memory address.  Note this assumes that @map
+ * contains the address.  To be sure the result is valid, check it forwards
+ * e.g. map__rip_2objdump(map->map_ip(map, map__objdump_2mem(map, ip))) == ip
+ *
+ * Return: Memory address.
+ */
+u64 map__objdump_2mem(struct map *map, u64 ip)
+{
+	if (!map->dso->adjust_symbols)
+		return map->unmap_ip(map, ip);
+
+	if (map->dso->rel)
+		return map->unmap_ip(map, ip + map->pgoff);
+
+	return ip + map->reloc;
+}
+
+void map_groups__init(struct map_groups *mg, struct machine *machine)
+{
+	int i;
+	for (i = 0; i < MAP__NR_TYPES; ++i) {
+		mg->maps[i] = RB_ROOT;
+		INIT_LIST_HEAD(&mg->removed_maps[i]);
+	}
+	mg->machine = machine;
+	mg->refcnt = 1;
+}
+
+static void maps__delete(struct rb_root *maps)
+{
+	struct rb_node *next = rb_first(maps);
+
+	while (next) {
+		struct map *pos = rb_entry(next, struct map, rb_node);
+
+		next = rb_next(&pos->rb_node);
+		rb_erase(&pos->rb_node, maps);
+		map__delete(pos);
+	}
+}
+
+static void maps__delete_removed(struct list_head *maps)
+{
+	struct map *pos, *n;
+
+	list_for_each_entry_safe(pos, n, maps, node) {
+		list_del(&pos->node);
+		map__delete(pos);
+	}
+}
+
+void map_groups__exit(struct map_groups *mg)
+{
+	int i;
+
+	for (i = 0; i < MAP__NR_TYPES; ++i) {
+		maps__delete(&mg->maps[i]);
+		maps__delete_removed(&mg->removed_maps[i]);
+	}
+}
+
+bool map_groups__empty(struct map_groups *mg)
+{
+	int i;
+
+	for (i = 0; i < MAP__NR_TYPES; ++i) {
+		if (maps__first(&mg->maps[i]))
+			return false;
+		if (!list_empty(&mg->removed_maps[i]))
+			return false;
+	}
+
+	return true;
+}
+
+struct map_groups *map_groups__new(struct machine *machine)
+{
+	struct map_groups *mg = malloc(sizeof(*mg));
+
+	if (mg != NULL)
+		map_groups__init(mg, machine);
+
+	return mg;
+}
+
+void map_groups__delete(struct map_groups *mg)
+{
+	map_groups__exit(mg);
+	free(mg);
+}
+
+void map_groups__put(struct map_groups *mg)
+{
+	if (--mg->refcnt == 0)
+		map_groups__delete(mg);
+}
+
+void map_groups__flush(struct map_groups *mg)
+{
+	int type;
+
+	for (type = 0; type < MAP__NR_TYPES; type++) {
+		struct rb_root *root = &mg->maps[type];
+		struct rb_node *next = rb_first(root);
+
+		while (next) {
+			struct map *pos = rb_entry(next, struct map, rb_node);
+			next = rb_next(&pos->rb_node);
+			rb_erase(&pos->rb_node, root);
+			/*
+			 * We may have references to this map, for
+			 * instance in some hist_entry instances, so
+			 * just move them to a separate list.
+			 */
+			list_add_tail(&pos->node, &mg->removed_maps[pos->type]);
+		}
+	}
+}
+
+struct symbol *map_groups__find_symbol(struct map_groups *mg,
+				       enum map_type type, u64 addr,
+				       struct map **mapp,
+				       symbol_filter_t filter)
+{
+	struct map *map = map_groups__find(mg, type, addr);
+
+	/* Ensure map is loaded before using map->map_ip */
+	if (map != NULL && map__load(map, filter) >= 0) {
+		if (mapp != NULL)
+			*mapp = map;
+		return map__find_symbol(map, map->map_ip(map, addr), filter);
+	}
+
+	return NULL;
+}
+
+struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg,
+					       enum map_type type,
+					       const char *name,
+					       struct map **mapp,
+					       symbol_filter_t filter)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(&mg->maps[type]); nd; nd = rb_next(nd)) {
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+		struct symbol *sym = map__find_symbol_by_name(pos, name, filter);
+
+		if (sym == NULL)
+			continue;
+		if (mapp != NULL)
+			*mapp = pos;
+		return sym;
+	}
+
+	return NULL;
+}
+
+int map_groups__find_ams(struct addr_map_symbol *ams, symbol_filter_t filter)
+{
+	if (ams->addr < ams->map->start || ams->addr >= ams->map->end) {
+		if (ams->map->groups == NULL)
+			return -1;
+		ams->map = map_groups__find(ams->map->groups, ams->map->type,
+					    ams->addr);
+		if (ams->map == NULL)
+			return -1;
+	}
+
+	ams->al_addr = ams->map->map_ip(ams->map, ams->addr);
+	ams->sym = map__find_symbol(ams->map, ams->al_addr, filter);
+
+	return ams->sym ? 0 : -1;
+}
+
+size_t __map_groups__fprintf_maps(struct map_groups *mg, enum map_type type,
+				  FILE *fp)
+{
+	size_t printed = fprintf(fp, "%s:\n", map_type__name[type]);
+	struct rb_node *nd;
+
+	for (nd = rb_first(&mg->maps[type]); nd; nd = rb_next(nd)) {
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+		printed += fprintf(fp, "Map:");
+		printed += map__fprintf(pos, fp);
+		if (verbose > 2) {
+			printed += dso__fprintf(pos->dso, type, fp);
+			printed += fprintf(fp, "--\n");
+		}
+	}
+
+	return printed;
+}
+
+static size_t map_groups__fprintf_maps(struct map_groups *mg, FILE *fp)
+{
+	size_t printed = 0, i;
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		printed += __map_groups__fprintf_maps(mg, i, fp);
+	return printed;
+}
+
+static size_t __map_groups__fprintf_removed_maps(struct map_groups *mg,
+						 enum map_type type, FILE *fp)
+{
+	struct map *pos;
+	size_t printed = 0;
+
+	list_for_each_entry(pos, &mg->removed_maps[type], node) {
+		printed += fprintf(fp, "Map:");
+		printed += map__fprintf(pos, fp);
+		if (verbose > 1) {
+			printed += dso__fprintf(pos->dso, type, fp);
+			printed += fprintf(fp, "--\n");
+		}
+	}
+	return printed;
+}
+
+static size_t map_groups__fprintf_removed_maps(struct map_groups *mg,
+					       FILE *fp)
+{
+	size_t printed = 0, i;
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		printed += __map_groups__fprintf_removed_maps(mg, i, fp);
+	return printed;
+}
+
+size_t map_groups__fprintf(struct map_groups *mg, FILE *fp)
+{
+	size_t printed = map_groups__fprintf_maps(mg, fp);
+	printed += fprintf(fp, "Removed maps:\n");
+	return printed + map_groups__fprintf_removed_maps(mg, fp);
+}
+
+int map_groups__fixup_overlappings(struct map_groups *mg, struct map *map,
+				   FILE *fp)
+{
+	struct rb_root *root = &mg->maps[map->type];
+	struct rb_node *next = rb_first(root);
+	int err = 0;
+
+	while (next) {
+		struct map *pos = rb_entry(next, struct map, rb_node);
+		next = rb_next(&pos->rb_node);
+
+		if (!map__overlap(pos, map))
+			continue;
+
+		if (verbose >= 2) {
+			fputs("overlapping maps:\n", fp);
+			map__fprintf(map, fp);
+			map__fprintf(pos, fp);
+		}
+
+		rb_erase(&pos->rb_node, root);
+		/*
+		 * Now check if we need to create new maps for areas not
+		 * overlapped by the new map:
+		 */
+		if (map->start > pos->start) {
+			struct map *before = map__clone(pos);
+
+			if (before == NULL) {
+				err = -ENOMEM;
+				goto move_map;
+			}
+
+			before->end = map->start;
+			map_groups__insert(mg, before);
+			if (verbose >= 2)
+				map__fprintf(before, fp);
+		}
+
+		if (map->end < pos->end) {
+			struct map *after = map__clone(pos);
+
+			if (after == NULL) {
+				err = -ENOMEM;
+				goto move_map;
+			}
+
+			after->start = map->end;
+			map_groups__insert(mg, after);
+			if (verbose >= 2)
+				map__fprintf(after, fp);
+		}
+move_map:
+		/*
+		 * If we have references, just move them to a separate list.
+		 */
+		if (pos->referenced)
+			list_add_tail(&pos->node, &mg->removed_maps[map->type]);
+		else
+			map__delete(pos);
+
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+/*
+ * XXX This should not really _copy_ te maps, but refcount them.
+ */
+int map_groups__clone(struct map_groups *mg,
+		      struct map_groups *parent, enum map_type type)
+{
+	struct rb_node *nd;
+	for (nd = rb_first(&parent->maps[type]); nd; nd = rb_next(nd)) {
+		struct map *map = rb_entry(nd, struct map, rb_node);
+		struct map *new = map__clone(map);
+		if (new == NULL)
+			return -ENOMEM;
+		map_groups__insert(mg, new);
+	}
+	return 0;
+}
+
+void maps__insert(struct rb_root *maps, struct map *map)
+{
+	struct rb_node **p = &maps->rb_node;
+	struct rb_node *parent = NULL;
+	const u64 ip = map->start;
+	struct map *m;
+
+	while (*p != NULL) {
+		parent = *p;
+		m = rb_entry(parent, struct map, rb_node);
+		if (ip < m->start)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&map->rb_node, parent, p);
+	rb_insert_color(&map->rb_node, maps);
+}
+
+void maps__remove(struct rb_root *maps, struct map *map)
+{
+	rb_erase(&map->rb_node, maps);
+}
+
+struct map *maps__find(struct rb_root *maps, u64 ip)
+{
+	struct rb_node **p = &maps->rb_node;
+	struct rb_node *parent = NULL;
+	struct map *m;
+
+	while (*p != NULL) {
+		parent = *p;
+		m = rb_entry(parent, struct map, rb_node);
+		if (ip < m->start)
+			p = &(*p)->rb_left;
+		else if (ip >= m->end)
+			p = &(*p)->rb_right;
+		else
+			return m;
+	}
+
+	return NULL;
+}
+
+struct map *maps__first(struct rb_root *maps)
+{
+	struct rb_node *first = rb_first(maps);
+
+	if (first)
+		return rb_entry(first, struct map, rb_node);
+	return NULL;
+}
+
+struct map *maps__next(struct map *map)
+{
+	struct rb_node *next = rb_next(&map->rb_node);
+
+	if (next)
+		return rb_entry(next, struct map, rb_node);
+	return NULL;
+}
+
+struct kmap *map__kmap(struct map *map)
+{
+	if (!map->dso || !map->dso->kernel) {
+		pr_err("Internal error: map__kmap with a non-kernel map\n");
+		return NULL;
+	}
+	return (struct kmap *)(map + 1);
+}
+
+struct map_groups *map__kmaps(struct map *map)
+{
+	struct kmap *kmap = map__kmap(map);
+
+	if (!kmap || !kmap->kmaps) {
+		pr_err("Internal error: map__kmaps with a non-kernel map\n");
+		return NULL;
+	}
+	return kmap->kmaps;
+}
diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h
new file mode 100644
index 000000000..ec19c59ca
--- /dev/null
+++ b/tools/perf/util/map.h
@@ -0,0 +1,235 @@
+#ifndef __PERF_MAP_H
+#define __PERF_MAP_H
+
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <linux/types.h>
+
+enum map_type {
+	MAP__FUNCTION = 0,
+	MAP__VARIABLE,
+};
+
+#define MAP__NR_TYPES (MAP__VARIABLE + 1)
+
+extern const char *map_type__name[MAP__NR_TYPES];
+
+struct dso;
+struct ip_callchain;
+struct ref_reloc_sym;
+struct map_groups;
+struct machine;
+struct perf_evsel;
+
+struct map {
+	union {
+		struct rb_node	rb_node;
+		struct list_head node;
+	};
+	u64			start;
+	u64			end;
+	u8 /* enum map_type */	type;
+	bool			referenced;
+	bool			erange_warned;
+	u32			priv;
+	u32			prot;
+	u32			flags;
+	u64			pgoff;
+	u64			reloc;
+	u32			maj, min; /* only valid for MMAP2 record */
+	u64			ino;      /* only valid for MMAP2 record */
+	u64			ino_generation;/* only valid for MMAP2 record */
+
+	/* ip -> dso rip */
+	u64			(*map_ip)(struct map *, u64);
+	/* dso rip -> ip */
+	u64			(*unmap_ip)(struct map *, u64);
+
+	struct dso		*dso;
+	struct map_groups	*groups;
+};
+
+struct kmap {
+	struct ref_reloc_sym	*ref_reloc_sym;
+	struct map_groups	*kmaps;
+};
+
+struct map_groups {
+	struct rb_root	 maps[MAP__NR_TYPES];
+	struct list_head removed_maps[MAP__NR_TYPES];
+	struct machine	 *machine;
+	int		 refcnt;
+};
+
+struct map_groups *map_groups__new(struct machine *machine);
+void map_groups__delete(struct map_groups *mg);
+bool map_groups__empty(struct map_groups *mg);
+
+static inline struct map_groups *map_groups__get(struct map_groups *mg)
+{
+	++mg->refcnt;
+	return mg;
+}
+
+void map_groups__put(struct map_groups *mg);
+
+struct kmap *map__kmap(struct map *map);
+struct map_groups *map__kmaps(struct map *map);
+
+static inline u64 map__map_ip(struct map *map, u64 ip)
+{
+	return ip - map->start + map->pgoff;
+}
+
+static inline u64 map__unmap_ip(struct map *map, u64 ip)
+{
+	return ip + map->start - map->pgoff;
+}
+
+static inline u64 identity__map_ip(struct map *map __maybe_unused, u64 ip)
+{
+	return ip;
+}
+
+
+/* rip/ip <-> addr suitable for passing to `objdump --start-address=` */
+u64 map__rip_2objdump(struct map *map, u64 rip);
+
+/* objdump address -> memory address */
+u64 map__objdump_2mem(struct map *map, u64 ip);
+
+struct symbol;
+struct thread;
+
+/* map__for_each_symbol - iterate over the symbols in the given map
+ *
+ * @map: the 'struct map *' in which symbols itereated
+ * @pos: the 'struct symbol *' to use as a loop cursor
+ * @n: the 'struct rb_node *' to use as a temporary storage
+ * Note: caller must ensure map->dso is not NULL (map is loaded).
+ */
+#define map__for_each_symbol(map, pos, n)	\
+	dso__for_each_symbol(map->dso, pos, n, map->type)
+
+/* map__for_each_symbol_with_name - iterate over the symbols in the given map
+ *                                  that have the given name
+ *
+ * @map: the 'struct map *' in which symbols itereated
+ * @sym_name: the symbol name
+ * @pos: the 'struct symbol *' to use as a loop cursor
+ * @filter: to use when loading the DSO
+ */
+#define __map__for_each_symbol_by_name(map, sym_name, pos, filter)	\
+	for (pos = map__find_symbol_by_name(map, sym_name, filter);	\
+	     pos && strcmp(pos->name, sym_name) == 0;		\
+	     pos = symbol__next_by_name(pos))
+
+#define map__for_each_symbol_by_name(map, sym_name, pos)		\
+	__map__for_each_symbol_by_name(map, sym_name, (pos), NULL)
+
+typedef int (*symbol_filter_t)(struct map *map, struct symbol *sym);
+
+void map__init(struct map *map, enum map_type type,
+	       u64 start, u64 end, u64 pgoff, struct dso *dso);
+struct map *map__new(struct machine *machine, u64 start, u64 len,
+		     u64 pgoff, u32 pid, u32 d_maj, u32 d_min, u64 ino,
+		     u64 ino_gen, u32 prot, u32 flags,
+		     char *filename, enum map_type type, struct thread *thread);
+struct map *map__new2(u64 start, struct dso *dso, enum map_type type);
+void map__delete(struct map *map);
+struct map *map__clone(struct map *map);
+int map__overlap(struct map *l, struct map *r);
+size_t map__fprintf(struct map *map, FILE *fp);
+size_t map__fprintf_dsoname(struct map *map, FILE *fp);
+int map__fprintf_srcline(struct map *map, u64 addr, const char *prefix,
+			 FILE *fp);
+
+int map__load(struct map *map, symbol_filter_t filter);
+struct symbol *map__find_symbol(struct map *map,
+				u64 addr, symbol_filter_t filter);
+struct symbol *map__find_symbol_by_name(struct map *map, const char *name,
+					symbol_filter_t filter);
+void map__fixup_start(struct map *map);
+void map__fixup_end(struct map *map);
+
+void map__reloc_vmlinux(struct map *map);
+
+size_t __map_groups__fprintf_maps(struct map_groups *mg, enum map_type type,
+				  FILE *fp);
+void maps__insert(struct rb_root *maps, struct map *map);
+void maps__remove(struct rb_root *maps, struct map *map);
+struct map *maps__find(struct rb_root *maps, u64 addr);
+struct map *maps__first(struct rb_root *maps);
+struct map *maps__next(struct map *map);
+void map_groups__init(struct map_groups *mg, struct machine *machine);
+void map_groups__exit(struct map_groups *mg);
+int map_groups__clone(struct map_groups *mg,
+		      struct map_groups *parent, enum map_type type);
+size_t map_groups__fprintf(struct map_groups *mg, FILE *fp);
+
+int maps__set_kallsyms_ref_reloc_sym(struct map **maps, const char *symbol_name,
+				     u64 addr);
+
+static inline void map_groups__insert(struct map_groups *mg, struct map *map)
+{
+	maps__insert(&mg->maps[map->type], map);
+	map->groups = mg;
+}
+
+static inline void map_groups__remove(struct map_groups *mg, struct map *map)
+{
+	maps__remove(&mg->maps[map->type], map);
+}
+
+static inline struct map *map_groups__find(struct map_groups *mg,
+					   enum map_type type, u64 addr)
+{
+	return maps__find(&mg->maps[type], addr);
+}
+
+static inline struct map *map_groups__first(struct map_groups *mg,
+					    enum map_type type)
+{
+	return maps__first(&mg->maps[type]);
+}
+
+static inline struct map *map_groups__next(struct map *map)
+{
+	return maps__next(map);
+}
+
+struct symbol *map_groups__find_symbol(struct map_groups *mg,
+				       enum map_type type, u64 addr,
+				       struct map **mapp,
+				       symbol_filter_t filter);
+
+struct symbol *map_groups__find_symbol_by_name(struct map_groups *mg,
+					       enum map_type type,
+					       const char *name,
+					       struct map **mapp,
+					       symbol_filter_t filter);
+
+struct addr_map_symbol;
+
+int map_groups__find_ams(struct addr_map_symbol *ams, symbol_filter_t filter);
+
+static inline
+struct symbol *map_groups__find_function_by_name(struct map_groups *mg,
+						 const char *name, struct map **mapp,
+						 symbol_filter_t filter)
+{
+	return map_groups__find_symbol_by_name(mg, MAP__FUNCTION, name, mapp, filter);
+}
+
+int map_groups__fixup_overlappings(struct map_groups *mg, struct map *map,
+				   FILE *fp);
+
+struct map *map_groups__find_by_name(struct map_groups *mg,
+				     enum map_type type, const char *name);
+
+void map_groups__flush(struct map_groups *mg);
+
+#endif /* __PERF_MAP_H */
diff --git a/tools/perf/util/ordered-events.c b/tools/perf/util/ordered-events.c
new file mode 100644
index 000000000..52be201b9
--- /dev/null
+++ b/tools/perf/util/ordered-events.c
@@ -0,0 +1,307 @@
+#include <linux/list.h>
+#include <linux/compiler.h>
+#include <linux/string.h>
+#include "ordered-events.h"
+#include "session.h"
+#include "asm/bug.h"
+#include "debug.h"
+
+#define pr_N(n, fmt, ...) \
+	eprintf(n, debug_ordered_events, fmt, ##__VA_ARGS__)
+
+#define pr(fmt, ...) pr_N(1, pr_fmt(fmt), ##__VA_ARGS__)
+
+static void queue_event(struct ordered_events *oe, struct ordered_event *new)
+{
+	struct ordered_event *last = oe->last;
+	u64 timestamp = new->timestamp;
+	struct list_head *p;
+
+	++oe->nr_events;
+	oe->last = new;
+
+	pr_oe_time2(timestamp, "queue_event nr_events %u\n", oe->nr_events);
+
+	if (!last) {
+		list_add(&new->list, &oe->events);
+		oe->max_timestamp = timestamp;
+		return;
+	}
+
+	/*
+	 * last event might point to some random place in the list as it's
+	 * the last queued event. We expect that the new event is close to
+	 * this.
+	 */
+	if (last->timestamp <= timestamp) {
+		while (last->timestamp <= timestamp) {
+			p = last->list.next;
+			if (p == &oe->events) {
+				list_add_tail(&new->list, &oe->events);
+				oe->max_timestamp = timestamp;
+				return;
+			}
+			last = list_entry(p, struct ordered_event, list);
+		}
+		list_add_tail(&new->list, &last->list);
+	} else {
+		while (last->timestamp > timestamp) {
+			p = last->list.prev;
+			if (p == &oe->events) {
+				list_add(&new->list, &oe->events);
+				return;
+			}
+			last = list_entry(p, struct ordered_event, list);
+		}
+		list_add(&new->list, &last->list);
+	}
+}
+
+static union perf_event *__dup_event(struct ordered_events *oe,
+				     union perf_event *event)
+{
+	union perf_event *new_event = NULL;
+
+	if (oe->cur_alloc_size < oe->max_alloc_size) {
+		new_event = memdup(event, event->header.size);
+		if (new_event)
+			oe->cur_alloc_size += event->header.size;
+	}
+
+	return new_event;
+}
+
+static union perf_event *dup_event(struct ordered_events *oe,
+				   union perf_event *event)
+{
+	return oe->copy_on_queue ? __dup_event(oe, event) : event;
+}
+
+static void free_dup_event(struct ordered_events *oe, union perf_event *event)
+{
+	if (oe->copy_on_queue) {
+		oe->cur_alloc_size -= event->header.size;
+		free(event);
+	}
+}
+
+#define MAX_SAMPLE_BUFFER	(64 * 1024 / sizeof(struct ordered_event))
+static struct ordered_event *alloc_event(struct ordered_events *oe,
+					 union perf_event *event)
+{
+	struct list_head *cache = &oe->cache;
+	struct ordered_event *new = NULL;
+	union perf_event *new_event;
+
+	new_event = dup_event(oe, event);
+	if (!new_event)
+		return NULL;
+
+	if (!list_empty(cache)) {
+		new = list_entry(cache->next, struct ordered_event, list);
+		list_del(&new->list);
+	} else if (oe->buffer) {
+		new = oe->buffer + oe->buffer_idx;
+		if (++oe->buffer_idx == MAX_SAMPLE_BUFFER)
+			oe->buffer = NULL;
+	} else if (oe->cur_alloc_size < oe->max_alloc_size) {
+		size_t size = MAX_SAMPLE_BUFFER * sizeof(*new);
+
+		oe->buffer = malloc(size);
+		if (!oe->buffer) {
+			free_dup_event(oe, new_event);
+			return NULL;
+		}
+
+		pr("alloc size %" PRIu64 "B (+%zu), max %" PRIu64 "B\n",
+		   oe->cur_alloc_size, size, oe->max_alloc_size);
+
+		oe->cur_alloc_size += size;
+		list_add(&oe->buffer->list, &oe->to_free);
+
+		/* First entry is abused to maintain the to_free list. */
+		oe->buffer_idx = 2;
+		new = oe->buffer + 1;
+	} else {
+		pr("allocation limit reached %" PRIu64 "B\n", oe->max_alloc_size);
+	}
+
+	new->event = new_event;
+	return new;
+}
+
+static struct ordered_event *
+ordered_events__new_event(struct ordered_events *oe, u64 timestamp,
+		    union perf_event *event)
+{
+	struct ordered_event *new;
+
+	new = alloc_event(oe, event);
+	if (new) {
+		new->timestamp = timestamp;
+		queue_event(oe, new);
+	}
+
+	return new;
+}
+
+void ordered_events__delete(struct ordered_events *oe, struct ordered_event *event)
+{
+	list_move(&event->list, &oe->cache);
+	oe->nr_events--;
+	free_dup_event(oe, event->event);
+}
+
+int ordered_events__queue(struct ordered_events *oe, union perf_event *event,
+			  struct perf_sample *sample, u64 file_offset)
+{
+	u64 timestamp = sample->time;
+	struct ordered_event *oevent;
+
+	if (!timestamp || timestamp == ~0ULL)
+		return -ETIME;
+
+	if (timestamp < oe->last_flush) {
+		pr_oe_time(timestamp,      "out of order event\n");
+		pr_oe_time(oe->last_flush, "last flush, last_flush_type %d\n",
+			   oe->last_flush_type);
+
+		oe->nr_unordered_events++;
+	}
+
+	oevent = ordered_events__new_event(oe, timestamp, event);
+	if (!oevent) {
+		ordered_events__flush(oe, OE_FLUSH__HALF);
+		oevent = ordered_events__new_event(oe, timestamp, event);
+	}
+
+	if (!oevent)
+		return -ENOMEM;
+
+	oevent->file_offset = file_offset;
+	return 0;
+}
+
+static int __ordered_events__flush(struct ordered_events *oe)
+{
+	struct list_head *head = &oe->events;
+	struct ordered_event *tmp, *iter;
+	u64 limit = oe->next_flush;
+	u64 last_ts = oe->last ? oe->last->timestamp : 0ULL;
+	bool show_progress = limit == ULLONG_MAX;
+	struct ui_progress prog;
+	int ret;
+
+	if (!limit)
+		return 0;
+
+	if (show_progress)
+		ui_progress__init(&prog, oe->nr_events, "Processing time ordered events...");
+
+	list_for_each_entry_safe(iter, tmp, head, list) {
+		if (session_done())
+			return 0;
+
+		if (iter->timestamp > limit)
+			break;
+		ret = oe->deliver(oe, iter);
+		if (ret)
+			return ret;
+
+		ordered_events__delete(oe, iter);
+		oe->last_flush = iter->timestamp;
+
+		if (show_progress)
+			ui_progress__update(&prog, 1);
+	}
+
+	if (list_empty(head))
+		oe->last = NULL;
+	else if (last_ts <= limit)
+		oe->last = list_entry(head->prev, struct ordered_event, list);
+
+	return 0;
+}
+
+int ordered_events__flush(struct ordered_events *oe, enum oe_flush how)
+{
+	static const char * const str[] = {
+		"NONE",
+		"FINAL",
+		"ROUND",
+		"HALF ",
+	};
+	int err;
+
+	if (oe->nr_events == 0)
+		return 0;
+
+	switch (how) {
+	case OE_FLUSH__FINAL:
+		oe->next_flush = ULLONG_MAX;
+		break;
+
+	case OE_FLUSH__HALF:
+	{
+		struct ordered_event *first, *last;
+		struct list_head *head = &oe->events;
+
+		first = list_entry(head->next, struct ordered_event, list);
+		last = oe->last;
+
+		/* Warn if we are called before any event got allocated. */
+		if (WARN_ONCE(!last || list_empty(head), "empty queue"))
+			return 0;
+
+		oe->next_flush  = first->timestamp;
+		oe->next_flush += (last->timestamp - first->timestamp) / 2;
+		break;
+	}
+
+	case OE_FLUSH__ROUND:
+	case OE_FLUSH__NONE:
+	default:
+		break;
+	};
+
+	pr_oe_time(oe->next_flush, "next_flush - ordered_events__flush PRE  %s, nr_events %u\n",
+		   str[how], oe->nr_events);
+	pr_oe_time(oe->max_timestamp, "max_timestamp\n");
+
+	err = __ordered_events__flush(oe);
+
+	if (!err) {
+		if (how == OE_FLUSH__ROUND)
+			oe->next_flush = oe->max_timestamp;
+
+		oe->last_flush_type = how;
+	}
+
+	pr_oe_time(oe->next_flush, "next_flush - ordered_events__flush POST %s, nr_events %u\n",
+		   str[how], oe->nr_events);
+	pr_oe_time(oe->last_flush, "last_flush\n");
+
+	return err;
+}
+
+void ordered_events__init(struct ordered_events *oe, ordered_events__deliver_t deliver)
+{
+	INIT_LIST_HEAD(&oe->events);
+	INIT_LIST_HEAD(&oe->cache);
+	INIT_LIST_HEAD(&oe->to_free);
+	oe->max_alloc_size = (u64) -1;
+	oe->cur_alloc_size = 0;
+	oe->deliver	   = deliver;
+}
+
+void ordered_events__free(struct ordered_events *oe)
+{
+	while (!list_empty(&oe->to_free)) {
+		struct ordered_event *event;
+
+		event = list_entry(oe->to_free.next, struct ordered_event, list);
+		list_del(&event->list);
+		free_dup_event(oe, event->event);
+		free(event);
+	}
+}
diff --git a/tools/perf/util/ordered-events.h b/tools/perf/util/ordered-events.h
new file mode 100644
index 000000000..f403991e3
--- /dev/null
+++ b/tools/perf/util/ordered-events.h
@@ -0,0 +1,64 @@
+#ifndef __ORDERED_EVENTS_H
+#define __ORDERED_EVENTS_H
+
+#include <linux/types.h>
+
+struct perf_sample;
+
+struct ordered_event {
+	u64			timestamp;
+	u64			file_offset;
+	union perf_event	*event;
+	struct list_head	list;
+};
+
+enum oe_flush {
+	OE_FLUSH__NONE,
+	OE_FLUSH__FINAL,
+	OE_FLUSH__ROUND,
+	OE_FLUSH__HALF,
+};
+
+struct ordered_events;
+
+typedef int (*ordered_events__deliver_t)(struct ordered_events *oe,
+					 struct ordered_event *event);
+
+struct ordered_events {
+	u64			last_flush;
+	u64			next_flush;
+	u64			max_timestamp;
+	u64			max_alloc_size;
+	u64			cur_alloc_size;
+	struct list_head	events;
+	struct list_head	cache;
+	struct list_head	to_free;
+	struct ordered_event	*buffer;
+	struct ordered_event	*last;
+	ordered_events__deliver_t deliver;
+	int			buffer_idx;
+	unsigned int		nr_events;
+	enum oe_flush		last_flush_type;
+	u32			nr_unordered_events;
+	bool                    copy_on_queue;
+};
+
+int ordered_events__queue(struct ordered_events *oe, union perf_event *event,
+			  struct perf_sample *sample, u64 file_offset);
+void ordered_events__delete(struct ordered_events *oe, struct ordered_event *event);
+int ordered_events__flush(struct ordered_events *oe, enum oe_flush how);
+void ordered_events__init(struct ordered_events *oe, ordered_events__deliver_t deliver);
+void ordered_events__free(struct ordered_events *oe);
+
+static inline
+void ordered_events__set_alloc_size(struct ordered_events *oe, u64 size)
+{
+	oe->max_alloc_size = size;
+}
+
+static inline
+void ordered_events__set_copy_on_queue(struct ordered_events *oe, bool copy)
+{
+	oe->copy_on_queue = copy;
+}
+#endif /* __ORDERED_EVENTS_H */
diff --git a/tools/perf/util/pager.c b/tools/perf/util/pager.c
new file mode 100644
index 000000000..31ee02d4e
--- /dev/null
+++ b/tools/perf/util/pager.c
@@ -0,0 +1,100 @@
+#include "cache.h"
+#include "run-command.h"
+#include "sigchain.h"
+
+/*
+ * This is split up from the rest of git so that we can do
+ * something different on Windows.
+ */
+
+static int spawned_pager;
+
+static void pager_preexec(void)
+{
+	/*
+	 * Work around bug in "less" by not starting it until we
+	 * have real input
+	 */
+	fd_set in;
+
+	FD_ZERO(&in);
+	FD_SET(0, &in);
+	select(1, &in, NULL, &in, NULL);
+
+	setenv("LESS", "FRSX", 0);
+}
+
+static const char *pager_argv[] = { "sh", "-c", NULL, NULL };
+static struct child_process pager_process;
+
+static void wait_for_pager(void)
+{
+	fflush(stdout);
+	fflush(stderr);
+	/* signal EOF to pager */
+	close(1);
+	close(2);
+	finish_command(&pager_process);
+}
+
+static void wait_for_pager_signal(int signo)
+{
+	wait_for_pager();
+	sigchain_pop(signo);
+	raise(signo);
+}
+
+void setup_pager(void)
+{
+	const char *pager = getenv("PERF_PAGER");
+
+	if (!isatty(1))
+		return;
+	if (!pager) {
+		if (!pager_program)
+			perf_config(perf_default_config, NULL);
+		pager = pager_program;
+	}
+	if (!pager)
+		pager = getenv("PAGER");
+	if (!(pager || access("/usr/bin/pager", X_OK)))
+		pager = "/usr/bin/pager";
+	if (!(pager || access("/usr/bin/less", X_OK)))
+		pager = "/usr/bin/less";
+	if (!pager)
+		pager = "cat";
+	if (!*pager || !strcmp(pager, "cat"))
+		return;
+
+	spawned_pager = 1; /* means we are emitting to terminal */
+
+	/* spawn the pager */
+	pager_argv[2] = pager;
+	pager_process.argv = pager_argv;
+	pager_process.in = -1;
+	pager_process.preexec_cb = pager_preexec;
+
+	if (start_command(&pager_process))
+		return;
+
+	/* original process continues, but writes to the pipe */
+	dup2(pager_process.in, 1);
+	if (isatty(2))
+		dup2(pager_process.in, 2);
+	close(pager_process.in);
+
+	/* this makes sure that the parent terminates after the pager */
+	sigchain_push_common(wait_for_pager_signal);
+	atexit(wait_for_pager);
+}
+
+int pager_in_use(void)
+{
+	const char *env;
+
+	if (spawned_pager)
+		return 1;
+
+	env = getenv("PERF_PAGER_IN_USE");
+	return env ? perf_config_bool("PERF_PAGER_IN_USE", env) : 0;
+}
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
new file mode 100644
index 000000000..be0655388
--- /dev/null
+++ b/tools/perf/util/parse-events.c
@@ -0,0 +1,1537 @@
+#include <linux/hw_breakpoint.h>
+#include "util.h"
+#include "../perf.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "parse-options.h"
+#include "parse-events.h"
+#include "exec_cmd.h"
+#include "string.h"
+#include "symbol.h"
+#include "cache.h"
+#include "header.h"
+#include "debug.h"
+#include <api/fs/debugfs.h>
+#include "parse-events-bison.h"
+#define YY_EXTRA_TYPE int
+#include "parse-events-flex.h"
+#include "pmu.h"
+#include "thread_map.h"
+
+#define MAX_NAME_LEN 100
+
+#ifdef PARSER_DEBUG
+extern int parse_events_debug;
+#endif
+int parse_events_parse(void *data, void *scanner);
+
+static struct perf_pmu_event_symbol *perf_pmu_events_list;
+/*
+ * The variable indicates the number of supported pmu event symbols.
+ * 0 means not initialized and ready to init
+ * -1 means failed to init, don't try anymore
+ * >0 is the number of supported pmu event symbols
+ */
+static int perf_pmu_events_list_num;
+
+struct event_symbol event_symbols_hw[PERF_COUNT_HW_MAX] = {
+	[PERF_COUNT_HW_CPU_CYCLES] = {
+		.symbol = "cpu-cycles",
+		.alias  = "cycles",
+	},
+	[PERF_COUNT_HW_INSTRUCTIONS] = {
+		.symbol = "instructions",
+		.alias  = "",
+	},
+	[PERF_COUNT_HW_CACHE_REFERENCES] = {
+		.symbol = "cache-references",
+		.alias  = "",
+	},
+	[PERF_COUNT_HW_CACHE_MISSES] = {
+		.symbol = "cache-misses",
+		.alias  = "",
+	},
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = {
+		.symbol = "branch-instructions",
+		.alias  = "branches",
+	},
+	[PERF_COUNT_HW_BRANCH_MISSES] = {
+		.symbol = "branch-misses",
+		.alias  = "",
+	},
+	[PERF_COUNT_HW_BUS_CYCLES] = {
+		.symbol = "bus-cycles",
+		.alias  = "",
+	},
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = {
+		.symbol = "stalled-cycles-frontend",
+		.alias  = "idle-cycles-frontend",
+	},
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = {
+		.symbol = "stalled-cycles-backend",
+		.alias  = "idle-cycles-backend",
+	},
+	[PERF_COUNT_HW_REF_CPU_CYCLES] = {
+		.symbol = "ref-cycles",
+		.alias  = "",
+	},
+};
+
+struct event_symbol event_symbols_sw[PERF_COUNT_SW_MAX] = {
+	[PERF_COUNT_SW_CPU_CLOCK] = {
+		.symbol = "cpu-clock",
+		.alias  = "",
+	},
+	[PERF_COUNT_SW_TASK_CLOCK] = {
+		.symbol = "task-clock",
+		.alias  = "",
+	},
+	[PERF_COUNT_SW_PAGE_FAULTS] = {
+		.symbol = "page-faults",
+		.alias  = "faults",
+	},
+	[PERF_COUNT_SW_CONTEXT_SWITCHES] = {
+		.symbol = "context-switches",
+		.alias  = "cs",
+	},
+	[PERF_COUNT_SW_CPU_MIGRATIONS] = {
+		.symbol = "cpu-migrations",
+		.alias  = "migrations",
+	},
+	[PERF_COUNT_SW_PAGE_FAULTS_MIN] = {
+		.symbol = "minor-faults",
+		.alias  = "",
+	},
+	[PERF_COUNT_SW_PAGE_FAULTS_MAJ] = {
+		.symbol = "major-faults",
+		.alias  = "",
+	},
+	[PERF_COUNT_SW_ALIGNMENT_FAULTS] = {
+		.symbol = "alignment-faults",
+		.alias  = "",
+	},
+	[PERF_COUNT_SW_EMULATION_FAULTS] = {
+		.symbol = "emulation-faults",
+		.alias  = "",
+	},
+	[PERF_COUNT_SW_DUMMY] = {
+		.symbol = "dummy",
+		.alias  = "",
+	},
+};
+
+#define __PERF_EVENT_FIELD(config, name) \
+	((config & PERF_EVENT_##name##_MASK) >> PERF_EVENT_##name##_SHIFT)
+
+#define PERF_EVENT_RAW(config)		__PERF_EVENT_FIELD(config, RAW)
+#define PERF_EVENT_CONFIG(config)	__PERF_EVENT_FIELD(config, CONFIG)
+#define PERF_EVENT_TYPE(config)		__PERF_EVENT_FIELD(config, TYPE)
+#define PERF_EVENT_ID(config)		__PERF_EVENT_FIELD(config, EVENT)
+
+#define for_each_subsystem(sys_dir, sys_dirent, sys_next)	       \
+	while (!readdir_r(sys_dir, &sys_dirent, &sys_next) && sys_next)	       \
+	if (sys_dirent.d_type == DT_DIR &&				       \
+	   (strcmp(sys_dirent.d_name, ".")) &&				       \
+	   (strcmp(sys_dirent.d_name, "..")))
+
+static int tp_event_has_id(struct dirent *sys_dir, struct dirent *evt_dir)
+{
+	char evt_path[MAXPATHLEN];
+	int fd;
+
+	snprintf(evt_path, MAXPATHLEN, "%s/%s/%s/id", tracing_events_path,
+			sys_dir->d_name, evt_dir->d_name);
+	fd = open(evt_path, O_RDONLY);
+	if (fd < 0)
+		return -EINVAL;
+	close(fd);
+
+	return 0;
+}
+
+#define for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next)	       \
+	while (!readdir_r(evt_dir, &evt_dirent, &evt_next) && evt_next)        \
+	if (evt_dirent.d_type == DT_DIR &&				       \
+	   (strcmp(evt_dirent.d_name, ".")) &&				       \
+	   (strcmp(evt_dirent.d_name, "..")) &&				       \
+	   (!tp_event_has_id(&sys_dirent, &evt_dirent)))
+
+#define MAX_EVENT_LENGTH 512
+
+
+struct tracepoint_path *tracepoint_id_to_path(u64 config)
+{
+	struct tracepoint_path *path = NULL;
+	DIR *sys_dir, *evt_dir;
+	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
+	char id_buf[24];
+	int fd;
+	u64 id;
+	char evt_path[MAXPATHLEN];
+	char dir_path[MAXPATHLEN];
+
+	sys_dir = opendir(tracing_events_path);
+	if (!sys_dir)
+		return NULL;
+
+	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+
+		snprintf(dir_path, MAXPATHLEN, "%s/%s", tracing_events_path,
+			 sys_dirent.d_name);
+		evt_dir = opendir(dir_path);
+		if (!evt_dir)
+			continue;
+
+		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+
+			snprintf(evt_path, MAXPATHLEN, "%s/%s/id", dir_path,
+				 evt_dirent.d_name);
+			fd = open(evt_path, O_RDONLY);
+			if (fd < 0)
+				continue;
+			if (read(fd, id_buf, sizeof(id_buf)) < 0) {
+				close(fd);
+				continue;
+			}
+			close(fd);
+			id = atoll(id_buf);
+			if (id == config) {
+				closedir(evt_dir);
+				closedir(sys_dir);
+				path = zalloc(sizeof(*path));
+				path->system = malloc(MAX_EVENT_LENGTH);
+				if (!path->system) {
+					free(path);
+					return NULL;
+				}
+				path->name = malloc(MAX_EVENT_LENGTH);
+				if (!path->name) {
+					zfree(&path->system);
+					free(path);
+					return NULL;
+				}
+				strncpy(path->system, sys_dirent.d_name,
+					MAX_EVENT_LENGTH);
+				strncpy(path->name, evt_dirent.d_name,
+					MAX_EVENT_LENGTH);
+				return path;
+			}
+		}
+		closedir(evt_dir);
+	}
+
+	closedir(sys_dir);
+	return NULL;
+}
+
+struct tracepoint_path *tracepoint_name_to_path(const char *name)
+{
+	struct tracepoint_path *path = zalloc(sizeof(*path));
+	char *str = strchr(name, ':');
+
+	if (path == NULL || str == NULL) {
+		free(path);
+		return NULL;
+	}
+
+	path->system = strndup(name, str - name);
+	path->name = strdup(str+1);
+
+	if (path->system == NULL || path->name == NULL) {
+		zfree(&path->system);
+		zfree(&path->name);
+		free(path);
+		path = NULL;
+	}
+
+	return path;
+}
+
+const char *event_type(int type)
+{
+	switch (type) {
+	case PERF_TYPE_HARDWARE:
+		return "hardware";
+
+	case PERF_TYPE_SOFTWARE:
+		return "software";
+
+	case PERF_TYPE_TRACEPOINT:
+		return "tracepoint";
+
+	case PERF_TYPE_HW_CACHE:
+		return "hardware-cache";
+
+	default:
+		break;
+	}
+
+	return "unknown";
+}
+
+
+
+static struct perf_evsel *
+__add_event(struct list_head *list, int *idx,
+	    struct perf_event_attr *attr,
+	    char *name, struct cpu_map *cpus)
+{
+	struct perf_evsel *evsel;
+
+	event_attr_init(attr);
+
+	evsel = perf_evsel__new_idx(attr, (*idx)++);
+	if (!evsel)
+		return NULL;
+
+	evsel->cpus = cpus;
+	if (name)
+		evsel->name = strdup(name);
+	list_add_tail(&evsel->node, list);
+	return evsel;
+}
+
+static int add_event(struct list_head *list, int *idx,
+		     struct perf_event_attr *attr, char *name)
+{
+	return __add_event(list, idx, attr, name, NULL) ? 0 : -ENOMEM;
+}
+
+static int parse_aliases(char *str, const char *names[][PERF_EVSEL__MAX_ALIASES], int size)
+{
+	int i, j;
+	int n, longest = -1;
+
+	for (i = 0; i < size; i++) {
+		for (j = 0; j < PERF_EVSEL__MAX_ALIASES && names[i][j]; j++) {
+			n = strlen(names[i][j]);
+			if (n > longest && !strncasecmp(str, names[i][j], n))
+				longest = n;
+		}
+		if (longest > 0)
+			return i;
+	}
+
+	return -1;
+}
+
+int parse_events_add_cache(struct list_head *list, int *idx,
+			   char *type, char *op_result1, char *op_result2)
+{
+	struct perf_event_attr attr;
+	char name[MAX_NAME_LEN];
+	int cache_type = -1, cache_op = -1, cache_result = -1;
+	char *op_result[2] = { op_result1, op_result2 };
+	int i, n;
+
+	/*
+	 * No fallback - if we cannot get a clear cache type
+	 * then bail out:
+	 */
+	cache_type = parse_aliases(type, perf_evsel__hw_cache,
+				   PERF_COUNT_HW_CACHE_MAX);
+	if (cache_type == -1)
+		return -EINVAL;
+
+	n = snprintf(name, MAX_NAME_LEN, "%s", type);
+
+	for (i = 0; (i < 2) && (op_result[i]); i++) {
+		char *str = op_result[i];
+
+		n += snprintf(name + n, MAX_NAME_LEN - n, "-%s", str);
+
+		if (cache_op == -1) {
+			cache_op = parse_aliases(str, perf_evsel__hw_cache_op,
+						 PERF_COUNT_HW_CACHE_OP_MAX);
+			if (cache_op >= 0) {
+				if (!perf_evsel__is_cache_op_valid(cache_type, cache_op))
+					return -EINVAL;
+				continue;
+			}
+		}
+
+		if (cache_result == -1) {
+			cache_result = parse_aliases(str, perf_evsel__hw_cache_result,
+						     PERF_COUNT_HW_CACHE_RESULT_MAX);
+			if (cache_result >= 0)
+				continue;
+		}
+	}
+
+	/*
+	 * Fall back to reads:
+	 */
+	if (cache_op == -1)
+		cache_op = PERF_COUNT_HW_CACHE_OP_READ;
+
+	/*
+	 * Fall back to accesses:
+	 */
+	if (cache_result == -1)
+		cache_result = PERF_COUNT_HW_CACHE_RESULT_ACCESS;
+
+	memset(&attr, 0, sizeof(attr));
+	attr.config = cache_type | (cache_op << 8) | (cache_result << 16);
+	attr.type = PERF_TYPE_HW_CACHE;
+	return add_event(list, idx, &attr, name);
+}
+
+static int add_tracepoint(struct list_head *list, int *idx,
+			  char *sys_name, char *evt_name)
+{
+	struct perf_evsel *evsel;
+
+	evsel = perf_evsel__newtp_idx(sys_name, evt_name, (*idx)++);
+	if (!evsel)
+		return -ENOMEM;
+
+	list_add_tail(&evsel->node, list);
+
+	return 0;
+}
+
+static int add_tracepoint_multi_event(struct list_head *list, int *idx,
+				      char *sys_name, char *evt_name)
+{
+	char evt_path[MAXPATHLEN];
+	struct dirent *evt_ent;
+	DIR *evt_dir;
+	int ret = 0;
+
+	snprintf(evt_path, MAXPATHLEN, "%s/%s", tracing_events_path, sys_name);
+	evt_dir = opendir(evt_path);
+	if (!evt_dir) {
+		perror("Can't open event dir");
+		return -1;
+	}
+
+	while (!ret && (evt_ent = readdir(evt_dir))) {
+		if (!strcmp(evt_ent->d_name, ".")
+		    || !strcmp(evt_ent->d_name, "..")
+		    || !strcmp(evt_ent->d_name, "enable")
+		    || !strcmp(evt_ent->d_name, "filter"))
+			continue;
+
+		if (!strglobmatch(evt_ent->d_name, evt_name))
+			continue;
+
+		ret = add_tracepoint(list, idx, sys_name, evt_ent->d_name);
+	}
+
+	closedir(evt_dir);
+	return ret;
+}
+
+static int add_tracepoint_event(struct list_head *list, int *idx,
+				char *sys_name, char *evt_name)
+{
+	return strpbrk(evt_name, "*?") ?
+	       add_tracepoint_multi_event(list, idx, sys_name, evt_name) :
+	       add_tracepoint(list, idx, sys_name, evt_name);
+}
+
+static int add_tracepoint_multi_sys(struct list_head *list, int *idx,
+				    char *sys_name, char *evt_name)
+{
+	struct dirent *events_ent;
+	DIR *events_dir;
+	int ret = 0;
+
+	events_dir = opendir(tracing_events_path);
+	if (!events_dir) {
+		perror("Can't open event dir");
+		return -1;
+	}
+
+	while (!ret && (events_ent = readdir(events_dir))) {
+		if (!strcmp(events_ent->d_name, ".")
+		    || !strcmp(events_ent->d_name, "..")
+		    || !strcmp(events_ent->d_name, "enable")
+		    || !strcmp(events_ent->d_name, "header_event")
+		    || !strcmp(events_ent->d_name, "header_page"))
+			continue;
+
+		if (!strglobmatch(events_ent->d_name, sys_name))
+			continue;
+
+		ret = add_tracepoint_event(list, idx, events_ent->d_name,
+					   evt_name);
+	}
+
+	closedir(events_dir);
+	return ret;
+}
+
+int parse_events_add_tracepoint(struct list_head *list, int *idx,
+				char *sys, char *event)
+{
+	if (strpbrk(sys, "*?"))
+		return add_tracepoint_multi_sys(list, idx, sys, event);
+	else
+		return add_tracepoint_event(list, idx, sys, event);
+}
+
+static int
+parse_breakpoint_type(const char *type, struct perf_event_attr *attr)
+{
+	int i;
+
+	for (i = 0; i < 3; i++) {
+		if (!type || !type[i])
+			break;
+
+#define CHECK_SET_TYPE(bit)		\
+do {					\
+	if (attr->bp_type & bit)	\
+		return -EINVAL;		\
+	else				\
+		attr->bp_type |= bit;	\
+} while (0)
+
+		switch (type[i]) {
+		case 'r':
+			CHECK_SET_TYPE(HW_BREAKPOINT_R);
+			break;
+		case 'w':
+			CHECK_SET_TYPE(HW_BREAKPOINT_W);
+			break;
+		case 'x':
+			CHECK_SET_TYPE(HW_BREAKPOINT_X);
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
+
+#undef CHECK_SET_TYPE
+
+	if (!attr->bp_type) /* Default */
+		attr->bp_type = HW_BREAKPOINT_R | HW_BREAKPOINT_W;
+
+	return 0;
+}
+
+int parse_events_add_breakpoint(struct list_head *list, int *idx,
+				void *ptr, char *type, u64 len)
+{
+	struct perf_event_attr attr;
+
+	memset(&attr, 0, sizeof(attr));
+	attr.bp_addr = (unsigned long) ptr;
+
+	if (parse_breakpoint_type(type, &attr))
+		return -EINVAL;
+
+	/* Provide some defaults if len is not specified */
+	if (!len) {
+		if (attr.bp_type == HW_BREAKPOINT_X)
+			len = sizeof(long);
+		else
+			len = HW_BREAKPOINT_LEN_4;
+	}
+
+	attr.bp_len = len;
+
+	attr.type = PERF_TYPE_BREAKPOINT;
+	attr.sample_period = 1;
+
+	return add_event(list, idx, &attr, NULL);
+}
+
+static int config_term(struct perf_event_attr *attr,
+		       struct parse_events_term *term)
+{
+#define CHECK_TYPE_VAL(type)					\
+do {								\
+	if (PARSE_EVENTS__TERM_TYPE_ ## type != term->type_val)	\
+		return -EINVAL;					\
+} while (0)
+
+	switch (term->type_term) {
+	case PARSE_EVENTS__TERM_TYPE_CONFIG:
+		CHECK_TYPE_VAL(NUM);
+		attr->config = term->val.num;
+		break;
+	case PARSE_EVENTS__TERM_TYPE_CONFIG1:
+		CHECK_TYPE_VAL(NUM);
+		attr->config1 = term->val.num;
+		break;
+	case PARSE_EVENTS__TERM_TYPE_CONFIG2:
+		CHECK_TYPE_VAL(NUM);
+		attr->config2 = term->val.num;
+		break;
+	case PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD:
+		CHECK_TYPE_VAL(NUM);
+		attr->sample_period = term->val.num;
+		break;
+	case PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE:
+		/*
+		 * TODO uncomment when the field is available
+		 * attr->branch_sample_type = term->val.num;
+		 */
+		break;
+	case PARSE_EVENTS__TERM_TYPE_NAME:
+		CHECK_TYPE_VAL(STR);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+#undef CHECK_TYPE_VAL
+}
+
+static int config_attr(struct perf_event_attr *attr,
+		       struct list_head *head, int fail)
+{
+	struct parse_events_term *term;
+
+	list_for_each_entry(term, head, list)
+		if (config_term(attr, term) && fail)
+			return -EINVAL;
+
+	return 0;
+}
+
+int parse_events_add_numeric(struct list_head *list, int *idx,
+			     u32 type, u64 config,
+			     struct list_head *head_config)
+{
+	struct perf_event_attr attr;
+
+	memset(&attr, 0, sizeof(attr));
+	attr.type = type;
+	attr.config = config;
+
+	if (head_config &&
+	    config_attr(&attr, head_config, 1))
+		return -EINVAL;
+
+	return add_event(list, idx, &attr, NULL);
+}
+
+static int parse_events__is_name_term(struct parse_events_term *term)
+{
+	return term->type_term == PARSE_EVENTS__TERM_TYPE_NAME;
+}
+
+static char *pmu_event_name(struct list_head *head_terms)
+{
+	struct parse_events_term *term;
+
+	list_for_each_entry(term, head_terms, list)
+		if (parse_events__is_name_term(term))
+			return term->val.str;
+
+	return NULL;
+}
+
+int parse_events_add_pmu(struct list_head *list, int *idx,
+			 char *name, struct list_head *head_config)
+{
+	struct perf_event_attr attr;
+	struct perf_pmu_info info;
+	struct perf_pmu *pmu;
+	struct perf_evsel *evsel;
+
+	pmu = perf_pmu__find(name);
+	if (!pmu)
+		return -EINVAL;
+
+	if (pmu->default_config) {
+		memcpy(&attr, pmu->default_config,
+		       sizeof(struct perf_event_attr));
+	} else {
+		memset(&attr, 0, sizeof(attr));
+	}
+
+	if (!head_config) {
+		attr.type = pmu->type;
+		evsel = __add_event(list, idx, &attr, NULL, pmu->cpus);
+		return evsel ? 0 : -ENOMEM;
+	}
+
+	if (perf_pmu__check_alias(pmu, head_config, &info))
+		return -EINVAL;
+
+	/*
+	 * Configure hardcoded terms first, no need to check
+	 * return value when called with fail == 0 ;)
+	 */
+	config_attr(&attr, head_config, 0);
+
+	if (perf_pmu__config(pmu, &attr, head_config))
+		return -EINVAL;
+
+	evsel = __add_event(list, idx, &attr, pmu_event_name(head_config),
+			    pmu->cpus);
+	if (evsel) {
+		evsel->unit = info.unit;
+		evsel->scale = info.scale;
+		evsel->per_pkg = info.per_pkg;
+		evsel->snapshot = info.snapshot;
+	}
+
+	return evsel ? 0 : -ENOMEM;
+}
+
+int parse_events__modifier_group(struct list_head *list,
+				 char *event_mod)
+{
+	return parse_events__modifier_event(list, event_mod, true);
+}
+
+void parse_events__set_leader(char *name, struct list_head *list)
+{
+	struct perf_evsel *leader;
+
+	__perf_evlist__set_leader(list);
+	leader = list_entry(list->next, struct perf_evsel, node);
+	leader->group_name = name ? strdup(name) : NULL;
+}
+
+/* list_event is assumed to point to malloc'ed memory */
+void parse_events_update_lists(struct list_head *list_event,
+			       struct list_head *list_all)
+{
+	/*
+	 * Called for single event definition. Update the
+	 * 'all event' list, and reinit the 'single event'
+	 * list, for next event definition.
+	 */
+	list_splice_tail(list_event, list_all);
+	free(list_event);
+}
+
+struct event_modifier {
+	int eu;
+	int ek;
+	int eh;
+	int eH;
+	int eG;
+	int eI;
+	int precise;
+	int exclude_GH;
+	int sample_read;
+	int pinned;
+};
+
+static int get_event_modifier(struct event_modifier *mod, char *str,
+			       struct perf_evsel *evsel)
+{
+	int eu = evsel ? evsel->attr.exclude_user : 0;
+	int ek = evsel ? evsel->attr.exclude_kernel : 0;
+	int eh = evsel ? evsel->attr.exclude_hv : 0;
+	int eH = evsel ? evsel->attr.exclude_host : 0;
+	int eG = evsel ? evsel->attr.exclude_guest : 0;
+	int eI = evsel ? evsel->attr.exclude_idle : 0;
+	int precise = evsel ? evsel->attr.precise_ip : 0;
+	int sample_read = 0;
+	int pinned = evsel ? evsel->attr.pinned : 0;
+
+	int exclude = eu | ek | eh;
+	int exclude_GH = evsel ? evsel->exclude_GH : 0;
+
+	memset(mod, 0, sizeof(*mod));
+
+	while (*str) {
+		if (*str == 'u') {
+			if (!exclude)
+				exclude = eu = ek = eh = 1;
+			eu = 0;
+		} else if (*str == 'k') {
+			if (!exclude)
+				exclude = eu = ek = eh = 1;
+			ek = 0;
+		} else if (*str == 'h') {
+			if (!exclude)
+				exclude = eu = ek = eh = 1;
+			eh = 0;
+		} else if (*str == 'G') {
+			if (!exclude_GH)
+				exclude_GH = eG = eH = 1;
+			eG = 0;
+		} else if (*str == 'H') {
+			if (!exclude_GH)
+				exclude_GH = eG = eH = 1;
+			eH = 0;
+		} else if (*str == 'I') {
+			eI = 1;
+		} else if (*str == 'p') {
+			precise++;
+			/* use of precise requires exclude_guest */
+			if (!exclude_GH)
+				eG = 1;
+		} else if (*str == 'S') {
+			sample_read = 1;
+		} else if (*str == 'D') {
+			pinned = 1;
+		} else
+			break;
+
+		++str;
+	}
+
+	/*
+	 * precise ip:
+	 *
+	 *  0 - SAMPLE_IP can have arbitrary skid
+	 *  1 - SAMPLE_IP must have constant skid
+	 *  2 - SAMPLE_IP requested to have 0 skid
+	 *  3 - SAMPLE_IP must have 0 skid
+	 *
+	 *  See also PERF_RECORD_MISC_EXACT_IP
+	 */
+	if (precise > 3)
+		return -EINVAL;
+
+	mod->eu = eu;
+	mod->ek = ek;
+	mod->eh = eh;
+	mod->eH = eH;
+	mod->eG = eG;
+	mod->eI = eI;
+	mod->precise = precise;
+	mod->exclude_GH = exclude_GH;
+	mod->sample_read = sample_read;
+	mod->pinned = pinned;
+
+	return 0;
+}
+
+/*
+ * Basic modifier sanity check to validate it contains only one
+ * instance of any modifier (apart from 'p') present.
+ */
+static int check_modifier(char *str)
+{
+	char *p = str;
+
+	/* The sizeof includes 0 byte as well. */
+	if (strlen(str) > (sizeof("ukhGHpppSDI") - 1))
+		return -1;
+
+	while (*p) {
+		if (*p != 'p' && strchr(p + 1, *p))
+			return -1;
+		p++;
+	}
+
+	return 0;
+}
+
+int parse_events__modifier_event(struct list_head *list, char *str, bool add)
+{
+	struct perf_evsel *evsel;
+	struct event_modifier mod;
+
+	if (str == NULL)
+		return 0;
+
+	if (check_modifier(str))
+		return -EINVAL;
+
+	if (!add && get_event_modifier(&mod, str, NULL))
+		return -EINVAL;
+
+	__evlist__for_each(list, evsel) {
+		if (add && get_event_modifier(&mod, str, evsel))
+			return -EINVAL;
+
+		evsel->attr.exclude_user   = mod.eu;
+		evsel->attr.exclude_kernel = mod.ek;
+		evsel->attr.exclude_hv     = mod.eh;
+		evsel->attr.precise_ip     = mod.precise;
+		evsel->attr.exclude_host   = mod.eH;
+		evsel->attr.exclude_guest  = mod.eG;
+		evsel->attr.exclude_idle   = mod.eI;
+		evsel->exclude_GH          = mod.exclude_GH;
+		evsel->sample_read         = mod.sample_read;
+
+		if (perf_evsel__is_group_leader(evsel))
+			evsel->attr.pinned = mod.pinned;
+	}
+
+	return 0;
+}
+
+int parse_events_name(struct list_head *list, char *name)
+{
+	struct perf_evsel *evsel;
+
+	__evlist__for_each(list, evsel) {
+		if (!evsel->name)
+			evsel->name = strdup(name);
+	}
+
+	return 0;
+}
+
+static int
+comp_pmu(const void *p1, const void *p2)
+{
+	struct perf_pmu_event_symbol *pmu1 = (struct perf_pmu_event_symbol *) p1;
+	struct perf_pmu_event_symbol *pmu2 = (struct perf_pmu_event_symbol *) p2;
+
+	return strcmp(pmu1->symbol, pmu2->symbol);
+}
+
+static void perf_pmu__parse_cleanup(void)
+{
+	if (perf_pmu_events_list_num > 0) {
+		struct perf_pmu_event_symbol *p;
+		int i;
+
+		for (i = 0; i < perf_pmu_events_list_num; i++) {
+			p = perf_pmu_events_list + i;
+			free(p->symbol);
+		}
+		free(perf_pmu_events_list);
+		perf_pmu_events_list = NULL;
+		perf_pmu_events_list_num = 0;
+	}
+}
+
+#define SET_SYMBOL(str, stype)		\
+do {					\
+	p->symbol = str;		\
+	if (!p->symbol)			\
+		goto err;		\
+	p->type = stype;		\
+} while (0)
+
+/*
+ * Read the pmu events list from sysfs
+ * Save it into perf_pmu_events_list
+ */
+static void perf_pmu__parse_init(void)
+{
+
+	struct perf_pmu *pmu = NULL;
+	struct perf_pmu_alias *alias;
+	int len = 0;
+
+	pmu = perf_pmu__find("cpu");
+	if ((pmu == NULL) || list_empty(&pmu->aliases)) {
+		perf_pmu_events_list_num = -1;
+		return;
+	}
+	list_for_each_entry(alias, &pmu->aliases, list) {
+		if (strchr(alias->name, '-'))
+			len++;
+		len++;
+	}
+	perf_pmu_events_list = malloc(sizeof(struct perf_pmu_event_symbol) * len);
+	if (!perf_pmu_events_list)
+		return;
+	perf_pmu_events_list_num = len;
+
+	len = 0;
+	list_for_each_entry(alias, &pmu->aliases, list) {
+		struct perf_pmu_event_symbol *p = perf_pmu_events_list + len;
+		char *tmp = strchr(alias->name, '-');
+
+		if (tmp != NULL) {
+			SET_SYMBOL(strndup(alias->name, tmp - alias->name),
+					PMU_EVENT_SYMBOL_PREFIX);
+			p++;
+			SET_SYMBOL(strdup(++tmp), PMU_EVENT_SYMBOL_SUFFIX);
+			len += 2;
+		} else {
+			SET_SYMBOL(strdup(alias->name), PMU_EVENT_SYMBOL);
+			len++;
+		}
+	}
+	qsort(perf_pmu_events_list, len,
+		sizeof(struct perf_pmu_event_symbol), comp_pmu);
+
+	return;
+err:
+	perf_pmu__parse_cleanup();
+}
+
+enum perf_pmu_event_symbol_type
+perf_pmu__parse_check(const char *name)
+{
+	struct perf_pmu_event_symbol p, *r;
+
+	/* scan kernel pmu events from sysfs if needed */
+	if (perf_pmu_events_list_num == 0)
+		perf_pmu__parse_init();
+	/*
+	 * name "cpu" could be prefix of cpu-cycles or cpu// events.
+	 * cpu-cycles has been handled by hardcode.
+	 * So it must be cpu// events, not kernel pmu event.
+	 */
+	if ((perf_pmu_events_list_num <= 0) || !strcmp(name, "cpu"))
+		return PMU_EVENT_SYMBOL_ERR;
+
+	p.symbol = strdup(name);
+	r = bsearch(&p, perf_pmu_events_list,
+			(size_t) perf_pmu_events_list_num,
+			sizeof(struct perf_pmu_event_symbol), comp_pmu);
+	free(p.symbol);
+	return r ? r->type : PMU_EVENT_SYMBOL_ERR;
+}
+
+static int parse_events__scanner(const char *str, void *data, int start_token)
+{
+	YY_BUFFER_STATE buffer;
+	void *scanner;
+	int ret;
+
+	ret = parse_events_lex_init_extra(start_token, &scanner);
+	if (ret)
+		return ret;
+
+	buffer = parse_events__scan_string(str, scanner);
+
+#ifdef PARSER_DEBUG
+	parse_events_debug = 1;
+#endif
+	ret = parse_events_parse(data, scanner);
+
+	parse_events__flush_buffer(buffer, scanner);
+	parse_events__delete_buffer(buffer, scanner);
+	parse_events_lex_destroy(scanner);
+	return ret;
+}
+
+/*
+ * parse event config string, return a list of event terms.
+ */
+int parse_events_terms(struct list_head *terms, const char *str)
+{
+	struct parse_events_terms data = {
+		.terms = NULL,
+	};
+	int ret;
+
+	ret = parse_events__scanner(str, &data, PE_START_TERMS);
+	if (!ret) {
+		list_splice(data.terms, terms);
+		zfree(&data.terms);
+		return 0;
+	}
+
+	if (data.terms)
+		parse_events__free_terms(data.terms);
+	return ret;
+}
+
+int parse_events(struct perf_evlist *evlist, const char *str)
+{
+	struct parse_events_evlist data = {
+		.list = LIST_HEAD_INIT(data.list),
+		.idx  = evlist->nr_entries,
+	};
+	int ret;
+
+	ret = parse_events__scanner(str, &data, PE_START_EVENTS);
+	perf_pmu__parse_cleanup();
+	if (!ret) {
+		int entries = data.idx - evlist->nr_entries;
+		perf_evlist__splice_list_tail(evlist, &data.list, entries);
+		evlist->nr_groups += data.nr_groups;
+		return 0;
+	}
+
+	/*
+	 * There are 2 users - builtin-record and builtin-test objects.
+	 * Both call perf_evlist__delete in case of error, so we dont
+	 * need to bother.
+	 */
+	return ret;
+}
+
+int parse_events_option(const struct option *opt, const char *str,
+			int unset __maybe_unused)
+{
+	struct perf_evlist *evlist = *(struct perf_evlist **)opt->value;
+	int ret = parse_events(evlist, str);
+
+	if (ret) {
+		fprintf(stderr, "invalid or unsupported event: '%s'\n", str);
+		fprintf(stderr, "Run 'perf list' for a list of valid events\n");
+	}
+	return ret;
+}
+
+int parse_filter(const struct option *opt, const char *str,
+		 int unset __maybe_unused)
+{
+	struct perf_evlist *evlist = *(struct perf_evlist **)opt->value;
+	struct perf_evsel *last = NULL;
+
+	if (evlist->nr_entries > 0)
+		last = perf_evlist__last(evlist);
+
+	if (last == NULL || last->attr.type != PERF_TYPE_TRACEPOINT) {
+		fprintf(stderr,
+			"--filter option should follow a -e tracepoint option\n");
+		return -1;
+	}
+
+	last->filter = strdup(str);
+	if (last->filter == NULL) {
+		fprintf(stderr, "not enough memory to hold filter string\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+static const char * const event_type_descriptors[] = {
+	"Hardware event",
+	"Software event",
+	"Tracepoint event",
+	"Hardware cache event",
+	"Raw hardware event descriptor",
+	"Hardware breakpoint",
+};
+
+static int cmp_string(const void *a, const void *b)
+{
+	const char * const *as = a;
+	const char * const *bs = b;
+
+	return strcmp(*as, *bs);
+}
+
+/*
+ * Print the events from <debugfs_mount_point>/tracing/events
+ */
+
+void print_tracepoint_events(const char *subsys_glob, const char *event_glob,
+			     bool name_only)
+{
+	DIR *sys_dir, *evt_dir;
+	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
+	char evt_path[MAXPATHLEN];
+	char dir_path[MAXPATHLEN];
+	char **evt_list = NULL;
+	unsigned int evt_i = 0, evt_num = 0;
+	bool evt_num_known = false;
+
+restart:
+	sys_dir = opendir(tracing_events_path);
+	if (!sys_dir)
+		return;
+
+	if (evt_num_known) {
+		evt_list = zalloc(sizeof(char *) * evt_num);
+		if (!evt_list)
+			goto out_close_sys_dir;
+	}
+
+	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+		if (subsys_glob != NULL &&
+		    !strglobmatch(sys_dirent.d_name, subsys_glob))
+			continue;
+
+		snprintf(dir_path, MAXPATHLEN, "%s/%s", tracing_events_path,
+			 sys_dirent.d_name);
+		evt_dir = opendir(dir_path);
+		if (!evt_dir)
+			continue;
+
+		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+			if (event_glob != NULL &&
+			    !strglobmatch(evt_dirent.d_name, event_glob))
+				continue;
+
+			if (!evt_num_known) {
+				evt_num++;
+				continue;
+			}
+
+			snprintf(evt_path, MAXPATHLEN, "%s:%s",
+				 sys_dirent.d_name, evt_dirent.d_name);
+
+			evt_list[evt_i] = strdup(evt_path);
+			if (evt_list[evt_i] == NULL)
+				goto out_close_evt_dir;
+			evt_i++;
+		}
+		closedir(evt_dir);
+	}
+	closedir(sys_dir);
+
+	if (!evt_num_known) {
+		evt_num_known = true;
+		goto restart;
+	}
+	qsort(evt_list, evt_num, sizeof(char *), cmp_string);
+	evt_i = 0;
+	while (evt_i < evt_num) {
+		if (name_only) {
+			printf("%s ", evt_list[evt_i++]);
+			continue;
+		}
+		printf("  %-50s [%s]\n", evt_list[evt_i++],
+				event_type_descriptors[PERF_TYPE_TRACEPOINT]);
+	}
+	if (evt_num)
+		printf("\n");
+
+out_free:
+	evt_num = evt_i;
+	for (evt_i = 0; evt_i < evt_num; evt_i++)
+		zfree(&evt_list[evt_i]);
+	zfree(&evt_list);
+	return;
+
+out_close_evt_dir:
+	closedir(evt_dir);
+out_close_sys_dir:
+	closedir(sys_dir);
+
+	printf("FATAL: not enough memory to print %s\n",
+			event_type_descriptors[PERF_TYPE_TRACEPOINT]);
+	if (evt_list)
+		goto out_free;
+}
+
+/*
+ * Check whether event is in <debugfs_mount_point>/tracing/events
+ */
+
+int is_valid_tracepoint(const char *event_string)
+{
+	DIR *sys_dir, *evt_dir;
+	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
+	char evt_path[MAXPATHLEN];
+	char dir_path[MAXPATHLEN];
+
+	sys_dir = opendir(tracing_events_path);
+	if (!sys_dir)
+		return 0;
+
+	for_each_subsystem(sys_dir, sys_dirent, sys_next) {
+
+		snprintf(dir_path, MAXPATHLEN, "%s/%s", tracing_events_path,
+			 sys_dirent.d_name);
+		evt_dir = opendir(dir_path);
+		if (!evt_dir)
+			continue;
+
+		for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) {
+			snprintf(evt_path, MAXPATHLEN, "%s:%s",
+				 sys_dirent.d_name, evt_dirent.d_name);
+			if (!strcmp(evt_path, event_string)) {
+				closedir(evt_dir);
+				closedir(sys_dir);
+				return 1;
+			}
+		}
+		closedir(evt_dir);
+	}
+	closedir(sys_dir);
+	return 0;
+}
+
+static bool is_event_supported(u8 type, unsigned config)
+{
+	bool ret = true;
+	int open_return;
+	struct perf_evsel *evsel;
+	struct perf_event_attr attr = {
+		.type = type,
+		.config = config,
+		.disabled = 1,
+	};
+	struct {
+		struct thread_map map;
+		int threads[1];
+	} tmap = {
+		.map.nr	 = 1,
+		.threads = { 0 },
+	};
+
+	evsel = perf_evsel__new(&attr);
+	if (evsel) {
+		open_return = perf_evsel__open(evsel, NULL, &tmap.map);
+		ret = open_return >= 0;
+
+		if (open_return == -EACCES) {
+			/*
+			 * This happens if the paranoid value
+			 * /proc/sys/kernel/perf_event_paranoid is set to 2
+			 * Re-run with exclude_kernel set; we don't do that
+			 * by default as some ARM machines do not support it.
+			 *
+			 */
+			evsel->attr.exclude_kernel = 1;
+			ret = perf_evsel__open(evsel, NULL, &tmap.map) >= 0;
+		}
+		perf_evsel__delete(evsel);
+	}
+
+	return ret;
+}
+
+int print_hwcache_events(const char *event_glob, bool name_only)
+{
+	unsigned int type, op, i, evt_i = 0, evt_num = 0;
+	char name[64];
+	char **evt_list = NULL;
+	bool evt_num_known = false;
+
+restart:
+	if (evt_num_known) {
+		evt_list = zalloc(sizeof(char *) * evt_num);
+		if (!evt_list)
+			goto out_enomem;
+	}
+
+	for (type = 0; type < PERF_COUNT_HW_CACHE_MAX; type++) {
+		for (op = 0; op < PERF_COUNT_HW_CACHE_OP_MAX; op++) {
+			/* skip invalid cache type */
+			if (!perf_evsel__is_cache_op_valid(type, op))
+				continue;
+
+			for (i = 0; i < PERF_COUNT_HW_CACHE_RESULT_MAX; i++) {
+				__perf_evsel__hw_cache_type_op_res_name(type, op, i,
+									name, sizeof(name));
+				if (event_glob != NULL && !strglobmatch(name, event_glob))
+					continue;
+
+				if (!is_event_supported(PERF_TYPE_HW_CACHE,
+							type | (op << 8) | (i << 16)))
+					continue;
+
+				if (!evt_num_known) {
+					evt_num++;
+					continue;
+				}
+
+				evt_list[evt_i] = strdup(name);
+				if (evt_list[evt_i] == NULL)
+					goto out_enomem;
+				evt_i++;
+			}
+		}
+	}
+
+	if (!evt_num_known) {
+		evt_num_known = true;
+		goto restart;
+	}
+	qsort(evt_list, evt_num, sizeof(char *), cmp_string);
+	evt_i = 0;
+	while (evt_i < evt_num) {
+		if (name_only) {
+			printf("%s ", evt_list[evt_i++]);
+			continue;
+		}
+		printf("  %-50s [%s]\n", evt_list[evt_i++],
+				event_type_descriptors[PERF_TYPE_HW_CACHE]);
+	}
+	if (evt_num)
+		printf("\n");
+
+out_free:
+	evt_num = evt_i;
+	for (evt_i = 0; evt_i < evt_num; evt_i++)
+		zfree(&evt_list[evt_i]);
+	zfree(&evt_list);
+	return evt_num;
+
+out_enomem:
+	printf("FATAL: not enough memory to print %s\n", event_type_descriptors[PERF_TYPE_HW_CACHE]);
+	if (evt_list)
+		goto out_free;
+	return evt_num;
+}
+
+void print_symbol_events(const char *event_glob, unsigned type,
+				struct event_symbol *syms, unsigned max,
+				bool name_only)
+{
+	unsigned int i, evt_i = 0, evt_num = 0;
+	char name[MAX_NAME_LEN];
+	char **evt_list = NULL;
+	bool evt_num_known = false;
+
+restart:
+	if (evt_num_known) {
+		evt_list = zalloc(sizeof(char *) * evt_num);
+		if (!evt_list)
+			goto out_enomem;
+		syms -= max;
+	}
+
+	for (i = 0; i < max; i++, syms++) {
+
+		if (event_glob != NULL &&
+		    !(strglobmatch(syms->symbol, event_glob) ||
+		      (syms->alias && strglobmatch(syms->alias, event_glob))))
+			continue;
+
+		if (!is_event_supported(type, i))
+			continue;
+
+		if (!evt_num_known) {
+			evt_num++;
+			continue;
+		}
+
+		if (!name_only && strlen(syms->alias))
+			snprintf(name, MAX_NAME_LEN, "%s OR %s", syms->symbol, syms->alias);
+		else
+			strncpy(name, syms->symbol, MAX_NAME_LEN);
+
+		evt_list[evt_i] = strdup(name);
+		if (evt_list[evt_i] == NULL)
+			goto out_enomem;
+		evt_i++;
+	}
+
+	if (!evt_num_known) {
+		evt_num_known = true;
+		goto restart;
+	}
+	qsort(evt_list, evt_num, sizeof(char *), cmp_string);
+	evt_i = 0;
+	while (evt_i < evt_num) {
+		if (name_only) {
+			printf("%s ", evt_list[evt_i++]);
+			continue;
+		}
+		printf("  %-50s [%s]\n", evt_list[evt_i++], event_type_descriptors[type]);
+	}
+	if (evt_num)
+		printf("\n");
+
+out_free:
+	evt_num = evt_i;
+	for (evt_i = 0; evt_i < evt_num; evt_i++)
+		zfree(&evt_list[evt_i]);
+	zfree(&evt_list);
+	return;
+
+out_enomem:
+	printf("FATAL: not enough memory to print %s\n", event_type_descriptors[type]);
+	if (evt_list)
+		goto out_free;
+}
+
+/*
+ * Print the help text for the event symbols:
+ */
+void print_events(const char *event_glob, bool name_only)
+{
+	print_symbol_events(event_glob, PERF_TYPE_HARDWARE,
+			    event_symbols_hw, PERF_COUNT_HW_MAX, name_only);
+
+	print_symbol_events(event_glob, PERF_TYPE_SOFTWARE,
+			    event_symbols_sw, PERF_COUNT_SW_MAX, name_only);
+
+	print_hwcache_events(event_glob, name_only);
+
+	print_pmu_events(event_glob, name_only);
+
+	if (event_glob != NULL)
+		return;
+
+	if (!name_only) {
+		printf("  %-50s [%s]\n",
+		       "rNNN",
+		       event_type_descriptors[PERF_TYPE_RAW]);
+		printf("  %-50s [%s]\n",
+		       "cpu/t1=v1[,t2=v2,t3 ...]/modifier",
+		       event_type_descriptors[PERF_TYPE_RAW]);
+		printf("   (see 'man perf-list' on how to encode it)\n");
+		printf("\n");
+
+		printf("  %-50s [%s]\n",
+		       "mem:<addr>[/len][:access]",
+			event_type_descriptors[PERF_TYPE_BREAKPOINT]);
+		printf("\n");
+	}
+
+	print_tracepoint_events(NULL, NULL, name_only);
+}
+
+int parse_events__is_hardcoded_term(struct parse_events_term *term)
+{
+	return term->type_term != PARSE_EVENTS__TERM_TYPE_USER;
+}
+
+static int new_term(struct parse_events_term **_term, int type_val,
+		    int type_term, char *config,
+		    char *str, u64 num)
+{
+	struct parse_events_term *term;
+
+	term = zalloc(sizeof(*term));
+	if (!term)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&term->list);
+	term->type_val  = type_val;
+	term->type_term = type_term;
+	term->config = config;
+
+	switch (type_val) {
+	case PARSE_EVENTS__TERM_TYPE_NUM:
+		term->val.num = num;
+		break;
+	case PARSE_EVENTS__TERM_TYPE_STR:
+		term->val.str = str;
+		break;
+	default:
+		free(term);
+		return -EINVAL;
+	}
+
+	*_term = term;
+	return 0;
+}
+
+int parse_events_term__num(struct parse_events_term **term,
+			   int type_term, char *config, u64 num)
+{
+	return new_term(term, PARSE_EVENTS__TERM_TYPE_NUM, type_term,
+			config, NULL, num);
+}
+
+int parse_events_term__str(struct parse_events_term **term,
+			   int type_term, char *config, char *str)
+{
+	return new_term(term, PARSE_EVENTS__TERM_TYPE_STR, type_term,
+			config, str, 0);
+}
+
+int parse_events_term__sym_hw(struct parse_events_term **term,
+			      char *config, unsigned idx)
+{
+	struct event_symbol *sym;
+
+	BUG_ON(idx >= PERF_COUNT_HW_MAX);
+	sym = &event_symbols_hw[idx];
+
+	if (config)
+		return new_term(term, PARSE_EVENTS__TERM_TYPE_STR,
+				PARSE_EVENTS__TERM_TYPE_USER, config,
+				(char *) sym->symbol, 0);
+	else
+		return new_term(term, PARSE_EVENTS__TERM_TYPE_STR,
+				PARSE_EVENTS__TERM_TYPE_USER,
+				(char *) "event", (char *) sym->symbol, 0);
+}
+
+int parse_events_term__clone(struct parse_events_term **new,
+			     struct parse_events_term *term)
+{
+	return new_term(new, term->type_val, term->type_term, term->config,
+			term->val.str, term->val.num);
+}
+
+void parse_events__free_terms(struct list_head *terms)
+{
+	struct parse_events_term *term, *h;
+
+	list_for_each_entry_safe(term, h, terms, list)
+		free(term);
+}
diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h
new file mode 100644
index 000000000..52a2dda4f
--- /dev/null
+++ b/tools/perf/util/parse-events.h
@@ -0,0 +1,136 @@
+#ifndef __PERF_PARSE_EVENTS_H
+#define __PERF_PARSE_EVENTS_H
+/*
+ * Parse symbolic events/counts passed in as options:
+ */
+
+#include <linux/list.h>
+#include <stdbool.h>
+#include <linux/types.h>
+#include <linux/perf_event.h>
+
+struct list_head;
+struct perf_evsel;
+struct perf_evlist;
+
+struct option;
+
+struct tracepoint_path {
+	char *system;
+	char *name;
+	struct tracepoint_path *next;
+};
+
+extern struct tracepoint_path *tracepoint_id_to_path(u64 config);
+extern struct tracepoint_path *tracepoint_name_to_path(const char *name);
+extern bool have_tracepoints(struct list_head *evlist);
+
+const char *event_type(int type);
+
+extern int parse_events_option(const struct option *opt, const char *str,
+			       int unset);
+extern int parse_events(struct perf_evlist *evlist, const char *str);
+extern int parse_events_terms(struct list_head *terms, const char *str);
+extern int parse_filter(const struct option *opt, const char *str, int unset);
+
+#define EVENTS_HELP_MAX (128*1024)
+
+enum perf_pmu_event_symbol_type {
+	PMU_EVENT_SYMBOL_ERR,		/* not a PMU EVENT */
+	PMU_EVENT_SYMBOL,		/* normal style PMU event */
+	PMU_EVENT_SYMBOL_PREFIX,	/* prefix of pre-suf style event */
+	PMU_EVENT_SYMBOL_SUFFIX,	/* suffix of pre-suf style event */
+};
+
+struct perf_pmu_event_symbol {
+	char	*symbol;
+	enum perf_pmu_event_symbol_type	type;
+};
+
+enum {
+	PARSE_EVENTS__TERM_TYPE_NUM,
+	PARSE_EVENTS__TERM_TYPE_STR,
+};
+
+enum {
+	PARSE_EVENTS__TERM_TYPE_USER,
+	PARSE_EVENTS__TERM_TYPE_CONFIG,
+	PARSE_EVENTS__TERM_TYPE_CONFIG1,
+	PARSE_EVENTS__TERM_TYPE_CONFIG2,
+	PARSE_EVENTS__TERM_TYPE_NAME,
+	PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD,
+	PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE,
+};
+
+struct parse_events_term {
+	char *config;
+	union {
+		char *str;
+		u64  num;
+	} val;
+	int type_val;
+	int type_term;
+	struct list_head list;
+	bool used;
+};
+
+struct parse_events_evlist {
+	struct list_head list;
+	int idx;
+	int nr_groups;
+};
+
+struct parse_events_terms {
+	struct list_head *terms;
+};
+
+int parse_events__is_hardcoded_term(struct parse_events_term *term);
+int parse_events_term__num(struct parse_events_term **_term,
+			   int type_term, char *config, u64 num);
+int parse_events_term__str(struct parse_events_term **_term,
+			   int type_term, char *config, char *str);
+int parse_events_term__sym_hw(struct parse_events_term **term,
+			      char *config, unsigned idx);
+int parse_events_term__clone(struct parse_events_term **new,
+			     struct parse_events_term *term);
+void parse_events__free_terms(struct list_head *terms);
+int parse_events__modifier_event(struct list_head *list, char *str, bool add);
+int parse_events__modifier_group(struct list_head *list, char *event_mod);
+int parse_events_name(struct list_head *list, char *name);
+int parse_events_add_tracepoint(struct list_head *list, int *idx,
+				char *sys, char *event);
+int parse_events_add_numeric(struct list_head *list, int *idx,
+			     u32 type, u64 config,
+			     struct list_head *head_config);
+int parse_events_add_cache(struct list_head *list, int *idx,
+			   char *type, char *op_result1, char *op_result2);
+int parse_events_add_breakpoint(struct list_head *list, int *idx,
+				void *ptr, char *type, u64 len);
+int parse_events_add_pmu(struct list_head *list, int *idx,
+			 char *pmu , struct list_head *head_config);
+enum perf_pmu_event_symbol_type
+perf_pmu__parse_check(const char *name);
+void parse_events__set_leader(char *name, struct list_head *list);
+void parse_events_update_lists(struct list_head *list_event,
+			       struct list_head *list_all);
+void parse_events_error(void *data, void *scanner, char const *msg);
+
+void print_events(const char *event_glob, bool name_only);
+
+struct event_symbol {
+	const char	*symbol;
+	const char	*alias;
+};
+extern struct event_symbol event_symbols_hw[];
+extern struct event_symbol event_symbols_sw[];
+void print_symbol_events(const char *event_glob, unsigned type,
+				struct event_symbol *syms, unsigned max,
+				bool name_only);
+void print_tracepoint_events(const char *subsys_glob, const char *event_glob,
+			     bool name_only);
+int print_hwcache_events(const char *event_glob, bool name_only);
+extern int is_valid_tracepoint(const char *event_string);
+
+int valid_event_mount(const char *eventfs);
+
+#endif /* __PERF_PARSE_EVENTS_H */
diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l
new file mode 100644
index 000000000..8895cf313
--- /dev/null
+++ b/tools/perf/util/parse-events.l
@@ -0,0 +1,247 @@
+
+%option reentrant
+%option bison-bridge
+%option prefix="parse_events_"
+%option stack
+
+%{
+#include <errno.h>
+#include "../perf.h"
+#include "parse-events-bison.h"
+#include "parse-events.h"
+
+char *parse_events_get_text(yyscan_t yyscanner);
+YYSTYPE *parse_events_get_lval(yyscan_t yyscanner);
+
+static int __value(YYSTYPE *yylval, char *str, int base, int token)
+{
+	u64 num;
+
+	errno = 0;
+	num = strtoull(str, NULL, base);
+	if (errno)
+		return PE_ERROR;
+
+	yylval->num = num;
+	return token;
+}
+
+static int value(yyscan_t scanner, int base)
+{
+	YYSTYPE *yylval = parse_events_get_lval(scanner);
+	char *text = parse_events_get_text(scanner);
+
+	return __value(yylval, text, base, PE_VALUE);
+}
+
+static int raw(yyscan_t scanner)
+{
+	YYSTYPE *yylval = parse_events_get_lval(scanner);
+	char *text = parse_events_get_text(scanner);
+
+	return __value(yylval, text + 1, 16, PE_RAW);
+}
+
+static int str(yyscan_t scanner, int token)
+{
+	YYSTYPE *yylval = parse_events_get_lval(scanner);
+	char *text = parse_events_get_text(scanner);
+
+	yylval->str = strdup(text);
+	return token;
+}
+
+static int pmu_str_check(yyscan_t scanner)
+{
+	YYSTYPE *yylval = parse_events_get_lval(scanner);
+	char *text = parse_events_get_text(scanner);
+
+	yylval->str = strdup(text);
+	switch (perf_pmu__parse_check(text)) {
+		case PMU_EVENT_SYMBOL_PREFIX:
+			return PE_PMU_EVENT_PRE;
+		case PMU_EVENT_SYMBOL_SUFFIX:
+			return PE_PMU_EVENT_SUF;
+		case PMU_EVENT_SYMBOL:
+			return PE_KERNEL_PMU_EVENT;
+		default:
+			return PE_NAME;
+	}
+}
+
+static int sym(yyscan_t scanner, int type, int config)
+{
+	YYSTYPE *yylval = parse_events_get_lval(scanner);
+
+	yylval->num = (type << 16) + config;
+	return type == PERF_TYPE_HARDWARE ? PE_VALUE_SYM_HW : PE_VALUE_SYM_SW;
+}
+
+static int term(yyscan_t scanner, int type)
+{
+	YYSTYPE *yylval = parse_events_get_lval(scanner);
+
+	yylval->num = type;
+	return PE_TERM;
+}
+
+%}
+
+%x mem
+%s config
+%x event
+
+group		[^,{}/]*[{][^}]*[}][^,{}/]*
+event_pmu	[^,{}/]+[/][^/]*[/][^,{}/]*
+event		[^,{}/]+
+
+num_dec		[0-9]+
+num_hex		0x[a-fA-F0-9]+
+num_raw_hex	[a-fA-F0-9]+
+name		[a-zA-Z_*?][a-zA-Z0-9_*?]*
+name_minus	[a-zA-Z_*?][a-zA-Z0-9\-_*?]*
+/* If you add a modifier you need to update check_modifier() */
+modifier_event	[ukhpGHSDI]+
+modifier_bp	[rwx]{1,3}
+
+%%
+
+%{
+	{
+		int start_token;
+
+		start_token = parse_events_get_extra(yyscanner);
+
+		if (start_token == PE_START_TERMS)
+			BEGIN(config);
+		else if (start_token == PE_START_EVENTS)
+			BEGIN(event);
+
+		if (start_token) {
+			parse_events_set_extra(NULL, yyscanner);
+			return start_token;
+		}
+         }
+%}
+
+<event>{
+
+{group}		{
+			BEGIN(INITIAL); yyless(0);
+		}
+
+{event_pmu}	|
+{event}		{
+			str(yyscanner, PE_EVENT_NAME);
+			BEGIN(INITIAL); yyless(0);
+			return PE_EVENT_NAME;
+		}
+
+.		|
+<<EOF>>		{
+			BEGIN(INITIAL); yyless(0);
+		}
+
+}
+
+<config>{
+config			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_CONFIG); }
+config1			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_CONFIG1); }
+config2			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_CONFIG2); }
+name			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_NAME); }
+period			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD); }
+branch_type		{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE); }
+,			{ return ','; }
+"/"			{ BEGIN(INITIAL); return '/'; }
+{name_minus}		{ return str(yyscanner, PE_NAME); }
+}
+
+<mem>{
+{modifier_bp}		{ return str(yyscanner, PE_MODIFIER_BP); }
+:			{ return ':'; }
+"/"			{ return '/'; }
+{num_dec}		{ return value(yyscanner, 10); }
+{num_hex}		{ return value(yyscanner, 16); }
+	/*
+	 * We need to separate 'mem:' scanner part, in order to get specific
+	 * modifier bits parsed out. Otherwise we would need to handle PE_NAME
+	 * and we'd need to parse it manually. During the escape from <mem>
+	 * state we need to put the escaping char back, so we dont miss it.
+	 */
+.			{ unput(*yytext); BEGIN(INITIAL); }
+	/*
+	 * We destroy the scanner after reaching EOF,
+	 * but anyway just to be sure get back to INIT state.
+	 */
+<<EOF>>			{ BEGIN(INITIAL); }
+}
+
+cpu-cycles|cycles				{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_CPU_CYCLES); }
+stalled-cycles-frontend|idle-cycles-frontend	{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_STALLED_CYCLES_FRONTEND); }
+stalled-cycles-backend|idle-cycles-backend	{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_STALLED_CYCLES_BACKEND); }
+instructions					{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_INSTRUCTIONS); }
+cache-references				{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_CACHE_REFERENCES); }
+cache-misses					{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_CACHE_MISSES); }
+branch-instructions|branches			{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_BRANCH_INSTRUCTIONS); }
+branch-misses					{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_BRANCH_MISSES); }
+bus-cycles					{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_BUS_CYCLES); }
+ref-cycles					{ return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_REF_CPU_CYCLES); }
+cpu-clock					{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CPU_CLOCK); }
+task-clock					{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_TASK_CLOCK); }
+page-faults|faults				{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_PAGE_FAULTS); }
+minor-faults					{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_PAGE_FAULTS_MIN); }
+major-faults					{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_PAGE_FAULTS_MAJ); }
+context-switches|cs				{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CONTEXT_SWITCHES); }
+cpu-migrations|migrations			{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CPU_MIGRATIONS); }
+alignment-faults				{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_ALIGNMENT_FAULTS); }
+emulation-faults				{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_EMULATION_FAULTS); }
+dummy						{ return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_DUMMY); }
+
+	/*
+	 * We have to handle the kernel PMU event cycles-ct/cycles-t/mem-loads/mem-stores separately.
+	 * Because the prefix cycles is mixed up with cpu-cycles.
+	 * loads and stores are mixed up with cache event
+	 */
+cycles-ct					{ return str(yyscanner, PE_KERNEL_PMU_EVENT); }
+cycles-t					{ return str(yyscanner, PE_KERNEL_PMU_EVENT); }
+mem-loads					{ return str(yyscanner, PE_KERNEL_PMU_EVENT); }
+mem-stores					{ return str(yyscanner, PE_KERNEL_PMU_EVENT); }
+
+L1-dcache|l1-d|l1d|L1-data		|
+L1-icache|l1-i|l1i|L1-instruction	|
+LLC|L2					|
+dTLB|d-tlb|Data-TLB			|
+iTLB|i-tlb|Instruction-TLB		|
+branch|branches|bpu|btb|bpc		|
+node					{ return str(yyscanner, PE_NAME_CACHE_TYPE); }
+
+load|loads|read				|
+store|stores|write			|
+prefetch|prefetches			|
+speculative-read|speculative-load	|
+refs|Reference|ops|access		|
+misses|miss				{ return str(yyscanner, PE_NAME_CACHE_OP_RESULT); }
+
+mem:			{ BEGIN(mem); return PE_PREFIX_MEM; }
+r{num_raw_hex}		{ return raw(yyscanner); }
+{num_dec}		{ return value(yyscanner, 10); }
+{num_hex}		{ return value(yyscanner, 16); }
+
+{modifier_event}	{ return str(yyscanner, PE_MODIFIER_EVENT); }
+{name}			{ return pmu_str_check(yyscanner); }
+"/"			{ BEGIN(config); return '/'; }
+-			{ return '-'; }
+,			{ BEGIN(event); return ','; }
+:			{ return ':'; }
+"{"			{ BEGIN(event); return '{'; }
+"}"			{ return '}'; }
+=			{ return '='; }
+\n			{ }
+.			{ }
+
+%%
+
+int parse_events_wrap(void *scanner __maybe_unused)
+{
+	return 1;
+}
diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y
new file mode 100644
index 000000000..72def077d
--- /dev/null
+++ b/tools/perf/util/parse-events.y
@@ -0,0 +1,526 @@
+%pure-parser
+%parse-param {void *_data}
+%parse-param {void *scanner}
+%lex-param {void* scanner}
+
+%{
+
+#define YYDEBUG 1
+
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/types.h>
+#include "util.h"
+#include "parse-events.h"
+#include "parse-events-bison.h"
+
+extern int parse_events_lex (YYSTYPE* lvalp, void* scanner);
+
+#define ABORT_ON(val) \
+do { \
+	if (val) \
+		YYABORT; \
+} while (0)
+
+#define ALLOC_LIST(list) \
+do { \
+	list = malloc(sizeof(*list)); \
+	ABORT_ON(!list);              \
+	INIT_LIST_HEAD(list);         \
+} while (0)
+
+static inc_group_count(struct list_head *list,
+		       struct parse_events_evlist *data)
+{
+	/* Count groups only have more than 1 members */
+	if (!list_is_last(list->next, list))
+		data->nr_groups++;
+}
+
+%}
+
+%token PE_START_EVENTS PE_START_TERMS
+%token PE_VALUE PE_VALUE_SYM_HW PE_VALUE_SYM_SW PE_RAW PE_TERM
+%token PE_EVENT_NAME
+%token PE_NAME
+%token PE_MODIFIER_EVENT PE_MODIFIER_BP
+%token PE_NAME_CACHE_TYPE PE_NAME_CACHE_OP_RESULT
+%token PE_PREFIX_MEM PE_PREFIX_RAW PE_PREFIX_GROUP
+%token PE_ERROR
+%token PE_PMU_EVENT_PRE PE_PMU_EVENT_SUF PE_KERNEL_PMU_EVENT
+%type <num> PE_VALUE
+%type <num> PE_VALUE_SYM_HW
+%type <num> PE_VALUE_SYM_SW
+%type <num> PE_RAW
+%type <num> PE_TERM
+%type <str> PE_NAME
+%type <str> PE_NAME_CACHE_TYPE
+%type <str> PE_NAME_CACHE_OP_RESULT
+%type <str> PE_MODIFIER_EVENT
+%type <str> PE_MODIFIER_BP
+%type <str> PE_EVENT_NAME
+%type <str> PE_PMU_EVENT_PRE PE_PMU_EVENT_SUF PE_KERNEL_PMU_EVENT
+%type <num> value_sym
+%type <head> event_config
+%type <term> event_term
+%type <head> event_pmu
+%type <head> event_legacy_symbol
+%type <head> event_legacy_cache
+%type <head> event_legacy_mem
+%type <head> event_legacy_tracepoint
+%type <head> event_legacy_numeric
+%type <head> event_legacy_raw
+%type <head> event_def
+%type <head> event_mod
+%type <head> event_name
+%type <head> event
+%type <head> events
+%type <head> group_def
+%type <head> group
+%type <head> groups
+
+%union
+{
+	char *str;
+	u64 num;
+	struct list_head *head;
+	struct parse_events_term *term;
+}
+%%
+
+start:
+PE_START_EVENTS start_events
+|
+PE_START_TERMS  start_terms
+
+start_events: groups
+{
+	struct parse_events_evlist *data = _data;
+
+	parse_events_update_lists($1, &data->list);
+}
+
+groups:
+groups ',' group
+{
+	struct list_head *list  = $1;
+	struct list_head *group = $3;
+
+	parse_events_update_lists(group, list);
+	$$ = list;
+}
+|
+groups ',' event
+{
+	struct list_head *list  = $1;
+	struct list_head *event = $3;
+
+	parse_events_update_lists(event, list);
+	$$ = list;
+}
+|
+group
+|
+event
+
+group:
+group_def ':' PE_MODIFIER_EVENT
+{
+	struct list_head *list = $1;
+
+	ABORT_ON(parse_events__modifier_group(list, $3));
+	$$ = list;
+}
+|
+group_def
+
+group_def:
+PE_NAME '{' events '}'
+{
+	struct list_head *list = $3;
+
+	inc_group_count(list, _data);
+	parse_events__set_leader($1, list);
+	$$ = list;
+}
+|
+'{' events '}'
+{
+	struct list_head *list = $2;
+
+	inc_group_count(list, _data);
+	parse_events__set_leader(NULL, list);
+	$$ = list;
+}
+
+events:
+events ',' event
+{
+	struct list_head *event = $3;
+	struct list_head *list  = $1;
+
+	parse_events_update_lists(event, list);
+	$$ = list;
+}
+|
+event
+
+event: event_mod
+
+event_mod:
+event_name PE_MODIFIER_EVENT
+{
+	struct list_head *list = $1;
+
+	/*
+	 * Apply modifier on all events added by single event definition
+	 * (there could be more events added for multiple tracepoint
+	 * definitions via '*?'.
+	 */
+	ABORT_ON(parse_events__modifier_event(list, $2, false));
+	$$ = list;
+}
+|
+event_name
+
+event_name:
+PE_EVENT_NAME event_def
+{
+	ABORT_ON(parse_events_name($2, $1));
+	free($1);
+	$$ = $2;
+}
+|
+event_def
+
+event_def: event_pmu |
+	   event_legacy_symbol |
+	   event_legacy_cache sep_dc |
+	   event_legacy_mem |
+	   event_legacy_tracepoint sep_dc |
+	   event_legacy_numeric sep_dc |
+	   event_legacy_raw sep_dc
+
+event_pmu:
+PE_NAME '/' event_config '/'
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_pmu(list, &data->idx, $1, $3));
+	parse_events__free_terms($3);
+	$$ = list;
+}
+|
+PE_NAME '/' '/'
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_pmu(list, &data->idx, $1, NULL));
+	$$ = list;
+}
+|
+PE_KERNEL_PMU_EVENT sep_dc
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *head;
+	struct parse_events_term *term;
+	struct list_head *list;
+
+	ALLOC_LIST(head);
+	ABORT_ON(parse_events_term__num(&term, PARSE_EVENTS__TERM_TYPE_USER,
+					$1, 1));
+	list_add_tail(&term->list, head);
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_pmu(list, &data->idx, "cpu", head));
+	parse_events__free_terms(head);
+	$$ = list;
+}
+|
+PE_PMU_EVENT_PRE '-' PE_PMU_EVENT_SUF sep_dc
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *head;
+	struct parse_events_term *term;
+	struct list_head *list;
+	char pmu_name[128];
+	snprintf(&pmu_name, 128, "%s-%s", $1, $3);
+
+	ALLOC_LIST(head);
+	ABORT_ON(parse_events_term__num(&term, PARSE_EVENTS__TERM_TYPE_USER,
+					&pmu_name, 1));
+	list_add_tail(&term->list, head);
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_pmu(list, &data->idx, "cpu", head));
+	parse_events__free_terms(head);
+	$$ = list;
+}
+
+value_sym:
+PE_VALUE_SYM_HW
+|
+PE_VALUE_SYM_SW
+
+event_legacy_symbol:
+value_sym '/' event_config '/'
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+	int type = $1 >> 16;
+	int config = $1 & 255;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_numeric(list, &data->idx,
+					  type, config, $3));
+	parse_events__free_terms($3);
+	$$ = list;
+}
+|
+value_sym sep_slash_dc
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+	int type = $1 >> 16;
+	int config = $1 & 255;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_numeric(list, &data->idx,
+					  type, config, NULL));
+	$$ = list;
+}
+
+event_legacy_cache:
+PE_NAME_CACHE_TYPE '-' PE_NAME_CACHE_OP_RESULT '-' PE_NAME_CACHE_OP_RESULT
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_cache(list, &data->idx, $1, $3, $5));
+	$$ = list;
+}
+|
+PE_NAME_CACHE_TYPE '-' PE_NAME_CACHE_OP_RESULT
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_cache(list, &data->idx, $1, $3, NULL));
+	$$ = list;
+}
+|
+PE_NAME_CACHE_TYPE
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_cache(list, &data->idx, $1, NULL, NULL));
+	$$ = list;
+}
+
+event_legacy_mem:
+PE_PREFIX_MEM PE_VALUE '/' PE_VALUE ':' PE_MODIFIER_BP sep_dc
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_breakpoint(list, &data->idx,
+					     (void *) $2, $6, $4));
+	$$ = list;
+}
+|
+PE_PREFIX_MEM PE_VALUE '/' PE_VALUE sep_dc
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_breakpoint(list, &data->idx,
+					     (void *) $2, NULL, $4));
+	$$ = list;
+}
+|
+PE_PREFIX_MEM PE_VALUE ':' PE_MODIFIER_BP sep_dc
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_breakpoint(list, &data->idx,
+					     (void *) $2, $4, 0));
+	$$ = list;
+}
+|
+PE_PREFIX_MEM PE_VALUE sep_dc
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_breakpoint(list, &data->idx,
+					     (void *) $2, NULL, 0));
+	$$ = list;
+}
+
+event_legacy_tracepoint:
+PE_NAME '-' PE_NAME ':' PE_NAME
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+	char sys_name[128];
+	snprintf(&sys_name, 128, "%s-%s", $1, $3);
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_tracepoint(list, &data->idx, &sys_name, $5));
+	$$ = list;
+}
+|
+PE_NAME ':' PE_NAME
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_tracepoint(list, &data->idx, $1, $3));
+	$$ = list;
+}
+
+event_legacy_numeric:
+PE_VALUE ':' PE_VALUE
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_numeric(list, &data->idx, (u32)$1, $3, NULL));
+	$$ = list;
+}
+
+event_legacy_raw:
+PE_RAW
+{
+	struct parse_events_evlist *data = _data;
+	struct list_head *list;
+
+	ALLOC_LIST(list);
+	ABORT_ON(parse_events_add_numeric(list, &data->idx,
+					  PERF_TYPE_RAW, $1, NULL));
+	$$ = list;
+}
+
+start_terms: event_config
+{
+	struct parse_events_terms *data = _data;
+	data->terms = $1;
+}
+
+event_config:
+event_config ',' event_term
+{
+	struct list_head *head = $1;
+	struct parse_events_term *term = $3;
+
+	ABORT_ON(!head);
+	list_add_tail(&term->list, head);
+	$$ = $1;
+}
+|
+event_term
+{
+	struct list_head *head = malloc(sizeof(*head));
+	struct parse_events_term *term = $1;
+
+	ABORT_ON(!head);
+	INIT_LIST_HEAD(head);
+	list_add_tail(&term->list, head);
+	$$ = head;
+}
+
+event_term:
+PE_NAME '=' PE_NAME
+{
+	struct parse_events_term *term;
+
+	ABORT_ON(parse_events_term__str(&term, PARSE_EVENTS__TERM_TYPE_USER,
+					$1, $3));
+	$$ = term;
+}
+|
+PE_NAME '=' PE_VALUE
+{
+	struct parse_events_term *term;
+
+	ABORT_ON(parse_events_term__num(&term, PARSE_EVENTS__TERM_TYPE_USER,
+					$1, $3));
+	$$ = term;
+}
+|
+PE_NAME '=' PE_VALUE_SYM_HW
+{
+	struct parse_events_term *term;
+	int config = $3 & 255;
+
+	ABORT_ON(parse_events_term__sym_hw(&term, $1, config));
+	$$ = term;
+}
+|
+PE_NAME
+{
+	struct parse_events_term *term;
+
+	ABORT_ON(parse_events_term__num(&term, PARSE_EVENTS__TERM_TYPE_USER,
+					$1, 1));
+	$$ = term;
+}
+|
+PE_VALUE_SYM_HW
+{
+	struct parse_events_term *term;
+	int config = $1 & 255;
+
+	ABORT_ON(parse_events_term__sym_hw(&term, NULL, config));
+	$$ = term;
+}
+|
+PE_TERM '=' PE_NAME
+{
+	struct parse_events_term *term;
+
+	ABORT_ON(parse_events_term__str(&term, (int)$1, NULL, $3));
+	$$ = term;
+}
+|
+PE_TERM '=' PE_VALUE
+{
+	struct parse_events_term *term;
+
+	ABORT_ON(parse_events_term__num(&term, (int)$1, NULL, $3));
+	$$ = term;
+}
+|
+PE_TERM
+{
+	struct parse_events_term *term;
+
+	ABORT_ON(parse_events_term__num(&term, (int)$1, NULL, 1));
+	$$ = term;
+}
+
+sep_dc: ':' |
+
+sep_slash_dc: '/' | ':' |
+
+%%
+
+void parse_events_error(void *data __maybe_unused, void *scanner __maybe_unused,
+			char const *msg __maybe_unused)
+{
+}
diff --git a/tools/perf/util/parse-options.c b/tools/perf/util/parse-options.c
new file mode 100644
index 000000000..01626be2a
--- /dev/null
+++ b/tools/perf/util/parse-options.c
@@ -0,0 +1,757 @@
+#include "util.h"
+#include "parse-options.h"
+#include "cache.h"
+#include "header.h"
+
+#define OPT_SHORT 1
+#define OPT_UNSET 2
+
+static int opterror(const struct option *opt, const char *reason, int flags)
+{
+	if (flags & OPT_SHORT)
+		return error("switch `%c' %s", opt->short_name, reason);
+	if (flags & OPT_UNSET)
+		return error("option `no-%s' %s", opt->long_name, reason);
+	return error("option `%s' %s", opt->long_name, reason);
+}
+
+static int get_arg(struct parse_opt_ctx_t *p, const struct option *opt,
+		   int flags, const char **arg)
+{
+	if (p->opt) {
+		*arg = p->opt;
+		p->opt = NULL;
+	} else if ((opt->flags & PARSE_OPT_LASTARG_DEFAULT) && (p->argc == 1 ||
+		    **(p->argv + 1) == '-')) {
+		*arg = (const char *)opt->defval;
+	} else if (p->argc > 1) {
+		p->argc--;
+		*arg = *++p->argv;
+	} else
+		return opterror(opt, "requires a value", flags);
+	return 0;
+}
+
+static int get_value(struct parse_opt_ctx_t *p,
+		     const struct option *opt, int flags)
+{
+	const char *s, *arg = NULL;
+	const int unset = flags & OPT_UNSET;
+	int err;
+
+	if (unset && p->opt)
+		return opterror(opt, "takes no value", flags);
+	if (unset && (opt->flags & PARSE_OPT_NONEG))
+		return opterror(opt, "isn't available", flags);
+	if (opt->flags & PARSE_OPT_DISABLED)
+		return opterror(opt, "is not usable", flags);
+
+	if (opt->flags & PARSE_OPT_EXCLUSIVE) {
+		if (p->excl_opt && p->excl_opt != opt) {
+			char msg[128];
+
+			if (((flags & OPT_SHORT) && p->excl_opt->short_name) ||
+			    p->excl_opt->long_name == NULL) {
+				scnprintf(msg, sizeof(msg), "cannot be used with switch `%c'",
+					  p->excl_opt->short_name);
+			} else {
+				scnprintf(msg, sizeof(msg), "cannot be used with %s",
+					  p->excl_opt->long_name);
+			}
+			opterror(opt, msg, flags);
+			return -3;
+		}
+		p->excl_opt = opt;
+	}
+	if (!(flags & OPT_SHORT) && p->opt) {
+		switch (opt->type) {
+		case OPTION_CALLBACK:
+			if (!(opt->flags & PARSE_OPT_NOARG))
+				break;
+			/* FALLTHROUGH */
+		case OPTION_BOOLEAN:
+		case OPTION_INCR:
+		case OPTION_BIT:
+		case OPTION_SET_UINT:
+		case OPTION_SET_PTR:
+			return opterror(opt, "takes no value", flags);
+		case OPTION_END:
+		case OPTION_ARGUMENT:
+		case OPTION_GROUP:
+		case OPTION_STRING:
+		case OPTION_INTEGER:
+		case OPTION_UINTEGER:
+		case OPTION_LONG:
+		case OPTION_U64:
+		default:
+			break;
+		}
+	}
+
+	switch (opt->type) {
+	case OPTION_BIT:
+		if (unset)
+			*(int *)opt->value &= ~opt->defval;
+		else
+			*(int *)opt->value |= opt->defval;
+		return 0;
+
+	case OPTION_BOOLEAN:
+		*(bool *)opt->value = unset ? false : true;
+		if (opt->set)
+			*(bool *)opt->set = true;
+		return 0;
+
+	case OPTION_INCR:
+		*(int *)opt->value = unset ? 0 : *(int *)opt->value + 1;
+		return 0;
+
+	case OPTION_SET_UINT:
+		*(unsigned int *)opt->value = unset ? 0 : opt->defval;
+		return 0;
+
+	case OPTION_SET_PTR:
+		*(void **)opt->value = unset ? NULL : (void *)opt->defval;
+		return 0;
+
+	case OPTION_STRING:
+		err = 0;
+		if (unset)
+			*(const char **)opt->value = NULL;
+		else if (opt->flags & PARSE_OPT_OPTARG && !p->opt)
+			*(const char **)opt->value = (const char *)opt->defval;
+		else
+			err = get_arg(p, opt, flags, (const char **)opt->value);
+
+		/* PARSE_OPT_NOEMPTY: Allow NULL but disallow empty string. */
+		if (opt->flags & PARSE_OPT_NOEMPTY) {
+			const char *val = *(const char **)opt->value;
+
+			if (!val)
+				return err;
+
+			/* Similar to unset if we are given an empty string. */
+			if (val[0] == '\0') {
+				*(const char **)opt->value = NULL;
+				return 0;
+			}
+		}
+
+		return err;
+
+	case OPTION_CALLBACK:
+		if (unset)
+			return (*opt->callback)(opt, NULL, 1) ? (-1) : 0;
+		if (opt->flags & PARSE_OPT_NOARG)
+			return (*opt->callback)(opt, NULL, 0) ? (-1) : 0;
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt)
+			return (*opt->callback)(opt, NULL, 0) ? (-1) : 0;
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		return (*opt->callback)(opt, arg, 0) ? (-1) : 0;
+
+	case OPTION_INTEGER:
+		if (unset) {
+			*(int *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(int *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(int *)opt->value = strtol(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	case OPTION_UINTEGER:
+		if (unset) {
+			*(unsigned int *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(unsigned int *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(unsigned int *)opt->value = strtol(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	case OPTION_LONG:
+		if (unset) {
+			*(long *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(long *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(long *)opt->value = strtol(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	case OPTION_U64:
+		if (unset) {
+			*(u64 *)opt->value = 0;
+			return 0;
+		}
+		if (opt->flags & PARSE_OPT_OPTARG && !p->opt) {
+			*(u64 *)opt->value = opt->defval;
+			return 0;
+		}
+		if (get_arg(p, opt, flags, &arg))
+			return -1;
+		*(u64 *)opt->value = strtoull(arg, (char **)&s, 10);
+		if (*s)
+			return opterror(opt, "expects a numerical value", flags);
+		return 0;
+
+	case OPTION_END:
+	case OPTION_ARGUMENT:
+	case OPTION_GROUP:
+	default:
+		die("should not happen, someone must be hit on the forehead");
+	}
+}
+
+static int parse_short_opt(struct parse_opt_ctx_t *p, const struct option *options)
+{
+	for (; options->type != OPTION_END; options++) {
+		if (options->short_name == *p->opt) {
+			p->opt = p->opt[1] ? p->opt + 1 : NULL;
+			return get_value(p, options, OPT_SHORT);
+		}
+	}
+	return -2;
+}
+
+static int parse_long_opt(struct parse_opt_ctx_t *p, const char *arg,
+                          const struct option *options)
+{
+	const char *arg_end = strchr(arg, '=');
+	const struct option *abbrev_option = NULL, *ambiguous_option = NULL;
+	int abbrev_flags = 0, ambiguous_flags = 0;
+
+	if (!arg_end)
+		arg_end = arg + strlen(arg);
+
+	for (; options->type != OPTION_END; options++) {
+		const char *rest;
+		int flags = 0;
+
+		if (!options->long_name)
+			continue;
+
+		rest = skip_prefix(arg, options->long_name);
+		if (options->type == OPTION_ARGUMENT) {
+			if (!rest)
+				continue;
+			if (*rest == '=')
+				return opterror(options, "takes no value", flags);
+			if (*rest)
+				continue;
+			p->out[p->cpidx++] = arg - 2;
+			return 0;
+		}
+		if (!rest) {
+			if (!prefixcmp(options->long_name, "no-")) {
+				/*
+				 * The long name itself starts with "no-", so
+				 * accept the option without "no-" so that users
+				 * do not have to enter "no-no-" to get the
+				 * negation.
+				 */
+				rest = skip_prefix(arg, options->long_name + 3);
+				if (rest) {
+					flags |= OPT_UNSET;
+					goto match;
+				}
+				/* Abbreviated case */
+				if (!prefixcmp(options->long_name + 3, arg)) {
+					flags |= OPT_UNSET;
+					goto is_abbreviated;
+				}
+			}
+			/* abbreviated? */
+			if (!strncmp(options->long_name, arg, arg_end - arg)) {
+is_abbreviated:
+				if (abbrev_option) {
+					/*
+					 * If this is abbreviated, it is
+					 * ambiguous. So when there is no
+					 * exact match later, we need to
+					 * error out.
+					 */
+					ambiguous_option = abbrev_option;
+					ambiguous_flags = abbrev_flags;
+				}
+				if (!(flags & OPT_UNSET) && *arg_end)
+					p->opt = arg_end + 1;
+				abbrev_option = options;
+				abbrev_flags = flags;
+				continue;
+			}
+			/* negated and abbreviated very much? */
+			if (!prefixcmp("no-", arg)) {
+				flags |= OPT_UNSET;
+				goto is_abbreviated;
+			}
+			/* negated? */
+			if (strncmp(arg, "no-", 3))
+				continue;
+			flags |= OPT_UNSET;
+			rest = skip_prefix(arg + 3, options->long_name);
+			/* abbreviated and negated? */
+			if (!rest && !prefixcmp(options->long_name, arg + 3))
+				goto is_abbreviated;
+			if (!rest)
+				continue;
+		}
+match:
+		if (*rest) {
+			if (*rest != '=')
+				continue;
+			p->opt = rest + 1;
+		}
+		return get_value(p, options, flags);
+	}
+
+	if (ambiguous_option)
+		return error("Ambiguous option: %s "
+			"(could be --%s%s or --%s%s)",
+			arg,
+			(ambiguous_flags & OPT_UNSET) ?  "no-" : "",
+			ambiguous_option->long_name,
+			(abbrev_flags & OPT_UNSET) ?  "no-" : "",
+			abbrev_option->long_name);
+	if (abbrev_option)
+		return get_value(p, abbrev_option, abbrev_flags);
+	return -2;
+}
+
+static void check_typos(const char *arg, const struct option *options)
+{
+	if (strlen(arg) < 3)
+		return;
+
+	if (!prefixcmp(arg, "no-")) {
+		error ("did you mean `--%s` (with two dashes ?)", arg);
+		exit(129);
+	}
+
+	for (; options->type != OPTION_END; options++) {
+		if (!options->long_name)
+			continue;
+		if (!prefixcmp(options->long_name, arg)) {
+			error ("did you mean `--%s` (with two dashes ?)", arg);
+			exit(129);
+		}
+	}
+}
+
+void parse_options_start(struct parse_opt_ctx_t *ctx,
+			 int argc, const char **argv, int flags)
+{
+	memset(ctx, 0, sizeof(*ctx));
+	ctx->argc = argc - 1;
+	ctx->argv = argv + 1;
+	ctx->out  = argv;
+	ctx->cpidx = ((flags & PARSE_OPT_KEEP_ARGV0) != 0);
+	ctx->flags = flags;
+	if ((flags & PARSE_OPT_KEEP_UNKNOWN) &&
+	    (flags & PARSE_OPT_STOP_AT_NON_OPTION))
+		die("STOP_AT_NON_OPTION and KEEP_UNKNOWN don't go together");
+}
+
+static int usage_with_options_internal(const char * const *,
+				       const struct option *, int);
+
+int parse_options_step(struct parse_opt_ctx_t *ctx,
+		       const struct option *options,
+		       const char * const usagestr[])
+{
+	int internal_help = !(ctx->flags & PARSE_OPT_NO_INTERNAL_HELP);
+	int excl_short_opt = 1;
+	const char *arg;
+
+	/* we must reset ->opt, unknown short option leave it dangling */
+	ctx->opt = NULL;
+
+	for (; ctx->argc; ctx->argc--, ctx->argv++) {
+		arg = ctx->argv[0];
+		if (*arg != '-' || !arg[1]) {
+			if (ctx->flags & PARSE_OPT_STOP_AT_NON_OPTION)
+				break;
+			ctx->out[ctx->cpidx++] = ctx->argv[0];
+			continue;
+		}
+
+		if (arg[1] != '-') {
+			ctx->opt = ++arg;
+			if (internal_help && *ctx->opt == 'h')
+				return usage_with_options_internal(usagestr, options, 0);
+			switch (parse_short_opt(ctx, options)) {
+			case -1:
+				return parse_options_usage(usagestr, options, arg, 1);
+			case -2:
+				goto unknown;
+			case -3:
+				goto exclusive;
+			default:
+				break;
+			}
+			if (ctx->opt)
+				check_typos(arg, options);
+			while (ctx->opt) {
+				if (internal_help && *ctx->opt == 'h')
+					return usage_with_options_internal(usagestr, options, 0);
+				arg = ctx->opt;
+				switch (parse_short_opt(ctx, options)) {
+				case -1:
+					return parse_options_usage(usagestr, options, arg, 1);
+				case -2:
+					/* fake a short option thing to hide the fact that we may have
+					 * started to parse aggregated stuff
+					 *
+					 * This is leaky, too bad.
+					 */
+					ctx->argv[0] = strdup(ctx->opt - 1);
+					*(char *)ctx->argv[0] = '-';
+					goto unknown;
+				case -3:
+					goto exclusive;
+				default:
+					break;
+				}
+			}
+			continue;
+		}
+
+		if (!arg[2]) { /* "--" */
+			if (!(ctx->flags & PARSE_OPT_KEEP_DASHDASH)) {
+				ctx->argc--;
+				ctx->argv++;
+			}
+			break;
+		}
+
+		arg += 2;
+		if (internal_help && !strcmp(arg, "help-all"))
+			return usage_with_options_internal(usagestr, options, 1);
+		if (internal_help && !strcmp(arg, "help"))
+			return usage_with_options_internal(usagestr, options, 0);
+		if (!strcmp(arg, "list-opts"))
+			return PARSE_OPT_LIST_OPTS;
+		if (!strcmp(arg, "list-cmds"))
+			return PARSE_OPT_LIST_SUBCMDS;
+		switch (parse_long_opt(ctx, arg, options)) {
+		case -1:
+			return parse_options_usage(usagestr, options, arg, 0);
+		case -2:
+			goto unknown;
+		case -3:
+			excl_short_opt = 0;
+			goto exclusive;
+		default:
+			break;
+		}
+		continue;
+unknown:
+		if (!(ctx->flags & PARSE_OPT_KEEP_UNKNOWN))
+			return PARSE_OPT_UNKNOWN;
+		ctx->out[ctx->cpidx++] = ctx->argv[0];
+		ctx->opt = NULL;
+	}
+	return PARSE_OPT_DONE;
+
+exclusive:
+	parse_options_usage(usagestr, options, arg, excl_short_opt);
+	if ((excl_short_opt && ctx->excl_opt->short_name) ||
+	    ctx->excl_opt->long_name == NULL) {
+		char opt = ctx->excl_opt->short_name;
+		parse_options_usage(NULL, options, &opt, 1);
+	} else {
+		parse_options_usage(NULL, options, ctx->excl_opt->long_name, 0);
+	}
+	return PARSE_OPT_HELP;
+}
+
+int parse_options_end(struct parse_opt_ctx_t *ctx)
+{
+	memmove(ctx->out + ctx->cpidx, ctx->argv, ctx->argc * sizeof(*ctx->out));
+	ctx->out[ctx->cpidx + ctx->argc] = NULL;
+	return ctx->cpidx + ctx->argc;
+}
+
+int parse_options_subcommand(int argc, const char **argv, const struct option *options,
+			const char *const subcommands[], const char *usagestr[], int flags)
+{
+	struct parse_opt_ctx_t ctx;
+
+	perf_header__set_cmdline(argc, argv);
+
+	/* build usage string if it's not provided */
+	if (subcommands && !usagestr[0]) {
+		struct strbuf buf = STRBUF_INIT;
+
+		strbuf_addf(&buf, "perf %s [<options>] {", argv[0]);
+		for (int i = 0; subcommands[i]; i++) {
+			if (i)
+				strbuf_addstr(&buf, "|");
+			strbuf_addstr(&buf, subcommands[i]);
+		}
+		strbuf_addstr(&buf, "}");
+
+		usagestr[0] = strdup(buf.buf);
+		strbuf_release(&buf);
+	}
+
+	parse_options_start(&ctx, argc, argv, flags);
+	switch (parse_options_step(&ctx, options, usagestr)) {
+	case PARSE_OPT_HELP:
+		exit(129);
+	case PARSE_OPT_DONE:
+		break;
+	case PARSE_OPT_LIST_OPTS:
+		while (options->type != OPTION_END) {
+			if (options->long_name)
+				printf("--%s ", options->long_name);
+			options++;
+		}
+		putchar('\n');
+		exit(130);
+	case PARSE_OPT_LIST_SUBCMDS:
+		if (subcommands) {
+			for (int i = 0; subcommands[i]; i++)
+				printf("%s ", subcommands[i]);
+		}
+		putchar('\n');
+		exit(130);
+	default: /* PARSE_OPT_UNKNOWN */
+		if (ctx.argv[0][1] == '-') {
+			error("unknown option `%s'", ctx.argv[0] + 2);
+		} else {
+			error("unknown switch `%c'", *ctx.opt);
+		}
+		usage_with_options(usagestr, options);
+	}
+
+	return parse_options_end(&ctx);
+}
+
+int parse_options(int argc, const char **argv, const struct option *options,
+		  const char * const usagestr[], int flags)
+{
+	return parse_options_subcommand(argc, argv, options, NULL,
+					(const char **) usagestr, flags);
+}
+
+#define USAGE_OPTS_WIDTH 24
+#define USAGE_GAP         2
+
+static void print_option_help(const struct option *opts, int full)
+{
+	size_t pos;
+	int pad;
+
+	if (opts->type == OPTION_GROUP) {
+		fputc('\n', stderr);
+		if (*opts->help)
+			fprintf(stderr, "%s\n", opts->help);
+		return;
+	}
+	if (!full && (opts->flags & PARSE_OPT_HIDDEN))
+		return;
+	if (opts->flags & PARSE_OPT_DISABLED)
+		return;
+
+	pos = fprintf(stderr, "    ");
+	if (opts->short_name)
+		pos += fprintf(stderr, "-%c", opts->short_name);
+	else
+		pos += fprintf(stderr, "    ");
+
+	if (opts->long_name && opts->short_name)
+		pos += fprintf(stderr, ", ");
+	if (opts->long_name)
+		pos += fprintf(stderr, "--%s", opts->long_name);
+
+	switch (opts->type) {
+	case OPTION_ARGUMENT:
+		break;
+	case OPTION_LONG:
+	case OPTION_U64:
+	case OPTION_INTEGER:
+	case OPTION_UINTEGER:
+		if (opts->flags & PARSE_OPT_OPTARG)
+			if (opts->long_name)
+				pos += fprintf(stderr, "[=<n>]");
+			else
+				pos += fprintf(stderr, "[<n>]");
+		else
+			pos += fprintf(stderr, " <n>");
+		break;
+	case OPTION_CALLBACK:
+		if (opts->flags & PARSE_OPT_NOARG)
+			break;
+		/* FALLTHROUGH */
+	case OPTION_STRING:
+		if (opts->argh) {
+			if (opts->flags & PARSE_OPT_OPTARG)
+				if (opts->long_name)
+					pos += fprintf(stderr, "[=<%s>]", opts->argh);
+				else
+					pos += fprintf(stderr, "[<%s>]", opts->argh);
+			else
+				pos += fprintf(stderr, " <%s>", opts->argh);
+		} else {
+			if (opts->flags & PARSE_OPT_OPTARG)
+				if (opts->long_name)
+					pos += fprintf(stderr, "[=...]");
+				else
+					pos += fprintf(stderr, "[...]");
+			else
+				pos += fprintf(stderr, " ...");
+		}
+		break;
+	default: /* OPTION_{BIT,BOOLEAN,SET_UINT,SET_PTR} */
+	case OPTION_END:
+	case OPTION_GROUP:
+	case OPTION_BIT:
+	case OPTION_BOOLEAN:
+	case OPTION_INCR:
+	case OPTION_SET_UINT:
+	case OPTION_SET_PTR:
+		break;
+	}
+
+	if (pos <= USAGE_OPTS_WIDTH)
+		pad = USAGE_OPTS_WIDTH - pos;
+	else {
+		fputc('\n', stderr);
+		pad = USAGE_OPTS_WIDTH;
+	}
+	fprintf(stderr, "%*s%s\n", pad + USAGE_GAP, "", opts->help);
+}
+
+int usage_with_options_internal(const char * const *usagestr,
+				const struct option *opts, int full)
+{
+	if (!usagestr)
+		return PARSE_OPT_HELP;
+
+	fprintf(stderr, "\n usage: %s\n", *usagestr++);
+	while (*usagestr && **usagestr)
+		fprintf(stderr, "    or: %s\n", *usagestr++);
+	while (*usagestr) {
+		fprintf(stderr, "%s%s\n",
+				**usagestr ? "    " : "",
+				*usagestr);
+		usagestr++;
+	}
+
+	if (opts->type != OPTION_GROUP)
+		fputc('\n', stderr);
+
+	for (  ; opts->type != OPTION_END; opts++)
+		print_option_help(opts, full);
+
+	fputc('\n', stderr);
+
+	return PARSE_OPT_HELP;
+}
+
+void usage_with_options(const char * const *usagestr,
+			const struct option *opts)
+{
+	exit_browser(false);
+	usage_with_options_internal(usagestr, opts, 0);
+	exit(129);
+}
+
+int parse_options_usage(const char * const *usagestr,
+			const struct option *opts,
+			const char *optstr, bool short_opt)
+{
+	if (!usagestr)
+		goto opt;
+
+	fprintf(stderr, "\n usage: %s\n", *usagestr++);
+	while (*usagestr && **usagestr)
+		fprintf(stderr, "    or: %s\n", *usagestr++);
+	while (*usagestr) {
+		fprintf(stderr, "%s%s\n",
+				**usagestr ? "    " : "",
+				*usagestr);
+		usagestr++;
+	}
+	fputc('\n', stderr);
+
+opt:
+	for (  ; opts->type != OPTION_END; opts++) {
+		if (short_opt) {
+			if (opts->short_name == *optstr)
+				break;
+			continue;
+		}
+
+		if (opts->long_name == NULL)
+			continue;
+
+		if (!prefixcmp(optstr, opts->long_name))
+			break;
+		if (!prefixcmp(optstr, "no-") &&
+		    !prefixcmp(optstr + 3, opts->long_name))
+			break;
+	}
+
+	if (opts->type != OPTION_END)
+		print_option_help(opts, 0);
+
+	return PARSE_OPT_HELP;
+}
+
+
+int parse_opt_verbosity_cb(const struct option *opt,
+			   const char *arg __maybe_unused,
+			   int unset)
+{
+	int *target = opt->value;
+
+	if (unset)
+		/* --no-quiet, --no-verbose */
+		*target = 0;
+	else if (opt->short_name == 'v') {
+		if (*target >= 0)
+			(*target)++;
+		else
+			*target = 1;
+	} else {
+		if (*target <= 0)
+			(*target)--;
+		else
+			*target = -1;
+	}
+	return 0;
+}
+
+void set_option_flag(struct option *opts, int shortopt, const char *longopt,
+		     int flag)
+{
+	for (; opts->type != OPTION_END; opts++) {
+		if ((shortopt && opts->short_name == shortopt) ||
+		    (opts->long_name && longopt &&
+		     !strcmp(opts->long_name, longopt))) {
+			opts->flags |= flag;
+			break;
+		}
+	}
+}
diff --git a/tools/perf/util/parse-options.h b/tools/perf/util/parse-options.h
new file mode 100644
index 000000000..59561fd86
--- /dev/null
+++ b/tools/perf/util/parse-options.h
@@ -0,0 +1,220 @@
+#ifndef __PERF_PARSE_OPTIONS_H
+#define __PERF_PARSE_OPTIONS_H
+
+#include <linux/kernel.h>
+#include <stdbool.h>
+
+enum parse_opt_type {
+	/* special types */
+	OPTION_END,
+	OPTION_ARGUMENT,
+	OPTION_GROUP,
+	/* options with no arguments */
+	OPTION_BIT,
+	OPTION_BOOLEAN,
+	OPTION_INCR,
+	OPTION_SET_UINT,
+	OPTION_SET_PTR,
+	/* options with arguments (usually) */
+	OPTION_STRING,
+	OPTION_INTEGER,
+	OPTION_LONG,
+	OPTION_CALLBACK,
+	OPTION_U64,
+	OPTION_UINTEGER,
+};
+
+enum parse_opt_flags {
+	PARSE_OPT_KEEP_DASHDASH = 1,
+	PARSE_OPT_STOP_AT_NON_OPTION = 2,
+	PARSE_OPT_KEEP_ARGV0 = 4,
+	PARSE_OPT_KEEP_UNKNOWN = 8,
+	PARSE_OPT_NO_INTERNAL_HELP = 16,
+};
+
+enum parse_opt_option_flags {
+	PARSE_OPT_OPTARG  = 1,
+	PARSE_OPT_NOARG   = 2,
+	PARSE_OPT_NONEG   = 4,
+	PARSE_OPT_HIDDEN  = 8,
+	PARSE_OPT_LASTARG_DEFAULT = 16,
+	PARSE_OPT_DISABLED = 32,
+	PARSE_OPT_EXCLUSIVE = 64,
+	PARSE_OPT_NOEMPTY  = 128,
+};
+
+struct option;
+typedef int parse_opt_cb(const struct option *, const char *arg, int unset);
+
+/*
+ * `type`::
+ *   holds the type of the option, you must have an OPTION_END last in your
+ *   array.
+ *
+ * `short_name`::
+ *   the character to use as a short option name, '\0' if none.
+ *
+ * `long_name`::
+ *   the long option name, without the leading dashes, NULL if none.
+ *
+ * `value`::
+ *   stores pointers to the values to be filled.
+ *
+ * `argh`::
+ *   token to explain the kind of argument this option wants. Keep it
+ *   homogenous across the repository.
+ *
+ * `help`::
+ *   the short help associated to what the option does.
+ *   Must never be NULL (except for OPTION_END).
+ *   OPTION_GROUP uses this pointer to store the group header.
+ *
+ * `flags`::
+ *   mask of parse_opt_option_flags.
+ *   PARSE_OPT_OPTARG: says that the argument is optionnal (not for BOOLEANs)
+ *   PARSE_OPT_NOARG: says that this option takes no argument, for CALLBACKs
+ *   PARSE_OPT_NONEG: says that this option cannot be negated
+ *   PARSE_OPT_HIDDEN this option is skipped in the default usage, showed in
+ *                    the long one.
+ *
+ * `callback`::
+ *   pointer to the callback to use for OPTION_CALLBACK.
+ *
+ * `defval`::
+ *   default value to fill (*->value) with for PARSE_OPT_OPTARG.
+ *   OPTION_{BIT,SET_UINT,SET_PTR} store the {mask,integer,pointer} to put in
+ *   the value when met.
+ *   CALLBACKS can use it like they want.
+ *
+ * `set`::
+ *   whether an option was set by the user
+ */
+struct option {
+	enum parse_opt_type type;
+	int short_name;
+	const char *long_name;
+	void *value;
+	const char *argh;
+	const char *help;
+
+	int flags;
+	parse_opt_cb *callback;
+	intptr_t defval;
+	bool *set;
+	void *data;
+};
+
+#define check_vtype(v, type) ( BUILD_BUG_ON_ZERO(!__builtin_types_compatible_p(typeof(v), type)) + v )
+
+#define OPT_END()                   { .type = OPTION_END }
+#define OPT_ARGUMENT(l, h)          { .type = OPTION_ARGUMENT, .long_name = (l), .help = (h) }
+#define OPT_GROUP(h)                { .type = OPTION_GROUP, .help = (h) }
+#define OPT_BIT(s, l, v, h, b)      { .type = OPTION_BIT, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h), .defval = (b) }
+#define OPT_BOOLEAN(s, l, v, h)     { .type = OPTION_BOOLEAN, .short_name = (s), .long_name = (l), .value = check_vtype(v, bool *), .help = (h) }
+#define OPT_BOOLEAN_SET(s, l, v, os, h) \
+	{ .type = OPTION_BOOLEAN, .short_name = (s), .long_name = (l), \
+	.value = check_vtype(v, bool *), .help = (h), \
+	.set = check_vtype(os, bool *)}
+#define OPT_INCR(s, l, v, h)        { .type = OPTION_INCR, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h) }
+#define OPT_SET_UINT(s, l, v, h, i)  { .type = OPTION_SET_UINT, .short_name = (s), .long_name = (l), .value = check_vtype(v, unsigned int *), .help = (h), .defval = (i) }
+#define OPT_SET_PTR(s, l, v, h, p)  { .type = OPTION_SET_PTR, .short_name = (s), .long_name = (l), .value = (v), .help = (h), .defval = (p) }
+#define OPT_INTEGER(s, l, v, h)     { .type = OPTION_INTEGER, .short_name = (s), .long_name = (l), .value = check_vtype(v, int *), .help = (h) }
+#define OPT_UINTEGER(s, l, v, h)    { .type = OPTION_UINTEGER, .short_name = (s), .long_name = (l), .value = check_vtype(v, unsigned int *), .help = (h) }
+#define OPT_LONG(s, l, v, h)        { .type = OPTION_LONG, .short_name = (s), .long_name = (l), .value = check_vtype(v, long *), .help = (h) }
+#define OPT_U64(s, l, v, h)         { .type = OPTION_U64, .short_name = (s), .long_name = (l), .value = check_vtype(v, u64 *), .help = (h) }
+#define OPT_STRING(s, l, v, a, h)   { .type = OPTION_STRING,  .short_name = (s), .long_name = (l), .value = check_vtype(v, const char **), (a), .help = (h) }
+#define OPT_STRING_NOEMPTY(s, l, v, a, h)   { .type = OPTION_STRING,  .short_name = (s), .long_name = (l), .value = check_vtype(v, const char **), (a), .help = (h), .flags = PARSE_OPT_NOEMPTY}
+#define OPT_DATE(s, l, v, h) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), .argh = "time", .help = (h), .callback = parse_opt_approxidate_cb }
+#define OPT_CALLBACK(s, l, v, a, h, f) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f) }
+#define OPT_CALLBACK_NOOPT(s, l, v, a, h, f) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .flags = PARSE_OPT_NOARG }
+#define OPT_CALLBACK_DEFAULT(s, l, v, a, h, f, d) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d, .flags = PARSE_OPT_LASTARG_DEFAULT }
+#define OPT_CALLBACK_DEFAULT_NOOPT(s, l, v, a, h, f, d) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l),\
+	.value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d,\
+	.flags = PARSE_OPT_LASTARG_DEFAULT | PARSE_OPT_NOARG}
+#define OPT_CALLBACK_OPTARG(s, l, v, d, a, h, f) \
+	{ .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), \
+	  .value = (v), (a), .help = (h), .callback = (f), \
+	  .flags = PARSE_OPT_OPTARG, .data = (d) }
+
+/* parse_options() will filter out the processed options and leave the
+ * non-option argments in argv[].
+ * Returns the number of arguments left in argv[].
+ */
+extern int parse_options(int argc, const char **argv,
+                         const struct option *options,
+                         const char * const usagestr[], int flags);
+
+extern int parse_options_subcommand(int argc, const char **argv,
+				const struct option *options,
+				const char *const subcommands[],
+				const char *usagestr[], int flags);
+
+extern NORETURN void usage_with_options(const char * const *usagestr,
+                                        const struct option *options);
+
+/*----- incremantal advanced APIs -----*/
+
+enum {
+	PARSE_OPT_HELP = -1,
+	PARSE_OPT_DONE,
+	PARSE_OPT_LIST_OPTS,
+	PARSE_OPT_LIST_SUBCMDS,
+	PARSE_OPT_UNKNOWN,
+};
+
+/*
+ * It's okay for the caller to consume argv/argc in the usual way.
+ * Other fields of that structure are private to parse-options and should not
+ * be modified in any way.
+ */
+struct parse_opt_ctx_t {
+	const char **argv;
+	const char **out;
+	int argc, cpidx;
+	const char *opt;
+	const struct option *excl_opt;
+	int flags;
+};
+
+extern int parse_options_usage(const char * const *usagestr,
+			       const struct option *opts,
+			       const char *optstr,
+			       bool short_opt);
+
+extern void parse_options_start(struct parse_opt_ctx_t *ctx,
+				int argc, const char **argv, int flags);
+
+extern int parse_options_step(struct parse_opt_ctx_t *ctx,
+			      const struct option *options,
+			      const char * const usagestr[]);
+
+extern int parse_options_end(struct parse_opt_ctx_t *ctx);
+
+
+/*----- some often used options -----*/
+extern int parse_opt_abbrev_cb(const struct option *, const char *, int);
+extern int parse_opt_approxidate_cb(const struct option *, const char *, int);
+extern int parse_opt_verbosity_cb(const struct option *, const char *, int);
+
+#define OPT__VERBOSE(var)  OPT_BOOLEAN('v', "verbose", (var), "be verbose")
+#define OPT__QUIET(var)    OPT_BOOLEAN('q', "quiet",   (var), "be quiet")
+#define OPT__VERBOSITY(var) \
+	{ OPTION_CALLBACK, 'v', "verbose", (var), NULL, "be more verbose", \
+	  PARSE_OPT_NOARG, &parse_opt_verbosity_cb, 0 }, \
+	{ OPTION_CALLBACK, 'q', "quiet", (var), NULL, "be more quiet", \
+	  PARSE_OPT_NOARG, &parse_opt_verbosity_cb, 0 }
+#define OPT__DRY_RUN(var)  OPT_BOOLEAN('n', "dry-run", (var), "dry run")
+#define OPT__ABBREV(var)  \
+	{ OPTION_CALLBACK, 0, "abbrev", (var), "n", \
+	  "use <n> digits to display SHA-1s", \
+	  PARSE_OPT_OPTARG, &parse_opt_abbrev_cb, 0 }
+
+extern const char *parse_options_fix_filename(const char *prefix, const char *file);
+
+void set_option_flag(struct option *opts, int sopt, const char *lopt, int flag);
+#endif /* __PERF_PARSE_OPTIONS_H */
diff --git a/tools/perf/util/path.c b/tools/perf/util/path.c
new file mode 100644
index 000000000..5d13cb45b
--- /dev/null
+++ b/tools/perf/util/path.c
@@ -0,0 +1,161 @@
+/*
+ * I'm tired of doing "vsnprintf()" etc just to open a
+ * file, so here's a "return static buffer with printf"
+ * interface for paths.
+ *
+ * It's obviously not thread-safe. Sue me. But it's quite
+ * useful for doing things like
+ *
+ *   f = open(mkpath("%s/%s.perf", base, name), O_RDONLY);
+ *
+ * which is what it's designed for.
+ */
+#include "cache.h"
+
+static char bad_path[] = "/bad-path/";
+/*
+ * Two hacks:
+ */
+
+static const char *get_perf_dir(void)
+{
+	return ".";
+}
+
+/*
+ * If libc has strlcpy() then that version will override this
+ * implementation:
+ */
+size_t __weak strlcpy(char *dest, const char *src, size_t size)
+{
+	size_t ret = strlen(src);
+
+	if (size) {
+		size_t len = (ret >= size) ? size - 1 : ret;
+
+		memcpy(dest, src, len);
+		dest[len] = '\0';
+	}
+
+	return ret;
+}
+
+static char *get_pathname(void)
+{
+	static char pathname_array[4][PATH_MAX];
+	static int idx;
+
+	return pathname_array[3 & ++idx];
+}
+
+static char *cleanup_path(char *path)
+{
+	/* Clean it up */
+	if (!memcmp(path, "./", 2)) {
+		path += 2;
+		while (*path == '/')
+			path++;
+	}
+	return path;
+}
+
+static char *perf_vsnpath(char *buf, size_t n, const char *fmt, va_list args)
+{
+	const char *perf_dir = get_perf_dir();
+	size_t len;
+
+	len = strlen(perf_dir);
+	if (n < len + 1)
+		goto bad;
+	memcpy(buf, perf_dir, len);
+	if (len && !is_dir_sep(perf_dir[len-1]))
+		buf[len++] = '/';
+	len += vsnprintf(buf + len, n - len, fmt, args);
+	if (len >= n)
+		goto bad;
+	return cleanup_path(buf);
+bad:
+	strlcpy(buf, bad_path, n);
+	return buf;
+}
+
+char *perf_pathdup(const char *fmt, ...)
+{
+	char path[PATH_MAX];
+	va_list args;
+	va_start(args, fmt);
+	(void)perf_vsnpath(path, sizeof(path), fmt, args);
+	va_end(args);
+	return xstrdup(path);
+}
+
+char *mkpath(const char *fmt, ...)
+{
+	va_list args;
+	unsigned len;
+	char *pathname = get_pathname();
+
+	va_start(args, fmt);
+	len = vsnprintf(pathname, PATH_MAX, fmt, args);
+	va_end(args);
+	if (len >= PATH_MAX)
+		return bad_path;
+	return cleanup_path(pathname);
+}
+
+char *perf_path(const char *fmt, ...)
+{
+	const char *perf_dir = get_perf_dir();
+	char *pathname = get_pathname();
+	va_list args;
+	unsigned len;
+
+	len = strlen(perf_dir);
+	if (len > PATH_MAX-100)
+		return bad_path;
+	memcpy(pathname, perf_dir, len);
+	if (len && perf_dir[len-1] != '/')
+		pathname[len++] = '/';
+	va_start(args, fmt);
+	len += vsnprintf(pathname + len, PATH_MAX - len, fmt, args);
+	va_end(args);
+	if (len >= PATH_MAX)
+		return bad_path;
+	return cleanup_path(pathname);
+}
+
+/* strip arbitrary amount of directory separators at end of path */
+static inline int chomp_trailing_dir_sep(const char *path, int len)
+{
+	while (len && is_dir_sep(path[len - 1]))
+		len--;
+	return len;
+}
+
+/*
+ * If path ends with suffix (complete path components), returns the
+ * part before suffix (sans trailing directory separators).
+ * Otherwise returns NULL.
+ */
+char *strip_path_suffix(const char *path, const char *suffix)
+{
+	int path_len = strlen(path), suffix_len = strlen(suffix);
+
+	while (suffix_len) {
+		if (!path_len)
+			return NULL;
+
+		if (is_dir_sep(path[path_len - 1])) {
+			if (!is_dir_sep(suffix[suffix_len - 1]))
+				return NULL;
+			path_len = chomp_trailing_dir_sep(path, path_len);
+			suffix_len = chomp_trailing_dir_sep(suffix, suffix_len);
+		}
+		else if (path[--path_len] != suffix[--suffix_len])
+			return NULL;
+	}
+
+	if (path_len && !is_dir_sep(path[path_len - 1]))
+		return NULL;
+	return strndup(path, chomp_trailing_dir_sep(path, path_len));
+}
diff --git a/tools/perf/util/perf_regs.c b/tools/perf/util/perf_regs.c
new file mode 100644
index 000000000..43168fb0d
--- /dev/null
+++ b/tools/perf/util/perf_regs.c
@@ -0,0 +1,27 @@
+#include <errno.h>
+#include "perf_regs.h"
+#include "event.h"
+
+int perf_reg_value(u64 *valp, struct regs_dump *regs, int id)
+{
+	int i, idx = 0;
+	u64 mask = regs->mask;
+
+	if (regs->cache_mask & (1 << id))
+		goto out;
+
+	if (!(mask & (1 << id)))
+		return -EINVAL;
+
+	for (i = 0; i < id; i++) {
+		if (mask & (1 << i))
+			idx++;
+	}
+
+	regs->cache_mask |= (1 << id);
+	regs->cache_regs[id] = regs->regs[idx];
+
+out:
+	*valp = regs->cache_regs[id];
+	return 0;
+}
diff --git a/tools/perf/util/perf_regs.h b/tools/perf/util/perf_regs.h
new file mode 100644
index 000000000..980dbf76b
--- /dev/null
+++ b/tools/perf/util/perf_regs.h
@@ -0,0 +1,29 @@
+#ifndef __PERF_REGS_H
+#define __PERF_REGS_H
+
+#include <linux/types.h>
+
+struct regs_dump;
+
+#ifdef HAVE_PERF_REGS_SUPPORT
+#include <perf_regs.h>
+
+int perf_reg_value(u64 *valp, struct regs_dump *regs, int id);
+
+#else
+#define PERF_REGS_MASK	0
+#define PERF_REGS_MAX	0
+
+static inline const char *perf_reg_name(int id __maybe_unused)
+{
+	return NULL;
+}
+
+static inline int perf_reg_value(u64 *valp __maybe_unused,
+				 struct regs_dump *regs __maybe_unused,
+				 int id __maybe_unused)
+{
+	return 0;
+}
+#endif /* HAVE_PERF_REGS_SUPPORT */
+#endif /* __PERF_REGS_H */
diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
new file mode 100644
index 000000000..48411674d
--- /dev/null
+++ b/tools/perf/util/pmu.c
@@ -0,0 +1,992 @@
+#include <linux/list.h>
+#include <sys/types.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <stdarg.h>
+#include <dirent.h>
+#include <api/fs/fs.h>
+#include <locale.h>
+#include "util.h"
+#include "pmu.h"
+#include "parse-events.h"
+#include "cpumap.h"
+
+struct perf_pmu_format {
+	char *name;
+	int value;
+	DECLARE_BITMAP(bits, PERF_PMU_FORMAT_BITS);
+	struct list_head list;
+};
+
+#define EVENT_SOURCE_DEVICE_PATH "/bus/event_source/devices/"
+
+int perf_pmu_parse(struct list_head *list, char *name);
+extern FILE *perf_pmu_in;
+
+static LIST_HEAD(pmus);
+
+/*
+ * Parse & process all the sysfs attributes located under
+ * the directory specified in 'dir' parameter.
+ */
+int perf_pmu__format_parse(char *dir, struct list_head *head)
+{
+	struct dirent *evt_ent;
+	DIR *format_dir;
+	int ret = 0;
+
+	format_dir = opendir(dir);
+	if (!format_dir)
+		return -EINVAL;
+
+	while (!ret && (evt_ent = readdir(format_dir))) {
+		char path[PATH_MAX];
+		char *name = evt_ent->d_name;
+		FILE *file;
+
+		if (!strcmp(name, ".") || !strcmp(name, ".."))
+			continue;
+
+		snprintf(path, PATH_MAX, "%s/%s", dir, name);
+
+		ret = -EINVAL;
+		file = fopen(path, "r");
+		if (!file)
+			break;
+
+		perf_pmu_in = file;
+		ret = perf_pmu_parse(head, name);
+		fclose(file);
+	}
+
+	closedir(format_dir);
+	return ret;
+}
+
+/*
+ * Reading/parsing the default pmu format definition, which should be
+ * located at:
+ * /sys/bus/event_source/devices/<dev>/format as sysfs group attributes.
+ */
+static int pmu_format(const char *name, struct list_head *format)
+{
+	struct stat st;
+	char path[PATH_MAX];
+	const char *sysfs = sysfs__mountpoint();
+
+	if (!sysfs)
+		return -1;
+
+	snprintf(path, PATH_MAX,
+		 "%s" EVENT_SOURCE_DEVICE_PATH "%s/format", sysfs, name);
+
+	if (stat(path, &st) < 0)
+		return 0;	/* no error if format does not exist */
+
+	if (perf_pmu__format_parse(path, format))
+		return -1;
+
+	return 0;
+}
+
+static int perf_pmu__parse_scale(struct perf_pmu_alias *alias, char *dir, char *name)
+{
+	struct stat st;
+	ssize_t sret;
+	char scale[128];
+	int fd, ret = -1;
+	char path[PATH_MAX];
+	const char *lc;
+
+	snprintf(path, PATH_MAX, "%s/%s.scale", dir, name);
+
+	fd = open(path, O_RDONLY);
+	if (fd == -1)
+		return -1;
+
+	if (fstat(fd, &st) < 0)
+		goto error;
+
+	sret = read(fd, scale, sizeof(scale)-1);
+	if (sret < 0)
+		goto error;
+
+	scale[sret] = '\0';
+	/*
+	 * save current locale
+	 */
+	lc = setlocale(LC_NUMERIC, NULL);
+
+	/*
+	 * force to C locale to ensure kernel
+	 * scale string is converted correctly.
+	 * kernel uses default C locale.
+	 */
+	setlocale(LC_NUMERIC, "C");
+
+	alias->scale = strtod(scale, NULL);
+
+	/* restore locale */
+	setlocale(LC_NUMERIC, lc);
+
+	ret = 0;
+error:
+	close(fd);
+	return ret;
+}
+
+static int perf_pmu__parse_unit(struct perf_pmu_alias *alias, char *dir, char *name)
+{
+	char path[PATH_MAX];
+	ssize_t sret;
+	int fd;
+
+	snprintf(path, PATH_MAX, "%s/%s.unit", dir, name);
+
+	fd = open(path, O_RDONLY);
+	if (fd == -1)
+		return -1;
+
+		sret = read(fd, alias->unit, UNIT_MAX_LEN);
+	if (sret < 0)
+		goto error;
+
+	close(fd);
+
+	alias->unit[sret] = '\0';
+
+	return 0;
+error:
+	close(fd);
+	alias->unit[0] = '\0';
+	return -1;
+}
+
+static int
+perf_pmu__parse_per_pkg(struct perf_pmu_alias *alias, char *dir, char *name)
+{
+	char path[PATH_MAX];
+	int fd;
+
+	snprintf(path, PATH_MAX, "%s/%s.per-pkg", dir, name);
+
+	fd = open(path, O_RDONLY);
+	if (fd == -1)
+		return -1;
+
+	close(fd);
+
+	alias->per_pkg = true;
+	return 0;
+}
+
+static int perf_pmu__parse_snapshot(struct perf_pmu_alias *alias,
+				    char *dir, char *name)
+{
+	char path[PATH_MAX];
+	int fd;
+
+	snprintf(path, PATH_MAX, "%s/%s.snapshot", dir, name);
+
+	fd = open(path, O_RDONLY);
+	if (fd == -1)
+		return -1;
+
+	alias->snapshot = true;
+	close(fd);
+	return 0;
+}
+
+static int perf_pmu__new_alias(struct list_head *list, char *dir, char *name, FILE *file)
+{
+	struct perf_pmu_alias *alias;
+	char buf[256];
+	int ret;
+
+	ret = fread(buf, 1, sizeof(buf), file);
+	if (ret == 0)
+		return -EINVAL;
+	buf[ret] = 0;
+
+	alias = malloc(sizeof(*alias));
+	if (!alias)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&alias->terms);
+	alias->scale = 1.0;
+	alias->unit[0] = '\0';
+	alias->per_pkg = false;
+
+	ret = parse_events_terms(&alias->terms, buf);
+	if (ret) {
+		free(alias);
+		return ret;
+	}
+
+	alias->name = strdup(name);
+	/*
+	 * load unit name and scale if available
+	 */
+	perf_pmu__parse_unit(alias, dir, name);
+	perf_pmu__parse_scale(alias, dir, name);
+	perf_pmu__parse_per_pkg(alias, dir, name);
+	perf_pmu__parse_snapshot(alias, dir, name);
+
+	list_add_tail(&alias->list, list);
+
+	return 0;
+}
+
+static inline bool pmu_alias_info_file(char *name)
+{
+	size_t len;
+
+	len = strlen(name);
+	if (len > 5 && !strcmp(name + len - 5, ".unit"))
+		return true;
+	if (len > 6 && !strcmp(name + len - 6, ".scale"))
+		return true;
+	if (len > 8 && !strcmp(name + len - 8, ".per-pkg"))
+		return true;
+	if (len > 9 && !strcmp(name + len - 9, ".snapshot"))
+		return true;
+
+	return false;
+}
+
+/*
+ * Process all the sysfs attributes located under the directory
+ * specified in 'dir' parameter.
+ */
+static int pmu_aliases_parse(char *dir, struct list_head *head)
+{
+	struct dirent *evt_ent;
+	DIR *event_dir;
+	int ret = 0;
+
+	event_dir = opendir(dir);
+	if (!event_dir)
+		return -EINVAL;
+
+	while (!ret && (evt_ent = readdir(event_dir))) {
+		char path[PATH_MAX];
+		char *name = evt_ent->d_name;
+		FILE *file;
+
+		if (!strcmp(name, ".") || !strcmp(name, ".."))
+			continue;
+
+		/*
+		 * skip info files parsed in perf_pmu__new_alias()
+		 */
+		if (pmu_alias_info_file(name))
+			continue;
+
+		snprintf(path, PATH_MAX, "%s/%s", dir, name);
+
+		ret = -EINVAL;
+		file = fopen(path, "r");
+		if (!file)
+			break;
+
+		ret = perf_pmu__new_alias(head, dir, name, file);
+		fclose(file);
+	}
+
+	closedir(event_dir);
+	return ret;
+}
+
+/*
+ * Reading the pmu event aliases definition, which should be located at:
+ * /sys/bus/event_source/devices/<dev>/events as sysfs group attributes.
+ */
+static int pmu_aliases(const char *name, struct list_head *head)
+{
+	struct stat st;
+	char path[PATH_MAX];
+	const char *sysfs = sysfs__mountpoint();
+
+	if (!sysfs)
+		return -1;
+
+	snprintf(path, PATH_MAX,
+		 "%s/bus/event_source/devices/%s/events", sysfs, name);
+
+	if (stat(path, &st) < 0)
+		return 0;	 /* no error if 'events' does not exist */
+
+	if (pmu_aliases_parse(path, head))
+		return -1;
+
+	return 0;
+}
+
+static int pmu_alias_terms(struct perf_pmu_alias *alias,
+			   struct list_head *terms)
+{
+	struct parse_events_term *term, *cloned;
+	LIST_HEAD(list);
+	int ret;
+
+	list_for_each_entry(term, &alias->terms, list) {
+		ret = parse_events_term__clone(&cloned, term);
+		if (ret) {
+			parse_events__free_terms(&list);
+			return ret;
+		}
+		list_add_tail(&cloned->list, &list);
+	}
+	list_splice(&list, terms);
+	return 0;
+}
+
+/*
+ * Reading/parsing the default pmu type value, which should be
+ * located at:
+ * /sys/bus/event_source/devices/<dev>/type as sysfs attribute.
+ */
+static int pmu_type(const char *name, __u32 *type)
+{
+	struct stat st;
+	char path[PATH_MAX];
+	FILE *file;
+	int ret = 0;
+	const char *sysfs = sysfs__mountpoint();
+
+	if (!sysfs)
+		return -1;
+
+	snprintf(path, PATH_MAX,
+		 "%s" EVENT_SOURCE_DEVICE_PATH "%s/type", sysfs, name);
+
+	if (stat(path, &st) < 0)
+		return -1;
+
+	file = fopen(path, "r");
+	if (!file)
+		return -EINVAL;
+
+	if (1 != fscanf(file, "%u", type))
+		ret = -1;
+
+	fclose(file);
+	return ret;
+}
+
+/* Add all pmus in sysfs to pmu list: */
+static void pmu_read_sysfs(void)
+{
+	char path[PATH_MAX];
+	DIR *dir;
+	struct dirent *dent;
+	const char *sysfs = sysfs__mountpoint();
+
+	if (!sysfs)
+		return;
+
+	snprintf(path, PATH_MAX,
+		 "%s" EVENT_SOURCE_DEVICE_PATH, sysfs);
+
+	dir = opendir(path);
+	if (!dir)
+		return;
+
+	while ((dent = readdir(dir))) {
+		if (!strcmp(dent->d_name, ".") || !strcmp(dent->d_name, ".."))
+			continue;
+		/* add to static LIST_HEAD(pmus): */
+		perf_pmu__find(dent->d_name);
+	}
+
+	closedir(dir);
+}
+
+static struct cpu_map *pmu_cpumask(const char *name)
+{
+	struct stat st;
+	char path[PATH_MAX];
+	FILE *file;
+	struct cpu_map *cpus;
+	const char *sysfs = sysfs__mountpoint();
+
+	if (!sysfs)
+		return NULL;
+
+	snprintf(path, PATH_MAX,
+		 "%s/bus/event_source/devices/%s/cpumask", sysfs, name);
+
+	if (stat(path, &st) < 0)
+		return NULL;
+
+	file = fopen(path, "r");
+	if (!file)
+		return NULL;
+
+	cpus = cpu_map__read(file);
+	fclose(file);
+	return cpus;
+}
+
+struct perf_event_attr *__attribute__((weak))
+perf_pmu__get_default_config(struct perf_pmu *pmu __maybe_unused)
+{
+	return NULL;
+}
+
+static struct perf_pmu *pmu_lookup(const char *name)
+{
+	struct perf_pmu *pmu;
+	LIST_HEAD(format);
+	LIST_HEAD(aliases);
+	__u32 type;
+
+	/*
+	 * The pmu data we store & need consists of the pmu
+	 * type value and format definitions. Load both right
+	 * now.
+	 */
+	if (pmu_format(name, &format))
+		return NULL;
+
+	if (pmu_aliases(name, &aliases))
+		return NULL;
+
+	if (pmu_type(name, &type))
+		return NULL;
+
+	pmu = zalloc(sizeof(*pmu));
+	if (!pmu)
+		return NULL;
+
+	pmu->cpus = pmu_cpumask(name);
+
+	INIT_LIST_HEAD(&pmu->format);
+	INIT_LIST_HEAD(&pmu->aliases);
+	list_splice(&format, &pmu->format);
+	list_splice(&aliases, &pmu->aliases);
+	pmu->name = strdup(name);
+	pmu->type = type;
+	list_add_tail(&pmu->list, &pmus);
+
+	pmu->default_config = perf_pmu__get_default_config(pmu);
+
+	return pmu;
+}
+
+static struct perf_pmu *pmu_find(const char *name)
+{
+	struct perf_pmu *pmu;
+
+	list_for_each_entry(pmu, &pmus, list)
+		if (!strcmp(pmu->name, name))
+			return pmu;
+
+	return NULL;
+}
+
+struct perf_pmu *perf_pmu__scan(struct perf_pmu *pmu)
+{
+	/*
+	 * pmu iterator: If pmu is NULL, we start at the begin,
+	 * otherwise return the next pmu. Returns NULL on end.
+	 */
+	if (!pmu) {
+		pmu_read_sysfs();
+		pmu = list_prepare_entry(pmu, &pmus, list);
+	}
+	list_for_each_entry_continue(pmu, &pmus, list)
+		return pmu;
+	return NULL;
+}
+
+struct perf_pmu *perf_pmu__find(const char *name)
+{
+	struct perf_pmu *pmu;
+
+	/*
+	 * Once PMU is loaded it stays in the list,
+	 * so we keep us from multiple reading/parsing
+	 * the pmu format definitions.
+	 */
+	pmu = pmu_find(name);
+	if (pmu)
+		return pmu;
+
+	return pmu_lookup(name);
+}
+
+static struct perf_pmu_format *
+pmu_find_format(struct list_head *formats, char *name)
+{
+	struct perf_pmu_format *format;
+
+	list_for_each_entry(format, formats, list)
+		if (!strcmp(format->name, name))
+			return format;
+
+	return NULL;
+}
+
+/*
+ * Sets value based on the format definition (format parameter)
+ * and unformated value (value parameter).
+ */
+static void pmu_format_value(unsigned long *format, __u64 value, __u64 *v,
+			     bool zero)
+{
+	unsigned long fbit, vbit;
+
+	for (fbit = 0, vbit = 0; fbit < PERF_PMU_FORMAT_BITS; fbit++) {
+
+		if (!test_bit(fbit, format))
+			continue;
+
+		if (value & (1llu << vbit++))
+			*v |= (1llu << fbit);
+		else if (zero)
+			*v &= ~(1llu << fbit);
+	}
+}
+
+/*
+ * Term is a string term, and might be a param-term. Try to look up it's value
+ * in the remaining terms.
+ * - We have a term like "base-or-format-term=param-term",
+ * - We need to find the value supplied for "param-term" (with param-term named
+ *   in a config string) later on in the term list.
+ */
+static int pmu_resolve_param_term(struct parse_events_term *term,
+				  struct list_head *head_terms,
+				  __u64 *value)
+{
+	struct parse_events_term *t;
+
+	list_for_each_entry(t, head_terms, list) {
+		if (t->type_val == PARSE_EVENTS__TERM_TYPE_NUM) {
+			if (!strcmp(t->config, term->config)) {
+				t->used = true;
+				*value = t->val.num;
+				return 0;
+			}
+		}
+	}
+
+	if (verbose)
+		printf("Required parameter '%s' not specified\n", term->config);
+
+	return -1;
+}
+
+/*
+ * Setup one of config[12] attr members based on the
+ * user input data - term parameter.
+ */
+static int pmu_config_term(struct list_head *formats,
+			   struct perf_event_attr *attr,
+			   struct parse_events_term *term,
+			   struct list_head *head_terms,
+			   bool zero)
+{
+	struct perf_pmu_format *format;
+	__u64 *vp;
+	__u64 val;
+
+	/*
+	 * If this is a parameter we've already used for parameterized-eval,
+	 * skip it in normal eval.
+	 */
+	if (term->used)
+		return 0;
+
+	/*
+	 * Hardcoded terms should be already in, so nothing
+	 * to be done for them.
+	 */
+	if (parse_events__is_hardcoded_term(term))
+		return 0;
+
+	format = pmu_find_format(formats, term->config);
+	if (!format) {
+		if (verbose)
+			printf("Invalid event/parameter '%s'\n", term->config);
+		return -EINVAL;
+	}
+
+	switch (format->value) {
+	case PERF_PMU_FORMAT_VALUE_CONFIG:
+		vp = &attr->config;
+		break;
+	case PERF_PMU_FORMAT_VALUE_CONFIG1:
+		vp = &attr->config1;
+		break;
+	case PERF_PMU_FORMAT_VALUE_CONFIG2:
+		vp = &attr->config2;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/*
+	 * Either directly use a numeric term, or try to translate string terms
+	 * using event parameters.
+	 */
+	if (term->type_val == PARSE_EVENTS__TERM_TYPE_NUM)
+		val = term->val.num;
+	else if (term->type_val == PARSE_EVENTS__TERM_TYPE_STR) {
+		if (strcmp(term->val.str, "?")) {
+			if (verbose)
+				pr_info("Invalid sysfs entry %s=%s\n",
+						term->config, term->val.str);
+			return -EINVAL;
+		}
+
+		if (pmu_resolve_param_term(term, head_terms, &val))
+			return -EINVAL;
+	} else
+		return -EINVAL;
+
+	pmu_format_value(format->bits, val, vp, zero);
+	return 0;
+}
+
+int perf_pmu__config_terms(struct list_head *formats,
+			   struct perf_event_attr *attr,
+			   struct list_head *head_terms,
+			   bool zero)
+{
+	struct parse_events_term *term;
+
+	list_for_each_entry(term, head_terms, list) {
+		if (pmu_config_term(formats, attr, term, head_terms, zero))
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+/*
+ * Configures event's 'attr' parameter based on the:
+ * 1) users input - specified in terms parameter
+ * 2) pmu format definitions - specified by pmu parameter
+ */
+int perf_pmu__config(struct perf_pmu *pmu, struct perf_event_attr *attr,
+		     struct list_head *head_terms)
+{
+	bool zero = !!pmu->default_config;
+
+	attr->type = pmu->type;
+	return perf_pmu__config_terms(&pmu->format, attr, head_terms, zero);
+}
+
+static struct perf_pmu_alias *pmu_find_alias(struct perf_pmu *pmu,
+					     struct parse_events_term *term)
+{
+	struct perf_pmu_alias *alias;
+	char *name;
+
+	if (parse_events__is_hardcoded_term(term))
+		return NULL;
+
+	if (term->type_val == PARSE_EVENTS__TERM_TYPE_NUM) {
+		if (term->val.num != 1)
+			return NULL;
+		if (pmu_find_format(&pmu->format, term->config))
+			return NULL;
+		name = term->config;
+	} else if (term->type_val == PARSE_EVENTS__TERM_TYPE_STR) {
+		if (strcasecmp(term->config, "event"))
+			return NULL;
+		name = term->val.str;
+	} else {
+		return NULL;
+	}
+
+	list_for_each_entry(alias, &pmu->aliases, list) {
+		if (!strcasecmp(alias->name, name))
+			return alias;
+	}
+	return NULL;
+}
+
+
+static int check_info_data(struct perf_pmu_alias *alias,
+			   struct perf_pmu_info *info)
+{
+	/*
+	 * Only one term in event definition can
+	 * define unit, scale and snapshot, fail
+	 * if there's more than one.
+	 */
+	if ((info->unit && alias->unit) ||
+	    (info->scale && alias->scale) ||
+	    (info->snapshot && alias->snapshot))
+		return -EINVAL;
+
+	if (alias->unit)
+		info->unit = alias->unit;
+
+	if (alias->scale)
+		info->scale = alias->scale;
+
+	if (alias->snapshot)
+		info->snapshot = alias->snapshot;
+
+	return 0;
+}
+
+/*
+ * Find alias in the terms list and replace it with the terms
+ * defined for the alias
+ */
+int perf_pmu__check_alias(struct perf_pmu *pmu, struct list_head *head_terms,
+			  struct perf_pmu_info *info)
+{
+	struct parse_events_term *term, *h;
+	struct perf_pmu_alias *alias;
+	int ret;
+
+	info->per_pkg = false;
+
+	/*
+	 * Mark unit and scale as not set
+	 * (different from default values, see below)
+	 */
+	info->unit     = NULL;
+	info->scale    = 0.0;
+	info->snapshot = false;
+
+	list_for_each_entry_safe(term, h, head_terms, list) {
+		alias = pmu_find_alias(pmu, term);
+		if (!alias)
+			continue;
+		ret = pmu_alias_terms(alias, &term->list);
+		if (ret)
+			return ret;
+
+		ret = check_info_data(alias, info);
+		if (ret)
+			return ret;
+
+		if (alias->per_pkg)
+			info->per_pkg = true;
+
+		list_del(&term->list);
+		free(term);
+	}
+
+	/*
+	 * if no unit or scale foundin aliases, then
+	 * set defaults as for evsel
+	 * unit cannot left to NULL
+	 */
+	if (info->unit == NULL)
+		info->unit   = "";
+
+	if (info->scale == 0.0)
+		info->scale  = 1.0;
+
+	return 0;
+}
+
+int perf_pmu__new_format(struct list_head *list, char *name,
+			 int config, unsigned long *bits)
+{
+	struct perf_pmu_format *format;
+
+	format = zalloc(sizeof(*format));
+	if (!format)
+		return -ENOMEM;
+
+	format->name = strdup(name);
+	format->value = config;
+	memcpy(format->bits, bits, sizeof(format->bits));
+
+	list_add_tail(&format->list, list);
+	return 0;
+}
+
+void perf_pmu__set_format(unsigned long *bits, long from, long to)
+{
+	long b;
+
+	if (!to)
+		to = from;
+
+	memset(bits, 0, BITS_TO_BYTES(PERF_PMU_FORMAT_BITS));
+	for (b = from; b <= to; b++)
+		set_bit(b, bits);
+}
+
+static int sub_non_neg(int a, int b)
+{
+	if (b > a)
+		return 0;
+	return a - b;
+}
+
+static char *format_alias(char *buf, int len, struct perf_pmu *pmu,
+			  struct perf_pmu_alias *alias)
+{
+	struct parse_events_term *term;
+	int used = snprintf(buf, len, "%s/%s", pmu->name, alias->name);
+
+	list_for_each_entry(term, &alias->terms, list) {
+		if (term->type_val == PARSE_EVENTS__TERM_TYPE_STR)
+			used += snprintf(buf + used, sub_non_neg(len, used),
+					",%s=%s", term->config,
+					term->val.str);
+	}
+
+	if (sub_non_neg(len, used) > 0) {
+		buf[used] = '/';
+		used++;
+	}
+	if (sub_non_neg(len, used) > 0) {
+		buf[used] = '\0';
+		used++;
+	} else
+		buf[len - 1] = '\0';
+
+	return buf;
+}
+
+static char *format_alias_or(char *buf, int len, struct perf_pmu *pmu,
+			     struct perf_pmu_alias *alias)
+{
+	snprintf(buf, len, "%s OR %s/%s/", alias->name, pmu->name, alias->name);
+	return buf;
+}
+
+static int cmp_string(const void *a, const void *b)
+{
+	const char * const *as = a;
+	const char * const *bs = b;
+	return strcmp(*as, *bs);
+}
+
+void print_pmu_events(const char *event_glob, bool name_only)
+{
+	struct perf_pmu *pmu;
+	struct perf_pmu_alias *alias;
+	char buf[1024];
+	int printed = 0;
+	int len, j;
+	char **aliases;
+
+	pmu = NULL;
+	len = 0;
+	while ((pmu = perf_pmu__scan(pmu)) != NULL) {
+		list_for_each_entry(alias, &pmu->aliases, list)
+			len++;
+		if (pmu->selectable)
+			len++;
+	}
+	aliases = zalloc(sizeof(char *) * len);
+	if (!aliases)
+		goto out_enomem;
+	pmu = NULL;
+	j = 0;
+	while ((pmu = perf_pmu__scan(pmu)) != NULL) {
+		list_for_each_entry(alias, &pmu->aliases, list) {
+			char *name = format_alias(buf, sizeof(buf), pmu, alias);
+			bool is_cpu = !strcmp(pmu->name, "cpu");
+
+			if (event_glob != NULL &&
+			    !(strglobmatch(name, event_glob) ||
+			      (!is_cpu && strglobmatch(alias->name,
+						       event_glob))))
+				continue;
+
+			if (is_cpu && !name_only)
+				name = format_alias_or(buf, sizeof(buf), pmu, alias);
+
+			aliases[j] = strdup(name);
+			if (aliases[j] == NULL)
+				goto out_enomem;
+			j++;
+		}
+		if (pmu->selectable) {
+			char *s;
+			if (asprintf(&s, "%s//", pmu->name) < 0)
+				goto out_enomem;
+			aliases[j] = s;
+			j++;
+		}
+	}
+	len = j;
+	qsort(aliases, len, sizeof(char *), cmp_string);
+	for (j = 0; j < len; j++) {
+		if (name_only) {
+			printf("%s ", aliases[j]);
+			continue;
+		}
+		printf("  %-50s [Kernel PMU event]\n", aliases[j]);
+		printed++;
+	}
+	if (printed)
+		printf("\n");
+out_free:
+	for (j = 0; j < len; j++)
+		zfree(&aliases[j]);
+	zfree(&aliases);
+	return;
+
+out_enomem:
+	printf("FATAL: not enough memory to print PMU events\n");
+	if (aliases)
+		goto out_free;
+}
+
+bool pmu_have_event(const char *pname, const char *name)
+{
+	struct perf_pmu *pmu;
+	struct perf_pmu_alias *alias;
+
+	pmu = NULL;
+	while ((pmu = perf_pmu__scan(pmu)) != NULL) {
+		if (strcmp(pname, pmu->name))
+			continue;
+		list_for_each_entry(alias, &pmu->aliases, list)
+			if (!strcmp(alias->name, name))
+				return true;
+	}
+	return false;
+}
+
+static FILE *perf_pmu__open_file(struct perf_pmu *pmu, const char *name)
+{
+	struct stat st;
+	char path[PATH_MAX];
+	const char *sysfs;
+
+	sysfs = sysfs__mountpoint();
+	if (!sysfs)
+		return NULL;
+
+	snprintf(path, PATH_MAX,
+		 "%s" EVENT_SOURCE_DEVICE_PATH "%s/%s", sysfs, pmu->name, name);
+
+	if (stat(path, &st) < 0)
+		return NULL;
+
+	return fopen(path, "r");
+}
+
+int perf_pmu__scan_file(struct perf_pmu *pmu, const char *name, const char *fmt,
+			...)
+{
+	va_list args;
+	FILE *file;
+	int ret = EOF;
+
+	va_start(args, fmt);
+	file = perf_pmu__open_file(pmu, name);
+	if (file) {
+		ret = vfscanf(file, fmt, args);
+		fclose(file);
+	}
+	va_end(args);
+	return ret;
+}
diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h
new file mode 100644
index 000000000..6b1249fbd
--- /dev/null
+++ b/tools/perf/util/pmu.h
@@ -0,0 +1,79 @@
+#ifndef __PMU_H
+#define __PMU_H
+
+#include <linux/bitmap.h>
+#include <linux/perf_event.h>
+#include <stdbool.h>
+
+enum {
+	PERF_PMU_FORMAT_VALUE_CONFIG,
+	PERF_PMU_FORMAT_VALUE_CONFIG1,
+	PERF_PMU_FORMAT_VALUE_CONFIG2,
+};
+
+#define PERF_PMU_FORMAT_BITS 64
+
+struct perf_event_attr;
+
+struct perf_pmu {
+	char *name;
+	__u32 type;
+	bool selectable;
+	struct perf_event_attr *default_config;
+	struct cpu_map *cpus;
+	struct list_head format;  /* HEAD struct perf_pmu_format -> list */
+	struct list_head aliases; /* HEAD struct perf_pmu_alias -> list */
+	struct list_head list;    /* ELEM */
+};
+
+struct perf_pmu_info {
+	const char *unit;
+	double scale;
+	bool per_pkg;
+	bool snapshot;
+};
+
+#define UNIT_MAX_LEN	31 /* max length for event unit name */
+
+struct perf_pmu_alias {
+	char *name;
+	struct list_head terms; /* HEAD struct parse_events_term -> list */
+	struct list_head list;  /* ELEM */
+	char unit[UNIT_MAX_LEN+1];
+	double scale;
+	bool per_pkg;
+	bool snapshot;
+};
+
+struct perf_pmu *perf_pmu__find(const char *name);
+int perf_pmu__config(struct perf_pmu *pmu, struct perf_event_attr *attr,
+		     struct list_head *head_terms);
+int perf_pmu__config_terms(struct list_head *formats,
+			   struct perf_event_attr *attr,
+			   struct list_head *head_terms,
+			   bool zero);
+int perf_pmu__check_alias(struct perf_pmu *pmu, struct list_head *head_terms,
+			  struct perf_pmu_info *info);
+struct list_head *perf_pmu__alias(struct perf_pmu *pmu,
+				  struct list_head *head_terms);
+int perf_pmu_wrap(void);
+void perf_pmu_error(struct list_head *list, char *name, char const *msg);
+
+int perf_pmu__new_format(struct list_head *list, char *name,
+			 int config, unsigned long *bits);
+void perf_pmu__set_format(unsigned long *bits, long from, long to);
+int perf_pmu__format_parse(char *dir, struct list_head *head);
+
+struct perf_pmu *perf_pmu__scan(struct perf_pmu *pmu);
+
+void print_pmu_events(const char *event_glob, bool name_only);
+bool pmu_have_event(const char *pname, const char *name);
+
+int perf_pmu__scan_file(struct perf_pmu *pmu, const char *name, const char *fmt,
+			...) __attribute__((format(scanf, 3, 4)));
+
+int perf_pmu__test(void);
+
+struct perf_event_attr *perf_pmu__get_default_config(struct perf_pmu *pmu);
+
+#endif /* __PMU_H */
diff --git a/tools/perf/util/pmu.l b/tools/perf/util/pmu.l
new file mode 100644
index 000000000..a15d9fbd7
--- /dev/null
+++ b/tools/perf/util/pmu.l
@@ -0,0 +1,43 @@
+%option prefix="perf_pmu_"
+
+%{
+#include <stdlib.h>
+#include <linux/bitops.h>
+#include "pmu.h"
+#include "pmu-bison.h"
+
+static int value(int base)
+{
+	long num;
+
+	errno = 0;
+	num = strtoul(perf_pmu_text, NULL, base);
+	if (errno)
+		return PP_ERROR;
+
+	perf_pmu_lval.num = num;
+	return PP_VALUE;
+}
+
+%}
+
+num_dec         [0-9]+
+
+%%
+
+{num_dec}	{ return value(10); }
+config		{ return PP_CONFIG; }
+config1		{ return PP_CONFIG1; }
+config2		{ return PP_CONFIG2; }
+-		{ return '-'; }
+:		{ return ':'; }
+,		{ return ','; }
+.		{ ; }
+\n		{ ; }
+
+%%
+
+int perf_pmu_wrap(void)
+{
+	return 1;
+}
diff --git a/tools/perf/util/pmu.y b/tools/perf/util/pmu.y
new file mode 100644
index 000000000..bfd7e8509
--- /dev/null
+++ b/tools/perf/util/pmu.y
@@ -0,0 +1,92 @@
+
+%parse-param {struct list_head *format}
+%parse-param {char *name}
+
+%{
+
+#include <linux/compiler.h>
+#include <linux/list.h>
+#include <linux/bitmap.h>
+#include <string.h>
+#include "pmu.h"
+
+extern int perf_pmu_lex (void);
+
+#define ABORT_ON(val) \
+do { \
+        if (val) \
+                YYABORT; \
+} while (0)
+
+%}
+
+%token PP_CONFIG PP_CONFIG1 PP_CONFIG2
+%token PP_VALUE PP_ERROR
+%type <num> PP_VALUE
+%type <bits> bit_term
+%type <bits> bits
+
+%union
+{
+	unsigned long num;
+	DECLARE_BITMAP(bits, PERF_PMU_FORMAT_BITS);
+}
+
+%%
+
+format:
+format format_term
+|
+format_term
+
+format_term:
+PP_CONFIG ':' bits
+{
+	ABORT_ON(perf_pmu__new_format(format, name,
+				      PERF_PMU_FORMAT_VALUE_CONFIG,
+				      $3));
+}
+|
+PP_CONFIG1 ':' bits
+{
+	ABORT_ON(perf_pmu__new_format(format, name,
+				      PERF_PMU_FORMAT_VALUE_CONFIG1,
+				      $3));
+}
+|
+PP_CONFIG2 ':' bits
+{
+	ABORT_ON(perf_pmu__new_format(format, name,
+				      PERF_PMU_FORMAT_VALUE_CONFIG2,
+				      $3));
+}
+
+bits:
+bits ',' bit_term
+{
+	bitmap_or($$, $1, $3, 64);
+}
+|
+bit_term
+{
+	memcpy($$, $1, sizeof($1));
+}
+
+bit_term:
+PP_VALUE '-' PP_VALUE
+{
+	perf_pmu__set_format($$, $1, $3);
+}
+|
+PP_VALUE
+{
+	perf_pmu__set_format($$, $1, 0);
+}
+
+%%
+
+void perf_pmu_error(struct list_head *list __maybe_unused,
+		    char *name __maybe_unused,
+		    char const *msg __maybe_unused)
+{
+}
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
new file mode 100644
index 000000000..d05b77cf3
--- /dev/null
+++ b/tools/perf/util/probe-event.c
@@ -0,0 +1,2847 @@
+/*
+ * probe-event.c : perf-probe definition to probe_events format converter
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <sys/utsname.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <limits.h>
+#include <elf.h>
+
+#include "util.h"
+#include "event.h"
+#include "strlist.h"
+#include "debug.h"
+#include "cache.h"
+#include "color.h"
+#include "symbol.h"
+#include "thread.h"
+#include <api/fs/debugfs.h>
+#include <api/fs/tracefs.h>
+#include "trace-event.h"	/* For __maybe_unused */
+#include "probe-event.h"
+#include "probe-finder.h"
+#include "session.h"
+
+#define MAX_CMDLEN 256
+#define PERFPROBE_GROUP "probe"
+
+bool probe_event_dry_run;	/* Dry run flag */
+
+#define semantic_error(msg ...) pr_err("Semantic error :" msg)
+
+/* If there is no space to write, returns -E2BIG. */
+static int e_snprintf(char *str, size_t size, const char *format, ...)
+	__attribute__((format(printf, 3, 4)));
+
+static int e_snprintf(char *str, size_t size, const char *format, ...)
+{
+	int ret;
+	va_list ap;
+	va_start(ap, format);
+	ret = vsnprintf(str, size, format, ap);
+	va_end(ap);
+	if (ret >= (int)size)
+		ret = -E2BIG;
+	return ret;
+}
+
+static char *synthesize_perf_probe_point(struct perf_probe_point *pp);
+static void clear_probe_trace_event(struct probe_trace_event *tev);
+static struct machine *host_machine;
+
+/* Initialize symbol maps and path of vmlinux/modules */
+static int init_symbol_maps(bool user_only)
+{
+	int ret;
+
+	symbol_conf.sort_by_name = true;
+	symbol_conf.allow_aliases = true;
+	ret = symbol__init(NULL);
+	if (ret < 0) {
+		pr_debug("Failed to init symbol map.\n");
+		goto out;
+	}
+
+	if (host_machine || user_only)	/* already initialized */
+		return 0;
+
+	if (symbol_conf.vmlinux_name)
+		pr_debug("Use vmlinux: %s\n", symbol_conf.vmlinux_name);
+
+	host_machine = machine__new_host();
+	if (!host_machine) {
+		pr_debug("machine__new_host() failed.\n");
+		symbol__exit();
+		ret = -1;
+	}
+out:
+	if (ret < 0)
+		pr_warning("Failed to init vmlinux path.\n");
+	return ret;
+}
+
+static void exit_symbol_maps(void)
+{
+	if (host_machine) {
+		machine__delete(host_machine);
+		host_machine = NULL;
+	}
+	symbol__exit();
+}
+
+static struct symbol *__find_kernel_function_by_name(const char *name,
+						     struct map **mapp)
+{
+	return machine__find_kernel_function_by_name(host_machine, name, mapp,
+						     NULL);
+}
+
+static struct symbol *__find_kernel_function(u64 addr, struct map **mapp)
+{
+	return machine__find_kernel_function(host_machine, addr, mapp, NULL);
+}
+
+static struct ref_reloc_sym *kernel_get_ref_reloc_sym(void)
+{
+	/* kmap->ref_reloc_sym should be set if host_machine is initialized */
+	struct kmap *kmap;
+
+	if (map__load(host_machine->vmlinux_maps[MAP__FUNCTION], NULL) < 0)
+		return NULL;
+
+	kmap = map__kmap(host_machine->vmlinux_maps[MAP__FUNCTION]);
+	if (!kmap)
+		return NULL;
+	return kmap->ref_reloc_sym;
+}
+
+static u64 kernel_get_symbol_address_by_name(const char *name, bool reloc)
+{
+	struct ref_reloc_sym *reloc_sym;
+	struct symbol *sym;
+	struct map *map;
+
+	/* ref_reloc_sym is just a label. Need a special fix*/
+	reloc_sym = kernel_get_ref_reloc_sym();
+	if (reloc_sym && strcmp(name, reloc_sym->name) == 0)
+		return (reloc) ? reloc_sym->addr : reloc_sym->unrelocated_addr;
+	else {
+		sym = __find_kernel_function_by_name(name, &map);
+		if (sym)
+			return map->unmap_ip(map, sym->start) -
+				((reloc) ? 0 : map->reloc);
+	}
+	return 0;
+}
+
+static struct map *kernel_get_module_map(const char *module)
+{
+	struct rb_node *nd;
+	struct map_groups *grp = &host_machine->kmaps;
+
+	/* A file path -- this is an offline module */
+	if (module && strchr(module, '/'))
+		return machine__new_module(host_machine, 0, module);
+
+	if (!module)
+		module = "kernel";
+
+	for (nd = rb_first(&grp->maps[MAP__FUNCTION]); nd; nd = rb_next(nd)) {
+		struct map *pos = rb_entry(nd, struct map, rb_node);
+		if (strncmp(pos->dso->short_name + 1, module,
+			    pos->dso->short_name_len - 2) == 0) {
+			return pos;
+		}
+	}
+	return NULL;
+}
+
+static struct map *get_target_map(const char *target, bool user)
+{
+	/* Init maps of given executable or kernel */
+	if (user)
+		return dso__new_map(target);
+	else
+		return kernel_get_module_map(target);
+}
+
+static void put_target_map(struct map *map, bool user)
+{
+	if (map && user) {
+		/* Only the user map needs to be released */
+		dso__delete(map->dso);
+		map__delete(map);
+	}
+}
+
+
+static struct dso *kernel_get_module_dso(const char *module)
+{
+	struct dso *dso;
+	struct map *map;
+	const char *vmlinux_name;
+
+	if (module) {
+		list_for_each_entry(dso, &host_machine->kernel_dsos.head,
+				    node) {
+			if (strncmp(dso->short_name + 1, module,
+				    dso->short_name_len - 2) == 0)
+				goto found;
+		}
+		pr_debug("Failed to find module %s.\n", module);
+		return NULL;
+	}
+
+	map = host_machine->vmlinux_maps[MAP__FUNCTION];
+	dso = map->dso;
+
+	vmlinux_name = symbol_conf.vmlinux_name;
+	if (vmlinux_name) {
+		if (dso__load_vmlinux(dso, map, vmlinux_name, false, NULL) <= 0)
+			return NULL;
+	} else {
+		if (dso__load_vmlinux_path(dso, map, NULL) <= 0) {
+			pr_debug("Failed to load kernel map.\n");
+			return NULL;
+		}
+	}
+found:
+	return dso;
+}
+
+const char *kernel_get_module_path(const char *module)
+{
+	struct dso *dso = kernel_get_module_dso(module);
+	return (dso) ? dso->long_name : NULL;
+}
+
+static int convert_exec_to_group(const char *exec, char **result)
+{
+	char *ptr1, *ptr2, *exec_copy;
+	char buf[64];
+	int ret;
+
+	exec_copy = strdup(exec);
+	if (!exec_copy)
+		return -ENOMEM;
+
+	ptr1 = basename(exec_copy);
+	if (!ptr1) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	ptr2 = strpbrk(ptr1, "-._");
+	if (ptr2)
+		*ptr2 = '\0';
+	ret = e_snprintf(buf, 64, "%s_%s", PERFPROBE_GROUP, ptr1);
+	if (ret < 0)
+		goto out;
+
+	*result = strdup(buf);
+	ret = *result ? 0 : -ENOMEM;
+
+out:
+	free(exec_copy);
+	return ret;
+}
+
+static void clear_perf_probe_point(struct perf_probe_point *pp)
+{
+	free(pp->file);
+	free(pp->function);
+	free(pp->lazy_line);
+}
+
+static void clear_probe_trace_events(struct probe_trace_event *tevs, int ntevs)
+{
+	int i;
+
+	for (i = 0; i < ntevs; i++)
+		clear_probe_trace_event(tevs + i);
+}
+
+#ifdef HAVE_DWARF_SUPPORT
+/*
+ * Some binaries like glibc have special symbols which are on the symbol
+ * table, but not in the debuginfo. If we can find the address of the
+ * symbol from map, we can translate the address back to the probe point.
+ */
+static int find_alternative_probe_point(struct debuginfo *dinfo,
+					struct perf_probe_point *pp,
+					struct perf_probe_point *result,
+					const char *target, bool uprobes)
+{
+	struct map *map = NULL;
+	struct symbol *sym;
+	u64 address = 0;
+	int ret = -ENOENT;
+
+	/* This can work only for function-name based one */
+	if (!pp->function || pp->file)
+		return -ENOTSUP;
+
+	map = get_target_map(target, uprobes);
+	if (!map)
+		return -EINVAL;
+
+	/* Find the address of given function */
+	map__for_each_symbol_by_name(map, pp->function, sym) {
+		if (uprobes)
+			address = sym->start;
+		else
+			address = map->unmap_ip(map, sym->start);
+		break;
+	}
+	if (!address) {
+		ret = -ENOENT;
+		goto out;
+	}
+	pr_debug("Symbol %s address found : %" PRIx64 "\n",
+			pp->function, address);
+
+	ret = debuginfo__find_probe_point(dinfo, (unsigned long)address,
+					  result);
+	if (ret <= 0)
+		ret = (!ret) ? -ENOENT : ret;
+	else {
+		result->offset += pp->offset;
+		result->line += pp->line;
+		result->retprobe = pp->retprobe;
+		ret = 0;
+	}
+
+out:
+	put_target_map(map, uprobes);
+	return ret;
+
+}
+
+static int get_alternative_probe_event(struct debuginfo *dinfo,
+				       struct perf_probe_event *pev,
+				       struct perf_probe_point *tmp,
+				       const char *target)
+{
+	int ret;
+
+	memcpy(tmp, &pev->point, sizeof(*tmp));
+	memset(&pev->point, 0, sizeof(pev->point));
+	ret = find_alternative_probe_point(dinfo, tmp, &pev->point,
+					   target, pev->uprobes);
+	if (ret < 0)
+		memcpy(&pev->point, tmp, sizeof(*tmp));
+
+	return ret;
+}
+
+static int get_alternative_line_range(struct debuginfo *dinfo,
+				      struct line_range *lr,
+				      const char *target, bool user)
+{
+	struct perf_probe_point pp = { .function = lr->function,
+				       .file = lr->file,
+				       .line = lr->start };
+	struct perf_probe_point result;
+	int ret, len = 0;
+
+	memset(&result, 0, sizeof(result));
+
+	if (lr->end != INT_MAX)
+		len = lr->end - lr->start;
+	ret = find_alternative_probe_point(dinfo, &pp, &result,
+					   target, user);
+	if (!ret) {
+		lr->function = result.function;
+		lr->file = result.file;
+		lr->start = result.line;
+		if (lr->end != INT_MAX)
+			lr->end = lr->start + len;
+		clear_perf_probe_point(&pp);
+	}
+	return ret;
+}
+
+/* Open new debuginfo of given module */
+static struct debuginfo *open_debuginfo(const char *module, bool silent)
+{
+	const char *path = module;
+	struct debuginfo *ret;
+
+	if (!module || !strchr(module, '/')) {
+		path = kernel_get_module_path(module);
+		if (!path) {
+			if (!silent)
+				pr_err("Failed to find path of %s module.\n",
+				       module ?: "kernel");
+			return NULL;
+		}
+	}
+	ret = debuginfo__new(path);
+	if (!ret && !silent) {
+		pr_warning("The %s file has no debug information.\n", path);
+		if (!module || !strtailcmp(path, ".ko"))
+			pr_warning("Rebuild with CONFIG_DEBUG_INFO=y, ");
+		else
+			pr_warning("Rebuild with -g, ");
+		pr_warning("or install an appropriate debuginfo package.\n");
+	}
+	return ret;
+}
+
+
+static int get_text_start_address(const char *exec, unsigned long *address)
+{
+	Elf *elf;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+	int fd, ret = -ENOENT;
+
+	fd = open(exec, O_RDONLY);
+	if (fd < 0)
+		return -errno;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL)
+		return -EINVAL;
+
+	if (gelf_getehdr(elf, &ehdr) == NULL)
+		goto out;
+
+	if (!elf_section_by_name(elf, &ehdr, &shdr, ".text", NULL))
+		goto out;
+
+	*address = shdr.sh_addr - shdr.sh_offset;
+	ret = 0;
+out:
+	elf_end(elf);
+	return ret;
+}
+
+/*
+ * Convert trace point to probe point with debuginfo
+ */
+static int find_perf_probe_point_from_dwarf(struct probe_trace_point *tp,
+					    struct perf_probe_point *pp,
+					    bool is_kprobe)
+{
+	struct debuginfo *dinfo = NULL;
+	unsigned long stext = 0;
+	u64 addr = tp->address;
+	int ret = -ENOENT;
+
+	/* convert the address to dwarf address */
+	if (!is_kprobe) {
+		if (!addr) {
+			ret = -EINVAL;
+			goto error;
+		}
+		ret = get_text_start_address(tp->module, &stext);
+		if (ret < 0)
+			goto error;
+		addr += stext;
+	} else {
+		addr = kernel_get_symbol_address_by_name(tp->symbol, false);
+		if (addr == 0)
+			goto error;
+		addr += tp->offset;
+	}
+
+	pr_debug("try to find information at %" PRIx64 " in %s\n", addr,
+		 tp->module ? : "kernel");
+
+	dinfo = open_debuginfo(tp->module, verbose == 0);
+	if (dinfo) {
+		ret = debuginfo__find_probe_point(dinfo,
+						 (unsigned long)addr, pp);
+		debuginfo__delete(dinfo);
+	} else
+		ret = -ENOENT;
+
+	if (ret > 0) {
+		pp->retprobe = tp->retprobe;
+		return 0;
+	}
+error:
+	pr_debug("Failed to find corresponding probes from debuginfo.\n");
+	return ret ? : -ENOENT;
+}
+
+static int add_exec_to_probe_trace_events(struct probe_trace_event *tevs,
+					  int ntevs, const char *exec)
+{
+	int i, ret = 0;
+	unsigned long stext = 0;
+
+	if (!exec)
+		return 0;
+
+	ret = get_text_start_address(exec, &stext);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < ntevs && ret >= 0; i++) {
+		/* point.address is the addres of point.symbol + point.offset */
+		tevs[i].point.address -= stext;
+		tevs[i].point.module = strdup(exec);
+		if (!tevs[i].point.module) {
+			ret = -ENOMEM;
+			break;
+		}
+		tevs[i].uprobes = true;
+	}
+
+	return ret;
+}
+
+static int add_module_to_probe_trace_events(struct probe_trace_event *tevs,
+					    int ntevs, const char *module)
+{
+	int i, ret = 0;
+	char *tmp;
+
+	if (!module)
+		return 0;
+
+	tmp = strrchr(module, '/');
+	if (tmp) {
+		/* This is a module path -- get the module name */
+		module = strdup(tmp + 1);
+		if (!module)
+			return -ENOMEM;
+		tmp = strchr(module, '.');
+		if (tmp)
+			*tmp = '\0';
+		tmp = (char *)module;	/* For free() */
+	}
+
+	for (i = 0; i < ntevs; i++) {
+		tevs[i].point.module = strdup(module);
+		if (!tevs[i].point.module) {
+			ret = -ENOMEM;
+			break;
+		}
+	}
+
+	free(tmp);
+	return ret;
+}
+
+/* Post processing the probe events */
+static int post_process_probe_trace_events(struct probe_trace_event *tevs,
+					   int ntevs, const char *module,
+					   bool uprobe)
+{
+	struct ref_reloc_sym *reloc_sym;
+	char *tmp;
+	int i;
+
+	if (uprobe)
+		return add_exec_to_probe_trace_events(tevs, ntevs, module);
+
+	/* Note that currently ref_reloc_sym based probe is not for drivers */
+	if (module)
+		return add_module_to_probe_trace_events(tevs, ntevs, module);
+
+	reloc_sym = kernel_get_ref_reloc_sym();
+	if (!reloc_sym) {
+		pr_warning("Relocated base symbol is not found!\n");
+		return -EINVAL;
+	}
+
+	for (i = 0; i < ntevs; i++) {
+		if (tevs[i].point.address && !tevs[i].point.retprobe) {
+			tmp = strdup(reloc_sym->name);
+			if (!tmp)
+				return -ENOMEM;
+			free(tevs[i].point.symbol);
+			tevs[i].point.symbol = tmp;
+			tevs[i].point.offset = tevs[i].point.address -
+					       reloc_sym->unrelocated_addr;
+		}
+	}
+	return 0;
+}
+
+/* Try to find perf_probe_event with debuginfo */
+static int try_to_find_probe_trace_events(struct perf_probe_event *pev,
+					  struct probe_trace_event **tevs,
+					  int max_tevs, const char *target)
+{
+	bool need_dwarf = perf_probe_event_need_dwarf(pev);
+	struct perf_probe_point tmp;
+	struct debuginfo *dinfo;
+	int ntevs, ret = 0;
+
+	dinfo = open_debuginfo(target, !need_dwarf);
+
+	if (!dinfo) {
+		if (need_dwarf)
+			return -ENOENT;
+		pr_debug("Could not open debuginfo. Try to use symbols.\n");
+		return 0;
+	}
+
+	pr_debug("Try to find probe point from debuginfo.\n");
+	/* Searching trace events corresponding to a probe event */
+	ntevs = debuginfo__find_trace_events(dinfo, pev, tevs, max_tevs);
+
+	if (ntevs == 0)	{  /* Not found, retry with an alternative */
+		ret = get_alternative_probe_event(dinfo, pev, &tmp, target);
+		if (!ret) {
+			ntevs = debuginfo__find_trace_events(dinfo, pev,
+							     tevs, max_tevs);
+			/*
+			 * Write back to the original probe_event for
+			 * setting appropriate (user given) event name
+			 */
+			clear_perf_probe_point(&pev->point);
+			memcpy(&pev->point, &tmp, sizeof(tmp));
+		}
+	}
+
+	debuginfo__delete(dinfo);
+
+	if (ntevs > 0) {	/* Succeeded to find trace events */
+		pr_debug("Found %d probe_trace_events.\n", ntevs);
+		ret = post_process_probe_trace_events(*tevs, ntevs,
+							target, pev->uprobes);
+		if (ret < 0) {
+			clear_probe_trace_events(*tevs, ntevs);
+			zfree(tevs);
+		}
+		return ret < 0 ? ret : ntevs;
+	}
+
+	if (ntevs == 0)	{	/* No error but failed to find probe point. */
+		pr_warning("Probe point '%s' not found.\n",
+			   synthesize_perf_probe_point(&pev->point));
+		return -ENOENT;
+	}
+	/* Error path : ntevs < 0 */
+	pr_debug("An error occurred in debuginfo analysis (%d).\n", ntevs);
+	if (ntevs == -EBADF) {
+		pr_warning("Warning: No dwarf info found in the vmlinux - "
+			"please rebuild kernel with CONFIG_DEBUG_INFO=y.\n");
+		if (!need_dwarf) {
+			pr_debug("Trying to use symbols.\n");
+			return 0;
+		}
+	}
+	return ntevs;
+}
+
+#define LINEBUF_SIZE 256
+#define NR_ADDITIONAL_LINES 2
+
+static int __show_one_line(FILE *fp, int l, bool skip, bool show_num)
+{
+	char buf[LINEBUF_SIZE], sbuf[STRERR_BUFSIZE];
+	const char *color = show_num ? "" : PERF_COLOR_BLUE;
+	const char *prefix = NULL;
+
+	do {
+		if (fgets(buf, LINEBUF_SIZE, fp) == NULL)
+			goto error;
+		if (skip)
+			continue;
+		if (!prefix) {
+			prefix = show_num ? "%7d  " : "         ";
+			color_fprintf(stdout, color, prefix, l);
+		}
+		color_fprintf(stdout, color, "%s", buf);
+
+	} while (strchr(buf, '\n') == NULL);
+
+	return 1;
+error:
+	if (ferror(fp)) {
+		pr_warning("File read error: %s\n",
+			   strerror_r(errno, sbuf, sizeof(sbuf)));
+		return -1;
+	}
+	return 0;
+}
+
+static int _show_one_line(FILE *fp, int l, bool skip, bool show_num)
+{
+	int rv = __show_one_line(fp, l, skip, show_num);
+	if (rv == 0) {
+		pr_warning("Source file is shorter than expected.\n");
+		rv = -1;
+	}
+	return rv;
+}
+
+#define show_one_line_with_num(f,l)	_show_one_line(f,l,false,true)
+#define show_one_line(f,l)		_show_one_line(f,l,false,false)
+#define skip_one_line(f,l)		_show_one_line(f,l,true,false)
+#define show_one_line_or_eof(f,l)	__show_one_line(f,l,false,false)
+
+/*
+ * Show line-range always requires debuginfo to find source file and
+ * line number.
+ */
+static int __show_line_range(struct line_range *lr, const char *module,
+			     bool user)
+{
+	int l = 1;
+	struct int_node *ln;
+	struct debuginfo *dinfo;
+	FILE *fp;
+	int ret;
+	char *tmp;
+	char sbuf[STRERR_BUFSIZE];
+
+	/* Search a line range */
+	dinfo = open_debuginfo(module, false);
+	if (!dinfo)
+		return -ENOENT;
+
+	ret = debuginfo__find_line_range(dinfo, lr);
+	if (!ret) {	/* Not found, retry with an alternative */
+		ret = get_alternative_line_range(dinfo, lr, module, user);
+		if (!ret)
+			ret = debuginfo__find_line_range(dinfo, lr);
+	}
+	debuginfo__delete(dinfo);
+	if (ret == 0 || ret == -ENOENT) {
+		pr_warning("Specified source line is not found.\n");
+		return -ENOENT;
+	} else if (ret < 0) {
+		pr_warning("Debuginfo analysis failed.\n");
+		return ret;
+	}
+
+	/* Convert source file path */
+	tmp = lr->path;
+	ret = get_real_path(tmp, lr->comp_dir, &lr->path);
+
+	/* Free old path when new path is assigned */
+	if (tmp != lr->path)
+		free(tmp);
+
+	if (ret < 0) {
+		pr_warning("Failed to find source file path.\n");
+		return ret;
+	}
+
+	setup_pager();
+
+	if (lr->function)
+		fprintf(stdout, "<%s@%s:%d>\n", lr->function, lr->path,
+			lr->start - lr->offset);
+	else
+		fprintf(stdout, "<%s:%d>\n", lr->path, lr->start);
+
+	fp = fopen(lr->path, "r");
+	if (fp == NULL) {
+		pr_warning("Failed to open %s: %s\n", lr->path,
+			   strerror_r(errno, sbuf, sizeof(sbuf)));
+		return -errno;
+	}
+	/* Skip to starting line number */
+	while (l < lr->start) {
+		ret = skip_one_line(fp, l++);
+		if (ret < 0)
+			goto end;
+	}
+
+	intlist__for_each(ln, lr->line_list) {
+		for (; ln->i > l; l++) {
+			ret = show_one_line(fp, l - lr->offset);
+			if (ret < 0)
+				goto end;
+		}
+		ret = show_one_line_with_num(fp, l++ - lr->offset);
+		if (ret < 0)
+			goto end;
+	}
+
+	if (lr->end == INT_MAX)
+		lr->end = l + NR_ADDITIONAL_LINES;
+	while (l <= lr->end) {
+		ret = show_one_line_or_eof(fp, l++ - lr->offset);
+		if (ret <= 0)
+			break;
+	}
+end:
+	fclose(fp);
+	return ret;
+}
+
+int show_line_range(struct line_range *lr, const char *module, bool user)
+{
+	int ret;
+
+	ret = init_symbol_maps(user);
+	if (ret < 0)
+		return ret;
+	ret = __show_line_range(lr, module, user);
+	exit_symbol_maps();
+
+	return ret;
+}
+
+static int show_available_vars_at(struct debuginfo *dinfo,
+				  struct perf_probe_event *pev,
+				  int max_vls, struct strfilter *_filter,
+				  bool externs, const char *target)
+{
+	char *buf;
+	int ret, i, nvars;
+	struct str_node *node;
+	struct variable_list *vls = NULL, *vl;
+	struct perf_probe_point tmp;
+	const char *var;
+
+	buf = synthesize_perf_probe_point(&pev->point);
+	if (!buf)
+		return -EINVAL;
+	pr_debug("Searching variables at %s\n", buf);
+
+	ret = debuginfo__find_available_vars_at(dinfo, pev, &vls,
+						max_vls, externs);
+	if (!ret) {  /* Not found, retry with an alternative */
+		ret = get_alternative_probe_event(dinfo, pev, &tmp, target);
+		if (!ret) {
+			ret = debuginfo__find_available_vars_at(dinfo, pev,
+						&vls, max_vls, externs);
+			/* Release the old probe_point */
+			clear_perf_probe_point(&tmp);
+		}
+	}
+	if (ret <= 0) {
+		if (ret == 0 || ret == -ENOENT) {
+			pr_err("Failed to find the address of %s\n", buf);
+			ret = -ENOENT;
+		} else
+			pr_warning("Debuginfo analysis failed.\n");
+		goto end;
+	}
+
+	/* Some variables are found */
+	fprintf(stdout, "Available variables at %s\n", buf);
+	for (i = 0; i < ret; i++) {
+		vl = &vls[i];
+		/*
+		 * A probe point might be converted to
+		 * several trace points.
+		 */
+		fprintf(stdout, "\t@<%s+%lu>\n", vl->point.symbol,
+			vl->point.offset);
+		zfree(&vl->point.symbol);
+		nvars = 0;
+		if (vl->vars) {
+			strlist__for_each(node, vl->vars) {
+				var = strchr(node->s, '\t') + 1;
+				if (strfilter__compare(_filter, var)) {
+					fprintf(stdout, "\t\t%s\n", node->s);
+					nvars++;
+				}
+			}
+			strlist__delete(vl->vars);
+		}
+		if (nvars == 0)
+			fprintf(stdout, "\t\t(No matched variables)\n");
+	}
+	free(vls);
+end:
+	free(buf);
+	return ret;
+}
+
+/* Show available variables on given probe point */
+int show_available_vars(struct perf_probe_event *pevs, int npevs,
+			int max_vls, const char *module,
+			struct strfilter *_filter, bool externs)
+{
+	int i, ret = 0;
+	struct debuginfo *dinfo;
+
+	ret = init_symbol_maps(pevs->uprobes);
+	if (ret < 0)
+		return ret;
+
+	dinfo = open_debuginfo(module, false);
+	if (!dinfo) {
+		ret = -ENOENT;
+		goto out;
+	}
+
+	setup_pager();
+
+	for (i = 0; i < npevs && ret >= 0; i++)
+		ret = show_available_vars_at(dinfo, &pevs[i], max_vls, _filter,
+					     externs, module);
+
+	debuginfo__delete(dinfo);
+out:
+	exit_symbol_maps();
+	return ret;
+}
+
+#else	/* !HAVE_DWARF_SUPPORT */
+
+static int
+find_perf_probe_point_from_dwarf(struct probe_trace_point *tp __maybe_unused,
+				 struct perf_probe_point *pp __maybe_unused,
+				 bool is_kprobe __maybe_unused)
+{
+	return -ENOSYS;
+}
+
+static int try_to_find_probe_trace_events(struct perf_probe_event *pev,
+				struct probe_trace_event **tevs __maybe_unused,
+				int max_tevs __maybe_unused,
+				const char *target __maybe_unused)
+{
+	if (perf_probe_event_need_dwarf(pev)) {
+		pr_warning("Debuginfo-analysis is not supported.\n");
+		return -ENOSYS;
+	}
+
+	return 0;
+}
+
+int show_line_range(struct line_range *lr __maybe_unused,
+		    const char *module __maybe_unused,
+		    bool user __maybe_unused)
+{
+	pr_warning("Debuginfo-analysis is not supported.\n");
+	return -ENOSYS;
+}
+
+int show_available_vars(struct perf_probe_event *pevs __maybe_unused,
+			int npevs __maybe_unused, int max_vls __maybe_unused,
+			const char *module __maybe_unused,
+			struct strfilter *filter __maybe_unused,
+			bool externs __maybe_unused)
+{
+	pr_warning("Debuginfo-analysis is not supported.\n");
+	return -ENOSYS;
+}
+#endif
+
+void line_range__clear(struct line_range *lr)
+{
+	free(lr->function);
+	free(lr->file);
+	free(lr->path);
+	free(lr->comp_dir);
+	intlist__delete(lr->line_list);
+	memset(lr, 0, sizeof(*lr));
+}
+
+int line_range__init(struct line_range *lr)
+{
+	memset(lr, 0, sizeof(*lr));
+	lr->line_list = intlist__new(NULL);
+	if (!lr->line_list)
+		return -ENOMEM;
+	else
+		return 0;
+}
+
+static int parse_line_num(char **ptr, int *val, const char *what)
+{
+	const char *start = *ptr;
+
+	errno = 0;
+	*val = strtol(*ptr, ptr, 0);
+	if (errno || *ptr == start) {
+		semantic_error("'%s' is not a valid number.\n", what);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/*
+ * Stuff 'lr' according to the line range described by 'arg'.
+ * The line range syntax is described by:
+ *
+ *         SRC[:SLN[+NUM|-ELN]]
+ *         FNC[@SRC][:SLN[+NUM|-ELN]]
+ */
+int parse_line_range_desc(const char *arg, struct line_range *lr)
+{
+	char *range, *file, *name = strdup(arg);
+	int err;
+
+	if (!name)
+		return -ENOMEM;
+
+	lr->start = 0;
+	lr->end = INT_MAX;
+
+	range = strchr(name, ':');
+	if (range) {
+		*range++ = '\0';
+
+		err = parse_line_num(&range, &lr->start, "start line");
+		if (err)
+			goto err;
+
+		if (*range == '+' || *range == '-') {
+			const char c = *range++;
+
+			err = parse_line_num(&range, &lr->end, "end line");
+			if (err)
+				goto err;
+
+			if (c == '+') {
+				lr->end += lr->start;
+				/*
+				 * Adjust the number of lines here.
+				 * If the number of lines == 1, the
+				 * the end of line should be equal to
+				 * the start of line.
+				 */
+				lr->end--;
+			}
+		}
+
+		pr_debug("Line range is %d to %d\n", lr->start, lr->end);
+
+		err = -EINVAL;
+		if (lr->start > lr->end) {
+			semantic_error("Start line must be smaller"
+				       " than end line.\n");
+			goto err;
+		}
+		if (*range != '\0') {
+			semantic_error("Tailing with invalid str '%s'.\n", range);
+			goto err;
+		}
+	}
+
+	file = strchr(name, '@');
+	if (file) {
+		*file = '\0';
+		lr->file = strdup(++file);
+		if (lr->file == NULL) {
+			err = -ENOMEM;
+			goto err;
+		}
+		lr->function = name;
+	} else if (strchr(name, '.'))
+		lr->file = name;
+	else
+		lr->function = name;
+
+	return 0;
+err:
+	free(name);
+	return err;
+}
+
+/* Check the name is good for event/group */
+static bool check_event_name(const char *name)
+{
+	if (!isalpha(*name) && *name != '_')
+		return false;
+	while (*++name != '\0') {
+		if (!isalpha(*name) && !isdigit(*name) && *name != '_')
+			return false;
+	}
+	return true;
+}
+
+/* Parse probepoint definition. */
+static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev)
+{
+	struct perf_probe_point *pp = &pev->point;
+	char *ptr, *tmp;
+	char c, nc = 0;
+	/*
+	 * <Syntax>
+	 * perf probe [EVENT=]SRC[:LN|;PTN]
+	 * perf probe [EVENT=]FUNC[@SRC][+OFFS|%return|:LN|;PAT]
+	 *
+	 * TODO:Group name support
+	 */
+	if (!arg)
+		return -EINVAL;
+
+	ptr = strpbrk(arg, ";=@+%");
+	if (ptr && *ptr == '=') {	/* Event name */
+		*ptr = '\0';
+		tmp = ptr + 1;
+		if (strchr(arg, ':')) {
+			semantic_error("Group name is not supported yet.\n");
+			return -ENOTSUP;
+		}
+		if (!check_event_name(arg)) {
+			semantic_error("%s is bad for event name -it must "
+				       "follow C symbol-naming rule.\n", arg);
+			return -EINVAL;
+		}
+		pev->event = strdup(arg);
+		if (pev->event == NULL)
+			return -ENOMEM;
+		pev->group = NULL;
+		arg = tmp;
+	}
+
+	ptr = strpbrk(arg, ";:+@%");
+	if (ptr) {
+		nc = *ptr;
+		*ptr++ = '\0';
+	}
+
+	tmp = strdup(arg);
+	if (tmp == NULL)
+		return -ENOMEM;
+
+	/* Check arg is function or file and copy it */
+	if (strchr(tmp, '.'))	/* File */
+		pp->file = tmp;
+	else			/* Function */
+		pp->function = tmp;
+
+	/* Parse other options */
+	while (ptr) {
+		arg = ptr;
+		c = nc;
+		if (c == ';') {	/* Lazy pattern must be the last part */
+			pp->lazy_line = strdup(arg);
+			if (pp->lazy_line == NULL)
+				return -ENOMEM;
+			break;
+		}
+		ptr = strpbrk(arg, ";:+@%");
+		if (ptr) {
+			nc = *ptr;
+			*ptr++ = '\0';
+		}
+		switch (c) {
+		case ':':	/* Line number */
+			pp->line = strtoul(arg, &tmp, 0);
+			if (*tmp != '\0') {
+				semantic_error("There is non-digit char"
+					       " in line number.\n");
+				return -EINVAL;
+			}
+			break;
+		case '+':	/* Byte offset from a symbol */
+			pp->offset = strtoul(arg, &tmp, 0);
+			if (*tmp != '\0') {
+				semantic_error("There is non-digit character"
+						" in offset.\n");
+				return -EINVAL;
+			}
+			break;
+		case '@':	/* File name */
+			if (pp->file) {
+				semantic_error("SRC@SRC is not allowed.\n");
+				return -EINVAL;
+			}
+			pp->file = strdup(arg);
+			if (pp->file == NULL)
+				return -ENOMEM;
+			break;
+		case '%':	/* Probe places */
+			if (strcmp(arg, "return") == 0) {
+				pp->retprobe = 1;
+			} else {	/* Others not supported yet */
+				semantic_error("%%%s is not supported.\n", arg);
+				return -ENOTSUP;
+			}
+			break;
+		default:	/* Buggy case */
+			pr_err("This program has a bug at %s:%d.\n",
+				__FILE__, __LINE__);
+			return -ENOTSUP;
+			break;
+		}
+	}
+
+	/* Exclusion check */
+	if (pp->lazy_line && pp->line) {
+		semantic_error("Lazy pattern can't be used with"
+			       " line number.\n");
+		return -EINVAL;
+	}
+
+	if (pp->lazy_line && pp->offset) {
+		semantic_error("Lazy pattern can't be used with offset.\n");
+		return -EINVAL;
+	}
+
+	if (pp->line && pp->offset) {
+		semantic_error("Offset can't be used with line number.\n");
+		return -EINVAL;
+	}
+
+	if (!pp->line && !pp->lazy_line && pp->file && !pp->function) {
+		semantic_error("File always requires line number or "
+			       "lazy pattern.\n");
+		return -EINVAL;
+	}
+
+	if (pp->offset && !pp->function) {
+		semantic_error("Offset requires an entry function.\n");
+		return -EINVAL;
+	}
+
+	if (pp->retprobe && !pp->function) {
+		semantic_error("Return probe requires an entry function.\n");
+		return -EINVAL;
+	}
+
+	if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe) {
+		semantic_error("Offset/Line/Lazy pattern can't be used with "
+			       "return probe.\n");
+		return -EINVAL;
+	}
+
+	pr_debug("symbol:%s file:%s line:%d offset:%lu return:%d lazy:%s\n",
+		 pp->function, pp->file, pp->line, pp->offset, pp->retprobe,
+		 pp->lazy_line);
+	return 0;
+}
+
+/* Parse perf-probe event argument */
+static int parse_perf_probe_arg(char *str, struct perf_probe_arg *arg)
+{
+	char *tmp, *goodname;
+	struct perf_probe_arg_field **fieldp;
+
+	pr_debug("parsing arg: %s into ", str);
+
+	tmp = strchr(str, '=');
+	if (tmp) {
+		arg->name = strndup(str, tmp - str);
+		if (arg->name == NULL)
+			return -ENOMEM;
+		pr_debug("name:%s ", arg->name);
+		str = tmp + 1;
+	}
+
+	tmp = strchr(str, ':');
+	if (tmp) {	/* Type setting */
+		*tmp = '\0';
+		arg->type = strdup(tmp + 1);
+		if (arg->type == NULL)
+			return -ENOMEM;
+		pr_debug("type:%s ", arg->type);
+	}
+
+	tmp = strpbrk(str, "-.[");
+	if (!is_c_varname(str) || !tmp) {
+		/* A variable, register, symbol or special value */
+		arg->var = strdup(str);
+		if (arg->var == NULL)
+			return -ENOMEM;
+		pr_debug("%s\n", arg->var);
+		return 0;
+	}
+
+	/* Structure fields or array element */
+	arg->var = strndup(str, tmp - str);
+	if (arg->var == NULL)
+		return -ENOMEM;
+	goodname = arg->var;
+	pr_debug("%s, ", arg->var);
+	fieldp = &arg->field;
+
+	do {
+		*fieldp = zalloc(sizeof(struct perf_probe_arg_field));
+		if (*fieldp == NULL)
+			return -ENOMEM;
+		if (*tmp == '[') {	/* Array */
+			str = tmp;
+			(*fieldp)->index = strtol(str + 1, &tmp, 0);
+			(*fieldp)->ref = true;
+			if (*tmp != ']' || tmp == str + 1) {
+				semantic_error("Array index must be a"
+						" number.\n");
+				return -EINVAL;
+			}
+			tmp++;
+			if (*tmp == '\0')
+				tmp = NULL;
+		} else {		/* Structure */
+			if (*tmp == '.') {
+				str = tmp + 1;
+				(*fieldp)->ref = false;
+			} else if (tmp[1] == '>') {
+				str = tmp + 2;
+				(*fieldp)->ref = true;
+			} else {
+				semantic_error("Argument parse error: %s\n",
+					       str);
+				return -EINVAL;
+			}
+			tmp = strpbrk(str, "-.[");
+		}
+		if (tmp) {
+			(*fieldp)->name = strndup(str, tmp - str);
+			if ((*fieldp)->name == NULL)
+				return -ENOMEM;
+			if (*str != '[')
+				goodname = (*fieldp)->name;
+			pr_debug("%s(%d), ", (*fieldp)->name, (*fieldp)->ref);
+			fieldp = &(*fieldp)->next;
+		}
+	} while (tmp);
+	(*fieldp)->name = strdup(str);
+	if ((*fieldp)->name == NULL)
+		return -ENOMEM;
+	if (*str != '[')
+		goodname = (*fieldp)->name;
+	pr_debug("%s(%d)\n", (*fieldp)->name, (*fieldp)->ref);
+
+	/* If no name is specified, set the last field name (not array index)*/
+	if (!arg->name) {
+		arg->name = strdup(goodname);
+		if (arg->name == NULL)
+			return -ENOMEM;
+	}
+	return 0;
+}
+
+/* Parse perf-probe event command */
+int parse_perf_probe_command(const char *cmd, struct perf_probe_event *pev)
+{
+	char **argv;
+	int argc, i, ret = 0;
+
+	argv = argv_split(cmd, &argc);
+	if (!argv) {
+		pr_debug("Failed to split arguments.\n");
+		return -ENOMEM;
+	}
+	if (argc - 1 > MAX_PROBE_ARGS) {
+		semantic_error("Too many probe arguments (%d).\n", argc - 1);
+		ret = -ERANGE;
+		goto out;
+	}
+	/* Parse probe point */
+	ret = parse_perf_probe_point(argv[0], pev);
+	if (ret < 0)
+		goto out;
+
+	/* Copy arguments and ensure return probe has no C argument */
+	pev->nargs = argc - 1;
+	pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs);
+	if (pev->args == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	for (i = 0; i < pev->nargs && ret >= 0; i++) {
+		ret = parse_perf_probe_arg(argv[i + 1], &pev->args[i]);
+		if (ret >= 0 &&
+		    is_c_varname(pev->args[i].var) && pev->point.retprobe) {
+			semantic_error("You can't specify local variable for"
+				       " kretprobe.\n");
+			ret = -EINVAL;
+		}
+	}
+out:
+	argv_free(argv);
+
+	return ret;
+}
+
+/* Return true if this perf_probe_event requires debuginfo */
+bool perf_probe_event_need_dwarf(struct perf_probe_event *pev)
+{
+	int i;
+
+	if (pev->point.file || pev->point.line || pev->point.lazy_line)
+		return true;
+
+	for (i = 0; i < pev->nargs; i++)
+		if (is_c_varname(pev->args[i].var))
+			return true;
+
+	return false;
+}
+
+/* Parse probe_events event into struct probe_point */
+static int parse_probe_trace_command(const char *cmd,
+				     struct probe_trace_event *tev)
+{
+	struct probe_trace_point *tp = &tev->point;
+	char pr;
+	char *p;
+	char *argv0_str = NULL, *fmt, *fmt1_str, *fmt2_str, *fmt3_str;
+	int ret, i, argc;
+	char **argv;
+
+	pr_debug("Parsing probe_events: %s\n", cmd);
+	argv = argv_split(cmd, &argc);
+	if (!argv) {
+		pr_debug("Failed to split arguments.\n");
+		return -ENOMEM;
+	}
+	if (argc < 2) {
+		semantic_error("Too few probe arguments.\n");
+		ret = -ERANGE;
+		goto out;
+	}
+
+	/* Scan event and group name. */
+	argv0_str = strdup(argv[0]);
+	if (argv0_str == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	fmt1_str = strtok_r(argv0_str, ":", &fmt);
+	fmt2_str = strtok_r(NULL, "/", &fmt);
+	fmt3_str = strtok_r(NULL, " \t", &fmt);
+	if (fmt1_str == NULL || strlen(fmt1_str) != 1 || fmt2_str == NULL
+	    || fmt3_str == NULL) {
+		semantic_error("Failed to parse event name: %s\n", argv[0]);
+		ret = -EINVAL;
+		goto out;
+	}
+	pr = fmt1_str[0];
+	tev->group = strdup(fmt2_str);
+	tev->event = strdup(fmt3_str);
+	if (tev->group == NULL || tev->event == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	pr_debug("Group:%s Event:%s probe:%c\n", tev->group, tev->event, pr);
+
+	tp->retprobe = (pr == 'r');
+
+	/* Scan module name(if there), function name and offset */
+	p = strchr(argv[1], ':');
+	if (p) {
+		tp->module = strndup(argv[1], p - argv[1]);
+		p++;
+	} else
+		p = argv[1];
+	fmt1_str = strtok_r(p, "+", &fmt);
+	if (fmt1_str[0] == '0')	/* only the address started with 0x */
+		tp->address = strtoul(fmt1_str, NULL, 0);
+	else {
+		/* Only the symbol-based probe has offset */
+		tp->symbol = strdup(fmt1_str);
+		if (tp->symbol == NULL) {
+			ret = -ENOMEM;
+			goto out;
+		}
+		fmt2_str = strtok_r(NULL, "", &fmt);
+		if (fmt2_str == NULL)
+			tp->offset = 0;
+		else
+			tp->offset = strtoul(fmt2_str, NULL, 10);
+	}
+
+	tev->nargs = argc - 2;
+	tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs);
+	if (tev->args == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	for (i = 0; i < tev->nargs; i++) {
+		p = strchr(argv[i + 2], '=');
+		if (p)	/* We don't need which register is assigned. */
+			*p++ = '\0';
+		else
+			p = argv[i + 2];
+		tev->args[i].name = strdup(argv[i + 2]);
+		/* TODO: parse regs and offset */
+		tev->args[i].value = strdup(p);
+		if (tev->args[i].name == NULL || tev->args[i].value == NULL) {
+			ret = -ENOMEM;
+			goto out;
+		}
+	}
+	ret = 0;
+out:
+	free(argv0_str);
+	argv_free(argv);
+	return ret;
+}
+
+/* Compose only probe arg */
+int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf, size_t len)
+{
+	struct perf_probe_arg_field *field = pa->field;
+	int ret;
+	char *tmp = buf;
+
+	if (pa->name && pa->var)
+		ret = e_snprintf(tmp, len, "%s=%s", pa->name, pa->var);
+	else
+		ret = e_snprintf(tmp, len, "%s", pa->name ? pa->name : pa->var);
+	if (ret <= 0)
+		goto error;
+	tmp += ret;
+	len -= ret;
+
+	while (field) {
+		if (field->name[0] == '[')
+			ret = e_snprintf(tmp, len, "%s", field->name);
+		else
+			ret = e_snprintf(tmp, len, "%s%s",
+					 field->ref ? "->" : ".", field->name);
+		if (ret <= 0)
+			goto error;
+		tmp += ret;
+		len -= ret;
+		field = field->next;
+	}
+
+	if (pa->type) {
+		ret = e_snprintf(tmp, len, ":%s", pa->type);
+		if (ret <= 0)
+			goto error;
+		tmp += ret;
+		len -= ret;
+	}
+
+	return tmp - buf;
+error:
+	pr_debug("Failed to synthesize perf probe argument: %d\n", ret);
+	return ret;
+}
+
+/* Compose only probe point (not argument) */
+static char *synthesize_perf_probe_point(struct perf_probe_point *pp)
+{
+	char *buf, *tmp;
+	char offs[32] = "", line[32] = "", file[32] = "";
+	int ret, len;
+
+	buf = zalloc(MAX_CMDLEN);
+	if (buf == NULL) {
+		ret = -ENOMEM;
+		goto error;
+	}
+	if (pp->offset) {
+		ret = e_snprintf(offs, 32, "+%lu", pp->offset);
+		if (ret <= 0)
+			goto error;
+	}
+	if (pp->line) {
+		ret = e_snprintf(line, 32, ":%d", pp->line);
+		if (ret <= 0)
+			goto error;
+	}
+	if (pp->file) {
+		tmp = pp->file;
+		len = strlen(tmp);
+		if (len > 30) {
+			tmp = strchr(pp->file + len - 30, '/');
+			tmp = tmp ? tmp + 1 : pp->file + len - 30;
+		}
+		ret = e_snprintf(file, 32, "@%s", tmp);
+		if (ret <= 0)
+			goto error;
+	}
+
+	if (pp->function)
+		ret = e_snprintf(buf, MAX_CMDLEN, "%s%s%s%s%s", pp->function,
+				 offs, pp->retprobe ? "%return" : "", line,
+				 file);
+	else
+		ret = e_snprintf(buf, MAX_CMDLEN, "%s%s", file, line);
+	if (ret <= 0)
+		goto error;
+
+	return buf;
+error:
+	pr_debug("Failed to synthesize perf probe point: %d\n", ret);
+	free(buf);
+	return NULL;
+}
+
+#if 0
+char *synthesize_perf_probe_command(struct perf_probe_event *pev)
+{
+	char *buf;
+	int i, len, ret;
+
+	buf = synthesize_perf_probe_point(&pev->point);
+	if (!buf)
+		return NULL;
+
+	len = strlen(buf);
+	for (i = 0; i < pev->nargs; i++) {
+		ret = e_snprintf(&buf[len], MAX_CMDLEN - len, " %s",
+				 pev->args[i].name);
+		if (ret <= 0) {
+			free(buf);
+			return NULL;
+		}
+		len += ret;
+	}
+
+	return buf;
+}
+#endif
+
+static int __synthesize_probe_trace_arg_ref(struct probe_trace_arg_ref *ref,
+					     char **buf, size_t *buflen,
+					     int depth)
+{
+	int ret;
+	if (ref->next) {
+		depth = __synthesize_probe_trace_arg_ref(ref->next, buf,
+							 buflen, depth + 1);
+		if (depth < 0)
+			goto out;
+	}
+
+	ret = e_snprintf(*buf, *buflen, "%+ld(", ref->offset);
+	if (ret < 0)
+		depth = ret;
+	else {
+		*buf += ret;
+		*buflen -= ret;
+	}
+out:
+	return depth;
+
+}
+
+static int synthesize_probe_trace_arg(struct probe_trace_arg *arg,
+				       char *buf, size_t buflen)
+{
+	struct probe_trace_arg_ref *ref = arg->ref;
+	int ret, depth = 0;
+	char *tmp = buf;
+
+	/* Argument name or separator */
+	if (arg->name)
+		ret = e_snprintf(buf, buflen, " %s=", arg->name);
+	else
+		ret = e_snprintf(buf, buflen, " ");
+	if (ret < 0)
+		return ret;
+	buf += ret;
+	buflen -= ret;
+
+	/* Special case: @XXX */
+	if (arg->value[0] == '@' && arg->ref)
+			ref = ref->next;
+
+	/* Dereferencing arguments */
+	if (ref) {
+		depth = __synthesize_probe_trace_arg_ref(ref, &buf,
+							  &buflen, 1);
+		if (depth < 0)
+			return depth;
+	}
+
+	/* Print argument value */
+	if (arg->value[0] == '@' && arg->ref)
+		ret = e_snprintf(buf, buflen, "%s%+ld", arg->value,
+				 arg->ref->offset);
+	else
+		ret = e_snprintf(buf, buflen, "%s", arg->value);
+	if (ret < 0)
+		return ret;
+	buf += ret;
+	buflen -= ret;
+
+	/* Closing */
+	while (depth--) {
+		ret = e_snprintf(buf, buflen, ")");
+		if (ret < 0)
+			return ret;
+		buf += ret;
+		buflen -= ret;
+	}
+	/* Print argument type */
+	if (arg->type) {
+		ret = e_snprintf(buf, buflen, ":%s", arg->type);
+		if (ret <= 0)
+			return ret;
+		buf += ret;
+	}
+
+	return buf - tmp;
+}
+
+char *synthesize_probe_trace_command(struct probe_trace_event *tev)
+{
+	struct probe_trace_point *tp = &tev->point;
+	char *buf;
+	int i, len, ret;
+
+	buf = zalloc(MAX_CMDLEN);
+	if (buf == NULL)
+		return NULL;
+
+	len = e_snprintf(buf, MAX_CMDLEN, "%c:%s/%s ", tp->retprobe ? 'r' : 'p',
+			 tev->group, tev->event);
+	if (len <= 0)
+		goto error;
+
+	/* Uprobes must have tp->address and tp->module */
+	if (tev->uprobes && (!tp->address || !tp->module))
+		goto error;
+
+	/* Use the tp->address for uprobes */
+	if (tev->uprobes)
+		ret = e_snprintf(buf + len, MAX_CMDLEN - len, "%s:0x%lx",
+				 tp->module, tp->address);
+	else
+		ret = e_snprintf(buf + len, MAX_CMDLEN - len, "%s%s%s+%lu",
+				 tp->module ?: "", tp->module ? ":" : "",
+				 tp->symbol, tp->offset);
+
+	if (ret <= 0)
+		goto error;
+	len += ret;
+
+	for (i = 0; i < tev->nargs; i++) {
+		ret = synthesize_probe_trace_arg(&tev->args[i], buf + len,
+						  MAX_CMDLEN - len);
+		if (ret <= 0)
+			goto error;
+		len += ret;
+	}
+
+	return buf;
+error:
+	free(buf);
+	return NULL;
+}
+
+static int find_perf_probe_point_from_map(struct probe_trace_point *tp,
+					  struct perf_probe_point *pp,
+					  bool is_kprobe)
+{
+	struct symbol *sym = NULL;
+	struct map *map;
+	u64 addr;
+	int ret = -ENOENT;
+
+	if (!is_kprobe) {
+		map = dso__new_map(tp->module);
+		if (!map)
+			goto out;
+		addr = tp->address;
+		sym = map__find_symbol(map, addr, NULL);
+	} else {
+		addr = kernel_get_symbol_address_by_name(tp->symbol, true);
+		if (addr) {
+			addr += tp->offset;
+			sym = __find_kernel_function(addr, &map);
+		}
+	}
+	if (!sym)
+		goto out;
+
+	pp->retprobe = tp->retprobe;
+	pp->offset = addr - map->unmap_ip(map, sym->start);
+	pp->function = strdup(sym->name);
+	ret = pp->function ? 0 : -ENOMEM;
+
+out:
+	if (map && !is_kprobe) {
+		dso__delete(map->dso);
+		map__delete(map);
+	}
+
+	return ret;
+}
+
+static int convert_to_perf_probe_point(struct probe_trace_point *tp,
+					struct perf_probe_point *pp,
+					bool is_kprobe)
+{
+	char buf[128];
+	int ret;
+
+	ret = find_perf_probe_point_from_dwarf(tp, pp, is_kprobe);
+	if (!ret)
+		return 0;
+	ret = find_perf_probe_point_from_map(tp, pp, is_kprobe);
+	if (!ret)
+		return 0;
+
+	pr_debug("Failed to find probe point from both of dwarf and map.\n");
+
+	if (tp->symbol) {
+		pp->function = strdup(tp->symbol);
+		pp->offset = tp->offset;
+	} else if (!tp->module && !is_kprobe) {
+		ret = e_snprintf(buf, 128, "0x%" PRIx64, (u64)tp->address);
+		if (ret < 0)
+			return ret;
+		pp->function = strdup(buf);
+		pp->offset = 0;
+	}
+	if (pp->function == NULL)
+		return -ENOMEM;
+
+	pp->retprobe = tp->retprobe;
+
+	return 0;
+}
+
+static int convert_to_perf_probe_event(struct probe_trace_event *tev,
+			       struct perf_probe_event *pev, bool is_kprobe)
+{
+	char buf[64] = "";
+	int i, ret;
+
+	/* Convert event/group name */
+	pev->event = strdup(tev->event);
+	pev->group = strdup(tev->group);
+	if (pev->event == NULL || pev->group == NULL)
+		return -ENOMEM;
+
+	/* Convert trace_point to probe_point */
+	ret = convert_to_perf_probe_point(&tev->point, &pev->point, is_kprobe);
+	if (ret < 0)
+		return ret;
+
+	/* Convert trace_arg to probe_arg */
+	pev->nargs = tev->nargs;
+	pev->args = zalloc(sizeof(struct perf_probe_arg) * pev->nargs);
+	if (pev->args == NULL)
+		return -ENOMEM;
+	for (i = 0; i < tev->nargs && ret >= 0; i++) {
+		if (tev->args[i].name)
+			pev->args[i].name = strdup(tev->args[i].name);
+		else {
+			ret = synthesize_probe_trace_arg(&tev->args[i],
+							  buf, 64);
+			pev->args[i].name = strdup(buf);
+		}
+		if (pev->args[i].name == NULL && ret >= 0)
+			ret = -ENOMEM;
+	}
+
+	if (ret < 0)
+		clear_perf_probe_event(pev);
+
+	return ret;
+}
+
+void clear_perf_probe_event(struct perf_probe_event *pev)
+{
+	struct perf_probe_arg_field *field, *next;
+	int i;
+
+	free(pev->event);
+	free(pev->group);
+	free(pev->target);
+	clear_perf_probe_point(&pev->point);
+
+	for (i = 0; i < pev->nargs; i++) {
+		free(pev->args[i].name);
+		free(pev->args[i].var);
+		free(pev->args[i].type);
+		field = pev->args[i].field;
+		while (field) {
+			next = field->next;
+			zfree(&field->name);
+			free(field);
+			field = next;
+		}
+	}
+	free(pev->args);
+	memset(pev, 0, sizeof(*pev));
+}
+
+static void clear_probe_trace_event(struct probe_trace_event *tev)
+{
+	struct probe_trace_arg_ref *ref, *next;
+	int i;
+
+	free(tev->event);
+	free(tev->group);
+	free(tev->point.symbol);
+	free(tev->point.module);
+	for (i = 0; i < tev->nargs; i++) {
+		free(tev->args[i].name);
+		free(tev->args[i].value);
+		free(tev->args[i].type);
+		ref = tev->args[i].ref;
+		while (ref) {
+			next = ref->next;
+			free(ref);
+			ref = next;
+		}
+	}
+	free(tev->args);
+	memset(tev, 0, sizeof(*tev));
+}
+
+static void print_open_warning(int err, bool is_kprobe)
+{
+	char sbuf[STRERR_BUFSIZE];
+
+	if (err == -ENOENT) {
+		const char *config;
+
+		if (!is_kprobe)
+			config = "CONFIG_UPROBE_EVENTS";
+		else
+			config = "CONFIG_KPROBE_EVENTS";
+
+		pr_warning("%cprobe_events file does not exist"
+			   " - please rebuild kernel with %s.\n",
+			   is_kprobe ? 'k' : 'u', config);
+	} else if (err == -ENOTSUP)
+		pr_warning("Tracefs or debugfs is not mounted.\n");
+	else
+		pr_warning("Failed to open %cprobe_events: %s\n",
+			   is_kprobe ? 'k' : 'u',
+			   strerror_r(-err, sbuf, sizeof(sbuf)));
+}
+
+static void print_both_open_warning(int kerr, int uerr)
+{
+	/* Both kprobes and uprobes are disabled, warn it. */
+	if (kerr == -ENOTSUP && uerr == -ENOTSUP)
+		pr_warning("Tracefs or debugfs is not mounted.\n");
+	else if (kerr == -ENOENT && uerr == -ENOENT)
+		pr_warning("Please rebuild kernel with CONFIG_KPROBE_EVENTS "
+			   "or/and CONFIG_UPROBE_EVENTS.\n");
+	else {
+		char sbuf[STRERR_BUFSIZE];
+		pr_warning("Failed to open kprobe events: %s.\n",
+			   strerror_r(-kerr, sbuf, sizeof(sbuf)));
+		pr_warning("Failed to open uprobe events: %s.\n",
+			   strerror_r(-uerr, sbuf, sizeof(sbuf)));
+	}
+}
+
+static int open_probe_events(const char *trace_file, bool readwrite)
+{
+	char buf[PATH_MAX];
+	const char *__debugfs;
+	const char *tracing_dir = "";
+	int ret;
+
+	__debugfs = tracefs_find_mountpoint();
+	if (__debugfs == NULL) {
+		tracing_dir = "tracing/";
+
+		__debugfs = debugfs_find_mountpoint();
+		if (__debugfs == NULL)
+			return -ENOTSUP;
+	}
+
+	ret = e_snprintf(buf, PATH_MAX, "%s/%s%s",
+			 __debugfs, tracing_dir, trace_file);
+	if (ret >= 0) {
+		pr_debug("Opening %s write=%d\n", buf, readwrite);
+		if (readwrite && !probe_event_dry_run)
+			ret = open(buf, O_RDWR, O_APPEND);
+		else
+			ret = open(buf, O_RDONLY, 0);
+
+		if (ret < 0)
+			ret = -errno;
+	}
+	return ret;
+}
+
+static int open_kprobe_events(bool readwrite)
+{
+	return open_probe_events("kprobe_events", readwrite);
+}
+
+static int open_uprobe_events(bool readwrite)
+{
+	return open_probe_events("uprobe_events", readwrite);
+}
+
+/* Get raw string list of current kprobe_events  or uprobe_events */
+static struct strlist *get_probe_trace_command_rawlist(int fd)
+{
+	int ret, idx;
+	FILE *fp;
+	char buf[MAX_CMDLEN];
+	char *p;
+	struct strlist *sl;
+
+	sl = strlist__new(true, NULL);
+
+	fp = fdopen(dup(fd), "r");
+	while (!feof(fp)) {
+		p = fgets(buf, MAX_CMDLEN, fp);
+		if (!p)
+			break;
+
+		idx = strlen(p) - 1;
+		if (p[idx] == '\n')
+			p[idx] = '\0';
+		ret = strlist__add(sl, buf);
+		if (ret < 0) {
+			pr_debug("strlist__add failed (%d)\n", ret);
+			strlist__delete(sl);
+			return NULL;
+		}
+	}
+	fclose(fp);
+
+	return sl;
+}
+
+struct kprobe_blacklist_node {
+	struct list_head list;
+	unsigned long start;
+	unsigned long end;
+	char *symbol;
+};
+
+static void kprobe_blacklist__delete(struct list_head *blacklist)
+{
+	struct kprobe_blacklist_node *node;
+
+	while (!list_empty(blacklist)) {
+		node = list_first_entry(blacklist,
+					struct kprobe_blacklist_node, list);
+		list_del(&node->list);
+		free(node->symbol);
+		free(node);
+	}
+}
+
+static int kprobe_blacklist__load(struct list_head *blacklist)
+{
+	struct kprobe_blacklist_node *node;
+	const char *__debugfs = debugfs_find_mountpoint();
+	char buf[PATH_MAX], *p;
+	FILE *fp;
+	int ret;
+
+	if (__debugfs == NULL)
+		return -ENOTSUP;
+
+	ret = e_snprintf(buf, PATH_MAX, "%s/kprobes/blacklist", __debugfs);
+	if (ret < 0)
+		return ret;
+
+	fp = fopen(buf, "r");
+	if (!fp)
+		return -errno;
+
+	ret = 0;
+	while (fgets(buf, PATH_MAX, fp)) {
+		node = zalloc(sizeof(*node));
+		if (!node) {
+			ret = -ENOMEM;
+			break;
+		}
+		INIT_LIST_HEAD(&node->list);
+		list_add_tail(&node->list, blacklist);
+		if (sscanf(buf, "0x%lx-0x%lx", &node->start, &node->end) != 2) {
+			ret = -EINVAL;
+			break;
+		}
+		p = strchr(buf, '\t');
+		if (p) {
+			p++;
+			if (p[strlen(p) - 1] == '\n')
+				p[strlen(p) - 1] = '\0';
+		} else
+			p = (char *)"unknown";
+		node->symbol = strdup(p);
+		if (!node->symbol) {
+			ret = -ENOMEM;
+			break;
+		}
+		pr_debug2("Blacklist: 0x%lx-0x%lx, %s\n",
+			  node->start, node->end, node->symbol);
+		ret++;
+	}
+	if (ret < 0)
+		kprobe_blacklist__delete(blacklist);
+	fclose(fp);
+
+	return ret;
+}
+
+static struct kprobe_blacklist_node *
+kprobe_blacklist__find_by_address(struct list_head *blacklist,
+				  unsigned long address)
+{
+	struct kprobe_blacklist_node *node;
+
+	list_for_each_entry(node, blacklist, list) {
+		if (node->start <= address && address <= node->end)
+			return node;
+	}
+
+	return NULL;
+}
+
+/* Show an event */
+static int show_perf_probe_event(struct perf_probe_event *pev,
+				 const char *module)
+{
+	int i, ret;
+	char buf[128];
+	char *place;
+
+	/* Synthesize only event probe point */
+	place = synthesize_perf_probe_point(&pev->point);
+	if (!place)
+		return -EINVAL;
+
+	ret = e_snprintf(buf, 128, "%s:%s", pev->group, pev->event);
+	if (ret < 0)
+		return ret;
+
+	pr_info("  %-20s (on %s", buf, place);
+	if (module)
+		pr_info(" in %s", module);
+
+	if (pev->nargs > 0) {
+		pr_info(" with");
+		for (i = 0; i < pev->nargs; i++) {
+			ret = synthesize_perf_probe_arg(&pev->args[i],
+							buf, 128);
+			if (ret < 0)
+				break;
+			pr_info(" %s", buf);
+		}
+	}
+	pr_info(")\n");
+	free(place);
+	return ret;
+}
+
+static int __show_perf_probe_events(int fd, bool is_kprobe)
+{
+	int ret = 0;
+	struct probe_trace_event tev;
+	struct perf_probe_event pev;
+	struct strlist *rawlist;
+	struct str_node *ent;
+
+	memset(&tev, 0, sizeof(tev));
+	memset(&pev, 0, sizeof(pev));
+
+	rawlist = get_probe_trace_command_rawlist(fd);
+	if (!rawlist)
+		return -ENOMEM;
+
+	strlist__for_each(ent, rawlist) {
+		ret = parse_probe_trace_command(ent->s, &tev);
+		if (ret >= 0) {
+			ret = convert_to_perf_probe_event(&tev, &pev,
+								is_kprobe);
+			if (ret >= 0)
+				ret = show_perf_probe_event(&pev,
+							    tev.point.module);
+		}
+		clear_perf_probe_event(&pev);
+		clear_probe_trace_event(&tev);
+		if (ret < 0)
+			break;
+	}
+	strlist__delete(rawlist);
+
+	return ret;
+}
+
+/* List up current perf-probe events */
+int show_perf_probe_events(void)
+{
+	int kp_fd, up_fd, ret;
+
+	setup_pager();
+
+	ret = init_symbol_maps(false);
+	if (ret < 0)
+		return ret;
+
+	kp_fd = open_kprobe_events(false);
+	if (kp_fd >= 0) {
+		ret = __show_perf_probe_events(kp_fd, true);
+		close(kp_fd);
+		if (ret < 0)
+			goto out;
+	}
+
+	up_fd = open_uprobe_events(false);
+	if (kp_fd < 0 && up_fd < 0) {
+		print_both_open_warning(kp_fd, up_fd);
+		ret = kp_fd;
+		goto out;
+	}
+
+	if (up_fd >= 0) {
+		ret = __show_perf_probe_events(up_fd, false);
+		close(up_fd);
+	}
+out:
+	exit_symbol_maps();
+	return ret;
+}
+
+/* Get current perf-probe event names */
+static struct strlist *get_probe_trace_event_names(int fd, bool include_group)
+{
+	char buf[128];
+	struct strlist *sl, *rawlist;
+	struct str_node *ent;
+	struct probe_trace_event tev;
+	int ret = 0;
+
+	memset(&tev, 0, sizeof(tev));
+	rawlist = get_probe_trace_command_rawlist(fd);
+	if (!rawlist)
+		return NULL;
+	sl = strlist__new(true, NULL);
+	strlist__for_each(ent, rawlist) {
+		ret = parse_probe_trace_command(ent->s, &tev);
+		if (ret < 0)
+			break;
+		if (include_group) {
+			ret = e_snprintf(buf, 128, "%s:%s", tev.group,
+					tev.event);
+			if (ret >= 0)
+				ret = strlist__add(sl, buf);
+		} else
+			ret = strlist__add(sl, tev.event);
+		clear_probe_trace_event(&tev);
+		if (ret < 0)
+			break;
+	}
+	strlist__delete(rawlist);
+
+	if (ret < 0) {
+		strlist__delete(sl);
+		return NULL;
+	}
+	return sl;
+}
+
+static int write_probe_trace_event(int fd, struct probe_trace_event *tev)
+{
+	int ret = 0;
+	char *buf = synthesize_probe_trace_command(tev);
+	char sbuf[STRERR_BUFSIZE];
+
+	if (!buf) {
+		pr_debug("Failed to synthesize probe trace event.\n");
+		return -EINVAL;
+	}
+
+	pr_debug("Writing event: %s\n", buf);
+	if (!probe_event_dry_run) {
+		ret = write(fd, buf, strlen(buf));
+		if (ret <= 0) {
+			ret = -errno;
+			pr_warning("Failed to write event: %s\n",
+				   strerror_r(errno, sbuf, sizeof(sbuf)));
+		}
+	}
+	free(buf);
+	return ret;
+}
+
+static int get_new_event_name(char *buf, size_t len, const char *base,
+			      struct strlist *namelist, bool allow_suffix)
+{
+	int i, ret;
+
+	/* Try no suffix */
+	ret = e_snprintf(buf, len, "%s", base);
+	if (ret < 0) {
+		pr_debug("snprintf() failed: %d\n", ret);
+		return ret;
+	}
+	if (!strlist__has_entry(namelist, buf))
+		return 0;
+
+	if (!allow_suffix) {
+		pr_warning("Error: event \"%s\" already exists. "
+			   "(Use -f to force duplicates.)\n", base);
+		return -EEXIST;
+	}
+
+	/* Try to add suffix */
+	for (i = 1; i < MAX_EVENT_INDEX; i++) {
+		ret = e_snprintf(buf, len, "%s_%d", base, i);
+		if (ret < 0) {
+			pr_debug("snprintf() failed: %d\n", ret);
+			return ret;
+		}
+		if (!strlist__has_entry(namelist, buf))
+			break;
+	}
+	if (i == MAX_EVENT_INDEX) {
+		pr_warning("Too many events are on the same function.\n");
+		ret = -ERANGE;
+	}
+
+	return ret;
+}
+
+/* Warn if the current kernel's uprobe implementation is old */
+static void warn_uprobe_event_compat(struct probe_trace_event *tev)
+{
+	int i;
+	char *buf = synthesize_probe_trace_command(tev);
+
+	/* Old uprobe event doesn't support memory dereference */
+	if (!tev->uprobes || tev->nargs == 0 || !buf)
+		goto out;
+
+	for (i = 0; i < tev->nargs; i++)
+		if (strglobmatch(tev->args[i].value, "[$@+-]*")) {
+			pr_warning("Please upgrade your kernel to at least "
+				   "3.14 to have access to feature %s\n",
+				   tev->args[i].value);
+			break;
+		}
+out:
+	free(buf);
+}
+
+static int __add_probe_trace_events(struct perf_probe_event *pev,
+				     struct probe_trace_event *tevs,
+				     int ntevs, bool allow_suffix)
+{
+	int i, fd, ret;
+	struct probe_trace_event *tev = NULL;
+	char buf[64];
+	const char *event, *group;
+	struct strlist *namelist;
+	LIST_HEAD(blacklist);
+	struct kprobe_blacklist_node *node;
+
+	if (pev->uprobes)
+		fd = open_uprobe_events(true);
+	else
+		fd = open_kprobe_events(true);
+
+	if (fd < 0) {
+		print_open_warning(fd, !pev->uprobes);
+		return fd;
+	}
+
+	/* Get current event names */
+	namelist = get_probe_trace_event_names(fd, false);
+	if (!namelist) {
+		pr_debug("Failed to get current event list.\n");
+		return -EIO;
+	}
+	/* Get kprobe blacklist if exists */
+	if (!pev->uprobes) {
+		ret = kprobe_blacklist__load(&blacklist);
+		if (ret < 0)
+			pr_debug("No kprobe blacklist support, ignored\n");
+	}
+
+	ret = 0;
+	pr_info("Added new event%s\n", (ntevs > 1) ? "s:" : ":");
+	for (i = 0; i < ntevs; i++) {
+		tev = &tevs[i];
+		/* Ensure that the address is NOT blacklisted */
+		node = kprobe_blacklist__find_by_address(&blacklist,
+							 tev->point.address);
+		if (node) {
+			pr_warning("Warning: Skipped probing on blacklisted function: %s\n", node->symbol);
+			continue;
+		}
+
+		if (pev->event)
+			event = pev->event;
+		else
+			if (pev->point.function)
+				event = pev->point.function;
+			else
+				event = tev->point.symbol;
+		if (pev->group)
+			group = pev->group;
+		else
+			group = PERFPROBE_GROUP;
+
+		/* Get an unused new event name */
+		ret = get_new_event_name(buf, 64, event,
+					 namelist, allow_suffix);
+		if (ret < 0)
+			break;
+		event = buf;
+
+		tev->event = strdup(event);
+		tev->group = strdup(group);
+		if (tev->event == NULL || tev->group == NULL) {
+			ret = -ENOMEM;
+			break;
+		}
+		ret = write_probe_trace_event(fd, tev);
+		if (ret < 0)
+			break;
+		/* Add added event name to namelist */
+		strlist__add(namelist, event);
+
+		/* Trick here - save current event/group */
+		event = pev->event;
+		group = pev->group;
+		pev->event = tev->event;
+		pev->group = tev->group;
+		show_perf_probe_event(pev, tev->point.module);
+		/* Trick here - restore current event/group */
+		pev->event = (char *)event;
+		pev->group = (char *)group;
+
+		/*
+		 * Probes after the first probe which comes from same
+		 * user input are always allowed to add suffix, because
+		 * there might be several addresses corresponding to
+		 * one code line.
+		 */
+		allow_suffix = true;
+	}
+	if (ret == -EINVAL && pev->uprobes)
+		warn_uprobe_event_compat(tev);
+
+	/* Note that it is possible to skip all events because of blacklist */
+	if (ret >= 0 && tev->event) {
+		/* Show how to use the event. */
+		pr_info("\nYou can now use it in all perf tools, such as:\n\n");
+		pr_info("\tperf record -e %s:%s -aR sleep 1\n\n", tev->group,
+			 tev->event);
+	}
+
+	kprobe_blacklist__delete(&blacklist);
+	strlist__delete(namelist);
+	close(fd);
+	return ret;
+}
+
+static int find_probe_functions(struct map *map, char *name)
+{
+	int found = 0;
+	struct symbol *sym;
+
+	map__for_each_symbol_by_name(map, name, sym) {
+		found++;
+	}
+
+	return found;
+}
+
+#define strdup_or_goto(str, label)	\
+	({ char *__p = strdup(str); if (!__p) goto label; __p; })
+
+/*
+ * Find probe function addresses from map.
+ * Return an error or the number of found probe_trace_event
+ */
+static int find_probe_trace_events_from_map(struct perf_probe_event *pev,
+					    struct probe_trace_event **tevs,
+					    int max_tevs, const char *target)
+{
+	struct map *map = NULL;
+	struct ref_reloc_sym *reloc_sym = NULL;
+	struct symbol *sym;
+	struct probe_trace_event *tev;
+	struct perf_probe_point *pp = &pev->point;
+	struct probe_trace_point *tp;
+	int num_matched_functions;
+	int ret, i;
+
+	map = get_target_map(target, pev->uprobes);
+	if (!map) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/*
+	 * Load matched symbols: Since the different local symbols may have
+	 * same name but different addresses, this lists all the symbols.
+	 */
+	num_matched_functions = find_probe_functions(map, pp->function);
+	if (num_matched_functions == 0) {
+		pr_err("Failed to find symbol %s in %s\n", pp->function,
+			target ? : "kernel");
+		ret = -ENOENT;
+		goto out;
+	} else if (num_matched_functions > max_tevs) {
+		pr_err("Too many functions matched in %s\n",
+			target ? : "kernel");
+		ret = -E2BIG;
+		goto out;
+	}
+
+	if (!pev->uprobes && !pp->retprobe) {
+		reloc_sym = kernel_get_ref_reloc_sym();
+		if (!reloc_sym) {
+			pr_warning("Relocated base symbol is not found!\n");
+			ret = -EINVAL;
+			goto out;
+		}
+	}
+
+	/* Setup result trace-probe-events */
+	*tevs = zalloc(sizeof(*tev) * num_matched_functions);
+	if (!*tevs) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = 0;
+
+	map__for_each_symbol_by_name(map, pp->function, sym) {
+		tev = (*tevs) + ret;
+		tp = &tev->point;
+		if (ret == num_matched_functions) {
+			pr_warning("Too many symbols are listed. Skip it.\n");
+			break;
+		}
+		ret++;
+
+		if (pp->offset > sym->end - sym->start) {
+			pr_warning("Offset %ld is bigger than the size of %s\n",
+				   pp->offset, sym->name);
+			ret = -ENOENT;
+			goto err_out;
+		}
+		/* Add one probe point */
+		tp->address = map->unmap_ip(map, sym->start) + pp->offset;
+		if (reloc_sym) {
+			tp->symbol = strdup_or_goto(reloc_sym->name, nomem_out);
+			tp->offset = tp->address - reloc_sym->addr;
+		} else {
+			tp->symbol = strdup_or_goto(sym->name, nomem_out);
+			tp->offset = pp->offset;
+		}
+		tp->retprobe = pp->retprobe;
+		if (target)
+			tev->point.module = strdup_or_goto(target, nomem_out);
+		tev->uprobes = pev->uprobes;
+		tev->nargs = pev->nargs;
+		if (tev->nargs) {
+			tev->args = zalloc(sizeof(struct probe_trace_arg) *
+					   tev->nargs);
+			if (tev->args == NULL)
+				goto nomem_out;
+		}
+		for (i = 0; i < tev->nargs; i++) {
+			if (pev->args[i].name)
+				tev->args[i].name =
+					strdup_or_goto(pev->args[i].name,
+							nomem_out);
+
+			tev->args[i].value = strdup_or_goto(pev->args[i].var,
+							    nomem_out);
+			if (pev->args[i].type)
+				tev->args[i].type =
+					strdup_or_goto(pev->args[i].type,
+							nomem_out);
+		}
+	}
+
+out:
+	put_target_map(map, pev->uprobes);
+	return ret;
+
+nomem_out:
+	ret = -ENOMEM;
+err_out:
+	clear_probe_trace_events(*tevs, num_matched_functions);
+	zfree(tevs);
+	goto out;
+}
+
+static int convert_to_probe_trace_events(struct perf_probe_event *pev,
+					  struct probe_trace_event **tevs,
+					  int max_tevs, const char *target)
+{
+	int ret;
+
+	if (pev->uprobes && !pev->group) {
+		/* Replace group name if not given */
+		ret = convert_exec_to_group(target, &pev->group);
+		if (ret != 0) {
+			pr_warning("Failed to make a group name.\n");
+			return ret;
+		}
+	}
+
+	/* Convert perf_probe_event with debuginfo */
+	ret = try_to_find_probe_trace_events(pev, tevs, max_tevs, target);
+	if (ret != 0)
+		return ret;	/* Found in debuginfo or got an error */
+
+	return find_probe_trace_events_from_map(pev, tevs, max_tevs, target);
+}
+
+struct __event_package {
+	struct perf_probe_event		*pev;
+	struct probe_trace_event	*tevs;
+	int				ntevs;
+};
+
+int add_perf_probe_events(struct perf_probe_event *pevs, int npevs,
+			  int max_tevs, bool force_add)
+{
+	int i, j, ret;
+	struct __event_package *pkgs;
+
+	ret = 0;
+	pkgs = zalloc(sizeof(struct __event_package) * npevs);
+
+	if (pkgs == NULL)
+		return -ENOMEM;
+
+	ret = init_symbol_maps(pevs->uprobes);
+	if (ret < 0) {
+		free(pkgs);
+		return ret;
+	}
+
+	/* Loop 1: convert all events */
+	for (i = 0; i < npevs; i++) {
+		pkgs[i].pev = &pevs[i];
+		/* Convert with or without debuginfo */
+		ret  = convert_to_probe_trace_events(pkgs[i].pev,
+						     &pkgs[i].tevs,
+						     max_tevs,
+						     pkgs[i].pev->target);
+		if (ret < 0)
+			goto end;
+		pkgs[i].ntevs = ret;
+	}
+
+	/* Loop 2: add all events */
+	for (i = 0; i < npevs; i++) {
+		ret = __add_probe_trace_events(pkgs[i].pev, pkgs[i].tevs,
+						pkgs[i].ntevs, force_add);
+		if (ret < 0)
+			break;
+	}
+end:
+	/* Loop 3: cleanup and free trace events  */
+	for (i = 0; i < npevs; i++) {
+		for (j = 0; j < pkgs[i].ntevs; j++)
+			clear_probe_trace_event(&pkgs[i].tevs[j]);
+		zfree(&pkgs[i].tevs);
+	}
+	free(pkgs);
+	exit_symbol_maps();
+
+	return ret;
+}
+
+static int __del_trace_probe_event(int fd, struct str_node *ent)
+{
+	char *p;
+	char buf[128];
+	int ret;
+
+	/* Convert from perf-probe event to trace-probe event */
+	ret = e_snprintf(buf, 128, "-:%s", ent->s);
+	if (ret < 0)
+		goto error;
+
+	p = strchr(buf + 2, ':');
+	if (!p) {
+		pr_debug("Internal error: %s should have ':' but not.\n",
+			 ent->s);
+		ret = -ENOTSUP;
+		goto error;
+	}
+	*p = '/';
+
+	pr_debug("Writing event: %s\n", buf);
+	ret = write(fd, buf, strlen(buf));
+	if (ret < 0) {
+		ret = -errno;
+		goto error;
+	}
+
+	pr_info("Removed event: %s\n", ent->s);
+	return 0;
+error:
+	pr_warning("Failed to delete event: %s\n",
+		   strerror_r(-ret, buf, sizeof(buf)));
+	return ret;
+}
+
+static int del_trace_probe_event(int fd, const char *buf,
+						  struct strlist *namelist)
+{
+	struct str_node *ent, *n;
+	int ret = -1;
+
+	if (strpbrk(buf, "*?")) { /* Glob-exp */
+		strlist__for_each_safe(ent, n, namelist)
+			if (strglobmatch(ent->s, buf)) {
+				ret = __del_trace_probe_event(fd, ent);
+				if (ret < 0)
+					break;
+				strlist__remove(namelist, ent);
+			}
+	} else {
+		ent = strlist__find(namelist, buf);
+		if (ent) {
+			ret = __del_trace_probe_event(fd, ent);
+			if (ret >= 0)
+				strlist__remove(namelist, ent);
+		}
+	}
+
+	return ret;
+}
+
+int del_perf_probe_events(struct strlist *dellist)
+{
+	int ret = -1, ufd = -1, kfd = -1;
+	char buf[128];
+	const char *group, *event;
+	char *p, *str;
+	struct str_node *ent;
+	struct strlist *namelist = NULL, *unamelist = NULL;
+
+	/* Get current event names */
+	kfd = open_kprobe_events(true);
+	if (kfd >= 0)
+		namelist = get_probe_trace_event_names(kfd, true);
+
+	ufd = open_uprobe_events(true);
+	if (ufd >= 0)
+		unamelist = get_probe_trace_event_names(ufd, true);
+
+	if (kfd < 0 && ufd < 0) {
+		print_both_open_warning(kfd, ufd);
+		goto error;
+	}
+
+	if (namelist == NULL && unamelist == NULL)
+		goto error;
+
+	strlist__for_each(ent, dellist) {
+		str = strdup(ent->s);
+		if (str == NULL) {
+			ret = -ENOMEM;
+			goto error;
+		}
+		pr_debug("Parsing: %s\n", str);
+		p = strchr(str, ':');
+		if (p) {
+			group = str;
+			*p = '\0';
+			event = p + 1;
+		} else {
+			group = "*";
+			event = str;
+		}
+
+		ret = e_snprintf(buf, 128, "%s:%s", group, event);
+		if (ret < 0) {
+			pr_err("Failed to copy event.");
+			free(str);
+			goto error;
+		}
+
+		pr_debug("Group: %s, Event: %s\n", group, event);
+
+		if (namelist)
+			ret = del_trace_probe_event(kfd, buf, namelist);
+
+		if (unamelist && ret != 0)
+			ret = del_trace_probe_event(ufd, buf, unamelist);
+
+		if (ret != 0)
+			pr_info("Info: Event \"%s\" does not exist.\n", buf);
+
+		free(str);
+	}
+
+error:
+	if (kfd >= 0) {
+		strlist__delete(namelist);
+		close(kfd);
+	}
+
+	if (ufd >= 0) {
+		strlist__delete(unamelist);
+		close(ufd);
+	}
+
+	return ret;
+}
+
+/* TODO: don't use a global variable for filter ... */
+static struct strfilter *available_func_filter;
+
+/*
+ * If a symbol corresponds to a function with global binding and
+ * matches filter return 0. For all others return 1.
+ */
+static int filter_available_functions(struct map *map __maybe_unused,
+				      struct symbol *sym)
+{
+	if (strfilter__compare(available_func_filter, sym->name))
+		return 0;
+	return 1;
+}
+
+int show_available_funcs(const char *target, struct strfilter *_filter,
+					bool user)
+{
+	struct map *map;
+	int ret;
+
+	ret = init_symbol_maps(user);
+	if (ret < 0)
+		return ret;
+
+	/* Get a symbol map */
+	if (user)
+		map = dso__new_map(target);
+	else
+		map = kernel_get_module_map(target);
+	if (!map) {
+		pr_err("Failed to get a map for %s\n", (target) ? : "kernel");
+		return -EINVAL;
+	}
+
+	/* Load symbols with given filter */
+	available_func_filter = _filter;
+	if (map__load(map, filter_available_functions)) {
+		pr_err("Failed to load symbols in %s\n", (target) ? : "kernel");
+		goto end;
+	}
+	if (!dso__sorted_by_name(map->dso, map->type))
+		dso__sort_by_name(map->dso, map->type);
+
+	/* Show all (filtered) symbols */
+	setup_pager();
+	dso__fprintf_symbols_by_name(map->dso, map->type, stdout);
+end:
+	if (user) {
+		dso__delete(map->dso);
+		map__delete(map);
+	}
+	exit_symbol_maps();
+
+	return ret;
+}
+
diff --git a/tools/perf/util/probe-event.h b/tools/perf/util/probe-event.h
new file mode 100644
index 000000000..d6b783447
--- /dev/null
+++ b/tools/perf/util/probe-event.h
@@ -0,0 +1,142 @@
+#ifndef _PROBE_EVENT_H
+#define _PROBE_EVENT_H
+
+#include <stdbool.h>
+#include "intlist.h"
+#include "strlist.h"
+#include "strfilter.h"
+
+extern bool probe_event_dry_run;
+
+/* kprobe-tracer and uprobe-tracer tracing point */
+struct probe_trace_point {
+	char		*symbol;	/* Base symbol */
+	char		*module;	/* Module name */
+	unsigned long	offset;		/* Offset from symbol */
+	unsigned long	address;	/* Actual address of the trace point */
+	bool		retprobe;	/* Return probe flag */
+};
+
+/* probe-tracer tracing argument referencing offset */
+struct probe_trace_arg_ref {
+	struct probe_trace_arg_ref	*next;	/* Next reference */
+	long				offset;	/* Offset value */
+};
+
+/* kprobe-tracer and uprobe-tracer tracing argument */
+struct probe_trace_arg {
+	char				*name;	/* Argument name */
+	char				*value;	/* Base value */
+	char				*type;	/* Type name */
+	struct probe_trace_arg_ref	*ref;	/* Referencing offset */
+};
+
+/* kprobe-tracer and uprobe-tracer tracing event (point + arg) */
+struct probe_trace_event {
+	char				*event;	/* Event name */
+	char				*group;	/* Group name */
+	struct probe_trace_point	point;	/* Trace point */
+	int				nargs;	/* Number of args */
+	bool				uprobes;	/* uprobes only */
+	struct probe_trace_arg		*args;	/* Arguments */
+};
+
+/* Perf probe probing point */
+struct perf_probe_point {
+	char		*file;		/* File path */
+	char		*function;	/* Function name */
+	int		line;		/* Line number */
+	bool		retprobe;	/* Return probe flag */
+	char		*lazy_line;	/* Lazy matching pattern */
+	unsigned long	offset;		/* Offset from function entry */
+};
+
+/* Perf probe probing argument field chain */
+struct perf_probe_arg_field {
+	struct perf_probe_arg_field	*next;	/* Next field */
+	char				*name;	/* Name of the field */
+	long				index;	/* Array index number */
+	bool				ref;	/* Referencing flag */
+};
+
+/* Perf probe probing argument */
+struct perf_probe_arg {
+	char				*name;	/* Argument name */
+	char				*var;	/* Variable name */
+	char				*type;	/* Type name */
+	struct perf_probe_arg_field	*field;	/* Structure fields */
+};
+
+/* Perf probe probing event (point + arg) */
+struct perf_probe_event {
+	char			*event;	/* Event name */
+	char			*group;	/* Group name */
+	struct perf_probe_point	point;	/* Probe point */
+	int			nargs;	/* Number of arguments */
+	bool			uprobes;	/* Uprobe event flag */
+	char			*target;	/* Target binary */
+	struct perf_probe_arg	*args;	/* Arguments */
+};
+
+/* Line range */
+struct line_range {
+	char			*file;		/* File name */
+	char			*function;	/* Function name */
+	int			start;		/* Start line number */
+	int			end;		/* End line number */
+	int			offset;		/* Start line offset */
+	char			*path;		/* Real path name */
+	char			*comp_dir;	/* Compile directory */
+	struct intlist		*line_list;	/* Visible lines */
+};
+
+/* List of variables */
+struct variable_list {
+	struct probe_trace_point	point;	/* Actual probepoint */
+	struct strlist			*vars;	/* Available variables */
+};
+
+/* Command string to events */
+extern int parse_perf_probe_command(const char *cmd,
+				    struct perf_probe_event *pev);
+
+/* Events to command string */
+extern char *synthesize_perf_probe_command(struct perf_probe_event *pev);
+extern char *synthesize_probe_trace_command(struct probe_trace_event *tev);
+extern int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf,
+				     size_t len);
+
+/* Check the perf_probe_event needs debuginfo */
+extern bool perf_probe_event_need_dwarf(struct perf_probe_event *pev);
+
+/* Release event contents */
+extern void clear_perf_probe_event(struct perf_probe_event *pev);
+
+/* Command string to line-range */
+extern int parse_line_range_desc(const char *cmd, struct line_range *lr);
+
+/* Release line range members */
+extern void line_range__clear(struct line_range *lr);
+
+/* Initialize line range */
+extern int line_range__init(struct line_range *lr);
+
+/* Internal use: Return kernel/module path */
+extern const char *kernel_get_module_path(const char *module);
+
+extern int add_perf_probe_events(struct perf_probe_event *pevs, int npevs,
+				 int max_probe_points, bool force_add);
+extern int del_perf_probe_events(struct strlist *dellist);
+extern int show_perf_probe_events(void);
+extern int show_line_range(struct line_range *lr, const char *module,
+			   bool user);
+extern int show_available_vars(struct perf_probe_event *pevs, int npevs,
+			       int max_probe_points, const char *module,
+			       struct strfilter *filter, bool externs);
+extern int show_available_funcs(const char *module, struct strfilter *filter,
+				bool user);
+
+/* Maximum index number of event-name postfix */
+#define MAX_EVENT_INDEX	1024
+
+#endif /*_PROBE_EVENT_H */
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
new file mode 100644
index 000000000..2a76e14db
--- /dev/null
+++ b/tools/perf/util/probe-finder.c
@@ -0,0 +1,1695 @@
+/*
+ * probe-finder.c : C expression to kprobe event converter
+ *
+ * Written by Masami Hiramatsu <mhiramat@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <sys/utsname.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <dwarf-regs.h>
+
+#include <linux/bitops.h>
+#include "event.h"
+#include "dso.h"
+#include "debug.h"
+#include "intlist.h"
+#include "util.h"
+#include "symbol.h"
+#include "probe-finder.h"
+
+/* Kprobe tracer basic type is up to u64 */
+#define MAX_BASIC_TYPE_BITS	64
+
+/* Dwarf FL wrappers */
+static char *debuginfo_path;	/* Currently dummy */
+
+static const Dwfl_Callbacks offline_callbacks = {
+	.find_debuginfo = dwfl_standard_find_debuginfo,
+	.debuginfo_path = &debuginfo_path,
+
+	.section_address = dwfl_offline_section_address,
+
+	/* We use this table for core files too.  */
+	.find_elf = dwfl_build_id_find_elf,
+};
+
+/* Get a Dwarf from offline image */
+static int debuginfo__init_offline_dwarf(struct debuginfo *dbg,
+					 const char *path)
+{
+	int fd;
+
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	dbg->dwfl = dwfl_begin(&offline_callbacks);
+	if (!dbg->dwfl)
+		goto error;
+
+	dbg->mod = dwfl_report_offline(dbg->dwfl, "", "", fd);
+	if (!dbg->mod)
+		goto error;
+
+	dbg->dbg = dwfl_module_getdwarf(dbg->mod, &dbg->bias);
+	if (!dbg->dbg)
+		goto error;
+
+	return 0;
+error:
+	if (dbg->dwfl)
+		dwfl_end(dbg->dwfl);
+	else
+		close(fd);
+	memset(dbg, 0, sizeof(*dbg));
+
+	return -ENOENT;
+}
+
+static struct debuginfo *__debuginfo__new(const char *path)
+{
+	struct debuginfo *dbg = zalloc(sizeof(*dbg));
+	if (!dbg)
+		return NULL;
+
+	if (debuginfo__init_offline_dwarf(dbg, path) < 0)
+		zfree(&dbg);
+	if (dbg)
+		pr_debug("Open Debuginfo file: %s\n", path);
+	return dbg;
+}
+
+enum dso_binary_type distro_dwarf_types[] = {
+	DSO_BINARY_TYPE__FEDORA_DEBUGINFO,
+	DSO_BINARY_TYPE__UBUNTU_DEBUGINFO,
+	DSO_BINARY_TYPE__OPENEMBEDDED_DEBUGINFO,
+	DSO_BINARY_TYPE__BUILDID_DEBUGINFO,
+	DSO_BINARY_TYPE__NOT_FOUND,
+};
+
+struct debuginfo *debuginfo__new(const char *path)
+{
+	enum dso_binary_type *type;
+	char buf[PATH_MAX], nil = '\0';
+	struct dso *dso;
+	struct debuginfo *dinfo = NULL;
+
+	/* Try to open distro debuginfo files */
+	dso = dso__new(path);
+	if (!dso)
+		goto out;
+
+	for (type = distro_dwarf_types;
+	     !dinfo && *type != DSO_BINARY_TYPE__NOT_FOUND;
+	     type++) {
+		if (dso__read_binary_type_filename(dso, *type, &nil,
+						   buf, PATH_MAX) < 0)
+			continue;
+		dinfo = __debuginfo__new(buf);
+	}
+	dso__delete(dso);
+
+out:
+	/* if failed to open all distro debuginfo, open given binary */
+	return dinfo ? : __debuginfo__new(path);
+}
+
+void debuginfo__delete(struct debuginfo *dbg)
+{
+	if (dbg) {
+		if (dbg->dwfl)
+			dwfl_end(dbg->dwfl);
+		free(dbg);
+	}
+}
+
+/*
+ * Probe finder related functions
+ */
+
+static struct probe_trace_arg_ref *alloc_trace_arg_ref(long offs)
+{
+	struct probe_trace_arg_ref *ref;
+	ref = zalloc(sizeof(struct probe_trace_arg_ref));
+	if (ref != NULL)
+		ref->offset = offs;
+	return ref;
+}
+
+/*
+ * Convert a location into trace_arg.
+ * If tvar == NULL, this just checks variable can be converted.
+ * If fentry == true and vr_die is a parameter, do huristic search
+ * for the location fuzzed by function entry mcount.
+ */
+static int convert_variable_location(Dwarf_Die *vr_die, Dwarf_Addr addr,
+				     Dwarf_Op *fb_ops, Dwarf_Die *sp_die,
+				     struct probe_trace_arg *tvar)
+{
+	Dwarf_Attribute attr;
+	Dwarf_Addr tmp = 0;
+	Dwarf_Op *op;
+	size_t nops;
+	unsigned int regn;
+	Dwarf_Word offs = 0;
+	bool ref = false;
+	const char *regs;
+	int ret;
+
+	if (dwarf_attr(vr_die, DW_AT_external, &attr) != NULL)
+		goto static_var;
+
+	/* TODO: handle more than 1 exprs */
+	if (dwarf_attr(vr_die, DW_AT_location, &attr) == NULL)
+		return -EINVAL;	/* Broken DIE ? */
+	if (dwarf_getlocation_addr(&attr, addr, &op, &nops, 1) <= 0) {
+		ret = dwarf_entrypc(sp_die, &tmp);
+		if (ret || addr != tmp ||
+		    dwarf_tag(vr_die) != DW_TAG_formal_parameter ||
+		    dwarf_highpc(sp_die, &tmp))
+			return -ENOENT;
+		/*
+		 * This is fuzzed by fentry mcount. We try to find the
+		 * parameter location at the earliest address.
+		 */
+		for (addr += 1; addr <= tmp; addr++) {
+			if (dwarf_getlocation_addr(&attr, addr, &op,
+						   &nops, 1) > 0)
+				goto found;
+		}
+		return -ENOENT;
+	}
+found:
+	if (nops == 0)
+		/* TODO: Support const_value */
+		return -ENOENT;
+
+	if (op->atom == DW_OP_addr) {
+static_var:
+		if (!tvar)
+			return 0;
+		/* Static variables on memory (not stack), make @varname */
+		ret = strlen(dwarf_diename(vr_die));
+		tvar->value = zalloc(ret + 2);
+		if (tvar->value == NULL)
+			return -ENOMEM;
+		snprintf(tvar->value, ret + 2, "@%s", dwarf_diename(vr_die));
+		tvar->ref = alloc_trace_arg_ref((long)offs);
+		if (tvar->ref == NULL)
+			return -ENOMEM;
+		return 0;
+	}
+
+	/* If this is based on frame buffer, set the offset */
+	if (op->atom == DW_OP_fbreg) {
+		if (fb_ops == NULL)
+			return -ENOTSUP;
+		ref = true;
+		offs = op->number;
+		op = &fb_ops[0];
+	}
+
+	if (op->atom >= DW_OP_breg0 && op->atom <= DW_OP_breg31) {
+		regn = op->atom - DW_OP_breg0;
+		offs += op->number;
+		ref = true;
+	} else if (op->atom >= DW_OP_reg0 && op->atom <= DW_OP_reg31) {
+		regn = op->atom - DW_OP_reg0;
+	} else if (op->atom == DW_OP_bregx) {
+		regn = op->number;
+		offs += op->number2;
+		ref = true;
+	} else if (op->atom == DW_OP_regx) {
+		regn = op->number;
+	} else {
+		pr_debug("DW_OP %x is not supported.\n", op->atom);
+		return -ENOTSUP;
+	}
+
+	if (!tvar)
+		return 0;
+
+	regs = get_arch_regstr(regn);
+	if (!regs) {
+		/* This should be a bug in DWARF or this tool */
+		pr_warning("Mapping for the register number %u "
+			   "missing on this architecture.\n", regn);
+		return -ERANGE;
+	}
+
+	tvar->value = strdup(regs);
+	if (tvar->value == NULL)
+		return -ENOMEM;
+
+	if (ref) {
+		tvar->ref = alloc_trace_arg_ref((long)offs);
+		if (tvar->ref == NULL)
+			return -ENOMEM;
+	}
+	return 0;
+}
+
+#define BYTES_TO_BITS(nb)	((nb) * BITS_PER_LONG / sizeof(long))
+
+static int convert_variable_type(Dwarf_Die *vr_die,
+				 struct probe_trace_arg *tvar,
+				 const char *cast)
+{
+	struct probe_trace_arg_ref **ref_ptr = &tvar->ref;
+	Dwarf_Die type;
+	char buf[16];
+	char sbuf[STRERR_BUFSIZE];
+	int bsize, boffs, total;
+	int ret;
+
+	/* TODO: check all types */
+	if (cast && strcmp(cast, "string") != 0) {
+		/* Non string type is OK */
+		tvar->type = strdup(cast);
+		return (tvar->type == NULL) ? -ENOMEM : 0;
+	}
+
+	bsize = dwarf_bitsize(vr_die);
+	if (bsize > 0) {
+		/* This is a bitfield */
+		boffs = dwarf_bitoffset(vr_die);
+		total = dwarf_bytesize(vr_die);
+		if (boffs < 0 || total < 0)
+			return -ENOENT;
+		ret = snprintf(buf, 16, "b%d@%d/%zd", bsize, boffs,
+				BYTES_TO_BITS(total));
+		goto formatted;
+	}
+
+	if (die_get_real_type(vr_die, &type) == NULL) {
+		pr_warning("Failed to get a type information of %s.\n",
+			   dwarf_diename(vr_die));
+		return -ENOENT;
+	}
+
+	pr_debug("%s type is %s.\n",
+		 dwarf_diename(vr_die), dwarf_diename(&type));
+
+	if (cast && strcmp(cast, "string") == 0) {	/* String type */
+		ret = dwarf_tag(&type);
+		if (ret != DW_TAG_pointer_type &&
+		    ret != DW_TAG_array_type) {
+			pr_warning("Failed to cast into string: "
+				   "%s(%s) is not a pointer nor array.\n",
+				   dwarf_diename(vr_die), dwarf_diename(&type));
+			return -EINVAL;
+		}
+		if (die_get_real_type(&type, &type) == NULL) {
+			pr_warning("Failed to get a type"
+				   " information.\n");
+			return -ENOENT;
+		}
+		if (ret == DW_TAG_pointer_type) {
+			while (*ref_ptr)
+				ref_ptr = &(*ref_ptr)->next;
+			/* Add new reference with offset +0 */
+			*ref_ptr = zalloc(sizeof(struct probe_trace_arg_ref));
+			if (*ref_ptr == NULL) {
+				pr_warning("Out of memory error\n");
+				return -ENOMEM;
+			}
+		}
+		if (!die_compare_name(&type, "char") &&
+		    !die_compare_name(&type, "unsigned char")) {
+			pr_warning("Failed to cast into string: "
+				   "%s is not (unsigned) char *.\n",
+				   dwarf_diename(vr_die));
+			return -EINVAL;
+		}
+		tvar->type = strdup(cast);
+		return (tvar->type == NULL) ? -ENOMEM : 0;
+	}
+
+	ret = dwarf_bytesize(&type);
+	if (ret <= 0)
+		/* No size ... try to use default type */
+		return 0;
+	ret = BYTES_TO_BITS(ret);
+
+	/* Check the bitwidth */
+	if (ret > MAX_BASIC_TYPE_BITS) {
+		pr_info("%s exceeds max-bitwidth. Cut down to %d bits.\n",
+			dwarf_diename(&type), MAX_BASIC_TYPE_BITS);
+		ret = MAX_BASIC_TYPE_BITS;
+	}
+	ret = snprintf(buf, 16, "%c%d",
+		       die_is_signed_type(&type) ? 's' : 'u', ret);
+
+formatted:
+	if (ret < 0 || ret >= 16) {
+		if (ret >= 16)
+			ret = -E2BIG;
+		pr_warning("Failed to convert variable type: %s\n",
+			   strerror_r(-ret, sbuf, sizeof(sbuf)));
+		return ret;
+	}
+	tvar->type = strdup(buf);
+	if (tvar->type == NULL)
+		return -ENOMEM;
+	return 0;
+}
+
+static int convert_variable_fields(Dwarf_Die *vr_die, const char *varname,
+				    struct perf_probe_arg_field *field,
+				    struct probe_trace_arg_ref **ref_ptr,
+				    Dwarf_Die *die_mem)
+{
+	struct probe_trace_arg_ref *ref = *ref_ptr;
+	Dwarf_Die type;
+	Dwarf_Word offs;
+	int ret, tag;
+
+	pr_debug("converting %s in %s\n", field->name, varname);
+	if (die_get_real_type(vr_die, &type) == NULL) {
+		pr_warning("Failed to get the type of %s.\n", varname);
+		return -ENOENT;
+	}
+	pr_debug2("Var real type: (%x)\n", (unsigned)dwarf_dieoffset(&type));
+	tag = dwarf_tag(&type);
+
+	if (field->name[0] == '[' &&
+	    (tag == DW_TAG_array_type || tag == DW_TAG_pointer_type)) {
+		if (field->next)
+			/* Save original type for next field */
+			memcpy(die_mem, &type, sizeof(*die_mem));
+		/* Get the type of this array */
+		if (die_get_real_type(&type, &type) == NULL) {
+			pr_warning("Failed to get the type of %s.\n", varname);
+			return -ENOENT;
+		}
+		pr_debug2("Array real type: (%x)\n",
+			 (unsigned)dwarf_dieoffset(&type));
+		if (tag == DW_TAG_pointer_type) {
+			ref = zalloc(sizeof(struct probe_trace_arg_ref));
+			if (ref == NULL)
+				return -ENOMEM;
+			if (*ref_ptr)
+				(*ref_ptr)->next = ref;
+			else
+				*ref_ptr = ref;
+		}
+		ref->offset += dwarf_bytesize(&type) * field->index;
+		if (!field->next)
+			/* Save vr_die for converting types */
+			memcpy(die_mem, vr_die, sizeof(*die_mem));
+		goto next;
+	} else if (tag == DW_TAG_pointer_type) {
+		/* Check the pointer and dereference */
+		if (!field->ref) {
+			pr_err("Semantic error: %s must be referred by '->'\n",
+			       field->name);
+			return -EINVAL;
+		}
+		/* Get the type pointed by this pointer */
+		if (die_get_real_type(&type, &type) == NULL) {
+			pr_warning("Failed to get the type of %s.\n", varname);
+			return -ENOENT;
+		}
+		/* Verify it is a data structure  */
+		tag = dwarf_tag(&type);
+		if (tag != DW_TAG_structure_type && tag != DW_TAG_union_type) {
+			pr_warning("%s is not a data structure nor an union.\n",
+				   varname);
+			return -EINVAL;
+		}
+
+		ref = zalloc(sizeof(struct probe_trace_arg_ref));
+		if (ref == NULL)
+			return -ENOMEM;
+		if (*ref_ptr)
+			(*ref_ptr)->next = ref;
+		else
+			*ref_ptr = ref;
+	} else {
+		/* Verify it is a data structure  */
+		if (tag != DW_TAG_structure_type && tag != DW_TAG_union_type) {
+			pr_warning("%s is not a data structure nor an union.\n",
+				   varname);
+			return -EINVAL;
+		}
+		if (field->name[0] == '[') {
+			pr_err("Semantic error: %s is not a pointer"
+			       " nor array.\n", varname);
+			return -EINVAL;
+		}
+		/* While prcessing unnamed field, we don't care about this */
+		if (field->ref && dwarf_diename(vr_die)) {
+			pr_err("Semantic error: %s must be referred by '.'\n",
+			       field->name);
+			return -EINVAL;
+		}
+		if (!ref) {
+			pr_warning("Structure on a register is not "
+				   "supported yet.\n");
+			return -ENOTSUP;
+		}
+	}
+
+	if (die_find_member(&type, field->name, die_mem) == NULL) {
+		pr_warning("%s(type:%s) has no member %s.\n", varname,
+			   dwarf_diename(&type), field->name);
+		return -EINVAL;
+	}
+
+	/* Get the offset of the field */
+	if (tag == DW_TAG_union_type) {
+		offs = 0;
+	} else {
+		ret = die_get_data_member_location(die_mem, &offs);
+		if (ret < 0) {
+			pr_warning("Failed to get the offset of %s.\n",
+				   field->name);
+			return ret;
+		}
+	}
+	ref->offset += (long)offs;
+
+	/* If this member is unnamed, we need to reuse this field */
+	if (!dwarf_diename(die_mem))
+		return convert_variable_fields(die_mem, varname, field,
+						&ref, die_mem);
+
+next:
+	/* Converting next field */
+	if (field->next)
+		return convert_variable_fields(die_mem, field->name,
+					field->next, &ref, die_mem);
+	else
+		return 0;
+}
+
+/* Show a variables in kprobe event format */
+static int convert_variable(Dwarf_Die *vr_die, struct probe_finder *pf)
+{
+	Dwarf_Die die_mem;
+	int ret;
+
+	pr_debug("Converting variable %s into trace event.\n",
+		 dwarf_diename(vr_die));
+
+	ret = convert_variable_location(vr_die, pf->addr, pf->fb_ops,
+					&pf->sp_die, pf->tvar);
+	if (ret == -ENOENT || ret == -EINVAL)
+		pr_err("Failed to find the location of %s at this address.\n"
+		       " Perhaps, it has been optimized out.\n", pf->pvar->var);
+	else if (ret == -ENOTSUP)
+		pr_err("Sorry, we don't support this variable location yet.\n");
+	else if (ret == 0 && pf->pvar->field) {
+		ret = convert_variable_fields(vr_die, pf->pvar->var,
+					      pf->pvar->field, &pf->tvar->ref,
+					      &die_mem);
+		vr_die = &die_mem;
+	}
+	if (ret == 0)
+		ret = convert_variable_type(vr_die, pf->tvar, pf->pvar->type);
+	/* *expr will be cached in libdw. Don't free it. */
+	return ret;
+}
+
+/* Find a variable in a scope DIE */
+static int find_variable(Dwarf_Die *sc_die, struct probe_finder *pf)
+{
+	Dwarf_Die vr_die;
+	char buf[32], *ptr;
+	int ret = 0;
+
+	if (!is_c_varname(pf->pvar->var)) {
+		/* Copy raw parameters */
+		pf->tvar->value = strdup(pf->pvar->var);
+		if (pf->tvar->value == NULL)
+			return -ENOMEM;
+		if (pf->pvar->type) {
+			pf->tvar->type = strdup(pf->pvar->type);
+			if (pf->tvar->type == NULL)
+				return -ENOMEM;
+		}
+		if (pf->pvar->name) {
+			pf->tvar->name = strdup(pf->pvar->name);
+			if (pf->tvar->name == NULL)
+				return -ENOMEM;
+		} else
+			pf->tvar->name = NULL;
+		return 0;
+	}
+
+	if (pf->pvar->name)
+		pf->tvar->name = strdup(pf->pvar->name);
+	else {
+		ret = synthesize_perf_probe_arg(pf->pvar, buf, 32);
+		if (ret < 0)
+			return ret;
+		ptr = strchr(buf, ':');	/* Change type separator to _ */
+		if (ptr)
+			*ptr = '_';
+		pf->tvar->name = strdup(buf);
+	}
+	if (pf->tvar->name == NULL)
+		return -ENOMEM;
+
+	pr_debug("Searching '%s' variable in context.\n", pf->pvar->var);
+	/* Search child die for local variables and parameters. */
+	if (!die_find_variable_at(sc_die, pf->pvar->var, pf->addr, &vr_die)) {
+		/* Search again in global variables */
+		if (!die_find_variable_at(&pf->cu_die, pf->pvar->var,
+						0, &vr_die)) {
+			pr_warning("Failed to find '%s' in this function.\n",
+				   pf->pvar->var);
+			ret = -ENOENT;
+		}
+	}
+	if (ret >= 0)
+		ret = convert_variable(&vr_die, pf);
+
+	return ret;
+}
+
+/* Convert subprogram DIE to trace point */
+static int convert_to_trace_point(Dwarf_Die *sp_die, Dwfl_Module *mod,
+				  Dwarf_Addr paddr, bool retprobe,
+				  struct probe_trace_point *tp)
+{
+	Dwarf_Addr eaddr, highaddr;
+	GElf_Sym sym;
+	const char *symbol;
+
+	/* Verify the address is correct */
+	if (dwarf_entrypc(sp_die, &eaddr) != 0) {
+		pr_warning("Failed to get entry address of %s\n",
+			   dwarf_diename(sp_die));
+		return -ENOENT;
+	}
+	if (dwarf_highpc(sp_die, &highaddr) != 0) {
+		pr_warning("Failed to get end address of %s\n",
+			   dwarf_diename(sp_die));
+		return -ENOENT;
+	}
+	if (paddr > highaddr) {
+		pr_warning("Offset specified is greater than size of %s\n",
+			   dwarf_diename(sp_die));
+		return -EINVAL;
+	}
+
+	symbol = dwarf_diename(sp_die);
+	if (!symbol) {
+		/* Try to get the symbol name from symtab */
+		symbol = dwfl_module_addrsym(mod, paddr, &sym, NULL);
+		if (!symbol) {
+			pr_warning("Failed to find symbol at 0x%lx\n",
+				   (unsigned long)paddr);
+			return -ENOENT;
+		}
+		eaddr = sym.st_value;
+	}
+	tp->offset = (unsigned long)(paddr - eaddr);
+	tp->address = (unsigned long)paddr;
+	tp->symbol = strdup(symbol);
+	if (!tp->symbol)
+		return -ENOMEM;
+
+	/* Return probe must be on the head of a subprogram */
+	if (retprobe) {
+		if (eaddr != paddr) {
+			pr_warning("Return probe must be on the head of"
+				   " a real function.\n");
+			return -EINVAL;
+		}
+		tp->retprobe = true;
+	}
+
+	return 0;
+}
+
+/* Call probe_finder callback with scope DIE */
+static int call_probe_finder(Dwarf_Die *sc_die, struct probe_finder *pf)
+{
+	Dwarf_Attribute fb_attr;
+	size_t nops;
+	int ret;
+
+	if (!sc_die) {
+		pr_err("Caller must pass a scope DIE. Program error.\n");
+		return -EINVAL;
+	}
+
+	/* If not a real subprogram, find a real one */
+	if (!die_is_func_def(sc_die)) {
+		if (!die_find_realfunc(&pf->cu_die, pf->addr, &pf->sp_die)) {
+			pr_warning("Failed to find probe point in any "
+				   "functions.\n");
+			return -ENOENT;
+		}
+	} else
+		memcpy(&pf->sp_die, sc_die, sizeof(Dwarf_Die));
+
+	/* Get the frame base attribute/ops from subprogram */
+	dwarf_attr(&pf->sp_die, DW_AT_frame_base, &fb_attr);
+	ret = dwarf_getlocation_addr(&fb_attr, pf->addr, &pf->fb_ops, &nops, 1);
+	if (ret <= 0 || nops == 0) {
+		pf->fb_ops = NULL;
+#if _ELFUTILS_PREREQ(0, 142)
+	} else if (nops == 1 && pf->fb_ops[0].atom == DW_OP_call_frame_cfa &&
+		   pf->cfi != NULL) {
+		Dwarf_Frame *frame;
+		if (dwarf_cfi_addrframe(pf->cfi, pf->addr, &frame) != 0 ||
+		    dwarf_frame_cfa(frame, &pf->fb_ops, &nops) != 0) {
+			pr_warning("Failed to get call frame on 0x%jx\n",
+				   (uintmax_t)pf->addr);
+			return -ENOENT;
+		}
+#endif
+	}
+
+	/* Call finder's callback handler */
+	ret = pf->callback(sc_die, pf);
+
+	/* *pf->fb_ops will be cached in libdw. Don't free it. */
+	pf->fb_ops = NULL;
+
+	return ret;
+}
+
+struct find_scope_param {
+	const char *function;
+	const char *file;
+	int line;
+	int diff;
+	Dwarf_Die *die_mem;
+	bool found;
+};
+
+static int find_best_scope_cb(Dwarf_Die *fn_die, void *data)
+{
+	struct find_scope_param *fsp = data;
+	const char *file;
+	int lno;
+
+	/* Skip if declared file name does not match */
+	if (fsp->file) {
+		file = dwarf_decl_file(fn_die);
+		if (!file || strcmp(fsp->file, file) != 0)
+			return 0;
+	}
+	/* If the function name is given, that's what user expects */
+	if (fsp->function) {
+		if (die_compare_name(fn_die, fsp->function)) {
+			memcpy(fsp->die_mem, fn_die, sizeof(Dwarf_Die));
+			fsp->found = true;
+			return 1;
+		}
+	} else {
+		/* With the line number, find the nearest declared DIE */
+		dwarf_decl_line(fn_die, &lno);
+		if (lno < fsp->line && fsp->diff > fsp->line - lno) {
+			/* Keep a candidate and continue */
+			fsp->diff = fsp->line - lno;
+			memcpy(fsp->die_mem, fn_die, sizeof(Dwarf_Die));
+			fsp->found = true;
+		}
+	}
+	return 0;
+}
+
+/* Find an appropriate scope fits to given conditions */
+static Dwarf_Die *find_best_scope(struct probe_finder *pf, Dwarf_Die *die_mem)
+{
+	struct find_scope_param fsp = {
+		.function = pf->pev->point.function,
+		.file = pf->fname,
+		.line = pf->lno,
+		.diff = INT_MAX,
+		.die_mem = die_mem,
+		.found = false,
+	};
+
+	cu_walk_functions_at(&pf->cu_die, pf->addr, find_best_scope_cb, &fsp);
+
+	return fsp.found ? die_mem : NULL;
+}
+
+static int probe_point_line_walker(const char *fname, int lineno,
+				   Dwarf_Addr addr, void *data)
+{
+	struct probe_finder *pf = data;
+	Dwarf_Die *sc_die, die_mem;
+	int ret;
+
+	if (lineno != pf->lno || strtailcmp(fname, pf->fname) != 0)
+		return 0;
+
+	pf->addr = addr;
+	sc_die = find_best_scope(pf, &die_mem);
+	if (!sc_die) {
+		pr_warning("Failed to find scope of probe point.\n");
+		return -ENOENT;
+	}
+
+	ret = call_probe_finder(sc_die, pf);
+
+	/* Continue if no error, because the line will be in inline function */
+	return ret < 0 ? ret : 0;
+}
+
+/* Find probe point from its line number */
+static int find_probe_point_by_line(struct probe_finder *pf)
+{
+	return die_walk_lines(&pf->cu_die, probe_point_line_walker, pf);
+}
+
+/* Find lines which match lazy pattern */
+static int find_lazy_match_lines(struct intlist *list,
+				 const char *fname, const char *pat)
+{
+	FILE *fp;
+	char *line = NULL;
+	size_t line_len;
+	ssize_t len;
+	int count = 0, linenum = 1;
+	char sbuf[STRERR_BUFSIZE];
+
+	fp = fopen(fname, "r");
+	if (!fp) {
+		pr_warning("Failed to open %s: %s\n", fname,
+			   strerror_r(errno, sbuf, sizeof(sbuf)));
+		return -errno;
+	}
+
+	while ((len = getline(&line, &line_len, fp)) > 0) {
+
+		if (line[len - 1] == '\n')
+			line[len - 1] = '\0';
+
+		if (strlazymatch(line, pat)) {
+			intlist__add(list, linenum);
+			count++;
+		}
+		linenum++;
+	}
+
+	if (ferror(fp))
+		count = -errno;
+	free(line);
+	fclose(fp);
+
+	if (count == 0)
+		pr_debug("No matched lines found in %s.\n", fname);
+	return count;
+}
+
+static int probe_point_lazy_walker(const char *fname, int lineno,
+				   Dwarf_Addr addr, void *data)
+{
+	struct probe_finder *pf = data;
+	Dwarf_Die *sc_die, die_mem;
+	int ret;
+
+	if (!intlist__has_entry(pf->lcache, lineno) ||
+	    strtailcmp(fname, pf->fname) != 0)
+		return 0;
+
+	pr_debug("Probe line found: line:%d addr:0x%llx\n",
+		 lineno, (unsigned long long)addr);
+	pf->addr = addr;
+	pf->lno = lineno;
+	sc_die = find_best_scope(pf, &die_mem);
+	if (!sc_die) {
+		pr_warning("Failed to find scope of probe point.\n");
+		return -ENOENT;
+	}
+
+	ret = call_probe_finder(sc_die, pf);
+
+	/*
+	 * Continue if no error, because the lazy pattern will match
+	 * to other lines
+	 */
+	return ret < 0 ? ret : 0;
+}
+
+/* Find probe points from lazy pattern  */
+static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf)
+{
+	int ret = 0;
+	char *fpath;
+
+	if (intlist__empty(pf->lcache)) {
+		const char *comp_dir;
+
+		comp_dir = cu_get_comp_dir(&pf->cu_die);
+		ret = get_real_path(pf->fname, comp_dir, &fpath);
+		if (ret < 0) {
+			pr_warning("Failed to find source file path.\n");
+			return ret;
+		}
+
+		/* Matching lazy line pattern */
+		ret = find_lazy_match_lines(pf->lcache, fpath,
+					    pf->pev->point.lazy_line);
+		free(fpath);
+		if (ret <= 0)
+			return ret;
+	}
+
+	return die_walk_lines(sp_die, probe_point_lazy_walker, pf);
+}
+
+static int probe_point_inline_cb(Dwarf_Die *in_die, void *data)
+{
+	struct probe_finder *pf = data;
+	struct perf_probe_point *pp = &pf->pev->point;
+	Dwarf_Addr addr;
+	int ret;
+
+	if (pp->lazy_line)
+		ret = find_probe_point_lazy(in_die, pf);
+	else {
+		/* Get probe address */
+		if (dwarf_entrypc(in_die, &addr) != 0) {
+			pr_warning("Failed to get entry address of %s.\n",
+				   dwarf_diename(in_die));
+			return -ENOENT;
+		}
+		pf->addr = addr;
+		pf->addr += pp->offset;
+		pr_debug("found inline addr: 0x%jx\n",
+			 (uintmax_t)pf->addr);
+
+		ret = call_probe_finder(in_die, pf);
+	}
+
+	return ret;
+}
+
+/* Callback parameter with return value for libdw */
+struct dwarf_callback_param {
+	void *data;
+	int retval;
+};
+
+/* Search function from function name */
+static int probe_point_search_cb(Dwarf_Die *sp_die, void *data)
+{
+	struct dwarf_callback_param *param = data;
+	struct probe_finder *pf = param->data;
+	struct perf_probe_point *pp = &pf->pev->point;
+
+	/* Check tag and diename */
+	if (!die_is_func_def(sp_die) ||
+	    !die_compare_name(sp_die, pp->function))
+		return DWARF_CB_OK;
+
+	/* Check declared file */
+	if (pp->file && strtailcmp(pp->file, dwarf_decl_file(sp_die)))
+		return DWARF_CB_OK;
+
+	pf->fname = dwarf_decl_file(sp_die);
+	if (pp->line) { /* Function relative line */
+		dwarf_decl_line(sp_die, &pf->lno);
+		pf->lno += pp->line;
+		param->retval = find_probe_point_by_line(pf);
+	} else if (die_is_func_instance(sp_die)) {
+		/* Instances always have the entry address */
+		dwarf_entrypc(sp_die, &pf->addr);
+		/* Real function */
+		if (pp->lazy_line)
+			param->retval = find_probe_point_lazy(sp_die, pf);
+		else {
+			pf->addr += pp->offset;
+			/* TODO: Check the address in this function */
+			param->retval = call_probe_finder(sp_die, pf);
+		}
+	} else
+		/* Inlined function: search instances */
+		param->retval = die_walk_instances(sp_die,
+					probe_point_inline_cb, (void *)pf);
+
+	return DWARF_CB_ABORT; /* Exit; no same symbol in this CU. */
+}
+
+static int find_probe_point_by_func(struct probe_finder *pf)
+{
+	struct dwarf_callback_param _param = {.data = (void *)pf,
+					      .retval = 0};
+	dwarf_getfuncs(&pf->cu_die, probe_point_search_cb, &_param, 0);
+	return _param.retval;
+}
+
+struct pubname_callback_param {
+	char *function;
+	char *file;
+	Dwarf_Die *cu_die;
+	Dwarf_Die *sp_die;
+	int found;
+};
+
+static int pubname_search_cb(Dwarf *dbg, Dwarf_Global *gl, void *data)
+{
+	struct pubname_callback_param *param = data;
+
+	if (dwarf_offdie(dbg, gl->die_offset, param->sp_die)) {
+		if (dwarf_tag(param->sp_die) != DW_TAG_subprogram)
+			return DWARF_CB_OK;
+
+		if (die_compare_name(param->sp_die, param->function)) {
+			if (!dwarf_offdie(dbg, gl->cu_offset, param->cu_die))
+				return DWARF_CB_OK;
+
+			if (param->file &&
+			    strtailcmp(param->file, dwarf_decl_file(param->sp_die)))
+				return DWARF_CB_OK;
+
+			param->found = 1;
+			return DWARF_CB_ABORT;
+		}
+	}
+
+	return DWARF_CB_OK;
+}
+
+/* Find probe points from debuginfo */
+static int debuginfo__find_probes(struct debuginfo *dbg,
+				  struct probe_finder *pf)
+{
+	struct perf_probe_point *pp = &pf->pev->point;
+	Dwarf_Off off, noff;
+	size_t cuhl;
+	Dwarf_Die *diep;
+	int ret = 0;
+
+#if _ELFUTILS_PREREQ(0, 142)
+	Elf *elf;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+
+	/* Get the call frame information from this dwarf */
+	elf = dwarf_getelf(dbg->dbg);
+	if (elf == NULL)
+		return -EINVAL;
+
+	if (gelf_getehdr(elf, &ehdr) == NULL)
+		return -EINVAL;
+
+	if (elf_section_by_name(elf, &ehdr, &shdr, ".eh_frame", NULL) &&
+	    shdr.sh_type == SHT_PROGBITS) {
+		pf->cfi = dwarf_getcfi_elf(elf);
+	} else {
+		pf->cfi = dwarf_getcfi(dbg->dbg);
+	}
+#endif
+
+	off = 0;
+	pf->lcache = intlist__new(NULL);
+	if (!pf->lcache)
+		return -ENOMEM;
+
+	/* Fastpath: lookup by function name from .debug_pubnames section */
+	if (pp->function) {
+		struct pubname_callback_param pubname_param = {
+			.function = pp->function,
+			.file	  = pp->file,
+			.cu_die	  = &pf->cu_die,
+			.sp_die	  = &pf->sp_die,
+			.found	  = 0,
+		};
+		struct dwarf_callback_param probe_param = {
+			.data = pf,
+		};
+
+		dwarf_getpubnames(dbg->dbg, pubname_search_cb,
+				  &pubname_param, 0);
+		if (pubname_param.found) {
+			ret = probe_point_search_cb(&pf->sp_die, &probe_param);
+			if (ret)
+				goto found;
+		}
+	}
+
+	/* Loop on CUs (Compilation Unit) */
+	while (!dwarf_nextcu(dbg->dbg, off, &noff, &cuhl, NULL, NULL, NULL)) {
+		/* Get the DIE(Debugging Information Entry) of this CU */
+		diep = dwarf_offdie(dbg->dbg, off + cuhl, &pf->cu_die);
+		if (!diep)
+			continue;
+
+		/* Check if target file is included. */
+		if (pp->file)
+			pf->fname = cu_find_realpath(&pf->cu_die, pp->file);
+		else
+			pf->fname = NULL;
+
+		if (!pp->file || pf->fname) {
+			if (pp->function)
+				ret = find_probe_point_by_func(pf);
+			else if (pp->lazy_line)
+				ret = find_probe_point_lazy(&pf->cu_die, pf);
+			else {
+				pf->lno = pp->line;
+				ret = find_probe_point_by_line(pf);
+			}
+			if (ret < 0)
+				break;
+		}
+		off = noff;
+	}
+
+found:
+	intlist__delete(pf->lcache);
+	pf->lcache = NULL;
+
+	return ret;
+}
+
+struct local_vars_finder {
+	struct probe_finder *pf;
+	struct perf_probe_arg *args;
+	int max_args;
+	int nargs;
+	int ret;
+};
+
+/* Collect available variables in this scope */
+static int copy_variables_cb(Dwarf_Die *die_mem, void *data)
+{
+	struct local_vars_finder *vf = data;
+	struct probe_finder *pf = vf->pf;
+	int tag;
+
+	tag = dwarf_tag(die_mem);
+	if (tag == DW_TAG_formal_parameter ||
+	    tag == DW_TAG_variable) {
+		if (convert_variable_location(die_mem, vf->pf->addr,
+					      vf->pf->fb_ops, &pf->sp_die,
+					      NULL) == 0) {
+			vf->args[vf->nargs].var = (char *)dwarf_diename(die_mem);
+			if (vf->args[vf->nargs].var == NULL) {
+				vf->ret = -ENOMEM;
+				return DIE_FIND_CB_END;
+			}
+			pr_debug(" %s", vf->args[vf->nargs].var);
+			vf->nargs++;
+		}
+	}
+
+	if (dwarf_haspc(die_mem, vf->pf->addr))
+		return DIE_FIND_CB_CONTINUE;
+	else
+		return DIE_FIND_CB_SIBLING;
+}
+
+static int expand_probe_args(Dwarf_Die *sc_die, struct probe_finder *pf,
+			     struct perf_probe_arg *args)
+{
+	Dwarf_Die die_mem;
+	int i;
+	int n = 0;
+	struct local_vars_finder vf = {.pf = pf, .args = args,
+				.max_args = MAX_PROBE_ARGS, .ret = 0};
+
+	for (i = 0; i < pf->pev->nargs; i++) {
+		/* var never be NULL */
+		if (strcmp(pf->pev->args[i].var, "$vars") == 0) {
+			pr_debug("Expanding $vars into:");
+			vf.nargs = n;
+			/* Special local variables */
+			die_find_child(sc_die, copy_variables_cb, (void *)&vf,
+				       &die_mem);
+			pr_debug(" (%d)\n", vf.nargs - n);
+			if (vf.ret < 0)
+				return vf.ret;
+			n = vf.nargs;
+		} else {
+			/* Copy normal argument */
+			args[n] = pf->pev->args[i];
+			n++;
+		}
+	}
+	return n;
+}
+
+/* Add a found probe point into trace event list */
+static int add_probe_trace_event(Dwarf_Die *sc_die, struct probe_finder *pf)
+{
+	struct trace_event_finder *tf =
+			container_of(pf, struct trace_event_finder, pf);
+	struct probe_trace_event *tev;
+	struct perf_probe_arg *args;
+	int ret, i;
+
+	/* Check number of tevs */
+	if (tf->ntevs == tf->max_tevs) {
+		pr_warning("Too many( > %d) probe point found.\n",
+			   tf->max_tevs);
+		return -ERANGE;
+	}
+	tev = &tf->tevs[tf->ntevs++];
+
+	/* Trace point should be converted from subprogram DIE */
+	ret = convert_to_trace_point(&pf->sp_die, tf->mod, pf->addr,
+				     pf->pev->point.retprobe, &tev->point);
+	if (ret < 0)
+		return ret;
+
+	pr_debug("Probe point found: %s+%lu\n", tev->point.symbol,
+		 tev->point.offset);
+
+	/* Expand special probe argument if exist */
+	args = zalloc(sizeof(struct perf_probe_arg) * MAX_PROBE_ARGS);
+	if (args == NULL)
+		return -ENOMEM;
+
+	ret = expand_probe_args(sc_die, pf, args);
+	if (ret < 0)
+		goto end;
+
+	tev->nargs = ret;
+	tev->args = zalloc(sizeof(struct probe_trace_arg) * tev->nargs);
+	if (tev->args == NULL) {
+		ret = -ENOMEM;
+		goto end;
+	}
+
+	/* Find each argument */
+	for (i = 0; i < tev->nargs; i++) {
+		pf->pvar = &args[i];
+		pf->tvar = &tev->args[i];
+		/* Variable should be found from scope DIE */
+		ret = find_variable(sc_die, pf);
+		if (ret != 0)
+			break;
+	}
+
+end:
+	free(args);
+	return ret;
+}
+
+/* Find probe_trace_events specified by perf_probe_event from debuginfo */
+int debuginfo__find_trace_events(struct debuginfo *dbg,
+				 struct perf_probe_event *pev,
+				 struct probe_trace_event **tevs, int max_tevs)
+{
+	struct trace_event_finder tf = {
+			.pf = {.pev = pev, .callback = add_probe_trace_event},
+			.mod = dbg->mod, .max_tevs = max_tevs};
+	int ret;
+
+	/* Allocate result tevs array */
+	*tevs = zalloc(sizeof(struct probe_trace_event) * max_tevs);
+	if (*tevs == NULL)
+		return -ENOMEM;
+
+	tf.tevs = *tevs;
+	tf.ntevs = 0;
+
+	ret = debuginfo__find_probes(dbg, &tf.pf);
+	if (ret < 0) {
+		zfree(tevs);
+		return ret;
+	}
+
+	return (ret < 0) ? ret : tf.ntevs;
+}
+
+#define MAX_VAR_LEN 64
+
+/* Collect available variables in this scope */
+static int collect_variables_cb(Dwarf_Die *die_mem, void *data)
+{
+	struct available_var_finder *af = data;
+	struct variable_list *vl;
+	char buf[MAX_VAR_LEN];
+	int tag, ret;
+
+	vl = &af->vls[af->nvls - 1];
+
+	tag = dwarf_tag(die_mem);
+	if (tag == DW_TAG_formal_parameter ||
+	    tag == DW_TAG_variable) {
+		ret = convert_variable_location(die_mem, af->pf.addr,
+						af->pf.fb_ops, &af->pf.sp_die,
+						NULL);
+		if (ret == 0) {
+			ret = die_get_varname(die_mem, buf, MAX_VAR_LEN);
+			pr_debug2("Add new var: %s\n", buf);
+			if (ret > 0)
+				strlist__add(vl->vars, buf);
+		}
+	}
+
+	if (af->child && dwarf_haspc(die_mem, af->pf.addr))
+		return DIE_FIND_CB_CONTINUE;
+	else
+		return DIE_FIND_CB_SIBLING;
+}
+
+/* Add a found vars into available variables list */
+static int add_available_vars(Dwarf_Die *sc_die, struct probe_finder *pf)
+{
+	struct available_var_finder *af =
+			container_of(pf, struct available_var_finder, pf);
+	struct variable_list *vl;
+	Dwarf_Die die_mem;
+	int ret;
+
+	/* Check number of tevs */
+	if (af->nvls == af->max_vls) {
+		pr_warning("Too many( > %d) probe point found.\n", af->max_vls);
+		return -ERANGE;
+	}
+	vl = &af->vls[af->nvls++];
+
+	/* Trace point should be converted from subprogram DIE */
+	ret = convert_to_trace_point(&pf->sp_die, af->mod, pf->addr,
+				     pf->pev->point.retprobe, &vl->point);
+	if (ret < 0)
+		return ret;
+
+	pr_debug("Probe point found: %s+%lu\n", vl->point.symbol,
+		 vl->point.offset);
+
+	/* Find local variables */
+	vl->vars = strlist__new(true, NULL);
+	if (vl->vars == NULL)
+		return -ENOMEM;
+	af->child = true;
+	die_find_child(sc_die, collect_variables_cb, (void *)af, &die_mem);
+
+	/* Find external variables */
+	if (!af->externs)
+		goto out;
+	/* Don't need to search child DIE for externs. */
+	af->child = false;
+	die_find_child(&pf->cu_die, collect_variables_cb, (void *)af, &die_mem);
+
+out:
+	if (strlist__empty(vl->vars)) {
+		strlist__delete(vl->vars);
+		vl->vars = NULL;
+	}
+
+	return ret;
+}
+
+/*
+ * Find available variables at given probe point
+ * Return the number of found probe points. Return 0 if there is no
+ * matched probe point. Return <0 if an error occurs.
+ */
+int debuginfo__find_available_vars_at(struct debuginfo *dbg,
+				      struct perf_probe_event *pev,
+				      struct variable_list **vls,
+				      int max_vls, bool externs)
+{
+	struct available_var_finder af = {
+			.pf = {.pev = pev, .callback = add_available_vars},
+			.mod = dbg->mod,
+			.max_vls = max_vls, .externs = externs};
+	int ret;
+
+	/* Allocate result vls array */
+	*vls = zalloc(sizeof(struct variable_list) * max_vls);
+	if (*vls == NULL)
+		return -ENOMEM;
+
+	af.vls = *vls;
+	af.nvls = 0;
+
+	ret = debuginfo__find_probes(dbg, &af.pf);
+	if (ret < 0) {
+		/* Free vlist for error */
+		while (af.nvls--) {
+			zfree(&af.vls[af.nvls].point.symbol);
+			strlist__delete(af.vls[af.nvls].vars);
+		}
+		zfree(vls);
+		return ret;
+	}
+
+	return (ret < 0) ? ret : af.nvls;
+}
+
+/* Reverse search */
+int debuginfo__find_probe_point(struct debuginfo *dbg, unsigned long addr,
+				struct perf_probe_point *ppt)
+{
+	Dwarf_Die cudie, spdie, indie;
+	Dwarf_Addr _addr = 0, baseaddr = 0;
+	const char *fname = NULL, *func = NULL, *basefunc = NULL, *tmp;
+	int baseline = 0, lineno = 0, ret = 0;
+
+	/* Find cu die */
+	if (!dwarf_addrdie(dbg->dbg, (Dwarf_Addr)addr, &cudie)) {
+		pr_warning("Failed to find debug information for address %lx\n",
+			   addr);
+		ret = -EINVAL;
+		goto end;
+	}
+
+	/* Find a corresponding line (filename and lineno) */
+	cu_find_lineinfo(&cudie, addr, &fname, &lineno);
+	/* Don't care whether it failed or not */
+
+	/* Find a corresponding function (name, baseline and baseaddr) */
+	if (die_find_realfunc(&cudie, (Dwarf_Addr)addr, &spdie)) {
+		/* Get function entry information */
+		func = basefunc = dwarf_diename(&spdie);
+		if (!func ||
+		    dwarf_entrypc(&spdie, &baseaddr) != 0 ||
+		    dwarf_decl_line(&spdie, &baseline) != 0) {
+			lineno = 0;
+			goto post;
+		}
+
+		fname = dwarf_decl_file(&spdie);
+		if (addr == (unsigned long)baseaddr) {
+			/* Function entry - Relative line number is 0 */
+			lineno = baseline;
+			goto post;
+		}
+
+		/* Track down the inline functions step by step */
+		while (die_find_top_inlinefunc(&spdie, (Dwarf_Addr)addr,
+						&indie)) {
+			/* There is an inline function */
+			if (dwarf_entrypc(&indie, &_addr) == 0 &&
+			    _addr == addr) {
+				/*
+				 * addr is at an inline function entry.
+				 * In this case, lineno should be the call-site
+				 * line number. (overwrite lineinfo)
+				 */
+				lineno = die_get_call_lineno(&indie);
+				fname = die_get_call_file(&indie);
+				break;
+			} else {
+				/*
+				 * addr is in an inline function body.
+				 * Since lineno points one of the lines
+				 * of the inline function, baseline should
+				 * be the entry line of the inline function.
+				 */
+				tmp = dwarf_diename(&indie);
+				if (!tmp ||
+				    dwarf_decl_line(&indie, &baseline) != 0)
+					break;
+				func = tmp;
+				spdie = indie;
+			}
+		}
+		/* Verify the lineno and baseline are in a same file */
+		tmp = dwarf_decl_file(&spdie);
+		if (!tmp || strcmp(tmp, fname) != 0)
+			lineno = 0;
+	}
+
+post:
+	/* Make a relative line number or an offset */
+	if (lineno)
+		ppt->line = lineno - baseline;
+	else if (basefunc) {
+		ppt->offset = addr - (unsigned long)baseaddr;
+		func = basefunc;
+	}
+
+	/* Duplicate strings */
+	if (func) {
+		ppt->function = strdup(func);
+		if (ppt->function == NULL) {
+			ret = -ENOMEM;
+			goto end;
+		}
+	}
+	if (fname) {
+		ppt->file = strdup(fname);
+		if (ppt->file == NULL) {
+			zfree(&ppt->function);
+			ret = -ENOMEM;
+			goto end;
+		}
+	}
+end:
+	if (ret == 0 && (fname || func))
+		ret = 1;	/* Found a point */
+	return ret;
+}
+
+/* Add a line and store the src path */
+static int line_range_add_line(const char *src, unsigned int lineno,
+			       struct line_range *lr)
+{
+	/* Copy source path */
+	if (!lr->path) {
+		lr->path = strdup(src);
+		if (lr->path == NULL)
+			return -ENOMEM;
+	}
+	return intlist__add(lr->line_list, lineno);
+}
+
+static int line_range_walk_cb(const char *fname, int lineno,
+			      Dwarf_Addr addr __maybe_unused,
+			      void *data)
+{
+	struct line_finder *lf = data;
+	int err;
+
+	if ((strtailcmp(fname, lf->fname) != 0) ||
+	    (lf->lno_s > lineno || lf->lno_e < lineno))
+		return 0;
+
+	err = line_range_add_line(fname, lineno, lf->lr);
+	if (err < 0 && err != -EEXIST)
+		return err;
+
+	return 0;
+}
+
+/* Find line range from its line number */
+static int find_line_range_by_line(Dwarf_Die *sp_die, struct line_finder *lf)
+{
+	int ret;
+
+	ret = die_walk_lines(sp_die ?: &lf->cu_die, line_range_walk_cb, lf);
+
+	/* Update status */
+	if (ret >= 0)
+		if (!intlist__empty(lf->lr->line_list))
+			ret = lf->found = 1;
+		else
+			ret = 0;	/* Lines are not found */
+	else {
+		zfree(&lf->lr->path);
+	}
+	return ret;
+}
+
+static int line_range_inline_cb(Dwarf_Die *in_die, void *data)
+{
+	int ret = find_line_range_by_line(in_die, data);
+
+	/*
+	 * We have to check all instances of inlined function, because
+	 * some execution paths can be optimized out depends on the
+	 * function argument of instances. However, if an error occurs,
+	 * it should be handled by the caller.
+	 */
+	return ret < 0 ? ret : 0;
+}
+
+/* Search function definition from function name */
+static int line_range_search_cb(Dwarf_Die *sp_die, void *data)
+{
+	struct dwarf_callback_param *param = data;
+	struct line_finder *lf = param->data;
+	struct line_range *lr = lf->lr;
+
+	/* Check declared file */
+	if (lr->file && strtailcmp(lr->file, dwarf_decl_file(sp_die)))
+		return DWARF_CB_OK;
+
+	if (die_is_func_def(sp_die) &&
+	    die_compare_name(sp_die, lr->function)) {
+		lf->fname = dwarf_decl_file(sp_die);
+		dwarf_decl_line(sp_die, &lr->offset);
+		pr_debug("fname: %s, lineno:%d\n", lf->fname, lr->offset);
+		lf->lno_s = lr->offset + lr->start;
+		if (lf->lno_s < 0)	/* Overflow */
+			lf->lno_s = INT_MAX;
+		lf->lno_e = lr->offset + lr->end;
+		if (lf->lno_e < 0)	/* Overflow */
+			lf->lno_e = INT_MAX;
+		pr_debug("New line range: %d to %d\n", lf->lno_s, lf->lno_e);
+		lr->start = lf->lno_s;
+		lr->end = lf->lno_e;
+		if (!die_is_func_instance(sp_die))
+			param->retval = die_walk_instances(sp_die,
+						line_range_inline_cb, lf);
+		else
+			param->retval = find_line_range_by_line(sp_die, lf);
+		return DWARF_CB_ABORT;
+	}
+	return DWARF_CB_OK;
+}
+
+static int find_line_range_by_func(struct line_finder *lf)
+{
+	struct dwarf_callback_param param = {.data = (void *)lf, .retval = 0};
+	dwarf_getfuncs(&lf->cu_die, line_range_search_cb, &param, 0);
+	return param.retval;
+}
+
+int debuginfo__find_line_range(struct debuginfo *dbg, struct line_range *lr)
+{
+	struct line_finder lf = {.lr = lr, .found = 0};
+	int ret = 0;
+	Dwarf_Off off = 0, noff;
+	size_t cuhl;
+	Dwarf_Die *diep;
+	const char *comp_dir;
+
+	/* Fastpath: lookup by function name from .debug_pubnames section */
+	if (lr->function) {
+		struct pubname_callback_param pubname_param = {
+			.function = lr->function, .file = lr->file,
+			.cu_die = &lf.cu_die, .sp_die = &lf.sp_die, .found = 0};
+		struct dwarf_callback_param line_range_param = {
+			.data = (void *)&lf, .retval = 0};
+
+		dwarf_getpubnames(dbg->dbg, pubname_search_cb,
+				  &pubname_param, 0);
+		if (pubname_param.found) {
+			line_range_search_cb(&lf.sp_die, &line_range_param);
+			if (lf.found)
+				goto found;
+		}
+	}
+
+	/* Loop on CUs (Compilation Unit) */
+	while (!lf.found && ret >= 0) {
+		if (dwarf_nextcu(dbg->dbg, off, &noff, &cuhl,
+				 NULL, NULL, NULL) != 0)
+			break;
+
+		/* Get the DIE(Debugging Information Entry) of this CU */
+		diep = dwarf_offdie(dbg->dbg, off + cuhl, &lf.cu_die);
+		if (!diep)
+			continue;
+
+		/* Check if target file is included. */
+		if (lr->file)
+			lf.fname = cu_find_realpath(&lf.cu_die, lr->file);
+		else
+			lf.fname = 0;
+
+		if (!lr->file || lf.fname) {
+			if (lr->function)
+				ret = find_line_range_by_func(&lf);
+			else {
+				lf.lno_s = lr->start;
+				lf.lno_e = lr->end;
+				ret = find_line_range_by_line(NULL, &lf);
+			}
+		}
+		off = noff;
+	}
+
+found:
+	/* Store comp_dir */
+	if (lf.found) {
+		comp_dir = cu_get_comp_dir(&lf.cu_die);
+		if (comp_dir) {
+			lr->comp_dir = strdup(comp_dir);
+			if (!lr->comp_dir)
+				ret = -ENOMEM;
+		}
+	}
+
+	pr_debug("path: %s\n", lr->path);
+	return (ret < 0) ? ret : lf.found;
+}
+
+/*
+ * Find a src file from a DWARF tag path. Prepend optional source path prefix
+ * and chop off leading directories that do not exist. Result is passed back as
+ * a newly allocated path on success.
+ * Return 0 if file was found and readable, -errno otherwise.
+ */
+int get_real_path(const char *raw_path, const char *comp_dir,
+			 char **new_path)
+{
+	const char *prefix = symbol_conf.source_prefix;
+
+	if (!prefix) {
+		if (raw_path[0] != '/' && comp_dir)
+			/* If not an absolute path, try to use comp_dir */
+			prefix = comp_dir;
+		else {
+			if (access(raw_path, R_OK) == 0) {
+				*new_path = strdup(raw_path);
+				return *new_path ? 0 : -ENOMEM;
+			} else
+				return -errno;
+		}
+	}
+
+	*new_path = malloc((strlen(prefix) + strlen(raw_path) + 2));
+	if (!*new_path)
+		return -ENOMEM;
+
+	for (;;) {
+		sprintf(*new_path, "%s/%s", prefix, raw_path);
+
+		if (access(*new_path, R_OK) == 0)
+			return 0;
+
+		if (!symbol_conf.source_prefix) {
+			/* In case of searching comp_dir, don't retry */
+			zfree(new_path);
+			return -errno;
+		}
+
+		switch (errno) {
+		case ENAMETOOLONG:
+		case ENOENT:
+		case EROFS:
+		case EFAULT:
+			raw_path = strchr(++raw_path, '/');
+			if (!raw_path) {
+				zfree(new_path);
+				return -ENOENT;
+			}
+			continue;
+
+		default:
+			zfree(new_path);
+			return -errno;
+		}
+	}
+}
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h
new file mode 100644
index 000000000..ebf8c8c81
--- /dev/null
+++ b/tools/perf/util/probe-finder.h
@@ -0,0 +1,116 @@
+#ifndef _PROBE_FINDER_H
+#define _PROBE_FINDER_H
+
+#include <stdbool.h>
+#include "util.h"
+#include "intlist.h"
+#include "probe-event.h"
+
+#define MAX_PROBE_BUFFER	1024
+#define MAX_PROBES		 128
+#define MAX_PROBE_ARGS		 128
+
+static inline int is_c_varname(const char *name)
+{
+	/* TODO */
+	return isalpha(name[0]) || name[0] == '_';
+}
+
+#ifdef HAVE_DWARF_SUPPORT
+
+#include "dwarf-aux.h"
+
+/* TODO: export debuginfo data structure even if no dwarf support */
+
+/* debug information structure */
+struct debuginfo {
+	Dwarf		*dbg;
+	Dwfl_Module	*mod;
+	Dwfl		*dwfl;
+	Dwarf_Addr	bias;
+};
+
+/* This also tries to open distro debuginfo */
+extern struct debuginfo *debuginfo__new(const char *path);
+extern void debuginfo__delete(struct debuginfo *dbg);
+
+/* Find probe_trace_events specified by perf_probe_event from debuginfo */
+extern int debuginfo__find_trace_events(struct debuginfo *dbg,
+					struct perf_probe_event *pev,
+					struct probe_trace_event **tevs,
+					int max_tevs);
+
+/* Find a perf_probe_point from debuginfo */
+extern int debuginfo__find_probe_point(struct debuginfo *dbg,
+				       unsigned long addr,
+				       struct perf_probe_point *ppt);
+
+/* Find a line range */
+extern int debuginfo__find_line_range(struct debuginfo *dbg,
+				      struct line_range *lr);
+
+/* Find available variables */
+extern int debuginfo__find_available_vars_at(struct debuginfo *dbg,
+					     struct perf_probe_event *pev,
+					     struct variable_list **vls,
+					     int max_points, bool externs);
+
+/* Find a src file from a DWARF tag path */
+int get_real_path(const char *raw_path, const char *comp_dir,
+			 char **new_path);
+
+struct probe_finder {
+	struct perf_probe_event	*pev;		/* Target probe event */
+
+	/* Callback when a probe point is found */
+	int (*callback)(Dwarf_Die *sc_die, struct probe_finder *pf);
+
+	/* For function searching */
+	int			lno;		/* Line number */
+	Dwarf_Addr		addr;		/* Address */
+	const char		*fname;		/* Real file name */
+	Dwarf_Die		cu_die;		/* Current CU */
+	Dwarf_Die		sp_die;
+	struct intlist		*lcache;	/* Line cache for lazy match */
+
+	/* For variable searching */
+#if _ELFUTILS_PREREQ(0, 142)
+	Dwarf_CFI		*cfi;		/* Call Frame Information */
+#endif
+	Dwarf_Op		*fb_ops;	/* Frame base attribute */
+	struct perf_probe_arg	*pvar;		/* Current target variable */
+	struct probe_trace_arg	*tvar;		/* Current result variable */
+};
+
+struct trace_event_finder {
+	struct probe_finder	pf;
+	Dwfl_Module		*mod;		/* For solving symbols */
+	struct probe_trace_event *tevs;		/* Found trace events */
+	int			ntevs;		/* Number of trace events */
+	int			max_tevs;	/* Max number of trace events */
+};
+
+struct available_var_finder {
+	struct probe_finder	pf;
+	Dwfl_Module		*mod;		/* For solving symbols */
+	struct variable_list	*vls;		/* Found variable lists */
+	int			nvls;		/* Number of variable lists */
+	int			max_vls;	/* Max no. of variable lists */
+	bool			externs;	/* Find external vars too */
+	bool			child;		/* Search child scopes */
+};
+
+struct line_finder {
+	struct line_range	*lr;		/* Target line range */
+
+	const char		*fname;		/* File name */
+	int			lno_s;		/* Start line number */
+	int			lno_e;		/* End line number */
+	Dwarf_Die		cu_die;		/* Current CU */
+	Dwarf_Die		sp_die;
+	int			found;
+};
+
+#endif /* HAVE_DWARF_SUPPORT */
+
+#endif /*_PROBE_FINDER_H */
diff --git a/tools/perf/util/pstack.c b/tools/perf/util/pstack.c
new file mode 100644
index 000000000..a126e6cc6
--- /dev/null
+++ b/tools/perf/util/pstack.c
@@ -0,0 +1,76 @@
+/*
+ * Simple pointer stack
+ *
+ * (c) 2010 Arnaldo Carvalho de Melo <acme@redhat.com>
+ */
+
+#include "util.h"
+#include "pstack.h"
+#include "debug.h"
+#include <linux/kernel.h>
+#include <stdlib.h>
+
+struct pstack {
+	unsigned short	top;
+	unsigned short	max_nr_entries;
+	void		*entries[0];
+};
+
+struct pstack *pstack__new(unsigned short max_nr_entries)
+{
+	struct pstack *pstack = zalloc((sizeof(*pstack) +
+				       max_nr_entries * sizeof(void *)));
+	if (pstack != NULL)
+		pstack->max_nr_entries = max_nr_entries;
+	return pstack;
+}
+
+void pstack__delete(struct pstack *pstack)
+{
+	free(pstack);
+}
+
+bool pstack__empty(const struct pstack *pstack)
+{
+	return pstack->top == 0;
+}
+
+void pstack__remove(struct pstack *pstack, void *key)
+{
+	unsigned short i = pstack->top, last_index = pstack->top - 1;
+
+	while (i-- != 0) {
+		if (pstack->entries[i] == key) {
+			if (i < last_index)
+				memmove(pstack->entries + i,
+					pstack->entries + i + 1,
+					(last_index - i) * sizeof(void *));
+			--pstack->top;
+			return;
+		}
+	}
+	pr_err("%s: %p not on the pstack!\n", __func__, key);
+}
+
+void pstack__push(struct pstack *pstack, void *key)
+{
+	if (pstack->top == pstack->max_nr_entries) {
+		pr_err("%s: top=%d, overflow!\n", __func__, pstack->top);
+		return;
+	}
+	pstack->entries[pstack->top++] = key;
+}
+
+void *pstack__pop(struct pstack *pstack)
+{
+	void *ret;
+
+	if (pstack->top == 0) {
+		pr_err("%s: underflow!\n", __func__);
+		return NULL;
+	}
+
+	ret = pstack->entries[--pstack->top];
+	pstack->entries[pstack->top] = NULL;
+	return ret;
+}
diff --git a/tools/perf/util/pstack.h b/tools/perf/util/pstack.h
new file mode 100644
index 000000000..c3cb6584d
--- /dev/null
+++ b/tools/perf/util/pstack.h
@@ -0,0 +1,14 @@
+#ifndef _PERF_PSTACK_
+#define _PERF_PSTACK_
+
+#include <stdbool.h>
+
+struct pstack;
+struct pstack *pstack__new(unsigned short max_nr_entries);
+void pstack__delete(struct pstack *pstack);
+bool pstack__empty(const struct pstack *pstack);
+void pstack__remove(struct pstack *pstack, void *key);
+void pstack__push(struct pstack *pstack, void *key);
+void *pstack__pop(struct pstack *pstack);
+
+#endif /* _PERF_PSTACK_ */
diff --git a/tools/perf/util/python-ext-sources b/tools/perf/util/python-ext-sources
new file mode 100644
index 000000000..4d28624a1
--- /dev/null
+++ b/tools/perf/util/python-ext-sources
@@ -0,0 +1,21 @@
+#
+# List of files needed by perf python extension
+#
+# Each source file must be placed on its own line so that it can be
+# processed by Makefile and util/setup.py accordingly.
+#
+
+util/python.c
+util/ctype.c
+util/evlist.c
+util/evsel.c
+util/cpumap.c
+../../lib/hweight.c
+util/thread_map.c
+util/util.c
+util/xyarray.c
+util/cgroup.c
+util/rblist.c
+util/strlist.c
+util/trace-event.c
+../../lib/rbtree.c
diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c
new file mode 100644
index 000000000..d906d0ad5
--- /dev/null
+++ b/tools/perf/util/python.c
@@ -0,0 +1,1074 @@
+#include <Python.h>
+#include <structmember.h>
+#include <inttypes.h>
+#include <poll.h>
+#include "evlist.h"
+#include "evsel.h"
+#include "event.h"
+#include "cpumap.h"
+#include "thread_map.h"
+
+/*
+ * Support debug printing even though util/debug.c is not linked.  That means
+ * implementing 'verbose' and 'eprintf'.
+ */
+int verbose;
+
+int eprintf(int level, int var, const char *fmt, ...)
+{
+	va_list args;
+	int ret = 0;
+
+	if (var >= level) {
+		va_start(args, fmt);
+		ret = vfprintf(stderr, fmt, args);
+		va_end(args);
+	}
+
+	return ret;
+}
+
+/* Define PyVarObject_HEAD_INIT for python 2.5 */
+#ifndef PyVarObject_HEAD_INIT
+# define PyVarObject_HEAD_INIT(type, size) PyObject_HEAD_INIT(type) size,
+#endif
+
+PyMODINIT_FUNC initperf(void);
+
+#define member_def(type, member, ptype, help) \
+	{ #member, ptype, \
+	  offsetof(struct pyrf_event, event) + offsetof(struct type, member), \
+	  0, help }
+
+#define sample_member_def(name, member, ptype, help) \
+	{ #name, ptype, \
+	  offsetof(struct pyrf_event, sample) + offsetof(struct perf_sample, member), \
+	  0, help }
+
+struct pyrf_event {
+	PyObject_HEAD
+	struct perf_sample sample;
+	union perf_event   event;
+};
+
+#define sample_members \
+	sample_member_def(sample_ip, ip, T_ULONGLONG, "event type"),			 \
+	sample_member_def(sample_pid, pid, T_INT, "event pid"),			 \
+	sample_member_def(sample_tid, tid, T_INT, "event tid"),			 \
+	sample_member_def(sample_time, time, T_ULONGLONG, "event timestamp"),		 \
+	sample_member_def(sample_addr, addr, T_ULONGLONG, "event addr"),		 \
+	sample_member_def(sample_id, id, T_ULONGLONG, "event id"),			 \
+	sample_member_def(sample_stream_id, stream_id, T_ULONGLONG, "event stream id"), \
+	sample_member_def(sample_period, period, T_ULONGLONG, "event period"),		 \
+	sample_member_def(sample_cpu, cpu, T_UINT, "event cpu"),
+
+static char pyrf_mmap_event__doc[] = PyDoc_STR("perf mmap event object.");
+
+static PyMemberDef pyrf_mmap_event__members[] = {
+	sample_members
+	member_def(perf_event_header, type, T_UINT, "event type"),
+	member_def(mmap_event, pid, T_UINT, "event pid"),
+	member_def(mmap_event, tid, T_UINT, "event tid"),
+	member_def(mmap_event, start, T_ULONGLONG, "start of the map"),
+	member_def(mmap_event, len, T_ULONGLONG, "map length"),
+	member_def(mmap_event, pgoff, T_ULONGLONG, "page offset"),
+	member_def(mmap_event, filename, T_STRING_INPLACE, "backing store"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_mmap_event__repr(struct pyrf_event *pevent)
+{
+	PyObject *ret;
+	char *s;
+
+	if (asprintf(&s, "{ type: mmap, pid: %u, tid: %u, start: %#" PRIx64 ", "
+			 "length: %#" PRIx64 ", offset: %#" PRIx64 ", "
+			 "filename: %s }",
+		     pevent->event.mmap.pid, pevent->event.mmap.tid,
+		     pevent->event.mmap.start, pevent->event.mmap.len,
+		     pevent->event.mmap.pgoff, pevent->event.mmap.filename) < 0) {
+		ret = PyErr_NoMemory();
+	} else {
+		ret = PyString_FromString(s);
+		free(s);
+	}
+	return ret;
+}
+
+static PyTypeObject pyrf_mmap_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.mmap_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_mmap_event__doc,
+	.tp_members	= pyrf_mmap_event__members,
+	.tp_repr	= (reprfunc)pyrf_mmap_event__repr,
+};
+
+static char pyrf_task_event__doc[] = PyDoc_STR("perf task (fork/exit) event object.");
+
+static PyMemberDef pyrf_task_event__members[] = {
+	sample_members
+	member_def(perf_event_header, type, T_UINT, "event type"),
+	member_def(fork_event, pid, T_UINT, "event pid"),
+	member_def(fork_event, ppid, T_UINT, "event ppid"),
+	member_def(fork_event, tid, T_UINT, "event tid"),
+	member_def(fork_event, ptid, T_UINT, "event ptid"),
+	member_def(fork_event, time, T_ULONGLONG, "timestamp"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_task_event__repr(struct pyrf_event *pevent)
+{
+	return PyString_FromFormat("{ type: %s, pid: %u, ppid: %u, tid: %u, "
+				   "ptid: %u, time: %" PRIu64 "}",
+				   pevent->event.header.type == PERF_RECORD_FORK ? "fork" : "exit",
+				   pevent->event.fork.pid,
+				   pevent->event.fork.ppid,
+				   pevent->event.fork.tid,
+				   pevent->event.fork.ptid,
+				   pevent->event.fork.time);
+}
+
+static PyTypeObject pyrf_task_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.task_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_task_event__doc,
+	.tp_members	= pyrf_task_event__members,
+	.tp_repr	= (reprfunc)pyrf_task_event__repr,
+};
+
+static char pyrf_comm_event__doc[] = PyDoc_STR("perf comm event object.");
+
+static PyMemberDef pyrf_comm_event__members[] = {
+	sample_members
+	member_def(perf_event_header, type, T_UINT, "event type"),
+	member_def(comm_event, pid, T_UINT, "event pid"),
+	member_def(comm_event, tid, T_UINT, "event tid"),
+	member_def(comm_event, comm, T_STRING_INPLACE, "process name"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_comm_event__repr(struct pyrf_event *pevent)
+{
+	return PyString_FromFormat("{ type: comm, pid: %u, tid: %u, comm: %s }",
+				   pevent->event.comm.pid,
+				   pevent->event.comm.tid,
+				   pevent->event.comm.comm);
+}
+
+static PyTypeObject pyrf_comm_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.comm_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_comm_event__doc,
+	.tp_members	= pyrf_comm_event__members,
+	.tp_repr	= (reprfunc)pyrf_comm_event__repr,
+};
+
+static char pyrf_throttle_event__doc[] = PyDoc_STR("perf throttle event object.");
+
+static PyMemberDef pyrf_throttle_event__members[] = {
+	sample_members
+	member_def(perf_event_header, type, T_UINT, "event type"),
+	member_def(throttle_event, time, T_ULONGLONG, "timestamp"),
+	member_def(throttle_event, id, T_ULONGLONG, "event id"),
+	member_def(throttle_event, stream_id, T_ULONGLONG, "event stream id"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_throttle_event__repr(struct pyrf_event *pevent)
+{
+	struct throttle_event *te = (struct throttle_event *)(&pevent->event.header + 1);
+
+	return PyString_FromFormat("{ type: %sthrottle, time: %" PRIu64 ", id: %" PRIu64
+				   ", stream_id: %" PRIu64 " }",
+				   pevent->event.header.type == PERF_RECORD_THROTTLE ? "" : "un",
+				   te->time, te->id, te->stream_id);
+}
+
+static PyTypeObject pyrf_throttle_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.throttle_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_throttle_event__doc,
+	.tp_members	= pyrf_throttle_event__members,
+	.tp_repr	= (reprfunc)pyrf_throttle_event__repr,
+};
+
+static char pyrf_lost_event__doc[] = PyDoc_STR("perf lost event object.");
+
+static PyMemberDef pyrf_lost_event__members[] = {
+	sample_members
+	member_def(lost_event, id, T_ULONGLONG, "event id"),
+	member_def(lost_event, lost, T_ULONGLONG, "number of lost events"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_lost_event__repr(struct pyrf_event *pevent)
+{
+	PyObject *ret;
+	char *s;
+
+	if (asprintf(&s, "{ type: lost, id: %#" PRIx64 ", "
+			 "lost: %#" PRIx64 " }",
+		     pevent->event.lost.id, pevent->event.lost.lost) < 0) {
+		ret = PyErr_NoMemory();
+	} else {
+		ret = PyString_FromString(s);
+		free(s);
+	}
+	return ret;
+}
+
+static PyTypeObject pyrf_lost_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.lost_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_lost_event__doc,
+	.tp_members	= pyrf_lost_event__members,
+	.tp_repr	= (reprfunc)pyrf_lost_event__repr,
+};
+
+static char pyrf_read_event__doc[] = PyDoc_STR("perf read event object.");
+
+static PyMemberDef pyrf_read_event__members[] = {
+	sample_members
+	member_def(read_event, pid, T_UINT, "event pid"),
+	member_def(read_event, tid, T_UINT, "event tid"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_read_event__repr(struct pyrf_event *pevent)
+{
+	return PyString_FromFormat("{ type: read, pid: %u, tid: %u }",
+				   pevent->event.read.pid,
+				   pevent->event.read.tid);
+	/*
+ 	 * FIXME: return the array of read values,
+ 	 * making this method useful ;-)
+ 	 */
+}
+
+static PyTypeObject pyrf_read_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.read_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_read_event__doc,
+	.tp_members	= pyrf_read_event__members,
+	.tp_repr	= (reprfunc)pyrf_read_event__repr,
+};
+
+static char pyrf_sample_event__doc[] = PyDoc_STR("perf sample event object.");
+
+static PyMemberDef pyrf_sample_event__members[] = {
+	sample_members
+	member_def(perf_event_header, type, T_UINT, "event type"),
+	{ .name = NULL, },
+};
+
+static PyObject *pyrf_sample_event__repr(struct pyrf_event *pevent)
+{
+	PyObject *ret;
+	char *s;
+
+	if (asprintf(&s, "{ type: sample }") < 0) {
+		ret = PyErr_NoMemory();
+	} else {
+		ret = PyString_FromString(s);
+		free(s);
+	}
+	return ret;
+}
+
+static PyTypeObject pyrf_sample_event__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.sample_event",
+	.tp_basicsize	= sizeof(struct pyrf_event),
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_sample_event__doc,
+	.tp_members	= pyrf_sample_event__members,
+	.tp_repr	= (reprfunc)pyrf_sample_event__repr,
+};
+
+static int pyrf_event__setup_types(void)
+{
+	int err;
+	pyrf_mmap_event__type.tp_new =
+	pyrf_task_event__type.tp_new =
+	pyrf_comm_event__type.tp_new =
+	pyrf_lost_event__type.tp_new =
+	pyrf_read_event__type.tp_new =
+	pyrf_sample_event__type.tp_new =
+	pyrf_throttle_event__type.tp_new = PyType_GenericNew;
+	err = PyType_Ready(&pyrf_mmap_event__type);
+	if (err < 0)
+		goto out;
+	err = PyType_Ready(&pyrf_lost_event__type);
+	if (err < 0)
+		goto out;
+	err = PyType_Ready(&pyrf_task_event__type);
+	if (err < 0)
+		goto out;
+	err = PyType_Ready(&pyrf_comm_event__type);
+	if (err < 0)
+		goto out;
+	err = PyType_Ready(&pyrf_throttle_event__type);
+	if (err < 0)
+		goto out;
+	err = PyType_Ready(&pyrf_read_event__type);
+	if (err < 0)
+		goto out;
+	err = PyType_Ready(&pyrf_sample_event__type);
+	if (err < 0)
+		goto out;
+out:
+	return err;
+}
+
+static PyTypeObject *pyrf_event__type[] = {
+	[PERF_RECORD_MMAP]	 = &pyrf_mmap_event__type,
+	[PERF_RECORD_LOST]	 = &pyrf_lost_event__type,
+	[PERF_RECORD_COMM]	 = &pyrf_comm_event__type,
+	[PERF_RECORD_EXIT]	 = &pyrf_task_event__type,
+	[PERF_RECORD_THROTTLE]	 = &pyrf_throttle_event__type,
+	[PERF_RECORD_UNTHROTTLE] = &pyrf_throttle_event__type,
+	[PERF_RECORD_FORK]	 = &pyrf_task_event__type,
+	[PERF_RECORD_READ]	 = &pyrf_read_event__type,
+	[PERF_RECORD_SAMPLE]	 = &pyrf_sample_event__type,
+};
+
+static PyObject *pyrf_event__new(union perf_event *event)
+{
+	struct pyrf_event *pevent;
+	PyTypeObject *ptype;
+
+	if (event->header.type < PERF_RECORD_MMAP ||
+	    event->header.type > PERF_RECORD_SAMPLE)
+		return NULL;
+
+	ptype = pyrf_event__type[event->header.type];
+	pevent = PyObject_New(struct pyrf_event, ptype);
+	if (pevent != NULL)
+		memcpy(&pevent->event, event, event->header.size);
+	return (PyObject *)pevent;
+}
+
+struct pyrf_cpu_map {
+	PyObject_HEAD
+
+	struct cpu_map *cpus;
+};
+
+static int pyrf_cpu_map__init(struct pyrf_cpu_map *pcpus,
+			      PyObject *args, PyObject *kwargs)
+{
+	static char *kwlist[] = { "cpustr", NULL };
+	char *cpustr = NULL;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|s",
+					 kwlist, &cpustr))
+		return -1;
+
+	pcpus->cpus = cpu_map__new(cpustr);
+	if (pcpus->cpus == NULL)
+		return -1;
+	return 0;
+}
+
+static void pyrf_cpu_map__delete(struct pyrf_cpu_map *pcpus)
+{
+	cpu_map__delete(pcpus->cpus);
+	pcpus->ob_type->tp_free((PyObject*)pcpus);
+}
+
+static Py_ssize_t pyrf_cpu_map__length(PyObject *obj)
+{
+	struct pyrf_cpu_map *pcpus = (void *)obj;
+
+	return pcpus->cpus->nr;
+}
+
+static PyObject *pyrf_cpu_map__item(PyObject *obj, Py_ssize_t i)
+{
+	struct pyrf_cpu_map *pcpus = (void *)obj;
+
+	if (i >= pcpus->cpus->nr)
+		return NULL;
+
+	return Py_BuildValue("i", pcpus->cpus->map[i]);
+}
+
+static PySequenceMethods pyrf_cpu_map__sequence_methods = {
+	.sq_length = pyrf_cpu_map__length,
+	.sq_item   = pyrf_cpu_map__item,
+};
+
+static char pyrf_cpu_map__doc[] = PyDoc_STR("cpu map object.");
+
+static PyTypeObject pyrf_cpu_map__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.cpu_map",
+	.tp_basicsize	= sizeof(struct pyrf_cpu_map),
+	.tp_dealloc	= (destructor)pyrf_cpu_map__delete,
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_cpu_map__doc,
+	.tp_as_sequence	= &pyrf_cpu_map__sequence_methods,
+	.tp_init	= (initproc)pyrf_cpu_map__init,
+};
+
+static int pyrf_cpu_map__setup_types(void)
+{
+	pyrf_cpu_map__type.tp_new = PyType_GenericNew;
+	return PyType_Ready(&pyrf_cpu_map__type);
+}
+
+struct pyrf_thread_map {
+	PyObject_HEAD
+
+	struct thread_map *threads;
+};
+
+static int pyrf_thread_map__init(struct pyrf_thread_map *pthreads,
+				 PyObject *args, PyObject *kwargs)
+{
+	static char *kwlist[] = { "pid", "tid", "uid", NULL };
+	int pid = -1, tid = -1, uid = UINT_MAX;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|iii",
+					 kwlist, &pid, &tid, &uid))
+		return -1;
+
+	pthreads->threads = thread_map__new(pid, tid, uid);
+	if (pthreads->threads == NULL)
+		return -1;
+	return 0;
+}
+
+static void pyrf_thread_map__delete(struct pyrf_thread_map *pthreads)
+{
+	thread_map__delete(pthreads->threads);
+	pthreads->ob_type->tp_free((PyObject*)pthreads);
+}
+
+static Py_ssize_t pyrf_thread_map__length(PyObject *obj)
+{
+	struct pyrf_thread_map *pthreads = (void *)obj;
+
+	return pthreads->threads->nr;
+}
+
+static PyObject *pyrf_thread_map__item(PyObject *obj, Py_ssize_t i)
+{
+	struct pyrf_thread_map *pthreads = (void *)obj;
+
+	if (i >= pthreads->threads->nr)
+		return NULL;
+
+	return Py_BuildValue("i", pthreads->threads->map[i]);
+}
+
+static PySequenceMethods pyrf_thread_map__sequence_methods = {
+	.sq_length = pyrf_thread_map__length,
+	.sq_item   = pyrf_thread_map__item,
+};
+
+static char pyrf_thread_map__doc[] = PyDoc_STR("thread map object.");
+
+static PyTypeObject pyrf_thread_map__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.thread_map",
+	.tp_basicsize	= sizeof(struct pyrf_thread_map),
+	.tp_dealloc	= (destructor)pyrf_thread_map__delete,
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_thread_map__doc,
+	.tp_as_sequence	= &pyrf_thread_map__sequence_methods,
+	.tp_init	= (initproc)pyrf_thread_map__init,
+};
+
+static int pyrf_thread_map__setup_types(void)
+{
+	pyrf_thread_map__type.tp_new = PyType_GenericNew;
+	return PyType_Ready(&pyrf_thread_map__type);
+}
+
+struct pyrf_evsel {
+	PyObject_HEAD
+
+	struct perf_evsel evsel;
+};
+
+static int pyrf_evsel__init(struct pyrf_evsel *pevsel,
+			    PyObject *args, PyObject *kwargs)
+{
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_HARDWARE,
+		.config = PERF_COUNT_HW_CPU_CYCLES,
+		.sample_type = PERF_SAMPLE_PERIOD | PERF_SAMPLE_TID,
+	};
+	static char *kwlist[] = {
+		"type",
+		"config",
+		"sample_freq",
+		"sample_period",
+		"sample_type",
+		"read_format",
+		"disabled",
+		"inherit",
+		"pinned",
+		"exclusive",
+		"exclude_user",
+		"exclude_kernel",
+		"exclude_hv",
+		"exclude_idle",
+		"mmap",
+		"comm",
+		"freq",
+		"inherit_stat",
+		"enable_on_exec",
+		"task",
+		"watermark",
+		"precise_ip",
+		"mmap_data",
+		"sample_id_all",
+		"wakeup_events",
+		"bp_type",
+		"bp_addr",
+		"bp_len",
+		 NULL
+	};
+	u64 sample_period = 0;
+	u32 disabled = 0,
+	    inherit = 0,
+	    pinned = 0,
+	    exclusive = 0,
+	    exclude_user = 0,
+	    exclude_kernel = 0,
+	    exclude_hv = 0,
+	    exclude_idle = 0,
+	    mmap = 0,
+	    comm = 0,
+	    freq = 1,
+	    inherit_stat = 0,
+	    enable_on_exec = 0,
+	    task = 0,
+	    watermark = 0,
+	    precise_ip = 0,
+	    mmap_data = 0,
+	    sample_id_all = 1;
+	int idx = 0;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs,
+					 "|iKiKKiiiiiiiiiiiiiiiiiiiiiKK", kwlist,
+					 &attr.type, &attr.config, &attr.sample_freq,
+					 &sample_period, &attr.sample_type,
+					 &attr.read_format, &disabled, &inherit,
+					 &pinned, &exclusive, &exclude_user,
+					 &exclude_kernel, &exclude_hv, &exclude_idle,
+					 &mmap, &comm, &freq, &inherit_stat,
+					 &enable_on_exec, &task, &watermark,
+					 &precise_ip, &mmap_data, &sample_id_all,
+					 &attr.wakeup_events, &attr.bp_type,
+					 &attr.bp_addr, &attr.bp_len, &idx))
+		return -1;
+
+	/* union... */
+	if (sample_period != 0) {
+		if (attr.sample_freq != 0)
+			return -1; /* FIXME: throw right exception */
+		attr.sample_period = sample_period;
+	}
+
+	/* Bitfields */
+	attr.disabled	    = disabled;
+	attr.inherit	    = inherit;
+	attr.pinned	    = pinned;
+	attr.exclusive	    = exclusive;
+	attr.exclude_user   = exclude_user;
+	attr.exclude_kernel = exclude_kernel;
+	attr.exclude_hv	    = exclude_hv;
+	attr.exclude_idle   = exclude_idle;
+	attr.mmap	    = mmap;
+	attr.comm	    = comm;
+	attr.freq	    = freq;
+	attr.inherit_stat   = inherit_stat;
+	attr.enable_on_exec = enable_on_exec;
+	attr.task	    = task;
+	attr.watermark	    = watermark;
+	attr.precise_ip	    = precise_ip;
+	attr.mmap_data	    = mmap_data;
+	attr.sample_id_all  = sample_id_all;
+
+	perf_evsel__init(&pevsel->evsel, &attr, idx);
+	return 0;
+}
+
+static void pyrf_evsel__delete(struct pyrf_evsel *pevsel)
+{
+	perf_evsel__exit(&pevsel->evsel);
+	pevsel->ob_type->tp_free((PyObject*)pevsel);
+}
+
+static PyObject *pyrf_evsel__open(struct pyrf_evsel *pevsel,
+				  PyObject *args, PyObject *kwargs)
+{
+	struct perf_evsel *evsel = &pevsel->evsel;
+	struct cpu_map *cpus = NULL;
+	struct thread_map *threads = NULL;
+	PyObject *pcpus = NULL, *pthreads = NULL;
+	int group = 0, inherit = 0;
+	static char *kwlist[] = { "cpus", "threads", "group", "inherit", NULL };
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|OOii", kwlist,
+					 &pcpus, &pthreads, &group, &inherit))
+		return NULL;
+
+	if (pthreads != NULL)
+		threads = ((struct pyrf_thread_map *)pthreads)->threads;
+
+	if (pcpus != NULL)
+		cpus = ((struct pyrf_cpu_map *)pcpus)->cpus;
+
+	evsel->attr.inherit = inherit;
+	/*
+	 * This will group just the fds for this single evsel, to group
+	 * multiple events, use evlist.open().
+	 */
+	if (perf_evsel__open(evsel, cpus, threads) < 0) {
+		PyErr_SetFromErrno(PyExc_OSError);
+		return NULL;
+	}
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+static PyMethodDef pyrf_evsel__methods[] = {
+	{
+		.ml_name  = "open",
+		.ml_meth  = (PyCFunction)pyrf_evsel__open,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("open the event selector file descriptor table.")
+	},
+	{ .ml_name = NULL, }
+};
+
+static char pyrf_evsel__doc[] = PyDoc_STR("perf event selector list object.");
+
+static PyTypeObject pyrf_evsel__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.evsel",
+	.tp_basicsize	= sizeof(struct pyrf_evsel),
+	.tp_dealloc	= (destructor)pyrf_evsel__delete,
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_doc		= pyrf_evsel__doc,
+	.tp_methods	= pyrf_evsel__methods,
+	.tp_init	= (initproc)pyrf_evsel__init,
+};
+
+static int pyrf_evsel__setup_types(void)
+{
+	pyrf_evsel__type.tp_new = PyType_GenericNew;
+	return PyType_Ready(&pyrf_evsel__type);
+}
+
+struct pyrf_evlist {
+	PyObject_HEAD
+
+	struct perf_evlist evlist;
+};
+
+static int pyrf_evlist__init(struct pyrf_evlist *pevlist,
+			     PyObject *args, PyObject *kwargs __maybe_unused)
+{
+	PyObject *pcpus = NULL, *pthreads = NULL;
+	struct cpu_map *cpus;
+	struct thread_map *threads;
+
+	if (!PyArg_ParseTuple(args, "OO", &pcpus, &pthreads))
+		return -1;
+
+	threads = ((struct pyrf_thread_map *)pthreads)->threads;
+	cpus = ((struct pyrf_cpu_map *)pcpus)->cpus;
+	perf_evlist__init(&pevlist->evlist, cpus, threads);
+	return 0;
+}
+
+static void pyrf_evlist__delete(struct pyrf_evlist *pevlist)
+{
+	perf_evlist__exit(&pevlist->evlist);
+	pevlist->ob_type->tp_free((PyObject*)pevlist);
+}
+
+static PyObject *pyrf_evlist__mmap(struct pyrf_evlist *pevlist,
+				   PyObject *args, PyObject *kwargs)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+	static char *kwlist[] = { "pages", "overwrite", NULL };
+	int pages = 128, overwrite = false;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|ii", kwlist,
+					 &pages, &overwrite))
+		return NULL;
+
+	if (perf_evlist__mmap(evlist, pages, overwrite) < 0) {
+		PyErr_SetFromErrno(PyExc_OSError);
+		return NULL;
+	}
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+static PyObject *pyrf_evlist__poll(struct pyrf_evlist *pevlist,
+				   PyObject *args, PyObject *kwargs)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+	static char *kwlist[] = { "timeout", NULL };
+	int timeout = -1, n;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|i", kwlist, &timeout))
+		return NULL;
+
+	n = perf_evlist__poll(evlist, timeout);
+	if (n < 0) {
+		PyErr_SetFromErrno(PyExc_OSError);
+		return NULL;
+	}
+
+	return Py_BuildValue("i", n);
+}
+
+static PyObject *pyrf_evlist__get_pollfd(struct pyrf_evlist *pevlist,
+					 PyObject *args __maybe_unused,
+					 PyObject *kwargs __maybe_unused)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+        PyObject *list = PyList_New(0);
+	int i;
+
+	for (i = 0; i < evlist->pollfd.nr; ++i) {
+		PyObject *file;
+		FILE *fp = fdopen(evlist->pollfd.entries[i].fd, "r");
+
+		if (fp == NULL)
+			goto free_list;
+
+		file = PyFile_FromFile(fp, "perf", "r", NULL);
+		if (file == NULL)
+			goto free_list;
+
+		if (PyList_Append(list, file) != 0) {
+			Py_DECREF(file);
+			goto free_list;
+		}
+
+		Py_DECREF(file);
+	}
+
+	return list;
+free_list:
+	return PyErr_NoMemory();
+}
+
+
+static PyObject *pyrf_evlist__add(struct pyrf_evlist *pevlist,
+				  PyObject *args,
+				  PyObject *kwargs __maybe_unused)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+	PyObject *pevsel;
+	struct perf_evsel *evsel;
+
+	if (!PyArg_ParseTuple(args, "O", &pevsel))
+		return NULL;
+
+	Py_INCREF(pevsel);
+	evsel = &((struct pyrf_evsel *)pevsel)->evsel;
+	evsel->idx = evlist->nr_entries;
+	perf_evlist__add(evlist, evsel);
+
+	return Py_BuildValue("i", evlist->nr_entries);
+}
+
+static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,
+					  PyObject *args, PyObject *kwargs)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+	union perf_event *event;
+	int sample_id_all = 1, cpu;
+	static char *kwlist[] = { "cpu", "sample_id_all", NULL };
+	int err;
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "i|i", kwlist,
+					 &cpu, &sample_id_all))
+		return NULL;
+
+	event = perf_evlist__mmap_read(evlist, cpu);
+	if (event != NULL) {
+		PyObject *pyevent = pyrf_event__new(event);
+		struct pyrf_event *pevent = (struct pyrf_event *)pyevent;
+
+		perf_evlist__mmap_consume(evlist, cpu);
+
+		if (pyevent == NULL)
+			return PyErr_NoMemory();
+
+		err = perf_evlist__parse_sample(evlist, event, &pevent->sample);
+		if (err)
+			return PyErr_Format(PyExc_OSError,
+					    "perf: can't parse sample, err=%d", err);
+		return pyevent;
+	}
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+static PyObject *pyrf_evlist__open(struct pyrf_evlist *pevlist,
+				   PyObject *args, PyObject *kwargs)
+{
+	struct perf_evlist *evlist = &pevlist->evlist;
+	int group = 0;
+	static char *kwlist[] = { "group", NULL };
+
+	if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|OOii", kwlist, &group))
+		return NULL;
+
+	if (group)
+		perf_evlist__set_leader(evlist);
+
+	if (perf_evlist__open(evlist) < 0) {
+		PyErr_SetFromErrno(PyExc_OSError);
+		return NULL;
+	}
+
+	Py_INCREF(Py_None);
+	return Py_None;
+}
+
+static PyMethodDef pyrf_evlist__methods[] = {
+	{
+		.ml_name  = "mmap",
+		.ml_meth  = (PyCFunction)pyrf_evlist__mmap,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("mmap the file descriptor table.")
+	},
+	{
+		.ml_name  = "open",
+		.ml_meth  = (PyCFunction)pyrf_evlist__open,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("open the file descriptors.")
+	},
+	{
+		.ml_name  = "poll",
+		.ml_meth  = (PyCFunction)pyrf_evlist__poll,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("poll the file descriptor table.")
+	},
+	{
+		.ml_name  = "get_pollfd",
+		.ml_meth  = (PyCFunction)pyrf_evlist__get_pollfd,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("get the poll file descriptor table.")
+	},
+	{
+		.ml_name  = "add",
+		.ml_meth  = (PyCFunction)pyrf_evlist__add,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("adds an event selector to the list.")
+	},
+	{
+		.ml_name  = "read_on_cpu",
+		.ml_meth  = (PyCFunction)pyrf_evlist__read_on_cpu,
+		.ml_flags = METH_VARARGS | METH_KEYWORDS,
+		.ml_doc	  = PyDoc_STR("reads an event.")
+	},
+	{ .ml_name = NULL, }
+};
+
+static Py_ssize_t pyrf_evlist__length(PyObject *obj)
+{
+	struct pyrf_evlist *pevlist = (void *)obj;
+
+	return pevlist->evlist.nr_entries;
+}
+
+static PyObject *pyrf_evlist__item(PyObject *obj, Py_ssize_t i)
+{
+	struct pyrf_evlist *pevlist = (void *)obj;
+	struct perf_evsel *pos;
+
+	if (i >= pevlist->evlist.nr_entries)
+		return NULL;
+
+	evlist__for_each(&pevlist->evlist, pos) {
+		if (i-- == 0)
+			break;
+	}
+
+	return Py_BuildValue("O", container_of(pos, struct pyrf_evsel, evsel));
+}
+
+static PySequenceMethods pyrf_evlist__sequence_methods = {
+	.sq_length = pyrf_evlist__length,
+	.sq_item   = pyrf_evlist__item,
+};
+
+static char pyrf_evlist__doc[] = PyDoc_STR("perf event selector list object.");
+
+static PyTypeObject pyrf_evlist__type = {
+	PyVarObject_HEAD_INIT(NULL, 0)
+	.tp_name	= "perf.evlist",
+	.tp_basicsize	= sizeof(struct pyrf_evlist),
+	.tp_dealloc	= (destructor)pyrf_evlist__delete,
+	.tp_flags	= Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE,
+	.tp_as_sequence	= &pyrf_evlist__sequence_methods,
+	.tp_doc		= pyrf_evlist__doc,
+	.tp_methods	= pyrf_evlist__methods,
+	.tp_init	= (initproc)pyrf_evlist__init,
+};
+
+static int pyrf_evlist__setup_types(void)
+{
+	pyrf_evlist__type.tp_new = PyType_GenericNew;
+	return PyType_Ready(&pyrf_evlist__type);
+}
+
+static struct {
+	const char *name;
+	int	    value;
+} perf__constants[] = {
+	{ "TYPE_HARDWARE",   PERF_TYPE_HARDWARE },
+	{ "TYPE_SOFTWARE",   PERF_TYPE_SOFTWARE },
+	{ "TYPE_TRACEPOINT", PERF_TYPE_TRACEPOINT },
+	{ "TYPE_HW_CACHE",   PERF_TYPE_HW_CACHE },
+	{ "TYPE_RAW",	     PERF_TYPE_RAW },
+	{ "TYPE_BREAKPOINT", PERF_TYPE_BREAKPOINT },
+
+	{ "COUNT_HW_CPU_CYCLES",	  PERF_COUNT_HW_CPU_CYCLES },
+	{ "COUNT_HW_INSTRUCTIONS",	  PERF_COUNT_HW_INSTRUCTIONS },
+	{ "COUNT_HW_CACHE_REFERENCES",	  PERF_COUNT_HW_CACHE_REFERENCES },
+	{ "COUNT_HW_CACHE_MISSES",	  PERF_COUNT_HW_CACHE_MISSES },
+	{ "COUNT_HW_BRANCH_INSTRUCTIONS", PERF_COUNT_HW_BRANCH_INSTRUCTIONS },
+	{ "COUNT_HW_BRANCH_MISSES",	  PERF_COUNT_HW_BRANCH_MISSES },
+	{ "COUNT_HW_BUS_CYCLES",	  PERF_COUNT_HW_BUS_CYCLES },
+	{ "COUNT_HW_CACHE_L1D",		  PERF_COUNT_HW_CACHE_L1D },
+	{ "COUNT_HW_CACHE_L1I",		  PERF_COUNT_HW_CACHE_L1I },
+	{ "COUNT_HW_CACHE_LL",	  	  PERF_COUNT_HW_CACHE_LL },
+	{ "COUNT_HW_CACHE_DTLB",	  PERF_COUNT_HW_CACHE_DTLB },
+	{ "COUNT_HW_CACHE_ITLB",	  PERF_COUNT_HW_CACHE_ITLB },
+	{ "COUNT_HW_CACHE_BPU",		  PERF_COUNT_HW_CACHE_BPU },
+	{ "COUNT_HW_CACHE_OP_READ",	  PERF_COUNT_HW_CACHE_OP_READ },
+	{ "COUNT_HW_CACHE_OP_WRITE",	  PERF_COUNT_HW_CACHE_OP_WRITE },
+	{ "COUNT_HW_CACHE_OP_PREFETCH",	  PERF_COUNT_HW_CACHE_OP_PREFETCH },
+	{ "COUNT_HW_CACHE_RESULT_ACCESS", PERF_COUNT_HW_CACHE_RESULT_ACCESS },
+	{ "COUNT_HW_CACHE_RESULT_MISS",   PERF_COUNT_HW_CACHE_RESULT_MISS },
+
+	{ "COUNT_HW_STALLED_CYCLES_FRONTEND",	  PERF_COUNT_HW_STALLED_CYCLES_FRONTEND },
+	{ "COUNT_HW_STALLED_CYCLES_BACKEND",	  PERF_COUNT_HW_STALLED_CYCLES_BACKEND },
+
+	{ "COUNT_SW_CPU_CLOCK",	       PERF_COUNT_SW_CPU_CLOCK },
+	{ "COUNT_SW_TASK_CLOCK",       PERF_COUNT_SW_TASK_CLOCK },
+	{ "COUNT_SW_PAGE_FAULTS",      PERF_COUNT_SW_PAGE_FAULTS },
+	{ "COUNT_SW_CONTEXT_SWITCHES", PERF_COUNT_SW_CONTEXT_SWITCHES },
+	{ "COUNT_SW_CPU_MIGRATIONS",   PERF_COUNT_SW_CPU_MIGRATIONS },
+	{ "COUNT_SW_PAGE_FAULTS_MIN",  PERF_COUNT_SW_PAGE_FAULTS_MIN },
+	{ "COUNT_SW_PAGE_FAULTS_MAJ",  PERF_COUNT_SW_PAGE_FAULTS_MAJ },
+	{ "COUNT_SW_ALIGNMENT_FAULTS", PERF_COUNT_SW_ALIGNMENT_FAULTS },
+	{ "COUNT_SW_EMULATION_FAULTS", PERF_COUNT_SW_EMULATION_FAULTS },
+	{ "COUNT_SW_DUMMY",            PERF_COUNT_SW_DUMMY },
+
+	{ "SAMPLE_IP",	      PERF_SAMPLE_IP },
+	{ "SAMPLE_TID",	      PERF_SAMPLE_TID },
+	{ "SAMPLE_TIME",      PERF_SAMPLE_TIME },
+	{ "SAMPLE_ADDR",      PERF_SAMPLE_ADDR },
+	{ "SAMPLE_READ",      PERF_SAMPLE_READ },
+	{ "SAMPLE_CALLCHAIN", PERF_SAMPLE_CALLCHAIN },
+	{ "SAMPLE_ID",	      PERF_SAMPLE_ID },
+	{ "SAMPLE_CPU",	      PERF_SAMPLE_CPU },
+	{ "SAMPLE_PERIOD",    PERF_SAMPLE_PERIOD },
+	{ "SAMPLE_STREAM_ID", PERF_SAMPLE_STREAM_ID },
+	{ "SAMPLE_RAW",	      PERF_SAMPLE_RAW },
+
+	{ "FORMAT_TOTAL_TIME_ENABLED", PERF_FORMAT_TOTAL_TIME_ENABLED },
+	{ "FORMAT_TOTAL_TIME_RUNNING", PERF_FORMAT_TOTAL_TIME_RUNNING },
+	{ "FORMAT_ID",		       PERF_FORMAT_ID },
+	{ "FORMAT_GROUP",	       PERF_FORMAT_GROUP },
+
+	{ "RECORD_MMAP",       PERF_RECORD_MMAP },
+	{ "RECORD_LOST",       PERF_RECORD_LOST },
+	{ "RECORD_COMM",       PERF_RECORD_COMM },
+	{ "RECORD_EXIT",       PERF_RECORD_EXIT },
+	{ "RECORD_THROTTLE",   PERF_RECORD_THROTTLE },
+	{ "RECORD_UNTHROTTLE", PERF_RECORD_UNTHROTTLE },
+	{ "RECORD_FORK",       PERF_RECORD_FORK },
+	{ "RECORD_READ",       PERF_RECORD_READ },
+	{ "RECORD_SAMPLE",     PERF_RECORD_SAMPLE },
+	{ .name = NULL, },
+};
+
+static PyMethodDef perf__methods[] = {
+	{ .ml_name = NULL, }
+};
+
+PyMODINIT_FUNC initperf(void)
+{
+	PyObject *obj;
+	int i;
+	PyObject *dict, *module = Py_InitModule("perf", perf__methods);
+
+	if (module == NULL ||
+	    pyrf_event__setup_types() < 0 ||
+	    pyrf_evlist__setup_types() < 0 ||
+	    pyrf_evsel__setup_types() < 0 ||
+	    pyrf_thread_map__setup_types() < 0 ||
+	    pyrf_cpu_map__setup_types() < 0)
+		return;
+
+	/* The page_size is placed in util object. */
+	page_size = sysconf(_SC_PAGE_SIZE);
+
+	Py_INCREF(&pyrf_evlist__type);
+	PyModule_AddObject(module, "evlist", (PyObject*)&pyrf_evlist__type);
+
+	Py_INCREF(&pyrf_evsel__type);
+	PyModule_AddObject(module, "evsel", (PyObject*)&pyrf_evsel__type);
+
+	Py_INCREF(&pyrf_thread_map__type);
+	PyModule_AddObject(module, "thread_map", (PyObject*)&pyrf_thread_map__type);
+
+	Py_INCREF(&pyrf_cpu_map__type);
+	PyModule_AddObject(module, "cpu_map", (PyObject*)&pyrf_cpu_map__type);
+
+	dict = PyModule_GetDict(module);
+	if (dict == NULL)
+		goto error;
+
+	for (i = 0; perf__constants[i].name != NULL; i++) {
+		obj = PyInt_FromLong(perf__constants[i].value);
+		if (obj == NULL)
+			goto error;
+		PyDict_SetItemString(dict, perf__constants[i].name, obj);
+		Py_DECREF(obj);
+	}
+
+error:
+	if (PyErr_Occurred())
+		PyErr_SetString(PyExc_ImportError, "perf: Init failed!");
+}
+
+/*
+ * Dummy, to avoid dragging all the test_attr infrastructure in the python
+ * binding.
+ */
+void test_attr__open(struct perf_event_attr *attr, pid_t pid, int cpu,
+                     int fd, int group_fd, unsigned long flags)
+{
+}
diff --git a/tools/perf/util/quote.c b/tools/perf/util/quote.c
new file mode 100644
index 000000000..01f03242b
--- /dev/null
+++ b/tools/perf/util/quote.c
@@ -0,0 +1,54 @@
+#include "cache.h"
+#include "quote.h"
+
+/* Help to copy the thing properly quoted for the shell safety.
+ * any single quote is replaced with '\'', any exclamation point
+ * is replaced with '\!', and the whole thing is enclosed in a
+ *
+ * E.g.
+ *  original     sq_quote     result
+ *  name     ==> name      ==> 'name'
+ *  a b      ==> a b       ==> 'a b'
+ *  a'b      ==> a'\''b    ==> 'a'\''b'
+ *  a!b      ==> a'\!'b    ==> 'a'\!'b'
+ */
+static inline int need_bs_quote(char c)
+{
+	return (c == '\'' || c == '!');
+}
+
+static void sq_quote_buf(struct strbuf *dst, const char *src)
+{
+	char *to_free = NULL;
+
+	if (dst->buf == src)
+		to_free = strbuf_detach(dst, NULL);
+
+	strbuf_addch(dst, '\'');
+	while (*src) {
+		size_t len = strcspn(src, "'!");
+		strbuf_add(dst, src, len);
+		src += len;
+		while (need_bs_quote(*src)) {
+			strbuf_addstr(dst, "'\\");
+			strbuf_addch(dst, *src++);
+			strbuf_addch(dst, '\'');
+		}
+	}
+	strbuf_addch(dst, '\'');
+	free(to_free);
+}
+
+void sq_quote_argv(struct strbuf *dst, const char** argv, size_t maxlen)
+{
+	int i;
+
+	/* Copy into destination buffer. */
+	strbuf_grow(dst, 255);
+	for (i = 0; argv[i]; ++i) {
+		strbuf_addch(dst, ' ');
+		sq_quote_buf(dst, argv[i]);
+		if (maxlen && dst->len > maxlen)
+			die("Too many or long arguments");
+	}
+}
diff --git a/tools/perf/util/quote.h b/tools/perf/util/quote.h
new file mode 100644
index 000000000..172889ea2
--- /dev/null
+++ b/tools/perf/util/quote.h
@@ -0,0 +1,29 @@
+#ifndef __PERF_QUOTE_H
+#define __PERF_QUOTE_H
+
+#include <stddef.h>
+#include <stdio.h>
+
+/* Help to copy the thing properly quoted for the shell safety.
+ * any single quote is replaced with '\'', any exclamation point
+ * is replaced with '\!', and the whole thing is enclosed in a
+ * single quote pair.
+ *
+ * For example, if you are passing the result to system() as an
+ * argument:
+ *
+ * sprintf(cmd, "foobar %s %s", sq_quote(arg0), sq_quote(arg1))
+ *
+ * would be appropriate.  If the system() is going to call ssh to
+ * run the command on the other side:
+ *
+ * sprintf(cmd, "git-diff-tree %s %s", sq_quote(arg0), sq_quote(arg1));
+ * sprintf(rcmd, "ssh %s %s", sq_util/quote.host), sq_quote(cmd));
+ *
+ * Note that the above examples leak memory!  Remember to free result from
+ * sq_quote() in a real application.
+ */
+
+extern void sq_quote_argv(struct strbuf *, const char **argv, size_t maxlen);
+
+#endif /* __PERF_QUOTE_H */
diff --git a/tools/perf/util/rblist.c b/tools/perf/util/rblist.c
new file mode 100644
index 000000000..0dfe27d99
--- /dev/null
+++ b/tools/perf/util/rblist.c
@@ -0,0 +1,128 @@
+/*
+ * Based on strlist.c by:
+ * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Licensed under the GPLv2.
+ */
+
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+#include "rblist.h"
+
+int rblist__add_node(struct rblist *rblist, const void *new_entry)
+{
+	struct rb_node **p = &rblist->entries.rb_node;
+	struct rb_node *parent = NULL, *new_node;
+
+	while (*p != NULL) {
+		int rc;
+
+		parent = *p;
+
+		rc = rblist->node_cmp(parent, new_entry);
+		if (rc > 0)
+			p = &(*p)->rb_left;
+		else if (rc < 0)
+			p = &(*p)->rb_right;
+		else
+			return -EEXIST;
+	}
+
+	new_node = rblist->node_new(rblist, new_entry);
+	if (new_node == NULL)
+		return -ENOMEM;
+
+	rb_link_node(new_node, parent, p);
+	rb_insert_color(new_node, &rblist->entries);
+	++rblist->nr_entries;
+
+	return 0;
+}
+
+void rblist__remove_node(struct rblist *rblist, struct rb_node *rb_node)
+{
+	rb_erase(rb_node, &rblist->entries);
+	--rblist->nr_entries;
+	rblist->node_delete(rblist, rb_node);
+}
+
+static struct rb_node *__rblist__findnew(struct rblist *rblist,
+					 const void *entry,
+					 bool create)
+{
+	struct rb_node **p = &rblist->entries.rb_node;
+	struct rb_node *parent = NULL, *new_node = NULL;
+
+	while (*p != NULL) {
+		int rc;
+
+		parent = *p;
+
+		rc = rblist->node_cmp(parent, entry);
+		if (rc > 0)
+			p = &(*p)->rb_left;
+		else if (rc < 0)
+			p = &(*p)->rb_right;
+		else
+			return parent;
+	}
+
+	if (create) {
+		new_node = rblist->node_new(rblist, entry);
+		if (new_node) {
+			rb_link_node(new_node, parent, p);
+			rb_insert_color(new_node, &rblist->entries);
+			++rblist->nr_entries;
+		}
+	}
+
+	return new_node;
+}
+
+struct rb_node *rblist__find(struct rblist *rblist, const void *entry)
+{
+	return __rblist__findnew(rblist, entry, false);
+}
+
+struct rb_node *rblist__findnew(struct rblist *rblist, const void *entry)
+{
+	return __rblist__findnew(rblist, entry, true);
+}
+
+void rblist__init(struct rblist *rblist)
+{
+	if (rblist != NULL) {
+		rblist->entries	 = RB_ROOT;
+		rblist->nr_entries = 0;
+	}
+
+	return;
+}
+
+void rblist__delete(struct rblist *rblist)
+{
+	if (rblist != NULL) {
+		struct rb_node *pos, *next = rb_first(&rblist->entries);
+
+		while (next) {
+			pos = next;
+			next = rb_next(pos);
+			rblist__remove_node(rblist, pos);
+		}
+		free(rblist);
+	}
+}
+
+struct rb_node *rblist__entry(const struct rblist *rblist, unsigned int idx)
+{
+	struct rb_node *node;
+
+	for (node = rb_first(&rblist->entries); node; node = rb_next(node)) {
+		if (!idx--)
+			return node;
+	}
+
+	return NULL;
+}
diff --git a/tools/perf/util/rblist.h b/tools/perf/util/rblist.h
new file mode 100644
index 000000000..ff9913b99
--- /dev/null
+++ b/tools/perf/util/rblist.h
@@ -0,0 +1,48 @@
+#ifndef __PERF_RBLIST_H
+#define __PERF_RBLIST_H
+
+#include <linux/rbtree.h>
+#include <stdbool.h>
+
+/*
+ * create node structs of the form:
+ * struct my_node {
+ *     struct rb_node rb_node;
+ *     ... my data ...
+ * };
+ *
+ * create list structs of the form:
+ * struct mylist {
+ *     struct rblist rblist;
+ *     ... my data ...
+ * };
+ */
+
+struct rblist {
+	struct rb_root entries;
+	unsigned int   nr_entries;
+
+	int (*node_cmp)(struct rb_node *rbn, const void *entry);
+	struct rb_node *(*node_new)(struct rblist *rlist, const void *new_entry);
+	void (*node_delete)(struct rblist *rblist, struct rb_node *rb_node);
+};
+
+void rblist__init(struct rblist *rblist);
+void rblist__delete(struct rblist *rblist);
+int rblist__add_node(struct rblist *rblist, const void *new_entry);
+void rblist__remove_node(struct rblist *rblist, struct rb_node *rb_node);
+struct rb_node *rblist__find(struct rblist *rblist, const void *entry);
+struct rb_node *rblist__findnew(struct rblist *rblist, const void *entry);
+struct rb_node *rblist__entry(const struct rblist *rblist, unsigned int idx);
+
+static inline bool rblist__empty(const struct rblist *rblist)
+{
+	return rblist->nr_entries == 0;
+}
+
+static inline unsigned int rblist__nr_entries(const struct rblist *rblist)
+{
+	return rblist->nr_entries;
+}
+
+#endif /* __PERF_RBLIST_H */
diff --git a/tools/perf/util/record.c b/tools/perf/util/record.c
new file mode 100644
index 000000000..8acd0df88
--- /dev/null
+++ b/tools/perf/util/record.c
@@ -0,0 +1,243 @@
+#include "evlist.h"
+#include "evsel.h"
+#include "cpumap.h"
+#include "parse-events.h"
+#include <api/fs/fs.h>
+#include "util.h"
+#include "cloexec.h"
+
+typedef void (*setup_probe_fn_t)(struct perf_evsel *evsel);
+
+static int perf_do_probe_api(setup_probe_fn_t fn, int cpu, const char *str)
+{
+	struct perf_evlist *evlist;
+	struct perf_evsel *evsel;
+	unsigned long flags = perf_event_open_cloexec_flag();
+	int err = -EAGAIN, fd;
+	static pid_t pid = -1;
+
+	evlist = perf_evlist__new();
+	if (!evlist)
+		return -ENOMEM;
+
+	if (parse_events(evlist, str))
+		goto out_delete;
+
+	evsel = perf_evlist__first(evlist);
+
+	while (1) {
+		fd = sys_perf_event_open(&evsel->attr, pid, cpu, -1, flags);
+		if (fd < 0) {
+			if (pid == -1 && errno == EACCES) {
+				pid = 0;
+				continue;
+			}
+			goto out_delete;
+		}
+		break;
+	}
+	close(fd);
+
+	fn(evsel);
+
+	fd = sys_perf_event_open(&evsel->attr, pid, cpu, -1, flags);
+	if (fd < 0) {
+		if (errno == EINVAL)
+			err = -EINVAL;
+		goto out_delete;
+	}
+	close(fd);
+	err = 0;
+
+out_delete:
+	perf_evlist__delete(evlist);
+	return err;
+}
+
+static bool perf_probe_api(setup_probe_fn_t fn)
+{
+	const char *try[] = {"cycles:u", "instructions:u", "cpu-clock:u", NULL};
+	struct cpu_map *cpus;
+	int cpu, ret, i = 0;
+
+	cpus = cpu_map__new(NULL);
+	if (!cpus)
+		return false;
+	cpu = cpus->map[0];
+	cpu_map__delete(cpus);
+
+	do {
+		ret = perf_do_probe_api(fn, cpu, try[i++]);
+		if (!ret)
+			return true;
+	} while (ret == -EAGAIN && try[i]);
+
+	return false;
+}
+
+static void perf_probe_sample_identifier(struct perf_evsel *evsel)
+{
+	evsel->attr.sample_type |= PERF_SAMPLE_IDENTIFIER;
+}
+
+static void perf_probe_comm_exec(struct perf_evsel *evsel)
+{
+	evsel->attr.comm_exec = 1;
+}
+
+bool perf_can_sample_identifier(void)
+{
+	return perf_probe_api(perf_probe_sample_identifier);
+}
+
+static bool perf_can_comm_exec(void)
+{
+	return perf_probe_api(perf_probe_comm_exec);
+}
+
+void perf_evlist__config(struct perf_evlist *evlist, struct record_opts *opts)
+{
+	struct perf_evsel *evsel;
+	bool use_sample_identifier = false;
+	bool use_comm_exec;
+
+	/*
+	 * Set the evsel leader links before we configure attributes,
+	 * since some might depend on this info.
+	 */
+	if (opts->group)
+		perf_evlist__set_leader(evlist);
+
+	if (evlist->cpus->map[0] < 0)
+		opts->no_inherit = true;
+
+	use_comm_exec = perf_can_comm_exec();
+
+	evlist__for_each(evlist, evsel) {
+		perf_evsel__config(evsel, opts);
+		if (evsel->tracking && use_comm_exec)
+			evsel->attr.comm_exec = 1;
+	}
+
+	if (evlist->nr_entries > 1) {
+		struct perf_evsel *first = perf_evlist__first(evlist);
+
+		evlist__for_each(evlist, evsel) {
+			if (evsel->attr.sample_type == first->attr.sample_type)
+				continue;
+			use_sample_identifier = perf_can_sample_identifier();
+			break;
+		}
+		evlist__for_each(evlist, evsel)
+			perf_evsel__set_sample_id(evsel, use_sample_identifier);
+	}
+
+	perf_evlist__set_id_pos(evlist);
+}
+
+static int get_max_rate(unsigned int *rate)
+{
+	return sysctl__read_int("kernel/perf_event_max_sample_rate", (int *)rate);
+}
+
+static int record_opts__config_freq(struct record_opts *opts)
+{
+	bool user_freq = opts->user_freq != UINT_MAX;
+	unsigned int max_rate;
+
+	if (opts->user_interval != ULLONG_MAX)
+		opts->default_interval = opts->user_interval;
+	if (user_freq)
+		opts->freq = opts->user_freq;
+
+	/*
+	 * User specified count overrides default frequency.
+	 */
+	if (opts->default_interval)
+		opts->freq = 0;
+	else if (opts->freq) {
+		opts->default_interval = opts->freq;
+	} else {
+		pr_err("frequency and count are zero, aborting\n");
+		return -1;
+	}
+
+	if (get_max_rate(&max_rate))
+		return 0;
+
+	/*
+	 * User specified frequency is over current maximum.
+	 */
+	if (user_freq && (max_rate < opts->freq)) {
+		pr_err("Maximum frequency rate (%u) reached.\n"
+		   "Please use -F freq option with lower value or consider\n"
+		   "tweaking /proc/sys/kernel/perf_event_max_sample_rate.\n",
+		   max_rate);
+		return -1;
+	}
+
+	/*
+	 * Default frequency is over current maximum.
+	 */
+	if (max_rate < opts->freq) {
+		pr_warning("Lowering default frequency rate to %u.\n"
+			   "Please consider tweaking "
+			   "/proc/sys/kernel/perf_event_max_sample_rate.\n",
+			   max_rate);
+		opts->freq = max_rate;
+	}
+
+	return 0;
+}
+
+int record_opts__config(struct record_opts *opts)
+{
+	return record_opts__config_freq(opts);
+}
+
+bool perf_evlist__can_select_event(struct perf_evlist *evlist, const char *str)
+{
+	struct perf_evlist *temp_evlist;
+	struct perf_evsel *evsel;
+	int err, fd, cpu;
+	bool ret = false;
+	pid_t pid = -1;
+
+	temp_evlist = perf_evlist__new();
+	if (!temp_evlist)
+		return false;
+
+	err = parse_events(temp_evlist, str);
+	if (err)
+		goto out_delete;
+
+	evsel = perf_evlist__last(temp_evlist);
+
+	if (!evlist || cpu_map__empty(evlist->cpus)) {
+		struct cpu_map *cpus = cpu_map__new(NULL);
+
+		cpu =  cpus ? cpus->map[0] : 0;
+		cpu_map__delete(cpus);
+	} else {
+		cpu = evlist->cpus->map[0];
+	}
+
+	while (1) {
+		fd = sys_perf_event_open(&evsel->attr, pid, cpu, -1,
+					 perf_event_open_cloexec_flag());
+		if (fd < 0) {
+			if (pid == -1 && errno == EACCES) {
+				pid = 0;
+				continue;
+			}
+			goto out_delete;
+		}
+		break;
+	}
+	close(fd);
+	ret = true;
+
+out_delete:
+	perf_evlist__delete(temp_evlist);
+	return ret;
+}
diff --git a/tools/perf/util/run-command.c b/tools/perf/util/run-command.c
new file mode 100644
index 000000000..34622b53e
--- /dev/null
+++ b/tools/perf/util/run-command.c
@@ -0,0 +1,219 @@
+#include "cache.h"
+#include "run-command.h"
+#include "exec_cmd.h"
+#include "debug.h"
+
+static inline void close_pair(int fd[2])
+{
+	close(fd[0]);
+	close(fd[1]);
+}
+
+static inline void dup_devnull(int to)
+{
+	int fd = open("/dev/null", O_RDWR);
+	dup2(fd, to);
+	close(fd);
+}
+
+int start_command(struct child_process *cmd)
+{
+	int need_in, need_out, need_err;
+	int fdin[2], fdout[2], fderr[2];
+	char sbuf[STRERR_BUFSIZE];
+
+	/*
+	 * In case of errors we must keep the promise to close FDs
+	 * that have been passed in via ->in and ->out.
+	 */
+
+	need_in = !cmd->no_stdin && cmd->in < 0;
+	if (need_in) {
+		if (pipe(fdin) < 0) {
+			if (cmd->out > 0)
+				close(cmd->out);
+			return -ERR_RUN_COMMAND_PIPE;
+		}
+		cmd->in = fdin[1];
+	}
+
+	need_out = !cmd->no_stdout
+		&& !cmd->stdout_to_stderr
+		&& cmd->out < 0;
+	if (need_out) {
+		if (pipe(fdout) < 0) {
+			if (need_in)
+				close_pair(fdin);
+			else if (cmd->in)
+				close(cmd->in);
+			return -ERR_RUN_COMMAND_PIPE;
+		}
+		cmd->out = fdout[0];
+	}
+
+	need_err = !cmd->no_stderr && cmd->err < 0;
+	if (need_err) {
+		if (pipe(fderr) < 0) {
+			if (need_in)
+				close_pair(fdin);
+			else if (cmd->in)
+				close(cmd->in);
+			if (need_out)
+				close_pair(fdout);
+			else if (cmd->out)
+				close(cmd->out);
+			return -ERR_RUN_COMMAND_PIPE;
+		}
+		cmd->err = fderr[0];
+	}
+
+	fflush(NULL);
+	cmd->pid = fork();
+	if (!cmd->pid) {
+		if (cmd->no_stdin)
+			dup_devnull(0);
+		else if (need_in) {
+			dup2(fdin[0], 0);
+			close_pair(fdin);
+		} else if (cmd->in) {
+			dup2(cmd->in, 0);
+			close(cmd->in);
+		}
+
+		if (cmd->no_stderr)
+			dup_devnull(2);
+		else if (need_err) {
+			dup2(fderr[1], 2);
+			close_pair(fderr);
+		}
+
+		if (cmd->no_stdout)
+			dup_devnull(1);
+		else if (cmd->stdout_to_stderr)
+			dup2(2, 1);
+		else if (need_out) {
+			dup2(fdout[1], 1);
+			close_pair(fdout);
+		} else if (cmd->out > 1) {
+			dup2(cmd->out, 1);
+			close(cmd->out);
+		}
+
+		if (cmd->dir && chdir(cmd->dir))
+			die("exec %s: cd to %s failed (%s)", cmd->argv[0],
+			    cmd->dir, strerror_r(errno, sbuf, sizeof(sbuf)));
+		if (cmd->env) {
+			for (; *cmd->env; cmd->env++) {
+				if (strchr(*cmd->env, '='))
+					putenv((char*)*cmd->env);
+				else
+					unsetenv(*cmd->env);
+			}
+		}
+		if (cmd->preexec_cb)
+			cmd->preexec_cb();
+		if (cmd->perf_cmd) {
+			execv_perf_cmd(cmd->argv);
+		} else {
+			execvp(cmd->argv[0], (char *const*) cmd->argv);
+		}
+		exit(127);
+	}
+
+	if (cmd->pid < 0) {
+		int err = errno;
+		if (need_in)
+			close_pair(fdin);
+		else if (cmd->in)
+			close(cmd->in);
+		if (need_out)
+			close_pair(fdout);
+		else if (cmd->out)
+			close(cmd->out);
+		if (need_err)
+			close_pair(fderr);
+		return err == ENOENT ?
+			-ERR_RUN_COMMAND_EXEC :
+			-ERR_RUN_COMMAND_FORK;
+	}
+
+	if (need_in)
+		close(fdin[0]);
+	else if (cmd->in)
+		close(cmd->in);
+
+	if (need_out)
+		close(fdout[1]);
+	else if (cmd->out)
+		close(cmd->out);
+
+	if (need_err)
+		close(fderr[1]);
+
+	return 0;
+}
+
+static int wait_or_whine(pid_t pid)
+{
+	char sbuf[STRERR_BUFSIZE];
+
+	for (;;) {
+		int status, code;
+		pid_t waiting = waitpid(pid, &status, 0);
+
+		if (waiting < 0) {
+			if (errno == EINTR)
+				continue;
+			error("waitpid failed (%s)",
+			      strerror_r(errno, sbuf, sizeof(sbuf)));
+			return -ERR_RUN_COMMAND_WAITPID;
+		}
+		if (waiting != pid)
+			return -ERR_RUN_COMMAND_WAITPID_WRONG_PID;
+		if (WIFSIGNALED(status))
+			return -ERR_RUN_COMMAND_WAITPID_SIGNAL;
+
+		if (!WIFEXITED(status))
+			return -ERR_RUN_COMMAND_WAITPID_NOEXIT;
+		code = WEXITSTATUS(status);
+		switch (code) {
+		case 127:
+			return -ERR_RUN_COMMAND_EXEC;
+		case 0:
+			return 0;
+		default:
+			return -code;
+		}
+	}
+}
+
+int finish_command(struct child_process *cmd)
+{
+	return wait_or_whine(cmd->pid);
+}
+
+int run_command(struct child_process *cmd)
+{
+	int code = start_command(cmd);
+	if (code)
+		return code;
+	return finish_command(cmd);
+}
+
+static void prepare_run_command_v_opt(struct child_process *cmd,
+				      const char **argv,
+				      int opt)
+{
+	memset(cmd, 0, sizeof(*cmd));
+	cmd->argv = argv;
+	cmd->no_stdin = opt & RUN_COMMAND_NO_STDIN ? 1 : 0;
+	cmd->perf_cmd = opt & RUN_PERF_CMD ? 1 : 0;
+	cmd->stdout_to_stderr = opt & RUN_COMMAND_STDOUT_TO_STDERR ? 1 : 0;
+}
+
+int run_command_v_opt(const char **argv, int opt)
+{
+	struct child_process cmd;
+	prepare_run_command_v_opt(&cmd, argv, opt);
+	return run_command(&cmd);
+}
diff --git a/tools/perf/util/run-command.h b/tools/perf/util/run-command.h
new file mode 100644
index 000000000..1ef264d50
--- /dev/null
+++ b/tools/perf/util/run-command.h
@@ -0,0 +1,58 @@
+#ifndef __PERF_RUN_COMMAND_H
+#define __PERF_RUN_COMMAND_H
+
+enum {
+	ERR_RUN_COMMAND_FORK = 10000,
+	ERR_RUN_COMMAND_EXEC,
+	ERR_RUN_COMMAND_PIPE,
+	ERR_RUN_COMMAND_WAITPID,
+	ERR_RUN_COMMAND_WAITPID_WRONG_PID,
+	ERR_RUN_COMMAND_WAITPID_SIGNAL,
+	ERR_RUN_COMMAND_WAITPID_NOEXIT,
+};
+#define IS_RUN_COMMAND_ERR(x) (-(x) >= ERR_RUN_COMMAND_FORK)
+
+struct child_process {
+	const char **argv;
+	pid_t pid;
+	/*
+	 * Using .in, .out, .err:
+	 * - Specify 0 for no redirections (child inherits stdin, stdout,
+	 *   stderr from parent).
+	 * - Specify -1 to have a pipe allocated as follows:
+	 *     .in: returns the writable pipe end; parent writes to it,
+	 *          the readable pipe end becomes child's stdin
+	 *     .out, .err: returns the readable pipe end; parent reads from
+	 *          it, the writable pipe end becomes child's stdout/stderr
+	 *   The caller of start_command() must close the returned FDs
+	 *   after it has completed reading from/writing to it!
+	 * - Specify > 0 to set a channel to a particular FD as follows:
+	 *     .in: a readable FD, becomes child's stdin
+	 *     .out: a writable FD, becomes child's stdout/stderr
+	 *     .err > 0 not supported
+	 *   The specified FD is closed by start_command(), even in case
+	 *   of errors!
+	 */
+	int in;
+	int out;
+	int err;
+	const char *dir;
+	const char *const *env;
+	unsigned no_stdin:1;
+	unsigned no_stdout:1;
+	unsigned no_stderr:1;
+	unsigned perf_cmd:1; /* if this is to be perf sub-command */
+	unsigned stdout_to_stderr:1;
+	void (*preexec_cb)(void);
+};
+
+int start_command(struct child_process *);
+int finish_command(struct child_process *);
+int run_command(struct child_process *);
+
+#define RUN_COMMAND_NO_STDIN 1
+#define RUN_PERF_CMD	     2	/*If this is to be perf sub-command */
+#define RUN_COMMAND_STDOUT_TO_STDERR 4
+int run_command_v_opt(const char **argv, int opt);
+
+#endif /* __PERF_RUN_COMMAND_H */
diff --git a/tools/perf/util/scripting-engines/Build b/tools/perf/util/scripting-engines/Build
new file mode 100644
index 000000000..6516e220c
--- /dev/null
+++ b/tools/perf/util/scripting-engines/Build
@@ -0,0 +1,6 @@
+libperf-$(CONFIG_LIBPERL)   += trace-event-perl.o
+libperf-$(CONFIG_LIBPYTHON) += trace-event-python.o
+
+CFLAGS_trace-event-perl.o += $(PERL_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow -Wno-undef -Wno-switch-default
+
+CFLAGS_trace-event-python.o += $(PYTHON_EMBED_CCOPTS) -Wno-redundant-decls -Wno-strict-prototypes -Wno-unused-parameter -Wno-shadow
diff --git a/tools/perf/util/scripting-engines/trace-event-perl.c b/tools/perf/util/scripting-engines/trace-event-perl.c
new file mode 100644
index 000000000..430b5d278
--- /dev/null
+++ b/tools/perf/util/scripting-engines/trace-event-perl.c
@@ -0,0 +1,632 @@
+/*
+ * trace-event-perl.  Feed perf script events to an embedded Perl interpreter.
+ *
+ * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+#include <linux/bitmap.h>
+
+#include "../util.h"
+#include <EXTERN.h>
+#include <perl.h>
+
+#include "../../perf.h"
+#include "../thread.h"
+#include "../event.h"
+#include "../trace-event.h"
+#include "../evsel.h"
+#include "../debug.h"
+
+void boot_Perf__Trace__Context(pTHX_ CV *cv);
+void boot_DynaLoader(pTHX_ CV *cv);
+typedef PerlInterpreter * INTERP;
+
+void xs_init(pTHX);
+
+void xs_init(pTHX)
+{
+	const char *file = __FILE__;
+	dXSUB_SYS;
+
+	newXS("Perf::Trace::Context::bootstrap", boot_Perf__Trace__Context,
+	      file);
+	newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file);
+}
+
+INTERP my_perl;
+
+#define FTRACE_MAX_EVENT				\
+	((1 << (sizeof(unsigned short) * 8)) - 1)
+
+static DECLARE_BITMAP(events_defined, FTRACE_MAX_EVENT);
+
+extern struct scripting_context *scripting_context;
+
+static char *cur_field_name;
+static int zero_flag_atom;
+
+static void define_symbolic_value(const char *ev_name,
+				  const char *field_name,
+				  const char *field_value,
+				  const char *field_str)
+{
+	unsigned long long value;
+	dSP;
+
+	value = eval_flag(field_value);
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+	XPUSHs(sv_2mortal(newSVuv(value)));
+	XPUSHs(sv_2mortal(newSVpv(field_str, 0)));
+
+	PUTBACK;
+	if (get_cv("main::define_symbolic_value", 0))
+		call_pv("main::define_symbolic_value", G_SCALAR);
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void define_symbolic_values(struct print_flag_sym *field,
+				   const char *ev_name,
+				   const char *field_name)
+{
+	define_symbolic_value(ev_name, field_name, field->value, field->str);
+	if (field->next)
+		define_symbolic_values(field->next, ev_name, field_name);
+}
+
+static void define_symbolic_field(const char *ev_name,
+				  const char *field_name)
+{
+	dSP;
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+
+	PUTBACK;
+	if (get_cv("main::define_symbolic_field", 0))
+		call_pv("main::define_symbolic_field", G_SCALAR);
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void define_flag_value(const char *ev_name,
+			      const char *field_name,
+			      const char *field_value,
+			      const char *field_str)
+{
+	unsigned long long value;
+	dSP;
+
+	value = eval_flag(field_value);
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+	XPUSHs(sv_2mortal(newSVuv(value)));
+	XPUSHs(sv_2mortal(newSVpv(field_str, 0)));
+
+	PUTBACK;
+	if (get_cv("main::define_flag_value", 0))
+		call_pv("main::define_flag_value", G_SCALAR);
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void define_flag_values(struct print_flag_sym *field,
+			       const char *ev_name,
+			       const char *field_name)
+{
+	define_flag_value(ev_name, field_name, field->value, field->str);
+	if (field->next)
+		define_flag_values(field->next, ev_name, field_name);
+}
+
+static void define_flag_field(const char *ev_name,
+			      const char *field_name,
+			      const char *delim)
+{
+	dSP;
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(ev_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(field_name, 0)));
+	XPUSHs(sv_2mortal(newSVpv(delim, 0)));
+
+	PUTBACK;
+	if (get_cv("main::define_flag_field", 0))
+		call_pv("main::define_flag_field", G_SCALAR);
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void define_event_symbols(struct event_format *event,
+				 const char *ev_name,
+				 struct print_arg *args)
+{
+	switch (args->type) {
+	case PRINT_NULL:
+		break;
+	case PRINT_ATOM:
+		define_flag_value(ev_name, cur_field_name, "0",
+				  args->atom.atom);
+		zero_flag_atom = 0;
+		break;
+	case PRINT_FIELD:
+		free(cur_field_name);
+		cur_field_name = strdup(args->field.name);
+		break;
+	case PRINT_FLAGS:
+		define_event_symbols(event, ev_name, args->flags.field);
+		define_flag_field(ev_name, cur_field_name, args->flags.delim);
+		define_flag_values(args->flags.flags, ev_name, cur_field_name);
+		break;
+	case PRINT_SYMBOL:
+		define_event_symbols(event, ev_name, args->symbol.field);
+		define_symbolic_field(ev_name, cur_field_name);
+		define_symbolic_values(args->symbol.symbols, ev_name,
+				       cur_field_name);
+		break;
+	case PRINT_HEX:
+		define_event_symbols(event, ev_name, args->hex.field);
+		define_event_symbols(event, ev_name, args->hex.size);
+		break;
+	case PRINT_INT_ARRAY:
+		define_event_symbols(event, ev_name, args->int_array.field);
+		define_event_symbols(event, ev_name, args->int_array.count);
+		define_event_symbols(event, ev_name, args->int_array.el_size);
+		break;
+	case PRINT_BSTRING:
+	case PRINT_DYNAMIC_ARRAY:
+	case PRINT_STRING:
+	case PRINT_BITMASK:
+		break;
+	case PRINT_TYPE:
+		define_event_symbols(event, ev_name, args->typecast.item);
+		break;
+	case PRINT_OP:
+		if (strcmp(args->op.op, ":") == 0)
+			zero_flag_atom = 1;
+		define_event_symbols(event, ev_name, args->op.left);
+		define_event_symbols(event, ev_name, args->op.right);
+		break;
+	case PRINT_FUNC:
+	default:
+		pr_err("Unsupported print arg type\n");
+		/* we should warn... */
+		return;
+	}
+
+	if (args->next)
+		define_event_symbols(event, ev_name, args->next);
+}
+
+static void perl_process_tracepoint(struct perf_sample *sample,
+				    struct perf_evsel *evsel,
+				    struct thread *thread)
+{
+	struct event_format *event = evsel->tp_format;
+	struct format_field *field;
+	static char handler[256];
+	unsigned long long val;
+	unsigned long s, ns;
+	int pid;
+	int cpu = sample->cpu;
+	void *data = sample->raw_data;
+	unsigned long long nsecs = sample->time;
+	const char *comm = thread__comm_str(thread);
+
+	dSP;
+
+	if (evsel->attr.type != PERF_TYPE_TRACEPOINT)
+		return;
+
+	if (!event)
+		die("ug! no event found for type %" PRIu64, (u64)evsel->attr.config);
+
+	pid = raw_field_value(event, "common_pid", data);
+
+	sprintf(handler, "%s::%s", event->system, event->name);
+
+	if (!test_and_set_bit(event->id, events_defined))
+		define_event_symbols(event, handler, event->print_fmt.args);
+
+	s = nsecs / NSECS_PER_SEC;
+	ns = nsecs - s * NSECS_PER_SEC;
+
+	scripting_context->event_data = data;
+	scripting_context->pevent = evsel->tp_format->pevent;
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+
+	XPUSHs(sv_2mortal(newSVpv(handler, 0)));
+	XPUSHs(sv_2mortal(newSViv(PTR2IV(scripting_context))));
+	XPUSHs(sv_2mortal(newSVuv(cpu)));
+	XPUSHs(sv_2mortal(newSVuv(s)));
+	XPUSHs(sv_2mortal(newSVuv(ns)));
+	XPUSHs(sv_2mortal(newSViv(pid)));
+	XPUSHs(sv_2mortal(newSVpv(comm, 0)));
+
+	/* common fields other than pid can be accessed via xsub fns */
+
+	for (field = event->format.fields; field; field = field->next) {
+		if (field->flags & FIELD_IS_STRING) {
+			int offset;
+			if (field->flags & FIELD_IS_DYNAMIC) {
+				offset = *(int *)(data + field->offset);
+				offset &= 0xffff;
+			} else
+				offset = field->offset;
+			XPUSHs(sv_2mortal(newSVpv((char *)data + offset, 0)));
+		} else { /* FIELD_IS_NUMERIC */
+			val = read_size(event, data + field->offset,
+					field->size);
+			if (field->flags & FIELD_IS_SIGNED) {
+				XPUSHs(sv_2mortal(newSViv(val)));
+			} else {
+				XPUSHs(sv_2mortal(newSVuv(val)));
+			}
+		}
+	}
+
+	PUTBACK;
+
+	if (get_cv(handler, 0))
+		call_pv(handler, G_SCALAR);
+	else if (get_cv("main::trace_unhandled", 0)) {
+		XPUSHs(sv_2mortal(newSVpv(handler, 0)));
+		XPUSHs(sv_2mortal(newSViv(PTR2IV(scripting_context))));
+		XPUSHs(sv_2mortal(newSVuv(cpu)));
+		XPUSHs(sv_2mortal(newSVuv(nsecs)));
+		XPUSHs(sv_2mortal(newSViv(pid)));
+		XPUSHs(sv_2mortal(newSVpv(comm, 0)));
+		call_pv("main::trace_unhandled", G_SCALAR);
+	}
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void perl_process_event_generic(union perf_event *event,
+				       struct perf_sample *sample,
+				       struct perf_evsel *evsel)
+{
+	dSP;
+
+	if (!get_cv("process_event", 0))
+		return;
+
+	ENTER;
+	SAVETMPS;
+	PUSHMARK(SP);
+	XPUSHs(sv_2mortal(newSVpvn((const char *)event, event->header.size)));
+	XPUSHs(sv_2mortal(newSVpvn((const char *)&evsel->attr, sizeof(evsel->attr))));
+	XPUSHs(sv_2mortal(newSVpvn((const char *)sample, sizeof(*sample))));
+	XPUSHs(sv_2mortal(newSVpvn((const char *)sample->raw_data, sample->raw_size)));
+	PUTBACK;
+	call_pv("process_event", G_SCALAR);
+	SPAGAIN;
+	PUTBACK;
+	FREETMPS;
+	LEAVE;
+}
+
+static void perl_process_event(union perf_event *event,
+			       struct perf_sample *sample,
+			       struct perf_evsel *evsel,
+			       struct addr_location *al)
+{
+	perl_process_tracepoint(sample, evsel, al->thread);
+	perl_process_event_generic(event, sample, evsel);
+}
+
+static void run_start_sub(void)
+{
+	dSP; /* access to Perl stack */
+	PUSHMARK(SP);
+
+	if (get_cv("main::trace_begin", 0))
+		call_pv("main::trace_begin", G_DISCARD | G_NOARGS);
+}
+
+/*
+ * Start trace script
+ */
+static int perl_start_script(const char *script, int argc, const char **argv)
+{
+	const char **command_line;
+	int i, err = 0;
+
+	command_line = malloc((argc + 2) * sizeof(const char *));
+	command_line[0] = "";
+	command_line[1] = script;
+	for (i = 2; i < argc + 2; i++)
+		command_line[i] = argv[i - 2];
+
+	my_perl = perl_alloc();
+	perl_construct(my_perl);
+
+	if (perl_parse(my_perl, xs_init, argc + 2, (char **)command_line,
+		       (char **)NULL)) {
+		err = -1;
+		goto error;
+	}
+
+	if (perl_run(my_perl)) {
+		err = -1;
+		goto error;
+	}
+
+	if (SvTRUE(ERRSV)) {
+		err = -1;
+		goto error;
+	}
+
+	run_start_sub();
+
+	free(command_line);
+	return 0;
+error:
+	perl_free(my_perl);
+	free(command_line);
+
+	return err;
+}
+
+static int perl_flush_script(void)
+{
+	return 0;
+}
+
+/*
+ * Stop trace script
+ */
+static int perl_stop_script(void)
+{
+	dSP; /* access to Perl stack */
+	PUSHMARK(SP);
+
+	if (get_cv("main::trace_end", 0))
+		call_pv("main::trace_end", G_DISCARD | G_NOARGS);
+
+	perl_destruct(my_perl);
+	perl_free(my_perl);
+
+	return 0;
+}
+
+static int perl_generate_script(struct pevent *pevent, const char *outfile)
+{
+	struct event_format *event = NULL;
+	struct format_field *f;
+	char fname[PATH_MAX];
+	int not_first, count;
+	FILE *ofp;
+
+	sprintf(fname, "%s.pl", outfile);
+	ofp = fopen(fname, "w");
+	if (ofp == NULL) {
+		fprintf(stderr, "couldn't open %s\n", fname);
+		return -1;
+	}
+
+	fprintf(ofp, "# perf script event handlers, "
+		"generated by perf script -g perl\n");
+
+	fprintf(ofp, "# Licensed under the terms of the GNU GPL"
+		" License version 2\n\n");
+
+	fprintf(ofp, "# The common_* event handler fields are the most useful "
+		"fields common to\n");
+
+	fprintf(ofp, "# all events.  They don't necessarily correspond to "
+		"the 'common_*' fields\n");
+
+	fprintf(ofp, "# in the format files.  Those fields not available as "
+		"handler params can\n");
+
+	fprintf(ofp, "# be retrieved using Perl functions of the form "
+		"common_*($context).\n");
+
+	fprintf(ofp, "# See Context.pm for the list of available "
+		"functions.\n\n");
+
+	fprintf(ofp, "use lib \"$ENV{'PERF_EXEC_PATH'}/scripts/perl/"
+		"Perf-Trace-Util/lib\";\n");
+
+	fprintf(ofp, "use lib \"./Perf-Trace-Util/lib\";\n");
+	fprintf(ofp, "use Perf::Trace::Core;\n");
+	fprintf(ofp, "use Perf::Trace::Context;\n");
+	fprintf(ofp, "use Perf::Trace::Util;\n\n");
+
+	fprintf(ofp, "sub trace_begin\n{\n\t# optional\n}\n\n");
+	fprintf(ofp, "sub trace_end\n{\n\t# optional\n}\n\n");
+
+	while ((event = trace_find_next_event(pevent, event))) {
+		fprintf(ofp, "sub %s::%s\n{\n", event->system, event->name);
+		fprintf(ofp, "\tmy (");
+
+		fprintf(ofp, "$event_name, ");
+		fprintf(ofp, "$context, ");
+		fprintf(ofp, "$common_cpu, ");
+		fprintf(ofp, "$common_secs, ");
+		fprintf(ofp, "$common_nsecs,\n");
+		fprintf(ofp, "\t    $common_pid, ");
+		fprintf(ofp, "$common_comm,\n\t    ");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t    ");
+
+			fprintf(ofp, "$%s", f->name);
+		}
+		fprintf(ofp, ") = @_;\n\n");
+
+		fprintf(ofp, "\tprint_header($event_name, $common_cpu, "
+			"$common_secs, $common_nsecs,\n\t             "
+			"$common_pid, $common_comm);\n\n");
+
+		fprintf(ofp, "\tprintf(\"");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (count && count % 4 == 0) {
+				fprintf(ofp, "\".\n\t       \"");
+			}
+			count++;
+
+			fprintf(ofp, "%s=", f->name);
+			if (f->flags & FIELD_IS_STRING ||
+			    f->flags & FIELD_IS_FLAG ||
+			    f->flags & FIELD_IS_SYMBOLIC)
+				fprintf(ofp, "%%s");
+			else if (f->flags & FIELD_IS_SIGNED)
+				fprintf(ofp, "%%d");
+			else
+				fprintf(ofp, "%%u");
+		}
+
+		fprintf(ofp, "\\n\",\n\t       ");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t       ");
+
+			if (f->flags & FIELD_IS_FLAG) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t       ");
+					count = 4;
+				}
+				fprintf(ofp, "flag_str(\"");
+				fprintf(ofp, "%s::%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", $%s)", f->name,
+					f->name);
+			} else if (f->flags & FIELD_IS_SYMBOLIC) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t       ");
+					count = 4;
+				}
+				fprintf(ofp, "symbol_str(\"");
+				fprintf(ofp, "%s::%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", $%s)", f->name,
+					f->name);
+			} else
+				fprintf(ofp, "$%s", f->name);
+		}
+
+		fprintf(ofp, ");\n");
+		fprintf(ofp, "}\n\n");
+	}
+
+	fprintf(ofp, "sub trace_unhandled\n{\n\tmy ($event_name, $context, "
+		"$common_cpu, $common_secs, $common_nsecs,\n\t    "
+		"$common_pid, $common_comm) = @_;\n\n");
+
+	fprintf(ofp, "\tprint_header($event_name, $common_cpu, "
+		"$common_secs, $common_nsecs,\n\t             $common_pid, "
+		"$common_comm);\n}\n\n");
+
+	fprintf(ofp, "sub print_header\n{\n"
+		"\tmy ($event_name, $cpu, $secs, $nsecs, $pid, $comm) = @_;\n\n"
+		"\tprintf(\"%%-20s %%5u %%05u.%%09u %%8u %%-20s \",\n\t       "
+		"$event_name, $cpu, $secs, $nsecs, $pid, $comm);\n}\n");
+
+	fprintf(ofp,
+		"\n# Packed byte string args of process_event():\n"
+		"#\n"
+		"# $event:\tunion perf_event\tutil/event.h\n"
+		"# $attr:\tstruct perf_event_attr\tlinux/perf_event.h\n"
+		"# $sample:\tstruct perf_sample\tutil/event.h\n"
+		"# $raw_data:\tperf_sample->raw_data\tutil/event.h\n"
+		"\n"
+		"sub process_event\n"
+		"{\n"
+		"\tmy ($event, $attr, $sample, $raw_data) = @_;\n"
+		"\n"
+		"\tmy @event\t= unpack(\"LSS\", $event);\n"
+		"\tmy @attr\t= unpack(\"LLQQQQQLLQQ\", $attr);\n"
+		"\tmy @sample\t= unpack(\"QLLQQQQQLL\", $sample);\n"
+		"\tmy @raw_data\t= unpack(\"C*\", $raw_data);\n"
+		"\n"
+		"\tuse Data::Dumper;\n"
+		"\tprint Dumper \\@event, \\@attr, \\@sample, \\@raw_data;\n"
+		"}\n");
+
+	fclose(ofp);
+
+	fprintf(stderr, "generated Perl script: %s\n", fname);
+
+	return 0;
+}
+
+struct scripting_ops perl_scripting_ops = {
+	.name = "Perl",
+	.start_script = perl_start_script,
+	.flush_script = perl_flush_script,
+	.stop_script = perl_stop_script,
+	.process_event = perl_process_event,
+	.generate_script = perl_generate_script,
+};
diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c
new file mode 100644
index 000000000..5544b8cdd
--- /dev/null
+++ b/tools/perf/util/scripting-engines/trace-event-python.c
@@ -0,0 +1,1209 @@
+/*
+ * trace-event-python.  Feed trace events to an embedded Python interpreter.
+ *
+ * Copyright (C) 2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <Python.h>
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdbool.h>
+#include <errno.h>
+#include <linux/bitmap.h>
+
+#include "../../perf.h"
+#include "../debug.h"
+#include "../callchain.h"
+#include "../evsel.h"
+#include "../util.h"
+#include "../event.h"
+#include "../thread.h"
+#include "../comm.h"
+#include "../machine.h"
+#include "../db-export.h"
+#include "../thread-stack.h"
+#include "../trace-event.h"
+#include "../machine.h"
+
+PyMODINIT_FUNC initperf_trace_context(void);
+
+#define FTRACE_MAX_EVENT				\
+	((1 << (sizeof(unsigned short) * 8)) - 1)
+
+static DECLARE_BITMAP(events_defined, FTRACE_MAX_EVENT);
+
+#define MAX_FIELDS	64
+#define N_COMMON_FIELDS	7
+
+extern struct scripting_context *scripting_context;
+
+static char *cur_field_name;
+static int zero_flag_atom;
+
+static PyObject *main_module, *main_dict;
+
+struct tables {
+	struct db_export	dbe;
+	PyObject		*evsel_handler;
+	PyObject		*machine_handler;
+	PyObject		*thread_handler;
+	PyObject		*comm_handler;
+	PyObject		*comm_thread_handler;
+	PyObject		*dso_handler;
+	PyObject		*symbol_handler;
+	PyObject		*branch_type_handler;
+	PyObject		*sample_handler;
+	PyObject		*call_path_handler;
+	PyObject		*call_return_handler;
+	bool			db_export_mode;
+};
+
+static struct tables tables_global;
+
+static void handler_call_die(const char *handler_name) NORETURN;
+static void handler_call_die(const char *handler_name)
+{
+	PyErr_Print();
+	Py_FatalError("problem in Python trace event handler");
+	// Py_FatalError does not return
+	// but we have to make the compiler happy
+	abort();
+}
+
+/*
+ * Insert val into into the dictionary and decrement the reference counter.
+ * This is necessary for dictionaries since PyDict_SetItemString() does not
+ * steal a reference, as opposed to PyTuple_SetItem().
+ */
+static void pydict_set_item_string_decref(PyObject *dict, const char *key, PyObject *val)
+{
+	PyDict_SetItemString(dict, key, val);
+	Py_DECREF(val);
+}
+
+static PyObject *get_handler(const char *handler_name)
+{
+	PyObject *handler;
+
+	handler = PyDict_GetItemString(main_dict, handler_name);
+	if (handler && !PyCallable_Check(handler))
+		return NULL;
+	return handler;
+}
+
+static void call_object(PyObject *handler, PyObject *args, const char *die_msg)
+{
+	PyObject *retval;
+
+	retval = PyObject_CallObject(handler, args);
+	if (retval == NULL)
+		handler_call_die(die_msg);
+	Py_DECREF(retval);
+}
+
+static void try_call_object(const char *handler_name, PyObject *args)
+{
+	PyObject *handler;
+
+	handler = get_handler(handler_name);
+	if (handler)
+		call_object(handler, args, handler_name);
+}
+
+static void define_value(enum print_arg_type field_type,
+			 const char *ev_name,
+			 const char *field_name,
+			 const char *field_value,
+			 const char *field_str)
+{
+	const char *handler_name = "define_flag_value";
+	PyObject *t;
+	unsigned long long value;
+	unsigned n = 0;
+
+	if (field_type == PRINT_SYMBOL)
+		handler_name = "define_symbolic_value";
+
+	t = PyTuple_New(4);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	value = eval_flag(field_value);
+
+	PyTuple_SetItem(t, n++, PyString_FromString(ev_name));
+	PyTuple_SetItem(t, n++, PyString_FromString(field_name));
+	PyTuple_SetItem(t, n++, PyInt_FromLong(value));
+	PyTuple_SetItem(t, n++, PyString_FromString(field_str));
+
+	try_call_object(handler_name, t);
+
+	Py_DECREF(t);
+}
+
+static void define_values(enum print_arg_type field_type,
+			  struct print_flag_sym *field,
+			  const char *ev_name,
+			  const char *field_name)
+{
+	define_value(field_type, ev_name, field_name, field->value,
+		     field->str);
+
+	if (field->next)
+		define_values(field_type, field->next, ev_name, field_name);
+}
+
+static void define_field(enum print_arg_type field_type,
+			 const char *ev_name,
+			 const char *field_name,
+			 const char *delim)
+{
+	const char *handler_name = "define_flag_field";
+	PyObject *t;
+	unsigned n = 0;
+
+	if (field_type == PRINT_SYMBOL)
+		handler_name = "define_symbolic_field";
+
+	if (field_type == PRINT_FLAGS)
+		t = PyTuple_New(3);
+	else
+		t = PyTuple_New(2);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	PyTuple_SetItem(t, n++, PyString_FromString(ev_name));
+	PyTuple_SetItem(t, n++, PyString_FromString(field_name));
+	if (field_type == PRINT_FLAGS)
+		PyTuple_SetItem(t, n++, PyString_FromString(delim));
+
+	try_call_object(handler_name, t);
+
+	Py_DECREF(t);
+}
+
+static void define_event_symbols(struct event_format *event,
+				 const char *ev_name,
+				 struct print_arg *args)
+{
+	switch (args->type) {
+	case PRINT_NULL:
+		break;
+	case PRINT_ATOM:
+		define_value(PRINT_FLAGS, ev_name, cur_field_name, "0",
+			     args->atom.atom);
+		zero_flag_atom = 0;
+		break;
+	case PRINT_FIELD:
+		free(cur_field_name);
+		cur_field_name = strdup(args->field.name);
+		break;
+	case PRINT_FLAGS:
+		define_event_symbols(event, ev_name, args->flags.field);
+		define_field(PRINT_FLAGS, ev_name, cur_field_name,
+			     args->flags.delim);
+		define_values(PRINT_FLAGS, args->flags.flags, ev_name,
+			      cur_field_name);
+		break;
+	case PRINT_SYMBOL:
+		define_event_symbols(event, ev_name, args->symbol.field);
+		define_field(PRINT_SYMBOL, ev_name, cur_field_name, NULL);
+		define_values(PRINT_SYMBOL, args->symbol.symbols, ev_name,
+			      cur_field_name);
+		break;
+	case PRINT_HEX:
+		define_event_symbols(event, ev_name, args->hex.field);
+		define_event_symbols(event, ev_name, args->hex.size);
+		break;
+	case PRINT_INT_ARRAY:
+		define_event_symbols(event, ev_name, args->int_array.field);
+		define_event_symbols(event, ev_name, args->int_array.count);
+		define_event_symbols(event, ev_name, args->int_array.el_size);
+		break;
+	case PRINT_STRING:
+		break;
+	case PRINT_TYPE:
+		define_event_symbols(event, ev_name, args->typecast.item);
+		break;
+	case PRINT_OP:
+		if (strcmp(args->op.op, ":") == 0)
+			zero_flag_atom = 1;
+		define_event_symbols(event, ev_name, args->op.left);
+		define_event_symbols(event, ev_name, args->op.right);
+		break;
+	default:
+		/* gcc warns for these? */
+	case PRINT_BSTRING:
+	case PRINT_DYNAMIC_ARRAY:
+	case PRINT_FUNC:
+	case PRINT_BITMASK:
+		/* we should warn... */
+		return;
+	}
+
+	if (args->next)
+		define_event_symbols(event, ev_name, args->next);
+}
+
+static PyObject *get_field_numeric_entry(struct event_format *event,
+		struct format_field *field, void *data)
+{
+	bool is_array = field->flags & FIELD_IS_ARRAY;
+	PyObject *obj, *list = NULL;
+	unsigned long long val;
+	unsigned int item_size, n_items, i;
+
+	if (is_array) {
+		list = PyList_New(field->arraylen);
+		item_size = field->size / field->arraylen;
+		n_items = field->arraylen;
+	} else {
+		item_size = field->size;
+		n_items = 1;
+	}
+
+	for (i = 0; i < n_items; i++) {
+
+		val = read_size(event, data + field->offset + i * item_size,
+				item_size);
+		if (field->flags & FIELD_IS_SIGNED) {
+			if ((long long)val >= LONG_MIN &&
+					(long long)val <= LONG_MAX)
+				obj = PyInt_FromLong(val);
+			else
+				obj = PyLong_FromLongLong(val);
+		} else {
+			if (val <= LONG_MAX)
+				obj = PyInt_FromLong(val);
+			else
+				obj = PyLong_FromUnsignedLongLong(val);
+		}
+		if (is_array)
+			PyList_SET_ITEM(list, i, obj);
+	}
+	if (is_array)
+		obj = list;
+	return obj;
+}
+
+
+static PyObject *python_process_callchain(struct perf_sample *sample,
+					 struct perf_evsel *evsel,
+					 struct addr_location *al)
+{
+	PyObject *pylist;
+
+	pylist = PyList_New(0);
+	if (!pylist)
+		Py_FatalError("couldn't create Python list");
+
+	if (!symbol_conf.use_callchain || !sample->callchain)
+		goto exit;
+
+	if (thread__resolve_callchain(al->thread, evsel,
+				      sample, NULL, NULL,
+				      PERF_MAX_STACK_DEPTH) != 0) {
+		pr_err("Failed to resolve callchain. Skipping\n");
+		goto exit;
+	}
+	callchain_cursor_commit(&callchain_cursor);
+
+
+	while (1) {
+		PyObject *pyelem;
+		struct callchain_cursor_node *node;
+		node = callchain_cursor_current(&callchain_cursor);
+		if (!node)
+			break;
+
+		pyelem = PyDict_New();
+		if (!pyelem)
+			Py_FatalError("couldn't create Python dictionary");
+
+
+		pydict_set_item_string_decref(pyelem, "ip",
+				PyLong_FromUnsignedLongLong(node->ip));
+
+		if (node->sym) {
+			PyObject *pysym  = PyDict_New();
+			if (!pysym)
+				Py_FatalError("couldn't create Python dictionary");
+			pydict_set_item_string_decref(pysym, "start",
+					PyLong_FromUnsignedLongLong(node->sym->start));
+			pydict_set_item_string_decref(pysym, "end",
+					PyLong_FromUnsignedLongLong(node->sym->end));
+			pydict_set_item_string_decref(pysym, "binding",
+					PyInt_FromLong(node->sym->binding));
+			pydict_set_item_string_decref(pysym, "name",
+					PyString_FromStringAndSize(node->sym->name,
+							node->sym->namelen));
+			pydict_set_item_string_decref(pyelem, "sym", pysym);
+		}
+
+		if (node->map) {
+			struct map *map = node->map;
+			const char *dsoname = "[unknown]";
+			if (map && map->dso && (map->dso->name || map->dso->long_name)) {
+				if (symbol_conf.show_kernel_path && map->dso->long_name)
+					dsoname = map->dso->long_name;
+				else if (map->dso->name)
+					dsoname = map->dso->name;
+			}
+			pydict_set_item_string_decref(pyelem, "dso",
+					PyString_FromString(dsoname));
+		}
+
+		callchain_cursor_advance(&callchain_cursor);
+		PyList_Append(pylist, pyelem);
+		Py_DECREF(pyelem);
+	}
+
+exit:
+	return pylist;
+}
+
+
+static void python_process_tracepoint(struct perf_sample *sample,
+				      struct perf_evsel *evsel,
+				      struct addr_location *al)
+{
+	struct event_format *event = evsel->tp_format;
+	PyObject *handler, *context, *t, *obj, *callchain;
+	PyObject *dict = NULL;
+	static char handler_name[256];
+	struct format_field *field;
+	unsigned long s, ns;
+	unsigned n = 0;
+	int pid;
+	int cpu = sample->cpu;
+	void *data = sample->raw_data;
+	unsigned long long nsecs = sample->time;
+	const char *comm = thread__comm_str(al->thread);
+
+	t = PyTuple_New(MAX_FIELDS);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	if (!event)
+		die("ug! no event found for type %d", (int)evsel->attr.config);
+
+	pid = raw_field_value(event, "common_pid", data);
+
+	sprintf(handler_name, "%s__%s", event->system, event->name);
+
+	if (!test_and_set_bit(event->id, events_defined))
+		define_event_symbols(event, handler_name, event->print_fmt.args);
+
+	handler = get_handler(handler_name);
+	if (!handler) {
+		dict = PyDict_New();
+		if (!dict)
+			Py_FatalError("couldn't create Python dict");
+	}
+	s = nsecs / NSECS_PER_SEC;
+	ns = nsecs - s * NSECS_PER_SEC;
+
+	scripting_context->event_data = data;
+	scripting_context->pevent = evsel->tp_format->pevent;
+
+	context = PyCObject_FromVoidPtr(scripting_context, NULL);
+
+	PyTuple_SetItem(t, n++, PyString_FromString(handler_name));
+	PyTuple_SetItem(t, n++, context);
+
+	/* ip unwinding */
+	callchain = python_process_callchain(sample, evsel, al);
+
+	if (handler) {
+		PyTuple_SetItem(t, n++, PyInt_FromLong(cpu));
+		PyTuple_SetItem(t, n++, PyInt_FromLong(s));
+		PyTuple_SetItem(t, n++, PyInt_FromLong(ns));
+		PyTuple_SetItem(t, n++, PyInt_FromLong(pid));
+		PyTuple_SetItem(t, n++, PyString_FromString(comm));
+		PyTuple_SetItem(t, n++, callchain);
+	} else {
+		pydict_set_item_string_decref(dict, "common_cpu", PyInt_FromLong(cpu));
+		pydict_set_item_string_decref(dict, "common_s", PyInt_FromLong(s));
+		pydict_set_item_string_decref(dict, "common_ns", PyInt_FromLong(ns));
+		pydict_set_item_string_decref(dict, "common_pid", PyInt_FromLong(pid));
+		pydict_set_item_string_decref(dict, "common_comm", PyString_FromString(comm));
+		pydict_set_item_string_decref(dict, "common_callchain", callchain);
+	}
+	for (field = event->format.fields; field; field = field->next) {
+		if (field->flags & FIELD_IS_STRING) {
+			int offset;
+			if (field->flags & FIELD_IS_DYNAMIC) {
+				offset = *(int *)(data + field->offset);
+				offset &= 0xffff;
+			} else
+				offset = field->offset;
+			obj = PyString_FromString((char *)data + offset);
+		} else { /* FIELD_IS_NUMERIC */
+			obj = get_field_numeric_entry(event, field, data);
+		}
+		if (handler)
+			PyTuple_SetItem(t, n++, obj);
+		else
+			pydict_set_item_string_decref(dict, field->name, obj);
+
+	}
+
+	if (!handler)
+		PyTuple_SetItem(t, n++, dict);
+
+	if (_PyTuple_Resize(&t, n) == -1)
+		Py_FatalError("error resizing Python tuple");
+
+	if (handler) {
+		call_object(handler, t, handler_name);
+	} else {
+		try_call_object("trace_unhandled", t);
+		Py_DECREF(dict);
+	}
+
+	Py_DECREF(t);
+}
+
+static PyObject *tuple_new(unsigned int sz)
+{
+	PyObject *t;
+
+	t = PyTuple_New(sz);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+	return t;
+}
+
+static int tuple_set_u64(PyObject *t, unsigned int pos, u64 val)
+{
+#if BITS_PER_LONG == 64
+	return PyTuple_SetItem(t, pos, PyInt_FromLong(val));
+#endif
+#if BITS_PER_LONG == 32
+	return PyTuple_SetItem(t, pos, PyLong_FromLongLong(val));
+#endif
+}
+
+static int tuple_set_s32(PyObject *t, unsigned int pos, s32 val)
+{
+	return PyTuple_SetItem(t, pos, PyInt_FromLong(val));
+}
+
+static int tuple_set_string(PyObject *t, unsigned int pos, const char *s)
+{
+	return PyTuple_SetItem(t, pos, PyString_FromString(s));
+}
+
+static int python_export_evsel(struct db_export *dbe, struct perf_evsel *evsel)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	PyObject *t;
+
+	t = tuple_new(2);
+
+	tuple_set_u64(t, 0, evsel->db_id);
+	tuple_set_string(t, 1, perf_evsel__name(evsel));
+
+	call_object(tables->evsel_handler, t, "evsel_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_machine(struct db_export *dbe,
+				 struct machine *machine)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	PyObject *t;
+
+	t = tuple_new(3);
+
+	tuple_set_u64(t, 0, machine->db_id);
+	tuple_set_s32(t, 1, machine->pid);
+	tuple_set_string(t, 2, machine->root_dir ? machine->root_dir : "");
+
+	call_object(tables->machine_handler, t, "machine_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_thread(struct db_export *dbe, struct thread *thread,
+				u64 main_thread_db_id, struct machine *machine)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	PyObject *t;
+
+	t = tuple_new(5);
+
+	tuple_set_u64(t, 0, thread->db_id);
+	tuple_set_u64(t, 1, machine->db_id);
+	tuple_set_u64(t, 2, main_thread_db_id);
+	tuple_set_s32(t, 3, thread->pid_);
+	tuple_set_s32(t, 4, thread->tid);
+
+	call_object(tables->thread_handler, t, "thread_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_comm(struct db_export *dbe, struct comm *comm)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	PyObject *t;
+
+	t = tuple_new(2);
+
+	tuple_set_u64(t, 0, comm->db_id);
+	tuple_set_string(t, 1, comm__str(comm));
+
+	call_object(tables->comm_handler, t, "comm_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_comm_thread(struct db_export *dbe, u64 db_id,
+				     struct comm *comm, struct thread *thread)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	PyObject *t;
+
+	t = tuple_new(3);
+
+	tuple_set_u64(t, 0, db_id);
+	tuple_set_u64(t, 1, comm->db_id);
+	tuple_set_u64(t, 2, thread->db_id);
+
+	call_object(tables->comm_thread_handler, t, "comm_thread_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_dso(struct db_export *dbe, struct dso *dso,
+			     struct machine *machine)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+	PyObject *t;
+
+	build_id__sprintf(dso->build_id, sizeof(dso->build_id), sbuild_id);
+
+	t = tuple_new(5);
+
+	tuple_set_u64(t, 0, dso->db_id);
+	tuple_set_u64(t, 1, machine->db_id);
+	tuple_set_string(t, 2, dso->short_name);
+	tuple_set_string(t, 3, dso->long_name);
+	tuple_set_string(t, 4, sbuild_id);
+
+	call_object(tables->dso_handler, t, "dso_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_symbol(struct db_export *dbe, struct symbol *sym,
+				struct dso *dso)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	u64 *sym_db_id = symbol__priv(sym);
+	PyObject *t;
+
+	t = tuple_new(6);
+
+	tuple_set_u64(t, 0, *sym_db_id);
+	tuple_set_u64(t, 1, dso->db_id);
+	tuple_set_u64(t, 2, sym->start);
+	tuple_set_u64(t, 3, sym->end);
+	tuple_set_s32(t, 4, sym->binding);
+	tuple_set_string(t, 5, sym->name);
+
+	call_object(tables->symbol_handler, t, "symbol_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_branch_type(struct db_export *dbe, u32 branch_type,
+				     const char *name)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	PyObject *t;
+
+	t = tuple_new(2);
+
+	tuple_set_s32(t, 0, branch_type);
+	tuple_set_string(t, 1, name);
+
+	call_object(tables->branch_type_handler, t, "branch_type_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_sample(struct db_export *dbe,
+				struct export_sample *es)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	PyObject *t;
+
+	t = tuple_new(21);
+
+	tuple_set_u64(t, 0, es->db_id);
+	tuple_set_u64(t, 1, es->evsel->db_id);
+	tuple_set_u64(t, 2, es->al->machine->db_id);
+	tuple_set_u64(t, 3, es->al->thread->db_id);
+	tuple_set_u64(t, 4, es->comm_db_id);
+	tuple_set_u64(t, 5, es->dso_db_id);
+	tuple_set_u64(t, 6, es->sym_db_id);
+	tuple_set_u64(t, 7, es->offset);
+	tuple_set_u64(t, 8, es->sample->ip);
+	tuple_set_u64(t, 9, es->sample->time);
+	tuple_set_s32(t, 10, es->sample->cpu);
+	tuple_set_u64(t, 11, es->addr_dso_db_id);
+	tuple_set_u64(t, 12, es->addr_sym_db_id);
+	tuple_set_u64(t, 13, es->addr_offset);
+	tuple_set_u64(t, 14, es->sample->addr);
+	tuple_set_u64(t, 15, es->sample->period);
+	tuple_set_u64(t, 16, es->sample->weight);
+	tuple_set_u64(t, 17, es->sample->transaction);
+	tuple_set_u64(t, 18, es->sample->data_src);
+	tuple_set_s32(t, 19, es->sample->flags & PERF_BRANCH_MASK);
+	tuple_set_s32(t, 20, !!(es->sample->flags & PERF_IP_FLAG_IN_TX));
+
+	call_object(tables->sample_handler, t, "sample_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_call_path(struct db_export *dbe, struct call_path *cp)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	PyObject *t;
+	u64 parent_db_id, sym_db_id;
+
+	parent_db_id = cp->parent ? cp->parent->db_id : 0;
+	sym_db_id = cp->sym ? *(u64 *)symbol__priv(cp->sym) : 0;
+
+	t = tuple_new(4);
+
+	tuple_set_u64(t, 0, cp->db_id);
+	tuple_set_u64(t, 1, parent_db_id);
+	tuple_set_u64(t, 2, sym_db_id);
+	tuple_set_u64(t, 3, cp->ip);
+
+	call_object(tables->call_path_handler, t, "call_path_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_export_call_return(struct db_export *dbe,
+				     struct call_return *cr)
+{
+	struct tables *tables = container_of(dbe, struct tables, dbe);
+	u64 comm_db_id = cr->comm ? cr->comm->db_id : 0;
+	PyObject *t;
+
+	t = tuple_new(11);
+
+	tuple_set_u64(t, 0, cr->db_id);
+	tuple_set_u64(t, 1, cr->thread->db_id);
+	tuple_set_u64(t, 2, comm_db_id);
+	tuple_set_u64(t, 3, cr->cp->db_id);
+	tuple_set_u64(t, 4, cr->call_time);
+	tuple_set_u64(t, 5, cr->return_time);
+	tuple_set_u64(t, 6, cr->branch_count);
+	tuple_set_u64(t, 7, cr->call_ref);
+	tuple_set_u64(t, 8, cr->return_ref);
+	tuple_set_u64(t, 9, cr->cp->parent->db_id);
+	tuple_set_s32(t, 10, cr->flags);
+
+	call_object(tables->call_return_handler, t, "call_return_table");
+
+	Py_DECREF(t);
+
+	return 0;
+}
+
+static int python_process_call_return(struct call_return *cr, void *data)
+{
+	struct db_export *dbe = data;
+
+	return db_export__call_return(dbe, cr);
+}
+
+static void python_process_general_event(struct perf_sample *sample,
+					 struct perf_evsel *evsel,
+					 struct addr_location *al)
+{
+	PyObject *handler, *t, *dict, *callchain, *dict_sample;
+	static char handler_name[64];
+	unsigned n = 0;
+
+	/*
+	 * Use the MAX_FIELDS to make the function expandable, though
+	 * currently there is only one item for the tuple.
+	 */
+	t = PyTuple_New(MAX_FIELDS);
+	if (!t)
+		Py_FatalError("couldn't create Python tuple");
+
+	dict = PyDict_New();
+	if (!dict)
+		Py_FatalError("couldn't create Python dictionary");
+
+	dict_sample = PyDict_New();
+	if (!dict_sample)
+		Py_FatalError("couldn't create Python dictionary");
+
+	snprintf(handler_name, sizeof(handler_name), "%s", "process_event");
+
+	handler = get_handler(handler_name);
+	if (!handler)
+		goto exit;
+
+	pydict_set_item_string_decref(dict, "ev_name", PyString_FromString(perf_evsel__name(evsel)));
+	pydict_set_item_string_decref(dict, "attr", PyString_FromStringAndSize(
+			(const char *)&evsel->attr, sizeof(evsel->attr)));
+
+	pydict_set_item_string_decref(dict_sample, "pid",
+			PyInt_FromLong(sample->pid));
+	pydict_set_item_string_decref(dict_sample, "tid",
+			PyInt_FromLong(sample->tid));
+	pydict_set_item_string_decref(dict_sample, "cpu",
+			PyInt_FromLong(sample->cpu));
+	pydict_set_item_string_decref(dict_sample, "ip",
+			PyLong_FromUnsignedLongLong(sample->ip));
+	pydict_set_item_string_decref(dict_sample, "time",
+			PyLong_FromUnsignedLongLong(sample->time));
+	pydict_set_item_string_decref(dict_sample, "period",
+			PyLong_FromUnsignedLongLong(sample->period));
+	pydict_set_item_string_decref(dict, "sample", dict_sample);
+
+	pydict_set_item_string_decref(dict, "raw_buf", PyString_FromStringAndSize(
+			(const char *)sample->raw_data, sample->raw_size));
+	pydict_set_item_string_decref(dict, "comm",
+			PyString_FromString(thread__comm_str(al->thread)));
+	if (al->map) {
+		pydict_set_item_string_decref(dict, "dso",
+			PyString_FromString(al->map->dso->name));
+	}
+	if (al->sym) {
+		pydict_set_item_string_decref(dict, "symbol",
+			PyString_FromString(al->sym->name));
+	}
+
+	/* ip unwinding */
+	callchain = python_process_callchain(sample, evsel, al);
+	pydict_set_item_string_decref(dict, "callchain", callchain);
+
+	PyTuple_SetItem(t, n++, dict);
+	if (_PyTuple_Resize(&t, n) == -1)
+		Py_FatalError("error resizing Python tuple");
+
+	call_object(handler, t, handler_name);
+exit:
+	Py_DECREF(dict);
+	Py_DECREF(t);
+}
+
+static void python_process_event(union perf_event *event,
+				 struct perf_sample *sample,
+				 struct perf_evsel *evsel,
+				 struct addr_location *al)
+{
+	struct tables *tables = &tables_global;
+
+	switch (evsel->attr.type) {
+	case PERF_TYPE_TRACEPOINT:
+		python_process_tracepoint(sample, evsel, al);
+		break;
+	/* Reserve for future process_hw/sw/raw APIs */
+	default:
+		if (tables->db_export_mode)
+			db_export__sample(&tables->dbe, event, sample, evsel, al);
+		else
+			python_process_general_event(sample, evsel, al);
+	}
+}
+
+static int run_start_sub(void)
+{
+	main_module = PyImport_AddModule("__main__");
+	if (main_module == NULL)
+		return -1;
+	Py_INCREF(main_module);
+
+	main_dict = PyModule_GetDict(main_module);
+	if (main_dict == NULL)
+		goto error;
+	Py_INCREF(main_dict);
+
+	try_call_object("trace_begin", NULL);
+
+	return 0;
+
+error:
+	Py_XDECREF(main_dict);
+	Py_XDECREF(main_module);
+	return -1;
+}
+
+#define SET_TABLE_HANDLER_(name, handler_name, table_name) do {		\
+	tables->handler_name = get_handler(#table_name);		\
+	if (tables->handler_name)					\
+		tables->dbe.export_ ## name = python_export_ ## name;	\
+} while (0)
+
+#define SET_TABLE_HANDLER(name) \
+	SET_TABLE_HANDLER_(name, name ## _handler, name ## _table)
+
+static void set_table_handlers(struct tables *tables)
+{
+	const char *perf_db_export_mode = "perf_db_export_mode";
+	const char *perf_db_export_calls = "perf_db_export_calls";
+	PyObject *db_export_mode, *db_export_calls;
+	bool export_calls = false;
+	int ret;
+
+	memset(tables, 0, sizeof(struct tables));
+	if (db_export__init(&tables->dbe))
+		Py_FatalError("failed to initialize export");
+
+	db_export_mode = PyDict_GetItemString(main_dict, perf_db_export_mode);
+	if (!db_export_mode)
+		return;
+
+	ret = PyObject_IsTrue(db_export_mode);
+	if (ret == -1)
+		handler_call_die(perf_db_export_mode);
+	if (!ret)
+		return;
+
+	tables->dbe.crp = NULL;
+	db_export_calls = PyDict_GetItemString(main_dict, perf_db_export_calls);
+	if (db_export_calls) {
+		ret = PyObject_IsTrue(db_export_calls);
+		if (ret == -1)
+			handler_call_die(perf_db_export_calls);
+		export_calls = !!ret;
+	}
+
+	if (export_calls) {
+		tables->dbe.crp =
+			call_return_processor__new(python_process_call_return,
+						   &tables->dbe);
+		if (!tables->dbe.crp)
+			Py_FatalError("failed to create calls processor");
+	}
+
+	tables->db_export_mode = true;
+	/*
+	 * Reserve per symbol space for symbol->db_id via symbol__priv()
+	 */
+	symbol_conf.priv_size = sizeof(u64);
+
+	SET_TABLE_HANDLER(evsel);
+	SET_TABLE_HANDLER(machine);
+	SET_TABLE_HANDLER(thread);
+	SET_TABLE_HANDLER(comm);
+	SET_TABLE_HANDLER(comm_thread);
+	SET_TABLE_HANDLER(dso);
+	SET_TABLE_HANDLER(symbol);
+	SET_TABLE_HANDLER(branch_type);
+	SET_TABLE_HANDLER(sample);
+	SET_TABLE_HANDLER(call_path);
+	SET_TABLE_HANDLER(call_return);
+}
+
+/*
+ * Start trace script
+ */
+static int python_start_script(const char *script, int argc, const char **argv)
+{
+	struct tables *tables = &tables_global;
+	const char **command_line;
+	char buf[PATH_MAX];
+	int i, err = 0;
+	FILE *fp;
+
+	command_line = malloc((argc + 1) * sizeof(const char *));
+	command_line[0] = script;
+	for (i = 1; i < argc + 1; i++)
+		command_line[i] = argv[i - 1];
+
+	Py_Initialize();
+
+	initperf_trace_context();
+
+	PySys_SetArgv(argc + 1, (char **)command_line);
+
+	fp = fopen(script, "r");
+	if (!fp) {
+		sprintf(buf, "Can't open python script \"%s\"", script);
+		perror(buf);
+		err = -1;
+		goto error;
+	}
+
+	err = PyRun_SimpleFile(fp, script);
+	if (err) {
+		fprintf(stderr, "Error running python script %s\n", script);
+		goto error;
+	}
+
+	err = run_start_sub();
+	if (err) {
+		fprintf(stderr, "Error starting python script %s\n", script);
+		goto error;
+	}
+
+	free(command_line);
+
+	set_table_handlers(tables);
+
+	if (tables->db_export_mode) {
+		err = db_export__branch_types(&tables->dbe);
+		if (err)
+			goto error;
+	}
+
+	return err;
+error:
+	Py_Finalize();
+	free(command_line);
+
+	return err;
+}
+
+static int python_flush_script(void)
+{
+	struct tables *tables = &tables_global;
+
+	return db_export__flush(&tables->dbe);
+}
+
+/*
+ * Stop trace script
+ */
+static int python_stop_script(void)
+{
+	struct tables *tables = &tables_global;
+
+	try_call_object("trace_end", NULL);
+
+	db_export__exit(&tables->dbe);
+
+	Py_XDECREF(main_dict);
+	Py_XDECREF(main_module);
+	Py_Finalize();
+
+	return 0;
+}
+
+static int python_generate_script(struct pevent *pevent, const char *outfile)
+{
+	struct event_format *event = NULL;
+	struct format_field *f;
+	char fname[PATH_MAX];
+	int not_first, count;
+	FILE *ofp;
+
+	sprintf(fname, "%s.py", outfile);
+	ofp = fopen(fname, "w");
+	if (ofp == NULL) {
+		fprintf(stderr, "couldn't open %s\n", fname);
+		return -1;
+	}
+	fprintf(ofp, "# perf script event handlers, "
+		"generated by perf script -g python\n");
+
+	fprintf(ofp, "# Licensed under the terms of the GNU GPL"
+		" License version 2\n\n");
+
+	fprintf(ofp, "# The common_* event handler fields are the most useful "
+		"fields common to\n");
+
+	fprintf(ofp, "# all events.  They don't necessarily correspond to "
+		"the 'common_*' fields\n");
+
+	fprintf(ofp, "# in the format files.  Those fields not available as "
+		"handler params can\n");
+
+	fprintf(ofp, "# be retrieved using Python functions of the form "
+		"common_*(context).\n");
+
+	fprintf(ofp, "# See the perf-trace-python Documentation for the list "
+		"of available functions.\n\n");
+
+	fprintf(ofp, "import os\n");
+	fprintf(ofp, "import sys\n\n");
+
+	fprintf(ofp, "sys.path.append(os.environ['PERF_EXEC_PATH'] + \\\n");
+	fprintf(ofp, "\t'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')\n");
+	fprintf(ofp, "\nfrom perf_trace_context import *\n");
+	fprintf(ofp, "from Core import *\n\n\n");
+
+	fprintf(ofp, "def trace_begin():\n");
+	fprintf(ofp, "\tprint \"in trace_begin\"\n\n");
+
+	fprintf(ofp, "def trace_end():\n");
+	fprintf(ofp, "\tprint \"in trace_end\"\n\n");
+
+	while ((event = trace_find_next_event(pevent, event))) {
+		fprintf(ofp, "def %s__%s(", event->system, event->name);
+		fprintf(ofp, "event_name, ");
+		fprintf(ofp, "context, ");
+		fprintf(ofp, "common_cpu,\n");
+		fprintf(ofp, "\tcommon_secs, ");
+		fprintf(ofp, "common_nsecs, ");
+		fprintf(ofp, "common_pid, ");
+		fprintf(ofp, "common_comm,\n\t");
+		fprintf(ofp, "common_callchain, ");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t");
+
+			fprintf(ofp, "%s", f->name);
+		}
+		fprintf(ofp, "):\n");
+
+		fprintf(ofp, "\t\tprint_header(event_name, common_cpu, "
+			"common_secs, common_nsecs,\n\t\t\t"
+			"common_pid, common_comm)\n\n");
+
+		fprintf(ofp, "\t\tprint \"");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+			if (count && count % 3 == 0) {
+				fprintf(ofp, "\" \\\n\t\t\"");
+			}
+			count++;
+
+			fprintf(ofp, "%s=", f->name);
+			if (f->flags & FIELD_IS_STRING ||
+			    f->flags & FIELD_IS_FLAG ||
+			    f->flags & FIELD_IS_ARRAY ||
+			    f->flags & FIELD_IS_SYMBOLIC)
+				fprintf(ofp, "%%s");
+			else if (f->flags & FIELD_IS_SIGNED)
+				fprintf(ofp, "%%d");
+			else
+				fprintf(ofp, "%%u");
+		}
+
+		fprintf(ofp, "\" %% \\\n\t\t(");
+
+		not_first = 0;
+		count = 0;
+
+		for (f = event->format.fields; f; f = f->next) {
+			if (not_first++)
+				fprintf(ofp, ", ");
+
+			if (++count % 5 == 0)
+				fprintf(ofp, "\n\t\t");
+
+			if (f->flags & FIELD_IS_FLAG) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t\t");
+					count = 4;
+				}
+				fprintf(ofp, "flag_str(\"");
+				fprintf(ofp, "%s__%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", %s)", f->name,
+					f->name);
+			} else if (f->flags & FIELD_IS_SYMBOLIC) {
+				if ((count - 1) % 5 != 0) {
+					fprintf(ofp, "\n\t\t");
+					count = 4;
+				}
+				fprintf(ofp, "symbol_str(\"");
+				fprintf(ofp, "%s__%s\", ", event->system,
+					event->name);
+				fprintf(ofp, "\"%s\", %s)", f->name,
+					f->name);
+			} else
+				fprintf(ofp, "%s", f->name);
+		}
+
+		fprintf(ofp, ")\n\n");
+
+		fprintf(ofp, "\t\tfor node in common_callchain:");
+		fprintf(ofp, "\n\t\t\tif 'sym' in node:");
+		fprintf(ofp, "\n\t\t\t\tprint \"\\t[%%x] %%s\" %% (node['ip'], node['sym']['name'])");
+		fprintf(ofp, "\n\t\t\telse:");
+		fprintf(ofp, "\n\t\t\t\tprint \"\t[%%x]\" %% (node['ip'])\n\n");
+		fprintf(ofp, "\t\tprint \"\\n\"\n\n");
+
+	}
+
+	fprintf(ofp, "def trace_unhandled(event_name, context, "
+		"event_fields_dict):\n");
+
+	fprintf(ofp, "\t\tprint ' '.join(['%%s=%%s'%%(k,str(v))"
+		"for k,v in sorted(event_fields_dict.items())])\n\n");
+
+	fprintf(ofp, "def print_header("
+		"event_name, cpu, secs, nsecs, pid, comm):\n"
+		"\tprint \"%%-20s %%5u %%05u.%%09u %%8u %%-20s \" %% \\\n\t"
+		"(event_name, cpu, secs, nsecs, pid, comm),\n");
+
+	fclose(ofp);
+
+	fprintf(stderr, "generated Python script: %s\n", fname);
+
+	return 0;
+}
+
+struct scripting_ops python_scripting_ops = {
+	.name = "Python",
+	.start_script = python_start_script,
+	.flush_script = python_flush_script,
+	.stop_script = python_stop_script,
+	.process_event = python_process_event,
+	.generate_script = python_generate_script,
+};
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
new file mode 100644
index 000000000..0c7401257
--- /dev/null
+++ b/tools/perf/util/session.c
@@ -0,0 +1,1818 @@
+#include <linux/kernel.h>
+#include <traceevent/event-parse.h>
+
+#include <byteswap.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/mman.h>
+
+#include "evlist.h"
+#include "evsel.h"
+#include "session.h"
+#include "tool.h"
+#include "sort.h"
+#include "util.h"
+#include "cpumap.h"
+#include "perf_regs.h"
+#include "asm/bug.h"
+
+static int machines__deliver_event(struct machines *machines,
+				   struct perf_evlist *evlist,
+				   union perf_event *event,
+				   struct perf_sample *sample,
+				   struct perf_tool *tool, u64 file_offset);
+
+static int perf_session__open(struct perf_session *session)
+{
+	struct perf_data_file *file = session->file;
+
+	if (perf_session__read_header(session) < 0) {
+		pr_err("incompatible file format (rerun with -v to learn more)");
+		return -1;
+	}
+
+	if (perf_data_file__is_pipe(file))
+		return 0;
+
+	if (!perf_evlist__valid_sample_type(session->evlist)) {
+		pr_err("non matching sample_type");
+		return -1;
+	}
+
+	if (!perf_evlist__valid_sample_id_all(session->evlist)) {
+		pr_err("non matching sample_id_all");
+		return -1;
+	}
+
+	if (!perf_evlist__valid_read_format(session->evlist)) {
+		pr_err("non matching read_format");
+		return -1;
+	}
+
+	return 0;
+}
+
+void perf_session__set_id_hdr_size(struct perf_session *session)
+{
+	u16 id_hdr_size = perf_evlist__id_hdr_size(session->evlist);
+
+	machines__set_id_hdr_size(&session->machines, id_hdr_size);
+}
+
+int perf_session__create_kernel_maps(struct perf_session *session)
+{
+	int ret = machine__create_kernel_maps(&session->machines.host);
+
+	if (ret >= 0)
+		ret = machines__create_guest_kernel_maps(&session->machines);
+	return ret;
+}
+
+static void perf_session__destroy_kernel_maps(struct perf_session *session)
+{
+	machines__destroy_kernel_maps(&session->machines);
+}
+
+static bool perf_session__has_comm_exec(struct perf_session *session)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(session->evlist, evsel) {
+		if (evsel->attr.comm_exec)
+			return true;
+	}
+
+	return false;
+}
+
+static void perf_session__set_comm_exec(struct perf_session *session)
+{
+	bool comm_exec = perf_session__has_comm_exec(session);
+
+	machines__set_comm_exec(&session->machines, comm_exec);
+}
+
+static int ordered_events__deliver_event(struct ordered_events *oe,
+					 struct ordered_event *event)
+{
+	struct perf_sample sample;
+	struct perf_session *session = container_of(oe, struct perf_session,
+						    ordered_events);
+	int ret = perf_evlist__parse_sample(session->evlist, event->event, &sample);
+
+	if (ret) {
+		pr_err("Can't parse sample, err = %d\n", ret);
+		return ret;
+	}
+
+	return machines__deliver_event(&session->machines, session->evlist, event->event,
+				       &sample, session->tool, event->file_offset);
+}
+
+struct perf_session *perf_session__new(struct perf_data_file *file,
+				       bool repipe, struct perf_tool *tool)
+{
+	struct perf_session *session = zalloc(sizeof(*session));
+
+	if (!session)
+		goto out;
+
+	session->repipe = repipe;
+	session->tool   = tool;
+	machines__init(&session->machines);
+	ordered_events__init(&session->ordered_events, ordered_events__deliver_event);
+
+	if (file) {
+		if (perf_data_file__open(file))
+			goto out_delete;
+
+		session->file = file;
+
+		if (perf_data_file__is_read(file)) {
+			if (perf_session__open(session) < 0)
+				goto out_close;
+
+			perf_session__set_id_hdr_size(session);
+			perf_session__set_comm_exec(session);
+		}
+	}
+
+	if (!file || perf_data_file__is_write(file)) {
+		/*
+		 * In O_RDONLY mode this will be performed when reading the
+		 * kernel MMAP event, in perf_event__process_mmap().
+		 */
+		if (perf_session__create_kernel_maps(session) < 0)
+			pr_warning("Cannot read kernel map\n");
+	}
+
+	if (tool && tool->ordering_requires_timestamps &&
+	    tool->ordered_events && !perf_evlist__sample_id_all(session->evlist)) {
+		dump_printf("WARNING: No sample_id_all support, falling back to unordered processing\n");
+		tool->ordered_events = false;
+	}
+
+	return session;
+
+ out_close:
+	perf_data_file__close(file);
+ out_delete:
+	perf_session__delete(session);
+ out:
+	return NULL;
+}
+
+static void perf_session__delete_threads(struct perf_session *session)
+{
+	machine__delete_threads(&session->machines.host);
+}
+
+static void perf_session_env__delete(struct perf_session_env *env)
+{
+	zfree(&env->hostname);
+	zfree(&env->os_release);
+	zfree(&env->version);
+	zfree(&env->arch);
+	zfree(&env->cpu_desc);
+	zfree(&env->cpuid);
+
+	zfree(&env->cmdline);
+	zfree(&env->sibling_cores);
+	zfree(&env->sibling_threads);
+	zfree(&env->numa_nodes);
+	zfree(&env->pmu_mappings);
+}
+
+void perf_session__delete(struct perf_session *session)
+{
+	perf_session__destroy_kernel_maps(session);
+	perf_session__delete_threads(session);
+	perf_session_env__delete(&session->header.env);
+	machines__exit(&session->machines);
+	if (session->file)
+		perf_data_file__close(session->file);
+	free(session);
+}
+
+static int process_event_synth_tracing_data_stub(struct perf_tool *tool
+						 __maybe_unused,
+						 union perf_event *event
+						 __maybe_unused,
+						 struct perf_session *session
+						__maybe_unused)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_event_synth_attr_stub(struct perf_tool *tool __maybe_unused,
+					 union perf_event *event __maybe_unused,
+					 struct perf_evlist **pevlist
+					 __maybe_unused)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_event_sample_stub(struct perf_tool *tool __maybe_unused,
+				     union perf_event *event __maybe_unused,
+				     struct perf_sample *sample __maybe_unused,
+				     struct perf_evsel *evsel __maybe_unused,
+				     struct machine *machine __maybe_unused)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_event_stub(struct perf_tool *tool __maybe_unused,
+			      union perf_event *event __maybe_unused,
+			      struct perf_sample *sample __maybe_unused,
+			      struct machine *machine __maybe_unused)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_build_id_stub(struct perf_tool *tool __maybe_unused,
+				 union perf_event *event __maybe_unused,
+				 struct perf_session *session __maybe_unused)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_finished_round_stub(struct perf_tool *tool __maybe_unused,
+				       union perf_event *event __maybe_unused,
+				       struct ordered_events *oe __maybe_unused)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+static int process_finished_round(struct perf_tool *tool,
+				  union perf_event *event,
+				  struct ordered_events *oe);
+
+static int process_id_index_stub(struct perf_tool *tool __maybe_unused,
+				 union perf_event *event __maybe_unused,
+				 struct perf_session *perf_session
+				 __maybe_unused)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
+void perf_tool__fill_defaults(struct perf_tool *tool)
+{
+	if (tool->sample == NULL)
+		tool->sample = process_event_sample_stub;
+	if (tool->mmap == NULL)
+		tool->mmap = process_event_stub;
+	if (tool->mmap2 == NULL)
+		tool->mmap2 = process_event_stub;
+	if (tool->comm == NULL)
+		tool->comm = process_event_stub;
+	if (tool->fork == NULL)
+		tool->fork = process_event_stub;
+	if (tool->exit == NULL)
+		tool->exit = process_event_stub;
+	if (tool->lost == NULL)
+		tool->lost = perf_event__process_lost;
+	if (tool->read == NULL)
+		tool->read = process_event_sample_stub;
+	if (tool->throttle == NULL)
+		tool->throttle = process_event_stub;
+	if (tool->unthrottle == NULL)
+		tool->unthrottle = process_event_stub;
+	if (tool->attr == NULL)
+		tool->attr = process_event_synth_attr_stub;
+	if (tool->tracing_data == NULL)
+		tool->tracing_data = process_event_synth_tracing_data_stub;
+	if (tool->build_id == NULL)
+		tool->build_id = process_build_id_stub;
+	if (tool->finished_round == NULL) {
+		if (tool->ordered_events)
+			tool->finished_round = process_finished_round;
+		else
+			tool->finished_round = process_finished_round_stub;
+	}
+	if (tool->id_index == NULL)
+		tool->id_index = process_id_index_stub;
+}
+
+static void swap_sample_id_all(union perf_event *event, void *data)
+{
+	void *end = (void *) event + event->header.size;
+	int size = end - data;
+
+	BUG_ON(size % sizeof(u64));
+	mem_bswap_64(data, size);
+}
+
+static void perf_event__all64_swap(union perf_event *event,
+				   bool sample_id_all __maybe_unused)
+{
+	struct perf_event_header *hdr = &event->header;
+	mem_bswap_64(hdr + 1, event->header.size - sizeof(*hdr));
+}
+
+static void perf_event__comm_swap(union perf_event *event, bool sample_id_all)
+{
+	event->comm.pid = bswap_32(event->comm.pid);
+	event->comm.tid = bswap_32(event->comm.tid);
+
+	if (sample_id_all) {
+		void *data = &event->comm.comm;
+
+		data += PERF_ALIGN(strlen(data) + 1, sizeof(u64));
+		swap_sample_id_all(event, data);
+	}
+}
+
+static void perf_event__mmap_swap(union perf_event *event,
+				  bool sample_id_all)
+{
+	event->mmap.pid	  = bswap_32(event->mmap.pid);
+	event->mmap.tid	  = bswap_32(event->mmap.tid);
+	event->mmap.start = bswap_64(event->mmap.start);
+	event->mmap.len	  = bswap_64(event->mmap.len);
+	event->mmap.pgoff = bswap_64(event->mmap.pgoff);
+
+	if (sample_id_all) {
+		void *data = &event->mmap.filename;
+
+		data += PERF_ALIGN(strlen(data) + 1, sizeof(u64));
+		swap_sample_id_all(event, data);
+	}
+}
+
+static void perf_event__mmap2_swap(union perf_event *event,
+				  bool sample_id_all)
+{
+	event->mmap2.pid   = bswap_32(event->mmap2.pid);
+	event->mmap2.tid   = bswap_32(event->mmap2.tid);
+	event->mmap2.start = bswap_64(event->mmap2.start);
+	event->mmap2.len   = bswap_64(event->mmap2.len);
+	event->mmap2.pgoff = bswap_64(event->mmap2.pgoff);
+	event->mmap2.maj   = bswap_32(event->mmap2.maj);
+	event->mmap2.min   = bswap_32(event->mmap2.min);
+	event->mmap2.ino   = bswap_64(event->mmap2.ino);
+
+	if (sample_id_all) {
+		void *data = &event->mmap2.filename;
+
+		data += PERF_ALIGN(strlen(data) + 1, sizeof(u64));
+		swap_sample_id_all(event, data);
+	}
+}
+static void perf_event__task_swap(union perf_event *event, bool sample_id_all)
+{
+	event->fork.pid	 = bswap_32(event->fork.pid);
+	event->fork.tid	 = bswap_32(event->fork.tid);
+	event->fork.ppid = bswap_32(event->fork.ppid);
+	event->fork.ptid = bswap_32(event->fork.ptid);
+	event->fork.time = bswap_64(event->fork.time);
+
+	if (sample_id_all)
+		swap_sample_id_all(event, &event->fork + 1);
+}
+
+static void perf_event__read_swap(union perf_event *event, bool sample_id_all)
+{
+	event->read.pid		 = bswap_32(event->read.pid);
+	event->read.tid		 = bswap_32(event->read.tid);
+	event->read.value	 = bswap_64(event->read.value);
+	event->read.time_enabled = bswap_64(event->read.time_enabled);
+	event->read.time_running = bswap_64(event->read.time_running);
+	event->read.id		 = bswap_64(event->read.id);
+
+	if (sample_id_all)
+		swap_sample_id_all(event, &event->read + 1);
+}
+
+static void perf_event__throttle_swap(union perf_event *event,
+				      bool sample_id_all)
+{
+	event->throttle.time	  = bswap_64(event->throttle.time);
+	event->throttle.id	  = bswap_64(event->throttle.id);
+	event->throttle.stream_id = bswap_64(event->throttle.stream_id);
+
+	if (sample_id_all)
+		swap_sample_id_all(event, &event->throttle + 1);
+}
+
+static u8 revbyte(u8 b)
+{
+	int rev = (b >> 4) | ((b & 0xf) << 4);
+	rev = ((rev & 0xcc) >> 2) | ((rev & 0x33) << 2);
+	rev = ((rev & 0xaa) >> 1) | ((rev & 0x55) << 1);
+	return (u8) rev;
+}
+
+/*
+ * XXX this is hack in attempt to carry flags bitfield
+ * throught endian village. ABI says:
+ *
+ * Bit-fields are allocated from right to left (least to most significant)
+ * on little-endian implementations and from left to right (most to least
+ * significant) on big-endian implementations.
+ *
+ * The above seems to be byte specific, so we need to reverse each
+ * byte of the bitfield. 'Internet' also says this might be implementation
+ * specific and we probably need proper fix and carry perf_event_attr
+ * bitfield flags in separate data file FEAT_ section. Thought this seems
+ * to work for now.
+ */
+static void swap_bitfield(u8 *p, unsigned len)
+{
+	unsigned i;
+
+	for (i = 0; i < len; i++) {
+		*p = revbyte(*p);
+		p++;
+	}
+}
+
+/* exported for swapping attributes in file header */
+void perf_event__attr_swap(struct perf_event_attr *attr)
+{
+	attr->type		= bswap_32(attr->type);
+	attr->size		= bswap_32(attr->size);
+	attr->config		= bswap_64(attr->config);
+	attr->sample_period	= bswap_64(attr->sample_period);
+	attr->sample_type	= bswap_64(attr->sample_type);
+	attr->read_format	= bswap_64(attr->read_format);
+	attr->wakeup_events	= bswap_32(attr->wakeup_events);
+	attr->bp_type		= bswap_32(attr->bp_type);
+	attr->bp_addr		= bswap_64(attr->bp_addr);
+	attr->bp_len		= bswap_64(attr->bp_len);
+	attr->branch_sample_type = bswap_64(attr->branch_sample_type);
+	attr->sample_regs_user	 = bswap_64(attr->sample_regs_user);
+	attr->sample_stack_user  = bswap_32(attr->sample_stack_user);
+
+	swap_bitfield((u8 *) (&attr->read_format + 1), sizeof(u64));
+}
+
+static void perf_event__hdr_attr_swap(union perf_event *event,
+				      bool sample_id_all __maybe_unused)
+{
+	size_t size;
+
+	perf_event__attr_swap(&event->attr.attr);
+
+	size = event->header.size;
+	size -= (void *)&event->attr.id - (void *)event;
+	mem_bswap_64(event->attr.id, size);
+}
+
+static void perf_event__event_type_swap(union perf_event *event,
+					bool sample_id_all __maybe_unused)
+{
+	event->event_type.event_type.event_id =
+		bswap_64(event->event_type.event_type.event_id);
+}
+
+static void perf_event__tracing_data_swap(union perf_event *event,
+					  bool sample_id_all __maybe_unused)
+{
+	event->tracing_data.size = bswap_32(event->tracing_data.size);
+}
+
+typedef void (*perf_event__swap_op)(union perf_event *event,
+				    bool sample_id_all);
+
+static perf_event__swap_op perf_event__swap_ops[] = {
+	[PERF_RECORD_MMAP]		  = perf_event__mmap_swap,
+	[PERF_RECORD_MMAP2]		  = perf_event__mmap2_swap,
+	[PERF_RECORD_COMM]		  = perf_event__comm_swap,
+	[PERF_RECORD_FORK]		  = perf_event__task_swap,
+	[PERF_RECORD_EXIT]		  = perf_event__task_swap,
+	[PERF_RECORD_LOST]		  = perf_event__all64_swap,
+	[PERF_RECORD_READ]		  = perf_event__read_swap,
+	[PERF_RECORD_THROTTLE]		  = perf_event__throttle_swap,
+	[PERF_RECORD_UNTHROTTLE]	  = perf_event__throttle_swap,
+	[PERF_RECORD_SAMPLE]		  = perf_event__all64_swap,
+	[PERF_RECORD_HEADER_ATTR]	  = perf_event__hdr_attr_swap,
+	[PERF_RECORD_HEADER_EVENT_TYPE]	  = perf_event__event_type_swap,
+	[PERF_RECORD_HEADER_TRACING_DATA] = perf_event__tracing_data_swap,
+	[PERF_RECORD_HEADER_BUILD_ID]	  = NULL,
+	[PERF_RECORD_ID_INDEX]		  = perf_event__all64_swap,
+	[PERF_RECORD_HEADER_MAX]	  = NULL,
+};
+
+/*
+ * When perf record finishes a pass on every buffers, it records this pseudo
+ * event.
+ * We record the max timestamp t found in the pass n.
+ * Assuming these timestamps are monotonic across cpus, we know that if
+ * a buffer still has events with timestamps below t, they will be all
+ * available and then read in the pass n + 1.
+ * Hence when we start to read the pass n + 2, we can safely flush every
+ * events with timestamps below t.
+ *
+ *    ============ PASS n =================
+ *       CPU 0         |   CPU 1
+ *                     |
+ *    cnt1 timestamps  |   cnt2 timestamps
+ *          1          |         2
+ *          2          |         3
+ *          -          |         4  <--- max recorded
+ *
+ *    ============ PASS n + 1 ==============
+ *       CPU 0         |   CPU 1
+ *                     |
+ *    cnt1 timestamps  |   cnt2 timestamps
+ *          3          |         5
+ *          4          |         6
+ *          5          |         7 <---- max recorded
+ *
+ *      Flush every events below timestamp 4
+ *
+ *    ============ PASS n + 2 ==============
+ *       CPU 0         |   CPU 1
+ *                     |
+ *    cnt1 timestamps  |   cnt2 timestamps
+ *          6          |         8
+ *          7          |         9
+ *          -          |         10
+ *
+ *      Flush every events below timestamp 7
+ *      etc...
+ */
+static int process_finished_round(struct perf_tool *tool __maybe_unused,
+				  union perf_event *event __maybe_unused,
+				  struct ordered_events *oe)
+{
+	return ordered_events__flush(oe, OE_FLUSH__ROUND);
+}
+
+int perf_session__queue_event(struct perf_session *s, union perf_event *event,
+			      struct perf_sample *sample, u64 file_offset)
+{
+	return ordered_events__queue(&s->ordered_events, event, sample, file_offset);
+}
+
+static void callchain__lbr_callstack_printf(struct perf_sample *sample)
+{
+	struct ip_callchain *callchain = sample->callchain;
+	struct branch_stack *lbr_stack = sample->branch_stack;
+	u64 kernel_callchain_nr = callchain->nr;
+	unsigned int i;
+
+	for (i = 0; i < kernel_callchain_nr; i++) {
+		if (callchain->ips[i] == PERF_CONTEXT_USER)
+			break;
+	}
+
+	if ((i != kernel_callchain_nr) && lbr_stack->nr) {
+		u64 total_nr;
+		/*
+		 * LBR callstack can only get user call chain,
+		 * i is kernel call chain number,
+		 * 1 is PERF_CONTEXT_USER.
+		 *
+		 * The user call chain is stored in LBR registers.
+		 * LBR are pair registers. The caller is stored
+		 * in "from" register, while the callee is stored
+		 * in "to" register.
+		 * For example, there is a call stack
+		 * "A"->"B"->"C"->"D".
+		 * The LBR registers will recorde like
+		 * "C"->"D", "B"->"C", "A"->"B".
+		 * So only the first "to" register and all "from"
+		 * registers are needed to construct the whole stack.
+		 */
+		total_nr = i + 1 + lbr_stack->nr + 1;
+		kernel_callchain_nr = i + 1;
+
+		printf("... LBR call chain: nr:%" PRIu64 "\n", total_nr);
+
+		for (i = 0; i < kernel_callchain_nr; i++)
+			printf("..... %2d: %016" PRIx64 "\n",
+			       i, callchain->ips[i]);
+
+		printf("..... %2d: %016" PRIx64 "\n",
+		       (int)(kernel_callchain_nr), lbr_stack->entries[0].to);
+		for (i = 0; i < lbr_stack->nr; i++)
+			printf("..... %2d: %016" PRIx64 "\n",
+			       (int)(i + kernel_callchain_nr + 1), lbr_stack->entries[i].from);
+	}
+}
+
+static void callchain__printf(struct perf_evsel *evsel,
+			      struct perf_sample *sample)
+{
+	unsigned int i;
+	struct ip_callchain *callchain = sample->callchain;
+
+	if (has_branch_callstack(evsel))
+		callchain__lbr_callstack_printf(sample);
+
+	printf("... FP chain: nr:%" PRIu64 "\n", callchain->nr);
+
+	for (i = 0; i < callchain->nr; i++)
+		printf("..... %2d: %016" PRIx64 "\n",
+		       i, callchain->ips[i]);
+}
+
+static void branch_stack__printf(struct perf_sample *sample)
+{
+	uint64_t i;
+
+	printf("... branch stack: nr:%" PRIu64 "\n", sample->branch_stack->nr);
+
+	for (i = 0; i < sample->branch_stack->nr; i++)
+		printf("..... %2"PRIu64": %016" PRIx64 " -> %016" PRIx64 "\n",
+			i, sample->branch_stack->entries[i].from,
+			sample->branch_stack->entries[i].to);
+}
+
+static void regs_dump__printf(u64 mask, u64 *regs)
+{
+	unsigned rid, i = 0;
+
+	for_each_set_bit(rid, (unsigned long *) &mask, sizeof(mask) * 8) {
+		u64 val = regs[i++];
+
+		printf(".... %-5s 0x%" PRIx64 "\n",
+		       perf_reg_name(rid), val);
+	}
+}
+
+static const char *regs_abi[] = {
+	[PERF_SAMPLE_REGS_ABI_NONE] = "none",
+	[PERF_SAMPLE_REGS_ABI_32] = "32-bit",
+	[PERF_SAMPLE_REGS_ABI_64] = "64-bit",
+};
+
+static inline const char *regs_dump_abi(struct regs_dump *d)
+{
+	if (d->abi > PERF_SAMPLE_REGS_ABI_64)
+		return "unknown";
+
+	return regs_abi[d->abi];
+}
+
+static void regs__printf(const char *type, struct regs_dump *regs)
+{
+	u64 mask = regs->mask;
+
+	printf("... %s regs: mask 0x%" PRIx64 " ABI %s\n",
+	       type,
+	       mask,
+	       regs_dump_abi(regs));
+
+	regs_dump__printf(mask, regs->regs);
+}
+
+static void regs_user__printf(struct perf_sample *sample)
+{
+	struct regs_dump *user_regs = &sample->user_regs;
+
+	if (user_regs->regs)
+		regs__printf("user", user_regs);
+}
+
+static void regs_intr__printf(struct perf_sample *sample)
+{
+	struct regs_dump *intr_regs = &sample->intr_regs;
+
+	if (intr_regs->regs)
+		regs__printf("intr", intr_regs);
+}
+
+static void stack_user__printf(struct stack_dump *dump)
+{
+	printf("... ustack: size %" PRIu64 ", offset 0x%x\n",
+	       dump->size, dump->offset);
+}
+
+static void perf_evlist__print_tstamp(struct perf_evlist *evlist,
+				       union perf_event *event,
+				       struct perf_sample *sample)
+{
+	u64 sample_type = __perf_evlist__combined_sample_type(evlist);
+
+	if (event->header.type != PERF_RECORD_SAMPLE &&
+	    !perf_evlist__sample_id_all(evlist)) {
+		fputs("-1 -1 ", stdout);
+		return;
+	}
+
+	if ((sample_type & PERF_SAMPLE_CPU))
+		printf("%u ", sample->cpu);
+
+	if (sample_type & PERF_SAMPLE_TIME)
+		printf("%" PRIu64 " ", sample->time);
+}
+
+static void sample_read__printf(struct perf_sample *sample, u64 read_format)
+{
+	printf("... sample_read:\n");
+
+	if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
+		printf("...... time enabled %016" PRIx64 "\n",
+		       sample->read.time_enabled);
+
+	if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
+		printf("...... time running %016" PRIx64 "\n",
+		       sample->read.time_running);
+
+	if (read_format & PERF_FORMAT_GROUP) {
+		u64 i;
+
+		printf(".... group nr %" PRIu64 "\n", sample->read.group.nr);
+
+		for (i = 0; i < sample->read.group.nr; i++) {
+			struct sample_read_value *value;
+
+			value = &sample->read.group.values[i];
+			printf("..... id %016" PRIx64
+			       ", value %016" PRIx64 "\n",
+			       value->id, value->value);
+		}
+	} else
+		printf("..... id %016" PRIx64 ", value %016" PRIx64 "\n",
+			sample->read.one.id, sample->read.one.value);
+}
+
+static void dump_event(struct perf_evlist *evlist, union perf_event *event,
+		       u64 file_offset, struct perf_sample *sample)
+{
+	if (!dump_trace)
+		return;
+
+	printf("\n%#" PRIx64 " [%#x]: event: %d\n",
+	       file_offset, event->header.size, event->header.type);
+
+	trace_event(event);
+
+	if (sample)
+		perf_evlist__print_tstamp(evlist, event, sample);
+
+	printf("%#" PRIx64 " [%#x]: PERF_RECORD_%s", file_offset,
+	       event->header.size, perf_event__name(event->header.type));
+}
+
+static void dump_sample(struct perf_evsel *evsel, union perf_event *event,
+			struct perf_sample *sample)
+{
+	u64 sample_type;
+
+	if (!dump_trace)
+		return;
+
+	printf("(IP, 0x%x): %d/%d: %#" PRIx64 " period: %" PRIu64 " addr: %#" PRIx64 "\n",
+	       event->header.misc, sample->pid, sample->tid, sample->ip,
+	       sample->period, sample->addr);
+
+	sample_type = evsel->attr.sample_type;
+
+	if (sample_type & PERF_SAMPLE_CALLCHAIN)
+		callchain__printf(evsel, sample);
+
+	if ((sample_type & PERF_SAMPLE_BRANCH_STACK) && !has_branch_callstack(evsel))
+		branch_stack__printf(sample);
+
+	if (sample_type & PERF_SAMPLE_REGS_USER)
+		regs_user__printf(sample);
+
+	if (sample_type & PERF_SAMPLE_REGS_INTR)
+		regs_intr__printf(sample);
+
+	if (sample_type & PERF_SAMPLE_STACK_USER)
+		stack_user__printf(&sample->user_stack);
+
+	if (sample_type & PERF_SAMPLE_WEIGHT)
+		printf("... weight: %" PRIu64 "\n", sample->weight);
+
+	if (sample_type & PERF_SAMPLE_DATA_SRC)
+		printf(" . data_src: 0x%"PRIx64"\n", sample->data_src);
+
+	if (sample_type & PERF_SAMPLE_TRANSACTION)
+		printf("... transaction: %" PRIx64 "\n", sample->transaction);
+
+	if (sample_type & PERF_SAMPLE_READ)
+		sample_read__printf(sample, evsel->attr.read_format);
+}
+
+static struct machine *machines__find_for_cpumode(struct machines *machines,
+					       union perf_event *event,
+					       struct perf_sample *sample)
+{
+	const u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
+	struct machine *machine;
+
+	if (perf_guest &&
+	    ((cpumode == PERF_RECORD_MISC_GUEST_KERNEL) ||
+	     (cpumode == PERF_RECORD_MISC_GUEST_USER))) {
+		u32 pid;
+
+		if (event->header.type == PERF_RECORD_MMAP
+		    || event->header.type == PERF_RECORD_MMAP2)
+			pid = event->mmap.pid;
+		else
+			pid = sample->pid;
+
+		machine = machines__find(machines, pid);
+		if (!machine)
+			machine = machines__find(machines, DEFAULT_GUEST_KERNEL_ID);
+		return machine;
+	}
+
+	return &machines->host;
+}
+
+static int deliver_sample_value(struct perf_evlist *evlist,
+				struct perf_tool *tool,
+				union perf_event *event,
+				struct perf_sample *sample,
+				struct sample_read_value *v,
+				struct machine *machine)
+{
+	struct perf_sample_id *sid = perf_evlist__id2sid(evlist, v->id);
+
+	if (sid) {
+		sample->id     = v->id;
+		sample->period = v->value - sid->period;
+		sid->period    = v->value;
+	}
+
+	if (!sid || sid->evsel == NULL) {
+		++evlist->stats.nr_unknown_id;
+		return 0;
+	}
+
+	return tool->sample(tool, event, sample, sid->evsel, machine);
+}
+
+static int deliver_sample_group(struct perf_evlist *evlist,
+				struct perf_tool *tool,
+				union  perf_event *event,
+				struct perf_sample *sample,
+				struct machine *machine)
+{
+	int ret = -EINVAL;
+	u64 i;
+
+	for (i = 0; i < sample->read.group.nr; i++) {
+		ret = deliver_sample_value(evlist, tool, event, sample,
+					   &sample->read.group.values[i],
+					   machine);
+		if (ret)
+			break;
+	}
+
+	return ret;
+}
+
+static int
+ perf_evlist__deliver_sample(struct perf_evlist *evlist,
+			     struct perf_tool *tool,
+			     union  perf_event *event,
+			     struct perf_sample *sample,
+			     struct perf_evsel *evsel,
+			     struct machine *machine)
+{
+	/* We know evsel != NULL. */
+	u64 sample_type = evsel->attr.sample_type;
+	u64 read_format = evsel->attr.read_format;
+
+	/* Standard sample delievery. */
+	if (!(sample_type & PERF_SAMPLE_READ))
+		return tool->sample(tool, event, sample, evsel, machine);
+
+	/* For PERF_SAMPLE_READ we have either single or group mode. */
+	if (read_format & PERF_FORMAT_GROUP)
+		return deliver_sample_group(evlist, tool, event, sample,
+					    machine);
+	else
+		return deliver_sample_value(evlist, tool, event, sample,
+					    &sample->read.one, machine);
+}
+
+static int machines__deliver_event(struct machines *machines,
+				   struct perf_evlist *evlist,
+				   union perf_event *event,
+				   struct perf_sample *sample,
+				   struct perf_tool *tool, u64 file_offset)
+{
+	struct perf_evsel *evsel;
+	struct machine *machine;
+
+	dump_event(evlist, event, file_offset, sample);
+
+	evsel = perf_evlist__id2evsel(evlist, sample->id);
+
+	machine = machines__find_for_cpumode(machines, event, sample);
+
+	switch (event->header.type) {
+	case PERF_RECORD_SAMPLE:
+		dump_sample(evsel, event, sample);
+		if (evsel == NULL) {
+			++evlist->stats.nr_unknown_id;
+			return 0;
+		}
+		if (machine == NULL) {
+			++evlist->stats.nr_unprocessable_samples;
+			return 0;
+		}
+		return perf_evlist__deliver_sample(evlist, tool, event, sample, evsel, machine);
+	case PERF_RECORD_MMAP:
+		return tool->mmap(tool, event, sample, machine);
+	case PERF_RECORD_MMAP2:
+		return tool->mmap2(tool, event, sample, machine);
+	case PERF_RECORD_COMM:
+		return tool->comm(tool, event, sample, machine);
+	case PERF_RECORD_FORK:
+		return tool->fork(tool, event, sample, machine);
+	case PERF_RECORD_EXIT:
+		return tool->exit(tool, event, sample, machine);
+	case PERF_RECORD_LOST:
+		if (tool->lost == perf_event__process_lost)
+			evlist->stats.total_lost += event->lost.lost;
+		return tool->lost(tool, event, sample, machine);
+	case PERF_RECORD_READ:
+		return tool->read(tool, event, sample, evsel, machine);
+	case PERF_RECORD_THROTTLE:
+		return tool->throttle(tool, event, sample, machine);
+	case PERF_RECORD_UNTHROTTLE:
+		return tool->unthrottle(tool, event, sample, machine);
+	default:
+		++evlist->stats.nr_unknown_events;
+		return -1;
+	}
+}
+
+static s64 perf_session__process_user_event(struct perf_session *session,
+					    union perf_event *event,
+					    u64 file_offset)
+{
+	struct ordered_events *oe = &session->ordered_events;
+	struct perf_tool *tool = session->tool;
+	int fd = perf_data_file__fd(session->file);
+	int err;
+
+	dump_event(session->evlist, event, file_offset, NULL);
+
+	/* These events are processed right away */
+	switch (event->header.type) {
+	case PERF_RECORD_HEADER_ATTR:
+		err = tool->attr(tool, event, &session->evlist);
+		if (err == 0) {
+			perf_session__set_id_hdr_size(session);
+			perf_session__set_comm_exec(session);
+		}
+		return err;
+	case PERF_RECORD_HEADER_EVENT_TYPE:
+		/*
+		 * Depreceated, but we need to handle it for sake
+		 * of old data files create in pipe mode.
+		 */
+		return 0;
+	case PERF_RECORD_HEADER_TRACING_DATA:
+		/* setup for reading amidst mmap */
+		lseek(fd, file_offset, SEEK_SET);
+		return tool->tracing_data(tool, event, session);
+	case PERF_RECORD_HEADER_BUILD_ID:
+		return tool->build_id(tool, event, session);
+	case PERF_RECORD_FINISHED_ROUND:
+		return tool->finished_round(tool, event, oe);
+	case PERF_RECORD_ID_INDEX:
+		return tool->id_index(tool, event, session);
+	default:
+		return -EINVAL;
+	}
+}
+
+int perf_session__deliver_synth_event(struct perf_session *session,
+				      union perf_event *event,
+				      struct perf_sample *sample)
+{
+	struct perf_evlist *evlist = session->evlist;
+	struct perf_tool *tool = session->tool;
+
+	events_stats__inc(&evlist->stats, event->header.type);
+
+	if (event->header.type >= PERF_RECORD_USER_TYPE_START)
+		return perf_session__process_user_event(session, event, 0);
+
+	return machines__deliver_event(&session->machines, evlist, event, sample, tool, 0);
+}
+
+static void event_swap(union perf_event *event, bool sample_id_all)
+{
+	perf_event__swap_op swap;
+
+	swap = perf_event__swap_ops[event->header.type];
+	if (swap)
+		swap(event, sample_id_all);
+}
+
+int perf_session__peek_event(struct perf_session *session, off_t file_offset,
+			     void *buf, size_t buf_sz,
+			     union perf_event **event_ptr,
+			     struct perf_sample *sample)
+{
+	union perf_event *event;
+	size_t hdr_sz, rest;
+	int fd;
+
+	if (session->one_mmap && !session->header.needs_swap) {
+		event = file_offset - session->one_mmap_offset +
+			session->one_mmap_addr;
+		goto out_parse_sample;
+	}
+
+	if (perf_data_file__is_pipe(session->file))
+		return -1;
+
+	fd = perf_data_file__fd(session->file);
+	hdr_sz = sizeof(struct perf_event_header);
+
+	if (buf_sz < hdr_sz)
+		return -1;
+
+	if (lseek(fd, file_offset, SEEK_SET) == (off_t)-1 ||
+	    readn(fd, &buf, hdr_sz) != (ssize_t)hdr_sz)
+		return -1;
+
+	event = (union perf_event *)buf;
+
+	if (session->header.needs_swap)
+		perf_event_header__bswap(&event->header);
+
+	if (event->header.size < hdr_sz)
+		return -1;
+
+	rest = event->header.size - hdr_sz;
+
+	if (readn(fd, &buf, rest) != (ssize_t)rest)
+		return -1;
+
+	if (session->header.needs_swap)
+		event_swap(event, perf_evlist__sample_id_all(session->evlist));
+
+out_parse_sample:
+
+	if (sample && event->header.type < PERF_RECORD_USER_TYPE_START &&
+	    perf_evlist__parse_sample(session->evlist, event, sample))
+		return -1;
+
+	*event_ptr = event;
+
+	return 0;
+}
+
+static s64 perf_session__process_event(struct perf_session *session,
+				       union perf_event *event, u64 file_offset)
+{
+	struct perf_evlist *evlist = session->evlist;
+	struct perf_tool *tool = session->tool;
+	struct perf_sample sample;
+	int ret;
+
+	if (session->header.needs_swap)
+		event_swap(event, perf_evlist__sample_id_all(evlist));
+
+	if (event->header.type >= PERF_RECORD_HEADER_MAX)
+		return -EINVAL;
+
+	events_stats__inc(&evlist->stats, event->header.type);
+
+	if (event->header.type >= PERF_RECORD_USER_TYPE_START)
+		return perf_session__process_user_event(session, event, file_offset);
+
+	/*
+	 * For all kernel events we get the sample data
+	 */
+	ret = perf_evlist__parse_sample(evlist, event, &sample);
+	if (ret)
+		return ret;
+
+	if (tool->ordered_events) {
+		ret = perf_session__queue_event(session, event, &sample, file_offset);
+		if (ret != -ETIME)
+			return ret;
+	}
+
+	return machines__deliver_event(&session->machines, evlist, event,
+				       &sample, tool, file_offset);
+}
+
+void perf_event_header__bswap(struct perf_event_header *hdr)
+{
+	hdr->type = bswap_32(hdr->type);
+	hdr->misc = bswap_16(hdr->misc);
+	hdr->size = bswap_16(hdr->size);
+}
+
+struct thread *perf_session__findnew(struct perf_session *session, pid_t pid)
+{
+	return machine__findnew_thread(&session->machines.host, -1, pid);
+}
+
+static struct thread *perf_session__register_idle_thread(struct perf_session *session)
+{
+	struct thread *thread;
+
+	thread = machine__findnew_thread(&session->machines.host, 0, 0);
+	if (thread == NULL || thread__set_comm(thread, "swapper", 0)) {
+		pr_err("problem inserting idle task.\n");
+		thread = NULL;
+	}
+
+	return thread;
+}
+
+static void perf_session__warn_about_errors(const struct perf_session *session)
+{
+	const struct events_stats *stats = &session->evlist->stats;
+	const struct ordered_events *oe = &session->ordered_events;
+
+	if (session->tool->lost == perf_event__process_lost &&
+	    stats->nr_events[PERF_RECORD_LOST] != 0) {
+		ui__warning("Processed %d events and lost %d chunks!\n\n"
+			    "Check IO/CPU overload!\n\n",
+			    stats->nr_events[0],
+			    stats->nr_events[PERF_RECORD_LOST]);
+	}
+
+	if (stats->nr_unknown_events != 0) {
+		ui__warning("Found %u unknown events!\n\n"
+			    "Is this an older tool processing a perf.data "
+			    "file generated by a more recent tool?\n\n"
+			    "If that is not the case, consider "
+			    "reporting to linux-kernel@vger.kernel.org.\n\n",
+			    stats->nr_unknown_events);
+	}
+
+	if (stats->nr_unknown_id != 0) {
+		ui__warning("%u samples with id not present in the header\n",
+			    stats->nr_unknown_id);
+	}
+
+	if (stats->nr_invalid_chains != 0) {
+		ui__warning("Found invalid callchains!\n\n"
+			    "%u out of %u events were discarded for this reason.\n\n"
+			    "Consider reporting to linux-kernel@vger.kernel.org.\n\n",
+			    stats->nr_invalid_chains,
+			    stats->nr_events[PERF_RECORD_SAMPLE]);
+	}
+
+	if (stats->nr_unprocessable_samples != 0) {
+		ui__warning("%u unprocessable samples recorded.\n"
+			    "Do you have a KVM guest running and not using 'perf kvm'?\n",
+			    stats->nr_unprocessable_samples);
+	}
+
+	if (oe->nr_unordered_events != 0)
+		ui__warning("%u out of order events recorded.\n", oe->nr_unordered_events);
+}
+
+volatile int session_done;
+
+static int __perf_session__process_pipe_events(struct perf_session *session)
+{
+	struct ordered_events *oe = &session->ordered_events;
+	struct perf_tool *tool = session->tool;
+	int fd = perf_data_file__fd(session->file);
+	union perf_event *event;
+	uint32_t size, cur_size = 0;
+	void *buf = NULL;
+	s64 skip = 0;
+	u64 head;
+	ssize_t err;
+	void *p;
+
+	perf_tool__fill_defaults(tool);
+
+	head = 0;
+	cur_size = sizeof(union perf_event);
+
+	buf = malloc(cur_size);
+	if (!buf)
+		return -errno;
+more:
+	event = buf;
+	err = readn(fd, event, sizeof(struct perf_event_header));
+	if (err <= 0) {
+		if (err == 0)
+			goto done;
+
+		pr_err("failed to read event header\n");
+		goto out_err;
+	}
+
+	if (session->header.needs_swap)
+		perf_event_header__bswap(&event->header);
+
+	size = event->header.size;
+	if (size < sizeof(struct perf_event_header)) {
+		pr_err("bad event header size\n");
+		goto out_err;
+	}
+
+	if (size > cur_size) {
+		void *new = realloc(buf, size);
+		if (!new) {
+			pr_err("failed to allocate memory to read event\n");
+			goto out_err;
+		}
+		buf = new;
+		cur_size = size;
+		event = buf;
+	}
+	p = event;
+	p += sizeof(struct perf_event_header);
+
+	if (size - sizeof(struct perf_event_header)) {
+		err = readn(fd, p, size - sizeof(struct perf_event_header));
+		if (err <= 0) {
+			if (err == 0) {
+				pr_err("unexpected end of event stream\n");
+				goto done;
+			}
+
+			pr_err("failed to read event data\n");
+			goto out_err;
+		}
+	}
+
+	if ((skip = perf_session__process_event(session, event, head)) < 0) {
+		pr_err("%#" PRIx64 " [%#x]: failed to process type: %d\n",
+		       head, event->header.size, event->header.type);
+		err = -EINVAL;
+		goto out_err;
+	}
+
+	head += size;
+
+	if (skip > 0)
+		head += skip;
+
+	if (!session_done())
+		goto more;
+done:
+	/* do the final flush for ordered samples */
+	err = ordered_events__flush(oe, OE_FLUSH__FINAL);
+out_err:
+	free(buf);
+	perf_session__warn_about_errors(session);
+	ordered_events__free(&session->ordered_events);
+	return err;
+}
+
+static union perf_event *
+fetch_mmaped_event(struct perf_session *session,
+		   u64 head, size_t mmap_size, char *buf)
+{
+	union perf_event *event;
+
+	/*
+	 * Ensure we have enough space remaining to read
+	 * the size of the event in the headers.
+	 */
+	if (head + sizeof(event->header) > mmap_size)
+		return NULL;
+
+	event = (union perf_event *)(buf + head);
+
+	if (session->header.needs_swap)
+		perf_event_header__bswap(&event->header);
+
+	if (head + event->header.size > mmap_size) {
+		/* We're not fetching the event so swap back again */
+		if (session->header.needs_swap)
+			perf_event_header__bswap(&event->header);
+		return NULL;
+	}
+
+	return event;
+}
+
+/*
+ * On 64bit we can mmap the data file in one go. No need for tiny mmap
+ * slices. On 32bit we use 32MB.
+ */
+#if BITS_PER_LONG == 64
+#define MMAP_SIZE ULLONG_MAX
+#define NUM_MMAPS 1
+#else
+#define MMAP_SIZE (32 * 1024 * 1024ULL)
+#define NUM_MMAPS 128
+#endif
+
+static int __perf_session__process_events(struct perf_session *session,
+					  u64 data_offset, u64 data_size,
+					  u64 file_size)
+{
+	struct ordered_events *oe = &session->ordered_events;
+	struct perf_tool *tool = session->tool;
+	int fd = perf_data_file__fd(session->file);
+	u64 head, page_offset, file_offset, file_pos, size;
+	int err, mmap_prot, mmap_flags, map_idx = 0;
+	size_t	mmap_size;
+	char *buf, *mmaps[NUM_MMAPS];
+	union perf_event *event;
+	struct ui_progress prog;
+	s64 skip;
+
+	perf_tool__fill_defaults(tool);
+
+	page_offset = page_size * (data_offset / page_size);
+	file_offset = page_offset;
+	head = data_offset - page_offset;
+
+	if (data_size && (data_offset + data_size < file_size))
+		file_size = data_offset + data_size;
+
+	ui_progress__init(&prog, file_size, "Processing events...");
+
+	mmap_size = MMAP_SIZE;
+	if (mmap_size > file_size) {
+		mmap_size = file_size;
+		session->one_mmap = true;
+	}
+
+	memset(mmaps, 0, sizeof(mmaps));
+
+	mmap_prot  = PROT_READ;
+	mmap_flags = MAP_SHARED;
+
+	if (session->header.needs_swap) {
+		mmap_prot  |= PROT_WRITE;
+		mmap_flags = MAP_PRIVATE;
+	}
+remap:
+	buf = mmap(NULL, mmap_size, mmap_prot, mmap_flags, fd,
+		   file_offset);
+	if (buf == MAP_FAILED) {
+		pr_err("failed to mmap file\n");
+		err = -errno;
+		goto out_err;
+	}
+	mmaps[map_idx] = buf;
+	map_idx = (map_idx + 1) & (ARRAY_SIZE(mmaps) - 1);
+	file_pos = file_offset + head;
+	if (session->one_mmap) {
+		session->one_mmap_addr = buf;
+		session->one_mmap_offset = file_offset;
+	}
+
+more:
+	event = fetch_mmaped_event(session, head, mmap_size, buf);
+	if (!event) {
+		if (mmaps[map_idx]) {
+			munmap(mmaps[map_idx], mmap_size);
+			mmaps[map_idx] = NULL;
+		}
+
+		page_offset = page_size * (head / page_size);
+		file_offset += page_offset;
+		head -= page_offset;
+		goto remap;
+	}
+
+	size = event->header.size;
+
+	if (size < sizeof(struct perf_event_header) ||
+	    (skip = perf_session__process_event(session, event, file_pos)) < 0) {
+		pr_err("%#" PRIx64 " [%#x]: failed to process type: %d\n",
+		       file_offset + head, event->header.size,
+		       event->header.type);
+		err = -EINVAL;
+		goto out_err;
+	}
+
+	if (skip)
+		size += skip;
+
+	head += size;
+	file_pos += size;
+
+	ui_progress__update(&prog, size);
+
+	if (session_done())
+		goto out;
+
+	if (file_pos < file_size)
+		goto more;
+
+out:
+	/* do the final flush for ordered samples */
+	err = ordered_events__flush(oe, OE_FLUSH__FINAL);
+out_err:
+	ui_progress__finish();
+	perf_session__warn_about_errors(session);
+	ordered_events__free(&session->ordered_events);
+	session->one_mmap = false;
+	return err;
+}
+
+int perf_session__process_events(struct perf_session *session)
+{
+	u64 size = perf_data_file__size(session->file);
+	int err;
+
+	if (perf_session__register_idle_thread(session) == NULL)
+		return -ENOMEM;
+
+	if (!perf_data_file__is_pipe(session->file))
+		err = __perf_session__process_events(session,
+						     session->header.data_offset,
+						     session->header.data_size, size);
+	else
+		err = __perf_session__process_pipe_events(session);
+
+	return err;
+}
+
+bool perf_session__has_traces(struct perf_session *session, const char *msg)
+{
+	struct perf_evsel *evsel;
+
+	evlist__for_each(session->evlist, evsel) {
+		if (evsel->attr.type == PERF_TYPE_TRACEPOINT)
+			return true;
+	}
+
+	pr_err("No trace sample to read. Did you call 'perf %s'?\n", msg);
+	return false;
+}
+
+int maps__set_kallsyms_ref_reloc_sym(struct map **maps,
+				     const char *symbol_name, u64 addr)
+{
+	char *bracket;
+	enum map_type i;
+	struct ref_reloc_sym *ref;
+
+	ref = zalloc(sizeof(struct ref_reloc_sym));
+	if (ref == NULL)
+		return -ENOMEM;
+
+	ref->name = strdup(symbol_name);
+	if (ref->name == NULL) {
+		free(ref);
+		return -ENOMEM;
+	}
+
+	bracket = strchr(ref->name, ']');
+	if (bracket)
+		*bracket = '\0';
+
+	ref->addr = addr;
+
+	for (i = 0; i < MAP__NR_TYPES; ++i) {
+		struct kmap *kmap = map__kmap(maps[i]);
+
+		if (!kmap)
+			continue;
+		kmap->ref_reloc_sym = ref;
+	}
+
+	return 0;
+}
+
+size_t perf_session__fprintf_dsos(struct perf_session *session, FILE *fp)
+{
+	return machines__fprintf_dsos(&session->machines, fp);
+}
+
+size_t perf_session__fprintf_dsos_buildid(struct perf_session *session, FILE *fp,
+					  bool (skip)(struct dso *dso, int parm), int parm)
+{
+	return machines__fprintf_dsos_buildid(&session->machines, fp, skip, parm);
+}
+
+size_t perf_session__fprintf_nr_events(struct perf_session *session, FILE *fp)
+{
+	size_t ret = fprintf(fp, "Aggregated stats:\n");
+
+	ret += events_stats__fprintf(&session->evlist->stats, fp);
+	return ret;
+}
+
+size_t perf_session__fprintf(struct perf_session *session, FILE *fp)
+{
+	/*
+	 * FIXME: Here we have to actually print all the machines in this
+	 * session, not just the host...
+	 */
+	return machine__fprintf(&session->machines.host, fp);
+}
+
+struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session,
+					      unsigned int type)
+{
+	struct perf_evsel *pos;
+
+	evlist__for_each(session->evlist, pos) {
+		if (pos->attr.type == type)
+			return pos;
+	}
+	return NULL;
+}
+
+void perf_evsel__print_ip(struct perf_evsel *evsel, struct perf_sample *sample,
+			  struct addr_location *al,
+			  unsigned int print_opts, unsigned int stack_depth)
+{
+	struct callchain_cursor_node *node;
+	int print_ip = print_opts & PRINT_IP_OPT_IP;
+	int print_sym = print_opts & PRINT_IP_OPT_SYM;
+	int print_dso = print_opts & PRINT_IP_OPT_DSO;
+	int print_symoffset = print_opts & PRINT_IP_OPT_SYMOFFSET;
+	int print_oneline = print_opts & PRINT_IP_OPT_ONELINE;
+	int print_srcline = print_opts & PRINT_IP_OPT_SRCLINE;
+	char s = print_oneline ? ' ' : '\t';
+
+	if (symbol_conf.use_callchain && sample->callchain) {
+		struct addr_location node_al;
+
+		if (thread__resolve_callchain(al->thread, evsel,
+					      sample, NULL, NULL,
+					      PERF_MAX_STACK_DEPTH) != 0) {
+			if (verbose)
+				error("Failed to resolve callchain. Skipping\n");
+			return;
+		}
+		callchain_cursor_commit(&callchain_cursor);
+
+		if (print_symoffset)
+			node_al = *al;
+
+		while (stack_depth) {
+			u64 addr = 0;
+
+			node = callchain_cursor_current(&callchain_cursor);
+			if (!node)
+				break;
+
+			if (node->sym && node->sym->ignore)
+				goto next;
+
+			if (print_ip)
+				printf("%c%16" PRIx64, s, node->ip);
+
+			if (node->map)
+				addr = node->map->map_ip(node->map, node->ip);
+
+			if (print_sym) {
+				printf(" ");
+				if (print_symoffset) {
+					node_al.addr = addr;
+					node_al.map  = node->map;
+					symbol__fprintf_symname_offs(node->sym, &node_al, stdout);
+				} else
+					symbol__fprintf_symname(node->sym, stdout);
+			}
+
+			if (print_dso) {
+				printf(" (");
+				map__fprintf_dsoname(node->map, stdout);
+				printf(")");
+			}
+
+			if (print_srcline)
+				map__fprintf_srcline(node->map, addr, "\n  ",
+						     stdout);
+
+			if (!print_oneline)
+				printf("\n");
+
+			stack_depth--;
+next:
+			callchain_cursor_advance(&callchain_cursor);
+		}
+
+	} else {
+		if (al->sym && al->sym->ignore)
+			return;
+
+		if (print_ip)
+			printf("%16" PRIx64, sample->ip);
+
+		if (print_sym) {
+			printf(" ");
+			if (print_symoffset)
+				symbol__fprintf_symname_offs(al->sym, al,
+							     stdout);
+			else
+				symbol__fprintf_symname(al->sym, stdout);
+		}
+
+		if (print_dso) {
+			printf(" (");
+			map__fprintf_dsoname(al->map, stdout);
+			printf(")");
+		}
+
+		if (print_srcline)
+			map__fprintf_srcline(al->map, al->addr, "\n  ", stdout);
+	}
+}
+
+int perf_session__cpu_bitmap(struct perf_session *session,
+			     const char *cpu_list, unsigned long *cpu_bitmap)
+{
+	int i, err = -1;
+	struct cpu_map *map;
+
+	for (i = 0; i < PERF_TYPE_MAX; ++i) {
+		struct perf_evsel *evsel;
+
+		evsel = perf_session__find_first_evtype(session, i);
+		if (!evsel)
+			continue;
+
+		if (!(evsel->attr.sample_type & PERF_SAMPLE_CPU)) {
+			pr_err("File does not contain CPU events. "
+			       "Remove -c option to proceed.\n");
+			return -1;
+		}
+	}
+
+	map = cpu_map__new(cpu_list);
+	if (map == NULL) {
+		pr_err("Invalid cpu_list\n");
+		return -1;
+	}
+
+	for (i = 0; i < map->nr; i++) {
+		int cpu = map->map[i];
+
+		if (cpu >= MAX_NR_CPUS) {
+			pr_err("Requested CPU %d too large. "
+			       "Consider raising MAX_NR_CPUS\n", cpu);
+			goto out_delete_map;
+		}
+
+		set_bit(cpu, cpu_bitmap);
+	}
+
+	err = 0;
+
+out_delete_map:
+	cpu_map__delete(map);
+	return err;
+}
+
+void perf_session__fprintf_info(struct perf_session *session, FILE *fp,
+				bool full)
+{
+	struct stat st;
+	int fd, ret;
+
+	if (session == NULL || fp == NULL)
+		return;
+
+	fd = perf_data_file__fd(session->file);
+
+	ret = fstat(fd, &st);
+	if (ret == -1)
+		return;
+
+	fprintf(fp, "# ========\n");
+	fprintf(fp, "# captured on: %s", ctime(&st.st_ctime));
+	perf_header__fprintf_info(session, fp, full);
+	fprintf(fp, "# ========\n#\n");
+}
+
+
+int __perf_session__set_tracepoints_handlers(struct perf_session *session,
+					     const struct perf_evsel_str_handler *assocs,
+					     size_t nr_assocs)
+{
+	struct perf_evsel *evsel;
+	size_t i;
+	int err;
+
+	for (i = 0; i < nr_assocs; i++) {
+		/*
+		 * Adding a handler for an event not in the session,
+		 * just ignore it.
+		 */
+		evsel = perf_evlist__find_tracepoint_by_name(session->evlist, assocs[i].name);
+		if (evsel == NULL)
+			continue;
+
+		err = -EEXIST;
+		if (evsel->handler != NULL)
+			goto out;
+		evsel->handler = assocs[i].handler;
+	}
+
+	err = 0;
+out:
+	return err;
+}
+
+int perf_event__process_id_index(struct perf_tool *tool __maybe_unused,
+				 union perf_event *event,
+				 struct perf_session *session)
+{
+	struct perf_evlist *evlist = session->evlist;
+	struct id_index_event *ie = &event->id_index;
+	size_t i, nr, max_nr;
+
+	max_nr = (ie->header.size - sizeof(struct id_index_event)) /
+		 sizeof(struct id_index_entry);
+	nr = ie->nr;
+	if (nr > max_nr)
+		return -EINVAL;
+
+	if (dump_trace)
+		fprintf(stdout, " nr: %zu\n", nr);
+
+	for (i = 0; i < nr; i++) {
+		struct id_index_entry *e = &ie->entries[i];
+		struct perf_sample_id *sid;
+
+		if (dump_trace) {
+			fprintf(stdout,	" ... id: %"PRIu64, e->id);
+			fprintf(stdout,	"  idx: %"PRIu64, e->idx);
+			fprintf(stdout,	"  cpu: %"PRId64, e->cpu);
+			fprintf(stdout,	"  tid: %"PRId64"\n", e->tid);
+		}
+
+		sid = perf_evlist__id2sid(evlist, e->id);
+		if (!sid)
+			return -ENOENT;
+		sid->idx = e->idx;
+		sid->cpu = e->cpu;
+		sid->tid = e->tid;
+	}
+	return 0;
+}
+
+int perf_event__synthesize_id_index(struct perf_tool *tool,
+				    perf_event__handler_t process,
+				    struct perf_evlist *evlist,
+				    struct machine *machine)
+{
+	union perf_event *ev;
+	struct perf_evsel *evsel;
+	size_t nr = 0, i = 0, sz, max_nr, n;
+	int err;
+
+	pr_debug2("Synthesizing id index\n");
+
+	max_nr = (UINT16_MAX - sizeof(struct id_index_event)) /
+		 sizeof(struct id_index_entry);
+
+	evlist__for_each(evlist, evsel)
+		nr += evsel->ids;
+
+	n = nr > max_nr ? max_nr : nr;
+	sz = sizeof(struct id_index_event) + n * sizeof(struct id_index_entry);
+	ev = zalloc(sz);
+	if (!ev)
+		return -ENOMEM;
+
+	ev->id_index.header.type = PERF_RECORD_ID_INDEX;
+	ev->id_index.header.size = sz;
+	ev->id_index.nr = n;
+
+	evlist__for_each(evlist, evsel) {
+		u32 j;
+
+		for (j = 0; j < evsel->ids; j++) {
+			struct id_index_entry *e;
+			struct perf_sample_id *sid;
+
+			if (i >= n) {
+				err = process(tool, ev, NULL, machine);
+				if (err)
+					goto out_err;
+				nr -= n;
+				i = 0;
+			}
+
+			e = &ev->id_index.entries[i++];
+
+			e->id = evsel->id[j];
+
+			sid = perf_evlist__id2sid(evlist, e->id);
+			if (!sid) {
+				free(ev);
+				return -ENOENT;
+			}
+
+			e->idx = sid->idx;
+			e->cpu = sid->cpu;
+			e->tid = sid->tid;
+		}
+	}
+
+	sz = sizeof(struct id_index_event) + nr * sizeof(struct id_index_entry);
+	ev->id_index.header.size = sz;
+	ev->id_index.nr = nr;
+
+	err = process(tool, ev, NULL, machine);
+out_err:
+	free(ev);
+
+	return err;
+}
diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h
new file mode 100644
index 000000000..d5fa7b791
--- /dev/null
+++ b/tools/perf/util/session.h
@@ -0,0 +1,133 @@
+#ifndef __PERF_SESSION_H
+#define __PERF_SESSION_H
+
+#include "trace-event.h"
+#include "event.h"
+#include "header.h"
+#include "machine.h"
+#include "symbol.h"
+#include "thread.h"
+#include "data.h"
+#include "ordered-events.h"
+#include <linux/rbtree.h>
+#include <linux/perf_event.h>
+
+struct ip_callchain;
+struct thread;
+
+struct perf_session {
+	struct perf_header	header;
+	struct machines		machines;
+	struct perf_evlist	*evlist;
+	struct trace_event	tevent;
+	bool			repipe;
+	bool			one_mmap;
+	void			*one_mmap_addr;
+	u64			one_mmap_offset;
+	struct ordered_events	ordered_events;
+	struct perf_data_file	*file;
+	struct perf_tool	*tool;
+};
+
+#define PRINT_IP_OPT_IP		(1<<0)
+#define PRINT_IP_OPT_SYM		(1<<1)
+#define PRINT_IP_OPT_DSO		(1<<2)
+#define PRINT_IP_OPT_SYMOFFSET	(1<<3)
+#define PRINT_IP_OPT_ONELINE	(1<<4)
+#define PRINT_IP_OPT_SRCLINE	(1<<5)
+
+struct perf_tool;
+
+struct perf_session *perf_session__new(struct perf_data_file *file,
+				       bool repipe, struct perf_tool *tool);
+void perf_session__delete(struct perf_session *session);
+
+void perf_event_header__bswap(struct perf_event_header *hdr);
+
+int perf_session__peek_event(struct perf_session *session, off_t file_offset,
+			     void *buf, size_t buf_sz,
+			     union perf_event **event_ptr,
+			     struct perf_sample *sample);
+
+int perf_session__process_events(struct perf_session *session);
+
+int perf_session__queue_event(struct perf_session *s, union perf_event *event,
+			      struct perf_sample *sample, u64 file_offset);
+
+void perf_tool__fill_defaults(struct perf_tool *tool);
+
+int perf_session__resolve_callchain(struct perf_session *session,
+				    struct perf_evsel *evsel,
+				    struct thread *thread,
+				    struct ip_callchain *chain,
+				    struct symbol **parent);
+
+bool perf_session__has_traces(struct perf_session *session, const char *msg);
+
+void perf_event__attr_swap(struct perf_event_attr *attr);
+
+int perf_session__create_kernel_maps(struct perf_session *session);
+
+void perf_session__set_id_hdr_size(struct perf_session *session);
+
+static inline
+struct machine *perf_session__find_machine(struct perf_session *session, pid_t pid)
+{
+	return machines__find(&session->machines, pid);
+}
+
+static inline
+struct machine *perf_session__findnew_machine(struct perf_session *session, pid_t pid)
+{
+	return machines__findnew(&session->machines, pid);
+}
+
+struct thread *perf_session__findnew(struct perf_session *session, pid_t pid);
+size_t perf_session__fprintf(struct perf_session *session, FILE *fp);
+
+size_t perf_session__fprintf_dsos(struct perf_session *session, FILE *fp);
+
+size_t perf_session__fprintf_dsos_buildid(struct perf_session *session, FILE *fp,
+					  bool (fn)(struct dso *dso, int parm), int parm);
+
+size_t perf_session__fprintf_nr_events(struct perf_session *session, FILE *fp);
+
+struct perf_evsel *perf_session__find_first_evtype(struct perf_session *session,
+					    unsigned int type);
+
+void perf_evsel__print_ip(struct perf_evsel *evsel, struct perf_sample *sample,
+			  struct addr_location *al,
+			  unsigned int print_opts, unsigned int stack_depth);
+
+int perf_session__cpu_bitmap(struct perf_session *session,
+			     const char *cpu_list, unsigned long *cpu_bitmap);
+
+void perf_session__fprintf_info(struct perf_session *s, FILE *fp, bool full);
+
+struct perf_evsel_str_handler;
+
+int __perf_session__set_tracepoints_handlers(struct perf_session *session,
+					     const struct perf_evsel_str_handler *assocs,
+					     size_t nr_assocs);
+
+#define perf_session__set_tracepoints_handlers(session, array) \
+	__perf_session__set_tracepoints_handlers(session, array, ARRAY_SIZE(array))
+
+extern volatile int session_done;
+
+#define session_done()	ACCESS_ONCE(session_done)
+
+int perf_session__deliver_synth_event(struct perf_session *session,
+				      union perf_event *event,
+				      struct perf_sample *sample);
+
+int perf_event__process_id_index(struct perf_tool *tool,
+				 union perf_event *event,
+				 struct perf_session *session);
+
+int perf_event__synthesize_id_index(struct perf_tool *tool,
+				    perf_event__handler_t process,
+				    struct perf_evlist *evlist,
+				    struct machine *machine);
+
+#endif /* __PERF_SESSION_H */
diff --git a/tools/perf/util/setup.py b/tools/perf/util/setup.py
new file mode 100644
index 000000000..183310376
--- /dev/null
+++ b/tools/perf/util/setup.py
@@ -0,0 +1,48 @@
+#!/usr/bin/python2
+
+from distutils.core import setup, Extension
+from os import getenv
+
+from distutils.command.build_ext   import build_ext   as _build_ext
+from distutils.command.install_lib import install_lib as _install_lib
+
+class build_ext(_build_ext):
+    def finalize_options(self):
+        _build_ext.finalize_options(self)
+        self.build_lib  = build_lib
+        self.build_temp = build_tmp
+
+class install_lib(_install_lib):
+    def finalize_options(self):
+        _install_lib.finalize_options(self)
+        self.build_dir = build_lib
+
+
+cflags = getenv('CFLAGS', '').split()
+# switch off several checks (need to be at the end of cflags list)
+cflags += ['-fno-strict-aliasing', '-Wno-write-strings', '-Wno-unused-parameter' ]
+
+build_lib = getenv('PYTHON_EXTBUILD_LIB')
+build_tmp = getenv('PYTHON_EXTBUILD_TMP')
+libtraceevent = getenv('LIBTRACEEVENT')
+libapikfs = getenv('LIBAPI')
+
+ext_sources = [f.strip() for f in file('util/python-ext-sources')
+				if len(f.strip()) > 0 and f[0] != '#']
+
+perf = Extension('perf',
+		  sources = ext_sources,
+		  include_dirs = ['util/include'],
+		  extra_compile_args = cflags,
+		  extra_objects = [libtraceevent, libapikfs],
+                 )
+
+setup(name='perf',
+      version='0.1',
+      description='Interface with the Linux profiling infrastructure',
+      author='Arnaldo Carvalho de Melo',
+      author_email='acme@redhat.com',
+      license='GPLv2',
+      url='http://perf.wiki.kernel.org',
+      ext_modules=[perf],
+      cmdclass={'build_ext': build_ext, 'install_lib': install_lib})
diff --git a/tools/perf/util/sigchain.c b/tools/perf/util/sigchain.c
new file mode 100644
index 000000000..ba785e9b1
--- /dev/null
+++ b/tools/perf/util/sigchain.c
@@ -0,0 +1,52 @@
+#include "sigchain.h"
+#include "cache.h"
+
+#define SIGCHAIN_MAX_SIGNALS 32
+
+struct sigchain_signal {
+	sigchain_fun *old;
+	int n;
+	int alloc;
+};
+static struct sigchain_signal signals[SIGCHAIN_MAX_SIGNALS];
+
+static void check_signum(int sig)
+{
+	if (sig < 1 || sig >= SIGCHAIN_MAX_SIGNALS)
+		die("BUG: signal out of range: %d", sig);
+}
+
+static int sigchain_push(int sig, sigchain_fun f)
+{
+	struct sigchain_signal *s = signals + sig;
+	check_signum(sig);
+
+	ALLOC_GROW(s->old, s->n + 1, s->alloc);
+	s->old[s->n] = signal(sig, f);
+	if (s->old[s->n] == SIG_ERR)
+		return -1;
+	s->n++;
+	return 0;
+}
+
+int sigchain_pop(int sig)
+{
+	struct sigchain_signal *s = signals + sig;
+	check_signum(sig);
+	if (s->n < 1)
+		return 0;
+
+	if (signal(sig, s->old[s->n - 1]) == SIG_ERR)
+		return -1;
+	s->n--;
+	return 0;
+}
+
+void sigchain_push_common(sigchain_fun f)
+{
+	sigchain_push(SIGINT, f);
+	sigchain_push(SIGHUP, f);
+	sigchain_push(SIGTERM, f);
+	sigchain_push(SIGQUIT, f);
+	sigchain_push(SIGPIPE, f);
+}
diff --git a/tools/perf/util/sigchain.h b/tools/perf/util/sigchain.h
new file mode 100644
index 000000000..959d64eb5
--- /dev/null
+++ b/tools/perf/util/sigchain.h
@@ -0,0 +1,10 @@
+#ifndef __PERF_SIGCHAIN_H
+#define __PERF_SIGCHAIN_H
+
+typedef void (*sigchain_fun)(int);
+
+int sigchain_pop(int sig);
+
+void sigchain_push_common(sigchain_fun f);
+
+#endif /* __PERF_SIGCHAIN_H */
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
new file mode 100644
index 000000000..4593f36ec
--- /dev/null
+++ b/tools/perf/util/sort.c
@@ -0,0 +1,1861 @@
+#include <sys/mman.h>
+#include "sort.h"
+#include "hist.h"
+#include "comm.h"
+#include "symbol.h"
+#include "evsel.h"
+
+regex_t		parent_regex;
+const char	default_parent_pattern[] = "^sys_|^do_page_fault";
+const char	*parent_pattern = default_parent_pattern;
+const char	default_sort_order[] = "comm,dso,symbol";
+const char	default_branch_sort_order[] = "comm,dso_from,symbol_from,dso_to,symbol_to";
+const char	default_mem_sort_order[] = "local_weight,mem,sym,dso,symbol_daddr,dso_daddr,snoop,tlb,locked";
+const char	default_top_sort_order[] = "dso,symbol";
+const char	default_diff_sort_order[] = "dso,symbol";
+const char	*sort_order;
+const char	*field_order;
+regex_t		ignore_callees_regex;
+int		have_ignore_callees = 0;
+int		sort__need_collapse = 0;
+int		sort__has_parent = 0;
+int		sort__has_sym = 0;
+int		sort__has_dso = 0;
+enum sort_mode	sort__mode = SORT_MODE__NORMAL;
+
+
+static int repsep_snprintf(char *bf, size_t size, const char *fmt, ...)
+{
+	int n;
+	va_list ap;
+
+	va_start(ap, fmt);
+	n = vsnprintf(bf, size, fmt, ap);
+	if (symbol_conf.field_sep && n > 0) {
+		char *sep = bf;
+
+		while (1) {
+			sep = strchr(sep, *symbol_conf.field_sep);
+			if (sep == NULL)
+				break;
+			*sep = '.';
+		}
+	}
+	va_end(ap);
+
+	if (n >= (int)size)
+		return size - 1;
+	return n;
+}
+
+static int64_t cmp_null(const void *l, const void *r)
+{
+	if (!l && !r)
+		return 0;
+	else if (!l)
+		return -1;
+	else
+		return 1;
+}
+
+/* --sort pid */
+
+static int64_t
+sort__thread_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	return right->thread->tid - left->thread->tid;
+}
+
+static int hist_entry__thread_snprintf(struct hist_entry *he, char *bf,
+				       size_t size, unsigned int width)
+{
+	const char *comm = thread__comm_str(he->thread);
+
+	width = max(7U, width) - 6;
+	return repsep_snprintf(bf, size, "%5d:%-*.*s", he->thread->tid,
+			       width, width, comm ?: "");
+}
+
+struct sort_entry sort_thread = {
+	.se_header	= "  Pid:Command",
+	.se_cmp		= sort__thread_cmp,
+	.se_snprintf	= hist_entry__thread_snprintf,
+	.se_width_idx	= HISTC_THREAD,
+};
+
+/* --sort comm */
+
+static int64_t
+sort__comm_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	/* Compare the addr that should be unique among comm */
+	return comm__str(right->comm) - comm__str(left->comm);
+}
+
+static int64_t
+sort__comm_collapse(struct hist_entry *left, struct hist_entry *right)
+{
+	/* Compare the addr that should be unique among comm */
+	return comm__str(right->comm) - comm__str(left->comm);
+}
+
+static int64_t
+sort__comm_sort(struct hist_entry *left, struct hist_entry *right)
+{
+	return strcmp(comm__str(right->comm), comm__str(left->comm));
+}
+
+static int hist_entry__comm_snprintf(struct hist_entry *he, char *bf,
+				     size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%-*.*s", width, width, comm__str(he->comm));
+}
+
+struct sort_entry sort_comm = {
+	.se_header	= "Command",
+	.se_cmp		= sort__comm_cmp,
+	.se_collapse	= sort__comm_collapse,
+	.se_sort	= sort__comm_sort,
+	.se_snprintf	= hist_entry__comm_snprintf,
+	.se_width_idx	= HISTC_COMM,
+};
+
+/* --sort dso */
+
+static int64_t _sort__dso_cmp(struct map *map_l, struct map *map_r)
+{
+	struct dso *dso_l = map_l ? map_l->dso : NULL;
+	struct dso *dso_r = map_r ? map_r->dso : NULL;
+	const char *dso_name_l, *dso_name_r;
+
+	if (!dso_l || !dso_r)
+		return cmp_null(dso_r, dso_l);
+
+	if (verbose) {
+		dso_name_l = dso_l->long_name;
+		dso_name_r = dso_r->long_name;
+	} else {
+		dso_name_l = dso_l->short_name;
+		dso_name_r = dso_r->short_name;
+	}
+
+	return strcmp(dso_name_l, dso_name_r);
+}
+
+static int64_t
+sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	return _sort__dso_cmp(right->ms.map, left->ms.map);
+}
+
+static int _hist_entry__dso_snprintf(struct map *map, char *bf,
+				     size_t size, unsigned int width)
+{
+	if (map && map->dso) {
+		const char *dso_name = !verbose ? map->dso->short_name :
+			map->dso->long_name;
+		return repsep_snprintf(bf, size, "%-*.*s", width, width, dso_name);
+	}
+
+	return repsep_snprintf(bf, size, "%-*.*s", width, width, "[unknown]");
+}
+
+static int hist_entry__dso_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	return _hist_entry__dso_snprintf(he->ms.map, bf, size, width);
+}
+
+struct sort_entry sort_dso = {
+	.se_header	= "Shared Object",
+	.se_cmp		= sort__dso_cmp,
+	.se_snprintf	= hist_entry__dso_snprintf,
+	.se_width_idx	= HISTC_DSO,
+};
+
+/* --sort symbol */
+
+static int64_t _sort__addr_cmp(u64 left_ip, u64 right_ip)
+{
+	return (int64_t)(right_ip - left_ip);
+}
+
+static int64_t _sort__sym_cmp(struct symbol *sym_l, struct symbol *sym_r)
+{
+	u64 ip_l, ip_r;
+
+	if (!sym_l || !sym_r)
+		return cmp_null(sym_l, sym_r);
+
+	if (sym_l == sym_r)
+		return 0;
+
+	ip_l = sym_l->start;
+	ip_r = sym_r->start;
+
+	return (int64_t)(ip_r - ip_l);
+}
+
+static int64_t
+sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	int64_t ret;
+
+	if (!left->ms.sym && !right->ms.sym)
+		return _sort__addr_cmp(left->ip, right->ip);
+
+	/*
+	 * comparing symbol address alone is not enough since it's a
+	 * relative address within a dso.
+	 */
+	if (!sort__has_dso) {
+		ret = sort__dso_cmp(left, right);
+		if (ret != 0)
+			return ret;
+	}
+
+	return _sort__sym_cmp(left->ms.sym, right->ms.sym);
+}
+
+static int64_t
+sort__sym_sort(struct hist_entry *left, struct hist_entry *right)
+{
+	if (!left->ms.sym || !right->ms.sym)
+		return cmp_null(left->ms.sym, right->ms.sym);
+
+	return strcmp(right->ms.sym->name, left->ms.sym->name);
+}
+
+static int _hist_entry__sym_snprintf(struct map *map, struct symbol *sym,
+				     u64 ip, char level, char *bf, size_t size,
+				     unsigned int width)
+{
+	size_t ret = 0;
+
+	if (verbose) {
+		char o = map ? dso__symtab_origin(map->dso) : '!';
+		ret += repsep_snprintf(bf, size, "%-#*llx %c ",
+				       BITS_PER_LONG / 4 + 2, ip, o);
+	}
+
+	ret += repsep_snprintf(bf + ret, size - ret, "[%c] ", level);
+	if (sym && map) {
+		if (map->type == MAP__VARIABLE) {
+			ret += repsep_snprintf(bf + ret, size - ret, "%s", sym->name);
+			ret += repsep_snprintf(bf + ret, size - ret, "+0x%llx",
+					ip - map->unmap_ip(map, sym->start));
+			ret += repsep_snprintf(bf + ret, size - ret, "%-*s",
+				       width - ret, "");
+		} else {
+			ret += repsep_snprintf(bf + ret, size - ret, "%-*s",
+					       width - ret,
+					       sym->name);
+		}
+	} else {
+		size_t len = BITS_PER_LONG / 4;
+		ret += repsep_snprintf(bf + ret, size - ret, "%-#.*llx",
+				       len, ip);
+		ret += repsep_snprintf(bf + ret, size - ret, "%-*s",
+				       width - ret, "");
+	}
+
+	if (ret > width)
+		bf[width] = '\0';
+
+	return width;
+}
+
+static int hist_entry__sym_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	return _hist_entry__sym_snprintf(he->ms.map, he->ms.sym, he->ip,
+					 he->level, bf, size, width);
+}
+
+struct sort_entry sort_sym = {
+	.se_header	= "Symbol",
+	.se_cmp		= sort__sym_cmp,
+	.se_sort	= sort__sym_sort,
+	.se_snprintf	= hist_entry__sym_snprintf,
+	.se_width_idx	= HISTC_SYMBOL,
+};
+
+/* --sort srcline */
+
+static int64_t
+sort__srcline_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	if (!left->srcline) {
+		if (!left->ms.map)
+			left->srcline = SRCLINE_UNKNOWN;
+		else {
+			struct map *map = left->ms.map;
+			left->srcline = get_srcline(map->dso,
+					   map__rip_2objdump(map, left->ip),
+						    left->ms.sym, true);
+		}
+	}
+	if (!right->srcline) {
+		if (!right->ms.map)
+			right->srcline = SRCLINE_UNKNOWN;
+		else {
+			struct map *map = right->ms.map;
+			right->srcline = get_srcline(map->dso,
+					     map__rip_2objdump(map, right->ip),
+						     right->ms.sym, true);
+		}
+	}
+	return strcmp(right->srcline, left->srcline);
+}
+
+static int hist_entry__srcline_snprintf(struct hist_entry *he, char *bf,
+					size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%-*.*s", width, width, he->srcline);
+}
+
+struct sort_entry sort_srcline = {
+	.se_header	= "Source:Line",
+	.se_cmp		= sort__srcline_cmp,
+	.se_snprintf	= hist_entry__srcline_snprintf,
+	.se_width_idx	= HISTC_SRCLINE,
+};
+
+/* --sort parent */
+
+static int64_t
+sort__parent_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct symbol *sym_l = left->parent;
+	struct symbol *sym_r = right->parent;
+
+	if (!sym_l || !sym_r)
+		return cmp_null(sym_l, sym_r);
+
+	return strcmp(sym_r->name, sym_l->name);
+}
+
+static int hist_entry__parent_snprintf(struct hist_entry *he, char *bf,
+				       size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%-*.*s", width, width,
+			      he->parent ? he->parent->name : "[other]");
+}
+
+struct sort_entry sort_parent = {
+	.se_header	= "Parent symbol",
+	.se_cmp		= sort__parent_cmp,
+	.se_snprintf	= hist_entry__parent_snprintf,
+	.se_width_idx	= HISTC_PARENT,
+};
+
+/* --sort cpu */
+
+static int64_t
+sort__cpu_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	return right->cpu - left->cpu;
+}
+
+static int hist_entry__cpu_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%*.*d", width, width, he->cpu);
+}
+
+struct sort_entry sort_cpu = {
+	.se_header      = "CPU",
+	.se_cmp	        = sort__cpu_cmp,
+	.se_snprintf    = hist_entry__cpu_snprintf,
+	.se_width_idx	= HISTC_CPU,
+};
+
+/* sort keys for branch stacks */
+
+static int64_t
+sort__dso_from_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	if (!left->branch_info || !right->branch_info)
+		return cmp_null(left->branch_info, right->branch_info);
+
+	return _sort__dso_cmp(left->branch_info->from.map,
+			      right->branch_info->from.map);
+}
+
+static int hist_entry__dso_from_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	if (he->branch_info)
+		return _hist_entry__dso_snprintf(he->branch_info->from.map,
+						 bf, size, width);
+	else
+		return repsep_snprintf(bf, size, "%-*.*s", width, width, "N/A");
+}
+
+static int64_t
+sort__dso_to_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	if (!left->branch_info || !right->branch_info)
+		return cmp_null(left->branch_info, right->branch_info);
+
+	return _sort__dso_cmp(left->branch_info->to.map,
+			      right->branch_info->to.map);
+}
+
+static int hist_entry__dso_to_snprintf(struct hist_entry *he, char *bf,
+				       size_t size, unsigned int width)
+{
+	if (he->branch_info)
+		return _hist_entry__dso_snprintf(he->branch_info->to.map,
+						 bf, size, width);
+	else
+		return repsep_snprintf(bf, size, "%-*.*s", width, width, "N/A");
+}
+
+static int64_t
+sort__sym_from_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct addr_map_symbol *from_l = &left->branch_info->from;
+	struct addr_map_symbol *from_r = &right->branch_info->from;
+
+	if (!left->branch_info || !right->branch_info)
+		return cmp_null(left->branch_info, right->branch_info);
+
+	from_l = &left->branch_info->from;
+	from_r = &right->branch_info->from;
+
+	if (!from_l->sym && !from_r->sym)
+		return _sort__addr_cmp(from_l->addr, from_r->addr);
+
+	return _sort__sym_cmp(from_l->sym, from_r->sym);
+}
+
+static int64_t
+sort__sym_to_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct addr_map_symbol *to_l, *to_r;
+
+	if (!left->branch_info || !right->branch_info)
+		return cmp_null(left->branch_info, right->branch_info);
+
+	to_l = &left->branch_info->to;
+	to_r = &right->branch_info->to;
+
+	if (!to_l->sym && !to_r->sym)
+		return _sort__addr_cmp(to_l->addr, to_r->addr);
+
+	return _sort__sym_cmp(to_l->sym, to_r->sym);
+}
+
+static int hist_entry__sym_from_snprintf(struct hist_entry *he, char *bf,
+					 size_t size, unsigned int width)
+{
+	if (he->branch_info) {
+		struct addr_map_symbol *from = &he->branch_info->from;
+
+		return _hist_entry__sym_snprintf(from->map, from->sym, from->addr,
+						 he->level, bf, size, width);
+	}
+
+	return repsep_snprintf(bf, size, "%-*.*s", width, width, "N/A");
+}
+
+static int hist_entry__sym_to_snprintf(struct hist_entry *he, char *bf,
+				       size_t size, unsigned int width)
+{
+	if (he->branch_info) {
+		struct addr_map_symbol *to = &he->branch_info->to;
+
+		return _hist_entry__sym_snprintf(to->map, to->sym, to->addr,
+						 he->level, bf, size, width);
+	}
+
+	return repsep_snprintf(bf, size, "%-*.*s", width, width, "N/A");
+}
+
+struct sort_entry sort_dso_from = {
+	.se_header	= "Source Shared Object",
+	.se_cmp		= sort__dso_from_cmp,
+	.se_snprintf	= hist_entry__dso_from_snprintf,
+	.se_width_idx	= HISTC_DSO_FROM,
+};
+
+struct sort_entry sort_dso_to = {
+	.se_header	= "Target Shared Object",
+	.se_cmp		= sort__dso_to_cmp,
+	.se_snprintf	= hist_entry__dso_to_snprintf,
+	.se_width_idx	= HISTC_DSO_TO,
+};
+
+struct sort_entry sort_sym_from = {
+	.se_header	= "Source Symbol",
+	.se_cmp		= sort__sym_from_cmp,
+	.se_snprintf	= hist_entry__sym_from_snprintf,
+	.se_width_idx	= HISTC_SYMBOL_FROM,
+};
+
+struct sort_entry sort_sym_to = {
+	.se_header	= "Target Symbol",
+	.se_cmp		= sort__sym_to_cmp,
+	.se_snprintf	= hist_entry__sym_to_snprintf,
+	.se_width_idx	= HISTC_SYMBOL_TO,
+};
+
+static int64_t
+sort__mispredict_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	unsigned char mp, p;
+
+	if (!left->branch_info || !right->branch_info)
+		return cmp_null(left->branch_info, right->branch_info);
+
+	mp = left->branch_info->flags.mispred != right->branch_info->flags.mispred;
+	p  = left->branch_info->flags.predicted != right->branch_info->flags.predicted;
+	return mp || p;
+}
+
+static int hist_entry__mispredict_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width){
+	static const char *out = "N/A";
+
+	if (he->branch_info) {
+		if (he->branch_info->flags.predicted)
+			out = "N";
+		else if (he->branch_info->flags.mispred)
+			out = "Y";
+	}
+
+	return repsep_snprintf(bf, size, "%-*.*s", width, width, out);
+}
+
+/* --sort daddr_sym */
+static int64_t
+sort__daddr_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	uint64_t l = 0, r = 0;
+
+	if (left->mem_info)
+		l = left->mem_info->daddr.addr;
+	if (right->mem_info)
+		r = right->mem_info->daddr.addr;
+
+	return (int64_t)(r - l);
+}
+
+static int hist_entry__daddr_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	uint64_t addr = 0;
+	struct map *map = NULL;
+	struct symbol *sym = NULL;
+
+	if (he->mem_info) {
+		addr = he->mem_info->daddr.addr;
+		map = he->mem_info->daddr.map;
+		sym = he->mem_info->daddr.sym;
+	}
+	return _hist_entry__sym_snprintf(map, sym, addr, he->level, bf, size,
+					 width);
+}
+
+static int64_t
+sort__dso_daddr_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	struct map *map_l = NULL;
+	struct map *map_r = NULL;
+
+	if (left->mem_info)
+		map_l = left->mem_info->daddr.map;
+	if (right->mem_info)
+		map_r = right->mem_info->daddr.map;
+
+	return _sort__dso_cmp(map_l, map_r);
+}
+
+static int hist_entry__dso_daddr_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	struct map *map = NULL;
+
+	if (he->mem_info)
+		map = he->mem_info->daddr.map;
+
+	return _hist_entry__dso_snprintf(map, bf, size, width);
+}
+
+static int64_t
+sort__locked_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	union perf_mem_data_src data_src_l;
+	union perf_mem_data_src data_src_r;
+
+	if (left->mem_info)
+		data_src_l = left->mem_info->data_src;
+	else
+		data_src_l.mem_lock = PERF_MEM_LOCK_NA;
+
+	if (right->mem_info)
+		data_src_r = right->mem_info->data_src;
+	else
+		data_src_r.mem_lock = PERF_MEM_LOCK_NA;
+
+	return (int64_t)(data_src_r.mem_lock - data_src_l.mem_lock);
+}
+
+static int hist_entry__locked_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	const char *out;
+	u64 mask = PERF_MEM_LOCK_NA;
+
+	if (he->mem_info)
+		mask = he->mem_info->data_src.mem_lock;
+
+	if (mask & PERF_MEM_LOCK_NA)
+		out = "N/A";
+	else if (mask & PERF_MEM_LOCK_LOCKED)
+		out = "Yes";
+	else
+		out = "No";
+
+	return repsep_snprintf(bf, size, "%-*s", width, out);
+}
+
+static int64_t
+sort__tlb_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	union perf_mem_data_src data_src_l;
+	union perf_mem_data_src data_src_r;
+
+	if (left->mem_info)
+		data_src_l = left->mem_info->data_src;
+	else
+		data_src_l.mem_dtlb = PERF_MEM_TLB_NA;
+
+	if (right->mem_info)
+		data_src_r = right->mem_info->data_src;
+	else
+		data_src_r.mem_dtlb = PERF_MEM_TLB_NA;
+
+	return (int64_t)(data_src_r.mem_dtlb - data_src_l.mem_dtlb);
+}
+
+static const char * const tlb_access[] = {
+	"N/A",
+	"HIT",
+	"MISS",
+	"L1",
+	"L2",
+	"Walker",
+	"Fault",
+};
+#define NUM_TLB_ACCESS (sizeof(tlb_access)/sizeof(const char *))
+
+static int hist_entry__tlb_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	char out[64];
+	size_t sz = sizeof(out) - 1; /* -1 for null termination */
+	size_t l = 0, i;
+	u64 m = PERF_MEM_TLB_NA;
+	u64 hit, miss;
+
+	out[0] = '\0';
+
+	if (he->mem_info)
+		m = he->mem_info->data_src.mem_dtlb;
+
+	hit = m & PERF_MEM_TLB_HIT;
+	miss = m & PERF_MEM_TLB_MISS;
+
+	/* already taken care of */
+	m &= ~(PERF_MEM_TLB_HIT|PERF_MEM_TLB_MISS);
+
+	for (i = 0; m && i < NUM_TLB_ACCESS; i++, m >>= 1) {
+		if (!(m & 0x1))
+			continue;
+		if (l) {
+			strcat(out, " or ");
+			l += 4;
+		}
+		strncat(out, tlb_access[i], sz - l);
+		l += strlen(tlb_access[i]);
+	}
+	if (*out == '\0')
+		strcpy(out, "N/A");
+	if (hit)
+		strncat(out, " hit", sz - l);
+	if (miss)
+		strncat(out, " miss", sz - l);
+
+	return repsep_snprintf(bf, size, "%-*s", width, out);
+}
+
+static int64_t
+sort__lvl_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	union perf_mem_data_src data_src_l;
+	union perf_mem_data_src data_src_r;
+
+	if (left->mem_info)
+		data_src_l = left->mem_info->data_src;
+	else
+		data_src_l.mem_lvl = PERF_MEM_LVL_NA;
+
+	if (right->mem_info)
+		data_src_r = right->mem_info->data_src;
+	else
+		data_src_r.mem_lvl = PERF_MEM_LVL_NA;
+
+	return (int64_t)(data_src_r.mem_lvl - data_src_l.mem_lvl);
+}
+
+static const char * const mem_lvl[] = {
+	"N/A",
+	"HIT",
+	"MISS",
+	"L1",
+	"LFB",
+	"L2",
+	"L3",
+	"Local RAM",
+	"Remote RAM (1 hop)",
+	"Remote RAM (2 hops)",
+	"Remote Cache (1 hop)",
+	"Remote Cache (2 hops)",
+	"I/O",
+	"Uncached",
+};
+#define NUM_MEM_LVL (sizeof(mem_lvl)/sizeof(const char *))
+
+static int hist_entry__lvl_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	char out[64];
+	size_t sz = sizeof(out) - 1; /* -1 for null termination */
+	size_t i, l = 0;
+	u64 m =  PERF_MEM_LVL_NA;
+	u64 hit, miss;
+
+	if (he->mem_info)
+		m  = he->mem_info->data_src.mem_lvl;
+
+	out[0] = '\0';
+
+	hit = m & PERF_MEM_LVL_HIT;
+	miss = m & PERF_MEM_LVL_MISS;
+
+	/* already taken care of */
+	m &= ~(PERF_MEM_LVL_HIT|PERF_MEM_LVL_MISS);
+
+	for (i = 0; m && i < NUM_MEM_LVL; i++, m >>= 1) {
+		if (!(m & 0x1))
+			continue;
+		if (l) {
+			strcat(out, " or ");
+			l += 4;
+		}
+		strncat(out, mem_lvl[i], sz - l);
+		l += strlen(mem_lvl[i]);
+	}
+	if (*out == '\0')
+		strcpy(out, "N/A");
+	if (hit)
+		strncat(out, " hit", sz - l);
+	if (miss)
+		strncat(out, " miss", sz - l);
+
+	return repsep_snprintf(bf, size, "%-*s", width, out);
+}
+
+static int64_t
+sort__snoop_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	union perf_mem_data_src data_src_l;
+	union perf_mem_data_src data_src_r;
+
+	if (left->mem_info)
+		data_src_l = left->mem_info->data_src;
+	else
+		data_src_l.mem_snoop = PERF_MEM_SNOOP_NA;
+
+	if (right->mem_info)
+		data_src_r = right->mem_info->data_src;
+	else
+		data_src_r.mem_snoop = PERF_MEM_SNOOP_NA;
+
+	return (int64_t)(data_src_r.mem_snoop - data_src_l.mem_snoop);
+}
+
+static const char * const snoop_access[] = {
+	"N/A",
+	"None",
+	"Miss",
+	"Hit",
+	"HitM",
+};
+#define NUM_SNOOP_ACCESS (sizeof(snoop_access)/sizeof(const char *))
+
+static int hist_entry__snoop_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	char out[64];
+	size_t sz = sizeof(out) - 1; /* -1 for null termination */
+	size_t i, l = 0;
+	u64 m = PERF_MEM_SNOOP_NA;
+
+	out[0] = '\0';
+
+	if (he->mem_info)
+		m = he->mem_info->data_src.mem_snoop;
+
+	for (i = 0; m && i < NUM_SNOOP_ACCESS; i++, m >>= 1) {
+		if (!(m & 0x1))
+			continue;
+		if (l) {
+			strcat(out, " or ");
+			l += 4;
+		}
+		strncat(out, snoop_access[i], sz - l);
+		l += strlen(snoop_access[i]);
+	}
+
+	if (*out == '\0')
+		strcpy(out, "N/A");
+
+	return repsep_snprintf(bf, size, "%-*s", width, out);
+}
+
+static inline  u64 cl_address(u64 address)
+{
+	/* return the cacheline of the address */
+	return (address & ~(cacheline_size - 1));
+}
+
+static int64_t
+sort__dcacheline_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	u64 l, r;
+	struct map *l_map, *r_map;
+
+	if (!left->mem_info)  return -1;
+	if (!right->mem_info) return 1;
+
+	/* group event types together */
+	if (left->cpumode > right->cpumode) return -1;
+	if (left->cpumode < right->cpumode) return 1;
+
+	l_map = left->mem_info->daddr.map;
+	r_map = right->mem_info->daddr.map;
+
+	/* if both are NULL, jump to sort on al_addr instead */
+	if (!l_map && !r_map)
+		goto addr;
+
+	if (!l_map) return -1;
+	if (!r_map) return 1;
+
+	if (l_map->maj > r_map->maj) return -1;
+	if (l_map->maj < r_map->maj) return 1;
+
+	if (l_map->min > r_map->min) return -1;
+	if (l_map->min < r_map->min) return 1;
+
+	if (l_map->ino > r_map->ino) return -1;
+	if (l_map->ino < r_map->ino) return 1;
+
+	if (l_map->ino_generation > r_map->ino_generation) return -1;
+	if (l_map->ino_generation < r_map->ino_generation) return 1;
+
+	/*
+	 * Addresses with no major/minor numbers are assumed to be
+	 * anonymous in userspace.  Sort those on pid then address.
+	 *
+	 * The kernel and non-zero major/minor mapped areas are
+	 * assumed to be unity mapped.  Sort those on address.
+	 */
+
+	if ((left->cpumode != PERF_RECORD_MISC_KERNEL) &&
+	    (!(l_map->flags & MAP_SHARED)) &&
+	    !l_map->maj && !l_map->min && !l_map->ino &&
+	    !l_map->ino_generation) {
+		/* userspace anonymous */
+
+		if (left->thread->pid_ > right->thread->pid_) return -1;
+		if (left->thread->pid_ < right->thread->pid_) return 1;
+	}
+
+addr:
+	/* al_addr does all the right addr - start + offset calculations */
+	l = cl_address(left->mem_info->daddr.al_addr);
+	r = cl_address(right->mem_info->daddr.al_addr);
+
+	if (l > r) return -1;
+	if (l < r) return 1;
+
+	return 0;
+}
+
+static int hist_entry__dcacheline_snprintf(struct hist_entry *he, char *bf,
+					  size_t size, unsigned int width)
+{
+
+	uint64_t addr = 0;
+	struct map *map = NULL;
+	struct symbol *sym = NULL;
+	char level = he->level;
+
+	if (he->mem_info) {
+		addr = cl_address(he->mem_info->daddr.al_addr);
+		map = he->mem_info->daddr.map;
+		sym = he->mem_info->daddr.sym;
+
+		/* print [s] for shared data mmaps */
+		if ((he->cpumode != PERF_RECORD_MISC_KERNEL) &&
+		     map && (map->type == MAP__VARIABLE) &&
+		    (map->flags & MAP_SHARED) &&
+		    (map->maj || map->min || map->ino ||
+		     map->ino_generation))
+			level = 's';
+		else if (!map)
+			level = 'X';
+	}
+	return _hist_entry__sym_snprintf(map, sym, addr, level, bf, size,
+					 width);
+}
+
+struct sort_entry sort_mispredict = {
+	.se_header	= "Branch Mispredicted",
+	.se_cmp		= sort__mispredict_cmp,
+	.se_snprintf	= hist_entry__mispredict_snprintf,
+	.se_width_idx	= HISTC_MISPREDICT,
+};
+
+static u64 he_weight(struct hist_entry *he)
+{
+	return he->stat.nr_events ? he->stat.weight / he->stat.nr_events : 0;
+}
+
+static int64_t
+sort__local_weight_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	return he_weight(left) - he_weight(right);
+}
+
+static int hist_entry__local_weight_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%-*llu", width, he_weight(he));
+}
+
+struct sort_entry sort_local_weight = {
+	.se_header	= "Local Weight",
+	.se_cmp		= sort__local_weight_cmp,
+	.se_snprintf	= hist_entry__local_weight_snprintf,
+	.se_width_idx	= HISTC_LOCAL_WEIGHT,
+};
+
+static int64_t
+sort__global_weight_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	return left->stat.weight - right->stat.weight;
+}
+
+static int hist_entry__global_weight_snprintf(struct hist_entry *he, char *bf,
+					      size_t size, unsigned int width)
+{
+	return repsep_snprintf(bf, size, "%-*llu", width, he->stat.weight);
+}
+
+struct sort_entry sort_global_weight = {
+	.se_header	= "Weight",
+	.se_cmp		= sort__global_weight_cmp,
+	.se_snprintf	= hist_entry__global_weight_snprintf,
+	.se_width_idx	= HISTC_GLOBAL_WEIGHT,
+};
+
+struct sort_entry sort_mem_daddr_sym = {
+	.se_header	= "Data Symbol",
+	.se_cmp		= sort__daddr_cmp,
+	.se_snprintf	= hist_entry__daddr_snprintf,
+	.se_width_idx	= HISTC_MEM_DADDR_SYMBOL,
+};
+
+struct sort_entry sort_mem_daddr_dso = {
+	.se_header	= "Data Object",
+	.se_cmp		= sort__dso_daddr_cmp,
+	.se_snprintf	= hist_entry__dso_daddr_snprintf,
+	.se_width_idx	= HISTC_MEM_DADDR_SYMBOL,
+};
+
+struct sort_entry sort_mem_locked = {
+	.se_header	= "Locked",
+	.se_cmp		= sort__locked_cmp,
+	.se_snprintf	= hist_entry__locked_snprintf,
+	.se_width_idx	= HISTC_MEM_LOCKED,
+};
+
+struct sort_entry sort_mem_tlb = {
+	.se_header	= "TLB access",
+	.se_cmp		= sort__tlb_cmp,
+	.se_snprintf	= hist_entry__tlb_snprintf,
+	.se_width_idx	= HISTC_MEM_TLB,
+};
+
+struct sort_entry sort_mem_lvl = {
+	.se_header	= "Memory access",
+	.se_cmp		= sort__lvl_cmp,
+	.se_snprintf	= hist_entry__lvl_snprintf,
+	.se_width_idx	= HISTC_MEM_LVL,
+};
+
+struct sort_entry sort_mem_snoop = {
+	.se_header	= "Snoop",
+	.se_cmp		= sort__snoop_cmp,
+	.se_snprintf	= hist_entry__snoop_snprintf,
+	.se_width_idx	= HISTC_MEM_SNOOP,
+};
+
+struct sort_entry sort_mem_dcacheline = {
+	.se_header	= "Data Cacheline",
+	.se_cmp		= sort__dcacheline_cmp,
+	.se_snprintf	= hist_entry__dcacheline_snprintf,
+	.se_width_idx	= HISTC_MEM_DCACHELINE,
+};
+
+static int64_t
+sort__abort_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	if (!left->branch_info || !right->branch_info)
+		return cmp_null(left->branch_info, right->branch_info);
+
+	return left->branch_info->flags.abort !=
+		right->branch_info->flags.abort;
+}
+
+static int hist_entry__abort_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	static const char *out = "N/A";
+
+	if (he->branch_info) {
+		if (he->branch_info->flags.abort)
+			out = "A";
+		else
+			out = ".";
+	}
+
+	return repsep_snprintf(bf, size, "%-*s", width, out);
+}
+
+struct sort_entry sort_abort = {
+	.se_header	= "Transaction abort",
+	.se_cmp		= sort__abort_cmp,
+	.se_snprintf	= hist_entry__abort_snprintf,
+	.se_width_idx	= HISTC_ABORT,
+};
+
+static int64_t
+sort__in_tx_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	if (!left->branch_info || !right->branch_info)
+		return cmp_null(left->branch_info, right->branch_info);
+
+	return left->branch_info->flags.in_tx !=
+		right->branch_info->flags.in_tx;
+}
+
+static int hist_entry__in_tx_snprintf(struct hist_entry *he, char *bf,
+				    size_t size, unsigned int width)
+{
+	static const char *out = "N/A";
+
+	if (he->branch_info) {
+		if (he->branch_info->flags.in_tx)
+			out = "T";
+		else
+			out = ".";
+	}
+
+	return repsep_snprintf(bf, size, "%-*s", width, out);
+}
+
+struct sort_entry sort_in_tx = {
+	.se_header	= "Branch in transaction",
+	.se_cmp		= sort__in_tx_cmp,
+	.se_snprintf	= hist_entry__in_tx_snprintf,
+	.se_width_idx	= HISTC_IN_TX,
+};
+
+static int64_t
+sort__transaction_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	return left->transaction - right->transaction;
+}
+
+static inline char *add_str(char *p, const char *str)
+{
+	strcpy(p, str);
+	return p + strlen(str);
+}
+
+static struct txbit {
+	unsigned flag;
+	const char *name;
+	int skip_for_len;
+} txbits[] = {
+	{ PERF_TXN_ELISION,        "EL ",        0 },
+	{ PERF_TXN_TRANSACTION,    "TX ",        1 },
+	{ PERF_TXN_SYNC,           "SYNC ",      1 },
+	{ PERF_TXN_ASYNC,          "ASYNC ",     0 },
+	{ PERF_TXN_RETRY,          "RETRY ",     0 },
+	{ PERF_TXN_CONFLICT,       "CON ",       0 },
+	{ PERF_TXN_CAPACITY_WRITE, "CAP-WRITE ", 1 },
+	{ PERF_TXN_CAPACITY_READ,  "CAP-READ ",  0 },
+	{ 0, NULL, 0 }
+};
+
+int hist_entry__transaction_len(void)
+{
+	int i;
+	int len = 0;
+
+	for (i = 0; txbits[i].name; i++) {
+		if (!txbits[i].skip_for_len)
+			len += strlen(txbits[i].name);
+	}
+	len += 4; /* :XX<space> */
+	return len;
+}
+
+static int hist_entry__transaction_snprintf(struct hist_entry *he, char *bf,
+					    size_t size, unsigned int width)
+{
+	u64 t = he->transaction;
+	char buf[128];
+	char *p = buf;
+	int i;
+
+	buf[0] = 0;
+	for (i = 0; txbits[i].name; i++)
+		if (txbits[i].flag & t)
+			p = add_str(p, txbits[i].name);
+	if (t && !(t & (PERF_TXN_SYNC|PERF_TXN_ASYNC)))
+		p = add_str(p, "NEITHER ");
+	if (t & PERF_TXN_ABORT_MASK) {
+		sprintf(p, ":%" PRIx64,
+			(t & PERF_TXN_ABORT_MASK) >>
+			PERF_TXN_ABORT_SHIFT);
+		p += strlen(p);
+	}
+
+	return repsep_snprintf(bf, size, "%-*s", width, buf);
+}
+
+struct sort_entry sort_transaction = {
+	.se_header	= "Transaction                ",
+	.se_cmp		= sort__transaction_cmp,
+	.se_snprintf	= hist_entry__transaction_snprintf,
+	.se_width_idx	= HISTC_TRANSACTION,
+};
+
+struct sort_dimension {
+	const char		*name;
+	struct sort_entry	*entry;
+	int			taken;
+};
+
+#define DIM(d, n, func) [d] = { .name = n, .entry = &(func) }
+
+static struct sort_dimension common_sort_dimensions[] = {
+	DIM(SORT_PID, "pid", sort_thread),
+	DIM(SORT_COMM, "comm", sort_comm),
+	DIM(SORT_DSO, "dso", sort_dso),
+	DIM(SORT_SYM, "symbol", sort_sym),
+	DIM(SORT_PARENT, "parent", sort_parent),
+	DIM(SORT_CPU, "cpu", sort_cpu),
+	DIM(SORT_SRCLINE, "srcline", sort_srcline),
+	DIM(SORT_LOCAL_WEIGHT, "local_weight", sort_local_weight),
+	DIM(SORT_GLOBAL_WEIGHT, "weight", sort_global_weight),
+	DIM(SORT_TRANSACTION, "transaction", sort_transaction),
+};
+
+#undef DIM
+
+#define DIM(d, n, func) [d - __SORT_BRANCH_STACK] = { .name = n, .entry = &(func) }
+
+static struct sort_dimension bstack_sort_dimensions[] = {
+	DIM(SORT_DSO_FROM, "dso_from", sort_dso_from),
+	DIM(SORT_DSO_TO, "dso_to", sort_dso_to),
+	DIM(SORT_SYM_FROM, "symbol_from", sort_sym_from),
+	DIM(SORT_SYM_TO, "symbol_to", sort_sym_to),
+	DIM(SORT_MISPREDICT, "mispredict", sort_mispredict),
+	DIM(SORT_IN_TX, "in_tx", sort_in_tx),
+	DIM(SORT_ABORT, "abort", sort_abort),
+};
+
+#undef DIM
+
+#define DIM(d, n, func) [d - __SORT_MEMORY_MODE] = { .name = n, .entry = &(func) }
+
+static struct sort_dimension memory_sort_dimensions[] = {
+	DIM(SORT_MEM_DADDR_SYMBOL, "symbol_daddr", sort_mem_daddr_sym),
+	DIM(SORT_MEM_DADDR_DSO, "dso_daddr", sort_mem_daddr_dso),
+	DIM(SORT_MEM_LOCKED, "locked", sort_mem_locked),
+	DIM(SORT_MEM_TLB, "tlb", sort_mem_tlb),
+	DIM(SORT_MEM_LVL, "mem", sort_mem_lvl),
+	DIM(SORT_MEM_SNOOP, "snoop", sort_mem_snoop),
+	DIM(SORT_MEM_DCACHELINE, "dcacheline", sort_mem_dcacheline),
+};
+
+#undef DIM
+
+struct hpp_dimension {
+	const char		*name;
+	struct perf_hpp_fmt	*fmt;
+	int			taken;
+};
+
+#define DIM(d, n) { .name = n, .fmt = &perf_hpp__format[d], }
+
+static struct hpp_dimension hpp_sort_dimensions[] = {
+	DIM(PERF_HPP__OVERHEAD, "overhead"),
+	DIM(PERF_HPP__OVERHEAD_SYS, "overhead_sys"),
+	DIM(PERF_HPP__OVERHEAD_US, "overhead_us"),
+	DIM(PERF_HPP__OVERHEAD_GUEST_SYS, "overhead_guest_sys"),
+	DIM(PERF_HPP__OVERHEAD_GUEST_US, "overhead_guest_us"),
+	DIM(PERF_HPP__OVERHEAD_ACC, "overhead_children"),
+	DIM(PERF_HPP__SAMPLES, "sample"),
+	DIM(PERF_HPP__PERIOD, "period"),
+};
+
+#undef DIM
+
+struct hpp_sort_entry {
+	struct perf_hpp_fmt hpp;
+	struct sort_entry *se;
+};
+
+bool perf_hpp__same_sort_entry(struct perf_hpp_fmt *a, struct perf_hpp_fmt *b)
+{
+	struct hpp_sort_entry *hse_a;
+	struct hpp_sort_entry *hse_b;
+
+	if (!perf_hpp__is_sort_entry(a) || !perf_hpp__is_sort_entry(b))
+		return false;
+
+	hse_a = container_of(a, struct hpp_sort_entry, hpp);
+	hse_b = container_of(b, struct hpp_sort_entry, hpp);
+
+	return hse_a->se == hse_b->se;
+}
+
+void perf_hpp__reset_sort_width(struct perf_hpp_fmt *fmt, struct hists *hists)
+{
+	struct hpp_sort_entry *hse;
+
+	if (!perf_hpp__is_sort_entry(fmt))
+		return;
+
+	hse = container_of(fmt, struct hpp_sort_entry, hpp);
+	hists__new_col_len(hists, hse->se->se_width_idx, strlen(fmt->name));
+}
+
+static int __sort__hpp_header(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+			      struct perf_evsel *evsel)
+{
+	struct hpp_sort_entry *hse;
+	size_t len = fmt->user_len;
+
+	hse = container_of(fmt, struct hpp_sort_entry, hpp);
+
+	if (!len)
+		len = hists__col_len(evsel__hists(evsel), hse->se->se_width_idx);
+
+	return scnprintf(hpp->buf, hpp->size, "%-*.*s", len, len, fmt->name);
+}
+
+static int __sort__hpp_width(struct perf_hpp_fmt *fmt,
+			     struct perf_hpp *hpp __maybe_unused,
+			     struct perf_evsel *evsel)
+{
+	struct hpp_sort_entry *hse;
+	size_t len = fmt->user_len;
+
+	hse = container_of(fmt, struct hpp_sort_entry, hpp);
+
+	if (!len)
+		len = hists__col_len(evsel__hists(evsel), hse->se->se_width_idx);
+
+	return len;
+}
+
+static int __sort__hpp_entry(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp,
+			     struct hist_entry *he)
+{
+	struct hpp_sort_entry *hse;
+	size_t len = fmt->user_len;
+
+	hse = container_of(fmt, struct hpp_sort_entry, hpp);
+
+	if (!len)
+		len = hists__col_len(he->hists, hse->se->se_width_idx);
+
+	return hse->se->se_snprintf(he, hpp->buf, hpp->size, len);
+}
+
+static int64_t __sort__hpp_cmp(struct perf_hpp_fmt *fmt,
+			       struct hist_entry *a, struct hist_entry *b)
+{
+	struct hpp_sort_entry *hse;
+
+	hse = container_of(fmt, struct hpp_sort_entry, hpp);
+	return hse->se->se_cmp(a, b);
+}
+
+static int64_t __sort__hpp_collapse(struct perf_hpp_fmt *fmt,
+				    struct hist_entry *a, struct hist_entry *b)
+{
+	struct hpp_sort_entry *hse;
+	int64_t (*collapse_fn)(struct hist_entry *, struct hist_entry *);
+
+	hse = container_of(fmt, struct hpp_sort_entry, hpp);
+	collapse_fn = hse->se->se_collapse ?: hse->se->se_cmp;
+	return collapse_fn(a, b);
+}
+
+static int64_t __sort__hpp_sort(struct perf_hpp_fmt *fmt,
+				struct hist_entry *a, struct hist_entry *b)
+{
+	struct hpp_sort_entry *hse;
+	int64_t (*sort_fn)(struct hist_entry *, struct hist_entry *);
+
+	hse = container_of(fmt, struct hpp_sort_entry, hpp);
+	sort_fn = hse->se->se_sort ?: hse->se->se_cmp;
+	return sort_fn(a, b);
+}
+
+static struct hpp_sort_entry *
+__sort_dimension__alloc_hpp(struct sort_dimension *sd)
+{
+	struct hpp_sort_entry *hse;
+
+	hse = malloc(sizeof(*hse));
+	if (hse == NULL) {
+		pr_err("Memory allocation failed\n");
+		return NULL;
+	}
+
+	hse->se = sd->entry;
+	hse->hpp.name = sd->entry->se_header;
+	hse->hpp.header = __sort__hpp_header;
+	hse->hpp.width = __sort__hpp_width;
+	hse->hpp.entry = __sort__hpp_entry;
+	hse->hpp.color = NULL;
+
+	hse->hpp.cmp = __sort__hpp_cmp;
+	hse->hpp.collapse = __sort__hpp_collapse;
+	hse->hpp.sort = __sort__hpp_sort;
+
+	INIT_LIST_HEAD(&hse->hpp.list);
+	INIT_LIST_HEAD(&hse->hpp.sort_list);
+	hse->hpp.elide = false;
+	hse->hpp.len = 0;
+	hse->hpp.user_len = 0;
+
+	return hse;
+}
+
+bool perf_hpp__is_sort_entry(struct perf_hpp_fmt *format)
+{
+	return format->header == __sort__hpp_header;
+}
+
+static int __sort_dimension__add_hpp_sort(struct sort_dimension *sd)
+{
+	struct hpp_sort_entry *hse = __sort_dimension__alloc_hpp(sd);
+
+	if (hse == NULL)
+		return -1;
+
+	perf_hpp__register_sort_field(&hse->hpp);
+	return 0;
+}
+
+static int __sort_dimension__add_hpp_output(struct sort_dimension *sd)
+{
+	struct hpp_sort_entry *hse = __sort_dimension__alloc_hpp(sd);
+
+	if (hse == NULL)
+		return -1;
+
+	perf_hpp__column_register(&hse->hpp);
+	return 0;
+}
+
+static int __sort_dimension__add(struct sort_dimension *sd)
+{
+	if (sd->taken)
+		return 0;
+
+	if (__sort_dimension__add_hpp_sort(sd) < 0)
+		return -1;
+
+	if (sd->entry->se_collapse)
+		sort__need_collapse = 1;
+
+	sd->taken = 1;
+
+	return 0;
+}
+
+static int __hpp_dimension__add(struct hpp_dimension *hd)
+{
+	if (!hd->taken) {
+		hd->taken = 1;
+
+		perf_hpp__register_sort_field(hd->fmt);
+	}
+	return 0;
+}
+
+static int __sort_dimension__add_output(struct sort_dimension *sd)
+{
+	if (sd->taken)
+		return 0;
+
+	if (__sort_dimension__add_hpp_output(sd) < 0)
+		return -1;
+
+	sd->taken = 1;
+	return 0;
+}
+
+static int __hpp_dimension__add_output(struct hpp_dimension *hd)
+{
+	if (!hd->taken) {
+		hd->taken = 1;
+
+		perf_hpp__column_register(hd->fmt);
+	}
+	return 0;
+}
+
+int sort_dimension__add(const char *tok)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(common_sort_dimensions); i++) {
+		struct sort_dimension *sd = &common_sort_dimensions[i];
+
+		if (strncasecmp(tok, sd->name, strlen(tok)))
+			continue;
+
+		if (sd->entry == &sort_parent) {
+			int ret = regcomp(&parent_regex, parent_pattern, REG_EXTENDED);
+			if (ret) {
+				char err[BUFSIZ];
+
+				regerror(ret, &parent_regex, err, sizeof(err));
+				pr_err("Invalid regex: %s\n%s", parent_pattern, err);
+				return -EINVAL;
+			}
+			sort__has_parent = 1;
+		} else if (sd->entry == &sort_sym) {
+			sort__has_sym = 1;
+			/*
+			 * perf diff displays the performance difference amongst
+			 * two or more perf.data files. Those files could come
+			 * from different binaries. So we should not compare
+			 * their ips, but the name of symbol.
+			 */
+			if (sort__mode == SORT_MODE__DIFF)
+				sd->entry->se_collapse = sort__sym_sort;
+
+		} else if (sd->entry == &sort_dso) {
+			sort__has_dso = 1;
+		}
+
+		return __sort_dimension__add(sd);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(hpp_sort_dimensions); i++) {
+		struct hpp_dimension *hd = &hpp_sort_dimensions[i];
+
+		if (strncasecmp(tok, hd->name, strlen(tok)))
+			continue;
+
+		return __hpp_dimension__add(hd);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(bstack_sort_dimensions); i++) {
+		struct sort_dimension *sd = &bstack_sort_dimensions[i];
+
+		if (strncasecmp(tok, sd->name, strlen(tok)))
+			continue;
+
+		if (sort__mode != SORT_MODE__BRANCH)
+			return -EINVAL;
+
+		if (sd->entry == &sort_sym_from || sd->entry == &sort_sym_to)
+			sort__has_sym = 1;
+
+		__sort_dimension__add(sd);
+		return 0;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(memory_sort_dimensions); i++) {
+		struct sort_dimension *sd = &memory_sort_dimensions[i];
+
+		if (strncasecmp(tok, sd->name, strlen(tok)))
+			continue;
+
+		if (sort__mode != SORT_MODE__MEMORY)
+			return -EINVAL;
+
+		if (sd->entry == &sort_mem_daddr_sym)
+			sort__has_sym = 1;
+
+		__sort_dimension__add(sd);
+		return 0;
+	}
+
+	return -ESRCH;
+}
+
+static const char *get_default_sort_order(void)
+{
+	const char *default_sort_orders[] = {
+		default_sort_order,
+		default_branch_sort_order,
+		default_mem_sort_order,
+		default_top_sort_order,
+		default_diff_sort_order,
+	};
+
+	BUG_ON(sort__mode >= ARRAY_SIZE(default_sort_orders));
+
+	return default_sort_orders[sort__mode];
+}
+
+static int setup_sort_order(void)
+{
+	char *new_sort_order;
+
+	/*
+	 * Append '+'-prefixed sort order to the default sort
+	 * order string.
+	 */
+	if (!sort_order || is_strict_order(sort_order))
+		return 0;
+
+	if (sort_order[1] == '\0') {
+		error("Invalid --sort key: `+'");
+		return -EINVAL;
+	}
+
+	/*
+	 * We allocate new sort_order string, but we never free it,
+	 * because it's checked over the rest of the code.
+	 */
+	if (asprintf(&new_sort_order, "%s,%s",
+		     get_default_sort_order(), sort_order + 1) < 0) {
+		error("Not enough memory to set up --sort");
+		return -ENOMEM;
+	}
+
+	sort_order = new_sort_order;
+	return 0;
+}
+
+static int __setup_sorting(void)
+{
+	char *tmp, *tok, *str;
+	const char *sort_keys;
+	int ret = 0;
+
+	ret = setup_sort_order();
+	if (ret)
+		return ret;
+
+	sort_keys = sort_order;
+	if (sort_keys == NULL) {
+		if (is_strict_order(field_order)) {
+			/*
+			 * If user specified field order but no sort order,
+			 * we'll honor it and not add default sort orders.
+			 */
+			return 0;
+		}
+
+		sort_keys = get_default_sort_order();
+	}
+
+	str = strdup(sort_keys);
+	if (str == NULL) {
+		error("Not enough memory to setup sort keys");
+		return -ENOMEM;
+	}
+
+	for (tok = strtok_r(str, ", ", &tmp);
+			tok; tok = strtok_r(NULL, ", ", &tmp)) {
+		ret = sort_dimension__add(tok);
+		if (ret == -EINVAL) {
+			error("Invalid --sort key: `%s'", tok);
+			break;
+		} else if (ret == -ESRCH) {
+			error("Unknown --sort key: `%s'", tok);
+			break;
+		}
+	}
+
+	free(str);
+	return ret;
+}
+
+void perf_hpp__set_elide(int idx, bool elide)
+{
+	struct perf_hpp_fmt *fmt;
+	struct hpp_sort_entry *hse;
+
+	perf_hpp__for_each_format(fmt) {
+		if (!perf_hpp__is_sort_entry(fmt))
+			continue;
+
+		hse = container_of(fmt, struct hpp_sort_entry, hpp);
+		if (hse->se->se_width_idx == idx) {
+			fmt->elide = elide;
+			break;
+		}
+	}
+}
+
+static bool __get_elide(struct strlist *list, const char *list_name, FILE *fp)
+{
+	if (list && strlist__nr_entries(list) == 1) {
+		if (fp != NULL)
+			fprintf(fp, "# %s: %s\n", list_name,
+				strlist__entry(list, 0)->s);
+		return true;
+	}
+	return false;
+}
+
+static bool get_elide(int idx, FILE *output)
+{
+	switch (idx) {
+	case HISTC_SYMBOL:
+		return __get_elide(symbol_conf.sym_list, "symbol", output);
+	case HISTC_DSO:
+		return __get_elide(symbol_conf.dso_list, "dso", output);
+	case HISTC_COMM:
+		return __get_elide(symbol_conf.comm_list, "comm", output);
+	default:
+		break;
+	}
+
+	if (sort__mode != SORT_MODE__BRANCH)
+		return false;
+
+	switch (idx) {
+	case HISTC_SYMBOL_FROM:
+		return __get_elide(symbol_conf.sym_from_list, "sym_from", output);
+	case HISTC_SYMBOL_TO:
+		return __get_elide(symbol_conf.sym_to_list, "sym_to", output);
+	case HISTC_DSO_FROM:
+		return __get_elide(symbol_conf.dso_from_list, "dso_from", output);
+	case HISTC_DSO_TO:
+		return __get_elide(symbol_conf.dso_to_list, "dso_to", output);
+	default:
+		break;
+	}
+
+	return false;
+}
+
+void sort__setup_elide(FILE *output)
+{
+	struct perf_hpp_fmt *fmt;
+	struct hpp_sort_entry *hse;
+
+	perf_hpp__for_each_format(fmt) {
+		if (!perf_hpp__is_sort_entry(fmt))
+			continue;
+
+		hse = container_of(fmt, struct hpp_sort_entry, hpp);
+		fmt->elide = get_elide(hse->se->se_width_idx, output);
+	}
+
+	/*
+	 * It makes no sense to elide all of sort entries.
+	 * Just revert them to show up again.
+	 */
+	perf_hpp__for_each_format(fmt) {
+		if (!perf_hpp__is_sort_entry(fmt))
+			continue;
+
+		if (!fmt->elide)
+			return;
+	}
+
+	perf_hpp__for_each_format(fmt) {
+		if (!perf_hpp__is_sort_entry(fmt))
+			continue;
+
+		fmt->elide = false;
+	}
+}
+
+static int output_field_add(char *tok)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(common_sort_dimensions); i++) {
+		struct sort_dimension *sd = &common_sort_dimensions[i];
+
+		if (strncasecmp(tok, sd->name, strlen(tok)))
+			continue;
+
+		return __sort_dimension__add_output(sd);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(hpp_sort_dimensions); i++) {
+		struct hpp_dimension *hd = &hpp_sort_dimensions[i];
+
+		if (strncasecmp(tok, hd->name, strlen(tok)))
+			continue;
+
+		return __hpp_dimension__add_output(hd);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(bstack_sort_dimensions); i++) {
+		struct sort_dimension *sd = &bstack_sort_dimensions[i];
+
+		if (strncasecmp(tok, sd->name, strlen(tok)))
+			continue;
+
+		return __sort_dimension__add_output(sd);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(memory_sort_dimensions); i++) {
+		struct sort_dimension *sd = &memory_sort_dimensions[i];
+
+		if (strncasecmp(tok, sd->name, strlen(tok)))
+			continue;
+
+		return __sort_dimension__add_output(sd);
+	}
+
+	return -ESRCH;
+}
+
+static void reset_dimensions(void)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(common_sort_dimensions); i++)
+		common_sort_dimensions[i].taken = 0;
+
+	for (i = 0; i < ARRAY_SIZE(hpp_sort_dimensions); i++)
+		hpp_sort_dimensions[i].taken = 0;
+
+	for (i = 0; i < ARRAY_SIZE(bstack_sort_dimensions); i++)
+		bstack_sort_dimensions[i].taken = 0;
+
+	for (i = 0; i < ARRAY_SIZE(memory_sort_dimensions); i++)
+		memory_sort_dimensions[i].taken = 0;
+}
+
+bool is_strict_order(const char *order)
+{
+	return order && (*order != '+');
+}
+
+static int __setup_output_field(void)
+{
+	char *tmp, *tok, *str, *strp;
+	int ret = -EINVAL;
+
+	if (field_order == NULL)
+		return 0;
+
+	reset_dimensions();
+
+	strp = str = strdup(field_order);
+	if (str == NULL) {
+		error("Not enough memory to setup output fields");
+		return -ENOMEM;
+	}
+
+	if (!is_strict_order(field_order))
+		strp++;
+
+	if (!strlen(strp)) {
+		error("Invalid --fields key: `+'");
+		goto out;
+	}
+
+	for (tok = strtok_r(strp, ", ", &tmp);
+			tok; tok = strtok_r(NULL, ", ", &tmp)) {
+		ret = output_field_add(tok);
+		if (ret == -EINVAL) {
+			error("Invalid --fields key: `%s'", tok);
+			break;
+		} else if (ret == -ESRCH) {
+			error("Unknown --fields key: `%s'", tok);
+			break;
+		}
+	}
+
+out:
+	free(str);
+	return ret;
+}
+
+int setup_sorting(void)
+{
+	int err;
+
+	err = __setup_sorting();
+	if (err < 0)
+		return err;
+
+	if (parent_pattern != default_parent_pattern) {
+		err = sort_dimension__add("parent");
+		if (err < 0)
+			return err;
+	}
+
+	reset_dimensions();
+
+	/*
+	 * perf diff doesn't use default hpp output fields.
+	 */
+	if (sort__mode != SORT_MODE__DIFF)
+		perf_hpp__init();
+
+	err = __setup_output_field();
+	if (err < 0)
+		return err;
+
+	/* copy sort keys to output fields */
+	perf_hpp__setup_output_field();
+	/* and then copy output fields to sort keys */
+	perf_hpp__append_sort_keys();
+
+	return 0;
+}
+
+void reset_output_field(void)
+{
+	sort__need_collapse = 0;
+	sort__has_parent = 0;
+	sort__has_sym = 0;
+	sort__has_dso = 0;
+
+	field_order = NULL;
+	sort_order = NULL;
+
+	reset_dimensions();
+	perf_hpp__reset_output_field();
+}
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
new file mode 100644
index 000000000..846036a92
--- /dev/null
+++ b/tools/perf/util/sort.h
@@ -0,0 +1,222 @@
+#ifndef __PERF_SORT_H
+#define __PERF_SORT_H
+#include "../builtin.h"
+
+#include "util.h"
+
+#include "color.h"
+#include <linux/list.h>
+#include "cache.h"
+#include <linux/rbtree.h>
+#include "symbol.h"
+#include "string.h"
+#include "callchain.h"
+#include "strlist.h"
+#include "values.h"
+
+#include "../perf.h"
+#include "debug.h"
+#include "header.h"
+
+#include "parse-options.h"
+#include "parse-events.h"
+#include "hist.h"
+#include "thread.h"
+
+extern regex_t parent_regex;
+extern const char *sort_order;
+extern const char *field_order;
+extern const char default_parent_pattern[];
+extern const char *parent_pattern;
+extern const char default_sort_order[];
+extern regex_t ignore_callees_regex;
+extern int have_ignore_callees;
+extern int sort__need_collapse;
+extern int sort__has_parent;
+extern int sort__has_sym;
+extern enum sort_mode sort__mode;
+extern struct sort_entry sort_comm;
+extern struct sort_entry sort_dso;
+extern struct sort_entry sort_sym;
+extern struct sort_entry sort_parent;
+extern struct sort_entry sort_dso_from;
+extern struct sort_entry sort_dso_to;
+extern struct sort_entry sort_sym_from;
+extern struct sort_entry sort_sym_to;
+extern enum sort_type sort__first_dimension;
+extern const char default_mem_sort_order[];
+
+struct he_stat {
+	u64			period;
+	u64			period_sys;
+	u64			period_us;
+	u64			period_guest_sys;
+	u64			period_guest_us;
+	u64			weight;
+	u32			nr_events;
+};
+
+struct hist_entry_diff {
+	bool	computed;
+
+	/* PERF_HPP__DELTA */
+	double	period_ratio_delta;
+
+	/* PERF_HPP__RATIO */
+	double	period_ratio;
+
+	/* HISTC_WEIGHTED_DIFF */
+	s64	wdiff;
+};
+
+/**
+ * struct hist_entry - histogram entry
+ *
+ * @row_offset - offset from the first callchain expanded to appear on screen
+ * @nr_rows - rows expanded in callchain, recalculated on folding/unfolding
+ */
+struct hist_entry {
+	struct rb_node		rb_node_in;
+	struct rb_node		rb_node;
+	union {
+		struct list_head node;
+		struct list_head head;
+	} pairs;
+	struct he_stat		stat;
+	struct he_stat		*stat_acc;
+	struct map_symbol	ms;
+	struct thread		*thread;
+	struct comm		*comm;
+	u64			ip;
+	u64			transaction;
+	s32			cpu;
+	u8			cpumode;
+
+	struct hist_entry_diff	diff;
+
+	/* We are added by hists__add_dummy_entry. */
+	bool			dummy;
+
+	/* XXX These two should move to some tree widget lib */
+	u16			row_offset;
+	u16			nr_rows;
+
+	bool			init_have_children;
+	char			level;
+	u8			filtered;
+	char			*srcline;
+	struct symbol		*parent;
+	unsigned long		position;
+	struct rb_root		sorted_chain;
+	struct branch_info	*branch_info;
+	struct hists		*hists;
+	struct mem_info		*mem_info;
+	struct callchain_root	callchain[0]; /* must be last member */
+};
+
+static inline bool hist_entry__has_pairs(struct hist_entry *he)
+{
+	return !list_empty(&he->pairs.node);
+}
+
+static inline struct hist_entry *hist_entry__next_pair(struct hist_entry *he)
+{
+	if (hist_entry__has_pairs(he))
+		return list_entry(he->pairs.node.next, struct hist_entry, pairs.node);
+	return NULL;
+}
+
+static inline void hist_entry__add_pair(struct hist_entry *pair,
+					struct hist_entry *he)
+{
+	list_add_tail(&pair->pairs.node, &he->pairs.head);
+}
+
+static inline float hist_entry__get_percent_limit(struct hist_entry *he)
+{
+	u64 period = he->stat.period;
+	u64 total_period = hists__total_period(he->hists);
+
+	if (unlikely(total_period == 0))
+		return 0;
+
+	if (symbol_conf.cumulate_callchain)
+		period = he->stat_acc->period;
+
+	return period * 100.0 / total_period;
+}
+
+
+enum sort_mode {
+	SORT_MODE__NORMAL,
+	SORT_MODE__BRANCH,
+	SORT_MODE__MEMORY,
+	SORT_MODE__TOP,
+	SORT_MODE__DIFF,
+};
+
+enum sort_type {
+	/* common sort keys */
+	SORT_PID,
+	SORT_COMM,
+	SORT_DSO,
+	SORT_SYM,
+	SORT_PARENT,
+	SORT_CPU,
+	SORT_SRCLINE,
+	SORT_LOCAL_WEIGHT,
+	SORT_GLOBAL_WEIGHT,
+	SORT_TRANSACTION,
+
+	/* branch stack specific sort keys */
+	__SORT_BRANCH_STACK,
+	SORT_DSO_FROM = __SORT_BRANCH_STACK,
+	SORT_DSO_TO,
+	SORT_SYM_FROM,
+	SORT_SYM_TO,
+	SORT_MISPREDICT,
+	SORT_ABORT,
+	SORT_IN_TX,
+
+	/* memory mode specific sort keys */
+	__SORT_MEMORY_MODE,
+	SORT_MEM_DADDR_SYMBOL = __SORT_MEMORY_MODE,
+	SORT_MEM_DADDR_DSO,
+	SORT_MEM_LOCKED,
+	SORT_MEM_TLB,
+	SORT_MEM_LVL,
+	SORT_MEM_SNOOP,
+	SORT_MEM_DCACHELINE,
+};
+
+/*
+ * configurable sorting bits
+ */
+
+struct sort_entry {
+	struct list_head list;
+
+	const char *se_header;
+
+	int64_t (*se_cmp)(struct hist_entry *, struct hist_entry *);
+	int64_t (*se_collapse)(struct hist_entry *, struct hist_entry *);
+	int64_t	(*se_sort)(struct hist_entry *, struct hist_entry *);
+	int	(*se_snprintf)(struct hist_entry *he, char *bf, size_t size,
+			       unsigned int width);
+	u8	se_width_idx;
+};
+
+extern struct sort_entry sort_thread;
+extern struct list_head hist_entry__sort_list;
+
+int setup_sorting(void);
+int setup_output_field(void);
+void reset_output_field(void);
+extern int sort_dimension__add(const char *);
+void sort__setup_elide(FILE *fp);
+void perf_hpp__set_elide(int idx, bool elide);
+
+int report_parse_ignore_callees_opt(const struct option *opt, const char *arg, int unset);
+
+bool is_strict_order(const char *order);
+#endif	/* __PERF_SORT_H */
diff --git a/tools/perf/util/srcline.c b/tools/perf/util/srcline.c
new file mode 100644
index 000000000..c93fb0c5b
--- /dev/null
+++ b/tools/perf/util/srcline.c
@@ -0,0 +1,308 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include <linux/kernel.h>
+
+#include "util/dso.h"
+#include "util/util.h"
+#include "util/debug.h"
+
+#include "symbol.h"
+
+#ifdef HAVE_LIBBFD_SUPPORT
+
+/*
+ * Implement addr2line using libbfd.
+ */
+#define PACKAGE "perf"
+#include <bfd.h>
+
+struct a2l_data {
+	const char 	*input;
+	u64	 	addr;
+
+	bool 		found;
+	const char 	*filename;
+	const char 	*funcname;
+	unsigned 	line;
+
+	bfd 		*abfd;
+	asymbol 	**syms;
+};
+
+static int bfd_error(const char *string)
+{
+	const char *errmsg;
+
+	errmsg = bfd_errmsg(bfd_get_error());
+	fflush(stdout);
+
+	if (string)
+		pr_debug("%s: %s\n", string, errmsg);
+	else
+		pr_debug("%s\n", errmsg);
+
+	return -1;
+}
+
+static int slurp_symtab(bfd *abfd, struct a2l_data *a2l)
+{
+	long storage;
+	long symcount;
+	asymbol **syms;
+	bfd_boolean dynamic = FALSE;
+
+	if ((bfd_get_file_flags(abfd) & HAS_SYMS) == 0)
+		return bfd_error(bfd_get_filename(abfd));
+
+	storage = bfd_get_symtab_upper_bound(abfd);
+	if (storage == 0L) {
+		storage = bfd_get_dynamic_symtab_upper_bound(abfd);
+		dynamic = TRUE;
+	}
+	if (storage < 0L)
+		return bfd_error(bfd_get_filename(abfd));
+
+	syms = malloc(storage);
+	if (dynamic)
+		symcount = bfd_canonicalize_dynamic_symtab(abfd, syms);
+	else
+		symcount = bfd_canonicalize_symtab(abfd, syms);
+
+	if (symcount < 0) {
+		free(syms);
+		return bfd_error(bfd_get_filename(abfd));
+	}
+
+	a2l->syms = syms;
+	return 0;
+}
+
+static void find_address_in_section(bfd *abfd, asection *section, void *data)
+{
+	bfd_vma pc, vma;
+	bfd_size_type size;
+	struct a2l_data *a2l = data;
+
+	if (a2l->found)
+		return;
+
+	if ((bfd_get_section_flags(abfd, section) & SEC_ALLOC) == 0)
+		return;
+
+	pc = a2l->addr;
+	vma = bfd_get_section_vma(abfd, section);
+	size = bfd_get_section_size(section);
+
+	if (pc < vma || pc >= vma + size)
+		return;
+
+	a2l->found = bfd_find_nearest_line(abfd, section, a2l->syms, pc - vma,
+					   &a2l->filename, &a2l->funcname,
+					   &a2l->line);
+}
+
+static struct a2l_data *addr2line_init(const char *path)
+{
+	bfd *abfd;
+	struct a2l_data *a2l = NULL;
+
+	abfd = bfd_openr(path, NULL);
+	if (abfd == NULL)
+		return NULL;
+
+	if (!bfd_check_format(abfd, bfd_object))
+		goto out;
+
+	a2l = zalloc(sizeof(*a2l));
+	if (a2l == NULL)
+		goto out;
+
+	a2l->abfd = abfd;
+	a2l->input = strdup(path);
+	if (a2l->input == NULL)
+		goto out;
+
+	if (slurp_symtab(abfd, a2l))
+		goto out;
+
+	return a2l;
+
+out:
+	if (a2l) {
+		zfree((char **)&a2l->input);
+		free(a2l);
+	}
+	bfd_close(abfd);
+	return NULL;
+}
+
+static void addr2line_cleanup(struct a2l_data *a2l)
+{
+	if (a2l->abfd)
+		bfd_close(a2l->abfd);
+	zfree((char **)&a2l->input);
+	zfree(&a2l->syms);
+	free(a2l);
+}
+
+static int addr2line(const char *dso_name, u64 addr,
+		     char **file, unsigned int *line, struct dso *dso)
+{
+	int ret = 0;
+	struct a2l_data *a2l = dso->a2l;
+
+	if (!a2l) {
+		dso->a2l = addr2line_init(dso_name);
+		a2l = dso->a2l;
+	}
+
+	if (a2l == NULL) {
+		pr_warning("addr2line_init failed for %s\n", dso_name);
+		return 0;
+	}
+
+	a2l->addr = addr;
+	a2l->found = false;
+
+	bfd_map_over_sections(a2l->abfd, find_address_in_section, a2l);
+
+	if (a2l->found && a2l->filename) {
+		*file = strdup(a2l->filename);
+		*line = a2l->line;
+
+		if (*file)
+			ret = 1;
+	}
+
+	return ret;
+}
+
+void dso__free_a2l(struct dso *dso)
+{
+	struct a2l_data *a2l = dso->a2l;
+
+	if (!a2l)
+		return;
+
+	addr2line_cleanup(a2l);
+
+	dso->a2l = NULL;
+}
+
+#else /* HAVE_LIBBFD_SUPPORT */
+
+static int addr2line(const char *dso_name, u64 addr,
+		     char **file, unsigned int *line_nr,
+		     struct dso *dso __maybe_unused)
+{
+	FILE *fp;
+	char cmd[PATH_MAX];
+	char *filename = NULL;
+	size_t len;
+	char *sep;
+	int ret = 0;
+
+	scnprintf(cmd, sizeof(cmd), "addr2line -e %s %016"PRIx64,
+		  dso_name, addr);
+
+	fp = popen(cmd, "r");
+	if (fp == NULL) {
+		pr_warning("popen failed for %s\n", dso_name);
+		return 0;
+	}
+
+	if (getline(&filename, &len, fp) < 0 || !len) {
+		pr_warning("addr2line has no output for %s\n", dso_name);
+		goto out;
+	}
+
+	sep = strchr(filename, '\n');
+	if (sep)
+		*sep = '\0';
+
+	if (!strcmp(filename, "??:0")) {
+		pr_debug("no debugging info in %s\n", dso_name);
+		free(filename);
+		goto out;
+	}
+
+	sep = strchr(filename, ':');
+	if (sep) {
+		*sep++ = '\0';
+		*file = filename;
+		*line_nr = strtoul(sep, NULL, 0);
+		ret = 1;
+	}
+out:
+	pclose(fp);
+	return ret;
+}
+
+void dso__free_a2l(struct dso *dso __maybe_unused)
+{
+}
+
+#endif /* HAVE_LIBBFD_SUPPORT */
+
+/*
+ * Number of addr2line failures (without success) before disabling it for that
+ * dso.
+ */
+#define A2L_FAIL_LIMIT 123
+
+char *get_srcline(struct dso *dso, u64 addr, struct symbol *sym,
+		  bool show_sym)
+{
+	char *file = NULL;
+	unsigned line = 0;
+	char *srcline;
+	const char *dso_name;
+
+	if (!dso->has_srcline)
+		goto out;
+
+	if (dso->symsrc_filename)
+		dso_name = dso->symsrc_filename;
+	else
+		dso_name = dso->long_name;
+
+	if (dso_name[0] == '[')
+		goto out;
+
+	if (!strncmp(dso_name, "/tmp/perf-", 10))
+		goto out;
+
+	if (!addr2line(dso_name, addr, &file, &line, dso))
+		goto out;
+
+	if (asprintf(&srcline, "%s:%u", basename(file), line) < 0) {
+		free(file);
+		goto out;
+	}
+
+	dso->a2l_fails = 0;
+
+	free(file);
+	return srcline;
+
+out:
+	if (dso->a2l_fails && ++dso->a2l_fails > A2L_FAIL_LIMIT) {
+		dso->has_srcline = 0;
+		dso__free_a2l(dso);
+	}
+	if (sym) {
+		if (asprintf(&srcline, "%s+%" PRIu64, show_sym ? sym->name : "",
+					addr - sym->start) < 0)
+			return SRCLINE_UNKNOWN;
+	} else if (asprintf(&srcline, "%s[%" PRIx64 "]", dso->short_name, addr) < 0)
+		return SRCLINE_UNKNOWN;
+	return srcline;
+}
+
+void free_srcline(char *srcline)
+{
+	if (srcline && strcmp(srcline, SRCLINE_UNKNOWN) != 0)
+		free(srcline);
+}
diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c
new file mode 100644
index 000000000..6506b3dfb
--- /dev/null
+++ b/tools/perf/util/stat.c
@@ -0,0 +1,63 @@
+#include <math.h>
+
+#include "stat.h"
+
+void update_stats(struct stats *stats, u64 val)
+{
+	double delta;
+
+	stats->n++;
+	delta = val - stats->mean;
+	stats->mean += delta / stats->n;
+	stats->M2 += delta*(val - stats->mean);
+
+	if (val > stats->max)
+		stats->max = val;
+
+	if (val < stats->min)
+		stats->min = val;
+}
+
+double avg_stats(struct stats *stats)
+{
+	return stats->mean;
+}
+
+/*
+ * http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance
+ *
+ *       (\Sum n_i^2) - ((\Sum n_i)^2)/n
+ * s^2 = -------------------------------
+ *                  n - 1
+ *
+ * http://en.wikipedia.org/wiki/Stddev
+ *
+ * The std dev of the mean is related to the std dev by:
+ *
+ *             s
+ * s_mean = -------
+ *          sqrt(n)
+ *
+ */
+double stddev_stats(struct stats *stats)
+{
+	double variance, variance_mean;
+
+	if (stats->n < 2)
+		return 0.0;
+
+	variance = stats->M2 / (stats->n - 1);
+	variance_mean = variance / stats->n;
+
+	return sqrt(variance_mean);
+}
+
+double rel_stddev_stats(double stddev, double avg)
+{
+	double pct = 0.0;
+
+	if (avg)
+		pct = 100.0 * stddev/avg;
+
+	return pct;
+}
diff --git a/tools/perf/util/stat.h b/tools/perf/util/stat.h
new file mode 100644
index 000000000..5667fc3e3
--- /dev/null
+++ b/tools/perf/util/stat.h
@@ -0,0 +1,25 @@
+#ifndef __PERF_STATS_H
+#define __PERF_STATS_H
+
+#include <linux/types.h>
+
+struct stats
+{
+	double n, mean, M2;
+	u64 max, min;
+};
+
+void update_stats(struct stats *stats, u64 val);
+double avg_stats(struct stats *stats);
+double stddev_stats(struct stats *stats);
+double rel_stddev_stats(double stddev, double avg);
+
+static inline void init_stats(struct stats *stats)
+{
+	stats->n    = 0.0;
+	stats->mean = 0.0;
+	stats->M2   = 0.0;
+	stats->min  = (u64) -1;
+	stats->max  = 0;
+}
+#endif
diff --git a/tools/perf/util/strbuf.c b/tools/perf/util/strbuf.c
new file mode 100644
index 000000000..4abe23550
--- /dev/null
+++ b/tools/perf/util/strbuf.c
@@ -0,0 +1,134 @@
+#include "cache.h"
+#include <linux/kernel.h>
+
+int prefixcmp(const char *str, const char *prefix)
+{
+	for (; ; str++, prefix++)
+		if (!*prefix)
+			return 0;
+		else if (*str != *prefix)
+			return (unsigned char)*prefix - (unsigned char)*str;
+}
+
+/*
+ * Used as the default ->buf value, so that people can always assume
+ * buf is non NULL and ->buf is NUL terminated even for a freshly
+ * initialized strbuf.
+ */
+char strbuf_slopbuf[1];
+
+void strbuf_init(struct strbuf *sb, ssize_t hint)
+{
+	sb->alloc = sb->len = 0;
+	sb->buf = strbuf_slopbuf;
+	if (hint)
+		strbuf_grow(sb, hint);
+}
+
+void strbuf_release(struct strbuf *sb)
+{
+	if (sb->alloc) {
+		zfree(&sb->buf);
+		strbuf_init(sb, 0);
+	}
+}
+
+char *strbuf_detach(struct strbuf *sb, size_t *sz)
+{
+	char *res = sb->alloc ? sb->buf : NULL;
+	if (sz)
+		*sz = sb->len;
+	strbuf_init(sb, 0);
+	return res;
+}
+
+void strbuf_grow(struct strbuf *sb, size_t extra)
+{
+	if (sb->len + extra + 1 <= sb->len)
+		die("you want to use way too much memory");
+	if (!sb->alloc)
+		sb->buf = NULL;
+	ALLOC_GROW(sb->buf, sb->len + extra + 1, sb->alloc);
+}
+
+static void strbuf_splice(struct strbuf *sb, size_t pos, size_t len,
+				   const void *data, size_t dlen)
+{
+	if (pos + len < pos)
+		die("you want to use way too much memory");
+	if (pos > sb->len)
+		die("`pos' is too far after the end of the buffer");
+	if (pos + len > sb->len)
+		die("`pos + len' is too far after the end of the buffer");
+
+	if (dlen >= len)
+		strbuf_grow(sb, dlen - len);
+	memmove(sb->buf + pos + dlen,
+			sb->buf + pos + len,
+			sb->len - pos - len);
+	memcpy(sb->buf + pos, data, dlen);
+	strbuf_setlen(sb, sb->len + dlen - len);
+}
+
+void strbuf_remove(struct strbuf *sb, size_t pos, size_t len)
+{
+	strbuf_splice(sb, pos, len, NULL, 0);
+}
+
+void strbuf_add(struct strbuf *sb, const void *data, size_t len)
+{
+	strbuf_grow(sb, len);
+	memcpy(sb->buf + sb->len, data, len);
+	strbuf_setlen(sb, sb->len + len);
+}
+
+void strbuf_addf(struct strbuf *sb, const char *fmt, ...)
+{
+	int len;
+	va_list ap;
+
+	if (!strbuf_avail(sb))
+		strbuf_grow(sb, 64);
+	va_start(ap, fmt);
+	len = vsnprintf(sb->buf + sb->len, sb->alloc - sb->len, fmt, ap);
+	va_end(ap);
+	if (len < 0)
+		die("your vsnprintf is broken");
+	if (len > strbuf_avail(sb)) {
+		strbuf_grow(sb, len);
+		va_start(ap, fmt);
+		len = vsnprintf(sb->buf + sb->len, sb->alloc - sb->len, fmt, ap);
+		va_end(ap);
+		if (len > strbuf_avail(sb)) {
+			die("this should not happen, your vsnprintf is broken");
+		}
+	}
+	strbuf_setlen(sb, sb->len + len);
+}
+
+ssize_t strbuf_read(struct strbuf *sb, int fd, ssize_t hint)
+{
+	size_t oldlen = sb->len;
+	size_t oldalloc = sb->alloc;
+
+	strbuf_grow(sb, hint ? hint : 8192);
+	for (;;) {
+		ssize_t cnt;
+
+		cnt = read(fd, sb->buf + sb->len, sb->alloc - sb->len - 1);
+		if (cnt < 0) {
+			if (oldalloc == 0)
+				strbuf_release(sb);
+			else
+				strbuf_setlen(sb, oldlen);
+			return -1;
+		}
+		if (!cnt)
+			break;
+		sb->len += cnt;
+		strbuf_grow(sb, 8192);
+	}
+
+	sb->buf[sb->len] = '\0';
+	return sb->len - oldlen;
+}
diff --git a/tools/perf/util/strbuf.h b/tools/perf/util/strbuf.h
new file mode 100644
index 000000000..436ac319f
--- /dev/null
+++ b/tools/perf/util/strbuf.h
@@ -0,0 +1,92 @@
+#ifndef __PERF_STRBUF_H
+#define __PERF_STRBUF_H
+
+/*
+ * Strbuf's can be use in many ways: as a byte array, or to store arbitrary
+ * long, overflow safe strings.
+ *
+ * Strbufs has some invariants that are very important to keep in mind:
+ *
+ * 1. the ->buf member is always malloc-ed, hence strbuf's can be used to
+ *    build complex strings/buffers whose final size isn't easily known.
+ *
+ *    It is NOT legal to copy the ->buf pointer away.
+ *    `strbuf_detach' is the operation that detachs a buffer from its shell
+ *    while keeping the shell valid wrt its invariants.
+ *
+ * 2. the ->buf member is a byte array that has at least ->len + 1 bytes
+ *    allocated. The extra byte is used to store a '\0', allowing the ->buf
+ *    member to be a valid C-string. Every strbuf function ensure this
+ *    invariant is preserved.
+ *
+ *    Note that it is OK to "play" with the buffer directly if you work it
+ *    that way:
+ *
+ *    strbuf_grow(sb, SOME_SIZE);
+ *       ... Here, the memory array starting at sb->buf, and of length
+ *       ... strbuf_avail(sb) is all yours, and you are sure that
+ *       ... strbuf_avail(sb) is at least SOME_SIZE.
+ *    strbuf_setlen(sb, sb->len + SOME_OTHER_SIZE);
+ *
+ *    Of course, SOME_OTHER_SIZE must be smaller or equal to strbuf_avail(sb).
+ *
+ *    Doing so is safe, though if it has to be done in many places, adding the
+ *    missing API to the strbuf module is the way to go.
+ *
+ *    XXX: do _not_ assume that the area that is yours is of size ->alloc - 1
+ *         even if it's true in the current implementation. Alloc is somehow a
+ *         "private" member that should not be messed with.
+ */
+
+#include <assert.h>
+
+extern char strbuf_slopbuf[];
+struct strbuf {
+	size_t alloc;
+	size_t len;
+	char *buf;
+};
+
+#define STRBUF_INIT  { 0, 0, strbuf_slopbuf }
+
+/*----- strbuf life cycle -----*/
+extern void strbuf_init(struct strbuf *buf, ssize_t hint);
+extern void strbuf_release(struct strbuf *);
+extern char *strbuf_detach(struct strbuf *, size_t *);
+
+/*----- strbuf size related -----*/
+static inline ssize_t strbuf_avail(const struct strbuf *sb) {
+	return sb->alloc ? sb->alloc - sb->len - 1 : 0;
+}
+
+extern void strbuf_grow(struct strbuf *, size_t);
+
+static inline void strbuf_setlen(struct strbuf *sb, size_t len) {
+	if (!sb->alloc)
+		strbuf_grow(sb, 0);
+	assert(len < sb->alloc);
+	sb->len = len;
+	sb->buf[len] = '\0';
+}
+
+/*----- add data in your buffer -----*/
+static inline void strbuf_addch(struct strbuf *sb, int c) {
+	strbuf_grow(sb, 1);
+	sb->buf[sb->len++] = c;
+	sb->buf[sb->len] = '\0';
+}
+
+extern void strbuf_remove(struct strbuf *, size_t pos, size_t len);
+
+extern void strbuf_add(struct strbuf *, const void *, size_t);
+static inline void strbuf_addstr(struct strbuf *sb, const char *s) {
+	strbuf_add(sb, s, strlen(s));
+}
+
+__attribute__((format(printf,2,3)))
+extern void strbuf_addf(struct strbuf *sb, const char *fmt, ...);
+
+/* XXX: if read fails, any partial read is undone */
+extern ssize_t strbuf_read(struct strbuf *, int fd, ssize_t hint);
+
+#endif /* __PERF_STRBUF_H */
diff --git a/tools/perf/util/strfilter.c b/tools/perf/util/strfilter.c
new file mode 100644
index 000000000..79a757a2a
--- /dev/null
+++ b/tools/perf/util/strfilter.c
@@ -0,0 +1,199 @@
+#include "util.h"
+#include "string.h"
+#include "strfilter.h"
+
+/* Operators */
+static const char *OP_and	= "&";	/* Logical AND */
+static const char *OP_or	= "|";	/* Logical OR */
+static const char *OP_not	= "!";	/* Logical NOT */
+
+#define is_operator(c)	((c) == '|' || (c) == '&' || (c) == '!')
+#define is_separator(c)	(is_operator(c) || (c) == '(' || (c) == ')')
+
+static void strfilter_node__delete(struct strfilter_node *node)
+{
+	if (node) {
+		if (node->p && !is_operator(*node->p))
+			zfree((char **)&node->p);
+		strfilter_node__delete(node->l);
+		strfilter_node__delete(node->r);
+		free(node);
+	}
+}
+
+void strfilter__delete(struct strfilter *filter)
+{
+	if (filter) {
+		strfilter_node__delete(filter->root);
+		free(filter);
+	}
+}
+
+static const char *get_token(const char *s, const char **e)
+{
+	const char *p;
+
+	while (isspace(*s))	/* Skip spaces */
+		s++;
+
+	if (*s == '\0') {
+		p = s;
+		goto end;
+	}
+
+	p = s + 1;
+	if (!is_separator(*s)) {
+		/* End search */
+retry:
+		while (*p && !is_separator(*p) && !isspace(*p))
+			p++;
+		/* Escape and special case: '!' is also used in glob pattern */
+		if (*(p - 1) == '\\' || (*p == '!' && *(p - 1) == '[')) {
+			p++;
+			goto retry;
+		}
+	}
+end:
+	*e = p;
+	return s;
+}
+
+static struct strfilter_node *strfilter_node__alloc(const char *op,
+						    struct strfilter_node *l,
+						    struct strfilter_node *r)
+{
+	struct strfilter_node *node = zalloc(sizeof(*node));
+
+	if (node) {
+		node->p = op;
+		node->l = l;
+		node->r = r;
+	}
+
+	return node;
+}
+
+static struct strfilter_node *strfilter_node__new(const char *s,
+						  const char **ep)
+{
+	struct strfilter_node root, *cur, *last_op;
+	const char *e;
+
+	if (!s)
+		return NULL;
+
+	memset(&root, 0, sizeof(root));
+	last_op = cur = &root;
+
+	s = get_token(s, &e);
+	while (*s != '\0' && *s != ')') {
+		switch (*s) {
+		case '&':	/* Exchg last OP->r with AND */
+			if (!cur->r || !last_op->r)
+				goto error;
+			cur = strfilter_node__alloc(OP_and, last_op->r, NULL);
+			if (!cur)
+				goto nomem;
+			last_op->r = cur;
+			last_op = cur;
+			break;
+		case '|':	/* Exchg the root with OR */
+			if (!cur->r || !root.r)
+				goto error;
+			cur = strfilter_node__alloc(OP_or, root.r, NULL);
+			if (!cur)
+				goto nomem;
+			root.r = cur;
+			last_op = cur;
+			break;
+		case '!':	/* Add NOT as a leaf node */
+			if (cur->r)
+				goto error;
+			cur->r = strfilter_node__alloc(OP_not, NULL, NULL);
+			if (!cur->r)
+				goto nomem;
+			cur = cur->r;
+			break;
+		case '(':	/* Recursively parses inside the parenthesis */
+			if (cur->r)
+				goto error;
+			cur->r = strfilter_node__new(s + 1, &s);
+			if (!s)
+				goto nomem;
+			if (!cur->r || *s != ')')
+				goto error;
+			e = s + 1;
+			break;
+		default:
+			if (cur->r)
+				goto error;
+			cur->r = strfilter_node__alloc(NULL, NULL, NULL);
+			if (!cur->r)
+				goto nomem;
+			cur->r->p = strndup(s, e - s);
+			if (!cur->r->p)
+				goto nomem;
+		}
+		s = get_token(e, &e);
+	}
+	if (!cur->r)
+		goto error;
+	*ep = s;
+	return root.r;
+nomem:
+	s = NULL;
+error:
+	*ep = s;
+	strfilter_node__delete(root.r);
+	return NULL;
+}
+
+/*
+ * Parse filter rule and return new strfilter.
+ * Return NULL if fail, and *ep == NULL if memory allocation failed.
+ */
+struct strfilter *strfilter__new(const char *rules, const char **err)
+{
+	struct strfilter *filter = zalloc(sizeof(*filter));
+	const char *ep = NULL;
+
+	if (filter)
+		filter->root = strfilter_node__new(rules, &ep);
+
+	if (!filter || !filter->root || *ep != '\0') {
+		if (err)
+			*err = ep;
+		strfilter__delete(filter);
+		filter = NULL;
+	}
+
+	return filter;
+}
+
+static bool strfilter_node__compare(struct strfilter_node *node,
+				    const char *str)
+{
+	if (!node || !node->p)
+		return false;
+
+	switch (*node->p) {
+	case '|':	/* OR */
+		return strfilter_node__compare(node->l, str) ||
+			strfilter_node__compare(node->r, str);
+	case '&':	/* AND */
+		return strfilter_node__compare(node->l, str) &&
+			strfilter_node__compare(node->r, str);
+	case '!':	/* NOT */
+		return !strfilter_node__compare(node->r, str);
+	default:
+		return strglobmatch(str, node->p);
+	}
+}
+
+/* Return true if STR matches the filter rules */
+bool strfilter__compare(struct strfilter *filter, const char *str)
+{
+	if (!filter)
+		return false;
+	return strfilter_node__compare(filter->root, str);
+}
diff --git a/tools/perf/util/strfilter.h b/tools/perf/util/strfilter.h
new file mode 100644
index 000000000..fe611f3c9
--- /dev/null
+++ b/tools/perf/util/strfilter.h
@@ -0,0 +1,48 @@
+#ifndef __PERF_STRFILTER_H
+#define __PERF_STRFILTER_H
+/* General purpose glob matching filter */
+
+#include <linux/list.h>
+#include <stdbool.h>
+
+/* A node of string filter */
+struct strfilter_node {
+	struct strfilter_node *l;	/* Tree left branche (for &,|) */
+	struct strfilter_node *r;	/* Tree right branche (for !,&,|) */
+	const char *p;		/* Operator or rule */
+};
+
+/* String filter */
+struct strfilter {
+	struct strfilter_node *root;
+};
+
+/**
+ * strfilter__new - Create a new string filter
+ * @rules: Filter rule, which is a combination of glob expressions.
+ * @err: Pointer which points an error detected on @rules
+ *
+ * Parse @rules and return new strfilter. Return NULL if an error detected.
+ * In that case, *@err will indicate where it is detected, and *@err is NULL
+ * if a memory allocation is failed.
+ */
+struct strfilter *strfilter__new(const char *rules, const char **err);
+
+/**
+ * strfilter__compare - compare given string and a string filter
+ * @filter: String filter
+ * @str: target string
+ *
+ * Compare @str and @filter. Return true if the str match the rule
+ */
+bool strfilter__compare(struct strfilter *filter, const char *str);
+
+/**
+ * strfilter__delete - delete a string filter
+ * @filter: String filter to delete
+ *
+ * Delete @filter.
+ */
+void strfilter__delete(struct strfilter *filter);
+
+#endif
diff --git a/tools/perf/util/string.c b/tools/perf/util/string.c
new file mode 100644
index 000000000..6afd6106c
--- /dev/null
+++ b/tools/perf/util/string.c
@@ -0,0 +1,359 @@
+#include "util.h"
+#include "linux/string.h"
+
+#define K 1024LL
+/*
+ * perf_atoll()
+ * Parse (\d+)(b|B|kb|KB|mb|MB|gb|GB|tb|TB) (e.g. "256MB")
+ * and return its numeric value
+ */
+s64 perf_atoll(const char *str)
+{
+	s64 length;
+	char *p;
+	char c;
+
+	if (!isdigit(str[0]))
+		goto out_err;
+
+	length = strtoll(str, &p, 10);
+	switch (c = *p++) {
+		case 'b': case 'B':
+			if (*p)
+				goto out_err;
+		case '\0':
+			return length;
+		default:
+			goto out_err;
+		/* two-letter suffices */
+		case 'k': case 'K':
+			length <<= 10;
+			break;
+		case 'm': case 'M':
+			length <<= 20;
+			break;
+		case 'g': case 'G':
+			length <<= 30;
+			break;
+		case 't': case 'T':
+			length <<= 40;
+			break;
+	}
+	/* we want the cases to match */
+	if (islower(c)) {
+		if (strcmp(p, "b") != 0)
+			goto out_err;
+	} else {
+		if (strcmp(p, "B") != 0)
+			goto out_err;
+	}
+	return length;
+
+out_err:
+	return -1;
+}
+
+/*
+ * Helper function for splitting a string into an argv-like array.
+ * originally copied from lib/argv_split.c
+ */
+static const char *skip_sep(const char *cp)
+{
+	while (*cp && isspace(*cp))
+		cp++;
+
+	return cp;
+}
+
+static const char *skip_arg(const char *cp)
+{
+	while (*cp && !isspace(*cp))
+		cp++;
+
+	return cp;
+}
+
+static int count_argc(const char *str)
+{
+	int count = 0;
+
+	while (*str) {
+		str = skip_sep(str);
+		if (*str) {
+			count++;
+			str = skip_arg(str);
+		}
+	}
+
+	return count;
+}
+
+/**
+ * argv_free - free an argv
+ * @argv - the argument vector to be freed
+ *
+ * Frees an argv and the strings it points to.
+ */
+void argv_free(char **argv)
+{
+	char **p;
+	for (p = argv; *p; p++)
+		zfree(p);
+
+	free(argv);
+}
+
+/**
+ * argv_split - split a string at whitespace, returning an argv
+ * @str: the string to be split
+ * @argcp: returned argument count
+ *
+ * Returns an array of pointers to strings which are split out from
+ * @str.  This is performed by strictly splitting on white-space; no
+ * quote processing is performed.  Multiple whitespace characters are
+ * considered to be a single argument separator.  The returned array
+ * is always NULL-terminated.  Returns NULL on memory allocation
+ * failure.
+ */
+char **argv_split(const char *str, int *argcp)
+{
+	int argc = count_argc(str);
+	char **argv = zalloc(sizeof(*argv) * (argc+1));
+	char **argvp;
+
+	if (argv == NULL)
+		goto out;
+
+	if (argcp)
+		*argcp = argc;
+
+	argvp = argv;
+
+	while (*str) {
+		str = skip_sep(str);
+
+		if (*str) {
+			const char *p = str;
+			char *t;
+
+			str = skip_arg(str);
+
+			t = strndup(p, str-p);
+			if (t == NULL)
+				goto fail;
+			*argvp++ = t;
+		}
+	}
+	*argvp = NULL;
+
+out:
+	return argv;
+
+fail:
+	argv_free(argv);
+	return NULL;
+}
+
+/* Character class matching */
+static bool __match_charclass(const char *pat, char c, const char **npat)
+{
+	bool complement = false, ret = true;
+
+	if (*pat == '!') {
+		complement = true;
+		pat++;
+	}
+	if (*pat++ == c)	/* First character is special */
+		goto end;
+
+	while (*pat && *pat != ']') {	/* Matching */
+		if (*pat == '-' && *(pat + 1) != ']') {	/* Range */
+			if (*(pat - 1) <= c && c <= *(pat + 1))
+				goto end;
+			if (*(pat - 1) > *(pat + 1))
+				goto error;
+			pat += 2;
+		} else if (*pat++ == c)
+			goto end;
+	}
+	if (!*pat)
+		goto error;
+	ret = false;
+
+end:
+	while (*pat && *pat != ']')	/* Searching closing */
+		pat++;
+	if (!*pat)
+		goto error;
+	*npat = pat + 1;
+	return complement ? !ret : ret;
+
+error:
+	return false;
+}
+
+/* Glob/lazy pattern matching */
+static bool __match_glob(const char *str, const char *pat, bool ignore_space)
+{
+	while (*str && *pat && *pat != '*') {
+		if (ignore_space) {
+			/* Ignore spaces for lazy matching */
+			if (isspace(*str)) {
+				str++;
+				continue;
+			}
+			if (isspace(*pat)) {
+				pat++;
+				continue;
+			}
+		}
+		if (*pat == '?') {	/* Matches any single character */
+			str++;
+			pat++;
+			continue;
+		} else if (*pat == '[')	/* Character classes/Ranges */
+			if (__match_charclass(pat + 1, *str, &pat)) {
+				str++;
+				continue;
+			} else
+				return false;
+		else if (*pat == '\\') /* Escaped char match as normal char */
+			pat++;
+		if (*str++ != *pat++)
+			return false;
+	}
+	/* Check wild card */
+	if (*pat == '*') {
+		while (*pat == '*')
+			pat++;
+		if (!*pat)	/* Tail wild card matches all */
+			return true;
+		while (*str)
+			if (__match_glob(str++, pat, ignore_space))
+				return true;
+	}
+	return !*str && !*pat;
+}
+
+/**
+ * strglobmatch - glob expression pattern matching
+ * @str: the target string to match
+ * @pat: the pattern string to match
+ *
+ * This returns true if the @str matches @pat. @pat can includes wildcards
+ * ('*','?') and character classes ([CHARS], complementation and ranges are
+ * also supported). Also, this supports escape character ('\') to use special
+ * characters as normal character.
+ *
+ * Note: if @pat syntax is broken, this always returns false.
+ */
+bool strglobmatch(const char *str, const char *pat)
+{
+	return __match_glob(str, pat, false);
+}
+
+/**
+ * strlazymatch - matching pattern strings lazily with glob pattern
+ * @str: the target string to match
+ * @pat: the pattern string to match
+ *
+ * This is similar to strglobmatch, except this ignores spaces in
+ * the target string.
+ */
+bool strlazymatch(const char *str, const char *pat)
+{
+	return __match_glob(str, pat, true);
+}
+
+/**
+ * strtailcmp - Compare the tail of two strings
+ * @s1: 1st string to be compared
+ * @s2: 2nd string to be compared
+ *
+ * Return 0 if whole of either string is same as another's tail part.
+ */
+int strtailcmp(const char *s1, const char *s2)
+{
+	int i1 = strlen(s1);
+	int i2 = strlen(s2);
+	while (--i1 >= 0 && --i2 >= 0) {
+		if (s1[i1] != s2[i2])
+			return s1[i1] - s2[i2];
+	}
+	return 0;
+}
+
+/**
+ * strxfrchar - Locate and replace character in @s
+ * @s:    The string to be searched/changed.
+ * @from: Source character to be replaced.
+ * @to:   Destination character.
+ *
+ * Return pointer to the changed string.
+ */
+char *strxfrchar(char *s, char from, char to)
+{
+	char *p = s;
+
+	while ((p = strchr(p, from)) != NULL)
+		*p++ = to;
+
+	return s;
+}
+
+/**
+ * ltrim - Removes leading whitespace from @s.
+ * @s: The string to be stripped.
+ *
+ * Return pointer to the first non-whitespace character in @s.
+ */
+char *ltrim(char *s)
+{
+	int len = strlen(s);
+
+	while (len && isspace(*s)) {
+		len--;
+		s++;
+	}
+
+	return s;
+}
+
+/**
+ * rtrim - Removes trailing whitespace from @s.
+ * @s: The string to be stripped.
+ *
+ * Note that the first trailing whitespace is replaced with a %NUL-terminator
+ * in the given string @s. Returns @s.
+ */
+char *rtrim(char *s)
+{
+	size_t size = strlen(s);
+	char *end;
+
+	if (!size)
+		return s;
+
+	end = s + size - 1;
+	while (end >= s && isspace(*end))
+		end--;
+	*(end + 1) = '\0';
+
+	return s;
+}
+
+/**
+ * memdup - duplicate region of memory
+ * @src: memory region to duplicate
+ * @len: memory region length
+ */
+void *memdup(const void *src, size_t len)
+{
+	void *p;
+
+	p = malloc(len);
+	if (p)
+		memcpy(p, src, len);
+
+	return p;
+}
diff --git a/tools/perf/util/strlist.c b/tools/perf/util/strlist.c
new file mode 100644
index 000000000..71f9d102b
--- /dev/null
+++ b/tools/perf/util/strlist.c
@@ -0,0 +1,173 @@
+/*
+ * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Licensed under the GPLv2.
+ */
+
+#include "strlist.h"
+#include "util.h"
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+static
+struct rb_node *strlist__node_new(struct rblist *rblist, const void *entry)
+{
+	const char *s = entry;
+	struct rb_node *rc = NULL;
+	struct strlist *strlist = container_of(rblist, struct strlist, rblist);
+	struct str_node *snode = malloc(sizeof(*snode));
+
+	if (snode != NULL) {
+		if (strlist->dupstr) {
+			s = strdup(s);
+			if (s == NULL)
+				goto out_delete;
+		}
+		snode->s = s;
+		rc = &snode->rb_node;
+	}
+
+	return rc;
+
+out_delete:
+	free(snode);
+	return NULL;
+}
+
+static void str_node__delete(struct str_node *snode, bool dupstr)
+{
+	if (dupstr)
+		zfree((char **)&snode->s);
+	free(snode);
+}
+
+static
+void strlist__node_delete(struct rblist *rblist, struct rb_node *rb_node)
+{
+	struct strlist *slist = container_of(rblist, struct strlist, rblist);
+	struct str_node *snode = container_of(rb_node, struct str_node, rb_node);
+
+	str_node__delete(snode, slist->dupstr);
+}
+
+static int strlist__node_cmp(struct rb_node *rb_node, const void *entry)
+{
+	const char *str = entry;
+	struct str_node *snode = container_of(rb_node, struct str_node, rb_node);
+
+	return strcmp(snode->s, str);
+}
+
+int strlist__add(struct strlist *slist, const char *new_entry)
+{
+	return rblist__add_node(&slist->rblist, new_entry);
+}
+
+int strlist__load(struct strlist *slist, const char *filename)
+{
+	char entry[1024];
+	int err;
+	FILE *fp = fopen(filename, "r");
+
+	if (fp == NULL)
+		return errno;
+
+	while (fgets(entry, sizeof(entry), fp) != NULL) {
+		const size_t len = strlen(entry);
+
+		if (len == 0)
+			continue;
+		entry[len - 1] = '\0';
+
+		err = strlist__add(slist, entry);
+		if (err != 0)
+			goto out;
+	}
+
+	err = 0;
+out:
+	fclose(fp);
+	return err;
+}
+
+void strlist__remove(struct strlist *slist, struct str_node *snode)
+{
+	rblist__remove_node(&slist->rblist, &snode->rb_node);
+}
+
+struct str_node *strlist__find(struct strlist *slist, const char *entry)
+{
+	struct str_node *snode = NULL;
+	struct rb_node *rb_node = rblist__find(&slist->rblist, entry);
+
+	if (rb_node)
+		snode = container_of(rb_node, struct str_node, rb_node);
+
+	return snode;
+}
+
+static int strlist__parse_list_entry(struct strlist *slist, const char *s)
+{
+	if (strncmp(s, "file://", 7) == 0)
+		return strlist__load(slist, s + 7);
+
+	return strlist__add(slist, s);
+}
+
+int strlist__parse_list(struct strlist *slist, const char *s)
+{
+	char *sep;
+	int err;
+
+	while ((sep = strchr(s, ',')) != NULL) {
+		*sep = '\0';
+		err = strlist__parse_list_entry(slist, s);
+		*sep = ',';
+		if (err != 0)
+			return err;
+		s = sep + 1;
+	}
+
+	return *s ? strlist__parse_list_entry(slist, s) : 0;
+}
+
+struct strlist *strlist__new(bool dupstr, const char *list)
+{
+	struct strlist *slist = malloc(sizeof(*slist));
+
+	if (slist != NULL) {
+		rblist__init(&slist->rblist);
+		slist->rblist.node_cmp    = strlist__node_cmp;
+		slist->rblist.node_new    = strlist__node_new;
+		slist->rblist.node_delete = strlist__node_delete;
+
+		slist->dupstr	 = dupstr;
+		if (list && strlist__parse_list(slist, list) != 0)
+			goto out_error;
+	}
+
+	return slist;
+out_error:
+	free(slist);
+	return NULL;
+}
+
+void strlist__delete(struct strlist *slist)
+{
+	if (slist != NULL)
+		rblist__delete(&slist->rblist);
+}
+
+struct str_node *strlist__entry(const struct strlist *slist, unsigned int idx)
+{
+	struct str_node *snode = NULL;
+	struct rb_node *rb_node;
+
+	rb_node = rblist__entry(&slist->rblist, idx);
+	if (rb_node)
+		snode = container_of(rb_node, struct str_node, rb_node);
+
+	return snode;
+}
diff --git a/tools/perf/util/strlist.h b/tools/perf/util/strlist.h
new file mode 100644
index 000000000..5c7f87069
--- /dev/null
+++ b/tools/perf/util/strlist.h
@@ -0,0 +1,79 @@
+#ifndef __PERF_STRLIST_H
+#define __PERF_STRLIST_H
+
+#include <linux/rbtree.h>
+#include <stdbool.h>
+
+#include "rblist.h"
+
+struct str_node {
+	struct rb_node rb_node;
+	const char     *s;
+};
+
+struct strlist {
+	struct rblist rblist;
+	bool	       dupstr;
+};
+
+struct strlist *strlist__new(bool dupstr, const char *slist);
+void strlist__delete(struct strlist *slist);
+
+void strlist__remove(struct strlist *slist, struct str_node *sn);
+int strlist__load(struct strlist *slist, const char *filename);
+int strlist__add(struct strlist *slist, const char *str);
+
+struct str_node *strlist__entry(const struct strlist *slist, unsigned int idx);
+struct str_node *strlist__find(struct strlist *slist, const char *entry);
+
+static inline bool strlist__has_entry(struct strlist *slist, const char *entry)
+{
+	return strlist__find(slist, entry) != NULL;
+}
+
+static inline bool strlist__empty(const struct strlist *slist)
+{
+	return rblist__empty(&slist->rblist);
+}
+
+static inline unsigned int strlist__nr_entries(const struct strlist *slist)
+{
+	return rblist__nr_entries(&slist->rblist);
+}
+
+/* For strlist iteration */
+static inline struct str_node *strlist__first(struct strlist *slist)
+{
+	struct rb_node *rn = rb_first(&slist->rblist.entries);
+	return rn ? rb_entry(rn, struct str_node, rb_node) : NULL;
+}
+static inline struct str_node *strlist__next(struct str_node *sn)
+{
+	struct rb_node *rn;
+	if (!sn)
+		return NULL;
+	rn = rb_next(&sn->rb_node);
+	return rn ? rb_entry(rn, struct str_node, rb_node) : NULL;
+}
+
+/**
+ * strlist_for_each      - iterate over a strlist
+ * @pos:	the &struct str_node to use as a loop cursor.
+ * @slist:	the &struct strlist for loop.
+ */
+#define strlist__for_each(pos, slist)	\
+	for (pos = strlist__first(slist); pos; pos = strlist__next(pos))
+
+/**
+ * strlist_for_each_safe - iterate over a strlist safe against removal of
+ *                         str_node
+ * @pos:	the &struct str_node to use as a loop cursor.
+ * @n:		another &struct str_node to use as temporary storage.
+ * @slist:	the &struct strlist for loop.
+ */
+#define strlist__for_each_safe(pos, n, slist)	\
+	for (pos = strlist__first(slist), n = strlist__next(pos); pos;\
+	     pos = n, n = strlist__next(n))
+
+int strlist__parse_list(struct strlist *slist, const char *s);
+#endif /* __PERF_STRLIST_H */
diff --git a/tools/perf/util/svghelper.c b/tools/perf/util/svghelper.c
new file mode 100644
index 000000000..283d3e73e
--- /dev/null
+++ b/tools/perf/util/svghelper.c
@@ -0,0 +1,808 @@
+/*
+ * svghelper.c - helper functions for outputting svg
+ *
+ * (C) Copyright 2009 Intel Corporation
+ *
+ * Authors:
+ *     Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <inttypes.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <string.h>
+#include <linux/bitmap.h>
+
+#include "perf.h"
+#include "svghelper.h"
+#include "util.h"
+#include "cpumap.h"
+
+static u64 first_time, last_time;
+static u64 turbo_frequency, max_freq;
+
+
+#define SLOT_MULT 30.0
+#define SLOT_HEIGHT 25.0
+#define SLOT_HALF (SLOT_HEIGHT / 2)
+
+int svg_page_width = 1000;
+u64 svg_highlight;
+const char *svg_highlight_name;
+
+#define MIN_TEXT_SIZE 0.01
+
+static u64 total_height;
+static FILE *svgfile;
+
+static double cpu2slot(int cpu)
+{
+	return 2 * cpu + 1;
+}
+
+static int *topology_map;
+
+static double cpu2y(int cpu)
+{
+	if (topology_map)
+		return cpu2slot(topology_map[cpu]) * SLOT_MULT;
+	else
+		return cpu2slot(cpu) * SLOT_MULT;
+}
+
+static double time2pixels(u64 __time)
+{
+	double X;
+
+	X = 1.0 * svg_page_width * (__time - first_time) / (last_time - first_time);
+	return X;
+}
+
+/*
+ * Round text sizes so that the svg viewer only needs a discrete
+ * number of renderings of the font
+ */
+static double round_text_size(double size)
+{
+	int loop = 100;
+	double target = 10.0;
+
+	if (size >= 10.0)
+		return size;
+	while (loop--) {
+		if (size >= target)
+			return target;
+		target = target / 2.0;
+	}
+	return size;
+}
+
+void open_svg(const char *filename, int cpus, int rows, u64 start, u64 end)
+{
+	int new_width;
+
+	svgfile = fopen(filename, "w");
+	if (!svgfile) {
+		fprintf(stderr, "Cannot open %s for output\n", filename);
+		return;
+	}
+	first_time = start;
+	first_time = first_time / 100000000 * 100000000;
+	last_time = end;
+
+	/*
+	 * if the recording is short, we default to a width of 1000, but
+	 * for longer recordings we want at least 200 units of width per second
+	 */
+	new_width = (last_time - first_time) / 5000000;
+
+	if (new_width > svg_page_width)
+		svg_page_width = new_width;
+
+	total_height = (1 + rows + cpu2slot(cpus)) * SLOT_MULT;
+	fprintf(svgfile, "<?xml version=\"1.0\" standalone=\"no\"?> \n");
+	fprintf(svgfile, "<!DOCTYPE svg SYSTEM \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n");
+	fprintf(svgfile, "<svg width=\"%i\" height=\"%" PRIu64 "\" version=\"1.1\" xmlns=\"http://www.w3.org/2000/svg\">\n", svg_page_width, total_height);
+
+	fprintf(svgfile, "<defs>\n  <style type=\"text/css\">\n    <![CDATA[\n");
+
+	fprintf(svgfile, "      rect          { stroke-width: 1; }\n");
+	fprintf(svgfile, "      rect.process  { fill:rgb(180,180,180); fill-opacity:0.9; stroke-width:1;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.process2 { fill:rgb(180,180,180); fill-opacity:0.9; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.process3 { fill:rgb(180,180,180); fill-opacity:0.5; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.sample   { fill:rgb(  0,  0,255); fill-opacity:0.8; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.sample_hi{ fill:rgb(255,128,  0); fill-opacity:0.8; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.error    { fill:rgb(255,  0,  0); fill-opacity:0.5; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.net      { fill:rgb(  0,128,  0); fill-opacity:0.5; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.disk     { fill:rgb(  0,  0,255); fill-opacity:0.5; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.sync     { fill:rgb(128,128,  0); fill-opacity:0.5; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.poll     { fill:rgb(  0,128,128); fill-opacity:0.2; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.blocked  { fill:rgb(255,  0,  0); fill-opacity:0.5; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.waiting  { fill:rgb(224,214,  0); fill-opacity:0.8; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.WAITING  { fill:rgb(255,214, 48); fill-opacity:0.6; stroke-width:0;   stroke:rgb(  0,  0,  0); } \n");
+	fprintf(svgfile, "      rect.cpu      { fill:rgb(192,192,192); fill-opacity:0.2; stroke-width:0.5; stroke:rgb(128,128,128); } \n");
+	fprintf(svgfile, "      rect.pstate   { fill:rgb(128,128,128); fill-opacity:0.8; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c1       { fill:rgb(255,214,214); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c2       { fill:rgb(255,172,172); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c3       { fill:rgb(255,130,130); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c4       { fill:rgb(255, 88, 88); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c5       { fill:rgb(255, 44, 44); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      rect.c6       { fill:rgb(255,  0,  0); fill-opacity:0.5; stroke-width:0; } \n");
+	fprintf(svgfile, "      line.pstate   { stroke:rgb(255,255,  0); stroke-opacity:0.8; stroke-width:2; } \n");
+
+	fprintf(svgfile, "    ]]>\n   </style>\n</defs>\n");
+}
+
+static double normalize_height(double height)
+{
+	if (height < 0.25)
+		return 0.25;
+	else if (height < 0.50)
+		return 0.50;
+	else if (height < 0.75)
+		return 0.75;
+	else
+		return 0.100;
+}
+
+void svg_ubox(int Yslot, u64 start, u64 end, double height, const char *type, int fd, int err, int merges)
+{
+	double w = time2pixels(end) - time2pixels(start);
+	height = normalize_height(height);
+
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<g>\n");
+	fprintf(svgfile, "<title>fd=%d error=%d merges=%d</title>\n", fd, err, merges);
+	fprintf(svgfile, "<rect x=\"%.8f\" width=\"%.8f\" y=\"%.1f\" height=\"%.1f\" class=\"%s\"/>\n",
+		time2pixels(start),
+		w,
+		Yslot * SLOT_MULT,
+		SLOT_HALF * height,
+		type);
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_lbox(int Yslot, u64 start, u64 end, double height, const char *type, int fd, int err, int merges)
+{
+	double w = time2pixels(end) - time2pixels(start);
+	height = normalize_height(height);
+
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<g>\n");
+	fprintf(svgfile, "<title>fd=%d error=%d merges=%d</title>\n", fd, err, merges);
+	fprintf(svgfile, "<rect x=\"%.8f\" width=\"%.8f\" y=\"%.1f\" height=\"%.1f\" class=\"%s\"/>\n",
+		time2pixels(start),
+		w,
+		Yslot * SLOT_MULT + SLOT_HEIGHT - SLOT_HALF * height,
+		SLOT_HALF * height,
+		type);
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_fbox(int Yslot, u64 start, u64 end, double height, const char *type, int fd, int err, int merges)
+{
+	double w = time2pixels(end) - time2pixels(start);
+	height = normalize_height(height);
+
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<g>\n");
+	fprintf(svgfile, "<title>fd=%d error=%d merges=%d</title>\n", fd, err, merges);
+	fprintf(svgfile, "<rect x=\"%.8f\" width=\"%.8f\" y=\"%.1f\" height=\"%.1f\" class=\"%s\"/>\n",
+		time2pixels(start),
+		w,
+		Yslot * SLOT_MULT + SLOT_HEIGHT - SLOT_HEIGHT * height,
+		SLOT_HEIGHT * height,
+		type);
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_box(int Yslot, u64 start, u64 end, const char *type)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<rect x=\"%.8f\" width=\"%.8f\" y=\"%.1f\" height=\"%.1f\" class=\"%s\"/>\n",
+		time2pixels(start), time2pixels(end)-time2pixels(start), Yslot * SLOT_MULT, SLOT_HEIGHT, type);
+}
+
+static char *time_to_string(u64 duration);
+void svg_blocked(int Yslot, int cpu, u64 start, u64 end, const char *backtrace)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<g>\n");
+	fprintf(svgfile, "<title>#%d blocked %s</title>\n", cpu,
+		time_to_string(end - start));
+	if (backtrace)
+		fprintf(svgfile, "<desc>Blocked on:\n%s</desc>\n", backtrace);
+	svg_box(Yslot, start, end, "blocked");
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_running(int Yslot, int cpu, u64 start, u64 end, const char *backtrace)
+{
+	double text_size;
+	const char *type;
+
+	if (!svgfile)
+		return;
+
+	if (svg_highlight && end - start > svg_highlight)
+		type = "sample_hi";
+	else
+		type = "sample";
+	fprintf(svgfile, "<g>\n");
+
+	fprintf(svgfile, "<title>#%d running %s</title>\n",
+		cpu, time_to_string(end - start));
+	if (backtrace)
+		fprintf(svgfile, "<desc>Switched because:\n%s</desc>\n", backtrace);
+	fprintf(svgfile, "<rect x=\"%.8f\" width=\"%.8f\" y=\"%.1f\" height=\"%.1f\" class=\"%s\"/>\n",
+		time2pixels(start), time2pixels(end)-time2pixels(start), Yslot * SLOT_MULT, SLOT_HEIGHT,
+		type);
+
+	text_size = (time2pixels(end)-time2pixels(start));
+	if (cpu > 9)
+		text_size = text_size/2;
+	if (text_size > 1.25)
+		text_size = 1.25;
+	text_size = round_text_size(text_size);
+
+	if (text_size > MIN_TEXT_SIZE)
+		fprintf(svgfile, "<text x=\"%.8f\" y=\"%.8f\" font-size=\"%.8fpt\">%i</text>\n",
+			time2pixels(start), Yslot *  SLOT_MULT + SLOT_HEIGHT - 1, text_size,  cpu + 1);
+
+	fprintf(svgfile, "</g>\n");
+}
+
+static char *time_to_string(u64 duration)
+{
+	static char text[80];
+
+	text[0] = 0;
+
+	if (duration < 1000) /* less than 1 usec */
+		return text;
+
+	if (duration < 1000 * 1000) { /* less than 1 msec */
+		sprintf(text, "%.1f us", duration / 1000.0);
+		return text;
+	}
+	sprintf(text, "%.1f ms", duration / 1000.0 / 1000);
+
+	return text;
+}
+
+void svg_waiting(int Yslot, int cpu, u64 start, u64 end, const char *backtrace)
+{
+	char *text;
+	const char *style;
+	double font_size;
+
+	if (!svgfile)
+		return;
+
+	style = "waiting";
+
+	if (end-start > 10 * 1000000) /* 10 msec */
+		style = "WAITING";
+
+	text = time_to_string(end-start);
+
+	font_size = 1.0 * (time2pixels(end)-time2pixels(start));
+
+	if (font_size > 3)
+		font_size = 3;
+
+	font_size = round_text_size(font_size);
+
+	fprintf(svgfile, "<g transform=\"translate(%.8f,%.8f)\">\n", time2pixels(start), Yslot * SLOT_MULT);
+	fprintf(svgfile, "<title>#%d waiting %s</title>\n", cpu, time_to_string(end - start));
+	if (backtrace)
+		fprintf(svgfile, "<desc>Waiting on:\n%s</desc>\n", backtrace);
+	fprintf(svgfile, "<rect x=\"0\" width=\"%.8f\" y=\"0\" height=\"%.1f\" class=\"%s\"/>\n",
+		time2pixels(end)-time2pixels(start), SLOT_HEIGHT, style);
+	if (font_size > MIN_TEXT_SIZE)
+		fprintf(svgfile, "<text transform=\"rotate(90)\" font-size=\"%.8fpt\"> %s</text>\n",
+			font_size, text);
+	fprintf(svgfile, "</g>\n");
+}
+
+static char *cpu_model(void)
+{
+	static char cpu_m[255];
+	char buf[256];
+	FILE *file;
+
+	cpu_m[0] = 0;
+	/* CPU type */
+	file = fopen("/proc/cpuinfo", "r");
+	if (file) {
+		while (fgets(buf, 255, file)) {
+			if (strstr(buf, "model name")) {
+				strncpy(cpu_m, &buf[13], 255);
+				break;
+			}
+		}
+		fclose(file);
+	}
+
+	/* CPU type */
+	file = fopen("/sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies", "r");
+	if (file) {
+		while (fgets(buf, 255, file)) {
+			unsigned int freq;
+			freq = strtoull(buf, NULL, 10);
+			if (freq > max_freq)
+				max_freq = freq;
+		}
+		fclose(file);
+	}
+	return cpu_m;
+}
+
+void svg_cpu_box(int cpu, u64 __max_freq, u64 __turbo_freq)
+{
+	char cpu_string[80];
+	if (!svgfile)
+		return;
+
+	max_freq = __max_freq;
+	turbo_frequency = __turbo_freq;
+
+	fprintf(svgfile, "<g>\n");
+
+	fprintf(svgfile, "<rect x=\"%.8f\" width=\"%.8f\" y=\"%.1f\" height=\"%.1f\" class=\"cpu\"/>\n",
+		time2pixels(first_time),
+		time2pixels(last_time)-time2pixels(first_time),
+		cpu2y(cpu), SLOT_MULT+SLOT_HEIGHT);
+
+	sprintf(cpu_string, "CPU %i", (int)cpu);
+	fprintf(svgfile, "<text x=\"%.8f\" y=\"%.8f\">%s</text>\n",
+		10+time2pixels(first_time), cpu2y(cpu) + SLOT_HEIGHT/2, cpu_string);
+
+	fprintf(svgfile, "<text transform=\"translate(%.8f,%.8f)\" font-size=\"1.25pt\">%s</text>\n",
+		10+time2pixels(first_time), cpu2y(cpu) + SLOT_MULT + SLOT_HEIGHT - 4, cpu_model());
+
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_process(int cpu, u64 start, u64 end, int pid, const char *name, const char *backtrace)
+{
+	double width;
+	const char *type;
+
+	if (!svgfile)
+		return;
+
+	if (svg_highlight && end - start >= svg_highlight)
+		type = "sample_hi";
+	else if (svg_highlight_name && strstr(name, svg_highlight_name))
+		type = "sample_hi";
+	else
+		type = "sample";
+
+	fprintf(svgfile, "<g transform=\"translate(%.8f,%.8f)\">\n", time2pixels(start), cpu2y(cpu));
+	fprintf(svgfile, "<title>%d %s running %s</title>\n", pid, name, time_to_string(end - start));
+	if (backtrace)
+		fprintf(svgfile, "<desc>Switched because:\n%s</desc>\n", backtrace);
+	fprintf(svgfile, "<rect x=\"0\" width=\"%.8f\" y=\"0\" height=\"%.1f\" class=\"%s\"/>\n",
+		time2pixels(end)-time2pixels(start), SLOT_MULT+SLOT_HEIGHT, type);
+	width = time2pixels(end)-time2pixels(start);
+	if (width > 6)
+		width = 6;
+
+	width = round_text_size(width);
+
+	if (width > MIN_TEXT_SIZE)
+		fprintf(svgfile, "<text transform=\"rotate(90)\" font-size=\"%.8fpt\">%s</text>\n",
+			width, name);
+
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_cstate(int cpu, u64 start, u64 end, int type)
+{
+	double width;
+	char style[128];
+
+	if (!svgfile)
+		return;
+
+
+	fprintf(svgfile, "<g>\n");
+
+	if (type > 6)
+		type = 6;
+	sprintf(style, "c%i", type);
+
+	fprintf(svgfile, "<rect class=\"%s\" x=\"%.8f\" width=\"%.8f\" y=\"%.1f\" height=\"%.1f\"/>\n",
+		style,
+		time2pixels(start), time2pixels(end)-time2pixels(start),
+		cpu2y(cpu), SLOT_MULT+SLOT_HEIGHT);
+
+	width = (time2pixels(end)-time2pixels(start))/2.0;
+	if (width > 6)
+		width = 6;
+
+	width = round_text_size(width);
+
+	if (width > MIN_TEXT_SIZE)
+		fprintf(svgfile, "<text x=\"%.8f\" y=\"%.8f\" font-size=\"%.8fpt\">C%i</text>\n",
+			time2pixels(start), cpu2y(cpu)+width, width, type);
+
+	fprintf(svgfile, "</g>\n");
+}
+
+static char *HzToHuman(unsigned long hz)
+{
+	static char buffer[1024];
+	unsigned long long Hz;
+
+	memset(buffer, 0, 1024);
+
+	Hz = hz;
+
+	/* default: just put the Number in */
+	sprintf(buffer, "%9lli", Hz);
+
+	if (Hz > 1000)
+		sprintf(buffer, " %6lli Mhz", (Hz+500)/1000);
+
+	if (Hz > 1500000)
+		sprintf(buffer, " %6.2f Ghz", (Hz+5000.0)/1000000);
+
+	if (Hz == turbo_frequency)
+		sprintf(buffer, "Turbo");
+
+	return buffer;
+}
+
+void svg_pstate(int cpu, u64 start, u64 end, u64 freq)
+{
+	double height = 0;
+
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<g>\n");
+
+	if (max_freq)
+		height = freq * 1.0 / max_freq * (SLOT_HEIGHT + SLOT_MULT);
+	height = 1 + cpu2y(cpu) + SLOT_MULT + SLOT_HEIGHT - height;
+	fprintf(svgfile, "<line x1=\"%.8f\" x2=\"%.8f\" y1=\"%.1f\" y2=\"%.1f\" class=\"pstate\"/>\n",
+		time2pixels(start), time2pixels(end), height, height);
+	fprintf(svgfile, "<text x=\"%.8f\" y=\"%.8f\" font-size=\"0.25pt\">%s</text>\n",
+		time2pixels(start), height+0.9, HzToHuman(freq));
+
+	fprintf(svgfile, "</g>\n");
+}
+
+
+void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2, const char *backtrace)
+{
+	double height;
+
+	if (!svgfile)
+		return;
+
+
+	fprintf(svgfile, "<g>\n");
+
+	fprintf(svgfile, "<title>%s wakes up %s</title>\n",
+		desc1 ? desc1 : "?",
+		desc2 ? desc2 : "?");
+
+	if (backtrace)
+		fprintf(svgfile, "<desc>%s</desc>\n", backtrace);
+
+	if (row1 < row2) {
+		if (row1) {
+			fprintf(svgfile, "<line x1=\"%.8f\" y1=\"%.2f\" x2=\"%.8f\" y2=\"%.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+				time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/32);
+			if (desc2)
+				fprintf(svgfile, "<g transform=\"translate(%.8f,%.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &gt;</text></g>\n",
+					time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT + SLOT_HEIGHT/48, desc2);
+		}
+		if (row2) {
+			fprintf(svgfile, "<line x1=\"%.8f\" y1=\"%.2f\" x2=\"%.8f\" y2=\"%.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+				time2pixels(start), row2 * SLOT_MULT - SLOT_MULT/32,  time2pixels(start), row2 * SLOT_MULT);
+			if (desc1)
+				fprintf(svgfile, "<g transform=\"translate(%.8f,%.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &gt;</text></g>\n",
+					time2pixels(start), row2 * SLOT_MULT - SLOT_MULT/32, desc1);
+		}
+	} else {
+		if (row2) {
+			fprintf(svgfile, "<line x1=\"%.8f\" y1=\"%.2f\" x2=\"%.8f\" y2=\"%.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+				time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/32);
+			if (desc1)
+				fprintf(svgfile, "<g transform=\"translate(%.8f,%.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &lt;</text></g>\n",
+					time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT + SLOT_MULT/48, desc1);
+		}
+		if (row1) {
+			fprintf(svgfile, "<line x1=\"%.8f\" y1=\"%.2f\" x2=\"%.8f\" y2=\"%.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+				time2pixels(start), row1 * SLOT_MULT - SLOT_MULT/32,  time2pixels(start), row1 * SLOT_MULT);
+			if (desc2)
+				fprintf(svgfile, "<g transform=\"translate(%.8f,%.8f)\"><text transform=\"rotate(90)\" font-size=\"0.02pt\">%s &lt;</text></g>\n",
+					time2pixels(start), row1 * SLOT_MULT - SLOT_HEIGHT/32, desc2);
+		}
+	}
+	height = row1 * SLOT_MULT;
+	if (row2 > row1)
+		height += SLOT_HEIGHT;
+	if (row1)
+		fprintf(svgfile, "<circle  cx=\"%.8f\" cy=\"%.2f\" r = \"0.01\"  style=\"fill:rgb(32,255,32)\"/>\n",
+			time2pixels(start), height);
+
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_wakeline(u64 start, int row1, int row2, const char *backtrace)
+{
+	double height;
+
+	if (!svgfile)
+		return;
+
+
+	fprintf(svgfile, "<g>\n");
+
+	if (backtrace)
+		fprintf(svgfile, "<desc>%s</desc>\n", backtrace);
+
+	if (row1 < row2)
+		fprintf(svgfile, "<line x1=\"%.8f\" y1=\"%.2f\" x2=\"%.8f\" y2=\"%.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+			time2pixels(start), row1 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row2 * SLOT_MULT);
+	else
+		fprintf(svgfile, "<line x1=\"%.8f\" y1=\"%.2f\" x2=\"%.8f\" y2=\"%.2f\" style=\"stroke:rgb(32,255,32);stroke-width:0.009\"/>\n",
+			time2pixels(start), row2 * SLOT_MULT + SLOT_HEIGHT,  time2pixels(start), row1 * SLOT_MULT);
+
+	height = row1 * SLOT_MULT;
+	if (row2 > row1)
+		height += SLOT_HEIGHT;
+	fprintf(svgfile, "<circle  cx=\"%.8f\" cy=\"%.2f\" r = \"0.01\"  style=\"fill:rgb(32,255,32)\"/>\n",
+			time2pixels(start), height);
+
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_interrupt(u64 start, int row, const char *backtrace)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<g>\n");
+
+	fprintf(svgfile, "<title>Wakeup from interrupt</title>\n");
+
+	if (backtrace)
+		fprintf(svgfile, "<desc>%s</desc>\n", backtrace);
+
+	fprintf(svgfile, "<circle  cx=\"%.8f\" cy=\"%.2f\" r = \"0.01\"  style=\"fill:rgb(255,128,128)\"/>\n",
+			time2pixels(start), row * SLOT_MULT);
+	fprintf(svgfile, "<circle  cx=\"%.8f\" cy=\"%.2f\" r = \"0.01\"  style=\"fill:rgb(255,128,128)\"/>\n",
+			time2pixels(start), row * SLOT_MULT + SLOT_HEIGHT);
+
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_text(int Yslot, u64 start, const char *text)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<text x=\"%.8f\" y=\"%.8f\">%s</text>\n",
+		time2pixels(start), Yslot * SLOT_MULT+SLOT_HEIGHT/2, text);
+}
+
+static void svg_legenda_box(int X, const char *text, const char *style)
+{
+	double boxsize;
+	boxsize = SLOT_HEIGHT / 2;
+
+	fprintf(svgfile, "<rect x=\"%i\" width=\"%.8f\" y=\"0\" height=\"%.1f\" class=\"%s\"/>\n",
+		X, boxsize, boxsize, style);
+	fprintf(svgfile, "<text transform=\"translate(%.8f, %.8f)\" font-size=\"%.8fpt\">%s</text>\n",
+		X + boxsize + 5, boxsize, 0.8 * boxsize, text);
+}
+
+void svg_io_legenda(void)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<g>\n");
+	svg_legenda_box(0,	"Disk", "disk");
+	svg_legenda_box(100,	"Network", "net");
+	svg_legenda_box(200,	"Sync", "sync");
+	svg_legenda_box(300,	"Poll", "poll");
+	svg_legenda_box(400,	"Error", "error");
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_legenda(void)
+{
+	if (!svgfile)
+		return;
+
+	fprintf(svgfile, "<g>\n");
+	svg_legenda_box(0,	"Running", "sample");
+	svg_legenda_box(100,	"Idle","c1");
+	svg_legenda_box(200,	"Deeper Idle", "c3");
+	svg_legenda_box(350,	"Deepest Idle", "c6");
+	svg_legenda_box(550,	"Sleeping", "process2");
+	svg_legenda_box(650,	"Waiting for cpu", "waiting");
+	svg_legenda_box(800,	"Blocked on IO", "blocked");
+	fprintf(svgfile, "</g>\n");
+}
+
+void svg_time_grid(double min_thickness)
+{
+	u64 i;
+
+	if (!svgfile)
+		return;
+
+	i = first_time;
+	while (i < last_time) {
+		int color = 220;
+		double thickness = 0.075;
+		if ((i % 100000000) == 0) {
+			thickness = 0.5;
+			color = 192;
+		}
+		if ((i % 1000000000) == 0) {
+			thickness = 2.0;
+			color = 128;
+		}
+
+		if (thickness >= min_thickness)
+			fprintf(svgfile, "<line x1=\"%.8f\" y1=\"%.2f\" x2=\"%.8f\" y2=\"%" PRIu64 "\" style=\"stroke:rgb(%i,%i,%i);stroke-width:%.3f\"/>\n",
+				time2pixels(i), SLOT_MULT/2, time2pixels(i),
+				total_height, color, color, color, thickness);
+
+		i += 10000000;
+	}
+}
+
+void svg_close(void)
+{
+	if (svgfile) {
+		fprintf(svgfile, "</svg>\n");
+		fclose(svgfile);
+		svgfile = NULL;
+	}
+}
+
+#define cpumask_bits(maskp) ((maskp)->bits)
+typedef struct { DECLARE_BITMAP(bits, MAX_NR_CPUS); } cpumask_t;
+
+struct topology {
+	cpumask_t *sib_core;
+	int sib_core_nr;
+	cpumask_t *sib_thr;
+	int sib_thr_nr;
+};
+
+static void scan_thread_topology(int *map, struct topology *t, int cpu, int *pos)
+{
+	int i;
+	int thr;
+
+	for (i = 0; i < t->sib_thr_nr; i++) {
+		if (!test_bit(cpu, cpumask_bits(&t->sib_thr[i])))
+			continue;
+
+		for_each_set_bit(thr,
+				 cpumask_bits(&t->sib_thr[i]),
+				 MAX_NR_CPUS)
+			if (map[thr] == -1)
+				map[thr] = (*pos)++;
+	}
+}
+
+static void scan_core_topology(int *map, struct topology *t)
+{
+	int pos = 0;
+	int i;
+	int cpu;
+
+	for (i = 0; i < t->sib_core_nr; i++)
+		for_each_set_bit(cpu,
+				 cpumask_bits(&t->sib_core[i]),
+				 MAX_NR_CPUS)
+			scan_thread_topology(map, t, cpu, &pos);
+}
+
+static int str_to_bitmap(char *s, cpumask_t *b)
+{
+	int i;
+	int ret = 0;
+	struct cpu_map *m;
+	int c;
+
+	m = cpu_map__new(s);
+	if (!m)
+		return -1;
+
+	for (i = 0; i < m->nr; i++) {
+		c = m->map[i];
+		if (c >= MAX_NR_CPUS) {
+			ret = -1;
+			break;
+		}
+
+		set_bit(c, cpumask_bits(b));
+	}
+
+	cpu_map__delete(m);
+
+	return ret;
+}
+
+int svg_build_topology_map(char *sib_core, int sib_core_nr,
+			   char *sib_thr, int sib_thr_nr)
+{
+	int i;
+	struct topology t;
+
+	t.sib_core_nr = sib_core_nr;
+	t.sib_thr_nr = sib_thr_nr;
+	t.sib_core = calloc(sib_core_nr, sizeof(cpumask_t));
+	t.sib_thr = calloc(sib_thr_nr, sizeof(cpumask_t));
+
+	if (!t.sib_core || !t.sib_thr) {
+		fprintf(stderr, "topology: no memory\n");
+		goto exit;
+	}
+
+	for (i = 0; i < sib_core_nr; i++) {
+		if (str_to_bitmap(sib_core, &t.sib_core[i])) {
+			fprintf(stderr, "topology: can't parse siblings map\n");
+			goto exit;
+		}
+
+		sib_core += strlen(sib_core) + 1;
+	}
+
+	for (i = 0; i < sib_thr_nr; i++) {
+		if (str_to_bitmap(sib_thr, &t.sib_thr[i])) {
+			fprintf(stderr, "topology: can't parse siblings map\n");
+			goto exit;
+		}
+
+		sib_thr += strlen(sib_thr) + 1;
+	}
+
+	topology_map = malloc(sizeof(int) * MAX_NR_CPUS);
+	if (!topology_map) {
+		fprintf(stderr, "topology: no memory\n");
+		goto exit;
+	}
+
+	for (i = 0; i < MAX_NR_CPUS; i++)
+		topology_map[i] = -1;
+
+	scan_core_topology(topology_map, &t);
+
+	return 0;
+
+exit:
+	zfree(&t.sib_core);
+	zfree(&t.sib_thr);
+
+	return -1;
+}
diff --git a/tools/perf/util/svghelper.h b/tools/perf/util/svghelper.h
new file mode 100644
index 000000000..9292a5291
--- /dev/null
+++ b/tools/perf/util/svghelper.h
@@ -0,0 +1,37 @@
+#ifndef __PERF_SVGHELPER_H
+#define __PERF_SVGHELPER_H
+
+#include <linux/types.h>
+
+extern void open_svg(const char *filename, int cpus, int rows, u64 start, u64 end);
+extern void svg_ubox(int Yslot, u64 start, u64 end, double height, const char *type, int fd, int err, int merges);
+extern void svg_lbox(int Yslot, u64 start, u64 end, double height, const char *type, int fd, int err, int merges);
+extern void svg_fbox(int Yslot, u64 start, u64 end, double height, const char *type, int fd, int err, int merges);
+extern void svg_box(int Yslot, u64 start, u64 end, const char *type);
+extern void svg_blocked(int Yslot, int cpu, u64 start, u64 end, const char *backtrace);
+extern void svg_running(int Yslot, int cpu, u64 start, u64 end, const char *backtrace);
+extern void svg_waiting(int Yslot, int cpu, u64 start, u64 end, const char *backtrace);
+extern void svg_cpu_box(int cpu, u64 max_frequency, u64 turbo_frequency);
+
+
+extern void svg_process(int cpu, u64 start, u64 end, int pid, const char *name, const char *backtrace);
+extern void svg_cstate(int cpu, u64 start, u64 end, int type);
+extern void svg_pstate(int cpu, u64 start, u64 end, u64 freq);
+
+
+extern void svg_time_grid(double min_thickness);
+extern void svg_io_legenda(void);
+extern void svg_legenda(void);
+extern void svg_wakeline(u64 start, int row1, int row2, const char *backtrace);
+extern void svg_partial_wakeline(u64 start, int row1, char *desc1, int row2, char *desc2, const char *backtrace);
+extern void svg_interrupt(u64 start, int row, const char *backtrace);
+extern void svg_text(int Yslot, u64 start, const char *text);
+extern void svg_close(void);
+extern int svg_build_topology_map(char *sib_core, int sib_core_nr,
+				  char *sib_thr, int sib_thr_nr);
+
+extern int svg_page_width;
+extern u64 svg_highlight;
+extern const char *svg_highlight_name;
+
+#endif /* __PERF_SVGHELPER_H */
diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
new file mode 100644
index 000000000..a7ab6063e
--- /dev/null
+++ b/tools/perf/util/symbol-elf.c
@@ -0,0 +1,1774 @@
+#include <fcntl.h>
+#include <stdio.h>
+#include <errno.h>
+#include <string.h>
+#include <unistd.h>
+#include <inttypes.h>
+
+#include "symbol.h"
+#include "machine.h"
+#include "vdso.h"
+#include <symbol/kallsyms.h>
+#include "debug.h"
+
+#ifndef EM_AARCH64
+#define EM_AARCH64	183  /* ARM 64 bit */
+#endif
+
+
+#ifdef HAVE_CPLUS_DEMANGLE_SUPPORT
+extern char *cplus_demangle(const char *, int);
+
+static inline char *bfd_demangle(void __maybe_unused *v, const char *c, int i)
+{
+	return cplus_demangle(c, i);
+}
+#else
+#ifdef NO_DEMANGLE
+static inline char *bfd_demangle(void __maybe_unused *v,
+				 const char __maybe_unused *c,
+				 int __maybe_unused i)
+{
+	return NULL;
+}
+#else
+#define PACKAGE 'perf'
+#include <bfd.h>
+#endif
+#endif
+
+#ifndef HAVE_ELF_GETPHDRNUM_SUPPORT
+static int elf_getphdrnum(Elf *elf, size_t *dst)
+{
+	GElf_Ehdr gehdr;
+	GElf_Ehdr *ehdr;
+
+	ehdr = gelf_getehdr(elf, &gehdr);
+	if (!ehdr)
+		return -1;
+
+	*dst = ehdr->e_phnum;
+
+	return 0;
+}
+#endif
+
+#ifndef NT_GNU_BUILD_ID
+#define NT_GNU_BUILD_ID 3
+#endif
+
+/**
+ * elf_symtab__for_each_symbol - iterate thru all the symbols
+ *
+ * @syms: struct elf_symtab instance to iterate
+ * @idx: uint32_t idx
+ * @sym: GElf_Sym iterator
+ */
+#define elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) \
+	for (idx = 0, gelf_getsym(syms, idx, &sym);\
+	     idx < nr_syms; \
+	     idx++, gelf_getsym(syms, idx, &sym))
+
+static inline uint8_t elf_sym__type(const GElf_Sym *sym)
+{
+	return GELF_ST_TYPE(sym->st_info);
+}
+
+#ifndef STT_GNU_IFUNC
+#define STT_GNU_IFUNC 10
+#endif
+
+static inline int elf_sym__is_function(const GElf_Sym *sym)
+{
+	return (elf_sym__type(sym) == STT_FUNC ||
+		elf_sym__type(sym) == STT_GNU_IFUNC) &&
+	       sym->st_name != 0 &&
+	       sym->st_shndx != SHN_UNDEF;
+}
+
+static inline bool elf_sym__is_object(const GElf_Sym *sym)
+{
+	return elf_sym__type(sym) == STT_OBJECT &&
+		sym->st_name != 0 &&
+		sym->st_shndx != SHN_UNDEF;
+}
+
+static inline int elf_sym__is_label(const GElf_Sym *sym)
+{
+	return elf_sym__type(sym) == STT_NOTYPE &&
+		sym->st_name != 0 &&
+		sym->st_shndx != SHN_UNDEF &&
+		sym->st_shndx != SHN_ABS;
+}
+
+static bool elf_sym__is_a(GElf_Sym *sym, enum map_type type)
+{
+	switch (type) {
+	case MAP__FUNCTION:
+		return elf_sym__is_function(sym);
+	case MAP__VARIABLE:
+		return elf_sym__is_object(sym);
+	default:
+		return false;
+	}
+}
+
+static inline const char *elf_sym__name(const GElf_Sym *sym,
+					const Elf_Data *symstrs)
+{
+	return symstrs->d_buf + sym->st_name;
+}
+
+static inline const char *elf_sec__name(const GElf_Shdr *shdr,
+					const Elf_Data *secstrs)
+{
+	return secstrs->d_buf + shdr->sh_name;
+}
+
+static inline int elf_sec__is_text(const GElf_Shdr *shdr,
+					const Elf_Data *secstrs)
+{
+	return strstr(elf_sec__name(shdr, secstrs), "text") != NULL;
+}
+
+static inline bool elf_sec__is_data(const GElf_Shdr *shdr,
+				    const Elf_Data *secstrs)
+{
+	return strstr(elf_sec__name(shdr, secstrs), "data") != NULL;
+}
+
+static bool elf_sec__is_a(GElf_Shdr *shdr, Elf_Data *secstrs,
+			  enum map_type type)
+{
+	switch (type) {
+	case MAP__FUNCTION:
+		return elf_sec__is_text(shdr, secstrs);
+	case MAP__VARIABLE:
+		return elf_sec__is_data(shdr, secstrs);
+	default:
+		return false;
+	}
+}
+
+static size_t elf_addr_to_index(Elf *elf, GElf_Addr addr)
+{
+	Elf_Scn *sec = NULL;
+	GElf_Shdr shdr;
+	size_t cnt = 1;
+
+	while ((sec = elf_nextscn(elf, sec)) != NULL) {
+		gelf_getshdr(sec, &shdr);
+
+		if ((addr >= shdr.sh_addr) &&
+		    (addr < (shdr.sh_addr + shdr.sh_size)))
+			return cnt;
+
+		++cnt;
+	}
+
+	return -1;
+}
+
+Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
+			     GElf_Shdr *shp, const char *name, size_t *idx)
+{
+	Elf_Scn *sec = NULL;
+	size_t cnt = 1;
+
+	/* Elf is corrupted/truncated, avoid calling elf_strptr. */
+	if (!elf_rawdata(elf_getscn(elf, ep->e_shstrndx), NULL))
+		return NULL;
+
+	while ((sec = elf_nextscn(elf, sec)) != NULL) {
+		char *str;
+
+		gelf_getshdr(sec, shp);
+		str = elf_strptr(elf, ep->e_shstrndx, shp->sh_name);
+		if (str && !strcmp(name, str)) {
+			if (idx)
+				*idx = cnt;
+			return sec;
+		}
+		++cnt;
+	}
+
+	return NULL;
+}
+
+#define elf_section__for_each_rel(reldata, pos, pos_mem, idx, nr_entries) \
+	for (idx = 0, pos = gelf_getrel(reldata, 0, &pos_mem); \
+	     idx < nr_entries; \
+	     ++idx, pos = gelf_getrel(reldata, idx, &pos_mem))
+
+#define elf_section__for_each_rela(reldata, pos, pos_mem, idx, nr_entries) \
+	for (idx = 0, pos = gelf_getrela(reldata, 0, &pos_mem); \
+	     idx < nr_entries; \
+	     ++idx, pos = gelf_getrela(reldata, idx, &pos_mem))
+
+/*
+ * We need to check if we have a .dynsym, so that we can handle the
+ * .plt, synthesizing its symbols, that aren't on the symtabs (be it
+ * .dynsym or .symtab).
+ * And always look at the original dso, not at debuginfo packages, that
+ * have the PLT data stripped out (shdr_rel_plt.sh_type == SHT_NOBITS).
+ */
+int dso__synthesize_plt_symbols(struct dso *dso, struct symsrc *ss, struct map *map,
+				symbol_filter_t filter)
+{
+	uint32_t nr_rel_entries, idx;
+	GElf_Sym sym;
+	u64 plt_offset;
+	GElf_Shdr shdr_plt;
+	struct symbol *f;
+	GElf_Shdr shdr_rel_plt, shdr_dynsym;
+	Elf_Data *reldata, *syms, *symstrs;
+	Elf_Scn *scn_plt_rel, *scn_symstrs, *scn_dynsym;
+	size_t dynsym_idx;
+	GElf_Ehdr ehdr;
+	char sympltname[1024];
+	Elf *elf;
+	int nr = 0, symidx, err = 0;
+
+	if (!ss->dynsym)
+		return 0;
+
+	elf = ss->elf;
+	ehdr = ss->ehdr;
+
+	scn_dynsym = ss->dynsym;
+	shdr_dynsym = ss->dynshdr;
+	dynsym_idx = ss->dynsym_idx;
+
+	if (scn_dynsym == NULL)
+		goto out_elf_end;
+
+	scn_plt_rel = elf_section_by_name(elf, &ehdr, &shdr_rel_plt,
+					  ".rela.plt", NULL);
+	if (scn_plt_rel == NULL) {
+		scn_plt_rel = elf_section_by_name(elf, &ehdr, &shdr_rel_plt,
+						  ".rel.plt", NULL);
+		if (scn_plt_rel == NULL)
+			goto out_elf_end;
+	}
+
+	err = -1;
+
+	if (shdr_rel_plt.sh_link != dynsym_idx)
+		goto out_elf_end;
+
+	if (elf_section_by_name(elf, &ehdr, &shdr_plt, ".plt", NULL) == NULL)
+		goto out_elf_end;
+
+	/*
+	 * Fetch the relocation section to find the idxes to the GOT
+	 * and the symbols in the .dynsym they refer to.
+	 */
+	reldata = elf_getdata(scn_plt_rel, NULL);
+	if (reldata == NULL)
+		goto out_elf_end;
+
+	syms = elf_getdata(scn_dynsym, NULL);
+	if (syms == NULL)
+		goto out_elf_end;
+
+	scn_symstrs = elf_getscn(elf, shdr_dynsym.sh_link);
+	if (scn_symstrs == NULL)
+		goto out_elf_end;
+
+	symstrs = elf_getdata(scn_symstrs, NULL);
+	if (symstrs == NULL)
+		goto out_elf_end;
+
+	if (symstrs->d_size == 0)
+		goto out_elf_end;
+
+	nr_rel_entries = shdr_rel_plt.sh_size / shdr_rel_plt.sh_entsize;
+	plt_offset = shdr_plt.sh_offset;
+
+	if (shdr_rel_plt.sh_type == SHT_RELA) {
+		GElf_Rela pos_mem, *pos;
+
+		elf_section__for_each_rela(reldata, pos, pos_mem, idx,
+					   nr_rel_entries) {
+			symidx = GELF_R_SYM(pos->r_info);
+			plt_offset += shdr_plt.sh_entsize;
+			gelf_getsym(syms, symidx, &sym);
+			snprintf(sympltname, sizeof(sympltname),
+				 "%s@plt", elf_sym__name(&sym, symstrs));
+
+			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
+					STB_GLOBAL, sympltname);
+			if (!f)
+				goto out_elf_end;
+
+			if (filter && filter(map, f))
+				symbol__delete(f);
+			else {
+				symbols__insert(&dso->symbols[map->type], f);
+				++nr;
+			}
+		}
+	} else if (shdr_rel_plt.sh_type == SHT_REL) {
+		GElf_Rel pos_mem, *pos;
+		elf_section__for_each_rel(reldata, pos, pos_mem, idx,
+					  nr_rel_entries) {
+			symidx = GELF_R_SYM(pos->r_info);
+			plt_offset += shdr_plt.sh_entsize;
+			gelf_getsym(syms, symidx, &sym);
+			snprintf(sympltname, sizeof(sympltname),
+				 "%s@plt", elf_sym__name(&sym, symstrs));
+
+			f = symbol__new(plt_offset, shdr_plt.sh_entsize,
+					STB_GLOBAL, sympltname);
+			if (!f)
+				goto out_elf_end;
+
+			if (filter && filter(map, f))
+				symbol__delete(f);
+			else {
+				symbols__insert(&dso->symbols[map->type], f);
+				++nr;
+			}
+		}
+	}
+
+	err = 0;
+out_elf_end:
+	if (err == 0)
+		return nr;
+	pr_debug("%s: problems reading %s PLT info.\n",
+		 __func__, dso->long_name);
+	return 0;
+}
+
+/*
+ * Align offset to 4 bytes as needed for note name and descriptor data.
+ */
+#define NOTE_ALIGN(n) (((n) + 3) & -4U)
+
+static int elf_read_build_id(Elf *elf, void *bf, size_t size)
+{
+	int err = -1;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+	Elf_Data *data;
+	Elf_Scn *sec;
+	Elf_Kind ek;
+	void *ptr;
+
+	if (size < BUILD_ID_SIZE)
+		goto out;
+
+	ek = elf_kind(elf);
+	if (ek != ELF_K_ELF)
+		goto out;
+
+	if (gelf_getehdr(elf, &ehdr) == NULL) {
+		pr_err("%s: cannot get elf header.\n", __func__);
+		goto out;
+	}
+
+	/*
+	 * Check following sections for notes:
+	 *   '.note.gnu.build-id'
+	 *   '.notes'
+	 *   '.note' (VDSO specific)
+	 */
+	do {
+		sec = elf_section_by_name(elf, &ehdr, &shdr,
+					  ".note.gnu.build-id", NULL);
+		if (sec)
+			break;
+
+		sec = elf_section_by_name(elf, &ehdr, &shdr,
+					  ".notes", NULL);
+		if (sec)
+			break;
+
+		sec = elf_section_by_name(elf, &ehdr, &shdr,
+					  ".note", NULL);
+		if (sec)
+			break;
+
+		return err;
+
+	} while (0);
+
+	data = elf_getdata(sec, NULL);
+	if (data == NULL)
+		goto out;
+
+	ptr = data->d_buf;
+	while (ptr < (data->d_buf + data->d_size)) {
+		GElf_Nhdr *nhdr = ptr;
+		size_t namesz = NOTE_ALIGN(nhdr->n_namesz),
+		       descsz = NOTE_ALIGN(nhdr->n_descsz);
+		const char *name;
+
+		ptr += sizeof(*nhdr);
+		name = ptr;
+		ptr += namesz;
+		if (nhdr->n_type == NT_GNU_BUILD_ID &&
+		    nhdr->n_namesz == sizeof("GNU")) {
+			if (memcmp(name, "GNU", sizeof("GNU")) == 0) {
+				size_t sz = min(size, descsz);
+				memcpy(bf, ptr, sz);
+				memset(bf + sz, 0, size - sz);
+				err = descsz;
+				break;
+			}
+		}
+		ptr += descsz;
+	}
+
+out:
+	return err;
+}
+
+int filename__read_build_id(const char *filename, void *bf, size_t size)
+{
+	int fd, err = -1;
+	Elf *elf;
+
+	if (size < BUILD_ID_SIZE)
+		goto out;
+
+	fd = open(filename, O_RDONLY);
+	if (fd < 0)
+		goto out;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL) {
+		pr_debug2("%s: cannot read %s ELF file.\n", __func__, filename);
+		goto out_close;
+	}
+
+	err = elf_read_build_id(elf, bf, size);
+
+	elf_end(elf);
+out_close:
+	close(fd);
+out:
+	return err;
+}
+
+int sysfs__read_build_id(const char *filename, void *build_id, size_t size)
+{
+	int fd, err = -1;
+
+	if (size < BUILD_ID_SIZE)
+		goto out;
+
+	fd = open(filename, O_RDONLY);
+	if (fd < 0)
+		goto out;
+
+	while (1) {
+		char bf[BUFSIZ];
+		GElf_Nhdr nhdr;
+		size_t namesz, descsz;
+
+		if (read(fd, &nhdr, sizeof(nhdr)) != sizeof(nhdr))
+			break;
+
+		namesz = NOTE_ALIGN(nhdr.n_namesz);
+		descsz = NOTE_ALIGN(nhdr.n_descsz);
+		if (nhdr.n_type == NT_GNU_BUILD_ID &&
+		    nhdr.n_namesz == sizeof("GNU")) {
+			if (read(fd, bf, namesz) != (ssize_t)namesz)
+				break;
+			if (memcmp(bf, "GNU", sizeof("GNU")) == 0) {
+				size_t sz = min(descsz, size);
+				if (read(fd, build_id, sz) == (ssize_t)sz) {
+					memset(build_id + sz, 0, size - sz);
+					err = 0;
+					break;
+				}
+			} else if (read(fd, bf, descsz) != (ssize_t)descsz)
+				break;
+		} else {
+			int n = namesz + descsz;
+			if (read(fd, bf, n) != n)
+				break;
+		}
+	}
+	close(fd);
+out:
+	return err;
+}
+
+int filename__read_debuglink(const char *filename, char *debuglink,
+			     size_t size)
+{
+	int fd, err = -1;
+	Elf *elf;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+	Elf_Data *data;
+	Elf_Scn *sec;
+	Elf_Kind ek;
+
+	fd = open(filename, O_RDONLY);
+	if (fd < 0)
+		goto out;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL) {
+		pr_debug2("%s: cannot read %s ELF file.\n", __func__, filename);
+		goto out_close;
+	}
+
+	ek = elf_kind(elf);
+	if (ek != ELF_K_ELF)
+		goto out_elf_end;
+
+	if (gelf_getehdr(elf, &ehdr) == NULL) {
+		pr_err("%s: cannot get elf header.\n", __func__);
+		goto out_elf_end;
+	}
+
+	sec = elf_section_by_name(elf, &ehdr, &shdr,
+				  ".gnu_debuglink", NULL);
+	if (sec == NULL)
+		goto out_elf_end;
+
+	data = elf_getdata(sec, NULL);
+	if (data == NULL)
+		goto out_elf_end;
+
+	/* the start of this section is a zero-terminated string */
+	strncpy(debuglink, data->d_buf, size);
+
+	err = 0;
+
+out_elf_end:
+	elf_end(elf);
+out_close:
+	close(fd);
+out:
+	return err;
+}
+
+static int dso__swap_init(struct dso *dso, unsigned char eidata)
+{
+	static unsigned int const endian = 1;
+
+	dso->needs_swap = DSO_SWAP__NO;
+
+	switch (eidata) {
+	case ELFDATA2LSB:
+		/* We are big endian, DSO is little endian. */
+		if (*(unsigned char const *)&endian != 1)
+			dso->needs_swap = DSO_SWAP__YES;
+		break;
+
+	case ELFDATA2MSB:
+		/* We are little endian, DSO is big endian. */
+		if (*(unsigned char const *)&endian != 0)
+			dso->needs_swap = DSO_SWAP__YES;
+		break;
+
+	default:
+		pr_err("unrecognized DSO data encoding %d\n", eidata);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int decompress_kmodule(struct dso *dso, const char *name,
+			      enum dso_binary_type type)
+{
+	int fd = -1;
+	char tmpbuf[] = "/tmp/perf-kmod-XXXXXX";
+	struct kmod_path m;
+
+	if (type != DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE_COMP &&
+	    type != DSO_BINARY_TYPE__GUEST_KMODULE_COMP &&
+	    type != DSO_BINARY_TYPE__BUILD_ID_CACHE)
+		return -1;
+
+	if (type == DSO_BINARY_TYPE__BUILD_ID_CACHE)
+		name = dso->long_name;
+
+	if (kmod_path__parse_ext(&m, name) || !m.comp)
+		return -1;
+
+	fd = mkstemp(tmpbuf);
+	if (fd < 0) {
+		dso->load_errno = errno;
+		goto out;
+	}
+
+	if (!decompress_to_file(m.ext, name, fd)) {
+		dso->load_errno = DSO_LOAD_ERRNO__DECOMPRESSION_FAILURE;
+		close(fd);
+		fd = -1;
+	}
+
+	unlink(tmpbuf);
+
+out:
+	free(m.ext);
+	return fd;
+}
+
+bool symsrc__possibly_runtime(struct symsrc *ss)
+{
+	return ss->dynsym || ss->opdsec;
+}
+
+bool symsrc__has_symtab(struct symsrc *ss)
+{
+	return ss->symtab != NULL;
+}
+
+void symsrc__destroy(struct symsrc *ss)
+{
+	zfree(&ss->name);
+	elf_end(ss->elf);
+	close(ss->fd);
+}
+
+int symsrc__init(struct symsrc *ss, struct dso *dso, const char *name,
+		 enum dso_binary_type type)
+{
+	int err = -1;
+	GElf_Ehdr ehdr;
+	Elf *elf;
+	int fd;
+
+	if (dso__needs_decompress(dso)) {
+		fd = decompress_kmodule(dso, name, type);
+		if (fd < 0)
+			return -1;
+	} else {
+		fd = open(name, O_RDONLY);
+		if (fd < 0) {
+			dso->load_errno = errno;
+			return -1;
+		}
+	}
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL) {
+		pr_debug("%s: cannot read %s ELF file.\n", __func__, name);
+		dso->load_errno = DSO_LOAD_ERRNO__INVALID_ELF;
+		goto out_close;
+	}
+
+	if (gelf_getehdr(elf, &ehdr) == NULL) {
+		dso->load_errno = DSO_LOAD_ERRNO__INVALID_ELF;
+		pr_debug("%s: cannot get elf header.\n", __func__);
+		goto out_elf_end;
+	}
+
+	if (dso__swap_init(dso, ehdr.e_ident[EI_DATA])) {
+		dso->load_errno = DSO_LOAD_ERRNO__INTERNAL_ERROR;
+		goto out_elf_end;
+	}
+
+	/* Always reject images with a mismatched build-id: */
+	if (dso->has_build_id) {
+		u8 build_id[BUILD_ID_SIZE];
+
+		if (elf_read_build_id(elf, build_id, BUILD_ID_SIZE) < 0) {
+			dso->load_errno = DSO_LOAD_ERRNO__CANNOT_READ_BUILDID;
+			goto out_elf_end;
+		}
+
+		if (!dso__build_id_equal(dso, build_id)) {
+			dso->load_errno = DSO_LOAD_ERRNO__MISMATCHING_BUILDID;
+			goto out_elf_end;
+		}
+	}
+
+	ss->is_64_bit = (gelf_getclass(elf) == ELFCLASS64);
+
+	ss->symtab = elf_section_by_name(elf, &ehdr, &ss->symshdr, ".symtab",
+			NULL);
+	if (ss->symshdr.sh_type != SHT_SYMTAB)
+		ss->symtab = NULL;
+
+	ss->dynsym_idx = 0;
+	ss->dynsym = elf_section_by_name(elf, &ehdr, &ss->dynshdr, ".dynsym",
+			&ss->dynsym_idx);
+	if (ss->dynshdr.sh_type != SHT_DYNSYM)
+		ss->dynsym = NULL;
+
+	ss->opdidx = 0;
+	ss->opdsec = elf_section_by_name(elf, &ehdr, &ss->opdshdr, ".opd",
+			&ss->opdidx);
+	if (ss->opdshdr.sh_type != SHT_PROGBITS)
+		ss->opdsec = NULL;
+
+	if (dso->kernel == DSO_TYPE_USER) {
+		GElf_Shdr shdr;
+		ss->adjust_symbols = (ehdr.e_type == ET_EXEC ||
+				ehdr.e_type == ET_REL ||
+				dso__is_vdso(dso) ||
+				elf_section_by_name(elf, &ehdr, &shdr,
+						     ".gnu.prelink_undo",
+						     NULL) != NULL);
+	} else {
+		ss->adjust_symbols = ehdr.e_type == ET_EXEC ||
+				     ehdr.e_type == ET_REL;
+	}
+
+	ss->name   = strdup(name);
+	if (!ss->name) {
+		dso->load_errno = errno;
+		goto out_elf_end;
+	}
+
+	ss->elf    = elf;
+	ss->fd     = fd;
+	ss->ehdr   = ehdr;
+	ss->type   = type;
+
+	return 0;
+
+out_elf_end:
+	elf_end(elf);
+out_close:
+	close(fd);
+	return err;
+}
+
+/**
+ * ref_reloc_sym_not_found - has kernel relocation symbol been found.
+ * @kmap: kernel maps and relocation reference symbol
+ *
+ * This function returns %true if we are dealing with the kernel maps and the
+ * relocation reference symbol has not yet been found.  Otherwise %false is
+ * returned.
+ */
+static bool ref_reloc_sym_not_found(struct kmap *kmap)
+{
+	return kmap && kmap->ref_reloc_sym && kmap->ref_reloc_sym->name &&
+	       !kmap->ref_reloc_sym->unrelocated_addr;
+}
+
+/**
+ * ref_reloc - kernel relocation offset.
+ * @kmap: kernel maps and relocation reference symbol
+ *
+ * This function returns the offset of kernel addresses as determined by using
+ * the relocation reference symbol i.e. if the kernel has not been relocated
+ * then the return value is zero.
+ */
+static u64 ref_reloc(struct kmap *kmap)
+{
+	if (kmap && kmap->ref_reloc_sym &&
+	    kmap->ref_reloc_sym->unrelocated_addr)
+		return kmap->ref_reloc_sym->addr -
+		       kmap->ref_reloc_sym->unrelocated_addr;
+	return 0;
+}
+
+static bool want_demangle(bool is_kernel_sym)
+{
+	return is_kernel_sym ? symbol_conf.demangle_kernel : symbol_conf.demangle;
+}
+
+int dso__load_sym(struct dso *dso, struct map *map,
+		  struct symsrc *syms_ss, struct symsrc *runtime_ss,
+		  symbol_filter_t filter, int kmodule)
+{
+	struct kmap *kmap = dso->kernel ? map__kmap(map) : NULL;
+	struct map_groups *kmaps = kmap ? map__kmaps(map) : NULL;
+	struct map *curr_map = map;
+	struct dso *curr_dso = dso;
+	Elf_Data *symstrs, *secstrs;
+	uint32_t nr_syms;
+	int err = -1;
+	uint32_t idx;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+	Elf_Data *syms, *opddata = NULL;
+	GElf_Sym sym;
+	Elf_Scn *sec, *sec_strndx;
+	Elf *elf;
+	int nr = 0;
+	bool remap_kernel = false, adjust_kernel_syms = false;
+
+	if (kmap && !kmaps)
+		return -1;
+
+	dso->symtab_type = syms_ss->type;
+	dso->is_64_bit = syms_ss->is_64_bit;
+	dso->rel = syms_ss->ehdr.e_type == ET_REL;
+
+	/*
+	 * Modules may already have symbols from kallsyms, but those symbols
+	 * have the wrong values for the dso maps, so remove them.
+	 */
+	if (kmodule && syms_ss->symtab)
+		symbols__delete(&dso->symbols[map->type]);
+
+	if (!syms_ss->symtab) {
+		/*
+		 * If the vmlinux is stripped, fail so we will fall back
+		 * to using kallsyms. The vmlinux runtime symbols aren't
+		 * of much use.
+		 */
+		if (dso->kernel)
+			goto out_elf_end;
+
+		syms_ss->symtab  = syms_ss->dynsym;
+		syms_ss->symshdr = syms_ss->dynshdr;
+	}
+
+	elf = syms_ss->elf;
+	ehdr = syms_ss->ehdr;
+	sec = syms_ss->symtab;
+	shdr = syms_ss->symshdr;
+
+	if (runtime_ss->opdsec)
+		opddata = elf_rawdata(runtime_ss->opdsec, NULL);
+
+	syms = elf_getdata(sec, NULL);
+	if (syms == NULL)
+		goto out_elf_end;
+
+	sec = elf_getscn(elf, shdr.sh_link);
+	if (sec == NULL)
+		goto out_elf_end;
+
+	symstrs = elf_getdata(sec, NULL);
+	if (symstrs == NULL)
+		goto out_elf_end;
+
+	sec_strndx = elf_getscn(runtime_ss->elf, runtime_ss->ehdr.e_shstrndx);
+	if (sec_strndx == NULL)
+		goto out_elf_end;
+
+	secstrs = elf_getdata(sec_strndx, NULL);
+	if (secstrs == NULL)
+		goto out_elf_end;
+
+	nr_syms = shdr.sh_size / shdr.sh_entsize;
+
+	memset(&sym, 0, sizeof(sym));
+
+	/*
+	 * The kernel relocation symbol is needed in advance in order to adjust
+	 * kernel maps correctly.
+	 */
+	if (ref_reloc_sym_not_found(kmap)) {
+		elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) {
+			const char *elf_name = elf_sym__name(&sym, symstrs);
+
+			if (strcmp(elf_name, kmap->ref_reloc_sym->name))
+				continue;
+			kmap->ref_reloc_sym->unrelocated_addr = sym.st_value;
+			map->reloc = kmap->ref_reloc_sym->addr -
+				     kmap->ref_reloc_sym->unrelocated_addr;
+			break;
+		}
+	}
+
+	dso->adjust_symbols = runtime_ss->adjust_symbols || ref_reloc(kmap);
+	/*
+	 * Initial kernel and module mappings do not map to the dso.  For
+	 * function mappings, flag the fixups.
+	 */
+	if (map->type == MAP__FUNCTION && (dso->kernel || kmodule)) {
+		remap_kernel = true;
+		adjust_kernel_syms = dso->adjust_symbols;
+	}
+	elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) {
+		struct symbol *f;
+		const char *elf_name = elf_sym__name(&sym, symstrs);
+		char *demangled = NULL;
+		int is_label = elf_sym__is_label(&sym);
+		const char *section_name;
+		bool used_opd = false;
+
+		if (!is_label && !elf_sym__is_a(&sym, map->type))
+			continue;
+
+		/* Reject ARM ELF "mapping symbols": these aren't unique and
+		 * don't identify functions, so will confuse the profile
+		 * output: */
+		if (ehdr.e_machine == EM_ARM || ehdr.e_machine == EM_AARCH64) {
+			if (elf_name[0] == '$' && strchr("adtx", elf_name[1])
+			    && (elf_name[2] == '\0' || elf_name[2] == '.'))
+				continue;
+		}
+
+		if (runtime_ss->opdsec && sym.st_shndx == runtime_ss->opdidx) {
+			u32 offset = sym.st_value - syms_ss->opdshdr.sh_addr;
+			u64 *opd = opddata->d_buf + offset;
+			sym.st_value = DSO__SWAP(dso, u64, *opd);
+			sym.st_shndx = elf_addr_to_index(runtime_ss->elf,
+					sym.st_value);
+			used_opd = true;
+		}
+		/*
+		 * When loading symbols in a data mapping, ABS symbols (which
+		 * has a value of SHN_ABS in its st_shndx) failed at
+		 * elf_getscn().  And it marks the loading as a failure so
+		 * already loaded symbols cannot be fixed up.
+		 *
+		 * I'm not sure what should be done. Just ignore them for now.
+		 * - Namhyung Kim
+		 */
+		if (sym.st_shndx == SHN_ABS)
+			continue;
+
+		sec = elf_getscn(runtime_ss->elf, sym.st_shndx);
+		if (!sec)
+			goto out_elf_end;
+
+		gelf_getshdr(sec, &shdr);
+
+		if (is_label && !elf_sec__is_a(&shdr, secstrs, map->type))
+			continue;
+
+		section_name = elf_sec__name(&shdr, secstrs);
+
+		/* On ARM, symbols for thumb functions have 1 added to
+		 * the symbol address as a flag - remove it */
+		if ((ehdr.e_machine == EM_ARM) &&
+		    (map->type == MAP__FUNCTION) &&
+		    (sym.st_value & 1))
+			--sym.st_value;
+
+		if (dso->kernel || kmodule) {
+			char dso_name[PATH_MAX];
+
+			/* Adjust symbol to map to file offset */
+			if (adjust_kernel_syms)
+				sym.st_value -= shdr.sh_addr - shdr.sh_offset;
+
+			if (strcmp(section_name,
+				   (curr_dso->short_name +
+				    dso->short_name_len)) == 0)
+				goto new_symbol;
+
+			if (strcmp(section_name, ".text") == 0) {
+				/*
+				 * The initial kernel mapping is based on
+				 * kallsyms and identity maps.  Overwrite it to
+				 * map to the kernel dso.
+				 */
+				if (remap_kernel && dso->kernel) {
+					remap_kernel = false;
+					map->start = shdr.sh_addr +
+						     ref_reloc(kmap);
+					map->end = map->start + shdr.sh_size;
+					map->pgoff = shdr.sh_offset;
+					map->map_ip = map__map_ip;
+					map->unmap_ip = map__unmap_ip;
+					/* Ensure maps are correctly ordered */
+					if (kmaps) {
+						map_groups__remove(kmaps, map);
+						map_groups__insert(kmaps, map);
+					}
+				}
+
+				/*
+				 * The initial module mapping is based on
+				 * /proc/modules mapped to offset zero.
+				 * Overwrite it to map to the module dso.
+				 */
+				if (remap_kernel && kmodule) {
+					remap_kernel = false;
+					map->pgoff = shdr.sh_offset;
+				}
+
+				curr_map = map;
+				curr_dso = dso;
+				goto new_symbol;
+			}
+
+			if (!kmap)
+				goto new_symbol;
+
+			snprintf(dso_name, sizeof(dso_name),
+				 "%s%s", dso->short_name, section_name);
+
+			curr_map = map_groups__find_by_name(kmaps, map->type, dso_name);
+			if (curr_map == NULL) {
+				u64 start = sym.st_value;
+
+				if (kmodule)
+					start += map->start + shdr.sh_offset;
+
+				curr_dso = dso__new(dso_name);
+				if (curr_dso == NULL)
+					goto out_elf_end;
+				curr_dso->kernel = dso->kernel;
+				curr_dso->long_name = dso->long_name;
+				curr_dso->long_name_len = dso->long_name_len;
+				curr_map = map__new2(start, curr_dso,
+						     map->type);
+				if (curr_map == NULL) {
+					dso__delete(curr_dso);
+					goto out_elf_end;
+				}
+				if (adjust_kernel_syms) {
+					curr_map->start = shdr.sh_addr +
+							  ref_reloc(kmap);
+					curr_map->end = curr_map->start +
+							shdr.sh_size;
+					curr_map->pgoff = shdr.sh_offset;
+				} else {
+					curr_map->map_ip = identity__map_ip;
+					curr_map->unmap_ip = identity__map_ip;
+				}
+				curr_dso->symtab_type = dso->symtab_type;
+				map_groups__insert(kmaps, curr_map);
+				/*
+				 * The new DSO should go to the kernel DSOS
+				 */
+				dsos__add(&map->groups->machine->kernel_dsos,
+					  curr_dso);
+				dso__set_loaded(curr_dso, map->type);
+			} else
+				curr_dso = curr_map->dso;
+
+			goto new_symbol;
+		}
+
+		if ((used_opd && runtime_ss->adjust_symbols)
+				|| (!used_opd && syms_ss->adjust_symbols)) {
+			pr_debug4("%s: adjusting symbol: st_value: %#" PRIx64 " "
+				  "sh_addr: %#" PRIx64 " sh_offset: %#" PRIx64 "\n", __func__,
+				  (u64)sym.st_value, (u64)shdr.sh_addr,
+				  (u64)shdr.sh_offset);
+			sym.st_value -= shdr.sh_addr - shdr.sh_offset;
+		}
+new_symbol:
+		/*
+		 * We need to figure out if the object was created from C++ sources
+		 * DWARF DW_compile_unit has this, but we don't always have access
+		 * to it...
+		 */
+		if (want_demangle(dso->kernel || kmodule)) {
+			int demangle_flags = DMGL_NO_OPTS;
+			if (verbose)
+				demangle_flags = DMGL_PARAMS | DMGL_ANSI;
+
+			demangled = bfd_demangle(NULL, elf_name, demangle_flags);
+			if (demangled != NULL)
+				elf_name = demangled;
+		}
+		f = symbol__new(sym.st_value, sym.st_size,
+				GELF_ST_BIND(sym.st_info), elf_name);
+		free(demangled);
+		if (!f)
+			goto out_elf_end;
+
+		if (filter && filter(curr_map, f))
+			symbol__delete(f);
+		else {
+			symbols__insert(&curr_dso->symbols[curr_map->type], f);
+			nr++;
+		}
+	}
+
+	/*
+	 * For misannotated, zeroed, ASM function sizes.
+	 */
+	if (nr > 0) {
+		if (!symbol_conf.allow_aliases)
+			symbols__fixup_duplicate(&dso->symbols[map->type]);
+		symbols__fixup_end(&dso->symbols[map->type]);
+		if (kmap) {
+			/*
+			 * We need to fixup this here too because we create new
+			 * maps here, for things like vsyscall sections.
+			 */
+			__map_groups__fixup_end(kmaps, map->type);
+		}
+	}
+	err = nr;
+out_elf_end:
+	return err;
+}
+
+static int elf_read_maps(Elf *elf, bool exe, mapfn_t mapfn, void *data)
+{
+	GElf_Phdr phdr;
+	size_t i, phdrnum;
+	int err;
+	u64 sz;
+
+	if (elf_getphdrnum(elf, &phdrnum))
+		return -1;
+
+	for (i = 0; i < phdrnum; i++) {
+		if (gelf_getphdr(elf, i, &phdr) == NULL)
+			return -1;
+		if (phdr.p_type != PT_LOAD)
+			continue;
+		if (exe) {
+			if (!(phdr.p_flags & PF_X))
+				continue;
+		} else {
+			if (!(phdr.p_flags & PF_R))
+				continue;
+		}
+		sz = min(phdr.p_memsz, phdr.p_filesz);
+		if (!sz)
+			continue;
+		err = mapfn(phdr.p_vaddr, sz, phdr.p_offset, data);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+int file__read_maps(int fd, bool exe, mapfn_t mapfn, void *data,
+		    bool *is_64_bit)
+{
+	int err;
+	Elf *elf;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL)
+		return -1;
+
+	if (is_64_bit)
+		*is_64_bit = (gelf_getclass(elf) == ELFCLASS64);
+
+	err = elf_read_maps(elf, exe, mapfn, data);
+
+	elf_end(elf);
+	return err;
+}
+
+enum dso_type dso__type_fd(int fd)
+{
+	enum dso_type dso_type = DSO__TYPE_UNKNOWN;
+	GElf_Ehdr ehdr;
+	Elf_Kind ek;
+	Elf *elf;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL)
+		goto out;
+
+	ek = elf_kind(elf);
+	if (ek != ELF_K_ELF)
+		goto out_end;
+
+	if (gelf_getclass(elf) == ELFCLASS64) {
+		dso_type = DSO__TYPE_64BIT;
+		goto out_end;
+	}
+
+	if (gelf_getehdr(elf, &ehdr) == NULL)
+		goto out_end;
+
+	if (ehdr.e_machine == EM_X86_64)
+		dso_type = DSO__TYPE_X32BIT;
+	else
+		dso_type = DSO__TYPE_32BIT;
+out_end:
+	elf_end(elf);
+out:
+	return dso_type;
+}
+
+static int copy_bytes(int from, off_t from_offs, int to, off_t to_offs, u64 len)
+{
+	ssize_t r;
+	size_t n;
+	int err = -1;
+	char *buf = malloc(page_size);
+
+	if (buf == NULL)
+		return -1;
+
+	if (lseek(to, to_offs, SEEK_SET) != to_offs)
+		goto out;
+
+	if (lseek(from, from_offs, SEEK_SET) != from_offs)
+		goto out;
+
+	while (len) {
+		n = page_size;
+		if (len < n)
+			n = len;
+		/* Use read because mmap won't work on proc files */
+		r = read(from, buf, n);
+		if (r < 0)
+			goto out;
+		if (!r)
+			break;
+		n = r;
+		r = write(to, buf, n);
+		if (r < 0)
+			goto out;
+		if ((size_t)r != n)
+			goto out;
+		len -= n;
+	}
+
+	err = 0;
+out:
+	free(buf);
+	return err;
+}
+
+struct kcore {
+	int fd;
+	int elfclass;
+	Elf *elf;
+	GElf_Ehdr ehdr;
+};
+
+static int kcore__open(struct kcore *kcore, const char *filename)
+{
+	GElf_Ehdr *ehdr;
+
+	kcore->fd = open(filename, O_RDONLY);
+	if (kcore->fd == -1)
+		return -1;
+
+	kcore->elf = elf_begin(kcore->fd, ELF_C_READ, NULL);
+	if (!kcore->elf)
+		goto out_close;
+
+	kcore->elfclass = gelf_getclass(kcore->elf);
+	if (kcore->elfclass == ELFCLASSNONE)
+		goto out_end;
+
+	ehdr = gelf_getehdr(kcore->elf, &kcore->ehdr);
+	if (!ehdr)
+		goto out_end;
+
+	return 0;
+
+out_end:
+	elf_end(kcore->elf);
+out_close:
+	close(kcore->fd);
+	return -1;
+}
+
+static int kcore__init(struct kcore *kcore, char *filename, int elfclass,
+		       bool temp)
+{
+	GElf_Ehdr *ehdr;
+
+	kcore->elfclass = elfclass;
+
+	if (temp)
+		kcore->fd = mkstemp(filename);
+	else
+		kcore->fd = open(filename, O_WRONLY | O_CREAT | O_EXCL, 0400);
+	if (kcore->fd == -1)
+		return -1;
+
+	kcore->elf = elf_begin(kcore->fd, ELF_C_WRITE, NULL);
+	if (!kcore->elf)
+		goto out_close;
+
+	if (!gelf_newehdr(kcore->elf, elfclass))
+		goto out_end;
+
+	ehdr = gelf_getehdr(kcore->elf, &kcore->ehdr);
+	if (!ehdr)
+		goto out_end;
+
+	return 0;
+
+out_end:
+	elf_end(kcore->elf);
+out_close:
+	close(kcore->fd);
+	unlink(filename);
+	return -1;
+}
+
+static void kcore__close(struct kcore *kcore)
+{
+	elf_end(kcore->elf);
+	close(kcore->fd);
+}
+
+static int kcore__copy_hdr(struct kcore *from, struct kcore *to, size_t count)
+{
+	GElf_Ehdr *ehdr = &to->ehdr;
+	GElf_Ehdr *kehdr = &from->ehdr;
+
+	memcpy(ehdr->e_ident, kehdr->e_ident, EI_NIDENT);
+	ehdr->e_type      = kehdr->e_type;
+	ehdr->e_machine   = kehdr->e_machine;
+	ehdr->e_version   = kehdr->e_version;
+	ehdr->e_entry     = 0;
+	ehdr->e_shoff     = 0;
+	ehdr->e_flags     = kehdr->e_flags;
+	ehdr->e_phnum     = count;
+	ehdr->e_shentsize = 0;
+	ehdr->e_shnum     = 0;
+	ehdr->e_shstrndx  = 0;
+
+	if (from->elfclass == ELFCLASS32) {
+		ehdr->e_phoff     = sizeof(Elf32_Ehdr);
+		ehdr->e_ehsize    = sizeof(Elf32_Ehdr);
+		ehdr->e_phentsize = sizeof(Elf32_Phdr);
+	} else {
+		ehdr->e_phoff     = sizeof(Elf64_Ehdr);
+		ehdr->e_ehsize    = sizeof(Elf64_Ehdr);
+		ehdr->e_phentsize = sizeof(Elf64_Phdr);
+	}
+
+	if (!gelf_update_ehdr(to->elf, ehdr))
+		return -1;
+
+	if (!gelf_newphdr(to->elf, count))
+		return -1;
+
+	return 0;
+}
+
+static int kcore__add_phdr(struct kcore *kcore, int idx, off_t offset,
+			   u64 addr, u64 len)
+{
+	GElf_Phdr gphdr;
+	GElf_Phdr *phdr;
+
+	phdr = gelf_getphdr(kcore->elf, idx, &gphdr);
+	if (!phdr)
+		return -1;
+
+	phdr->p_type	= PT_LOAD;
+	phdr->p_flags	= PF_R | PF_W | PF_X;
+	phdr->p_offset	= offset;
+	phdr->p_vaddr	= addr;
+	phdr->p_paddr	= 0;
+	phdr->p_filesz	= len;
+	phdr->p_memsz	= len;
+	phdr->p_align	= page_size;
+
+	if (!gelf_update_phdr(kcore->elf, idx, phdr))
+		return -1;
+
+	return 0;
+}
+
+static off_t kcore__write(struct kcore *kcore)
+{
+	return elf_update(kcore->elf, ELF_C_WRITE);
+}
+
+struct phdr_data {
+	off_t offset;
+	u64 addr;
+	u64 len;
+};
+
+struct kcore_copy_info {
+	u64 stext;
+	u64 etext;
+	u64 first_symbol;
+	u64 last_symbol;
+	u64 first_module;
+	u64 last_module_symbol;
+	struct phdr_data kernel_map;
+	struct phdr_data modules_map;
+};
+
+static int kcore_copy__process_kallsyms(void *arg, const char *name, char type,
+					u64 start)
+{
+	struct kcore_copy_info *kci = arg;
+
+	if (!symbol_type__is_a(type, MAP__FUNCTION))
+		return 0;
+
+	if (strchr(name, '[')) {
+		if (start > kci->last_module_symbol)
+			kci->last_module_symbol = start;
+		return 0;
+	}
+
+	if (!kci->first_symbol || start < kci->first_symbol)
+		kci->first_symbol = start;
+
+	if (!kci->last_symbol || start > kci->last_symbol)
+		kci->last_symbol = start;
+
+	if (!strcmp(name, "_stext")) {
+		kci->stext = start;
+		return 0;
+	}
+
+	if (!strcmp(name, "_etext")) {
+		kci->etext = start;
+		return 0;
+	}
+
+	return 0;
+}
+
+static int kcore_copy__parse_kallsyms(struct kcore_copy_info *kci,
+				      const char *dir)
+{
+	char kallsyms_filename[PATH_MAX];
+
+	scnprintf(kallsyms_filename, PATH_MAX, "%s/kallsyms", dir);
+
+	if (symbol__restricted_filename(kallsyms_filename, "/proc/kallsyms"))
+		return -1;
+
+	if (kallsyms__parse(kallsyms_filename, kci,
+			    kcore_copy__process_kallsyms) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int kcore_copy__process_modules(void *arg,
+				       const char *name __maybe_unused,
+				       u64 start)
+{
+	struct kcore_copy_info *kci = arg;
+
+	if (!kci->first_module || start < kci->first_module)
+		kci->first_module = start;
+
+	return 0;
+}
+
+static int kcore_copy__parse_modules(struct kcore_copy_info *kci,
+				     const char *dir)
+{
+	char modules_filename[PATH_MAX];
+
+	scnprintf(modules_filename, PATH_MAX, "%s/modules", dir);
+
+	if (symbol__restricted_filename(modules_filename, "/proc/modules"))
+		return -1;
+
+	if (modules__parse(modules_filename, kci,
+			   kcore_copy__process_modules) < 0)
+		return -1;
+
+	return 0;
+}
+
+static void kcore_copy__map(struct phdr_data *p, u64 start, u64 end, u64 pgoff,
+			    u64 s, u64 e)
+{
+	if (p->addr || s < start || s >= end)
+		return;
+
+	p->addr = s;
+	p->offset = (s - start) + pgoff;
+	p->len = e < end ? e - s : end - s;
+}
+
+static int kcore_copy__read_map(u64 start, u64 len, u64 pgoff, void *data)
+{
+	struct kcore_copy_info *kci = data;
+	u64 end = start + len;
+
+	kcore_copy__map(&kci->kernel_map, start, end, pgoff, kci->stext,
+			kci->etext);
+
+	kcore_copy__map(&kci->modules_map, start, end, pgoff, kci->first_module,
+			kci->last_module_symbol);
+
+	return 0;
+}
+
+static int kcore_copy__read_maps(struct kcore_copy_info *kci, Elf *elf)
+{
+	if (elf_read_maps(elf, true, kcore_copy__read_map, kci) < 0)
+		return -1;
+
+	return 0;
+}
+
+static int kcore_copy__calc_maps(struct kcore_copy_info *kci, const char *dir,
+				 Elf *elf)
+{
+	if (kcore_copy__parse_kallsyms(kci, dir))
+		return -1;
+
+	if (kcore_copy__parse_modules(kci, dir))
+		return -1;
+
+	if (kci->stext)
+		kci->stext = round_down(kci->stext, page_size);
+	else
+		kci->stext = round_down(kci->first_symbol, page_size);
+
+	if (kci->etext) {
+		kci->etext = round_up(kci->etext, page_size);
+	} else if (kci->last_symbol) {
+		kci->etext = round_up(kci->last_symbol, page_size);
+		kci->etext += page_size;
+	}
+
+	kci->first_module = round_down(kci->first_module, page_size);
+
+	if (kci->last_module_symbol) {
+		kci->last_module_symbol = round_up(kci->last_module_symbol,
+						   page_size);
+		kci->last_module_symbol += page_size;
+	}
+
+	if (!kci->stext || !kci->etext)
+		return -1;
+
+	if (kci->first_module && !kci->last_module_symbol)
+		return -1;
+
+	return kcore_copy__read_maps(kci, elf);
+}
+
+static int kcore_copy__copy_file(const char *from_dir, const char *to_dir,
+				 const char *name)
+{
+	char from_filename[PATH_MAX];
+	char to_filename[PATH_MAX];
+
+	scnprintf(from_filename, PATH_MAX, "%s/%s", from_dir, name);
+	scnprintf(to_filename, PATH_MAX, "%s/%s", to_dir, name);
+
+	return copyfile_mode(from_filename, to_filename, 0400);
+}
+
+static int kcore_copy__unlink(const char *dir, const char *name)
+{
+	char filename[PATH_MAX];
+
+	scnprintf(filename, PATH_MAX, "%s/%s", dir, name);
+
+	return unlink(filename);
+}
+
+static int kcore_copy__compare_fds(int from, int to)
+{
+	char *buf_from;
+	char *buf_to;
+	ssize_t ret;
+	size_t len;
+	int err = -1;
+
+	buf_from = malloc(page_size);
+	buf_to = malloc(page_size);
+	if (!buf_from || !buf_to)
+		goto out;
+
+	while (1) {
+		/* Use read because mmap won't work on proc files */
+		ret = read(from, buf_from, page_size);
+		if (ret < 0)
+			goto out;
+
+		if (!ret)
+			break;
+
+		len = ret;
+
+		if (readn(to, buf_to, len) != (int)len)
+			goto out;
+
+		if (memcmp(buf_from, buf_to, len))
+			goto out;
+	}
+
+	err = 0;
+out:
+	free(buf_to);
+	free(buf_from);
+	return err;
+}
+
+static int kcore_copy__compare_files(const char *from_filename,
+				     const char *to_filename)
+{
+	int from, to, err = -1;
+
+	from = open(from_filename, O_RDONLY);
+	if (from < 0)
+		return -1;
+
+	to = open(to_filename, O_RDONLY);
+	if (to < 0)
+		goto out_close_from;
+
+	err = kcore_copy__compare_fds(from, to);
+
+	close(to);
+out_close_from:
+	close(from);
+	return err;
+}
+
+static int kcore_copy__compare_file(const char *from_dir, const char *to_dir,
+				    const char *name)
+{
+	char from_filename[PATH_MAX];
+	char to_filename[PATH_MAX];
+
+	scnprintf(from_filename, PATH_MAX, "%s/%s", from_dir, name);
+	scnprintf(to_filename, PATH_MAX, "%s/%s", to_dir, name);
+
+	return kcore_copy__compare_files(from_filename, to_filename);
+}
+
+/**
+ * kcore_copy - copy kallsyms, modules and kcore from one directory to another.
+ * @from_dir: from directory
+ * @to_dir: to directory
+ *
+ * This function copies kallsyms, modules and kcore files from one directory to
+ * another.  kallsyms and modules are copied entirely.  Only code segments are
+ * copied from kcore.  It is assumed that two segments suffice: one for the
+ * kernel proper and one for all the modules.  The code segments are determined
+ * from kallsyms and modules files.  The kernel map starts at _stext or the
+ * lowest function symbol, and ends at _etext or the highest function symbol.
+ * The module map starts at the lowest module address and ends at the highest
+ * module symbol.  Start addresses are rounded down to the nearest page.  End
+ * addresses are rounded up to the nearest page.  An extra page is added to the
+ * highest kernel symbol and highest module symbol to, hopefully, encompass that
+ * symbol too.  Because it contains only code sections, the resulting kcore is
+ * unusual.  One significant peculiarity is that the mapping (start -> pgoff)
+ * is not the same for the kernel map and the modules map.  That happens because
+ * the data is copied adjacently whereas the original kcore has gaps.  Finally,
+ * kallsyms and modules files are compared with their copies to check that
+ * modules have not been loaded or unloaded while the copies were taking place.
+ *
+ * Return: %0 on success, %-1 on failure.
+ */
+int kcore_copy(const char *from_dir, const char *to_dir)
+{
+	struct kcore kcore;
+	struct kcore extract;
+	size_t count = 2;
+	int idx = 0, err = -1;
+	off_t offset = page_size, sz, modules_offset = 0;
+	struct kcore_copy_info kci = { .stext = 0, };
+	char kcore_filename[PATH_MAX];
+	char extract_filename[PATH_MAX];
+
+	if (kcore_copy__copy_file(from_dir, to_dir, "kallsyms"))
+		return -1;
+
+	if (kcore_copy__copy_file(from_dir, to_dir, "modules"))
+		goto out_unlink_kallsyms;
+
+	scnprintf(kcore_filename, PATH_MAX, "%s/kcore", from_dir);
+	scnprintf(extract_filename, PATH_MAX, "%s/kcore", to_dir);
+
+	if (kcore__open(&kcore, kcore_filename))
+		goto out_unlink_modules;
+
+	if (kcore_copy__calc_maps(&kci, from_dir, kcore.elf))
+		goto out_kcore_close;
+
+	if (kcore__init(&extract, extract_filename, kcore.elfclass, false))
+		goto out_kcore_close;
+
+	if (!kci.modules_map.addr)
+		count -= 1;
+
+	if (kcore__copy_hdr(&kcore, &extract, count))
+		goto out_extract_close;
+
+	if (kcore__add_phdr(&extract, idx++, offset, kci.kernel_map.addr,
+			    kci.kernel_map.len))
+		goto out_extract_close;
+
+	if (kci.modules_map.addr) {
+		modules_offset = offset + kci.kernel_map.len;
+		if (kcore__add_phdr(&extract, idx, modules_offset,
+				    kci.modules_map.addr, kci.modules_map.len))
+			goto out_extract_close;
+	}
+
+	sz = kcore__write(&extract);
+	if (sz < 0 || sz > offset)
+		goto out_extract_close;
+
+	if (copy_bytes(kcore.fd, kci.kernel_map.offset, extract.fd, offset,
+		       kci.kernel_map.len))
+		goto out_extract_close;
+
+	if (modules_offset && copy_bytes(kcore.fd, kci.modules_map.offset,
+					 extract.fd, modules_offset,
+					 kci.modules_map.len))
+		goto out_extract_close;
+
+	if (kcore_copy__compare_file(from_dir, to_dir, "modules"))
+		goto out_extract_close;
+
+	if (kcore_copy__compare_file(from_dir, to_dir, "kallsyms"))
+		goto out_extract_close;
+
+	err = 0;
+
+out_extract_close:
+	kcore__close(&extract);
+	if (err)
+		unlink(extract_filename);
+out_kcore_close:
+	kcore__close(&kcore);
+out_unlink_modules:
+	if (err)
+		kcore_copy__unlink(to_dir, "modules");
+out_unlink_kallsyms:
+	if (err)
+		kcore_copy__unlink(to_dir, "kallsyms");
+
+	return err;
+}
+
+int kcore_extract__create(struct kcore_extract *kce)
+{
+	struct kcore kcore;
+	struct kcore extract;
+	size_t count = 1;
+	int idx = 0, err = -1;
+	off_t offset = page_size, sz;
+
+	if (kcore__open(&kcore, kce->kcore_filename))
+		return -1;
+
+	strcpy(kce->extract_filename, PERF_KCORE_EXTRACT);
+	if (kcore__init(&extract, kce->extract_filename, kcore.elfclass, true))
+		goto out_kcore_close;
+
+	if (kcore__copy_hdr(&kcore, &extract, count))
+		goto out_extract_close;
+
+	if (kcore__add_phdr(&extract, idx, offset, kce->addr, kce->len))
+		goto out_extract_close;
+
+	sz = kcore__write(&extract);
+	if (sz < 0 || sz > offset)
+		goto out_extract_close;
+
+	if (copy_bytes(kcore.fd, kce->offs, extract.fd, offset, kce->len))
+		goto out_extract_close;
+
+	err = 0;
+
+out_extract_close:
+	kcore__close(&extract);
+	if (err)
+		unlink(kce->extract_filename);
+out_kcore_close:
+	kcore__close(&kcore);
+
+	return err;
+}
+
+void kcore_extract__delete(struct kcore_extract *kce)
+{
+	unlink(kce->extract_filename);
+}
+
+void symbol__elf_init(void)
+{
+	elf_version(EV_CURRENT);
+}
diff --git a/tools/perf/util/symbol-minimal.c b/tools/perf/util/symbol-minimal.c
new file mode 100644
index 000000000..fd8477cac
--- /dev/null
+++ b/tools/perf/util/symbol-minimal.c
@@ -0,0 +1,377 @@
+#include "symbol.h"
+#include "util.h"
+
+#include <stdio.h>
+#include <fcntl.h>
+#include <string.h>
+#include <byteswap.h>
+#include <sys/stat.h>
+
+
+static bool check_need_swap(int file_endian)
+{
+	const int data = 1;
+	u8 *check = (u8 *)&data;
+	int host_endian;
+
+	if (check[0] == 1)
+		host_endian = ELFDATA2LSB;
+	else
+		host_endian = ELFDATA2MSB;
+
+	return host_endian != file_endian;
+}
+
+#define NOTE_ALIGN(sz) (((sz) + 3) & ~3)
+
+#define NT_GNU_BUILD_ID	3
+
+static int read_build_id(void *note_data, size_t note_len, void *bf,
+			 size_t size, bool need_swap)
+{
+	struct {
+		u32 n_namesz;
+		u32 n_descsz;
+		u32 n_type;
+	} *nhdr;
+	void *ptr;
+
+	ptr = note_data;
+	while (ptr < (note_data + note_len)) {
+		const char *name;
+		size_t namesz, descsz;
+
+		nhdr = ptr;
+		if (need_swap) {
+			nhdr->n_namesz = bswap_32(nhdr->n_namesz);
+			nhdr->n_descsz = bswap_32(nhdr->n_descsz);
+			nhdr->n_type = bswap_32(nhdr->n_type);
+		}
+
+		namesz = NOTE_ALIGN(nhdr->n_namesz);
+		descsz = NOTE_ALIGN(nhdr->n_descsz);
+
+		ptr += sizeof(*nhdr);
+		name = ptr;
+		ptr += namesz;
+		if (nhdr->n_type == NT_GNU_BUILD_ID &&
+		    nhdr->n_namesz == sizeof("GNU")) {
+			if (memcmp(name, "GNU", sizeof("GNU")) == 0) {
+				size_t sz = min(size, descsz);
+				memcpy(bf, ptr, sz);
+				memset(bf + sz, 0, size - sz);
+				return 0;
+			}
+		}
+		ptr += descsz;
+	}
+
+	return -1;
+}
+
+int filename__read_debuglink(const char *filename __maybe_unused,
+			     char *debuglink __maybe_unused,
+			     size_t size __maybe_unused)
+{
+	return -1;
+}
+
+/*
+ * Just try PT_NOTE header otherwise fails
+ */
+int filename__read_build_id(const char *filename, void *bf, size_t size)
+{
+	FILE *fp;
+	int ret = -1;
+	bool need_swap = false;
+	u8 e_ident[EI_NIDENT];
+	size_t buf_size;
+	void *buf;
+	int i;
+
+	fp = fopen(filename, "r");
+	if (fp == NULL)
+		return -1;
+
+	if (fread(e_ident, sizeof(e_ident), 1, fp) != 1)
+		goto out;
+
+	if (memcmp(e_ident, ELFMAG, SELFMAG) ||
+	    e_ident[EI_VERSION] != EV_CURRENT)
+		goto out;
+
+	need_swap = check_need_swap(e_ident[EI_DATA]);
+
+	/* for simplicity */
+	fseek(fp, 0, SEEK_SET);
+
+	if (e_ident[EI_CLASS] == ELFCLASS32) {
+		Elf32_Ehdr ehdr;
+		Elf32_Phdr *phdr;
+
+		if (fread(&ehdr, sizeof(ehdr), 1, fp) != 1)
+			goto out;
+
+		if (need_swap) {
+			ehdr.e_phoff = bswap_32(ehdr.e_phoff);
+			ehdr.e_phentsize = bswap_16(ehdr.e_phentsize);
+			ehdr.e_phnum = bswap_16(ehdr.e_phnum);
+		}
+
+		buf_size = ehdr.e_phentsize * ehdr.e_phnum;
+		buf = malloc(buf_size);
+		if (buf == NULL)
+			goto out;
+
+		fseek(fp, ehdr.e_phoff, SEEK_SET);
+		if (fread(buf, buf_size, 1, fp) != 1)
+			goto out_free;
+
+		for (i = 0, phdr = buf; i < ehdr.e_phnum; i++, phdr++) {
+			void *tmp;
+			long offset;
+
+			if (need_swap) {
+				phdr->p_type = bswap_32(phdr->p_type);
+				phdr->p_offset = bswap_32(phdr->p_offset);
+				phdr->p_filesz = bswap_32(phdr->p_filesz);
+			}
+
+			if (phdr->p_type != PT_NOTE)
+				continue;
+
+			buf_size = phdr->p_filesz;
+			offset = phdr->p_offset;
+			tmp = realloc(buf, buf_size);
+			if (tmp == NULL)
+				goto out_free;
+
+			buf = tmp;
+			fseek(fp, offset, SEEK_SET);
+			if (fread(buf, buf_size, 1, fp) != 1)
+				goto out_free;
+
+			ret = read_build_id(buf, buf_size, bf, size, need_swap);
+			if (ret == 0)
+				ret = size;
+			break;
+		}
+	} else {
+		Elf64_Ehdr ehdr;
+		Elf64_Phdr *phdr;
+
+		if (fread(&ehdr, sizeof(ehdr), 1, fp) != 1)
+			goto out;
+
+		if (need_swap) {
+			ehdr.e_phoff = bswap_64(ehdr.e_phoff);
+			ehdr.e_phentsize = bswap_16(ehdr.e_phentsize);
+			ehdr.e_phnum = bswap_16(ehdr.e_phnum);
+		}
+
+		buf_size = ehdr.e_phentsize * ehdr.e_phnum;
+		buf = malloc(buf_size);
+		if (buf == NULL)
+			goto out;
+
+		fseek(fp, ehdr.e_phoff, SEEK_SET);
+		if (fread(buf, buf_size, 1, fp) != 1)
+			goto out_free;
+
+		for (i = 0, phdr = buf; i < ehdr.e_phnum; i++, phdr++) {
+			void *tmp;
+			long offset;
+
+			if (need_swap) {
+				phdr->p_type = bswap_32(phdr->p_type);
+				phdr->p_offset = bswap_64(phdr->p_offset);
+				phdr->p_filesz = bswap_64(phdr->p_filesz);
+			}
+
+			if (phdr->p_type != PT_NOTE)
+				continue;
+
+			buf_size = phdr->p_filesz;
+			offset = phdr->p_offset;
+			tmp = realloc(buf, buf_size);
+			if (tmp == NULL)
+				goto out_free;
+
+			buf = tmp;
+			fseek(fp, offset, SEEK_SET);
+			if (fread(buf, buf_size, 1, fp) != 1)
+				goto out_free;
+
+			ret = read_build_id(buf, buf_size, bf, size, need_swap);
+			if (ret == 0)
+				ret = size;
+			break;
+		}
+	}
+out_free:
+	free(buf);
+out:
+	fclose(fp);
+	return ret;
+}
+
+int sysfs__read_build_id(const char *filename, void *build_id, size_t size)
+{
+	int fd;
+	int ret = -1;
+	struct stat stbuf;
+	size_t buf_size;
+	void *buf;
+
+	fd = open(filename, O_RDONLY);
+	if (fd < 0)
+		return -1;
+
+	if (fstat(fd, &stbuf) < 0)
+		goto out;
+
+	buf_size = stbuf.st_size;
+	buf = malloc(buf_size);
+	if (buf == NULL)
+		goto out;
+
+	if (read(fd, buf, buf_size) != (ssize_t) buf_size)
+		goto out_free;
+
+	ret = read_build_id(buf, buf_size, build_id, size, false);
+out_free:
+	free(buf);
+out:
+	close(fd);
+	return ret;
+}
+
+int symsrc__init(struct symsrc *ss, struct dso *dso, const char *name,
+	         enum dso_binary_type type)
+{
+	int fd = open(name, O_RDONLY);
+	if (fd < 0)
+		goto out_errno;
+
+	ss->name = strdup(name);
+	if (!ss->name)
+		goto out_close;
+
+	ss->fd = fd;
+	ss->type = type;
+
+	return 0;
+out_close:
+	close(fd);
+out_errno:
+	dso->load_errno = errno;
+	return -1;
+}
+
+bool symsrc__possibly_runtime(struct symsrc *ss __maybe_unused)
+{
+	/* Assume all sym sources could be a runtime image. */
+	return true;
+}
+
+bool symsrc__has_symtab(struct symsrc *ss __maybe_unused)
+{
+	return false;
+}
+
+void symsrc__destroy(struct symsrc *ss)
+{
+	zfree(&ss->name);
+	close(ss->fd);
+}
+
+int dso__synthesize_plt_symbols(struct dso *dso __maybe_unused,
+				struct symsrc *ss __maybe_unused,
+				struct map *map __maybe_unused,
+				symbol_filter_t filter __maybe_unused)
+{
+	return 0;
+}
+
+static int fd__is_64_bit(int fd)
+{
+	u8 e_ident[EI_NIDENT];
+
+	if (lseek(fd, 0, SEEK_SET))
+		return -1;
+
+	if (readn(fd, e_ident, sizeof(e_ident)) != sizeof(e_ident))
+		return -1;
+
+	if (memcmp(e_ident, ELFMAG, SELFMAG) ||
+	    e_ident[EI_VERSION] != EV_CURRENT)
+		return -1;
+
+	return e_ident[EI_CLASS] == ELFCLASS64;
+}
+
+enum dso_type dso__type_fd(int fd)
+{
+	Elf64_Ehdr ehdr;
+	int ret;
+
+	ret = fd__is_64_bit(fd);
+	if (ret < 0)
+		return DSO__TYPE_UNKNOWN;
+
+	if (ret)
+		return DSO__TYPE_64BIT;
+
+	if (readn(fd, &ehdr, sizeof(ehdr)) != sizeof(ehdr))
+		return DSO__TYPE_UNKNOWN;
+
+	if (ehdr.e_machine == EM_X86_64)
+		return DSO__TYPE_X32BIT;
+
+	return DSO__TYPE_32BIT;
+}
+
+int dso__load_sym(struct dso *dso, struct map *map __maybe_unused,
+		  struct symsrc *ss,
+		  struct symsrc *runtime_ss __maybe_unused,
+		  symbol_filter_t filter __maybe_unused,
+		  int kmodule __maybe_unused)
+{
+	unsigned char *build_id[BUILD_ID_SIZE];
+	int ret;
+
+	ret = fd__is_64_bit(ss->fd);
+	if (ret >= 0)
+		dso->is_64_bit = ret;
+
+	if (filename__read_build_id(ss->name, build_id, BUILD_ID_SIZE) > 0) {
+		dso__set_build_id(dso, build_id);
+	}
+	return 0;
+}
+
+int file__read_maps(int fd __maybe_unused, bool exe __maybe_unused,
+		    mapfn_t mapfn __maybe_unused, void *data __maybe_unused,
+		    bool *is_64_bit __maybe_unused)
+{
+	return -1;
+}
+
+int kcore_extract__create(struct kcore_extract *kce __maybe_unused)
+{
+	return -1;
+}
+
+void kcore_extract__delete(struct kcore_extract *kce __maybe_unused)
+{
+}
+
+int kcore_copy(const char *from_dir __maybe_unused,
+	       const char *to_dir __maybe_unused)
+{
+	return -1;
+}
+
+void symbol__elf_init(void)
+{
+}
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
new file mode 100644
index 000000000..201f6c4ca
--- /dev/null
+++ b/tools/perf/util/symbol.c
@@ -0,0 +1,2001 @@
+#include <dirent.h>
+#include <errno.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/param.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <inttypes.h>
+#include "build-id.h"
+#include "util.h"
+#include "debug.h"
+#include "machine.h"
+#include "symbol.h"
+#include "strlist.h"
+#include "intlist.h"
+#include "header.h"
+
+#include <elf.h>
+#include <limits.h>
+#include <symbol/kallsyms.h>
+#include <sys/utsname.h>
+
+static int dso__load_kernel_sym(struct dso *dso, struct map *map,
+				symbol_filter_t filter);
+static int dso__load_guest_kernel_sym(struct dso *dso, struct map *map,
+			symbol_filter_t filter);
+int vmlinux_path__nr_entries;
+char **vmlinux_path;
+
+struct symbol_conf symbol_conf = {
+	.use_modules		= true,
+	.try_vmlinux_path	= true,
+	.annotate_src		= true,
+	.demangle		= true,
+	.demangle_kernel	= false,
+	.cumulate_callchain	= true,
+	.show_hist_headers	= true,
+	.symfs			= "",
+};
+
+static enum dso_binary_type binary_type_symtab[] = {
+	DSO_BINARY_TYPE__KALLSYMS,
+	DSO_BINARY_TYPE__GUEST_KALLSYMS,
+	DSO_BINARY_TYPE__JAVA_JIT,
+	DSO_BINARY_TYPE__DEBUGLINK,
+	DSO_BINARY_TYPE__BUILD_ID_CACHE,
+	DSO_BINARY_TYPE__FEDORA_DEBUGINFO,
+	DSO_BINARY_TYPE__UBUNTU_DEBUGINFO,
+	DSO_BINARY_TYPE__BUILDID_DEBUGINFO,
+	DSO_BINARY_TYPE__SYSTEM_PATH_DSO,
+	DSO_BINARY_TYPE__GUEST_KMODULE,
+	DSO_BINARY_TYPE__GUEST_KMODULE_COMP,
+	DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE,
+	DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE_COMP,
+	DSO_BINARY_TYPE__OPENEMBEDDED_DEBUGINFO,
+	DSO_BINARY_TYPE__NOT_FOUND,
+};
+
+#define DSO_BINARY_TYPE__SYMTAB_CNT ARRAY_SIZE(binary_type_symtab)
+
+bool symbol_type__is_a(char symbol_type, enum map_type map_type)
+{
+	symbol_type = toupper(symbol_type);
+
+	switch (map_type) {
+	case MAP__FUNCTION:
+		return symbol_type == 'T' || symbol_type == 'W';
+	case MAP__VARIABLE:
+		return symbol_type == 'D';
+	default:
+		return false;
+	}
+}
+
+static int prefix_underscores_count(const char *str)
+{
+	const char *tail = str;
+
+	while (*tail == '_')
+		tail++;
+
+	return tail - str;
+}
+
+#define SYMBOL_A 0
+#define SYMBOL_B 1
+
+static int choose_best_symbol(struct symbol *syma, struct symbol *symb)
+{
+	s64 a;
+	s64 b;
+	size_t na, nb;
+
+	/* Prefer a symbol with non zero length */
+	a = syma->end - syma->start;
+	b = symb->end - symb->start;
+	if ((b == 0) && (a > 0))
+		return SYMBOL_A;
+	else if ((a == 0) && (b > 0))
+		return SYMBOL_B;
+
+	/* Prefer a non weak symbol over a weak one */
+	a = syma->binding == STB_WEAK;
+	b = symb->binding == STB_WEAK;
+	if (b && !a)
+		return SYMBOL_A;
+	if (a && !b)
+		return SYMBOL_B;
+
+	/* Prefer a global symbol over a non global one */
+	a = syma->binding == STB_GLOBAL;
+	b = symb->binding == STB_GLOBAL;
+	if (a && !b)
+		return SYMBOL_A;
+	if (b && !a)
+		return SYMBOL_B;
+
+	/* Prefer a symbol with less underscores */
+	a = prefix_underscores_count(syma->name);
+	b = prefix_underscores_count(symb->name);
+	if (b > a)
+		return SYMBOL_A;
+	else if (a > b)
+		return SYMBOL_B;
+
+	/* Choose the symbol with the longest name */
+	na = strlen(syma->name);
+	nb = strlen(symb->name);
+	if (na > nb)
+		return SYMBOL_A;
+	else if (na < nb)
+		return SYMBOL_B;
+
+	/* Avoid "SyS" kernel syscall aliases */
+	if (na >= 3 && !strncmp(syma->name, "SyS", 3))
+		return SYMBOL_B;
+	if (na >= 10 && !strncmp(syma->name, "compat_SyS", 10))
+		return SYMBOL_B;
+
+	return SYMBOL_A;
+}
+
+void symbols__fixup_duplicate(struct rb_root *symbols)
+{
+	struct rb_node *nd;
+	struct symbol *curr, *next;
+
+	nd = rb_first(symbols);
+
+	while (nd) {
+		curr = rb_entry(nd, struct symbol, rb_node);
+again:
+		nd = rb_next(&curr->rb_node);
+		next = rb_entry(nd, struct symbol, rb_node);
+
+		if (!nd)
+			break;
+
+		if (curr->start != next->start)
+			continue;
+
+		if (choose_best_symbol(curr, next) == SYMBOL_A) {
+			rb_erase(&next->rb_node, symbols);
+			symbol__delete(next);
+			goto again;
+		} else {
+			nd = rb_next(&curr->rb_node);
+			rb_erase(&curr->rb_node, symbols);
+			symbol__delete(curr);
+		}
+	}
+}
+
+void symbols__fixup_end(struct rb_root *symbols)
+{
+	struct rb_node *nd, *prevnd = rb_first(symbols);
+	struct symbol *curr, *prev;
+
+	if (prevnd == NULL)
+		return;
+
+	curr = rb_entry(prevnd, struct symbol, rb_node);
+
+	for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) {
+		prev = curr;
+		curr = rb_entry(nd, struct symbol, rb_node);
+
+		if (prev->end == prev->start && prev->end != curr->start)
+			prev->end = curr->start;
+	}
+
+	/* Last entry */
+	if (curr->end == curr->start)
+		curr->end = roundup(curr->start, 4096);
+}
+
+void __map_groups__fixup_end(struct map_groups *mg, enum map_type type)
+{
+	struct map *prev, *curr;
+	struct rb_node *nd, *prevnd = rb_first(&mg->maps[type]);
+
+	if (prevnd == NULL)
+		return;
+
+	curr = rb_entry(prevnd, struct map, rb_node);
+
+	for (nd = rb_next(prevnd); nd; nd = rb_next(nd)) {
+		prev = curr;
+		curr = rb_entry(nd, struct map, rb_node);
+		prev->end = curr->start;
+	}
+
+	/*
+	 * We still haven't the actual symbols, so guess the
+	 * last map final address.
+	 */
+	curr->end = ~0ULL;
+}
+
+struct symbol *symbol__new(u64 start, u64 len, u8 binding, const char *name)
+{
+	size_t namelen = strlen(name) + 1;
+	struct symbol *sym = calloc(1, (symbol_conf.priv_size +
+					sizeof(*sym) + namelen));
+	if (sym == NULL)
+		return NULL;
+
+	if (symbol_conf.priv_size)
+		sym = ((void *)sym) + symbol_conf.priv_size;
+
+	sym->start   = start;
+	sym->end     = len ? start + len : start;
+	sym->binding = binding;
+	sym->namelen = namelen - 1;
+
+	pr_debug4("%s: %s %#" PRIx64 "-%#" PRIx64 "\n",
+		  __func__, name, start, sym->end);
+	memcpy(sym->name, name, namelen);
+
+	return sym;
+}
+
+void symbol__delete(struct symbol *sym)
+{
+	free(((void *)sym) - symbol_conf.priv_size);
+}
+
+size_t symbol__fprintf(struct symbol *sym, FILE *fp)
+{
+	return fprintf(fp, " %" PRIx64 "-%" PRIx64 " %c %s\n",
+		       sym->start, sym->end,
+		       sym->binding == STB_GLOBAL ? 'g' :
+		       sym->binding == STB_LOCAL  ? 'l' : 'w',
+		       sym->name);
+}
+
+size_t symbol__fprintf_symname_offs(const struct symbol *sym,
+				    const struct addr_location *al, FILE *fp)
+{
+	unsigned long offset;
+	size_t length;
+
+	if (sym && sym->name) {
+		length = fprintf(fp, "%s", sym->name);
+		if (al) {
+			if (al->addr < sym->end)
+				offset = al->addr - sym->start;
+			else
+				offset = al->addr - al->map->start - sym->start;
+			length += fprintf(fp, "+0x%lx", offset);
+		}
+		return length;
+	} else
+		return fprintf(fp, "[unknown]");
+}
+
+size_t symbol__fprintf_symname(const struct symbol *sym, FILE *fp)
+{
+	return symbol__fprintf_symname_offs(sym, NULL, fp);
+}
+
+void symbols__delete(struct rb_root *symbols)
+{
+	struct symbol *pos;
+	struct rb_node *next = rb_first(symbols);
+
+	while (next) {
+		pos = rb_entry(next, struct symbol, rb_node);
+		next = rb_next(&pos->rb_node);
+		rb_erase(&pos->rb_node, symbols);
+		symbol__delete(pos);
+	}
+}
+
+void symbols__insert(struct rb_root *symbols, struct symbol *sym)
+{
+	struct rb_node **p = &symbols->rb_node;
+	struct rb_node *parent = NULL;
+	const u64 ip = sym->start;
+	struct symbol *s;
+
+	while (*p != NULL) {
+		parent = *p;
+		s = rb_entry(parent, struct symbol, rb_node);
+		if (ip < s->start)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&sym->rb_node, parent, p);
+	rb_insert_color(&sym->rb_node, symbols);
+}
+
+static struct symbol *symbols__find(struct rb_root *symbols, u64 ip)
+{
+	struct rb_node *n;
+
+	if (symbols == NULL)
+		return NULL;
+
+	n = symbols->rb_node;
+
+	while (n) {
+		struct symbol *s = rb_entry(n, struct symbol, rb_node);
+
+		if (ip < s->start)
+			n = n->rb_left;
+		else if (ip >= s->end)
+			n = n->rb_right;
+		else
+			return s;
+	}
+
+	return NULL;
+}
+
+static struct symbol *symbols__first(struct rb_root *symbols)
+{
+	struct rb_node *n = rb_first(symbols);
+
+	if (n)
+		return rb_entry(n, struct symbol, rb_node);
+
+	return NULL;
+}
+
+static struct symbol *symbols__next(struct symbol *sym)
+{
+	struct rb_node *n = rb_next(&sym->rb_node);
+
+	if (n)
+		return rb_entry(n, struct symbol, rb_node);
+
+	return NULL;
+}
+
+struct symbol_name_rb_node {
+	struct rb_node	rb_node;
+	struct symbol	sym;
+};
+
+static void symbols__insert_by_name(struct rb_root *symbols, struct symbol *sym)
+{
+	struct rb_node **p = &symbols->rb_node;
+	struct rb_node *parent = NULL;
+	struct symbol_name_rb_node *symn, *s;
+
+	symn = container_of(sym, struct symbol_name_rb_node, sym);
+
+	while (*p != NULL) {
+		parent = *p;
+		s = rb_entry(parent, struct symbol_name_rb_node, rb_node);
+		if (strcmp(sym->name, s->sym.name) < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&symn->rb_node, parent, p);
+	rb_insert_color(&symn->rb_node, symbols);
+}
+
+static void symbols__sort_by_name(struct rb_root *symbols,
+				  struct rb_root *source)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(source); nd; nd = rb_next(nd)) {
+		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
+		symbols__insert_by_name(symbols, pos);
+	}
+}
+
+static struct symbol *symbols__find_by_name(struct rb_root *symbols,
+					    const char *name)
+{
+	struct rb_node *n;
+	struct symbol_name_rb_node *s;
+
+	if (symbols == NULL)
+		return NULL;
+
+	n = symbols->rb_node;
+
+	while (n) {
+		int cmp;
+
+		s = rb_entry(n, struct symbol_name_rb_node, rb_node);
+		cmp = strcmp(name, s->sym.name);
+
+		if (cmp < 0)
+			n = n->rb_left;
+		else if (cmp > 0)
+			n = n->rb_right;
+		else
+			break;
+	}
+
+	if (n == NULL)
+		return NULL;
+
+	/* return first symbol that has same name (if any) */
+	for (n = rb_prev(n); n; n = rb_prev(n)) {
+		struct symbol_name_rb_node *tmp;
+
+		tmp = rb_entry(n, struct symbol_name_rb_node, rb_node);
+		if (strcmp(tmp->sym.name, s->sym.name))
+			break;
+
+		s = tmp;
+	}
+
+	return &s->sym;
+}
+
+struct symbol *dso__find_symbol(struct dso *dso,
+				enum map_type type, u64 addr)
+{
+	return symbols__find(&dso->symbols[type], addr);
+}
+
+struct symbol *dso__first_symbol(struct dso *dso, enum map_type type)
+{
+	return symbols__first(&dso->symbols[type]);
+}
+
+struct symbol *dso__next_symbol(struct symbol *sym)
+{
+	return symbols__next(sym);
+}
+
+struct symbol *symbol__next_by_name(struct symbol *sym)
+{
+	struct symbol_name_rb_node *s = container_of(sym, struct symbol_name_rb_node, sym);
+	struct rb_node *n = rb_next(&s->rb_node);
+
+	return n ? &rb_entry(n, struct symbol_name_rb_node, rb_node)->sym : NULL;
+}
+
+ /*
+  * Teturns first symbol that matched with @name.
+  */
+struct symbol *dso__find_symbol_by_name(struct dso *dso, enum map_type type,
+					const char *name)
+{
+	return symbols__find_by_name(&dso->symbol_names[type], name);
+}
+
+void dso__sort_by_name(struct dso *dso, enum map_type type)
+{
+	dso__set_sorted_by_name(dso, type);
+	return symbols__sort_by_name(&dso->symbol_names[type],
+				     &dso->symbols[type]);
+}
+
+size_t dso__fprintf_symbols_by_name(struct dso *dso,
+				    enum map_type type, FILE *fp)
+{
+	size_t ret = 0;
+	struct rb_node *nd;
+	struct symbol_name_rb_node *pos;
+
+	for (nd = rb_first(&dso->symbol_names[type]); nd; nd = rb_next(nd)) {
+		pos = rb_entry(nd, struct symbol_name_rb_node, rb_node);
+		fprintf(fp, "%s\n", pos->sym.name);
+	}
+
+	return ret;
+}
+
+int modules__parse(const char *filename, void *arg,
+		   int (*process_module)(void *arg, const char *name,
+					 u64 start))
+{
+	char *line = NULL;
+	size_t n;
+	FILE *file;
+	int err = 0;
+
+	file = fopen(filename, "r");
+	if (file == NULL)
+		return -1;
+
+	while (1) {
+		char name[PATH_MAX];
+		u64 start;
+		char *sep;
+		ssize_t line_len;
+
+		line_len = getline(&line, &n, file);
+		if (line_len < 0) {
+			if (feof(file))
+				break;
+			err = -1;
+			goto out;
+		}
+
+		if (!line) {
+			err = -1;
+			goto out;
+		}
+
+		line[--line_len] = '\0'; /* \n */
+
+		sep = strrchr(line, 'x');
+		if (sep == NULL)
+			continue;
+
+		hex2u64(sep + 1, &start);
+
+		sep = strchr(line, ' ');
+		if (sep == NULL)
+			continue;
+
+		*sep = '\0';
+
+		scnprintf(name, sizeof(name), "[%s]", line);
+
+		err = process_module(arg, name, start);
+		if (err)
+			break;
+	}
+out:
+	free(line);
+	fclose(file);
+	return err;
+}
+
+struct process_kallsyms_args {
+	struct map *map;
+	struct dso *dso;
+};
+
+/*
+ * These are symbols in the kernel image, so make sure that
+ * sym is from a kernel DSO.
+ */
+bool symbol__is_idle(struct symbol *sym)
+{
+	const char * const idle_symbols[] = {
+		"cpu_idle",
+		"cpu_startup_entry",
+		"intel_idle",
+		"default_idle",
+		"native_safe_halt",
+		"enter_idle",
+		"exit_idle",
+		"mwait_idle",
+		"mwait_idle_with_hints",
+		"poll_idle",
+		"ppc64_runlatch_off",
+		"pseries_dedicated_idle_sleep",
+		NULL
+	};
+
+	int i;
+
+	if (!sym)
+		return false;
+
+	for (i = 0; idle_symbols[i]; i++) {
+		if (!strcmp(idle_symbols[i], sym->name))
+			return true;
+	}
+
+	return false;
+}
+
+static int map__process_kallsym_symbol(void *arg, const char *name,
+				       char type, u64 start)
+{
+	struct symbol *sym;
+	struct process_kallsyms_args *a = arg;
+	struct rb_root *root = &a->dso->symbols[a->map->type];
+
+	if (!symbol_type__is_a(type, a->map->type))
+		return 0;
+
+	/*
+	 * module symbols are not sorted so we add all
+	 * symbols, setting length to 0, and rely on
+	 * symbols__fixup_end() to fix it up.
+	 */
+	sym = symbol__new(start, 0, kallsyms2elf_type(type), name);
+	if (sym == NULL)
+		return -ENOMEM;
+	/*
+	 * We will pass the symbols to the filter later, in
+	 * map__split_kallsyms, when we have split the maps per module
+	 */
+	symbols__insert(root, sym);
+
+	return 0;
+}
+
+/*
+ * Loads the function entries in /proc/kallsyms into kernel_map->dso,
+ * so that we can in the next step set the symbol ->end address and then
+ * call kernel_maps__split_kallsyms.
+ */
+static int dso__load_all_kallsyms(struct dso *dso, const char *filename,
+				  struct map *map)
+{
+	struct process_kallsyms_args args = { .map = map, .dso = dso, };
+	return kallsyms__parse(filename, &args, map__process_kallsym_symbol);
+}
+
+static int dso__split_kallsyms_for_kcore(struct dso *dso, struct map *map,
+					 symbol_filter_t filter)
+{
+	struct map_groups *kmaps = map__kmaps(map);
+	struct map *curr_map;
+	struct symbol *pos;
+	int count = 0, moved = 0;
+	struct rb_root *root = &dso->symbols[map->type];
+	struct rb_node *next = rb_first(root);
+
+	if (!kmaps)
+		return -1;
+
+	while (next) {
+		char *module;
+
+		pos = rb_entry(next, struct symbol, rb_node);
+		next = rb_next(&pos->rb_node);
+
+		module = strchr(pos->name, '\t');
+		if (module)
+			*module = '\0';
+
+		curr_map = map_groups__find(kmaps, map->type, pos->start);
+
+		if (!curr_map || (filter && filter(curr_map, pos))) {
+			rb_erase(&pos->rb_node, root);
+			symbol__delete(pos);
+		} else {
+			pos->start -= curr_map->start - curr_map->pgoff;
+			if (pos->end)
+				pos->end -= curr_map->start - curr_map->pgoff;
+			if (curr_map != map) {
+				rb_erase(&pos->rb_node, root);
+				symbols__insert(
+					&curr_map->dso->symbols[curr_map->type],
+					pos);
+				++moved;
+			} else {
+				++count;
+			}
+		}
+	}
+
+	/* Symbols have been adjusted */
+	dso->adjust_symbols = 1;
+
+	return count + moved;
+}
+
+/*
+ * Split the symbols into maps, making sure there are no overlaps, i.e. the
+ * kernel range is broken in several maps, named [kernel].N, as we don't have
+ * the original ELF section names vmlinux have.
+ */
+static int dso__split_kallsyms(struct dso *dso, struct map *map, u64 delta,
+			       symbol_filter_t filter)
+{
+	struct map_groups *kmaps = map__kmaps(map);
+	struct machine *machine;
+	struct map *curr_map = map;
+	struct symbol *pos;
+	int count = 0, moved = 0;
+	struct rb_root *root = &dso->symbols[map->type];
+	struct rb_node *next = rb_first(root);
+	int kernel_range = 0;
+
+	if (!kmaps)
+		return -1;
+
+	machine = kmaps->machine;
+
+	while (next) {
+		char *module;
+
+		pos = rb_entry(next, struct symbol, rb_node);
+		next = rb_next(&pos->rb_node);
+
+		module = strchr(pos->name, '\t');
+		if (module) {
+			if (!symbol_conf.use_modules)
+				goto discard_symbol;
+
+			*module++ = '\0';
+
+			if (strcmp(curr_map->dso->short_name, module)) {
+				if (curr_map != map &&
+				    dso->kernel == DSO_TYPE_GUEST_KERNEL &&
+				    machine__is_default_guest(machine)) {
+					/*
+					 * We assume all symbols of a module are
+					 * continuous in * kallsyms, so curr_map
+					 * points to a module and all its
+					 * symbols are in its kmap. Mark it as
+					 * loaded.
+					 */
+					dso__set_loaded(curr_map->dso,
+							curr_map->type);
+				}
+
+				curr_map = map_groups__find_by_name(kmaps,
+							map->type, module);
+				if (curr_map == NULL) {
+					pr_debug("%s/proc/{kallsyms,modules} "
+					         "inconsistency while looking "
+						 "for \"%s\" module!\n",
+						 machine->root_dir, module);
+					curr_map = map;
+					goto discard_symbol;
+				}
+
+				if (curr_map->dso->loaded &&
+				    !machine__is_default_guest(machine))
+					goto discard_symbol;
+			}
+			/*
+			 * So that we look just like we get from .ko files,
+			 * i.e. not prelinked, relative to map->start.
+			 */
+			pos->start = curr_map->map_ip(curr_map, pos->start);
+			pos->end   = curr_map->map_ip(curr_map, pos->end);
+		} else if (curr_map != map) {
+			char dso_name[PATH_MAX];
+			struct dso *ndso;
+
+			if (delta) {
+				/* Kernel was relocated at boot time */
+				pos->start -= delta;
+				pos->end -= delta;
+			}
+
+			if (count == 0) {
+				curr_map = map;
+				goto filter_symbol;
+			}
+
+			if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
+				snprintf(dso_name, sizeof(dso_name),
+					"[guest.kernel].%d",
+					kernel_range++);
+			else
+				snprintf(dso_name, sizeof(dso_name),
+					"[kernel].%d",
+					kernel_range++);
+
+			ndso = dso__new(dso_name);
+			if (ndso == NULL)
+				return -1;
+
+			ndso->kernel = dso->kernel;
+
+			curr_map = map__new2(pos->start, ndso, map->type);
+			if (curr_map == NULL) {
+				dso__delete(ndso);
+				return -1;
+			}
+
+			curr_map->map_ip = curr_map->unmap_ip = identity__map_ip;
+			map_groups__insert(kmaps, curr_map);
+			++kernel_range;
+		} else if (delta) {
+			/* Kernel was relocated at boot time */
+			pos->start -= delta;
+			pos->end -= delta;
+		}
+filter_symbol:
+		if (filter && filter(curr_map, pos)) {
+discard_symbol:		rb_erase(&pos->rb_node, root);
+			symbol__delete(pos);
+		} else {
+			if (curr_map != map) {
+				rb_erase(&pos->rb_node, root);
+				symbols__insert(&curr_map->dso->symbols[curr_map->type], pos);
+				++moved;
+			} else
+				++count;
+		}
+	}
+
+	if (curr_map != map &&
+	    dso->kernel == DSO_TYPE_GUEST_KERNEL &&
+	    machine__is_default_guest(kmaps->machine)) {
+		dso__set_loaded(curr_map->dso, curr_map->type);
+	}
+
+	return count + moved;
+}
+
+bool symbol__restricted_filename(const char *filename,
+				 const char *restricted_filename)
+{
+	bool restricted = false;
+
+	if (symbol_conf.kptr_restrict) {
+		char *r = realpath(filename, NULL);
+
+		if (r != NULL) {
+			restricted = strcmp(r, restricted_filename) == 0;
+			free(r);
+			return restricted;
+		}
+	}
+
+	return restricted;
+}
+
+struct module_info {
+	struct rb_node rb_node;
+	char *name;
+	u64 start;
+};
+
+static void add_module(struct module_info *mi, struct rb_root *modules)
+{
+	struct rb_node **p = &modules->rb_node;
+	struct rb_node *parent = NULL;
+	struct module_info *m;
+
+	while (*p != NULL) {
+		parent = *p;
+		m = rb_entry(parent, struct module_info, rb_node);
+		if (strcmp(mi->name, m->name) < 0)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+	rb_link_node(&mi->rb_node, parent, p);
+	rb_insert_color(&mi->rb_node, modules);
+}
+
+static void delete_modules(struct rb_root *modules)
+{
+	struct module_info *mi;
+	struct rb_node *next = rb_first(modules);
+
+	while (next) {
+		mi = rb_entry(next, struct module_info, rb_node);
+		next = rb_next(&mi->rb_node);
+		rb_erase(&mi->rb_node, modules);
+		zfree(&mi->name);
+		free(mi);
+	}
+}
+
+static struct module_info *find_module(const char *name,
+				       struct rb_root *modules)
+{
+	struct rb_node *n = modules->rb_node;
+
+	while (n) {
+		struct module_info *m;
+		int cmp;
+
+		m = rb_entry(n, struct module_info, rb_node);
+		cmp = strcmp(name, m->name);
+		if (cmp < 0)
+			n = n->rb_left;
+		else if (cmp > 0)
+			n = n->rb_right;
+		else
+			return m;
+	}
+
+	return NULL;
+}
+
+static int __read_proc_modules(void *arg, const char *name, u64 start)
+{
+	struct rb_root *modules = arg;
+	struct module_info *mi;
+
+	mi = zalloc(sizeof(struct module_info));
+	if (!mi)
+		return -ENOMEM;
+
+	mi->name = strdup(name);
+	mi->start = start;
+
+	if (!mi->name) {
+		free(mi);
+		return -ENOMEM;
+	}
+
+	add_module(mi, modules);
+
+	return 0;
+}
+
+static int read_proc_modules(const char *filename, struct rb_root *modules)
+{
+	if (symbol__restricted_filename(filename, "/proc/modules"))
+		return -1;
+
+	if (modules__parse(filename, modules, __read_proc_modules)) {
+		delete_modules(modules);
+		return -1;
+	}
+
+	return 0;
+}
+
+int compare_proc_modules(const char *from, const char *to)
+{
+	struct rb_root from_modules = RB_ROOT;
+	struct rb_root to_modules = RB_ROOT;
+	struct rb_node *from_node, *to_node;
+	struct module_info *from_m, *to_m;
+	int ret = -1;
+
+	if (read_proc_modules(from, &from_modules))
+		return -1;
+
+	if (read_proc_modules(to, &to_modules))
+		goto out_delete_from;
+
+	from_node = rb_first(&from_modules);
+	to_node = rb_first(&to_modules);
+	while (from_node) {
+		if (!to_node)
+			break;
+
+		from_m = rb_entry(from_node, struct module_info, rb_node);
+		to_m = rb_entry(to_node, struct module_info, rb_node);
+
+		if (from_m->start != to_m->start ||
+		    strcmp(from_m->name, to_m->name))
+			break;
+
+		from_node = rb_next(from_node);
+		to_node = rb_next(to_node);
+	}
+
+	if (!from_node && !to_node)
+		ret = 0;
+
+	delete_modules(&to_modules);
+out_delete_from:
+	delete_modules(&from_modules);
+
+	return ret;
+}
+
+static int do_validate_kcore_modules(const char *filename, struct map *map,
+				  struct map_groups *kmaps)
+{
+	struct rb_root modules = RB_ROOT;
+	struct map *old_map;
+	int err;
+
+	err = read_proc_modules(filename, &modules);
+	if (err)
+		return err;
+
+	old_map = map_groups__first(kmaps, map->type);
+	while (old_map) {
+		struct map *next = map_groups__next(old_map);
+		struct module_info *mi;
+
+		if (old_map == map || old_map->start == map->start) {
+			/* The kernel map */
+			old_map = next;
+			continue;
+		}
+
+		/* Module must be in memory at the same address */
+		mi = find_module(old_map->dso->short_name, &modules);
+		if (!mi || mi->start != old_map->start) {
+			err = -EINVAL;
+			goto out;
+		}
+
+		old_map = next;
+	}
+out:
+	delete_modules(&modules);
+	return err;
+}
+
+/*
+ * If kallsyms is referenced by name then we look for filename in the same
+ * directory.
+ */
+static bool filename_from_kallsyms_filename(char *filename,
+					    const char *base_name,
+					    const char *kallsyms_filename)
+{
+	char *name;
+
+	strcpy(filename, kallsyms_filename);
+	name = strrchr(filename, '/');
+	if (!name)
+		return false;
+
+	name += 1;
+
+	if (!strcmp(name, "kallsyms")) {
+		strcpy(name, base_name);
+		return true;
+	}
+
+	return false;
+}
+
+static int validate_kcore_modules(const char *kallsyms_filename,
+				  struct map *map)
+{
+	struct map_groups *kmaps = map__kmaps(map);
+	char modules_filename[PATH_MAX];
+
+	if (!kmaps)
+		return -EINVAL;
+
+	if (!filename_from_kallsyms_filename(modules_filename, "modules",
+					     kallsyms_filename))
+		return -EINVAL;
+
+	if (do_validate_kcore_modules(modules_filename, map, kmaps))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int validate_kcore_addresses(const char *kallsyms_filename,
+				    struct map *map)
+{
+	struct kmap *kmap = map__kmap(map);
+
+	if (!kmap)
+		return -EINVAL;
+
+	if (kmap->ref_reloc_sym && kmap->ref_reloc_sym->name) {
+		u64 start;
+
+		start = kallsyms__get_function_start(kallsyms_filename,
+						     kmap->ref_reloc_sym->name);
+		if (start != kmap->ref_reloc_sym->addr)
+			return -EINVAL;
+	}
+
+	return validate_kcore_modules(kallsyms_filename, map);
+}
+
+struct kcore_mapfn_data {
+	struct dso *dso;
+	enum map_type type;
+	struct list_head maps;
+};
+
+static int kcore_mapfn(u64 start, u64 len, u64 pgoff, void *data)
+{
+	struct kcore_mapfn_data *md = data;
+	struct map *map;
+
+	map = map__new2(start, md->dso, md->type);
+	if (map == NULL)
+		return -ENOMEM;
+
+	map->end = map->start + len;
+	map->pgoff = pgoff;
+
+	list_add(&map->node, &md->maps);
+
+	return 0;
+}
+
+static int dso__load_kcore(struct dso *dso, struct map *map,
+			   const char *kallsyms_filename)
+{
+	struct map_groups *kmaps = map__kmaps(map);
+	struct machine *machine;
+	struct kcore_mapfn_data md;
+	struct map *old_map, *new_map, *replacement_map = NULL;
+	bool is_64_bit;
+	int err, fd;
+	char kcore_filename[PATH_MAX];
+	struct symbol *sym;
+
+	if (!kmaps)
+		return -EINVAL;
+
+	machine = kmaps->machine;
+
+	/* This function requires that the map is the kernel map */
+	if (map != machine->vmlinux_maps[map->type])
+		return -EINVAL;
+
+	if (!filename_from_kallsyms_filename(kcore_filename, "kcore",
+					     kallsyms_filename))
+		return -EINVAL;
+
+	/* Modules and kernel must be present at their original addresses */
+	if (validate_kcore_addresses(kallsyms_filename, map))
+		return -EINVAL;
+
+	md.dso = dso;
+	md.type = map->type;
+	INIT_LIST_HEAD(&md.maps);
+
+	fd = open(kcore_filename, O_RDONLY);
+	if (fd < 0)
+		return -EINVAL;
+
+	/* Read new maps into temporary lists */
+	err = file__read_maps(fd, md.type == MAP__FUNCTION, kcore_mapfn, &md,
+			      &is_64_bit);
+	if (err)
+		goto out_err;
+	dso->is_64_bit = is_64_bit;
+
+	if (list_empty(&md.maps)) {
+		err = -EINVAL;
+		goto out_err;
+	}
+
+	/* Remove old maps */
+	old_map = map_groups__first(kmaps, map->type);
+	while (old_map) {
+		struct map *next = map_groups__next(old_map);
+
+		if (old_map != map)
+			map_groups__remove(kmaps, old_map);
+		old_map = next;
+	}
+
+	/* Find the kernel map using the first symbol */
+	sym = dso__first_symbol(dso, map->type);
+	list_for_each_entry(new_map, &md.maps, node) {
+		if (sym && sym->start >= new_map->start &&
+		    sym->start < new_map->end) {
+			replacement_map = new_map;
+			break;
+		}
+	}
+
+	if (!replacement_map)
+		replacement_map = list_entry(md.maps.next, struct map, node);
+
+	/* Add new maps */
+	while (!list_empty(&md.maps)) {
+		new_map = list_entry(md.maps.next, struct map, node);
+		list_del(&new_map->node);
+		if (new_map == replacement_map) {
+			map->start	= new_map->start;
+			map->end	= new_map->end;
+			map->pgoff	= new_map->pgoff;
+			map->map_ip	= new_map->map_ip;
+			map->unmap_ip	= new_map->unmap_ip;
+			map__delete(new_map);
+			/* Ensure maps are correctly ordered */
+			map_groups__remove(kmaps, map);
+			map_groups__insert(kmaps, map);
+		} else {
+			map_groups__insert(kmaps, new_map);
+		}
+	}
+
+	/*
+	 * Set the data type and long name so that kcore can be read via
+	 * dso__data_read_addr().
+	 */
+	if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
+		dso->binary_type = DSO_BINARY_TYPE__GUEST_KCORE;
+	else
+		dso->binary_type = DSO_BINARY_TYPE__KCORE;
+	dso__set_long_name(dso, strdup(kcore_filename), true);
+
+	close(fd);
+
+	if (map->type == MAP__FUNCTION)
+		pr_debug("Using %s for kernel object code\n", kcore_filename);
+	else
+		pr_debug("Using %s for kernel data\n", kcore_filename);
+
+	return 0;
+
+out_err:
+	while (!list_empty(&md.maps)) {
+		map = list_entry(md.maps.next, struct map, node);
+		list_del(&map->node);
+		map__delete(map);
+	}
+	close(fd);
+	return -EINVAL;
+}
+
+/*
+ * If the kernel is relocated at boot time, kallsyms won't match.  Compute the
+ * delta based on the relocation reference symbol.
+ */
+static int kallsyms__delta(struct map *map, const char *filename, u64 *delta)
+{
+	struct kmap *kmap = map__kmap(map);
+	u64 addr;
+
+	if (!kmap)
+		return -1;
+
+	if (!kmap->ref_reloc_sym || !kmap->ref_reloc_sym->name)
+		return 0;
+
+	addr = kallsyms__get_function_start(filename,
+					    kmap->ref_reloc_sym->name);
+	if (!addr)
+		return -1;
+
+	*delta = addr - kmap->ref_reloc_sym->addr;
+	return 0;
+}
+
+int dso__load_kallsyms(struct dso *dso, const char *filename,
+		       struct map *map, symbol_filter_t filter)
+{
+	u64 delta = 0;
+
+	if (symbol__restricted_filename(filename, "/proc/kallsyms"))
+		return -1;
+
+	if (dso__load_all_kallsyms(dso, filename, map) < 0)
+		return -1;
+
+	if (kallsyms__delta(map, filename, &delta))
+		return -1;
+
+	symbols__fixup_duplicate(&dso->symbols[map->type]);
+	symbols__fixup_end(&dso->symbols[map->type]);
+
+	if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
+		dso->symtab_type = DSO_BINARY_TYPE__GUEST_KALLSYMS;
+	else
+		dso->symtab_type = DSO_BINARY_TYPE__KALLSYMS;
+
+	if (!dso__load_kcore(dso, map, filename))
+		return dso__split_kallsyms_for_kcore(dso, map, filter);
+	else
+		return dso__split_kallsyms(dso, map, delta, filter);
+}
+
+static int dso__load_perf_map(struct dso *dso, struct map *map,
+			      symbol_filter_t filter)
+{
+	char *line = NULL;
+	size_t n;
+	FILE *file;
+	int nr_syms = 0;
+
+	file = fopen(dso->long_name, "r");
+	if (file == NULL)
+		goto out_failure;
+
+	while (!feof(file)) {
+		u64 start, size;
+		struct symbol *sym;
+		int line_len, len;
+
+		line_len = getline(&line, &n, file);
+		if (line_len < 0)
+			break;
+
+		if (!line)
+			goto out_failure;
+
+		line[--line_len] = '\0'; /* \n */
+
+		len = hex2u64(line, &start);
+
+		len++;
+		if (len + 2 >= line_len)
+			continue;
+
+		len += hex2u64(line + len, &size);
+
+		len++;
+		if (len + 2 >= line_len)
+			continue;
+
+		sym = symbol__new(start, size, STB_GLOBAL, line + len);
+
+		if (sym == NULL)
+			goto out_delete_line;
+
+		if (filter && filter(map, sym))
+			symbol__delete(sym);
+		else {
+			symbols__insert(&dso->symbols[map->type], sym);
+			nr_syms++;
+		}
+	}
+
+	free(line);
+	fclose(file);
+
+	return nr_syms;
+
+out_delete_line:
+	free(line);
+out_failure:
+	return -1;
+}
+
+static bool dso__is_compatible_symtab_type(struct dso *dso, bool kmod,
+					   enum dso_binary_type type)
+{
+	switch (type) {
+	case DSO_BINARY_TYPE__JAVA_JIT:
+	case DSO_BINARY_TYPE__DEBUGLINK:
+	case DSO_BINARY_TYPE__SYSTEM_PATH_DSO:
+	case DSO_BINARY_TYPE__FEDORA_DEBUGINFO:
+	case DSO_BINARY_TYPE__UBUNTU_DEBUGINFO:
+	case DSO_BINARY_TYPE__BUILDID_DEBUGINFO:
+	case DSO_BINARY_TYPE__OPENEMBEDDED_DEBUGINFO:
+		return !kmod && dso->kernel == DSO_TYPE_USER;
+
+	case DSO_BINARY_TYPE__KALLSYMS:
+	case DSO_BINARY_TYPE__VMLINUX:
+	case DSO_BINARY_TYPE__KCORE:
+		return dso->kernel == DSO_TYPE_KERNEL;
+
+	case DSO_BINARY_TYPE__GUEST_KALLSYMS:
+	case DSO_BINARY_TYPE__GUEST_VMLINUX:
+	case DSO_BINARY_TYPE__GUEST_KCORE:
+		return dso->kernel == DSO_TYPE_GUEST_KERNEL;
+
+	case DSO_BINARY_TYPE__GUEST_KMODULE:
+	case DSO_BINARY_TYPE__GUEST_KMODULE_COMP:
+	case DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE:
+	case DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE_COMP:
+		/*
+		 * kernel modules know their symtab type - it's set when
+		 * creating a module dso in machine__new_module().
+		 */
+		return kmod && dso->symtab_type == type;
+
+	case DSO_BINARY_TYPE__BUILD_ID_CACHE:
+		return true;
+
+	case DSO_BINARY_TYPE__NOT_FOUND:
+	default:
+		return false;
+	}
+}
+
+int dso__load(struct dso *dso, struct map *map, symbol_filter_t filter)
+{
+	char *name;
+	int ret = -1;
+	u_int i;
+	struct machine *machine;
+	char *root_dir = (char *) "";
+	int ss_pos = 0;
+	struct symsrc ss_[2];
+	struct symsrc *syms_ss = NULL, *runtime_ss = NULL;
+	bool kmod;
+
+	dso__set_loaded(dso, map->type);
+
+	if (dso->kernel == DSO_TYPE_KERNEL)
+		return dso__load_kernel_sym(dso, map, filter);
+	else if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
+		return dso__load_guest_kernel_sym(dso, map, filter);
+
+	if (map->groups && map->groups->machine)
+		machine = map->groups->machine;
+	else
+		machine = NULL;
+
+	dso->adjust_symbols = 0;
+
+	if (strncmp(dso->name, "/tmp/perf-", 10) == 0) {
+		struct stat st;
+
+		if (lstat(dso->name, &st) < 0)
+			return -1;
+
+		if (st.st_uid && (st.st_uid != geteuid())) {
+			pr_warning("File %s not owned by current user or root, "
+				"ignoring it.\n", dso->name);
+			return -1;
+		}
+
+		ret = dso__load_perf_map(dso, map, filter);
+		dso->symtab_type = ret > 0 ? DSO_BINARY_TYPE__JAVA_JIT :
+					     DSO_BINARY_TYPE__NOT_FOUND;
+		return ret;
+	}
+
+	if (machine)
+		root_dir = machine->root_dir;
+
+	name = malloc(PATH_MAX);
+	if (!name)
+		return -1;
+
+	kmod = dso->symtab_type == DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE ||
+		dso->symtab_type == DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE_COMP ||
+		dso->symtab_type == DSO_BINARY_TYPE__GUEST_KMODULE ||
+		dso->symtab_type == DSO_BINARY_TYPE__GUEST_KMODULE_COMP;
+
+	/*
+	 * Iterate over candidate debug images.
+	 * Keep track of "interesting" ones (those which have a symtab, dynsym,
+	 * and/or opd section) for processing.
+	 */
+	for (i = 0; i < DSO_BINARY_TYPE__SYMTAB_CNT; i++) {
+		struct symsrc *ss = &ss_[ss_pos];
+		bool next_slot = false;
+
+		enum dso_binary_type symtab_type = binary_type_symtab[i];
+
+		if (!dso__is_compatible_symtab_type(dso, kmod, symtab_type))
+			continue;
+
+		if (dso__read_binary_type_filename(dso, symtab_type,
+						   root_dir, name, PATH_MAX))
+			continue;
+
+		/* Name is now the name of the next image to try */
+		if (symsrc__init(ss, dso, name, symtab_type) < 0)
+			continue;
+
+		if (!syms_ss && symsrc__has_symtab(ss)) {
+			syms_ss = ss;
+			next_slot = true;
+			if (!dso->symsrc_filename)
+				dso->symsrc_filename = strdup(name);
+		}
+
+		if (!runtime_ss && symsrc__possibly_runtime(ss)) {
+			runtime_ss = ss;
+			next_slot = true;
+		}
+
+		if (next_slot) {
+			ss_pos++;
+
+			if (syms_ss && runtime_ss)
+				break;
+		} else {
+			symsrc__destroy(ss);
+		}
+
+	}
+
+	if (!runtime_ss && !syms_ss)
+		goto out_free;
+
+	if (runtime_ss && !syms_ss) {
+		syms_ss = runtime_ss;
+	}
+
+	/* We'll have to hope for the best */
+	if (!runtime_ss && syms_ss)
+		runtime_ss = syms_ss;
+
+	if (syms_ss)
+		ret = dso__load_sym(dso, map, syms_ss, runtime_ss, filter, kmod);
+	else
+		ret = -1;
+
+	if (ret > 0) {
+		int nr_plt;
+
+		nr_plt = dso__synthesize_plt_symbols(dso, runtime_ss, map, filter);
+		if (nr_plt > 0)
+			ret += nr_plt;
+	}
+
+	for (; ss_pos > 0; ss_pos--)
+		symsrc__destroy(&ss_[ss_pos - 1]);
+out_free:
+	free(name);
+	if (ret < 0 && strstr(dso->name, " (deleted)") != NULL)
+		return 0;
+	return ret;
+}
+
+struct map *map_groups__find_by_name(struct map_groups *mg,
+				     enum map_type type, const char *name)
+{
+	struct rb_node *nd;
+
+	for (nd = rb_first(&mg->maps[type]); nd; nd = rb_next(nd)) {
+		struct map *map = rb_entry(nd, struct map, rb_node);
+
+		if (map->dso && strcmp(map->dso->short_name, name) == 0)
+			return map;
+	}
+
+	return NULL;
+}
+
+int dso__load_vmlinux(struct dso *dso, struct map *map,
+		      const char *vmlinux, bool vmlinux_allocated,
+		      symbol_filter_t filter)
+{
+	int err = -1;
+	struct symsrc ss;
+	char symfs_vmlinux[PATH_MAX];
+	enum dso_binary_type symtab_type;
+
+	if (vmlinux[0] == '/')
+		snprintf(symfs_vmlinux, sizeof(symfs_vmlinux), "%s", vmlinux);
+	else
+		symbol__join_symfs(symfs_vmlinux, vmlinux);
+
+	if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
+		symtab_type = DSO_BINARY_TYPE__GUEST_VMLINUX;
+	else
+		symtab_type = DSO_BINARY_TYPE__VMLINUX;
+
+	if (symsrc__init(&ss, dso, symfs_vmlinux, symtab_type))
+		return -1;
+
+	err = dso__load_sym(dso, map, &ss, &ss, filter, 0);
+	symsrc__destroy(&ss);
+
+	if (err > 0) {
+		if (dso->kernel == DSO_TYPE_GUEST_KERNEL)
+			dso->binary_type = DSO_BINARY_TYPE__GUEST_VMLINUX;
+		else
+			dso->binary_type = DSO_BINARY_TYPE__VMLINUX;
+		dso__set_long_name(dso, vmlinux, vmlinux_allocated);
+		dso__set_loaded(dso, map->type);
+		pr_debug("Using %s for symbols\n", symfs_vmlinux);
+	}
+
+	return err;
+}
+
+int dso__load_vmlinux_path(struct dso *dso, struct map *map,
+			   symbol_filter_t filter)
+{
+	int i, err = 0;
+	char *filename = NULL;
+
+	if (!symbol_conf.ignore_vmlinux_buildid)
+		filename = dso__build_id_filename(dso, NULL, 0);
+	if (filename != NULL) {
+		err = dso__load_vmlinux(dso, map, filename, true, filter);
+		if (err > 0)
+			goto out;
+		free(filename);
+	}
+
+	pr_debug("Looking at the vmlinux_path (%d entries long)\n",
+		 vmlinux_path__nr_entries + 1);
+
+	for (i = 0; i < vmlinux_path__nr_entries; ++i) {
+		err = dso__load_vmlinux(dso, map, vmlinux_path[i], false, filter);
+		if (err > 0)
+			break;
+	}
+out:
+	return err;
+}
+
+static int find_matching_kcore(struct map *map, char *dir, size_t dir_sz)
+{
+	char kallsyms_filename[PATH_MAX];
+	struct dirent *dent;
+	int ret = -1;
+	DIR *d;
+
+	d = opendir(dir);
+	if (!d)
+		return -1;
+
+	while (1) {
+		dent = readdir(d);
+		if (!dent)
+			break;
+		if (dent->d_type != DT_DIR)
+			continue;
+		scnprintf(kallsyms_filename, sizeof(kallsyms_filename),
+			  "%s/%s/kallsyms", dir, dent->d_name);
+		if (!validate_kcore_addresses(kallsyms_filename, map)) {
+			strlcpy(dir, kallsyms_filename, dir_sz);
+			ret = 0;
+			break;
+		}
+	}
+
+	closedir(d);
+
+	return ret;
+}
+
+static char *dso__find_kallsyms(struct dso *dso, struct map *map)
+{
+	u8 host_build_id[BUILD_ID_SIZE];
+	char sbuild_id[BUILD_ID_SIZE * 2 + 1];
+	bool is_host = false;
+	char path[PATH_MAX];
+
+	if (!dso->has_build_id) {
+		/*
+		 * Last resort, if we don't have a build-id and couldn't find
+		 * any vmlinux file, try the running kernel kallsyms table.
+		 */
+		goto proc_kallsyms;
+	}
+
+	if (sysfs__read_build_id("/sys/kernel/notes", host_build_id,
+				 sizeof(host_build_id)) == 0)
+		is_host = dso__build_id_equal(dso, host_build_id);
+
+	build_id__sprintf(dso->build_id, sizeof(dso->build_id), sbuild_id);
+
+	scnprintf(path, sizeof(path), "%s/[kernel.kcore]/%s", buildid_dir,
+		  sbuild_id);
+
+	/* Use /proc/kallsyms if possible */
+	if (is_host) {
+		DIR *d;
+		int fd;
+
+		/* If no cached kcore go with /proc/kallsyms */
+		d = opendir(path);
+		if (!d)
+			goto proc_kallsyms;
+		closedir(d);
+
+		/*
+		 * Do not check the build-id cache, until we know we cannot use
+		 * /proc/kcore.
+		 */
+		fd = open("/proc/kcore", O_RDONLY);
+		if (fd != -1) {
+			close(fd);
+			/* If module maps match go with /proc/kallsyms */
+			if (!validate_kcore_addresses("/proc/kallsyms", map))
+				goto proc_kallsyms;
+		}
+
+		/* Find kallsyms in build-id cache with kcore */
+		if (!find_matching_kcore(map, path, sizeof(path)))
+			return strdup(path);
+
+		goto proc_kallsyms;
+	}
+
+	/* Find kallsyms in build-id cache with kcore */
+	if (!find_matching_kcore(map, path, sizeof(path)))
+		return strdup(path);
+
+	scnprintf(path, sizeof(path), "%s/[kernel.kallsyms]/%s",
+		  buildid_dir, sbuild_id);
+
+	if (access(path, F_OK)) {
+		pr_err("No kallsyms or vmlinux with build-id %s was found\n",
+		       sbuild_id);
+		return NULL;
+	}
+
+	return strdup(path);
+
+proc_kallsyms:
+	return strdup("/proc/kallsyms");
+}
+
+static int dso__load_kernel_sym(struct dso *dso, struct map *map,
+				symbol_filter_t filter)
+{
+	int err;
+	const char *kallsyms_filename = NULL;
+	char *kallsyms_allocated_filename = NULL;
+	/*
+	 * Step 1: if the user specified a kallsyms or vmlinux filename, use
+	 * it and only it, reporting errors to the user if it cannot be used.
+	 *
+	 * For instance, try to analyse an ARM perf.data file _without_ a
+	 * build-id, or if the user specifies the wrong path to the right
+	 * vmlinux file, obviously we can't fallback to another vmlinux (a
+	 * x86_86 one, on the machine where analysis is being performed, say),
+	 * or worse, /proc/kallsyms.
+	 *
+	 * If the specified file _has_ a build-id and there is a build-id
+	 * section in the perf.data file, we will still do the expected
+	 * validation in dso__load_vmlinux and will bail out if they don't
+	 * match.
+	 */
+	if (symbol_conf.kallsyms_name != NULL) {
+		kallsyms_filename = symbol_conf.kallsyms_name;
+		goto do_kallsyms;
+	}
+
+	if (!symbol_conf.ignore_vmlinux && symbol_conf.vmlinux_name != NULL) {
+		return dso__load_vmlinux(dso, map, symbol_conf.vmlinux_name,
+					 false, filter);
+	}
+
+	if (!symbol_conf.ignore_vmlinux && vmlinux_path != NULL) {
+		err = dso__load_vmlinux_path(dso, map, filter);
+		if (err > 0)
+			return err;
+	}
+
+	/* do not try local files if a symfs was given */
+	if (symbol_conf.symfs[0] != 0)
+		return -1;
+
+	kallsyms_allocated_filename = dso__find_kallsyms(dso, map);
+	if (!kallsyms_allocated_filename)
+		return -1;
+
+	kallsyms_filename = kallsyms_allocated_filename;
+
+do_kallsyms:
+	err = dso__load_kallsyms(dso, kallsyms_filename, map, filter);
+	if (err > 0)
+		pr_debug("Using %s for symbols\n", kallsyms_filename);
+	free(kallsyms_allocated_filename);
+
+	if (err > 0 && !dso__is_kcore(dso)) {
+		dso->binary_type = DSO_BINARY_TYPE__KALLSYMS;
+		dso__set_long_name(dso, "[kernel.kallsyms]", false);
+		map__fixup_start(map);
+		map__fixup_end(map);
+	}
+
+	return err;
+}
+
+static int dso__load_guest_kernel_sym(struct dso *dso, struct map *map,
+				      symbol_filter_t filter)
+{
+	int err;
+	const char *kallsyms_filename = NULL;
+	struct machine *machine;
+	char path[PATH_MAX];
+
+	if (!map->groups) {
+		pr_debug("Guest kernel map hasn't the point to groups\n");
+		return -1;
+	}
+	machine = map->groups->machine;
+
+	if (machine__is_default_guest(machine)) {
+		/*
+		 * if the user specified a vmlinux filename, use it and only
+		 * it, reporting errors to the user if it cannot be used.
+		 * Or use file guest_kallsyms inputted by user on commandline
+		 */
+		if (symbol_conf.default_guest_vmlinux_name != NULL) {
+			err = dso__load_vmlinux(dso, map,
+						symbol_conf.default_guest_vmlinux_name,
+						false, filter);
+			return err;
+		}
+
+		kallsyms_filename = symbol_conf.default_guest_kallsyms;
+		if (!kallsyms_filename)
+			return -1;
+	} else {
+		sprintf(path, "%s/proc/kallsyms", machine->root_dir);
+		kallsyms_filename = path;
+	}
+
+	err = dso__load_kallsyms(dso, kallsyms_filename, map, filter);
+	if (err > 0)
+		pr_debug("Using %s for symbols\n", kallsyms_filename);
+	if (err > 0 && !dso__is_kcore(dso)) {
+		dso->binary_type = DSO_BINARY_TYPE__GUEST_KALLSYMS;
+		machine__mmap_name(machine, path, sizeof(path));
+		dso__set_long_name(dso, strdup(path), true);
+		map__fixup_start(map);
+		map__fixup_end(map);
+	}
+
+	return err;
+}
+
+static void vmlinux_path__exit(void)
+{
+	while (--vmlinux_path__nr_entries >= 0)
+		zfree(&vmlinux_path[vmlinux_path__nr_entries]);
+
+	zfree(&vmlinux_path);
+}
+
+static int vmlinux_path__init(struct perf_session_env *env)
+{
+	struct utsname uts;
+	char bf[PATH_MAX];
+	char *kernel_version;
+
+	vmlinux_path = malloc(sizeof(char *) * 6);
+	if (vmlinux_path == NULL)
+		return -1;
+
+	vmlinux_path[vmlinux_path__nr_entries] = strdup("vmlinux");
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+	vmlinux_path[vmlinux_path__nr_entries] = strdup("/boot/vmlinux");
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+
+	/* only try kernel version if no symfs was given */
+	if (symbol_conf.symfs[0] != 0)
+		return 0;
+
+	if (env) {
+		kernel_version = env->os_release;
+	} else {
+		if (uname(&uts) < 0)
+			goto out_fail;
+
+		kernel_version = uts.release;
+	}
+
+	snprintf(bf, sizeof(bf), "/boot/vmlinux-%s", kernel_version);
+	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+	snprintf(bf, sizeof(bf), "/usr/lib/debug/boot/vmlinux-%s",
+		 kernel_version);
+	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+        ++vmlinux_path__nr_entries;
+	snprintf(bf, sizeof(bf), "/lib/modules/%s/build/vmlinux", kernel_version);
+	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+	snprintf(bf, sizeof(bf), "/usr/lib/debug/lib/modules/%s/vmlinux",
+		 kernel_version);
+	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		goto out_fail;
+	++vmlinux_path__nr_entries;
+
+	return 0;
+
+out_fail:
+	vmlinux_path__exit();
+	return -1;
+}
+
+int setup_list(struct strlist **list, const char *list_str,
+		      const char *list_name)
+{
+	if (list_str == NULL)
+		return 0;
+
+	*list = strlist__new(true, list_str);
+	if (!*list) {
+		pr_err("problems parsing %s list\n", list_name);
+		return -1;
+	}
+	return 0;
+}
+
+int setup_intlist(struct intlist **list, const char *list_str,
+		  const char *list_name)
+{
+	if (list_str == NULL)
+		return 0;
+
+	*list = intlist__new(list_str);
+	if (!*list) {
+		pr_err("problems parsing %s list\n", list_name);
+		return -1;
+	}
+	return 0;
+}
+
+static bool symbol__read_kptr_restrict(void)
+{
+	bool value = false;
+
+	if (geteuid() != 0) {
+		FILE *fp = fopen("/proc/sys/kernel/kptr_restrict", "r");
+		if (fp != NULL) {
+			char line[8];
+
+			if (fgets(line, sizeof(line), fp) != NULL)
+				value = atoi(line) != 0;
+
+			fclose(fp);
+		}
+	}
+
+	return value;
+}
+
+int symbol__init(struct perf_session_env *env)
+{
+	const char *symfs;
+
+	if (symbol_conf.initialized)
+		return 0;
+
+	symbol_conf.priv_size = PERF_ALIGN(symbol_conf.priv_size, sizeof(u64));
+
+	symbol__elf_init();
+
+	if (symbol_conf.sort_by_name)
+		symbol_conf.priv_size += (sizeof(struct symbol_name_rb_node) -
+					  sizeof(struct symbol));
+
+	if (symbol_conf.try_vmlinux_path && vmlinux_path__init(env) < 0)
+		return -1;
+
+	if (symbol_conf.field_sep && *symbol_conf.field_sep == '.') {
+		pr_err("'.' is the only non valid --field-separator argument\n");
+		return -1;
+	}
+
+	if (setup_list(&symbol_conf.dso_list,
+		       symbol_conf.dso_list_str, "dso") < 0)
+		return -1;
+
+	if (setup_list(&symbol_conf.comm_list,
+		       symbol_conf.comm_list_str, "comm") < 0)
+		goto out_free_dso_list;
+
+	if (setup_intlist(&symbol_conf.pid_list,
+		       symbol_conf.pid_list_str, "pid") < 0)
+		goto out_free_comm_list;
+
+	if (setup_intlist(&symbol_conf.tid_list,
+		       symbol_conf.tid_list_str, "tid") < 0)
+		goto out_free_pid_list;
+
+	if (setup_list(&symbol_conf.sym_list,
+		       symbol_conf.sym_list_str, "symbol") < 0)
+		goto out_free_tid_list;
+
+	/*
+	 * A path to symbols of "/" is identical to ""
+	 * reset here for simplicity.
+	 */
+	symfs = realpath(symbol_conf.symfs, NULL);
+	if (symfs == NULL)
+		symfs = symbol_conf.symfs;
+	if (strcmp(symfs, "/") == 0)
+		symbol_conf.symfs = "";
+	if (symfs != symbol_conf.symfs)
+		free((void *)symfs);
+
+	symbol_conf.kptr_restrict = symbol__read_kptr_restrict();
+
+	symbol_conf.initialized = true;
+	return 0;
+
+out_free_tid_list:
+	intlist__delete(symbol_conf.tid_list);
+out_free_pid_list:
+	intlist__delete(symbol_conf.pid_list);
+out_free_comm_list:
+	strlist__delete(symbol_conf.comm_list);
+out_free_dso_list:
+	strlist__delete(symbol_conf.dso_list);
+	return -1;
+}
+
+void symbol__exit(void)
+{
+	if (!symbol_conf.initialized)
+		return;
+	strlist__delete(symbol_conf.sym_list);
+	strlist__delete(symbol_conf.dso_list);
+	strlist__delete(symbol_conf.comm_list);
+	intlist__delete(symbol_conf.tid_list);
+	intlist__delete(symbol_conf.pid_list);
+	vmlinux_path__exit();
+	symbol_conf.sym_list = symbol_conf.dso_list = symbol_conf.comm_list = NULL;
+	symbol_conf.initialized = false;
+}
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
new file mode 100644
index 000000000..095615001
--- /dev/null
+++ b/tools/perf/util/symbol.h
@@ -0,0 +1,306 @@
+#ifndef __PERF_SYMBOL
+#define __PERF_SYMBOL 1
+
+#include <linux/types.h>
+#include <stdbool.h>
+#include <stdint.h>
+#include "map.h"
+#include "../perf.h"
+#include <linux/list.h>
+#include <linux/rbtree.h>
+#include <stdio.h>
+#include <byteswap.h>
+#include <libgen.h>
+#include "build-id.h"
+#include "event.h"
+#include "util.h"
+
+#ifdef HAVE_LIBELF_SUPPORT
+#include <libelf.h>
+#include <gelf.h>
+#endif
+#include <elf.h>
+
+#include "dso.h"
+
+/*
+ * libelf 0.8.x and earlier do not support ELF_C_READ_MMAP;
+ * for newer versions we can use mmap to reduce memory usage:
+ */
+#ifdef HAVE_LIBELF_MMAP_SUPPORT
+# define PERF_ELF_C_READ_MMAP ELF_C_READ_MMAP
+#else
+# define PERF_ELF_C_READ_MMAP ELF_C_READ
+#endif
+
+#ifdef HAVE_LIBELF_SUPPORT
+extern Elf_Scn *elf_section_by_name(Elf *elf, GElf_Ehdr *ep,
+				GElf_Shdr *shp, const char *name, size_t *idx);
+#endif
+
+#ifndef DMGL_PARAMS
+#define DMGL_NO_OPTS     0              /* For readability... */
+#define DMGL_PARAMS      (1 << 0)       /* Include function args */
+#define DMGL_ANSI        (1 << 1)       /* Include const, volatile, etc */
+#endif
+
+/** struct symbol - symtab entry
+ *
+ * @ignore - resolvable but tools ignore it (e.g. idle routines)
+ */
+struct symbol {
+	struct rb_node	rb_node;
+	u64		start;
+	u64		end;
+	u16		namelen;
+	u8		binding;
+	bool		ignore;
+	char		name[0];
+};
+
+void symbol__delete(struct symbol *sym);
+void symbols__delete(struct rb_root *symbols);
+
+/* symbols__for_each_entry - iterate over symbols (rb_root)
+ *
+ * @symbols: the rb_root of symbols
+ * @pos: the 'struct symbol *' to use as a loop cursor
+ * @nd: the 'struct rb_node *' to use as a temporary storage
+ */
+#define symbols__for_each_entry(symbols, pos, nd)			\
+	for (nd = rb_first(symbols);					\
+	     nd && (pos = rb_entry(nd, struct symbol, rb_node));	\
+	     nd = rb_next(nd))
+
+static inline size_t symbol__size(const struct symbol *sym)
+{
+	return sym->end - sym->start;
+}
+
+struct strlist;
+struct intlist;
+
+struct symbol_conf {
+	unsigned short	priv_size;
+	unsigned short	nr_events;
+	bool		try_vmlinux_path,
+			ignore_vmlinux,
+			ignore_vmlinux_buildid,
+			show_kernel_path,
+			use_modules,
+			allow_aliases,
+			sort_by_name,
+			show_nr_samples,
+			show_total_period,
+			use_callchain,
+			cumulate_callchain,
+			exclude_other,
+			show_cpu_utilization,
+			initialized,
+			kptr_restrict,
+			annotate_asm_raw,
+			annotate_src,
+			event_group,
+			demangle,
+			demangle_kernel,
+			filter_relative,
+			show_hist_headers,
+			branch_callstack;
+	const char	*vmlinux_name,
+			*kallsyms_name,
+			*source_prefix,
+			*field_sep;
+	const char	*default_guest_vmlinux_name,
+			*default_guest_kallsyms,
+			*default_guest_modules;
+	const char	*guestmount;
+	const char	*dso_list_str,
+			*comm_list_str,
+			*pid_list_str,
+			*tid_list_str,
+			*sym_list_str,
+			*col_width_list_str;
+       struct strlist	*dso_list,
+			*comm_list,
+			*sym_list,
+			*dso_from_list,
+			*dso_to_list,
+			*sym_from_list,
+			*sym_to_list;
+	struct intlist	*pid_list,
+			*tid_list;
+	const char	*symfs;
+};
+
+extern struct symbol_conf symbol_conf;
+
+static inline int __symbol__join_symfs(char *bf, size_t size, const char *path)
+{
+	return path__join(bf, size, symbol_conf.symfs, path);
+}
+
+#define symbol__join_symfs(bf, path) __symbol__join_symfs(bf, sizeof(bf), path)
+
+extern int vmlinux_path__nr_entries;
+extern char **vmlinux_path;
+
+static inline void *symbol__priv(struct symbol *sym)
+{
+	return ((void *)sym) - symbol_conf.priv_size;
+}
+
+struct ref_reloc_sym {
+	const char	*name;
+	u64		addr;
+	u64		unrelocated_addr;
+};
+
+struct map_symbol {
+	struct map    *map;
+	struct symbol *sym;
+	bool	      unfolded;
+	bool	      has_children;
+};
+
+struct addr_map_symbol {
+	struct map    *map;
+	struct symbol *sym;
+	u64	      addr;
+	u64	      al_addr;
+};
+
+struct branch_info {
+	struct addr_map_symbol from;
+	struct addr_map_symbol to;
+	struct branch_flags flags;
+};
+
+struct mem_info {
+	struct addr_map_symbol iaddr;
+	struct addr_map_symbol daddr;
+	union perf_mem_data_src data_src;
+};
+
+struct addr_location {
+	struct machine *machine;
+	struct thread *thread;
+	struct map    *map;
+	struct symbol *sym;
+	u64	      addr;
+	char	      level;
+	u8	      filtered;
+	u8	      cpumode;
+	s32	      cpu;
+};
+
+struct symsrc {
+	char *name;
+	int fd;
+	enum dso_binary_type type;
+
+#ifdef HAVE_LIBELF_SUPPORT
+	Elf *elf;
+	GElf_Ehdr ehdr;
+
+	Elf_Scn *opdsec;
+	size_t opdidx;
+	GElf_Shdr opdshdr;
+
+	Elf_Scn *symtab;
+	GElf_Shdr symshdr;
+
+	Elf_Scn *dynsym;
+	size_t dynsym_idx;
+	GElf_Shdr dynshdr;
+
+	bool adjust_symbols;
+	bool is_64_bit;
+#endif
+};
+
+void symsrc__destroy(struct symsrc *ss);
+int symsrc__init(struct symsrc *ss, struct dso *dso, const char *name,
+		 enum dso_binary_type type);
+bool symsrc__has_symtab(struct symsrc *ss);
+bool symsrc__possibly_runtime(struct symsrc *ss);
+
+int dso__load(struct dso *dso, struct map *map, symbol_filter_t filter);
+int dso__load_vmlinux(struct dso *dso, struct map *map,
+		      const char *vmlinux, bool vmlinux_allocated,
+		      symbol_filter_t filter);
+int dso__load_vmlinux_path(struct dso *dso, struct map *map,
+			   symbol_filter_t filter);
+int dso__load_kallsyms(struct dso *dso, const char *filename, struct map *map,
+		       symbol_filter_t filter);
+
+struct symbol *dso__find_symbol(struct dso *dso, enum map_type type,
+				u64 addr);
+struct symbol *dso__find_symbol_by_name(struct dso *dso, enum map_type type,
+					const char *name);
+struct symbol *symbol__next_by_name(struct symbol *sym);
+
+struct symbol *dso__first_symbol(struct dso *dso, enum map_type type);
+struct symbol *dso__next_symbol(struct symbol *sym);
+
+enum dso_type dso__type_fd(int fd);
+
+int filename__read_build_id(const char *filename, void *bf, size_t size);
+int sysfs__read_build_id(const char *filename, void *bf, size_t size);
+int modules__parse(const char *filename, void *arg,
+		   int (*process_module)(void *arg, const char *name,
+					 u64 start));
+int filename__read_debuglink(const char *filename, char *debuglink,
+			     size_t size);
+
+struct perf_session_env;
+int symbol__init(struct perf_session_env *env);
+void symbol__exit(void);
+void symbol__elf_init(void);
+struct symbol *symbol__new(u64 start, u64 len, u8 binding, const char *name);
+size_t symbol__fprintf_symname_offs(const struct symbol *sym,
+				    const struct addr_location *al, FILE *fp);
+size_t symbol__fprintf_symname(const struct symbol *sym, FILE *fp);
+size_t symbol__fprintf(struct symbol *sym, FILE *fp);
+bool symbol_type__is_a(char symbol_type, enum map_type map_type);
+bool symbol__restricted_filename(const char *filename,
+				 const char *restricted_filename);
+bool symbol__is_idle(struct symbol *sym);
+
+int dso__load_sym(struct dso *dso, struct map *map, struct symsrc *syms_ss,
+		  struct symsrc *runtime_ss, symbol_filter_t filter,
+		  int kmodule);
+int dso__synthesize_plt_symbols(struct dso *dso, struct symsrc *ss,
+				struct map *map, symbol_filter_t filter);
+
+void symbols__insert(struct rb_root *symbols, struct symbol *sym);
+void symbols__fixup_duplicate(struct rb_root *symbols);
+void symbols__fixup_end(struct rb_root *symbols);
+void __map_groups__fixup_end(struct map_groups *mg, enum map_type type);
+
+typedef int (*mapfn_t)(u64 start, u64 len, u64 pgoff, void *data);
+int file__read_maps(int fd, bool exe, mapfn_t mapfn, void *data,
+		    bool *is_64_bit);
+
+#define PERF_KCORE_EXTRACT "/tmp/perf-kcore-XXXXXX"
+
+struct kcore_extract {
+	char *kcore_filename;
+	u64 addr;
+	u64 offs;
+	u64 len;
+	char extract_filename[sizeof(PERF_KCORE_EXTRACT)];
+	int fd;
+};
+
+int kcore_extract__create(struct kcore_extract *kce);
+void kcore_extract__delete(struct kcore_extract *kce);
+
+int kcore_copy(const char *from_dir, const char *to_dir);
+int compare_proc_modules(const char *from, const char *to);
+
+int setup_list(struct strlist **list, const char *list_str,
+	       const char *list_name);
+int setup_intlist(struct intlist **list, const char *list_str,
+		  const char *list_name);
+
+#endif /* __PERF_SYMBOL */
diff --git a/tools/perf/util/target.c b/tools/perf/util/target.c
new file mode 100644
index 000000000..a53603b27
--- /dev/null
+++ b/tools/perf/util/target.c
@@ -0,0 +1,155 @@
+/*
+ * Helper functions for handling target threads/cpus
+ *
+ * Copyright (C) 2012, LG Electronics, Namhyung Kim <namhyung.kim@lge.com>
+ *
+ * Released under the GPL v2.
+ */
+
+#include "target.h"
+#include "debug.h"
+
+#include <pwd.h>
+#include <string.h>
+
+
+enum target_errno target__validate(struct target *target)
+{
+	enum target_errno ret = TARGET_ERRNO__SUCCESS;
+
+	if (target->pid)
+		target->tid = target->pid;
+
+	/* CPU and PID are mutually exclusive */
+	if (target->tid && target->cpu_list) {
+		target->cpu_list = NULL;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__PID_OVERRIDE_CPU;
+	}
+
+	/* UID and PID are mutually exclusive */
+	if (target->tid && target->uid_str) {
+		target->uid_str = NULL;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__PID_OVERRIDE_UID;
+	}
+
+	/* UID and CPU are mutually exclusive */
+	if (target->uid_str && target->cpu_list) {
+		target->cpu_list = NULL;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__UID_OVERRIDE_CPU;
+	}
+
+	/* PID and SYSTEM are mutually exclusive */
+	if (target->tid && target->system_wide) {
+		target->system_wide = false;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__PID_OVERRIDE_SYSTEM;
+	}
+
+	/* UID and SYSTEM are mutually exclusive */
+	if (target->uid_str && target->system_wide) {
+		target->system_wide = false;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__UID_OVERRIDE_SYSTEM;
+	}
+
+	/* THREAD and SYSTEM/CPU are mutually exclusive */
+	if (target->per_thread && (target->system_wide || target->cpu_list)) {
+		target->per_thread = false;
+		if (ret == TARGET_ERRNO__SUCCESS)
+			ret = TARGET_ERRNO__SYSTEM_OVERRIDE_THREAD;
+	}
+
+	return ret;
+}
+
+enum target_errno target__parse_uid(struct target *target)
+{
+	struct passwd pwd, *result;
+	char buf[1024];
+	const char *str = target->uid_str;
+
+	target->uid = UINT_MAX;
+	if (str == NULL)
+		return TARGET_ERRNO__SUCCESS;
+
+	/* Try user name first */
+	getpwnam_r(str, &pwd, buf, sizeof(buf), &result);
+
+	if (result == NULL) {
+		/*
+		 * The user name not found. Maybe it's a UID number.
+		 */
+		char *endptr;
+		int uid = strtol(str, &endptr, 10);
+
+		if (*endptr != '\0')
+			return TARGET_ERRNO__INVALID_UID;
+
+		getpwuid_r(uid, &pwd, buf, sizeof(buf), &result);
+
+		if (result == NULL)
+			return TARGET_ERRNO__USER_NOT_FOUND;
+	}
+
+	target->uid = result->pw_uid;
+	return TARGET_ERRNO__SUCCESS;
+}
+
+/*
+ * This must have a same ordering as the enum target_errno.
+ */
+static const char *target__error_str[] = {
+	"PID/TID switch overriding CPU",
+	"PID/TID switch overriding UID",
+	"UID switch overriding CPU",
+	"PID/TID switch overriding SYSTEM",
+	"UID switch overriding SYSTEM",
+	"SYSTEM/CPU switch overriding PER-THREAD",
+	"Invalid User: %s",
+	"Problems obtaining information for user %s",
+};
+
+int target__strerror(struct target *target, int errnum,
+			  char *buf, size_t buflen)
+{
+	int idx;
+	const char *msg;
+
+	BUG_ON(buflen == 0);
+
+	if (errnum >= 0) {
+		const char *err = strerror_r(errnum, buf, buflen);
+
+		if (err != buf)
+			scnprintf(buf, buflen, "%s", err);
+
+		return 0;
+	}
+
+	if (errnum <  __TARGET_ERRNO__START || errnum >= __TARGET_ERRNO__END)
+		return -1;
+
+	idx = errnum - __TARGET_ERRNO__START;
+	msg = target__error_str[idx];
+
+	switch (errnum) {
+	case TARGET_ERRNO__PID_OVERRIDE_CPU ...
+	     TARGET_ERRNO__SYSTEM_OVERRIDE_THREAD:
+		snprintf(buf, buflen, "%s", msg);
+		break;
+
+	case TARGET_ERRNO__INVALID_UID:
+	case TARGET_ERRNO__USER_NOT_FOUND:
+		snprintf(buf, buflen, msg, target->uid_str);
+		break;
+
+	default:
+		/* cannot reach here */
+		break;
+	}
+
+	return 0;
+}
diff --git a/tools/perf/util/target.h b/tools/perf/util/target.h
new file mode 100644
index 000000000..7381b1ca4
--- /dev/null
+++ b/tools/perf/util/target.h
@@ -0,0 +1,79 @@
+#ifndef _PERF_TARGET_H
+#define _PERF_TARGET_H
+
+#include <stdbool.h>
+#include <sys/types.h>
+
+struct target {
+	const char   *pid;
+	const char   *tid;
+	const char   *cpu_list;
+	const char   *uid_str;
+	uid_t	     uid;
+	bool	     system_wide;
+	bool	     uses_mmap;
+	bool	     default_per_cpu;
+	bool	     per_thread;
+};
+
+enum target_errno {
+	TARGET_ERRNO__SUCCESS		= 0,
+
+	/*
+	 * Choose an arbitrary negative big number not to clash with standard
+	 * errno since SUS requires the errno has distinct positive values.
+	 * See 'Issue 6' in the link below.
+	 *
+	 * http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/errno.h.html
+	 */
+	__TARGET_ERRNO__START		= -10000,
+
+	/* for target__validate() */
+	TARGET_ERRNO__PID_OVERRIDE_CPU	= __TARGET_ERRNO__START,
+	TARGET_ERRNO__PID_OVERRIDE_UID,
+	TARGET_ERRNO__UID_OVERRIDE_CPU,
+	TARGET_ERRNO__PID_OVERRIDE_SYSTEM,
+	TARGET_ERRNO__UID_OVERRIDE_SYSTEM,
+	TARGET_ERRNO__SYSTEM_OVERRIDE_THREAD,
+
+	/* for target__parse_uid() */
+	TARGET_ERRNO__INVALID_UID,
+	TARGET_ERRNO__USER_NOT_FOUND,
+
+	__TARGET_ERRNO__END,
+};
+
+enum target_errno target__validate(struct target *target);
+enum target_errno target__parse_uid(struct target *target);
+
+int target__strerror(struct target *target, int errnum, char *buf, size_t buflen);
+
+static inline bool target__has_task(struct target *target)
+{
+	return target->tid || target->pid || target->uid_str;
+}
+
+static inline bool target__has_cpu(struct target *target)
+{
+	return target->system_wide || target->cpu_list;
+}
+
+static inline bool target__none(struct target *target)
+{
+	return !target__has_task(target) && !target__has_cpu(target);
+}
+
+static inline bool target__uses_dummy_map(struct target *target)
+{
+	bool use_dummy = false;
+
+	if (target->default_per_cpu)
+		use_dummy = target->per_thread ? true : false;
+	else if (target__has_task(target) ||
+	         (!target__has_cpu(target) && !target->uses_mmap))
+		use_dummy = true;
+
+	return use_dummy;
+}
+
+#endif /* _PERF_TARGET_H */
diff --git a/tools/perf/util/thread-stack.c b/tools/perf/util/thread-stack.c
new file mode 100644
index 000000000..9ed59a452
--- /dev/null
+++ b/tools/perf/util/thread-stack.c
@@ -0,0 +1,747 @@
+/*
+ * thread-stack.c: Synthesize a thread's stack using call / return events
+ * Copyright (c) 2014, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ */
+
+#include <linux/rbtree.h>
+#include <linux/list.h>
+#include "thread.h"
+#include "event.h"
+#include "machine.h"
+#include "util.h"
+#include "debug.h"
+#include "symbol.h"
+#include "comm.h"
+#include "thread-stack.h"
+
+#define CALL_PATH_BLOCK_SHIFT 8
+#define CALL_PATH_BLOCK_SIZE (1 << CALL_PATH_BLOCK_SHIFT)
+#define CALL_PATH_BLOCK_MASK (CALL_PATH_BLOCK_SIZE - 1)
+
+struct call_path_block {
+	struct call_path cp[CALL_PATH_BLOCK_SIZE];
+	struct list_head node;
+};
+
+/**
+ * struct call_path_root - root of all call paths.
+ * @call_path: root call path
+ * @blocks: list of blocks to store call paths
+ * @next: next free space
+ * @sz: number of spaces
+ */
+struct call_path_root {
+	struct call_path call_path;
+	struct list_head blocks;
+	size_t next;
+	size_t sz;
+};
+
+/**
+ * struct call_return_processor - provides a call-back to consume call-return
+ *                                information.
+ * @cpr: call path root
+ * @process: call-back that accepts call/return information
+ * @data: anonymous data for call-back
+ */
+struct call_return_processor {
+	struct call_path_root *cpr;
+	int (*process)(struct call_return *cr, void *data);
+	void *data;
+};
+
+#define STACK_GROWTH 2048
+
+/**
+ * struct thread_stack_entry - thread stack entry.
+ * @ret_addr: return address
+ * @timestamp: timestamp (if known)
+ * @ref: external reference (e.g. db_id of sample)
+ * @branch_count: the branch count when the entry was created
+ * @cp: call path
+ * @no_call: a 'call' was not seen
+ */
+struct thread_stack_entry {
+	u64 ret_addr;
+	u64 timestamp;
+	u64 ref;
+	u64 branch_count;
+	struct call_path *cp;
+	bool no_call;
+};
+
+/**
+ * struct thread_stack - thread stack constructed from 'call' and 'return'
+ *                       branch samples.
+ * @stack: array that holds the stack
+ * @cnt: number of entries in the stack
+ * @sz: current maximum stack size
+ * @trace_nr: current trace number
+ * @branch_count: running branch count
+ * @kernel_start: kernel start address
+ * @last_time: last timestamp
+ * @crp: call/return processor
+ * @comm: current comm
+ */
+struct thread_stack {
+	struct thread_stack_entry *stack;
+	size_t cnt;
+	size_t sz;
+	u64 trace_nr;
+	u64 branch_count;
+	u64 kernel_start;
+	u64 last_time;
+	struct call_return_processor *crp;
+	struct comm *comm;
+};
+
+static int thread_stack__grow(struct thread_stack *ts)
+{
+	struct thread_stack_entry *new_stack;
+	size_t sz, new_sz;
+
+	new_sz = ts->sz + STACK_GROWTH;
+	sz = new_sz * sizeof(struct thread_stack_entry);
+
+	new_stack = realloc(ts->stack, sz);
+	if (!new_stack)
+		return -ENOMEM;
+
+	ts->stack = new_stack;
+	ts->sz = new_sz;
+
+	return 0;
+}
+
+static struct thread_stack *thread_stack__new(struct thread *thread,
+					      struct call_return_processor *crp)
+{
+	struct thread_stack *ts;
+
+	ts = zalloc(sizeof(struct thread_stack));
+	if (!ts)
+		return NULL;
+
+	if (thread_stack__grow(ts)) {
+		free(ts);
+		return NULL;
+	}
+
+	if (thread->mg && thread->mg->machine)
+		ts->kernel_start = machine__kernel_start(thread->mg->machine);
+	else
+		ts->kernel_start = 1ULL << 63;
+	ts->crp = crp;
+
+	return ts;
+}
+
+static int thread_stack__push(struct thread_stack *ts, u64 ret_addr)
+{
+	int err = 0;
+
+	if (ts->cnt == ts->sz) {
+		err = thread_stack__grow(ts);
+		if (err) {
+			pr_warning("Out of memory: discarding thread stack\n");
+			ts->cnt = 0;
+		}
+	}
+
+	ts->stack[ts->cnt++].ret_addr = ret_addr;
+
+	return err;
+}
+
+static void thread_stack__pop(struct thread_stack *ts, u64 ret_addr)
+{
+	size_t i;
+
+	/*
+	 * In some cases there may be functions which are not seen to return.
+	 * For example when setjmp / longjmp has been used.  Or the perf context
+	 * switch in the kernel which doesn't stop and start tracing in exactly
+	 * the same code path.  When that happens the return address will be
+	 * further down the stack.  If the return address is not found at all,
+	 * we assume the opposite (i.e. this is a return for a call that wasn't
+	 * seen for some reason) and leave the stack alone.
+	 */
+	for (i = ts->cnt; i; ) {
+		if (ts->stack[--i].ret_addr == ret_addr) {
+			ts->cnt = i;
+			return;
+		}
+	}
+}
+
+static bool thread_stack__in_kernel(struct thread_stack *ts)
+{
+	if (!ts->cnt)
+		return false;
+
+	return ts->stack[ts->cnt - 1].cp->in_kernel;
+}
+
+static int thread_stack__call_return(struct thread *thread,
+				     struct thread_stack *ts, size_t idx,
+				     u64 timestamp, u64 ref, bool no_return)
+{
+	struct call_return_processor *crp = ts->crp;
+	struct thread_stack_entry *tse;
+	struct call_return cr = {
+		.thread = thread,
+		.comm = ts->comm,
+		.db_id = 0,
+	};
+
+	tse = &ts->stack[idx];
+	cr.cp = tse->cp;
+	cr.call_time = tse->timestamp;
+	cr.return_time = timestamp;
+	cr.branch_count = ts->branch_count - tse->branch_count;
+	cr.call_ref = tse->ref;
+	cr.return_ref = ref;
+	if (tse->no_call)
+		cr.flags |= CALL_RETURN_NO_CALL;
+	if (no_return)
+		cr.flags |= CALL_RETURN_NO_RETURN;
+
+	return crp->process(&cr, crp->data);
+}
+
+static int thread_stack__flush(struct thread *thread, struct thread_stack *ts)
+{
+	struct call_return_processor *crp = ts->crp;
+	int err;
+
+	if (!crp) {
+		ts->cnt = 0;
+		return 0;
+	}
+
+	while (ts->cnt) {
+		err = thread_stack__call_return(thread, ts, --ts->cnt,
+						ts->last_time, 0, true);
+		if (err) {
+			pr_err("Error flushing thread stack!\n");
+			ts->cnt = 0;
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+int thread_stack__event(struct thread *thread, u32 flags, u64 from_ip,
+			u64 to_ip, u16 insn_len, u64 trace_nr)
+{
+	if (!thread)
+		return -EINVAL;
+
+	if (!thread->ts) {
+		thread->ts = thread_stack__new(thread, NULL);
+		if (!thread->ts) {
+			pr_warning("Out of memory: no thread stack\n");
+			return -ENOMEM;
+		}
+		thread->ts->trace_nr = trace_nr;
+	}
+
+	/*
+	 * When the trace is discontinuous, the trace_nr changes.  In that case
+	 * the stack might be completely invalid.  Better to report nothing than
+	 * to report something misleading, so flush the stack.
+	 */
+	if (trace_nr != thread->ts->trace_nr) {
+		if (thread->ts->trace_nr)
+			thread_stack__flush(thread, thread->ts);
+		thread->ts->trace_nr = trace_nr;
+	}
+
+	/* Stop here if thread_stack__process() is in use */
+	if (thread->ts->crp)
+		return 0;
+
+	if (flags & PERF_IP_FLAG_CALL) {
+		u64 ret_addr;
+
+		if (!to_ip)
+			return 0;
+		ret_addr = from_ip + insn_len;
+		if (ret_addr == to_ip)
+			return 0; /* Zero-length calls are excluded */
+		return thread_stack__push(thread->ts, ret_addr);
+	} else if (flags & PERF_IP_FLAG_RETURN) {
+		if (!from_ip)
+			return 0;
+		thread_stack__pop(thread->ts, to_ip);
+	}
+
+	return 0;
+}
+
+void thread_stack__set_trace_nr(struct thread *thread, u64 trace_nr)
+{
+	if (!thread || !thread->ts)
+		return;
+
+	if (trace_nr != thread->ts->trace_nr) {
+		if (thread->ts->trace_nr)
+			thread_stack__flush(thread, thread->ts);
+		thread->ts->trace_nr = trace_nr;
+	}
+}
+
+void thread_stack__free(struct thread *thread)
+{
+	if (thread->ts) {
+		thread_stack__flush(thread, thread->ts);
+		zfree(&thread->ts->stack);
+		zfree(&thread->ts);
+	}
+}
+
+void thread_stack__sample(struct thread *thread, struct ip_callchain *chain,
+			  size_t sz, u64 ip)
+{
+	size_t i;
+
+	if (!thread || !thread->ts)
+		chain->nr = 1;
+	else
+		chain->nr = min(sz, thread->ts->cnt + 1);
+
+	chain->ips[0] = ip;
+
+	for (i = 1; i < chain->nr; i++)
+		chain->ips[i] = thread->ts->stack[thread->ts->cnt - i].ret_addr;
+}
+
+static void call_path__init(struct call_path *cp, struct call_path *parent,
+			    struct symbol *sym, u64 ip, bool in_kernel)
+{
+	cp->parent = parent;
+	cp->sym = sym;
+	cp->ip = sym ? 0 : ip;
+	cp->db_id = 0;
+	cp->in_kernel = in_kernel;
+	RB_CLEAR_NODE(&cp->rb_node);
+	cp->children = RB_ROOT;
+}
+
+static struct call_path_root *call_path_root__new(void)
+{
+	struct call_path_root *cpr;
+
+	cpr = zalloc(sizeof(struct call_path_root));
+	if (!cpr)
+		return NULL;
+	call_path__init(&cpr->call_path, NULL, NULL, 0, false);
+	INIT_LIST_HEAD(&cpr->blocks);
+	return cpr;
+}
+
+static void call_path_root__free(struct call_path_root *cpr)
+{
+	struct call_path_block *pos, *n;
+
+	list_for_each_entry_safe(pos, n, &cpr->blocks, node) {
+		list_del(&pos->node);
+		free(pos);
+	}
+	free(cpr);
+}
+
+static struct call_path *call_path__new(struct call_path_root *cpr,
+					struct call_path *parent,
+					struct symbol *sym, u64 ip,
+					bool in_kernel)
+{
+	struct call_path_block *cpb;
+	struct call_path *cp;
+	size_t n;
+
+	if (cpr->next < cpr->sz) {
+		cpb = list_last_entry(&cpr->blocks, struct call_path_block,
+				      node);
+	} else {
+		cpb = zalloc(sizeof(struct call_path_block));
+		if (!cpb)
+			return NULL;
+		list_add_tail(&cpb->node, &cpr->blocks);
+		cpr->sz += CALL_PATH_BLOCK_SIZE;
+	}
+
+	n = cpr->next++ & CALL_PATH_BLOCK_MASK;
+	cp = &cpb->cp[n];
+
+	call_path__init(cp, parent, sym, ip, in_kernel);
+
+	return cp;
+}
+
+static struct call_path *call_path__findnew(struct call_path_root *cpr,
+					    struct call_path *parent,
+					    struct symbol *sym, u64 ip, u64 ks)
+{
+	struct rb_node **p;
+	struct rb_node *node_parent = NULL;
+	struct call_path *cp;
+	bool in_kernel = ip >= ks;
+
+	if (sym)
+		ip = 0;
+
+	if (!parent)
+		return call_path__new(cpr, parent, sym, ip, in_kernel);
+
+	p = &parent->children.rb_node;
+	while (*p != NULL) {
+		node_parent = *p;
+		cp = rb_entry(node_parent, struct call_path, rb_node);
+
+		if (cp->sym == sym && cp->ip == ip)
+			return cp;
+
+		if (sym < cp->sym || (sym == cp->sym && ip < cp->ip))
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	cp = call_path__new(cpr, parent, sym, ip, in_kernel);
+	if (!cp)
+		return NULL;
+
+	rb_link_node(&cp->rb_node, node_parent, p);
+	rb_insert_color(&cp->rb_node, &parent->children);
+
+	return cp;
+}
+
+struct call_return_processor *
+call_return_processor__new(int (*process)(struct call_return *cr, void *data),
+			   void *data)
+{
+	struct call_return_processor *crp;
+
+	crp = zalloc(sizeof(struct call_return_processor));
+	if (!crp)
+		return NULL;
+	crp->cpr = call_path_root__new();
+	if (!crp->cpr)
+		goto out_free;
+	crp->process = process;
+	crp->data = data;
+	return crp;
+
+out_free:
+	free(crp);
+	return NULL;
+}
+
+void call_return_processor__free(struct call_return_processor *crp)
+{
+	if (crp) {
+		call_path_root__free(crp->cpr);
+		free(crp);
+	}
+}
+
+static int thread_stack__push_cp(struct thread_stack *ts, u64 ret_addr,
+				 u64 timestamp, u64 ref, struct call_path *cp,
+				 bool no_call)
+{
+	struct thread_stack_entry *tse;
+	int err;
+
+	if (ts->cnt == ts->sz) {
+		err = thread_stack__grow(ts);
+		if (err)
+			return err;
+	}
+
+	tse = &ts->stack[ts->cnt++];
+	tse->ret_addr = ret_addr;
+	tse->timestamp = timestamp;
+	tse->ref = ref;
+	tse->branch_count = ts->branch_count;
+	tse->cp = cp;
+	tse->no_call = no_call;
+
+	return 0;
+}
+
+static int thread_stack__pop_cp(struct thread *thread, struct thread_stack *ts,
+				u64 ret_addr, u64 timestamp, u64 ref,
+				struct symbol *sym)
+{
+	int err;
+
+	if (!ts->cnt)
+		return 1;
+
+	if (ts->cnt == 1) {
+		struct thread_stack_entry *tse = &ts->stack[0];
+
+		if (tse->cp->sym == sym)
+			return thread_stack__call_return(thread, ts, --ts->cnt,
+							 timestamp, ref, false);
+	}
+
+	if (ts->stack[ts->cnt - 1].ret_addr == ret_addr) {
+		return thread_stack__call_return(thread, ts, --ts->cnt,
+						 timestamp, ref, false);
+	} else {
+		size_t i = ts->cnt - 1;
+
+		while (i--) {
+			if (ts->stack[i].ret_addr != ret_addr)
+				continue;
+			i += 1;
+			while (ts->cnt > i) {
+				err = thread_stack__call_return(thread, ts,
+								--ts->cnt,
+								timestamp, ref,
+								true);
+				if (err)
+					return err;
+			}
+			return thread_stack__call_return(thread, ts, --ts->cnt,
+							 timestamp, ref, false);
+		}
+	}
+
+	return 1;
+}
+
+static int thread_stack__bottom(struct thread *thread, struct thread_stack *ts,
+				struct perf_sample *sample,
+				struct addr_location *from_al,
+				struct addr_location *to_al, u64 ref)
+{
+	struct call_path_root *cpr = ts->crp->cpr;
+	struct call_path *cp;
+	struct symbol *sym;
+	u64 ip;
+
+	if (sample->ip) {
+		ip = sample->ip;
+		sym = from_al->sym;
+	} else if (sample->addr) {
+		ip = sample->addr;
+		sym = to_al->sym;
+	} else {
+		return 0;
+	}
+
+	cp = call_path__findnew(cpr, &cpr->call_path, sym, ip,
+				ts->kernel_start);
+	if (!cp)
+		return -ENOMEM;
+
+	return thread_stack__push_cp(thread->ts, ip, sample->time, ref, cp,
+				     true);
+}
+
+static int thread_stack__no_call_return(struct thread *thread,
+					struct thread_stack *ts,
+					struct perf_sample *sample,
+					struct addr_location *from_al,
+					struct addr_location *to_al, u64 ref)
+{
+	struct call_path_root *cpr = ts->crp->cpr;
+	struct call_path *cp, *parent;
+	u64 ks = ts->kernel_start;
+	int err;
+
+	if (sample->ip >= ks && sample->addr < ks) {
+		/* Return to userspace, so pop all kernel addresses */
+		while (thread_stack__in_kernel(ts)) {
+			err = thread_stack__call_return(thread, ts, --ts->cnt,
+							sample->time, ref,
+							true);
+			if (err)
+				return err;
+		}
+
+		/* If the stack is empty, push the userspace address */
+		if (!ts->cnt) {
+			cp = call_path__findnew(cpr, &cpr->call_path,
+						to_al->sym, sample->addr,
+						ts->kernel_start);
+			if (!cp)
+				return -ENOMEM;
+			return thread_stack__push_cp(ts, 0, sample->time, ref,
+						     cp, true);
+		}
+	} else if (thread_stack__in_kernel(ts) && sample->ip < ks) {
+		/* Return to userspace, so pop all kernel addresses */
+		while (thread_stack__in_kernel(ts)) {
+			err = thread_stack__call_return(thread, ts, --ts->cnt,
+							sample->time, ref,
+							true);
+			if (err)
+				return err;
+		}
+	}
+
+	if (ts->cnt)
+		parent = ts->stack[ts->cnt - 1].cp;
+	else
+		parent = &cpr->call_path;
+
+	/* This 'return' had no 'call', so push and pop top of stack */
+	cp = call_path__findnew(cpr, parent, from_al->sym, sample->ip,
+				ts->kernel_start);
+	if (!cp)
+		return -ENOMEM;
+
+	err = thread_stack__push_cp(ts, sample->addr, sample->time, ref, cp,
+				    true);
+	if (err)
+		return err;
+
+	return thread_stack__pop_cp(thread, ts, sample->addr, sample->time, ref,
+				    to_al->sym);
+}
+
+static int thread_stack__trace_begin(struct thread *thread,
+				     struct thread_stack *ts, u64 timestamp,
+				     u64 ref)
+{
+	struct thread_stack_entry *tse;
+	int err;
+
+	if (!ts->cnt)
+		return 0;
+
+	/* Pop trace end */
+	tse = &ts->stack[ts->cnt - 1];
+	if (tse->cp->sym == NULL && tse->cp->ip == 0) {
+		err = thread_stack__call_return(thread, ts, --ts->cnt,
+						timestamp, ref, false);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int thread_stack__trace_end(struct thread_stack *ts,
+				   struct perf_sample *sample, u64 ref)
+{
+	struct call_path_root *cpr = ts->crp->cpr;
+	struct call_path *cp;
+	u64 ret_addr;
+
+	/* No point having 'trace end' on the bottom of the stack */
+	if (!ts->cnt || (ts->cnt == 1 && ts->stack[0].ref == ref))
+		return 0;
+
+	cp = call_path__findnew(cpr, ts->stack[ts->cnt - 1].cp, NULL, 0,
+				ts->kernel_start);
+	if (!cp)
+		return -ENOMEM;
+
+	ret_addr = sample->ip + sample->insn_len;
+
+	return thread_stack__push_cp(ts, ret_addr, sample->time, ref, cp,
+				     false);
+}
+
+int thread_stack__process(struct thread *thread, struct comm *comm,
+			  struct perf_sample *sample,
+			  struct addr_location *from_al,
+			  struct addr_location *to_al, u64 ref,
+			  struct call_return_processor *crp)
+{
+	struct thread_stack *ts = thread->ts;
+	int err = 0;
+
+	if (ts) {
+		if (!ts->crp) {
+			/* Supersede thread_stack__event() */
+			thread_stack__free(thread);
+			thread->ts = thread_stack__new(thread, crp);
+			if (!thread->ts)
+				return -ENOMEM;
+			ts = thread->ts;
+			ts->comm = comm;
+		}
+	} else {
+		thread->ts = thread_stack__new(thread, crp);
+		if (!thread->ts)
+			return -ENOMEM;
+		ts = thread->ts;
+		ts->comm = comm;
+	}
+
+	/* Flush stack on exec */
+	if (ts->comm != comm && thread->pid_ == thread->tid) {
+		err = thread_stack__flush(thread, ts);
+		if (err)
+			return err;
+		ts->comm = comm;
+	}
+
+	/* If the stack is empty, put the current symbol on the stack */
+	if (!ts->cnt) {
+		err = thread_stack__bottom(thread, ts, sample, from_al, to_al,
+					   ref);
+		if (err)
+			return err;
+	}
+
+	ts->branch_count += 1;
+	ts->last_time = sample->time;
+
+	if (sample->flags & PERF_IP_FLAG_CALL) {
+		struct call_path_root *cpr = ts->crp->cpr;
+		struct call_path *cp;
+		u64 ret_addr;
+
+		if (!sample->ip || !sample->addr)
+			return 0;
+
+		ret_addr = sample->ip + sample->insn_len;
+		if (ret_addr == sample->addr)
+			return 0; /* Zero-length calls are excluded */
+
+		cp = call_path__findnew(cpr, ts->stack[ts->cnt - 1].cp,
+					to_al->sym, sample->addr,
+					ts->kernel_start);
+		if (!cp)
+			return -ENOMEM;
+		err = thread_stack__push_cp(ts, ret_addr, sample->time, ref,
+					    cp, false);
+	} else if (sample->flags & PERF_IP_FLAG_RETURN) {
+		if (!sample->ip || !sample->addr)
+			return 0;
+
+		err = thread_stack__pop_cp(thread, ts, sample->addr,
+					   sample->time, ref, from_al->sym);
+		if (err) {
+			if (err < 0)
+				return err;
+			err = thread_stack__no_call_return(thread, ts, sample,
+							   from_al, to_al, ref);
+		}
+	} else if (sample->flags & PERF_IP_FLAG_TRACE_BEGIN) {
+		err = thread_stack__trace_begin(thread, ts, sample->time, ref);
+	} else if (sample->flags & PERF_IP_FLAG_TRACE_END) {
+		err = thread_stack__trace_end(ts, sample, ref);
+	}
+
+	return err;
+}
diff --git a/tools/perf/util/thread-stack.h b/tools/perf/util/thread-stack.h
new file mode 100644
index 000000000..b843bbef8
--- /dev/null
+++ b/tools/perf/util/thread-stack.h
@@ -0,0 +1,111 @@
+/*
+ * thread-stack.h: Synthesize a thread's stack using call / return events
+ * Copyright (c) 2014, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ */
+
+#ifndef __PERF_THREAD_STACK_H
+#define __PERF_THREAD_STACK_H
+
+#include <sys/types.h>
+
+#include <linux/types.h>
+#include <linux/rbtree.h>
+
+struct thread;
+struct comm;
+struct ip_callchain;
+struct symbol;
+struct dso;
+struct call_return_processor;
+struct comm;
+struct perf_sample;
+struct addr_location;
+
+/*
+ * Call/Return flags.
+ *
+ * CALL_RETURN_NO_CALL: 'return' but no matching 'call'
+ * CALL_RETURN_NO_RETURN: 'call' but no matching 'return'
+ */
+enum {
+	CALL_RETURN_NO_CALL	= 1 << 0,
+	CALL_RETURN_NO_RETURN	= 1 << 1,
+};
+
+/**
+ * struct call_return - paired call/return information.
+ * @thread: thread in which call/return occurred
+ * @comm: comm in which call/return occurred
+ * @cp: call path
+ * @call_time: timestamp of call (if known)
+ * @return_time: timestamp of return (if known)
+ * @branch_count: number of branches seen between call and return
+ * @call_ref: external reference to 'call' sample (e.g. db_id)
+ * @return_ref:  external reference to 'return' sample (e.g. db_id)
+ * @db_id: id used for db-export
+ * @flags: Call/Return flags
+ */
+struct call_return {
+	struct thread *thread;
+	struct comm *comm;
+	struct call_path *cp;
+	u64 call_time;
+	u64 return_time;
+	u64 branch_count;
+	u64 call_ref;
+	u64 return_ref;
+	u64 db_id;
+	u32 flags;
+};
+
+/**
+ * struct call_path - node in list of calls leading to a function call.
+ * @parent: call path to the parent function call
+ * @sym: symbol of function called
+ * @ip: only if sym is null, the ip of the function
+ * @db_id: id used for db-export
+ * @in_kernel: whether function is a in the kernel
+ * @rb_node: node in parent's tree of called functions
+ * @children: tree of call paths of functions called
+ *
+ * In combination with the call_return structure, the call_path structure
+ * defines a context-sensitve call-graph.
+ */
+struct call_path {
+	struct call_path *parent;
+	struct symbol *sym;
+	u64 ip;
+	u64 db_id;
+	bool in_kernel;
+	struct rb_node rb_node;
+	struct rb_root children;
+};
+
+int thread_stack__event(struct thread *thread, u32 flags, u64 from_ip,
+			u64 to_ip, u16 insn_len, u64 trace_nr);
+void thread_stack__set_trace_nr(struct thread *thread, u64 trace_nr);
+void thread_stack__sample(struct thread *thread, struct ip_callchain *chain,
+			  size_t sz, u64 ip);
+void thread_stack__free(struct thread *thread);
+
+struct call_return_processor *
+call_return_processor__new(int (*process)(struct call_return *cr, void *data),
+			   void *data);
+void call_return_processor__free(struct call_return_processor *crp);
+int thread_stack__process(struct thread *thread, struct comm *comm,
+			  struct perf_sample *sample,
+			  struct addr_location *from_al,
+			  struct addr_location *to_al, u64 ref,
+			  struct call_return_processor *crp);
+
+#endif
diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
new file mode 100644
index 000000000..1c8fbc958
--- /dev/null
+++ b/tools/perf/util/thread.c
@@ -0,0 +1,232 @@
+#include "../perf.h"
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include "session.h"
+#include "thread.h"
+#include "thread-stack.h"
+#include "util.h"
+#include "debug.h"
+#include "comm.h"
+#include "unwind.h"
+
+int thread__init_map_groups(struct thread *thread, struct machine *machine)
+{
+	struct thread *leader;
+	pid_t pid = thread->pid_;
+
+	if (pid == thread->tid || pid == -1) {
+		thread->mg = map_groups__new(machine);
+	} else {
+		leader = machine__findnew_thread(machine, pid, pid);
+		if (leader)
+			thread->mg = map_groups__get(leader->mg);
+	}
+
+	return thread->mg ? 0 : -1;
+}
+
+struct thread *thread__new(pid_t pid, pid_t tid)
+{
+	char *comm_str;
+	struct comm *comm;
+	struct thread *thread = zalloc(sizeof(*thread));
+
+	if (thread != NULL) {
+		thread->pid_ = pid;
+		thread->tid = tid;
+		thread->ppid = -1;
+		thread->cpu = -1;
+		INIT_LIST_HEAD(&thread->comm_list);
+
+		if (unwind__prepare_access(thread) < 0)
+			goto err_thread;
+
+		comm_str = malloc(32);
+		if (!comm_str)
+			goto err_thread;
+
+		snprintf(comm_str, 32, ":%d", tid);
+		comm = comm__new(comm_str, 0, false);
+		free(comm_str);
+		if (!comm)
+			goto err_thread;
+
+		list_add(&comm->list, &thread->comm_list);
+
+	}
+
+	return thread;
+
+err_thread:
+	free(thread);
+	return NULL;
+}
+
+void thread__delete(struct thread *thread)
+{
+	struct comm *comm, *tmp;
+
+	thread_stack__free(thread);
+
+	if (thread->mg) {
+		map_groups__put(thread->mg);
+		thread->mg = NULL;
+	}
+	list_for_each_entry_safe(comm, tmp, &thread->comm_list, list) {
+		list_del(&comm->list);
+		comm__free(comm);
+	}
+	unwind__finish_access(thread);
+
+	free(thread);
+}
+
+struct thread *thread__get(struct thread *thread)
+{
+	++thread->refcnt;
+	return thread;
+}
+
+void thread__put(struct thread *thread)
+{
+	if (thread && --thread->refcnt == 0) {
+		list_del_init(&thread->node);
+		thread__delete(thread);
+	}
+}
+
+struct comm *thread__comm(const struct thread *thread)
+{
+	if (list_empty(&thread->comm_list))
+		return NULL;
+
+	return list_first_entry(&thread->comm_list, struct comm, list);
+}
+
+struct comm *thread__exec_comm(const struct thread *thread)
+{
+	struct comm *comm, *last = NULL;
+
+	list_for_each_entry(comm, &thread->comm_list, list) {
+		if (comm->exec)
+			return comm;
+		last = comm;
+	}
+
+	return last;
+}
+
+int __thread__set_comm(struct thread *thread, const char *str, u64 timestamp,
+		       bool exec)
+{
+	struct comm *new, *curr = thread__comm(thread);
+	int err;
+
+	/* Override the default :tid entry */
+	if (!thread->comm_set) {
+		err = comm__override(curr, str, timestamp, exec);
+		if (err)
+			return err;
+	} else {
+		new = comm__new(str, timestamp, exec);
+		if (!new)
+			return -ENOMEM;
+		list_add(&new->list, &thread->comm_list);
+
+		if (exec)
+			unwind__flush_access(thread);
+	}
+
+	thread->comm_set = true;
+
+	return 0;
+}
+
+const char *thread__comm_str(const struct thread *thread)
+{
+	const struct comm *comm = thread__comm(thread);
+
+	if (!comm)
+		return NULL;
+
+	return comm__str(comm);
+}
+
+/* CHECKME: it should probably better return the max comm len from its comm list */
+int thread__comm_len(struct thread *thread)
+{
+	if (!thread->comm_len) {
+		const char *comm = thread__comm_str(thread);
+		if (!comm)
+			return 0;
+		thread->comm_len = strlen(comm);
+	}
+
+	return thread->comm_len;
+}
+
+size_t thread__fprintf(struct thread *thread, FILE *fp)
+{
+	return fprintf(fp, "Thread %d %s\n", thread->tid, thread__comm_str(thread)) +
+	       map_groups__fprintf(thread->mg, fp);
+}
+
+void thread__insert_map(struct thread *thread, struct map *map)
+{
+	map_groups__fixup_overlappings(thread->mg, map, stderr);
+	map_groups__insert(thread->mg, map);
+}
+
+static int thread__clone_map_groups(struct thread *thread,
+				    struct thread *parent)
+{
+	int i;
+
+	/* This is new thread, we share map groups for process. */
+	if (thread->pid_ == parent->pid_)
+		return 0;
+
+	/* But this one is new process, copy maps. */
+	for (i = 0; i < MAP__NR_TYPES; ++i)
+		if (map_groups__clone(thread->mg, parent->mg, i) < 0)
+			return -ENOMEM;
+
+	return 0;
+}
+
+int thread__fork(struct thread *thread, struct thread *parent, u64 timestamp)
+{
+	int err;
+
+	if (parent->comm_set) {
+		const char *comm = thread__comm_str(parent);
+		if (!comm)
+			return -ENOMEM;
+		err = thread__set_comm(thread, comm, timestamp);
+		if (err)
+			return err;
+	}
+
+	thread->ppid = parent->tid;
+	return thread__clone_map_groups(thread, parent);
+}
+
+void thread__find_cpumode_addr_location(struct thread *thread,
+					enum map_type type, u64 addr,
+					struct addr_location *al)
+{
+	size_t i;
+	const u8 const cpumodes[] = {
+		PERF_RECORD_MISC_USER,
+		PERF_RECORD_MISC_KERNEL,
+		PERF_RECORD_MISC_GUEST_USER,
+		PERF_RECORD_MISC_GUEST_KERNEL
+	};
+
+	for (i = 0; i < ARRAY_SIZE(cpumodes); i++) {
+		thread__find_addr_location(thread, cpumodes[i], type, addr, al);
+		if (al->map)
+			break;
+	}
+}
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
new file mode 100644
index 000000000..9b8a54dc3
--- /dev/null
+++ b/tools/perf/util/thread.h
@@ -0,0 +1,117 @@
+#ifndef __PERF_THREAD_H
+#define __PERF_THREAD_H
+
+#include <linux/rbtree.h>
+#include <linux/list.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include "symbol.h"
+#include <strlist.h>
+#include <intlist.h>
+
+struct thread_stack;
+
+struct thread {
+	union {
+		struct rb_node	 rb_node;
+		struct list_head node;
+	};
+	struct map_groups	*mg;
+	pid_t			pid_; /* Not all tools update this */
+	pid_t			tid;
+	pid_t			ppid;
+	int			cpu;
+	int			refcnt;
+	char			shortname[3];
+	bool			comm_set;
+	bool			dead; /* if set thread has exited */
+	struct list_head	comm_list;
+	int			comm_len;
+	u64			db_id;
+
+	void			*priv;
+	struct thread_stack	*ts;
+};
+
+struct machine;
+struct comm;
+
+struct thread *thread__new(pid_t pid, pid_t tid);
+int thread__init_map_groups(struct thread *thread, struct machine *machine);
+void thread__delete(struct thread *thread);
+
+struct thread *thread__get(struct thread *thread);
+void thread__put(struct thread *thread);
+
+static inline void __thread__zput(struct thread **thread)
+{
+	thread__put(*thread);
+	*thread = NULL;
+}
+
+#define thread__zput(thread) __thread__zput(&thread)
+
+static inline void thread__exited(struct thread *thread)
+{
+	thread->dead = true;
+}
+
+int __thread__set_comm(struct thread *thread, const char *comm, u64 timestamp,
+		       bool exec);
+static inline int thread__set_comm(struct thread *thread, const char *comm,
+				   u64 timestamp)
+{
+	return __thread__set_comm(thread, comm, timestamp, false);
+}
+
+int thread__comm_len(struct thread *thread);
+struct comm *thread__comm(const struct thread *thread);
+struct comm *thread__exec_comm(const struct thread *thread);
+const char *thread__comm_str(const struct thread *thread);
+void thread__insert_map(struct thread *thread, struct map *map);
+int thread__fork(struct thread *thread, struct thread *parent, u64 timestamp);
+size_t thread__fprintf(struct thread *thread, FILE *fp);
+
+void thread__find_addr_map(struct thread *thread,
+			   u8 cpumode, enum map_type type, u64 addr,
+			   struct addr_location *al);
+
+void thread__find_addr_location(struct thread *thread,
+				u8 cpumode, enum map_type type, u64 addr,
+				struct addr_location *al);
+
+void thread__find_cpumode_addr_location(struct thread *thread,
+					enum map_type type, u64 addr,
+					struct addr_location *al);
+
+static inline void *thread__priv(struct thread *thread)
+{
+	return thread->priv;
+}
+
+static inline void thread__set_priv(struct thread *thread, void *p)
+{
+	thread->priv = p;
+}
+
+static inline bool thread__is_filtered(struct thread *thread)
+{
+	if (symbol_conf.comm_list &&
+	    !strlist__has_entry(symbol_conf.comm_list, thread__comm_str(thread))) {
+		return true;
+	}
+
+	if (symbol_conf.pid_list &&
+	    !intlist__has_entry(symbol_conf.pid_list, thread->pid_)) {
+		return true;
+	}
+
+	if (symbol_conf.tid_list &&
+	    !intlist__has_entry(symbol_conf.tid_list, thread->tid)) {
+		return true;
+	}
+
+	return false;
+}
+
+#endif	/* __PERF_THREAD_H */
diff --git a/tools/perf/util/thread_map.c b/tools/perf/util/thread_map.c
new file mode 100644
index 000000000..f93b97347
--- /dev/null
+++ b/tools/perf/util/thread_map.c
@@ -0,0 +1,299 @@
+#include <dirent.h>
+#include <limits.h>
+#include <stdbool.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+#include "strlist.h"
+#include <string.h>
+#include "thread_map.h"
+#include "util.h"
+
+/* Skip "." and ".." directories */
+static int filter(const struct dirent *dir)
+{
+	if (dir->d_name[0] == '.')
+		return 0;
+	else
+		return 1;
+}
+
+struct thread_map *thread_map__new_by_pid(pid_t pid)
+{
+	struct thread_map *threads;
+	char name[256];
+	int items;
+	struct dirent **namelist = NULL;
+	int i;
+
+	sprintf(name, "/proc/%d/task", pid);
+	items = scandir(name, &namelist, filter, NULL);
+	if (items <= 0)
+		return NULL;
+
+	threads = malloc(sizeof(*threads) + sizeof(pid_t) * items);
+	if (threads != NULL) {
+		for (i = 0; i < items; i++)
+			threads->map[i] = atoi(namelist[i]->d_name);
+		threads->nr = items;
+	}
+
+	for (i=0; i<items; i++)
+		zfree(&namelist[i]);
+	free(namelist);
+
+	return threads;
+}
+
+struct thread_map *thread_map__new_by_tid(pid_t tid)
+{
+	struct thread_map *threads = malloc(sizeof(*threads) + sizeof(pid_t));
+
+	if (threads != NULL) {
+		threads->map[0] = tid;
+		threads->nr	= 1;
+	}
+
+	return threads;
+}
+
+struct thread_map *thread_map__new_by_uid(uid_t uid)
+{
+	DIR *proc;
+	int max_threads = 32, items, i;
+	char path[256];
+	struct dirent dirent, *next, **namelist = NULL;
+	struct thread_map *threads = malloc(sizeof(*threads) +
+					    max_threads * sizeof(pid_t));
+	if (threads == NULL)
+		goto out;
+
+	proc = opendir("/proc");
+	if (proc == NULL)
+		goto out_free_threads;
+
+	threads->nr = 0;
+
+	while (!readdir_r(proc, &dirent, &next) && next) {
+		char *end;
+		bool grow = false;
+		struct stat st;
+		pid_t pid = strtol(dirent.d_name, &end, 10);
+
+		if (*end) /* only interested in proper numerical dirents */
+			continue;
+
+		snprintf(path, sizeof(path), "/proc/%s", dirent.d_name);
+
+		if (stat(path, &st) != 0)
+			continue;
+
+		if (st.st_uid != uid)
+			continue;
+
+		snprintf(path, sizeof(path), "/proc/%d/task", pid);
+		items = scandir(path, &namelist, filter, NULL);
+		if (items <= 0)
+			goto out_free_closedir;
+
+		while (threads->nr + items >= max_threads) {
+			max_threads *= 2;
+			grow = true;
+		}
+
+		if (grow) {
+			struct thread_map *tmp;
+
+			tmp = realloc(threads, (sizeof(*threads) +
+						max_threads * sizeof(pid_t)));
+			if (tmp == NULL)
+				goto out_free_namelist;
+
+			threads = tmp;
+		}
+
+		for (i = 0; i < items; i++)
+			threads->map[threads->nr + i] = atoi(namelist[i]->d_name);
+
+		for (i = 0; i < items; i++)
+			zfree(&namelist[i]);
+		free(namelist);
+
+		threads->nr += items;
+	}
+
+out_closedir:
+	closedir(proc);
+out:
+	return threads;
+
+out_free_threads:
+	free(threads);
+	return NULL;
+
+out_free_namelist:
+	for (i = 0; i < items; i++)
+		zfree(&namelist[i]);
+	free(namelist);
+
+out_free_closedir:
+	zfree(&threads);
+	goto out_closedir;
+}
+
+struct thread_map *thread_map__new(pid_t pid, pid_t tid, uid_t uid)
+{
+	if (pid != -1)
+		return thread_map__new_by_pid(pid);
+
+	if (tid == -1 && uid != UINT_MAX)
+		return thread_map__new_by_uid(uid);
+
+	return thread_map__new_by_tid(tid);
+}
+
+static struct thread_map *thread_map__new_by_pid_str(const char *pid_str)
+{
+	struct thread_map *threads = NULL, *nt;
+	char name[256];
+	int items, total_tasks = 0;
+	struct dirent **namelist = NULL;
+	int i, j = 0;
+	pid_t pid, prev_pid = INT_MAX;
+	char *end_ptr;
+	struct str_node *pos;
+	struct strlist *slist = strlist__new(false, pid_str);
+
+	if (!slist)
+		return NULL;
+
+	strlist__for_each(pos, slist) {
+		pid = strtol(pos->s, &end_ptr, 10);
+
+		if (pid == INT_MIN || pid == INT_MAX ||
+		    (*end_ptr != '\0' && *end_ptr != ','))
+			goto out_free_threads;
+
+		if (pid == prev_pid)
+			continue;
+
+		sprintf(name, "/proc/%d/task", pid);
+		items = scandir(name, &namelist, filter, NULL);
+		if (items <= 0)
+			goto out_free_threads;
+
+		total_tasks += items;
+		nt = realloc(threads, (sizeof(*threads) +
+				       sizeof(pid_t) * total_tasks));
+		if (nt == NULL)
+			goto out_free_namelist;
+
+		threads = nt;
+
+		for (i = 0; i < items; i++) {
+			threads->map[j++] = atoi(namelist[i]->d_name);
+			zfree(&namelist[i]);
+		}
+		threads->nr = total_tasks;
+		free(namelist);
+	}
+
+out:
+	strlist__delete(slist);
+	return threads;
+
+out_free_namelist:
+	for (i = 0; i < items; i++)
+		zfree(&namelist[i]);
+	free(namelist);
+
+out_free_threads:
+	zfree(&threads);
+	goto out;
+}
+
+struct thread_map *thread_map__new_dummy(void)
+{
+	struct thread_map *threads = malloc(sizeof(*threads) + sizeof(pid_t));
+
+	if (threads != NULL) {
+		threads->map[0]	= -1;
+		threads->nr	= 1;
+	}
+	return threads;
+}
+
+static struct thread_map *thread_map__new_by_tid_str(const char *tid_str)
+{
+	struct thread_map *threads = NULL, *nt;
+	int ntasks = 0;
+	pid_t tid, prev_tid = INT_MAX;
+	char *end_ptr;
+	struct str_node *pos;
+	struct strlist *slist;
+
+	/* perf-stat expects threads to be generated even if tid not given */
+	if (!tid_str)
+		return thread_map__new_dummy();
+
+	slist = strlist__new(false, tid_str);
+	if (!slist)
+		return NULL;
+
+	strlist__for_each(pos, slist) {
+		tid = strtol(pos->s, &end_ptr, 10);
+
+		if (tid == INT_MIN || tid == INT_MAX ||
+		    (*end_ptr != '\0' && *end_ptr != ','))
+			goto out_free_threads;
+
+		if (tid == prev_tid)
+			continue;
+
+		ntasks++;
+		nt = realloc(threads, sizeof(*threads) + sizeof(pid_t) * ntasks);
+
+		if (nt == NULL)
+			goto out_free_threads;
+
+		threads = nt;
+		threads->map[ntasks - 1] = tid;
+		threads->nr		 = ntasks;
+	}
+out:
+	return threads;
+
+out_free_threads:
+	zfree(&threads);
+	goto out;
+}
+
+struct thread_map *thread_map__new_str(const char *pid, const char *tid,
+				       uid_t uid)
+{
+	if (pid)
+		return thread_map__new_by_pid_str(pid);
+
+	if (!tid && uid != UINT_MAX)
+		return thread_map__new_by_uid(uid);
+
+	return thread_map__new_by_tid_str(tid);
+}
+
+void thread_map__delete(struct thread_map *threads)
+{
+	free(threads);
+}
+
+size_t thread_map__fprintf(struct thread_map *threads, FILE *fp)
+{
+	int i;
+	size_t printed = fprintf(fp, "%d thread%s: ",
+				 threads->nr, threads->nr > 1 ? "s" : "");
+	for (i = 0; i < threads->nr; ++i)
+		printed += fprintf(fp, "%s%d", i ? ", " : "", threads->map[i]);
+
+	return printed + fprintf(fp, "\n");
+}
diff --git a/tools/perf/util/thread_map.h b/tools/perf/util/thread_map.h
new file mode 100644
index 000000000..95313f43c
--- /dev/null
+++ b/tools/perf/util/thread_map.h
@@ -0,0 +1,30 @@
+#ifndef __PERF_THREAD_MAP_H
+#define __PERF_THREAD_MAP_H
+
+#include <sys/types.h>
+#include <stdio.h>
+
+struct thread_map {
+	int nr;
+	pid_t map[];
+};
+
+struct thread_map *thread_map__new_dummy(void);
+struct thread_map *thread_map__new_by_pid(pid_t pid);
+struct thread_map *thread_map__new_by_tid(pid_t tid);
+struct thread_map *thread_map__new_by_uid(uid_t uid);
+struct thread_map *thread_map__new(pid_t pid, pid_t tid, uid_t uid);
+
+struct thread_map *thread_map__new_str(const char *pid,
+		const char *tid, uid_t uid);
+
+void thread_map__delete(struct thread_map *threads);
+
+size_t thread_map__fprintf(struct thread_map *threads, FILE *fp);
+
+static inline int thread_map__nr(struct thread_map *threads)
+{
+	return threads ? threads->nr : 1;
+}
+
+#endif	/* __PERF_THREAD_MAP_H */
diff --git a/tools/perf/util/tool.h b/tools/perf/util/tool.h
new file mode 100644
index 000000000..51d9e56c0
--- /dev/null
+++ b/tools/perf/util/tool.h
@@ -0,0 +1,52 @@
+#ifndef __PERF_TOOL_H
+#define __PERF_TOOL_H
+
+#include <stdbool.h>
+
+struct perf_session;
+union perf_event;
+struct perf_evlist;
+struct perf_evsel;
+struct perf_sample;
+struct perf_tool;
+struct machine;
+struct ordered_events;
+
+typedef int (*event_sample)(struct perf_tool *tool, union perf_event *event,
+			    struct perf_sample *sample,
+			    struct perf_evsel *evsel, struct machine *machine);
+
+typedef int (*event_op)(struct perf_tool *tool, union perf_event *event,
+			struct perf_sample *sample, struct machine *machine);
+
+typedef int (*event_attr_op)(struct perf_tool *tool,
+			     union perf_event *event,
+			     struct perf_evlist **pevlist);
+
+typedef int (*event_op2)(struct perf_tool *tool, union perf_event *event,
+			 struct perf_session *session);
+
+typedef int (*event_oe)(struct perf_tool *tool, union perf_event *event,
+			struct ordered_events *oe);
+
+struct perf_tool {
+	event_sample	sample,
+			read;
+	event_op	mmap,
+			mmap2,
+			comm,
+			fork,
+			exit,
+			lost,
+			throttle,
+			unthrottle;
+	event_attr_op	attr;
+	event_op2	tracing_data;
+	event_oe	finished_round;
+	event_op2	build_id,
+			id_index;
+	bool		ordered_events;
+	bool		ordering_requires_timestamps;
+};
+
+#endif /* __PERF_TOOL_H */
diff --git a/tools/perf/util/top.c b/tools/perf/util/top.c
new file mode 100644
index 000000000..8e517def9
--- /dev/null
+++ b/tools/perf/util/top.c
@@ -0,0 +1,117 @@
+/*
+ * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Refactored from builtin-top.c, see that files for further copyright notes.
+ *
+ * Released under the GPL v2. (and only v2, not any later version)
+ */
+
+#include "cpumap.h"
+#include "event.h"
+#include "evlist.h"
+#include "evsel.h"
+#include "parse-events.h"
+#include "symbol.h"
+#include "top.h"
+#include <inttypes.h>
+
+#define SNPRINTF(buf, size, fmt, args...) \
+({ \
+	size_t r = snprintf(buf, size, fmt, ## args); \
+	r > size ?  size : r; \
+})
+
+size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size)
+{
+	float samples_per_sec;
+	float ksamples_per_sec;
+	float esamples_percent;
+	struct record_opts *opts = &top->record_opts;
+	struct target *target = &opts->target;
+	size_t ret = 0;
+
+	if (top->samples) {
+		samples_per_sec = top->samples / top->delay_secs;
+		ksamples_per_sec = top->kernel_samples / top->delay_secs;
+		esamples_percent = (100.0 * top->exact_samples) / top->samples;
+	} else {
+		samples_per_sec = ksamples_per_sec = esamples_percent = 0.0;
+	}
+
+	if (!perf_guest) {
+		float ksamples_percent = 0.0;
+
+		if (samples_per_sec)
+			ksamples_percent = (100.0 * ksamples_per_sec) /
+							samples_per_sec;
+		ret = SNPRINTF(bf, size,
+			       "   PerfTop:%8.0f irqs/sec  kernel:%4.1f%%"
+			       "  exact: %4.1f%% [", samples_per_sec,
+			       ksamples_percent, esamples_percent);
+	} else {
+		float us_samples_per_sec = top->us_samples / top->delay_secs;
+		float guest_kernel_samples_per_sec = top->guest_kernel_samples / top->delay_secs;
+		float guest_us_samples_per_sec = top->guest_us_samples / top->delay_secs;
+
+		ret = SNPRINTF(bf, size,
+			       "   PerfTop:%8.0f irqs/sec  kernel:%4.1f%% us:%4.1f%%"
+			       " guest kernel:%4.1f%% guest us:%4.1f%%"
+			       " exact: %4.1f%% [", samples_per_sec,
+			       100.0 - (100.0 * ((samples_per_sec - ksamples_per_sec) /
+						 samples_per_sec)),
+			       100.0 - (100.0 * ((samples_per_sec - us_samples_per_sec) /
+						 samples_per_sec)),
+			       100.0 - (100.0 * ((samples_per_sec -
+						  guest_kernel_samples_per_sec) /
+						 samples_per_sec)),
+			       100.0 - (100.0 * ((samples_per_sec -
+						  guest_us_samples_per_sec) /
+						 samples_per_sec)),
+			       esamples_percent);
+	}
+
+	if (top->evlist->nr_entries == 1) {
+		struct perf_evsel *first = perf_evlist__first(top->evlist);
+		ret += SNPRINTF(bf + ret, size - ret, "%" PRIu64 "%s ",
+				(uint64_t)first->attr.sample_period,
+				opts->freq ? "Hz" : "");
+	}
+
+	ret += SNPRINTF(bf + ret, size - ret, "%s", perf_evsel__name(top->sym_evsel));
+
+	ret += SNPRINTF(bf + ret, size - ret, "], ");
+
+	if (target->pid)
+		ret += SNPRINTF(bf + ret, size - ret, " (target_pid: %s",
+				target->pid);
+	else if (target->tid)
+		ret += SNPRINTF(bf + ret, size - ret, " (target_tid: %s",
+				target->tid);
+	else if (target->uid_str != NULL)
+		ret += SNPRINTF(bf + ret, size - ret, " (uid: %s",
+				target->uid_str);
+	else
+		ret += SNPRINTF(bf + ret, size - ret, " (all");
+
+	if (target->cpu_list)
+		ret += SNPRINTF(bf + ret, size - ret, ", CPU%s: %s)",
+				top->evlist->cpus->nr > 1 ? "s" : "",
+				target->cpu_list);
+	else {
+		if (target->tid)
+			ret += SNPRINTF(bf + ret, size - ret, ")");
+		else
+			ret += SNPRINTF(bf + ret, size - ret, ", %d CPU%s)",
+					top->evlist->cpus->nr,
+					top->evlist->cpus->nr > 1 ? "s" : "");
+	}
+
+	return ret;
+}
+
+void perf_top__reset_sample_counters(struct perf_top *top)
+{
+	top->samples = top->us_samples = top->kernel_samples =
+	top->exact_samples = top->guest_kernel_samples =
+	top->guest_us_samples = 0;
+}
diff --git a/tools/perf/util/top.h b/tools/perf/util/top.h
new file mode 100644
index 000000000..f92c37abb
--- /dev/null
+++ b/tools/perf/util/top.h
@@ -0,0 +1,47 @@
+#ifndef __PERF_TOP_H
+#define __PERF_TOP_H 1
+
+#include "tool.h"
+#include <linux/types.h>
+#include <stddef.h>
+#include <stdbool.h>
+#include <termios.h>
+
+struct perf_evlist;
+struct perf_evsel;
+struct perf_session;
+
+struct perf_top {
+	struct perf_tool   tool;
+	struct perf_evlist *evlist;
+	struct record_opts record_opts;
+	/*
+	 * Symbols will be added here in perf_event__process_sample and will
+	 * get out after decayed.
+	 */
+	u64		   samples;
+	u64		   kernel_samples, us_samples;
+	u64		   exact_samples;
+	u64		   guest_us_samples, guest_kernel_samples;
+	int		   print_entries, count_filter, delay_secs;
+	int		   max_stack;
+	bool		   hide_kernel_symbols, hide_user_symbols, zero;
+	bool		   use_tui, use_stdio;
+	bool		   kptr_restrict_warned;
+	bool		   vmlinux_warned;
+	bool		   dump_symtab;
+	struct hist_entry  *sym_filter_entry;
+	struct perf_evsel  *sym_evsel;
+	struct perf_session *session;
+	struct winsize	   winsize;
+	int		   realtime_prio;
+	int		   sym_pcnt_filter;
+	const char	   *sym_filter;
+	float		   min_percent;
+};
+
+#define CONSOLE_CLEAR ""
+
+size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size);
+void perf_top__reset_sample_counters(struct perf_top *top);
+#endif /* __PERF_TOP_H */
diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c
new file mode 100644
index 000000000..eb7271601
--- /dev/null
+++ b/tools/perf/util/trace-event-info.c
@@ -0,0 +1,595 @@
+/*
+ * Copyright (C) 2008,2009, Steven Rostedt <srostedt@redhat.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License (not later!)
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+#include "util.h"
+#include <dirent.h>
+#include <mntent.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/wait.h>
+#include <pthread.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <errno.h>
+#include <stdbool.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+
+#include "../perf.h"
+#include "trace-event.h"
+#include <api/fs/debugfs.h>
+#include "evsel.h"
+#include "debug.h"
+
+#define VERSION "0.5"
+
+static int output_fd;
+
+
+int bigendian(void)
+{
+	unsigned char str[] = { 0x1, 0x2, 0x3, 0x4, 0x0, 0x0, 0x0, 0x0};
+	unsigned int *ptr;
+
+	ptr = (unsigned int *)(void *)str;
+	return *ptr == 0x01020304;
+}
+
+/* unfortunately, you can not stat debugfs or proc files for size */
+static int record_file(const char *file, ssize_t hdr_sz)
+{
+	unsigned long long size = 0;
+	char buf[BUFSIZ], *sizep;
+	off_t hdr_pos = lseek(output_fd, 0, SEEK_CUR);
+	int r, fd;
+	int err = -EIO;
+
+	fd = open(file, O_RDONLY);
+	if (fd < 0) {
+		pr_debug("Can't read '%s'", file);
+		return -errno;
+	}
+
+	/* put in zeros for file size, then fill true size later */
+	if (hdr_sz) {
+		if (write(output_fd, &size, hdr_sz) != hdr_sz)
+			goto out;
+	}
+
+	do {
+		r = read(fd, buf, BUFSIZ);
+		if (r > 0) {
+			size += r;
+			if (write(output_fd, buf, r) != r)
+				goto out;
+		}
+	} while (r > 0);
+
+	/* ugh, handle big-endian hdr_size == 4 */
+	sizep = (char*)&size;
+	if (bigendian())
+		sizep += sizeof(u64) - hdr_sz;
+
+	if (hdr_sz && pwrite(output_fd, sizep, hdr_sz, hdr_pos) < 0) {
+		pr_debug("writing file size failed\n");
+		goto out;
+	}
+
+	err = 0;
+out:
+	close(fd);
+	return err;
+}
+
+static int record_header_files(void)
+{
+	char *path;
+	struct stat st;
+	int err = -EIO;
+
+	path = get_tracing_file("events/header_page");
+	if (!path) {
+		pr_debug("can't get tracing/events/header_page");
+		return -ENOMEM;
+	}
+
+	if (stat(path, &st) < 0) {
+		pr_debug("can't read '%s'", path);
+		goto out;
+	}
+
+	if (write(output_fd, "header_page", 12) != 12) {
+		pr_debug("can't write header_page\n");
+		goto out;
+	}
+
+	if (record_file(path, 8) < 0) {
+		pr_debug("can't record header_page file\n");
+		goto out;
+	}
+
+	put_tracing_file(path);
+
+	path = get_tracing_file("events/header_event");
+	if (!path) {
+		pr_debug("can't get tracing/events/header_event");
+		err = -ENOMEM;
+		goto out;
+	}
+
+	if (stat(path, &st) < 0) {
+		pr_debug("can't read '%s'", path);
+		goto out;
+	}
+
+	if (write(output_fd, "header_event", 13) != 13) {
+		pr_debug("can't write header_event\n");
+		goto out;
+	}
+
+	if (record_file(path, 8) < 0) {
+		pr_debug("can't record header_event file\n");
+		goto out;
+	}
+
+	err = 0;
+out:
+	put_tracing_file(path);
+	return err;
+}
+
+static bool name_in_tp_list(char *sys, struct tracepoint_path *tps)
+{
+	while (tps) {
+		if (!strcmp(sys, tps->name))
+			return true;
+		tps = tps->next;
+	}
+
+	return false;
+}
+
+static int copy_event_system(const char *sys, struct tracepoint_path *tps)
+{
+	struct dirent *dent;
+	struct stat st;
+	char *format;
+	DIR *dir;
+	int count = 0;
+	int ret;
+	int err;
+
+	dir = opendir(sys);
+	if (!dir) {
+		pr_debug("can't read directory '%s'", sys);
+		return -errno;
+	}
+
+	while ((dent = readdir(dir))) {
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
+		    strcmp(dent->d_name, "..") == 0 ||
+		    !name_in_tp_list(dent->d_name, tps))
+			continue;
+		if (asprintf(&format, "%s/%s/format", sys, dent->d_name) < 0) {
+			err = -ENOMEM;
+			goto out;
+		}
+		ret = stat(format, &st);
+		free(format);
+		if (ret < 0)
+			continue;
+		count++;
+	}
+
+	if (write(output_fd, &count, 4) != 4) {
+		err = -EIO;
+		pr_debug("can't write count\n");
+		goto out;
+	}
+
+	rewinddir(dir);
+	while ((dent = readdir(dir))) {
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
+		    strcmp(dent->d_name, "..") == 0 ||
+		    !name_in_tp_list(dent->d_name, tps))
+			continue;
+		if (asprintf(&format, "%s/%s/format", sys, dent->d_name) < 0) {
+			err = -ENOMEM;
+			goto out;
+		}
+		ret = stat(format, &st);
+
+		if (ret >= 0) {
+			err = record_file(format, 8);
+			if (err) {
+				free(format);
+				goto out;
+			}
+		}
+		free(format);
+	}
+	err = 0;
+out:
+	closedir(dir);
+	return err;
+}
+
+static int record_ftrace_files(struct tracepoint_path *tps)
+{
+	char *path;
+	int ret;
+
+	path = get_tracing_file("events/ftrace");
+	if (!path) {
+		pr_debug("can't get tracing/events/ftrace");
+		return -ENOMEM;
+	}
+
+	ret = copy_event_system(path, tps);
+
+	put_tracing_file(path);
+
+	return ret;
+}
+
+static bool system_in_tp_list(char *sys, struct tracepoint_path *tps)
+{
+	while (tps) {
+		if (!strcmp(sys, tps->system))
+			return true;
+		tps = tps->next;
+	}
+
+	return false;
+}
+
+static int record_event_files(struct tracepoint_path *tps)
+{
+	struct dirent *dent;
+	struct stat st;
+	char *path;
+	char *sys;
+	DIR *dir;
+	int count = 0;
+	int ret;
+	int err;
+
+	path = get_tracing_file("events");
+	if (!path) {
+		pr_debug("can't get tracing/events");
+		return -ENOMEM;
+	}
+
+	dir = opendir(path);
+	if (!dir) {
+		err = -errno;
+		pr_debug("can't read directory '%s'", path);
+		goto out;
+	}
+
+	while ((dent = readdir(dir))) {
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
+		    strcmp(dent->d_name, "..") == 0 ||
+		    strcmp(dent->d_name, "ftrace") == 0 ||
+		    !system_in_tp_list(dent->d_name, tps))
+			continue;
+		count++;
+	}
+
+	if (write(output_fd, &count, 4) != 4) {
+		err = -EIO;
+		pr_debug("can't write count\n");
+		goto out;
+	}
+
+	rewinddir(dir);
+	while ((dent = readdir(dir))) {
+		if (dent->d_type != DT_DIR ||
+		    strcmp(dent->d_name, ".") == 0 ||
+		    strcmp(dent->d_name, "..") == 0 ||
+		    strcmp(dent->d_name, "ftrace") == 0 ||
+		    !system_in_tp_list(dent->d_name, tps))
+			continue;
+		if (asprintf(&sys, "%s/%s", path, dent->d_name) < 0) {
+			err = -ENOMEM;
+			goto out;
+		}
+		ret = stat(sys, &st);
+		if (ret >= 0) {
+			ssize_t size = strlen(dent->d_name) + 1;
+
+			if (write(output_fd, dent->d_name, size) != size ||
+			    copy_event_system(sys, tps) < 0) {
+				err = -EIO;
+				free(sys);
+				goto out;
+			}
+		}
+		free(sys);
+	}
+	err = 0;
+out:
+	closedir(dir);
+	put_tracing_file(path);
+
+	return err;
+}
+
+static int record_proc_kallsyms(void)
+{
+	unsigned int size;
+	const char *path = "/proc/kallsyms";
+	struct stat st;
+	int ret, err = 0;
+
+	ret = stat(path, &st);
+	if (ret < 0) {
+		/* not found */
+		size = 0;
+		if (write(output_fd, &size, 4) != 4)
+			err = -EIO;
+		return err;
+	}
+	return record_file(path, 4);
+}
+
+static int record_ftrace_printk(void)
+{
+	unsigned int size;
+	char *path;
+	struct stat st;
+	int ret, err = 0;
+
+	path = get_tracing_file("printk_formats");
+	if (!path) {
+		pr_debug("can't get tracing/printk_formats");
+		return -ENOMEM;
+	}
+
+	ret = stat(path, &st);
+	if (ret < 0) {
+		/* not found */
+		size = 0;
+		if (write(output_fd, &size, 4) != 4)
+			err = -EIO;
+		goto out;
+	}
+	err = record_file(path, 4);
+
+out:
+	put_tracing_file(path);
+	return err;
+}
+
+static void
+put_tracepoints_path(struct tracepoint_path *tps)
+{
+	while (tps) {
+		struct tracepoint_path *t = tps;
+
+		tps = tps->next;
+		zfree(&t->name);
+		zfree(&t->system);
+		free(t);
+	}
+}
+
+static struct tracepoint_path *
+get_tracepoints_path(struct list_head *pattrs)
+{
+	struct tracepoint_path path, *ppath = &path;
+	struct perf_evsel *pos;
+	int nr_tracepoints = 0;
+
+	list_for_each_entry(pos, pattrs, node) {
+		if (pos->attr.type != PERF_TYPE_TRACEPOINT)
+			continue;
+		++nr_tracepoints;
+
+		if (pos->name) {
+			ppath->next = tracepoint_name_to_path(pos->name);
+			if (ppath->next)
+				goto next;
+
+			if (strchr(pos->name, ':') == NULL)
+				goto try_id;
+
+			goto error;
+		}
+
+try_id:
+		ppath->next = tracepoint_id_to_path(pos->attr.config);
+		if (!ppath->next) {
+error:
+			pr_debug("No memory to alloc tracepoints list\n");
+			put_tracepoints_path(&path);
+			return NULL;
+		}
+next:
+		ppath = ppath->next;
+	}
+
+	return nr_tracepoints > 0 ? path.next : NULL;
+}
+
+bool have_tracepoints(struct list_head *pattrs)
+{
+	struct perf_evsel *pos;
+
+	list_for_each_entry(pos, pattrs, node)
+		if (pos->attr.type == PERF_TYPE_TRACEPOINT)
+			return true;
+
+	return false;
+}
+
+static int tracing_data_header(void)
+{
+	char buf[20];
+	ssize_t size;
+
+	/* just guessing this is someone's birthday.. ;) */
+	buf[0] = 23;
+	buf[1] = 8;
+	buf[2] = 68;
+	memcpy(buf + 3, "tracing", 7);
+
+	if (write(output_fd, buf, 10) != 10)
+		return -1;
+
+	size = strlen(VERSION) + 1;
+	if (write(output_fd, VERSION, size) != size)
+		return -1;
+
+	/* save endian */
+	if (bigendian())
+		buf[0] = 1;
+	else
+		buf[0] = 0;
+
+	if (write(output_fd, buf, 1) != 1)
+		return -1;
+
+	/* save size of long */
+	buf[0] = sizeof(long);
+	if (write(output_fd, buf, 1) != 1)
+		return -1;
+
+	/* save page_size */
+	if (write(output_fd, &page_size, 4) != 4)
+		return -1;
+
+	return 0;
+}
+
+struct tracing_data *tracing_data_get(struct list_head *pattrs,
+				      int fd, bool temp)
+{
+	struct tracepoint_path *tps;
+	struct tracing_data *tdata;
+	int err;
+
+	output_fd = fd;
+
+	tps = get_tracepoints_path(pattrs);
+	if (!tps)
+		return NULL;
+
+	tdata = malloc(sizeof(*tdata));
+	if (!tdata)
+		return NULL;
+
+	tdata->temp = temp;
+	tdata->size = 0;
+
+	if (temp) {
+		int temp_fd;
+
+		snprintf(tdata->temp_file, sizeof(tdata->temp_file),
+			 "/tmp/perf-XXXXXX");
+		if (!mkstemp(tdata->temp_file)) {
+			pr_debug("Can't make temp file");
+			return NULL;
+		}
+
+		temp_fd = open(tdata->temp_file, O_RDWR);
+		if (temp_fd < 0) {
+			pr_debug("Can't read '%s'", tdata->temp_file);
+			return NULL;
+		}
+
+		/*
+		 * Set the temp file the default output, so all the
+		 * tracing data are stored into it.
+		 */
+		output_fd = temp_fd;
+	}
+
+	err = tracing_data_header();
+	if (err)
+		goto out;
+	err = record_header_files();
+	if (err)
+		goto out;
+	err = record_ftrace_files(tps);
+	if (err)
+		goto out;
+	err = record_event_files(tps);
+	if (err)
+		goto out;
+	err = record_proc_kallsyms();
+	if (err)
+		goto out;
+	err = record_ftrace_printk();
+
+out:
+	/*
+	 * All tracing data are stored by now, we can restore
+	 * the default output file in case we used temp file.
+	 */
+	if (temp) {
+		tdata->size = lseek(output_fd, 0, SEEK_CUR);
+		close(output_fd);
+		output_fd = fd;
+	}
+
+	if (err)
+		zfree(&tdata);
+
+	put_tracepoints_path(tps);
+	return tdata;
+}
+
+int tracing_data_put(struct tracing_data *tdata)
+{
+	int err = 0;
+
+	if (tdata->temp) {
+		err = record_file(tdata->temp_file, 0);
+		unlink(tdata->temp_file);
+	}
+
+	free(tdata);
+	return err;
+}
+
+int read_tracing_data(int fd, struct list_head *pattrs)
+{
+	int err;
+	struct tracing_data *tdata;
+
+	/*
+	 * We work over the real file, so we can write data
+	 * directly, no temp file is needed.
+	 */
+	tdata = tracing_data_get(pattrs, fd, false);
+	if (!tdata)
+		return -ENOMEM;
+
+	err = tracing_data_put(tdata);
+	return err;
+}
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
new file mode 100644
index 000000000..25d6c737b
--- /dev/null
+++ b/tools/perf/util/trace-event-parse.c
@@ -0,0 +1,269 @@
+/*
+ * Copyright (C) 2009, Steven Rostedt <srostedt@redhat.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License (not later!)
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <errno.h>
+
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+
+static int get_common_field(struct scripting_context *context,
+			    int *offset, int *size, const char *type)
+{
+	struct pevent *pevent = context->pevent;
+	struct event_format *event;
+	struct format_field *field;
+
+	if (!*size) {
+		if (!pevent->events)
+			return 0;
+
+		event = pevent->events[0];
+		field = pevent_find_common_field(event, type);
+		if (!field)
+			return 0;
+		*offset = field->offset;
+		*size = field->size;
+	}
+
+	return pevent_read_number(pevent, context->event_data + *offset, *size);
+}
+
+int common_lock_depth(struct scripting_context *context)
+{
+	static int offset;
+	static int size;
+	int ret;
+
+	ret = get_common_field(context, &size, &offset,
+			       "common_lock_depth");
+	if (ret < 0)
+		return -1;
+
+	return ret;
+}
+
+int common_flags(struct scripting_context *context)
+{
+	static int offset;
+	static int size;
+	int ret;
+
+	ret = get_common_field(context, &size, &offset,
+			       "common_flags");
+	if (ret < 0)
+		return -1;
+
+	return ret;
+}
+
+int common_pc(struct scripting_context *context)
+{
+	static int offset;
+	static int size;
+	int ret;
+
+	ret = get_common_field(context, &size, &offset,
+			       "common_preempt_count");
+	if (ret < 0)
+		return -1;
+
+	return ret;
+}
+
+unsigned long long
+raw_field_value(struct event_format *event, const char *name, void *data)
+{
+	struct format_field *field;
+	unsigned long long val;
+
+	field = pevent_find_any_field(event, name);
+	if (!field)
+		return 0ULL;
+
+	pevent_read_number_field(field, data, &val);
+
+	return val;
+}
+
+unsigned long long read_size(struct event_format *event, void *ptr, int size)
+{
+	return pevent_read_number(event->pevent, ptr, size);
+}
+
+void event_format__fprintf(struct event_format *event,
+			   int cpu, void *data, int size, FILE *fp)
+{
+	struct pevent_record record;
+	struct trace_seq s;
+
+	memset(&record, 0, sizeof(record));
+	record.cpu = cpu;
+	record.size = size;
+	record.data = data;
+
+	trace_seq_init(&s);
+	pevent_event_info(&s, event, &record);
+	trace_seq_do_fprintf(&s, fp);
+	trace_seq_destroy(&s);
+}
+
+void event_format__print(struct event_format *event,
+			 int cpu, void *data, int size)
+{
+	return event_format__fprintf(event, cpu, data, size, stdout);
+}
+
+void parse_proc_kallsyms(struct pevent *pevent,
+			 char *file, unsigned int size __maybe_unused)
+{
+	unsigned long long addr;
+	char *func;
+	char *line;
+	char *next = NULL;
+	char *addr_str;
+	char *mod;
+	char *fmt = NULL;
+
+	line = strtok_r(file, "\n", &next);
+	while (line) {
+		mod = NULL;
+		addr_str = strtok_r(line, " ", &fmt);
+		addr = strtoull(addr_str, NULL, 16);
+		/* skip character */
+		strtok_r(NULL, " ", &fmt);
+		func = strtok_r(NULL, "\t", &fmt);
+		mod = strtok_r(NULL, "]", &fmt);
+		/* truncate the extra '[' */
+		if (mod)
+			mod = mod + 1;
+
+		pevent_register_function(pevent, func, addr, mod);
+
+		line = strtok_r(NULL, "\n", &next);
+	}
+}
+
+void parse_ftrace_printk(struct pevent *pevent,
+			 char *file, unsigned int size __maybe_unused)
+{
+	unsigned long long addr;
+	char *printk;
+	char *line;
+	char *next = NULL;
+	char *addr_str;
+	char *fmt;
+
+	line = strtok_r(file, "\n", &next);
+	while (line) {
+		addr_str = strtok_r(line, ":", &fmt);
+		if (!addr_str) {
+			warning("printk format with empty entry");
+			break;
+		}
+		addr = strtoull(addr_str, NULL, 16);
+		/* fmt still has a space, skip it */
+		printk = strdup(fmt+1);
+		line = strtok_r(NULL, "\n", &next);
+		pevent_register_print_string(pevent, printk, addr);
+	}
+}
+
+int parse_ftrace_file(struct pevent *pevent, char *buf, unsigned long size)
+{
+	return pevent_parse_event(pevent, buf, size, "ftrace");
+}
+
+int parse_event_file(struct pevent *pevent,
+		     char *buf, unsigned long size, char *sys)
+{
+	return pevent_parse_event(pevent, buf, size, sys);
+}
+
+struct event_format *trace_find_next_event(struct pevent *pevent,
+					   struct event_format *event)
+{
+	static int idx;
+
+	if (!pevent || !pevent->events)
+		return NULL;
+
+	if (!event) {
+		idx = 0;
+		return pevent->events[0];
+	}
+
+	if (idx < pevent->nr_events && event == pevent->events[idx]) {
+		idx++;
+		if (idx == pevent->nr_events)
+			return NULL;
+		return pevent->events[idx];
+	}
+
+	for (idx = 1; idx < pevent->nr_events; idx++) {
+		if (event == pevent->events[idx - 1])
+			return pevent->events[idx];
+	}
+	return NULL;
+}
+
+struct flag {
+	const char *name;
+	unsigned long long value;
+};
+
+static const struct flag flags[] = {
+	{ "HI_SOFTIRQ", 0 },
+	{ "TIMER_SOFTIRQ", 1 },
+	{ "NET_TX_SOFTIRQ", 2 },
+	{ "NET_RX_SOFTIRQ", 3 },
+	{ "BLOCK_SOFTIRQ", 4 },
+	{ "BLOCK_IOPOLL_SOFTIRQ", 5 },
+	{ "TASKLET_SOFTIRQ", 6 },
+	{ "SCHED_SOFTIRQ", 7 },
+	{ "HRTIMER_SOFTIRQ", 8 },
+	{ "RCU_SOFTIRQ", 9 },
+
+	{ "HRTIMER_NORESTART", 0 },
+	{ "HRTIMER_RESTART", 1 },
+};
+
+unsigned long long eval_flag(const char *flag)
+{
+	int i;
+
+	/*
+	 * Some flags in the format files do not get converted.
+	 * If the flag is not numeric, see if it is something that
+	 * we already know about.
+	 */
+	if (isdigit(flag[0]))
+		return strtoull(flag, NULL, 0);
+
+	for (i = 0; i < (int)(sizeof(flags)/sizeof(flags[0])); i++)
+		if (strcmp(flags[i].name, flag) == 0)
+			return flags[i].value;
+
+	return 0;
+}
diff --git a/tools/perf/util/trace-event-read.c b/tools/perf/util/trace-event-read.c
new file mode 100644
index 000000000..54d9e9b54
--- /dev/null
+++ b/tools/perf/util/trace-event-read.c
@@ -0,0 +1,444 @@
+/*
+ * Copyright (C) 2009, Steven Rostedt <srostedt@redhat.com>
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License (not later!)
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+#include <dirent.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdarg.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/wait.h>
+#include <sys/mman.h>
+#include <pthread.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <errno.h>
+
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+#include "debug.h"
+
+static int input_fd;
+
+static ssize_t trace_data_size;
+static bool repipe;
+
+static int __do_read(int fd, void *buf, int size)
+{
+	int rsize = size;
+
+	while (size) {
+		int ret = read(fd, buf, size);
+
+		if (ret <= 0)
+			return -1;
+
+		if (repipe) {
+			int retw = write(STDOUT_FILENO, buf, ret);
+
+			if (retw <= 0 || retw != ret) {
+				pr_debug("repiping input file");
+				return -1;
+			}
+		}
+
+		size -= ret;
+		buf += ret;
+	}
+
+	return rsize;
+}
+
+static int do_read(void *data, int size)
+{
+	int r;
+
+	r = __do_read(input_fd, data, size);
+	if (r <= 0) {
+		pr_debug("reading input file (size expected=%d received=%d)",
+			 size, r);
+		return -1;
+	}
+
+	trace_data_size += r;
+
+	return r;
+}
+
+/* If it fails, the next read will report it */
+static void skip(int size)
+{
+	char buf[BUFSIZ];
+	int r;
+
+	while (size) {
+		r = size > BUFSIZ ? BUFSIZ : size;
+		do_read(buf, r);
+		size -= r;
+	};
+}
+
+static unsigned int read4(struct pevent *pevent)
+{
+	unsigned int data;
+
+	if (do_read(&data, 4) < 0)
+		return 0;
+	return __data2host4(pevent, data);
+}
+
+static unsigned long long read8(struct pevent *pevent)
+{
+	unsigned long long data;
+
+	if (do_read(&data, 8) < 0)
+		return 0;
+	return __data2host8(pevent, data);
+}
+
+static char *read_string(void)
+{
+	char buf[BUFSIZ];
+	char *str = NULL;
+	int size = 0;
+	off_t r;
+	char c;
+
+	for (;;) {
+		r = read(input_fd, &c, 1);
+		if (r < 0) {
+			pr_debug("reading input file");
+			goto out;
+		}
+
+		if (!r) {
+			pr_debug("no data");
+			goto out;
+		}
+
+		if (repipe) {
+			int retw = write(STDOUT_FILENO, &c, 1);
+
+			if (retw <= 0 || retw != r) {
+				pr_debug("repiping input file string");
+				goto out;
+			}
+		}
+
+		buf[size++] = c;
+
+		if (!c)
+			break;
+	}
+
+	trace_data_size += size;
+
+	str = malloc(size);
+	if (str)
+		memcpy(str, buf, size);
+out:
+	return str;
+}
+
+static int read_proc_kallsyms(struct pevent *pevent)
+{
+	unsigned int size;
+	char *buf;
+
+	size = read4(pevent);
+	if (!size)
+		return 0;
+
+	buf = malloc(size + 1);
+	if (buf == NULL)
+		return -1;
+
+	if (do_read(buf, size) < 0) {
+		free(buf);
+		return -1;
+	}
+	buf[size] = '\0';
+
+	parse_proc_kallsyms(pevent, buf, size);
+
+	free(buf);
+	return 0;
+}
+
+static int read_ftrace_printk(struct pevent *pevent)
+{
+	unsigned int size;
+	char *buf;
+
+	/* it can have 0 size */
+	size = read4(pevent);
+	if (!size)
+		return 0;
+
+	buf = malloc(size);
+	if (buf == NULL)
+		return -1;
+
+	if (do_read(buf, size) < 0) {
+		free(buf);
+		return -1;
+	}
+
+	parse_ftrace_printk(pevent, buf, size);
+
+	free(buf);
+	return 0;
+}
+
+static int read_header_files(struct pevent *pevent)
+{
+	unsigned long long size;
+	char *header_page;
+	char buf[BUFSIZ];
+	int ret = 0;
+
+	if (do_read(buf, 12) < 0)
+		return -1;
+
+	if (memcmp(buf, "header_page", 12) != 0) {
+		pr_debug("did not read header page");
+		return -1;
+	}
+
+	size = read8(pevent);
+
+	header_page = malloc(size);
+	if (header_page == NULL)
+		return -1;
+
+	if (do_read(header_page, size) < 0) {
+		pr_debug("did not read header page");
+		free(header_page);
+		return -1;
+	}
+
+	if (!pevent_parse_header_page(pevent, header_page, size,
+				      pevent_get_long_size(pevent))) {
+		/*
+		 * The commit field in the page is of type long,
+		 * use that instead, since it represents the kernel.
+		 */
+		pevent_set_long_size(pevent, pevent->header_page_size_size);
+	}
+	free(header_page);
+
+	if (do_read(buf, 13) < 0)
+		return -1;
+
+	if (memcmp(buf, "header_event", 13) != 0) {
+		pr_debug("did not read header event");
+		return -1;
+	}
+
+	size = read8(pevent);
+	skip(size);
+
+	return ret;
+}
+
+static int read_ftrace_file(struct pevent *pevent, unsigned long long size)
+{
+	char *buf;
+
+	buf = malloc(size);
+	if (buf == NULL)
+		return -1;
+
+	if (do_read(buf, size) < 0) {
+		free(buf);
+		return -1;
+	}
+
+	parse_ftrace_file(pevent, buf, size);
+	free(buf);
+	return 0;
+}
+
+static int read_event_file(struct pevent *pevent, char *sys,
+			    unsigned long long size)
+{
+	char *buf;
+
+	buf = malloc(size);
+	if (buf == NULL)
+		return -1;
+
+	if (do_read(buf, size) < 0) {
+		free(buf);
+		return -1;
+	}
+
+	parse_event_file(pevent, buf, size, sys);
+	free(buf);
+	return 0;
+}
+
+static int read_ftrace_files(struct pevent *pevent)
+{
+	unsigned long long size;
+	int count;
+	int i;
+	int ret;
+
+	count = read4(pevent);
+
+	for (i = 0; i < count; i++) {
+		size = read8(pevent);
+		ret = read_ftrace_file(pevent, size);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+static int read_event_files(struct pevent *pevent)
+{
+	unsigned long long size;
+	char *sys;
+	int systems;
+	int count;
+	int i,x;
+	int ret;
+
+	systems = read4(pevent);
+
+	for (i = 0; i < systems; i++) {
+		sys = read_string();
+		if (sys == NULL)
+			return -1;
+
+		count = read4(pevent);
+
+		for (x=0; x < count; x++) {
+			size = read8(pevent);
+			ret = read_event_file(pevent, sys, size);
+			if (ret)
+				return ret;
+		}
+	}
+	return 0;
+}
+
+ssize_t trace_report(int fd, struct trace_event *tevent, bool __repipe)
+{
+	char buf[BUFSIZ];
+	char test[] = { 23, 8, 68 };
+	char *version;
+	int show_version = 0;
+	int show_funcs = 0;
+	int show_printk = 0;
+	ssize_t size = -1;
+	int file_bigendian;
+	int host_bigendian;
+	int file_long_size;
+	int file_page_size;
+	struct pevent *pevent = NULL;
+	int err;
+
+	repipe = __repipe;
+	input_fd = fd;
+
+	if (do_read(buf, 3) < 0)
+		return -1;
+	if (memcmp(buf, test, 3) != 0) {
+		pr_debug("no trace data in the file");
+		return -1;
+	}
+
+	if (do_read(buf, 7) < 0)
+		return -1;
+	if (memcmp(buf, "tracing", 7) != 0) {
+		pr_debug("not a trace file (missing 'tracing' tag)");
+		return -1;
+	}
+
+	version = read_string();
+	if (version == NULL)
+		return -1;
+	if (show_version)
+		printf("version = %s\n", version);
+	free(version);
+
+	if (do_read(buf, 1) < 0)
+		return -1;
+	file_bigendian = buf[0];
+	host_bigendian = bigendian();
+
+	if (trace_event__init(tevent)) {
+		pr_debug("trace_event__init failed");
+		goto out;
+	}
+
+	pevent = tevent->pevent;
+
+	pevent_set_flag(pevent, PEVENT_NSEC_OUTPUT);
+	pevent_set_file_bigendian(pevent, file_bigendian);
+	pevent_set_host_bigendian(pevent, host_bigendian);
+
+	if (do_read(buf, 1) < 0)
+		goto out;
+	file_long_size = buf[0];
+
+	file_page_size = read4(pevent);
+	if (!file_page_size)
+		goto out;
+
+	pevent_set_long_size(pevent, file_long_size);
+	pevent_set_page_size(pevent, file_page_size);
+
+	err = read_header_files(pevent);
+	if (err)
+		goto out;
+	err = read_ftrace_files(pevent);
+	if (err)
+		goto out;
+	err = read_event_files(pevent);
+	if (err)
+		goto out;
+	err = read_proc_kallsyms(pevent);
+	if (err)
+		goto out;
+	err = read_ftrace_printk(pevent);
+	if (err)
+		goto out;
+
+	size = trace_data_size;
+	repipe = false;
+
+	if (show_funcs) {
+		pevent_print_funcs(pevent);
+	} else if (show_printk) {
+		pevent_print_printk(pevent);
+	}
+
+	pevent = NULL;
+
+out:
+	if (pevent)
+		trace_event__cleanup(tevent);
+	return size;
+}
diff --git a/tools/perf/util/trace-event-scripting.c b/tools/perf/util/trace-event-scripting.c
new file mode 100644
index 000000000..9df61059a
--- /dev/null
+++ b/tools/perf/util/trace-event-scripting.c
@@ -0,0 +1,177 @@
+/*
+ * trace-event-scripting.  Scripting engine common and initialization code.
+ *
+ * Copyright (C) 2009-2010 Tom Zanussi <tzanussi@gmail.com>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+
+#include "../perf.h"
+#include "util.h"
+#include "trace-event.h"
+
+struct scripting_context *scripting_context;
+
+static int flush_script_unsupported(void)
+{
+	return 0;
+}
+
+static int stop_script_unsupported(void)
+{
+	return 0;
+}
+
+static void process_event_unsupported(union perf_event *event __maybe_unused,
+				      struct perf_sample *sample __maybe_unused,
+				      struct perf_evsel *evsel __maybe_unused,
+				      struct addr_location *al __maybe_unused)
+{
+}
+
+static void print_python_unsupported_msg(void)
+{
+	fprintf(stderr, "Python scripting not supported."
+		"  Install libpython and rebuild perf to enable it.\n"
+		"For example:\n  # apt-get install python-dev (ubuntu)"
+		"\n  # yum install python-devel (Fedora)"
+		"\n  etc.\n");
+}
+
+static int python_start_script_unsupported(const char *script __maybe_unused,
+					   int argc __maybe_unused,
+					   const char **argv __maybe_unused)
+{
+	print_python_unsupported_msg();
+
+	return -1;
+}
+
+static int python_generate_script_unsupported(struct pevent *pevent
+					      __maybe_unused,
+					      const char *outfile
+					      __maybe_unused)
+{
+	print_python_unsupported_msg();
+
+	return -1;
+}
+
+struct scripting_ops python_scripting_unsupported_ops = {
+	.name = "Python",
+	.start_script = python_start_script_unsupported,
+	.flush_script = flush_script_unsupported,
+	.stop_script = stop_script_unsupported,
+	.process_event = process_event_unsupported,
+	.generate_script = python_generate_script_unsupported,
+};
+
+static void register_python_scripting(struct scripting_ops *scripting_ops)
+{
+	int err;
+	err = script_spec_register("Python", scripting_ops);
+	if (err)
+		die("error registering Python script extension");
+
+	err = script_spec_register("py", scripting_ops);
+	if (err)
+		die("error registering py script extension");
+
+	scripting_context = malloc(sizeof(struct scripting_context));
+}
+
+#ifdef NO_LIBPYTHON
+void setup_python_scripting(void)
+{
+	register_python_scripting(&python_scripting_unsupported_ops);
+}
+#else
+extern struct scripting_ops python_scripting_ops;
+
+void setup_python_scripting(void)
+{
+	register_python_scripting(&python_scripting_ops);
+}
+#endif
+
+static void print_perl_unsupported_msg(void)
+{
+	fprintf(stderr, "Perl scripting not supported."
+		"  Install libperl and rebuild perf to enable it.\n"
+		"For example:\n  # apt-get install libperl-dev (ubuntu)"
+		"\n  # yum install 'perl(ExtUtils::Embed)' (Fedora)"
+		"\n  etc.\n");
+}
+
+static int perl_start_script_unsupported(const char *script __maybe_unused,
+					 int argc __maybe_unused,
+					 const char **argv __maybe_unused)
+{
+	print_perl_unsupported_msg();
+
+	return -1;
+}
+
+static int perl_generate_script_unsupported(struct pevent *pevent
+					    __maybe_unused,
+					    const char *outfile __maybe_unused)
+{
+	print_perl_unsupported_msg();
+
+	return -1;
+}
+
+struct scripting_ops perl_scripting_unsupported_ops = {
+	.name = "Perl",
+	.start_script = perl_start_script_unsupported,
+	.flush_script = flush_script_unsupported,
+	.stop_script = stop_script_unsupported,
+	.process_event = process_event_unsupported,
+	.generate_script = perl_generate_script_unsupported,
+};
+
+static void register_perl_scripting(struct scripting_ops *scripting_ops)
+{
+	int err;
+	err = script_spec_register("Perl", scripting_ops);
+	if (err)
+		die("error registering Perl script extension");
+
+	err = script_spec_register("pl", scripting_ops);
+	if (err)
+		die("error registering pl script extension");
+
+	scripting_context = malloc(sizeof(struct scripting_context));
+}
+
+#ifdef NO_LIBPERL
+void setup_perl_scripting(void)
+{
+	register_perl_scripting(&perl_scripting_unsupported_ops);
+}
+#else
+extern struct scripting_ops perl_scripting_ops;
+
+void setup_perl_scripting(void)
+{
+	register_perl_scripting(&perl_scripting_ops);
+}
+#endif
diff --git a/tools/perf/util/trace-event.c b/tools/perf/util/trace-event.c
new file mode 100644
index 000000000..6322d3716
--- /dev/null
+++ b/tools/perf/util/trace-event.c
@@ -0,0 +1,82 @@
+
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <errno.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <linux/kernel.h>
+#include <traceevent/event-parse.h>
+#include "trace-event.h"
+#include "util.h"
+
+/*
+ * global trace_event object used by trace_event__tp_format
+ *
+ * TODO There's no cleanup call for this. Add some sort of
+ * __exit function support and call trace_event__cleanup
+ * there.
+ */
+static struct trace_event tevent;
+
+int trace_event__init(struct trace_event *t)
+{
+	struct pevent *pevent = pevent_alloc();
+
+	if (pevent) {
+		t->plugin_list = traceevent_load_plugins(pevent);
+		t->pevent  = pevent;
+	}
+
+	return pevent ? 0 : -1;
+}
+
+void trace_event__cleanup(struct trace_event *t)
+{
+	traceevent_unload_plugins(t->plugin_list, t->pevent);
+	pevent_free(t->pevent);
+}
+
+static struct event_format*
+tp_format(const char *sys, const char *name)
+{
+	struct pevent *pevent = tevent.pevent;
+	struct event_format *event = NULL;
+	char path[PATH_MAX];
+	size_t size;
+	char *data;
+
+	scnprintf(path, PATH_MAX, "%s/%s/%s/format",
+		  tracing_events_path, sys, name);
+
+	if (filename__read_str(path, &data, &size))
+		return NULL;
+
+	pevent_parse_format(pevent, &event, data, size, sys);
+
+	free(data);
+	return event;
+}
+
+struct event_format*
+trace_event__tp_format(const char *sys, const char *name)
+{
+	static bool initialized;
+
+	if (!initialized) {
+		int be = traceevent_host_bigendian();
+		struct pevent *pevent;
+
+		if (trace_event__init(&tevent))
+			return NULL;
+
+		pevent = tevent.pevent;
+		pevent_set_flag(pevent, PEVENT_NSEC_OUTPUT);
+		pevent_set_file_bigendian(pevent, be);
+		pevent_set_host_bigendian(pevent, be);
+		initialized = true;
+	}
+
+	return tp_format(sys, name);
+}
diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h
new file mode 100644
index 000000000..d5168f0be
--- /dev/null
+++ b/tools/perf/util/trace-event.h
@@ -0,0 +1,93 @@
+#ifndef _PERF_UTIL_TRACE_EVENT_H
+#define _PERF_UTIL_TRACE_EVENT_H
+
+#include <traceevent/event-parse.h>
+#include "parse-events.h"
+
+struct machine;
+struct perf_sample;
+union perf_event;
+struct perf_tool;
+struct thread;
+struct plugin_list;
+
+struct trace_event {
+	struct pevent		*pevent;
+	struct plugin_list	*plugin_list;
+};
+
+int trace_event__init(struct trace_event *t);
+void trace_event__cleanup(struct trace_event *t);
+struct event_format*
+trace_event__tp_format(const char *sys, const char *name);
+
+int bigendian(void);
+
+void event_format__fprintf(struct event_format *event,
+			   int cpu, void *data, int size, FILE *fp);
+
+void event_format__print(struct event_format *event,
+			 int cpu, void *data, int size);
+
+int parse_ftrace_file(struct pevent *pevent, char *buf, unsigned long size);
+int parse_event_file(struct pevent *pevent,
+		     char *buf, unsigned long size, char *sys);
+
+unsigned long long
+raw_field_value(struct event_format *event, const char *name, void *data);
+
+void parse_proc_kallsyms(struct pevent *pevent, char *file, unsigned int size);
+void parse_ftrace_printk(struct pevent *pevent, char *file, unsigned int size);
+
+ssize_t trace_report(int fd, struct trace_event *tevent, bool repipe);
+
+struct event_format *trace_find_next_event(struct pevent *pevent,
+					   struct event_format *event);
+unsigned long long read_size(struct event_format *event, void *ptr, int size);
+unsigned long long eval_flag(const char *flag);
+
+int read_tracing_data(int fd, struct list_head *pattrs);
+
+struct tracing_data {
+	/* size is only valid if temp is 'true' */
+	ssize_t size;
+	bool temp;
+	char temp_file[50];
+};
+
+struct tracing_data *tracing_data_get(struct list_head *pattrs,
+				      int fd, bool temp);
+int tracing_data_put(struct tracing_data *tdata);
+
+
+struct addr_location;
+
+struct perf_session;
+
+struct scripting_ops {
+	const char *name;
+	int (*start_script) (const char *script, int argc, const char **argv);
+	int (*flush_script) (void);
+	int (*stop_script) (void);
+	void (*process_event) (union perf_event *event,
+			       struct perf_sample *sample,
+			       struct perf_evsel *evsel,
+			       struct addr_location *al);
+	int (*generate_script) (struct pevent *pevent, const char *outfile);
+};
+
+int script_spec_register(const char *spec, struct scripting_ops *ops);
+
+void setup_perl_scripting(void);
+void setup_python_scripting(void);
+
+struct scripting_context {
+	struct pevent *pevent;
+	void *event_data;
+};
+
+int common_pc(struct scripting_context *context);
+int common_flags(struct scripting_context *context);
+int common_lock_depth(struct scripting_context *context);
+
+#endif /* _PERF_UTIL_TRACE_EVENT_H */
diff --git a/tools/perf/util/tsc.c b/tools/perf/util/tsc.c
new file mode 100644
index 000000000..4d4210d4e
--- /dev/null
+++ b/tools/perf/util/tsc.c
@@ -0,0 +1,30 @@
+#include <linux/compiler.h>
+#include <linux/types.h>
+
+#include "tsc.h"
+
+u64 perf_time_to_tsc(u64 ns, struct perf_tsc_conversion *tc)
+{
+	u64 t, quot, rem;
+
+	t = ns - tc->time_zero;
+	quot = t / tc->time_mult;
+	rem  = t % tc->time_mult;
+	return (quot << tc->time_shift) +
+	       (rem << tc->time_shift) / tc->time_mult;
+}
+
+u64 tsc_to_perf_time(u64 cyc, struct perf_tsc_conversion *tc)
+{
+	u64 quot, rem;
+
+	quot = cyc >> tc->time_shift;
+	rem  = cyc & ((1 << tc->time_shift) - 1);
+	return tc->time_zero + quot * tc->time_mult +
+	       ((rem * tc->time_mult) >> tc->time_shift);
+}
+
+u64 __weak rdtsc(void)
+{
+	return 0;
+}
diff --git a/tools/perf/util/tsc.h b/tools/perf/util/tsc.h
new file mode 100644
index 000000000..a8b78f1b3
--- /dev/null
+++ b/tools/perf/util/tsc.h
@@ -0,0 +1,12 @@
+#ifndef __PERF_TSC_H
+#define __PERF_TSC_H
+
+#include <linux/types.h>
+
+#include "../arch/x86/util/tsc.h"
+
+u64 perf_time_to_tsc(u64 ns, struct perf_tsc_conversion *tc);
+u64 tsc_to_perf_time(u64 cyc, struct perf_tsc_conversion *tc);
+u64 rdtsc(void);
+
+#endif
diff --git a/tools/perf/util/unwind-libdw.c b/tools/perf/util/unwind-libdw.c
new file mode 100644
index 000000000..2dcfe9a7c
--- /dev/null
+++ b/tools/perf/util/unwind-libdw.c
@@ -0,0 +1,211 @@
+#include <linux/compiler.h>
+#include <elfutils/libdw.h>
+#include <elfutils/libdwfl.h>
+#include <inttypes.h>
+#include <errno.h>
+#include "debug.h"
+#include "unwind.h"
+#include "unwind-libdw.h"
+#include "machine.h"
+#include "thread.h"
+#include <linux/types.h>
+#include "event.h"
+#include "perf_regs.h"
+
+static char *debuginfo_path;
+
+static const Dwfl_Callbacks offline_callbacks = {
+	.find_debuginfo		= dwfl_standard_find_debuginfo,
+	.debuginfo_path		= &debuginfo_path,
+	.section_address	= dwfl_offline_section_address,
+};
+
+static int __report_module(struct addr_location *al, u64 ip,
+			    struct unwind_info *ui)
+{
+	Dwfl_Module *mod;
+	struct dso *dso = NULL;
+
+	thread__find_addr_location(ui->thread,
+				   PERF_RECORD_MISC_USER,
+				   MAP__FUNCTION, ip, al);
+
+	if (al->map)
+		dso = al->map->dso;
+
+	if (!dso)
+		return 0;
+
+	mod = dwfl_addrmodule(ui->dwfl, ip);
+	if (!mod)
+		mod = dwfl_report_elf(ui->dwfl, dso->short_name,
+				      dso->long_name, -1, al->map->start,
+				      false);
+
+	return mod && dwfl_addrmodule(ui->dwfl, ip) == mod ? 0 : -1;
+}
+
+static int report_module(u64 ip, struct unwind_info *ui)
+{
+	struct addr_location al;
+
+	return __report_module(&al, ip, ui);
+}
+
+static int entry(u64 ip, struct unwind_info *ui)
+
+{
+	struct unwind_entry e;
+	struct addr_location al;
+
+	if (__report_module(&al, ip, ui))
+		return -1;
+
+	e.ip  = ip;
+	e.map = al.map;
+	e.sym = al.sym;
+
+	pr_debug("unwind: %s:ip = 0x%" PRIx64 " (0x%" PRIx64 ")\n",
+		 al.sym ? al.sym->name : "''",
+		 ip,
+		 al.map ? al.map->map_ip(al.map, ip) : (u64) 0);
+
+	return ui->cb(&e, ui->arg);
+}
+
+static pid_t next_thread(Dwfl *dwfl, void *arg, void **thread_argp)
+{
+	/* We want only single thread to be processed. */
+	if (*thread_argp != NULL)
+		return 0;
+
+	*thread_argp = arg;
+	return dwfl_pid(dwfl);
+}
+
+static int access_dso_mem(struct unwind_info *ui, Dwarf_Addr addr,
+			  Dwarf_Word *data)
+{
+	struct addr_location al;
+	ssize_t size;
+
+	thread__find_addr_map(ui->thread, PERF_RECORD_MISC_USER,
+			      MAP__FUNCTION, addr, &al);
+	if (!al.map) {
+		pr_debug("unwind: no map for %lx\n", (unsigned long)addr);
+		return -1;
+	}
+
+	if (!al.map->dso)
+		return -1;
+
+	size = dso__data_read_addr(al.map->dso, al.map, ui->machine,
+				   addr, (u8 *) data, sizeof(*data));
+
+	return !(size == sizeof(*data));
+}
+
+static bool memory_read(Dwfl *dwfl __maybe_unused, Dwarf_Addr addr, Dwarf_Word *result,
+			void *arg)
+{
+	struct unwind_info *ui = arg;
+	struct stack_dump *stack = &ui->sample->user_stack;
+	u64 start, end;
+	int offset;
+	int ret;
+
+	ret = perf_reg_value(&start, &ui->sample->user_regs, PERF_REG_SP);
+	if (ret)
+		return false;
+
+	end = start + stack->size;
+
+	/* Check overflow. */
+	if (addr + sizeof(Dwarf_Word) < addr)
+		return false;
+
+	if (addr < start || addr + sizeof(Dwarf_Word) > end) {
+		ret = access_dso_mem(ui, addr, result);
+		if (ret) {
+			pr_debug("unwind: access_mem 0x%" PRIx64 " not inside range"
+				 " 0x%" PRIx64 "-0x%" PRIx64 "\n",
+				addr, start, end);
+			return false;
+		}
+		return true;
+	}
+
+	offset  = addr - start;
+	*result = *(Dwarf_Word *)&stack->data[offset];
+	pr_debug("unwind: access_mem addr 0x%" PRIx64 ", val %lx, offset %d\n",
+		 addr, (unsigned long)*result, offset);
+	return true;
+}
+
+static const Dwfl_Thread_Callbacks callbacks = {
+	.next_thread		= next_thread,
+	.memory_read		= memory_read,
+	.set_initial_registers	= libdw__arch_set_initial_registers,
+};
+
+static int
+frame_callback(Dwfl_Frame *state, void *arg)
+{
+	struct unwind_info *ui = arg;
+	Dwarf_Addr pc;
+
+	if (!dwfl_frame_pc(state, &pc, NULL)) {
+		pr_err("%s", dwfl_errmsg(-1));
+		return DWARF_CB_ABORT;
+	}
+
+	return entry(pc, ui) || !(--ui->max_stack) ?
+	       DWARF_CB_ABORT : DWARF_CB_OK;
+}
+
+int unwind__get_entries(unwind_entry_cb_t cb, void *arg,
+			struct thread *thread,
+			struct perf_sample *data,
+			int max_stack)
+{
+	struct unwind_info ui = {
+		.sample		= data,
+		.thread		= thread,
+		.machine	= thread->mg->machine,
+		.cb		= cb,
+		.arg		= arg,
+		.max_stack	= max_stack,
+	};
+	Dwarf_Word ip;
+	int err = -EINVAL;
+
+	if (!data->user_regs.regs)
+		return -EINVAL;
+
+	ui.dwfl = dwfl_begin(&offline_callbacks);
+	if (!ui.dwfl)
+		goto out;
+
+	err = perf_reg_value(&ip, &data->user_regs, PERF_REG_IP);
+	if (err)
+		goto out;
+
+	err = report_module(ip, &ui);
+	if (err)
+		goto out;
+
+	if (!dwfl_attach_state(ui.dwfl, EM_NONE, thread->tid, &callbacks, &ui))
+		goto out;
+
+	err = dwfl_getthread_frames(ui.dwfl, thread->tid, frame_callback, &ui);
+
+	if (err && !ui.max_stack)
+		err = 0;
+
+ out:
+	if (err)
+		pr_debug("unwind: failed with '%s'\n", dwfl_errmsg(-1));
+
+	dwfl_end(ui.dwfl);
+	return 0;
+}
diff --git a/tools/perf/util/unwind-libdw.h b/tools/perf/util/unwind-libdw.h
new file mode 100644
index 000000000..417a1426f
--- /dev/null
+++ b/tools/perf/util/unwind-libdw.h
@@ -0,0 +1,21 @@
+#ifndef __PERF_UNWIND_LIBDW_H
+#define __PERF_UNWIND_LIBDW_H
+
+#include <elfutils/libdwfl.h>
+#include "event.h"
+#include "thread.h"
+#include "unwind.h"
+
+bool libdw__arch_set_initial_registers(Dwfl_Thread *thread, void *arg);
+
+struct unwind_info {
+	Dwfl			*dwfl;
+	struct perf_sample      *sample;
+	struct machine          *machine;
+	struct thread           *thread;
+	unwind_entry_cb_t	cb;
+	void			*arg;
+	int			max_stack;
+};
+
+#endif /* __PERF_UNWIND_LIBDW_H */
diff --git a/tools/perf/util/unwind-libunwind.c b/tools/perf/util/unwind-libunwind.c
new file mode 100644
index 000000000..7b09a443a
--- /dev/null
+++ b/tools/perf/util/unwind-libunwind.c
@@ -0,0 +1,653 @@
+/*
+ * Post mortem Dwarf CFI based unwinding on top of regs and stack dumps.
+ *
+ * Lots of this code have been borrowed or heavily inspired from parts of
+ * the libunwind 0.99 code which are (amongst other contributors I may have
+ * forgotten):
+ *
+ * Copyright (C) 2002-2007 Hewlett-Packard Co
+ *	Contributed by David Mosberger-Tang <davidm@hpl.hp.com>
+ *
+ * And the bugs have been added by:
+ *
+ * Copyright (C) 2010, Frederic Weisbecker <fweisbec@gmail.com>
+ * Copyright (C) 2012, Jiri Olsa <jolsa@redhat.com>
+ *
+ */
+
+#include <elf.h>
+#include <gelf.h>
+#include <fcntl.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/mman.h>
+#include <linux/list.h>
+#include <libunwind.h>
+#include <libunwind-ptrace.h>
+#include "callchain.h"
+#include "thread.h"
+#include "session.h"
+#include "perf_regs.h"
+#include "unwind.h"
+#include "symbol.h"
+#include "util.h"
+#include "debug.h"
+
+extern int
+UNW_OBJ(dwarf_search_unwind_table) (unw_addr_space_t as,
+				    unw_word_t ip,
+				    unw_dyn_info_t *di,
+				    unw_proc_info_t *pi,
+				    int need_unwind_info, void *arg);
+
+#define dwarf_search_unwind_table UNW_OBJ(dwarf_search_unwind_table)
+
+extern int
+UNW_OBJ(dwarf_find_debug_frame) (int found, unw_dyn_info_t *di_debug,
+				 unw_word_t ip,
+				 unw_word_t segbase,
+				 const char *obj_name, unw_word_t start,
+				 unw_word_t end);
+
+#define dwarf_find_debug_frame UNW_OBJ(dwarf_find_debug_frame)
+
+#define DW_EH_PE_FORMAT_MASK	0x0f	/* format of the encoded value */
+#define DW_EH_PE_APPL_MASK	0x70	/* how the value is to be applied */
+
+/* Pointer-encoding formats: */
+#define DW_EH_PE_omit		0xff
+#define DW_EH_PE_ptr		0x00	/* pointer-sized unsigned value */
+#define DW_EH_PE_udata4		0x03	/* unsigned 32-bit value */
+#define DW_EH_PE_udata8		0x04	/* unsigned 64-bit value */
+#define DW_EH_PE_sdata4		0x0b	/* signed 32-bit value */
+#define DW_EH_PE_sdata8		0x0c	/* signed 64-bit value */
+
+/* Pointer-encoding application: */
+#define DW_EH_PE_absptr		0x00	/* absolute value */
+#define DW_EH_PE_pcrel		0x10	/* rel. to addr. of encoded value */
+
+/*
+ * The following are not documented by LSB v1.3, yet they are used by
+ * GCC, presumably they aren't documented by LSB since they aren't
+ * used on Linux:
+ */
+#define DW_EH_PE_funcrel	0x40	/* start-of-procedure-relative */
+#define DW_EH_PE_aligned	0x50	/* aligned pointer */
+
+/* Flags intentionaly not handled, since they're not needed:
+ * #define DW_EH_PE_indirect      0x80
+ * #define DW_EH_PE_uleb128       0x01
+ * #define DW_EH_PE_udata2        0x02
+ * #define DW_EH_PE_sleb128       0x09
+ * #define DW_EH_PE_sdata2        0x0a
+ * #define DW_EH_PE_textrel       0x20
+ * #define DW_EH_PE_datarel       0x30
+ */
+
+struct unwind_info {
+	struct perf_sample	*sample;
+	struct machine		*machine;
+	struct thread		*thread;
+};
+
+#define dw_read(ptr, type, end) ({	\
+	type *__p = (type *) ptr;	\
+	type  __v;			\
+	if ((__p + 1) > (type *) end)	\
+		return -EINVAL;		\
+	__v = *__p++;			\
+	ptr = (typeof(ptr)) __p;	\
+	__v;				\
+	})
+
+static int __dw_read_encoded_value(u8 **p, u8 *end, u64 *val,
+				   u8 encoding)
+{
+	u8 *cur = *p;
+	*val = 0;
+
+	switch (encoding) {
+	case DW_EH_PE_omit:
+		*val = 0;
+		goto out;
+	case DW_EH_PE_ptr:
+		*val = dw_read(cur, unsigned long, end);
+		goto out;
+	default:
+		break;
+	}
+
+	switch (encoding & DW_EH_PE_APPL_MASK) {
+	case DW_EH_PE_absptr:
+		break;
+	case DW_EH_PE_pcrel:
+		*val = (unsigned long) cur;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if ((encoding & 0x07) == 0x00)
+		encoding |= DW_EH_PE_udata4;
+
+	switch (encoding & DW_EH_PE_FORMAT_MASK) {
+	case DW_EH_PE_sdata4:
+		*val += dw_read(cur, s32, end);
+		break;
+	case DW_EH_PE_udata4:
+		*val += dw_read(cur, u32, end);
+		break;
+	case DW_EH_PE_sdata8:
+		*val += dw_read(cur, s64, end);
+		break;
+	case DW_EH_PE_udata8:
+		*val += dw_read(cur, u64, end);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+ out:
+	*p = cur;
+	return 0;
+}
+
+#define dw_read_encoded_value(ptr, end, enc) ({			\
+	u64 __v;						\
+	if (__dw_read_encoded_value(&ptr, end, &__v, enc)) {	\
+		return -EINVAL;                                 \
+	}                                                       \
+	__v;                                                    \
+	})
+
+static u64 elf_section_offset(int fd, const char *name)
+{
+	Elf *elf;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+	u64 offset = 0;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL)
+		return 0;
+
+	do {
+		if (gelf_getehdr(elf, &ehdr) == NULL)
+			break;
+
+		if (!elf_section_by_name(elf, &ehdr, &shdr, name, NULL))
+			break;
+
+		offset = shdr.sh_offset;
+	} while (0);
+
+	elf_end(elf);
+	return offset;
+}
+
+#ifndef NO_LIBUNWIND_DEBUG_FRAME
+static int elf_is_exec(int fd, const char *name)
+{
+	Elf *elf;
+	GElf_Ehdr ehdr;
+	int retval = 0;
+
+	elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL);
+	if (elf == NULL)
+		return 0;
+	if (gelf_getehdr(elf, &ehdr) == NULL)
+		goto out;
+
+	retval = (ehdr.e_type == ET_EXEC);
+
+out:
+	elf_end(elf);
+	pr_debug("unwind: elf_is_exec(%s): %d\n", name, retval);
+	return retval;
+}
+#endif
+
+struct table_entry {
+	u32 start_ip_offset;
+	u32 fde_offset;
+};
+
+struct eh_frame_hdr {
+	unsigned char version;
+	unsigned char eh_frame_ptr_enc;
+	unsigned char fde_count_enc;
+	unsigned char table_enc;
+
+	/*
+	 * The rest of the header is variable-length and consists of the
+	 * following members:
+	 *
+	 *	encoded_t eh_frame_ptr;
+	 *	encoded_t fde_count;
+	 */
+
+	/* A single encoded pointer should not be more than 8 bytes. */
+	u64 enc[2];
+
+	/*
+	 * struct {
+	 *    encoded_t start_ip;
+	 *    encoded_t fde_addr;
+	 * } binary_search_table[fde_count];
+	 */
+	char data[0];
+} __packed;
+
+static int unwind_spec_ehframe(struct dso *dso, struct machine *machine,
+			       u64 offset, u64 *table_data, u64 *segbase,
+			       u64 *fde_count)
+{
+	struct eh_frame_hdr hdr;
+	u8 *enc = (u8 *) &hdr.enc;
+	u8 *end = (u8 *) &hdr.data;
+	ssize_t r;
+
+	r = dso__data_read_offset(dso, machine, offset,
+				  (u8 *) &hdr, sizeof(hdr));
+	if (r != sizeof(hdr))
+		return -EINVAL;
+
+	/* We dont need eh_frame_ptr, just skip it. */
+	dw_read_encoded_value(enc, end, hdr.eh_frame_ptr_enc);
+
+	*fde_count  = dw_read_encoded_value(enc, end, hdr.fde_count_enc);
+	*segbase    = offset;
+	*table_data = (enc - (u8 *) &hdr) + offset;
+	return 0;
+}
+
+static int read_unwind_spec_eh_frame(struct dso *dso, struct machine *machine,
+				     u64 *table_data, u64 *segbase,
+				     u64 *fde_count)
+{
+	int ret = -EINVAL, fd;
+	u64 offset = dso->data.eh_frame_hdr_offset;
+
+	if (offset == 0) {
+		fd = dso__data_fd(dso, machine);
+		if (fd < 0)
+			return -EINVAL;
+
+		/* Check the .eh_frame section for unwinding info */
+		offset = elf_section_offset(fd, ".eh_frame_hdr");
+		dso->data.eh_frame_hdr_offset = offset;
+	}
+
+	if (offset)
+		ret = unwind_spec_ehframe(dso, machine, offset,
+					  table_data, segbase,
+					  fde_count);
+
+	return ret;
+}
+
+#ifndef NO_LIBUNWIND_DEBUG_FRAME
+static int read_unwind_spec_debug_frame(struct dso *dso,
+					struct machine *machine, u64 *offset)
+{
+	int fd;
+	u64 ofs = dso->data.debug_frame_offset;
+
+	if (ofs == 0) {
+		fd = dso__data_fd(dso, machine);
+		if (fd < 0)
+			return -EINVAL;
+
+		/* Check the .debug_frame section for unwinding info */
+		ofs = elf_section_offset(fd, ".debug_frame");
+		dso->data.debug_frame_offset = ofs;
+	}
+
+	*offset = ofs;
+	if (*offset)
+		return 0;
+
+	return -EINVAL;
+}
+#endif
+
+static struct map *find_map(unw_word_t ip, struct unwind_info *ui)
+{
+	struct addr_location al;
+
+	thread__find_addr_map(ui->thread, PERF_RECORD_MISC_USER,
+			      MAP__FUNCTION, ip, &al);
+	return al.map;
+}
+
+static int
+find_proc_info(unw_addr_space_t as, unw_word_t ip, unw_proc_info_t *pi,
+	       int need_unwind_info, void *arg)
+{
+	struct unwind_info *ui = arg;
+	struct map *map;
+	unw_dyn_info_t di;
+	u64 table_data, segbase, fde_count;
+
+	map = find_map(ip, ui);
+	if (!map || !map->dso)
+		return -EINVAL;
+
+	pr_debug("unwind: find_proc_info dso %s\n", map->dso->name);
+
+	/* Check the .eh_frame section for unwinding info */
+	if (!read_unwind_spec_eh_frame(map->dso, ui->machine,
+				       &table_data, &segbase, &fde_count)) {
+		memset(&di, 0, sizeof(di));
+		di.format   = UNW_INFO_FORMAT_REMOTE_TABLE;
+		di.start_ip = map->start;
+		di.end_ip   = map->end;
+		di.u.rti.segbase    = map->start + segbase;
+		di.u.rti.table_data = map->start + table_data;
+		di.u.rti.table_len  = fde_count * sizeof(struct table_entry)
+				      / sizeof(unw_word_t);
+		return dwarf_search_unwind_table(as, ip, &di, pi,
+						 need_unwind_info, arg);
+	}
+
+#ifndef NO_LIBUNWIND_DEBUG_FRAME
+	/* Check the .debug_frame section for unwinding info */
+	if (!read_unwind_spec_debug_frame(map->dso, ui->machine, &segbase)) {
+		int fd = dso__data_fd(map->dso, ui->machine);
+		int is_exec = elf_is_exec(fd, map->dso->name);
+		unw_word_t base = is_exec ? 0 : map->start;
+
+		memset(&di, 0, sizeof(di));
+		if (dwarf_find_debug_frame(0, &di, ip, base, map->dso->name,
+					   map->start, map->end))
+			return dwarf_search_unwind_table(as, ip, &di, pi,
+							 need_unwind_info, arg);
+	}
+#endif
+
+	return -EINVAL;
+}
+
+static int access_fpreg(unw_addr_space_t __maybe_unused as,
+			unw_regnum_t __maybe_unused num,
+			unw_fpreg_t __maybe_unused *val,
+			int __maybe_unused __write,
+			void __maybe_unused *arg)
+{
+	pr_err("unwind: access_fpreg unsupported\n");
+	return -UNW_EINVAL;
+}
+
+static int get_dyn_info_list_addr(unw_addr_space_t __maybe_unused as,
+				  unw_word_t __maybe_unused *dil_addr,
+				  void __maybe_unused *arg)
+{
+	return -UNW_ENOINFO;
+}
+
+static int resume(unw_addr_space_t __maybe_unused as,
+		  unw_cursor_t __maybe_unused *cu,
+		  void __maybe_unused *arg)
+{
+	pr_err("unwind: resume unsupported\n");
+	return -UNW_EINVAL;
+}
+
+static int
+get_proc_name(unw_addr_space_t __maybe_unused as,
+	      unw_word_t __maybe_unused addr,
+		char __maybe_unused *bufp, size_t __maybe_unused buf_len,
+		unw_word_t __maybe_unused *offp, void __maybe_unused *arg)
+{
+	pr_err("unwind: get_proc_name unsupported\n");
+	return -UNW_EINVAL;
+}
+
+static int access_dso_mem(struct unwind_info *ui, unw_word_t addr,
+			  unw_word_t *data)
+{
+	struct addr_location al;
+	ssize_t size;
+
+	thread__find_addr_map(ui->thread, PERF_RECORD_MISC_USER,
+			      MAP__FUNCTION, addr, &al);
+	if (!al.map) {
+		pr_debug("unwind: no map for %lx\n", (unsigned long)addr);
+		return -1;
+	}
+
+	if (!al.map->dso)
+		return -1;
+
+	size = dso__data_read_addr(al.map->dso, al.map, ui->machine,
+				   addr, (u8 *) data, sizeof(*data));
+
+	return !(size == sizeof(*data));
+}
+
+static int access_mem(unw_addr_space_t __maybe_unused as,
+		      unw_word_t addr, unw_word_t *valp,
+		      int __write, void *arg)
+{
+	struct unwind_info *ui = arg;
+	struct stack_dump *stack = &ui->sample->user_stack;
+	u64 start, end;
+	int offset;
+	int ret;
+
+	/* Don't support write, probably not needed. */
+	if (__write || !stack || !ui->sample->user_regs.regs) {
+		*valp = 0;
+		return 0;
+	}
+
+	ret = perf_reg_value(&start, &ui->sample->user_regs, PERF_REG_SP);
+	if (ret)
+		return ret;
+
+	end = start + stack->size;
+
+	/* Check overflow. */
+	if (addr + sizeof(unw_word_t) < addr)
+		return -EINVAL;
+
+	if (addr < start || addr + sizeof(unw_word_t) >= end) {
+		ret = access_dso_mem(ui, addr, valp);
+		if (ret) {
+			pr_debug("unwind: access_mem %p not inside range"
+				 " 0x%" PRIx64 "-0x%" PRIx64 "\n",
+				 (void *) addr, start, end);
+			*valp = 0;
+			return ret;
+		}
+		return 0;
+	}
+
+	offset = addr - start;
+	*valp  = *(unw_word_t *)&stack->data[offset];
+	pr_debug("unwind: access_mem addr %p val %lx, offset %d\n",
+		 (void *) addr, (unsigned long)*valp, offset);
+	return 0;
+}
+
+static int access_reg(unw_addr_space_t __maybe_unused as,
+		      unw_regnum_t regnum, unw_word_t *valp,
+		      int __write, void *arg)
+{
+	struct unwind_info *ui = arg;
+	int id, ret;
+	u64 val;
+
+	/* Don't support write, I suspect we don't need it. */
+	if (__write) {
+		pr_err("unwind: access_reg w %d\n", regnum);
+		return 0;
+	}
+
+	if (!ui->sample->user_regs.regs) {
+		*valp = 0;
+		return 0;
+	}
+
+	id = libunwind__arch_reg_id(regnum);
+	if (id < 0)
+		return -EINVAL;
+
+	ret = perf_reg_value(&val, &ui->sample->user_regs, id);
+	if (ret) {
+		pr_err("unwind: can't read reg %d\n", regnum);
+		return ret;
+	}
+
+	*valp = (unw_word_t) val;
+	pr_debug("unwind: reg %d, val %lx\n", regnum, (unsigned long)*valp);
+	return 0;
+}
+
+static void put_unwind_info(unw_addr_space_t __maybe_unused as,
+			    unw_proc_info_t *pi __maybe_unused,
+			    void *arg __maybe_unused)
+{
+	pr_debug("unwind: put_unwind_info called\n");
+}
+
+static int entry(u64 ip, struct thread *thread,
+		 unwind_entry_cb_t cb, void *arg)
+{
+	struct unwind_entry e;
+	struct addr_location al;
+
+	thread__find_addr_location(thread, PERF_RECORD_MISC_USER,
+				   MAP__FUNCTION, ip, &al);
+
+	e.ip = ip;
+	e.map = al.map;
+	e.sym = al.sym;
+
+	pr_debug("unwind: %s:ip = 0x%" PRIx64 " (0x%" PRIx64 ")\n",
+		 al.sym ? al.sym->name : "''",
+		 ip,
+		 al.map ? al.map->map_ip(al.map, ip) : (u64) 0);
+
+	return cb(&e, arg);
+}
+
+static void display_error(int err)
+{
+	switch (err) {
+	case UNW_EINVAL:
+		pr_err("unwind: Only supports local.\n");
+		break;
+	case UNW_EUNSPEC:
+		pr_err("unwind: Unspecified error.\n");
+		break;
+	case UNW_EBADREG:
+		pr_err("unwind: Register unavailable.\n");
+		break;
+	default:
+		break;
+	}
+}
+
+static unw_accessors_t accessors = {
+	.find_proc_info		= find_proc_info,
+	.put_unwind_info	= put_unwind_info,
+	.get_dyn_info_list_addr	= get_dyn_info_list_addr,
+	.access_mem		= access_mem,
+	.access_reg		= access_reg,
+	.access_fpreg		= access_fpreg,
+	.resume			= resume,
+	.get_proc_name		= get_proc_name,
+};
+
+int unwind__prepare_access(struct thread *thread)
+{
+	unw_addr_space_t addr_space;
+
+	if (callchain_param.record_mode != CALLCHAIN_DWARF)
+		return 0;
+
+	addr_space = unw_create_addr_space(&accessors, 0);
+	if (!addr_space) {
+		pr_err("unwind: Can't create unwind address space.\n");
+		return -ENOMEM;
+	}
+
+	unw_set_caching_policy(addr_space, UNW_CACHE_GLOBAL);
+	thread__set_priv(thread, addr_space);
+
+	return 0;
+}
+
+void unwind__flush_access(struct thread *thread)
+{
+	unw_addr_space_t addr_space;
+
+	if (callchain_param.record_mode != CALLCHAIN_DWARF)
+		return;
+
+	addr_space = thread__priv(thread);
+	unw_flush_cache(addr_space, 0, 0);
+}
+
+void unwind__finish_access(struct thread *thread)
+{
+	unw_addr_space_t addr_space;
+
+	if (callchain_param.record_mode != CALLCHAIN_DWARF)
+		return;
+
+	addr_space = thread__priv(thread);
+	unw_destroy_addr_space(addr_space);
+}
+
+static int get_entries(struct unwind_info *ui, unwind_entry_cb_t cb,
+		       void *arg, int max_stack)
+{
+	unw_addr_space_t addr_space;
+	unw_cursor_t c;
+	int ret;
+
+	addr_space = thread__priv(ui->thread);
+	if (addr_space == NULL)
+		return -1;
+
+	ret = unw_init_remote(&c, addr_space, ui);
+	if (ret)
+		display_error(ret);
+
+	while (!ret && (unw_step(&c) > 0) && max_stack--) {
+		unw_word_t ip;
+
+		unw_get_reg(&c, UNW_REG_IP, &ip);
+		ret = ip ? entry(ip, ui->thread, cb, arg) : 0;
+	}
+
+	return ret;
+}
+
+int unwind__get_entries(unwind_entry_cb_t cb, void *arg,
+			struct thread *thread,
+			struct perf_sample *data, int max_stack)
+{
+	u64 ip;
+	struct unwind_info ui = {
+		.sample       = data,
+		.thread       = thread,
+		.machine      = thread->mg->machine,
+	};
+	int ret;
+
+	if (!data->user_regs.regs)
+		return -EINVAL;
+
+	ret = perf_reg_value(&ip, &data->user_regs, PERF_REG_IP);
+	if (ret)
+		return ret;
+
+	ret = entry(ip, thread, cb, arg);
+	if (ret)
+		return -ENOMEM;
+
+	return --max_stack > 0 ? get_entries(&ui, cb, arg, max_stack) : 0;
+}
diff --git a/tools/perf/util/unwind.h b/tools/perf/util/unwind.h
new file mode 100644
index 000000000..12790cf94
--- /dev/null
+++ b/tools/perf/util/unwind.h
@@ -0,0 +1,55 @@
+#ifndef __UNWIND_H
+#define __UNWIND_H
+
+#include <linux/types.h>
+#include "event.h"
+#include "symbol.h"
+#include "thread.h"
+
+struct unwind_entry {
+	struct map	*map;
+	struct symbol	*sym;
+	u64		ip;
+};
+
+typedef int (*unwind_entry_cb_t)(struct unwind_entry *entry, void *arg);
+
+#ifdef HAVE_DWARF_UNWIND_SUPPORT
+int unwind__get_entries(unwind_entry_cb_t cb, void *arg,
+			struct thread *thread,
+			struct perf_sample *data, int max_stack);
+/* libunwind specific */
+#ifdef HAVE_LIBUNWIND_SUPPORT
+int libunwind__arch_reg_id(int regnum);
+int unwind__prepare_access(struct thread *thread);
+void unwind__flush_access(struct thread *thread);
+void unwind__finish_access(struct thread *thread);
+#else
+static inline int unwind__prepare_access(struct thread *thread __maybe_unused)
+{
+	return 0;
+}
+
+static inline void unwind__flush_access(struct thread *thread __maybe_unused) {}
+static inline void unwind__finish_access(struct thread *thread __maybe_unused) {}
+#endif
+#else
+static inline int
+unwind__get_entries(unwind_entry_cb_t cb __maybe_unused,
+		    void *arg __maybe_unused,
+		    struct thread *thread __maybe_unused,
+		    struct perf_sample *data __maybe_unused,
+		    int max_stack __maybe_unused)
+{
+	return 0;
+}
+
+static inline int unwind__prepare_access(struct thread *thread __maybe_unused)
+{
+	return 0;
+}
+
+static inline void unwind__flush_access(struct thread *thread __maybe_unused) {}
+static inline void unwind__finish_access(struct thread *thread __maybe_unused) {}
+#endif /* HAVE_DWARF_UNWIND_SUPPORT */
+#endif /* __UNWIND_H */
diff --git a/tools/perf/util/usage.c b/tools/perf/util/usage.c
new file mode 100644
index 000000000..4007aca8e
--- /dev/null
+++ b/tools/perf/util/usage.c
@@ -0,0 +1,84 @@
+/*
+ * usage.c
+ *
+ * Various reporting routines.
+ * Originally copied from GIT source.
+ *
+ * Copyright (C) Linus Torvalds, 2005
+ */
+#include "util.h"
+#include "debug.h"
+
+static void report(const char *prefix, const char *err, va_list params)
+{
+	char msg[1024];
+	vsnprintf(msg, sizeof(msg), err, params);
+	fprintf(stderr, " %s%s\n", prefix, msg);
+}
+
+static NORETURN void usage_builtin(const char *err)
+{
+	fprintf(stderr, "\n Usage: %s\n", err);
+	exit(129);
+}
+
+static NORETURN void die_builtin(const char *err, va_list params)
+{
+	report(" Fatal: ", err, params);
+	exit(128);
+}
+
+static void error_builtin(const char *err, va_list params)
+{
+	report(" Error: ", err, params);
+}
+
+static void warn_builtin(const char *warn, va_list params)
+{
+	report(" Warning: ", warn, params);
+}
+
+/* If we are in a dlopen()ed .so write to a global variable would segfault
+ * (ugh), so keep things static. */
+static void (*usage_routine)(const char *err) NORETURN = usage_builtin;
+static void (*die_routine)(const char *err, va_list params) NORETURN = die_builtin;
+static void (*error_routine)(const char *err, va_list params) = error_builtin;
+static void (*warn_routine)(const char *err, va_list params) = warn_builtin;
+
+void set_die_routine(void (*routine)(const char *err, va_list params) NORETURN)
+{
+	die_routine = routine;
+}
+
+void usage(const char *err)
+{
+	usage_routine(err);
+}
+
+void die(const char *err, ...)
+{
+	va_list params;
+
+	va_start(params, err);
+	die_routine(err, params);
+	va_end(params);
+}
+
+int error(const char *err, ...)
+{
+	va_list params;
+
+	va_start(params, err);
+	error_routine(err, params);
+	va_end(params);
+	return -1;
+}
+
+void warning(const char *warn, ...)
+{
+	va_list params;
+
+	va_start(params, warn);
+	warn_routine(warn, params);
+	va_end(params);
+}
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
new file mode 100644
index 000000000..4ee6d0d4c
--- /dev/null
+++ b/tools/perf/util/util.c
@@ -0,0 +1,617 @@
+#include "../perf.h"
+#include "util.h"
+#include "debug.h"
+#include <api/fs/fs.h>
+#include <sys/mman.h>
+#ifdef HAVE_BACKTRACE_SUPPORT
+#include <execinfo.h>
+#endif
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <limits.h>
+#include <byteswap.h>
+#include <linux/kernel.h>
+#include <unistd.h>
+#include "callchain.h"
+
+struct callchain_param	callchain_param = {
+	.mode	= CHAIN_GRAPH_REL,
+	.min_percent = 0.5,
+	.order  = ORDER_CALLEE,
+	.key	= CCKEY_FUNCTION
+};
+
+/*
+ * XXX We need to find a better place for these things...
+ */
+unsigned int page_size;
+int cacheline_size;
+
+bool test_attr__enabled;
+
+bool perf_host  = true;
+bool perf_guest = false;
+
+char tracing_events_path[PATH_MAX + 1] = "/sys/kernel/debug/tracing/events";
+
+void event_attr_init(struct perf_event_attr *attr)
+{
+	if (!perf_host)
+		attr->exclude_host  = 1;
+	if (!perf_guest)
+		attr->exclude_guest = 1;
+	/* to capture ABI version */
+	attr->size = sizeof(*attr);
+}
+
+int mkdir_p(char *path, mode_t mode)
+{
+	struct stat st;
+	int err;
+	char *d = path;
+
+	if (*d != '/')
+		return -1;
+
+	if (stat(path, &st) == 0)
+		return 0;
+
+	while (*++d == '/');
+
+	while ((d = strchr(d, '/'))) {
+		*d = '\0';
+		err = stat(path, &st) && mkdir(path, mode);
+		*d++ = '/';
+		if (err)
+			return -1;
+		while (*d == '/')
+			++d;
+	}
+	return (stat(path, &st) && mkdir(path, mode)) ? -1 : 0;
+}
+
+static int slow_copyfile(const char *from, const char *to, mode_t mode)
+{
+	int err = -1;
+	char *line = NULL;
+	size_t n;
+	FILE *from_fp = fopen(from, "r"), *to_fp;
+	mode_t old_umask;
+
+	if (from_fp == NULL)
+		goto out;
+
+	old_umask = umask(mode ^ 0777);
+	to_fp = fopen(to, "w");
+	umask(old_umask);
+	if (to_fp == NULL)
+		goto out_fclose_from;
+
+	while (getline(&line, &n, from_fp) > 0)
+		if (fputs(line, to_fp) == EOF)
+			goto out_fclose_to;
+	err = 0;
+out_fclose_to:
+	fclose(to_fp);
+	free(line);
+out_fclose_from:
+	fclose(from_fp);
+out:
+	return err;
+}
+
+int copyfile_mode(const char *from, const char *to, mode_t mode)
+{
+	int fromfd, tofd;
+	struct stat st;
+	void *addr;
+	int err = -1;
+
+	if (stat(from, &st))
+		goto out;
+
+	if (st.st_size == 0) /* /proc? do it slowly... */
+		return slow_copyfile(from, to, mode);
+
+	fromfd = open(from, O_RDONLY);
+	if (fromfd < 0)
+		goto out;
+
+	tofd = creat(to, mode);
+	if (tofd < 0)
+		goto out_close_from;
+
+	addr = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fromfd, 0);
+	if (addr == MAP_FAILED)
+		goto out_close_to;
+
+	if (write(tofd, addr, st.st_size) == st.st_size)
+		err = 0;
+
+	munmap(addr, st.st_size);
+out_close_to:
+	close(tofd);
+	if (err)
+		unlink(to);
+out_close_from:
+	close(fromfd);
+out:
+	return err;
+}
+
+int copyfile(const char *from, const char *to)
+{
+	return copyfile_mode(from, to, 0755);
+}
+
+unsigned long convert_unit(unsigned long value, char *unit)
+{
+	*unit = ' ';
+
+	if (value > 1000) {
+		value /= 1000;
+		*unit = 'K';
+	}
+
+	if (value > 1000) {
+		value /= 1000;
+		*unit = 'M';
+	}
+
+	if (value > 1000) {
+		value /= 1000;
+		*unit = 'G';
+	}
+
+	return value;
+}
+
+static ssize_t ion(bool is_read, int fd, void *buf, size_t n)
+{
+	void *buf_start = buf;
+	size_t left = n;
+
+	while (left) {
+		ssize_t ret = is_read ? read(fd, buf, left) :
+					write(fd, buf, left);
+
+		if (ret < 0 && errno == EINTR)
+			continue;
+		if (ret <= 0)
+			return ret;
+
+		left -= ret;
+		buf  += ret;
+	}
+
+	BUG_ON((size_t)(buf - buf_start) != n);
+	return n;
+}
+
+/*
+ * Read exactly 'n' bytes or return an error.
+ */
+ssize_t readn(int fd, void *buf, size_t n)
+{
+	return ion(true, fd, buf, n);
+}
+
+/*
+ * Write exactly 'n' bytes or return an error.
+ */
+ssize_t writen(int fd, void *buf, size_t n)
+{
+	return ion(false, fd, buf, n);
+}
+
+size_t hex_width(u64 v)
+{
+	size_t n = 1;
+
+	while ((v >>= 4))
+		++n;
+
+	return n;
+}
+
+static int hex(char ch)
+{
+	if ((ch >= '0') && (ch <= '9'))
+		return ch - '0';
+	if ((ch >= 'a') && (ch <= 'f'))
+		return ch - 'a' + 10;
+	if ((ch >= 'A') && (ch <= 'F'))
+		return ch - 'A' + 10;
+	return -1;
+}
+
+/*
+ * While we find nice hex chars, build a long_val.
+ * Return number of chars processed.
+ */
+int hex2u64(const char *ptr, u64 *long_val)
+{
+	const char *p = ptr;
+	*long_val = 0;
+
+	while (*p) {
+		const int hex_val = hex(*p);
+
+		if (hex_val < 0)
+			break;
+
+		*long_val = (*long_val << 4) | hex_val;
+		p++;
+	}
+
+	return p - ptr;
+}
+
+/* Obtain a backtrace and print it to stdout. */
+#ifdef HAVE_BACKTRACE_SUPPORT
+void dump_stack(void)
+{
+	void *array[16];
+	size_t size = backtrace(array, ARRAY_SIZE(array));
+	char **strings = backtrace_symbols(array, size);
+	size_t i;
+
+	printf("Obtained %zd stack frames.\n", size);
+
+	for (i = 0; i < size; i++)
+		printf("%s\n", strings[i]);
+
+	free(strings);
+}
+#else
+void dump_stack(void) {}
+#endif
+
+void sighandler_dump_stack(int sig)
+{
+	psignal(sig, "perf");
+	dump_stack();
+	exit(sig);
+}
+
+void get_term_dimensions(struct winsize *ws)
+{
+	char *s = getenv("LINES");
+
+	if (s != NULL) {
+		ws->ws_row = atoi(s);
+		s = getenv("COLUMNS");
+		if (s != NULL) {
+			ws->ws_col = atoi(s);
+			if (ws->ws_row && ws->ws_col)
+				return;
+		}
+	}
+#ifdef TIOCGWINSZ
+	if (ioctl(1, TIOCGWINSZ, ws) == 0 &&
+	    ws->ws_row && ws->ws_col)
+		return;
+#endif
+	ws->ws_row = 25;
+	ws->ws_col = 80;
+}
+
+void set_term_quiet_input(struct termios *old)
+{
+	struct termios tc;
+
+	tcgetattr(0, old);
+	tc = *old;
+	tc.c_lflag &= ~(ICANON | ECHO);
+	tc.c_cc[VMIN] = 0;
+	tc.c_cc[VTIME] = 0;
+	tcsetattr(0, TCSANOW, &tc);
+}
+
+static void set_tracing_events_path(const char *tracing, const char *mountpoint)
+{
+	snprintf(tracing_events_path, sizeof(tracing_events_path), "%s/%s%s",
+		 mountpoint, tracing, "events");
+}
+
+static const char *__perf_tracefs_mount(const char *mountpoint)
+{
+	const char *mnt;
+
+	mnt = tracefs_mount(mountpoint);
+	if (!mnt)
+		return NULL;
+
+	set_tracing_events_path("", mnt);
+
+	return mnt;
+}
+
+static const char *__perf_debugfs_mount(const char *mountpoint)
+{
+	const char *mnt;
+
+	mnt = debugfs_mount(mountpoint);
+	if (!mnt)
+		return NULL;
+
+	set_tracing_events_path("tracing/", mnt);
+
+	return mnt;
+}
+
+const char *perf_debugfs_mount(const char *mountpoint)
+{
+	const char *mnt;
+
+	mnt = __perf_tracefs_mount(mountpoint);
+	if (mnt)
+		return mnt;
+
+	mnt = __perf_debugfs_mount(mountpoint);
+
+	return mnt;
+}
+
+void perf_debugfs_set_path(const char *mntpt)
+{
+	snprintf(debugfs_mountpoint, strlen(debugfs_mountpoint), "%s", mntpt);
+	set_tracing_events_path("tracing/", mntpt);
+}
+
+static const char *find_tracefs(void)
+{
+	const char *path = __perf_tracefs_mount(NULL);
+
+	return path;
+}
+
+static const char *find_debugfs(void)
+{
+	const char *path = __perf_debugfs_mount(NULL);
+
+	if (!path)
+		fprintf(stderr, "Your kernel does not support the debugfs filesystem");
+
+	return path;
+}
+
+/*
+ * Finds the path to the debugfs/tracing
+ * Allocates the string and stores it.
+ */
+const char *find_tracing_dir(void)
+{
+	const char *tracing_dir = "";
+	static char *tracing;
+	static int tracing_found;
+	const char *debugfs;
+
+	if (tracing_found)
+		return tracing;
+
+	debugfs = find_tracefs();
+	if (!debugfs) {
+		tracing_dir = "/tracing";
+		debugfs = find_debugfs();
+		if (!debugfs)
+			return NULL;
+	}
+
+	if (asprintf(&tracing, "%s%s", debugfs, tracing_dir) < 0)
+		return NULL;
+
+	tracing_found = 1;
+	return tracing;
+}
+
+char *get_tracing_file(const char *name)
+{
+	const char *tracing;
+	char *file;
+
+	tracing = find_tracing_dir();
+	if (!tracing)
+		return NULL;
+
+	if (asprintf(&file, "%s/%s", tracing, name) < 0)
+		return NULL;
+
+	return file;
+}
+
+void put_tracing_file(char *file)
+{
+	free(file);
+}
+
+int parse_nsec_time(const char *str, u64 *ptime)
+{
+	u64 time_sec, time_nsec;
+	char *end;
+
+	time_sec = strtoul(str, &end, 10);
+	if (*end != '.' && *end != '\0')
+		return -1;
+
+	if (*end == '.') {
+		int i;
+		char nsec_buf[10];
+
+		if (strlen(++end) > 9)
+			return -1;
+
+		strncpy(nsec_buf, end, 9);
+		nsec_buf[9] = '\0';
+
+		/* make it nsec precision */
+		for (i = strlen(nsec_buf); i < 9; i++)
+			nsec_buf[i] = '0';
+
+		time_nsec = strtoul(nsec_buf, &end, 10);
+		if (*end != '\0')
+			return -1;
+	} else
+		time_nsec = 0;
+
+	*ptime = time_sec * NSEC_PER_SEC + time_nsec;
+	return 0;
+}
+
+unsigned long parse_tag_value(const char *str, struct parse_tag *tags)
+{
+	struct parse_tag *i = tags;
+
+	while (i->tag) {
+		char *s;
+
+		s = strchr(str, i->tag);
+		if (s) {
+			unsigned long int value;
+			char *endptr;
+
+			value = strtoul(str, &endptr, 10);
+			if (s != endptr)
+				break;
+
+			if (value > ULONG_MAX / i->mult)
+				break;
+			value *= i->mult;
+			return value;
+		}
+		i++;
+	}
+
+	return (unsigned long) -1;
+}
+
+int filename__read_str(const char *filename, char **buf, size_t *sizep)
+{
+	size_t size = 0, alloc_size = 0;
+	void *bf = NULL, *nbf;
+	int fd, n, err = 0;
+	char sbuf[STRERR_BUFSIZE];
+
+	fd = open(filename, O_RDONLY);
+	if (fd < 0)
+		return -errno;
+
+	do {
+		if (size == alloc_size) {
+			alloc_size += BUFSIZ;
+			nbf = realloc(bf, alloc_size);
+			if (!nbf) {
+				err = -ENOMEM;
+				break;
+			}
+
+			bf = nbf;
+		}
+
+		n = read(fd, bf + size, alloc_size - size);
+		if (n < 0) {
+			if (size) {
+				pr_warning("read failed %d: %s\n", errno,
+					 strerror_r(errno, sbuf, sizeof(sbuf)));
+				err = 0;
+			} else
+				err = -errno;
+
+			break;
+		}
+
+		size += n;
+	} while (n > 0);
+
+	if (!err) {
+		*sizep = size;
+		*buf   = bf;
+	} else
+		free(bf);
+
+	close(fd);
+	return err;
+}
+
+const char *get_filename_for_perf_kvm(void)
+{
+	const char *filename;
+
+	if (perf_host && !perf_guest)
+		filename = strdup("perf.data.host");
+	else if (!perf_host && perf_guest)
+		filename = strdup("perf.data.guest");
+	else
+		filename = strdup("perf.data.kvm");
+
+	return filename;
+}
+
+int perf_event_paranoid(void)
+{
+	int value;
+
+	if (sysctl__read_int("kernel/perf_event_paranoid", &value))
+		return INT_MAX;
+
+	return value;
+}
+
+void mem_bswap_32(void *src, int byte_size)
+{
+	u32 *m = src;
+	while (byte_size > 0) {
+		*m = bswap_32(*m);
+		byte_size -= sizeof(u32);
+		++m;
+	}
+}
+
+void mem_bswap_64(void *src, int byte_size)
+{
+	u64 *m = src;
+
+	while (byte_size > 0) {
+		*m = bswap_64(*m);
+		byte_size -= sizeof(u64);
+		++m;
+	}
+}
+
+bool find_process(const char *name)
+{
+	size_t len = strlen(name);
+	DIR *dir;
+	struct dirent *d;
+	int ret = -1;
+
+	dir = opendir(procfs__mountpoint());
+	if (!dir)
+		return -1;
+
+	/* Walk through the directory. */
+	while (ret && (d = readdir(dir)) != NULL) {
+		char path[PATH_MAX];
+		char *data;
+		size_t size;
+
+		if ((d->d_type != DT_DIR) ||
+		     !strcmp(".", d->d_name) ||
+		     !strcmp("..", d->d_name))
+			continue;
+
+		scnprintf(path, sizeof(path), "%s/%s/comm",
+			  procfs__mountpoint(), d->d_name);
+
+		if (filename__read_str(path, &data, &size))
+			continue;
+
+		ret = strncmp(name, data, len);
+		free(data);
+	}
+
+	closedir(dir);
+	return ret ? false : true;
+}
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
new file mode 100644
index 000000000..1ff23e04a
--- /dev/null
+++ b/tools/perf/util/util.h
@@ -0,0 +1,336 @@
+#ifndef GIT_COMPAT_UTIL_H
+#define GIT_COMPAT_UTIL_H
+
+#ifndef FLEX_ARRAY
+/*
+ * See if our compiler is known to support flexible array members.
+ */
+#if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)
+# define FLEX_ARRAY /* empty */
+#elif defined(__GNUC__)
+# if (__GNUC__ >= 3)
+#  define FLEX_ARRAY /* empty */
+# else
+#  define FLEX_ARRAY 0 /* older GNU extension */
+# endif
+#endif
+
+/*
+ * Otherwise, default to safer but a bit wasteful traditional style
+ */
+#ifndef FLEX_ARRAY
+# define FLEX_ARRAY 1
+#endif
+#endif
+
+#define ARRAY_SIZE(x) (sizeof(x)/sizeof(x[0]))
+
+#ifdef __GNUC__
+#define TYPEOF(x) (__typeof__(x))
+#else
+#define TYPEOF(x)
+#endif
+
+#define MSB(x, bits) ((x) & TYPEOF(x)(~0ULL << (sizeof(x) * 8 - (bits))))
+#define HAS_MULTI_BITS(i)  ((i) & ((i) - 1))  /* checks if an integer has more than 1 bit set */
+
+/* Approximation of the length of the decimal representation of this type. */
+#define decimal_length(x)	((int)(sizeof(x) * 2.56 + 0.5) + 1)
+
+#define _ALL_SOURCE 1
+#define _BSD_SOURCE 1
+/* glibc 2.20 deprecates _BSD_SOURCE in favour of _DEFAULT_SOURCE */
+#define _DEFAULT_SOURCE 1
+#define HAS_BOOL
+
+#include <unistd.h>
+#include <stdio.h>
+#include <sys/stat.h>
+#include <sys/statfs.h>
+#include <fcntl.h>
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdlib.h>
+#include <stdarg.h>
+#include <string.h>
+#include <errno.h>
+#include <limits.h>
+#include <sys/param.h>
+#include <sys/types.h>
+#include <dirent.h>
+#include <sys/time.h>
+#include <time.h>
+#include <signal.h>
+#include <fnmatch.h>
+#include <assert.h>
+#include <regex.h>
+#include <utime.h>
+#include <sys/wait.h>
+#include <poll.h>
+#include <sys/socket.h>
+#include <sys/ioctl.h>
+#include <inttypes.h>
+#include <linux/kernel.h>
+#include <linux/magic.h>
+#include <linux/types.h>
+#include <sys/ttydefaults.h>
+#include <api/fs/debugfs.h>
+#include <api/fs/tracefs.h>
+#include <termios.h>
+#include <linux/bitops.h>
+#include <termios.h>
+
+extern const char *graph_line;
+extern const char *graph_dotted_line;
+extern char buildid_dir[];
+extern char tracing_events_path[];
+extern void perf_debugfs_set_path(const char *mountpoint);
+const char *perf_debugfs_mount(const char *mountpoint);
+const char *find_tracing_dir(void);
+char *get_tracing_file(const char *name);
+void put_tracing_file(char *file);
+
+/* On most systems <limits.h> would have given us this, but
+ * not on some systems (e.g. GNU/Hurd).
+ */
+#ifndef PATH_MAX
+#define PATH_MAX 4096
+#endif
+
+#ifndef PRIuMAX
+#define PRIuMAX "llu"
+#endif
+
+#ifndef PRIu32
+#define PRIu32 "u"
+#endif
+
+#ifndef PRIx32
+#define PRIx32 "x"
+#endif
+
+#ifndef PATH_SEP
+#define PATH_SEP ':'
+#endif
+
+#ifndef STRIP_EXTENSION
+#define STRIP_EXTENSION ""
+#endif
+
+#ifndef has_dos_drive_prefix
+#define has_dos_drive_prefix(path) 0
+#endif
+
+#ifndef is_dir_sep
+#define is_dir_sep(c) ((c) == '/')
+#endif
+
+#ifdef __GNUC__
+#define NORETURN __attribute__((__noreturn__))
+#else
+#define NORETURN
+#ifndef __attribute__
+#define __attribute__(x)
+#endif
+#endif
+
+#define PERF_GTK_DSO  "libperf-gtk.so"
+
+/* General helper functions */
+extern void usage(const char *err) NORETURN;
+extern void die(const char *err, ...) NORETURN __attribute__((format (printf, 1, 2)));
+extern int error(const char *err, ...) __attribute__((format (printf, 1, 2)));
+extern void warning(const char *err, ...) __attribute__((format (printf, 1, 2)));
+
+#include "../../../include/linux/stringify.h"
+
+#define DIE_IF(cnd)	\
+	do { if (cnd)	\
+		die(" at (" __FILE__ ":" __stringify(__LINE__) "): "	\
+		    __stringify(cnd) "\n");				\
+	} while (0)
+
+
+extern void set_die_routine(void (*routine)(const char *err, va_list params) NORETURN);
+
+extern int prefixcmp(const char *str, const char *prefix);
+extern void set_buildid_dir(const char *dir);
+
+static inline const char *skip_prefix(const char *str, const char *prefix)
+{
+	size_t len = strlen(prefix);
+	return strncmp(str, prefix, len) ? NULL : str + len;
+}
+
+#ifdef __GLIBC_PREREQ
+#if __GLIBC_PREREQ(2, 1)
+#define HAVE_STRCHRNUL
+#endif
+#endif
+
+#ifndef HAVE_STRCHRNUL
+#define strchrnul gitstrchrnul
+static inline char *gitstrchrnul(const char *s, int c)
+{
+	while (*s && *s != c)
+		s++;
+	return (char *)s;
+}
+#endif
+
+/*
+ * Wrappers:
+ */
+extern char *xstrdup(const char *str);
+extern void *xrealloc(void *ptr, size_t size) __attribute__((weak));
+
+
+static inline void *zalloc(size_t size)
+{
+	return calloc(1, size);
+}
+
+#define zfree(ptr) ({ free(*ptr); *ptr = NULL; })
+
+static inline int has_extension(const char *filename, const char *ext)
+{
+	size_t len = strlen(filename);
+	size_t extlen = strlen(ext);
+
+	return len > extlen && !memcmp(filename + len - extlen, ext, extlen);
+}
+
+/* Sane ctype - no locale, and works with signed chars */
+#undef isascii
+#undef isspace
+#undef isdigit
+#undef isxdigit
+#undef isalpha
+#undef isprint
+#undef isalnum
+#undef islower
+#undef isupper
+#undef tolower
+#undef toupper
+
+#ifndef NSEC_PER_MSEC
+#define NSEC_PER_MSEC	1000000L
+#endif
+
+int parse_nsec_time(const char *str, u64 *ptime);
+
+extern unsigned char sane_ctype[256];
+#define GIT_SPACE		0x01
+#define GIT_DIGIT		0x02
+#define GIT_ALPHA		0x04
+#define GIT_GLOB_SPECIAL	0x08
+#define GIT_REGEX_SPECIAL	0x10
+#define GIT_PRINT_EXTRA		0x20
+#define GIT_PRINT		0x3E
+#define sane_istest(x,mask) ((sane_ctype[(unsigned char)(x)] & (mask)) != 0)
+#define isascii(x) (((x) & ~0x7f) == 0)
+#define isspace(x) sane_istest(x,GIT_SPACE)
+#define isdigit(x) sane_istest(x,GIT_DIGIT)
+#define isxdigit(x)	\
+	(sane_istest(toupper(x), GIT_ALPHA | GIT_DIGIT) && toupper(x) < 'G')
+#define isalpha(x) sane_istest(x,GIT_ALPHA)
+#define isalnum(x) sane_istest(x,GIT_ALPHA | GIT_DIGIT)
+#define isprint(x) sane_istest(x,GIT_PRINT)
+#define islower(x) (sane_istest(x,GIT_ALPHA) && (x & 0x20))
+#define isupper(x) (sane_istest(x,GIT_ALPHA) && !(x & 0x20))
+#define tolower(x) sane_case((unsigned char)(x), 0x20)
+#define toupper(x) sane_case((unsigned char)(x), 0)
+
+static inline int sane_case(int x, int high)
+{
+	if (sane_istest(x, GIT_ALPHA))
+		x = (x & ~0x20) | high;
+	return x;
+}
+
+int mkdir_p(char *path, mode_t mode);
+int copyfile(const char *from, const char *to);
+int copyfile_mode(const char *from, const char *to, mode_t mode);
+
+s64 perf_atoll(const char *str);
+char **argv_split(const char *str, int *argcp);
+void argv_free(char **argv);
+bool strglobmatch(const char *str, const char *pat);
+bool strlazymatch(const char *str, const char *pat);
+int strtailcmp(const char *s1, const char *s2);
+char *strxfrchar(char *s, char from, char to);
+unsigned long convert_unit(unsigned long value, char *unit);
+ssize_t readn(int fd, void *buf, size_t n);
+ssize_t writen(int fd, void *buf, size_t n);
+
+struct perf_event_attr;
+
+void event_attr_init(struct perf_event_attr *attr);
+
+#define _STR(x) #x
+#define STR(x) _STR(x)
+
+size_t hex_width(u64 v);
+int hex2u64(const char *ptr, u64 *val);
+
+char *ltrim(char *s);
+char *rtrim(char *s);
+
+void dump_stack(void);
+void sighandler_dump_stack(int sig);
+
+extern unsigned int page_size;
+extern int cacheline_size;
+
+void get_term_dimensions(struct winsize *ws);
+void set_term_quiet_input(struct termios *old);
+
+struct parse_tag {
+	char tag;
+	int mult;
+};
+
+unsigned long parse_tag_value(const char *str, struct parse_tag *tags);
+
+#define SRCLINE_UNKNOWN  ((char *) "??:0")
+
+static inline int path__join(char *bf, size_t size,
+			     const char *path1, const char *path2)
+{
+	return scnprintf(bf, size, "%s%s%s", path1, path1[0] ? "/" : "", path2);
+}
+
+static inline int path__join3(char *bf, size_t size,
+			      const char *path1, const char *path2,
+			      const char *path3)
+{
+	return scnprintf(bf, size, "%s%s%s%s%s",
+			 path1, path1[0] ? "/" : "",
+			 path2, path2[0] ? "/" : "", path3);
+}
+
+struct dso;
+struct symbol;
+
+char *get_srcline(struct dso *dso, u64 addr, struct symbol *sym,
+		  bool show_sym);
+void free_srcline(char *srcline);
+
+int filename__read_str(const char *filename, char **buf, size_t *sizep);
+int perf_event_paranoid(void);
+
+void mem_bswap_64(void *src, int byte_size);
+void mem_bswap_32(void *src, int byte_size);
+
+const char *get_filename_for_perf_kvm(void);
+bool find_process(const char *name);
+
+#ifdef HAVE_ZLIB_SUPPORT
+int gzip_decompress_to_file(const char *input, int output_fd);
+#endif
+
+#ifdef HAVE_LZMA_SUPPORT
+int lzma_decompress_to_file(const char *input, int output_fd);
+#endif
+
+#endif /* GIT_COMPAT_UTIL_H */
diff --git a/tools/perf/util/values.c b/tools/perf/util/values.c
new file mode 100644
index 000000000..0fb3c1fcd
--- /dev/null
+++ b/tools/perf/util/values.c
@@ -0,0 +1,232 @@
+#include <stdlib.h>
+
+#include "util.h"
+#include "values.h"
+
+void perf_read_values_init(struct perf_read_values *values)
+{
+	values->threads_max = 16;
+	values->pid = malloc(values->threads_max * sizeof(*values->pid));
+	values->tid = malloc(values->threads_max * sizeof(*values->tid));
+	values->value = malloc(values->threads_max * sizeof(*values->value));
+	if (!values->pid || !values->tid || !values->value)
+		die("failed to allocate read_values threads arrays");
+	values->threads = 0;
+
+	values->counters_max = 16;
+	values->counterrawid = malloc(values->counters_max
+				      * sizeof(*values->counterrawid));
+	values->countername = malloc(values->counters_max
+				     * sizeof(*values->countername));
+	if (!values->counterrawid || !values->countername)
+		die("failed to allocate read_values counters arrays");
+	values->counters = 0;
+}
+
+void perf_read_values_destroy(struct perf_read_values *values)
+{
+	int i;
+
+	if (!values->threads_max || !values->counters_max)
+		return;
+
+	for (i = 0; i < values->threads; i++)
+		zfree(&values->value[i]);
+	zfree(&values->value);
+	zfree(&values->pid);
+	zfree(&values->tid);
+	zfree(&values->counterrawid);
+	for (i = 0; i < values->counters; i++)
+		zfree(&values->countername[i]);
+	zfree(&values->countername);
+}
+
+static void perf_read_values__enlarge_threads(struct perf_read_values *values)
+{
+	values->threads_max *= 2;
+	values->pid = realloc(values->pid,
+			      values->threads_max * sizeof(*values->pid));
+	values->tid = realloc(values->tid,
+			      values->threads_max * sizeof(*values->tid));
+	values->value = realloc(values->value,
+				values->threads_max * sizeof(*values->value));
+	if (!values->pid || !values->tid || !values->value)
+		die("failed to enlarge read_values threads arrays");
+}
+
+static int perf_read_values__findnew_thread(struct perf_read_values *values,
+					    u32 pid, u32 tid)
+{
+	int i;
+
+	for (i = 0; i < values->threads; i++)
+		if (values->pid[i] == pid && values->tid[i] == tid)
+			return i;
+
+	if (values->threads == values->threads_max)
+		perf_read_values__enlarge_threads(values);
+
+	i = values->threads++;
+	values->pid[i] = pid;
+	values->tid[i] = tid;
+	values->value[i] = malloc(values->counters_max * sizeof(**values->value));
+	if (!values->value[i])
+		die("failed to allocate read_values counters array");
+
+	return i;
+}
+
+static void perf_read_values__enlarge_counters(struct perf_read_values *values)
+{
+	int i;
+
+	values->counters_max *= 2;
+	values->counterrawid = realloc(values->counterrawid,
+				       values->counters_max * sizeof(*values->counterrawid));
+	values->countername = realloc(values->countername,
+				      values->counters_max * sizeof(*values->countername));
+	if (!values->counterrawid || !values->countername)
+		die("failed to enlarge read_values counters arrays");
+
+	for (i = 0; i < values->threads; i++) {
+		values->value[i] = realloc(values->value[i],
+					   values->counters_max * sizeof(**values->value));
+		if (!values->value[i])
+			die("failed to enlarge read_values counters arrays");
+	}
+}
+
+static int perf_read_values__findnew_counter(struct perf_read_values *values,
+					     u64 rawid, const char *name)
+{
+	int i;
+
+	for (i = 0; i < values->counters; i++)
+		if (values->counterrawid[i] == rawid)
+			return i;
+
+	if (values->counters == values->counters_max)
+		perf_read_values__enlarge_counters(values);
+
+	i = values->counters++;
+	values->counterrawid[i] = rawid;
+	values->countername[i] = strdup(name);
+
+	return i;
+}
+
+void perf_read_values_add_value(struct perf_read_values *values,
+				u32 pid, u32 tid,
+				u64 rawid, const char *name, u64 value)
+{
+	int tindex, cindex;
+
+	tindex = perf_read_values__findnew_thread(values, pid, tid);
+	cindex = perf_read_values__findnew_counter(values, rawid, name);
+
+	values->value[tindex][cindex] = value;
+}
+
+static void perf_read_values__display_pretty(FILE *fp,
+					     struct perf_read_values *values)
+{
+	int i, j;
+	int pidwidth, tidwidth;
+	int *counterwidth;
+
+	counterwidth = malloc(values->counters * sizeof(*counterwidth));
+	if (!counterwidth)
+		die("failed to allocate counterwidth array");
+	tidwidth = 3;
+	pidwidth = 3;
+	for (j = 0; j < values->counters; j++)
+		counterwidth[j] = strlen(values->countername[j]);
+	for (i = 0; i < values->threads; i++) {
+		int width;
+
+		width = snprintf(NULL, 0, "%d", values->pid[i]);
+		if (width > pidwidth)
+			pidwidth = width;
+		width = snprintf(NULL, 0, "%d", values->tid[i]);
+		if (width > tidwidth)
+			tidwidth = width;
+		for (j = 0; j < values->counters; j++) {
+			width = snprintf(NULL, 0, "%" PRIu64, values->value[i][j]);
+			if (width > counterwidth[j])
+				counterwidth[j] = width;
+		}
+	}
+
+	fprintf(fp, "# %*s  %*s", pidwidth, "PID", tidwidth, "TID");
+	for (j = 0; j < values->counters; j++)
+		fprintf(fp, "  %*s", counterwidth[j], values->countername[j]);
+	fprintf(fp, "\n");
+
+	for (i = 0; i < values->threads; i++) {
+		fprintf(fp, "  %*d  %*d", pidwidth, values->pid[i],
+			tidwidth, values->tid[i]);
+		for (j = 0; j < values->counters; j++)
+			fprintf(fp, "  %*" PRIu64,
+				counterwidth[j], values->value[i][j]);
+		fprintf(fp, "\n");
+	}
+	free(counterwidth);
+}
+
+static void perf_read_values__display_raw(FILE *fp,
+					  struct perf_read_values *values)
+{
+	int width, pidwidth, tidwidth, namewidth, rawwidth, countwidth;
+	int i, j;
+
+	tidwidth = 3; /* TID */
+	pidwidth = 3; /* PID */
+	namewidth = 4; /* "Name" */
+	rawwidth = 3; /* "Raw" */
+	countwidth = 5; /* "Count" */
+
+	for (i = 0; i < values->threads; i++) {
+		width = snprintf(NULL, 0, "%d", values->pid[i]);
+		if (width > pidwidth)
+			pidwidth = width;
+		width = snprintf(NULL, 0, "%d", values->tid[i]);
+		if (width > tidwidth)
+			tidwidth = width;
+	}
+	for (j = 0; j < values->counters; j++) {
+		width = strlen(values->countername[j]);
+		if (width > namewidth)
+			namewidth = width;
+		width = snprintf(NULL, 0, "%" PRIx64, values->counterrawid[j]);
+		if (width > rawwidth)
+			rawwidth = width;
+	}
+	for (i = 0; i < values->threads; i++) {
+		for (j = 0; j < values->counters; j++) {
+			width = snprintf(NULL, 0, "%" PRIu64, values->value[i][j]);
+			if (width > countwidth)
+				countwidth = width;
+		}
+	}
+
+	fprintf(fp, "# %*s  %*s  %*s  %*s  %*s\n",
+		pidwidth, "PID", tidwidth, "TID",
+		namewidth, "Name", rawwidth, "Raw",
+		countwidth, "Count");
+	for (i = 0; i < values->threads; i++)
+		for (j = 0; j < values->counters; j++)
+			fprintf(fp, "  %*d  %*d  %*s  %*" PRIx64 "  %*" PRIu64,
+				pidwidth, values->pid[i],
+				tidwidth, values->tid[i],
+				namewidth, values->countername[j],
+				rawwidth, values->counterrawid[j],
+				countwidth, values->value[i][j]);
+}
+
+void perf_read_values_display(FILE *fp, struct perf_read_values *values, int raw)
+{
+	if (raw)
+		perf_read_values__display_raw(fp, values);
+	else
+		perf_read_values__display_pretty(fp, values);
+}
diff --git a/tools/perf/util/values.h b/tools/perf/util/values.h
new file mode 100644
index 000000000..b21a80c6c
--- /dev/null
+++ b/tools/perf/util/values.h
@@ -0,0 +1,27 @@
+#ifndef __PERF_VALUES_H
+#define __PERF_VALUES_H
+
+#include <linux/types.h>
+
+struct perf_read_values {
+	int threads;
+	int threads_max;
+	u32 *pid, *tid;
+	int counters;
+	int counters_max;
+	u64 *counterrawid;
+	char **countername;
+	u64 **value;
+};
+
+void perf_read_values_init(struct perf_read_values *values);
+void perf_read_values_destroy(struct perf_read_values *values);
+
+void perf_read_values_add_value(struct perf_read_values *values,
+				u32 pid, u32 tid,
+				u64 rawid, const char *name, u64 value);
+
+void perf_read_values_display(FILE *fp, struct perf_read_values *values,
+			      int raw);
+
+#endif /* __PERF_VALUES_H */
diff --git a/tools/perf/util/vdso.c b/tools/perf/util/vdso.c
new file mode 100644
index 000000000..5c7dd7969
--- /dev/null
+++ b/tools/perf/util/vdso.c
@@ -0,0 +1,321 @@
+
+#include <unistd.h>
+#include <stdio.h>
+#include <string.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <stdlib.h>
+#include <linux/kernel.h>
+
+#include "vdso.h"
+#include "util.h"
+#include "symbol.h"
+#include "machine.h"
+#include "thread.h"
+#include "linux/string.h"
+#include "debug.h"
+
+/*
+ * Include definition of find_vdso_map() also used in perf-read-vdso.c for
+ * building perf-read-vdso32 and perf-read-vdsox32.
+ */
+#include "find-vdso-map.c"
+
+#define VDSO__TEMP_FILE_NAME "/tmp/perf-vdso.so-XXXXXX"
+
+struct vdso_file {
+	bool found;
+	bool error;
+	char temp_file_name[sizeof(VDSO__TEMP_FILE_NAME)];
+	const char *dso_name;
+	const char *read_prog;
+};
+
+struct vdso_info {
+	struct vdso_file vdso;
+#if BITS_PER_LONG == 64
+	struct vdso_file vdso32;
+	struct vdso_file vdsox32;
+#endif
+};
+
+static struct vdso_info *vdso_info__new(void)
+{
+	static const struct vdso_info vdso_info_init = {
+		.vdso    = {
+			.temp_file_name = VDSO__TEMP_FILE_NAME,
+			.dso_name = DSO__NAME_VDSO,
+		},
+#if BITS_PER_LONG == 64
+		.vdso32  = {
+			.temp_file_name = VDSO__TEMP_FILE_NAME,
+			.dso_name = DSO__NAME_VDSO32,
+			.read_prog = "perf-read-vdso32",
+		},
+		.vdsox32  = {
+			.temp_file_name = VDSO__TEMP_FILE_NAME,
+			.dso_name = DSO__NAME_VDSOX32,
+			.read_prog = "perf-read-vdsox32",
+		},
+#endif
+	};
+
+	return memdup(&vdso_info_init, sizeof(vdso_info_init));
+}
+
+static char *get_file(struct vdso_file *vdso_file)
+{
+	char *vdso = NULL;
+	char *buf = NULL;
+	void *start, *end;
+	size_t size;
+	int fd;
+
+	if (vdso_file->found)
+		return vdso_file->temp_file_name;
+
+	if (vdso_file->error || find_vdso_map(&start, &end))
+		return NULL;
+
+	size = end - start;
+
+	buf = memdup(start, size);
+	if (!buf)
+		return NULL;
+
+	fd = mkstemp(vdso_file->temp_file_name);
+	if (fd < 0)
+		goto out;
+
+	if (size == (size_t) write(fd, buf, size))
+		vdso = vdso_file->temp_file_name;
+
+	close(fd);
+
+ out:
+	free(buf);
+
+	vdso_file->found = (vdso != NULL);
+	vdso_file->error = !vdso_file->found;
+	return vdso;
+}
+
+void vdso__exit(struct machine *machine)
+{
+	struct vdso_info *vdso_info = machine->vdso_info;
+
+	if (!vdso_info)
+		return;
+
+	if (vdso_info->vdso.found)
+		unlink(vdso_info->vdso.temp_file_name);
+#if BITS_PER_LONG == 64
+	if (vdso_info->vdso32.found)
+		unlink(vdso_info->vdso32.temp_file_name);
+	if (vdso_info->vdsox32.found)
+		unlink(vdso_info->vdsox32.temp_file_name);
+#endif
+
+	zfree(&machine->vdso_info);
+}
+
+static struct dso *vdso__new(struct machine *machine, const char *short_name,
+			     const char *long_name)
+{
+	struct dso *dso;
+
+	dso = dso__new(short_name);
+	if (dso != NULL) {
+		dsos__add(&machine->user_dsos, dso);
+		dso__set_long_name(dso, long_name, false);
+	}
+
+	return dso;
+}
+
+#if BITS_PER_LONG == 64
+
+static enum dso_type machine__thread_dso_type(struct machine *machine,
+					      struct thread *thread)
+{
+	enum dso_type dso_type = DSO__TYPE_UNKNOWN;
+	struct map *map;
+	struct dso *dso;
+
+	map = map_groups__first(thread->mg, MAP__FUNCTION);
+	for (; map ; map = map_groups__next(map)) {
+		dso = map->dso;
+		if (!dso || dso->long_name[0] != '/')
+			continue;
+		dso_type = dso__type(dso, machine);
+		if (dso_type != DSO__TYPE_UNKNOWN)
+			break;
+	}
+
+	return dso_type;
+}
+
+static int vdso__do_copy_compat(FILE *f, int fd)
+{
+	char buf[4096];
+	size_t count;
+
+	while (1) {
+		count = fread(buf, 1, sizeof(buf), f);
+		if (ferror(f))
+			return -errno;
+		if (feof(f))
+			break;
+		if (count && writen(fd, buf, count) != (ssize_t)count)
+			return -errno;
+	}
+
+	return 0;
+}
+
+static int vdso__copy_compat(const char *prog, int fd)
+{
+	FILE *f;
+	int err;
+
+	f = popen(prog, "r");
+	if (!f)
+		return -errno;
+
+	err = vdso__do_copy_compat(f, fd);
+
+	if (pclose(f) == -1)
+		return -errno;
+
+	return err;
+}
+
+static int vdso__create_compat_file(const char *prog, char *temp_name)
+{
+	int fd, err;
+
+	fd = mkstemp(temp_name);
+	if (fd < 0)
+		return -errno;
+
+	err = vdso__copy_compat(prog, fd);
+
+	if (close(fd) == -1)
+		return -errno;
+
+	return err;
+}
+
+static const char *vdso__get_compat_file(struct vdso_file *vdso_file)
+{
+	int err;
+
+	if (vdso_file->found)
+		return vdso_file->temp_file_name;
+
+	if (vdso_file->error)
+		return NULL;
+
+	err = vdso__create_compat_file(vdso_file->read_prog,
+				       vdso_file->temp_file_name);
+	if (err) {
+		pr_err("%s failed, error %d\n", vdso_file->read_prog, err);
+		vdso_file->error = true;
+		return NULL;
+	}
+
+	vdso_file->found = true;
+
+	return vdso_file->temp_file_name;
+}
+
+static struct dso *vdso__findnew_compat(struct machine *machine,
+					struct vdso_file *vdso_file)
+{
+	const char *file_name;
+	struct dso *dso;
+
+	dso = dsos__find(&machine->user_dsos, vdso_file->dso_name, true);
+	if (dso)
+		return dso;
+
+	file_name = vdso__get_compat_file(vdso_file);
+	if (!file_name)
+		return NULL;
+
+	return vdso__new(machine, vdso_file->dso_name, file_name);
+}
+
+static int vdso__dso_findnew_compat(struct machine *machine,
+				    struct thread *thread,
+				    struct vdso_info *vdso_info,
+				    struct dso **dso)
+{
+	enum dso_type dso_type;
+
+	dso_type = machine__thread_dso_type(machine, thread);
+
+#ifndef HAVE_PERF_READ_VDSO32
+	if (dso_type == DSO__TYPE_32BIT)
+		return 0;
+#endif
+#ifndef HAVE_PERF_READ_VDSOX32
+	if (dso_type == DSO__TYPE_X32BIT)
+		return 0;
+#endif
+
+	switch (dso_type) {
+	case DSO__TYPE_32BIT:
+		*dso = vdso__findnew_compat(machine, &vdso_info->vdso32);
+		return 1;
+	case DSO__TYPE_X32BIT:
+		*dso = vdso__findnew_compat(machine, &vdso_info->vdsox32);
+		return 1;
+	case DSO__TYPE_UNKNOWN:
+	case DSO__TYPE_64BIT:
+	default:
+		return 0;
+	}
+}
+
+#endif
+
+struct dso *vdso__dso_findnew(struct machine *machine,
+			      struct thread *thread __maybe_unused)
+{
+	struct vdso_info *vdso_info;
+	struct dso *dso;
+
+	if (!machine->vdso_info)
+		machine->vdso_info = vdso_info__new();
+
+	vdso_info = machine->vdso_info;
+	if (!vdso_info)
+		return NULL;
+
+#if BITS_PER_LONG == 64
+	if (vdso__dso_findnew_compat(machine, thread, vdso_info, &dso))
+		return dso;
+#endif
+
+	dso = dsos__find(&machine->user_dsos, DSO__NAME_VDSO, true);
+	if (!dso) {
+		char *file;
+
+		file = get_file(&vdso_info->vdso);
+		if (!file)
+			return NULL;
+
+		dso = vdso__new(machine, DSO__NAME_VDSO, file);
+	}
+
+	return dso;
+}
+
+bool dso__is_vdso(struct dso *dso)
+{
+	return !strcmp(dso->short_name, DSO__NAME_VDSO) ||
+	       !strcmp(dso->short_name, DSO__NAME_VDSO32) ||
+	       !strcmp(dso->short_name, DSO__NAME_VDSOX32);
+}
diff --git a/tools/perf/util/vdso.h b/tools/perf/util/vdso.h
new file mode 100644
index 000000000..d97da1616
--- /dev/null
+++ b/tools/perf/util/vdso.h
@@ -0,0 +1,29 @@
+#ifndef __PERF_VDSO__
+#define __PERF_VDSO__
+
+#include <linux/types.h>
+#include <string.h>
+#include <stdbool.h>
+
+#define VDSO__MAP_NAME "[vdso]"
+
+#define DSO__NAME_VDSO    "[vdso]"
+#define DSO__NAME_VDSO32  "[vdso32]"
+#define DSO__NAME_VDSOX32 "[vdsox32]"
+
+static inline bool is_vdso_map(const char *filename)
+{
+	return !strcmp(filename, VDSO__MAP_NAME);
+}
+
+struct dso;
+
+bool dso__is_vdso(struct dso *dso);
+
+struct machine;
+struct thread;
+
+struct dso *vdso__dso_findnew(struct machine *machine, struct thread *thread);
+void vdso__exit(struct machine *machine);
+
+#endif /* __PERF_VDSO__ */
diff --git a/tools/perf/util/wrapper.c b/tools/perf/util/wrapper.c
new file mode 100644
index 000000000..19f15b650
--- /dev/null
+++ b/tools/perf/util/wrapper.c
@@ -0,0 +1,41 @@
+/*
+ * Various trivial helper wrappers around standard functions
+ */
+#include "cache.h"
+
+/*
+ * There's no pack memory to release - but stay close to the Git
+ * version so wrap this away:
+ */
+static inline void release_pack_memory(size_t size __maybe_unused,
+				       int flag __maybe_unused)
+{
+}
+
+char *xstrdup(const char *str)
+{
+	char *ret = strdup(str);
+	if (!ret) {
+		release_pack_memory(strlen(str) + 1, -1);
+		ret = strdup(str);
+		if (!ret)
+			die("Out of memory, strdup failed");
+	}
+	return ret;
+}
+
+void *xrealloc(void *ptr, size_t size)
+{
+	void *ret = realloc(ptr, size);
+	if (!ret && !size)
+		ret = realloc(ptr, 1);
+	if (!ret) {
+		release_pack_memory(size, -1);
+		ret = realloc(ptr, size);
+		if (!ret && !size)
+			ret = realloc(ptr, 1);
+		if (!ret)
+			die("Out of memory, realloc failed");
+	}
+	return ret;
+}
diff --git a/tools/perf/util/xyarray.c b/tools/perf/util/xyarray.c
new file mode 100644
index 000000000..22afbf6c5
--- /dev/null
+++ b/tools/perf/util/xyarray.c
@@ -0,0 +1,20 @@
+#include "xyarray.h"
+#include "util.h"
+
+struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size)
+{
+	size_t row_size = ylen * entry_size;
+	struct xyarray *xy = zalloc(sizeof(*xy) + xlen * row_size);
+
+	if (xy != NULL) {
+		xy->entry_size = entry_size;
+		xy->row_size   = row_size;
+	}
+
+	return xy;
+}
+
+void xyarray__delete(struct xyarray *xy)
+{
+	free(xy);
+}
diff --git a/tools/perf/util/xyarray.h b/tools/perf/util/xyarray.h
new file mode 100644
index 000000000..c488a0727
--- /dev/null
+++ b/tools/perf/util/xyarray.h
@@ -0,0 +1,20 @@
+#ifndef _PERF_XYARRAY_H_
+#define _PERF_XYARRAY_H_ 1
+
+#include <sys/types.h>
+
+struct xyarray {
+	size_t row_size;
+	size_t entry_size;
+	char contents[];
+};
+
+struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size);
+void xyarray__delete(struct xyarray *xy);
+
+static inline void *xyarray__entry(struct xyarray *xy, int x, int y)
+{
+	return &xy->contents[x * xy->row_size + y * xy->entry_size];
+}
+
+#endif /* _PERF_XYARRAY_H_ */
diff --git a/tools/perf/util/zlib.c b/tools/perf/util/zlib.c
new file mode 100644
index 000000000..495a449fc
--- /dev/null
+++ b/tools/perf/util/zlib.c
@@ -0,0 +1,78 @@
+#include <stdio.h>
+#include <unistd.h>
+#include <sys/stat.h>
+#include <sys/mman.h>
+#include <zlib.h>
+
+#include "util/util.h"
+#include "util/debug.h"
+
+
+#define CHUNK_SIZE  16384
+
+int gzip_decompress_to_file(const char *input, int output_fd)
+{
+	int ret = Z_STREAM_ERROR;
+	int input_fd;
+	void *ptr;
+	int len;
+	struct stat stbuf;
+	unsigned char buf[CHUNK_SIZE];
+	z_stream zs = {
+		.zalloc		= Z_NULL,
+		.zfree		= Z_NULL,
+		.opaque		= Z_NULL,
+		.avail_in	= 0,
+		.next_in	= Z_NULL,
+	};
+
+	input_fd = open(input, O_RDONLY);
+	if (input_fd < 0)
+		return -1;
+
+	if (fstat(input_fd, &stbuf) < 0)
+		goto out_close;
+
+	ptr = mmap(NULL, stbuf.st_size, PROT_READ, MAP_PRIVATE, input_fd, 0);
+	if (ptr == MAP_FAILED)
+		goto out_close;
+
+	if (inflateInit2(&zs, 16 + MAX_WBITS) != Z_OK)
+		goto out_unmap;
+
+	zs.next_in = ptr;
+	zs.avail_in = stbuf.st_size;
+
+	do {
+		zs.next_out = buf;
+		zs.avail_out = CHUNK_SIZE;
+
+		ret = inflate(&zs, Z_NO_FLUSH);
+		switch (ret) {
+		case Z_NEED_DICT:
+			ret = Z_DATA_ERROR;
+			/* fall through */
+		case Z_DATA_ERROR:
+		case Z_MEM_ERROR:
+			goto out;
+		default:
+			break;
+		}
+
+		len = CHUNK_SIZE - zs.avail_out;
+		if (writen(output_fd, buf, len) != len) {
+			ret = Z_DATA_ERROR;
+			goto out;
+		}
+
+	} while (ret != Z_STREAM_END);
+
+out:
+	inflateEnd(&zs);
+out_unmap:
+	munmap(ptr, stbuf.st_size);
+out_close:
+	close(input_fd);
+
+	return ret == Z_STREAM_END ? 0 : -1;
+}