diff options
| author | André Fabian Silva Delgado <emulatorman@parabola.nu> | 2015-08-05 17:04:01 -0300 |
|---|---|---|
| committer | André Fabian Silva Delgado <emulatorman@parabola.nu> | 2015-08-05 17:04:01 -0300 |
| commit | 57f0f512b273f60d52568b8c6b77e17f5636edc0 (patch) | |
| tree | 5e910f0e82173f4ef4f51111366a3f1299037a7b /arch/s390 | |
Initial import
Diffstat (limited to 'arch/s390')
370 files changed, 82458 insertions, 0 deletions
diff --git a/arch/s390/Kbuild b/arch/s390/Kbuild new file mode 100644 index 000000000..2938934c6 --- /dev/null +++ b/arch/s390/Kbuild @@ -0,0 +1,8 @@ +obj-y += kernel/ +obj-y += mm/ +obj-$(CONFIG_KVM) += kvm/ +obj-$(CONFIG_CRYPTO_HW) += crypto/ +obj-$(CONFIG_S390_HYPFS_FS) += hypfs/ +obj-$(CONFIG_APPLDATA_BASE) += appldata/ +obj-y += net/ +obj-$(CONFIG_PCI) += pci/ diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig new file mode 100644 index 000000000..b06dc3839 --- /dev/null +++ b/arch/s390/Kconfig @@ -0,0 +1,798 @@ +config MMU + def_bool y + +config ZONE_DMA + def_bool y + +config LOCKDEP_SUPPORT + def_bool y + +config STACKTRACE_SUPPORT + def_bool y + +config HAVE_LATENCYTOP_SUPPORT + def_bool y + +config RWSEM_GENERIC_SPINLOCK + bool + +config RWSEM_XCHGADD_ALGORITHM + def_bool y + +config ARCH_HAS_ILOG2_U32 + def_bool n + +config ARCH_HAS_ILOG2_U64 + def_bool n + +config GENERIC_HWEIGHT + def_bool y + +config GENERIC_BUG + def_bool y if BUG + +config GENERIC_BUG_RELATIVE_POINTERS + def_bool y + +config ARCH_DMA_ADDR_T_64BIT + def_bool y + +config GENERIC_LOCKBREAK + def_bool y if SMP && PREEMPT + +config PGSTE + def_bool y if KVM + +config ARCH_SUPPORTS_DEBUG_PAGEALLOC + def_bool y + +config KEXEC + def_bool y + +config AUDIT_ARCH + def_bool y + +config NO_IOPORT_MAP + def_bool y + +config PCI_QUIRKS + def_bool n + +config ARCH_SUPPORTS_UPROBES + def_bool y + +config S390 + def_bool y + select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE + select ARCH_HAS_DEBUG_STRICT_USER_COPY_CHECKS + select ARCH_HAS_ELF_RANDOMIZE + select ARCH_HAS_GCOV_PROFILE_ALL + select ARCH_HAS_SG_CHAIN + select ARCH_HAVE_NMI_SAFE_CMPXCHG + select ARCH_INLINE_READ_LOCK + select ARCH_INLINE_READ_LOCK_BH + select ARCH_INLINE_READ_LOCK_IRQ + select ARCH_INLINE_READ_LOCK_IRQSAVE + select ARCH_INLINE_READ_TRYLOCK + select ARCH_INLINE_READ_UNLOCK + select ARCH_INLINE_READ_UNLOCK_BH + select ARCH_INLINE_READ_UNLOCK_IRQ + select ARCH_INLINE_READ_UNLOCK_IRQRESTORE + select ARCH_INLINE_SPIN_LOCK + select ARCH_INLINE_SPIN_LOCK_BH + select ARCH_INLINE_SPIN_LOCK_IRQ + select ARCH_INLINE_SPIN_LOCK_IRQSAVE + select ARCH_INLINE_SPIN_TRYLOCK + select ARCH_INLINE_SPIN_TRYLOCK_BH + select ARCH_INLINE_SPIN_UNLOCK + select ARCH_INLINE_SPIN_UNLOCK_BH + select ARCH_INLINE_SPIN_UNLOCK_IRQ + select ARCH_INLINE_SPIN_UNLOCK_IRQRESTORE + select ARCH_INLINE_WRITE_LOCK + select ARCH_INLINE_WRITE_LOCK_BH + select ARCH_INLINE_WRITE_LOCK_IRQ + select ARCH_INLINE_WRITE_LOCK_IRQSAVE + select ARCH_INLINE_WRITE_TRYLOCK + select ARCH_INLINE_WRITE_UNLOCK + select ARCH_INLINE_WRITE_UNLOCK_BH + select ARCH_INLINE_WRITE_UNLOCK_IRQ + select ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE + select ARCH_SAVE_PAGE_KEYS if HIBERNATION + select ARCH_SUPPORTS_ATOMIC_RMW + select ARCH_USE_CMPXCHG_LOCKREF + select ARCH_WANT_IPC_PARSE_VERSION + select BUILDTIME_EXTABLE_SORT + select CLONE_BACKWARDS2 + select DYNAMIC_FTRACE if FUNCTION_TRACER + select GENERIC_CLOCKEVENTS + select GENERIC_CPU_DEVICES if !SMP + select GENERIC_FIND_FIRST_BIT + select GENERIC_SMP_IDLE_THREAD + select GENERIC_TIME_VSYSCALL + select HAVE_ALIGNED_STRUCT_PAGE if SLUB + select HAVE_ARCH_AUDITSYSCALL + select HAVE_ARCH_JUMP_LABEL + select HAVE_ARCH_SECCOMP_FILTER + select HAVE_ARCH_TRACEHOOK + select HAVE_ARCH_TRANSPARENT_HUGEPAGE + select HAVE_BPF_JIT if PACK_STACK && HAVE_MARCH_Z196_FEATURES + select HAVE_CMPXCHG_DOUBLE + select HAVE_CMPXCHG_LOCAL + select HAVE_DEBUG_KMEMLEAK + select HAVE_DYNAMIC_FTRACE + select HAVE_DYNAMIC_FTRACE_WITH_REGS + select HAVE_FTRACE_MCOUNT_RECORD + select HAVE_FUNCTION_GRAPH_TRACER + select HAVE_FUNCTION_TRACER + select HAVE_FUTEX_CMPXCHG if FUTEX + select HAVE_KERNEL_BZIP2 + select HAVE_KERNEL_GZIP + select HAVE_KERNEL_LZ4 + select HAVE_KERNEL_LZMA + select HAVE_KERNEL_LZO + select HAVE_KERNEL_XZ + select HAVE_KPROBES + select HAVE_KRETPROBES + select HAVE_KVM + select HAVE_LIVEPATCH + select HAVE_MEMBLOCK + select HAVE_MEMBLOCK_NODE_MAP + select HAVE_MEMBLOCK_PHYS_MAP + select HAVE_MOD_ARCH_SPECIFIC + select HAVE_OPROFILE + select HAVE_PERF_EVENTS + select HAVE_REGS_AND_STACK_ACCESS_API + select HAVE_SYSCALL_TRACEPOINTS + select HAVE_VIRT_CPU_ACCOUNTING + select MODULES_USE_ELF_RELA + select NO_BOOTMEM + select OLD_SIGACTION + select OLD_SIGSUSPEND3 + select SYSCTL_EXCEPTION_TRACE + select TTY + select VIRT_CPU_ACCOUNTING + select VIRT_TO_BUS + +config SCHED_OMIT_FRAME_POINTER + def_bool y + +config PGTABLE_LEVELS + int + default 4 if 64BIT + default 2 + +source "init/Kconfig" + +source "kernel/Kconfig.freezer" + +source "kernel/livepatch/Kconfig" + +menu "Processor type and features" + +config HAVE_MARCH_Z900_FEATURES + def_bool n + +config HAVE_MARCH_Z990_FEATURES + def_bool n + select HAVE_MARCH_Z900_FEATURES + +config HAVE_MARCH_Z9_109_FEATURES + def_bool n + select HAVE_MARCH_Z990_FEATURES + +config HAVE_MARCH_Z10_FEATURES + def_bool n + select HAVE_MARCH_Z9_109_FEATURES + +config HAVE_MARCH_Z196_FEATURES + def_bool n + select HAVE_MARCH_Z10_FEATURES + +config HAVE_MARCH_ZEC12_FEATURES + def_bool n + select HAVE_MARCH_Z196_FEATURES + +config HAVE_MARCH_Z13_FEATURES + def_bool n + select HAVE_MARCH_ZEC12_FEATURES + +choice + prompt "Processor type" + default MARCH_Z900 + +config MARCH_Z900 + bool "IBM zSeries model z800 and z900" + select HAVE_MARCH_Z900_FEATURES + help + Select this to enable optimizations for model z800/z900 (2064 and + 2066 series). This will enable some optimizations that are not + available on older ESA/390 (31 Bit) only CPUs. + +config MARCH_Z990 + bool "IBM zSeries model z890 and z990" + select HAVE_MARCH_Z990_FEATURES + help + Select this to enable optimizations for model z890/z990 (2084 and + 2086 series). The kernel will be slightly faster but will not work + on older machines. + +config MARCH_Z9_109 + bool "IBM System z9" + select HAVE_MARCH_Z9_109_FEATURES + help + Select this to enable optimizations for IBM System z9 (2094 and + 2096 series). The kernel will be slightly faster but will not work + on older machines. + +config MARCH_Z10 + bool "IBM System z10" + select HAVE_MARCH_Z10_FEATURES + help + Select this to enable optimizations for IBM System z10 (2097 and + 2098 series). The kernel will be slightly faster but will not work + on older machines. + +config MARCH_Z196 + bool "IBM zEnterprise 114 and 196" + select HAVE_MARCH_Z196_FEATURES + help + Select this to enable optimizations for IBM zEnterprise 114 and 196 + (2818 and 2817 series). The kernel will be slightly faster but will + not work on older machines. + +config MARCH_ZEC12 + bool "IBM zBC12 and zEC12" + select HAVE_MARCH_ZEC12_FEATURES + help + Select this to enable optimizations for IBM zBC12 and zEC12 (2828 and + 2827 series). The kernel will be slightly faster but will not work on + older machines. + +config MARCH_Z13 + bool "IBM z13" + select HAVE_MARCH_Z13_FEATURES + help + Select this to enable optimizations for IBM z13 (2964 series). + The kernel will be slightly faster but will not work on older + machines. + +endchoice + +config MARCH_Z900_TUNE + def_bool TUNE_Z900 || MARCH_Z900 && TUNE_DEFAULT + +config MARCH_Z990_TUNE + def_bool TUNE_Z990 || MARCH_Z990 && TUNE_DEFAULT + +config MARCH_Z9_109_TUNE + def_bool TUNE_Z9_109 || MARCH_Z9_109 && TUNE_DEFAULT + +config MARCH_Z10_TUNE + def_bool TUNE_Z10 || MARCH_Z10 && TUNE_DEFAULT + +config MARCH_Z196_TUNE + def_bool TUNE_Z196 || MARCH_Z196 && TUNE_DEFAULT + +config MARCH_ZEC12_TUNE + def_bool TUNE_ZEC12 || MARCH_ZEC12 && TUNE_DEFAULT + +config MARCH_Z13_TUNE + def_bool TUNE_Z13 || MARCH_Z13 && TUNE_DEFAULT + +choice + prompt "Tune code generation" + default TUNE_DEFAULT + help + Cause the compiler to tune (-mtune) the generated code for a machine. + This will make the code run faster on the selected machine but + somewhat slower on other machines. + This option only changes how the compiler emits instructions, not the + selection of instructions itself, so the resulting kernel will run on + all other machines. + +config TUNE_DEFAULT + bool "Default" + help + Tune the generated code for the target processor for which the kernel + will be compiled. + +config TUNE_Z900 + bool "IBM zSeries model z800 and z900" + +config TUNE_Z990 + bool "IBM zSeries model z890 and z990" + +config TUNE_Z9_109 + bool "IBM System z9" + +config TUNE_Z10 + bool "IBM System z10" + +config TUNE_Z196 + bool "IBM zEnterprise 114 and 196" + +config TUNE_ZEC12 + bool "IBM zBC12 and zEC12" + +config TUNE_Z13 + bool "IBM z13" + +endchoice + +config 64BIT + def_bool y + +config COMPAT + def_bool y + prompt "Kernel support for 31 bit emulation" + select COMPAT_BINFMT_ELF if BINFMT_ELF + select ARCH_WANT_OLD_COMPAT_IPC + select COMPAT_OLD_SIGACTION + depends on MULTIUSER + help + Select this option if you want to enable your system kernel to + handle system-calls from ELF binaries for 31 bit ESA. This option + (and some other stuff like libraries and such) is needed for + executing 31 bit applications. It is safe to say "Y". + +config SYSVIPC_COMPAT + def_bool y if COMPAT && SYSVIPC + +config KEYS_COMPAT + def_bool y if COMPAT && KEYS + +config SMP + def_bool y + prompt "Symmetric multi-processing support" + ---help--- + This enables support for systems with more than one CPU. If you have + a system with only one CPU, like most personal computers, say N. If + you have a system with more than one CPU, say Y. + + If you say N here, the kernel will run on uni- and multiprocessor + machines, but will use only one CPU of a multiprocessor machine. If + you say Y here, the kernel will run on many, but not all, + uniprocessor machines. On a uniprocessor machine, the kernel + will run faster if you say N here. + + See also the SMP-HOWTO available at + <http://www.tldp.org/docs.html#howto>. + + Even if you don't know what to do here, say Y. + +config NR_CPUS + int "Maximum number of CPUs (2-512)" + range 2 512 + depends on SMP + default "64" + help + This allows you to specify the maximum number of CPUs which this + kernel will support. The maximum supported value is 512 and the + minimum value which makes sense is 2. + + This is purely to save memory - each supported CPU adds + approximately sixteen kilobytes to the kernel image. + +config HOTPLUG_CPU + def_bool y + prompt "Support for hot-pluggable CPUs" + depends on SMP + help + Say Y here to be able to turn CPUs off and on. CPUs + can be controlled through /sys/devices/system/cpu/cpu#. + Say N if you want to disable CPU hotplug. + +config SCHED_SMT + def_bool n + +config SCHED_MC + def_bool n + +config SCHED_BOOK + def_bool n + +config SCHED_TOPOLOGY + def_bool y + prompt "Topology scheduler support" + depends on SMP + select SCHED_SMT + select SCHED_MC + select SCHED_BOOK + help + Topology scheduler support improves the CPU scheduler's decision + making when dealing with machines that have multi-threading, + multiple cores or multiple books. + +source kernel/Kconfig.preempt + +source kernel/Kconfig.hz + +endmenu + +menu "Memory setup" + +config ARCH_SPARSEMEM_ENABLE + def_bool y + select SPARSEMEM_VMEMMAP_ENABLE + select SPARSEMEM_VMEMMAP + +config ARCH_SPARSEMEM_DEFAULT + def_bool y + +config ARCH_SELECT_MEMORY_MODEL + def_bool y + +config ARCH_ENABLE_MEMORY_HOTPLUG + def_bool y if SPARSEMEM + +config ARCH_ENABLE_MEMORY_HOTREMOVE + def_bool y + +config ARCH_ENABLE_SPLIT_PMD_PTLOCK + def_bool y + +config FORCE_MAX_ZONEORDER + int + default "9" + +source "mm/Kconfig" + +config PACK_STACK + def_bool y + prompt "Pack kernel stack" + help + This option enables the compiler option -mkernel-backchain if it + is available. If the option is available the compiler supports + the new stack layout which dramatically reduces the minimum stack + frame size. With an old compiler a non-leaf function needs a + minimum of 96 bytes on 31 bit and 160 bytes on 64 bit. With + -mkernel-backchain the minimum size drops to 16 byte on 31 bit + and 24 byte on 64 bit. + + Say Y if you are unsure. + +config CHECK_STACK + def_bool y + prompt "Detect kernel stack overflow" + help + This option enables the compiler option -mstack-guard and + -mstack-size if they are available. If the compiler supports them + it will emit additional code to each function prolog to trigger + an illegal operation if the kernel stack is about to overflow. + + Say N if you are unsure. + +config STACK_GUARD + int "Size of the guard area (128-1024)" + range 128 1024 + depends on CHECK_STACK + default "256" + help + This allows you to specify the size of the guard area at the lower + end of the kernel stack. If the kernel stack points into the guard + area on function entry an illegal operation is triggered. The size + needs to be a power of 2. Please keep in mind that the size of an + interrupt frame is 184 bytes for 31 bit and 328 bytes on 64 bit. + The minimum size for the stack guard should be 256 for 31 bit and + 512 for 64 bit. + +config WARN_DYNAMIC_STACK + def_bool n + prompt "Emit compiler warnings for function with dynamic stack usage" + help + This option enables the compiler option -mwarn-dynamicstack. If the + compiler supports this options generates warnings for functions + that dynamically allocate stack space using alloca. + + Say N if you are unsure. + +endmenu + +menu "I/O subsystem" + +config QDIO + def_tristate y + prompt "QDIO support" + ---help--- + This driver provides the Queued Direct I/O base support for + IBM System z. + + To compile this driver as a module, choose M here: the + module will be called qdio. + + If unsure, say Y. + +menuconfig PCI + bool "PCI support" + select HAVE_DMA_ATTRS + select PCI_MSI + help + Enable PCI support. + +if PCI + +config PCI_NR_FUNCTIONS + int "Maximum number of PCI functions (1-4096)" + range 1 4096 + default "64" + help + This allows you to specify the maximum number of PCI functions which + this kernel will support. + +config PCI_NR_MSI + int "Maximum number of MSI interrupts (64-32768)" + range 64 32768 + default "256" + help + This defines the number of virtual interrupts the kernel will + provide for MSI interrupts. If you configure your system to have + too few drivers will fail to allocate MSI interrupts for all + PCI devices. + +source "drivers/pci/Kconfig" +source "drivers/pci/pcie/Kconfig" +source "drivers/pci/hotplug/Kconfig" + +endif # PCI + +config PCI_DOMAINS + def_bool PCI + +config HAS_IOMEM + def_bool PCI + +config IOMMU_HELPER + def_bool PCI + +config HAS_DMA + def_bool PCI + select HAVE_DMA_API_DEBUG + +config NEED_SG_DMA_LENGTH + def_bool PCI + +config NEED_DMA_MAP_STATE + def_bool PCI + +config CHSC_SCH + def_tristate m + prompt "Support for CHSC subchannels" + help + This driver allows usage of CHSC subchannels. A CHSC subchannel + is usually present on LPAR only. + The driver creates a device /dev/chsc, which may be used to + obtain I/O configuration information about the machine and + to issue asynchronous chsc commands (DANGEROUS). + You will usually only want to use this interface on a special + LPAR designated for system management. + + To compile this driver as a module, choose M here: the + module will be called chsc_sch. + + If unsure, say N. + +config SCM_BUS + def_bool y + prompt "SCM bus driver" + help + Bus driver for Storage Class Memory. + +config EADM_SCH + def_tristate m + prompt "Support for EADM subchannels" + depends on SCM_BUS + help + This driver allows usage of EADM subchannels. EADM subchannels act + as a communication vehicle for SCM increments. + + To compile this driver as a module, choose M here: the + module will be called eadm_sch. + +endmenu + +menu "Dump support" + +config CRASH_DUMP + bool "kernel crash dumps" + depends on SMP + select KEXEC + help + Generate crash dump after being started by kexec. + Crash dump kernels are loaded in the main kernel with kexec-tools + into a specially reserved region and then later executed after + a crash by kdump/kexec. + Refer to <file:Documentation/s390/zfcpdump.txt> for more details on this. + This option also enables s390 zfcpdump. + See also <file:Documentation/s390/zfcpdump.txt> + +endmenu + +menu "Executable file formats / Emulations" + +source "fs/Kconfig.binfmt" + +config SECCOMP + def_bool y + prompt "Enable seccomp to safely compute untrusted bytecode" + depends on PROC_FS + help + This kernel feature is useful for number crunching applications + that may need to compute untrusted bytecode during their + execution. By using pipes or other transports made available to + the process as file descriptors supporting the read/write + syscalls, it's possible to isolate those applications in + their own address space using seccomp. Once seccomp is + enabled via /proc/<pid>/seccomp, it cannot be disabled + and the task is only allowed to execute a few safe syscalls + defined by each seccomp mode. + + If unsure, say Y. + +endmenu + +menu "Power Management" + +config ARCH_HIBERNATION_POSSIBLE + def_bool y + +source "kernel/power/Kconfig" + +endmenu + +source "net/Kconfig" + +config PCMCIA + def_bool n + +config CCW + def_bool y + +source "drivers/Kconfig" + +source "fs/Kconfig" + +source "arch/s390/Kconfig.debug" + +source "security/Kconfig" + +source "crypto/Kconfig" + +source "lib/Kconfig" + +menu "Virtualization" + +config PFAULT + def_bool y + prompt "Pseudo page fault support" + help + Select this option, if you want to use PFAULT pseudo page fault + handling under VM. If running native or in LPAR, this option + has no effect. If your VM does not support PFAULT, PAGEEX + pseudo page fault handling will be used. + Note that VM 4.2 supports PFAULT but has a bug in its + implementation that causes some problems. + Everybody who wants to run Linux under VM != VM4.2 should select + this option. + +config SHARED_KERNEL + bool "VM shared kernel support" + depends on !JUMP_LABEL + help + Select this option, if you want to share the text segment of the + Linux kernel between different VM guests. This reduces memory + usage with lots of guests but greatly increases kernel size. + Also if a kernel was IPL'ed from a shared segment the kexec system + call will not work. + You should only select this option if you know what you are + doing and want to exploit this feature. + +config CMM + def_tristate n + prompt "Cooperative memory management" + help + Select this option, if you want to enable the kernel interface + to reduce the memory size of the system. This is accomplished + by allocating pages of memory and put them "on hold". This only + makes sense for a system running under VM where the unused pages + will be reused by VM for other guest systems. The interface + allows an external monitor to balance memory of many systems. + Everybody who wants to run Linux under VM should select this + option. + +config CMM_IUCV + def_bool y + prompt "IUCV special message interface to cooperative memory management" + depends on CMM && (SMSGIUCV=y || CMM=SMSGIUCV) + help + Select this option to enable the special message interface to + the cooperative memory management. + +config APPLDATA_BASE + def_bool n + prompt "Linux - VM Monitor Stream, base infrastructure" + depends on PROC_FS + help + This provides a kernel interface for creating and updating z/VM APPLDATA + monitor records. The monitor records are updated at certain time + intervals, once the timer is started. + Writing 1 or 0 to /proc/appldata/timer starts(1) or stops(0) the timer, + i.e. enables or disables monitoring on the Linux side. + A custom interval value (in seconds) can be written to + /proc/appldata/interval. + + Defaults are 60 seconds interval and timer off. + The /proc entries can also be read from, showing the current settings. + +config APPLDATA_MEM + def_tristate m + prompt "Monitor memory management statistics" + depends on APPLDATA_BASE && VM_EVENT_COUNTERS + help + This provides memory management related data to the Linux - VM Monitor + Stream, like paging/swapping rate, memory utilisation, etc. + Writing 1 or 0 to /proc/appldata/memory creates(1) or removes(0) a z/VM + APPLDATA monitor record, i.e. enables or disables monitoring this record + on the z/VM side. + + Default is disabled. + The /proc entry can also be read from, showing the current settings. + + This can also be compiled as a module, which will be called + appldata_mem.o. + +config APPLDATA_OS + def_tristate m + prompt "Monitor OS statistics" + depends on APPLDATA_BASE + help + This provides OS related data to the Linux - VM Monitor Stream, like + CPU utilisation, etc. + Writing 1 or 0 to /proc/appldata/os creates(1) or removes(0) a z/VM + APPLDATA monitor record, i.e. enables or disables monitoring this record + on the z/VM side. + + Default is disabled. + This can also be compiled as a module, which will be called + appldata_os.o. + +config APPLDATA_NET_SUM + def_tristate m + prompt "Monitor overall network statistics" + depends on APPLDATA_BASE && NET + help + This provides network related data to the Linux - VM Monitor Stream, + currently there is only a total sum of network I/O statistics, no + per-interface data. + Writing 1 or 0 to /proc/appldata/net_sum creates(1) or removes(0) a z/VM + APPLDATA monitor record, i.e. enables or disables monitoring this record + on the z/VM side. + + Default is disabled. + This can also be compiled as a module, which will be called + appldata_net_sum.o. + +config S390_HYPFS_FS + def_bool y + prompt "s390 hypervisor file system support" + select SYS_HYPERVISOR + help + This is a virtual file system intended to provide accounting + information in an s390 hypervisor environment. + +source "arch/s390/kvm/Kconfig" + +config S390_GUEST + def_bool y + prompt "s390 support for virtio devices" + select TTY + select VIRTUALIZATION + select VIRTIO + select VIRTIO_CONSOLE + help + Enabling this option adds support for virtio based paravirtual device + drivers on s390. + + Select this option if you want to run the kernel as a guest under + the KVM hypervisor. + +endmenu diff --git a/arch/s390/Kconfig.debug b/arch/s390/Kconfig.debug new file mode 100644 index 000000000..c56878e12 --- /dev/null +++ b/arch/s390/Kconfig.debug @@ -0,0 +1,35 @@ +menu "Kernel hacking" + +config TRACE_IRQFLAGS_SUPPORT + def_bool y + +source "lib/Kconfig.debug" + +config STRICT_DEVMEM + def_bool y + prompt "Filter access to /dev/mem" + ---help--- + This option restricts access to /dev/mem. If this option is + disabled, you allow userspace access to all memory, including + kernel and userspace memory. Accidental memory access is likely + to be disastrous. + Memory access is required for experts who want to debug the kernel. + + If you are unsure, say Y. + +config S390_PTDUMP + bool "Export kernel pagetable layout to userspace via debugfs" + depends on DEBUG_KERNEL + select DEBUG_FS + ---help--- + Say Y here if you want to show the kernel pagetable layout in a + debugfs file. This information is only useful for kernel developers + who are working in architecture specific areas of the kernel. + It is probably not a good idea to enable this feature in a production + kernel. + If in doubt, say "N" + +config DEBUG_SET_MODULE_RONX + def_bool y + depends on MODULES +endmenu diff --git a/arch/s390/Makefile b/arch/s390/Makefile new file mode 100644 index 000000000..667b1bca5 --- /dev/null +++ b/arch/s390/Makefile @@ -0,0 +1,130 @@ +# +# s390/Makefile +# +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. Remember to do have actions +# for "archclean" and "archdep" for cleaning up and making dependencies for +# this architecture +# +# This file is subject to the terms and conditions of the GNU General Public +# License. See the file "COPYING" in the main directory of this archive +# for more details. +# +# Copyright (C) 1994 by Linus Torvalds +# + +LD_BFD := elf64-s390 +LDFLAGS := -m elf64_s390 +KBUILD_AFLAGS_MODULE += -fPIC +KBUILD_CFLAGS_MODULE += -fPIC +KBUILD_CFLAGS += -m64 +KBUILD_AFLAGS += -m64 +UTS_MACHINE := s390x +STACK_SIZE := 16384 +CHECKFLAGS += -D__s390__ -D__s390x__ + +export LD_BFD + +mflags-$(CONFIG_MARCH_Z900) := -march=z900 +mflags-$(CONFIG_MARCH_Z990) := -march=z990 +mflags-$(CONFIG_MARCH_Z9_109) := -march=z9-109 +mflags-$(CONFIG_MARCH_Z10) := -march=z10 +mflags-$(CONFIG_MARCH_Z196) := -march=z196 +mflags-$(CONFIG_MARCH_ZEC12) := -march=zEC12 +mflags-$(CONFIG_MARCH_Z13) := -march=z13 + +aflags-y += $(mflags-y) +cflags-y += $(mflags-y) + +cflags-$(CONFIG_MARCH_Z900_TUNE) += -mtune=z900 +cflags-$(CONFIG_MARCH_Z990_TUNE) += -mtune=z990 +cflags-$(CONFIG_MARCH_Z9_109_TUNE) += -mtune=z9-109 +cflags-$(CONFIG_MARCH_Z10_TUNE) += -mtune=z10 +cflags-$(CONFIG_MARCH_Z196_TUNE) += -mtune=z196 +cflags-$(CONFIG_MARCH_ZEC12_TUNE) += -mtune=zEC12 +cflags-$(CONFIG_MARCH_Z13_TUNE) += -mtune=z13 + +#KBUILD_IMAGE is necessary for make rpm +KBUILD_IMAGE :=arch/s390/boot/image + +# +# Prevent tail-call optimizations, to get clearer backtraces: +# +cflags-$(CONFIG_FRAME_POINTER) += -fno-optimize-sibling-calls + +# old style option for packed stacks +ifeq ($(call cc-option-yn,-mkernel-backchain),y) +cflags-$(CONFIG_PACK_STACK) += -mkernel-backchain -D__PACK_STACK +aflags-$(CONFIG_PACK_STACK) += -D__PACK_STACK +endif + +# new style option for packed stacks +ifeq ($(call cc-option-yn,-mpacked-stack),y) +cflags-$(CONFIG_PACK_STACK) += -mpacked-stack -D__PACK_STACK +aflags-$(CONFIG_PACK_STACK) += -D__PACK_STACK +endif + +ifeq ($(call cc-option-yn,-mstack-size=8192 -mstack-guard=128),y) +cflags-$(CONFIG_CHECK_STACK) += -mstack-size=$(STACK_SIZE) +ifneq ($(call cc-option-yn,-mstack-size=8192),y) +cflags-$(CONFIG_CHECK_STACK) += -mstack-guard=$(CONFIG_STACK_GUARD) +endif +endif + +ifeq ($(call cc-option-yn,-mwarn-dynamicstack),y) +cflags-$(CONFIG_WARN_DYNAMIC_STACK) += -mwarn-dynamicstack +endif + +ifdef CONFIG_FUNCTION_TRACER +# make use of hotpatch feature if the compiler supports it +cc_hotpatch := -mhotpatch=0,3 +ifeq ($(call cc-option-yn,$(cc_hotpatch)),y) +CC_FLAGS_FTRACE := $(cc_hotpatch) +KBUILD_AFLAGS += -DCC_USING_HOTPATCH +KBUILD_CFLAGS += -DCC_USING_HOTPATCH +endif +endif + +KBUILD_CFLAGS += -mbackchain -msoft-float $(cflags-y) +KBUILD_CFLAGS += -pipe -fno-strength-reduce -Wno-sign-compare +KBUILD_AFLAGS += $(aflags-y) + +OBJCOPYFLAGS := -O binary + +head-y := arch/s390/kernel/head.o +head-y += arch/s390/kernel/head64.o + +# See arch/s390/Kbuild for content of core part of the kernel +core-y += arch/s390/ + +libs-y += arch/s390/lib/ +drivers-y += drivers/s390/ + +# must be linked after kernel +drivers-$(CONFIG_OPROFILE) += arch/s390/oprofile/ + +boot := arch/s390/boot + +all: image bzImage + +install: vmlinux + $(Q)$(MAKE) $(build)=$(boot) $@ + +image bzImage: vmlinux + $(Q)$(MAKE) $(build)=$(boot) $(boot)/$@ + +zfcpdump: + $(Q)$(MAKE) $(build)=$(boot) $(boot)/$@ + +vdso_install: + $(Q)$(MAKE) $(build)=arch/$(ARCH)/kernel/vdso64 $@ + $(Q)$(MAKE) $(build)=arch/$(ARCH)/kernel/vdso32 $@ + +archclean: + $(Q)$(MAKE) $(clean)=$(boot) + +# Don't use tabs in echo arguments +define archhelp + echo '* image - Kernel image for IPL ($(boot)/image)' + echo '* bzImage - Compressed kernel image for IPL ($(boot)/bzImage)' +endef diff --git a/arch/s390/appldata/Makefile b/arch/s390/appldata/Makefile new file mode 100644 index 000000000..99f1cf071 --- /dev/null +++ b/arch/s390/appldata/Makefile @@ -0,0 +1,8 @@ +# +# Makefile for the Linux - z/VM Monitor Stream. +# + +obj-$(CONFIG_APPLDATA_BASE) += appldata_base.o +obj-$(CONFIG_APPLDATA_MEM) += appldata_mem.o +obj-$(CONFIG_APPLDATA_OS) += appldata_os.o +obj-$(CONFIG_APPLDATA_NET_SUM) += appldata_net_sum.o diff --git a/arch/s390/appldata/appldata.h b/arch/s390/appldata/appldata.h new file mode 100644 index 000000000..4a67f2b5f --- /dev/null +++ b/arch/s390/appldata/appldata.h @@ -0,0 +1,47 @@ +/* + * Definitions and interface for Linux - z/VM Monitor Stream. + * + * Copyright IBM Corp. 2003, 2008 + * + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> + */ + +#define APPLDATA_MAX_REC_SIZE 4024 /* Maximum size of the */ + /* data buffer */ +#define APPLDATA_MAX_PROCS 100 + +#define APPLDATA_PROC_NAME_LENGTH 16 /* Max. length of /proc name */ + +#define APPLDATA_RECORD_MEM_ID 0x01 /* IDs to identify the */ +#define APPLDATA_RECORD_OS_ID 0x02 /* individual records, */ +#define APPLDATA_RECORD_NET_SUM_ID 0x03 /* must be < 256 ! */ +#define APPLDATA_RECORD_PROC_ID 0x04 + +#define CTL_APPLDATA_TIMER 2121 /* sysctl IDs, must be unique */ +#define CTL_APPLDATA_INTERVAL 2122 +#define CTL_APPLDATA_MEM 2123 +#define CTL_APPLDATA_OS 2124 +#define CTL_APPLDATA_NET_SUM 2125 +#define CTL_APPLDATA_PROC 2126 + +struct appldata_ops { + struct list_head list; + struct ctl_table_header *sysctl_header; + struct ctl_table *ctl_table; + int active; /* monitoring status */ + + /* fill in from here */ + char name[APPLDATA_PROC_NAME_LENGTH]; /* name of /proc fs node */ + unsigned char record_nr; /* Record Nr. for Product ID */ + void (*callback)(void *data); /* callback function */ + void *data; /* record data */ + unsigned int size; /* size of record */ + struct module *owner; /* THIS_MODULE */ + char mod_lvl[2]; /* modification level, EBCDIC */ +}; + +extern int appldata_register_ops(struct appldata_ops *ops); +extern void appldata_unregister_ops(struct appldata_ops *ops); +extern int appldata_diag(char record_nr, u16 function, unsigned long buffer, + u16 length, char *mod_lvl); + diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c new file mode 100644 index 000000000..15c94246b --- /dev/null +++ b/arch/s390/appldata/appldata_base.c @@ -0,0 +1,574 @@ +/* + * Base infrastructure for Linux-z/VM Monitor Stream, Stage 1. + * Exports appldata_register_ops() and appldata_unregister_ops() for the + * data gathering modules. + * + * Copyright IBM Corp. 2003, 2009 + * + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> + */ + +#define KMSG_COMPONENT "appldata" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/errno.h> +#include <linux/interrupt.h> +#include <linux/proc_fs.h> +#include <linux/mm.h> +#include <linux/swap.h> +#include <linux/pagemap.h> +#include <linux/sysctl.h> +#include <linux/notifier.h> +#include <linux/cpu.h> +#include <linux/workqueue.h> +#include <linux/suspend.h> +#include <linux/platform_device.h> +#include <asm/appldata.h> +#include <asm/vtimer.h> +#include <asm/uaccess.h> +#include <asm/io.h> +#include <asm/smp.h> + +#include "appldata.h" + + +#define APPLDATA_CPU_INTERVAL 10000 /* default (CPU) time for + sampling interval in + milliseconds */ + +#define TOD_MICRO 0x01000 /* nr. of TOD clock units + for 1 microsecond */ + +static struct platform_device *appldata_pdev; + +/* + * /proc entries (sysctl) + */ +static const char appldata_proc_name[APPLDATA_PROC_NAME_LENGTH] = "appldata"; +static int appldata_timer_handler(struct ctl_table *ctl, int write, + void __user *buffer, size_t *lenp, loff_t *ppos); +static int appldata_interval_handler(struct ctl_table *ctl, int write, + void __user *buffer, + size_t *lenp, loff_t *ppos); + +static struct ctl_table_header *appldata_sysctl_header; +static struct ctl_table appldata_table[] = { + { + .procname = "timer", + .mode = S_IRUGO | S_IWUSR, + .proc_handler = appldata_timer_handler, + }, + { + .procname = "interval", + .mode = S_IRUGO | S_IWUSR, + .proc_handler = appldata_interval_handler, + }, + { }, +}; + +static struct ctl_table appldata_dir_table[] = { + { + .procname = appldata_proc_name, + .maxlen = 0, + .mode = S_IRUGO | S_IXUGO, + .child = appldata_table, + }, + { }, +}; + +/* + * Timer + */ +static struct vtimer_list appldata_timer; + +static DEFINE_SPINLOCK(appldata_timer_lock); +static int appldata_interval = APPLDATA_CPU_INTERVAL; +static int appldata_timer_active; +static int appldata_timer_suspended = 0; + +/* + * Work queue + */ +static struct workqueue_struct *appldata_wq; +static void appldata_work_fn(struct work_struct *work); +static DECLARE_WORK(appldata_work, appldata_work_fn); + + +/* + * Ops list + */ +static DEFINE_MUTEX(appldata_ops_mutex); +static LIST_HEAD(appldata_ops_list); + + +/*************************** timer, work, DIAG *******************************/ +/* + * appldata_timer_function() + * + * schedule work and reschedule timer + */ +static void appldata_timer_function(unsigned long data) +{ + queue_work(appldata_wq, (struct work_struct *) data); +} + +/* + * appldata_work_fn() + * + * call data gathering function for each (active) module + */ +static void appldata_work_fn(struct work_struct *work) +{ + struct list_head *lh; + struct appldata_ops *ops; + + mutex_lock(&appldata_ops_mutex); + list_for_each(lh, &appldata_ops_list) { + ops = list_entry(lh, struct appldata_ops, list); + if (ops->active == 1) { + ops->callback(ops->data); + } + } + mutex_unlock(&appldata_ops_mutex); +} + +/* + * appldata_diag() + * + * prepare parameter list, issue DIAG 0xDC + */ +int appldata_diag(char record_nr, u16 function, unsigned long buffer, + u16 length, char *mod_lvl) +{ + struct appldata_product_id id = { + .prod_nr = {0xD3, 0xC9, 0xD5, 0xE4, + 0xE7, 0xD2, 0xD9}, /* "LINUXKR" */ + .prod_fn = 0xD5D3, /* "NL" */ + .version_nr = 0xF2F6, /* "26" */ + .release_nr = 0xF0F1, /* "01" */ + }; + + id.record_nr = record_nr; + id.mod_lvl = (mod_lvl[0]) << 8 | mod_lvl[1]; + return appldata_asm(&id, function, (void *) buffer, length); +} +/************************ timer, work, DIAG <END> ****************************/ + + +/****************************** /proc stuff **********************************/ + +#define APPLDATA_ADD_TIMER 0 +#define APPLDATA_DEL_TIMER 1 +#define APPLDATA_MOD_TIMER 2 + +/* + * __appldata_vtimer_setup() + * + * Add, delete or modify virtual timers on all online cpus. + * The caller needs to get the appldata_timer_lock spinlock. + */ +static void __appldata_vtimer_setup(int cmd) +{ + u64 timer_interval = (u64) appldata_interval * 1000 * TOD_MICRO; + + switch (cmd) { + case APPLDATA_ADD_TIMER: + if (appldata_timer_active) + break; + appldata_timer.expires = timer_interval; + add_virt_timer_periodic(&appldata_timer); + appldata_timer_active = 1; + break; + case APPLDATA_DEL_TIMER: + del_virt_timer(&appldata_timer); + if (!appldata_timer_active) + break; + appldata_timer_active = 0; + break; + case APPLDATA_MOD_TIMER: + if (!appldata_timer_active) + break; + mod_virt_timer_periodic(&appldata_timer, timer_interval); + } +} + +/* + * appldata_timer_handler() + * + * Start/Stop timer, show status of timer (0 = not active, 1 = active) + */ +static int +appldata_timer_handler(struct ctl_table *ctl, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + unsigned int len; + char buf[2]; + + if (!*lenp || *ppos) { + *lenp = 0; + return 0; + } + if (!write) { + strncpy(buf, appldata_timer_active ? "1\n" : "0\n", + ARRAY_SIZE(buf)); + len = strnlen(buf, ARRAY_SIZE(buf)); + if (len > *lenp) + len = *lenp; + if (copy_to_user(buffer, buf, len)) + return -EFAULT; + goto out; + } + len = *lenp; + if (copy_from_user(buf, buffer, len > sizeof(buf) ? sizeof(buf) : len)) + return -EFAULT; + spin_lock(&appldata_timer_lock); + if (buf[0] == '1') + __appldata_vtimer_setup(APPLDATA_ADD_TIMER); + else if (buf[0] == '0') + __appldata_vtimer_setup(APPLDATA_DEL_TIMER); + spin_unlock(&appldata_timer_lock); +out: + *lenp = len; + *ppos += len; + return 0; +} + +/* + * appldata_interval_handler() + * + * Set (CPU) timer interval for collection of data (in milliseconds), show + * current timer interval. + */ +static int +appldata_interval_handler(struct ctl_table *ctl, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + unsigned int len; + int interval; + char buf[16]; + + if (!*lenp || *ppos) { + *lenp = 0; + return 0; + } + if (!write) { + len = sprintf(buf, "%i\n", appldata_interval); + if (len > *lenp) + len = *lenp; + if (copy_to_user(buffer, buf, len)) + return -EFAULT; + goto out; + } + len = *lenp; + if (copy_from_user(buf, buffer, len > sizeof(buf) ? sizeof(buf) : len)) + return -EFAULT; + interval = 0; + sscanf(buf, "%i", &interval); + if (interval <= 0) + return -EINVAL; + + spin_lock(&appldata_timer_lock); + appldata_interval = interval; + __appldata_vtimer_setup(APPLDATA_MOD_TIMER); + spin_unlock(&appldata_timer_lock); +out: + *lenp = len; + *ppos += len; + return 0; +} + +/* + * appldata_generic_handler() + * + * Generic start/stop monitoring and DIAG, show status of + * monitoring (0 = not in process, 1 = in process) + */ +static int +appldata_generic_handler(struct ctl_table *ctl, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + struct appldata_ops *ops = NULL, *tmp_ops; + unsigned int len; + int rc, found; + char buf[2]; + struct list_head *lh; + + found = 0; + mutex_lock(&appldata_ops_mutex); + list_for_each(lh, &appldata_ops_list) { + tmp_ops = list_entry(lh, struct appldata_ops, list); + if (&tmp_ops->ctl_table[2] == ctl) { + found = 1; + } + } + if (!found) { + mutex_unlock(&appldata_ops_mutex); + return -ENODEV; + } + ops = ctl->data; + if (!try_module_get(ops->owner)) { // protect this function + mutex_unlock(&appldata_ops_mutex); + return -ENODEV; + } + mutex_unlock(&appldata_ops_mutex); + + if (!*lenp || *ppos) { + *lenp = 0; + module_put(ops->owner); + return 0; + } + if (!write) { + strncpy(buf, ops->active ? "1\n" : "0\n", ARRAY_SIZE(buf)); + len = strnlen(buf, ARRAY_SIZE(buf)); + if (len > *lenp) + len = *lenp; + if (copy_to_user(buffer, buf, len)) { + module_put(ops->owner); + return -EFAULT; + } + goto out; + } + len = *lenp; + if (copy_from_user(buf, buffer, + len > sizeof(buf) ? sizeof(buf) : len)) { + module_put(ops->owner); + return -EFAULT; + } + + mutex_lock(&appldata_ops_mutex); + if ((buf[0] == '1') && (ops->active == 0)) { + // protect work queue callback + if (!try_module_get(ops->owner)) { + mutex_unlock(&appldata_ops_mutex); + module_put(ops->owner); + return -ENODEV; + } + ops->callback(ops->data); // init record + rc = appldata_diag(ops->record_nr, + APPLDATA_START_INTERVAL_REC, + (unsigned long) ops->data, ops->size, + ops->mod_lvl); + if (rc != 0) { + pr_err("Starting the data collection for %s " + "failed with rc=%d\n", ops->name, rc); + module_put(ops->owner); + } else + ops->active = 1; + } else if ((buf[0] == '0') && (ops->active == 1)) { + ops->active = 0; + rc = appldata_diag(ops->record_nr, APPLDATA_STOP_REC, + (unsigned long) ops->data, ops->size, + ops->mod_lvl); + if (rc != 0) + pr_err("Stopping the data collection for %s " + "failed with rc=%d\n", ops->name, rc); + module_put(ops->owner); + } + mutex_unlock(&appldata_ops_mutex); +out: + *lenp = len; + *ppos += len; + module_put(ops->owner); + return 0; +} + +/*************************** /proc stuff <END> *******************************/ + + +/************************* module-ops management *****************************/ +/* + * appldata_register_ops() + * + * update ops list, register /proc/sys entries + */ +int appldata_register_ops(struct appldata_ops *ops) +{ + if (ops->size > APPLDATA_MAX_REC_SIZE) + return -EINVAL; + + ops->ctl_table = kzalloc(4 * sizeof(struct ctl_table), GFP_KERNEL); + if (!ops->ctl_table) + return -ENOMEM; + + mutex_lock(&appldata_ops_mutex); + list_add(&ops->list, &appldata_ops_list); + mutex_unlock(&appldata_ops_mutex); + + ops->ctl_table[0].procname = appldata_proc_name; + ops->ctl_table[0].maxlen = 0; + ops->ctl_table[0].mode = S_IRUGO | S_IXUGO; + ops->ctl_table[0].child = &ops->ctl_table[2]; + + ops->ctl_table[2].procname = ops->name; + ops->ctl_table[2].mode = S_IRUGO | S_IWUSR; + ops->ctl_table[2].proc_handler = appldata_generic_handler; + ops->ctl_table[2].data = ops; + + ops->sysctl_header = register_sysctl_table(ops->ctl_table); + if (!ops->sysctl_header) + goto out; + return 0; +out: + mutex_lock(&appldata_ops_mutex); + list_del(&ops->list); + mutex_unlock(&appldata_ops_mutex); + kfree(ops->ctl_table); + return -ENOMEM; +} + +/* + * appldata_unregister_ops() + * + * update ops list, unregister /proc entries, stop DIAG if necessary + */ +void appldata_unregister_ops(struct appldata_ops *ops) +{ + mutex_lock(&appldata_ops_mutex); + list_del(&ops->list); + mutex_unlock(&appldata_ops_mutex); + unregister_sysctl_table(ops->sysctl_header); + kfree(ops->ctl_table); +} +/********************** module-ops management <END> **************************/ + + +/**************************** suspend / resume *******************************/ +static int appldata_freeze(struct device *dev) +{ + struct appldata_ops *ops; + int rc; + struct list_head *lh; + + spin_lock(&appldata_timer_lock); + if (appldata_timer_active) { + __appldata_vtimer_setup(APPLDATA_DEL_TIMER); + appldata_timer_suspended = 1; + } + spin_unlock(&appldata_timer_lock); + + mutex_lock(&appldata_ops_mutex); + list_for_each(lh, &appldata_ops_list) { + ops = list_entry(lh, struct appldata_ops, list); + if (ops->active == 1) { + rc = appldata_diag(ops->record_nr, APPLDATA_STOP_REC, + (unsigned long) ops->data, ops->size, + ops->mod_lvl); + if (rc != 0) + pr_err("Stopping the data collection for %s " + "failed with rc=%d\n", ops->name, rc); + } + } + mutex_unlock(&appldata_ops_mutex); + return 0; +} + +static int appldata_restore(struct device *dev) +{ + struct appldata_ops *ops; + int rc; + struct list_head *lh; + + spin_lock(&appldata_timer_lock); + if (appldata_timer_suspended) { + __appldata_vtimer_setup(APPLDATA_ADD_TIMER); + appldata_timer_suspended = 0; + } + spin_unlock(&appldata_timer_lock); + + mutex_lock(&appldata_ops_mutex); + list_for_each(lh, &appldata_ops_list) { + ops = list_entry(lh, struct appldata_ops, list); + if (ops->active == 1) { + ops->callback(ops->data); // init record + rc = appldata_diag(ops->record_nr, + APPLDATA_START_INTERVAL_REC, + (unsigned long) ops->data, ops->size, + ops->mod_lvl); + if (rc != 0) { + pr_err("Starting the data collection for %s " + "failed with rc=%d\n", ops->name, rc); + } + } + } + mutex_unlock(&appldata_ops_mutex); + return 0; +} + +static int appldata_thaw(struct device *dev) +{ + return appldata_restore(dev); +} + +static const struct dev_pm_ops appldata_pm_ops = { + .freeze = appldata_freeze, + .thaw = appldata_thaw, + .restore = appldata_restore, +}; + +static struct platform_driver appldata_pdrv = { + .driver = { + .name = "appldata", + .pm = &appldata_pm_ops, + }, +}; +/************************* suspend / resume <END> ****************************/ + + +/******************************* init / exit *********************************/ + +/* + * appldata_init() + * + * init timer, register /proc entries + */ +static int __init appldata_init(void) +{ + int rc; + + init_virt_timer(&appldata_timer); + appldata_timer.function = appldata_timer_function; + appldata_timer.data = (unsigned long) &appldata_work; + + rc = platform_driver_register(&appldata_pdrv); + if (rc) + return rc; + + appldata_pdev = platform_device_register_simple("appldata", -1, NULL, + 0); + if (IS_ERR(appldata_pdev)) { + rc = PTR_ERR(appldata_pdev); + goto out_driver; + } + appldata_wq = create_singlethread_workqueue("appldata"); + if (!appldata_wq) { + rc = -ENOMEM; + goto out_device; + } + + appldata_sysctl_header = register_sysctl_table(appldata_dir_table); + return 0; + +out_device: + platform_device_unregister(appldata_pdev); +out_driver: + platform_driver_unregister(&appldata_pdrv); + return rc; +} + +__initcall(appldata_init); + +/**************************** init / exit <END> ******************************/ + +EXPORT_SYMBOL_GPL(appldata_register_ops); +EXPORT_SYMBOL_GPL(appldata_unregister_ops); +EXPORT_SYMBOL_GPL(appldata_diag); + +#ifdef CONFIG_SWAP +EXPORT_SYMBOL_GPL(si_swapinfo); +#endif +EXPORT_SYMBOL_GPL(nr_threads); +EXPORT_SYMBOL_GPL(nr_running); +EXPORT_SYMBOL_GPL(nr_iowait); diff --git a/arch/s390/appldata/appldata_mem.c b/arch/s390/appldata/appldata_mem.c new file mode 100644 index 000000000..edcf2a706 --- /dev/null +++ b/arch/s390/appldata/appldata_mem.c @@ -0,0 +1,164 @@ +/* + * Data gathering module for Linux-VM Monitor Stream, Stage 1. + * Collects data related to memory management. + * + * Copyright IBM Corp. 2003, 2006 + * + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <linux/kernel_stat.h> +#include <linux/pagemap.h> +#include <linux/swap.h> +#include <linux/slab.h> +#include <asm/io.h> + +#include "appldata.h" + + +#define P2K(x) ((x) << (PAGE_SHIFT - 10)) /* Converts #Pages to KB */ + +/* + * Memory data + * + * This is accessed as binary data by z/VM. If changes to it can't be avoided, + * the structure version (product ID, see appldata_base.c) needs to be changed + * as well and all documentation and z/VM applications using it must be + * updated. + * + * The record layout is documented in the Linux for zSeries Device Drivers + * book: + * http://oss.software.ibm.com/developerworks/opensource/linux390/index.shtml + */ +struct appldata_mem_data { + u64 timestamp; + u32 sync_count_1; /* after VM collected the record data, */ + u32 sync_count_2; /* sync_count_1 and sync_count_2 should be the + same. If not, the record has been updated on + the Linux side while VM was collecting the + (possibly corrupt) data */ + + u64 pgpgin; /* data read from disk */ + u64 pgpgout; /* data written to disk */ + u64 pswpin; /* pages swapped in */ + u64 pswpout; /* pages swapped out */ + + u64 sharedram; /* sharedram is currently set to 0 */ + + u64 totalram; /* total main memory size */ + u64 freeram; /* free main memory size */ + u64 totalhigh; /* total high memory size */ + u64 freehigh; /* free high memory size */ + + u64 bufferram; /* memory reserved for buffers, free cache */ + u64 cached; /* size of (used) cache, w/o buffers */ + u64 totalswap; /* total swap space size */ + u64 freeswap; /* free swap space */ + +// New in 2.6 --> + u64 pgalloc; /* page allocations */ + u64 pgfault; /* page faults (major+minor) */ + u64 pgmajfault; /* page faults (major only) */ +// <-- New in 2.6 + +} __packed; + + +/* + * appldata_get_mem_data() + * + * gather memory data + */ +static void appldata_get_mem_data(void *data) +{ + /* + * don't put large structures on the stack, we are + * serialized through the appldata_ops_mutex and can use static + */ + static struct sysinfo val; + unsigned long ev[NR_VM_EVENT_ITEMS]; + struct appldata_mem_data *mem_data; + + mem_data = data; + mem_data->sync_count_1++; + + all_vm_events(ev); + mem_data->pgpgin = ev[PGPGIN] >> 1; + mem_data->pgpgout = ev[PGPGOUT] >> 1; + mem_data->pswpin = ev[PSWPIN]; + mem_data->pswpout = ev[PSWPOUT]; + mem_data->pgalloc = ev[PGALLOC_NORMAL]; + mem_data->pgalloc += ev[PGALLOC_DMA]; + mem_data->pgfault = ev[PGFAULT]; + mem_data->pgmajfault = ev[PGMAJFAULT]; + + si_meminfo(&val); + mem_data->sharedram = val.sharedram; + mem_data->totalram = P2K(val.totalram); + mem_data->freeram = P2K(val.freeram); + mem_data->totalhigh = P2K(val.totalhigh); + mem_data->freehigh = P2K(val.freehigh); + mem_data->bufferram = P2K(val.bufferram); + mem_data->cached = P2K(global_page_state(NR_FILE_PAGES) + - val.bufferram); + + si_swapinfo(&val); + mem_data->totalswap = P2K(val.totalswap); + mem_data->freeswap = P2K(val.freeswap); + + mem_data->timestamp = get_tod_clock(); + mem_data->sync_count_2++; +} + + +static struct appldata_ops ops = { + .name = "mem", + .record_nr = APPLDATA_RECORD_MEM_ID, + .size = sizeof(struct appldata_mem_data), + .callback = &appldata_get_mem_data, + .owner = THIS_MODULE, + .mod_lvl = {0xF0, 0xF0}, /* EBCDIC "00" */ +}; + + +/* + * appldata_mem_init() + * + * init_data, register ops + */ +static int __init appldata_mem_init(void) +{ + int ret; + + ops.data = kzalloc(sizeof(struct appldata_mem_data), GFP_KERNEL); + if (!ops.data) + return -ENOMEM; + + ret = appldata_register_ops(&ops); + if (ret) + kfree(ops.data); + + return ret; +} + +/* + * appldata_mem_exit() + * + * unregister ops + */ +static void __exit appldata_mem_exit(void) +{ + appldata_unregister_ops(&ops); + kfree(ops.data); +} + + +module_init(appldata_mem_init); +module_exit(appldata_mem_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gerald Schaefer"); +MODULE_DESCRIPTION("Linux-VM Monitor Stream, MEMORY statistics"); diff --git a/arch/s390/appldata/appldata_net_sum.c b/arch/s390/appldata/appldata_net_sum.c new file mode 100644 index 000000000..66037d262 --- /dev/null +++ b/arch/s390/appldata/appldata_net_sum.c @@ -0,0 +1,166 @@ +/* + * Data gathering module for Linux-VM Monitor Stream, Stage 1. + * Collects accumulated network statistics (Packets received/transmitted, + * dropped, errors, ...). + * + * Copyright IBM Corp. 2003, 2006 + * + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <linux/kernel_stat.h> +#include <linux/netdevice.h> +#include <net/net_namespace.h> + +#include "appldata.h" + + +/* + * Network data + * + * This is accessed as binary data by z/VM. If changes to it can't be avoided, + * the structure version (product ID, see appldata_base.c) needs to be changed + * as well and all documentation and z/VM applications using it must be updated. + * + * The record layout is documented in the Linux for zSeries Device Drivers + * book: + * http://oss.software.ibm.com/developerworks/opensource/linux390/index.shtml + */ +struct appldata_net_sum_data { + u64 timestamp; + u32 sync_count_1; /* after VM collected the record data, */ + u32 sync_count_2; /* sync_count_1 and sync_count_2 should be the + same. If not, the record has been updated on + the Linux side while VM was collecting the + (possibly corrupt) data */ + + u32 nr_interfaces; /* nr. of network interfaces being monitored */ + + u32 padding; /* next value is 64-bit aligned, so these */ + /* 4 byte would be padded out by compiler */ + + u64 rx_packets; /* total packets received */ + u64 tx_packets; /* total packets transmitted */ + u64 rx_bytes; /* total bytes received */ + u64 tx_bytes; /* total bytes transmitted */ + u64 rx_errors; /* bad packets received */ + u64 tx_errors; /* packet transmit problems */ + u64 rx_dropped; /* no space in linux buffers */ + u64 tx_dropped; /* no space available in linux */ + u64 collisions; /* collisions while transmitting */ +} __packed; + + +/* + * appldata_get_net_sum_data() + * + * gather accumulated network statistics + */ +static void appldata_get_net_sum_data(void *data) +{ + int i; + struct appldata_net_sum_data *net_data; + struct net_device *dev; + unsigned long rx_packets, tx_packets, rx_bytes, tx_bytes, rx_errors, + tx_errors, rx_dropped, tx_dropped, collisions; + + net_data = data; + net_data->sync_count_1++; + + i = 0; + rx_packets = 0; + tx_packets = 0; + rx_bytes = 0; + tx_bytes = 0; + rx_errors = 0; + tx_errors = 0; + rx_dropped = 0; + tx_dropped = 0; + collisions = 0; + + rcu_read_lock(); + for_each_netdev_rcu(&init_net, dev) { + const struct rtnl_link_stats64 *stats; + struct rtnl_link_stats64 temp; + + stats = dev_get_stats(dev, &temp); + rx_packets += stats->rx_packets; + tx_packets += stats->tx_packets; + rx_bytes += stats->rx_bytes; + tx_bytes += stats->tx_bytes; + rx_errors += stats->rx_errors; + tx_errors += stats->tx_errors; + rx_dropped += stats->rx_dropped; + tx_dropped += stats->tx_dropped; + collisions += stats->collisions; + i++; + } + rcu_read_unlock(); + + net_data->nr_interfaces = i; + net_data->rx_packets = rx_packets; + net_data->tx_packets = tx_packets; + net_data->rx_bytes = rx_bytes; + net_data->tx_bytes = tx_bytes; + net_data->rx_errors = rx_errors; + net_data->tx_errors = tx_errors; + net_data->rx_dropped = rx_dropped; + net_data->tx_dropped = tx_dropped; + net_data->collisions = collisions; + + net_data->timestamp = get_tod_clock(); + net_data->sync_count_2++; +} + + +static struct appldata_ops ops = { + .name = "net_sum", + .record_nr = APPLDATA_RECORD_NET_SUM_ID, + .size = sizeof(struct appldata_net_sum_data), + .callback = &appldata_get_net_sum_data, + .owner = THIS_MODULE, + .mod_lvl = {0xF0, 0xF0}, /* EBCDIC "00" */ +}; + + +/* + * appldata_net_init() + * + * init data, register ops + */ +static int __init appldata_net_init(void) +{ + int ret; + + ops.data = kzalloc(sizeof(struct appldata_net_sum_data), GFP_KERNEL); + if (!ops.data) + return -ENOMEM; + + ret = appldata_register_ops(&ops); + if (ret) + kfree(ops.data); + + return ret; +} + +/* + * appldata_net_exit() + * + * unregister ops + */ +static void __exit appldata_net_exit(void) +{ + appldata_unregister_ops(&ops); + kfree(ops.data); +} + + +module_init(appldata_net_init); +module_exit(appldata_net_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gerald Schaefer"); +MODULE_DESCRIPTION("Linux-VM Monitor Stream, accumulated network statistics"); diff --git a/arch/s390/appldata/appldata_os.c b/arch/s390/appldata/appldata_os.c new file mode 100644 index 000000000..69b23b25a --- /dev/null +++ b/arch/s390/appldata/appldata_os.c @@ -0,0 +1,218 @@ +/* + * Data gathering module for Linux-VM Monitor Stream, Stage 1. + * Collects misc. OS related data (CPU utilization, running processes). + * + * Copyright IBM Corp. 2003, 2006 + * + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> + */ + +#define KMSG_COMPONENT "appldata" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/errno.h> +#include <linux/kernel_stat.h> +#include <linux/netdevice.h> +#include <linux/sched.h> +#include <asm/appldata.h> +#include <asm/smp.h> + +#include "appldata.h" + + +#define LOAD_INT(x) ((x) >> FSHIFT) +#define LOAD_FRAC(x) LOAD_INT(((x) & (FIXED_1-1)) * 100) + +/* + * OS data + * + * This is accessed as binary data by z/VM. If changes to it can't be avoided, + * the structure version (product ID, see appldata_base.c) needs to be changed + * as well and all documentation and z/VM applications using it must be + * updated. + * + * The record layout is documented in the Linux for zSeries Device Drivers + * book: + * http://oss.software.ibm.com/developerworks/opensource/linux390/index.shtml + */ +struct appldata_os_per_cpu { + u32 per_cpu_user; /* timer ticks spent in user mode */ + u32 per_cpu_nice; /* ... spent with modified priority */ + u32 per_cpu_system; /* ... spent in kernel mode */ + u32 per_cpu_idle; /* ... spent in idle mode */ + + /* New in 2.6 */ + u32 per_cpu_irq; /* ... spent in interrupts */ + u32 per_cpu_softirq; /* ... spent in softirqs */ + u32 per_cpu_iowait; /* ... spent while waiting for I/O */ + + /* New in modification level 01 */ + u32 per_cpu_steal; /* ... stolen by hypervisor */ + u32 cpu_id; /* number of this CPU */ +} __attribute__((packed)); + +struct appldata_os_data { + u64 timestamp; + u32 sync_count_1; /* after VM collected the record data, */ + u32 sync_count_2; /* sync_count_1 and sync_count_2 should be the + same. If not, the record has been updated on + the Linux side while VM was collecting the + (possibly corrupt) data */ + + u32 nr_cpus; /* number of (virtual) CPUs */ + u32 per_cpu_size; /* size of the per-cpu data struct */ + u32 cpu_offset; /* offset of the first per-cpu data struct */ + + u32 nr_running; /* number of runnable threads */ + u32 nr_threads; /* number of threads */ + u32 avenrun[3]; /* average nr. of running processes during */ + /* the last 1, 5 and 15 minutes */ + + /* New in 2.6 */ + u32 nr_iowait; /* number of blocked threads + (waiting for I/O) */ + + /* per cpu data */ + struct appldata_os_per_cpu os_cpu[0]; +} __attribute__((packed)); + +static struct appldata_os_data *appldata_os_data; + +static struct appldata_ops ops = { + .name = "os", + .record_nr = APPLDATA_RECORD_OS_ID, + .owner = THIS_MODULE, + .mod_lvl = {0xF0, 0xF1}, /* EBCDIC "01" */ +}; + + +/* + * appldata_get_os_data() + * + * gather OS data + */ +static void appldata_get_os_data(void *data) +{ + int i, j, rc; + struct appldata_os_data *os_data; + unsigned int new_size; + + os_data = data; + os_data->sync_count_1++; + + os_data->nr_threads = nr_threads; + os_data->nr_running = nr_running(); + os_data->nr_iowait = nr_iowait(); + os_data->avenrun[0] = avenrun[0] + (FIXED_1/200); + os_data->avenrun[1] = avenrun[1] + (FIXED_1/200); + os_data->avenrun[2] = avenrun[2] + (FIXED_1/200); + + j = 0; + for_each_online_cpu(i) { + os_data->os_cpu[j].per_cpu_user = + cputime_to_jiffies(kcpustat_cpu(i).cpustat[CPUTIME_USER]); + os_data->os_cpu[j].per_cpu_nice = + cputime_to_jiffies(kcpustat_cpu(i).cpustat[CPUTIME_NICE]); + os_data->os_cpu[j].per_cpu_system = + cputime_to_jiffies(kcpustat_cpu(i).cpustat[CPUTIME_SYSTEM]); + os_data->os_cpu[j].per_cpu_idle = + cputime_to_jiffies(kcpustat_cpu(i).cpustat[CPUTIME_IDLE]); + os_data->os_cpu[j].per_cpu_irq = + cputime_to_jiffies(kcpustat_cpu(i).cpustat[CPUTIME_IRQ]); + os_data->os_cpu[j].per_cpu_softirq = + cputime_to_jiffies(kcpustat_cpu(i).cpustat[CPUTIME_SOFTIRQ]); + os_data->os_cpu[j].per_cpu_iowait = + cputime_to_jiffies(kcpustat_cpu(i).cpustat[CPUTIME_IOWAIT]); + os_data->os_cpu[j].per_cpu_steal = + cputime_to_jiffies(kcpustat_cpu(i).cpustat[CPUTIME_STEAL]); + os_data->os_cpu[j].cpu_id = i; + j++; + } + + os_data->nr_cpus = j; + + new_size = sizeof(struct appldata_os_data) + + (os_data->nr_cpus * sizeof(struct appldata_os_per_cpu)); + if (ops.size != new_size) { + if (ops.active) { + rc = appldata_diag(APPLDATA_RECORD_OS_ID, + APPLDATA_START_INTERVAL_REC, + (unsigned long) ops.data, new_size, + ops.mod_lvl); + if (rc != 0) + pr_err("Starting a new OS data collection " + "failed with rc=%d\n", rc); + + rc = appldata_diag(APPLDATA_RECORD_OS_ID, + APPLDATA_STOP_REC, + (unsigned long) ops.data, ops.size, + ops.mod_lvl); + if (rc != 0) + pr_err("Stopping a faulty OS data " + "collection failed with rc=%d\n", rc); + } + ops.size = new_size; + } + os_data->timestamp = get_tod_clock(); + os_data->sync_count_2++; +} + + +/* + * appldata_os_init() + * + * init data, register ops + */ +static int __init appldata_os_init(void) +{ + int rc, max_size; + + max_size = sizeof(struct appldata_os_data) + + (num_possible_cpus() * sizeof(struct appldata_os_per_cpu)); + if (max_size > APPLDATA_MAX_REC_SIZE) { + pr_err("Maximum OS record size %i exceeds the maximum " + "record size %i\n", max_size, APPLDATA_MAX_REC_SIZE); + rc = -ENOMEM; + goto out; + } + + appldata_os_data = kzalloc(max_size, GFP_KERNEL | GFP_DMA); + if (appldata_os_data == NULL) { + rc = -ENOMEM; + goto out; + } + + appldata_os_data->per_cpu_size = sizeof(struct appldata_os_per_cpu); + appldata_os_data->cpu_offset = offsetof(struct appldata_os_data, + os_cpu); + + ops.data = appldata_os_data; + ops.callback = &appldata_get_os_data; + rc = appldata_register_ops(&ops); + if (rc != 0) + kfree(appldata_os_data); +out: + return rc; +} + +/* + * appldata_os_exit() + * + * unregister ops + */ +static void __exit appldata_os_exit(void) +{ + appldata_unregister_ops(&ops); + kfree(appldata_os_data); +} + + +module_init(appldata_os_init); +module_exit(appldata_os_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Gerald Schaefer"); +MODULE_DESCRIPTION("Linux-VM Monitor Stream, OS statistics"); diff --git a/arch/s390/boot/.gitignore b/arch/s390/boot/.gitignore new file mode 100644 index 000000000..017d5912a --- /dev/null +++ b/arch/s390/boot/.gitignore @@ -0,0 +1,2 @@ +image +bzImage diff --git a/arch/s390/boot/Makefile b/arch/s390/boot/Makefile new file mode 100644 index 000000000..9a42ecec5 --- /dev/null +++ b/arch/s390/boot/Makefile @@ -0,0 +1,26 @@ +# +# Makefile for the linux s390-specific parts of the memory manager. +# + +COMPILE_VERSION := __linux_compile_version_id__`hostname | \ + tr -c '[0-9A-Za-z]' '_'`__`date | \ + tr -c '[0-9A-Za-z]' '_'`_t + +ccflags-y := -DCOMPILE_VERSION=$(COMPILE_VERSION) -gstabs -I. + +targets := image +targets += bzImage +subdir- := compressed + +$(obj)/image: vmlinux FORCE + $(call if_changed,objcopy) + +$(obj)/bzImage: $(obj)/compressed/vmlinux FORCE + $(call if_changed,objcopy) + +$(obj)/compressed/vmlinux: FORCE + $(Q)$(MAKE) $(build)=$(obj)/compressed $@ + +install: $(CONFIGURE) $(obj)/bzImage + sh -x $(srctree)/$(obj)/install.sh $(KERNELRELEASE) $(obj)/bzImage \ + System.map "$(INSTALL_PATH)" diff --git a/arch/s390/boot/compressed/.gitignore b/arch/s390/boot/compressed/.gitignore new file mode 100644 index 000000000..ae06b9b4c --- /dev/null +++ b/arch/s390/boot/compressed/.gitignore @@ -0,0 +1,3 @@ +sizes.h +vmlinux +vmlinux.lds diff --git a/arch/s390/boot/compressed/Makefile b/arch/s390/boot/compressed/Makefile new file mode 100644 index 000000000..d4788111c --- /dev/null +++ b/arch/s390/boot/compressed/Makefile @@ -0,0 +1,69 @@ +# +# linux/arch/s390/boot/compressed/Makefile +# +# create a compressed vmlinux image from the original vmlinux +# + +targets := vmlinux.lds vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 +targets += vmlinux.bin.xz vmlinux.bin.lzma vmlinux.bin.lzo vmlinux.bin.lz4 +targets += misc.o piggy.o sizes.h head.o + +KBUILD_CFLAGS := -m64 -D__KERNEL__ $(LINUX_INCLUDE) -O2 +KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING +KBUILD_CFLAGS += $(cflags-y) -fno-delete-null-pointer-checks +KBUILD_CFLAGS += $(call cc-option,-mpacked-stack) +KBUILD_CFLAGS += $(call cc-option,-ffreestanding) + +GCOV_PROFILE := n + +OBJECTS := $(addprefix $(objtree)/arch/s390/kernel/, head.o sclp.o ebcdic.o) +OBJECTS += $(obj)/head.o $(obj)/misc.o $(obj)/piggy.o + +LDFLAGS_vmlinux := --oformat $(LD_BFD) -e startup -T +$(obj)/vmlinux: $(obj)/vmlinux.lds $(OBJECTS) + $(call if_changed,ld) + @: + +sed-sizes := -e 's/^\([0-9a-fA-F]*\) . \(__bss_start\|_end\)$$/\#define SZ\2 0x\1/p' + +quiet_cmd_sizes = GEN $@ + cmd_sizes = $(NM) $< | sed -n $(sed-sizes) > $@ + +$(obj)/sizes.h: vmlinux + $(call if_changed,sizes) + +AFLAGS_head.o += -I$(obj) +$(obj)/head.o: $(obj)/sizes.h + +CFLAGS_misc.o += -I$(obj) +$(obj)/misc.o: $(obj)/sizes.h + +OBJCOPYFLAGS_vmlinux.bin := -R .comment -S +$(obj)/vmlinux.bin: vmlinux + $(call if_changed,objcopy) + +vmlinux.bin.all-y := $(obj)/vmlinux.bin + +suffix-$(CONFIG_KERNEL_GZIP) := gz +suffix-$(CONFIG_KERNEL_BZIP2) := bz2 +suffix-$(CONFIG_KERNEL_LZ4) := lz4 +suffix-$(CONFIG_KERNEL_LZMA) := lzma +suffix-$(CONFIG_KERNEL_LZO) := lzo +suffix-$(CONFIG_KERNEL_XZ) := xz + +$(obj)/vmlinux.bin.gz: $(vmlinux.bin.all-y) + $(call if_changed,gzip) +$(obj)/vmlinux.bin.bz2: $(vmlinux.bin.all-y) + $(call if_changed,bzip2) +$(obj)/vmlinux.bin.lz4: $(vmlinux.bin.all-y) + $(call if_changed,lz4) +$(obj)/vmlinux.bin.lzma: $(vmlinux.bin.all-y) + $(call if_changed,lzma) +$(obj)/vmlinux.bin.lzo: $(vmlinux.bin.all-y) + $(call if_changed,lzo) +$(obj)/vmlinux.bin.xz: $(vmlinux.bin.all-y) + $(call if_changed,xzkern) + +LDFLAGS_piggy.o := -r --format binary --oformat $(LD_BFD) -T +$(obj)/piggy.o: $(obj)/vmlinux.scr $(obj)/vmlinux.bin.$(suffix-y) + $(call if_changed,ld) diff --git a/arch/s390/boot/compressed/head.S b/arch/s390/boot/compressed/head.S new file mode 100644 index 000000000..f86a4eef2 --- /dev/null +++ b/arch/s390/boot/compressed/head.S @@ -0,0 +1,48 @@ +/* + * Startup glue code to uncompress the kernel + * + * Copyright IBM Corp. 2010 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#include <linux/init.h> +#include <linux/linkage.h> +#include <asm/asm-offsets.h> +#include <asm/thread_info.h> +#include <asm/page.h> +#include "sizes.h" + +__HEAD +ENTRY(startup_continue) + basr %r13,0 # get base +.LPG1: + # setup stack + lg %r15,.Lstack-.LPG1(%r13) + aghi %r15,-160 + brasl %r14,decompress_kernel + # setup registers for memory mover & branch to target + lgr %r4,%r2 + lg %r2,.Loffset-.LPG1(%r13) + la %r4,0(%r2,%r4) + lg %r3,.Lmvsize-.LPG1(%r13) + lgr %r5,%r3 + # move the memory mover someplace safe + la %r1,0x200 + mvc 0(mover_end-mover,%r1),mover-.LPG1(%r13) + # decompress image is started at 0x11000 + lgr %r6,%r2 + br %r1 +mover: + mvcle %r2,%r4,0 + jo mover + br %r6 +mover_end: + + .align 8 +.Lstack: + .quad 0x8000 + (1<<(PAGE_SHIFT+THREAD_ORDER)) +.Loffset: + .quad 0x11000 +.Lmvsize: + .quad SZ__bss_start diff --git a/arch/s390/boot/compressed/misc.c b/arch/s390/boot/compressed/misc.c new file mode 100644 index 000000000..42506b371 --- /dev/null +++ b/arch/s390/boot/compressed/misc.c @@ -0,0 +1,174 @@ +/* + * Definitions and wrapper functions for kernel decompressor + * + * Copyright IBM Corp. 2010 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#include <asm/uaccess.h> +#include <asm/page.h> +#include <asm/sclp.h> +#include <asm/ipl.h> +#include "sizes.h" + +/* + * gzip declarations + */ +#define STATIC static + +#undef memset +#undef memcpy +#undef memmove +#define memmove memmove +#define memzero(s, n) memset((s), 0, (n)) + +/* Symbols defined by linker scripts */ +extern char input_data[]; +extern int input_len; +extern char _text, _end; +extern char _bss, _ebss; + +static void error(char *m); + +static unsigned long free_mem_ptr; +static unsigned long free_mem_end_ptr; + +#ifdef CONFIG_HAVE_KERNEL_BZIP2 +#define HEAP_SIZE 0x400000 +#else +#define HEAP_SIZE 0x10000 +#endif + +#ifdef CONFIG_KERNEL_GZIP +#include "../../../../lib/decompress_inflate.c" +#endif + +#ifdef CONFIG_KERNEL_BZIP2 +#include "../../../../lib/decompress_bunzip2.c" +#endif + +#ifdef CONFIG_KERNEL_LZ4 +#include "../../../../lib/decompress_unlz4.c" +#endif + +#ifdef CONFIG_KERNEL_LZMA +#include "../../../../lib/decompress_unlzma.c" +#endif + +#ifdef CONFIG_KERNEL_LZO +#include "../../../../lib/decompress_unlzo.c" +#endif + +#ifdef CONFIG_KERNEL_XZ +#include "../../../../lib/decompress_unxz.c" +#endif + +static int puts(const char *s) +{ + _sclp_print_early(s); + return 0; +} + +void *memset(void *s, int c, size_t n) +{ + char *xs; + + xs = s; + while (n--) + *xs++ = c; + return s; +} + +void *memcpy(void *dest, const void *src, size_t n) +{ + const char *s = src; + char *d = dest; + + while (n--) + *d++ = *s++; + return dest; +} + +void *memmove(void *dest, const void *src, size_t n) +{ + const char *s = src; + char *d = dest; + + if (d <= s) { + while (n--) + *d++ = *s++; + } else { + d += n; + s += n; + while (n--) + *--d = *--s; + } + return dest; +} + +static void error(char *x) +{ + unsigned long long psw = 0x000a0000deadbeefULL; + + puts("\n\n"); + puts(x); + puts("\n\n -- System halted"); + + asm volatile("lpsw %0" : : "Q" (psw)); +} + +/* + * Safe guard the ipl parameter block against a memory area that will be + * overwritten. The validity check for the ipl parameter block is complex + * (see cio_get_iplinfo and ipl_save_parameters) but if the pointer to + * the ipl parameter block intersects with the passed memory area we can + * safely assume that we can read from that memory. In that case just copy + * the memory to IPL_PARMBLOCK_ORIGIN even if there is no ipl parameter + * block. + */ +static void check_ipl_parmblock(void *start, unsigned long size) +{ + void *src, *dst; + + src = (void *)(unsigned long) S390_lowcore.ipl_parmblock_ptr; + if (src + PAGE_SIZE <= start || src >= start + size) + return; + dst = (void *) IPL_PARMBLOCK_ORIGIN; + memmove(dst, src, PAGE_SIZE); + S390_lowcore.ipl_parmblock_ptr = IPL_PARMBLOCK_ORIGIN; +} + +unsigned long decompress_kernel(void) +{ + unsigned long output_addr; + unsigned char *output; + + output_addr = ((unsigned long) &_end + HEAP_SIZE + 4095UL) & -4096UL; + check_ipl_parmblock((void *) 0, output_addr + SZ__bss_start); + memset(&_bss, 0, &_ebss - &_bss); + free_mem_ptr = (unsigned long)&_end; + free_mem_end_ptr = free_mem_ptr + HEAP_SIZE; + output = (unsigned char *) output_addr; + +#ifdef CONFIG_BLK_DEV_INITRD + /* + * Move the initrd right behind the end of the decompressed + * kernel image. + */ + if (INITRD_START && INITRD_SIZE && + INITRD_START < (unsigned long) output + SZ__bss_start) { + check_ipl_parmblock(output + SZ__bss_start, + INITRD_START + INITRD_SIZE); + memmove(output + SZ__bss_start, + (void *) INITRD_START, INITRD_SIZE); + INITRD_START = (unsigned long) output + SZ__bss_start; + } +#endif + + puts("Uncompressing Linux... "); + decompress(input_data, input_len, NULL, NULL, output, NULL, error); + puts("Ok, booting the kernel.\n"); + return (unsigned long) output; +} + diff --git a/arch/s390/boot/compressed/vmlinux.lds.S b/arch/s390/boot/compressed/vmlinux.lds.S new file mode 100644 index 000000000..747735f83 --- /dev/null +++ b/arch/s390/boot/compressed/vmlinux.lds.S @@ -0,0 +1,50 @@ +#include <asm-generic/vmlinux.lds.h> + +OUTPUT_FORMAT("elf64-s390", "elf64-s390", "elf64-s390") +OUTPUT_ARCH(s390:64-bit) + +ENTRY(startup) + +SECTIONS +{ + /* Be careful parts of head_64.S assume startup_32 is at + * address 0. + */ + . = 0; + .head.text : { + _head = . ; + HEAD_TEXT + _ehead = . ; + } + .rodata.compressed : { + *(.rodata.compressed) + } + .text : { + _text = .; /* Text */ + *(.text) + *(.text.*) + _etext = . ; + } + .rodata : { + _rodata = . ; + *(.rodata) /* read-only data */ + *(.rodata.*) + _erodata = . ; + } + .data : { + _data = . ; + *(.data) + *(.data.*) + _edata = . ; + } + . = ALIGN(256); + .bss : { + _bss = . ; + *(.bss) + *(.bss.*) + *(COMMON) + . = ALIGN(8); /* For convenience during zeroing */ + _ebss = .; + } + _end = .; +} diff --git a/arch/s390/boot/compressed/vmlinux.scr b/arch/s390/boot/compressed/vmlinux.scr new file mode 100644 index 000000000..f02382ae5 --- /dev/null +++ b/arch/s390/boot/compressed/vmlinux.scr @@ -0,0 +1,10 @@ +SECTIONS +{ + .rodata.compressed : { + input_len = .; + LONG(input_data_end - input_data) input_data = .; + *(.data) + output_len = . - 4; + input_data_end = .; + } +} diff --git a/arch/s390/boot/install.sh b/arch/s390/boot/install.sh new file mode 100644 index 000000000..aed306969 --- /dev/null +++ b/arch/s390/boot/install.sh @@ -0,0 +1,38 @@ +#!/bin/sh +# +# arch/s390x/boot/install.sh +# +# This file is subject to the terms and conditions of the GNU General Public +# License. See the file "COPYING" in the main directory of this archive +# for more details. +# +# Copyright (C) 1995 by Linus Torvalds +# +# Adapted from code in arch/i386/boot/Makefile by H. Peter Anvin +# +# "make install" script for s390 architecture +# +# Arguments: +# $1 - kernel version +# $2 - kernel image file +# $3 - kernel map file +# $4 - default install path (blank if root directory) +# + +# User may have a custom install script + +if [ -x ~/bin/${INSTALLKERNEL} ]; then exec ~/bin/${INSTALLKERNEL} "$@"; fi +if [ -x /sbin/${INSTALLKERNEL} ]; then exec /sbin/${INSTALLKERNEL} "$@"; fi + +# Default install - same as make zlilo + +if [ -f $4/vmlinuz ]; then + mv $4/vmlinuz $4/vmlinuz.old +fi + +if [ -f $4/System.map ]; then + mv $4/System.map $4/System.old +fi + +cat $2 > $4/vmlinuz +cp $3 $4/System.map diff --git a/arch/s390/configs/default_defconfig b/arch/s390/configs/default_defconfig new file mode 100644 index 000000000..64707750c --- /dev/null +++ b/arch/s390/configs/default_defconfig @@ -0,0 +1,678 @@ +CONFIG_SYSVIPC=y +CONFIG_POSIX_MQUEUE=y +CONFIG_FHANDLE=y +CONFIG_AUDIT=y +CONFIG_NO_HZ=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_BSD_PROCESS_ACCT=y +CONFIG_BSD_PROCESS_ACCT_V3=y +CONFIG_TASKSTATS=y +CONFIG_TASK_DELAY_ACCT=y +CONFIG_TASK_XACCT=y +CONFIG_TASK_IO_ACCOUNTING=y +CONFIG_RCU_FAST_NO_HZ=y +CONFIG_IKCONFIG=y +CONFIG_IKCONFIG_PROC=y +CONFIG_CGROUP_FREEZER=y +CONFIG_CGROUP_DEVICE=y +CONFIG_CPUSETS=y +CONFIG_CGROUP_CPUACCT=y +CONFIG_RESOURCE_COUNTERS=y +CONFIG_CGROUP_PERF=y +CONFIG_CFS_BANDWIDTH=y +CONFIG_RT_GROUP_SCHED=y +CONFIG_BLK_CGROUP=y +CONFIG_SCHED_AUTOGROUP=y +CONFIG_BLK_DEV_INITRD=y +# CONFIG_COMPAT_BRK is not set +CONFIG_PROFILING=y +CONFIG_OPROFILE=m +CONFIG_KPROBES=y +CONFIG_JUMP_LABEL=y +CONFIG_MODULES=y +CONFIG_MODULE_FORCE_LOAD=y +CONFIG_MODULE_UNLOAD=y +CONFIG_MODULE_FORCE_UNLOAD=y +CONFIG_MODVERSIONS=y +CONFIG_MODULE_SRCVERSION_ALL=y +CONFIG_BLK_DEV_THROTTLING=y +CONFIG_PARTITION_ADVANCED=y +CONFIG_IBM_PARTITION=y +CONFIG_BSD_DISKLABEL=y +CONFIG_MINIX_SUBPARTITION=y +CONFIG_SOLARIS_X86_PARTITION=y +CONFIG_UNIXWARE_DISKLABEL=y +CONFIG_CFQ_GROUP_IOSCHED=y +CONFIG_DEFAULT_DEADLINE=y +CONFIG_MARCH_Z196=y +CONFIG_TUNE_ZEC12=y +CONFIG_NR_CPUS=256 +CONFIG_PREEMPT=y +CONFIG_HZ_100=y +CONFIG_MEMORY_HOTPLUG=y +CONFIG_MEMORY_HOTREMOVE=y +CONFIG_KSM=y +CONFIG_TRANSPARENT_HUGEPAGE=y +CONFIG_PCI=y +CONFIG_PCI_DEBUG=y +CONFIG_HOTPLUG_PCI=y +CONFIG_HOTPLUG_PCI_S390=y +CONFIG_CHSC_SCH=y +CONFIG_CRASH_DUMP=y +# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set +CONFIG_BINFMT_MISC=m +CONFIG_HIBERNATION=y +CONFIG_NET=y +CONFIG_PACKET=y +CONFIG_PACKET_DIAG=m +CONFIG_UNIX=y +CONFIG_UNIX_DIAG=m +CONFIG_XFRM_USER=m +CONFIG_NET_KEY=m +CONFIG_INET=y +CONFIG_IP_MULTICAST=y +CONFIG_IP_ADVANCED_ROUTER=y +CONFIG_IP_MULTIPLE_TABLES=y +CONFIG_IP_ROUTE_MULTIPATH=y +CONFIG_IP_ROUTE_VERBOSE=y +CONFIG_NET_IPIP=m +CONFIG_NET_IPGRE_DEMUX=m +CONFIG_NET_IPGRE=m +CONFIG_NET_IPGRE_BROADCAST=y +CONFIG_IP_MROUTE=y +CONFIG_IP_MROUTE_MULTIPLE_TABLES=y +CONFIG_IP_PIMSM_V1=y +CONFIG_IP_PIMSM_V2=y +CONFIG_SYN_COOKIES=y +CONFIG_NET_IPVTI=m +CONFIG_INET_AH=m +CONFIG_INET_ESP=m +CONFIG_INET_IPCOMP=m +CONFIG_INET_XFRM_MODE_TRANSPORT=m +CONFIG_INET_XFRM_MODE_TUNNEL=m +CONFIG_INET_XFRM_MODE_BEET=m +CONFIG_INET_DIAG=m +CONFIG_INET_UDP_DIAG=m +CONFIG_TCP_CONG_ADVANCED=y +CONFIG_TCP_CONG_HSTCP=m +CONFIG_TCP_CONG_HYBLA=m +CONFIG_TCP_CONG_SCALABLE=m +CONFIG_TCP_CONG_LP=m +CONFIG_TCP_CONG_VENO=m +CONFIG_TCP_CONG_YEAH=m +CONFIG_TCP_CONG_ILLINOIS=m +CONFIG_IPV6=y +CONFIG_IPV6_ROUTER_PREF=y +CONFIG_INET6_AH=m +CONFIG_INET6_ESP=m +CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_MIP6=m +CONFIG_INET6_XFRM_MODE_TRANSPORT=m +CONFIG_INET6_XFRM_MODE_TUNNEL=m +CONFIG_INET6_XFRM_MODE_BEET=m +CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION=m +CONFIG_IPV6_VTI=m +CONFIG_IPV6_SIT=m +CONFIG_IPV6_GRE=m +CONFIG_IPV6_MULTIPLE_TABLES=y +CONFIG_IPV6_SUBTREES=y +CONFIG_NETFILTER=y +CONFIG_NF_CONNTRACK=m +CONFIG_NF_CONNTRACK_SECMARK=y +CONFIG_NF_CONNTRACK_EVENTS=y +CONFIG_NF_CONNTRACK_TIMEOUT=y +CONFIG_NF_CONNTRACK_TIMESTAMP=y +CONFIG_NF_CT_PROTO_DCCP=m +CONFIG_NF_CT_PROTO_UDPLITE=m +CONFIG_NF_CONNTRACK_AMANDA=m +CONFIG_NF_CONNTRACK_FTP=m +CONFIG_NF_CONNTRACK_H323=m +CONFIG_NF_CONNTRACK_IRC=m +CONFIG_NF_CONNTRACK_NETBIOS_NS=m +CONFIG_NF_CONNTRACK_SNMP=m +CONFIG_NF_CONNTRACK_PPTP=m +CONFIG_NF_CONNTRACK_SANE=m +CONFIG_NF_CONNTRACK_SIP=m +CONFIG_NF_CONNTRACK_TFTP=m +CONFIG_NF_CT_NETLINK=m +CONFIG_NF_CT_NETLINK_TIMEOUT=m +CONFIG_NF_TABLES=m +CONFIG_NFT_EXTHDR=m +CONFIG_NFT_META=m +CONFIG_NFT_CT=m +CONFIG_NFT_RBTREE=m +CONFIG_NFT_HASH=m +CONFIG_NFT_COUNTER=m +CONFIG_NFT_LOG=m +CONFIG_NFT_LIMIT=m +CONFIG_NFT_NAT=m +CONFIG_NFT_COMPAT=m +CONFIG_NETFILTER_XT_SET=m +CONFIG_NETFILTER_XT_TARGET_AUDIT=m +CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m +CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m +CONFIG_NETFILTER_XT_TARGET_CONNMARK=m +CONFIG_NETFILTER_XT_TARGET_CONNSECMARK=m +CONFIG_NETFILTER_XT_TARGET_CT=m +CONFIG_NETFILTER_XT_TARGET_DSCP=m +CONFIG_NETFILTER_XT_TARGET_HMARK=m +CONFIG_NETFILTER_XT_TARGET_IDLETIMER=m +CONFIG_NETFILTER_XT_TARGET_LOG=m +CONFIG_NETFILTER_XT_TARGET_MARK=m +CONFIG_NETFILTER_XT_TARGET_NFLOG=m +CONFIG_NETFILTER_XT_TARGET_NFQUEUE=m +CONFIG_NETFILTER_XT_TARGET_TEE=m +CONFIG_NETFILTER_XT_TARGET_TPROXY=m +CONFIG_NETFILTER_XT_TARGET_TRACE=m +CONFIG_NETFILTER_XT_TARGET_SECMARK=m +CONFIG_NETFILTER_XT_TARGET_TCPMSS=m +CONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP=m +CONFIG_NETFILTER_XT_MATCH_ADDRTYPE=m +CONFIG_NETFILTER_XT_MATCH_BPF=m +CONFIG_NETFILTER_XT_MATCH_CLUSTER=m +CONFIG_NETFILTER_XT_MATCH_COMMENT=m +CONFIG_NETFILTER_XT_MATCH_CONNBYTES=m +CONFIG_NETFILTER_XT_MATCH_CONNLABEL=m +CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=m +CONFIG_NETFILTER_XT_MATCH_CONNMARK=m +CONFIG_NETFILTER_XT_MATCH_CONNTRACK=m +CONFIG_NETFILTER_XT_MATCH_CPU=m +CONFIG_NETFILTER_XT_MATCH_DCCP=m +CONFIG_NETFILTER_XT_MATCH_DEVGROUP=m +CONFIG_NETFILTER_XT_MATCH_DSCP=m +CONFIG_NETFILTER_XT_MATCH_ESP=m +CONFIG_NETFILTER_XT_MATCH_HASHLIMIT=m +CONFIG_NETFILTER_XT_MATCH_HELPER=m +CONFIG_NETFILTER_XT_MATCH_IPRANGE=m +CONFIG_NETFILTER_XT_MATCH_IPVS=m +CONFIG_NETFILTER_XT_MATCH_LENGTH=m +CONFIG_NETFILTER_XT_MATCH_LIMIT=m +CONFIG_NETFILTER_XT_MATCH_MAC=m +CONFIG_NETFILTER_XT_MATCH_MARK=m +CONFIG_NETFILTER_XT_MATCH_MULTIPORT=m +CONFIG_NETFILTER_XT_MATCH_NFACCT=m +CONFIG_NETFILTER_XT_MATCH_OSF=m +CONFIG_NETFILTER_XT_MATCH_OWNER=m +CONFIG_NETFILTER_XT_MATCH_POLICY=m +CONFIG_NETFILTER_XT_MATCH_PHYSDEV=m +CONFIG_NETFILTER_XT_MATCH_PKTTYPE=m +CONFIG_NETFILTER_XT_MATCH_QUOTA=m +CONFIG_NETFILTER_XT_MATCH_RATEEST=m +CONFIG_NETFILTER_XT_MATCH_REALM=m +CONFIG_NETFILTER_XT_MATCH_RECENT=m +CONFIG_NETFILTER_XT_MATCH_SOCKET=m +CONFIG_NETFILTER_XT_MATCH_STATE=m +CONFIG_NETFILTER_XT_MATCH_STATISTIC=m +CONFIG_NETFILTER_XT_MATCH_STRING=m +CONFIG_NETFILTER_XT_MATCH_TCPMSS=m +CONFIG_NETFILTER_XT_MATCH_TIME=m +CONFIG_NETFILTER_XT_MATCH_U32=m +CONFIG_IP_SET=m +CONFIG_IP_SET_BITMAP_IP=m +CONFIG_IP_SET_BITMAP_IPMAC=m +CONFIG_IP_SET_BITMAP_PORT=m +CONFIG_IP_SET_HASH_IP=m +CONFIG_IP_SET_HASH_IPPORT=m +CONFIG_IP_SET_HASH_IPPORTIP=m +CONFIG_IP_SET_HASH_IPPORTNET=m +CONFIG_IP_SET_HASH_NETPORTNET=m +CONFIG_IP_SET_HASH_NET=m +CONFIG_IP_SET_HASH_NETNET=m +CONFIG_IP_SET_HASH_NETPORT=m +CONFIG_IP_SET_HASH_NETIFACE=m +CONFIG_IP_SET_LIST_SET=m +CONFIG_IP_VS=m +CONFIG_IP_VS_PROTO_TCP=y +CONFIG_IP_VS_PROTO_UDP=y +CONFIG_IP_VS_PROTO_ESP=y +CONFIG_IP_VS_PROTO_AH=y +CONFIG_IP_VS_RR=m +CONFIG_IP_VS_WRR=m +CONFIG_IP_VS_LC=m +CONFIG_IP_VS_WLC=m +CONFIG_IP_VS_LBLC=m +CONFIG_IP_VS_LBLCR=m +CONFIG_IP_VS_DH=m +CONFIG_IP_VS_SH=m +CONFIG_IP_VS_SED=m +CONFIG_IP_VS_NQ=m +CONFIG_IP_VS_FTP=m +CONFIG_IP_VS_PE_SIP=m +CONFIG_NF_CONNTRACK_IPV4=m +# CONFIG_NF_CONNTRACK_PROC_COMPAT is not set +CONFIG_NF_TABLES_IPV4=m +CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_CHAIN_NAT_IPV4=m +CONFIG_NF_TABLES_ARP=m +CONFIG_NF_NAT_IPV4=m +CONFIG_IP_NF_IPTABLES=m +CONFIG_IP_NF_MATCH_AH=m +CONFIG_IP_NF_MATCH_ECN=m +CONFIG_IP_NF_MATCH_RPFILTER=m +CONFIG_IP_NF_MATCH_TTL=m +CONFIG_IP_NF_FILTER=m +CONFIG_IP_NF_TARGET_REJECT=m +CONFIG_IP_NF_MANGLE=m +CONFIG_IP_NF_TARGET_CLUSTERIP=m +CONFIG_IP_NF_TARGET_ECN=m +CONFIG_IP_NF_TARGET_TTL=m +CONFIG_IP_NF_RAW=m +CONFIG_IP_NF_SECURITY=m +CONFIG_IP_NF_ARPTABLES=m +CONFIG_IP_NF_ARPFILTER=m +CONFIG_IP_NF_ARP_MANGLE=m +CONFIG_NF_CONNTRACK_IPV6=m +CONFIG_NF_TABLES_IPV6=m +CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_CHAIN_NAT_IPV6=m +CONFIG_NF_NAT_IPV6=m +CONFIG_IP6_NF_IPTABLES=m +CONFIG_IP6_NF_MATCH_AH=m +CONFIG_IP6_NF_MATCH_EUI64=m +CONFIG_IP6_NF_MATCH_FRAG=m +CONFIG_IP6_NF_MATCH_OPTS=m +CONFIG_IP6_NF_MATCH_HL=m +CONFIG_IP6_NF_MATCH_IPV6HEADER=m +CONFIG_IP6_NF_MATCH_MH=m +CONFIG_IP6_NF_MATCH_RPFILTER=m +CONFIG_IP6_NF_MATCH_RT=m +CONFIG_IP6_NF_TARGET_HL=m +CONFIG_IP6_NF_FILTER=m +CONFIG_IP6_NF_TARGET_REJECT=m +CONFIG_IP6_NF_MANGLE=m +CONFIG_IP6_NF_RAW=m +CONFIG_IP6_NF_SECURITY=m +CONFIG_NF_TABLES_BRIDGE=m +CONFIG_NET_SCTPPROBE=m +CONFIG_RDS=m +CONFIG_RDS_RDMA=m +CONFIG_RDS_TCP=m +CONFIG_RDS_DEBUG=y +CONFIG_L2TP=m +CONFIG_L2TP_DEBUGFS=m +CONFIG_L2TP_V3=y +CONFIG_L2TP_IP=m +CONFIG_L2TP_ETH=m +CONFIG_BRIDGE=m +CONFIG_VLAN_8021Q=m +CONFIG_VLAN_8021Q_GVRP=y +CONFIG_NET_SCHED=y +CONFIG_NET_SCH_CBQ=m +CONFIG_NET_SCH_HTB=m +CONFIG_NET_SCH_HFSC=m +CONFIG_NET_SCH_PRIO=m +CONFIG_NET_SCH_MULTIQ=m +CONFIG_NET_SCH_RED=m +CONFIG_NET_SCH_SFB=m +CONFIG_NET_SCH_SFQ=m +CONFIG_NET_SCH_TEQL=m +CONFIG_NET_SCH_TBF=m +CONFIG_NET_SCH_GRED=m +CONFIG_NET_SCH_DSMARK=m +CONFIG_NET_SCH_NETEM=m +CONFIG_NET_SCH_DRR=m +CONFIG_NET_SCH_MQPRIO=m +CONFIG_NET_SCH_CHOKE=m +CONFIG_NET_SCH_QFQ=m +CONFIG_NET_SCH_CODEL=m +CONFIG_NET_SCH_FQ_CODEL=m +CONFIG_NET_SCH_INGRESS=m +CONFIG_NET_SCH_PLUG=m +CONFIG_NET_CLS_BASIC=m +CONFIG_NET_CLS_TCINDEX=m +CONFIG_NET_CLS_ROUTE4=m +CONFIG_NET_CLS_FW=m +CONFIG_NET_CLS_U32=m +CONFIG_CLS_U32_PERF=y +CONFIG_CLS_U32_MARK=y +CONFIG_NET_CLS_RSVP=m +CONFIG_NET_CLS_RSVP6=m +CONFIG_NET_CLS_FLOW=m +CONFIG_NET_CLS_CGROUP=y +CONFIG_NET_CLS_BPF=m +CONFIG_NET_CLS_ACT=y +CONFIG_NET_ACT_POLICE=m +CONFIG_NET_ACT_GACT=m +CONFIG_GACT_PROB=y +CONFIG_NET_ACT_MIRRED=m +CONFIG_NET_ACT_IPT=m +CONFIG_NET_ACT_NAT=m +CONFIG_NET_ACT_PEDIT=m +CONFIG_NET_ACT_SIMP=m +CONFIG_NET_ACT_SKBEDIT=m +CONFIG_NET_ACT_CSUM=m +CONFIG_DNS_RESOLVER=y +CONFIG_BPF_JIT=y +CONFIG_NET_PKTGEN=m +CONFIG_NET_TCPPROBE=m +CONFIG_DEVTMPFS=y +CONFIG_CONNECTOR=y +CONFIG_BLK_DEV_LOOP=m +CONFIG_BLK_DEV_CRYPTOLOOP=m +CONFIG_BLK_DEV_NBD=m +CONFIG_BLK_DEV_OSD=m +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_SIZE=32768 +CONFIG_BLK_DEV_XIP=y +CONFIG_CDROM_PKTCDVD=m +CONFIG_ATA_OVER_ETH=m +CONFIG_VIRTIO_BLK=y +CONFIG_ENCLOSURE_SERVICES=m +CONFIG_RAID_ATTRS=m +CONFIG_SCSI=y +CONFIG_BLK_DEV_SD=y +CONFIG_CHR_DEV_ST=m +CONFIG_CHR_DEV_OSST=m +CONFIG_BLK_DEV_SR=m +CONFIG_CHR_DEV_SG=y +CONFIG_CHR_DEV_SCH=m +CONFIG_SCSI_ENCLOSURE=m +CONFIG_SCSI_CONSTANTS=y +CONFIG_SCSI_LOGGING=y +CONFIG_SCSI_SPI_ATTRS=m +CONFIG_SCSI_FC_ATTRS=y +CONFIG_SCSI_SAS_LIBSAS=m +CONFIG_SCSI_SRP_ATTRS=m +CONFIG_ISCSI_TCP=m +CONFIG_SCSI_DEBUG=m +CONFIG_ZFCP=y +CONFIG_SCSI_VIRTIO=m +CONFIG_SCSI_DH=m +CONFIG_SCSI_DH_RDAC=m +CONFIG_SCSI_DH_HP_SW=m +CONFIG_SCSI_DH_EMC=m +CONFIG_SCSI_DH_ALUA=m +CONFIG_SCSI_OSD_INITIATOR=m +CONFIG_SCSI_OSD_ULD=m +CONFIG_MD=y +CONFIG_BLK_DEV_MD=y +CONFIG_MD_LINEAR=m +CONFIG_MD_RAID0=m +CONFIG_MD_MULTIPATH=m +CONFIG_MD_FAULTY=m +CONFIG_BLK_DEV_DM=m +CONFIG_DM_CRYPT=m +CONFIG_DM_SNAPSHOT=m +CONFIG_DM_MIRROR=m +CONFIG_DM_LOG_USERSPACE=m +CONFIG_DM_RAID=m +CONFIG_DM_ZERO=m +CONFIG_DM_MULTIPATH=m +CONFIG_DM_MULTIPATH_QL=m +CONFIG_DM_MULTIPATH_ST=m +CONFIG_DM_DELAY=m +CONFIG_DM_UEVENT=y +CONFIG_DM_FLAKEY=m +CONFIG_DM_VERITY=m +CONFIG_DM_SWITCH=m +CONFIG_NETDEVICES=y +CONFIG_BONDING=m +CONFIG_DUMMY=m +CONFIG_EQUALIZER=m +CONFIG_IFB=m +CONFIG_MACVLAN=m +CONFIG_MACVTAP=m +CONFIG_VXLAN=m +CONFIG_TUN=m +CONFIG_VETH=m +CONFIG_VIRTIO_NET=m +CONFIG_NLMON=m +CONFIG_VHOST_NET=m +# CONFIG_NET_VENDOR_ARC is not set +# CONFIG_NET_VENDOR_CHELSIO is not set +# CONFIG_NET_VENDOR_INTEL is not set +# CONFIG_NET_VENDOR_MARVELL is not set +CONFIG_MLX4_EN=m +# CONFIG_NET_VENDOR_NATSEMI is not set +CONFIG_PPP=m +CONFIG_PPP_BSDCOMP=m +CONFIG_PPP_DEFLATE=m +CONFIG_PPP_MPPE=m +CONFIG_PPPOE=m +CONFIG_PPTP=m +CONFIG_PPPOL2TP=m +CONFIG_PPP_ASYNC=m +CONFIG_PPP_SYNC_TTY=m +# CONFIG_INPUT_MOUSEDEV_PSAUX is not set +# CONFIG_INPUT_KEYBOARD is not set +# CONFIG_INPUT_MOUSE is not set +# CONFIG_SERIO is not set +CONFIG_DEVPTS_MULTIPLE_INSTANCES=y +CONFIG_LEGACY_PTY_COUNT=0 +CONFIG_HW_RANDOM_VIRTIO=m +CONFIG_RAW_DRIVER=m +CONFIG_HANGCHECK_TIMER=m +CONFIG_TN3270_FS=y +CONFIG_WATCHDOG=y +CONFIG_WATCHDOG_NOWAYOUT=y +CONFIG_SOFT_WATCHDOG=m +CONFIG_DIAG288_WATCHDOG=m +# CONFIG_HID is not set +# CONFIG_USB_SUPPORT is not set +CONFIG_INFINIBAND=m +CONFIG_INFINIBAND_USER_ACCESS=m +CONFIG_MLX4_INFINIBAND=m +CONFIG_VIRTIO_BALLOON=m +# CONFIG_IOMMU_SUPPORT is not set +CONFIG_EXT2_FS=y +CONFIG_EXT2_FS_XATTR=y +CONFIG_EXT2_FS_POSIX_ACL=y +CONFIG_EXT2_FS_SECURITY=y +CONFIG_EXT2_FS_XIP=y +CONFIG_EXT3_FS=y +# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set +CONFIG_EXT3_FS_POSIX_ACL=y +CONFIG_EXT3_FS_SECURITY=y +CONFIG_EXT4_FS=y +CONFIG_EXT4_FS_POSIX_ACL=y +CONFIG_EXT4_FS_SECURITY=y +CONFIG_JBD_DEBUG=y +CONFIG_JBD2_DEBUG=y +CONFIG_JFS_FS=m +CONFIG_JFS_POSIX_ACL=y +CONFIG_JFS_SECURITY=y +CONFIG_JFS_STATISTICS=y +CONFIG_XFS_FS=y +CONFIG_XFS_QUOTA=y +CONFIG_XFS_POSIX_ACL=y +CONFIG_XFS_RT=y +CONFIG_XFS_DEBUG=y +CONFIG_GFS2_FS=m +CONFIG_OCFS2_FS=m +CONFIG_BTRFS_FS=y +CONFIG_BTRFS_FS_POSIX_ACL=y +CONFIG_NILFS2_FS=m +CONFIG_FANOTIFY=y +CONFIG_QUOTA_NETLINK_INTERFACE=y +CONFIG_QFMT_V1=m +CONFIG_QFMT_V2=m +CONFIG_AUTOFS4_FS=m +CONFIG_FUSE_FS=m +CONFIG_CUSE=m +CONFIG_FSCACHE=m +CONFIG_CACHEFILES=m +CONFIG_ISO9660_FS=y +CONFIG_JOLIET=y +CONFIG_ZISOFS=y +CONFIG_UDF_FS=m +CONFIG_MSDOS_FS=m +CONFIG_VFAT_FS=m +CONFIG_NTFS_FS=m +CONFIG_NTFS_RW=y +CONFIG_PROC_KCORE=y +CONFIG_TMPFS=y +CONFIG_TMPFS_POSIX_ACL=y +CONFIG_HUGETLBFS=y +CONFIG_CONFIGFS_FS=m +CONFIG_ECRYPT_FS=m +CONFIG_CRAMFS=m +CONFIG_SQUASHFS=m +CONFIG_SQUASHFS_XATTR=y +CONFIG_SQUASHFS_LZO=y +CONFIG_SQUASHFS_XZ=y +CONFIG_ROMFS_FS=m +CONFIG_NFS_FS=m +CONFIG_NFS_V3_ACL=y +CONFIG_NFS_V4=m +CONFIG_NFS_SWAP=y +CONFIG_NFSD=m +CONFIG_NFSD_V3_ACL=y +CONFIG_NFSD_V4=y +CONFIG_NFSD_V4_SECURITY_LABEL=y +CONFIG_CIFS=m +CONFIG_CIFS_STATS=y +CONFIG_CIFS_STATS2=y +CONFIG_CIFS_WEAK_PW_HASH=y +CONFIG_CIFS_UPCALL=y +CONFIG_CIFS_XATTR=y +CONFIG_CIFS_POSIX=y +# CONFIG_CIFS_DEBUG is not set +CONFIG_CIFS_DFS_UPCALL=y +CONFIG_NLS_DEFAULT="utf8" +CONFIG_NLS_CODEPAGE_437=m +CONFIG_NLS_CODEPAGE_850=m +CONFIG_NLS_ASCII=m +CONFIG_NLS_ISO8859_1=m +CONFIG_NLS_ISO8859_15=m +CONFIG_NLS_UTF8=m +CONFIG_DLM=m +CONFIG_PRINTK_TIME=y +CONFIG_DYNAMIC_DEBUG=y +CONFIG_DEBUG_INFO=y +# CONFIG_ENABLE_MUST_CHECK is not set +CONFIG_FRAME_WARN=1024 +CONFIG_READABLE_ASM=y +CONFIG_UNUSED_SYMBOLS=y +CONFIG_MAGIC_SYSRQ=y +CONFIG_DEBUG_KERNEL=y +CONFIG_DEBUG_PAGEALLOC=y +CONFIG_DEBUG_OBJECTS=y +CONFIG_DEBUG_OBJECTS_SELFTEST=y +CONFIG_DEBUG_OBJECTS_FREE=y +CONFIG_DEBUG_OBJECTS_TIMERS=y +CONFIG_DEBUG_OBJECTS_WORK=y +CONFIG_DEBUG_OBJECTS_RCU_HEAD=y +CONFIG_DEBUG_OBJECTS_PERCPU_COUNTER=y +CONFIG_SLUB_DEBUG_ON=y +CONFIG_SLUB_STATS=y +CONFIG_DEBUG_STACK_USAGE=y +CONFIG_DEBUG_VM=y +CONFIG_DEBUG_VM_RB=y +CONFIG_MEMORY_NOTIFIER_ERROR_INJECT=m +CONFIG_DEBUG_PER_CPU_MAPS=y +CONFIG_DEBUG_SHIRQ=y +CONFIG_DETECT_HUNG_TASK=y +CONFIG_PANIC_ON_OOPS=y +CONFIG_TIMER_STATS=y +CONFIG_DEBUG_RT_MUTEXES=y +CONFIG_DEBUG_WW_MUTEX_SLOWPATH=y +CONFIG_PROVE_LOCKING=y +CONFIG_LOCK_STAT=y +CONFIG_DEBUG_LOCKDEP=y +CONFIG_DEBUG_ATOMIC_SLEEP=y +CONFIG_DEBUG_LOCKING_API_SELFTESTS=y +CONFIG_DEBUG_LIST=y +CONFIG_DEBUG_SG=y +CONFIG_DEBUG_NOTIFIERS=y +CONFIG_DEBUG_CREDENTIALS=y +CONFIG_PROVE_RCU=y +CONFIG_RCU_TORTURE_TEST=m +CONFIG_RCU_CPU_STALL_TIMEOUT=300 +CONFIG_NOTIFIER_ERROR_INJECTION=m +CONFIG_CPU_NOTIFIER_ERROR_INJECT=m +CONFIG_PM_NOTIFIER_ERROR_INJECT=m +CONFIG_FAULT_INJECTION=y +CONFIG_FAILSLAB=y +CONFIG_FAIL_PAGE_ALLOC=y +CONFIG_FAIL_MAKE_REQUEST=y +CONFIG_FAIL_IO_TIMEOUT=y +CONFIG_FAULT_INJECTION_DEBUG_FS=y +CONFIG_FAULT_INJECTION_STACKTRACE_FILTER=y +CONFIG_LATENCYTOP=y +CONFIG_DEBUG_STRICT_USER_COPY_CHECKS=y +CONFIG_IRQSOFF_TRACER=y +CONFIG_PREEMPT_TRACER=y +CONFIG_SCHED_TRACER=y +CONFIG_FTRACE_SYSCALLS=y +CONFIG_STACK_TRACER=y +CONFIG_BLK_DEV_IO_TRACE=y +CONFIG_UPROBE_EVENT=y +CONFIG_LKDTM=m +CONFIG_TEST_LIST_SORT=y +CONFIG_KPROBES_SANITY_TEST=y +CONFIG_RBTREE_TEST=y +CONFIG_INTERVAL_TREE_TEST=m +CONFIG_PERCPU_TEST=m +CONFIG_ATOMIC64_SELFTEST=y +CONFIG_TEST_STRING_HELPERS=y +CONFIG_TEST_KSTRTOX=y +CONFIG_DMA_API_DEBUG=y +CONFIG_TEST_BPF=m +# CONFIG_STRICT_DEVMEM is not set +CONFIG_S390_PTDUMP=y +CONFIG_ENCRYPTED_KEYS=m +CONFIG_KEYS_DEBUG_PROC_KEYS=y +CONFIG_SECURITY=y +CONFIG_SECURITY_NETWORK=y +CONFIG_SECURITY_SELINUX=y +CONFIG_SECURITY_SELINUX_BOOTPARAM=y +CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=0 +CONFIG_SECURITY_SELINUX_DISABLE=y +CONFIG_IMA=y +CONFIG_IMA_APPRAISE=y +CONFIG_CRYPTO_USER=m +# CONFIG_CRYPTO_MANAGER_DISABLE_TESTS is not set +CONFIG_CRYPTO_CRYPTD=m +CONFIG_CRYPTO_TEST=m +CONFIG_CRYPTO_CCM=m +CONFIG_CRYPTO_GCM=m +CONFIG_CRYPTO_CTS=m +CONFIG_CRYPTO_LRW=m +CONFIG_CRYPTO_PCBC=m +CONFIG_CRYPTO_XTS=m +CONFIG_CRYPTO_XCBC=m +CONFIG_CRYPTO_VMAC=m +CONFIG_CRYPTO_CRC32=m +CONFIG_CRYPTO_MICHAEL_MIC=m +CONFIG_CRYPTO_RMD128=m +CONFIG_CRYPTO_RMD160=m +CONFIG_CRYPTO_RMD256=m +CONFIG_CRYPTO_RMD320=m +CONFIG_CRYPTO_SHA512=m +CONFIG_CRYPTO_TGR192=m +CONFIG_CRYPTO_WP512=m +CONFIG_CRYPTO_ANUBIS=m +CONFIG_CRYPTO_BLOWFISH=m +CONFIG_CRYPTO_CAMELLIA=m +CONFIG_CRYPTO_CAST5=m +CONFIG_CRYPTO_CAST6=m +CONFIG_CRYPTO_FCRYPT=m +CONFIG_CRYPTO_KHAZAD=m +CONFIG_CRYPTO_SALSA20=m +CONFIG_CRYPTO_SEED=m +CONFIG_CRYPTO_SERPENT=m +CONFIG_CRYPTO_TEA=m +CONFIG_CRYPTO_TWOFISH=m +CONFIG_CRYPTO_ZLIB=y +CONFIG_CRYPTO_LZO=m +CONFIG_CRYPTO_LZ4=m +CONFIG_CRYPTO_LZ4HC=m +CONFIG_CRYPTO_USER_API_HASH=m +CONFIG_CRYPTO_USER_API_SKCIPHER=m +CONFIG_ZCRYPT=m +CONFIG_CRYPTO_SHA1_S390=m +CONFIG_CRYPTO_SHA256_S390=m +CONFIG_CRYPTO_SHA512_S390=m +CONFIG_CRYPTO_DES_S390=m +CONFIG_CRYPTO_AES_S390=m +CONFIG_CRYPTO_GHASH_S390=m +CONFIG_ASYMMETRIC_KEY_TYPE=m +CONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=m +CONFIG_X509_CERTIFICATE_PARSER=m +CONFIG_CRC7=m +CONFIG_CRC8=m +CONFIG_CORDIC=m +CONFIG_CMM=m +CONFIG_APPLDATA_BASE=y +CONFIG_KVM=m +CONFIG_KVM_S390_UCONTROL=y diff --git a/arch/s390/configs/gcov_defconfig b/arch/s390/configs/gcov_defconfig new file mode 100644 index 000000000..5c3097272 --- /dev/null +++ b/arch/s390/configs/gcov_defconfig @@ -0,0 +1,625 @@ +CONFIG_SYSVIPC=y +CONFIG_POSIX_MQUEUE=y +CONFIG_FHANDLE=y +CONFIG_AUDIT=y +CONFIG_NO_HZ=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_BSD_PROCESS_ACCT=y +CONFIG_BSD_PROCESS_ACCT_V3=y +CONFIG_TASKSTATS=y +CONFIG_TASK_DELAY_ACCT=y +CONFIG_TASK_XACCT=y +CONFIG_TASK_IO_ACCOUNTING=y +CONFIG_RCU_FAST_NO_HZ=y +CONFIG_IKCONFIG=y +CONFIG_IKCONFIG_PROC=y +CONFIG_CGROUP_FREEZER=y +CONFIG_CGROUP_DEVICE=y +CONFIG_CPUSETS=y +CONFIG_CGROUP_CPUACCT=y +CONFIG_RESOURCE_COUNTERS=y +CONFIG_CGROUP_PERF=y +CONFIG_BLK_CGROUP=y +CONFIG_SCHED_AUTOGROUP=y +CONFIG_BLK_DEV_INITRD=y +# CONFIG_COMPAT_BRK is not set +CONFIG_PROFILING=y +CONFIG_OPROFILE=m +CONFIG_KPROBES=y +CONFIG_JUMP_LABEL=y +CONFIG_GCOV_KERNEL=y +CONFIG_GCOV_PROFILE_ALL=y +CONFIG_MODULES=y +CONFIG_MODULE_FORCE_LOAD=y +CONFIG_MODULE_UNLOAD=y +CONFIG_MODULE_FORCE_UNLOAD=y +CONFIG_MODVERSIONS=y +CONFIG_MODULE_SRCVERSION_ALL=y +CONFIG_BLK_DEV_THROTTLING=y +CONFIG_PARTITION_ADVANCED=y +CONFIG_IBM_PARTITION=y +CONFIG_BSD_DISKLABEL=y +CONFIG_MINIX_SUBPARTITION=y +CONFIG_SOLARIS_X86_PARTITION=y +CONFIG_UNIXWARE_DISKLABEL=y +CONFIG_CFQ_GROUP_IOSCHED=y +CONFIG_DEFAULT_DEADLINE=y +CONFIG_MARCH_Z196=y +CONFIG_TUNE_ZEC12=y +CONFIG_NR_CPUS=256 +CONFIG_HZ_100=y +CONFIG_MEMORY_HOTPLUG=y +CONFIG_MEMORY_HOTREMOVE=y +CONFIG_KSM=y +CONFIG_TRANSPARENT_HUGEPAGE=y +CONFIG_PCI=y +CONFIG_HOTPLUG_PCI=y +CONFIG_HOTPLUG_PCI_S390=y +CONFIG_CHSC_SCH=y +CONFIG_CRASH_DUMP=y +# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set +CONFIG_BINFMT_MISC=m +CONFIG_HIBERNATION=y +CONFIG_NET=y +CONFIG_PACKET=y +CONFIG_PACKET_DIAG=m +CONFIG_UNIX=y +CONFIG_UNIX_DIAG=m +CONFIG_XFRM_USER=m +CONFIG_NET_KEY=m +CONFIG_INET=y +CONFIG_IP_MULTICAST=y +CONFIG_IP_ADVANCED_ROUTER=y +CONFIG_IP_MULTIPLE_TABLES=y +CONFIG_IP_ROUTE_MULTIPATH=y +CONFIG_IP_ROUTE_VERBOSE=y +CONFIG_NET_IPIP=m +CONFIG_NET_IPGRE_DEMUX=m +CONFIG_NET_IPGRE=m +CONFIG_NET_IPGRE_BROADCAST=y +CONFIG_IP_MROUTE=y +CONFIG_IP_MROUTE_MULTIPLE_TABLES=y +CONFIG_IP_PIMSM_V1=y +CONFIG_IP_PIMSM_V2=y +CONFIG_SYN_COOKIES=y +CONFIG_NET_IPVTI=m +CONFIG_INET_AH=m +CONFIG_INET_ESP=m +CONFIG_INET_IPCOMP=m +CONFIG_INET_XFRM_MODE_TRANSPORT=m +CONFIG_INET_XFRM_MODE_TUNNEL=m +CONFIG_INET_XFRM_MODE_BEET=m +CONFIG_INET_DIAG=m +CONFIG_INET_UDP_DIAG=m +CONFIG_TCP_CONG_ADVANCED=y +CONFIG_TCP_CONG_HSTCP=m +CONFIG_TCP_CONG_HYBLA=m +CONFIG_TCP_CONG_SCALABLE=m +CONFIG_TCP_CONG_LP=m +CONFIG_TCP_CONG_VENO=m +CONFIG_TCP_CONG_YEAH=m +CONFIG_TCP_CONG_ILLINOIS=m +CONFIG_IPV6=y +CONFIG_IPV6_ROUTER_PREF=y +CONFIG_INET6_AH=m +CONFIG_INET6_ESP=m +CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_MIP6=m +CONFIG_INET6_XFRM_MODE_TRANSPORT=m +CONFIG_INET6_XFRM_MODE_TUNNEL=m +CONFIG_INET6_XFRM_MODE_BEET=m +CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION=m +CONFIG_IPV6_VTI=m +CONFIG_IPV6_SIT=m +CONFIG_IPV6_GRE=m +CONFIG_IPV6_MULTIPLE_TABLES=y +CONFIG_IPV6_SUBTREES=y +CONFIG_NETFILTER=y +CONFIG_NF_CONNTRACK=m +CONFIG_NF_CONNTRACK_SECMARK=y +CONFIG_NF_CONNTRACK_EVENTS=y +CONFIG_NF_CONNTRACK_TIMEOUT=y +CONFIG_NF_CONNTRACK_TIMESTAMP=y +CONFIG_NF_CT_PROTO_DCCP=m +CONFIG_NF_CT_PROTO_UDPLITE=m +CONFIG_NF_CONNTRACK_AMANDA=m +CONFIG_NF_CONNTRACK_FTP=m +CONFIG_NF_CONNTRACK_H323=m +CONFIG_NF_CONNTRACK_IRC=m +CONFIG_NF_CONNTRACK_NETBIOS_NS=m +CONFIG_NF_CONNTRACK_SNMP=m +CONFIG_NF_CONNTRACK_PPTP=m +CONFIG_NF_CONNTRACK_SANE=m +CONFIG_NF_CONNTRACK_SIP=m +CONFIG_NF_CONNTRACK_TFTP=m +CONFIG_NF_CT_NETLINK=m +CONFIG_NF_CT_NETLINK_TIMEOUT=m +CONFIG_NF_TABLES=m +CONFIG_NFT_EXTHDR=m +CONFIG_NFT_META=m +CONFIG_NFT_CT=m +CONFIG_NFT_RBTREE=m +CONFIG_NFT_HASH=m +CONFIG_NFT_COUNTER=m +CONFIG_NFT_LOG=m +CONFIG_NFT_LIMIT=m +CONFIG_NFT_NAT=m +CONFIG_NFT_COMPAT=m +CONFIG_NETFILTER_XT_SET=m +CONFIG_NETFILTER_XT_TARGET_AUDIT=m +CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m +CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m +CONFIG_NETFILTER_XT_TARGET_CONNMARK=m +CONFIG_NETFILTER_XT_TARGET_CONNSECMARK=m +CONFIG_NETFILTER_XT_TARGET_CT=m +CONFIG_NETFILTER_XT_TARGET_DSCP=m +CONFIG_NETFILTER_XT_TARGET_HMARK=m +CONFIG_NETFILTER_XT_TARGET_IDLETIMER=m +CONFIG_NETFILTER_XT_TARGET_LOG=m +CONFIG_NETFILTER_XT_TARGET_MARK=m +CONFIG_NETFILTER_XT_TARGET_NFLOG=m +CONFIG_NETFILTER_XT_TARGET_NFQUEUE=m +CONFIG_NETFILTER_XT_TARGET_TEE=m +CONFIG_NETFILTER_XT_TARGET_TPROXY=m +CONFIG_NETFILTER_XT_TARGET_TRACE=m +CONFIG_NETFILTER_XT_TARGET_SECMARK=m +CONFIG_NETFILTER_XT_TARGET_TCPMSS=m +CONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP=m +CONFIG_NETFILTER_XT_MATCH_ADDRTYPE=m +CONFIG_NETFILTER_XT_MATCH_BPF=m +CONFIG_NETFILTER_XT_MATCH_CLUSTER=m +CONFIG_NETFILTER_XT_MATCH_COMMENT=m +CONFIG_NETFILTER_XT_MATCH_CONNBYTES=m +CONFIG_NETFILTER_XT_MATCH_CONNLABEL=m +CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=m +CONFIG_NETFILTER_XT_MATCH_CONNMARK=m +CONFIG_NETFILTER_XT_MATCH_CONNTRACK=m +CONFIG_NETFILTER_XT_MATCH_CPU=m +CONFIG_NETFILTER_XT_MATCH_DCCP=m +CONFIG_NETFILTER_XT_MATCH_DEVGROUP=m +CONFIG_NETFILTER_XT_MATCH_DSCP=m +CONFIG_NETFILTER_XT_MATCH_ESP=m +CONFIG_NETFILTER_XT_MATCH_HASHLIMIT=m +CONFIG_NETFILTER_XT_MATCH_HELPER=m +CONFIG_NETFILTER_XT_MATCH_IPRANGE=m +CONFIG_NETFILTER_XT_MATCH_IPVS=m +CONFIG_NETFILTER_XT_MATCH_LENGTH=m +CONFIG_NETFILTER_XT_MATCH_LIMIT=m +CONFIG_NETFILTER_XT_MATCH_MAC=m +CONFIG_NETFILTER_XT_MATCH_MARK=m +CONFIG_NETFILTER_XT_MATCH_MULTIPORT=m +CONFIG_NETFILTER_XT_MATCH_NFACCT=m +CONFIG_NETFILTER_XT_MATCH_OSF=m +CONFIG_NETFILTER_XT_MATCH_OWNER=m +CONFIG_NETFILTER_XT_MATCH_POLICY=m +CONFIG_NETFILTER_XT_MATCH_PHYSDEV=m +CONFIG_NETFILTER_XT_MATCH_PKTTYPE=m +CONFIG_NETFILTER_XT_MATCH_QUOTA=m +CONFIG_NETFILTER_XT_MATCH_RATEEST=m +CONFIG_NETFILTER_XT_MATCH_REALM=m +CONFIG_NETFILTER_XT_MATCH_RECENT=m +CONFIG_NETFILTER_XT_MATCH_SOCKET=m +CONFIG_NETFILTER_XT_MATCH_STATE=m +CONFIG_NETFILTER_XT_MATCH_STATISTIC=m +CONFIG_NETFILTER_XT_MATCH_STRING=m +CONFIG_NETFILTER_XT_MATCH_TCPMSS=m +CONFIG_NETFILTER_XT_MATCH_TIME=m +CONFIG_NETFILTER_XT_MATCH_U32=m +CONFIG_IP_SET=m +CONFIG_IP_SET_BITMAP_IP=m +CONFIG_IP_SET_BITMAP_IPMAC=m +CONFIG_IP_SET_BITMAP_PORT=m +CONFIG_IP_SET_HASH_IP=m +CONFIG_IP_SET_HASH_IPPORT=m +CONFIG_IP_SET_HASH_IPPORTIP=m +CONFIG_IP_SET_HASH_IPPORTNET=m +CONFIG_IP_SET_HASH_NETPORTNET=m +CONFIG_IP_SET_HASH_NET=m +CONFIG_IP_SET_HASH_NETNET=m +CONFIG_IP_SET_HASH_NETPORT=m +CONFIG_IP_SET_HASH_NETIFACE=m +CONFIG_IP_SET_LIST_SET=m +CONFIG_IP_VS=m +CONFIG_IP_VS_PROTO_TCP=y +CONFIG_IP_VS_PROTO_UDP=y +CONFIG_IP_VS_PROTO_ESP=y +CONFIG_IP_VS_PROTO_AH=y +CONFIG_IP_VS_RR=m +CONFIG_IP_VS_WRR=m +CONFIG_IP_VS_LC=m +CONFIG_IP_VS_WLC=m +CONFIG_IP_VS_LBLC=m +CONFIG_IP_VS_LBLCR=m +CONFIG_IP_VS_DH=m +CONFIG_IP_VS_SH=m +CONFIG_IP_VS_SED=m +CONFIG_IP_VS_NQ=m +CONFIG_IP_VS_FTP=m +CONFIG_IP_VS_PE_SIP=m +CONFIG_NF_CONNTRACK_IPV4=m +# CONFIG_NF_CONNTRACK_PROC_COMPAT is not set +CONFIG_NF_TABLES_IPV4=m +CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_CHAIN_NAT_IPV4=m +CONFIG_NF_TABLES_ARP=m +CONFIG_NF_NAT_IPV4=m +CONFIG_IP_NF_IPTABLES=m +CONFIG_IP_NF_MATCH_AH=m +CONFIG_IP_NF_MATCH_ECN=m +CONFIG_IP_NF_MATCH_RPFILTER=m +CONFIG_IP_NF_MATCH_TTL=m +CONFIG_IP_NF_FILTER=m +CONFIG_IP_NF_TARGET_REJECT=m +CONFIG_IP_NF_MANGLE=m +CONFIG_IP_NF_TARGET_CLUSTERIP=m +CONFIG_IP_NF_TARGET_ECN=m +CONFIG_IP_NF_TARGET_TTL=m +CONFIG_IP_NF_RAW=m +CONFIG_IP_NF_SECURITY=m +CONFIG_IP_NF_ARPTABLES=m +CONFIG_IP_NF_ARPFILTER=m +CONFIG_IP_NF_ARP_MANGLE=m +CONFIG_NF_CONNTRACK_IPV6=m +CONFIG_NF_TABLES_IPV6=m +CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_CHAIN_NAT_IPV6=m +CONFIG_NF_NAT_IPV6=m +CONFIG_IP6_NF_IPTABLES=m +CONFIG_IP6_NF_MATCH_AH=m +CONFIG_IP6_NF_MATCH_EUI64=m +CONFIG_IP6_NF_MATCH_FRAG=m +CONFIG_IP6_NF_MATCH_OPTS=m +CONFIG_IP6_NF_MATCH_HL=m +CONFIG_IP6_NF_MATCH_IPV6HEADER=m +CONFIG_IP6_NF_MATCH_MH=m +CONFIG_IP6_NF_MATCH_RPFILTER=m +CONFIG_IP6_NF_MATCH_RT=m +CONFIG_IP6_NF_TARGET_HL=m +CONFIG_IP6_NF_FILTER=m +CONFIG_IP6_NF_TARGET_REJECT=m +CONFIG_IP6_NF_MANGLE=m +CONFIG_IP6_NF_RAW=m +CONFIG_IP6_NF_SECURITY=m +CONFIG_NF_TABLES_BRIDGE=m +CONFIG_NET_SCTPPROBE=m +CONFIG_RDS=m +CONFIG_RDS_RDMA=m +CONFIG_RDS_TCP=m +CONFIG_L2TP=m +CONFIG_L2TP_DEBUGFS=m +CONFIG_L2TP_V3=y +CONFIG_L2TP_IP=m +CONFIG_L2TP_ETH=m +CONFIG_BRIDGE=m +CONFIG_VLAN_8021Q=m +CONFIG_VLAN_8021Q_GVRP=y +CONFIG_NET_SCHED=y +CONFIG_NET_SCH_CBQ=m +CONFIG_NET_SCH_HTB=m +CONFIG_NET_SCH_HFSC=m +CONFIG_NET_SCH_PRIO=m +CONFIG_NET_SCH_MULTIQ=m +CONFIG_NET_SCH_RED=m +CONFIG_NET_SCH_SFB=m +CONFIG_NET_SCH_SFQ=m +CONFIG_NET_SCH_TEQL=m +CONFIG_NET_SCH_TBF=m +CONFIG_NET_SCH_GRED=m +CONFIG_NET_SCH_DSMARK=m +CONFIG_NET_SCH_NETEM=m +CONFIG_NET_SCH_DRR=m +CONFIG_NET_SCH_MQPRIO=m +CONFIG_NET_SCH_CHOKE=m +CONFIG_NET_SCH_QFQ=m +CONFIG_NET_SCH_CODEL=m +CONFIG_NET_SCH_FQ_CODEL=m +CONFIG_NET_SCH_INGRESS=m +CONFIG_NET_SCH_PLUG=m +CONFIG_NET_CLS_BASIC=m +CONFIG_NET_CLS_TCINDEX=m +CONFIG_NET_CLS_ROUTE4=m +CONFIG_NET_CLS_FW=m +CONFIG_NET_CLS_U32=m +CONFIG_CLS_U32_PERF=y +CONFIG_CLS_U32_MARK=y +CONFIG_NET_CLS_RSVP=m +CONFIG_NET_CLS_RSVP6=m +CONFIG_NET_CLS_FLOW=m +CONFIG_NET_CLS_CGROUP=y +CONFIG_NET_CLS_BPF=m +CONFIG_NET_CLS_ACT=y +CONFIG_NET_ACT_POLICE=m +CONFIG_NET_ACT_GACT=m +CONFIG_GACT_PROB=y +CONFIG_NET_ACT_MIRRED=m +CONFIG_NET_ACT_IPT=m +CONFIG_NET_ACT_NAT=m +CONFIG_NET_ACT_PEDIT=m +CONFIG_NET_ACT_SIMP=m +CONFIG_NET_ACT_SKBEDIT=m +CONFIG_NET_ACT_CSUM=m +CONFIG_DNS_RESOLVER=y +CONFIG_BPF_JIT=y +CONFIG_NET_PKTGEN=m +CONFIG_NET_TCPPROBE=m +CONFIG_DEVTMPFS=y +CONFIG_CONNECTOR=y +CONFIG_BLK_DEV_LOOP=m +CONFIG_BLK_DEV_CRYPTOLOOP=m +CONFIG_BLK_DEV_NBD=m +CONFIG_BLK_DEV_OSD=m +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_SIZE=32768 +CONFIG_BLK_DEV_XIP=y +CONFIG_CDROM_PKTCDVD=m +CONFIG_ATA_OVER_ETH=m +CONFIG_VIRTIO_BLK=y +CONFIG_ENCLOSURE_SERVICES=m +CONFIG_RAID_ATTRS=m +CONFIG_SCSI=y +CONFIG_BLK_DEV_SD=y +CONFIG_CHR_DEV_ST=m +CONFIG_CHR_DEV_OSST=m +CONFIG_BLK_DEV_SR=m +CONFIG_CHR_DEV_SG=y +CONFIG_CHR_DEV_SCH=m +CONFIG_SCSI_ENCLOSURE=m +CONFIG_SCSI_CONSTANTS=y +CONFIG_SCSI_LOGGING=y +CONFIG_SCSI_SPI_ATTRS=m +CONFIG_SCSI_FC_ATTRS=y +CONFIG_SCSI_SAS_LIBSAS=m +CONFIG_SCSI_SRP_ATTRS=m +CONFIG_ISCSI_TCP=m +CONFIG_SCSI_DEBUG=m +CONFIG_ZFCP=y +CONFIG_SCSI_VIRTIO=m +CONFIG_SCSI_DH=m +CONFIG_SCSI_DH_RDAC=m +CONFIG_SCSI_DH_HP_SW=m +CONFIG_SCSI_DH_EMC=m +CONFIG_SCSI_DH_ALUA=m +CONFIG_SCSI_OSD_INITIATOR=m +CONFIG_SCSI_OSD_ULD=m +CONFIG_MD=y +CONFIG_BLK_DEV_MD=y +CONFIG_MD_LINEAR=m +CONFIG_MD_RAID0=m +CONFIG_MD_MULTIPATH=m +CONFIG_MD_FAULTY=m +CONFIG_BLK_DEV_DM=m +CONFIG_DM_CRYPT=m +CONFIG_DM_SNAPSHOT=m +CONFIG_DM_MIRROR=m +CONFIG_DM_LOG_USERSPACE=m +CONFIG_DM_RAID=m +CONFIG_DM_ZERO=m +CONFIG_DM_MULTIPATH=m +CONFIG_DM_MULTIPATH_QL=m +CONFIG_DM_MULTIPATH_ST=m +CONFIG_DM_DELAY=m +CONFIG_DM_UEVENT=y +CONFIG_DM_FLAKEY=m +CONFIG_DM_VERITY=m +CONFIG_DM_SWITCH=m +CONFIG_NETDEVICES=y +CONFIG_BONDING=m +CONFIG_DUMMY=m +CONFIG_EQUALIZER=m +CONFIG_IFB=m +CONFIG_MACVLAN=m +CONFIG_MACVTAP=m +CONFIG_VXLAN=m +CONFIG_TUN=m +CONFIG_VETH=m +CONFIG_VIRTIO_NET=m +CONFIG_NLMON=m +CONFIG_VHOST_NET=m +# CONFIG_NET_VENDOR_ARC is not set +# CONFIG_NET_VENDOR_CHELSIO is not set +# CONFIG_NET_VENDOR_INTEL is not set +# CONFIG_NET_VENDOR_MARVELL is not set +CONFIG_MLX4_EN=m +# CONFIG_NET_VENDOR_NATSEMI is not set +CONFIG_PPP=m +CONFIG_PPP_BSDCOMP=m +CONFIG_PPP_DEFLATE=m +CONFIG_PPP_MPPE=m +CONFIG_PPPOE=m +CONFIG_PPTP=m +CONFIG_PPPOL2TP=m +CONFIG_PPP_ASYNC=m +CONFIG_PPP_SYNC_TTY=m +# CONFIG_INPUT_MOUSEDEV_PSAUX is not set +# CONFIG_INPUT_KEYBOARD is not set +# CONFIG_INPUT_MOUSE is not set +# CONFIG_SERIO is not set +CONFIG_DEVPTS_MULTIPLE_INSTANCES=y +CONFIG_LEGACY_PTY_COUNT=0 +CONFIG_HW_RANDOM_VIRTIO=m +CONFIG_RAW_DRIVER=m +CONFIG_HANGCHECK_TIMER=m +CONFIG_TN3270_FS=y +CONFIG_WATCHDOG=y +CONFIG_WATCHDOG_NOWAYOUT=y +CONFIG_SOFT_WATCHDOG=m +CONFIG_DIAG288_WATCHDOG=m +# CONFIG_HID is not set +# CONFIG_USB_SUPPORT is not set +CONFIG_INFINIBAND=m +CONFIG_INFINIBAND_USER_ACCESS=m +CONFIG_MLX4_INFINIBAND=m +CONFIG_VIRTIO_BALLOON=m +# CONFIG_IOMMU_SUPPORT is not set +CONFIG_EXT2_FS=y +CONFIG_EXT2_FS_XATTR=y +CONFIG_EXT2_FS_POSIX_ACL=y +CONFIG_EXT2_FS_SECURITY=y +CONFIG_EXT2_FS_XIP=y +CONFIG_EXT3_FS=y +# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set +CONFIG_EXT3_FS_POSIX_ACL=y +CONFIG_EXT3_FS_SECURITY=y +CONFIG_EXT4_FS=y +CONFIG_EXT4_FS_POSIX_ACL=y +CONFIG_EXT4_FS_SECURITY=y +CONFIG_JBD_DEBUG=y +CONFIG_JBD2_DEBUG=y +CONFIG_JFS_FS=m +CONFIG_JFS_POSIX_ACL=y +CONFIG_JFS_SECURITY=y +CONFIG_JFS_STATISTICS=y +CONFIG_XFS_FS=y +CONFIG_XFS_QUOTA=y +CONFIG_XFS_POSIX_ACL=y +CONFIG_XFS_RT=y +CONFIG_GFS2_FS=m +CONFIG_OCFS2_FS=m +CONFIG_BTRFS_FS=y +CONFIG_BTRFS_FS_POSIX_ACL=y +CONFIG_NILFS2_FS=m +CONFIG_FANOTIFY=y +CONFIG_QUOTA_NETLINK_INTERFACE=y +CONFIG_QFMT_V1=m +CONFIG_QFMT_V2=m +CONFIG_AUTOFS4_FS=m +CONFIG_FUSE_FS=m +CONFIG_CUSE=m +CONFIG_FSCACHE=m +CONFIG_CACHEFILES=m +CONFIG_ISO9660_FS=y +CONFIG_JOLIET=y +CONFIG_ZISOFS=y +CONFIG_UDF_FS=m +CONFIG_MSDOS_FS=m +CONFIG_VFAT_FS=m +CONFIG_NTFS_FS=m +CONFIG_NTFS_RW=y +CONFIG_PROC_KCORE=y +CONFIG_TMPFS=y +CONFIG_TMPFS_POSIX_ACL=y +CONFIG_HUGETLBFS=y +CONFIG_CONFIGFS_FS=m +CONFIG_ECRYPT_FS=m +CONFIG_CRAMFS=m +CONFIG_SQUASHFS=m +CONFIG_SQUASHFS_XATTR=y +CONFIG_SQUASHFS_LZO=y +CONFIG_SQUASHFS_XZ=y +CONFIG_ROMFS_FS=m +CONFIG_NFS_FS=m +CONFIG_NFS_V3_ACL=y +CONFIG_NFS_V4=m +CONFIG_NFS_SWAP=y +CONFIG_NFSD=m +CONFIG_NFSD_V3_ACL=y +CONFIG_NFSD_V4=y +CONFIG_NFSD_V4_SECURITY_LABEL=y +CONFIG_CIFS=m +CONFIG_CIFS_STATS=y +CONFIG_CIFS_STATS2=y +CONFIG_CIFS_WEAK_PW_HASH=y +CONFIG_CIFS_UPCALL=y +CONFIG_CIFS_XATTR=y +CONFIG_CIFS_POSIX=y +# CONFIG_CIFS_DEBUG is not set +CONFIG_CIFS_DFS_UPCALL=y +CONFIG_NLS_DEFAULT="utf8" +CONFIG_NLS_CODEPAGE_437=m +CONFIG_NLS_CODEPAGE_850=m +CONFIG_NLS_ASCII=m +CONFIG_NLS_ISO8859_1=m +CONFIG_NLS_ISO8859_15=m +CONFIG_NLS_UTF8=m +CONFIG_DLM=m +CONFIG_PRINTK_TIME=y +CONFIG_DEBUG_INFO=y +# CONFIG_ENABLE_MUST_CHECK is not set +CONFIG_FRAME_WARN=1024 +CONFIG_UNUSED_SYMBOLS=y +CONFIG_MAGIC_SYSRQ=y +CONFIG_DEBUG_KERNEL=y +CONFIG_MEMORY_NOTIFIER_ERROR_INJECT=m +CONFIG_PANIC_ON_OOPS=y +CONFIG_TIMER_STATS=y +CONFIG_RCU_TORTURE_TEST=m +CONFIG_RCU_CPU_STALL_TIMEOUT=60 +CONFIG_NOTIFIER_ERROR_INJECTION=m +CONFIG_CPU_NOTIFIER_ERROR_INJECT=m +CONFIG_PM_NOTIFIER_ERROR_INJECT=m +CONFIG_LATENCYTOP=y +CONFIG_BLK_DEV_IO_TRACE=y +# CONFIG_KPROBE_EVENT is not set +CONFIG_LKDTM=m +CONFIG_RBTREE_TEST=m +CONFIG_INTERVAL_TREE_TEST=m +CONFIG_PERCPU_TEST=m +CONFIG_ATOMIC64_SELFTEST=y +# CONFIG_STRICT_DEVMEM is not set +CONFIG_S390_PTDUMP=y +CONFIG_ENCRYPTED_KEYS=m +CONFIG_KEYS_DEBUG_PROC_KEYS=y +CONFIG_SECURITY=y +CONFIG_SECURITY_NETWORK=y +CONFIG_SECURITY_SELINUX=y +CONFIG_SECURITY_SELINUX_BOOTPARAM=y +CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=0 +CONFIG_SECURITY_SELINUX_DISABLE=y +CONFIG_IMA=y +CONFIG_IMA_APPRAISE=y +CONFIG_CRYPTO_USER=m +# CONFIG_CRYPTO_MANAGER_DISABLE_TESTS is not set +CONFIG_CRYPTO_CRYPTD=m +CONFIG_CRYPTO_TEST=m +CONFIG_CRYPTO_CCM=m +CONFIG_CRYPTO_GCM=m +CONFIG_CRYPTO_CTS=m +CONFIG_CRYPTO_LRW=m +CONFIG_CRYPTO_PCBC=m +CONFIG_CRYPTO_XTS=m +CONFIG_CRYPTO_XCBC=m +CONFIG_CRYPTO_VMAC=m +CONFIG_CRYPTO_CRC32=m +CONFIG_CRYPTO_MICHAEL_MIC=m +CONFIG_CRYPTO_RMD128=m +CONFIG_CRYPTO_RMD160=m +CONFIG_CRYPTO_RMD256=m +CONFIG_CRYPTO_RMD320=m +CONFIG_CRYPTO_SHA512=m +CONFIG_CRYPTO_TGR192=m +CONFIG_CRYPTO_WP512=m +CONFIG_CRYPTO_ANUBIS=m +CONFIG_CRYPTO_BLOWFISH=m +CONFIG_CRYPTO_CAMELLIA=m +CONFIG_CRYPTO_CAST5=m +CONFIG_CRYPTO_CAST6=m +CONFIG_CRYPTO_FCRYPT=m +CONFIG_CRYPTO_KHAZAD=m +CONFIG_CRYPTO_SALSA20=m +CONFIG_CRYPTO_SEED=m +CONFIG_CRYPTO_SERPENT=m +CONFIG_CRYPTO_TEA=m +CONFIG_CRYPTO_TWOFISH=m +CONFIG_CRYPTO_ZLIB=y +CONFIG_CRYPTO_LZO=m +CONFIG_CRYPTO_LZ4=m +CONFIG_CRYPTO_LZ4HC=m +CONFIG_CRYPTO_USER_API_HASH=m +CONFIG_CRYPTO_USER_API_SKCIPHER=m +CONFIG_ZCRYPT=m +CONFIG_CRYPTO_SHA1_S390=m +CONFIG_CRYPTO_SHA256_S390=m +CONFIG_CRYPTO_SHA512_S390=m +CONFIG_CRYPTO_DES_S390=m +CONFIG_CRYPTO_AES_S390=m +CONFIG_CRYPTO_GHASH_S390=m +CONFIG_ASYMMETRIC_KEY_TYPE=m +CONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=m +CONFIG_X509_CERTIFICATE_PARSER=m +CONFIG_CRC7=m +CONFIG_CRC8=m +CONFIG_CORDIC=m +CONFIG_CMM=m +CONFIG_APPLDATA_BASE=y +CONFIG_KVM=m +CONFIG_KVM_S390_UCONTROL=y diff --git a/arch/s390/configs/performance_defconfig b/arch/s390/configs/performance_defconfig new file mode 100644 index 000000000..bda70f1ff --- /dev/null +++ b/arch/s390/configs/performance_defconfig @@ -0,0 +1,620 @@ +CONFIG_SYSVIPC=y +CONFIG_POSIX_MQUEUE=y +CONFIG_FHANDLE=y +CONFIG_AUDIT=y +CONFIG_NO_HZ=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_BSD_PROCESS_ACCT=y +CONFIG_BSD_PROCESS_ACCT_V3=y +CONFIG_TASKSTATS=y +CONFIG_TASK_DELAY_ACCT=y +CONFIG_TASK_XACCT=y +CONFIG_TASK_IO_ACCOUNTING=y +CONFIG_RCU_FAST_NO_HZ=y +CONFIG_IKCONFIG=y +CONFIG_IKCONFIG_PROC=y +CONFIG_CGROUP_FREEZER=y +CONFIG_CGROUP_DEVICE=y +CONFIG_CPUSETS=y +CONFIG_CGROUP_CPUACCT=y +CONFIG_RESOURCE_COUNTERS=y +CONFIG_CGROUP_PERF=y +CONFIG_BLK_CGROUP=y +CONFIG_SCHED_AUTOGROUP=y +CONFIG_BLK_DEV_INITRD=y +# CONFIG_COMPAT_BRK is not set +CONFIG_PROFILING=y +CONFIG_OPROFILE=m +CONFIG_KPROBES=y +CONFIG_JUMP_LABEL=y +CONFIG_MODULES=y +CONFIG_MODULE_FORCE_LOAD=y +CONFIG_MODULE_UNLOAD=y +CONFIG_MODULE_FORCE_UNLOAD=y +CONFIG_MODVERSIONS=y +CONFIG_MODULE_SRCVERSION_ALL=y +CONFIG_BLK_DEV_THROTTLING=y +CONFIG_PARTITION_ADVANCED=y +CONFIG_IBM_PARTITION=y +CONFIG_BSD_DISKLABEL=y +CONFIG_MINIX_SUBPARTITION=y +CONFIG_SOLARIS_X86_PARTITION=y +CONFIG_UNIXWARE_DISKLABEL=y +CONFIG_CFQ_GROUP_IOSCHED=y +CONFIG_DEFAULT_DEADLINE=y +CONFIG_MARCH_Z196=y +CONFIG_TUNE_ZEC12=y +CONFIG_NR_CPUS=256 +CONFIG_HZ_100=y +CONFIG_MEMORY_HOTPLUG=y +CONFIG_MEMORY_HOTREMOVE=y +CONFIG_KSM=y +CONFIG_TRANSPARENT_HUGEPAGE=y +CONFIG_PCI=y +CONFIG_HOTPLUG_PCI=y +CONFIG_HOTPLUG_PCI_S390=y +CONFIG_CHSC_SCH=y +CONFIG_CRASH_DUMP=y +# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set +CONFIG_BINFMT_MISC=m +CONFIG_HIBERNATION=y +CONFIG_NET=y +CONFIG_PACKET=y +CONFIG_PACKET_DIAG=m +CONFIG_UNIX=y +CONFIG_UNIX_DIAG=m +CONFIG_XFRM_USER=m +CONFIG_NET_KEY=m +CONFIG_INET=y +CONFIG_IP_MULTICAST=y +CONFIG_IP_ADVANCED_ROUTER=y +CONFIG_IP_MULTIPLE_TABLES=y +CONFIG_IP_ROUTE_MULTIPATH=y +CONFIG_IP_ROUTE_VERBOSE=y +CONFIG_NET_IPIP=m +CONFIG_NET_IPGRE_DEMUX=m +CONFIG_NET_IPGRE=m +CONFIG_NET_IPGRE_BROADCAST=y +CONFIG_IP_MROUTE=y +CONFIG_IP_MROUTE_MULTIPLE_TABLES=y +CONFIG_IP_PIMSM_V1=y +CONFIG_IP_PIMSM_V2=y +CONFIG_SYN_COOKIES=y +CONFIG_NET_IPVTI=m +CONFIG_INET_AH=m +CONFIG_INET_ESP=m +CONFIG_INET_IPCOMP=m +CONFIG_INET_XFRM_MODE_TRANSPORT=m +CONFIG_INET_XFRM_MODE_TUNNEL=m +CONFIG_INET_XFRM_MODE_BEET=m +CONFIG_INET_DIAG=m +CONFIG_INET_UDP_DIAG=m +CONFIG_TCP_CONG_ADVANCED=y +CONFIG_TCP_CONG_HSTCP=m +CONFIG_TCP_CONG_HYBLA=m +CONFIG_TCP_CONG_SCALABLE=m +CONFIG_TCP_CONG_LP=m +CONFIG_TCP_CONG_VENO=m +CONFIG_TCP_CONG_YEAH=m +CONFIG_TCP_CONG_ILLINOIS=m +CONFIG_IPV6=y +CONFIG_IPV6_ROUTER_PREF=y +CONFIG_INET6_AH=m +CONFIG_INET6_ESP=m +CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_MIP6=m +CONFIG_INET6_XFRM_MODE_TRANSPORT=m +CONFIG_INET6_XFRM_MODE_TUNNEL=m +CONFIG_INET6_XFRM_MODE_BEET=m +CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION=m +CONFIG_IPV6_VTI=m +CONFIG_IPV6_SIT=m +CONFIG_IPV6_GRE=m +CONFIG_IPV6_MULTIPLE_TABLES=y +CONFIG_IPV6_SUBTREES=y +CONFIG_NETFILTER=y +CONFIG_NF_CONNTRACK=m +CONFIG_NF_CONNTRACK_SECMARK=y +CONFIG_NF_CONNTRACK_EVENTS=y +CONFIG_NF_CONNTRACK_TIMEOUT=y +CONFIG_NF_CONNTRACK_TIMESTAMP=y +CONFIG_NF_CT_PROTO_DCCP=m +CONFIG_NF_CT_PROTO_UDPLITE=m +CONFIG_NF_CONNTRACK_AMANDA=m +CONFIG_NF_CONNTRACK_FTP=m +CONFIG_NF_CONNTRACK_H323=m +CONFIG_NF_CONNTRACK_IRC=m +CONFIG_NF_CONNTRACK_NETBIOS_NS=m +CONFIG_NF_CONNTRACK_SNMP=m +CONFIG_NF_CONNTRACK_PPTP=m +CONFIG_NF_CONNTRACK_SANE=m +CONFIG_NF_CONNTRACK_SIP=m +CONFIG_NF_CONNTRACK_TFTP=m +CONFIG_NF_CT_NETLINK=m +CONFIG_NF_CT_NETLINK_TIMEOUT=m +CONFIG_NF_TABLES=m +CONFIG_NFT_EXTHDR=m +CONFIG_NFT_META=m +CONFIG_NFT_CT=m +CONFIG_NFT_RBTREE=m +CONFIG_NFT_HASH=m +CONFIG_NFT_COUNTER=m +CONFIG_NFT_LOG=m +CONFIG_NFT_LIMIT=m +CONFIG_NFT_NAT=m +CONFIG_NFT_COMPAT=m +CONFIG_NETFILTER_XT_SET=m +CONFIG_NETFILTER_XT_TARGET_AUDIT=m +CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m +CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m +CONFIG_NETFILTER_XT_TARGET_CONNMARK=m +CONFIG_NETFILTER_XT_TARGET_CONNSECMARK=m +CONFIG_NETFILTER_XT_TARGET_CT=m +CONFIG_NETFILTER_XT_TARGET_DSCP=m +CONFIG_NETFILTER_XT_TARGET_HMARK=m +CONFIG_NETFILTER_XT_TARGET_IDLETIMER=m +CONFIG_NETFILTER_XT_TARGET_LOG=m +CONFIG_NETFILTER_XT_TARGET_MARK=m +CONFIG_NETFILTER_XT_TARGET_NFLOG=m +CONFIG_NETFILTER_XT_TARGET_NFQUEUE=m +CONFIG_NETFILTER_XT_TARGET_TEE=m +CONFIG_NETFILTER_XT_TARGET_TPROXY=m +CONFIG_NETFILTER_XT_TARGET_TRACE=m +CONFIG_NETFILTER_XT_TARGET_SECMARK=m +CONFIG_NETFILTER_XT_TARGET_TCPMSS=m +CONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP=m +CONFIG_NETFILTER_XT_MATCH_ADDRTYPE=m +CONFIG_NETFILTER_XT_MATCH_BPF=m +CONFIG_NETFILTER_XT_MATCH_CLUSTER=m +CONFIG_NETFILTER_XT_MATCH_COMMENT=m +CONFIG_NETFILTER_XT_MATCH_CONNBYTES=m +CONFIG_NETFILTER_XT_MATCH_CONNLABEL=m +CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=m +CONFIG_NETFILTER_XT_MATCH_CONNMARK=m +CONFIG_NETFILTER_XT_MATCH_CONNTRACK=m +CONFIG_NETFILTER_XT_MATCH_CPU=m +CONFIG_NETFILTER_XT_MATCH_DCCP=m +CONFIG_NETFILTER_XT_MATCH_DEVGROUP=m +CONFIG_NETFILTER_XT_MATCH_DSCP=m +CONFIG_NETFILTER_XT_MATCH_ESP=m +CONFIG_NETFILTER_XT_MATCH_HASHLIMIT=m +CONFIG_NETFILTER_XT_MATCH_HELPER=m +CONFIG_NETFILTER_XT_MATCH_IPRANGE=m +CONFIG_NETFILTER_XT_MATCH_IPVS=m +CONFIG_NETFILTER_XT_MATCH_LENGTH=m +CONFIG_NETFILTER_XT_MATCH_LIMIT=m +CONFIG_NETFILTER_XT_MATCH_MAC=m +CONFIG_NETFILTER_XT_MATCH_MARK=m +CONFIG_NETFILTER_XT_MATCH_MULTIPORT=m +CONFIG_NETFILTER_XT_MATCH_NFACCT=m +CONFIG_NETFILTER_XT_MATCH_OSF=m +CONFIG_NETFILTER_XT_MATCH_OWNER=m +CONFIG_NETFILTER_XT_MATCH_POLICY=m +CONFIG_NETFILTER_XT_MATCH_PHYSDEV=m +CONFIG_NETFILTER_XT_MATCH_PKTTYPE=m +CONFIG_NETFILTER_XT_MATCH_QUOTA=m +CONFIG_NETFILTER_XT_MATCH_RATEEST=m +CONFIG_NETFILTER_XT_MATCH_REALM=m +CONFIG_NETFILTER_XT_MATCH_RECENT=m +CONFIG_NETFILTER_XT_MATCH_SOCKET=m +CONFIG_NETFILTER_XT_MATCH_STATE=m +CONFIG_NETFILTER_XT_MATCH_STATISTIC=m +CONFIG_NETFILTER_XT_MATCH_STRING=m +CONFIG_NETFILTER_XT_MATCH_TCPMSS=m +CONFIG_NETFILTER_XT_MATCH_TIME=m +CONFIG_NETFILTER_XT_MATCH_U32=m +CONFIG_IP_SET=m +CONFIG_IP_SET_BITMAP_IP=m +CONFIG_IP_SET_BITMAP_IPMAC=m +CONFIG_IP_SET_BITMAP_PORT=m +CONFIG_IP_SET_HASH_IP=m +CONFIG_IP_SET_HASH_IPPORT=m +CONFIG_IP_SET_HASH_IPPORTIP=m +CONFIG_IP_SET_HASH_IPPORTNET=m +CONFIG_IP_SET_HASH_NETPORTNET=m +CONFIG_IP_SET_HASH_NET=m +CONFIG_IP_SET_HASH_NETNET=m +CONFIG_IP_SET_HASH_NETPORT=m +CONFIG_IP_SET_HASH_NETIFACE=m +CONFIG_IP_SET_LIST_SET=m +CONFIG_IP_VS=m +CONFIG_IP_VS_PROTO_TCP=y +CONFIG_IP_VS_PROTO_UDP=y +CONFIG_IP_VS_PROTO_ESP=y +CONFIG_IP_VS_PROTO_AH=y +CONFIG_IP_VS_RR=m +CONFIG_IP_VS_WRR=m +CONFIG_IP_VS_LC=m +CONFIG_IP_VS_WLC=m +CONFIG_IP_VS_LBLC=m +CONFIG_IP_VS_LBLCR=m +CONFIG_IP_VS_DH=m +CONFIG_IP_VS_SH=m +CONFIG_IP_VS_SED=m +CONFIG_IP_VS_NQ=m +CONFIG_IP_VS_FTP=m +CONFIG_IP_VS_PE_SIP=m +CONFIG_NF_CONNTRACK_IPV4=m +# CONFIG_NF_CONNTRACK_PROC_COMPAT is not set +CONFIG_NF_TABLES_IPV4=m +CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_CHAIN_NAT_IPV4=m +CONFIG_NF_TABLES_ARP=m +CONFIG_NF_NAT_IPV4=m +CONFIG_IP_NF_IPTABLES=m +CONFIG_IP_NF_MATCH_AH=m +CONFIG_IP_NF_MATCH_ECN=m +CONFIG_IP_NF_MATCH_RPFILTER=m +CONFIG_IP_NF_MATCH_TTL=m +CONFIG_IP_NF_FILTER=m +CONFIG_IP_NF_TARGET_REJECT=m +CONFIG_IP_NF_MANGLE=m +CONFIG_IP_NF_TARGET_CLUSTERIP=m +CONFIG_IP_NF_TARGET_ECN=m +CONFIG_IP_NF_TARGET_TTL=m +CONFIG_IP_NF_RAW=m +CONFIG_IP_NF_SECURITY=m +CONFIG_IP_NF_ARPTABLES=m +CONFIG_IP_NF_ARPFILTER=m +CONFIG_IP_NF_ARP_MANGLE=m +CONFIG_NF_CONNTRACK_IPV6=m +CONFIG_NF_TABLES_IPV6=m +CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_CHAIN_NAT_IPV6=m +CONFIG_NF_NAT_IPV6=m +CONFIG_IP6_NF_IPTABLES=m +CONFIG_IP6_NF_MATCH_AH=m +CONFIG_IP6_NF_MATCH_EUI64=m +CONFIG_IP6_NF_MATCH_FRAG=m +CONFIG_IP6_NF_MATCH_OPTS=m +CONFIG_IP6_NF_MATCH_HL=m +CONFIG_IP6_NF_MATCH_IPV6HEADER=m +CONFIG_IP6_NF_MATCH_MH=m +CONFIG_IP6_NF_MATCH_RPFILTER=m +CONFIG_IP6_NF_MATCH_RT=m +CONFIG_IP6_NF_TARGET_HL=m +CONFIG_IP6_NF_FILTER=m +CONFIG_IP6_NF_TARGET_REJECT=m +CONFIG_IP6_NF_MANGLE=m +CONFIG_IP6_NF_RAW=m +CONFIG_IP6_NF_SECURITY=m +CONFIG_NF_TABLES_BRIDGE=m +CONFIG_NET_SCTPPROBE=m +CONFIG_RDS=m +CONFIG_RDS_RDMA=m +CONFIG_RDS_TCP=m +CONFIG_L2TP=m +CONFIG_L2TP_DEBUGFS=m +CONFIG_L2TP_V3=y +CONFIG_L2TP_IP=m +CONFIG_L2TP_ETH=m +CONFIG_BRIDGE=m +CONFIG_VLAN_8021Q=m +CONFIG_VLAN_8021Q_GVRP=y +CONFIG_NET_SCHED=y +CONFIG_NET_SCH_CBQ=m +CONFIG_NET_SCH_HTB=m +CONFIG_NET_SCH_HFSC=m +CONFIG_NET_SCH_PRIO=m +CONFIG_NET_SCH_MULTIQ=m +CONFIG_NET_SCH_RED=m +CONFIG_NET_SCH_SFB=m +CONFIG_NET_SCH_SFQ=m +CONFIG_NET_SCH_TEQL=m +CONFIG_NET_SCH_TBF=m +CONFIG_NET_SCH_GRED=m +CONFIG_NET_SCH_DSMARK=m +CONFIG_NET_SCH_NETEM=m +CONFIG_NET_SCH_DRR=m +CONFIG_NET_SCH_MQPRIO=m +CONFIG_NET_SCH_CHOKE=m +CONFIG_NET_SCH_QFQ=m +CONFIG_NET_SCH_CODEL=m +CONFIG_NET_SCH_FQ_CODEL=m +CONFIG_NET_SCH_INGRESS=m +CONFIG_NET_SCH_PLUG=m +CONFIG_NET_CLS_BASIC=m +CONFIG_NET_CLS_TCINDEX=m +CONFIG_NET_CLS_ROUTE4=m +CONFIG_NET_CLS_FW=m +CONFIG_NET_CLS_U32=m +CONFIG_CLS_U32_PERF=y +CONFIG_CLS_U32_MARK=y +CONFIG_NET_CLS_RSVP=m +CONFIG_NET_CLS_RSVP6=m +CONFIG_NET_CLS_FLOW=m +CONFIG_NET_CLS_CGROUP=y +CONFIG_NET_CLS_BPF=m +CONFIG_NET_CLS_ACT=y +CONFIG_NET_ACT_POLICE=m +CONFIG_NET_ACT_GACT=m +CONFIG_GACT_PROB=y +CONFIG_NET_ACT_MIRRED=m +CONFIG_NET_ACT_IPT=m +CONFIG_NET_ACT_NAT=m +CONFIG_NET_ACT_PEDIT=m +CONFIG_NET_ACT_SIMP=m +CONFIG_NET_ACT_SKBEDIT=m +CONFIG_NET_ACT_CSUM=m +CONFIG_DNS_RESOLVER=y +CONFIG_BPF_JIT=y +CONFIG_NET_PKTGEN=m +CONFIG_NET_TCPPROBE=m +CONFIG_DEVTMPFS=y +CONFIG_CONNECTOR=y +CONFIG_BLK_DEV_LOOP=m +CONFIG_BLK_DEV_CRYPTOLOOP=m +CONFIG_BLK_DEV_NBD=m +CONFIG_BLK_DEV_OSD=m +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_SIZE=32768 +CONFIG_BLK_DEV_XIP=y +CONFIG_CDROM_PKTCDVD=m +CONFIG_ATA_OVER_ETH=m +CONFIG_VIRTIO_BLK=y +CONFIG_ENCLOSURE_SERVICES=m +CONFIG_RAID_ATTRS=m +CONFIG_SCSI=y +CONFIG_BLK_DEV_SD=y +CONFIG_CHR_DEV_ST=m +CONFIG_CHR_DEV_OSST=m +CONFIG_BLK_DEV_SR=m +CONFIG_CHR_DEV_SG=y +CONFIG_CHR_DEV_SCH=m +CONFIG_SCSI_ENCLOSURE=m +CONFIG_SCSI_CONSTANTS=y +CONFIG_SCSI_LOGGING=y +CONFIG_SCSI_SPI_ATTRS=m +CONFIG_SCSI_FC_ATTRS=y +CONFIG_SCSI_SAS_LIBSAS=m +CONFIG_SCSI_SRP_ATTRS=m +CONFIG_ISCSI_TCP=m +CONFIG_SCSI_DEBUG=m +CONFIG_ZFCP=y +CONFIG_SCSI_VIRTIO=m +CONFIG_SCSI_DH=m +CONFIG_SCSI_DH_RDAC=m +CONFIG_SCSI_DH_HP_SW=m +CONFIG_SCSI_DH_EMC=m +CONFIG_SCSI_DH_ALUA=m +CONFIG_SCSI_OSD_INITIATOR=m +CONFIG_SCSI_OSD_ULD=m +CONFIG_MD=y +CONFIG_BLK_DEV_MD=y +CONFIG_MD_LINEAR=m +CONFIG_MD_RAID0=m +CONFIG_MD_MULTIPATH=m +CONFIG_MD_FAULTY=m +CONFIG_BLK_DEV_DM=m +CONFIG_DM_CRYPT=m +CONFIG_DM_SNAPSHOT=m +CONFIG_DM_MIRROR=m +CONFIG_DM_LOG_USERSPACE=m +CONFIG_DM_RAID=m +CONFIG_DM_ZERO=m +CONFIG_DM_MULTIPATH=m +CONFIG_DM_MULTIPATH_QL=m +CONFIG_DM_MULTIPATH_ST=m +CONFIG_DM_DELAY=m +CONFIG_DM_UEVENT=y +CONFIG_DM_FLAKEY=m +CONFIG_DM_VERITY=m +CONFIG_DM_SWITCH=m +CONFIG_NETDEVICES=y +CONFIG_BONDING=m +CONFIG_DUMMY=m +CONFIG_EQUALIZER=m +CONFIG_IFB=m +CONFIG_MACVLAN=m +CONFIG_MACVTAP=m +CONFIG_VXLAN=m +CONFIG_TUN=m +CONFIG_VETH=m +CONFIG_VIRTIO_NET=m +CONFIG_NLMON=m +CONFIG_VHOST_NET=m +# CONFIG_NET_VENDOR_ARC is not set +# CONFIG_NET_VENDOR_CHELSIO is not set +# CONFIG_NET_VENDOR_INTEL is not set +# CONFIG_NET_VENDOR_MARVELL is not set +CONFIG_MLX4_EN=m +# CONFIG_NET_VENDOR_NATSEMI is not set +CONFIG_PPP=m +CONFIG_PPP_BSDCOMP=m +CONFIG_PPP_DEFLATE=m +CONFIG_PPP_MPPE=m +CONFIG_PPPOE=m +CONFIG_PPTP=m +CONFIG_PPPOL2TP=m +CONFIG_PPP_ASYNC=m +CONFIG_PPP_SYNC_TTY=m +# CONFIG_INPUT_MOUSEDEV_PSAUX is not set +# CONFIG_INPUT_KEYBOARD is not set +# CONFIG_INPUT_MOUSE is not set +# CONFIG_SERIO is not set +CONFIG_DEVPTS_MULTIPLE_INSTANCES=y +CONFIG_LEGACY_PTY_COUNT=0 +CONFIG_HW_RANDOM_VIRTIO=m +CONFIG_RAW_DRIVER=m +CONFIG_HANGCHECK_TIMER=m +CONFIG_TN3270_FS=y +CONFIG_WATCHDOG=y +CONFIG_WATCHDOG_NOWAYOUT=y +CONFIG_SOFT_WATCHDOG=m +CONFIG_DIAG288_WATCHDOG=m +# CONFIG_HID is not set +# CONFIG_USB_SUPPORT is not set +CONFIG_INFINIBAND=m +CONFIG_INFINIBAND_USER_ACCESS=m +CONFIG_MLX4_INFINIBAND=m +CONFIG_VIRTIO_BALLOON=m +# CONFIG_IOMMU_SUPPORT is not set +CONFIG_EXT2_FS=y +CONFIG_EXT2_FS_XATTR=y +CONFIG_EXT2_FS_POSIX_ACL=y +CONFIG_EXT2_FS_SECURITY=y +CONFIG_EXT2_FS_XIP=y +CONFIG_EXT3_FS=y +# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set +CONFIG_EXT3_FS_POSIX_ACL=y +CONFIG_EXT3_FS_SECURITY=y +CONFIG_EXT4_FS=y +CONFIG_EXT4_FS_POSIX_ACL=y +CONFIG_EXT4_FS_SECURITY=y +CONFIG_JBD_DEBUG=y +CONFIG_JBD2_DEBUG=y +CONFIG_JFS_FS=m +CONFIG_JFS_POSIX_ACL=y +CONFIG_JFS_SECURITY=y +CONFIG_JFS_STATISTICS=y +CONFIG_XFS_FS=y +CONFIG_XFS_QUOTA=y +CONFIG_XFS_POSIX_ACL=y +CONFIG_XFS_RT=y +CONFIG_GFS2_FS=m +CONFIG_OCFS2_FS=m +CONFIG_BTRFS_FS=y +CONFIG_BTRFS_FS_POSIX_ACL=y +CONFIG_NILFS2_FS=m +CONFIG_FANOTIFY=y +CONFIG_QUOTA_NETLINK_INTERFACE=y +CONFIG_QFMT_V1=m +CONFIG_QFMT_V2=m +CONFIG_AUTOFS4_FS=m +CONFIG_FUSE_FS=m +CONFIG_CUSE=m +CONFIG_FSCACHE=m +CONFIG_CACHEFILES=m +CONFIG_ISO9660_FS=y +CONFIG_JOLIET=y +CONFIG_ZISOFS=y +CONFIG_UDF_FS=m +CONFIG_MSDOS_FS=m +CONFIG_VFAT_FS=m +CONFIG_NTFS_FS=m +CONFIG_NTFS_RW=y +CONFIG_PROC_KCORE=y +CONFIG_TMPFS=y +CONFIG_TMPFS_POSIX_ACL=y +CONFIG_HUGETLBFS=y +CONFIG_CONFIGFS_FS=m +CONFIG_ECRYPT_FS=m +CONFIG_CRAMFS=m +CONFIG_SQUASHFS=m +CONFIG_SQUASHFS_XATTR=y +CONFIG_SQUASHFS_LZO=y +CONFIG_SQUASHFS_XZ=y +CONFIG_ROMFS_FS=m +CONFIG_NFS_FS=m +CONFIG_NFS_V3_ACL=y +CONFIG_NFS_V4=m +CONFIG_NFS_SWAP=y +CONFIG_NFSD=m +CONFIG_NFSD_V3_ACL=y +CONFIG_NFSD_V4=y +CONFIG_NFSD_V4_SECURITY_LABEL=y +CONFIG_CIFS=m +CONFIG_CIFS_STATS=y +CONFIG_CIFS_STATS2=y +CONFIG_CIFS_WEAK_PW_HASH=y +CONFIG_CIFS_UPCALL=y +CONFIG_CIFS_XATTR=y +CONFIG_CIFS_POSIX=y +# CONFIG_CIFS_DEBUG is not set +CONFIG_CIFS_DFS_UPCALL=y +CONFIG_NLS_DEFAULT="utf8" +CONFIG_NLS_CODEPAGE_437=m +CONFIG_NLS_CODEPAGE_850=m +CONFIG_NLS_ASCII=m +CONFIG_NLS_ISO8859_1=m +CONFIG_NLS_ISO8859_15=m +CONFIG_NLS_UTF8=m +CONFIG_DLM=m +CONFIG_PRINTK_TIME=y +CONFIG_DEBUG_INFO=y +# CONFIG_ENABLE_MUST_CHECK is not set +CONFIG_FRAME_WARN=1024 +CONFIG_UNUSED_SYMBOLS=y +CONFIG_MAGIC_SYSRQ=y +CONFIG_DEBUG_KERNEL=y +CONFIG_PANIC_ON_OOPS=y +CONFIG_TIMER_STATS=y +CONFIG_RCU_TORTURE_TEST=m +CONFIG_RCU_CPU_STALL_TIMEOUT=60 +CONFIG_LATENCYTOP=y +CONFIG_SCHED_TRACER=y +CONFIG_FTRACE_SYSCALLS=y +CONFIG_STACK_TRACER=y +CONFIG_BLK_DEV_IO_TRACE=y +CONFIG_UPROBE_EVENT=y +CONFIG_LKDTM=m +CONFIG_PERCPU_TEST=m +CONFIG_ATOMIC64_SELFTEST=y +# CONFIG_STRICT_DEVMEM is not set +CONFIG_S390_PTDUMP=y +CONFIG_ENCRYPTED_KEYS=m +CONFIG_KEYS_DEBUG_PROC_KEYS=y +CONFIG_SECURITY=y +CONFIG_SECURITY_NETWORK=y +CONFIG_SECURITY_SELINUX=y +CONFIG_SECURITY_SELINUX_BOOTPARAM=y +CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=0 +CONFIG_SECURITY_SELINUX_DISABLE=y +CONFIG_IMA=y +CONFIG_IMA_APPRAISE=y +CONFIG_CRYPTO_USER=m +# CONFIG_CRYPTO_MANAGER_DISABLE_TESTS is not set +CONFIG_CRYPTO_CRYPTD=m +CONFIG_CRYPTO_TEST=m +CONFIG_CRYPTO_CCM=m +CONFIG_CRYPTO_GCM=m +CONFIG_CRYPTO_CTS=m +CONFIG_CRYPTO_LRW=m +CONFIG_CRYPTO_PCBC=m +CONFIG_CRYPTO_XTS=m +CONFIG_CRYPTO_XCBC=m +CONFIG_CRYPTO_VMAC=m +CONFIG_CRYPTO_CRC32=m +CONFIG_CRYPTO_MICHAEL_MIC=m +CONFIG_CRYPTO_RMD128=m +CONFIG_CRYPTO_RMD160=m +CONFIG_CRYPTO_RMD256=m +CONFIG_CRYPTO_RMD320=m +CONFIG_CRYPTO_SHA512=m +CONFIG_CRYPTO_TGR192=m +CONFIG_CRYPTO_WP512=m +CONFIG_CRYPTO_ANUBIS=m +CONFIG_CRYPTO_BLOWFISH=m +CONFIG_CRYPTO_CAMELLIA=m +CONFIG_CRYPTO_CAST5=m +CONFIG_CRYPTO_CAST6=m +CONFIG_CRYPTO_FCRYPT=m +CONFIG_CRYPTO_KHAZAD=m +CONFIG_CRYPTO_SALSA20=m +CONFIG_CRYPTO_SEED=m +CONFIG_CRYPTO_SERPENT=m +CONFIG_CRYPTO_TEA=m +CONFIG_CRYPTO_TWOFISH=m +CONFIG_CRYPTO_ZLIB=y +CONFIG_CRYPTO_LZO=m +CONFIG_CRYPTO_LZ4=m +CONFIG_CRYPTO_LZ4HC=m +CONFIG_CRYPTO_USER_API_HASH=m +CONFIG_CRYPTO_USER_API_SKCIPHER=m +CONFIG_ZCRYPT=m +CONFIG_CRYPTO_SHA1_S390=m +CONFIG_CRYPTO_SHA256_S390=m +CONFIG_CRYPTO_SHA512_S390=m +CONFIG_CRYPTO_DES_S390=m +CONFIG_CRYPTO_AES_S390=m +CONFIG_CRYPTO_GHASH_S390=m +CONFIG_ASYMMETRIC_KEY_TYPE=m +CONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=m +CONFIG_X509_CERTIFICATE_PARSER=m +CONFIG_CRC7=m +CONFIG_CRC8=m +CONFIG_CORDIC=m +CONFIG_CMM=m +CONFIG_APPLDATA_BASE=y +CONFIG_KVM=m +CONFIG_KVM_S390_UCONTROL=y diff --git a/arch/s390/configs/zfcpdump_defconfig b/arch/s390/configs/zfcpdump_defconfig new file mode 100644 index 000000000..1b0184a0f --- /dev/null +++ b/arch/s390/configs/zfcpdump_defconfig @@ -0,0 +1,82 @@ +# CONFIG_SWAP is not set +CONFIG_NO_HZ=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_RCU_FAST_NO_HZ=y +CONFIG_BLK_DEV_INITRD=y +CONFIG_CC_OPTIMIZE_FOR_SIZE=y +# CONFIG_COMPAT_BRK is not set +CONFIG_PARTITION_ADVANCED=y +CONFIG_IBM_PARTITION=y +CONFIG_DEFAULT_DEADLINE=y +CONFIG_MARCH_Z196=y +CONFIG_TUNE_ZEC12=y +# CONFIG_COMPAT is not set +CONFIG_NR_CPUS=2 +# CONFIG_HOTPLUG_CPU is not set +CONFIG_HZ_100=y +# CONFIG_COMPACTION is not set +# CONFIG_MIGRATION is not set +# CONFIG_CHECK_STACK is not set +# CONFIG_CHSC_SCH is not set +# CONFIG_SCM_BUS is not set +CONFIG_CRASH_DUMP=y +# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set +# CONFIG_SECCOMP is not set +CONFIG_NET=y +# CONFIG_IUCV is not set +CONFIG_ATM=y +CONFIG_ATM_LANE=y +CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug" +CONFIG_DEVTMPFS=y +# CONFIG_FIRMWARE_IN_KERNEL is not set +# CONFIG_BLK_DEV_XPRAM is not set +# CONFIG_DCSSBLK is not set +# CONFIG_DASD is not set +CONFIG_ENCLOSURE_SERVICES=y +CONFIG_SCSI=y +CONFIG_BLK_DEV_SD=y +CONFIG_SCSI_ENCLOSURE=y +CONFIG_SCSI_CONSTANTS=y +CONFIG_SCSI_LOGGING=y +CONFIG_SCSI_FC_ATTRS=y +CONFIG_SCSI_SRP_ATTRS=y +CONFIG_ZFCP=y +# CONFIG_INPUT_MOUSEDEV_PSAUX is not set +# CONFIG_INPUT_KEYBOARD is not set +# CONFIG_INPUT_MOUSE is not set +# CONFIG_SERIO is not set +# CONFIG_HVC_IUCV is not set +CONFIG_RAW_DRIVER=y +# CONFIG_SCLP_ASYNC is not set +# CONFIG_HMC_DRV is not set +# CONFIG_S390_TAPE is not set +# CONFIG_VMCP is not set +# CONFIG_MONWRITER is not set +# CONFIG_S390_VMUR is not set +# CONFIG_HID is not set +CONFIG_MEMSTICK=y +CONFIG_MEMSTICK_DEBUG=y +CONFIG_MEMSTICK_UNSAFE_RESUME=y +CONFIG_MSPRO_BLOCK=y +# CONFIG_IOMMU_SUPPORT is not set +CONFIG_EXT2_FS=y +CONFIG_EXT3_FS=y +# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set +CONFIG_EXT4_FS=y +CONFIG_EXT4_FS_POSIX_ACL=y +CONFIG_EXT4_FS_SECURITY=y +# CONFIG_INOTIFY_USER is not set +CONFIG_CONFIGFS_FS=y +CONFIG_PRINTK_TIME=y +CONFIG_DEBUG_INFO=y +CONFIG_DEBUG_FS=y +CONFIG_DEBUG_KERNEL=y +CONFIG_PANIC_ON_OOPS=y +# CONFIG_SCHED_DEBUG is not set +CONFIG_RCU_CPU_STALL_TIMEOUT=60 +# CONFIG_FTRACE is not set +# CONFIG_STRICT_DEVMEM is not set +# CONFIG_PFAULT is not set +# CONFIG_S390_HYPFS_FS is not set +# CONFIG_VIRTUALIZATION is not set +# CONFIG_S390_GUEST is not set diff --git a/arch/s390/crypto/Makefile b/arch/s390/crypto/Makefile new file mode 100644 index 000000000..7f0b7cda6 --- /dev/null +++ b/arch/s390/crypto/Makefile @@ -0,0 +1,11 @@ +# +# Cryptographic API +# + +obj-$(CONFIG_CRYPTO_SHA1_S390) += sha1_s390.o sha_common.o +obj-$(CONFIG_CRYPTO_SHA256_S390) += sha256_s390.o sha_common.o +obj-$(CONFIG_CRYPTO_SHA512_S390) += sha512_s390.o sha_common.o +obj-$(CONFIG_CRYPTO_DES_S390) += des_s390.o +obj-$(CONFIG_CRYPTO_AES_S390) += aes_s390.o +obj-$(CONFIG_S390_PRNG) += prng.o +obj-$(CONFIG_CRYPTO_GHASH_S390) += ghash_s390.o diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c new file mode 100644 index 000000000..5566ce80a --- /dev/null +++ b/arch/s390/crypto/aes_s390.c @@ -0,0 +1,985 @@ +/* + * Cryptographic API. + * + * s390 implementation of the AES Cipher Algorithm. + * + * s390 Version: + * Copyright IBM Corp. 2005, 2007 + * Author(s): Jan Glauber (jang@de.ibm.com) + * Sebastian Siewior (sebastian@breakpoint.cc> SW-Fallback + * + * Derived from "crypto/aes_generic.c" + * + * 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. + * + */ + +#define KMSG_COMPONENT "aes_s390" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <crypto/aes.h> +#include <crypto/algapi.h> +#include <linux/err.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/spinlock.h> +#include "crypt_s390.h" + +#define AES_KEYLEN_128 1 +#define AES_KEYLEN_192 2 +#define AES_KEYLEN_256 4 + +static u8 *ctrblk; +static DEFINE_SPINLOCK(ctrblk_lock); +static char keylen_flag; + +struct s390_aes_ctx { + u8 key[AES_MAX_KEY_SIZE]; + long enc; + long dec; + int key_len; + union { + struct crypto_blkcipher *blk; + struct crypto_cipher *cip; + } fallback; +}; + +struct pcc_param { + u8 key[32]; + u8 tweak[16]; + u8 block[16]; + u8 bit[16]; + u8 xts[16]; +}; + +struct s390_xts_ctx { + u8 key[32]; + u8 pcc_key[32]; + long enc; + long dec; + int key_len; + struct crypto_blkcipher *fallback; +}; + +/* + * Check if the key_len is supported by the HW. + * Returns 0 if it is, a positive number if it is not and software fallback is + * required or a negative number in case the key size is not valid + */ +static int need_fallback(unsigned int key_len) +{ + switch (key_len) { + case 16: + if (!(keylen_flag & AES_KEYLEN_128)) + return 1; + break; + case 24: + if (!(keylen_flag & AES_KEYLEN_192)) + return 1; + break; + case 32: + if (!(keylen_flag & AES_KEYLEN_256)) + return 1; + break; + default: + return -1; + break; + } + return 0; +} + +static int setkey_fallback_cip(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + int ret; + + sctx->fallback.cip->base.crt_flags &= ~CRYPTO_TFM_REQ_MASK; + sctx->fallback.cip->base.crt_flags |= (tfm->crt_flags & + CRYPTO_TFM_REQ_MASK); + + ret = crypto_cipher_setkey(sctx->fallback.cip, in_key, key_len); + if (ret) { + tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK; + tfm->crt_flags |= (sctx->fallback.cip->base.crt_flags & + CRYPTO_TFM_RES_MASK); + } + return ret; +} + +static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + u32 *flags = &tfm->crt_flags; + int ret; + + ret = need_fallback(key_len); + if (ret < 0) { + *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + + sctx->key_len = key_len; + if (!ret) { + memcpy(sctx->key, in_key, key_len); + return 0; + } + + return setkey_fallback_cip(tfm, in_key, key_len); +} + +static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + if (unlikely(need_fallback(sctx->key_len))) { + crypto_cipher_encrypt_one(sctx->fallback.cip, out, in); + return; + } + + switch (sctx->key_len) { + case 16: + crypt_s390_km(KM_AES_128_ENCRYPT, &sctx->key, out, in, + AES_BLOCK_SIZE); + break; + case 24: + crypt_s390_km(KM_AES_192_ENCRYPT, &sctx->key, out, in, + AES_BLOCK_SIZE); + break; + case 32: + crypt_s390_km(KM_AES_256_ENCRYPT, &sctx->key, out, in, + AES_BLOCK_SIZE); + break; + } +} + +static void aes_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + if (unlikely(need_fallback(sctx->key_len))) { + crypto_cipher_decrypt_one(sctx->fallback.cip, out, in); + return; + } + + switch (sctx->key_len) { + case 16: + crypt_s390_km(KM_AES_128_DECRYPT, &sctx->key, out, in, + AES_BLOCK_SIZE); + break; + case 24: + crypt_s390_km(KM_AES_192_DECRYPT, &sctx->key, out, in, + AES_BLOCK_SIZE); + break; + case 32: + crypt_s390_km(KM_AES_256_DECRYPT, &sctx->key, out, in, + AES_BLOCK_SIZE); + break; + } +} + +static int fallback_init_cip(struct crypto_tfm *tfm) +{ + const char *name = tfm->__crt_alg->cra_name; + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + sctx->fallback.cip = crypto_alloc_cipher(name, 0, + CRYPTO_ALG_ASYNC | CRYPTO_ALG_NEED_FALLBACK); + + if (IS_ERR(sctx->fallback.cip)) { + pr_err("Allocating AES fallback algorithm %s failed\n", + name); + return PTR_ERR(sctx->fallback.cip); + } + + return 0; +} + +static void fallback_exit_cip(struct crypto_tfm *tfm) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + crypto_free_cipher(sctx->fallback.cip); + sctx->fallback.cip = NULL; +} + +static struct crypto_alg aes_alg = { + .cra_name = "aes", + .cra_driver_name = "aes-s390", + .cra_priority = CRYPT_S390_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_CIPHER | + CRYPTO_ALG_NEED_FALLBACK, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_aes_ctx), + .cra_module = THIS_MODULE, + .cra_init = fallback_init_cip, + .cra_exit = fallback_exit_cip, + .cra_u = { + .cipher = { + .cia_min_keysize = AES_MIN_KEY_SIZE, + .cia_max_keysize = AES_MAX_KEY_SIZE, + .cia_setkey = aes_set_key, + .cia_encrypt = aes_encrypt, + .cia_decrypt = aes_decrypt, + } + } +}; + +static int setkey_fallback_blk(struct crypto_tfm *tfm, const u8 *key, + unsigned int len) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + unsigned int ret; + + sctx->fallback.blk->base.crt_flags &= ~CRYPTO_TFM_REQ_MASK; + sctx->fallback.blk->base.crt_flags |= (tfm->crt_flags & + CRYPTO_TFM_REQ_MASK); + + ret = crypto_blkcipher_setkey(sctx->fallback.blk, key, len); + if (ret) { + tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK; + tfm->crt_flags |= (sctx->fallback.blk->base.crt_flags & + CRYPTO_TFM_RES_MASK); + } + return ret; +} + +static int fallback_blk_dec(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + unsigned int ret; + struct crypto_blkcipher *tfm; + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + + tfm = desc->tfm; + desc->tfm = sctx->fallback.blk; + + ret = crypto_blkcipher_decrypt_iv(desc, dst, src, nbytes); + + desc->tfm = tfm; + return ret; +} + +static int fallback_blk_enc(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + unsigned int ret; + struct crypto_blkcipher *tfm; + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + + tfm = desc->tfm; + desc->tfm = sctx->fallback.blk; + + ret = crypto_blkcipher_encrypt_iv(desc, dst, src, nbytes); + + desc->tfm = tfm; + return ret; +} + +static int ecb_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + int ret; + + ret = need_fallback(key_len); + if (ret > 0) { + sctx->key_len = key_len; + return setkey_fallback_blk(tfm, in_key, key_len); + } + + switch (key_len) { + case 16: + sctx->enc = KM_AES_128_ENCRYPT; + sctx->dec = KM_AES_128_DECRYPT; + break; + case 24: + sctx->enc = KM_AES_192_ENCRYPT; + sctx->dec = KM_AES_192_DECRYPT; + break; + case 32: + sctx->enc = KM_AES_256_ENCRYPT; + sctx->dec = KM_AES_256_DECRYPT; + break; + } + + return aes_set_key(tfm, in_key, key_len); +} + +static int ecb_aes_crypt(struct blkcipher_desc *desc, long func, void *param, + struct blkcipher_walk *walk) +{ + int ret = blkcipher_walk_virt(desc, walk); + unsigned int nbytes; + + while ((nbytes = walk->nbytes)) { + /* only use complete blocks */ + unsigned int n = nbytes & ~(AES_BLOCK_SIZE - 1); + u8 *out = walk->dst.virt.addr; + u8 *in = walk->src.virt.addr; + + ret = crypt_s390_km(func, param, out, in, n); + if (ret < 0 || ret != n) + return -EIO; + + nbytes &= AES_BLOCK_SIZE - 1; + ret = blkcipher_walk_done(desc, walk, nbytes); + } + + return ret; +} + +static int ecb_aes_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + if (unlikely(need_fallback(sctx->key_len))) + return fallback_blk_enc(desc, dst, src, nbytes); + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_aes_crypt(desc, sctx->enc, sctx->key, &walk); +} + +static int ecb_aes_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + if (unlikely(need_fallback(sctx->key_len))) + return fallback_blk_dec(desc, dst, src, nbytes); + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_aes_crypt(desc, sctx->dec, sctx->key, &walk); +} + +static int fallback_init_blk(struct crypto_tfm *tfm) +{ + const char *name = tfm->__crt_alg->cra_name; + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + sctx->fallback.blk = crypto_alloc_blkcipher(name, 0, + CRYPTO_ALG_ASYNC | CRYPTO_ALG_NEED_FALLBACK); + + if (IS_ERR(sctx->fallback.blk)) { + pr_err("Allocating AES fallback algorithm %s failed\n", + name); + return PTR_ERR(sctx->fallback.blk); + } + + return 0; +} + +static void fallback_exit_blk(struct crypto_tfm *tfm) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + crypto_free_blkcipher(sctx->fallback.blk); + sctx->fallback.blk = NULL; +} + +static struct crypto_alg ecb_aes_alg = { + .cra_name = "ecb(aes)", + .cra_driver_name = "ecb-aes-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER | + CRYPTO_ALG_NEED_FALLBACK, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_aes_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_init = fallback_init_blk, + .cra_exit = fallback_exit_blk, + .cra_u = { + .blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .setkey = ecb_aes_set_key, + .encrypt = ecb_aes_encrypt, + .decrypt = ecb_aes_decrypt, + } + } +}; + +static int cbc_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + int ret; + + ret = need_fallback(key_len); + if (ret > 0) { + sctx->key_len = key_len; + return setkey_fallback_blk(tfm, in_key, key_len); + } + + switch (key_len) { + case 16: + sctx->enc = KMC_AES_128_ENCRYPT; + sctx->dec = KMC_AES_128_DECRYPT; + break; + case 24: + sctx->enc = KMC_AES_192_ENCRYPT; + sctx->dec = KMC_AES_192_DECRYPT; + break; + case 32: + sctx->enc = KMC_AES_256_ENCRYPT; + sctx->dec = KMC_AES_256_DECRYPT; + break; + } + + return aes_set_key(tfm, in_key, key_len); +} + +static int cbc_aes_crypt(struct blkcipher_desc *desc, long func, + struct blkcipher_walk *walk) +{ + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + int ret = blkcipher_walk_virt(desc, walk); + unsigned int nbytes = walk->nbytes; + struct { + u8 iv[AES_BLOCK_SIZE]; + u8 key[AES_MAX_KEY_SIZE]; + } param; + + if (!nbytes) + goto out; + + memcpy(param.iv, walk->iv, AES_BLOCK_SIZE); + memcpy(param.key, sctx->key, sctx->key_len); + do { + /* only use complete blocks */ + unsigned int n = nbytes & ~(AES_BLOCK_SIZE - 1); + u8 *out = walk->dst.virt.addr; + u8 *in = walk->src.virt.addr; + + ret = crypt_s390_kmc(func, ¶m, out, in, n); + if (ret < 0 || ret != n) + return -EIO; + + nbytes &= AES_BLOCK_SIZE - 1; + ret = blkcipher_walk_done(desc, walk, nbytes); + } while ((nbytes = walk->nbytes)); + memcpy(walk->iv, param.iv, AES_BLOCK_SIZE); + +out: + return ret; +} + +static int cbc_aes_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + if (unlikely(need_fallback(sctx->key_len))) + return fallback_blk_enc(desc, dst, src, nbytes); + + blkcipher_walk_init(&walk, dst, src, nbytes); + return cbc_aes_crypt(desc, sctx->enc, &walk); +} + +static int cbc_aes_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + if (unlikely(need_fallback(sctx->key_len))) + return fallback_blk_dec(desc, dst, src, nbytes); + + blkcipher_walk_init(&walk, dst, src, nbytes); + return cbc_aes_crypt(desc, sctx->dec, &walk); +} + +static struct crypto_alg cbc_aes_alg = { + .cra_name = "cbc(aes)", + .cra_driver_name = "cbc-aes-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER | + CRYPTO_ALG_NEED_FALLBACK, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_aes_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_init = fallback_init_blk, + .cra_exit = fallback_exit_blk, + .cra_u = { + .blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = cbc_aes_set_key, + .encrypt = cbc_aes_encrypt, + .decrypt = cbc_aes_decrypt, + } + } +}; + +static int xts_fallback_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int len) +{ + struct s390_xts_ctx *xts_ctx = crypto_tfm_ctx(tfm); + unsigned int ret; + + xts_ctx->fallback->base.crt_flags &= ~CRYPTO_TFM_REQ_MASK; + xts_ctx->fallback->base.crt_flags |= (tfm->crt_flags & + CRYPTO_TFM_REQ_MASK); + + ret = crypto_blkcipher_setkey(xts_ctx->fallback, key, len); + if (ret) { + tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK; + tfm->crt_flags |= (xts_ctx->fallback->base.crt_flags & + CRYPTO_TFM_RES_MASK); + } + return ret; +} + +static int xts_fallback_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_xts_ctx *xts_ctx = crypto_blkcipher_ctx(desc->tfm); + struct crypto_blkcipher *tfm; + unsigned int ret; + + tfm = desc->tfm; + desc->tfm = xts_ctx->fallback; + + ret = crypto_blkcipher_decrypt_iv(desc, dst, src, nbytes); + + desc->tfm = tfm; + return ret; +} + +static int xts_fallback_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_xts_ctx *xts_ctx = crypto_blkcipher_ctx(desc->tfm); + struct crypto_blkcipher *tfm; + unsigned int ret; + + tfm = desc->tfm; + desc->tfm = xts_ctx->fallback; + + ret = crypto_blkcipher_encrypt_iv(desc, dst, src, nbytes); + + desc->tfm = tfm; + return ret; +} + +static int xts_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct s390_xts_ctx *xts_ctx = crypto_tfm_ctx(tfm); + u32 *flags = &tfm->crt_flags; + + switch (key_len) { + case 32: + xts_ctx->enc = KM_XTS_128_ENCRYPT; + xts_ctx->dec = KM_XTS_128_DECRYPT; + memcpy(xts_ctx->key + 16, in_key, 16); + memcpy(xts_ctx->pcc_key + 16, in_key + 16, 16); + break; + case 48: + xts_ctx->enc = 0; + xts_ctx->dec = 0; + xts_fallback_setkey(tfm, in_key, key_len); + break; + case 64: + xts_ctx->enc = KM_XTS_256_ENCRYPT; + xts_ctx->dec = KM_XTS_256_DECRYPT; + memcpy(xts_ctx->key, in_key, 32); + memcpy(xts_ctx->pcc_key, in_key + 32, 32); + break; + default: + *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + xts_ctx->key_len = key_len; + return 0; +} + +static int xts_aes_crypt(struct blkcipher_desc *desc, long func, + struct s390_xts_ctx *xts_ctx, + struct blkcipher_walk *walk) +{ + unsigned int offset = (xts_ctx->key_len >> 1) & 0x10; + int ret = blkcipher_walk_virt(desc, walk); + unsigned int nbytes = walk->nbytes; + unsigned int n; + u8 *in, *out; + struct pcc_param pcc_param; + struct { + u8 key[32]; + u8 init[16]; + } xts_param; + + if (!nbytes) + goto out; + + memset(pcc_param.block, 0, sizeof(pcc_param.block)); + memset(pcc_param.bit, 0, sizeof(pcc_param.bit)); + memset(pcc_param.xts, 0, sizeof(pcc_param.xts)); + memcpy(pcc_param.tweak, walk->iv, sizeof(pcc_param.tweak)); + memcpy(pcc_param.key, xts_ctx->pcc_key, 32); + ret = crypt_s390_pcc(func, &pcc_param.key[offset]); + if (ret < 0) + return -EIO; + + memcpy(xts_param.key, xts_ctx->key, 32); + memcpy(xts_param.init, pcc_param.xts, 16); + do { + /* only use complete blocks */ + n = nbytes & ~(AES_BLOCK_SIZE - 1); + out = walk->dst.virt.addr; + in = walk->src.virt.addr; + + ret = crypt_s390_km(func, &xts_param.key[offset], out, in, n); + if (ret < 0 || ret != n) + return -EIO; + + nbytes &= AES_BLOCK_SIZE - 1; + ret = blkcipher_walk_done(desc, walk, nbytes); + } while ((nbytes = walk->nbytes)); +out: + return ret; +} + +static int xts_aes_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_xts_ctx *xts_ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + if (unlikely(xts_ctx->key_len == 48)) + return xts_fallback_encrypt(desc, dst, src, nbytes); + + blkcipher_walk_init(&walk, dst, src, nbytes); + return xts_aes_crypt(desc, xts_ctx->enc, xts_ctx, &walk); +} + +static int xts_aes_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_xts_ctx *xts_ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + if (unlikely(xts_ctx->key_len == 48)) + return xts_fallback_decrypt(desc, dst, src, nbytes); + + blkcipher_walk_init(&walk, dst, src, nbytes); + return xts_aes_crypt(desc, xts_ctx->dec, xts_ctx, &walk); +} + +static int xts_fallback_init(struct crypto_tfm *tfm) +{ + const char *name = tfm->__crt_alg->cra_name; + struct s390_xts_ctx *xts_ctx = crypto_tfm_ctx(tfm); + + xts_ctx->fallback = crypto_alloc_blkcipher(name, 0, + CRYPTO_ALG_ASYNC | CRYPTO_ALG_NEED_FALLBACK); + + if (IS_ERR(xts_ctx->fallback)) { + pr_err("Allocating XTS fallback algorithm %s failed\n", + name); + return PTR_ERR(xts_ctx->fallback); + } + return 0; +} + +static void xts_fallback_exit(struct crypto_tfm *tfm) +{ + struct s390_xts_ctx *xts_ctx = crypto_tfm_ctx(tfm); + + crypto_free_blkcipher(xts_ctx->fallback); + xts_ctx->fallback = NULL; +} + +static struct crypto_alg xts_aes_alg = { + .cra_name = "xts(aes)", + .cra_driver_name = "xts-aes-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER | + CRYPTO_ALG_NEED_FALLBACK, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_xts_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_init = xts_fallback_init, + .cra_exit = xts_fallback_exit, + .cra_u = { + .blkcipher = { + .min_keysize = 2 * AES_MIN_KEY_SIZE, + .max_keysize = 2 * AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = xts_aes_set_key, + .encrypt = xts_aes_encrypt, + .decrypt = xts_aes_decrypt, + } + } +}; + +static int xts_aes_alg_reg; + +static int ctr_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); + + switch (key_len) { + case 16: + sctx->enc = KMCTR_AES_128_ENCRYPT; + sctx->dec = KMCTR_AES_128_DECRYPT; + break; + case 24: + sctx->enc = KMCTR_AES_192_ENCRYPT; + sctx->dec = KMCTR_AES_192_DECRYPT; + break; + case 32: + sctx->enc = KMCTR_AES_256_ENCRYPT; + sctx->dec = KMCTR_AES_256_DECRYPT; + break; + } + + return aes_set_key(tfm, in_key, key_len); +} + +static unsigned int __ctrblk_init(u8 *ctrptr, unsigned int nbytes) +{ + unsigned int i, n; + + /* only use complete blocks, max. PAGE_SIZE */ + n = (nbytes > PAGE_SIZE) ? PAGE_SIZE : nbytes & ~(AES_BLOCK_SIZE - 1); + for (i = AES_BLOCK_SIZE; i < n; i += AES_BLOCK_SIZE) { + memcpy(ctrptr + i, ctrptr + i - AES_BLOCK_SIZE, + AES_BLOCK_SIZE); + crypto_inc(ctrptr + i, AES_BLOCK_SIZE); + } + return n; +} + +static int ctr_aes_crypt(struct blkcipher_desc *desc, long func, + struct s390_aes_ctx *sctx, struct blkcipher_walk *walk) +{ + int ret = blkcipher_walk_virt_block(desc, walk, AES_BLOCK_SIZE); + unsigned int n, nbytes; + u8 buf[AES_BLOCK_SIZE], ctrbuf[AES_BLOCK_SIZE]; + u8 *out, *in, *ctrptr = ctrbuf; + + if (!walk->nbytes) + return ret; + + if (spin_trylock(&ctrblk_lock)) + ctrptr = ctrblk; + + memcpy(ctrptr, walk->iv, AES_BLOCK_SIZE); + while ((nbytes = walk->nbytes) >= AES_BLOCK_SIZE) { + out = walk->dst.virt.addr; + in = walk->src.virt.addr; + while (nbytes >= AES_BLOCK_SIZE) { + if (ctrptr == ctrblk) + n = __ctrblk_init(ctrptr, nbytes); + else + n = AES_BLOCK_SIZE; + ret = crypt_s390_kmctr(func, sctx->key, out, in, + n, ctrptr); + if (ret < 0 || ret != n) { + if (ctrptr == ctrblk) + spin_unlock(&ctrblk_lock); + return -EIO; + } + if (n > AES_BLOCK_SIZE) + memcpy(ctrptr, ctrptr + n - AES_BLOCK_SIZE, + AES_BLOCK_SIZE); + crypto_inc(ctrptr, AES_BLOCK_SIZE); + out += n; + in += n; + nbytes -= n; + } + ret = blkcipher_walk_done(desc, walk, nbytes); + } + if (ctrptr == ctrblk) { + if (nbytes) + memcpy(ctrbuf, ctrptr, AES_BLOCK_SIZE); + else + memcpy(walk->iv, ctrptr, AES_BLOCK_SIZE); + spin_unlock(&ctrblk_lock); + } else { + if (!nbytes) + memcpy(walk->iv, ctrptr, AES_BLOCK_SIZE); + } + /* + * final block may be < AES_BLOCK_SIZE, copy only nbytes + */ + if (nbytes) { + out = walk->dst.virt.addr; + in = walk->src.virt.addr; + ret = crypt_s390_kmctr(func, sctx->key, buf, in, + AES_BLOCK_SIZE, ctrbuf); + if (ret < 0 || ret != AES_BLOCK_SIZE) + return -EIO; + memcpy(out, buf, nbytes); + crypto_inc(ctrbuf, AES_BLOCK_SIZE); + ret = blkcipher_walk_done(desc, walk, 0); + memcpy(walk->iv, ctrbuf, AES_BLOCK_SIZE); + } + + return ret; +} + +static int ctr_aes_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ctr_aes_crypt(desc, sctx->enc, sctx, &walk); +} + +static int ctr_aes_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_aes_ctx *sctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ctr_aes_crypt(desc, sctx->dec, sctx, &walk); +} + +static struct crypto_alg ctr_aes_alg = { + .cra_name = "ctr(aes)", + .cra_driver_name = "ctr-aes-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct s390_aes_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = ctr_aes_set_key, + .encrypt = ctr_aes_encrypt, + .decrypt = ctr_aes_decrypt, + } + } +}; + +static int ctr_aes_alg_reg; + +static int __init aes_s390_init(void) +{ + int ret; + + if (crypt_s390_func_available(KM_AES_128_ENCRYPT, CRYPT_S390_MSA)) + keylen_flag |= AES_KEYLEN_128; + if (crypt_s390_func_available(KM_AES_192_ENCRYPT, CRYPT_S390_MSA)) + keylen_flag |= AES_KEYLEN_192; + if (crypt_s390_func_available(KM_AES_256_ENCRYPT, CRYPT_S390_MSA)) + keylen_flag |= AES_KEYLEN_256; + + if (!keylen_flag) + return -EOPNOTSUPP; + + /* z9 109 and z9 BC/EC only support 128 bit key length */ + if (keylen_flag == AES_KEYLEN_128) + pr_info("AES hardware acceleration is only available for" + " 128-bit keys\n"); + + ret = crypto_register_alg(&aes_alg); + if (ret) + goto aes_err; + + ret = crypto_register_alg(&ecb_aes_alg); + if (ret) + goto ecb_aes_err; + + ret = crypto_register_alg(&cbc_aes_alg); + if (ret) + goto cbc_aes_err; + + if (crypt_s390_func_available(KM_XTS_128_ENCRYPT, + CRYPT_S390_MSA | CRYPT_S390_MSA4) && + crypt_s390_func_available(KM_XTS_256_ENCRYPT, + CRYPT_S390_MSA | CRYPT_S390_MSA4)) { + ret = crypto_register_alg(&xts_aes_alg); + if (ret) + goto xts_aes_err; + xts_aes_alg_reg = 1; + } + + if (crypt_s390_func_available(KMCTR_AES_128_ENCRYPT, + CRYPT_S390_MSA | CRYPT_S390_MSA4) && + crypt_s390_func_available(KMCTR_AES_192_ENCRYPT, + CRYPT_S390_MSA | CRYPT_S390_MSA4) && + crypt_s390_func_available(KMCTR_AES_256_ENCRYPT, + CRYPT_S390_MSA | CRYPT_S390_MSA4)) { + ctrblk = (u8 *) __get_free_page(GFP_KERNEL); + if (!ctrblk) { + ret = -ENOMEM; + goto ctr_aes_err; + } + ret = crypto_register_alg(&ctr_aes_alg); + if (ret) { + free_page((unsigned long) ctrblk); + goto ctr_aes_err; + } + ctr_aes_alg_reg = 1; + } + +out: + return ret; + +ctr_aes_err: + crypto_unregister_alg(&xts_aes_alg); +xts_aes_err: + crypto_unregister_alg(&cbc_aes_alg); +cbc_aes_err: + crypto_unregister_alg(&ecb_aes_alg); +ecb_aes_err: + crypto_unregister_alg(&aes_alg); +aes_err: + goto out; +} + +static void __exit aes_s390_fini(void) +{ + if (ctr_aes_alg_reg) { + crypto_unregister_alg(&ctr_aes_alg); + free_page((unsigned long) ctrblk); + } + if (xts_aes_alg_reg) + crypto_unregister_alg(&xts_aes_alg); + crypto_unregister_alg(&cbc_aes_alg); + crypto_unregister_alg(&ecb_aes_alg); + crypto_unregister_alg(&aes_alg); +} + +module_init(aes_s390_init); +module_exit(aes_s390_fini); + +MODULE_ALIAS_CRYPTO("aes-all"); + +MODULE_DESCRIPTION("Rijndael (AES) Cipher Algorithm"); +MODULE_LICENSE("GPL"); diff --git a/arch/s390/crypto/crypt_s390.h b/arch/s390/crypto/crypt_s390.h new file mode 100644 index 000000000..d9c4c313f --- /dev/null +++ b/arch/s390/crypto/crypt_s390.h @@ -0,0 +1,493 @@ +/* + * Cryptographic API. + * + * Support for s390 cryptographic instructions. + * + * Copyright IBM Corp. 2003, 2015 + * Author(s): Thomas Spatzier + * Jan Glauber (jan.glauber@de.ibm.com) + * Harald Freudenberger (freude@de.ibm.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. + * + */ +#ifndef _CRYPTO_ARCH_S390_CRYPT_S390_H +#define _CRYPTO_ARCH_S390_CRYPT_S390_H + +#include <asm/errno.h> +#include <asm/facility.h> + +#define CRYPT_S390_OP_MASK 0xFF00 +#define CRYPT_S390_FUNC_MASK 0x00FF + +#define CRYPT_S390_PRIORITY 300 +#define CRYPT_S390_COMPOSITE_PRIORITY 400 + +#define CRYPT_S390_MSA 0x1 +#define CRYPT_S390_MSA3 0x2 +#define CRYPT_S390_MSA4 0x4 +#define CRYPT_S390_MSA5 0x8 + +/* s390 cryptographic operations */ +enum crypt_s390_operations { + CRYPT_S390_KM = 0x0100, + CRYPT_S390_KMC = 0x0200, + CRYPT_S390_KIMD = 0x0300, + CRYPT_S390_KLMD = 0x0400, + CRYPT_S390_KMAC = 0x0500, + CRYPT_S390_KMCTR = 0x0600, + CRYPT_S390_PPNO = 0x0700 +}; + +/* + * function codes for KM (CIPHER MESSAGE) instruction + * 0x80 is the decipher modifier bit + */ +enum crypt_s390_km_func { + KM_QUERY = CRYPT_S390_KM | 0x0, + KM_DEA_ENCRYPT = CRYPT_S390_KM | 0x1, + KM_DEA_DECRYPT = CRYPT_S390_KM | 0x1 | 0x80, + KM_TDEA_128_ENCRYPT = CRYPT_S390_KM | 0x2, + KM_TDEA_128_DECRYPT = CRYPT_S390_KM | 0x2 | 0x80, + KM_TDEA_192_ENCRYPT = CRYPT_S390_KM | 0x3, + KM_TDEA_192_DECRYPT = CRYPT_S390_KM | 0x3 | 0x80, + KM_AES_128_ENCRYPT = CRYPT_S390_KM | 0x12, + KM_AES_128_DECRYPT = CRYPT_S390_KM | 0x12 | 0x80, + KM_AES_192_ENCRYPT = CRYPT_S390_KM | 0x13, + KM_AES_192_DECRYPT = CRYPT_S390_KM | 0x13 | 0x80, + KM_AES_256_ENCRYPT = CRYPT_S390_KM | 0x14, + KM_AES_256_DECRYPT = CRYPT_S390_KM | 0x14 | 0x80, + KM_XTS_128_ENCRYPT = CRYPT_S390_KM | 0x32, + KM_XTS_128_DECRYPT = CRYPT_S390_KM | 0x32 | 0x80, + KM_XTS_256_ENCRYPT = CRYPT_S390_KM | 0x34, + KM_XTS_256_DECRYPT = CRYPT_S390_KM | 0x34 | 0x80, +}; + +/* + * function codes for KMC (CIPHER MESSAGE WITH CHAINING) + * instruction + */ +enum crypt_s390_kmc_func { + KMC_QUERY = CRYPT_S390_KMC | 0x0, + KMC_DEA_ENCRYPT = CRYPT_S390_KMC | 0x1, + KMC_DEA_DECRYPT = CRYPT_S390_KMC | 0x1 | 0x80, + KMC_TDEA_128_ENCRYPT = CRYPT_S390_KMC | 0x2, + KMC_TDEA_128_DECRYPT = CRYPT_S390_KMC | 0x2 | 0x80, + KMC_TDEA_192_ENCRYPT = CRYPT_S390_KMC | 0x3, + KMC_TDEA_192_DECRYPT = CRYPT_S390_KMC | 0x3 | 0x80, + KMC_AES_128_ENCRYPT = CRYPT_S390_KMC | 0x12, + KMC_AES_128_DECRYPT = CRYPT_S390_KMC | 0x12 | 0x80, + KMC_AES_192_ENCRYPT = CRYPT_S390_KMC | 0x13, + KMC_AES_192_DECRYPT = CRYPT_S390_KMC | 0x13 | 0x80, + KMC_AES_256_ENCRYPT = CRYPT_S390_KMC | 0x14, + KMC_AES_256_DECRYPT = CRYPT_S390_KMC | 0x14 | 0x80, + KMC_PRNG = CRYPT_S390_KMC | 0x43, +}; + +/* + * function codes for KMCTR (CIPHER MESSAGE WITH COUNTER) + * instruction + */ +enum crypt_s390_kmctr_func { + KMCTR_QUERY = CRYPT_S390_KMCTR | 0x0, + KMCTR_DEA_ENCRYPT = CRYPT_S390_KMCTR | 0x1, + KMCTR_DEA_DECRYPT = CRYPT_S390_KMCTR | 0x1 | 0x80, + KMCTR_TDEA_128_ENCRYPT = CRYPT_S390_KMCTR | 0x2, + KMCTR_TDEA_128_DECRYPT = CRYPT_S390_KMCTR | 0x2 | 0x80, + KMCTR_TDEA_192_ENCRYPT = CRYPT_S390_KMCTR | 0x3, + KMCTR_TDEA_192_DECRYPT = CRYPT_S390_KMCTR | 0x3 | 0x80, + KMCTR_AES_128_ENCRYPT = CRYPT_S390_KMCTR | 0x12, + KMCTR_AES_128_DECRYPT = CRYPT_S390_KMCTR | 0x12 | 0x80, + KMCTR_AES_192_ENCRYPT = CRYPT_S390_KMCTR | 0x13, + KMCTR_AES_192_DECRYPT = CRYPT_S390_KMCTR | 0x13 | 0x80, + KMCTR_AES_256_ENCRYPT = CRYPT_S390_KMCTR | 0x14, + KMCTR_AES_256_DECRYPT = CRYPT_S390_KMCTR | 0x14 | 0x80, +}; + +/* + * function codes for KIMD (COMPUTE INTERMEDIATE MESSAGE DIGEST) + * instruction + */ +enum crypt_s390_kimd_func { + KIMD_QUERY = CRYPT_S390_KIMD | 0, + KIMD_SHA_1 = CRYPT_S390_KIMD | 1, + KIMD_SHA_256 = CRYPT_S390_KIMD | 2, + KIMD_SHA_512 = CRYPT_S390_KIMD | 3, + KIMD_GHASH = CRYPT_S390_KIMD | 65, +}; + +/* + * function codes for KLMD (COMPUTE LAST MESSAGE DIGEST) + * instruction + */ +enum crypt_s390_klmd_func { + KLMD_QUERY = CRYPT_S390_KLMD | 0, + KLMD_SHA_1 = CRYPT_S390_KLMD | 1, + KLMD_SHA_256 = CRYPT_S390_KLMD | 2, + KLMD_SHA_512 = CRYPT_S390_KLMD | 3, +}; + +/* + * function codes for KMAC (COMPUTE MESSAGE AUTHENTICATION CODE) + * instruction + */ +enum crypt_s390_kmac_func { + KMAC_QUERY = CRYPT_S390_KMAC | 0, + KMAC_DEA = CRYPT_S390_KMAC | 1, + KMAC_TDEA_128 = CRYPT_S390_KMAC | 2, + KMAC_TDEA_192 = CRYPT_S390_KMAC | 3 +}; + +/* + * function codes for PPNO (PERFORM PSEUDORANDOM NUMBER + * OPERATION) instruction + */ +enum crypt_s390_ppno_func { + PPNO_QUERY = CRYPT_S390_PPNO | 0, + PPNO_SHA512_DRNG_GEN = CRYPT_S390_PPNO | 3, + PPNO_SHA512_DRNG_SEED = CRYPT_S390_PPNO | 0x83 +}; + +/** + * crypt_s390_km: + * @func: the function code passed to KM; see crypt_s390_km_func + * @param: address of parameter block; see POP for details on each func + * @dest: address of destination memory area + * @src: address of source memory area + * @src_len: length of src operand in bytes + * + * Executes the KM (CIPHER MESSAGE) operation of the CPU. + * + * Returns -1 for failure, 0 for the query func, number of processed + * bytes for encryption/decryption funcs + */ +static inline int crypt_s390_km(long func, void *param, + u8 *dest, const u8 *src, long src_len) +{ + register long __func asm("0") = func & CRYPT_S390_FUNC_MASK; + register void *__param asm("1") = param; + register const u8 *__src asm("2") = src; + register long __src_len asm("3") = src_len; + register u8 *__dest asm("4") = dest; + int ret; + + asm volatile( + "0: .insn rre,0xb92e0000,%3,%1\n" /* KM opcode */ + "1: brc 1,0b\n" /* handle partial completion */ + " la %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "=d" (ret), "+a" (__src), "+d" (__src_len), "+a" (__dest) + : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); + if (ret < 0) + return ret; + return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len; +} + +/** + * crypt_s390_kmc: + * @func: the function code passed to KM; see crypt_s390_kmc_func + * @param: address of parameter block; see POP for details on each func + * @dest: address of destination memory area + * @src: address of source memory area + * @src_len: length of src operand in bytes + * + * Executes the KMC (CIPHER MESSAGE WITH CHAINING) operation of the CPU. + * + * Returns -1 for failure, 0 for the query func, number of processed + * bytes for encryption/decryption funcs + */ +static inline int crypt_s390_kmc(long func, void *param, + u8 *dest, const u8 *src, long src_len) +{ + register long __func asm("0") = func & CRYPT_S390_FUNC_MASK; + register void *__param asm("1") = param; + register const u8 *__src asm("2") = src; + register long __src_len asm("3") = src_len; + register u8 *__dest asm("4") = dest; + int ret; + + asm volatile( + "0: .insn rre,0xb92f0000,%3,%1\n" /* KMC opcode */ + "1: brc 1,0b\n" /* handle partial completion */ + " la %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "=d" (ret), "+a" (__src), "+d" (__src_len), "+a" (__dest) + : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); + if (ret < 0) + return ret; + return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len; +} + +/** + * crypt_s390_kimd: + * @func: the function code passed to KM; see crypt_s390_kimd_func + * @param: address of parameter block; see POP for details on each func + * @src: address of source memory area + * @src_len: length of src operand in bytes + * + * Executes the KIMD (COMPUTE INTERMEDIATE MESSAGE DIGEST) operation + * of the CPU. + * + * Returns -1 for failure, 0 for the query func, number of processed + * bytes for digest funcs + */ +static inline int crypt_s390_kimd(long func, void *param, + const u8 *src, long src_len) +{ + register long __func asm("0") = func & CRYPT_S390_FUNC_MASK; + register void *__param asm("1") = param; + register const u8 *__src asm("2") = src; + register long __src_len asm("3") = src_len; + int ret; + + asm volatile( + "0: .insn rre,0xb93e0000,%1,%1\n" /* KIMD opcode */ + "1: brc 1,0b\n" /* handle partial completion */ + " la %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "=d" (ret), "+a" (__src), "+d" (__src_len) + : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); + if (ret < 0) + return ret; + return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len; +} + +/** + * crypt_s390_klmd: + * @func: the function code passed to KM; see crypt_s390_klmd_func + * @param: address of parameter block; see POP for details on each func + * @src: address of source memory area + * @src_len: length of src operand in bytes + * + * Executes the KLMD (COMPUTE LAST MESSAGE DIGEST) operation of the CPU. + * + * Returns -1 for failure, 0 for the query func, number of processed + * bytes for digest funcs + */ +static inline int crypt_s390_klmd(long func, void *param, + const u8 *src, long src_len) +{ + register long __func asm("0") = func & CRYPT_S390_FUNC_MASK; + register void *__param asm("1") = param; + register const u8 *__src asm("2") = src; + register long __src_len asm("3") = src_len; + int ret; + + asm volatile( + "0: .insn rre,0xb93f0000,%1,%1\n" /* KLMD opcode */ + "1: brc 1,0b\n" /* handle partial completion */ + " la %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "=d" (ret), "+a" (__src), "+d" (__src_len) + : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); + if (ret < 0) + return ret; + return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len; +} + +/** + * crypt_s390_kmac: + * @func: the function code passed to KM; see crypt_s390_klmd_func + * @param: address of parameter block; see POP for details on each func + * @src: address of source memory area + * @src_len: length of src operand in bytes + * + * Executes the KMAC (COMPUTE MESSAGE AUTHENTICATION CODE) operation + * of the CPU. + * + * Returns -1 for failure, 0 for the query func, number of processed + * bytes for digest funcs + */ +static inline int crypt_s390_kmac(long func, void *param, + const u8 *src, long src_len) +{ + register long __func asm("0") = func & CRYPT_S390_FUNC_MASK; + register void *__param asm("1") = param; + register const u8 *__src asm("2") = src; + register long __src_len asm("3") = src_len; + int ret; + + asm volatile( + "0: .insn rre,0xb91e0000,%1,%1\n" /* KLAC opcode */ + "1: brc 1,0b\n" /* handle partial completion */ + " la %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "=d" (ret), "+a" (__src), "+d" (__src_len) + : "d" (__func), "a" (__param), "0" (-1) : "cc", "memory"); + if (ret < 0) + return ret; + return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len; +} + +/** + * crypt_s390_kmctr: + * @func: the function code passed to KMCTR; see crypt_s390_kmctr_func + * @param: address of parameter block; see POP for details on each func + * @dest: address of destination memory area + * @src: address of source memory area + * @src_len: length of src operand in bytes + * @counter: address of counter value + * + * Executes the KMCTR (CIPHER MESSAGE WITH COUNTER) operation of the CPU. + * + * Returns -1 for failure, 0 for the query func, number of processed + * bytes for encryption/decryption funcs + */ +static inline int crypt_s390_kmctr(long func, void *param, u8 *dest, + const u8 *src, long src_len, u8 *counter) +{ + register long __func asm("0") = func & CRYPT_S390_FUNC_MASK; + register void *__param asm("1") = param; + register const u8 *__src asm("2") = src; + register long __src_len asm("3") = src_len; + register u8 *__dest asm("4") = dest; + register u8 *__ctr asm("6") = counter; + int ret = -1; + + asm volatile( + "0: .insn rrf,0xb92d0000,%3,%1,%4,0\n" /* KMCTR opcode */ + "1: brc 1,0b\n" /* handle partial completion */ + " la %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "+d" (ret), "+a" (__src), "+d" (__src_len), "+a" (__dest), + "+a" (__ctr) + : "d" (__func), "a" (__param) : "cc", "memory"); + if (ret < 0) + return ret; + return (func & CRYPT_S390_FUNC_MASK) ? src_len - __src_len : __src_len; +} + +/** + * crypt_s390_ppno: + * @func: the function code passed to PPNO; see crypt_s390_ppno_func + * @param: address of parameter block; see POP for details on each func + * @dest: address of destination memory area + * @dest_len: size of destination memory area in bytes + * @seed: address of seed data + * @seed_len: size of seed data in bytes + * + * Executes the PPNO (PERFORM PSEUDORANDOM NUMBER OPERATION) + * operation of the CPU. + * + * Returns -1 for failure, 0 for the query func, number of random + * bytes stored in dest buffer for generate function + */ +static inline int crypt_s390_ppno(long func, void *param, + u8 *dest, long dest_len, + const u8 *seed, long seed_len) +{ + register long __func asm("0") = func & CRYPT_S390_FUNC_MASK; + register void *__param asm("1") = param; /* param block (240 bytes) */ + register u8 *__dest asm("2") = dest; /* buf for recv random bytes */ + register long __dest_len asm("3") = dest_len; /* requested random bytes */ + register const u8 *__seed asm("4") = seed; /* buf with seed data */ + register long __seed_len asm("5") = seed_len; /* bytes in seed buf */ + int ret = -1; + + asm volatile ( + "0: .insn rre,0xb93c0000,%1,%5\n" /* PPNO opcode */ + "1: brc 1,0b\n" /* handle partial completion */ + " la %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "+d" (ret), "+a"(__dest), "+d"(__dest_len) + : "d"(__func), "a"(__param), "a"(__seed), "d"(__seed_len) + : "cc", "memory"); + if (ret < 0) + return ret; + return (func & CRYPT_S390_FUNC_MASK) ? dest_len - __dest_len : 0; +} + +/** + * crypt_s390_func_available: + * @func: the function code of the specific function; 0 if op in general + * + * Tests if a specific crypto function is implemented on the machine. + * + * Returns 1 if func available; 0 if func or op in general not available + */ +static inline int crypt_s390_func_available(int func, + unsigned int facility_mask) +{ + unsigned char status[16]; + int ret; + + if (facility_mask & CRYPT_S390_MSA && !test_facility(17)) + return 0; + if (facility_mask & CRYPT_S390_MSA3 && !test_facility(76)) + return 0; + if (facility_mask & CRYPT_S390_MSA4 && !test_facility(77)) + return 0; + if (facility_mask & CRYPT_S390_MSA5 && !test_facility(57)) + return 0; + + switch (func & CRYPT_S390_OP_MASK) { + case CRYPT_S390_KM: + ret = crypt_s390_km(KM_QUERY, &status, NULL, NULL, 0); + break; + case CRYPT_S390_KMC: + ret = crypt_s390_kmc(KMC_QUERY, &status, NULL, NULL, 0); + break; + case CRYPT_S390_KIMD: + ret = crypt_s390_kimd(KIMD_QUERY, &status, NULL, 0); + break; + case CRYPT_S390_KLMD: + ret = crypt_s390_klmd(KLMD_QUERY, &status, NULL, 0); + break; + case CRYPT_S390_KMAC: + ret = crypt_s390_kmac(KMAC_QUERY, &status, NULL, 0); + break; + case CRYPT_S390_KMCTR: + ret = crypt_s390_kmctr(KMCTR_QUERY, &status, + NULL, NULL, 0, NULL); + break; + case CRYPT_S390_PPNO: + ret = crypt_s390_ppno(PPNO_QUERY, &status, + NULL, 0, NULL, 0); + break; + default: + return 0; + } + if (ret < 0) + return 0; + func &= CRYPT_S390_FUNC_MASK; + func &= 0x7f; /* mask modifier bit */ + return (status[func >> 3] & (0x80 >> (func & 7))) != 0; +} + +/** + * crypt_s390_pcc: + * @func: the function code passed to KM; see crypt_s390_km_func + * @param: address of parameter block; see POP for details on each func + * + * Executes the PCC (PERFORM CRYPTOGRAPHIC COMPUTATION) operation of the CPU. + * + * Returns -1 for failure, 0 for success. + */ +static inline int crypt_s390_pcc(long func, void *param) +{ + register long __func asm("0") = func & 0x7f; /* encrypt or decrypt */ + register void *__param asm("1") = param; + int ret = -1; + + asm volatile( + "0: .insn rre,0xb92c0000,0,0\n" /* PCC opcode */ + "1: brc 1,0b\n" /* handle partial completion */ + " la %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "+d" (ret) + : "d" (__func), "a" (__param) : "cc", "memory"); + return ret; +} + +#endif /* _CRYPTO_ARCH_S390_CRYPT_S390_H */ diff --git a/arch/s390/crypto/des_s390.c b/arch/s390/crypto/des_s390.c new file mode 100644 index 000000000..9e05cc453 --- /dev/null +++ b/arch/s390/crypto/des_s390.c @@ -0,0 +1,626 @@ +/* + * Cryptographic API. + * + * s390 implementation of the DES Cipher Algorithm. + * + * Copyright IBM Corp. 2003, 2011 + * Author(s): Thomas Spatzier + * Jan Glauber (jan.glauber@de.ibm.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. + * + */ + +#include <linux/init.h> +#include <linux/module.h> +#include <linux/crypto.h> +#include <crypto/algapi.h> +#include <crypto/des.h> + +#include "crypt_s390.h" + +#define DES3_KEY_SIZE (3 * DES_KEY_SIZE) + +static u8 *ctrblk; +static DEFINE_SPINLOCK(ctrblk_lock); + +struct s390_des_ctx { + u8 iv[DES_BLOCK_SIZE]; + u8 key[DES3_KEY_SIZE]; +}; + +static int des_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int key_len) +{ + struct s390_des_ctx *ctx = crypto_tfm_ctx(tfm); + u32 *flags = &tfm->crt_flags; + u32 tmp[DES_EXPKEY_WORDS]; + + /* check for weak keys */ + if (!des_ekey(tmp, key) && (*flags & CRYPTO_TFM_REQ_WEAK_KEY)) { + *flags |= CRYPTO_TFM_RES_WEAK_KEY; + return -EINVAL; + } + + memcpy(ctx->key, key, key_len); + return 0; +} + +static void des_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) +{ + struct s390_des_ctx *ctx = crypto_tfm_ctx(tfm); + + crypt_s390_km(KM_DEA_ENCRYPT, ctx->key, out, in, DES_BLOCK_SIZE); +} + +static void des_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) +{ + struct s390_des_ctx *ctx = crypto_tfm_ctx(tfm); + + crypt_s390_km(KM_DEA_DECRYPT, ctx->key, out, in, DES_BLOCK_SIZE); +} + +static struct crypto_alg des_alg = { + .cra_name = "des", + .cra_driver_name = "des-s390", + .cra_priority = CRYPT_S390_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_CIPHER, + .cra_blocksize = DES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_des_ctx), + .cra_module = THIS_MODULE, + .cra_u = { + .cipher = { + .cia_min_keysize = DES_KEY_SIZE, + .cia_max_keysize = DES_KEY_SIZE, + .cia_setkey = des_setkey, + .cia_encrypt = des_encrypt, + .cia_decrypt = des_decrypt, + } + } +}; + +static int ecb_desall_crypt(struct blkcipher_desc *desc, long func, + u8 *key, struct blkcipher_walk *walk) +{ + int ret = blkcipher_walk_virt(desc, walk); + unsigned int nbytes; + + while ((nbytes = walk->nbytes)) { + /* only use complete blocks */ + unsigned int n = nbytes & ~(DES_BLOCK_SIZE - 1); + u8 *out = walk->dst.virt.addr; + u8 *in = walk->src.virt.addr; + + ret = crypt_s390_km(func, key, out, in, n); + if (ret < 0 || ret != n) + return -EIO; + + nbytes &= DES_BLOCK_SIZE - 1; + ret = blkcipher_walk_done(desc, walk, nbytes); + } + + return ret; +} + +static int cbc_desall_crypt(struct blkcipher_desc *desc, long func, + struct blkcipher_walk *walk) +{ + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + int ret = blkcipher_walk_virt(desc, walk); + unsigned int nbytes = walk->nbytes; + struct { + u8 iv[DES_BLOCK_SIZE]; + u8 key[DES3_KEY_SIZE]; + } param; + + if (!nbytes) + goto out; + + memcpy(param.iv, walk->iv, DES_BLOCK_SIZE); + memcpy(param.key, ctx->key, DES3_KEY_SIZE); + do { + /* only use complete blocks */ + unsigned int n = nbytes & ~(DES_BLOCK_SIZE - 1); + u8 *out = walk->dst.virt.addr; + u8 *in = walk->src.virt.addr; + + ret = crypt_s390_kmc(func, ¶m, out, in, n); + if (ret < 0 || ret != n) + return -EIO; + + nbytes &= DES_BLOCK_SIZE - 1; + ret = blkcipher_walk_done(desc, walk, nbytes); + } while ((nbytes = walk->nbytes)); + memcpy(walk->iv, param.iv, DES_BLOCK_SIZE); + +out: + return ret; +} + +static int ecb_des_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_desall_crypt(desc, KM_DEA_ENCRYPT, ctx->key, &walk); +} + +static int ecb_des_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_desall_crypt(desc, KM_DEA_DECRYPT, ctx->key, &walk); +} + +static struct crypto_alg ecb_des_alg = { + .cra_name = "ecb(des)", + .cra_driver_name = "ecb-des-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = DES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_des_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES_KEY_SIZE, + .max_keysize = DES_KEY_SIZE, + .setkey = des_setkey, + .encrypt = ecb_des_encrypt, + .decrypt = ecb_des_decrypt, + } + } +}; + +static int cbc_des_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return cbc_desall_crypt(desc, KMC_DEA_ENCRYPT, &walk); +} + +static int cbc_des_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return cbc_desall_crypt(desc, KMC_DEA_DECRYPT, &walk); +} + +static struct crypto_alg cbc_des_alg = { + .cra_name = "cbc(des)", + .cra_driver_name = "cbc-des-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = DES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_des_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES_KEY_SIZE, + .max_keysize = DES_KEY_SIZE, + .ivsize = DES_BLOCK_SIZE, + .setkey = des_setkey, + .encrypt = cbc_des_encrypt, + .decrypt = cbc_des_decrypt, + } + } +}; + +/* + * RFC2451: + * + * For DES-EDE3, there is no known need to reject weak or + * complementation keys. Any weakness is obviated by the use of + * multiple keys. + * + * However, if the first two or last two independent 64-bit keys are + * equal (k1 == k2 or k2 == k3), then the DES3 operation is simply the + * same as DES. Implementers MUST reject keys that exhibit this + * property. + * + */ +static int des3_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int key_len) +{ + struct s390_des_ctx *ctx = crypto_tfm_ctx(tfm); + u32 *flags = &tfm->crt_flags; + + if (!(crypto_memneq(key, &key[DES_KEY_SIZE], DES_KEY_SIZE) && + crypto_memneq(&key[DES_KEY_SIZE], &key[DES_KEY_SIZE * 2], + DES_KEY_SIZE)) && + (*flags & CRYPTO_TFM_REQ_WEAK_KEY)) { + *flags |= CRYPTO_TFM_RES_WEAK_KEY; + return -EINVAL; + } + memcpy(ctx->key, key, key_len); + return 0; +} + +static void des3_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct s390_des_ctx *ctx = crypto_tfm_ctx(tfm); + + crypt_s390_km(KM_TDEA_192_ENCRYPT, ctx->key, dst, src, DES_BLOCK_SIZE); +} + +static void des3_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct s390_des_ctx *ctx = crypto_tfm_ctx(tfm); + + crypt_s390_km(KM_TDEA_192_DECRYPT, ctx->key, dst, src, DES_BLOCK_SIZE); +} + +static struct crypto_alg des3_alg = { + .cra_name = "des3_ede", + .cra_driver_name = "des3_ede-s390", + .cra_priority = CRYPT_S390_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_CIPHER, + .cra_blocksize = DES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_des_ctx), + .cra_module = THIS_MODULE, + .cra_u = { + .cipher = { + .cia_min_keysize = DES3_KEY_SIZE, + .cia_max_keysize = DES3_KEY_SIZE, + .cia_setkey = des3_setkey, + .cia_encrypt = des3_encrypt, + .cia_decrypt = des3_decrypt, + } + } +}; + +static int ecb_des3_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_desall_crypt(desc, KM_TDEA_192_ENCRYPT, ctx->key, &walk); +} + +static int ecb_des3_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ecb_desall_crypt(desc, KM_TDEA_192_DECRYPT, ctx->key, &walk); +} + +static struct crypto_alg ecb_des3_alg = { + .cra_name = "ecb(des3_ede)", + .cra_driver_name = "ecb-des3_ede-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = DES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_des_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES3_KEY_SIZE, + .max_keysize = DES3_KEY_SIZE, + .setkey = des3_setkey, + .encrypt = ecb_des3_encrypt, + .decrypt = ecb_des3_decrypt, + } + } +}; + +static int cbc_des3_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return cbc_desall_crypt(desc, KMC_TDEA_192_ENCRYPT, &walk); +} + +static int cbc_des3_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return cbc_desall_crypt(desc, KMC_TDEA_192_DECRYPT, &walk); +} + +static struct crypto_alg cbc_des3_alg = { + .cra_name = "cbc(des3_ede)", + .cra_driver_name = "cbc-des3_ede-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = DES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_des_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES3_KEY_SIZE, + .max_keysize = DES3_KEY_SIZE, + .ivsize = DES_BLOCK_SIZE, + .setkey = des3_setkey, + .encrypt = cbc_des3_encrypt, + .decrypt = cbc_des3_decrypt, + } + } +}; + +static unsigned int __ctrblk_init(u8 *ctrptr, unsigned int nbytes) +{ + unsigned int i, n; + + /* align to block size, max. PAGE_SIZE */ + n = (nbytes > PAGE_SIZE) ? PAGE_SIZE : nbytes & ~(DES_BLOCK_SIZE - 1); + for (i = DES_BLOCK_SIZE; i < n; i += DES_BLOCK_SIZE) { + memcpy(ctrptr + i, ctrptr + i - DES_BLOCK_SIZE, DES_BLOCK_SIZE); + crypto_inc(ctrptr + i, DES_BLOCK_SIZE); + } + return n; +} + +static int ctr_desall_crypt(struct blkcipher_desc *desc, long func, + struct s390_des_ctx *ctx, + struct blkcipher_walk *walk) +{ + int ret = blkcipher_walk_virt_block(desc, walk, DES_BLOCK_SIZE); + unsigned int n, nbytes; + u8 buf[DES_BLOCK_SIZE], ctrbuf[DES_BLOCK_SIZE]; + u8 *out, *in, *ctrptr = ctrbuf; + + if (!walk->nbytes) + return ret; + + if (spin_trylock(&ctrblk_lock)) + ctrptr = ctrblk; + + memcpy(ctrptr, walk->iv, DES_BLOCK_SIZE); + while ((nbytes = walk->nbytes) >= DES_BLOCK_SIZE) { + out = walk->dst.virt.addr; + in = walk->src.virt.addr; + while (nbytes >= DES_BLOCK_SIZE) { + if (ctrptr == ctrblk) + n = __ctrblk_init(ctrptr, nbytes); + else + n = DES_BLOCK_SIZE; + ret = crypt_s390_kmctr(func, ctx->key, out, in, + n, ctrptr); + if (ret < 0 || ret != n) { + if (ctrptr == ctrblk) + spin_unlock(&ctrblk_lock); + return -EIO; + } + if (n > DES_BLOCK_SIZE) + memcpy(ctrptr, ctrptr + n - DES_BLOCK_SIZE, + DES_BLOCK_SIZE); + crypto_inc(ctrptr, DES_BLOCK_SIZE); + out += n; + in += n; + nbytes -= n; + } + ret = blkcipher_walk_done(desc, walk, nbytes); + } + if (ctrptr == ctrblk) { + if (nbytes) + memcpy(ctrbuf, ctrptr, DES_BLOCK_SIZE); + else + memcpy(walk->iv, ctrptr, DES_BLOCK_SIZE); + spin_unlock(&ctrblk_lock); + } else { + if (!nbytes) + memcpy(walk->iv, ctrptr, DES_BLOCK_SIZE); + } + /* final block may be < DES_BLOCK_SIZE, copy only nbytes */ + if (nbytes) { + out = walk->dst.virt.addr; + in = walk->src.virt.addr; + ret = crypt_s390_kmctr(func, ctx->key, buf, in, + DES_BLOCK_SIZE, ctrbuf); + if (ret < 0 || ret != DES_BLOCK_SIZE) + return -EIO; + memcpy(out, buf, nbytes); + crypto_inc(ctrbuf, DES_BLOCK_SIZE); + ret = blkcipher_walk_done(desc, walk, 0); + memcpy(walk->iv, ctrbuf, DES_BLOCK_SIZE); + } + return ret; +} + +static int ctr_des_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ctr_desall_crypt(desc, KMCTR_DEA_ENCRYPT, ctx, &walk); +} + +static int ctr_des_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ctr_desall_crypt(desc, KMCTR_DEA_DECRYPT, ctx, &walk); +} + +static struct crypto_alg ctr_des_alg = { + .cra_name = "ctr(des)", + .cra_driver_name = "ctr-des-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct s390_des_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES_KEY_SIZE, + .max_keysize = DES_KEY_SIZE, + .ivsize = DES_BLOCK_SIZE, + .setkey = des_setkey, + .encrypt = ctr_des_encrypt, + .decrypt = ctr_des_decrypt, + } + } +}; + +static int ctr_des3_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ctr_desall_crypt(desc, KMCTR_TDEA_192_ENCRYPT, ctx, &walk); +} + +static int ctr_des3_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + + blkcipher_walk_init(&walk, dst, src, nbytes); + return ctr_desall_crypt(desc, KMCTR_TDEA_192_DECRYPT, ctx, &walk); +} + +static struct crypto_alg ctr_des3_alg = { + .cra_name = "ctr(des3_ede)", + .cra_driver_name = "ctr-des3_ede-s390", + .cra_priority = CRYPT_S390_COMPOSITE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct s390_des_ctx), + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES3_KEY_SIZE, + .max_keysize = DES3_KEY_SIZE, + .ivsize = DES_BLOCK_SIZE, + .setkey = des3_setkey, + .encrypt = ctr_des3_encrypt, + .decrypt = ctr_des3_decrypt, + } + } +}; + +static int __init des_s390_init(void) +{ + int ret; + + if (!crypt_s390_func_available(KM_DEA_ENCRYPT, CRYPT_S390_MSA) || + !crypt_s390_func_available(KM_TDEA_192_ENCRYPT, CRYPT_S390_MSA)) + return -EOPNOTSUPP; + + ret = crypto_register_alg(&des_alg); + if (ret) + goto des_err; + ret = crypto_register_alg(&ecb_des_alg); + if (ret) + goto ecb_des_err; + ret = crypto_register_alg(&cbc_des_alg); + if (ret) + goto cbc_des_err; + ret = crypto_register_alg(&des3_alg); + if (ret) + goto des3_err; + ret = crypto_register_alg(&ecb_des3_alg); + if (ret) + goto ecb_des3_err; + ret = crypto_register_alg(&cbc_des3_alg); + if (ret) + goto cbc_des3_err; + + if (crypt_s390_func_available(KMCTR_DEA_ENCRYPT, + CRYPT_S390_MSA | CRYPT_S390_MSA4) && + crypt_s390_func_available(KMCTR_TDEA_192_ENCRYPT, + CRYPT_S390_MSA | CRYPT_S390_MSA4)) { + ret = crypto_register_alg(&ctr_des_alg); + if (ret) + goto ctr_des_err; + ret = crypto_register_alg(&ctr_des3_alg); + if (ret) + goto ctr_des3_err; + ctrblk = (u8 *) __get_free_page(GFP_KERNEL); + if (!ctrblk) { + ret = -ENOMEM; + goto ctr_mem_err; + } + } +out: + return ret; + +ctr_mem_err: + crypto_unregister_alg(&ctr_des3_alg); +ctr_des3_err: + crypto_unregister_alg(&ctr_des_alg); +ctr_des_err: + crypto_unregister_alg(&cbc_des3_alg); +cbc_des3_err: + crypto_unregister_alg(&ecb_des3_alg); +ecb_des3_err: + crypto_unregister_alg(&des3_alg); +des3_err: + crypto_unregister_alg(&cbc_des_alg); +cbc_des_err: + crypto_unregister_alg(&ecb_des_alg); +ecb_des_err: + crypto_unregister_alg(&des_alg); +des_err: + goto out; +} + +static void __exit des_s390_exit(void) +{ + if (ctrblk) { + crypto_unregister_alg(&ctr_des_alg); + crypto_unregister_alg(&ctr_des3_alg); + free_page((unsigned long) ctrblk); + } + crypto_unregister_alg(&cbc_des3_alg); + crypto_unregister_alg(&ecb_des3_alg); + crypto_unregister_alg(&des3_alg); + crypto_unregister_alg(&cbc_des_alg); + crypto_unregister_alg(&ecb_des_alg); + crypto_unregister_alg(&des_alg); +} + +module_init(des_s390_init); +module_exit(des_s390_exit); + +MODULE_ALIAS_CRYPTO("des"); +MODULE_ALIAS_CRYPTO("des3_ede"); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("DES & Triple DES EDE Cipher Algorithms"); diff --git a/arch/s390/crypto/ghash_s390.c b/arch/s390/crypto/ghash_s390.c new file mode 100644 index 000000000..b258110da --- /dev/null +++ b/arch/s390/crypto/ghash_s390.c @@ -0,0 +1,167 @@ +/* + * Cryptographic API. + * + * s390 implementation of the GHASH algorithm for GCM (Galois/Counter Mode). + * + * Copyright IBM Corp. 2011 + * Author(s): Gerald Schaefer <gerald.schaefer@de.ibm.com> + */ + +#include <crypto/internal/hash.h> +#include <linux/module.h> + +#include "crypt_s390.h" + +#define GHASH_BLOCK_SIZE 16 +#define GHASH_DIGEST_SIZE 16 + +struct ghash_ctx { + u8 key[GHASH_BLOCK_SIZE]; +}; + +struct ghash_desc_ctx { + u8 icv[GHASH_BLOCK_SIZE]; + u8 key[GHASH_BLOCK_SIZE]; + u8 buffer[GHASH_BLOCK_SIZE]; + u32 bytes; +}; + +static int ghash_init(struct shash_desc *desc) +{ + struct ghash_desc_ctx *dctx = shash_desc_ctx(desc); + struct ghash_ctx *ctx = crypto_shash_ctx(desc->tfm); + + memset(dctx, 0, sizeof(*dctx)); + memcpy(dctx->key, ctx->key, GHASH_BLOCK_SIZE); + + return 0; +} + +static int ghash_setkey(struct crypto_shash *tfm, + const u8 *key, unsigned int keylen) +{ + struct ghash_ctx *ctx = crypto_shash_ctx(tfm); + + if (keylen != GHASH_BLOCK_SIZE) { + crypto_shash_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN); + return -EINVAL; + } + + memcpy(ctx->key, key, GHASH_BLOCK_SIZE); + + return 0; +} + +static int ghash_update(struct shash_desc *desc, + const u8 *src, unsigned int srclen) +{ + struct ghash_desc_ctx *dctx = shash_desc_ctx(desc); + unsigned int n; + u8 *buf = dctx->buffer; + int ret; + + if (dctx->bytes) { + u8 *pos = buf + (GHASH_BLOCK_SIZE - dctx->bytes); + + n = min(srclen, dctx->bytes); + dctx->bytes -= n; + srclen -= n; + + memcpy(pos, src, n); + src += n; + + if (!dctx->bytes) { + ret = crypt_s390_kimd(KIMD_GHASH, dctx, buf, + GHASH_BLOCK_SIZE); + if (ret != GHASH_BLOCK_SIZE) + return -EIO; + } + } + + n = srclen & ~(GHASH_BLOCK_SIZE - 1); + if (n) { + ret = crypt_s390_kimd(KIMD_GHASH, dctx, src, n); + if (ret != n) + return -EIO; + src += n; + srclen -= n; + } + + if (srclen) { + dctx->bytes = GHASH_BLOCK_SIZE - srclen; + memcpy(buf, src, srclen); + } + + return 0; +} + +static int ghash_flush(struct ghash_desc_ctx *dctx) +{ + u8 *buf = dctx->buffer; + int ret; + + if (dctx->bytes) { + u8 *pos = buf + (GHASH_BLOCK_SIZE - dctx->bytes); + + memset(pos, 0, dctx->bytes); + + ret = crypt_s390_kimd(KIMD_GHASH, dctx, buf, GHASH_BLOCK_SIZE); + if (ret != GHASH_BLOCK_SIZE) + return -EIO; + + dctx->bytes = 0; + } + + return 0; +} + +static int ghash_final(struct shash_desc *desc, u8 *dst) +{ + struct ghash_desc_ctx *dctx = shash_desc_ctx(desc); + int ret; + + ret = ghash_flush(dctx); + if (!ret) + memcpy(dst, dctx->icv, GHASH_BLOCK_SIZE); + return ret; +} + +static struct shash_alg ghash_alg = { + .digestsize = GHASH_DIGEST_SIZE, + .init = ghash_init, + .update = ghash_update, + .final = ghash_final, + .setkey = ghash_setkey, + .descsize = sizeof(struct ghash_desc_ctx), + .base = { + .cra_name = "ghash", + .cra_driver_name = "ghash-s390", + .cra_priority = CRYPT_S390_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = GHASH_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct ghash_ctx), + .cra_module = THIS_MODULE, + }, +}; + +static int __init ghash_mod_init(void) +{ + if (!crypt_s390_func_available(KIMD_GHASH, + CRYPT_S390_MSA | CRYPT_S390_MSA4)) + return -EOPNOTSUPP; + + return crypto_register_shash(&ghash_alg); +} + +static void __exit ghash_mod_exit(void) +{ + crypto_unregister_shash(&ghash_alg); +} + +module_init(ghash_mod_init); +module_exit(ghash_mod_exit); + +MODULE_ALIAS_CRYPTO("ghash"); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("GHASH Message Digest Algorithm, s390 implementation"); diff --git a/arch/s390/crypto/prng.c b/arch/s390/crypto/prng.c new file mode 100644 index 000000000..9d5192c94 --- /dev/null +++ b/arch/s390/crypto/prng.c @@ -0,0 +1,919 @@ +/* + * Copyright IBM Corp. 2006, 2015 + * Author(s): Jan Glauber <jan.glauber@de.ibm.com> + * Harald Freudenberger <freude@de.ibm.com> + * Driver for the s390 pseudo random number generator + */ + +#define KMSG_COMPONENT "prng" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/fs.h> +#include <linux/fips.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/device.h> +#include <linux/miscdevice.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/mutex.h> +#include <linux/random.h> +#include <linux/slab.h> +#include <asm/debug.h> +#include <asm/uaccess.h> +#include <asm/timex.h> + +#include "crypt_s390.h" + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("IBM Corporation"); +MODULE_DESCRIPTION("s390 PRNG interface"); + + +#define PRNG_MODE_AUTO 0 +#define PRNG_MODE_TDES 1 +#define PRNG_MODE_SHA512 2 + +static unsigned int prng_mode = PRNG_MODE_AUTO; +module_param_named(mode, prng_mode, int, 0); +MODULE_PARM_DESC(prng_mode, "PRNG mode: 0 - auto, 1 - TDES, 2 - SHA512"); + + +#define PRNG_CHUNKSIZE_TDES_MIN 8 +#define PRNG_CHUNKSIZE_TDES_MAX (64*1024) +#define PRNG_CHUNKSIZE_SHA512_MIN 64 +#define PRNG_CHUNKSIZE_SHA512_MAX (64*1024) + +static unsigned int prng_chunk_size = 256; +module_param_named(chunksize, prng_chunk_size, int, 0); +MODULE_PARM_DESC(prng_chunk_size, "PRNG read chunk size in bytes"); + + +#define PRNG_RESEED_LIMIT_TDES 4096 +#define PRNG_RESEED_LIMIT_TDES_LOWER 4096 +#define PRNG_RESEED_LIMIT_SHA512 100000 +#define PRNG_RESEED_LIMIT_SHA512_LOWER 10000 + +static unsigned int prng_reseed_limit; +module_param_named(reseed_limit, prng_reseed_limit, int, 0); +MODULE_PARM_DESC(prng_reseed_limit, "PRNG reseed limit"); + + +/* + * Any one who considers arithmetical methods of producing random digits is, + * of course, in a state of sin. -- John von Neumann + */ + +static int prng_errorflag; + +#define PRNG_GEN_ENTROPY_FAILED 1 +#define PRNG_SELFTEST_FAILED 2 +#define PRNG_INSTANTIATE_FAILED 3 +#define PRNG_SEED_FAILED 4 +#define PRNG_RESEED_FAILED 5 +#define PRNG_GEN_FAILED 6 + +struct prng_ws_s { + u8 parm_block[32]; + u32 reseed_counter; + u64 byte_counter; +}; + +struct ppno_ws_s { + u32 res; + u32 reseed_counter; + u64 stream_bytes; + u8 V[112]; + u8 C[112]; +}; + +struct prng_data_s { + struct mutex mutex; + union { + struct prng_ws_s prngws; + struct ppno_ws_s ppnows; + }; + u8 *buf; + u32 rest; + u8 *prev; +}; + +static struct prng_data_s *prng_data; + +/* initial parameter block for tdes mode, copied from libica */ +static const u8 initial_parm_block[32] __initconst = { + 0x0F, 0x2B, 0x8E, 0x63, 0x8C, 0x8E, 0xD2, 0x52, + 0x64, 0xB7, 0xA0, 0x7B, 0x75, 0x28, 0xB8, 0xF4, + 0x75, 0x5F, 0xD2, 0xA6, 0x8D, 0x97, 0x11, 0xFF, + 0x49, 0xD8, 0x23, 0xF3, 0x7E, 0x21, 0xEC, 0xA0 }; + + +/*** helper functions ***/ + +static int generate_entropy(u8 *ebuf, size_t nbytes) +{ + int n, ret = 0; + u8 *pg, *h, hash[32]; + + pg = (u8 *) __get_free_page(GFP_KERNEL); + if (!pg) { + prng_errorflag = PRNG_GEN_ENTROPY_FAILED; + return -ENOMEM; + } + + while (nbytes) { + /* fill page with urandom bytes */ + get_random_bytes(pg, PAGE_SIZE); + /* exor page with stckf values */ + for (n = 0; n < PAGE_SIZE / sizeof(u64); n++) { + u64 *p = ((u64 *)pg) + n; + *p ^= get_tod_clock_fast(); + } + n = (nbytes < sizeof(hash)) ? nbytes : sizeof(hash); + if (n < sizeof(hash)) + h = hash; + else + h = ebuf; + /* generate sha256 from this page */ + if (crypt_s390_kimd(KIMD_SHA_256, h, + pg, PAGE_SIZE) != PAGE_SIZE) { + prng_errorflag = PRNG_GEN_ENTROPY_FAILED; + ret = -EIO; + goto out; + } + if (n < sizeof(hash)) + memcpy(ebuf, hash, n); + ret += n; + ebuf += n; + nbytes -= n; + } + +out: + free_page((unsigned long)pg); + return ret; +} + + +/*** tdes functions ***/ + +static void prng_tdes_add_entropy(void) +{ + __u64 entropy[4]; + unsigned int i; + int ret; + + for (i = 0; i < 16; i++) { + ret = crypt_s390_kmc(KMC_PRNG, prng_data->prngws.parm_block, + (char *)entropy, (char *)entropy, + sizeof(entropy)); + BUG_ON(ret < 0 || ret != sizeof(entropy)); + memcpy(prng_data->prngws.parm_block, entropy, sizeof(entropy)); + } +} + + +static void prng_tdes_seed(int nbytes) +{ + char buf[16]; + int i = 0; + + BUG_ON(nbytes > sizeof(buf)); + + get_random_bytes(buf, nbytes); + + /* Add the entropy */ + while (nbytes >= 8) { + *((__u64 *)prng_data->prngws.parm_block) ^= *((__u64 *)(buf+i)); + prng_tdes_add_entropy(); + i += 8; + nbytes -= 8; + } + prng_tdes_add_entropy(); + prng_data->prngws.reseed_counter = 0; +} + + +static int __init prng_tdes_instantiate(void) +{ + int datalen; + + pr_debug("prng runs in TDES mode with " + "chunksize=%d and reseed_limit=%u\n", + prng_chunk_size, prng_reseed_limit); + + /* memory allocation, prng_data struct init, mutex init */ + datalen = sizeof(struct prng_data_s) + prng_chunk_size; + prng_data = kzalloc(datalen, GFP_KERNEL); + if (!prng_data) { + prng_errorflag = PRNG_INSTANTIATE_FAILED; + return -ENOMEM; + } + mutex_init(&prng_data->mutex); + prng_data->buf = ((u8 *)prng_data) + sizeof(struct prng_data_s); + memcpy(prng_data->prngws.parm_block, initial_parm_block, 32); + + /* initialize the PRNG, add 128 bits of entropy */ + prng_tdes_seed(16); + + return 0; +} + + +static void prng_tdes_deinstantiate(void) +{ + pr_debug("The prng module stopped " + "after running in triple DES mode\n"); + kzfree(prng_data); +} + + +/*** sha512 functions ***/ + +static int __init prng_sha512_selftest(void) +{ + /* NIST DRBG testvector for Hash Drbg, Sha-512, Count #0 */ + static const u8 seed[] __initconst = { + 0x6b, 0x50, 0xa7, 0xd8, 0xf8, 0xa5, 0x5d, 0x7a, + 0x3d, 0xf8, 0xbb, 0x40, 0xbc, 0xc3, 0xb7, 0x22, + 0xd8, 0x70, 0x8d, 0xe6, 0x7f, 0xda, 0x01, 0x0b, + 0x03, 0xc4, 0xc8, 0x4d, 0x72, 0x09, 0x6f, 0x8c, + 0x3e, 0xc6, 0x49, 0xcc, 0x62, 0x56, 0xd9, 0xfa, + 0x31, 0xdb, 0x7a, 0x29, 0x04, 0xaa, 0xf0, 0x25 }; + static const u8 V0[] __initconst = { + 0x00, 0xad, 0xe3, 0x6f, 0x9a, 0x01, 0xc7, 0x76, + 0x61, 0x34, 0x35, 0xf5, 0x4e, 0x24, 0x74, 0x22, + 0x21, 0x9a, 0x29, 0x89, 0xc7, 0x93, 0x2e, 0x60, + 0x1e, 0xe8, 0x14, 0x24, 0x8d, 0xd5, 0x03, 0xf1, + 0x65, 0x5d, 0x08, 0x22, 0x72, 0xd5, 0xad, 0x95, + 0xe1, 0x23, 0x1e, 0x8a, 0xa7, 0x13, 0xd9, 0x2b, + 0x5e, 0xbc, 0xbb, 0x80, 0xab, 0x8d, 0xe5, 0x79, + 0xab, 0x5b, 0x47, 0x4e, 0xdd, 0xee, 0x6b, 0x03, + 0x8f, 0x0f, 0x5c, 0x5e, 0xa9, 0x1a, 0x83, 0xdd, + 0xd3, 0x88, 0xb2, 0x75, 0x4b, 0xce, 0x83, 0x36, + 0x57, 0x4b, 0xf1, 0x5c, 0xca, 0x7e, 0x09, 0xc0, + 0xd3, 0x89, 0xc6, 0xe0, 0xda, 0xc4, 0x81, 0x7e, + 0x5b, 0xf9, 0xe1, 0x01, 0xc1, 0x92, 0x05, 0xea, + 0xf5, 0x2f, 0xc6, 0xc6, 0xc7, 0x8f, 0xbc, 0xf4 }; + static const u8 C0[] __initconst = { + 0x00, 0xf4, 0xa3, 0xe5, 0xa0, 0x72, 0x63, 0x95, + 0xc6, 0x4f, 0x48, 0xd0, 0x8b, 0x5b, 0x5f, 0x8e, + 0x6b, 0x96, 0x1f, 0x16, 0xed, 0xbc, 0x66, 0x94, + 0x45, 0x31, 0xd7, 0x47, 0x73, 0x22, 0xa5, 0x86, + 0xce, 0xc0, 0x4c, 0xac, 0x63, 0xb8, 0x39, 0x50, + 0xbf, 0xe6, 0x59, 0x6c, 0x38, 0x58, 0x99, 0x1f, + 0x27, 0xa7, 0x9d, 0x71, 0x2a, 0xb3, 0x7b, 0xf9, + 0xfb, 0x17, 0x86, 0xaa, 0x99, 0x81, 0xaa, 0x43, + 0xe4, 0x37, 0xd3, 0x1e, 0x6e, 0xe5, 0xe6, 0xee, + 0xc2, 0xed, 0x95, 0x4f, 0x53, 0x0e, 0x46, 0x8a, + 0xcc, 0x45, 0xa5, 0xdb, 0x69, 0x0d, 0x81, 0xc9, + 0x32, 0x92, 0xbc, 0x8f, 0x33, 0xe6, 0xf6, 0x09, + 0x7c, 0x8e, 0x05, 0x19, 0x0d, 0xf1, 0xb6, 0xcc, + 0xf3, 0x02, 0x21, 0x90, 0x25, 0xec, 0xed, 0x0e }; + static const u8 random[] __initconst = { + 0x95, 0xb7, 0xf1, 0x7e, 0x98, 0x02, 0xd3, 0x57, + 0x73, 0x92, 0xc6, 0xa9, 0xc0, 0x80, 0x83, 0xb6, + 0x7d, 0xd1, 0x29, 0x22, 0x65, 0xb5, 0xf4, 0x2d, + 0x23, 0x7f, 0x1c, 0x55, 0xbb, 0x9b, 0x10, 0xbf, + 0xcf, 0xd8, 0x2c, 0x77, 0xa3, 0x78, 0xb8, 0x26, + 0x6a, 0x00, 0x99, 0x14, 0x3b, 0x3c, 0x2d, 0x64, + 0x61, 0x1e, 0xee, 0xb6, 0x9a, 0xcd, 0xc0, 0x55, + 0x95, 0x7c, 0x13, 0x9e, 0x8b, 0x19, 0x0c, 0x7a, + 0x06, 0x95, 0x5f, 0x2c, 0x79, 0x7c, 0x27, 0x78, + 0xde, 0x94, 0x03, 0x96, 0xa5, 0x01, 0xf4, 0x0e, + 0x91, 0x39, 0x6a, 0xcf, 0x8d, 0x7e, 0x45, 0xeb, + 0xdb, 0xb5, 0x3b, 0xbf, 0x8c, 0x97, 0x52, 0x30, + 0xd2, 0xf0, 0xff, 0x91, 0x06, 0xc7, 0x61, 0x19, + 0xae, 0x49, 0x8e, 0x7f, 0xbc, 0x03, 0xd9, 0x0f, + 0x8e, 0x4c, 0x51, 0x62, 0x7a, 0xed, 0x5c, 0x8d, + 0x42, 0x63, 0xd5, 0xd2, 0xb9, 0x78, 0x87, 0x3a, + 0x0d, 0xe5, 0x96, 0xee, 0x6d, 0xc7, 0xf7, 0xc2, + 0x9e, 0x37, 0xee, 0xe8, 0xb3, 0x4c, 0x90, 0xdd, + 0x1c, 0xf6, 0xa9, 0xdd, 0xb2, 0x2b, 0x4c, 0xbd, + 0x08, 0x6b, 0x14, 0xb3, 0x5d, 0xe9, 0x3d, 0xa2, + 0xd5, 0xcb, 0x18, 0x06, 0x69, 0x8c, 0xbd, 0x7b, + 0xbb, 0x67, 0xbf, 0xe3, 0xd3, 0x1f, 0xd2, 0xd1, + 0xdb, 0xd2, 0xa1, 0xe0, 0x58, 0xa3, 0xeb, 0x99, + 0xd7, 0xe5, 0x1f, 0x1a, 0x93, 0x8e, 0xed, 0x5e, + 0x1c, 0x1d, 0xe2, 0x3a, 0x6b, 0x43, 0x45, 0xd3, + 0x19, 0x14, 0x09, 0xf9, 0x2f, 0x39, 0xb3, 0x67, + 0x0d, 0x8d, 0xbf, 0xb6, 0x35, 0xd8, 0xe6, 0xa3, + 0x69, 0x32, 0xd8, 0x10, 0x33, 0xd1, 0x44, 0x8d, + 0x63, 0xb4, 0x03, 0xdd, 0xf8, 0x8e, 0x12, 0x1b, + 0x6e, 0x81, 0x9a, 0xc3, 0x81, 0x22, 0x6c, 0x13, + 0x21, 0xe4, 0xb0, 0x86, 0x44, 0xf6, 0x72, 0x7c, + 0x36, 0x8c, 0x5a, 0x9f, 0x7a, 0x4b, 0x3e, 0xe2 }; + + int ret = 0; + u8 buf[sizeof(random)]; + struct ppno_ws_s ws; + + memset(&ws, 0, sizeof(ws)); + + /* initial seed */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_SEED, + &ws, NULL, 0, + seed, sizeof(seed)); + if (ret < 0) { + pr_err("The prng self test seed operation for the " + "SHA-512 mode failed with rc=%d\n", ret); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + + /* check working states V and C */ + if (memcmp(ws.V, V0, sizeof(V0)) != 0 + || memcmp(ws.C, C0, sizeof(C0)) != 0) { + pr_err("The prng self test state test " + "for the SHA-512 mode failed\n"); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + + /* generate random bytes */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_GEN, + &ws, buf, sizeof(buf), + NULL, 0); + if (ret < 0) { + pr_err("The prng self test generate operation for " + "the SHA-512 mode failed with rc=%d\n", ret); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_GEN, + &ws, buf, sizeof(buf), + NULL, 0); + if (ret < 0) { + pr_err("The prng self test generate operation for " + "the SHA-512 mode failed with rc=%d\n", ret); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + + /* check against expected data */ + if (memcmp(buf, random, sizeof(random)) != 0) { + pr_err("The prng self test data test " + "for the SHA-512 mode failed\n"); + prng_errorflag = PRNG_SELFTEST_FAILED; + return -EIO; + } + + return 0; +} + + +static int __init prng_sha512_instantiate(void) +{ + int ret, datalen; + u8 seed[64]; + + pr_debug("prng runs in SHA-512 mode " + "with chunksize=%d and reseed_limit=%u\n", + prng_chunk_size, prng_reseed_limit); + + /* memory allocation, prng_data struct init, mutex init */ + datalen = sizeof(struct prng_data_s) + prng_chunk_size; + if (fips_enabled) + datalen += prng_chunk_size; + prng_data = kzalloc(datalen, GFP_KERNEL); + if (!prng_data) { + prng_errorflag = PRNG_INSTANTIATE_FAILED; + return -ENOMEM; + } + mutex_init(&prng_data->mutex); + prng_data->buf = ((u8 *)prng_data) + sizeof(struct prng_data_s); + + /* selftest */ + ret = prng_sha512_selftest(); + if (ret) + goto outfree; + + /* generate initial seed bytestring, first 48 bytes of entropy */ + ret = generate_entropy(seed, 48); + if (ret != 48) + goto outfree; + /* followed by 16 bytes of unique nonce */ + get_tod_clock_ext(seed + 48); + + /* initial seed of the ppno drng */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_SEED, + &prng_data->ppnows, NULL, 0, + seed, sizeof(seed)); + if (ret < 0) { + prng_errorflag = PRNG_SEED_FAILED; + ret = -EIO; + goto outfree; + } + + /* if fips mode is enabled, generate a first block of random + bytes for the FIPS 140-2 Conditional Self Test */ + if (fips_enabled) { + prng_data->prev = prng_data->buf + prng_chunk_size; + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_GEN, + &prng_data->ppnows, + prng_data->prev, + prng_chunk_size, + NULL, 0); + if (ret < 0 || ret != prng_chunk_size) { + prng_errorflag = PRNG_GEN_FAILED; + ret = -EIO; + goto outfree; + } + } + + return 0; + +outfree: + kfree(prng_data); + return ret; +} + + +static void prng_sha512_deinstantiate(void) +{ + pr_debug("The prng module stopped after running in SHA-512 mode\n"); + kzfree(prng_data); +} + + +static int prng_sha512_reseed(void) +{ + int ret; + u8 seed[32]; + + /* generate 32 bytes of fresh entropy */ + ret = generate_entropy(seed, sizeof(seed)); + if (ret != sizeof(seed)) + return ret; + + /* do a reseed of the ppno drng with this bytestring */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_SEED, + &prng_data->ppnows, NULL, 0, + seed, sizeof(seed)); + if (ret) { + prng_errorflag = PRNG_RESEED_FAILED; + return -EIO; + } + + return 0; +} + + +static int prng_sha512_generate(u8 *buf, size_t nbytes) +{ + int ret; + + /* reseed needed ? */ + if (prng_data->ppnows.reseed_counter > prng_reseed_limit) { + ret = prng_sha512_reseed(); + if (ret) + return ret; + } + + /* PPNO generate */ + ret = crypt_s390_ppno(PPNO_SHA512_DRNG_GEN, + &prng_data->ppnows, buf, nbytes, + NULL, 0); + if (ret < 0 || ret != nbytes) { + prng_errorflag = PRNG_GEN_FAILED; + return -EIO; + } + + /* FIPS 140-2 Conditional Self Test */ + if (fips_enabled) { + if (!memcmp(prng_data->prev, buf, nbytes)) { + prng_errorflag = PRNG_GEN_FAILED; + return -EILSEQ; + } + memcpy(prng_data->prev, buf, nbytes); + } + + return ret; +} + + +/*** file io functions ***/ + +static int prng_open(struct inode *inode, struct file *file) +{ + return nonseekable_open(inode, file); +} + + +static ssize_t prng_tdes_read(struct file *file, char __user *ubuf, + size_t nbytes, loff_t *ppos) +{ + int chunk, n, tmp, ret = 0; + + /* lock prng_data struct */ + if (mutex_lock_interruptible(&prng_data->mutex)) + return -ERESTARTSYS; + + while (nbytes) { + if (need_resched()) { + if (signal_pending(current)) { + if (ret == 0) + ret = -ERESTARTSYS; + break; + } + /* give mutex free before calling schedule() */ + mutex_unlock(&prng_data->mutex); + schedule(); + /* occopy mutex again */ + if (mutex_lock_interruptible(&prng_data->mutex)) { + if (ret == 0) + ret = -ERESTARTSYS; + return ret; + } + } + + /* + * we lose some random bytes if an attacker issues + * reads < 8 bytes, but we don't care + */ + chunk = min_t(int, nbytes, prng_chunk_size); + + /* PRNG only likes multiples of 8 bytes */ + n = (chunk + 7) & -8; + + if (prng_data->prngws.reseed_counter > prng_reseed_limit) + prng_tdes_seed(8); + + /* if the CPU supports PRNG stckf is present too */ + *((unsigned long long *)prng_data->buf) = get_tod_clock_fast(); + + /* + * Beside the STCKF the input for the TDES-EDE is the output + * of the last operation. We differ here from X9.17 since we + * only store one timestamp into the buffer. Padding the whole + * buffer with timestamps does not improve security, since + * successive stckf have nearly constant offsets. + * If an attacker knows the first timestamp it would be + * trivial to guess the additional values. One timestamp + * is therefore enough and still guarantees unique input values. + * + * Note: you can still get strict X9.17 conformity by setting + * prng_chunk_size to 8 bytes. + */ + tmp = crypt_s390_kmc(KMC_PRNG, prng_data->prngws.parm_block, + prng_data->buf, prng_data->buf, n); + if (tmp < 0 || tmp != n) { + ret = -EIO; + break; + } + + prng_data->prngws.byte_counter += n; + prng_data->prngws.reseed_counter += n; + + if (copy_to_user(ubuf, prng_data->buf, chunk)) + return -EFAULT; + + nbytes -= chunk; + ret += chunk; + ubuf += chunk; + } + + /* unlock prng_data struct */ + mutex_unlock(&prng_data->mutex); + + return ret; +} + + +static ssize_t prng_sha512_read(struct file *file, char __user *ubuf, + size_t nbytes, loff_t *ppos) +{ + int n, ret = 0; + u8 *p; + + /* if errorflag is set do nothing and return 'broken pipe' */ + if (prng_errorflag) + return -EPIPE; + + /* lock prng_data struct */ + if (mutex_lock_interruptible(&prng_data->mutex)) + return -ERESTARTSYS; + + while (nbytes) { + if (need_resched()) { + if (signal_pending(current)) { + if (ret == 0) + ret = -ERESTARTSYS; + break; + } + /* give mutex free before calling schedule() */ + mutex_unlock(&prng_data->mutex); + schedule(); + /* occopy mutex again */ + if (mutex_lock_interruptible(&prng_data->mutex)) { + if (ret == 0) + ret = -ERESTARTSYS; + return ret; + } + } + if (prng_data->rest) { + /* push left over random bytes from the previous read */ + p = prng_data->buf + prng_chunk_size - prng_data->rest; + n = (nbytes < prng_data->rest) ? + nbytes : prng_data->rest; + prng_data->rest -= n; + } else { + /* generate one chunk of random bytes into read buf */ + p = prng_data->buf; + n = prng_sha512_generate(p, prng_chunk_size); + if (n < 0) { + ret = n; + break; + } + if (nbytes < prng_chunk_size) { + n = nbytes; + prng_data->rest = prng_chunk_size - n; + } else { + n = prng_chunk_size; + prng_data->rest = 0; + } + } + if (copy_to_user(ubuf, p, n)) { + ret = -EFAULT; + break; + } + ubuf += n; + nbytes -= n; + ret += n; + } + + /* unlock prng_data struct */ + mutex_unlock(&prng_data->mutex); + + return ret; +} + + +/*** sysfs stuff ***/ + +static const struct file_operations prng_sha512_fops = { + .owner = THIS_MODULE, + .open = &prng_open, + .release = NULL, + .read = &prng_sha512_read, + .llseek = noop_llseek, +}; +static const struct file_operations prng_tdes_fops = { + .owner = THIS_MODULE, + .open = &prng_open, + .release = NULL, + .read = &prng_tdes_read, + .llseek = noop_llseek, +}; + +static struct miscdevice prng_sha512_dev = { + .name = "prandom", + .minor = MISC_DYNAMIC_MINOR, + .fops = &prng_sha512_fops, +}; +static struct miscdevice prng_tdes_dev = { + .name = "prandom", + .minor = MISC_DYNAMIC_MINOR, + .fops = &prng_tdes_fops, +}; + + +/* chunksize attribute (ro) */ +static ssize_t prng_chunksize_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%u\n", prng_chunk_size); +} +static DEVICE_ATTR(chunksize, 0444, prng_chunksize_show, NULL); + +/* counter attribute (ro) */ +static ssize_t prng_counter_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + u64 counter; + + if (mutex_lock_interruptible(&prng_data->mutex)) + return -ERESTARTSYS; + if (prng_mode == PRNG_MODE_SHA512) + counter = prng_data->ppnows.stream_bytes; + else + counter = prng_data->prngws.byte_counter; + mutex_unlock(&prng_data->mutex); + + return snprintf(buf, PAGE_SIZE, "%llu\n", counter); +} +static DEVICE_ATTR(byte_counter, 0444, prng_counter_show, NULL); + +/* errorflag attribute (ro) */ +static ssize_t prng_errorflag_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%d\n", prng_errorflag); +} +static DEVICE_ATTR(errorflag, 0444, prng_errorflag_show, NULL); + +/* mode attribute (ro) */ +static ssize_t prng_mode_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (prng_mode == PRNG_MODE_TDES) + return snprintf(buf, PAGE_SIZE, "TDES\n"); + else + return snprintf(buf, PAGE_SIZE, "SHA512\n"); +} +static DEVICE_ATTR(mode, 0444, prng_mode_show, NULL); + +/* reseed attribute (w) */ +static ssize_t prng_reseed_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + if (mutex_lock_interruptible(&prng_data->mutex)) + return -ERESTARTSYS; + prng_sha512_reseed(); + mutex_unlock(&prng_data->mutex); + + return count; +} +static DEVICE_ATTR(reseed, 0200, NULL, prng_reseed_store); + +/* reseed limit attribute (rw) */ +static ssize_t prng_reseed_limit_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%u\n", prng_reseed_limit); +} +static ssize_t prng_reseed_limit_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + unsigned limit; + + if (sscanf(buf, "%u\n", &limit) != 1) + return -EINVAL; + + if (prng_mode == PRNG_MODE_SHA512) { + if (limit < PRNG_RESEED_LIMIT_SHA512_LOWER) + return -EINVAL; + } else { + if (limit < PRNG_RESEED_LIMIT_TDES_LOWER) + return -EINVAL; + } + + prng_reseed_limit = limit; + + return count; +} +static DEVICE_ATTR(reseed_limit, 0644, + prng_reseed_limit_show, prng_reseed_limit_store); + +/* strength attribute (ro) */ +static ssize_t prng_strength_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return snprintf(buf, PAGE_SIZE, "256\n"); +} +static DEVICE_ATTR(strength, 0444, prng_strength_show, NULL); + +static struct attribute *prng_sha512_dev_attrs[] = { + &dev_attr_errorflag.attr, + &dev_attr_chunksize.attr, + &dev_attr_byte_counter.attr, + &dev_attr_mode.attr, + &dev_attr_reseed.attr, + &dev_attr_reseed_limit.attr, + &dev_attr_strength.attr, + NULL +}; +static struct attribute *prng_tdes_dev_attrs[] = { + &dev_attr_chunksize.attr, + &dev_attr_byte_counter.attr, + &dev_attr_mode.attr, + NULL +}; + +static struct attribute_group prng_sha512_dev_attr_group = { + .attrs = prng_sha512_dev_attrs +}; +static struct attribute_group prng_tdes_dev_attr_group = { + .attrs = prng_tdes_dev_attrs +}; + + +/*** module init and exit ***/ + +static int __init prng_init(void) +{ + int ret; + + /* check if the CPU has a PRNG */ + if (!crypt_s390_func_available(KMC_PRNG, CRYPT_S390_MSA)) + return -EOPNOTSUPP; + + /* choose prng mode */ + if (prng_mode != PRNG_MODE_TDES) { + /* check for MSA5 support for PPNO operations */ + if (!crypt_s390_func_available(PPNO_SHA512_DRNG_GEN, + CRYPT_S390_MSA5)) { + if (prng_mode == PRNG_MODE_SHA512) { + pr_err("The prng module cannot " + "start in SHA-512 mode\n"); + return -EOPNOTSUPP; + } + prng_mode = PRNG_MODE_TDES; + } else + prng_mode = PRNG_MODE_SHA512; + } + + if (prng_mode == PRNG_MODE_SHA512) { + + /* SHA512 mode */ + + if (prng_chunk_size < PRNG_CHUNKSIZE_SHA512_MIN + || prng_chunk_size > PRNG_CHUNKSIZE_SHA512_MAX) + return -EINVAL; + prng_chunk_size = (prng_chunk_size + 0x3f) & ~0x3f; + + if (prng_reseed_limit == 0) + prng_reseed_limit = PRNG_RESEED_LIMIT_SHA512; + else if (prng_reseed_limit < PRNG_RESEED_LIMIT_SHA512_LOWER) + return -EINVAL; + + ret = prng_sha512_instantiate(); + if (ret) + goto out; + + ret = misc_register(&prng_sha512_dev); + if (ret) { + prng_sha512_deinstantiate(); + goto out; + } + ret = sysfs_create_group(&prng_sha512_dev.this_device->kobj, + &prng_sha512_dev_attr_group); + if (ret) { + misc_deregister(&prng_sha512_dev); + prng_sha512_deinstantiate(); + goto out; + } + + } else { + + /* TDES mode */ + + if (prng_chunk_size < PRNG_CHUNKSIZE_TDES_MIN + || prng_chunk_size > PRNG_CHUNKSIZE_TDES_MAX) + return -EINVAL; + prng_chunk_size = (prng_chunk_size + 0x07) & ~0x07; + + if (prng_reseed_limit == 0) + prng_reseed_limit = PRNG_RESEED_LIMIT_TDES; + else if (prng_reseed_limit < PRNG_RESEED_LIMIT_TDES_LOWER) + return -EINVAL; + + ret = prng_tdes_instantiate(); + if (ret) + goto out; + + ret = misc_register(&prng_tdes_dev); + if (ret) { + prng_tdes_deinstantiate(); + goto out; + } + ret = sysfs_create_group(&prng_tdes_dev.this_device->kobj, + &prng_tdes_dev_attr_group); + if (ret) { + misc_deregister(&prng_tdes_dev); + prng_tdes_deinstantiate(); + goto out; + } + + } + +out: + return ret; +} + + +static void __exit prng_exit(void) +{ + if (prng_mode == PRNG_MODE_SHA512) { + sysfs_remove_group(&prng_sha512_dev.this_device->kobj, + &prng_sha512_dev_attr_group); + misc_deregister(&prng_sha512_dev); + prng_sha512_deinstantiate(); + } else { + sysfs_remove_group(&prng_tdes_dev.this_device->kobj, + &prng_tdes_dev_attr_group); + misc_deregister(&prng_tdes_dev); + prng_tdes_deinstantiate(); + } +} + + +module_init(prng_init); +module_exit(prng_exit); diff --git a/arch/s390/crypto/sha.h b/arch/s390/crypto/sha.h new file mode 100644 index 000000000..f4e9dc716 --- /dev/null +++ b/arch/s390/crypto/sha.h @@ -0,0 +1,37 @@ +/* + * Cryptographic API. + * + * s390 generic implementation of the SHA Secure Hash Algorithms. + * + * Copyright IBM Corp. 2007 + * Author(s): Jan Glauber (jang@de.ibm.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. + * + */ +#ifndef _CRYPTO_ARCH_S390_SHA_H +#define _CRYPTO_ARCH_S390_SHA_H + +#include <linux/crypto.h> +#include <crypto/sha.h> + +/* must be big enough for the largest SHA variant */ +#define SHA_MAX_STATE_SIZE 16 +#define SHA_MAX_BLOCK_SIZE SHA512_BLOCK_SIZE + +struct s390_sha_ctx { + u64 count; /* message length in bytes */ + u32 state[SHA_MAX_STATE_SIZE]; + u8 buf[2 * SHA_MAX_BLOCK_SIZE]; + int func; /* KIMD function to use */ +}; + +struct shash_desc; + +int s390_sha_update(struct shash_desc *desc, const u8 *data, unsigned int len); +int s390_sha_final(struct shash_desc *desc, u8 *out); + +#endif diff --git a/arch/s390/crypto/sha1_s390.c b/arch/s390/crypto/sha1_s390.c new file mode 100644 index 000000000..5b2bee323 --- /dev/null +++ b/arch/s390/crypto/sha1_s390.c @@ -0,0 +1,108 @@ +/* + * Cryptographic API. + * + * s390 implementation of the SHA1 Secure Hash Algorithm. + * + * Derived from cryptoapi implementation, adapted for in-place + * scatterlist interface. Originally based on the public domain + * implementation written by Steve Reid. + * + * s390 Version: + * Copyright IBM Corp. 2003, 2007 + * Author(s): Thomas Spatzier + * Jan Glauber (jan.glauber@de.ibm.com) + * + * Derived from "crypto/sha1_generic.c" + * Copyright (c) Alan Smithee. + * Copyright (c) Andrew McDonald <andrew@mcdonald.org.uk> + * Copyright (c) Jean-Francois Dive <jef@linuxbe.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. + * + */ +#include <crypto/internal/hash.h> +#include <linux/init.h> +#include <linux/module.h> +#include <crypto/sha.h> + +#include "crypt_s390.h" +#include "sha.h" + +static int sha1_init(struct shash_desc *desc) +{ + struct s390_sha_ctx *sctx = shash_desc_ctx(desc); + + sctx->state[0] = SHA1_H0; + sctx->state[1] = SHA1_H1; + sctx->state[2] = SHA1_H2; + sctx->state[3] = SHA1_H3; + sctx->state[4] = SHA1_H4; + sctx->count = 0; + sctx->func = KIMD_SHA_1; + + return 0; +} + +static int sha1_export(struct shash_desc *desc, void *out) +{ + struct s390_sha_ctx *sctx = shash_desc_ctx(desc); + struct sha1_state *octx = out; + + octx->count = sctx->count; + memcpy(octx->state, sctx->state, sizeof(octx->state)); + memcpy(octx->buffer, sctx->buf, sizeof(octx->buffer)); + return 0; +} + +static int sha1_import(struct shash_desc *desc, const void *in) +{ + struct s390_sha_ctx *sctx = shash_desc_ctx(desc); + const struct sha1_state *ictx = in; + + sctx->count = ictx->count; + memcpy(sctx->state, ictx->state, sizeof(ictx->state)); + memcpy(sctx->buf, ictx->buffer, sizeof(ictx->buffer)); + sctx->func = KIMD_SHA_1; + return 0; +} + +static struct shash_alg alg = { + .digestsize = SHA1_DIGEST_SIZE, + .init = sha1_init, + .update = s390_sha_update, + .final = s390_sha_final, + .export = sha1_export, + .import = sha1_import, + .descsize = sizeof(struct s390_sha_ctx), + .statesize = sizeof(struct sha1_state), + .base = { + .cra_name = "sha1", + .cra_driver_name= "sha1-s390", + .cra_priority = CRYPT_S390_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA1_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static int __init sha1_s390_init(void) +{ + if (!crypt_s390_func_available(KIMD_SHA_1, CRYPT_S390_MSA)) + return -EOPNOTSUPP; + return crypto_register_shash(&alg); +} + +static void __exit sha1_s390_fini(void) +{ + crypto_unregister_shash(&alg); +} + +module_init(sha1_s390_init); +module_exit(sha1_s390_fini); + +MODULE_ALIAS_CRYPTO("sha1"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("SHA1 Secure Hash Algorithm"); diff --git a/arch/s390/crypto/sha256_s390.c b/arch/s390/crypto/sha256_s390.c new file mode 100644 index 000000000..b74ff1581 --- /dev/null +++ b/arch/s390/crypto/sha256_s390.c @@ -0,0 +1,149 @@ +/* + * Cryptographic API. + * + * s390 implementation of the SHA256 and SHA224 Secure Hash Algorithm. + * + * s390 Version: + * Copyright IBM Corp. 2005, 2011 + * Author(s): Jan Glauber (jang@de.ibm.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. + * + */ +#include <crypto/internal/hash.h> +#include <linux/init.h> +#include <linux/module.h> +#include <crypto/sha.h> + +#include "crypt_s390.h" +#include "sha.h" + +static int sha256_init(struct shash_desc *desc) +{ + struct s390_sha_ctx *sctx = shash_desc_ctx(desc); + + sctx->state[0] = SHA256_H0; + sctx->state[1] = SHA256_H1; + sctx->state[2] = SHA256_H2; + sctx->state[3] = SHA256_H3; + sctx->state[4] = SHA256_H4; + sctx->state[5] = SHA256_H5; + sctx->state[6] = SHA256_H6; + sctx->state[7] = SHA256_H7; + sctx->count = 0; + sctx->func = KIMD_SHA_256; + + return 0; +} + +static int sha256_export(struct shash_desc *desc, void *out) +{ + struct s390_sha_ctx *sctx = shash_desc_ctx(desc); + struct sha256_state *octx = out; + + octx->count = sctx->count; + memcpy(octx->state, sctx->state, sizeof(octx->state)); + memcpy(octx->buf, sctx->buf, sizeof(octx->buf)); + return 0; +} + +static int sha256_import(struct shash_desc *desc, const void *in) +{ + struct s390_sha_ctx *sctx = shash_desc_ctx(desc); + const struct sha256_state *ictx = in; + + sctx->count = ictx->count; + memcpy(sctx->state, ictx->state, sizeof(ictx->state)); + memcpy(sctx->buf, ictx->buf, sizeof(ictx->buf)); + sctx->func = KIMD_SHA_256; + return 0; +} + +static struct shash_alg sha256_alg = { + .digestsize = SHA256_DIGEST_SIZE, + .init = sha256_init, + .update = s390_sha_update, + .final = s390_sha_final, + .export = sha256_export, + .import = sha256_import, + .descsize = sizeof(struct s390_sha_ctx), + .statesize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha256", + .cra_driver_name= "sha256-s390", + .cra_priority = CRYPT_S390_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA256_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static int sha224_init(struct shash_desc *desc) +{ + struct s390_sha_ctx *sctx = shash_desc_ctx(desc); + + sctx->state[0] = SHA224_H0; + sctx->state[1] = SHA224_H1; + sctx->state[2] = SHA224_H2; + sctx->state[3] = SHA224_H3; + sctx->state[4] = SHA224_H4; + sctx->state[5] = SHA224_H5; + sctx->state[6] = SHA224_H6; + sctx->state[7] = SHA224_H7; + sctx->count = 0; + sctx->func = KIMD_SHA_256; + + return 0; +} + +static struct shash_alg sha224_alg = { + .digestsize = SHA224_DIGEST_SIZE, + .init = sha224_init, + .update = s390_sha_update, + .final = s390_sha_final, + .export = sha256_export, + .import = sha256_import, + .descsize = sizeof(struct s390_sha_ctx), + .statesize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha224", + .cra_driver_name= "sha224-s390", + .cra_priority = CRYPT_S390_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA224_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static int __init sha256_s390_init(void) +{ + int ret; + + if (!crypt_s390_func_available(KIMD_SHA_256, CRYPT_S390_MSA)) + return -EOPNOTSUPP; + ret = crypto_register_shash(&sha256_alg); + if (ret < 0) + goto out; + ret = crypto_register_shash(&sha224_alg); + if (ret < 0) + crypto_unregister_shash(&sha256_alg); +out: + return ret; +} + +static void __exit sha256_s390_fini(void) +{ + crypto_unregister_shash(&sha224_alg); + crypto_unregister_shash(&sha256_alg); +} + +module_init(sha256_s390_init); +module_exit(sha256_s390_fini); + +MODULE_ALIAS_CRYPTO("sha256"); +MODULE_ALIAS_CRYPTO("sha224"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("SHA256 and SHA224 Secure Hash Algorithm"); diff --git a/arch/s390/crypto/sha512_s390.c b/arch/s390/crypto/sha512_s390.c new file mode 100644 index 000000000..0c36989ba --- /dev/null +++ b/arch/s390/crypto/sha512_s390.c @@ -0,0 +1,155 @@ +/* + * Cryptographic API. + * + * s390 implementation of the SHA512 and SHA38 Secure Hash Algorithm. + * + * Copyright IBM Corp. 2007 + * Author(s): Jan Glauber (jang@de.ibm.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. + * + */ +#include <crypto/internal/hash.h> +#include <crypto/sha.h> +#include <linux/errno.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> + +#include "sha.h" +#include "crypt_s390.h" + +static int sha512_init(struct shash_desc *desc) +{ + struct s390_sha_ctx *ctx = shash_desc_ctx(desc); + + *(__u64 *)&ctx->state[0] = 0x6a09e667f3bcc908ULL; + *(__u64 *)&ctx->state[2] = 0xbb67ae8584caa73bULL; + *(__u64 *)&ctx->state[4] = 0x3c6ef372fe94f82bULL; + *(__u64 *)&ctx->state[6] = 0xa54ff53a5f1d36f1ULL; + *(__u64 *)&ctx->state[8] = 0x510e527fade682d1ULL; + *(__u64 *)&ctx->state[10] = 0x9b05688c2b3e6c1fULL; + *(__u64 *)&ctx->state[12] = 0x1f83d9abfb41bd6bULL; + *(__u64 *)&ctx->state[14] = 0x5be0cd19137e2179ULL; + ctx->count = 0; + ctx->func = KIMD_SHA_512; + + return 0; +} + +static int sha512_export(struct shash_desc *desc, void *out) +{ + struct s390_sha_ctx *sctx = shash_desc_ctx(desc); + struct sha512_state *octx = out; + + octx->count[0] = sctx->count; + octx->count[1] = 0; + memcpy(octx->state, sctx->state, sizeof(octx->state)); + memcpy(octx->buf, sctx->buf, sizeof(octx->buf)); + return 0; +} + +static int sha512_import(struct shash_desc *desc, const void *in) +{ + struct s390_sha_ctx *sctx = shash_desc_ctx(desc); + const struct sha512_state *ictx = in; + + if (unlikely(ictx->count[1])) + return -ERANGE; + sctx->count = ictx->count[0]; + + memcpy(sctx->state, ictx->state, sizeof(ictx->state)); + memcpy(sctx->buf, ictx->buf, sizeof(ictx->buf)); + sctx->func = KIMD_SHA_512; + return 0; +} + +static struct shash_alg sha512_alg = { + .digestsize = SHA512_DIGEST_SIZE, + .init = sha512_init, + .update = s390_sha_update, + .final = s390_sha_final, + .export = sha512_export, + .import = sha512_import, + .descsize = sizeof(struct s390_sha_ctx), + .statesize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha512", + .cra_driver_name= "sha512-s390", + .cra_priority = CRYPT_S390_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA512_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +MODULE_ALIAS_CRYPTO("sha512"); + +static int sha384_init(struct shash_desc *desc) +{ + struct s390_sha_ctx *ctx = shash_desc_ctx(desc); + + *(__u64 *)&ctx->state[0] = 0xcbbb9d5dc1059ed8ULL; + *(__u64 *)&ctx->state[2] = 0x629a292a367cd507ULL; + *(__u64 *)&ctx->state[4] = 0x9159015a3070dd17ULL; + *(__u64 *)&ctx->state[6] = 0x152fecd8f70e5939ULL; + *(__u64 *)&ctx->state[8] = 0x67332667ffc00b31ULL; + *(__u64 *)&ctx->state[10] = 0x8eb44a8768581511ULL; + *(__u64 *)&ctx->state[12] = 0xdb0c2e0d64f98fa7ULL; + *(__u64 *)&ctx->state[14] = 0x47b5481dbefa4fa4ULL; + ctx->count = 0; + ctx->func = KIMD_SHA_512; + + return 0; +} + +static struct shash_alg sha384_alg = { + .digestsize = SHA384_DIGEST_SIZE, + .init = sha384_init, + .update = s390_sha_update, + .final = s390_sha_final, + .export = sha512_export, + .import = sha512_import, + .descsize = sizeof(struct s390_sha_ctx), + .statesize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha384", + .cra_driver_name= "sha384-s390", + .cra_priority = CRYPT_S390_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA384_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct s390_sha_ctx), + .cra_module = THIS_MODULE, + } +}; + +MODULE_ALIAS_CRYPTO("sha384"); + +static int __init init(void) +{ + int ret; + + if (!crypt_s390_func_available(KIMD_SHA_512, CRYPT_S390_MSA)) + return -EOPNOTSUPP; + if ((ret = crypto_register_shash(&sha512_alg)) < 0) + goto out; + if ((ret = crypto_register_shash(&sha384_alg)) < 0) + crypto_unregister_shash(&sha512_alg); +out: + return ret; +} + +static void __exit fini(void) +{ + crypto_unregister_shash(&sha512_alg); + crypto_unregister_shash(&sha384_alg); +} + +module_init(init); +module_exit(fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("SHA512 and SHA-384 Secure Hash Algorithm"); diff --git a/arch/s390/crypto/sha_common.c b/arch/s390/crypto/sha_common.c new file mode 100644 index 000000000..8620b0ec9 --- /dev/null +++ b/arch/s390/crypto/sha_common.c @@ -0,0 +1,106 @@ +/* + * Cryptographic API. + * + * s390 generic implementation of the SHA Secure Hash Algorithms. + * + * Copyright IBM Corp. 2007 + * Author(s): Jan Glauber (jang@de.ibm.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. + * + */ + +#include <crypto/internal/hash.h> +#include <linux/module.h> +#include "sha.h" +#include "crypt_s390.h" + +int s390_sha_update(struct shash_desc *desc, const u8 *data, unsigned int len) +{ + struct s390_sha_ctx *ctx = shash_desc_ctx(desc); + unsigned int bsize = crypto_shash_blocksize(desc->tfm); + unsigned int index; + int ret; + + /* how much is already in the buffer? */ + index = ctx->count & (bsize - 1); + ctx->count += len; + + if ((index + len) < bsize) + goto store; + + /* process one stored block */ + if (index) { + memcpy(ctx->buf + index, data, bsize - index); + ret = crypt_s390_kimd(ctx->func, ctx->state, ctx->buf, bsize); + if (ret != bsize) + return -EIO; + data += bsize - index; + len -= bsize - index; + index = 0; + } + + /* process as many blocks as possible */ + if (len >= bsize) { + ret = crypt_s390_kimd(ctx->func, ctx->state, data, + len & ~(bsize - 1)); + if (ret != (len & ~(bsize - 1))) + return -EIO; + data += ret; + len -= ret; + } +store: + if (len) + memcpy(ctx->buf + index , data, len); + + return 0; +} +EXPORT_SYMBOL_GPL(s390_sha_update); + +int s390_sha_final(struct shash_desc *desc, u8 *out) +{ + struct s390_sha_ctx *ctx = shash_desc_ctx(desc); + unsigned int bsize = crypto_shash_blocksize(desc->tfm); + u64 bits; + unsigned int index, end, plen; + int ret; + + /* SHA-512 uses 128 bit padding length */ + plen = (bsize > SHA256_BLOCK_SIZE) ? 16 : 8; + + /* must perform manual padding */ + index = ctx->count & (bsize - 1); + end = (index < bsize - plen) ? bsize : (2 * bsize); + + /* start pad with 1 */ + ctx->buf[index] = 0x80; + index++; + + /* pad with zeros */ + memset(ctx->buf + index, 0x00, end - index - 8); + + /* + * Append message length. Well, SHA-512 wants a 128 bit length value, + * nevertheless we use u64, should be enough for now... + */ + bits = ctx->count * 8; + memcpy(ctx->buf + end - 8, &bits, sizeof(bits)); + + ret = crypt_s390_kimd(ctx->func, ctx->state, ctx->buf, end); + if (ret != end) + return -EIO; + + /* copy digest to out */ + memcpy(out, ctx->state, crypto_shash_digestsize(desc->tfm)); + /* wipe context */ + memset(ctx, 0, sizeof *ctx); + + return 0; +} +EXPORT_SYMBOL_GPL(s390_sha_final); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("s390 SHA cipher common functions"); diff --git a/arch/s390/defconfig b/arch/s390/defconfig new file mode 100644 index 000000000..83ef702d2 --- /dev/null +++ b/arch/s390/defconfig @@ -0,0 +1,203 @@ +CONFIG_SYSVIPC=y +CONFIG_POSIX_MQUEUE=y +CONFIG_FHANDLE=y +CONFIG_AUDIT=y +CONFIG_NO_HZ=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_TASKSTATS=y +CONFIG_TASK_DELAY_ACCT=y +CONFIG_TASK_XACCT=y +CONFIG_TASK_IO_ACCOUNTING=y +CONFIG_RCU_FAST_NO_HZ=y +CONFIG_IKCONFIG=y +CONFIG_IKCONFIG_PROC=y +CONFIG_CGROUPS=y +CONFIG_CPUSETS=y +CONFIG_CGROUP_CPUACCT=y +CONFIG_MEMCG=y +CONFIG_MEMCG_SWAP=y +CONFIG_CGROUP_SCHED=y +CONFIG_RT_GROUP_SCHED=y +CONFIG_BLK_CGROUP=y +CONFIG_NAMESPACES=y +CONFIG_BLK_DEV_INITRD=y +CONFIG_EXPERT=y +CONFIG_BPF_SYSCALL=y +# CONFIG_COMPAT_BRK is not set +CONFIG_PROFILING=y +CONFIG_OPROFILE=y +CONFIG_KPROBES=y +CONFIG_JUMP_LABEL=y +CONFIG_MODULES=y +CONFIG_MODULE_UNLOAD=y +CONFIG_MODVERSIONS=y +CONFIG_PARTITION_ADVANCED=y +CONFIG_IBM_PARTITION=y +CONFIG_DEFAULT_DEADLINE=y +CONFIG_MARCH_Z196=y +CONFIG_NR_CPUS=256 +CONFIG_HZ_100=y +CONFIG_MEMORY_HOTPLUG=y +CONFIG_MEMORY_HOTREMOVE=y +CONFIG_KSM=y +CONFIG_TRANSPARENT_HUGEPAGE=y +CONFIG_CMA=y +CONFIG_CRASH_DUMP=y +CONFIG_BINFMT_MISC=m +CONFIG_HIBERNATION=y +CONFIG_NET=y +CONFIG_PACKET=y +CONFIG_UNIX=y +CONFIG_NET_KEY=y +CONFIG_INET=y +CONFIG_IP_MULTICAST=y +# CONFIG_INET_LRO is not set +CONFIG_IPV6=y +CONFIG_L2TP=m +CONFIG_L2TP_DEBUGFS=m +CONFIG_VLAN_8021Q=y +CONFIG_NET_SCHED=y +CONFIG_NET_SCH_CBQ=m +CONFIG_NET_SCH_PRIO=m +CONFIG_NET_SCH_RED=m +CONFIG_NET_SCH_SFQ=m +CONFIG_NET_SCH_TEQL=m +CONFIG_NET_SCH_TBF=m +CONFIG_NET_SCH_GRED=m +CONFIG_NET_SCH_DSMARK=m +CONFIG_NET_CLS_TCINDEX=m +CONFIG_NET_CLS_ROUTE4=m +CONFIG_NET_CLS_FW=m +CONFIG_NET_CLS_U32=m +CONFIG_CLS_U32_MARK=y +CONFIG_NET_CLS_RSVP=m +CONFIG_NET_CLS_RSVP6=m +CONFIG_NET_CLS_ACT=y +CONFIG_NET_ACT_POLICE=y +CONFIG_BPF_JIT=y +CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug" +CONFIG_DEVTMPFS=y +CONFIG_BLK_DEV_LOOP=m +CONFIG_BLK_DEV_NBD=m +CONFIG_BLK_DEV_RAM=y +CONFIG_VIRTIO_BLK=y +CONFIG_SCSI=y +CONFIG_BLK_DEV_SD=y +CONFIG_CHR_DEV_ST=y +CONFIG_BLK_DEV_SR=y +CONFIG_BLK_DEV_SR_VENDOR=y +CONFIG_CHR_DEV_SG=y +CONFIG_SCSI_CONSTANTS=y +CONFIG_SCSI_LOGGING=y +CONFIG_SCSI_SCAN_ASYNC=y +CONFIG_SCSI_FC_ATTRS=y +CONFIG_ZFCP=y +CONFIG_SCSI_VIRTIO=y +CONFIG_NETDEVICES=y +CONFIG_BONDING=m +CONFIG_DUMMY=m +CONFIG_EQUALIZER=m +CONFIG_TUN=m +CONFIG_VIRTIO_NET=y +# CONFIG_INPUT is not set +# CONFIG_SERIO is not set +CONFIG_RAW_DRIVER=m +CONFIG_VIRTIO_BALLOON=y +CONFIG_EXT4_FS=y +CONFIG_EXT4_FS_POSIX_ACL=y +CONFIG_EXT4_FS_SECURITY=y +CONFIG_XFS_FS=y +CONFIG_XFS_QUOTA=y +CONFIG_XFS_POSIX_ACL=y +CONFIG_XFS_RT=y +CONFIG_BTRFS_FS=y +CONFIG_BTRFS_FS_POSIX_ACL=y +CONFIG_FANOTIFY=y +CONFIG_FUSE_FS=y +CONFIG_PROC_KCORE=y +CONFIG_TMPFS=y +CONFIG_TMPFS_POSIX_ACL=y +CONFIG_HUGETLBFS=y +# CONFIG_NETWORK_FILESYSTEMS is not set +CONFIG_UNUSED_SYMBOLS=y +CONFIG_DEBUG_SECTION_MISMATCH=y +CONFIG_DEBUG_FORCE_WEAK_PER_CPU=y +CONFIG_MAGIC_SYSRQ=y +CONFIG_DEBUG_PAGEALLOC=y +CONFIG_DETECT_HUNG_TASK=y +CONFIG_TIMER_STATS=y +CONFIG_DEBUG_RT_MUTEXES=y +CONFIG_PROVE_LOCKING=y +CONFIG_LOCK_STAT=y +CONFIG_DEBUG_LOCKDEP=y +CONFIG_DEBUG_ATOMIC_SLEEP=y +CONFIG_DEBUG_LIST=y +CONFIG_DEBUG_PI_LIST=y +CONFIG_DEBUG_SG=y +CONFIG_DEBUG_NOTIFIERS=y +CONFIG_PROVE_RCU=y +CONFIG_RCU_CPU_STALL_TIMEOUT=60 +CONFIG_RCU_TRACE=y +CONFIG_LATENCYTOP=y +CONFIG_DEBUG_STRICT_USER_COPY_CHECKS=y +CONFIG_BLK_DEV_IO_TRACE=y +CONFIG_KPROBES_SANITY_TEST=y +# CONFIG_STRICT_DEVMEM is not set +CONFIG_S390_PTDUMP=y +CONFIG_CRYPTO_CRYPTD=m +CONFIG_CRYPTO_AUTHENC=m +CONFIG_CRYPTO_TEST=m +CONFIG_CRYPTO_CCM=m +CONFIG_CRYPTO_GCM=m +CONFIG_CRYPTO_CBC=y +CONFIG_CRYPTO_CTS=m +CONFIG_CRYPTO_ECB=m +CONFIG_CRYPTO_LRW=m +CONFIG_CRYPTO_PCBC=m +CONFIG_CRYPTO_XTS=m +CONFIG_CRYPTO_CMAC=m +CONFIG_CRYPTO_XCBC=m +CONFIG_CRYPTO_VMAC=m +CONFIG_CRYPTO_CRC32=m +CONFIG_CRYPTO_MD4=m +CONFIG_CRYPTO_MICHAEL_MIC=m +CONFIG_CRYPTO_RMD128=m +CONFIG_CRYPTO_RMD160=m +CONFIG_CRYPTO_RMD256=m +CONFIG_CRYPTO_RMD320=m +CONFIG_CRYPTO_SHA256=y +CONFIG_CRYPTO_SHA512=m +CONFIG_CRYPTO_TGR192=m +CONFIG_CRYPTO_WP512=m +CONFIG_CRYPTO_ANUBIS=m +CONFIG_CRYPTO_ARC4=m +CONFIG_CRYPTO_BLOWFISH=m +CONFIG_CRYPTO_CAMELLIA=m +CONFIG_CRYPTO_CAST5=m +CONFIG_CRYPTO_CAST6=m +CONFIG_CRYPTO_FCRYPT=m +CONFIG_CRYPTO_KHAZAD=m +CONFIG_CRYPTO_SALSA20=m +CONFIG_CRYPTO_SEED=m +CONFIG_CRYPTO_SERPENT=m +CONFIG_CRYPTO_TEA=m +CONFIG_CRYPTO_TWOFISH=m +CONFIG_CRYPTO_DEFLATE=m +CONFIG_CRYPTO_ZLIB=m +CONFIG_CRYPTO_LZO=m +CONFIG_CRYPTO_LZ4=m +CONFIG_CRYPTO_LZ4HC=m +CONFIG_ZCRYPT=m +CONFIG_CRYPTO_SHA1_S390=m +CONFIG_CRYPTO_SHA256_S390=m +CONFIG_CRYPTO_SHA512_S390=m +CONFIG_CRYPTO_DES_S390=m +CONFIG_CRYPTO_AES_S390=m +CONFIG_CRC7=m +# CONFIG_XZ_DEC_X86 is not set +# CONFIG_XZ_DEC_POWERPC is not set +# CONFIG_XZ_DEC_IA64 is not set +# CONFIG_XZ_DEC_ARM is not set +# CONFIG_XZ_DEC_ARMTHUMB is not set +# CONFIG_XZ_DEC_SPARC is not set +CONFIG_CMM=m diff --git a/arch/s390/hypfs/Makefile b/arch/s390/hypfs/Makefile new file mode 100644 index 000000000..2ee25ba25 --- /dev/null +++ b/arch/s390/hypfs/Makefile @@ -0,0 +1,8 @@ +# +# Makefile for the linux hypfs filesystem routines. +# + +obj-$(CONFIG_S390_HYPFS_FS) += s390_hypfs.o + +s390_hypfs-objs := inode.o hypfs_diag.o hypfs_vm.o hypfs_dbfs.o hypfs_sprp.o +s390_hypfs-objs += hypfs_diag0c.o diff --git a/arch/s390/hypfs/hypfs.h b/arch/s390/hypfs/hypfs.h new file mode 100644 index 000000000..eecde500e --- /dev/null +++ b/arch/s390/hypfs/hypfs.h @@ -0,0 +1,76 @@ +/* + * Hypervisor filesystem for Linux on s390. + * + * Copyright IBM Corp. 2006 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#ifndef _HYPFS_H_ +#define _HYPFS_H_ + +#include <linux/fs.h> +#include <linux/types.h> +#include <linux/debugfs.h> +#include <linux/workqueue.h> +#include <linux/kref.h> +#include <asm/hypfs.h> + +#define REG_FILE_MODE 0440 +#define UPDATE_FILE_MODE 0220 +#define DIR_MODE 0550 + +extern struct dentry *hypfs_mkdir(struct dentry *parent, const char *name); + +extern struct dentry *hypfs_create_u64(struct dentry *dir, const char *name, + __u64 value); + +extern struct dentry *hypfs_create_str(struct dentry *dir, const char *name, + char *string); + +/* LPAR Hypervisor */ +extern int hypfs_diag_init(void); +extern void hypfs_diag_exit(void); +extern int hypfs_diag_create_files(struct dentry *root); + +/* VM Hypervisor */ +extern int hypfs_vm_init(void); +extern void hypfs_vm_exit(void); +extern int hypfs_vm_create_files(struct dentry *root); + +/* VM diagnose 0c */ +int hypfs_diag0c_init(void); +void hypfs_diag0c_exit(void); + +/* Set Partition-Resource Parameter */ +int hypfs_sprp_init(void); +void hypfs_sprp_exit(void); + +/* debugfs interface */ +struct hypfs_dbfs_file; + +struct hypfs_dbfs_data { + void *buf; + void *buf_free_ptr; + size_t size; + struct hypfs_dbfs_file *dbfs_file; +}; + +struct hypfs_dbfs_file { + const char *name; + int (*data_create)(void **data, void **data_free_ptr, + size_t *size); + void (*data_free)(const void *buf_free_ptr); + long (*unlocked_ioctl) (struct file *, unsigned int, + unsigned long); + + /* Private data for hypfs_dbfs.c */ + struct mutex lock; + struct dentry *dentry; +}; + +extern int hypfs_dbfs_init(void); +extern void hypfs_dbfs_exit(void); +extern int hypfs_dbfs_create_file(struct hypfs_dbfs_file *df); +extern void hypfs_dbfs_remove_file(struct hypfs_dbfs_file *df); + +#endif /* _HYPFS_H_ */ diff --git a/arch/s390/hypfs/hypfs_dbfs.c b/arch/s390/hypfs/hypfs_dbfs.c new file mode 100644 index 000000000..752f6df3e --- /dev/null +++ b/arch/s390/hypfs/hypfs_dbfs.c @@ -0,0 +1,104 @@ +/* + * Hypervisor filesystem for Linux on s390 - debugfs interface + * + * Copyright IBM Corp. 2010 + * Author(s): Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#include <linux/slab.h> +#include "hypfs.h" + +static struct dentry *dbfs_dir; + +static struct hypfs_dbfs_data *hypfs_dbfs_data_alloc(struct hypfs_dbfs_file *f) +{ + struct hypfs_dbfs_data *data; + + data = kmalloc(sizeof(*data), GFP_KERNEL); + if (!data) + return NULL; + data->dbfs_file = f; + return data; +} + +static void hypfs_dbfs_data_free(struct hypfs_dbfs_data *data) +{ + data->dbfs_file->data_free(data->buf_free_ptr); + kfree(data); +} + +static ssize_t dbfs_read(struct file *file, char __user *buf, + size_t size, loff_t *ppos) +{ + struct hypfs_dbfs_data *data; + struct hypfs_dbfs_file *df; + ssize_t rc; + + if (*ppos != 0) + return 0; + + df = file_inode(file)->i_private; + mutex_lock(&df->lock); + data = hypfs_dbfs_data_alloc(df); + if (!data) { + mutex_unlock(&df->lock); + return -ENOMEM; + } + rc = df->data_create(&data->buf, &data->buf_free_ptr, &data->size); + if (rc) { + mutex_unlock(&df->lock); + kfree(data); + return rc; + } + mutex_unlock(&df->lock); + + rc = simple_read_from_buffer(buf, size, ppos, data->buf, data->size); + hypfs_dbfs_data_free(data); + return rc; +} + +static long dbfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + struct hypfs_dbfs_file *df = file_inode(file)->i_private; + long rc; + + mutex_lock(&df->lock); + if (df->unlocked_ioctl) + rc = df->unlocked_ioctl(file, cmd, arg); + else + rc = -ENOTTY; + mutex_unlock(&df->lock); + return rc; +} + +static const struct file_operations dbfs_ops = { + .read = dbfs_read, + .llseek = no_llseek, + .unlocked_ioctl = dbfs_ioctl, +}; + +int hypfs_dbfs_create_file(struct hypfs_dbfs_file *df) +{ + df->dentry = debugfs_create_file(df->name, 0400, dbfs_dir, df, + &dbfs_ops); + if (IS_ERR(df->dentry)) + return PTR_ERR(df->dentry); + mutex_init(&df->lock); + return 0; +} + +void hypfs_dbfs_remove_file(struct hypfs_dbfs_file *df) +{ + debugfs_remove(df->dentry); +} + +int hypfs_dbfs_init(void) +{ + dbfs_dir = debugfs_create_dir("s390_hypfs", NULL); + return PTR_ERR_OR_ZERO(dbfs_dir); +} + +void hypfs_dbfs_exit(void) +{ + debugfs_remove(dbfs_dir); +} diff --git a/arch/s390/hypfs/hypfs_diag.c b/arch/s390/hypfs/hypfs_diag.c new file mode 100644 index 000000000..5eeffeefa --- /dev/null +++ b/arch/s390/hypfs/hypfs_diag.c @@ -0,0 +1,769 @@ +/* + * Hypervisor filesystem for Linux on s390. Diag 204 and 224 + * implementation. + * + * Copyright IBM Corp. 2006, 2008 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#define KMSG_COMPONENT "hypfs" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/string.h> +#include <linux/vmalloc.h> +#include <linux/mm.h> +#include <asm/ebcdic.h> +#include "hypfs.h" + +#define LPAR_NAME_LEN 8 /* lpar name len in diag 204 data */ +#define CPU_NAME_LEN 16 /* type name len of cpus in diag224 name table */ +#define TMP_SIZE 64 /* size of temporary buffers */ + +#define DBFS_D204_HDR_VERSION 0 + +/* diag 204 subcodes */ +enum diag204_sc { + SUBC_STIB4 = 4, + SUBC_RSI = 5, + SUBC_STIB6 = 6, + SUBC_STIB7 = 7 +}; + +/* The two available diag 204 data formats */ +enum diag204_format { + INFO_SIMPLE = 0, + INFO_EXT = 0x00010000 +}; + +/* bit is set in flags, when physical cpu info is included in diag 204 data */ +#define LPAR_PHYS_FLG 0x80 + +static char *diag224_cpu_names; /* diag 224 name table */ +static enum diag204_sc diag204_store_sc; /* used subcode for store */ +static enum diag204_format diag204_info_type; /* used diag 204 data format */ + +static void *diag204_buf; /* 4K aligned buffer for diag204 data */ +static void *diag204_buf_vmalloc; /* vmalloc pointer for diag204 data */ +static int diag204_buf_pages; /* number of pages for diag204 data */ + +static struct dentry *dbfs_d204_file; + +/* + * DIAG 204 data structures and member access functions. + * + * Since we have two different diag 204 data formats for old and new s390 + * machines, we do not access the structs directly, but use getter functions for + * each struct member instead. This should make the code more readable. + */ + +/* Time information block */ + +struct info_blk_hdr { + __u8 npar; + __u8 flags; + __u16 tslice; + __u16 phys_cpus; + __u16 this_part; + __u64 curtod; +} __attribute__ ((packed)); + +struct x_info_blk_hdr { + __u8 npar; + __u8 flags; + __u16 tslice; + __u16 phys_cpus; + __u16 this_part; + __u64 curtod1; + __u64 curtod2; + char reserved[40]; +} __attribute__ ((packed)); + +static inline int info_blk_hdr__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct info_blk_hdr); + else /* INFO_EXT */ + return sizeof(struct x_info_blk_hdr); +} + +static inline __u8 info_blk_hdr__npar(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct info_blk_hdr *)hdr)->npar; + else /* INFO_EXT */ + return ((struct x_info_blk_hdr *)hdr)->npar; +} + +static inline __u8 info_blk_hdr__flags(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct info_blk_hdr *)hdr)->flags; + else /* INFO_EXT */ + return ((struct x_info_blk_hdr *)hdr)->flags; +} + +static inline __u16 info_blk_hdr__pcpus(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct info_blk_hdr *)hdr)->phys_cpus; + else /* INFO_EXT */ + return ((struct x_info_blk_hdr *)hdr)->phys_cpus; +} + +/* Partition header */ + +struct part_hdr { + __u8 pn; + __u8 cpus; + char reserved[6]; + char part_name[LPAR_NAME_LEN]; +} __attribute__ ((packed)); + +struct x_part_hdr { + __u8 pn; + __u8 cpus; + __u8 rcpus; + __u8 pflag; + __u32 mlu; + char part_name[LPAR_NAME_LEN]; + char lpc_name[8]; + char os_name[8]; + __u64 online_cs; + __u64 online_es; + __u8 upid; + char reserved1[3]; + __u32 group_mlu; + char group_name[8]; + char reserved2[32]; +} __attribute__ ((packed)); + +static inline int part_hdr__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct part_hdr); + else /* INFO_EXT */ + return sizeof(struct x_part_hdr); +} + +static inline __u8 part_hdr__rcpus(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct part_hdr *)hdr)->cpus; + else /* INFO_EXT */ + return ((struct x_part_hdr *)hdr)->rcpus; +} + +static inline void part_hdr__part_name(enum diag204_format type, void *hdr, + char *name) +{ + if (type == INFO_SIMPLE) + memcpy(name, ((struct part_hdr *)hdr)->part_name, + LPAR_NAME_LEN); + else /* INFO_EXT */ + memcpy(name, ((struct x_part_hdr *)hdr)->part_name, + LPAR_NAME_LEN); + EBCASC(name, LPAR_NAME_LEN); + name[LPAR_NAME_LEN] = 0; + strim(name); +} + +struct cpu_info { + __u16 cpu_addr; + char reserved1[2]; + __u8 ctidx; + __u8 cflag; + __u16 weight; + __u64 acc_time; + __u64 lp_time; +} __attribute__ ((packed)); + +struct x_cpu_info { + __u16 cpu_addr; + char reserved1[2]; + __u8 ctidx; + __u8 cflag; + __u16 weight; + __u64 acc_time; + __u64 lp_time; + __u16 min_weight; + __u16 cur_weight; + __u16 max_weight; + char reseved2[2]; + __u64 online_time; + __u64 wait_time; + __u32 pma_weight; + __u32 polar_weight; + char reserved3[40]; +} __attribute__ ((packed)); + +/* CPU info block */ + +static inline int cpu_info__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct cpu_info); + else /* INFO_EXT */ + return sizeof(struct x_cpu_info); +} + +static inline __u8 cpu_info__ctidx(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct cpu_info *)hdr)->ctidx; + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->ctidx; +} + +static inline __u16 cpu_info__cpu_addr(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct cpu_info *)hdr)->cpu_addr; + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->cpu_addr; +} + +static inline __u64 cpu_info__acc_time(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct cpu_info *)hdr)->acc_time; + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->acc_time; +} + +static inline __u64 cpu_info__lp_time(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct cpu_info *)hdr)->lp_time; + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->lp_time; +} + +static inline __u64 cpu_info__online_time(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return 0; /* online_time not available in simple info */ + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->online_time; +} + +/* Physical header */ + +struct phys_hdr { + char reserved1[1]; + __u8 cpus; + char reserved2[6]; + char mgm_name[8]; +} __attribute__ ((packed)); + +struct x_phys_hdr { + char reserved1[1]; + __u8 cpus; + char reserved2[6]; + char mgm_name[8]; + char reserved3[80]; +} __attribute__ ((packed)); + +static inline int phys_hdr__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct phys_hdr); + else /* INFO_EXT */ + return sizeof(struct x_phys_hdr); +} + +static inline __u8 phys_hdr__cpus(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct phys_hdr *)hdr)->cpus; + else /* INFO_EXT */ + return ((struct x_phys_hdr *)hdr)->cpus; +} + +/* Physical CPU info block */ + +struct phys_cpu { + __u16 cpu_addr; + char reserved1[2]; + __u8 ctidx; + char reserved2[3]; + __u64 mgm_time; + char reserved3[8]; +} __attribute__ ((packed)); + +struct x_phys_cpu { + __u16 cpu_addr; + char reserved1[2]; + __u8 ctidx; + char reserved2[3]; + __u64 mgm_time; + char reserved3[80]; +} __attribute__ ((packed)); + +static inline int phys_cpu__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct phys_cpu); + else /* INFO_EXT */ + return sizeof(struct x_phys_cpu); +} + +static inline __u16 phys_cpu__cpu_addr(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct phys_cpu *)hdr)->cpu_addr; + else /* INFO_EXT */ + return ((struct x_phys_cpu *)hdr)->cpu_addr; +} + +static inline __u64 phys_cpu__mgm_time(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct phys_cpu *)hdr)->mgm_time; + else /* INFO_EXT */ + return ((struct x_phys_cpu *)hdr)->mgm_time; +} + +static inline __u64 phys_cpu__ctidx(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct phys_cpu *)hdr)->ctidx; + else /* INFO_EXT */ + return ((struct x_phys_cpu *)hdr)->ctidx; +} + +/* Diagnose 204 functions */ + +static int diag204(unsigned long subcode, unsigned long size, void *addr) +{ + register unsigned long _subcode asm("0") = subcode; + register unsigned long _size asm("1") = size; + + asm volatile( + " diag %2,%0,0x204\n" + "0:\n" + EX_TABLE(0b,0b) + : "+d" (_subcode), "+d" (_size) : "d" (addr) : "memory"); + if (_subcode) + return -1; + return _size; +} + +/* + * For the old diag subcode 4 with simple data format we have to use real + * memory. If we use subcode 6 or 7 with extended data format, we can (and + * should) use vmalloc, since we need a lot of memory in that case. Currently + * up to 93 pages! + */ + +static void diag204_free_buffer(void) +{ + if (!diag204_buf) + return; + if (diag204_buf_vmalloc) { + vfree(diag204_buf_vmalloc); + diag204_buf_vmalloc = NULL; + } else { + free_pages((unsigned long) diag204_buf, 0); + } + diag204_buf = NULL; +} + +static void *page_align_ptr(void *ptr) +{ + return (void *) PAGE_ALIGN((unsigned long) ptr); +} + +static void *diag204_alloc_vbuf(int pages) +{ + /* The buffer has to be page aligned! */ + diag204_buf_vmalloc = vmalloc(PAGE_SIZE * (pages + 1)); + if (!diag204_buf_vmalloc) + return ERR_PTR(-ENOMEM); + diag204_buf = page_align_ptr(diag204_buf_vmalloc); + diag204_buf_pages = pages; + return diag204_buf; +} + +static void *diag204_alloc_rbuf(void) +{ + diag204_buf = (void*)__get_free_pages(GFP_KERNEL,0); + if (!diag204_buf) + return ERR_PTR(-ENOMEM); + diag204_buf_pages = 1; + return diag204_buf; +} + +static void *diag204_get_buffer(enum diag204_format fmt, int *pages) +{ + if (diag204_buf) { + *pages = diag204_buf_pages; + return diag204_buf; + } + if (fmt == INFO_SIMPLE) { + *pages = 1; + return diag204_alloc_rbuf(); + } else {/* INFO_EXT */ + *pages = diag204((unsigned long)SUBC_RSI | + (unsigned long)INFO_EXT, 0, NULL); + if (*pages <= 0) + return ERR_PTR(-ENOSYS); + else + return diag204_alloc_vbuf(*pages); + } +} + +/* + * diag204_probe() has to find out, which type of diagnose 204 implementation + * we have on our machine. Currently there are three possible scanarios: + * - subcode 4 + simple data format (only one page) + * - subcode 4-6 + extended data format + * - subcode 4-7 + extended data format + * + * Subcode 5 is used to retrieve the size of the data, provided by subcodes + * 6 and 7. Subcode 7 basically has the same function as subcode 6. In addition + * to subcode 6 it provides also information about secondary cpus. + * In order to get as much information as possible, we first try + * subcode 7, then 6 and if both fail, we use subcode 4. + */ + +static int diag204_probe(void) +{ + void *buf; + int pages, rc; + + buf = diag204_get_buffer(INFO_EXT, &pages); + if (!IS_ERR(buf)) { + if (diag204((unsigned long)SUBC_STIB7 | + (unsigned long)INFO_EXT, pages, buf) >= 0) { + diag204_store_sc = SUBC_STIB7; + diag204_info_type = INFO_EXT; + goto out; + } + if (diag204((unsigned long)SUBC_STIB6 | + (unsigned long)INFO_EXT, pages, buf) >= 0) { + diag204_store_sc = SUBC_STIB6; + diag204_info_type = INFO_EXT; + goto out; + } + diag204_free_buffer(); + } + + /* subcodes 6 and 7 failed, now try subcode 4 */ + + buf = diag204_get_buffer(INFO_SIMPLE, &pages); + if (IS_ERR(buf)) { + rc = PTR_ERR(buf); + goto fail_alloc; + } + if (diag204((unsigned long)SUBC_STIB4 | + (unsigned long)INFO_SIMPLE, pages, buf) >= 0) { + diag204_store_sc = SUBC_STIB4; + diag204_info_type = INFO_SIMPLE; + goto out; + } else { + rc = -ENOSYS; + goto fail_store; + } +out: + rc = 0; +fail_store: + diag204_free_buffer(); +fail_alloc: + return rc; +} + +static int diag204_do_store(void *buf, int pages) +{ + int rc; + + rc = diag204((unsigned long) diag204_store_sc | + (unsigned long) diag204_info_type, pages, buf); + return rc < 0 ? -ENOSYS : 0; +} + +static void *diag204_store(void) +{ + void *buf; + int pages, rc; + + buf = diag204_get_buffer(diag204_info_type, &pages); + if (IS_ERR(buf)) + goto out; + rc = diag204_do_store(buf, pages); + if (rc) + return ERR_PTR(rc); +out: + return buf; +} + +/* Diagnose 224 functions */ + +static int diag224(void *ptr) +{ + int rc = -EOPNOTSUPP; + + asm volatile( + " diag %1,%2,0x224\n" + "0: lhi %0,0x0\n" + "1:\n" + EX_TABLE(0b,1b) + : "+d" (rc) :"d" (0), "d" (ptr) : "memory"); + return rc; +} + +static int diag224_get_name_table(void) +{ + /* memory must be below 2GB */ + diag224_cpu_names = kmalloc(PAGE_SIZE, GFP_KERNEL | GFP_DMA); + if (!diag224_cpu_names) + return -ENOMEM; + if (diag224(diag224_cpu_names)) { + kfree(diag224_cpu_names); + return -EOPNOTSUPP; + } + EBCASC(diag224_cpu_names + 16, (*diag224_cpu_names + 1) * 16); + return 0; +} + +static void diag224_delete_name_table(void) +{ + kfree(diag224_cpu_names); +} + +static int diag224_idx2name(int index, char *name) +{ + memcpy(name, diag224_cpu_names + ((index + 1) * CPU_NAME_LEN), + CPU_NAME_LEN); + name[CPU_NAME_LEN] = 0; + strim(name); + return 0; +} + +struct dbfs_d204_hdr { + u64 len; /* Length of d204 buffer without header */ + u16 version; /* Version of header */ + u8 sc; /* Used subcode */ + char reserved[53]; +} __attribute__ ((packed)); + +struct dbfs_d204 { + struct dbfs_d204_hdr hdr; /* 64 byte header */ + char buf[]; /* d204 buffer */ +} __attribute__ ((packed)); + +static int dbfs_d204_create(void **data, void **data_free_ptr, size_t *size) +{ + struct dbfs_d204 *d204; + int rc, buf_size; + void *base; + + buf_size = PAGE_SIZE * (diag204_buf_pages + 1) + sizeof(d204->hdr); + base = vzalloc(buf_size); + if (!base) + return -ENOMEM; + d204 = page_align_ptr(base + sizeof(d204->hdr)) - sizeof(d204->hdr); + rc = diag204_do_store(d204->buf, diag204_buf_pages); + if (rc) { + vfree(base); + return rc; + } + d204->hdr.version = DBFS_D204_HDR_VERSION; + d204->hdr.len = PAGE_SIZE * diag204_buf_pages; + d204->hdr.sc = diag204_store_sc; + *data = d204; + *data_free_ptr = base; + *size = d204->hdr.len + sizeof(struct dbfs_d204_hdr); + return 0; +} + +static struct hypfs_dbfs_file dbfs_file_d204 = { + .name = "diag_204", + .data_create = dbfs_d204_create, + .data_free = vfree, +}; + +__init int hypfs_diag_init(void) +{ + int rc; + + if (diag204_probe()) { + pr_err("The hardware system does not support hypfs\n"); + return -ENODATA; + } + if (diag204_info_type == INFO_EXT) { + rc = hypfs_dbfs_create_file(&dbfs_file_d204); + if (rc) + return rc; + } + if (MACHINE_IS_LPAR) { + rc = diag224_get_name_table(); + if (rc) { + pr_err("The hardware system does not provide all " + "functions required by hypfs\n"); + debugfs_remove(dbfs_d204_file); + return rc; + } + } + return 0; +} + +void hypfs_diag_exit(void) +{ + debugfs_remove(dbfs_d204_file); + diag224_delete_name_table(); + diag204_free_buffer(); + hypfs_dbfs_remove_file(&dbfs_file_d204); +} + +/* + * Functions to create the directory structure + * ******************************************* + */ + +static int hypfs_create_cpu_files(struct dentry *cpus_dir, void *cpu_info) +{ + struct dentry *cpu_dir; + char buffer[TMP_SIZE]; + void *rc; + + snprintf(buffer, TMP_SIZE, "%d", cpu_info__cpu_addr(diag204_info_type, + cpu_info)); + cpu_dir = hypfs_mkdir(cpus_dir, buffer); + rc = hypfs_create_u64(cpu_dir, "mgmtime", + cpu_info__acc_time(diag204_info_type, cpu_info) - + cpu_info__lp_time(diag204_info_type, cpu_info)); + if (IS_ERR(rc)) + return PTR_ERR(rc); + rc = hypfs_create_u64(cpu_dir, "cputime", + cpu_info__lp_time(diag204_info_type, cpu_info)); + if (IS_ERR(rc)) + return PTR_ERR(rc); + if (diag204_info_type == INFO_EXT) { + rc = hypfs_create_u64(cpu_dir, "onlinetime", + cpu_info__online_time(diag204_info_type, + cpu_info)); + if (IS_ERR(rc)) + return PTR_ERR(rc); + } + diag224_idx2name(cpu_info__ctidx(diag204_info_type, cpu_info), buffer); + rc = hypfs_create_str(cpu_dir, "type", buffer); + return PTR_RET(rc); +} + +static void *hypfs_create_lpar_files(struct dentry *systems_dir, void *part_hdr) +{ + struct dentry *cpus_dir; + struct dentry *lpar_dir; + char lpar_name[LPAR_NAME_LEN + 1]; + void *cpu_info; + int i; + + part_hdr__part_name(diag204_info_type, part_hdr, lpar_name); + lpar_name[LPAR_NAME_LEN] = 0; + lpar_dir = hypfs_mkdir(systems_dir, lpar_name); + if (IS_ERR(lpar_dir)) + return lpar_dir; + cpus_dir = hypfs_mkdir(lpar_dir, "cpus"); + if (IS_ERR(cpus_dir)) + return cpus_dir; + cpu_info = part_hdr + part_hdr__size(diag204_info_type); + for (i = 0; i < part_hdr__rcpus(diag204_info_type, part_hdr); i++) { + int rc; + rc = hypfs_create_cpu_files(cpus_dir, cpu_info); + if (rc) + return ERR_PTR(rc); + cpu_info += cpu_info__size(diag204_info_type); + } + return cpu_info; +} + +static int hypfs_create_phys_cpu_files(struct dentry *cpus_dir, void *cpu_info) +{ + struct dentry *cpu_dir; + char buffer[TMP_SIZE]; + void *rc; + + snprintf(buffer, TMP_SIZE, "%i", phys_cpu__cpu_addr(diag204_info_type, + cpu_info)); + cpu_dir = hypfs_mkdir(cpus_dir, buffer); + if (IS_ERR(cpu_dir)) + return PTR_ERR(cpu_dir); + rc = hypfs_create_u64(cpu_dir, "mgmtime", + phys_cpu__mgm_time(diag204_info_type, cpu_info)); + if (IS_ERR(rc)) + return PTR_ERR(rc); + diag224_idx2name(phys_cpu__ctidx(diag204_info_type, cpu_info), buffer); + rc = hypfs_create_str(cpu_dir, "type", buffer); + return PTR_RET(rc); +} + +static void *hypfs_create_phys_files(struct dentry *parent_dir, void *phys_hdr) +{ + int i; + void *cpu_info; + struct dentry *cpus_dir; + + cpus_dir = hypfs_mkdir(parent_dir, "cpus"); + if (IS_ERR(cpus_dir)) + return cpus_dir; + cpu_info = phys_hdr + phys_hdr__size(diag204_info_type); + for (i = 0; i < phys_hdr__cpus(diag204_info_type, phys_hdr); i++) { + int rc; + rc = hypfs_create_phys_cpu_files(cpus_dir, cpu_info); + if (rc) + return ERR_PTR(rc); + cpu_info += phys_cpu__size(diag204_info_type); + } + return cpu_info; +} + +int hypfs_diag_create_files(struct dentry *root) +{ + struct dentry *systems_dir, *hyp_dir; + void *time_hdr, *part_hdr; + int i, rc; + void *buffer, *ptr; + + buffer = diag204_store(); + if (IS_ERR(buffer)) + return PTR_ERR(buffer); + + systems_dir = hypfs_mkdir(root, "systems"); + if (IS_ERR(systems_dir)) { + rc = PTR_ERR(systems_dir); + goto err_out; + } + time_hdr = (struct x_info_blk_hdr *)buffer; + part_hdr = time_hdr + info_blk_hdr__size(diag204_info_type); + for (i = 0; i < info_blk_hdr__npar(diag204_info_type, time_hdr); i++) { + part_hdr = hypfs_create_lpar_files(systems_dir, part_hdr); + if (IS_ERR(part_hdr)) { + rc = PTR_ERR(part_hdr); + goto err_out; + } + } + if (info_blk_hdr__flags(diag204_info_type, time_hdr) & LPAR_PHYS_FLG) { + ptr = hypfs_create_phys_files(root, part_hdr); + if (IS_ERR(ptr)) { + rc = PTR_ERR(ptr); + goto err_out; + } + } + hyp_dir = hypfs_mkdir(root, "hyp"); + if (IS_ERR(hyp_dir)) { + rc = PTR_ERR(hyp_dir); + goto err_out; + } + ptr = hypfs_create_str(hyp_dir, "type", "LPAR Hypervisor"); + if (IS_ERR(ptr)) { + rc = PTR_ERR(ptr); + goto err_out; + } + rc = 0; + +err_out: + return rc; +} diff --git a/arch/s390/hypfs/hypfs_diag0c.c b/arch/s390/hypfs/hypfs_diag0c.c new file mode 100644 index 000000000..24c747a0f --- /dev/null +++ b/arch/s390/hypfs/hypfs_diag0c.c @@ -0,0 +1,135 @@ +/* + * Hypervisor filesystem for Linux on s390 + * + * Diag 0C implementation + * + * Copyright IBM Corp. 2014 + */ + +#include <linux/slab.h> +#include <linux/cpu.h> +#include <asm/hypfs.h> +#include "hypfs.h" + +#define DBFS_D0C_HDR_VERSION 0 + +/* + * Execute diagnose 0c in 31 bit mode + */ +static void diag0c(struct hypfs_diag0c_entry *entry) +{ + asm volatile ( + " sam31\n" + " diag %0,%0,0x0c\n" + " sam64\n" + : /* no output register */ + : "a" (entry) + : "memory"); +} + +/* + * Get hypfs_diag0c_entry from CPU vector and store diag0c data + */ +static void diag0c_fn(void *data) +{ + diag0c(((void **) data)[smp_processor_id()]); +} + +/* + * Allocate buffer and store diag 0c data + */ +static void *diag0c_store(unsigned int *count) +{ + struct hypfs_diag0c_data *diag0c_data; + unsigned int cpu_count, cpu, i; + void **cpu_vec; + + get_online_cpus(); + cpu_count = num_online_cpus(); + cpu_vec = kmalloc(sizeof(*cpu_vec) * num_possible_cpus(), GFP_KERNEL); + if (!cpu_vec) + goto fail_put_online_cpus; + /* Note: Diag 0c needs 8 byte alignment and real storage */ + diag0c_data = kzalloc(sizeof(struct hypfs_diag0c_hdr) + + cpu_count * sizeof(struct hypfs_diag0c_entry), + GFP_KERNEL | GFP_DMA); + if (!diag0c_data) + goto fail_kfree_cpu_vec; + i = 0; + /* Fill CPU vector for each online CPU */ + for_each_online_cpu(cpu) { + diag0c_data->entry[i].cpu = cpu; + cpu_vec[cpu] = &diag0c_data->entry[i++]; + } + /* Collect data all CPUs */ + on_each_cpu(diag0c_fn, cpu_vec, 1); + *count = cpu_count; + kfree(cpu_vec); + put_online_cpus(); + return diag0c_data; + +fail_kfree_cpu_vec: + kfree(cpu_vec); +fail_put_online_cpus: + put_online_cpus(); + return ERR_PTR(-ENOMEM); +} + +/* + * Hypfs DBFS callback: Free diag 0c data + */ +static void dbfs_diag0c_free(const void *data) +{ + kfree(data); +} + +/* + * Hypfs DBFS callback: Create diag 0c data + */ +static int dbfs_diag0c_create(void **data, void **data_free_ptr, size_t *size) +{ + struct hypfs_diag0c_data *diag0c_data; + unsigned int count; + + diag0c_data = diag0c_store(&count); + if (IS_ERR(diag0c_data)) + return PTR_ERR(diag0c_data); + memset(&diag0c_data->hdr, 0, sizeof(diag0c_data->hdr)); + get_tod_clock_ext(diag0c_data->hdr.tod_ext); + diag0c_data->hdr.len = count * sizeof(struct hypfs_diag0c_entry); + diag0c_data->hdr.version = DBFS_D0C_HDR_VERSION; + diag0c_data->hdr.count = count; + *data = diag0c_data; + *data_free_ptr = diag0c_data; + *size = diag0c_data->hdr.len + sizeof(struct hypfs_diag0c_hdr); + return 0; +} + +/* + * Hypfs DBFS file structure + */ +static struct hypfs_dbfs_file dbfs_file_0c = { + .name = "diag_0c", + .data_create = dbfs_diag0c_create, + .data_free = dbfs_diag0c_free, +}; + +/* + * Initialize diag 0c interface for z/VM + */ +int __init hypfs_diag0c_init(void) +{ + if (!MACHINE_IS_VM) + return 0; + return hypfs_dbfs_create_file(&dbfs_file_0c); +} + +/* + * Shutdown diag 0c interface for z/VM + */ +void hypfs_diag0c_exit(void) +{ + if (!MACHINE_IS_VM) + return; + hypfs_dbfs_remove_file(&dbfs_file_0c); +} diff --git a/arch/s390/hypfs/hypfs_sprp.c b/arch/s390/hypfs/hypfs_sprp.c new file mode 100644 index 000000000..f043c3c7e --- /dev/null +++ b/arch/s390/hypfs/hypfs_sprp.c @@ -0,0 +1,141 @@ +/* + * Hypervisor filesystem for Linux on s390. + * Set Partition-Resource Parameter interface. + * + * Copyright IBM Corp. 2013 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#include <linux/compat.h> +#include <linux/errno.h> +#include <linux/gfp.h> +#include <linux/string.h> +#include <linux/types.h> +#include <linux/uaccess.h> +#include <asm/compat.h> +#include <asm/sclp.h> +#include "hypfs.h" + +#define DIAG304_SET_WEIGHTS 0 +#define DIAG304_QUERY_PRP 1 +#define DIAG304_SET_CAPPING 2 + +#define DIAG304_CMD_MAX 2 + +static unsigned long hypfs_sprp_diag304(void *data, unsigned long cmd) +{ + register unsigned long _data asm("2") = (unsigned long) data; + register unsigned long _rc asm("3"); + register unsigned long _cmd asm("4") = cmd; + + asm volatile("diag %1,%2,0x304\n" + : "=d" (_rc) : "d" (_data), "d" (_cmd) : "memory"); + + return _rc; +} + +static void hypfs_sprp_free(const void *data) +{ + free_page((unsigned long) data); +} + +static int hypfs_sprp_create(void **data_ptr, void **free_ptr, size_t *size) +{ + unsigned long rc; + void *data; + + data = (void *) get_zeroed_page(GFP_KERNEL); + if (!data) + return -ENOMEM; + rc = hypfs_sprp_diag304(data, DIAG304_QUERY_PRP); + if (rc != 1) { + *data_ptr = *free_ptr = NULL; + *size = 0; + free_page((unsigned long) data); + return -EIO; + } + *data_ptr = *free_ptr = data; + *size = PAGE_SIZE; + return 0; +} + +static int __hypfs_sprp_ioctl(void __user *user_area) +{ + struct hypfs_diag304 diag304; + unsigned long cmd; + void __user *udata; + void *data; + int rc; + + if (copy_from_user(&diag304, user_area, sizeof(diag304))) + return -EFAULT; + if ((diag304.args[0] >> 8) != 0 || diag304.args[1] > DIAG304_CMD_MAX) + return -EINVAL; + + data = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA); + if (!data) + return -ENOMEM; + + udata = (void __user *)(unsigned long) diag304.data; + if (diag304.args[1] == DIAG304_SET_WEIGHTS || + diag304.args[1] == DIAG304_SET_CAPPING) + if (copy_from_user(data, udata, PAGE_SIZE)) { + rc = -EFAULT; + goto out; + } + + cmd = *(unsigned long *) &diag304.args[0]; + diag304.rc = hypfs_sprp_diag304(data, cmd); + + if (diag304.args[1] == DIAG304_QUERY_PRP) + if (copy_to_user(udata, data, PAGE_SIZE)) { + rc = -EFAULT; + goto out; + } + + rc = copy_to_user(user_area, &diag304, sizeof(diag304)) ? -EFAULT : 0; +out: + free_page((unsigned long) data); + return rc; +} + +static long hypfs_sprp_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + void __user *argp; + + if (!capable(CAP_SYS_ADMIN)) + return -EACCES; + if (is_compat_task()) + argp = compat_ptr(arg); + else + argp = (void __user *) arg; + switch (cmd) { + case HYPFS_DIAG304: + return __hypfs_sprp_ioctl(argp); + default: /* unknown ioctl number */ + return -ENOTTY; + } + return 0; +} + +static struct hypfs_dbfs_file hypfs_sprp_file = { + .name = "diag_304", + .data_create = hypfs_sprp_create, + .data_free = hypfs_sprp_free, + .unlocked_ioctl = hypfs_sprp_ioctl, +}; + +int hypfs_sprp_init(void) +{ + if (!sclp_has_sprp()) + return 0; + return hypfs_dbfs_create_file(&hypfs_sprp_file); +} + +void hypfs_sprp_exit(void) +{ + if (!sclp_has_sprp()) + return; + hypfs_dbfs_remove_file(&hypfs_sprp_file); +} diff --git a/arch/s390/hypfs/hypfs_vm.c b/arch/s390/hypfs/hypfs_vm.c new file mode 100644 index 000000000..afbe07907 --- /dev/null +++ b/arch/s390/hypfs/hypfs_vm.c @@ -0,0 +1,287 @@ +/* + * Hypervisor filesystem for Linux on s390. z/VM implementation. + * + * Copyright IBM Corp. 2006 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/string.h> +#include <linux/vmalloc.h> +#include <asm/ebcdic.h> +#include <asm/timex.h> +#include "hypfs.h" + +#define NAME_LEN 8 +#define DBFS_D2FC_HDR_VERSION 0 + +static char local_guest[] = " "; +static char all_guests[] = "* "; +static char *guest_query; + +struct diag2fc_data { + __u32 version; + __u32 flags; + __u64 used_cpu; + __u64 el_time; + __u64 mem_min_kb; + __u64 mem_max_kb; + __u64 mem_share_kb; + __u64 mem_used_kb; + __u32 pcpus; + __u32 lcpus; + __u32 vcpus; + __u32 ocpus; + __u32 cpu_max; + __u32 cpu_shares; + __u32 cpu_use_samp; + __u32 cpu_delay_samp; + __u32 page_wait_samp; + __u32 idle_samp; + __u32 other_samp; + __u32 total_samp; + char guest_name[NAME_LEN]; +}; + +struct diag2fc_parm_list { + char userid[NAME_LEN]; + char aci_grp[NAME_LEN]; + __u64 addr; + __u32 size; + __u32 fmt; +}; + +static int diag2fc(int size, char* query, void *addr) +{ + unsigned long residual_cnt; + unsigned long rc; + struct diag2fc_parm_list parm_list; + + memcpy(parm_list.userid, query, NAME_LEN); + ASCEBC(parm_list.userid, NAME_LEN); + parm_list.addr = (unsigned long) addr ; + parm_list.size = size; + parm_list.fmt = 0x02; + memset(parm_list.aci_grp, 0x40, NAME_LEN); + rc = -1; + + asm volatile( + " diag %0,%1,0x2fc\n" + "0:\n" + EX_TABLE(0b,0b) + : "=d" (residual_cnt), "+d" (rc) : "0" (&parm_list) : "memory"); + + if ((rc != 0 ) && (rc != -2)) + return rc; + else + return -residual_cnt; +} + +/* + * Allocate buffer for "query" and store diag 2fc at "offset" + */ +static void *diag2fc_store(char *query, unsigned int *count, int offset) +{ + void *data; + int size; + + do { + size = diag2fc(0, query, NULL); + if (size < 0) + return ERR_PTR(-EACCES); + data = vmalloc(size + offset); + if (!data) + return ERR_PTR(-ENOMEM); + if (diag2fc(size, query, data + offset) == 0) + break; + vfree(data); + } while (1); + *count = (size / sizeof(struct diag2fc_data)); + + return data; +} + +static void diag2fc_free(const void *data) +{ + vfree(data); +} + +#define ATTRIBUTE(dir, name, member) \ +do { \ + void *rc; \ + rc = hypfs_create_u64(dir, name, member); \ + if (IS_ERR(rc)) \ + return PTR_ERR(rc); \ +} while(0) + +static int hpyfs_vm_create_guest(struct dentry *systems_dir, + struct diag2fc_data *data) +{ + char guest_name[NAME_LEN + 1] = {}; + struct dentry *guest_dir, *cpus_dir, *samples_dir, *mem_dir; + int dedicated_flag, capped_value; + + capped_value = (data->flags & 0x00000006) >> 1; + dedicated_flag = (data->flags & 0x00000008) >> 3; + + /* guest dir */ + memcpy(guest_name, data->guest_name, NAME_LEN); + EBCASC(guest_name, NAME_LEN); + strim(guest_name); + guest_dir = hypfs_mkdir(systems_dir, guest_name); + if (IS_ERR(guest_dir)) + return PTR_ERR(guest_dir); + ATTRIBUTE(guest_dir, "onlinetime_us", data->el_time); + + /* logical cpu information */ + cpus_dir = hypfs_mkdir(guest_dir, "cpus"); + if (IS_ERR(cpus_dir)) + return PTR_ERR(cpus_dir); + ATTRIBUTE(cpus_dir, "cputime_us", data->used_cpu); + ATTRIBUTE(cpus_dir, "capped", capped_value); + ATTRIBUTE(cpus_dir, "dedicated", dedicated_flag); + ATTRIBUTE(cpus_dir, "count", data->vcpus); + /* + * Note: The "weight_min" attribute got the wrong name. + * The value represents the number of non-stopped (operating) + * CPUS. + */ + ATTRIBUTE(cpus_dir, "weight_min", data->ocpus); + ATTRIBUTE(cpus_dir, "weight_max", data->cpu_max); + ATTRIBUTE(cpus_dir, "weight_cur", data->cpu_shares); + + /* memory information */ + mem_dir = hypfs_mkdir(guest_dir, "mem"); + if (IS_ERR(mem_dir)) + return PTR_ERR(mem_dir); + ATTRIBUTE(mem_dir, "min_KiB", data->mem_min_kb); + ATTRIBUTE(mem_dir, "max_KiB", data->mem_max_kb); + ATTRIBUTE(mem_dir, "used_KiB", data->mem_used_kb); + ATTRIBUTE(mem_dir, "share_KiB", data->mem_share_kb); + + /* samples */ + samples_dir = hypfs_mkdir(guest_dir, "samples"); + if (IS_ERR(samples_dir)) + return PTR_ERR(samples_dir); + ATTRIBUTE(samples_dir, "cpu_using", data->cpu_use_samp); + ATTRIBUTE(samples_dir, "cpu_delay", data->cpu_delay_samp); + ATTRIBUTE(samples_dir, "mem_delay", data->page_wait_samp); + ATTRIBUTE(samples_dir, "idle", data->idle_samp); + ATTRIBUTE(samples_dir, "other", data->other_samp); + ATTRIBUTE(samples_dir, "total", data->total_samp); + return 0; +} + +int hypfs_vm_create_files(struct dentry *root) +{ + struct dentry *dir, *file; + struct diag2fc_data *data; + unsigned int count = 0; + int rc, i; + + data = diag2fc_store(guest_query, &count, 0); + if (IS_ERR(data)) + return PTR_ERR(data); + + /* Hpervisor Info */ + dir = hypfs_mkdir(root, "hyp"); + if (IS_ERR(dir)) { + rc = PTR_ERR(dir); + goto failed; + } + file = hypfs_create_str(dir, "type", "z/VM Hypervisor"); + if (IS_ERR(file)) { + rc = PTR_ERR(file); + goto failed; + } + + /* physical cpus */ + dir = hypfs_mkdir(root, "cpus"); + if (IS_ERR(dir)) { + rc = PTR_ERR(dir); + goto failed; + } + file = hypfs_create_u64(dir, "count", data->lcpus); + if (IS_ERR(file)) { + rc = PTR_ERR(file); + goto failed; + } + + /* guests */ + dir = hypfs_mkdir(root, "systems"); + if (IS_ERR(dir)) { + rc = PTR_ERR(dir); + goto failed; + } + + for (i = 0; i < count; i++) { + rc = hpyfs_vm_create_guest(dir, &(data[i])); + if (rc) + goto failed; + } + diag2fc_free(data); + return 0; + +failed: + diag2fc_free(data); + return rc; +} + +struct dbfs_d2fc_hdr { + u64 len; /* Length of d2fc buffer without header */ + u16 version; /* Version of header */ + char tod_ext[STORE_CLOCK_EXT_SIZE]; /* TOD clock for d2fc */ + u64 count; /* Number of VM guests in d2fc buffer */ + char reserved[30]; +} __attribute__ ((packed)); + +struct dbfs_d2fc { + struct dbfs_d2fc_hdr hdr; /* 64 byte header */ + char buf[]; /* d2fc buffer */ +} __attribute__ ((packed)); + +static int dbfs_diag2fc_create(void **data, void **data_free_ptr, size_t *size) +{ + struct dbfs_d2fc *d2fc; + unsigned int count; + + d2fc = diag2fc_store(guest_query, &count, sizeof(d2fc->hdr)); + if (IS_ERR(d2fc)) + return PTR_ERR(d2fc); + get_tod_clock_ext(d2fc->hdr.tod_ext); + d2fc->hdr.len = count * sizeof(struct diag2fc_data); + d2fc->hdr.version = DBFS_D2FC_HDR_VERSION; + d2fc->hdr.count = count; + memset(&d2fc->hdr.reserved, 0, sizeof(d2fc->hdr.reserved)); + *data = d2fc; + *data_free_ptr = d2fc; + *size = d2fc->hdr.len + sizeof(struct dbfs_d2fc_hdr); + return 0; +} + +static struct hypfs_dbfs_file dbfs_file_2fc = { + .name = "diag_2fc", + .data_create = dbfs_diag2fc_create, + .data_free = diag2fc_free, +}; + +int hypfs_vm_init(void) +{ + if (!MACHINE_IS_VM) + return 0; + if (diag2fc(0, all_guests, NULL) > 0) + guest_query = all_guests; + else if (diag2fc(0, local_guest, NULL) > 0) + guest_query = local_guest; + else + return -EACCES; + return hypfs_dbfs_create_file(&dbfs_file_2fc); +} + +void hypfs_vm_exit(void) +{ + if (!MACHINE_IS_VM) + return; + hypfs_dbfs_remove_file(&dbfs_file_2fc); +} diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c new file mode 100644 index 000000000..2eeb0a0f5 --- /dev/null +++ b/arch/s390/hypfs/inode.c @@ -0,0 +1,522 @@ +/* + * Hypervisor filesystem for Linux on s390. + * + * Copyright IBM Corp. 2006, 2008 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#define KMSG_COMPONENT "hypfs" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/vfs.h> +#include <linux/slab.h> +#include <linux/pagemap.h> +#include <linux/time.h> +#include <linux/parser.h> +#include <linux/sysfs.h> +#include <linux/module.h> +#include <linux/seq_file.h> +#include <linux/mount.h> +#include <linux/uio.h> +#include <asm/ebcdic.h> +#include "hypfs.h" + +#define HYPFS_MAGIC 0x687970 /* ASCII 'hyp' */ +#define TMP_SIZE 64 /* size of temporary buffers */ + +static struct dentry *hypfs_create_update_file(struct dentry *dir); + +struct hypfs_sb_info { + kuid_t uid; /* uid used for files and dirs */ + kgid_t gid; /* gid used for files and dirs */ + struct dentry *update_file; /* file to trigger update */ + time_t last_update; /* last update time in secs since 1970 */ + struct mutex lock; /* lock to protect update process */ +}; + +static const struct file_operations hypfs_file_ops; +static struct file_system_type hypfs_type; +static const struct super_operations hypfs_s_ops; + +/* start of list of all dentries, which have to be deleted on update */ +static struct dentry *hypfs_last_dentry; + +static void hypfs_update_update(struct super_block *sb) +{ + struct hypfs_sb_info *sb_info = sb->s_fs_info; + struct inode *inode = d_inode(sb_info->update_file); + + sb_info->last_update = get_seconds(); + inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; +} + +/* directory tree removal functions */ + +static void hypfs_add_dentry(struct dentry *dentry) +{ + dentry->d_fsdata = hypfs_last_dentry; + hypfs_last_dentry = dentry; +} + +static inline int hypfs_positive(struct dentry *dentry) +{ + return d_really_is_positive(dentry) && !d_unhashed(dentry); +} + +static void hypfs_remove(struct dentry *dentry) +{ + struct dentry *parent; + + parent = dentry->d_parent; + mutex_lock(&d_inode(parent)->i_mutex); + if (hypfs_positive(dentry)) { + if (d_is_dir(dentry)) + simple_rmdir(d_inode(parent), dentry); + else + simple_unlink(d_inode(parent), dentry); + } + d_delete(dentry); + dput(dentry); + mutex_unlock(&d_inode(parent)->i_mutex); +} + +static void hypfs_delete_tree(struct dentry *root) +{ + while (hypfs_last_dentry) { + struct dentry *next_dentry; + next_dentry = hypfs_last_dentry->d_fsdata; + hypfs_remove(hypfs_last_dentry); + hypfs_last_dentry = next_dentry; + } +} + +static struct inode *hypfs_make_inode(struct super_block *sb, umode_t mode) +{ + struct inode *ret = new_inode(sb); + + if (ret) { + struct hypfs_sb_info *hypfs_info = sb->s_fs_info; + ret->i_ino = get_next_ino(); + ret->i_mode = mode; + ret->i_uid = hypfs_info->uid; + ret->i_gid = hypfs_info->gid; + ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME; + if (S_ISDIR(mode)) + set_nlink(ret, 2); + } + return ret; +} + +static void hypfs_evict_inode(struct inode *inode) +{ + clear_inode(inode); + kfree(inode->i_private); +} + +static int hypfs_open(struct inode *inode, struct file *filp) +{ + char *data = file_inode(filp)->i_private; + struct hypfs_sb_info *fs_info; + + if (filp->f_mode & FMODE_WRITE) { + if (!(inode->i_mode & S_IWUGO)) + return -EACCES; + } + if (filp->f_mode & FMODE_READ) { + if (!(inode->i_mode & S_IRUGO)) + return -EACCES; + } + + fs_info = inode->i_sb->s_fs_info; + if(data) { + mutex_lock(&fs_info->lock); + filp->private_data = kstrdup(data, GFP_KERNEL); + if (!filp->private_data) { + mutex_unlock(&fs_info->lock); + return -ENOMEM; + } + mutex_unlock(&fs_info->lock); + } + return nonseekable_open(inode, filp); +} + +static ssize_t hypfs_read_iter(struct kiocb *iocb, struct iov_iter *to) +{ + struct file *file = iocb->ki_filp; + char *data = file->private_data; + size_t available = strlen(data); + loff_t pos = iocb->ki_pos; + size_t count; + + if (pos < 0) + return -EINVAL; + if (pos >= available || !iov_iter_count(to)) + return 0; + count = copy_to_iter(data + pos, available - pos, to); + if (!count) + return -EFAULT; + iocb->ki_pos = pos + count; + file_accessed(file); + return count; +} + +static ssize_t hypfs_write_iter(struct kiocb *iocb, struct iov_iter *from) +{ + int rc; + struct super_block *sb = file_inode(iocb->ki_filp)->i_sb; + struct hypfs_sb_info *fs_info = sb->s_fs_info; + size_t count = iov_iter_count(from); + + /* + * Currently we only allow one update per second for two reasons: + * 1. diag 204 is VERY expensive + * 2. If several processes do updates in parallel and then read the + * hypfs data, the likelihood of collisions is reduced, if we restrict + * the minimum update interval. A collision occurs, if during the + * data gathering of one process another process triggers an update + * If the first process wants to ensure consistent data, it has + * to restart data collection in this case. + */ + mutex_lock(&fs_info->lock); + if (fs_info->last_update == get_seconds()) { + rc = -EBUSY; + goto out; + } + hypfs_delete_tree(sb->s_root); + if (MACHINE_IS_VM) + rc = hypfs_vm_create_files(sb->s_root); + else + rc = hypfs_diag_create_files(sb->s_root); + if (rc) { + pr_err("Updating the hypfs tree failed\n"); + hypfs_delete_tree(sb->s_root); + goto out; + } + hypfs_update_update(sb); + rc = count; + iov_iter_advance(from, count); +out: + mutex_unlock(&fs_info->lock); + return rc; +} + +static int hypfs_release(struct inode *inode, struct file *filp) +{ + kfree(filp->private_data); + return 0; +} + +enum { opt_uid, opt_gid, opt_err }; + +static const match_table_t hypfs_tokens = { + {opt_uid, "uid=%u"}, + {opt_gid, "gid=%u"}, + {opt_err, NULL} +}; + +static int hypfs_parse_options(char *options, struct super_block *sb) +{ + char *str; + substring_t args[MAX_OPT_ARGS]; + kuid_t uid; + kgid_t gid; + + if (!options) + return 0; + while ((str = strsep(&options, ",")) != NULL) { + int token, option; + struct hypfs_sb_info *hypfs_info = sb->s_fs_info; + + if (!*str) + continue; + token = match_token(str, hypfs_tokens, args); + switch (token) { + case opt_uid: + if (match_int(&args[0], &option)) + return -EINVAL; + uid = make_kuid(current_user_ns(), option); + if (!uid_valid(uid)) + return -EINVAL; + hypfs_info->uid = uid; + break; + case opt_gid: + if (match_int(&args[0], &option)) + return -EINVAL; + gid = make_kgid(current_user_ns(), option); + if (!gid_valid(gid)) + return -EINVAL; + hypfs_info->gid = gid; + break; + case opt_err: + default: + pr_err("%s is not a valid mount option\n", str); + return -EINVAL; + } + } + return 0; +} + +static int hypfs_show_options(struct seq_file *s, struct dentry *root) +{ + struct hypfs_sb_info *hypfs_info = root->d_sb->s_fs_info; + + seq_printf(s, ",uid=%u", from_kuid_munged(&init_user_ns, hypfs_info->uid)); + seq_printf(s, ",gid=%u", from_kgid_munged(&init_user_ns, hypfs_info->gid)); + return 0; +} + +static int hypfs_fill_super(struct super_block *sb, void *data, int silent) +{ + struct inode *root_inode; + struct dentry *root_dentry; + int rc = 0; + struct hypfs_sb_info *sbi; + + sbi = kzalloc(sizeof(struct hypfs_sb_info), GFP_KERNEL); + if (!sbi) + return -ENOMEM; + mutex_init(&sbi->lock); + sbi->uid = current_uid(); + sbi->gid = current_gid(); + sb->s_fs_info = sbi; + sb->s_blocksize = PAGE_CACHE_SIZE; + sb->s_blocksize_bits = PAGE_CACHE_SHIFT; + sb->s_magic = HYPFS_MAGIC; + sb->s_op = &hypfs_s_ops; + if (hypfs_parse_options(data, sb)) + return -EINVAL; + root_inode = hypfs_make_inode(sb, S_IFDIR | 0755); + if (!root_inode) + return -ENOMEM; + root_inode->i_op = &simple_dir_inode_operations; + root_inode->i_fop = &simple_dir_operations; + sb->s_root = root_dentry = d_make_root(root_inode); + if (!root_dentry) + return -ENOMEM; + if (MACHINE_IS_VM) + rc = hypfs_vm_create_files(root_dentry); + else + rc = hypfs_diag_create_files(root_dentry); + if (rc) + return rc; + sbi->update_file = hypfs_create_update_file(root_dentry); + if (IS_ERR(sbi->update_file)) + return PTR_ERR(sbi->update_file); + hypfs_update_update(sb); + pr_info("Hypervisor filesystem mounted\n"); + return 0; +} + +static struct dentry *hypfs_mount(struct file_system_type *fst, int flags, + const char *devname, void *data) +{ + return mount_single(fst, flags, data, hypfs_fill_super); +} + +static void hypfs_kill_super(struct super_block *sb) +{ + struct hypfs_sb_info *sb_info = sb->s_fs_info; + + if (sb->s_root) + hypfs_delete_tree(sb->s_root); + if (sb_info->update_file) + hypfs_remove(sb_info->update_file); + kfree(sb->s_fs_info); + sb->s_fs_info = NULL; + kill_litter_super(sb); +} + +static struct dentry *hypfs_create_file(struct dentry *parent, const char *name, + char *data, umode_t mode) +{ + struct dentry *dentry; + struct inode *inode; + + mutex_lock(&d_inode(parent)->i_mutex); + dentry = lookup_one_len(name, parent, strlen(name)); + if (IS_ERR(dentry)) { + dentry = ERR_PTR(-ENOMEM); + goto fail; + } + inode = hypfs_make_inode(parent->d_sb, mode); + if (!inode) { + dput(dentry); + dentry = ERR_PTR(-ENOMEM); + goto fail; + } + if (S_ISREG(mode)) { + inode->i_fop = &hypfs_file_ops; + if (data) + inode->i_size = strlen(data); + else + inode->i_size = 0; + } else if (S_ISDIR(mode)) { + inode->i_op = &simple_dir_inode_operations; + inode->i_fop = &simple_dir_operations; + inc_nlink(d_inode(parent)); + } else + BUG(); + inode->i_private = data; + d_instantiate(dentry, inode); + dget(dentry); +fail: + mutex_unlock(&d_inode(parent)->i_mutex); + return dentry; +} + +struct dentry *hypfs_mkdir(struct dentry *parent, const char *name) +{ + struct dentry *dentry; + + dentry = hypfs_create_file(parent, name, NULL, S_IFDIR | DIR_MODE); + if (IS_ERR(dentry)) + return dentry; + hypfs_add_dentry(dentry); + return dentry; +} + +static struct dentry *hypfs_create_update_file(struct dentry *dir) +{ + struct dentry *dentry; + + dentry = hypfs_create_file(dir, "update", NULL, + S_IFREG | UPDATE_FILE_MODE); + /* + * We do not put the update file on the 'delete' list with + * hypfs_add_dentry(), since it should not be removed when the tree + * is updated. + */ + return dentry; +} + +struct dentry *hypfs_create_u64(struct dentry *dir, + const char *name, __u64 value) +{ + char *buffer; + char tmp[TMP_SIZE]; + struct dentry *dentry; + + snprintf(tmp, TMP_SIZE, "%llu\n", (unsigned long long int)value); + buffer = kstrdup(tmp, GFP_KERNEL); + if (!buffer) + return ERR_PTR(-ENOMEM); + dentry = + hypfs_create_file(dir, name, buffer, S_IFREG | REG_FILE_MODE); + if (IS_ERR(dentry)) { + kfree(buffer); + return ERR_PTR(-ENOMEM); + } + hypfs_add_dentry(dentry); + return dentry; +} + +struct dentry *hypfs_create_str(struct dentry *dir, + const char *name, char *string) +{ + char *buffer; + struct dentry *dentry; + + buffer = kmalloc(strlen(string) + 2, GFP_KERNEL); + if (!buffer) + return ERR_PTR(-ENOMEM); + sprintf(buffer, "%s\n", string); + dentry = + hypfs_create_file(dir, name, buffer, S_IFREG | REG_FILE_MODE); + if (IS_ERR(dentry)) { + kfree(buffer); + return ERR_PTR(-ENOMEM); + } + hypfs_add_dentry(dentry); + return dentry; +} + +static const struct file_operations hypfs_file_ops = { + .open = hypfs_open, + .release = hypfs_release, + .read_iter = hypfs_read_iter, + .write_iter = hypfs_write_iter, + .llseek = no_llseek, +}; + +static struct file_system_type hypfs_type = { + .owner = THIS_MODULE, + .name = "s390_hypfs", + .mount = hypfs_mount, + .kill_sb = hypfs_kill_super +}; +MODULE_ALIAS_FS("s390_hypfs"); + +static const struct super_operations hypfs_s_ops = { + .statfs = simple_statfs, + .evict_inode = hypfs_evict_inode, + .show_options = hypfs_show_options, +}; + +static int __init hypfs_init(void) +{ + int rc; + + rc = hypfs_dbfs_init(); + if (rc) + return rc; + if (hypfs_diag_init()) { + rc = -ENODATA; + goto fail_dbfs_exit; + } + if (hypfs_vm_init()) { + rc = -ENODATA; + goto fail_hypfs_diag_exit; + } + if (hypfs_sprp_init()) { + rc = -ENODATA; + goto fail_hypfs_vm_exit; + } + if (hypfs_diag0c_init()) { + rc = -ENODATA; + goto fail_hypfs_sprp_exit; + } + rc = sysfs_create_mount_point(hypervisor_kobj, "s390"); + if (rc) + goto fail_hypfs_diag0c_exit; + rc = register_filesystem(&hypfs_type); + if (rc) + goto fail_filesystem; + return 0; + +fail_filesystem: + sysfs_remove_mount_point(hypervisor_kobj, "s390"); +fail_hypfs_diag0c_exit: + hypfs_diag0c_exit(); +fail_hypfs_sprp_exit: + hypfs_sprp_exit(); +fail_hypfs_vm_exit: + hypfs_vm_exit(); +fail_hypfs_diag_exit: + hypfs_diag_exit(); +fail_dbfs_exit: + hypfs_dbfs_exit(); + pr_err("Initialization of hypfs failed with rc=%i\n", rc); + return rc; +} + +static void __exit hypfs_exit(void) +{ + unregister_filesystem(&hypfs_type); + sysfs_remove_mount_point(hypervisor_kobj, "s390"); + hypfs_diag0c_exit(); + hypfs_sprp_exit(); + hypfs_vm_exit(); + hypfs_diag_exit(); + hypfs_dbfs_exit(); +} + +module_init(hypfs_init) +module_exit(hypfs_exit) + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Michael Holzheu <holzheu@de.ibm.com>"); +MODULE_DESCRIPTION("s390 Hypervisor Filesystem"); diff --git a/arch/s390/include/asm/Kbuild b/arch/s390/include/asm/Kbuild new file mode 100644 index 000000000..c631f98fd --- /dev/null +++ b/arch/s390/include/asm/Kbuild @@ -0,0 +1,8 @@ + + +generic-y += clkdev.h +generic-y += irq_work.h +generic-y += mcs_spinlock.h +generic-y += preempt.h +generic-y += scatterlist.h +generic-y += trace_clock.h diff --git a/arch/s390/include/asm/airq.h b/arch/s390/include/asm/airq.h new file mode 100644 index 000000000..bd93ff666 --- /dev/null +++ b/arch/s390/include/asm/airq.h @@ -0,0 +1,103 @@ +/* + * Copyright IBM Corp. 2002, 2007 + * Author(s): Ingo Adlung <adlung@de.ibm.com> + * Cornelia Huck <cornelia.huck@de.ibm.com> + * Arnd Bergmann <arndb@de.ibm.com> + * Peter Oberparleiter <peter.oberparleiter@de.ibm.com> + */ + +#ifndef _ASM_S390_AIRQ_H +#define _ASM_S390_AIRQ_H + +#include <linux/bit_spinlock.h> + +struct airq_struct { + struct hlist_node list; /* Handler queueing. */ + void (*handler)(struct airq_struct *); /* Thin-interrupt handler */ + u8 *lsi_ptr; /* Local-Summary-Indicator pointer */ + u8 lsi_mask; /* Local-Summary-Indicator mask */ + u8 isc; /* Interrupt-subclass */ + u8 flags; +}; + +#define AIRQ_PTR_ALLOCATED 0x01 + +int register_adapter_interrupt(struct airq_struct *airq); +void unregister_adapter_interrupt(struct airq_struct *airq); + +/* Adapter interrupt bit vector */ +struct airq_iv { + unsigned long *vector; /* Adapter interrupt bit vector */ + unsigned long *avail; /* Allocation bit mask for the bit vector */ + unsigned long *bitlock; /* Lock bit mask for the bit vector */ + unsigned long *ptr; /* Pointer associated with each bit */ + unsigned int *data; /* 32 bit value associated with each bit */ + unsigned long bits; /* Number of bits in the vector */ + unsigned long end; /* Number of highest allocated bit + 1 */ + spinlock_t lock; /* Lock to protect alloc & free */ +}; + +#define AIRQ_IV_ALLOC 1 /* Use an allocation bit mask */ +#define AIRQ_IV_BITLOCK 2 /* Allocate the lock bit mask */ +#define AIRQ_IV_PTR 4 /* Allocate the ptr array */ +#define AIRQ_IV_DATA 8 /* Allocate the data array */ + +struct airq_iv *airq_iv_create(unsigned long bits, unsigned long flags); +void airq_iv_release(struct airq_iv *iv); +unsigned long airq_iv_alloc(struct airq_iv *iv, unsigned long num); +void airq_iv_free(struct airq_iv *iv, unsigned long bit, unsigned long num); +unsigned long airq_iv_scan(struct airq_iv *iv, unsigned long start, + unsigned long end); + +static inline unsigned long airq_iv_alloc_bit(struct airq_iv *iv) +{ + return airq_iv_alloc(iv, 1); +} + +static inline void airq_iv_free_bit(struct airq_iv *iv, unsigned long bit) +{ + airq_iv_free(iv, bit, 1); +} + +static inline unsigned long airq_iv_end(struct airq_iv *iv) +{ + return iv->end; +} + +static inline void airq_iv_lock(struct airq_iv *iv, unsigned long bit) +{ + const unsigned long be_to_le = BITS_PER_LONG - 1; + bit_spin_lock(bit ^ be_to_le, iv->bitlock); +} + +static inline void airq_iv_unlock(struct airq_iv *iv, unsigned long bit) +{ + const unsigned long be_to_le = BITS_PER_LONG - 1; + bit_spin_unlock(bit ^ be_to_le, iv->bitlock); +} + +static inline void airq_iv_set_data(struct airq_iv *iv, unsigned long bit, + unsigned int data) +{ + iv->data[bit] = data; +} + +static inline unsigned int airq_iv_get_data(struct airq_iv *iv, + unsigned long bit) +{ + return iv->data[bit]; +} + +static inline void airq_iv_set_ptr(struct airq_iv *iv, unsigned long bit, + unsigned long ptr) +{ + iv->ptr[bit] = ptr; +} + +static inline unsigned long airq_iv_get_ptr(struct airq_iv *iv, + unsigned long bit) +{ + return iv->ptr[bit]; +} + +#endif /* _ASM_S390_AIRQ_H */ diff --git a/arch/s390/include/asm/appldata.h b/arch/s390/include/asm/appldata.h new file mode 100644 index 000000000..16887c5fd --- /dev/null +++ b/arch/s390/include/asm/appldata.h @@ -0,0 +1,64 @@ +/* + * Copyright IBM Corp. 2006 + * + * Author(s): Melissa Howland <melissah@us.ibm.com> + */ + +#ifndef _ASM_S390_APPLDATA_H +#define _ASM_S390_APPLDATA_H + +#include <asm/io.h> + +#define APPLDATA_START_INTERVAL_REC 0x80 +#define APPLDATA_STOP_REC 0x81 +#define APPLDATA_GEN_EVENT_REC 0x82 +#define APPLDATA_START_CONFIG_REC 0x83 + +/* + * Parameter list for DIAGNOSE X'DC' + */ +struct appldata_parameter_list { + u16 diag; + u8 function; + u8 parlist_length; + u32 unused01; + u16 reserved; + u16 buffer_length; + u32 unused02; + u64 product_id_addr; + u64 buffer_addr; +} __attribute__ ((packed)); + +struct appldata_product_id { + char prod_nr[7]; /* product number */ + u16 prod_fn; /* product function */ + u8 record_nr; /* record number */ + u16 version_nr; /* version */ + u16 release_nr; /* release */ + u16 mod_lvl; /* modification level */ +} __attribute__ ((packed)); + +static inline int appldata_asm(struct appldata_product_id *id, + unsigned short fn, void *buffer, + unsigned short length) +{ + struct appldata_parameter_list parm_list; + int ry; + + if (!MACHINE_IS_VM) + return -EOPNOTSUPP; + parm_list.diag = 0xdc; + parm_list.function = fn; + parm_list.parlist_length = sizeof(parm_list); + parm_list.buffer_length = length; + parm_list.product_id_addr = (unsigned long) id; + parm_list.buffer_addr = virt_to_phys(buffer); + asm volatile( + " diag %1,%0,0xdc" + : "=d" (ry) + : "d" (&parm_list), "m" (parm_list), "m" (*id) + : "cc"); + return ry; +} + +#endif /* _ASM_S390_APPLDATA_H */ diff --git a/arch/s390/include/asm/asm-offsets.h b/arch/s390/include/asm/asm-offsets.h new file mode 100644 index 000000000..d370ee36a --- /dev/null +++ b/arch/s390/include/asm/asm-offsets.h @@ -0,0 +1 @@ +#include <generated/asm-offsets.h> diff --git a/arch/s390/include/asm/atomic.h b/arch/s390/include/asm/atomic.h new file mode 100644 index 000000000..adbe3802e --- /dev/null +++ b/arch/s390/include/asm/atomic.h @@ -0,0 +1,320 @@ +/* + * Copyright IBM Corp. 1999, 2009 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Denis Joseph Barrow, + * Arnd Bergmann <arndb@de.ibm.com>, + * + * Atomic operations that C can't guarantee us. + * Useful for resource counting etc. + * s390 uses 'Compare And Swap' for atomicity in SMP environment. + * + */ + +#ifndef __ARCH_S390_ATOMIC__ +#define __ARCH_S390_ATOMIC__ + +#include <linux/compiler.h> +#include <linux/types.h> +#include <asm/barrier.h> +#include <asm/cmpxchg.h> + +#define ATOMIC_INIT(i) { (i) } + +#define __ATOMIC_NO_BARRIER "\n" + +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES + +#define __ATOMIC_OR "lao" +#define __ATOMIC_AND "lan" +#define __ATOMIC_ADD "laa" +#define __ATOMIC_BARRIER "bcr 14,0\n" + +#define __ATOMIC_LOOP(ptr, op_val, op_string, __barrier) \ +({ \ + int old_val; \ + \ + typecheck(atomic_t *, ptr); \ + asm volatile( \ + __barrier \ + op_string " %0,%2,%1\n" \ + __barrier \ + : "=d" (old_val), "+Q" ((ptr)->counter) \ + : "d" (op_val) \ + : "cc", "memory"); \ + old_val; \ +}) + +#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +#define __ATOMIC_OR "or" +#define __ATOMIC_AND "nr" +#define __ATOMIC_ADD "ar" +#define __ATOMIC_BARRIER "\n" + +#define __ATOMIC_LOOP(ptr, op_val, op_string, __barrier) \ +({ \ + int old_val, new_val; \ + \ + typecheck(atomic_t *, ptr); \ + asm volatile( \ + " l %0,%2\n" \ + "0: lr %1,%0\n" \ + op_string " %1,%3\n" \ + " cs %0,%1,%2\n" \ + " jl 0b" \ + : "=&d" (old_val), "=&d" (new_val), "+Q" ((ptr)->counter)\ + : "d" (op_val) \ + : "cc", "memory"); \ + old_val; \ +}) + +#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +static inline int atomic_read(const atomic_t *v) +{ + int c; + + asm volatile( + " l %0,%1\n" + : "=d" (c) : "Q" (v->counter)); + return c; +} + +static inline void atomic_set(atomic_t *v, int i) +{ + asm volatile( + " st %1,%0\n" + : "=Q" (v->counter) : "d" (i)); +} + +static inline int atomic_add_return(int i, atomic_t *v) +{ + return __ATOMIC_LOOP(v, i, __ATOMIC_ADD, __ATOMIC_BARRIER) + i; +} + +static inline void atomic_add(int i, atomic_t *v) +{ +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES + if (__builtin_constant_p(i) && (i > -129) && (i < 128)) { + asm volatile( + "asi %0,%1\n" + : "+Q" (v->counter) + : "i" (i) + : "cc", "memory"); + return; + } +#endif + __ATOMIC_LOOP(v, i, __ATOMIC_ADD, __ATOMIC_NO_BARRIER); +} + +#define atomic_add_negative(_i, _v) (atomic_add_return(_i, _v) < 0) +#define atomic_inc(_v) atomic_add(1, _v) +#define atomic_inc_return(_v) atomic_add_return(1, _v) +#define atomic_inc_and_test(_v) (atomic_add_return(1, _v) == 0) +#define atomic_sub(_i, _v) atomic_add(-(int)(_i), _v) +#define atomic_sub_return(_i, _v) atomic_add_return(-(int)(_i), _v) +#define atomic_sub_and_test(_i, _v) (atomic_sub_return(_i, _v) == 0) +#define atomic_dec(_v) atomic_sub(1, _v) +#define atomic_dec_return(_v) atomic_sub_return(1, _v) +#define atomic_dec_and_test(_v) (atomic_sub_return(1, _v) == 0) + +static inline void atomic_clear_mask(unsigned int mask, atomic_t *v) +{ + __ATOMIC_LOOP(v, ~mask, __ATOMIC_AND, __ATOMIC_NO_BARRIER); +} + +static inline void atomic_set_mask(unsigned int mask, atomic_t *v) +{ + __ATOMIC_LOOP(v, mask, __ATOMIC_OR, __ATOMIC_NO_BARRIER); +} + +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + +static inline int atomic_cmpxchg(atomic_t *v, int old, int new) +{ + asm volatile( + " cs %0,%2,%1" + : "+d" (old), "+Q" (v->counter) + : "d" (new) + : "cc", "memory"); + return old; +} + +static inline int __atomic_add_unless(atomic_t *v, int a, int u) +{ + int c, old; + c = atomic_read(v); + for (;;) { + if (unlikely(c == u)) + break; + old = atomic_cmpxchg(v, c, c + a); + if (likely(old == c)) + break; + c = old; + } + return c; +} + + +#undef __ATOMIC_LOOP + +#define ATOMIC64_INIT(i) { (i) } + +#define __ATOMIC64_NO_BARRIER "\n" + +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES + +#define __ATOMIC64_OR "laog" +#define __ATOMIC64_AND "lang" +#define __ATOMIC64_ADD "laag" +#define __ATOMIC64_BARRIER "bcr 14,0\n" + +#define __ATOMIC64_LOOP(ptr, op_val, op_string, __barrier) \ +({ \ + long long old_val; \ + \ + typecheck(atomic64_t *, ptr); \ + asm volatile( \ + __barrier \ + op_string " %0,%2,%1\n" \ + __barrier \ + : "=d" (old_val), "+Q" ((ptr)->counter) \ + : "d" (op_val) \ + : "cc", "memory"); \ + old_val; \ +}) + +#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +#define __ATOMIC64_OR "ogr" +#define __ATOMIC64_AND "ngr" +#define __ATOMIC64_ADD "agr" +#define __ATOMIC64_BARRIER "\n" + +#define __ATOMIC64_LOOP(ptr, op_val, op_string, __barrier) \ +({ \ + long long old_val, new_val; \ + \ + typecheck(atomic64_t *, ptr); \ + asm volatile( \ + " lg %0,%2\n" \ + "0: lgr %1,%0\n" \ + op_string " %1,%3\n" \ + " csg %0,%1,%2\n" \ + " jl 0b" \ + : "=&d" (old_val), "=&d" (new_val), "+Q" ((ptr)->counter)\ + : "d" (op_val) \ + : "cc", "memory"); \ + old_val; \ +}) + +#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +static inline long long atomic64_read(const atomic64_t *v) +{ + long long c; + + asm volatile( + " lg %0,%1\n" + : "=d" (c) : "Q" (v->counter)); + return c; +} + +static inline void atomic64_set(atomic64_t *v, long long i) +{ + asm volatile( + " stg %1,%0\n" + : "=Q" (v->counter) : "d" (i)); +} + +static inline long long atomic64_add_return(long long i, atomic64_t *v) +{ + return __ATOMIC64_LOOP(v, i, __ATOMIC64_ADD, __ATOMIC64_BARRIER) + i; +} + +static inline void atomic64_add(long long i, atomic64_t *v) +{ +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES + if (__builtin_constant_p(i) && (i > -129) && (i < 128)) { + asm volatile( + "agsi %0,%1\n" + : "+Q" (v->counter) + : "i" (i) + : "cc", "memory"); + return; + } +#endif + __ATOMIC64_LOOP(v, i, __ATOMIC64_ADD, __ATOMIC64_NO_BARRIER); +} + +static inline void atomic64_clear_mask(unsigned long mask, atomic64_t *v) +{ + __ATOMIC64_LOOP(v, ~mask, __ATOMIC64_AND, __ATOMIC64_NO_BARRIER); +} + +static inline void atomic64_set_mask(unsigned long mask, atomic64_t *v) +{ + __ATOMIC64_LOOP(v, mask, __ATOMIC64_OR, __ATOMIC64_NO_BARRIER); +} + +#define atomic64_xchg(v, new) (xchg(&((v)->counter), new)) + +static inline long long atomic64_cmpxchg(atomic64_t *v, + long long old, long long new) +{ + asm volatile( + " csg %0,%2,%1" + : "+d" (old), "+Q" (v->counter) + : "d" (new) + : "cc", "memory"); + return old; +} + +#undef __ATOMIC64_LOOP + +static inline int atomic64_add_unless(atomic64_t *v, long long i, long long u) +{ + long long c, old; + + c = atomic64_read(v); + for (;;) { + if (unlikely(c == u)) + break; + old = atomic64_cmpxchg(v, c, c + i); + if (likely(old == c)) + break; + c = old; + } + return c != u; +} + +static inline long long atomic64_dec_if_positive(atomic64_t *v) +{ + long long c, old, dec; + + c = atomic64_read(v); + for (;;) { + dec = c - 1; + if (unlikely(dec < 0)) + break; + old = atomic64_cmpxchg((v), c, dec); + if (likely(old == c)) + break; + c = old; + } + return dec; +} + +#define atomic64_add_negative(_i, _v) (atomic64_add_return(_i, _v) < 0) +#define atomic64_inc(_v) atomic64_add(1, _v) +#define atomic64_inc_return(_v) atomic64_add_return(1, _v) +#define atomic64_inc_and_test(_v) (atomic64_add_return(1, _v) == 0) +#define atomic64_sub_return(_i, _v) atomic64_add_return(-(long long)(_i), _v) +#define atomic64_sub(_i, _v) atomic64_add(-(long long)(_i), _v) +#define atomic64_sub_and_test(_i, _v) (atomic64_sub_return(_i, _v) == 0) +#define atomic64_dec(_v) atomic64_sub(1, _v) +#define atomic64_dec_return(_v) atomic64_sub_return(1, _v) +#define atomic64_dec_and_test(_v) (atomic64_sub_return(1, _v) == 0) +#define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0) + +#endif /* __ARCH_S390_ATOMIC__ */ diff --git a/arch/s390/include/asm/barrier.h b/arch/s390/include/asm/barrier.h new file mode 100644 index 000000000..8d724718e --- /dev/null +++ b/arch/s390/include/asm/barrier.h @@ -0,0 +1,56 @@ +/* + * Copyright IBM Corp. 1999, 2009 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef __ASM_BARRIER_H +#define __ASM_BARRIER_H + +/* + * Force strict CPU ordering. + * And yes, this is required on UP too when we're talking + * to devices. + */ + +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES +/* Fast-BCR without checkpoint synchronization */ +#define __ASM_BARRIER "bcr 14,0\n" +#else +#define __ASM_BARRIER "bcr 15,0\n" +#endif + +#define mb() do { asm volatile(__ASM_BARRIER : : : "memory"); } while (0) + +#define rmb() mb() +#define wmb() mb() +#define dma_rmb() rmb() +#define dma_wmb() wmb() +#define smp_mb() mb() +#define smp_rmb() rmb() +#define smp_wmb() wmb() + +#define read_barrier_depends() do { } while (0) +#define smp_read_barrier_depends() do { } while (0) + +#define smp_mb__before_atomic() smp_mb() +#define smp_mb__after_atomic() smp_mb() + +#define set_mb(var, value) do { var = value; mb(); } while (0) + +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ___p1; \ +}) + +#endif /* __ASM_BARRIER_H */ diff --git a/arch/s390/include/asm/bitops.h b/arch/s390/include/asm/bitops.h new file mode 100644 index 000000000..9b68e98a7 --- /dev/null +++ b/arch/s390/include/asm/bitops.h @@ -0,0 +1,454 @@ +/* + * Copyright IBM Corp. 1999,2013 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>, + * + * The description below was taken in large parts from the powerpc + * bitops header file: + * Within a word, bits are numbered LSB first. Lot's of places make + * this assumption by directly testing bits with (val & (1<<nr)). + * This can cause confusion for large (> 1 word) bitmaps on a + * big-endian system because, unlike little endian, the number of each + * bit depends on the word size. + * + * The bitop functions are defined to work on unsigned longs, so for an + * s390x system the bits end up numbered: + * |63..............0|127............64|191...........128|255...........192| + * and on s390: + * |31.....0|63....32|95....64|127...96|159..128|191..160|223..192|255..224| + * + * There are a few little-endian macros used mostly for filesystem + * bitmaps, these work on similar bit arrays layouts, but + * byte-oriented: + * |7...0|15...8|23...16|31...24|39...32|47...40|55...48|63...56| + * + * The main difference is that bit 3-5 (64b) or 3-4 (32b) in the bit + * number field needs to be reversed compared to the big-endian bit + * fields. This can be achieved by XOR with 0x38 (64b) or 0x18 (32b). + * + * We also have special functions which work with an MSB0 encoding: + * on an s390x system the bits are numbered: + * |0..............63|64............127|128...........191|192...........255| + * and on s390: + * |0.....31|32....63|64....95|96...127|128..159|160..191|192..223|224..255| + * + * The main difference is that bit 0-63 (64b) or 0-31 (32b) in the bit + * number field needs to be reversed compared to the LSB0 encoded bit + * fields. This can be achieved by XOR with 0x3f (64b) or 0x1f (32b). + * + */ + +#ifndef _S390_BITOPS_H +#define _S390_BITOPS_H + +#ifndef _LINUX_BITOPS_H +#error only <linux/bitops.h> can be included directly +#endif + +#include <linux/typecheck.h> +#include <linux/compiler.h> +#include <asm/barrier.h> + +#define __BITOPS_NO_BARRIER "\n" + +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES + +#define __BITOPS_OR "laog" +#define __BITOPS_AND "lang" +#define __BITOPS_XOR "laxg" +#define __BITOPS_BARRIER "bcr 14,0\n" + +#define __BITOPS_LOOP(__addr, __val, __op_string, __barrier) \ +({ \ + unsigned long __old; \ + \ + typecheck(unsigned long *, (__addr)); \ + asm volatile( \ + __barrier \ + __op_string " %0,%2,%1\n" \ + __barrier \ + : "=d" (__old), "+Q" (*(__addr)) \ + : "d" (__val) \ + : "cc", "memory"); \ + __old; \ +}) + +#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +#define __BITOPS_OR "ogr" +#define __BITOPS_AND "ngr" +#define __BITOPS_XOR "xgr" +#define __BITOPS_BARRIER "\n" + +#define __BITOPS_LOOP(__addr, __val, __op_string, __barrier) \ +({ \ + unsigned long __old, __new; \ + \ + typecheck(unsigned long *, (__addr)); \ + asm volatile( \ + " lg %0,%2\n" \ + "0: lgr %1,%0\n" \ + __op_string " %1,%3\n" \ + " csg %0,%1,%2\n" \ + " jl 0b" \ + : "=&d" (__old), "=&d" (__new), "+Q" (*(__addr))\ + : "d" (__val) \ + : "cc", "memory"); \ + __old; \ +}) + +#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +#define __BITOPS_WORDS(bits) (((bits) + BITS_PER_LONG - 1) / BITS_PER_LONG) + +static inline unsigned long * +__bitops_word(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned long addr; + + addr = (unsigned long)ptr + ((nr ^ (nr & (BITS_PER_LONG - 1))) >> 3); + return (unsigned long *)addr; +} + +static inline unsigned char * +__bitops_byte(unsigned long nr, volatile unsigned long *ptr) +{ + return ((unsigned char *)ptr) + ((nr ^ (BITS_PER_LONG - 8)) >> 3); +} + +static inline void set_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned long *addr = __bitops_word(nr, ptr); + unsigned long mask; + +#ifdef CONFIG_HAVE_MARCH_ZEC12_FEATURES + if (__builtin_constant_p(nr)) { + unsigned char *caddr = __bitops_byte(nr, ptr); + + asm volatile( + "oi %0,%b1\n" + : "+Q" (*caddr) + : "i" (1 << (nr & 7)) + : "cc", "memory"); + return; + } +#endif + mask = 1UL << (nr & (BITS_PER_LONG - 1)); + __BITOPS_LOOP(addr, mask, __BITOPS_OR, __BITOPS_NO_BARRIER); +} + +static inline void clear_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned long *addr = __bitops_word(nr, ptr); + unsigned long mask; + +#ifdef CONFIG_HAVE_MARCH_ZEC12_FEATURES + if (__builtin_constant_p(nr)) { + unsigned char *caddr = __bitops_byte(nr, ptr); + + asm volatile( + "ni %0,%b1\n" + : "+Q" (*caddr) + : "i" (~(1 << (nr & 7))) + : "cc", "memory"); + return; + } +#endif + mask = ~(1UL << (nr & (BITS_PER_LONG - 1))); + __BITOPS_LOOP(addr, mask, __BITOPS_AND, __BITOPS_NO_BARRIER); +} + +static inline void change_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned long *addr = __bitops_word(nr, ptr); + unsigned long mask; + +#ifdef CONFIG_HAVE_MARCH_ZEC12_FEATURES + if (__builtin_constant_p(nr)) { + unsigned char *caddr = __bitops_byte(nr, ptr); + + asm volatile( + "xi %0,%b1\n" + : "+Q" (*caddr) + : "i" (1 << (nr & 7)) + : "cc", "memory"); + return; + } +#endif + mask = 1UL << (nr & (BITS_PER_LONG - 1)); + __BITOPS_LOOP(addr, mask, __BITOPS_XOR, __BITOPS_NO_BARRIER); +} + +static inline int +test_and_set_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned long *addr = __bitops_word(nr, ptr); + unsigned long old, mask; + + mask = 1UL << (nr & (BITS_PER_LONG - 1)); + old = __BITOPS_LOOP(addr, mask, __BITOPS_OR, __BITOPS_BARRIER); + return (old & mask) != 0; +} + +static inline int +test_and_clear_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned long *addr = __bitops_word(nr, ptr); + unsigned long old, mask; + + mask = ~(1UL << (nr & (BITS_PER_LONG - 1))); + old = __BITOPS_LOOP(addr, mask, __BITOPS_AND, __BITOPS_BARRIER); + return (old & ~mask) != 0; +} + +static inline int +test_and_change_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned long *addr = __bitops_word(nr, ptr); + unsigned long old, mask; + + mask = 1UL << (nr & (BITS_PER_LONG - 1)); + old = __BITOPS_LOOP(addr, mask, __BITOPS_XOR, __BITOPS_BARRIER); + return (old & mask) != 0; +} + +static inline void __set_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned char *addr = __bitops_byte(nr, ptr); + + *addr |= 1 << (nr & 7); +} + +static inline void +__clear_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned char *addr = __bitops_byte(nr, ptr); + + *addr &= ~(1 << (nr & 7)); +} + +static inline void __change_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned char *addr = __bitops_byte(nr, ptr); + + *addr ^= 1 << (nr & 7); +} + +static inline int +__test_and_set_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned char *addr = __bitops_byte(nr, ptr); + unsigned char ch; + + ch = *addr; + *addr |= 1 << (nr & 7); + return (ch >> (nr & 7)) & 1; +} + +static inline int +__test_and_clear_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned char *addr = __bitops_byte(nr, ptr); + unsigned char ch; + + ch = *addr; + *addr &= ~(1 << (nr & 7)); + return (ch >> (nr & 7)) & 1; +} + +static inline int +__test_and_change_bit(unsigned long nr, volatile unsigned long *ptr) +{ + unsigned char *addr = __bitops_byte(nr, ptr); + unsigned char ch; + + ch = *addr; + *addr ^= 1 << (nr & 7); + return (ch >> (nr & 7)) & 1; +} + +static inline int test_bit(unsigned long nr, const volatile unsigned long *ptr) +{ + const volatile unsigned char *addr; + + addr = ((const volatile unsigned char *)ptr); + addr += (nr ^ (BITS_PER_LONG - 8)) >> 3; + return (*addr >> (nr & 7)) & 1; +} + +/* + * Functions which use MSB0 bit numbering. + * On an s390x system the bits are numbered: + * |0..............63|64............127|128...........191|192...........255| + * and on s390: + * |0.....31|32....63|64....95|96...127|128..159|160..191|192..223|224..255| + */ +unsigned long find_first_bit_inv(const unsigned long *addr, unsigned long size); +unsigned long find_next_bit_inv(const unsigned long *addr, unsigned long size, + unsigned long offset); + +static inline void set_bit_inv(unsigned long nr, volatile unsigned long *ptr) +{ + return set_bit(nr ^ (BITS_PER_LONG - 1), ptr); +} + +static inline void clear_bit_inv(unsigned long nr, volatile unsigned long *ptr) +{ + return clear_bit(nr ^ (BITS_PER_LONG - 1), ptr); +} + +static inline void __set_bit_inv(unsigned long nr, volatile unsigned long *ptr) +{ + return __set_bit(nr ^ (BITS_PER_LONG - 1), ptr); +} + +static inline void __clear_bit_inv(unsigned long nr, volatile unsigned long *ptr) +{ + return __clear_bit(nr ^ (BITS_PER_LONG - 1), ptr); +} + +static inline int test_bit_inv(unsigned long nr, + const volatile unsigned long *ptr) +{ + return test_bit(nr ^ (BITS_PER_LONG - 1), ptr); +} + +#ifdef CONFIG_HAVE_MARCH_Z9_109_FEATURES + +/** + * __flogr - find leftmost one + * @word - The word to search + * + * Returns the bit number of the most significant bit set, + * where the most significant bit has bit number 0. + * If no bit is set this function returns 64. + */ +static inline unsigned char __flogr(unsigned long word) +{ + if (__builtin_constant_p(word)) { + unsigned long bit = 0; + + if (!word) + return 64; + if (!(word & 0xffffffff00000000UL)) { + word <<= 32; + bit += 32; + } + if (!(word & 0xffff000000000000UL)) { + word <<= 16; + bit += 16; + } + if (!(word & 0xff00000000000000UL)) { + word <<= 8; + bit += 8; + } + if (!(word & 0xf000000000000000UL)) { + word <<= 4; + bit += 4; + } + if (!(word & 0xc000000000000000UL)) { + word <<= 2; + bit += 2; + } + if (!(word & 0x8000000000000000UL)) { + word <<= 1; + bit += 1; + } + return bit; + } else { + register unsigned long bit asm("4") = word; + register unsigned long out asm("5"); + + asm volatile( + " flogr %[bit],%[bit]\n" + : [bit] "+d" (bit), [out] "=d" (out) : : "cc"); + return bit; + } +} + +/** + * __ffs - find first bit in word. + * @word: The word to search + * + * Undefined if no bit exists, so code should check against 0 first. + */ +static inline unsigned long __ffs(unsigned long word) +{ + return __flogr(-word & word) ^ (BITS_PER_LONG - 1); +} + +/** + * ffs - find first bit set + * @word: the word to search + * + * This is defined the same way as the libc and + * compiler builtin ffs routines (man ffs). + */ +static inline int ffs(int word) +{ + unsigned long mask = 2 * BITS_PER_LONG - 1; + unsigned int val = (unsigned int)word; + + return (1 + (__flogr(-val & val) ^ (BITS_PER_LONG - 1))) & mask; +} + +/** + * __fls - find last (most-significant) set bit in a long word + * @word: the word to search + * + * Undefined if no set bit exists, so code should check against 0 first. + */ +static inline unsigned long __fls(unsigned long word) +{ + return __flogr(word) ^ (BITS_PER_LONG - 1); +} + +/** + * fls64 - find last set bit in a 64-bit word + * @word: the word to search + * + * This is defined in a similar way as the libc and compiler builtin + * ffsll, but returns the position of the most significant set bit. + * + * fls64(value) returns 0 if value is 0 or the position of the last + * set bit if value is nonzero. The last (most significant) bit is + * at position 64. + */ +static inline int fls64(unsigned long word) +{ + unsigned long mask = 2 * BITS_PER_LONG - 1; + + return (1 + (__flogr(word) ^ (BITS_PER_LONG - 1))) & mask; +} + +/** + * fls - find last (most-significant) bit set + * @word: the word to search + * + * This is defined the same way as ffs. + * Note fls(0) = 0, fls(1) = 1, fls(0x80000000) = 32. + */ +static inline int fls(int word) +{ + return fls64((unsigned int)word); +} + +#else /* CONFIG_HAVE_MARCH_Z9_109_FEATURES */ + +#include <asm-generic/bitops/__ffs.h> +#include <asm-generic/bitops/ffs.h> +#include <asm-generic/bitops/__fls.h> +#include <asm-generic/bitops/fls.h> +#include <asm-generic/bitops/fls64.h> + +#endif /* CONFIG_HAVE_MARCH_Z9_109_FEATURES */ + +#include <asm-generic/bitops/ffz.h> +#include <asm-generic/bitops/find.h> +#include <asm-generic/bitops/hweight.h> +#include <asm-generic/bitops/lock.h> +#include <asm-generic/bitops/sched.h> +#include <asm-generic/bitops/le.h> +#include <asm-generic/bitops/ext2-atomic-setbit.h> + +#endif /* _S390_BITOPS_H */ diff --git a/arch/s390/include/asm/bug.h b/arch/s390/include/asm/bug.h new file mode 100644 index 000000000..bf90d1fd9 --- /dev/null +++ b/arch/s390/include/asm/bug.h @@ -0,0 +1,71 @@ +#ifndef _ASM_S390_BUG_H +#define _ASM_S390_BUG_H + +#include <linux/kernel.h> + +#ifdef CONFIG_BUG + +#ifdef CONFIG_DEBUG_BUGVERBOSE + +#define __EMIT_BUG(x) do { \ + asm volatile( \ + "0: j 0b+2\n" \ + "1:\n" \ + ".section .rodata.str,\"aMS\",@progbits,1\n" \ + "2: .asciz \""__FILE__"\"\n" \ + ".previous\n" \ + ".section __bug_table,\"a\"\n" \ + "3: .long 1b-3b,2b-3b\n" \ + " .short %0,%1\n" \ + " .org 3b+%2\n" \ + ".previous\n" \ + : : "i" (__LINE__), \ + "i" (x), \ + "i" (sizeof(struct bug_entry))); \ +} while (0) + +#else /* CONFIG_DEBUG_BUGVERBOSE */ + +#define __EMIT_BUG(x) do { \ + asm volatile( \ + "0: j 0b+2\n" \ + "1:\n" \ + ".section __bug_table,\"a\"\n" \ + "2: .long 1b-2b\n" \ + " .short %0\n" \ + " .org 2b+%1\n" \ + ".previous\n" \ + : : "i" (x), \ + "i" (sizeof(struct bug_entry))); \ +} while (0) + +#endif /* CONFIG_DEBUG_BUGVERBOSE */ + +#define BUG() do { \ + __EMIT_BUG(0); \ + unreachable(); \ +} while (0) + +#define __WARN_TAINT(taint) do { \ + __EMIT_BUG(BUGFLAG_TAINT(taint)); \ +} while (0) + +#define WARN_ON(x) ({ \ + int __ret_warn_on = !!(x); \ + if (__builtin_constant_p(__ret_warn_on)) { \ + if (__ret_warn_on) \ + __WARN(); \ + } else { \ + if (unlikely(__ret_warn_on)) \ + __WARN(); \ + } \ + unlikely(__ret_warn_on); \ +}) + +#define HAVE_ARCH_BUG +#define HAVE_ARCH_WARN_ON +#endif /* CONFIG_BUG */ + +#include <asm-generic/bug.h> + +#endif /* _ASM_S390_BUG_H */ diff --git a/arch/s390/include/asm/bugs.h b/arch/s390/include/asm/bugs.h new file mode 100644 index 000000000..0f5bd894f --- /dev/null +++ b/arch/s390/include/asm/bugs.h @@ -0,0 +1,20 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "include/asm-i386/bugs.h" + * Copyright (C) 1994 Linus Torvalds + */ + +/* + * This is included by init/main.c to check for architecture-dependent bugs. + * + * Needs: + * void check_bugs(void); + */ + +static inline void check_bugs(void) +{ + /* s390 has no bugs ... */ +} diff --git a/arch/s390/include/asm/cache.h b/arch/s390/include/asm/cache.h new file mode 100644 index 000000000..4d7ccac5f --- /dev/null +++ b/arch/s390/include/asm/cache.h @@ -0,0 +1,18 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * + * Derived from "include/asm-i386/cache.h" + * Copyright (C) 1992, Linus Torvalds + */ + +#ifndef __ARCH_S390_CACHE_H +#define __ARCH_S390_CACHE_H + +#define L1_CACHE_BYTES 256 +#define L1_CACHE_SHIFT 8 +#define NET_SKB_PAD 32 + +#define __read_mostly __attribute__((__section__(".data..read_mostly"))) + +#endif diff --git a/arch/s390/include/asm/cacheflush.h b/arch/s390/include/asm/cacheflush.h new file mode 100644 index 000000000..58fae7d09 --- /dev/null +++ b/arch/s390/include/asm/cacheflush.h @@ -0,0 +1,12 @@ +#ifndef _S390_CACHEFLUSH_H +#define _S390_CACHEFLUSH_H + +/* Caches aren't brain-dead on the s390. */ +#include <asm-generic/cacheflush.h> + +int set_memory_ro(unsigned long addr, int numpages); +int set_memory_rw(unsigned long addr, int numpages); +int set_memory_nx(unsigned long addr, int numpages); +int set_memory_x(unsigned long addr, int numpages); + +#endif /* _S390_CACHEFLUSH_H */ diff --git a/arch/s390/include/asm/ccwdev.h b/arch/s390/include/asm/ccwdev.h new file mode 100644 index 000000000..b80e456d6 --- /dev/null +++ b/arch/s390/include/asm/ccwdev.h @@ -0,0 +1,233 @@ +/* + * Copyright IBM Corp. 2002, 2009 + * + * Author(s): Arnd Bergmann <arndb@de.ibm.com> + * + * Interface for CCW device drivers + */ +#ifndef _S390_CCWDEV_H_ +#define _S390_CCWDEV_H_ + +#include <linux/device.h> +#include <linux/mod_devicetable.h> +#include <asm/fcx.h> +#include <asm/irq.h> +#include <asm/schid.h> + +/* structs from asm/cio.h */ +struct irb; +struct ccw1; +struct ccw_dev_id; + +/* simplified initializers for struct ccw_device: + * CCW_DEVICE and CCW_DEVICE_DEVTYPE initialize one + * entry in your MODULE_DEVICE_TABLE and set the match_flag correctly */ +#define CCW_DEVICE(cu, cum) \ + .cu_type=(cu), .cu_model=(cum), \ + .match_flags=(CCW_DEVICE_ID_MATCH_CU_TYPE \ + | (cum ? CCW_DEVICE_ID_MATCH_CU_MODEL : 0)) + +#define CCW_DEVICE_DEVTYPE(cu, cum, dev, devm) \ + .cu_type=(cu), .cu_model=(cum), .dev_type=(dev), .dev_model=(devm),\ + .match_flags=CCW_DEVICE_ID_MATCH_CU_TYPE \ + | ((cum) ? CCW_DEVICE_ID_MATCH_CU_MODEL : 0) \ + | CCW_DEVICE_ID_MATCH_DEVICE_TYPE \ + | ((devm) ? CCW_DEVICE_ID_MATCH_DEVICE_MODEL : 0) + +/* scan through an array of device ids and return the first + * entry that matches the device. + * + * the array must end with an entry containing zero match_flags + */ +static inline const struct ccw_device_id * +ccw_device_id_match(const struct ccw_device_id *array, + const struct ccw_device_id *match) +{ + const struct ccw_device_id *id = array; + + for (id = array; id->match_flags; id++) { + if ((id->match_flags & CCW_DEVICE_ID_MATCH_CU_TYPE) + && (id->cu_type != match->cu_type)) + continue; + + if ((id->match_flags & CCW_DEVICE_ID_MATCH_CU_MODEL) + && (id->cu_model != match->cu_model)) + continue; + + if ((id->match_flags & CCW_DEVICE_ID_MATCH_DEVICE_TYPE) + && (id->dev_type != match->dev_type)) + continue; + + if ((id->match_flags & CCW_DEVICE_ID_MATCH_DEVICE_MODEL) + && (id->dev_model != match->dev_model)) + continue; + + return id; + } + + return NULL; +} + +/** + * struct ccw_device - channel attached device + * @ccwlock: pointer to device lock + * @id: id of this device + * @drv: ccw driver for this device + * @dev: embedded device structure + * @online: online status of device + * @handler: interrupt handler + * + * @handler is a member of the device rather than the driver since a driver + * can have different interrupt handlers for different ccw devices + * (multi-subchannel drivers). + */ +struct ccw_device { + spinlock_t *ccwlock; +/* private: */ + struct ccw_device_private *private; /* cio private information */ +/* public: */ + struct ccw_device_id id; + struct ccw_driver *drv; + struct device dev; + int online; + void (*handler) (struct ccw_device *, unsigned long, struct irb *); +}; + +/* + * Possible events used by the path_event notifier. + */ +#define PE_NONE 0x0 +#define PE_PATH_GONE 0x1 /* A path is no longer available. */ +#define PE_PATH_AVAILABLE 0x2 /* A path has become available and + was successfully verified. */ +#define PE_PATHGROUP_ESTABLISHED 0x4 /* A pathgroup was reset and had + to be established again. */ + +/* + * Possible CIO actions triggered by the unit check handler. + */ +enum uc_todo { + UC_TODO_RETRY, + UC_TODO_RETRY_ON_NEW_PATH, + UC_TODO_STOP +}; + +/** + * struct ccw driver - device driver for channel attached devices + * @ids: ids supported by this driver + * @probe: function called on probe + * @remove: function called on remove + * @set_online: called when setting device online + * @set_offline: called when setting device offline + * @notify: notify driver of device state changes + * @path_event: notify driver of channel path events + * @shutdown: called at device shutdown + * @prepare: prepare for pm state transition + * @complete: undo work done in @prepare + * @freeze: callback for freezing during hibernation snapshotting + * @thaw: undo work done in @freeze + * @restore: callback for restoring after hibernation + * @uc_handler: callback for unit check handler + * @driver: embedded device driver structure + * @int_class: interruption class to use for accounting interrupts + */ +struct ccw_driver { + struct ccw_device_id *ids; + int (*probe) (struct ccw_device *); + void (*remove) (struct ccw_device *); + int (*set_online) (struct ccw_device *); + int (*set_offline) (struct ccw_device *); + int (*notify) (struct ccw_device *, int); + void (*path_event) (struct ccw_device *, int *); + void (*shutdown) (struct ccw_device *); + int (*prepare) (struct ccw_device *); + void (*complete) (struct ccw_device *); + int (*freeze)(struct ccw_device *); + int (*thaw) (struct ccw_device *); + int (*restore)(struct ccw_device *); + enum uc_todo (*uc_handler) (struct ccw_device *, struct irb *); + struct device_driver driver; + enum interruption_class int_class; +}; + +extern struct ccw_device *get_ccwdev_by_busid(struct ccw_driver *cdrv, + const char *bus_id); + +/* devices drivers call these during module load and unload. + * When a driver is registered, its probe method is called + * when new devices for its type pop up */ +extern int ccw_driver_register (struct ccw_driver *driver); +extern void ccw_driver_unregister (struct ccw_driver *driver); + +struct ccw1; + +extern int ccw_device_set_options_mask(struct ccw_device *, unsigned long); +extern int ccw_device_set_options(struct ccw_device *, unsigned long); +extern void ccw_device_clear_options(struct ccw_device *, unsigned long); +int ccw_device_is_pathgroup(struct ccw_device *cdev); +int ccw_device_is_multipath(struct ccw_device *cdev); + +/* Allow for i/o completion notification after primary interrupt status. */ +#define CCWDEV_EARLY_NOTIFICATION 0x0001 +/* Report all interrupt conditions. */ +#define CCWDEV_REPORT_ALL 0x0002 +/* Try to perform path grouping. */ +#define CCWDEV_DO_PATHGROUP 0x0004 +/* Allow forced onlining of boxed devices. */ +#define CCWDEV_ALLOW_FORCE 0x0008 +/* Try to use multipath mode. */ +#define CCWDEV_DO_MULTIPATH 0x0010 + +extern int ccw_device_start(struct ccw_device *, struct ccw1 *, + unsigned long, __u8, unsigned long); +extern int ccw_device_start_timeout(struct ccw_device *, struct ccw1 *, + unsigned long, __u8, unsigned long, int); +extern int ccw_device_start_key(struct ccw_device *, struct ccw1 *, + unsigned long, __u8, __u8, unsigned long); +extern int ccw_device_start_timeout_key(struct ccw_device *, struct ccw1 *, + unsigned long, __u8, __u8, + unsigned long, int); + + +extern int ccw_device_resume(struct ccw_device *); +extern int ccw_device_halt(struct ccw_device *, unsigned long); +extern int ccw_device_clear(struct ccw_device *, unsigned long); +int ccw_device_tm_start_key(struct ccw_device *cdev, struct tcw *tcw, + unsigned long intparm, u8 lpm, u8 key); +int ccw_device_tm_start_key(struct ccw_device *, struct tcw *, + unsigned long, u8, u8); +int ccw_device_tm_start_timeout_key(struct ccw_device *, struct tcw *, + unsigned long, u8, u8, int); +int ccw_device_tm_start(struct ccw_device *, struct tcw *, + unsigned long, u8); +int ccw_device_tm_start_timeout(struct ccw_device *, struct tcw *, + unsigned long, u8, int); +int ccw_device_tm_intrg(struct ccw_device *cdev); + +int ccw_device_get_mdc(struct ccw_device *cdev, u8 mask); + +extern int ccw_device_set_online(struct ccw_device *cdev); +extern int ccw_device_set_offline(struct ccw_device *cdev); + + +extern struct ciw *ccw_device_get_ciw(struct ccw_device *, __u32 cmd); +extern __u8 ccw_device_get_path_mask(struct ccw_device *); +extern void ccw_device_get_id(struct ccw_device *, struct ccw_dev_id *); + +#define get_ccwdev_lock(x) (x)->ccwlock + +#define to_ccwdev(n) container_of(n, struct ccw_device, dev) +#define to_ccwdrv(n) container_of(n, struct ccw_driver, driver) + +extern struct ccw_device *ccw_device_create_console(struct ccw_driver *); +extern void ccw_device_destroy_console(struct ccw_device *); +extern int ccw_device_enable_console(struct ccw_device *); +extern void ccw_device_wait_idle(struct ccw_device *); +extern int ccw_device_force_console(struct ccw_device *); + +int ccw_device_siosl(struct ccw_device *); + +extern void ccw_device_get_schid(struct ccw_device *, struct subchannel_id *); + +struct channel_path_desc *ccw_device_get_chp_desc(struct ccw_device *, int); +#endif /* _S390_CCWDEV_H_ */ diff --git a/arch/s390/include/asm/ccwgroup.h b/arch/s390/include/asm/ccwgroup.h new file mode 100644 index 000000000..057ce0ca6 --- /dev/null +++ b/arch/s390/include/asm/ccwgroup.h @@ -0,0 +1,73 @@ +#ifndef S390_CCWGROUP_H +#define S390_CCWGROUP_H + +struct ccw_device; +struct ccw_driver; + +/** + * struct ccwgroup_device - ccw group device + * @state: online/offline state + * @count: number of attached slave devices + * @dev: embedded device structure + * @cdev: variable number of slave devices, allocated as needed + * @ungroup_work: work to be done when a ccwgroup notifier has action + * type %BUS_NOTIFY_UNBIND_DRIVER + */ +struct ccwgroup_device { + enum { + CCWGROUP_OFFLINE, + CCWGROUP_ONLINE, + } state; +/* private: */ + atomic_t onoff; + struct mutex reg_mutex; +/* public: */ + unsigned int count; + struct device dev; + struct work_struct ungroup_work; + struct ccw_device *cdev[0]; +}; + +/** + * struct ccwgroup_driver - driver for ccw group devices + * @setup: function called during device creation to setup the device + * @remove: function called on remove + * @set_online: function called when device is set online + * @set_offline: function called when device is set offline + * @shutdown: function called when device is shut down + * @prepare: prepare for pm state transition + * @complete: undo work done in @prepare + * @freeze: callback for freezing during hibernation snapshotting + * @thaw: undo work done in @freeze + * @restore: callback for restoring after hibernation + * @driver: embedded driver structure + */ +struct ccwgroup_driver { + int (*setup) (struct ccwgroup_device *); + void (*remove) (struct ccwgroup_device *); + int (*set_online) (struct ccwgroup_device *); + int (*set_offline) (struct ccwgroup_device *); + void (*shutdown)(struct ccwgroup_device *); + int (*prepare) (struct ccwgroup_device *); + void (*complete) (struct ccwgroup_device *); + int (*freeze)(struct ccwgroup_device *); + int (*thaw) (struct ccwgroup_device *); + int (*restore)(struct ccwgroup_device *); + + struct device_driver driver; +}; + +extern int ccwgroup_driver_register (struct ccwgroup_driver *cdriver); +extern void ccwgroup_driver_unregister (struct ccwgroup_driver *cdriver); +int ccwgroup_create_dev(struct device *root, struct ccwgroup_driver *gdrv, + int num_devices, const char *buf); + +extern int ccwgroup_set_online(struct ccwgroup_device *gdev); +extern int ccwgroup_set_offline(struct ccwgroup_device *gdev); + +extern int ccwgroup_probe_ccwdev(struct ccw_device *cdev); +extern void ccwgroup_remove_ccwdev(struct ccw_device *cdev); + +#define to_ccwgroupdev(x) container_of((x), struct ccwgroup_device, dev) +#define to_ccwgroupdrv(x) container_of((x), struct ccwgroup_driver, driver) +#endif diff --git a/arch/s390/include/asm/checksum.h b/arch/s390/include/asm/checksum.h new file mode 100644 index 000000000..740364856 --- /dev/null +++ b/arch/s390/include/asm/checksum.h @@ -0,0 +1,140 @@ +/* + * S390 fast network checksum routines + * + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Ulrich Hild (first version) + * Martin Schwidefsky (heavily optimized CKSM version) + * D.J. Barrow (third attempt) + */ + +#ifndef _S390_CHECKSUM_H +#define _S390_CHECKSUM_H + +#include <asm/uaccess.h> + +/* + * computes the checksum of a memory block at buff, length len, + * and adds in "sum" (32-bit) + * + * returns a 32-bit number suitable for feeding into itself + * or csum_tcpudp_magic + * + * this function must be called with even lengths, except + * for the last fragment, which may be odd + * + * it's best to have buff aligned on a 32-bit boundary + */ +static inline __wsum +csum_partial(const void *buff, int len, __wsum sum) +{ + register unsigned long reg2 asm("2") = (unsigned long) buff; + register unsigned long reg3 asm("3") = (unsigned long) len; + + asm volatile( + "0: cksm %0,%1\n" /* do checksum on longs */ + " jo 0b\n" + : "+d" (sum), "+d" (reg2), "+d" (reg3) : : "cc", "memory"); + return sum; +} + +/* + * the same as csum_partial_copy, but copies from user space. + * + * here even more important to align src and dst on a 32-bit (or even + * better 64-bit) boundary + * + * Copy from userspace and compute checksum. + */ +static inline __wsum +csum_partial_copy_from_user(const void __user *src, void *dst, + int len, __wsum sum, + int *err_ptr) +{ + if (unlikely(copy_from_user(dst, src, len))) + *err_ptr = -EFAULT; + return csum_partial(dst, len, sum); +} + + +static inline __wsum +csum_partial_copy_nocheck (const void *src, void *dst, int len, __wsum sum) +{ + memcpy(dst,src,len); + return csum_partial(dst, len, sum); +} + +/* + * Fold a partial checksum without adding pseudo headers + */ +static inline __sum16 csum_fold(__wsum sum) +{ + u32 csum = (__force u32) sum; + + csum += (csum >> 16) + (csum << 16); + csum >>= 16; + return (__force __sum16) ~csum; +} + +/* + * This is a version of ip_compute_csum() optimized for IP headers, + * which always checksum on 4 octet boundaries. + * + */ +static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) +{ + return csum_fold(csum_partial(iph, ihl*4, 0)); +} + +/* + * computes the checksum of the TCP/UDP pseudo-header + * returns a 32-bit checksum + */ +static inline __wsum +csum_tcpudp_nofold(__be32 saddr, __be32 daddr, + unsigned short len, unsigned short proto, + __wsum sum) +{ + __u32 csum = (__force __u32)sum; + + csum += (__force __u32)saddr; + if (csum < (__force __u32)saddr) + csum++; + + csum += (__force __u32)daddr; + if (csum < (__force __u32)daddr) + csum++; + + csum += len + proto; + if (csum < len + proto) + csum++; + + return (__force __wsum)csum; +} + +/* + * computes the checksum of the TCP/UDP pseudo-header + * returns a 16-bit checksum, already complemented + */ + +static inline __sum16 +csum_tcpudp_magic(__be32 saddr, __be32 daddr, + unsigned short len, unsigned short proto, + __wsum sum) +{ + return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); +} + +/* + * this routine is used for miscellaneous IP-like checksums, mainly + * in icmp.c + */ + +static inline __sum16 ip_compute_csum(const void *buff, int len) +{ + return csum_fold(csum_partial(buff, len, 0)); +} + +#endif /* _S390_CHECKSUM_H */ + + diff --git a/arch/s390/include/asm/chpid.h b/arch/s390/include/asm/chpid.h new file mode 100644 index 000000000..7298eec98 --- /dev/null +++ b/arch/s390/include/asm/chpid.h @@ -0,0 +1,50 @@ +/* + * Copyright IBM Corp. 2007, 2012 + * Author(s): Peter Oberparleiter <peter.oberparleiter@de.ibm.com> + */ +#ifndef _ASM_S390_CHPID_H +#define _ASM_S390_CHPID_H + +#include <uapi/asm/chpid.h> +#include <asm/cio.h> + +struct channel_path_desc { + u8 flags; + u8 lsn; + u8 desc; + u8 chpid; + u8 swla; + u8 zeroes; + u8 chla; + u8 chpp; +} __packed; + +static inline void chp_id_init(struct chp_id *chpid) +{ + memset(chpid, 0, sizeof(struct chp_id)); +} + +static inline int chp_id_is_equal(struct chp_id *a, struct chp_id *b) +{ + return (a->id == b->id) && (a->cssid == b->cssid); +} + +static inline void chp_id_next(struct chp_id *chpid) +{ + if (chpid->id < __MAX_CHPID) + chpid->id++; + else { + chpid->id = 0; + chpid->cssid++; + } +} + +static inline int chp_id_is_valid(struct chp_id *chpid) +{ + return (chpid->cssid <= __MAX_CSSID); +} + + +#define chp_id_for_each(c) \ + for (chp_id_init(c); chp_id_is_valid(c); chp_id_next(c)) +#endif /* _ASM_S390_CHPID_H */ diff --git a/arch/s390/include/asm/cio.h b/arch/s390/include/asm/cio.h new file mode 100644 index 000000000..096339207 --- /dev/null +++ b/arch/s390/include/asm/cio.h @@ -0,0 +1,315 @@ +/* + * Common interface for I/O on S/390 + */ +#ifndef _ASM_S390_CIO_H_ +#define _ASM_S390_CIO_H_ + +#include <linux/spinlock.h> +#include <asm/types.h> + +#define LPM_ANYPATH 0xff +#define __MAX_CSSID 0 +#define __MAX_SUBCHANNEL 65535 +#define __MAX_SSID 3 + +#include <asm/scsw.h> + +/** + * struct ccw1 - channel command word + * @cmd_code: command code + * @flags: flags, like IDA addressing, etc. + * @count: byte count + * @cda: data address + * + * The ccw is the basic structure to build channel programs that perform + * operations with the device or the control unit. Only Format-1 channel + * command words are supported. + */ +struct ccw1 { + __u8 cmd_code; + __u8 flags; + __u16 count; + __u32 cda; +} __attribute__ ((packed,aligned(8))); + +#define CCW_FLAG_DC 0x80 +#define CCW_FLAG_CC 0x40 +#define CCW_FLAG_SLI 0x20 +#define CCW_FLAG_SKIP 0x10 +#define CCW_FLAG_PCI 0x08 +#define CCW_FLAG_IDA 0x04 +#define CCW_FLAG_SUSPEND 0x02 + +#define CCW_CMD_READ_IPL 0x02 +#define CCW_CMD_NOOP 0x03 +#define CCW_CMD_BASIC_SENSE 0x04 +#define CCW_CMD_TIC 0x08 +#define CCW_CMD_STLCK 0x14 +#define CCW_CMD_SENSE_PGID 0x34 +#define CCW_CMD_SUSPEND_RECONN 0x5B +#define CCW_CMD_RDC 0x64 +#define CCW_CMD_RELEASE 0x94 +#define CCW_CMD_SET_PGID 0xAF +#define CCW_CMD_SENSE_ID 0xE4 +#define CCW_CMD_DCTL 0xF3 + +#define SENSE_MAX_COUNT 0x20 + +/** + * struct erw - extended report word + * @res0: reserved + * @auth: authorization check + * @pvrf: path-verification-required flag + * @cpt: channel-path timeout + * @fsavf: failing storage address validity flag + * @cons: concurrent sense + * @scavf: secondary ccw address validity flag + * @fsaf: failing storage address format + * @scnt: sense count, if @cons == %1 + * @res16: reserved + */ +struct erw { + __u32 res0 : 3; + __u32 auth : 1; + __u32 pvrf : 1; + __u32 cpt : 1; + __u32 fsavf : 1; + __u32 cons : 1; + __u32 scavf : 1; + __u32 fsaf : 1; + __u32 scnt : 6; + __u32 res16 : 16; +} __attribute__ ((packed)); + +/** + * struct erw_eadm - EADM Subchannel extended report word + * @b: aob error + * @r: arsb error + */ +struct erw_eadm { + __u32 : 16; + __u32 b : 1; + __u32 r : 1; + __u32 : 14; +} __packed; + +/** + * struct sublog - subchannel logout area + * @res0: reserved + * @esf: extended status flags + * @lpum: last path used mask + * @arep: ancillary report + * @fvf: field-validity flags + * @sacc: storage access code + * @termc: termination code + * @devsc: device-status check + * @serr: secondary error + * @ioerr: i/o-error alert + * @seqc: sequence code + */ +struct sublog { + __u32 res0 : 1; + __u32 esf : 7; + __u32 lpum : 8; + __u32 arep : 1; + __u32 fvf : 5; + __u32 sacc : 2; + __u32 termc : 2; + __u32 devsc : 1; + __u32 serr : 1; + __u32 ioerr : 1; + __u32 seqc : 3; +} __attribute__ ((packed)); + +/** + * struct esw0 - Format 0 Extended Status Word (ESW) + * @sublog: subchannel logout + * @erw: extended report word + * @faddr: failing storage address + * @saddr: secondary ccw address + */ +struct esw0 { + struct sublog sublog; + struct erw erw; + __u32 faddr[2]; + __u32 saddr; +} __attribute__ ((packed)); + +/** + * struct esw1 - Format 1 Extended Status Word (ESW) + * @zero0: reserved zeros + * @lpum: last path used mask + * @zero16: reserved zeros + * @erw: extended report word + * @zeros: three fullwords of zeros + */ +struct esw1 { + __u8 zero0; + __u8 lpum; + __u16 zero16; + struct erw erw; + __u32 zeros[3]; +} __attribute__ ((packed)); + +/** + * struct esw2 - Format 2 Extended Status Word (ESW) + * @zero0: reserved zeros + * @lpum: last path used mask + * @dcti: device-connect-time interval + * @erw: extended report word + * @zeros: three fullwords of zeros + */ +struct esw2 { + __u8 zero0; + __u8 lpum; + __u16 dcti; + struct erw erw; + __u32 zeros[3]; +} __attribute__ ((packed)); + +/** + * struct esw3 - Format 3 Extended Status Word (ESW) + * @zero0: reserved zeros + * @lpum: last path used mask + * @res: reserved + * @erw: extended report word + * @zeros: three fullwords of zeros + */ +struct esw3 { + __u8 zero0; + __u8 lpum; + __u16 res; + struct erw erw; + __u32 zeros[3]; +} __attribute__ ((packed)); + +/** + * struct esw_eadm - EADM Subchannel Extended Status Word (ESW) + * @sublog: subchannel logout + * @erw: extended report word + */ +struct esw_eadm { + __u32 sublog; + struct erw_eadm erw; + __u32 : 32; + __u32 : 32; + __u32 : 32; +} __packed; + +/** + * struct irb - interruption response block + * @scsw: subchannel status word + * @esw: extended status word + * @ecw: extended control word + * + * The irb that is handed to the device driver when an interrupt occurs. For + * solicited interrupts, the common I/O layer already performs checks whether + * a field is valid; a field not being valid is always passed as %0. + * If a unit check occurred, @ecw may contain sense data; this is retrieved + * by the common I/O layer itself if the device doesn't support concurrent + * sense (so that the device driver never needs to perform basic sene itself). + * For unsolicited interrupts, the irb is passed as-is (expect for sense data, + * if applicable). + */ +struct irb { + union scsw scsw; + union { + struct esw0 esw0; + struct esw1 esw1; + struct esw2 esw2; + struct esw3 esw3; + struct esw_eadm eadm; + } esw; + __u8 ecw[32]; +} __attribute__ ((packed,aligned(4))); + +/** + * struct ciw - command information word (CIW) layout + * @et: entry type + * @reserved: reserved bits + * @ct: command type + * @cmd: command code + * @count: command count + */ +struct ciw { + __u32 et : 2; + __u32 reserved : 2; + __u32 ct : 4; + __u32 cmd : 8; + __u32 count : 16; +} __attribute__ ((packed)); + +#define CIW_TYPE_RCD 0x0 /* read configuration data */ +#define CIW_TYPE_SII 0x1 /* set interface identifier */ +#define CIW_TYPE_RNI 0x2 /* read node identifier */ + +/* + * Flags used as input parameters for do_IO() + */ +#define DOIO_ALLOW_SUSPEND 0x0001 /* allow for channel prog. suspend */ +#define DOIO_DENY_PREFETCH 0x0002 /* don't allow for CCW prefetch */ +#define DOIO_SUPPRESS_INTER 0x0004 /* suppress intermediate inter. */ + /* ... for suspended CCWs */ +/* Device or subchannel gone. */ +#define CIO_GONE 0x0001 +/* No path to device. */ +#define CIO_NO_PATH 0x0002 +/* Device has appeared. */ +#define CIO_OPER 0x0004 +/* Sick revalidation of device. */ +#define CIO_REVALIDATE 0x0008 +/* Device did not respond in time. */ +#define CIO_BOXED 0x0010 + +/** + * struct ccw_dev_id - unique identifier for ccw devices + * @ssid: subchannel set id + * @devno: device number + * + * This structure is not directly based on any hardware structure. The + * hardware identifies a device by its device number and its subchannel, + * which is in turn identified by its id. In order to get a unique identifier + * for ccw devices across subchannel sets, @struct ccw_dev_id has been + * introduced. + */ +struct ccw_dev_id { + u8 ssid; + u16 devno; +}; + +/** + * ccw_device_id_is_equal() - compare two ccw_dev_ids + * @dev_id1: a ccw_dev_id + * @dev_id2: another ccw_dev_id + * Returns: + * %1 if the two structures are equal field-by-field, + * %0 if not. + * Context: + * any + */ +static inline int ccw_dev_id_is_equal(struct ccw_dev_id *dev_id1, + struct ccw_dev_id *dev_id2) +{ + if ((dev_id1->ssid == dev_id2->ssid) && + (dev_id1->devno == dev_id2->devno)) + return 1; + return 0; +} + +void channel_subsystem_reinit(void); +extern void css_schedule_reprobe(void); + +extern void reipl_ccw_dev(struct ccw_dev_id *id); + +struct cio_iplinfo { + u16 devno; + int is_qdio; +}; + +extern int cio_get_iplinfo(struct cio_iplinfo *iplinfo); + +/* Function from drivers/s390/cio/chsc.c */ +int chsc_sstpc(void *page, unsigned int op, u16 ctrl); +int chsc_sstpi(void *page, void *result, size_t size); + +#endif diff --git a/arch/s390/include/asm/clp.h b/arch/s390/include/asm/clp.h new file mode 100644 index 000000000..a0e71a501 --- /dev/null +++ b/arch/s390/include/asm/clp.h @@ -0,0 +1,28 @@ +#ifndef _ASM_S390_CLP_H +#define _ASM_S390_CLP_H + +/* CLP common request & response block size */ +#define CLP_BLK_SIZE PAGE_SIZE + +struct clp_req_hdr { + u16 len; + u16 cmd; +} __packed; + +struct clp_rsp_hdr { + u16 len; + u16 rsp; +} __packed; + +/* CLP Response Codes */ +#define CLP_RC_OK 0x0010 /* Command request successfully */ +#define CLP_RC_CMD 0x0020 /* Command code not recognized */ +#define CLP_RC_PERM 0x0030 /* Command not authorized */ +#define CLP_RC_FMT 0x0040 /* Invalid command request format */ +#define CLP_RC_LEN 0x0050 /* Invalid command request length */ +#define CLP_RC_8K 0x0060 /* Command requires 8K LPCB */ +#define CLP_RC_RESNOT0 0x0070 /* Reserved field not zero */ +#define CLP_RC_NODATA 0x0080 /* No data available */ +#define CLP_RC_FC_UNKNOWN 0x0100 /* Function code not recognized */ + +#endif diff --git a/arch/s390/include/asm/cmb.h b/arch/s390/include/asm/cmb.h new file mode 100644 index 000000000..806eac12e --- /dev/null +++ b/arch/s390/include/asm/cmb.h @@ -0,0 +1,12 @@ +#ifndef S390_CMB_H +#define S390_CMB_H + +#include <uapi/asm/cmb.h> + +struct ccw_device; +extern int enable_cmf(struct ccw_device *cdev); +extern int disable_cmf(struct ccw_device *cdev); +extern u64 cmf_read(struct ccw_device *cdev, int index); +extern int cmf_readall(struct ccw_device *cdev, struct cmbdata *data); + +#endif /* S390_CMB_H */ diff --git a/arch/s390/include/asm/cmpxchg.h b/arch/s390/include/asm/cmpxchg.h new file mode 100644 index 000000000..4eadec466 --- /dev/null +++ b/arch/s390/include/asm/cmpxchg.h @@ -0,0 +1,91 @@ +/* + * Copyright IBM Corp. 1999, 2011 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>, + */ + +#ifndef __ASM_CMPXCHG_H +#define __ASM_CMPXCHG_H + +#include <linux/mmdebug.h> +#include <linux/types.h> +#include <linux/bug.h> + +#define cmpxchg(ptr, o, n) \ +({ \ + __typeof__(*(ptr)) __o = (o); \ + __typeof__(*(ptr)) __n = (n); \ + (__typeof__(*(ptr))) __sync_val_compare_and_swap((ptr),__o,__n);\ +}) + +#define cmpxchg64 cmpxchg +#define cmpxchg_local cmpxchg +#define cmpxchg64_local cmpxchg + +#define xchg(ptr, x) \ +({ \ + __typeof__(ptr) __ptr = (ptr); \ + __typeof__(*(ptr)) __old; \ + do { \ + __old = *__ptr; \ + } while (!__sync_bool_compare_and_swap(__ptr, __old, x)); \ + __old; \ +}) + +#define __HAVE_ARCH_CMPXCHG + +#define __cmpxchg_double_op(p1, p2, o1, o2, n1, n2, insn) \ +({ \ + register __typeof__(*(p1)) __old1 asm("2") = (o1); \ + register __typeof__(*(p2)) __old2 asm("3") = (o2); \ + register __typeof__(*(p1)) __new1 asm("4") = (n1); \ + register __typeof__(*(p2)) __new2 asm("5") = (n2); \ + int cc; \ + asm volatile( \ + insn " %[old],%[new],%[ptr]\n" \ + " ipm %[cc]\n" \ + " srl %[cc],28" \ + : [cc] "=d" (cc), [old] "+d" (__old1), "+d" (__old2) \ + : [new] "d" (__new1), "d" (__new2), \ + [ptr] "Q" (*(p1)), "Q" (*(p2)) \ + : "memory", "cc"); \ + !cc; \ +}) + +#define __cmpxchg_double_4(p1, p2, o1, o2, n1, n2) \ + __cmpxchg_double_op(p1, p2, o1, o2, n1, n2, "cds") + +#define __cmpxchg_double_8(p1, p2, o1, o2, n1, n2) \ + __cmpxchg_double_op(p1, p2, o1, o2, n1, n2, "cdsg") + +extern void __cmpxchg_double_called_with_bad_pointer(void); + +#define __cmpxchg_double(p1, p2, o1, o2, n1, n2) \ +({ \ + int __ret; \ + switch (sizeof(*(p1))) { \ + case 4: \ + __ret = __cmpxchg_double_4(p1, p2, o1, o2, n1, n2); \ + break; \ + case 8: \ + __ret = __cmpxchg_double_8(p1, p2, o1, o2, n1, n2); \ + break; \ + default: \ + __cmpxchg_double_called_with_bad_pointer(); \ + } \ + __ret; \ +}) + +#define cmpxchg_double(p1, p2, o1, o2, n1, n2) \ +({ \ + __typeof__(p1) __p1 = (p1); \ + __typeof__(p2) __p2 = (p2); \ + BUILD_BUG_ON(sizeof(*(p1)) != sizeof(long)); \ + BUILD_BUG_ON(sizeof(*(p2)) != sizeof(long)); \ + VM_BUG_ON((unsigned long)((__p1) + 1) != (unsigned long)(__p2));\ + __cmpxchg_double_8(__p1, __p2, o1, o2, n1, n2); \ +}) + +#define system_has_cmpxchg_double() 1 + +#endif /* __ASM_CMPXCHG_H */ diff --git a/arch/s390/include/asm/compat.h b/arch/s390/include/asm/compat.h new file mode 100644 index 000000000..d350ed9d0 --- /dev/null +++ b/arch/s390/include/asm/compat.h @@ -0,0 +1,359 @@ +#ifndef _ASM_S390X_COMPAT_H +#define _ASM_S390X_COMPAT_H +/* + * Architecture specific compatibility types + */ +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/thread_info.h> + +#define __TYPE_IS_PTR(t) (!__builtin_types_compatible_p(typeof(0?(t)0:0ULL), u64)) + +#define __SC_DELOUSE(t,v) ({ \ + BUILD_BUG_ON(sizeof(t) > 4 && !__TYPE_IS_PTR(t)); \ + (t)(__TYPE_IS_PTR(t) ? ((v) & 0x7fffffff) : (v)); \ +}) + +#define PSW32_MASK_PER 0x40000000UL +#define PSW32_MASK_DAT 0x04000000UL +#define PSW32_MASK_IO 0x02000000UL +#define PSW32_MASK_EXT 0x01000000UL +#define PSW32_MASK_KEY 0x00F00000UL +#define PSW32_MASK_BASE 0x00080000UL /* Always one */ +#define PSW32_MASK_MCHECK 0x00040000UL +#define PSW32_MASK_WAIT 0x00020000UL +#define PSW32_MASK_PSTATE 0x00010000UL +#define PSW32_MASK_ASC 0x0000C000UL +#define PSW32_MASK_CC 0x00003000UL +#define PSW32_MASK_PM 0x00000f00UL +#define PSW32_MASK_RI 0x00000080UL + +#define PSW32_MASK_USER 0x0000FF00UL + +#define PSW32_ADDR_AMODE 0x80000000UL +#define PSW32_ADDR_INSN 0x7FFFFFFFUL + +#define PSW32_DEFAULT_KEY (((u32) PAGE_DEFAULT_ACC) << 20) + +#define PSW32_ASC_PRIMARY 0x00000000UL +#define PSW32_ASC_ACCREG 0x00004000UL +#define PSW32_ASC_SECONDARY 0x00008000UL +#define PSW32_ASC_HOME 0x0000C000UL + +#define PSW32_USER_BITS (PSW32_MASK_DAT | PSW32_MASK_IO | PSW32_MASK_EXT | \ + PSW32_DEFAULT_KEY | PSW32_MASK_BASE | \ + PSW32_MASK_MCHECK | PSW32_MASK_PSTATE | \ + PSW32_ASC_PRIMARY) + +#define COMPAT_USER_HZ 100 +#define COMPAT_UTS_MACHINE "s390\0\0\0\0" + +typedef u32 compat_size_t; +typedef s32 compat_ssize_t; +typedef s32 compat_time_t; +typedef s32 compat_clock_t; +typedef s32 compat_pid_t; +typedef u16 __compat_uid_t; +typedef u16 __compat_gid_t; +typedef u32 __compat_uid32_t; +typedef u32 __compat_gid32_t; +typedef u16 compat_mode_t; +typedef u32 compat_ino_t; +typedef u16 compat_dev_t; +typedef s32 compat_off_t; +typedef s64 compat_loff_t; +typedef u16 compat_nlink_t; +typedef u16 compat_ipc_pid_t; +typedef s32 compat_daddr_t; +typedef u32 compat_caddr_t; +typedef __kernel_fsid_t compat_fsid_t; +typedef s32 compat_key_t; +typedef s32 compat_timer_t; + +typedef s32 compat_int_t; +typedef s32 compat_long_t; +typedef s64 compat_s64; +typedef u32 compat_uint_t; +typedef u32 compat_ulong_t; +typedef u64 compat_u64; +typedef u32 compat_uptr_t; + +typedef struct { + u32 mask; + u32 addr; +} __aligned(8) psw_compat_t; + +typedef struct { + psw_compat_t psw; + u32 gprs[NUM_GPRS]; + u32 acrs[NUM_ACRS]; + u32 orig_gpr2; +} s390_compat_regs; + +typedef struct { + u32 gprs_high[NUM_GPRS]; +} s390_compat_regs_high; + +struct compat_timespec { + compat_time_t tv_sec; + s32 tv_nsec; +}; + +struct compat_timeval { + compat_time_t tv_sec; + s32 tv_usec; +}; + +struct compat_stat { + compat_dev_t st_dev; + u16 __pad1; + compat_ino_t st_ino; + compat_mode_t st_mode; + compat_nlink_t st_nlink; + __compat_uid_t st_uid; + __compat_gid_t st_gid; + compat_dev_t st_rdev; + u16 __pad2; + u32 st_size; + u32 st_blksize; + u32 st_blocks; + u32 st_atime; + u32 st_atime_nsec; + u32 st_mtime; + u32 st_mtime_nsec; + u32 st_ctime; + u32 st_ctime_nsec; + u32 __unused4; + u32 __unused5; +}; + +struct compat_flock { + short l_type; + short l_whence; + compat_off_t l_start; + compat_off_t l_len; + compat_pid_t l_pid; +}; + +#define F_GETLK64 12 +#define F_SETLK64 13 +#define F_SETLKW64 14 + +struct compat_flock64 { + short l_type; + short l_whence; + compat_loff_t l_start; + compat_loff_t l_len; + compat_pid_t l_pid; +}; + +struct compat_statfs { + u32 f_type; + u32 f_bsize; + u32 f_blocks; + u32 f_bfree; + u32 f_bavail; + u32 f_files; + u32 f_ffree; + compat_fsid_t f_fsid; + u32 f_namelen; + u32 f_frsize; + u32 f_flags; + u32 f_spare[4]; +}; + +struct compat_statfs64 { + u32 f_type; + u32 f_bsize; + u64 f_blocks; + u64 f_bfree; + u64 f_bavail; + u64 f_files; + u64 f_ffree; + compat_fsid_t f_fsid; + u32 f_namelen; + u32 f_frsize; + u32 f_flags; + u32 f_spare[4]; +}; + +#define COMPAT_RLIM_OLD_INFINITY 0x7fffffff +#define COMPAT_RLIM_INFINITY 0xffffffff + +typedef u32 compat_old_sigset_t; /* at least 32 bits */ + +#define _COMPAT_NSIG 64 +#define _COMPAT_NSIG_BPW 32 + +typedef u32 compat_sigset_word; + +typedef union compat_sigval { + compat_int_t sival_int; + compat_uptr_t sival_ptr; +} compat_sigval_t; + +typedef struct compat_siginfo { + int si_signo; + int si_errno; + int si_code; + + union { + int _pad[128/sizeof(int) - 3]; + + /* kill() */ + struct { + pid_t _pid; /* sender's pid */ + uid_t _uid; /* sender's uid */ + } _kill; + + /* POSIX.1b timers */ + struct { + compat_timer_t _tid; /* timer id */ + int _overrun; /* overrun count */ + compat_sigval_t _sigval; /* same as below */ + int _sys_private; /* not to be passed to user */ + } _timer; + + /* POSIX.1b signals */ + struct { + pid_t _pid; /* sender's pid */ + uid_t _uid; /* sender's uid */ + compat_sigval_t _sigval; + } _rt; + + /* SIGCHLD */ + struct { + pid_t _pid; /* which child */ + uid_t _uid; /* sender's uid */ + int _status;/* exit code */ + compat_clock_t _utime; + compat_clock_t _stime; + } _sigchld; + + /* SIGILL, SIGFPE, SIGSEGV, SIGBUS */ + struct { + __u32 _addr; /* faulting insn/memory ref. - pointer */ + } _sigfault; + + /* SIGPOLL */ + struct { + int _band; /* POLL_IN, POLL_OUT, POLL_MSG */ + int _fd; + } _sigpoll; + } _sifields; +} compat_siginfo_t; + +/* + * How these fields are to be accessed. + */ +#define si_pid _sifields._kill._pid +#define si_uid _sifields._kill._uid +#define si_status _sifields._sigchld._status +#define si_utime _sifields._sigchld._utime +#define si_stime _sifields._sigchld._stime +#define si_value _sifields._rt._sigval +#define si_int _sifields._rt._sigval.sival_int +#define si_ptr _sifields._rt._sigval.sival_ptr +#define si_addr _sifields._sigfault._addr +#define si_band _sifields._sigpoll._band +#define si_fd _sifields._sigpoll._fd +#define si_tid _sifields._timer._tid +#define si_overrun _sifields._timer._overrun + +#define COMPAT_OFF_T_MAX 0x7fffffff +#define COMPAT_LOFF_T_MAX 0x7fffffffffffffffL + +/* + * A pointer passed in from user mode. This should not + * be used for syscall parameters, just declare them + * as pointers because the syscall entry code will have + * appropriately converted them already. + */ + +static inline void __user *compat_ptr(compat_uptr_t uptr) +{ + return (void __user *)(unsigned long)(uptr & 0x7fffffffUL); +} + +static inline compat_uptr_t ptr_to_compat(void __user *uptr) +{ + return (u32)(unsigned long)uptr; +} + +#ifdef CONFIG_COMPAT + +static inline int is_compat_task(void) +{ + return is_32bit_task(); +} + +static inline void __user *arch_compat_alloc_user_space(long len) +{ + unsigned long stack; + + stack = KSTK_ESP(current); + if (is_compat_task()) + stack &= 0x7fffffffUL; + return (void __user *) (stack - len); +} + +#endif + +struct compat_ipc64_perm { + compat_key_t key; + __compat_uid32_t uid; + __compat_gid32_t gid; + __compat_uid32_t cuid; + __compat_gid32_t cgid; + compat_mode_t mode; + unsigned short __pad1; + unsigned short seq; + unsigned short __pad2; + unsigned int __unused1; + unsigned int __unused2; +}; + +struct compat_semid64_ds { + struct compat_ipc64_perm sem_perm; + compat_time_t sem_otime; + compat_ulong_t __pad1; + compat_time_t sem_ctime; + compat_ulong_t __pad2; + compat_ulong_t sem_nsems; + compat_ulong_t __unused1; + compat_ulong_t __unused2; +}; + +struct compat_msqid64_ds { + struct compat_ipc64_perm msg_perm; + compat_time_t msg_stime; + compat_ulong_t __pad1; + compat_time_t msg_rtime; + compat_ulong_t __pad2; + compat_time_t msg_ctime; + compat_ulong_t __pad3; + compat_ulong_t msg_cbytes; + compat_ulong_t msg_qnum; + compat_ulong_t msg_qbytes; + compat_pid_t msg_lspid; + compat_pid_t msg_lrpid; + compat_ulong_t __unused1; + compat_ulong_t __unused2; +}; + +struct compat_shmid64_ds { + struct compat_ipc64_perm shm_perm; + compat_size_t shm_segsz; + compat_time_t shm_atime; + compat_ulong_t __pad1; + compat_time_t shm_dtime; + compat_ulong_t __pad2; + compat_time_t shm_ctime; + compat_ulong_t __pad3; + compat_pid_t shm_cpid; + compat_pid_t shm_lpid; + compat_ulong_t shm_nattch; + compat_ulong_t __unused1; + compat_ulong_t __unused2; +}; +#endif /* _ASM_S390X_COMPAT_H */ diff --git a/arch/s390/include/asm/cpcmd.h b/arch/s390/include/asm/cpcmd.h new file mode 100644 index 000000000..3dfadb5d6 --- /dev/null +++ b/arch/s390/include/asm/cpcmd.h @@ -0,0 +1,32 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + * Christian Borntraeger (cborntra@de.ibm.com), + */ + +#ifndef _ASM_S390_CPCMD_H +#define _ASM_S390_CPCMD_H + +/* + * the lowlevel function for cpcmd + * the caller of __cpcmd has to ensure that the response buffer is below 2 GB + */ +extern int __cpcmd(const char *cmd, char *response, int rlen, int *response_code); + +/* + * cpcmd is the in-kernel interface for issuing CP commands + * + * cmd: null-terminated command string, max 240 characters + * response: response buffer for VM's textual response + * rlen: size of the response buffer, cpcmd will not exceed this size + * but will cap the output, if its too large. Everything that + * did not fit into the buffer will be silently dropped + * response_code: return pointer for VM's error code + * return value: the size of the response. The caller can check if the buffer + * was large enough by comparing the return value and rlen + * NOTE: If the response buffer is not below 2 GB, cpcmd can sleep + */ +extern int cpcmd(const char *cmd, char *response, int rlen, int *response_code); + +#endif /* _ASM_S390_CPCMD_H */ diff --git a/arch/s390/include/asm/cpu.h b/arch/s390/include/asm/cpu.h new file mode 100644 index 000000000..f5a8e2fcd --- /dev/null +++ b/arch/s390/include/asm/cpu.h @@ -0,0 +1,26 @@ +/* + * Copyright IBM Corp. 2000, 2009 + * Author(s): Hartmut Penner <hp@de.ibm.com>, + * Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Christian Ehrhardt <ehrhardt@de.ibm.com>, + */ + +#ifndef _ASM_S390_CPU_H +#define _ASM_S390_CPU_H + +#define MAX_CPU_ADDRESS 255 + +#ifndef __ASSEMBLY__ + +#include <linux/types.h> + +struct cpuid +{ + unsigned int version : 8; + unsigned int ident : 24; + unsigned int machine : 16; + unsigned int unused : 16; +} __attribute__ ((packed, aligned(8))); + +#endif /* __ASSEMBLY__ */ +#endif /* _ASM_S390_CPU_H */ diff --git a/arch/s390/include/asm/cpu_mf.h b/arch/s390/include/asm/cpu_mf.h new file mode 100644 index 000000000..5243a8679 --- /dev/null +++ b/arch/s390/include/asm/cpu_mf.h @@ -0,0 +1,297 @@ +/* + * CPU-measurement facilities + * + * Copyright IBM Corp. 2012 + * Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com> + * Jan Glauber <jang@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. + */ +#ifndef _ASM_S390_CPU_MF_H +#define _ASM_S390_CPU_MF_H + +#include <linux/errno.h> +#include <asm/facility.h> + +#define CPU_MF_INT_SF_IAE (1 << 31) /* invalid entry address */ +#define CPU_MF_INT_SF_ISE (1 << 30) /* incorrect SDBT entry */ +#define CPU_MF_INT_SF_PRA (1 << 29) /* program request alert */ +#define CPU_MF_INT_SF_SACA (1 << 23) /* sampler auth. change alert */ +#define CPU_MF_INT_SF_LSDA (1 << 22) /* loss of sample data alert */ +#define CPU_MF_INT_CF_CACA (1 << 7) /* counter auth. change alert */ +#define CPU_MF_INT_CF_LCDA (1 << 6) /* loss of counter data alert */ +#define CPU_MF_INT_RI_HALTED (1 << 5) /* run-time instr. halted */ +#define CPU_MF_INT_RI_BUF_FULL (1 << 4) /* run-time instr. program + buffer full */ + +#define CPU_MF_INT_CF_MASK (CPU_MF_INT_CF_CACA|CPU_MF_INT_CF_LCDA) +#define CPU_MF_INT_SF_MASK (CPU_MF_INT_SF_IAE|CPU_MF_INT_SF_ISE| \ + CPU_MF_INT_SF_PRA|CPU_MF_INT_SF_SACA| \ + CPU_MF_INT_SF_LSDA) +#define CPU_MF_INT_RI_MASK (CPU_MF_INT_RI_HALTED|CPU_MF_INT_RI_BUF_FULL) + +/* CPU measurement facility support */ +static inline int cpum_cf_avail(void) +{ + return MACHINE_HAS_LPP && test_facility(67); +} + +static inline int cpum_sf_avail(void) +{ + return MACHINE_HAS_LPP && test_facility(68); +} + + +struct cpumf_ctr_info { + u16 cfvn; + u16 auth_ctl; + u16 enable_ctl; + u16 act_ctl; + u16 max_cpu; + u16 csvn; + u16 max_cg; + u16 reserved1; + u32 reserved2[12]; +} __packed; + +/* QUERY SAMPLING INFORMATION block */ +struct hws_qsi_info_block { /* Bit(s) */ + unsigned int b0_13:14; /* 0-13: zeros */ + unsigned int as:1; /* 14: basic-sampling authorization */ + unsigned int ad:1; /* 15: diag-sampling authorization */ + unsigned int b16_21:6; /* 16-21: zeros */ + unsigned int es:1; /* 22: basic-sampling enable control */ + unsigned int ed:1; /* 23: diag-sampling enable control */ + unsigned int b24_29:6; /* 24-29: zeros */ + unsigned int cs:1; /* 30: basic-sampling activation control */ + unsigned int cd:1; /* 31: diag-sampling activation control */ + unsigned int bsdes:16; /* 4-5: size of basic sampling entry */ + unsigned int dsdes:16; /* 6-7: size of diagnostic sampling entry */ + unsigned long min_sampl_rate; /* 8-15: minimum sampling interval */ + unsigned long max_sampl_rate; /* 16-23: maximum sampling interval*/ + unsigned long tear; /* 24-31: TEAR contents */ + unsigned long dear; /* 32-39: DEAR contents */ + unsigned int rsvrd0; /* 40-43: reserved */ + unsigned int cpu_speed; /* 44-47: CPU speed */ + unsigned long long rsvrd1; /* 48-55: reserved */ + unsigned long long rsvrd2; /* 56-63: reserved */ +} __packed; + +/* SET SAMPLING CONTROLS request block */ +struct hws_lsctl_request_block { + unsigned int s:1; /* 0: maximum buffer indicator */ + unsigned int h:1; /* 1: part. level reserved for VM use*/ + unsigned long long b2_53:52;/* 2-53: zeros */ + unsigned int es:1; /* 54: basic-sampling enable control */ + unsigned int ed:1; /* 55: diag-sampling enable control */ + unsigned int b56_61:6; /* 56-61: - zeros */ + unsigned int cs:1; /* 62: basic-sampling activation control */ + unsigned int cd:1; /* 63: diag-sampling activation control */ + unsigned long interval; /* 8-15: sampling interval */ + unsigned long tear; /* 16-23: TEAR contents */ + unsigned long dear; /* 24-31: DEAR contents */ + /* 32-63: */ + unsigned long rsvrd1; /* reserved */ + unsigned long rsvrd2; /* reserved */ + unsigned long rsvrd3; /* reserved */ + unsigned long rsvrd4; /* reserved */ +} __packed; + +struct hws_basic_entry { + unsigned int def:16; /* 0-15 Data Entry Format */ + unsigned int R:4; /* 16-19 reserved */ + unsigned int U:4; /* 20-23 Number of unique instruct. */ + unsigned int z:2; /* zeros */ + unsigned int T:1; /* 26 PSW DAT mode */ + unsigned int W:1; /* 27 PSW wait state */ + unsigned int P:1; /* 28 PSW Problem state */ + unsigned int AS:2; /* 29-30 PSW address-space control */ + unsigned int I:1; /* 31 entry valid or invalid */ + unsigned int:16; + unsigned int prim_asn:16; /* primary ASN */ + unsigned long long ia; /* Instruction Address */ + unsigned long long gpp; /* Guest Program Parameter */ + unsigned long long hpp; /* Host Program Parameter */ +} __packed; + +struct hws_diag_entry { + unsigned int def:16; /* 0-15 Data Entry Format */ + unsigned int R:14; /* 16-19 and 20-30 reserved */ + unsigned int I:1; /* 31 entry valid or invalid */ + u8 data[]; /* Machine-dependent sample data */ +} __packed; + +struct hws_combined_entry { + struct hws_basic_entry basic; /* Basic-sampling data entry */ + struct hws_diag_entry diag; /* Diagnostic-sampling data entry */ +} __packed; + +struct hws_trailer_entry { + union { + struct { + unsigned int f:1; /* 0 - Block Full Indicator */ + unsigned int a:1; /* 1 - Alert request control */ + unsigned int t:1; /* 2 - Timestamp format */ + unsigned long long:61; /* 3 - 63: Reserved */ + }; + unsigned long long flags; /* 0 - 63: All indicators */ + }; + unsigned long long overflow; /* 64 - sample Overflow count */ + unsigned char timestamp[16]; /* 16 - 31 timestamp */ + unsigned long long reserved1; /* 32 -Reserved */ + unsigned long long reserved2; /* */ + unsigned long long progusage1; /* 48 - reserved for programming use */ + unsigned long long progusage2; /* */ +} __packed; + +/* Query counter information */ +static inline int qctri(struct cpumf_ctr_info *info) +{ + int rc = -EINVAL; + + asm volatile ( + "0: .insn s,0xb28e0000,%1\n" + "1: lhi %0,0\n" + "2:\n" + EX_TABLE(1b, 2b) + : "+d" (rc), "=Q" (*info)); + return rc; +} + +/* Load CPU-counter-set controls */ +static inline int lcctl(u64 ctl) +{ + int cc; + + asm volatile ( + " .insn s,0xb2840000,%1\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (cc) : "m" (ctl) : "cc"); + return cc; +} + +/* Extract CPU counter */ +static inline int ecctr(u64 ctr, u64 *val) +{ + register u64 content asm("4") = 0; + int cc; + + asm volatile ( + " .insn rre,0xb2e40000,%0,%2\n" + " ipm %1\n" + " srl %1,28\n" + : "=d" (content), "=d" (cc) : "d" (ctr) : "cc"); + if (!cc) + *val = content; + return cc; +} + +/* Store CPU counter multiple for the MT utilization counter set */ +static inline int stcctm5(u64 num, u64 *val) +{ + typedef struct { u64 _[num]; } addrtype; + int cc; + + asm volatile ( + " .insn rsy,0xeb0000000017,%2,5,%1\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (cc), "=Q" (*(addrtype *) val) : "d" (num) : "cc"); + return cc; +} + +/* Query sampling information */ +static inline int qsi(struct hws_qsi_info_block *info) +{ + int cc; + cc = 1; + + asm volatile( + "0: .insn s,0xb2860000,0(%1)\n" + "1: lhi %0,0\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "=d" (cc), "+a" (info) + : "m" (*info) + : "cc", "memory"); + + return cc ? -EINVAL : 0; +} + +/* Load sampling controls */ +static inline int lsctl(struct hws_lsctl_request_block *req) +{ + int cc; + + cc = 1; + asm volatile( + "0: .insn s,0xb2870000,0(%1)\n" + "1: ipm %0\n" + " srl %0,28\n" + "2:\n" + EX_TABLE(0b, 2b) EX_TABLE(1b, 2b) + : "+d" (cc), "+a" (req) + : "m" (*req) + : "cc", "memory"); + + return cc ? -EINVAL : 0; +} + +/* Sampling control helper functions */ + +#include <linux/time.h> + +static inline unsigned long freq_to_sample_rate(struct hws_qsi_info_block *qsi, + unsigned long freq) +{ + return (USEC_PER_SEC / freq) * qsi->cpu_speed; +} + +static inline unsigned long sample_rate_to_freq(struct hws_qsi_info_block *qsi, + unsigned long rate) +{ + return USEC_PER_SEC * qsi->cpu_speed / rate; +} + +#define SDB_TE_ALERT_REQ_MASK 0x4000000000000000UL +#define SDB_TE_BUFFER_FULL_MASK 0x8000000000000000UL + +/* Return TOD timestamp contained in an trailer entry */ +static inline unsigned long long trailer_timestamp(struct hws_trailer_entry *te) +{ + /* TOD in STCKE format */ + if (te->t) + return *((unsigned long long *) &te->timestamp[1]); + + /* TOD in STCK format */ + return *((unsigned long long *) &te->timestamp[0]); +} + +/* Return pointer to trailer entry of an sample data block */ +static inline unsigned long *trailer_entry_ptr(unsigned long v) +{ + void *ret; + + ret = (void *) v; + ret += PAGE_SIZE; + ret -= sizeof(struct hws_trailer_entry); + + return (unsigned long *) ret; +} + +/* Return if the entry in the sample data block table (sdbt) + * is a link to the next sdbt */ +static inline int is_link_entry(unsigned long *s) +{ + return *s & 0x1ul ? 1 : 0; +} + +/* Return pointer to the linked sdbt */ +static inline unsigned long *get_next_sdbt(unsigned long *s) +{ + return (unsigned long *) (*s & ~0x1ul); +} +#endif /* _ASM_S390_CPU_MF_H */ diff --git a/arch/s390/include/asm/cputime.h b/arch/s390/include/asm/cputime.h new file mode 100644 index 000000000..221b454c7 --- /dev/null +++ b/arch/s390/include/asm/cputime.h @@ -0,0 +1,148 @@ +/* + * Copyright IBM Corp. 2004 + * + * Author: Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef _S390_CPUTIME_H +#define _S390_CPUTIME_H + +#include <linux/types.h> +#include <asm/div64.h> + +#define CPUTIME_PER_USEC 4096ULL +#define CPUTIME_PER_SEC (CPUTIME_PER_USEC * USEC_PER_SEC) + +/* We want to use full resolution of the CPU timer: 2**-12 micro-seconds. */ + +typedef unsigned long long __nocast cputime_t; +typedef unsigned long long __nocast cputime64_t; + +#define cmpxchg_cputime(ptr, old, new) cmpxchg64(ptr, old, new) + +static inline unsigned long __div(unsigned long long n, unsigned long base) +{ + return n / base; +} + +#define cputime_one_jiffy jiffies_to_cputime(1) + +/* + * Convert cputime to jiffies and back. + */ +static inline unsigned long cputime_to_jiffies(const cputime_t cputime) +{ + return __div((__force unsigned long long) cputime, CPUTIME_PER_SEC / HZ); +} + +static inline cputime_t jiffies_to_cputime(const unsigned int jif) +{ + return (__force cputime_t)(jif * (CPUTIME_PER_SEC / HZ)); +} + +static inline u64 cputime64_to_jiffies64(cputime64_t cputime) +{ + unsigned long long jif = (__force unsigned long long) cputime; + do_div(jif, CPUTIME_PER_SEC / HZ); + return jif; +} + +static inline cputime64_t jiffies64_to_cputime64(const u64 jif) +{ + return (__force cputime64_t)(jif * (CPUTIME_PER_SEC / HZ)); +} + +/* + * Convert cputime to microseconds and back. + */ +static inline unsigned int cputime_to_usecs(const cputime_t cputime) +{ + return (__force unsigned long long) cputime >> 12; +} + +static inline cputime_t usecs_to_cputime(const unsigned int m) +{ + return (__force cputime_t)(m * CPUTIME_PER_USEC); +} + +#define usecs_to_cputime64(m) usecs_to_cputime(m) + +/* + * Convert cputime to milliseconds and back. + */ +static inline unsigned int cputime_to_secs(const cputime_t cputime) +{ + return __div((__force unsigned long long) cputime, CPUTIME_PER_SEC / 2) >> 1; +} + +static inline cputime_t secs_to_cputime(const unsigned int s) +{ + return (__force cputime_t)(s * CPUTIME_PER_SEC); +} + +/* + * Convert cputime to timespec and back. + */ +static inline cputime_t timespec_to_cputime(const struct timespec *value) +{ + unsigned long long ret = value->tv_sec * CPUTIME_PER_SEC; + return (__force cputime_t)(ret + __div(value->tv_nsec * CPUTIME_PER_USEC, NSEC_PER_USEC)); +} + +static inline void cputime_to_timespec(const cputime_t cputime, + struct timespec *value) +{ + unsigned long long __cputime = (__force unsigned long long) cputime; + value->tv_nsec = (__cputime % CPUTIME_PER_SEC) * NSEC_PER_USEC / CPUTIME_PER_USEC; + value->tv_sec = __cputime / CPUTIME_PER_SEC; +} + +/* + * Convert cputime to timeval and back. + * Since cputime and timeval have the same resolution (microseconds) + * this is easy. + */ +static inline cputime_t timeval_to_cputime(const struct timeval *value) +{ + unsigned long long ret = value->tv_sec * CPUTIME_PER_SEC; + return (__force cputime_t)(ret + value->tv_usec * CPUTIME_PER_USEC); +} + +static inline void cputime_to_timeval(const cputime_t cputime, + struct timeval *value) +{ + unsigned long long __cputime = (__force unsigned long long) cputime; + value->tv_usec = (__cputime % CPUTIME_PER_SEC) / CPUTIME_PER_USEC; + value->tv_sec = __cputime / CPUTIME_PER_SEC; +} + +/* + * Convert cputime to clock and back. + */ +static inline clock_t cputime_to_clock_t(cputime_t cputime) +{ + unsigned long long clock = (__force unsigned long long) cputime; + do_div(clock, CPUTIME_PER_SEC / USER_HZ); + return clock; +} + +static inline cputime_t clock_t_to_cputime(unsigned long x) +{ + return (__force cputime_t)(x * (CPUTIME_PER_SEC / USER_HZ)); +} + +/* + * Convert cputime64 to clock. + */ +static inline clock_t cputime64_to_clock_t(cputime64_t cputime) +{ + unsigned long long clock = (__force unsigned long long) cputime; + do_div(clock, CPUTIME_PER_SEC / USER_HZ); + return clock; +} + +cputime64_t arch_cpu_idle_time(int cpu); + +#define arch_idle_time(cpu) arch_cpu_idle_time(cpu) + +#endif /* _S390_CPUTIME_H */ diff --git a/arch/s390/include/asm/crw.h b/arch/s390/include/asm/crw.h new file mode 100644 index 000000000..7c31d3e25 --- /dev/null +++ b/arch/s390/include/asm/crw.h @@ -0,0 +1,69 @@ +/* + * Data definitions for channel report processing + * Copyright IBM Corp. 2000, 2009 + * Author(s): Ingo Adlung <adlung@de.ibm.com>, + * Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Cornelia Huck <cornelia.huck@de.ibm.com>, + * Heiko Carstens <heiko.carstens@de.ibm.com>, + */ + +#ifndef _ASM_S390_CRW_H +#define _ASM_S390_CRW_H + +#include <linux/types.h> + +/* + * Channel Report Word + */ +struct crw { + __u32 res1 : 1; /* reserved zero */ + __u32 slct : 1; /* solicited */ + __u32 oflw : 1; /* overflow */ + __u32 chn : 1; /* chained */ + __u32 rsc : 4; /* reporting source code */ + __u32 anc : 1; /* ancillary report */ + __u32 res2 : 1; /* reserved zero */ + __u32 erc : 6; /* error-recovery code */ + __u32 rsid : 16; /* reporting-source ID */ +} __attribute__ ((packed)); + +typedef void (*crw_handler_t)(struct crw *, struct crw *, int); + +extern int crw_register_handler(int rsc, crw_handler_t handler); +extern void crw_unregister_handler(int rsc); +extern void crw_handle_channel_report(void); +void crw_wait_for_channel_report(void); + +#define NR_RSCS 16 + +#define CRW_RSC_MONITOR 0x2 /* monitoring facility */ +#define CRW_RSC_SCH 0x3 /* subchannel */ +#define CRW_RSC_CPATH 0x4 /* channel path */ +#define CRW_RSC_CONFIG 0x9 /* configuration-alert facility */ +#define CRW_RSC_CSS 0xB /* channel subsystem */ + +#define CRW_ERC_EVENT 0x00 /* event information pending */ +#define CRW_ERC_AVAIL 0x01 /* available */ +#define CRW_ERC_INIT 0x02 /* initialized */ +#define CRW_ERC_TERROR 0x03 /* temporary error */ +#define CRW_ERC_IPARM 0x04 /* installed parm initialized */ +#define CRW_ERC_TERM 0x05 /* terminal */ +#define CRW_ERC_PERRN 0x06 /* perm. error, fac. not init */ +#define CRW_ERC_PERRI 0x07 /* perm. error, facility init */ +#define CRW_ERC_PMOD 0x08 /* installed parameters modified */ + +static inline int stcrw(struct crw *pcrw) +{ + int ccode; + + asm volatile( + " stcrw 0(%2)\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (ccode), "=m" (*pcrw) + : "a" (pcrw) + : "cc" ); + return ccode; +} + +#endif /* _ASM_S390_CRW_H */ diff --git a/arch/s390/include/asm/css_chars.h b/arch/s390/include/asm/css_chars.h new file mode 100644 index 000000000..09d1dd46b --- /dev/null +++ b/arch/s390/include/asm/css_chars.h @@ -0,0 +1,38 @@ +#ifndef _ASM_CSS_CHARS_H +#define _ASM_CSS_CHARS_H + +#include <linux/types.h> + +struct css_general_char { + u64 : 12; + u32 dynio : 1; /* bit 12 */ + u32 : 4; + u32 eadm : 1; /* bit 17 */ + u32 : 23; + u32 aif : 1; /* bit 41 */ + u32 : 3; + u32 mcss : 1; /* bit 45 */ + u32 fcs : 1; /* bit 46 */ + u32 : 1; + u32 ext_mb : 1; /* bit 48 */ + u32 : 7; + u32 aif_tdd : 1; /* bit 56 */ + u32 : 1; + u32 qebsm : 1; /* bit 58 */ + u32 : 8; + u32 aif_osa : 1; /* bit 67 */ + u32 : 12; + u32 eadm_rf : 1; /* bit 80 */ + u32 : 1; + u32 cib : 1; /* bit 82 */ + u32 : 5; + u32 fcx : 1; /* bit 88 */ + u32 : 19; + u32 alt_ssi : 1; /* bit 108 */ + u32:1; + u32 narf:1; /* bit 110 */ +} __packed; + +extern struct css_general_char css_general_characteristics; + +#endif diff --git a/arch/s390/include/asm/ctl_reg.h b/arch/s390/include/asm/ctl_reg.h new file mode 100644 index 000000000..cfad7fca0 --- /dev/null +++ b/arch/s390/include/asm/ctl_reg.h @@ -0,0 +1,72 @@ +/* + * Copyright IBM Corp. 1999, 2009 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef __ASM_CTL_REG_H +#define __ASM_CTL_REG_H + +#include <linux/bug.h> + +#define __ctl_load(array, low, high) { \ + typedef struct { char _[sizeof(array)]; } addrtype; \ + \ + BUILD_BUG_ON(sizeof(addrtype) != (high - low + 1) * sizeof(long));\ + asm volatile( \ + " lctlg %1,%2,%0\n" \ + : : "Q" (*(addrtype *)(&array)), "i" (low), "i" (high));\ +} + +#define __ctl_store(array, low, high) { \ + typedef struct { char _[sizeof(array)]; } addrtype; \ + \ + BUILD_BUG_ON(sizeof(addrtype) != (high - low + 1) * sizeof(long));\ + asm volatile( \ + " stctg %1,%2,%0\n" \ + : "=Q" (*(addrtype *)(&array)) \ + : "i" (low), "i" (high)); \ +} + +static inline void __ctl_set_bit(unsigned int cr, unsigned int bit) +{ + unsigned long reg; + + __ctl_store(reg, cr, cr); + reg |= 1UL << bit; + __ctl_load(reg, cr, cr); +} + +static inline void __ctl_clear_bit(unsigned int cr, unsigned int bit) +{ + unsigned long reg; + + __ctl_store(reg, cr, cr); + reg &= ~(1UL << bit); + __ctl_load(reg, cr, cr); +} + +void smp_ctl_set_bit(int cr, int bit); +void smp_ctl_clear_bit(int cr, int bit); + +union ctlreg0 { + unsigned long val; + struct { + unsigned long : 32; + unsigned long : 3; + unsigned long lap : 1; /* Low-address-protection control */ + unsigned long : 4; + unsigned long edat : 1; /* Enhanced-DAT-enablement control */ + unsigned long : 23; + }; +}; + +#ifdef CONFIG_SMP +# define ctl_set_bit(cr, bit) smp_ctl_set_bit(cr, bit) +# define ctl_clear_bit(cr, bit) smp_ctl_clear_bit(cr, bit) +#else +# define ctl_set_bit(cr, bit) __ctl_set_bit(cr, bit) +# define ctl_clear_bit(cr, bit) __ctl_clear_bit(cr, bit) +#endif + +#endif /* __ASM_CTL_REG_H */ diff --git a/arch/s390/include/asm/current.h b/arch/s390/include/asm/current.h new file mode 100644 index 000000000..b80941f30 --- /dev/null +++ b/arch/s390/include/asm/current.h @@ -0,0 +1,18 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "include/asm-i386/current.h" + */ + +#ifndef _S390_CURRENT_H +#define _S390_CURRENT_H + +#include <asm/lowcore.h> + +struct task_struct; + +#define current ((struct task_struct *const)S390_lowcore.current_task) + +#endif /* !(_S390_CURRENT_H) */ diff --git a/arch/s390/include/asm/debug.h b/arch/s390/include/asm/debug.h new file mode 100644 index 000000000..0206c8052 --- /dev/null +++ b/arch/s390/include/asm/debug.h @@ -0,0 +1,262 @@ +/* + * S/390 debug facility + * + * Copyright IBM Corp. 1999, 2000 + */ +#ifndef DEBUG_H +#define DEBUG_H + +#include <linux/string.h> +#include <linux/spinlock.h> +#include <linux/kernel.h> +#include <linux/time.h> +#include <uapi/asm/debug.h> + +#define DEBUG_MAX_LEVEL 6 /* debug levels range from 0 to 6 */ +#define DEBUG_OFF_LEVEL -1 /* level where debug is switched off */ +#define DEBUG_FLUSH_ALL -1 /* parameter to flush all areas */ +#define DEBUG_MAX_VIEWS 10 /* max number of views in proc fs */ +#define DEBUG_MAX_NAME_LEN 64 /* max length for a debugfs file name */ +#define DEBUG_DEFAULT_LEVEL 3 /* initial debug level */ + +#define DEBUG_DIR_ROOT "s390dbf" /* name of debug root directory in proc fs */ + +#define DEBUG_DATA(entry) (char*)(entry + 1) /* data is stored behind */ + /* the entry information */ + +typedef struct __debug_entry debug_entry_t; + +struct debug_view; + +typedef struct debug_info { + struct debug_info* next; + struct debug_info* prev; + atomic_t ref_count; + spinlock_t lock; + int level; + int nr_areas; + int pages_per_area; + int buf_size; + int entry_size; + debug_entry_t*** areas; + int active_area; + int *active_pages; + int *active_entries; + struct dentry* debugfs_root_entry; + struct dentry* debugfs_entries[DEBUG_MAX_VIEWS]; + struct debug_view* views[DEBUG_MAX_VIEWS]; + char name[DEBUG_MAX_NAME_LEN]; + umode_t mode; +} debug_info_t; + +typedef int (debug_header_proc_t) (debug_info_t* id, + struct debug_view* view, + int area, + debug_entry_t* entry, + char* out_buf); + +typedef int (debug_format_proc_t) (debug_info_t* id, + struct debug_view* view, char* out_buf, + const char* in_buf); +typedef int (debug_prolog_proc_t) (debug_info_t* id, + struct debug_view* view, + char* out_buf); +typedef int (debug_input_proc_t) (debug_info_t* id, + struct debug_view* view, + struct file* file, + const char __user *user_buf, + size_t in_buf_size, loff_t* offset); + +int debug_dflt_header_fn(debug_info_t* id, struct debug_view* view, + int area, debug_entry_t* entry, char* out_buf); + +struct debug_view { + char name[DEBUG_MAX_NAME_LEN]; + debug_prolog_proc_t* prolog_proc; + debug_header_proc_t* header_proc; + debug_format_proc_t* format_proc; + debug_input_proc_t* input_proc; + void* private_data; +}; + +extern struct debug_view debug_hex_ascii_view; +extern struct debug_view debug_raw_view; +extern struct debug_view debug_sprintf_view; + +/* do NOT use the _common functions */ + +debug_entry_t* debug_event_common(debug_info_t* id, int level, + const void* data, int length); + +debug_entry_t* debug_exception_common(debug_info_t* id, int level, + const void* data, int length); + +/* Debug Feature API: */ + +debug_info_t *debug_register(const char *name, int pages, int nr_areas, + int buf_size); + +debug_info_t *debug_register_mode(const char *name, int pages, int nr_areas, + int buf_size, umode_t mode, uid_t uid, + gid_t gid); + +void debug_unregister(debug_info_t* id); + +void debug_set_level(debug_info_t* id, int new_level); + +void debug_set_critical(void); +void debug_stop_all(void); + +static inline bool debug_level_enabled(debug_info_t* id, int level) +{ + return level <= id->level; +} + +static inline debug_entry_t* +debug_event(debug_info_t* id, int level, void* data, int length) +{ + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; + return debug_event_common(id,level,data,length); +} + +static inline debug_entry_t* +debug_int_event(debug_info_t* id, int level, unsigned int tag) +{ + unsigned int t=tag; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; + return debug_event_common(id,level,&t,sizeof(unsigned int)); +} + +static inline debug_entry_t * +debug_long_event (debug_info_t* id, int level, unsigned long tag) +{ + unsigned long t=tag; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; + return debug_event_common(id,level,&t,sizeof(unsigned long)); +} + +static inline debug_entry_t* +debug_text_event(debug_info_t* id, int level, const char* txt) +{ + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; + return debug_event_common(id,level,txt,strlen(txt)); +} + +/* + * IMPORTANT: Use "%s" in sprintf format strings with care! Only pointers are + * stored in the s390dbf. See Documentation/s390/s390dbf.txt for more details! + */ +extern debug_entry_t * +__debug_sprintf_event(debug_info_t *id, int level, char *string, ...) + __attribute__ ((format(printf, 3, 4))); + +#define debug_sprintf_event(_id, _level, _fmt, ...) \ +({ \ + debug_entry_t *__ret; \ + debug_info_t *__id = _id; \ + int __level = _level; \ + if ((!__id) || (__level > __id->level)) \ + __ret = NULL; \ + else \ + __ret = __debug_sprintf_event(__id, __level, \ + _fmt, ## __VA_ARGS__); \ + __ret; \ +}) + +static inline debug_entry_t* +debug_exception(debug_info_t* id, int level, void* data, int length) +{ + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; + return debug_exception_common(id,level,data,length); +} + +static inline debug_entry_t* +debug_int_exception(debug_info_t* id, int level, unsigned int tag) +{ + unsigned int t=tag; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; + return debug_exception_common(id,level,&t,sizeof(unsigned int)); +} + +static inline debug_entry_t * +debug_long_exception (debug_info_t* id, int level, unsigned long tag) +{ + unsigned long t=tag; + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; + return debug_exception_common(id,level,&t,sizeof(unsigned long)); +} + +static inline debug_entry_t* +debug_text_exception(debug_info_t* id, int level, const char* txt) +{ + if ((!id) || (level > id->level) || (id->pages_per_area == 0)) + return NULL; + return debug_exception_common(id,level,txt,strlen(txt)); +} + +/* + * IMPORTANT: Use "%s" in sprintf format strings with care! Only pointers are + * stored in the s390dbf. See Documentation/s390/s390dbf.txt for more details! + */ +extern debug_entry_t * +__debug_sprintf_exception(debug_info_t *id, int level, char *string, ...) + __attribute__ ((format(printf, 3, 4))); + +#define debug_sprintf_exception(_id, _level, _fmt, ...) \ +({ \ + debug_entry_t *__ret; \ + debug_info_t *__id = _id; \ + int __level = _level; \ + if ((!__id) || (__level > __id->level)) \ + __ret = NULL; \ + else \ + __ret = __debug_sprintf_exception(__id, __level, \ + _fmt, ## __VA_ARGS__);\ + __ret; \ +}) + +int debug_register_view(debug_info_t* id, struct debug_view* view); +int debug_unregister_view(debug_info_t* id, struct debug_view* view); + +/* + define the debug levels: + - 0 No debugging output to console or syslog + - 1 Log internal errors to syslog, ignore check conditions + - 2 Log internal errors and check conditions to syslog + - 3 Log internal errors to console, log check conditions to syslog + - 4 Log internal errors and check conditions to console + - 5 panic on internal errors, log check conditions to console + - 6 panic on both, internal errors and check conditions + */ + +#ifndef DEBUG_LEVEL +#define DEBUG_LEVEL 4 +#endif + +#define INTERNAL_ERRMSG(x,y...) "E" __FILE__ "%d: " x, __LINE__, y +#define INTERNAL_WRNMSG(x,y...) "W" __FILE__ "%d: " x, __LINE__, y +#define INTERNAL_INFMSG(x,y...) "I" __FILE__ "%d: " x, __LINE__, y +#define INTERNAL_DEBMSG(x,y...) "D" __FILE__ "%d: " x, __LINE__, y + +#if DEBUG_LEVEL > 0 +#define PRINT_DEBUG(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) +#define PRINT_INFO(x...) printk ( KERN_INFO PRINTK_HEADER x ) +#define PRINT_WARN(x...) printk ( KERN_WARNING PRINTK_HEADER x ) +#define PRINT_ERR(x...) printk ( KERN_ERR PRINTK_HEADER x ) +#define PRINT_FATAL(x...) panic ( PRINTK_HEADER x ) +#else +#define PRINT_DEBUG(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) +#define PRINT_INFO(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) +#define PRINT_WARN(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) +#define PRINT_ERR(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) +#define PRINT_FATAL(x...) printk ( KERN_DEBUG PRINTK_HEADER x ) +#endif /* DASD_DEBUG */ + +#endif /* DEBUG_H */ diff --git a/arch/s390/include/asm/delay.h b/arch/s390/include/asm/delay.h new file mode 100644 index 000000000..3f6e4095f --- /dev/null +++ b/arch/s390/include/asm/delay.h @@ -0,0 +1,24 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "include/asm-i386/delay.h" + * Copyright (C) 1993 Linus Torvalds + * + * Delay routines calling functions in arch/s390/lib/delay.c + */ + +#ifndef _S390_DELAY_H +#define _S390_DELAY_H + +void __ndelay(unsigned long long nsecs); +void __udelay(unsigned long long usecs); +void udelay_simple(unsigned long long usecs); +void __delay(unsigned long loops); + +#define ndelay(n) __ndelay((unsigned long long) (n)) +#define udelay(n) __udelay((unsigned long long) (n)) +#define mdelay(n) __udelay((unsigned long long) (n) * 1000) + +#endif /* defined(_S390_DELAY_H) */ diff --git a/arch/s390/include/asm/device.h b/arch/s390/include/asm/device.h new file mode 100644 index 000000000..d8f9872b0 --- /dev/null +++ b/arch/s390/include/asm/device.h @@ -0,0 +1,7 @@ +/* + * Arch specific extensions to struct device + * + * This file is released under the GPLv2 + */ +#include <asm-generic/device.h> + diff --git a/arch/s390/include/asm/diag.h b/arch/s390/include/asm/diag.h new file mode 100644 index 000000000..7e91c5807 --- /dev/null +++ b/arch/s390/include/asm/diag.h @@ -0,0 +1,52 @@ +/* + * s390 diagnose functions + * + * Copyright IBM Corp. 2007 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#ifndef _ASM_S390_DIAG_H +#define _ASM_S390_DIAG_H + +/* + * Diagnose 10: Release page range + */ +static inline void diag10_range(unsigned long start_pfn, unsigned long num_pfn) +{ + unsigned long start_addr, end_addr; + + start_addr = start_pfn << PAGE_SHIFT; + end_addr = (start_pfn + num_pfn - 1) << PAGE_SHIFT; + + asm volatile( + "0: diag %0,%1,0x10\n" + "1:\n" + EX_TABLE(0b, 1b) + EX_TABLE(1b, 1b) + : : "a" (start_addr), "a" (end_addr)); +} + +/* + * Diagnose 14: Input spool file manipulation + */ +extern int diag14(unsigned long rx, unsigned long ry1, unsigned long subcode); + +/* + * Diagnose 210: Get information about a virtual device + */ +struct diag210 { + u16 vrdcdvno; /* device number (input) */ + u16 vrdclen; /* data block length (input) */ + u8 vrdcvcla; /* virtual device class (output) */ + u8 vrdcvtyp; /* virtual device type (output) */ + u8 vrdcvsta; /* virtual device status (output) */ + u8 vrdcvfla; /* virtual device flags (output) */ + u8 vrdcrccl; /* real device class (output) */ + u8 vrdccrty; /* real device type (output) */ + u8 vrdccrmd; /* real device model (output) */ + u8 vrdccrft; /* real device feature (output) */ +} __attribute__((packed, aligned(4))); + +extern int diag210(struct diag210 *addr); + +#endif /* _ASM_S390_DIAG_H */ diff --git a/arch/s390/include/asm/dis.h b/arch/s390/include/asm/dis.h new file mode 100644 index 000000000..60323c219 --- /dev/null +++ b/arch/s390/include/asm/dis.h @@ -0,0 +1,53 @@ +/* + * Disassemble s390 instructions. + * + * Copyright IBM Corp. 2007 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + */ + +#ifndef __ASM_S390_DIS_H__ +#define __ASM_S390_DIS_H__ + +/* Type of operand */ +#define OPERAND_GPR 0x1 /* Operand printed as %rx */ +#define OPERAND_FPR 0x2 /* Operand printed as %fx */ +#define OPERAND_AR 0x4 /* Operand printed as %ax */ +#define OPERAND_CR 0x8 /* Operand printed as %cx */ +#define OPERAND_VR 0x10 /* Operand printed as %vx */ +#define OPERAND_DISP 0x20 /* Operand printed as displacement */ +#define OPERAND_BASE 0x40 /* Operand printed as base register */ +#define OPERAND_INDEX 0x80 /* Operand printed as index register */ +#define OPERAND_PCREL 0x100 /* Operand printed as pc-relative symbol */ +#define OPERAND_SIGNED 0x200 /* Operand printed as signed value */ +#define OPERAND_LENGTH 0x400 /* Operand printed as length (+1) */ + + +struct s390_operand { + int bits; /* The number of bits in the operand. */ + int shift; /* The number of bits to shift. */ + int flags; /* One bit syntax flags. */ +}; + +struct s390_insn { + const char name[5]; + unsigned char opfrag; + unsigned char format; +}; + + +static inline int insn_length(unsigned char code) +{ + return ((((int) code + 64) >> 7) + 1) << 1; +} + +void show_code(struct pt_regs *regs); +void print_fn_code(unsigned char *code, unsigned long len); +int insn_to_mnemonic(unsigned char *instruction, char *buf, unsigned int len); +struct s390_insn *find_insn(unsigned char *code); + +static inline int is_known_insn(unsigned char *code) +{ + return !!find_insn(code); +} + +#endif /* __ASM_S390_DIS_H__ */ diff --git a/arch/s390/include/asm/div64.h b/arch/s390/include/asm/div64.h new file mode 100644 index 000000000..6cd978cef --- /dev/null +++ b/arch/s390/include/asm/div64.h @@ -0,0 +1 @@ +#include <asm-generic/div64.h> diff --git a/arch/s390/include/asm/dma-mapping.h b/arch/s390/include/asm/dma-mapping.h new file mode 100644 index 000000000..9d395961e --- /dev/null +++ b/arch/s390/include/asm/dma-mapping.h @@ -0,0 +1,90 @@ +#ifndef _ASM_S390_DMA_MAPPING_H +#define _ASM_S390_DMA_MAPPING_H + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/mm.h> +#include <linux/scatterlist.h> +#include <linux/dma-attrs.h> +#include <linux/dma-debug.h> +#include <linux/io.h> + +#define DMA_ERROR_CODE (~(dma_addr_t) 0x0) + +extern struct dma_map_ops s390_dma_ops; + +static inline struct dma_map_ops *get_dma_ops(struct device *dev) +{ + return &s390_dma_ops; +} + +extern int dma_set_mask(struct device *dev, u64 mask); + +static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size, + enum dma_data_direction direction) +{ +} + +#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f) +#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) + +#include <asm-generic/dma-mapping-common.h> + +static inline int dma_supported(struct device *dev, u64 mask) +{ + struct dma_map_ops *dma_ops = get_dma_ops(dev); + + if (dma_ops->dma_supported == NULL) + return 1; + return dma_ops->dma_supported(dev, mask); +} + +static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size) +{ + if (!dev->dma_mask) + return false; + return addr + size - 1 <= *dev->dma_mask; +} + +static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) +{ + struct dma_map_ops *dma_ops = get_dma_ops(dev); + + debug_dma_mapping_error(dev, dma_addr); + if (dma_ops->mapping_error) + return dma_ops->mapping_error(dev, dma_addr); + return dma_addr == DMA_ERROR_CODE; +} + +#define dma_alloc_coherent(d, s, h, f) dma_alloc_attrs(d, s, h, f, NULL) + +static inline void *dma_alloc_attrs(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t flags, + struct dma_attrs *attrs) +{ + struct dma_map_ops *ops = get_dma_ops(dev); + void *cpu_addr; + + BUG_ON(!ops); + + cpu_addr = ops->alloc(dev, size, dma_handle, flags, attrs); + debug_dma_alloc_coherent(dev, size, *dma_handle, cpu_addr); + + return cpu_addr; +} + +#define dma_free_coherent(d, s, c, h) dma_free_attrs(d, s, c, h, NULL) + +static inline void dma_free_attrs(struct device *dev, size_t size, + void *cpu_addr, dma_addr_t dma_handle, + struct dma_attrs *attrs) +{ + struct dma_map_ops *ops = get_dma_ops(dev); + + BUG_ON(!ops); + + debug_dma_free_coherent(dev, size, cpu_addr, dma_handle); + ops->free(dev, size, cpu_addr, dma_handle, attrs); +} + +#endif /* _ASM_S390_DMA_MAPPING_H */ diff --git a/arch/s390/include/asm/dma.h b/arch/s390/include/asm/dma.h new file mode 100644 index 000000000..bb9bdcd20 --- /dev/null +++ b/arch/s390/include/asm/dma.h @@ -0,0 +1,19 @@ +#ifndef _ASM_S390_DMA_H +#define _ASM_S390_DMA_H + +#include <asm/io.h> + +/* + * MAX_DMA_ADDRESS is ambiguous because on s390 its completely unrelated + * to DMA. It _is_ used for the s390 memory zone split at 2GB caused + * by the 31 bit heritage. + */ +#define MAX_DMA_ADDRESS 0x80000000 + +#ifdef CONFIG_PCI +extern int isa_dma_bridge_buggy; +#else +#define isa_dma_bridge_buggy (0) +#endif + +#endif /* _ASM_S390_DMA_H */ diff --git a/arch/s390/include/asm/eadm.h b/arch/s390/include/asm/eadm.h new file mode 100644 index 000000000..67026300c --- /dev/null +++ b/arch/s390/include/asm/eadm.h @@ -0,0 +1,117 @@ +#ifndef _ASM_S390_EADM_H +#define _ASM_S390_EADM_H + +#include <linux/types.h> +#include <linux/device.h> + +struct arqb { + u64 data; + u16 fmt:4; + u16:12; + u16 cmd_code; + u16:16; + u16 msb_count; + u32 reserved[12]; +} __packed; + +#define ARQB_CMD_MOVE 1 + +struct arsb { + u16 fmt:4; + u32:28; + u8 ef; + u8:8; + u8 ecbi; + u8:8; + u8 fvf; + u16:16; + u8 eqc; + u32:32; + u64 fail_msb; + u64 fail_aidaw; + u64 fail_ms; + u64 fail_scm; + u32 reserved[4]; +} __packed; + +#define EQC_WR_PROHIBIT 22 + +struct msb { + u8 fmt:4; + u8 oc:4; + u8 flags; + u16:12; + u16 bs:4; + u32 blk_count; + u64 data_addr; + u64 scm_addr; + u64:64; +} __packed; + +struct aidaw { + u8 flags; + u32 :24; + u32 :32; + u64 data_addr; +} __packed; + +#define MSB_OC_CLEAR 0 +#define MSB_OC_READ 1 +#define MSB_OC_WRITE 2 +#define MSB_OC_RELEASE 3 + +#define MSB_FLAG_BNM 0x80 +#define MSB_FLAG_IDA 0x40 + +#define MSB_BS_4K 0 +#define MSB_BS_1M 1 + +#define AOB_NR_MSB 124 + +struct aob { + struct arqb request; + struct arsb response; + struct msb msb[AOB_NR_MSB]; +} __packed __aligned(PAGE_SIZE); + +struct aob_rq_header { + struct scm_device *scmdev; + char data[0]; +}; + +struct scm_device { + u64 address; + u64 size; + unsigned int nr_max_block; + struct device dev; + struct { + unsigned int persistence:4; + unsigned int oper_state:4; + unsigned int data_state:4; + unsigned int rank:4; + unsigned int release:1; + unsigned int res_id:8; + } __packed attrs; +}; + +#define OP_STATE_GOOD 1 +#define OP_STATE_TEMP_ERR 2 +#define OP_STATE_PERM_ERR 3 + +enum scm_event {SCM_CHANGE, SCM_AVAIL}; + +struct scm_driver { + struct device_driver drv; + int (*probe) (struct scm_device *scmdev); + int (*remove) (struct scm_device *scmdev); + void (*notify) (struct scm_device *scmdev, enum scm_event event); + void (*handler) (struct scm_device *scmdev, void *data, int error); +}; + +int scm_driver_register(struct scm_driver *scmdrv); +void scm_driver_unregister(struct scm_driver *scmdrv); + +int eadm_start_aob(struct aob *aob); +void scm_irq_handler(struct aob *aob, int error); + +#endif /* _ASM_S390_EADM_H */ diff --git a/arch/s390/include/asm/ebcdic.h b/arch/s390/include/asm/ebcdic.h new file mode 100644 index 000000000..c5befc5a3 --- /dev/null +++ b/arch/s390/include/asm/ebcdic.h @@ -0,0 +1,48 @@ +/* + * EBCDIC -> ASCII, ASCII -> EBCDIC conversion routines. + * + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef _EBCDIC_H +#define _EBCDIC_H + +#ifndef _S390_TYPES_H +#include <types.h> +#endif + +extern __u8 _ascebc_500[256]; /* ASCII -> EBCDIC 500 conversion table */ +extern __u8 _ebcasc_500[256]; /* EBCDIC 500 -> ASCII conversion table */ +extern __u8 _ascebc[256]; /* ASCII -> EBCDIC conversion table */ +extern __u8 _ebcasc[256]; /* EBCDIC -> ASCII conversion table */ +extern __u8 _ebc_tolower[256]; /* EBCDIC -> lowercase */ +extern __u8 _ebc_toupper[256]; /* EBCDIC -> uppercase */ + +static inline void +codepage_convert(const __u8 *codepage, volatile __u8 * addr, unsigned long nr) +{ + if (nr-- <= 0) + return; + asm volatile( + " bras 1,1f\n" + " tr 0(1,%0),0(%2)\n" + "0: tr 0(256,%0),0(%2)\n" + " la %0,256(%0)\n" + "1: ahi %1,-256\n" + " jnm 0b\n" + " ex %1,0(1)" + : "+&a" (addr), "+&a" (nr) + : "a" (codepage) : "cc", "memory", "1"); +} + +#define ASCEBC(addr,nr) codepage_convert(_ascebc, addr, nr) +#define EBCASC(addr,nr) codepage_convert(_ebcasc, addr, nr) +#define ASCEBC_500(addr,nr) codepage_convert(_ascebc_500, addr, nr) +#define EBCASC_500(addr,nr) codepage_convert(_ebcasc_500, addr, nr) +#define EBC_TOLOWER(addr,nr) codepage_convert(_ebc_tolower, addr, nr) +#define EBC_TOUPPER(addr,nr) codepage_convert(_ebc_toupper, addr, nr) + +#endif + diff --git a/arch/s390/include/asm/elf.h b/arch/s390/include/asm/elf.h new file mode 100644 index 000000000..3ad48f22d --- /dev/null +++ b/arch/s390/include/asm/elf.h @@ -0,0 +1,227 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/elf.h" + */ + +#ifndef __ASMS390_ELF_H +#define __ASMS390_ELF_H + +/* s390 relocations defined by the ABIs */ +#define R_390_NONE 0 /* No reloc. */ +#define R_390_8 1 /* Direct 8 bit. */ +#define R_390_12 2 /* Direct 12 bit. */ +#define R_390_16 3 /* Direct 16 bit. */ +#define R_390_32 4 /* Direct 32 bit. */ +#define R_390_PC32 5 /* PC relative 32 bit. */ +#define R_390_GOT12 6 /* 12 bit GOT offset. */ +#define R_390_GOT32 7 /* 32 bit GOT offset. */ +#define R_390_PLT32 8 /* 32 bit PC relative PLT address. */ +#define R_390_COPY 9 /* Copy symbol at runtime. */ +#define R_390_GLOB_DAT 10 /* Create GOT entry. */ +#define R_390_JMP_SLOT 11 /* Create PLT entry. */ +#define R_390_RELATIVE 12 /* Adjust by program base. */ +#define R_390_GOTOFF32 13 /* 32 bit offset to GOT. */ +#define R_390_GOTPC 14 /* 32 bit PC rel. offset to GOT. */ +#define R_390_GOT16 15 /* 16 bit GOT offset. */ +#define R_390_PC16 16 /* PC relative 16 bit. */ +#define R_390_PC16DBL 17 /* PC relative 16 bit shifted by 1. */ +#define R_390_PLT16DBL 18 /* 16 bit PC rel. PLT shifted by 1. */ +#define R_390_PC32DBL 19 /* PC relative 32 bit shifted by 1. */ +#define R_390_PLT32DBL 20 /* 32 bit PC rel. PLT shifted by 1. */ +#define R_390_GOTPCDBL 21 /* 32 bit PC rel. GOT shifted by 1. */ +#define R_390_64 22 /* Direct 64 bit. */ +#define R_390_PC64 23 /* PC relative 64 bit. */ +#define R_390_GOT64 24 /* 64 bit GOT offset. */ +#define R_390_PLT64 25 /* 64 bit PC relative PLT address. */ +#define R_390_GOTENT 26 /* 32 bit PC rel. to GOT entry >> 1. */ +#define R_390_GOTOFF16 27 /* 16 bit offset to GOT. */ +#define R_390_GOTOFF64 28 /* 64 bit offset to GOT. */ +#define R_390_GOTPLT12 29 /* 12 bit offset to jump slot. */ +#define R_390_GOTPLT16 30 /* 16 bit offset to jump slot. */ +#define R_390_GOTPLT32 31 /* 32 bit offset to jump slot. */ +#define R_390_GOTPLT64 32 /* 64 bit offset to jump slot. */ +#define R_390_GOTPLTENT 33 /* 32 bit rel. offset to jump slot. */ +#define R_390_PLTOFF16 34 /* 16 bit offset from GOT to PLT. */ +#define R_390_PLTOFF32 35 /* 32 bit offset from GOT to PLT. */ +#define R_390_PLTOFF64 36 /* 16 bit offset from GOT to PLT. */ +#define R_390_TLS_LOAD 37 /* Tag for load insn in TLS code. */ +#define R_390_TLS_GDCALL 38 /* Tag for function call in general + dynamic TLS code. */ +#define R_390_TLS_LDCALL 39 /* Tag for function call in local + dynamic TLS code. */ +#define R_390_TLS_GD32 40 /* Direct 32 bit for general dynamic + thread local data. */ +#define R_390_TLS_GD64 41 /* Direct 64 bit for general dynamic + thread local data. */ +#define R_390_TLS_GOTIE12 42 /* 12 bit GOT offset for static TLS + block offset. */ +#define R_390_TLS_GOTIE32 43 /* 32 bit GOT offset for static TLS + block offset. */ +#define R_390_TLS_GOTIE64 44 /* 64 bit GOT offset for static TLS + block offset. */ +#define R_390_TLS_LDM32 45 /* Direct 32 bit for local dynamic + thread local data in LD code. */ +#define R_390_TLS_LDM64 46 /* Direct 64 bit for local dynamic + thread local data in LD code. */ +#define R_390_TLS_IE32 47 /* 32 bit address of GOT entry for + negated static TLS block offset. */ +#define R_390_TLS_IE64 48 /* 64 bit address of GOT entry for + negated static TLS block offset. */ +#define R_390_TLS_IEENT 49 /* 32 bit rel. offset to GOT entry for + negated static TLS block offset. */ +#define R_390_TLS_LE32 50 /* 32 bit negated offset relative to + static TLS block. */ +#define R_390_TLS_LE64 51 /* 64 bit negated offset relative to + static TLS block. */ +#define R_390_TLS_LDO32 52 /* 32 bit offset relative to TLS + block. */ +#define R_390_TLS_LDO64 53 /* 64 bit offset relative to TLS + block. */ +#define R_390_TLS_DTPMOD 54 /* ID of module containing symbol. */ +#define R_390_TLS_DTPOFF 55 /* Offset in TLS block. */ +#define R_390_TLS_TPOFF 56 /* Negate offset in static TLS + block. */ +#define R_390_20 57 /* Direct 20 bit. */ +#define R_390_GOT20 58 /* 20 bit GOT offset. */ +#define R_390_GOTPLT20 59 /* 20 bit offset to jump slot. */ +#define R_390_TLS_GOTIE20 60 /* 20 bit GOT offset for static TLS + block offset. */ +/* Keep this the last entry. */ +#define R_390_NUM 61 + +/* Bits present in AT_HWCAP. */ +#define HWCAP_S390_ESAN3 1 +#define HWCAP_S390_ZARCH 2 +#define HWCAP_S390_STFLE 4 +#define HWCAP_S390_MSA 8 +#define HWCAP_S390_LDISP 16 +#define HWCAP_S390_EIMM 32 +#define HWCAP_S390_DFP 64 +#define HWCAP_S390_HPAGE 128 +#define HWCAP_S390_ETF3EH 256 +#define HWCAP_S390_HIGH_GPRS 512 +#define HWCAP_S390_TE 1024 +#define HWCAP_S390_VXRS 2048 + +/* + * These are used to set parameters in the core dumps. + */ +#define ELF_CLASS ELFCLASS64 +#define ELF_DATA ELFDATA2MSB +#define ELF_ARCH EM_S390 + +/* + * ELF register definitions.. + */ + +#include <asm/ptrace.h> +#include <asm/compat.h> +#include <asm/syscall.h> +#include <asm/user.h> + +typedef s390_fp_regs elf_fpregset_t; +typedef s390_regs elf_gregset_t; + +typedef s390_fp_regs compat_elf_fpregset_t; +typedef s390_compat_regs compat_elf_gregset_t; + +#include <linux/sched.h> /* for task_struct */ +#include <asm/mmu_context.h> + +#include <asm/vdso.h> + +extern unsigned int vdso_enabled; + +/* + * This is used to ensure we don't load something for the wrong architecture. + */ +#define elf_check_arch(x) \ + (((x)->e_machine == EM_S390 || (x)->e_machine == EM_S390_OLD) \ + && (x)->e_ident[EI_CLASS] == ELF_CLASS) +#define compat_elf_check_arch(x) \ + (((x)->e_machine == EM_S390 || (x)->e_machine == EM_S390_OLD) \ + && (x)->e_ident[EI_CLASS] == ELF_CLASS) +#define compat_start_thread start_thread31 + +/* For SVR4/S390 the function pointer to be registered with `atexit` is + passed in R14. */ +#define ELF_PLAT_INIT(_r, load_addr) \ + do { \ + _r->gprs[14] = 0; \ + } while (0) + +#define CORE_DUMP_USE_REGSET +#define ELF_EXEC_PAGESIZE 4096 + +/* This is the location that an ET_DYN program is loaded if exec'ed. Typical + use of this is to invoke "./ld.so someprog" to test out a new version of + the loader. We need to make sure that it is out of the way of the program + that it will "exec", and that there is sufficient room for the brk. 64-bit + tasks are aligned to 4GB. */ +#define ELF_ET_DYN_BASE (is_32bit_task() ? \ + (STACK_TOP / 3 * 2) : \ + (STACK_TOP / 3 * 2) & ~((1UL << 32) - 1)) + +/* This yields a mask that user programs can use to figure out what + instruction set this CPU supports. */ + +extern unsigned long elf_hwcap; +#define ELF_HWCAP (elf_hwcap) + +/* This yields a string that ld.so will use to load implementation + specific libraries for optimization. This is more specific in + intent than poking at uname or /proc/cpuinfo. + + For the moment, we have only optimizations for the Intel generations, + but that could change... */ + +#define ELF_PLATFORM_SIZE 8 +extern char elf_platform[]; +#define ELF_PLATFORM (elf_platform) + +#ifndef CONFIG_COMPAT +#define SET_PERSONALITY(ex) \ +do { \ + set_personality(PER_LINUX | \ + (current->personality & (~PER_MASK))); \ + current_thread_info()->sys_call_table = \ + (unsigned long) &sys_call_table; \ +} while (0) +#else /* CONFIG_COMPAT */ +#define SET_PERSONALITY(ex) \ +do { \ + if (personality(current->personality) != PER_LINUX32) \ + set_personality(PER_LINUX | \ + (current->personality & ~PER_MASK)); \ + if ((ex).e_ident[EI_CLASS] == ELFCLASS32) { \ + set_thread_flag(TIF_31BIT); \ + current_thread_info()->sys_call_table = \ + (unsigned long) &sys_call_table_emu; \ + } else { \ + clear_thread_flag(TIF_31BIT); \ + current_thread_info()->sys_call_table = \ + (unsigned long) &sys_call_table; \ + } \ +} while (0) +#endif /* CONFIG_COMPAT */ + +extern unsigned long mmap_rnd_mask; + +#define STACK_RND_MASK (test_thread_flag(TIF_31BIT) ? 0x7ff : mmap_rnd_mask) + +#define ARCH_DLINFO \ +do { \ + if (vdso_enabled) \ + NEW_AUX_ENT(AT_SYSINFO_EHDR, \ + (unsigned long)current->mm->context.vdso_base); \ +} while (0) + +struct linux_binprm; + +#define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 +int arch_setup_additional_pages(struct linux_binprm *, int); + +void *fill_cpu_elf_notes(void *ptr, struct save_area *sa, __vector128 *vxrs); + +#endif diff --git a/arch/s390/include/asm/emergency-restart.h b/arch/s390/include/asm/emergency-restart.h new file mode 100644 index 000000000..108d8c48e --- /dev/null +++ b/arch/s390/include/asm/emergency-restart.h @@ -0,0 +1,6 @@ +#ifndef _ASM_EMERGENCY_RESTART_H +#define _ASM_EMERGENCY_RESTART_H + +#include <asm-generic/emergency-restart.h> + +#endif /* _ASM_EMERGENCY_RESTART_H */ diff --git a/arch/s390/include/asm/etr.h b/arch/s390/include/asm/etr.h new file mode 100644 index 000000000..629b79a93 --- /dev/null +++ b/arch/s390/include/asm/etr.h @@ -0,0 +1,256 @@ +/* + * Copyright IBM Corp. 2006 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ +#ifndef __S390_ETR_H +#define __S390_ETR_H + +/* ETR attachment control register */ +struct etr_eacr { + unsigned int e0 : 1; /* port 0 stepping control */ + unsigned int e1 : 1; /* port 1 stepping control */ + unsigned int _pad0 : 5; /* must be 00100 */ + unsigned int dp : 1; /* data port control */ + unsigned int p0 : 1; /* port 0 change recognition control */ + unsigned int p1 : 1; /* port 1 change recognition control */ + unsigned int _pad1 : 3; /* must be 000 */ + unsigned int ea : 1; /* ETR alert control */ + unsigned int es : 1; /* ETR sync check control */ + unsigned int sl : 1; /* switch to local control */ +} __attribute__ ((packed)); + +/* Port state returned by steai */ +enum etr_psc { + etr_psc_operational = 0, + etr_psc_semi_operational = 1, + etr_psc_protocol_error = 4, + etr_psc_no_symbols = 8, + etr_psc_no_signal = 12, + etr_psc_pps_mode = 13 +}; + +/* Logical port state returned by stetr */ +enum etr_lpsc { + etr_lpsc_operational_step = 0, + etr_lpsc_operational_alt = 1, + etr_lpsc_semi_operational = 2, + etr_lpsc_protocol_error = 4, + etr_lpsc_no_symbol_sync = 8, + etr_lpsc_no_signal = 12, + etr_lpsc_pps_mode = 13 +}; + +/* ETR status words */ +struct etr_esw { + struct etr_eacr eacr; /* attachment control register */ + unsigned int y : 1; /* stepping mode */ + unsigned int _pad0 : 5; /* must be 00000 */ + unsigned int p : 1; /* stepping port number */ + unsigned int q : 1; /* data port number */ + unsigned int psc0 : 4; /* port 0 state code */ + unsigned int psc1 : 4; /* port 1 state code */ +} __attribute__ ((packed)); + +/* Second level data register status word */ +struct etr_slsw { + unsigned int vv1 : 1; /* copy of validity bit data frame 1 */ + unsigned int vv2 : 1; /* copy of validity bit data frame 2 */ + unsigned int vv3 : 1; /* copy of validity bit data frame 3 */ + unsigned int vv4 : 1; /* copy of validity bit data frame 4 */ + unsigned int _pad0 : 19; /* must by all zeroes */ + unsigned int n : 1; /* EAF port number */ + unsigned int v1 : 1; /* validity bit ETR data frame 1 */ + unsigned int v2 : 1; /* validity bit ETR data frame 2 */ + unsigned int v3 : 1; /* validity bit ETR data frame 3 */ + unsigned int v4 : 1; /* validity bit ETR data frame 4 */ + unsigned int _pad1 : 4; /* must be 0000 */ +} __attribute__ ((packed)); + +/* ETR data frames */ +struct etr_edf1 { + unsigned int u : 1; /* untuned bit */ + unsigned int _pad0 : 1; /* must be 0 */ + unsigned int r : 1; /* service request bit */ + unsigned int _pad1 : 4; /* must be 0000 */ + unsigned int a : 1; /* time adjustment bit */ + unsigned int net_id : 8; /* ETR network id */ + unsigned int etr_id : 8; /* id of ETR which sends data frames */ + unsigned int etr_pn : 8; /* port number of ETR output port */ +} __attribute__ ((packed)); + +struct etr_edf2 { + unsigned int etv : 32; /* Upper 32 bits of TOD. */ +} __attribute__ ((packed)); + +struct etr_edf3 { + unsigned int rc : 8; /* failure reason code */ + unsigned int _pad0 : 3; /* must be 000 */ + unsigned int c : 1; /* ETR coupled bit */ + unsigned int tc : 4; /* ETR type code */ + unsigned int blto : 8; /* biased local time offset */ + /* (blto - 128) * 15 = minutes */ + unsigned int buo : 8; /* biased utc offset */ + /* (buo - 128) = leap seconds */ +} __attribute__ ((packed)); + +struct etr_edf4 { + unsigned int ed : 8; /* ETS device dependent data */ + unsigned int _pad0 : 1; /* must be 0 */ + unsigned int buc : 5; /* biased ut1 correction */ + /* (buc - 16) * 0.1 seconds */ + unsigned int em : 6; /* ETS error magnitude */ + unsigned int dc : 6; /* ETS drift code */ + unsigned int sc : 6; /* ETS steering code */ +} __attribute__ ((packed)); + +/* + * ETR attachment information block, two formats + * format 1 has 4 reserved words with a size of 64 bytes + * format 2 has 16 reserved words with a size of 96 bytes + */ +struct etr_aib { + struct etr_esw esw; + struct etr_slsw slsw; + unsigned long long tsp; + struct etr_edf1 edf1; + struct etr_edf2 edf2; + struct etr_edf3 edf3; + struct etr_edf4 edf4; + unsigned int reserved[16]; +} __attribute__ ((packed,aligned(8))); + +/* ETR interruption parameter */ +struct etr_irq_parm { + unsigned int _pad0 : 8; + unsigned int pc0 : 1; /* port 0 state change */ + unsigned int pc1 : 1; /* port 1 state change */ + unsigned int _pad1 : 3; + unsigned int eai : 1; /* ETR alert indication */ + unsigned int _pad2 : 18; +} __attribute__ ((packed)); + +/* Query TOD offset result */ +struct etr_ptff_qto { + unsigned long long physical_clock; + unsigned long long tod_offset; + unsigned long long logical_tod_offset; + unsigned long long tod_epoch_difference; +} __attribute__ ((packed)); + +/* Inline assembly helper functions */ +static inline int etr_setr(struct etr_eacr *ctrl) +{ + int rc = -EOPNOTSUPP; + + asm volatile( + " .insn s,0xb2160000,%1\n" + "0: la %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "+d" (rc) : "Q" (*ctrl)); + return rc; +} + +/* Stores a format 1 aib with 64 bytes */ +static inline int etr_stetr(struct etr_aib *aib) +{ + int rc = -EOPNOTSUPP; + + asm volatile( + " .insn s,0xb2170000,%1\n" + "0: la %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "+d" (rc) : "Q" (*aib)); + return rc; +} + +/* Stores a format 2 aib with 96 bytes for specified port */ +static inline int etr_steai(struct etr_aib *aib, unsigned int func) +{ + register unsigned int reg0 asm("0") = func; + int rc = -EOPNOTSUPP; + + asm volatile( + " .insn s,0xb2b30000,%1\n" + "0: la %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "+d" (rc) : "Q" (*aib), "d" (reg0)); + return rc; +} + +/* Function codes for the steai instruction. */ +#define ETR_STEAI_STEPPING_PORT 0x10 +#define ETR_STEAI_ALTERNATE_PORT 0x11 +#define ETR_STEAI_PORT_0 0x12 +#define ETR_STEAI_PORT_1 0x13 + +static inline int etr_ptff(void *ptff_block, unsigned int func) +{ + register unsigned int reg0 asm("0") = func; + register unsigned long reg1 asm("1") = (unsigned long) ptff_block; + int rc = -EOPNOTSUPP; + + asm volatile( + " .word 0x0104\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (rc), "=m" (ptff_block) + : "d" (reg0), "d" (reg1), "m" (ptff_block) : "cc"); + return rc; +} + +/* Function codes for the ptff instruction. */ +#define ETR_PTFF_QAF 0x00 /* query available functions */ +#define ETR_PTFF_QTO 0x01 /* query tod offset */ +#define ETR_PTFF_QSI 0x02 /* query steering information */ +#define ETR_PTFF_ATO 0x40 /* adjust tod offset */ +#define ETR_PTFF_STO 0x41 /* set tod offset */ +#define ETR_PTFF_SFS 0x42 /* set fine steering rate */ +#define ETR_PTFF_SGS 0x43 /* set gross steering rate */ + +/* Functions needed by the machine check handler */ +void etr_switch_to_local(void); +void etr_sync_check(void); + +/* STP interruption parameter */ +struct stp_irq_parm { + unsigned int _pad0 : 14; + unsigned int tsc : 1; /* Timing status change */ + unsigned int lac : 1; /* Link availability change */ + unsigned int tcpc : 1; /* Time control parameter change */ + unsigned int _pad2 : 15; +} __attribute__ ((packed)); + +#define STP_OP_SYNC 1 +#define STP_OP_CTRL 3 + +struct stp_sstpi { + unsigned int rsvd0; + unsigned int rsvd1 : 8; + unsigned int stratum : 8; + unsigned int vbits : 16; + unsigned int leaps : 16; + unsigned int tmd : 4; + unsigned int ctn : 4; + unsigned int rsvd2 : 3; + unsigned int c : 1; + unsigned int tst : 4; + unsigned int tzo : 16; + unsigned int dsto : 16; + unsigned int ctrl : 16; + unsigned int rsvd3 : 16; + unsigned int tto; + unsigned int rsvd4; + unsigned int ctnid[3]; + unsigned int rsvd5; + unsigned int todoff[4]; + unsigned int rsvd6[48]; +} __attribute__ ((packed)); + +/* Functions needed by the machine check handler */ +void stp_sync_check(void); +void stp_island_check(void); + +#endif /* __S390_ETR_H */ diff --git a/arch/s390/include/asm/exec.h b/arch/s390/include/asm/exec.h new file mode 100644 index 000000000..c4a93d632 --- /dev/null +++ b/arch/s390/include/asm/exec.h @@ -0,0 +1,12 @@ +/* + * Copyright IBM Corp. 1999, 2009 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef __ASM_EXEC_H +#define __ASM_EXEC_H + +extern unsigned long arch_align_stack(unsigned long sp); + +#endif /* __ASM_EXEC_H */ diff --git a/arch/s390/include/asm/extmem.h b/arch/s390/include/asm/extmem.h new file mode 100644 index 000000000..6276002d7 --- /dev/null +++ b/arch/s390/include/asm/extmem.h @@ -0,0 +1,31 @@ +/* + * definitions for external memory segment support + * Copyright IBM Corp. 2003 + */ + +#ifndef _ASM_S390X_DCSS_H +#define _ASM_S390X_DCSS_H +#ifndef __ASSEMBLY__ + +/* possible values for segment type as returned by segment_info */ +#define SEG_TYPE_SW 0 +#define SEG_TYPE_EW 1 +#define SEG_TYPE_SR 2 +#define SEG_TYPE_ER 3 +#define SEG_TYPE_SN 4 +#define SEG_TYPE_EN 5 +#define SEG_TYPE_SC 6 +#define SEG_TYPE_EWEN 7 + +#define SEGMENT_SHARED 0 +#define SEGMENT_EXCLUSIVE 1 + +int segment_load (char *name, int segtype, unsigned long *addr, unsigned long *length); +void segment_unload(char *name); +void segment_save(char *name); +int segment_type (char* name); +int segment_modify_shared (char *name, int do_nonshared); +void segment_warning(int rc, char *seg_name); + +#endif +#endif diff --git a/arch/s390/include/asm/facility.h b/arch/s390/include/asm/facility.h new file mode 100644 index 000000000..0aa6a7ed9 --- /dev/null +++ b/arch/s390/include/asm/facility.h @@ -0,0 +1,67 @@ +/* + * Copyright IBM Corp. 1999, 2009 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef __ASM_FACILITY_H +#define __ASM_FACILITY_H + +#include <linux/string.h> +#include <linux/preempt.h> +#include <asm/lowcore.h> + +#define MAX_FACILITY_BIT (256*8) /* stfle_fac_list has 256 bytes */ + +static inline int __test_facility(unsigned long nr, void *facilities) +{ + unsigned char *ptr; + + if (nr >= MAX_FACILITY_BIT) + return 0; + ptr = (unsigned char *) facilities + (nr >> 3); + return (*ptr & (0x80 >> (nr & 7))) != 0; +} + +/* + * The test_facility function uses the bit odering where the MSB is bit 0. + * That makes it easier to query facility bits with the bit number as + * documented in the Principles of Operation. + */ +static inline int test_facility(unsigned long nr) +{ + return __test_facility(nr, &S390_lowcore.stfle_fac_list); +} + +/** + * stfle - Store facility list extended + * @stfle_fac_list: array where facility list can be stored + * @size: size of passed in array in double words + */ +static inline void stfle(u64 *stfle_fac_list, int size) +{ + unsigned long nr; + + preempt_disable(); + asm volatile( + " .insn s,0xb2b10000,0(0)\n" /* stfl */ + "0:\n" + EX_TABLE(0b, 0b) + : "+m" (S390_lowcore.stfl_fac_list)); + nr = 4; /* bytes stored by stfl */ + memcpy(stfle_fac_list, &S390_lowcore.stfl_fac_list, 4); + if (S390_lowcore.stfl_fac_list & 0x01000000) { + /* More facility bits available with stfle */ + register unsigned long reg0 asm("0") = size - 1; + + asm volatile(".insn s,0xb2b00000,0(%1)" /* stfle */ + : "+d" (reg0) + : "a" (stfle_fac_list) + : "memory", "cc"); + nr = (reg0 + 1) * 8; /* # bytes stored by stfle */ + } + memset((char *) stfle_fac_list + nr, 0, size * 8 - nr); + preempt_enable(); +} + +#endif /* __ASM_FACILITY_H */ diff --git a/arch/s390/include/asm/fb.h b/arch/s390/include/asm/fb.h new file mode 100644 index 000000000..c7df38030 --- /dev/null +++ b/arch/s390/include/asm/fb.h @@ -0,0 +1,12 @@ +#ifndef _ASM_FB_H_ +#define _ASM_FB_H_ +#include <linux/fb.h> + +#define fb_pgprotect(...) do {} while (0) + +static inline int fb_is_primary_device(struct fb_info *info) +{ + return 0; +} + +#endif /* _ASM_FB_H_ */ diff --git a/arch/s390/include/asm/fcx.h b/arch/s390/include/asm/fcx.h new file mode 100644 index 000000000..7ecb92b46 --- /dev/null +++ b/arch/s390/include/asm/fcx.h @@ -0,0 +1,311 @@ +/* + * Functions for assembling fcx enabled I/O control blocks. + * + * Copyright IBM Corp. 2008 + * Author(s): Peter Oberparleiter <peter.oberparleiter@de.ibm.com> + */ + +#ifndef _ASM_S390_FCX_H +#define _ASM_S390_FCX_H _ASM_S390_FCX_H + +#include <linux/types.h> + +#define TCW_FORMAT_DEFAULT 0 +#define TCW_TIDAW_FORMAT_DEFAULT 0 +#define TCW_FLAGS_INPUT_TIDA (1 << (23 - 5)) +#define TCW_FLAGS_TCCB_TIDA (1 << (23 - 6)) +#define TCW_FLAGS_OUTPUT_TIDA (1 << (23 - 7)) +#define TCW_FLAGS_TIDAW_FORMAT(x) ((x) & 3) << (23 - 9) +#define TCW_FLAGS_GET_TIDAW_FORMAT(x) (((x) >> (23 - 9)) & 3) + +/** + * struct tcw - Transport Control Word (TCW) + * @format: TCW format + * @flags: TCW flags + * @tccbl: Transport-Command-Control-Block Length + * @r: Read Operations + * @w: Write Operations + * @output: Output-Data Address + * @input: Input-Data Address + * @tsb: Transport-Status-Block Address + * @tccb: Transport-Command-Control-Block Address + * @output_count: Output Count + * @input_count: Input Count + * @intrg: Interrogate TCW Address + */ +struct tcw { + u32 format:2; + u32 :6; + u32 flags:24; + u32 :8; + u32 tccbl:6; + u32 r:1; + u32 w:1; + u32 :16; + u64 output; + u64 input; + u64 tsb; + u64 tccb; + u32 output_count; + u32 input_count; + u32 :32; + u32 :32; + u32 :32; + u32 intrg; +} __attribute__ ((packed, aligned(64))); + +#define TIDAW_FLAGS_LAST (1 << (7 - 0)) +#define TIDAW_FLAGS_SKIP (1 << (7 - 1)) +#define TIDAW_FLAGS_DATA_INT (1 << (7 - 2)) +#define TIDAW_FLAGS_TTIC (1 << (7 - 3)) +#define TIDAW_FLAGS_INSERT_CBC (1 << (7 - 4)) + +/** + * struct tidaw - Transport-Indirect-Addressing Word (TIDAW) + * @flags: TIDAW flags. Can be an arithmetic OR of the following constants: + * %TIDAW_FLAGS_LAST, %TIDAW_FLAGS_SKIP, %TIDAW_FLAGS_DATA_INT, + * %TIDAW_FLAGS_TTIC, %TIDAW_FLAGS_INSERT_CBC + * @count: Count + * @addr: Address + */ +struct tidaw { + u32 flags:8; + u32 :24; + u32 count; + u64 addr; +} __attribute__ ((packed, aligned(16))); + +/** + * struct tsa_iostat - I/O-Status Transport-Status Area (IO-Stat TSA) + * @dev_time: Device Time + * @def_time: Defer Time + * @queue_time: Queue Time + * @dev_busy_time: Device-Busy Time + * @dev_act_time: Device-Active-Only Time + * @sense: Sense Data (if present) + */ +struct tsa_iostat { + u32 dev_time; + u32 def_time; + u32 queue_time; + u32 dev_busy_time; + u32 dev_act_time; + u8 sense[32]; +} __attribute__ ((packed)); + +/** + * struct tsa_ddpcs - Device-Detected-Program-Check Transport-Status Area (DDPC TSA) + * @rc: Reason Code + * @rcq: Reason Code Qualifier + * @sense: Sense Data (if present) + */ +struct tsa_ddpc { + u32 :24; + u32 rc:8; + u8 rcq[16]; + u8 sense[32]; +} __attribute__ ((packed)); + +#define TSA_INTRG_FLAGS_CU_STATE_VALID (1 << (7 - 0)) +#define TSA_INTRG_FLAGS_DEV_STATE_VALID (1 << (7 - 1)) +#define TSA_INTRG_FLAGS_OP_STATE_VALID (1 << (7 - 2)) + +/** + * struct tsa_intrg - Interrogate Transport-Status Area (Intrg. TSA) + * @format: Format + * @flags: Flags. Can be an arithmetic OR of the following constants: + * %TSA_INTRG_FLAGS_CU_STATE_VALID, %TSA_INTRG_FLAGS_DEV_STATE_VALID, + * %TSA_INTRG_FLAGS_OP_STATE_VALID + * @cu_state: Controle-Unit State + * @dev_state: Device State + * @op_state: Operation State + * @sd_info: State-Dependent Information + * @dl_id: Device-Level Identifier + * @dd_data: Device-Dependent Data + */ +struct tsa_intrg { + u32 format:8; + u32 flags:8; + u32 cu_state:8; + u32 dev_state:8; + u32 op_state:8; + u32 :24; + u8 sd_info[12]; + u32 dl_id; + u8 dd_data[28]; +} __attribute__ ((packed)); + +#define TSB_FORMAT_NONE 0 +#define TSB_FORMAT_IOSTAT 1 +#define TSB_FORMAT_DDPC 2 +#define TSB_FORMAT_INTRG 3 + +#define TSB_FLAGS_DCW_OFFSET_VALID (1 << (7 - 0)) +#define TSB_FLAGS_COUNT_VALID (1 << (7 - 1)) +#define TSB_FLAGS_CACHE_MISS (1 << (7 - 2)) +#define TSB_FLAGS_TIME_VALID (1 << (7 - 3)) +#define TSB_FLAGS_FORMAT(x) ((x) & 7) +#define TSB_FORMAT(t) ((t)->flags & 7) + +/** + * struct tsb - Transport-Status Block (TSB) + * @length: Length + * @flags: Flags. Can be an arithmetic OR of the following constants: + * %TSB_FLAGS_DCW_OFFSET_VALID, %TSB_FLAGS_COUNT_VALID, %TSB_FLAGS_CACHE_MISS, + * %TSB_FLAGS_TIME_VALID + * @dcw_offset: DCW Offset + * @count: Count + * @tsa: Transport-Status-Area + */ +struct tsb { + u32 length:8; + u32 flags:8; + u32 dcw_offset:16; + u32 count; + u32 :32; + union { + struct tsa_iostat iostat; + struct tsa_ddpc ddpc; + struct tsa_intrg intrg; + } __attribute__ ((packed)) tsa; +} __attribute__ ((packed, aligned(8))); + +#define DCW_INTRG_FORMAT_DEFAULT 0 + +#define DCW_INTRG_RC_UNSPECIFIED 0 +#define DCW_INTRG_RC_TIMEOUT 1 + +#define DCW_INTRG_RCQ_UNSPECIFIED 0 +#define DCW_INTRG_RCQ_PRIMARY 1 +#define DCW_INTRG_RCQ_SECONDARY 2 + +#define DCW_INTRG_FLAGS_MPM (1 << (7 - 0)) +#define DCW_INTRG_FLAGS_PPR (1 << (7 - 1)) +#define DCW_INTRG_FLAGS_CRIT (1 << (7 - 2)) + +/** + * struct dcw_intrg_data - Interrogate DCW data + * @format: Format. Should be %DCW_INTRG_FORMAT_DEFAULT + * @rc: Reason Code. Can be one of %DCW_INTRG_RC_UNSPECIFIED, + * %DCW_INTRG_RC_TIMEOUT + * @rcq: Reason Code Qualifier: Can be one of %DCW_INTRG_RCQ_UNSPECIFIED, + * %DCW_INTRG_RCQ_PRIMARY, %DCW_INTRG_RCQ_SECONDARY + * @lpm: Logical-Path Mask + * @pam: Path-Available Mask + * @pim: Path-Installed Mask + * @timeout: Timeout + * @flags: Flags. Can be an arithmetic OR of %DCW_INTRG_FLAGS_MPM, + * %DCW_INTRG_FLAGS_PPR, %DCW_INTRG_FLAGS_CRIT + * @time: Time + * @prog_id: Program Identifier + * @prog_data: Program-Dependent Data + */ +struct dcw_intrg_data { + u32 format:8; + u32 rc:8; + u32 rcq:8; + u32 lpm:8; + u32 pam:8; + u32 pim:8; + u32 timeout:16; + u32 flags:8; + u32 :24; + u32 :32; + u64 time; + u64 prog_id; + u8 prog_data[0]; +} __attribute__ ((packed)); + +#define DCW_FLAGS_CC (1 << (7 - 1)) + +#define DCW_CMD_WRITE 0x01 +#define DCW_CMD_READ 0x02 +#define DCW_CMD_CONTROL 0x03 +#define DCW_CMD_SENSE 0x04 +#define DCW_CMD_SENSE_ID 0xe4 +#define DCW_CMD_INTRG 0x40 + +/** + * struct dcw - Device-Command Word (DCW) + * @cmd: Command Code. Can be one of %DCW_CMD_WRITE, %DCW_CMD_READ, + * %DCW_CMD_CONTROL, %DCW_CMD_SENSE, %DCW_CMD_SENSE_ID, %DCW_CMD_INTRG + * @flags: Flags. Can be an arithmetic OR of %DCW_FLAGS_CC + * @cd_count: Control-Data Count + * @count: Count + * @cd: Control Data + */ +struct dcw { + u32 cmd:8; + u32 flags:8; + u32 :8; + u32 cd_count:8; + u32 count; + u8 cd[0]; +} __attribute__ ((packed)); + +#define TCCB_FORMAT_DEFAULT 0x7f +#define TCCB_MAX_DCW 30 +#define TCCB_MAX_SIZE (sizeof(struct tccb_tcah) + \ + TCCB_MAX_DCW * sizeof(struct dcw) + \ + sizeof(struct tccb_tcat)) +#define TCCB_SAC_DEFAULT 0x1ffe +#define TCCB_SAC_INTRG 0x1fff + +/** + * struct tccb_tcah - Transport-Command-Area Header (TCAH) + * @format: Format. Should be %TCCB_FORMAT_DEFAULT + * @tcal: Transport-Command-Area Length + * @sac: Service-Action Code. Can be one of %TCCB_SAC_DEFAULT, %TCCB_SAC_INTRG + * @prio: Priority + */ +struct tccb_tcah { + u32 format:8; + u32 :24; + u32 :24; + u32 tcal:8; + u32 sac:16; + u32 :8; + u32 prio:8; + u32 :32; +} __attribute__ ((packed)); + +/** + * struct tccb_tcat - Transport-Command-Area Trailer (TCAT) + * @count: Transport Count + */ +struct tccb_tcat { + u32 :32; + u32 count; +} __attribute__ ((packed)); + +/** + * struct tccb - (partial) Transport-Command-Control Block (TCCB) + * @tcah: TCAH + * @tca: Transport-Command Area + */ +struct tccb { + struct tccb_tcah tcah; + u8 tca[0]; +} __attribute__ ((packed, aligned(8))); + +struct tcw *tcw_get_intrg(struct tcw *tcw); +void *tcw_get_data(struct tcw *tcw); +struct tccb *tcw_get_tccb(struct tcw *tcw); +struct tsb *tcw_get_tsb(struct tcw *tcw); + +void tcw_init(struct tcw *tcw, int r, int w); +void tcw_finalize(struct tcw *tcw, int num_tidaws); + +void tcw_set_intrg(struct tcw *tcw, struct tcw *intrg_tcw); +void tcw_set_data(struct tcw *tcw, void *data, int use_tidal); +void tcw_set_tccb(struct tcw *tcw, struct tccb *tccb); +void tcw_set_tsb(struct tcw *tcw, struct tsb *tsb); + +void tccb_init(struct tccb *tccb, size_t tccb_size, u32 sac); +void tsb_init(struct tsb *tsb); +struct dcw *tccb_add_dcw(struct tccb *tccb, size_t tccb_size, u8 cmd, u8 flags, + void *cd, u8 cd_count, u32 count); +struct tidaw *tcw_add_tidaw(struct tcw *tcw, int num_tidaws, u8 flags, + void *addr, u32 count); + +#endif /* _ASM_S390_FCX_H */ diff --git a/arch/s390/include/asm/ftrace.h b/arch/s390/include/asm/ftrace.h new file mode 100644 index 000000000..836c56290 --- /dev/null +++ b/arch/s390/include/asm/ftrace.h @@ -0,0 +1,84 @@ +#ifndef _ASM_S390_FTRACE_H +#define _ASM_S390_FTRACE_H + +#define ARCH_SUPPORTS_FTRACE_OPS 1 + +#ifdef CC_USING_HOTPATCH +#define MCOUNT_INSN_SIZE 6 +#else +#define MCOUNT_INSN_SIZE 24 +#define MCOUNT_RETURN_FIXUP 18 +#endif + +#ifndef __ASSEMBLY__ + +#define ftrace_return_address(n) __builtin_return_address(n) + +void _mcount(void); +void ftrace_caller(void); + +extern char ftrace_graph_caller_end; +extern unsigned long ftrace_plt; + +struct dyn_arch_ftrace { }; + +#define MCOUNT_ADDR ((unsigned long)_mcount) +#define FTRACE_ADDR ((unsigned long)ftrace_caller) + +#define KPROBE_ON_FTRACE_NOP 0 +#define KPROBE_ON_FTRACE_CALL 1 + +static inline unsigned long ftrace_call_adjust(unsigned long addr) +{ + return addr; +} + +struct ftrace_insn { + u16 opc; + s32 disp; +} __packed; + +static inline void ftrace_generate_nop_insn(struct ftrace_insn *insn) +{ +#ifdef CONFIG_FUNCTION_TRACER +#ifdef CC_USING_HOTPATCH + /* brcl 0,0 */ + insn->opc = 0xc004; + insn->disp = 0; +#else + /* jg .+24 */ + insn->opc = 0xc0f4; + insn->disp = MCOUNT_INSN_SIZE / 2; +#endif +#endif +} + +static inline int is_ftrace_nop(struct ftrace_insn *insn) +{ +#ifdef CONFIG_FUNCTION_TRACER +#ifdef CC_USING_HOTPATCH + if (insn->disp == 0) + return 1; +#else + if (insn->disp == MCOUNT_INSN_SIZE / 2) + return 1; +#endif +#endif + return 0; +} + +static inline void ftrace_generate_call_insn(struct ftrace_insn *insn, + unsigned long ip) +{ +#ifdef CONFIG_FUNCTION_TRACER + unsigned long target; + + /* brasl r0,ftrace_caller */ + target = is_module_addr((void *) ip) ? ftrace_plt : FTRACE_ADDR; + insn->opc = 0xc005; + insn->disp = (target - ip) / 2; +#endif +} + +#endif /* __ASSEMBLY__ */ +#endif /* _ASM_S390_FTRACE_H */ diff --git a/arch/s390/include/asm/futex.h b/arch/s390/include/asm/futex.h new file mode 100644 index 000000000..a4811aa03 --- /dev/null +++ b/arch/s390/include/asm/futex.h @@ -0,0 +1,96 @@ +#ifndef _ASM_S390_FUTEX_H +#define _ASM_S390_FUTEX_H + +#include <linux/uaccess.h> +#include <linux/futex.h> +#include <asm/mmu_context.h> +#include <asm/errno.h> + +#define __futex_atomic_op(insn, ret, oldval, newval, uaddr, oparg) \ + asm volatile( \ + " sacf 256\n" \ + "0: l %1,0(%6)\n" \ + "1:"insn \ + "2: cs %1,%2,0(%6)\n" \ + "3: jl 1b\n" \ + " lhi %0,0\n" \ + "4: sacf 768\n" \ + EX_TABLE(0b,4b) EX_TABLE(2b,4b) EX_TABLE(3b,4b) \ + : "=d" (ret), "=&d" (oldval), "=&d" (newval), \ + "=m" (*uaddr) \ + : "0" (-EFAULT), "d" (oparg), "a" (uaddr), \ + "m" (*uaddr) : "cc"); + +static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr) +{ + int op = (encoded_op >> 28) & 7; + int cmp = (encoded_op >> 24) & 15; + int oparg = (encoded_op << 8) >> 20; + int cmparg = (encoded_op << 20) >> 20; + int oldval = 0, newval, ret; + + load_kernel_asce(); + if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) + oparg = 1 << oparg; + + pagefault_disable(); + switch (op) { + case FUTEX_OP_SET: + __futex_atomic_op("lr %2,%5\n", + ret, oldval, newval, uaddr, oparg); + break; + case FUTEX_OP_ADD: + __futex_atomic_op("lr %2,%1\nar %2,%5\n", + ret, oldval, newval, uaddr, oparg); + break; + case FUTEX_OP_OR: + __futex_atomic_op("lr %2,%1\nor %2,%5\n", + ret, oldval, newval, uaddr, oparg); + break; + case FUTEX_OP_ANDN: + __futex_atomic_op("lr %2,%1\nnr %2,%5\n", + ret, oldval, newval, uaddr, oparg); + break; + case FUTEX_OP_XOR: + __futex_atomic_op("lr %2,%1\nxr %2,%5\n", + ret, oldval, newval, uaddr, oparg); + break; + default: + ret = -ENOSYS; + } + pagefault_enable(); + + if (!ret) { + switch (cmp) { + case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break; + case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break; + case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break; + case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break; + case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break; + case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break; + default: ret = -ENOSYS; + } + } + return ret; +} + +static inline int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, + u32 oldval, u32 newval) +{ + int ret; + + load_kernel_asce(); + asm volatile( + " sacf 256\n" + "0: cs %1,%4,0(%5)\n" + "1: la %0,0\n" + "2: sacf 768\n" + EX_TABLE(0b,2b) EX_TABLE(1b,2b) + : "=d" (ret), "+d" (oldval), "=m" (*uaddr) + : "0" (-EFAULT), "d" (newval), "a" (uaddr), "m" (*uaddr) + : "cc", "memory"); + *uval = oldval; + return ret; +} + +#endif /* _ASM_S390_FUTEX_H */ diff --git a/arch/s390/include/asm/hardirq.h b/arch/s390/include/asm/hardirq.h new file mode 100644 index 000000000..b7eabaaef --- /dev/null +++ b/arch/s390/include/asm/hardirq.h @@ -0,0 +1,26 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + * Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com) + * + * Derived from "include/asm-i386/hardirq.h" + */ + +#ifndef __ASM_HARDIRQ_H +#define __ASM_HARDIRQ_H + +#include <asm/lowcore.h> + +#define local_softirq_pending() (S390_lowcore.softirq_pending) + +#define __ARCH_IRQ_STAT +#define __ARCH_HAS_DO_SOFTIRQ +#define __ARCH_IRQ_EXIT_IRQS_DISABLED + +static inline void ack_bad_irq(unsigned int irq) +{ + printk(KERN_CRIT "unexpected IRQ trap at vector %02x\n", irq); +} + +#endif /* __ASM_HARDIRQ_H */ diff --git a/arch/s390/include/asm/hugetlb.h b/arch/s390/include/asm/hugetlb.h new file mode 100644 index 000000000..11eae5f55 --- /dev/null +++ b/arch/s390/include/asm/hugetlb.h @@ -0,0 +1,115 @@ +/* + * IBM System z Huge TLB Page Support for Kernel. + * + * Copyright IBM Corp. 2008 + * Author(s): Gerald Schaefer <gerald.schaefer@de.ibm.com> + */ + +#ifndef _ASM_S390_HUGETLB_H +#define _ASM_S390_HUGETLB_H + +#include <asm/page.h> +#include <asm/pgtable.h> + + +#define is_hugepage_only_range(mm, addr, len) 0 +#define hugetlb_free_pgd_range free_pgd_range + +void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pte); +pte_t huge_ptep_get(pte_t *ptep); +pte_t huge_ptep_get_and_clear(struct mm_struct *mm, + unsigned long addr, pte_t *ptep); + +/* + * If the arch doesn't supply something else, assume that hugepage + * size aligned regions are ok without further preparation. + */ +static inline int prepare_hugepage_range(struct file *file, + unsigned long addr, unsigned long len) +{ + if (len & ~HPAGE_MASK) + return -EINVAL; + if (addr & ~HPAGE_MASK) + return -EINVAL; + return 0; +} + +#define hugetlb_prefault_arch_hook(mm) do { } while (0) +#define arch_clear_hugepage_flags(page) do { } while (0) + +int arch_prepare_hugepage(struct page *page); +void arch_release_hugepage(struct page *page); + +static inline void huge_pte_clear(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) +{ + pte_val(*ptep) = _SEGMENT_ENTRY_EMPTY; +} + +static inline void huge_ptep_clear_flush(struct vm_area_struct *vma, + unsigned long address, pte_t *ptep) +{ + huge_ptep_get_and_clear(vma->vm_mm, address, ptep); +} + +static inline int huge_ptep_set_access_flags(struct vm_area_struct *vma, + unsigned long addr, pte_t *ptep, + pte_t pte, int dirty) +{ + int changed = !pte_same(huge_ptep_get(ptep), pte); + if (changed) { + huge_ptep_get_and_clear(vma->vm_mm, addr, ptep); + set_huge_pte_at(vma->vm_mm, addr, ptep, pte); + } + return changed; +} + +static inline void huge_ptep_set_wrprotect(struct mm_struct *mm, + unsigned long addr, pte_t *ptep) +{ + pte_t pte = huge_ptep_get_and_clear(mm, addr, ptep); + set_huge_pte_at(mm, addr, ptep, pte_wrprotect(pte)); +} + +static inline pte_t mk_huge_pte(struct page *page, pgprot_t pgprot) +{ + return mk_pte(page, pgprot); +} + +static inline int huge_pte_none(pte_t pte) +{ + return pte_none(pte); +} + +static inline int huge_pte_write(pte_t pte) +{ + return pte_write(pte); +} + +static inline int huge_pte_dirty(pte_t pte) +{ + return pte_dirty(pte); +} + +static inline pte_t huge_pte_mkwrite(pte_t pte) +{ + return pte_mkwrite(pte); +} + +static inline pte_t huge_pte_mkdirty(pte_t pte) +{ + return pte_mkdirty(pte); +} + +static inline pte_t huge_pte_wrprotect(pte_t pte) +{ + return pte_wrprotect(pte); +} + +static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot) +{ + return pte_modify(pte, newprot); +} + +#endif /* _ASM_S390_HUGETLB_H */ diff --git a/arch/s390/include/asm/hw_irq.h b/arch/s390/include/asm/hw_irq.h new file mode 100644 index 000000000..ee96a8b69 --- /dev/null +++ b/arch/s390/include/asm/hw_irq.h @@ -0,0 +1,11 @@ +#ifndef _HW_IRQ_H +#define _HW_IRQ_H + +#include <linux/msi.h> +#include <linux/pci.h> + +void __init init_airq_interrupts(void); +void __init init_cio_interrupts(void); +void __init init_ext_interrupts(void); + +#endif diff --git a/arch/s390/include/asm/idals.h b/arch/s390/include/asm/idals.h new file mode 100644 index 000000000..a7b2d7504 --- /dev/null +++ b/arch/s390/include/asm/idals.h @@ -0,0 +1,232 @@ +/* + * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com> + * Martin Schwidefsky <schwidefsky@de.ibm.com> + * Bugreports.to..: <Linux390@de.ibm.com> + * Copyright IBM Corp. 2000 + * + * History of changes + * 07/24/00 new file + * 05/04/02 code restructuring. + */ + +#ifndef _S390_IDALS_H +#define _S390_IDALS_H + +#include <linux/errno.h> +#include <linux/err.h> +#include <linux/types.h> +#include <linux/slab.h> +#include <asm/cio.h> +#include <asm/uaccess.h> + +#define IDA_SIZE_LOG 12 /* 11 for 2k , 12 for 4k */ +#define IDA_BLOCK_SIZE (1L<<IDA_SIZE_LOG) + +/* + * Test if an address/length pair needs an idal list. + */ +static inline int +idal_is_needed(void *vaddr, unsigned int length) +{ + return ((__pa(vaddr) + length - 1) >> 31) != 0; +} + + +/* + * Return the number of idal words needed for an address/length pair. + */ +static inline unsigned int idal_nr_words(void *vaddr, unsigned int length) +{ + return ((__pa(vaddr) & (IDA_BLOCK_SIZE-1)) + length + + (IDA_BLOCK_SIZE-1)) >> IDA_SIZE_LOG; +} + +/* + * Create the list of idal words for an address/length pair. + */ +static inline unsigned long *idal_create_words(unsigned long *idaws, + void *vaddr, unsigned int length) +{ + unsigned long paddr; + unsigned int cidaw; + + paddr = __pa(vaddr); + cidaw = ((paddr & (IDA_BLOCK_SIZE-1)) + length + + (IDA_BLOCK_SIZE-1)) >> IDA_SIZE_LOG; + *idaws++ = paddr; + paddr &= -IDA_BLOCK_SIZE; + while (--cidaw > 0) { + paddr += IDA_BLOCK_SIZE; + *idaws++ = paddr; + } + return idaws; +} + +/* + * Sets the address of the data in CCW. + * If necessary it allocates an IDAL and sets the appropriate flags. + */ +static inline int +set_normalized_cda(struct ccw1 * ccw, void *vaddr) +{ + unsigned int nridaws; + unsigned long *idal; + + if (ccw->flags & CCW_FLAG_IDA) + return -EINVAL; + nridaws = idal_nr_words(vaddr, ccw->count); + if (nridaws > 0) { + idal = kmalloc(nridaws * sizeof(unsigned long), + GFP_ATOMIC | GFP_DMA ); + if (idal == NULL) + return -ENOMEM; + idal_create_words(idal, vaddr, ccw->count); + ccw->flags |= CCW_FLAG_IDA; + vaddr = idal; + } + ccw->cda = (__u32)(unsigned long) vaddr; + return 0; +} + +/* + * Releases any allocated IDAL related to the CCW. + */ +static inline void +clear_normalized_cda(struct ccw1 * ccw) +{ + if (ccw->flags & CCW_FLAG_IDA) { + kfree((void *)(unsigned long) ccw->cda); + ccw->flags &= ~CCW_FLAG_IDA; + } + ccw->cda = 0; +} + +/* + * Idal buffer extension + */ +struct idal_buffer { + size_t size; + size_t page_order; + void *data[0]; +}; + +/* + * Allocate an idal buffer + */ +static inline struct idal_buffer * +idal_buffer_alloc(size_t size, int page_order) +{ + struct idal_buffer *ib; + int nr_chunks, nr_ptrs, i; + + nr_ptrs = (size + IDA_BLOCK_SIZE - 1) >> IDA_SIZE_LOG; + nr_chunks = (4096 << page_order) >> IDA_SIZE_LOG; + ib = kmalloc(sizeof(struct idal_buffer) + nr_ptrs*sizeof(void *), + GFP_DMA | GFP_KERNEL); + if (ib == NULL) + return ERR_PTR(-ENOMEM); + ib->size = size; + ib->page_order = page_order; + for (i = 0; i < nr_ptrs; i++) { + if ((i & (nr_chunks - 1)) != 0) { + ib->data[i] = ib->data[i-1] + IDA_BLOCK_SIZE; + continue; + } + ib->data[i] = (void *) + __get_free_pages(GFP_KERNEL, page_order); + if (ib->data[i] != NULL) + continue; + // Not enough memory + while (i >= nr_chunks) { + i -= nr_chunks; + free_pages((unsigned long) ib->data[i], + ib->page_order); + } + kfree(ib); + return ERR_PTR(-ENOMEM); + } + return ib; +} + +/* + * Free an idal buffer. + */ +static inline void +idal_buffer_free(struct idal_buffer *ib) +{ + int nr_chunks, nr_ptrs, i; + + nr_ptrs = (ib->size + IDA_BLOCK_SIZE - 1) >> IDA_SIZE_LOG; + nr_chunks = (4096 << ib->page_order) >> IDA_SIZE_LOG; + for (i = 0; i < nr_ptrs; i += nr_chunks) + free_pages((unsigned long) ib->data[i], ib->page_order); + kfree(ib); +} + +/* + * Test if a idal list is really needed. + */ +static inline int +__idal_buffer_is_needed(struct idal_buffer *ib) +{ + return ib->size > (4096ul << ib->page_order) || + idal_is_needed(ib->data[0], ib->size); +} + +/* + * Set channel data address to idal buffer. + */ +static inline void +idal_buffer_set_cda(struct idal_buffer *ib, struct ccw1 *ccw) +{ + if (__idal_buffer_is_needed(ib)) { + // setup idals; + ccw->cda = (u32)(addr_t) ib->data; + ccw->flags |= CCW_FLAG_IDA; + } else + // we do not need idals - use direct addressing + ccw->cda = (u32)(addr_t) ib->data[0]; + ccw->count = ib->size; +} + +/* + * Copy count bytes from an idal buffer to user memory + */ +static inline size_t +idal_buffer_to_user(struct idal_buffer *ib, void __user *to, size_t count) +{ + size_t left; + int i; + + BUG_ON(count > ib->size); + for (i = 0; count > IDA_BLOCK_SIZE; i++) { + left = copy_to_user(to, ib->data[i], IDA_BLOCK_SIZE); + if (left) + return left + count - IDA_BLOCK_SIZE; + to = (void __user *) to + IDA_BLOCK_SIZE; + count -= IDA_BLOCK_SIZE; + } + return copy_to_user(to, ib->data[i], count); +} + +/* + * Copy count bytes from user memory to an idal buffer + */ +static inline size_t +idal_buffer_from_user(struct idal_buffer *ib, const void __user *from, size_t count) +{ + size_t left; + int i; + + BUG_ON(count > ib->size); + for (i = 0; count > IDA_BLOCK_SIZE; i++) { + left = copy_from_user(ib->data[i], from, IDA_BLOCK_SIZE); + if (left) + return left + count - IDA_BLOCK_SIZE; + from = (void __user *) from + IDA_BLOCK_SIZE; + count -= IDA_BLOCK_SIZE; + } + return copy_from_user(ib->data[i], from, count); +} + +#endif diff --git a/arch/s390/include/asm/idle.h b/arch/s390/include/asm/idle.h new file mode 100644 index 000000000..113cd963d --- /dev/null +++ b/arch/s390/include/asm/idle.h @@ -0,0 +1,27 @@ +/* + * Copyright IBM Corp. 2014 + * + * Author: Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef _S390_IDLE_H +#define _S390_IDLE_H + +#include <linux/types.h> +#include <linux/device.h> +#include <linux/seqlock.h> + +struct s390_idle_data { + seqcount_t seqcount; + unsigned long long idle_count; + unsigned long long idle_time; + unsigned long long clock_idle_enter; + unsigned long long clock_idle_exit; + unsigned long long timer_idle_enter; + unsigned long long timer_idle_exit; +}; + +extern struct device_attribute dev_attr_idle_count; +extern struct device_attribute dev_attr_idle_time_us; + +#endif /* _S390_IDLE_H */ diff --git a/arch/s390/include/asm/io.h b/arch/s390/include/asm/io.h new file mode 100644 index 000000000..30fd5c846 --- /dev/null +++ b/arch/s390/include/asm/io.h @@ -0,0 +1,77 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "include/asm-i386/io.h" + */ + +#ifndef _S390_IO_H +#define _S390_IO_H + +#include <linux/kernel.h> +#include <asm/page.h> +#include <asm/pci_io.h> + +#define xlate_dev_mem_ptr xlate_dev_mem_ptr +void *xlate_dev_mem_ptr(phys_addr_t phys); +#define unxlate_dev_mem_ptr unxlate_dev_mem_ptr +void unxlate_dev_mem_ptr(phys_addr_t phys, void *addr); + +/* + * Convert a virtual cached pointer to an uncached pointer + */ +#define xlate_dev_kmem_ptr(p) p + +#define IO_SPACE_LIMIT 0 + +#ifdef CONFIG_PCI + +#define ioremap_nocache(addr, size) ioremap(addr, size) +#define ioremap_wc ioremap_nocache + +static inline void __iomem *ioremap(unsigned long offset, unsigned long size) +{ + return (void __iomem *) offset; +} + +static inline void iounmap(volatile void __iomem *addr) +{ +} + +static inline void __iomem *ioport_map(unsigned long port, unsigned int nr) +{ + return NULL; +} + +static inline void ioport_unmap(void __iomem *p) +{ +} + +/* + * s390 needs a private implementation of pci_iomap since ioremap with its + * offset parameter isn't sufficient. That's because BAR spaces are not + * disjunctive on s390 so we need the bar parameter of pci_iomap to find + * the corresponding device and create the mapping cookie. + */ +#define pci_iomap pci_iomap +#define pci_iounmap pci_iounmap + +#define memcpy_fromio(dst, src, count) zpci_memcpy_fromio(dst, src, count) +#define memcpy_toio(dst, src, count) zpci_memcpy_toio(dst, src, count) +#define memset_io(dst, val, count) zpci_memset_io(dst, val, count) + +#define __raw_readb zpci_read_u8 +#define __raw_readw zpci_read_u16 +#define __raw_readl zpci_read_u32 +#define __raw_readq zpci_read_u64 +#define __raw_writeb zpci_write_u8 +#define __raw_writew zpci_write_u16 +#define __raw_writel zpci_write_u32 +#define __raw_writeq zpci_write_u64 + +#endif /* CONFIG_PCI */ + +#include <asm-generic/io.h> + +#endif diff --git a/arch/s390/include/asm/ipl.h b/arch/s390/include/asm/ipl.h new file mode 100644 index 000000000..ece606c2e --- /dev/null +++ b/arch/s390/include/asm/ipl.h @@ -0,0 +1,182 @@ +/* + * s390 (re)ipl support + * + * Copyright IBM Corp. 2007 + */ + +#ifndef _ASM_S390_IPL_H +#define _ASM_S390_IPL_H + +#include <asm/lowcore.h> +#include <asm/types.h> +#include <asm/cio.h> +#include <asm/setup.h> + +#define IPL_PARMBLOCK_ORIGIN 0x2000 + +#define IPL_PARM_BLK_FCP_LEN (sizeof(struct ipl_list_hdr) + \ + sizeof(struct ipl_block_fcp)) + +#define IPL_PARM_BLK0_FCP_LEN (sizeof(struct ipl_block_fcp) + 16) + +#define IPL_PARM_BLK_CCW_LEN (sizeof(struct ipl_list_hdr) + \ + sizeof(struct ipl_block_ccw)) + +#define IPL_PARM_BLK0_CCW_LEN (sizeof(struct ipl_block_ccw) + 16) + +#define IPL_MAX_SUPPORTED_VERSION (0) + +#define IPL_PARMBLOCK_START ((struct ipl_parameter_block *) \ + IPL_PARMBLOCK_ORIGIN) +#define IPL_PARMBLOCK_SIZE (IPL_PARMBLOCK_START->hdr.len) + +struct ipl_list_hdr { + u32 len; + u8 reserved1[3]; + u8 version; + u32 blk0_len; + u8 pbt; + u8 flags; + u16 reserved2; + u8 loadparm[8]; +} __attribute__((packed)); + +struct ipl_block_fcp { + u8 reserved1[305-1]; + u8 opt; + u8 reserved2[3]; + u16 reserved3; + u16 devno; + u8 reserved4[4]; + u64 wwpn; + u64 lun; + u32 bootprog; + u8 reserved5[12]; + u64 br_lba; + u32 scp_data_len; + u8 reserved6[260]; + u8 scp_data[]; +} __attribute__((packed)); + +#define DIAG308_VMPARM_SIZE 64 +#define DIAG308_SCPDATA_SIZE (PAGE_SIZE - (sizeof(struct ipl_list_hdr) + \ + offsetof(struct ipl_block_fcp, scp_data))) + +struct ipl_block_ccw { + u8 reserved1[84]; + u8 reserved2[2]; + u16 devno; + u8 vm_flags; + u8 reserved3[3]; + u32 vm_parm_len; + u8 nss_name[8]; + u8 vm_parm[DIAG308_VMPARM_SIZE]; + u8 reserved4[8]; +} __attribute__((packed)); + +struct ipl_parameter_block { + struct ipl_list_hdr hdr; + union { + struct ipl_block_fcp fcp; + struct ipl_block_ccw ccw; + } ipl_info; +} __attribute__((packed,aligned(4096))); + +/* + * IPL validity flags + */ +extern u32 ipl_flags; +extern u32 dump_prefix_page; + +struct dump_save_areas { + struct save_area_ext **areas; + int count; +}; + +extern struct dump_save_areas dump_save_areas; +struct save_area_ext *dump_save_area_create(int cpu); + +extern void do_reipl(void); +extern void do_halt(void); +extern void do_poff(void); +extern void ipl_save_parameters(void); +extern void ipl_update_parameters(void); +extern size_t append_ipl_vmparm(char *, size_t); +extern size_t append_ipl_scpdata(char *, size_t); + +enum { + IPL_DEVNO_VALID = 1, + IPL_PARMBLOCK_VALID = 2, + IPL_NSS_VALID = 4, +}; + +enum ipl_type { + IPL_TYPE_UNKNOWN = 1, + IPL_TYPE_CCW = 2, + IPL_TYPE_FCP = 4, + IPL_TYPE_FCP_DUMP = 8, + IPL_TYPE_NSS = 16, +}; + +struct ipl_info +{ + enum ipl_type type; + union { + struct { + struct ccw_dev_id dev_id; + } ccw; + struct { + struct ccw_dev_id dev_id; + u64 wwpn; + u64 lun; + } fcp; + struct { + char name[NSS_NAME_SIZE + 1]; + } nss; + } data; +}; + +extern struct ipl_info ipl_info; +extern void setup_ipl(void); + +/* + * DIAG 308 support + */ +enum diag308_subcode { + DIAG308_REL_HSA = 2, + DIAG308_IPL = 3, + DIAG308_DUMP = 4, + DIAG308_SET = 5, + DIAG308_STORE = 6, +}; + +enum diag308_ipl_type { + DIAG308_IPL_TYPE_FCP = 0, + DIAG308_IPL_TYPE_CCW = 2, +}; + +enum diag308_opt { + DIAG308_IPL_OPT_IPL = 0x10, + DIAG308_IPL_OPT_DUMP = 0x20, +}; + +enum diag308_flags { + DIAG308_FLAGS_LP_VALID = 0x80, +}; + +enum diag308_vm_flags { + DIAG308_VM_FLAGS_NSS_VALID = 0x80, + DIAG308_VM_FLAGS_VP_VALID = 0x40, +}; + +enum diag308_rc { + DIAG308_RC_OK = 0x0001, + DIAG308_RC_NOCONFIG = 0x0102, +}; + +extern int diag308(unsigned long subcode, void *addr); +extern void diag308_reset(void); +extern void store_status(void); +extern void lgr_info_log(void); + +#endif /* _ASM_S390_IPL_H */ diff --git a/arch/s390/include/asm/irq.h b/arch/s390/include/asm/irq.h new file mode 100644 index 000000000..ff95d15a2 --- /dev/null +++ b/arch/s390/include/asm/irq.h @@ -0,0 +1,106 @@ +#ifndef _ASM_IRQ_H +#define _ASM_IRQ_H + +#define EXT_INTERRUPT 0 +#define IO_INTERRUPT 1 +#define THIN_INTERRUPT 2 + +#define NR_IRQS_BASE 3 + +#ifdef CONFIG_PCI_NR_MSI +# define NR_IRQS (NR_IRQS_BASE + CONFIG_PCI_NR_MSI) +#else +# define NR_IRQS NR_IRQS_BASE +#endif + +/* External interruption codes */ +#define EXT_IRQ_INTERRUPT_KEY 0x0040 +#define EXT_IRQ_CLK_COMP 0x1004 +#define EXT_IRQ_CPU_TIMER 0x1005 +#define EXT_IRQ_WARNING_TRACK 0x1007 +#define EXT_IRQ_MALFUNC_ALERT 0x1200 +#define EXT_IRQ_EMERGENCY_SIG 0x1201 +#define EXT_IRQ_EXTERNAL_CALL 0x1202 +#define EXT_IRQ_TIMING_ALERT 0x1406 +#define EXT_IRQ_MEASURE_ALERT 0x1407 +#define EXT_IRQ_SERVICE_SIG 0x2401 +#define EXT_IRQ_CP_SERVICE 0x2603 +#define EXT_IRQ_IUCV 0x4000 + +#ifndef __ASSEMBLY__ + +#include <linux/hardirq.h> +#include <linux/percpu.h> +#include <linux/cache.h> +#include <linux/types.h> + +enum interruption_class { + IRQEXT_CLK, + IRQEXT_EXC, + IRQEXT_EMS, + IRQEXT_TMR, + IRQEXT_TLA, + IRQEXT_PFL, + IRQEXT_DSD, + IRQEXT_VRT, + IRQEXT_SCP, + IRQEXT_IUC, + IRQEXT_CMS, + IRQEXT_CMC, + IRQEXT_CMR, + IRQEXT_FTP, + IRQIO_CIO, + IRQIO_QAI, + IRQIO_DAS, + IRQIO_C15, + IRQIO_C70, + IRQIO_TAP, + IRQIO_VMR, + IRQIO_LCS, + IRQIO_CTC, + IRQIO_APB, + IRQIO_ADM, + IRQIO_CSC, + IRQIO_PCI, + IRQIO_MSI, + IRQIO_VIR, + IRQIO_VAI, + NMI_NMI, + CPU_RST, + NR_ARCH_IRQS +}; + +struct irq_stat { + unsigned int irqs[NR_ARCH_IRQS]; +}; + +DECLARE_PER_CPU_SHARED_ALIGNED(struct irq_stat, irq_stat); + +static __always_inline void inc_irq_stat(enum interruption_class irq) +{ + __this_cpu_inc(irq_stat.irqs[irq]); +} + +struct ext_code { + unsigned short subcode; + unsigned short code; +}; + +typedef void (*ext_int_handler_t)(struct ext_code, unsigned int, unsigned long); + +int register_external_irq(u16 code, ext_int_handler_t handler); +int unregister_external_irq(u16 code, ext_int_handler_t handler); + +enum irq_subclass { + IRQ_SUBCLASS_MEASUREMENT_ALERT = 5, + IRQ_SUBCLASS_SERVICE_SIGNAL = 9, +}; + +void irq_subclass_register(enum irq_subclass subclass); +void irq_subclass_unregister(enum irq_subclass subclass); + +#define irq_canonicalize(irq) (irq) + +#endif /* __ASSEMBLY__ */ + +#endif /* _ASM_IRQ_H */ diff --git a/arch/s390/include/asm/irq_regs.h b/arch/s390/include/asm/irq_regs.h new file mode 100644 index 000000000..3dd9c0b70 --- /dev/null +++ b/arch/s390/include/asm/irq_regs.h @@ -0,0 +1 @@ +#include <asm-generic/irq_regs.h> diff --git a/arch/s390/include/asm/irqflags.h b/arch/s390/include/asm/irqflags.h new file mode 100644 index 000000000..16aa0c779 --- /dev/null +++ b/arch/s390/include/asm/irqflags.h @@ -0,0 +1,72 @@ +/* + * Copyright IBM Corp. 2006, 2010 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef __ASM_IRQFLAGS_H +#define __ASM_IRQFLAGS_H + +#include <linux/types.h> + +/* store then OR system mask. */ +#define __arch_local_irq_stosm(__or) \ +({ \ + unsigned long __mask; \ + asm volatile( \ + " stosm %0,%1" \ + : "=Q" (__mask) : "i" (__or) : "memory"); \ + __mask; \ +}) + +/* store then AND system mask. */ +#define __arch_local_irq_stnsm(__and) \ +({ \ + unsigned long __mask; \ + asm volatile( \ + " stnsm %0,%1" \ + : "=Q" (__mask) : "i" (__and) : "memory"); \ + __mask; \ +}) + +/* set system mask. */ +static inline notrace void __arch_local_irq_ssm(unsigned long flags) +{ + asm volatile("ssm %0" : : "Q" (flags) : "memory"); +} + +static inline notrace unsigned long arch_local_save_flags(void) +{ + return __arch_local_irq_stnsm(0xff); +} + +static inline notrace unsigned long arch_local_irq_save(void) +{ + return __arch_local_irq_stnsm(0xfc); +} + +static inline notrace void arch_local_irq_disable(void) +{ + arch_local_irq_save(); +} + +static inline notrace void arch_local_irq_enable(void) +{ + __arch_local_irq_stosm(0x03); +} + +static inline notrace void arch_local_irq_restore(unsigned long flags) +{ + __arch_local_irq_ssm(flags); +} + +static inline notrace bool arch_irqs_disabled_flags(unsigned long flags) +{ + return !(flags & (3UL << (BITS_PER_LONG - 8))); +} + +static inline notrace bool arch_irqs_disabled(void) +{ + return arch_irqs_disabled_flags(arch_local_save_flags()); +} + +#endif /* __ASM_IRQFLAGS_H */ diff --git a/arch/s390/include/asm/isc.h b/arch/s390/include/asm/isc.h new file mode 100644 index 000000000..68d7d6830 --- /dev/null +++ b/arch/s390/include/asm/isc.h @@ -0,0 +1,28 @@ +#ifndef _ASM_S390_ISC_H +#define _ASM_S390_ISC_H + +#include <linux/types.h> + +/* + * I/O interruption subclasses used by drivers. + * Please add all used iscs here so that it is possible to distribute + * isc usage between drivers. + * Reminder: 0 is highest priority, 7 lowest. + */ +#define MAX_ISC 7 + +/* Regular I/O interrupts. */ +#define IO_SCH_ISC 3 /* regular I/O subchannels */ +#define CONSOLE_ISC 1 /* console I/O subchannel */ +#define EADM_SCH_ISC 4 /* EADM subchannels */ +#define CHSC_SCH_ISC 7 /* CHSC subchannels */ +/* Adapter interrupts. */ +#define QDIO_AIRQ_ISC IO_SCH_ISC /* I/O subchannel in qdio mode */ +#define PCI_ISC 2 /* PCI I/O subchannels */ +#define AP_ISC 6 /* adjunct processor (crypto) devices */ + +/* Functions for registration of I/O interruption subclasses */ +void isc_register(unsigned int isc); +void isc_unregister(unsigned int isc); + +#endif /* _ASM_S390_ISC_H */ diff --git a/arch/s390/include/asm/itcw.h b/arch/s390/include/asm/itcw.h new file mode 100644 index 000000000..fb1bedd3d --- /dev/null +++ b/arch/s390/include/asm/itcw.h @@ -0,0 +1,30 @@ +/* + * Functions for incremental construction of fcx enabled I/O control blocks. + * + * Copyright IBM Corp. 2008 + * Author(s): Peter Oberparleiter <peter.oberparleiter@de.ibm.com> + */ + +#ifndef _ASM_S390_ITCW_H +#define _ASM_S390_ITCW_H + +#include <linux/types.h> +#include <asm/fcx.h> + +#define ITCW_OP_READ 0 +#define ITCW_OP_WRITE 1 + +struct itcw; + +struct tcw *itcw_get_tcw(struct itcw *itcw); +size_t itcw_calc_size(int intrg, int max_tidaws, int intrg_max_tidaws); +struct itcw *itcw_init(void *buffer, size_t size, int op, int intrg, + int max_tidaws, int intrg_max_tidaws); +struct dcw *itcw_add_dcw(struct itcw *itcw, u8 cmd, u8 flags, void *cd, + u8 cd_count, u32 count); +struct tidaw *itcw_add_tidaw(struct itcw *itcw, u8 flags, void *addr, + u32 count); +void itcw_set_data(struct itcw *itcw, void *addr, int use_tidal); +void itcw_finalize(struct itcw *itcw); + +#endif /* _ASM_S390_ITCW_H */ diff --git a/arch/s390/include/asm/jump_label.h b/arch/s390/include/asm/jump_label.h new file mode 100644 index 000000000..69972b795 --- /dev/null +++ b/arch/s390/include/asm/jump_label.h @@ -0,0 +1,37 @@ +#ifndef _ASM_S390_JUMP_LABEL_H +#define _ASM_S390_JUMP_LABEL_H + +#ifndef __ASSEMBLY__ + +#include <linux/types.h> + +#define JUMP_LABEL_NOP_SIZE 6 +#define JUMP_LABEL_NOP_OFFSET 2 + +/* + * We use a brcl 0,2 instruction for jump labels at compile time so it + * can be easily distinguished from a hotpatch generated instruction. + */ +static __always_inline bool arch_static_branch(struct static_key *key) +{ + asm_volatile_goto("0: brcl 0,"__stringify(JUMP_LABEL_NOP_OFFSET)"\n" + ".pushsection __jump_table, \"aw\"\n" + ".balign 8\n" + ".quad 0b, %l[label], %0\n" + ".popsection\n" + : : "X" (key) : : label); + return false; +label: + return true; +} + +typedef unsigned long jump_label_t; + +struct jump_entry { + jump_label_t code; + jump_label_t target; + jump_label_t key; +}; + +#endif /* __ASSEMBLY__ */ +#endif diff --git a/arch/s390/include/asm/kdebug.h b/arch/s390/include/asm/kdebug.h new file mode 100644 index 000000000..5c1abd476 --- /dev/null +++ b/arch/s390/include/asm/kdebug.h @@ -0,0 +1,27 @@ +#ifndef _S390_KDEBUG_H +#define _S390_KDEBUG_H + +/* + * Feb 2006 Ported to s390 <grundym@us.ibm.com> + */ + +struct pt_regs; + +enum die_val { + DIE_OOPS = 1, + DIE_BPT, + DIE_SSTEP, + DIE_PANIC, + DIE_NMI, + DIE_DIE, + DIE_NMIWATCHDOG, + DIE_KERNELDEBUG, + DIE_TRAP, + DIE_GPF, + DIE_CALL, + DIE_NMI_IPI, +}; + +extern void die(struct pt_regs *, const char *); + +#endif diff --git a/arch/s390/include/asm/kexec.h b/arch/s390/include/asm/kexec.h new file mode 100644 index 000000000..2f924bc30 --- /dev/null +++ b/arch/s390/include/asm/kexec.h @@ -0,0 +1,66 @@ +/* + * Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger <adelsberger@de.ibm.com> + * + */ + +#ifndef _S390_KEXEC_H +#define _S390_KEXEC_H + +#include <asm/processor.h> +#include <asm/page.h> +/* + * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return. + * I.e. Maximum page that is mapped directly into kernel memory, + * and kmap is not required. + */ + +/* Maximum physical address we can use pages from */ +#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL) + +/* Maximum address we can reach in physical address mode */ +#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL) + +/* Maximum address we can use for the control pages */ +/* Not more than 2GB */ +#define KEXEC_CONTROL_MEMORY_LIMIT (1UL<<31) + +/* Allocate control page with GFP_DMA */ +#define KEXEC_CONTROL_MEMORY_GFP GFP_DMA + +/* Maximum address we can use for the crash control pages */ +#define KEXEC_CRASH_CONTROL_MEMORY_LIMIT (-1UL) + +/* Allocate one page for the pdp and the second for the code */ +#define KEXEC_CONTROL_PAGE_SIZE 4096 + +/* Alignment of crashkernel memory */ +#define KEXEC_CRASH_MEM_ALIGN HPAGE_SIZE + +/* The native architecture */ +#define KEXEC_ARCH KEXEC_ARCH_S390 + +/* + * Size for s390x ELF notes per CPU + * + * Seven notes plus zero note at the end: prstatus, fpregset, timer, + * tod_cmp, tod_reg, control regs, and prefix + */ +#define KEXEC_NOTE_BYTES \ + (ALIGN(sizeof(struct elf_note), 4) * 8 + \ + ALIGN(sizeof("CORE"), 4) * 7 + \ + ALIGN(sizeof(struct elf_prstatus), 4) + \ + ALIGN(sizeof(elf_fpregset_t), 4) + \ + ALIGN(sizeof(u64), 4) + \ + ALIGN(sizeof(u64), 4) + \ + ALIGN(sizeof(u32), 4) + \ + ALIGN(sizeof(u64) * 16, 4) + \ + ALIGN(sizeof(u32), 4) \ + ) + +/* Provide a dummy definition to avoid build failures. */ +static inline void crash_setup_regs(struct pt_regs *newregs, + struct pt_regs *oldregs) { } + +#endif /*_S390_KEXEC_H */ diff --git a/arch/s390/include/asm/kmap_types.h b/arch/s390/include/asm/kmap_types.h new file mode 100644 index 000000000..0a8862233 --- /dev/null +++ b/arch/s390/include/asm/kmap_types.h @@ -0,0 +1,6 @@ +#ifndef _ASM_KMAP_TYPES_H +#define _ASM_KMAP_TYPES_H + +#include <asm-generic/kmap_types.h> + +#endif diff --git a/arch/s390/include/asm/kprobes.h b/arch/s390/include/asm/kprobes.h new file mode 100644 index 000000000..b47ad3b64 --- /dev/null +++ b/arch/s390/include/asm/kprobes.h @@ -0,0 +1,94 @@ +#ifndef _ASM_S390_KPROBES_H +#define _ASM_S390_KPROBES_H +/* + * Kernel Probes (KProbes) + * + * 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. + * + * Copyright IBM Corp. 2002, 2006 + * + * 2002-Oct Created by Vamsi Krishna S <vamsi_krishna@in.ibm.com> Kernel + * Probes initial implementation ( includes suggestions from + * Rusty Russell). + * 2004-Nov Modified for PPC64 by Ananth N Mavinakayanahalli + * <ananth@in.ibm.com> + * 2005-Dec Used as a template for s390 by Mike Grundy + * <grundym@us.ibm.com> + */ +#include <linux/types.h> +#include <linux/ptrace.h> +#include <linux/percpu.h> + +#define __ARCH_WANT_KPROBES_INSN_SLOT + +struct pt_regs; +struct kprobe; + +typedef u16 kprobe_opcode_t; +#define BREAKPOINT_INSTRUCTION 0x0002 + +/* Maximum instruction size is 3 (16bit) halfwords: */ +#define MAX_INSN_SIZE 0x0003 +#define MAX_STACK_SIZE 64 +#define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \ + (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) \ + ? (MAX_STACK_SIZE) \ + : (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) + +#define kretprobe_blacklist_size 0 + +#define KPROBE_SWAP_INST 0x10 + +#define FIXUP_PSW_NORMAL 0x08 +#define FIXUP_BRANCH_NOT_TAKEN 0x04 +#define FIXUP_RETURN_REGISTER 0x02 +#define FIXUP_NOT_REQUIRED 0x01 + +/* Architecture specific copy of original instruction */ +struct arch_specific_insn { + /* copy of original instruction */ + kprobe_opcode_t *insn; + unsigned int is_ftrace_insn : 1; +}; + +struct prev_kprobe { + struct kprobe *kp; + unsigned long status; +}; + +/* per-cpu kprobe control block */ +struct kprobe_ctlblk { + unsigned long kprobe_status; + unsigned long kprobe_saved_imask; + unsigned long kprobe_saved_ctl[3]; + struct prev_kprobe prev_kprobe; + struct pt_regs jprobe_saved_regs; + kprobe_opcode_t jprobes_stack[MAX_STACK_SIZE]; +}; + +void arch_remove_kprobe(struct kprobe *p); +void kretprobe_trampoline(void); + +int kprobe_fault_handler(struct pt_regs *regs, int trapnr); +int kprobe_exceptions_notify(struct notifier_block *self, + unsigned long val, void *data); + +int probe_is_prohibited_opcode(u16 *insn); +int probe_get_fixup_type(u16 *insn); +int probe_is_insn_relative_long(u16 *insn); + +#define flush_insn_slot(p) do { } while (0) + +#endif /* _ASM_S390_KPROBES_H */ diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h new file mode 100644 index 000000000..d01fc588b --- /dev/null +++ b/arch/s390/include/asm/kvm_host.h @@ -0,0 +1,642 @@ +/* + * definition for kernel virtual machines on s390 + * + * Copyright IBM Corp. 2008, 2009 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + */ + + +#ifndef ASM_KVM_HOST_H +#define ASM_KVM_HOST_H + +#include <linux/types.h> +#include <linux/hrtimer.h> +#include <linux/interrupt.h> +#include <linux/kvm_types.h> +#include <linux/kvm_host.h> +#include <linux/kvm.h> +#include <asm/debug.h> +#include <asm/cpu.h> +#include <asm/isc.h> + +#define KVM_MAX_VCPUS 64 +#define KVM_USER_MEM_SLOTS 32 + +/* + * These seem to be used for allocating ->chip in the routing table, + * which we don't use. 4096 is an out-of-thin-air value. If we need + * to look at ->chip later on, we'll need to revisit this. + */ +#define KVM_NR_IRQCHIPS 1 +#define KVM_IRQCHIP_NUM_PINS 4096 + +#define SIGP_CTRL_C 0x80 +#define SIGP_CTRL_SCN_MASK 0x3f + +struct sca_entry { + __u8 reserved0; + __u8 sigp_ctrl; + __u16 reserved[3]; + __u64 sda; + __u64 reserved2[2]; +} __attribute__((packed)); + +union ipte_control { + unsigned long val; + struct { + unsigned long k : 1; + unsigned long kh : 31; + unsigned long kg : 32; + }; +}; + +struct sca_block { + union ipte_control ipte_control; + __u64 reserved[5]; + __u64 mcn; + __u64 reserved2; + struct sca_entry cpu[64]; +} __attribute__((packed)); + +#define CPUSTAT_STOPPED 0x80000000 +#define CPUSTAT_WAIT 0x10000000 +#define CPUSTAT_ECALL_PEND 0x08000000 +#define CPUSTAT_STOP_INT 0x04000000 +#define CPUSTAT_IO_INT 0x02000000 +#define CPUSTAT_EXT_INT 0x01000000 +#define CPUSTAT_RUNNING 0x00800000 +#define CPUSTAT_RETAINED 0x00400000 +#define CPUSTAT_TIMING_SUB 0x00020000 +#define CPUSTAT_SIE_SUB 0x00010000 +#define CPUSTAT_RRF 0x00008000 +#define CPUSTAT_SLSV 0x00004000 +#define CPUSTAT_SLSR 0x00002000 +#define CPUSTAT_ZARCH 0x00000800 +#define CPUSTAT_MCDS 0x00000100 +#define CPUSTAT_SM 0x00000080 +#define CPUSTAT_IBS 0x00000040 +#define CPUSTAT_G 0x00000008 +#define CPUSTAT_GED 0x00000004 +#define CPUSTAT_J 0x00000002 +#define CPUSTAT_P 0x00000001 + +struct kvm_s390_sie_block { + atomic_t cpuflags; /* 0x0000 */ + __u32 : 1; /* 0x0004 */ + __u32 prefix : 18; + __u32 : 1; + __u32 ibc : 12; + __u8 reserved08[4]; /* 0x0008 */ +#define PROG_IN_SIE (1<<0) + __u32 prog0c; /* 0x000c */ + __u8 reserved10[16]; /* 0x0010 */ +#define PROG_BLOCK_SIE 0x00000001 + atomic_t prog20; /* 0x0020 */ + __u8 reserved24[4]; /* 0x0024 */ + __u64 cputm; /* 0x0028 */ + __u64 ckc; /* 0x0030 */ + __u64 epoch; /* 0x0038 */ + __u8 reserved40[4]; /* 0x0040 */ +#define LCTL_CR0 0x8000 +#define LCTL_CR6 0x0200 +#define LCTL_CR9 0x0040 +#define LCTL_CR10 0x0020 +#define LCTL_CR11 0x0010 +#define LCTL_CR14 0x0002 + __u16 lctl; /* 0x0044 */ + __s16 icpua; /* 0x0046 */ +#define ICTL_PINT 0x20000000 +#define ICTL_LPSW 0x00400000 +#define ICTL_STCTL 0x00040000 +#define ICTL_ISKE 0x00004000 +#define ICTL_SSKE 0x00002000 +#define ICTL_RRBE 0x00001000 +#define ICTL_TPROT 0x00000200 + __u32 ictl; /* 0x0048 */ + __u32 eca; /* 0x004c */ +#define ICPT_INST 0x04 +#define ICPT_PROGI 0x08 +#define ICPT_INSTPROGI 0x0C +#define ICPT_OPEREXC 0x2C +#define ICPT_PARTEXEC 0x38 +#define ICPT_IOINST 0x40 + __u8 icptcode; /* 0x0050 */ + __u8 icptstatus; /* 0x0051 */ + __u16 ihcpu; /* 0x0052 */ + __u8 reserved54[2]; /* 0x0054 */ + __u16 ipa; /* 0x0056 */ + __u32 ipb; /* 0x0058 */ + __u32 scaoh; /* 0x005c */ + __u8 reserved60; /* 0x0060 */ + __u8 ecb; /* 0x0061 */ + __u8 ecb2; /* 0x0062 */ +#define ECB3_AES 0x04 +#define ECB3_DEA 0x08 + __u8 ecb3; /* 0x0063 */ + __u32 scaol; /* 0x0064 */ + __u8 reserved68[4]; /* 0x0068 */ + __u32 todpr; /* 0x006c */ + __u8 reserved70[32]; /* 0x0070 */ + psw_t gpsw; /* 0x0090 */ + __u64 gg14; /* 0x00a0 */ + __u64 gg15; /* 0x00a8 */ + __u8 reservedb0[20]; /* 0x00b0 */ + __u16 extcpuaddr; /* 0x00c4 */ + __u16 eic; /* 0x00c6 */ + __u32 reservedc8; /* 0x00c8 */ + __u16 pgmilc; /* 0x00cc */ + __u16 iprcc; /* 0x00ce */ + __u32 dxc; /* 0x00d0 */ + __u16 mcn; /* 0x00d4 */ + __u8 perc; /* 0x00d6 */ + __u8 peratmid; /* 0x00d7 */ + __u64 peraddr; /* 0x00d8 */ + __u8 eai; /* 0x00e0 */ + __u8 peraid; /* 0x00e1 */ + __u8 oai; /* 0x00e2 */ + __u8 armid; /* 0x00e3 */ + __u8 reservede4[4]; /* 0x00e4 */ + __u64 tecmc; /* 0x00e8 */ + __u8 reservedf0[12]; /* 0x00f0 */ +#define CRYCB_FORMAT1 0x00000001 +#define CRYCB_FORMAT2 0x00000003 + __u32 crycbd; /* 0x00fc */ + __u64 gcr[16]; /* 0x0100 */ + __u64 gbea; /* 0x0180 */ + __u8 reserved188[24]; /* 0x0188 */ + __u32 fac; /* 0x01a0 */ + __u8 reserved1a4[20]; /* 0x01a4 */ + __u64 cbrlo; /* 0x01b8 */ + __u8 reserved1c0[8]; /* 0x01c0 */ + __u32 ecd; /* 0x01c8 */ + __u8 reserved1cc[18]; /* 0x01cc */ + __u64 pp; /* 0x01de */ + __u8 reserved1e6[2]; /* 0x01e6 */ + __u64 itdba; /* 0x01e8 */ + __u8 reserved1f0[16]; /* 0x01f0 */ +} __attribute__((packed)); + +struct kvm_s390_itdb { + __u8 data[256]; +} __packed; + +struct kvm_s390_vregs { + __vector128 vrs[32]; + __u8 reserved200[512]; /* for future vector expansion */ +} __packed; + +struct sie_page { + struct kvm_s390_sie_block sie_block; + __u8 reserved200[1024]; /* 0x0200 */ + struct kvm_s390_itdb itdb; /* 0x0600 */ + __u8 reserved700[1280]; /* 0x0700 */ + struct kvm_s390_vregs vregs; /* 0x0c00 */ +} __packed; + +struct kvm_vcpu_stat { + u32 exit_userspace; + u32 exit_null; + u32 exit_external_request; + u32 exit_external_interrupt; + u32 exit_stop_request; + u32 exit_validity; + u32 exit_instruction; + u32 halt_successful_poll; + u32 halt_wakeup; + u32 instruction_lctl; + u32 instruction_lctlg; + u32 instruction_stctl; + u32 instruction_stctg; + u32 exit_program_interruption; + u32 exit_instr_and_program; + u32 deliver_external_call; + u32 deliver_emergency_signal; + u32 deliver_service_signal; + u32 deliver_virtio_interrupt; + u32 deliver_stop_signal; + u32 deliver_prefix_signal; + u32 deliver_restart_signal; + u32 deliver_program_int; + u32 deliver_io_int; + u32 exit_wait_state; + u32 instruction_pfmf; + u32 instruction_stidp; + u32 instruction_spx; + u32 instruction_stpx; + u32 instruction_stap; + u32 instruction_storage_key; + u32 instruction_ipte_interlock; + u32 instruction_stsch; + u32 instruction_chsc; + u32 instruction_stsi; + u32 instruction_stfl; + u32 instruction_tprot; + u32 instruction_essa; + u32 instruction_sigp_sense; + u32 instruction_sigp_sense_running; + u32 instruction_sigp_external_call; + u32 instruction_sigp_emergency; + u32 instruction_sigp_cond_emergency; + u32 instruction_sigp_start; + u32 instruction_sigp_stop; + u32 instruction_sigp_stop_store_status; + u32 instruction_sigp_store_status; + u32 instruction_sigp_store_adtl_status; + u32 instruction_sigp_arch; + u32 instruction_sigp_prefix; + u32 instruction_sigp_restart; + u32 instruction_sigp_init_cpu_reset; + u32 instruction_sigp_cpu_reset; + u32 instruction_sigp_unknown; + u32 diagnose_10; + u32 diagnose_44; + u32 diagnose_9c; +}; + +#define PGM_OPERATION 0x01 +#define PGM_PRIVILEGED_OP 0x02 +#define PGM_EXECUTE 0x03 +#define PGM_PROTECTION 0x04 +#define PGM_ADDRESSING 0x05 +#define PGM_SPECIFICATION 0x06 +#define PGM_DATA 0x07 +#define PGM_FIXED_POINT_OVERFLOW 0x08 +#define PGM_FIXED_POINT_DIVIDE 0x09 +#define PGM_DECIMAL_OVERFLOW 0x0a +#define PGM_DECIMAL_DIVIDE 0x0b +#define PGM_HFP_EXPONENT_OVERFLOW 0x0c +#define PGM_HFP_EXPONENT_UNDERFLOW 0x0d +#define PGM_HFP_SIGNIFICANCE 0x0e +#define PGM_HFP_DIVIDE 0x0f +#define PGM_SEGMENT_TRANSLATION 0x10 +#define PGM_PAGE_TRANSLATION 0x11 +#define PGM_TRANSLATION_SPEC 0x12 +#define PGM_SPECIAL_OPERATION 0x13 +#define PGM_OPERAND 0x15 +#define PGM_TRACE_TABEL 0x16 +#define PGM_VECTOR_PROCESSING 0x1b +#define PGM_SPACE_SWITCH 0x1c +#define PGM_HFP_SQUARE_ROOT 0x1d +#define PGM_PC_TRANSLATION_SPEC 0x1f +#define PGM_AFX_TRANSLATION 0x20 +#define PGM_ASX_TRANSLATION 0x21 +#define PGM_LX_TRANSLATION 0x22 +#define PGM_EX_TRANSLATION 0x23 +#define PGM_PRIMARY_AUTHORITY 0x24 +#define PGM_SECONDARY_AUTHORITY 0x25 +#define PGM_LFX_TRANSLATION 0x26 +#define PGM_LSX_TRANSLATION 0x27 +#define PGM_ALET_SPECIFICATION 0x28 +#define PGM_ALEN_TRANSLATION 0x29 +#define PGM_ALE_SEQUENCE 0x2a +#define PGM_ASTE_VALIDITY 0x2b +#define PGM_ASTE_SEQUENCE 0x2c +#define PGM_EXTENDED_AUTHORITY 0x2d +#define PGM_LSTE_SEQUENCE 0x2e +#define PGM_ASTE_INSTANCE 0x2f +#define PGM_STACK_FULL 0x30 +#define PGM_STACK_EMPTY 0x31 +#define PGM_STACK_SPECIFICATION 0x32 +#define PGM_STACK_TYPE 0x33 +#define PGM_STACK_OPERATION 0x34 +#define PGM_ASCE_TYPE 0x38 +#define PGM_REGION_FIRST_TRANS 0x39 +#define PGM_REGION_SECOND_TRANS 0x3a +#define PGM_REGION_THIRD_TRANS 0x3b +#define PGM_MONITOR 0x40 +#define PGM_PER 0x80 +#define PGM_CRYPTO_OPERATION 0x119 + +/* irq types in order of priority */ +enum irq_types { + IRQ_PEND_MCHK_EX = 0, + IRQ_PEND_SVC, + IRQ_PEND_PROG, + IRQ_PEND_MCHK_REP, + IRQ_PEND_EXT_IRQ_KEY, + IRQ_PEND_EXT_MALFUNC, + IRQ_PEND_EXT_EMERGENCY, + IRQ_PEND_EXT_EXTERNAL, + IRQ_PEND_EXT_CLOCK_COMP, + IRQ_PEND_EXT_CPU_TIMER, + IRQ_PEND_EXT_TIMING, + IRQ_PEND_EXT_SERVICE, + IRQ_PEND_EXT_HOST, + IRQ_PEND_PFAULT_INIT, + IRQ_PEND_PFAULT_DONE, + IRQ_PEND_VIRTIO, + IRQ_PEND_IO_ISC_0, + IRQ_PEND_IO_ISC_1, + IRQ_PEND_IO_ISC_2, + IRQ_PEND_IO_ISC_3, + IRQ_PEND_IO_ISC_4, + IRQ_PEND_IO_ISC_5, + IRQ_PEND_IO_ISC_6, + IRQ_PEND_IO_ISC_7, + IRQ_PEND_SIGP_STOP, + IRQ_PEND_RESTART, + IRQ_PEND_SET_PREFIX, + IRQ_PEND_COUNT +}; + +/* We have 2M for virtio device descriptor pages. Smallest amount of + * memory per page is 24 bytes (1 queue), so (2048*1024) / 24 = 87381 + */ +#define KVM_S390_MAX_VIRTIO_IRQS 87381 + +/* + * Repressible (non-floating) machine check interrupts + * subclass bits in MCIC + */ +#define MCHK_EXTD_BIT 58 +#define MCHK_DEGR_BIT 56 +#define MCHK_WARN_BIT 55 +#define MCHK_REP_MASK ((1UL << MCHK_DEGR_BIT) | \ + (1UL << MCHK_EXTD_BIT) | \ + (1UL << MCHK_WARN_BIT)) + +/* Exigent machine check interrupts subclass bits in MCIC */ +#define MCHK_SD_BIT 63 +#define MCHK_PD_BIT 62 +#define MCHK_EX_MASK ((1UL << MCHK_SD_BIT) | (1UL << MCHK_PD_BIT)) + +#define IRQ_PEND_EXT_MASK ((1UL << IRQ_PEND_EXT_IRQ_KEY) | \ + (1UL << IRQ_PEND_EXT_CLOCK_COMP) | \ + (1UL << IRQ_PEND_EXT_CPU_TIMER) | \ + (1UL << IRQ_PEND_EXT_MALFUNC) | \ + (1UL << IRQ_PEND_EXT_EMERGENCY) | \ + (1UL << IRQ_PEND_EXT_EXTERNAL) | \ + (1UL << IRQ_PEND_EXT_TIMING) | \ + (1UL << IRQ_PEND_EXT_HOST) | \ + (1UL << IRQ_PEND_EXT_SERVICE) | \ + (1UL << IRQ_PEND_VIRTIO) | \ + (1UL << IRQ_PEND_PFAULT_INIT) | \ + (1UL << IRQ_PEND_PFAULT_DONE)) + +#define IRQ_PEND_IO_MASK ((1UL << IRQ_PEND_IO_ISC_0) | \ + (1UL << IRQ_PEND_IO_ISC_1) | \ + (1UL << IRQ_PEND_IO_ISC_2) | \ + (1UL << IRQ_PEND_IO_ISC_3) | \ + (1UL << IRQ_PEND_IO_ISC_4) | \ + (1UL << IRQ_PEND_IO_ISC_5) | \ + (1UL << IRQ_PEND_IO_ISC_6) | \ + (1UL << IRQ_PEND_IO_ISC_7)) + +#define IRQ_PEND_MCHK_MASK ((1UL << IRQ_PEND_MCHK_REP) | \ + (1UL << IRQ_PEND_MCHK_EX)) + +struct kvm_s390_interrupt_info { + struct list_head list; + u64 type; + union { + struct kvm_s390_io_info io; + struct kvm_s390_ext_info ext; + struct kvm_s390_pgm_info pgm; + struct kvm_s390_emerg_info emerg; + struct kvm_s390_extcall_info extcall; + struct kvm_s390_prefix_info prefix; + struct kvm_s390_stop_info stop; + struct kvm_s390_mchk_info mchk; + }; +}; + +struct kvm_s390_irq_payload { + struct kvm_s390_io_info io; + struct kvm_s390_ext_info ext; + struct kvm_s390_pgm_info pgm; + struct kvm_s390_emerg_info emerg; + struct kvm_s390_extcall_info extcall; + struct kvm_s390_prefix_info prefix; + struct kvm_s390_stop_info stop; + struct kvm_s390_mchk_info mchk; +}; + +struct kvm_s390_local_interrupt { + spinlock_t lock; + struct kvm_s390_float_interrupt *float_int; + wait_queue_head_t *wq; + atomic_t *cpuflags; + DECLARE_BITMAP(sigp_emerg_pending, KVM_MAX_VCPUS); + struct kvm_s390_irq_payload irq; + unsigned long pending_irqs; +}; + +#define FIRQ_LIST_IO_ISC_0 0 +#define FIRQ_LIST_IO_ISC_1 1 +#define FIRQ_LIST_IO_ISC_2 2 +#define FIRQ_LIST_IO_ISC_3 3 +#define FIRQ_LIST_IO_ISC_4 4 +#define FIRQ_LIST_IO_ISC_5 5 +#define FIRQ_LIST_IO_ISC_6 6 +#define FIRQ_LIST_IO_ISC_7 7 +#define FIRQ_LIST_PFAULT 8 +#define FIRQ_LIST_VIRTIO 9 +#define FIRQ_LIST_COUNT 10 +#define FIRQ_CNTR_IO 0 +#define FIRQ_CNTR_SERVICE 1 +#define FIRQ_CNTR_VIRTIO 2 +#define FIRQ_CNTR_PFAULT 3 +#define FIRQ_MAX_COUNT 4 + +struct kvm_s390_float_interrupt { + unsigned long pending_irqs; + spinlock_t lock; + struct list_head lists[FIRQ_LIST_COUNT]; + int counters[FIRQ_MAX_COUNT]; + struct kvm_s390_mchk_info mchk; + struct kvm_s390_ext_info srv_signal; + int next_rr_cpu; + unsigned long idle_mask[BITS_TO_LONGS(KVM_MAX_VCPUS)]; +}; + +struct kvm_hw_wp_info_arch { + unsigned long addr; + unsigned long phys_addr; + int len; + char *old_data; +}; + +struct kvm_hw_bp_info_arch { + unsigned long addr; + int len; +}; + +/* + * Only the upper 16 bits of kvm_guest_debug->control are arch specific. + * Further KVM_GUESTDBG flags which an be used from userspace can be found in + * arch/s390/include/uapi/asm/kvm.h + */ +#define KVM_GUESTDBG_EXIT_PENDING 0x10000000 + +#define guestdbg_enabled(vcpu) \ + (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) +#define guestdbg_sstep_enabled(vcpu) \ + (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) +#define guestdbg_hw_bp_enabled(vcpu) \ + (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) +#define guestdbg_exit_pending(vcpu) (guestdbg_enabled(vcpu) && \ + (vcpu->guest_debug & KVM_GUESTDBG_EXIT_PENDING)) + +struct kvm_guestdbg_info_arch { + unsigned long cr0; + unsigned long cr9; + unsigned long cr10; + unsigned long cr11; + struct kvm_hw_bp_info_arch *hw_bp_info; + struct kvm_hw_wp_info_arch *hw_wp_info; + int nr_hw_bp; + int nr_hw_wp; + unsigned long last_bp; +}; + +struct kvm_vcpu_arch { + struct kvm_s390_sie_block *sie_block; + s390_fp_regs host_fpregs; + unsigned int host_acrs[NUM_ACRS]; + s390_fp_regs guest_fpregs; + struct kvm_s390_vregs *host_vregs; + struct kvm_s390_local_interrupt local_int; + struct hrtimer ckc_timer; + struct kvm_s390_pgm_info pgm; + union { + struct cpuid cpu_id; + u64 stidp_data; + }; + struct gmap *gmap; + struct kvm_guestdbg_info_arch guestdbg; + unsigned long pfault_token; + unsigned long pfault_select; + unsigned long pfault_compare; +}; + +struct kvm_vm_stat { + u32 remote_tlb_flush; +}; + +struct kvm_arch_memory_slot { +}; + +struct s390_map_info { + struct list_head list; + __u64 guest_addr; + __u64 addr; + struct page *page; +}; + +struct s390_io_adapter { + unsigned int id; + int isc; + bool maskable; + bool masked; + bool swap; + struct rw_semaphore maps_lock; + struct list_head maps; + atomic_t nr_maps; +}; + +#define MAX_S390_IO_ADAPTERS ((MAX_ISC + 1) * 8) +#define MAX_S390_ADAPTER_MAPS 256 + +/* maximum size of facilities and facility mask is 2k bytes */ +#define S390_ARCH_FAC_LIST_SIZE_BYTE (1<<11) +#define S390_ARCH_FAC_LIST_SIZE_U64 \ + (S390_ARCH_FAC_LIST_SIZE_BYTE / sizeof(u64)) +#define S390_ARCH_FAC_MASK_SIZE_BYTE S390_ARCH_FAC_LIST_SIZE_BYTE +#define S390_ARCH_FAC_MASK_SIZE_U64 \ + (S390_ARCH_FAC_MASK_SIZE_BYTE / sizeof(u64)) + +struct kvm_s390_fac { + /* facility list requested by guest */ + __u64 list[S390_ARCH_FAC_LIST_SIZE_U64]; + /* facility mask supported by kvm & hosting machine */ + __u64 mask[S390_ARCH_FAC_LIST_SIZE_U64]; +}; + +struct kvm_s390_cpu_model { + struct kvm_s390_fac *fac; + struct cpuid cpu_id; + unsigned short ibc; +}; + +struct kvm_s390_crypto { + struct kvm_s390_crypto_cb *crycb; + __u32 crycbd; + __u8 aes_kw; + __u8 dea_kw; +}; + +struct kvm_s390_crypto_cb { + __u8 reserved00[72]; /* 0x0000 */ + __u8 dea_wrapping_key_mask[24]; /* 0x0048 */ + __u8 aes_wrapping_key_mask[32]; /* 0x0060 */ + __u8 reserved80[128]; /* 0x0080 */ +}; + +struct kvm_arch{ + struct sca_block *sca; + debug_info_t *dbf; + struct kvm_s390_float_interrupt float_int; + struct kvm_device *flic; + struct gmap *gmap; + int css_support; + int use_irqchip; + int use_cmma; + int user_cpu_state_ctrl; + int user_sigp; + int user_stsi; + struct s390_io_adapter *adapters[MAX_S390_IO_ADAPTERS]; + wait_queue_head_t ipte_wq; + int ipte_lock_count; + struct mutex ipte_mutex; + spinlock_t start_stop_lock; + struct kvm_s390_cpu_model model; + struct kvm_s390_crypto crypto; + u64 epoch; +}; + +#define KVM_HVA_ERR_BAD (-1UL) +#define KVM_HVA_ERR_RO_BAD (-2UL) + +static inline bool kvm_is_error_hva(unsigned long addr) +{ + return IS_ERR_VALUE(addr); +} + +#define ASYNC_PF_PER_VCPU 64 +struct kvm_arch_async_pf { + unsigned long pfault_token; +}; + +bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu); + +void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, + struct kvm_async_pf *work); + +void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu, + struct kvm_async_pf *work); + +void kvm_arch_async_page_present(struct kvm_vcpu *vcpu, + struct kvm_async_pf *work); + +extern int sie64a(struct kvm_s390_sie_block *, u64 *); +extern char sie_exit; + +static inline void kvm_arch_hardware_disable(void) {} +static inline void kvm_arch_check_processor_compat(void *rtn) {} +static inline void kvm_arch_exit(void) {} +static inline void kvm_arch_sync_events(struct kvm *kvm) {} +static inline void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) {} +static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {} +static inline void kvm_arch_free_memslot(struct kvm *kvm, + struct kvm_memory_slot *free, struct kvm_memory_slot *dont) {} +static inline void kvm_arch_memslots_updated(struct kvm *kvm) {} +static inline void kvm_arch_flush_shadow_all(struct kvm *kvm) {} +static inline void kvm_arch_flush_shadow_memslot(struct kvm *kvm, + struct kvm_memory_slot *slot) {} + +#endif diff --git a/arch/s390/include/asm/kvm_para.h b/arch/s390/include/asm/kvm_para.h new file mode 100644 index 000000000..e0f842308 --- /dev/null +++ b/arch/s390/include/asm/kvm_para.h @@ -0,0 +1,157 @@ +/* + * definition for paravirtual devices on s390 + * + * Copyright IBM Corp. 2008 + * + * 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. + * + * Author(s): Christian Borntraeger <borntraeger@de.ibm.com> + */ +/* + * Hypercalls for KVM on s390. The calling convention is similar to the + * s390 ABI, so we use R2-R6 for parameters 1-5. In addition we use R1 + * as hypercall number and R7 as parameter 6. The return value is + * written to R2. We use the diagnose instruction as hypercall. To avoid + * conflicts with existing diagnoses for LPAR and z/VM, we do not use + * the instruction encoded number, but specify the number in R1 and + * use 0x500 as KVM hypercall + * + * Copyright IBM Corp. 2007,2008 + * Author(s): Christian Borntraeger <borntraeger@de.ibm.com> + * + * This work is licensed under the terms of the GNU GPL, version 2. + */ +#ifndef __S390_KVM_PARA_H +#define __S390_KVM_PARA_H + +#include <uapi/asm/kvm_para.h> + + + +static inline long kvm_hypercall0(unsigned long nr) +{ + register unsigned long __nr asm("1") = nr; + register long __rc asm("2"); + + asm volatile ("diag 2,4,0x500\n" + : "=d" (__rc) : "d" (__nr): "memory", "cc"); + return __rc; +} + +static inline long kvm_hypercall1(unsigned long nr, unsigned long p1) +{ + register unsigned long __nr asm("1") = nr; + register unsigned long __p1 asm("2") = p1; + register long __rc asm("2"); + + asm volatile ("diag 2,4,0x500\n" + : "=d" (__rc) : "d" (__nr), "0" (__p1) : "memory", "cc"); + return __rc; +} + +static inline long kvm_hypercall2(unsigned long nr, unsigned long p1, + unsigned long p2) +{ + register unsigned long __nr asm("1") = nr; + register unsigned long __p1 asm("2") = p1; + register unsigned long __p2 asm("3") = p2; + register long __rc asm("2"); + + asm volatile ("diag 2,4,0x500\n" + : "=d" (__rc) : "d" (__nr), "0" (__p1), "d" (__p2) + : "memory", "cc"); + return __rc; +} + +static inline long kvm_hypercall3(unsigned long nr, unsigned long p1, + unsigned long p2, unsigned long p3) +{ + register unsigned long __nr asm("1") = nr; + register unsigned long __p1 asm("2") = p1; + register unsigned long __p2 asm("3") = p2; + register unsigned long __p3 asm("4") = p3; + register long __rc asm("2"); + + asm volatile ("diag 2,4,0x500\n" + : "=d" (__rc) : "d" (__nr), "0" (__p1), "d" (__p2), + "d" (__p3) : "memory", "cc"); + return __rc; +} + + +static inline long kvm_hypercall4(unsigned long nr, unsigned long p1, + unsigned long p2, unsigned long p3, + unsigned long p4) +{ + register unsigned long __nr asm("1") = nr; + register unsigned long __p1 asm("2") = p1; + register unsigned long __p2 asm("3") = p2; + register unsigned long __p3 asm("4") = p3; + register unsigned long __p4 asm("5") = p4; + register long __rc asm("2"); + + asm volatile ("diag 2,4,0x500\n" + : "=d" (__rc) : "d" (__nr), "0" (__p1), "d" (__p2), + "d" (__p3), "d" (__p4) : "memory", "cc"); + return __rc; +} + +static inline long kvm_hypercall5(unsigned long nr, unsigned long p1, + unsigned long p2, unsigned long p3, + unsigned long p4, unsigned long p5) +{ + register unsigned long __nr asm("1") = nr; + register unsigned long __p1 asm("2") = p1; + register unsigned long __p2 asm("3") = p2; + register unsigned long __p3 asm("4") = p3; + register unsigned long __p4 asm("5") = p4; + register unsigned long __p5 asm("6") = p5; + register long __rc asm("2"); + + asm volatile ("diag 2,4,0x500\n" + : "=d" (__rc) : "d" (__nr), "0" (__p1), "d" (__p2), + "d" (__p3), "d" (__p4), "d" (__p5) : "memory", "cc"); + return __rc; +} + +static inline long kvm_hypercall6(unsigned long nr, unsigned long p1, + unsigned long p2, unsigned long p3, + unsigned long p4, unsigned long p5, + unsigned long p6) +{ + register unsigned long __nr asm("1") = nr; + register unsigned long __p1 asm("2") = p1; + register unsigned long __p2 asm("3") = p2; + register unsigned long __p3 asm("4") = p3; + register unsigned long __p4 asm("5") = p4; + register unsigned long __p5 asm("6") = p5; + register unsigned long __p6 asm("7") = p6; + register long __rc asm("2"); + + asm volatile ("diag 2,4,0x500\n" + : "=d" (__rc) : "d" (__nr), "0" (__p1), "d" (__p2), + "d" (__p3), "d" (__p4), "d" (__p5), "d" (__p6) + : "memory", "cc"); + return __rc; +} + +/* kvm on s390 is always paravirtualization enabled */ +static inline int kvm_para_available(void) +{ + return 1; +} + +/* No feature bits are currently assigned for kvm on s390 */ +static inline unsigned int kvm_arch_para_features(void) +{ + return 0; +} + +static inline bool kvm_check_and_clear_guest_paused(void) +{ + return false; +} + +#endif /* __S390_KVM_PARA_H */ diff --git a/arch/s390/include/asm/linkage.h b/arch/s390/include/asm/linkage.h new file mode 100644 index 000000000..fc8a82847 --- /dev/null +++ b/arch/s390/include/asm/linkage.h @@ -0,0 +1,9 @@ +#ifndef __ASM_LINKAGE_H +#define __ASM_LINKAGE_H + +#include <linux/stringify.h> + +#define __ALIGN .align 4, 0x07 +#define __ALIGN_STR __stringify(__ALIGN) + +#endif diff --git a/arch/s390/include/asm/livepatch.h b/arch/s390/include/asm/livepatch.h new file mode 100644 index 000000000..7aa799134 --- /dev/null +++ b/arch/s390/include/asm/livepatch.h @@ -0,0 +1,43 @@ +/* + * livepatch.h - s390-specific Kernel Live Patching Core + * + * Copyright (c) 2013-2015 SUSE + * Authors: Jiri Kosina + * Vojtech Pavlik + * Jiri Slaby + */ + +/* + * 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. + */ + +#ifndef ASM_LIVEPATCH_H +#define ASM_LIVEPATCH_H + +#include <linux/module.h> + +#ifdef CONFIG_LIVEPATCH +static inline int klp_check_compiler_support(void) +{ + return 0; +} + +static inline int klp_write_module_reloc(struct module *mod, unsigned long + type, unsigned long loc, unsigned long value) +{ + /* not supported yet */ + return -ENOSYS; +} + +static inline void klp_arch_set_pc(struct pt_regs *regs, unsigned long ip) +{ + regs->psw.addr = ip; +} +#else +#error Live patching support is disabled; check CONFIG_LIVEPATCH +#endif + +#endif diff --git a/arch/s390/include/asm/local.h b/arch/s390/include/asm/local.h new file mode 100644 index 000000000..c11c530f7 --- /dev/null +++ b/arch/s390/include/asm/local.h @@ -0,0 +1 @@ +#include <asm-generic/local.h> diff --git a/arch/s390/include/asm/local64.h b/arch/s390/include/asm/local64.h new file mode 100644 index 000000000..36c93b5cc --- /dev/null +++ b/arch/s390/include/asm/local64.h @@ -0,0 +1 @@ +#include <asm-generic/local64.h> diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h new file mode 100644 index 000000000..663f23e37 --- /dev/null +++ b/arch/s390/include/asm/lowcore.h @@ -0,0 +1,217 @@ +/* + * Copyright IBM Corp. 1999, 2012 + * Author(s): Hartmut Penner <hp@de.ibm.com>, + * Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Denis Joseph Barrow, + */ + +#ifndef _ASM_S390_LOWCORE_H +#define _ASM_S390_LOWCORE_H + +#include <linux/types.h> +#include <asm/ptrace.h> +#include <asm/cpu.h> +#include <asm/types.h> + +#define LC_ORDER 1 +#define LC_PAGES 2 + +struct save_area { + u64 fp_regs[16]; + u64 gp_regs[16]; + u8 psw[16]; + u8 pad1[8]; + u32 pref_reg; + u32 fp_ctrl_reg; + u8 pad2[4]; + u32 tod_reg; + u64 timer; + u64 clk_cmp; + u8 pad3[8]; + u32 acc_regs[16]; + u64 ctrl_regs[16]; +} __packed; + +struct save_area_ext { + struct save_area sa; + __vector128 vx_regs[32]; +}; + +struct _lowcore { + __u8 pad_0x0000[0x0014-0x0000]; /* 0x0000 */ + __u32 ipl_parmblock_ptr; /* 0x0014 */ + __u8 pad_0x0018[0x0080-0x0018]; /* 0x0018 */ + __u32 ext_params; /* 0x0080 */ + __u16 ext_cpu_addr; /* 0x0084 */ + __u16 ext_int_code; /* 0x0086 */ + __u16 svc_ilc; /* 0x0088 */ + __u16 svc_code; /* 0x008a */ + __u16 pgm_ilc; /* 0x008c */ + __u16 pgm_code; /* 0x008e */ + __u32 data_exc_code; /* 0x0090 */ + __u16 mon_class_num; /* 0x0094 */ + __u8 per_code; /* 0x0096 */ + __u8 per_atmid; /* 0x0097 */ + __u64 per_address; /* 0x0098 */ + __u8 exc_access_id; /* 0x00a0 */ + __u8 per_access_id; /* 0x00a1 */ + __u8 op_access_id; /* 0x00a2 */ + __u8 ar_mode_id; /* 0x00a3 */ + __u8 pad_0x00a4[0x00a8-0x00a4]; /* 0x00a4 */ + __u64 trans_exc_code; /* 0x00a8 */ + __u64 monitor_code; /* 0x00b0 */ + __u16 subchannel_id; /* 0x00b8 */ + __u16 subchannel_nr; /* 0x00ba */ + __u32 io_int_parm; /* 0x00bc */ + __u32 io_int_word; /* 0x00c0 */ + __u8 pad_0x00c4[0x00c8-0x00c4]; /* 0x00c4 */ + __u32 stfl_fac_list; /* 0x00c8 */ + __u8 pad_0x00cc[0x00e8-0x00cc]; /* 0x00cc */ + __u32 mcck_interruption_code[2]; /* 0x00e8 */ + __u8 pad_0x00f0[0x00f4-0x00f0]; /* 0x00f0 */ + __u32 external_damage_code; /* 0x00f4 */ + __u64 failing_storage_address; /* 0x00f8 */ + __u8 pad_0x0100[0x0110-0x0100]; /* 0x0100 */ + __u64 breaking_event_addr; /* 0x0110 */ + __u8 pad_0x0118[0x0120-0x0118]; /* 0x0118 */ + psw_t restart_old_psw; /* 0x0120 */ + psw_t external_old_psw; /* 0x0130 */ + psw_t svc_old_psw; /* 0x0140 */ + psw_t program_old_psw; /* 0x0150 */ + psw_t mcck_old_psw; /* 0x0160 */ + psw_t io_old_psw; /* 0x0170 */ + __u8 pad_0x0180[0x01a0-0x0180]; /* 0x0180 */ + psw_t restart_psw; /* 0x01a0 */ + psw_t external_new_psw; /* 0x01b0 */ + psw_t svc_new_psw; /* 0x01c0 */ + psw_t program_new_psw; /* 0x01d0 */ + psw_t mcck_new_psw; /* 0x01e0 */ + psw_t io_new_psw; /* 0x01f0 */ + + /* Save areas. */ + __u64 save_area_sync[8]; /* 0x0200 */ + __u64 save_area_async[8]; /* 0x0240 */ + __u64 save_area_restart[1]; /* 0x0280 */ + + /* CPU flags. */ + __u64 cpu_flags; /* 0x0288 */ + + /* Return psws. */ + psw_t return_psw; /* 0x0290 */ + psw_t return_mcck_psw; /* 0x02a0 */ + + /* CPU accounting and timing values. */ + __u64 sync_enter_timer; /* 0x02b0 */ + __u64 async_enter_timer; /* 0x02b8 */ + __u64 mcck_enter_timer; /* 0x02c0 */ + __u64 exit_timer; /* 0x02c8 */ + __u64 user_timer; /* 0x02d0 */ + __u64 system_timer; /* 0x02d8 */ + __u64 steal_timer; /* 0x02e0 */ + __u64 last_update_timer; /* 0x02e8 */ + __u64 last_update_clock; /* 0x02f0 */ + __u64 int_clock; /* 0x02f8 */ + __u64 mcck_clock; /* 0x0300 */ + __u64 clock_comparator; /* 0x0308 */ + + /* Current process. */ + __u64 current_task; /* 0x0310 */ + __u64 thread_info; /* 0x0318 */ + __u64 kernel_stack; /* 0x0320 */ + + /* Interrupt, panic and restart stack. */ + __u64 async_stack; /* 0x0328 */ + __u64 panic_stack; /* 0x0330 */ + __u64 restart_stack; /* 0x0338 */ + + /* Restart function and parameter. */ + __u64 restart_fn; /* 0x0340 */ + __u64 restart_data; /* 0x0348 */ + __u64 restart_source; /* 0x0350 */ + + /* Address space pointer. */ + __u64 kernel_asce; /* 0x0358 */ + __u64 user_asce; /* 0x0360 */ + __u64 current_pid; /* 0x0368 */ + + /* SMP info area */ + __u32 cpu_nr; /* 0x0370 */ + __u32 softirq_pending; /* 0x0374 */ + __u64 percpu_offset; /* 0x0378 */ + __u64 vdso_per_cpu_data; /* 0x0380 */ + __u64 machine_flags; /* 0x0388 */ + __u8 pad_0x0390[0x0398-0x0390]; /* 0x0390 */ + __u64 gmap; /* 0x0398 */ + __u32 spinlock_lockval; /* 0x03a0 */ + __u8 pad_0x03a0[0x0400-0x03a4]; /* 0x03a4 */ + + /* Per cpu primary space access list */ + __u32 paste[16]; /* 0x0400 */ + + __u8 pad_0x04c0[0x0e00-0x0440]; /* 0x0440 */ + + /* + * 0xe00 contains the address of the IPL Parameter Information + * block. Dump tools need IPIB for IPL after dump. + * Note: do not change the position of any fields in 0x0e00-0x0f00 + */ + __u64 ipib; /* 0x0e00 */ + __u32 ipib_checksum; /* 0x0e08 */ + __u64 vmcore_info; /* 0x0e0c */ + __u8 pad_0x0e14[0x0e18-0x0e14]; /* 0x0e14 */ + __u64 os_info; /* 0x0e18 */ + __u8 pad_0x0e20[0x0f00-0x0e20]; /* 0x0e20 */ + + /* Extended facility list */ + __u64 stfle_fac_list[32]; /* 0x0f00 */ + __u8 pad_0x1000[0x11b0-0x1000]; /* 0x1000 */ + + /* Pointer to vector register save area */ + __u64 vector_save_area_addr; /* 0x11b0 */ + + /* 64 bit extparam used for pfault/diag 250: defined by architecture */ + __u64 ext_params2; /* 0x11B8 */ + __u8 pad_0x11c0[0x1200-0x11C0]; /* 0x11C0 */ + + /* CPU register save area: defined by architecture */ + __u64 floating_pt_save_area[16]; /* 0x1200 */ + __u64 gpregs_save_area[16]; /* 0x1280 */ + psw_t psw_save_area; /* 0x1300 */ + __u8 pad_0x1310[0x1318-0x1310]; /* 0x1310 */ + __u32 prefixreg_save_area; /* 0x1318 */ + __u32 fpt_creg_save_area; /* 0x131c */ + __u8 pad_0x1320[0x1324-0x1320]; /* 0x1320 */ + __u32 tod_progreg_save_area; /* 0x1324 */ + __u32 cpu_timer_save_area[2]; /* 0x1328 */ + __u32 clock_comp_save_area[2]; /* 0x1330 */ + __u8 pad_0x1338[0x1340-0x1338]; /* 0x1338 */ + __u32 access_regs_save_area[16]; /* 0x1340 */ + __u64 cregs_save_area[16]; /* 0x1380 */ + __u8 pad_0x1400[0x1800-0x1400]; /* 0x1400 */ + + /* Transaction abort diagnostic block */ + __u8 pgm_tdb[256]; /* 0x1800 */ + __u8 pad_0x1900[0x1c00-0x1900]; /* 0x1900 */ + + /* Software defined save area for vector registers */ + __u8 vector_save_area[1024]; /* 0x1c00 */ +} __packed; + +#define S390_lowcore (*((struct _lowcore *) 0)) + +extern struct _lowcore *lowcore_ptr[]; + +static inline void set_prefix(__u32 address) +{ + asm volatile("spx %0" : : "m" (address) : "memory"); +} + +static inline __u32 store_prefix(void) +{ + __u32 address; + + asm volatile("stpx %0" : "=m" (address)); + return address; +} + +#endif /* _ASM_S390_LOWCORE_H */ diff --git a/arch/s390/include/asm/mathemu.h b/arch/s390/include/asm/mathemu.h new file mode 100644 index 000000000..614dfaf47 --- /dev/null +++ b/arch/s390/include/asm/mathemu.h @@ -0,0 +1,28 @@ +/* + * IEEE floating point emulation. + * + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#ifndef __MATHEMU__ +#define __MATHEMU__ + +extern int math_emu_b3(__u8 *, struct pt_regs *); +extern int math_emu_ed(__u8 *, struct pt_regs *); +extern int math_emu_ldr(__u8 *); +extern int math_emu_ler(__u8 *); +extern int math_emu_std(__u8 *, struct pt_regs *); +extern int math_emu_ld(__u8 *, struct pt_regs *); +extern int math_emu_ste(__u8 *, struct pt_regs *); +extern int math_emu_le(__u8 *, struct pt_regs *); +extern int math_emu_lfpc(__u8 *, struct pt_regs *); +extern int math_emu_stfpc(__u8 *, struct pt_regs *); +extern int math_emu_srnm(__u8 *, struct pt_regs *); + +#endif /* __MATHEMU__ */ + + + + diff --git a/arch/s390/include/asm/mman.h b/arch/s390/include/asm/mman.h new file mode 100644 index 000000000..b55a59e1d --- /dev/null +++ b/arch/s390/include/asm/mman.h @@ -0,0 +1,15 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/mman.h" + */ +#ifndef __S390_MMAN_H__ +#define __S390_MMAN_H__ + +#include <uapi/asm/mman.h> + +#ifndef __ASSEMBLY__ +int s390_mmap_check(unsigned long addr, unsigned long len, unsigned long flags); +#define arch_mmap_check(addr, len, flags) s390_mmap_check(addr, len, flags) +#endif +#endif /* __S390_MMAN_H__ */ diff --git a/arch/s390/include/asm/mmu.h b/arch/s390/include/asm/mmu.h new file mode 100644 index 000000000..d29ad9545 --- /dev/null +++ b/arch/s390/include/asm/mmu.h @@ -0,0 +1,44 @@ +#ifndef __MMU_H +#define __MMU_H + +#include <linux/cpumask.h> +#include <linux/errno.h> + +typedef struct { + cpumask_t cpu_attach_mask; + atomic_t attach_count; + unsigned int flush_mm; + spinlock_t list_lock; + struct list_head pgtable_list; + struct list_head gmap_list; + unsigned long asce_bits; + unsigned long asce_limit; + unsigned long vdso_base; + /* The mmu context allocates 4K page tables. */ + unsigned int alloc_pgste:1; + /* The mmu context uses extended page tables. */ + unsigned int has_pgste:1; + /* The mmu context uses storage keys. */ + unsigned int use_skey:1; +} mm_context_t; + +#define INIT_MM_CONTEXT(name) \ + .context.list_lock = __SPIN_LOCK_UNLOCKED(name.context.list_lock), \ + .context.pgtable_list = LIST_HEAD_INIT(name.context.pgtable_list), \ + .context.gmap_list = LIST_HEAD_INIT(name.context.gmap_list), + +static inline int tprot(unsigned long addr) +{ + int rc = -EFAULT; + + asm volatile( + " tprot 0(%1),0\n" + "0: ipm %0\n" + " srl %0,28\n" + "1:\n" + EX_TABLE(0b,1b) + : "+d" (rc) : "a" (addr) : "cc"); + return rc; +} + +#endif diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h new file mode 100644 index 000000000..fb1b93ea3 --- /dev/null +++ b/arch/s390/include/asm/mmu_context.h @@ -0,0 +1,133 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/mmu_context.h" + */ + +#ifndef __S390_MMU_CONTEXT_H +#define __S390_MMU_CONTEXT_H + +#include <asm/pgalloc.h> +#include <asm/uaccess.h> +#include <asm/tlbflush.h> +#include <asm/ctl_reg.h> + +static inline int init_new_context(struct task_struct *tsk, + struct mm_struct *mm) +{ + cpumask_clear(&mm->context.cpu_attach_mask); + atomic_set(&mm->context.attach_count, 0); + mm->context.flush_mm = 0; + mm->context.asce_bits = _ASCE_TABLE_LENGTH | _ASCE_USER_BITS; + mm->context.asce_bits |= _ASCE_TYPE_REGION3; +#ifdef CONFIG_PGSTE + mm->context.alloc_pgste = page_table_allocate_pgste; + mm->context.has_pgste = 0; + mm->context.use_skey = 0; +#endif + mm->context.asce_limit = STACK_TOP_MAX; + crst_table_init((unsigned long *) mm->pgd, pgd_entry_type(mm)); + return 0; +} + +#define destroy_context(mm) do { } while (0) + +static inline void set_user_asce(struct mm_struct *mm) +{ + S390_lowcore.user_asce = mm->context.asce_bits | __pa(mm->pgd); + if (current->thread.mm_segment.ar4) + __ctl_load(S390_lowcore.user_asce, 7, 7); + set_cpu_flag(CIF_ASCE); +} + +static inline void clear_user_asce(void) +{ + S390_lowcore.user_asce = S390_lowcore.kernel_asce; + + __ctl_load(S390_lowcore.user_asce, 1, 1); + __ctl_load(S390_lowcore.user_asce, 7, 7); +} + +static inline void load_kernel_asce(void) +{ + unsigned long asce; + + __ctl_store(asce, 1, 1); + if (asce != S390_lowcore.kernel_asce) + __ctl_load(S390_lowcore.kernel_asce, 1, 1); + set_cpu_flag(CIF_ASCE); +} + +static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, + struct task_struct *tsk) +{ + int cpu = smp_processor_id(); + + S390_lowcore.user_asce = next->context.asce_bits | __pa(next->pgd); + if (prev == next) + return; + if (MACHINE_HAS_TLB_LC) + cpumask_set_cpu(cpu, &next->context.cpu_attach_mask); + /* Clear old ASCE by loading the kernel ASCE. */ + __ctl_load(S390_lowcore.kernel_asce, 1, 1); + __ctl_load(S390_lowcore.kernel_asce, 7, 7); + atomic_inc(&next->context.attach_count); + atomic_dec(&prev->context.attach_count); + if (MACHINE_HAS_TLB_LC) + cpumask_clear_cpu(cpu, &prev->context.cpu_attach_mask); +} + +#define finish_arch_post_lock_switch finish_arch_post_lock_switch +static inline void finish_arch_post_lock_switch(void) +{ + struct task_struct *tsk = current; + struct mm_struct *mm = tsk->mm; + + load_kernel_asce(); + if (mm) { + preempt_disable(); + while (atomic_read(&mm->context.attach_count) >> 16) + cpu_relax(); + + cpumask_set_cpu(smp_processor_id(), mm_cpumask(mm)); + if (mm->context.flush_mm) + __tlb_flush_mm(mm); + preempt_enable(); + } + set_fs(current->thread.mm_segment); +} + +#define enter_lazy_tlb(mm,tsk) do { } while (0) +#define deactivate_mm(tsk,mm) do { } while (0) + +static inline void activate_mm(struct mm_struct *prev, + struct mm_struct *next) +{ + switch_mm(prev, next, current); + cpumask_set_cpu(smp_processor_id(), mm_cpumask(next)); + set_user_asce(next); +} + +static inline void arch_dup_mmap(struct mm_struct *oldmm, + struct mm_struct *mm) +{ + if (oldmm->context.asce_limit < mm->context.asce_limit) + crst_table_downgrade(mm, oldmm->context.asce_limit); +} + +static inline void arch_exit_mmap(struct mm_struct *mm) +{ +} + +static inline void arch_unmap(struct mm_struct *mm, + struct vm_area_struct *vma, + unsigned long start, unsigned long end) +{ +} + +static inline void arch_bprm_mm_init(struct mm_struct *mm, + struct vm_area_struct *vma) +{ +} + +#endif /* __S390_MMU_CONTEXT_H */ diff --git a/arch/s390/include/asm/module.h b/arch/s390/include/asm/module.h new file mode 100644 index 000000000..df1f861a8 --- /dev/null +++ b/arch/s390/include/asm/module.h @@ -0,0 +1,34 @@ +#ifndef _ASM_S390_MODULE_H +#define _ASM_S390_MODULE_H + +#include <asm-generic/module.h> + +/* + * This file contains the s390 architecture specific module code. + */ + +struct mod_arch_syminfo +{ + unsigned long got_offset; + unsigned long plt_offset; + int got_initialized; + int plt_initialized; +}; + +struct mod_arch_specific +{ + /* Starting offset of got in the module core memory. */ + unsigned long got_offset; + /* Starting offset of plt in the module core memory. */ + unsigned long plt_offset; + /* Size of the got. */ + unsigned long got_size; + /* Size of the plt. */ + unsigned long plt_size; + /* Number of symbols in syminfo. */ + int nsyms; + /* Additional symbol information (got and plt offsets). */ + struct mod_arch_syminfo *syminfo; +}; + +#endif /* _ASM_S390_MODULE_H */ diff --git a/arch/s390/include/asm/mutex.h b/arch/s390/include/asm/mutex.h new file mode 100644 index 000000000..458c1f7fb --- /dev/null +++ b/arch/s390/include/asm/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/arch/s390/include/asm/nmi.h b/arch/s390/include/asm/nmi.h new file mode 100644 index 000000000..3027a5a72 --- /dev/null +++ b/arch/s390/include/asm/nmi.h @@ -0,0 +1,66 @@ +/* + * Machine check handler definitions + * + * Copyright IBM Corp. 2000, 2009 + * Author(s): Ingo Adlung <adlung@de.ibm.com>, + * Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Cornelia Huck <cornelia.huck@de.ibm.com>, + * Heiko Carstens <heiko.carstens@de.ibm.com>, + */ + +#ifndef _ASM_S390_NMI_H +#define _ASM_S390_NMI_H + +#include <linux/types.h> + +struct mci { + __u32 sd : 1; /* 00 system damage */ + __u32 pd : 1; /* 01 instruction-processing damage */ + __u32 sr : 1; /* 02 system recovery */ + __u32 : 1; /* 03 */ + __u32 cd : 1; /* 04 timing-facility damage */ + __u32 ed : 1; /* 05 external damage */ + __u32 : 1; /* 06 */ + __u32 dg : 1; /* 07 degradation */ + __u32 w : 1; /* 08 warning pending */ + __u32 cp : 1; /* 09 channel-report pending */ + __u32 sp : 1; /* 10 service-processor damage */ + __u32 ck : 1; /* 11 channel-subsystem damage */ + __u32 : 2; /* 12-13 */ + __u32 b : 1; /* 14 backed up */ + __u32 : 1; /* 15 */ + __u32 se : 1; /* 16 storage error uncorrected */ + __u32 sc : 1; /* 17 storage error corrected */ + __u32 ke : 1; /* 18 storage-key error uncorrected */ + __u32 ds : 1; /* 19 storage degradation */ + __u32 wp : 1; /* 20 psw mwp validity */ + __u32 ms : 1; /* 21 psw mask and key validity */ + __u32 pm : 1; /* 22 psw program mask and cc validity */ + __u32 ia : 1; /* 23 psw instruction address validity */ + __u32 fa : 1; /* 24 failing storage address validity */ + __u32 vr : 1; /* 25 vector register validity */ + __u32 ec : 1; /* 26 external damage code validity */ + __u32 fp : 1; /* 27 floating point register validity */ + __u32 gr : 1; /* 28 general register validity */ + __u32 cr : 1; /* 29 control register validity */ + __u32 : 1; /* 30 */ + __u32 st : 1; /* 31 storage logical validity */ + __u32 ie : 1; /* 32 indirect storage error */ + __u32 ar : 1; /* 33 access register validity */ + __u32 da : 1; /* 34 delayed access exception */ + __u32 : 7; /* 35-41 */ + __u32 pr : 1; /* 42 tod programmable register validity */ + __u32 fc : 1; /* 43 fp control register validity */ + __u32 ap : 1; /* 44 ancillary report */ + __u32 : 1; /* 45 */ + __u32 ct : 1; /* 46 cpu timer validity */ + __u32 cc : 1; /* 47 clock comparator validity */ + __u32 : 16; /* 47-63 */ +}; + +struct pt_regs; + +extern void s390_handle_mcck(void); +extern void s390_do_machine_check(struct pt_regs *regs); + +#endif /* _ASM_S390_NMI_H */ diff --git a/arch/s390/include/asm/os_info.h b/arch/s390/include/asm/os_info.h new file mode 100644 index 000000000..295f2c4f1 --- /dev/null +++ b/arch/s390/include/asm/os_info.h @@ -0,0 +1,49 @@ +/* + * OS info memory interface + * + * Copyright IBM Corp. 2012 + * Author(s): Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ +#ifndef _ASM_S390_OS_INFO_H +#define _ASM_S390_OS_INFO_H + +#define OS_INFO_VERSION_MAJOR 1 +#define OS_INFO_VERSION_MINOR 1 +#define OS_INFO_MAGIC 0x4f53494e464f535aULL /* OSINFOSZ */ + +#define OS_INFO_VMCOREINFO 0 +#define OS_INFO_REIPL_BLOCK 1 + +struct os_info_entry { + u64 addr; + u64 size; + u32 csum; +} __packed; + +struct os_info { + u64 magic; + u32 csum; + u16 version_major; + u16 version_minor; + u64 crashkernel_addr; + u64 crashkernel_size; + struct os_info_entry entry[2]; + u8 reserved[4024]; +} __packed; + +void os_info_init(void); +void os_info_entry_add(int nr, void *ptr, u64 len); +void os_info_crashkernel_add(unsigned long base, unsigned long size); +u32 os_info_csum(struct os_info *os_info); + +#ifdef CONFIG_CRASH_DUMP +void *os_info_old_entry(int nr, unsigned long *size); +int copy_from_oldmem(void *dest, void *src, size_t count); +#else +static inline void *os_info_old_entry(int nr, unsigned long *size) +{ + return NULL; +} +#endif + +#endif /* _ASM_S390_OS_INFO_H */ diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h new file mode 100644 index 000000000..53eacbd4f --- /dev/null +++ b/arch/s390/include/asm/page.h @@ -0,0 +1,156 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + * Author(s): Hartmut Penner (hp@de.ibm.com) + */ + +#ifndef _S390_PAGE_H +#define _S390_PAGE_H + +#include <linux/const.h> +#include <asm/types.h> + +/* PAGE_SHIFT determines the page size */ +#define PAGE_SHIFT 12 +#define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT) +#define PAGE_MASK (~(PAGE_SIZE-1)) +#define PAGE_DEFAULT_ACC 0 +#define PAGE_DEFAULT_KEY (PAGE_DEFAULT_ACC << 4) + +#define HPAGE_SHIFT 20 +#define HPAGE_SIZE (1UL << HPAGE_SHIFT) +#define HPAGE_MASK (~(HPAGE_SIZE - 1)) +#define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) + +#define ARCH_HAS_SETCLEAR_HUGE_PTE +#define ARCH_HAS_HUGE_PTE_TYPE +#define ARCH_HAS_PREPARE_HUGEPAGE +#define ARCH_HAS_HUGEPAGE_CLEAR_FLUSH + +#include <asm/setup.h> +#ifndef __ASSEMBLY__ + +static inline void storage_key_init_range(unsigned long start, unsigned long end) +{ +#if PAGE_DEFAULT_KEY + __storage_key_init_range(start, end); +#endif +} + +#define clear_page(page) memset((page), 0, PAGE_SIZE) + +/* + * copy_page uses the mvcl instruction with 0xb0 padding byte in order to + * bypass caches when copying a page. Especially when copying huge pages + * this keeps L1 and L2 data caches alive. + */ +static inline void copy_page(void *to, void *from) +{ + register void *reg2 asm ("2") = to; + register unsigned long reg3 asm ("3") = 0x1000; + register void *reg4 asm ("4") = from; + register unsigned long reg5 asm ("5") = 0xb0001000; + asm volatile( + " mvcl 2,4" + : "+d" (reg2), "+d" (reg3), "+d" (reg4), "+d" (reg5) + : : "memory", "cc"); +} + +#define clear_user_page(page, vaddr, pg) clear_page(page) +#define copy_user_page(to, from, vaddr, pg) copy_page(to, from) + +#define __alloc_zeroed_user_highpage(movableflags, vma, vaddr) \ + alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO | movableflags, vma, vaddr) +#define __HAVE_ARCH_ALLOC_ZEROED_USER_HIGHPAGE + +/* + * These are used to make use of C type-checking.. + */ + +typedef struct { unsigned long pgprot; } pgprot_t; +typedef struct { unsigned long pgste; } pgste_t; +typedef struct { unsigned long pte; } pte_t; +typedef struct { unsigned long pmd; } pmd_t; +typedef struct { unsigned long pud; } pud_t; +typedef struct { unsigned long pgd; } pgd_t; +typedef pte_t *pgtable_t; + +#define pgprot_val(x) ((x).pgprot) +#define pgste_val(x) ((x).pgste) +#define pte_val(x) ((x).pte) +#define pmd_val(x) ((x).pmd) +#define pud_val(x) ((x).pud) +#define pgd_val(x) ((x).pgd) + +#define __pgste(x) ((pgste_t) { (x) } ) +#define __pte(x) ((pte_t) { (x) } ) +#define __pmd(x) ((pmd_t) { (x) } ) +#define __pud(x) ((pud_t) { (x) } ) +#define __pgd(x) ((pgd_t) { (x) } ) +#define __pgprot(x) ((pgprot_t) { (x) } ) + +static inline void page_set_storage_key(unsigned long addr, + unsigned char skey, int mapped) +{ + if (!mapped) + asm volatile(".insn rrf,0xb22b0000,%0,%1,8,0" + : : "d" (skey), "a" (addr)); + else + asm volatile("sske %0,%1" : : "d" (skey), "a" (addr)); +} + +static inline unsigned char page_get_storage_key(unsigned long addr) +{ + unsigned char skey; + + asm volatile("iske %0,%1" : "=d" (skey) : "a" (addr)); + return skey; +} + +static inline int page_reset_referenced(unsigned long addr) +{ + unsigned int ipm; + + asm volatile( + " rrbe 0,%1\n" + " ipm %0\n" + : "=d" (ipm) : "a" (addr) : "cc"); + return !!(ipm & 0x20000000); +} + +/* Bits int the storage key */ +#define _PAGE_CHANGED 0x02 /* HW changed bit */ +#define _PAGE_REFERENCED 0x04 /* HW referenced bit */ +#define _PAGE_FP_BIT 0x08 /* HW fetch protection bit */ +#define _PAGE_ACC_BITS 0xf0 /* HW access control bits */ + +struct page; +void arch_free_page(struct page *page, int order); +void arch_alloc_page(struct page *page, int order); +void arch_set_page_states(int make_stable); + +static inline int devmem_is_allowed(unsigned long pfn) +{ + return 0; +} + +#define HAVE_ARCH_FREE_PAGE +#define HAVE_ARCH_ALLOC_PAGE + +#endif /* !__ASSEMBLY__ */ + +#define __PAGE_OFFSET 0x0UL +#define PAGE_OFFSET 0x0UL +#define __pa(x) (unsigned long)(x) +#define __va(x) (void *)(unsigned long)(x) +#define virt_to_page(kaddr) pfn_to_page(__pa(kaddr) >> PAGE_SHIFT) +#define page_to_phys(page) (page_to_pfn(page) << PAGE_SHIFT) +#define virt_addr_valid(kaddr) pfn_valid(__pa(kaddr) >> PAGE_SHIFT) + +#define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + +#include <asm-generic/memory_model.h> +#include <asm-generic/getorder.h> + +#endif /* _S390_PAGE_H */ diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h new file mode 100644 index 000000000..a648338c4 --- /dev/null +++ b/arch/s390/include/asm/pci.h @@ -0,0 +1,191 @@ +#ifndef __ASM_S390_PCI_H +#define __ASM_S390_PCI_H + +/* must be set before including asm-generic/pci.h */ +#define PCI_DMA_BUS_IS_PHYS (0) +/* must be set before including pci_clp.h */ +#define PCI_BAR_COUNT 6 + +#include <linux/pci.h> +#include <linux/mutex.h> +#include <asm-generic/pci.h> +#include <asm-generic/pci-dma-compat.h> +#include <asm/pci_clp.h> +#include <asm/pci_debug.h> + +#define PCIBIOS_MIN_IO 0x1000 +#define PCIBIOS_MIN_MEM 0x10000000 + +#define pcibios_assign_all_busses() (0) + +void __iomem *pci_iomap(struct pci_dev *, int, unsigned long); +void pci_iounmap(struct pci_dev *, void __iomem *); +int pci_domain_nr(struct pci_bus *); +int pci_proc_domain(struct pci_bus *); + +#define ZPCI_BUS_NR 0 /* default bus number */ +#define ZPCI_DEVFN 0 /* default device number */ + +/* PCI Function Controls */ +#define ZPCI_FC_FN_ENABLED 0x80 +#define ZPCI_FC_ERROR 0x40 +#define ZPCI_FC_BLOCKED 0x20 +#define ZPCI_FC_DMA_ENABLED 0x10 + +struct zpci_fmb { + u32 format : 8; + u32 dma_valid : 1; + u32 : 23; + u32 samples; + u64 last_update; + /* hardware counters */ + u64 ld_ops; + u64 st_ops; + u64 stb_ops; + u64 rpcit_ops; + u64 dma_rbytes; + u64 dma_wbytes; +} __packed __aligned(16); + +enum zpci_state { + ZPCI_FN_STATE_RESERVED, + ZPCI_FN_STATE_STANDBY, + ZPCI_FN_STATE_CONFIGURED, + ZPCI_FN_STATE_ONLINE, + NR_ZPCI_FN_STATES, +}; + +struct zpci_bar_struct { + struct resource *res; /* bus resource */ + u32 val; /* bar start & 3 flag bits */ + u16 map_idx; /* index into bar mapping array */ + u8 size; /* order 2 exponent */ +}; + +/* Private data per function */ +struct zpci_dev { + struct pci_dev *pdev; + struct pci_bus *bus; + struct list_head entry; /* list of all zpci_devices, needed for hotplug, etc. */ + + enum zpci_state state; + u32 fid; /* function ID, used by sclp */ + u32 fh; /* function handle, used by insn's */ + u16 vfn; /* virtual function number */ + u16 pchid; /* physical channel ID */ + u8 pfgid; /* function group ID */ + u8 pft; /* pci function type */ + u16 domain; + + struct mutex lock; + u8 pfip[CLP_PFIP_NR_SEGMENTS]; /* pci function internal path */ + u32 uid; /* user defined id */ + u8 util_str[CLP_UTIL_STR_LEN]; /* utility string */ + + /* IRQ stuff */ + u64 msi_addr; /* MSI address */ + unsigned int max_msi; /* maximum number of MSI's */ + struct airq_iv *aibv; /* adapter interrupt bit vector */ + unsigned int aisb; /* number of the summary bit */ + + /* DMA stuff */ + unsigned long *dma_table; + spinlock_t dma_table_lock; + int tlb_refresh; + + spinlock_t iommu_bitmap_lock; + unsigned long *iommu_bitmap; + unsigned long iommu_size; + unsigned long iommu_pages; + unsigned int next_bit; + + char res_name[16]; + struct zpci_bar_struct bars[PCI_BAR_COUNT]; + + u64 start_dma; /* Start of available DMA addresses */ + u64 end_dma; /* End of available DMA addresses */ + u64 dma_mask; /* DMA address space mask */ + + /* Function measurement block */ + struct zpci_fmb *fmb; + u16 fmb_update; /* update interval */ + /* software counters */ + atomic64_t allocated_pages; + atomic64_t mapped_pages; + atomic64_t unmapped_pages; + + enum pci_bus_speed max_bus_speed; + + struct dentry *debugfs_dev; + struct dentry *debugfs_perf; +}; + +static inline bool zdev_enabled(struct zpci_dev *zdev) +{ + return (zdev->fh & (1UL << 31)) ? true : false; +} + +extern const struct attribute_group *zpci_attr_groups[]; + +/* ----------------------------------------------------------------------------- + Prototypes +----------------------------------------------------------------------------- */ +/* Base stuff */ +int zpci_create_device(struct zpci_dev *); +int zpci_enable_device(struct zpci_dev *); +int zpci_disable_device(struct zpci_dev *); +void zpci_stop_device(struct zpci_dev *); +int zpci_register_ioat(struct zpci_dev *, u8, u64, u64, u64); +int zpci_unregister_ioat(struct zpci_dev *, u8); + +/* CLP */ +int clp_scan_pci_devices(void); +int clp_rescan_pci_devices(void); +int clp_rescan_pci_devices_simple(void); +int clp_add_pci_device(u32, u32, int); +int clp_enable_fh(struct zpci_dev *, u8); +int clp_disable_fh(struct zpci_dev *); + +#ifdef CONFIG_PCI +/* Error handling and recovery */ +void zpci_event_error(void *); +void zpci_event_availability(void *); +void zpci_rescan(void); +bool zpci_is_enabled(void); +#else /* CONFIG_PCI */ +static inline void zpci_event_error(void *e) {} +static inline void zpci_event_availability(void *e) {} +static inline void zpci_rescan(void) {} +#endif /* CONFIG_PCI */ + +#ifdef CONFIG_HOTPLUG_PCI_S390 +int zpci_init_slot(struct zpci_dev *); +void zpci_exit_slot(struct zpci_dev *); +#else /* CONFIG_HOTPLUG_PCI_S390 */ +static inline int zpci_init_slot(struct zpci_dev *zdev) +{ + return 0; +} +static inline void zpci_exit_slot(struct zpci_dev *zdev) {} +#endif /* CONFIG_HOTPLUG_PCI_S390 */ + +/* Helpers */ +struct zpci_dev *get_zdev(struct pci_dev *); +struct zpci_dev *get_zdev_by_fid(u32); + +/* DMA */ +int zpci_dma_init(void); +void zpci_dma_exit(void); + +/* FMB */ +int zpci_fmb_enable_device(struct zpci_dev *); +int zpci_fmb_disable_device(struct zpci_dev *); + +/* Debug */ +int zpci_debug_init(void); +void zpci_debug_exit(void); +void zpci_debug_init_device(struct zpci_dev *); +void zpci_debug_exit_device(struct zpci_dev *); +void zpci_debug_info(struct zpci_dev *, struct seq_file *); + +#endif diff --git a/arch/s390/include/asm/pci_clp.h b/arch/s390/include/asm/pci_clp.h new file mode 100644 index 000000000..dd78f92f1 --- /dev/null +++ b/arch/s390/include/asm/pci_clp.h @@ -0,0 +1,186 @@ +#ifndef _ASM_S390_PCI_CLP_H +#define _ASM_S390_PCI_CLP_H + +#include <asm/clp.h> + +/* + * Call Logical Processor - Command Codes + */ +#define CLP_LIST_PCI 0x0002 +#define CLP_QUERY_PCI_FN 0x0003 +#define CLP_QUERY_PCI_FNGRP 0x0004 +#define CLP_SET_PCI_FN 0x0005 + +/* PCI function handle list entry */ +struct clp_fh_list_entry { + u16 device_id; + u16 vendor_id; + u32 config_state : 1; + u32 : 31; + u32 fid; /* PCI function id */ + u32 fh; /* PCI function handle */ +} __packed; + +#define CLP_RC_SETPCIFN_FH 0x0101 /* Invalid PCI fn handle */ +#define CLP_RC_SETPCIFN_FHOP 0x0102 /* Fn handle not valid for op */ +#define CLP_RC_SETPCIFN_DMAAS 0x0103 /* Invalid DMA addr space */ +#define CLP_RC_SETPCIFN_RES 0x0104 /* Insufficient resources */ +#define CLP_RC_SETPCIFN_ALRDY 0x0105 /* Fn already in requested state */ +#define CLP_RC_SETPCIFN_ERR 0x0106 /* Fn in permanent error state */ +#define CLP_RC_SETPCIFN_RECPND 0x0107 /* Error recovery pending */ +#define CLP_RC_SETPCIFN_BUSY 0x0108 /* Fn busy */ +#define CLP_RC_LISTPCI_BADRT 0x010a /* Resume token not recognized */ +#define CLP_RC_QUERYPCIFG_PFGID 0x010b /* Unrecognized PFGID */ + +/* request or response block header length */ +#define LIST_PCI_HDR_LEN 32 + +/* Number of function handles fitting in response block */ +#define CLP_FH_LIST_NR_ENTRIES \ + ((CLP_BLK_SIZE - 2 * LIST_PCI_HDR_LEN) \ + / sizeof(struct clp_fh_list_entry)) + +#define CLP_SET_ENABLE_PCI_FN 0 /* Yes, 0 enables it */ +#define CLP_SET_DISABLE_PCI_FN 1 /* Yes, 1 disables it */ + +#define CLP_UTIL_STR_LEN 64 +#define CLP_PFIP_NR_SEGMENTS 4 + +/* List PCI functions request */ +struct clp_req_list_pci { + struct clp_req_hdr hdr; + u32 fmt : 4; /* cmd request block format */ + u32 : 28; + u64 reserved1; + u64 resume_token; + u64 reserved2; +} __packed; + +/* List PCI functions response */ +struct clp_rsp_list_pci { + struct clp_rsp_hdr hdr; + u32 fmt : 4; /* cmd request block format */ + u32 : 28; + u64 reserved1; + u64 resume_token; + u32 reserved2; + u16 max_fn; + u8 reserved3; + u8 entry_size; + struct clp_fh_list_entry fh_list[CLP_FH_LIST_NR_ENTRIES]; +} __packed; + +/* Query PCI function request */ +struct clp_req_query_pci { + struct clp_req_hdr hdr; + u32 fmt : 4; /* cmd request block format */ + u32 : 28; + u64 reserved1; + u32 fh; /* function handle */ + u32 reserved2; + u64 reserved3; +} __packed; + +/* Query PCI function response */ +struct clp_rsp_query_pci { + struct clp_rsp_hdr hdr; + u32 fmt : 4; /* cmd request block format */ + u32 : 28; + u64 : 64; + u16 vfn; /* virtual fn number */ + u16 : 7; + u16 util_str_avail : 1; /* utility string available? */ + u16 pfgid : 8; /* pci function group id */ + u32 fid; /* pci function id */ + u8 bar_size[PCI_BAR_COUNT]; + u16 pchid; + u32 bar[PCI_BAR_COUNT]; + u8 pfip[CLP_PFIP_NR_SEGMENTS]; /* pci function internal path */ + u32 : 24; + u8 pft; /* pci function type */ + u64 sdma; /* start dma as */ + u64 edma; /* end dma as */ + u32 reserved[11]; + u32 uid; /* user defined id */ + u8 util_str[CLP_UTIL_STR_LEN]; /* utility string */ +} __packed; + +/* Query PCI function group request */ +struct clp_req_query_pci_grp { + struct clp_req_hdr hdr; + u32 fmt : 4; /* cmd request block format */ + u32 : 28; + u64 reserved1; + u32 : 24; + u32 pfgid : 8; /* function group id */ + u32 reserved2; + u64 reserved3; +} __packed; + +/* Query PCI function group response */ +struct clp_rsp_query_pci_grp { + struct clp_rsp_hdr hdr; + u32 fmt : 4; /* cmd request block format */ + u32 : 28; + u64 reserved1; + u16 : 4; + u16 noi : 12; /* number of interrupts */ + u8 version; + u8 : 6; + u8 frame : 1; + u8 refresh : 1; /* TLB refresh mode */ + u16 reserved2; + u16 mui; + u64 reserved3; + u64 dasm; /* dma address space mask */ + u64 msia; /* MSI address */ + u64 reserved4; + u64 reserved5; +} __packed; + +/* Set PCI function request */ +struct clp_req_set_pci { + struct clp_req_hdr hdr; + u32 fmt : 4; /* cmd request block format */ + u32 : 28; + u64 reserved1; + u32 fh; /* function handle */ + u16 reserved2; + u8 oc; /* operation controls */ + u8 ndas; /* number of dma spaces */ + u64 reserved3; +} __packed; + +/* Set PCI function response */ +struct clp_rsp_set_pci { + struct clp_rsp_hdr hdr; + u32 fmt : 4; /* cmd request block format */ + u32 : 28; + u64 reserved1; + u32 fh; /* function handle */ + u32 reserved3; + u64 reserved4; +} __packed; + +/* Combined request/response block structures used by clp insn */ +struct clp_req_rsp_list_pci { + struct clp_req_list_pci request; + struct clp_rsp_list_pci response; +} __packed; + +struct clp_req_rsp_set_pci { + struct clp_req_set_pci request; + struct clp_rsp_set_pci response; +} __packed; + +struct clp_req_rsp_query_pci { + struct clp_req_query_pci request; + struct clp_rsp_query_pci response; +} __packed; + +struct clp_req_rsp_query_pci_grp { + struct clp_req_query_pci_grp request; + struct clp_rsp_query_pci_grp response; +} __packed; + +#endif diff --git a/arch/s390/include/asm/pci_debug.h b/arch/s390/include/asm/pci_debug.h new file mode 100644 index 000000000..ac24b26fc --- /dev/null +++ b/arch/s390/include/asm/pci_debug.h @@ -0,0 +1,28 @@ +#ifndef _S390_ASM_PCI_DEBUG_H +#define _S390_ASM_PCI_DEBUG_H + +#include <asm/debug.h> + +extern debug_info_t *pci_debug_msg_id; +extern debug_info_t *pci_debug_err_id; + +#define zpci_dbg(imp, fmt, args...) \ + debug_sprintf_event(pci_debug_msg_id, imp, fmt, ##args) + +#define zpci_err(text...) \ + do { \ + char debug_buffer[16]; \ + snprintf(debug_buffer, 16, text); \ + debug_text_event(pci_debug_err_id, 0, debug_buffer); \ + } while (0) + +static inline void zpci_err_hex(void *addr, int len) +{ + while (len > 0) { + debug_event(pci_debug_err_id, 0, (void *) addr, len); + len -= pci_debug_err_id->buf_size; + addr += pci_debug_err_id->buf_size; + } +} + +#endif diff --git a/arch/s390/include/asm/pci_dma.h b/arch/s390/include/asm/pci_dma.h new file mode 100644 index 000000000..30b4c179c --- /dev/null +++ b/arch/s390/include/asm/pci_dma.h @@ -0,0 +1,196 @@ +#ifndef _ASM_S390_PCI_DMA_H +#define _ASM_S390_PCI_DMA_H + +/* I/O Translation Anchor (IOTA) */ +enum zpci_ioat_dtype { + ZPCI_IOTA_STO = 0, + ZPCI_IOTA_RTTO = 1, + ZPCI_IOTA_RSTO = 2, + ZPCI_IOTA_RFTO = 3, + ZPCI_IOTA_PFAA = 4, + ZPCI_IOTA_IOPFAA = 5, + ZPCI_IOTA_IOPTO = 7 +}; + +#define ZPCI_IOTA_IOT_ENABLED 0x800UL +#define ZPCI_IOTA_DT_ST (ZPCI_IOTA_STO << 2) +#define ZPCI_IOTA_DT_RT (ZPCI_IOTA_RTTO << 2) +#define ZPCI_IOTA_DT_RS (ZPCI_IOTA_RSTO << 2) +#define ZPCI_IOTA_DT_RF (ZPCI_IOTA_RFTO << 2) +#define ZPCI_IOTA_DT_PF (ZPCI_IOTA_PFAA << 2) +#define ZPCI_IOTA_FS_4K 0 +#define ZPCI_IOTA_FS_1M 1 +#define ZPCI_IOTA_FS_2G 2 +#define ZPCI_KEY (PAGE_DEFAULT_KEY << 5) + +#define ZPCI_IOTA_STO_FLAG (ZPCI_IOTA_IOT_ENABLED | ZPCI_KEY | ZPCI_IOTA_DT_ST) +#define ZPCI_IOTA_RTTO_FLAG (ZPCI_IOTA_IOT_ENABLED | ZPCI_KEY | ZPCI_IOTA_DT_RT) +#define ZPCI_IOTA_RSTO_FLAG (ZPCI_IOTA_IOT_ENABLED | ZPCI_KEY | ZPCI_IOTA_DT_RS) +#define ZPCI_IOTA_RFTO_FLAG (ZPCI_IOTA_IOT_ENABLED | ZPCI_KEY | ZPCI_IOTA_DT_RF) +#define ZPCI_IOTA_RFAA_FLAG (ZPCI_IOTA_IOT_ENABLED | ZPCI_KEY | ZPCI_IOTA_DT_PF | ZPCI_IOTA_FS_2G) + +/* I/O Region and segment tables */ +#define ZPCI_INDEX_MASK 0x7ffUL + +#define ZPCI_TABLE_TYPE_MASK 0xc +#define ZPCI_TABLE_TYPE_RFX 0xc +#define ZPCI_TABLE_TYPE_RSX 0x8 +#define ZPCI_TABLE_TYPE_RTX 0x4 +#define ZPCI_TABLE_TYPE_SX 0x0 + +#define ZPCI_TABLE_LEN_RFX 0x3 +#define ZPCI_TABLE_LEN_RSX 0x3 +#define ZPCI_TABLE_LEN_RTX 0x3 + +#define ZPCI_TABLE_OFFSET_MASK 0xc0 +#define ZPCI_TABLE_SIZE 0x4000 +#define ZPCI_TABLE_ALIGN ZPCI_TABLE_SIZE +#define ZPCI_TABLE_ENTRY_SIZE (sizeof(unsigned long)) +#define ZPCI_TABLE_ENTRIES (ZPCI_TABLE_SIZE / ZPCI_TABLE_ENTRY_SIZE) + +#define ZPCI_TABLE_BITS 11 +#define ZPCI_PT_BITS 8 +#define ZPCI_ST_SHIFT (ZPCI_PT_BITS + PAGE_SHIFT) +#define ZPCI_RT_SHIFT (ZPCI_ST_SHIFT + ZPCI_TABLE_BITS) + +#define ZPCI_RTE_FLAG_MASK 0x3fffUL +#define ZPCI_RTE_ADDR_MASK (~ZPCI_RTE_FLAG_MASK) +#define ZPCI_STE_FLAG_MASK 0x7ffUL +#define ZPCI_STE_ADDR_MASK (~ZPCI_STE_FLAG_MASK) + +/* I/O Page tables */ +#define ZPCI_PTE_VALID_MASK 0x400 +#define ZPCI_PTE_INVALID 0x400 +#define ZPCI_PTE_VALID 0x000 +#define ZPCI_PT_SIZE 0x800 +#define ZPCI_PT_ALIGN ZPCI_PT_SIZE +#define ZPCI_PT_ENTRIES (ZPCI_PT_SIZE / ZPCI_TABLE_ENTRY_SIZE) +#define ZPCI_PT_MASK (ZPCI_PT_ENTRIES - 1) + +#define ZPCI_PTE_FLAG_MASK 0xfffUL +#define ZPCI_PTE_ADDR_MASK (~ZPCI_PTE_FLAG_MASK) + +/* Shared bits */ +#define ZPCI_TABLE_VALID 0x00 +#define ZPCI_TABLE_INVALID 0x20 +#define ZPCI_TABLE_PROTECTED 0x200 +#define ZPCI_TABLE_UNPROTECTED 0x000 + +#define ZPCI_TABLE_VALID_MASK 0x20 +#define ZPCI_TABLE_PROT_MASK 0x200 + +static inline unsigned int calc_rtx(dma_addr_t ptr) +{ + return ((unsigned long) ptr >> ZPCI_RT_SHIFT) & ZPCI_INDEX_MASK; +} + +static inline unsigned int calc_sx(dma_addr_t ptr) +{ + return ((unsigned long) ptr >> ZPCI_ST_SHIFT) & ZPCI_INDEX_MASK; +} + +static inline unsigned int calc_px(dma_addr_t ptr) +{ + return ((unsigned long) ptr >> PAGE_SHIFT) & ZPCI_PT_MASK; +} + +static inline void set_pt_pfaa(unsigned long *entry, void *pfaa) +{ + *entry &= ZPCI_PTE_FLAG_MASK; + *entry |= ((unsigned long) pfaa & ZPCI_PTE_ADDR_MASK); +} + +static inline void set_rt_sto(unsigned long *entry, void *sto) +{ + *entry &= ZPCI_RTE_FLAG_MASK; + *entry |= ((unsigned long) sto & ZPCI_RTE_ADDR_MASK); + *entry |= ZPCI_TABLE_TYPE_RTX; +} + +static inline void set_st_pto(unsigned long *entry, void *pto) +{ + *entry &= ZPCI_STE_FLAG_MASK; + *entry |= ((unsigned long) pto & ZPCI_STE_ADDR_MASK); + *entry |= ZPCI_TABLE_TYPE_SX; +} + +static inline void validate_rt_entry(unsigned long *entry) +{ + *entry &= ~ZPCI_TABLE_VALID_MASK; + *entry &= ~ZPCI_TABLE_OFFSET_MASK; + *entry |= ZPCI_TABLE_VALID; + *entry |= ZPCI_TABLE_LEN_RTX; +} + +static inline void validate_st_entry(unsigned long *entry) +{ + *entry &= ~ZPCI_TABLE_VALID_MASK; + *entry |= ZPCI_TABLE_VALID; +} + +static inline void invalidate_table_entry(unsigned long *entry) +{ + *entry &= ~ZPCI_TABLE_VALID_MASK; + *entry |= ZPCI_TABLE_INVALID; +} + +static inline void invalidate_pt_entry(unsigned long *entry) +{ + WARN_ON_ONCE((*entry & ZPCI_PTE_VALID_MASK) == ZPCI_PTE_INVALID); + *entry &= ~ZPCI_PTE_VALID_MASK; + *entry |= ZPCI_PTE_INVALID; +} + +static inline void validate_pt_entry(unsigned long *entry) +{ + WARN_ON_ONCE((*entry & ZPCI_PTE_VALID_MASK) == ZPCI_PTE_VALID); + *entry &= ~ZPCI_PTE_VALID_MASK; + *entry |= ZPCI_PTE_VALID; +} + +static inline void entry_set_protected(unsigned long *entry) +{ + *entry &= ~ZPCI_TABLE_PROT_MASK; + *entry |= ZPCI_TABLE_PROTECTED; +} + +static inline void entry_clr_protected(unsigned long *entry) +{ + *entry &= ~ZPCI_TABLE_PROT_MASK; + *entry |= ZPCI_TABLE_UNPROTECTED; +} + +static inline int reg_entry_isvalid(unsigned long entry) +{ + return (entry & ZPCI_TABLE_VALID_MASK) == ZPCI_TABLE_VALID; +} + +static inline int pt_entry_isvalid(unsigned long entry) +{ + return (entry & ZPCI_PTE_VALID_MASK) == ZPCI_PTE_VALID; +} + +static inline int entry_isprotected(unsigned long entry) +{ + return (entry & ZPCI_TABLE_PROT_MASK) == ZPCI_TABLE_PROTECTED; +} + +static inline unsigned long *get_rt_sto(unsigned long entry) +{ + return ((entry & ZPCI_TABLE_TYPE_MASK) == ZPCI_TABLE_TYPE_RTX) + ? (unsigned long *) (entry & ZPCI_RTE_ADDR_MASK) + : NULL; +} + +static inline unsigned long *get_st_pto(unsigned long entry) +{ + return ((entry & ZPCI_TABLE_TYPE_MASK) == ZPCI_TABLE_TYPE_SX) + ? (unsigned long *) (entry & ZPCI_STE_ADDR_MASK) + : NULL; +} + +/* Prototypes */ +int zpci_dma_init_device(struct zpci_dev *); +void zpci_dma_exit_device(struct zpci_dev *); + +#endif diff --git a/arch/s390/include/asm/pci_insn.h b/arch/s390/include/asm/pci_insn.h new file mode 100644 index 000000000..649eb62c5 --- /dev/null +++ b/arch/s390/include/asm/pci_insn.h @@ -0,0 +1,86 @@ +#ifndef _ASM_S390_PCI_INSN_H +#define _ASM_S390_PCI_INSN_H + +/* Load/Store status codes */ +#define ZPCI_PCI_ST_FUNC_NOT_ENABLED 4 +#define ZPCI_PCI_ST_FUNC_IN_ERR 8 +#define ZPCI_PCI_ST_BLOCKED 12 +#define ZPCI_PCI_ST_INSUF_RES 16 +#define ZPCI_PCI_ST_INVAL_AS 20 +#define ZPCI_PCI_ST_FUNC_ALREADY_ENABLED 24 +#define ZPCI_PCI_ST_DMA_AS_NOT_ENABLED 28 +#define ZPCI_PCI_ST_2ND_OP_IN_INV_AS 36 +#define ZPCI_PCI_ST_FUNC_NOT_AVAIL 40 +#define ZPCI_PCI_ST_ALREADY_IN_RQ_STATE 44 + +/* Load/Store return codes */ +#define ZPCI_PCI_LS_OK 0 +#define ZPCI_PCI_LS_ERR 1 +#define ZPCI_PCI_LS_BUSY 2 +#define ZPCI_PCI_LS_INVAL_HANDLE 3 + +/* Load/Store address space identifiers */ +#define ZPCI_PCIAS_MEMIO_0 0 +#define ZPCI_PCIAS_MEMIO_1 1 +#define ZPCI_PCIAS_MEMIO_2 2 +#define ZPCI_PCIAS_MEMIO_3 3 +#define ZPCI_PCIAS_MEMIO_4 4 +#define ZPCI_PCIAS_MEMIO_5 5 +#define ZPCI_PCIAS_CFGSPC 15 + +/* Modify PCI Function Controls */ +#define ZPCI_MOD_FC_REG_INT 2 +#define ZPCI_MOD_FC_DEREG_INT 3 +#define ZPCI_MOD_FC_REG_IOAT 4 +#define ZPCI_MOD_FC_DEREG_IOAT 5 +#define ZPCI_MOD_FC_REREG_IOAT 6 +#define ZPCI_MOD_FC_RESET_ERROR 7 +#define ZPCI_MOD_FC_RESET_BLOCK 9 +#define ZPCI_MOD_FC_SET_MEASURE 10 + +/* FIB function controls */ +#define ZPCI_FIB_FC_ENABLED 0x80 +#define ZPCI_FIB_FC_ERROR 0x40 +#define ZPCI_FIB_FC_LS_BLOCKED 0x20 +#define ZPCI_FIB_FC_DMAAS_REG 0x10 + +/* FIB function controls */ +#define ZPCI_FIB_FC_ENABLED 0x80 +#define ZPCI_FIB_FC_ERROR 0x40 +#define ZPCI_FIB_FC_LS_BLOCKED 0x20 +#define ZPCI_FIB_FC_DMAAS_REG 0x10 + +/* Function Information Block */ +struct zpci_fib { + u32 fmt : 8; /* format */ + u32 : 24; + u32 : 32; + u8 fc; /* function controls */ + u64 : 56; + u64 pba; /* PCI base address */ + u64 pal; /* PCI address limit */ + u64 iota; /* I/O Translation Anchor */ + u32 : 1; + u32 isc : 3; /* Interrupt subclass */ + u32 noi : 12; /* Number of interrupts */ + u32 : 2; + u32 aibvo : 6; /* Adapter interrupt bit vector offset */ + u32 sum : 1; /* Adapter int summary bit enabled */ + u32 : 1; + u32 aisbo : 6; /* Adapter int summary bit offset */ + u32 : 32; + u64 aibv; /* Adapter int bit vector address */ + u64 aisb; /* Adapter int summary bit address */ + u64 fmb_addr; /* Function measurement block address and key */ + u32 : 32; + u32 gd; +} __packed __aligned(8); + +int zpci_mod_fc(u64 req, struct zpci_fib *fib); +int zpci_refresh_trans(u64 fn, u64 addr, u64 range); +int zpci_load(u64 *data, u64 req, u64 offset); +int zpci_store(u64 data, u64 req, u64 offset); +int zpci_store_block(const u64 *data, u64 req, u64 offset); +void zpci_set_irq_ctrl(u16 ctl, char *unused, u8 isc); + +#endif diff --git a/arch/s390/include/asm/pci_io.h b/arch/s390/include/asm/pci_io.h new file mode 100644 index 000000000..1a9a98de5 --- /dev/null +++ b/arch/s390/include/asm/pci_io.h @@ -0,0 +1,201 @@ +#ifndef _ASM_S390_PCI_IO_H +#define _ASM_S390_PCI_IO_H + +#ifdef CONFIG_PCI + +#include <linux/kernel.h> +#include <linux/slab.h> +#include <asm/pci_insn.h> + +/* I/O Map */ +#define ZPCI_IOMAP_MAX_ENTRIES 0x7fff +#define ZPCI_IOMAP_ADDR_BASE 0x8000000000000000ULL +#define ZPCI_IOMAP_ADDR_IDX_MASK 0x7fff000000000000ULL +#define ZPCI_IOMAP_ADDR_OFF_MASK 0x0000ffffffffffffULL + +struct zpci_iomap_entry { + u32 fh; + u8 bar; + u16 count; +}; + +extern struct zpci_iomap_entry *zpci_iomap_start; + +#define ZPCI_IDX(addr) \ + (((__force u64) addr & ZPCI_IOMAP_ADDR_IDX_MASK) >> 48) +#define ZPCI_OFFSET(addr) \ + ((__force u64) addr & ZPCI_IOMAP_ADDR_OFF_MASK) + +#define ZPCI_CREATE_REQ(handle, space, len) \ + ((u64) handle << 32 | space << 16 | len) + +#define zpci_read(LENGTH, RETTYPE) \ +static inline RETTYPE zpci_read_##RETTYPE(const volatile void __iomem *addr) \ +{ \ + struct zpci_iomap_entry *entry = &zpci_iomap_start[ZPCI_IDX(addr)]; \ + u64 req = ZPCI_CREATE_REQ(entry->fh, entry->bar, LENGTH); \ + u64 data; \ + int rc; \ + \ + rc = zpci_load(&data, req, ZPCI_OFFSET(addr)); \ + if (rc) \ + data = -1ULL; \ + return (RETTYPE) data; \ +} + +#define zpci_write(LENGTH, VALTYPE) \ +static inline void zpci_write_##VALTYPE(VALTYPE val, \ + const volatile void __iomem *addr) \ +{ \ + struct zpci_iomap_entry *entry = &zpci_iomap_start[ZPCI_IDX(addr)]; \ + u64 req = ZPCI_CREATE_REQ(entry->fh, entry->bar, LENGTH); \ + u64 data = (VALTYPE) val; \ + \ + zpci_store(data, req, ZPCI_OFFSET(addr)); \ +} + +zpci_read(8, u64) +zpci_read(4, u32) +zpci_read(2, u16) +zpci_read(1, u8) +zpci_write(8, u64) +zpci_write(4, u32) +zpci_write(2, u16) +zpci_write(1, u8) + +static inline int zpci_write_single(u64 req, const u64 *data, u64 offset, u8 len) +{ + u64 val; + + switch (len) { + case 1: + val = (u64) *((u8 *) data); + break; + case 2: + val = (u64) *((u16 *) data); + break; + case 4: + val = (u64) *((u32 *) data); + break; + case 8: + val = (u64) *((u64 *) data); + break; + default: + val = 0; /* let FW report error */ + break; + } + return zpci_store(val, req, offset); +} + +static inline int zpci_read_single(u64 req, u64 *dst, u64 offset, u8 len) +{ + u64 data; + int cc; + + cc = zpci_load(&data, req, offset); + if (cc) + goto out; + + switch (len) { + case 1: + *((u8 *) dst) = (u8) data; + break; + case 2: + *((u16 *) dst) = (u16) data; + break; + case 4: + *((u32 *) dst) = (u32) data; + break; + case 8: + *((u64 *) dst) = (u64) data; + break; + } +out: + return cc; +} + +static inline int zpci_write_block(u64 req, const u64 *data, u64 offset) +{ + return zpci_store_block(data, req, offset); +} + +static inline u8 zpci_get_max_write_size(u64 src, u64 dst, int len, int max) +{ + int count = len > max ? max : len, size = 1; + + while (!(src & 0x1) && !(dst & 0x1) && ((size << 1) <= count)) { + dst = dst >> 1; + src = src >> 1; + size = size << 1; + } + return size; +} + +static inline int zpci_memcpy_fromio(void *dst, + const volatile void __iomem *src, + unsigned long n) +{ + struct zpci_iomap_entry *entry = &zpci_iomap_start[ZPCI_IDX(src)]; + u64 req, offset = ZPCI_OFFSET(src); + int size, rc = 0; + + while (n > 0) { + size = zpci_get_max_write_size((u64 __force) src, + (u64) dst, n, 8); + req = ZPCI_CREATE_REQ(entry->fh, entry->bar, size); + rc = zpci_read_single(req, dst, offset, size); + if (rc) + break; + offset += size; + dst += size; + n -= size; + } + return rc; +} + +static inline int zpci_memcpy_toio(volatile void __iomem *dst, + const void *src, unsigned long n) +{ + struct zpci_iomap_entry *entry = &zpci_iomap_start[ZPCI_IDX(dst)]; + u64 req, offset = ZPCI_OFFSET(dst); + int size, rc = 0; + + if (!src) + return -EINVAL; + + while (n > 0) { + size = zpci_get_max_write_size((u64 __force) dst, + (u64) src, n, 128); + req = ZPCI_CREATE_REQ(entry->fh, entry->bar, size); + + if (size > 8) /* main path */ + rc = zpci_write_block(req, src, offset); + else + rc = zpci_write_single(req, src, offset, size); + if (rc) + break; + offset += size; + src += size; + n -= size; + } + return rc; +} + +static inline int zpci_memset_io(volatile void __iomem *dst, + unsigned char val, size_t count) +{ + u8 *src = kmalloc(count, GFP_KERNEL); + int rc; + + if (src == NULL) + return -ENOMEM; + memset(src, val, count); + + rc = zpci_memcpy_toio(dst, src, count); + kfree(src); + return rc; +} + +#endif /* CONFIG_PCI */ + +#endif /* _ASM_S390_PCI_IO_H */ diff --git a/arch/s390/include/asm/percpu.h b/arch/s390/include/asm/percpu.h new file mode 100644 index 000000000..6d6556ca2 --- /dev/null +++ b/arch/s390/include/asm/percpu.h @@ -0,0 +1,186 @@ +#ifndef __ARCH_S390_PERCPU__ +#define __ARCH_S390_PERCPU__ + +#include <linux/preempt.h> +#include <asm/cmpxchg.h> + +/* + * s390 uses its own implementation for per cpu data, the offset of + * the cpu local data area is cached in the cpu's lowcore memory. + */ +#define __my_cpu_offset S390_lowcore.percpu_offset + +/* + * For 64 bit module code, the module may be more than 4G above the + * per cpu area, use weak definitions to force the compiler to + * generate external references. + */ +#if defined(CONFIG_SMP) && defined(MODULE) +#define ARCH_NEEDS_WEAK_PER_CPU +#endif + +/* + * We use a compare-and-swap loop since that uses less cpu cycles than + * disabling and enabling interrupts like the generic variant would do. + */ +#define arch_this_cpu_to_op_simple(pcp, val, op) \ +({ \ + typedef typeof(pcp) pcp_op_T__; \ + pcp_op_T__ old__, new__, prev__; \ + pcp_op_T__ *ptr__; \ + preempt_disable(); \ + ptr__ = raw_cpu_ptr(&(pcp)); \ + prev__ = *ptr__; \ + do { \ + old__ = prev__; \ + new__ = old__ op (val); \ + prev__ = cmpxchg(ptr__, old__, new__); \ + } while (prev__ != old__); \ + preempt_enable(); \ + new__; \ +}) + +#define this_cpu_add_1(pcp, val) arch_this_cpu_to_op_simple(pcp, val, +) +#define this_cpu_add_2(pcp, val) arch_this_cpu_to_op_simple(pcp, val, +) +#define this_cpu_add_return_1(pcp, val) arch_this_cpu_to_op_simple(pcp, val, +) +#define this_cpu_add_return_2(pcp, val) arch_this_cpu_to_op_simple(pcp, val, +) +#define this_cpu_and_1(pcp, val) arch_this_cpu_to_op_simple(pcp, val, &) +#define this_cpu_and_2(pcp, val) arch_this_cpu_to_op_simple(pcp, val, &) +#define this_cpu_or_1(pcp, val) arch_this_cpu_to_op_simple(pcp, val, |) +#define this_cpu_or_2(pcp, val) arch_this_cpu_to_op_simple(pcp, val, |) + +#ifndef CONFIG_HAVE_MARCH_Z196_FEATURES + +#define this_cpu_add_4(pcp, val) arch_this_cpu_to_op_simple(pcp, val, +) +#define this_cpu_add_8(pcp, val) arch_this_cpu_to_op_simple(pcp, val, +) +#define this_cpu_add_return_4(pcp, val) arch_this_cpu_to_op_simple(pcp, val, +) +#define this_cpu_add_return_8(pcp, val) arch_this_cpu_to_op_simple(pcp, val, +) +#define this_cpu_and_4(pcp, val) arch_this_cpu_to_op_simple(pcp, val, &) +#define this_cpu_and_8(pcp, val) arch_this_cpu_to_op_simple(pcp, val, &) +#define this_cpu_or_4(pcp, val) arch_this_cpu_to_op_simple(pcp, val, |) +#define this_cpu_or_8(pcp, val) arch_this_cpu_to_op_simple(pcp, val, |) + +#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +#define arch_this_cpu_add(pcp, val, op1, op2, szcast) \ +{ \ + typedef typeof(pcp) pcp_op_T__; \ + pcp_op_T__ val__ = (val); \ + pcp_op_T__ old__, *ptr__; \ + preempt_disable(); \ + ptr__ = raw_cpu_ptr(&(pcp)); \ + if (__builtin_constant_p(val__) && \ + ((szcast)val__ > -129) && ((szcast)val__ < 128)) { \ + asm volatile( \ + op2 " %[ptr__],%[val__]\n" \ + : [ptr__] "+Q" (*ptr__) \ + : [val__] "i" ((szcast)val__) \ + : "cc"); \ + } else { \ + asm volatile( \ + op1 " %[old__],%[val__],%[ptr__]\n" \ + : [old__] "=d" (old__), [ptr__] "+Q" (*ptr__) \ + : [val__] "d" (val__) \ + : "cc"); \ + } \ + preempt_enable(); \ +} + +#define this_cpu_add_4(pcp, val) arch_this_cpu_add(pcp, val, "laa", "asi", int) +#define this_cpu_add_8(pcp, val) arch_this_cpu_add(pcp, val, "laag", "agsi", long) + +#define arch_this_cpu_add_return(pcp, val, op) \ +({ \ + typedef typeof(pcp) pcp_op_T__; \ + pcp_op_T__ val__ = (val); \ + pcp_op_T__ old__, *ptr__; \ + preempt_disable(); \ + ptr__ = raw_cpu_ptr(&(pcp)); \ + asm volatile( \ + op " %[old__],%[val__],%[ptr__]\n" \ + : [old__] "=d" (old__), [ptr__] "+Q" (*ptr__) \ + : [val__] "d" (val__) \ + : "cc"); \ + preempt_enable(); \ + old__ + val__; \ +}) + +#define this_cpu_add_return_4(pcp, val) arch_this_cpu_add_return(pcp, val, "laa") +#define this_cpu_add_return_8(pcp, val) arch_this_cpu_add_return(pcp, val, "laag") + +#define arch_this_cpu_to_op(pcp, val, op) \ +{ \ + typedef typeof(pcp) pcp_op_T__; \ + pcp_op_T__ val__ = (val); \ + pcp_op_T__ old__, *ptr__; \ + preempt_disable(); \ + ptr__ = raw_cpu_ptr(&(pcp)); \ + asm volatile( \ + op " %[old__],%[val__],%[ptr__]\n" \ + : [old__] "=d" (old__), [ptr__] "+Q" (*ptr__) \ + : [val__] "d" (val__) \ + : "cc"); \ + preempt_enable(); \ +} + +#define this_cpu_and_4(pcp, val) arch_this_cpu_to_op(pcp, val, "lan") +#define this_cpu_and_8(pcp, val) arch_this_cpu_to_op(pcp, val, "lang") +#define this_cpu_or_4(pcp, val) arch_this_cpu_to_op(pcp, val, "lao") +#define this_cpu_or_8(pcp, val) arch_this_cpu_to_op(pcp, val, "laog") + +#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +#define arch_this_cpu_cmpxchg(pcp, oval, nval) \ +({ \ + typedef typeof(pcp) pcp_op_T__; \ + pcp_op_T__ ret__; \ + pcp_op_T__ *ptr__; \ + preempt_disable(); \ + ptr__ = raw_cpu_ptr(&(pcp)); \ + ret__ = cmpxchg(ptr__, oval, nval); \ + preempt_enable(); \ + ret__; \ +}) + +#define this_cpu_cmpxchg_1(pcp, oval, nval) arch_this_cpu_cmpxchg(pcp, oval, nval) +#define this_cpu_cmpxchg_2(pcp, oval, nval) arch_this_cpu_cmpxchg(pcp, oval, nval) +#define this_cpu_cmpxchg_4(pcp, oval, nval) arch_this_cpu_cmpxchg(pcp, oval, nval) +#define this_cpu_cmpxchg_8(pcp, oval, nval) arch_this_cpu_cmpxchg(pcp, oval, nval) + +#define arch_this_cpu_xchg(pcp, nval) \ +({ \ + typeof(pcp) *ptr__; \ + typeof(pcp) ret__; \ + preempt_disable(); \ + ptr__ = raw_cpu_ptr(&(pcp)); \ + ret__ = xchg(ptr__, nval); \ + preempt_enable(); \ + ret__; \ +}) + +#define this_cpu_xchg_1(pcp, nval) arch_this_cpu_xchg(pcp, nval) +#define this_cpu_xchg_2(pcp, nval) arch_this_cpu_xchg(pcp, nval) +#define this_cpu_xchg_4(pcp, nval) arch_this_cpu_xchg(pcp, nval) +#define this_cpu_xchg_8(pcp, nval) arch_this_cpu_xchg(pcp, nval) + +#define arch_this_cpu_cmpxchg_double(pcp1, pcp2, o1, o2, n1, n2) \ +({ \ + typeof(pcp1) o1__ = (o1), n1__ = (n1); \ + typeof(pcp2) o2__ = (o2), n2__ = (n2); \ + typeof(pcp1) *p1__; \ + typeof(pcp2) *p2__; \ + int ret__; \ + preempt_disable(); \ + p1__ = raw_cpu_ptr(&(pcp1)); \ + p2__ = raw_cpu_ptr(&(pcp2)); \ + ret__ = __cmpxchg_double(p1__, p2__, o1__, o2__, n1__, n2__); \ + preempt_enable(); \ + ret__; \ +}) + +#define this_cpu_cmpxchg_double_4 arch_this_cpu_cmpxchg_double +#define this_cpu_cmpxchg_double_8 arch_this_cpu_cmpxchg_double + +#include <asm-generic/percpu.h> + +#endif /* __ARCH_S390_PERCPU__ */ diff --git a/arch/s390/include/asm/perf_event.h b/arch/s390/include/asm/perf_event.h new file mode 100644 index 000000000..4cb19fe76 --- /dev/null +++ b/arch/s390/include/asm/perf_event.h @@ -0,0 +1,93 @@ +/* + * Performance event support - s390 specific definitions. + * + * Copyright IBM Corp. 2009, 2013 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + * Hendrik Brueckner <brueckner@linux.vnet.ibm.com> + */ + +#ifndef _ASM_S390_PERF_EVENT_H +#define _ASM_S390_PERF_EVENT_H + +#include <linux/perf_event.h> +#include <linux/device.h> +#include <asm/cpu_mf.h> + +/* Per-CPU flags for PMU states */ +#define PMU_F_RESERVED 0x1000 +#define PMU_F_ENABLED 0x2000 +#define PMU_F_IN_USE 0x4000 +#define PMU_F_ERR_IBE 0x0100 +#define PMU_F_ERR_LSDA 0x0200 +#define PMU_F_ERR_MASK (PMU_F_ERR_IBE|PMU_F_ERR_LSDA) + +/* Perf defintions for PMU event attributes in sysfs */ +extern __init const struct attribute_group **cpumf_cf_event_group(void); +extern ssize_t cpumf_events_sysfs_show(struct device *dev, + struct device_attribute *attr, + char *page); +#define EVENT_VAR(_cat, _name) event_attr_##_cat##_##_name +#define EVENT_PTR(_cat, _name) (&EVENT_VAR(_cat, _name).attr.attr) + +#define CPUMF_EVENT_ATTR(cat, name, id) \ + PMU_EVENT_ATTR(name, EVENT_VAR(cat, name), id, cpumf_events_sysfs_show) +#define CPUMF_EVENT_PTR(cat, name) EVENT_PTR(cat, name) + + +/* Perf callbacks */ +struct pt_regs; +extern unsigned long perf_instruction_pointer(struct pt_regs *regs); +extern unsigned long perf_misc_flags(struct pt_regs *regs); +#define perf_misc_flags(regs) perf_misc_flags(regs) + +/* Perf pt_regs extension for sample-data-entry indicators */ +struct perf_sf_sde_regs { + unsigned char in_guest:1; /* guest sample */ + unsigned long reserved:63; /* reserved */ +}; + +/* Perf PMU definitions for the counter facility */ +#define PERF_CPUM_CF_MAX_CTR 256 + +/* Perf PMU definitions for the sampling facility */ +#define PERF_CPUM_SF_MAX_CTR 2 +#define PERF_EVENT_CPUM_SF 0xB0000UL /* Event: Basic-sampling */ +#define PERF_EVENT_CPUM_SF_DIAG 0xBD000UL /* Event: Combined-sampling */ +#define PERF_CPUM_SF_BASIC_MODE 0x0001 /* Basic-sampling flag */ +#define PERF_CPUM_SF_DIAG_MODE 0x0002 /* Diagnostic-sampling flag */ +#define PERF_CPUM_SF_MODE_MASK (PERF_CPUM_SF_BASIC_MODE| \ + PERF_CPUM_SF_DIAG_MODE) +#define PERF_CPUM_SF_FULL_BLOCKS 0x0004 /* Process full SDBs only */ + +#define REG_NONE 0 +#define REG_OVERFLOW 1 +#define OVERFLOW_REG(hwc) ((hwc)->extra_reg.config) +#define SFB_ALLOC_REG(hwc) ((hwc)->extra_reg.alloc) +#define RAWSAMPLE_REG(hwc) ((hwc)->config) +#define TEAR_REG(hwc) ((hwc)->last_tag) +#define SAMPL_RATE(hwc) ((hwc)->event_base) +#define SAMPL_FLAGS(hwc) ((hwc)->config_base) +#define SAMPL_DIAG_MODE(hwc) (SAMPL_FLAGS(hwc) & PERF_CPUM_SF_DIAG_MODE) +#define SDB_FULL_BLOCKS(hwc) (SAMPL_FLAGS(hwc) & PERF_CPUM_SF_FULL_BLOCKS) + +/* Structure for sampling data entries to be passed as perf raw sample data + * to user space. Note that raw sample data must be aligned and, thus, might + * be padded with zeros. + */ +struct sf_raw_sample { +#define SF_RAW_SAMPLE_BASIC PERF_CPUM_SF_BASIC_MODE +#define SF_RAW_SAMPLE_DIAG PERF_CPUM_SF_DIAG_MODE + u64 format; + u32 size; /* Size of sf_raw_sample */ + u16 bsdes; /* Basic-sampling data entry size */ + u16 dsdes; /* Diagnostic-sampling data entry size */ + struct hws_basic_entry basic; /* Basic-sampling data entry */ + struct hws_diag_entry diag; /* Diagnostic-sampling data entry */ + u8 padding[]; /* Padding to next multiple of 8 */ +} __packed; + +/* Perf hardware reserve and release functions */ +int perf_reserve_sampling(void); +void perf_release_sampling(void); + +#endif /* _ASM_S390_PERF_EVENT_H */ diff --git a/arch/s390/include/asm/pgalloc.h b/arch/s390/include/asm/pgalloc.h new file mode 100644 index 000000000..7b7858f15 --- /dev/null +++ b/arch/s390/include/asm/pgalloc.h @@ -0,0 +1,132 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + * Author(s): Hartmut Penner (hp@de.ibm.com) + * Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "include/asm-i386/pgalloc.h" + * Copyright (C) 1994 Linus Torvalds + */ + +#ifndef _S390_PGALLOC_H +#define _S390_PGALLOC_H + +#include <linux/threads.h> +#include <linux/gfp.h> +#include <linux/mm.h> + +unsigned long *crst_table_alloc(struct mm_struct *); +void crst_table_free(struct mm_struct *, unsigned long *); + +unsigned long *page_table_alloc(struct mm_struct *); +void page_table_free(struct mm_struct *, unsigned long *); +void page_table_free_rcu(struct mmu_gather *, unsigned long *, unsigned long); +extern int page_table_allocate_pgste; + +int set_guest_storage_key(struct mm_struct *mm, unsigned long addr, + unsigned long key, bool nq); +unsigned long get_guest_storage_key(struct mm_struct *mm, unsigned long addr); + +static inline void clear_table(unsigned long *s, unsigned long val, size_t n) +{ + typedef struct { char _[n]; } addrtype; + + *s = val; + n = (n / 256) - 1; + asm volatile( + " mvc 8(248,%0),0(%0)\n" + "0: mvc 256(256,%0),0(%0)\n" + " la %0,256(%0)\n" + " brct %1,0b\n" + : "+a" (s), "+d" (n), "=m" (*(addrtype *) s) + : "m" (*(addrtype *) s)); +} + +static inline void crst_table_init(unsigned long *crst, unsigned long entry) +{ + clear_table(crst, entry, sizeof(unsigned long)*2048); +} + +static inline unsigned long pgd_entry_type(struct mm_struct *mm) +{ + if (mm->context.asce_limit <= (1UL << 31)) + return _SEGMENT_ENTRY_EMPTY; + if (mm->context.asce_limit <= (1UL << 42)) + return _REGION3_ENTRY_EMPTY; + return _REGION2_ENTRY_EMPTY; +} + +int crst_table_upgrade(struct mm_struct *, unsigned long limit); +void crst_table_downgrade(struct mm_struct *, unsigned long limit); + +static inline pud_t *pud_alloc_one(struct mm_struct *mm, unsigned long address) +{ + unsigned long *table = crst_table_alloc(mm); + if (table) + crst_table_init(table, _REGION3_ENTRY_EMPTY); + return (pud_t *) table; +} +#define pud_free(mm, pud) crst_table_free(mm, (unsigned long *) pud) + +static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long vmaddr) +{ + unsigned long *table = crst_table_alloc(mm); + + if (!table) + return NULL; + crst_table_init(table, _SEGMENT_ENTRY_EMPTY); + if (!pgtable_pmd_page_ctor(virt_to_page(table))) { + crst_table_free(mm, table); + return NULL; + } + return (pmd_t *) table; +} + +static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd) +{ + pgtable_pmd_page_dtor(virt_to_page(pmd)); + crst_table_free(mm, (unsigned long *) pmd); +} + +static inline void pgd_populate(struct mm_struct *mm, pgd_t *pgd, pud_t *pud) +{ + pgd_val(*pgd) = _REGION2_ENTRY | __pa(pud); +} + +static inline void pud_populate(struct mm_struct *mm, pud_t *pud, pmd_t *pmd) +{ + pud_val(*pud) = _REGION3_ENTRY | __pa(pmd); +} + +static inline pgd_t *pgd_alloc(struct mm_struct *mm) +{ + spin_lock_init(&mm->context.list_lock); + INIT_LIST_HEAD(&mm->context.pgtable_list); + INIT_LIST_HEAD(&mm->context.gmap_list); + return (pgd_t *) crst_table_alloc(mm); +} +#define pgd_free(mm, pgd) crst_table_free(mm, (unsigned long *) pgd) + +static inline void pmd_populate(struct mm_struct *mm, + pmd_t *pmd, pgtable_t pte) +{ + pmd_val(*pmd) = _SEGMENT_ENTRY + __pa(pte); +} + +#define pmd_populate_kernel(mm, pmd, pte) pmd_populate(mm, pmd, pte) + +#define pmd_pgtable(pmd) \ + (pgtable_t)(pmd_val(pmd) & -sizeof(pte_t)*PTRS_PER_PTE) + +/* + * page table entry allocation/free routines. + */ +#define pte_alloc_one_kernel(mm, vmaddr) ((pte_t *) page_table_alloc(mm)) +#define pte_alloc_one(mm, vmaddr) ((pte_t *) page_table_alloc(mm)) + +#define pte_free_kernel(mm, pte) page_table_free(mm, (unsigned long *) pte) +#define pte_free(mm, pte) page_table_free(mm, (unsigned long *) pte) + +extern void rcu_table_freelist_finish(void); + +#endif /* _S390_PGALLOC_H */ diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h new file mode 100644 index 000000000..ef24a212e --- /dev/null +++ b/arch/s390/include/asm/pgtable.h @@ -0,0 +1,1637 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + * Author(s): Hartmut Penner (hp@de.ibm.com) + * Ulrich Weigand (weigand@de.ibm.com) + * Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "include/asm-i386/pgtable.h" + */ + +#ifndef _ASM_S390_PGTABLE_H +#define _ASM_S390_PGTABLE_H + +/* + * The Linux memory management assumes a three-level page table setup. + * For s390 64 bit we use up to four of the five levels the hardware + * provides (region first tables are not used). + * + * The "pgd_xxx()" functions are trivial for a folded two-level + * setup: the pgd is never bad, and a pmd always exists (as it's folded + * into the pgd entry) + * + * This file contains the functions and defines necessary to modify and use + * the S390 page table tree. + */ +#ifndef __ASSEMBLY__ +#include <linux/sched.h> +#include <linux/mm_types.h> +#include <linux/page-flags.h> +#include <linux/radix-tree.h> +#include <asm/bug.h> +#include <asm/page.h> + +extern pgd_t swapper_pg_dir[] __attribute__ ((aligned (4096))); +extern void paging_init(void); +extern void vmem_map_init(void); + +/* + * The S390 doesn't have any external MMU info: the kernel page + * tables contain all the necessary information. + */ +#define update_mmu_cache(vma, address, ptep) do { } while (0) +#define update_mmu_cache_pmd(vma, address, ptep) do { } while (0) + +/* + * ZERO_PAGE is a global shared page that is always zero; used + * for zero-mapped memory areas etc.. + */ + +extern unsigned long empty_zero_page; +extern unsigned long zero_page_mask; + +#define ZERO_PAGE(vaddr) \ + (virt_to_page((void *)(empty_zero_page + \ + (((unsigned long)(vaddr)) &zero_page_mask)))) +#define __HAVE_COLOR_ZERO_PAGE + +/* TODO: s390 cannot support io_remap_pfn_range... */ +#endif /* !__ASSEMBLY__ */ + +/* + * PMD_SHIFT determines the size of the area a second-level page + * table can map + * PGDIR_SHIFT determines what a third-level page table entry can map + */ +#define PMD_SHIFT 20 +#define PUD_SHIFT 31 +#define PGDIR_SHIFT 42 + +#define PMD_SIZE (1UL << PMD_SHIFT) +#define PMD_MASK (~(PMD_SIZE-1)) +#define PUD_SIZE (1UL << PUD_SHIFT) +#define PUD_MASK (~(PUD_SIZE-1)) +#define PGDIR_SIZE (1UL << PGDIR_SHIFT) +#define PGDIR_MASK (~(PGDIR_SIZE-1)) + +/* + * entries per page directory level: the S390 is two-level, so + * we don't really have any PMD directory physically. + * for S390 segment-table entries are combined to one PGD + * that leads to 1024 pte per pgd + */ +#define PTRS_PER_PTE 256 +#define PTRS_PER_PMD 2048 +#define PTRS_PER_PUD 2048 +#define PTRS_PER_PGD 2048 + +#define FIRST_USER_ADDRESS 0UL + +#define pte_ERROR(e) \ + printk("%s:%d: bad pte %p.\n", __FILE__, __LINE__, (void *) pte_val(e)) +#define pmd_ERROR(e) \ + printk("%s:%d: bad pmd %p.\n", __FILE__, __LINE__, (void *) pmd_val(e)) +#define pud_ERROR(e) \ + printk("%s:%d: bad pud %p.\n", __FILE__, __LINE__, (void *) pud_val(e)) +#define pgd_ERROR(e) \ + printk("%s:%d: bad pgd %p.\n", __FILE__, __LINE__, (void *) pgd_val(e)) + +#ifndef __ASSEMBLY__ +/* + * The vmalloc and module area will always be on the topmost area of the + * kernel mapping. We reserve 128GB (64bit) for vmalloc and modules. + * On 64 bit kernels we have a 2GB area at the top of the vmalloc area where + * modules will reside. That makes sure that inter module branches always + * happen without trampolines and in addition the placement within a 2GB frame + * is branch prediction unit friendly. + */ +extern unsigned long VMALLOC_START; +extern unsigned long VMALLOC_END; +extern struct page *vmemmap; + +#define VMEM_MAX_PHYS ((unsigned long) vmemmap) + +extern unsigned long MODULES_VADDR; +extern unsigned long MODULES_END; +#define MODULES_VADDR MODULES_VADDR +#define MODULES_END MODULES_END +#define MODULES_LEN (1UL << 31) + +static inline int is_module_addr(void *addr) +{ + BUILD_BUG_ON(MODULES_LEN > (1UL << 31)); + if (addr < (void *)MODULES_VADDR) + return 0; + if (addr > (void *)MODULES_END) + return 0; + return 1; +} + +/* + * A 64 bit pagetable entry of S390 has following format: + * | PFRA |0IPC| OS | + * 0000000000111111111122222222223333333333444444444455555555556666 + * 0123456789012345678901234567890123456789012345678901234567890123 + * + * I Page-Invalid Bit: Page is not available for address-translation + * P Page-Protection Bit: Store access not possible for page + * C Change-bit override: HW is not required to set change bit + * + * A 64 bit segmenttable entry of S390 has following format: + * | P-table origin | TT + * 0000000000111111111122222222223333333333444444444455555555556666 + * 0123456789012345678901234567890123456789012345678901234567890123 + * + * I Segment-Invalid Bit: Segment is not available for address-translation + * C Common-Segment Bit: Segment is not private (PoP 3-30) + * P Page-Protection Bit: Store access not possible for page + * TT Type 00 + * + * A 64 bit region table entry of S390 has following format: + * | S-table origin | TF TTTL + * 0000000000111111111122222222223333333333444444444455555555556666 + * 0123456789012345678901234567890123456789012345678901234567890123 + * + * I Segment-Invalid Bit: Segment is not available for address-translation + * TT Type 01 + * TF + * TL Table length + * + * The 64 bit regiontable origin of S390 has following format: + * | region table origon | DTTL + * 0000000000111111111122222222223333333333444444444455555555556666 + * 0123456789012345678901234567890123456789012345678901234567890123 + * + * X Space-Switch event: + * G Segment-Invalid Bit: + * P Private-Space Bit: + * S Storage-Alteration: + * R Real space + * TL Table-Length: + * + * A storage key has the following format: + * | ACC |F|R|C|0| + * 0 3 4 5 6 7 + * ACC: access key + * F : fetch protection bit + * R : referenced bit + * C : changed bit + */ + +/* Hardware bits in the page table entry */ +#define _PAGE_PROTECT 0x200 /* HW read-only bit */ +#define _PAGE_INVALID 0x400 /* HW invalid bit */ +#define _PAGE_LARGE 0x800 /* Bit to mark a large pte */ + +/* Software bits in the page table entry */ +#define _PAGE_PRESENT 0x001 /* SW pte present bit */ +#define _PAGE_YOUNG 0x004 /* SW pte young bit */ +#define _PAGE_DIRTY 0x008 /* SW pte dirty bit */ +#define _PAGE_READ 0x010 /* SW pte read bit */ +#define _PAGE_WRITE 0x020 /* SW pte write bit */ +#define _PAGE_SPECIAL 0x040 /* SW associated with special page */ +#define _PAGE_UNUSED 0x080 /* SW bit for pgste usage state */ +#define __HAVE_ARCH_PTE_SPECIAL + +/* Set of bits not changed in pte_modify */ +#define _PAGE_CHG_MASK (PAGE_MASK | _PAGE_SPECIAL | _PAGE_DIRTY | \ + _PAGE_YOUNG) + +/* + * handle_pte_fault uses pte_present and pte_none to find out the pte type + * WITHOUT holding the page table lock. The _PAGE_PRESENT bit is used to + * distinguish present from not-present ptes. It is changed only with the page + * table lock held. + * + * The following table gives the different possible bit combinations for + * the pte hardware and software bits in the last 12 bits of a pte + * (. unassigned bit, x don't care, t swap type): + * + * 842100000000 + * 000084210000 + * 000000008421 + * .IR.uswrdy.p + * empty .10.00000000 + * swap .11..ttttt.0 + * prot-none, clean, old .11.xx0000.1 + * prot-none, clean, young .11.xx0001.1 + * prot-none, dirty, old .10.xx0010.1 + * prot-none, dirty, young .10.xx0011.1 + * read-only, clean, old .11.xx0100.1 + * read-only, clean, young .01.xx0101.1 + * read-only, dirty, old .11.xx0110.1 + * read-only, dirty, young .01.xx0111.1 + * read-write, clean, old .11.xx1100.1 + * read-write, clean, young .01.xx1101.1 + * read-write, dirty, old .10.xx1110.1 + * read-write, dirty, young .00.xx1111.1 + * HW-bits: R read-only, I invalid + * SW-bits: p present, y young, d dirty, r read, w write, s special, + * u unused, l large + * + * pte_none is true for the bit pattern .10.00000000, pte == 0x400 + * pte_swap is true for the bit pattern .11..ooooo.0, (pte & 0x201) == 0x200 + * pte_present is true for the bit pattern .xx.xxxxxx.1, (pte & 0x001) == 0x001 + */ + +/* Bits in the segment/region table address-space-control-element */ +#define _ASCE_ORIGIN ~0xfffUL/* segment table origin */ +#define _ASCE_PRIVATE_SPACE 0x100 /* private space control */ +#define _ASCE_ALT_EVENT 0x80 /* storage alteration event control */ +#define _ASCE_SPACE_SWITCH 0x40 /* space switch event */ +#define _ASCE_REAL_SPACE 0x20 /* real space control */ +#define _ASCE_TYPE_MASK 0x0c /* asce table type mask */ +#define _ASCE_TYPE_REGION1 0x0c /* region first table type */ +#define _ASCE_TYPE_REGION2 0x08 /* region second table type */ +#define _ASCE_TYPE_REGION3 0x04 /* region third table type */ +#define _ASCE_TYPE_SEGMENT 0x00 /* segment table type */ +#define _ASCE_TABLE_LENGTH 0x03 /* region table length */ + +/* Bits in the region table entry */ +#define _REGION_ENTRY_ORIGIN ~0xfffUL/* region/segment table origin */ +#define _REGION_ENTRY_PROTECT 0x200 /* region protection bit */ +#define _REGION_ENTRY_INVALID 0x20 /* invalid region table entry */ +#define _REGION_ENTRY_TYPE_MASK 0x0c /* region/segment table type mask */ +#define _REGION_ENTRY_TYPE_R1 0x0c /* region first table type */ +#define _REGION_ENTRY_TYPE_R2 0x08 /* region second table type */ +#define _REGION_ENTRY_TYPE_R3 0x04 /* region third table type */ +#define _REGION_ENTRY_LENGTH 0x03 /* region third length */ + +#define _REGION1_ENTRY (_REGION_ENTRY_TYPE_R1 | _REGION_ENTRY_LENGTH) +#define _REGION1_ENTRY_EMPTY (_REGION_ENTRY_TYPE_R1 | _REGION_ENTRY_INVALID) +#define _REGION2_ENTRY (_REGION_ENTRY_TYPE_R2 | _REGION_ENTRY_LENGTH) +#define _REGION2_ENTRY_EMPTY (_REGION_ENTRY_TYPE_R2 | _REGION_ENTRY_INVALID) +#define _REGION3_ENTRY (_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_LENGTH) +#define _REGION3_ENTRY_EMPTY (_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_INVALID) + +#define _REGION3_ENTRY_LARGE 0x400 /* RTTE-format control, large page */ +#define _REGION3_ENTRY_RO 0x200 /* page protection bit */ + +/* Bits in the segment table entry */ +#define _SEGMENT_ENTRY_BITS 0xfffffffffffffe33UL +#define _SEGMENT_ENTRY_BITS_LARGE 0xfffffffffff0ff33UL +#define _SEGMENT_ENTRY_ORIGIN_LARGE ~0xfffffUL /* large page address */ +#define _SEGMENT_ENTRY_ORIGIN ~0x7ffUL/* segment table origin */ +#define _SEGMENT_ENTRY_PROTECT 0x200 /* page protection bit */ +#define _SEGMENT_ENTRY_INVALID 0x20 /* invalid segment table entry */ + +#define _SEGMENT_ENTRY (0) +#define _SEGMENT_ENTRY_EMPTY (_SEGMENT_ENTRY_INVALID) + +#define _SEGMENT_ENTRY_DIRTY 0x2000 /* SW segment dirty bit */ +#define _SEGMENT_ENTRY_YOUNG 0x1000 /* SW segment young bit */ +#define _SEGMENT_ENTRY_SPLIT 0x0800 /* THP splitting bit */ +#define _SEGMENT_ENTRY_LARGE 0x0400 /* STE-format control, large page */ +#define _SEGMENT_ENTRY_READ 0x0002 /* SW segment read bit */ +#define _SEGMENT_ENTRY_WRITE 0x0001 /* SW segment write bit */ + +/* + * Segment table entry encoding (R = read-only, I = invalid, y = young bit): + * dy..R...I...wr + * prot-none, clean, old 00..1...1...00 + * prot-none, clean, young 01..1...1...00 + * prot-none, dirty, old 10..1...1...00 + * prot-none, dirty, young 11..1...1...00 + * read-only, clean, old 00..1...1...01 + * read-only, clean, young 01..1...0...01 + * read-only, dirty, old 10..1...1...01 + * read-only, dirty, young 11..1...0...01 + * read-write, clean, old 00..1...1...11 + * read-write, clean, young 01..1...0...11 + * read-write, dirty, old 10..0...1...11 + * read-write, dirty, young 11..0...0...11 + * The segment table origin is used to distinguish empty (origin==0) from + * read-write, old segment table entries (origin!=0) + * HW-bits: R read-only, I invalid + * SW-bits: y young, d dirty, r read, w write + */ + +#define _SEGMENT_ENTRY_SPLIT_BIT 11 /* THP splitting bit number */ + +/* Page status table bits for virtualization */ +#define PGSTE_ACC_BITS 0xf000000000000000UL +#define PGSTE_FP_BIT 0x0800000000000000UL +#define PGSTE_PCL_BIT 0x0080000000000000UL +#define PGSTE_HR_BIT 0x0040000000000000UL +#define PGSTE_HC_BIT 0x0020000000000000UL +#define PGSTE_GR_BIT 0x0004000000000000UL +#define PGSTE_GC_BIT 0x0002000000000000UL +#define PGSTE_UC_BIT 0x0000800000000000UL /* user dirty (migration) */ +#define PGSTE_IN_BIT 0x0000400000000000UL /* IPTE notify bit */ + +/* Guest Page State used for virtualization */ +#define _PGSTE_GPS_ZERO 0x0000000080000000UL +#define _PGSTE_GPS_USAGE_MASK 0x0000000003000000UL +#define _PGSTE_GPS_USAGE_STABLE 0x0000000000000000UL +#define _PGSTE_GPS_USAGE_UNUSED 0x0000000001000000UL + +/* + * A user page table pointer has the space-switch-event bit, the + * private-space-control bit and the storage-alteration-event-control + * bit set. A kernel page table pointer doesn't need them. + */ +#define _ASCE_USER_BITS (_ASCE_SPACE_SWITCH | _ASCE_PRIVATE_SPACE | \ + _ASCE_ALT_EVENT) + +/* + * Page protection definitions. + */ +#define PAGE_NONE __pgprot(_PAGE_PRESENT | _PAGE_INVALID) +#define PAGE_READ __pgprot(_PAGE_PRESENT | _PAGE_READ | \ + _PAGE_INVALID | _PAGE_PROTECT) +#define PAGE_WRITE __pgprot(_PAGE_PRESENT | _PAGE_READ | _PAGE_WRITE | \ + _PAGE_INVALID | _PAGE_PROTECT) + +#define PAGE_SHARED __pgprot(_PAGE_PRESENT | _PAGE_READ | _PAGE_WRITE | \ + _PAGE_YOUNG | _PAGE_DIRTY) +#define PAGE_KERNEL __pgprot(_PAGE_PRESENT | _PAGE_READ | _PAGE_WRITE | \ + _PAGE_YOUNG | _PAGE_DIRTY) +#define PAGE_KERNEL_RO __pgprot(_PAGE_PRESENT | _PAGE_READ | _PAGE_YOUNG | \ + _PAGE_PROTECT) + +/* + * On s390 the page table entry has an invalid bit and a read-only bit. + * Read permission implies execute permission and write permission + * implies read permission. + */ + /*xwr*/ +#define __P000 PAGE_NONE +#define __P001 PAGE_READ +#define __P010 PAGE_READ +#define __P011 PAGE_READ +#define __P100 PAGE_READ +#define __P101 PAGE_READ +#define __P110 PAGE_READ +#define __P111 PAGE_READ + +#define __S000 PAGE_NONE +#define __S001 PAGE_READ +#define __S010 PAGE_WRITE +#define __S011 PAGE_WRITE +#define __S100 PAGE_READ +#define __S101 PAGE_READ +#define __S110 PAGE_WRITE +#define __S111 PAGE_WRITE + +/* + * Segment entry (large page) protection definitions. + */ +#define SEGMENT_NONE __pgprot(_SEGMENT_ENTRY_INVALID | \ + _SEGMENT_ENTRY_PROTECT) +#define SEGMENT_READ __pgprot(_SEGMENT_ENTRY_PROTECT | \ + _SEGMENT_ENTRY_READ) +#define SEGMENT_WRITE __pgprot(_SEGMENT_ENTRY_READ | \ + _SEGMENT_ENTRY_WRITE) + +static inline int mm_has_pgste(struct mm_struct *mm) +{ +#ifdef CONFIG_PGSTE + if (unlikely(mm->context.has_pgste)) + return 1; +#endif + return 0; +} + +static inline int mm_alloc_pgste(struct mm_struct *mm) +{ +#ifdef CONFIG_PGSTE + if (unlikely(mm->context.alloc_pgste)) + return 1; +#endif + return 0; +} + +/* + * In the case that a guest uses storage keys + * faults should no longer be backed by zero pages + */ +#define mm_forbids_zeropage mm_use_skey +static inline int mm_use_skey(struct mm_struct *mm) +{ +#ifdef CONFIG_PGSTE + if (mm->context.use_skey) + return 1; +#endif + return 0; +} + +/* + * pgd/pmd/pte query functions + */ +static inline int pgd_present(pgd_t pgd) +{ + if ((pgd_val(pgd) & _REGION_ENTRY_TYPE_MASK) < _REGION_ENTRY_TYPE_R2) + return 1; + return (pgd_val(pgd) & _REGION_ENTRY_ORIGIN) != 0UL; +} + +static inline int pgd_none(pgd_t pgd) +{ + if ((pgd_val(pgd) & _REGION_ENTRY_TYPE_MASK) < _REGION_ENTRY_TYPE_R2) + return 0; + return (pgd_val(pgd) & _REGION_ENTRY_INVALID) != 0UL; +} + +static inline int pgd_bad(pgd_t pgd) +{ + /* + * With dynamic page table levels the pgd can be a region table + * entry or a segment table entry. Check for the bit that are + * invalid for either table entry. + */ + unsigned long mask = + ~_SEGMENT_ENTRY_ORIGIN & ~_REGION_ENTRY_INVALID & + ~_REGION_ENTRY_TYPE_MASK & ~_REGION_ENTRY_LENGTH; + return (pgd_val(pgd) & mask) != 0; +} + +static inline int pud_present(pud_t pud) +{ + if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) < _REGION_ENTRY_TYPE_R3) + return 1; + return (pud_val(pud) & _REGION_ENTRY_ORIGIN) != 0UL; +} + +static inline int pud_none(pud_t pud) +{ + if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) < _REGION_ENTRY_TYPE_R3) + return 0; + return (pud_val(pud) & _REGION_ENTRY_INVALID) != 0UL; +} + +static inline int pud_large(pud_t pud) +{ + if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) != _REGION_ENTRY_TYPE_R3) + return 0; + return !!(pud_val(pud) & _REGION3_ENTRY_LARGE); +} + +static inline int pud_bad(pud_t pud) +{ + /* + * With dynamic page table levels the pud can be a region table + * entry or a segment table entry. Check for the bit that are + * invalid for either table entry. + */ + unsigned long mask = + ~_SEGMENT_ENTRY_ORIGIN & ~_REGION_ENTRY_INVALID & + ~_REGION_ENTRY_TYPE_MASK & ~_REGION_ENTRY_LENGTH; + return (pud_val(pud) & mask) != 0; +} + +static inline int pmd_present(pmd_t pmd) +{ + return pmd_val(pmd) != _SEGMENT_ENTRY_INVALID; +} + +static inline int pmd_none(pmd_t pmd) +{ + return pmd_val(pmd) == _SEGMENT_ENTRY_INVALID; +} + +static inline int pmd_large(pmd_t pmd) +{ + return (pmd_val(pmd) & _SEGMENT_ENTRY_LARGE) != 0; +} + +static inline unsigned long pmd_pfn(pmd_t pmd) +{ + unsigned long origin_mask; + + origin_mask = _SEGMENT_ENTRY_ORIGIN; + if (pmd_large(pmd)) + origin_mask = _SEGMENT_ENTRY_ORIGIN_LARGE; + return (pmd_val(pmd) & origin_mask) >> PAGE_SHIFT; +} + +static inline int pmd_bad(pmd_t pmd) +{ + if (pmd_large(pmd)) + return (pmd_val(pmd) & ~_SEGMENT_ENTRY_BITS_LARGE) != 0; + return (pmd_val(pmd) & ~_SEGMENT_ENTRY_BITS) != 0; +} + +#define __HAVE_ARCH_PMDP_SPLITTING_FLUSH +extern void pmdp_splitting_flush(struct vm_area_struct *vma, + unsigned long addr, pmd_t *pmdp); + +#define __HAVE_ARCH_PMDP_SET_ACCESS_FLAGS +extern int pmdp_set_access_flags(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp, + pmd_t entry, int dirty); + +#define __HAVE_ARCH_PMDP_CLEAR_YOUNG_FLUSH +extern int pmdp_clear_flush_young(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp); + +#define __HAVE_ARCH_PMD_WRITE +static inline int pmd_write(pmd_t pmd) +{ + return (pmd_val(pmd) & _SEGMENT_ENTRY_WRITE) != 0; +} + +static inline int pmd_dirty(pmd_t pmd) +{ + int dirty = 1; + if (pmd_large(pmd)) + dirty = (pmd_val(pmd) & _SEGMENT_ENTRY_DIRTY) != 0; + return dirty; +} + +static inline int pmd_young(pmd_t pmd) +{ + int young = 1; + if (pmd_large(pmd)) + young = (pmd_val(pmd) & _SEGMENT_ENTRY_YOUNG) != 0; + return young; +} + +static inline int pte_present(pte_t pte) +{ + /* Bit pattern: (pte & 0x001) == 0x001 */ + return (pte_val(pte) & _PAGE_PRESENT) != 0; +} + +static inline int pte_none(pte_t pte) +{ + /* Bit pattern: pte == 0x400 */ + return pte_val(pte) == _PAGE_INVALID; +} + +static inline int pte_swap(pte_t pte) +{ + /* Bit pattern: (pte & 0x201) == 0x200 */ + return (pte_val(pte) & (_PAGE_PROTECT | _PAGE_PRESENT)) + == _PAGE_PROTECT; +} + +static inline int pte_special(pte_t pte) +{ + return (pte_val(pte) & _PAGE_SPECIAL); +} + +#define __HAVE_ARCH_PTE_SAME +static inline int pte_same(pte_t a, pte_t b) +{ + return pte_val(a) == pte_val(b); +} + +static inline pgste_t pgste_get_lock(pte_t *ptep) +{ + unsigned long new = 0; +#ifdef CONFIG_PGSTE + unsigned long old; + + preempt_disable(); + asm( + " lg %0,%2\n" + "0: lgr %1,%0\n" + " nihh %0,0xff7f\n" /* clear PCL bit in old */ + " oihh %1,0x0080\n" /* set PCL bit in new */ + " csg %0,%1,%2\n" + " jl 0b\n" + : "=&d" (old), "=&d" (new), "=Q" (ptep[PTRS_PER_PTE]) + : "Q" (ptep[PTRS_PER_PTE]) : "cc", "memory"); +#endif + return __pgste(new); +} + +static inline void pgste_set_unlock(pte_t *ptep, pgste_t pgste) +{ +#ifdef CONFIG_PGSTE + asm( + " nihh %1,0xff7f\n" /* clear PCL bit */ + " stg %1,%0\n" + : "=Q" (ptep[PTRS_PER_PTE]) + : "d" (pgste_val(pgste)), "Q" (ptep[PTRS_PER_PTE]) + : "cc", "memory"); + preempt_enable(); +#endif +} + +static inline pgste_t pgste_get(pte_t *ptep) +{ + unsigned long pgste = 0; +#ifdef CONFIG_PGSTE + pgste = *(unsigned long *)(ptep + PTRS_PER_PTE); +#endif + return __pgste(pgste); +} + +static inline void pgste_set(pte_t *ptep, pgste_t pgste) +{ +#ifdef CONFIG_PGSTE + *(pgste_t *)(ptep + PTRS_PER_PTE) = pgste; +#endif +} + +static inline pgste_t pgste_update_all(pte_t *ptep, pgste_t pgste, + struct mm_struct *mm) +{ +#ifdef CONFIG_PGSTE + unsigned long address, bits, skey; + + if (!mm_use_skey(mm) || pte_val(*ptep) & _PAGE_INVALID) + return pgste; + address = pte_val(*ptep) & PAGE_MASK; + skey = (unsigned long) page_get_storage_key(address); + bits = skey & (_PAGE_CHANGED | _PAGE_REFERENCED); + /* Transfer page changed & referenced bit to guest bits in pgste */ + pgste_val(pgste) |= bits << 48; /* GR bit & GC bit */ + /* Copy page access key and fetch protection bit to pgste */ + pgste_val(pgste) &= ~(PGSTE_ACC_BITS | PGSTE_FP_BIT); + pgste_val(pgste) |= (skey & (_PAGE_ACC_BITS | _PAGE_FP_BIT)) << 56; +#endif + return pgste; + +} + +static inline void pgste_set_key(pte_t *ptep, pgste_t pgste, pte_t entry, + struct mm_struct *mm) +{ +#ifdef CONFIG_PGSTE + unsigned long address; + unsigned long nkey; + + if (!mm_use_skey(mm) || pte_val(entry) & _PAGE_INVALID) + return; + VM_BUG_ON(!(pte_val(*ptep) & _PAGE_INVALID)); + address = pte_val(entry) & PAGE_MASK; + /* + * Set page access key and fetch protection bit from pgste. + * The guest C/R information is still in the PGSTE, set real + * key C/R to 0. + */ + nkey = (pgste_val(pgste) & (PGSTE_ACC_BITS | PGSTE_FP_BIT)) >> 56; + nkey |= (pgste_val(pgste) & (PGSTE_GR_BIT | PGSTE_GC_BIT)) >> 48; + page_set_storage_key(address, nkey, 0); +#endif +} + +static inline pgste_t pgste_set_pte(pte_t *ptep, pgste_t pgste, pte_t entry) +{ + if ((pte_val(entry) & _PAGE_PRESENT) && + (pte_val(entry) & _PAGE_WRITE) && + !(pte_val(entry) & _PAGE_INVALID)) { + if (!MACHINE_HAS_ESOP) { + /* + * Without enhanced suppression-on-protection force + * the dirty bit on for all writable ptes. + */ + pte_val(entry) |= _PAGE_DIRTY; + pte_val(entry) &= ~_PAGE_PROTECT; + } + if (!(pte_val(entry) & _PAGE_PROTECT)) + /* This pte allows write access, set user-dirty */ + pgste_val(pgste) |= PGSTE_UC_BIT; + } + *ptep = entry; + return pgste; +} + +/** + * struct gmap_struct - guest address space + * @crst_list: list of all crst tables used in the guest address space + * @mm: pointer to the parent mm_struct + * @guest_to_host: radix tree with guest to host address translation + * @host_to_guest: radix tree with pointer to segment table entries + * @guest_table_lock: spinlock to protect all entries in the guest page table + * @table: pointer to the page directory + * @asce: address space control element for gmap page table + * @pfault_enabled: defines if pfaults are applicable for the guest + */ +struct gmap { + struct list_head list; + struct list_head crst_list; + struct mm_struct *mm; + struct radix_tree_root guest_to_host; + struct radix_tree_root host_to_guest; + spinlock_t guest_table_lock; + unsigned long *table; + unsigned long asce; + unsigned long asce_end; + void *private; + bool pfault_enabled; +}; + +/** + * struct gmap_notifier - notify function block for page invalidation + * @notifier_call: address of callback function + */ +struct gmap_notifier { + struct list_head list; + void (*notifier_call)(struct gmap *gmap, unsigned long gaddr); +}; + +struct gmap *gmap_alloc(struct mm_struct *mm, unsigned long limit); +void gmap_free(struct gmap *gmap); +void gmap_enable(struct gmap *gmap); +void gmap_disable(struct gmap *gmap); +int gmap_map_segment(struct gmap *gmap, unsigned long from, + unsigned long to, unsigned long len); +int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len); +unsigned long __gmap_translate(struct gmap *, unsigned long gaddr); +unsigned long gmap_translate(struct gmap *, unsigned long gaddr); +int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr); +int gmap_fault(struct gmap *, unsigned long gaddr, unsigned int fault_flags); +void gmap_discard(struct gmap *, unsigned long from, unsigned long to); +void __gmap_zap(struct gmap *, unsigned long gaddr); +bool gmap_test_and_clear_dirty(unsigned long address, struct gmap *); + + +void gmap_register_ipte_notifier(struct gmap_notifier *); +void gmap_unregister_ipte_notifier(struct gmap_notifier *); +int gmap_ipte_notify(struct gmap *, unsigned long start, unsigned long len); +void gmap_do_ipte_notify(struct mm_struct *, unsigned long addr, pte_t *); + +static inline pgste_t pgste_ipte_notify(struct mm_struct *mm, + unsigned long addr, + pte_t *ptep, pgste_t pgste) +{ +#ifdef CONFIG_PGSTE + if (pgste_val(pgste) & PGSTE_IN_BIT) { + pgste_val(pgste) &= ~PGSTE_IN_BIT; + gmap_do_ipte_notify(mm, addr, ptep); + } +#endif + return pgste; +} + +/* + * Certain architectures need to do special things when PTEs + * within a page table are directly modified. Thus, the following + * hook is made available. + */ +static inline void set_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t entry) +{ + pgste_t pgste; + + if (mm_has_pgste(mm)) { + pgste = pgste_get_lock(ptep); + pgste_val(pgste) &= ~_PGSTE_GPS_ZERO; + pgste_set_key(ptep, pgste, entry, mm); + pgste = pgste_set_pte(ptep, pgste, entry); + pgste_set_unlock(ptep, pgste); + } else { + *ptep = entry; + } +} + +/* + * query functions pte_write/pte_dirty/pte_young only work if + * pte_present() is true. Undefined behaviour if not.. + */ +static inline int pte_write(pte_t pte) +{ + return (pte_val(pte) & _PAGE_WRITE) != 0; +} + +static inline int pte_dirty(pte_t pte) +{ + return (pte_val(pte) & _PAGE_DIRTY) != 0; +} + +static inline int pte_young(pte_t pte) +{ + return (pte_val(pte) & _PAGE_YOUNG) != 0; +} + +#define __HAVE_ARCH_PTE_UNUSED +static inline int pte_unused(pte_t pte) +{ + return pte_val(pte) & _PAGE_UNUSED; +} + +/* + * pgd/pmd/pte modification functions + */ + +static inline void pgd_clear(pgd_t *pgd) +{ + if ((pgd_val(*pgd) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R2) + pgd_val(*pgd) = _REGION2_ENTRY_EMPTY; +} + +static inline void pud_clear(pud_t *pud) +{ + if ((pud_val(*pud) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R3) + pud_val(*pud) = _REGION3_ENTRY_EMPTY; +} + +static inline void pmd_clear(pmd_t *pmdp) +{ + pmd_val(*pmdp) = _SEGMENT_ENTRY_INVALID; +} + +static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) +{ + pte_val(*ptep) = _PAGE_INVALID; +} + +/* + * The following pte modification functions only work if + * pte_present() is true. Undefined behaviour if not.. + */ +static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) +{ + pte_val(pte) &= _PAGE_CHG_MASK; + pte_val(pte) |= pgprot_val(newprot); + /* + * newprot for PAGE_NONE, PAGE_READ and PAGE_WRITE has the + * invalid bit set, clear it again for readable, young pages + */ + if ((pte_val(pte) & _PAGE_YOUNG) && (pte_val(pte) & _PAGE_READ)) + pte_val(pte) &= ~_PAGE_INVALID; + /* + * newprot for PAGE_READ and PAGE_WRITE has the page protection + * bit set, clear it again for writable, dirty pages + */ + if ((pte_val(pte) & _PAGE_DIRTY) && (pte_val(pte) & _PAGE_WRITE)) + pte_val(pte) &= ~_PAGE_PROTECT; + return pte; +} + +static inline pte_t pte_wrprotect(pte_t pte) +{ + pte_val(pte) &= ~_PAGE_WRITE; + pte_val(pte) |= _PAGE_PROTECT; + return pte; +} + +static inline pte_t pte_mkwrite(pte_t pte) +{ + pte_val(pte) |= _PAGE_WRITE; + if (pte_val(pte) & _PAGE_DIRTY) + pte_val(pte) &= ~_PAGE_PROTECT; + return pte; +} + +static inline pte_t pte_mkclean(pte_t pte) +{ + pte_val(pte) &= ~_PAGE_DIRTY; + pte_val(pte) |= _PAGE_PROTECT; + return pte; +} + +static inline pte_t pte_mkdirty(pte_t pte) +{ + pte_val(pte) |= _PAGE_DIRTY; + if (pte_val(pte) & _PAGE_WRITE) + pte_val(pte) &= ~_PAGE_PROTECT; + return pte; +} + +static inline pte_t pte_mkold(pte_t pte) +{ + pte_val(pte) &= ~_PAGE_YOUNG; + pte_val(pte) |= _PAGE_INVALID; + return pte; +} + +static inline pte_t pte_mkyoung(pte_t pte) +{ + pte_val(pte) |= _PAGE_YOUNG; + if (pte_val(pte) & _PAGE_READ) + pte_val(pte) &= ~_PAGE_INVALID; + return pte; +} + +static inline pte_t pte_mkspecial(pte_t pte) +{ + pte_val(pte) |= _PAGE_SPECIAL; + return pte; +} + +#ifdef CONFIG_HUGETLB_PAGE +static inline pte_t pte_mkhuge(pte_t pte) +{ + pte_val(pte) |= _PAGE_LARGE; + return pte; +} +#endif + +static inline void __ptep_ipte(unsigned long address, pte_t *ptep) +{ + unsigned long pto = (unsigned long) ptep; + + /* Invalidation + global TLB flush for the pte */ + asm volatile( + " ipte %2,%3" + : "=m" (*ptep) : "m" (*ptep), "a" (pto), "a" (address)); +} + +static inline void __ptep_ipte_local(unsigned long address, pte_t *ptep) +{ + unsigned long pto = (unsigned long) ptep; + + /* Invalidation + local TLB flush for the pte */ + asm volatile( + " .insn rrf,0xb2210000,%2,%3,0,1" + : "=m" (*ptep) : "m" (*ptep), "a" (pto), "a" (address)); +} + +static inline void __ptep_ipte_range(unsigned long address, int nr, pte_t *ptep) +{ + unsigned long pto = (unsigned long) ptep; + + /* Invalidate a range of ptes + global TLB flush of the ptes */ + do { + asm volatile( + " .insn rrf,0xb2210000,%2,%0,%1,0" + : "+a" (address), "+a" (nr) : "a" (pto) : "memory"); + } while (nr != 255); +} + +static inline void ptep_flush_direct(struct mm_struct *mm, + unsigned long address, pte_t *ptep) +{ + int active, count; + + if (pte_val(*ptep) & _PAGE_INVALID) + return; + active = (mm == current->active_mm) ? 1 : 0; + count = atomic_add_return(0x10000, &mm->context.attach_count); + if (MACHINE_HAS_TLB_LC && (count & 0xffff) <= active && + cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) + __ptep_ipte_local(address, ptep); + else + __ptep_ipte(address, ptep); + atomic_sub(0x10000, &mm->context.attach_count); +} + +static inline void ptep_flush_lazy(struct mm_struct *mm, + unsigned long address, pte_t *ptep) +{ + int active, count; + + if (pte_val(*ptep) & _PAGE_INVALID) + return; + active = (mm == current->active_mm) ? 1 : 0; + count = atomic_add_return(0x10000, &mm->context.attach_count); + if ((count & 0xffff) <= active) { + pte_val(*ptep) |= _PAGE_INVALID; + mm->context.flush_mm = 1; + } else + __ptep_ipte(address, ptep); + atomic_sub(0x10000, &mm->context.attach_count); +} + +/* + * Get (and clear) the user dirty bit for a pte. + */ +static inline int ptep_test_and_clear_user_dirty(struct mm_struct *mm, + unsigned long addr, + pte_t *ptep) +{ + pgste_t pgste; + pte_t pte; + int dirty; + + if (!mm_has_pgste(mm)) + return 0; + pgste = pgste_get_lock(ptep); + dirty = !!(pgste_val(pgste) & PGSTE_UC_BIT); + pgste_val(pgste) &= ~PGSTE_UC_BIT; + pte = *ptep; + if (dirty && (pte_val(pte) & _PAGE_PRESENT)) { + pgste = pgste_ipte_notify(mm, addr, ptep, pgste); + __ptep_ipte(addr, ptep); + if (MACHINE_HAS_ESOP || !(pte_val(pte) & _PAGE_WRITE)) + pte_val(pte) |= _PAGE_PROTECT; + else + pte_val(pte) |= _PAGE_INVALID; + *ptep = pte; + } + pgste_set_unlock(ptep, pgste); + return dirty; +} + +#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG +static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, + unsigned long addr, pte_t *ptep) +{ + pgste_t pgste; + pte_t pte, oldpte; + int young; + + if (mm_has_pgste(vma->vm_mm)) { + pgste = pgste_get_lock(ptep); + pgste = pgste_ipte_notify(vma->vm_mm, addr, ptep, pgste); + } + + oldpte = pte = *ptep; + ptep_flush_direct(vma->vm_mm, addr, ptep); + young = pte_young(pte); + pte = pte_mkold(pte); + + if (mm_has_pgste(vma->vm_mm)) { + pgste = pgste_update_all(&oldpte, pgste, vma->vm_mm); + pgste = pgste_set_pte(ptep, pgste, pte); + pgste_set_unlock(ptep, pgste); + } else + *ptep = pte; + + return young; +} + +#define __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH +static inline int ptep_clear_flush_young(struct vm_area_struct *vma, + unsigned long address, pte_t *ptep) +{ + return ptep_test_and_clear_young(vma, address, ptep); +} + +/* + * This is hard to understand. ptep_get_and_clear and ptep_clear_flush + * both clear the TLB for the unmapped pte. The reason is that + * ptep_get_and_clear is used in common code (e.g. change_pte_range) + * to modify an active pte. The sequence is + * 1) ptep_get_and_clear + * 2) set_pte_at + * 3) flush_tlb_range + * On s390 the tlb needs to get flushed with the modification of the pte + * if the pte is active. The only way how this can be implemented is to + * have ptep_get_and_clear do the tlb flush. In exchange flush_tlb_range + * is a nop. + */ +#define __HAVE_ARCH_PTEP_GET_AND_CLEAR +static inline pte_t ptep_get_and_clear(struct mm_struct *mm, + unsigned long address, pte_t *ptep) +{ + pgste_t pgste; + pte_t pte; + + if (mm_has_pgste(mm)) { + pgste = pgste_get_lock(ptep); + pgste = pgste_ipte_notify(mm, address, ptep, pgste); + } + + pte = *ptep; + ptep_flush_lazy(mm, address, ptep); + pte_val(*ptep) = _PAGE_INVALID; + + if (mm_has_pgste(mm)) { + pgste = pgste_update_all(&pte, pgste, mm); + pgste_set_unlock(ptep, pgste); + } + return pte; +} + +#define __HAVE_ARCH_PTEP_MODIFY_PROT_TRANSACTION +static inline pte_t ptep_modify_prot_start(struct mm_struct *mm, + unsigned long address, + pte_t *ptep) +{ + pgste_t pgste; + pte_t pte; + + if (mm_has_pgste(mm)) { + pgste = pgste_get_lock(ptep); + pgste_ipte_notify(mm, address, ptep, pgste); + } + + pte = *ptep; + ptep_flush_lazy(mm, address, ptep); + + if (mm_has_pgste(mm)) { + pgste = pgste_update_all(&pte, pgste, mm); + pgste_set(ptep, pgste); + } + return pte; +} + +static inline void ptep_modify_prot_commit(struct mm_struct *mm, + unsigned long address, + pte_t *ptep, pte_t pte) +{ + pgste_t pgste; + + if (mm_has_pgste(mm)) { + pgste = pgste_get(ptep); + pgste_set_key(ptep, pgste, pte, mm); + pgste = pgste_set_pte(ptep, pgste, pte); + pgste_set_unlock(ptep, pgste); + } else + *ptep = pte; +} + +#define __HAVE_ARCH_PTEP_CLEAR_FLUSH +static inline pte_t ptep_clear_flush(struct vm_area_struct *vma, + unsigned long address, pte_t *ptep) +{ + pgste_t pgste; + pte_t pte; + + if (mm_has_pgste(vma->vm_mm)) { + pgste = pgste_get_lock(ptep); + pgste = pgste_ipte_notify(vma->vm_mm, address, ptep, pgste); + } + + pte = *ptep; + ptep_flush_direct(vma->vm_mm, address, ptep); + pte_val(*ptep) = _PAGE_INVALID; + + if (mm_has_pgste(vma->vm_mm)) { + if ((pgste_val(pgste) & _PGSTE_GPS_USAGE_MASK) == + _PGSTE_GPS_USAGE_UNUSED) + pte_val(pte) |= _PAGE_UNUSED; + pgste = pgste_update_all(&pte, pgste, vma->vm_mm); + pgste_set_unlock(ptep, pgste); + } + return pte; +} + +/* + * The batched pte unmap code uses ptep_get_and_clear_full to clear the + * ptes. Here an optimization is possible. tlb_gather_mmu flushes all + * tlbs of an mm if it can guarantee that the ptes of the mm_struct + * cannot be accessed while the batched unmap is running. In this case + * full==1 and a simple pte_clear is enough. See tlb.h. + */ +#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL +static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, + unsigned long address, + pte_t *ptep, int full) +{ + pgste_t pgste; + pte_t pte; + + if (!full && mm_has_pgste(mm)) { + pgste = pgste_get_lock(ptep); + pgste = pgste_ipte_notify(mm, address, ptep, pgste); + } + + pte = *ptep; + if (!full) + ptep_flush_lazy(mm, address, ptep); + pte_val(*ptep) = _PAGE_INVALID; + + if (!full && mm_has_pgste(mm)) { + pgste = pgste_update_all(&pte, pgste, mm); + pgste_set_unlock(ptep, pgste); + } + return pte; +} + +#define __HAVE_ARCH_PTEP_SET_WRPROTECT +static inline pte_t ptep_set_wrprotect(struct mm_struct *mm, + unsigned long address, pte_t *ptep) +{ + pgste_t pgste; + pte_t pte = *ptep; + + if (pte_write(pte)) { + if (mm_has_pgste(mm)) { + pgste = pgste_get_lock(ptep); + pgste = pgste_ipte_notify(mm, address, ptep, pgste); + } + + ptep_flush_lazy(mm, address, ptep); + pte = pte_wrprotect(pte); + + if (mm_has_pgste(mm)) { + pgste = pgste_set_pte(ptep, pgste, pte); + pgste_set_unlock(ptep, pgste); + } else + *ptep = pte; + } + return pte; +} + +#define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS +static inline int ptep_set_access_flags(struct vm_area_struct *vma, + unsigned long address, pte_t *ptep, + pte_t entry, int dirty) +{ + pgste_t pgste; + + if (pte_same(*ptep, entry)) + return 0; + if (mm_has_pgste(vma->vm_mm)) { + pgste = pgste_get_lock(ptep); + pgste = pgste_ipte_notify(vma->vm_mm, address, ptep, pgste); + } + + ptep_flush_direct(vma->vm_mm, address, ptep); + + if (mm_has_pgste(vma->vm_mm)) { + pgste_set_key(ptep, pgste, entry, vma->vm_mm); + pgste = pgste_set_pte(ptep, pgste, entry); + pgste_set_unlock(ptep, pgste); + } else + *ptep = entry; + return 1; +} + +/* + * Conversion functions: convert a page and protection to a page entry, + * and a page entry and page directory to the page they refer to. + */ +static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot) +{ + pte_t __pte; + pte_val(__pte) = physpage + pgprot_val(pgprot); + return pte_mkyoung(__pte); +} + +static inline pte_t mk_pte(struct page *page, pgprot_t pgprot) +{ + unsigned long physpage = page_to_phys(page); + pte_t __pte = mk_pte_phys(physpage, pgprot); + + if (pte_write(__pte) && PageDirty(page)) + __pte = pte_mkdirty(__pte); + return __pte; +} + +#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1)) +#define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1)) +#define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1)) +#define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE-1)) + +#define pgd_offset(mm, address) ((mm)->pgd + pgd_index(address)) +#define pgd_offset_k(address) pgd_offset(&init_mm, address) + +#define pmd_deref(pmd) (pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN) +#define pud_deref(pud) (pud_val(pud) & _REGION_ENTRY_ORIGIN) +#define pgd_deref(pgd) (pgd_val(pgd) & _REGION_ENTRY_ORIGIN) + +static inline pud_t *pud_offset(pgd_t *pgd, unsigned long address) +{ + pud_t *pud = (pud_t *) pgd; + if ((pgd_val(*pgd) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R2) + pud = (pud_t *) pgd_deref(*pgd); + return pud + pud_index(address); +} + +static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address) +{ + pmd_t *pmd = (pmd_t *) pud; + if ((pud_val(*pud) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R3) + pmd = (pmd_t *) pud_deref(*pud); + return pmd + pmd_index(address); +} + +#define pfn_pte(pfn,pgprot) mk_pte_phys(__pa((pfn) << PAGE_SHIFT),(pgprot)) +#define pte_pfn(x) (pte_val(x) >> PAGE_SHIFT) +#define pte_page(x) pfn_to_page(pte_pfn(x)) + +#define pmd_page(pmd) pfn_to_page(pmd_pfn(pmd)) + +/* Find an entry in the lowest level page table.. */ +#define pte_offset(pmd, addr) ((pte_t *) pmd_deref(*(pmd)) + pte_index(addr)) +#define pte_offset_kernel(pmd, address) pte_offset(pmd,address) +#define pte_offset_map(pmd, address) pte_offset_kernel(pmd, address) +#define pte_unmap(pte) do { } while (0) + +#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLB_PAGE) +static inline unsigned long massage_pgprot_pmd(pgprot_t pgprot) +{ + /* + * pgprot is PAGE_NONE, PAGE_READ, or PAGE_WRITE (see __Pxxx / __Sxxx) + * Convert to segment table entry format. + */ + if (pgprot_val(pgprot) == pgprot_val(PAGE_NONE)) + return pgprot_val(SEGMENT_NONE); + if (pgprot_val(pgprot) == pgprot_val(PAGE_READ)) + return pgprot_val(SEGMENT_READ); + return pgprot_val(SEGMENT_WRITE); +} + +static inline pmd_t pmd_wrprotect(pmd_t pmd) +{ + pmd_val(pmd) &= ~_SEGMENT_ENTRY_WRITE; + pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT; + return pmd; +} + +static inline pmd_t pmd_mkwrite(pmd_t pmd) +{ + pmd_val(pmd) |= _SEGMENT_ENTRY_WRITE; + if (pmd_large(pmd) && !(pmd_val(pmd) & _SEGMENT_ENTRY_DIRTY)) + return pmd; + pmd_val(pmd) &= ~_SEGMENT_ENTRY_PROTECT; + return pmd; +} + +static inline pmd_t pmd_mkclean(pmd_t pmd) +{ + if (pmd_large(pmd)) { + pmd_val(pmd) &= ~_SEGMENT_ENTRY_DIRTY; + pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT; + } + return pmd; +} + +static inline pmd_t pmd_mkdirty(pmd_t pmd) +{ + if (pmd_large(pmd)) { + pmd_val(pmd) |= _SEGMENT_ENTRY_DIRTY; + if (pmd_val(pmd) & _SEGMENT_ENTRY_WRITE) + pmd_val(pmd) &= ~_SEGMENT_ENTRY_PROTECT; + } + return pmd; +} + +static inline pmd_t pmd_mkyoung(pmd_t pmd) +{ + if (pmd_large(pmd)) { + pmd_val(pmd) |= _SEGMENT_ENTRY_YOUNG; + if (pmd_val(pmd) & _SEGMENT_ENTRY_READ) + pmd_val(pmd) &= ~_SEGMENT_ENTRY_INVALID; + } + return pmd; +} + +static inline pmd_t pmd_mkold(pmd_t pmd) +{ + if (pmd_large(pmd)) { + pmd_val(pmd) &= ~_SEGMENT_ENTRY_YOUNG; + pmd_val(pmd) |= _SEGMENT_ENTRY_INVALID; + } + return pmd; +} + +static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot) +{ + if (pmd_large(pmd)) { + pmd_val(pmd) &= _SEGMENT_ENTRY_ORIGIN_LARGE | + _SEGMENT_ENTRY_DIRTY | _SEGMENT_ENTRY_YOUNG | + _SEGMENT_ENTRY_LARGE | _SEGMENT_ENTRY_SPLIT; + pmd_val(pmd) |= massage_pgprot_pmd(newprot); + if (!(pmd_val(pmd) & _SEGMENT_ENTRY_DIRTY)) + pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT; + if (!(pmd_val(pmd) & _SEGMENT_ENTRY_YOUNG)) + pmd_val(pmd) |= _SEGMENT_ENTRY_INVALID; + return pmd; + } + pmd_val(pmd) &= _SEGMENT_ENTRY_ORIGIN; + pmd_val(pmd) |= massage_pgprot_pmd(newprot); + return pmd; +} + +static inline pmd_t mk_pmd_phys(unsigned long physpage, pgprot_t pgprot) +{ + pmd_t __pmd; + pmd_val(__pmd) = physpage + massage_pgprot_pmd(pgprot); + return __pmd; +} + +#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLB_PAGE */ + +static inline void __pmdp_csp(pmd_t *pmdp) +{ + register unsigned long reg2 asm("2") = pmd_val(*pmdp); + register unsigned long reg3 asm("3") = pmd_val(*pmdp) | + _SEGMENT_ENTRY_INVALID; + register unsigned long reg4 asm("4") = ((unsigned long) pmdp) + 5; + + asm volatile( + " csp %1,%3" + : "=m" (*pmdp) + : "d" (reg2), "d" (reg3), "d" (reg4), "m" (*pmdp) : "cc"); +} + +static inline void __pmdp_idte(unsigned long address, pmd_t *pmdp) +{ + unsigned long sto; + + sto = (unsigned long) pmdp - pmd_index(address) * sizeof(pmd_t); + asm volatile( + " .insn rrf,0xb98e0000,%2,%3,0,0" + : "=m" (*pmdp) + : "m" (*pmdp), "a" (sto), "a" ((address & HPAGE_MASK)) + : "cc" ); +} + +static inline void __pmdp_idte_local(unsigned long address, pmd_t *pmdp) +{ + unsigned long sto; + + sto = (unsigned long) pmdp - pmd_index(address) * sizeof(pmd_t); + asm volatile( + " .insn rrf,0xb98e0000,%2,%3,0,1" + : "=m" (*pmdp) + : "m" (*pmdp), "a" (sto), "a" ((address & HPAGE_MASK)) + : "cc" ); +} + +static inline void pmdp_flush_direct(struct mm_struct *mm, + unsigned long address, pmd_t *pmdp) +{ + int active, count; + + if (pmd_val(*pmdp) & _SEGMENT_ENTRY_INVALID) + return; + if (!MACHINE_HAS_IDTE) { + __pmdp_csp(pmdp); + return; + } + active = (mm == current->active_mm) ? 1 : 0; + count = atomic_add_return(0x10000, &mm->context.attach_count); + if (MACHINE_HAS_TLB_LC && (count & 0xffff) <= active && + cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) + __pmdp_idte_local(address, pmdp); + else + __pmdp_idte(address, pmdp); + atomic_sub(0x10000, &mm->context.attach_count); +} + +static inline void pmdp_flush_lazy(struct mm_struct *mm, + unsigned long address, pmd_t *pmdp) +{ + int active, count; + + if (pmd_val(*pmdp) & _SEGMENT_ENTRY_INVALID) + return; + active = (mm == current->active_mm) ? 1 : 0; + count = atomic_add_return(0x10000, &mm->context.attach_count); + if ((count & 0xffff) <= active) { + pmd_val(*pmdp) |= _SEGMENT_ENTRY_INVALID; + mm->context.flush_mm = 1; + } else if (MACHINE_HAS_IDTE) + __pmdp_idte(address, pmdp); + else + __pmdp_csp(pmdp); + atomic_sub(0x10000, &mm->context.attach_count); +} + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + +#define __HAVE_ARCH_PGTABLE_DEPOSIT +extern void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp, + pgtable_t pgtable); + +#define __HAVE_ARCH_PGTABLE_WITHDRAW +extern pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp); + +static inline int pmd_trans_splitting(pmd_t pmd) +{ + return (pmd_val(pmd) & _SEGMENT_ENTRY_LARGE) && + (pmd_val(pmd) & _SEGMENT_ENTRY_SPLIT); +} + +static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp, pmd_t entry) +{ + *pmdp = entry; +} + +static inline pmd_t pmd_mkhuge(pmd_t pmd) +{ + pmd_val(pmd) |= _SEGMENT_ENTRY_LARGE; + pmd_val(pmd) |= _SEGMENT_ENTRY_YOUNG; + pmd_val(pmd) |= _SEGMENT_ENTRY_PROTECT; + return pmd; +} + +#define __HAVE_ARCH_PMDP_TEST_AND_CLEAR_YOUNG +static inline int pmdp_test_and_clear_young(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp) +{ + pmd_t pmd; + + pmd = *pmdp; + pmdp_flush_direct(vma->vm_mm, address, pmdp); + *pmdp = pmd_mkold(pmd); + return pmd_young(pmd); +} + +#define __HAVE_ARCH_PMDP_GET_AND_CLEAR +static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm, + unsigned long address, pmd_t *pmdp) +{ + pmd_t pmd = *pmdp; + + pmdp_flush_direct(mm, address, pmdp); + pmd_clear(pmdp); + return pmd; +} + +#define __HAVE_ARCH_PMDP_GET_AND_CLEAR_FULL +static inline pmd_t pmdp_get_and_clear_full(struct mm_struct *mm, + unsigned long address, + pmd_t *pmdp, int full) +{ + pmd_t pmd = *pmdp; + + if (!full) + pmdp_flush_lazy(mm, address, pmdp); + pmd_clear(pmdp); + return pmd; +} + +#define __HAVE_ARCH_PMDP_CLEAR_FLUSH +static inline pmd_t pmdp_clear_flush(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp) +{ + return pmdp_get_and_clear(vma->vm_mm, address, pmdp); +} + +#define __HAVE_ARCH_PMDP_INVALIDATE +static inline void pmdp_invalidate(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp) +{ + pmdp_flush_direct(vma->vm_mm, address, pmdp); +} + +#define __HAVE_ARCH_PMDP_SET_WRPROTECT +static inline void pmdp_set_wrprotect(struct mm_struct *mm, + unsigned long address, pmd_t *pmdp) +{ + pmd_t pmd = *pmdp; + + if (pmd_write(pmd)) { + pmdp_flush_direct(mm, address, pmdp); + set_pmd_at(mm, address, pmdp, pmd_wrprotect(pmd)); + } +} + +#define pfn_pmd(pfn, pgprot) mk_pmd_phys(__pa((pfn) << PAGE_SHIFT), (pgprot)) +#define mk_pmd(page, pgprot) pfn_pmd(page_to_pfn(page), (pgprot)) + +static inline int pmd_trans_huge(pmd_t pmd) +{ + return pmd_val(pmd) & _SEGMENT_ENTRY_LARGE; +} + +static inline int has_transparent_hugepage(void) +{ + return MACHINE_HAS_HPAGE ? 1 : 0; +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ + +/* + * 64 bit swap entry format: + * A page-table entry has some bits we have to treat in a special way. + * Bits 52 and bit 55 have to be zero, otherwise an specification + * exception will occur instead of a page translation exception. The + * specifiation exception has the bad habit not to store necessary + * information in the lowcore. + * Bits 54 and 63 are used to indicate the page type. + * A swap pte is indicated by bit pattern (pte & 0x201) == 0x200 + * This leaves the bits 0-51 and bits 56-62 to store type and offset. + * We use the 5 bits from 57-61 for the type and the 52 bits from 0-51 + * for the offset. + * | offset |01100|type |00| + * |0000000000111111111122222222223333333333444444444455|55555|55566|66| + * |0123456789012345678901234567890123456789012345678901|23456|78901|23| + */ + +#define __SWP_OFFSET_MASK ((1UL << 52) - 1) +#define __SWP_OFFSET_SHIFT 12 +#define __SWP_TYPE_MASK ((1UL << 5) - 1) +#define __SWP_TYPE_SHIFT 2 + +static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset) +{ + pte_t pte; + + pte_val(pte) = _PAGE_INVALID | _PAGE_PROTECT; + pte_val(pte) |= (offset & __SWP_OFFSET_MASK) << __SWP_OFFSET_SHIFT; + pte_val(pte) |= (type & __SWP_TYPE_MASK) << __SWP_TYPE_SHIFT; + return pte; +} + +static inline unsigned long __swp_type(swp_entry_t entry) +{ + return (entry.val >> __SWP_TYPE_SHIFT) & __SWP_TYPE_MASK; +} + +static inline unsigned long __swp_offset(swp_entry_t entry) +{ + return (entry.val >> __SWP_OFFSET_SHIFT) & __SWP_OFFSET_MASK; +} + +static inline swp_entry_t __swp_entry(unsigned long type, unsigned long offset) +{ + return (swp_entry_t) { pte_val(mk_swap_pte(type, offset)) }; +} + +#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) +#define __swp_entry_to_pte(x) ((pte_t) { (x).val }) + +#endif /* !__ASSEMBLY__ */ + +#define kern_addr_valid(addr) (1) + +extern int vmem_add_mapping(unsigned long start, unsigned long size); +extern int vmem_remove_mapping(unsigned long start, unsigned long size); +extern int s390_enable_sie(void); +extern int s390_enable_skey(void); +extern void s390_reset_cmma(struct mm_struct *mm); + +/* s390 has a private copy of get unmapped area to deal with cache synonyms */ +#define HAVE_ARCH_UNMAPPED_AREA +#define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN + +/* + * No page table caches to initialise + */ +static inline void pgtable_cache_init(void) { } +static inline void check_pgt_cache(void) { } + +#include <asm-generic/pgtable.h> + +#endif /* _S390_PAGE_H */ diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h new file mode 100644 index 000000000..dedb62185 --- /dev/null +++ b/arch/s390/include/asm/processor.h @@ -0,0 +1,360 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Hartmut Penner (hp@de.ibm.com), + * Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "include/asm-i386/processor.h" + * Copyright (C) 1994, Linus Torvalds + */ + +#ifndef __ASM_S390_PROCESSOR_H +#define __ASM_S390_PROCESSOR_H + +#define CIF_MCCK_PENDING 0 /* machine check handling is pending */ +#define CIF_ASCE 1 /* user asce needs fixup / uaccess */ +#define CIF_NOHZ_DELAY 2 /* delay HZ disable for a tick */ + +#define _CIF_MCCK_PENDING (1<<CIF_MCCK_PENDING) +#define _CIF_ASCE (1<<CIF_ASCE) +#define _CIF_NOHZ_DELAY (1<<CIF_NOHZ_DELAY) + +#ifndef __ASSEMBLY__ + +#include <linux/linkage.h> +#include <linux/irqflags.h> +#include <asm/cpu.h> +#include <asm/page.h> +#include <asm/ptrace.h> +#include <asm/setup.h> +#include <asm/runtime_instr.h> + +static inline void set_cpu_flag(int flag) +{ + S390_lowcore.cpu_flags |= (1U << flag); +} + +static inline void clear_cpu_flag(int flag) +{ + S390_lowcore.cpu_flags &= ~(1U << flag); +} + +static inline int test_cpu_flag(int flag) +{ + return !!(S390_lowcore.cpu_flags & (1U << flag)); +} + +#define arch_needs_cpu() test_cpu_flag(CIF_NOHZ_DELAY) + +/* + * Default implementation of macro that returns current + * instruction pointer ("program counter"). + */ +#define current_text_addr() ({ void *pc; asm("basr %0,0" : "=a" (pc)); pc; }) + +static inline void get_cpu_id(struct cpuid *ptr) +{ + asm volatile("stidp %0" : "=Q" (*ptr)); +} + +extern void s390_adjust_jiffies(void); +extern const struct seq_operations cpuinfo_op; +extern int sysctl_ieee_emulation_warnings; +extern void execve_tail(void); + +/* + * User space process size: 2GB for 31 bit, 4TB or 8PT for 64 bit. + */ + +#define TASK_SIZE_OF(tsk) ((tsk)->mm->context.asce_limit) +#define TASK_UNMAPPED_BASE (test_thread_flag(TIF_31BIT) ? \ + (1UL << 30) : (1UL << 41)) +#define TASK_SIZE TASK_SIZE_OF(current) +#define TASK_MAX_SIZE (1UL << 53) + +#define STACK_TOP (1UL << (test_thread_flag(TIF_31BIT) ? 31:42)) +#define STACK_TOP_MAX (1UL << 42) + +#define HAVE_ARCH_PICK_MMAP_LAYOUT + +typedef struct { + __u32 ar4; +} mm_segment_t; + +/* + * Thread structure + */ +struct thread_struct { + s390_fp_regs fp_regs; + unsigned int acrs[NUM_ACRS]; + unsigned long ksp; /* kernel stack pointer */ + mm_segment_t mm_segment; + unsigned long gmap_addr; /* address of last gmap fault. */ + unsigned int gmap_pfault; /* signal of a pending guest pfault */ + struct per_regs per_user; /* User specified PER registers */ + struct per_event per_event; /* Cause of the last PER trap */ + unsigned long per_flags; /* Flags to control debug behavior */ + /* pfault_wait is used to block the process on a pfault event */ + unsigned long pfault_wait; + struct list_head list; + /* cpu runtime instrumentation */ + struct runtime_instr_cb *ri_cb; + int ri_signum; + unsigned char trap_tdb[256]; /* Transaction abort diagnose block */ + __vector128 *vxrs; /* Vector register save area */ +}; + +/* Flag to disable transactions. */ +#define PER_FLAG_NO_TE 1UL +/* Flag to enable random transaction aborts. */ +#define PER_FLAG_TE_ABORT_RAND 2UL +/* Flag to specify random transaction abort mode: + * - abort each transaction at a random instruction before TEND if set. + * - abort random transactions at a random instruction if cleared. + */ +#define PER_FLAG_TE_ABORT_RAND_TEND 4UL + +typedef struct thread_struct thread_struct; + +/* + * Stack layout of a C stack frame. + */ +#ifndef __PACK_STACK +struct stack_frame { + unsigned long back_chain; + unsigned long empty1[5]; + unsigned long gprs[10]; + unsigned int empty2[8]; +}; +#else +struct stack_frame { + unsigned long empty1[5]; + unsigned int empty2[8]; + unsigned long gprs[10]; + unsigned long back_chain; +}; +#endif + +#define ARCH_MIN_TASKALIGN 8 + +#define INIT_THREAD { \ + .ksp = sizeof(init_stack) + (unsigned long) &init_stack, \ +} + +/* + * Do necessary setup to start up a new thread. + */ +#define start_thread(regs, new_psw, new_stackp) do { \ + regs->psw.mask = PSW_USER_BITS | PSW_MASK_EA | PSW_MASK_BA; \ + regs->psw.addr = new_psw | PSW_ADDR_AMODE; \ + regs->gprs[15] = new_stackp; \ + execve_tail(); \ +} while (0) + +#define start_thread31(regs, new_psw, new_stackp) do { \ + regs->psw.mask = PSW_USER_BITS | PSW_MASK_BA; \ + regs->psw.addr = new_psw | PSW_ADDR_AMODE; \ + regs->gprs[15] = new_stackp; \ + crst_table_downgrade(current->mm, 1UL << 31); \ + execve_tail(); \ +} while (0) + +/* Forward declaration, a strange C thing */ +struct task_struct; +struct mm_struct; +struct seq_file; + +void show_cacheinfo(struct seq_file *m); + +/* Free all resources held by a thread. */ +extern void release_thread(struct task_struct *); + +/* + * Return saved PC of a blocked thread. + */ +extern unsigned long thread_saved_pc(struct task_struct *t); + +unsigned long get_wchan(struct task_struct *p); +#define task_pt_regs(tsk) ((struct pt_regs *) \ + (task_stack_page(tsk) + THREAD_SIZE) - 1) +#define KSTK_EIP(tsk) (task_pt_regs(tsk)->psw.addr) +#define KSTK_ESP(tsk) (task_pt_regs(tsk)->gprs[15]) + +/* Has task runtime instrumentation enabled ? */ +#define is_ri_task(tsk) (!!(tsk)->thread.ri_cb) + +static inline unsigned short stap(void) +{ + unsigned short cpu_address; + + asm volatile("stap %0" : "=m" (cpu_address)); + return cpu_address; +} + +/* + * Give up the time slice of the virtual PU. + */ +void cpu_relax(void); + +#define cpu_relax_lowlatency() barrier() + +static inline void psw_set_key(unsigned int key) +{ + asm volatile("spka 0(%0)" : : "d" (key)); +} + +/* + * Set PSW to specified value. + */ +static inline void __load_psw(psw_t psw) +{ + asm volatile("lpswe %0" : : "Q" (psw) : "cc"); +} + +/* + * Set PSW mask to specified value, while leaving the + * PSW addr pointing to the next instruction. + */ +static inline void __load_psw_mask (unsigned long mask) +{ + unsigned long addr; + psw_t psw; + + psw.mask = mask; + + asm volatile( + " larl %0,1f\n" + " stg %0,%O1+8(%R1)\n" + " lpswe %1\n" + "1:" + : "=&d" (addr), "=Q" (psw) : "Q" (psw) : "memory", "cc"); +} + +/* + * Rewind PSW instruction address by specified number of bytes. + */ +static inline unsigned long __rewind_psw(psw_t psw, unsigned long ilc) +{ + unsigned long mask; + + mask = (psw.mask & PSW_MASK_EA) ? -1UL : + (psw.mask & PSW_MASK_BA) ? (1UL << 31) - 1 : + (1UL << 24) - 1; + return (psw.addr - ilc) & mask; +} + +/* + * Function to stop a processor until the next interrupt occurs + */ +void enabled_wait(void); + +/* + * Function to drop a processor into disabled wait state + */ +static inline void __noreturn disabled_wait(unsigned long code) +{ + unsigned long ctl_buf; + psw_t dw_psw; + + dw_psw.mask = PSW_MASK_BASE | PSW_MASK_WAIT | PSW_MASK_BA | PSW_MASK_EA; + dw_psw.addr = code; + /* + * Store status and then load disabled wait psw, + * the processor is dead afterwards + */ + asm volatile( + " stctg 0,0,0(%2)\n" + " ni 4(%2),0xef\n" /* switch off protection */ + " lctlg 0,0,0(%2)\n" + " lghi 1,0x1000\n" + " stpt 0x328(1)\n" /* store timer */ + " stckc 0x330(1)\n" /* store clock comparator */ + " stpx 0x318(1)\n" /* store prefix register */ + " stam 0,15,0x340(1)\n"/* store access registers */ + " stfpc 0x31c(1)\n" /* store fpu control */ + " std 0,0x200(1)\n" /* store f0 */ + " std 1,0x208(1)\n" /* store f1 */ + " std 2,0x210(1)\n" /* store f2 */ + " std 3,0x218(1)\n" /* store f3 */ + " std 4,0x220(1)\n" /* store f4 */ + " std 5,0x228(1)\n" /* store f5 */ + " std 6,0x230(1)\n" /* store f6 */ + " std 7,0x238(1)\n" /* store f7 */ + " std 8,0x240(1)\n" /* store f8 */ + " std 9,0x248(1)\n" /* store f9 */ + " std 10,0x250(1)\n" /* store f10 */ + " std 11,0x258(1)\n" /* store f11 */ + " std 12,0x260(1)\n" /* store f12 */ + " std 13,0x268(1)\n" /* store f13 */ + " std 14,0x270(1)\n" /* store f14 */ + " std 15,0x278(1)\n" /* store f15 */ + " stmg 0,15,0x280(1)\n"/* store general registers */ + " stctg 0,15,0x380(1)\n"/* store control registers */ + " oi 0x384(1),0x10\n"/* fake protection bit */ + " lpswe 0(%1)" + : "=m" (ctl_buf) + : "a" (&dw_psw), "a" (&ctl_buf), "m" (dw_psw) : "cc", "0", "1"); + while (1); +} + +/* + * Use to set psw mask except for the first byte which + * won't be changed by this function. + */ +static inline void +__set_psw_mask(unsigned long mask) +{ + __load_psw_mask(mask | (arch_local_save_flags() & ~(-1UL >> 8))); +} + +#define local_mcck_enable() \ + __set_psw_mask(PSW_KERNEL_BITS | PSW_MASK_DAT | PSW_MASK_MCHECK) +#define local_mcck_disable() \ + __set_psw_mask(PSW_KERNEL_BITS | PSW_MASK_DAT) + +/* + * Basic Machine Check/Program Check Handler. + */ + +extern void s390_base_mcck_handler(void); +extern void s390_base_pgm_handler(void); +extern void s390_base_ext_handler(void); + +extern void (*s390_base_mcck_handler_fn)(void); +extern void (*s390_base_pgm_handler_fn)(void); +extern void (*s390_base_ext_handler_fn)(void); + +#define ARCH_LOW_ADDRESS_LIMIT 0x7fffffffUL + +extern int memcpy_real(void *, void *, size_t); +extern void memcpy_absolute(void *, void *, size_t); + +#define mem_assign_absolute(dest, val) { \ + __typeof__(dest) __tmp = (val); \ + \ + BUILD_BUG_ON(sizeof(__tmp) != sizeof(val)); \ + memcpy_absolute(&(dest), &__tmp, sizeof(__tmp)); \ +} + +/* + * Helper macro for exception table entries + */ +#define EX_TABLE(_fault, _target) \ + ".section __ex_table,\"a\"\n" \ + ".align 4\n" \ + ".long (" #_fault ") - .\n" \ + ".long (" #_target ") - .\n" \ + ".previous\n" + +#else /* __ASSEMBLY__ */ + +#define EX_TABLE(_fault, _target) \ + .section __ex_table,"a" ; \ + .align 4 ; \ + .long (_fault) - . ; \ + .long (_target) - . ; \ + .previous + +#endif /* __ASSEMBLY__ */ + +#endif /* __ASM_S390_PROCESSOR_H */ diff --git a/arch/s390/include/asm/ptrace.h b/arch/s390/include/asm/ptrace.h new file mode 100644 index 000000000..6feda2599 --- /dev/null +++ b/arch/s390/include/asm/ptrace.h @@ -0,0 +1,177 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + * Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com) + */ +#ifndef _S390_PTRACE_H +#define _S390_PTRACE_H + +#include <uapi/asm/ptrace.h> + +#define PIF_SYSCALL 0 /* inside a system call */ +#define PIF_PER_TRAP 1 /* deliver sigtrap on return to user */ + +#define _PIF_SYSCALL (1<<PIF_SYSCALL) +#define _PIF_PER_TRAP (1<<PIF_PER_TRAP) + +#ifndef __ASSEMBLY__ + +#define PSW_KERNEL_BITS (PSW_DEFAULT_KEY | PSW_MASK_BASE | PSW_ASC_HOME | \ + PSW_MASK_EA | PSW_MASK_BA) +#define PSW_USER_BITS (PSW_MASK_DAT | PSW_MASK_IO | PSW_MASK_EXT | \ + PSW_DEFAULT_KEY | PSW_MASK_BASE | PSW_MASK_MCHECK | \ + PSW_MASK_PSTATE | PSW_ASC_PRIMARY) + +struct psw_bits { + unsigned long long : 1; + unsigned long long r : 1; /* PER-Mask */ + unsigned long long : 3; + unsigned long long t : 1; /* DAT Mode */ + unsigned long long i : 1; /* Input/Output Mask */ + unsigned long long e : 1; /* External Mask */ + unsigned long long key : 4; /* PSW Key */ + unsigned long long : 1; + unsigned long long m : 1; /* Machine-Check Mask */ + unsigned long long w : 1; /* Wait State */ + unsigned long long p : 1; /* Problem State */ + unsigned long long as : 2; /* Address Space Control */ + unsigned long long cc : 2; /* Condition Code */ + unsigned long long pm : 4; /* Program Mask */ + unsigned long long ri : 1; /* Runtime Instrumentation */ + unsigned long long : 6; + unsigned long long eaba : 2; /* Addressing Mode */ + unsigned long long : 31; + unsigned long long ia : 64;/* Instruction Address */ +}; + +enum { + PSW_AMODE_24BIT = 0, + PSW_AMODE_31BIT = 1, + PSW_AMODE_64BIT = 3 +}; + +enum { + PSW_AS_PRIMARY = 0, + PSW_AS_ACCREG = 1, + PSW_AS_SECONDARY = 2, + PSW_AS_HOME = 3 +}; + +#define psw_bits(__psw) (*({ \ + typecheck(psw_t, __psw); \ + &(*(struct psw_bits *)(&(__psw))); \ +})) + +/* + * The pt_regs struct defines the way the registers are stored on + * the stack during a system call. + */ +struct pt_regs +{ + unsigned long args[1]; + psw_t psw; + unsigned long gprs[NUM_GPRS]; + unsigned long orig_gpr2; + unsigned int int_code; + unsigned int int_parm; + unsigned long int_parm_long; + unsigned long flags; +}; + +/* + * Program event recording (PER) register set. + */ +struct per_regs { + unsigned long control; /* PER control bits */ + unsigned long start; /* PER starting address */ + unsigned long end; /* PER ending address */ +}; + +/* + * PER event contains information about the cause of the last PER exception. + */ +struct per_event { + unsigned short cause; /* PER code, ATMID and AI */ + unsigned long address; /* PER address */ + unsigned char paid; /* PER access identification */ +}; + +/* + * Simplified per_info structure used to decode the ptrace user space ABI. + */ +struct per_struct_kernel { + unsigned long cr9; /* PER control bits */ + unsigned long cr10; /* PER starting address */ + unsigned long cr11; /* PER ending address */ + unsigned long bits; /* Obsolete software bits */ + unsigned long starting_addr; /* User specified start address */ + unsigned long ending_addr; /* User specified end address */ + unsigned short perc_atmid; /* PER trap ATMID */ + unsigned long address; /* PER trap instruction address */ + unsigned char access_id; /* PER trap access identification */ +}; + +#define PER_EVENT_MASK 0xEB000000UL + +#define PER_EVENT_BRANCH 0x80000000UL +#define PER_EVENT_IFETCH 0x40000000UL +#define PER_EVENT_STORE 0x20000000UL +#define PER_EVENT_STORE_REAL 0x08000000UL +#define PER_EVENT_TRANSACTION_END 0x02000000UL +#define PER_EVENT_NULLIFICATION 0x01000000UL + +#define PER_CONTROL_MASK 0x00e00000UL + +#define PER_CONTROL_BRANCH_ADDRESS 0x00800000UL +#define PER_CONTROL_SUSPENSION 0x00400000UL +#define PER_CONTROL_ALTERATION 0x00200000UL + +static inline void set_pt_regs_flag(struct pt_regs *regs, int flag) +{ + regs->flags |= (1U << flag); +} + +static inline void clear_pt_regs_flag(struct pt_regs *regs, int flag) +{ + regs->flags &= ~(1U << flag); +} + +static inline int test_pt_regs_flag(struct pt_regs *regs, int flag) +{ + return !!(regs->flags & (1U << flag)); +} + +/* + * These are defined as per linux/ptrace.h, which see. + */ +#define arch_has_single_step() (1) +#define arch_has_block_step() (1) + +#define user_mode(regs) (((regs)->psw.mask & PSW_MASK_PSTATE) != 0) +#define instruction_pointer(regs) ((regs)->psw.addr & PSW_ADDR_INSN) +#define user_stack_pointer(regs)((regs)->gprs[15]) +#define profile_pc(regs) instruction_pointer(regs) + +static inline long regs_return_value(struct pt_regs *regs) +{ + return regs->gprs[2]; +} + +static inline void instruction_pointer_set(struct pt_regs *regs, + unsigned long val) +{ + regs->psw.addr = val | PSW_ADDR_AMODE; +} + +int regs_query_register_offset(const char *name); +const char *regs_query_register_name(unsigned int offset); +unsigned long regs_get_register(struct pt_regs *regs, unsigned int offset); +unsigned long regs_get_kernel_stack_nth(struct pt_regs *regs, unsigned int n); + +static inline unsigned long kernel_stack_pointer(struct pt_regs *regs) +{ + return regs->gprs[15] & PSW_ADDR_INSN; +} + +#endif /* __ASSEMBLY__ */ +#endif /* _S390_PTRACE_H */ diff --git a/arch/s390/include/asm/qdio.h b/arch/s390/include/asm/qdio.h new file mode 100644 index 000000000..998b61cd0 --- /dev/null +++ b/arch/s390/include/asm/qdio.h @@ -0,0 +1,430 @@ +/* + * Copyright IBM Corp. 2000, 2008 + * Author(s): Utz Bacher <utz.bacher@de.ibm.com> + * Jan Glauber <jang@linux.vnet.ibm.com> + * + */ +#ifndef __QDIO_H__ +#define __QDIO_H__ + +#include <linux/interrupt.h> +#include <asm/cio.h> +#include <asm/ccwdev.h> + +/* only use 4 queues to save some cachelines */ +#define QDIO_MAX_QUEUES_PER_IRQ 4 +#define QDIO_MAX_BUFFERS_PER_Q 128 +#define QDIO_MAX_BUFFERS_MASK (QDIO_MAX_BUFFERS_PER_Q - 1) +#define QDIO_MAX_ELEMENTS_PER_BUFFER 16 +#define QDIO_SBAL_SIZE 256 + +#define QDIO_QETH_QFMT 0 +#define QDIO_ZFCP_QFMT 1 +#define QDIO_IQDIO_QFMT 2 + +/** + * struct qdesfmt0 - queue descriptor, format 0 + * @sliba: storage list information block address + * @sla: storage list address + * @slsba: storage list state block address + * @akey: access key for DLIB + * @bkey: access key for SL + * @ckey: access key for SBALs + * @dkey: access key for SLSB + */ +struct qdesfmt0 { + u64 sliba; + u64 sla; + u64 slsba; + u32 : 32; + u32 akey : 4; + u32 bkey : 4; + u32 ckey : 4; + u32 dkey : 4; + u32 : 16; +} __attribute__ ((packed)); + +#define QDR_AC_MULTI_BUFFER_ENABLE 0x01 + +/** + * struct qdr - queue description record (QDR) + * @qfmt: queue format + * @pfmt: implementation dependent parameter format + * @ac: adapter characteristics + * @iqdcnt: input queue descriptor count + * @oqdcnt: output queue descriptor count + * @iqdsz: inpout queue descriptor size + * @oqdsz: output queue descriptor size + * @qiba: queue information block address + * @qkey: queue information block key + * @qdf0: queue descriptions + */ +struct qdr { + u32 qfmt : 8; + u32 pfmt : 8; + u32 : 8; + u32 ac : 8; + u32 : 8; + u32 iqdcnt : 8; + u32 : 8; + u32 oqdcnt : 8; + u32 : 8; + u32 iqdsz : 8; + u32 : 8; + u32 oqdsz : 8; + /* private: */ + u32 res[9]; + /* public: */ + u64 qiba; + u32 : 32; + u32 qkey : 4; + u32 : 28; + struct qdesfmt0 qdf0[126]; +} __attribute__ ((packed, aligned(4096))); + +#define QIB_AC_OUTBOUND_PCI_SUPPORTED 0x40 +#define QIB_RFLAGS_ENABLE_QEBSM 0x80 +#define QIB_RFLAGS_ENABLE_DATA_DIV 0x02 + +/** + * struct qib - queue information block (QIB) + * @qfmt: queue format + * @pfmt: implementation dependent parameter format + * @rflags: QEBSM + * @ac: adapter characteristics + * @isliba: absolute address of first input SLIB + * @osliba: absolute address of first output SLIB + * @ebcnam: adapter identifier in EBCDIC + * @parm: implementation dependent parameters + */ +struct qib { + u32 qfmt : 8; + u32 pfmt : 8; + u32 rflags : 8; + u32 ac : 8; + u32 : 32; + u64 isliba; + u64 osliba; + u32 : 32; + u32 : 32; + u8 ebcnam[8]; + /* private: */ + u8 res[88]; + /* public: */ + u8 parm[QDIO_MAX_BUFFERS_PER_Q]; +} __attribute__ ((packed, aligned(256))); + +/** + * struct slibe - storage list information block element (SLIBE) + * @parms: implementation dependent parameters + */ +struct slibe { + u64 parms; +}; + +/** + * struct qaob - queue asynchronous operation block + * @res0: reserved parameters + * @res1: reserved parameter + * @res2: reserved parameter + * @res3: reserved parameter + * @aorc: asynchronous operation return code + * @flags: internal flags + * @cbtbs: control block type + * @sb_count: number of storage blocks + * @sba: storage block element addresses + * @dcount: size of storage block elements + * @user0: user defineable value + * @res4: reserved paramater + * @user1: user defineable value + * @user2: user defineable value + */ +struct qaob { + u64 res0[6]; + u8 res1; + u8 res2; + u8 res3; + u8 aorc; + u8 flags; + u16 cbtbs; + u8 sb_count; + u64 sba[QDIO_MAX_ELEMENTS_PER_BUFFER]; + u16 dcount[QDIO_MAX_ELEMENTS_PER_BUFFER]; + u64 user0; + u64 res4[2]; + u64 user1; + u64 user2; +} __attribute__ ((packed, aligned(256))); + +/** + * struct slib - storage list information block (SLIB) + * @nsliba: next SLIB address (if any) + * @sla: SL address + * @slsba: SLSB address + * @slibe: SLIB elements + */ +struct slib { + u64 nsliba; + u64 sla; + u64 slsba; + /* private: */ + u8 res[1000]; + /* public: */ + struct slibe slibe[QDIO_MAX_BUFFERS_PER_Q]; +} __attribute__ ((packed, aligned(2048))); + +#define SBAL_EFLAGS_LAST_ENTRY 0x40 +#define SBAL_EFLAGS_CONTIGUOUS 0x20 +#define SBAL_EFLAGS_FIRST_FRAG 0x04 +#define SBAL_EFLAGS_MIDDLE_FRAG 0x08 +#define SBAL_EFLAGS_LAST_FRAG 0x0c +#define SBAL_EFLAGS_MASK 0x6f + +#define SBAL_SFLAGS0_PCI_REQ 0x40 +#define SBAL_SFLAGS0_DATA_CONTINUATION 0x20 + +/* Awesome OpenFCP extensions */ +#define SBAL_SFLAGS0_TYPE_STATUS 0x00 +#define SBAL_SFLAGS0_TYPE_WRITE 0x08 +#define SBAL_SFLAGS0_TYPE_READ 0x10 +#define SBAL_SFLAGS0_TYPE_WRITE_READ 0x18 +#define SBAL_SFLAGS0_MORE_SBALS 0x04 +#define SBAL_SFLAGS0_COMMAND 0x02 +#define SBAL_SFLAGS0_LAST_SBAL 0x00 +#define SBAL_SFLAGS0_ONLY_SBAL SBAL_SFLAGS0_COMMAND +#define SBAL_SFLAGS0_MIDDLE_SBAL SBAL_SFLAGS0_MORE_SBALS +#define SBAL_SFLAGS0_FIRST_SBAL (SBAL_SFLAGS0_MORE_SBALS | SBAL_SFLAGS0_COMMAND) + +/** + * struct qdio_buffer_element - SBAL entry + * @eflags: SBAL entry flags + * @scount: SBAL count + * @sflags: whole SBAL flags + * @length: length + * @addr: address +*/ +struct qdio_buffer_element { + u8 eflags; + /* private: */ + u8 res1; + /* public: */ + u8 scount; + u8 sflags; + u32 length; + void *addr; +} __attribute__ ((packed, aligned(16))); + +/** + * struct qdio_buffer - storage block address list (SBAL) + * @element: SBAL entries + */ +struct qdio_buffer { + struct qdio_buffer_element element[QDIO_MAX_ELEMENTS_PER_BUFFER]; +} __attribute__ ((packed, aligned(256))); + +/** + * struct sl_element - storage list entry + * @sbal: absolute SBAL address + */ +struct sl_element { + unsigned long sbal; +} __attribute__ ((packed)); + +/** + * struct sl - storage list (SL) + * @element: SL entries + */ +struct sl { + struct sl_element element[QDIO_MAX_BUFFERS_PER_Q]; +} __attribute__ ((packed, aligned(1024))); + +/** + * struct slsb - storage list state block (SLSB) + * @val: state per buffer + */ +struct slsb { + u8 val[QDIO_MAX_BUFFERS_PER_Q]; +} __attribute__ ((packed, aligned(256))); + +/** + * struct qdio_outbuf_state - SBAL related asynchronous operation information + * (for communication with upper layer programs) + * (only required for use with completion queues) + * @flags: flags indicating state of buffer + * @aob: pointer to QAOB used for the particular SBAL + * @user: pointer to upper layer program's state information related to SBAL + * (stored in user1 data of QAOB) + */ +struct qdio_outbuf_state { + u8 flags; + struct qaob *aob; + void *user; +}; + +#define QDIO_OUTBUF_STATE_FLAG_NONE 0x00 +#define QDIO_OUTBUF_STATE_FLAG_PENDING 0x01 + +#define CHSC_AC1_INITIATE_INPUTQ 0x80 + + +/* qdio adapter-characteristics-1 flag */ +#define AC1_SIGA_INPUT_NEEDED 0x40 /* process input queues */ +#define AC1_SIGA_OUTPUT_NEEDED 0x20 /* process output queues */ +#define AC1_SIGA_SYNC_NEEDED 0x10 /* ask hypervisor to sync */ +#define AC1_AUTOMATIC_SYNC_ON_THININT 0x08 /* set by hypervisor */ +#define AC1_AUTOMATIC_SYNC_ON_OUT_PCI 0x04 /* set by hypervisor */ +#define AC1_SC_QEBSM_AVAILABLE 0x02 /* available for subchannel */ +#define AC1_SC_QEBSM_ENABLED 0x01 /* enabled for subchannel */ + +#define CHSC_AC2_MULTI_BUFFER_AVAILABLE 0x0080 +#define CHSC_AC2_MULTI_BUFFER_ENABLED 0x0040 +#define CHSC_AC2_DATA_DIV_AVAILABLE 0x0010 +#define CHSC_AC2_DATA_DIV_ENABLED 0x0002 + +#define CHSC_AC3_FORMAT2_CQ_AVAILABLE 0x8000 + +struct qdio_ssqd_desc { + u8 flags; + u8:8; + u16 sch; + u8 qfmt; + u8 parm; + u8 qdioac1; + u8 sch_class; + u8 pcnt; + u8 icnt; + u8:8; + u8 ocnt; + u8:8; + u8 mbccnt; + u16 qdioac2; + u64 sch_token; + u8 mro; + u8 mri; + u16 qdioac3; + u16:16; + u8:8; + u8 mmwc; +} __attribute__ ((packed)); + +/* params are: ccw_device, qdio_error, queue_number, + first element processed, number of elements processed, int_parm */ +typedef void qdio_handler_t(struct ccw_device *, unsigned int, int, + int, int, unsigned long); + +/* qdio errors reported to the upper-layer program */ +#define QDIO_ERROR_ACTIVATE 0x0001 +#define QDIO_ERROR_GET_BUF_STATE 0x0002 +#define QDIO_ERROR_SET_BUF_STATE 0x0004 +#define QDIO_ERROR_SLSB_STATE 0x0100 + +#define QDIO_ERROR_FATAL 0x00ff +#define QDIO_ERROR_TEMPORARY 0xff00 + +/* for qdio_cleanup */ +#define QDIO_FLAG_CLEANUP_USING_CLEAR 0x01 +#define QDIO_FLAG_CLEANUP_USING_HALT 0x02 + +/** + * struct qdio_initialize - qdio initialization data + * @cdev: associated ccw device + * @q_format: queue format + * @adapter_name: name for the adapter + * @qib_param_field_format: format for qib_parm_field + * @qib_param_field: pointer to 128 bytes or NULL, if no param field + * @qib_rflags: rflags to set + * @input_slib_elements: pointer to no_input_qs * 128 words of data or NULL + * @output_slib_elements: pointer to no_output_qs * 128 words of data or NULL + * @no_input_qs: number of input queues + * @no_output_qs: number of output queues + * @input_handler: handler to be called for input queues + * @output_handler: handler to be called for output queues + * @queue_start_poll_array: polling handlers (one per input queue or NULL) + * @int_parm: interruption parameter + * @input_sbal_addr_array: address of no_input_qs * 128 pointers + * @output_sbal_addr_array: address of no_output_qs * 128 pointers + * @output_sbal_state_array: no_output_qs * 128 state info (for CQ or NULL) + */ +struct qdio_initialize { + struct ccw_device *cdev; + unsigned char q_format; + unsigned char qdr_ac; + unsigned char adapter_name[8]; + unsigned int qib_param_field_format; + unsigned char *qib_param_field; + unsigned char qib_rflags; + unsigned long *input_slib_elements; + unsigned long *output_slib_elements; + unsigned int no_input_qs; + unsigned int no_output_qs; + qdio_handler_t *input_handler; + qdio_handler_t *output_handler; + void (**queue_start_poll_array) (struct ccw_device *, int, + unsigned long); + int scan_threshold; + unsigned long int_parm; + void **input_sbal_addr_array; + void **output_sbal_addr_array; + struct qdio_outbuf_state *output_sbal_state_array; +}; + +/** + * enum qdio_brinfo_entry_type - type of address entry for qdio_brinfo_desc() + * @l3_ipv6_addr: entry contains IPv6 address + * @l3_ipv4_addr: entry contains IPv4 address + * @l2_addr_lnid: entry contains MAC address and VLAN ID + */ +enum qdio_brinfo_entry_type {l3_ipv6_addr, l3_ipv4_addr, l2_addr_lnid}; + +/** + * struct qdio_brinfo_entry_XXX - Address entry for qdio_brinfo_desc() + * @nit: Network interface token + * @addr: Address of one of the three types + * + * The struct is passed to the callback function by qdio_brinfo_desc() + */ +struct qdio_brinfo_entry_l3_ipv6 { + u64 nit; + struct { unsigned char _s6_addr[16]; } addr; +} __packed; +struct qdio_brinfo_entry_l3_ipv4 { + u64 nit; + struct { uint32_t _s_addr; } addr; +} __packed; +struct qdio_brinfo_entry_l2 { + u64 nit; + struct { u8 mac[6]; u16 lnid; } addr_lnid; +} __packed; + +#define QDIO_STATE_INACTIVE 0x00000002 /* after qdio_cleanup */ +#define QDIO_STATE_ESTABLISHED 0x00000004 /* after qdio_establish */ +#define QDIO_STATE_ACTIVE 0x00000008 /* after qdio_activate */ +#define QDIO_STATE_STOPPED 0x00000010 /* after queues went down */ + +#define QDIO_FLAG_SYNC_INPUT 0x01 +#define QDIO_FLAG_SYNC_OUTPUT 0x02 +#define QDIO_FLAG_PCI_OUT 0x10 + +int qdio_alloc_buffers(struct qdio_buffer **buf, unsigned int count); +void qdio_free_buffers(struct qdio_buffer **buf, unsigned int count); +void qdio_reset_buffers(struct qdio_buffer **buf, unsigned int count); + +extern int qdio_allocate(struct qdio_initialize *); +extern int qdio_establish(struct qdio_initialize *); +extern int qdio_activate(struct ccw_device *); +extern void qdio_release_aob(struct qaob *); +extern int do_QDIO(struct ccw_device *, unsigned int, int, unsigned int, + unsigned int); +extern int qdio_start_irq(struct ccw_device *, int); +extern int qdio_stop_irq(struct ccw_device *, int); +extern int qdio_get_next_buffers(struct ccw_device *, int, int *, int *); +extern int qdio_shutdown(struct ccw_device *, int); +extern int qdio_free(struct ccw_device *); +extern int qdio_get_ssqd_desc(struct ccw_device *, struct qdio_ssqd_desc *); +extern int qdio_pnso_brinfo(struct subchannel_id schid, + int cnc, u16 *response, + void (*cb)(void *priv, enum qdio_brinfo_entry_type type, + void *entry), + void *priv); + +#endif /* __QDIO_H__ */ diff --git a/arch/s390/include/asm/reset.h b/arch/s390/include/asm/reset.h new file mode 100644 index 000000000..72786067b --- /dev/null +++ b/arch/s390/include/asm/reset.h @@ -0,0 +1,20 @@ +/* + * Copyright IBM Corp. 2006 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#ifndef _ASM_S390_RESET_H +#define _ASM_S390_RESET_H + +#include <linux/list.h> + +struct reset_call { + struct list_head list; + void (*fn)(void); +}; + +extern void register_reset_call(struct reset_call *reset); +extern void unregister_reset_call(struct reset_call *reset); +extern void s390_reset_system(void (*fn_pre)(void), + void (*fn_post)(void *), void *data); +#endif /* _ASM_S390_RESET_H */ diff --git a/arch/s390/include/asm/runtime_instr.h b/arch/s390/include/asm/runtime_instr.h new file mode 100644 index 000000000..402ad6df4 --- /dev/null +++ b/arch/s390/include/asm/runtime_instr.h @@ -0,0 +1,90 @@ +#ifndef _RUNTIME_INSTR_H +#define _RUNTIME_INSTR_H + +#define S390_RUNTIME_INSTR_START 0x1 +#define S390_RUNTIME_INSTR_STOP 0x2 + +struct runtime_instr_cb { + __u64 buf_current; + __u64 buf_origin; + __u64 buf_limit; + + __u32 valid : 1; + __u32 pstate : 1; + __u32 pstate_set_buf : 1; + __u32 home_space : 1; + __u32 altered : 1; + __u32 : 3; + __u32 pstate_sample : 1; + __u32 sstate_sample : 1; + __u32 pstate_collect : 1; + __u32 sstate_collect : 1; + __u32 : 1; + __u32 halted_int : 1; + __u32 int_requested : 1; + __u32 buffer_full_int : 1; + __u32 key : 4; + __u32 : 9; + __u32 rgs : 3; + + __u32 mode : 4; + __u32 next : 1; + __u32 mae : 1; + __u32 : 2; + __u32 call_type_br : 1; + __u32 return_type_br : 1; + __u32 other_type_br : 1; + __u32 bc_other_type : 1; + __u32 emit : 1; + __u32 tx_abort : 1; + __u32 : 2; + __u32 bp_xn : 1; + __u32 bp_xt : 1; + __u32 bp_ti : 1; + __u32 bp_ni : 1; + __u32 suppr_y : 1; + __u32 suppr_z : 1; + + __u32 dc_miss_extra : 1; + __u32 lat_lev_ignore : 1; + __u32 ic_lat_lev : 4; + __u32 dc_lat_lev : 4; + + __u64 reserved1; + __u64 scaling_factor; + __u64 rsic; + __u64 reserved2; +} __packed __aligned(8); + +extern struct runtime_instr_cb runtime_instr_empty_cb; + +static inline void load_runtime_instr_cb(struct runtime_instr_cb *cb) +{ + asm volatile(".insn rsy,0xeb0000000060,0,0,%0" /* LRIC */ + : : "Q" (*cb)); +} + +static inline void store_runtime_instr_cb(struct runtime_instr_cb *cb) +{ + asm volatile(".insn rsy,0xeb0000000061,0,0,%0" /* STRIC */ + : "=Q" (*cb) : : "cc"); +} + +static inline void save_ri_cb(struct runtime_instr_cb *cb_prev) +{ + if (cb_prev) + store_runtime_instr_cb(cb_prev); +} + +static inline void restore_ri_cb(struct runtime_instr_cb *cb_next, + struct runtime_instr_cb *cb_prev) +{ + if (cb_next) + load_runtime_instr_cb(cb_next); + else if (cb_prev) + load_runtime_instr_cb(&runtime_instr_empty_cb); +} + +void exit_thread_runtime_instr(void); + +#endif /* _RUNTIME_INSTR_H */ diff --git a/arch/s390/include/asm/rwsem.h b/arch/s390/include/asm/rwsem.h new file mode 100644 index 000000000..4b43ee7e6 --- /dev/null +++ b/arch/s390/include/asm/rwsem.h @@ -0,0 +1,237 @@ +#ifndef _S390_RWSEM_H +#define _S390_RWSEM_H + +/* + * S390 version + * Copyright IBM Corp. 2002 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Based on asm-alpha/semaphore.h and asm-i386/rwsem.h + */ + +/* + * + * The MSW of the count is the negated number of active writers and waiting + * lockers, and the LSW is the total number of active locks + * + * The lock count is initialized to 0 (no active and no waiting lockers). + * + * When a writer subtracts WRITE_BIAS, it'll get 0xffff0001 for the case of an + * uncontended lock. This can be determined because XADD returns the old value. + * Readers increment by 1 and see a positive value when uncontended, negative + * if there are writers (and maybe) readers waiting (in which case it goes to + * sleep). + * + * The value of WAITING_BIAS supports up to 32766 waiting processes. This can + * be extended to 65534 by manually checking the whole MSW rather than relying + * on the S flag. + * + * The value of ACTIVE_BIAS supports up to 65535 active processes. + * + * This should be totally fair - if anything is waiting, a process that wants a + * lock will go to the back of the queue. When the currently active lock is + * released, if there's a writer at the front of the queue, then that and only + * that will be woken up; if there's a bunch of consequtive readers at the + * front, then they'll all be woken up, but no other readers will be. + */ + +#ifndef _LINUX_RWSEM_H +#error "please don't include asm/rwsem.h directly, use linux/rwsem.h instead" +#endif + +#define RWSEM_UNLOCKED_VALUE 0x0000000000000000L +#define RWSEM_ACTIVE_BIAS 0x0000000000000001L +#define RWSEM_ACTIVE_MASK 0x00000000ffffffffL +#define RWSEM_WAITING_BIAS (-0x0000000100000000L) +#define RWSEM_ACTIVE_READ_BIAS RWSEM_ACTIVE_BIAS +#define RWSEM_ACTIVE_WRITE_BIAS (RWSEM_WAITING_BIAS + RWSEM_ACTIVE_BIAS) + +/* + * lock for reading + */ +static inline void __down_read(struct rw_semaphore *sem) +{ + signed long old, new; + + asm volatile( + " lg %0,%2\n" + "0: lgr %1,%0\n" + " aghi %1,%4\n" + " csg %0,%1,%2\n" + " jl 0b" + : "=&d" (old), "=&d" (new), "=Q" (sem->count) + : "Q" (sem->count), "i" (RWSEM_ACTIVE_READ_BIAS) + : "cc", "memory"); + if (old < 0) + rwsem_down_read_failed(sem); +} + +/* + * trylock for reading -- returns 1 if successful, 0 if contention + */ +static inline int __down_read_trylock(struct rw_semaphore *sem) +{ + signed long old, new; + + asm volatile( + " lg %0,%2\n" + "0: ltgr %1,%0\n" + " jm 1f\n" + " aghi %1,%4\n" + " csg %0,%1,%2\n" + " jl 0b\n" + "1:" + : "=&d" (old), "=&d" (new), "=Q" (sem->count) + : "Q" (sem->count), "i" (RWSEM_ACTIVE_READ_BIAS) + : "cc", "memory"); + return old >= 0 ? 1 : 0; +} + +/* + * lock for writing + */ +static inline void __down_write_nested(struct rw_semaphore *sem, int subclass) +{ + signed long old, new, tmp; + + tmp = RWSEM_ACTIVE_WRITE_BIAS; + asm volatile( + " lg %0,%2\n" + "0: lgr %1,%0\n" + " ag %1,%4\n" + " csg %0,%1,%2\n" + " jl 0b" + : "=&d" (old), "=&d" (new), "=Q" (sem->count) + : "Q" (sem->count), "m" (tmp) + : "cc", "memory"); + if (old != 0) + rwsem_down_write_failed(sem); +} + +static inline void __down_write(struct rw_semaphore *sem) +{ + __down_write_nested(sem, 0); +} + +/* + * trylock for writing -- returns 1 if successful, 0 if contention + */ +static inline int __down_write_trylock(struct rw_semaphore *sem) +{ + signed long old; + + asm volatile( + " lg %0,%1\n" + "0: ltgr %0,%0\n" + " jnz 1f\n" + " csg %0,%3,%1\n" + " jl 0b\n" + "1:" + : "=&d" (old), "=Q" (sem->count) + : "Q" (sem->count), "d" (RWSEM_ACTIVE_WRITE_BIAS) + : "cc", "memory"); + return (old == RWSEM_UNLOCKED_VALUE) ? 1 : 0; +} + +/* + * unlock after reading + */ +static inline void __up_read(struct rw_semaphore *sem) +{ + signed long old, new; + + asm volatile( + " lg %0,%2\n" + "0: lgr %1,%0\n" + " aghi %1,%4\n" + " csg %0,%1,%2\n" + " jl 0b" + : "=&d" (old), "=&d" (new), "=Q" (sem->count) + : "Q" (sem->count), "i" (-RWSEM_ACTIVE_READ_BIAS) + : "cc", "memory"); + if (new < 0) + if ((new & RWSEM_ACTIVE_MASK) == 0) + rwsem_wake(sem); +} + +/* + * unlock after writing + */ +static inline void __up_write(struct rw_semaphore *sem) +{ + signed long old, new, tmp; + + tmp = -RWSEM_ACTIVE_WRITE_BIAS; + asm volatile( + " lg %0,%2\n" + "0: lgr %1,%0\n" + " ag %1,%4\n" + " csg %0,%1,%2\n" + " jl 0b" + : "=&d" (old), "=&d" (new), "=Q" (sem->count) + : "Q" (sem->count), "m" (tmp) + : "cc", "memory"); + if (new < 0) + if ((new & RWSEM_ACTIVE_MASK) == 0) + rwsem_wake(sem); +} + +/* + * downgrade write lock to read lock + */ +static inline void __downgrade_write(struct rw_semaphore *sem) +{ + signed long old, new, tmp; + + tmp = -RWSEM_WAITING_BIAS; + asm volatile( + " lg %0,%2\n" + "0: lgr %1,%0\n" + " ag %1,%4\n" + " csg %0,%1,%2\n" + " jl 0b" + : "=&d" (old), "=&d" (new), "=Q" (sem->count) + : "Q" (sem->count), "m" (tmp) + : "cc", "memory"); + if (new > 1) + rwsem_downgrade_wake(sem); +} + +/* + * implement atomic add functionality + */ +static inline void rwsem_atomic_add(long delta, struct rw_semaphore *sem) +{ + signed long old, new; + + asm volatile( + " lg %0,%2\n" + "0: lgr %1,%0\n" + " agr %1,%4\n" + " csg %0,%1,%2\n" + " jl 0b" + : "=&d" (old), "=&d" (new), "=Q" (sem->count) + : "Q" (sem->count), "d" (delta) + : "cc", "memory"); +} + +/* + * implement exchange and add functionality + */ +static inline long rwsem_atomic_update(long delta, struct rw_semaphore *sem) +{ + signed long old, new; + + asm volatile( + " lg %0,%2\n" + "0: lgr %1,%0\n" + " agr %1,%4\n" + " csg %0,%1,%2\n" + " jl 0b" + : "=&d" (old), "=&d" (new), "=Q" (sem->count) + : "Q" (sem->count), "d" (delta) + : "cc", "memory"); + return new; +} + +#endif /* _S390_RWSEM_H */ diff --git a/arch/s390/include/asm/schid.h b/arch/s390/include/asm/schid.h new file mode 100644 index 000000000..40b47dfa9 --- /dev/null +++ b/arch/s390/include/asm/schid.h @@ -0,0 +1,21 @@ +#ifndef ASM_SCHID_H +#define ASM_SCHID_H + +#include <linux/string.h> +#include <uapi/asm/schid.h> + +/* Helper function for sane state of pre-allocated subchannel_id. */ +static inline void +init_subchannel_id(struct subchannel_id *schid) +{ + memset(schid, 0, sizeof(struct subchannel_id)); + schid->one = 1; +} + +static inline int +schid_equal(struct subchannel_id *schid1, struct subchannel_id *schid2) +{ + return !memcmp(schid1, schid2, sizeof(struct subchannel_id)); +} + +#endif /* ASM_SCHID_H */ diff --git a/arch/s390/include/asm/sclp.h b/arch/s390/include/asm/sclp.h new file mode 100644 index 000000000..f1096bab5 --- /dev/null +++ b/arch/s390/include/asm/sclp.h @@ -0,0 +1,78 @@ +/* + * Copyright IBM Corp. 2007 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#ifndef _ASM_S390_SCLP_H +#define _ASM_S390_SCLP_H + +#include <linux/types.h> +#include <asm/chpid.h> +#include <asm/cpu.h> + +#define SCLP_CHP_INFO_MASK_SIZE 32 + +struct sclp_chp_info { + u8 recognized[SCLP_CHP_INFO_MASK_SIZE]; + u8 standby[SCLP_CHP_INFO_MASK_SIZE]; + u8 configured[SCLP_CHP_INFO_MASK_SIZE]; +}; + +#define LOADPARM_LEN 8 + +struct sclp_ipl_info { + int is_valid; + int has_dump; + char loadparm[LOADPARM_LEN]; +}; + +struct sclp_cpu_entry { + u8 core_id; + u8 reserved0[2]; + u8 : 3; + u8 siif : 1; + u8 sigpif : 1; + u8 : 3; + u8 reserved2[10]; + u8 type; + u8 reserved1; +} __attribute__((packed)); + +struct sclp_cpu_info { + unsigned int configured; + unsigned int standby; + unsigned int combined; + int has_cpu_type; + struct sclp_cpu_entry cpu[MAX_CPU_ADDRESS + 1]; +}; + +int sclp_get_cpu_info(struct sclp_cpu_info *info); +int sclp_cpu_configure(u8 cpu); +int sclp_cpu_deconfigure(u8 cpu); +unsigned long long sclp_get_rnmax(void); +unsigned long long sclp_get_rzm(void); +unsigned int sclp_get_max_cpu(void); +unsigned int sclp_get_mtid(u8 cpu_type); +unsigned int sclp_get_mtid_max(void); +unsigned int sclp_get_mtid_prev(void); +int sclp_sdias_blk_count(void); +int sclp_sdias_copy(void *dest, int blk_num, int nr_blks); +int sclp_chp_configure(struct chp_id chpid); +int sclp_chp_deconfigure(struct chp_id chpid); +int sclp_chp_read_info(struct sclp_chp_info *info); +void sclp_get_ipl_info(struct sclp_ipl_info *info); +bool __init sclp_has_linemode(void); +bool __init sclp_has_vt220(void); +bool sclp_has_sprp(void); +int sclp_pci_configure(u32 fid); +int sclp_pci_deconfigure(u32 fid); +int memcpy_hsa(void *dest, unsigned long src, size_t count, int mode); +unsigned long sclp_get_hsa_size(void); +void sclp_early_detect(void); +int sclp_has_siif(void); +int sclp_has_sigpif(void); +unsigned int sclp_get_ibc(void); + +long _sclp_print_early(const char *); + +#endif /* _ASM_S390_SCLP_H */ diff --git a/arch/s390/include/asm/scsw.h b/arch/s390/include/asm/scsw.h new file mode 100644 index 000000000..4af99cdad --- /dev/null +++ b/arch/s390/include/asm/scsw.h @@ -0,0 +1,988 @@ +/* + * Helper functions for scsw access. + * + * Copyright IBM Corp. 2008, 2012 + * Author(s): Peter Oberparleiter <peter.oberparleiter@de.ibm.com> + */ + +#ifndef _ASM_S390_SCSW_H_ +#define _ASM_S390_SCSW_H_ + +#include <linux/types.h> +#include <asm/css_chars.h> +#include <asm/cio.h> + +/** + * struct cmd_scsw - command-mode subchannel status word + * @key: subchannel key + * @sctl: suspend control + * @eswf: esw format + * @cc: deferred condition code + * @fmt: format + * @pfch: prefetch + * @isic: initial-status interruption control + * @alcc: address-limit checking control + * @ssi: suppress-suspended interruption + * @zcc: zero condition code + * @ectl: extended control + * @pno: path not operational + * @res: reserved + * @fctl: function control + * @actl: activity control + * @stctl: status control + * @cpa: channel program address + * @dstat: device status + * @cstat: subchannel status + * @count: residual count + */ +struct cmd_scsw { + __u32 key : 4; + __u32 sctl : 1; + __u32 eswf : 1; + __u32 cc : 2; + __u32 fmt : 1; + __u32 pfch : 1; + __u32 isic : 1; + __u32 alcc : 1; + __u32 ssi : 1; + __u32 zcc : 1; + __u32 ectl : 1; + __u32 pno : 1; + __u32 res : 1; + __u32 fctl : 3; + __u32 actl : 7; + __u32 stctl : 5; + __u32 cpa; + __u32 dstat : 8; + __u32 cstat : 8; + __u32 count : 16; +} __attribute__ ((packed)); + +/** + * struct tm_scsw - transport-mode subchannel status word + * @key: subchannel key + * @eswf: esw format + * @cc: deferred condition code + * @fmt: format + * @x: IRB-format control + * @q: interrogate-complete + * @ectl: extended control + * @pno: path not operational + * @fctl: function control + * @actl: activity control + * @stctl: status control + * @tcw: TCW address + * @dstat: device status + * @cstat: subchannel status + * @fcxs: FCX status + * @schxs: subchannel-extended status + */ +struct tm_scsw { + u32 key:4; + u32 :1; + u32 eswf:1; + u32 cc:2; + u32 fmt:3; + u32 x:1; + u32 q:1; + u32 :1; + u32 ectl:1; + u32 pno:1; + u32 :1; + u32 fctl:3; + u32 actl:7; + u32 stctl:5; + u32 tcw; + u32 dstat:8; + u32 cstat:8; + u32 fcxs:8; + u32 schxs:8; +} __attribute__ ((packed)); + +/** + * struct eadm_scsw - subchannel status word for eadm subchannels + * @key: subchannel key + * @eswf: esw format + * @cc: deferred condition code + * @ectl: extended control + * @fctl: function control + * @actl: activity control + * @stctl: status control + * @aob: AOB address + * @dstat: device status + * @cstat: subchannel status + */ +struct eadm_scsw { + u32 key:4; + u32:1; + u32 eswf:1; + u32 cc:2; + u32:6; + u32 ectl:1; + u32:2; + u32 fctl:3; + u32 actl:7; + u32 stctl:5; + u32 aob; + u32 dstat:8; + u32 cstat:8; + u32:16; +} __packed; + +/** + * union scsw - subchannel status word + * @cmd: command-mode SCSW + * @tm: transport-mode SCSW + * @eadm: eadm SCSW + */ +union scsw { + struct cmd_scsw cmd; + struct tm_scsw tm; + struct eadm_scsw eadm; +} __packed; + +#define SCSW_FCTL_CLEAR_FUNC 0x1 +#define SCSW_FCTL_HALT_FUNC 0x2 +#define SCSW_FCTL_START_FUNC 0x4 + +#define SCSW_ACTL_SUSPENDED 0x1 +#define SCSW_ACTL_DEVACT 0x2 +#define SCSW_ACTL_SCHACT 0x4 +#define SCSW_ACTL_CLEAR_PEND 0x8 +#define SCSW_ACTL_HALT_PEND 0x10 +#define SCSW_ACTL_START_PEND 0x20 +#define SCSW_ACTL_RESUME_PEND 0x40 + +#define SCSW_STCTL_STATUS_PEND 0x1 +#define SCSW_STCTL_SEC_STATUS 0x2 +#define SCSW_STCTL_PRIM_STATUS 0x4 +#define SCSW_STCTL_INTER_STATUS 0x8 +#define SCSW_STCTL_ALERT_STATUS 0x10 + +#define DEV_STAT_ATTENTION 0x80 +#define DEV_STAT_STAT_MOD 0x40 +#define DEV_STAT_CU_END 0x20 +#define DEV_STAT_BUSY 0x10 +#define DEV_STAT_CHN_END 0x08 +#define DEV_STAT_DEV_END 0x04 +#define DEV_STAT_UNIT_CHECK 0x02 +#define DEV_STAT_UNIT_EXCEP 0x01 + +#define SCHN_STAT_PCI 0x80 +#define SCHN_STAT_INCORR_LEN 0x40 +#define SCHN_STAT_PROG_CHECK 0x20 +#define SCHN_STAT_PROT_CHECK 0x10 +#define SCHN_STAT_CHN_DATA_CHK 0x08 +#define SCHN_STAT_CHN_CTRL_CHK 0x04 +#define SCHN_STAT_INTF_CTRL_CHK 0x02 +#define SCHN_STAT_CHAIN_CHECK 0x01 + +/* + * architectured values for first sense byte + */ +#define SNS0_CMD_REJECT 0x80 +#define SNS_CMD_REJECT SNS0_CMD_REJEC +#define SNS0_INTERVENTION_REQ 0x40 +#define SNS0_BUS_OUT_CHECK 0x20 +#define SNS0_EQUIPMENT_CHECK 0x10 +#define SNS0_DATA_CHECK 0x08 +#define SNS0_OVERRUN 0x04 +#define SNS0_INCOMPL_DOMAIN 0x01 + +/* + * architectured values for second sense byte + */ +#define SNS1_PERM_ERR 0x80 +#define SNS1_INV_TRACK_FORMAT 0x40 +#define SNS1_EOC 0x20 +#define SNS1_MESSAGE_TO_OPER 0x10 +#define SNS1_NO_REC_FOUND 0x08 +#define SNS1_FILE_PROTECTED 0x04 +#define SNS1_WRITE_INHIBITED 0x02 +#define SNS1_INPRECISE_END 0x01 + +/* + * architectured values for third sense byte + */ +#define SNS2_REQ_INH_WRITE 0x80 +#define SNS2_CORRECTABLE 0x40 +#define SNS2_FIRST_LOG_ERR 0x20 +#define SNS2_ENV_DATA_PRESENT 0x10 +#define SNS2_INPRECISE_END 0x04 + +/** + * scsw_is_tm - check for transport mode scsw + * @scsw: pointer to scsw + * + * Return non-zero if the specified scsw is a transport mode scsw, zero + * otherwise. + */ +static inline int scsw_is_tm(union scsw *scsw) +{ + return css_general_characteristics.fcx && (scsw->tm.x == 1); +} + +/** + * scsw_key - return scsw key field + * @scsw: pointer to scsw + * + * Return the value of the key field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_key(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.key; + else + return scsw->cmd.key; +} + +/** + * scsw_eswf - return scsw eswf field + * @scsw: pointer to scsw + * + * Return the value of the eswf field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_eswf(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.eswf; + else + return scsw->cmd.eswf; +} + +/** + * scsw_cc - return scsw cc field + * @scsw: pointer to scsw + * + * Return the value of the cc field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_cc(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.cc; + else + return scsw->cmd.cc; +} + +/** + * scsw_ectl - return scsw ectl field + * @scsw: pointer to scsw + * + * Return the value of the ectl field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_ectl(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.ectl; + else + return scsw->cmd.ectl; +} + +/** + * scsw_pno - return scsw pno field + * @scsw: pointer to scsw + * + * Return the value of the pno field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_pno(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.pno; + else + return scsw->cmd.pno; +} + +/** + * scsw_fctl - return scsw fctl field + * @scsw: pointer to scsw + * + * Return the value of the fctl field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_fctl(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.fctl; + else + return scsw->cmd.fctl; +} + +/** + * scsw_actl - return scsw actl field + * @scsw: pointer to scsw + * + * Return the value of the actl field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_actl(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.actl; + else + return scsw->cmd.actl; +} + +/** + * scsw_stctl - return scsw stctl field + * @scsw: pointer to scsw + * + * Return the value of the stctl field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_stctl(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.stctl; + else + return scsw->cmd.stctl; +} + +/** + * scsw_dstat - return scsw dstat field + * @scsw: pointer to scsw + * + * Return the value of the dstat field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_dstat(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.dstat; + else + return scsw->cmd.dstat; +} + +/** + * scsw_cstat - return scsw cstat field + * @scsw: pointer to scsw + * + * Return the value of the cstat field of the specified scsw, regardless of + * whether it is a transport mode or command mode scsw. + */ +static inline u32 scsw_cstat(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw->tm.cstat; + else + return scsw->cmd.cstat; +} + +/** + * scsw_cmd_is_valid_key - check key field validity + * @scsw: pointer to scsw + * + * Return non-zero if the key field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_key(union scsw *scsw) +{ + return (scsw->cmd.fctl & SCSW_FCTL_START_FUNC); +} + +/** + * scsw_cmd_is_valid_sctl - check fctl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the fctl field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_sctl(union scsw *scsw) +{ + return (scsw->cmd.fctl & SCSW_FCTL_START_FUNC); +} + +/** + * scsw_cmd_is_valid_eswf - check eswf field validity + * @scsw: pointer to scsw + * + * Return non-zero if the eswf field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_eswf(union scsw *scsw) +{ + return (scsw->cmd.stctl & SCSW_STCTL_STATUS_PEND); +} + +/** + * scsw_cmd_is_valid_cc - check cc field validity + * @scsw: pointer to scsw + * + * Return non-zero if the cc field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_cc(union scsw *scsw) +{ + return (scsw->cmd.fctl & SCSW_FCTL_START_FUNC) && + (scsw->cmd.stctl & SCSW_STCTL_STATUS_PEND); +} + +/** + * scsw_cmd_is_valid_fmt - check fmt field validity + * @scsw: pointer to scsw + * + * Return non-zero if the fmt field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_fmt(union scsw *scsw) +{ + return (scsw->cmd.fctl & SCSW_FCTL_START_FUNC); +} + +/** + * scsw_cmd_is_valid_pfch - check pfch field validity + * @scsw: pointer to scsw + * + * Return non-zero if the pfch field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_pfch(union scsw *scsw) +{ + return (scsw->cmd.fctl & SCSW_FCTL_START_FUNC); +} + +/** + * scsw_cmd_is_valid_isic - check isic field validity + * @scsw: pointer to scsw + * + * Return non-zero if the isic field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_isic(union scsw *scsw) +{ + return (scsw->cmd.fctl & SCSW_FCTL_START_FUNC); +} + +/** + * scsw_cmd_is_valid_alcc - check alcc field validity + * @scsw: pointer to scsw + * + * Return non-zero if the alcc field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_alcc(union scsw *scsw) +{ + return (scsw->cmd.fctl & SCSW_FCTL_START_FUNC); +} + +/** + * scsw_cmd_is_valid_ssi - check ssi field validity + * @scsw: pointer to scsw + * + * Return non-zero if the ssi field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_ssi(union scsw *scsw) +{ + return (scsw->cmd.fctl & SCSW_FCTL_START_FUNC); +} + +/** + * scsw_cmd_is_valid_zcc - check zcc field validity + * @scsw: pointer to scsw + * + * Return non-zero if the zcc field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_zcc(union scsw *scsw) +{ + return (scsw->cmd.fctl & SCSW_FCTL_START_FUNC) && + (scsw->cmd.stctl & SCSW_STCTL_INTER_STATUS); +} + +/** + * scsw_cmd_is_valid_ectl - check ectl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the ectl field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_ectl(union scsw *scsw) +{ + return (scsw->cmd.stctl & SCSW_STCTL_STATUS_PEND) && + !(scsw->cmd.stctl & SCSW_STCTL_INTER_STATUS) && + (scsw->cmd.stctl & SCSW_STCTL_ALERT_STATUS); +} + +/** + * scsw_cmd_is_valid_pno - check pno field validity + * @scsw: pointer to scsw + * + * Return non-zero if the pno field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_pno(union scsw *scsw) +{ + return (scsw->cmd.fctl != 0) && + (scsw->cmd.stctl & SCSW_STCTL_STATUS_PEND) && + (!(scsw->cmd.stctl & SCSW_STCTL_INTER_STATUS) || + ((scsw->cmd.stctl & SCSW_STCTL_INTER_STATUS) && + (scsw->cmd.actl & SCSW_ACTL_SUSPENDED))); +} + +/** + * scsw_cmd_is_valid_fctl - check fctl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the fctl field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_fctl(union scsw *scsw) +{ + /* Only valid if pmcw.dnv == 1*/ + return 1; +} + +/** + * scsw_cmd_is_valid_actl - check actl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the actl field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_actl(union scsw *scsw) +{ + /* Only valid if pmcw.dnv == 1*/ + return 1; +} + +/** + * scsw_cmd_is_valid_stctl - check stctl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the stctl field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_stctl(union scsw *scsw) +{ + /* Only valid if pmcw.dnv == 1*/ + return 1; +} + +/** + * scsw_cmd_is_valid_dstat - check dstat field validity + * @scsw: pointer to scsw + * + * Return non-zero if the dstat field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_dstat(union scsw *scsw) +{ + return (scsw->cmd.stctl & SCSW_STCTL_STATUS_PEND) && + (scsw->cmd.cc != 3); +} + +/** + * scsw_cmd_is_valid_cstat - check cstat field validity + * @scsw: pointer to scsw + * + * Return non-zero if the cstat field of the specified command mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_cmd_is_valid_cstat(union scsw *scsw) +{ + return (scsw->cmd.stctl & SCSW_STCTL_STATUS_PEND) && + (scsw->cmd.cc != 3); +} + +/** + * scsw_tm_is_valid_key - check key field validity + * @scsw: pointer to scsw + * + * Return non-zero if the key field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_key(union scsw *scsw) +{ + return (scsw->tm.fctl & SCSW_FCTL_START_FUNC); +} + +/** + * scsw_tm_is_valid_eswf - check eswf field validity + * @scsw: pointer to scsw + * + * Return non-zero if the eswf field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_eswf(union scsw *scsw) +{ + return (scsw->tm.stctl & SCSW_STCTL_STATUS_PEND); +} + +/** + * scsw_tm_is_valid_cc - check cc field validity + * @scsw: pointer to scsw + * + * Return non-zero if the cc field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_cc(union scsw *scsw) +{ + return (scsw->tm.fctl & SCSW_FCTL_START_FUNC) && + (scsw->tm.stctl & SCSW_STCTL_STATUS_PEND); +} + +/** + * scsw_tm_is_valid_fmt - check fmt field validity + * @scsw: pointer to scsw + * + * Return non-zero if the fmt field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_fmt(union scsw *scsw) +{ + return 1; +} + +/** + * scsw_tm_is_valid_x - check x field validity + * @scsw: pointer to scsw + * + * Return non-zero if the x field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_x(union scsw *scsw) +{ + return 1; +} + +/** + * scsw_tm_is_valid_q - check q field validity + * @scsw: pointer to scsw + * + * Return non-zero if the q field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_q(union scsw *scsw) +{ + return 1; +} + +/** + * scsw_tm_is_valid_ectl - check ectl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the ectl field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_ectl(union scsw *scsw) +{ + return (scsw->tm.stctl & SCSW_STCTL_STATUS_PEND) && + !(scsw->tm.stctl & SCSW_STCTL_INTER_STATUS) && + (scsw->tm.stctl & SCSW_STCTL_ALERT_STATUS); +} + +/** + * scsw_tm_is_valid_pno - check pno field validity + * @scsw: pointer to scsw + * + * Return non-zero if the pno field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_pno(union scsw *scsw) +{ + return (scsw->tm.fctl != 0) && + (scsw->tm.stctl & SCSW_STCTL_STATUS_PEND) && + (!(scsw->tm.stctl & SCSW_STCTL_INTER_STATUS) || + ((scsw->tm.stctl & SCSW_STCTL_INTER_STATUS) && + (scsw->tm.actl & SCSW_ACTL_SUSPENDED))); +} + +/** + * scsw_tm_is_valid_fctl - check fctl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the fctl field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_fctl(union scsw *scsw) +{ + /* Only valid if pmcw.dnv == 1*/ + return 1; +} + +/** + * scsw_tm_is_valid_actl - check actl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the actl field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_actl(union scsw *scsw) +{ + /* Only valid if pmcw.dnv == 1*/ + return 1; +} + +/** + * scsw_tm_is_valid_stctl - check stctl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the stctl field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_stctl(union scsw *scsw) +{ + /* Only valid if pmcw.dnv == 1*/ + return 1; +} + +/** + * scsw_tm_is_valid_dstat - check dstat field validity + * @scsw: pointer to scsw + * + * Return non-zero if the dstat field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_dstat(union scsw *scsw) +{ + return (scsw->tm.stctl & SCSW_STCTL_STATUS_PEND) && + (scsw->tm.cc != 3); +} + +/** + * scsw_tm_is_valid_cstat - check cstat field validity + * @scsw: pointer to scsw + * + * Return non-zero if the cstat field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_cstat(union scsw *scsw) +{ + return (scsw->tm.stctl & SCSW_STCTL_STATUS_PEND) && + (scsw->tm.cc != 3); +} + +/** + * scsw_tm_is_valid_fcxs - check fcxs field validity + * @scsw: pointer to scsw + * + * Return non-zero if the fcxs field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_fcxs(union scsw *scsw) +{ + return 1; +} + +/** + * scsw_tm_is_valid_schxs - check schxs field validity + * @scsw: pointer to scsw + * + * Return non-zero if the schxs field of the specified transport mode scsw is + * valid, zero otherwise. + */ +static inline int scsw_tm_is_valid_schxs(union scsw *scsw) +{ + return (scsw->tm.cstat & (SCHN_STAT_PROG_CHECK | + SCHN_STAT_INTF_CTRL_CHK | + SCHN_STAT_PROT_CHECK | + SCHN_STAT_CHN_DATA_CHK)); +} + +/** + * scsw_is_valid_actl - check actl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the actl field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_actl(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_actl(scsw); + else + return scsw_cmd_is_valid_actl(scsw); +} + +/** + * scsw_is_valid_cc - check cc field validity + * @scsw: pointer to scsw + * + * Return non-zero if the cc field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_cc(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_cc(scsw); + else + return scsw_cmd_is_valid_cc(scsw); +} + +/** + * scsw_is_valid_cstat - check cstat field validity + * @scsw: pointer to scsw + * + * Return non-zero if the cstat field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_cstat(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_cstat(scsw); + else + return scsw_cmd_is_valid_cstat(scsw); +} + +/** + * scsw_is_valid_dstat - check dstat field validity + * @scsw: pointer to scsw + * + * Return non-zero if the dstat field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_dstat(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_dstat(scsw); + else + return scsw_cmd_is_valid_dstat(scsw); +} + +/** + * scsw_is_valid_ectl - check ectl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the ectl field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_ectl(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_ectl(scsw); + else + return scsw_cmd_is_valid_ectl(scsw); +} + +/** + * scsw_is_valid_eswf - check eswf field validity + * @scsw: pointer to scsw + * + * Return non-zero if the eswf field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_eswf(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_eswf(scsw); + else + return scsw_cmd_is_valid_eswf(scsw); +} + +/** + * scsw_is_valid_fctl - check fctl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the fctl field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_fctl(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_fctl(scsw); + else + return scsw_cmd_is_valid_fctl(scsw); +} + +/** + * scsw_is_valid_key - check key field validity + * @scsw: pointer to scsw + * + * Return non-zero if the key field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_key(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_key(scsw); + else + return scsw_cmd_is_valid_key(scsw); +} + +/** + * scsw_is_valid_pno - check pno field validity + * @scsw: pointer to scsw + * + * Return non-zero if the pno field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_pno(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_pno(scsw); + else + return scsw_cmd_is_valid_pno(scsw); +} + +/** + * scsw_is_valid_stctl - check stctl field validity + * @scsw: pointer to scsw + * + * Return non-zero if the stctl field of the specified scsw is valid, + * regardless of whether it is a transport mode or command mode scsw. + * Return zero if the field does not contain a valid value. + */ +static inline int scsw_is_valid_stctl(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_valid_stctl(scsw); + else + return scsw_cmd_is_valid_stctl(scsw); +} + +/** + * scsw_cmd_is_solicited - check for solicited scsw + * @scsw: pointer to scsw + * + * Return non-zero if the command mode scsw indicates that the associated + * status condition is solicited, zero if it is unsolicited. + */ +static inline int scsw_cmd_is_solicited(union scsw *scsw) +{ + return (scsw->cmd.cc != 0) || (scsw->cmd.stctl != + (SCSW_STCTL_STATUS_PEND | SCSW_STCTL_ALERT_STATUS)); +} + +/** + * scsw_tm_is_solicited - check for solicited scsw + * @scsw: pointer to scsw + * + * Return non-zero if the transport mode scsw indicates that the associated + * status condition is solicited, zero if it is unsolicited. + */ +static inline int scsw_tm_is_solicited(union scsw *scsw) +{ + return (scsw->tm.cc != 0) || (scsw->tm.stctl != + (SCSW_STCTL_STATUS_PEND | SCSW_STCTL_ALERT_STATUS)); +} + +/** + * scsw_is_solicited - check for solicited scsw + * @scsw: pointer to scsw + * + * Return non-zero if the transport or command mode scsw indicates that the + * associated status condition is solicited, zero if it is unsolicited. + */ +static inline int scsw_is_solicited(union scsw *scsw) +{ + if (scsw_is_tm(scsw)) + return scsw_tm_is_solicited(scsw); + else + return scsw_cmd_is_solicited(scsw); +} + +#endif /* _ASM_S390_SCSW_H_ */ diff --git a/arch/s390/include/asm/seccomp.h b/arch/s390/include/asm/seccomp.h new file mode 100644 index 000000000..781a9cf9b --- /dev/null +++ b/arch/s390/include/asm/seccomp.h @@ -0,0 +1,16 @@ +#ifndef _ASM_S390_SECCOMP_H +#define _ASM_S390_SECCOMP_H + +#include <linux/unistd.h> + +#define __NR_seccomp_read __NR_read +#define __NR_seccomp_write __NR_write +#define __NR_seccomp_exit __NR_exit +#define __NR_seccomp_sigreturn __NR_sigreturn + +#define __NR_seccomp_read_32 __NR_read +#define __NR_seccomp_write_32 __NR_write +#define __NR_seccomp_exit_32 __NR_exit +#define __NR_seccomp_sigreturn_32 __NR_sigreturn + +#endif /* _ASM_S390_SECCOMP_H */ diff --git a/arch/s390/include/asm/sections.h b/arch/s390/include/asm/sections.h new file mode 100644 index 000000000..fbd9116eb --- /dev/null +++ b/arch/s390/include/asm/sections.h @@ -0,0 +1,8 @@ +#ifndef _S390_SECTIONS_H +#define _S390_SECTIONS_H + +#include <asm-generic/sections.h> + +extern char _eshared[], _ehead[]; + +#endif diff --git a/arch/s390/include/asm/segment.h b/arch/s390/include/asm/segment.h new file mode 100644 index 000000000..8bfce3475 --- /dev/null +++ b/arch/s390/include/asm/segment.h @@ -0,0 +1,4 @@ +#ifndef _ASM_SEGMENT_H +#define _ASM_SEGMENT_H + +#endif diff --git a/arch/s390/include/asm/serial.h b/arch/s390/include/asm/serial.h new file mode 100644 index 000000000..5b3e48ef5 --- /dev/null +++ b/arch/s390/include/asm/serial.h @@ -0,0 +1,6 @@ +#ifndef _ASM_S390_SERIAL_H +#define _ASM_S390_SERIAL_H + +#define BASE_BAUD 0 + +#endif /* _ASM_S390_SERIAL_H */ diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h new file mode 100644 index 000000000..b8ffc1bd0 --- /dev/null +++ b/arch/s390/include/asm/setup.h @@ -0,0 +1,119 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2010 + */ +#ifndef _ASM_S390_SETUP_H +#define _ASM_S390_SETUP_H + +#include <uapi/asm/setup.h> + + +#define PARMAREA 0x10400 + +#ifndef __ASSEMBLY__ + +#include <asm/lowcore.h> +#include <asm/types.h> + +#define IPL_DEVICE (*(unsigned long *) (0x10400)) +#define INITRD_START (*(unsigned long *) (0x10408)) +#define INITRD_SIZE (*(unsigned long *) (0x10410)) +#define OLDMEM_BASE (*(unsigned long *) (0x10418)) +#define OLDMEM_SIZE (*(unsigned long *) (0x10420)) +#define COMMAND_LINE ((char *) (0x10480)) + +extern int memory_end_set; +extern unsigned long memory_end; +extern unsigned long max_physmem_end; + +extern void detect_memory_memblock(void); + +/* + * Machine features detected in head.S + */ + +#define MACHINE_FLAG_VM (1UL << 0) +#define MACHINE_FLAG_IEEE (1UL << 1) +#define MACHINE_FLAG_CSP (1UL << 2) +#define MACHINE_FLAG_MVPG (1UL << 3) +#define MACHINE_FLAG_DIAG44 (1UL << 4) +#define MACHINE_FLAG_IDTE (1UL << 5) +#define MACHINE_FLAG_DIAG9C (1UL << 6) +#define MACHINE_FLAG_KVM (1UL << 8) +#define MACHINE_FLAG_ESOP (1UL << 9) +#define MACHINE_FLAG_EDAT1 (1UL << 10) +#define MACHINE_FLAG_EDAT2 (1UL << 11) +#define MACHINE_FLAG_LPAR (1UL << 12) +#define MACHINE_FLAG_LPP (1UL << 13) +#define MACHINE_FLAG_TOPOLOGY (1UL << 14) +#define MACHINE_FLAG_TE (1UL << 15) +#define MACHINE_FLAG_TLB_LC (1UL << 17) +#define MACHINE_FLAG_VX (1UL << 18) +#define MACHINE_FLAG_CAD (1UL << 19) + +#define MACHINE_IS_VM (S390_lowcore.machine_flags & MACHINE_FLAG_VM) +#define MACHINE_IS_KVM (S390_lowcore.machine_flags & MACHINE_FLAG_KVM) +#define MACHINE_IS_LPAR (S390_lowcore.machine_flags & MACHINE_FLAG_LPAR) + +#define MACHINE_HAS_DIAG9C (S390_lowcore.machine_flags & MACHINE_FLAG_DIAG9C) +#define MACHINE_HAS_ESOP (S390_lowcore.machine_flags & MACHINE_FLAG_ESOP) +#define MACHINE_HAS_PFMF MACHINE_HAS_EDAT1 +#define MACHINE_HAS_HPAGE MACHINE_HAS_EDAT1 + +#define MACHINE_HAS_IDTE (S390_lowcore.machine_flags & MACHINE_FLAG_IDTE) +#define MACHINE_HAS_DIAG44 (S390_lowcore.machine_flags & MACHINE_FLAG_DIAG44) +#define MACHINE_HAS_EDAT1 (S390_lowcore.machine_flags & MACHINE_FLAG_EDAT1) +#define MACHINE_HAS_EDAT2 (S390_lowcore.machine_flags & MACHINE_FLAG_EDAT2) +#define MACHINE_HAS_LPP (S390_lowcore.machine_flags & MACHINE_FLAG_LPP) +#define MACHINE_HAS_TOPOLOGY (S390_lowcore.machine_flags & MACHINE_FLAG_TOPOLOGY) +#define MACHINE_HAS_TE (S390_lowcore.machine_flags & MACHINE_FLAG_TE) +#define MACHINE_HAS_TLB_LC (S390_lowcore.machine_flags & MACHINE_FLAG_TLB_LC) +#define MACHINE_HAS_VX (S390_lowcore.machine_flags & MACHINE_FLAG_VX) +#define MACHINE_HAS_CAD (S390_lowcore.machine_flags & MACHINE_FLAG_CAD) + +/* + * Console mode. Override with conmode= + */ +extern unsigned int console_mode; +extern unsigned int console_devno; +extern unsigned int console_irq; + +extern char vmhalt_cmd[]; +extern char vmpoff_cmd[]; + +#define CONSOLE_IS_UNDEFINED (console_mode == 0) +#define CONSOLE_IS_SCLP (console_mode == 1) +#define CONSOLE_IS_3215 (console_mode == 2) +#define CONSOLE_IS_3270 (console_mode == 3) +#define SET_CONSOLE_SCLP do { console_mode = 1; } while (0) +#define SET_CONSOLE_3215 do { console_mode = 2; } while (0) +#define SET_CONSOLE_3270 do { console_mode = 3; } while (0) + +#define NSS_NAME_SIZE 8 +extern char kernel_nss_name[]; + +#ifdef CONFIG_PFAULT +extern int pfault_init(void); +extern void pfault_fini(void); +#else /* CONFIG_PFAULT */ +#define pfault_init() ({-1;}) +#define pfault_fini() do { } while (0) +#endif /* CONFIG_PFAULT */ + +extern void cmma_init(void); + +extern void (*_machine_restart)(char *command); +extern void (*_machine_halt)(void); +extern void (*_machine_power_off)(void); + +#else /* __ASSEMBLY__ */ + +#define IPL_DEVICE 0x10400 +#define INITRD_START 0x10408 +#define INITRD_SIZE 0x10410 +#define OLDMEM_BASE 0x10418 +#define OLDMEM_SIZE 0x10420 +#define COMMAND_LINE 0x10480 + +#endif /* __ASSEMBLY__ */ +#endif /* _ASM_S390_SETUP_H */ diff --git a/arch/s390/include/asm/sfp-machine.h b/arch/s390/include/asm/sfp-machine.h new file mode 100644 index 000000000..4e16aede4 --- /dev/null +++ b/arch/s390/include/asm/sfp-machine.h @@ -0,0 +1,142 @@ +/* Machine-dependent software floating-point definitions. + S/390 kernel version. + Copyright (C) 1997,1998,1999 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Richard Henderson (rth@cygnus.com), + Jakub Jelinek (jj@ultra.linux.cz), + David S. Miller (davem@redhat.com) and + Peter Maydell (pmaydell@chiark.greenend.org.uk). + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Library General Public License as + published by the Free Software Foundation; either version 2 of the + License, or (at your option) any later version. + + The GNU C Library 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 + Library General Public License for more details. + + You should have received a copy of the GNU Library General Public + License along with the GNU C Library; see the file COPYING.LIB. If + not, write to the Free Software Foundation, Inc., + 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ + +#ifndef _SFP_MACHINE_H +#define _SFP_MACHINE_H + + +#define _FP_W_TYPE_SIZE 32 +#define _FP_W_TYPE unsigned int +#define _FP_WS_TYPE signed int +#define _FP_I_TYPE int + +#define _FP_MUL_MEAT_S(R,X,Y) \ + _FP_MUL_MEAT_1_wide(_FP_WFRACBITS_S,R,X,Y,umul_ppmm) +#define _FP_MUL_MEAT_D(R,X,Y) \ + _FP_MUL_MEAT_2_wide(_FP_WFRACBITS_D,R,X,Y,umul_ppmm) +#define _FP_MUL_MEAT_Q(R,X,Y) \ + _FP_MUL_MEAT_4_wide(_FP_WFRACBITS_Q,R,X,Y,umul_ppmm) + +#define _FP_DIV_MEAT_S(R,X,Y) _FP_DIV_MEAT_1_udiv(S,R,X,Y) +#define _FP_DIV_MEAT_D(R,X,Y) _FP_DIV_MEAT_2_udiv(D,R,X,Y) +#define _FP_DIV_MEAT_Q(R,X,Y) _FP_DIV_MEAT_4_udiv(Q,R,X,Y) + +#define _FP_NANFRAC_S ((_FP_QNANBIT_S << 1) - 1) +#define _FP_NANFRAC_D ((_FP_QNANBIT_D << 1) - 1), -1 +#define _FP_NANFRAC_Q ((_FP_QNANBIT_Q << 1) - 1), -1, -1, -1 +#define _FP_NANSIGN_S 0 +#define _FP_NANSIGN_D 0 +#define _FP_NANSIGN_Q 0 + +#define _FP_KEEPNANFRACP 1 + +/* + * If one NaN is signaling and the other is not, + * we choose that one, otherwise we choose X. + */ +#define _FP_CHOOSENAN(fs, wc, R, X, Y, OP) \ + do { \ + if ((_FP_FRAC_HIGH_RAW_##fs(X) & _FP_QNANBIT_##fs) \ + && !(_FP_FRAC_HIGH_RAW_##fs(Y) & _FP_QNANBIT_##fs)) \ + { \ + R##_s = Y##_s; \ + _FP_FRAC_COPY_##wc(R,Y); \ + } \ + else \ + { \ + R##_s = X##_s; \ + _FP_FRAC_COPY_##wc(R,X); \ + } \ + R##_c = FP_CLS_NAN; \ + } while (0) + +/* Some assembly to speed things up. */ +#define __FP_FRAC_ADD_3(r2,r1,r0,x2,x1,x0,y2,y1,y0) ({ \ + unsigned int __r2 = (x2) + (y2); \ + unsigned int __r1 = (x1); \ + unsigned int __r0 = (x0); \ + asm volatile( \ + " alr %2,%3\n" \ + " brc 12,0f\n" \ + " lhi 0,1\n" \ + " alr %1,0\n" \ + " brc 12,0f\n" \ + " alr %0,0\n" \ + "0:" \ + : "+&d" (__r2), "+&d" (__r1), "+&d" (__r0) \ + : "d" (y0), "i" (1) : "cc", "0" ); \ + asm volatile( \ + " alr %1,%2\n" \ + " brc 12,0f\n" \ + " ahi %0,1\n" \ + "0:" \ + : "+&d" (__r2), "+&d" (__r1) \ + : "d" (y1) : "cc"); \ + (r2) = __r2; \ + (r1) = __r1; \ + (r0) = __r0; \ +}) + +#define __FP_FRAC_SUB_3(r2,r1,r0,x2,x1,x0,y2,y1,y0) ({ \ + unsigned int __r2 = (x2) - (y2); \ + unsigned int __r1 = (x1); \ + unsigned int __r0 = (x0); \ + asm volatile( \ + " slr %2,%3\n" \ + " brc 3,0f\n" \ + " lhi 0,1\n" \ + " slr %1,0\n" \ + " brc 3,0f\n" \ + " slr %0,0\n" \ + "0:" \ + : "+&d" (__r2), "+&d" (__r1), "+&d" (__r0) \ + : "d" (y0) : "cc", "0"); \ + asm volatile( \ + " slr %1,%2\n" \ + " brc 3,0f\n" \ + " ahi %0,-1\n" \ + "0:" \ + : "+&d" (__r2), "+&d" (__r1) \ + : "d" (y1) : "cc"); \ + (r2) = __r2; \ + (r1) = __r1; \ + (r0) = __r0; \ +}) + +#define __FP_FRAC_DEC_3(x2,x1,x0,y2,y1,y0) __FP_FRAC_SUB_3(x2,x1,x0,x2,x1,x0,y2,y1,y0) + +/* Obtain the current rounding mode. */ +#define FP_ROUNDMODE mode + +/* Exception flags. */ +#define FP_EX_INVALID 0x800000 +#define FP_EX_DIVZERO 0x400000 +#define FP_EX_OVERFLOW 0x200000 +#define FP_EX_UNDERFLOW 0x100000 +#define FP_EX_INEXACT 0x080000 + +/* We write the results always */ +#define FP_INHIBIT_RESULTS 0 + +#endif diff --git a/arch/s390/include/asm/sfp-util.h b/arch/s390/include/asm/sfp-util.h new file mode 100644 index 000000000..c8b7cf9d6 --- /dev/null +++ b/arch/s390/include/asm/sfp-util.h @@ -0,0 +1,67 @@ +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/types.h> +#include <asm/byteorder.h> + +#define add_ssaaaa(sh, sl, ah, al, bh, bl) ({ \ + unsigned int __sh = (ah); \ + unsigned int __sl = (al); \ + asm volatile( \ + " alr %1,%3\n" \ + " brc 12,0f\n" \ + " ahi %0,1\n" \ + "0: alr %0,%2" \ + : "+&d" (__sh), "+d" (__sl) \ + : "d" (bh), "d" (bl) : "cc"); \ + (sh) = __sh; \ + (sl) = __sl; \ +}) + +#define sub_ddmmss(sh, sl, ah, al, bh, bl) ({ \ + unsigned int __sh = (ah); \ + unsigned int __sl = (al); \ + asm volatile( \ + " slr %1,%3\n" \ + " brc 3,0f\n" \ + " ahi %0,-1\n" \ + "0: slr %0,%2" \ + : "+&d" (__sh), "+d" (__sl) \ + : "d" (bh), "d" (bl) : "cc"); \ + (sh) = __sh; \ + (sl) = __sl; \ +}) + +/* a umul b = a mul b + (a>=2<<31) ? b<<32:0 + (b>=2<<31) ? a<<32:0 */ +#define umul_ppmm(wh, wl, u, v) ({ \ + unsigned int __wh = u; \ + unsigned int __wl = v; \ + asm volatile( \ + " ltr 1,%0\n" \ + " mr 0,%1\n" \ + " jnm 0f\n" \ + " alr 0,%1\n" \ + "0: ltr %1,%1\n" \ + " jnm 1f\n" \ + " alr 0,%0\n" \ + "1: lr %0,0\n" \ + " lr %1,1\n" \ + : "+d" (__wh), "+d" (__wl) \ + : : "0", "1", "cc"); \ + wh = __wh; \ + wl = __wl; \ +}) + +#define udiv_qrnnd(q, r, n1, n0, d) \ + do { unsigned long __n; \ + unsigned int __r, __d; \ + __n = ((unsigned long)(n1) << 32) + n0; \ + __d = (d); \ + (q) = __n / __d; \ + (r) = __n % __d; \ + } while (0) + +#define UDIV_NEEDS_NORMALIZATION 0 + +#define abort() BUG() + +#define __BYTE_ORDER __BIG_ENDIAN diff --git a/arch/s390/include/asm/shmparam.h b/arch/s390/include/asm/shmparam.h new file mode 100644 index 000000000..e98518273 --- /dev/null +++ b/arch/s390/include/asm/shmparam.h @@ -0,0 +1,11 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/shmparam.h" + */ +#ifndef _ASM_S390_SHMPARAM_H +#define _ASM_S390_SHMPARAM_H + +#define SHMLBA PAGE_SIZE /* attach addr a multiple of this */ + +#endif /* _ASM_S390_SHMPARAM_H */ diff --git a/arch/s390/include/asm/signal.h b/arch/s390/include/asm/signal.h new file mode 100644 index 000000000..abf9e5735 --- /dev/null +++ b/arch/s390/include/asm/signal.h @@ -0,0 +1,25 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/signal.h" + */ +#ifndef _ASMS390_SIGNAL_H +#define _ASMS390_SIGNAL_H + +#include <uapi/asm/signal.h> + +/* Most things should be clean enough to redefine this at will, if care + is taken to make libc match. */ +#include <asm/sigcontext.h> +#define _NSIG _SIGCONTEXT_NSIG +#define _NSIG_BPW _SIGCONTEXT_NSIG_BPW +#define _NSIG_WORDS _SIGCONTEXT_NSIG_WORDS + +typedef unsigned long old_sigset_t; /* at least 32 bits */ + +typedef struct { + unsigned long sig[_NSIG_WORDS]; +} sigset_t; + +#define __ARCH_HAS_SA_RESTORER +#endif diff --git a/arch/s390/include/asm/sigp.h b/arch/s390/include/asm/sigp.h new file mode 100644 index 000000000..ec60cf7fa --- /dev/null +++ b/arch/s390/include/asm/sigp.h @@ -0,0 +1,57 @@ +#ifndef __S390_ASM_SIGP_H +#define __S390_ASM_SIGP_H + +/* SIGP order codes */ +#define SIGP_SENSE 1 +#define SIGP_EXTERNAL_CALL 2 +#define SIGP_EMERGENCY_SIGNAL 3 +#define SIGP_START 4 +#define SIGP_STOP 5 +#define SIGP_RESTART 6 +#define SIGP_STOP_AND_STORE_STATUS 9 +#define SIGP_INITIAL_CPU_RESET 11 +#define SIGP_CPU_RESET 12 +#define SIGP_SET_PREFIX 13 +#define SIGP_STORE_STATUS_AT_ADDRESS 14 +#define SIGP_SET_ARCHITECTURE 18 +#define SIGP_COND_EMERGENCY_SIGNAL 19 +#define SIGP_SENSE_RUNNING 21 +#define SIGP_SET_MULTI_THREADING 22 +#define SIGP_STORE_ADDITIONAL_STATUS 23 + +/* SIGP condition codes */ +#define SIGP_CC_ORDER_CODE_ACCEPTED 0 +#define SIGP_CC_STATUS_STORED 1 +#define SIGP_CC_BUSY 2 +#define SIGP_CC_NOT_OPERATIONAL 3 + +/* SIGP cpu status bits */ + +#define SIGP_STATUS_CHECK_STOP 0x00000010UL +#define SIGP_STATUS_STOPPED 0x00000040UL +#define SIGP_STATUS_EXT_CALL_PENDING 0x00000080UL +#define SIGP_STATUS_INVALID_PARAMETER 0x00000100UL +#define SIGP_STATUS_INCORRECT_STATE 0x00000200UL +#define SIGP_STATUS_NOT_RUNNING 0x00000400UL + +#ifndef __ASSEMBLY__ + +static inline int __pcpu_sigp(u16 addr, u8 order, unsigned long parm, + u32 *status) +{ + register unsigned long reg1 asm ("1") = parm; + int cc; + + asm volatile( + " sigp %1,%2,0(%3)\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (cc), "+d" (reg1) : "d" (addr), "a" (order) : "cc"); + if (status && cc == 1) + *status = reg1; + return cc; +} + +#endif /* __ASSEMBLY__ */ + +#endif /* __S390_ASM_SIGP_H */ diff --git a/arch/s390/include/asm/smp.h b/arch/s390/include/asm/smp.h new file mode 100644 index 000000000..b3bd0282d --- /dev/null +++ b/arch/s390/include/asm/smp.h @@ -0,0 +1,80 @@ +/* + * Copyright IBM Corp. 1999, 2012 + * Author(s): Denis Joseph Barrow, + * Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Heiko Carstens <heiko.carstens@de.ibm.com>, + */ +#ifndef __ASM_SMP_H +#define __ASM_SMP_H + +#include <asm/sigp.h> + +#ifdef CONFIG_SMP + +#include <asm/lowcore.h> + +#define raw_smp_processor_id() (S390_lowcore.cpu_nr) + +extern struct mutex smp_cpu_state_mutex; +extern unsigned int smp_cpu_mt_shift; +extern unsigned int smp_cpu_mtid; + +extern int __cpu_up(unsigned int cpu, struct task_struct *tidle); + +extern void arch_send_call_function_single_ipi(int cpu); +extern void arch_send_call_function_ipi_mask(const struct cpumask *mask); + +extern void smp_call_online_cpu(void (*func)(void *), void *); +extern void smp_call_ipl_cpu(void (*func)(void *), void *); + +extern int smp_find_processor_id(u16 address); +extern int smp_store_status(int cpu); +extern int smp_vcpu_scheduled(int cpu); +extern void smp_yield_cpu(int cpu); +extern void smp_cpu_set_polarization(int cpu, int val); +extern int smp_cpu_get_polarization(int cpu); +extern void smp_fill_possible_mask(void); + +#else /* CONFIG_SMP */ + +#define smp_cpu_mtid 0 + +static inline void smp_call_ipl_cpu(void (*func)(void *), void *data) +{ + func(data); +} + +static inline void smp_call_online_cpu(void (*func)(void *), void *data) +{ + func(data); +} + +static inline int smp_find_processor_id(u16 address) { return 0; } +static inline int smp_store_status(int cpu) { return 0; } +static inline int smp_vcpu_scheduled(int cpu) { return 1; } +static inline void smp_yield_cpu(int cpu) { } +static inline void smp_fill_possible_mask(void) { } + +#endif /* CONFIG_SMP */ + +static inline void smp_stop_cpu(void) +{ + u16 pcpu = stap(); + + for (;;) { + __pcpu_sigp(pcpu, SIGP_STOP, 0, NULL); + cpu_relax(); + } +} + +#ifdef CONFIG_HOTPLUG_CPU +extern int smp_rescan_cpus(void); +extern void __noreturn cpu_die(void); +extern void __cpu_die(unsigned int cpu); +extern int __cpu_disable(void); +#else +static inline int smp_rescan_cpus(void) { return 0; } +static inline void cpu_die(void) { } +#endif + +#endif /* __ASM_SMP_H */ diff --git a/arch/s390/include/asm/sparsemem.h b/arch/s390/include/asm/sparsemem.h new file mode 100644 index 000000000..487428b6d --- /dev/null +++ b/arch/s390/include/asm/sparsemem.h @@ -0,0 +1,7 @@ +#ifndef _ASM_S390_SPARSEMEM_H +#define _ASM_S390_SPARSEMEM_H + +#define SECTION_SIZE_BITS 28 +#define MAX_PHYSMEM_BITS 46 + +#endif /* _ASM_S390_SPARSEMEM_H */ diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h new file mode 100644 index 000000000..0e37cd041 --- /dev/null +++ b/arch/s390/include/asm/spinlock.h @@ -0,0 +1,280 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "include/asm-i386/spinlock.h" + */ + +#ifndef __ASM_SPINLOCK_H +#define __ASM_SPINLOCK_H + +#include <linux/smp.h> + +#define SPINLOCK_LOCKVAL (S390_lowcore.spinlock_lockval) + +extern int spin_retry; + +static inline int +_raw_compare_and_swap(unsigned int *lock, unsigned int old, unsigned int new) +{ + return __sync_bool_compare_and_swap(lock, old, new); +} + +/* + * Simple spin lock operations. There are two variants, one clears IRQ's + * on the local processor, one does not. + * + * We make no fairness assumptions. They have a cost. + * + * (the type definitions are in asm/spinlock_types.h) + */ + +void arch_lock_relax(unsigned int cpu); + +void arch_spin_lock_wait(arch_spinlock_t *); +int arch_spin_trylock_retry(arch_spinlock_t *); +void arch_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags); + +static inline void arch_spin_relax(arch_spinlock_t *lock) +{ + arch_lock_relax(lock->lock); +} + +static inline u32 arch_spin_lockval(int cpu) +{ + return ~cpu; +} + +static inline int arch_spin_value_unlocked(arch_spinlock_t lock) +{ + return lock.lock == 0; +} + +static inline int arch_spin_is_locked(arch_spinlock_t *lp) +{ + return ACCESS_ONCE(lp->lock) != 0; +} + +static inline int arch_spin_trylock_once(arch_spinlock_t *lp) +{ + barrier(); + return likely(arch_spin_value_unlocked(*lp) && + _raw_compare_and_swap(&lp->lock, 0, SPINLOCK_LOCKVAL)); +} + +static inline void arch_spin_lock(arch_spinlock_t *lp) +{ + if (!arch_spin_trylock_once(lp)) + arch_spin_lock_wait(lp); +} + +static inline void arch_spin_lock_flags(arch_spinlock_t *lp, + unsigned long flags) +{ + if (!arch_spin_trylock_once(lp)) + arch_spin_lock_wait_flags(lp, flags); +} + +static inline int arch_spin_trylock(arch_spinlock_t *lp) +{ + if (!arch_spin_trylock_once(lp)) + return arch_spin_trylock_retry(lp); + return 1; +} + +static inline void arch_spin_unlock(arch_spinlock_t *lp) +{ + typecheck(unsigned int, lp->lock); + asm volatile( + __ASM_BARRIER + "st %1,%0\n" + : "+Q" (lp->lock) + : "d" (0) + : "cc", "memory"); +} + +static inline void arch_spin_unlock_wait(arch_spinlock_t *lock) +{ + while (arch_spin_is_locked(lock)) + arch_spin_relax(lock); +} + +/* + * Read-write spinlocks, allowing multiple readers + * but only one writer. + * + * NOTE! it is quite common to have readers in interrupts + * but no interrupt writers. For those circumstances we + * can "mix" irq-safe locks - any writer needs to get a + * irq-safe write-lock, but readers can get non-irqsafe + * read-locks. + */ + +/** + * read_can_lock - would read_trylock() succeed? + * @lock: the rwlock in question. + */ +#define arch_read_can_lock(x) ((int)(x)->lock >= 0) + +/** + * write_can_lock - would write_trylock() succeed? + * @lock: the rwlock in question. + */ +#define arch_write_can_lock(x) ((x)->lock == 0) + +extern int _raw_read_trylock_retry(arch_rwlock_t *lp); +extern int _raw_write_trylock_retry(arch_rwlock_t *lp); + +#define arch_read_lock_flags(lock, flags) arch_read_lock(lock) +#define arch_write_lock_flags(lock, flags) arch_write_lock(lock) + +static inline int arch_read_trylock_once(arch_rwlock_t *rw) +{ + unsigned int old = ACCESS_ONCE(rw->lock); + return likely((int) old >= 0 && + _raw_compare_and_swap(&rw->lock, old, old + 1)); +} + +static inline int arch_write_trylock_once(arch_rwlock_t *rw) +{ + unsigned int old = ACCESS_ONCE(rw->lock); + return likely(old == 0 && + _raw_compare_and_swap(&rw->lock, 0, 0x80000000)); +} + +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES + +#define __RAW_OP_OR "lao" +#define __RAW_OP_AND "lan" +#define __RAW_OP_ADD "laa" + +#define __RAW_LOCK(ptr, op_val, op_string) \ +({ \ + unsigned int old_val; \ + \ + typecheck(unsigned int *, ptr); \ + asm volatile( \ + op_string " %0,%2,%1\n" \ + "bcr 14,0\n" \ + : "=d" (old_val), "+Q" (*ptr) \ + : "d" (op_val) \ + : "cc", "memory"); \ + old_val; \ +}) + +#define __RAW_UNLOCK(ptr, op_val, op_string) \ +({ \ + unsigned int old_val; \ + \ + typecheck(unsigned int *, ptr); \ + asm volatile( \ + "bcr 14,0\n" \ + op_string " %0,%2,%1\n" \ + : "=d" (old_val), "+Q" (*ptr) \ + : "d" (op_val) \ + : "cc", "memory"); \ + old_val; \ +}) + +extern void _raw_read_lock_wait(arch_rwlock_t *lp); +extern void _raw_write_lock_wait(arch_rwlock_t *lp, unsigned int prev); + +static inline void arch_read_lock(arch_rwlock_t *rw) +{ + unsigned int old; + + old = __RAW_LOCK(&rw->lock, 1, __RAW_OP_ADD); + if ((int) old < 0) + _raw_read_lock_wait(rw); +} + +static inline void arch_read_unlock(arch_rwlock_t *rw) +{ + __RAW_UNLOCK(&rw->lock, -1, __RAW_OP_ADD); +} + +static inline void arch_write_lock(arch_rwlock_t *rw) +{ + unsigned int old; + + old = __RAW_LOCK(&rw->lock, 0x80000000, __RAW_OP_OR); + if (old != 0) + _raw_write_lock_wait(rw, old); + rw->owner = SPINLOCK_LOCKVAL; +} + +static inline void arch_write_unlock(arch_rwlock_t *rw) +{ + rw->owner = 0; + __RAW_UNLOCK(&rw->lock, 0x7fffffff, __RAW_OP_AND); +} + +#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +extern void _raw_read_lock_wait(arch_rwlock_t *lp); +extern void _raw_write_lock_wait(arch_rwlock_t *lp); + +static inline void arch_read_lock(arch_rwlock_t *rw) +{ + if (!arch_read_trylock_once(rw)) + _raw_read_lock_wait(rw); +} + +static inline void arch_read_unlock(arch_rwlock_t *rw) +{ + unsigned int old; + + do { + old = ACCESS_ONCE(rw->lock); + } while (!_raw_compare_and_swap(&rw->lock, old, old - 1)); +} + +static inline void arch_write_lock(arch_rwlock_t *rw) +{ + if (!arch_write_trylock_once(rw)) + _raw_write_lock_wait(rw); + rw->owner = SPINLOCK_LOCKVAL; +} + +static inline void arch_write_unlock(arch_rwlock_t *rw) +{ + typecheck(unsigned int, rw->lock); + + rw->owner = 0; + asm volatile( + __ASM_BARRIER + "st %1,%0\n" + : "+Q" (rw->lock) + : "d" (0) + : "cc", "memory"); +} + +#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +static inline int arch_read_trylock(arch_rwlock_t *rw) +{ + if (!arch_read_trylock_once(rw)) + return _raw_read_trylock_retry(rw); + return 1; +} + +static inline int arch_write_trylock(arch_rwlock_t *rw) +{ + if (!arch_write_trylock_once(rw) && !_raw_write_trylock_retry(rw)) + return 0; + rw->owner = SPINLOCK_LOCKVAL; + return 1; +} + +static inline void arch_read_relax(arch_rwlock_t *rw) +{ + arch_lock_relax(rw->owner); +} + +static inline void arch_write_relax(arch_rwlock_t *rw) +{ + arch_lock_relax(rw->owner); +} + +#endif /* __ASM_SPINLOCK_H */ diff --git a/arch/s390/include/asm/spinlock_types.h b/arch/s390/include/asm/spinlock_types.h new file mode 100644 index 000000000..d84b69392 --- /dev/null +++ b/arch/s390/include/asm/spinlock_types.h @@ -0,0 +1,21 @@ +#ifndef __ASM_SPINLOCK_TYPES_H +#define __ASM_SPINLOCK_TYPES_H + +#ifndef __LINUX_SPINLOCK_TYPES_H +# error "please don't include this file directly" +#endif + +typedef struct { + unsigned int lock; +} __attribute__ ((aligned (4))) arch_spinlock_t; + +#define __ARCH_SPIN_LOCK_UNLOCKED { .lock = 0, } + +typedef struct { + unsigned int lock; + unsigned int owner; +} arch_rwlock_t; + +#define __ARCH_RW_LOCK_UNLOCKED { 0 } + +#endif diff --git a/arch/s390/include/asm/string.h b/arch/s390/include/asm/string.h new file mode 100644 index 000000000..8662f5c8e --- /dev/null +++ b/arch/s390/include/asm/string.h @@ -0,0 +1,142 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + */ + +#ifndef _S390_STRING_H_ +#define _S390_STRING_H_ + +#ifndef _LINUX_TYPES_H +#include <linux/types.h> +#endif + +#define __HAVE_ARCH_MEMCHR /* inline & arch function */ +#define __HAVE_ARCH_MEMCMP /* arch function */ +#define __HAVE_ARCH_MEMCPY /* gcc builtin & arch function */ +#define __HAVE_ARCH_MEMSCAN /* inline & arch function */ +#define __HAVE_ARCH_MEMSET /* gcc builtin & arch function */ +#define __HAVE_ARCH_STRCAT /* inline & arch function */ +#define __HAVE_ARCH_STRCMP /* arch function */ +#define __HAVE_ARCH_STRCPY /* inline & arch function */ +#define __HAVE_ARCH_STRLCAT /* arch function */ +#define __HAVE_ARCH_STRLCPY /* arch function */ +#define __HAVE_ARCH_STRLEN /* inline & arch function */ +#define __HAVE_ARCH_STRNCAT /* arch function */ +#define __HAVE_ARCH_STRNCPY /* arch function */ +#define __HAVE_ARCH_STRNLEN /* inline & arch function */ +#define __HAVE_ARCH_STRRCHR /* arch function */ +#define __HAVE_ARCH_STRSTR /* arch function */ + +/* Prototypes for non-inlined arch strings functions. */ +extern int memcmp(const void *, const void *, size_t); +extern void *memcpy(void *, const void *, size_t); +extern void *memset(void *, int, size_t); +extern int strcmp(const char *,const char *); +extern size_t strlcat(char *, const char *, size_t); +extern size_t strlcpy(char *, const char *, size_t); +extern char *strncat(char *, const char *, size_t); +extern char *strncpy(char *, const char *, size_t); +extern char *strrchr(const char *, int); +extern char *strstr(const char *, const char *); + +#undef __HAVE_ARCH_MEMMOVE +#undef __HAVE_ARCH_STRCHR +#undef __HAVE_ARCH_STRNCHR +#undef __HAVE_ARCH_STRNCMP +#undef __HAVE_ARCH_STRPBRK +#undef __HAVE_ARCH_STRSEP +#undef __HAVE_ARCH_STRSPN + +#if !defined(IN_ARCH_STRING_C) + +static inline void *memchr(const void * s, int c, size_t n) +{ + register int r0 asm("0") = (char) c; + const void *ret = s + n; + + asm volatile( + "0: srst %0,%1\n" + " jo 0b\n" + " jl 1f\n" + " la %0,0\n" + "1:" + : "+a" (ret), "+&a" (s) : "d" (r0) : "cc"); + return (void *) ret; +} + +static inline void *memscan(void *s, int c, size_t n) +{ + register int r0 asm("0") = (char) c; + const void *ret = s + n; + + asm volatile( + "0: srst %0,%1\n" + " jo 0b\n" + : "+a" (ret), "+&a" (s) : "d" (r0) : "cc"); + return (void *) ret; +} + +static inline char *strcat(char *dst, const char *src) +{ + register int r0 asm("0") = 0; + unsigned long dummy; + char *ret = dst; + + asm volatile( + "0: srst %0,%1\n" + " jo 0b\n" + "1: mvst %0,%2\n" + " jo 1b" + : "=&a" (dummy), "+a" (dst), "+a" (src) + : "d" (r0), "0" (0) : "cc", "memory" ); + return ret; +} + +static inline char *strcpy(char *dst, const char *src) +{ + register int r0 asm("0") = 0; + char *ret = dst; + + asm volatile( + "0: mvst %0,%1\n" + " jo 0b" + : "+&a" (dst), "+&a" (src) : "d" (r0) + : "cc", "memory"); + return ret; +} + +static inline size_t strlen(const char *s) +{ + register unsigned long r0 asm("0") = 0; + const char *tmp = s; + + asm volatile( + "0: srst %0,%1\n" + " jo 0b" + : "+d" (r0), "+a" (tmp) : : "cc"); + return r0 - (unsigned long) s; +} + +static inline size_t strnlen(const char * s, size_t n) +{ + register int r0 asm("0") = 0; + const char *tmp = s; + const char *end = s + n; + + asm volatile( + "0: srst %0,%1\n" + " jo 0b" + : "+a" (end), "+a" (tmp) : "d" (r0) : "cc"); + return end - s; +} +#else /* IN_ARCH_STRING_C */ +void *memchr(const void * s, int c, size_t n); +void *memscan(void *s, int c, size_t n); +char *strcat(char *dst, const char *src); +char *strcpy(char *dst, const char *src); +size_t strlen(const char *s); +size_t strnlen(const char * s, size_t n); +#endif /* !IN_ARCH_STRING_C */ + +#endif /* __S390_STRING_H_ */ diff --git a/arch/s390/include/asm/switch_to.h b/arch/s390/include/asm/switch_to.h new file mode 100644 index 000000000..d62e7a696 --- /dev/null +++ b/arch/s390/include/asm/switch_to.h @@ -0,0 +1,175 @@ +/* + * Copyright IBM Corp. 1999, 2009 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef __ASM_SWITCH_TO_H +#define __ASM_SWITCH_TO_H + +#include <linux/thread_info.h> +#include <asm/ptrace.h> + +extern struct task_struct *__switch_to(void *, void *); +extern void update_cr_regs(struct task_struct *task); + +static inline int test_fp_ctl(u32 fpc) +{ + u32 orig_fpc; + int rc; + + asm volatile( + " efpc %1\n" + " sfpc %2\n" + "0: sfpc %1\n" + " la %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "=d" (rc), "=d" (orig_fpc) + : "d" (fpc), "0" (-EINVAL)); + return rc; +} + +static inline void save_fp_ctl(u32 *fpc) +{ + asm volatile( + " stfpc %0\n" + : "+Q" (*fpc)); +} + +static inline int restore_fp_ctl(u32 *fpc) +{ + int rc; + + asm volatile( + " lfpc %1\n" + "0: la %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "=d" (rc) : "Q" (*fpc), "0" (-EINVAL)); + return rc; +} + +static inline void save_fp_regs(freg_t *fprs) +{ + asm volatile("std 0,%0" : "=Q" (fprs[0])); + asm volatile("std 2,%0" : "=Q" (fprs[2])); + asm volatile("std 4,%0" : "=Q" (fprs[4])); + asm volatile("std 6,%0" : "=Q" (fprs[6])); + asm volatile("std 1,%0" : "=Q" (fprs[1])); + asm volatile("std 3,%0" : "=Q" (fprs[3])); + asm volatile("std 5,%0" : "=Q" (fprs[5])); + asm volatile("std 7,%0" : "=Q" (fprs[7])); + asm volatile("std 8,%0" : "=Q" (fprs[8])); + asm volatile("std 9,%0" : "=Q" (fprs[9])); + asm volatile("std 10,%0" : "=Q" (fprs[10])); + asm volatile("std 11,%0" : "=Q" (fprs[11])); + asm volatile("std 12,%0" : "=Q" (fprs[12])); + asm volatile("std 13,%0" : "=Q" (fprs[13])); + asm volatile("std 14,%0" : "=Q" (fprs[14])); + asm volatile("std 15,%0" : "=Q" (fprs[15])); +} + +static inline void restore_fp_regs(freg_t *fprs) +{ + asm volatile("ld 0,%0" : : "Q" (fprs[0])); + asm volatile("ld 2,%0" : : "Q" (fprs[2])); + asm volatile("ld 4,%0" : : "Q" (fprs[4])); + asm volatile("ld 6,%0" : : "Q" (fprs[6])); + asm volatile("ld 1,%0" : : "Q" (fprs[1])); + asm volatile("ld 3,%0" : : "Q" (fprs[3])); + asm volatile("ld 5,%0" : : "Q" (fprs[5])); + asm volatile("ld 7,%0" : : "Q" (fprs[7])); + asm volatile("ld 8,%0" : : "Q" (fprs[8])); + asm volatile("ld 9,%0" : : "Q" (fprs[9])); + asm volatile("ld 10,%0" : : "Q" (fprs[10])); + asm volatile("ld 11,%0" : : "Q" (fprs[11])); + asm volatile("ld 12,%0" : : "Q" (fprs[12])); + asm volatile("ld 13,%0" : : "Q" (fprs[13])); + asm volatile("ld 14,%0" : : "Q" (fprs[14])); + asm volatile("ld 15,%0" : : "Q" (fprs[15])); +} + +static inline void save_vx_regs(__vector128 *vxrs) +{ + typedef struct { __vector128 _[__NUM_VXRS]; } addrtype; + + asm volatile( + " la 1,%0\n" + " .word 0xe70f,0x1000,0x003e\n" /* vstm 0,15,0(1) */ + " .word 0xe70f,0x1100,0x0c3e\n" /* vstm 16,31,256(1) */ + : "=Q" (*(addrtype *) vxrs) : : "1"); +} + +static inline void save_vx_regs_safe(__vector128 *vxrs) +{ + unsigned long cr0, flags; + + flags = arch_local_irq_save(); + __ctl_store(cr0, 0, 0); + __ctl_set_bit(0, 17); + __ctl_set_bit(0, 18); + save_vx_regs(vxrs); + __ctl_load(cr0, 0, 0); + arch_local_irq_restore(flags); +} + +static inline void restore_vx_regs(__vector128 *vxrs) +{ + typedef struct { __vector128 _[__NUM_VXRS]; } addrtype; + + asm volatile( + " la 1,%0\n" + " .word 0xe70f,0x1000,0x0036\n" /* vlm 0,15,0(1) */ + " .word 0xe70f,0x1100,0x0c36\n" /* vlm 16,31,256(1) */ + : : "Q" (*(addrtype *) vxrs) : "1"); +} + +static inline void save_fp_vx_regs(struct task_struct *task) +{ + if (task->thread.vxrs) + save_vx_regs(task->thread.vxrs); + else + save_fp_regs(task->thread.fp_regs.fprs); +} + +static inline void restore_fp_vx_regs(struct task_struct *task) +{ + if (task->thread.vxrs) + restore_vx_regs(task->thread.vxrs); + else + restore_fp_regs(task->thread.fp_regs.fprs); +} + +static inline void save_access_regs(unsigned int *acrs) +{ + typedef struct { int _[NUM_ACRS]; } acrstype; + + asm volatile("stam 0,15,%0" : "=Q" (*(acrstype *)acrs)); +} + +static inline void restore_access_regs(unsigned int *acrs) +{ + typedef struct { int _[NUM_ACRS]; } acrstype; + + asm volatile("lam 0,15,%0" : : "Q" (*(acrstype *)acrs)); +} + +#define switch_to(prev,next,last) do { \ + if (prev->mm) { \ + save_fp_ctl(&prev->thread.fp_regs.fpc); \ + save_fp_vx_regs(prev); \ + save_access_regs(&prev->thread.acrs[0]); \ + save_ri_cb(prev->thread.ri_cb); \ + } \ + if (next->mm) { \ + update_cr_regs(next); \ + restore_fp_ctl(&next->thread.fp_regs.fpc); \ + restore_fp_vx_regs(next); \ + restore_access_regs(&next->thread.acrs[0]); \ + restore_ri_cb(next->thread.ri_cb, prev->thread.ri_cb); \ + } \ + prev = __switch_to(prev,next); \ +} while (0) + +#endif /* __ASM_SWITCH_TO_H */ diff --git a/arch/s390/include/asm/syscall.h b/arch/s390/include/asm/syscall.h new file mode 100644 index 000000000..6ba0bf928 --- /dev/null +++ b/arch/s390/include/asm/syscall.h @@ -0,0 +1,100 @@ +/* + * Access to user system call parameters and results + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.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. + */ + +#ifndef _ASM_SYSCALL_H +#define _ASM_SYSCALL_H 1 + +#include <uapi/linux/audit.h> +#include <linux/sched.h> +#include <linux/err.h> +#include <asm/ptrace.h> + +/* + * The syscall table always contains 32 bit pointers since we know that the + * address of the function to be called is (way) below 4GB. So the "int" + * type here is what we want [need] for both 32 bit and 64 bit systems. + */ +extern const unsigned int sys_call_table[]; +extern const unsigned int sys_call_table_emu[]; + +static inline long syscall_get_nr(struct task_struct *task, + struct pt_regs *regs) +{ + return test_pt_regs_flag(regs, PIF_SYSCALL) ? + (regs->int_code & 0xffff) : -1; +} + +static inline void syscall_rollback(struct task_struct *task, + struct pt_regs *regs) +{ + regs->gprs[2] = regs->orig_gpr2; +} + +static inline long syscall_get_error(struct task_struct *task, + struct pt_regs *regs) +{ + return IS_ERR_VALUE(regs->gprs[2]) ? regs->gprs[2] : 0; +} + +static inline long syscall_get_return_value(struct task_struct *task, + struct pt_regs *regs) +{ + return regs->gprs[2]; +} + +static inline void syscall_set_return_value(struct task_struct *task, + struct pt_regs *regs, + int error, long val) +{ + regs->gprs[2] = error ? error : val; +} + +static inline void syscall_get_arguments(struct task_struct *task, + struct pt_regs *regs, + unsigned int i, unsigned int n, + unsigned long *args) +{ + unsigned long mask = -1UL; + + BUG_ON(i + n > 6); +#ifdef CONFIG_COMPAT + if (test_tsk_thread_flag(task, TIF_31BIT)) + mask = 0xffffffff; +#endif + while (n-- > 0) + if (i + n > 0) + args[n] = regs->gprs[2 + i + n] & mask; + if (i == 0) + args[0] = regs->orig_gpr2 & mask; +} + +static inline void syscall_set_arguments(struct task_struct *task, + struct pt_regs *regs, + unsigned int i, unsigned int n, + const unsigned long *args) +{ + BUG_ON(i + n > 6); + while (n-- > 0) + if (i + n > 0) + regs->gprs[2 + i + n] = args[n]; + if (i == 0) + regs->orig_gpr2 = args[0]; +} + +static inline int syscall_get_arch(void) +{ +#ifdef CONFIG_COMPAT + if (test_tsk_thread_flag(current, TIF_31BIT)) + return AUDIT_ARCH_S390; +#endif + return AUDIT_ARCH_S390X; +} +#endif /* _ASM_SYSCALL_H */ diff --git a/arch/s390/include/asm/sysinfo.h b/arch/s390/include/asm/sysinfo.h new file mode 100644 index 000000000..f7054a892 --- /dev/null +++ b/arch/s390/include/asm/sysinfo.h @@ -0,0 +1,179 @@ +/* + * definition for store system information stsi + * + * Copyright IBM Corp. 2001, 2008 + * + * 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. + * + * Author(s): Ulrich Weigand <weigand@de.ibm.com> + * Christian Borntraeger <borntraeger@de.ibm.com> + */ + +#ifndef __ASM_S390_SYSINFO_H +#define __ASM_S390_SYSINFO_H + +#include <asm/bitsperlong.h> +#include <linux/uuid.h> + +struct sysinfo_1_1_1 { + unsigned char p:1; + unsigned char :6; + unsigned char t:1; + unsigned char :8; + unsigned char ccr; + unsigned char cai; + char reserved_0[28]; + char manufacturer[16]; + char type[4]; + char reserved_1[12]; + char model_capacity[16]; + char sequence[16]; + char plant[4]; + char model[16]; + char model_perm_cap[16]; + char model_temp_cap[16]; + unsigned int model_cap_rating; + unsigned int model_perm_cap_rating; + unsigned int model_temp_cap_rating; + unsigned char typepct[5]; + unsigned char reserved_2[3]; + unsigned int ncr; + unsigned int npr; + unsigned int ntr; +}; + +struct sysinfo_1_2_1 { + char reserved_0[80]; + char sequence[16]; + char plant[4]; + char reserved_1[2]; + unsigned short cpu_address; +}; + +struct sysinfo_1_2_2 { + char format; + char reserved_0[1]; + unsigned short acc_offset; + char reserved_1[20]; + unsigned int nominal_cap; + unsigned int secondary_cap; + unsigned int capability; + unsigned short cpus_total; + unsigned short cpus_configured; + unsigned short cpus_standby; + unsigned short cpus_reserved; + unsigned short adjustment[0]; +}; + +struct sysinfo_1_2_2_extension { + unsigned int alt_capability; + unsigned short alt_adjustment[0]; +}; + +struct sysinfo_2_2_1 { + char reserved_0[80]; + char sequence[16]; + char plant[4]; + unsigned short cpu_id; + unsigned short cpu_address; +}; + +struct sysinfo_2_2_2 { + char reserved_0[32]; + unsigned short lpar_number; + char reserved_1; + unsigned char characteristics; + unsigned short cpus_total; + unsigned short cpus_configured; + unsigned short cpus_standby; + unsigned short cpus_reserved; + char name[8]; + unsigned int caf; + char reserved_2[8]; + unsigned char mt_installed; + unsigned char mt_general; + unsigned char mt_psmtid; + char reserved_3[5]; + unsigned short cpus_dedicated; + unsigned short cpus_shared; +}; + +#define LPAR_CHAR_DEDICATED (1 << 7) +#define LPAR_CHAR_SHARED (1 << 6) +#define LPAR_CHAR_LIMITED (1 << 5) + +struct sysinfo_3_2_2 { + char reserved_0[31]; + unsigned char :4; + unsigned char count:4; + struct { + char reserved_0[4]; + unsigned short cpus_total; + unsigned short cpus_configured; + unsigned short cpus_standby; + unsigned short cpus_reserved; + char name[8]; + unsigned int caf; + char cpi[16]; + char reserved_1[3]; + char ext_name_encoding; + unsigned int reserved_2; + uuid_be uuid; + } vm[8]; + char reserved_3[1504]; + char ext_names[8][256]; +}; + +extern int topology_max_mnest; + +#define TOPOLOGY_CORE_BITS 64 +#define TOPOLOGY_NR_MAG 6 + +struct topology_core { + unsigned char nl; + unsigned char reserved0[3]; + unsigned char :6; + unsigned char pp:2; + unsigned char reserved1; + unsigned short origin; + unsigned long mask[TOPOLOGY_CORE_BITS / BITS_PER_LONG]; +}; + +struct topology_container { + unsigned char nl; + unsigned char reserved[6]; + unsigned char id; +}; + +union topology_entry { + unsigned char nl; + struct topology_core cpu; + struct topology_container container; +}; + +struct sysinfo_15_1_x { + unsigned char reserved0[2]; + unsigned short length; + unsigned char mag[TOPOLOGY_NR_MAG]; + unsigned char reserved1; + unsigned char mnest; + unsigned char reserved2[4]; + union topology_entry tle[0]; +}; + +int stsi(void *sysinfo, int fc, int sel1, int sel2); + +/* + * Service level reporting interface. + */ +struct service_level { + struct list_head list; + void (*seq_print)(struct seq_file *, struct service_level *); +}; + +int register_service_level(struct service_level *); +int unregister_service_level(struct service_level *); + +#endif /* __ASM_S390_SYSINFO_H */ diff --git a/arch/s390/include/asm/termios.h b/arch/s390/include/asm/termios.h new file mode 100644 index 000000000..db028d17f --- /dev/null +++ b/arch/s390/include/asm/termios.h @@ -0,0 +1,25 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/termios.h" + */ +#ifndef _S390_TERMIOS_H +#define _S390_TERMIOS_H + +#include <uapi/asm/termios.h> + + +/* intr=^C quit=^\ erase=del kill=^U + eof=^D vtime=\0 vmin=\1 sxtc=\0 + start=^Q stop=^S susp=^Z eol=\0 + reprint=^R discard=^U werase=^W lnext=^V + eol2=\0 +*/ +#define INIT_C_CC "\003\034\177\025\004\0\1\0\021\023\032\0\022\017\027\026\0" + +#define user_termios_to_kernel_termios(k, u) copy_from_user(k, u, sizeof(struct termios2)) +#define kernel_termios_to_user_termios(u, k) copy_to_user(u, k, sizeof(struct termios2)) + +#include <asm-generic/termios-base.h> + +#endif /* _S390_TERMIOS_H */ diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h new file mode 100644 index 000000000..4c27ec764 --- /dev/null +++ b/arch/s390/include/asm/thread_info.h @@ -0,0 +1,99 @@ +/* + * S390 version + * Copyright IBM Corp. 2002, 2006 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#ifndef _ASM_THREAD_INFO_H +#define _ASM_THREAD_INFO_H + +/* + * Size of kernel stack for each process + */ +#define THREAD_ORDER 2 +#define ASYNC_ORDER 2 + +#define THREAD_SIZE (PAGE_SIZE << THREAD_ORDER) +#define ASYNC_SIZE (PAGE_SIZE << ASYNC_ORDER) + +#ifndef __ASSEMBLY__ +#include <asm/lowcore.h> +#include <asm/page.h> +#include <asm/processor.h> + +/* + * low level task data that entry.S needs immediate access to + * - this struct should fit entirely inside of one cache line + * - this struct shares the supervisor stack pages + * - if the contents of this structure are changed, the assembly constants must also be changed + */ +struct thread_info { + struct task_struct *task; /* main task structure */ + unsigned long flags; /* low level flags */ + unsigned long sys_call_table; /* System call table address */ + unsigned int cpu; /* current CPU */ + int preempt_count; /* 0 => preemptable, <0 => BUG */ + unsigned int system_call; + __u64 user_timer; + __u64 system_timer; + unsigned long last_break; /* last breaking-event-address. */ +}; + +/* + * macros/functions for gaining access to the thread information structure + */ +#define INIT_THREAD_INFO(tsk) \ +{ \ + .task = &tsk, \ + .flags = 0, \ + .cpu = 0, \ + .preempt_count = INIT_PREEMPT_COUNT, \ +} + +#define init_thread_info (init_thread_union.thread_info) +#define init_stack (init_thread_union.stack) + +/* how to get the thread information struct from C */ +static inline struct thread_info *current_thread_info(void) +{ + return (struct thread_info *) S390_lowcore.thread_info; +} + +void arch_release_task_struct(struct task_struct *tsk); + +#define THREAD_SIZE_ORDER THREAD_ORDER + +#endif + +/* + * thread information flags bit numbers + */ +#define TIF_NOTIFY_RESUME 0 /* callback before returning to user */ +#define TIF_SIGPENDING 1 /* signal pending */ +#define TIF_NEED_RESCHED 2 /* rescheduling necessary */ +#define TIF_SYSCALL_TRACE 3 /* syscall trace active */ +#define TIF_SYSCALL_AUDIT 4 /* syscall auditing active */ +#define TIF_SECCOMP 5 /* secure computing */ +#define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ +#define TIF_UPROBE 7 /* breakpointed or single-stepping */ +#define TIF_31BIT 16 /* 32bit process */ +#define TIF_MEMDIE 17 /* is terminating due to OOM killer */ +#define TIF_RESTORE_SIGMASK 18 /* restore signal mask in do_signal() */ +#define TIF_SINGLE_STEP 19 /* This task is single stepped */ +#define TIF_BLOCK_STEP 20 /* This task is block stepped */ +#define TIF_UPROBE_SINGLESTEP 21 /* This task is uprobe single stepped */ + +#define _TIF_NOTIFY_RESUME (1<<TIF_NOTIFY_RESUME) +#define _TIF_SIGPENDING (1<<TIF_SIGPENDING) +#define _TIF_NEED_RESCHED (1<<TIF_NEED_RESCHED) +#define _TIF_SYSCALL_TRACE (1<<TIF_SYSCALL_TRACE) +#define _TIF_SYSCALL_AUDIT (1<<TIF_SYSCALL_AUDIT) +#define _TIF_SECCOMP (1<<TIF_SECCOMP) +#define _TIF_SYSCALL_TRACEPOINT (1<<TIF_SYSCALL_TRACEPOINT) +#define _TIF_UPROBE (1<<TIF_UPROBE) +#define _TIF_31BIT (1<<TIF_31BIT) +#define _TIF_SINGLE_STEP (1<<TIF_SINGLE_STEP) + +#define is_32bit_task() (test_thread_flag(TIF_31BIT)) + +#endif /* _ASM_THREAD_INFO_H */ diff --git a/arch/s390/include/asm/timex.h b/arch/s390/include/asm/timex.h new file mode 100644 index 000000000..98eb2a579 --- /dev/null +++ b/arch/s390/include/asm/timex.h @@ -0,0 +1,163 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * + * Derived from "include/asm-i386/timex.h" + * Copyright (C) 1992, Linus Torvalds + */ + +#ifndef _ASM_S390_TIMEX_H +#define _ASM_S390_TIMEX_H + +#include <asm/lowcore.h> + +/* The value of the TOD clock for 1.1.1970. */ +#define TOD_UNIX_EPOCH 0x7d91048bca000000ULL + +/* Inline functions for clock register access. */ +static inline int set_tod_clock(__u64 time) +{ + int cc; + + asm volatile( + " sck %1\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (cc) : "Q" (time) : "cc"); + return cc; +} + +static inline int store_tod_clock(__u64 *time) +{ + int cc; + + asm volatile( + " stck %1\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (cc), "=Q" (*time) : : "cc"); + return cc; +} + +static inline void set_clock_comparator(__u64 time) +{ + asm volatile("sckc %0" : : "Q" (time)); +} + +static inline void store_clock_comparator(__u64 *time) +{ + asm volatile("stckc %0" : "=Q" (*time)); +} + +void clock_comparator_work(void); + +static inline unsigned long long local_tick_disable(void) +{ + unsigned long long old; + + old = S390_lowcore.clock_comparator; + S390_lowcore.clock_comparator = -1ULL; + set_clock_comparator(S390_lowcore.clock_comparator); + return old; +} + +static inline void local_tick_enable(unsigned long long comp) +{ + S390_lowcore.clock_comparator = comp; + set_clock_comparator(S390_lowcore.clock_comparator); +} + +#define CLOCK_TICK_RATE 1193180 /* Underlying HZ */ +#define STORE_CLOCK_EXT_SIZE 16 /* stcke writes 16 bytes */ + +typedef unsigned long long cycles_t; + +static inline void get_tod_clock_ext(char *clk) +{ + typedef struct { char _[STORE_CLOCK_EXT_SIZE]; } addrtype; + + asm volatile("stcke %0" : "=Q" (*(addrtype *) clk) : : "cc"); +} + +static inline unsigned long long get_tod_clock(void) +{ + unsigned char clk[STORE_CLOCK_EXT_SIZE]; + + get_tod_clock_ext(clk); + return *((unsigned long long *)&clk[1]); +} + +static inline unsigned long long get_tod_clock_fast(void) +{ +#ifdef CONFIG_HAVE_MARCH_Z9_109_FEATURES + unsigned long long clk; + + asm volatile("stckf %0" : "=Q" (clk) : : "cc"); + return clk; +#else + return get_tod_clock(); +#endif +} + +static inline cycles_t get_cycles(void) +{ + return (cycles_t) get_tod_clock() >> 2; +} + +int get_sync_clock(unsigned long long *clock); +void init_cpu_timer(void); +unsigned long long monotonic_clock(void); + +void tod_to_timeval(__u64, struct timespec *); + +static inline +void stck_to_timespec(unsigned long long stck, struct timespec *ts) +{ + tod_to_timeval(stck - TOD_UNIX_EPOCH, ts); +} + +extern u64 sched_clock_base_cc; + +/** + * get_clock_monotonic - returns current time in clock rate units + * + * The caller must ensure that preemption is disabled. + * The clock and sched_clock_base get changed via stop_machine. + * Therefore preemption must be disabled when calling this + * function, otherwise the returned value is not guaranteed to + * be monotonic. + */ +static inline unsigned long long get_tod_clock_monotonic(void) +{ + return get_tod_clock() - sched_clock_base_cc; +} + +/** + * tod_to_ns - convert a TOD format value to nanoseconds + * @todval: to be converted TOD format value + * Returns: number of nanoseconds that correspond to the TOD format value + * + * Converting a 64 Bit TOD format value to nanoseconds means that the value + * must be divided by 4.096. In order to achieve that we multiply with 125 + * and divide by 512: + * + * ns = (todval * 125) >> 9; + * + * In order to avoid an overflow with the multiplication we can rewrite this. + * With a split todval == 2^32 * th + tl (th upper 32 bits, tl lower 32 bits) + * we end up with + * + * ns = ((2^32 * th + tl) * 125 ) >> 9; + * -> ns = (2^23 * th * 125) + ((tl * 125) >> 9); + * + */ +static inline unsigned long long tod_to_ns(unsigned long long todval) +{ + unsigned long long ns; + + ns = ((todval >> 32) << 23) * 125; + ns += ((todval & 0xffffffff) * 125) >> 9; + return ns; +} + +#endif diff --git a/arch/s390/include/asm/tlb.h b/arch/s390/include/asm/tlb.h new file mode 100644 index 000000000..7a92e69c5 --- /dev/null +++ b/arch/s390/include/asm/tlb.h @@ -0,0 +1,148 @@ +#ifndef _S390_TLB_H +#define _S390_TLB_H + +/* + * TLB flushing on s390 is complicated. The following requirement + * from the principles of operation is the most arduous: + * + * "A valid table entry must not be changed while it is attached + * to any CPU and may be used for translation by that CPU except to + * (1) invalidate the entry by using INVALIDATE PAGE TABLE ENTRY, + * or INVALIDATE DAT TABLE ENTRY, (2) alter bits 56-63 of a page + * table entry, or (3) make a change by means of a COMPARE AND SWAP + * AND PURGE instruction that purges the TLB." + * + * The modification of a pte of an active mm struct therefore is + * a two step process: i) invalidate the pte, ii) store the new pte. + * This is true for the page protection bit as well. + * The only possible optimization is to flush at the beginning of + * a tlb_gather_mmu cycle if the mm_struct is currently not in use. + * + * Pages used for the page tables is a different story. FIXME: more + */ + +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/swap.h> +#include <asm/processor.h> +#include <asm/pgalloc.h> +#include <asm/tlbflush.h> + +struct mmu_gather { + struct mm_struct *mm; + struct mmu_table_batch *batch; + unsigned int fullmm; + unsigned long start, end; +}; + +struct mmu_table_batch { + struct rcu_head rcu; + unsigned int nr; + void *tables[0]; +}; + +#define MAX_TABLE_BATCH \ + ((PAGE_SIZE - sizeof(struct mmu_table_batch)) / sizeof(void *)) + +extern void tlb_table_flush(struct mmu_gather *tlb); +extern void tlb_remove_table(struct mmu_gather *tlb, void *table); + +static inline void tlb_gather_mmu(struct mmu_gather *tlb, + struct mm_struct *mm, + unsigned long start, + unsigned long end) +{ + tlb->mm = mm; + tlb->start = start; + tlb->end = end; + tlb->fullmm = !(start | (end+1)); + tlb->batch = NULL; +} + +static inline void tlb_flush_mmu_tlbonly(struct mmu_gather *tlb) +{ + __tlb_flush_mm_lazy(tlb->mm); +} + +static inline void tlb_flush_mmu_free(struct mmu_gather *tlb) +{ + tlb_table_flush(tlb); +} + + +static inline void tlb_flush_mmu(struct mmu_gather *tlb) +{ + tlb_flush_mmu_tlbonly(tlb); + tlb_flush_mmu_free(tlb); +} + +static inline void tlb_finish_mmu(struct mmu_gather *tlb, + unsigned long start, unsigned long end) +{ + tlb_flush_mmu(tlb); +} + +/* + * Release the page cache reference for a pte removed by + * tlb_ptep_clear_flush. In both flush modes the tlb for a page cache page + * has already been freed, so just do free_page_and_swap_cache. + */ +static inline int __tlb_remove_page(struct mmu_gather *tlb, struct page *page) +{ + free_page_and_swap_cache(page); + return 1; /* avoid calling tlb_flush_mmu */ +} + +static inline void tlb_remove_page(struct mmu_gather *tlb, struct page *page) +{ + free_page_and_swap_cache(page); +} + +/* + * pte_free_tlb frees a pte table and clears the CRSTE for the + * page table from the tlb. + */ +static inline void pte_free_tlb(struct mmu_gather *tlb, pgtable_t pte, + unsigned long address) +{ + page_table_free_rcu(tlb, (unsigned long *) pte, address); +} + +/* + * pmd_free_tlb frees a pmd table and clears the CRSTE for the + * segment table entry from the tlb. + * If the mm uses a two level page table the single pmd is freed + * as the pgd. pmd_free_tlb checks the asce_limit against 2GB + * to avoid the double free of the pmd in this case. + */ +static inline void pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd, + unsigned long address) +{ + if (tlb->mm->context.asce_limit <= (1UL << 31)) + return; + pgtable_pmd_page_dtor(virt_to_page(pmd)); + tlb_remove_table(tlb, pmd); +} + +/* + * pud_free_tlb frees a pud table and clears the CRSTE for the + * region third table entry from the tlb. + * If the mm uses a three level page table the single pud is freed + * as the pgd. pud_free_tlb checks the asce_limit against 4TB + * to avoid the double free of the pud in this case. + */ +static inline void pud_free_tlb(struct mmu_gather *tlb, pud_t *pud, + unsigned long address) +{ + if (tlb->mm->context.asce_limit <= (1UL << 42)) + return; + tlb_remove_table(tlb, pud); +} + +#define tlb_start_vma(tlb, vma) do { } while (0) +#define tlb_end_vma(tlb, vma) do { } while (0) +#define tlb_remove_tlb_entry(tlb, ptep, addr) do { } while (0) +#define tlb_remove_pmd_tlb_entry(tlb, pmdp, addr) do { } while (0) +#define tlb_migrate_finish(mm) do { } while (0) + +#endif /* _S390_TLB_H */ diff --git a/arch/s390/include/asm/tlbflush.h b/arch/s390/include/asm/tlbflush.h new file mode 100644 index 000000000..ca148f7c3 --- /dev/null +++ b/arch/s390/include/asm/tlbflush.h @@ -0,0 +1,204 @@ +#ifndef _S390_TLBFLUSH_H +#define _S390_TLBFLUSH_H + +#include <linux/mm.h> +#include <linux/sched.h> +#include <asm/processor.h> +#include <asm/pgalloc.h> + +/* + * Flush all TLB entries on the local CPU. + */ +static inline void __tlb_flush_local(void) +{ + asm volatile("ptlb" : : : "memory"); +} + +/* + * Flush TLB entries for a specific ASCE on all CPUs + */ +static inline void __tlb_flush_idte(unsigned long asce) +{ + /* Global TLB flush for the mm */ + asm volatile( + " .insn rrf,0xb98e0000,0,%0,%1,0" + : : "a" (2048), "a" (asce) : "cc"); +} + +/* + * Flush TLB entries for a specific ASCE on the local CPU + */ +static inline void __tlb_flush_idte_local(unsigned long asce) +{ + /* Local TLB flush for the mm */ + asm volatile( + " .insn rrf,0xb98e0000,0,%0,%1,1" + : : "a" (2048), "a" (asce) : "cc"); +} + +#ifdef CONFIG_SMP +void smp_ptlb_all(void); + +/* + * Flush all TLB entries on all CPUs. + */ +static inline void __tlb_flush_global(void) +{ + register unsigned long reg2 asm("2"); + register unsigned long reg3 asm("3"); + register unsigned long reg4 asm("4"); + long dummy; + + dummy = 0; + reg2 = reg3 = 0; + reg4 = ((unsigned long) &dummy) + 1; + asm volatile( + " csp %0,%2" + : : "d" (reg2), "d" (reg3), "d" (reg4), "m" (dummy) : "cc" ); +} + +/* + * Flush TLB entries for a specific mm on all CPUs (in case gmap is used + * this implicates multiple ASCEs!). + */ +static inline void __tlb_flush_full(struct mm_struct *mm) +{ + preempt_disable(); + atomic_add(0x10000, &mm->context.attach_count); + if (cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) { + /* Local TLB flush */ + __tlb_flush_local(); + } else { + /* Global TLB flush */ + __tlb_flush_global(); + /* Reset TLB flush mask */ + if (MACHINE_HAS_TLB_LC) + cpumask_copy(mm_cpumask(mm), + &mm->context.cpu_attach_mask); + } + atomic_sub(0x10000, &mm->context.attach_count); + preempt_enable(); +} + +/* + * Flush TLB entries for a specific ASCE on all CPUs. + */ +static inline void __tlb_flush_asce(struct mm_struct *mm, unsigned long asce) +{ + int active, count; + + preempt_disable(); + active = (mm == current->active_mm) ? 1 : 0; + count = atomic_add_return(0x10000, &mm->context.attach_count); + if (MACHINE_HAS_TLB_LC && (count & 0xffff) <= active && + cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) { + __tlb_flush_idte_local(asce); + } else { + if (MACHINE_HAS_IDTE) + __tlb_flush_idte(asce); + else + __tlb_flush_global(); + /* Reset TLB flush mask */ + if (MACHINE_HAS_TLB_LC) + cpumask_copy(mm_cpumask(mm), + &mm->context.cpu_attach_mask); + } + atomic_sub(0x10000, &mm->context.attach_count); + preempt_enable(); +} + +static inline void __tlb_flush_kernel(void) +{ + if (MACHINE_HAS_IDTE) + __tlb_flush_idte((unsigned long) init_mm.pgd | + init_mm.context.asce_bits); + else + __tlb_flush_global(); +} +#else +#define __tlb_flush_global() __tlb_flush_local() +#define __tlb_flush_full(mm) __tlb_flush_local() + +/* + * Flush TLB entries for a specific ASCE on all CPUs. + */ +static inline void __tlb_flush_asce(struct mm_struct *mm, unsigned long asce) +{ + if (MACHINE_HAS_TLB_LC) + __tlb_flush_idte_local(asce); + else + __tlb_flush_local(); +} + +static inline void __tlb_flush_kernel(void) +{ + if (MACHINE_HAS_TLB_LC) + __tlb_flush_idte_local((unsigned long) init_mm.pgd | + init_mm.context.asce_bits); + else + __tlb_flush_local(); +} +#endif + +static inline void __tlb_flush_mm(struct mm_struct * mm) +{ + /* + * If the machine has IDTE we prefer to do a per mm flush + * on all cpus instead of doing a local flush if the mm + * only ran on the local cpu. + */ + if (MACHINE_HAS_IDTE && list_empty(&mm->context.gmap_list)) + __tlb_flush_asce(mm, (unsigned long) mm->pgd | + mm->context.asce_bits); + else + __tlb_flush_full(mm); +} + +static inline void __tlb_flush_mm_lazy(struct mm_struct * mm) +{ + if (mm->context.flush_mm) { + __tlb_flush_mm(mm); + mm->context.flush_mm = 0; + } +} + +/* + * TLB flushing: + * flush_tlb() - flushes the current mm struct TLBs + * flush_tlb_all() - flushes all processes TLBs + * flush_tlb_mm(mm) - flushes the specified mm context TLB's + * flush_tlb_page(vma, vmaddr) - flushes one page + * flush_tlb_range(vma, start, end) - flushes a range of pages + * flush_tlb_kernel_range(start, end) - flushes a range of kernel pages + */ + +/* + * flush_tlb_mm goes together with ptep_set_wrprotect for the + * copy_page_range operation and flush_tlb_range is related to + * ptep_get_and_clear for change_protection. ptep_set_wrprotect and + * ptep_get_and_clear do not flush the TLBs directly if the mm has + * only one user. At the end of the update the flush_tlb_mm and + * flush_tlb_range functions need to do the flush. + */ +#define flush_tlb() do { } while (0) +#define flush_tlb_all() do { } while (0) +#define flush_tlb_page(vma, addr) do { } while (0) + +static inline void flush_tlb_mm(struct mm_struct *mm) +{ + __tlb_flush_mm_lazy(mm); +} + +static inline void flush_tlb_range(struct vm_area_struct *vma, + unsigned long start, unsigned long end) +{ + __tlb_flush_mm_lazy(vma->vm_mm); +} + +static inline void flush_tlb_kernel_range(unsigned long start, + unsigned long end) +{ + __tlb_flush_kernel(); +} + +#endif /* _S390_TLBFLUSH_H */ diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h new file mode 100644 index 000000000..b1453a2ae --- /dev/null +++ b/arch/s390/include/asm/topology.h @@ -0,0 +1,56 @@ +#ifndef _ASM_S390_TOPOLOGY_H +#define _ASM_S390_TOPOLOGY_H + +#include <linux/cpumask.h> + +struct sysinfo_15_1_x; +struct cpu; + +#ifdef CONFIG_SCHED_BOOK + +struct cpu_topology_s390 { + unsigned short thread_id; + unsigned short core_id; + unsigned short socket_id; + unsigned short book_id; + cpumask_t thread_mask; + cpumask_t core_mask; + cpumask_t book_mask; +}; + +DECLARE_PER_CPU(struct cpu_topology_s390, cpu_topology); + +#define topology_physical_package_id(cpu) (per_cpu(cpu_topology, cpu).socket_id) +#define topology_thread_id(cpu) (per_cpu(cpu_topology, cpu).thread_id) +#define topology_thread_cpumask(cpu) (&per_cpu(cpu_topology, cpu).thread_mask) +#define topology_core_id(cpu) (per_cpu(cpu_topology, cpu).core_id) +#define topology_core_cpumask(cpu) (&per_cpu(cpu_topology, cpu).core_mask) +#define topology_book_id(cpu) (per_cpu(cpu_topology, cpu).book_id) +#define topology_book_cpumask(cpu) (&per_cpu(cpu_topology, cpu).book_mask) + +#define mc_capable() 1 + +int topology_cpu_init(struct cpu *); +int topology_set_cpu_management(int fc); +void topology_schedule_update(void); +void store_topology(struct sysinfo_15_1_x *info); +void topology_expect_change(void); +const struct cpumask *cpu_coregroup_mask(int cpu); + +#else /* CONFIG_SCHED_BOOK */ + +static inline void topology_schedule_update(void) { } +static inline int topology_cpu_init(struct cpu *cpu) { return 0; } +static inline void topology_expect_change(void) { } + +#endif /* CONFIG_SCHED_BOOK */ + +#define POLARIZATION_UNKNOWN (-1) +#define POLARIZATION_HRZ (0) +#define POLARIZATION_VL (1) +#define POLARIZATION_VM (2) +#define POLARIZATION_VH (3) + +#include <asm-generic/topology.h> + +#endif /* _ASM_S390_TOPOLOGY_H */ diff --git a/arch/s390/include/asm/types.h b/arch/s390/include/asm/types.h new file mode 100644 index 000000000..6740f4f97 --- /dev/null +++ b/arch/s390/include/asm/types.h @@ -0,0 +1,11 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/types.h" + */ +#ifndef _S390_TYPES_H +#define _S390_TYPES_H + +#include <uapi/asm/types.h> + +#endif /* _S390_TYPES_H */ diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h new file mode 100644 index 000000000..d64a7a621 --- /dev/null +++ b/arch/s390/include/asm/uaccess.h @@ -0,0 +1,377 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + * Author(s): Hartmut Penner (hp@de.ibm.com), + * Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "include/asm-i386/uaccess.h" + */ +#ifndef __S390_UACCESS_H +#define __S390_UACCESS_H + +/* + * User space memory access functions + */ +#include <linux/sched.h> +#include <linux/errno.h> +#include <asm/ctl_reg.h> + +#define VERIFY_READ 0 +#define VERIFY_WRITE 1 + + +/* + * The fs value determines whether argument validity checking should be + * performed or not. If get_fs() == USER_DS, checking is performed, with + * get_fs() == KERNEL_DS, checking is bypassed. + * + * For historical reasons, these macros are grossly misnamed. + */ + +#define MAKE_MM_SEG(a) ((mm_segment_t) { (a) }) + + +#define KERNEL_DS MAKE_MM_SEG(0) +#define USER_DS MAKE_MM_SEG(1) + +#define get_ds() (KERNEL_DS) +#define get_fs() (current->thread.mm_segment) + +#define set_fs(x) \ +({ \ + unsigned long __pto; \ + current->thread.mm_segment = (x); \ + __pto = current->thread.mm_segment.ar4 ? \ + S390_lowcore.user_asce : S390_lowcore.kernel_asce; \ + __ctl_load(__pto, 7, 7); \ +}) + +#define segment_eq(a,b) ((a).ar4 == (b).ar4) + +static inline int __range_ok(unsigned long addr, unsigned long size) +{ + return 1; +} + +#define __access_ok(addr, size) \ +({ \ + __chk_user_ptr(addr); \ + __range_ok((unsigned long)(addr), (size)); \ +}) + +#define access_ok(type, addr, size) __access_ok(addr, size) + +/* + * The exception table consists of pairs of addresses: the first is the + * address of an instruction that is allowed to fault, and the second is + * the address at which the program should continue. No registers are + * modified, so it is entirely up to the continuation code to figure out + * what to do. + * + * All the routines below use bits of fixup code that are out of line + * with the main instruction path. This means when everything is well, + * we don't even have to jump over them. Further, they do not intrude + * on our cache or tlb entries. + */ + +struct exception_table_entry +{ + int insn, fixup; +}; + +static inline unsigned long extable_insn(const struct exception_table_entry *x) +{ + return (unsigned long)&x->insn + x->insn; +} + +static inline unsigned long extable_fixup(const struct exception_table_entry *x) +{ + return (unsigned long)&x->fixup + x->fixup; +} + +#define ARCH_HAS_SORT_EXTABLE +#define ARCH_HAS_SEARCH_EXTABLE + +/** + * __copy_from_user: - Copy a block of data from user space, with less checking. + * @to: Destination address, in kernel space. + * @from: Source address, in user space. + * @n: Number of bytes to copy. + * + * Context: User context only. This function may sleep. + * + * Copy data from user space to kernel space. Caller must check + * the specified block with access_ok() before calling this function. + * + * Returns number of bytes that could not be copied. + * On success, this will be zero. + * + * If some data could not be copied, this function will pad the copied + * data to the requested size using zero bytes. + */ +unsigned long __must_check __copy_from_user(void *to, const void __user *from, + unsigned long n); + +/** + * __copy_to_user: - Copy a block of data into user space, with less checking. + * @to: Destination address, in user space. + * @from: Source address, in kernel space. + * @n: Number of bytes to copy. + * + * Context: User context only. This function may sleep. + * + * Copy data from kernel space to user space. Caller must check + * the specified block with access_ok() before calling this function. + * + * Returns number of bytes that could not be copied. + * On success, this will be zero. + */ +unsigned long __must_check __copy_to_user(void __user *to, const void *from, + unsigned long n); + +#define __copy_to_user_inatomic __copy_to_user +#define __copy_from_user_inatomic __copy_from_user + +#ifdef CONFIG_HAVE_MARCH_Z10_FEATURES + +#define __put_get_user_asm(to, from, size, spec) \ +({ \ + register unsigned long __reg0 asm("0") = spec; \ + int __rc; \ + \ + asm volatile( \ + "0: mvcos %1,%3,%2\n" \ + "1: xr %0,%0\n" \ + "2:\n" \ + ".pushsection .fixup, \"ax\"\n" \ + "3: lhi %0,%5\n" \ + " jg 2b\n" \ + ".popsection\n" \ + EX_TABLE(0b,3b) EX_TABLE(1b,3b) \ + : "=d" (__rc), "=Q" (*(to)) \ + : "d" (size), "Q" (*(from)), \ + "d" (__reg0), "K" (-EFAULT) \ + : "cc"); \ + __rc; \ +}) + +#define __put_user_fn(x, ptr, size) __put_get_user_asm(ptr, x, size, 0x810000UL) +#define __get_user_fn(x, ptr, size) __put_get_user_asm(x, ptr, size, 0x81UL) + +#else /* CONFIG_HAVE_MARCH_Z10_FEATURES */ + +static inline int __put_user_fn(void *x, void __user *ptr, unsigned long size) +{ + size = __copy_to_user(ptr, x, size); + return size ? -EFAULT : 0; +} + +static inline int __get_user_fn(void *x, const void __user *ptr, unsigned long size) +{ + size = __copy_from_user(x, ptr, size); + return size ? -EFAULT : 0; +} + +#endif /* CONFIG_HAVE_MARCH_Z10_FEATURES */ + +/* + * These are the main single-value transfer routines. They automatically + * use the right size if we just have the right pointer type. + */ +#define __put_user(x, ptr) \ +({ \ + __typeof__(*(ptr)) __x = (x); \ + int __pu_err = -EFAULT; \ + __chk_user_ptr(ptr); \ + switch (sizeof (*(ptr))) { \ + case 1: \ + case 2: \ + case 4: \ + case 8: \ + __pu_err = __put_user_fn(&__x, ptr, \ + sizeof(*(ptr))); \ + break; \ + default: \ + __put_user_bad(); \ + break; \ + } \ + __pu_err; \ +}) + +#define put_user(x, ptr) \ +({ \ + might_fault(); \ + __put_user(x, ptr); \ +}) + + +int __put_user_bad(void) __attribute__((noreturn)); + +#define __get_user(x, ptr) \ +({ \ + int __gu_err = -EFAULT; \ + __chk_user_ptr(ptr); \ + switch (sizeof(*(ptr))) { \ + case 1: { \ + unsigned char __x; \ + __gu_err = __get_user_fn(&__x, ptr, \ + sizeof(*(ptr))); \ + (x) = *(__force __typeof__(*(ptr)) *) &__x; \ + break; \ + }; \ + case 2: { \ + unsigned short __x; \ + __gu_err = __get_user_fn(&__x, ptr, \ + sizeof(*(ptr))); \ + (x) = *(__force __typeof__(*(ptr)) *) &__x; \ + break; \ + }; \ + case 4: { \ + unsigned int __x; \ + __gu_err = __get_user_fn(&__x, ptr, \ + sizeof(*(ptr))); \ + (x) = *(__force __typeof__(*(ptr)) *) &__x; \ + break; \ + }; \ + case 8: { \ + unsigned long long __x; \ + __gu_err = __get_user_fn(&__x, ptr, \ + sizeof(*(ptr))); \ + (x) = *(__force __typeof__(*(ptr)) *) &__x; \ + break; \ + }; \ + default: \ + __get_user_bad(); \ + break; \ + } \ + __gu_err; \ +}) + +#define get_user(x, ptr) \ +({ \ + might_fault(); \ + __get_user(x, ptr); \ +}) + +int __get_user_bad(void) __attribute__((noreturn)); + +#define __put_user_unaligned __put_user +#define __get_user_unaligned __get_user + +/** + * copy_to_user: - Copy a block of data into user space. + * @to: Destination address, in user space. + * @from: Source address, in kernel space. + * @n: Number of bytes to copy. + * + * Context: User context only. This function may sleep. + * + * Copy data from kernel space to user space. + * + * Returns number of bytes that could not be copied. + * On success, this will be zero. + */ +static inline unsigned long __must_check +copy_to_user(void __user *to, const void *from, unsigned long n) +{ + might_fault(); + return __copy_to_user(to, from, n); +} + +void copy_from_user_overflow(void) +#ifdef CONFIG_DEBUG_STRICT_USER_COPY_CHECKS +__compiletime_warning("copy_from_user() buffer size is not provably correct") +#endif +; + +/** + * copy_from_user: - Copy a block of data from user space. + * @to: Destination address, in kernel space. + * @from: Source address, in user space. + * @n: Number of bytes to copy. + * + * Context: User context only. This function may sleep. + * + * Copy data from user space to kernel space. + * + * Returns number of bytes that could not be copied. + * On success, this will be zero. + * + * If some data could not be copied, this function will pad the copied + * data to the requested size using zero bytes. + */ +static inline unsigned long __must_check +copy_from_user(void *to, const void __user *from, unsigned long n) +{ + unsigned int sz = __compiletime_object_size(to); + + might_fault(); + if (unlikely(sz != -1 && sz < n)) { + copy_from_user_overflow(); + return n; + } + return __copy_from_user(to, from, n); +} + +unsigned long __must_check +__copy_in_user(void __user *to, const void __user *from, unsigned long n); + +static inline unsigned long __must_check +copy_in_user(void __user *to, const void __user *from, unsigned long n) +{ + might_fault(); + return __copy_in_user(to, from, n); +} + +/* + * Copy a null terminated string from userspace. + */ + +long __strncpy_from_user(char *dst, const char __user *src, long count); + +static inline long __must_check +strncpy_from_user(char *dst, const char __user *src, long count) +{ + might_fault(); + return __strncpy_from_user(dst, src, count); +} + +unsigned long __must_check __strnlen_user(const char __user *src, unsigned long count); + +static inline unsigned long strnlen_user(const char __user *src, unsigned long n) +{ + might_fault(); + return __strnlen_user(src, n); +} + +/** + * strlen_user: - Get the size of a string in user space. + * @str: The string to measure. + * + * Context: User context only. This function may sleep. + * + * Get the size of a NUL-terminated string in user space. + * + * Returns the size of the string INCLUDING the terminating NUL. + * On exception, returns 0. + * + * If there is a limit on the length of a valid string, you may wish to + * consider using strnlen_user() instead. + */ +#define strlen_user(str) strnlen_user(str, ~0UL) + +/* + * Zero Userspace + */ +unsigned long __must_check __clear_user(void __user *to, unsigned long size); + +static inline unsigned long __must_check clear_user(void __user *to, unsigned long n) +{ + might_fault(); + return __clear_user(to, n); +} + +int copy_to_user_real(void __user *dest, void *src, unsigned long count); +void s390_kernel_write(void *dst, const void *src, size_t size); + +#endif /* __S390_UACCESS_H */ diff --git a/arch/s390/include/asm/unaligned.h b/arch/s390/include/asm/unaligned.h new file mode 100644 index 000000000..da9627afe --- /dev/null +++ b/arch/s390/include/asm/unaligned.h @@ -0,0 +1,13 @@ +#ifndef _ASM_S390_UNALIGNED_H +#define _ASM_S390_UNALIGNED_H + +/* + * The S390 can do unaligned accesses itself. + */ +#include <linux/unaligned/access_ok.h> +#include <linux/unaligned/generic.h> + +#define get_unaligned __get_unaligned_be +#define put_unaligned __put_unaligned_be + +#endif /* _ASM_S390_UNALIGNED_H */ diff --git a/arch/s390/include/asm/unistd.h b/arch/s390/include/asm/unistd.h new file mode 100644 index 000000000..91f56b1d8 --- /dev/null +++ b/arch/s390/include/asm/unistd.h @@ -0,0 +1,49 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/unistd.h" + */ +#ifndef _ASM_S390_UNISTD_H_ +#define _ASM_S390_UNISTD_H_ + +#include <uapi/asm/unistd.h> + + +#define __IGNORE_time + +/* Ignore NUMA system calls. Not wired up on s390. */ +#define __IGNORE_mbind +#define __IGNORE_get_mempolicy +#define __IGNORE_set_mempolicy +#define __IGNORE_migrate_pages +#define __IGNORE_move_pages + +/* Ignore system calls that are also reachable via sys_socket */ +#define __IGNORE_recvmmsg +#define __IGNORE_sendmmsg + +#define __ARCH_WANT_OLD_READDIR +#define __ARCH_WANT_SYS_ALARM +#define __ARCH_WANT_SYS_GETHOSTNAME +#define __ARCH_WANT_SYS_PAUSE +#define __ARCH_WANT_SYS_SIGNAL +#define __ARCH_WANT_SYS_UTIME +#define __ARCH_WANT_SYS_SOCKETCALL +#define __ARCH_WANT_SYS_IPC +#define __ARCH_WANT_SYS_FADVISE64 +#define __ARCH_WANT_SYS_GETPGRP +#define __ARCH_WANT_SYS_LLSEEK +#define __ARCH_WANT_SYS_NICE +#define __ARCH_WANT_SYS_OLD_GETRLIMIT +#define __ARCH_WANT_SYS_OLD_MMAP +#define __ARCH_WANT_SYS_OLDUMOUNT +#define __ARCH_WANT_SYS_SIGPENDING +#define __ARCH_WANT_SYS_SIGPROCMASK +# ifdef CONFIG_COMPAT +# define __ARCH_WANT_COMPAT_SYS_TIME +# endif +#define __ARCH_WANT_SYS_FORK +#define __ARCH_WANT_SYS_VFORK +#define __ARCH_WANT_SYS_CLONE + +#endif /* _ASM_S390_UNISTD_H_ */ diff --git a/arch/s390/include/asm/uprobes.h b/arch/s390/include/asm/uprobes.h new file mode 100644 index 000000000..1411dff7f --- /dev/null +++ b/arch/s390/include/asm/uprobes.h @@ -0,0 +1,42 @@ +/* + * User-space Probes (UProbes) for s390 + * + * Copyright IBM Corp. 2014 + * Author(s): Jan Willeke, + */ + +#ifndef _ASM_UPROBES_H +#define _ASM_UPROBES_H + +#include <linux/notifier.h> + +typedef u16 uprobe_opcode_t; + +#define UPROBE_XOL_SLOT_BYTES 256 /* cache aligned */ + +#define UPROBE_SWBP_INSN 0x0002 +#define UPROBE_SWBP_INSN_SIZE 2 + +struct arch_uprobe { + union{ + uprobe_opcode_t insn[3]; + uprobe_opcode_t ixol[3]; + }; + unsigned int saved_per : 1; + unsigned int saved_int_code; +}; + +struct arch_uprobe_task { +}; + +int arch_uprobe_analyze_insn(struct arch_uprobe *aup, struct mm_struct *mm, + unsigned long addr); +int arch_uprobe_pre_xol(struct arch_uprobe *aup, struct pt_regs *regs); +int arch_uprobe_post_xol(struct arch_uprobe *aup, struct pt_regs *regs); +bool arch_uprobe_xol_was_trapped(struct task_struct *tsk); +int arch_uprobe_exception_notify(struct notifier_block *self, unsigned long val, + void *data); +void arch_uprobe_abort_xol(struct arch_uprobe *ap, struct pt_regs *regs); +unsigned long arch_uretprobe_hijack_return_addr(unsigned long trampoline, + struct pt_regs *regs); +#endif /* _ASM_UPROBES_H */ diff --git a/arch/s390/include/asm/user.h b/arch/s390/include/asm/user.h new file mode 100644 index 000000000..6ed1d1886 --- /dev/null +++ b/arch/s390/include/asm/user.h @@ -0,0 +1,74 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/usr.h" + */ + +#ifndef _S390_USER_H +#define _S390_USER_H + +#include <asm/page.h> +#include <asm/ptrace.h> +/* Core file format: The core file is written in such a way that gdb + can understand it and provide useful information to the user (under + linux we use the 'trad-core' bfd). There are quite a number of + obstacles to being able to view the contents of the floating point + registers, and until these are solved you will not be able to view the + contents of them. Actually, you can read in the core file and look at + the contents of the user struct to find out what the floating point + registers contain. + The actual file contents are as follows: + UPAGE: 1 page consisting of a user struct that tells gdb what is present + in the file. Directly after this is a copy of the task_struct, which + is currently not used by gdb, but it may come in useful at some point. + All of the registers are stored as part of the upage. The upage should + always be only one page. + DATA: The data area is stored. We use current->end_text to + current->brk to pick up all of the user variables, plus any memory + that may have been malloced. No attempt is made to determine if a page + is demand-zero or if a page is totally unused, we just cover the entire + range. All of the addresses are rounded in such a way that an integral + number of pages is written. + STACK: We need the stack information in order to get a meaningful + backtrace. We need to write the data from (esp) to + current->start_stack, so we round each of these off in order to be able + to write an integer number of pages. + The minimum core file size is 3 pages, or 12288 bytes. +*/ + + +/* + * This is the old layout of "struct pt_regs", and + * is still the layout used by user mode (the new + * pt_regs doesn't have all registers as the kernel + * doesn't use the extra segment registers) + */ + +/* When the kernel dumps core, it starts by dumping the user struct - + this will be used by gdb to figure out where the data and stack segments + are within the file, and what virtual addresses to use. */ +struct user { +/* We start with the registers, to mimic the way that "memory" is returned + from the ptrace(3,...) function. */ + struct user_regs_struct regs; /* Where the registers are actually stored */ +/* The rest of this junk is to help gdb figure out what goes where */ + unsigned long int u_tsize; /* Text segment size (pages). */ + unsigned long int u_dsize; /* Data segment size (pages). */ + unsigned long int u_ssize; /* Stack segment size (pages). */ + unsigned long start_code; /* Starting virtual address of text. */ + unsigned long start_stack; /* Starting virtual address of stack area. + This is actually the bottom of the stack, + the top of the stack is always found in the + esp register. */ + long int signal; /* Signal that caused the core dump. */ + unsigned long u_ar0; /* Used by gdb to help find the values for */ + /* the registers. */ + unsigned long magic; /* To uniquely identify a core file */ + char u_comm[32]; /* User command that was responsible */ +}; +#define NBPG PAGE_SIZE +#define UPAGES 1 +#define HOST_TEXT_START_ADDR (u.start_code) +#define HOST_STACK_END_ADDR (u.start_stack + u.u_ssize * NBPG) + +#endif /* _S390_USER_H */ diff --git a/arch/s390/include/asm/vdso.h b/arch/s390/include/asm/vdso.h new file mode 100644 index 000000000..787acd4f9 --- /dev/null +++ b/arch/s390/include/asm/vdso.h @@ -0,0 +1,49 @@ +#ifndef __S390_VDSO_H__ +#define __S390_VDSO_H__ + +/* Default link addresses for the vDSOs */ +#define VDSO32_LBASE 0 +#define VDSO64_LBASE 0 + +#define VDSO_VERSION_STRING LINUX_2.6.29 + +#ifndef __ASSEMBLY__ + +/* + * Note about the vdso_data and vdso_per_cpu_data structures: + * + * NEVER USE THEM IN USERSPACE CODE DIRECTLY. The layout of the + * structure is supposed to be known only to the function in the vdso + * itself and may change without notice. + */ + +struct vdso_data { + __u64 tb_update_count; /* Timebase atomicity ctr 0x00 */ + __u64 xtime_tod_stamp; /* TOD clock for xtime 0x08 */ + __u64 xtime_clock_sec; /* Kernel time 0x10 */ + __u64 xtime_clock_nsec; /* 0x18 */ + __u64 xtime_coarse_sec; /* Coarse kernel time 0x20 */ + __u64 xtime_coarse_nsec; /* 0x28 */ + __u64 wtom_clock_sec; /* Wall to monotonic clock 0x30 */ + __u64 wtom_clock_nsec; /* 0x38 */ + __u64 wtom_coarse_sec; /* Coarse wall to monotonic 0x40 */ + __u64 wtom_coarse_nsec; /* 0x48 */ + __u32 tz_minuteswest; /* Minutes west of Greenwich 0x50 */ + __u32 tz_dsttime; /* Type of dst correction 0x54 */ + __u32 ectg_available; /* ECTG instruction present 0x58 */ + __u32 tk_mult; /* Mult. used for xtime_nsec 0x5c */ + __u32 tk_shift; /* Shift used for xtime_nsec 0x60 */ +}; + +struct vdso_per_cpu_data { + __u64 ectg_timer_base; + __u64 ectg_user_time; +}; + +extern struct vdso_data *vdso_data; + +int vdso_alloc_per_cpu(struct _lowcore *lowcore); +void vdso_free_per_cpu(struct _lowcore *lowcore); + +#endif /* __ASSEMBLY__ */ +#endif /* __S390_VDSO_H__ */ diff --git a/arch/s390/include/asm/vga.h b/arch/s390/include/asm/vga.h new file mode 100644 index 000000000..d375526c2 --- /dev/null +++ b/arch/s390/include/asm/vga.h @@ -0,0 +1,6 @@ +#ifndef _ASM_S390_VGA_H +#define _ASM_S390_VGA_H + +/* Avoid compile errors due to missing asm/vga.h */ + +#endif /* _ASM_S390_VGA_H */ diff --git a/arch/s390/include/asm/vtime.h b/arch/s390/include/asm/vtime.h new file mode 100644 index 000000000..af9896c53 --- /dev/null +++ b/arch/s390/include/asm/vtime.h @@ -0,0 +1,7 @@ +#ifndef _S390_VTIME_H +#define _S390_VTIME_H + +#define __ARCH_HAS_VTIME_ACCOUNT +#define __ARCH_HAS_VTIME_TASK_SWITCH + +#endif /* _S390_VTIME_H */ diff --git a/arch/s390/include/asm/vtimer.h b/arch/s390/include/asm/vtimer.h new file mode 100644 index 000000000..10a179af6 --- /dev/null +++ b/arch/s390/include/asm/vtimer.h @@ -0,0 +1,31 @@ +/* + * Copyright IBM Corp. 2003, 2012 + * Virtual CPU timer + * + * Author(s): Jan Glauber <jan.glauber@de.ibm.com> + */ + +#ifndef _ASM_S390_TIMER_H +#define _ASM_S390_TIMER_H + +#define VTIMER_MAX_SLICE (0x7fffffffffffffffULL) + +struct vtimer_list { + struct list_head entry; + u64 expires; + u64 interval; + void (*function)(unsigned long); + unsigned long data; +}; + +extern void init_virt_timer(struct vtimer_list *timer); +extern void add_virt_timer(struct vtimer_list *timer); +extern void add_virt_timer_periodic(struct vtimer_list *timer); +extern int mod_virt_timer(struct vtimer_list *timer, u64 expires); +extern int mod_virt_timer_periodic(struct vtimer_list *timer, u64 expires); +extern int del_virt_timer(struct vtimer_list *timer); + +extern void init_cpu_vtimer(void); +extern void vtime_init(void); + +#endif /* _ASM_S390_TIMER_H */ diff --git a/arch/s390/include/asm/xor.h b/arch/s390/include/asm/xor.h new file mode 100644 index 000000000..c82eb12a5 --- /dev/null +++ b/arch/s390/include/asm/xor.h @@ -0,0 +1 @@ +#include <asm-generic/xor.h> diff --git a/arch/s390/include/uapi/asm/Kbuild b/arch/s390/include/uapi/asm/Kbuild new file mode 100644 index 000000000..08fe6dad9 --- /dev/null +++ b/arch/s390/include/uapi/asm/Kbuild @@ -0,0 +1,52 @@ +# UAPI Header export list +include include/uapi/asm-generic/Kbuild.asm + +header-y += auxvec.h +header-y += bitsperlong.h +header-y += byteorder.h +header-y += chpid.h +header-y += chsc.h +header-y += cmb.h +header-y += dasd.h +header-y += debug.h +header-y += errno.h +header-y += fcntl.h +header-y += ioctl.h +header-y += ioctls.h +header-y += ipcbuf.h +header-y += kvm.h +header-y += kvm_para.h +header-y += kvm_perf.h +header-y += kvm_virtio.h +header-y += mman.h +header-y += monwriter.h +header-y += msgbuf.h +header-y += param.h +header-y += poll.h +header-y += posix_types.h +header-y += ptrace.h +header-y += qeth.h +header-y += resource.h +header-y += schid.h +header-y += sembuf.h +header-y += setup.h +header-y += shmbuf.h +header-y += sigcontext.h +header-y += siginfo.h +header-y += signal.h +header-y += socket.h +header-y += sockios.h +header-y += sclp_ctl.h +header-y += sie.h +header-y += stat.h +header-y += statfs.h +header-y += swab.h +header-y += tape390.h +header-y += termbits.h +header-y += termios.h +header-y += types.h +header-y += ucontext.h +header-y += unistd.h +header-y += virtio-ccw.h +header-y += vtoc.h +header-y += zcrypt.h diff --git a/arch/s390/include/uapi/asm/auxvec.h b/arch/s390/include/uapi/asm/auxvec.h new file mode 100644 index 000000000..a1f153e89 --- /dev/null +++ b/arch/s390/include/uapi/asm/auxvec.h @@ -0,0 +1,6 @@ +#ifndef __ASMS390_AUXVEC_H +#define __ASMS390_AUXVEC_H + +#define AT_SYSINFO_EHDR 33 + +#endif diff --git a/arch/s390/include/uapi/asm/bitsperlong.h b/arch/s390/include/uapi/asm/bitsperlong.h new file mode 100644 index 000000000..6b235aea9 --- /dev/null +++ b/arch/s390/include/uapi/asm/bitsperlong.h @@ -0,0 +1,13 @@ +#ifndef __ASM_S390_BITSPERLONG_H +#define __ASM_S390_BITSPERLONG_H + +#ifndef __s390x__ +#define __BITS_PER_LONG 32 +#else +#define __BITS_PER_LONG 64 +#endif + +#include <asm-generic/bitsperlong.h> + +#endif /* __ASM_S390_BITSPERLONG_H */ + diff --git a/arch/s390/include/uapi/asm/byteorder.h b/arch/s390/include/uapi/asm/byteorder.h new file mode 100644 index 000000000..a332e59e2 --- /dev/null +++ b/arch/s390/include/uapi/asm/byteorder.h @@ -0,0 +1,6 @@ +#ifndef _S390_BYTEORDER_H +#define _S390_BYTEORDER_H + +#include <linux/byteorder/big_endian.h> + +#endif /* _S390_BYTEORDER_H */ diff --git a/arch/s390/include/uapi/asm/chpid.h b/arch/s390/include/uapi/asm/chpid.h new file mode 100644 index 000000000..6b4fb29cc --- /dev/null +++ b/arch/s390/include/uapi/asm/chpid.h @@ -0,0 +1,22 @@ +/* + * Copyright IBM Corp. 2007, 2012 + * Author(s): Peter Oberparleiter <peter.oberparleiter@de.ibm.com> + */ + +#ifndef _UAPI_ASM_S390_CHPID_H +#define _UAPI_ASM_S390_CHPID_H + +#include <linux/string.h> +#include <linux/types.h> + +#define __MAX_CHPID 255 + +struct chp_id { + __u8 reserved1; + __u8 cssid; + __u8 reserved2; + __u8 id; +} __attribute__((packed)); + + +#endif /* _UAPI_ASM_S390_CHPID_H */ diff --git a/arch/s390/include/uapi/asm/chsc.h b/arch/s390/include/uapi/asm/chsc.h new file mode 100644 index 000000000..65dc69472 --- /dev/null +++ b/arch/s390/include/uapi/asm/chsc.h @@ -0,0 +1,143 @@ +/* + * ioctl interface for /dev/chsc + * + * Copyright IBM Corp. 2008, 2012 + * Author(s): Cornelia Huck <cornelia.huck@de.ibm.com> + */ + +#ifndef _ASM_CHSC_H +#define _ASM_CHSC_H + +#include <linux/types.h> +#include <linux/ioctl.h> +#include <asm/chpid.h> +#include <asm/schid.h> + +#define CHSC_SIZE 0x1000 + +struct chsc_async_header { + __u16 length; + __u16 code; + __u32 cmd_dependend; + __u32 key : 4; + __u32 : 28; + struct subchannel_id sid; +} __attribute__ ((packed)); + +struct chsc_async_area { + struct chsc_async_header header; + __u8 data[CHSC_SIZE - sizeof(struct chsc_async_header)]; +} __attribute__ ((packed)); + +struct chsc_header { + __u16 length; + __u16 code; +} __attribute__ ((packed)); + +struct chsc_sync_area { + struct chsc_header header; + __u8 data[CHSC_SIZE - sizeof(struct chsc_header)]; +} __attribute__ ((packed)); + +struct chsc_response_struct { + __u16 length; + __u16 code; + __u32 parms; + __u8 data[CHSC_SIZE - 2 * sizeof(__u16) - sizeof(__u32)]; +} __attribute__ ((packed)); + +struct chsc_chp_cd { + struct chp_id chpid; + int m; + int fmt; + struct chsc_response_struct cpcb; +}; + +struct chsc_cu_cd { + __u16 cun; + __u8 cssid; + int m; + int fmt; + struct chsc_response_struct cucb; +}; + +struct chsc_sch_cud { + struct subchannel_id schid; + int fmt; + struct chsc_response_struct scub; +}; + +struct conf_id { + int m; + __u8 cssid; + __u8 ssid; +}; + +struct chsc_conf_info { + struct conf_id id; + int fmt; + struct chsc_response_struct scid; +}; + +struct ccl_parm_chpid { + int m; + struct chp_id chp; +}; + +struct ccl_parm_cssids { + __u8 f_cssid; + __u8 l_cssid; +}; + +struct chsc_comp_list { + struct { + enum { + CCL_CU_ON_CHP = 1, + CCL_CHP_TYPE_CAP = 2, + CCL_CSS_IMG = 4, + CCL_CSS_IMG_CONF_CHAR = 5, + CCL_IOP_CHP = 6, + } ctype; + int fmt; + struct ccl_parm_chpid chpid; + struct ccl_parm_cssids cssids; + } req; + struct chsc_response_struct sccl; +}; + +struct chsc_dcal { + struct { + enum { + DCAL_CSS_IID_PN = 4, + } atype; + __u32 list_parm[2]; + int fmt; + } req; + struct chsc_response_struct sdcal; +}; + +struct chsc_cpd_info { + struct chp_id chpid; + int m; + int fmt; + int rfmt; + int c; + struct chsc_response_struct chpdb; +}; + +#define CHSC_IOCTL_MAGIC 'c' + +#define CHSC_START _IOWR(CHSC_IOCTL_MAGIC, 0x81, struct chsc_async_area) +#define CHSC_INFO_CHANNEL_PATH _IOWR(CHSC_IOCTL_MAGIC, 0x82, \ + struct chsc_chp_cd) +#define CHSC_INFO_CU _IOWR(CHSC_IOCTL_MAGIC, 0x83, struct chsc_cu_cd) +#define CHSC_INFO_SCH_CU _IOWR(CHSC_IOCTL_MAGIC, 0x84, struct chsc_sch_cud) +#define CHSC_INFO_CI _IOWR(CHSC_IOCTL_MAGIC, 0x85, struct chsc_conf_info) +#define CHSC_INFO_CCL _IOWR(CHSC_IOCTL_MAGIC, 0x86, struct chsc_comp_list) +#define CHSC_INFO_CPD _IOWR(CHSC_IOCTL_MAGIC, 0x87, struct chsc_cpd_info) +#define CHSC_INFO_DCAL _IOWR(CHSC_IOCTL_MAGIC, 0x88, struct chsc_dcal) +#define CHSC_START_SYNC _IOWR(CHSC_IOCTL_MAGIC, 0x89, struct chsc_sync_area) +#define CHSC_ON_CLOSE_SET _IOWR(CHSC_IOCTL_MAGIC, 0x8a, struct chsc_async_area) +#define CHSC_ON_CLOSE_REMOVE _IO(CHSC_IOCTL_MAGIC, 0x8b) + +#endif diff --git a/arch/s390/include/uapi/asm/cmb.h b/arch/s390/include/uapi/asm/cmb.h new file mode 100644 index 000000000..0c086d00d --- /dev/null +++ b/arch/s390/include/uapi/asm/cmb.h @@ -0,0 +1,53 @@ +#ifndef _UAPIS390_CMB_H +#define _UAPIS390_CMB_H + +#include <linux/types.h> + +/** + * struct cmbdata - channel measurement block data for user space + * @size: size of the stored data + * @elapsed_time: time since last sampling + * @ssch_rsch_count: number of ssch and rsch + * @sample_count: number of samples + * @device_connect_time: time of device connect + * @function_pending_time: time of function pending + * @device_disconnect_time: time of device disconnect + * @control_unit_queuing_time: time of control unit queuing + * @device_active_only_time: time of device active only + * @device_busy_time: time of device busy (ext. format) + * @initial_command_response_time: initial command response time (ext. format) + * + * All values are stored as 64 bit for simplicity, especially + * in 32 bit emulation mode. All time values are normalized to + * nanoseconds. + * Currently, two formats are known, which differ by the size of + * this structure, i.e. the last two members are only set when + * the extended channel measurement facility (first shipped in + * z990 machines) is activated. + * Potentially, more fields could be added, which would result in a + * new ioctl number. + */ +struct cmbdata { + __u64 size; + __u64 elapsed_time; + /* basic and exended format: */ + __u64 ssch_rsch_count; + __u64 sample_count; + __u64 device_connect_time; + __u64 function_pending_time; + __u64 device_disconnect_time; + __u64 control_unit_queuing_time; + __u64 device_active_only_time; + /* extended format only: */ + __u64 device_busy_time; + __u64 initial_command_response_time; +}; + +/* enable channel measurement */ +#define BIODASDCMFENABLE _IO(DASD_IOCTL_LETTER, 32) +/* enable channel measurement */ +#define BIODASDCMFDISABLE _IO(DASD_IOCTL_LETTER, 33) +/* read channel measurement data */ +#define BIODASDREADALLCMB _IOWR(DASD_IOCTL_LETTER, 33, struct cmbdata) + +#endif /* _UAPIS390_CMB_H */ diff --git a/arch/s390/include/uapi/asm/dasd.h b/arch/s390/include/uapi/asm/dasd.h new file mode 100644 index 000000000..5812a3b2d --- /dev/null +++ b/arch/s390/include/uapi/asm/dasd.h @@ -0,0 +1,295 @@ +/* + * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com> + * Bugreports.to..: <Linux390@de.ibm.com> + * Copyright IBM Corp. 1999, 2000 + * EMC Symmetrix ioctl Copyright EMC Corporation, 2008 + * Author.........: Nigel Hislop <hislop_nigel@emc.com> + * + * This file is the interface of the DASD device driver, which is exported to user space + * any future changes wrt the API will result in a change of the APIVERSION reported + * to userspace by the DASDAPIVER-ioctl + * + */ + +#ifndef DASD_H +#define DASD_H +#include <linux/types.h> +#include <linux/ioctl.h> + +#define DASD_IOCTL_LETTER 'D' + +#define DASD_API_VERSION 6 + +/* + * struct dasd_information2_t + * represents any data about the device, which is visible to userspace. + * including foramt and featueres. + */ +typedef struct dasd_information2_t { + unsigned int devno; /* S/390 devno */ + unsigned int real_devno; /* for aliases */ + unsigned int schid; /* S/390 subchannel identifier */ + unsigned int cu_type : 16; /* from SenseID */ + unsigned int cu_model : 8; /* from SenseID */ + unsigned int dev_type : 16; /* from SenseID */ + unsigned int dev_model : 8; /* from SenseID */ + unsigned int open_count; + unsigned int req_queue_len; + unsigned int chanq_len; /* length of chanq */ + char type[4]; /* from discipline.name, 'none' for unknown */ + unsigned int status; /* current device level */ + unsigned int label_block; /* where to find the VOLSER */ + unsigned int FBA_layout; /* fixed block size (like AIXVOL) */ + unsigned int characteristics_size; + unsigned int confdata_size; + char characteristics[64]; /* from read_device_characteristics */ + char configuration_data[256]; /* from read_configuration_data */ + unsigned int format; /* format info like formatted/cdl/ldl/... */ + unsigned int features; /* dasd features like 'ro',... */ + unsigned int reserved0; /* reserved for further use ,... */ + unsigned int reserved1; /* reserved for further use ,... */ + unsigned int reserved2; /* reserved for further use ,... */ + unsigned int reserved3; /* reserved for further use ,... */ + unsigned int reserved4; /* reserved for further use ,... */ + unsigned int reserved5; /* reserved for further use ,... */ + unsigned int reserved6; /* reserved for further use ,... */ + unsigned int reserved7; /* reserved for further use ,... */ +} dasd_information2_t; + +/* + * values to be used for dasd_information_t.format + * 0x00: NOT formatted + * 0x01: Linux disc layout + * 0x02: Common disc layout + */ +#define DASD_FORMAT_NONE 0 +#define DASD_FORMAT_LDL 1 +#define DASD_FORMAT_CDL 2 +/* + * values to be used for dasd_information_t.features + * 0x00: default features + * 0x01: readonly (ro) + * 0x02: use diag discipline (diag) + * 0x04: set the device initially online (internal use only) + * 0x08: enable ERP related logging + * 0x20: give access to raw eckd data + */ +#define DASD_FEATURE_DEFAULT 0x00 +#define DASD_FEATURE_READONLY 0x01 +#define DASD_FEATURE_USEDIAG 0x02 +#define DASD_FEATURE_INITIAL_ONLINE 0x04 +#define DASD_FEATURE_ERPLOG 0x08 +#define DASD_FEATURE_FAILFAST 0x10 +#define DASD_FEATURE_FAILONSLCK 0x20 +#define DASD_FEATURE_USERAW 0x40 + +#define DASD_PARTN_BITS 2 + +/* + * struct dasd_information_t + * represents any data about the data, which is visible to userspace + */ +typedef struct dasd_information_t { + unsigned int devno; /* S/390 devno */ + unsigned int real_devno; /* for aliases */ + unsigned int schid; /* S/390 subchannel identifier */ + unsigned int cu_type : 16; /* from SenseID */ + unsigned int cu_model : 8; /* from SenseID */ + unsigned int dev_type : 16; /* from SenseID */ + unsigned int dev_model : 8; /* from SenseID */ + unsigned int open_count; + unsigned int req_queue_len; + unsigned int chanq_len; /* length of chanq */ + char type[4]; /* from discipline.name, 'none' for unknown */ + unsigned int status; /* current device level */ + unsigned int label_block; /* where to find the VOLSER */ + unsigned int FBA_layout; /* fixed block size (like AIXVOL) */ + unsigned int characteristics_size; + unsigned int confdata_size; + char characteristics[64]; /* from read_device_characteristics */ + char configuration_data[256]; /* from read_configuration_data */ +} dasd_information_t; + +/* + * Read Subsystem Data - Performance Statistics + */ +typedef struct dasd_rssd_perf_stats_t { + unsigned char invalid:1; + unsigned char format:3; + unsigned char data_format:4; + unsigned char unit_address; + unsigned short device_status; + unsigned int nr_read_normal; + unsigned int nr_read_normal_hits; + unsigned int nr_write_normal; + unsigned int nr_write_fast_normal_hits; + unsigned int nr_read_seq; + unsigned int nr_read_seq_hits; + unsigned int nr_write_seq; + unsigned int nr_write_fast_seq_hits; + unsigned int nr_read_cache; + unsigned int nr_read_cache_hits; + unsigned int nr_write_cache; + unsigned int nr_write_fast_cache_hits; + unsigned int nr_inhibit_cache; + unsigned int nr_bybass_cache; + unsigned int nr_seq_dasd_to_cache; + unsigned int nr_dasd_to_cache; + unsigned int nr_cache_to_dasd; + unsigned int nr_delayed_fast_write; + unsigned int nr_normal_fast_write; + unsigned int nr_seq_fast_write; + unsigned int nr_cache_miss; + unsigned char status2; + unsigned int nr_quick_write_promotes; + unsigned char reserved; + unsigned short ssid; + unsigned char reseved2[96]; +} __attribute__((packed)) dasd_rssd_perf_stats_t; + +/* + * struct profile_info_t + * holds the profinling information + */ +typedef struct dasd_profile_info_t { + unsigned int dasd_io_reqs; /* number of requests processed at all */ + unsigned int dasd_io_sects; /* number of sectors processed at all */ + unsigned int dasd_io_secs[32]; /* histogram of request's sizes */ + unsigned int dasd_io_times[32]; /* histogram of requests's times */ + unsigned int dasd_io_timps[32]; /* histogram of requests's times per sector */ + unsigned int dasd_io_time1[32]; /* histogram of time from build to start */ + unsigned int dasd_io_time2[32]; /* histogram of time from start to irq */ + unsigned int dasd_io_time2ps[32]; /* histogram of time from start to irq */ + unsigned int dasd_io_time3[32]; /* histogram of time from irq to end */ + unsigned int dasd_io_nr_req[32]; /* histogram of # of requests in chanq */ +} dasd_profile_info_t; + +/* + * struct format_data_t + * represents all data necessary to format a dasd + */ +typedef struct format_data_t { + unsigned int start_unit; /* from track */ + unsigned int stop_unit; /* to track */ + unsigned int blksize; /* sectorsize */ + unsigned int intensity; +} format_data_t; + +/* + * values to be used for format_data_t.intensity + * 0/8: normal format + * 1/9: also write record zero + * 3/11: also write home address + * 4/12: invalidate track + */ +#define DASD_FMT_INT_FMT_R0 1 /* write record zero */ +#define DASD_FMT_INT_FMT_HA 2 /* write home address, also set FMT_R0 ! */ +#define DASD_FMT_INT_INVAL 4 /* invalidate tracks */ +#define DASD_FMT_INT_COMPAT 8 /* use OS/390 compatible disk layout */ + + +/* + * struct attrib_data_t + * represents the operation (cache) bits for the device. + * Used in DE to influence caching of the DASD. + */ +typedef struct attrib_data_t { + unsigned char operation:3; /* cache operation mode */ + unsigned char reserved:5; /* cache operation mode */ + __u16 nr_cyl; /* no of cyliners for read ahaed */ + __u8 reserved2[29]; /* for future use */ +} __attribute__ ((packed)) attrib_data_t; + +/* definition of operation (cache) bits within attributes of DE */ +#define DASD_NORMAL_CACHE 0x0 +#define DASD_BYPASS_CACHE 0x1 +#define DASD_INHIBIT_LOAD 0x2 +#define DASD_SEQ_ACCESS 0x3 +#define DASD_SEQ_PRESTAGE 0x4 +#define DASD_REC_ACCESS 0x5 + +/* + * Perform EMC Symmetrix I/O + */ +typedef struct dasd_symmio_parms { + unsigned char reserved[8]; /* compat with older releases */ + unsigned long long psf_data; /* char * cast to u64 */ + unsigned long long rssd_result; /* char * cast to u64 */ + int psf_data_len; + int rssd_result_len; +} __attribute__ ((packed)) dasd_symmio_parms_t; + +/* + * Data returned by Sense Path Group ID (SNID) + */ +struct dasd_snid_data { + struct { + __u8 group:2; + __u8 reserve:2; + __u8 mode:1; + __u8 res:3; + } __attribute__ ((packed)) path_state; + __u8 pgid[11]; +} __attribute__ ((packed)); + +struct dasd_snid_ioctl_data { + struct dasd_snid_data data; + __u8 path_mask; +} __attribute__ ((packed)); + + +/******************************************************************************** + * SECTION: Definition of IOCTLs + * + * Here ist how the ioctl-nr should be used: + * 0 - 31 DASD driver itself + * 32 - 239 still open + * 240 - 255 reserved for EMC + *******************************************************************************/ + +/* Disable the volume (for Linux) */ +#define BIODASDDISABLE _IO(DASD_IOCTL_LETTER,0) +/* Enable the volume (for Linux) */ +#define BIODASDENABLE _IO(DASD_IOCTL_LETTER,1) +/* Issue a reserve/release command, rsp. */ +#define BIODASDRSRV _IO(DASD_IOCTL_LETTER,2) /* reserve */ +#define BIODASDRLSE _IO(DASD_IOCTL_LETTER,3) /* release */ +#define BIODASDSLCK _IO(DASD_IOCTL_LETTER,4) /* steal lock */ +/* reset profiling information of a device */ +#define BIODASDPRRST _IO(DASD_IOCTL_LETTER,5) +/* Quiesce IO on device */ +#define BIODASDQUIESCE _IO(DASD_IOCTL_LETTER,6) +/* Resume IO on device */ +#define BIODASDRESUME _IO(DASD_IOCTL_LETTER,7) +/* Abort all I/O on a device */ +#define BIODASDABORTIO _IO(DASD_IOCTL_LETTER, 240) +/* Allow I/O on a device */ +#define BIODASDALLOWIO _IO(DASD_IOCTL_LETTER, 241) + + +/* retrieve API version number */ +#define DASDAPIVER _IOR(DASD_IOCTL_LETTER,0,int) +/* Get information on a dasd device */ +#define BIODASDINFO _IOR(DASD_IOCTL_LETTER,1,dasd_information_t) +/* retrieve profiling information of a device */ +#define BIODASDPRRD _IOR(DASD_IOCTL_LETTER,2,dasd_profile_info_t) +/* Get information on a dasd device (enhanced) */ +#define BIODASDINFO2 _IOR(DASD_IOCTL_LETTER,3,dasd_information2_t) +/* Performance Statistics Read */ +#define BIODASDPSRD _IOR(DASD_IOCTL_LETTER,4,dasd_rssd_perf_stats_t) +/* Get Attributes (cache operations) */ +#define BIODASDGATTR _IOR(DASD_IOCTL_LETTER,5,attrib_data_t) + + +/* #define BIODASDFORMAT _IOW(IOCTL_LETTER,0,format_data_t) , deprecated */ +#define BIODASDFMT _IOW(DASD_IOCTL_LETTER,1,format_data_t) +/* Set Attributes (cache operations) */ +#define BIODASDSATTR _IOW(DASD_IOCTL_LETTER,2,attrib_data_t) + +/* Get Sense Path Group ID (SNID) data */ +#define BIODASDSNID _IOWR(DASD_IOCTL_LETTER, 1, struct dasd_snid_ioctl_data) + +#define BIODASDSYMMIO _IOWR(DASD_IOCTL_LETTER, 240, dasd_symmio_parms_t) + +#endif /* DASD_H */ + diff --git a/arch/s390/include/uapi/asm/debug.h b/arch/s390/include/uapi/asm/debug.h new file mode 100644 index 000000000..c59fc7912 --- /dev/null +++ b/arch/s390/include/uapi/asm/debug.h @@ -0,0 +1,34 @@ +/* + * S/390 debug facility + * + * Copyright IBM Corp. 1999, 2000 + */ + +#ifndef _UAPIDEBUG_H +#define _UAPIDEBUG_H + +#include <linux/fs.h> + +/* Note: + * struct __debug_entry must be defined outside of #ifdef __KERNEL__ + * in order to allow a user program to analyze the 'raw'-view. + */ + +struct __debug_entry{ + union { + struct { + unsigned long long clock:52; + unsigned long long exception:1; + unsigned long long level:3; + unsigned long long cpuid:8; + } fields; + + unsigned long long stck; + } id; + void* caller; +} __attribute__((packed)); + + +#define __DEBUG_FEATURE_VERSION 2 /* version of debug feature */ + +#endif /* _UAPIDEBUG_H */ diff --git a/arch/s390/include/uapi/asm/errno.h b/arch/s390/include/uapi/asm/errno.h new file mode 100644 index 000000000..395e97d80 --- /dev/null +++ b/arch/s390/include/uapi/asm/errno.h @@ -0,0 +1,11 @@ +/* + * S390 version + * + */ + +#ifndef _S390_ERRNO_H +#define _S390_ERRNO_H + +#include <asm-generic/errno.h> + +#endif diff --git a/arch/s390/include/uapi/asm/fcntl.h b/arch/s390/include/uapi/asm/fcntl.h new file mode 100644 index 000000000..46ab12db5 --- /dev/null +++ b/arch/s390/include/uapi/asm/fcntl.h @@ -0,0 +1 @@ +#include <asm-generic/fcntl.h> diff --git a/arch/s390/include/uapi/asm/hypfs.h b/arch/s390/include/uapi/asm/hypfs.h new file mode 100644 index 000000000..b3fe12d8d --- /dev/null +++ b/arch/s390/include/uapi/asm/hypfs.h @@ -0,0 +1,54 @@ +/* + * Structures for hypfs interface + * + * Copyright IBM Corp. 2013 + * + * Author: Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#ifndef _ASM_HYPFS_H +#define _ASM_HYPFS_H + +#include <linux/types.h> + +/* + * IOCTL for binary interface /sys/kernel/debug/diag_304 + */ +struct hypfs_diag304 { + __u32 args[2]; + __u64 data; + __u64 rc; +} __attribute__((packed)); + +#define HYPFS_IOCTL_MAGIC 0x10 + +#define HYPFS_DIAG304 \ + _IOWR(HYPFS_IOCTL_MAGIC, 0x20, struct hypfs_diag304) + +/* + * Structures for binary interface /sys/kernel/debug/diag_0c + */ +struct hypfs_diag0c_hdr { + __u64 len; /* Length of diag0c buffer without header */ + __u16 version; /* Version of header */ + char reserved1[6]; /* Reserved */ + char tod_ext[16]; /* TOD clock for diag0c */ + __u64 count; /* Number of entries (CPUs) in diag0c array */ + char reserved2[24]; /* Reserved */ +}; + +struct hypfs_diag0c_entry { + char date[8]; /* MM/DD/YY in EBCDIC */ + char time[8]; /* HH:MM:SS in EBCDIC */ + __u64 virtcpu; /* Virtual time consumed by the virt CPU (us) */ + __u64 totalproc; /* Total of virtual and simulation time (us) */ + __u32 cpu; /* Linux logical CPU number */ + __u32 reserved; /* Align to 8 byte */ +}; + +struct hypfs_diag0c_data { + struct hypfs_diag0c_hdr hdr; /* 64 byte header */ + struct hypfs_diag0c_entry entry[]; /* diag0c entry array */ +}; + +#endif diff --git a/arch/s390/include/uapi/asm/ioctl.h b/arch/s390/include/uapi/asm/ioctl.h new file mode 100644 index 000000000..b279fe06d --- /dev/null +++ b/arch/s390/include/uapi/asm/ioctl.h @@ -0,0 +1 @@ +#include <asm-generic/ioctl.h> diff --git a/arch/s390/include/uapi/asm/ioctls.h b/arch/s390/include/uapi/asm/ioctls.h new file mode 100644 index 000000000..960a4c1eb --- /dev/null +++ b/arch/s390/include/uapi/asm/ioctls.h @@ -0,0 +1,8 @@ +#ifndef __ARCH_S390_IOCTLS_H__ +#define __ARCH_S390_IOCTLS_H__ + +#define FIOQSIZE 0x545E + +#include <asm-generic/ioctls.h> + +#endif diff --git a/arch/s390/include/uapi/asm/ipcbuf.h b/arch/s390/include/uapi/asm/ipcbuf.h new file mode 100644 index 000000000..37f293d12 --- /dev/null +++ b/arch/s390/include/uapi/asm/ipcbuf.h @@ -0,0 +1,31 @@ +#ifndef __S390_IPCBUF_H__ +#define __S390_IPCBUF_H__ + +/* + * The user_ipc_perm structure for S/390 architecture. + * Note extra padding because this structure is passed back and forth + * between kernel and user space. + * + * Pad space is left for: + * - 32-bit mode_t and seq + * - 2 miscellaneous 32-bit values + */ + +struct ipc64_perm +{ + __kernel_key_t key; + __kernel_uid32_t uid; + __kernel_gid32_t gid; + __kernel_uid32_t cuid; + __kernel_gid32_t cgid; + __kernel_mode_t mode; + unsigned short __pad1; + unsigned short seq; +#ifndef __s390x__ + unsigned short __pad2; +#endif /* ! __s390x__ */ + unsigned long __unused1; + unsigned long __unused2; +}; + +#endif /* __S390_IPCBUF_H__ */ diff --git a/arch/s390/include/uapi/asm/kvm.h b/arch/s390/include/uapi/asm/kvm.h new file mode 100644 index 000000000..ef1a5fcc6 --- /dev/null +++ b/arch/s390/include/uapi/asm/kvm.h @@ -0,0 +1,182 @@ +#ifndef __LINUX_KVM_S390_H +#define __LINUX_KVM_S390_H +/* + * KVM s390 specific structures and definitions + * + * Copyright IBM Corp. 2008 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + * Christian Borntraeger <borntraeger@de.ibm.com> + */ +#include <linux/types.h> + +#define __KVM_S390 +#define __KVM_HAVE_GUEST_DEBUG + +/* Device control API: s390-specific devices */ +#define KVM_DEV_FLIC_GET_ALL_IRQS 1 +#define KVM_DEV_FLIC_ENQUEUE 2 +#define KVM_DEV_FLIC_CLEAR_IRQS 3 +#define KVM_DEV_FLIC_APF_ENABLE 4 +#define KVM_DEV_FLIC_APF_DISABLE_WAIT 5 +#define KVM_DEV_FLIC_ADAPTER_REGISTER 6 +#define KVM_DEV_FLIC_ADAPTER_MODIFY 7 +/* + * We can have up to 4*64k pending subchannels + 8 adapter interrupts, + * as well as up to ASYNC_PF_PER_VCPU*KVM_MAX_VCPUS pfault done interrupts. + * There are also sclp and machine checks. This gives us + * sizeof(kvm_s390_irq)*(4*65536+8+64*64+1+1) = 72 * 266250 = 19170000 + * Lets round up to 8192 pages. + */ +#define KVM_S390_MAX_FLOAT_IRQS 266250 +#define KVM_S390_FLIC_MAX_BUFFER 0x2000000 + +struct kvm_s390_io_adapter { + __u32 id; + __u8 isc; + __u8 maskable; + __u8 swap; + __u8 pad; +}; + +#define KVM_S390_IO_ADAPTER_MASK 1 +#define KVM_S390_IO_ADAPTER_MAP 2 +#define KVM_S390_IO_ADAPTER_UNMAP 3 + +struct kvm_s390_io_adapter_req { + __u32 id; + __u8 type; + __u8 mask; + __u16 pad0; + __u64 addr; +}; + +/* kvm attr_group on vm fd */ +#define KVM_S390_VM_MEM_CTRL 0 +#define KVM_S390_VM_TOD 1 +#define KVM_S390_VM_CRYPTO 2 +#define KVM_S390_VM_CPU_MODEL 3 + +/* kvm attributes for mem_ctrl */ +#define KVM_S390_VM_MEM_ENABLE_CMMA 0 +#define KVM_S390_VM_MEM_CLR_CMMA 1 +#define KVM_S390_VM_MEM_LIMIT_SIZE 2 + +/* kvm attributes for KVM_S390_VM_TOD */ +#define KVM_S390_VM_TOD_LOW 0 +#define KVM_S390_VM_TOD_HIGH 1 + +/* kvm attributes for KVM_S390_VM_CPU_MODEL */ +/* processor related attributes are r/w */ +#define KVM_S390_VM_CPU_PROCESSOR 0 +struct kvm_s390_vm_cpu_processor { + __u64 cpuid; + __u16 ibc; + __u8 pad[6]; + __u64 fac_list[256]; +}; + +/* machine related attributes are r/o */ +#define KVM_S390_VM_CPU_MACHINE 1 +struct kvm_s390_vm_cpu_machine { + __u64 cpuid; + __u32 ibc; + __u8 pad[4]; + __u64 fac_mask[256]; + __u64 fac_list[256]; +}; + +/* kvm attributes for crypto */ +#define KVM_S390_VM_CRYPTO_ENABLE_AES_KW 0 +#define KVM_S390_VM_CRYPTO_ENABLE_DEA_KW 1 +#define KVM_S390_VM_CRYPTO_DISABLE_AES_KW 2 +#define KVM_S390_VM_CRYPTO_DISABLE_DEA_KW 3 + +/* for KVM_GET_REGS and KVM_SET_REGS */ +struct kvm_regs { + /* general purpose regs for s390 */ + __u64 gprs[16]; +}; + +/* for KVM_GET_SREGS and KVM_SET_SREGS */ +struct kvm_sregs { + __u32 acrs[16]; + __u64 crs[16]; +}; + +/* for KVM_GET_FPU and KVM_SET_FPU */ +struct kvm_fpu { + __u32 fpc; + __u64 fprs[16]; +}; + +#define KVM_GUESTDBG_USE_HW_BP 0x00010000 + +#define KVM_HW_BP 1 +#define KVM_HW_WP_WRITE 2 +#define KVM_SINGLESTEP 4 + +struct kvm_debug_exit_arch { + __u64 addr; + __u8 type; + __u8 pad[7]; /* Should be set to 0 */ +}; + +struct kvm_hw_breakpoint { + __u64 addr; + __u64 phys_addr; + __u64 len; + __u8 type; + __u8 pad[7]; /* Should be set to 0 */ +}; + +/* for KVM_SET_GUEST_DEBUG */ +struct kvm_guest_debug_arch { + __u32 nr_hw_bp; + __u32 pad; /* Should be set to 0 */ + struct kvm_hw_breakpoint __user *hw_bp; +}; + +/* for KVM_SYNC_PFAULT and KVM_REG_S390_PFTOKEN */ +#define KVM_S390_PFAULT_TOKEN_INVALID 0xffffffffffffffffULL + +#define KVM_SYNC_PREFIX (1UL << 0) +#define KVM_SYNC_GPRS (1UL << 1) +#define KVM_SYNC_ACRS (1UL << 2) +#define KVM_SYNC_CRS (1UL << 3) +#define KVM_SYNC_ARCH0 (1UL << 4) +#define KVM_SYNC_PFAULT (1UL << 5) +#define KVM_SYNC_VRS (1UL << 6) +/* definition of registers in kvm_run */ +struct kvm_sync_regs { + __u64 prefix; /* prefix register */ + __u64 gprs[16]; /* general purpose registers */ + __u32 acrs[16]; /* access registers */ + __u64 crs[16]; /* control registers */ + __u64 todpr; /* tod programmable register [ARCH0] */ + __u64 cputm; /* cpu timer [ARCH0] */ + __u64 ckc; /* clock comparator [ARCH0] */ + __u64 pp; /* program parameter [ARCH0] */ + __u64 gbea; /* guest breaking-event address [ARCH0] */ + __u64 pft; /* pfault token [PFAULT] */ + __u64 pfs; /* pfault select [PFAULT] */ + __u64 pfc; /* pfault compare [PFAULT] */ + __u64 vrs[32][2]; /* vector registers */ + __u8 reserved[512]; /* for future vector expansion */ + __u32 fpc; /* only valid with vector registers */ +}; + +#define KVM_REG_S390_TODPR (KVM_REG_S390 | KVM_REG_SIZE_U32 | 0x1) +#define KVM_REG_S390_EPOCHDIFF (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x2) +#define KVM_REG_S390_CPU_TIMER (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x3) +#define KVM_REG_S390_CLOCK_COMP (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x4) +#define KVM_REG_S390_PFTOKEN (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x5) +#define KVM_REG_S390_PFCOMPARE (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x6) +#define KVM_REG_S390_PFSELECT (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x7) +#define KVM_REG_S390_PP (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x8) +#define KVM_REG_S390_GBEA (KVM_REG_S390 | KVM_REG_SIZE_U64 | 0x9) +#endif diff --git a/arch/s390/include/uapi/asm/kvm_para.h b/arch/s390/include/uapi/asm/kvm_para.h new file mode 100644 index 000000000..ff1f4e7b3 --- /dev/null +++ b/arch/s390/include/uapi/asm/kvm_para.h @@ -0,0 +1,11 @@ +/* + * User API definitions for paravirtual devices on s390 + * + * Copyright IBM Corp. 2008 + * + * 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. + * + * Author(s): Christian Borntraeger <borntraeger@de.ibm.com> + */ diff --git a/arch/s390/include/uapi/asm/kvm_perf.h b/arch/s390/include/uapi/asm/kvm_perf.h new file mode 100644 index 000000000..397282727 --- /dev/null +++ b/arch/s390/include/uapi/asm/kvm_perf.h @@ -0,0 +1,25 @@ +/* + * Definitions for perf-kvm on s390 + * + * 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. + */ + +#ifndef __LINUX_KVM_PERF_S390_H +#define __LINUX_KVM_PERF_S390_H + +#include <asm/sie.h> + +#define DECODE_STR_LEN 40 + +#define VCPU_ID "id" + +#define KVM_ENTRY_TRACE "kvm:kvm_s390_sie_enter" +#define KVM_EXIT_TRACE "kvm:kvm_s390_sie_exit" +#define KVM_EXIT_REASON "icptcode" + +#endif diff --git a/arch/s390/include/uapi/asm/kvm_virtio.h b/arch/s390/include/uapi/asm/kvm_virtio.h new file mode 100644 index 000000000..44a438ca9 --- /dev/null +++ b/arch/s390/include/uapi/asm/kvm_virtio.h @@ -0,0 +1,64 @@ +/* + * definition for virtio for kvm on s390 + * + * Copyright IBM Corp. 2008 + * + * 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. + * + * Author(s): Christian Borntraeger <borntraeger@de.ibm.com> + */ + +#ifndef __KVM_S390_VIRTIO_H +#define __KVM_S390_VIRTIO_H + +#include <linux/types.h> + +struct kvm_device_desc { + /* The device type: console, network, disk etc. Type 0 terminates. */ + __u8 type; + /* The number of virtqueues (first in config array) */ + __u8 num_vq; + /* + * The number of bytes of feature bits. Multiply by 2: one for host + * features and one for guest acknowledgements. + */ + __u8 feature_len; + /* The number of bytes of the config array after virtqueues. */ + __u8 config_len; + /* A status byte, written by the Guest. */ + __u8 status; + __u8 config[0]; +}; + +/* + * This is how we expect the device configuration field for a virtqueue + * to be laid out in config space. + */ +struct kvm_vqconfig { + /* The token returned with an interrupt. Set by the guest */ + __u64 token; + /* The address of the virtio ring */ + __u64 address; + /* The number of entries in the virtio_ring */ + __u16 num; + +}; + +#define KVM_S390_VIRTIO_NOTIFY 0 +#define KVM_S390_VIRTIO_RESET 1 +#define KVM_S390_VIRTIO_SET_STATUS 2 + +/* The alignment to use between consumer and producer parts of vring. + * This is pagesize for historical reasons. */ +#define KVM_S390_VIRTIO_RING_ALIGN 4096 + + +/* These values are supposed to be in ext_params on an interrupt */ +#define VIRTIO_PARAM_MASK 0xff +#define VIRTIO_PARAM_VRING_INTERRUPT 0x0 +#define VIRTIO_PARAM_CONFIG_CHANGED 0x1 +#define VIRTIO_PARAM_DEV_ADD 0x2 + +#endif diff --git a/arch/s390/include/uapi/asm/mman.h b/arch/s390/include/uapi/asm/mman.h new file mode 100644 index 000000000..de23da1f4 --- /dev/null +++ b/arch/s390/include/uapi/asm/mman.h @@ -0,0 +1,6 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/mman.h" + */ +#include <asm-generic/mman.h> diff --git a/arch/s390/include/uapi/asm/monwriter.h b/arch/s390/include/uapi/asm/monwriter.h new file mode 100644 index 000000000..f845c8e2f --- /dev/null +++ b/arch/s390/include/uapi/asm/monwriter.h @@ -0,0 +1,31 @@ +/* + * Copyright IBM Corp. 2006 + * Character device driver for writing z/VM APPLDATA monitor records + * Version 1.0 + * Author(s): Melissa Howland <melissah@us.ibm.com> + * + */ + +#ifndef _ASM_390_MONWRITER_H +#define _ASM_390_MONWRITER_H + +/* mon_function values */ +#define MONWRITE_START_INTERVAL 0x00 /* start interval recording */ +#define MONWRITE_STOP_INTERVAL 0x01 /* stop interval or config recording */ +#define MONWRITE_GEN_EVENT 0x02 /* generate event record */ +#define MONWRITE_START_CONFIG 0x03 /* start configuration recording */ + +/* the header the app uses in its write() data */ +struct monwrite_hdr { + unsigned char mon_function; + unsigned short applid; + unsigned char record_num; + unsigned short version; + unsigned short release; + unsigned short mod_level; + unsigned short datalen; + unsigned char hdrlen; + +} __attribute__((packed)); + +#endif /* _ASM_390_MONWRITER_H */ diff --git a/arch/s390/include/uapi/asm/msgbuf.h b/arch/s390/include/uapi/asm/msgbuf.h new file mode 100644 index 000000000..1bbdee927 --- /dev/null +++ b/arch/s390/include/uapi/asm/msgbuf.h @@ -0,0 +1,37 @@ +#ifndef _S390_MSGBUF_H +#define _S390_MSGBUF_H + +/* + * The msqid64_ds structure for S/390 architecture. + * Note extra padding because this structure is passed back and forth + * between kernel and user space. + * + * Pad space is left for: + * - 64-bit time_t to solve y2038 problem + * - 2 miscellaneous 32-bit values + */ + +struct msqid64_ds { + struct ipc64_perm msg_perm; + __kernel_time_t msg_stime; /* last msgsnd time */ +#ifndef __s390x__ + unsigned long __unused1; +#endif /* ! __s390x__ */ + __kernel_time_t msg_rtime; /* last msgrcv time */ +#ifndef __s390x__ + unsigned long __unused2; +#endif /* ! __s390x__ */ + __kernel_time_t msg_ctime; /* last change time */ +#ifndef __s390x__ + unsigned long __unused3; +#endif /* ! __s390x__ */ + unsigned long msg_cbytes; /* current number of bytes on queue */ + unsigned long msg_qnum; /* number of messages in queue */ + unsigned long msg_qbytes; /* max number of bytes on queue */ + __kernel_pid_t msg_lspid; /* pid of last msgsnd */ + __kernel_pid_t msg_lrpid; /* last receive pid */ + unsigned long __unused4; + unsigned long __unused5; +}; + +#endif /* _S390_MSGBUF_H */ diff --git a/arch/s390/include/uapi/asm/param.h b/arch/s390/include/uapi/asm/param.h new file mode 100644 index 000000000..c616821bf --- /dev/null +++ b/arch/s390/include/uapi/asm/param.h @@ -0,0 +1,6 @@ +#ifndef _ASMS390_PARAM_H +#define _ASMS390_PARAM_H + +#include <asm-generic/param.h> + +#endif /* _ASMS390_PARAM_H */ diff --git a/arch/s390/include/uapi/asm/poll.h b/arch/s390/include/uapi/asm/poll.h new file mode 100644 index 000000000..c98509d31 --- /dev/null +++ b/arch/s390/include/uapi/asm/poll.h @@ -0,0 +1 @@ +#include <asm-generic/poll.h> diff --git a/arch/s390/include/uapi/asm/posix_types.h b/arch/s390/include/uapi/asm/posix_types.h new file mode 100644 index 000000000..bf2a2ad2f --- /dev/null +++ b/arch/s390/include/uapi/asm/posix_types.h @@ -0,0 +1,51 @@ +/* + * S390 version + * + */ + +#ifndef __ARCH_S390_POSIX_TYPES_H +#define __ARCH_S390_POSIX_TYPES_H + +/* + * This file is generally used by user-level software, so you need to + * be a little careful about namespace pollution etc. Also, we cannot + * assume GCC is being used. + */ + +typedef unsigned long __kernel_size_t; +typedef long __kernel_ssize_t; +#define __kernel_size_t __kernel_size_t + +typedef unsigned short __kernel_old_dev_t; +#define __kernel_old_dev_t __kernel_old_dev_t + +#ifndef __s390x__ + +typedef unsigned long __kernel_ino_t; +typedef unsigned short __kernel_mode_t; +typedef unsigned short __kernel_ipc_pid_t; +typedef unsigned short __kernel_uid_t; +typedef unsigned short __kernel_gid_t; +typedef int __kernel_ptrdiff_t; + +#else /* __s390x__ */ + +typedef unsigned int __kernel_ino_t; +typedef unsigned int __kernel_mode_t; +typedef int __kernel_ipc_pid_t; +typedef unsigned int __kernel_uid_t; +typedef unsigned int __kernel_gid_t; +typedef long __kernel_ptrdiff_t; +typedef unsigned long __kernel_sigset_t; /* at least 32 bits */ + +#endif /* __s390x__ */ + +#define __kernel_ino_t __kernel_ino_t +#define __kernel_mode_t __kernel_mode_t +#define __kernel_ipc_pid_t __kernel_ipc_pid_t +#define __kernel_uid_t __kernel_uid_t +#define __kernel_gid_t __kernel_gid_t + +#include <asm-generic/posix_types.h> + +#endif diff --git a/arch/s390/include/uapi/asm/ptrace.h b/arch/s390/include/uapi/asm/ptrace.h new file mode 100644 index 000000000..a150f4fab --- /dev/null +++ b/arch/s390/include/uapi/asm/ptrace.h @@ -0,0 +1,459 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + * Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com) + */ + +#ifndef _UAPI_S390_PTRACE_H +#define _UAPI_S390_PTRACE_H + +/* + * Offsets in the user_regs_struct. They are used for the ptrace + * system call and in entry.S + */ +#ifndef __s390x__ + +#define PT_PSWMASK 0x00 +#define PT_PSWADDR 0x04 +#define PT_GPR0 0x08 +#define PT_GPR1 0x0C +#define PT_GPR2 0x10 +#define PT_GPR3 0x14 +#define PT_GPR4 0x18 +#define PT_GPR5 0x1C +#define PT_GPR6 0x20 +#define PT_GPR7 0x24 +#define PT_GPR8 0x28 +#define PT_GPR9 0x2C +#define PT_GPR10 0x30 +#define PT_GPR11 0x34 +#define PT_GPR12 0x38 +#define PT_GPR13 0x3C +#define PT_GPR14 0x40 +#define PT_GPR15 0x44 +#define PT_ACR0 0x48 +#define PT_ACR1 0x4C +#define PT_ACR2 0x50 +#define PT_ACR3 0x54 +#define PT_ACR4 0x58 +#define PT_ACR5 0x5C +#define PT_ACR6 0x60 +#define PT_ACR7 0x64 +#define PT_ACR8 0x68 +#define PT_ACR9 0x6C +#define PT_ACR10 0x70 +#define PT_ACR11 0x74 +#define PT_ACR12 0x78 +#define PT_ACR13 0x7C +#define PT_ACR14 0x80 +#define PT_ACR15 0x84 +#define PT_ORIGGPR2 0x88 +#define PT_FPC 0x90 +/* + * A nasty fact of life that the ptrace api + * only supports passing of longs. + */ +#define PT_FPR0_HI 0x98 +#define PT_FPR0_LO 0x9C +#define PT_FPR1_HI 0xA0 +#define PT_FPR1_LO 0xA4 +#define PT_FPR2_HI 0xA8 +#define PT_FPR2_LO 0xAC +#define PT_FPR3_HI 0xB0 +#define PT_FPR3_LO 0xB4 +#define PT_FPR4_HI 0xB8 +#define PT_FPR4_LO 0xBC +#define PT_FPR5_HI 0xC0 +#define PT_FPR5_LO 0xC4 +#define PT_FPR6_HI 0xC8 +#define PT_FPR6_LO 0xCC +#define PT_FPR7_HI 0xD0 +#define PT_FPR7_LO 0xD4 +#define PT_FPR8_HI 0xD8 +#define PT_FPR8_LO 0XDC +#define PT_FPR9_HI 0xE0 +#define PT_FPR9_LO 0xE4 +#define PT_FPR10_HI 0xE8 +#define PT_FPR10_LO 0xEC +#define PT_FPR11_HI 0xF0 +#define PT_FPR11_LO 0xF4 +#define PT_FPR12_HI 0xF8 +#define PT_FPR12_LO 0xFC +#define PT_FPR13_HI 0x100 +#define PT_FPR13_LO 0x104 +#define PT_FPR14_HI 0x108 +#define PT_FPR14_LO 0x10C +#define PT_FPR15_HI 0x110 +#define PT_FPR15_LO 0x114 +#define PT_CR_9 0x118 +#define PT_CR_10 0x11C +#define PT_CR_11 0x120 +#define PT_IEEE_IP 0x13C +#define PT_LASTOFF PT_IEEE_IP +#define PT_ENDREGS 0x140-1 + +#define GPR_SIZE 4 +#define CR_SIZE 4 + +#define STACK_FRAME_OVERHEAD 96 /* size of minimum stack frame */ + +#else /* __s390x__ */ + +#define PT_PSWMASK 0x00 +#define PT_PSWADDR 0x08 +#define PT_GPR0 0x10 +#define PT_GPR1 0x18 +#define PT_GPR2 0x20 +#define PT_GPR3 0x28 +#define PT_GPR4 0x30 +#define PT_GPR5 0x38 +#define PT_GPR6 0x40 +#define PT_GPR7 0x48 +#define PT_GPR8 0x50 +#define PT_GPR9 0x58 +#define PT_GPR10 0x60 +#define PT_GPR11 0x68 +#define PT_GPR12 0x70 +#define PT_GPR13 0x78 +#define PT_GPR14 0x80 +#define PT_GPR15 0x88 +#define PT_ACR0 0x90 +#define PT_ACR1 0x94 +#define PT_ACR2 0x98 +#define PT_ACR3 0x9C +#define PT_ACR4 0xA0 +#define PT_ACR5 0xA4 +#define PT_ACR6 0xA8 +#define PT_ACR7 0xAC +#define PT_ACR8 0xB0 +#define PT_ACR9 0xB4 +#define PT_ACR10 0xB8 +#define PT_ACR11 0xBC +#define PT_ACR12 0xC0 +#define PT_ACR13 0xC4 +#define PT_ACR14 0xC8 +#define PT_ACR15 0xCC +#define PT_ORIGGPR2 0xD0 +#define PT_FPC 0xD8 +#define PT_FPR0 0xE0 +#define PT_FPR1 0xE8 +#define PT_FPR2 0xF0 +#define PT_FPR3 0xF8 +#define PT_FPR4 0x100 +#define PT_FPR5 0x108 +#define PT_FPR6 0x110 +#define PT_FPR7 0x118 +#define PT_FPR8 0x120 +#define PT_FPR9 0x128 +#define PT_FPR10 0x130 +#define PT_FPR11 0x138 +#define PT_FPR12 0x140 +#define PT_FPR13 0x148 +#define PT_FPR14 0x150 +#define PT_FPR15 0x158 +#define PT_CR_9 0x160 +#define PT_CR_10 0x168 +#define PT_CR_11 0x170 +#define PT_IEEE_IP 0x1A8 +#define PT_LASTOFF PT_IEEE_IP +#define PT_ENDREGS 0x1B0-1 + +#define GPR_SIZE 8 +#define CR_SIZE 8 + +#define STACK_FRAME_OVERHEAD 160 /* size of minimum stack frame */ + +#endif /* __s390x__ */ + +#define NUM_GPRS 16 +#define NUM_FPRS 16 +#define NUM_CRS 16 +#define NUM_ACRS 16 + +#define NUM_CR_WORDS 3 + +#define FPR_SIZE 8 +#define FPC_SIZE 4 +#define FPC_PAD_SIZE 4 /* gcc insists on aligning the fpregs */ +#define ACR_SIZE 4 + + +#define PTRACE_OLDSETOPTIONS 21 + +#ifndef __ASSEMBLY__ +#include <linux/stddef.h> +#include <linux/types.h> + +typedef union +{ + float f; + double d; + __u64 ui; + struct + { + __u32 hi; + __u32 lo; + } fp; +} freg_t; + +typedef struct +{ + __u32 fpc; + __u32 pad; + freg_t fprs[NUM_FPRS]; +} s390_fp_regs; + +#define FPC_EXCEPTION_MASK 0xF8000000 +#define FPC_FLAGS_MASK 0x00F80000 +#define FPC_DXC_MASK 0x0000FF00 +#define FPC_RM_MASK 0x00000003 + +/* this typedef defines how a Program Status Word looks like */ +typedef struct +{ + unsigned long mask; + unsigned long addr; +} __attribute__ ((aligned(8))) psw_t; + +#ifndef __s390x__ + +#define PSW_MASK_PER 0x40000000UL +#define PSW_MASK_DAT 0x04000000UL +#define PSW_MASK_IO 0x02000000UL +#define PSW_MASK_EXT 0x01000000UL +#define PSW_MASK_KEY 0x00F00000UL +#define PSW_MASK_BASE 0x00080000UL /* always one */ +#define PSW_MASK_MCHECK 0x00040000UL +#define PSW_MASK_WAIT 0x00020000UL +#define PSW_MASK_PSTATE 0x00010000UL +#define PSW_MASK_ASC 0x0000C000UL +#define PSW_MASK_CC 0x00003000UL +#define PSW_MASK_PM 0x00000F00UL +#define PSW_MASK_RI 0x00000000UL +#define PSW_MASK_EA 0x00000000UL +#define PSW_MASK_BA 0x00000000UL + +#define PSW_MASK_USER 0x0000FF00UL + +#define PSW_ADDR_AMODE 0x80000000UL +#define PSW_ADDR_INSN 0x7FFFFFFFUL + +#define PSW_DEFAULT_KEY (((unsigned long) PAGE_DEFAULT_ACC) << 20) + +#define PSW_ASC_PRIMARY 0x00000000UL +#define PSW_ASC_ACCREG 0x00004000UL +#define PSW_ASC_SECONDARY 0x00008000UL +#define PSW_ASC_HOME 0x0000C000UL + +#else /* __s390x__ */ + +#define PSW_MASK_PER 0x4000000000000000UL +#define PSW_MASK_DAT 0x0400000000000000UL +#define PSW_MASK_IO 0x0200000000000000UL +#define PSW_MASK_EXT 0x0100000000000000UL +#define PSW_MASK_BASE 0x0000000000000000UL +#define PSW_MASK_KEY 0x00F0000000000000UL +#define PSW_MASK_MCHECK 0x0004000000000000UL +#define PSW_MASK_WAIT 0x0002000000000000UL +#define PSW_MASK_PSTATE 0x0001000000000000UL +#define PSW_MASK_ASC 0x0000C00000000000UL +#define PSW_MASK_CC 0x0000300000000000UL +#define PSW_MASK_PM 0x00000F0000000000UL +#define PSW_MASK_RI 0x0000008000000000UL +#define PSW_MASK_EA 0x0000000100000000UL +#define PSW_MASK_BA 0x0000000080000000UL + +#define PSW_MASK_USER 0x0000FF0180000000UL + +#define PSW_ADDR_AMODE 0x0000000000000000UL +#define PSW_ADDR_INSN 0xFFFFFFFFFFFFFFFFUL + +#define PSW_DEFAULT_KEY (((unsigned long) PAGE_DEFAULT_ACC) << 52) + +#define PSW_ASC_PRIMARY 0x0000000000000000UL +#define PSW_ASC_ACCREG 0x0000400000000000UL +#define PSW_ASC_SECONDARY 0x0000800000000000UL +#define PSW_ASC_HOME 0x0000C00000000000UL + +#endif /* __s390x__ */ + + +/* + * The s390_regs structure is used to define the elf_gregset_t. + */ +typedef struct +{ + psw_t psw; + unsigned long gprs[NUM_GPRS]; + unsigned int acrs[NUM_ACRS]; + unsigned long orig_gpr2; +} s390_regs; + +/* + * Now for the user space program event recording (trace) definitions. + * The following structures are used only for the ptrace interface, don't + * touch or even look at it if you don't want to modify the user-space + * ptrace interface. In particular stay away from it for in-kernel PER. + */ +typedef struct +{ + unsigned long cr[NUM_CR_WORDS]; +} per_cr_words; + +#define PER_EM_MASK 0xE8000000UL + +typedef struct +{ +#ifdef __s390x__ + unsigned : 32; +#endif /* __s390x__ */ + unsigned em_branching : 1; + unsigned em_instruction_fetch : 1; + /* + * Switching on storage alteration automatically fixes + * the storage alteration event bit in the users std. + */ + unsigned em_storage_alteration : 1; + unsigned em_gpr_alt_unused : 1; + unsigned em_store_real_address : 1; + unsigned : 3; + unsigned branch_addr_ctl : 1; + unsigned : 1; + unsigned storage_alt_space_ctl : 1; + unsigned : 21; + unsigned long starting_addr; + unsigned long ending_addr; +} per_cr_bits; + +typedef struct +{ + unsigned short perc_atmid; + unsigned long address; + unsigned char access_id; +} per_lowcore_words; + +typedef struct +{ + unsigned perc_branching : 1; + unsigned perc_instruction_fetch : 1; + unsigned perc_storage_alteration : 1; + unsigned perc_gpr_alt_unused : 1; + unsigned perc_store_real_address : 1; + unsigned : 3; + unsigned atmid_psw_bit_31 : 1; + unsigned atmid_validity_bit : 1; + unsigned atmid_psw_bit_32 : 1; + unsigned atmid_psw_bit_5 : 1; + unsigned atmid_psw_bit_16 : 1; + unsigned atmid_psw_bit_17 : 1; + unsigned si : 2; + unsigned long address; + unsigned : 4; + unsigned access_id : 4; +} per_lowcore_bits; + +typedef struct +{ + union { + per_cr_words words; + per_cr_bits bits; + } control_regs; + /* + * Use these flags instead of setting em_instruction_fetch + * directly they are used so that single stepping can be + * switched on & off while not affecting other tracing + */ + unsigned single_step : 1; + unsigned instruction_fetch : 1; + unsigned : 30; + /* + * These addresses are copied into cr10 & cr11 if single + * stepping is switched off + */ + unsigned long starting_addr; + unsigned long ending_addr; + union { + per_lowcore_words words; + per_lowcore_bits bits; + } lowcore; +} per_struct; + +typedef struct +{ + unsigned int len; + unsigned long kernel_addr; + unsigned long process_addr; +} ptrace_area; + +/* + * S/390 specific non posix ptrace requests. I chose unusual values so + * they are unlikely to clash with future ptrace definitions. + */ +#define PTRACE_PEEKUSR_AREA 0x5000 +#define PTRACE_POKEUSR_AREA 0x5001 +#define PTRACE_PEEKTEXT_AREA 0x5002 +#define PTRACE_PEEKDATA_AREA 0x5003 +#define PTRACE_POKETEXT_AREA 0x5004 +#define PTRACE_POKEDATA_AREA 0x5005 +#define PTRACE_GET_LAST_BREAK 0x5006 +#define PTRACE_PEEK_SYSTEM_CALL 0x5007 +#define PTRACE_POKE_SYSTEM_CALL 0x5008 +#define PTRACE_ENABLE_TE 0x5009 +#define PTRACE_DISABLE_TE 0x5010 +#define PTRACE_TE_ABORT_RAND 0x5011 + +/* + * The numbers chosen here are somewhat arbitrary but absolutely MUST + * not overlap with any of the number assigned in <linux/ptrace.h>. + */ +#define PTRACE_SINGLEBLOCK 12 /* resume execution until next branch */ + +/* + * PT_PROT definition is loosely based on hppa bsd definition in + * gdb/hppab-nat.c + */ +#define PTRACE_PROT 21 + +typedef enum +{ + ptprot_set_access_watchpoint, + ptprot_set_write_watchpoint, + ptprot_disable_watchpoint +} ptprot_flags; + +typedef struct +{ + unsigned long lowaddr; + unsigned long hiaddr; + ptprot_flags prot; +} ptprot_area; + +/* Sequence of bytes for breakpoint illegal instruction. */ +#define S390_BREAKPOINT {0x0,0x1} +#define S390_BREAKPOINT_U16 ((__u16)0x0001) +#define S390_SYSCALL_OPCODE ((__u16)0x0a00) +#define S390_SYSCALL_SIZE 2 + +/* + * The user_regs_struct defines the way the user registers are + * store on the stack for signal handling. + */ +struct user_regs_struct +{ + psw_t psw; + unsigned long gprs[NUM_GPRS]; + unsigned int acrs[NUM_ACRS]; + unsigned long orig_gpr2; + s390_fp_regs fp_regs; + /* + * These per registers are in here so that gdb can modify them + * itself as there is no "official" ptrace interface for hardware + * watchpoints. This is the way intel does it. + */ + per_struct per_info; + unsigned long ieee_instruction_pointer; /* obsolete, always 0 */ +}; + +#endif /* __ASSEMBLY__ */ + +#endif /* _UAPI_S390_PTRACE_H */ diff --git a/arch/s390/include/uapi/asm/qeth.h b/arch/s390/include/uapi/asm/qeth.h new file mode 100644 index 000000000..3a896cf52 --- /dev/null +++ b/arch/s390/include/uapi/asm/qeth.h @@ -0,0 +1,115 @@ +/* + * ioctl definitions for qeth driver + * + * Copyright IBM Corp. 2004 + * + * Author(s): Thomas Spatzier <tspat@de.ibm.com> + * + */ +#ifndef __ASM_S390_QETH_IOCTL_H__ +#define __ASM_S390_QETH_IOCTL_H__ +#include <linux/types.h> +#include <linux/ioctl.h> + +#define SIOC_QETH_ARP_SET_NO_ENTRIES (SIOCDEVPRIVATE) +#define SIOC_QETH_ARP_QUERY_INFO (SIOCDEVPRIVATE + 1) +#define SIOC_QETH_ARP_ADD_ENTRY (SIOCDEVPRIVATE + 2) +#define SIOC_QETH_ARP_REMOVE_ENTRY (SIOCDEVPRIVATE + 3) +#define SIOC_QETH_ARP_FLUSH_CACHE (SIOCDEVPRIVATE + 4) +#define SIOC_QETH_ADP_SET_SNMP_CONTROL (SIOCDEVPRIVATE + 5) +#define SIOC_QETH_GET_CARD_TYPE (SIOCDEVPRIVATE + 6) +#define SIOC_QETH_QUERY_OAT (SIOCDEVPRIVATE + 7) + +struct qeth_arp_cache_entry { + __u8 macaddr[6]; + __u8 reserved1[2]; + __u8 ipaddr[16]; /* for both IPv4 and IPv6 */ + __u8 reserved2[32]; +} __attribute__ ((packed)); + +enum qeth_arp_ipaddrtype { + QETHARP_IP_ADDR_V4 = 1, + QETHARP_IP_ADDR_V6 = 2, +}; +struct qeth_arp_entrytype { + __u8 mac; + __u8 ip; +} __attribute__((packed)); + +#define QETH_QARP_MEDIASPECIFIC_BYTES 32 +#define QETH_QARP_MACADDRTYPE_BYTES 1 +struct qeth_arp_qi_entry7 { + __u8 media_specific[QETH_QARP_MEDIASPECIFIC_BYTES]; + struct qeth_arp_entrytype type; + __u8 macaddr[6]; + __u8 ipaddr[4]; +} __attribute__((packed)); + +struct qeth_arp_qi_entry7_ipv6 { + __u8 media_specific[QETH_QARP_MEDIASPECIFIC_BYTES]; + struct qeth_arp_entrytype type; + __u8 macaddr[6]; + __u8 ipaddr[16]; +} __attribute__((packed)); + +struct qeth_arp_qi_entry7_short { + struct qeth_arp_entrytype type; + __u8 macaddr[6]; + __u8 ipaddr[4]; +} __attribute__((packed)); + +struct qeth_arp_qi_entry7_short_ipv6 { + struct qeth_arp_entrytype type; + __u8 macaddr[6]; + __u8 ipaddr[16]; +} __attribute__((packed)); + +struct qeth_arp_qi_entry5 { + __u8 media_specific[QETH_QARP_MEDIASPECIFIC_BYTES]; + struct qeth_arp_entrytype type; + __u8 ipaddr[4]; +} __attribute__((packed)); + +struct qeth_arp_qi_entry5_ipv6 { + __u8 media_specific[QETH_QARP_MEDIASPECIFIC_BYTES]; + struct qeth_arp_entrytype type; + __u8 ipaddr[16]; +} __attribute__((packed)); + +struct qeth_arp_qi_entry5_short { + struct qeth_arp_entrytype type; + __u8 ipaddr[4]; +} __attribute__((packed)); + +struct qeth_arp_qi_entry5_short_ipv6 { + struct qeth_arp_entrytype type; + __u8 ipaddr[16]; +} __attribute__((packed)); +/* + * can be set by user if no "media specific information" is wanted + * -> saves a lot of space in user space buffer + */ +#define QETH_QARP_STRIP_ENTRIES 0x8000 +#define QETH_QARP_WITH_IPV6 0x4000 +#define QETH_QARP_REQUEST_MASK 0x00ff + +/* data sent to user space as result of query arp ioctl */ +#define QETH_QARP_USER_DATA_SIZE 20000 +#define QETH_QARP_MASK_OFFSET 4 +#define QETH_QARP_ENTRIES_OFFSET 6 +struct qeth_arp_query_user_data { + union { + __u32 data_len; /* set by user space program */ + __u32 no_entries; /* set by kernel */ + } u; + __u16 mask_bits; + char *entries; +} __attribute__((packed)); + +struct qeth_query_oat_data { + __u32 command; + __u32 buffer_len; + __u32 response_len; + __u64 ptr; +}; +#endif /* __ASM_S390_QETH_IOCTL_H__ */ diff --git a/arch/s390/include/uapi/asm/resource.h b/arch/s390/include/uapi/asm/resource.h new file mode 100644 index 000000000..ec23d1c73 --- /dev/null +++ b/arch/s390/include/uapi/asm/resource.h @@ -0,0 +1,13 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/resources.h" + */ + +#ifndef _S390_RESOURCE_H +#define _S390_RESOURCE_H + +#include <asm-generic/resource.h> + +#endif + diff --git a/arch/s390/include/uapi/asm/schid.h b/arch/s390/include/uapi/asm/schid.h new file mode 100644 index 000000000..32f3ab2a8 --- /dev/null +++ b/arch/s390/include/uapi/asm/schid.h @@ -0,0 +1,16 @@ +#ifndef _UAPIASM_SCHID_H +#define _UAPIASM_SCHID_H + +#include <linux/types.h> + +struct subchannel_id { + __u32 cssid : 8; + __u32 : 4; + __u32 m : 1; + __u32 ssid : 2; + __u32 one : 1; + __u32 sch_no : 16; +} __attribute__ ((packed, aligned(4))); + + +#endif /* _UAPIASM_SCHID_H */ diff --git a/arch/s390/include/uapi/asm/sclp_ctl.h b/arch/s390/include/uapi/asm/sclp_ctl.h new file mode 100644 index 000000000..f2818613e --- /dev/null +++ b/arch/s390/include/uapi/asm/sclp_ctl.h @@ -0,0 +1,24 @@ +/* + * IOCTL interface for SCLP + * + * Copyright IBM Corp. 2012 + * + * Author: Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#ifndef _ASM_SCLP_CTL_H +#define _ASM_SCLP_CTL_H + +#include <linux/types.h> + +struct sclp_ctl_sccb { + __u32 cmdw; + __u64 sccb; +} __attribute__((packed)); + +#define SCLP_CTL_IOCTL_MAGIC 0x10 + +#define SCLP_CTL_SCCB \ + _IOWR(SCLP_CTL_IOCTL_MAGIC, 0x10, struct sclp_ctl_sccb) + +#endif diff --git a/arch/s390/include/uapi/asm/sembuf.h b/arch/s390/include/uapi/asm/sembuf.h new file mode 100644 index 000000000..32626b0ca --- /dev/null +++ b/arch/s390/include/uapi/asm/sembuf.h @@ -0,0 +1,29 @@ +#ifndef _S390_SEMBUF_H +#define _S390_SEMBUF_H + +/* + * The semid64_ds structure for S/390 architecture. + * Note extra padding because this structure is passed back and forth + * between kernel and user space. + * + * Pad space is left for: + * - 64-bit time_t to solve y2038 problem (for !__s390x__) + * - 2 miscellaneous 32-bit values + */ + +struct semid64_ds { + struct ipc64_perm sem_perm; /* permissions .. see ipc.h */ + __kernel_time_t sem_otime; /* last semop time */ +#ifndef __s390x__ + unsigned long __unused1; +#endif /* ! __s390x__ */ + __kernel_time_t sem_ctime; /* last change time */ +#ifndef __s390x__ + unsigned long __unused2; +#endif /* ! __s390x__ */ + unsigned long sem_nsems; /* no. of semaphores in array */ + unsigned long __unused3; + unsigned long __unused4; +}; + +#endif /* _S390_SEMBUF_H */ diff --git a/arch/s390/include/uapi/asm/setup.h b/arch/s390/include/uapi/asm/setup.h new file mode 100644 index 000000000..5a637e3e3 --- /dev/null +++ b/arch/s390/include/uapi/asm/setup.h @@ -0,0 +1,13 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2010 + */ + +#ifndef _UAPI_ASM_S390_SETUP_H +#define _UAPI_ASM_S390_SETUP_H + +#define COMMAND_LINE_SIZE 4096 + +#define ARCH_COMMAND_LINE_SIZE 896 + +#endif /* _UAPI_ASM_S390_SETUP_H */ diff --git a/arch/s390/include/uapi/asm/shmbuf.h b/arch/s390/include/uapi/asm/shmbuf.h new file mode 100644 index 000000000..eed2e280c --- /dev/null +++ b/arch/s390/include/uapi/asm/shmbuf.h @@ -0,0 +1,48 @@ +#ifndef _S390_SHMBUF_H +#define _S390_SHMBUF_H + +/* + * The shmid64_ds structure for S/390 architecture. + * Note extra padding because this structure is passed back and forth + * between kernel and user space. + * + * Pad space is left for: + * - 64-bit time_t to solve y2038 problem (for !__s390x__) + * - 2 miscellaneous 32-bit values + */ + +struct shmid64_ds { + struct ipc64_perm shm_perm; /* operation perms */ + size_t shm_segsz; /* size of segment (bytes) */ + __kernel_time_t shm_atime; /* last attach time */ +#ifndef __s390x__ + unsigned long __unused1; +#endif /* ! __s390x__ */ + __kernel_time_t shm_dtime; /* last detach time */ +#ifndef __s390x__ + unsigned long __unused2; +#endif /* ! __s390x__ */ + __kernel_time_t shm_ctime; /* last change time */ +#ifndef __s390x__ + unsigned long __unused3; +#endif /* ! __s390x__ */ + __kernel_pid_t shm_cpid; /* pid of creator */ + __kernel_pid_t shm_lpid; /* pid of last operator */ + unsigned long shm_nattch; /* no. of current attaches */ + unsigned long __unused4; + unsigned long __unused5; +}; + +struct shminfo64 { + unsigned long shmmax; + unsigned long shmmin; + unsigned long shmmni; + unsigned long shmseg; + unsigned long shmall; + unsigned long __unused1; + unsigned long __unused2; + unsigned long __unused3; + unsigned long __unused4; +}; + +#endif /* _S390_SHMBUF_H */ diff --git a/arch/s390/include/uapi/asm/sie.h b/arch/s390/include/uapi/asm/sie.h new file mode 100644 index 000000000..ee69c0854 --- /dev/null +++ b/arch/s390/include/uapi/asm/sie.h @@ -0,0 +1,244 @@ +#ifndef _UAPI_ASM_S390_SIE_H +#define _UAPI_ASM_S390_SIE_H + +#define diagnose_codes \ + { 0x10, "DIAG (0x10) release pages" }, \ + { 0x44, "DIAG (0x44) time slice end" }, \ + { 0x9c, "DIAG (0x9c) time slice end directed" }, \ + { 0x204, "DIAG (0x204) logical-cpu utilization" }, \ + { 0x258, "DIAG (0x258) page-reference services" }, \ + { 0x308, "DIAG (0x308) ipl functions" }, \ + { 0x500, "DIAG (0x500) KVM virtio functions" }, \ + { 0x501, "DIAG (0x501) KVM breakpoint" } + +#define sigp_order_codes \ + { 0x01, "SIGP sense" }, \ + { 0x02, "SIGP external call" }, \ + { 0x03, "SIGP emergency signal" }, \ + { 0x05, "SIGP stop" }, \ + { 0x06, "SIGP restart" }, \ + { 0x09, "SIGP stop and store status" }, \ + { 0x0b, "SIGP initial cpu reset" }, \ + { 0x0d, "SIGP set prefix" }, \ + { 0x0e, "SIGP store status at address" }, \ + { 0x12, "SIGP set architecture" }, \ + { 0x15, "SIGP sense running" } + +#define icpt_prog_codes \ + { 0x0001, "Prog Operation" }, \ + { 0x0002, "Prog Privileged Operation" }, \ + { 0x0003, "Prog Execute" }, \ + { 0x0004, "Prog Protection" }, \ + { 0x0005, "Prog Addressing" }, \ + { 0x0006, "Prog Specification" }, \ + { 0x0007, "Prog Data" }, \ + { 0x0008, "Prog Fixedpoint overflow" }, \ + { 0x0009, "Prog Fixedpoint divide" }, \ + { 0x000A, "Prog Decimal overflow" }, \ + { 0x000B, "Prog Decimal divide" }, \ + { 0x000C, "Prog HFP exponent overflow" }, \ + { 0x000D, "Prog HFP exponent underflow" }, \ + { 0x000E, "Prog HFP significance" }, \ + { 0x000F, "Prog HFP divide" }, \ + { 0x0010, "Prog Segment translation" }, \ + { 0x0011, "Prog Page translation" }, \ + { 0x0012, "Prog Translation specification" }, \ + { 0x0013, "Prog Special operation" }, \ + { 0x0015, "Prog Operand" }, \ + { 0x0016, "Prog Trace table" }, \ + { 0x0017, "Prog ASNtranslation specification" }, \ + { 0x001C, "Prog Spaceswitch event" }, \ + { 0x001D, "Prog HFP square root" }, \ + { 0x001F, "Prog PCtranslation specification" }, \ + { 0x0020, "Prog AFX translation" }, \ + { 0x0021, "Prog ASX translation" }, \ + { 0x0022, "Prog LX translation" }, \ + { 0x0023, "Prog EX translation" }, \ + { 0x0024, "Prog Primary authority" }, \ + { 0x0025, "Prog Secondary authority" }, \ + { 0x0026, "Prog LFXtranslation exception" }, \ + { 0x0027, "Prog LSXtranslation exception" }, \ + { 0x0028, "Prog ALET specification" }, \ + { 0x0029, "Prog ALEN translation" }, \ + { 0x002A, "Prog ALE sequence" }, \ + { 0x002B, "Prog ASTE validity" }, \ + { 0x002C, "Prog ASTE sequence" }, \ + { 0x002D, "Prog Extended authority" }, \ + { 0x002E, "Prog LSTE sequence" }, \ + { 0x002F, "Prog ASTE instance" }, \ + { 0x0030, "Prog Stack full" }, \ + { 0x0031, "Prog Stack empty" }, \ + { 0x0032, "Prog Stack specification" }, \ + { 0x0033, "Prog Stack type" }, \ + { 0x0034, "Prog Stack operation" }, \ + { 0x0039, "Prog Region first translation" }, \ + { 0x003A, "Prog Region second translation" }, \ + { 0x003B, "Prog Region third translation" }, \ + { 0x0040, "Prog Monitor event" }, \ + { 0x0080, "Prog PER event" }, \ + { 0x0119, "Prog Crypto operation" } + +#define exit_code_ipa0(ipa0, opcode, mnemonic) \ + { (ipa0 << 8 | opcode), #ipa0 " " mnemonic } +#define exit_code(opcode, mnemonic) \ + { opcode, mnemonic } + +#define icpt_insn_codes \ + exit_code_ipa0(0x01, 0x01, "PR"), \ + exit_code_ipa0(0x01, 0x04, "PTFF"), \ + exit_code_ipa0(0x01, 0x07, "SCKPF"), \ + exit_code_ipa0(0xAA, 0x00, "RINEXT"), \ + exit_code_ipa0(0xAA, 0x01, "RION"), \ + exit_code_ipa0(0xAA, 0x02, "TRIC"), \ + exit_code_ipa0(0xAA, 0x03, "RIOFF"), \ + exit_code_ipa0(0xAA, 0x04, "RIEMIT"), \ + exit_code_ipa0(0xB2, 0x02, "STIDP"), \ + exit_code_ipa0(0xB2, 0x04, "SCK"), \ + exit_code_ipa0(0xB2, 0x05, "STCK"), \ + exit_code_ipa0(0xB2, 0x06, "SCKC"), \ + exit_code_ipa0(0xB2, 0x07, "STCKC"), \ + exit_code_ipa0(0xB2, 0x08, "SPT"), \ + exit_code_ipa0(0xB2, 0x09, "STPT"), \ + exit_code_ipa0(0xB2, 0x0d, "PTLB"), \ + exit_code_ipa0(0xB2, 0x10, "SPX"), \ + exit_code_ipa0(0xB2, 0x11, "STPX"), \ + exit_code_ipa0(0xB2, 0x12, "STAP"), \ + exit_code_ipa0(0xB2, 0x14, "SIE"), \ + exit_code_ipa0(0xB2, 0x16, "SETR"), \ + exit_code_ipa0(0xB2, 0x17, "STETR"), \ + exit_code_ipa0(0xB2, 0x18, "PC"), \ + exit_code_ipa0(0xB2, 0x20, "SERVC"), \ + exit_code_ipa0(0xB2, 0x21, "IPTE"), \ + exit_code_ipa0(0xB2, 0x28, "PT"), \ + exit_code_ipa0(0xB2, 0x29, "ISKE"), \ + exit_code_ipa0(0xB2, 0x2a, "RRBE"), \ + exit_code_ipa0(0xB2, 0x2b, "SSKE"), \ + exit_code_ipa0(0xB2, 0x2c, "TB"), \ + exit_code_ipa0(0xB2, 0x2e, "PGIN"), \ + exit_code_ipa0(0xB2, 0x2f, "PGOUT"), \ + exit_code_ipa0(0xB2, 0x30, "CSCH"), \ + exit_code_ipa0(0xB2, 0x31, "HSCH"), \ + exit_code_ipa0(0xB2, 0x32, "MSCH"), \ + exit_code_ipa0(0xB2, 0x33, "SSCH"), \ + exit_code_ipa0(0xB2, 0x34, "STSCH"), \ + exit_code_ipa0(0xB2, 0x35, "TSCH"), \ + exit_code_ipa0(0xB2, 0x36, "TPI"), \ + exit_code_ipa0(0xB2, 0x37, "SAL"), \ + exit_code_ipa0(0xB2, 0x38, "RSCH"), \ + exit_code_ipa0(0xB2, 0x39, "STCRW"), \ + exit_code_ipa0(0xB2, 0x3a, "STCPS"), \ + exit_code_ipa0(0xB2, 0x3b, "RCHP"), \ + exit_code_ipa0(0xB2, 0x3c, "SCHM"), \ + exit_code_ipa0(0xB2, 0x40, "BAKR"), \ + exit_code_ipa0(0xB2, 0x48, "PALB"), \ + exit_code_ipa0(0xB2, 0x4c, "TAR"), \ + exit_code_ipa0(0xB2, 0x50, "CSP"), \ + exit_code_ipa0(0xB2, 0x54, "MVPG"), \ + exit_code_ipa0(0xB2, 0x58, "BSG"), \ + exit_code_ipa0(0xB2, 0x5a, "BSA"), \ + exit_code_ipa0(0xB2, 0x5f, "CHSC"), \ + exit_code_ipa0(0xB2, 0x74, "SIGA"), \ + exit_code_ipa0(0xB2, 0x76, "XSCH"), \ + exit_code_ipa0(0xB2, 0x78, "STCKE"), \ + exit_code_ipa0(0xB2, 0x7c, "STCKF"), \ + exit_code_ipa0(0xB2, 0x7d, "STSI"), \ + exit_code_ipa0(0xB2, 0xb0, "STFLE"), \ + exit_code_ipa0(0xB2, 0xb1, "STFL"), \ + exit_code_ipa0(0xB2, 0xb2, "LPSWE"), \ + exit_code_ipa0(0xB2, 0xf8, "TEND"), \ + exit_code_ipa0(0xB2, 0xfc, "TABORT"), \ + exit_code_ipa0(0xB9, 0x1e, "KMAC"), \ + exit_code_ipa0(0xB9, 0x28, "PCKMO"), \ + exit_code_ipa0(0xB9, 0x2a, "KMF"), \ + exit_code_ipa0(0xB9, 0x2b, "KMO"), \ + exit_code_ipa0(0xB9, 0x2d, "KMCTR"), \ + exit_code_ipa0(0xB9, 0x2e, "KM"), \ + exit_code_ipa0(0xB9, 0x2f, "KMC"), \ + exit_code_ipa0(0xB9, 0x3e, "KIMD"), \ + exit_code_ipa0(0xB9, 0x3f, "KLMD"), \ + exit_code_ipa0(0xB9, 0x8a, "CSPG"), \ + exit_code_ipa0(0xB9, 0x8d, "EPSW"), \ + exit_code_ipa0(0xB9, 0x8e, "IDTE"), \ + exit_code_ipa0(0xB9, 0x8f, "CRDTE"), \ + exit_code_ipa0(0xB9, 0x9c, "EQBS"), \ + exit_code_ipa0(0xB9, 0xa2, "PTF"), \ + exit_code_ipa0(0xB9, 0xab, "ESSA"), \ + exit_code_ipa0(0xB9, 0xae, "RRBM"), \ + exit_code_ipa0(0xB9, 0xaf, "PFMF"), \ + exit_code_ipa0(0xE3, 0x03, "LRAG"), \ + exit_code_ipa0(0xE3, 0x13, "LRAY"), \ + exit_code_ipa0(0xE3, 0x25, "NTSTG"), \ + exit_code_ipa0(0xE5, 0x00, "LASP"), \ + exit_code_ipa0(0xE5, 0x01, "TPROT"), \ + exit_code_ipa0(0xE5, 0x60, "TBEGIN"), \ + exit_code_ipa0(0xE5, 0x61, "TBEGINC"), \ + exit_code_ipa0(0xEB, 0x25, "STCTG"), \ + exit_code_ipa0(0xEB, 0x2f, "LCTLG"), \ + exit_code_ipa0(0xEB, 0x60, "LRIC"), \ + exit_code_ipa0(0xEB, 0x61, "STRIC"), \ + exit_code_ipa0(0xEB, 0x62, "MRIC"), \ + exit_code_ipa0(0xEB, 0x8a, "SQBS"), \ + exit_code_ipa0(0xC8, 0x01, "ECTG"), \ + exit_code(0x0a, "SVC"), \ + exit_code(0x80, "SSM"), \ + exit_code(0x82, "LPSW"), \ + exit_code(0x83, "DIAG"), \ + exit_code(0xae, "SIGP"), \ + exit_code(0xac, "STNSM"), \ + exit_code(0xad, "STOSM"), \ + exit_code(0xb1, "LRA"), \ + exit_code(0xb6, "STCTL"), \ + exit_code(0xb7, "LCTL"), \ + exit_code(0xee, "PLO") + +#define sie_intercept_code \ + { 0x00, "Host interruption" }, \ + { 0x04, "Instruction" }, \ + { 0x08, "Program interruption" }, \ + { 0x0c, "Instruction and program interruption" }, \ + { 0x10, "External request" }, \ + { 0x14, "External interruption" }, \ + { 0x18, "I/O request" }, \ + { 0x1c, "Wait state" }, \ + { 0x20, "Validity" }, \ + { 0x28, "Stop request" }, \ + { 0x2c, "Operation exception" }, \ + { 0x38, "Partial-execution" }, \ + { 0x3c, "I/O interruption" }, \ + { 0x40, "I/O instruction" }, \ + { 0x48, "Timing subset" } + +/* + * This is the simple interceptable instructions decoder. + * + * It will be used as userspace interface and it can be used in places + * that does not allow to use general decoder functions, + * such as trace events declarations. + * + * Some userspace tools may want to parse this code + * and would be confused by switch(), if() and other statements, + * but they can understand conditional operator. + */ +#define INSN_DECODE_IPA0(ipa0, insn, rshift, mask) \ + (insn >> 56) == (ipa0) ? \ + ((ipa0 << 8) | ((insn >> rshift) & mask)) : + +#define INSN_DECODE(insn) (insn >> 56) + +/* + * The macro icpt_insn_decoder() takes an intercepted instruction + * and returns a key, which can be used to find a mnemonic name + * of the instruction in the icpt_insn_codes table. + */ +#define icpt_insn_decoder(insn) ( \ + INSN_DECODE_IPA0(0x01, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xaa, insn, 48, 0x0f) \ + INSN_DECODE_IPA0(0xb2, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xb9, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xe3, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xe5, insn, 48, 0xff) \ + INSN_DECODE_IPA0(0xeb, insn, 16, 0xff) \ + INSN_DECODE_IPA0(0xc8, insn, 48, 0x0f) \ + INSN_DECODE(insn)) + +#endif /* _UAPI_ASM_S390_SIE_H */ diff --git a/arch/s390/include/uapi/asm/sigcontext.h b/arch/s390/include/uapi/asm/sigcontext.h new file mode 100644 index 000000000..5f0b8d7dd --- /dev/null +++ b/arch/s390/include/uapi/asm/sigcontext.h @@ -0,0 +1,84 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + */ + +#ifndef _ASM_S390_SIGCONTEXT_H +#define _ASM_S390_SIGCONTEXT_H + +#include <linux/compiler.h> +#include <linux/types.h> + +#define __NUM_GPRS 16 +#define __NUM_FPRS 16 +#define __NUM_ACRS 16 +#define __NUM_VXRS 32 +#define __NUM_VXRS_LOW 16 +#define __NUM_VXRS_HIGH 16 + +#ifndef __s390x__ + +/* Has to be at least _NSIG_WORDS from asm/signal.h */ +#define _SIGCONTEXT_NSIG 64 +#define _SIGCONTEXT_NSIG_BPW 32 +/* Size of stack frame allocated when calling signal handler. */ +#define __SIGNAL_FRAMESIZE 96 + +#else /* __s390x__ */ + +/* Has to be at least _NSIG_WORDS from asm/signal.h */ +#define _SIGCONTEXT_NSIG 64 +#define _SIGCONTEXT_NSIG_BPW 64 +/* Size of stack frame allocated when calling signal handler. */ +#define __SIGNAL_FRAMESIZE 160 + +#endif /* __s390x__ */ + +#define _SIGCONTEXT_NSIG_WORDS (_SIGCONTEXT_NSIG / _SIGCONTEXT_NSIG_BPW) +#define _SIGMASK_COPY_SIZE (sizeof(unsigned long)*_SIGCONTEXT_NSIG_WORDS) + +typedef struct +{ + unsigned long mask; + unsigned long addr; +} __attribute__ ((aligned(8))) _psw_t; + +typedef struct +{ + _psw_t psw; + unsigned long gprs[__NUM_GPRS]; + unsigned int acrs[__NUM_ACRS]; +} _s390_regs_common; + +typedef struct +{ + unsigned int fpc; + unsigned int pad; + double fprs[__NUM_FPRS]; +} _s390_fp_regs; + +typedef struct +{ + _s390_regs_common regs; + _s390_fp_regs fpregs; +} _sigregs; + +typedef struct +{ +#ifndef __s390x__ + unsigned long gprs_high[__NUM_GPRS]; +#endif + unsigned long long vxrs_low[__NUM_VXRS_LOW]; + __vector128 vxrs_high[__NUM_VXRS_HIGH]; + unsigned char __reserved[128]; +} _sigregs_ext; + +struct sigcontext +{ + unsigned long oldmask[_SIGCONTEXT_NSIG_WORDS]; + _sigregs __user *sregs; +}; + + +#endif + diff --git a/arch/s390/include/uapi/asm/siginfo.h b/arch/s390/include/uapi/asm/siginfo.h new file mode 100644 index 000000000..91fd3e4b7 --- /dev/null +++ b/arch/s390/include/uapi/asm/siginfo.h @@ -0,0 +1,16 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/siginfo.h" + */ + +#ifndef _S390_SIGINFO_H +#define _S390_SIGINFO_H + +#ifdef __s390x__ +#define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) +#endif + +#include <asm-generic/siginfo.h> + +#endif diff --git a/arch/s390/include/uapi/asm/signal.h b/arch/s390/include/uapi/asm/signal.h new file mode 100644 index 000000000..2f43cfbf5 --- /dev/null +++ b/arch/s390/include/uapi/asm/signal.h @@ -0,0 +1,129 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/signal.h" + */ + +#ifndef _UAPI_ASMS390_SIGNAL_H +#define _UAPI_ASMS390_SIGNAL_H + +#include <linux/types.h> +#include <linux/time.h> + +/* Avoid too many header ordering problems. */ +struct siginfo; +struct pt_regs; + +#ifndef __KERNEL__ +/* Here we must cater to libcs that poke about in kernel headers. */ + +#define NSIG 32 +typedef unsigned long sigset_t; + +#endif /* __KERNEL__ */ + +#define SIGHUP 1 +#define SIGINT 2 +#define SIGQUIT 3 +#define SIGILL 4 +#define SIGTRAP 5 +#define SIGABRT 6 +#define SIGIOT 6 +#define SIGBUS 7 +#define SIGFPE 8 +#define SIGKILL 9 +#define SIGUSR1 10 +#define SIGSEGV 11 +#define SIGUSR2 12 +#define SIGPIPE 13 +#define SIGALRM 14 +#define SIGTERM 15 +#define SIGSTKFLT 16 +#define SIGCHLD 17 +#define SIGCONT 18 +#define SIGSTOP 19 +#define SIGTSTP 20 +#define SIGTTIN 21 +#define SIGTTOU 22 +#define SIGURG 23 +#define SIGXCPU 24 +#define SIGXFSZ 25 +#define SIGVTALRM 26 +#define SIGPROF 27 +#define SIGWINCH 28 +#define SIGIO 29 +#define SIGPOLL SIGIO +/* +#define SIGLOST 29 +*/ +#define SIGPWR 30 +#define SIGSYS 31 +#define SIGUNUSED 31 + +/* These should not be considered constants from userland. */ +#define SIGRTMIN 32 +#define SIGRTMAX _NSIG + +/* + * SA_FLAGS values: + * + * SA_ONSTACK indicates that a registered stack_t will be used. + * SA_RESTART flag to get restarting signals (which were the default long ago) + * SA_NOCLDSTOP flag to turn off SIGCHLD when children stop. + * SA_RESETHAND clears the handler when the signal is delivered. + * SA_NOCLDWAIT flag on SIGCHLD to inhibit zombies. + * SA_NODEFER prevents the current signal from being masked in the handler. + * + * SA_ONESHOT and SA_NOMASK are the historical Linux names for the Single + * Unix names RESETHAND and NODEFER respectively. + */ +#define SA_NOCLDSTOP 0x00000001 +#define SA_NOCLDWAIT 0x00000002 +#define SA_SIGINFO 0x00000004 +#define SA_ONSTACK 0x08000000 +#define SA_RESTART 0x10000000 +#define SA_NODEFER 0x40000000 +#define SA_RESETHAND 0x80000000 + +#define SA_NOMASK SA_NODEFER +#define SA_ONESHOT SA_RESETHAND + +#define SA_RESTORER 0x04000000 + +#define MINSIGSTKSZ 2048 +#define SIGSTKSZ 8192 + +#include <asm-generic/signal-defs.h> + +#ifndef __KERNEL__ +/* Here we must cater to libcs that poke about in kernel headers. */ + +struct sigaction { + union { + __sighandler_t _sa_handler; + void (*_sa_sigaction)(int, struct siginfo *, void *); + } _u; +#ifndef __s390x__ /* lovely */ + sigset_t sa_mask; + unsigned long sa_flags; + void (*sa_restorer)(void); +#else /* __s390x__ */ + unsigned long sa_flags; + void (*sa_restorer)(void); + sigset_t sa_mask; +#endif /* __s390x__ */ +}; + +#define sa_handler _u._sa_handler +#define sa_sigaction _u._sa_sigaction + +#endif /* __KERNEL__ */ + +typedef struct sigaltstack { + void __user *ss_sp; + int ss_flags; + size_t ss_size; +} stack_t; + + +#endif /* _UAPI_ASMS390_SIGNAL_H */ diff --git a/arch/s390/include/uapi/asm/socket.h b/arch/s390/include/uapi/asm/socket.h new file mode 100644 index 000000000..296942d56 --- /dev/null +++ b/arch/s390/include/uapi/asm/socket.h @@ -0,0 +1,94 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/socket.h" + */ + +#ifndef _ASM_SOCKET_H +#define _ASM_SOCKET_H + +#include <asm/sockios.h> + +/* For setsockopt(2) */ +#define SOL_SOCKET 1 + +#define SO_DEBUG 1 +#define SO_REUSEADDR 2 +#define SO_TYPE 3 +#define SO_ERROR 4 +#define SO_DONTROUTE 5 +#define SO_BROADCAST 6 +#define SO_SNDBUF 7 +#define SO_RCVBUF 8 +#define SO_SNDBUFFORCE 32 +#define SO_RCVBUFFORCE 33 +#define SO_KEEPALIVE 9 +#define SO_OOBINLINE 10 +#define SO_NO_CHECK 11 +#define SO_PRIORITY 12 +#define SO_LINGER 13 +#define SO_BSDCOMPAT 14 +#define SO_REUSEPORT 15 +#define SO_PASSCRED 16 +#define SO_PEERCRED 17 +#define SO_RCVLOWAT 18 +#define SO_SNDLOWAT 19 +#define SO_RCVTIMEO 20 +#define SO_SNDTIMEO 21 + +/* Security levels - as per NRL IPv6 - don't actually do anything */ +#define SO_SECURITY_AUTHENTICATION 22 +#define SO_SECURITY_ENCRYPTION_TRANSPORT 23 +#define SO_SECURITY_ENCRYPTION_NETWORK 24 + +#define SO_BINDTODEVICE 25 + +/* Socket filtering */ +#define SO_ATTACH_FILTER 26 +#define SO_DETACH_FILTER 27 +#define SO_GET_FILTER SO_ATTACH_FILTER + +#define SO_PEERNAME 28 +#define SO_TIMESTAMP 29 +#define SCM_TIMESTAMP SO_TIMESTAMP + +#define SO_ACCEPTCONN 30 + +#define SO_PEERSEC 31 +#define SO_PASSSEC 34 +#define SO_TIMESTAMPNS 35 +#define SCM_TIMESTAMPNS SO_TIMESTAMPNS + +#define SO_MARK 36 + +#define SO_TIMESTAMPING 37 +#define SCM_TIMESTAMPING SO_TIMESTAMPING + +#define SO_PROTOCOL 38 +#define SO_DOMAIN 39 + +#define SO_RXQ_OVFL 40 + +#define SO_WIFI_STATUS 41 +#define SCM_WIFI_STATUS SO_WIFI_STATUS +#define SO_PEEK_OFF 42 + +/* Instruct lower device to use last 4-bytes of skb data as FCS */ +#define SO_NOFCS 43 + +#define SO_LOCK_FILTER 44 + +#define SO_SELECT_ERR_QUEUE 45 + +#define SO_BUSY_POLL 46 + +#define SO_MAX_PACING_RATE 47 + +#define SO_BPF_EXTENSIONS 48 + +#define SO_INCOMING_CPU 49 + +#define SO_ATTACH_BPF 50 +#define SO_DETACH_BPF SO_DETACH_FILTER + +#endif /* _ASM_SOCKET_H */ diff --git a/arch/s390/include/uapi/asm/sockios.h b/arch/s390/include/uapi/asm/sockios.h new file mode 100644 index 000000000..6f60eee73 --- /dev/null +++ b/arch/s390/include/uapi/asm/sockios.h @@ -0,0 +1,6 @@ +#ifndef _ASM_S390_SOCKIOS_H +#define _ASM_S390_SOCKIOS_H + +#include <asm-generic/sockios.h> + +#endif diff --git a/arch/s390/include/uapi/asm/stat.h b/arch/s390/include/uapi/asm/stat.h new file mode 100644 index 000000000..b4ca97d91 --- /dev/null +++ b/arch/s390/include/uapi/asm/stat.h @@ -0,0 +1,103 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/stat.h" + */ + +#ifndef _S390_STAT_H +#define _S390_STAT_H + +#ifndef __s390x__ +struct __old_kernel_stat { + unsigned short st_dev; + unsigned short st_ino; + unsigned short st_mode; + unsigned short st_nlink; + unsigned short st_uid; + unsigned short st_gid; + unsigned short st_rdev; + unsigned long st_size; + unsigned long st_atime; + unsigned long st_mtime; + unsigned long st_ctime; +}; + +struct stat { + unsigned short st_dev; + unsigned short __pad1; + unsigned long st_ino; + unsigned short st_mode; + unsigned short st_nlink; + unsigned short st_uid; + unsigned short st_gid; + unsigned short st_rdev; + unsigned short __pad2; + unsigned long st_size; + unsigned long st_blksize; + unsigned long st_blocks; + unsigned long st_atime; + unsigned long st_atime_nsec; + unsigned long st_mtime; + unsigned long st_mtime_nsec; + unsigned long st_ctime; + unsigned long st_ctime_nsec; + unsigned long __unused4; + unsigned long __unused5; +}; + +/* This matches struct stat64 in glibc2.1, hence the absolutely + * insane amounts of padding around dev_t's. + */ +struct stat64 { + unsigned long long st_dev; + unsigned int __pad1; +#define STAT64_HAS_BROKEN_ST_INO 1 + unsigned long __st_ino; + unsigned int st_mode; + unsigned int st_nlink; + unsigned long st_uid; + unsigned long st_gid; + unsigned long long st_rdev; + unsigned int __pad3; + long long st_size; + unsigned long st_blksize; + unsigned char __pad4[4]; + unsigned long __pad5; /* future possible st_blocks high bits */ + unsigned long st_blocks; /* Number 512-byte blocks allocated. */ + unsigned long st_atime; + unsigned long st_atime_nsec; + unsigned long st_mtime; + unsigned long st_mtime_nsec; + unsigned long st_ctime; + unsigned long st_ctime_nsec; /* will be high 32 bits of ctime someday */ + unsigned long long st_ino; +}; + +#else /* __s390x__ */ + +struct stat { + unsigned long st_dev; + unsigned long st_ino; + unsigned long st_nlink; + unsigned int st_mode; + unsigned int st_uid; + unsigned int st_gid; + unsigned int __pad1; + unsigned long st_rdev; + unsigned long st_size; + unsigned long st_atime; + unsigned long st_atime_nsec; + unsigned long st_mtime; + unsigned long st_mtime_nsec; + unsigned long st_ctime; + unsigned long st_ctime_nsec; + unsigned long st_blksize; + long st_blocks; + unsigned long __unused[3]; +}; + +#endif /* __s390x__ */ + +#define STAT_HAVE_NSEC 1 + +#endif diff --git a/arch/s390/include/uapi/asm/statfs.h b/arch/s390/include/uapi/asm/statfs.h new file mode 100644 index 000000000..471eb0918 --- /dev/null +++ b/arch/s390/include/uapi/asm/statfs.h @@ -0,0 +1,50 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/statfs.h" + */ + +#ifndef _S390_STATFS_H +#define _S390_STATFS_H + +/* + * We can't use <asm-generic/statfs.h> because in 64-bit mode + * we mix ints of different sizes in our struct statfs. + */ + +#ifndef __KERNEL_STRICT_NAMES +#include <linux/types.h> +typedef __kernel_fsid_t fsid_t; +#endif + +struct statfs { + unsigned int f_type; + unsigned int f_bsize; + unsigned long f_blocks; + unsigned long f_bfree; + unsigned long f_bavail; + unsigned long f_files; + unsigned long f_ffree; + __kernel_fsid_t f_fsid; + unsigned int f_namelen; + unsigned int f_frsize; + unsigned int f_flags; + unsigned int f_spare[4]; +}; + +struct statfs64 { + unsigned int f_type; + unsigned int f_bsize; + unsigned long long f_blocks; + unsigned long long f_bfree; + unsigned long long f_bavail; + unsigned long long f_files; + unsigned long long f_ffree; + __kernel_fsid_t f_fsid; + unsigned int f_namelen; + unsigned int f_frsize; + unsigned int f_flags; + unsigned int f_spare[4]; +}; + +#endif diff --git a/arch/s390/include/uapi/asm/swab.h b/arch/s390/include/uapi/asm/swab.h new file mode 100644 index 000000000..da3bfe5cc --- /dev/null +++ b/arch/s390/include/uapi/asm/swab.h @@ -0,0 +1,89 @@ +#ifndef _S390_SWAB_H +#define _S390_SWAB_H + +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#include <linux/types.h> + +#ifndef __s390x__ +# define __SWAB_64_THRU_32__ +#endif + +#ifdef __s390x__ +static inline __u64 __arch_swab64p(const __u64 *x) +{ + __u64 result; + + asm volatile("lrvg %0,%1" : "=d" (result) : "m" (*x)); + return result; +} +#define __arch_swab64p __arch_swab64p + +static inline __u64 __arch_swab64(__u64 x) +{ + __u64 result; + + asm volatile("lrvgr %0,%1" : "=d" (result) : "d" (x)); + return result; +} +#define __arch_swab64 __arch_swab64 + +static inline void __arch_swab64s(__u64 *x) +{ + *x = __arch_swab64p(x); +} +#define __arch_swab64s __arch_swab64s +#endif /* __s390x__ */ + +static inline __u32 __arch_swab32p(const __u32 *x) +{ + __u32 result; + + asm volatile( +#ifndef __s390x__ + " icm %0,8,%O1+3(%R1)\n" + " icm %0,4,%O1+2(%R1)\n" + " icm %0,2,%O1+1(%R1)\n" + " ic %0,%1" + : "=&d" (result) : "Q" (*x) : "cc"); +#else /* __s390x__ */ + " lrv %0,%1" + : "=d" (result) : "m" (*x)); +#endif /* __s390x__ */ + return result; +} +#define __arch_swab32p __arch_swab32p + +#ifdef __s390x__ +static inline __u32 __arch_swab32(__u32 x) +{ + __u32 result; + + asm volatile("lrvr %0,%1" : "=d" (result) : "d" (x)); + return result; +} +#define __arch_swab32 __arch_swab32 +#endif /* __s390x__ */ + +static inline __u16 __arch_swab16p(const __u16 *x) +{ + __u16 result; + + asm volatile( +#ifndef __s390x__ + " icm %0,2,%O1+1(%R1)\n" + " ic %0,%1\n" + : "=&d" (result) : "Q" (*x) : "cc"); +#else /* __s390x__ */ + " lrvh %0,%1" + : "=d" (result) : "m" (*x)); +#endif /* __s390x__ */ + return result; +} +#define __arch_swab16p __arch_swab16p + +#endif /* _S390_SWAB_H */ diff --git a/arch/s390/include/uapi/asm/tape390.h b/arch/s390/include/uapi/asm/tape390.h new file mode 100644 index 000000000..b2bc4bab7 --- /dev/null +++ b/arch/s390/include/uapi/asm/tape390.h @@ -0,0 +1,102 @@ +/************************************************************************* + * + * enables user programs to display messages and control encryption + * on s390 tape devices + * + * Copyright IBM Corp. 2001, 2006 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + * + *************************************************************************/ + +#ifndef _TAPE390_H +#define _TAPE390_H + +#define TAPE390_DISPLAY _IOW('d', 1, struct display_struct) + +/* + * The TAPE390_DISPLAY ioctl calls the Load Display command + * which transfers 17 bytes of data from the channel to the subsystem: + * - 1 format control byte, and + * - two 8-byte messages + * + * Format control byte: + * 0-2: New Message Overlay + * 3: Alternate Messages + * 4: Blink Message + * 5: Display Low/High Message + * 6: Reserved + * 7: Automatic Load Request + * + */ + +typedef struct display_struct { + char cntrl; + char message1[8]; + char message2[8]; +} display_struct; + +/* + * Tape encryption support + */ + +struct tape390_crypt_info { + char capability; + char status; + char medium_status; +} __attribute__ ((packed)); + + +/* Macros for "capable" field */ +#define TAPE390_CRYPT_SUPPORTED_MASK 0x01 +#define TAPE390_CRYPT_SUPPORTED(x) \ + ((x.capability & TAPE390_CRYPT_SUPPORTED_MASK)) + +/* Macros for "status" field */ +#define TAPE390_CRYPT_ON_MASK 0x01 +#define TAPE390_CRYPT_ON(x) (((x.status) & TAPE390_CRYPT_ON_MASK)) + +/* Macros for "medium status" field */ +#define TAPE390_MEDIUM_LOADED_MASK 0x01 +#define TAPE390_MEDIUM_ENCRYPTED_MASK 0x02 +#define TAPE390_MEDIUM_ENCRYPTED(x) \ + (((x.medium_status) & TAPE390_MEDIUM_ENCRYPTED_MASK)) +#define TAPE390_MEDIUM_LOADED(x) \ + (((x.medium_status) & TAPE390_MEDIUM_LOADED_MASK)) + +/* + * The TAPE390_CRYPT_SET ioctl is used to switch on/off encryption. + * The "encryption_capable" and "tape_status" fields are ignored for this ioctl! + */ +#define TAPE390_CRYPT_SET _IOW('d', 2, struct tape390_crypt_info) + +/* + * The TAPE390_CRYPT_QUERY ioctl is used to query the encryption state. + */ +#define TAPE390_CRYPT_QUERY _IOR('d', 3, struct tape390_crypt_info) + +/* Values for "kekl1/2_type" and "kekl1/2_type_on_tape" fields */ +#define TAPE390_KEKL_TYPE_NONE 0 +#define TAPE390_KEKL_TYPE_LABEL 1 +#define TAPE390_KEKL_TYPE_HASH 2 + +struct tape390_kekl { + unsigned char type; + unsigned char type_on_tape; + char label[65]; +} __attribute__ ((packed)); + +struct tape390_kekl_pair { + struct tape390_kekl kekl[2]; +} __attribute__ ((packed)); + +/* + * The TAPE390_KEKL_SET ioctl is used to set Key Encrypting Key labels. + */ +#define TAPE390_KEKL_SET _IOW('d', 4, struct tape390_kekl_pair) + +/* + * The TAPE390_KEKL_QUERY ioctl is used to query Key Encrypting Key labels. + */ +#define TAPE390_KEKL_QUERY _IOR('d', 5, struct tape390_kekl_pair) + +#endif diff --git a/arch/s390/include/uapi/asm/termbits.h b/arch/s390/include/uapi/asm/termbits.h new file mode 100644 index 000000000..71bf6ac6a --- /dev/null +++ b/arch/s390/include/uapi/asm/termbits.h @@ -0,0 +1,6 @@ +#ifndef _ASM_S390_TERMBITS_H +#define _ASM_S390_TERMBITS_H + +#include <asm-generic/termbits.h> + +#endif diff --git a/arch/s390/include/uapi/asm/termios.h b/arch/s390/include/uapi/asm/termios.h new file mode 100644 index 000000000..554f973db --- /dev/null +++ b/arch/s390/include/uapi/asm/termios.h @@ -0,0 +1,49 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/termios.h" + */ + +#ifndef _UAPI_S390_TERMIOS_H +#define _UAPI_S390_TERMIOS_H + +#include <asm/termbits.h> +#include <asm/ioctls.h> + +struct winsize { + unsigned short ws_row; + unsigned short ws_col; + unsigned short ws_xpixel; + unsigned short ws_ypixel; +}; + +#define NCC 8 +struct termio { + unsigned short c_iflag; /* input mode flags */ + unsigned short c_oflag; /* output mode flags */ + unsigned short c_cflag; /* control mode flags */ + unsigned short c_lflag; /* local mode flags */ + unsigned char c_line; /* line discipline */ + unsigned char c_cc[NCC]; /* control characters */ +}; + +/* modem lines */ +#define TIOCM_LE 0x001 +#define TIOCM_DTR 0x002 +#define TIOCM_RTS 0x004 +#define TIOCM_ST 0x008 +#define TIOCM_SR 0x010 +#define TIOCM_CTS 0x020 +#define TIOCM_CAR 0x040 +#define TIOCM_RNG 0x080 +#define TIOCM_DSR 0x100 +#define TIOCM_CD TIOCM_CAR +#define TIOCM_RI TIOCM_RNG +#define TIOCM_OUT1 0x2000 +#define TIOCM_OUT2 0x4000 +#define TIOCM_LOOP 0x8000 + +/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */ + + +#endif /* _UAPI_S390_TERMIOS_H */ diff --git a/arch/s390/include/uapi/asm/types.h b/arch/s390/include/uapi/asm/types.h new file mode 100644 index 000000000..3c3951e34 --- /dev/null +++ b/arch/s390/include/uapi/asm/types.h @@ -0,0 +1,26 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/types.h" + */ + +#ifndef _UAPI_S390_TYPES_H +#define _UAPI_S390_TYPES_H + +#include <asm-generic/int-ll64.h> + +#ifndef __ASSEMBLY__ + +/* A address type so that arithmetic can be done on it & it can be upgraded to + 64 bit when necessary +*/ +typedef unsigned long addr_t; +typedef __signed__ long saddr_t; + +typedef struct { + __u32 u[4]; +} __vector128; + +#endif /* __ASSEMBLY__ */ + +#endif /* _UAPI_S390_TYPES_H */ diff --git a/arch/s390/include/uapi/asm/ucontext.h b/arch/s390/include/uapi/asm/ucontext.h new file mode 100644 index 000000000..64a69aa5d --- /dev/null +++ b/arch/s390/include/uapi/asm/ucontext.h @@ -0,0 +1,40 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/ucontext.h" + */ + +#ifndef _ASM_S390_UCONTEXT_H +#define _ASM_S390_UCONTEXT_H + +#define UC_GPRS_HIGH 1 /* uc_mcontext_ext has valid high gprs */ +#define UC_VXRS 2 /* uc_mcontext_ext has valid vector regs */ + +/* + * The struct ucontext_extended describes how the registers are stored + * on a rt signal frame. Please note that the structure is not fixed, + * if new CPU registers are added to the user state the size of the + * struct ucontext_extended will increase. + */ +struct ucontext_extended { + unsigned long uc_flags; + struct ucontext *uc_link; + stack_t uc_stack; + _sigregs uc_mcontext; + sigset_t uc_sigmask; + /* Allow for uc_sigmask growth. Glibc uses a 1024-bit sigset_t. */ + unsigned char __unused[128 - sizeof(sigset_t)]; + _sigregs_ext uc_mcontext_ext; +}; + +struct ucontext { + unsigned long uc_flags; + struct ucontext *uc_link; + stack_t uc_stack; + _sigregs uc_mcontext; + sigset_t uc_sigmask; + /* Allow for uc_sigmask growth. Glibc uses a 1024-bit sigset_t. */ + unsigned char __unused[128 - sizeof(sigset_t)]; +}; + +#endif /* !_ASM_S390_UCONTEXT_H */ diff --git a/arch/s390/include/uapi/asm/unistd.h b/arch/s390/include/uapi/asm/unistd.h new file mode 100644 index 000000000..67878af25 --- /dev/null +++ b/arch/s390/include/uapi/asm/unistd.h @@ -0,0 +1,385 @@ +/* + * S390 version + * + * Derived from "include/asm-i386/unistd.h" + */ + +#ifndef _UAPI_ASM_S390_UNISTD_H_ +#define _UAPI_ASM_S390_UNISTD_H_ + +/* + * This file contains the system call numbers. + */ + +#define __NR_exit 1 +#define __NR_fork 2 +#define __NR_read 3 +#define __NR_write 4 +#define __NR_open 5 +#define __NR_close 6 +#define __NR_restart_syscall 7 +#define __NR_creat 8 +#define __NR_link 9 +#define __NR_unlink 10 +#define __NR_execve 11 +#define __NR_chdir 12 +#define __NR_mknod 14 +#define __NR_chmod 15 +#define __NR_lseek 19 +#define __NR_getpid 20 +#define __NR_mount 21 +#define __NR_umount 22 +#define __NR_ptrace 26 +#define __NR_alarm 27 +#define __NR_pause 29 +#define __NR_utime 30 +#define __NR_access 33 +#define __NR_nice 34 +#define __NR_sync 36 +#define __NR_kill 37 +#define __NR_rename 38 +#define __NR_mkdir 39 +#define __NR_rmdir 40 +#define __NR_dup 41 +#define __NR_pipe 42 +#define __NR_times 43 +#define __NR_brk 45 +#define __NR_signal 48 +#define __NR_acct 51 +#define __NR_umount2 52 +#define __NR_ioctl 54 +#define __NR_fcntl 55 +#define __NR_setpgid 57 +#define __NR_umask 60 +#define __NR_chroot 61 +#define __NR_ustat 62 +#define __NR_dup2 63 +#define __NR_getppid 64 +#define __NR_getpgrp 65 +#define __NR_setsid 66 +#define __NR_sigaction 67 +#define __NR_sigsuspend 72 +#define __NR_sigpending 73 +#define __NR_sethostname 74 +#define __NR_setrlimit 75 +#define __NR_getrusage 77 +#define __NR_gettimeofday 78 +#define __NR_settimeofday 79 +#define __NR_symlink 83 +#define __NR_readlink 85 +#define __NR_uselib 86 +#define __NR_swapon 87 +#define __NR_reboot 88 +#define __NR_readdir 89 +#define __NR_mmap 90 +#define __NR_munmap 91 +#define __NR_truncate 92 +#define __NR_ftruncate 93 +#define __NR_fchmod 94 +#define __NR_getpriority 96 +#define __NR_setpriority 97 +#define __NR_statfs 99 +#define __NR_fstatfs 100 +#define __NR_socketcall 102 +#define __NR_syslog 103 +#define __NR_setitimer 104 +#define __NR_getitimer 105 +#define __NR_stat 106 +#define __NR_lstat 107 +#define __NR_fstat 108 +#define __NR_lookup_dcookie 110 +#define __NR_vhangup 111 +#define __NR_idle 112 +#define __NR_wait4 114 +#define __NR_swapoff 115 +#define __NR_sysinfo 116 +#define __NR_ipc 117 +#define __NR_fsync 118 +#define __NR_sigreturn 119 +#define __NR_clone 120 +#define __NR_setdomainname 121 +#define __NR_uname 122 +#define __NR_adjtimex 124 +#define __NR_mprotect 125 +#define __NR_sigprocmask 126 +#define __NR_create_module 127 +#define __NR_init_module 128 +#define __NR_delete_module 129 +#define __NR_get_kernel_syms 130 +#define __NR_quotactl 131 +#define __NR_getpgid 132 +#define __NR_fchdir 133 +#define __NR_bdflush 134 +#define __NR_sysfs 135 +#define __NR_personality 136 +#define __NR_afs_syscall 137 /* Syscall for Andrew File System */ +#define __NR_getdents 141 +#define __NR_flock 143 +#define __NR_msync 144 +#define __NR_readv 145 +#define __NR_writev 146 +#define __NR_getsid 147 +#define __NR_fdatasync 148 +#define __NR__sysctl 149 +#define __NR_mlock 150 +#define __NR_munlock 151 +#define __NR_mlockall 152 +#define __NR_munlockall 153 +#define __NR_sched_setparam 154 +#define __NR_sched_getparam 155 +#define __NR_sched_setscheduler 156 +#define __NR_sched_getscheduler 157 +#define __NR_sched_yield 158 +#define __NR_sched_get_priority_max 159 +#define __NR_sched_get_priority_min 160 +#define __NR_sched_rr_get_interval 161 +#define __NR_nanosleep 162 +#define __NR_mremap 163 +#define __NR_query_module 167 +#define __NR_poll 168 +#define __NR_nfsservctl 169 +#define __NR_prctl 172 +#define __NR_rt_sigreturn 173 +#define __NR_rt_sigaction 174 +#define __NR_rt_sigprocmask 175 +#define __NR_rt_sigpending 176 +#define __NR_rt_sigtimedwait 177 +#define __NR_rt_sigqueueinfo 178 +#define __NR_rt_sigsuspend 179 +#define __NR_pread64 180 +#define __NR_pwrite64 181 +#define __NR_getcwd 183 +#define __NR_capget 184 +#define __NR_capset 185 +#define __NR_sigaltstack 186 +#define __NR_sendfile 187 +#define __NR_getpmsg 188 +#define __NR_putpmsg 189 +#define __NR_vfork 190 +#define __NR_pivot_root 217 +#define __NR_mincore 218 +#define __NR_madvise 219 +#define __NR_getdents64 220 +#define __NR_readahead 222 +#define __NR_setxattr 224 +#define __NR_lsetxattr 225 +#define __NR_fsetxattr 226 +#define __NR_getxattr 227 +#define __NR_lgetxattr 228 +#define __NR_fgetxattr 229 +#define __NR_listxattr 230 +#define __NR_llistxattr 231 +#define __NR_flistxattr 232 +#define __NR_removexattr 233 +#define __NR_lremovexattr 234 +#define __NR_fremovexattr 235 +#define __NR_gettid 236 +#define __NR_tkill 237 +#define __NR_futex 238 +#define __NR_sched_setaffinity 239 +#define __NR_sched_getaffinity 240 +#define __NR_tgkill 241 +/* Number 242 is reserved for tux */ +#define __NR_io_setup 243 +#define __NR_io_destroy 244 +#define __NR_io_getevents 245 +#define __NR_io_submit 246 +#define __NR_io_cancel 247 +#define __NR_exit_group 248 +#define __NR_epoll_create 249 +#define __NR_epoll_ctl 250 +#define __NR_epoll_wait 251 +#define __NR_set_tid_address 252 +#define __NR_fadvise64 253 +#define __NR_timer_create 254 +#define __NR_timer_settime (__NR_timer_create+1) +#define __NR_timer_gettime (__NR_timer_create+2) +#define __NR_timer_getoverrun (__NR_timer_create+3) +#define __NR_timer_delete (__NR_timer_create+4) +#define __NR_clock_settime (__NR_timer_create+5) +#define __NR_clock_gettime (__NR_timer_create+6) +#define __NR_clock_getres (__NR_timer_create+7) +#define __NR_clock_nanosleep (__NR_timer_create+8) +/* Number 263 is reserved for vserver */ +#define __NR_statfs64 265 +#define __NR_fstatfs64 266 +#define __NR_remap_file_pages 267 +/* Number 268 is reserved for new sys_mbind */ +/* Number 269 is reserved for new sys_get_mempolicy */ +/* Number 270 is reserved for new sys_set_mempolicy */ +#define __NR_mq_open 271 +#define __NR_mq_unlink 272 +#define __NR_mq_timedsend 273 +#define __NR_mq_timedreceive 274 +#define __NR_mq_notify 275 +#define __NR_mq_getsetattr 276 +#define __NR_kexec_load 277 +#define __NR_add_key 278 +#define __NR_request_key 279 +#define __NR_keyctl 280 +#define __NR_waitid 281 +#define __NR_ioprio_set 282 +#define __NR_ioprio_get 283 +#define __NR_inotify_init 284 +#define __NR_inotify_add_watch 285 +#define __NR_inotify_rm_watch 286 +/* Number 287 is reserved for new sys_migrate_pages */ +#define __NR_openat 288 +#define __NR_mkdirat 289 +#define __NR_mknodat 290 +#define __NR_fchownat 291 +#define __NR_futimesat 292 +#define __NR_unlinkat 294 +#define __NR_renameat 295 +#define __NR_linkat 296 +#define __NR_symlinkat 297 +#define __NR_readlinkat 298 +#define __NR_fchmodat 299 +#define __NR_faccessat 300 +#define __NR_pselect6 301 +#define __NR_ppoll 302 +#define __NR_unshare 303 +#define __NR_set_robust_list 304 +#define __NR_get_robust_list 305 +#define __NR_splice 306 +#define __NR_sync_file_range 307 +#define __NR_tee 308 +#define __NR_vmsplice 309 +/* Number 310 is reserved for new sys_move_pages */ +#define __NR_getcpu 311 +#define __NR_epoll_pwait 312 +#define __NR_utimes 313 +#define __NR_fallocate 314 +#define __NR_utimensat 315 +#define __NR_signalfd 316 +#define __NR_timerfd 317 +#define __NR_eventfd 318 +#define __NR_timerfd_create 319 +#define __NR_timerfd_settime 320 +#define __NR_timerfd_gettime 321 +#define __NR_signalfd4 322 +#define __NR_eventfd2 323 +#define __NR_inotify_init1 324 +#define __NR_pipe2 325 +#define __NR_dup3 326 +#define __NR_epoll_create1 327 +#define __NR_preadv 328 +#define __NR_pwritev 329 +#define __NR_rt_tgsigqueueinfo 330 +#define __NR_perf_event_open 331 +#define __NR_fanotify_init 332 +#define __NR_fanotify_mark 333 +#define __NR_prlimit64 334 +#define __NR_name_to_handle_at 335 +#define __NR_open_by_handle_at 336 +#define __NR_clock_adjtime 337 +#define __NR_syncfs 338 +#define __NR_setns 339 +#define __NR_process_vm_readv 340 +#define __NR_process_vm_writev 341 +#define __NR_s390_runtime_instr 342 +#define __NR_kcmp 343 +#define __NR_finit_module 344 +#define __NR_sched_setattr 345 +#define __NR_sched_getattr 346 +#define __NR_renameat2 347 +#define __NR_seccomp 348 +#define __NR_getrandom 349 +#define __NR_memfd_create 350 +#define __NR_bpf 351 +#define __NR_s390_pci_mmio_write 352 +#define __NR_s390_pci_mmio_read 353 +#define __NR_execveat 354 +#define NR_syscalls 355 + +/* + * There are some system calls that are not present on 64 bit, some + * have a different name although they do the same (e.g. __NR_chown32 + * is __NR_chown on 64 bit). + */ +#ifndef __s390x__ + +#define __NR_time 13 +#define __NR_lchown 16 +#define __NR_setuid 23 +#define __NR_getuid 24 +#define __NR_stime 25 +#define __NR_setgid 46 +#define __NR_getgid 47 +#define __NR_geteuid 49 +#define __NR_getegid 50 +#define __NR_setreuid 70 +#define __NR_setregid 71 +#define __NR_getrlimit 76 +#define __NR_getgroups 80 +#define __NR_setgroups 81 +#define __NR_fchown 95 +#define __NR_ioperm 101 +#define __NR_setfsuid 138 +#define __NR_setfsgid 139 +#define __NR__llseek 140 +#define __NR__newselect 142 +#define __NR_setresuid 164 +#define __NR_getresuid 165 +#define __NR_setresgid 170 +#define __NR_getresgid 171 +#define __NR_chown 182 +#define __NR_ugetrlimit 191 /* SuS compliant getrlimit */ +#define __NR_mmap2 192 +#define __NR_truncate64 193 +#define __NR_ftruncate64 194 +#define __NR_stat64 195 +#define __NR_lstat64 196 +#define __NR_fstat64 197 +#define __NR_lchown32 198 +#define __NR_getuid32 199 +#define __NR_getgid32 200 +#define __NR_geteuid32 201 +#define __NR_getegid32 202 +#define __NR_setreuid32 203 +#define __NR_setregid32 204 +#define __NR_getgroups32 205 +#define __NR_setgroups32 206 +#define __NR_fchown32 207 +#define __NR_setresuid32 208 +#define __NR_getresuid32 209 +#define __NR_setresgid32 210 +#define __NR_getresgid32 211 +#define __NR_chown32 212 +#define __NR_setuid32 213 +#define __NR_setgid32 214 +#define __NR_setfsuid32 215 +#define __NR_setfsgid32 216 +#define __NR_fcntl64 221 +#define __NR_sendfile64 223 +#define __NR_fadvise64_64 264 +#define __NR_fstatat64 293 + +#else + +#define __NR_select 142 +#define __NR_getrlimit 191 /* SuS compliant getrlimit */ +#define __NR_lchown 198 +#define __NR_getuid 199 +#define __NR_getgid 200 +#define __NR_geteuid 201 +#define __NR_getegid 202 +#define __NR_setreuid 203 +#define __NR_setregid 204 +#define __NR_getgroups 205 +#define __NR_setgroups 206 +#define __NR_fchown 207 +#define __NR_setresuid 208 +#define __NR_getresuid 209 +#define __NR_setresgid 210 +#define __NR_getresgid 211 +#define __NR_chown 212 +#define __NR_setuid 213 +#define __NR_setgid 214 +#define __NR_setfsuid 215 +#define __NR_setfsgid 216 +#define __NR_newfstatat 293 + +#endif + +#endif /* _UAPI_ASM_S390_UNISTD_H_ */ diff --git a/arch/s390/include/uapi/asm/virtio-ccw.h b/arch/s390/include/uapi/asm/virtio-ccw.h new file mode 100644 index 000000000..a9a4ebf79 --- /dev/null +++ b/arch/s390/include/uapi/asm/virtio-ccw.h @@ -0,0 +1,21 @@ +/* + * Definitions for virtio-ccw devices. + * + * Copyright IBM Corp. 2013 + * + * 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. + * + * Author(s): Cornelia Huck <cornelia.huck@de.ibm.com> + */ +#ifndef __KVM_VIRTIO_CCW_H +#define __KVM_VIRTIO_CCW_H + +/* Alignment of vring buffers. */ +#define KVM_VIRTIO_CCW_RING_ALIGN 4096 + +/* Subcode for diagnose 500 (virtio hypercall). */ +#define KVM_S390_VIRTIO_CCW_NOTIFY 3 + +#endif diff --git a/arch/s390/include/uapi/asm/vtoc.h b/arch/s390/include/uapi/asm/vtoc.h new file mode 100644 index 000000000..221419de2 --- /dev/null +++ b/arch/s390/include/uapi/asm/vtoc.h @@ -0,0 +1,213 @@ +/* + * This file contains volume label definitions for DASD devices. + * + * Copyright IBM Corp. 2005 + * + * Author(s): Volker Sameske <sameske@de.ibm.com> + * + */ + +#ifndef _ASM_S390_VTOC_H +#define _ASM_S390_VTOC_H + +#include <linux/types.h> + +struct vtoc_ttr +{ + __u16 tt; + __u8 r; +} __attribute__ ((packed)); + +struct vtoc_cchhb +{ + __u16 cc; + __u16 hh; + __u8 b; +} __attribute__ ((packed)); + +struct vtoc_cchh +{ + __u16 cc; + __u16 hh; +} __attribute__ ((packed)); + +struct vtoc_labeldate +{ + __u8 year; + __u16 day; +} __attribute__ ((packed)); + +struct vtoc_volume_label_cdl +{ + char volkey[4]; /* volume key = volume label */ + char vollbl[4]; /* volume label */ + char volid[6]; /* volume identifier */ + __u8 security; /* security byte */ + struct vtoc_cchhb vtoc; /* VTOC address */ + char res1[5]; /* reserved */ + char cisize[4]; /* CI-size for FBA,... */ + /* ...blanks for CKD */ + char blkperci[4]; /* no of blocks per CI (FBA), blanks for CKD */ + char labperci[4]; /* no of labels per CI (FBA), blanks for CKD */ + char res2[4]; /* reserved */ + char lvtoc[14]; /* owner code for LVTOC */ + char res3[29]; /* reserved */ +} __attribute__ ((packed)); + +struct vtoc_volume_label_ldl { + char vollbl[4]; /* volume label */ + char volid[6]; /* volume identifier */ + char res3[69]; /* reserved */ + char ldl_version; /* version number, valid for ldl format */ + __u64 formatted_blocks; /* valid when ldl_version >= f2 */ +} __attribute__ ((packed)); + +struct vtoc_extent +{ + __u8 typeind; /* extent type indicator */ + __u8 seqno; /* extent sequence number */ + struct vtoc_cchh llimit; /* starting point of this extent */ + struct vtoc_cchh ulimit; /* ending point of this extent */ +} __attribute__ ((packed)); + +struct vtoc_dev_const +{ + __u16 DS4DSCYL; /* number of logical cyls */ + __u16 DS4DSTRK; /* number of tracks in a logical cylinder */ + __u16 DS4DEVTK; /* device track length */ + __u8 DS4DEVI; /* non-last keyed record overhead */ + __u8 DS4DEVL; /* last keyed record overhead */ + __u8 DS4DEVK; /* non-keyed record overhead differential */ + __u8 DS4DEVFG; /* flag byte */ + __u16 DS4DEVTL; /* device tolerance */ + __u8 DS4DEVDT; /* number of DSCB's per track */ + __u8 DS4DEVDB; /* number of directory blocks per track */ +} __attribute__ ((packed)); + +struct vtoc_format1_label +{ + char DS1DSNAM[44]; /* data set name */ + __u8 DS1FMTID; /* format identifier */ + char DS1DSSN[6]; /* data set serial number */ + __u16 DS1VOLSQ; /* volume sequence number */ + struct vtoc_labeldate DS1CREDT; /* creation date: ydd */ + struct vtoc_labeldate DS1EXPDT; /* expiration date */ + __u8 DS1NOEPV; /* number of extents on volume */ + __u8 DS1NOBDB; /* no. of bytes used in last direction blk */ + __u8 DS1FLAG1; /* flag 1 */ + char DS1SYSCD[13]; /* system code */ + struct vtoc_labeldate DS1REFD; /* date last referenced */ + __u8 DS1SMSFG; /* system managed storage indicators */ + __u8 DS1SCXTF; /* sec. space extension flag byte */ + __u16 DS1SCXTV; /* secondary space extension value */ + __u8 DS1DSRG1; /* data set organisation byte 1 */ + __u8 DS1DSRG2; /* data set organisation byte 2 */ + __u8 DS1RECFM; /* record format */ + __u8 DS1OPTCD; /* option code */ + __u16 DS1BLKL; /* block length */ + __u16 DS1LRECL; /* record length */ + __u8 DS1KEYL; /* key length */ + __u16 DS1RKP; /* relative key position */ + __u8 DS1DSIND; /* data set indicators */ + __u8 DS1SCAL1; /* secondary allocation flag byte */ + char DS1SCAL3[3]; /* secondary allocation quantity */ + struct vtoc_ttr DS1LSTAR; /* last used track and block on track */ + __u16 DS1TRBAL; /* space remaining on last used track */ + __u16 res1; /* reserved */ + struct vtoc_extent DS1EXT1; /* first extent description */ + struct vtoc_extent DS1EXT2; /* second extent description */ + struct vtoc_extent DS1EXT3; /* third extent description */ + struct vtoc_cchhb DS1PTRDS; /* possible pointer to f2 or f3 DSCB */ +} __attribute__ ((packed)); + +struct vtoc_format4_label +{ + char DS4KEYCD[44]; /* key code for VTOC labels: 44 times 0x04 */ + __u8 DS4IDFMT; /* format identifier */ + struct vtoc_cchhb DS4HPCHR; /* highest address of a format 1 DSCB */ + __u16 DS4DSREC; /* number of available DSCB's */ + struct vtoc_cchh DS4HCCHH; /* CCHH of next available alternate track */ + __u16 DS4NOATK; /* number of remaining alternate tracks */ + __u8 DS4VTOCI; /* VTOC indicators */ + __u8 DS4NOEXT; /* number of extents in VTOC */ + __u8 DS4SMSFG; /* system managed storage indicators */ + __u8 DS4DEVAC; /* number of alternate cylinders. + * Subtract from first two bytes of + * DS4DEVSZ to get number of usable + * cylinders. can be zero. valid + * only if DS4DEVAV on. */ + struct vtoc_dev_const DS4DEVCT; /* device constants */ + char DS4AMTIM[8]; /* VSAM time stamp */ + char DS4AMCAT[3]; /* VSAM catalog indicator */ + char DS4R2TIM[8]; /* VSAM volume/catalog match time stamp */ + char res1[5]; /* reserved */ + char DS4F6PTR[5]; /* pointer to first format 6 DSCB */ + struct vtoc_extent DS4VTOCE; /* VTOC extent description */ + char res2[10]; /* reserved */ + __u8 DS4EFLVL; /* extended free-space management level */ + struct vtoc_cchhb DS4EFPTR; /* pointer to extended free-space info */ + char res3; /* reserved */ + __u32 DS4DCYL; /* number of logical cyls */ + char res4[2]; /* reserved */ + __u8 DS4DEVF2; /* device flags */ + char res5; /* reserved */ +} __attribute__ ((packed)); + +struct vtoc_ds5ext +{ + __u16 t; /* RTA of the first track of free extent */ + __u16 fc; /* number of whole cylinders in free ext. */ + __u8 ft; /* number of remaining free tracks */ +} __attribute__ ((packed)); + +struct vtoc_format5_label +{ + char DS5KEYID[4]; /* key identifier */ + struct vtoc_ds5ext DS5AVEXT; /* first available (free-space) extent. */ + struct vtoc_ds5ext DS5EXTAV[7]; /* seven available extents */ + __u8 DS5FMTID; /* format identifier */ + struct vtoc_ds5ext DS5MAVET[18]; /* eighteen available extents */ + struct vtoc_cchhb DS5PTRDS; /* pointer to next format5 DSCB */ +} __attribute__ ((packed)); + +struct vtoc_ds7ext +{ + __u32 a; /* starting RTA value */ + __u32 b; /* ending RTA value + 1 */ +} __attribute__ ((packed)); + +struct vtoc_format7_label +{ + char DS7KEYID[4]; /* key identifier */ + struct vtoc_ds7ext DS7EXTNT[5]; /* space for 5 extent descriptions */ + __u8 DS7FMTID; /* format identifier */ + struct vtoc_ds7ext DS7ADEXT[11]; /* space for 11 extent descriptions */ + char res1[2]; /* reserved */ + struct vtoc_cchhb DS7PTRDS; /* pointer to next FMT7 DSCB */ +} __attribute__ ((packed)); + +struct vtoc_cms_label { + __u8 label_id[4]; /* Label identifier */ + __u8 vol_id[6]; /* Volid */ + __u16 version_id; /* Version identifier */ + __u32 block_size; /* Disk block size */ + __u32 origin_ptr; /* Disk origin pointer */ + __u32 usable_count; /* Number of usable cylinders/blocks */ + __u32 formatted_count; /* Maximum number of formatted cylinders/ + * blocks */ + __u32 block_count; /* Disk size in CMS blocks */ + __u32 used_count; /* Number of CMS blocks in use */ + __u32 fst_size; /* File Status Table (FST) size */ + __u32 fst_count; /* Number of FSTs per CMS block */ + __u8 format_date[6]; /* Disk FORMAT date */ + __u8 reserved1[2]; + __u32 disk_offset; /* Disk offset when reserved*/ + __u32 map_block; /* Allocation Map Block with next hole */ + __u32 hblk_disp; /* Displacement into HBLK data of next hole */ + __u32 user_disp; /* Displacement into user part of Allocation + * map */ + __u8 reserved2[4]; + __u8 segment_name[8]; /* Name of shared segment */ +} __attribute__ ((packed)); + +#endif /* _ASM_S390_VTOC_H */ diff --git a/arch/s390/include/uapi/asm/zcrypt.h b/arch/s390/include/uapi/asm/zcrypt.h new file mode 100644 index 000000000..f2b18eaca --- /dev/null +++ b/arch/s390/include/uapi/asm/zcrypt.h @@ -0,0 +1,341 @@ +/* + * include/asm-s390/zcrypt.h + * + * zcrypt 2.1.0 (user-visible header) + * + * Copyright IBM Corp. 2001, 2006 + * Author(s): Robert Burroughs + * Eric Rossman (edrossma@us.ibm.com) + * + * Hotplug & misc device support: Jochen Roehrig (roehrig@de.ibm.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, 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., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#ifndef __ASM_S390_ZCRYPT_H +#define __ASM_S390_ZCRYPT_H + +#define ZCRYPT_VERSION 2 +#define ZCRYPT_RELEASE 1 +#define ZCRYPT_VARIANT 1 + +#include <linux/ioctl.h> +#include <linux/compiler.h> + +/** + * struct ica_rsa_modexpo + * + * Requirements: + * - outputdatalength is at least as large as inputdatalength. + * - All key parts are right justified in their fields, padded on + * the left with zeroes. + * - length(b_key) = inputdatalength + * - length(n_modulus) = inputdatalength + */ +struct ica_rsa_modexpo { + char __user * inputdata; + unsigned int inputdatalength; + char __user * outputdata; + unsigned int outputdatalength; + char __user * b_key; + char __user * n_modulus; +}; + +/** + * struct ica_rsa_modexpo_crt + * + * Requirements: + * - inputdatalength is even. + * - outputdatalength is at least as large as inputdatalength. + * - All key parts are right justified in their fields, padded on + * the left with zeroes. + * - length(bp_key) = inputdatalength/2 + 8 + * - length(bq_key) = inputdatalength/2 + * - length(np_key) = inputdatalength/2 + 8 + * - length(nq_key) = inputdatalength/2 + * - length(u_mult_inv) = inputdatalength/2 + 8 + */ +struct ica_rsa_modexpo_crt { + char __user * inputdata; + unsigned int inputdatalength; + char __user * outputdata; + unsigned int outputdatalength; + char __user * bp_key; + char __user * bq_key; + char __user * np_prime; + char __user * nq_prime; + char __user * u_mult_inv; +}; + +/** + * CPRBX + * Note that all shorts and ints are big-endian. + * All pointer fields are 16 bytes long, and mean nothing. + * + * A request CPRB is followed by a request_parameter_block. + * + * The request (or reply) parameter block is organized thus: + * function code + * VUD block + * key block + */ +struct CPRBX { + unsigned short cprb_len; /* CPRB length 220 */ + unsigned char cprb_ver_id; /* CPRB version id. 0x02 */ + unsigned char pad_000[3]; /* Alignment pad bytes */ + unsigned char func_id[2]; /* function id 0x5432 */ + unsigned char cprb_flags[4]; /* Flags */ + unsigned int req_parml; /* request parameter buffer len */ + unsigned int req_datal; /* request data buffer */ + unsigned int rpl_msgbl; /* reply message block length */ + unsigned int rpld_parml; /* replied parameter block len */ + unsigned int rpl_datal; /* reply data block len */ + unsigned int rpld_datal; /* replied data block len */ + unsigned int req_extbl; /* request extension block len */ + unsigned char pad_001[4]; /* reserved */ + unsigned int rpld_extbl; /* replied extension block len */ + unsigned char padx000[16 - sizeof (char *)]; + unsigned char * req_parmb; /* request parm block 'address' */ + unsigned char padx001[16 - sizeof (char *)]; + unsigned char * req_datab; /* request data block 'address' */ + unsigned char padx002[16 - sizeof (char *)]; + unsigned char * rpl_parmb; /* reply parm block 'address' */ + unsigned char padx003[16 - sizeof (char *)]; + unsigned char * rpl_datab; /* reply data block 'address' */ + unsigned char padx004[16 - sizeof (char *)]; + unsigned char * req_extb; /* request extension block 'addr'*/ + unsigned char padx005[16 - sizeof (char *)]; + unsigned char * rpl_extb; /* reply extension block 'address'*/ + unsigned short ccp_rtcode; /* server return code */ + unsigned short ccp_rscode; /* server reason code */ + unsigned int mac_data_len; /* Mac Data Length */ + unsigned char logon_id[8]; /* Logon Identifier */ + unsigned char mac_value[8]; /* Mac Value */ + unsigned char mac_content_flgs;/* Mac content flag byte */ + unsigned char pad_002; /* Alignment */ + unsigned short domain; /* Domain */ + unsigned char usage_domain[4];/* Usage domain */ + unsigned char cntrl_domain[4];/* Control domain */ + unsigned char S390enf_mask[4];/* S/390 enforcement mask */ + unsigned char pad_004[36]; /* reserved */ +} __attribute__((packed)); + +/** + * xcRB + */ +struct ica_xcRB { + unsigned short agent_ID; + unsigned int user_defined; + unsigned short request_ID; + unsigned int request_control_blk_length; + unsigned char padding1[16 - sizeof (char *)]; + char __user * request_control_blk_addr; + unsigned int request_data_length; + char padding2[16 - sizeof (char *)]; + char __user * request_data_address; + unsigned int reply_control_blk_length; + char padding3[16 - sizeof (char *)]; + char __user * reply_control_blk_addr; + unsigned int reply_data_length; + char padding4[16 - sizeof (char *)]; + char __user * reply_data_addr; + unsigned short priority_window; + unsigned int status; +} __attribute__((packed)); + +/** + * struct ep11_cprb - EP11 connectivity programming request block + * @cprb_len: CPRB header length [0x0020] + * @cprb_ver_id: CPRB version id. [0x04] + * @pad_000: Alignment pad bytes + * @flags: Admin cmd [0x80] or functional cmd [0x00] + * @func_id: Function id / subtype [0x5434] + * @source_id: Source id [originator id] + * @target_id: Target id [usage/ctrl domain id] + * @ret_code: Return code + * @reserved1: Reserved + * @reserved2: Reserved + * @payload_len: Payload length + */ +struct ep11_cprb { + uint16_t cprb_len; + unsigned char cprb_ver_id; + unsigned char pad_000[2]; + unsigned char flags; + unsigned char func_id[2]; + uint32_t source_id; + uint32_t target_id; + uint32_t ret_code; + uint32_t reserved1; + uint32_t reserved2; + uint32_t payload_len; +} __attribute__((packed)); + +/** + * struct ep11_target_dev - EP11 target device list + * @ap_id: AP device id + * @dom_id: Usage domain id + */ +struct ep11_target_dev { + uint16_t ap_id; + uint16_t dom_id; +}; + +/** + * struct ep11_urb - EP11 user request block + * @targets_num: Number of target adapters + * @targets: Addr to target adapter list + * @weight: Level of request priority + * @req_no: Request id/number + * @req_len: Request length + * @req: Addr to request block + * @resp_len: Response length + * @resp: Addr to response block + */ +struct ep11_urb { + uint16_t targets_num; + uint64_t targets; + uint64_t weight; + uint64_t req_no; + uint64_t req_len; + uint64_t req; + uint64_t resp_len; + uint64_t resp; +} __attribute__((packed)); + +#define AUTOSELECT ((unsigned int)0xFFFFFFFF) + +#define ZCRYPT_IOCTL_MAGIC 'z' + +/** + * Interface notes: + * + * The ioctl()s which are implemented (along with relevant details) + * are: + * + * ICARSAMODEXPO + * Perform an RSA operation using a Modulus-Exponent pair + * This takes an ica_rsa_modexpo struct as its arg. + * + * NOTE: please refer to the comments preceding this structure + * for the implementation details for the contents of the + * block + * + * ICARSACRT + * Perform an RSA operation using a Chinese-Remainder Theorem key + * This takes an ica_rsa_modexpo_crt struct as its arg. + * + * NOTE: please refer to the comments preceding this structure + * for the implementation details for the contents of the + * block + * + * ZSECSENDCPRB + * Send an arbitrary CPRB to a crypto card. + * + * ZSENDEP11CPRB + * Send an arbitrary EP11 CPRB to an EP11 coprocessor crypto card. + * + * Z90STAT_STATUS_MASK + * Return an 64 element array of unsigned chars for the status of + * all devices. + * 0x01: PCICA + * 0x02: PCICC + * 0x03: PCIXCC_MCL2 + * 0x04: PCIXCC_MCL3 + * 0x05: CEX2C + * 0x06: CEX2A + * 0x0d: device is disabled via the proc filesystem + * + * Z90STAT_QDEPTH_MASK + * Return an 64 element array of unsigned chars for the queue + * depth of all devices. + * + * Z90STAT_PERDEV_REQCNT + * Return an 64 element array of unsigned integers for the number + * of successfully completed requests per device since the device + * was detected and made available. + * + * Z90STAT_REQUESTQ_COUNT + * Return an integer count of the number of entries waiting to be + * sent to a device. + * + * Z90STAT_PENDINGQ_COUNT + * Return an integer count of the number of entries sent to all + * devices awaiting the reply. + * + * Z90STAT_TOTALOPEN_COUNT + * Return an integer count of the number of open file handles. + * + * Z90STAT_DOMAIN_INDEX + * Return the integer value of the Cryptographic Domain. + * + * The following ioctls are deprecated and should be no longer used: + * + * Z90STAT_TOTALCOUNT + * Return an integer count of all device types together. + * + * Z90STAT_PCICACOUNT + * Return an integer count of all PCICAs. + * + * Z90STAT_PCICCCOUNT + * Return an integer count of all PCICCs. + * + * Z90STAT_PCIXCCMCL2COUNT + * Return an integer count of all MCL2 PCIXCCs. + * + * Z90STAT_PCIXCCMCL3COUNT + * Return an integer count of all MCL3 PCIXCCs. + * + * Z90STAT_CEX2CCOUNT + * Return an integer count of all CEX2Cs. + * + * Z90STAT_CEX2ACOUNT + * Return an integer count of all CEX2As. + * + * ICAZ90STATUS + * Return some device driver status in a ica_z90_status struct + * This takes an ica_z90_status struct as its arg. + * + * Z90STAT_PCIXCCCOUNT + * Return an integer count of all PCIXCCs (MCL2 + MCL3). + * This is DEPRECATED now that MCL3 PCIXCCs are treated differently from + * MCL2 PCIXCCs. + */ + +/** + * Supported ioctl calls + */ +#define ICARSAMODEXPO _IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x05, 0) +#define ICARSACRT _IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x06, 0) +#define ZSECSENDCPRB _IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x81, 0) +#define ZSENDEP11CPRB _IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x04, 0) + +/* New status calls */ +#define Z90STAT_TOTALCOUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x40, int) +#define Z90STAT_PCICACOUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x41, int) +#define Z90STAT_PCICCCOUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x42, int) +#define Z90STAT_PCIXCCMCL2COUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x4b, int) +#define Z90STAT_PCIXCCMCL3COUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x4c, int) +#define Z90STAT_CEX2CCOUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x4d, int) +#define Z90STAT_CEX2ACOUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x4e, int) +#define Z90STAT_REQUESTQ_COUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x44, int) +#define Z90STAT_PENDINGQ_COUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x45, int) +#define Z90STAT_TOTALOPEN_COUNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x46, int) +#define Z90STAT_DOMAIN_INDEX _IOR(ZCRYPT_IOCTL_MAGIC, 0x47, int) +#define Z90STAT_STATUS_MASK _IOR(ZCRYPT_IOCTL_MAGIC, 0x48, char[64]) +#define Z90STAT_QDEPTH_MASK _IOR(ZCRYPT_IOCTL_MAGIC, 0x49, char[64]) +#define Z90STAT_PERDEV_REQCNT _IOR(ZCRYPT_IOCTL_MAGIC, 0x4a, int[64]) + +#endif /* __ASM_S390_ZCRYPT_H */ diff --git a/arch/s390/kernel/.gitignore b/arch/s390/kernel/.gitignore new file mode 100644 index 000000000..c5f676c3c --- /dev/null +++ b/arch/s390/kernel/.gitignore @@ -0,0 +1 @@ +vmlinux.lds diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile new file mode 100644 index 000000000..ffb87617a --- /dev/null +++ b/arch/s390/kernel/Makefile @@ -0,0 +1,60 @@ +# +# Makefile for the linux kernel. +# + +ifdef CONFIG_FUNCTION_TRACER +# Don't trace early setup code and tracing code +CFLAGS_REMOVE_early.o = $(CC_FLAGS_FTRACE) +CFLAGS_REMOVE_ftrace.o = $(CC_FLAGS_FTRACE) +endif + +# +# Passing null pointers is ok for smp code, since we access the lowcore here. +# +CFLAGS_smp.o := -Wno-nonnull + +# +# Disable tailcall optimizations for stack / callchain walking functions +# since this might generate broken code when accessing register 15 and +# passing its content to other functions. +# +CFLAGS_stacktrace.o += -fno-optimize-sibling-calls +CFLAGS_dumpstack.o += -fno-optimize-sibling-calls + +# +# Pass UTS_MACHINE for user_regset definition +# +CFLAGS_ptrace.o += -DUTS_MACHINE='"$(UTS_MACHINE)"' + +CFLAGS_sysinfo.o += -w + +obj-y := traps.o time.o process.o base.o early.o setup.o idle.o vtime.o +obj-y += processor.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o nmi.o +obj-y += debug.o irq.o ipl.o dis.o diag.o sclp.o vdso.o +obj-y += sysinfo.o jump_label.o lgr.o os_info.o machine_kexec.o pgm_check.o +obj-y += runtime_instr.o cache.o dumpstack.o +obj-y += entry.o reipl.o relocate_kernel.o + +extra-y += head.o head64.o vmlinux.lds + +obj-$(CONFIG_MODULES) += s390_ksyms.o module.o +obj-$(CONFIG_SMP) += smp.o +obj-$(CONFIG_SCHED_BOOK) += topology.o +obj-$(CONFIG_HIBERNATION) += suspend.o swsusp.o +obj-$(CONFIG_AUDIT) += audit.o +compat-obj-$(CONFIG_AUDIT) += compat_audit.o +obj-$(CONFIG_COMPAT) += compat_linux.o compat_signal.o +obj-$(CONFIG_COMPAT) += compat_wrapper.o $(compat-obj-y) + +obj-$(CONFIG_STACKTRACE) += stacktrace.o +obj-$(CONFIG_KPROBES) += kprobes.o +obj-$(CONFIG_FUNCTION_TRACER) += mcount.o ftrace.o +obj-$(CONFIG_CRASH_DUMP) += crash_dump.o +obj-$(CONFIG_UPROBES) += uprobes.o + +obj-$(CONFIG_PERF_EVENTS) += perf_event.o perf_cpum_cf.o perf_cpum_sf.o +obj-$(CONFIG_PERF_EVENTS) += perf_cpum_cf_events.o + +# vdso +obj-y += vdso64/ +obj-$(CONFIG_COMPAT) += vdso32/ diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c new file mode 100644 index 000000000..c7d1b9d09 --- /dev/null +++ b/arch/s390/kernel/asm-offsets.c @@ -0,0 +1,184 @@ +/* + * Generate definitions needed by assembly language modules. + * This code generates raw asm output which is post-processed to extract + * and format the required data. + */ + +#define ASM_OFFSETS_C + +#include <linux/kbuild.h> +#include <linux/kvm_host.h> +#include <linux/sched.h> +#include <asm/idle.h> +#include <asm/vdso.h> +#include <asm/pgtable.h> + +/* + * Make sure that the compiler is new enough. We want a compiler that + * is known to work with the "Q" assembler constraint. + */ +#if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 3) +#error Your compiler is too old; please use version 4.3 or newer +#endif + +int main(void) +{ + DEFINE(__THREAD_info, offsetof(struct task_struct, stack)); + DEFINE(__THREAD_ksp, offsetof(struct task_struct, thread.ksp)); + DEFINE(__THREAD_mm_segment, offsetof(struct task_struct, thread.mm_segment)); + BLANK(); + DEFINE(__TASK_pid, offsetof(struct task_struct, pid)); + BLANK(); + DEFINE(__THREAD_per_cause, offsetof(struct task_struct, thread.per_event.cause)); + DEFINE(__THREAD_per_address, offsetof(struct task_struct, thread.per_event.address)); + DEFINE(__THREAD_per_paid, offsetof(struct task_struct, thread.per_event.paid)); + BLANK(); + DEFINE(__TI_task, offsetof(struct thread_info, task)); + DEFINE(__TI_flags, offsetof(struct thread_info, flags)); + DEFINE(__TI_sysc_table, offsetof(struct thread_info, sys_call_table)); + DEFINE(__TI_cpu, offsetof(struct thread_info, cpu)); + DEFINE(__TI_precount, offsetof(struct thread_info, preempt_count)); + DEFINE(__TI_user_timer, offsetof(struct thread_info, user_timer)); + DEFINE(__TI_system_timer, offsetof(struct thread_info, system_timer)); + DEFINE(__TI_last_break, offsetof(struct thread_info, last_break)); + BLANK(); + DEFINE(__PT_ARGS, offsetof(struct pt_regs, args)); + DEFINE(__PT_PSW, offsetof(struct pt_regs, psw)); + DEFINE(__PT_GPRS, offsetof(struct pt_regs, gprs)); + DEFINE(__PT_ORIG_GPR2, offsetof(struct pt_regs, orig_gpr2)); + DEFINE(__PT_INT_CODE, offsetof(struct pt_regs, int_code)); + DEFINE(__PT_INT_PARM, offsetof(struct pt_regs, int_parm)); + DEFINE(__PT_INT_PARM_LONG, offsetof(struct pt_regs, int_parm_long)); + DEFINE(__PT_FLAGS, offsetof(struct pt_regs, flags)); + DEFINE(__PT_SIZE, sizeof(struct pt_regs)); + BLANK(); + DEFINE(__SF_BACKCHAIN, offsetof(struct stack_frame, back_chain)); + DEFINE(__SF_GPRS, offsetof(struct stack_frame, gprs)); + DEFINE(__SF_EMPTY, offsetof(struct stack_frame, empty1)); + BLANK(); + /* timeval/timezone offsets for use by vdso */ + DEFINE(__VDSO_UPD_COUNT, offsetof(struct vdso_data, tb_update_count)); + DEFINE(__VDSO_XTIME_STAMP, offsetof(struct vdso_data, xtime_tod_stamp)); + DEFINE(__VDSO_XTIME_SEC, offsetof(struct vdso_data, xtime_clock_sec)); + DEFINE(__VDSO_XTIME_NSEC, offsetof(struct vdso_data, xtime_clock_nsec)); + DEFINE(__VDSO_XTIME_CRS_SEC, offsetof(struct vdso_data, xtime_coarse_sec)); + DEFINE(__VDSO_XTIME_CRS_NSEC, offsetof(struct vdso_data, xtime_coarse_nsec)); + DEFINE(__VDSO_WTOM_SEC, offsetof(struct vdso_data, wtom_clock_sec)); + DEFINE(__VDSO_WTOM_NSEC, offsetof(struct vdso_data, wtom_clock_nsec)); + DEFINE(__VDSO_WTOM_CRS_SEC, offsetof(struct vdso_data, wtom_coarse_sec)); + DEFINE(__VDSO_WTOM_CRS_NSEC, offsetof(struct vdso_data, wtom_coarse_nsec)); + DEFINE(__VDSO_TIMEZONE, offsetof(struct vdso_data, tz_minuteswest)); + DEFINE(__VDSO_ECTG_OK, offsetof(struct vdso_data, ectg_available)); + DEFINE(__VDSO_TK_MULT, offsetof(struct vdso_data, tk_mult)); + DEFINE(__VDSO_TK_SHIFT, offsetof(struct vdso_data, tk_shift)); + DEFINE(__VDSO_ECTG_BASE, offsetof(struct vdso_per_cpu_data, ectg_timer_base)); + DEFINE(__VDSO_ECTG_USER, offsetof(struct vdso_per_cpu_data, ectg_user_time)); + /* constants used by the vdso */ + DEFINE(__CLOCK_REALTIME, CLOCK_REALTIME); + DEFINE(__CLOCK_MONOTONIC, CLOCK_MONOTONIC); + DEFINE(__CLOCK_REALTIME_COARSE, CLOCK_REALTIME_COARSE); + DEFINE(__CLOCK_MONOTONIC_COARSE, CLOCK_MONOTONIC_COARSE); + DEFINE(__CLOCK_THREAD_CPUTIME_ID, CLOCK_THREAD_CPUTIME_ID); + DEFINE(__CLOCK_REALTIME_RES, MONOTONIC_RES_NSEC); + DEFINE(__CLOCK_COARSE_RES, LOW_RES_NSEC); + BLANK(); + /* idle data offsets */ + DEFINE(__CLOCK_IDLE_ENTER, offsetof(struct s390_idle_data, clock_idle_enter)); + DEFINE(__CLOCK_IDLE_EXIT, offsetof(struct s390_idle_data, clock_idle_exit)); + DEFINE(__TIMER_IDLE_ENTER, offsetof(struct s390_idle_data, timer_idle_enter)); + DEFINE(__TIMER_IDLE_EXIT, offsetof(struct s390_idle_data, timer_idle_exit)); + /* lowcore offsets */ + DEFINE(__LC_EXT_PARAMS, offsetof(struct _lowcore, ext_params)); + DEFINE(__LC_EXT_CPU_ADDR, offsetof(struct _lowcore, ext_cpu_addr)); + DEFINE(__LC_EXT_INT_CODE, offsetof(struct _lowcore, ext_int_code)); + DEFINE(__LC_SVC_ILC, offsetof(struct _lowcore, svc_ilc)); + DEFINE(__LC_SVC_INT_CODE, offsetof(struct _lowcore, svc_code)); + DEFINE(__LC_PGM_ILC, offsetof(struct _lowcore, pgm_ilc)); + DEFINE(__LC_PGM_INT_CODE, offsetof(struct _lowcore, pgm_code)); + DEFINE(__LC_TRANS_EXC_CODE, offsetof(struct _lowcore, trans_exc_code)); + DEFINE(__LC_MON_CLASS_NR, offsetof(struct _lowcore, mon_class_num)); + DEFINE(__LC_PER_CODE, offsetof(struct _lowcore, per_code)); + DEFINE(__LC_PER_ATMID, offsetof(struct _lowcore, per_atmid)); + DEFINE(__LC_PER_ADDRESS, offsetof(struct _lowcore, per_address)); + DEFINE(__LC_EXC_ACCESS_ID, offsetof(struct _lowcore, exc_access_id)); + DEFINE(__LC_PER_ACCESS_ID, offsetof(struct _lowcore, per_access_id)); + DEFINE(__LC_OP_ACCESS_ID, offsetof(struct _lowcore, op_access_id)); + DEFINE(__LC_AR_MODE_ID, offsetof(struct _lowcore, ar_mode_id)); + DEFINE(__LC_MON_CODE, offsetof(struct _lowcore, monitor_code)); + DEFINE(__LC_SUBCHANNEL_ID, offsetof(struct _lowcore, subchannel_id)); + DEFINE(__LC_SUBCHANNEL_NR, offsetof(struct _lowcore, subchannel_nr)); + DEFINE(__LC_IO_INT_PARM, offsetof(struct _lowcore, io_int_parm)); + DEFINE(__LC_IO_INT_WORD, offsetof(struct _lowcore, io_int_word)); + DEFINE(__LC_STFL_FAC_LIST, offsetof(struct _lowcore, stfl_fac_list)); + DEFINE(__LC_MCCK_CODE, offsetof(struct _lowcore, mcck_interruption_code)); + DEFINE(__LC_MCCK_EXT_DAM_CODE, offsetof(struct _lowcore, external_damage_code)); + DEFINE(__LC_RST_OLD_PSW, offsetof(struct _lowcore, restart_old_psw)); + DEFINE(__LC_EXT_OLD_PSW, offsetof(struct _lowcore, external_old_psw)); + DEFINE(__LC_SVC_OLD_PSW, offsetof(struct _lowcore, svc_old_psw)); + DEFINE(__LC_PGM_OLD_PSW, offsetof(struct _lowcore, program_old_psw)); + DEFINE(__LC_MCK_OLD_PSW, offsetof(struct _lowcore, mcck_old_psw)); + DEFINE(__LC_IO_OLD_PSW, offsetof(struct _lowcore, io_old_psw)); + DEFINE(__LC_RST_NEW_PSW, offsetof(struct _lowcore, restart_psw)); + DEFINE(__LC_EXT_NEW_PSW, offsetof(struct _lowcore, external_new_psw)); + DEFINE(__LC_SVC_NEW_PSW, offsetof(struct _lowcore, svc_new_psw)); + DEFINE(__LC_PGM_NEW_PSW, offsetof(struct _lowcore, program_new_psw)); + DEFINE(__LC_MCK_NEW_PSW, offsetof(struct _lowcore, mcck_new_psw)); + DEFINE(__LC_IO_NEW_PSW, offsetof(struct _lowcore, io_new_psw)); + BLANK(); + DEFINE(__LC_SAVE_AREA_SYNC, offsetof(struct _lowcore, save_area_sync)); + DEFINE(__LC_SAVE_AREA_ASYNC, offsetof(struct _lowcore, save_area_async)); + DEFINE(__LC_SAVE_AREA_RESTART, offsetof(struct _lowcore, save_area_restart)); + DEFINE(__LC_CPU_FLAGS, offsetof(struct _lowcore, cpu_flags)); + DEFINE(__LC_RETURN_PSW, offsetof(struct _lowcore, return_psw)); + DEFINE(__LC_RETURN_MCCK_PSW, offsetof(struct _lowcore, return_mcck_psw)); + DEFINE(__LC_SYNC_ENTER_TIMER, offsetof(struct _lowcore, sync_enter_timer)); + DEFINE(__LC_ASYNC_ENTER_TIMER, offsetof(struct _lowcore, async_enter_timer)); + DEFINE(__LC_MCCK_ENTER_TIMER, offsetof(struct _lowcore, mcck_enter_timer)); + DEFINE(__LC_EXIT_TIMER, offsetof(struct _lowcore, exit_timer)); + DEFINE(__LC_USER_TIMER, offsetof(struct _lowcore, user_timer)); + DEFINE(__LC_SYSTEM_TIMER, offsetof(struct _lowcore, system_timer)); + DEFINE(__LC_STEAL_TIMER, offsetof(struct _lowcore, steal_timer)); + DEFINE(__LC_LAST_UPDATE_TIMER, offsetof(struct _lowcore, last_update_timer)); + DEFINE(__LC_LAST_UPDATE_CLOCK, offsetof(struct _lowcore, last_update_clock)); + DEFINE(__LC_CURRENT, offsetof(struct _lowcore, current_task)); + DEFINE(__LC_CURRENT_PID, offsetof(struct _lowcore, current_pid)); + DEFINE(__LC_THREAD_INFO, offsetof(struct _lowcore, thread_info)); + DEFINE(__LC_KERNEL_STACK, offsetof(struct _lowcore, kernel_stack)); + DEFINE(__LC_ASYNC_STACK, offsetof(struct _lowcore, async_stack)); + DEFINE(__LC_PANIC_STACK, offsetof(struct _lowcore, panic_stack)); + DEFINE(__LC_RESTART_STACK, offsetof(struct _lowcore, restart_stack)); + DEFINE(__LC_RESTART_FN, offsetof(struct _lowcore, restart_fn)); + DEFINE(__LC_RESTART_DATA, offsetof(struct _lowcore, restart_data)); + DEFINE(__LC_RESTART_SOURCE, offsetof(struct _lowcore, restart_source)); + DEFINE(__LC_KERNEL_ASCE, offsetof(struct _lowcore, kernel_asce)); + DEFINE(__LC_USER_ASCE, offsetof(struct _lowcore, user_asce)); + DEFINE(__LC_INT_CLOCK, offsetof(struct _lowcore, int_clock)); + DEFINE(__LC_MCCK_CLOCK, offsetof(struct _lowcore, mcck_clock)); + DEFINE(__LC_MACHINE_FLAGS, offsetof(struct _lowcore, machine_flags)); + DEFINE(__LC_DUMP_REIPL, offsetof(struct _lowcore, ipib)); + BLANK(); + DEFINE(__LC_CPU_TIMER_SAVE_AREA, offsetof(struct _lowcore, cpu_timer_save_area)); + DEFINE(__LC_CLOCK_COMP_SAVE_AREA, offsetof(struct _lowcore, clock_comp_save_area)); + DEFINE(__LC_PSW_SAVE_AREA, offsetof(struct _lowcore, psw_save_area)); + DEFINE(__LC_PREFIX_SAVE_AREA, offsetof(struct _lowcore, prefixreg_save_area)); + DEFINE(__LC_AREGS_SAVE_AREA, offsetof(struct _lowcore, access_regs_save_area)); + DEFINE(__LC_FPREGS_SAVE_AREA, offsetof(struct _lowcore, floating_pt_save_area)); + DEFINE(__LC_GPREGS_SAVE_AREA, offsetof(struct _lowcore, gpregs_save_area)); + DEFINE(__LC_CREGS_SAVE_AREA, offsetof(struct _lowcore, cregs_save_area)); + DEFINE(__LC_DATA_EXC_CODE, offsetof(struct _lowcore, data_exc_code)); + DEFINE(__LC_MCCK_FAIL_STOR_ADDR, offsetof(struct _lowcore, failing_storage_address)); + DEFINE(__LC_VX_SAVE_AREA_ADDR, offsetof(struct _lowcore, vector_save_area_addr)); + DEFINE(__LC_EXT_PARAMS2, offsetof(struct _lowcore, ext_params2)); + DEFINE(SAVE_AREA_BASE, offsetof(struct _lowcore, floating_pt_save_area)); + DEFINE(__LC_PASTE, offsetof(struct _lowcore, paste)); + DEFINE(__LC_FP_CREG_SAVE_AREA, offsetof(struct _lowcore, fpt_creg_save_area)); + DEFINE(__LC_LAST_BREAK, offsetof(struct _lowcore, breaking_event_addr)); + DEFINE(__LC_VDSO_PER_CPU, offsetof(struct _lowcore, vdso_per_cpu_data)); + DEFINE(__LC_GMAP, offsetof(struct _lowcore, gmap)); + DEFINE(__LC_PGM_TDB, offsetof(struct _lowcore, pgm_tdb)); + DEFINE(__THREAD_trap_tdb, offsetof(struct task_struct, thread.trap_tdb)); + DEFINE(__GMAP_ASCE, offsetof(struct gmap, asce)); + DEFINE(__SIE_PROG0C, offsetof(struct kvm_s390_sie_block, prog0c)); + DEFINE(__SIE_PROG20, offsetof(struct kvm_s390_sie_block, prog20)); + return 0; +} diff --git a/arch/s390/kernel/audit.c b/arch/s390/kernel/audit.c new file mode 100644 index 000000000..f4932c22e --- /dev/null +++ b/arch/s390/kernel/audit.c @@ -0,0 +1,78 @@ +#include <linux/init.h> +#include <linux/types.h> +#include <linux/audit.h> +#include <asm/unistd.h> +#include "audit.h" + +static unsigned dir_class[] = { +#include <asm-generic/audit_dir_write.h> +~0U +}; + +static unsigned read_class[] = { +#include <asm-generic/audit_read.h> +~0U +}; + +static unsigned write_class[] = { +#include <asm-generic/audit_write.h> +~0U +}; + +static unsigned chattr_class[] = { +#include <asm-generic/audit_change_attr.h> +~0U +}; + +static unsigned signal_class[] = { +#include <asm-generic/audit_signal.h> +~0U +}; + +int audit_classify_arch(int arch) +{ +#ifdef CONFIG_COMPAT + if (arch == AUDIT_ARCH_S390) + return 1; +#endif + return 0; +} + +int audit_classify_syscall(int abi, unsigned syscall) +{ +#ifdef CONFIG_COMPAT + if (abi == AUDIT_ARCH_S390) + return s390_classify_syscall(syscall); +#endif + switch(syscall) { + case __NR_open: + return 2; + case __NR_openat: + return 3; + case __NR_socketcall: + return 4; + case __NR_execve: + return 5; + default: + return 0; + } +} + +static int __init audit_classes_init(void) +{ +#ifdef CONFIG_COMPAT + audit_register_class(AUDIT_CLASS_WRITE_32, s390_write_class); + audit_register_class(AUDIT_CLASS_READ_32, s390_read_class); + audit_register_class(AUDIT_CLASS_DIR_WRITE_32, s390_dir_class); + audit_register_class(AUDIT_CLASS_CHATTR_32, s390_chattr_class); + audit_register_class(AUDIT_CLASS_SIGNAL_32, s390_signal_class); +#endif + audit_register_class(AUDIT_CLASS_WRITE, write_class); + audit_register_class(AUDIT_CLASS_READ, read_class); + audit_register_class(AUDIT_CLASS_DIR_WRITE, dir_class); + audit_register_class(AUDIT_CLASS_CHATTR, chattr_class); + audit_register_class(AUDIT_CLASS_SIGNAL, signal_class); + return 0; +} + +__initcall(audit_classes_init); diff --git a/arch/s390/kernel/audit.h b/arch/s390/kernel/audit.h new file mode 100644 index 000000000..12b56f4b5 --- /dev/null +++ b/arch/s390/kernel/audit.h @@ -0,0 +1,15 @@ +#ifndef __ARCH_S390_KERNEL_AUDIT_H +#define __ARCH_S390_KERNEL_AUDIT_H + +#include <linux/types.h> + +#ifdef CONFIG_COMPAT +extern int s390_classify_syscall(unsigned); +extern __u32 s390_dir_class[]; +extern __u32 s390_write_class[]; +extern __u32 s390_read_class[]; +extern __u32 s390_chattr_class[]; +extern __u32 s390_signal_class[]; +#endif /* CONFIG_COMPAT */ + +#endif /* __ARCH_S390_KERNEL_AUDIT_H */ diff --git a/arch/s390/kernel/base.S b/arch/s390/kernel/base.S new file mode 100644 index 000000000..daed3fde4 --- /dev/null +++ b/arch/s390/kernel/base.S @@ -0,0 +1,131 @@ +/* + * arch/s390/kernel/base.S + * + * Copyright IBM Corp. 2006, 2007 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + * Michael Holzheu <holzheu@de.ibm.com> + */ + +#include <linux/linkage.h> +#include <asm/asm-offsets.h> +#include <asm/ptrace.h> +#include <asm/sigp.h> + +ENTRY(s390_base_mcck_handler) + basr %r13,0 +0: lg %r15,__LC_PANIC_STACK # load panic stack + aghi %r15,-STACK_FRAME_OVERHEAD + larl %r1,s390_base_mcck_handler_fn + lg %r1,0(%r1) + ltgr %r1,%r1 + jz 1f + basr %r14,%r1 +1: la %r1,4095 + lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1) + lpswe __LC_MCK_OLD_PSW + + .section .bss + .align 8 + .globl s390_base_mcck_handler_fn +s390_base_mcck_handler_fn: + .quad 0 + .previous + +ENTRY(s390_base_ext_handler) + stmg %r0,%r15,__LC_SAVE_AREA_ASYNC + basr %r13,0 +0: aghi %r15,-STACK_FRAME_OVERHEAD + larl %r1,s390_base_ext_handler_fn + lg %r1,0(%r1) + ltgr %r1,%r1 + jz 1f + basr %r14,%r1 +1: lmg %r0,%r15,__LC_SAVE_AREA_ASYNC + ni __LC_EXT_OLD_PSW+1,0xfd # clear wait state bit + lpswe __LC_EXT_OLD_PSW + + .section .bss + .align 8 + .globl s390_base_ext_handler_fn +s390_base_ext_handler_fn: + .quad 0 + .previous + +ENTRY(s390_base_pgm_handler) + stmg %r0,%r15,__LC_SAVE_AREA_SYNC + basr %r13,0 +0: aghi %r15,-STACK_FRAME_OVERHEAD + larl %r1,s390_base_pgm_handler_fn + lg %r1,0(%r1) + ltgr %r1,%r1 + jz 1f + basr %r14,%r1 + lmg %r0,%r15,__LC_SAVE_AREA_SYNC + lpswe __LC_PGM_OLD_PSW +1: lpswe disabled_wait_psw-0b(%r13) + + .align 8 +disabled_wait_psw: + .quad 0x0002000180000000,0x0000000000000000 + s390_base_pgm_handler + + .section .bss + .align 8 + .globl s390_base_pgm_handler_fn +s390_base_pgm_handler_fn: + .quad 0 + .previous + +# +# Calls diag 308 subcode 1 and continues execution +# +# The following conditions must be ensured before calling this function: +# * Prefix register = 0 +# * Lowcore protection is disabled +# +ENTRY(diag308_reset) + larl %r4,.Lctlregs # Save control registers + stctg %c0,%c15,0(%r4) + larl %r4,.Lfpctl # Floating point control register + stfpc 0(%r4) + larl %r4,.Lcontinue_psw # Save PSW flags + epsw %r2,%r3 + stm %r2,%r3,0(%r4) + larl %r4,.Lrestart_psw # Setup restart PSW at absolute 0 + lghi %r3,0 + lg %r4,0(%r4) # Save PSW + sturg %r4,%r3 # Use sturg, because of large pages + lghi %r1,1 + lghi %r0,0 + diag %r0,%r1,0x308 +.Lrestart_part2: + lhi %r0,0 # Load r0 with zero + lhi %r1,2 # Use mode 2 = ESAME (dump) + sigp %r1,%r0,SIGP_SET_ARCHITECTURE # Switch to ESAME mode + sam64 # Switch to 64 bit addressing mode + larl %r4,.Lctlregs # Restore control registers + lctlg %c0,%c15,0(%r4) + larl %r4,.Lfpctl # Restore floating point ctl register + lfpc 0(%r4) + larl %r4,.Lcontinue_psw # Restore PSW flags + lpswe 0(%r4) +.Lcontinue: + br %r14 +.align 16 +.Lrestart_psw: + .long 0x00080000,0x80000000 + .Lrestart_part2 + + .section .data..nosave,"aw",@progbits +.align 8 +.Lcontinue_psw: + .quad 0,.Lcontinue + .previous + + .section .bss +.align 8 +.Lctlregs: + .rept 16 + .quad 0 + .endr +.Lfpctl: + .long 0 + .previous diff --git a/arch/s390/kernel/cache.c b/arch/s390/kernel/cache.c new file mode 100644 index 000000000..bff5e3b6d --- /dev/null +++ b/arch/s390/kernel/cache.c @@ -0,0 +1,181 @@ +/* + * Extract CPU cache information and expose them via sysfs. + * + * Copyright IBM Corp. 2012 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#include <linux/seq_file.h> +#include <linux/cpu.h> +#include <linux/cacheinfo.h> +#include <asm/facility.h> + +enum { + CACHE_SCOPE_NOTEXISTS, + CACHE_SCOPE_PRIVATE, + CACHE_SCOPE_SHARED, + CACHE_SCOPE_RESERVED, +}; + +enum { + CTYPE_SEPARATE, + CTYPE_DATA, + CTYPE_INSTRUCTION, + CTYPE_UNIFIED, +}; + +enum { + EXTRACT_TOPOLOGY, + EXTRACT_LINE_SIZE, + EXTRACT_SIZE, + EXTRACT_ASSOCIATIVITY, +}; + +enum { + CACHE_TI_UNIFIED = 0, + CACHE_TI_DATA = 0, + CACHE_TI_INSTRUCTION, +}; + +struct cache_info { + unsigned char : 4; + unsigned char scope : 2; + unsigned char type : 2; +}; + +#define CACHE_MAX_LEVEL 8 +union cache_topology { + struct cache_info ci[CACHE_MAX_LEVEL]; + unsigned long long raw; +}; + +static const char * const cache_type_string[] = { + "", + "Instruction", + "Data", + "", + "Unified", +}; + +static const enum cache_type cache_type_map[] = { + [CTYPE_SEPARATE] = CACHE_TYPE_SEPARATE, + [CTYPE_DATA] = CACHE_TYPE_DATA, + [CTYPE_INSTRUCTION] = CACHE_TYPE_INST, + [CTYPE_UNIFIED] = CACHE_TYPE_UNIFIED, +}; + +void show_cacheinfo(struct seq_file *m) +{ + struct cpu_cacheinfo *this_cpu_ci; + struct cacheinfo *cache; + int idx; + + if (!test_facility(34)) + return; + get_online_cpus(); + this_cpu_ci = get_cpu_cacheinfo(cpumask_any(cpu_online_mask)); + for (idx = 0; idx < this_cpu_ci->num_leaves; idx++) { + cache = this_cpu_ci->info_list + idx; + seq_printf(m, "cache%-11d: ", idx); + seq_printf(m, "level=%d ", cache->level); + seq_printf(m, "type=%s ", cache_type_string[cache->type]); + seq_printf(m, "scope=%s ", + cache->disable_sysfs ? "Shared" : "Private"); + seq_printf(m, "size=%dK ", cache->size >> 10); + seq_printf(m, "line_size=%u ", cache->coherency_line_size); + seq_printf(m, "associativity=%d", cache->ways_of_associativity); + seq_puts(m, "\n"); + } + put_online_cpus(); +} + +static inline enum cache_type get_cache_type(struct cache_info *ci, int level) +{ + if (level >= CACHE_MAX_LEVEL) + return CACHE_TYPE_NOCACHE; + ci += level; + if (ci->scope != CACHE_SCOPE_SHARED && ci->scope != CACHE_SCOPE_PRIVATE) + return CACHE_TYPE_NOCACHE; + return cache_type_map[ci->type]; +} + +static inline unsigned long ecag(int ai, int li, int ti) +{ + unsigned long cmd, val; + + cmd = ai << 4 | li << 1 | ti; + asm volatile(".insn rsy,0xeb000000004c,%0,0,0(%1)" /* ecag */ + : "=d" (val) : "a" (cmd)); + return val; +} + +static void ci_leaf_init(struct cacheinfo *this_leaf, int private, + enum cache_type type, unsigned int level, int cpu) +{ + int ti, num_sets; + + if (type == CACHE_TYPE_INST) + ti = CACHE_TI_INSTRUCTION; + else + ti = CACHE_TI_UNIFIED; + this_leaf->level = level + 1; + this_leaf->type = type; + this_leaf->coherency_line_size = ecag(EXTRACT_LINE_SIZE, level, ti); + this_leaf->ways_of_associativity = ecag(EXTRACT_ASSOCIATIVITY, level, ti); + this_leaf->size = ecag(EXTRACT_SIZE, level, ti); + num_sets = this_leaf->size / this_leaf->coherency_line_size; + num_sets /= this_leaf->ways_of_associativity; + this_leaf->number_of_sets = num_sets; + cpumask_set_cpu(cpu, &this_leaf->shared_cpu_map); + if (!private) + this_leaf->disable_sysfs = true; +} + +int init_cache_level(unsigned int cpu) +{ + struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); + unsigned int level = 0, leaves = 0; + union cache_topology ct; + enum cache_type ctype; + + if (!this_cpu_ci) + return -EINVAL; + ct.raw = ecag(EXTRACT_TOPOLOGY, 0, 0); + do { + ctype = get_cache_type(&ct.ci[0], level); + if (ctype == CACHE_TYPE_NOCACHE) + break; + /* Separate instruction and data caches */ + leaves += (ctype == CACHE_TYPE_SEPARATE) ? 2 : 1; + } while (++level < CACHE_MAX_LEVEL); + this_cpu_ci->num_levels = level; + this_cpu_ci->num_leaves = leaves; + return 0; +} + +int populate_cache_leaves(unsigned int cpu) +{ + struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); + struct cacheinfo *this_leaf = this_cpu_ci->info_list; + unsigned int level, idx, pvt; + union cache_topology ct; + enum cache_type ctype; + + if (!test_facility(34)) + return -EOPNOTSUPP; + ct.raw = ecag(EXTRACT_TOPOLOGY, 0, 0); + for (idx = 0, level = 0; level < this_cpu_ci->num_levels && + idx < this_cpu_ci->num_leaves; idx++, level++) { + if (!this_leaf) + return -EINVAL; + pvt = (ct.ci[level].scope == CACHE_SCOPE_PRIVATE) ? 1 : 0; + ctype = get_cache_type(&ct.ci[0], level); + if (ctype == CACHE_TYPE_SEPARATE) { + ci_leaf_init(this_leaf++, pvt, CACHE_TYPE_DATA, level, cpu); + ci_leaf_init(this_leaf++, pvt, CACHE_TYPE_INST, level, cpu); + } else { + ci_leaf_init(this_leaf++, pvt, ctype, level, cpu); + } + } + return 0; +} diff --git a/arch/s390/kernel/compat_audit.c b/arch/s390/kernel/compat_audit.c new file mode 100644 index 000000000..d6487bf87 --- /dev/null +++ b/arch/s390/kernel/compat_audit.c @@ -0,0 +1,44 @@ +#undef __s390x__ +#include <asm/unistd.h> +#include "audit.h" + +unsigned s390_dir_class[] = { +#include <asm-generic/audit_dir_write.h> +~0U +}; + +unsigned s390_chattr_class[] = { +#include <asm-generic/audit_change_attr.h> +~0U +}; + +unsigned s390_write_class[] = { +#include <asm-generic/audit_write.h> +~0U +}; + +unsigned s390_read_class[] = { +#include <asm-generic/audit_read.h> +~0U +}; + +unsigned s390_signal_class[] = { +#include <asm-generic/audit_signal.h> +~0U +}; + +int s390_classify_syscall(unsigned syscall) +{ + switch(syscall) { + case __NR_open: + return 2; + case __NR_openat: + return 3; + case __NR_socketcall: + return 4; + case __NR_execve: + return 5; + default: + return 1; + } +} diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c new file mode 100644 index 000000000..437e61159 --- /dev/null +++ b/arch/s390/kernel/compat_linux.c @@ -0,0 +1,520 @@ +/* + * S390 version + * Copyright IBM Corp. 2000 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + * Gerhard Tonn (ton@de.ibm.com) + * Thomas Spatzier (tspat@de.ibm.com) + * + * Conversion between 31bit and 64bit native syscalls. + * + * Heavily inspired by the 32-bit Sparc compat code which is + * Copyright (C) 1997,1998 Jakub Jelinek (jj@sunsite.mff.cuni.cz) + * Copyright (C) 1997 David S. Miller (davem@caip.rutgers.edu) + * + */ + + +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/file.h> +#include <linux/signal.h> +#include <linux/resource.h> +#include <linux/times.h> +#include <linux/smp.h> +#include <linux/sem.h> +#include <linux/msg.h> +#include <linux/shm.h> +#include <linux/uio.h> +#include <linux/quota.h> +#include <linux/module.h> +#include <linux/poll.h> +#include <linux/personality.h> +#include <linux/stat.h> +#include <linux/filter.h> +#include <linux/highmem.h> +#include <linux/highuid.h> +#include <linux/mman.h> +#include <linux/ipv6.h> +#include <linux/in.h> +#include <linux/icmpv6.h> +#include <linux/syscalls.h> +#include <linux/sysctl.h> +#include <linux/binfmts.h> +#include <linux/capability.h> +#include <linux/compat.h> +#include <linux/vfs.h> +#include <linux/ptrace.h> +#include <linux/fadvise.h> +#include <linux/ipc.h> +#include <linux/slab.h> + +#include <asm/types.h> +#include <asm/uaccess.h> + +#include <net/scm.h> +#include <net/sock.h> + +#include "compat_linux.h" + +/* For this source file, we want overflow handling. */ + +#undef high2lowuid +#undef high2lowgid +#undef low2highuid +#undef low2highgid +#undef SET_UID16 +#undef SET_GID16 +#undef NEW_TO_OLD_UID +#undef NEW_TO_OLD_GID +#undef SET_OLDSTAT_UID +#undef SET_OLDSTAT_GID +#undef SET_STAT_UID +#undef SET_STAT_GID + +#define high2lowuid(uid) ((uid) > 65535) ? (u16)overflowuid : (u16)(uid) +#define high2lowgid(gid) ((gid) > 65535) ? (u16)overflowgid : (u16)(gid) +#define low2highuid(uid) ((uid) == (u16)-1) ? (uid_t)-1 : (uid_t)(uid) +#define low2highgid(gid) ((gid) == (u16)-1) ? (gid_t)-1 : (gid_t)(gid) +#define SET_UID16(var, uid) var = high2lowuid(uid) +#define SET_GID16(var, gid) var = high2lowgid(gid) +#define NEW_TO_OLD_UID(uid) high2lowuid(uid) +#define NEW_TO_OLD_GID(gid) high2lowgid(gid) +#define SET_OLDSTAT_UID(stat, uid) (stat).st_uid = high2lowuid(uid) +#define SET_OLDSTAT_GID(stat, gid) (stat).st_gid = high2lowgid(gid) +#define SET_STAT_UID(stat, uid) (stat).st_uid = high2lowuid(uid) +#define SET_STAT_GID(stat, gid) (stat).st_gid = high2lowgid(gid) + +COMPAT_SYSCALL_DEFINE3(s390_chown16, const char __user *, filename, + u16, user, u16, group) +{ + return sys_chown(filename, low2highuid(user), low2highgid(group)); +} + +COMPAT_SYSCALL_DEFINE3(s390_lchown16, const char __user *, + filename, u16, user, u16, group) +{ + return sys_lchown(filename, low2highuid(user), low2highgid(group)); +} + +COMPAT_SYSCALL_DEFINE3(s390_fchown16, unsigned int, fd, u16, user, u16, group) +{ + return sys_fchown(fd, low2highuid(user), low2highgid(group)); +} + +COMPAT_SYSCALL_DEFINE2(s390_setregid16, u16, rgid, u16, egid) +{ + return sys_setregid(low2highgid(rgid), low2highgid(egid)); +} + +COMPAT_SYSCALL_DEFINE1(s390_setgid16, u16, gid) +{ + return sys_setgid((gid_t)gid); +} + +COMPAT_SYSCALL_DEFINE2(s390_setreuid16, u16, ruid, u16, euid) +{ + return sys_setreuid(low2highuid(ruid), low2highuid(euid)); +} + +COMPAT_SYSCALL_DEFINE1(s390_setuid16, u16, uid) +{ + return sys_setuid((uid_t)uid); +} + +COMPAT_SYSCALL_DEFINE3(s390_setresuid16, u16, ruid, u16, euid, u16, suid) +{ + return sys_setresuid(low2highuid(ruid), low2highuid(euid), + low2highuid(suid)); +} + +COMPAT_SYSCALL_DEFINE3(s390_getresuid16, u16 __user *, ruidp, + u16 __user *, euidp, u16 __user *, suidp) +{ + const struct cred *cred = current_cred(); + int retval; + u16 ruid, euid, suid; + + ruid = high2lowuid(from_kuid_munged(cred->user_ns, cred->uid)); + euid = high2lowuid(from_kuid_munged(cred->user_ns, cred->euid)); + suid = high2lowuid(from_kuid_munged(cred->user_ns, cred->suid)); + + if (!(retval = put_user(ruid, ruidp)) && + !(retval = put_user(euid, euidp))) + retval = put_user(suid, suidp); + + return retval; +} + +COMPAT_SYSCALL_DEFINE3(s390_setresgid16, u16, rgid, u16, egid, u16, sgid) +{ + return sys_setresgid(low2highgid(rgid), low2highgid(egid), + low2highgid(sgid)); +} + +COMPAT_SYSCALL_DEFINE3(s390_getresgid16, u16 __user *, rgidp, + u16 __user *, egidp, u16 __user *, sgidp) +{ + const struct cred *cred = current_cred(); + int retval; + u16 rgid, egid, sgid; + + rgid = high2lowgid(from_kgid_munged(cred->user_ns, cred->gid)); + egid = high2lowgid(from_kgid_munged(cred->user_ns, cred->egid)); + sgid = high2lowgid(from_kgid_munged(cred->user_ns, cred->sgid)); + + if (!(retval = put_user(rgid, rgidp)) && + !(retval = put_user(egid, egidp))) + retval = put_user(sgid, sgidp); + + return retval; +} + +COMPAT_SYSCALL_DEFINE1(s390_setfsuid16, u16, uid) +{ + return sys_setfsuid((uid_t)uid); +} + +COMPAT_SYSCALL_DEFINE1(s390_setfsgid16, u16, gid) +{ + return sys_setfsgid((gid_t)gid); +} + +static int groups16_to_user(u16 __user *grouplist, struct group_info *group_info) +{ + struct user_namespace *user_ns = current_user_ns(); + int i; + u16 group; + kgid_t kgid; + + for (i = 0; i < group_info->ngroups; i++) { + kgid = GROUP_AT(group_info, i); + group = (u16)from_kgid_munged(user_ns, kgid); + if (put_user(group, grouplist+i)) + return -EFAULT; + } + + return 0; +} + +static int groups16_from_user(struct group_info *group_info, u16 __user *grouplist) +{ + struct user_namespace *user_ns = current_user_ns(); + int i; + u16 group; + kgid_t kgid; + + for (i = 0; i < group_info->ngroups; i++) { + if (get_user(group, grouplist+i)) + return -EFAULT; + + kgid = make_kgid(user_ns, (gid_t)group); + if (!gid_valid(kgid)) + return -EINVAL; + + GROUP_AT(group_info, i) = kgid; + } + + return 0; +} + +COMPAT_SYSCALL_DEFINE2(s390_getgroups16, int, gidsetsize, u16 __user *, grouplist) +{ + const struct cred *cred = current_cred(); + int i; + + if (gidsetsize < 0) + return -EINVAL; + + get_group_info(cred->group_info); + i = cred->group_info->ngroups; + if (gidsetsize) { + if (i > gidsetsize) { + i = -EINVAL; + goto out; + } + if (groups16_to_user(grouplist, cred->group_info)) { + i = -EFAULT; + goto out; + } + } +out: + put_group_info(cred->group_info); + return i; +} + +COMPAT_SYSCALL_DEFINE2(s390_setgroups16, int, gidsetsize, u16 __user *, grouplist) +{ + struct group_info *group_info; + int retval; + + if (!may_setgroups()) + return -EPERM; + if ((unsigned)gidsetsize > NGROUPS_MAX) + return -EINVAL; + + group_info = groups_alloc(gidsetsize); + if (!group_info) + return -ENOMEM; + retval = groups16_from_user(group_info, grouplist); + if (retval) { + put_group_info(group_info); + return retval; + } + + retval = set_current_groups(group_info); + put_group_info(group_info); + + return retval; +} + +COMPAT_SYSCALL_DEFINE0(s390_getuid16) +{ + return high2lowuid(from_kuid_munged(current_user_ns(), current_uid())); +} + +COMPAT_SYSCALL_DEFINE0(s390_geteuid16) +{ + return high2lowuid(from_kuid_munged(current_user_ns(), current_euid())); +} + +COMPAT_SYSCALL_DEFINE0(s390_getgid16) +{ + return high2lowgid(from_kgid_munged(current_user_ns(), current_gid())); +} + +COMPAT_SYSCALL_DEFINE0(s390_getegid16) +{ + return high2lowgid(from_kgid_munged(current_user_ns(), current_egid())); +} + +#ifdef CONFIG_SYSVIPC +COMPAT_SYSCALL_DEFINE5(s390_ipc, uint, call, int, first, compat_ulong_t, second, + compat_ulong_t, third, compat_uptr_t, ptr) +{ + if (call >> 16) /* hack for backward compatibility */ + return -EINVAL; + return compat_sys_ipc(call, first, second, third, ptr, third); +} +#endif + +COMPAT_SYSCALL_DEFINE3(s390_truncate64, const char __user *, path, u32, high, u32, low) +{ + return sys_truncate(path, (unsigned long)high << 32 | low); +} + +COMPAT_SYSCALL_DEFINE3(s390_ftruncate64, unsigned int, fd, u32, high, u32, low) +{ + return sys_ftruncate(fd, (unsigned long)high << 32 | low); +} + +COMPAT_SYSCALL_DEFINE5(s390_pread64, unsigned int, fd, char __user *, ubuf, + compat_size_t, count, u32, high, u32, low) +{ + if ((compat_ssize_t) count < 0) + return -EINVAL; + return sys_pread64(fd, ubuf, count, (unsigned long)high << 32 | low); +} + +COMPAT_SYSCALL_DEFINE5(s390_pwrite64, unsigned int, fd, const char __user *, ubuf, + compat_size_t, count, u32, high, u32, low) +{ + if ((compat_ssize_t) count < 0) + return -EINVAL; + return sys_pwrite64(fd, ubuf, count, (unsigned long)high << 32 | low); +} + +COMPAT_SYSCALL_DEFINE4(s390_readahead, int, fd, u32, high, u32, low, s32, count) +{ + return sys_readahead(fd, (unsigned long)high << 32 | low, count); +} + +struct stat64_emu31 { + unsigned long long st_dev; + unsigned int __pad1; +#define STAT64_HAS_BROKEN_ST_INO 1 + u32 __st_ino; + unsigned int st_mode; + unsigned int st_nlink; + u32 st_uid; + u32 st_gid; + unsigned long long st_rdev; + unsigned int __pad3; + long st_size; + u32 st_blksize; + unsigned char __pad4[4]; + u32 __pad5; /* future possible st_blocks high bits */ + u32 st_blocks; /* Number 512-byte blocks allocated. */ + u32 st_atime; + u32 __pad6; + u32 st_mtime; + u32 __pad7; + u32 st_ctime; + u32 __pad8; /* will be high 32 bits of ctime someday */ + unsigned long st_ino; +}; + +static int cp_stat64(struct stat64_emu31 __user *ubuf, struct kstat *stat) +{ + struct stat64_emu31 tmp; + + memset(&tmp, 0, sizeof(tmp)); + + tmp.st_dev = huge_encode_dev(stat->dev); + tmp.st_ino = stat->ino; + tmp.__st_ino = (u32)stat->ino; + tmp.st_mode = stat->mode; + tmp.st_nlink = (unsigned int)stat->nlink; + tmp.st_uid = from_kuid_munged(current_user_ns(), stat->uid); + tmp.st_gid = from_kgid_munged(current_user_ns(), stat->gid); + tmp.st_rdev = huge_encode_dev(stat->rdev); + tmp.st_size = stat->size; + tmp.st_blksize = (u32)stat->blksize; + tmp.st_blocks = (u32)stat->blocks; + tmp.st_atime = (u32)stat->atime.tv_sec; + tmp.st_mtime = (u32)stat->mtime.tv_sec; + tmp.st_ctime = (u32)stat->ctime.tv_sec; + + return copy_to_user(ubuf,&tmp,sizeof(tmp)) ? -EFAULT : 0; +} + +COMPAT_SYSCALL_DEFINE2(s390_stat64, const char __user *, filename, struct stat64_emu31 __user *, statbuf) +{ + struct kstat stat; + int ret = vfs_stat(filename, &stat); + if (!ret) + ret = cp_stat64(statbuf, &stat); + return ret; +} + +COMPAT_SYSCALL_DEFINE2(s390_lstat64, const char __user *, filename, struct stat64_emu31 __user *, statbuf) +{ + struct kstat stat; + int ret = vfs_lstat(filename, &stat); + if (!ret) + ret = cp_stat64(statbuf, &stat); + return ret; +} + +COMPAT_SYSCALL_DEFINE2(s390_fstat64, unsigned int, fd, struct stat64_emu31 __user *, statbuf) +{ + struct kstat stat; + int ret = vfs_fstat(fd, &stat); + if (!ret) + ret = cp_stat64(statbuf, &stat); + return ret; +} + +COMPAT_SYSCALL_DEFINE4(s390_fstatat64, unsigned int, dfd, const char __user *, filename, + struct stat64_emu31 __user *, statbuf, int, flag) +{ + struct kstat stat; + int error; + + error = vfs_fstatat(dfd, filename, &stat, flag); + if (error) + return error; + return cp_stat64(statbuf, &stat); +} + +/* + * Linux/i386 didn't use to be able to handle more than + * 4 system call parameters, so these system calls used a memory + * block for parameter passing.. + */ + +struct mmap_arg_struct_emu31 { + compat_ulong_t addr; + compat_ulong_t len; + compat_ulong_t prot; + compat_ulong_t flags; + compat_ulong_t fd; + compat_ulong_t offset; +}; + +COMPAT_SYSCALL_DEFINE1(s390_old_mmap, struct mmap_arg_struct_emu31 __user *, arg) +{ + struct mmap_arg_struct_emu31 a; + + if (copy_from_user(&a, arg, sizeof(a))) + return -EFAULT; + if (a.offset & ~PAGE_MASK) + return -EINVAL; + return sys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd, + a.offset >> PAGE_SHIFT); +} + +COMPAT_SYSCALL_DEFINE1(s390_mmap2, struct mmap_arg_struct_emu31 __user *, arg) +{ + struct mmap_arg_struct_emu31 a; + + if (copy_from_user(&a, arg, sizeof(a))) + return -EFAULT; + return sys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd, a.offset); +} + +COMPAT_SYSCALL_DEFINE3(s390_read, unsigned int, fd, char __user *, buf, compat_size_t, count) +{ + if ((compat_ssize_t) count < 0) + return -EINVAL; + + return sys_read(fd, buf, count); +} + +COMPAT_SYSCALL_DEFINE3(s390_write, unsigned int, fd, const char __user *, buf, compat_size_t, count) +{ + if ((compat_ssize_t) count < 0) + return -EINVAL; + + return sys_write(fd, buf, count); +} + +/* + * 31 bit emulation wrapper functions for sys_fadvise64/fadvise64_64. + * These need to rewrite the advise values for POSIX_FADV_{DONTNEED,NOREUSE} + * because the 31 bit values differ from the 64 bit values. + */ + +COMPAT_SYSCALL_DEFINE5(s390_fadvise64, int, fd, u32, high, u32, low, compat_size_t, len, int, advise) +{ + if (advise == 4) + advise = POSIX_FADV_DONTNEED; + else if (advise == 5) + advise = POSIX_FADV_NOREUSE; + return sys_fadvise64(fd, (unsigned long)high << 32 | low, len, advise); +} + +struct fadvise64_64_args { + int fd; + long long offset; + long long len; + int advice; +}; + +COMPAT_SYSCALL_DEFINE1(s390_fadvise64_64, struct fadvise64_64_args __user *, args) +{ + struct fadvise64_64_args a; + + if ( copy_from_user(&a, args, sizeof(a)) ) + return -EFAULT; + if (a.advice == 4) + a.advice = POSIX_FADV_DONTNEED; + else if (a.advice == 5) + a.advice = POSIX_FADV_NOREUSE; + return sys_fadvise64_64(a.fd, a.offset, a.len, a.advice); +} + +COMPAT_SYSCALL_DEFINE6(s390_sync_file_range, int, fd, u32, offhigh, u32, offlow, + u32, nhigh, u32, nlow, unsigned int, flags) +{ + return sys_sync_file_range(fd, ((loff_t)offhigh << 32) + offlow, + ((u64)nhigh << 32) + nlow, flags); +} + +COMPAT_SYSCALL_DEFINE6(s390_fallocate, int, fd, int, mode, u32, offhigh, u32, offlow, + u32, lenhigh, u32, lenlow) +{ + return sys_fallocate(fd, mode, ((loff_t)offhigh << 32) + offlow, + ((u64)lenhigh << 32) + lenlow); +} diff --git a/arch/s390/kernel/compat_linux.h b/arch/s390/kernel/compat_linux.h new file mode 100644 index 000000000..a0a886c04 --- /dev/null +++ b/arch/s390/kernel/compat_linux.h @@ -0,0 +1,129 @@ +#ifndef _ASM_S390X_S390_H +#define _ASM_S390X_S390_H + +#include <linux/compat.h> +#include <linux/socket.h> +#include <linux/syscalls.h> + +/* Macro that masks the high order bit of an 32 bit pointer and converts it*/ +/* to a 64 bit pointer */ +#define A(__x) ((unsigned long)((__x) & 0x7FFFFFFFUL)) +#define AA(__x) \ + ((unsigned long)(__x)) + +/* Now 32bit compatibility types */ +struct ipc_kludge_32 { + __u32 msgp; /* pointer */ + __s32 msgtyp; +}; + +/* asm/sigcontext.h */ +typedef union +{ + __u64 d; + __u32 f; +} freg_t32; + +typedef struct +{ + unsigned int fpc; + unsigned int pad; + freg_t32 fprs[__NUM_FPRS]; +} _s390_fp_regs32; + +typedef struct +{ + __u32 mask; + __u32 addr; +} _psw_t32 __attribute__ ((aligned(8))); + +typedef struct +{ + _psw_t32 psw; + __u32 gprs[__NUM_GPRS]; + __u32 acrs[__NUM_ACRS]; +} _s390_regs_common32; + +typedef struct +{ + _s390_regs_common32 regs; + _s390_fp_regs32 fpregs; +} _sigregs32; + +typedef struct +{ + __u32 gprs_high[__NUM_GPRS]; + __u64 vxrs_low[__NUM_VXRS_LOW]; + __vector128 vxrs_high[__NUM_VXRS_HIGH]; + __u8 __reserved[128]; +} _sigregs_ext32; + +#define _SIGCONTEXT_NSIG32 64 +#define _SIGCONTEXT_NSIG_BPW32 32 +#define __SIGNAL_FRAMESIZE32 96 +#define _SIGMASK_COPY_SIZE32 (sizeof(u32)*2) + +struct sigcontext32 +{ + __u32 oldmask[_COMPAT_NSIG_WORDS]; + __u32 sregs; /* pointer */ +}; + +/* asm/signal.h */ + +/* asm/ucontext.h */ +struct ucontext32 { + __u32 uc_flags; + __u32 uc_link; /* pointer */ + compat_stack_t uc_stack; + _sigregs32 uc_mcontext; + compat_sigset_t uc_sigmask; + /* Allow for uc_sigmask growth. Glibc uses a 1024-bit sigset_t. */ + unsigned char __unused[128 - sizeof(compat_sigset_t)]; + _sigregs_ext32 uc_mcontext_ext; +}; + +struct stat64_emu31; +struct mmap_arg_struct_emu31; +struct fadvise64_64_args; + +long compat_sys_s390_chown16(const char __user *filename, u16 user, u16 group); +long compat_sys_s390_lchown16(const char __user *filename, u16 user, u16 group); +long compat_sys_s390_fchown16(unsigned int fd, u16 user, u16 group); +long compat_sys_s390_setregid16(u16 rgid, u16 egid); +long compat_sys_s390_setgid16(u16 gid); +long compat_sys_s390_setreuid16(u16 ruid, u16 euid); +long compat_sys_s390_setuid16(u16 uid); +long compat_sys_s390_setresuid16(u16 ruid, u16 euid, u16 suid); +long compat_sys_s390_getresuid16(u16 __user *ruid, u16 __user *euid, u16 __user *suid); +long compat_sys_s390_setresgid16(u16 rgid, u16 egid, u16 sgid); +long compat_sys_s390_getresgid16(u16 __user *rgid, u16 __user *egid, u16 __user *sgid); +long compat_sys_s390_setfsuid16(u16 uid); +long compat_sys_s390_setfsgid16(u16 gid); +long compat_sys_s390_getgroups16(int gidsetsize, u16 __user *grouplist); +long compat_sys_s390_setgroups16(int gidsetsize, u16 __user *grouplist); +long compat_sys_s390_getuid16(void); +long compat_sys_s390_geteuid16(void); +long compat_sys_s390_getgid16(void); +long compat_sys_s390_getegid16(void); +long compat_sys_s390_truncate64(const char __user *path, u32 high, u32 low); +long compat_sys_s390_ftruncate64(unsigned int fd, u32 high, u32 low); +long compat_sys_s390_pread64(unsigned int fd, char __user *ubuf, compat_size_t count, u32 high, u32 low); +long compat_sys_s390_pwrite64(unsigned int fd, const char __user *ubuf, compat_size_t count, u32 high, u32 low); +long compat_sys_s390_readahead(int fd, u32 high, u32 low, s32 count); +long compat_sys_s390_stat64(const char __user *filename, struct stat64_emu31 __user *statbuf); +long compat_sys_s390_lstat64(const char __user *filename, struct stat64_emu31 __user *statbuf); +long compat_sys_s390_fstat64(unsigned int fd, struct stat64_emu31 __user *statbuf); +long compat_sys_s390_fstatat64(unsigned int dfd, const char __user *filename, struct stat64_emu31 __user *statbuf, int flag); +long compat_sys_s390_old_mmap(struct mmap_arg_struct_emu31 __user *arg); +long compat_sys_s390_mmap2(struct mmap_arg_struct_emu31 __user *arg); +long compat_sys_s390_read(unsigned int fd, char __user * buf, compat_size_t count); +long compat_sys_s390_write(unsigned int fd, const char __user * buf, compat_size_t count); +long compat_sys_s390_fadvise64(int fd, u32 high, u32 low, compat_size_t len, int advise); +long compat_sys_s390_fadvise64_64(struct fadvise64_64_args __user *args); +long compat_sys_s390_sync_file_range(int fd, u32 offhigh, u32 offlow, u32 nhigh, u32 nlow, unsigned int flags); +long compat_sys_s390_fallocate(int fd, int mode, u32 offhigh, u32 offlow, u32 lenhigh, u32 lenlow); +long compat_sys_sigreturn(void); +long compat_sys_rt_sigreturn(void); + +#endif /* _ASM_S390X_S390_H */ diff --git a/arch/s390/kernel/compat_ptrace.h b/arch/s390/kernel/compat_ptrace.h new file mode 100644 index 000000000..12b823833 --- /dev/null +++ b/arch/s390/kernel/compat_ptrace.h @@ -0,0 +1,63 @@ +#ifndef _PTRACE32_H +#define _PTRACE32_H + +#include <asm/ptrace.h> /* needed for NUM_CR_WORDS */ +#include "compat_linux.h" /* needed for psw_compat_t */ + +struct compat_per_struct_kernel { + __u32 cr9; /* PER control bits */ + __u32 cr10; /* PER starting address */ + __u32 cr11; /* PER ending address */ + __u32 bits; /* Obsolete software bits */ + __u32 starting_addr; /* User specified start address */ + __u32 ending_addr; /* User specified end address */ + __u16 perc_atmid; /* PER trap ATMID */ + __u32 address; /* PER trap instruction address */ + __u8 access_id; /* PER trap access identification */ +}; + +struct compat_user_regs_struct +{ + psw_compat_t psw; + u32 gprs[NUM_GPRS]; + u32 acrs[NUM_ACRS]; + u32 orig_gpr2; + /* nb: there's a 4-byte hole here */ + s390_fp_regs fp_regs; + /* + * These per registers are in here so that gdb can modify them + * itself as there is no "official" ptrace interface for hardware + * watchpoints. This is the way intel does it. + */ + struct compat_per_struct_kernel per_info; + u32 ieee_instruction_pointer; /* obsolete, always 0 */ +}; + +struct compat_user { + /* We start with the registers, to mimic the way that "memory" + is returned from the ptrace(3,...) function. */ + struct compat_user_regs_struct regs; + /* The rest of this junk is to help gdb figure out what goes where */ + u32 u_tsize; /* Text segment size (pages). */ + u32 u_dsize; /* Data segment size (pages). */ + u32 u_ssize; /* Stack segment size (pages). */ + u32 start_code; /* Starting virtual address of text. */ + u32 start_stack; /* Starting virtual address of stack area. + This is actually the bottom of the stack, + the top of the stack is always found in the + esp register. */ + s32 signal; /* Signal that caused the core dump. */ + u32 u_ar0; /* Used by gdb to help find the values for */ + /* the registers. */ + u32 magic; /* To uniquely identify a core file */ + char u_comm[32]; /* User command that was responsible */ +}; + +typedef struct +{ + __u32 len; + __u32 kernel_addr; + __u32 process_addr; +} compat_ptrace_area; + +#endif /* _PTRACE32_H */ diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c new file mode 100644 index 000000000..fe8d6924e --- /dev/null +++ b/arch/s390/kernel/compat_signal.c @@ -0,0 +1,549 @@ +/* + * Copyright IBM Corp. 2000, 2006 + * Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com) + * Gerhard Tonn (ton@de.ibm.com) + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * 1997-11-28 Modified for POSIX.1b signals by Richard Henderson + */ + +#include <linux/compat.h> +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/smp.h> +#include <linux/kernel.h> +#include <linux/signal.h> +#include <linux/errno.h> +#include <linux/wait.h> +#include <linux/ptrace.h> +#include <linux/unistd.h> +#include <linux/stddef.h> +#include <linux/tty.h> +#include <linux/personality.h> +#include <linux/binfmts.h> +#include <asm/ucontext.h> +#include <asm/uaccess.h> +#include <asm/lowcore.h> +#include <asm/switch_to.h> +#include "compat_linux.h" +#include "compat_ptrace.h" +#include "entry.h" + +typedef struct +{ + __u8 callee_used_stack[__SIGNAL_FRAMESIZE32]; + struct sigcontext32 sc; + _sigregs32 sregs; + int signo; + _sigregs_ext32 sregs_ext; + __u16 svc_insn; /* Offset of svc_insn is NOT fixed! */ +} sigframe32; + +typedef struct +{ + __u8 callee_used_stack[__SIGNAL_FRAMESIZE32]; + __u16 svc_insn; + compat_siginfo_t info; + struct ucontext32 uc; +} rt_sigframe32; + +int copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from) +{ + int err; + + /* If you change siginfo_t structure, please be sure + this code is fixed accordingly. + It should never copy any pad contained in the structure + to avoid security leaks, but must copy the generic + 3 ints plus the relevant union member. + This routine must convert siginfo from 64bit to 32bit as well + at the same time. */ + err = __put_user(from->si_signo, &to->si_signo); + err |= __put_user(from->si_errno, &to->si_errno); + err |= __put_user((short)from->si_code, &to->si_code); + if (from->si_code < 0) + err |= __copy_to_user(&to->_sifields._pad, &from->_sifields._pad, SI_PAD_SIZE); + else { + switch (from->si_code >> 16) { + case __SI_RT >> 16: /* This is not generated by the kernel as of now. */ + case __SI_MESGQ >> 16: + err |= __put_user(from->si_int, &to->si_int); + /* fallthrough */ + case __SI_KILL >> 16: + err |= __put_user(from->si_pid, &to->si_pid); + err |= __put_user(from->si_uid, &to->si_uid); + break; + case __SI_CHLD >> 16: + err |= __put_user(from->si_pid, &to->si_pid); + err |= __put_user(from->si_uid, &to->si_uid); + err |= __put_user(from->si_utime, &to->si_utime); + err |= __put_user(from->si_stime, &to->si_stime); + err |= __put_user(from->si_status, &to->si_status); + break; + case __SI_FAULT >> 16: + err |= __put_user((unsigned long) from->si_addr, + &to->si_addr); + break; + case __SI_POLL >> 16: + err |= __put_user(from->si_band, &to->si_band); + err |= __put_user(from->si_fd, &to->si_fd); + break; + case __SI_TIMER >> 16: + err |= __put_user(from->si_tid, &to->si_tid); + err |= __put_user(from->si_overrun, &to->si_overrun); + err |= __put_user(from->si_int, &to->si_int); + break; + default: + break; + } + } + return err ? -EFAULT : 0; +} + +int copy_siginfo_from_user32(siginfo_t *to, compat_siginfo_t __user *from) +{ + int err; + u32 tmp; + + err = __get_user(to->si_signo, &from->si_signo); + err |= __get_user(to->si_errno, &from->si_errno); + err |= __get_user(to->si_code, &from->si_code); + + if (to->si_code < 0) + err |= __copy_from_user(&to->_sifields._pad, &from->_sifields._pad, SI_PAD_SIZE); + else { + switch (to->si_code >> 16) { + case __SI_RT >> 16: /* This is not generated by the kernel as of now. */ + case __SI_MESGQ >> 16: + err |= __get_user(to->si_int, &from->si_int); + /* fallthrough */ + case __SI_KILL >> 16: + err |= __get_user(to->si_pid, &from->si_pid); + err |= __get_user(to->si_uid, &from->si_uid); + break; + case __SI_CHLD >> 16: + err |= __get_user(to->si_pid, &from->si_pid); + err |= __get_user(to->si_uid, &from->si_uid); + err |= __get_user(to->si_utime, &from->si_utime); + err |= __get_user(to->si_stime, &from->si_stime); + err |= __get_user(to->si_status, &from->si_status); + break; + case __SI_FAULT >> 16: + err |= __get_user(tmp, &from->si_addr); + to->si_addr = (void __force __user *) + (u64) (tmp & PSW32_ADDR_INSN); + break; + case __SI_POLL >> 16: + err |= __get_user(to->si_band, &from->si_band); + err |= __get_user(to->si_fd, &from->si_fd); + break; + case __SI_TIMER >> 16: + err |= __get_user(to->si_tid, &from->si_tid); + err |= __get_user(to->si_overrun, &from->si_overrun); + err |= __get_user(to->si_int, &from->si_int); + break; + default: + break; + } + } + return err ? -EFAULT : 0; +} + +/* Store registers needed to create the signal frame */ +static void store_sigregs(void) +{ + int i; + + save_access_regs(current->thread.acrs); + save_fp_ctl(¤t->thread.fp_regs.fpc); + if (current->thread.vxrs) { + save_vx_regs(current->thread.vxrs); + for (i = 0; i < __NUM_FPRS; i++) + current->thread.fp_regs.fprs[i] = + *(freg_t *)(current->thread.vxrs + i); + } else + save_fp_regs(current->thread.fp_regs.fprs); +} + +/* Load registers after signal return */ +static void load_sigregs(void) +{ + int i; + + restore_access_regs(current->thread.acrs); + /* restore_fp_ctl is done in restore_sigregs */ + if (current->thread.vxrs) { + for (i = 0; i < __NUM_FPRS; i++) + *(freg_t *)(current->thread.vxrs + i) = + current->thread.fp_regs.fprs[i]; + restore_vx_regs(current->thread.vxrs); + } else + restore_fp_regs(current->thread.fp_regs.fprs); +} + +static int save_sigregs32(struct pt_regs *regs, _sigregs32 __user *sregs) +{ + _sigregs32 user_sregs; + int i; + + user_sregs.regs.psw.mask = (__u32)(regs->psw.mask >> 32); + user_sregs.regs.psw.mask &= PSW32_MASK_USER | PSW32_MASK_RI; + user_sregs.regs.psw.mask |= PSW32_USER_BITS; + user_sregs.regs.psw.addr = (__u32) regs->psw.addr | + (__u32)(regs->psw.mask & PSW_MASK_BA); + for (i = 0; i < NUM_GPRS; i++) + user_sregs.regs.gprs[i] = (__u32) regs->gprs[i]; + memcpy(&user_sregs.regs.acrs, current->thread.acrs, + sizeof(user_sregs.regs.acrs)); + memcpy(&user_sregs.fpregs, ¤t->thread.fp_regs, + sizeof(user_sregs.fpregs)); + if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs32))) + return -EFAULT; + return 0; +} + +static int restore_sigregs32(struct pt_regs *regs,_sigregs32 __user *sregs) +{ + _sigregs32 user_sregs; + int i; + + /* Alwys make any pending restarted system call return -EINTR */ + current->restart_block.fn = do_no_restart_syscall; + + if (__copy_from_user(&user_sregs, &sregs->regs, sizeof(user_sregs))) + return -EFAULT; + + if (!is_ri_task(current) && (user_sregs.regs.psw.mask & PSW32_MASK_RI)) + return -EINVAL; + + /* Loading the floating-point-control word can fail. Do that first. */ + if (restore_fp_ctl(&user_sregs.fpregs.fpc)) + return -EINVAL; + + /* Use regs->psw.mask instead of PSW_USER_BITS to preserve PER bit. */ + regs->psw.mask = (regs->psw.mask & ~(PSW_MASK_USER | PSW_MASK_RI)) | + (__u64)(user_sregs.regs.psw.mask & PSW32_MASK_USER) << 32 | + (__u64)(user_sregs.regs.psw.mask & PSW32_MASK_RI) << 32 | + (__u64)(user_sregs.regs.psw.addr & PSW32_ADDR_AMODE); + /* Check for invalid user address space control. */ + if ((regs->psw.mask & PSW_MASK_ASC) == PSW_ASC_HOME) + regs->psw.mask = PSW_ASC_PRIMARY | + (regs->psw.mask & ~PSW_MASK_ASC); + regs->psw.addr = (__u64)(user_sregs.regs.psw.addr & PSW32_ADDR_INSN); + for (i = 0; i < NUM_GPRS; i++) + regs->gprs[i] = (__u64) user_sregs.regs.gprs[i]; + memcpy(¤t->thread.acrs, &user_sregs.regs.acrs, + sizeof(current->thread.acrs)); + + memcpy(¤t->thread.fp_regs, &user_sregs.fpregs, + sizeof(current->thread.fp_regs)); + + clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */ + return 0; +} + +static int save_sigregs_ext32(struct pt_regs *regs, + _sigregs_ext32 __user *sregs_ext) +{ + __u32 gprs_high[NUM_GPRS]; + __u64 vxrs[__NUM_VXRS_LOW]; + int i; + + /* Save high gprs to signal stack */ + for (i = 0; i < NUM_GPRS; i++) + gprs_high[i] = regs->gprs[i] >> 32; + if (__copy_to_user(&sregs_ext->gprs_high, &gprs_high, + sizeof(sregs_ext->gprs_high))) + return -EFAULT; + + /* Save vector registers to signal stack */ + if (current->thread.vxrs) { + for (i = 0; i < __NUM_VXRS_LOW; i++) + vxrs[i] = *((__u64 *)(current->thread.vxrs + i) + 1); + if (__copy_to_user(&sregs_ext->vxrs_low, vxrs, + sizeof(sregs_ext->vxrs_low)) || + __copy_to_user(&sregs_ext->vxrs_high, + current->thread.vxrs + __NUM_VXRS_LOW, + sizeof(sregs_ext->vxrs_high))) + return -EFAULT; + } + return 0; +} + +static int restore_sigregs_ext32(struct pt_regs *regs, + _sigregs_ext32 __user *sregs_ext) +{ + __u32 gprs_high[NUM_GPRS]; + __u64 vxrs[__NUM_VXRS_LOW]; + int i; + + /* Restore high gprs from signal stack */ + if (__copy_from_user(&gprs_high, &sregs_ext->gprs_high, + sizeof(&sregs_ext->gprs_high))) + return -EFAULT; + for (i = 0; i < NUM_GPRS; i++) + *(__u32 *)®s->gprs[i] = gprs_high[i]; + + /* Restore vector registers from signal stack */ + if (current->thread.vxrs) { + if (__copy_from_user(vxrs, &sregs_ext->vxrs_low, + sizeof(sregs_ext->vxrs_low)) || + __copy_from_user(current->thread.vxrs + __NUM_VXRS_LOW, + &sregs_ext->vxrs_high, + sizeof(sregs_ext->vxrs_high))) + return -EFAULT; + for (i = 0; i < __NUM_VXRS_LOW; i++) + *((__u64 *)(current->thread.vxrs + i) + 1) = vxrs[i]; + } + return 0; +} + +COMPAT_SYSCALL_DEFINE0(sigreturn) +{ + struct pt_regs *regs = task_pt_regs(current); + sigframe32 __user *frame = (sigframe32 __user *)regs->gprs[15]; + sigset_t set; + + if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE32)) + goto badframe; + set_current_blocked(&set); + if (restore_sigregs32(regs, &frame->sregs)) + goto badframe; + if (restore_sigregs_ext32(regs, &frame->sregs_ext)) + goto badframe; + load_sigregs(); + return regs->gprs[2]; +badframe: + force_sig(SIGSEGV, current); + return 0; +} + +COMPAT_SYSCALL_DEFINE0(rt_sigreturn) +{ + struct pt_regs *regs = task_pt_regs(current); + rt_sigframe32 __user *frame = (rt_sigframe32 __user *)regs->gprs[15]; + sigset_t set; + + if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) + goto badframe; + set_current_blocked(&set); + if (compat_restore_altstack(&frame->uc.uc_stack)) + goto badframe; + if (restore_sigregs32(regs, &frame->uc.uc_mcontext)) + goto badframe; + if (restore_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext)) + goto badframe; + load_sigregs(); + return regs->gprs[2]; +badframe: + force_sig(SIGSEGV, current); + return 0; +} + +/* + * Set up a signal frame. + */ + + +/* + * Determine which stack to use.. + */ +static inline void __user * +get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size) +{ + unsigned long sp; + + /* Default to using normal stack */ + sp = (unsigned long) A(regs->gprs[15]); + + /* Overflow on alternate signal stack gives SIGSEGV. */ + if (on_sig_stack(sp) && !on_sig_stack((sp - frame_size) & -8UL)) + return (void __user *) -1UL; + + /* This is the X/Open sanctioned signal stack switching. */ + if (ka->sa.sa_flags & SA_ONSTACK) { + if (! sas_ss_flags(sp)) + sp = current->sas_ss_sp + current->sas_ss_size; + } + + return (void __user *)((sp - frame_size) & -8ul); +} + +static int setup_frame32(struct ksignal *ksig, sigset_t *set, + struct pt_regs *regs) +{ + int sig = ksig->sig; + sigframe32 __user *frame; + struct sigcontext32 sc; + unsigned long restorer; + size_t frame_size; + + /* + * gprs_high are always present for 31-bit compat tasks. + * The space for vector registers is only allocated if + * the machine supports it + */ + frame_size = sizeof(*frame) - sizeof(frame->sregs_ext.__reserved); + if (!MACHINE_HAS_VX) + frame_size -= sizeof(frame->sregs_ext.vxrs_low) + + sizeof(frame->sregs_ext.vxrs_high); + frame = get_sigframe(&ksig->ka, regs, frame_size); + if (frame == (void __user *) -1UL) + return -EFAULT; + + /* Set up backchain. */ + if (__put_user(regs->gprs[15], (unsigned int __user *) frame)) + return -EFAULT; + + /* Create struct sigcontext32 on the signal stack */ + memcpy(&sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE32); + sc.sregs = (__u32)(unsigned long __force) &frame->sregs; + if (__copy_to_user(&frame->sc, &sc, sizeof(frame->sc))) + return -EFAULT; + + /* Store registers needed to create the signal frame */ + store_sigregs(); + + /* Create _sigregs32 on the signal stack */ + if (save_sigregs32(regs, &frame->sregs)) + return -EFAULT; + + /* Place signal number on stack to allow backtrace from handler. */ + if (__put_user(regs->gprs[2], (int __force __user *) &frame->signo)) + return -EFAULT; + + /* Create _sigregs_ext32 on the signal stack */ + if (save_sigregs_ext32(regs, &frame->sregs_ext)) + return -EFAULT; + + /* Set up to return from userspace. If provided, use a stub + already in userspace. */ + if (ksig->ka.sa.sa_flags & SA_RESTORER) { + restorer = (unsigned long __force) + ksig->ka.sa.sa_restorer | PSW32_ADDR_AMODE; + } else { + /* Signal frames without vectors registers are short ! */ + __u16 __user *svc = (void __user *) frame + frame_size - 2; + if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn, svc)) + return -EFAULT; + restorer = (unsigned long __force) svc | PSW32_ADDR_AMODE; + } + + /* Set up registers for signal handler */ + regs->gprs[14] = restorer; + regs->gprs[15] = (__force __u64) frame; + /* Force 31 bit amode and default user address space control. */ + regs->psw.mask = PSW_MASK_BA | + (PSW_USER_BITS & PSW_MASK_ASC) | + (regs->psw.mask & ~PSW_MASK_ASC); + regs->psw.addr = (__force __u64) ksig->ka.sa.sa_handler; + + regs->gprs[2] = sig; + regs->gprs[3] = (__force __u64) &frame->sc; + + /* We forgot to include these in the sigcontext. + To avoid breaking binary compatibility, they are passed as args. */ + if (sig == SIGSEGV || sig == SIGBUS || sig == SIGILL || + sig == SIGTRAP || sig == SIGFPE) { + /* set extra registers only for synchronous signals */ + regs->gprs[4] = regs->int_code & 127; + regs->gprs[5] = regs->int_parm_long; + regs->gprs[6] = task_thread_info(current)->last_break; + } + + return 0; +} + +static int setup_rt_frame32(struct ksignal *ksig, sigset_t *set, + struct pt_regs *regs) +{ + rt_sigframe32 __user *frame; + unsigned long restorer; + size_t frame_size; + u32 uc_flags; + + frame_size = sizeof(*frame) - + sizeof(frame->uc.uc_mcontext_ext.__reserved); + /* + * gprs_high are always present for 31-bit compat tasks. + * The space for vector registers is only allocated if + * the machine supports it + */ + uc_flags = UC_GPRS_HIGH; + if (MACHINE_HAS_VX) { + if (current->thread.vxrs) + uc_flags |= UC_VXRS; + } else + frame_size -= sizeof(frame->uc.uc_mcontext_ext.vxrs_low) + + sizeof(frame->uc.uc_mcontext_ext.vxrs_high); + frame = get_sigframe(&ksig->ka, regs, frame_size); + if (frame == (void __user *) -1UL) + return -EFAULT; + + /* Set up backchain. */ + if (__put_user(regs->gprs[15], (unsigned int __force __user *) frame)) + return -EFAULT; + + /* Set up to return from userspace. If provided, use a stub + already in userspace. */ + if (ksig->ka.sa.sa_flags & SA_RESTORER) { + restorer = (unsigned long __force) + ksig->ka.sa.sa_restorer | PSW32_ADDR_AMODE; + } else { + __u16 __user *svc = &frame->svc_insn; + if (__put_user(S390_SYSCALL_OPCODE | __NR_rt_sigreturn, svc)) + return -EFAULT; + restorer = (unsigned long __force) svc | PSW32_ADDR_AMODE; + } + + /* Create siginfo on the signal stack */ + if (copy_siginfo_to_user32(&frame->info, &ksig->info)) + return -EFAULT; + + /* Store registers needed to create the signal frame */ + store_sigregs(); + + /* Create ucontext on the signal stack. */ + if (__put_user(uc_flags, &frame->uc.uc_flags) || + __put_user(0, &frame->uc.uc_link) || + __compat_save_altstack(&frame->uc.uc_stack, regs->gprs[15]) || + save_sigregs32(regs, &frame->uc.uc_mcontext) || + __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)) || + save_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext)) + return -EFAULT; + + /* Set up registers for signal handler */ + regs->gprs[14] = restorer; + regs->gprs[15] = (__force __u64) frame; + /* Force 31 bit amode and default user address space control. */ + regs->psw.mask = PSW_MASK_BA | + (PSW_USER_BITS & PSW_MASK_ASC) | + (regs->psw.mask & ~PSW_MASK_ASC); + regs->psw.addr = (__u64 __force) ksig->ka.sa.sa_handler; + + regs->gprs[2] = ksig->sig; + regs->gprs[3] = (__force __u64) &frame->info; + regs->gprs[4] = (__force __u64) &frame->uc; + regs->gprs[5] = task_thread_info(current)->last_break; + return 0; +} + +/* + * OK, we're invoking a handler + */ + +void handle_signal32(struct ksignal *ksig, sigset_t *oldset, + struct pt_regs *regs) +{ + int ret; + + /* Set up the stack frame */ + if (ksig->ka.sa.sa_flags & SA_SIGINFO) + ret = setup_rt_frame32(ksig, oldset, regs); + else + ret = setup_frame32(ksig, oldset, regs); + + signal_setup_done(ret, ksig, test_thread_flag(TIF_SINGLE_STEP)); +} + diff --git a/arch/s390/kernel/compat_wrapper.c b/arch/s390/kernel/compat_wrapper.c new file mode 100644 index 000000000..d7fa2f0f1 --- /dev/null +++ b/arch/s390/kernel/compat_wrapper.c @@ -0,0 +1,222 @@ +/* + * Compat system call wrappers. + * + * Copyright IBM Corp. 2014 + */ + +#include <linux/syscalls.h> +#include <linux/compat.h> +#include "entry.h" + +#define COMPAT_SYSCALL_WRAP1(name, ...) \ + COMPAT_SYSCALL_WRAPx(1, _##name, __VA_ARGS__) +#define COMPAT_SYSCALL_WRAP2(name, ...) \ + COMPAT_SYSCALL_WRAPx(2, _##name, __VA_ARGS__) +#define COMPAT_SYSCALL_WRAP3(name, ...) \ + COMPAT_SYSCALL_WRAPx(3, _##name, __VA_ARGS__) +#define COMPAT_SYSCALL_WRAP4(name, ...) \ + COMPAT_SYSCALL_WRAPx(4, _##name, __VA_ARGS__) +#define COMPAT_SYSCALL_WRAP5(name, ...) \ + COMPAT_SYSCALL_WRAPx(5, _##name, __VA_ARGS__) +#define COMPAT_SYSCALL_WRAP6(name, ...) \ + COMPAT_SYSCALL_WRAPx(6, _##name, __VA_ARGS__) + +#define __SC_COMPAT_TYPE(t, a) \ + __typeof(__builtin_choose_expr(sizeof(t) > 4, 0L, (t)0)) a + +#define __SC_COMPAT_CAST(t, a) \ +({ \ + long __ReS = a; \ + \ + BUILD_BUG_ON((sizeof(t) > 4) && !__TYPE_IS_L(t) && \ + !__TYPE_IS_UL(t) && !__TYPE_IS_PTR(t)); \ + if (__TYPE_IS_L(t)) \ + __ReS = (s32)a; \ + if (__TYPE_IS_UL(t)) \ + __ReS = (u32)a; \ + if (__TYPE_IS_PTR(t)) \ + __ReS = a & 0x7fffffff; \ + (t)__ReS; \ +}) + +/* + * The COMPAT_SYSCALL_WRAP macro generates system call wrappers to be used by + * compat tasks. These wrappers will only be used for system calls where only + * the system call arguments need sign or zero extension or zeroing of the upper + * 33 bits of pointers. + * Note: since the wrapper function will afterwards call a system call which + * again performs zero and sign extension for all system call arguments with + * a size of less than eight bytes, these compat wrappers only touch those + * system call arguments with a size of eight bytes ((unsigned) long and + * pointers). Zero and sign extension for e.g. int parameters will be done by + * the regular system call wrappers. + */ +#define COMPAT_SYSCALL_WRAPx(x, name, ...) \ + asmlinkage long sys##name(__MAP(x,__SC_DECL,__VA_ARGS__)); \ + asmlinkage long compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__));\ + asmlinkage long compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__)) \ + { \ + return sys##name(__MAP(x,__SC_COMPAT_CAST,__VA_ARGS__)); \ + } + +COMPAT_SYSCALL_WRAP1(exit, int, error_code); +COMPAT_SYSCALL_WRAP1(close, unsigned int, fd); +COMPAT_SYSCALL_WRAP2(creat, const char __user *, pathname, umode_t, mode); +COMPAT_SYSCALL_WRAP2(link, const char __user *, oldname, const char __user *, newname); +COMPAT_SYSCALL_WRAP1(unlink, const char __user *, pathname); +COMPAT_SYSCALL_WRAP1(chdir, const char __user *, filename); +COMPAT_SYSCALL_WRAP3(mknod, const char __user *, filename, umode_t, mode, unsigned, dev); +COMPAT_SYSCALL_WRAP2(chmod, const char __user *, filename, umode_t, mode); +COMPAT_SYSCALL_WRAP1(oldumount, char __user *, name); +COMPAT_SYSCALL_WRAP1(alarm, unsigned int, seconds); +COMPAT_SYSCALL_WRAP2(access, const char __user *, filename, int, mode); +COMPAT_SYSCALL_WRAP1(nice, int, increment); +COMPAT_SYSCALL_WRAP2(kill, int, pid, int, sig); +COMPAT_SYSCALL_WRAP2(rename, const char __user *, oldname, const char __user *, newname); +COMPAT_SYSCALL_WRAP2(mkdir, const char __user *, pathname, umode_t, mode); +COMPAT_SYSCALL_WRAP1(rmdir, const char __user *, pathname); +COMPAT_SYSCALL_WRAP1(dup, unsigned int, fildes); +COMPAT_SYSCALL_WRAP1(pipe, int __user *, fildes); +COMPAT_SYSCALL_WRAP1(brk, unsigned long, brk); +COMPAT_SYSCALL_WRAP2(signal, int, sig, __sighandler_t, handler); +COMPAT_SYSCALL_WRAP1(acct, const char __user *, name); +COMPAT_SYSCALL_WRAP2(umount, char __user *, name, int, flags); +COMPAT_SYSCALL_WRAP2(setpgid, pid_t, pid, pid_t, pgid); +COMPAT_SYSCALL_WRAP1(umask, int, mask); +COMPAT_SYSCALL_WRAP1(chroot, const char __user *, filename); +COMPAT_SYSCALL_WRAP2(dup2, unsigned int, oldfd, unsigned int, newfd); +COMPAT_SYSCALL_WRAP3(sigsuspend, int, unused1, int, unused2, old_sigset_t, mask); +COMPAT_SYSCALL_WRAP2(sethostname, char __user *, name, int, len); +COMPAT_SYSCALL_WRAP2(symlink, const char __user *, old, const char __user *, new); +COMPAT_SYSCALL_WRAP3(readlink, const char __user *, path, char __user *, buf, int, bufsiz); +COMPAT_SYSCALL_WRAP1(uselib, const char __user *, library); +COMPAT_SYSCALL_WRAP2(swapon, const char __user *, specialfile, int, swap_flags); +COMPAT_SYSCALL_WRAP4(reboot, int, magic1, int, magic2, unsigned int, cmd, void __user *, arg); +COMPAT_SYSCALL_WRAP2(munmap, unsigned long, addr, size_t, len); +COMPAT_SYSCALL_WRAP2(fchmod, unsigned int, fd, umode_t, mode); +COMPAT_SYSCALL_WRAP2(getpriority, int, which, int, who); +COMPAT_SYSCALL_WRAP3(setpriority, int, which, int, who, int, niceval); +COMPAT_SYSCALL_WRAP3(syslog, int, type, char __user *, buf, int, len); +COMPAT_SYSCALL_WRAP1(swapoff, const char __user *, specialfile); +COMPAT_SYSCALL_WRAP1(fsync, unsigned int, fd); +COMPAT_SYSCALL_WRAP2(setdomainname, char __user *, name, int, len); +COMPAT_SYSCALL_WRAP1(newuname, struct new_utsname __user *, name); +COMPAT_SYSCALL_WRAP3(mprotect, unsigned long, start, size_t, len, unsigned long, prot); +COMPAT_SYSCALL_WRAP3(init_module, void __user *, umod, unsigned long, len, const char __user *, uargs); +COMPAT_SYSCALL_WRAP2(delete_module, const char __user *, name_user, unsigned int, flags); +COMPAT_SYSCALL_WRAP4(quotactl, unsigned int, cmd, const char __user *, special, qid_t, id, void __user *, addr); +COMPAT_SYSCALL_WRAP1(getpgid, pid_t, pid); +COMPAT_SYSCALL_WRAP1(fchdir, unsigned int, fd); +COMPAT_SYSCALL_WRAP2(bdflush, int, func, long, data); +COMPAT_SYSCALL_WRAP3(sysfs, int, option, unsigned long, arg1, unsigned long, arg2); +COMPAT_SYSCALL_WRAP1(s390_personality, unsigned int, personality); +COMPAT_SYSCALL_WRAP5(llseek, unsigned int, fd, unsigned long, high, unsigned long, low, loff_t __user *, result, unsigned int, whence); +COMPAT_SYSCALL_WRAP2(flock, unsigned int, fd, unsigned int, cmd); +COMPAT_SYSCALL_WRAP3(msync, unsigned long, start, size_t, len, int, flags); +COMPAT_SYSCALL_WRAP1(getsid, pid_t, pid); +COMPAT_SYSCALL_WRAP1(fdatasync, unsigned int, fd); +COMPAT_SYSCALL_WRAP2(mlock, unsigned long, start, size_t, len); +COMPAT_SYSCALL_WRAP2(munlock, unsigned long, start, size_t, len); +COMPAT_SYSCALL_WRAP1(mlockall, int, flags); +COMPAT_SYSCALL_WRAP2(sched_setparam, pid_t, pid, struct sched_param __user *, param); +COMPAT_SYSCALL_WRAP2(sched_getparam, pid_t, pid, struct sched_param __user *, param); +COMPAT_SYSCALL_WRAP3(sched_setscheduler, pid_t, pid, int, policy, struct sched_param __user *, param); +COMPAT_SYSCALL_WRAP1(sched_getscheduler, pid_t, pid); +COMPAT_SYSCALL_WRAP1(sched_get_priority_max, int, policy); +COMPAT_SYSCALL_WRAP1(sched_get_priority_min, int, policy); +COMPAT_SYSCALL_WRAP5(mremap, unsigned long, addr, unsigned long, old_len, unsigned long, new_len, unsigned long, flags, unsigned long, new_addr); +COMPAT_SYSCALL_WRAP3(poll, struct pollfd __user *, ufds, unsigned int, nfds, int, timeout); +COMPAT_SYSCALL_WRAP5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, unsigned long, arg4, unsigned long, arg5); +COMPAT_SYSCALL_WRAP2(getcwd, char __user *, buf, unsigned long, size); +COMPAT_SYSCALL_WRAP2(capget, cap_user_header_t, header, cap_user_data_t, dataptr); +COMPAT_SYSCALL_WRAP2(capset, cap_user_header_t, header, const cap_user_data_t, data); +COMPAT_SYSCALL_WRAP3(lchown, const char __user *, filename, uid_t, user, gid_t, group); +COMPAT_SYSCALL_WRAP2(setreuid, uid_t, ruid, uid_t, euid); +COMPAT_SYSCALL_WRAP2(setregid, gid_t, rgid, gid_t, egid); +COMPAT_SYSCALL_WRAP2(getgroups, int, gidsetsize, gid_t __user *, grouplist); +COMPAT_SYSCALL_WRAP2(setgroups, int, gidsetsize, gid_t __user *, grouplist); +COMPAT_SYSCALL_WRAP3(fchown, unsigned int, fd, uid_t, user, gid_t, group); +COMPAT_SYSCALL_WRAP3(setresuid, uid_t, ruid, uid_t, euid, uid_t, suid); +COMPAT_SYSCALL_WRAP3(getresuid, uid_t __user *, ruid, uid_t __user *, euid, uid_t __user *, suid); +COMPAT_SYSCALL_WRAP3(setresgid, gid_t, rgid, gid_t, egid, gid_t, sgid); +COMPAT_SYSCALL_WRAP3(getresgid, gid_t __user *, rgid, gid_t __user *, egid, gid_t __user *, sgid); +COMPAT_SYSCALL_WRAP3(chown, const char __user *, filename, uid_t, user, gid_t, group); +COMPAT_SYSCALL_WRAP1(setuid, uid_t, uid); +COMPAT_SYSCALL_WRAP1(setgid, gid_t, gid); +COMPAT_SYSCALL_WRAP1(setfsuid, uid_t, uid); +COMPAT_SYSCALL_WRAP1(setfsgid, gid_t, gid); +COMPAT_SYSCALL_WRAP2(pivot_root, const char __user *, new_root, const char __user *, put_old); +COMPAT_SYSCALL_WRAP3(mincore, unsigned long, start, size_t, len, unsigned char __user *, vec); +COMPAT_SYSCALL_WRAP3(madvise, unsigned long, start, size_t, len, int, behavior); +COMPAT_SYSCALL_WRAP5(setxattr, const char __user *, path, const char __user *, name, const void __user *, value, size_t, size, int, flags); +COMPAT_SYSCALL_WRAP5(lsetxattr, const char __user *, path, const char __user *, name, const void __user *, value, size_t, size, int, flags); +COMPAT_SYSCALL_WRAP5(fsetxattr, int, fd, const char __user *, name, const void __user *, value, size_t, size, int, flags); +COMPAT_SYSCALL_WRAP3(getdents64, unsigned int, fd, struct linux_dirent64 __user *, dirent, unsigned int, count); +COMPAT_SYSCALL_WRAP4(getxattr, const char __user *, path, const char __user *, name, void __user *, value, size_t, size); +COMPAT_SYSCALL_WRAP4(lgetxattr, const char __user *, path, const char __user *, name, void __user *, value, size_t, size); +COMPAT_SYSCALL_WRAP4(fgetxattr, int, fd, const char __user *, name, void __user *, value, size_t, size); +COMPAT_SYSCALL_WRAP3(listxattr, const char __user *, path, char __user *, list, size_t, size); +COMPAT_SYSCALL_WRAP3(llistxattr, const char __user *, path, char __user *, list, size_t, size); +COMPAT_SYSCALL_WRAP3(flistxattr, int, fd, char __user *, list, size_t, size); +COMPAT_SYSCALL_WRAP2(removexattr, const char __user *, path, const char __user *, name); +COMPAT_SYSCALL_WRAP2(lremovexattr, const char __user *, path, const char __user *, name); +COMPAT_SYSCALL_WRAP2(fremovexattr, int, fd, const char __user *, name); +COMPAT_SYSCALL_WRAP1(exit_group, int, error_code); +COMPAT_SYSCALL_WRAP1(set_tid_address, int __user *, tidptr); +COMPAT_SYSCALL_WRAP1(epoll_create, int, size); +COMPAT_SYSCALL_WRAP4(epoll_ctl, int, epfd, int, op, int, fd, struct epoll_event __user *, event); +COMPAT_SYSCALL_WRAP4(epoll_wait, int, epfd, struct epoll_event __user *, events, int, maxevents, int, timeout); +COMPAT_SYSCALL_WRAP1(timer_getoverrun, timer_t, timer_id); +COMPAT_SYSCALL_WRAP1(timer_delete, compat_timer_t, compat_timer_id); +COMPAT_SYSCALL_WRAP1(io_destroy, aio_context_t, ctx); +COMPAT_SYSCALL_WRAP3(io_cancel, aio_context_t, ctx_id, struct iocb __user *, iocb, struct io_event __user *, result); +COMPAT_SYSCALL_WRAP1(mq_unlink, const char __user *, name); +COMPAT_SYSCALL_WRAP5(add_key, const char __user *, tp, const char __user *, dsc, const void __user *, pld, size_t, len, key_serial_t, id); +COMPAT_SYSCALL_WRAP4(request_key, const char __user *, tp, const char __user *, dsc, const char __user *, info, key_serial_t, id); +COMPAT_SYSCALL_WRAP5(remap_file_pages, unsigned long, start, unsigned long, size, unsigned long, prot, unsigned long, pgoff, unsigned long, flags); +COMPAT_SYSCALL_WRAP3(ioprio_set, int, which, int, who, int, ioprio); +COMPAT_SYSCALL_WRAP2(ioprio_get, int, which, int, who); +COMPAT_SYSCALL_WRAP3(inotify_add_watch, int, fd, const char __user *, path, u32, mask); +COMPAT_SYSCALL_WRAP2(inotify_rm_watch, int, fd, __s32, wd); +COMPAT_SYSCALL_WRAP3(mkdirat, int, dfd, const char __user *, pathname, umode_t, mode); +COMPAT_SYSCALL_WRAP4(mknodat, int, dfd, const char __user *, filename, umode_t, mode, unsigned, dev); +COMPAT_SYSCALL_WRAP5(fchownat, int, dfd, const char __user *, filename, uid_t, user, gid_t, group, int, flag); +COMPAT_SYSCALL_WRAP3(unlinkat, int, dfd, const char __user *, pathname, int, flag); +COMPAT_SYSCALL_WRAP4(renameat, int, olddfd, const char __user *, oldname, int, newdfd, const char __user *, newname); +COMPAT_SYSCALL_WRAP5(linkat, int, olddfd, const char __user *, oldname, int, newdfd, const char __user *, newname, int, flags); +COMPAT_SYSCALL_WRAP3(symlinkat, const char __user *, oldname, int, newdfd, const char __user *, newname); +COMPAT_SYSCALL_WRAP4(readlinkat, int, dfd, const char __user *, path, char __user *, buf, int, bufsiz); +COMPAT_SYSCALL_WRAP3(fchmodat, int, dfd, const char __user *, filename, umode_t, mode); +COMPAT_SYSCALL_WRAP3(faccessat, int, dfd, const char __user *, filename, int, mode); +COMPAT_SYSCALL_WRAP1(unshare, unsigned long, unshare_flags); +COMPAT_SYSCALL_WRAP6(splice, int, fd_in, loff_t __user *, off_in, int, fd_out, loff_t __user *, off_out, size_t, len, unsigned int, flags); +COMPAT_SYSCALL_WRAP4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags); +COMPAT_SYSCALL_WRAP3(getcpu, unsigned __user *, cpu, unsigned __user *, node, struct getcpu_cache __user *, cache); +COMPAT_SYSCALL_WRAP1(eventfd, unsigned int, count); +COMPAT_SYSCALL_WRAP2(timerfd_create, int, clockid, int, flags); +COMPAT_SYSCALL_WRAP2(eventfd2, unsigned int, count, int, flags); +COMPAT_SYSCALL_WRAP1(inotify_init1, int, flags); +COMPAT_SYSCALL_WRAP2(pipe2, int __user *, fildes, int, flags); +COMPAT_SYSCALL_WRAP3(dup3, unsigned int, oldfd, unsigned int, newfd, int, flags); +COMPAT_SYSCALL_WRAP1(epoll_create1, int, flags); +COMPAT_SYSCALL_WRAP2(tkill, int, pid, int, sig); +COMPAT_SYSCALL_WRAP3(tgkill, int, tgid, int, pid, int, sig); +COMPAT_SYSCALL_WRAP5(perf_event_open, struct perf_event_attr __user *, attr_uptr, pid_t, pid, int, cpu, int, group_fd, unsigned long, flags); +COMPAT_SYSCALL_WRAP5(clone, unsigned long, newsp, unsigned long, clone_flags, int __user *, parent_tidptr, int __user *, child_tidptr, int, tls_val); +COMPAT_SYSCALL_WRAP2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags); +COMPAT_SYSCALL_WRAP4(prlimit64, pid_t, pid, unsigned int, resource, const struct rlimit64 __user *, new_rlim, struct rlimit64 __user *, old_rlim); +COMPAT_SYSCALL_WRAP5(name_to_handle_at, int, dfd, const char __user *, name, struct file_handle __user *, handle, int __user *, mnt_id, int, flag); +COMPAT_SYSCALL_WRAP1(syncfs, int, fd); +COMPAT_SYSCALL_WRAP2(setns, int, fd, int, nstype); +COMPAT_SYSCALL_WRAP2(s390_runtime_instr, int, command, int, signum); +COMPAT_SYSCALL_WRAP5(kcmp, pid_t, pid1, pid_t, pid2, int, type, unsigned long, idx1, unsigned long, idx2); +COMPAT_SYSCALL_WRAP3(finit_module, int, fd, const char __user *, uargs, int, flags); +COMPAT_SYSCALL_WRAP3(sched_setattr, pid_t, pid, struct sched_attr __user *, attr, unsigned int, flags); +COMPAT_SYSCALL_WRAP4(sched_getattr, pid_t, pid, struct sched_attr __user *, attr, unsigned int, size, unsigned int, flags); +COMPAT_SYSCALL_WRAP5(renameat2, int, olddfd, const char __user *, oldname, int, newdfd, const char __user *, newname, unsigned int, flags); +COMPAT_SYSCALL_WRAP3(seccomp, unsigned int, op, unsigned int, flags, const char __user *, uargs) +COMPAT_SYSCALL_WRAP3(getrandom, char __user *, buf, size_t, count, unsigned int, flags) +COMPAT_SYSCALL_WRAP2(memfd_create, const char __user *, uname, unsigned int, flags) +COMPAT_SYSCALL_WRAP3(bpf, int, cmd, union bpf_attr *, attr, unsigned int, size); +COMPAT_SYSCALL_WRAP3(s390_pci_mmio_write, const unsigned long, mmio_addr, const void __user *, user_buffer, const size_t, length); +COMPAT_SYSCALL_WRAP3(s390_pci_mmio_read, const unsigned long, mmio_addr, void __user *, user_buffer, const size_t, length); diff --git a/arch/s390/kernel/cpcmd.c b/arch/s390/kernel/cpcmd.c new file mode 100644 index 000000000..199ec92ef --- /dev/null +++ b/arch/s390/kernel/cpcmd.c @@ -0,0 +1,113 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2007 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + * Christian Borntraeger (cborntra@de.ibm.com), + */ + +#define KMSG_COMPONENT "cpcmd" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/stddef.h> +#include <linux/string.h> +#include <asm/ebcdic.h> +#include <asm/cpcmd.h> +#include <asm/io.h> + +static DEFINE_SPINLOCK(cpcmd_lock); +static char cpcmd_buf[241]; + +static int diag8_noresponse(int cmdlen) +{ + register unsigned long reg2 asm ("2") = (addr_t) cpcmd_buf; + register unsigned long reg3 asm ("3") = cmdlen; + + asm volatile( + " sam31\n" + " diag %1,%0,0x8\n" + " sam64\n" + : "+d" (reg3) : "d" (reg2) : "cc"); + return reg3; +} + +static int diag8_response(int cmdlen, char *response, int *rlen) +{ + register unsigned long reg2 asm ("2") = (addr_t) cpcmd_buf; + register unsigned long reg3 asm ("3") = (addr_t) response; + register unsigned long reg4 asm ("4") = cmdlen | 0x40000000L; + register unsigned long reg5 asm ("5") = *rlen; + + asm volatile( + " sam31\n" + " diag %2,%0,0x8\n" + " sam64\n" + " brc 8,1f\n" + " agr %1,%4\n" + "1:\n" + : "+d" (reg4), "+d" (reg5) + : "d" (reg2), "d" (reg3), "d" (*rlen) : "cc"); + *rlen = reg5; + return reg4; +} + +/* + * __cpcmd has some restrictions over cpcmd + * - the response buffer must reside below 2GB (if any) + * - __cpcmd is unlocked and therefore not SMP-safe + */ +int __cpcmd(const char *cmd, char *response, int rlen, int *response_code) +{ + int cmdlen; + int rc; + int response_len; + + cmdlen = strlen(cmd); + BUG_ON(cmdlen > 240); + memcpy(cpcmd_buf, cmd, cmdlen); + ASCEBC(cpcmd_buf, cmdlen); + + if (response) { + memset(response, 0, rlen); + response_len = rlen; + rc = diag8_response(cmdlen, response, &rlen); + EBCASC(response, response_len); + } else { + rc = diag8_noresponse(cmdlen); + } + if (response_code) + *response_code = rc; + return rlen; +} +EXPORT_SYMBOL(__cpcmd); + +int cpcmd(const char *cmd, char *response, int rlen, int *response_code) +{ + char *lowbuf; + int len; + unsigned long flags; + + if ((virt_to_phys(response) != (unsigned long) response) || + (((unsigned long)response + rlen) >> 31)) { + lowbuf = kmalloc(rlen, GFP_KERNEL | GFP_DMA); + if (!lowbuf) { + pr_warning("The cpcmd kernel function failed to " + "allocate a response buffer\n"); + return -ENOMEM; + } + spin_lock_irqsave(&cpcmd_lock, flags); + len = __cpcmd(cmd, lowbuf, rlen, response_code); + spin_unlock_irqrestore(&cpcmd_lock, flags); + memcpy(response, lowbuf, rlen); + kfree(lowbuf); + } else { + spin_lock_irqsave(&cpcmd_lock, flags); + len = __cpcmd(cmd, response, rlen, response_code); + spin_unlock_irqrestore(&cpcmd_lock, flags); + } + return len; +} +EXPORT_SYMBOL(cpcmd); diff --git a/arch/s390/kernel/crash_dump.c b/arch/s390/kernel/crash_dump.c new file mode 100644 index 000000000..49b74454d --- /dev/null +++ b/arch/s390/kernel/crash_dump.c @@ -0,0 +1,687 @@ +/* + * S390 kdump implementation + * + * Copyright IBM Corp. 2011 + * Author(s): Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#include <linux/crash_dump.h> +#include <asm/lowcore.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/gfp.h> +#include <linux/slab.h> +#include <linux/bootmem.h> +#include <linux/elf.h> +#include <linux/memblock.h> +#include <asm/os_info.h> +#include <asm/elf.h> +#include <asm/ipl.h> +#include <asm/sclp.h> + +#define PTR_ADD(x, y) (((char *) (x)) + ((unsigned long) (y))) +#define PTR_SUB(x, y) (((char *) (x)) - ((unsigned long) (y))) +#define PTR_DIFF(x, y) ((unsigned long)(((char *) (x)) - ((unsigned long) (y)))) + +static struct memblock_region oldmem_region; + +static struct memblock_type oldmem_type = { + .cnt = 1, + .max = 1, + .total_size = 0, + .regions = &oldmem_region, +}; + +#define for_each_dump_mem_range(i, nid, p_start, p_end, p_nid) \ + for (i = 0, __next_mem_range(&i, nid, &memblock.physmem, \ + &oldmem_type, p_start, \ + p_end, p_nid); \ + i != (u64)ULLONG_MAX; \ + __next_mem_range(&i, nid, &memblock.physmem, \ + &oldmem_type, \ + p_start, p_end, p_nid)) + +struct dump_save_areas dump_save_areas; + +/* + * Allocate and add a save area for a CPU + */ +struct save_area_ext *dump_save_area_create(int cpu) +{ + struct save_area_ext **save_areas, *save_area; + + save_area = kmalloc(sizeof(*save_area), GFP_KERNEL); + if (!save_area) + return NULL; + if (cpu + 1 > dump_save_areas.count) { + dump_save_areas.count = cpu + 1; + save_areas = krealloc(dump_save_areas.areas, + dump_save_areas.count * sizeof(void *), + GFP_KERNEL | __GFP_ZERO); + if (!save_areas) { + kfree(save_area); + return NULL; + } + dump_save_areas.areas = save_areas; + } + dump_save_areas.areas[cpu] = save_area; + return save_area; +} + +/* + * Return physical address for virtual address + */ +static inline void *load_real_addr(void *addr) +{ + unsigned long real_addr; + + asm volatile( + " lra %0,0(%1)\n" + " jz 0f\n" + " la %0,0\n" + "0:" + : "=a" (real_addr) : "a" (addr) : "cc"); + return (void *)real_addr; +} + +/* + * Copy real to virtual or real memory + */ +static int copy_from_realmem(void *dest, void *src, size_t count) +{ + unsigned long size; + + if (!count) + return 0; + if (!is_vmalloc_or_module_addr(dest)) + return memcpy_real(dest, src, count); + do { + size = min(count, PAGE_SIZE - (__pa(dest) & ~PAGE_MASK)); + if (memcpy_real(load_real_addr(dest), src, size)) + return -EFAULT; + count -= size; + dest += size; + src += size; + } while (count); + return 0; +} + +/* + * Pointer to ELF header in new kernel + */ +static void *elfcorehdr_newmem; + +/* + * Copy one page from zfcpdump "oldmem" + * + * For pages below HSA size memory from the HSA is copied. Otherwise + * real memory copy is used. + */ +static ssize_t copy_oldmem_page_zfcpdump(char *buf, size_t csize, + unsigned long src, int userbuf) +{ + int rc; + + if (src < sclp_get_hsa_size()) { + rc = memcpy_hsa(buf, src, csize, userbuf); + } else { + if (userbuf) + rc = copy_to_user_real((void __force __user *) buf, + (void *) src, csize); + else + rc = memcpy_real(buf, (void *) src, csize); + } + return rc ? rc : csize; +} + +/* + * Copy one page from kdump "oldmem" + * + * For the kdump reserved memory this functions performs a swap operation: + * - [OLDMEM_BASE - OLDMEM_BASE + OLDMEM_SIZE] is mapped to [0 - OLDMEM_SIZE]. + * - [0 - OLDMEM_SIZE] is mapped to [OLDMEM_BASE - OLDMEM_BASE + OLDMEM_SIZE] + */ +static ssize_t copy_oldmem_page_kdump(char *buf, size_t csize, + unsigned long src, int userbuf) + +{ + int rc; + + if (src < OLDMEM_SIZE) + src += OLDMEM_BASE; + else if (src > OLDMEM_BASE && + src < OLDMEM_BASE + OLDMEM_SIZE) + src -= OLDMEM_BASE; + if (userbuf) + rc = copy_to_user_real((void __force __user *) buf, + (void *) src, csize); + else + rc = copy_from_realmem(buf, (void *) src, csize); + return (rc == 0) ? rc : csize; +} + +/* + * Copy one page from "oldmem" + */ +ssize_t copy_oldmem_page(unsigned long pfn, char *buf, size_t csize, + unsigned long offset, int userbuf) +{ + unsigned long src; + + if (!csize) + return 0; + src = (pfn << PAGE_SHIFT) + offset; + if (OLDMEM_BASE) + return copy_oldmem_page_kdump(buf, csize, src, userbuf); + else + return copy_oldmem_page_zfcpdump(buf, csize, src, userbuf); +} + +/* + * Remap "oldmem" for kdump + * + * For the kdump reserved memory this functions performs a swap operation: + * [0 - OLDMEM_SIZE] is mapped to [OLDMEM_BASE - OLDMEM_BASE + OLDMEM_SIZE] + */ +static int remap_oldmem_pfn_range_kdump(struct vm_area_struct *vma, + unsigned long from, unsigned long pfn, + unsigned long size, pgprot_t prot) +{ + unsigned long size_old; + int rc; + + if (pfn < OLDMEM_SIZE >> PAGE_SHIFT) { + size_old = min(size, OLDMEM_SIZE - (pfn << PAGE_SHIFT)); + rc = remap_pfn_range(vma, from, + pfn + (OLDMEM_BASE >> PAGE_SHIFT), + size_old, prot); + if (rc || size == size_old) + return rc; + size -= size_old; + from += size_old; + pfn += size_old >> PAGE_SHIFT; + } + return remap_pfn_range(vma, from, pfn, size, prot); +} + +/* + * Remap "oldmem" for zfcpdump + * + * We only map available memory above HSA size. Memory below HSA size + * is read on demand using the copy_oldmem_page() function. + */ +static int remap_oldmem_pfn_range_zfcpdump(struct vm_area_struct *vma, + unsigned long from, + unsigned long pfn, + unsigned long size, pgprot_t prot) +{ + unsigned long hsa_end = sclp_get_hsa_size(); + unsigned long size_hsa; + + if (pfn < hsa_end >> PAGE_SHIFT) { + size_hsa = min(size, hsa_end - (pfn << PAGE_SHIFT)); + if (size == size_hsa) + return 0; + size -= size_hsa; + from += size_hsa; + pfn += size_hsa >> PAGE_SHIFT; + } + return remap_pfn_range(vma, from, pfn, size, prot); +} + +/* + * Remap "oldmem" for kdump or zfcpdump + */ +int remap_oldmem_pfn_range(struct vm_area_struct *vma, unsigned long from, + unsigned long pfn, unsigned long size, pgprot_t prot) +{ + if (OLDMEM_BASE) + return remap_oldmem_pfn_range_kdump(vma, from, pfn, size, prot); + else + return remap_oldmem_pfn_range_zfcpdump(vma, from, pfn, size, + prot); +} + +/* + * Copy memory from old kernel + */ +int copy_from_oldmem(void *dest, void *src, size_t count) +{ + unsigned long copied = 0; + int rc; + + if (OLDMEM_BASE) { + if ((unsigned long) src < OLDMEM_SIZE) { + copied = min(count, OLDMEM_SIZE - (unsigned long) src); + rc = copy_from_realmem(dest, src + OLDMEM_BASE, copied); + if (rc) + return rc; + } + } else { + unsigned long hsa_end = sclp_get_hsa_size(); + if ((unsigned long) src < hsa_end) { + copied = min(count, hsa_end - (unsigned long) src); + rc = memcpy_hsa(dest, (unsigned long) src, copied, 0); + if (rc) + return rc; + } + } + return copy_from_realmem(dest + copied, src + copied, count - copied); +} + +/* + * Alloc memory and panic in case of ENOMEM + */ +static void *kzalloc_panic(int len) +{ + void *rc; + + rc = kzalloc(len, GFP_KERNEL); + if (!rc) + panic("s390 kdump kzalloc (%d) failed", len); + return rc; +} + +/* + * Initialize ELF note + */ +static void *nt_init(void *buf, Elf64_Word type, void *desc, int d_len, + const char *name) +{ + Elf64_Nhdr *note; + u64 len; + + note = (Elf64_Nhdr *)buf; + note->n_namesz = strlen(name) + 1; + note->n_descsz = d_len; + note->n_type = type; + len = sizeof(Elf64_Nhdr); + + memcpy(buf + len, name, note->n_namesz); + len = roundup(len + note->n_namesz, 4); + + memcpy(buf + len, desc, note->n_descsz); + len = roundup(len + note->n_descsz, 4); + + return PTR_ADD(buf, len); +} + +/* + * Initialize prstatus note + */ +static void *nt_prstatus(void *ptr, struct save_area *sa) +{ + struct elf_prstatus nt_prstatus; + static int cpu_nr = 1; + + memset(&nt_prstatus, 0, sizeof(nt_prstatus)); + memcpy(&nt_prstatus.pr_reg.gprs, sa->gp_regs, sizeof(sa->gp_regs)); + memcpy(&nt_prstatus.pr_reg.psw, sa->psw, sizeof(sa->psw)); + memcpy(&nt_prstatus.pr_reg.acrs, sa->acc_regs, sizeof(sa->acc_regs)); + nt_prstatus.pr_pid = cpu_nr; + cpu_nr++; + + return nt_init(ptr, NT_PRSTATUS, &nt_prstatus, sizeof(nt_prstatus), + "CORE"); +} + +/* + * Initialize fpregset (floating point) note + */ +static void *nt_fpregset(void *ptr, struct save_area *sa) +{ + elf_fpregset_t nt_fpregset; + + memset(&nt_fpregset, 0, sizeof(nt_fpregset)); + memcpy(&nt_fpregset.fpc, &sa->fp_ctrl_reg, sizeof(sa->fp_ctrl_reg)); + memcpy(&nt_fpregset.fprs, &sa->fp_regs, sizeof(sa->fp_regs)); + + return nt_init(ptr, NT_PRFPREG, &nt_fpregset, sizeof(nt_fpregset), + "CORE"); +} + +/* + * Initialize timer note + */ +static void *nt_s390_timer(void *ptr, struct save_area *sa) +{ + return nt_init(ptr, NT_S390_TIMER, &sa->timer, sizeof(sa->timer), + KEXEC_CORE_NOTE_NAME); +} + +/* + * Initialize TOD clock comparator note + */ +static void *nt_s390_tod_cmp(void *ptr, struct save_area *sa) +{ + return nt_init(ptr, NT_S390_TODCMP, &sa->clk_cmp, + sizeof(sa->clk_cmp), KEXEC_CORE_NOTE_NAME); +} + +/* + * Initialize TOD programmable register note + */ +static void *nt_s390_tod_preg(void *ptr, struct save_area *sa) +{ + return nt_init(ptr, NT_S390_TODPREG, &sa->tod_reg, + sizeof(sa->tod_reg), KEXEC_CORE_NOTE_NAME); +} + +/* + * Initialize control register note + */ +static void *nt_s390_ctrs(void *ptr, struct save_area *sa) +{ + return nt_init(ptr, NT_S390_CTRS, &sa->ctrl_regs, + sizeof(sa->ctrl_regs), KEXEC_CORE_NOTE_NAME); +} + +/* + * Initialize prefix register note + */ +static void *nt_s390_prefix(void *ptr, struct save_area *sa) +{ + return nt_init(ptr, NT_S390_PREFIX, &sa->pref_reg, + sizeof(sa->pref_reg), KEXEC_CORE_NOTE_NAME); +} + +/* + * Initialize vxrs high note (full 128 bit VX registers 16-31) + */ +static void *nt_s390_vx_high(void *ptr, __vector128 *vx_regs) +{ + return nt_init(ptr, NT_S390_VXRS_HIGH, &vx_regs[16], + 16 * sizeof(__vector128), KEXEC_CORE_NOTE_NAME); +} + +/* + * Initialize vxrs low note (lower halves of VX registers 0-15) + */ +static void *nt_s390_vx_low(void *ptr, __vector128 *vx_regs) +{ + Elf64_Nhdr *note; + u64 len; + int i; + + note = (Elf64_Nhdr *)ptr; + note->n_namesz = strlen(KEXEC_CORE_NOTE_NAME) + 1; + note->n_descsz = 16 * 8; + note->n_type = NT_S390_VXRS_LOW; + len = sizeof(Elf64_Nhdr); + + memcpy(ptr + len, KEXEC_CORE_NOTE_NAME, note->n_namesz); + len = roundup(len + note->n_namesz, 4); + + ptr += len; + /* Copy lower halves of SIMD registers 0-15 */ + for (i = 0; i < 16; i++) { + memcpy(ptr, &vx_regs[i].u[2], 8); + ptr += 8; + } + return ptr; +} + +/* + * Fill ELF notes for one CPU with save area registers + */ +void *fill_cpu_elf_notes(void *ptr, struct save_area *sa, __vector128 *vx_regs) +{ + ptr = nt_prstatus(ptr, sa); + ptr = nt_fpregset(ptr, sa); + ptr = nt_s390_timer(ptr, sa); + ptr = nt_s390_tod_cmp(ptr, sa); + ptr = nt_s390_tod_preg(ptr, sa); + ptr = nt_s390_ctrs(ptr, sa); + ptr = nt_s390_prefix(ptr, sa); + if (MACHINE_HAS_VX && vx_regs) { + ptr = nt_s390_vx_low(ptr, vx_regs); + ptr = nt_s390_vx_high(ptr, vx_regs); + } + return ptr; +} + +/* + * Initialize prpsinfo note (new kernel) + */ +static void *nt_prpsinfo(void *ptr) +{ + struct elf_prpsinfo prpsinfo; + + memset(&prpsinfo, 0, sizeof(prpsinfo)); + prpsinfo.pr_sname = 'R'; + strcpy(prpsinfo.pr_fname, "vmlinux"); + return nt_init(ptr, NT_PRPSINFO, &prpsinfo, sizeof(prpsinfo), + KEXEC_CORE_NOTE_NAME); +} + +/* + * Get vmcoreinfo using lowcore->vmcore_info (new kernel) + */ +static void *get_vmcoreinfo_old(unsigned long *size) +{ + char nt_name[11], *vmcoreinfo; + Elf64_Nhdr note; + void *addr; + + if (copy_from_oldmem(&addr, &S390_lowcore.vmcore_info, sizeof(addr))) + return NULL; + memset(nt_name, 0, sizeof(nt_name)); + if (copy_from_oldmem(¬e, addr, sizeof(note))) + return NULL; + if (copy_from_oldmem(nt_name, addr + sizeof(note), sizeof(nt_name) - 1)) + return NULL; + if (strcmp(nt_name, "VMCOREINFO") != 0) + return NULL; + vmcoreinfo = kzalloc_panic(note.n_descsz); + if (copy_from_oldmem(vmcoreinfo, addr + 24, note.n_descsz)) + return NULL; + *size = note.n_descsz; + return vmcoreinfo; +} + +/* + * Initialize vmcoreinfo note (new kernel) + */ +static void *nt_vmcoreinfo(void *ptr) +{ + unsigned long size; + void *vmcoreinfo; + + vmcoreinfo = os_info_old_entry(OS_INFO_VMCOREINFO, &size); + if (!vmcoreinfo) + vmcoreinfo = get_vmcoreinfo_old(&size); + if (!vmcoreinfo) + return ptr; + return nt_init(ptr, 0, vmcoreinfo, size, "VMCOREINFO"); +} + +/* + * Initialize ELF header (new kernel) + */ +static void *ehdr_init(Elf64_Ehdr *ehdr, int mem_chunk_cnt) +{ + memset(ehdr, 0, sizeof(*ehdr)); + memcpy(ehdr->e_ident, ELFMAG, SELFMAG); + ehdr->e_ident[EI_CLASS] = ELFCLASS64; + ehdr->e_ident[EI_DATA] = ELFDATA2MSB; + ehdr->e_ident[EI_VERSION] = EV_CURRENT; + memset(ehdr->e_ident + EI_PAD, 0, EI_NIDENT - EI_PAD); + ehdr->e_type = ET_CORE; + ehdr->e_machine = EM_S390; + ehdr->e_version = EV_CURRENT; + ehdr->e_phoff = sizeof(Elf64_Ehdr); + ehdr->e_ehsize = sizeof(Elf64_Ehdr); + ehdr->e_phentsize = sizeof(Elf64_Phdr); + ehdr->e_phnum = mem_chunk_cnt + 1; + return ehdr + 1; +} + +/* + * Return CPU count for ELF header (new kernel) + */ +static int get_cpu_cnt(void) +{ + int i, cpus = 0; + + for (i = 0; i < dump_save_areas.count; i++) { + if (dump_save_areas.areas[i]->sa.pref_reg == 0) + continue; + cpus++; + } + return cpus; +} + +/* + * Return memory chunk count for ELF header (new kernel) + */ +static int get_mem_chunk_cnt(void) +{ + int cnt = 0; + u64 idx; + + for_each_dump_mem_range(idx, NUMA_NO_NODE, NULL, NULL, NULL) + cnt++; + return cnt; +} + +/* + * Initialize ELF loads (new kernel) + */ +static void loads_init(Elf64_Phdr *phdr, u64 loads_offset) +{ + phys_addr_t start, end; + u64 idx; + + for_each_dump_mem_range(idx, NUMA_NO_NODE, &start, &end, NULL) { + phdr->p_filesz = end - start; + phdr->p_type = PT_LOAD; + phdr->p_offset = start; + phdr->p_vaddr = start; + phdr->p_paddr = start; + phdr->p_memsz = end - start; + phdr->p_flags = PF_R | PF_W | PF_X; + phdr->p_align = PAGE_SIZE; + phdr++; + } +} + +/* + * Initialize notes (new kernel) + */ +static void *notes_init(Elf64_Phdr *phdr, void *ptr, u64 notes_offset) +{ + struct save_area_ext *sa_ext; + void *ptr_start = ptr; + int i; + + ptr = nt_prpsinfo(ptr); + + for (i = 0; i < dump_save_areas.count; i++) { + sa_ext = dump_save_areas.areas[i]; + if (sa_ext->sa.pref_reg == 0) + continue; + ptr = fill_cpu_elf_notes(ptr, &sa_ext->sa, sa_ext->vx_regs); + } + ptr = nt_vmcoreinfo(ptr); + memset(phdr, 0, sizeof(*phdr)); + phdr->p_type = PT_NOTE; + phdr->p_offset = notes_offset; + phdr->p_filesz = (unsigned long) PTR_SUB(ptr, ptr_start); + phdr->p_memsz = phdr->p_filesz; + return ptr; +} + +/* + * Create ELF core header (new kernel) + */ +int elfcorehdr_alloc(unsigned long long *addr, unsigned long long *size) +{ + Elf64_Phdr *phdr_notes, *phdr_loads; + int mem_chunk_cnt; + void *ptr, *hdr; + u32 alloc_size; + u64 hdr_off; + + /* If we are not in kdump or zfcpdump mode return */ + if (!OLDMEM_BASE && ipl_info.type != IPL_TYPE_FCP_DUMP) + return 0; + /* If elfcorehdr= has been passed via cmdline, we use that one */ + if (elfcorehdr_addr != ELFCORE_ADDR_MAX) + return 0; + /* If we cannot get HSA size for zfcpdump return error */ + if (ipl_info.type == IPL_TYPE_FCP_DUMP && !sclp_get_hsa_size()) + return -ENODEV; + + /* For kdump, exclude previous crashkernel memory */ + if (OLDMEM_BASE) { + oldmem_region.base = OLDMEM_BASE; + oldmem_region.size = OLDMEM_SIZE; + oldmem_type.total_size = OLDMEM_SIZE; + } + + mem_chunk_cnt = get_mem_chunk_cnt(); + + alloc_size = 0x1000 + get_cpu_cnt() * 0x4a0 + + mem_chunk_cnt * sizeof(Elf64_Phdr); + hdr = kzalloc_panic(alloc_size); + /* Init elf header */ + ptr = ehdr_init(hdr, mem_chunk_cnt); + /* Init program headers */ + phdr_notes = ptr; + ptr = PTR_ADD(ptr, sizeof(Elf64_Phdr)); + phdr_loads = ptr; + ptr = PTR_ADD(ptr, sizeof(Elf64_Phdr) * mem_chunk_cnt); + /* Init notes */ + hdr_off = PTR_DIFF(ptr, hdr); + ptr = notes_init(phdr_notes, ptr, ((unsigned long) hdr) + hdr_off); + /* Init loads */ + hdr_off = PTR_DIFF(ptr, hdr); + loads_init(phdr_loads, hdr_off); + *addr = (unsigned long long) hdr; + elfcorehdr_newmem = hdr; + *size = (unsigned long long) hdr_off; + BUG_ON(elfcorehdr_size > alloc_size); + return 0; +} + +/* + * Free ELF core header (new kernel) + */ +void elfcorehdr_free(unsigned long long addr) +{ + if (!elfcorehdr_newmem) + return; + kfree((void *)(unsigned long)addr); +} + +/* + * Read from ELF header + */ +ssize_t elfcorehdr_read(char *buf, size_t count, u64 *ppos) +{ + void *src = (void *)(unsigned long)*ppos; + + src = elfcorehdr_newmem ? src : src - OLDMEM_BASE; + memcpy(buf, src, count); + *ppos += count; + return count; +} + +/* + * Read from ELF notes data + */ +ssize_t elfcorehdr_read_notes(char *buf, size_t count, u64 *ppos) +{ + void *src = (void *)(unsigned long)*ppos; + int rc; + + if (elfcorehdr_newmem) { + memcpy(buf, src, count); + } else { + rc = copy_from_oldmem(buf, src, count); + if (rc) + return rc; + } + *ppos += count; + return count; +} diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c new file mode 100644 index 000000000..c1f21aca7 --- /dev/null +++ b/arch/s390/kernel/debug.c @@ -0,0 +1,1540 @@ +/* + * S/390 debug facility + * + * Copyright IBM Corp. 1999, 2012 + * + * Author(s): Michael Holzheu (holzheu@de.ibm.com), + * Holger Smolinski (Holger.Smolinski@de.ibm.com) + * + * Bugreports to: <Linux390@de.ibm.com> + */ + +#define KMSG_COMPONENT "s390dbf" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/stddef.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/ctype.h> +#include <linux/string.h> +#include <linux/sysctl.h> +#include <asm/uaccess.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/fs.h> +#include <linux/debugfs.h> + +#include <asm/debug.h> + +#define DEBUG_PROLOG_ENTRY -1 + +#define ALL_AREAS 0 /* copy all debug areas */ +#define NO_AREAS 1 /* copy no debug areas */ + +/* typedefs */ + +typedef struct file_private_info { + loff_t offset; /* offset of last read in file */ + int act_area; /* number of last formated area */ + int act_page; /* act page in given area */ + int act_entry; /* last formated entry (offset */ + /* relative to beginning of last */ + /* formated page) */ + size_t act_entry_offset; /* up to this offset we copied */ + /* in last read the last formated */ + /* entry to userland */ + char temp_buf[2048]; /* buffer for output */ + debug_info_t *debug_info_org; /* original debug information */ + debug_info_t *debug_info_snap; /* snapshot of debug information */ + struct debug_view *view; /* used view of debug info */ +} file_private_info_t; + +typedef struct +{ + char *string; + /* + * This assumes that all args are converted into longs + * on L/390 this is the case for all types of parameter + * except of floats, and long long (32 bit) + * + */ + long args[0]; +} debug_sprintf_entry_t; + + +/* internal function prototyes */ + +static int debug_init(void); +static ssize_t debug_output(struct file *file, char __user *user_buf, + size_t user_len, loff_t * offset); +static ssize_t debug_input(struct file *file, const char __user *user_buf, + size_t user_len, loff_t * offset); +static int debug_open(struct inode *inode, struct file *file); +static int debug_close(struct inode *inode, struct file *file); +static debug_info_t *debug_info_create(const char *name, int pages_per_area, + int nr_areas, int buf_size, umode_t mode); +static void debug_info_get(debug_info_t *); +static void debug_info_put(debug_info_t *); +static int debug_prolog_level_fn(debug_info_t * id, + struct debug_view *view, char *out_buf); +static int debug_input_level_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_buf_size, loff_t * offset); +static int debug_prolog_pages_fn(debug_info_t * id, + struct debug_view *view, char *out_buf); +static int debug_input_pages_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_buf_size, loff_t * offset); +static int debug_input_flush_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_buf_size, loff_t * offset); +static int debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view, + char *out_buf, const char *in_buf); +static int debug_raw_format_fn(debug_info_t * id, + struct debug_view *view, char *out_buf, + const char *in_buf); +static int debug_raw_header_fn(debug_info_t * id, struct debug_view *view, + int area, debug_entry_t * entry, char *out_buf); + +static int debug_sprintf_format_fn(debug_info_t * id, struct debug_view *view, + char *out_buf, debug_sprintf_entry_t *curr_event); + +/* globals */ + +struct debug_view debug_raw_view = { + "raw", + NULL, + &debug_raw_header_fn, + &debug_raw_format_fn, + NULL, + NULL +}; +EXPORT_SYMBOL(debug_raw_view); + +struct debug_view debug_hex_ascii_view = { + "hex_ascii", + NULL, + &debug_dflt_header_fn, + &debug_hex_ascii_format_fn, + NULL, + NULL +}; +EXPORT_SYMBOL(debug_hex_ascii_view); + +static struct debug_view debug_level_view = { + "level", + &debug_prolog_level_fn, + NULL, + NULL, + &debug_input_level_fn, + NULL +}; + +static struct debug_view debug_pages_view = { + "pages", + &debug_prolog_pages_fn, + NULL, + NULL, + &debug_input_pages_fn, + NULL +}; + +static struct debug_view debug_flush_view = { + "flush", + NULL, + NULL, + NULL, + &debug_input_flush_fn, + NULL +}; + +struct debug_view debug_sprintf_view = { + "sprintf", + NULL, + &debug_dflt_header_fn, + (debug_format_proc_t*)&debug_sprintf_format_fn, + NULL, + NULL +}; +EXPORT_SYMBOL(debug_sprintf_view); + +/* used by dump analysis tools to determine version of debug feature */ +static unsigned int __used debug_feature_version = __DEBUG_FEATURE_VERSION; + +/* static globals */ + +static debug_info_t *debug_area_first = NULL; +static debug_info_t *debug_area_last = NULL; +static DEFINE_MUTEX(debug_mutex); + +static int initialized; +static int debug_critical; + +static const struct file_operations debug_file_ops = { + .owner = THIS_MODULE, + .read = debug_output, + .write = debug_input, + .open = debug_open, + .release = debug_close, + .llseek = no_llseek, +}; + +static struct dentry *debug_debugfs_root_entry; + +/* functions */ + +/* + * debug_areas_alloc + * - Debug areas are implemented as a threedimensonal array: + * areas[areanumber][pagenumber][pageoffset] + */ + +static debug_entry_t*** +debug_areas_alloc(int pages_per_area, int nr_areas) +{ + debug_entry_t*** areas; + int i,j; + + areas = kmalloc(nr_areas * + sizeof(debug_entry_t**), + GFP_KERNEL); + if (!areas) + goto fail_malloc_areas; + for (i = 0; i < nr_areas; i++) { + areas[i] = kmalloc(pages_per_area * + sizeof(debug_entry_t*),GFP_KERNEL); + if (!areas[i]) { + goto fail_malloc_areas2; + } + for(j = 0; j < pages_per_area; j++) { + areas[i][j] = kzalloc(PAGE_SIZE, GFP_KERNEL); + if(!areas[i][j]) { + for(j--; j >=0 ; j--) { + kfree(areas[i][j]); + } + kfree(areas[i]); + goto fail_malloc_areas2; + } + } + } + return areas; + +fail_malloc_areas2: + for(i--; i >= 0; i--){ + for(j=0; j < pages_per_area;j++){ + kfree(areas[i][j]); + } + kfree(areas[i]); + } + kfree(areas); +fail_malloc_areas: + return NULL; + +} + + +/* + * debug_info_alloc + * - alloc new debug-info + */ + +static debug_info_t* +debug_info_alloc(const char *name, int pages_per_area, int nr_areas, + int buf_size, int level, int mode) +{ + debug_info_t* rc; + + /* alloc everything */ + + rc = kmalloc(sizeof(debug_info_t), GFP_KERNEL); + if(!rc) + goto fail_malloc_rc; + rc->active_entries = kcalloc(nr_areas, sizeof(int), GFP_KERNEL); + if(!rc->active_entries) + goto fail_malloc_active_entries; + rc->active_pages = kcalloc(nr_areas, sizeof(int), GFP_KERNEL); + if(!rc->active_pages) + goto fail_malloc_active_pages; + if((mode == ALL_AREAS) && (pages_per_area != 0)){ + rc->areas = debug_areas_alloc(pages_per_area, nr_areas); + if(!rc->areas) + goto fail_malloc_areas; + } else { + rc->areas = NULL; + } + + /* initialize members */ + + spin_lock_init(&rc->lock); + rc->pages_per_area = pages_per_area; + rc->nr_areas = nr_areas; + rc->active_area = 0; + rc->level = level; + rc->buf_size = buf_size; + rc->entry_size = sizeof(debug_entry_t) + buf_size; + strlcpy(rc->name, name, sizeof(rc->name)); + memset(rc->views, 0, DEBUG_MAX_VIEWS * sizeof(struct debug_view *)); + memset(rc->debugfs_entries, 0 ,DEBUG_MAX_VIEWS * + sizeof(struct dentry*)); + atomic_set(&(rc->ref_count), 0); + + return rc; + +fail_malloc_areas: + kfree(rc->active_pages); +fail_malloc_active_pages: + kfree(rc->active_entries); +fail_malloc_active_entries: + kfree(rc); +fail_malloc_rc: + return NULL; +} + +/* + * debug_areas_free + * - free all debug areas + */ + +static void +debug_areas_free(debug_info_t* db_info) +{ + int i,j; + + if(!db_info->areas) + return; + for (i = 0; i < db_info->nr_areas; i++) { + for(j = 0; j < db_info->pages_per_area; j++) { + kfree(db_info->areas[i][j]); + } + kfree(db_info->areas[i]); + } + kfree(db_info->areas); + db_info->areas = NULL; +} + +/* + * debug_info_free + * - free memory debug-info + */ + +static void +debug_info_free(debug_info_t* db_info){ + debug_areas_free(db_info); + kfree(db_info->active_entries); + kfree(db_info->active_pages); + kfree(db_info); +} + +/* + * debug_info_create + * - create new debug-info + */ + +static debug_info_t* +debug_info_create(const char *name, int pages_per_area, int nr_areas, + int buf_size, umode_t mode) +{ + debug_info_t* rc; + + rc = debug_info_alloc(name, pages_per_area, nr_areas, buf_size, + DEBUG_DEFAULT_LEVEL, ALL_AREAS); + if(!rc) + goto out; + + rc->mode = mode & ~S_IFMT; + + /* create root directory */ + rc->debugfs_root_entry = debugfs_create_dir(rc->name, + debug_debugfs_root_entry); + + /* append new element to linked list */ + if (!debug_area_first) { + /* first element in list */ + debug_area_first = rc; + rc->prev = NULL; + } else { + /* append element to end of list */ + debug_area_last->next = rc; + rc->prev = debug_area_last; + } + debug_area_last = rc; + rc->next = NULL; + + debug_info_get(rc); +out: + return rc; +} + +/* + * debug_info_copy + * - copy debug-info + */ + +static debug_info_t* +debug_info_copy(debug_info_t* in, int mode) +{ + int i,j; + debug_info_t* rc; + unsigned long flags; + + /* get a consistent copy of the debug areas */ + do { + rc = debug_info_alloc(in->name, in->pages_per_area, + in->nr_areas, in->buf_size, in->level, mode); + spin_lock_irqsave(&in->lock, flags); + if(!rc) + goto out; + /* has something changed in the meantime ? */ + if((rc->pages_per_area == in->pages_per_area) && + (rc->nr_areas == in->nr_areas)) { + break; + } + spin_unlock_irqrestore(&in->lock, flags); + debug_info_free(rc); + } while (1); + + if (mode == NO_AREAS) + goto out; + + for(i = 0; i < in->nr_areas; i++){ + for(j = 0; j < in->pages_per_area; j++) { + memcpy(rc->areas[i][j], in->areas[i][j],PAGE_SIZE); + } + } +out: + spin_unlock_irqrestore(&in->lock, flags); + return rc; +} + +/* + * debug_info_get + * - increments reference count for debug-info + */ + +static void +debug_info_get(debug_info_t * db_info) +{ + if (db_info) + atomic_inc(&db_info->ref_count); +} + +/* + * debug_info_put: + * - decreases reference count for debug-info and frees it if necessary + */ + +static void +debug_info_put(debug_info_t *db_info) +{ + int i; + + if (!db_info) + return; + if (atomic_dec_and_test(&db_info->ref_count)) { + for (i = 0; i < DEBUG_MAX_VIEWS; i++) { + if (!db_info->views[i]) + continue; + debugfs_remove(db_info->debugfs_entries[i]); + } + debugfs_remove(db_info->debugfs_root_entry); + if(db_info == debug_area_first) + debug_area_first = db_info->next; + if(db_info == debug_area_last) + debug_area_last = db_info->prev; + if(db_info->prev) db_info->prev->next = db_info->next; + if(db_info->next) db_info->next->prev = db_info->prev; + debug_info_free(db_info); + } +} + +/* + * debug_format_entry: + * - format one debug entry and return size of formated data + */ + +static int +debug_format_entry(file_private_info_t *p_info) +{ + debug_info_t *id_snap = p_info->debug_info_snap; + struct debug_view *view = p_info->view; + debug_entry_t *act_entry; + size_t len = 0; + if(p_info->act_entry == DEBUG_PROLOG_ENTRY){ + /* print prolog */ + if (view->prolog_proc) + len += view->prolog_proc(id_snap,view,p_info->temp_buf); + goto out; + } + if (!id_snap->areas) /* this is true, if we have a prolog only view */ + goto out; /* or if 'pages_per_area' is 0 */ + act_entry = (debug_entry_t *) ((char*)id_snap->areas[p_info->act_area] + [p_info->act_page] + p_info->act_entry); + + if (act_entry->id.stck == 0LL) + goto out; /* empty entry */ + if (view->header_proc) + len += view->header_proc(id_snap, view, p_info->act_area, + act_entry, p_info->temp_buf + len); + if (view->format_proc) + len += view->format_proc(id_snap, view, p_info->temp_buf + len, + DEBUG_DATA(act_entry)); +out: + return len; +} + +/* + * debug_next_entry: + * - goto next entry in p_info + */ + +static inline int +debug_next_entry(file_private_info_t *p_info) +{ + debug_info_t *id; + + id = p_info->debug_info_snap; + if(p_info->act_entry == DEBUG_PROLOG_ENTRY){ + p_info->act_entry = 0; + p_info->act_page = 0; + goto out; + } + if(!id->areas) + return 1; + p_info->act_entry += id->entry_size; + /* switch to next page, if we reached the end of the page */ + if (p_info->act_entry > (PAGE_SIZE - id->entry_size)){ + /* next page */ + p_info->act_entry = 0; + p_info->act_page += 1; + if((p_info->act_page % id->pages_per_area) == 0) { + /* next area */ + p_info->act_area++; + p_info->act_page=0; + } + if(p_info->act_area >= id->nr_areas) + return 1; + } +out: + return 0; +} + +/* + * debug_output: + * - called for user read() + * - copies formated debug entries to the user buffer + */ + +static ssize_t +debug_output(struct file *file, /* file descriptor */ + char __user *user_buf, /* user buffer */ + size_t len, /* length of buffer */ + loff_t *offset) /* offset in the file */ +{ + size_t count = 0; + size_t entry_offset; + file_private_info_t *p_info; + + p_info = ((file_private_info_t *) file->private_data); + if (*offset != p_info->offset) + return -EPIPE; + if(p_info->act_area >= p_info->debug_info_snap->nr_areas) + return 0; + entry_offset = p_info->act_entry_offset; + while(count < len){ + int formatted_line_size; + int formatted_line_residue; + int user_buf_residue; + size_t copy_size; + + formatted_line_size = debug_format_entry(p_info); + formatted_line_residue = formatted_line_size - entry_offset; + user_buf_residue = len-count; + copy_size = min(user_buf_residue, formatted_line_residue); + if(copy_size){ + if (copy_to_user(user_buf + count, p_info->temp_buf + + entry_offset, copy_size)) + return -EFAULT; + count += copy_size; + entry_offset += copy_size; + } + if(copy_size == formatted_line_residue){ + entry_offset = 0; + if(debug_next_entry(p_info)) + goto out; + } + } +out: + p_info->offset = *offset + count; + p_info->act_entry_offset = entry_offset; + *offset = p_info->offset; + return count; +} + +/* + * debug_input: + * - called for user write() + * - calls input function of view + */ + +static ssize_t +debug_input(struct file *file, const char __user *user_buf, size_t length, + loff_t *offset) +{ + int rc = 0; + file_private_info_t *p_info; + + mutex_lock(&debug_mutex); + p_info = ((file_private_info_t *) file->private_data); + if (p_info->view->input_proc) + rc = p_info->view->input_proc(p_info->debug_info_org, + p_info->view, file, user_buf, + length, offset); + else + rc = -EPERM; + mutex_unlock(&debug_mutex); + return rc; /* number of input characters */ +} + +/* + * debug_open: + * - called for user open() + * - copies formated output to private_data area of the file + * handle + */ + +static int +debug_open(struct inode *inode, struct file *file) +{ + int i, rc = 0; + file_private_info_t *p_info; + debug_info_t *debug_info, *debug_info_snapshot; + + mutex_lock(&debug_mutex); + debug_info = file_inode(file)->i_private; + /* find debug view */ + for (i = 0; i < DEBUG_MAX_VIEWS; i++) { + if (!debug_info->views[i]) + continue; + else if (debug_info->debugfs_entries[i] == + file->f_path.dentry) { + goto found; /* found view ! */ + } + } + /* no entry found */ + rc = -EINVAL; + goto out; + +found: + + /* Make snapshot of current debug areas to get it consistent. */ + /* To copy all the areas is only needed, if we have a view which */ + /* formats the debug areas. */ + + if(!debug_info->views[i]->format_proc && + !debug_info->views[i]->header_proc){ + debug_info_snapshot = debug_info_copy(debug_info, NO_AREAS); + } else { + debug_info_snapshot = debug_info_copy(debug_info, ALL_AREAS); + } + + if(!debug_info_snapshot){ + rc = -ENOMEM; + goto out; + } + p_info = kmalloc(sizeof(file_private_info_t), + GFP_KERNEL); + if(!p_info){ + debug_info_free(debug_info_snapshot); + rc = -ENOMEM; + goto out; + } + p_info->offset = 0; + p_info->debug_info_snap = debug_info_snapshot; + p_info->debug_info_org = debug_info; + p_info->view = debug_info->views[i]; + p_info->act_area = 0; + p_info->act_page = 0; + p_info->act_entry = DEBUG_PROLOG_ENTRY; + p_info->act_entry_offset = 0; + file->private_data = p_info; + debug_info_get(debug_info); + nonseekable_open(inode, file); +out: + mutex_unlock(&debug_mutex); + return rc; +} + +/* + * debug_close: + * - called for user close() + * - deletes private_data area of the file handle + */ + +static int +debug_close(struct inode *inode, struct file *file) +{ + file_private_info_t *p_info; + p_info = (file_private_info_t *) file->private_data; + if(p_info->debug_info_snap) + debug_info_free(p_info->debug_info_snap); + debug_info_put(p_info->debug_info_org); + kfree(file->private_data); + return 0; /* success */ +} + +/* + * debug_register_mode: + * - Creates and initializes debug area for the caller + * The mode parameter allows to specify access rights for the s390dbf files + * - Returns handle for debug area + */ + +debug_info_t *debug_register_mode(const char *name, int pages_per_area, + int nr_areas, int buf_size, umode_t mode, + uid_t uid, gid_t gid) +{ + debug_info_t *rc = NULL; + + /* Since debugfs currently does not support uid/gid other than root, */ + /* we do not allow gid/uid != 0 until we get support for that. */ + if ((uid != 0) || (gid != 0)) + pr_warning("Root becomes the owner of all s390dbf files " + "in sysfs\n"); + BUG_ON(!initialized); + mutex_lock(&debug_mutex); + + /* create new debug_info */ + + rc = debug_info_create(name, pages_per_area, nr_areas, buf_size, mode); + if(!rc) + goto out; + debug_register_view(rc, &debug_level_view); + debug_register_view(rc, &debug_flush_view); + debug_register_view(rc, &debug_pages_view); +out: + if (!rc){ + pr_err("Registering debug feature %s failed\n", name); + } + mutex_unlock(&debug_mutex); + return rc; +} +EXPORT_SYMBOL(debug_register_mode); + +/* + * debug_register: + * - creates and initializes debug area for the caller + * - returns handle for debug area + */ + +debug_info_t *debug_register(const char *name, int pages_per_area, + int nr_areas, int buf_size) +{ + return debug_register_mode(name, pages_per_area, nr_areas, buf_size, + S_IRUSR | S_IWUSR, 0, 0); +} +EXPORT_SYMBOL(debug_register); + +/* + * debug_unregister: + * - give back debug area + */ + +void +debug_unregister(debug_info_t * id) +{ + if (!id) + goto out; + mutex_lock(&debug_mutex); + debug_info_put(id); + mutex_unlock(&debug_mutex); + +out: + return; +} +EXPORT_SYMBOL(debug_unregister); + +/* + * debug_set_size: + * - set area size (number of pages) and number of areas + */ +static int +debug_set_size(debug_info_t* id, int nr_areas, int pages_per_area) +{ + unsigned long flags; + debug_entry_t *** new_areas; + int rc=0; + + if(!id || (nr_areas <= 0) || (pages_per_area < 0)) + return -EINVAL; + if(pages_per_area > 0){ + new_areas = debug_areas_alloc(pages_per_area, nr_areas); + if(!new_areas) { + pr_info("Allocating memory for %i pages failed\n", + pages_per_area); + rc = -ENOMEM; + goto out; + } + } else { + new_areas = NULL; + } + spin_lock_irqsave(&id->lock,flags); + debug_areas_free(id); + id->areas = new_areas; + id->nr_areas = nr_areas; + id->pages_per_area = pages_per_area; + id->active_area = 0; + memset(id->active_entries,0,sizeof(int)*id->nr_areas); + memset(id->active_pages, 0, sizeof(int)*id->nr_areas); + spin_unlock_irqrestore(&id->lock,flags); + pr_info("%s: set new size (%i pages)\n" ,id->name, pages_per_area); +out: + return rc; +} + +/* + * debug_set_level: + * - set actual debug level + */ + +void +debug_set_level(debug_info_t* id, int new_level) +{ + unsigned long flags; + if(!id) + return; + spin_lock_irqsave(&id->lock,flags); + if(new_level == DEBUG_OFF_LEVEL){ + id->level = DEBUG_OFF_LEVEL; + pr_info("%s: switched off\n",id->name); + } else if ((new_level > DEBUG_MAX_LEVEL) || (new_level < 0)) { + pr_info("%s: level %i is out of range (%i - %i)\n", + id->name, new_level, 0, DEBUG_MAX_LEVEL); + } else { + id->level = new_level; + } + spin_unlock_irqrestore(&id->lock,flags); +} +EXPORT_SYMBOL(debug_set_level); + +/* + * proceed_active_entry: + * - set active entry to next in the ring buffer + */ + +static inline void +proceed_active_entry(debug_info_t * id) +{ + if ((id->active_entries[id->active_area] += id->entry_size) + > (PAGE_SIZE - id->entry_size)){ + id->active_entries[id->active_area] = 0; + id->active_pages[id->active_area] = + (id->active_pages[id->active_area] + 1) % + id->pages_per_area; + } +} + +/* + * proceed_active_area: + * - set active area to next in the ring buffer + */ + +static inline void +proceed_active_area(debug_info_t * id) +{ + id->active_area++; + id->active_area = id->active_area % id->nr_areas; +} + +/* + * get_active_entry: + */ + +static inline debug_entry_t* +get_active_entry(debug_info_t * id) +{ + return (debug_entry_t *) (((char *) id->areas[id->active_area] + [id->active_pages[id->active_area]]) + + id->active_entries[id->active_area]); +} + +/* + * debug_finish_entry: + * - set timestamp, caller address, cpu number etc. + */ + +static inline void +debug_finish_entry(debug_info_t * id, debug_entry_t* active, int level, + int exception) +{ + active->id.stck = get_tod_clock_fast(); + active->id.fields.cpuid = smp_processor_id(); + active->caller = __builtin_return_address(0); + active->id.fields.exception = exception; + active->id.fields.level = level; + proceed_active_entry(id); + if(exception) + proceed_active_area(id); +} + +static int debug_stoppable=1; +static int debug_active=1; + +#define CTL_S390DBF_STOPPABLE 5678 +#define CTL_S390DBF_ACTIVE 5679 + +/* + * proc handler for the running debug_active sysctl + * always allow read, allow write only if debug_stoppable is set or + * if debug_active is already off + */ +static int +s390dbf_procactive(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + if (!write || debug_stoppable || !debug_active) + return proc_dointvec(table, write, buffer, lenp, ppos); + else + return 0; +} + + +static struct ctl_table s390dbf_table[] = { + { + .procname = "debug_stoppable", + .data = &debug_stoppable, + .maxlen = sizeof(int), + .mode = S_IRUGO | S_IWUSR, + .proc_handler = proc_dointvec, + }, + { + .procname = "debug_active", + .data = &debug_active, + .maxlen = sizeof(int), + .mode = S_IRUGO | S_IWUSR, + .proc_handler = s390dbf_procactive, + }, + { } +}; + +static struct ctl_table s390dbf_dir_table[] = { + { + .procname = "s390dbf", + .maxlen = 0, + .mode = S_IRUGO | S_IXUGO, + .child = s390dbf_table, + }, + { } +}; + +static struct ctl_table_header *s390dbf_sysctl_header; + +void +debug_stop_all(void) +{ + if (debug_stoppable) + debug_active = 0; +} +EXPORT_SYMBOL(debug_stop_all); + +void debug_set_critical(void) +{ + debug_critical = 1; +} + +/* + * debug_event_common: + * - write debug entry with given size + */ + +debug_entry_t* +debug_event_common(debug_info_t * id, int level, const void *buf, int len) +{ + unsigned long flags; + debug_entry_t *active; + + if (!debug_active || !id->areas) + return NULL; + if (debug_critical) { + if (!spin_trylock_irqsave(&id->lock, flags)) + return NULL; + } else + spin_lock_irqsave(&id->lock, flags); + active = get_active_entry(id); + memset(DEBUG_DATA(active), 0, id->buf_size); + memcpy(DEBUG_DATA(active), buf, min(len, id->buf_size)); + debug_finish_entry(id, active, level, 0); + spin_unlock_irqrestore(&id->lock, flags); + + return active; +} +EXPORT_SYMBOL(debug_event_common); + +/* + * debug_exception_common: + * - write debug entry with given size and switch to next debug area + */ + +debug_entry_t +*debug_exception_common(debug_info_t * id, int level, const void *buf, int len) +{ + unsigned long flags; + debug_entry_t *active; + + if (!debug_active || !id->areas) + return NULL; + if (debug_critical) { + if (!spin_trylock_irqsave(&id->lock, flags)) + return NULL; + } else + spin_lock_irqsave(&id->lock, flags); + active = get_active_entry(id); + memset(DEBUG_DATA(active), 0, id->buf_size); + memcpy(DEBUG_DATA(active), buf, min(len, id->buf_size)); + debug_finish_entry(id, active, level, 1); + spin_unlock_irqrestore(&id->lock, flags); + + return active; +} +EXPORT_SYMBOL(debug_exception_common); + +/* + * counts arguments in format string for sprintf view + */ + +static inline int +debug_count_numargs(char *string) +{ + int numargs=0; + + while(*string) { + if(*string++=='%') + numargs++; + } + return(numargs); +} + +/* + * debug_sprintf_event: + */ + +debug_entry_t* +__debug_sprintf_event(debug_info_t *id, int level, char *string, ...) +{ + va_list ap; + int numargs,idx; + unsigned long flags; + debug_sprintf_entry_t *curr_event; + debug_entry_t *active; + + if (!debug_active || !id->areas) + return NULL; + numargs=debug_count_numargs(string); + + if (debug_critical) { + if (!spin_trylock_irqsave(&id->lock, flags)) + return NULL; + } else + spin_lock_irqsave(&id->lock, flags); + active = get_active_entry(id); + curr_event=(debug_sprintf_entry_t *) DEBUG_DATA(active); + va_start(ap,string); + curr_event->string=string; + for(idx=0;idx<min(numargs,(int)(id->buf_size / sizeof(long))-1);idx++) + curr_event->args[idx]=va_arg(ap,long); + va_end(ap); + debug_finish_entry(id, active, level, 0); + spin_unlock_irqrestore(&id->lock, flags); + + return active; +} +EXPORT_SYMBOL(__debug_sprintf_event); + +/* + * debug_sprintf_exception: + */ + +debug_entry_t* +__debug_sprintf_exception(debug_info_t *id, int level, char *string, ...) +{ + va_list ap; + int numargs,idx; + unsigned long flags; + debug_sprintf_entry_t *curr_event; + debug_entry_t *active; + + if (!debug_active || !id->areas) + return NULL; + + numargs=debug_count_numargs(string); + + if (debug_critical) { + if (!spin_trylock_irqsave(&id->lock, flags)) + return NULL; + } else + spin_lock_irqsave(&id->lock, flags); + active = get_active_entry(id); + curr_event=(debug_sprintf_entry_t *)DEBUG_DATA(active); + va_start(ap,string); + curr_event->string=string; + for(idx=0;idx<min(numargs,(int)(id->buf_size / sizeof(long))-1);idx++) + curr_event->args[idx]=va_arg(ap,long); + va_end(ap); + debug_finish_entry(id, active, level, 1); + spin_unlock_irqrestore(&id->lock, flags); + + return active; +} +EXPORT_SYMBOL(__debug_sprintf_exception); + +/* + * debug_register_view: + */ + +int +debug_register_view(debug_info_t * id, struct debug_view *view) +{ + int rc = 0; + int i; + unsigned long flags; + umode_t mode; + struct dentry *pde; + + if (!id) + goto out; + mode = (id->mode | S_IFREG) & ~S_IXUGO; + if (!(view->prolog_proc || view->format_proc || view->header_proc)) + mode &= ~(S_IRUSR | S_IRGRP | S_IROTH); + if (!view->input_proc) + mode &= ~(S_IWUSR | S_IWGRP | S_IWOTH); + pde = debugfs_create_file(view->name, mode, id->debugfs_root_entry, + id , &debug_file_ops); + if (!pde){ + pr_err("Registering view %s/%s failed due to out of " + "memory\n", id->name,view->name); + rc = -1; + goto out; + } + spin_lock_irqsave(&id->lock, flags); + for (i = 0; i < DEBUG_MAX_VIEWS; i++) { + if (!id->views[i]) + break; + } + if (i == DEBUG_MAX_VIEWS) { + pr_err("Registering view %s/%s would exceed the maximum " + "number of views %i\n", id->name, view->name, i); + rc = -1; + } else { + id->views[i] = view; + id->debugfs_entries[i] = pde; + } + spin_unlock_irqrestore(&id->lock, flags); + if (rc) + debugfs_remove(pde); +out: + return rc; +} +EXPORT_SYMBOL(debug_register_view); + +/* + * debug_unregister_view: + */ + +int +debug_unregister_view(debug_info_t * id, struct debug_view *view) +{ + struct dentry *dentry = NULL; + unsigned long flags; + int i, rc = 0; + + if (!id) + goto out; + spin_lock_irqsave(&id->lock, flags); + for (i = 0; i < DEBUG_MAX_VIEWS; i++) { + if (id->views[i] == view) + break; + } + if (i == DEBUG_MAX_VIEWS) + rc = -1; + else { + dentry = id->debugfs_entries[i]; + id->views[i] = NULL; + id->debugfs_entries[i] = NULL; + } + spin_unlock_irqrestore(&id->lock, flags); + debugfs_remove(dentry); +out: + return rc; +} +EXPORT_SYMBOL(debug_unregister_view); + +static inline char * +debug_get_user_string(const char __user *user_buf, size_t user_len) +{ + char* buffer; + + buffer = kmalloc(user_len + 1, GFP_KERNEL); + if (!buffer) + return ERR_PTR(-ENOMEM); + if (copy_from_user(buffer, user_buf, user_len) != 0) { + kfree(buffer); + return ERR_PTR(-EFAULT); + } + /* got the string, now strip linefeed. */ + if (buffer[user_len - 1] == '\n') + buffer[user_len - 1] = 0; + else + buffer[user_len] = 0; + return buffer; +} + +static inline int +debug_get_uint(char *buf) +{ + int rc; + + buf = skip_spaces(buf); + rc = simple_strtoul(buf, &buf, 10); + if(*buf){ + rc = -EINVAL; + } + return rc; +} + +/* + * functions for debug-views + *********************************** +*/ + +/* + * prints out actual debug level + */ + +static int +debug_prolog_pages_fn(debug_info_t * id, + struct debug_view *view, char *out_buf) +{ + return sprintf(out_buf, "%i\n", id->pages_per_area); +} + +/* + * reads new size (number of pages per debug area) + */ + +static int +debug_input_pages_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_len, loff_t * offset) +{ + char *str; + int rc,new_pages; + + if (user_len > 0x10000) + user_len = 0x10000; + if (*offset != 0){ + rc = -EPIPE; + goto out; + } + str = debug_get_user_string(user_buf,user_len); + if(IS_ERR(str)){ + rc = PTR_ERR(str); + goto out; + } + new_pages = debug_get_uint(str); + if(new_pages < 0){ + rc = -EINVAL; + goto free_str; + } + rc = debug_set_size(id,id->nr_areas, new_pages); + if(rc != 0){ + rc = -EINVAL; + goto free_str; + } + rc = user_len; +free_str: + kfree(str); +out: + *offset += user_len; + return rc; /* number of input characters */ +} + +/* + * prints out actual debug level + */ + +static int +debug_prolog_level_fn(debug_info_t * id, struct debug_view *view, char *out_buf) +{ + int rc = 0; + + if(id->level == DEBUG_OFF_LEVEL) { + rc = sprintf(out_buf,"-\n"); + } + else { + rc = sprintf(out_buf, "%i\n", id->level); + } + return rc; +} + +/* + * reads new debug level + */ + +static int +debug_input_level_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_len, loff_t * offset) +{ + char *str; + int rc,new_level; + + if (user_len > 0x10000) + user_len = 0x10000; + if (*offset != 0){ + rc = -EPIPE; + goto out; + } + str = debug_get_user_string(user_buf,user_len); + if(IS_ERR(str)){ + rc = PTR_ERR(str); + goto out; + } + if(str[0] == '-'){ + debug_set_level(id, DEBUG_OFF_LEVEL); + rc = user_len; + goto free_str; + } else { + new_level = debug_get_uint(str); + } + if(new_level < 0) { + pr_warning("%s is not a valid level for a debug " + "feature\n", str); + rc = -EINVAL; + } else { + debug_set_level(id, new_level); + rc = user_len; + } +free_str: + kfree(str); +out: + *offset += user_len; + return rc; /* number of input characters */ +} + + +/* + * flushes debug areas + */ + +static void debug_flush(debug_info_t* id, int area) +{ + unsigned long flags; + int i,j; + + if(!id || !id->areas) + return; + spin_lock_irqsave(&id->lock,flags); + if(area == DEBUG_FLUSH_ALL){ + id->active_area = 0; + memset(id->active_entries, 0, id->nr_areas * sizeof(int)); + for (i = 0; i < id->nr_areas; i++) { + id->active_pages[i] = 0; + for(j = 0; j < id->pages_per_area; j++) { + memset(id->areas[i][j], 0, PAGE_SIZE); + } + } + } else if(area >= 0 && area < id->nr_areas) { + id->active_entries[area] = 0; + id->active_pages[area] = 0; + for(i = 0; i < id->pages_per_area; i++) { + memset(id->areas[area][i],0,PAGE_SIZE); + } + } + spin_unlock_irqrestore(&id->lock,flags); +} + +/* + * view function: flushes debug areas + */ + +static int +debug_input_flush_fn(debug_info_t * id, struct debug_view *view, + struct file *file, const char __user *user_buf, + size_t user_len, loff_t * offset) +{ + char input_buf[1]; + int rc = user_len; + + if (user_len > 0x10000) + user_len = 0x10000; + if (*offset != 0){ + rc = -EPIPE; + goto out; + } + if (copy_from_user(input_buf, user_buf, 1)){ + rc = -EFAULT; + goto out; + } + if(input_buf[0] == '-') { + debug_flush(id, DEBUG_FLUSH_ALL); + goto out; + } + if (isdigit(input_buf[0])) { + int area = ((int) input_buf[0] - (int) '0'); + debug_flush(id, area); + goto out; + } + + pr_info("Flushing debug data failed because %c is not a valid " + "area\n", input_buf[0]); + +out: + *offset += user_len; + return rc; /* number of input characters */ +} + +/* + * prints debug header in raw format + */ + +static int +debug_raw_header_fn(debug_info_t * id, struct debug_view *view, + int area, debug_entry_t * entry, char *out_buf) +{ + int rc; + + rc = sizeof(debug_entry_t); + memcpy(out_buf,entry,sizeof(debug_entry_t)); + return rc; +} + +/* + * prints debug data in raw format + */ + +static int +debug_raw_format_fn(debug_info_t * id, struct debug_view *view, + char *out_buf, const char *in_buf) +{ + int rc; + + rc = id->buf_size; + memcpy(out_buf, in_buf, id->buf_size); + return rc; +} + +/* + * prints debug data in hex/ascii format + */ + +static int +debug_hex_ascii_format_fn(debug_info_t * id, struct debug_view *view, + char *out_buf, const char *in_buf) +{ + int i, rc = 0; + + for (i = 0; i < id->buf_size; i++) { + rc += sprintf(out_buf + rc, "%02x ", + ((unsigned char *) in_buf)[i]); + } + rc += sprintf(out_buf + rc, "| "); + for (i = 0; i < id->buf_size; i++) { + unsigned char c = in_buf[i]; + if (isascii(c) && isprint(c)) + rc += sprintf(out_buf + rc, "%c", c); + else + rc += sprintf(out_buf + rc, "."); + } + rc += sprintf(out_buf + rc, "\n"); + return rc; +} + +/* + * prints header for debug entry + */ + +int +debug_dflt_header_fn(debug_info_t * id, struct debug_view *view, + int area, debug_entry_t * entry, char *out_buf) +{ + struct timespec time_spec; + char *except_str; + unsigned long caller; + int rc = 0; + unsigned int level; + + level = entry->id.fields.level; + stck_to_timespec(entry->id.stck, &time_spec); + + if (entry->id.fields.exception) + except_str = "*"; + else + except_str = "-"; + caller = ((unsigned long) entry->caller) & PSW_ADDR_INSN; + rc += sprintf(out_buf, "%02i %011lu:%06lu %1u %1s %02i %p ", + area, time_spec.tv_sec, time_spec.tv_nsec / 1000, level, + except_str, entry->id.fields.cpuid, (void *) caller); + return rc; +} +EXPORT_SYMBOL(debug_dflt_header_fn); + +/* + * prints debug data sprintf-formated: + * debug_sprinf_event/exception calls must be used together with this view + */ + +#define DEBUG_SPRINTF_MAX_ARGS 10 + +static int +debug_sprintf_format_fn(debug_info_t * id, struct debug_view *view, + char *out_buf, debug_sprintf_entry_t *curr_event) +{ + int num_longs, num_used_args = 0,i, rc = 0; + int index[DEBUG_SPRINTF_MAX_ARGS]; + + /* count of longs fit into one entry */ + num_longs = id->buf_size / sizeof(long); + + if(num_longs < 1) + goto out; /* bufsize of entry too small */ + if(num_longs == 1) { + /* no args, we use only the string */ + strcpy(out_buf, curr_event->string); + rc = strlen(curr_event->string); + goto out; + } + + /* number of arguments used for sprintf (without the format string) */ + num_used_args = min(DEBUG_SPRINTF_MAX_ARGS, (num_longs - 1)); + + memset(index,0, DEBUG_SPRINTF_MAX_ARGS * sizeof(int)); + + for(i = 0; i < num_used_args; i++) + index[i] = i; + + rc = sprintf(out_buf, curr_event->string, curr_event->args[index[0]], + curr_event->args[index[1]], curr_event->args[index[2]], + curr_event->args[index[3]], curr_event->args[index[4]], + curr_event->args[index[5]], curr_event->args[index[6]], + curr_event->args[index[7]], curr_event->args[index[8]], + curr_event->args[index[9]]); + +out: + + return rc; +} + +/* + * debug_init: + * - is called exactly once to initialize the debug feature + */ +static int __init debug_init(void) +{ + s390dbf_sysctl_header = register_sysctl_table(s390dbf_dir_table); + mutex_lock(&debug_mutex); + debug_debugfs_root_entry = debugfs_create_dir(DEBUG_DIR_ROOT, NULL); + initialized = 1; + mutex_unlock(&debug_mutex); + return 0; +} +postcore_initcall(debug_init); diff --git a/arch/s390/kernel/diag.c b/arch/s390/kernel/diag.c new file mode 100644 index 000000000..2f69243bf --- /dev/null +++ b/arch/s390/kernel/diag.c @@ -0,0 +1,66 @@ +/* + * Implementation of s390 diagnose codes + * + * Copyright IBM Corp. 2007 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#include <linux/module.h> +#include <asm/diag.h> + +/* + * Diagnose 14: Input spool file manipulation + */ +int diag14(unsigned long rx, unsigned long ry1, unsigned long subcode) +{ + register unsigned long _ry1 asm("2") = ry1; + register unsigned long _ry2 asm("3") = subcode; + int rc = 0; + + asm volatile( + " sam31\n" + " diag %2,2,0x14\n" + " sam64\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (rc), "+d" (_ry2) + : "d" (rx), "d" (_ry1) + : "cc"); + + return rc; +} +EXPORT_SYMBOL(diag14); + +/* + * Diagnose 210: Get information about a virtual device + */ +int diag210(struct diag210 *addr) +{ + /* + * diag 210 needs its data below the 2GB border, so we + * use a static data area to be sure + */ + static struct diag210 diag210_tmp; + static DEFINE_SPINLOCK(diag210_lock); + unsigned long flags; + int ccode; + + spin_lock_irqsave(&diag210_lock, flags); + diag210_tmp = *addr; + + asm volatile( + " lhi %0,-1\n" + " sam31\n" + " diag %1,0,0x210\n" + "0: ipm %0\n" + " srl %0,28\n" + "1: sam64\n" + EX_TABLE(0b, 1b) + : "=&d" (ccode) : "a" (&diag210_tmp) : "cc", "memory"); + + *addr = diag210_tmp; + spin_unlock_irqrestore(&diag210_lock, flags); + + return ccode; +} +EXPORT_SYMBOL(diag210); diff --git a/arch/s390/kernel/dis.c b/arch/s390/kernel/dis.c new file mode 100644 index 000000000..8140d10c6 --- /dev/null +++ b/arch/s390/kernel/dis.c @@ -0,0 +1,2049 @@ +/* + * Disassemble s390 instructions. + * + * Copyright IBM Corp. 2007 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + */ + +#include <linux/sched.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/ptrace.h> +#include <linux/timer.h> +#include <linux/mm.h> +#include <linux/smp.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/delay.h> +#include <linux/module.h> +#include <linux/kallsyms.h> +#include <linux/reboot.h> +#include <linux/kprobes.h> +#include <linux/kdebug.h> + +#include <asm/uaccess.h> +#include <asm/dis.h> +#include <asm/io.h> +#include <linux/atomic.h> +#include <asm/mathemu.h> +#include <asm/cpcmd.h> +#include <asm/lowcore.h> +#include <asm/debug.h> +#include <asm/irq.h> + +enum { + UNUSED, /* Indicates the end of the operand list */ + R_8, /* GPR starting at position 8 */ + R_12, /* GPR starting at position 12 */ + R_16, /* GPR starting at position 16 */ + R_20, /* GPR starting at position 20 */ + R_24, /* GPR starting at position 24 */ + R_28, /* GPR starting at position 28 */ + R_32, /* GPR starting at position 32 */ + F_8, /* FPR starting at position 8 */ + F_12, /* FPR starting at position 12 */ + F_16, /* FPR starting at position 16 */ + F_20, /* FPR starting at position 16 */ + F_24, /* FPR starting at position 24 */ + F_28, /* FPR starting at position 28 */ + F_32, /* FPR starting at position 32 */ + A_8, /* Access reg. starting at position 8 */ + A_12, /* Access reg. starting at position 12 */ + A_24, /* Access reg. starting at position 24 */ + A_28, /* Access reg. starting at position 28 */ + C_8, /* Control reg. starting at position 8 */ + C_12, /* Control reg. starting at position 12 */ + V_8, /* Vector reg. starting at position 8, extension bit at 36 */ + V_12, /* Vector reg. starting at position 12, extension bit at 37 */ + V_16, /* Vector reg. starting at position 16, extension bit at 38 */ + V_32, /* Vector reg. starting at position 32, extension bit at 39 */ + W_12, /* Vector reg. at bit 12, extension at bit 37, used as index */ + B_16, /* Base register starting at position 16 */ + B_32, /* Base register starting at position 32 */ + X_12, /* Index register starting at position 12 */ + D_20, /* Displacement starting at position 20 */ + D_36, /* Displacement starting at position 36 */ + D20_20, /* 20 bit displacement starting at 20 */ + L4_8, /* 4 bit length starting at position 8 */ + L4_12, /* 4 bit length starting at position 12 */ + L8_8, /* 8 bit length starting at position 8 */ + U4_8, /* 4 bit unsigned value starting at 8 */ + U4_12, /* 4 bit unsigned value starting at 12 */ + U4_16, /* 4 bit unsigned value starting at 16 */ + U4_20, /* 4 bit unsigned value starting at 20 */ + U4_24, /* 4 bit unsigned value starting at 24 */ + U4_28, /* 4 bit unsigned value starting at 28 */ + U4_32, /* 4 bit unsigned value starting at 32 */ + U4_36, /* 4 bit unsigned value starting at 36 */ + U8_8, /* 8 bit unsigned value starting at 8 */ + U8_16, /* 8 bit unsigned value starting at 16 */ + U8_24, /* 8 bit unsigned value starting at 24 */ + U8_32, /* 8 bit unsigned value starting at 32 */ + I8_8, /* 8 bit signed value starting at 8 */ + I8_16, /* 8 bit signed value starting at 16 */ + I8_24, /* 8 bit signed value starting at 24 */ + I8_32, /* 8 bit signed value starting at 32 */ + J12_12, /* PC relative offset at 12 */ + I16_16, /* 16 bit signed value starting at 16 */ + I16_32, /* 32 bit signed value starting at 16 */ + U16_16, /* 16 bit unsigned value starting at 16 */ + U16_32, /* 32 bit unsigned value starting at 16 */ + J16_16, /* PC relative jump offset at 16 */ + J16_32, /* PC relative offset at 16 */ + I24_24, /* 24 bit signed value starting at 24 */ + J32_16, /* PC relative long offset at 16 */ + I32_16, /* 32 bit signed value starting at 16 */ + U32_16, /* 32 bit unsigned value starting at 16 */ + M_16, /* 4 bit optional mask starting at 16 */ + M_20, /* 4 bit optional mask starting at 20 */ + M_24, /* 4 bit optional mask starting at 24 */ + M_28, /* 4 bit optional mask starting at 28 */ + M_32, /* 4 bit optional mask starting at 32 */ + RO_28, /* optional GPR starting at position 28 */ +}; + +/* + * Enumeration of the different instruction formats. + * For details consult the principles of operation. + */ +enum { + INSTR_INVALID, + INSTR_E, + INSTR_IE_UU, + INSTR_MII_UPI, + INSTR_RIE_R0IU, INSTR_RIE_R0UU, INSTR_RIE_RRP, INSTR_RIE_RRPU, + INSTR_RIE_RRUUU, INSTR_RIE_RUPI, INSTR_RIE_RUPU, INSTR_RIE_RRI0, + INSTR_RIL_RI, INSTR_RIL_RP, INSTR_RIL_RU, INSTR_RIL_UP, + INSTR_RIS_R0RDU, INSTR_RIS_R0UU, INSTR_RIS_RURDI, INSTR_RIS_RURDU, + INSTR_RI_RI, INSTR_RI_RP, INSTR_RI_RU, INSTR_RI_UP, + INSTR_RRE_00, INSTR_RRE_0R, INSTR_RRE_AA, INSTR_RRE_AR, INSTR_RRE_F0, + INSTR_RRE_FF, INSTR_RRE_FR, INSTR_RRE_R0, INSTR_RRE_RA, INSTR_RRE_RF, + INSTR_RRE_RR, INSTR_RRE_RR_OPT, + INSTR_RRF_0UFF, INSTR_RRF_F0FF, INSTR_RRF_F0FF2, INSTR_RRF_F0FR, + INSTR_RRF_FFRU, INSTR_RRF_FUFF, INSTR_RRF_FUFF2, INSTR_RRF_M0RR, + INSTR_RRF_R0RR, INSTR_RRF_R0RR2, INSTR_RRF_RMRR, INSTR_RRF_RURR, + INSTR_RRF_U0FF, INSTR_RRF_U0RF, INSTR_RRF_U0RR, INSTR_RRF_UUFF, + INSTR_RRF_UUFR, INSTR_RRF_UURF, + INSTR_RRR_F0FF, INSTR_RRS_RRRDU, + INSTR_RR_FF, INSTR_RR_R0, INSTR_RR_RR, INSTR_RR_U0, INSTR_RR_UR, + INSTR_RSE_CCRD, INSTR_RSE_RRRD, INSTR_RSE_RURD, + INSTR_RSI_RRP, + INSTR_RSL_LRDFU, INSTR_RSL_R0RD, + INSTR_RSY_AARD, INSTR_RSY_CCRD, INSTR_RSY_RRRD, INSTR_RSY_RURD, + INSTR_RSY_RDRM, INSTR_RSY_RMRD, + INSTR_RS_AARD, INSTR_RS_CCRD, INSTR_RS_R0RD, INSTR_RS_RRRD, + INSTR_RS_RURD, + INSTR_RXE_FRRD, INSTR_RXE_RRRD, INSTR_RXE_RRRDM, + INSTR_RXF_FRRDF, + INSTR_RXY_FRRD, INSTR_RXY_RRRD, INSTR_RXY_URRD, + INSTR_RX_FRRD, INSTR_RX_RRRD, INSTR_RX_URRD, + INSTR_SIL_RDI, INSTR_SIL_RDU, + INSTR_SIY_IRD, INSTR_SIY_URD, + INSTR_SI_URD, + INSTR_SMI_U0RDP, + INSTR_SSE_RDRD, + INSTR_SSF_RRDRD, INSTR_SSF_RRDRD2, + INSTR_SS_L0RDRD, INSTR_SS_LIRDRD, INSTR_SS_LLRDRD, INSTR_SS_RRRDRD, + INSTR_SS_RRRDRD2, INSTR_SS_RRRDRD3, + INSTR_S_00, INSTR_S_RD, + INSTR_VRI_V0IM, INSTR_VRI_V0I0, INSTR_VRI_V0IIM, INSTR_VRI_VVIM, + INSTR_VRI_VVV0IM, INSTR_VRI_VVV0I0, INSTR_VRI_VVIMM, + INSTR_VRR_VV00MMM, INSTR_VRR_VV000MM, INSTR_VRR_VV0000M, + INSTR_VRR_VV00000, INSTR_VRR_VVV0M0M, INSTR_VRR_VV00M0M, + INSTR_VRR_VVV000M, INSTR_VRR_VVV000V, INSTR_VRR_VVV0000, + INSTR_VRR_VVV0MMM, INSTR_VRR_VVV00MM, INSTR_VRR_VVVMM0V, + INSTR_VRR_VVVM0MV, INSTR_VRR_VVVM00V, INSTR_VRR_VRR0000, + INSTR_VRS_VVRDM, INSTR_VRS_VVRD0, INSTR_VRS_VRRDM, INSTR_VRS_VRRD0, + INSTR_VRS_RVRDM, + INSTR_VRV_VVRDM, INSTR_VRV_VWRDM, + INSTR_VRX_VRRDM, INSTR_VRX_VRRD0, +}; + +static const struct s390_operand operands[] = +{ + [UNUSED] = { 0, 0, 0 }, + [R_8] = { 4, 8, OPERAND_GPR }, + [R_12] = { 4, 12, OPERAND_GPR }, + [R_16] = { 4, 16, OPERAND_GPR }, + [R_20] = { 4, 20, OPERAND_GPR }, + [R_24] = { 4, 24, OPERAND_GPR }, + [R_28] = { 4, 28, OPERAND_GPR }, + [R_32] = { 4, 32, OPERAND_GPR }, + [F_8] = { 4, 8, OPERAND_FPR }, + [F_12] = { 4, 12, OPERAND_FPR }, + [F_16] = { 4, 16, OPERAND_FPR }, + [F_20] = { 4, 16, OPERAND_FPR }, + [F_24] = { 4, 24, OPERAND_FPR }, + [F_28] = { 4, 28, OPERAND_FPR }, + [F_32] = { 4, 32, OPERAND_FPR }, + [A_8] = { 4, 8, OPERAND_AR }, + [A_12] = { 4, 12, OPERAND_AR }, + [A_24] = { 4, 24, OPERAND_AR }, + [A_28] = { 4, 28, OPERAND_AR }, + [C_8] = { 4, 8, OPERAND_CR }, + [C_12] = { 4, 12, OPERAND_CR }, + [V_8] = { 4, 8, OPERAND_VR }, + [V_12] = { 4, 12, OPERAND_VR }, + [V_16] = { 4, 16, OPERAND_VR }, + [V_32] = { 4, 32, OPERAND_VR }, + [W_12] = { 4, 12, OPERAND_INDEX | OPERAND_VR }, + [B_16] = { 4, 16, OPERAND_BASE | OPERAND_GPR }, + [B_32] = { 4, 32, OPERAND_BASE | OPERAND_GPR }, + [X_12] = { 4, 12, OPERAND_INDEX | OPERAND_GPR }, + [D_20] = { 12, 20, OPERAND_DISP }, + [D_36] = { 12, 36, OPERAND_DISP }, + [D20_20] = { 20, 20, OPERAND_DISP | OPERAND_SIGNED }, + [L4_8] = { 4, 8, OPERAND_LENGTH }, + [L4_12] = { 4, 12, OPERAND_LENGTH }, + [L8_8] = { 8, 8, OPERAND_LENGTH }, + [U4_8] = { 4, 8, 0 }, + [U4_12] = { 4, 12, 0 }, + [U4_16] = { 4, 16, 0 }, + [U4_20] = { 4, 20, 0 }, + [U4_24] = { 4, 24, 0 }, + [U4_28] = { 4, 28, 0 }, + [U4_32] = { 4, 32, 0 }, + [U4_36] = { 4, 36, 0 }, + [U8_8] = { 8, 8, 0 }, + [U8_16] = { 8, 16, 0 }, + [U8_24] = { 8, 24, 0 }, + [U8_32] = { 8, 32, 0 }, + [J12_12] = { 12, 12, OPERAND_PCREL }, + [I8_8] = { 8, 8, OPERAND_SIGNED }, + [I8_16] = { 8, 16, OPERAND_SIGNED }, + [I8_24] = { 8, 24, OPERAND_SIGNED }, + [I8_32] = { 8, 32, OPERAND_SIGNED }, + [I16_32] = { 16, 32, OPERAND_SIGNED }, + [I16_16] = { 16, 16, OPERAND_SIGNED }, + [U16_16] = { 16, 16, 0 }, + [U16_32] = { 16, 32, 0 }, + [J16_16] = { 16, 16, OPERAND_PCREL }, + [J16_32] = { 16, 32, OPERAND_PCREL }, + [I24_24] = { 24, 24, OPERAND_SIGNED }, + [J32_16] = { 32, 16, OPERAND_PCREL }, + [I32_16] = { 32, 16, OPERAND_SIGNED }, + [U32_16] = { 32, 16, 0 }, + [M_16] = { 4, 16, 0 }, + [M_20] = { 4, 20, 0 }, + [M_24] = { 4, 24, 0 }, + [M_28] = { 4, 28, 0 }, + [M_32] = { 4, 32, 0 }, + [RO_28] = { 4, 28, OPERAND_GPR } +}; + +static const unsigned char formats[][7] = { + [INSTR_E] = { 0xff, 0,0,0,0,0,0 }, + [INSTR_IE_UU] = { 0xff, U4_24,U4_28,0,0,0,0 }, + [INSTR_MII_UPI] = { 0xff, U4_8,J12_12,I24_24 }, + [INSTR_RIE_R0IU] = { 0xff, R_8,I16_16,U4_32,0,0,0 }, + [INSTR_RIE_R0UU] = { 0xff, R_8,U16_16,U4_32,0,0,0 }, + [INSTR_RIE_RRI0] = { 0xff, R_8,R_12,I16_16,0,0,0 }, + [INSTR_RIE_RRPU] = { 0xff, R_8,R_12,U4_32,J16_16,0,0 }, + [INSTR_RIE_RRP] = { 0xff, R_8,R_12,J16_16,0,0,0 }, + [INSTR_RIE_RRUUU] = { 0xff, R_8,R_12,U8_16,U8_24,U8_32,0 }, + [INSTR_RIE_RUPI] = { 0xff, R_8,I8_32,U4_12,J16_16,0,0 }, + [INSTR_RIE_RUPU] = { 0xff, R_8,U8_32,U4_12,J16_16,0,0 }, + [INSTR_RIL_RI] = { 0x0f, R_8,I32_16,0,0,0,0 }, + [INSTR_RIL_RP] = { 0x0f, R_8,J32_16,0,0,0,0 }, + [INSTR_RIL_RU] = { 0x0f, R_8,U32_16,0,0,0,0 }, + [INSTR_RIL_UP] = { 0x0f, U4_8,J32_16,0,0,0,0 }, + [INSTR_RIS_R0RDU] = { 0xff, R_8,U8_32,D_20,B_16,0,0 }, + [INSTR_RIS_RURDI] = { 0xff, R_8,I8_32,U4_12,D_20,B_16,0 }, + [INSTR_RIS_RURDU] = { 0xff, R_8,U8_32,U4_12,D_20,B_16,0 }, + [INSTR_RI_RI] = { 0x0f, R_8,I16_16,0,0,0,0 }, + [INSTR_RI_RP] = { 0x0f, R_8,J16_16,0,0,0,0 }, + [INSTR_RI_RU] = { 0x0f, R_8,U16_16,0,0,0,0 }, + [INSTR_RI_UP] = { 0x0f, U4_8,J16_16,0,0,0,0 }, + [INSTR_RRE_00] = { 0xff, 0,0,0,0,0,0 }, + [INSTR_RRE_0R] = { 0xff, R_28,0,0,0,0,0 }, + [INSTR_RRE_AA] = { 0xff, A_24,A_28,0,0,0,0 }, + [INSTR_RRE_AR] = { 0xff, A_24,R_28,0,0,0,0 }, + [INSTR_RRE_F0] = { 0xff, F_24,0,0,0,0,0 }, + [INSTR_RRE_FF] = { 0xff, F_24,F_28,0,0,0,0 }, + [INSTR_RRE_FR] = { 0xff, F_24,R_28,0,0,0,0 }, + [INSTR_RRE_R0] = { 0xff, R_24,0,0,0,0,0 }, + [INSTR_RRE_RA] = { 0xff, R_24,A_28,0,0,0,0 }, + [INSTR_RRE_RF] = { 0xff, R_24,F_28,0,0,0,0 }, + [INSTR_RRE_RR] = { 0xff, R_24,R_28,0,0,0,0 }, + [INSTR_RRE_RR_OPT]= { 0xff, R_24,RO_28,0,0,0,0 }, + [INSTR_RRF_0UFF] = { 0xff, F_24,F_28,U4_20,0,0,0 }, + [INSTR_RRF_F0FF2] = { 0xff, F_24,F_16,F_28,0,0,0 }, + [INSTR_RRF_F0FF] = { 0xff, F_16,F_24,F_28,0,0,0 }, + [INSTR_RRF_F0FR] = { 0xff, F_24,F_16,R_28,0,0,0 }, + [INSTR_RRF_FFRU] = { 0xff, F_24,F_16,R_28,U4_20,0,0 }, + [INSTR_RRF_FUFF] = { 0xff, F_24,F_16,F_28,U4_20,0,0 }, + [INSTR_RRF_FUFF2] = { 0xff, F_24,F_28,F_16,U4_20,0,0 }, + [INSTR_RRF_M0RR] = { 0xff, R_24,R_28,M_16,0,0,0 }, + [INSTR_RRF_R0RR] = { 0xff, R_24,R_16,R_28,0,0,0 }, + [INSTR_RRF_R0RR2] = { 0xff, R_24,R_28,R_16,0,0,0 }, + [INSTR_RRF_RMRR] = { 0xff, R_24,R_16,R_28,M_20,0,0 }, + [INSTR_RRF_RURR] = { 0xff, R_24,R_28,R_16,U4_20,0,0 }, + [INSTR_RRF_U0FF] = { 0xff, F_24,U4_16,F_28,0,0,0 }, + [INSTR_RRF_U0RF] = { 0xff, R_24,U4_16,F_28,0,0,0 }, + [INSTR_RRF_U0RR] = { 0xff, R_24,R_28,U4_16,0,0,0 }, + [INSTR_RRF_UUFF] = { 0xff, F_24,U4_16,F_28,U4_20,0,0 }, + [INSTR_RRF_UUFR] = { 0xff, F_24,U4_16,R_28,U4_20,0,0 }, + [INSTR_RRF_UURF] = { 0xff, R_24,U4_16,F_28,U4_20,0,0 }, + [INSTR_RRR_F0FF] = { 0xff, F_24,F_28,F_16,0,0,0 }, + [INSTR_RRS_RRRDU] = { 0xff, R_8,R_12,U4_32,D_20,B_16,0 }, + [INSTR_RR_FF] = { 0xff, F_8,F_12,0,0,0,0 }, + [INSTR_RR_R0] = { 0xff, R_8, 0,0,0,0,0 }, + [INSTR_RR_RR] = { 0xff, R_8,R_12,0,0,0,0 }, + [INSTR_RR_U0] = { 0xff, U8_8, 0,0,0,0,0 }, + [INSTR_RR_UR] = { 0xff, U4_8,R_12,0,0,0,0 }, + [INSTR_RSE_CCRD] = { 0xff, C_8,C_12,D_20,B_16,0,0 }, + [INSTR_RSE_RRRD] = { 0xff, R_8,R_12,D_20,B_16,0,0 }, + [INSTR_RSE_RURD] = { 0xff, R_8,U4_12,D_20,B_16,0,0 }, + [INSTR_RSI_RRP] = { 0xff, R_8,R_12,J16_16,0,0,0 }, + [INSTR_RSL_LRDFU] = { 0xff, F_32,D_20,L4_8,B_16,U4_36,0 }, + [INSTR_RSL_R0RD] = { 0xff, D_20,L4_8,B_16,0,0,0 }, + [INSTR_RSY_AARD] = { 0xff, A_8,A_12,D20_20,B_16,0,0 }, + [INSTR_RSY_CCRD] = { 0xff, C_8,C_12,D20_20,B_16,0,0 }, + [INSTR_RSY_RDRM] = { 0xff, R_8,D20_20,B_16,U4_12,0,0 }, + [INSTR_RSY_RMRD] = { 0xff, R_8,U4_12,D20_20,B_16,0,0 }, + [INSTR_RSY_RRRD] = { 0xff, R_8,R_12,D20_20,B_16,0,0 }, + [INSTR_RSY_RURD] = { 0xff, R_8,U4_12,D20_20,B_16,0,0 }, + [INSTR_RS_AARD] = { 0xff, A_8,A_12,D_20,B_16,0,0 }, + [INSTR_RS_CCRD] = { 0xff, C_8,C_12,D_20,B_16,0,0 }, + [INSTR_RS_R0RD] = { 0xff, R_8,D_20,B_16,0,0,0 }, + [INSTR_RS_RRRD] = { 0xff, R_8,R_12,D_20,B_16,0,0 }, + [INSTR_RS_RURD] = { 0xff, R_8,U4_12,D_20,B_16,0,0 }, + [INSTR_RXE_FRRD] = { 0xff, F_8,D_20,X_12,B_16,0,0 }, + [INSTR_RXE_RRRD] = { 0xff, R_8,D_20,X_12,B_16,0,0 }, + [INSTR_RXE_RRRDM] = { 0xff, R_8,D_20,X_12,B_16,M_32,0 }, + [INSTR_RXF_FRRDF] = { 0xff, F_32,F_8,D_20,X_12,B_16,0 }, + [INSTR_RXY_FRRD] = { 0xff, F_8,D20_20,X_12,B_16,0,0 }, + [INSTR_RXY_RRRD] = { 0xff, R_8,D20_20,X_12,B_16,0,0 }, + [INSTR_RXY_URRD] = { 0xff, U4_8,D20_20,X_12,B_16,0,0 }, + [INSTR_RX_FRRD] = { 0xff, F_8,D_20,X_12,B_16,0,0 }, + [INSTR_RX_RRRD] = { 0xff, R_8,D_20,X_12,B_16,0,0 }, + [INSTR_RX_URRD] = { 0xff, U4_8,D_20,X_12,B_16,0,0 }, + [INSTR_SIL_RDI] = { 0xff, D_20,B_16,I16_32,0,0,0 }, + [INSTR_SIL_RDU] = { 0xff, D_20,B_16,U16_32,0,0,0 }, + [INSTR_SIY_IRD] = { 0xff, D20_20,B_16,I8_8,0,0,0 }, + [INSTR_SIY_URD] = { 0xff, D20_20,B_16,U8_8,0,0,0 }, + [INSTR_SI_URD] = { 0xff, D_20,B_16,U8_8,0,0,0 }, + [INSTR_SMI_U0RDP] = { 0xff, U4_8,J16_32,D_20,B_16,0,0 }, + [INSTR_SSE_RDRD] = { 0xff, D_20,B_16,D_36,B_32,0,0 }, + [INSTR_SSF_RRDRD] = { 0x0f, D_20,B_16,D_36,B_32,R_8,0 }, + [INSTR_SSF_RRDRD2]= { 0x0f, R_8,D_20,B_16,D_36,B_32,0 }, + [INSTR_SS_L0RDRD] = { 0xff, D_20,L8_8,B_16,D_36,B_32,0 }, + [INSTR_SS_LIRDRD] = { 0xff, D_20,L4_8,B_16,D_36,B_32,U4_12 }, + [INSTR_SS_LLRDRD] = { 0xff, D_20,L4_8,B_16,D_36,L4_12,B_32 }, + [INSTR_SS_RRRDRD2]= { 0xff, R_8,D_20,B_16,R_12,D_36,B_32 }, + [INSTR_SS_RRRDRD3]= { 0xff, R_8,R_12,D_20,B_16,D_36,B_32 }, + [INSTR_SS_RRRDRD] = { 0xff, D_20,R_8,B_16,D_36,B_32,R_12 }, + [INSTR_S_00] = { 0xff, 0,0,0,0,0,0 }, + [INSTR_S_RD] = { 0xff, D_20,B_16,0,0,0,0 }, + [INSTR_VRI_V0IM] = { 0xff, V_8,I16_16,M_32,0,0,0 }, + [INSTR_VRI_V0I0] = { 0xff, V_8,I16_16,0,0,0,0 }, + [INSTR_VRI_V0IIM] = { 0xff, V_8,I8_16,I8_24,M_32,0,0 }, + [INSTR_VRI_VVIM] = { 0xff, V_8,I16_16,V_12,M_32,0,0 }, + [INSTR_VRI_VVV0IM]= { 0xff, V_8,V_12,V_16,I8_24,M_32,0 }, + [INSTR_VRI_VVV0I0]= { 0xff, V_8,V_12,V_16,I8_24,0,0 }, + [INSTR_VRI_VVIMM] = { 0xff, V_8,V_12,I16_16,M_32,M_28,0 }, + [INSTR_VRR_VV00MMM]={ 0xff, V_8,V_12,M_32,M_28,M_24,0 }, + [INSTR_VRR_VV000MM]={ 0xff, V_8,V_12,M_32,M_28,0,0 }, + [INSTR_VRR_VV0000M]={ 0xff, V_8,V_12,M_32,0,0,0 }, + [INSTR_VRR_VV00000]={ 0xff, V_8,V_12,0,0,0,0 }, + [INSTR_VRR_VVV0M0M]={ 0xff, V_8,V_12,V_16,M_32,M_24,0 }, + [INSTR_VRR_VV00M0M]={ 0xff, V_8,V_12,M_32,M_24,0,0 }, + [INSTR_VRR_VVV000M]={ 0xff, V_8,V_12,V_16,M_32,0,0 }, + [INSTR_VRR_VVV000V]={ 0xff, V_8,V_12,V_16,V_32,0,0 }, + [INSTR_VRR_VVV0000]={ 0xff, V_8,V_12,V_16,0,0,0 }, + [INSTR_VRR_VVV0MMM]={ 0xff, V_8,V_12,V_16,M_32,M_28,M_24 }, + [INSTR_VRR_VVV00MM]={ 0xff, V_8,V_12,V_16,M_32,M_28,0 }, + [INSTR_VRR_VVVMM0V]={ 0xff, V_8,V_12,V_16,V_32,M_20,M_24 }, + [INSTR_VRR_VVVM0MV]={ 0xff, V_8,V_12,V_16,V_32,M_28,M_20 }, + [INSTR_VRR_VVVM00V]={ 0xff, V_8,V_12,V_16,V_32,M_20,0 }, + [INSTR_VRR_VRR0000]={ 0xff, V_8,R_12,R_16,0,0,0 }, + [INSTR_VRS_VVRDM] = { 0xff, V_8,V_12,D_20,B_16,M_32,0 }, + [INSTR_VRS_VVRD0] = { 0xff, V_8,V_12,D_20,B_16,0,0 }, + [INSTR_VRS_VRRDM] = { 0xff, V_8,R_12,D_20,B_16,M_32,0 }, + [INSTR_VRS_VRRD0] = { 0xff, V_8,R_12,D_20,B_16,0,0 }, + [INSTR_VRS_RVRDM] = { 0xff, R_8,V_12,D_20,B_16,M_32,0 }, + [INSTR_VRV_VVRDM] = { 0xff, V_8,V_12,D_20,B_16,M_32,0 }, + [INSTR_VRV_VWRDM] = { 0xff, V_8,D_20,W_12,B_16,M_32,0 }, + [INSTR_VRX_VRRDM] = { 0xff, V_8,D_20,X_12,B_16,M_32,0 }, + [INSTR_VRX_VRRD0] = { 0xff, V_8,D_20,X_12,B_16,0,0 }, +}; + +enum { + LONG_INSN_ALGHSIK, + LONG_INSN_ALHHHR, + LONG_INSN_ALHHLR, + LONG_INSN_ALHSIK, + LONG_INSN_ALSIHN, + LONG_INSN_CDFBRA, + LONG_INSN_CDGBRA, + LONG_INSN_CDGTRA, + LONG_INSN_CDLFBR, + LONG_INSN_CDLFTR, + LONG_INSN_CDLGBR, + LONG_INSN_CDLGTR, + LONG_INSN_CEFBRA, + LONG_INSN_CEGBRA, + LONG_INSN_CELFBR, + LONG_INSN_CELGBR, + LONG_INSN_CFDBRA, + LONG_INSN_CFEBRA, + LONG_INSN_CFXBRA, + LONG_INSN_CGDBRA, + LONG_INSN_CGDTRA, + LONG_INSN_CGEBRA, + LONG_INSN_CGXBRA, + LONG_INSN_CGXTRA, + LONG_INSN_CLFDBR, + LONG_INSN_CLFDTR, + LONG_INSN_CLFEBR, + LONG_INSN_CLFHSI, + LONG_INSN_CLFXBR, + LONG_INSN_CLFXTR, + LONG_INSN_CLGDBR, + LONG_INSN_CLGDTR, + LONG_INSN_CLGEBR, + LONG_INSN_CLGFRL, + LONG_INSN_CLGHRL, + LONG_INSN_CLGHSI, + LONG_INSN_CLGXBR, + LONG_INSN_CLGXTR, + LONG_INSN_CLHHSI, + LONG_INSN_CXFBRA, + LONG_INSN_CXGBRA, + LONG_INSN_CXGTRA, + LONG_INSN_CXLFBR, + LONG_INSN_CXLFTR, + LONG_INSN_CXLGBR, + LONG_INSN_CXLGTR, + LONG_INSN_FIDBRA, + LONG_INSN_FIEBRA, + LONG_INSN_FIXBRA, + LONG_INSN_LDXBRA, + LONG_INSN_LEDBRA, + LONG_INSN_LEXBRA, + LONG_INSN_LLGFAT, + LONG_INSN_LLGFRL, + LONG_INSN_LLGHRL, + LONG_INSN_LLGTAT, + LONG_INSN_POPCNT, + LONG_INSN_RIEMIT, + LONG_INSN_RINEXT, + LONG_INSN_RISBGN, + LONG_INSN_RISBHG, + LONG_INSN_RISBLG, + LONG_INSN_SLHHHR, + LONG_INSN_SLHHLR, + LONG_INSN_TABORT, + LONG_INSN_TBEGIN, + LONG_INSN_TBEGINC, + LONG_INSN_PCISTG, + LONG_INSN_MPCIFC, + LONG_INSN_STPCIFC, + LONG_INSN_PCISTB, + LONG_INSN_VPOPCT, + LONG_INSN_VERLLV, + LONG_INSN_VESRAV, + LONG_INSN_VESRLV, + LONG_INSN_VSBCBI, + LONG_INSN_STCCTM +}; + +static char *long_insn_name[] = { + [LONG_INSN_ALGHSIK] = "alghsik", + [LONG_INSN_ALHHHR] = "alhhhr", + [LONG_INSN_ALHHLR] = "alhhlr", + [LONG_INSN_ALHSIK] = "alhsik", + [LONG_INSN_ALSIHN] = "alsihn", + [LONG_INSN_CDFBRA] = "cdfbra", + [LONG_INSN_CDGBRA] = "cdgbra", + [LONG_INSN_CDGTRA] = "cdgtra", + [LONG_INSN_CDLFBR] = "cdlfbr", + [LONG_INSN_CDLFTR] = "cdlftr", + [LONG_INSN_CDLGBR] = "cdlgbr", + [LONG_INSN_CDLGTR] = "cdlgtr", + [LONG_INSN_CEFBRA] = "cefbra", + [LONG_INSN_CEGBRA] = "cegbra", + [LONG_INSN_CELFBR] = "celfbr", + [LONG_INSN_CELGBR] = "celgbr", + [LONG_INSN_CFDBRA] = "cfdbra", + [LONG_INSN_CFEBRA] = "cfebra", + [LONG_INSN_CFXBRA] = "cfxbra", + [LONG_INSN_CGDBRA] = "cgdbra", + [LONG_INSN_CGDTRA] = "cgdtra", + [LONG_INSN_CGEBRA] = "cgebra", + [LONG_INSN_CGXBRA] = "cgxbra", + [LONG_INSN_CGXTRA] = "cgxtra", + [LONG_INSN_CLFDBR] = "clfdbr", + [LONG_INSN_CLFDTR] = "clfdtr", + [LONG_INSN_CLFEBR] = "clfebr", + [LONG_INSN_CLFHSI] = "clfhsi", + [LONG_INSN_CLFXBR] = "clfxbr", + [LONG_INSN_CLFXTR] = "clfxtr", + [LONG_INSN_CLGDBR] = "clgdbr", + [LONG_INSN_CLGDTR] = "clgdtr", + [LONG_INSN_CLGEBR] = "clgebr", + [LONG_INSN_CLGFRL] = "clgfrl", + [LONG_INSN_CLGHRL] = "clghrl", + [LONG_INSN_CLGHSI] = "clghsi", + [LONG_INSN_CLGXBR] = "clgxbr", + [LONG_INSN_CLGXTR] = "clgxtr", + [LONG_INSN_CLHHSI] = "clhhsi", + [LONG_INSN_CXFBRA] = "cxfbra", + [LONG_INSN_CXGBRA] = "cxgbra", + [LONG_INSN_CXGTRA] = "cxgtra", + [LONG_INSN_CXLFBR] = "cxlfbr", + [LONG_INSN_CXLFTR] = "cxlftr", + [LONG_INSN_CXLGBR] = "cxlgbr", + [LONG_INSN_CXLGTR] = "cxlgtr", + [LONG_INSN_FIDBRA] = "fidbra", + [LONG_INSN_FIEBRA] = "fiebra", + [LONG_INSN_FIXBRA] = "fixbra", + [LONG_INSN_LDXBRA] = "ldxbra", + [LONG_INSN_LEDBRA] = "ledbra", + [LONG_INSN_LEXBRA] = "lexbra", + [LONG_INSN_LLGFAT] = "llgfat", + [LONG_INSN_LLGFRL] = "llgfrl", + [LONG_INSN_LLGHRL] = "llghrl", + [LONG_INSN_LLGTAT] = "llgtat", + [LONG_INSN_POPCNT] = "popcnt", + [LONG_INSN_RIEMIT] = "riemit", + [LONG_INSN_RINEXT] = "rinext", + [LONG_INSN_RISBGN] = "risbgn", + [LONG_INSN_RISBHG] = "risbhg", + [LONG_INSN_RISBLG] = "risblg", + [LONG_INSN_SLHHHR] = "slhhhr", + [LONG_INSN_SLHHLR] = "slhhlr", + [LONG_INSN_TABORT] = "tabort", + [LONG_INSN_TBEGIN] = "tbegin", + [LONG_INSN_TBEGINC] = "tbeginc", + [LONG_INSN_PCISTG] = "pcistg", + [LONG_INSN_MPCIFC] = "mpcifc", + [LONG_INSN_STPCIFC] = "stpcifc", + [LONG_INSN_PCISTB] = "pcistb", + [LONG_INSN_VPOPCT] = "vpopct", + [LONG_INSN_VERLLV] = "verllv", + [LONG_INSN_VESRAV] = "vesrav", + [LONG_INSN_VESRLV] = "vesrlv", + [LONG_INSN_VSBCBI] = "vsbcbi", + [LONG_INSN_STCCTM] = "stcctm", +}; + +static struct s390_insn opcode[] = { + { "bprp", 0xc5, INSTR_MII_UPI }, + { "bpp", 0xc7, INSTR_SMI_U0RDP }, + { "trtr", 0xd0, INSTR_SS_L0RDRD }, + { "lmd", 0xef, INSTR_SS_RRRDRD3 }, + { "spm", 0x04, INSTR_RR_R0 }, + { "balr", 0x05, INSTR_RR_RR }, + { "bctr", 0x06, INSTR_RR_RR }, + { "bcr", 0x07, INSTR_RR_UR }, + { "svc", 0x0a, INSTR_RR_U0 }, + { "bsm", 0x0b, INSTR_RR_RR }, + { "bassm", 0x0c, INSTR_RR_RR }, + { "basr", 0x0d, INSTR_RR_RR }, + { "mvcl", 0x0e, INSTR_RR_RR }, + { "clcl", 0x0f, INSTR_RR_RR }, + { "lpr", 0x10, INSTR_RR_RR }, + { "lnr", 0x11, INSTR_RR_RR }, + { "ltr", 0x12, INSTR_RR_RR }, + { "lcr", 0x13, INSTR_RR_RR }, + { "nr", 0x14, INSTR_RR_RR }, + { "clr", 0x15, INSTR_RR_RR }, + { "or", 0x16, INSTR_RR_RR }, + { "xr", 0x17, INSTR_RR_RR }, + { "lr", 0x18, INSTR_RR_RR }, + { "cr", 0x19, INSTR_RR_RR }, + { "ar", 0x1a, INSTR_RR_RR }, + { "sr", 0x1b, INSTR_RR_RR }, + { "mr", 0x1c, INSTR_RR_RR }, + { "dr", 0x1d, INSTR_RR_RR }, + { "alr", 0x1e, INSTR_RR_RR }, + { "slr", 0x1f, INSTR_RR_RR }, + { "lpdr", 0x20, INSTR_RR_FF }, + { "lndr", 0x21, INSTR_RR_FF }, + { "ltdr", 0x22, INSTR_RR_FF }, + { "lcdr", 0x23, INSTR_RR_FF }, + { "hdr", 0x24, INSTR_RR_FF }, + { "ldxr", 0x25, INSTR_RR_FF }, + { "mxr", 0x26, INSTR_RR_FF }, + { "mxdr", 0x27, INSTR_RR_FF }, + { "ldr", 0x28, INSTR_RR_FF }, + { "cdr", 0x29, INSTR_RR_FF }, + { "adr", 0x2a, INSTR_RR_FF }, + { "sdr", 0x2b, INSTR_RR_FF }, + { "mdr", 0x2c, INSTR_RR_FF }, + { "ddr", 0x2d, INSTR_RR_FF }, + { "awr", 0x2e, INSTR_RR_FF }, + { "swr", 0x2f, INSTR_RR_FF }, + { "lper", 0x30, INSTR_RR_FF }, + { "lner", 0x31, INSTR_RR_FF }, + { "lter", 0x32, INSTR_RR_FF }, + { "lcer", 0x33, INSTR_RR_FF }, + { "her", 0x34, INSTR_RR_FF }, + { "ledr", 0x35, INSTR_RR_FF }, + { "axr", 0x36, INSTR_RR_FF }, + { "sxr", 0x37, INSTR_RR_FF }, + { "ler", 0x38, INSTR_RR_FF }, + { "cer", 0x39, INSTR_RR_FF }, + { "aer", 0x3a, INSTR_RR_FF }, + { "ser", 0x3b, INSTR_RR_FF }, + { "mder", 0x3c, INSTR_RR_FF }, + { "der", 0x3d, INSTR_RR_FF }, + { "aur", 0x3e, INSTR_RR_FF }, + { "sur", 0x3f, INSTR_RR_FF }, + { "sth", 0x40, INSTR_RX_RRRD }, + { "la", 0x41, INSTR_RX_RRRD }, + { "stc", 0x42, INSTR_RX_RRRD }, + { "ic", 0x43, INSTR_RX_RRRD }, + { "ex", 0x44, INSTR_RX_RRRD }, + { "bal", 0x45, INSTR_RX_RRRD }, + { "bct", 0x46, INSTR_RX_RRRD }, + { "bc", 0x47, INSTR_RX_URRD }, + { "lh", 0x48, INSTR_RX_RRRD }, + { "ch", 0x49, INSTR_RX_RRRD }, + { "ah", 0x4a, INSTR_RX_RRRD }, + { "sh", 0x4b, INSTR_RX_RRRD }, + { "mh", 0x4c, INSTR_RX_RRRD }, + { "bas", 0x4d, INSTR_RX_RRRD }, + { "cvd", 0x4e, INSTR_RX_RRRD }, + { "cvb", 0x4f, INSTR_RX_RRRD }, + { "st", 0x50, INSTR_RX_RRRD }, + { "lae", 0x51, INSTR_RX_RRRD }, + { "n", 0x54, INSTR_RX_RRRD }, + { "cl", 0x55, INSTR_RX_RRRD }, + { "o", 0x56, INSTR_RX_RRRD }, + { "x", 0x57, INSTR_RX_RRRD }, + { "l", 0x58, INSTR_RX_RRRD }, + { "c", 0x59, INSTR_RX_RRRD }, + { "a", 0x5a, INSTR_RX_RRRD }, + { "s", 0x5b, INSTR_RX_RRRD }, + { "m", 0x5c, INSTR_RX_RRRD }, + { "d", 0x5d, INSTR_RX_RRRD }, + { "al", 0x5e, INSTR_RX_RRRD }, + { "sl", 0x5f, INSTR_RX_RRRD }, + { "std", 0x60, INSTR_RX_FRRD }, + { "mxd", 0x67, INSTR_RX_FRRD }, + { "ld", 0x68, INSTR_RX_FRRD }, + { "cd", 0x69, INSTR_RX_FRRD }, + { "ad", 0x6a, INSTR_RX_FRRD }, + { "sd", 0x6b, INSTR_RX_FRRD }, + { "md", 0x6c, INSTR_RX_FRRD }, + { "dd", 0x6d, INSTR_RX_FRRD }, + { "aw", 0x6e, INSTR_RX_FRRD }, + { "sw", 0x6f, INSTR_RX_FRRD }, + { "ste", 0x70, INSTR_RX_FRRD }, + { "ms", 0x71, INSTR_RX_RRRD }, + { "le", 0x78, INSTR_RX_FRRD }, + { "ce", 0x79, INSTR_RX_FRRD }, + { "ae", 0x7a, INSTR_RX_FRRD }, + { "se", 0x7b, INSTR_RX_FRRD }, + { "mde", 0x7c, INSTR_RX_FRRD }, + { "de", 0x7d, INSTR_RX_FRRD }, + { "au", 0x7e, INSTR_RX_FRRD }, + { "su", 0x7f, INSTR_RX_FRRD }, + { "ssm", 0x80, INSTR_S_RD }, + { "lpsw", 0x82, INSTR_S_RD }, + { "diag", 0x83, INSTR_RS_RRRD }, + { "brxh", 0x84, INSTR_RSI_RRP }, + { "brxle", 0x85, INSTR_RSI_RRP }, + { "bxh", 0x86, INSTR_RS_RRRD }, + { "bxle", 0x87, INSTR_RS_RRRD }, + { "srl", 0x88, INSTR_RS_R0RD }, + { "sll", 0x89, INSTR_RS_R0RD }, + { "sra", 0x8a, INSTR_RS_R0RD }, + { "sla", 0x8b, INSTR_RS_R0RD }, + { "srdl", 0x8c, INSTR_RS_R0RD }, + { "sldl", 0x8d, INSTR_RS_R0RD }, + { "srda", 0x8e, INSTR_RS_R0RD }, + { "slda", 0x8f, INSTR_RS_R0RD }, + { "stm", 0x90, INSTR_RS_RRRD }, + { "tm", 0x91, INSTR_SI_URD }, + { "mvi", 0x92, INSTR_SI_URD }, + { "ts", 0x93, INSTR_S_RD }, + { "ni", 0x94, INSTR_SI_URD }, + { "cli", 0x95, INSTR_SI_URD }, + { "oi", 0x96, INSTR_SI_URD }, + { "xi", 0x97, INSTR_SI_URD }, + { "lm", 0x98, INSTR_RS_RRRD }, + { "trace", 0x99, INSTR_RS_RRRD }, + { "lam", 0x9a, INSTR_RS_AARD }, + { "stam", 0x9b, INSTR_RS_AARD }, + { "mvcle", 0xa8, INSTR_RS_RRRD }, + { "clcle", 0xa9, INSTR_RS_RRRD }, + { "stnsm", 0xac, INSTR_SI_URD }, + { "stosm", 0xad, INSTR_SI_URD }, + { "sigp", 0xae, INSTR_RS_RRRD }, + { "mc", 0xaf, INSTR_SI_URD }, + { "lra", 0xb1, INSTR_RX_RRRD }, + { "stctl", 0xb6, INSTR_RS_CCRD }, + { "lctl", 0xb7, INSTR_RS_CCRD }, + { "cs", 0xba, INSTR_RS_RRRD }, + { "cds", 0xbb, INSTR_RS_RRRD }, + { "clm", 0xbd, INSTR_RS_RURD }, + { "stcm", 0xbe, INSTR_RS_RURD }, + { "icm", 0xbf, INSTR_RS_RURD }, + { "mvn", 0xd1, INSTR_SS_L0RDRD }, + { "mvc", 0xd2, INSTR_SS_L0RDRD }, + { "mvz", 0xd3, INSTR_SS_L0RDRD }, + { "nc", 0xd4, INSTR_SS_L0RDRD }, + { "clc", 0xd5, INSTR_SS_L0RDRD }, + { "oc", 0xd6, INSTR_SS_L0RDRD }, + { "xc", 0xd7, INSTR_SS_L0RDRD }, + { "mvck", 0xd9, INSTR_SS_RRRDRD }, + { "mvcp", 0xda, INSTR_SS_RRRDRD }, + { "mvcs", 0xdb, INSTR_SS_RRRDRD }, + { "tr", 0xdc, INSTR_SS_L0RDRD }, + { "trt", 0xdd, INSTR_SS_L0RDRD }, + { "ed", 0xde, INSTR_SS_L0RDRD }, + { "edmk", 0xdf, INSTR_SS_L0RDRD }, + { "pku", 0xe1, INSTR_SS_L0RDRD }, + { "unpku", 0xe2, INSTR_SS_L0RDRD }, + { "mvcin", 0xe8, INSTR_SS_L0RDRD }, + { "pka", 0xe9, INSTR_SS_L0RDRD }, + { "unpka", 0xea, INSTR_SS_L0RDRD }, + { "plo", 0xee, INSTR_SS_RRRDRD2 }, + { "srp", 0xf0, INSTR_SS_LIRDRD }, + { "mvo", 0xf1, INSTR_SS_LLRDRD }, + { "pack", 0xf2, INSTR_SS_LLRDRD }, + { "unpk", 0xf3, INSTR_SS_LLRDRD }, + { "zap", 0xf8, INSTR_SS_LLRDRD }, + { "cp", 0xf9, INSTR_SS_LLRDRD }, + { "ap", 0xfa, INSTR_SS_LLRDRD }, + { "sp", 0xfb, INSTR_SS_LLRDRD }, + { "mp", 0xfc, INSTR_SS_LLRDRD }, + { "dp", 0xfd, INSTR_SS_LLRDRD }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_01[] = { + { "ptff", 0x04, INSTR_E }, + { "pfpo", 0x0a, INSTR_E }, + { "sam64", 0x0e, INSTR_E }, + { "pr", 0x01, INSTR_E }, + { "upt", 0x02, INSTR_E }, + { "sckpf", 0x07, INSTR_E }, + { "tam", 0x0b, INSTR_E }, + { "sam24", 0x0c, INSTR_E }, + { "sam31", 0x0d, INSTR_E }, + { "trap2", 0xff, INSTR_E }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_a5[] = { + { "iihh", 0x00, INSTR_RI_RU }, + { "iihl", 0x01, INSTR_RI_RU }, + { "iilh", 0x02, INSTR_RI_RU }, + { "iill", 0x03, INSTR_RI_RU }, + { "nihh", 0x04, INSTR_RI_RU }, + { "nihl", 0x05, INSTR_RI_RU }, + { "nilh", 0x06, INSTR_RI_RU }, + { "nill", 0x07, INSTR_RI_RU }, + { "oihh", 0x08, INSTR_RI_RU }, + { "oihl", 0x09, INSTR_RI_RU }, + { "oilh", 0x0a, INSTR_RI_RU }, + { "oill", 0x0b, INSTR_RI_RU }, + { "llihh", 0x0c, INSTR_RI_RU }, + { "llihl", 0x0d, INSTR_RI_RU }, + { "llilh", 0x0e, INSTR_RI_RU }, + { "llill", 0x0f, INSTR_RI_RU }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_a7[] = { + { "tmhh", 0x02, INSTR_RI_RU }, + { "tmhl", 0x03, INSTR_RI_RU }, + { "brctg", 0x07, INSTR_RI_RP }, + { "lghi", 0x09, INSTR_RI_RI }, + { "aghi", 0x0b, INSTR_RI_RI }, + { "mghi", 0x0d, INSTR_RI_RI }, + { "cghi", 0x0f, INSTR_RI_RI }, + { "tmlh", 0x00, INSTR_RI_RU }, + { "tmll", 0x01, INSTR_RI_RU }, + { "brc", 0x04, INSTR_RI_UP }, + { "bras", 0x05, INSTR_RI_RP }, + { "brct", 0x06, INSTR_RI_RP }, + { "lhi", 0x08, INSTR_RI_RI }, + { "ahi", 0x0a, INSTR_RI_RI }, + { "mhi", 0x0c, INSTR_RI_RI }, + { "chi", 0x0e, INSTR_RI_RI }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_aa[] = { + { { 0, LONG_INSN_RINEXT }, 0x00, INSTR_RI_RI }, + { "rion", 0x01, INSTR_RI_RI }, + { "tric", 0x02, INSTR_RI_RI }, + { "rioff", 0x03, INSTR_RI_RI }, + { { 0, LONG_INSN_RIEMIT }, 0x04, INSTR_RI_RI }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_b2[] = { + { "stckf", 0x7c, INSTR_S_RD }, + { "lpp", 0x80, INSTR_S_RD }, + { "lcctl", 0x84, INSTR_S_RD }, + { "lpctl", 0x85, INSTR_S_RD }, + { "qsi", 0x86, INSTR_S_RD }, + { "lsctl", 0x87, INSTR_S_RD }, + { "qctri", 0x8e, INSTR_S_RD }, + { "stfle", 0xb0, INSTR_S_RD }, + { "lpswe", 0xb2, INSTR_S_RD }, + { "srnmb", 0xb8, INSTR_S_RD }, + { "srnmt", 0xb9, INSTR_S_RD }, + { "lfas", 0xbd, INSTR_S_RD }, + { "scctr", 0xe0, INSTR_RRE_RR }, + { "spctr", 0xe1, INSTR_RRE_RR }, + { "ecctr", 0xe4, INSTR_RRE_RR }, + { "epctr", 0xe5, INSTR_RRE_RR }, + { "ppa", 0xe8, INSTR_RRF_U0RR }, + { "etnd", 0xec, INSTR_RRE_R0 }, + { "ecpga", 0xed, INSTR_RRE_RR }, + { "tend", 0xf8, INSTR_S_00 }, + { "niai", 0xfa, INSTR_IE_UU }, + { { 0, LONG_INSN_TABORT }, 0xfc, INSTR_S_RD }, + { "stidp", 0x02, INSTR_S_RD }, + { "sck", 0x04, INSTR_S_RD }, + { "stck", 0x05, INSTR_S_RD }, + { "sckc", 0x06, INSTR_S_RD }, + { "stckc", 0x07, INSTR_S_RD }, + { "spt", 0x08, INSTR_S_RD }, + { "stpt", 0x09, INSTR_S_RD }, + { "spka", 0x0a, INSTR_S_RD }, + { "ipk", 0x0b, INSTR_S_00 }, + { "ptlb", 0x0d, INSTR_S_00 }, + { "spx", 0x10, INSTR_S_RD }, + { "stpx", 0x11, INSTR_S_RD }, + { "stap", 0x12, INSTR_S_RD }, + { "sie", 0x14, INSTR_S_RD }, + { "pc", 0x18, INSTR_S_RD }, + { "sac", 0x19, INSTR_S_RD }, + { "cfc", 0x1a, INSTR_S_RD }, + { "servc", 0x20, INSTR_RRE_RR }, + { "ipte", 0x21, INSTR_RRE_RR }, + { "ipm", 0x22, INSTR_RRE_R0 }, + { "ivsk", 0x23, INSTR_RRE_RR }, + { "iac", 0x24, INSTR_RRE_R0 }, + { "ssar", 0x25, INSTR_RRE_R0 }, + { "epar", 0x26, INSTR_RRE_R0 }, + { "esar", 0x27, INSTR_RRE_R0 }, + { "pt", 0x28, INSTR_RRE_RR }, + { "iske", 0x29, INSTR_RRE_RR }, + { "rrbe", 0x2a, INSTR_RRE_RR }, + { "sske", 0x2b, INSTR_RRF_M0RR }, + { "tb", 0x2c, INSTR_RRE_0R }, + { "dxr", 0x2d, INSTR_RRE_FF }, + { "pgin", 0x2e, INSTR_RRE_RR }, + { "pgout", 0x2f, INSTR_RRE_RR }, + { "csch", 0x30, INSTR_S_00 }, + { "hsch", 0x31, INSTR_S_00 }, + { "msch", 0x32, INSTR_S_RD }, + { "ssch", 0x33, INSTR_S_RD }, + { "stsch", 0x34, INSTR_S_RD }, + { "tsch", 0x35, INSTR_S_RD }, + { "tpi", 0x36, INSTR_S_RD }, + { "sal", 0x37, INSTR_S_00 }, + { "rsch", 0x38, INSTR_S_00 }, + { "stcrw", 0x39, INSTR_S_RD }, + { "stcps", 0x3a, INSTR_S_RD }, + { "rchp", 0x3b, INSTR_S_00 }, + { "schm", 0x3c, INSTR_S_00 }, + { "bakr", 0x40, INSTR_RRE_RR }, + { "cksm", 0x41, INSTR_RRE_RR }, + { "sqdr", 0x44, INSTR_RRE_FF }, + { "sqer", 0x45, INSTR_RRE_FF }, + { "stura", 0x46, INSTR_RRE_RR }, + { "msta", 0x47, INSTR_RRE_R0 }, + { "palb", 0x48, INSTR_RRE_00 }, + { "ereg", 0x49, INSTR_RRE_RR }, + { "esta", 0x4a, INSTR_RRE_RR }, + { "lura", 0x4b, INSTR_RRE_RR }, + { "tar", 0x4c, INSTR_RRE_AR }, + { "cpya", 0x4d, INSTR_RRE_AA }, + { "sar", 0x4e, INSTR_RRE_AR }, + { "ear", 0x4f, INSTR_RRE_RA }, + { "csp", 0x50, INSTR_RRE_RR }, + { "msr", 0x52, INSTR_RRE_RR }, + { "mvpg", 0x54, INSTR_RRE_RR }, + { "mvst", 0x55, INSTR_RRE_RR }, + { "cuse", 0x57, INSTR_RRE_RR }, + { "bsg", 0x58, INSTR_RRE_RR }, + { "bsa", 0x5a, INSTR_RRE_RR }, + { "clst", 0x5d, INSTR_RRE_RR }, + { "srst", 0x5e, INSTR_RRE_RR }, + { "cmpsc", 0x63, INSTR_RRE_RR }, + { "siga", 0x74, INSTR_S_RD }, + { "xsch", 0x76, INSTR_S_00 }, + { "rp", 0x77, INSTR_S_RD }, + { "stcke", 0x78, INSTR_S_RD }, + { "sacf", 0x79, INSTR_S_RD }, + { "stsi", 0x7d, INSTR_S_RD }, + { "srnm", 0x99, INSTR_S_RD }, + { "stfpc", 0x9c, INSTR_S_RD }, + { "lfpc", 0x9d, INSTR_S_RD }, + { "tre", 0xa5, INSTR_RRE_RR }, + { "cuutf", 0xa6, INSTR_RRF_M0RR }, + { "cutfu", 0xa7, INSTR_RRF_M0RR }, + { "stfl", 0xb1, INSTR_S_RD }, + { "trap4", 0xff, INSTR_S_RD }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_b3[] = { + { "maylr", 0x38, INSTR_RRF_F0FF }, + { "mylr", 0x39, INSTR_RRF_F0FF }, + { "mayr", 0x3a, INSTR_RRF_F0FF }, + { "myr", 0x3b, INSTR_RRF_F0FF }, + { "mayhr", 0x3c, INSTR_RRF_F0FF }, + { "myhr", 0x3d, INSTR_RRF_F0FF }, + { "lpdfr", 0x70, INSTR_RRE_FF }, + { "lndfr", 0x71, INSTR_RRE_FF }, + { "cpsdr", 0x72, INSTR_RRF_F0FF2 }, + { "lcdfr", 0x73, INSTR_RRE_FF }, + { "sfasr", 0x85, INSTR_RRE_R0 }, + { { 0, LONG_INSN_CELFBR }, 0x90, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CDLFBR }, 0x91, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CXLFBR }, 0x92, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CEFBRA }, 0x94, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CDFBRA }, 0x95, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CXFBRA }, 0x96, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CFEBRA }, 0x98, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CFDBRA }, 0x99, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CFXBRA }, 0x9a, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CLFEBR }, 0x9c, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CLFDBR }, 0x9d, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CLFXBR }, 0x9e, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CELGBR }, 0xa0, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CDLGBR }, 0xa1, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CXLGBR }, 0xa2, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CEGBRA }, 0xa4, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CDGBRA }, 0xa5, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CXGBRA }, 0xa6, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CGEBRA }, 0xa8, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CGDBRA }, 0xa9, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CGXBRA }, 0xaa, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CLGEBR }, 0xac, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CLGDBR }, 0xad, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CLGXBR }, 0xae, INSTR_RRF_UUFR }, + { "ldgr", 0xc1, INSTR_RRE_FR }, + { "cegr", 0xc4, INSTR_RRE_FR }, + { "cdgr", 0xc5, INSTR_RRE_FR }, + { "cxgr", 0xc6, INSTR_RRE_FR }, + { "cger", 0xc8, INSTR_RRF_U0RF }, + { "cgdr", 0xc9, INSTR_RRF_U0RF }, + { "cgxr", 0xca, INSTR_RRF_U0RF }, + { "lgdr", 0xcd, INSTR_RRE_RF }, + { "mdtra", 0xd0, INSTR_RRF_FUFF2 }, + { "ddtra", 0xd1, INSTR_RRF_FUFF2 }, + { "adtra", 0xd2, INSTR_RRF_FUFF2 }, + { "sdtra", 0xd3, INSTR_RRF_FUFF2 }, + { "ldetr", 0xd4, INSTR_RRF_0UFF }, + { "ledtr", 0xd5, INSTR_RRF_UUFF }, + { "ltdtr", 0xd6, INSTR_RRE_FF }, + { "fidtr", 0xd7, INSTR_RRF_UUFF }, + { "mxtra", 0xd8, INSTR_RRF_FUFF2 }, + { "dxtra", 0xd9, INSTR_RRF_FUFF2 }, + { "axtra", 0xda, INSTR_RRF_FUFF2 }, + { "sxtra", 0xdb, INSTR_RRF_FUFF2 }, + { "lxdtr", 0xdc, INSTR_RRF_0UFF }, + { "ldxtr", 0xdd, INSTR_RRF_UUFF }, + { "ltxtr", 0xde, INSTR_RRE_FF }, + { "fixtr", 0xdf, INSTR_RRF_UUFF }, + { "kdtr", 0xe0, INSTR_RRE_FF }, + { { 0, LONG_INSN_CGDTRA }, 0xe1, INSTR_RRF_UURF }, + { "cudtr", 0xe2, INSTR_RRE_RF }, + { "csdtr", 0xe3, INSTR_RRE_RF }, + { "cdtr", 0xe4, INSTR_RRE_FF }, + { "eedtr", 0xe5, INSTR_RRE_RF }, + { "esdtr", 0xe7, INSTR_RRE_RF }, + { "kxtr", 0xe8, INSTR_RRE_FF }, + { { 0, LONG_INSN_CGXTRA }, 0xe9, INSTR_RRF_UUFR }, + { "cuxtr", 0xea, INSTR_RRE_RF }, + { "csxtr", 0xeb, INSTR_RRE_RF }, + { "cxtr", 0xec, INSTR_RRE_FF }, + { "eextr", 0xed, INSTR_RRE_RF }, + { "esxtr", 0xef, INSTR_RRE_RF }, + { { 0, LONG_INSN_CDGTRA }, 0xf1, INSTR_RRF_UUFR }, + { "cdutr", 0xf2, INSTR_RRE_FR }, + { "cdstr", 0xf3, INSTR_RRE_FR }, + { "cedtr", 0xf4, INSTR_RRE_FF }, + { "qadtr", 0xf5, INSTR_RRF_FUFF }, + { "iedtr", 0xf6, INSTR_RRF_F0FR }, + { "rrdtr", 0xf7, INSTR_RRF_FFRU }, + { { 0, LONG_INSN_CXGTRA }, 0xf9, INSTR_RRF_UURF }, + { "cxutr", 0xfa, INSTR_RRE_FR }, + { "cxstr", 0xfb, INSTR_RRE_FR }, + { "cextr", 0xfc, INSTR_RRE_FF }, + { "qaxtr", 0xfd, INSTR_RRF_FUFF }, + { "iextr", 0xfe, INSTR_RRF_F0FR }, + { "rrxtr", 0xff, INSTR_RRF_FFRU }, + { "lpebr", 0x00, INSTR_RRE_FF }, + { "lnebr", 0x01, INSTR_RRE_FF }, + { "ltebr", 0x02, INSTR_RRE_FF }, + { "lcebr", 0x03, INSTR_RRE_FF }, + { "ldebr", 0x04, INSTR_RRE_FF }, + { "lxdbr", 0x05, INSTR_RRE_FF }, + { "lxebr", 0x06, INSTR_RRE_FF }, + { "mxdbr", 0x07, INSTR_RRE_FF }, + { "kebr", 0x08, INSTR_RRE_FF }, + { "cebr", 0x09, INSTR_RRE_FF }, + { "aebr", 0x0a, INSTR_RRE_FF }, + { "sebr", 0x0b, INSTR_RRE_FF }, + { "mdebr", 0x0c, INSTR_RRE_FF }, + { "debr", 0x0d, INSTR_RRE_FF }, + { "maebr", 0x0e, INSTR_RRF_F0FF }, + { "msebr", 0x0f, INSTR_RRF_F0FF }, + { "lpdbr", 0x10, INSTR_RRE_FF }, + { "lndbr", 0x11, INSTR_RRE_FF }, + { "ltdbr", 0x12, INSTR_RRE_FF }, + { "lcdbr", 0x13, INSTR_RRE_FF }, + { "sqebr", 0x14, INSTR_RRE_FF }, + { "sqdbr", 0x15, INSTR_RRE_FF }, + { "sqxbr", 0x16, INSTR_RRE_FF }, + { "meebr", 0x17, INSTR_RRE_FF }, + { "kdbr", 0x18, INSTR_RRE_FF }, + { "cdbr", 0x19, INSTR_RRE_FF }, + { "adbr", 0x1a, INSTR_RRE_FF }, + { "sdbr", 0x1b, INSTR_RRE_FF }, + { "mdbr", 0x1c, INSTR_RRE_FF }, + { "ddbr", 0x1d, INSTR_RRE_FF }, + { "madbr", 0x1e, INSTR_RRF_F0FF }, + { "msdbr", 0x1f, INSTR_RRF_F0FF }, + { "lder", 0x24, INSTR_RRE_FF }, + { "lxdr", 0x25, INSTR_RRE_FF }, + { "lxer", 0x26, INSTR_RRE_FF }, + { "maer", 0x2e, INSTR_RRF_F0FF }, + { "mser", 0x2f, INSTR_RRF_F0FF }, + { "sqxr", 0x36, INSTR_RRE_FF }, + { "meer", 0x37, INSTR_RRE_FF }, + { "madr", 0x3e, INSTR_RRF_F0FF }, + { "msdr", 0x3f, INSTR_RRF_F0FF }, + { "lpxbr", 0x40, INSTR_RRE_FF }, + { "lnxbr", 0x41, INSTR_RRE_FF }, + { "ltxbr", 0x42, INSTR_RRE_FF }, + { "lcxbr", 0x43, INSTR_RRE_FF }, + { { 0, LONG_INSN_LEDBRA }, 0x44, INSTR_RRF_UUFF }, + { { 0, LONG_INSN_LDXBRA }, 0x45, INSTR_RRF_UUFF }, + { { 0, LONG_INSN_LEXBRA }, 0x46, INSTR_RRF_UUFF }, + { { 0, LONG_INSN_FIXBRA }, 0x47, INSTR_RRF_UUFF }, + { "kxbr", 0x48, INSTR_RRE_FF }, + { "cxbr", 0x49, INSTR_RRE_FF }, + { "axbr", 0x4a, INSTR_RRE_FF }, + { "sxbr", 0x4b, INSTR_RRE_FF }, + { "mxbr", 0x4c, INSTR_RRE_FF }, + { "dxbr", 0x4d, INSTR_RRE_FF }, + { "tbedr", 0x50, INSTR_RRF_U0FF }, + { "tbdr", 0x51, INSTR_RRF_U0FF }, + { "diebr", 0x53, INSTR_RRF_FUFF }, + { { 0, LONG_INSN_FIEBRA }, 0x57, INSTR_RRF_UUFF }, + { "thder", 0x58, INSTR_RRE_FF }, + { "thdr", 0x59, INSTR_RRE_FF }, + { "didbr", 0x5b, INSTR_RRF_FUFF }, + { { 0, LONG_INSN_FIDBRA }, 0x5f, INSTR_RRF_UUFF }, + { "lpxr", 0x60, INSTR_RRE_FF }, + { "lnxr", 0x61, INSTR_RRE_FF }, + { "ltxr", 0x62, INSTR_RRE_FF }, + { "lcxr", 0x63, INSTR_RRE_FF }, + { "lxr", 0x65, INSTR_RRE_FF }, + { "lexr", 0x66, INSTR_RRE_FF }, + { "fixr", 0x67, INSTR_RRE_FF }, + { "cxr", 0x69, INSTR_RRE_FF }, + { "lzer", 0x74, INSTR_RRE_F0 }, + { "lzdr", 0x75, INSTR_RRE_F0 }, + { "lzxr", 0x76, INSTR_RRE_F0 }, + { "fier", 0x77, INSTR_RRE_FF }, + { "fidr", 0x7f, INSTR_RRE_FF }, + { "sfpc", 0x84, INSTR_RRE_RR_OPT }, + { "efpc", 0x8c, INSTR_RRE_RR_OPT }, + { "cefbr", 0x94, INSTR_RRE_RF }, + { "cdfbr", 0x95, INSTR_RRE_RF }, + { "cxfbr", 0x96, INSTR_RRE_RF }, + { "cfebr", 0x98, INSTR_RRF_U0RF }, + { "cfdbr", 0x99, INSTR_RRF_U0RF }, + { "cfxbr", 0x9a, INSTR_RRF_U0RF }, + { "cefr", 0xb4, INSTR_RRE_FR }, + { "cdfr", 0xb5, INSTR_RRE_FR }, + { "cxfr", 0xb6, INSTR_RRE_FR }, + { "cfer", 0xb8, INSTR_RRF_U0RF }, + { "cfdr", 0xb9, INSTR_RRF_U0RF }, + { "cfxr", 0xba, INSTR_RRF_U0RF }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_b9[] = { + { "lpgr", 0x00, INSTR_RRE_RR }, + { "lngr", 0x01, INSTR_RRE_RR }, + { "ltgr", 0x02, INSTR_RRE_RR }, + { "lcgr", 0x03, INSTR_RRE_RR }, + { "lgr", 0x04, INSTR_RRE_RR }, + { "lurag", 0x05, INSTR_RRE_RR }, + { "lgbr", 0x06, INSTR_RRE_RR }, + { "lghr", 0x07, INSTR_RRE_RR }, + { "agr", 0x08, INSTR_RRE_RR }, + { "sgr", 0x09, INSTR_RRE_RR }, + { "algr", 0x0a, INSTR_RRE_RR }, + { "slgr", 0x0b, INSTR_RRE_RR }, + { "msgr", 0x0c, INSTR_RRE_RR }, + { "dsgr", 0x0d, INSTR_RRE_RR }, + { "eregg", 0x0e, INSTR_RRE_RR }, + { "lrvgr", 0x0f, INSTR_RRE_RR }, + { "lpgfr", 0x10, INSTR_RRE_RR }, + { "lngfr", 0x11, INSTR_RRE_RR }, + { "ltgfr", 0x12, INSTR_RRE_RR }, + { "lcgfr", 0x13, INSTR_RRE_RR }, + { "lgfr", 0x14, INSTR_RRE_RR }, + { "llgfr", 0x16, INSTR_RRE_RR }, + { "llgtr", 0x17, INSTR_RRE_RR }, + { "agfr", 0x18, INSTR_RRE_RR }, + { "sgfr", 0x19, INSTR_RRE_RR }, + { "algfr", 0x1a, INSTR_RRE_RR }, + { "slgfr", 0x1b, INSTR_RRE_RR }, + { "msgfr", 0x1c, INSTR_RRE_RR }, + { "dsgfr", 0x1d, INSTR_RRE_RR }, + { "cgr", 0x20, INSTR_RRE_RR }, + { "clgr", 0x21, INSTR_RRE_RR }, + { "sturg", 0x25, INSTR_RRE_RR }, + { "lbr", 0x26, INSTR_RRE_RR }, + { "lhr", 0x27, INSTR_RRE_RR }, + { "cgfr", 0x30, INSTR_RRE_RR }, + { "clgfr", 0x31, INSTR_RRE_RR }, + { "cfdtr", 0x41, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CLGDTR }, 0x42, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CLFDTR }, 0x43, INSTR_RRF_UURF }, + { "bctgr", 0x46, INSTR_RRE_RR }, + { "cfxtr", 0x49, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CLGXTR }, 0x4a, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CLFXTR }, 0x4b, INSTR_RRF_UUFR }, + { "cdftr", 0x51, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CDLGTR }, 0x52, INSTR_RRF_UUFR }, + { { 0, LONG_INSN_CDLFTR }, 0x53, INSTR_RRF_UUFR }, + { "cxftr", 0x59, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CXLGTR }, 0x5a, INSTR_RRF_UURF }, + { { 0, LONG_INSN_CXLFTR }, 0x5b, INSTR_RRF_UUFR }, + { "cgrt", 0x60, INSTR_RRF_U0RR }, + { "clgrt", 0x61, INSTR_RRF_U0RR }, + { "crt", 0x72, INSTR_RRF_U0RR }, + { "clrt", 0x73, INSTR_RRF_U0RR }, + { "ngr", 0x80, INSTR_RRE_RR }, + { "ogr", 0x81, INSTR_RRE_RR }, + { "xgr", 0x82, INSTR_RRE_RR }, + { "flogr", 0x83, INSTR_RRE_RR }, + { "llgcr", 0x84, INSTR_RRE_RR }, + { "llghr", 0x85, INSTR_RRE_RR }, + { "mlgr", 0x86, INSTR_RRE_RR }, + { "dlgr", 0x87, INSTR_RRE_RR }, + { "alcgr", 0x88, INSTR_RRE_RR }, + { "slbgr", 0x89, INSTR_RRE_RR }, + { "cspg", 0x8a, INSTR_RRE_RR }, + { "idte", 0x8e, INSTR_RRF_R0RR }, + { "crdte", 0x8f, INSTR_RRF_RMRR }, + { "llcr", 0x94, INSTR_RRE_RR }, + { "llhr", 0x95, INSTR_RRE_RR }, + { "esea", 0x9d, INSTR_RRE_R0 }, + { "ptf", 0xa2, INSTR_RRE_R0 }, + { "lptea", 0xaa, INSTR_RRF_RURR }, + { "rrbm", 0xae, INSTR_RRE_RR }, + { "pfmf", 0xaf, INSTR_RRE_RR }, + { "cu14", 0xb0, INSTR_RRF_M0RR }, + { "cu24", 0xb1, INSTR_RRF_M0RR }, + { "cu41", 0xb2, INSTR_RRE_RR }, + { "cu42", 0xb3, INSTR_RRE_RR }, + { "trtre", 0xbd, INSTR_RRF_M0RR }, + { "srstu", 0xbe, INSTR_RRE_RR }, + { "trte", 0xbf, INSTR_RRF_M0RR }, + { "ahhhr", 0xc8, INSTR_RRF_R0RR2 }, + { "shhhr", 0xc9, INSTR_RRF_R0RR2 }, + { { 0, LONG_INSN_ALHHHR }, 0xca, INSTR_RRF_R0RR2 }, + { { 0, LONG_INSN_SLHHHR }, 0xcb, INSTR_RRF_R0RR2 }, + { "chhr", 0xcd, INSTR_RRE_RR }, + { "clhhr", 0xcf, INSTR_RRE_RR }, + { { 0, LONG_INSN_PCISTG }, 0xd0, INSTR_RRE_RR }, + { "pcilg", 0xd2, INSTR_RRE_RR }, + { "rpcit", 0xd3, INSTR_RRE_RR }, + { "ahhlr", 0xd8, INSTR_RRF_R0RR2 }, + { "shhlr", 0xd9, INSTR_RRF_R0RR2 }, + { { 0, LONG_INSN_ALHHLR }, 0xda, INSTR_RRF_R0RR2 }, + { { 0, LONG_INSN_SLHHLR }, 0xdb, INSTR_RRF_R0RR2 }, + { "chlr", 0xdd, INSTR_RRE_RR }, + { "clhlr", 0xdf, INSTR_RRE_RR }, + { { 0, LONG_INSN_POPCNT }, 0xe1, INSTR_RRE_RR }, + { "locgr", 0xe2, INSTR_RRF_M0RR }, + { "ngrk", 0xe4, INSTR_RRF_R0RR2 }, + { "ogrk", 0xe6, INSTR_RRF_R0RR2 }, + { "xgrk", 0xe7, INSTR_RRF_R0RR2 }, + { "agrk", 0xe8, INSTR_RRF_R0RR2 }, + { "sgrk", 0xe9, INSTR_RRF_R0RR2 }, + { "algrk", 0xea, INSTR_RRF_R0RR2 }, + { "slgrk", 0xeb, INSTR_RRF_R0RR2 }, + { "locr", 0xf2, INSTR_RRF_M0RR }, + { "nrk", 0xf4, INSTR_RRF_R0RR2 }, + { "ork", 0xf6, INSTR_RRF_R0RR2 }, + { "xrk", 0xf7, INSTR_RRF_R0RR2 }, + { "ark", 0xf8, INSTR_RRF_R0RR2 }, + { "srk", 0xf9, INSTR_RRF_R0RR2 }, + { "alrk", 0xfa, INSTR_RRF_R0RR2 }, + { "slrk", 0xfb, INSTR_RRF_R0RR2 }, + { "kmac", 0x1e, INSTR_RRE_RR }, + { "lrvr", 0x1f, INSTR_RRE_RR }, + { "km", 0x2e, INSTR_RRE_RR }, + { "kmc", 0x2f, INSTR_RRE_RR }, + { "kimd", 0x3e, INSTR_RRE_RR }, + { "klmd", 0x3f, INSTR_RRE_RR }, + { "epsw", 0x8d, INSTR_RRE_RR }, + { "trtt", 0x90, INSTR_RRF_M0RR }, + { "trto", 0x91, INSTR_RRF_M0RR }, + { "trot", 0x92, INSTR_RRF_M0RR }, + { "troo", 0x93, INSTR_RRF_M0RR }, + { "mlr", 0x96, INSTR_RRE_RR }, + { "dlr", 0x97, INSTR_RRE_RR }, + { "alcr", 0x98, INSTR_RRE_RR }, + { "slbr", 0x99, INSTR_RRE_RR }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_c0[] = { + { "lgfi", 0x01, INSTR_RIL_RI }, + { "xihf", 0x06, INSTR_RIL_RU }, + { "xilf", 0x07, INSTR_RIL_RU }, + { "iihf", 0x08, INSTR_RIL_RU }, + { "iilf", 0x09, INSTR_RIL_RU }, + { "nihf", 0x0a, INSTR_RIL_RU }, + { "nilf", 0x0b, INSTR_RIL_RU }, + { "oihf", 0x0c, INSTR_RIL_RU }, + { "oilf", 0x0d, INSTR_RIL_RU }, + { "llihf", 0x0e, INSTR_RIL_RU }, + { "llilf", 0x0f, INSTR_RIL_RU }, + { "larl", 0x00, INSTR_RIL_RP }, + { "brcl", 0x04, INSTR_RIL_UP }, + { "brasl", 0x05, INSTR_RIL_RP }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_c2[] = { + { "msgfi", 0x00, INSTR_RIL_RI }, + { "msfi", 0x01, INSTR_RIL_RI }, + { "slgfi", 0x04, INSTR_RIL_RU }, + { "slfi", 0x05, INSTR_RIL_RU }, + { "agfi", 0x08, INSTR_RIL_RI }, + { "afi", 0x09, INSTR_RIL_RI }, + { "algfi", 0x0a, INSTR_RIL_RU }, + { "alfi", 0x0b, INSTR_RIL_RU }, + { "cgfi", 0x0c, INSTR_RIL_RI }, + { "cfi", 0x0d, INSTR_RIL_RI }, + { "clgfi", 0x0e, INSTR_RIL_RU }, + { "clfi", 0x0f, INSTR_RIL_RU }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_c4[] = { + { "llhrl", 0x02, INSTR_RIL_RP }, + { "lghrl", 0x04, INSTR_RIL_RP }, + { "lhrl", 0x05, INSTR_RIL_RP }, + { { 0, LONG_INSN_LLGHRL }, 0x06, INSTR_RIL_RP }, + { "sthrl", 0x07, INSTR_RIL_RP }, + { "lgrl", 0x08, INSTR_RIL_RP }, + { "stgrl", 0x0b, INSTR_RIL_RP }, + { "lgfrl", 0x0c, INSTR_RIL_RP }, + { "lrl", 0x0d, INSTR_RIL_RP }, + { { 0, LONG_INSN_LLGFRL }, 0x0e, INSTR_RIL_RP }, + { "strl", 0x0f, INSTR_RIL_RP }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_c6[] = { + { "exrl", 0x00, INSTR_RIL_RP }, + { "pfdrl", 0x02, INSTR_RIL_UP }, + { "cghrl", 0x04, INSTR_RIL_RP }, + { "chrl", 0x05, INSTR_RIL_RP }, + { { 0, LONG_INSN_CLGHRL }, 0x06, INSTR_RIL_RP }, + { "clhrl", 0x07, INSTR_RIL_RP }, + { "cgrl", 0x08, INSTR_RIL_RP }, + { "clgrl", 0x0a, INSTR_RIL_RP }, + { "cgfrl", 0x0c, INSTR_RIL_RP }, + { "crl", 0x0d, INSTR_RIL_RP }, + { { 0, LONG_INSN_CLGFRL }, 0x0e, INSTR_RIL_RP }, + { "clrl", 0x0f, INSTR_RIL_RP }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_c8[] = { + { "mvcos", 0x00, INSTR_SSF_RRDRD }, + { "ectg", 0x01, INSTR_SSF_RRDRD }, + { "csst", 0x02, INSTR_SSF_RRDRD }, + { "lpd", 0x04, INSTR_SSF_RRDRD2 }, + { "lpdg", 0x05, INSTR_SSF_RRDRD2 }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_cc[] = { + { "brcth", 0x06, INSTR_RIL_RP }, + { "aih", 0x08, INSTR_RIL_RI }, + { "alsih", 0x0a, INSTR_RIL_RI }, + { { 0, LONG_INSN_ALSIHN }, 0x0b, INSTR_RIL_RI }, + { "cih", 0x0d, INSTR_RIL_RI }, + { "clih", 0x0f, INSTR_RIL_RI }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_e3[] = { + { "ltg", 0x02, INSTR_RXY_RRRD }, + { "lrag", 0x03, INSTR_RXY_RRRD }, + { "lg", 0x04, INSTR_RXY_RRRD }, + { "cvby", 0x06, INSTR_RXY_RRRD }, + { "ag", 0x08, INSTR_RXY_RRRD }, + { "sg", 0x09, INSTR_RXY_RRRD }, + { "alg", 0x0a, INSTR_RXY_RRRD }, + { "slg", 0x0b, INSTR_RXY_RRRD }, + { "msg", 0x0c, INSTR_RXY_RRRD }, + { "dsg", 0x0d, INSTR_RXY_RRRD }, + { "cvbg", 0x0e, INSTR_RXY_RRRD }, + { "lrvg", 0x0f, INSTR_RXY_RRRD }, + { "lt", 0x12, INSTR_RXY_RRRD }, + { "lray", 0x13, INSTR_RXY_RRRD }, + { "lgf", 0x14, INSTR_RXY_RRRD }, + { "lgh", 0x15, INSTR_RXY_RRRD }, + { "llgf", 0x16, INSTR_RXY_RRRD }, + { "llgt", 0x17, INSTR_RXY_RRRD }, + { "agf", 0x18, INSTR_RXY_RRRD }, + { "sgf", 0x19, INSTR_RXY_RRRD }, + { "algf", 0x1a, INSTR_RXY_RRRD }, + { "slgf", 0x1b, INSTR_RXY_RRRD }, + { "msgf", 0x1c, INSTR_RXY_RRRD }, + { "dsgf", 0x1d, INSTR_RXY_RRRD }, + { "cg", 0x20, INSTR_RXY_RRRD }, + { "clg", 0x21, INSTR_RXY_RRRD }, + { "stg", 0x24, INSTR_RXY_RRRD }, + { "ntstg", 0x25, INSTR_RXY_RRRD }, + { "cvdy", 0x26, INSTR_RXY_RRRD }, + { "cvdg", 0x2e, INSTR_RXY_RRRD }, + { "strvg", 0x2f, INSTR_RXY_RRRD }, + { "cgf", 0x30, INSTR_RXY_RRRD }, + { "clgf", 0x31, INSTR_RXY_RRRD }, + { "ltgf", 0x32, INSTR_RXY_RRRD }, + { "cgh", 0x34, INSTR_RXY_RRRD }, + { "pfd", 0x36, INSTR_RXY_URRD }, + { "strvh", 0x3f, INSTR_RXY_RRRD }, + { "bctg", 0x46, INSTR_RXY_RRRD }, + { "sty", 0x50, INSTR_RXY_RRRD }, + { "msy", 0x51, INSTR_RXY_RRRD }, + { "ny", 0x54, INSTR_RXY_RRRD }, + { "cly", 0x55, INSTR_RXY_RRRD }, + { "oy", 0x56, INSTR_RXY_RRRD }, + { "xy", 0x57, INSTR_RXY_RRRD }, + { "ly", 0x58, INSTR_RXY_RRRD }, + { "cy", 0x59, INSTR_RXY_RRRD }, + { "ay", 0x5a, INSTR_RXY_RRRD }, + { "sy", 0x5b, INSTR_RXY_RRRD }, + { "mfy", 0x5c, INSTR_RXY_RRRD }, + { "aly", 0x5e, INSTR_RXY_RRRD }, + { "sly", 0x5f, INSTR_RXY_RRRD }, + { "sthy", 0x70, INSTR_RXY_RRRD }, + { "lay", 0x71, INSTR_RXY_RRRD }, + { "stcy", 0x72, INSTR_RXY_RRRD }, + { "icy", 0x73, INSTR_RXY_RRRD }, + { "laey", 0x75, INSTR_RXY_RRRD }, + { "lb", 0x76, INSTR_RXY_RRRD }, + { "lgb", 0x77, INSTR_RXY_RRRD }, + { "lhy", 0x78, INSTR_RXY_RRRD }, + { "chy", 0x79, INSTR_RXY_RRRD }, + { "ahy", 0x7a, INSTR_RXY_RRRD }, + { "shy", 0x7b, INSTR_RXY_RRRD }, + { "mhy", 0x7c, INSTR_RXY_RRRD }, + { "ng", 0x80, INSTR_RXY_RRRD }, + { "og", 0x81, INSTR_RXY_RRRD }, + { "xg", 0x82, INSTR_RXY_RRRD }, + { "lgat", 0x85, INSTR_RXY_RRRD }, + { "mlg", 0x86, INSTR_RXY_RRRD }, + { "dlg", 0x87, INSTR_RXY_RRRD }, + { "alcg", 0x88, INSTR_RXY_RRRD }, + { "slbg", 0x89, INSTR_RXY_RRRD }, + { "stpq", 0x8e, INSTR_RXY_RRRD }, + { "lpq", 0x8f, INSTR_RXY_RRRD }, + { "llgc", 0x90, INSTR_RXY_RRRD }, + { "llgh", 0x91, INSTR_RXY_RRRD }, + { "llc", 0x94, INSTR_RXY_RRRD }, + { "llh", 0x95, INSTR_RXY_RRRD }, + { { 0, LONG_INSN_LLGTAT }, 0x9c, INSTR_RXY_RRRD }, + { { 0, LONG_INSN_LLGFAT }, 0x9d, INSTR_RXY_RRRD }, + { "lat", 0x9f, INSTR_RXY_RRRD }, + { "lbh", 0xc0, INSTR_RXY_RRRD }, + { "llch", 0xc2, INSTR_RXY_RRRD }, + { "stch", 0xc3, INSTR_RXY_RRRD }, + { "lhh", 0xc4, INSTR_RXY_RRRD }, + { "llhh", 0xc6, INSTR_RXY_RRRD }, + { "sthh", 0xc7, INSTR_RXY_RRRD }, + { "lfhat", 0xc8, INSTR_RXY_RRRD }, + { "lfh", 0xca, INSTR_RXY_RRRD }, + { "stfh", 0xcb, INSTR_RXY_RRRD }, + { "chf", 0xcd, INSTR_RXY_RRRD }, + { "clhf", 0xcf, INSTR_RXY_RRRD }, + { { 0, LONG_INSN_MPCIFC }, 0xd0, INSTR_RXY_RRRD }, + { { 0, LONG_INSN_STPCIFC }, 0xd4, INSTR_RXY_RRRD }, + { "lrv", 0x1e, INSTR_RXY_RRRD }, + { "lrvh", 0x1f, INSTR_RXY_RRRD }, + { "strv", 0x3e, INSTR_RXY_RRRD }, + { "ml", 0x96, INSTR_RXY_RRRD }, + { "dl", 0x97, INSTR_RXY_RRRD }, + { "alc", 0x98, INSTR_RXY_RRRD }, + { "slb", 0x99, INSTR_RXY_RRRD }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_e5[] = { + { "strag", 0x02, INSTR_SSE_RDRD }, + { "mvhhi", 0x44, INSTR_SIL_RDI }, + { "mvghi", 0x48, INSTR_SIL_RDI }, + { "mvhi", 0x4c, INSTR_SIL_RDI }, + { "chhsi", 0x54, INSTR_SIL_RDI }, + { { 0, LONG_INSN_CLHHSI }, 0x55, INSTR_SIL_RDU }, + { "cghsi", 0x58, INSTR_SIL_RDI }, + { { 0, LONG_INSN_CLGHSI }, 0x59, INSTR_SIL_RDU }, + { "chsi", 0x5c, INSTR_SIL_RDI }, + { { 0, LONG_INSN_CLFHSI }, 0x5d, INSTR_SIL_RDU }, + { { 0, LONG_INSN_TBEGIN }, 0x60, INSTR_SIL_RDU }, + { { 0, LONG_INSN_TBEGINC }, 0x61, INSTR_SIL_RDU }, + { "lasp", 0x00, INSTR_SSE_RDRD }, + { "tprot", 0x01, INSTR_SSE_RDRD }, + { "mvcsk", 0x0e, INSTR_SSE_RDRD }, + { "mvcdk", 0x0f, INSTR_SSE_RDRD }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_e7[] = { + { "lcbb", 0x27, INSTR_RXE_RRRDM }, + { "vgef", 0x13, INSTR_VRV_VVRDM }, + { "vgeg", 0x12, INSTR_VRV_VVRDM }, + { "vgbm", 0x44, INSTR_VRI_V0I0 }, + { "vgm", 0x46, INSTR_VRI_V0IIM }, + { "vl", 0x06, INSTR_VRX_VRRD0 }, + { "vlr", 0x56, INSTR_VRR_VV00000 }, + { "vlrp", 0x05, INSTR_VRX_VRRDM }, + { "vleb", 0x00, INSTR_VRX_VRRDM }, + { "vleh", 0x01, INSTR_VRX_VRRDM }, + { "vlef", 0x03, INSTR_VRX_VRRDM }, + { "vleg", 0x02, INSTR_VRX_VRRDM }, + { "vleib", 0x40, INSTR_VRI_V0IM }, + { "vleih", 0x41, INSTR_VRI_V0IM }, + { "vleif", 0x43, INSTR_VRI_V0IM }, + { "vleig", 0x42, INSTR_VRI_V0IM }, + { "vlgv", 0x21, INSTR_VRS_RVRDM }, + { "vllez", 0x04, INSTR_VRX_VRRDM }, + { "vlm", 0x36, INSTR_VRS_VVRD0 }, + { "vlbb", 0x07, INSTR_VRX_VRRDM }, + { "vlvg", 0x22, INSTR_VRS_VRRDM }, + { "vlvgp", 0x62, INSTR_VRR_VRR0000 }, + { "vll", 0x37, INSTR_VRS_VRRD0 }, + { "vmrh", 0x61, INSTR_VRR_VVV000M }, + { "vmrl", 0x60, INSTR_VRR_VVV000M }, + { "vpk", 0x94, INSTR_VRR_VVV000M }, + { "vpks", 0x97, INSTR_VRR_VVV0M0M }, + { "vpkls", 0x95, INSTR_VRR_VVV0M0M }, + { "vperm", 0x8c, INSTR_VRR_VVV000V }, + { "vpdi", 0x84, INSTR_VRR_VVV000M }, + { "vrep", 0x4d, INSTR_VRI_VVIM }, + { "vrepi", 0x45, INSTR_VRI_V0IM }, + { "vscef", 0x1b, INSTR_VRV_VWRDM }, + { "vsceg", 0x1a, INSTR_VRV_VWRDM }, + { "vsel", 0x8d, INSTR_VRR_VVV000V }, + { "vseg", 0x5f, INSTR_VRR_VV0000M }, + { "vst", 0x0e, INSTR_VRX_VRRD0 }, + { "vsteb", 0x08, INSTR_VRX_VRRDM }, + { "vsteh", 0x09, INSTR_VRX_VRRDM }, + { "vstef", 0x0b, INSTR_VRX_VRRDM }, + { "vsteg", 0x0a, INSTR_VRX_VRRDM }, + { "vstm", 0x3e, INSTR_VRS_VVRD0 }, + { "vstl", 0x3f, INSTR_VRS_VRRD0 }, + { "vuph", 0xd7, INSTR_VRR_VV0000M }, + { "vuplh", 0xd5, INSTR_VRR_VV0000M }, + { "vupl", 0xd6, INSTR_VRR_VV0000M }, + { "vupll", 0xd4, INSTR_VRR_VV0000M }, + { "va", 0xf3, INSTR_VRR_VVV000M }, + { "vacc", 0xf1, INSTR_VRR_VVV000M }, + { "vac", 0xbb, INSTR_VRR_VVVM00V }, + { "vaccc", 0xb9, INSTR_VRR_VVVM00V }, + { "vn", 0x68, INSTR_VRR_VVV0000 }, + { "vnc", 0x69, INSTR_VRR_VVV0000 }, + { "vavg", 0xf2, INSTR_VRR_VVV000M }, + { "vavgl", 0xf0, INSTR_VRR_VVV000M }, + { "vcksm", 0x66, INSTR_VRR_VVV0000 }, + { "vec", 0xdb, INSTR_VRR_VV0000M }, + { "vecl", 0xd9, INSTR_VRR_VV0000M }, + { "vceq", 0xf8, INSTR_VRR_VVV0M0M }, + { "vch", 0xfb, INSTR_VRR_VVV0M0M }, + { "vchl", 0xf9, INSTR_VRR_VVV0M0M }, + { "vclz", 0x53, INSTR_VRR_VV0000M }, + { "vctz", 0x52, INSTR_VRR_VV0000M }, + { "vx", 0x6d, INSTR_VRR_VVV0000 }, + { "vgfm", 0xb4, INSTR_VRR_VVV000M }, + { "vgfma", 0xbc, INSTR_VRR_VVVM00V }, + { "vlc", 0xde, INSTR_VRR_VV0000M }, + { "vlp", 0xdf, INSTR_VRR_VV0000M }, + { "vmx", 0xff, INSTR_VRR_VVV000M }, + { "vmxl", 0xfd, INSTR_VRR_VVV000M }, + { "vmn", 0xfe, INSTR_VRR_VVV000M }, + { "vmnl", 0xfc, INSTR_VRR_VVV000M }, + { "vmal", 0xaa, INSTR_VRR_VVVM00V }, + { "vmae", 0xae, INSTR_VRR_VVVM00V }, + { "vmale", 0xac, INSTR_VRR_VVVM00V }, + { "vmah", 0xab, INSTR_VRR_VVVM00V }, + { "vmalh", 0xa9, INSTR_VRR_VVVM00V }, + { "vmao", 0xaf, INSTR_VRR_VVVM00V }, + { "vmalo", 0xad, INSTR_VRR_VVVM00V }, + { "vmh", 0xa3, INSTR_VRR_VVV000M }, + { "vmlh", 0xa1, INSTR_VRR_VVV000M }, + { "vml", 0xa2, INSTR_VRR_VVV000M }, + { "vme", 0xa6, INSTR_VRR_VVV000M }, + { "vmle", 0xa4, INSTR_VRR_VVV000M }, + { "vmo", 0xa7, INSTR_VRR_VVV000M }, + { "vmlo", 0xa5, INSTR_VRR_VVV000M }, + { "vno", 0x6b, INSTR_VRR_VVV0000 }, + { "vo", 0x6a, INSTR_VRR_VVV0000 }, + { { 0, LONG_INSN_VPOPCT }, 0x50, INSTR_VRR_VV0000M }, + { { 0, LONG_INSN_VERLLV }, 0x73, INSTR_VRR_VVV000M }, + { "verll", 0x33, INSTR_VRS_VVRDM }, + { "verim", 0x72, INSTR_VRI_VVV0IM }, + { "veslv", 0x70, INSTR_VRR_VVV000M }, + { "vesl", 0x30, INSTR_VRS_VVRDM }, + { { 0, LONG_INSN_VESRAV }, 0x7a, INSTR_VRR_VVV000M }, + { "vesra", 0x3a, INSTR_VRS_VVRDM }, + { { 0, LONG_INSN_VESRLV }, 0x78, INSTR_VRR_VVV000M }, + { "vesrl", 0x38, INSTR_VRS_VVRDM }, + { "vsl", 0x74, INSTR_VRR_VVV0000 }, + { "vslb", 0x75, INSTR_VRR_VVV0000 }, + { "vsldb", 0x77, INSTR_VRI_VVV0I0 }, + { "vsra", 0x7e, INSTR_VRR_VVV0000 }, + { "vsrab", 0x7f, INSTR_VRR_VVV0000 }, + { "vsrl", 0x7c, INSTR_VRR_VVV0000 }, + { "vsrlb", 0x7d, INSTR_VRR_VVV0000 }, + { "vs", 0xf7, INSTR_VRR_VVV000M }, + { "vscb", 0xf5, INSTR_VRR_VVV000M }, + { "vsb", 0xbf, INSTR_VRR_VVVM00V }, + { { 0, LONG_INSN_VSBCBI }, 0xbd, INSTR_VRR_VVVM00V }, + { "vsumg", 0x65, INSTR_VRR_VVV000M }, + { "vsumq", 0x67, INSTR_VRR_VVV000M }, + { "vsum", 0x64, INSTR_VRR_VVV000M }, + { "vtm", 0xd8, INSTR_VRR_VV00000 }, + { "vfae", 0x82, INSTR_VRR_VVV0M0M }, + { "vfee", 0x80, INSTR_VRR_VVV0M0M }, + { "vfene", 0x81, INSTR_VRR_VVV0M0M }, + { "vistr", 0x5c, INSTR_VRR_VV00M0M }, + { "vstrc", 0x8a, INSTR_VRR_VVVMM0V }, + { "vfa", 0xe3, INSTR_VRR_VVV00MM }, + { "wfc", 0xcb, INSTR_VRR_VV000MM }, + { "wfk", 0xca, INSTR_VRR_VV000MM }, + { "vfce", 0xe8, INSTR_VRR_VVV0MMM }, + { "vfch", 0xeb, INSTR_VRR_VVV0MMM }, + { "vfche", 0xea, INSTR_VRR_VVV0MMM }, + { "vcdg", 0xc3, INSTR_VRR_VV00MMM }, + { "vcdlg", 0xc1, INSTR_VRR_VV00MMM }, + { "vcgd", 0xc2, INSTR_VRR_VV00MMM }, + { "vclgd", 0xc0, INSTR_VRR_VV00MMM }, + { "vfd", 0xe5, INSTR_VRR_VVV00MM }, + { "vfi", 0xc7, INSTR_VRR_VV00MMM }, + { "vlde", 0xc4, INSTR_VRR_VV000MM }, + { "vled", 0xc5, INSTR_VRR_VV00MMM }, + { "vfm", 0xe7, INSTR_VRR_VVV00MM }, + { "vfma", 0x8f, INSTR_VRR_VVVM0MV }, + { "vfms", 0x8e, INSTR_VRR_VVVM0MV }, + { "vfpso", 0xcc, INSTR_VRR_VV00MMM }, + { "vfsq", 0xce, INSTR_VRR_VV000MM }, + { "vfs", 0xe2, INSTR_VRR_VVV00MM }, + { "vftci", 0x4a, INSTR_VRI_VVIMM }, +}; + +static struct s390_insn opcode_eb[] = { + { "lmg", 0x04, INSTR_RSY_RRRD }, + { "srag", 0x0a, INSTR_RSY_RRRD }, + { "slag", 0x0b, INSTR_RSY_RRRD }, + { "srlg", 0x0c, INSTR_RSY_RRRD }, + { "sllg", 0x0d, INSTR_RSY_RRRD }, + { "tracg", 0x0f, INSTR_RSY_RRRD }, + { "csy", 0x14, INSTR_RSY_RRRD }, + { "rllg", 0x1c, INSTR_RSY_RRRD }, + { "clmh", 0x20, INSTR_RSY_RURD }, + { "clmy", 0x21, INSTR_RSY_RURD }, + { "clt", 0x23, INSTR_RSY_RURD }, + { "stmg", 0x24, INSTR_RSY_RRRD }, + { "stctg", 0x25, INSTR_RSY_CCRD }, + { "stmh", 0x26, INSTR_RSY_RRRD }, + { "clgt", 0x2b, INSTR_RSY_RURD }, + { "stcmh", 0x2c, INSTR_RSY_RURD }, + { "stcmy", 0x2d, INSTR_RSY_RURD }, + { "lctlg", 0x2f, INSTR_RSY_CCRD }, + { "csg", 0x30, INSTR_RSY_RRRD }, + { "cdsy", 0x31, INSTR_RSY_RRRD }, + { "cdsg", 0x3e, INSTR_RSY_RRRD }, + { "bxhg", 0x44, INSTR_RSY_RRRD }, + { "bxleg", 0x45, INSTR_RSY_RRRD }, + { "ecag", 0x4c, INSTR_RSY_RRRD }, + { "tmy", 0x51, INSTR_SIY_URD }, + { "mviy", 0x52, INSTR_SIY_URD }, + { "niy", 0x54, INSTR_SIY_URD }, + { "cliy", 0x55, INSTR_SIY_URD }, + { "oiy", 0x56, INSTR_SIY_URD }, + { "xiy", 0x57, INSTR_SIY_URD }, + { "asi", 0x6a, INSTR_SIY_IRD }, + { "alsi", 0x6e, INSTR_SIY_IRD }, + { "agsi", 0x7a, INSTR_SIY_IRD }, + { "algsi", 0x7e, INSTR_SIY_IRD }, + { "icmh", 0x80, INSTR_RSY_RURD }, + { "icmy", 0x81, INSTR_RSY_RURD }, + { "clclu", 0x8f, INSTR_RSY_RRRD }, + { "stmy", 0x90, INSTR_RSY_RRRD }, + { "lmh", 0x96, INSTR_RSY_RRRD }, + { "lmy", 0x98, INSTR_RSY_RRRD }, + { "lamy", 0x9a, INSTR_RSY_AARD }, + { "stamy", 0x9b, INSTR_RSY_AARD }, + { { 0, LONG_INSN_PCISTB }, 0xd0, INSTR_RSY_RRRD }, + { "sic", 0xd1, INSTR_RSY_RRRD }, + { "srak", 0xdc, INSTR_RSY_RRRD }, + { "slak", 0xdd, INSTR_RSY_RRRD }, + { "srlk", 0xde, INSTR_RSY_RRRD }, + { "sllk", 0xdf, INSTR_RSY_RRRD }, + { "locg", 0xe2, INSTR_RSY_RDRM }, + { "stocg", 0xe3, INSTR_RSY_RDRM }, + { "lang", 0xe4, INSTR_RSY_RRRD }, + { "laog", 0xe6, INSTR_RSY_RRRD }, + { "laxg", 0xe7, INSTR_RSY_RRRD }, + { "laag", 0xe8, INSTR_RSY_RRRD }, + { "laalg", 0xea, INSTR_RSY_RRRD }, + { "loc", 0xf2, INSTR_RSY_RDRM }, + { "stoc", 0xf3, INSTR_RSY_RDRM }, + { "lan", 0xf4, INSTR_RSY_RRRD }, + { "lao", 0xf6, INSTR_RSY_RRRD }, + { "lax", 0xf7, INSTR_RSY_RRRD }, + { "laa", 0xf8, INSTR_RSY_RRRD }, + { "laal", 0xfa, INSTR_RSY_RRRD }, + { "lric", 0x60, INSTR_RSY_RDRM }, + { "stric", 0x61, INSTR_RSY_RDRM }, + { "mric", 0x62, INSTR_RSY_RDRM }, + { { 0, LONG_INSN_STCCTM }, 0x17, INSTR_RSY_RMRD }, + { "rll", 0x1d, INSTR_RSY_RRRD }, + { "mvclu", 0x8e, INSTR_RSY_RRRD }, + { "tp", 0xc0, INSTR_RSL_R0RD }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_ec[] = { + { "brxhg", 0x44, INSTR_RIE_RRP }, + { "brxlg", 0x45, INSTR_RIE_RRP }, + { { 0, LONG_INSN_RISBLG }, 0x51, INSTR_RIE_RRUUU }, + { "rnsbg", 0x54, INSTR_RIE_RRUUU }, + { "risbg", 0x55, INSTR_RIE_RRUUU }, + { "rosbg", 0x56, INSTR_RIE_RRUUU }, + { "rxsbg", 0x57, INSTR_RIE_RRUUU }, + { { 0, LONG_INSN_RISBGN }, 0x59, INSTR_RIE_RRUUU }, + { { 0, LONG_INSN_RISBHG }, 0x5D, INSTR_RIE_RRUUU }, + { "cgrj", 0x64, INSTR_RIE_RRPU }, + { "clgrj", 0x65, INSTR_RIE_RRPU }, + { "cgit", 0x70, INSTR_RIE_R0IU }, + { "clgit", 0x71, INSTR_RIE_R0UU }, + { "cit", 0x72, INSTR_RIE_R0IU }, + { "clfit", 0x73, INSTR_RIE_R0UU }, + { "crj", 0x76, INSTR_RIE_RRPU }, + { "clrj", 0x77, INSTR_RIE_RRPU }, + { "cgij", 0x7c, INSTR_RIE_RUPI }, + { "clgij", 0x7d, INSTR_RIE_RUPU }, + { "cij", 0x7e, INSTR_RIE_RUPI }, + { "clij", 0x7f, INSTR_RIE_RUPU }, + { "ahik", 0xd8, INSTR_RIE_RRI0 }, + { "aghik", 0xd9, INSTR_RIE_RRI0 }, + { { 0, LONG_INSN_ALHSIK }, 0xda, INSTR_RIE_RRI0 }, + { { 0, LONG_INSN_ALGHSIK }, 0xdb, INSTR_RIE_RRI0 }, + { "cgrb", 0xe4, INSTR_RRS_RRRDU }, + { "clgrb", 0xe5, INSTR_RRS_RRRDU }, + { "crb", 0xf6, INSTR_RRS_RRRDU }, + { "clrb", 0xf7, INSTR_RRS_RRRDU }, + { "cgib", 0xfc, INSTR_RIS_RURDI }, + { "clgib", 0xfd, INSTR_RIS_RURDU }, + { "cib", 0xfe, INSTR_RIS_RURDI }, + { "clib", 0xff, INSTR_RIS_RURDU }, + { "", 0, INSTR_INVALID } +}; + +static struct s390_insn opcode_ed[] = { + { "mayl", 0x38, INSTR_RXF_FRRDF }, + { "myl", 0x39, INSTR_RXF_FRRDF }, + { "may", 0x3a, INSTR_RXF_FRRDF }, + { "my", 0x3b, INSTR_RXF_FRRDF }, + { "mayh", 0x3c, INSTR_RXF_FRRDF }, + { "myh", 0x3d, INSTR_RXF_FRRDF }, + { "sldt", 0x40, INSTR_RXF_FRRDF }, + { "srdt", 0x41, INSTR_RXF_FRRDF }, + { "slxt", 0x48, INSTR_RXF_FRRDF }, + { "srxt", 0x49, INSTR_RXF_FRRDF }, + { "tdcet", 0x50, INSTR_RXE_FRRD }, + { "tdget", 0x51, INSTR_RXE_FRRD }, + { "tdcdt", 0x54, INSTR_RXE_FRRD }, + { "tdgdt", 0x55, INSTR_RXE_FRRD }, + { "tdcxt", 0x58, INSTR_RXE_FRRD }, + { "tdgxt", 0x59, INSTR_RXE_FRRD }, + { "ley", 0x64, INSTR_RXY_FRRD }, + { "ldy", 0x65, INSTR_RXY_FRRD }, + { "stey", 0x66, INSTR_RXY_FRRD }, + { "stdy", 0x67, INSTR_RXY_FRRD }, + { "czdt", 0xa8, INSTR_RSL_LRDFU }, + { "czxt", 0xa9, INSTR_RSL_LRDFU }, + { "cdzt", 0xaa, INSTR_RSL_LRDFU }, + { "cxzt", 0xab, INSTR_RSL_LRDFU }, + { "ldeb", 0x04, INSTR_RXE_FRRD }, + { "lxdb", 0x05, INSTR_RXE_FRRD }, + { "lxeb", 0x06, INSTR_RXE_FRRD }, + { "mxdb", 0x07, INSTR_RXE_FRRD }, + { "keb", 0x08, INSTR_RXE_FRRD }, + { "ceb", 0x09, INSTR_RXE_FRRD }, + { "aeb", 0x0a, INSTR_RXE_FRRD }, + { "seb", 0x0b, INSTR_RXE_FRRD }, + { "mdeb", 0x0c, INSTR_RXE_FRRD }, + { "deb", 0x0d, INSTR_RXE_FRRD }, + { "maeb", 0x0e, INSTR_RXF_FRRDF }, + { "mseb", 0x0f, INSTR_RXF_FRRDF }, + { "tceb", 0x10, INSTR_RXE_FRRD }, + { "tcdb", 0x11, INSTR_RXE_FRRD }, + { "tcxb", 0x12, INSTR_RXE_FRRD }, + { "sqeb", 0x14, INSTR_RXE_FRRD }, + { "sqdb", 0x15, INSTR_RXE_FRRD }, + { "meeb", 0x17, INSTR_RXE_FRRD }, + { "kdb", 0x18, INSTR_RXE_FRRD }, + { "cdb", 0x19, INSTR_RXE_FRRD }, + { "adb", 0x1a, INSTR_RXE_FRRD }, + { "sdb", 0x1b, INSTR_RXE_FRRD }, + { "mdb", 0x1c, INSTR_RXE_FRRD }, + { "ddb", 0x1d, INSTR_RXE_FRRD }, + { "madb", 0x1e, INSTR_RXF_FRRDF }, + { "msdb", 0x1f, INSTR_RXF_FRRDF }, + { "lde", 0x24, INSTR_RXE_FRRD }, + { "lxd", 0x25, INSTR_RXE_FRRD }, + { "lxe", 0x26, INSTR_RXE_FRRD }, + { "mae", 0x2e, INSTR_RXF_FRRDF }, + { "mse", 0x2f, INSTR_RXF_FRRDF }, + { "sqe", 0x34, INSTR_RXE_FRRD }, + { "sqd", 0x35, INSTR_RXE_FRRD }, + { "mee", 0x37, INSTR_RXE_FRRD }, + { "mad", 0x3e, INSTR_RXF_FRRDF }, + { "msd", 0x3f, INSTR_RXF_FRRDF }, + { "", 0, INSTR_INVALID } +}; + +/* Extracts an operand value from an instruction. */ +static unsigned int extract_operand(unsigned char *code, + const struct s390_operand *operand) +{ + unsigned char *cp; + unsigned int val; + int bits; + + /* Extract fragments of the operand byte for byte. */ + cp = code + operand->shift / 8; + bits = (operand->shift & 7) + operand->bits; + val = 0; + do { + val <<= 8; + val |= (unsigned int) *cp++; + bits -= 8; + } while (bits > 0); + val >>= -bits; + val &= ((1U << (operand->bits - 1)) << 1) - 1; + + /* Check for special long displacement case. */ + if (operand->bits == 20 && operand->shift == 20) + val = (val & 0xff) << 12 | (val & 0xfff00) >> 8; + + /* Check for register extensions bits for vector registers. */ + if (operand->flags & OPERAND_VR) { + if (operand->shift == 8) + val |= (code[4] & 8) << 1; + else if (operand->shift == 12) + val |= (code[4] & 4) << 2; + else if (operand->shift == 16) + val |= (code[4] & 2) << 3; + else if (operand->shift == 32) + val |= (code[4] & 1) << 4; + } + + /* Sign extend value if the operand is signed or pc relative. */ + if ((operand->flags & (OPERAND_SIGNED | OPERAND_PCREL)) && + (val & (1U << (operand->bits - 1)))) + val |= (-1U << (operand->bits - 1)) << 1; + + /* Double value if the operand is pc relative. */ + if (operand->flags & OPERAND_PCREL) + val <<= 1; + + /* Length x in an instructions has real length x + 1. */ + if (operand->flags & OPERAND_LENGTH) + val++; + return val; +} + +struct s390_insn *find_insn(unsigned char *code) +{ + unsigned char opfrag = code[1]; + unsigned char opmask; + struct s390_insn *table; + + switch (code[0]) { + case 0x01: + table = opcode_01; + break; + case 0xa5: + table = opcode_a5; + break; + case 0xa7: + table = opcode_a7; + break; + case 0xaa: + table = opcode_aa; + break; + case 0xb2: + table = opcode_b2; + break; + case 0xb3: + table = opcode_b3; + break; + case 0xb9: + table = opcode_b9; + break; + case 0xc0: + table = opcode_c0; + break; + case 0xc2: + table = opcode_c2; + break; + case 0xc4: + table = opcode_c4; + break; + case 0xc6: + table = opcode_c6; + break; + case 0xc8: + table = opcode_c8; + break; + case 0xcc: + table = opcode_cc; + break; + case 0xe3: + table = opcode_e3; + opfrag = code[5]; + break; + case 0xe5: + table = opcode_e5; + break; + case 0xe7: + table = opcode_e7; + opfrag = code[5]; + break; + case 0xeb: + table = opcode_eb; + opfrag = code[5]; + break; + case 0xec: + table = opcode_ec; + opfrag = code[5]; + break; + case 0xed: + table = opcode_ed; + opfrag = code[5]; + break; + default: + table = opcode; + opfrag = code[0]; + break; + } + while (table->format != INSTR_INVALID) { + opmask = formats[table->format][0]; + if (table->opfrag == (opfrag & opmask)) + return table; + table++; + } + return NULL; +} + +/** + * insn_to_mnemonic - decode an s390 instruction + * @instruction: instruction to decode + * @buf: buffer to fill with mnemonic + * @len: length of buffer + * + * Decode the instruction at @instruction and store the corresponding + * mnemonic into @buf of length @len. + * @buf is left unchanged if the instruction could not be decoded. + * Returns: + * %0 on success, %-ENOENT if the instruction was not found. + */ +int insn_to_mnemonic(unsigned char *instruction, char *buf, unsigned int len) +{ + struct s390_insn *insn; + + insn = find_insn(instruction); + if (!insn) + return -ENOENT; + if (insn->name[0] == '\0') + snprintf(buf, len, "%s", + long_insn_name[(int) insn->name[1]]); + else + snprintf(buf, len, "%.5s", insn->name); + return 0; +} +EXPORT_SYMBOL_GPL(insn_to_mnemonic); + +static int print_insn(char *buffer, unsigned char *code, unsigned long addr) +{ + struct s390_insn *insn; + const unsigned char *ops; + const struct s390_operand *operand; + unsigned int value; + char separator; + char *ptr; + int i; + + ptr = buffer; + insn = find_insn(code); + if (insn) { + if (insn->name[0] == '\0') + ptr += sprintf(ptr, "%s\t", + long_insn_name[(int) insn->name[1]]); + else + ptr += sprintf(ptr, "%.5s\t", insn->name); + /* Extract the operands. */ + separator = 0; + for (ops = formats[insn->format] + 1, i = 0; + *ops != 0 && i < 6; ops++, i++) { + operand = operands + *ops; + value = extract_operand(code, operand); + if ((operand->flags & OPERAND_INDEX) && value == 0) + continue; + if ((operand->flags & OPERAND_BASE) && + value == 0 && separator == '(') { + separator = ','; + continue; + } + if (separator) + ptr += sprintf(ptr, "%c", separator); + if (operand->flags & OPERAND_GPR) + ptr += sprintf(ptr, "%%r%i", value); + else if (operand->flags & OPERAND_FPR) + ptr += sprintf(ptr, "%%f%i", value); + else if (operand->flags & OPERAND_AR) + ptr += sprintf(ptr, "%%a%i", value); + else if (operand->flags & OPERAND_CR) + ptr += sprintf(ptr, "%%c%i", value); + else if (operand->flags & OPERAND_VR) + ptr += sprintf(ptr, "%%v%i", value); + else if (operand->flags & OPERAND_PCREL) + ptr += sprintf(ptr, "%lx", (signed int) value + + addr); + else if (operand->flags & OPERAND_SIGNED) + ptr += sprintf(ptr, "%i", value); + else + ptr += sprintf(ptr, "%u", value); + if (operand->flags & OPERAND_DISP) + separator = '('; + else if (operand->flags & OPERAND_BASE) { + ptr += sprintf(ptr, ")"); + separator = ','; + } else + separator = ','; + } + } else + ptr += sprintf(ptr, "unknown"); + return (int) (ptr - buffer); +} + +void show_code(struct pt_regs *regs) +{ + char *mode = user_mode(regs) ? "User" : "Krnl"; + unsigned char code[64]; + char buffer[64], *ptr; + mm_segment_t old_fs; + unsigned long addr; + int start, end, opsize, hops, i; + + /* Get a snapshot of the 64 bytes surrounding the fault address. */ + old_fs = get_fs(); + set_fs(user_mode(regs) ? USER_DS : KERNEL_DS); + for (start = 32; start && regs->psw.addr >= 34 - start; start -= 2) { + addr = regs->psw.addr - 34 + start; + if (__copy_from_user(code + start - 2, + (char __user *) addr, 2)) + break; + } + for (end = 32; end < 64; end += 2) { + addr = regs->psw.addr + end - 32; + if (__copy_from_user(code + end, + (char __user *) addr, 2)) + break; + } + set_fs(old_fs); + /* Code snapshot useable ? */ + if ((regs->psw.addr & 1) || start >= end) { + printk("%s Code: Bad PSW.\n", mode); + return; + } + /* Find a starting point for the disassembly. */ + while (start < 32) { + for (i = 0, hops = 0; start + i < 32 && hops < 3; hops++) { + if (!find_insn(code + start + i)) + break; + i += insn_length(code[start + i]); + } + if (start + i == 32) + /* Looks good, sequence ends at PSW. */ + break; + start += 2; + } + /* Decode the instructions. */ + ptr = buffer; + ptr += sprintf(ptr, "%s Code:", mode); + hops = 0; + while (start < end && hops < 8) { + opsize = insn_length(code[start]); + if (start + opsize == 32) + *ptr++ = '#'; + else if (start == 32) + *ptr++ = '>'; + else + *ptr++ = ' '; + addr = regs->psw.addr + start - 32; + ptr += sprintf(ptr, "%016lx: ", addr); + if (start + opsize >= end) + break; + for (i = 0; i < opsize; i++) + ptr += sprintf(ptr, "%02x", code[start + i]); + *ptr++ = '\t'; + if (i < 6) + *ptr++ = '\t'; + ptr += print_insn(ptr, code + start, addr); + start += opsize; + printk(buffer); + ptr = buffer; + ptr += sprintf(ptr, "\n "); + hops++; + } + printk("\n"); +} + +void print_fn_code(unsigned char *code, unsigned long len) +{ + char buffer[64], *ptr; + int opsize, i; + + while (len) { + ptr = buffer; + opsize = insn_length(*code); + if (opsize > len) + break; + ptr += sprintf(ptr, "%p: ", code); + for (i = 0; i < opsize; i++) + ptr += sprintf(ptr, "%02x", code[i]); + *ptr++ = '\t'; + if (i < 4) + *ptr++ = '\t'; + ptr += print_insn(ptr, code, (unsigned long) code); + *ptr++ = '\n'; + *ptr++ = 0; + printk(buffer); + code += opsize; + len -= opsize; + } +} diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c new file mode 100644 index 000000000..dc8e20473 --- /dev/null +++ b/arch/s390/kernel/dumpstack.c @@ -0,0 +1,204 @@ +/* + * Stack dumping functions + * + * Copyright IBM Corp. 1999, 2013 + */ + +#include <linux/kallsyms.h> +#include <linux/hardirq.h> +#include <linux/kprobes.h> +#include <linux/utsname.h> +#include <linux/export.h> +#include <linux/kdebug.h> +#include <linux/ptrace.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <asm/processor.h> +#include <asm/debug.h> +#include <asm/dis.h> +#include <asm/ipl.h> + +/* + * For show_trace we have tree different stack to consider: + * - the panic stack which is used if the kernel stack has overflown + * - the asynchronous interrupt stack (cpu related) + * - the synchronous kernel stack (process related) + * The stack trace can start at any of the three stack and can potentially + * touch all of them. The order is: panic stack, async stack, sync stack. + */ +static unsigned long +__show_trace(unsigned long sp, unsigned long low, unsigned long high) +{ + struct stack_frame *sf; + struct pt_regs *regs; + unsigned long addr; + + while (1) { + sp = sp & PSW_ADDR_INSN; + if (sp < low || sp > high - sizeof(*sf)) + return sp; + sf = (struct stack_frame *) sp; + addr = sf->gprs[8] & PSW_ADDR_INSN; + printk("([<%016lx>] %pSR)\n", addr, (void *)addr); + /* Follow the backchain. */ + while (1) { + low = sp; + sp = sf->back_chain & PSW_ADDR_INSN; + if (!sp) + break; + if (sp <= low || sp > high - sizeof(*sf)) + return sp; + sf = (struct stack_frame *) sp; + addr = sf->gprs[8] & PSW_ADDR_INSN; + printk(" [<%016lx>] %pSR\n", addr, (void *)addr); + } + /* Zero backchain detected, check for interrupt frame. */ + sp = (unsigned long) (sf + 1); + if (sp <= low || sp > high - sizeof(*regs)) + return sp; + regs = (struct pt_regs *) sp; + addr = regs->psw.addr & PSW_ADDR_INSN; + printk(" [<%016lx>] %pSR\n", addr, (void *)addr); + low = sp; + sp = regs->gprs[15]; + } +} + +static void show_trace(struct task_struct *task, unsigned long *stack) +{ + const unsigned long frame_size = + STACK_FRAME_OVERHEAD + sizeof(struct pt_regs); + register unsigned long __r15 asm ("15"); + unsigned long sp; + + sp = (unsigned long) stack; + if (!sp) + sp = task ? task->thread.ksp : __r15; + printk("Call Trace:\n"); +#ifdef CONFIG_CHECK_STACK + sp = __show_trace(sp, + S390_lowcore.panic_stack + frame_size - 4096, + S390_lowcore.panic_stack + frame_size); +#endif + sp = __show_trace(sp, + S390_lowcore.async_stack + frame_size - ASYNC_SIZE, + S390_lowcore.async_stack + frame_size); + if (task) + __show_trace(sp, (unsigned long) task_stack_page(task), + (unsigned long) task_stack_page(task) + THREAD_SIZE); + else + __show_trace(sp, S390_lowcore.thread_info, + S390_lowcore.thread_info + THREAD_SIZE); + if (!task) + task = current; + debug_show_held_locks(task); +} + +void show_stack(struct task_struct *task, unsigned long *sp) +{ + register unsigned long *__r15 asm ("15"); + unsigned long *stack; + int i; + + if (!sp) + stack = task ? (unsigned long *) task->thread.ksp : __r15; + else + stack = sp; + + for (i = 0; i < 20; i++) { + if (((addr_t) stack & (THREAD_SIZE-1)) == 0) + break; + if ((i * sizeof(long) % 32) == 0) + printk("%s ", i == 0 ? "" : "\n"); + printk("%016lx ", *stack++); + } + printk("\n"); + show_trace(task, sp); +} + +static void show_last_breaking_event(struct pt_regs *regs) +{ + printk("Last Breaking-Event-Address:\n"); + printk(" [<%016lx>] %pSR\n", regs->args[0], (void *)regs->args[0]); +} + +static inline int mask_bits(struct pt_regs *regs, unsigned long bits) +{ + return (regs->psw.mask & bits) / ((~bits + 1) & bits); +} + +void show_registers(struct pt_regs *regs) +{ + char *mode; + + mode = user_mode(regs) ? "User" : "Krnl"; + printk("%s PSW : %p %p", mode, (void *)regs->psw.mask, (void *)regs->psw.addr); + if (!user_mode(regs)) + printk(" (%pSR)", (void *)regs->psw.addr); + printk("\n"); + printk(" R:%x T:%x IO:%x EX:%x Key:%x M:%x W:%x " + "P:%x AS:%x CC:%x PM:%x", mask_bits(regs, PSW_MASK_PER), + mask_bits(regs, PSW_MASK_DAT), mask_bits(regs, PSW_MASK_IO), + mask_bits(regs, PSW_MASK_EXT), mask_bits(regs, PSW_MASK_KEY), + mask_bits(regs, PSW_MASK_MCHECK), mask_bits(regs, PSW_MASK_WAIT), + mask_bits(regs, PSW_MASK_PSTATE), mask_bits(regs, PSW_MASK_ASC), + mask_bits(regs, PSW_MASK_CC), mask_bits(regs, PSW_MASK_PM)); + printk(" EA:%x", mask_bits(regs, PSW_MASK_EA | PSW_MASK_BA)); + printk("\n%s GPRS: %016lx %016lx %016lx %016lx\n", mode, + regs->gprs[0], regs->gprs[1], regs->gprs[2], regs->gprs[3]); + printk(" %016lx %016lx %016lx %016lx\n", + regs->gprs[4], regs->gprs[5], regs->gprs[6], regs->gprs[7]); + printk(" %016lx %016lx %016lx %016lx\n", + regs->gprs[8], regs->gprs[9], regs->gprs[10], regs->gprs[11]); + printk(" %016lx %016lx %016lx %016lx\n", + regs->gprs[12], regs->gprs[13], regs->gprs[14], regs->gprs[15]); + show_code(regs); +} + +void show_regs(struct pt_regs *regs) +{ + show_regs_print_info(KERN_DEFAULT); + show_registers(regs); + /* Show stack backtrace if pt_regs is from kernel mode */ + if (!user_mode(regs)) + show_trace(NULL, (unsigned long *) regs->gprs[15]); + show_last_breaking_event(regs); +} + +static DEFINE_SPINLOCK(die_lock); + +void die(struct pt_regs *regs, const char *str) +{ + static int die_counter; + + oops_enter(); + lgr_info_log(); + debug_stop_all(); + console_verbose(); + spin_lock_irq(&die_lock); + bust_spinlocks(1); + printk("%s: %04x ilc:%d [#%d] ", str, regs->int_code & 0xffff, + regs->int_code >> 17, ++die_counter); +#ifdef CONFIG_PREEMPT + printk("PREEMPT "); +#endif +#ifdef CONFIG_SMP + printk("SMP "); +#endif +#ifdef CONFIG_DEBUG_PAGEALLOC + printk("DEBUG_PAGEALLOC"); +#endif + printk("\n"); + notify_die(DIE_OOPS, str, regs, 0, regs->int_code & 0xffff, SIGSEGV); + print_modules(); + show_regs(regs); + bust_spinlocks(0); + add_taint(TAINT_DIE, LOCKDEP_NOW_UNRELIABLE); + spin_unlock_irq(&die_lock); + if (in_interrupt()) + panic("Fatal exception in interrupt"); + if (panic_on_oops) + panic("Fatal exception: panic_on_oops"); + oops_exit(); + do_exit(SIGSEGV); +} diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c new file mode 100644 index 000000000..549a73a4b --- /dev/null +++ b/arch/s390/kernel/early.c @@ -0,0 +1,444 @@ +/* + * Copyright IBM Corp. 2007, 2009 + * Author(s): Hongjie Yang <hongjie@us.ibm.com>, + * Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#define KMSG_COMPONENT "setup" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/compiler.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <linux/string.h> +#include <linux/ctype.h> +#include <linux/lockdep.h> +#include <linux/module.h> +#include <linux/pfn.h> +#include <linux/uaccess.h> +#include <linux/kernel.h> +#include <asm/ebcdic.h> +#include <asm/ipl.h> +#include <asm/lowcore.h> +#include <asm/processor.h> +#include <asm/sections.h> +#include <asm/setup.h> +#include <asm/sysinfo.h> +#include <asm/cpcmd.h> +#include <asm/sclp.h> +#include <asm/facility.h> +#include "entry.h" + +/* + * Create a Kernel NSS if the SAVESYS= parameter is defined + */ +#define DEFSYS_CMD_SIZE 128 +#define SAVESYS_CMD_SIZE 32 + +char kernel_nss_name[NSS_NAME_SIZE + 1]; + +static void __init setup_boot_command_line(void); + +/* + * Get the TOD clock running. + */ +static void __init reset_tod_clock(void) +{ + u64 time; + + if (store_tod_clock(&time) == 0) + return; + /* TOD clock not running. Set the clock to Unix Epoch. */ + if (set_tod_clock(TOD_UNIX_EPOCH) != 0 || store_tod_clock(&time) != 0) + disabled_wait(0); + + sched_clock_base_cc = TOD_UNIX_EPOCH; + S390_lowcore.last_update_clock = sched_clock_base_cc; +} + +#ifdef CONFIG_SHARED_KERNEL +int __init savesys_ipl_nss(char *cmd, const int cmdlen); + +asm( + " .section .init.text,\"ax\",@progbits\n" + " .align 4\n" + " .type savesys_ipl_nss, @function\n" + "savesys_ipl_nss:\n" + " stmg 6,15,48(15)\n" + " lgr 14,3\n" + " sam31\n" + " diag 2,14,0x8\n" + " sam64\n" + " lgr 2,14\n" + " lmg 6,15,48(15)\n" + " br 14\n" + " .size savesys_ipl_nss, .-savesys_ipl_nss\n" + " .previous\n"); + +static __initdata char upper_command_line[COMMAND_LINE_SIZE]; + +static noinline __init void create_kernel_nss(void) +{ + unsigned int i, stext_pfn, eshared_pfn, end_pfn, min_size; +#ifdef CONFIG_BLK_DEV_INITRD + unsigned int sinitrd_pfn, einitrd_pfn; +#endif + int response; + int hlen; + size_t len; + char *savesys_ptr; + char defsys_cmd[DEFSYS_CMD_SIZE]; + char savesys_cmd[SAVESYS_CMD_SIZE]; + + /* Do nothing if we are not running under VM */ + if (!MACHINE_IS_VM) + return; + + /* Convert COMMAND_LINE to upper case */ + for (i = 0; i < strlen(boot_command_line); i++) + upper_command_line[i] = toupper(boot_command_line[i]); + + savesys_ptr = strstr(upper_command_line, "SAVESYS="); + + if (!savesys_ptr) + return; + + savesys_ptr += 8; /* Point to the beginning of the NSS name */ + for (i = 0; i < NSS_NAME_SIZE; i++) { + if (savesys_ptr[i] == ' ' || savesys_ptr[i] == '\0') + break; + kernel_nss_name[i] = savesys_ptr[i]; + } + + stext_pfn = PFN_DOWN(__pa(&_stext)); + eshared_pfn = PFN_DOWN(__pa(&_eshared)); + end_pfn = PFN_UP(__pa(&_end)); + min_size = end_pfn << 2; + + hlen = snprintf(defsys_cmd, DEFSYS_CMD_SIZE, + "DEFSYS %s 00000-%.5X EW %.5X-%.5X SR %.5X-%.5X", + kernel_nss_name, stext_pfn - 1, stext_pfn, + eshared_pfn - 1, eshared_pfn, end_pfn); + +#ifdef CONFIG_BLK_DEV_INITRD + if (INITRD_START && INITRD_SIZE) { + sinitrd_pfn = PFN_DOWN(__pa(INITRD_START)); + einitrd_pfn = PFN_UP(__pa(INITRD_START + INITRD_SIZE)); + min_size = einitrd_pfn << 2; + hlen += snprintf(defsys_cmd + hlen, DEFSYS_CMD_SIZE - hlen, + " EW %.5X-%.5X", sinitrd_pfn, einitrd_pfn); + } +#endif + + snprintf(defsys_cmd + hlen, DEFSYS_CMD_SIZE - hlen, + " EW MINSIZE=%.7iK PARMREGS=0-13", min_size); + defsys_cmd[DEFSYS_CMD_SIZE - 1] = '\0'; + snprintf(savesys_cmd, SAVESYS_CMD_SIZE, "SAVESYS %s \n IPL %s", + kernel_nss_name, kernel_nss_name); + savesys_cmd[SAVESYS_CMD_SIZE - 1] = '\0'; + + __cpcmd(defsys_cmd, NULL, 0, &response); + + if (response != 0) { + pr_err("Defining the Linux kernel NSS failed with rc=%d\n", + response); + kernel_nss_name[0] = '\0'; + return; + } + + len = strlen(savesys_cmd); + ASCEBC(savesys_cmd, len); + response = savesys_ipl_nss(savesys_cmd, len); + + /* On success: response is equal to the command size, + * max SAVESYS_CMD_SIZE + * On error: response contains the numeric portion of cp error message. + * for SAVESYS it will be >= 263 + * for missing privilege class, it will be 1 + */ + if (response > SAVESYS_CMD_SIZE || response == 1) { + pr_err("Saving the Linux kernel NSS failed with rc=%d\n", + response); + kernel_nss_name[0] = '\0'; + return; + } + + /* re-initialize cputime accounting. */ + sched_clock_base_cc = get_tod_clock(); + S390_lowcore.last_update_clock = sched_clock_base_cc; + S390_lowcore.last_update_timer = 0x7fffffffffffffffULL; + S390_lowcore.user_timer = 0; + S390_lowcore.system_timer = 0; + asm volatile("SPT 0(%0)" : : "a" (&S390_lowcore.last_update_timer)); + + /* re-setup boot command line with new ipl vm parms */ + ipl_update_parameters(); + setup_boot_command_line(); + + ipl_flags = IPL_NSS_VALID; +} + +#else /* CONFIG_SHARED_KERNEL */ + +static inline void create_kernel_nss(void) { } + +#endif /* CONFIG_SHARED_KERNEL */ + +/* + * Clear bss memory + */ +static noinline __init void clear_bss_section(void) +{ + memset(__bss_start, 0, __bss_stop - __bss_start); +} + +/* + * Initialize storage key for kernel pages + */ +static noinline __init void init_kernel_storage_key(void) +{ +#if PAGE_DEFAULT_KEY + unsigned long end_pfn, init_pfn; + + end_pfn = PFN_UP(__pa(&_end)); + + for (init_pfn = 0 ; init_pfn < end_pfn; init_pfn++) + page_set_storage_key(init_pfn << PAGE_SHIFT, + PAGE_DEFAULT_KEY, 0); +#endif +} + +static __initdata char sysinfo_page[PAGE_SIZE] __aligned(PAGE_SIZE); + +static noinline __init void detect_machine_type(void) +{ + struct sysinfo_3_2_2 *vmms = (struct sysinfo_3_2_2 *)&sysinfo_page; + + /* Check current-configuration-level */ + if (stsi(NULL, 0, 0, 0) <= 2) { + S390_lowcore.machine_flags |= MACHINE_FLAG_LPAR; + return; + } + /* Get virtual-machine cpu information. */ + if (stsi(vmms, 3, 2, 2) || !vmms->count) + return; + + /* Running under KVM? If not we assume z/VM */ + if (!memcmp(vmms->vm[0].cpi, "\xd2\xe5\xd4", 3)) + S390_lowcore.machine_flags |= MACHINE_FLAG_KVM; + else + S390_lowcore.machine_flags |= MACHINE_FLAG_VM; +} + +static __init void setup_topology(void) +{ + int max_mnest; + + if (!test_facility(11)) + return; + S390_lowcore.machine_flags |= MACHINE_FLAG_TOPOLOGY; + for (max_mnest = 6; max_mnest > 1; max_mnest--) { + if (stsi(&sysinfo_page, 15, 1, max_mnest) == 0) + break; + } + topology_max_mnest = max_mnest; +} + +static void early_pgm_check_handler(void) +{ + const struct exception_table_entry *fixup; + unsigned long cr0, cr0_new; + unsigned long addr; + + addr = S390_lowcore.program_old_psw.addr; + fixup = search_exception_tables(addr & PSW_ADDR_INSN); + if (!fixup) + disabled_wait(0); + /* Disable low address protection before storing into lowcore. */ + __ctl_store(cr0, 0, 0); + cr0_new = cr0 & ~(1UL << 28); + __ctl_load(cr0_new, 0, 0); + S390_lowcore.program_old_psw.addr = extable_fixup(fixup)|PSW_ADDR_AMODE; + __ctl_load(cr0, 0, 0); +} + +static noinline __init void setup_lowcore_early(void) +{ + psw_t psw; + + psw.mask = PSW_MASK_BASE | PSW_DEFAULT_KEY | PSW_MASK_EA | PSW_MASK_BA; + psw.addr = PSW_ADDR_AMODE | (unsigned long) s390_base_ext_handler; + S390_lowcore.external_new_psw = psw; + psw.addr = PSW_ADDR_AMODE | (unsigned long) s390_base_pgm_handler; + S390_lowcore.program_new_psw = psw; + s390_base_pgm_handler_fn = early_pgm_check_handler; +} + +static noinline __init void setup_facility_list(void) +{ + stfle(S390_lowcore.stfle_fac_list, + ARRAY_SIZE(S390_lowcore.stfle_fac_list)); +} + +static __init void detect_diag9c(void) +{ + unsigned int cpu_address; + int rc; + + cpu_address = stap(); + asm volatile( + " diag %2,0,0x9c\n" + "0: la %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "=d" (rc) : "0" (-EOPNOTSUPP), "d" (cpu_address) : "cc"); + if (!rc) + S390_lowcore.machine_flags |= MACHINE_FLAG_DIAG9C; +} + +static __init void detect_diag44(void) +{ + int rc; + + asm volatile( + " diag 0,0,0x44\n" + "0: la %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "=d" (rc) : "0" (-EOPNOTSUPP) : "cc"); + if (!rc) + S390_lowcore.machine_flags |= MACHINE_FLAG_DIAG44; +} + +static __init void detect_machine_facilities(void) +{ + if (test_facility(8)) { + S390_lowcore.machine_flags |= MACHINE_FLAG_EDAT1; + __ctl_set_bit(0, 23); + } + if (test_facility(78)) + S390_lowcore.machine_flags |= MACHINE_FLAG_EDAT2; + if (test_facility(3)) + S390_lowcore.machine_flags |= MACHINE_FLAG_IDTE; + if (test_facility(40)) + S390_lowcore.machine_flags |= MACHINE_FLAG_LPP; + if (test_facility(50) && test_facility(73)) + S390_lowcore.machine_flags |= MACHINE_FLAG_TE; + if (test_facility(51)) + S390_lowcore.machine_flags |= MACHINE_FLAG_TLB_LC; + if (test_facility(129)) + S390_lowcore.machine_flags |= MACHINE_FLAG_VX; +} + +static int __init cad_setup(char *str) +{ + int val; + + get_option(&str, &val); + if (val && test_facility(128)) + S390_lowcore.machine_flags |= MACHINE_FLAG_CAD; + return 0; +} +early_param("cad", cad_setup); + +static int __init cad_init(void) +{ + if (MACHINE_HAS_CAD) + /* Enable problem state CAD. */ + __ctl_set_bit(2, 3); + return 0; +} +early_initcall(cad_init); + +static __init void rescue_initrd(void) +{ +#ifdef CONFIG_BLK_DEV_INITRD + unsigned long min_initrd_addr = (unsigned long) _end + (4UL << 20); + /* + * Just like in case of IPL from VM reader we make sure there is a + * gap of 4MB between end of kernel and start of initrd. + * That way we can also be sure that saving an NSS will succeed, + * which however only requires different segments. + */ + if (!INITRD_START || !INITRD_SIZE) + return; + if (INITRD_START >= min_initrd_addr) + return; + memmove((void *) min_initrd_addr, (void *) INITRD_START, INITRD_SIZE); + INITRD_START = min_initrd_addr; +#endif +} + +/* Set up boot command line */ +static void __init append_to_cmdline(size_t (*ipl_data)(char *, size_t)) +{ + char *parm, *delim; + size_t rc, len; + + len = strlen(boot_command_line); + + delim = boot_command_line + len; /* '\0' character position */ + parm = boot_command_line + len + 1; /* append right after '\0' */ + + rc = ipl_data(parm, COMMAND_LINE_SIZE - len - 1); + if (rc) { + if (*parm == '=') + memmove(boot_command_line, parm + 1, rc); + else + *delim = ' '; /* replace '\0' with space */ + } +} + +static inline int has_ebcdic_char(const char *str) +{ + int i; + + for (i = 0; str[i]; i++) + if (str[i] & 0x80) + return 1; + return 0; +} + +static void __init setup_boot_command_line(void) +{ + COMMAND_LINE[ARCH_COMMAND_LINE_SIZE - 1] = 0; + /* convert arch command line to ascii if necessary */ + if (has_ebcdic_char(COMMAND_LINE)) + EBCASC(COMMAND_LINE, ARCH_COMMAND_LINE_SIZE); + /* copy arch command line */ + strlcpy(boot_command_line, strstrip(COMMAND_LINE), + ARCH_COMMAND_LINE_SIZE); + + /* append IPL PARM data to the boot command line */ + if (MACHINE_IS_VM) + append_to_cmdline(append_ipl_vmparm); + + append_to_cmdline(append_ipl_scpdata); +} + +/* + * Save ipl parameters, clear bss memory, initialize storage keys + * and create a kernel NSS at startup if the SAVESYS= parm is defined + */ +void __init startup_init(void) +{ + reset_tod_clock(); + ipl_save_parameters(); + rescue_initrd(); + clear_bss_section(); + init_kernel_storage_key(); + lockdep_init(); + lockdep_off(); + setup_lowcore_early(); + setup_facility_list(); + detect_machine_type(); + ipl_update_parameters(); + setup_boot_command_line(); + create_kernel_nss(); + detect_diag9c(); + detect_diag44(); + detect_machine_facilities(); + setup_topology(); + sclp_early_detect(); + lockdep_on(); +} diff --git a/arch/s390/kernel/ebcdic.c b/arch/s390/kernel/ebcdic.c new file mode 100644 index 000000000..b971c6be6 --- /dev/null +++ b/arch/s390/kernel/ebcdic.c @@ -0,0 +1,400 @@ +/* + * ECBDIC -> ASCII, ASCII -> ECBDIC, + * upper to lower case (EBCDIC) conversion tables. + * + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + * Martin Peschke <peschke@fh-brandenburg.de> + */ + +#include <linux/module.h> +#include <asm/types.h> +#include <asm/ebcdic.h> + +/* + * ASCII (IBM PC 437) -> EBCDIC 037 + */ +__u8 _ascebc[256] = +{ + /*00 NUL SOH STX ETX EOT ENQ ACK BEL */ + 0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F, + /*08 BS HT LF VT FF CR SO SI */ + /* ->NL */ + 0x16, 0x05, 0x15, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, + /*10 DLE DC1 DC2 DC3 DC4 NAK SYN ETB */ + 0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26, + /*18 CAN EM SUB ESC FS GS RS US */ + /* ->IGS ->IRS ->IUS */ + 0x18, 0x19, 0x3F, 0x27, 0x22, 0x1D, 0x1E, 0x1F, + /*20 SP ! " # $ % & ' */ + 0x40, 0x5A, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D, + /*28 ( ) * + , - . / */ + 0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61, + /*30 0 1 2 3 4 5 6 7 */ + 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, + /*38 8 9 : ; < = > ? */ + 0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F, + /*40 @ A B C D E F G */ + 0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, + /*48 H I J K L M N O */ + 0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, + /*50 P Q R S T U V W */ + 0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, + /*58 X Y Z [ \ ] ^ _ */ + 0xE7, 0xE8, 0xE9, 0xBA, 0xE0, 0xBB, 0xB0, 0x6D, + /*60 ` a b c d e f g */ + 0x79, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, + /*68 h i j k l m n o */ + 0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, + /*70 p q r s t u v w */ + 0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, + /*78 x y z { | } ~ DL */ + 0xA7, 0xA8, 0xA9, 0xC0, 0x4F, 0xD0, 0xA1, 0x07, + /*80*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*88*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*90*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*98*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*A0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*A8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*B0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*B8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*C0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*C8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*D0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*D8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*E0 sz */ + 0x3F, 0x59, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*E8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*F0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*F8*/ + 0x90, 0x3F, 0x3F, 0x3F, 0x3F, 0xEA, 0x3F, 0xFF +}; + +/* + * EBCDIC 037 -> ASCII (IBM PC 437) + */ +__u8 _ebcasc[256] = +{ + /* 0x00 NUL SOH STX ETX *SEL HT *RNL DEL */ + 0x00, 0x01, 0x02, 0x03, 0x07, 0x09, 0x07, 0x7F, + /* 0x08 -GE -SPS -RPT VT FF CR SO SI */ + 0x07, 0x07, 0x07, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, + /* 0x10 DLE DC1 DC2 DC3 -RES -NL BS -POC + -ENP ->LF */ + 0x10, 0x11, 0x12, 0x13, 0x07, 0x0A, 0x08, 0x07, + /* 0x18 CAN EM -UBS -CU1 -IFS -IGS -IRS -ITB + -IUS */ + 0x18, 0x19, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, + /* 0x20 -DS -SOS FS -WUS -BYP LF ETB ESC + -INP */ + 0x07, 0x07, 0x1C, 0x07, 0x07, 0x0A, 0x17, 0x1B, + /* 0x28 -SA -SFE -SM -CSP -MFA ENQ ACK BEL + -SW */ + 0x07, 0x07, 0x07, 0x07, 0x07, 0x05, 0x06, 0x07, + /* 0x30 ---- ---- SYN -IR -PP -TRN -NBS EOT */ + 0x07, 0x07, 0x16, 0x07, 0x07, 0x07, 0x07, 0x04, + /* 0x38 -SBS -IT -RFF -CU3 DC4 NAK ---- SUB */ + 0x07, 0x07, 0x07, 0x07, 0x14, 0x15, 0x07, 0x1A, + /* 0x40 SP RSP ä ---- */ + 0x20, 0xFF, 0x83, 0x84, 0x85, 0xA0, 0x07, 0x86, + /* 0x48 . < ( + | */ + 0x87, 0xA4, 0x9B, 0x2E, 0x3C, 0x28, 0x2B, 0x7C, + /* 0x50 & ---- */ + 0x26, 0x82, 0x88, 0x89, 0x8A, 0xA1, 0x8C, 0x07, + /* 0x58 ß ! $ * ) ; */ + 0x8D, 0xE1, 0x21, 0x24, 0x2A, 0x29, 0x3B, 0xAA, + /* 0x60 - / ---- Ä ---- ---- ---- */ + 0x2D, 0x2F, 0x07, 0x8E, 0x07, 0x07, 0x07, 0x8F, + /* 0x68 ---- , % _ > ? */ + 0x80, 0xA5, 0x07, 0x2C, 0x25, 0x5F, 0x3E, 0x3F, + /* 0x70 ---- ---- ---- ---- ---- ---- ---- */ + 0x07, 0x90, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, + /* 0x78 * ` : # @ ' = " */ + 0x70, 0x60, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22, + /* 0x80 * a b c d e f g */ + 0x07, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, + /* 0x88 h i ---- ---- ---- */ + 0x68, 0x69, 0xAE, 0xAF, 0x07, 0x07, 0x07, 0xF1, + /* 0x90 ° j k l m n o p */ + 0xF8, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, + /* 0x98 q r ---- ---- */ + 0x71, 0x72, 0xA6, 0xA7, 0x91, 0x07, 0x92, 0x07, + /* 0xA0 ~ s t u v w x */ + 0xE6, 0x7E, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, + /* 0xA8 y z ---- ---- ---- ---- */ + 0x79, 0x7A, 0xAD, 0xAB, 0x07, 0x07, 0x07, 0x07, + /* 0xB0 ^ ---- § ---- */ + 0x5E, 0x9C, 0x9D, 0xFA, 0x07, 0x07, 0x07, 0xAC, + /* 0xB8 ---- [ ] ---- ---- ---- ---- */ + 0xAB, 0x07, 0x5B, 0x5D, 0x07, 0x07, 0x07, 0x07, + /* 0xC0 { A B C D E F G */ + 0x7B, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, + /* 0xC8 H I ---- ö ---- */ + 0x48, 0x49, 0x07, 0x93, 0x94, 0x95, 0xA2, 0x07, + /* 0xD0 } J K L M N O P */ + 0x7D, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, + /* 0xD8 Q R ---- ü */ + 0x51, 0x52, 0x07, 0x96, 0x81, 0x97, 0xA3, 0x98, + /* 0xE0 \ S T U V W X */ + 0x5C, 0xF6, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, + /* 0xE8 Y Z ---- Ö ---- ---- ---- */ + 0x59, 0x5A, 0xFD, 0x07, 0x99, 0x07, 0x07, 0x07, + /* 0xF0 0 1 2 3 4 5 6 7 */ + 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, + /* 0xF8 8 9 ---- ---- Ü ---- ---- ---- */ + 0x38, 0x39, 0x07, 0x07, 0x9A, 0x07, 0x07, 0x07 +}; + + +/* + * ASCII (IBM PC 437) -> EBCDIC 500 + */ +__u8 _ascebc_500[256] = +{ + /*00 NUL SOH STX ETX EOT ENQ ACK BEL */ + 0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F, + /*08 BS HT LF VT FF CR SO SI */ + /* ->NL */ + 0x16, 0x05, 0x15, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, + /*10 DLE DC1 DC2 DC3 DC4 NAK SYN ETB */ + 0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26, + /*18 CAN EM SUB ESC FS GS RS US */ + /* ->IGS ->IRS ->IUS */ + 0x18, 0x19, 0x3F, 0x27, 0x22, 0x1D, 0x1E, 0x1F, + /*20 SP ! " # $ % & ' */ + 0x40, 0x4F, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D, + /*28 ( ) * + , - . / */ + 0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61, + /*30 0 1 2 3 4 5 6 7 */ + 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, + /*38 8 9 : ; < = > ? */ + 0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F, + /*40 @ A B C D E F G */ + 0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, + /*48 H I J K L M N O */ + 0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, + /*50 P Q R S T U V W */ + 0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, + /*58 X Y Z [ \ ] ^ _ */ + 0xE7, 0xE8, 0xE9, 0x4A, 0xE0, 0x5A, 0x5F, 0x6D, + /*60 ` a b c d e f g */ + 0x79, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, + /*68 h i j k l m n o */ + 0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, + /*70 p q r s t u v w */ + 0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, + /*78 x y z { | } ~ DL */ + 0xA7, 0xA8, 0xA9, 0xC0, 0xBB, 0xD0, 0xA1, 0x07, + /*80*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*88*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*90*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*98*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*A0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*A8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*B0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*B8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*C0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*C8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*D0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*D8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*E0 sz */ + 0x3F, 0x59, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*E8*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*F0*/ + 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, + /*F8*/ + 0x90, 0x3F, 0x3F, 0x3F, 0x3F, 0xEA, 0x3F, 0xFF +}; + +/* + * EBCDIC 500 -> ASCII (IBM PC 437) + */ +__u8 _ebcasc_500[256] = +{ + /* 0x00 NUL SOH STX ETX *SEL HT *RNL DEL */ + 0x00, 0x01, 0x02, 0x03, 0x07, 0x09, 0x07, 0x7F, + /* 0x08 -GE -SPS -RPT VT FF CR SO SI */ + 0x07, 0x07, 0x07, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, + /* 0x10 DLE DC1 DC2 DC3 -RES -NL BS -POC + -ENP ->LF */ + 0x10, 0x11, 0x12, 0x13, 0x07, 0x0A, 0x08, 0x07, + /* 0x18 CAN EM -UBS -CU1 -IFS -IGS -IRS -ITB + -IUS */ + 0x18, 0x19, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, + /* 0x20 -DS -SOS FS -WUS -BYP LF ETB ESC + -INP */ + 0x07, 0x07, 0x1C, 0x07, 0x07, 0x0A, 0x17, 0x1B, + /* 0x28 -SA -SFE -SM -CSP -MFA ENQ ACK BEL + -SW */ + 0x07, 0x07, 0x07, 0x07, 0x07, 0x05, 0x06, 0x07, + /* 0x30 ---- ---- SYN -IR -PP -TRN -NBS EOT */ + 0x07, 0x07, 0x16, 0x07, 0x07, 0x07, 0x07, 0x04, + /* 0x38 -SBS -IT -RFF -CU3 DC4 NAK ---- SUB */ + 0x07, 0x07, 0x07, 0x07, 0x14, 0x15, 0x07, 0x1A, + /* 0x40 SP RSP ä ---- */ + 0x20, 0xFF, 0x83, 0x84, 0x85, 0xA0, 0x07, 0x86, + /* 0x48 [ . < ( + ! */ + 0x87, 0xA4, 0x5B, 0x2E, 0x3C, 0x28, 0x2B, 0x21, + /* 0x50 & ---- */ + 0x26, 0x82, 0x88, 0x89, 0x8A, 0xA1, 0x8C, 0x07, + /* 0x58 ß ] $ * ) ; ^ */ + 0x8D, 0xE1, 0x5D, 0x24, 0x2A, 0x29, 0x3B, 0x5E, + /* 0x60 - / ---- Ä ---- ---- ---- */ + 0x2D, 0x2F, 0x07, 0x8E, 0x07, 0x07, 0x07, 0x8F, + /* 0x68 ---- , % _ > ? */ + 0x80, 0xA5, 0x07, 0x2C, 0x25, 0x5F, 0x3E, 0x3F, + /* 0x70 ---- ---- ---- ---- ---- ---- ---- */ + 0x07, 0x90, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, + /* 0x78 * ` : # @ ' = " */ + 0x70, 0x60, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22, + /* 0x80 * a b c d e f g */ + 0x07, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, + /* 0x88 h i ---- ---- ---- */ + 0x68, 0x69, 0xAE, 0xAF, 0x07, 0x07, 0x07, 0xF1, + /* 0x90 ° j k l m n o p */ + 0xF8, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, + /* 0x98 q r ---- ---- */ + 0x71, 0x72, 0xA6, 0xA7, 0x91, 0x07, 0x92, 0x07, + /* 0xA0 ~ s t u v w x */ + 0xE6, 0x7E, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, + /* 0xA8 y z ---- ---- ---- ---- */ + 0x79, 0x7A, 0xAD, 0xAB, 0x07, 0x07, 0x07, 0x07, + /* 0xB0 ---- § ---- */ + 0x9B, 0x9C, 0x9D, 0xFA, 0x07, 0x07, 0x07, 0xAC, + /* 0xB8 ---- | ---- ---- ---- ---- */ + 0xAB, 0x07, 0xAA, 0x7C, 0x07, 0x07, 0x07, 0x07, + /* 0xC0 { A B C D E F G */ + 0x7B, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, + /* 0xC8 H I ---- ö ---- */ + 0x48, 0x49, 0x07, 0x93, 0x94, 0x95, 0xA2, 0x07, + /* 0xD0 } J K L M N O P */ + 0x7D, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, + /* 0xD8 Q R ---- ü */ + 0x51, 0x52, 0x07, 0x96, 0x81, 0x97, 0xA3, 0x98, + /* 0xE0 \ S T U V W X */ + 0x5C, 0xF6, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, + /* 0xE8 Y Z ---- Ö ---- ---- ---- */ + 0x59, 0x5A, 0xFD, 0x07, 0x99, 0x07, 0x07, 0x07, + /* 0xF0 0 1 2 3 4 5 6 7 */ + 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, + /* 0xF8 8 9 ---- ---- Ü ---- ---- ---- */ + 0x38, 0x39, 0x07, 0x07, 0x9A, 0x07, 0x07, 0x07 +}; + + +/* + * EBCDIC 037/500 conversion table: + * from upper to lower case + */ +__u8 _ebc_tolower[256] = +{ + 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, + 0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, + 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, + 0x18, 0x19, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x1F, + 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, + 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F, + 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, + 0x38, 0x39, 0x3A, 0x3B, 0x3C, 0x3D, 0x3E, 0x3F, + 0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, + 0x48, 0x49, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, + 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, + 0x58, 0x59, 0x5A, 0x5B, 0x5C, 0x5D, 0x5E, 0x5F, + 0x60, 0x61, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, + 0x48, 0x49, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, + 0x70, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, + 0x58, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F, + 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, + 0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x8D, 0x8E, 0x8F, + 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, + 0x98, 0x99, 0x9A, 0x9B, 0x9C, 0x9D, 0x9C, 0x9F, + 0xA0, 0xA1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, + 0xA8, 0xA9, 0xAA, 0xAB, 0x8C, 0x8D, 0x8E, 0xAF, + 0xB0, 0xB1, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7, + 0xB8, 0xB9, 0xBA, 0xBB, 0xBC, 0xBD, 0xBE, 0xBF, + 0xC0, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, + 0x88, 0x89, 0xCA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF, + 0xD0, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, + 0x98, 0x99, 0xDA, 0xDB, 0xDC, 0xDD, 0xDE, 0xDF, + 0xE0, 0xE1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, + 0xA8, 0xA9, 0xEA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF, + 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, + 0xF8, 0xF9, 0xFA, 0xDB, 0xDC, 0xDD, 0xDE, 0xFF +}; + + +/* + * EBCDIC 037/500 conversion table: + * from lower to upper case + */ +__u8 _ebc_toupper[256] = +{ + 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, + 0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, + 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, + 0x18, 0x19, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x1F, + 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, + 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F, + 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, + 0x38, 0x39, 0x3A, 0x3B, 0x3C, 0x3D, 0x3E, 0x3F, + 0x40, 0x41, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, + 0x68, 0x69, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, + 0x50, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, + 0x78, 0x59, 0x5A, 0x5B, 0x5C, 0x5D, 0x5E, 0x5F, + 0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, + 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, + 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, + 0x78, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F, + 0x80, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, + 0xC8, 0xC9, 0x8A, 0x8B, 0xAC, 0xAD, 0xAE, 0x8F, + 0x90, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, 0xD7, + 0xD8, 0xD9, 0x9A, 0x9B, 0x9E, 0x9D, 0x9E, 0x9F, + 0xA0, 0xA1, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, + 0xE8, 0xE9, 0xAA, 0xAB, 0xAC, 0xAD, 0xAE, 0xAF, + 0xB0, 0xB1, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7, + 0xB8, 0xB9, 0xBA, 0xBB, 0xBC, 0xBD, 0xBE, 0xBF, + 0xC0, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, + 0xC8, 0xC9, 0xCA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF, + 0xD0, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, 0xD7, + 0xD8, 0xD9, 0xDA, 0xFB, 0xFC, 0xFD, 0xFE, 0xDF, + 0xE0, 0xE1, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, + 0xE8, 0xE9, 0xEA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF, + 0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, + 0xF8, 0xF9, 0xFA, 0xFB, 0xFC, 0xFD, 0xFE, 0xFF +}; + +EXPORT_SYMBOL(_ascebc_500); +EXPORT_SYMBOL(_ebcasc_500); +EXPORT_SYMBOL(_ascebc); +EXPORT_SYMBOL(_ebcasc); +EXPORT_SYMBOL(_ebc_tolower); +EXPORT_SYMBOL(_ebc_toupper); + diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S new file mode 100644 index 000000000..99b44acbf --- /dev/null +++ b/arch/s390/kernel/entry.S @@ -0,0 +1,1059 @@ +/* + * S390 low-level entry points. + * + * Copyright IBM Corp. 1999, 2012 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + * Hartmut Penner (hp@de.ibm.com), + * Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com), + * Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#include <linux/init.h> +#include <linux/linkage.h> +#include <asm/processor.h> +#include <asm/cache.h> +#include <asm/errno.h> +#include <asm/ptrace.h> +#include <asm/thread_info.h> +#include <asm/asm-offsets.h> +#include <asm/unistd.h> +#include <asm/page.h> +#include <asm/sigp.h> +#include <asm/irq.h> + +__PT_R0 = __PT_GPRS +__PT_R1 = __PT_GPRS + 8 +__PT_R2 = __PT_GPRS + 16 +__PT_R3 = __PT_GPRS + 24 +__PT_R4 = __PT_GPRS + 32 +__PT_R5 = __PT_GPRS + 40 +__PT_R6 = __PT_GPRS + 48 +__PT_R7 = __PT_GPRS + 56 +__PT_R8 = __PT_GPRS + 64 +__PT_R9 = __PT_GPRS + 72 +__PT_R10 = __PT_GPRS + 80 +__PT_R11 = __PT_GPRS + 88 +__PT_R12 = __PT_GPRS + 96 +__PT_R13 = __PT_GPRS + 104 +__PT_R14 = __PT_GPRS + 112 +__PT_R15 = __PT_GPRS + 120 + +STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER +STACK_SIZE = 1 << STACK_SHIFT +STACK_INIT = STACK_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE + +_TIF_WORK = (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \ + _TIF_UPROBE) +_TIF_TRACE = (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \ + _TIF_SYSCALL_TRACEPOINT) +_CIF_WORK = (_CIF_MCCK_PENDING | _CIF_ASCE) +_PIF_WORK = (_PIF_PER_TRAP) + +#define BASED(name) name-system_call(%r13) + + .macro TRACE_IRQS_ON +#ifdef CONFIG_TRACE_IRQFLAGS + basr %r2,%r0 + brasl %r14,trace_hardirqs_on_caller +#endif + .endm + + .macro TRACE_IRQS_OFF +#ifdef CONFIG_TRACE_IRQFLAGS + basr %r2,%r0 + brasl %r14,trace_hardirqs_off_caller +#endif + .endm + + .macro LOCKDEP_SYS_EXIT +#ifdef CONFIG_LOCKDEP + tm __PT_PSW+1(%r11),0x01 # returning to user ? + jz .+10 + brasl %r14,lockdep_sys_exit +#endif + .endm + + .macro LPP newpp +#if IS_ENABLED(CONFIG_KVM) + tm __LC_MACHINE_FLAGS+6,0x20 # MACHINE_FLAG_LPP + jz .+8 + .insn s,0xb2800000,\newpp +#endif + .endm + + .macro HANDLE_SIE_INTERCEPT scratch,reason +#if IS_ENABLED(CONFIG_KVM) + tmhh %r8,0x0001 # interrupting from user ? + jnz .+62 + lgr \scratch,%r9 + slg \scratch,BASED(.Lsie_critical) + clg \scratch,BASED(.Lsie_critical_length) + .if \reason==1 + # Some program interrupts are suppressing (e.g. protection). + # We must also check the instruction after SIE in that case. + # do_protection_exception will rewind to .Lrewind_pad + jh .+42 + .else + jhe .+42 + .endif + lg %r14,__SF_EMPTY(%r15) # get control block pointer + LPP __SF_EMPTY+16(%r15) # set host id + ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE + lctlg %c1,%c1,__LC_USER_ASCE # load primary asce + larl %r9,sie_exit # skip forward to sie_exit + mvi __SF_EMPTY+31(%r15),\reason # set exit reason +#endif + .endm + + .macro CHECK_STACK stacksize,savearea +#ifdef CONFIG_CHECK_STACK + tml %r15,\stacksize - CONFIG_STACK_GUARD + lghi %r14,\savearea + jz stack_overflow +#endif + .endm + + .macro SWITCH_ASYNC savearea,stack,shift + tmhh %r8,0x0001 # interrupting from user ? + jnz 1f + lgr %r14,%r9 + slg %r14,BASED(.Lcritical_start) + clg %r14,BASED(.Lcritical_length) + jhe 0f + lghi %r11,\savearea # inside critical section, do cleanup + brasl %r14,cleanup_critical + tmhh %r8,0x0001 # retest problem state after cleanup + jnz 1f +0: lg %r14,\stack # are we already on the target stack? + slgr %r14,%r15 + srag %r14,%r14,\shift + jnz 1f + CHECK_STACK 1<<\shift,\savearea + aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) + j 2f +1: lg %r15,\stack # load target stack +2: la %r11,STACK_FRAME_OVERHEAD(%r15) + .endm + + .macro UPDATE_VTIME scratch,enter_timer + lg \scratch,__LC_EXIT_TIMER + slg \scratch,\enter_timer + alg \scratch,__LC_USER_TIMER + stg \scratch,__LC_USER_TIMER + lg \scratch,__LC_LAST_UPDATE_TIMER + slg \scratch,__LC_EXIT_TIMER + alg \scratch,__LC_SYSTEM_TIMER + stg \scratch,__LC_SYSTEM_TIMER + mvc __LC_LAST_UPDATE_TIMER(8),\enter_timer + .endm + + .macro LAST_BREAK scratch + srag \scratch,%r10,23 + jz .+10 + stg %r10,__TI_last_break(%r12) + .endm + + .macro REENABLE_IRQS + stg %r8,__LC_RETURN_PSW + ni __LC_RETURN_PSW,0xbf + ssm __LC_RETURN_PSW + .endm + + .macro STCK savearea +#ifdef CONFIG_HAVE_MARCH_Z9_109_FEATURES + .insn s,0xb27c0000,\savearea # store clock fast +#else + .insn s,0xb2050000,\savearea # store clock +#endif + .endm + + .section .kprobes.text, "ax" + +/* + * Scheduler resume function, called by switch_to + * gpr2 = (task_struct *) prev + * gpr3 = (task_struct *) next + * Returns: + * gpr2 = prev + */ +ENTRY(__switch_to) + stmg %r6,%r15,__SF_GPRS(%r15) # store gprs of prev task + stg %r15,__THREAD_ksp(%r2) # store kernel stack of prev + lg %r4,__THREAD_info(%r2) # get thread_info of prev + lg %r5,__THREAD_info(%r3) # get thread_info of next + lgr %r15,%r5 + aghi %r15,STACK_INIT # end of kernel stack of next + stg %r3,__LC_CURRENT # store task struct of next + stg %r5,__LC_THREAD_INFO # store thread info of next + stg %r15,__LC_KERNEL_STACK # store end of kernel stack + lctl %c4,%c4,__TASK_pid(%r3) # load pid to control reg. 4 + mvc __LC_CURRENT_PID+4(4,%r0),__TASK_pid(%r3) # store pid of next + lg %r15,__THREAD_ksp(%r3) # load kernel stack of next + lmg %r6,%r15,__SF_GPRS(%r15) # load gprs of next task + br %r14 + +.L__critical_start: +/* + * SVC interrupt handler routine. System calls are synchronous events and + * are executed with interrupts enabled. + */ + +ENTRY(system_call) + stpt __LC_SYNC_ENTER_TIMER +.Lsysc_stmg: + stmg %r8,%r15,__LC_SAVE_AREA_SYNC + lg %r10,__LC_LAST_BREAK + lg %r12,__LC_THREAD_INFO + lghi %r14,_PIF_SYSCALL +.Lsysc_per: + lg %r15,__LC_KERNEL_STACK + la %r11,STACK_FRAME_OVERHEAD(%r15) # pointer to pt_regs +.Lsysc_vtime: + UPDATE_VTIME %r13,__LC_SYNC_ENTER_TIMER + LAST_BREAK %r13 + stmg %r0,%r7,__PT_R0(%r11) + mvc __PT_R8(64,%r11),__LC_SAVE_AREA_SYNC + mvc __PT_PSW(16,%r11),__LC_SVC_OLD_PSW + mvc __PT_INT_CODE(4,%r11),__LC_SVC_ILC + stg %r14,__PT_FLAGS(%r11) +.Lsysc_do_svc: + lg %r10,__TI_sysc_table(%r12) # address of system call table + llgh %r8,__PT_INT_CODE+2(%r11) + slag %r8,%r8,2 # shift and test for svc 0 + jnz .Lsysc_nr_ok + # svc 0: system call number in %r1 + llgfr %r1,%r1 # clear high word in r1 + cghi %r1,NR_syscalls + jnl .Lsysc_nr_ok + sth %r1,__PT_INT_CODE+2(%r11) + slag %r8,%r1,2 +.Lsysc_nr_ok: + xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + stg %r2,__PT_ORIG_GPR2(%r11) + stg %r7,STACK_FRAME_OVERHEAD(%r15) + lgf %r9,0(%r8,%r10) # get system call add. + tm __TI_flags+7(%r12),_TIF_TRACE + jnz .Lsysc_tracesys + basr %r14,%r9 # call sys_xxxx + stg %r2,__PT_R2(%r11) # store return value + +.Lsysc_return: + LOCKDEP_SYS_EXIT +.Lsysc_tif: + tm __PT_PSW+1(%r11),0x01 # returning to user ? + jno .Lsysc_restore + tm __PT_FLAGS+7(%r11),_PIF_WORK + jnz .Lsysc_work + tm __TI_flags+7(%r12),_TIF_WORK + jnz .Lsysc_work # check for work + tm __LC_CPU_FLAGS+7,_CIF_WORK + jnz .Lsysc_work +.Lsysc_restore: + lg %r14,__LC_VDSO_PER_CPU + lmg %r0,%r10,__PT_R0(%r11) + mvc __LC_RETURN_PSW(16),__PT_PSW(%r11) + stpt __LC_EXIT_TIMER + mvc __VDSO_ECTG_BASE(16,%r14),__LC_EXIT_TIMER + lmg %r11,%r15,__PT_R11(%r11) + lpswe __LC_RETURN_PSW +.Lsysc_done: + +# +# One of the work bits is on. Find out which one. +# +.Lsysc_work: + tm __LC_CPU_FLAGS+7,_CIF_MCCK_PENDING + jo .Lsysc_mcck_pending + tm __TI_flags+7(%r12),_TIF_NEED_RESCHED + jo .Lsysc_reschedule +#ifdef CONFIG_UPROBES + tm __TI_flags+7(%r12),_TIF_UPROBE + jo .Lsysc_uprobe_notify +#endif + tm __PT_FLAGS+7(%r11),_PIF_PER_TRAP + jo .Lsysc_singlestep + tm __TI_flags+7(%r12),_TIF_SIGPENDING + jo .Lsysc_sigpending + tm __TI_flags+7(%r12),_TIF_NOTIFY_RESUME + jo .Lsysc_notify_resume + tm __LC_CPU_FLAGS+7,_CIF_ASCE + jo .Lsysc_uaccess + j .Lsysc_return # beware of critical section cleanup + +# +# _TIF_NEED_RESCHED is set, call schedule +# +.Lsysc_reschedule: + larl %r14,.Lsysc_return + jg schedule + +# +# _CIF_MCCK_PENDING is set, call handler +# +.Lsysc_mcck_pending: + larl %r14,.Lsysc_return + jg s390_handle_mcck # TIF bit will be cleared by handler + +# +# _CIF_ASCE is set, load user space asce +# +.Lsysc_uaccess: + ni __LC_CPU_FLAGS+7,255-_CIF_ASCE + lctlg %c1,%c1,__LC_USER_ASCE # load primary asce + j .Lsysc_return + +# +# _TIF_SIGPENDING is set, call do_signal +# +.Lsysc_sigpending: + lgr %r2,%r11 # pass pointer to pt_regs + brasl %r14,do_signal + tm __PT_FLAGS+7(%r11),_PIF_SYSCALL + jno .Lsysc_return + lmg %r2,%r7,__PT_R2(%r11) # load svc arguments + lg %r10,__TI_sysc_table(%r12) # address of system call table + lghi %r8,0 # svc 0 returns -ENOSYS + llgh %r1,__PT_INT_CODE+2(%r11) # load new svc number + cghi %r1,NR_syscalls + jnl .Lsysc_nr_ok # invalid svc number -> do svc 0 + slag %r8,%r1,2 + j .Lsysc_nr_ok # restart svc + +# +# _TIF_NOTIFY_RESUME is set, call do_notify_resume +# +.Lsysc_notify_resume: + lgr %r2,%r11 # pass pointer to pt_regs + larl %r14,.Lsysc_return + jg do_notify_resume + +# +# _TIF_UPROBE is set, call uprobe_notify_resume +# +#ifdef CONFIG_UPROBES +.Lsysc_uprobe_notify: + lgr %r2,%r11 # pass pointer to pt_regs + larl %r14,.Lsysc_return + jg uprobe_notify_resume +#endif + +# +# _PIF_PER_TRAP is set, call do_per_trap +# +.Lsysc_singlestep: + ni __PT_FLAGS+7(%r11),255-_PIF_PER_TRAP + lgr %r2,%r11 # pass pointer to pt_regs + larl %r14,.Lsysc_return + jg do_per_trap + +# +# call tracehook_report_syscall_entry/tracehook_report_syscall_exit before +# and after the system call +# +.Lsysc_tracesys: + lgr %r2,%r11 # pass pointer to pt_regs + la %r3,0 + llgh %r0,__PT_INT_CODE+2(%r11) + stg %r0,__PT_R2(%r11) + brasl %r14,do_syscall_trace_enter + lghi %r0,NR_syscalls + clgr %r0,%r2 + jnh .Lsysc_tracenogo + sllg %r8,%r2,2 + lgf %r9,0(%r8,%r10) +.Lsysc_tracego: + lmg %r3,%r7,__PT_R3(%r11) + stg %r7,STACK_FRAME_OVERHEAD(%r15) + lg %r2,__PT_ORIG_GPR2(%r11) + basr %r14,%r9 # call sys_xxx + stg %r2,__PT_R2(%r11) # store return value +.Lsysc_tracenogo: + tm __TI_flags+7(%r12),_TIF_TRACE + jz .Lsysc_return + lgr %r2,%r11 # pass pointer to pt_regs + larl %r14,.Lsysc_return + jg do_syscall_trace_exit + +# +# a new process exits the kernel with ret_from_fork +# +ENTRY(ret_from_fork) + la %r11,STACK_FRAME_OVERHEAD(%r15) + lg %r12,__LC_THREAD_INFO + brasl %r14,schedule_tail + TRACE_IRQS_ON + ssm __LC_SVC_NEW_PSW # reenable interrupts + tm __PT_PSW+1(%r11),0x01 # forking a kernel thread ? + jne .Lsysc_tracenogo + # it's a kernel thread + lmg %r9,%r10,__PT_R9(%r11) # load gprs +ENTRY(kernel_thread_starter) + la %r2,0(%r10) + basr %r14,%r9 + j .Lsysc_tracenogo + +/* + * Program check handler routine + */ + +ENTRY(pgm_check_handler) + stpt __LC_SYNC_ENTER_TIMER + stmg %r8,%r15,__LC_SAVE_AREA_SYNC + lg %r10,__LC_LAST_BREAK + lg %r12,__LC_THREAD_INFO + larl %r13,system_call + lmg %r8,%r9,__LC_PGM_OLD_PSW + HANDLE_SIE_INTERCEPT %r14,1 + tmhh %r8,0x0001 # test problem state bit + jnz 1f # -> fault in user space + tmhh %r8,0x4000 # PER bit set in old PSW ? + jnz 0f # -> enabled, can't be a double fault + tm __LC_PGM_ILC+3,0x80 # check for per exception + jnz .Lpgm_svcper # -> single stepped svc +0: CHECK_STACK STACK_SIZE,__LC_SAVE_AREA_SYNC + aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) + j 2f +1: UPDATE_VTIME %r14,__LC_SYNC_ENTER_TIMER + LAST_BREAK %r14 + lg %r15,__LC_KERNEL_STACK + lg %r14,__TI_task(%r12) + lghi %r13,__LC_PGM_TDB + tm __LC_PGM_ILC+2,0x02 # check for transaction abort + jz 2f + mvc __THREAD_trap_tdb(256,%r14),0(%r13) +2: la %r11,STACK_FRAME_OVERHEAD(%r15) + stmg %r0,%r7,__PT_R0(%r11) + mvc __PT_R8(64,%r11),__LC_SAVE_AREA_SYNC + stmg %r8,%r9,__PT_PSW(%r11) + mvc __PT_INT_CODE(4,%r11),__LC_PGM_ILC + mvc __PT_INT_PARM_LONG(8,%r11),__LC_TRANS_EXC_CODE + xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) + stg %r10,__PT_ARGS(%r11) + tm __LC_PGM_ILC+3,0x80 # check for per exception + jz 0f + tmhh %r8,0x0001 # kernel per event ? + jz .Lpgm_kprobe + oi __PT_FLAGS+7(%r11),_PIF_PER_TRAP + mvc __THREAD_per_address(8,%r14),__LC_PER_ADDRESS + mvc __THREAD_per_cause(2,%r14),__LC_PER_CODE + mvc __THREAD_per_paid(1,%r14),__LC_PER_ACCESS_ID +0: REENABLE_IRQS + xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + larl %r1,pgm_check_table + llgh %r10,__PT_INT_CODE+2(%r11) + nill %r10,0x007f + sll %r10,2 + je .Lsysc_return + lgf %r1,0(%r10,%r1) # load address of handler routine + lgr %r2,%r11 # pass pointer to pt_regs + basr %r14,%r1 # branch to interrupt-handler + j .Lsysc_return + +# +# PER event in supervisor state, must be kprobes +# +.Lpgm_kprobe: + REENABLE_IRQS + xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + lgr %r2,%r11 # pass pointer to pt_regs + brasl %r14,do_per_trap + j .Lsysc_return + +# +# single stepped system call +# +.Lpgm_svcper: + mvc __LC_RETURN_PSW(8),__LC_SVC_NEW_PSW + larl %r14,.Lsysc_per + stg %r14,__LC_RETURN_PSW+8 + lghi %r14,_PIF_SYSCALL | _PIF_PER_TRAP + lpswe __LC_RETURN_PSW # branch to .Lsysc_per and enable irqs + +/* + * IO interrupt handler routine + */ +ENTRY(io_int_handler) + STCK __LC_INT_CLOCK + stpt __LC_ASYNC_ENTER_TIMER + stmg %r8,%r15,__LC_SAVE_AREA_ASYNC + lg %r10,__LC_LAST_BREAK + lg %r12,__LC_THREAD_INFO + larl %r13,system_call + lmg %r8,%r9,__LC_IO_OLD_PSW + HANDLE_SIE_INTERCEPT %r14,2 + SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_STACK,STACK_SHIFT + tmhh %r8,0x0001 # interrupting from user? + jz .Lio_skip + UPDATE_VTIME %r14,__LC_ASYNC_ENTER_TIMER + LAST_BREAK %r14 +.Lio_skip: + stmg %r0,%r7,__PT_R0(%r11) + mvc __PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC + stmg %r8,%r9,__PT_PSW(%r11) + mvc __PT_INT_CODE(12,%r11),__LC_SUBCHANNEL_ID + xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) + TRACE_IRQS_OFF + xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) +.Lio_loop: + lgr %r2,%r11 # pass pointer to pt_regs + lghi %r3,IO_INTERRUPT + tm __PT_INT_CODE+8(%r11),0x80 # adapter interrupt ? + jz .Lio_call + lghi %r3,THIN_INTERRUPT +.Lio_call: + brasl %r14,do_IRQ + tm __LC_MACHINE_FLAGS+6,0x10 # MACHINE_FLAG_LPAR + jz .Lio_return + tpi 0 + jz .Lio_return + mvc __PT_INT_CODE(12,%r11),__LC_SUBCHANNEL_ID + j .Lio_loop +.Lio_return: + LOCKDEP_SYS_EXIT + TRACE_IRQS_ON +.Lio_tif: + tm __TI_flags+7(%r12),_TIF_WORK + jnz .Lio_work # there is work to do (signals etc.) + tm __LC_CPU_FLAGS+7,_CIF_WORK + jnz .Lio_work +.Lio_restore: + lg %r14,__LC_VDSO_PER_CPU + lmg %r0,%r10,__PT_R0(%r11) + mvc __LC_RETURN_PSW(16),__PT_PSW(%r11) + stpt __LC_EXIT_TIMER + mvc __VDSO_ECTG_BASE(16,%r14),__LC_EXIT_TIMER + lmg %r11,%r15,__PT_R11(%r11) + lpswe __LC_RETURN_PSW +.Lio_done: + +# +# There is work todo, find out in which context we have been interrupted: +# 1) if we return to user space we can do all _TIF_WORK work +# 2) if we return to kernel code and kvm is enabled check if we need to +# modify the psw to leave SIE +# 3) if we return to kernel code and preemptive scheduling is enabled check +# the preemption counter and if it is zero call preempt_schedule_irq +# Before any work can be done, a switch to the kernel stack is required. +# +.Lio_work: + tm __PT_PSW+1(%r11),0x01 # returning to user ? + jo .Lio_work_user # yes -> do resched & signal +#ifdef CONFIG_PREEMPT + # check for preemptive scheduling + icm %r0,15,__TI_precount(%r12) + jnz .Lio_restore # preemption is disabled + tm __TI_flags+7(%r12),_TIF_NEED_RESCHED + jno .Lio_restore + # switch to kernel stack + lg %r1,__PT_R15(%r11) + aghi %r1,-(STACK_FRAME_OVERHEAD + __PT_SIZE) + mvc STACK_FRAME_OVERHEAD(__PT_SIZE,%r1),0(%r11) + xc __SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1) + la %r11,STACK_FRAME_OVERHEAD(%r1) + lgr %r15,%r1 + # TRACE_IRQS_ON already done at .Lio_return, call + # TRACE_IRQS_OFF to keep things symmetrical + TRACE_IRQS_OFF + brasl %r14,preempt_schedule_irq + j .Lio_return +#else + j .Lio_restore +#endif + +# +# Need to do work before returning to userspace, switch to kernel stack +# +.Lio_work_user: + lg %r1,__LC_KERNEL_STACK + mvc STACK_FRAME_OVERHEAD(__PT_SIZE,%r1),0(%r11) + xc __SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1) + la %r11,STACK_FRAME_OVERHEAD(%r1) + lgr %r15,%r1 + +# +# One of the work bits is on. Find out which one. +# +.Lio_work_tif: + tm __LC_CPU_FLAGS+7,_CIF_MCCK_PENDING + jo .Lio_mcck_pending + tm __TI_flags+7(%r12),_TIF_NEED_RESCHED + jo .Lio_reschedule + tm __TI_flags+7(%r12),_TIF_SIGPENDING + jo .Lio_sigpending + tm __TI_flags+7(%r12),_TIF_NOTIFY_RESUME + jo .Lio_notify_resume + tm __LC_CPU_FLAGS+7,_CIF_ASCE + jo .Lio_uaccess + j .Lio_return # beware of critical section cleanup + +# +# _CIF_MCCK_PENDING is set, call handler +# +.Lio_mcck_pending: + # TRACE_IRQS_ON already done at .Lio_return + brasl %r14,s390_handle_mcck # TIF bit will be cleared by handler + TRACE_IRQS_OFF + j .Lio_return + +# +# _CIF_ASCE is set, load user space asce +# +.Lio_uaccess: + ni __LC_CPU_FLAGS+7,255-_CIF_ASCE + lctlg %c1,%c1,__LC_USER_ASCE # load primary asce + j .Lio_return + +# +# _TIF_NEED_RESCHED is set, call schedule +# +.Lio_reschedule: + # TRACE_IRQS_ON already done at .Lio_return + ssm __LC_SVC_NEW_PSW # reenable interrupts + brasl %r14,schedule # call scheduler + ssm __LC_PGM_NEW_PSW # disable I/O and ext. interrupts + TRACE_IRQS_OFF + j .Lio_return + +# +# _TIF_SIGPENDING or is set, call do_signal +# +.Lio_sigpending: + # TRACE_IRQS_ON already done at .Lio_return + ssm __LC_SVC_NEW_PSW # reenable interrupts + lgr %r2,%r11 # pass pointer to pt_regs + brasl %r14,do_signal + ssm __LC_PGM_NEW_PSW # disable I/O and ext. interrupts + TRACE_IRQS_OFF + j .Lio_return + +# +# _TIF_NOTIFY_RESUME or is set, call do_notify_resume +# +.Lio_notify_resume: + # TRACE_IRQS_ON already done at .Lio_return + ssm __LC_SVC_NEW_PSW # reenable interrupts + lgr %r2,%r11 # pass pointer to pt_regs + brasl %r14,do_notify_resume + ssm __LC_PGM_NEW_PSW # disable I/O and ext. interrupts + TRACE_IRQS_OFF + j .Lio_return + +/* + * External interrupt handler routine + */ +ENTRY(ext_int_handler) + STCK __LC_INT_CLOCK + stpt __LC_ASYNC_ENTER_TIMER + stmg %r8,%r15,__LC_SAVE_AREA_ASYNC + lg %r10,__LC_LAST_BREAK + lg %r12,__LC_THREAD_INFO + larl %r13,system_call + lmg %r8,%r9,__LC_EXT_OLD_PSW + HANDLE_SIE_INTERCEPT %r14,3 + SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_STACK,STACK_SHIFT + tmhh %r8,0x0001 # interrupting from user ? + jz .Lext_skip + UPDATE_VTIME %r14,__LC_ASYNC_ENTER_TIMER + LAST_BREAK %r14 +.Lext_skip: + stmg %r0,%r7,__PT_R0(%r11) + mvc __PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC + stmg %r8,%r9,__PT_PSW(%r11) + lghi %r1,__LC_EXT_PARAMS2 + mvc __PT_INT_CODE(4,%r11),__LC_EXT_CPU_ADDR + mvc __PT_INT_PARM(4,%r11),__LC_EXT_PARAMS + mvc __PT_INT_PARM_LONG(8,%r11),0(%r1) + xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) + TRACE_IRQS_OFF + xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + lgr %r2,%r11 # pass pointer to pt_regs + lghi %r3,EXT_INTERRUPT + brasl %r14,do_IRQ + j .Lio_return + +/* + * Load idle PSW. The second "half" of this function is in .Lcleanup_idle. + */ +ENTRY(psw_idle) + stg %r3,__SF_EMPTY(%r15) + larl %r1,.Lpsw_idle_lpsw+4 + stg %r1,__SF_EMPTY+8(%r15) + STCK __CLOCK_IDLE_ENTER(%r2) + stpt __TIMER_IDLE_ENTER(%r2) +.Lpsw_idle_lpsw: + lpswe __SF_EMPTY(%r15) + br %r14 +.Lpsw_idle_end: + +.L__critical_end: + +/* + * Machine check handler routines + */ +ENTRY(mcck_int_handler) + STCK __LC_MCCK_CLOCK + la %r1,4095 # revalidate r1 + spt __LC_CPU_TIMER_SAVE_AREA-4095(%r1) # revalidate cpu timer + lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1)# revalidate gprs + lg %r10,__LC_LAST_BREAK + lg %r12,__LC_THREAD_INFO + larl %r13,system_call + lmg %r8,%r9,__LC_MCK_OLD_PSW + HANDLE_SIE_INTERCEPT %r14,4 + tm __LC_MCCK_CODE,0x80 # system damage? + jo .Lmcck_panic # yes -> rest of mcck code invalid + lghi %r14,__LC_CPU_TIMER_SAVE_AREA + mvc __LC_MCCK_ENTER_TIMER(8),0(%r14) + tm __LC_MCCK_CODE+5,0x02 # stored cpu timer value valid? + jo 3f + la %r14,__LC_SYNC_ENTER_TIMER + clc 0(8,%r14),__LC_ASYNC_ENTER_TIMER + jl 0f + la %r14,__LC_ASYNC_ENTER_TIMER +0: clc 0(8,%r14),__LC_EXIT_TIMER + jl 1f + la %r14,__LC_EXIT_TIMER +1: clc 0(8,%r14),__LC_LAST_UPDATE_TIMER + jl 2f + la %r14,__LC_LAST_UPDATE_TIMER +2: spt 0(%r14) + mvc __LC_MCCK_ENTER_TIMER(8),0(%r14) +3: tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? + jno .Lmcck_panic # no -> skip cleanup critical + SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_PANIC_STACK,PAGE_SHIFT + tm %r8,0x0001 # interrupting from user ? + jz .Lmcck_skip + UPDATE_VTIME %r14,__LC_MCCK_ENTER_TIMER + LAST_BREAK %r14 +.Lmcck_skip: + lghi %r14,__LC_GPREGS_SAVE_AREA+64 + stmg %r0,%r7,__PT_R0(%r11) + mvc __PT_R8(64,%r11),0(%r14) + stmg %r8,%r9,__PT_PSW(%r11) + xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11) + xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + lgr %r2,%r11 # pass pointer to pt_regs + brasl %r14,s390_do_machine_check + tm __PT_PSW+1(%r11),0x01 # returning to user ? + jno .Lmcck_return + lg %r1,__LC_KERNEL_STACK # switch to kernel stack + mvc STACK_FRAME_OVERHEAD(__PT_SIZE,%r1),0(%r11) + xc __SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1) + la %r11,STACK_FRAME_OVERHEAD(%r1) + lgr %r15,%r1 + ssm __LC_PGM_NEW_PSW # turn dat on, keep irqs off + tm __LC_CPU_FLAGS+7,_CIF_MCCK_PENDING + jno .Lmcck_return + TRACE_IRQS_OFF + brasl %r14,s390_handle_mcck + TRACE_IRQS_ON +.Lmcck_return: + lg %r14,__LC_VDSO_PER_CPU + lmg %r0,%r10,__PT_R0(%r11) + mvc __LC_RETURN_MCCK_PSW(16),__PT_PSW(%r11) # move return PSW + tm __LC_RETURN_MCCK_PSW+1,0x01 # returning to user ? + jno 0f + stpt __LC_EXIT_TIMER + mvc __VDSO_ECTG_BASE(16,%r14),__LC_EXIT_TIMER +0: lmg %r11,%r15,__PT_R11(%r11) + lpswe __LC_RETURN_MCCK_PSW + +.Lmcck_panic: + lg %r14,__LC_PANIC_STACK + slgr %r14,%r15 + srag %r14,%r14,PAGE_SHIFT + jz 0f + lg %r15,__LC_PANIC_STACK +0: aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE) + j .Lmcck_skip + +# +# PSW restart interrupt handler +# +ENTRY(restart_int_handler) + stg %r15,__LC_SAVE_AREA_RESTART + lg %r15,__LC_RESTART_STACK + aghi %r15,-__PT_SIZE # create pt_regs on stack + xc 0(__PT_SIZE,%r15),0(%r15) + stmg %r0,%r14,__PT_R0(%r15) + mvc __PT_R15(8,%r15),__LC_SAVE_AREA_RESTART + mvc __PT_PSW(16,%r15),__LC_RST_OLD_PSW # store restart old psw + aghi %r15,-STACK_FRAME_OVERHEAD # create stack frame on stack + xc 0(STACK_FRAME_OVERHEAD,%r15),0(%r15) + lg %r1,__LC_RESTART_FN # load fn, parm & source cpu + lg %r2,__LC_RESTART_DATA + lg %r3,__LC_RESTART_SOURCE + ltgr %r3,%r3 # test source cpu address + jm 1f # negative -> skip source stop +0: sigp %r4,%r3,SIGP_SENSE # sigp sense to source cpu + brc 10,0b # wait for status stored +1: basr %r14,%r1 # call function + stap __SF_EMPTY(%r15) # store cpu address + llgh %r3,__SF_EMPTY(%r15) +2: sigp %r4,%r3,SIGP_STOP # sigp stop to current cpu + brc 2,2b +3: j 3b + + .section .kprobes.text, "ax" + +#ifdef CONFIG_CHECK_STACK +/* + * The synchronous or the asynchronous stack overflowed. We are dead. + * No need to properly save the registers, we are going to panic anyway. + * Setup a pt_regs so that show_trace can provide a good call trace. + */ +stack_overflow: + lg %r15,__LC_PANIC_STACK # change to panic stack + la %r11,STACK_FRAME_OVERHEAD(%r15) + stmg %r0,%r7,__PT_R0(%r11) + stmg %r8,%r9,__PT_PSW(%r11) + mvc __PT_R8(64,%r11),0(%r14) + stg %r10,__PT_ORIG_GPR2(%r11) # store last break to orig_gpr2 + xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) + lgr %r2,%r11 # pass pointer to pt_regs + jg kernel_stack_overflow +#endif + + .align 8 +.Lcleanup_table: + .quad system_call + .quad .Lsysc_do_svc + .quad .Lsysc_tif + .quad .Lsysc_restore + .quad .Lsysc_done + .quad .Lio_tif + .quad .Lio_restore + .quad .Lio_done + .quad psw_idle + .quad .Lpsw_idle_end + +cleanup_critical: + clg %r9,BASED(.Lcleanup_table) # system_call + jl 0f + clg %r9,BASED(.Lcleanup_table+8) # .Lsysc_do_svc + jl .Lcleanup_system_call + clg %r9,BASED(.Lcleanup_table+16) # .Lsysc_tif + jl 0f + clg %r9,BASED(.Lcleanup_table+24) # .Lsysc_restore + jl .Lcleanup_sysc_tif + clg %r9,BASED(.Lcleanup_table+32) # .Lsysc_done + jl .Lcleanup_sysc_restore + clg %r9,BASED(.Lcleanup_table+40) # .Lio_tif + jl 0f + clg %r9,BASED(.Lcleanup_table+48) # .Lio_restore + jl .Lcleanup_io_tif + clg %r9,BASED(.Lcleanup_table+56) # .Lio_done + jl .Lcleanup_io_restore + clg %r9,BASED(.Lcleanup_table+64) # psw_idle + jl 0f + clg %r9,BASED(.Lcleanup_table+72) # .Lpsw_idle_end + jl .Lcleanup_idle +0: br %r14 + + +.Lcleanup_system_call: + # check if stpt has been executed + clg %r9,BASED(.Lcleanup_system_call_insn) + jh 0f + mvc __LC_SYNC_ENTER_TIMER(8),__LC_ASYNC_ENTER_TIMER + cghi %r11,__LC_SAVE_AREA_ASYNC + je 0f + mvc __LC_SYNC_ENTER_TIMER(8),__LC_MCCK_ENTER_TIMER +0: # check if stmg has been executed + clg %r9,BASED(.Lcleanup_system_call_insn+8) + jh 0f + mvc __LC_SAVE_AREA_SYNC(64),0(%r11) +0: # check if base register setup + TIF bit load has been done + clg %r9,BASED(.Lcleanup_system_call_insn+16) + jhe 0f + # set up saved registers r10 and r12 + stg %r10,16(%r11) # r10 last break + stg %r12,32(%r11) # r12 thread-info pointer +0: # check if the user time update has been done + clg %r9,BASED(.Lcleanup_system_call_insn+24) + jh 0f + lg %r15,__LC_EXIT_TIMER + slg %r15,__LC_SYNC_ENTER_TIMER + alg %r15,__LC_USER_TIMER + stg %r15,__LC_USER_TIMER +0: # check if the system time update has been done + clg %r9,BASED(.Lcleanup_system_call_insn+32) + jh 0f + lg %r15,__LC_LAST_UPDATE_TIMER + slg %r15,__LC_EXIT_TIMER + alg %r15,__LC_SYSTEM_TIMER + stg %r15,__LC_SYSTEM_TIMER +0: # update accounting time stamp + mvc __LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER + # do LAST_BREAK + lg %r9,16(%r11) + srag %r9,%r9,23 + jz 0f + mvc __TI_last_break(8,%r12),16(%r11) +0: # set up saved register r11 + lg %r15,__LC_KERNEL_STACK + la %r9,STACK_FRAME_OVERHEAD(%r15) + stg %r9,24(%r11) # r11 pt_regs pointer + # fill pt_regs + mvc __PT_R8(64,%r9),__LC_SAVE_AREA_SYNC + stmg %r0,%r7,__PT_R0(%r9) + mvc __PT_PSW(16,%r9),__LC_SVC_OLD_PSW + mvc __PT_INT_CODE(4,%r9),__LC_SVC_ILC + xc __PT_FLAGS(8,%r9),__PT_FLAGS(%r9) + mvi __PT_FLAGS+7(%r9),_PIF_SYSCALL + # setup saved register r15 + stg %r15,56(%r11) # r15 stack pointer + # set new psw address and exit + larl %r9,.Lsysc_do_svc + br %r14 +.Lcleanup_system_call_insn: + .quad system_call + .quad .Lsysc_stmg + .quad .Lsysc_per + .quad .Lsysc_vtime+18 + .quad .Lsysc_vtime+42 + +.Lcleanup_sysc_tif: + larl %r9,.Lsysc_tif + br %r14 + +.Lcleanup_sysc_restore: + clg %r9,BASED(.Lcleanup_sysc_restore_insn) + je 0f + lg %r9,24(%r11) # get saved pointer to pt_regs + mvc __LC_RETURN_PSW(16),__PT_PSW(%r9) + mvc 0(64,%r11),__PT_R8(%r9) + lmg %r0,%r7,__PT_R0(%r9) +0: lmg %r8,%r9,__LC_RETURN_PSW + br %r14 +.Lcleanup_sysc_restore_insn: + .quad .Lsysc_done - 4 + +.Lcleanup_io_tif: + larl %r9,.Lio_tif + br %r14 + +.Lcleanup_io_restore: + clg %r9,BASED(.Lcleanup_io_restore_insn) + je 0f + lg %r9,24(%r11) # get saved r11 pointer to pt_regs + mvc __LC_RETURN_PSW(16),__PT_PSW(%r9) + mvc 0(64,%r11),__PT_R8(%r9) + lmg %r0,%r7,__PT_R0(%r9) +0: lmg %r8,%r9,__LC_RETURN_PSW + br %r14 +.Lcleanup_io_restore_insn: + .quad .Lio_done - 4 + +.Lcleanup_idle: + # copy interrupt clock & cpu timer + mvc __CLOCK_IDLE_EXIT(8,%r2),__LC_INT_CLOCK + mvc __TIMER_IDLE_EXIT(8,%r2),__LC_ASYNC_ENTER_TIMER + cghi %r11,__LC_SAVE_AREA_ASYNC + je 0f + mvc __CLOCK_IDLE_EXIT(8,%r2),__LC_MCCK_CLOCK + mvc __TIMER_IDLE_EXIT(8,%r2),__LC_MCCK_ENTER_TIMER +0: # check if stck & stpt have been executed + clg %r9,BASED(.Lcleanup_idle_insn) + jhe 1f + mvc __CLOCK_IDLE_ENTER(8,%r2),__CLOCK_IDLE_EXIT(%r2) + mvc __TIMER_IDLE_ENTER(8,%r2),__TIMER_IDLE_EXIT(%r2) +1: # account system time going idle + lg %r9,__LC_STEAL_TIMER + alg %r9,__CLOCK_IDLE_ENTER(%r2) + slg %r9,__LC_LAST_UPDATE_CLOCK + stg %r9,__LC_STEAL_TIMER + mvc __LC_LAST_UPDATE_CLOCK(8),__CLOCK_IDLE_EXIT(%r2) + lg %r9,__LC_SYSTEM_TIMER + alg %r9,__LC_LAST_UPDATE_TIMER + slg %r9,__TIMER_IDLE_ENTER(%r2) + stg %r9,__LC_SYSTEM_TIMER + mvc __LC_LAST_UPDATE_TIMER(8),__TIMER_IDLE_EXIT(%r2) + # prepare return psw + nihh %r8,0xfcfd # clear irq & wait state bits + lg %r9,48(%r11) # return from psw_idle + br %r14 +.Lcleanup_idle_insn: + .quad .Lpsw_idle_lpsw + +/* + * Integer constants + */ + .align 8 +.Lcritical_start: + .quad .L__critical_start +.Lcritical_length: + .quad .L__critical_end - .L__critical_start + + +#if IS_ENABLED(CONFIG_KVM) +/* + * sie64a calling convention: + * %r2 pointer to sie control block + * %r3 guest register save area + */ +ENTRY(sie64a) + stmg %r6,%r14,__SF_GPRS(%r15) # save kernel registers + stg %r2,__SF_EMPTY(%r15) # save control block pointer + stg %r3,__SF_EMPTY+8(%r15) # save guest register save area + xc __SF_EMPTY+16(16,%r15),__SF_EMPTY+16(%r15) # host id & reason + lmg %r0,%r13,0(%r3) # load guest gprs 0-13 + lg %r14,__LC_GMAP # get gmap pointer + ltgr %r14,%r14 + jz .Lsie_gmap + lctlg %c1,%c1,__GMAP_ASCE(%r14) # load primary asce +.Lsie_gmap: + lg %r14,__SF_EMPTY(%r15) # get control block pointer + oi __SIE_PROG0C+3(%r14),1 # we are going into SIE now + tm __SIE_PROG20+3(%r14),1 # last exit... + jnz .Lsie_done + LPP __SF_EMPTY(%r15) # set guest id + sie 0(%r14) +.Lsie_done: + LPP __SF_EMPTY+16(%r15) # set host id + ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE + lctlg %c1,%c1,__LC_USER_ASCE # load primary asce +# some program checks are suppressing. C code (e.g. do_protection_exception) +# will rewind the PSW by the ILC, which is 4 bytes in case of SIE. Other +# instructions between sie64a and .Lsie_done should not cause program +# interrupts. So lets use a nop (47 00 00 00) as a landing pad. +# See also HANDLE_SIE_INTERCEPT +.Lrewind_pad: + nop 0 + .globl sie_exit +sie_exit: + lg %r14,__SF_EMPTY+8(%r15) # load guest register save area + stmg %r0,%r13,0(%r14) # save guest gprs 0-13 + lmg %r6,%r14,__SF_GPRS(%r15) # restore kernel registers + lg %r2,__SF_EMPTY+24(%r15) # return exit reason code + br %r14 +.Lsie_fault: + lghi %r14,-EFAULT + stg %r14,__SF_EMPTY+24(%r15) # set exit reason code + j sie_exit + + .align 8 +.Lsie_critical: + .quad .Lsie_gmap +.Lsie_critical_length: + .quad .Lsie_done - .Lsie_gmap + + EX_TABLE(.Lrewind_pad,.Lsie_fault) + EX_TABLE(sie_exit,.Lsie_fault) +#endif + + .section .rodata, "a" +#define SYSCALL(esame,emu) .long esame + .globl sys_call_table +sys_call_table: +#include "syscalls.S" +#undef SYSCALL + +#ifdef CONFIG_COMPAT + +#define SYSCALL(esame,emu) .long emu + .globl sys_call_table_emu +sys_call_table_emu: +#include "syscalls.S" +#undef SYSCALL +#endif diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h new file mode 100644 index 000000000..834df047d --- /dev/null +++ b/arch/s390/kernel/entry.h @@ -0,0 +1,81 @@ +#ifndef _ENTRY_H +#define _ENTRY_H + +#include <linux/types.h> +#include <linux/signal.h> +#include <asm/ptrace.h> +#include <asm/idle.h> + +extern void *restart_stack; +extern unsigned long suspend_zero_pages; + +void system_call(void); +void pgm_check_handler(void); +void ext_int_handler(void); +void io_int_handler(void); +void mcck_int_handler(void); +void restart_int_handler(void); +void restart_call_handler(void); +void psw_idle(struct s390_idle_data *, unsigned long); + +asmlinkage long do_syscall_trace_enter(struct pt_regs *regs); +asmlinkage void do_syscall_trace_exit(struct pt_regs *regs); + +int alloc_vector_registers(struct task_struct *tsk); + +void do_protection_exception(struct pt_regs *regs); +void do_dat_exception(struct pt_regs *regs); + +void addressing_exception(struct pt_regs *regs); +void data_exception(struct pt_regs *regs); +void default_trap_handler(struct pt_regs *regs); +void divide_exception(struct pt_regs *regs); +void execute_exception(struct pt_regs *regs); +void hfp_divide_exception(struct pt_regs *regs); +void hfp_overflow_exception(struct pt_regs *regs); +void hfp_significance_exception(struct pt_regs *regs); +void hfp_sqrt_exception(struct pt_regs *regs); +void hfp_underflow_exception(struct pt_regs *regs); +void illegal_op(struct pt_regs *regs); +void operand_exception(struct pt_regs *regs); +void overflow_exception(struct pt_regs *regs); +void privileged_op(struct pt_regs *regs); +void space_switch_exception(struct pt_regs *regs); +void special_op_exception(struct pt_regs *regs); +void specification_exception(struct pt_regs *regs); +void transaction_exception(struct pt_regs *regs); +void translation_exception(struct pt_regs *regs); +void vector_exception(struct pt_regs *regs); + +void do_per_trap(struct pt_regs *regs); +void do_report_trap(struct pt_regs *regs, int si_signo, int si_code, char *str); +void syscall_trace(struct pt_regs *regs, int entryexit); +void kernel_stack_overflow(struct pt_regs * regs); +void do_signal(struct pt_regs *regs); +void handle_signal32(struct ksignal *ksig, sigset_t *oldset, + struct pt_regs *regs); +void do_notify_resume(struct pt_regs *regs); + +void __init init_IRQ(void); +void do_IRQ(struct pt_regs *regs, int irq); +void do_restart(void); +void __init startup_init(void); +void die(struct pt_regs *regs, const char *str); +int setup_profiling_timer(unsigned int multiplier); +void __init time_init(void); +int pfn_is_nosave(unsigned long); +void s390_early_resume(void); +unsigned long prepare_ftrace_return(unsigned long parent, unsigned long ip); + +struct s390_mmap_arg_struct; +struct fadvise64_64_args; +struct old_sigaction; + +long sys_rt_sigreturn(void); +long sys_sigreturn(void); + +long sys_s390_personality(unsigned int personality); +long sys_s390_runtime_instr(int command, int signum); +long sys_s390_pci_mmio_write(unsigned long, const void __user *, size_t); +long sys_s390_pci_mmio_read(unsigned long, void __user *, size_t); +#endif /* _ENTRY_H */ diff --git a/arch/s390/kernel/ftrace.c b/arch/s390/kernel/ftrace.c new file mode 100644 index 000000000..e0eaf1113 --- /dev/null +++ b/arch/s390/kernel/ftrace.c @@ -0,0 +1,244 @@ +/* + * Dynamic function tracer architecture backend. + * + * Copyright IBM Corp. 2009,2014 + * + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>, + * Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#include <linux/moduleloader.h> +#include <linux/hardirq.h> +#include <linux/uaccess.h> +#include <linux/ftrace.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/kprobes.h> +#include <trace/syscall.h> +#include <asm/asm-offsets.h> +#include <asm/cacheflush.h> +#include "entry.h" + +/* + * The mcount code looks like this: + * stg %r14,8(%r15) # offset 0 + * larl %r1,<&counter> # offset 6 + * brasl %r14,_mcount # offset 12 + * lg %r14,8(%r15) # offset 18 + * Total length is 24 bytes. Only the first instruction will be patched + * by ftrace_make_call / ftrace_make_nop. + * The enabled ftrace code block looks like this: + * > brasl %r0,ftrace_caller # offset 0 + * larl %r1,<&counter> # offset 6 + * brasl %r14,_mcount # offset 12 + * lg %r14,8(%r15) # offset 18 + * The ftrace function gets called with a non-standard C function call ABI + * where r0 contains the return address. It is also expected that the called + * function only clobbers r0 and r1, but restores r2-r15. + * For module code we can't directly jump to ftrace caller, but need a + * trampoline (ftrace_plt), which clobbers also r1. + * The return point of the ftrace function has offset 24, so execution + * continues behind the mcount block. + * The disabled ftrace code block looks like this: + * > jg .+24 # offset 0 + * larl %r1,<&counter> # offset 6 + * brasl %r14,_mcount # offset 12 + * lg %r14,8(%r15) # offset 18 + * The jg instruction branches to offset 24 to skip as many instructions + * as possible. + * In case we use gcc's hotpatch feature the original and also the disabled + * function prologue contains only a single six byte instruction and looks + * like this: + * > brcl 0,0 # offset 0 + * To enable ftrace the code gets patched like above and afterwards looks + * like this: + * > brasl %r0,ftrace_caller # offset 0 + */ + +unsigned long ftrace_plt; + +static inline void ftrace_generate_orig_insn(struct ftrace_insn *insn) +{ +#ifdef CC_USING_HOTPATCH + /* brcl 0,0 */ + insn->opc = 0xc004; + insn->disp = 0; +#else + /* stg r14,8(r15) */ + insn->opc = 0xe3e0; + insn->disp = 0xf0080024; +#endif +} + +static inline int is_kprobe_on_ftrace(struct ftrace_insn *insn) +{ +#ifdef CONFIG_KPROBES + if (insn->opc == BREAKPOINT_INSTRUCTION) + return 1; +#endif + return 0; +} + +static inline void ftrace_generate_kprobe_nop_insn(struct ftrace_insn *insn) +{ +#ifdef CONFIG_KPROBES + insn->opc = BREAKPOINT_INSTRUCTION; + insn->disp = KPROBE_ON_FTRACE_NOP; +#endif +} + +static inline void ftrace_generate_kprobe_call_insn(struct ftrace_insn *insn) +{ +#ifdef CONFIG_KPROBES + insn->opc = BREAKPOINT_INSTRUCTION; + insn->disp = KPROBE_ON_FTRACE_CALL; +#endif +} + +int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr, + unsigned long addr) +{ + return 0; +} + +int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec, + unsigned long addr) +{ + struct ftrace_insn orig, new, old; + + if (probe_kernel_read(&old, (void *) rec->ip, sizeof(old))) + return -EFAULT; + if (addr == MCOUNT_ADDR) { + /* Initial code replacement */ + ftrace_generate_orig_insn(&orig); + ftrace_generate_nop_insn(&new); + } else if (is_kprobe_on_ftrace(&old)) { + /* + * If we find a breakpoint instruction, a kprobe has been + * placed at the beginning of the function. We write the + * constant KPROBE_ON_FTRACE_NOP into the remaining four + * bytes of the original instruction so that the kprobes + * handler can execute a nop, if it reaches this breakpoint. + */ + ftrace_generate_kprobe_call_insn(&orig); + ftrace_generate_kprobe_nop_insn(&new); + } else { + /* Replace ftrace call with a nop. */ + ftrace_generate_call_insn(&orig, rec->ip); + ftrace_generate_nop_insn(&new); + } + /* Verify that the to be replaced code matches what we expect. */ + if (memcmp(&orig, &old, sizeof(old))) + return -EINVAL; + s390_kernel_write((void *) rec->ip, &new, sizeof(new)); + return 0; +} + +int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) +{ + struct ftrace_insn orig, new, old; + + if (probe_kernel_read(&old, (void *) rec->ip, sizeof(old))) + return -EFAULT; + if (is_kprobe_on_ftrace(&old)) { + /* + * If we find a breakpoint instruction, a kprobe has been + * placed at the beginning of the function. We write the + * constant KPROBE_ON_FTRACE_CALL into the remaining four + * bytes of the original instruction so that the kprobes + * handler can execute a brasl if it reaches this breakpoint. + */ + ftrace_generate_kprobe_nop_insn(&orig); + ftrace_generate_kprobe_call_insn(&new); + } else { + /* Replace nop with an ftrace call. */ + ftrace_generate_nop_insn(&orig); + ftrace_generate_call_insn(&new, rec->ip); + } + /* Verify that the to be replaced code matches what we expect. */ + if (memcmp(&orig, &old, sizeof(old))) + return -EINVAL; + s390_kernel_write((void *) rec->ip, &new, sizeof(new)); + return 0; +} + +int ftrace_update_ftrace_func(ftrace_func_t func) +{ + return 0; +} + +int __init ftrace_dyn_arch_init(void) +{ + return 0; +} + +static int __init ftrace_plt_init(void) +{ + unsigned int *ip; + + ftrace_plt = (unsigned long) module_alloc(PAGE_SIZE); + if (!ftrace_plt) + panic("cannot allocate ftrace plt\n"); + ip = (unsigned int *) ftrace_plt; + ip[0] = 0x0d10e310; /* basr 1,0; lg 1,10(1); br 1 */ + ip[1] = 0x100a0004; + ip[2] = 0x07f10000; + ip[3] = FTRACE_ADDR >> 32; + ip[4] = FTRACE_ADDR & 0xffffffff; + set_memory_ro(ftrace_plt, 1); + return 0; +} +device_initcall(ftrace_plt_init); + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +/* + * Hook the return address and push it in the stack of return addresses + * in current thread info. + */ +unsigned long prepare_ftrace_return(unsigned long parent, unsigned long ip) +{ + struct ftrace_graph_ent trace; + + if (unlikely(ftrace_graph_is_dead())) + goto out; + if (unlikely(atomic_read(¤t->tracing_graph_pause))) + goto out; + ip = (ip & PSW_ADDR_INSN) - MCOUNT_INSN_SIZE; + trace.func = ip; + trace.depth = current->curr_ret_stack + 1; + /* Only trace if the calling function expects to. */ + if (!ftrace_graph_entry(&trace)) + goto out; + if (ftrace_push_return_trace(parent, ip, &trace.depth, 0) == -EBUSY) + goto out; + parent = (unsigned long) return_to_handler; +out: + return parent; +} +NOKPROBE_SYMBOL(prepare_ftrace_return); + +/* + * Patch the kernel code at ftrace_graph_caller location. The instruction + * there is branch relative on condition. To enable the ftrace graph code + * block, we simply patch the mask field of the instruction to zero and + * turn the instruction into a nop. + * To disable the ftrace graph code the mask field will be patched to + * all ones, which turns the instruction into an unconditional branch. + */ +int ftrace_enable_ftrace_graph_caller(void) +{ + u8 op = 0x04; /* set mask field to zero */ + + s390_kernel_write(__va(ftrace_graph_caller)+1, &op, sizeof(op)); + return 0; +} + +int ftrace_disable_ftrace_graph_caller(void) +{ + u8 op = 0xf4; /* set mask field to all ones */ + + s390_kernel_write(__va(ftrace_graph_caller)+1, &op, sizeof(op)); + return 0; +} + +#endif /* CONFIG_FUNCTION_GRAPH_TRACER */ diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S new file mode 100644 index 000000000..59b7c6470 --- /dev/null +++ b/arch/s390/kernel/head.S @@ -0,0 +1,454 @@ +/* + * Copyright IBM Corp. 1999, 2010 + * + * Author(s): Hartmut Penner <hp@de.ibm.com> + * Martin Schwidefsky <schwidefsky@de.ibm.com> + * Rob van der Heij <rvdhei@iae.nl> + * Heiko Carstens <heiko.carstens@de.ibm.com> + * + * There are 5 different IPL methods + * 1) load the image directly into ram at address 0 and do an PSW restart + * 2) linload will load the image from address 0x10000 to memory 0x10000 + * and start the code thru LPSW 0x0008000080010000 (VM only, deprecated) + * 3) generate the tape ipl header, store the generated image on a tape + * and ipl from it + * In case of SL tape you need to IPL 5 times to get past VOL1 etc + * 4) generate the vm reader ipl header, move the generated image to the + * VM reader (use option NOH!) and do a ipl from reader (VM only) + * 5) direct call of start by the SALIPL loader + * We use the cpuid to distinguish between VM and native ipl + * params for kernel are pushed to 0x10400 (see setup.h) + * + */ + +#include <linux/init.h> +#include <linux/linkage.h> +#include <asm/asm-offsets.h> +#include <asm/thread_info.h> +#include <asm/page.h> + +#define ARCH_OFFSET 4 + +__HEAD + +#define IPL_BS 0x730 + .org 0 + .long 0x00080000,0x80000000+iplstart # The first 24 bytes are loaded + .long 0x02000018,0x60000050 # by ipl to addresses 0-23. + .long 0x02000068,0x60000050 # (a PSW and two CCWs). + .fill 80-24,1,0x40 # bytes 24-79 are discarded !! + .long 0x020000f0,0x60000050 # The next 160 byte are loaded + .long 0x02000140,0x60000050 # to addresses 0x18-0xb7 + .long 0x02000190,0x60000050 # They form the continuation + .long 0x020001e0,0x60000050 # of the CCW program started + .long 0x02000230,0x60000050 # by ipl and load the range + .long 0x02000280,0x60000050 # 0x0f0-0x730 from the image + .long 0x020002d0,0x60000050 # to the range 0x0f0-0x730 + .long 0x02000320,0x60000050 # in memory. At the end of + .long 0x02000370,0x60000050 # the channel program the PSW + .long 0x020003c0,0x60000050 # at location 0 is loaded. + .long 0x02000410,0x60000050 # Initial processing starts + .long 0x02000460,0x60000050 # at 0x200 = iplstart. + .long 0x020004b0,0x60000050 + .long 0x02000500,0x60000050 + .long 0x02000550,0x60000050 + .long 0x020005a0,0x60000050 + .long 0x020005f0,0x60000050 + .long 0x02000640,0x60000050 + .long 0x02000690,0x60000050 + .long 0x020006e0,0x20000050 + + .org 0x200 +# +# subroutine to set architecture mode +# +.Lsetmode: + mvi __LC_AR_MODE_ID,1 # set esame flag + slr %r0,%r0 # set cpuid to zero + lhi %r1,2 # mode 2 = esame (dump) + sigp %r1,%r0,0x12 # switch to esame mode + bras %r13,0f + .fill 16,4,0x0 +0: lmh %r0,%r15,0(%r13) # clear high-order half of gprs + sam31 # switch to 31 bit addressing mode + br %r14 + +# +# subroutine to wait for end I/O +# +.Lirqwait: + mvc 0x1f0(16),.Lnewpsw # set up IO interrupt psw + lpsw .Lwaitpsw +.Lioint: + br %r14 + .align 8 +.Lnewpsw: + .quad 0x0000000080000000,.Lioint +.Lwaitpsw: + .long 0x020a0000,0x80000000+.Lioint + +# +# subroutine for loading cards from the reader +# +.Lloader: + la %r4,0(%r14) + la %r3,.Lorb # r2 = address of orb into r2 + la %r5,.Lirb # r4 = address of irb + la %r6,.Lccws + la %r7,20 +.Linit: + st %r2,4(%r6) # initialize CCW data addresses + la %r2,0x50(%r2) + la %r6,8(%r6) + bct 7,.Linit + + lctl %c6,%c6,.Lcr6 # set IO subclass mask + slr %r2,%r2 +.Lldlp: + ssch 0(%r3) # load chunk of 1600 bytes + bnz .Llderr +.Lwait4irq: + bas %r14,.Lirqwait + c %r1,0xb8 # compare subchannel number + bne .Lwait4irq + tsch 0(%r5) + + slr %r0,%r0 + ic %r0,8(%r5) # get device status + chi %r0,8 # channel end ? + be .Lcont + chi %r0,12 # channel end + device end ? + be .Lcont + + l %r0,4(%r5) + s %r0,8(%r3) # r0/8 = number of ccws executed + mhi %r0,10 # *10 = number of bytes in ccws + lh %r3,10(%r5) # get residual count + sr %r0,%r3 # #ccws*80-residual=#bytes read + ar %r2,%r0 + + br %r4 # r2 contains the total size + +.Lcont: + ahi %r2,0x640 # add 0x640 to total size + la %r6,.Lccws + la %r7,20 +.Lincr: + l %r0,4(%r6) # update CCW data addresses + ahi %r0,0x640 + st %r0,4(%r6) + ahi %r6,8 + bct 7,.Lincr + + b .Lldlp +.Llderr: + lpsw .Lcrash + + .align 8 +.Lorb: .long 0x00000000,0x0080ff00,.Lccws +.Lirb: .long 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 +.Lcr6: .long 0xff000000 +.Lloadp:.long 0,0 + .align 8 +.Lcrash:.long 0x000a0000,0x00000000 + + .align 8 +.Lccws: .rept 19 + .long 0x02600050,0x00000000 + .endr + .long 0x02200050,0x00000000 + +iplstart: + bas %r14,.Lsetmode # Immediately switch to 64 bit mode + lh %r1,0xb8 # test if subchannel number + bct %r1,.Lnoload # is valid + l %r1,0xb8 # load ipl subchannel number + la %r2,IPL_BS # load start address + bas %r14,.Lloader # load rest of ipl image + l %r12,.Lparm # pointer to parameter area + st %r1,IPL_DEVICE+ARCH_OFFSET-PARMAREA(%r12) # save ipl device number + +# +# load parameter file from ipl device +# +.Lagain1: + l %r2,.Linitrd # ramdisk loc. is temp + bas %r14,.Lloader # load parameter file + ltr %r2,%r2 # got anything ? + bz .Lnopf + chi %r2,895 + bnh .Lnotrunc + la %r2,895 +.Lnotrunc: + l %r4,.Linitrd + clc 0(3,%r4),.L_hdr # if it is HDRx + bz .Lagain1 # skip dataset header + clc 0(3,%r4),.L_eof # if it is EOFx + bz .Lagain1 # skip dateset trailer + la %r5,0(%r4,%r2) + lr %r3,%r2 + la %r3,COMMAND_LINE-PARMAREA(%r12) # load adr. of command line + mvc 0(256,%r3),0(%r4) + mvc 256(256,%r3),256(%r4) + mvc 512(256,%r3),512(%r4) + mvc 768(122,%r3),768(%r4) + slr %r0,%r0 + b .Lcntlp +.Ldelspc: + ic %r0,0(%r2,%r3) + chi %r0,0x20 # is it a space ? + be .Lcntlp + ahi %r2,1 + b .Leolp +.Lcntlp: + brct %r2,.Ldelspc +.Leolp: + slr %r0,%r0 + stc %r0,0(%r2,%r3) # terminate buffer +.Lnopf: + +# +# load ramdisk from ipl device +# +.Lagain2: + l %r2,.Linitrd # addr of ramdisk + st %r2,INITRD_START+ARCH_OFFSET-PARMAREA(%r12) + bas %r14,.Lloader # load ramdisk + st %r2,INITRD_SIZE+ARCH_OFFSET-PARMAREA(%r12) # store size of rd + ltr %r2,%r2 + bnz .Lrdcont + st %r2,INITRD_START+ARCH_OFFSET-PARMAREA(%r12) # no ramdisk found +.Lrdcont: + l %r2,.Linitrd + + clc 0(3,%r2),.L_hdr # skip HDRx and EOFx + bz .Lagain2 + clc 0(3,%r2),.L_eof + bz .Lagain2 + +# +# reset files in VM reader +# + stidp .Lcpuid # store cpuid + tm .Lcpuid,0xff # running VM ? + bno .Lnoreset + la %r2,.Lreset + lhi %r3,26 + diag %r2,%r3,8 + la %r5,.Lirb + stsch 0(%r5) # check if irq is pending + tm 30(%r5),0x0f # by verifying if any of the + bnz .Lwaitforirq # activity or status control + tm 31(%r5),0xff # bits is set in the schib + bz .Lnoreset +.Lwaitforirq: + bas %r14,.Lirqwait # wait for IO interrupt + c %r1,0xb8 # compare subchannel number + bne .Lwaitforirq + la %r5,.Lirb + tsch 0(%r5) +.Lnoreset: + b .Lnoload + +# +# everything loaded, go for it +# +.Lnoload: + l %r1,.Lstartup + br %r1 + +.Linitrd:.long _end # default address of initrd +.Lparm: .long PARMAREA +.Lstartup: .long startup +.Lreset:.byte 0xc3,0xc8,0xc1,0xd5,0xc7,0xc5,0x40,0xd9,0xc4,0xd9,0x40 + .byte 0xc1,0xd3,0xd3,0x40,0xd2,0xc5,0xc5,0xd7,0x40,0xd5,0xd6 + .byte 0xc8,0xd6,0xd3,0xc4 # "change rdr all keep nohold" +.L_eof: .long 0xc5d6c600 /* C'EOF' */ +.L_hdr: .long 0xc8c4d900 /* C'HDR' */ + .align 8 +.Lcpuid:.fill 8,1,0 + +# +# SALIPL loader support. Based on a patch by Rob van der Heij. +# This entry point is called directly from the SALIPL loader and +# doesn't need a builtin ipl record. +# + .org 0x800 +ENTRY(start) + stm %r0,%r15,0x07b0 # store registers + bas %r14,.Lsetmode # Immediately switch to 64 bit mode + basr %r12,%r0 +.base: + l %r11,.parm + l %r8,.cmd # pointer to command buffer + + ltr %r9,%r9 # do we have SALIPL parameters? + bp .sk8x8 + + mvc 0(64,%r8),0x00b0 # copy saved registers + xc 64(240-64,%r8),0(%r8) # remainder of buffer + tr 0(64,%r8),.lowcase + b .gotr +.sk8x8: + mvc 0(240,%r8),0(%r9) # copy iplparms into buffer +.gotr: + slr %r0,%r0 + st %r0,INITRD_SIZE+ARCH_OFFSET-PARMAREA(%r11) + st %r0,INITRD_START+ARCH_OFFSET-PARMAREA(%r11) + j startup # continue with startup +.cmd: .long COMMAND_LINE # address of command line buffer +.parm: .long PARMAREA +.lowcase: + .byte 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07 + .byte 0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,0x0f + .byte 0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17 + .byte 0x18,0x19,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f + .byte 0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27 + .byte 0x28,0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f + .byte 0x30,0x31,0x32,0x33,0x34,0x35,0x36,0x37 + .byte 0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x3e,0x3f + .byte 0x40,0x41,0x42,0x43,0x44,0x45,0x46,0x47 + .byte 0x48,0x49,0x4a,0x4b,0x4c,0x4d,0x4e,0x4f + .byte 0x50,0x51,0x52,0x53,0x54,0x55,0x56,0x57 + .byte 0x58,0x59,0x5a,0x5b,0x5c,0x5d,0x5e,0x5f + .byte 0x60,0x61,0x62,0x63,0x64,0x65,0x66,0x67 + .byte 0x68,0x69,0x6a,0x6b,0x6c,0x6d,0x6e,0x6f + .byte 0x70,0x71,0x72,0x73,0x74,0x75,0x76,0x77 + .byte 0x78,0x79,0x7a,0x7b,0x7c,0x7d,0x7e,0x7f + + .byte 0x80,0x81,0x82,0x83,0x84,0x85,0x86,0x87 + .byte 0x88,0x89,0x8a,0x8b,0x8c,0x8d,0x8e,0x8f + .byte 0x90,0x91,0x92,0x93,0x94,0x95,0x96,0x97 + .byte 0x98,0x99,0x9a,0x9b,0x9c,0x9d,0x9e,0x9f + .byte 0xa0,0xa1,0xa2,0xa3,0xa4,0xa5,0xa6,0xa7 + .byte 0xa8,0xa9,0xaa,0xab,0xac,0xad,0xae,0xaf + .byte 0xb0,0xb1,0xb2,0xb3,0xb4,0xb5,0xb6,0xb7 + .byte 0xb8,0xb9,0xba,0xbb,0xbc,0xbd,0xbe,0xbf + .byte 0xc0,0x81,0x82,0x83,0x84,0x85,0x86,0x87 # .abcdefg + .byte 0x88,0x89,0xca,0xcb,0xcc,0xcd,0xce,0xcf # hi + .byte 0xd0,0x91,0x92,0x93,0x94,0x95,0x96,0x97 # .jklmnop + .byte 0x98,0x99,0xda,0xdb,0xdc,0xdd,0xde,0xdf # qr + .byte 0xe0,0xe1,0xa2,0xa3,0xa4,0xa5,0xa6,0xa7 # ..stuvwx + .byte 0xa8,0xa9,0xea,0xeb,0xec,0xed,0xee,0xef # yz + .byte 0xf0,0xf1,0xf2,0xf3,0xf4,0xf5,0xf6,0xf7 + .byte 0xf8,0xf9,0xfa,0xfb,0xfc,0xfd,0xfe,0xff + +# +# startup-code at 0x10000, running in absolute addressing mode +# this is called either by the ipl loader or directly by PSW restart +# or linload or SALIPL +# + .org 0x10000 +ENTRY(startup) + j .Lep_startup_normal + .org 0x10008 +# +# This is a list of s390 kernel entry points. At address 0x1000f the number of +# valid entry points is stored. +# +# IMPORTANT: Do not change this table, it is s390 kernel ABI! +# + .ascii "S390EP" + .byte 0x00,0x01 +# +# kdump startup-code at 0x10010, running in 64 bit absolute addressing mode +# + .org 0x10010 +ENTRY(startup_kdump) + j .Lep_startup_kdump +.Lep_startup_normal: + mvi __LC_AR_MODE_ID,1 # set esame flag + slr %r0,%r0 # set cpuid to zero + lhi %r1,2 # mode 2 = esame (dump) + sigp %r1,%r0,0x12 # switch to esame mode + bras %r13,0f + .fill 16,4,0x0 +0: lmh %r0,%r15,0(%r13) # clear high-order half of gprs + sam31 # switch to 31 bit addressing mode + basr %r13,0 # get base +.LPG0: + xc 0x200(256),0x200 # partially clear lowcore + xc 0x300(256),0x300 + xc 0xe00(256),0xe00 + stck __LC_LAST_UPDATE_CLOCK + spt 6f-.LPG0(%r13) + mvc __LC_LAST_UPDATE_TIMER(8),6f-.LPG0(%r13) + xc __LC_STFL_FAC_LIST(8),__LC_STFL_FAC_LIST + # check capabilities against MARCH_{G5,Z900,Z990,Z9_109,Z10} + .insn s,0xb2b10000,0 # store facilities @ __LC_STFL_FAC_LIST + tm __LC_STFL_FAC_LIST,0x01 # stfle available ? + jz 0f + la %r0,1 + .insn s,0xb2b00000,__LC_STFL_FAC_LIST # store facility list extended + # verify if all required facilities are supported by the machine +0: la %r1,__LC_STFL_FAC_LIST + la %r2,3f+8-.LPG0(%r13) + l %r3,0(%r2) +1: l %r0,0(%r1) + n %r0,4(%r2) + cl %r0,4(%r2) + jne 2f + la %r1,4(%r1) + la %r2,4(%r2) + ahi %r3,-1 + jnz 1b + j 4f +2: l %r15,.Lstack-.LPG0(%r13) + ahi %r15,-96 + la %r2,.Lals_string-.LPG0(%r13) + l %r3,.Lsclp_print-.LPG0(%r13) + basr %r14,%r3 + lpsw 3f-.LPG0(%r13) # machine type not good enough, crash +.Lals_string: + .asciz "The Linux kernel requires more recent processor hardware" +.Lsclp_print: + .long _sclp_print_early +.Lstack: + .long 0x8000 + (1<<(PAGE_SHIFT+THREAD_ORDER)) + .align 16 +3: .long 0x000a0000,0x8badcccc + +# List of facilities that are required. If not all facilities are present +# the kernel will crash. Format is number of facility words with bits set, +# followed by the facility words. + +#if defined(CONFIG_MARCH_Z13) + .long 3, 0xc100eff2, 0xf46ce800, 0x00400000 +#elif defined(CONFIG_MARCH_ZEC12) + .long 3, 0xc100eff2, 0xf46ce800, 0x00400000 +#elif defined(CONFIG_MARCH_Z196) + .long 2, 0xc100eff2, 0xf46c0000 +#elif defined(CONFIG_MARCH_Z10) + .long 2, 0xc100eff2, 0xf0680000 +#elif defined(CONFIG_MARCH_Z9_109) + .long 1, 0xc100efc2 +#elif defined(CONFIG_MARCH_Z990) + .long 1, 0xc0002000 +#elif defined(CONFIG_MARCH_Z900) + .long 1, 0xc0000000 +#endif +4: + /* Continue with 64bit startup code in head64.S */ + sam64 # switch to 64 bit mode + jg startup_continue + + .align 8 +6: .long 0x7fffffff,0xffffffff + +#include "head_kdump.S" + +# +# params at 10400 (setup.h) +# + .org PARMAREA + .long 0,0 # IPL_DEVICE + .long 0,0 # INITRD_START + .long 0,0 # INITRD_SIZE + .long 0,0 # OLDMEM_BASE + .long 0,0 # OLDMEM_SIZE + + .org COMMAND_LINE + .byte "root=/dev/ram0 ro" + .byte 0 + + .org 0x11000 diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S new file mode 100644 index 000000000..d7c005075 --- /dev/null +++ b/arch/s390/kernel/head64.S @@ -0,0 +1,105 @@ +/* + * Copyright IBM Corp. 1999, 2010 + * + * Author(s): Hartmut Penner <hp@de.ibm.com> + * Martin Schwidefsky <schwidefsky@de.ibm.com> + * Rob van der Heij <rvdhei@iae.nl> + * Heiko Carstens <heiko.carstens@de.ibm.com> + * + */ + +#include <linux/init.h> +#include <linux/linkage.h> +#include <asm/asm-offsets.h> +#include <asm/thread_info.h> +#include <asm/page.h> + +__HEAD +ENTRY(startup_continue) + larl %r1,sched_clock_base_cc + mvc 0(8,%r1),__LC_LAST_UPDATE_CLOCK + larl %r13,.LPG1 # get base + lctlg %c0,%c15,.Lctl-.LPG1(%r13) # load control registers + lg %r12,.Lparmaddr-.LPG1(%r13) # pointer to parameter area + # move IPL device to lowcore + lghi %r0,__LC_PASTE + stg %r0,__LC_VDSO_PER_CPU +# +# Setup stack +# + larl %r15,init_thread_union + stg %r15,__LC_THREAD_INFO # cache thread info in lowcore + lg %r14,__TI_task(%r15) # cache current in lowcore + stg %r14,__LC_CURRENT + aghi %r15,1<<(PAGE_SHIFT+THREAD_ORDER) # init_task_union + THREAD_SIZE + stg %r15,__LC_KERNEL_STACK # set end of kernel stack + aghi %r15,-160 +# +# Save ipl parameters, clear bss memory, initialize storage key for kernel pages, +# and create a kernel NSS if the SAVESYS= parm is defined +# + brasl %r14,startup_init + lpswe .Lentry-.LPG1(13) # jump to _stext in primary-space, + # virtual and never return ... + .align 16 +.LPG1: +.Lentry:.quad 0x0000000180000000,_stext +.Lctl: .quad 0x04040000 # cr0: AFP registers & secondary space + .quad 0 # cr1: primary space segment table + .quad .Lduct # cr2: dispatchable unit control table + .quad 0 # cr3: instruction authorization + .quad 0 # cr4: instruction authorization + .quad .Lduct # cr5: primary-aste origin + .quad 0 # cr6: I/O interrupts + .quad 0 # cr7: secondary space segment table + .quad 0 # cr8: access registers translation + .quad 0 # cr9: tracing off + .quad 0 # cr10: tracing off + .quad 0 # cr11: tracing off + .quad 0 # cr12: tracing off + .quad 0 # cr13: home space segment table + .quad 0xc0000000 # cr14: machine check handling off + .quad .Llinkage_stack # cr15: linkage stack operations +.Lpcmsk:.quad 0x0000000180000000 +.L4malign:.quad 0xffffffffffc00000 +.Lscan2g:.quad 0x80000000 + 0x20000 - 8 # 2GB + 128K - 8 +.Lnop: .long 0x07000700 +.Lparmaddr: + .quad PARMAREA + .align 64 +.Lduct: .long 0,.Laste,.Laste,0,.Lduald,0,0,0 + .long 0,0,0,0,0,0,0,0 +.Laste: .quad 0,0xffffffffffffffff,0,0,0,0,0,0 + .align 128 +.Lduald:.rept 8 + .long 0x80000000,0,0,0 # invalid access-list entries + .endr +.Llinkage_stack: + .long 0,0,0x89000000,0,0,0,0x8a000000,0 + +ENTRY(_ehead) + + .org 0x100000 - 0x11000 # head.o ends at 0x11000 +# +# startup-code, running in absolute addressing mode +# +ENTRY(_stext) + basr %r13,0 # get base +.LPG3: +# check control registers + stctg %c0,%c15,0(%r15) + oi 6(%r15),0x60 # enable sigp emergency & external call + oi 4(%r15),0x10 # switch on low address proctection + lctlg %c0,%c15,0(%r15) + + lam 0,15,.Laregs-.LPG3(%r13) # load acrs needed by uaccess + brasl %r14,start_kernel # go to C code +# +# We returned from start_kernel ?!? PANIK +# + basr %r13,0 + lpswe .Ldw-.(%r13) # load disabled wait psw + + .align 8 +.Ldw: .quad 0x0002000180000000,0x0000000000000000 +.Laregs:.long 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 diff --git a/arch/s390/kernel/head_kdump.S b/arch/s390/kernel/head_kdump.S new file mode 100644 index 000000000..d05950f02 --- /dev/null +++ b/arch/s390/kernel/head_kdump.S @@ -0,0 +1,100 @@ +/* + * S390 kdump lowlevel functions (new kernel) + * + * Copyright IBM Corp. 2011 + * Author(s): Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#include <asm/sigp.h> + +#define DATAMOVER_ADDR 0x4000 +#define COPY_PAGE_ADDR 0x6000 + +#ifdef CONFIG_CRASH_DUMP + +# +# kdump entry (new kernel - not yet relocated) +# +# Note: This code has to be position independent +# + +.align 2 +.Lep_startup_kdump: + lhi %r1,2 # mode 2 = esame (dump) + sigp %r1,%r0,SIGP_SET_ARCHITECTURE # Switch to esame mode + sam64 # Switch to 64 bit addressing + basr %r13,0 +.Lbase: + larl %r2,.Lbase_addr # Check, if we have been + lg %r2,0(%r2) # already relocated: + clgr %r2,%r13 # + jne .Lrelocate # No : Start data mover + lghi %r2,0 # Yes: Start kdump kernel + brasl %r14,startup_kdump_relocated + +.Lrelocate: + larl %r4,startup + lg %r2,0x418(%r4) # Get kdump base + lg %r3,0x420(%r4) # Get kdump size + + larl %r10,.Lcopy_start # Source of data mover + lghi %r8,DATAMOVER_ADDR # Target of data mover + mvc 0(256,%r8),0(%r10) # Copy data mover code + + agr %r8,%r2 # Copy data mover to + mvc 0(256,%r8),0(%r10) # reserved mem + + lghi %r14,DATAMOVER_ADDR # Jump to copied data mover + basr %r14,%r14 +.Lbase_addr: + .quad .Lbase + +# +# kdump data mover code (runs at address DATAMOVER_ADDR) +# +# r2: kdump base address +# r3: kdump size +# +.Lcopy_start: + basr %r13,0 # Base +0: + lgr %r11,%r2 # Save kdump base address + lgr %r12,%r2 + agr %r12,%r3 # Compute kdump end address + + lghi %r5,0 + lghi %r10,COPY_PAGE_ADDR # Load copy page address +1: + mvc 0(256,%r10),0(%r5) # Copy old kernel to tmp + mvc 0(256,%r5),0(%r11) # Copy new kernel to old + mvc 0(256,%r11),0(%r10) # Copy tmp to new + aghi %r11,256 + aghi %r5,256 + clgr %r11,%r12 + jl 1b + + lg %r14,.Lstartup_kdump-0b(%r13) + basr %r14,%r14 # Start relocated kernel +.Lstartup_kdump: + .long 0x00000000,0x00000000 + startup_kdump_relocated +.Lcopy_end: + +# +# Startup of kdump (relocated new kernel) +# +.align 2 +startup_kdump_relocated: + basr %r13,0 +0: lpswe .Lrestart_psw-0b(%r13) # Start new kernel... +.align 8 +.Lrestart_psw: + .quad 0x0000000080000000,0x0000000000000000 + startup +#else +.align 2 +.Lep_startup_kdump: + larl %r13,startup_kdump_crash + lpswe 0(%r13) +.align 8 +startup_kdump_crash: + .quad 0x0002000080000000,0x0000000000000000 + startup_kdump_crash +#endif /* CONFIG_CRASH_DUMP */ diff --git a/arch/s390/kernel/idle.c b/arch/s390/kernel/idle.c new file mode 100644 index 000000000..7a55c29b0 --- /dev/null +++ b/arch/s390/kernel/idle.c @@ -0,0 +1,125 @@ +/* + * Idle functions for s390. + * + * Copyright IBM Corp. 2014 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#include <linux/kernel.h> +#include <linux/kernel_stat.h> +#include <linux/kprobes.h> +#include <linux/notifier.h> +#include <linux/init.h> +#include <linux/cpu.h> +#include <asm/cputime.h> +#include <asm/nmi.h> +#include <asm/smp.h> +#include "entry.h" + +static DEFINE_PER_CPU(struct s390_idle_data, s390_idle); + +void enabled_wait(void) +{ + struct s390_idle_data *idle = this_cpu_ptr(&s390_idle); + unsigned long long idle_time; + unsigned long psw_mask; + + trace_hardirqs_on(); + + /* Wait for external, I/O or machine check interrupt. */ + psw_mask = PSW_KERNEL_BITS | PSW_MASK_WAIT | PSW_MASK_DAT | + PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK; + clear_cpu_flag(CIF_NOHZ_DELAY); + + /* Call the assembler magic in entry.S */ + psw_idle(idle, psw_mask); + + trace_hardirqs_off(); + + /* Account time spent with enabled wait psw loaded as idle time. */ + write_seqcount_begin(&idle->seqcount); + idle_time = idle->clock_idle_exit - idle->clock_idle_enter; + idle->clock_idle_enter = idle->clock_idle_exit = 0ULL; + idle->idle_time += idle_time; + idle->idle_count++; + account_idle_time(idle_time); + write_seqcount_end(&idle->seqcount); +} +NOKPROBE_SYMBOL(enabled_wait); + +static ssize_t show_idle_count(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id); + unsigned long long idle_count; + unsigned int seq; + + do { + seq = read_seqcount_begin(&idle->seqcount); + idle_count = ACCESS_ONCE(idle->idle_count); + if (ACCESS_ONCE(idle->clock_idle_enter)) + idle_count++; + } while (read_seqcount_retry(&idle->seqcount, seq)); + return sprintf(buf, "%llu\n", idle_count); +} +DEVICE_ATTR(idle_count, 0444, show_idle_count, NULL); + +static ssize_t show_idle_time(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id); + unsigned long long now, idle_time, idle_enter, idle_exit; + unsigned int seq; + + do { + now = get_tod_clock(); + seq = read_seqcount_begin(&idle->seqcount); + idle_time = ACCESS_ONCE(idle->idle_time); + idle_enter = ACCESS_ONCE(idle->clock_idle_enter); + idle_exit = ACCESS_ONCE(idle->clock_idle_exit); + } while (read_seqcount_retry(&idle->seqcount, seq)); + idle_time += idle_enter ? ((idle_exit ? : now) - idle_enter) : 0; + return sprintf(buf, "%llu\n", idle_time >> 12); +} +DEVICE_ATTR(idle_time_us, 0444, show_idle_time, NULL); + +cputime64_t arch_cpu_idle_time(int cpu) +{ + struct s390_idle_data *idle = &per_cpu(s390_idle, cpu); + unsigned long long now, idle_enter, idle_exit; + unsigned int seq; + + do { + now = get_tod_clock(); + seq = read_seqcount_begin(&idle->seqcount); + idle_enter = ACCESS_ONCE(idle->clock_idle_enter); + idle_exit = ACCESS_ONCE(idle->clock_idle_exit); + } while (read_seqcount_retry(&idle->seqcount, seq)); + return idle_enter ? ((idle_exit ?: now) - idle_enter) : 0; +} + +void arch_cpu_idle_enter(void) +{ + local_mcck_disable(); +} + +void arch_cpu_idle(void) +{ + if (!test_cpu_flag(CIF_MCCK_PENDING)) + /* Halt the cpu and keep track of cpu time accounting. */ + enabled_wait(); + local_irq_enable(); +} + +void arch_cpu_idle_exit(void) +{ + local_mcck_enable(); + if (test_cpu_flag(CIF_MCCK_PENDING)) + s390_handle_mcck(); +} + +void arch_cpu_idle_dead(void) +{ + cpu_die(); +} diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c new file mode 100644 index 000000000..52fbef91d --- /dev/null +++ b/arch/s390/kernel/ipl.c @@ -0,0 +1,2064 @@ +/* + * ipl/reipl/dump support for Linux on s390. + * + * Copyright IBM Corp. 2005, 2012 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + * Heiko Carstens <heiko.carstens@de.ibm.com> + * Volker Sameske <sameske@de.ibm.com> + */ + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/device.h> +#include <linux/delay.h> +#include <linux/reboot.h> +#include <linux/ctype.h> +#include <linux/fs.h> +#include <linux/gfp.h> +#include <linux/crash_dump.h> +#include <linux/debug_locks.h> +#include <asm/ipl.h> +#include <asm/smp.h> +#include <asm/setup.h> +#include <asm/cpcmd.h> +#include <asm/cio.h> +#include <asm/ebcdic.h> +#include <asm/reset.h> +#include <asm/sclp.h> +#include <asm/checksum.h> +#include <asm/debug.h> +#include <asm/os_info.h> +#include "entry.h" + +#define IPL_PARM_BLOCK_VERSION 0 + +#define IPL_UNKNOWN_STR "unknown" +#define IPL_CCW_STR "ccw" +#define IPL_FCP_STR "fcp" +#define IPL_FCP_DUMP_STR "fcp_dump" +#define IPL_NSS_STR "nss" + +#define DUMP_CCW_STR "ccw" +#define DUMP_FCP_STR "fcp" +#define DUMP_NONE_STR "none" + +/* + * Four shutdown trigger types are supported: + * - panic + * - halt + * - power off + * - reipl + * - restart + */ +#define ON_PANIC_STR "on_panic" +#define ON_HALT_STR "on_halt" +#define ON_POFF_STR "on_poff" +#define ON_REIPL_STR "on_reboot" +#define ON_RESTART_STR "on_restart" + +struct shutdown_action; +struct shutdown_trigger { + char *name; + struct shutdown_action *action; +}; + +/* + * The following shutdown action types are supported: + */ +#define SHUTDOWN_ACTION_IPL_STR "ipl" +#define SHUTDOWN_ACTION_REIPL_STR "reipl" +#define SHUTDOWN_ACTION_DUMP_STR "dump" +#define SHUTDOWN_ACTION_VMCMD_STR "vmcmd" +#define SHUTDOWN_ACTION_STOP_STR "stop" +#define SHUTDOWN_ACTION_DUMP_REIPL_STR "dump_reipl" + +struct shutdown_action { + char *name; + void (*fn) (struct shutdown_trigger *trigger); + int (*init) (void); + int init_rc; +}; + +static char *ipl_type_str(enum ipl_type type) +{ + switch (type) { + case IPL_TYPE_CCW: + return IPL_CCW_STR; + case IPL_TYPE_FCP: + return IPL_FCP_STR; + case IPL_TYPE_FCP_DUMP: + return IPL_FCP_DUMP_STR; + case IPL_TYPE_NSS: + return IPL_NSS_STR; + case IPL_TYPE_UNKNOWN: + default: + return IPL_UNKNOWN_STR; + } +} + +enum dump_type { + DUMP_TYPE_NONE = 1, + DUMP_TYPE_CCW = 2, + DUMP_TYPE_FCP = 4, +}; + +static char *dump_type_str(enum dump_type type) +{ + switch (type) { + case DUMP_TYPE_NONE: + return DUMP_NONE_STR; + case DUMP_TYPE_CCW: + return DUMP_CCW_STR; + case DUMP_TYPE_FCP: + return DUMP_FCP_STR; + default: + return NULL; + } +} + +/* + * Must be in data section since the bss section + * is not cleared when these are accessed. + */ +static u16 ipl_devno __attribute__((__section__(".data"))) = 0; +u32 ipl_flags __attribute__((__section__(".data"))) = 0; + +enum ipl_method { + REIPL_METHOD_CCW_CIO, + REIPL_METHOD_CCW_DIAG, + REIPL_METHOD_CCW_VM, + REIPL_METHOD_FCP_RO_DIAG, + REIPL_METHOD_FCP_RW_DIAG, + REIPL_METHOD_FCP_RO_VM, + REIPL_METHOD_FCP_DUMP, + REIPL_METHOD_NSS, + REIPL_METHOD_NSS_DIAG, + REIPL_METHOD_DEFAULT, +}; + +enum dump_method { + DUMP_METHOD_NONE, + DUMP_METHOD_CCW_CIO, + DUMP_METHOD_CCW_DIAG, + DUMP_METHOD_CCW_VM, + DUMP_METHOD_FCP_DIAG, +}; + +static int diag308_set_works = 0; + +static struct ipl_parameter_block ipl_block; + +static int reipl_capabilities = IPL_TYPE_UNKNOWN; + +static enum ipl_type reipl_type = IPL_TYPE_UNKNOWN; +static enum ipl_method reipl_method = REIPL_METHOD_DEFAULT; +static struct ipl_parameter_block *reipl_block_fcp; +static struct ipl_parameter_block *reipl_block_ccw; +static struct ipl_parameter_block *reipl_block_nss; +static struct ipl_parameter_block *reipl_block_actual; + +static int dump_capabilities = DUMP_TYPE_NONE; +static enum dump_type dump_type = DUMP_TYPE_NONE; +static enum dump_method dump_method = DUMP_METHOD_NONE; +static struct ipl_parameter_block *dump_block_fcp; +static struct ipl_parameter_block *dump_block_ccw; + +static struct sclp_ipl_info sclp_ipl_info; + +int diag308(unsigned long subcode, void *addr) +{ + register unsigned long _addr asm("0") = (unsigned long) addr; + register unsigned long _rc asm("1") = 0; + + asm volatile( + " diag %0,%2,0x308\n" + "0:\n" + EX_TABLE(0b,0b) + : "+d" (_addr), "+d" (_rc) + : "d" (subcode) : "cc", "memory"); + return _rc; +} +EXPORT_SYMBOL_GPL(diag308); + +/* SYSFS */ + +#define IPL_ATTR_SHOW_FN(_prefix, _name, _format, args...) \ +static ssize_t sys_##_prefix##_##_name##_show(struct kobject *kobj, \ + struct kobj_attribute *attr, \ + char *page) \ +{ \ + return snprintf(page, PAGE_SIZE, _format, ##args); \ +} + +#define DEFINE_IPL_ATTR_RO(_prefix, _name, _format, _value) \ +IPL_ATTR_SHOW_FN(_prefix, _name, _format, _value) \ +static struct kobj_attribute sys_##_prefix##_##_name##_attr = \ + __ATTR(_name, S_IRUGO, sys_##_prefix##_##_name##_show, NULL) + +#define DEFINE_IPL_ATTR_RW(_prefix, _name, _fmt_out, _fmt_in, _value) \ +IPL_ATTR_SHOW_FN(_prefix, _name, _fmt_out, (unsigned long long) _value) \ +static ssize_t sys_##_prefix##_##_name##_store(struct kobject *kobj, \ + struct kobj_attribute *attr, \ + const char *buf, size_t len) \ +{ \ + unsigned long long value; \ + if (sscanf(buf, _fmt_in, &value) != 1) \ + return -EINVAL; \ + _value = value; \ + return len; \ +} \ +static struct kobj_attribute sys_##_prefix##_##_name##_attr = \ + __ATTR(_name,(S_IRUGO | S_IWUSR), \ + sys_##_prefix##_##_name##_show, \ + sys_##_prefix##_##_name##_store) + +#define DEFINE_IPL_ATTR_STR_RW(_prefix, _name, _fmt_out, _fmt_in, _value)\ +IPL_ATTR_SHOW_FN(_prefix, _name, _fmt_out, _value) \ +static ssize_t sys_##_prefix##_##_name##_store(struct kobject *kobj, \ + struct kobj_attribute *attr, \ + const char *buf, size_t len) \ +{ \ + strncpy(_value, buf, sizeof(_value) - 1); \ + strim(_value); \ + return len; \ +} \ +static struct kobj_attribute sys_##_prefix##_##_name##_attr = \ + __ATTR(_name,(S_IRUGO | S_IWUSR), \ + sys_##_prefix##_##_name##_show, \ + sys_##_prefix##_##_name##_store) + +static void make_attrs_ro(struct attribute **attrs) +{ + while (*attrs) { + (*attrs)->mode = S_IRUGO; + attrs++; + } +} + +/* + * ipl section + */ + +static __init enum ipl_type get_ipl_type(void) +{ + struct ipl_parameter_block *ipl = IPL_PARMBLOCK_START; + + if (ipl_flags & IPL_NSS_VALID) + return IPL_TYPE_NSS; + if (!(ipl_flags & IPL_DEVNO_VALID)) + return IPL_TYPE_UNKNOWN; + if (!(ipl_flags & IPL_PARMBLOCK_VALID)) + return IPL_TYPE_CCW; + if (ipl->hdr.version > IPL_MAX_SUPPORTED_VERSION) + return IPL_TYPE_UNKNOWN; + if (ipl->hdr.pbt != DIAG308_IPL_TYPE_FCP) + return IPL_TYPE_UNKNOWN; + if (ipl->ipl_info.fcp.opt == DIAG308_IPL_OPT_DUMP) + return IPL_TYPE_FCP_DUMP; + return IPL_TYPE_FCP; +} + +struct ipl_info ipl_info; +EXPORT_SYMBOL_GPL(ipl_info); + +static ssize_t ipl_type_show(struct kobject *kobj, struct kobj_attribute *attr, + char *page) +{ + return sprintf(page, "%s\n", ipl_type_str(ipl_info.type)); +} + +static struct kobj_attribute sys_ipl_type_attr = __ATTR_RO(ipl_type); + +/* VM IPL PARM routines */ +static size_t reipl_get_ascii_vmparm(char *dest, size_t size, + const struct ipl_parameter_block *ipb) +{ + int i; + size_t len; + char has_lowercase = 0; + + len = 0; + if ((ipb->ipl_info.ccw.vm_flags & DIAG308_VM_FLAGS_VP_VALID) && + (ipb->ipl_info.ccw.vm_parm_len > 0)) { + + len = min_t(size_t, size - 1, ipb->ipl_info.ccw.vm_parm_len); + memcpy(dest, ipb->ipl_info.ccw.vm_parm, len); + /* If at least one character is lowercase, we assume mixed + * case; otherwise we convert everything to lowercase. + */ + for (i = 0; i < len; i++) + if ((dest[i] > 0x80 && dest[i] < 0x8a) || /* a-i */ + (dest[i] > 0x90 && dest[i] < 0x9a) || /* j-r */ + (dest[i] > 0xa1 && dest[i] < 0xaa)) { /* s-z */ + has_lowercase = 1; + break; + } + if (!has_lowercase) + EBC_TOLOWER(dest, len); + EBCASC(dest, len); + } + dest[len] = 0; + + return len; +} + +size_t append_ipl_vmparm(char *dest, size_t size) +{ + size_t rc; + + rc = 0; + if (diag308_set_works && (ipl_block.hdr.pbt == DIAG308_IPL_TYPE_CCW)) + rc = reipl_get_ascii_vmparm(dest, size, &ipl_block); + else + dest[0] = 0; + return rc; +} + +static ssize_t ipl_vm_parm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + char parm[DIAG308_VMPARM_SIZE + 1] = {}; + + append_ipl_vmparm(parm, sizeof(parm)); + return sprintf(page, "%s\n", parm); +} + +static size_t scpdata_length(const char* buf, size_t count) +{ + while (count) { + if (buf[count - 1] != '\0' && buf[count - 1] != ' ') + break; + count--; + } + return count; +} + +static size_t reipl_append_ascii_scpdata(char *dest, size_t size, + const struct ipl_parameter_block *ipb) +{ + size_t count; + size_t i; + int has_lowercase; + + count = min(size - 1, scpdata_length(ipb->ipl_info.fcp.scp_data, + ipb->ipl_info.fcp.scp_data_len)); + if (!count) + goto out; + + has_lowercase = 0; + for (i = 0; i < count; i++) { + if (!isascii(ipb->ipl_info.fcp.scp_data[i])) { + count = 0; + goto out; + } + if (!has_lowercase && islower(ipb->ipl_info.fcp.scp_data[i])) + has_lowercase = 1; + } + + if (has_lowercase) + memcpy(dest, ipb->ipl_info.fcp.scp_data, count); + else + for (i = 0; i < count; i++) + dest[i] = tolower(ipb->ipl_info.fcp.scp_data[i]); +out: + dest[count] = '\0'; + return count; +} + +size_t append_ipl_scpdata(char *dest, size_t len) +{ + size_t rc; + + rc = 0; + if (ipl_block.hdr.pbt == DIAG308_IPL_TYPE_FCP) + rc = reipl_append_ascii_scpdata(dest, len, &ipl_block); + else + dest[0] = 0; + return rc; +} + + +static struct kobj_attribute sys_ipl_vm_parm_attr = + __ATTR(parm, S_IRUGO, ipl_vm_parm_show, NULL); + +static ssize_t sys_ipl_device_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + struct ipl_parameter_block *ipl = IPL_PARMBLOCK_START; + + switch (ipl_info.type) { + case IPL_TYPE_CCW: + return sprintf(page, "0.0.%04x\n", ipl_devno); + case IPL_TYPE_FCP: + case IPL_TYPE_FCP_DUMP: + return sprintf(page, "0.0.%04x\n", ipl->ipl_info.fcp.devno); + default: + return 0; + } +} + +static struct kobj_attribute sys_ipl_device_attr = + __ATTR(device, S_IRUGO, sys_ipl_device_show, NULL); + +static ssize_t ipl_parameter_read(struct file *filp, struct kobject *kobj, + struct bin_attribute *attr, char *buf, + loff_t off, size_t count) +{ + return memory_read_from_buffer(buf, count, &off, IPL_PARMBLOCK_START, + IPL_PARMBLOCK_SIZE); +} +static struct bin_attribute ipl_parameter_attr = + __BIN_ATTR(binary_parameter, S_IRUGO, ipl_parameter_read, NULL, + PAGE_SIZE); + +static ssize_t ipl_scp_data_read(struct file *filp, struct kobject *kobj, + struct bin_attribute *attr, char *buf, + loff_t off, size_t count) +{ + unsigned int size = IPL_PARMBLOCK_START->ipl_info.fcp.scp_data_len; + void *scp_data = &IPL_PARMBLOCK_START->ipl_info.fcp.scp_data; + + return memory_read_from_buffer(buf, count, &off, scp_data, size); +} +static struct bin_attribute ipl_scp_data_attr = + __BIN_ATTR(scp_data, S_IRUGO, ipl_scp_data_read, NULL, PAGE_SIZE); + +static struct bin_attribute *ipl_fcp_bin_attrs[] = { + &ipl_parameter_attr, + &ipl_scp_data_attr, + NULL, +}; + +/* FCP ipl device attributes */ + +DEFINE_IPL_ATTR_RO(ipl_fcp, wwpn, "0x%016llx\n", (unsigned long long) + IPL_PARMBLOCK_START->ipl_info.fcp.wwpn); +DEFINE_IPL_ATTR_RO(ipl_fcp, lun, "0x%016llx\n", (unsigned long long) + IPL_PARMBLOCK_START->ipl_info.fcp.lun); +DEFINE_IPL_ATTR_RO(ipl_fcp, bootprog, "%lld\n", (unsigned long long) + IPL_PARMBLOCK_START->ipl_info.fcp.bootprog); +DEFINE_IPL_ATTR_RO(ipl_fcp, br_lba, "%lld\n", (unsigned long long) + IPL_PARMBLOCK_START->ipl_info.fcp.br_lba); + +static ssize_t ipl_ccw_loadparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + char loadparm[LOADPARM_LEN + 1] = {}; + + if (!sclp_ipl_info.is_valid) + return sprintf(page, "#unknown#\n"); + memcpy(loadparm, &sclp_ipl_info.loadparm, LOADPARM_LEN); + EBCASC(loadparm, LOADPARM_LEN); + strim(loadparm); + return sprintf(page, "%s\n", loadparm); +} + +static struct kobj_attribute sys_ipl_ccw_loadparm_attr = + __ATTR(loadparm, 0444, ipl_ccw_loadparm_show, NULL); + +static struct attribute *ipl_fcp_attrs[] = { + &sys_ipl_type_attr.attr, + &sys_ipl_device_attr.attr, + &sys_ipl_fcp_wwpn_attr.attr, + &sys_ipl_fcp_lun_attr.attr, + &sys_ipl_fcp_bootprog_attr.attr, + &sys_ipl_fcp_br_lba_attr.attr, + &sys_ipl_ccw_loadparm_attr.attr, + NULL, +}; + +static struct attribute_group ipl_fcp_attr_group = { + .attrs = ipl_fcp_attrs, + .bin_attrs = ipl_fcp_bin_attrs, +}; + +/* CCW ipl device attributes */ + +static struct attribute *ipl_ccw_attrs_vm[] = { + &sys_ipl_type_attr.attr, + &sys_ipl_device_attr.attr, + &sys_ipl_ccw_loadparm_attr.attr, + &sys_ipl_vm_parm_attr.attr, + NULL, +}; + +static struct attribute *ipl_ccw_attrs_lpar[] = { + &sys_ipl_type_attr.attr, + &sys_ipl_device_attr.attr, + &sys_ipl_ccw_loadparm_attr.attr, + NULL, +}; + +static struct attribute_group ipl_ccw_attr_group_vm = { + .attrs = ipl_ccw_attrs_vm, +}; + +static struct attribute_group ipl_ccw_attr_group_lpar = { + .attrs = ipl_ccw_attrs_lpar +}; + +/* NSS ipl device attributes */ + +DEFINE_IPL_ATTR_RO(ipl_nss, name, "%s\n", kernel_nss_name); + +static struct attribute *ipl_nss_attrs[] = { + &sys_ipl_type_attr.attr, + &sys_ipl_nss_name_attr.attr, + &sys_ipl_ccw_loadparm_attr.attr, + &sys_ipl_vm_parm_attr.attr, + NULL, +}; + +static struct attribute_group ipl_nss_attr_group = { + .attrs = ipl_nss_attrs, +}; + +/* UNKNOWN ipl device attributes */ + +static struct attribute *ipl_unknown_attrs[] = { + &sys_ipl_type_attr.attr, + NULL, +}; + +static struct attribute_group ipl_unknown_attr_group = { + .attrs = ipl_unknown_attrs, +}; + +static struct kset *ipl_kset; + +static void __ipl_run(void *unused) +{ + diag308(DIAG308_IPL, NULL); + if (MACHINE_IS_VM) + __cpcmd("IPL", NULL, 0, NULL); + else if (ipl_info.type == IPL_TYPE_CCW) + reipl_ccw_dev(&ipl_info.data.ccw.dev_id); +} + +static void ipl_run(struct shutdown_trigger *trigger) +{ + smp_call_ipl_cpu(__ipl_run, NULL); +} + +static int __init ipl_init(void) +{ + int rc; + + ipl_kset = kset_create_and_add("ipl", NULL, firmware_kobj); + if (!ipl_kset) { + rc = -ENOMEM; + goto out; + } + switch (ipl_info.type) { + case IPL_TYPE_CCW: + if (MACHINE_IS_VM) + rc = sysfs_create_group(&ipl_kset->kobj, + &ipl_ccw_attr_group_vm); + else + rc = sysfs_create_group(&ipl_kset->kobj, + &ipl_ccw_attr_group_lpar); + break; + case IPL_TYPE_FCP: + case IPL_TYPE_FCP_DUMP: + rc = sysfs_create_group(&ipl_kset->kobj, &ipl_fcp_attr_group); + break; + case IPL_TYPE_NSS: + rc = sysfs_create_group(&ipl_kset->kobj, &ipl_nss_attr_group); + break; + default: + rc = sysfs_create_group(&ipl_kset->kobj, + &ipl_unknown_attr_group); + break; + } +out: + if (rc) + panic("ipl_init failed: rc = %i\n", rc); + + return 0; +} + +static struct shutdown_action __refdata ipl_action = { + .name = SHUTDOWN_ACTION_IPL_STR, + .fn = ipl_run, + .init = ipl_init, +}; + +/* + * reipl shutdown action: Reboot Linux on shutdown. + */ + +/* VM IPL PARM attributes */ +static ssize_t reipl_generic_vmparm_show(struct ipl_parameter_block *ipb, + char *page) +{ + char vmparm[DIAG308_VMPARM_SIZE + 1] = {}; + + reipl_get_ascii_vmparm(vmparm, sizeof(vmparm), ipb); + return sprintf(page, "%s\n", vmparm); +} + +static ssize_t reipl_generic_vmparm_store(struct ipl_parameter_block *ipb, + size_t vmparm_max, + const char *buf, size_t len) +{ + int i, ip_len; + + /* ignore trailing newline */ + ip_len = len; + if ((len > 0) && (buf[len - 1] == '\n')) + ip_len--; + + if (ip_len > vmparm_max) + return -EINVAL; + + /* parm is used to store kernel options, check for common chars */ + for (i = 0; i < ip_len; i++) + if (!(isalnum(buf[i]) || isascii(buf[i]) || isprint(buf[i]))) + return -EINVAL; + + memset(ipb->ipl_info.ccw.vm_parm, 0, DIAG308_VMPARM_SIZE); + ipb->ipl_info.ccw.vm_parm_len = ip_len; + if (ip_len > 0) { + ipb->ipl_info.ccw.vm_flags |= DIAG308_VM_FLAGS_VP_VALID; + memcpy(ipb->ipl_info.ccw.vm_parm, buf, ip_len); + ASCEBC(ipb->ipl_info.ccw.vm_parm, ip_len); + } else { + ipb->ipl_info.ccw.vm_flags &= ~DIAG308_VM_FLAGS_VP_VALID; + } + + return len; +} + +/* NSS wrapper */ +static ssize_t reipl_nss_vmparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return reipl_generic_vmparm_show(reipl_block_nss, page); +} + +static ssize_t reipl_nss_vmparm_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return reipl_generic_vmparm_store(reipl_block_nss, 56, buf, len); +} + +/* CCW wrapper */ +static ssize_t reipl_ccw_vmparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return reipl_generic_vmparm_show(reipl_block_ccw, page); +} + +static ssize_t reipl_ccw_vmparm_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return reipl_generic_vmparm_store(reipl_block_ccw, 64, buf, len); +} + +static struct kobj_attribute sys_reipl_nss_vmparm_attr = + __ATTR(parm, S_IRUGO | S_IWUSR, reipl_nss_vmparm_show, + reipl_nss_vmparm_store); +static struct kobj_attribute sys_reipl_ccw_vmparm_attr = + __ATTR(parm, S_IRUGO | S_IWUSR, reipl_ccw_vmparm_show, + reipl_ccw_vmparm_store); + +/* FCP reipl device attributes */ + +static ssize_t reipl_fcp_scpdata_read(struct file *filp, struct kobject *kobj, + struct bin_attribute *attr, + char *buf, loff_t off, size_t count) +{ + size_t size = reipl_block_fcp->ipl_info.fcp.scp_data_len; + void *scp_data = reipl_block_fcp->ipl_info.fcp.scp_data; + + return memory_read_from_buffer(buf, count, &off, scp_data, size); +} + +static ssize_t reipl_fcp_scpdata_write(struct file *filp, struct kobject *kobj, + struct bin_attribute *attr, + char *buf, loff_t off, size_t count) +{ + size_t padding; + size_t scpdata_len; + + if (off < 0) + return -EINVAL; + + if (off >= DIAG308_SCPDATA_SIZE) + return -ENOSPC; + + if (count > DIAG308_SCPDATA_SIZE - off) + count = DIAG308_SCPDATA_SIZE - off; + + memcpy(reipl_block_fcp->ipl_info.fcp.scp_data, buf + off, count); + scpdata_len = off + count; + + if (scpdata_len % 8) { + padding = 8 - (scpdata_len % 8); + memset(reipl_block_fcp->ipl_info.fcp.scp_data + scpdata_len, + 0, padding); + scpdata_len += padding; + } + + reipl_block_fcp->ipl_info.fcp.scp_data_len = scpdata_len; + reipl_block_fcp->hdr.len = IPL_PARM_BLK_FCP_LEN + scpdata_len; + reipl_block_fcp->hdr.blk0_len = IPL_PARM_BLK0_FCP_LEN + scpdata_len; + + return count; +} +static struct bin_attribute sys_reipl_fcp_scp_data_attr = + __BIN_ATTR(scp_data, (S_IRUGO | S_IWUSR), reipl_fcp_scpdata_read, + reipl_fcp_scpdata_write, PAGE_SIZE); + +static struct bin_attribute *reipl_fcp_bin_attrs[] = { + &sys_reipl_fcp_scp_data_attr, + NULL, +}; + +DEFINE_IPL_ATTR_RW(reipl_fcp, wwpn, "0x%016llx\n", "%llx\n", + reipl_block_fcp->ipl_info.fcp.wwpn); +DEFINE_IPL_ATTR_RW(reipl_fcp, lun, "0x%016llx\n", "%llx\n", + reipl_block_fcp->ipl_info.fcp.lun); +DEFINE_IPL_ATTR_RW(reipl_fcp, bootprog, "%lld\n", "%lld\n", + reipl_block_fcp->ipl_info.fcp.bootprog); +DEFINE_IPL_ATTR_RW(reipl_fcp, br_lba, "%lld\n", "%lld\n", + reipl_block_fcp->ipl_info.fcp.br_lba); +DEFINE_IPL_ATTR_RW(reipl_fcp, device, "0.0.%04llx\n", "0.0.%llx\n", + reipl_block_fcp->ipl_info.fcp.devno); + +static void reipl_get_ascii_loadparm(char *loadparm, + struct ipl_parameter_block *ibp) +{ + memcpy(loadparm, ibp->hdr.loadparm, LOADPARM_LEN); + EBCASC(loadparm, LOADPARM_LEN); + loadparm[LOADPARM_LEN] = 0; + strim(loadparm); +} + +static ssize_t reipl_generic_loadparm_show(struct ipl_parameter_block *ipb, + char *page) +{ + char buf[LOADPARM_LEN + 1]; + + reipl_get_ascii_loadparm(buf, ipb); + return sprintf(page, "%s\n", buf); +} + +static ssize_t reipl_generic_loadparm_store(struct ipl_parameter_block *ipb, + const char *buf, size_t len) +{ + int i, lp_len; + + /* ignore trailing newline */ + lp_len = len; + if ((len > 0) && (buf[len - 1] == '\n')) + lp_len--; + /* loadparm can have max 8 characters and must not start with a blank */ + if ((lp_len > LOADPARM_LEN) || ((lp_len > 0) && (buf[0] == ' '))) + return -EINVAL; + /* loadparm can only contain "a-z,A-Z,0-9,SP,." */ + for (i = 0; i < lp_len; i++) { + if (isalpha(buf[i]) || isdigit(buf[i]) || (buf[i] == ' ') || + (buf[i] == '.')) + continue; + return -EINVAL; + } + /* initialize loadparm with blanks */ + memset(ipb->hdr.loadparm, ' ', LOADPARM_LEN); + /* copy and convert to ebcdic */ + memcpy(ipb->hdr.loadparm, buf, lp_len); + ASCEBC(ipb->hdr.loadparm, LOADPARM_LEN); + return len; +} + +/* FCP wrapper */ +static ssize_t reipl_fcp_loadparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return reipl_generic_loadparm_show(reipl_block_fcp, page); +} + +static ssize_t reipl_fcp_loadparm_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return reipl_generic_loadparm_store(reipl_block_fcp, buf, len); +} + +static struct kobj_attribute sys_reipl_fcp_loadparm_attr = + __ATTR(loadparm, S_IRUGO | S_IWUSR, reipl_fcp_loadparm_show, + reipl_fcp_loadparm_store); + +static struct attribute *reipl_fcp_attrs[] = { + &sys_reipl_fcp_device_attr.attr, + &sys_reipl_fcp_wwpn_attr.attr, + &sys_reipl_fcp_lun_attr.attr, + &sys_reipl_fcp_bootprog_attr.attr, + &sys_reipl_fcp_br_lba_attr.attr, + &sys_reipl_fcp_loadparm_attr.attr, + NULL, +}; + +static struct attribute_group reipl_fcp_attr_group = { + .attrs = reipl_fcp_attrs, + .bin_attrs = reipl_fcp_bin_attrs, +}; + +/* CCW reipl device attributes */ + +DEFINE_IPL_ATTR_RW(reipl_ccw, device, "0.0.%04llx\n", "0.0.%llx\n", + reipl_block_ccw->ipl_info.ccw.devno); + +/* NSS wrapper */ +static ssize_t reipl_nss_loadparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return reipl_generic_loadparm_show(reipl_block_nss, page); +} + +static ssize_t reipl_nss_loadparm_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return reipl_generic_loadparm_store(reipl_block_nss, buf, len); +} + +/* CCW wrapper */ +static ssize_t reipl_ccw_loadparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return reipl_generic_loadparm_show(reipl_block_ccw, page); +} + +static ssize_t reipl_ccw_loadparm_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return reipl_generic_loadparm_store(reipl_block_ccw, buf, len); +} + +static struct kobj_attribute sys_reipl_ccw_loadparm_attr = + __ATTR(loadparm, S_IRUGO | S_IWUSR, reipl_ccw_loadparm_show, + reipl_ccw_loadparm_store); + +static struct attribute *reipl_ccw_attrs_vm[] = { + &sys_reipl_ccw_device_attr.attr, + &sys_reipl_ccw_loadparm_attr.attr, + &sys_reipl_ccw_vmparm_attr.attr, + NULL, +}; + +static struct attribute *reipl_ccw_attrs_lpar[] = { + &sys_reipl_ccw_device_attr.attr, + &sys_reipl_ccw_loadparm_attr.attr, + NULL, +}; + +static struct attribute_group reipl_ccw_attr_group_vm = { + .name = IPL_CCW_STR, + .attrs = reipl_ccw_attrs_vm, +}; + +static struct attribute_group reipl_ccw_attr_group_lpar = { + .name = IPL_CCW_STR, + .attrs = reipl_ccw_attrs_lpar, +}; + + +/* NSS reipl device attributes */ +static void reipl_get_ascii_nss_name(char *dst, + struct ipl_parameter_block *ipb) +{ + memcpy(dst, ipb->ipl_info.ccw.nss_name, NSS_NAME_SIZE); + EBCASC(dst, NSS_NAME_SIZE); + dst[NSS_NAME_SIZE] = 0; +} + +static ssize_t reipl_nss_name_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + char nss_name[NSS_NAME_SIZE + 1] = {}; + + reipl_get_ascii_nss_name(nss_name, reipl_block_nss); + return sprintf(page, "%s\n", nss_name); +} + +static ssize_t reipl_nss_name_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + int nss_len; + + /* ignore trailing newline */ + nss_len = len; + if ((len > 0) && (buf[len - 1] == '\n')) + nss_len--; + + if (nss_len > NSS_NAME_SIZE) + return -EINVAL; + + memset(reipl_block_nss->ipl_info.ccw.nss_name, 0x40, NSS_NAME_SIZE); + if (nss_len > 0) { + reipl_block_nss->ipl_info.ccw.vm_flags |= + DIAG308_VM_FLAGS_NSS_VALID; + memcpy(reipl_block_nss->ipl_info.ccw.nss_name, buf, nss_len); + ASCEBC(reipl_block_nss->ipl_info.ccw.nss_name, nss_len); + EBC_TOUPPER(reipl_block_nss->ipl_info.ccw.nss_name, nss_len); + } else { + reipl_block_nss->ipl_info.ccw.vm_flags &= + ~DIAG308_VM_FLAGS_NSS_VALID; + } + + return len; +} + +static struct kobj_attribute sys_reipl_nss_name_attr = + __ATTR(name, S_IRUGO | S_IWUSR, reipl_nss_name_show, + reipl_nss_name_store); + +static struct kobj_attribute sys_reipl_nss_loadparm_attr = + __ATTR(loadparm, S_IRUGO | S_IWUSR, reipl_nss_loadparm_show, + reipl_nss_loadparm_store); + +static struct attribute *reipl_nss_attrs[] = { + &sys_reipl_nss_name_attr.attr, + &sys_reipl_nss_loadparm_attr.attr, + &sys_reipl_nss_vmparm_attr.attr, + NULL, +}; + +static struct attribute_group reipl_nss_attr_group = { + .name = IPL_NSS_STR, + .attrs = reipl_nss_attrs, +}; + +static void set_reipl_block_actual(struct ipl_parameter_block *reipl_block) +{ + reipl_block_actual = reipl_block; + os_info_entry_add(OS_INFO_REIPL_BLOCK, reipl_block_actual, + reipl_block->hdr.len); +} + +/* reipl type */ + +static int reipl_set_type(enum ipl_type type) +{ + if (!(reipl_capabilities & type)) + return -EINVAL; + + switch(type) { + case IPL_TYPE_CCW: + if (diag308_set_works) + reipl_method = REIPL_METHOD_CCW_DIAG; + else if (MACHINE_IS_VM) + reipl_method = REIPL_METHOD_CCW_VM; + else + reipl_method = REIPL_METHOD_CCW_CIO; + set_reipl_block_actual(reipl_block_ccw); + break; + case IPL_TYPE_FCP: + if (diag308_set_works) + reipl_method = REIPL_METHOD_FCP_RW_DIAG; + else if (MACHINE_IS_VM) + reipl_method = REIPL_METHOD_FCP_RO_VM; + else + reipl_method = REIPL_METHOD_FCP_RO_DIAG; + set_reipl_block_actual(reipl_block_fcp); + break; + case IPL_TYPE_FCP_DUMP: + reipl_method = REIPL_METHOD_FCP_DUMP; + break; + case IPL_TYPE_NSS: + if (diag308_set_works) + reipl_method = REIPL_METHOD_NSS_DIAG; + else + reipl_method = REIPL_METHOD_NSS; + set_reipl_block_actual(reipl_block_nss); + break; + case IPL_TYPE_UNKNOWN: + reipl_method = REIPL_METHOD_DEFAULT; + break; + default: + BUG(); + } + reipl_type = type; + return 0; +} + +static ssize_t reipl_type_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return sprintf(page, "%s\n", ipl_type_str(reipl_type)); +} + +static ssize_t reipl_type_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + int rc = -EINVAL; + + if (strncmp(buf, IPL_CCW_STR, strlen(IPL_CCW_STR)) == 0) + rc = reipl_set_type(IPL_TYPE_CCW); + else if (strncmp(buf, IPL_FCP_STR, strlen(IPL_FCP_STR)) == 0) + rc = reipl_set_type(IPL_TYPE_FCP); + else if (strncmp(buf, IPL_NSS_STR, strlen(IPL_NSS_STR)) == 0) + rc = reipl_set_type(IPL_TYPE_NSS); + return (rc != 0) ? rc : len; +} + +static struct kobj_attribute reipl_type_attr = + __ATTR(reipl_type, 0644, reipl_type_show, reipl_type_store); + +static struct kset *reipl_kset; +static struct kset *reipl_fcp_kset; + +static void get_ipl_string(char *dst, struct ipl_parameter_block *ipb, + const enum ipl_method m) +{ + char loadparm[LOADPARM_LEN + 1] = {}; + char vmparm[DIAG308_VMPARM_SIZE + 1] = {}; + char nss_name[NSS_NAME_SIZE + 1] = {}; + size_t pos = 0; + + reipl_get_ascii_loadparm(loadparm, ipb); + reipl_get_ascii_nss_name(nss_name, ipb); + reipl_get_ascii_vmparm(vmparm, sizeof(vmparm), ipb); + + switch (m) { + case REIPL_METHOD_CCW_VM: + pos = sprintf(dst, "IPL %X CLEAR", ipb->ipl_info.ccw.devno); + break; + case REIPL_METHOD_NSS: + pos = sprintf(dst, "IPL %s", nss_name); + break; + default: + break; + } + if (strlen(loadparm) > 0) + pos += sprintf(dst + pos, " LOADPARM '%s'", loadparm); + if (strlen(vmparm) > 0) + sprintf(dst + pos, " PARM %s", vmparm); +} + +static void __reipl_run(void *unused) +{ + struct ccw_dev_id devid; + static char buf[128]; + + switch (reipl_method) { + case REIPL_METHOD_CCW_CIO: + devid.devno = reipl_block_ccw->ipl_info.ccw.devno; + devid.ssid = 0; + reipl_ccw_dev(&devid); + break; + case REIPL_METHOD_CCW_VM: + get_ipl_string(buf, reipl_block_ccw, REIPL_METHOD_CCW_VM); + __cpcmd(buf, NULL, 0, NULL); + break; + case REIPL_METHOD_CCW_DIAG: + diag308(DIAG308_SET, reipl_block_ccw); + diag308(DIAG308_IPL, NULL); + break; + case REIPL_METHOD_FCP_RW_DIAG: + diag308(DIAG308_SET, reipl_block_fcp); + diag308(DIAG308_IPL, NULL); + break; + case REIPL_METHOD_FCP_RO_DIAG: + diag308(DIAG308_IPL, NULL); + break; + case REIPL_METHOD_FCP_RO_VM: + __cpcmd("IPL", NULL, 0, NULL); + break; + case REIPL_METHOD_NSS_DIAG: + diag308(DIAG308_SET, reipl_block_nss); + diag308(DIAG308_IPL, NULL); + break; + case REIPL_METHOD_NSS: + get_ipl_string(buf, reipl_block_nss, REIPL_METHOD_NSS); + __cpcmd(buf, NULL, 0, NULL); + break; + case REIPL_METHOD_DEFAULT: + if (MACHINE_IS_VM) + __cpcmd("IPL", NULL, 0, NULL); + diag308(DIAG308_IPL, NULL); + break; + case REIPL_METHOD_FCP_DUMP: + break; + } + disabled_wait((unsigned long) __builtin_return_address(0)); +} + +static void reipl_run(struct shutdown_trigger *trigger) +{ + smp_call_ipl_cpu(__reipl_run, NULL); +} + +static void reipl_block_ccw_init(struct ipl_parameter_block *ipb) +{ + ipb->hdr.len = IPL_PARM_BLK_CCW_LEN; + ipb->hdr.version = IPL_PARM_BLOCK_VERSION; + ipb->hdr.blk0_len = IPL_PARM_BLK0_CCW_LEN; + ipb->hdr.pbt = DIAG308_IPL_TYPE_CCW; +} + +static void reipl_block_ccw_fill_parms(struct ipl_parameter_block *ipb) +{ + /* LOADPARM */ + /* check if read scp info worked and set loadparm */ + if (sclp_ipl_info.is_valid) + memcpy(ipb->hdr.loadparm, &sclp_ipl_info.loadparm, LOADPARM_LEN); + else + /* read scp info failed: set empty loadparm (EBCDIC blanks) */ + memset(ipb->hdr.loadparm, 0x40, LOADPARM_LEN); + ipb->hdr.flags = DIAG308_FLAGS_LP_VALID; + + /* VM PARM */ + if (MACHINE_IS_VM && diag308_set_works && + (ipl_block.ipl_info.ccw.vm_flags & DIAG308_VM_FLAGS_VP_VALID)) { + + ipb->ipl_info.ccw.vm_flags |= DIAG308_VM_FLAGS_VP_VALID; + ipb->ipl_info.ccw.vm_parm_len = + ipl_block.ipl_info.ccw.vm_parm_len; + memcpy(ipb->ipl_info.ccw.vm_parm, + ipl_block.ipl_info.ccw.vm_parm, DIAG308_VMPARM_SIZE); + } +} + +static int __init reipl_nss_init(void) +{ + int rc; + + if (!MACHINE_IS_VM) + return 0; + + reipl_block_nss = (void *) get_zeroed_page(GFP_KERNEL); + if (!reipl_block_nss) + return -ENOMEM; + + if (!diag308_set_works) + sys_reipl_nss_vmparm_attr.attr.mode = S_IRUGO; + + rc = sysfs_create_group(&reipl_kset->kobj, &reipl_nss_attr_group); + if (rc) + return rc; + + reipl_block_ccw_init(reipl_block_nss); + if (ipl_info.type == IPL_TYPE_NSS) { + memset(reipl_block_nss->ipl_info.ccw.nss_name, + ' ', NSS_NAME_SIZE); + memcpy(reipl_block_nss->ipl_info.ccw.nss_name, + kernel_nss_name, strlen(kernel_nss_name)); + ASCEBC(reipl_block_nss->ipl_info.ccw.nss_name, NSS_NAME_SIZE); + reipl_block_nss->ipl_info.ccw.vm_flags |= + DIAG308_VM_FLAGS_NSS_VALID; + + reipl_block_ccw_fill_parms(reipl_block_nss); + } + + reipl_capabilities |= IPL_TYPE_NSS; + return 0; +} + +static int __init reipl_ccw_init(void) +{ + int rc; + + reipl_block_ccw = (void *) get_zeroed_page(GFP_KERNEL); + if (!reipl_block_ccw) + return -ENOMEM; + + if (MACHINE_IS_VM) { + if (!diag308_set_works) + sys_reipl_ccw_vmparm_attr.attr.mode = S_IRUGO; + rc = sysfs_create_group(&reipl_kset->kobj, + &reipl_ccw_attr_group_vm); + } else { + if(!diag308_set_works) + sys_reipl_ccw_loadparm_attr.attr.mode = S_IRUGO; + rc = sysfs_create_group(&reipl_kset->kobj, + &reipl_ccw_attr_group_lpar); + } + if (rc) + return rc; + + reipl_block_ccw_init(reipl_block_ccw); + if (ipl_info.type == IPL_TYPE_CCW) { + reipl_block_ccw->ipl_info.ccw.devno = ipl_devno; + reipl_block_ccw_fill_parms(reipl_block_ccw); + } + + reipl_capabilities |= IPL_TYPE_CCW; + return 0; +} + +static int __init reipl_fcp_init(void) +{ + int rc; + + if (!diag308_set_works) { + if (ipl_info.type == IPL_TYPE_FCP) { + make_attrs_ro(reipl_fcp_attrs); + sys_reipl_fcp_scp_data_attr.attr.mode = S_IRUGO; + } else + return 0; + } + + reipl_block_fcp = (void *) get_zeroed_page(GFP_KERNEL); + if (!reipl_block_fcp) + return -ENOMEM; + + /* sysfs: create fcp kset for mixing attr group and bin attrs */ + reipl_fcp_kset = kset_create_and_add(IPL_FCP_STR, NULL, + &reipl_kset->kobj); + if (!reipl_fcp_kset) { + free_page((unsigned long) reipl_block_fcp); + return -ENOMEM; + } + + rc = sysfs_create_group(&reipl_fcp_kset->kobj, &reipl_fcp_attr_group); + if (rc) { + kset_unregister(reipl_fcp_kset); + free_page((unsigned long) reipl_block_fcp); + return rc; + } + + if (ipl_info.type == IPL_TYPE_FCP) { + memcpy(reipl_block_fcp, IPL_PARMBLOCK_START, PAGE_SIZE); + /* + * Fix loadparm: There are systems where the (SCSI) LOADPARM + * is invalid in the SCSI IPL parameter block, so take it + * always from sclp_ipl_info. + */ + memcpy(reipl_block_fcp->hdr.loadparm, sclp_ipl_info.loadparm, + LOADPARM_LEN); + } else { + reipl_block_fcp->hdr.len = IPL_PARM_BLK_FCP_LEN; + reipl_block_fcp->hdr.version = IPL_PARM_BLOCK_VERSION; + reipl_block_fcp->hdr.blk0_len = IPL_PARM_BLK0_FCP_LEN; + reipl_block_fcp->hdr.pbt = DIAG308_IPL_TYPE_FCP; + reipl_block_fcp->ipl_info.fcp.opt = DIAG308_IPL_OPT_IPL; + } + reipl_capabilities |= IPL_TYPE_FCP; + return 0; +} + +static int __init reipl_type_init(void) +{ + enum ipl_type reipl_type = ipl_info.type; + struct ipl_parameter_block *reipl_block; + unsigned long size; + + reipl_block = os_info_old_entry(OS_INFO_REIPL_BLOCK, &size); + if (!reipl_block) + goto out; + /* + * If we have an OS info reipl block, this will be used + */ + if (reipl_block->hdr.pbt == DIAG308_IPL_TYPE_FCP) { + memcpy(reipl_block_fcp, reipl_block, size); + reipl_type = IPL_TYPE_FCP; + } else if (reipl_block->hdr.pbt == DIAG308_IPL_TYPE_CCW) { + memcpy(reipl_block_ccw, reipl_block, size); + reipl_type = IPL_TYPE_CCW; + } +out: + return reipl_set_type(reipl_type); +} + +static int __init reipl_init(void) +{ + int rc; + + reipl_kset = kset_create_and_add("reipl", NULL, firmware_kobj); + if (!reipl_kset) + return -ENOMEM; + rc = sysfs_create_file(&reipl_kset->kobj, &reipl_type_attr.attr); + if (rc) { + kset_unregister(reipl_kset); + return rc; + } + rc = reipl_ccw_init(); + if (rc) + return rc; + rc = reipl_fcp_init(); + if (rc) + return rc; + rc = reipl_nss_init(); + if (rc) + return rc; + return reipl_type_init(); +} + +static struct shutdown_action __refdata reipl_action = { + .name = SHUTDOWN_ACTION_REIPL_STR, + .fn = reipl_run, + .init = reipl_init, +}; + +/* + * dump shutdown action: Dump Linux on shutdown. + */ + +/* FCP dump device attributes */ + +DEFINE_IPL_ATTR_RW(dump_fcp, wwpn, "0x%016llx\n", "%llx\n", + dump_block_fcp->ipl_info.fcp.wwpn); +DEFINE_IPL_ATTR_RW(dump_fcp, lun, "0x%016llx\n", "%llx\n", + dump_block_fcp->ipl_info.fcp.lun); +DEFINE_IPL_ATTR_RW(dump_fcp, bootprog, "%lld\n", "%lld\n", + dump_block_fcp->ipl_info.fcp.bootprog); +DEFINE_IPL_ATTR_RW(dump_fcp, br_lba, "%lld\n", "%lld\n", + dump_block_fcp->ipl_info.fcp.br_lba); +DEFINE_IPL_ATTR_RW(dump_fcp, device, "0.0.%04llx\n", "0.0.%llx\n", + dump_block_fcp->ipl_info.fcp.devno); + +static struct attribute *dump_fcp_attrs[] = { + &sys_dump_fcp_device_attr.attr, + &sys_dump_fcp_wwpn_attr.attr, + &sys_dump_fcp_lun_attr.attr, + &sys_dump_fcp_bootprog_attr.attr, + &sys_dump_fcp_br_lba_attr.attr, + NULL, +}; + +static struct attribute_group dump_fcp_attr_group = { + .name = IPL_FCP_STR, + .attrs = dump_fcp_attrs, +}; + +/* CCW dump device attributes */ + +DEFINE_IPL_ATTR_RW(dump_ccw, device, "0.0.%04llx\n", "0.0.%llx\n", + dump_block_ccw->ipl_info.ccw.devno); + +static struct attribute *dump_ccw_attrs[] = { + &sys_dump_ccw_device_attr.attr, + NULL, +}; + +static struct attribute_group dump_ccw_attr_group = { + .name = IPL_CCW_STR, + .attrs = dump_ccw_attrs, +}; + +/* dump type */ + +static int dump_set_type(enum dump_type type) +{ + if (!(dump_capabilities & type)) + return -EINVAL; + switch (type) { + case DUMP_TYPE_CCW: + if (diag308_set_works) + dump_method = DUMP_METHOD_CCW_DIAG; + else if (MACHINE_IS_VM) + dump_method = DUMP_METHOD_CCW_VM; + else + dump_method = DUMP_METHOD_CCW_CIO; + break; + case DUMP_TYPE_FCP: + dump_method = DUMP_METHOD_FCP_DIAG; + break; + default: + dump_method = DUMP_METHOD_NONE; + } + dump_type = type; + return 0; +} + +static ssize_t dump_type_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return sprintf(page, "%s\n", dump_type_str(dump_type)); +} + +static ssize_t dump_type_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + int rc = -EINVAL; + + if (strncmp(buf, DUMP_NONE_STR, strlen(DUMP_NONE_STR)) == 0) + rc = dump_set_type(DUMP_TYPE_NONE); + else if (strncmp(buf, DUMP_CCW_STR, strlen(DUMP_CCW_STR)) == 0) + rc = dump_set_type(DUMP_TYPE_CCW); + else if (strncmp(buf, DUMP_FCP_STR, strlen(DUMP_FCP_STR)) == 0) + rc = dump_set_type(DUMP_TYPE_FCP); + return (rc != 0) ? rc : len; +} + +static struct kobj_attribute dump_type_attr = + __ATTR(dump_type, 0644, dump_type_show, dump_type_store); + +static struct kset *dump_kset; + +static void diag308_dump(void *dump_block) +{ + diag308(DIAG308_SET, dump_block); + while (1) { + if (diag308(DIAG308_DUMP, NULL) != 0x302) + break; + udelay_simple(USEC_PER_SEC); + } +} + +static void __dump_run(void *unused) +{ + struct ccw_dev_id devid; + static char buf[100]; + + switch (dump_method) { + case DUMP_METHOD_CCW_CIO: + devid.devno = dump_block_ccw->ipl_info.ccw.devno; + devid.ssid = 0; + reipl_ccw_dev(&devid); + break; + case DUMP_METHOD_CCW_VM: + sprintf(buf, "STORE STATUS"); + __cpcmd(buf, NULL, 0, NULL); + sprintf(buf, "IPL %X", dump_block_ccw->ipl_info.ccw.devno); + __cpcmd(buf, NULL, 0, NULL); + break; + case DUMP_METHOD_CCW_DIAG: + diag308_dump(dump_block_ccw); + break; + case DUMP_METHOD_FCP_DIAG: + diag308_dump(dump_block_fcp); + break; + default: + break; + } +} + +static void dump_run(struct shutdown_trigger *trigger) +{ + if (dump_method == DUMP_METHOD_NONE) + return; + smp_send_stop(); + smp_call_ipl_cpu(__dump_run, NULL); +} + +static int __init dump_ccw_init(void) +{ + int rc; + + dump_block_ccw = (void *) get_zeroed_page(GFP_KERNEL); + if (!dump_block_ccw) + return -ENOMEM; + rc = sysfs_create_group(&dump_kset->kobj, &dump_ccw_attr_group); + if (rc) { + free_page((unsigned long)dump_block_ccw); + return rc; + } + dump_block_ccw->hdr.len = IPL_PARM_BLK_CCW_LEN; + dump_block_ccw->hdr.version = IPL_PARM_BLOCK_VERSION; + dump_block_ccw->hdr.blk0_len = IPL_PARM_BLK0_CCW_LEN; + dump_block_ccw->hdr.pbt = DIAG308_IPL_TYPE_CCW; + dump_capabilities |= DUMP_TYPE_CCW; + return 0; +} + +static int __init dump_fcp_init(void) +{ + int rc; + + if (!sclp_ipl_info.has_dump) + return 0; /* LDIPL DUMP is not installed */ + if (!diag308_set_works) + return 0; + dump_block_fcp = (void *) get_zeroed_page(GFP_KERNEL); + if (!dump_block_fcp) + return -ENOMEM; + rc = sysfs_create_group(&dump_kset->kobj, &dump_fcp_attr_group); + if (rc) { + free_page((unsigned long)dump_block_fcp); + return rc; + } + dump_block_fcp->hdr.len = IPL_PARM_BLK_FCP_LEN; + dump_block_fcp->hdr.version = IPL_PARM_BLOCK_VERSION; + dump_block_fcp->hdr.blk0_len = IPL_PARM_BLK0_FCP_LEN; + dump_block_fcp->hdr.pbt = DIAG308_IPL_TYPE_FCP; + dump_block_fcp->ipl_info.fcp.opt = DIAG308_IPL_OPT_DUMP; + dump_capabilities |= DUMP_TYPE_FCP; + return 0; +} + +static int __init dump_init(void) +{ + int rc; + + dump_kset = kset_create_and_add("dump", NULL, firmware_kobj); + if (!dump_kset) + return -ENOMEM; + rc = sysfs_create_file(&dump_kset->kobj, &dump_type_attr.attr); + if (rc) { + kset_unregister(dump_kset); + return rc; + } + rc = dump_ccw_init(); + if (rc) + return rc; + rc = dump_fcp_init(); + if (rc) + return rc; + dump_set_type(DUMP_TYPE_NONE); + return 0; +} + +static struct shutdown_action __refdata dump_action = { + .name = SHUTDOWN_ACTION_DUMP_STR, + .fn = dump_run, + .init = dump_init, +}; + +static void dump_reipl_run(struct shutdown_trigger *trigger) +{ + unsigned long ipib = (unsigned long) reipl_block_actual; + unsigned int csum; + + csum = csum_partial(reipl_block_actual, reipl_block_actual->hdr.len, 0); + mem_assign_absolute(S390_lowcore.ipib, ipib); + mem_assign_absolute(S390_lowcore.ipib_checksum, csum); + dump_run(trigger); +} + +static int __init dump_reipl_init(void) +{ + if (!diag308_set_works) + return -EOPNOTSUPP; + else + return 0; +} + +static struct shutdown_action __refdata dump_reipl_action = { + .name = SHUTDOWN_ACTION_DUMP_REIPL_STR, + .fn = dump_reipl_run, + .init = dump_reipl_init, +}; + +/* + * vmcmd shutdown action: Trigger vm command on shutdown. + */ + +static char vmcmd_on_reboot[128]; +static char vmcmd_on_panic[128]; +static char vmcmd_on_halt[128]; +static char vmcmd_on_poff[128]; +static char vmcmd_on_restart[128]; + +DEFINE_IPL_ATTR_STR_RW(vmcmd, on_reboot, "%s\n", "%s\n", vmcmd_on_reboot); +DEFINE_IPL_ATTR_STR_RW(vmcmd, on_panic, "%s\n", "%s\n", vmcmd_on_panic); +DEFINE_IPL_ATTR_STR_RW(vmcmd, on_halt, "%s\n", "%s\n", vmcmd_on_halt); +DEFINE_IPL_ATTR_STR_RW(vmcmd, on_poff, "%s\n", "%s\n", vmcmd_on_poff); +DEFINE_IPL_ATTR_STR_RW(vmcmd, on_restart, "%s\n", "%s\n", vmcmd_on_restart); + +static struct attribute *vmcmd_attrs[] = { + &sys_vmcmd_on_reboot_attr.attr, + &sys_vmcmd_on_panic_attr.attr, + &sys_vmcmd_on_halt_attr.attr, + &sys_vmcmd_on_poff_attr.attr, + &sys_vmcmd_on_restart_attr.attr, + NULL, +}; + +static struct attribute_group vmcmd_attr_group = { + .attrs = vmcmd_attrs, +}; + +static struct kset *vmcmd_kset; + +static void vmcmd_run(struct shutdown_trigger *trigger) +{ + char *cmd; + + if (strcmp(trigger->name, ON_REIPL_STR) == 0) + cmd = vmcmd_on_reboot; + else if (strcmp(trigger->name, ON_PANIC_STR) == 0) + cmd = vmcmd_on_panic; + else if (strcmp(trigger->name, ON_HALT_STR) == 0) + cmd = vmcmd_on_halt; + else if (strcmp(trigger->name, ON_POFF_STR) == 0) + cmd = vmcmd_on_poff; + else if (strcmp(trigger->name, ON_RESTART_STR) == 0) + cmd = vmcmd_on_restart; + else + return; + + if (strlen(cmd) == 0) + return; + __cpcmd(cmd, NULL, 0, NULL); +} + +static int vmcmd_init(void) +{ + if (!MACHINE_IS_VM) + return -EOPNOTSUPP; + vmcmd_kset = kset_create_and_add("vmcmd", NULL, firmware_kobj); + if (!vmcmd_kset) + return -ENOMEM; + return sysfs_create_group(&vmcmd_kset->kobj, &vmcmd_attr_group); +} + +static struct shutdown_action vmcmd_action = {SHUTDOWN_ACTION_VMCMD_STR, + vmcmd_run, vmcmd_init}; + +/* + * stop shutdown action: Stop Linux on shutdown. + */ + +static void stop_run(struct shutdown_trigger *trigger) +{ + if (strcmp(trigger->name, ON_PANIC_STR) == 0 || + strcmp(trigger->name, ON_RESTART_STR) == 0) + disabled_wait((unsigned long) __builtin_return_address(0)); + smp_stop_cpu(); +} + +static struct shutdown_action stop_action = {SHUTDOWN_ACTION_STOP_STR, + stop_run, NULL}; + +/* action list */ + +static struct shutdown_action *shutdown_actions_list[] = { + &ipl_action, &reipl_action, &dump_reipl_action, &dump_action, + &vmcmd_action, &stop_action}; +#define SHUTDOWN_ACTIONS_COUNT (sizeof(shutdown_actions_list) / sizeof(void *)) + +/* + * Trigger section + */ + +static struct kset *shutdown_actions_kset; + +static int set_trigger(const char *buf, struct shutdown_trigger *trigger, + size_t len) +{ + int i; + + for (i = 0; i < SHUTDOWN_ACTIONS_COUNT; i++) { + if (sysfs_streq(buf, shutdown_actions_list[i]->name)) { + if (shutdown_actions_list[i]->init_rc) { + return shutdown_actions_list[i]->init_rc; + } else { + trigger->action = shutdown_actions_list[i]; + return len; + } + } + } + return -EINVAL; +} + +/* on reipl */ + +static struct shutdown_trigger on_reboot_trigger = {ON_REIPL_STR, + &reipl_action}; + +static ssize_t on_reboot_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return sprintf(page, "%s\n", on_reboot_trigger.action->name); +} + +static ssize_t on_reboot_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return set_trigger(buf, &on_reboot_trigger, len); +} +static struct kobj_attribute on_reboot_attr = __ATTR_RW(on_reboot); + +static void do_machine_restart(char *__unused) +{ + smp_send_stop(); + on_reboot_trigger.action->fn(&on_reboot_trigger); + reipl_run(NULL); +} +void (*_machine_restart)(char *command) = do_machine_restart; + +/* on panic */ + +static struct shutdown_trigger on_panic_trigger = {ON_PANIC_STR, &stop_action}; + +static ssize_t on_panic_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return sprintf(page, "%s\n", on_panic_trigger.action->name); +} + +static ssize_t on_panic_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return set_trigger(buf, &on_panic_trigger, len); +} +static struct kobj_attribute on_panic_attr = __ATTR_RW(on_panic); + +static void do_panic(void) +{ + lgr_info_log(); + on_panic_trigger.action->fn(&on_panic_trigger); + stop_run(&on_panic_trigger); +} + +/* on restart */ + +static struct shutdown_trigger on_restart_trigger = {ON_RESTART_STR, + &stop_action}; + +static ssize_t on_restart_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return sprintf(page, "%s\n", on_restart_trigger.action->name); +} + +static ssize_t on_restart_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return set_trigger(buf, &on_restart_trigger, len); +} +static struct kobj_attribute on_restart_attr = __ATTR_RW(on_restart); + +static void __do_restart(void *ignore) +{ + __arch_local_irq_stosm(0x04); /* enable DAT */ + smp_send_stop(); +#ifdef CONFIG_CRASH_DUMP + crash_kexec(NULL); +#endif + on_restart_trigger.action->fn(&on_restart_trigger); + stop_run(&on_restart_trigger); +} + +void do_restart(void) +{ + tracing_off(); + debug_locks_off(); + lgr_info_log(); + smp_call_online_cpu(__do_restart, NULL); +} + +/* on halt */ + +static struct shutdown_trigger on_halt_trigger = {ON_HALT_STR, &stop_action}; + +static ssize_t on_halt_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return sprintf(page, "%s\n", on_halt_trigger.action->name); +} + +static ssize_t on_halt_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return set_trigger(buf, &on_halt_trigger, len); +} +static struct kobj_attribute on_halt_attr = __ATTR_RW(on_halt); + +static void do_machine_halt(void) +{ + smp_send_stop(); + on_halt_trigger.action->fn(&on_halt_trigger); + stop_run(&on_halt_trigger); +} +void (*_machine_halt)(void) = do_machine_halt; + +/* on power off */ + +static struct shutdown_trigger on_poff_trigger = {ON_POFF_STR, &stop_action}; + +static ssize_t on_poff_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return sprintf(page, "%s\n", on_poff_trigger.action->name); +} + +static ssize_t on_poff_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return set_trigger(buf, &on_poff_trigger, len); +} +static struct kobj_attribute on_poff_attr = __ATTR_RW(on_poff); + +static void do_machine_power_off(void) +{ + smp_send_stop(); + on_poff_trigger.action->fn(&on_poff_trigger); + stop_run(&on_poff_trigger); +} +void (*_machine_power_off)(void) = do_machine_power_off; + +static struct attribute *shutdown_action_attrs[] = { + &on_restart_attr.attr, + &on_reboot_attr.attr, + &on_panic_attr.attr, + &on_halt_attr.attr, + &on_poff_attr.attr, + NULL, +}; + +static struct attribute_group shutdown_action_attr_group = { + .attrs = shutdown_action_attrs, +}; + +static void __init shutdown_triggers_init(void) +{ + shutdown_actions_kset = kset_create_and_add("shutdown_actions", NULL, + firmware_kobj); + if (!shutdown_actions_kset) + goto fail; + if (sysfs_create_group(&shutdown_actions_kset->kobj, + &shutdown_action_attr_group)) + goto fail; + return; +fail: + panic("shutdown_triggers_init failed\n"); +} + +static void __init shutdown_actions_init(void) +{ + int i; + + for (i = 0; i < SHUTDOWN_ACTIONS_COUNT; i++) { + if (!shutdown_actions_list[i]->init) + continue; + shutdown_actions_list[i]->init_rc = + shutdown_actions_list[i]->init(); + } +} + +static int __init s390_ipl_init(void) +{ + char str[8] = {0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40}; + + sclp_get_ipl_info(&sclp_ipl_info); + /* + * Fix loadparm: There are systems where the (SCSI) LOADPARM + * returned by read SCP info is invalid (contains EBCDIC blanks) + * when the system has been booted via diag308. In that case we use + * the value from diag308, if available. + * + * There are also systems where diag308 store does not work in + * case the system is booted from HMC. Fortunately in this case + * READ SCP info provides the correct value. + */ + if (memcmp(sclp_ipl_info.loadparm, str, sizeof(str)) == 0 && + diag308_set_works) + memcpy(sclp_ipl_info.loadparm, ipl_block.hdr.loadparm, + LOADPARM_LEN); + shutdown_actions_init(); + shutdown_triggers_init(); + return 0; +} + +__initcall(s390_ipl_init); + +static void __init strncpy_skip_quote(char *dst, char *src, int n) +{ + int sx, dx; + + dx = 0; + for (sx = 0; src[sx] != 0; sx++) { + if (src[sx] == '"') + continue; + dst[dx++] = src[sx]; + if (dx >= n) + break; + } +} + +static int __init vmcmd_on_reboot_setup(char *str) +{ + if (!MACHINE_IS_VM) + return 1; + strncpy_skip_quote(vmcmd_on_reboot, str, 127); + vmcmd_on_reboot[127] = 0; + on_reboot_trigger.action = &vmcmd_action; + return 1; +} +__setup("vmreboot=", vmcmd_on_reboot_setup); + +static int __init vmcmd_on_panic_setup(char *str) +{ + if (!MACHINE_IS_VM) + return 1; + strncpy_skip_quote(vmcmd_on_panic, str, 127); + vmcmd_on_panic[127] = 0; + on_panic_trigger.action = &vmcmd_action; + return 1; +} +__setup("vmpanic=", vmcmd_on_panic_setup); + +static int __init vmcmd_on_halt_setup(char *str) +{ + if (!MACHINE_IS_VM) + return 1; + strncpy_skip_quote(vmcmd_on_halt, str, 127); + vmcmd_on_halt[127] = 0; + on_halt_trigger.action = &vmcmd_action; + return 1; +} +__setup("vmhalt=", vmcmd_on_halt_setup); + +static int __init vmcmd_on_poff_setup(char *str) +{ + if (!MACHINE_IS_VM) + return 1; + strncpy_skip_quote(vmcmd_on_poff, str, 127); + vmcmd_on_poff[127] = 0; + on_poff_trigger.action = &vmcmd_action; + return 1; +} +__setup("vmpoff=", vmcmd_on_poff_setup); + +static int on_panic_notify(struct notifier_block *self, + unsigned long event, void *data) +{ + do_panic(); + return NOTIFY_OK; +} + +static struct notifier_block on_panic_nb = { + .notifier_call = on_panic_notify, + .priority = INT_MIN, +}; + +void __init setup_ipl(void) +{ + ipl_info.type = get_ipl_type(); + switch (ipl_info.type) { + case IPL_TYPE_CCW: + ipl_info.data.ccw.dev_id.devno = ipl_devno; + ipl_info.data.ccw.dev_id.ssid = 0; + break; + case IPL_TYPE_FCP: + case IPL_TYPE_FCP_DUMP: + ipl_info.data.fcp.dev_id.devno = + IPL_PARMBLOCK_START->ipl_info.fcp.devno; + ipl_info.data.fcp.dev_id.ssid = 0; + ipl_info.data.fcp.wwpn = IPL_PARMBLOCK_START->ipl_info.fcp.wwpn; + ipl_info.data.fcp.lun = IPL_PARMBLOCK_START->ipl_info.fcp.lun; + break; + case IPL_TYPE_NSS: + strncpy(ipl_info.data.nss.name, kernel_nss_name, + sizeof(ipl_info.data.nss.name)); + break; + case IPL_TYPE_UNKNOWN: + /* We have no info to copy */ + break; + } + atomic_notifier_chain_register(&panic_notifier_list, &on_panic_nb); +} + +void __init ipl_update_parameters(void) +{ + int rc; + + rc = diag308(DIAG308_STORE, &ipl_block); + if ((rc == DIAG308_RC_OK) || (rc == DIAG308_RC_NOCONFIG)) + diag308_set_works = 1; +} + +void __init ipl_save_parameters(void) +{ + struct cio_iplinfo iplinfo; + void *src, *dst; + + if (cio_get_iplinfo(&iplinfo)) + return; + + ipl_devno = iplinfo.devno; + ipl_flags |= IPL_DEVNO_VALID; + if (!iplinfo.is_qdio) + return; + ipl_flags |= IPL_PARMBLOCK_VALID; + src = (void *)(unsigned long)S390_lowcore.ipl_parmblock_ptr; + dst = (void *)IPL_PARMBLOCK_ORIGIN; + memmove(dst, src, PAGE_SIZE); + S390_lowcore.ipl_parmblock_ptr = IPL_PARMBLOCK_ORIGIN; +} + +static LIST_HEAD(rcall); +static DEFINE_MUTEX(rcall_mutex); + +void register_reset_call(struct reset_call *reset) +{ + mutex_lock(&rcall_mutex); + list_add(&reset->list, &rcall); + mutex_unlock(&rcall_mutex); +} +EXPORT_SYMBOL_GPL(register_reset_call); + +void unregister_reset_call(struct reset_call *reset) +{ + mutex_lock(&rcall_mutex); + list_del(&reset->list); + mutex_unlock(&rcall_mutex); +} +EXPORT_SYMBOL_GPL(unregister_reset_call); + +static void do_reset_calls(void) +{ + struct reset_call *reset; + + if (diag308_set_works) { + diag308_reset(); + return; + } + list_for_each_entry(reset, &rcall, list) + reset->fn(); +} + +u32 dump_prefix_page; + +void s390_reset_system(void (*fn_pre)(void), + void (*fn_post)(void *), void *data) +{ + struct _lowcore *lc; + + lc = (struct _lowcore *)(unsigned long) store_prefix(); + + /* Stack for interrupt/machine check handler */ + lc->panic_stack = S390_lowcore.panic_stack; + + /* Save prefix page address for dump case */ + dump_prefix_page = (u32)(unsigned long) lc; + + /* Disable prefixing */ + set_prefix(0); + + /* Disable lowcore protection */ + __ctl_clear_bit(0,28); + + /* Set new machine check handler */ + S390_lowcore.mcck_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_DAT; + S390_lowcore.mcck_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long) s390_base_mcck_handler; + + /* Set new program check handler */ + S390_lowcore.program_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_DAT; + S390_lowcore.program_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long) s390_base_pgm_handler; + + /* + * Clear subchannel ID and number to signal new kernel that no CCW or + * SCSI IPL has been done (for kexec and kdump) + */ + S390_lowcore.subchannel_id = 0; + S390_lowcore.subchannel_nr = 0; + + /* Store status at absolute zero */ + store_status(); + + /* Call function before reset */ + if (fn_pre) + fn_pre(); + do_reset_calls(); + /* Call function after reset */ + if (fn_post) + fn_post(data); +} diff --git a/arch/s390/kernel/irq.c b/arch/s390/kernel/irq.c new file mode 100644 index 000000000..e9d9addfa --- /dev/null +++ b/arch/s390/kernel/irq.c @@ -0,0 +1,310 @@ +/* + * Copyright IBM Corp. 2004, 2011 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Holger Smolinski <Holger.Smolinski@de.ibm.com>, + * Thomas Spatzier <tspat@de.ibm.com>, + * + * This file contains interrupt related functions. + */ + +#include <linux/kernel_stat.h> +#include <linux/interrupt.h> +#include <linux/seq_file.h> +#include <linux/proc_fs.h> +#include <linux/profile.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/ftrace.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/cpu.h> +#include <linux/irq.h> +#include <asm/irq_regs.h> +#include <asm/cputime.h> +#include <asm/lowcore.h> +#include <asm/irq.h> +#include <asm/hw_irq.h> +#include "entry.h" + +DEFINE_PER_CPU_SHARED_ALIGNED(struct irq_stat, irq_stat); +EXPORT_PER_CPU_SYMBOL_GPL(irq_stat); + +struct irq_class { + int irq; + char *name; + char *desc; +}; + +/* + * The list of "main" irq classes on s390. This is the list of interrupts + * that appear both in /proc/stat ("intr" line) and /proc/interrupts. + * Historically only external and I/O interrupts have been part of /proc/stat. + * We can't add the split external and I/O sub classes since the first field + * in the "intr" line in /proc/stat is supposed to be the sum of all other + * fields. + * Since the external and I/O interrupt fields are already sums we would end + * up with having a sum which accounts each interrupt twice. + */ +static const struct irq_class irqclass_main_desc[NR_IRQS_BASE] = { + {.irq = EXT_INTERRUPT, .name = "EXT"}, + {.irq = IO_INTERRUPT, .name = "I/O"}, + {.irq = THIN_INTERRUPT, .name = "AIO"}, +}; + +/* + * The list of split external and I/O interrupts that appear only in + * /proc/interrupts. + * In addition this list contains non external / I/O events like NMIs. + */ +static const struct irq_class irqclass_sub_desc[] = { + {.irq = IRQEXT_CLK, .name = "CLK", .desc = "[EXT] Clock Comparator"}, + {.irq = IRQEXT_EXC, .name = "EXC", .desc = "[EXT] External Call"}, + {.irq = IRQEXT_EMS, .name = "EMS", .desc = "[EXT] Emergency Signal"}, + {.irq = IRQEXT_TMR, .name = "TMR", .desc = "[EXT] CPU Timer"}, + {.irq = IRQEXT_TLA, .name = "TAL", .desc = "[EXT] Timing Alert"}, + {.irq = IRQEXT_PFL, .name = "PFL", .desc = "[EXT] Pseudo Page Fault"}, + {.irq = IRQEXT_DSD, .name = "DSD", .desc = "[EXT] DASD Diag"}, + {.irq = IRQEXT_VRT, .name = "VRT", .desc = "[EXT] Virtio"}, + {.irq = IRQEXT_SCP, .name = "SCP", .desc = "[EXT] Service Call"}, + {.irq = IRQEXT_IUC, .name = "IUC", .desc = "[EXT] IUCV"}, + {.irq = IRQEXT_CMS, .name = "CMS", .desc = "[EXT] CPU-Measurement: Sampling"}, + {.irq = IRQEXT_CMC, .name = "CMC", .desc = "[EXT] CPU-Measurement: Counter"}, + {.irq = IRQEXT_CMR, .name = "CMR", .desc = "[EXT] CPU-Measurement: RI"}, + {.irq = IRQEXT_FTP, .name = "FTP", .desc = "[EXT] HMC FTP Service"}, + {.irq = IRQIO_CIO, .name = "CIO", .desc = "[I/O] Common I/O Layer Interrupt"}, + {.irq = IRQIO_QAI, .name = "QAI", .desc = "[I/O] QDIO Adapter Interrupt"}, + {.irq = IRQIO_DAS, .name = "DAS", .desc = "[I/O] DASD"}, + {.irq = IRQIO_C15, .name = "C15", .desc = "[I/O] 3215"}, + {.irq = IRQIO_C70, .name = "C70", .desc = "[I/O] 3270"}, + {.irq = IRQIO_TAP, .name = "TAP", .desc = "[I/O] Tape"}, + {.irq = IRQIO_VMR, .name = "VMR", .desc = "[I/O] Unit Record Devices"}, + {.irq = IRQIO_LCS, .name = "LCS", .desc = "[I/O] LCS"}, + {.irq = IRQIO_CTC, .name = "CTC", .desc = "[I/O] CTC"}, + {.irq = IRQIO_APB, .name = "APB", .desc = "[I/O] AP Bus"}, + {.irq = IRQIO_ADM, .name = "ADM", .desc = "[I/O] EADM Subchannel"}, + {.irq = IRQIO_CSC, .name = "CSC", .desc = "[I/O] CHSC Subchannel"}, + {.irq = IRQIO_PCI, .name = "PCI", .desc = "[I/O] PCI Interrupt" }, + {.irq = IRQIO_MSI, .name = "MSI", .desc = "[I/O] MSI Interrupt" }, + {.irq = IRQIO_VIR, .name = "VIR", .desc = "[I/O] Virtual I/O Devices"}, + {.irq = IRQIO_VAI, .name = "VAI", .desc = "[I/O] Virtual I/O Devices AI"}, + {.irq = NMI_NMI, .name = "NMI", .desc = "[NMI] Machine Check"}, + {.irq = CPU_RST, .name = "RST", .desc = "[CPU] CPU Restart"}, +}; + +void __init init_IRQ(void) +{ + BUILD_BUG_ON(ARRAY_SIZE(irqclass_sub_desc) != NR_ARCH_IRQS); + init_cio_interrupts(); + init_airq_interrupts(); + init_ext_interrupts(); +} + +void do_IRQ(struct pt_regs *regs, int irq) +{ + struct pt_regs *old_regs; + + old_regs = set_irq_regs(regs); + irq_enter(); + if (S390_lowcore.int_clock >= S390_lowcore.clock_comparator) + /* Serve timer interrupts first. */ + clock_comparator_work(); + generic_handle_irq(irq); + irq_exit(); + set_irq_regs(old_regs); +} + +/* + * show_interrupts is needed by /proc/interrupts. + */ +int show_interrupts(struct seq_file *p, void *v) +{ + int index = *(loff_t *) v; + int cpu, irq; + + get_online_cpus(); + if (index == 0) { + seq_puts(p, " "); + for_each_online_cpu(cpu) + seq_printf(p, "CPU%d ", cpu); + seq_putc(p, '\n'); + } + if (index < NR_IRQS) { + if (index >= NR_IRQS_BASE) + goto out; + seq_printf(p, "%s: ", irqclass_main_desc[index].name); + irq = irqclass_main_desc[index].irq; + for_each_online_cpu(cpu) + seq_printf(p, "%10u ", kstat_irqs_cpu(irq, cpu)); + seq_putc(p, '\n'); + goto out; + } + for (index = 0; index < NR_ARCH_IRQS; index++) { + seq_printf(p, "%s: ", irqclass_sub_desc[index].name); + irq = irqclass_sub_desc[index].irq; + for_each_online_cpu(cpu) + seq_printf(p, "%10u ", + per_cpu(irq_stat, cpu).irqs[irq]); + if (irqclass_sub_desc[index].desc) + seq_printf(p, " %s", irqclass_sub_desc[index].desc); + seq_putc(p, '\n'); + } +out: + put_online_cpus(); + return 0; +} + +unsigned int arch_dynirq_lower_bound(unsigned int from) +{ + return from < NR_IRQS_BASE ? NR_IRQS_BASE : from; +} + +/* + * Switch to the asynchronous interrupt stack for softirq execution. + */ +void do_softirq_own_stack(void) +{ + unsigned long old, new; + + /* Get current stack pointer. */ + asm volatile("la %0,0(15)" : "=a" (old)); + /* Check against async. stack address range. */ + new = S390_lowcore.async_stack; + if (((new - old) >> (PAGE_SHIFT + THREAD_ORDER)) != 0) { + /* Need to switch to the async. stack. */ + new -= STACK_FRAME_OVERHEAD; + ((struct stack_frame *) new)->back_chain = old; + asm volatile(" la 15,0(%0)\n" + " basr 14,%2\n" + " la 15,0(%1)\n" + : : "a" (new), "a" (old), + "a" (__do_softirq) + : "0", "1", "2", "3", "4", "5", "14", + "cc", "memory" ); + } else { + /* We are already on the async stack. */ + __do_softirq(); + } +} + +/* + * ext_int_hash[index] is the list head for all external interrupts that hash + * to this index. + */ +static struct hlist_head ext_int_hash[32] ____cacheline_aligned; + +struct ext_int_info { + ext_int_handler_t handler; + struct hlist_node entry; + struct rcu_head rcu; + u16 code; +}; + +/* ext_int_hash_lock protects the handler lists for external interrupts */ +static DEFINE_SPINLOCK(ext_int_hash_lock); + +static inline int ext_hash(u16 code) +{ + BUILD_BUG_ON(!is_power_of_2(ARRAY_SIZE(ext_int_hash))); + + return (code + (code >> 9)) & (ARRAY_SIZE(ext_int_hash) - 1); +} + +int register_external_irq(u16 code, ext_int_handler_t handler) +{ + struct ext_int_info *p; + unsigned long flags; + int index; + + p = kmalloc(sizeof(*p), GFP_ATOMIC); + if (!p) + return -ENOMEM; + p->code = code; + p->handler = handler; + index = ext_hash(code); + + spin_lock_irqsave(&ext_int_hash_lock, flags); + hlist_add_head_rcu(&p->entry, &ext_int_hash[index]); + spin_unlock_irqrestore(&ext_int_hash_lock, flags); + return 0; +} +EXPORT_SYMBOL(register_external_irq); + +int unregister_external_irq(u16 code, ext_int_handler_t handler) +{ + struct ext_int_info *p; + unsigned long flags; + int index = ext_hash(code); + + spin_lock_irqsave(&ext_int_hash_lock, flags); + hlist_for_each_entry_rcu(p, &ext_int_hash[index], entry) { + if (p->code == code && p->handler == handler) { + hlist_del_rcu(&p->entry); + kfree_rcu(p, rcu); + } + } + spin_unlock_irqrestore(&ext_int_hash_lock, flags); + return 0; +} +EXPORT_SYMBOL(unregister_external_irq); + +static irqreturn_t do_ext_interrupt(int irq, void *dummy) +{ + struct pt_regs *regs = get_irq_regs(); + struct ext_code ext_code; + struct ext_int_info *p; + int index; + + ext_code = *(struct ext_code *) ®s->int_code; + if (ext_code.code != EXT_IRQ_CLK_COMP) + set_cpu_flag(CIF_NOHZ_DELAY); + + index = ext_hash(ext_code.code); + rcu_read_lock(); + hlist_for_each_entry_rcu(p, &ext_int_hash[index], entry) { + if (unlikely(p->code != ext_code.code)) + continue; + p->handler(ext_code, regs->int_parm, regs->int_parm_long); + } + rcu_read_unlock(); + return IRQ_HANDLED; +} + +static struct irqaction external_interrupt = { + .name = "EXT", + .handler = do_ext_interrupt, +}; + +void __init init_ext_interrupts(void) +{ + int idx; + + for (idx = 0; idx < ARRAY_SIZE(ext_int_hash); idx++) + INIT_HLIST_HEAD(&ext_int_hash[idx]); + + irq_set_chip_and_handler(EXT_INTERRUPT, + &dummy_irq_chip, handle_percpu_irq); + setup_irq(EXT_INTERRUPT, &external_interrupt); +} + +static DEFINE_SPINLOCK(irq_subclass_lock); +static unsigned char irq_subclass_refcount[64]; + +void irq_subclass_register(enum irq_subclass subclass) +{ + spin_lock(&irq_subclass_lock); + if (!irq_subclass_refcount[subclass]) + ctl_set_bit(0, subclass); + irq_subclass_refcount[subclass]++; + spin_unlock(&irq_subclass_lock); +} +EXPORT_SYMBOL(irq_subclass_register); + +void irq_subclass_unregister(enum irq_subclass subclass) +{ + spin_lock(&irq_subclass_lock); + irq_subclass_refcount[subclass]--; + if (!irq_subclass_refcount[subclass]) + ctl_clear_bit(0, subclass); + spin_unlock(&irq_subclass_lock); +} +EXPORT_SYMBOL(irq_subclass_unregister); diff --git a/arch/s390/kernel/jump_label.c b/arch/s390/kernel/jump_label.c new file mode 100644 index 000000000..a90299600 --- /dev/null +++ b/arch/s390/kernel/jump_label.c @@ -0,0 +1,109 @@ +/* + * Jump label s390 support + * + * Copyright IBM Corp. 2011 + * Author(s): Jan Glauber <jang@linux.vnet.ibm.com> + */ +#include <linux/module.h> +#include <linux/uaccess.h> +#include <linux/stop_machine.h> +#include <linux/jump_label.h> +#include <asm/ipl.h> + +#ifdef HAVE_JUMP_LABEL + +struct insn { + u16 opcode; + s32 offset; +} __packed; + +struct insn_args { + struct jump_entry *entry; + enum jump_label_type type; +}; + +static void jump_label_make_nop(struct jump_entry *entry, struct insn *insn) +{ + /* brcl 0,0 */ + insn->opcode = 0xc004; + insn->offset = 0; +} + +static void jump_label_make_branch(struct jump_entry *entry, struct insn *insn) +{ + /* brcl 15,offset */ + insn->opcode = 0xc0f4; + insn->offset = (entry->target - entry->code) >> 1; +} + +static void jump_label_bug(struct jump_entry *entry, struct insn *expected, + struct insn *new) +{ + unsigned char *ipc = (unsigned char *)entry->code; + unsigned char *ipe = (unsigned char *)expected; + unsigned char *ipn = (unsigned char *)new; + + pr_emerg("Jump label code mismatch at %pS [%p]\n", ipc, ipc); + pr_emerg("Found: %02x %02x %02x %02x %02x %02x\n", + ipc[0], ipc[1], ipc[2], ipc[3], ipc[4], ipc[5]); + pr_emerg("Expected: %02x %02x %02x %02x %02x %02x\n", + ipe[0], ipe[1], ipe[2], ipe[3], ipe[4], ipe[5]); + pr_emerg("New: %02x %02x %02x %02x %02x %02x\n", + ipn[0], ipn[1], ipn[2], ipn[3], ipn[4], ipn[5]); + panic("Corrupted kernel text"); +} + +static struct insn orignop = { + .opcode = 0xc004, + .offset = JUMP_LABEL_NOP_OFFSET >> 1, +}; + +static void __jump_label_transform(struct jump_entry *entry, + enum jump_label_type type, + int init) +{ + struct insn old, new; + + if (type == JUMP_LABEL_ENABLE) { + jump_label_make_nop(entry, &old); + jump_label_make_branch(entry, &new); + } else { + jump_label_make_branch(entry, &old); + jump_label_make_nop(entry, &new); + } + if (init) { + if (memcmp((void *)entry->code, &orignop, sizeof(orignop))) + jump_label_bug(entry, &orignop, &new); + } else { + if (memcmp((void *)entry->code, &old, sizeof(old))) + jump_label_bug(entry, &old, &new); + } + s390_kernel_write((void *)entry->code, &new, sizeof(new)); +} + +static int __sm_arch_jump_label_transform(void *data) +{ + struct insn_args *args = data; + + __jump_label_transform(args->entry, args->type, 0); + return 0; +} + +void arch_jump_label_transform(struct jump_entry *entry, + enum jump_label_type type) +{ + struct insn_args args; + + args.entry = entry; + args.type = type; + + stop_machine(__sm_arch_jump_label_transform, &args, NULL); +} + +void arch_jump_label_transform_static(struct jump_entry *entry, + enum jump_label_type type) +{ + __jump_label_transform(entry, type, 1); +} + +#endif diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c new file mode 100644 index 000000000..389db56a2 --- /dev/null +++ b/arch/s390/kernel/kprobes.c @@ -0,0 +1,733 @@ +/* + * Kernel Probes (KProbes) + * + * 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. + * + * Copyright IBM Corp. 2002, 2006 + * + * s390 port, used ppc64 as template. Mike Grundy <grundym@us.ibm.com> + */ + +#include <linux/kprobes.h> +#include <linux/ptrace.h> +#include <linux/preempt.h> +#include <linux/stop_machine.h> +#include <linux/kdebug.h> +#include <linux/uaccess.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/hardirq.h> +#include <linux/ftrace.h> +#include <asm/cacheflush.h> +#include <asm/sections.h> +#include <asm/dis.h> + +DEFINE_PER_CPU(struct kprobe *, current_kprobe); +DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk); + +struct kretprobe_blackpoint kretprobe_blacklist[] = { }; + +DEFINE_INSN_CACHE_OPS(dmainsn); + +static void *alloc_dmainsn_page(void) +{ + return (void *)__get_free_page(GFP_KERNEL | GFP_DMA); +} + +static void free_dmainsn_page(void *page) +{ + free_page((unsigned long)page); +} + +struct kprobe_insn_cache kprobe_dmainsn_slots = { + .mutex = __MUTEX_INITIALIZER(kprobe_dmainsn_slots.mutex), + .alloc = alloc_dmainsn_page, + .free = free_dmainsn_page, + .pages = LIST_HEAD_INIT(kprobe_dmainsn_slots.pages), + .insn_size = MAX_INSN_SIZE, +}; + +static void copy_instruction(struct kprobe *p) +{ + unsigned long ip = (unsigned long) p->addr; + s64 disp, new_disp; + u64 addr, new_addr; + + if (ftrace_location(ip) == ip) { + /* + * If kprobes patches the instruction that is morphed by + * ftrace make sure that kprobes always sees the branch + * "jg .+24" that skips the mcount block or the "brcl 0,0" + * in case of hotpatch. + */ + ftrace_generate_nop_insn((struct ftrace_insn *)p->ainsn.insn); + p->ainsn.is_ftrace_insn = 1; + } else + memcpy(p->ainsn.insn, p->addr, insn_length(*p->addr >> 8)); + p->opcode = p->ainsn.insn[0]; + if (!probe_is_insn_relative_long(p->ainsn.insn)) + return; + /* + * For pc-relative instructions in RIL-b or RIL-c format patch the + * RI2 displacement field. We have already made sure that the insn + * slot for the patched instruction is within the same 2GB area + * as the original instruction (either kernel image or module area). + * Therefore the new displacement will always fit. + */ + disp = *(s32 *)&p->ainsn.insn[1]; + addr = (u64)(unsigned long)p->addr; + new_addr = (u64)(unsigned long)p->ainsn.insn; + new_disp = ((addr + (disp * 2)) - new_addr) / 2; + *(s32 *)&p->ainsn.insn[1] = new_disp; +} +NOKPROBE_SYMBOL(copy_instruction); + +static inline int is_kernel_addr(void *addr) +{ + return addr < (void *)_end; +} + +static int s390_get_insn_slot(struct kprobe *p) +{ + /* + * Get an insn slot that is within the same 2GB area like the original + * instruction. That way instructions with a 32bit signed displacement + * field can be patched and executed within the insn slot. + */ + p->ainsn.insn = NULL; + if (is_kernel_addr(p->addr)) + p->ainsn.insn = get_dmainsn_slot(); + else if (is_module_addr(p->addr)) + p->ainsn.insn = get_insn_slot(); + return p->ainsn.insn ? 0 : -ENOMEM; +} +NOKPROBE_SYMBOL(s390_get_insn_slot); + +static void s390_free_insn_slot(struct kprobe *p) +{ + if (!p->ainsn.insn) + return; + if (is_kernel_addr(p->addr)) + free_dmainsn_slot(p->ainsn.insn, 0); + else + free_insn_slot(p->ainsn.insn, 0); + p->ainsn.insn = NULL; +} +NOKPROBE_SYMBOL(s390_free_insn_slot); + +int arch_prepare_kprobe(struct kprobe *p) +{ + if ((unsigned long) p->addr & 0x01) + return -EINVAL; + /* Make sure the probe isn't going on a difficult instruction */ + if (probe_is_prohibited_opcode(p->addr)) + return -EINVAL; + if (s390_get_insn_slot(p)) + return -ENOMEM; + copy_instruction(p); + return 0; +} +NOKPROBE_SYMBOL(arch_prepare_kprobe); + +int arch_check_ftrace_location(struct kprobe *p) +{ + return 0; +} + +struct swap_insn_args { + struct kprobe *p; + unsigned int arm_kprobe : 1; +}; + +static int swap_instruction(void *data) +{ + struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + unsigned long status = kcb->kprobe_status; + struct swap_insn_args *args = data; + struct ftrace_insn new_insn, *insn; + struct kprobe *p = args->p; + size_t len; + + new_insn.opc = args->arm_kprobe ? BREAKPOINT_INSTRUCTION : p->opcode; + len = sizeof(new_insn.opc); + if (!p->ainsn.is_ftrace_insn) + goto skip_ftrace; + len = sizeof(new_insn); + insn = (struct ftrace_insn *) p->addr; + if (args->arm_kprobe) { + if (is_ftrace_nop(insn)) + new_insn.disp = KPROBE_ON_FTRACE_NOP; + else + new_insn.disp = KPROBE_ON_FTRACE_CALL; + } else { + ftrace_generate_call_insn(&new_insn, (unsigned long)p->addr); + if (insn->disp == KPROBE_ON_FTRACE_NOP) + ftrace_generate_nop_insn(&new_insn); + } +skip_ftrace: + kcb->kprobe_status = KPROBE_SWAP_INST; + s390_kernel_write(p->addr, &new_insn, len); + kcb->kprobe_status = status; + return 0; +} +NOKPROBE_SYMBOL(swap_instruction); + +void arch_arm_kprobe(struct kprobe *p) +{ + struct swap_insn_args args = {.p = p, .arm_kprobe = 1}; + + stop_machine(swap_instruction, &args, NULL); +} +NOKPROBE_SYMBOL(arch_arm_kprobe); + +void arch_disarm_kprobe(struct kprobe *p) +{ + struct swap_insn_args args = {.p = p, .arm_kprobe = 0}; + + stop_machine(swap_instruction, &args, NULL); +} +NOKPROBE_SYMBOL(arch_disarm_kprobe); + +void arch_remove_kprobe(struct kprobe *p) +{ + s390_free_insn_slot(p); +} +NOKPROBE_SYMBOL(arch_remove_kprobe); + +static void enable_singlestep(struct kprobe_ctlblk *kcb, + struct pt_regs *regs, + unsigned long ip) +{ + struct per_regs per_kprobe; + + /* Set up the PER control registers %cr9-%cr11 */ + per_kprobe.control = PER_EVENT_IFETCH; + per_kprobe.start = ip; + per_kprobe.end = ip; + + /* Save control regs and psw mask */ + __ctl_store(kcb->kprobe_saved_ctl, 9, 11); + kcb->kprobe_saved_imask = regs->psw.mask & + (PSW_MASK_PER | PSW_MASK_IO | PSW_MASK_EXT); + + /* Set PER control regs, turns on single step for the given address */ + __ctl_load(per_kprobe, 9, 11); + regs->psw.mask |= PSW_MASK_PER; + regs->psw.mask &= ~(PSW_MASK_IO | PSW_MASK_EXT); + regs->psw.addr = ip | PSW_ADDR_AMODE; +} +NOKPROBE_SYMBOL(enable_singlestep); + +static void disable_singlestep(struct kprobe_ctlblk *kcb, + struct pt_regs *regs, + unsigned long ip) +{ + /* Restore control regs and psw mask, set new psw address */ + __ctl_load(kcb->kprobe_saved_ctl, 9, 11); + regs->psw.mask &= ~PSW_MASK_PER; + regs->psw.mask |= kcb->kprobe_saved_imask; + regs->psw.addr = ip | PSW_ADDR_AMODE; +} +NOKPROBE_SYMBOL(disable_singlestep); + +/* + * Activate a kprobe by storing its pointer to current_kprobe. The + * previous kprobe is stored in kcb->prev_kprobe. A stack of up to + * two kprobes can be active, see KPROBE_REENTER. + */ +static void push_kprobe(struct kprobe_ctlblk *kcb, struct kprobe *p) +{ + kcb->prev_kprobe.kp = __this_cpu_read(current_kprobe); + kcb->prev_kprobe.status = kcb->kprobe_status; + __this_cpu_write(current_kprobe, p); +} +NOKPROBE_SYMBOL(push_kprobe); + +/* + * Deactivate a kprobe by backing up to the previous state. If the + * current state is KPROBE_REENTER prev_kprobe.kp will be non-NULL, + * for any other state prev_kprobe.kp will be NULL. + */ +static void pop_kprobe(struct kprobe_ctlblk *kcb) +{ + __this_cpu_write(current_kprobe, kcb->prev_kprobe.kp); + kcb->kprobe_status = kcb->prev_kprobe.status; +} +NOKPROBE_SYMBOL(pop_kprobe); + +void arch_prepare_kretprobe(struct kretprobe_instance *ri, struct pt_regs *regs) +{ + ri->ret_addr = (kprobe_opcode_t *) regs->gprs[14]; + + /* Replace the return addr with trampoline addr */ + regs->gprs[14] = (unsigned long) &kretprobe_trampoline; +} +NOKPROBE_SYMBOL(arch_prepare_kretprobe); + +static void kprobe_reenter_check(struct kprobe_ctlblk *kcb, struct kprobe *p) +{ + switch (kcb->kprobe_status) { + case KPROBE_HIT_SSDONE: + case KPROBE_HIT_ACTIVE: + kprobes_inc_nmissed_count(p); + break; + case KPROBE_HIT_SS: + case KPROBE_REENTER: + default: + /* + * A kprobe on the code path to single step an instruction + * is a BUG. The code path resides in the .kprobes.text + * section and is executed with interrupts disabled. + */ + printk(KERN_EMERG "Invalid kprobe detected at %p.\n", p->addr); + dump_kprobe(p); + BUG(); + } +} +NOKPROBE_SYMBOL(kprobe_reenter_check); + +static int kprobe_handler(struct pt_regs *regs) +{ + struct kprobe_ctlblk *kcb; + struct kprobe *p; + + /* + * We want to disable preemption for the entire duration of kprobe + * processing. That includes the calls to the pre/post handlers + * and single stepping the kprobe instruction. + */ + preempt_disable(); + kcb = get_kprobe_ctlblk(); + p = get_kprobe((void *)((regs->psw.addr & PSW_ADDR_INSN) - 2)); + + if (p) { + if (kprobe_running()) { + /* + * We have hit a kprobe while another is still + * active. This can happen in the pre and post + * handler. Single step the instruction of the + * new probe but do not call any handler function + * of this secondary kprobe. + * push_kprobe and pop_kprobe saves and restores + * the currently active kprobe. + */ + kprobe_reenter_check(kcb, p); + push_kprobe(kcb, p); + kcb->kprobe_status = KPROBE_REENTER; + } else { + /* + * If we have no pre-handler or it returned 0, we + * continue with single stepping. If we have a + * pre-handler and it returned non-zero, it prepped + * for calling the break_handler below on re-entry + * for jprobe processing, so get out doing nothing + * more here. + */ + push_kprobe(kcb, p); + kcb->kprobe_status = KPROBE_HIT_ACTIVE; + if (p->pre_handler && p->pre_handler(p, regs)) + return 1; + kcb->kprobe_status = KPROBE_HIT_SS; + } + enable_singlestep(kcb, regs, (unsigned long) p->ainsn.insn); + return 1; + } else if (kprobe_running()) { + p = __this_cpu_read(current_kprobe); + if (p->break_handler && p->break_handler(p, regs)) { + /* + * Continuation after the jprobe completed and + * caused the jprobe_return trap. The jprobe + * break_handler "returns" to the original + * function that still has the kprobe breakpoint + * installed. We continue with single stepping. + */ + kcb->kprobe_status = KPROBE_HIT_SS; + enable_singlestep(kcb, regs, + (unsigned long) p->ainsn.insn); + return 1; + } /* else: + * No kprobe at this address and the current kprobe + * has no break handler (no jprobe!). The kernel just + * exploded, let the standard trap handler pick up the + * pieces. + */ + } /* else: + * No kprobe at this address and no active kprobe. The trap has + * not been caused by a kprobe breakpoint. The race of breakpoint + * vs. kprobe remove does not exist because on s390 as we use + * stop_machine to arm/disarm the breakpoints. + */ + preempt_enable_no_resched(); + return 0; +} +NOKPROBE_SYMBOL(kprobe_handler); + +/* + * Function return probe trampoline: + * - init_kprobes() establishes a probepoint here + * - When the probed function returns, this probe + * causes the handlers to fire + */ +static void __used kretprobe_trampoline_holder(void) +{ + asm volatile(".global kretprobe_trampoline\n" + "kretprobe_trampoline: bcr 0,0\n"); +} + +/* + * Called when the probe at kretprobe trampoline is hit + */ +static int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) +{ + struct kretprobe_instance *ri; + struct hlist_head *head, empty_rp; + struct hlist_node *tmp; + unsigned long flags, orig_ret_address; + unsigned long trampoline_address; + kprobe_opcode_t *correct_ret_addr; + + INIT_HLIST_HEAD(&empty_rp); + kretprobe_hash_lock(current, &head, &flags); + + /* + * It is possible to have multiple instances associated with a given + * task either because an multiple functions in the call path + * have a return probe installed on them, and/or more than one return + * return probe was registered for a target function. + * + * We can handle this because: + * - instances are always inserted at the head of the list + * - when multiple return probes are registered for the same + * function, the first instance's ret_addr will point to the + * real return address, and all the rest will point to + * kretprobe_trampoline + */ + ri = NULL; + orig_ret_address = 0; + correct_ret_addr = NULL; + trampoline_address = (unsigned long) &kretprobe_trampoline; + hlist_for_each_entry_safe(ri, tmp, head, hlist) { + if (ri->task != current) + /* another task is sharing our hash bucket */ + continue; + + orig_ret_address = (unsigned long) ri->ret_addr; + + if (orig_ret_address != trampoline_address) + /* + * This is the real return address. Any other + * instances associated with this task are for + * other calls deeper on the call stack + */ + break; + } + + kretprobe_assert(ri, orig_ret_address, trampoline_address); + + correct_ret_addr = ri->ret_addr; + hlist_for_each_entry_safe(ri, tmp, head, hlist) { + if (ri->task != current) + /* another task is sharing our hash bucket */ + continue; + + orig_ret_address = (unsigned long) ri->ret_addr; + + if (ri->rp && ri->rp->handler) { + ri->ret_addr = correct_ret_addr; + ri->rp->handler(ri, regs); + } + + recycle_rp_inst(ri, &empty_rp); + + if (orig_ret_address != trampoline_address) + /* + * This is the real return address. Any other + * instances associated with this task are for + * other calls deeper on the call stack + */ + break; + } + + regs->psw.addr = orig_ret_address | PSW_ADDR_AMODE; + + pop_kprobe(get_kprobe_ctlblk()); + kretprobe_hash_unlock(current, &flags); + preempt_enable_no_resched(); + + hlist_for_each_entry_safe(ri, tmp, &empty_rp, hlist) { + hlist_del(&ri->hlist); + kfree(ri); + } + /* + * By returning a non-zero value, we are telling + * kprobe_handler() that we don't want the post_handler + * to run (and have re-enabled preemption) + */ + return 1; +} +NOKPROBE_SYMBOL(trampoline_probe_handler); + +/* + * Called after single-stepping. p->addr is the address of the + * instruction whose first byte has been replaced by the "breakpoint" + * instruction. To avoid the SMP problems that can occur when we + * temporarily put back the original opcode to single-step, we + * single-stepped a copy of the instruction. The address of this + * copy is p->ainsn.insn. + */ +static void resume_execution(struct kprobe *p, struct pt_regs *regs) +{ + struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + unsigned long ip = regs->psw.addr & PSW_ADDR_INSN; + int fixup = probe_get_fixup_type(p->ainsn.insn); + + /* Check if the kprobes location is an enabled ftrace caller */ + if (p->ainsn.is_ftrace_insn) { + struct ftrace_insn *insn = (struct ftrace_insn *) p->addr; + struct ftrace_insn call_insn; + + ftrace_generate_call_insn(&call_insn, (unsigned long) p->addr); + /* + * A kprobe on an enabled ftrace call site actually single + * stepped an unconditional branch (ftrace nop equivalent). + * Now we need to fixup things and pretend that a brasl r0,... + * was executed instead. + */ + if (insn->disp == KPROBE_ON_FTRACE_CALL) { + ip += call_insn.disp * 2 - MCOUNT_INSN_SIZE; + regs->gprs[0] = (unsigned long)p->addr + sizeof(*insn); + } + } + + if (fixup & FIXUP_PSW_NORMAL) + ip += (unsigned long) p->addr - (unsigned long) p->ainsn.insn; + + if (fixup & FIXUP_BRANCH_NOT_TAKEN) { + int ilen = insn_length(p->ainsn.insn[0] >> 8); + if (ip - (unsigned long) p->ainsn.insn == ilen) + ip = (unsigned long) p->addr + ilen; + } + + if (fixup & FIXUP_RETURN_REGISTER) { + int reg = (p->ainsn.insn[0] & 0xf0) >> 4; + regs->gprs[reg] += (unsigned long) p->addr - + (unsigned long) p->ainsn.insn; + } + + disable_singlestep(kcb, regs, ip); +} +NOKPROBE_SYMBOL(resume_execution); + +static int post_kprobe_handler(struct pt_regs *regs) +{ + struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + struct kprobe *p = kprobe_running(); + + if (!p) + return 0; + + if (kcb->kprobe_status != KPROBE_REENTER && p->post_handler) { + kcb->kprobe_status = KPROBE_HIT_SSDONE; + p->post_handler(p, regs, 0); + } + + resume_execution(p, regs); + pop_kprobe(kcb); + preempt_enable_no_resched(); + + /* + * if somebody else is singlestepping across a probe point, psw mask + * will have PER set, in which case, continue the remaining processing + * of do_single_step, as if this is not a probe hit. + */ + if (regs->psw.mask & PSW_MASK_PER) + return 0; + + return 1; +} +NOKPROBE_SYMBOL(post_kprobe_handler); + +static int kprobe_trap_handler(struct pt_regs *regs, int trapnr) +{ + struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + struct kprobe *p = kprobe_running(); + const struct exception_table_entry *entry; + + switch(kcb->kprobe_status) { + case KPROBE_SWAP_INST: + /* We are here because the instruction replacement failed */ + return 0; + case KPROBE_HIT_SS: + case KPROBE_REENTER: + /* + * We are here because the instruction being single + * stepped caused a page fault. We reset the current + * kprobe and the nip points back to the probe address + * and allow the page fault handler to continue as a + * normal page fault. + */ + disable_singlestep(kcb, regs, (unsigned long) p->addr); + pop_kprobe(kcb); + preempt_enable_no_resched(); + break; + case KPROBE_HIT_ACTIVE: + case KPROBE_HIT_SSDONE: + /* + * We increment the nmissed count for accounting, + * we can also use npre/npostfault count for accounting + * these specific fault cases. + */ + kprobes_inc_nmissed_count(p); + + /* + * We come here because instructions in the pre/post + * handler caused the page_fault, this could happen + * if handler tries to access user space by + * copy_from_user(), get_user() etc. Let the + * user-specified handler try to fix it first. + */ + if (p->fault_handler && p->fault_handler(p, regs, trapnr)) + return 1; + + /* + * In case the user-specified fault handler returned + * zero, try to fix up. + */ + entry = search_exception_tables(regs->psw.addr & PSW_ADDR_INSN); + if (entry) { + regs->psw.addr = extable_fixup(entry) | PSW_ADDR_AMODE; + return 1; + } + + /* + * fixup_exception() could not handle it, + * Let do_page_fault() fix it. + */ + break; + default: + break; + } + return 0; +} +NOKPROBE_SYMBOL(kprobe_trap_handler); + +int kprobe_fault_handler(struct pt_regs *regs, int trapnr) +{ + int ret; + + if (regs->psw.mask & (PSW_MASK_IO | PSW_MASK_EXT)) + local_irq_disable(); + ret = kprobe_trap_handler(regs, trapnr); + if (regs->psw.mask & (PSW_MASK_IO | PSW_MASK_EXT)) + local_irq_restore(regs->psw.mask & ~PSW_MASK_PER); + return ret; +} +NOKPROBE_SYMBOL(kprobe_fault_handler); + +/* + * Wrapper routine to for handling exceptions. + */ +int kprobe_exceptions_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + struct die_args *args = (struct die_args *) data; + struct pt_regs *regs = args->regs; + int ret = NOTIFY_DONE; + + if (regs->psw.mask & (PSW_MASK_IO | PSW_MASK_EXT)) + local_irq_disable(); + + switch (val) { + case DIE_BPT: + if (kprobe_handler(regs)) + ret = NOTIFY_STOP; + break; + case DIE_SSTEP: + if (post_kprobe_handler(regs)) + ret = NOTIFY_STOP; + break; + case DIE_TRAP: + if (!preemptible() && kprobe_running() && + kprobe_trap_handler(regs, args->trapnr)) + ret = NOTIFY_STOP; + break; + default: + break; + } + + if (regs->psw.mask & (PSW_MASK_IO | PSW_MASK_EXT)) + local_irq_restore(regs->psw.mask & ~PSW_MASK_PER); + + return ret; +} +NOKPROBE_SYMBOL(kprobe_exceptions_notify); + +int setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs) +{ + struct jprobe *jp = container_of(p, struct jprobe, kp); + struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + unsigned long stack; + + memcpy(&kcb->jprobe_saved_regs, regs, sizeof(struct pt_regs)); + + /* setup return addr to the jprobe handler routine */ + regs->psw.addr = (unsigned long) jp->entry | PSW_ADDR_AMODE; + regs->psw.mask &= ~(PSW_MASK_IO | PSW_MASK_EXT); + + /* r15 is the stack pointer */ + stack = (unsigned long) regs->gprs[15]; + + memcpy(kcb->jprobes_stack, (void *) stack, MIN_STACK_SIZE(stack)); + return 1; +} +NOKPROBE_SYMBOL(setjmp_pre_handler); + +void jprobe_return(void) +{ + asm volatile(".word 0x0002"); +} +NOKPROBE_SYMBOL(jprobe_return); + +int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs) +{ + struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + unsigned long stack; + + stack = (unsigned long) kcb->jprobe_saved_regs.gprs[15]; + + /* Put the regs back */ + memcpy(regs, &kcb->jprobe_saved_regs, sizeof(struct pt_regs)); + /* put the stack back */ + memcpy((void *) stack, kcb->jprobes_stack, MIN_STACK_SIZE(stack)); + preempt_enable_no_resched(); + return 1; +} +NOKPROBE_SYMBOL(longjmp_break_handler); + +static struct kprobe trampoline = { + .addr = (kprobe_opcode_t *) &kretprobe_trampoline, + .pre_handler = trampoline_probe_handler +}; + +int __init arch_init_kprobes(void) +{ + return register_kprobe(&trampoline); +} + +int arch_trampoline_kprobe(struct kprobe *p) +{ + return p->addr == (kprobe_opcode_t *) &kretprobe_trampoline; +} +NOKPROBE_SYMBOL(arch_trampoline_kprobe); diff --git a/arch/s390/kernel/lgr.c b/arch/s390/kernel/lgr.c new file mode 100644 index 000000000..6ea6d6933 --- /dev/null +++ b/arch/s390/kernel/lgr.c @@ -0,0 +1,186 @@ +/* + * Linux Guest Relocation (LGR) detection + * + * Copyright IBM Corp. 2012 + * Author(s): Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#include <linux/module.h> +#include <linux/timer.h> +#include <linux/slab.h> +#include <asm/facility.h> +#include <asm/sysinfo.h> +#include <asm/ebcdic.h> +#include <asm/debug.h> +#include <asm/ipl.h> + +#define LGR_TIMER_INTERVAL_SECS (30 * 60) +#define VM_LEVEL_MAX 2 /* Maximum is 8, but we only record two levels */ + +/* + * LGR info: Contains stfle and stsi data + */ +struct lgr_info { + /* Bit field with facility information: 4 DWORDs are stored */ + u64 stfle_fac_list[4]; + /* Level of system (1 = CEC, 2 = LPAR, 3 = z/VM */ + u32 level; + /* Level 1: CEC info (stsi 1.1.1) */ + char manufacturer[16]; + char type[4]; + char sequence[16]; + char plant[4]; + char model[16]; + /* Level 2: LPAR info (stsi 2.2.2) */ + u16 lpar_number; + char name[8]; + /* Level 3: VM info (stsi 3.2.2) */ + u8 vm_count; + struct { + char name[8]; + char cpi[16]; + } vm[VM_LEVEL_MAX]; +} __packed __aligned(8); + +/* + * LGR globals + */ +static char lgr_page[PAGE_SIZE] __aligned(PAGE_SIZE); +static struct lgr_info lgr_info_last; +static struct lgr_info lgr_info_cur; +static struct debug_info *lgr_dbf; + +/* + * Copy buffer and then convert it to ASCII + */ +static void cpascii(char *dst, char *src, int size) +{ + memcpy(dst, src, size); + EBCASC(dst, size); +} + +/* + * Fill LGR info with 1.1.1 stsi data + */ +static void lgr_stsi_1_1_1(struct lgr_info *lgr_info) +{ + struct sysinfo_1_1_1 *si = (void *) lgr_page; + + if (stsi(si, 1, 1, 1)) + return; + cpascii(lgr_info->manufacturer, si->manufacturer, + sizeof(si->manufacturer)); + cpascii(lgr_info->type, si->type, sizeof(si->type)); + cpascii(lgr_info->model, si->model, sizeof(si->model)); + cpascii(lgr_info->sequence, si->sequence, sizeof(si->sequence)); + cpascii(lgr_info->plant, si->plant, sizeof(si->plant)); +} + +/* + * Fill LGR info with 2.2.2 stsi data + */ +static void lgr_stsi_2_2_2(struct lgr_info *lgr_info) +{ + struct sysinfo_2_2_2 *si = (void *) lgr_page; + + if (stsi(si, 2, 2, 2)) + return; + cpascii(lgr_info->name, si->name, sizeof(si->name)); + memcpy(&lgr_info->lpar_number, &si->lpar_number, + sizeof(lgr_info->lpar_number)); +} + +/* + * Fill LGR info with 3.2.2 stsi data + */ +static void lgr_stsi_3_2_2(struct lgr_info *lgr_info) +{ + struct sysinfo_3_2_2 *si = (void *) lgr_page; + int i; + + if (stsi(si, 3, 2, 2)) + return; + for (i = 0; i < min_t(u8, si->count, VM_LEVEL_MAX); i++) { + cpascii(lgr_info->vm[i].name, si->vm[i].name, + sizeof(si->vm[i].name)); + cpascii(lgr_info->vm[i].cpi, si->vm[i].cpi, + sizeof(si->vm[i].cpi)); + } + lgr_info->vm_count = si->count; +} + +/* + * Fill LGR info with current data + */ +static void lgr_info_get(struct lgr_info *lgr_info) +{ + int level; + + memset(lgr_info, 0, sizeof(*lgr_info)); + stfle(lgr_info->stfle_fac_list, ARRAY_SIZE(lgr_info->stfle_fac_list)); + level = stsi(NULL, 0, 0, 0); + lgr_info->level = level; + if (level >= 1) + lgr_stsi_1_1_1(lgr_info); + if (level >= 2) + lgr_stsi_2_2_2(lgr_info); + if (level >= 3) + lgr_stsi_3_2_2(lgr_info); +} + +/* + * Check if LGR info has changed and if yes log new LGR info to s390dbf + */ +void lgr_info_log(void) +{ + static DEFINE_SPINLOCK(lgr_info_lock); + unsigned long flags; + + if (!spin_trylock_irqsave(&lgr_info_lock, flags)) + return; + lgr_info_get(&lgr_info_cur); + if (memcmp(&lgr_info_last, &lgr_info_cur, sizeof(lgr_info_cur)) != 0) { + debug_event(lgr_dbf, 1, &lgr_info_cur, sizeof(lgr_info_cur)); + lgr_info_last = lgr_info_cur; + } + spin_unlock_irqrestore(&lgr_info_lock, flags); +} +EXPORT_SYMBOL_GPL(lgr_info_log); + +static void lgr_timer_set(void); + +/* + * LGR timer callback + */ +static void lgr_timer_fn(unsigned long ignored) +{ + lgr_info_log(); + lgr_timer_set(); +} + +static struct timer_list lgr_timer = + TIMER_DEFERRED_INITIALIZER(lgr_timer_fn, 0, 0); + +/* + * Setup next LGR timer + */ +static void lgr_timer_set(void) +{ + mod_timer(&lgr_timer, jiffies + LGR_TIMER_INTERVAL_SECS * HZ); +} + +/* + * Initialize LGR: Add s390dbf, write initial lgr_info and setup timer + */ +static int __init lgr_init(void) +{ + lgr_dbf = debug_register("lgr", 1, 1, sizeof(struct lgr_info)); + if (!lgr_dbf) + return -ENOMEM; + debug_register_view(lgr_dbf, &debug_hex_ascii_view); + lgr_info_get(&lgr_info_last); + debug_event(lgr_dbf, 1, &lgr_info_last, sizeof(lgr_info_last)); + lgr_timer_set(); + return 0; +} +module_init(lgr_init); diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c new file mode 100644 index 000000000..fb0901ec4 --- /dev/null +++ b/arch/s390/kernel/machine_kexec.c @@ -0,0 +1,275 @@ +/* + * Copyright IBM Corp. 2005, 2011 + * + * Author(s): Rolf Adelsberger, + * Heiko Carstens <heiko.carstens@de.ibm.com> + * Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#include <linux/device.h> +#include <linux/mm.h> +#include <linux/kexec.h> +#include <linux/delay.h> +#include <linux/reboot.h> +#include <linux/ftrace.h> +#include <linux/debug_locks.h> +#include <linux/suspend.h> +#include <asm/cio.h> +#include <asm/setup.h> +#include <asm/pgtable.h> +#include <asm/pgalloc.h> +#include <asm/smp.h> +#include <asm/reset.h> +#include <asm/ipl.h> +#include <asm/diag.h> +#include <asm/elf.h> +#include <asm/asm-offsets.h> +#include <asm/os_info.h> +#include <asm/switch_to.h> + +typedef void (*relocate_kernel_t)(kimage_entry_t *, unsigned long); + +extern const unsigned char relocate_kernel[]; +extern const unsigned long long relocate_kernel_len; + +#ifdef CONFIG_CRASH_DUMP + +/* + * Create ELF notes for one CPU + */ +static void add_elf_notes(int cpu) +{ + struct save_area *sa = (void *) 4608 + store_prefix(); + void *ptr; + + memcpy((void *) (4608UL + sa->pref_reg), sa, sizeof(*sa)); + ptr = (u64 *) per_cpu_ptr(crash_notes, cpu); + ptr = fill_cpu_elf_notes(ptr, sa, NULL); + memset(ptr, 0, sizeof(struct elf_note)); +} + +/* + * Initialize CPU ELF notes + */ +static void setup_regs(void) +{ + unsigned long sa = S390_lowcore.prefixreg_save_area + SAVE_AREA_BASE; + struct _lowcore *lc; + int cpu, this_cpu; + + /* Get lowcore pointer from store status of this CPU (absolute zero) */ + lc = (struct _lowcore *)(unsigned long)S390_lowcore.prefixreg_save_area; + this_cpu = smp_find_processor_id(stap()); + add_elf_notes(this_cpu); + for_each_online_cpu(cpu) { + if (cpu == this_cpu) + continue; + if (smp_store_status(cpu)) + continue; + add_elf_notes(cpu); + } + if (MACHINE_HAS_VX) + save_vx_regs_safe((void *) lc->vector_save_area_addr); + /* Copy dump CPU store status info to absolute zero */ + memcpy((void *) SAVE_AREA_BASE, (void *) sa, sizeof(struct save_area)); +} + +/* + * PM notifier callback for kdump + */ +static int machine_kdump_pm_cb(struct notifier_block *nb, unsigned long action, + void *ptr) +{ + switch (action) { + case PM_SUSPEND_PREPARE: + case PM_HIBERNATION_PREPARE: + if (crashk_res.start) + crash_map_reserved_pages(); + break; + case PM_POST_SUSPEND: + case PM_POST_HIBERNATION: + if (crashk_res.start) + crash_unmap_reserved_pages(); + break; + default: + return NOTIFY_DONE; + } + return NOTIFY_OK; +} + +static int __init machine_kdump_pm_init(void) +{ + pm_notifier(machine_kdump_pm_cb, 0); + return 0; +} +arch_initcall(machine_kdump_pm_init); + +/* + * Start kdump: We expect here that a store status has been done on our CPU + */ +static void __do_machine_kdump(void *image) +{ + int (*start_kdump)(int) = (void *)((struct kimage *) image)->start; + + __load_psw_mask(PSW_MASK_BASE | PSW_DEFAULT_KEY | PSW_MASK_EA | PSW_MASK_BA); + start_kdump(1); +} +#endif + +/* + * Check if kdump checksums are valid: We call purgatory with parameter "0" + */ +static int kdump_csum_valid(struct kimage *image) +{ +#ifdef CONFIG_CRASH_DUMP + int (*start_kdump)(int) = (void *)image->start; + int rc; + + __arch_local_irq_stnsm(0xfb); /* disable DAT */ + rc = start_kdump(0); + __arch_local_irq_stosm(0x04); /* enable DAT */ + return rc ? 0 : -EINVAL; +#else + return -EINVAL; +#endif +} + +/* + * Map or unmap crashkernel memory + */ +static void crash_map_pages(int enable) +{ + unsigned long size = resource_size(&crashk_res); + + BUG_ON(crashk_res.start % KEXEC_CRASH_MEM_ALIGN || + size % KEXEC_CRASH_MEM_ALIGN); + if (enable) + vmem_add_mapping(crashk_res.start, size); + else { + vmem_remove_mapping(crashk_res.start, size); + if (size) + os_info_crashkernel_add(crashk_res.start, size); + else + os_info_crashkernel_add(0, 0); + } +} + +/* + * Map crashkernel memory + */ +void crash_map_reserved_pages(void) +{ + crash_map_pages(1); +} + +/* + * Unmap crashkernel memory + */ +void crash_unmap_reserved_pages(void) +{ + crash_map_pages(0); +} + +/* + * Give back memory to hypervisor before new kdump is loaded + */ +static int machine_kexec_prepare_kdump(void) +{ +#ifdef CONFIG_CRASH_DUMP + if (MACHINE_IS_VM) + diag10_range(PFN_DOWN(crashk_res.start), + PFN_DOWN(crashk_res.end - crashk_res.start + 1)); + return 0; +#else + return -EINVAL; +#endif +} + +int machine_kexec_prepare(struct kimage *image) +{ + void *reboot_code_buffer; + + /* Can't replace kernel image since it is read-only. */ + if (ipl_flags & IPL_NSS_VALID) + return -EOPNOTSUPP; + + if (image->type == KEXEC_TYPE_CRASH) + return machine_kexec_prepare_kdump(); + + /* We don't support anything but the default image type for now. */ + if (image->type != KEXEC_TYPE_DEFAULT) + return -EINVAL; + + /* Get the destination where the assembler code should be copied to.*/ + reboot_code_buffer = (void *) page_to_phys(image->control_code_page); + + /* Then copy it */ + memcpy(reboot_code_buffer, relocate_kernel, relocate_kernel_len); + return 0; +} + +void machine_kexec_cleanup(struct kimage *image) +{ +} + +void arch_crash_save_vmcoreinfo(void) +{ + VMCOREINFO_SYMBOL(lowcore_ptr); + VMCOREINFO_SYMBOL(high_memory); + VMCOREINFO_LENGTH(lowcore_ptr, NR_CPUS); +} + +void machine_shutdown(void) +{ +} + +void machine_crash_shutdown(struct pt_regs *regs) +{ +} + +/* + * Do normal kexec + */ +static void __do_machine_kexec(void *data) +{ + relocate_kernel_t data_mover; + struct kimage *image = data; + + data_mover = (relocate_kernel_t) page_to_phys(image->control_code_page); + + /* Call the moving routine */ + (*data_mover)(&image->head, image->start); +} + +/* + * Reset system and call either kdump or normal kexec + */ +static void __machine_kexec(void *data) +{ + __arch_local_irq_stosm(0x04); /* enable DAT */ + pfault_fini(); + tracing_off(); + debug_locks_off(); +#ifdef CONFIG_CRASH_DUMP + if (((struct kimage *) data)->type == KEXEC_TYPE_CRASH) { + + lgr_info_log(); + s390_reset_system(setup_regs, __do_machine_kdump, data); + } else +#endif + s390_reset_system(NULL, __do_machine_kexec, data); + disabled_wait((unsigned long) __builtin_return_address(0)); +} + +/* + * Do either kdump or normal kexec. In case of kdump we first ask + * purgatory, if kdump checksums are valid. + */ +void machine_kexec(struct kimage *image) +{ + if (image->type == KEXEC_TYPE_CRASH && !kdump_csum_valid(image)) + return; + tracer_disable(); + smp_send_stop(); + smp_call_ipl_cpu(__machine_kexec, image); +} diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S new file mode 100644 index 000000000..e499370fb --- /dev/null +++ b/arch/s390/kernel/mcount.S @@ -0,0 +1,82 @@ +/* + * Copyright IBM Corp. 2008, 2009 + * + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>, + * + */ + +#include <linux/linkage.h> +#include <asm/asm-offsets.h> +#include <asm/ftrace.h> +#include <asm/ptrace.h> + + .section .kprobes.text, "ax" + +ENTRY(ftrace_stub) + br %r14 + +#define STACK_FRAME_SIZE (STACK_FRAME_OVERHEAD + __PT_SIZE) +#define STACK_PTREGS (STACK_FRAME_OVERHEAD) +#define STACK_PTREGS_GPRS (STACK_PTREGS + __PT_GPRS) +#define STACK_PTREGS_PSW (STACK_PTREGS + __PT_PSW) + +ENTRY(_mcount) + br %r14 + +ENTRY(ftrace_caller) + .globl ftrace_regs_caller + .set ftrace_regs_caller,ftrace_caller + lgr %r1,%r15 +#ifndef CC_USING_HOTPATCH + aghi %r0,MCOUNT_RETURN_FIXUP +#endif + aghi %r15,-STACK_FRAME_SIZE + stg %r1,__SF_BACKCHAIN(%r15) + stg %r1,(STACK_PTREGS_GPRS+15*8)(%r15) + stg %r0,(STACK_PTREGS_PSW+8)(%r15) + stmg %r2,%r14,(STACK_PTREGS_GPRS+2*8)(%r15) +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES + aghik %r2,%r0,-MCOUNT_INSN_SIZE + lgrl %r4,function_trace_op + lgrl %r1,ftrace_trace_function +#else + lgr %r2,%r0 + aghi %r2,-MCOUNT_INSN_SIZE + larl %r4,function_trace_op + lg %r4,0(%r4) + larl %r1,ftrace_trace_function + lg %r1,0(%r1) +#endif + lgr %r3,%r14 + la %r5,STACK_PTREGS(%r15) + basr %r14,%r1 +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +# The j instruction gets runtime patched to a nop instruction. +# See ftrace_enable_ftrace_graph_caller. +ENTRY(ftrace_graph_caller) + j ftrace_graph_caller_end + lg %r2,(STACK_PTREGS_GPRS+14*8)(%r15) + lg %r3,(STACK_PTREGS_PSW+8)(%r15) + brasl %r14,prepare_ftrace_return + stg %r2,(STACK_PTREGS_GPRS+14*8)(%r15) +ftrace_graph_caller_end: + .globl ftrace_graph_caller_end +#endif + lg %r1,(STACK_PTREGS_PSW+8)(%r15) + lmg %r2,%r15,(STACK_PTREGS_GPRS+2*8)(%r15) + br %r1 + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER + +ENTRY(return_to_handler) + stmg %r2,%r5,32(%r15) + lgr %r1,%r15 + aghi %r15,-STACK_FRAME_OVERHEAD + stg %r1,__SF_BACKCHAIN(%r15) + brasl %r14,ftrace_return_to_handler + aghi %r15,STACK_FRAME_OVERHEAD + lgr %r14,%r2 + lmg %r2,%r5,32(%r15) + br %r14 + +#endif diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c new file mode 100644 index 000000000..0c1a67931 --- /dev/null +++ b/arch/s390/kernel/module.c @@ -0,0 +1,431 @@ +/* + * Kernel module help for s390. + * + * S390 version + * Copyright IBM Corp. 2002, 2003 + * Author(s): Arnd Bergmann (arndb@de.ibm.com) + * Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * based on i386 version + * Copyright (C) 2001 Rusty Russell. + * + * 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 <linux/module.h> +#include <linux/elf.h> +#include <linux/vmalloc.h> +#include <linux/fs.h> +#include <linux/string.h> +#include <linux/kernel.h> +#include <linux/moduleloader.h> +#include <linux/bug.h> + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(fmt , ...) +#endif + +#define PLT_ENTRY_SIZE 20 + +void *module_alloc(unsigned long size) +{ + if (PAGE_ALIGN(size) > MODULES_LEN) + return NULL; + return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END, + GFP_KERNEL, PAGE_KERNEL, 0, NUMA_NO_NODE, + __builtin_return_address(0)); +} + +void module_arch_freeing_init(struct module *mod) +{ + vfree(mod->arch.syminfo); + mod->arch.syminfo = NULL; +} + +static void check_rela(Elf_Rela *rela, struct module *me) +{ + struct mod_arch_syminfo *info; + + info = me->arch.syminfo + ELF_R_SYM (rela->r_info); + switch (ELF_R_TYPE (rela->r_info)) { + case R_390_GOT12: /* 12 bit GOT offset. */ + case R_390_GOT16: /* 16 bit GOT offset. */ + case R_390_GOT20: /* 20 bit GOT offset. */ + case R_390_GOT32: /* 32 bit GOT offset. */ + case R_390_GOT64: /* 64 bit GOT offset. */ + case R_390_GOTENT: /* 32 bit PC rel. to GOT entry shifted by 1. */ + case R_390_GOTPLT12: /* 12 bit offset to jump slot. */ + case R_390_GOTPLT16: /* 16 bit offset to jump slot. */ + case R_390_GOTPLT20: /* 20 bit offset to jump slot. */ + case R_390_GOTPLT32: /* 32 bit offset to jump slot. */ + case R_390_GOTPLT64: /* 64 bit offset to jump slot. */ + case R_390_GOTPLTENT: /* 32 bit rel. offset to jump slot >> 1. */ + if (info->got_offset == -1UL) { + info->got_offset = me->arch.got_size; + me->arch.got_size += sizeof(void*); + } + break; + case R_390_PLT16DBL: /* 16 bit PC rel. PLT shifted by 1. */ + case R_390_PLT32DBL: /* 32 bit PC rel. PLT shifted by 1. */ + case R_390_PLT32: /* 32 bit PC relative PLT address. */ + case R_390_PLT64: /* 64 bit PC relative PLT address. */ + case R_390_PLTOFF16: /* 16 bit offset from GOT to PLT. */ + case R_390_PLTOFF32: /* 32 bit offset from GOT to PLT. */ + case R_390_PLTOFF64: /* 16 bit offset from GOT to PLT. */ + if (info->plt_offset == -1UL) { + info->plt_offset = me->arch.plt_size; + me->arch.plt_size += PLT_ENTRY_SIZE; + } + break; + case R_390_COPY: + case R_390_GLOB_DAT: + case R_390_JMP_SLOT: + case R_390_RELATIVE: + /* Only needed if we want to support loading of + modules linked with -shared. */ + break; + } +} + +/* + * Account for GOT and PLT relocations. We can't add sections for + * got and plt but we can increase the core module size. + */ +int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs, + char *secstrings, struct module *me) +{ + Elf_Shdr *symtab; + Elf_Sym *symbols; + Elf_Rela *rela; + char *strings; + int nrela, i, j; + + /* Find symbol table and string table. */ + symtab = NULL; + for (i = 0; i < hdr->e_shnum; i++) + switch (sechdrs[i].sh_type) { + case SHT_SYMTAB: + symtab = sechdrs + i; + break; + } + if (!symtab) { + printk(KERN_ERR "module %s: no symbol table\n", me->name); + return -ENOEXEC; + } + + /* Allocate one syminfo structure per symbol. */ + me->arch.nsyms = symtab->sh_size / sizeof(Elf_Sym); + me->arch.syminfo = vmalloc(me->arch.nsyms * + sizeof(struct mod_arch_syminfo)); + if (!me->arch.syminfo) + return -ENOMEM; + symbols = (void *) hdr + symtab->sh_offset; + strings = (void *) hdr + sechdrs[symtab->sh_link].sh_offset; + for (i = 0; i < me->arch.nsyms; i++) { + if (symbols[i].st_shndx == SHN_UNDEF && + strcmp(strings + symbols[i].st_name, + "_GLOBAL_OFFSET_TABLE_") == 0) + /* "Define" it as absolute. */ + symbols[i].st_shndx = SHN_ABS; + me->arch.syminfo[i].got_offset = -1UL; + me->arch.syminfo[i].plt_offset = -1UL; + me->arch.syminfo[i].got_initialized = 0; + me->arch.syminfo[i].plt_initialized = 0; + } + + /* Search for got/plt relocations. */ + me->arch.got_size = me->arch.plt_size = 0; + for (i = 0; i < hdr->e_shnum; i++) { + if (sechdrs[i].sh_type != SHT_RELA) + continue; + nrela = sechdrs[i].sh_size / sizeof(Elf_Rela); + rela = (void *) hdr + sechdrs[i].sh_offset; + for (j = 0; j < nrela; j++) + check_rela(rela + j, me); + } + + /* Increase core size by size of got & plt and set start + offsets for got and plt. */ + me->core_size = ALIGN(me->core_size, 4); + me->arch.got_offset = me->core_size; + me->core_size += me->arch.got_size; + me->arch.plt_offset = me->core_size; + me->core_size += me->arch.plt_size; + return 0; +} + +static int apply_rela_bits(Elf_Addr loc, Elf_Addr val, + int sign, int bits, int shift) +{ + unsigned long umax; + long min, max; + + if (val & ((1UL << shift) - 1)) + return -ENOEXEC; + if (sign) { + val = (Elf_Addr)(((long) val) >> shift); + min = -(1L << (bits - 1)); + max = (1L << (bits - 1)) - 1; + if ((long) val < min || (long) val > max) + return -ENOEXEC; + } else { + val >>= shift; + umax = ((1UL << (bits - 1)) << 1) - 1; + if ((unsigned long) val > umax) + return -ENOEXEC; + } + + if (bits == 8) + *(unsigned char *) loc = val; + else if (bits == 12) + *(unsigned short *) loc = (val & 0xfff) | + (*(unsigned short *) loc & 0xf000); + else if (bits == 16) + *(unsigned short *) loc = val; + else if (bits == 20) + *(unsigned int *) loc = (val & 0xfff) << 16 | + (val & 0xff000) >> 4 | + (*(unsigned int *) loc & 0xf00000ff); + else if (bits == 32) + *(unsigned int *) loc = val; + else if (bits == 64) + *(unsigned long *) loc = val; + return 0; +} + +static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab, + const char *strtab, struct module *me) +{ + struct mod_arch_syminfo *info; + Elf_Addr loc, val; + int r_type, r_sym; + int rc = -ENOEXEC; + + /* This is where to make the change */ + loc = base + rela->r_offset; + /* This is the symbol it is referring to. Note that all + undefined symbols have been resolved. */ + r_sym = ELF_R_SYM(rela->r_info); + r_type = ELF_R_TYPE(rela->r_info); + info = me->arch.syminfo + r_sym; + val = symtab[r_sym].st_value; + + switch (r_type) { + case R_390_NONE: /* No relocation. */ + rc = 0; + break; + case R_390_8: /* Direct 8 bit. */ + case R_390_12: /* Direct 12 bit. */ + case R_390_16: /* Direct 16 bit. */ + case R_390_20: /* Direct 20 bit. */ + case R_390_32: /* Direct 32 bit. */ + case R_390_64: /* Direct 64 bit. */ + val += rela->r_addend; + if (r_type == R_390_8) + rc = apply_rela_bits(loc, val, 0, 8, 0); + else if (r_type == R_390_12) + rc = apply_rela_bits(loc, val, 0, 12, 0); + else if (r_type == R_390_16) + rc = apply_rela_bits(loc, val, 0, 16, 0); + else if (r_type == R_390_20) + rc = apply_rela_bits(loc, val, 1, 20, 0); + else if (r_type == R_390_32) + rc = apply_rela_bits(loc, val, 0, 32, 0); + else if (r_type == R_390_64) + rc = apply_rela_bits(loc, val, 0, 64, 0); + break; + case R_390_PC16: /* PC relative 16 bit. */ + case R_390_PC16DBL: /* PC relative 16 bit shifted by 1. */ + case R_390_PC32DBL: /* PC relative 32 bit shifted by 1. */ + case R_390_PC32: /* PC relative 32 bit. */ + case R_390_PC64: /* PC relative 64 bit. */ + val += rela->r_addend - loc; + if (r_type == R_390_PC16) + rc = apply_rela_bits(loc, val, 1, 16, 0); + else if (r_type == R_390_PC16DBL) + rc = apply_rela_bits(loc, val, 1, 16, 1); + else if (r_type == R_390_PC32DBL) + rc = apply_rela_bits(loc, val, 1, 32, 1); + else if (r_type == R_390_PC32) + rc = apply_rela_bits(loc, val, 1, 32, 0); + else if (r_type == R_390_PC64) + rc = apply_rela_bits(loc, val, 1, 64, 0); + break; + case R_390_GOT12: /* 12 bit GOT offset. */ + case R_390_GOT16: /* 16 bit GOT offset. */ + case R_390_GOT20: /* 20 bit GOT offset. */ + case R_390_GOT32: /* 32 bit GOT offset. */ + case R_390_GOT64: /* 64 bit GOT offset. */ + case R_390_GOTENT: /* 32 bit PC rel. to GOT entry shifted by 1. */ + case R_390_GOTPLT12: /* 12 bit offset to jump slot. */ + case R_390_GOTPLT20: /* 20 bit offset to jump slot. */ + case R_390_GOTPLT16: /* 16 bit offset to jump slot. */ + case R_390_GOTPLT32: /* 32 bit offset to jump slot. */ + case R_390_GOTPLT64: /* 64 bit offset to jump slot. */ + case R_390_GOTPLTENT: /* 32 bit rel. offset to jump slot >> 1. */ + if (info->got_initialized == 0) { + Elf_Addr *gotent; + + gotent = me->module_core + me->arch.got_offset + + info->got_offset; + *gotent = val; + info->got_initialized = 1; + } + val = info->got_offset + rela->r_addend; + if (r_type == R_390_GOT12 || + r_type == R_390_GOTPLT12) + rc = apply_rela_bits(loc, val, 0, 12, 0); + else if (r_type == R_390_GOT16 || + r_type == R_390_GOTPLT16) + rc = apply_rela_bits(loc, val, 0, 16, 0); + else if (r_type == R_390_GOT20 || + r_type == R_390_GOTPLT20) + rc = apply_rela_bits(loc, val, 1, 20, 0); + else if (r_type == R_390_GOT32 || + r_type == R_390_GOTPLT32) + rc = apply_rela_bits(loc, val, 0, 32, 0); + else if (r_type == R_390_GOT64 || + r_type == R_390_GOTPLT64) + rc = apply_rela_bits(loc, val, 0, 64, 0); + else if (r_type == R_390_GOTENT || + r_type == R_390_GOTPLTENT) { + val += (Elf_Addr) me->module_core - loc; + rc = apply_rela_bits(loc, val, 1, 32, 1); + } + break; + case R_390_PLT16DBL: /* 16 bit PC rel. PLT shifted by 1. */ + case R_390_PLT32DBL: /* 32 bit PC rel. PLT shifted by 1. */ + case R_390_PLT32: /* 32 bit PC relative PLT address. */ + case R_390_PLT64: /* 64 bit PC relative PLT address. */ + case R_390_PLTOFF16: /* 16 bit offset from GOT to PLT. */ + case R_390_PLTOFF32: /* 32 bit offset from GOT to PLT. */ + case R_390_PLTOFF64: /* 16 bit offset from GOT to PLT. */ + if (info->plt_initialized == 0) { + unsigned int *ip; + ip = me->module_core + me->arch.plt_offset + + info->plt_offset; + ip[0] = 0x0d10e310; /* basr 1,0; lg 1,10(1); br 1 */ + ip[1] = 0x100a0004; + ip[2] = 0x07f10000; + ip[3] = (unsigned int) (val >> 32); + ip[4] = (unsigned int) val; + info->plt_initialized = 1; + } + if (r_type == R_390_PLTOFF16 || + r_type == R_390_PLTOFF32 || + r_type == R_390_PLTOFF64) + val = me->arch.plt_offset - me->arch.got_offset + + info->plt_offset + rela->r_addend; + else { + if (!((r_type == R_390_PLT16DBL && + val - loc + 0xffffUL < 0x1ffffeUL) || + (r_type == R_390_PLT32DBL && + val - loc + 0xffffffffULL < 0x1fffffffeULL))) + val = (Elf_Addr) me->module_core + + me->arch.plt_offset + + info->plt_offset; + val += rela->r_addend - loc; + } + if (r_type == R_390_PLT16DBL) + rc = apply_rela_bits(loc, val, 1, 16, 1); + else if (r_type == R_390_PLTOFF16) + rc = apply_rela_bits(loc, val, 0, 16, 0); + else if (r_type == R_390_PLT32DBL) + rc = apply_rela_bits(loc, val, 1, 32, 1); + else if (r_type == R_390_PLT32 || + r_type == R_390_PLTOFF32) + rc = apply_rela_bits(loc, val, 0, 32, 0); + else if (r_type == R_390_PLT64 || + r_type == R_390_PLTOFF64) + rc = apply_rela_bits(loc, val, 0, 64, 0); + break; + case R_390_GOTOFF16: /* 16 bit offset to GOT. */ + case R_390_GOTOFF32: /* 32 bit offset to GOT. */ + case R_390_GOTOFF64: /* 64 bit offset to GOT. */ + val = val + rela->r_addend - + ((Elf_Addr) me->module_core + me->arch.got_offset); + if (r_type == R_390_GOTOFF16) + rc = apply_rela_bits(loc, val, 0, 16, 0); + else if (r_type == R_390_GOTOFF32) + rc = apply_rela_bits(loc, val, 0, 32, 0); + else if (r_type == R_390_GOTOFF64) + rc = apply_rela_bits(loc, val, 0, 64, 0); + break; + case R_390_GOTPC: /* 32 bit PC relative offset to GOT. */ + case R_390_GOTPCDBL: /* 32 bit PC rel. off. to GOT shifted by 1. */ + val = (Elf_Addr) me->module_core + me->arch.got_offset + + rela->r_addend - loc; + if (r_type == R_390_GOTPC) + rc = apply_rela_bits(loc, val, 1, 32, 0); + else if (r_type == R_390_GOTPCDBL) + rc = apply_rela_bits(loc, val, 1, 32, 1); + break; + case R_390_COPY: + case R_390_GLOB_DAT: /* Create GOT entry. */ + case R_390_JMP_SLOT: /* Create PLT entry. */ + case R_390_RELATIVE: /* Adjust by program base. */ + /* Only needed if we want to support loading of + modules linked with -shared. */ + return -ENOEXEC; + default: + printk(KERN_ERR "module %s: unknown relocation: %u\n", + me->name, r_type); + return -ENOEXEC; + } + if (rc) { + printk(KERN_ERR "module %s: relocation error for symbol %s " + "(r_type %i, value 0x%lx)\n", + me->name, strtab + symtab[r_sym].st_name, + r_type, (unsigned long) val); + return rc; + } + return 0; +} + +int apply_relocate_add(Elf_Shdr *sechdrs, const char *strtab, + unsigned int symindex, unsigned int relsec, + struct module *me) +{ + Elf_Addr base; + Elf_Sym *symtab; + Elf_Rela *rela; + unsigned long i, n; + int rc; + + DEBUGP("Applying relocate section %u to %u\n", + relsec, sechdrs[relsec].sh_info); + base = sechdrs[sechdrs[relsec].sh_info].sh_addr; + symtab = (Elf_Sym *) sechdrs[symindex].sh_addr; + rela = (Elf_Rela *) sechdrs[relsec].sh_addr; + n = sechdrs[relsec].sh_size / sizeof(Elf_Rela); + + for (i = 0; i < n; i++, rela++) { + rc = apply_rela(rela, base, symtab, strtab, me); + if (rc) + return rc; + } + return 0; +} + +int module_finalize(const Elf_Ehdr *hdr, + const Elf_Shdr *sechdrs, + struct module *me) +{ + jump_label_apply_nops(me); + vfree(me->arch.syminfo); + me->arch.syminfo = NULL; + return 0; +} diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c new file mode 100644 index 000000000..505c17c0a --- /dev/null +++ b/arch/s390/kernel/nmi.c @@ -0,0 +1,352 @@ +/* + * Machine check handler + * + * Copyright IBM Corp. 2000, 2009 + * Author(s): Ingo Adlung <adlung@de.ibm.com>, + * Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Cornelia Huck <cornelia.huck@de.ibm.com>, + * Heiko Carstens <heiko.carstens@de.ibm.com>, + */ + +#include <linux/kernel_stat.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <linux/hardirq.h> +#include <linux/time.h> +#include <linux/module.h> +#include <asm/lowcore.h> +#include <asm/smp.h> +#include <asm/etr.h> +#include <asm/cputime.h> +#include <asm/nmi.h> +#include <asm/crw.h> +#include <asm/switch_to.h> + +struct mcck_struct { + int kill_task; + int channel_report; + int warning; + unsigned long long mcck_code; +}; + +static DEFINE_PER_CPU(struct mcck_struct, cpu_mcck); + +static void s390_handle_damage(char *msg) +{ + smp_send_stop(); + disabled_wait((unsigned long) __builtin_return_address(0)); + while (1); +} + +/* + * Main machine check handler function. Will be called with interrupts enabled + * or disabled and machine checks enabled or disabled. + */ +void s390_handle_mcck(void) +{ + unsigned long flags; + struct mcck_struct mcck; + + /* + * Disable machine checks and get the current state of accumulated + * machine checks. Afterwards delete the old state and enable machine + * checks again. + */ + local_irq_save(flags); + local_mcck_disable(); + mcck = *this_cpu_ptr(&cpu_mcck); + memset(this_cpu_ptr(&cpu_mcck), 0, sizeof(mcck)); + clear_cpu_flag(CIF_MCCK_PENDING); + local_mcck_enable(); + local_irq_restore(flags); + + if (mcck.channel_report) + crw_handle_channel_report(); + /* + * A warning may remain for a prolonged period on the bare iron. + * (actually until the machine is powered off, or the problem is gone) + * So we just stop listening for the WARNING MCH and avoid continuously + * being interrupted. One caveat is however, that we must do this per + * processor and cannot use the smp version of ctl_clear_bit(). + * On VM we only get one interrupt per virtally presented machinecheck. + * Though one suffices, we may get one interrupt per (virtual) cpu. + */ + if (mcck.warning) { /* WARNING pending ? */ + static int mchchk_wng_posted = 0; + + /* Use single cpu clear, as we cannot handle smp here. */ + __ctl_clear_bit(14, 24); /* Disable WARNING MCH */ + if (xchg(&mchchk_wng_posted, 1) == 0) + kill_cad_pid(SIGPWR, 1); + } + if (mcck.kill_task) { + local_irq_enable(); + printk(KERN_EMERG "mcck: Terminating task because of machine " + "malfunction (code 0x%016llx).\n", mcck.mcck_code); + printk(KERN_EMERG "mcck: task: %s, pid: %d.\n", + current->comm, current->pid); + do_exit(SIGSEGV); + } +} +EXPORT_SYMBOL_GPL(s390_handle_mcck); + +/* + * returns 0 if all registers could be validated + * returns 1 otherwise + */ +static int notrace s390_revalidate_registers(struct mci *mci) +{ + int kill_task; + u64 zero; + void *fpt_save_area, *fpt_creg_save_area; + + kill_task = 0; + zero = 0; + + if (!mci->gr) { + /* + * General purpose registers couldn't be restored and have + * unknown contents. Process needs to be terminated. + */ + kill_task = 1; + } + if (!mci->fp) { + /* + * Floating point registers can't be restored and + * therefore the process needs to be terminated. + */ + kill_task = 1; + } + fpt_save_area = &S390_lowcore.floating_pt_save_area; + fpt_creg_save_area = &S390_lowcore.fpt_creg_save_area; + if (!mci->fc) { + /* + * Floating point control register can't be restored. + * Task will be terminated. + */ + asm volatile("lfpc 0(%0)" : : "a" (&zero), "m" (zero)); + kill_task = 1; + } else + asm volatile("lfpc 0(%0)" : : "a" (fpt_creg_save_area)); + + asm volatile( + " ld 0,0(%0)\n" + " ld 1,8(%0)\n" + " ld 2,16(%0)\n" + " ld 3,24(%0)\n" + " ld 4,32(%0)\n" + " ld 5,40(%0)\n" + " ld 6,48(%0)\n" + " ld 7,56(%0)\n" + " ld 8,64(%0)\n" + " ld 9,72(%0)\n" + " ld 10,80(%0)\n" + " ld 11,88(%0)\n" + " ld 12,96(%0)\n" + " ld 13,104(%0)\n" + " ld 14,112(%0)\n" + " ld 15,120(%0)\n" + : : "a" (fpt_save_area)); + /* Revalidate vector registers */ + if (MACHINE_HAS_VX && current->thread.vxrs) { + if (!mci->vr) { + /* + * Vector registers can't be restored and therefore + * the process needs to be terminated. + */ + kill_task = 1; + } + restore_vx_regs((__vector128 *) + S390_lowcore.vector_save_area_addr); + } + /* Revalidate access registers */ + asm volatile( + " lam 0,15,0(%0)" + : : "a" (&S390_lowcore.access_regs_save_area)); + if (!mci->ar) { + /* + * Access registers have unknown contents. + * Terminating task. + */ + kill_task = 1; + } + /* Revalidate control registers */ + if (!mci->cr) { + /* + * Control registers have unknown contents. + * Can't recover and therefore stopping machine. + */ + s390_handle_damage("invalid control registers."); + } else { + asm volatile( + " lctlg 0,15,0(%0)" + : : "a" (&S390_lowcore.cregs_save_area)); + } + /* + * We don't even try to revalidate the TOD register, since we simply + * can't write something sensible into that register. + */ + /* + * See if we can revalidate the TOD programmable register with its + * old contents (should be zero) otherwise set it to zero. + */ + if (!mci->pr) + asm volatile( + " sr 0,0\n" + " sckpf" + : : : "0", "cc"); + else + asm volatile( + " l 0,0(%0)\n" + " sckpf" + : : "a" (&S390_lowcore.tod_progreg_save_area) + : "0", "cc"); + /* Revalidate clock comparator register */ + set_clock_comparator(S390_lowcore.clock_comparator); + /* Check if old PSW is valid */ + if (!mci->wp) + /* + * Can't tell if we come from user or kernel mode + * -> stopping machine. + */ + s390_handle_damage("old psw invalid."); + + if (!mci->ms || !mci->pm || !mci->ia) + kill_task = 1; + + return kill_task; +} + +#define MAX_IPD_COUNT 29 +#define MAX_IPD_TIME (5 * 60 * USEC_PER_SEC) /* 5 minutes */ + +#define ED_STP_ISLAND 6 /* External damage STP island check */ +#define ED_STP_SYNC 7 /* External damage STP sync check */ +#define ED_ETR_SYNC 12 /* External damage ETR sync check */ +#define ED_ETR_SWITCH 13 /* External damage ETR switch to local */ + +/* + * machine check handler. + */ +void notrace s390_do_machine_check(struct pt_regs *regs) +{ + static int ipd_count; + static DEFINE_SPINLOCK(ipd_lock); + static unsigned long long last_ipd; + struct mcck_struct *mcck; + unsigned long long tmp; + struct mci *mci; + int umode; + + nmi_enter(); + inc_irq_stat(NMI_NMI); + mci = (struct mci *) &S390_lowcore.mcck_interruption_code; + mcck = this_cpu_ptr(&cpu_mcck); + umode = user_mode(regs); + + if (mci->sd) { + /* System damage -> stopping machine */ + s390_handle_damage("received system damage machine check."); + } + if (mci->pd) { + if (mci->b) { + /* Processing backup -> verify if we can survive this */ + u64 z_mcic, o_mcic, t_mcic; + z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<29); + o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 | + 1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 | + 1ULL<<30 | 1ULL<<21 | 1ULL<<20 | 1ULL<<17 | + 1ULL<<16); + t_mcic = *(u64 *)mci; + + if (((t_mcic & z_mcic) != 0) || + ((t_mcic & o_mcic) != o_mcic)) { + s390_handle_damage("processing backup machine " + "check with damage."); + } + + /* + * Nullifying exigent condition, therefore we might + * retry this instruction. + */ + spin_lock(&ipd_lock); + tmp = get_tod_clock(); + if (((tmp - last_ipd) >> 12) < MAX_IPD_TIME) + ipd_count++; + else + ipd_count = 1; + last_ipd = tmp; + if (ipd_count == MAX_IPD_COUNT) + s390_handle_damage("too many ipd retries."); + spin_unlock(&ipd_lock); + } else { + /* Processing damage -> stopping machine */ + s390_handle_damage("received instruction processing " + "damage machine check."); + } + } + if (s390_revalidate_registers(mci)) { + if (umode) { + /* + * Couldn't restore all register contents while in + * user mode -> mark task for termination. + */ + mcck->kill_task = 1; + mcck->mcck_code = *(unsigned long long *) mci; + set_cpu_flag(CIF_MCCK_PENDING); + } else { + /* + * Couldn't restore all register contents while in + * kernel mode -> stopping machine. + */ + s390_handle_damage("unable to revalidate registers."); + } + } + if (mci->cd) { + /* Timing facility damage */ + s390_handle_damage("TOD clock damaged"); + } + if (mci->ed && mci->ec) { + /* External damage */ + if (S390_lowcore.external_damage_code & (1U << ED_ETR_SYNC)) + etr_sync_check(); + if (S390_lowcore.external_damage_code & (1U << ED_ETR_SWITCH)) + etr_switch_to_local(); + if (S390_lowcore.external_damage_code & (1U << ED_STP_SYNC)) + stp_sync_check(); + if (S390_lowcore.external_damage_code & (1U << ED_STP_ISLAND)) + stp_island_check(); + } + if (mci->se) + /* Storage error uncorrected */ + s390_handle_damage("received storage error uncorrected " + "machine check."); + if (mci->ke) + /* Storage key-error uncorrected */ + s390_handle_damage("received storage key-error uncorrected " + "machine check."); + if (mci->ds && mci->fa) + /* Storage degradation */ + s390_handle_damage("received storage degradation machine " + "check."); + if (mci->cp) { + /* Channel report word pending */ + mcck->channel_report = 1; + set_cpu_flag(CIF_MCCK_PENDING); + } + if (mci->w) { + /* Warning pending */ + mcck->warning = 1; + set_cpu_flag(CIF_MCCK_PENDING); + } + nmi_exit(); +} + +static int __init machine_check_init(void) +{ + ctl_set_bit(14, 25); /* enable external damage MCH */ + ctl_set_bit(14, 27); /* enable system recovery MCH */ + ctl_set_bit(14, 24); /* enable warning MCH */ + return 0; +} +arch_initcall(machine_check_init); diff --git a/arch/s390/kernel/os_info.c b/arch/s390/kernel/os_info.c new file mode 100644 index 000000000..d112fc66f --- /dev/null +++ b/arch/s390/kernel/os_info.c @@ -0,0 +1,168 @@ +/* + * OS info memory interface + * + * Copyright IBM Corp. 2012 + * Author(s): Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#define KMSG_COMPONENT "os_info" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/crash_dump.h> +#include <linux/kernel.h> +#include <linux/slab.h> +#include <asm/checksum.h> +#include <asm/lowcore.h> +#include <asm/os_info.h> + +/* + * OS info structure has to be page aligned + */ +static struct os_info os_info __page_aligned_data; + +/* + * Compute checksum over OS info structure + */ +u32 os_info_csum(struct os_info *os_info) +{ + int size = sizeof(*os_info) - offsetof(struct os_info, version_major); + return csum_partial(&os_info->version_major, size, 0); +} + +/* + * Add crashkernel info to OS info and update checksum + */ +void os_info_crashkernel_add(unsigned long base, unsigned long size) +{ + os_info.crashkernel_addr = (u64)(unsigned long)base; + os_info.crashkernel_size = (u64)(unsigned long)size; + os_info.csum = os_info_csum(&os_info); +} + +/* + * Add OS info entry and update checksum + */ +void os_info_entry_add(int nr, void *ptr, u64 size) +{ + os_info.entry[nr].addr = (u64)(unsigned long)ptr; + os_info.entry[nr].size = size; + os_info.entry[nr].csum = csum_partial(ptr, size, 0); + os_info.csum = os_info_csum(&os_info); +} + +/* + * Initialize OS info struture and set lowcore pointer + */ +void __init os_info_init(void) +{ + void *ptr = &os_info; + + os_info.version_major = OS_INFO_VERSION_MAJOR; + os_info.version_minor = OS_INFO_VERSION_MINOR; + os_info.magic = OS_INFO_MAGIC; + os_info.csum = os_info_csum(&os_info); + mem_assign_absolute(S390_lowcore.os_info, (unsigned long) ptr); +} + +#ifdef CONFIG_CRASH_DUMP + +static struct os_info *os_info_old; + +/* + * Allocate and copy OS info entry from oldmem + */ +static void os_info_old_alloc(int nr, int align) +{ + unsigned long addr, size = 0; + char *buf, *buf_align, *msg; + u32 csum; + + addr = os_info_old->entry[nr].addr; + if (!addr) { + msg = "not available"; + goto fail; + } + size = os_info_old->entry[nr].size; + buf = kmalloc(size + align - 1, GFP_KERNEL); + if (!buf) { + msg = "alloc failed"; + goto fail; + } + buf_align = PTR_ALIGN(buf, align); + if (copy_from_oldmem(buf_align, (void *) addr, size)) { + msg = "copy failed"; + goto fail_free; + } + csum = csum_partial(buf_align, size, 0); + if (csum != os_info_old->entry[nr].csum) { + msg = "checksum failed"; + goto fail_free; + } + os_info_old->entry[nr].addr = (u64)(unsigned long)buf_align; + msg = "copied"; + goto out; +fail_free: + kfree(buf); +fail: + os_info_old->entry[nr].addr = 0; +out: + pr_info("entry %i: %s (addr=0x%lx size=%lu)\n", + nr, msg, addr, size); +} + +/* + * Initialize os info and os info entries from oldmem + */ +static void os_info_old_init(void) +{ + static int os_info_init; + unsigned long addr; + + if (os_info_init) + return; + if (!OLDMEM_BASE) + goto fail; + if (copy_from_oldmem(&addr, &S390_lowcore.os_info, sizeof(addr))) + goto fail; + if (addr == 0 || addr % PAGE_SIZE) + goto fail; + os_info_old = kzalloc(sizeof(*os_info_old), GFP_KERNEL); + if (!os_info_old) + goto fail; + if (copy_from_oldmem(os_info_old, (void *) addr, sizeof(*os_info_old))) + goto fail_free; + if (os_info_old->magic != OS_INFO_MAGIC) + goto fail_free; + if (os_info_old->csum != os_info_csum(os_info_old)) + goto fail_free; + if (os_info_old->version_major > OS_INFO_VERSION_MAJOR) + goto fail_free; + os_info_old_alloc(OS_INFO_VMCOREINFO, 1); + os_info_old_alloc(OS_INFO_REIPL_BLOCK, 1); + pr_info("crashkernel: addr=0x%lx size=%lu\n", + (unsigned long) os_info_old->crashkernel_addr, + (unsigned long) os_info_old->crashkernel_size); + os_info_init = 1; + return; +fail_free: + kfree(os_info_old); +fail: + os_info_init = 1; + os_info_old = NULL; +} + +/* + * Return pointer to os infor entry and its size + */ +void *os_info_old_entry(int nr, unsigned long *size) +{ + os_info_old_init(); + + if (!os_info_old) + return NULL; + if (!os_info_old->entry[nr].addr) + return NULL; + *size = (unsigned long) os_info_old->entry[nr].size; + return (void *)(unsigned long)os_info_old->entry[nr].addr; +} +#endif diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c new file mode 100644 index 000000000..56fdad479 --- /dev/null +++ b/arch/s390/kernel/perf_cpum_cf.c @@ -0,0 +1,696 @@ +/* + * Performance event support for s390x - CPU-measurement Counter Facility + * + * Copyright IBM Corp. 2012 + * Author(s): Hendrik Brueckner <brueckner@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. + */ +#define KMSG_COMPONENT "cpum_cf" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/kernel_stat.h> +#include <linux/perf_event.h> +#include <linux/percpu.h> +#include <linux/notifier.h> +#include <linux/init.h> +#include <linux/export.h> +#include <asm/ctl_reg.h> +#include <asm/irq.h> +#include <asm/cpu_mf.h> + +/* CPU-measurement counter facility supports these CPU counter sets: + * For CPU counter sets: + * Basic counter set: 0-31 + * Problem-state counter set: 32-63 + * Crypto-activity counter set: 64-127 + * Extented counter set: 128-159 + */ +enum cpumf_ctr_set { + /* CPU counter sets */ + CPUMF_CTR_SET_BASIC = 0, + CPUMF_CTR_SET_USER = 1, + CPUMF_CTR_SET_CRYPTO = 2, + CPUMF_CTR_SET_EXT = 3, + + /* Maximum number of counter sets */ + CPUMF_CTR_SET_MAX, +}; + +#define CPUMF_LCCTL_ENABLE_SHIFT 16 +#define CPUMF_LCCTL_ACTCTL_SHIFT 0 +static const u64 cpumf_state_ctl[CPUMF_CTR_SET_MAX] = { + [CPUMF_CTR_SET_BASIC] = 0x02, + [CPUMF_CTR_SET_USER] = 0x04, + [CPUMF_CTR_SET_CRYPTO] = 0x08, + [CPUMF_CTR_SET_EXT] = 0x01, +}; + +static void ctr_set_enable(u64 *state, int ctr_set) +{ + *state |= cpumf_state_ctl[ctr_set] << CPUMF_LCCTL_ENABLE_SHIFT; +} +static void ctr_set_disable(u64 *state, int ctr_set) +{ + *state &= ~(cpumf_state_ctl[ctr_set] << CPUMF_LCCTL_ENABLE_SHIFT); +} +static void ctr_set_start(u64 *state, int ctr_set) +{ + *state |= cpumf_state_ctl[ctr_set] << CPUMF_LCCTL_ACTCTL_SHIFT; +} +static void ctr_set_stop(u64 *state, int ctr_set) +{ + *state &= ~(cpumf_state_ctl[ctr_set] << CPUMF_LCCTL_ACTCTL_SHIFT); +} + +/* Local CPUMF event structure */ +struct cpu_hw_events { + struct cpumf_ctr_info info; + atomic_t ctr_set[CPUMF_CTR_SET_MAX]; + u64 state, tx_state; + unsigned int flags; +}; +static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { + .ctr_set = { + [CPUMF_CTR_SET_BASIC] = ATOMIC_INIT(0), + [CPUMF_CTR_SET_USER] = ATOMIC_INIT(0), + [CPUMF_CTR_SET_CRYPTO] = ATOMIC_INIT(0), + [CPUMF_CTR_SET_EXT] = ATOMIC_INIT(0), + }, + .state = 0, + .flags = 0, +}; + +static int get_counter_set(u64 event) +{ + int set = -1; + + if (event < 32) + set = CPUMF_CTR_SET_BASIC; + else if (event < 64) + set = CPUMF_CTR_SET_USER; + else if (event < 128) + set = CPUMF_CTR_SET_CRYPTO; + else if (event < 256) + set = CPUMF_CTR_SET_EXT; + + return set; +} + +static int validate_event(const struct hw_perf_event *hwc) +{ + switch (hwc->config_base) { + case CPUMF_CTR_SET_BASIC: + case CPUMF_CTR_SET_USER: + case CPUMF_CTR_SET_CRYPTO: + case CPUMF_CTR_SET_EXT: + /* check for reserved counters */ + if ((hwc->config >= 6 && hwc->config <= 31) || + (hwc->config >= 38 && hwc->config <= 63) || + (hwc->config >= 80 && hwc->config <= 127)) + return -EOPNOTSUPP; + break; + default: + return -EINVAL; + } + + return 0; +} + +static int validate_ctr_version(const struct hw_perf_event *hwc) +{ + struct cpu_hw_events *cpuhw; + int err = 0; + + cpuhw = &get_cpu_var(cpu_hw_events); + + /* check required version for counter sets */ + switch (hwc->config_base) { + case CPUMF_CTR_SET_BASIC: + case CPUMF_CTR_SET_USER: + if (cpuhw->info.cfvn < 1) + err = -EOPNOTSUPP; + break; + case CPUMF_CTR_SET_CRYPTO: + case CPUMF_CTR_SET_EXT: + if (cpuhw->info.csvn < 1) + err = -EOPNOTSUPP; + if ((cpuhw->info.csvn == 1 && hwc->config > 159) || + (cpuhw->info.csvn == 2 && hwc->config > 175) || + (cpuhw->info.csvn > 2 && hwc->config > 255)) + err = -EOPNOTSUPP; + break; + } + + put_cpu_var(cpu_hw_events); + return err; +} + +static int validate_ctr_auth(const struct hw_perf_event *hwc) +{ + struct cpu_hw_events *cpuhw; + u64 ctrs_state; + int err = 0; + + cpuhw = &get_cpu_var(cpu_hw_events); + + /* check authorization for cpu counter sets */ + ctrs_state = cpumf_state_ctl[hwc->config_base]; + if (!(ctrs_state & cpuhw->info.auth_ctl)) + err = -EPERM; + + put_cpu_var(cpu_hw_events); + return err; +} + +/* + * Change the CPUMF state to active. + * Enable and activate the CPU-counter sets according + * to the per-cpu control state. + */ +static void cpumf_pmu_enable(struct pmu *pmu) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + int err; + + if (cpuhw->flags & PMU_F_ENABLED) + return; + + err = lcctl(cpuhw->state); + if (err) { + pr_err("Enabling the performance measuring unit " + "failed with rc=%x\n", err); + return; + } + + cpuhw->flags |= PMU_F_ENABLED; +} + +/* + * Change the CPUMF state to inactive. + * Disable and enable (inactive) the CPU-counter sets according + * to the per-cpu control state. + */ +static void cpumf_pmu_disable(struct pmu *pmu) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + int err; + u64 inactive; + + if (!(cpuhw->flags & PMU_F_ENABLED)) + return; + + inactive = cpuhw->state & ~((1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1); + err = lcctl(inactive); + if (err) { + pr_err("Disabling the performance measuring unit " + "failed with rc=%x\n", err); + return; + } + + cpuhw->flags &= ~PMU_F_ENABLED; +} + + +/* Number of perf events counting hardware events */ +static atomic_t num_events = ATOMIC_INIT(0); +/* Used to avoid races in calling reserve/release_cpumf_hardware */ +static DEFINE_MUTEX(pmc_reserve_mutex); + +/* CPU-measurement alerts for the counter facility */ +static void cpumf_measurement_alert(struct ext_code ext_code, + unsigned int alert, unsigned long unused) +{ + struct cpu_hw_events *cpuhw; + + if (!(alert & CPU_MF_INT_CF_MASK)) + return; + + inc_irq_stat(IRQEXT_CMC); + cpuhw = this_cpu_ptr(&cpu_hw_events); + + /* Measurement alerts are shared and might happen when the PMU + * is not reserved. Ignore these alerts in this case. */ + if (!(cpuhw->flags & PMU_F_RESERVED)) + return; + + /* counter authorization change alert */ + if (alert & CPU_MF_INT_CF_CACA) + qctri(&cpuhw->info); + + /* loss of counter data alert */ + if (alert & CPU_MF_INT_CF_LCDA) + pr_err("CPU[%i] Counter data was lost\n", smp_processor_id()); +} + +#define PMC_INIT 0 +#define PMC_RELEASE 1 +static void setup_pmc_cpu(void *flags) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + + switch (*((int *) flags)) { + case PMC_INIT: + memset(&cpuhw->info, 0, sizeof(cpuhw->info)); + qctri(&cpuhw->info); + cpuhw->flags |= PMU_F_RESERVED; + break; + + case PMC_RELEASE: + cpuhw->flags &= ~PMU_F_RESERVED; + break; + } + + /* Disable CPU counter sets */ + lcctl(0); +} + +/* Initialize the CPU-measurement facility */ +static int reserve_pmc_hardware(void) +{ + int flags = PMC_INIT; + + on_each_cpu(setup_pmc_cpu, &flags, 1); + irq_subclass_register(IRQ_SUBCLASS_MEASUREMENT_ALERT); + + return 0; +} + +/* Release the CPU-measurement facility */ +static void release_pmc_hardware(void) +{ + int flags = PMC_RELEASE; + + on_each_cpu(setup_pmc_cpu, &flags, 1); + irq_subclass_unregister(IRQ_SUBCLASS_MEASUREMENT_ALERT); +} + +/* Release the PMU if event is the last perf event */ +static void hw_perf_event_destroy(struct perf_event *event) +{ + if (!atomic_add_unless(&num_events, -1, 1)) { + mutex_lock(&pmc_reserve_mutex); + if (atomic_dec_return(&num_events) == 0) + release_pmc_hardware(); + mutex_unlock(&pmc_reserve_mutex); + } +} + +/* CPUMF <-> perf event mappings for kernel+userspace (basic set) */ +static const int cpumf_generic_events_basic[] = { + [PERF_COUNT_HW_CPU_CYCLES] = 0, + [PERF_COUNT_HW_INSTRUCTIONS] = 1, + [PERF_COUNT_HW_CACHE_REFERENCES] = -1, + [PERF_COUNT_HW_CACHE_MISSES] = -1, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = -1, + [PERF_COUNT_HW_BRANCH_MISSES] = -1, + [PERF_COUNT_HW_BUS_CYCLES] = -1, +}; +/* CPUMF <-> perf event mappings for userspace (problem-state set) */ +static const int cpumf_generic_events_user[] = { + [PERF_COUNT_HW_CPU_CYCLES] = 32, + [PERF_COUNT_HW_INSTRUCTIONS] = 33, + [PERF_COUNT_HW_CACHE_REFERENCES] = -1, + [PERF_COUNT_HW_CACHE_MISSES] = -1, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = -1, + [PERF_COUNT_HW_BRANCH_MISSES] = -1, + [PERF_COUNT_HW_BUS_CYCLES] = -1, +}; + +static int __hw_perf_event_init(struct perf_event *event) +{ + struct perf_event_attr *attr = &event->attr; + struct hw_perf_event *hwc = &event->hw; + int err; + u64 ev; + + switch (attr->type) { + case PERF_TYPE_RAW: + /* Raw events are used to access counters directly, + * hence do not permit excludes */ + if (attr->exclude_kernel || attr->exclude_user || + attr->exclude_hv) + return -EOPNOTSUPP; + ev = attr->config; + break; + + case PERF_TYPE_HARDWARE: + ev = attr->config; + /* Count user space (problem-state) only */ + if (!attr->exclude_user && attr->exclude_kernel) { + if (ev >= ARRAY_SIZE(cpumf_generic_events_user)) + return -EOPNOTSUPP; + ev = cpumf_generic_events_user[ev]; + + /* No support for kernel space counters only */ + } else if (!attr->exclude_kernel && attr->exclude_user) { + return -EOPNOTSUPP; + + /* Count user and kernel space */ + } else { + if (ev >= ARRAY_SIZE(cpumf_generic_events_basic)) + return -EOPNOTSUPP; + ev = cpumf_generic_events_basic[ev]; + } + break; + + default: + return -ENOENT; + } + + if (ev == -1) + return -ENOENT; + + if (ev >= PERF_CPUM_CF_MAX_CTR) + return -EINVAL; + + /* Use the hardware perf event structure to store the counter number + * in 'config' member and the counter set to which the counter belongs + * in the 'config_base'. The counter set (config_base) is then used + * to enable/disable the counters. + */ + hwc->config = ev; + hwc->config_base = get_counter_set(ev); + + /* Validate the counter that is assigned to this event. + * Because the counter facility can use numerous counters at the + * same time without constraints, it is not necessary to explicity + * validate event groups (event->group_leader != event). + */ + err = validate_event(hwc); + if (err) + return err; + + /* Initialize for using the CPU-measurement counter facility */ + if (!atomic_inc_not_zero(&num_events)) { + mutex_lock(&pmc_reserve_mutex); + if (atomic_read(&num_events) == 0 && reserve_pmc_hardware()) + err = -EBUSY; + else + atomic_inc(&num_events); + mutex_unlock(&pmc_reserve_mutex); + } + event->destroy = hw_perf_event_destroy; + + /* Finally, validate version and authorization of the counter set */ + err = validate_ctr_auth(hwc); + if (!err) + err = validate_ctr_version(hwc); + + return err; +} + +static int cpumf_pmu_event_init(struct perf_event *event) +{ + int err; + + switch (event->attr.type) { + case PERF_TYPE_HARDWARE: + case PERF_TYPE_HW_CACHE: + case PERF_TYPE_RAW: + err = __hw_perf_event_init(event); + break; + default: + return -ENOENT; + } + + if (unlikely(err) && event->destroy) + event->destroy(event); + + return err; +} + +static int hw_perf_event_reset(struct perf_event *event) +{ + u64 prev, new; + int err; + + do { + prev = local64_read(&event->hw.prev_count); + err = ecctr(event->hw.config, &new); + if (err) { + if (err != 3) + break; + /* The counter is not (yet) available. This + * might happen if the counter set to which + * this counter belongs is in the disabled + * state. + */ + new = 0; + } + } while (local64_cmpxchg(&event->hw.prev_count, prev, new) != prev); + + return err; +} + +static int hw_perf_event_update(struct perf_event *event) +{ + u64 prev, new, delta; + int err; + + do { + prev = local64_read(&event->hw.prev_count); + err = ecctr(event->hw.config, &new); + if (err) + goto out; + } while (local64_cmpxchg(&event->hw.prev_count, prev, new) != prev); + + delta = (prev <= new) ? new - prev + : (-1ULL - prev) + new + 1; /* overflow */ + local64_add(delta, &event->count); +out: + return err; +} + +static void cpumf_pmu_read(struct perf_event *event) +{ + if (event->hw.state & PERF_HES_STOPPED) + return; + + hw_perf_event_update(event); +} + +static void cpumf_pmu_start(struct perf_event *event, int flags) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + + if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED))) + return; + + if (WARN_ON_ONCE(hwc->config == -1)) + return; + + if (flags & PERF_EF_RELOAD) + WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE)); + + hwc->state = 0; + + /* (Re-)enable and activate the counter set */ + ctr_set_enable(&cpuhw->state, hwc->config_base); + ctr_set_start(&cpuhw->state, hwc->config_base); + + /* The counter set to which this counter belongs can be already active. + * Because all counters in a set are active, the event->hw.prev_count + * needs to be synchronized. At this point, the counter set can be in + * the inactive or disabled state. + */ + hw_perf_event_reset(event); + + /* increment refcount for this counter set */ + atomic_inc(&cpuhw->ctr_set[hwc->config_base]); +} + +static void cpumf_pmu_stop(struct perf_event *event, int flags) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + struct hw_perf_event *hwc = &event->hw; + + if (!(hwc->state & PERF_HES_STOPPED)) { + /* Decrement reference count for this counter set and if this + * is the last used counter in the set, clear activation + * control and set the counter set state to inactive. + */ + if (!atomic_dec_return(&cpuhw->ctr_set[hwc->config_base])) + ctr_set_stop(&cpuhw->state, hwc->config_base); + event->hw.state |= PERF_HES_STOPPED; + } + + if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) { + hw_perf_event_update(event); + event->hw.state |= PERF_HES_UPTODATE; + } +} + +static int cpumf_pmu_add(struct perf_event *event, int flags) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + + /* Check authorization for the counter set to which this + * counter belongs. + * For group events transaction, the authorization check is + * done in cpumf_pmu_commit_txn(). + */ + if (!(cpuhw->flags & PERF_EVENT_TXN)) + if (validate_ctr_auth(&event->hw)) + return -EPERM; + + ctr_set_enable(&cpuhw->state, event->hw.config_base); + event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED; + + if (flags & PERF_EF_START) + cpumf_pmu_start(event, PERF_EF_RELOAD); + + perf_event_update_userpage(event); + + return 0; +} + +static void cpumf_pmu_del(struct perf_event *event, int flags) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + + cpumf_pmu_stop(event, PERF_EF_UPDATE); + + /* Check if any counter in the counter set is still used. If not used, + * change the counter set to the disabled state. This also clears the + * content of all counters in the set. + * + * When a new perf event has been added but not yet started, this can + * clear enable control and resets all counters in a set. Therefore, + * cpumf_pmu_start() always has to reenable a counter set. + */ + if (!atomic_read(&cpuhw->ctr_set[event->hw.config_base])) + ctr_set_disable(&cpuhw->state, event->hw.config_base); + + perf_event_update_userpage(event); +} + +/* + * Start group events scheduling transaction. + * Set flags to perform a single test at commit time. + */ +static void cpumf_pmu_start_txn(struct pmu *pmu) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + + perf_pmu_disable(pmu); + cpuhw->flags |= PERF_EVENT_TXN; + cpuhw->tx_state = cpuhw->state; +} + +/* + * Stop and cancel a group events scheduling tranctions. + * Assumes cpumf_pmu_del() is called for each successful added + * cpumf_pmu_add() during the transaction. + */ +static void cpumf_pmu_cancel_txn(struct pmu *pmu) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + + WARN_ON(cpuhw->tx_state != cpuhw->state); + + cpuhw->flags &= ~PERF_EVENT_TXN; + perf_pmu_enable(pmu); +} + +/* + * Commit the group events scheduling transaction. On success, the + * transaction is closed. On error, the transaction is kept open + * until cpumf_pmu_cancel_txn() is called. + */ +static int cpumf_pmu_commit_txn(struct pmu *pmu) +{ + struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events); + u64 state; + + /* check if the updated state can be scheduled */ + state = cpuhw->state & ~((1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1); + state >>= CPUMF_LCCTL_ENABLE_SHIFT; + if ((state & cpuhw->info.auth_ctl) != state) + return -EPERM; + + cpuhw->flags &= ~PERF_EVENT_TXN; + perf_pmu_enable(pmu); + return 0; +} + +/* Performance monitoring unit for s390x */ +static struct pmu cpumf_pmu = { + .pmu_enable = cpumf_pmu_enable, + .pmu_disable = cpumf_pmu_disable, + .event_init = cpumf_pmu_event_init, + .add = cpumf_pmu_add, + .del = cpumf_pmu_del, + .start = cpumf_pmu_start, + .stop = cpumf_pmu_stop, + .read = cpumf_pmu_read, + .start_txn = cpumf_pmu_start_txn, + .commit_txn = cpumf_pmu_commit_txn, + .cancel_txn = cpumf_pmu_cancel_txn, +}; + +static int cpumf_pmu_notifier(struct notifier_block *self, unsigned long action, + void *hcpu) +{ + unsigned int cpu = (long) hcpu; + int flags; + + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_ONLINE: + flags = PMC_INIT; + smp_call_function_single(cpu, setup_pmc_cpu, &flags, 1); + break; + case CPU_DOWN_PREPARE: + flags = PMC_RELEASE; + smp_call_function_single(cpu, setup_pmc_cpu, &flags, 1); + break; + default: + break; + } + + return NOTIFY_OK; +} + +static int __init cpumf_pmu_init(void) +{ + int rc; + + if (!cpum_cf_avail()) + return -ENODEV; + + /* clear bit 15 of cr0 to unauthorize problem-state to + * extract measurement counters */ + ctl_clear_bit(0, 48); + + /* register handler for measurement-alert interruptions */ + rc = register_external_irq(EXT_IRQ_MEASURE_ALERT, + cpumf_measurement_alert); + if (rc) { + pr_err("Registering for CPU-measurement alerts " + "failed with rc=%i\n", rc); + goto out; + } + + /* The CPU measurement counter facility does not have overflow + * interrupts to do sampling. Sampling must be provided by + * external means, for example, by timers. + */ + cpumf_pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT; + + cpumf_pmu.attr_groups = cpumf_cf_event_group(); + rc = perf_pmu_register(&cpumf_pmu, "cpum_cf", PERF_TYPE_RAW); + if (rc) { + pr_err("Registering the cpum_cf PMU failed with rc=%i\n", rc); + unregister_external_irq(EXT_IRQ_MEASURE_ALERT, + cpumf_measurement_alert); + goto out; + } + perf_cpu_notifier(cpumf_pmu_notifier); +out: + return rc; +} +early_initcall(cpumf_pmu_init); diff --git a/arch/s390/kernel/perf_cpum_cf_events.c b/arch/s390/kernel/perf_cpum_cf_events.c new file mode 100644 index 000000000..4554a4bae --- /dev/null +++ b/arch/s390/kernel/perf_cpum_cf_events.c @@ -0,0 +1,322 @@ +/* + * Perf PMU sysfs events attributes for available CPU-measurement counters + * + */ + +#include <linux/slab.h> +#include <linux/perf_event.h> + + +/* BEGIN: CPUM_CF COUNTER DEFINITIONS =================================== */ + +CPUMF_EVENT_ATTR(cf, CPU_CYCLES, 0x0000); +CPUMF_EVENT_ATTR(cf, INSTRUCTIONS, 0x0001); +CPUMF_EVENT_ATTR(cf, L1I_DIR_WRITES, 0x0002); +CPUMF_EVENT_ATTR(cf, L1I_PENALTY_CYCLES, 0x0003); +CPUMF_EVENT_ATTR(cf, PROBLEM_STATE_CPU_CYCLES, 0x0020); +CPUMF_EVENT_ATTR(cf, PROBLEM_STATE_INSTRUCTIONS, 0x0021); +CPUMF_EVENT_ATTR(cf, PROBLEM_STATE_L1I_DIR_WRITES, 0x0022); +CPUMF_EVENT_ATTR(cf, PROBLEM_STATE_L1I_PENALTY_CYCLES, 0x0023); +CPUMF_EVENT_ATTR(cf, PROBLEM_STATE_L1D_DIR_WRITES, 0x0024); +CPUMF_EVENT_ATTR(cf, PROBLEM_STATE_L1D_PENALTY_CYCLES, 0x0025); +CPUMF_EVENT_ATTR(cf, L1D_DIR_WRITES, 0x0004); +CPUMF_EVENT_ATTR(cf, L1D_PENALTY_CYCLES, 0x0005); +CPUMF_EVENT_ATTR(cf, PRNG_FUNCTIONS, 0x0040); +CPUMF_EVENT_ATTR(cf, PRNG_CYCLES, 0x0041); +CPUMF_EVENT_ATTR(cf, PRNG_BLOCKED_FUNCTIONS, 0x0042); +CPUMF_EVENT_ATTR(cf, PRNG_BLOCKED_CYCLES, 0x0043); +CPUMF_EVENT_ATTR(cf, SHA_FUNCTIONS, 0x0044); +CPUMF_EVENT_ATTR(cf, SHA_CYCLES, 0x0045); +CPUMF_EVENT_ATTR(cf, SHA_BLOCKED_FUNCTIONS, 0x0046); +CPUMF_EVENT_ATTR(cf, SHA_BLOCKED_CYCLES, 0x0047); +CPUMF_EVENT_ATTR(cf, DEA_FUNCTIONS, 0x0048); +CPUMF_EVENT_ATTR(cf, DEA_CYCLES, 0x0049); +CPUMF_EVENT_ATTR(cf, DEA_BLOCKED_FUNCTIONS, 0x004a); +CPUMF_EVENT_ATTR(cf, DEA_BLOCKED_CYCLES, 0x004b); +CPUMF_EVENT_ATTR(cf, AES_FUNCTIONS, 0x004c); +CPUMF_EVENT_ATTR(cf, AES_CYCLES, 0x004d); +CPUMF_EVENT_ATTR(cf, AES_BLOCKED_FUNCTIONS, 0x004e); +CPUMF_EVENT_ATTR(cf, AES_BLOCKED_CYCLES, 0x004f); +CPUMF_EVENT_ATTR(cf_z10, L1I_L2_SOURCED_WRITES, 0x0080); +CPUMF_EVENT_ATTR(cf_z10, L1D_L2_SOURCED_WRITES, 0x0081); +CPUMF_EVENT_ATTR(cf_z10, L1I_L3_LOCAL_WRITES, 0x0082); +CPUMF_EVENT_ATTR(cf_z10, L1D_L3_LOCAL_WRITES, 0x0083); +CPUMF_EVENT_ATTR(cf_z10, L1I_L3_REMOTE_WRITES, 0x0084); +CPUMF_EVENT_ATTR(cf_z10, L1D_L3_REMOTE_WRITES, 0x0085); +CPUMF_EVENT_ATTR(cf_z10, L1D_LMEM_SOURCED_WRITES, 0x0086); +CPUMF_EVENT_ATTR(cf_z10, L1I_LMEM_SOURCED_WRITES, 0x0087); +CPUMF_EVENT_ATTR(cf_z10, L1D_RO_EXCL_WRITES, 0x0088); +CPUMF_EVENT_ATTR(cf_z10, L1I_CACHELINE_INVALIDATES, 0x0089); +CPUMF_EVENT_ATTR(cf_z10, ITLB1_WRITES, 0x008a); +CPUMF_EVENT_ATTR(cf_z10, DTLB1_WRITES, 0x008b); +CPUMF_EVENT_ATTR(cf_z10, TLB2_PTE_WRITES, 0x008c); +CPUMF_EVENT_ATTR(cf_z10, TLB2_CRSTE_WRITES, 0x008d); +CPUMF_EVENT_ATTR(cf_z10, TLB2_CRSTE_HPAGE_WRITES, 0x008e); +CPUMF_EVENT_ATTR(cf_z10, ITLB1_MISSES, 0x0091); +CPUMF_EVENT_ATTR(cf_z10, DTLB1_MISSES, 0x0092); +CPUMF_EVENT_ATTR(cf_z10, L2C_STORES_SENT, 0x0093); +CPUMF_EVENT_ATTR(cf_z196, L1D_L2_SOURCED_WRITES, 0x0080); +CPUMF_EVENT_ATTR(cf_z196, L1I_L2_SOURCED_WRITES, 0x0081); +CPUMF_EVENT_ATTR(cf_z196, DTLB1_MISSES, 0x0082); +CPUMF_EVENT_ATTR(cf_z196, ITLB1_MISSES, 0x0083); +CPUMF_EVENT_ATTR(cf_z196, L2C_STORES_SENT, 0x0085); +CPUMF_EVENT_ATTR(cf_z196, L1D_OFFBOOK_L3_SOURCED_WRITES, 0x0086); +CPUMF_EVENT_ATTR(cf_z196, L1D_ONBOOK_L4_SOURCED_WRITES, 0x0087); +CPUMF_EVENT_ATTR(cf_z196, L1I_ONBOOK_L4_SOURCED_WRITES, 0x0088); +CPUMF_EVENT_ATTR(cf_z196, L1D_RO_EXCL_WRITES, 0x0089); +CPUMF_EVENT_ATTR(cf_z196, L1D_OFFBOOK_L4_SOURCED_WRITES, 0x008a); +CPUMF_EVENT_ATTR(cf_z196, L1I_OFFBOOK_L4_SOURCED_WRITES, 0x008b); +CPUMF_EVENT_ATTR(cf_z196, DTLB1_HPAGE_WRITES, 0x008c); +CPUMF_EVENT_ATTR(cf_z196, L1D_LMEM_SOURCED_WRITES, 0x008d); +CPUMF_EVENT_ATTR(cf_z196, L1I_LMEM_SOURCED_WRITES, 0x008e); +CPUMF_EVENT_ATTR(cf_z196, L1I_OFFBOOK_L3_SOURCED_WRITES, 0x008f); +CPUMF_EVENT_ATTR(cf_z196, DTLB1_WRITES, 0x0090); +CPUMF_EVENT_ATTR(cf_z196, ITLB1_WRITES, 0x0091); +CPUMF_EVENT_ATTR(cf_z196, TLB2_PTE_WRITES, 0x0092); +CPUMF_EVENT_ATTR(cf_z196, TLB2_CRSTE_HPAGE_WRITES, 0x0093); +CPUMF_EVENT_ATTR(cf_z196, TLB2_CRSTE_WRITES, 0x0094); +CPUMF_EVENT_ATTR(cf_z196, L1D_ONCHIP_L3_SOURCED_WRITES, 0x0096); +CPUMF_EVENT_ATTR(cf_z196, L1D_OFFCHIP_L3_SOURCED_WRITES, 0x0098); +CPUMF_EVENT_ATTR(cf_z196, L1I_ONCHIP_L3_SOURCED_WRITES, 0x0099); +CPUMF_EVENT_ATTR(cf_z196, L1I_OFFCHIP_L3_SOURCED_WRITES, 0x009b); +CPUMF_EVENT_ATTR(cf_zec12, DTLB1_MISSES, 0x0080); +CPUMF_EVENT_ATTR(cf_zec12, ITLB1_MISSES, 0x0081); +CPUMF_EVENT_ATTR(cf_zec12, L1D_L2I_SOURCED_WRITES, 0x0082); +CPUMF_EVENT_ATTR(cf_zec12, L1I_L2I_SOURCED_WRITES, 0x0083); +CPUMF_EVENT_ATTR(cf_zec12, L1D_L2D_SOURCED_WRITES, 0x0084); +CPUMF_EVENT_ATTR(cf_zec12, DTLB1_WRITES, 0x0085); +CPUMF_EVENT_ATTR(cf_zec12, L1D_LMEM_SOURCED_WRITES, 0x0087); +CPUMF_EVENT_ATTR(cf_zec12, L1I_LMEM_SOURCED_WRITES, 0x0089); +CPUMF_EVENT_ATTR(cf_zec12, L1D_RO_EXCL_WRITES, 0x008a); +CPUMF_EVENT_ATTR(cf_zec12, DTLB1_HPAGE_WRITES, 0x008b); +CPUMF_EVENT_ATTR(cf_zec12, ITLB1_WRITES, 0x008c); +CPUMF_EVENT_ATTR(cf_zec12, TLB2_PTE_WRITES, 0x008d); +CPUMF_EVENT_ATTR(cf_zec12, TLB2_CRSTE_HPAGE_WRITES, 0x008e); +CPUMF_EVENT_ATTR(cf_zec12, TLB2_CRSTE_WRITES, 0x008f); +CPUMF_EVENT_ATTR(cf_zec12, L1D_ONCHIP_L3_SOURCED_WRITES, 0x0090); +CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFCHIP_L3_SOURCED_WRITES, 0x0091); +CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFBOOK_L3_SOURCED_WRITES, 0x0092); +CPUMF_EVENT_ATTR(cf_zec12, L1D_ONBOOK_L4_SOURCED_WRITES, 0x0093); +CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFBOOK_L4_SOURCED_WRITES, 0x0094); +CPUMF_EVENT_ATTR(cf_zec12, TX_NC_TEND, 0x0095); +CPUMF_EVENT_ATTR(cf_zec12, L1D_ONCHIP_L3_SOURCED_WRITES_IV, 0x0096); +CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFCHIP_L3_SOURCED_WRITES_IV, 0x0097); +CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFBOOK_L3_SOURCED_WRITES_IV, 0x0098); +CPUMF_EVENT_ATTR(cf_zec12, L1I_ONCHIP_L3_SOURCED_WRITES, 0x0099); +CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFCHIP_L3_SOURCED_WRITES, 0x009a); +CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFBOOK_L3_SOURCED_WRITES, 0x009b); +CPUMF_EVENT_ATTR(cf_zec12, L1I_ONBOOK_L4_SOURCED_WRITES, 0x009c); +CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFBOOK_L4_SOURCED_WRITES, 0x009d); +CPUMF_EVENT_ATTR(cf_zec12, TX_C_TEND, 0x009e); +CPUMF_EVENT_ATTR(cf_zec12, L1I_ONCHIP_L3_SOURCED_WRITES_IV, 0x009f); +CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFCHIP_L3_SOURCED_WRITES_IV, 0x00a0); +CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFBOOK_L3_SOURCED_WRITES_IV, 0x00a1); +CPUMF_EVENT_ATTR(cf_zec12, TX_NC_TABORT, 0x00b1); +CPUMF_EVENT_ATTR(cf_zec12, TX_C_TABORT_NO_SPECIAL, 0x00b2); +CPUMF_EVENT_ATTR(cf_zec12, TX_C_TABORT_SPECIAL, 0x00b3); + +static struct attribute *cpumcf_pmu_event_attr[] = { + CPUMF_EVENT_PTR(cf, CPU_CYCLES), + CPUMF_EVENT_PTR(cf, INSTRUCTIONS), + CPUMF_EVENT_PTR(cf, L1I_DIR_WRITES), + CPUMF_EVENT_PTR(cf, L1I_PENALTY_CYCLES), + CPUMF_EVENT_PTR(cf, PROBLEM_STATE_CPU_CYCLES), + CPUMF_EVENT_PTR(cf, PROBLEM_STATE_INSTRUCTIONS), + CPUMF_EVENT_PTR(cf, PROBLEM_STATE_L1I_DIR_WRITES), + CPUMF_EVENT_PTR(cf, PROBLEM_STATE_L1I_PENALTY_CYCLES), + CPUMF_EVENT_PTR(cf, PROBLEM_STATE_L1D_DIR_WRITES), + CPUMF_EVENT_PTR(cf, PROBLEM_STATE_L1D_PENALTY_CYCLES), + CPUMF_EVENT_PTR(cf, L1D_DIR_WRITES), + CPUMF_EVENT_PTR(cf, L1D_PENALTY_CYCLES), + CPUMF_EVENT_PTR(cf, PRNG_FUNCTIONS), + CPUMF_EVENT_PTR(cf, PRNG_CYCLES), + CPUMF_EVENT_PTR(cf, PRNG_BLOCKED_FUNCTIONS), + CPUMF_EVENT_PTR(cf, PRNG_BLOCKED_CYCLES), + CPUMF_EVENT_PTR(cf, SHA_FUNCTIONS), + CPUMF_EVENT_PTR(cf, SHA_CYCLES), + CPUMF_EVENT_PTR(cf, SHA_BLOCKED_FUNCTIONS), + CPUMF_EVENT_PTR(cf, SHA_BLOCKED_CYCLES), + CPUMF_EVENT_PTR(cf, DEA_FUNCTIONS), + CPUMF_EVENT_PTR(cf, DEA_CYCLES), + CPUMF_EVENT_PTR(cf, DEA_BLOCKED_FUNCTIONS), + CPUMF_EVENT_PTR(cf, DEA_BLOCKED_CYCLES), + CPUMF_EVENT_PTR(cf, AES_FUNCTIONS), + CPUMF_EVENT_PTR(cf, AES_CYCLES), + CPUMF_EVENT_PTR(cf, AES_BLOCKED_FUNCTIONS), + CPUMF_EVENT_PTR(cf, AES_BLOCKED_CYCLES), + NULL, +}; + +static struct attribute *cpumcf_z10_pmu_event_attr[] __initdata = { + CPUMF_EVENT_PTR(cf_z10, L1I_L2_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z10, L1D_L2_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z10, L1I_L3_LOCAL_WRITES), + CPUMF_EVENT_PTR(cf_z10, L1D_L3_LOCAL_WRITES), + CPUMF_EVENT_PTR(cf_z10, L1I_L3_REMOTE_WRITES), + CPUMF_EVENT_PTR(cf_z10, L1D_L3_REMOTE_WRITES), + CPUMF_EVENT_PTR(cf_z10, L1D_LMEM_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z10, L1I_LMEM_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z10, L1D_RO_EXCL_WRITES), + CPUMF_EVENT_PTR(cf_z10, L1I_CACHELINE_INVALIDATES), + CPUMF_EVENT_PTR(cf_z10, ITLB1_WRITES), + CPUMF_EVENT_PTR(cf_z10, DTLB1_WRITES), + CPUMF_EVENT_PTR(cf_z10, TLB2_PTE_WRITES), + CPUMF_EVENT_PTR(cf_z10, TLB2_CRSTE_WRITES), + CPUMF_EVENT_PTR(cf_z10, TLB2_CRSTE_HPAGE_WRITES), + CPUMF_EVENT_PTR(cf_z10, ITLB1_MISSES), + CPUMF_EVENT_PTR(cf_z10, DTLB1_MISSES), + CPUMF_EVENT_PTR(cf_z10, L2C_STORES_SENT), + NULL, +}; + +static struct attribute *cpumcf_z196_pmu_event_attr[] __initdata = { + CPUMF_EVENT_PTR(cf_z196, L1D_L2_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1I_L2_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, DTLB1_MISSES), + CPUMF_EVENT_PTR(cf_z196, ITLB1_MISSES), + CPUMF_EVENT_PTR(cf_z196, L2C_STORES_SENT), + CPUMF_EVENT_PTR(cf_z196, L1D_OFFBOOK_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1D_ONBOOK_L4_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1I_ONBOOK_L4_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1D_RO_EXCL_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1D_OFFBOOK_L4_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1I_OFFBOOK_L4_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, DTLB1_HPAGE_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1D_LMEM_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1I_LMEM_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1I_OFFBOOK_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, DTLB1_WRITES), + CPUMF_EVENT_PTR(cf_z196, ITLB1_WRITES), + CPUMF_EVENT_PTR(cf_z196, TLB2_PTE_WRITES), + CPUMF_EVENT_PTR(cf_z196, TLB2_CRSTE_HPAGE_WRITES), + CPUMF_EVENT_PTR(cf_z196, TLB2_CRSTE_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1D_ONCHIP_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1D_OFFCHIP_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1I_ONCHIP_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_z196, L1I_OFFCHIP_L3_SOURCED_WRITES), + NULL, +}; + +static struct attribute *cpumcf_zec12_pmu_event_attr[] __initdata = { + CPUMF_EVENT_PTR(cf_zec12, DTLB1_MISSES), + CPUMF_EVENT_PTR(cf_zec12, ITLB1_MISSES), + CPUMF_EVENT_PTR(cf_zec12, L1D_L2I_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1I_L2I_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1D_L2D_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, DTLB1_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1D_LMEM_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1I_LMEM_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1D_RO_EXCL_WRITES), + CPUMF_EVENT_PTR(cf_zec12, DTLB1_HPAGE_WRITES), + CPUMF_EVENT_PTR(cf_zec12, ITLB1_WRITES), + CPUMF_EVENT_PTR(cf_zec12, TLB2_PTE_WRITES), + CPUMF_EVENT_PTR(cf_zec12, TLB2_CRSTE_HPAGE_WRITES), + CPUMF_EVENT_PTR(cf_zec12, TLB2_CRSTE_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1D_ONCHIP_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1D_OFFCHIP_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1D_OFFBOOK_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1D_ONBOOK_L4_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1D_OFFBOOK_L4_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, TX_NC_TEND), + CPUMF_EVENT_PTR(cf_zec12, L1D_ONCHIP_L3_SOURCED_WRITES_IV), + CPUMF_EVENT_PTR(cf_zec12, L1D_OFFCHIP_L3_SOURCED_WRITES_IV), + CPUMF_EVENT_PTR(cf_zec12, L1D_OFFBOOK_L3_SOURCED_WRITES_IV), + CPUMF_EVENT_PTR(cf_zec12, L1I_ONCHIP_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1I_OFFCHIP_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1I_OFFBOOK_L3_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1I_ONBOOK_L4_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, L1I_OFFBOOK_L4_SOURCED_WRITES), + CPUMF_EVENT_PTR(cf_zec12, TX_C_TEND), + CPUMF_EVENT_PTR(cf_zec12, L1I_ONCHIP_L3_SOURCED_WRITES_IV), + CPUMF_EVENT_PTR(cf_zec12, L1I_OFFCHIP_L3_SOURCED_WRITES_IV), + CPUMF_EVENT_PTR(cf_zec12, L1I_OFFBOOK_L3_SOURCED_WRITES_IV), + CPUMF_EVENT_PTR(cf_zec12, TX_NC_TABORT), + CPUMF_EVENT_PTR(cf_zec12, TX_C_TABORT_NO_SPECIAL), + CPUMF_EVENT_PTR(cf_zec12, TX_C_TABORT_SPECIAL), + NULL, +}; + +/* END: CPUM_CF COUNTER DEFINITIONS ===================================== */ + +static struct attribute_group cpumsf_pmu_events_group = { + .name = "events", + .attrs = cpumcf_pmu_event_attr, +}; + +PMU_FORMAT_ATTR(event, "config:0-63"); + +static struct attribute *cpumsf_pmu_format_attr[] = { + &format_attr_event.attr, + NULL, +}; + +static struct attribute_group cpumsf_pmu_format_group = { + .name = "format", + .attrs = cpumsf_pmu_format_attr, +}; + +static const struct attribute_group *cpumsf_pmu_attr_groups[] = { + &cpumsf_pmu_events_group, + &cpumsf_pmu_format_group, + NULL, +}; + + +static __init struct attribute **merge_attr(struct attribute **a, + struct attribute **b) +{ + struct attribute **new; + int j, i; + + for (j = 0; a[j]; j++) + ; + for (i = 0; b[i]; i++) + j++; + j++; + + new = kmalloc(sizeof(struct attribute *) * j, GFP_KERNEL); + if (!new) + return NULL; + j = 0; + for (i = 0; a[i]; i++) + new[j++] = a[i]; + for (i = 0; b[i]; i++) + new[j++] = b[i]; + new[j] = NULL; + + return new; +} + +__init const struct attribute_group **cpumf_cf_event_group(void) +{ + struct attribute **combined, **model; + struct cpuid cpu_id; + + get_cpu_id(&cpu_id); + switch (cpu_id.machine) { + case 0x2097: + case 0x2098: + model = cpumcf_z10_pmu_event_attr; + break; + case 0x2817: + case 0x2818: + model = cpumcf_z196_pmu_event_attr; + break; + case 0x2827: + case 0x2828: + model = cpumcf_zec12_pmu_event_attr; + break; + default: + model = NULL; + break; + }; + + if (!model) + goto out; + + combined = merge_attr(cpumcf_pmu_event_attr, model); + if (combined) + cpumsf_pmu_events_group.attrs = combined; +out: + return cpumsf_pmu_attr_groups; +} diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c new file mode 100644 index 000000000..e6a1578fc --- /dev/null +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -0,0 +1,1639 @@ +/* + * Performance event support for the System z CPU-measurement Sampling Facility + * + * Copyright IBM Corp. 2013 + * Author(s): Hendrik Brueckner <brueckner@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. + */ +#define KMSG_COMPONENT "cpum_sf" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/kernel_stat.h> +#include <linux/perf_event.h> +#include <linux/percpu.h> +#include <linux/notifier.h> +#include <linux/export.h> +#include <linux/slab.h> +#include <linux/mm.h> +#include <linux/moduleparam.h> +#include <asm/cpu_mf.h> +#include <asm/irq.h> +#include <asm/debug.h> +#include <asm/timex.h> + +/* Minimum number of sample-data-block-tables: + * At least one table is required for the sampling buffer structure. + * A single table contains up to 511 pointers to sample-data-blocks. + */ +#define CPUM_SF_MIN_SDBT 1 + +/* Number of sample-data-blocks per sample-data-block-table (SDBT): + * A table contains SDB pointers (8 bytes) and one table-link entry + * that points to the origin of the next SDBT. + */ +#define CPUM_SF_SDB_PER_TABLE ((PAGE_SIZE - 8) / 8) + +/* Maximum page offset for an SDBT table-link entry: + * If this page offset is reached, a table-link entry to the next SDBT + * must be added. + */ +#define CPUM_SF_SDBT_TL_OFFSET (CPUM_SF_SDB_PER_TABLE * 8) +static inline int require_table_link(const void *sdbt) +{ + return ((unsigned long) sdbt & ~PAGE_MASK) == CPUM_SF_SDBT_TL_OFFSET; +} + +/* Minimum and maximum sampling buffer sizes: + * + * This number represents the maximum size of the sampling buffer taking + * the number of sample-data-block-tables into account. Note that these + * numbers apply to the basic-sampling function only. + * The maximum number of SDBs is increased by CPUM_SF_SDB_DIAG_FACTOR if + * the diagnostic-sampling function is active. + * + * Sampling buffer size Buffer characteristics + * --------------------------------------------------- + * 64KB == 16 pages (4KB per page) + * 1 page for SDB-tables + * 15 pages for SDBs + * + * 32MB == 8192 pages (4KB per page) + * 16 pages for SDB-tables + * 8176 pages for SDBs + */ +static unsigned long __read_mostly CPUM_SF_MIN_SDB = 15; +static unsigned long __read_mostly CPUM_SF_MAX_SDB = 8176; +static unsigned long __read_mostly CPUM_SF_SDB_DIAG_FACTOR = 1; + +struct sf_buffer { + unsigned long *sdbt; /* Sample-data-block-table origin */ + /* buffer characteristics (required for buffer increments) */ + unsigned long num_sdb; /* Number of sample-data-blocks */ + unsigned long num_sdbt; /* Number of sample-data-block-tables */ + unsigned long *tail; /* last sample-data-block-table */ +}; + +struct cpu_hw_sf { + /* CPU-measurement sampling information block */ + struct hws_qsi_info_block qsi; + /* CPU-measurement sampling control block */ + struct hws_lsctl_request_block lsctl; + struct sf_buffer sfb; /* Sampling buffer */ + unsigned int flags; /* Status flags */ + struct perf_event *event; /* Scheduled perf event */ +}; +static DEFINE_PER_CPU(struct cpu_hw_sf, cpu_hw_sf); + +/* Debug feature */ +static debug_info_t *sfdbg; + +/* + * sf_disable() - Switch off sampling facility + */ +static int sf_disable(void) +{ + struct hws_lsctl_request_block sreq; + + memset(&sreq, 0, sizeof(sreq)); + return lsctl(&sreq); +} + +/* + * sf_buffer_available() - Check for an allocated sampling buffer + */ +static int sf_buffer_available(struct cpu_hw_sf *cpuhw) +{ + return !!cpuhw->sfb.sdbt; +} + +/* + * deallocate sampling facility buffer + */ +static void free_sampling_buffer(struct sf_buffer *sfb) +{ + unsigned long *sdbt, *curr; + + if (!sfb->sdbt) + return; + + sdbt = sfb->sdbt; + curr = sdbt; + + /* Free the SDBT after all SDBs are processed... */ + while (1) { + if (!*curr || !sdbt) + break; + + /* Process table-link entries */ + if (is_link_entry(curr)) { + curr = get_next_sdbt(curr); + if (sdbt) + free_page((unsigned long) sdbt); + + /* If the origin is reached, sampling buffer is freed */ + if (curr == sfb->sdbt) + break; + else + sdbt = curr; + } else { + /* Process SDB pointer */ + if (*curr) { + free_page(*curr); + curr++; + } + } + } + + debug_sprintf_event(sfdbg, 5, + "free_sampling_buffer: freed sdbt=%p\n", sfb->sdbt); + memset(sfb, 0, sizeof(*sfb)); +} + +static int alloc_sample_data_block(unsigned long *sdbt, gfp_t gfp_flags) +{ + unsigned long sdb, *trailer; + + /* Allocate and initialize sample-data-block */ + sdb = get_zeroed_page(gfp_flags); + if (!sdb) + return -ENOMEM; + trailer = trailer_entry_ptr(sdb); + *trailer = SDB_TE_ALERT_REQ_MASK; + + /* Link SDB into the sample-data-block-table */ + *sdbt = sdb; + + return 0; +} + +/* + * realloc_sampling_buffer() - extend sampler memory + * + * Allocates new sample-data-blocks and adds them to the specified sampling + * buffer memory. + * + * Important: This modifies the sampling buffer and must be called when the + * sampling facility is disabled. + * + * Returns zero on success, non-zero otherwise. + */ +static int realloc_sampling_buffer(struct sf_buffer *sfb, + unsigned long num_sdb, gfp_t gfp_flags) +{ + int i, rc; + unsigned long *new, *tail; + + if (!sfb->sdbt || !sfb->tail) + return -EINVAL; + + if (!is_link_entry(sfb->tail)) + return -EINVAL; + + /* Append to the existing sampling buffer, overwriting the table-link + * register. + * The tail variables always points to the "tail" (last and table-link) + * entry in an SDB-table. + */ + tail = sfb->tail; + + /* Do a sanity check whether the table-link entry points to + * the sampling buffer origin. + */ + if (sfb->sdbt != get_next_sdbt(tail)) { + debug_sprintf_event(sfdbg, 3, "realloc_sampling_buffer: " + "sampling buffer is not linked: origin=%p" + "tail=%p\n", + (void *) sfb->sdbt, (void *) tail); + return -EINVAL; + } + + /* Allocate remaining SDBs */ + rc = 0; + for (i = 0; i < num_sdb; i++) { + /* Allocate a new SDB-table if it is full. */ + if (require_table_link(tail)) { + new = (unsigned long *) get_zeroed_page(gfp_flags); + if (!new) { + rc = -ENOMEM; + break; + } + sfb->num_sdbt++; + /* Link current page to tail of chain */ + *tail = (unsigned long)(void *) new + 1; + tail = new; + } + + /* Allocate a new sample-data-block. + * If there is not enough memory, stop the realloc process + * and simply use what was allocated. If this is a temporary + * issue, a new realloc call (if required) might succeed. + */ + rc = alloc_sample_data_block(tail, gfp_flags); + if (rc) + break; + sfb->num_sdb++; + tail++; + } + + /* Link sampling buffer to its origin */ + *tail = (unsigned long) sfb->sdbt + 1; + sfb->tail = tail; + + debug_sprintf_event(sfdbg, 4, "realloc_sampling_buffer: new buffer" + " settings: sdbt=%lu sdb=%lu\n", + sfb->num_sdbt, sfb->num_sdb); + return rc; +} + +/* + * allocate_sampling_buffer() - allocate sampler memory + * + * Allocates and initializes a sampling buffer structure using the + * specified number of sample-data-blocks (SDB). For each allocation, + * a 4K page is used. The number of sample-data-block-tables (SDBT) + * are calculated from SDBs. + * Also set the ALERT_REQ mask in each SDBs trailer. + * + * Returns zero on success, non-zero otherwise. + */ +static int alloc_sampling_buffer(struct sf_buffer *sfb, unsigned long num_sdb) +{ + int rc; + + if (sfb->sdbt) + return -EINVAL; + + /* Allocate the sample-data-block-table origin */ + sfb->sdbt = (unsigned long *) get_zeroed_page(GFP_KERNEL); + if (!sfb->sdbt) + return -ENOMEM; + sfb->num_sdb = 0; + sfb->num_sdbt = 1; + + /* Link the table origin to point to itself to prepare for + * realloc_sampling_buffer() invocation. + */ + sfb->tail = sfb->sdbt; + *sfb->tail = (unsigned long)(void *) sfb->sdbt + 1; + + /* Allocate requested number of sample-data-blocks */ + rc = realloc_sampling_buffer(sfb, num_sdb, GFP_KERNEL); + if (rc) { + free_sampling_buffer(sfb); + debug_sprintf_event(sfdbg, 4, "alloc_sampling_buffer: " + "realloc_sampling_buffer failed with rc=%i\n", rc); + } else + debug_sprintf_event(sfdbg, 4, + "alloc_sampling_buffer: tear=%p dear=%p\n", + sfb->sdbt, (void *) *sfb->sdbt); + return rc; +} + +static void sfb_set_limits(unsigned long min, unsigned long max) +{ + struct hws_qsi_info_block si; + + CPUM_SF_MIN_SDB = min; + CPUM_SF_MAX_SDB = max; + + memset(&si, 0, sizeof(si)); + if (!qsi(&si)) + CPUM_SF_SDB_DIAG_FACTOR = DIV_ROUND_UP(si.dsdes, si.bsdes); +} + +static unsigned long sfb_max_limit(struct hw_perf_event *hwc) +{ + return SAMPL_DIAG_MODE(hwc) ? CPUM_SF_MAX_SDB * CPUM_SF_SDB_DIAG_FACTOR + : CPUM_SF_MAX_SDB; +} + +static unsigned long sfb_pending_allocs(struct sf_buffer *sfb, + struct hw_perf_event *hwc) +{ + if (!sfb->sdbt) + return SFB_ALLOC_REG(hwc); + if (SFB_ALLOC_REG(hwc) > sfb->num_sdb) + return SFB_ALLOC_REG(hwc) - sfb->num_sdb; + return 0; +} + +static int sfb_has_pending_allocs(struct sf_buffer *sfb, + struct hw_perf_event *hwc) +{ + return sfb_pending_allocs(sfb, hwc) > 0; +} + +static void sfb_account_allocs(unsigned long num, struct hw_perf_event *hwc) +{ + /* Limit the number of SDBs to not exceed the maximum */ + num = min_t(unsigned long, num, sfb_max_limit(hwc) - SFB_ALLOC_REG(hwc)); + if (num) + SFB_ALLOC_REG(hwc) += num; +} + +static void sfb_init_allocs(unsigned long num, struct hw_perf_event *hwc) +{ + SFB_ALLOC_REG(hwc) = 0; + sfb_account_allocs(num, hwc); +} + +static size_t event_sample_size(struct hw_perf_event *hwc) +{ + struct sf_raw_sample *sfr = (struct sf_raw_sample *) RAWSAMPLE_REG(hwc); + size_t sample_size; + + /* The sample size depends on the sampling function: The basic-sampling + * function must be always enabled, diagnostic-sampling function is + * optional. + */ + sample_size = sfr->bsdes; + if (SAMPL_DIAG_MODE(hwc)) + sample_size += sfr->dsdes; + + return sample_size; +} + +static void deallocate_buffers(struct cpu_hw_sf *cpuhw) +{ + if (cpuhw->sfb.sdbt) + free_sampling_buffer(&cpuhw->sfb); +} + +static int allocate_buffers(struct cpu_hw_sf *cpuhw, struct hw_perf_event *hwc) +{ + unsigned long n_sdb, freq, factor; + size_t sfr_size, sample_size; + struct sf_raw_sample *sfr; + + /* Allocate raw sample buffer + * + * The raw sample buffer is used to temporarily store sampling data + * entries for perf raw sample processing. The buffer size mainly + * depends on the size of diagnostic-sampling data entries which is + * machine-specific. The exact size calculation includes: + * 1. The first 4 bytes of diagnostic-sampling data entries are + * already reflected in the sf_raw_sample structure. Subtract + * these bytes. + * 2. The perf raw sample data must be 8-byte aligned (u64) and + * perf's internal data size must be considered too. So add + * an additional u32 for correct alignment and subtract before + * allocating the buffer. + * 3. Store the raw sample buffer pointer in the perf event + * hardware structure. + */ + sfr_size = ALIGN((sizeof(*sfr) - sizeof(sfr->diag) + cpuhw->qsi.dsdes) + + sizeof(u32), sizeof(u64)); + sfr_size -= sizeof(u32); + sfr = kzalloc(sfr_size, GFP_KERNEL); + if (!sfr) + return -ENOMEM; + sfr->size = sfr_size; + sfr->bsdes = cpuhw->qsi.bsdes; + sfr->dsdes = cpuhw->qsi.dsdes; + RAWSAMPLE_REG(hwc) = (unsigned long) sfr; + + /* Calculate sampling buffers using 4K pages + * + * 1. Determine the sample data size which depends on the used + * sampling functions, for example, basic-sampling or + * basic-sampling with diagnostic-sampling. + * + * 2. Use the sampling frequency as input. The sampling buffer is + * designed for almost one second. This can be adjusted through + * the "factor" variable. + * In any case, alloc_sampling_buffer() sets the Alert Request + * Control indicator to trigger a measurement-alert to harvest + * sample-data-blocks (sdb). + * + * 3. Compute the number of sample-data-blocks and ensure a minimum + * of CPUM_SF_MIN_SDB. Also ensure the upper limit does not + * exceed a "calculated" maximum. The symbolic maximum is + * designed for basic-sampling only and needs to be increased if + * diagnostic-sampling is active. + * See also the remarks for these symbolic constants. + * + * 4. Compute the number of sample-data-block-tables (SDBT) and + * ensure a minimum of CPUM_SF_MIN_SDBT (one table can manage up + * to 511 SDBs). + */ + sample_size = event_sample_size(hwc); + freq = sample_rate_to_freq(&cpuhw->qsi, SAMPL_RATE(hwc)); + factor = 1; + n_sdb = DIV_ROUND_UP(freq, factor * ((PAGE_SIZE-64) / sample_size)); + if (n_sdb < CPUM_SF_MIN_SDB) + n_sdb = CPUM_SF_MIN_SDB; + + /* If there is already a sampling buffer allocated, it is very likely + * that the sampling facility is enabled too. If the event to be + * initialized requires a greater sampling buffer, the allocation must + * be postponed. Changing the sampling buffer requires the sampling + * facility to be in the disabled state. So, account the number of + * required SDBs and let cpumsf_pmu_enable() resize the buffer just + * before the event is started. + */ + sfb_init_allocs(n_sdb, hwc); + if (sf_buffer_available(cpuhw)) + return 0; + + debug_sprintf_event(sfdbg, 3, + "allocate_buffers: rate=%lu f=%lu sdb=%lu/%lu" + " sample_size=%lu cpuhw=%p\n", + SAMPL_RATE(hwc), freq, n_sdb, sfb_max_limit(hwc), + sample_size, cpuhw); + + return alloc_sampling_buffer(&cpuhw->sfb, + sfb_pending_allocs(&cpuhw->sfb, hwc)); +} + +static unsigned long min_percent(unsigned int percent, unsigned long base, + unsigned long min) +{ + return min_t(unsigned long, min, DIV_ROUND_UP(percent * base, 100)); +} + +static unsigned long compute_sfb_extent(unsigned long ratio, unsigned long base) +{ + /* Use a percentage-based approach to extend the sampling facility + * buffer. Accept up to 5% sample data loss. + * Vary the extents between 1% to 5% of the current number of + * sample-data-blocks. + */ + if (ratio <= 5) + return 0; + if (ratio <= 25) + return min_percent(1, base, 1); + if (ratio <= 50) + return min_percent(1, base, 1); + if (ratio <= 75) + return min_percent(2, base, 2); + if (ratio <= 100) + return min_percent(3, base, 3); + if (ratio <= 250) + return min_percent(4, base, 4); + + return min_percent(5, base, 8); +} + +static void sfb_account_overflows(struct cpu_hw_sf *cpuhw, + struct hw_perf_event *hwc) +{ + unsigned long ratio, num; + + if (!OVERFLOW_REG(hwc)) + return; + + /* The sample_overflow contains the average number of sample data + * that has been lost because sample-data-blocks were full. + * + * Calculate the total number of sample data entries that has been + * discarded. Then calculate the ratio of lost samples to total samples + * per second in percent. + */ + ratio = DIV_ROUND_UP(100 * OVERFLOW_REG(hwc) * cpuhw->sfb.num_sdb, + sample_rate_to_freq(&cpuhw->qsi, SAMPL_RATE(hwc))); + + /* Compute number of sample-data-blocks */ + num = compute_sfb_extent(ratio, cpuhw->sfb.num_sdb); + if (num) + sfb_account_allocs(num, hwc); + + debug_sprintf_event(sfdbg, 5, "sfb: overflow: overflow=%llu ratio=%lu" + " num=%lu\n", OVERFLOW_REG(hwc), ratio, num); + OVERFLOW_REG(hwc) = 0; +} + +/* extend_sampling_buffer() - Extend sampling buffer + * @sfb: Sampling buffer structure (for local CPU) + * @hwc: Perf event hardware structure + * + * Use this function to extend the sampling buffer based on the overflow counter + * and postponed allocation extents stored in the specified Perf event hardware. + * + * Important: This function disables the sampling facility in order to safely + * change the sampling buffer structure. Do not call this function + * when the PMU is active. + */ +static void extend_sampling_buffer(struct sf_buffer *sfb, + struct hw_perf_event *hwc) +{ + unsigned long num, num_old; + int rc; + + num = sfb_pending_allocs(sfb, hwc); + if (!num) + return; + num_old = sfb->num_sdb; + + /* Disable the sampling facility to reset any states and also + * clear pending measurement alerts. + */ + sf_disable(); + + /* Extend the sampling buffer. + * This memory allocation typically happens in an atomic context when + * called by perf. Because this is a reallocation, it is fine if the + * new SDB-request cannot be satisfied immediately. + */ + rc = realloc_sampling_buffer(sfb, num, GFP_ATOMIC); + if (rc) + debug_sprintf_event(sfdbg, 5, "sfb: extend: realloc " + "failed with rc=%i\n", rc); + + if (sfb_has_pending_allocs(sfb, hwc)) + debug_sprintf_event(sfdbg, 5, "sfb: extend: " + "req=%lu alloc=%lu remaining=%lu\n", + num, sfb->num_sdb - num_old, + sfb_pending_allocs(sfb, hwc)); +} + + +/* Number of perf events counting hardware events */ +static atomic_t num_events; +/* Used to avoid races in calling reserve/release_cpumf_hardware */ +static DEFINE_MUTEX(pmc_reserve_mutex); + +#define PMC_INIT 0 +#define PMC_RELEASE 1 +#define PMC_FAILURE 2 +static void setup_pmc_cpu(void *flags) +{ + int err; + struct cpu_hw_sf *cpusf = this_cpu_ptr(&cpu_hw_sf); + + err = 0; + switch (*((int *) flags)) { + case PMC_INIT: + memset(cpusf, 0, sizeof(*cpusf)); + err = qsi(&cpusf->qsi); + if (err) + break; + cpusf->flags |= PMU_F_RESERVED; + err = sf_disable(); + if (err) + pr_err("Switching off the sampling facility failed " + "with rc=%i\n", err); + debug_sprintf_event(sfdbg, 5, + "setup_pmc_cpu: initialized: cpuhw=%p\n", cpusf); + break; + case PMC_RELEASE: + cpusf->flags &= ~PMU_F_RESERVED; + err = sf_disable(); + if (err) { + pr_err("Switching off the sampling facility failed " + "with rc=%i\n", err); + } else + deallocate_buffers(cpusf); + debug_sprintf_event(sfdbg, 5, + "setup_pmc_cpu: released: cpuhw=%p\n", cpusf); + break; + } + if (err) + *((int *) flags) |= PMC_FAILURE; +} + +static void release_pmc_hardware(void) +{ + int flags = PMC_RELEASE; + + irq_subclass_unregister(IRQ_SUBCLASS_MEASUREMENT_ALERT); + on_each_cpu(setup_pmc_cpu, &flags, 1); + perf_release_sampling(); +} + +static int reserve_pmc_hardware(void) +{ + int flags = PMC_INIT; + int err; + + err = perf_reserve_sampling(); + if (err) + return err; + on_each_cpu(setup_pmc_cpu, &flags, 1); + if (flags & PMC_FAILURE) { + release_pmc_hardware(); + return -ENODEV; + } + irq_subclass_register(IRQ_SUBCLASS_MEASUREMENT_ALERT); + + return 0; +} + +static void hw_perf_event_destroy(struct perf_event *event) +{ + /* Free raw sample buffer */ + if (RAWSAMPLE_REG(&event->hw)) + kfree((void *) RAWSAMPLE_REG(&event->hw)); + + /* Release PMC if this is the last perf event */ + if (!atomic_add_unless(&num_events, -1, 1)) { + mutex_lock(&pmc_reserve_mutex); + if (atomic_dec_return(&num_events) == 0) + release_pmc_hardware(); + mutex_unlock(&pmc_reserve_mutex); + } +} + +static void hw_init_period(struct hw_perf_event *hwc, u64 period) +{ + hwc->sample_period = period; + hwc->last_period = hwc->sample_period; + local64_set(&hwc->period_left, hwc->sample_period); +} + +static void hw_reset_registers(struct hw_perf_event *hwc, + unsigned long *sdbt_origin) +{ + struct sf_raw_sample *sfr; + + /* (Re)set to first sample-data-block-table */ + TEAR_REG(hwc) = (unsigned long) sdbt_origin; + + /* (Re)set raw sampling buffer register */ + sfr = (struct sf_raw_sample *) RAWSAMPLE_REG(hwc); + memset(&sfr->basic, 0, sizeof(sfr->basic)); + memset(&sfr->diag, 0, sfr->dsdes); +} + +static unsigned long hw_limit_rate(const struct hws_qsi_info_block *si, + unsigned long rate) +{ + return clamp_t(unsigned long, rate, + si->min_sampl_rate, si->max_sampl_rate); +} + +static int __hw_perf_event_init(struct perf_event *event) +{ + struct cpu_hw_sf *cpuhw; + struct hws_qsi_info_block si; + struct perf_event_attr *attr = &event->attr; + struct hw_perf_event *hwc = &event->hw; + unsigned long rate; + int cpu, err; + + /* Reserve CPU-measurement sampling facility */ + err = 0; + if (!atomic_inc_not_zero(&num_events)) { + mutex_lock(&pmc_reserve_mutex); + if (atomic_read(&num_events) == 0 && reserve_pmc_hardware()) + err = -EBUSY; + else + atomic_inc(&num_events); + mutex_unlock(&pmc_reserve_mutex); + } + event->destroy = hw_perf_event_destroy; + + if (err) + goto out; + + /* Access per-CPU sampling information (query sampling info) */ + /* + * The event->cpu value can be -1 to count on every CPU, for example, + * when attaching to a task. If this is specified, use the query + * sampling info from the current CPU, otherwise use event->cpu to + * retrieve the per-CPU information. + * Later, cpuhw indicates whether to allocate sampling buffers for a + * particular CPU (cpuhw!=NULL) or each online CPU (cpuw==NULL). + */ + memset(&si, 0, sizeof(si)); + cpuhw = NULL; + if (event->cpu == -1) + qsi(&si); + else { + /* Event is pinned to a particular CPU, retrieve the per-CPU + * sampling structure for accessing the CPU-specific QSI. + */ + cpuhw = &per_cpu(cpu_hw_sf, event->cpu); + si = cpuhw->qsi; + } + + /* Check sampling facility authorization and, if not authorized, + * fall back to other PMUs. It is safe to check any CPU because + * the authorization is identical for all configured CPUs. + */ + if (!si.as) { + err = -ENOENT; + goto out; + } + + /* Always enable basic sampling */ + SAMPL_FLAGS(hwc) = PERF_CPUM_SF_BASIC_MODE; + + /* Check if diagnostic sampling is requested. Deny if the required + * sampling authorization is missing. + */ + if (attr->config == PERF_EVENT_CPUM_SF_DIAG) { + if (!si.ad) { + err = -EPERM; + goto out; + } + SAMPL_FLAGS(hwc) |= PERF_CPUM_SF_DIAG_MODE; + } + + /* Check and set other sampling flags */ + if (attr->config1 & PERF_CPUM_SF_FULL_BLOCKS) + SAMPL_FLAGS(hwc) |= PERF_CPUM_SF_FULL_BLOCKS; + + /* The sampling information (si) contains information about the + * min/max sampling intervals and the CPU speed. So calculate the + * correct sampling interval and avoid the whole period adjust + * feedback loop. + */ + rate = 0; + if (attr->freq) { + rate = freq_to_sample_rate(&si, attr->sample_freq); + rate = hw_limit_rate(&si, rate); + attr->freq = 0; + attr->sample_period = rate; + } else { + /* The min/max sampling rates specifies the valid range + * of sample periods. If the specified sample period is + * out of range, limit the period to the range boundary. + */ + rate = hw_limit_rate(&si, hwc->sample_period); + + /* The perf core maintains a maximum sample rate that is + * configurable through the sysctl interface. Ensure the + * sampling rate does not exceed this value. This also helps + * to avoid throttling when pushing samples with + * perf_event_overflow(). + */ + if (sample_rate_to_freq(&si, rate) > + sysctl_perf_event_sample_rate) { + err = -EINVAL; + debug_sprintf_event(sfdbg, 1, "Sampling rate exceeds maximum perf sample rate\n"); + goto out; + } + } + SAMPL_RATE(hwc) = rate; + hw_init_period(hwc, SAMPL_RATE(hwc)); + + /* Initialize sample data overflow accounting */ + hwc->extra_reg.reg = REG_OVERFLOW; + OVERFLOW_REG(hwc) = 0; + + /* Allocate the per-CPU sampling buffer using the CPU information + * from the event. If the event is not pinned to a particular + * CPU (event->cpu == -1; or cpuhw == NULL), allocate sampling + * buffers for each online CPU. + */ + if (cpuhw) + /* Event is pinned to a particular CPU */ + err = allocate_buffers(cpuhw, hwc); + else { + /* Event is not pinned, allocate sampling buffer on + * each online CPU + */ + for_each_online_cpu(cpu) { + cpuhw = &per_cpu(cpu_hw_sf, cpu); + err = allocate_buffers(cpuhw, hwc); + if (err) + break; + } + } +out: + return err; +} + +static int cpumsf_pmu_event_init(struct perf_event *event) +{ + int err; + + /* No support for taken branch sampling */ + if (has_branch_stack(event)) + return -EOPNOTSUPP; + + switch (event->attr.type) { + case PERF_TYPE_RAW: + if ((event->attr.config != PERF_EVENT_CPUM_SF) && + (event->attr.config != PERF_EVENT_CPUM_SF_DIAG)) + return -ENOENT; + break; + case PERF_TYPE_HARDWARE: + /* Support sampling of CPU cycles in addition to the + * counter facility. However, the counter facility + * is more precise and, hence, restrict this PMU to + * sampling events only. + */ + if (event->attr.config != PERF_COUNT_HW_CPU_CYCLES) + return -ENOENT; + if (!is_sampling_event(event)) + return -ENOENT; + break; + default: + return -ENOENT; + } + + /* Check online status of the CPU to which the event is pinned */ + if (event->cpu >= nr_cpumask_bits || + (event->cpu >= 0 && !cpu_online(event->cpu))) + return -ENODEV; + + /* Force reset of idle/hv excludes regardless of what the + * user requested. + */ + if (event->attr.exclude_hv) + event->attr.exclude_hv = 0; + if (event->attr.exclude_idle) + event->attr.exclude_idle = 0; + + err = __hw_perf_event_init(event); + if (unlikely(err)) + if (event->destroy) + event->destroy(event); + return err; +} + +static void cpumsf_pmu_enable(struct pmu *pmu) +{ + struct cpu_hw_sf *cpuhw = this_cpu_ptr(&cpu_hw_sf); + struct hw_perf_event *hwc; + int err; + + if (cpuhw->flags & PMU_F_ENABLED) + return; + + if (cpuhw->flags & PMU_F_ERR_MASK) + return; + + /* Check whether to extent the sampling buffer. + * + * Two conditions trigger an increase of the sampling buffer for a + * perf event: + * 1. Postponed buffer allocations from the event initialization. + * 2. Sampling overflows that contribute to pending allocations. + * + * Note that the extend_sampling_buffer() function disables the sampling + * facility, but it can be fully re-enabled using sampling controls that + * have been saved in cpumsf_pmu_disable(). + */ + if (cpuhw->event) { + hwc = &cpuhw->event->hw; + /* Account number of overflow-designated buffer extents */ + sfb_account_overflows(cpuhw, hwc); + if (sfb_has_pending_allocs(&cpuhw->sfb, hwc)) + extend_sampling_buffer(&cpuhw->sfb, hwc); + } + + /* (Re)enable the PMU and sampling facility */ + cpuhw->flags |= PMU_F_ENABLED; + barrier(); + + err = lsctl(&cpuhw->lsctl); + if (err) { + cpuhw->flags &= ~PMU_F_ENABLED; + pr_err("Loading sampling controls failed: op=%i err=%i\n", + 1, err); + return; + } + + debug_sprintf_event(sfdbg, 6, "pmu_enable: es=%i cs=%i ed=%i cd=%i " + "tear=%p dear=%p\n", cpuhw->lsctl.es, cpuhw->lsctl.cs, + cpuhw->lsctl.ed, cpuhw->lsctl.cd, + (void *) cpuhw->lsctl.tear, (void *) cpuhw->lsctl.dear); +} + +static void cpumsf_pmu_disable(struct pmu *pmu) +{ + struct cpu_hw_sf *cpuhw = this_cpu_ptr(&cpu_hw_sf); + struct hws_lsctl_request_block inactive; + struct hws_qsi_info_block si; + int err; + + if (!(cpuhw->flags & PMU_F_ENABLED)) + return; + + if (cpuhw->flags & PMU_F_ERR_MASK) + return; + + /* Switch off sampling activation control */ + inactive = cpuhw->lsctl; + inactive.cs = 0; + inactive.cd = 0; + + err = lsctl(&inactive); + if (err) { + pr_err("Loading sampling controls failed: op=%i err=%i\n", + 2, err); + return; + } + + /* Save state of TEAR and DEAR register contents */ + if (!qsi(&si)) { + /* TEAR/DEAR values are valid only if the sampling facility is + * enabled. Note that cpumsf_pmu_disable() might be called even + * for a disabled sampling facility because cpumsf_pmu_enable() + * controls the enable/disable state. + */ + if (si.es) { + cpuhw->lsctl.tear = si.tear; + cpuhw->lsctl.dear = si.dear; + } + } else + debug_sprintf_event(sfdbg, 3, "cpumsf_pmu_disable: " + "qsi() failed with err=%i\n", err); + + cpuhw->flags &= ~PMU_F_ENABLED; +} + +/* perf_exclude_event() - Filter event + * @event: The perf event + * @regs: pt_regs structure + * @sde_regs: Sample-data-entry (sde) regs structure + * + * Filter perf events according to their exclude specification. + * + * Return non-zero if the event shall be excluded. + */ +static int perf_exclude_event(struct perf_event *event, struct pt_regs *regs, + struct perf_sf_sde_regs *sde_regs) +{ + if (event->attr.exclude_user && user_mode(regs)) + return 1; + if (event->attr.exclude_kernel && !user_mode(regs)) + return 1; + if (event->attr.exclude_guest && sde_regs->in_guest) + return 1; + if (event->attr.exclude_host && !sde_regs->in_guest) + return 1; + return 0; +} + +/* perf_push_sample() - Push samples to perf + * @event: The perf event + * @sample: Hardware sample data + * + * Use the hardware sample data to create perf event sample. The sample + * is the pushed to the event subsystem and the function checks for + * possible event overflows. If an event overflow occurs, the PMU is + * stopped. + * + * Return non-zero if an event overflow occurred. + */ +static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr) +{ + int overflow; + struct pt_regs regs; + struct perf_sf_sde_regs *sde_regs; + struct perf_sample_data data; + struct perf_raw_record raw; + + /* Setup perf sample */ + perf_sample_data_init(&data, 0, event->hw.last_period); + raw.size = sfr->size; + raw.data = sfr; + data.raw = &raw; + + /* Setup pt_regs to look like an CPU-measurement external interrupt + * using the Program Request Alert code. The regs.int_parm_long + * field which is unused contains additional sample-data-entry related + * indicators. + */ + memset(®s, 0, sizeof(regs)); + regs.int_code = 0x1407; + regs.int_parm = CPU_MF_INT_SF_PRA; + sde_regs = (struct perf_sf_sde_regs *) ®s.int_parm_long; + + regs.psw.addr = sfr->basic.ia; + if (sfr->basic.T) + regs.psw.mask |= PSW_MASK_DAT; + if (sfr->basic.W) + regs.psw.mask |= PSW_MASK_WAIT; + if (sfr->basic.P) + regs.psw.mask |= PSW_MASK_PSTATE; + switch (sfr->basic.AS) { + case 0x0: + regs.psw.mask |= PSW_ASC_PRIMARY; + break; + case 0x1: + regs.psw.mask |= PSW_ASC_ACCREG; + break; + case 0x2: + regs.psw.mask |= PSW_ASC_SECONDARY; + break; + case 0x3: + regs.psw.mask |= PSW_ASC_HOME; + break; + } + + /* The host-program-parameter (hpp) contains the sie control + * block that is set by sie64a() in entry64.S. Check if hpp + * refers to a valid control block and set sde_regs flags + * accordingly. This would allow to use hpp values for other + * purposes too. + * For now, simply use a non-zero value as guest indicator. + */ + if (sfr->basic.hpp) + sde_regs->in_guest = 1; + + overflow = 0; + if (perf_exclude_event(event, ®s, sde_regs)) + goto out; + if (perf_event_overflow(event, &data, ®s)) { + overflow = 1; + event->pmu->stop(event, 0); + } + perf_event_update_userpage(event); +out: + return overflow; +} + +static void perf_event_count_update(struct perf_event *event, u64 count) +{ + local64_add(count, &event->count); +} + +static int sample_format_is_valid(struct hws_combined_entry *sample, + unsigned int flags) +{ + if (likely(flags & PERF_CPUM_SF_BASIC_MODE)) + /* Only basic-sampling data entries with data-entry-format + * version of 0x0001 can be processed. + */ + if (sample->basic.def != 0x0001) + return 0; + if (flags & PERF_CPUM_SF_DIAG_MODE) + /* The data-entry-format number of diagnostic-sampling data + * entries can vary. Because diagnostic data is just passed + * through, do only a sanity check on the DEF. + */ + if (sample->diag.def < 0x8001) + return 0; + return 1; +} + +static int sample_is_consistent(struct hws_combined_entry *sample, + unsigned long flags) +{ + /* This check applies only to basic-sampling data entries of potentially + * combined-sampling data entries. Invalid entries cannot be processed + * by the PMU and, thus, do not deliver an associated + * diagnostic-sampling data entry. + */ + if (unlikely(!(flags & PERF_CPUM_SF_BASIC_MODE))) + return 0; + /* + * Samples are skipped, if they are invalid or for which the + * instruction address is not predictable, i.e., the wait-state bit is + * set. + */ + if (sample->basic.I || sample->basic.W) + return 0; + return 1; +} + +static void reset_sample_slot(struct hws_combined_entry *sample, + unsigned long flags) +{ + if (likely(flags & PERF_CPUM_SF_BASIC_MODE)) + sample->basic.def = 0; + if (flags & PERF_CPUM_SF_DIAG_MODE) + sample->diag.def = 0; +} + +static void sfr_store_sample(struct sf_raw_sample *sfr, + struct hws_combined_entry *sample) +{ + if (likely(sfr->format & PERF_CPUM_SF_BASIC_MODE)) + sfr->basic = sample->basic; + if (sfr->format & PERF_CPUM_SF_DIAG_MODE) + memcpy(&sfr->diag, &sample->diag, sfr->dsdes); +} + +static void debug_sample_entry(struct hws_combined_entry *sample, + struct hws_trailer_entry *te, + unsigned long flags) +{ + debug_sprintf_event(sfdbg, 4, "hw_collect_samples: Found unknown " + "sampling data entry: te->f=%i basic.def=%04x (%p)" + " diag.def=%04x (%p)\n", te->f, + sample->basic.def, &sample->basic, + (flags & PERF_CPUM_SF_DIAG_MODE) + ? sample->diag.def : 0xFFFF, + (flags & PERF_CPUM_SF_DIAG_MODE) + ? &sample->diag : NULL); +} + +/* hw_collect_samples() - Walk through a sample-data-block and collect samples + * @event: The perf event + * @sdbt: Sample-data-block table + * @overflow: Event overflow counter + * + * Walks through a sample-data-block and collects sampling data entries that are + * then pushed to the perf event subsystem. Depending on the sampling function, + * there can be either basic-sampling or combined-sampling data entries. A + * combined-sampling data entry consists of a basic- and a diagnostic-sampling + * data entry. The sampling function is determined by the flags in the perf + * event hardware structure. The function always works with a combined-sampling + * data entry but ignores the the diagnostic portion if it is not available. + * + * Note that the implementation focuses on basic-sampling data entries and, if + * such an entry is not valid, the entire combined-sampling data entry is + * ignored. + * + * The overflow variables counts the number of samples that has been discarded + * due to a perf event overflow. + */ +static void hw_collect_samples(struct perf_event *event, unsigned long *sdbt, + unsigned long long *overflow) +{ + unsigned long flags = SAMPL_FLAGS(&event->hw); + struct hws_combined_entry *sample; + struct hws_trailer_entry *te; + struct sf_raw_sample *sfr; + size_t sample_size; + + /* Prepare and initialize raw sample data */ + sfr = (struct sf_raw_sample *) RAWSAMPLE_REG(&event->hw); + sfr->format = flags & PERF_CPUM_SF_MODE_MASK; + + sample_size = event_sample_size(&event->hw); + te = (struct hws_trailer_entry *) trailer_entry_ptr(*sdbt); + sample = (struct hws_combined_entry *) *sdbt; + while ((unsigned long *) sample < (unsigned long *) te) { + /* Check for an empty sample */ + if (!sample->basic.def) + break; + + /* Update perf event period */ + perf_event_count_update(event, SAMPL_RATE(&event->hw)); + + /* Check sampling data entry */ + if (sample_format_is_valid(sample, flags)) { + /* If an event overflow occurred, the PMU is stopped to + * throttle event delivery. Remaining sample data is + * discarded. + */ + if (!*overflow) { + if (sample_is_consistent(sample, flags)) { + /* Deliver sample data to perf */ + sfr_store_sample(sfr, sample); + *overflow = perf_push_sample(event, sfr); + } + } else + /* Count discarded samples */ + *overflow += 1; + } else { + debug_sample_entry(sample, te, flags); + /* Sample slot is not yet written or other record. + * + * This condition can occur if the buffer was reused + * from a combined basic- and diagnostic-sampling. + * If only basic-sampling is then active, entries are + * written into the larger diagnostic entries. + * This is typically the case for sample-data-blocks + * that are not full. Stop processing if the first + * invalid format was detected. + */ + if (!te->f) + break; + } + + /* Reset sample slot and advance to next sample */ + reset_sample_slot(sample, flags); + sample += sample_size; + } +} + +/* hw_perf_event_update() - Process sampling buffer + * @event: The perf event + * @flush_all: Flag to also flush partially filled sample-data-blocks + * + * Processes the sampling buffer and create perf event samples. + * The sampling buffer position are retrieved and saved in the TEAR_REG + * register of the specified perf event. + * + * Only full sample-data-blocks are processed. Specify the flash_all flag + * to also walk through partially filled sample-data-blocks. It is ignored + * if PERF_CPUM_SF_FULL_BLOCKS is set. The PERF_CPUM_SF_FULL_BLOCKS flag + * enforces the processing of full sample-data-blocks only (trailer entries + * with the block-full-indicator bit set). + */ +static void hw_perf_event_update(struct perf_event *event, int flush_all) +{ + struct hw_perf_event *hwc = &event->hw; + struct hws_trailer_entry *te; + unsigned long *sdbt; + unsigned long long event_overflow, sampl_overflow, num_sdb, te_flags; + int done; + + if (flush_all && SDB_FULL_BLOCKS(hwc)) + flush_all = 0; + + sdbt = (unsigned long *) TEAR_REG(hwc); + done = event_overflow = sampl_overflow = num_sdb = 0; + while (!done) { + /* Get the trailer entry of the sample-data-block */ + te = (struct hws_trailer_entry *) trailer_entry_ptr(*sdbt); + + /* Leave loop if no more work to do (block full indicator) */ + if (!te->f) { + done = 1; + if (!flush_all) + break; + } + + /* Check the sample overflow count */ + if (te->overflow) + /* Account sample overflows and, if a particular limit + * is reached, extend the sampling buffer. + * For details, see sfb_account_overflows(). + */ + sampl_overflow += te->overflow; + + /* Timestamps are valid for full sample-data-blocks only */ + debug_sprintf_event(sfdbg, 6, "hw_perf_event_update: sdbt=%p " + "overflow=%llu timestamp=0x%llx\n", + sdbt, te->overflow, + (te->f) ? trailer_timestamp(te) : 0ULL); + + /* Collect all samples from a single sample-data-block and + * flag if an (perf) event overflow happened. If so, the PMU + * is stopped and remaining samples will be discarded. + */ + hw_collect_samples(event, sdbt, &event_overflow); + num_sdb++; + + /* Reset trailer (using compare-double-and-swap) */ + do { + te_flags = te->flags & ~SDB_TE_BUFFER_FULL_MASK; + te_flags |= SDB_TE_ALERT_REQ_MASK; + } while (!cmpxchg_double(&te->flags, &te->overflow, + te->flags, te->overflow, + te_flags, 0ULL)); + + /* Advance to next sample-data-block */ + sdbt++; + if (is_link_entry(sdbt)) + sdbt = get_next_sdbt(sdbt); + + /* Update event hardware registers */ + TEAR_REG(hwc) = (unsigned long) sdbt; + + /* Stop processing sample-data if all samples of the current + * sample-data-block were flushed even if it was not full. + */ + if (flush_all && done) + break; + + /* If an event overflow happened, discard samples by + * processing any remaining sample-data-blocks. + */ + if (event_overflow) + flush_all = 1; + } + + /* Account sample overflows in the event hardware structure */ + if (sampl_overflow) + OVERFLOW_REG(hwc) = DIV_ROUND_UP(OVERFLOW_REG(hwc) + + sampl_overflow, 1 + num_sdb); + if (sampl_overflow || event_overflow) + debug_sprintf_event(sfdbg, 4, "hw_perf_event_update: " + "overflow stats: sample=%llu event=%llu\n", + sampl_overflow, event_overflow); +} + +static void cpumsf_pmu_read(struct perf_event *event) +{ + /* Nothing to do ... updates are interrupt-driven */ +} + +/* Activate sampling control. + * Next call of pmu_enable() starts sampling. + */ +static void cpumsf_pmu_start(struct perf_event *event, int flags) +{ + struct cpu_hw_sf *cpuhw = this_cpu_ptr(&cpu_hw_sf); + + if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED))) + return; + + if (flags & PERF_EF_RELOAD) + WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE)); + + perf_pmu_disable(event->pmu); + event->hw.state = 0; + cpuhw->lsctl.cs = 1; + if (SAMPL_DIAG_MODE(&event->hw)) + cpuhw->lsctl.cd = 1; + perf_pmu_enable(event->pmu); +} + +/* Deactivate sampling control. + * Next call of pmu_enable() stops sampling. + */ +static void cpumsf_pmu_stop(struct perf_event *event, int flags) +{ + struct cpu_hw_sf *cpuhw = this_cpu_ptr(&cpu_hw_sf); + + if (event->hw.state & PERF_HES_STOPPED) + return; + + perf_pmu_disable(event->pmu); + cpuhw->lsctl.cs = 0; + cpuhw->lsctl.cd = 0; + event->hw.state |= PERF_HES_STOPPED; + + if ((flags & PERF_EF_UPDATE) && !(event->hw.state & PERF_HES_UPTODATE)) { + hw_perf_event_update(event, 1); + event->hw.state |= PERF_HES_UPTODATE; + } + perf_pmu_enable(event->pmu); +} + +static int cpumsf_pmu_add(struct perf_event *event, int flags) +{ + struct cpu_hw_sf *cpuhw = this_cpu_ptr(&cpu_hw_sf); + int err; + + if (cpuhw->flags & PMU_F_IN_USE) + return -EAGAIN; + + if (!cpuhw->sfb.sdbt) + return -EINVAL; + + err = 0; + perf_pmu_disable(event->pmu); + + event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED; + + /* Set up sampling controls. Always program the sampling register + * using the SDB-table start. Reset TEAR_REG event hardware register + * that is used by hw_perf_event_update() to store the sampling buffer + * position after samples have been flushed. + */ + cpuhw->lsctl.s = 0; + cpuhw->lsctl.h = 1; + cpuhw->lsctl.tear = (unsigned long) cpuhw->sfb.sdbt; + cpuhw->lsctl.dear = *(unsigned long *) cpuhw->sfb.sdbt; + cpuhw->lsctl.interval = SAMPL_RATE(&event->hw); + hw_reset_registers(&event->hw, cpuhw->sfb.sdbt); + + /* Ensure sampling functions are in the disabled state. If disabled, + * switch on sampling enable control. */ + if (WARN_ON_ONCE(cpuhw->lsctl.es == 1 || cpuhw->lsctl.ed == 1)) { + err = -EAGAIN; + goto out; + } + cpuhw->lsctl.es = 1; + if (SAMPL_DIAG_MODE(&event->hw)) + cpuhw->lsctl.ed = 1; + + /* Set in_use flag and store event */ + cpuhw->event = event; + cpuhw->flags |= PMU_F_IN_USE; + + if (flags & PERF_EF_START) + cpumsf_pmu_start(event, PERF_EF_RELOAD); +out: + perf_event_update_userpage(event); + perf_pmu_enable(event->pmu); + return err; +} + +static void cpumsf_pmu_del(struct perf_event *event, int flags) +{ + struct cpu_hw_sf *cpuhw = this_cpu_ptr(&cpu_hw_sf); + + perf_pmu_disable(event->pmu); + cpumsf_pmu_stop(event, PERF_EF_UPDATE); + + cpuhw->lsctl.es = 0; + cpuhw->lsctl.ed = 0; + cpuhw->flags &= ~PMU_F_IN_USE; + cpuhw->event = NULL; + + perf_event_update_userpage(event); + perf_pmu_enable(event->pmu); +} + +CPUMF_EVENT_ATTR(SF, SF_CYCLES_BASIC, PERF_EVENT_CPUM_SF); +CPUMF_EVENT_ATTR(SF, SF_CYCLES_BASIC_DIAG, PERF_EVENT_CPUM_SF_DIAG); + +static struct attribute *cpumsf_pmu_events_attr[] = { + CPUMF_EVENT_PTR(SF, SF_CYCLES_BASIC), + NULL, + NULL, +}; + +PMU_FORMAT_ATTR(event, "config:0-63"); + +static struct attribute *cpumsf_pmu_format_attr[] = { + &format_attr_event.attr, + NULL, +}; + +static struct attribute_group cpumsf_pmu_events_group = { + .name = "events", + .attrs = cpumsf_pmu_events_attr, +}; +static struct attribute_group cpumsf_pmu_format_group = { + .name = "format", + .attrs = cpumsf_pmu_format_attr, +}; +static const struct attribute_group *cpumsf_pmu_attr_groups[] = { + &cpumsf_pmu_events_group, + &cpumsf_pmu_format_group, + NULL, +}; + +static struct pmu cpumf_sampling = { + .pmu_enable = cpumsf_pmu_enable, + .pmu_disable = cpumsf_pmu_disable, + + .event_init = cpumsf_pmu_event_init, + .add = cpumsf_pmu_add, + .del = cpumsf_pmu_del, + + .start = cpumsf_pmu_start, + .stop = cpumsf_pmu_stop, + .read = cpumsf_pmu_read, + + .attr_groups = cpumsf_pmu_attr_groups, +}; + +static void cpumf_measurement_alert(struct ext_code ext_code, + unsigned int alert, unsigned long unused) +{ + struct cpu_hw_sf *cpuhw; + + if (!(alert & CPU_MF_INT_SF_MASK)) + return; + inc_irq_stat(IRQEXT_CMS); + cpuhw = this_cpu_ptr(&cpu_hw_sf); + + /* Measurement alerts are shared and might happen when the PMU + * is not reserved. Ignore these alerts in this case. */ + if (!(cpuhw->flags & PMU_F_RESERVED)) + return; + + /* The processing below must take care of multiple alert events that + * might be indicated concurrently. */ + + /* Program alert request */ + if (alert & CPU_MF_INT_SF_PRA) { + if (cpuhw->flags & PMU_F_IN_USE) + hw_perf_event_update(cpuhw->event, 0); + else + WARN_ON_ONCE(!(cpuhw->flags & PMU_F_IN_USE)); + } + + /* Report measurement alerts only for non-PRA codes */ + if (alert != CPU_MF_INT_SF_PRA) + debug_sprintf_event(sfdbg, 6, "measurement alert: 0x%x\n", alert); + + /* Sampling authorization change request */ + if (alert & CPU_MF_INT_SF_SACA) + qsi(&cpuhw->qsi); + + /* Loss of sample data due to high-priority machine activities */ + if (alert & CPU_MF_INT_SF_LSDA) { + pr_err("Sample data was lost\n"); + cpuhw->flags |= PMU_F_ERR_LSDA; + sf_disable(); + } + + /* Invalid sampling buffer entry */ + if (alert & (CPU_MF_INT_SF_IAE|CPU_MF_INT_SF_ISE)) { + pr_err("A sampling buffer entry is incorrect (alert=0x%x)\n", + alert); + cpuhw->flags |= PMU_F_ERR_IBE; + sf_disable(); + } +} + +static int cpumf_pmu_notifier(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + unsigned int cpu = (long) hcpu; + int flags; + + /* Ignore the notification if no events are scheduled on the PMU. + * This might be racy... + */ + if (!atomic_read(&num_events)) + return NOTIFY_OK; + + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_ONLINE: + case CPU_ONLINE_FROZEN: + flags = PMC_INIT; + smp_call_function_single(cpu, setup_pmc_cpu, &flags, 1); + break; + case CPU_DOWN_PREPARE: + flags = PMC_RELEASE; + smp_call_function_single(cpu, setup_pmc_cpu, &flags, 1); + break; + default: + break; + } + + return NOTIFY_OK; +} + +static int param_get_sfb_size(char *buffer, const struct kernel_param *kp) +{ + if (!cpum_sf_avail()) + return -ENODEV; + return sprintf(buffer, "%lu,%lu", CPUM_SF_MIN_SDB, CPUM_SF_MAX_SDB); +} + +static int param_set_sfb_size(const char *val, const struct kernel_param *kp) +{ + int rc; + unsigned long min, max; + + if (!cpum_sf_avail()) + return -ENODEV; + if (!val || !strlen(val)) + return -EINVAL; + + /* Valid parameter values: "min,max" or "max" */ + min = CPUM_SF_MIN_SDB; + max = CPUM_SF_MAX_SDB; + if (strchr(val, ',')) + rc = (sscanf(val, "%lu,%lu", &min, &max) == 2) ? 0 : -EINVAL; + else + rc = kstrtoul(val, 10, &max); + + if (min < 2 || min >= max || max > get_num_physpages()) + rc = -EINVAL; + if (rc) + return rc; + + sfb_set_limits(min, max); + pr_info("The sampling buffer limits have changed to: " + "min=%lu max=%lu (diag=x%lu)\n", + CPUM_SF_MIN_SDB, CPUM_SF_MAX_SDB, CPUM_SF_SDB_DIAG_FACTOR); + return 0; +} + +#define param_check_sfb_size(name, p) __param_check(name, p, void) +static struct kernel_param_ops param_ops_sfb_size = { + .set = param_set_sfb_size, + .get = param_get_sfb_size, +}; + +#define RS_INIT_FAILURE_QSI 0x0001 +#define RS_INIT_FAILURE_BSDES 0x0002 +#define RS_INIT_FAILURE_ALRT 0x0003 +#define RS_INIT_FAILURE_PERF 0x0004 +static void __init pr_cpumsf_err(unsigned int reason) +{ + pr_err("Sampling facility support for perf is not available: " + "reason=%04x\n", reason); +} + +static int __init init_cpum_sampling_pmu(void) +{ + struct hws_qsi_info_block si; + int err; + + if (!cpum_sf_avail()) + return -ENODEV; + + memset(&si, 0, sizeof(si)); + if (qsi(&si)) { + pr_cpumsf_err(RS_INIT_FAILURE_QSI); + return -ENODEV; + } + + if (si.bsdes != sizeof(struct hws_basic_entry)) { + pr_cpumsf_err(RS_INIT_FAILURE_BSDES); + return -EINVAL; + } + + if (si.ad) { + sfb_set_limits(CPUM_SF_MIN_SDB, CPUM_SF_MAX_SDB); + cpumsf_pmu_events_attr[1] = + CPUMF_EVENT_PTR(SF, SF_CYCLES_BASIC_DIAG); + } + + sfdbg = debug_register(KMSG_COMPONENT, 2, 1, 80); + if (!sfdbg) + pr_err("Registering for s390dbf failed\n"); + debug_register_view(sfdbg, &debug_sprintf_view); + + err = register_external_irq(EXT_IRQ_MEASURE_ALERT, + cpumf_measurement_alert); + if (err) { + pr_cpumsf_err(RS_INIT_FAILURE_ALRT); + goto out; + } + + err = perf_pmu_register(&cpumf_sampling, "cpum_sf", PERF_TYPE_RAW); + if (err) { + pr_cpumsf_err(RS_INIT_FAILURE_PERF); + unregister_external_irq(EXT_IRQ_MEASURE_ALERT, + cpumf_measurement_alert); + goto out; + } + perf_cpu_notifier(cpumf_pmu_notifier); +out: + return err; +} +arch_initcall(init_cpum_sampling_pmu); +core_param(cpum_sfb_size, CPUM_SF_MAX_SDB, sfb_size, 0640); diff --git a/arch/s390/kernel/perf_event.c b/arch/s390/kernel/perf_event.c new file mode 100644 index 000000000..61595c1f0 --- /dev/null +++ b/arch/s390/kernel/perf_event.c @@ -0,0 +1,324 @@ +/* + * Performance event support for s390x + * + * Copyright IBM Corp. 2012, 2013 + * Author(s): Hendrik Brueckner <brueckner@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. + */ +#define KMSG_COMPONENT "perf" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/perf_event.h> +#include <linux/kvm_host.h> +#include <linux/percpu.h> +#include <linux/export.h> +#include <linux/seq_file.h> +#include <linux/spinlock.h> +#include <linux/sysfs.h> +#include <asm/irq.h> +#include <asm/cpu_mf.h> +#include <asm/lowcore.h> +#include <asm/processor.h> +#include <asm/sysinfo.h> + +const char *perf_pmu_name(void) +{ + if (cpum_cf_avail() || cpum_sf_avail()) + return "CPU-Measurement Facilities (CPU-MF)"; + return "pmu"; +} +EXPORT_SYMBOL(perf_pmu_name); + +int perf_num_counters(void) +{ + int num = 0; + + if (cpum_cf_avail()) + num += PERF_CPUM_CF_MAX_CTR; + if (cpum_sf_avail()) + num += PERF_CPUM_SF_MAX_CTR; + + return num; +} +EXPORT_SYMBOL(perf_num_counters); + +static struct kvm_s390_sie_block *sie_block(struct pt_regs *regs) +{ + struct stack_frame *stack = (struct stack_frame *) regs->gprs[15]; + + if (!stack) + return NULL; + + return (struct kvm_s390_sie_block *) stack->empty1[0]; +} + +static bool is_in_guest(struct pt_regs *regs) +{ + if (user_mode(regs)) + return false; +#if IS_ENABLED(CONFIG_KVM) + return instruction_pointer(regs) == (unsigned long) &sie_exit; +#else + return false; +#endif +} + +static unsigned long guest_is_user_mode(struct pt_regs *regs) +{ + return sie_block(regs)->gpsw.mask & PSW_MASK_PSTATE; +} + +static unsigned long instruction_pointer_guest(struct pt_regs *regs) +{ + return sie_block(regs)->gpsw.addr & PSW_ADDR_INSN; +} + +unsigned long perf_instruction_pointer(struct pt_regs *regs) +{ + return is_in_guest(regs) ? instruction_pointer_guest(regs) + : instruction_pointer(regs); +} + +static unsigned long perf_misc_guest_flags(struct pt_regs *regs) +{ + return guest_is_user_mode(regs) ? PERF_RECORD_MISC_GUEST_USER + : PERF_RECORD_MISC_GUEST_KERNEL; +} + +static unsigned long perf_misc_flags_sf(struct pt_regs *regs) +{ + struct perf_sf_sde_regs *sde_regs; + unsigned long flags; + + sde_regs = (struct perf_sf_sde_regs *) ®s->int_parm_long; + if (sde_regs->in_guest) + flags = user_mode(regs) ? PERF_RECORD_MISC_GUEST_USER + : PERF_RECORD_MISC_GUEST_KERNEL; + else + flags = user_mode(regs) ? PERF_RECORD_MISC_USER + : PERF_RECORD_MISC_KERNEL; + return flags; +} + +unsigned long perf_misc_flags(struct pt_regs *regs) +{ + /* Check if the cpum_sf PMU has created the pt_regs structure. + * In this case, perf misc flags can be easily extracted. Otherwise, + * do regular checks on the pt_regs content. + */ + if (regs->int_code == 0x1407 && regs->int_parm == CPU_MF_INT_SF_PRA) + if (!regs->gprs[15]) + return perf_misc_flags_sf(regs); + + if (is_in_guest(regs)) + return perf_misc_guest_flags(regs); + + return user_mode(regs) ? PERF_RECORD_MISC_USER + : PERF_RECORD_MISC_KERNEL; +} + +static void print_debug_cf(void) +{ + struct cpumf_ctr_info cf_info; + int cpu = smp_processor_id(); + + memset(&cf_info, 0, sizeof(cf_info)); + if (!qctri(&cf_info)) + pr_info("CPU[%i] CPUM_CF: ver=%u.%u A=%04x E=%04x C=%04x\n", + cpu, cf_info.cfvn, cf_info.csvn, + cf_info.auth_ctl, cf_info.enable_ctl, cf_info.act_ctl); +} + +static void print_debug_sf(void) +{ + struct hws_qsi_info_block si; + int cpu = smp_processor_id(); + + memset(&si, 0, sizeof(si)); + if (qsi(&si)) + return; + + pr_info("CPU[%i] CPUM_SF: basic=%i diag=%i min=%lu max=%lu cpu_speed=%u\n", + cpu, si.as, si.ad, si.min_sampl_rate, si.max_sampl_rate, + si.cpu_speed); + + if (si.as) + pr_info("CPU[%i] CPUM_SF: Basic-sampling: a=%i e=%i c=%i" + " bsdes=%i tear=%016lx dear=%016lx\n", cpu, + si.as, si.es, si.cs, si.bsdes, si.tear, si.dear); + if (si.ad) + pr_info("CPU[%i] CPUM_SF: Diagnostic-sampling: a=%i e=%i c=%i" + " dsdes=%i tear=%016lx dear=%016lx\n", cpu, + si.ad, si.ed, si.cd, si.dsdes, si.tear, si.dear); +} + +void perf_event_print_debug(void) +{ + unsigned long flags; + + local_irq_save(flags); + if (cpum_cf_avail()) + print_debug_cf(); + if (cpum_sf_avail()) + print_debug_sf(); + local_irq_restore(flags); +} + +/* Service level infrastructure */ +static void sl_print_counter(struct seq_file *m) +{ + struct cpumf_ctr_info ci; + + memset(&ci, 0, sizeof(ci)); + if (qctri(&ci)) + return; + + seq_printf(m, "CPU-MF: Counter facility: version=%u.%u " + "authorization=%04x\n", ci.cfvn, ci.csvn, ci.auth_ctl); +} + +static void sl_print_sampling(struct seq_file *m) +{ + struct hws_qsi_info_block si; + + memset(&si, 0, sizeof(si)); + if (qsi(&si)) + return; + + if (!si.as && !si.ad) + return; + + seq_printf(m, "CPU-MF: Sampling facility: min_rate=%lu max_rate=%lu" + " cpu_speed=%u\n", si.min_sampl_rate, si.max_sampl_rate, + si.cpu_speed); + if (si.as) + seq_printf(m, "CPU-MF: Sampling facility: mode=basic" + " sample_size=%u\n", si.bsdes); + if (si.ad) + seq_printf(m, "CPU-MF: Sampling facility: mode=diagnostic" + " sample_size=%u\n", si.dsdes); +} + +static void service_level_perf_print(struct seq_file *m, + struct service_level *sl) +{ + if (cpum_cf_avail()) + sl_print_counter(m); + if (cpum_sf_avail()) + sl_print_sampling(m); +} + +static struct service_level service_level_perf = { + .seq_print = service_level_perf_print, +}; + +static int __init service_level_perf_register(void) +{ + return register_service_level(&service_level_perf); +} +arch_initcall(service_level_perf_register); + +/* See also arch/s390/kernel/traps.c */ +static unsigned long __store_trace(struct perf_callchain_entry *entry, + unsigned long sp, + unsigned long low, unsigned long high) +{ + struct stack_frame *sf; + struct pt_regs *regs; + + while (1) { + sp = sp & PSW_ADDR_INSN; + if (sp < low || sp > high - sizeof(*sf)) + return sp; + sf = (struct stack_frame *) sp; + perf_callchain_store(entry, sf->gprs[8] & PSW_ADDR_INSN); + /* Follow the backchain. */ + while (1) { + low = sp; + sp = sf->back_chain & PSW_ADDR_INSN; + if (!sp) + break; + if (sp <= low || sp > high - sizeof(*sf)) + return sp; + sf = (struct stack_frame *) sp; + perf_callchain_store(entry, + sf->gprs[8] & PSW_ADDR_INSN); + } + /* Zero backchain detected, check for interrupt frame. */ + sp = (unsigned long) (sf + 1); + if (sp <= low || sp > high - sizeof(*regs)) + return sp; + regs = (struct pt_regs *) sp; + perf_callchain_store(entry, sf->gprs[8] & PSW_ADDR_INSN); + low = sp; + sp = regs->gprs[15]; + } +} + +void perf_callchain_kernel(struct perf_callchain_entry *entry, + struct pt_regs *regs) +{ + unsigned long head; + struct stack_frame *head_sf; + + if (user_mode(regs)) + return; + + head = regs->gprs[15]; + head_sf = (struct stack_frame *) head; + + if (!head_sf || !head_sf->back_chain) + return; + + head = head_sf->back_chain; + head = __store_trace(entry, head, S390_lowcore.async_stack - ASYNC_SIZE, + S390_lowcore.async_stack); + + __store_trace(entry, head, S390_lowcore.thread_info, + S390_lowcore.thread_info + THREAD_SIZE); +} + +/* Perf defintions for PMU event attributes in sysfs */ +ssize_t cpumf_events_sysfs_show(struct device *dev, + struct device_attribute *attr, char *page) +{ + struct perf_pmu_events_attr *pmu_attr; + + pmu_attr = container_of(attr, struct perf_pmu_events_attr, attr); + return sprintf(page, "event=0x%04llx,name=%s\n", + pmu_attr->id, attr->attr.name); +} + +/* Reserve/release functions for sharing perf hardware */ +static DEFINE_SPINLOCK(perf_hw_owner_lock); +static void *perf_sampling_owner; + +int perf_reserve_sampling(void) +{ + int err; + + err = 0; + spin_lock(&perf_hw_owner_lock); + if (perf_sampling_owner) { + pr_warn("The sampling facility is already reserved by %p\n", + perf_sampling_owner); + err = -EBUSY; + } else + perf_sampling_owner = __builtin_return_address(0); + spin_unlock(&perf_hw_owner_lock); + return err; +} +EXPORT_SYMBOL(perf_reserve_sampling); + +void perf_release_sampling(void) +{ + spin_lock(&perf_hw_owner_lock); + WARN_ON(!perf_sampling_owner); + perf_sampling_owner = NULL; + spin_unlock(&perf_hw_owner_lock); +} +EXPORT_SYMBOL(perf_release_sampling); diff --git a/arch/s390/kernel/pgm_check.S b/arch/s390/kernel/pgm_check.S new file mode 100644 index 000000000..036aa01d0 --- /dev/null +++ b/arch/s390/kernel/pgm_check.S @@ -0,0 +1,146 @@ +/* + * Program check table. + * + * Copyright IBM Corp. 2012 + */ + +#include <linux/linkage.h> + +#define PGM_CHECK(handler) .long handler +#define PGM_CHECK_DEFAULT PGM_CHECK(default_trap_handler) + +/* + * The program check table contains exactly 128 (0x00-0x7f) entries. Each + * line defines the function to be called corresponding to the program check + * interruption code. + */ +.section .rodata, "a" +ENTRY(pgm_check_table) +PGM_CHECK_DEFAULT /* 00 */ +PGM_CHECK(illegal_op) /* 01 */ +PGM_CHECK(privileged_op) /* 02 */ +PGM_CHECK(execute_exception) /* 03 */ +PGM_CHECK(do_protection_exception) /* 04 */ +PGM_CHECK(addressing_exception) /* 05 */ +PGM_CHECK(specification_exception) /* 06 */ +PGM_CHECK(data_exception) /* 07 */ +PGM_CHECK(overflow_exception) /* 08 */ +PGM_CHECK(divide_exception) /* 09 */ +PGM_CHECK(overflow_exception) /* 0a */ +PGM_CHECK(divide_exception) /* 0b */ +PGM_CHECK(hfp_overflow_exception) /* 0c */ +PGM_CHECK(hfp_underflow_exception) /* 0d */ +PGM_CHECK(hfp_significance_exception) /* 0e */ +PGM_CHECK(hfp_divide_exception) /* 0f */ +PGM_CHECK(do_dat_exception) /* 10 */ +PGM_CHECK(do_dat_exception) /* 11 */ +PGM_CHECK(translation_exception) /* 12 */ +PGM_CHECK(special_op_exception) /* 13 */ +PGM_CHECK_DEFAULT /* 14 */ +PGM_CHECK(operand_exception) /* 15 */ +PGM_CHECK_DEFAULT /* 16 */ +PGM_CHECK_DEFAULT /* 17 */ +PGM_CHECK(transaction_exception) /* 18 */ +PGM_CHECK_DEFAULT /* 19 */ +PGM_CHECK_DEFAULT /* 1a */ +PGM_CHECK(vector_exception) /* 1b */ +PGM_CHECK(space_switch_exception) /* 1c */ +PGM_CHECK(hfp_sqrt_exception) /* 1d */ +PGM_CHECK_DEFAULT /* 1e */ +PGM_CHECK_DEFAULT /* 1f */ +PGM_CHECK_DEFAULT /* 20 */ +PGM_CHECK_DEFAULT /* 21 */ +PGM_CHECK_DEFAULT /* 22 */ +PGM_CHECK_DEFAULT /* 23 */ +PGM_CHECK_DEFAULT /* 24 */ +PGM_CHECK_DEFAULT /* 25 */ +PGM_CHECK_DEFAULT /* 26 */ +PGM_CHECK_DEFAULT /* 27 */ +PGM_CHECK_DEFAULT /* 28 */ +PGM_CHECK_DEFAULT /* 29 */ +PGM_CHECK_DEFAULT /* 2a */ +PGM_CHECK_DEFAULT /* 2b */ +PGM_CHECK_DEFAULT /* 2c */ +PGM_CHECK_DEFAULT /* 2d */ +PGM_CHECK_DEFAULT /* 2e */ +PGM_CHECK_DEFAULT /* 2f */ +PGM_CHECK_DEFAULT /* 30 */ +PGM_CHECK_DEFAULT /* 31 */ +PGM_CHECK_DEFAULT /* 32 */ +PGM_CHECK_DEFAULT /* 33 */ +PGM_CHECK_DEFAULT /* 34 */ +PGM_CHECK_DEFAULT /* 35 */ +PGM_CHECK_DEFAULT /* 36 */ +PGM_CHECK_DEFAULT /* 37 */ +PGM_CHECK(do_dat_exception) /* 38 */ +PGM_CHECK(do_dat_exception) /* 39 */ +PGM_CHECK(do_dat_exception) /* 3a */ +PGM_CHECK(do_dat_exception) /* 3b */ +PGM_CHECK_DEFAULT /* 3c */ +PGM_CHECK_DEFAULT /* 3d */ +PGM_CHECK_DEFAULT /* 3e */ +PGM_CHECK_DEFAULT /* 3f */ +PGM_CHECK_DEFAULT /* 40 */ +PGM_CHECK_DEFAULT /* 41 */ +PGM_CHECK_DEFAULT /* 42 */ +PGM_CHECK_DEFAULT /* 43 */ +PGM_CHECK_DEFAULT /* 44 */ +PGM_CHECK_DEFAULT /* 45 */ +PGM_CHECK_DEFAULT /* 46 */ +PGM_CHECK_DEFAULT /* 47 */ +PGM_CHECK_DEFAULT /* 48 */ +PGM_CHECK_DEFAULT /* 49 */ +PGM_CHECK_DEFAULT /* 4a */ +PGM_CHECK_DEFAULT /* 4b */ +PGM_CHECK_DEFAULT /* 4c */ +PGM_CHECK_DEFAULT /* 4d */ +PGM_CHECK_DEFAULT /* 4e */ +PGM_CHECK_DEFAULT /* 4f */ +PGM_CHECK_DEFAULT /* 50 */ +PGM_CHECK_DEFAULT /* 51 */ +PGM_CHECK_DEFAULT /* 52 */ +PGM_CHECK_DEFAULT /* 53 */ +PGM_CHECK_DEFAULT /* 54 */ +PGM_CHECK_DEFAULT /* 55 */ +PGM_CHECK_DEFAULT /* 56 */ +PGM_CHECK_DEFAULT /* 57 */ +PGM_CHECK_DEFAULT /* 58 */ +PGM_CHECK_DEFAULT /* 59 */ +PGM_CHECK_DEFAULT /* 5a */ +PGM_CHECK_DEFAULT /* 5b */ +PGM_CHECK_DEFAULT /* 5c */ +PGM_CHECK_DEFAULT /* 5d */ +PGM_CHECK_DEFAULT /* 5e */ +PGM_CHECK_DEFAULT /* 5f */ +PGM_CHECK_DEFAULT /* 60 */ +PGM_CHECK_DEFAULT /* 61 */ +PGM_CHECK_DEFAULT /* 62 */ +PGM_CHECK_DEFAULT /* 63 */ +PGM_CHECK_DEFAULT /* 64 */ +PGM_CHECK_DEFAULT /* 65 */ +PGM_CHECK_DEFAULT /* 66 */ +PGM_CHECK_DEFAULT /* 67 */ +PGM_CHECK_DEFAULT /* 68 */ +PGM_CHECK_DEFAULT /* 69 */ +PGM_CHECK_DEFAULT /* 6a */ +PGM_CHECK_DEFAULT /* 6b */ +PGM_CHECK_DEFAULT /* 6c */ +PGM_CHECK_DEFAULT /* 6d */ +PGM_CHECK_DEFAULT /* 6e */ +PGM_CHECK_DEFAULT /* 6f */ +PGM_CHECK_DEFAULT /* 70 */ +PGM_CHECK_DEFAULT /* 71 */ +PGM_CHECK_DEFAULT /* 72 */ +PGM_CHECK_DEFAULT /* 73 */ +PGM_CHECK_DEFAULT /* 74 */ +PGM_CHECK_DEFAULT /* 75 */ +PGM_CHECK_DEFAULT /* 76 */ +PGM_CHECK_DEFAULT /* 77 */ +PGM_CHECK_DEFAULT /* 78 */ +PGM_CHECK_DEFAULT /* 79 */ +PGM_CHECK_DEFAULT /* 7a */ +PGM_CHECK_DEFAULT /* 7b */ +PGM_CHECK_DEFAULT /* 7c */ +PGM_CHECK_DEFAULT /* 7d */ +PGM_CHECK_DEFAULT /* 7e */ +PGM_CHECK_DEFAULT /* 7f */ diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c new file mode 100644 index 000000000..dc5edc29b --- /dev/null +++ b/arch/s390/kernel/process.c @@ -0,0 +1,226 @@ +/* + * This file handles the architecture dependent parts of process handling. + * + * Copyright IBM Corp. 1999, 2009 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Hartmut Penner <hp@de.ibm.com>, + * Denis Joseph Barrow, + */ + +#include <linux/compiler.h> +#include <linux/cpu.h> +#include <linux/sched.h> +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/elfcore.h> +#include <linux/smp.h> +#include <linux/slab.h> +#include <linux/interrupt.h> +#include <linux/tick.h> +#include <linux/personality.h> +#include <linux/syscalls.h> +#include <linux/compat.h> +#include <linux/kprobes.h> +#include <linux/random.h> +#include <linux/module.h> +#include <asm/io.h> +#include <asm/processor.h> +#include <asm/vtimer.h> +#include <asm/exec.h> +#include <asm/irq.h> +#include <asm/nmi.h> +#include <asm/smp.h> +#include <asm/switch_to.h> +#include <asm/runtime_instr.h> +#include "entry.h" + +asmlinkage void ret_from_fork(void) asm ("ret_from_fork"); + +/* + * Return saved PC of a blocked thread. used in kernel/sched. + * resume in entry.S does not create a new stack frame, it + * just stores the registers %r6-%r15 to the frame given by + * schedule. We want to return the address of the caller of + * schedule, so we have to walk the backchain one time to + * find the frame schedule() store its return address. + */ +unsigned long thread_saved_pc(struct task_struct *tsk) +{ + struct stack_frame *sf, *low, *high; + + if (!tsk || !task_stack_page(tsk)) + return 0; + low = task_stack_page(tsk); + high = (struct stack_frame *) task_pt_regs(tsk); + sf = (struct stack_frame *) (tsk->thread.ksp & PSW_ADDR_INSN); + if (sf <= low || sf > high) + return 0; + sf = (struct stack_frame *) (sf->back_chain & PSW_ADDR_INSN); + if (sf <= low || sf > high) + return 0; + return sf->gprs[8]; +} + +extern void kernel_thread_starter(void); + +/* + * Free current thread data structures etc.. + */ +void exit_thread(void) +{ + exit_thread_runtime_instr(); +} + +void flush_thread(void) +{ +} + +void release_thread(struct task_struct *dead_task) +{ +} + +void arch_release_task_struct(struct task_struct *tsk) +{ + if (tsk->thread.vxrs) + kfree(tsk->thread.vxrs); +} + +int copy_thread(unsigned long clone_flags, unsigned long new_stackp, + unsigned long arg, struct task_struct *p) +{ + struct thread_info *ti; + struct fake_frame + { + struct stack_frame sf; + struct pt_regs childregs; + } *frame; + + frame = container_of(task_pt_regs(p), struct fake_frame, childregs); + p->thread.ksp = (unsigned long) frame; + /* Save access registers to new thread structure. */ + save_access_regs(&p->thread.acrs[0]); + /* start new process with ar4 pointing to the correct address space */ + p->thread.mm_segment = get_fs(); + /* Don't copy debug registers */ + memset(&p->thread.per_user, 0, sizeof(p->thread.per_user)); + memset(&p->thread.per_event, 0, sizeof(p->thread.per_event)); + clear_tsk_thread_flag(p, TIF_SINGLE_STEP); + /* Initialize per thread user and system timer values */ + ti = task_thread_info(p); + ti->user_timer = 0; + ti->system_timer = 0; + + frame->sf.back_chain = 0; + /* new return point is ret_from_fork */ + frame->sf.gprs[8] = (unsigned long) ret_from_fork; + /* fake return stack for resume(), don't go back to schedule */ + frame->sf.gprs[9] = (unsigned long) frame; + + /* Store access registers to kernel stack of new process. */ + if (unlikely(p->flags & PF_KTHREAD)) { + /* kernel thread */ + memset(&frame->childregs, 0, sizeof(struct pt_regs)); + frame->childregs.psw.mask = PSW_KERNEL_BITS | PSW_MASK_DAT | + PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK; + frame->childregs.psw.addr = PSW_ADDR_AMODE | + (unsigned long) kernel_thread_starter; + frame->childregs.gprs[9] = new_stackp; /* function */ + frame->childregs.gprs[10] = arg; + frame->childregs.gprs[11] = (unsigned long) do_exit; + frame->childregs.orig_gpr2 = -1; + + return 0; + } + frame->childregs = *current_pt_regs(); + frame->childregs.gprs[2] = 0; /* child returns 0 on fork. */ + frame->childregs.flags = 0; + if (new_stackp) + frame->childregs.gprs[15] = new_stackp; + + /* Don't copy runtime instrumentation info */ + p->thread.ri_cb = NULL; + p->thread.ri_signum = 0; + frame->childregs.psw.mask &= ~PSW_MASK_RI; + + /* Save the fpu registers to new thread structure. */ + save_fp_ctl(&p->thread.fp_regs.fpc); + save_fp_regs(p->thread.fp_regs.fprs); + p->thread.fp_regs.pad = 0; + p->thread.vxrs = NULL; + /* Set a new TLS ? */ + if (clone_flags & CLONE_SETTLS) { + unsigned long tls = frame->childregs.gprs[6]; + if (is_compat_task()) { + p->thread.acrs[0] = (unsigned int)tls; + } else { + p->thread.acrs[0] = (unsigned int)(tls >> 32); + p->thread.acrs[1] = (unsigned int)tls; + } + } + return 0; +} + +asmlinkage void execve_tail(void) +{ + current->thread.fp_regs.fpc = 0; + asm volatile("sfpc %0,%0" : : "d" (0)); +} + +/* + * fill in the FPU structure for a core dump. + */ +int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs) +{ + save_fp_ctl(&fpregs->fpc); + save_fp_regs(fpregs->fprs); + return 1; +} +EXPORT_SYMBOL(dump_fpu); + +unsigned long get_wchan(struct task_struct *p) +{ + struct stack_frame *sf, *low, *high; + unsigned long return_address; + int count; + + if (!p || p == current || p->state == TASK_RUNNING || !task_stack_page(p)) + return 0; + low = task_stack_page(p); + high = (struct stack_frame *) task_pt_regs(p); + sf = (struct stack_frame *) (p->thread.ksp & PSW_ADDR_INSN); + if (sf <= low || sf > high) + return 0; + for (count = 0; count < 16; count++) { + sf = (struct stack_frame *) (sf->back_chain & PSW_ADDR_INSN); + if (sf <= low || sf > high) + return 0; + return_address = sf->gprs[8] & PSW_ADDR_INSN; + if (!in_sched_functions(return_address)) + return return_address; + } + return 0; +} + +unsigned long arch_align_stack(unsigned long sp) +{ + if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) + sp -= get_random_int() & ~PAGE_MASK; + return sp & ~0xf; +} + +static inline unsigned long brk_rnd(void) +{ + /* 8MB for 32bit, 1GB for 64bit */ + if (is_32bit_task()) + return (get_random_int() & 0x7ffUL) << PAGE_SHIFT; + else + return (get_random_int() & 0x3ffffUL) << PAGE_SHIFT; +} + +unsigned long arch_randomize_brk(struct mm_struct *mm) +{ + unsigned long ret; + + ret = PAGE_ALIGN(mm->brk + brk_rnd()); + return (ret > mm->brk) ? ret : mm->brk; +} diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c new file mode 100644 index 000000000..dc488e13b --- /dev/null +++ b/arch/s390/kernel/processor.c @@ -0,0 +1,103 @@ +/* + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#define KMSG_COMPONENT "cpu" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/seq_file.h> +#include <linux/delay.h> +#include <linux/cpu.h> +#include <asm/elf.h> +#include <asm/lowcore.h> +#include <asm/param.h> +#include <asm/smp.h> + +static DEFINE_PER_CPU(struct cpuid, cpu_id); + +void notrace cpu_relax(void) +{ + if (!smp_cpu_mtid && MACHINE_HAS_DIAG44) + asm volatile("diag 0,0,0x44"); + barrier(); +} +EXPORT_SYMBOL(cpu_relax); + +/* + * cpu_init - initializes state that is per-CPU. + */ +void cpu_init(void) +{ + struct cpuid *id = this_cpu_ptr(&cpu_id); + + get_cpu_id(id); + atomic_inc(&init_mm.mm_count); + current->active_mm = &init_mm; + BUG_ON(current->mm); + enter_lazy_tlb(&init_mm, current); +} + +/* + * show_cpuinfo - Get information on one CPU for use by procfs. + */ +static int show_cpuinfo(struct seq_file *m, void *v) +{ + static const char *hwcap_str[] = { + "esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp", + "edat", "etf3eh", "highgprs", "te", "vx" + }; + unsigned long n = (unsigned long) v - 1; + int i; + + if (!n) { + s390_adjust_jiffies(); + seq_printf(m, "vendor_id : IBM/S390\n" + "# processors : %i\n" + "bogomips per cpu: %lu.%02lu\n", + num_online_cpus(), loops_per_jiffy/(500000/HZ), + (loops_per_jiffy/(5000/HZ))%100); + seq_puts(m, "features\t: "); + for (i = 0; i < ARRAY_SIZE(hwcap_str); i++) + if (hwcap_str[i] && (elf_hwcap & (1UL << i))) + seq_printf(m, "%s ", hwcap_str[i]); + seq_puts(m, "\n"); + show_cacheinfo(m); + } + get_online_cpus(); + if (cpu_online(n)) { + struct cpuid *id = &per_cpu(cpu_id, n); + seq_printf(m, "processor %li: " + "version = %02X, " + "identification = %06X, " + "machine = %04X\n", + n, id->version, id->ident, id->machine); + } + put_online_cpus(); + return 0; +} + +static void *c_start(struct seq_file *m, loff_t *pos) +{ + return *pos < nr_cpu_ids ? (void *)((unsigned long) *pos + 1) : NULL; +} + +static void *c_next(struct seq_file *m, void *v, loff_t *pos) +{ + ++*pos; + return c_start(m, pos); +} + +static void c_stop(struct seq_file *m, void *v) +{ +} + +const struct seq_operations cpuinfo_op = { + .start = c_start, + .next = c_next, + .stop = c_stop, + .show = show_cpuinfo, +}; + diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c new file mode 100644 index 000000000..d363c9c32 --- /dev/null +++ b/arch/s390/kernel/ptrace.c @@ -0,0 +1,1533 @@ +/* + * Ptrace user space interface. + * + * Copyright IBM Corp. 1999, 2010 + * Author(s): Denis Joseph Barrow + * Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/smp.h> +#include <linux/errno.h> +#include <linux/ptrace.h> +#include <linux/user.h> +#include <linux/security.h> +#include <linux/audit.h> +#include <linux/signal.h> +#include <linux/elf.h> +#include <linux/regset.h> +#include <linux/tracehook.h> +#include <linux/seccomp.h> +#include <linux/compat.h> +#include <trace/syscall.h> +#include <asm/segment.h> +#include <asm/page.h> +#include <asm/pgtable.h> +#include <asm/pgalloc.h> +#include <asm/uaccess.h> +#include <asm/unistd.h> +#include <asm/switch_to.h> +#include "entry.h" + +#ifdef CONFIG_COMPAT +#include "compat_ptrace.h" +#endif + +#define CREATE_TRACE_POINTS +#include <trace/events/syscalls.h> + +void update_cr_regs(struct task_struct *task) +{ + struct pt_regs *regs = task_pt_regs(task); + struct thread_struct *thread = &task->thread; + struct per_regs old, new; + + /* Take care of the enable/disable of transactional execution. */ + if (MACHINE_HAS_TE || MACHINE_HAS_VX) { + unsigned long cr, cr_new; + + __ctl_store(cr, 0, 0); + cr_new = cr; + if (MACHINE_HAS_TE) { + /* Set or clear transaction execution TXC bit 8. */ + cr_new |= (1UL << 55); + if (task->thread.per_flags & PER_FLAG_NO_TE) + cr_new &= ~(1UL << 55); + } + if (MACHINE_HAS_VX) { + /* Enable/disable of vector extension */ + cr_new &= ~(1UL << 17); + if (task->thread.vxrs) + cr_new |= (1UL << 17); + } + if (cr_new != cr) + __ctl_load(cr_new, 0, 0); + if (MACHINE_HAS_TE) { + /* Set/clear transaction execution TDC bits 62/63. */ + __ctl_store(cr, 2, 2); + cr_new = cr & ~3UL; + if (task->thread.per_flags & PER_FLAG_TE_ABORT_RAND) { + if (task->thread.per_flags & + PER_FLAG_TE_ABORT_RAND_TEND) + cr_new |= 1UL; + else + cr_new |= 2UL; + } + if (cr_new != cr) + __ctl_load(cr_new, 2, 2); + } + } + /* Copy user specified PER registers */ + new.control = thread->per_user.control; + new.start = thread->per_user.start; + new.end = thread->per_user.end; + + /* merge TIF_SINGLE_STEP into user specified PER registers. */ + if (test_tsk_thread_flag(task, TIF_SINGLE_STEP) || + test_tsk_thread_flag(task, TIF_UPROBE_SINGLESTEP)) { + if (test_tsk_thread_flag(task, TIF_BLOCK_STEP)) + new.control |= PER_EVENT_BRANCH; + else + new.control |= PER_EVENT_IFETCH; + new.control |= PER_CONTROL_SUSPENSION; + new.control |= PER_EVENT_TRANSACTION_END; + if (test_tsk_thread_flag(task, TIF_UPROBE_SINGLESTEP)) + new.control |= PER_EVENT_IFETCH; + new.start = 0; + new.end = PSW_ADDR_INSN; + } + + /* Take care of the PER enablement bit in the PSW. */ + if (!(new.control & PER_EVENT_MASK)) { + regs->psw.mask &= ~PSW_MASK_PER; + return; + } + regs->psw.mask |= PSW_MASK_PER; + __ctl_store(old, 9, 11); + if (memcmp(&new, &old, sizeof(struct per_regs)) != 0) + __ctl_load(new, 9, 11); +} + +void user_enable_single_step(struct task_struct *task) +{ + clear_tsk_thread_flag(task, TIF_BLOCK_STEP); + set_tsk_thread_flag(task, TIF_SINGLE_STEP); +} + +void user_disable_single_step(struct task_struct *task) +{ + clear_tsk_thread_flag(task, TIF_BLOCK_STEP); + clear_tsk_thread_flag(task, TIF_SINGLE_STEP); +} + +void user_enable_block_step(struct task_struct *task) +{ + set_tsk_thread_flag(task, TIF_SINGLE_STEP); + set_tsk_thread_flag(task, TIF_BLOCK_STEP); +} + +/* + * Called by kernel/ptrace.c when detaching.. + * + * Clear all debugging related fields. + */ +void ptrace_disable(struct task_struct *task) +{ + memset(&task->thread.per_user, 0, sizeof(task->thread.per_user)); + memset(&task->thread.per_event, 0, sizeof(task->thread.per_event)); + clear_tsk_thread_flag(task, TIF_SINGLE_STEP); + clear_pt_regs_flag(task_pt_regs(task), PIF_PER_TRAP); + task->thread.per_flags = 0; +} + +#define __ADDR_MASK 7 + +static inline unsigned long __peek_user_per(struct task_struct *child, + addr_t addr) +{ + struct per_struct_kernel *dummy = NULL; + + if (addr == (addr_t) &dummy->cr9) + /* Control bits of the active per set. */ + return test_thread_flag(TIF_SINGLE_STEP) ? + PER_EVENT_IFETCH : child->thread.per_user.control; + else if (addr == (addr_t) &dummy->cr10) + /* Start address of the active per set. */ + return test_thread_flag(TIF_SINGLE_STEP) ? + 0 : child->thread.per_user.start; + else if (addr == (addr_t) &dummy->cr11) + /* End address of the active per set. */ + return test_thread_flag(TIF_SINGLE_STEP) ? + PSW_ADDR_INSN : child->thread.per_user.end; + else if (addr == (addr_t) &dummy->bits) + /* Single-step bit. */ + return test_thread_flag(TIF_SINGLE_STEP) ? + (1UL << (BITS_PER_LONG - 1)) : 0; + else if (addr == (addr_t) &dummy->starting_addr) + /* Start address of the user specified per set. */ + return child->thread.per_user.start; + else if (addr == (addr_t) &dummy->ending_addr) + /* End address of the user specified per set. */ + return child->thread.per_user.end; + else if (addr == (addr_t) &dummy->perc_atmid) + /* PER code, ATMID and AI of the last PER trap */ + return (unsigned long) + child->thread.per_event.cause << (BITS_PER_LONG - 16); + else if (addr == (addr_t) &dummy->address) + /* Address of the last PER trap */ + return child->thread.per_event.address; + else if (addr == (addr_t) &dummy->access_id) + /* Access id of the last PER trap */ + return (unsigned long) + child->thread.per_event.paid << (BITS_PER_LONG - 8); + return 0; +} + +/* + * Read the word at offset addr from the user area of a process. The + * trouble here is that the information is littered over different + * locations. The process registers are found on the kernel stack, + * the floating point stuff and the trace settings are stored in + * the task structure. In addition the different structures in + * struct user contain pad bytes that should be read as zeroes. + * Lovely... + */ +static unsigned long __peek_user(struct task_struct *child, addr_t addr) +{ + struct user *dummy = NULL; + addr_t offset, tmp; + + if (addr < (addr_t) &dummy->regs.acrs) { + /* + * psw and gprs are stored on the stack + */ + tmp = *(addr_t *)((addr_t) &task_pt_regs(child)->psw + addr); + if (addr == (addr_t) &dummy->regs.psw.mask) { + /* Return a clean psw mask. */ + tmp &= PSW_MASK_USER | PSW_MASK_RI; + tmp |= PSW_USER_BITS; + } + + } else if (addr < (addr_t) &dummy->regs.orig_gpr2) { + /* + * access registers are stored in the thread structure + */ + offset = addr - (addr_t) &dummy->regs.acrs; + /* + * Very special case: old & broken 64 bit gdb reading + * from acrs[15]. Result is a 64 bit value. Read the + * 32 bit acrs[15] value and shift it by 32. Sick... + */ + if (addr == (addr_t) &dummy->regs.acrs[15]) + tmp = ((unsigned long) child->thread.acrs[15]) << 32; + else + tmp = *(addr_t *)((addr_t) &child->thread.acrs + offset); + + } else if (addr == (addr_t) &dummy->regs.orig_gpr2) { + /* + * orig_gpr2 is stored on the kernel stack + */ + tmp = (addr_t) task_pt_regs(child)->orig_gpr2; + + } else if (addr < (addr_t) &dummy->regs.fp_regs) { + /* + * prevent reads of padding hole between + * orig_gpr2 and fp_regs on s390. + */ + tmp = 0; + + } else if (addr == (addr_t) &dummy->regs.fp_regs.fpc) { + /* + * floating point control reg. is in the thread structure + */ + tmp = child->thread.fp_regs.fpc; + tmp <<= BITS_PER_LONG - 32; + + } else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) { + /* + * floating point regs. are either in child->thread.fp_regs + * or the child->thread.vxrs array + */ + offset = addr - (addr_t) &dummy->regs.fp_regs.fprs; + if (child->thread.vxrs) + tmp = *(addr_t *) + ((addr_t) child->thread.vxrs + 2*offset); + else + tmp = *(addr_t *) + ((addr_t) &child->thread.fp_regs.fprs + offset); + + } else if (addr < (addr_t) (&dummy->regs.per_info + 1)) { + /* + * Handle access to the per_info structure. + */ + addr -= (addr_t) &dummy->regs.per_info; + tmp = __peek_user_per(child, addr); + + } else + tmp = 0; + + return tmp; +} + +static int +peek_user(struct task_struct *child, addr_t addr, addr_t data) +{ + addr_t tmp, mask; + + /* + * Stupid gdb peeks/pokes the access registers in 64 bit with + * an alignment of 4. Programmers from hell... + */ + mask = __ADDR_MASK; + if (addr >= (addr_t) &((struct user *) NULL)->regs.acrs && + addr < (addr_t) &((struct user *) NULL)->regs.orig_gpr2) + mask = 3; + if ((addr & mask) || addr > sizeof(struct user) - __ADDR_MASK) + return -EIO; + + tmp = __peek_user(child, addr); + return put_user(tmp, (addr_t __user *) data); +} + +static inline void __poke_user_per(struct task_struct *child, + addr_t addr, addr_t data) +{ + struct per_struct_kernel *dummy = NULL; + + /* + * There are only three fields in the per_info struct that the + * debugger user can write to. + * 1) cr9: the debugger wants to set a new PER event mask + * 2) starting_addr: the debugger wants to set a new starting + * address to use with the PER event mask. + * 3) ending_addr: the debugger wants to set a new ending + * address to use with the PER event mask. + * The user specified PER event mask and the start and end + * addresses are used only if single stepping is not in effect. + * Writes to any other field in per_info are ignored. + */ + if (addr == (addr_t) &dummy->cr9) + /* PER event mask of the user specified per set. */ + child->thread.per_user.control = + data & (PER_EVENT_MASK | PER_CONTROL_MASK); + else if (addr == (addr_t) &dummy->starting_addr) + /* Starting address of the user specified per set. */ + child->thread.per_user.start = data; + else if (addr == (addr_t) &dummy->ending_addr) + /* Ending address of the user specified per set. */ + child->thread.per_user.end = data; +} + +/* + * Write a word to the user area of a process at location addr. This + * operation does have an additional problem compared to peek_user. + * Stores to the program status word and on the floating point + * control register needs to get checked for validity. + */ +static int __poke_user(struct task_struct *child, addr_t addr, addr_t data) +{ + struct user *dummy = NULL; + addr_t offset; + + if (addr < (addr_t) &dummy->regs.acrs) { + /* + * psw and gprs are stored on the stack + */ + if (addr == (addr_t) &dummy->regs.psw.mask) { + unsigned long mask = PSW_MASK_USER; + + mask |= is_ri_task(child) ? PSW_MASK_RI : 0; + if ((data ^ PSW_USER_BITS) & ~mask) + /* Invalid psw mask. */ + return -EINVAL; + if ((data & PSW_MASK_ASC) == PSW_ASC_HOME) + /* Invalid address-space-control bits */ + return -EINVAL; + if ((data & PSW_MASK_EA) && !(data & PSW_MASK_BA)) + /* Invalid addressing mode bits */ + return -EINVAL; + } + *(addr_t *)((addr_t) &task_pt_regs(child)->psw + addr) = data; + + } else if (addr < (addr_t) (&dummy->regs.orig_gpr2)) { + /* + * access registers are stored in the thread structure + */ + offset = addr - (addr_t) &dummy->regs.acrs; + /* + * Very special case: old & broken 64 bit gdb writing + * to acrs[15] with a 64 bit value. Ignore the lower + * half of the value and write the upper 32 bit to + * acrs[15]. Sick... + */ + if (addr == (addr_t) &dummy->regs.acrs[15]) + child->thread.acrs[15] = (unsigned int) (data >> 32); + else + *(addr_t *)((addr_t) &child->thread.acrs + offset) = data; + + } else if (addr == (addr_t) &dummy->regs.orig_gpr2) { + /* + * orig_gpr2 is stored on the kernel stack + */ + task_pt_regs(child)->orig_gpr2 = data; + + } else if (addr < (addr_t) &dummy->regs.fp_regs) { + /* + * prevent writes of padding hole between + * orig_gpr2 and fp_regs on s390. + */ + return 0; + + } else if (addr == (addr_t) &dummy->regs.fp_regs.fpc) { + /* + * floating point control reg. is in the thread structure + */ + if ((unsigned int) data != 0 || + test_fp_ctl(data >> (BITS_PER_LONG - 32))) + return -EINVAL; + child->thread.fp_regs.fpc = data >> (BITS_PER_LONG - 32); + + } else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) { + /* + * floating point regs. are either in child->thread.fp_regs + * or the child->thread.vxrs array + */ + offset = addr - (addr_t) &dummy->regs.fp_regs.fprs; + if (child->thread.vxrs) + *(addr_t *)((addr_t) + child->thread.vxrs + 2*offset) = data; + else + *(addr_t *)((addr_t) + &child->thread.fp_regs.fprs + offset) = data; + + } else if (addr < (addr_t) (&dummy->regs.per_info + 1)) { + /* + * Handle access to the per_info structure. + */ + addr -= (addr_t) &dummy->regs.per_info; + __poke_user_per(child, addr, data); + + } + + return 0; +} + +static int poke_user(struct task_struct *child, addr_t addr, addr_t data) +{ + addr_t mask; + + /* + * Stupid gdb peeks/pokes the access registers in 64 bit with + * an alignment of 4. Programmers from hell indeed... + */ + mask = __ADDR_MASK; + if (addr >= (addr_t) &((struct user *) NULL)->regs.acrs && + addr < (addr_t) &((struct user *) NULL)->regs.orig_gpr2) + mask = 3; + if ((addr & mask) || addr > sizeof(struct user) - __ADDR_MASK) + return -EIO; + + return __poke_user(child, addr, data); +} + +long arch_ptrace(struct task_struct *child, long request, + unsigned long addr, unsigned long data) +{ + ptrace_area parea; + int copied, ret; + + switch (request) { + case PTRACE_PEEKUSR: + /* read the word at location addr in the USER area. */ + return peek_user(child, addr, data); + + case PTRACE_POKEUSR: + /* write the word at location addr in the USER area */ + return poke_user(child, addr, data); + + case PTRACE_PEEKUSR_AREA: + case PTRACE_POKEUSR_AREA: + if (copy_from_user(&parea, (void __force __user *) addr, + sizeof(parea))) + return -EFAULT; + addr = parea.kernel_addr; + data = parea.process_addr; + copied = 0; + while (copied < parea.len) { + if (request == PTRACE_PEEKUSR_AREA) + ret = peek_user(child, addr, data); + else { + addr_t utmp; + if (get_user(utmp, + (addr_t __force __user *) data)) + return -EFAULT; + ret = poke_user(child, addr, utmp); + } + if (ret) + return ret; + addr += sizeof(unsigned long); + data += sizeof(unsigned long); + copied += sizeof(unsigned long); + } + return 0; + case PTRACE_GET_LAST_BREAK: + put_user(task_thread_info(child)->last_break, + (unsigned long __user *) data); + return 0; + case PTRACE_ENABLE_TE: + if (!MACHINE_HAS_TE) + return -EIO; + child->thread.per_flags &= ~PER_FLAG_NO_TE; + return 0; + case PTRACE_DISABLE_TE: + if (!MACHINE_HAS_TE) + return -EIO; + child->thread.per_flags |= PER_FLAG_NO_TE; + child->thread.per_flags &= ~PER_FLAG_TE_ABORT_RAND; + return 0; + case PTRACE_TE_ABORT_RAND: + if (!MACHINE_HAS_TE || (child->thread.per_flags & PER_FLAG_NO_TE)) + return -EIO; + switch (data) { + case 0UL: + child->thread.per_flags &= ~PER_FLAG_TE_ABORT_RAND; + break; + case 1UL: + child->thread.per_flags |= PER_FLAG_TE_ABORT_RAND; + child->thread.per_flags |= PER_FLAG_TE_ABORT_RAND_TEND; + break; + case 2UL: + child->thread.per_flags |= PER_FLAG_TE_ABORT_RAND; + child->thread.per_flags &= ~PER_FLAG_TE_ABORT_RAND_TEND; + break; + default: + return -EINVAL; + } + return 0; + default: + /* Removing high order bit from addr (only for 31 bit). */ + addr &= PSW_ADDR_INSN; + return ptrace_request(child, request, addr, data); + } +} + +#ifdef CONFIG_COMPAT +/* + * Now the fun part starts... a 31 bit program running in the + * 31 bit emulation tracing another program. PTRACE_PEEKTEXT, + * PTRACE_PEEKDATA, PTRACE_POKETEXT and PTRACE_POKEDATA are easy + * to handle, the difference to the 64 bit versions of the requests + * is that the access is done in multiples of 4 byte instead of + * 8 bytes (sizeof(unsigned long) on 31/64 bit). + * The ugly part are PTRACE_PEEKUSR, PTRACE_PEEKUSR_AREA, + * PTRACE_POKEUSR and PTRACE_POKEUSR_AREA. If the traced program + * is a 31 bit program too, the content of struct user can be + * emulated. A 31 bit program peeking into the struct user of + * a 64 bit program is a no-no. + */ + +/* + * Same as peek_user_per but for a 31 bit program. + */ +static inline __u32 __peek_user_per_compat(struct task_struct *child, + addr_t addr) +{ + struct compat_per_struct_kernel *dummy32 = NULL; + + if (addr == (addr_t) &dummy32->cr9) + /* Control bits of the active per set. */ + return (__u32) test_thread_flag(TIF_SINGLE_STEP) ? + PER_EVENT_IFETCH : child->thread.per_user.control; + else if (addr == (addr_t) &dummy32->cr10) + /* Start address of the active per set. */ + return (__u32) test_thread_flag(TIF_SINGLE_STEP) ? + 0 : child->thread.per_user.start; + else if (addr == (addr_t) &dummy32->cr11) + /* End address of the active per set. */ + return test_thread_flag(TIF_SINGLE_STEP) ? + PSW32_ADDR_INSN : child->thread.per_user.end; + else if (addr == (addr_t) &dummy32->bits) + /* Single-step bit. */ + return (__u32) test_thread_flag(TIF_SINGLE_STEP) ? + 0x80000000 : 0; + else if (addr == (addr_t) &dummy32->starting_addr) + /* Start address of the user specified per set. */ + return (__u32) child->thread.per_user.start; + else if (addr == (addr_t) &dummy32->ending_addr) + /* End address of the user specified per set. */ + return (__u32) child->thread.per_user.end; + else if (addr == (addr_t) &dummy32->perc_atmid) + /* PER code, ATMID and AI of the last PER trap */ + return (__u32) child->thread.per_event.cause << 16; + else if (addr == (addr_t) &dummy32->address) + /* Address of the last PER trap */ + return (__u32) child->thread.per_event.address; + else if (addr == (addr_t) &dummy32->access_id) + /* Access id of the last PER trap */ + return (__u32) child->thread.per_event.paid << 24; + return 0; +} + +/* + * Same as peek_user but for a 31 bit program. + */ +static u32 __peek_user_compat(struct task_struct *child, addr_t addr) +{ + struct compat_user *dummy32 = NULL; + addr_t offset; + __u32 tmp; + + if (addr < (addr_t) &dummy32->regs.acrs) { + struct pt_regs *regs = task_pt_regs(child); + /* + * psw and gprs are stored on the stack + */ + if (addr == (addr_t) &dummy32->regs.psw.mask) { + /* Fake a 31 bit psw mask. */ + tmp = (__u32)(regs->psw.mask >> 32); + tmp &= PSW32_MASK_USER | PSW32_MASK_RI; + tmp |= PSW32_USER_BITS; + } else if (addr == (addr_t) &dummy32->regs.psw.addr) { + /* Fake a 31 bit psw address. */ + tmp = (__u32) regs->psw.addr | + (__u32)(regs->psw.mask & PSW_MASK_BA); + } else { + /* gpr 0-15 */ + tmp = *(__u32 *)((addr_t) ®s->psw + addr*2 + 4); + } + } else if (addr < (addr_t) (&dummy32->regs.orig_gpr2)) { + /* + * access registers are stored in the thread structure + */ + offset = addr - (addr_t) &dummy32->regs.acrs; + tmp = *(__u32*)((addr_t) &child->thread.acrs + offset); + + } else if (addr == (addr_t) (&dummy32->regs.orig_gpr2)) { + /* + * orig_gpr2 is stored on the kernel stack + */ + tmp = *(__u32*)((addr_t) &task_pt_regs(child)->orig_gpr2 + 4); + + } else if (addr < (addr_t) &dummy32->regs.fp_regs) { + /* + * prevent reads of padding hole between + * orig_gpr2 and fp_regs on s390. + */ + tmp = 0; + + } else if (addr == (addr_t) &dummy32->regs.fp_regs.fpc) { + /* + * floating point control reg. is in the thread structure + */ + tmp = child->thread.fp_regs.fpc; + + } else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) { + /* + * floating point regs. are either in child->thread.fp_regs + * or the child->thread.vxrs array + */ + offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs; + if (child->thread.vxrs) + tmp = *(__u32 *) + ((addr_t) child->thread.vxrs + 2*offset); + else + tmp = *(__u32 *) + ((addr_t) &child->thread.fp_regs.fprs + offset); + + } else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) { + /* + * Handle access to the per_info structure. + */ + addr -= (addr_t) &dummy32->regs.per_info; + tmp = __peek_user_per_compat(child, addr); + + } else + tmp = 0; + + return tmp; +} + +static int peek_user_compat(struct task_struct *child, + addr_t addr, addr_t data) +{ + __u32 tmp; + + if (!is_compat_task() || (addr & 3) || addr > sizeof(struct user) - 3) + return -EIO; + + tmp = __peek_user_compat(child, addr); + return put_user(tmp, (__u32 __user *) data); +} + +/* + * Same as poke_user_per but for a 31 bit program. + */ +static inline void __poke_user_per_compat(struct task_struct *child, + addr_t addr, __u32 data) +{ + struct compat_per_struct_kernel *dummy32 = NULL; + + if (addr == (addr_t) &dummy32->cr9) + /* PER event mask of the user specified per set. */ + child->thread.per_user.control = + data & (PER_EVENT_MASK | PER_CONTROL_MASK); + else if (addr == (addr_t) &dummy32->starting_addr) + /* Starting address of the user specified per set. */ + child->thread.per_user.start = data; + else if (addr == (addr_t) &dummy32->ending_addr) + /* Ending address of the user specified per set. */ + child->thread.per_user.end = data; +} + +/* + * Same as poke_user but for a 31 bit program. + */ +static int __poke_user_compat(struct task_struct *child, + addr_t addr, addr_t data) +{ + struct compat_user *dummy32 = NULL; + __u32 tmp = (__u32) data; + addr_t offset; + + if (addr < (addr_t) &dummy32->regs.acrs) { + struct pt_regs *regs = task_pt_regs(child); + /* + * psw, gprs, acrs and orig_gpr2 are stored on the stack + */ + if (addr == (addr_t) &dummy32->regs.psw.mask) { + __u32 mask = PSW32_MASK_USER; + + mask |= is_ri_task(child) ? PSW32_MASK_RI : 0; + /* Build a 64 bit psw mask from 31 bit mask. */ + if ((tmp ^ PSW32_USER_BITS) & ~mask) + /* Invalid psw mask. */ + return -EINVAL; + if ((data & PSW32_MASK_ASC) == PSW32_ASC_HOME) + /* Invalid address-space-control bits */ + return -EINVAL; + regs->psw.mask = (regs->psw.mask & ~PSW_MASK_USER) | + (regs->psw.mask & PSW_MASK_BA) | + (__u64)(tmp & mask) << 32; + } else if (addr == (addr_t) &dummy32->regs.psw.addr) { + /* Build a 64 bit psw address from 31 bit address. */ + regs->psw.addr = (__u64) tmp & PSW32_ADDR_INSN; + /* Transfer 31 bit amode bit to psw mask. */ + regs->psw.mask = (regs->psw.mask & ~PSW_MASK_BA) | + (__u64)(tmp & PSW32_ADDR_AMODE); + } else { + /* gpr 0-15 */ + *(__u32*)((addr_t) ®s->psw + addr*2 + 4) = tmp; + } + } else if (addr < (addr_t) (&dummy32->regs.orig_gpr2)) { + /* + * access registers are stored in the thread structure + */ + offset = addr - (addr_t) &dummy32->regs.acrs; + *(__u32*)((addr_t) &child->thread.acrs + offset) = tmp; + + } else if (addr == (addr_t) (&dummy32->regs.orig_gpr2)) { + /* + * orig_gpr2 is stored on the kernel stack + */ + *(__u32*)((addr_t) &task_pt_regs(child)->orig_gpr2 + 4) = tmp; + + } else if (addr < (addr_t) &dummy32->regs.fp_regs) { + /* + * prevent writess of padding hole between + * orig_gpr2 and fp_regs on s390. + */ + return 0; + + } else if (addr == (addr_t) &dummy32->regs.fp_regs.fpc) { + /* + * floating point control reg. is in the thread structure + */ + if (test_fp_ctl(tmp)) + return -EINVAL; + child->thread.fp_regs.fpc = data; + + } else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) { + /* + * floating point regs. are either in child->thread.fp_regs + * or the child->thread.vxrs array + */ + offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs; + if (child->thread.vxrs) + *(__u32 *)((addr_t) + child->thread.vxrs + 2*offset) = tmp; + else + *(__u32 *)((addr_t) + &child->thread.fp_regs.fprs + offset) = tmp; + + } else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) { + /* + * Handle access to the per_info structure. + */ + addr -= (addr_t) &dummy32->regs.per_info; + __poke_user_per_compat(child, addr, data); + } + + return 0; +} + +static int poke_user_compat(struct task_struct *child, + addr_t addr, addr_t data) +{ + if (!is_compat_task() || (addr & 3) || + addr > sizeof(struct compat_user) - 3) + return -EIO; + + return __poke_user_compat(child, addr, data); +} + +long compat_arch_ptrace(struct task_struct *child, compat_long_t request, + compat_ulong_t caddr, compat_ulong_t cdata) +{ + unsigned long addr = caddr; + unsigned long data = cdata; + compat_ptrace_area parea; + int copied, ret; + + switch (request) { + case PTRACE_PEEKUSR: + /* read the word at location addr in the USER area. */ + return peek_user_compat(child, addr, data); + + case PTRACE_POKEUSR: + /* write the word at location addr in the USER area */ + return poke_user_compat(child, addr, data); + + case PTRACE_PEEKUSR_AREA: + case PTRACE_POKEUSR_AREA: + if (copy_from_user(&parea, (void __force __user *) addr, + sizeof(parea))) + return -EFAULT; + addr = parea.kernel_addr; + data = parea.process_addr; + copied = 0; + while (copied < parea.len) { + if (request == PTRACE_PEEKUSR_AREA) + ret = peek_user_compat(child, addr, data); + else { + __u32 utmp; + if (get_user(utmp, + (__u32 __force __user *) data)) + return -EFAULT; + ret = poke_user_compat(child, addr, utmp); + } + if (ret) + return ret; + addr += sizeof(unsigned int); + data += sizeof(unsigned int); + copied += sizeof(unsigned int); + } + return 0; + case PTRACE_GET_LAST_BREAK: + put_user(task_thread_info(child)->last_break, + (unsigned int __user *) data); + return 0; + } + return compat_ptrace_request(child, request, addr, data); +} +#endif + +asmlinkage long do_syscall_trace_enter(struct pt_regs *regs) +{ + long ret = 0; + + /* Do the secure computing check first. */ + if (secure_computing()) { + /* seccomp failures shouldn't expose any additional code. */ + ret = -1; + goto out; + } + + /* + * The sysc_tracesys code in entry.S stored the system + * call number to gprs[2]. + */ + if (test_thread_flag(TIF_SYSCALL_TRACE) && + (tracehook_report_syscall_entry(regs) || + regs->gprs[2] >= NR_syscalls)) { + /* + * Tracing decided this syscall should not happen or the + * debugger stored an invalid system call number. Skip + * the system call and the system call restart handling. + */ + clear_pt_regs_flag(regs, PIF_SYSCALL); + ret = -1; + } + + if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) + trace_sys_enter(regs, regs->gprs[2]); + + audit_syscall_entry(regs->gprs[2], regs->orig_gpr2, + regs->gprs[3], regs->gprs[4], + regs->gprs[5]); +out: + return ret ?: regs->gprs[2]; +} + +asmlinkage void do_syscall_trace_exit(struct pt_regs *regs) +{ + audit_syscall_exit(regs); + + if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) + trace_sys_exit(regs, regs->gprs[2]); + + if (test_thread_flag(TIF_SYSCALL_TRACE)) + tracehook_report_syscall_exit(regs, 0); +} + +/* + * user_regset definitions. + */ + +static int s390_regs_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + if (target == current) + save_access_regs(target->thread.acrs); + + if (kbuf) { + unsigned long *k = kbuf; + while (count > 0) { + *k++ = __peek_user(target, pos); + count -= sizeof(*k); + pos += sizeof(*k); + } + } else { + unsigned long __user *u = ubuf; + while (count > 0) { + if (__put_user(__peek_user(target, pos), u++)) + return -EFAULT; + count -= sizeof(*u); + pos += sizeof(*u); + } + } + return 0; +} + +static int s390_regs_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + int rc = 0; + + if (target == current) + save_access_regs(target->thread.acrs); + + if (kbuf) { + const unsigned long *k = kbuf; + while (count > 0 && !rc) { + rc = __poke_user(target, pos, *k++); + count -= sizeof(*k); + pos += sizeof(*k); + } + } else { + const unsigned long __user *u = ubuf; + while (count > 0 && !rc) { + unsigned long word; + rc = __get_user(word, u++); + if (rc) + break; + rc = __poke_user(target, pos, word); + count -= sizeof(*u); + pos += sizeof(*u); + } + } + + if (rc == 0 && target == current) + restore_access_regs(target->thread.acrs); + + return rc; +} + +static int s390_fpregs_get(struct task_struct *target, + const struct user_regset *regset, unsigned int pos, + unsigned int count, void *kbuf, void __user *ubuf) +{ + if (target == current) { + save_fp_ctl(&target->thread.fp_regs.fpc); + save_fp_regs(target->thread.fp_regs.fprs); + } else if (target->thread.vxrs) { + int i; + + for (i = 0; i < __NUM_VXRS_LOW; i++) + target->thread.fp_regs.fprs[i] = + *(freg_t *)(target->thread.vxrs + i); + } + return user_regset_copyout(&pos, &count, &kbuf, &ubuf, + &target->thread.fp_regs, 0, -1); +} + +static int s390_fpregs_set(struct task_struct *target, + const struct user_regset *regset, unsigned int pos, + unsigned int count, const void *kbuf, + const void __user *ubuf) +{ + int rc = 0; + + if (target == current) { + save_fp_ctl(&target->thread.fp_regs.fpc); + save_fp_regs(target->thread.fp_regs.fprs); + } + + /* If setting FPC, must validate it first. */ + if (count > 0 && pos < offsetof(s390_fp_regs, fprs)) { + u32 ufpc[2] = { target->thread.fp_regs.fpc, 0 }; + rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &ufpc, + 0, offsetof(s390_fp_regs, fprs)); + if (rc) + return rc; + if (ufpc[1] != 0 || test_fp_ctl(ufpc[0])) + return -EINVAL; + target->thread.fp_regs.fpc = ufpc[0]; + } + + if (rc == 0 && count > 0) + rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, + target->thread.fp_regs.fprs, + offsetof(s390_fp_regs, fprs), -1); + + if (rc == 0) { + if (target == current) { + restore_fp_ctl(&target->thread.fp_regs.fpc); + restore_fp_regs(target->thread.fp_regs.fprs); + } else if (target->thread.vxrs) { + int i; + + for (i = 0; i < __NUM_VXRS_LOW; i++) + *(freg_t *)(target->thread.vxrs + i) = + target->thread.fp_regs.fprs[i]; + } + } + + return rc; +} + +static int s390_last_break_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + if (count > 0) { + if (kbuf) { + unsigned long *k = kbuf; + *k = task_thread_info(target)->last_break; + } else { + unsigned long __user *u = ubuf; + if (__put_user(task_thread_info(target)->last_break, u)) + return -EFAULT; + } + } + return 0; +} + +static int s390_last_break_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + return 0; +} + +static int s390_tdb_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + struct pt_regs *regs = task_pt_regs(target); + unsigned char *data; + + if (!(regs->int_code & 0x200)) + return -ENODATA; + data = target->thread.trap_tdb; + return user_regset_copyout(&pos, &count, &kbuf, &ubuf, data, 0, 256); +} + +static int s390_tdb_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + return 0; +} + +static int s390_vxrs_low_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + __u64 vxrs[__NUM_VXRS_LOW]; + int i; + + if (!MACHINE_HAS_VX) + return -ENODEV; + if (target->thread.vxrs) { + if (target == current) + save_vx_regs(target->thread.vxrs); + for (i = 0; i < __NUM_VXRS_LOW; i++) + vxrs[i] = *((__u64 *)(target->thread.vxrs + i) + 1); + } else + memset(vxrs, 0, sizeof(vxrs)); + return user_regset_copyout(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1); +} + +static int s390_vxrs_low_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + __u64 vxrs[__NUM_VXRS_LOW]; + int i, rc; + + if (!MACHINE_HAS_VX) + return -ENODEV; + if (!target->thread.vxrs) { + rc = alloc_vector_registers(target); + if (rc) + return rc; + } else if (target == current) + save_vx_regs(target->thread.vxrs); + + rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1); + if (rc == 0) { + for (i = 0; i < __NUM_VXRS_LOW; i++) + *((__u64 *)(target->thread.vxrs + i) + 1) = vxrs[i]; + if (target == current) + restore_vx_regs(target->thread.vxrs); + } + + return rc; +} + +static int s390_vxrs_high_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + __vector128 vxrs[__NUM_VXRS_HIGH]; + + if (!MACHINE_HAS_VX) + return -ENODEV; + if (target->thread.vxrs) { + if (target == current) + save_vx_regs(target->thread.vxrs); + memcpy(vxrs, target->thread.vxrs + __NUM_VXRS_LOW, + sizeof(vxrs)); + } else + memset(vxrs, 0, sizeof(vxrs)); + return user_regset_copyout(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1); +} + +static int s390_vxrs_high_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + int rc; + + if (!MACHINE_HAS_VX) + return -ENODEV; + if (!target->thread.vxrs) { + rc = alloc_vector_registers(target); + if (rc) + return rc; + } else if (target == current) + save_vx_regs(target->thread.vxrs); + + rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, + target->thread.vxrs + __NUM_VXRS_LOW, 0, -1); + if (rc == 0 && target == current) + restore_vx_regs(target->thread.vxrs); + + return rc; +} + +static int s390_system_call_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + unsigned int *data = &task_thread_info(target)->system_call; + return user_regset_copyout(&pos, &count, &kbuf, &ubuf, + data, 0, sizeof(unsigned int)); +} + +static int s390_system_call_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + unsigned int *data = &task_thread_info(target)->system_call; + return user_regset_copyin(&pos, &count, &kbuf, &ubuf, + data, 0, sizeof(unsigned int)); +} + +static const struct user_regset s390_regsets[] = { + { + .core_note_type = NT_PRSTATUS, + .n = sizeof(s390_regs) / sizeof(long), + .size = sizeof(long), + .align = sizeof(long), + .get = s390_regs_get, + .set = s390_regs_set, + }, + { + .core_note_type = NT_PRFPREG, + .n = sizeof(s390_fp_regs) / sizeof(long), + .size = sizeof(long), + .align = sizeof(long), + .get = s390_fpregs_get, + .set = s390_fpregs_set, + }, + { + .core_note_type = NT_S390_SYSTEM_CALL, + .n = 1, + .size = sizeof(unsigned int), + .align = sizeof(unsigned int), + .get = s390_system_call_get, + .set = s390_system_call_set, + }, + { + .core_note_type = NT_S390_LAST_BREAK, + .n = 1, + .size = sizeof(long), + .align = sizeof(long), + .get = s390_last_break_get, + .set = s390_last_break_set, + }, + { + .core_note_type = NT_S390_TDB, + .n = 1, + .size = 256, + .align = 1, + .get = s390_tdb_get, + .set = s390_tdb_set, + }, + { + .core_note_type = NT_S390_VXRS_LOW, + .n = __NUM_VXRS_LOW, + .size = sizeof(__u64), + .align = sizeof(__u64), + .get = s390_vxrs_low_get, + .set = s390_vxrs_low_set, + }, + { + .core_note_type = NT_S390_VXRS_HIGH, + .n = __NUM_VXRS_HIGH, + .size = sizeof(__vector128), + .align = sizeof(__vector128), + .get = s390_vxrs_high_get, + .set = s390_vxrs_high_set, + }, +}; + +static const struct user_regset_view user_s390_view = { + .name = UTS_MACHINE, + .e_machine = EM_S390, + .regsets = s390_regsets, + .n = ARRAY_SIZE(s390_regsets) +}; + +#ifdef CONFIG_COMPAT +static int s390_compat_regs_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + if (target == current) + save_access_regs(target->thread.acrs); + + if (kbuf) { + compat_ulong_t *k = kbuf; + while (count > 0) { + *k++ = __peek_user_compat(target, pos); + count -= sizeof(*k); + pos += sizeof(*k); + } + } else { + compat_ulong_t __user *u = ubuf; + while (count > 0) { + if (__put_user(__peek_user_compat(target, pos), u++)) + return -EFAULT; + count -= sizeof(*u); + pos += sizeof(*u); + } + } + return 0; +} + +static int s390_compat_regs_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + int rc = 0; + + if (target == current) + save_access_regs(target->thread.acrs); + + if (kbuf) { + const compat_ulong_t *k = kbuf; + while (count > 0 && !rc) { + rc = __poke_user_compat(target, pos, *k++); + count -= sizeof(*k); + pos += sizeof(*k); + } + } else { + const compat_ulong_t __user *u = ubuf; + while (count > 0 && !rc) { + compat_ulong_t word; + rc = __get_user(word, u++); + if (rc) + break; + rc = __poke_user_compat(target, pos, word); + count -= sizeof(*u); + pos += sizeof(*u); + } + } + + if (rc == 0 && target == current) + restore_access_regs(target->thread.acrs); + + return rc; +} + +static int s390_compat_regs_high_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + compat_ulong_t *gprs_high; + + gprs_high = (compat_ulong_t *) + &task_pt_regs(target)->gprs[pos / sizeof(compat_ulong_t)]; + if (kbuf) { + compat_ulong_t *k = kbuf; + while (count > 0) { + *k++ = *gprs_high; + gprs_high += 2; + count -= sizeof(*k); + } + } else { + compat_ulong_t __user *u = ubuf; + while (count > 0) { + if (__put_user(*gprs_high, u++)) + return -EFAULT; + gprs_high += 2; + count -= sizeof(*u); + } + } + return 0; +} + +static int s390_compat_regs_high_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + compat_ulong_t *gprs_high; + int rc = 0; + + gprs_high = (compat_ulong_t *) + &task_pt_regs(target)->gprs[pos / sizeof(compat_ulong_t)]; + if (kbuf) { + const compat_ulong_t *k = kbuf; + while (count > 0) { + *gprs_high = *k++; + *gprs_high += 2; + count -= sizeof(*k); + } + } else { + const compat_ulong_t __user *u = ubuf; + while (count > 0 && !rc) { + unsigned long word; + rc = __get_user(word, u++); + if (rc) + break; + *gprs_high = word; + *gprs_high += 2; + count -= sizeof(*u); + } + } + + return rc; +} + +static int s390_compat_last_break_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + compat_ulong_t last_break; + + if (count > 0) { + last_break = task_thread_info(target)->last_break; + if (kbuf) { + unsigned long *k = kbuf; + *k = last_break; + } else { + unsigned long __user *u = ubuf; + if (__put_user(last_break, u)) + return -EFAULT; + } + } + return 0; +} + +static int s390_compat_last_break_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + return 0; +} + +static const struct user_regset s390_compat_regsets[] = { + { + .core_note_type = NT_PRSTATUS, + .n = sizeof(s390_compat_regs) / sizeof(compat_long_t), + .size = sizeof(compat_long_t), + .align = sizeof(compat_long_t), + .get = s390_compat_regs_get, + .set = s390_compat_regs_set, + }, + { + .core_note_type = NT_PRFPREG, + .n = sizeof(s390_fp_regs) / sizeof(compat_long_t), + .size = sizeof(compat_long_t), + .align = sizeof(compat_long_t), + .get = s390_fpregs_get, + .set = s390_fpregs_set, + }, + { + .core_note_type = NT_S390_SYSTEM_CALL, + .n = 1, + .size = sizeof(compat_uint_t), + .align = sizeof(compat_uint_t), + .get = s390_system_call_get, + .set = s390_system_call_set, + }, + { + .core_note_type = NT_S390_LAST_BREAK, + .n = 1, + .size = sizeof(long), + .align = sizeof(long), + .get = s390_compat_last_break_get, + .set = s390_compat_last_break_set, + }, + { + .core_note_type = NT_S390_TDB, + .n = 1, + .size = 256, + .align = 1, + .get = s390_tdb_get, + .set = s390_tdb_set, + }, + { + .core_note_type = NT_S390_VXRS_LOW, + .n = __NUM_VXRS_LOW, + .size = sizeof(__u64), + .align = sizeof(__u64), + .get = s390_vxrs_low_get, + .set = s390_vxrs_low_set, + }, + { + .core_note_type = NT_S390_VXRS_HIGH, + .n = __NUM_VXRS_HIGH, + .size = sizeof(__vector128), + .align = sizeof(__vector128), + .get = s390_vxrs_high_get, + .set = s390_vxrs_high_set, + }, + { + .core_note_type = NT_S390_HIGH_GPRS, + .n = sizeof(s390_compat_regs_high) / sizeof(compat_long_t), + .size = sizeof(compat_long_t), + .align = sizeof(compat_long_t), + .get = s390_compat_regs_high_get, + .set = s390_compat_regs_high_set, + }, +}; + +static const struct user_regset_view user_s390_compat_view = { + .name = "s390", + .e_machine = EM_S390, + .regsets = s390_compat_regsets, + .n = ARRAY_SIZE(s390_compat_regsets) +}; +#endif + +const struct user_regset_view *task_user_regset_view(struct task_struct *task) +{ +#ifdef CONFIG_COMPAT + if (test_tsk_thread_flag(task, TIF_31BIT)) + return &user_s390_compat_view; +#endif + return &user_s390_view; +} + +static const char *gpr_names[NUM_GPRS] = { + "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", + "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15", +}; + +unsigned long regs_get_register(struct pt_regs *regs, unsigned int offset) +{ + if (offset >= NUM_GPRS) + return 0; + return regs->gprs[offset]; +} + +int regs_query_register_offset(const char *name) +{ + unsigned long offset; + + if (!name || *name != 'r') + return -EINVAL; + if (kstrtoul(name + 1, 10, &offset)) + return -EINVAL; + if (offset >= NUM_GPRS) + return -EINVAL; + return offset; +} + +const char *regs_query_register_name(unsigned int offset) +{ + if (offset >= NUM_GPRS) + return NULL; + return gpr_names[offset]; +} + +static int regs_within_kernel_stack(struct pt_regs *regs, unsigned long addr) +{ + unsigned long ksp = kernel_stack_pointer(regs); + + return (addr & ~(THREAD_SIZE - 1)) == (ksp & ~(THREAD_SIZE - 1)); +} + +/** + * regs_get_kernel_stack_nth() - get Nth entry of the stack + * @regs:pt_regs which contains kernel stack pointer. + * @n:stack entry number. + * + * regs_get_kernel_stack_nth() returns @n th entry of the kernel stack which + * is specifined by @regs. If the @n th entry is NOT in the kernel stack, + * this returns 0. + */ +unsigned long regs_get_kernel_stack_nth(struct pt_regs *regs, unsigned int n) +{ + unsigned long addr; + + addr = kernel_stack_pointer(regs) + n * sizeof(long); + if (!regs_within_kernel_stack(regs, addr)) + return 0; + return *(unsigned long *)addr; +} diff --git a/arch/s390/kernel/reipl.S b/arch/s390/kernel/reipl.S new file mode 100644 index 000000000..52aab0bd8 --- /dev/null +++ b/arch/s390/kernel/reipl.S @@ -0,0 +1,155 @@ +/* + * Copyright IBM Corp 2000, 2011 + * Author(s): Holger Smolinski <Holger.Smolinski@de.ibm.com>, + * Denis Joseph Barrow, + */ + +#include <linux/linkage.h> +#include <asm/asm-offsets.h> +#include <asm/sigp.h> + +# +# store_status +# +# Prerequisites to run this function: +# - Prefix register is set to zero +# - Original prefix register is stored in "dump_prefix_page" +# - Lowcore protection is off +# +ENTRY(store_status) + /* Save register one and load save area base */ + stg %r1,__LC_SAVE_AREA_RESTART + lghi %r1,SAVE_AREA_BASE + /* General purpose registers */ + stmg %r0,%r15,__LC_GPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + lg %r2,__LC_SAVE_AREA_RESTART + stg %r2,__LC_GPREGS_SAVE_AREA-SAVE_AREA_BASE+8(%r1) + /* Control registers */ + stctg %c0,%c15,__LC_CREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* Access registers */ + stam %a0,%a15,__LC_AREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* Floating point registers */ + std %f0, 0x00 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f1, 0x08 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f2, 0x10 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f3, 0x18 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f4, 0x20 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f5, 0x28 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f6, 0x30 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f7, 0x38 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f8, 0x40 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f9, 0x48 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f10,0x50 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f11,0x58 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f12,0x60 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f13,0x68 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f14,0x70 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f15,0x78 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* Floating point control register */ + stfpc __LC_FP_CREG_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* CPU timer */ + stpt __LC_CPU_TIMER_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* Saved prefix register */ + larl %r2,dump_prefix_page + mvc __LC_PREFIX_SAVE_AREA-SAVE_AREA_BASE(4,%r1),0(%r2) + /* Clock comparator - seven bytes */ + larl %r2,.Lclkcmp + stckc 0(%r2) + mvc __LC_CLOCK_COMP_SAVE_AREA-SAVE_AREA_BASE + 1(7,%r1),1(%r2) + /* Program status word */ + epsw %r2,%r3 + st %r2,__LC_PSW_SAVE_AREA-SAVE_AREA_BASE + 0(%r1) + st %r3,__LC_PSW_SAVE_AREA-SAVE_AREA_BASE + 4(%r1) + larl %r2,store_status + stg %r2,__LC_PSW_SAVE_AREA-SAVE_AREA_BASE + 8(%r1) + br %r14 + + .section .bss + .align 8 +.Lclkcmp: .quad 0x0000000000000000 + .previous + +# +# do_reipl_asm +# Parameter: r2 = schid of reipl device +# + +ENTRY(do_reipl_asm) + basr %r13,0 +.Lpg0: lpswe .Lnewpsw-.Lpg0(%r13) +.Lpg1: brasl %r14,store_status + + lctlg %c6,%c6,.Lall-.Lpg0(%r13) + lgr %r1,%r2 + mvc __LC_PGM_NEW_PSW(16),.Lpcnew-.Lpg0(%r13) + stsch .Lschib-.Lpg0(%r13) + oi .Lschib+5-.Lpg0(%r13),0x84 +.Lecs: xi .Lschib+27-.Lpg0(%r13),0x01 + msch .Lschib-.Lpg0(%r13) + lghi %r0,5 +.Lssch: ssch .Liplorb-.Lpg0(%r13) + jz .L001 + brct %r0,.Lssch + bas %r14,.Ldisab-.Lpg0(%r13) +.L001: mvc __LC_IO_NEW_PSW(16),.Lionew-.Lpg0(%r13) +.Ltpi: lpswe .Lwaitpsw-.Lpg0(%r13) +.Lcont: c %r1,__LC_SUBCHANNEL_ID + jnz .Ltpi + clc __LC_IO_INT_PARM(4),.Liplorb-.Lpg0(%r13) + jnz .Ltpi + tsch .Liplirb-.Lpg0(%r13) + tm .Liplirb+9-.Lpg0(%r13),0xbf + jz .L002 + bas %r14,.Ldisab-.Lpg0(%r13) +.L002: tm .Liplirb+8-.Lpg0(%r13),0xf3 + jz .L003 + bas %r14,.Ldisab-.Lpg0(%r13) +.L003: st %r1,__LC_SUBCHANNEL_ID + lhi %r1,0 # mode 0 = esa + slr %r0,%r0 # set cpuid to zero + sigp %r1,%r0,SIGP_SET_ARCHITECTURE # switch to esa mode + lpsw 0 +.Ldisab: sll %r14,1 + srl %r14,1 # need to kill hi bit to avoid specification exceptions. + st %r14,.Ldispsw+12-.Lpg0(%r13) + lpswe .Ldispsw-.Lpg0(%r13) + .align 8 +.Lall: .quad 0x00000000ff000000 + .align 16 +/* + * These addresses have to be 31 bit otherwise + * the sigp will throw a specifcation exception + * when switching to ESA mode as bit 31 be set + * in the ESA psw. + * Bit 31 of the addresses has to be 0 for the + * 31bit lpswe instruction a fact they appear to have + * omitted from the pop. + */ +.Lnewpsw: .quad 0x0000000080000000 + .quad .Lpg1 +.Lpcnew: .quad 0x0000000080000000 + .quad .Lecs +.Lionew: .quad 0x0000000080000000 + .quad .Lcont +.Lwaitpsw: .quad 0x0202000080000000 + .quad .Ltpi +.Ldispsw: .quad 0x0002000080000000 + .quad 0x0000000000000000 +.Liplccws: .long 0x02000000,0x60000018 + .long 0x08000008,0x20000001 +.Liplorb: .long 0x0049504c,0x0040ff80 + .long 0x00000000+.Liplccws +.Lschib: .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 +.Liplirb: .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 + .long 0x00000000,0x00000000 diff --git a/arch/s390/kernel/relocate_kernel.S b/arch/s390/kernel/relocate_kernel.S new file mode 100644 index 000000000..cfac28330 --- /dev/null +++ b/arch/s390/kernel/relocate_kernel.S @@ -0,0 +1,121 @@ +/* + * Copyright IBM Corp. 2005 + * + * Author(s): Rolf Adelsberger, + * Heiko Carstens <heiko.carstens@de.ibm.com> + * + */ + +#include <linux/linkage.h> +#include <asm/sigp.h> + +/* + * moves the new kernel to its destination... + * %r2 = pointer to first kimage_entry_t + * %r3 = start address - where to jump to after the job is done... + * + * %r5 will be used as temp. storage + * %r6 holds the destination address + * %r7 = PAGE_SIZE + * %r8 holds the source address + * %r9 = PAGE_SIZE + * + * 0xf000 is a page_mask + */ + + .text +ENTRY(relocate_kernel) + basr %r13,0 # base address + .base: + stnsm sys_msk-.base(%r13),0xfb # disable DAT + stctg %c0,%c15,ctlregs-.base(%r13) + stmg %r0,%r15,gprregs-.base(%r13) + lghi %r0,3 + sllg %r0,%r0,31 + stg %r0,0x1d0(%r0) + la %r0,.back_pgm-.base(%r13) + stg %r0,0x1d8(%r0) + la %r1,load_psw-.base(%r13) + mvc 0(8,%r0),0(%r1) + la %r0,.back-.base(%r13) + st %r0,4(%r0) + oi 4(%r0),0x80 + lghi %r0,0 + diag %r0,%r0,0x308 + .back: + lhi %r1,1 # mode 1 = esame + sigp %r1,%r0,SIGP_SET_ARCHITECTURE # switch to esame mode + sam64 # switch to 64 bit addressing mode + basr %r13,0 + .back_base: + oi have_diag308-.back_base(%r13),0x01 + lctlg %c0,%c15,ctlregs-.back_base(%r13) + lmg %r0,%r15,gprregs-.back_base(%r13) + j .top + .back_pgm: + lmg %r0,%r15,gprregs-.base(%r13) + .top: + lghi %r7,4096 # load PAGE_SIZE in r7 + lghi %r9,4096 # load PAGE_SIZE in r9 + lg %r5,0(%r2) # read another word for indirection page + aghi %r2,8 # increment pointer + tml %r5,0x1 # is it a destination page? + je .indir_check # NO, goto "indir_check" + lgr %r6,%r5 # r6 = r5 + nill %r6,0xf000 # mask it out and... + j .top # ...next iteration + .indir_check: + tml %r5,0x2 # is it a indirection page? + je .done_test # NO, goto "done_test" + nill %r5,0xf000 # YES, mask out, + lgr %r2,%r5 # move it into the right register, + j .top # and read next... + .done_test: + tml %r5,0x4 # is it the done indicator? + je .source_test # NO! Well, then it should be the source indicator... + j .done # ok, lets finish it here... + .source_test: + tml %r5,0x8 # it should be a source indicator... + je .top # NO, ignore it... + lgr %r8,%r5 # r8 = r5 + nill %r8,0xf000 # masking + 0: mvcle %r6,%r8,0x0 # copy PAGE_SIZE bytes from r8 to r6 - pad with 0 + jo 0b + j .top + .done: + sgr %r0,%r0 # clear register r0 + la %r4,load_psw-.base(%r13) # load psw-address into the register + o %r3,4(%r4) # or load address into psw + st %r3,4(%r4) + mvc 0(8,%r0),0(%r4) # copy psw to absolute address 0 + tm have_diag308-.base(%r13),0x01 + jno .no_diag308 + diag %r0,%r0,0x308 + .no_diag308: + sam31 # 31 bit mode + sr %r1,%r1 # erase register r1 + sr %r2,%r2 # erase register r2 + sigp %r1,%r2,SIGP_SET_ARCHITECTURE # set cpuid to zero + lpsw 0 # hopefully start new kernel... + + .align 8 + load_psw: + .long 0x00080000,0x80000000 + sys_msk: + .quad 0 + ctlregs: + .rept 16 + .quad 0 + .endr + gprregs: + .rept 16 + .quad 0 + .endr + have_diag308: + .byte 0 + .align 8 + relocate_kernel_end: + .align 8 + .globl relocate_kernel_len + relocate_kernel_len: + .quad relocate_kernel_end - relocate_kernel diff --git a/arch/s390/kernel/runtime_instr.c b/arch/s390/kernel/runtime_instr.c new file mode 100644 index 000000000..26b4ae96f --- /dev/null +++ b/arch/s390/kernel/runtime_instr.c @@ -0,0 +1,149 @@ +/* + * Copyright IBM Corp. 2012 + * Author(s): Jan Glauber <jang@linux.vnet.ibm.com> + */ + +#include <linux/kernel.h> +#include <linux/syscalls.h> +#include <linux/signal.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <linux/kernel_stat.h> +#include <asm/runtime_instr.h> +#include <asm/cpu_mf.h> +#include <asm/irq.h> + +/* empty control block to disable RI by loading it */ +struct runtime_instr_cb runtime_instr_empty_cb; + +static int runtime_instr_avail(void) +{ + return test_facility(64); +} + +static void disable_runtime_instr(void) +{ + struct pt_regs *regs = task_pt_regs(current); + + load_runtime_instr_cb(&runtime_instr_empty_cb); + + /* + * Make sure the RI bit is deleted from the PSW. If the user did not + * switch off RI before the system call the process will get a + * specification exception otherwise. + */ + regs->psw.mask &= ~PSW_MASK_RI; +} + +static void init_runtime_instr_cb(struct runtime_instr_cb *cb) +{ + cb->buf_limit = 0xfff; + cb->int_requested = 1; + cb->pstate = 1; + cb->pstate_set_buf = 1; + cb->pstate_sample = 1; + cb->pstate_collect = 1; + cb->key = PAGE_DEFAULT_KEY; + cb->valid = 1; +} + +void exit_thread_runtime_instr(void) +{ + struct task_struct *task = current; + + if (!task->thread.ri_cb) + return; + disable_runtime_instr(); + kfree(task->thread.ri_cb); + task->thread.ri_signum = 0; + task->thread.ri_cb = NULL; +} + +static void runtime_instr_int_handler(struct ext_code ext_code, + unsigned int param32, unsigned long param64) +{ + struct siginfo info; + + if (!(param32 & CPU_MF_INT_RI_MASK)) + return; + + inc_irq_stat(IRQEXT_CMR); + + if (!current->thread.ri_cb) + return; + if (current->thread.ri_signum < SIGRTMIN || + current->thread.ri_signum > SIGRTMAX) { + WARN_ON_ONCE(1); + return; + } + + memset(&info, 0, sizeof(info)); + info.si_signo = current->thread.ri_signum; + info.si_code = SI_QUEUE; + if (param32 & CPU_MF_INT_RI_BUF_FULL) + info.si_int = ENOBUFS; + else if (param32 & CPU_MF_INT_RI_HALTED) + info.si_int = ECANCELED; + else + return; /* unknown reason */ + + send_sig_info(current->thread.ri_signum, &info, current); +} + +SYSCALL_DEFINE2(s390_runtime_instr, int, command, int, signum) +{ + struct runtime_instr_cb *cb; + + if (!runtime_instr_avail()) + return -EOPNOTSUPP; + + if (command == S390_RUNTIME_INSTR_STOP) { + preempt_disable(); + exit_thread_runtime_instr(); + preempt_enable(); + return 0; + } + + if (command != S390_RUNTIME_INSTR_START || + (signum < SIGRTMIN || signum > SIGRTMAX)) + return -EINVAL; + + if (!current->thread.ri_cb) { + cb = kzalloc(sizeof(*cb), GFP_KERNEL); + if (!cb) + return -ENOMEM; + } else { + cb = current->thread.ri_cb; + memset(cb, 0, sizeof(*cb)); + } + + init_runtime_instr_cb(cb); + current->thread.ri_signum = signum; + + /* now load the control block to make it available */ + preempt_disable(); + current->thread.ri_cb = cb; + load_runtime_instr_cb(cb); + preempt_enable(); + return 0; +} + +static int __init runtime_instr_init(void) +{ + int rc; + + if (!runtime_instr_avail()) + return 0; + + irq_subclass_register(IRQ_SUBCLASS_MEASUREMENT_ALERT); + rc = register_external_irq(EXT_IRQ_MEASURE_ALERT, + runtime_instr_int_handler); + if (rc) + irq_subclass_unregister(IRQ_SUBCLASS_MEASUREMENT_ALERT); + else + pr_info("Runtime instrumentation facility initialized\n"); + return rc; +} +device_initcall(runtime_instr_init); diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c new file mode 100644 index 000000000..9f6046793 --- /dev/null +++ b/arch/s390/kernel/s390_ksyms.c @@ -0,0 +1,13 @@ +#include <linux/module.h> +#include <linux/kvm_host.h> +#include <asm/ftrace.h> + +#ifdef CONFIG_FUNCTION_TRACER +EXPORT_SYMBOL(_mcount); +#endif +#if IS_ENABLED(CONFIG_KVM) +EXPORT_SYMBOL(sie64a); +EXPORT_SYMBOL(sie_exit); +#endif +EXPORT_SYMBOL(memcpy); +EXPORT_SYMBOL(memset); diff --git a/arch/s390/kernel/sclp.S b/arch/s390/kernel/sclp.S new file mode 100644 index 000000000..43c3169ea --- /dev/null +++ b/arch/s390/kernel/sclp.S @@ -0,0 +1,351 @@ +/* + * Mini SCLP driver. + * + * Copyright IBM Corp. 2004, 2009 + * + * Author(s): Peter Oberparleiter <Peter.Oberparleiter@de.ibm.com>, + * Heiko Carstens <heiko.carstens@de.ibm.com>, + * + */ + +#include <linux/linkage.h> +#include <asm/irq.h> + +LC_EXT_NEW_PSW = 0x58 # addr of ext int handler +LC_EXT_NEW_PSW_64 = 0x1b0 # addr of ext int handler 64 bit +LC_EXT_INT_PARAM = 0x80 # addr of ext int parameter +LC_EXT_INT_CODE = 0x86 # addr of ext int code +LC_AR_MODE_ID = 0xa3 + +# +# Subroutine which waits synchronously until either an external interruption +# or a timeout occurs. +# +# Parameters: +# R2 = 0 for no timeout, non-zero for timeout in (approximated) seconds +# +# Returns: +# R2 = 0 on interrupt, 2 on timeout +# R3 = external interruption parameter if R2=0 +# + +_sclp_wait_int: + stm %r6,%r15,24(%r15) # save registers + basr %r13,0 # get base register +.LbaseS1: + ahi %r15,-96 # create stack frame + la %r8,LC_EXT_NEW_PSW # register int handler + la %r9,.LextpswS1-.LbaseS1(%r13) + tm LC_AR_MODE_ID,1 + jno .Lesa1 + la %r8,LC_EXT_NEW_PSW_64 # register int handler 64 bit + la %r9,.LextpswS1_64-.LbaseS1(%r13) +.Lesa1: + mvc .LoldpswS1-.LbaseS1(16,%r13),0(%r8) + mvc 0(16,%r8),0(%r9) + epsw %r6,%r7 # set current addressing mode + nill %r6,0x1 # in new psw (31 or 64 bit mode) + nilh %r7,0x8000 + stm %r6,%r7,0(%r8) + lhi %r6,0x0200 # cr mask for ext int (cr0.54) + ltr %r2,%r2 + jz .LsetctS1 + ahi %r6,0x0800 # cr mask for clock int (cr0.52) + stck .LtimeS1-.LbaseS1(%r13) # initiate timeout + al %r2,.LtimeS1-.LbaseS1(%r13) + st %r2,.LtimeS1-.LbaseS1(%r13) + sckc .LtimeS1-.LbaseS1(%r13) + +.LsetctS1: + stctl %c0,%c0,.LctlS1-.LbaseS1(%r13) # enable required interrupts + l %r0,.LctlS1-.LbaseS1(%r13) + lhi %r1,~(0x200 | 0x800) # clear old values + nr %r1,%r0 + or %r1,%r6 # set new value + st %r1,.LctlS1-.LbaseS1(%r13) + lctl %c0,%c0,.LctlS1-.LbaseS1(%r13) + st %r0,.LctlS1-.LbaseS1(%r13) + lhi %r2,2 # return code for timeout +.LloopS1: + lpsw .LwaitpswS1-.LbaseS1(%r13) # wait until interrupt +.LwaitS1: + lh %r7,LC_EXT_INT_CODE + chi %r7,EXT_IRQ_CLK_COMP # timeout? + je .LtimeoutS1 + chi %r7,EXT_IRQ_SERVICE_SIG # service int? + jne .LloopS1 + sr %r2,%r2 + l %r3,LC_EXT_INT_PARAM +.LtimeoutS1: + lctl %c0,%c0,.LctlS1-.LbaseS1(%r13) # restore interrupt setting + # restore old handler + mvc 0(16,%r8),.LoldpswS1-.LbaseS1(%r13) + lm %r6,%r15,120(%r15) # restore registers + br %r14 # return to caller + + .align 8 +.LoldpswS1: + .long 0, 0, 0, 0 # old ext int PSW +.LextpswS1: + .long 0x00080000, 0x80000000+.LwaitS1 # PSW to handle ext int +.LextpswS1_64: + .quad 0, .LwaitS1 # PSW to handle ext int, 64 bit +.LwaitpswS1: + .long 0x010a0000, 0x00000000+.LloopS1 # PSW to wait for ext int +.LtimeS1: + .quad 0 # current time +.LctlS1: + .long 0 # CT0 contents + +# +# Subroutine to synchronously issue a service call. +# +# Parameters: +# R2 = command word +# R3 = sccb address +# +# Returns: +# R2 = 0 on success, 1 on failure +# R3 = sccb response code if R2 = 0 +# + +_sclp_servc: + stm %r6,%r15,24(%r15) # save registers + ahi %r15,-96 # create stack frame + lr %r6,%r2 # save command word + lr %r7,%r3 # save sccb address +.LretryS2: + lhi %r2,1 # error return code + .insn rre,0xb2200000,%r6,%r7 # servc + brc 1,.LendS2 # exit if not operational + brc 8,.LnotbusyS2 # go on if not busy + sr %r2,%r2 # wait until no longer busy + bras %r14,_sclp_wait_int + j .LretryS2 # retry +.LnotbusyS2: + sr %r2,%r2 # wait until result + bras %r14,_sclp_wait_int + sr %r2,%r2 + lh %r3,6(%r7) +.LendS2: + lm %r6,%r15,120(%r15) # restore registers + br %r14 + +# +# Subroutine to set up the SCLP interface. +# +# Parameters: +# R2 = 0 to activate, non-zero to deactivate +# +# Returns: +# R2 = 0 on success, non-zero on failure +# + +_sclp_setup: + stm %r6,%r15,24(%r15) # save registers + ahi %r15,-96 # create stack frame + basr %r13,0 # get base register +.LbaseS3: + l %r6,.LsccbS0-.LbaseS3(%r13) # prepare init mask sccb + mvc 0(.LinitendS3-.LinitsccbS3,%r6),.LinitsccbS3-.LbaseS3(%r13) + ltr %r2,%r2 # initialization? + jz .LdoinitS3 # go ahead + # clear masks + xc .LinitmaskS3-.LinitsccbS3(8,%r6),.LinitmaskS3-.LinitsccbS3(%r6) +.LdoinitS3: + l %r2,.LwritemaskS3-.LbaseS3(%r13)# get command word + lr %r3,%r6 # get sccb address + bras %r14,_sclp_servc # issue service call + ltr %r2,%r2 # servc successful? + jnz .LerrorS3 + chi %r3,0x20 # write mask successful? + jne .LerrorS3 + # check masks + la %r2,.LinitmaskS3-.LinitsccbS3(%r6) + l %r1,0(%r2) # receive mask ok? + n %r1,12(%r2) + cl %r1,0(%r2) + jne .LerrorS3 + l %r1,4(%r2) # send mask ok? + n %r1,8(%r2) + cl %r1,4(%r2) + sr %r2,%r2 + je .LendS3 +.LerrorS3: + lhi %r2,1 # error return code +.LendS3: + lm %r6,%r15,120(%r15) # restore registers + br %r14 +.LwritemaskS3: + .long 0x00780005 # SCLP command for write mask +.LinitsccbS3: + .word .LinitendS3-.LinitsccbS3 + .byte 0,0,0,0 + .word 0 + .word 0 + .word 4 +.LinitmaskS3: + .long 0x80000000 + .long 0x40000000 + .long 0 + .long 0 +.LinitendS3: + +# +# Subroutine which prints a given text to the SCLP console. +# +# Parameters: +# R2 = address of nil-terminated ASCII text +# +# Returns: +# R2 = 0 on success, 1 on failure +# + +_sclp_print: + stm %r6,%r15,24(%r15) # save registers + ahi %r15,-96 # create stack frame + basr %r13,0 # get base register +.LbaseS4: + l %r8,.LsccbS0-.LbaseS4(%r13) # prepare write data sccb + mvc 0(.LmtoS4-.LwritesccbS4,%r8),.LwritesccbS4-.LbaseS4(%r13) + la %r7,.LmtoS4-.LwritesccbS4(%r8) # current mto addr + sr %r0,%r0 + l %r10,.Lascebc-.LbaseS4(%r13) # address of translation table +.LinitmtoS4: + # initialize mto + mvc 0(.LmtoendS4-.LmtoS4,%r7),.LmtoS4-.LbaseS4(%r13) + lhi %r6,.LmtoendS4-.LmtoS4 # current mto length +.LloopS4: + ic %r0,0(%r2) # get character + ahi %r2,1 + ltr %r0,%r0 # end of string? + jz .LfinalizemtoS4 + chi %r0,0x0a # end of line (NL)? + jz .LfinalizemtoS4 + stc %r0,0(%r6,%r7) # copy to mto + la %r11,0(%r6,%r7) + tr 0(1,%r11),0(%r10) # translate to EBCDIC + ahi %r6,1 + j .LloopS4 +.LfinalizemtoS4: + sth %r6,0(%r7) # update mto length + lh %r9,.LmdbS4-.LwritesccbS4(%r8) # update mdb length + ar %r9,%r6 + sth %r9,.LmdbS4-.LwritesccbS4(%r8) + lh %r9,.LevbufS4-.LwritesccbS4(%r8)# update evbuf length + ar %r9,%r6 + sth %r9,.LevbufS4-.LwritesccbS4(%r8) + lh %r9,0(%r8) # update sccb length + ar %r9,%r6 + sth %r9,0(%r8) + ar %r7,%r6 # update current mto address + ltr %r0,%r0 # more characters? + jnz .LinitmtoS4 + l %r2,.LwritedataS4-.LbaseS4(%r13)# write data + lr %r3,%r8 + bras %r14,_sclp_servc + ltr %r2,%r2 # servc successful? + jnz .LendS4 + chi %r3,0x20 # write data successful? + je .LendS4 + lhi %r2,1 # error return code +.LendS4: + lm %r6,%r15,120(%r15) # restore registers + br %r14 + +# +# Function which prints a given text to the SCLP console. +# +# Parameters: +# R2 = address of nil-terminated ASCII text +# +# Returns: +# R2 = 0 on success, 1 on failure +# + +ENTRY(_sclp_print_early) + stm %r6,%r15,24(%r15) # save registers + ahi %r15,-96 # create stack frame + tm LC_AR_MODE_ID,1 + jno .Lesa2 + ahi %r15,-80 + stmh %r6,%r15,96(%r15) # store upper register halves +.Lesa2: + lr %r10,%r2 # save string pointer + lhi %r2,0 + bras %r14,_sclp_setup # enable console + ltr %r2,%r2 + jnz .LendS5 + lr %r2,%r10 + bras %r14,_sclp_print # print string + ltr %r2,%r2 + jnz .LendS5 + lhi %r2,1 + bras %r14,_sclp_setup # disable console +.LendS5: + tm LC_AR_MODE_ID,1 + jno .Lesa3 + lgfr %r2,%r2 # sign extend return value + lmh %r6,%r15,96(%r15) # restore upper register halves + ahi %r15,80 +.Lesa3: + lm %r6,%r15,120(%r15) # restore registers + br %r14 + +.LwritedataS4: + .long 0x00760005 # SCLP command for write data +.LwritesccbS4: + # sccb + .word .LmtoS4-.LwritesccbS4 + .byte 0 + .byte 0,0,0 + .word 0 + + # evbuf +.LevbufS4: + .word .LmtoS4-.LevbufS4 + .byte 0x02 + .byte 0 + .word 0 + +.LmdbS4: + # mdb + .word .LmtoS4-.LmdbS4 + .word 1 + .long 0xd4c4c240 + .long 1 + + # go +.LgoS4: + .word .LmtoS4-.LgoS4 + .word 1 + .long 0 + .byte 0,0,0,0,0,0,0,0 + .byte 0,0,0 + .byte 0 + .byte 0,0,0,0,0,0,0 + .byte 0 + .word 0 + .byte 0,0,0,0,0,0,0,0,0,0 + .byte 0,0,0,0,0,0,0,0 + .byte 0,0,0,0,0,0,0,0 + +.LmtoS4: + .word .LmtoendS4-.LmtoS4 + .word 4 + .word 0x1000 + .byte 0 + .byte 0,0,0 +.LmtoendS4: + + # Global constants +.LsccbS0: + .long _sclp_work_area +.Lascebc: + .long _ascebc + +.section .data,"aw",@progbits + .balign 4096 +_sclp_work_area: + .fill 4096 +.previous diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c new file mode 100644 index 000000000..7262fe438 --- /dev/null +++ b/arch/s390/kernel/setup.c @@ -0,0 +1,892 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2012 + * Author(s): Hartmut Penner (hp@de.ibm.com), + * Martin Schwidefsky (schwidefsky@de.ibm.com) + * + * Derived from "arch/i386/kernel/setup.c" + * Copyright (C) 1995, Linus Torvalds + */ + +/* + * This file handles the architecture-dependent parts of initialization + */ + +#define KMSG_COMPONENT "setup" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/errno.h> +#include <linux/export.h> +#include <linux/sched.h> +#include <linux/kernel.h> +#include <linux/memblock.h> +#include <linux/mm.h> +#include <linux/stddef.h> +#include <linux/unistd.h> +#include <linux/ptrace.h> +#include <linux/random.h> +#include <linux/user.h> +#include <linux/tty.h> +#include <linux/ioport.h> +#include <linux/delay.h> +#include <linux/init.h> +#include <linux/initrd.h> +#include <linux/bootmem.h> +#include <linux/root_dev.h> +#include <linux/console.h> +#include <linux/kernel_stat.h> +#include <linux/device.h> +#include <linux/notifier.h> +#include <linux/pfn.h> +#include <linux/ctype.h> +#include <linux/reboot.h> +#include <linux/topology.h> +#include <linux/kexec.h> +#include <linux/crash_dump.h> +#include <linux/memory.h> +#include <linux/compat.h> + +#include <asm/ipl.h> +#include <asm/facility.h> +#include <asm/smp.h> +#include <asm/mmu_context.h> +#include <asm/cpcmd.h> +#include <asm/lowcore.h> +#include <asm/irq.h> +#include <asm/page.h> +#include <asm/ptrace.h> +#include <asm/sections.h> +#include <asm/ebcdic.h> +#include <asm/kvm_virtio.h> +#include <asm/diag.h> +#include <asm/os_info.h> +#include <asm/sclp.h> +#include <asm/sysinfo.h> +#include "entry.h" + +/* + * Machine setup.. + */ +unsigned int console_mode = 0; +EXPORT_SYMBOL(console_mode); + +unsigned int console_devno = -1; +EXPORT_SYMBOL(console_devno); + +unsigned int console_irq = -1; +EXPORT_SYMBOL(console_irq); + +unsigned long elf_hwcap = 0; +char elf_platform[ELF_PLATFORM_SIZE]; + +int __initdata memory_end_set; +unsigned long __initdata memory_end; +unsigned long __initdata max_physmem_end; + +unsigned long VMALLOC_START; +EXPORT_SYMBOL(VMALLOC_START); + +unsigned long VMALLOC_END; +EXPORT_SYMBOL(VMALLOC_END); + +struct page *vmemmap; +EXPORT_SYMBOL(vmemmap); + +unsigned long MODULES_VADDR; +unsigned long MODULES_END; + +/* An array with a pointer to the lowcore of every CPU. */ +struct _lowcore *lowcore_ptr[NR_CPUS]; +EXPORT_SYMBOL(lowcore_ptr); + +/* + * This is set up by the setup-routine at boot-time + * for S390 need to find out, what we have to setup + * using address 0x10400 ... + */ + +#include <asm/setup.h> + +/* + * condev= and conmode= setup parameter. + */ + +static int __init condev_setup(char *str) +{ + int vdev; + + vdev = simple_strtoul(str, &str, 0); + if (vdev >= 0 && vdev < 65536) { + console_devno = vdev; + console_irq = -1; + } + return 1; +} + +__setup("condev=", condev_setup); + +static void __init set_preferred_console(void) +{ + if (MACHINE_IS_KVM) { + if (sclp_has_vt220()) + add_preferred_console("ttyS", 1, NULL); + else if (sclp_has_linemode()) + add_preferred_console("ttyS", 0, NULL); + else + add_preferred_console("hvc", 0, NULL); + } else if (CONSOLE_IS_3215 || CONSOLE_IS_SCLP) + add_preferred_console("ttyS", 0, NULL); + else if (CONSOLE_IS_3270) + add_preferred_console("tty3270", 0, NULL); +} + +static int __init conmode_setup(char *str) +{ +#if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE) + if (strncmp(str, "hwc", 4) == 0 || strncmp(str, "sclp", 5) == 0) + SET_CONSOLE_SCLP; +#endif +#if defined(CONFIG_TN3215_CONSOLE) + if (strncmp(str, "3215", 5) == 0) + SET_CONSOLE_3215; +#endif +#if defined(CONFIG_TN3270_CONSOLE) + if (strncmp(str, "3270", 5) == 0) + SET_CONSOLE_3270; +#endif + set_preferred_console(); + return 1; +} + +__setup("conmode=", conmode_setup); + +static void __init conmode_default(void) +{ + char query_buffer[1024]; + char *ptr; + + if (MACHINE_IS_VM) { + cpcmd("QUERY CONSOLE", query_buffer, 1024, NULL); + console_devno = simple_strtoul(query_buffer + 5, NULL, 16); + ptr = strstr(query_buffer, "SUBCHANNEL ="); + console_irq = simple_strtoul(ptr + 13, NULL, 16); + cpcmd("QUERY TERM", query_buffer, 1024, NULL); + ptr = strstr(query_buffer, "CONMODE"); + /* + * Set the conmode to 3215 so that the device recognition + * will set the cu_type of the console to 3215. If the + * conmode is 3270 and we don't set it back then both + * 3215 and the 3270 driver will try to access the console + * device (3215 as console and 3270 as normal tty). + */ + cpcmd("TERM CONMODE 3215", NULL, 0, NULL); + if (ptr == NULL) { +#if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE) + SET_CONSOLE_SCLP; +#endif + return; + } + if (strncmp(ptr + 8, "3270", 4) == 0) { +#if defined(CONFIG_TN3270_CONSOLE) + SET_CONSOLE_3270; +#elif defined(CONFIG_TN3215_CONSOLE) + SET_CONSOLE_3215; +#elif defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE) + SET_CONSOLE_SCLP; +#endif + } else if (strncmp(ptr + 8, "3215", 4) == 0) { +#if defined(CONFIG_TN3215_CONSOLE) + SET_CONSOLE_3215; +#elif defined(CONFIG_TN3270_CONSOLE) + SET_CONSOLE_3270; +#elif defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE) + SET_CONSOLE_SCLP; +#endif + } + } else { +#if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE) + SET_CONSOLE_SCLP; +#endif + } +} + +#ifdef CONFIG_CRASH_DUMP +static void __init setup_zfcpdump(void) +{ + if (ipl_info.type != IPL_TYPE_FCP_DUMP) + return; + if (OLDMEM_BASE) + return; + strcat(boot_command_line, " cio_ignore=all,!ipldev,!condev"); + console_loglevel = 2; +} +#else +static inline void setup_zfcpdump(void) {} +#endif /* CONFIG_CRASH_DUMP */ + + /* + * Reboot, halt and power_off stubs. They just call _machine_restart, + * _machine_halt or _machine_power_off. + */ + +void machine_restart(char *command) +{ + if ((!in_interrupt() && !in_atomic()) || oops_in_progress) + /* + * Only unblank the console if we are called in enabled + * context or a bust_spinlocks cleared the way for us. + */ + console_unblank(); + _machine_restart(command); +} + +void machine_halt(void) +{ + if (!in_interrupt() || oops_in_progress) + /* + * Only unblank the console if we are called in enabled + * context or a bust_spinlocks cleared the way for us. + */ + console_unblank(); + _machine_halt(); +} + +void machine_power_off(void) +{ + if (!in_interrupt() || oops_in_progress) + /* + * Only unblank the console if we are called in enabled + * context or a bust_spinlocks cleared the way for us. + */ + console_unblank(); + _machine_power_off(); +} + +/* + * Dummy power off function. + */ +void (*pm_power_off)(void) = machine_power_off; +EXPORT_SYMBOL_GPL(pm_power_off); + +static int __init early_parse_mem(char *p) +{ + memory_end = memparse(p, &p); + memory_end &= PAGE_MASK; + memory_end_set = 1; + return 0; +} +early_param("mem", early_parse_mem); + +static int __init parse_vmalloc(char *arg) +{ + if (!arg) + return -EINVAL; + VMALLOC_END = (memparse(arg, &arg) + PAGE_SIZE - 1) & PAGE_MASK; + return 0; +} +early_param("vmalloc", parse_vmalloc); + +void *restart_stack __attribute__((__section__(".data"))); + +static void __init setup_lowcore(void) +{ + struct _lowcore *lc; + + /* + * Setup lowcore for boot cpu + */ + BUILD_BUG_ON(sizeof(struct _lowcore) != LC_PAGES * 4096); + lc = __alloc_bootmem_low(LC_PAGES * PAGE_SIZE, LC_PAGES * PAGE_SIZE, 0); + lc->restart_psw.mask = PSW_KERNEL_BITS; + lc->restart_psw.addr = + PSW_ADDR_AMODE | (unsigned long) restart_int_handler; + lc->external_new_psw.mask = PSW_KERNEL_BITS | + PSW_MASK_DAT | PSW_MASK_MCHECK; + lc->external_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long) ext_int_handler; + lc->svc_new_psw.mask = PSW_KERNEL_BITS | + PSW_MASK_DAT | PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK; + lc->svc_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) system_call; + lc->program_new_psw.mask = PSW_KERNEL_BITS | + PSW_MASK_DAT | PSW_MASK_MCHECK; + lc->program_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long) pgm_check_handler; + lc->mcck_new_psw.mask = PSW_KERNEL_BITS; + lc->mcck_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long) mcck_int_handler; + lc->io_new_psw.mask = PSW_KERNEL_BITS | + PSW_MASK_DAT | PSW_MASK_MCHECK; + lc->io_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) io_int_handler; + lc->clock_comparator = -1ULL; + lc->kernel_stack = ((unsigned long) &init_thread_union) + + THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); + lc->async_stack = (unsigned long) + __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0) + + ASYNC_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); + lc->panic_stack = (unsigned long) + __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0) + + PAGE_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); + lc->current_task = (unsigned long) init_thread_union.thread_info.task; + lc->thread_info = (unsigned long) &init_thread_union; + lc->machine_flags = S390_lowcore.machine_flags; + lc->stfl_fac_list = S390_lowcore.stfl_fac_list; + memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list, + MAX_FACILITY_BIT/8); + if (MACHINE_HAS_VX) + lc->vector_save_area_addr = + (unsigned long) &lc->vector_save_area; + lc->vdso_per_cpu_data = (unsigned long) &lc->paste[0]; + lc->sync_enter_timer = S390_lowcore.sync_enter_timer; + lc->async_enter_timer = S390_lowcore.async_enter_timer; + lc->exit_timer = S390_lowcore.exit_timer; + lc->user_timer = S390_lowcore.user_timer; + lc->system_timer = S390_lowcore.system_timer; + lc->steal_timer = S390_lowcore.steal_timer; + lc->last_update_timer = S390_lowcore.last_update_timer; + lc->last_update_clock = S390_lowcore.last_update_clock; + + restart_stack = __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0); + restart_stack += ASYNC_SIZE; + + /* + * Set up PSW restart to call ipl.c:do_restart(). Copy the relevant + * restart data to the absolute zero lowcore. This is necessary if + * PSW restart is done on an offline CPU that has lowcore zero. + */ + lc->restart_stack = (unsigned long) restart_stack; + lc->restart_fn = (unsigned long) do_restart; + lc->restart_data = 0; + lc->restart_source = -1UL; + + /* Setup absolute zero lowcore */ + mem_assign_absolute(S390_lowcore.restart_stack, lc->restart_stack); + mem_assign_absolute(S390_lowcore.restart_fn, lc->restart_fn); + mem_assign_absolute(S390_lowcore.restart_data, lc->restart_data); + mem_assign_absolute(S390_lowcore.restart_source, lc->restart_source); + mem_assign_absolute(S390_lowcore.restart_psw, lc->restart_psw); + +#ifdef CONFIG_SMP + lc->spinlock_lockval = arch_spin_lockval(0); +#endif + + set_prefix((u32)(unsigned long) lc); + lowcore_ptr[0] = lc; +} + +static struct resource code_resource = { + .name = "Kernel code", + .flags = IORESOURCE_BUSY | IORESOURCE_MEM, +}; + +static struct resource data_resource = { + .name = "Kernel data", + .flags = IORESOURCE_BUSY | IORESOURCE_MEM, +}; + +static struct resource bss_resource = { + .name = "Kernel bss", + .flags = IORESOURCE_BUSY | IORESOURCE_MEM, +}; + +static struct resource __initdata *standard_resources[] = { + &code_resource, + &data_resource, + &bss_resource, +}; + +static void __init setup_resources(void) +{ + struct resource *res, *std_res, *sub_res; + struct memblock_region *reg; + int j; + + code_resource.start = (unsigned long) &_text; + code_resource.end = (unsigned long) &_etext - 1; + data_resource.start = (unsigned long) &_etext; + data_resource.end = (unsigned long) &_edata - 1; + bss_resource.start = (unsigned long) &__bss_start; + bss_resource.end = (unsigned long) &__bss_stop - 1; + + for_each_memblock(memory, reg) { + res = alloc_bootmem_low(sizeof(*res)); + res->flags = IORESOURCE_BUSY | IORESOURCE_MEM; + + res->name = "System RAM"; + res->start = reg->base; + res->end = reg->base + reg->size - 1; + request_resource(&iomem_resource, res); + + for (j = 0; j < ARRAY_SIZE(standard_resources); j++) { + std_res = standard_resources[j]; + if (std_res->start < res->start || + std_res->start > res->end) + continue; + if (std_res->end > res->end) { + sub_res = alloc_bootmem_low(sizeof(*sub_res)); + *sub_res = *std_res; + sub_res->end = res->end; + std_res->start = res->end + 1; + request_resource(res, sub_res); + } else { + request_resource(res, std_res); + } + } + } +} + +static void __init setup_memory_end(void) +{ + unsigned long vmax, vmalloc_size, tmp; + + /* Choose kernel address space layout: 2, 3, or 4 levels. */ + vmalloc_size = VMALLOC_END ?: (128UL << 30) - MODULES_LEN; + tmp = (memory_end ?: max_physmem_end) / PAGE_SIZE; + tmp = tmp * (sizeof(struct page) + PAGE_SIZE); + if (tmp + vmalloc_size + MODULES_LEN <= (1UL << 42)) + vmax = 1UL << 42; /* 3-level kernel page table */ + else + vmax = 1UL << 53; /* 4-level kernel page table */ + /* module area is at the end of the kernel address space. */ + MODULES_END = vmax; + MODULES_VADDR = MODULES_END - MODULES_LEN; + VMALLOC_END = MODULES_VADDR; + VMALLOC_START = vmax - vmalloc_size; + + /* Split remaining virtual space between 1:1 mapping & vmemmap array */ + tmp = VMALLOC_START / (PAGE_SIZE + sizeof(struct page)); + /* vmemmap contains a multiple of PAGES_PER_SECTION struct pages */ + tmp = SECTION_ALIGN_UP(tmp); + tmp = VMALLOC_START - tmp * sizeof(struct page); + tmp &= ~((vmax >> 11) - 1); /* align to page table level */ + tmp = min(tmp, 1UL << MAX_PHYSMEM_BITS); + vmemmap = (struct page *) tmp; + + /* Take care that memory_end is set and <= vmemmap */ + memory_end = min(memory_end ?: max_physmem_end, tmp); + max_pfn = max_low_pfn = PFN_DOWN(memory_end); + memblock_remove(memory_end, ULONG_MAX); + + pr_notice("Max memory size: %luMB\n", memory_end >> 20); +} + +static void __init setup_vmcoreinfo(void) +{ + mem_assign_absolute(S390_lowcore.vmcore_info, paddr_vmcoreinfo_note()); +} + +#ifdef CONFIG_CRASH_DUMP + +/* + * When kdump is enabled, we have to ensure that no memory from + * the area [0 - crashkernel memory size] and + * [crashk_res.start - crashk_res.end] is set offline. + */ +static int kdump_mem_notifier(struct notifier_block *nb, + unsigned long action, void *data) +{ + struct memory_notify *arg = data; + + if (action != MEM_GOING_OFFLINE) + return NOTIFY_OK; + if (arg->start_pfn < PFN_DOWN(resource_size(&crashk_res))) + return NOTIFY_BAD; + if (arg->start_pfn > PFN_DOWN(crashk_res.end)) + return NOTIFY_OK; + if (arg->start_pfn + arg->nr_pages - 1 < PFN_DOWN(crashk_res.start)) + return NOTIFY_OK; + return NOTIFY_BAD; +} + +static struct notifier_block kdump_mem_nb = { + .notifier_call = kdump_mem_notifier, +}; + +#endif + +/* + * Make sure that the area behind memory_end is protected + */ +static void reserve_memory_end(void) +{ +#ifdef CONFIG_CRASH_DUMP + if (ipl_info.type == IPL_TYPE_FCP_DUMP && + !OLDMEM_BASE && sclp_get_hsa_size()) { + memory_end = sclp_get_hsa_size(); + memory_end &= PAGE_MASK; + memory_end_set = 1; + } +#endif + if (!memory_end_set) + return; + memblock_reserve(memory_end, ULONG_MAX); +} + +/* + * Make sure that oldmem, where the dump is stored, is protected + */ +static void reserve_oldmem(void) +{ +#ifdef CONFIG_CRASH_DUMP + if (OLDMEM_BASE) + /* Forget all memory above the running kdump system */ + memblock_reserve(OLDMEM_SIZE, (phys_addr_t)ULONG_MAX); +#endif +} + +/* + * Make sure that oldmem, where the dump is stored, is protected + */ +static void remove_oldmem(void) +{ +#ifdef CONFIG_CRASH_DUMP + if (OLDMEM_BASE) + /* Forget all memory above the running kdump system */ + memblock_remove(OLDMEM_SIZE, (phys_addr_t)ULONG_MAX); +#endif +} + +/* + * Reserve memory for kdump kernel to be loaded with kexec + */ +static void __init reserve_crashkernel(void) +{ +#ifdef CONFIG_CRASH_DUMP + unsigned long long crash_base, crash_size; + phys_addr_t low, high; + int rc; + + rc = parse_crashkernel(boot_command_line, memory_end, &crash_size, + &crash_base); + + crash_base = ALIGN(crash_base, KEXEC_CRASH_MEM_ALIGN); + crash_size = ALIGN(crash_size, KEXEC_CRASH_MEM_ALIGN); + if (rc || crash_size == 0) + return; + + if (memblock.memory.regions[0].size < crash_size) { + pr_info("crashkernel reservation failed: %s\n", + "first memory chunk must be at least crashkernel size"); + return; + } + + low = crash_base ?: OLDMEM_BASE; + high = low + crash_size; + if (low >= OLDMEM_BASE && high <= OLDMEM_BASE + OLDMEM_SIZE) { + /* The crashkernel fits into OLDMEM, reuse OLDMEM */ + crash_base = low; + } else { + /* Find suitable area in free memory */ + low = max_t(unsigned long, crash_size, sclp_get_hsa_size()); + high = crash_base ? crash_base + crash_size : ULONG_MAX; + + if (crash_base && crash_base < low) { + pr_info("crashkernel reservation failed: %s\n", + "crash_base too low"); + return; + } + low = crash_base ?: low; + crash_base = memblock_find_in_range(low, high, crash_size, + KEXEC_CRASH_MEM_ALIGN); + } + + if (!crash_base) { + pr_info("crashkernel reservation failed: %s\n", + "no suitable area found"); + return; + } + + if (register_memory_notifier(&kdump_mem_nb)) + return; + + if (!OLDMEM_BASE && MACHINE_IS_VM) + diag10_range(PFN_DOWN(crash_base), PFN_DOWN(crash_size)); + crashk_res.start = crash_base; + crashk_res.end = crash_base + crash_size - 1; + insert_resource(&iomem_resource, &crashk_res); + memblock_remove(crash_base, crash_size); + pr_info("Reserving %lluMB of memory at %lluMB " + "for crashkernel (System RAM: %luMB)\n", + crash_size >> 20, crash_base >> 20, + (unsigned long)memblock.memory.total_size >> 20); + os_info_crashkernel_add(crash_base, crash_size); +#endif +} + +/* + * Reserve the initrd from being used by memblock + */ +static void __init reserve_initrd(void) +{ +#ifdef CONFIG_BLK_DEV_INITRD + initrd_start = INITRD_START; + initrd_end = initrd_start + INITRD_SIZE; + memblock_reserve(INITRD_START, INITRD_SIZE); +#endif +} + +/* + * Check for initrd being in usable memory + */ +static void __init check_initrd(void) +{ +#ifdef CONFIG_BLK_DEV_INITRD + if (INITRD_START && INITRD_SIZE && + !memblock_is_region_memory(INITRD_START, INITRD_SIZE)) { + pr_err("initrd does not fit memory.\n"); + memblock_free(INITRD_START, INITRD_SIZE); + initrd_start = initrd_end = 0; + } +#endif +} + +/* + * Reserve all kernel text + */ +static void __init reserve_kernel(void) +{ + unsigned long start_pfn; + start_pfn = PFN_UP(__pa(&_end)); + + /* + * Reserve memory used for lowcore/command line/kernel image. + */ + memblock_reserve(0, (unsigned long)_ehead); + memblock_reserve((unsigned long)_stext, PFN_PHYS(start_pfn) + - (unsigned long)_stext); +} + +static void __init reserve_elfcorehdr(void) +{ +#ifdef CONFIG_CRASH_DUMP + if (is_kdump_kernel()) + memblock_reserve(elfcorehdr_addr - OLDMEM_BASE, + PAGE_ALIGN(elfcorehdr_size)); +#endif +} + +static void __init setup_memory(void) +{ + struct memblock_region *reg; + + /* + * Init storage key for present memory + */ + for_each_memblock(memory, reg) { + storage_key_init_range(reg->base, reg->base + reg->size); + } + psw_set_key(PAGE_DEFAULT_KEY); + + /* Only cosmetics */ + memblock_enforce_memory_limit(memblock_end_of_DRAM()); +} + +/* + * Setup hardware capabilities. + */ +static void __init setup_hwcaps(void) +{ + static const int stfl_bits[6] = { 0, 2, 7, 17, 19, 21 }; + struct cpuid cpu_id; + int i; + + /* + * The store facility list bits numbers as found in the principles + * of operation are numbered with bit 1UL<<31 as number 0 to + * bit 1UL<<0 as number 31. + * Bit 0: instructions named N3, "backported" to esa-mode + * Bit 2: z/Architecture mode is active + * Bit 7: the store-facility-list-extended facility is installed + * Bit 17: the message-security assist is installed + * Bit 19: the long-displacement facility is installed + * Bit 21: the extended-immediate facility is installed + * Bit 22: extended-translation facility 3 is installed + * Bit 30: extended-translation facility 3 enhancement facility + * These get translated to: + * HWCAP_S390_ESAN3 bit 0, HWCAP_S390_ZARCH bit 1, + * HWCAP_S390_STFLE bit 2, HWCAP_S390_MSA bit 3, + * HWCAP_S390_LDISP bit 4, HWCAP_S390_EIMM bit 5 and + * HWCAP_S390_ETF3EH bit 8 (22 && 30). + */ + for (i = 0; i < 6; i++) + if (test_facility(stfl_bits[i])) + elf_hwcap |= 1UL << i; + + if (test_facility(22) && test_facility(30)) + elf_hwcap |= HWCAP_S390_ETF3EH; + + /* + * Check for additional facilities with store-facility-list-extended. + * stfle stores doublewords (8 byte) with bit 1ULL<<63 as bit 0 + * and 1ULL<<0 as bit 63. Bits 0-31 contain the same information + * as stored by stfl, bits 32-xxx contain additional facilities. + * How many facility words are stored depends on the number of + * doublewords passed to the instruction. The additional facilities + * are: + * Bit 42: decimal floating point facility is installed + * Bit 44: perform floating point operation facility is installed + * translated to: + * HWCAP_S390_DFP bit 6 (42 && 44). + */ + if ((elf_hwcap & (1UL << 2)) && test_facility(42) && test_facility(44)) + elf_hwcap |= HWCAP_S390_DFP; + + /* + * Huge page support HWCAP_S390_HPAGE is bit 7. + */ + if (MACHINE_HAS_HPAGE) + elf_hwcap |= HWCAP_S390_HPAGE; + + /* + * 64-bit register support for 31-bit processes + * HWCAP_S390_HIGH_GPRS is bit 9. + */ + elf_hwcap |= HWCAP_S390_HIGH_GPRS; + + /* + * Transactional execution support HWCAP_S390_TE is bit 10. + */ + if (test_facility(50) && test_facility(73)) + elf_hwcap |= HWCAP_S390_TE; + + /* + * Vector extension HWCAP_S390_VXRS is bit 11. + */ + if (test_facility(129)) + elf_hwcap |= HWCAP_S390_VXRS; + get_cpu_id(&cpu_id); + add_device_randomness(&cpu_id, sizeof(cpu_id)); + switch (cpu_id.machine) { + case 0x9672: + strcpy(elf_platform, "g5"); + break; + case 0x2064: + case 0x2066: + default: /* Use "z900" as default for 64 bit kernels. */ + strcpy(elf_platform, "z900"); + break; + case 0x2084: + case 0x2086: + strcpy(elf_platform, "z990"); + break; + case 0x2094: + case 0x2096: + strcpy(elf_platform, "z9-109"); + break; + case 0x2097: + case 0x2098: + strcpy(elf_platform, "z10"); + break; + case 0x2817: + case 0x2818: + strcpy(elf_platform, "z196"); + break; + case 0x2827: + case 0x2828: + strcpy(elf_platform, "zEC12"); + break; + case 0x2964: + strcpy(elf_platform, "z13"); + break; + } +} + +/* + * Add system information as device randomness + */ +static void __init setup_randomness(void) +{ + struct sysinfo_3_2_2 *vmms; + + vmms = (struct sysinfo_3_2_2 *) alloc_page(GFP_KERNEL); + if (vmms && stsi(vmms, 3, 2, 2) == 0 && vmms->count) + add_device_randomness(&vmms, vmms->count); + free_page((unsigned long) vmms); +} + +/* + * Setup function called from init/main.c just after the banner + * was printed. + */ + +void __init setup_arch(char **cmdline_p) +{ + /* + * print what head.S has found out about the machine + */ + if (MACHINE_IS_VM) + pr_info("Linux is running as a z/VM " + "guest operating system in 64-bit mode\n"); + else if (MACHINE_IS_KVM) + pr_info("Linux is running under KVM in 64-bit mode\n"); + else if (MACHINE_IS_LPAR) + pr_info("Linux is running natively in 64-bit mode\n"); + + /* Have one command line that is parsed and saved in /proc/cmdline */ + /* boot_command_line has been already set up in early.c */ + *cmdline_p = boot_command_line; + + ROOT_DEV = Root_RAM0; + + /* Is init_mm really needed? */ + init_mm.start_code = PAGE_OFFSET; + init_mm.end_code = (unsigned long) &_etext; + init_mm.end_data = (unsigned long) &_edata; + init_mm.brk = (unsigned long) &_end; + + parse_early_param(); + os_info_init(); + setup_ipl(); + + /* Do some memory reservations *before* memory is added to memblock */ + reserve_memory_end(); + reserve_oldmem(); + reserve_kernel(); + reserve_initrd(); + reserve_elfcorehdr(); + memblock_allow_resize(); + + /* Get information about *all* installed memory */ + detect_memory_memblock(); + + remove_oldmem(); + + /* + * Make sure all chunks are MAX_ORDER aligned so we don't need the + * extra checks that HOLES_IN_ZONE would require. + * + * Is this still required? + */ + memblock_trim_memory(1UL << (MAX_ORDER - 1 + PAGE_SHIFT)); + + setup_memory_end(); + setup_memory(); + + check_initrd(); + reserve_crashkernel(); + + setup_resources(); + setup_vmcoreinfo(); + setup_lowcore(); + smp_fill_possible_mask(); + cpu_init(); + + /* + * Setup capabilities (ELF_HWCAP & ELF_PLATFORM). + */ + setup_hwcaps(); + + /* + * Create kernel page tables and switch to virtual addressing. + */ + paging_init(); + + /* Setup default console */ + conmode_default(); + set_preferred_console(); + + /* Setup zfcpdump support */ + setup_zfcpdump(); + + /* Add system specific data to the random pool */ + setup_randomness(); +} diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c new file mode 100644 index 000000000..c551f22ce --- /dev/null +++ b/arch/s390/kernel/signal.c @@ -0,0 +1,559 @@ +/* + * Copyright IBM Corp. 1999, 2006 + * Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com) + * + * Based on Intel version + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * 1997-11-28 Modified for POSIX.1b signals by Richard Henderson + */ + +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/smp.h> +#include <linux/kernel.h> +#include <linux/signal.h> +#include <linux/errno.h> +#include <linux/wait.h> +#include <linux/ptrace.h> +#include <linux/unistd.h> +#include <linux/stddef.h> +#include <linux/tty.h> +#include <linux/personality.h> +#include <linux/binfmts.h> +#include <linux/tracehook.h> +#include <linux/syscalls.h> +#include <linux/compat.h> +#include <asm/ucontext.h> +#include <asm/uaccess.h> +#include <asm/lowcore.h> +#include <asm/switch_to.h> +#include "entry.h" + +/* + * Layout of an old-style signal-frame: + * ----------------------------------------- + * | save area (_SIGNAL_FRAMESIZE) | + * ----------------------------------------- + * | struct sigcontext | + * | oldmask | + * | _sigregs * | + * ----------------------------------------- + * | _sigregs with | + * | _s390_regs_common | + * | _s390_fp_regs | + * ----------------------------------------- + * | int signo | + * ----------------------------------------- + * | _sigregs_ext with | + * | gprs_high 64 byte (opt) | + * | vxrs_low 128 byte (opt) | + * | vxrs_high 256 byte (opt) | + * | reserved 128 byte (opt) | + * ----------------------------------------- + * | __u16 svc_insn | + * ----------------------------------------- + * The svc_insn entry with the sigreturn system call opcode does not + * have a fixed position and moves if gprs_high or vxrs exist. + * Future extensions will be added to _sigregs_ext. + */ +struct sigframe +{ + __u8 callee_used_stack[__SIGNAL_FRAMESIZE]; + struct sigcontext sc; + _sigregs sregs; + int signo; + _sigregs_ext sregs_ext; + __u16 svc_insn; /* Offset of svc_insn is NOT fixed! */ +}; + +/* + * Layout of an rt signal-frame: + * ----------------------------------------- + * | save area (_SIGNAL_FRAMESIZE) | + * ----------------------------------------- + * | svc __NR_rt_sigreturn 2 byte | + * ----------------------------------------- + * | struct siginfo | + * ----------------------------------------- + * | struct ucontext_extended with | + * | unsigned long uc_flags | + * | struct ucontext *uc_link | + * | stack_t uc_stack | + * | _sigregs uc_mcontext with | + * | _s390_regs_common | + * | _s390_fp_regs | + * | sigset_t uc_sigmask | + * | _sigregs_ext uc_mcontext_ext | + * | gprs_high 64 byte (opt) | + * | vxrs_low 128 byte (opt) | + * | vxrs_high 256 byte (opt)| + * | reserved 128 byte (opt) | + * ----------------------------------------- + * Future extensions will be added to _sigregs_ext. + */ +struct rt_sigframe +{ + __u8 callee_used_stack[__SIGNAL_FRAMESIZE]; + __u16 svc_insn; + struct siginfo info; + struct ucontext_extended uc; +}; + +/* Store registers needed to create the signal frame */ +static void store_sigregs(void) +{ + save_access_regs(current->thread.acrs); + save_fp_ctl(¤t->thread.fp_regs.fpc); + if (current->thread.vxrs) { + int i; + + save_vx_regs(current->thread.vxrs); + for (i = 0; i < __NUM_FPRS; i++) + current->thread.fp_regs.fprs[i] = + *(freg_t *)(current->thread.vxrs + i); + } else + save_fp_regs(current->thread.fp_regs.fprs); +} + +/* Load registers after signal return */ +static void load_sigregs(void) +{ + restore_access_regs(current->thread.acrs); + /* restore_fp_ctl is done in restore_sigregs */ + if (current->thread.vxrs) { + int i; + + for (i = 0; i < __NUM_FPRS; i++) + *(freg_t *)(current->thread.vxrs + i) = + current->thread.fp_regs.fprs[i]; + restore_vx_regs(current->thread.vxrs); + } else + restore_fp_regs(current->thread.fp_regs.fprs); +} + +/* Returns non-zero on fault. */ +static int save_sigregs(struct pt_regs *regs, _sigregs __user *sregs) +{ + _sigregs user_sregs; + + /* Copy a 'clean' PSW mask to the user to avoid leaking + information about whether PER is currently on. */ + user_sregs.regs.psw.mask = PSW_USER_BITS | + (regs->psw.mask & (PSW_MASK_USER | PSW_MASK_RI)); + user_sregs.regs.psw.addr = regs->psw.addr; + memcpy(&user_sregs.regs.gprs, ®s->gprs, sizeof(sregs->regs.gprs)); + memcpy(&user_sregs.regs.acrs, current->thread.acrs, + sizeof(user_sregs.regs.acrs)); + memcpy(&user_sregs.fpregs, ¤t->thread.fp_regs, + sizeof(user_sregs.fpregs)); + if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs))) + return -EFAULT; + return 0; +} + +static int restore_sigregs(struct pt_regs *regs, _sigregs __user *sregs) +{ + _sigregs user_sregs; + + /* Alwys make any pending restarted system call return -EINTR */ + current->restart_block.fn = do_no_restart_syscall; + + if (__copy_from_user(&user_sregs, sregs, sizeof(user_sregs))) + return -EFAULT; + + if (!is_ri_task(current) && (user_sregs.regs.psw.mask & PSW_MASK_RI)) + return -EINVAL; + + /* Loading the floating-point-control word can fail. Do that first. */ + if (restore_fp_ctl(&user_sregs.fpregs.fpc)) + return -EINVAL; + + /* Use regs->psw.mask instead of PSW_USER_BITS to preserve PER bit. */ + regs->psw.mask = (regs->psw.mask & ~(PSW_MASK_USER | PSW_MASK_RI)) | + (user_sregs.regs.psw.mask & (PSW_MASK_USER | PSW_MASK_RI)); + /* Check for invalid user address space control. */ + if ((regs->psw.mask & PSW_MASK_ASC) == PSW_ASC_HOME) + regs->psw.mask = PSW_ASC_PRIMARY | + (regs->psw.mask & ~PSW_MASK_ASC); + /* Check for invalid amode */ + if (regs->psw.mask & PSW_MASK_EA) + regs->psw.mask |= PSW_MASK_BA; + regs->psw.addr = user_sregs.regs.psw.addr; + memcpy(®s->gprs, &user_sregs.regs.gprs, sizeof(sregs->regs.gprs)); + memcpy(¤t->thread.acrs, &user_sregs.regs.acrs, + sizeof(current->thread.acrs)); + + memcpy(¤t->thread.fp_regs, &user_sregs.fpregs, + sizeof(current->thread.fp_regs)); + + clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */ + return 0; +} + +/* Returns non-zero on fault. */ +static int save_sigregs_ext(struct pt_regs *regs, + _sigregs_ext __user *sregs_ext) +{ + __u64 vxrs[__NUM_VXRS_LOW]; + int i; + + /* Save vector registers to signal stack */ + if (current->thread.vxrs) { + for (i = 0; i < __NUM_VXRS_LOW; i++) + vxrs[i] = *((__u64 *)(current->thread.vxrs + i) + 1); + if (__copy_to_user(&sregs_ext->vxrs_low, vxrs, + sizeof(sregs_ext->vxrs_low)) || + __copy_to_user(&sregs_ext->vxrs_high, + current->thread.vxrs + __NUM_VXRS_LOW, + sizeof(sregs_ext->vxrs_high))) + return -EFAULT; + } + return 0; +} + +static int restore_sigregs_ext(struct pt_regs *regs, + _sigregs_ext __user *sregs_ext) +{ + __u64 vxrs[__NUM_VXRS_LOW]; + int i; + + /* Restore vector registers from signal stack */ + if (current->thread.vxrs) { + if (__copy_from_user(vxrs, &sregs_ext->vxrs_low, + sizeof(sregs_ext->vxrs_low)) || + __copy_from_user(current->thread.vxrs + __NUM_VXRS_LOW, + &sregs_ext->vxrs_high, + sizeof(sregs_ext->vxrs_high))) + return -EFAULT; + for (i = 0; i < __NUM_VXRS_LOW; i++) + *((__u64 *)(current->thread.vxrs + i) + 1) = vxrs[i]; + } + return 0; +} + +SYSCALL_DEFINE0(sigreturn) +{ + struct pt_regs *regs = task_pt_regs(current); + struct sigframe __user *frame = + (struct sigframe __user *) regs->gprs[15]; + sigset_t set; + + if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE)) + goto badframe; + set_current_blocked(&set); + if (restore_sigregs(regs, &frame->sregs)) + goto badframe; + if (restore_sigregs_ext(regs, &frame->sregs_ext)) + goto badframe; + load_sigregs(); + return regs->gprs[2]; +badframe: + force_sig(SIGSEGV, current); + return 0; +} + +SYSCALL_DEFINE0(rt_sigreturn) +{ + struct pt_regs *regs = task_pt_regs(current); + struct rt_sigframe __user *frame = + (struct rt_sigframe __user *)regs->gprs[15]; + sigset_t set; + + if (__copy_from_user(&set.sig, &frame->uc.uc_sigmask, sizeof(set))) + goto badframe; + set_current_blocked(&set); + if (restore_altstack(&frame->uc.uc_stack)) + goto badframe; + if (restore_sigregs(regs, &frame->uc.uc_mcontext)) + goto badframe; + if (restore_sigregs_ext(regs, &frame->uc.uc_mcontext_ext)) + goto badframe; + load_sigregs(); + return regs->gprs[2]; +badframe: + force_sig(SIGSEGV, current); + return 0; +} + +/* + * Determine which stack to use.. + */ +static inline void __user * +get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size) +{ + unsigned long sp; + + /* Default to using normal stack */ + sp = regs->gprs[15]; + + /* Overflow on alternate signal stack gives SIGSEGV. */ + if (on_sig_stack(sp) && !on_sig_stack((sp - frame_size) & -8UL)) + return (void __user *) -1UL; + + /* This is the X/Open sanctioned signal stack switching. */ + if (ka->sa.sa_flags & SA_ONSTACK) { + if (! sas_ss_flags(sp)) + sp = current->sas_ss_sp + current->sas_ss_size; + } + + return (void __user *)((sp - frame_size) & -8ul); +} + +static int setup_frame(int sig, struct k_sigaction *ka, + sigset_t *set, struct pt_regs * regs) +{ + struct sigframe __user *frame; + struct sigcontext sc; + unsigned long restorer; + size_t frame_size; + + /* + * gprs_high are only present for a 31-bit task running on + * a 64-bit kernel (see compat_signal.c) but the space for + * gprs_high need to be allocated if vector registers are + * included in the signal frame on a 31-bit system. + */ + frame_size = sizeof(*frame) - sizeof(frame->sregs_ext); + if (MACHINE_HAS_VX) + frame_size += sizeof(frame->sregs_ext); + frame = get_sigframe(ka, regs, frame_size); + if (frame == (void __user *) -1UL) + return -EFAULT; + + /* Set up backchain. */ + if (__put_user(regs->gprs[15], (addr_t __user *) frame)) + return -EFAULT; + + /* Create struct sigcontext on the signal stack */ + memcpy(&sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE); + sc.sregs = (_sigregs __user __force *) &frame->sregs; + if (__copy_to_user(&frame->sc, &sc, sizeof(frame->sc))) + return -EFAULT; + + /* Store registers needed to create the signal frame */ + store_sigregs(); + + /* Create _sigregs on the signal stack */ + if (save_sigregs(regs, &frame->sregs)) + return -EFAULT; + + /* Place signal number on stack to allow backtrace from handler. */ + if (__put_user(regs->gprs[2], (int __user *) &frame->signo)) + return -EFAULT; + + /* Create _sigregs_ext on the signal stack */ + if (save_sigregs_ext(regs, &frame->sregs_ext)) + return -EFAULT; + + /* Set up to return from userspace. If provided, use a stub + already in userspace. */ + if (ka->sa.sa_flags & SA_RESTORER) { + restorer = (unsigned long) ka->sa.sa_restorer | PSW_ADDR_AMODE; + } else { + /* Signal frame without vector registers are short ! */ + __u16 __user *svc = (void __user *) frame + frame_size - 2; + if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn, svc)) + return -EFAULT; + restorer = (unsigned long) svc | PSW_ADDR_AMODE; + } + + /* Set up registers for signal handler */ + regs->gprs[14] = restorer; + regs->gprs[15] = (unsigned long) frame; + /* Force default amode and default user address space control. */ + regs->psw.mask = PSW_MASK_EA | PSW_MASK_BA | + (PSW_USER_BITS & PSW_MASK_ASC) | + (regs->psw.mask & ~PSW_MASK_ASC); + regs->psw.addr = (unsigned long) ka->sa.sa_handler | PSW_ADDR_AMODE; + + regs->gprs[2] = sig; + regs->gprs[3] = (unsigned long) &frame->sc; + + /* We forgot to include these in the sigcontext. + To avoid breaking binary compatibility, they are passed as args. */ + if (sig == SIGSEGV || sig == SIGBUS || sig == SIGILL || + sig == SIGTRAP || sig == SIGFPE) { + /* set extra registers only for synchronous signals */ + regs->gprs[4] = regs->int_code & 127; + regs->gprs[5] = regs->int_parm_long; + regs->gprs[6] = task_thread_info(current)->last_break; + } + return 0; +} + +static int setup_rt_frame(struct ksignal *ksig, sigset_t *set, + struct pt_regs *regs) +{ + struct rt_sigframe __user *frame; + unsigned long uc_flags, restorer; + size_t frame_size; + + frame_size = sizeof(struct rt_sigframe) - sizeof(_sigregs_ext); + /* + * gprs_high are only present for a 31-bit task running on + * a 64-bit kernel (see compat_signal.c) but the space for + * gprs_high need to be allocated if vector registers are + * included in the signal frame on a 31-bit system. + */ + uc_flags = 0; + if (MACHINE_HAS_VX) { + frame_size += sizeof(_sigregs_ext); + if (current->thread.vxrs) + uc_flags |= UC_VXRS; + } + frame = get_sigframe(&ksig->ka, regs, frame_size); + if (frame == (void __user *) -1UL) + return -EFAULT; + + /* Set up backchain. */ + if (__put_user(regs->gprs[15], (addr_t __user *) frame)) + return -EFAULT; + + /* Set up to return from userspace. If provided, use a stub + already in userspace. */ + if (ksig->ka.sa.sa_flags & SA_RESTORER) { + restorer = (unsigned long) + ksig->ka.sa.sa_restorer | PSW_ADDR_AMODE; + } else { + __u16 __user *svc = &frame->svc_insn; + if (__put_user(S390_SYSCALL_OPCODE | __NR_rt_sigreturn, svc)) + return -EFAULT; + restorer = (unsigned long) svc | PSW_ADDR_AMODE; + } + + /* Create siginfo on the signal stack */ + if (copy_siginfo_to_user(&frame->info, &ksig->info)) + return -EFAULT; + + /* Store registers needed to create the signal frame */ + store_sigregs(); + + /* Create ucontext on the signal stack. */ + if (__put_user(uc_flags, &frame->uc.uc_flags) || + __put_user(NULL, &frame->uc.uc_link) || + __save_altstack(&frame->uc.uc_stack, regs->gprs[15]) || + save_sigregs(regs, &frame->uc.uc_mcontext) || + __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)) || + save_sigregs_ext(regs, &frame->uc.uc_mcontext_ext)) + return -EFAULT; + + /* Set up registers for signal handler */ + regs->gprs[14] = restorer; + regs->gprs[15] = (unsigned long) frame; + /* Force default amode and default user address space control. */ + regs->psw.mask = PSW_MASK_EA | PSW_MASK_BA | + (PSW_USER_BITS & PSW_MASK_ASC) | + (regs->psw.mask & ~PSW_MASK_ASC); + regs->psw.addr = (unsigned long) ksig->ka.sa.sa_handler | PSW_ADDR_AMODE; + + regs->gprs[2] = ksig->sig; + regs->gprs[3] = (unsigned long) &frame->info; + regs->gprs[4] = (unsigned long) &frame->uc; + regs->gprs[5] = task_thread_info(current)->last_break; + return 0; +} + +static void handle_signal(struct ksignal *ksig, sigset_t *oldset, + struct pt_regs *regs) +{ + int ret; + + /* Set up the stack frame */ + if (ksig->ka.sa.sa_flags & SA_SIGINFO) + ret = setup_rt_frame(ksig, oldset, regs); + else + ret = setup_frame(ksig->sig, &ksig->ka, oldset, regs); + + signal_setup_done(ret, ksig, test_thread_flag(TIF_SINGLE_STEP)); +} + +/* + * Note that 'init' is a special process: it doesn't get signals it doesn't + * want to handle. Thus you cannot kill init even with a SIGKILL even by + * mistake. + * + * Note that we go through the signals twice: once to check the signals that + * the kernel can handle, and then we build all the user-level signal handling + * stack-frames in one go after that. + */ +void do_signal(struct pt_regs *regs) +{ + struct ksignal ksig; + sigset_t *oldset = sigmask_to_save(); + + /* + * Get signal to deliver. When running under ptrace, at this point + * the debugger may change all our registers, including the system + * call information. + */ + current_thread_info()->system_call = + test_pt_regs_flag(regs, PIF_SYSCALL) ? regs->int_code : 0; + + if (get_signal(&ksig)) { + /* Whee! Actually deliver the signal. */ + if (current_thread_info()->system_call) { + regs->int_code = current_thread_info()->system_call; + /* Check for system call restarting. */ + switch (regs->gprs[2]) { + case -ERESTART_RESTARTBLOCK: + case -ERESTARTNOHAND: + regs->gprs[2] = -EINTR; + break; + case -ERESTARTSYS: + if (!(ksig.ka.sa.sa_flags & SA_RESTART)) { + regs->gprs[2] = -EINTR; + break; + } + /* fallthrough */ + case -ERESTARTNOINTR: + regs->gprs[2] = regs->orig_gpr2; + regs->psw.addr = + __rewind_psw(regs->psw, + regs->int_code >> 16); + break; + } + } + /* No longer in a system call */ + clear_pt_regs_flag(regs, PIF_SYSCALL); + + if (is_compat_task()) + handle_signal32(&ksig, oldset, regs); + else + handle_signal(&ksig, oldset, regs); + return; + } + + /* No handlers present - check for system call restart */ + clear_pt_regs_flag(regs, PIF_SYSCALL); + if (current_thread_info()->system_call) { + regs->int_code = current_thread_info()->system_call; + switch (regs->gprs[2]) { + case -ERESTART_RESTARTBLOCK: + /* Restart with sys_restart_syscall */ + regs->int_code = __NR_restart_syscall; + /* fallthrough */ + case -ERESTARTNOHAND: + case -ERESTARTSYS: + case -ERESTARTNOINTR: + /* Restart system call with magic TIF bit. */ + regs->gprs[2] = regs->orig_gpr2; + set_pt_regs_flag(regs, PIF_SYSCALL); + if (test_thread_flag(TIF_SINGLE_STEP)) + clear_pt_regs_flag(regs, PIF_PER_TRAP); + break; + } + } + + /* + * If there's no signal to deliver, we just put the saved sigmask back. + */ + restore_saved_sigmask(); +} + +void do_notify_resume(struct pt_regs *regs) +{ + clear_thread_flag(TIF_NOTIFY_RESUME); + tracehook_notify_resume(regs); +} diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c new file mode 100644 index 000000000..efd2c1968 --- /dev/null +++ b/arch/s390/kernel/smp.c @@ -0,0 +1,1161 @@ +/* + * SMP related functions + * + * Copyright IBM Corp. 1999, 2012 + * Author(s): Denis Joseph Barrow, + * Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Heiko Carstens <heiko.carstens@de.ibm.com>, + * + * based on other smp stuff by + * (c) 1995 Alan Cox, CymruNET Ltd <alan@cymru.net> + * (c) 1998 Ingo Molnar + * + * The code outside of smp.c uses logical cpu numbers, only smp.c does + * the translation of logical to physical cpu ids. All new code that + * operates on physical cpu numbers needs to go into smp.c. + */ + +#define KMSG_COMPONENT "cpu" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/workqueue.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/mm.h> +#include <linux/err.h> +#include <linux/spinlock.h> +#include <linux/kernel_stat.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/irqflags.h> +#include <linux/cpu.h> +#include <linux/slab.h> +#include <linux/crash_dump.h> +#include <asm/asm-offsets.h> +#include <asm/switch_to.h> +#include <asm/facility.h> +#include <asm/ipl.h> +#include <asm/setup.h> +#include <asm/irq.h> +#include <asm/tlbflush.h> +#include <asm/vtimer.h> +#include <asm/lowcore.h> +#include <asm/sclp.h> +#include <asm/vdso.h> +#include <asm/debug.h> +#include <asm/os_info.h> +#include <asm/sigp.h> +#include <asm/idle.h> +#include "entry.h" + +enum { + ec_schedule = 0, + ec_call_function_single, + ec_stop_cpu, +}; + +enum { + CPU_STATE_STANDBY, + CPU_STATE_CONFIGURED, +}; + +static DEFINE_PER_CPU(struct cpu *, cpu_device); + +struct pcpu { + struct _lowcore *lowcore; /* lowcore page(s) for the cpu */ + unsigned long ec_mask; /* bit mask for ec_xxx functions */ + signed char state; /* physical cpu state */ + signed char polarization; /* physical polarization */ + u16 address; /* physical cpu address */ +}; + +static u8 boot_cpu_type; +static struct pcpu pcpu_devices[NR_CPUS]; + +unsigned int smp_cpu_mt_shift; +EXPORT_SYMBOL(smp_cpu_mt_shift); + +unsigned int smp_cpu_mtid; +EXPORT_SYMBOL(smp_cpu_mtid); + +static unsigned int smp_max_threads __initdata = -1U; + +static int __init early_nosmt(char *s) +{ + smp_max_threads = 1; + return 0; +} +early_param("nosmt", early_nosmt); + +static int __init early_smt(char *s) +{ + get_option(&s, &smp_max_threads); + return 0; +} +early_param("smt", early_smt); + +/* + * The smp_cpu_state_mutex must be held when changing the state or polarization + * member of a pcpu data structure within the pcpu_devices arreay. + */ +DEFINE_MUTEX(smp_cpu_state_mutex); + +/* + * Signal processor helper functions. + */ +static inline int __pcpu_sigp_relax(u16 addr, u8 order, unsigned long parm, + u32 *status) +{ + int cc; + + while (1) { + cc = __pcpu_sigp(addr, order, parm, NULL); + if (cc != SIGP_CC_BUSY) + return cc; + cpu_relax(); + } +} + +static int pcpu_sigp_retry(struct pcpu *pcpu, u8 order, u32 parm) +{ + int cc, retry; + + for (retry = 0; ; retry++) { + cc = __pcpu_sigp(pcpu->address, order, parm, NULL); + if (cc != SIGP_CC_BUSY) + break; + if (retry >= 3) + udelay(10); + } + return cc; +} + +static inline int pcpu_stopped(struct pcpu *pcpu) +{ + u32 uninitialized_var(status); + + if (__pcpu_sigp(pcpu->address, SIGP_SENSE, + 0, &status) != SIGP_CC_STATUS_STORED) + return 0; + return !!(status & (SIGP_STATUS_CHECK_STOP|SIGP_STATUS_STOPPED)); +} + +static inline int pcpu_running(struct pcpu *pcpu) +{ + if (__pcpu_sigp(pcpu->address, SIGP_SENSE_RUNNING, + 0, NULL) != SIGP_CC_STATUS_STORED) + return 1; + /* Status stored condition code is equivalent to cpu not running. */ + return 0; +} + +/* + * Find struct pcpu by cpu address. + */ +static struct pcpu *pcpu_find_address(const struct cpumask *mask, u16 address) +{ + int cpu; + + for_each_cpu(cpu, mask) + if (pcpu_devices[cpu].address == address) + return pcpu_devices + cpu; + return NULL; +} + +static void pcpu_ec_call(struct pcpu *pcpu, int ec_bit) +{ + int order; + + if (test_and_set_bit(ec_bit, &pcpu->ec_mask)) + return; + order = pcpu_running(pcpu) ? SIGP_EXTERNAL_CALL : SIGP_EMERGENCY_SIGNAL; + pcpu_sigp_retry(pcpu, order, 0); +} + +#define ASYNC_FRAME_OFFSET (ASYNC_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE) +#define PANIC_FRAME_OFFSET (PAGE_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE) + +static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu) +{ + unsigned long async_stack, panic_stack; + struct _lowcore *lc; + + if (pcpu != &pcpu_devices[0]) { + pcpu->lowcore = (struct _lowcore *) + __get_free_pages(GFP_KERNEL | GFP_DMA, LC_ORDER); + async_stack = __get_free_pages(GFP_KERNEL, ASYNC_ORDER); + panic_stack = __get_free_page(GFP_KERNEL); + if (!pcpu->lowcore || !panic_stack || !async_stack) + goto out; + } else { + async_stack = pcpu->lowcore->async_stack - ASYNC_FRAME_OFFSET; + panic_stack = pcpu->lowcore->panic_stack - PANIC_FRAME_OFFSET; + } + lc = pcpu->lowcore; + memcpy(lc, &S390_lowcore, 512); + memset((char *) lc + 512, 0, sizeof(*lc) - 512); + lc->async_stack = async_stack + ASYNC_FRAME_OFFSET; + lc->panic_stack = panic_stack + PANIC_FRAME_OFFSET; + lc->cpu_nr = cpu; + lc->spinlock_lockval = arch_spin_lockval(cpu); + if (MACHINE_HAS_VX) + lc->vector_save_area_addr = + (unsigned long) &lc->vector_save_area; + if (vdso_alloc_per_cpu(lc)) + goto out; + lowcore_ptr[cpu] = lc; + pcpu_sigp_retry(pcpu, SIGP_SET_PREFIX, (u32)(unsigned long) lc); + return 0; +out: + if (pcpu != &pcpu_devices[0]) { + free_page(panic_stack); + free_pages(async_stack, ASYNC_ORDER); + free_pages((unsigned long) pcpu->lowcore, LC_ORDER); + } + return -ENOMEM; +} + +#ifdef CONFIG_HOTPLUG_CPU + +static void pcpu_free_lowcore(struct pcpu *pcpu) +{ + pcpu_sigp_retry(pcpu, SIGP_SET_PREFIX, 0); + lowcore_ptr[pcpu - pcpu_devices] = NULL; + vdso_free_per_cpu(pcpu->lowcore); + if (pcpu == &pcpu_devices[0]) + return; + free_page(pcpu->lowcore->panic_stack-PANIC_FRAME_OFFSET); + free_pages(pcpu->lowcore->async_stack-ASYNC_FRAME_OFFSET, ASYNC_ORDER); + free_pages((unsigned long) pcpu->lowcore, LC_ORDER); +} + +#endif /* CONFIG_HOTPLUG_CPU */ + +static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu) +{ + struct _lowcore *lc = pcpu->lowcore; + + if (MACHINE_HAS_TLB_LC) + cpumask_set_cpu(cpu, &init_mm.context.cpu_attach_mask); + cpumask_set_cpu(cpu, mm_cpumask(&init_mm)); + atomic_inc(&init_mm.context.attach_count); + lc->cpu_nr = cpu; + lc->spinlock_lockval = arch_spin_lockval(cpu); + lc->percpu_offset = __per_cpu_offset[cpu]; + lc->kernel_asce = S390_lowcore.kernel_asce; + lc->machine_flags = S390_lowcore.machine_flags; + lc->user_timer = lc->system_timer = lc->steal_timer = 0; + __ctl_store(lc->cregs_save_area, 0, 15); + save_access_regs((unsigned int *) lc->access_regs_save_area); + memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list, + MAX_FACILITY_BIT/8); +} + +static void pcpu_attach_task(struct pcpu *pcpu, struct task_struct *tsk) +{ + struct _lowcore *lc = pcpu->lowcore; + struct thread_info *ti = task_thread_info(tsk); + + lc->kernel_stack = (unsigned long) task_stack_page(tsk) + + THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs); + lc->thread_info = (unsigned long) task_thread_info(tsk); + lc->current_task = (unsigned long) tsk; + lc->user_timer = ti->user_timer; + lc->system_timer = ti->system_timer; + lc->steal_timer = 0; +} + +static void pcpu_start_fn(struct pcpu *pcpu, void (*func)(void *), void *data) +{ + struct _lowcore *lc = pcpu->lowcore; + + lc->restart_stack = lc->kernel_stack; + lc->restart_fn = (unsigned long) func; + lc->restart_data = (unsigned long) data; + lc->restart_source = -1UL; + pcpu_sigp_retry(pcpu, SIGP_RESTART, 0); +} + +/* + * Call function via PSW restart on pcpu and stop the current cpu. + */ +static void pcpu_delegate(struct pcpu *pcpu, void (*func)(void *), + void *data, unsigned long stack) +{ + struct _lowcore *lc = lowcore_ptr[pcpu - pcpu_devices]; + unsigned long source_cpu = stap(); + + __load_psw_mask(PSW_KERNEL_BITS); + if (pcpu->address == source_cpu) + func(data); /* should not return */ + /* Stop target cpu (if func returns this stops the current cpu). */ + pcpu_sigp_retry(pcpu, SIGP_STOP, 0); + /* Restart func on the target cpu and stop the current cpu. */ + mem_assign_absolute(lc->restart_stack, stack); + mem_assign_absolute(lc->restart_fn, (unsigned long) func); + mem_assign_absolute(lc->restart_data, (unsigned long) data); + mem_assign_absolute(lc->restart_source, source_cpu); + asm volatile( + "0: sigp 0,%0,%2 # sigp restart to target cpu\n" + " brc 2,0b # busy, try again\n" + "1: sigp 0,%1,%3 # sigp stop to current cpu\n" + " brc 2,1b # busy, try again\n" + : : "d" (pcpu->address), "d" (source_cpu), + "K" (SIGP_RESTART), "K" (SIGP_STOP) + : "0", "1", "cc"); + for (;;) ; +} + +/* + * Enable additional logical cpus for multi-threading. + */ +static int pcpu_set_smt(unsigned int mtid) +{ + register unsigned long reg1 asm ("1") = (unsigned long) mtid; + int cc; + + if (smp_cpu_mtid == mtid) + return 0; + asm volatile( + " sigp %1,0,%2 # sigp set multi-threading\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (cc) : "d" (reg1), "K" (SIGP_SET_MULTI_THREADING) + : "cc"); + if (cc == 0) { + smp_cpu_mtid = mtid; + smp_cpu_mt_shift = 0; + while (smp_cpu_mtid >= (1U << smp_cpu_mt_shift)) + smp_cpu_mt_shift++; + pcpu_devices[0].address = stap(); + } + return cc; +} + +/* + * Call function on an online CPU. + */ +void smp_call_online_cpu(void (*func)(void *), void *data) +{ + struct pcpu *pcpu; + + /* Use the current cpu if it is online. */ + pcpu = pcpu_find_address(cpu_online_mask, stap()); + if (!pcpu) + /* Use the first online cpu. */ + pcpu = pcpu_devices + cpumask_first(cpu_online_mask); + pcpu_delegate(pcpu, func, data, (unsigned long) restart_stack); +} + +/* + * Call function on the ipl CPU. + */ +void smp_call_ipl_cpu(void (*func)(void *), void *data) +{ + pcpu_delegate(&pcpu_devices[0], func, data, + pcpu_devices->lowcore->panic_stack - + PANIC_FRAME_OFFSET + PAGE_SIZE); +} + +int smp_find_processor_id(u16 address) +{ + int cpu; + + for_each_present_cpu(cpu) + if (pcpu_devices[cpu].address == address) + return cpu; + return -1; +} + +int smp_vcpu_scheduled(int cpu) +{ + return pcpu_running(pcpu_devices + cpu); +} + +void smp_yield_cpu(int cpu) +{ + if (MACHINE_HAS_DIAG9C) + asm volatile("diag %0,0,0x9c" + : : "d" (pcpu_devices[cpu].address)); + else if (MACHINE_HAS_DIAG44) + asm volatile("diag 0,0,0x44"); +} + +/* + * Send cpus emergency shutdown signal. This gives the cpus the + * opportunity to complete outstanding interrupts. + */ +static void smp_emergency_stop(cpumask_t *cpumask) +{ + u64 end; + int cpu; + + end = get_tod_clock() + (1000000UL << 12); + for_each_cpu(cpu, cpumask) { + struct pcpu *pcpu = pcpu_devices + cpu; + set_bit(ec_stop_cpu, &pcpu->ec_mask); + while (__pcpu_sigp(pcpu->address, SIGP_EMERGENCY_SIGNAL, + 0, NULL) == SIGP_CC_BUSY && + get_tod_clock() < end) + cpu_relax(); + } + while (get_tod_clock() < end) { + for_each_cpu(cpu, cpumask) + if (pcpu_stopped(pcpu_devices + cpu)) + cpumask_clear_cpu(cpu, cpumask); + if (cpumask_empty(cpumask)) + break; + cpu_relax(); + } +} + +/* + * Stop all cpus but the current one. + */ +void smp_send_stop(void) +{ + cpumask_t cpumask; + int cpu; + + /* Disable all interrupts/machine checks */ + __load_psw_mask(PSW_KERNEL_BITS | PSW_MASK_DAT); + trace_hardirqs_off(); + + debug_set_critical(); + cpumask_copy(&cpumask, cpu_online_mask); + cpumask_clear_cpu(smp_processor_id(), &cpumask); + + if (oops_in_progress) + smp_emergency_stop(&cpumask); + + /* stop all processors */ + for_each_cpu(cpu, &cpumask) { + struct pcpu *pcpu = pcpu_devices + cpu; + pcpu_sigp_retry(pcpu, SIGP_STOP, 0); + while (!pcpu_stopped(pcpu)) + cpu_relax(); + } +} + +/* + * This is the main routine where commands issued by other + * cpus are handled. + */ +static void smp_handle_ext_call(void) +{ + unsigned long bits; + + /* handle bit signal external calls */ + bits = xchg(&pcpu_devices[smp_processor_id()].ec_mask, 0); + if (test_bit(ec_stop_cpu, &bits)) + smp_stop_cpu(); + if (test_bit(ec_schedule, &bits)) + scheduler_ipi(); + if (test_bit(ec_call_function_single, &bits)) + generic_smp_call_function_single_interrupt(); +} + +static void do_ext_call_interrupt(struct ext_code ext_code, + unsigned int param32, unsigned long param64) +{ + inc_irq_stat(ext_code.code == 0x1202 ? IRQEXT_EXC : IRQEXT_EMS); + smp_handle_ext_call(); +} + +void arch_send_call_function_ipi_mask(const struct cpumask *mask) +{ + int cpu; + + for_each_cpu(cpu, mask) + pcpu_ec_call(pcpu_devices + cpu, ec_call_function_single); +} + +void arch_send_call_function_single_ipi(int cpu) +{ + pcpu_ec_call(pcpu_devices + cpu, ec_call_function_single); +} + +/* + * this function sends a 'reschedule' IPI to another CPU. + * it goes straight through and wastes no time serializing + * anything. Worst case is that we lose a reschedule ... + */ +void smp_send_reschedule(int cpu) +{ + pcpu_ec_call(pcpu_devices + cpu, ec_schedule); +} + +/* + * parameter area for the set/clear control bit callbacks + */ +struct ec_creg_mask_parms { + unsigned long orval; + unsigned long andval; + int cr; +}; + +/* + * callback for setting/clearing control bits + */ +static void smp_ctl_bit_callback(void *info) +{ + struct ec_creg_mask_parms *pp = info; + unsigned long cregs[16]; + + __ctl_store(cregs, 0, 15); + cregs[pp->cr] = (cregs[pp->cr] & pp->andval) | pp->orval; + __ctl_load(cregs, 0, 15); +} + +/* + * Set a bit in a control register of all cpus + */ +void smp_ctl_set_bit(int cr, int bit) +{ + struct ec_creg_mask_parms parms = { 1UL << bit, -1UL, cr }; + + on_each_cpu(smp_ctl_bit_callback, &parms, 1); +} +EXPORT_SYMBOL(smp_ctl_set_bit); + +/* + * Clear a bit in a control register of all cpus + */ +void smp_ctl_clear_bit(int cr, int bit) +{ + struct ec_creg_mask_parms parms = { 0, ~(1UL << bit), cr }; + + on_each_cpu(smp_ctl_bit_callback, &parms, 1); +} +EXPORT_SYMBOL(smp_ctl_clear_bit); + +#ifdef CONFIG_CRASH_DUMP + +static inline void __smp_store_cpu_state(int cpu, u16 address, int is_boot_cpu) +{ + void *lc = pcpu_devices[0].lowcore; + struct save_area_ext *sa_ext; + unsigned long vx_sa; + + sa_ext = dump_save_area_create(cpu); + if (!sa_ext) + panic("could not allocate memory for save area\n"); + if (is_boot_cpu) { + /* Copy the registers of the boot CPU. */ + copy_oldmem_page(1, (void *) &sa_ext->sa, sizeof(sa_ext->sa), + SAVE_AREA_BASE - PAGE_SIZE, 0); + if (MACHINE_HAS_VX) + save_vx_regs_safe(sa_ext->vx_regs); + return; + } + /* Get the registers of a non-boot cpu. */ + __pcpu_sigp_relax(address, SIGP_STOP_AND_STORE_STATUS, 0, NULL); + memcpy_real(&sa_ext->sa, lc + SAVE_AREA_BASE, sizeof(sa_ext->sa)); + if (!MACHINE_HAS_VX) + return; + /* Get the VX registers */ + vx_sa = __get_free_page(GFP_KERNEL); + if (!vx_sa) + panic("could not allocate memory for VX save area\n"); + __pcpu_sigp_relax(address, SIGP_STORE_ADDITIONAL_STATUS, vx_sa, NULL); + memcpy(sa_ext->vx_regs, (void *) vx_sa, sizeof(sa_ext->vx_regs)); + free_page(vx_sa); +} + +/* + * Collect CPU state of the previous, crashed system. + * There are four cases: + * 1) standard zfcp dump + * condition: OLDMEM_BASE == NULL && ipl_info.type == IPL_TYPE_FCP_DUMP + * The state for all CPUs except the boot CPU needs to be collected + * with sigp stop-and-store-status. The boot CPU state is located in + * the absolute lowcore of the memory stored in the HSA. The zcore code + * will allocate the save area and copy the boot CPU state from the HSA. + * 2) stand-alone kdump for SCSI (zfcp dump with swapped memory) + * condition: OLDMEM_BASE != NULL && ipl_info.type == IPL_TYPE_FCP_DUMP + * The state for all CPUs except the boot CPU needs to be collected + * with sigp stop-and-store-status. The firmware or the boot-loader + * stored the registers of the boot CPU in the absolute lowcore in the + * memory of the old system. + * 3) kdump and the old kernel did not store the CPU state, + * or stand-alone kdump for DASD + * condition: OLDMEM_BASE != NULL && !is_kdump_kernel() + * The state for all CPUs except the boot CPU needs to be collected + * with sigp stop-and-store-status. The kexec code or the boot-loader + * stored the registers of the boot CPU in the memory of the old system. + * 4) kdump and the old kernel stored the CPU state + * condition: OLDMEM_BASE != NULL && is_kdump_kernel() + * The state of all CPUs is stored in ELF sections in the memory of the + * old system. The ELF sections are picked up by the crash_dump code + * via elfcorehdr_addr. + */ +static void __init smp_store_cpu_states(struct sclp_cpu_info *info) +{ + unsigned int cpu, address, i, j; + int is_boot_cpu; + + if (is_kdump_kernel()) + /* Previous system stored the CPU states. Nothing to do. */ + return; + if (!(OLDMEM_BASE || ipl_info.type == IPL_TYPE_FCP_DUMP)) + /* No previous system present, normal boot. */ + return; + /* Set multi-threading state to the previous system. */ + pcpu_set_smt(sclp_get_mtid_prev()); + /* Collect CPU states. */ + cpu = 0; + for (i = 0; i < info->configured; i++) { + /* Skip CPUs with different CPU type. */ + if (info->has_cpu_type && info->cpu[i].type != boot_cpu_type) + continue; + for (j = 0; j <= smp_cpu_mtid; j++, cpu++) { + address = (info->cpu[i].core_id << smp_cpu_mt_shift) + j; + is_boot_cpu = (address == pcpu_devices[0].address); + if (is_boot_cpu && !OLDMEM_BASE) + /* Skip boot CPU for standard zfcp dump. */ + continue; + /* Get state for this CPu. */ + __smp_store_cpu_state(cpu, address, is_boot_cpu); + } + } +} + +int smp_store_status(int cpu) +{ + unsigned long vx_sa; + struct pcpu *pcpu; + + pcpu = pcpu_devices + cpu; + if (__pcpu_sigp_relax(pcpu->address, SIGP_STOP_AND_STORE_STATUS, + 0, NULL) != SIGP_CC_ORDER_CODE_ACCEPTED) + return -EIO; + if (!MACHINE_HAS_VX) + return 0; + vx_sa = __pa(pcpu->lowcore->vector_save_area_addr); + __pcpu_sigp_relax(pcpu->address, SIGP_STORE_ADDITIONAL_STATUS, + vx_sa, NULL); + return 0; +} + +#endif /* CONFIG_CRASH_DUMP */ + +void smp_cpu_set_polarization(int cpu, int val) +{ + pcpu_devices[cpu].polarization = val; +} + +int smp_cpu_get_polarization(int cpu) +{ + return pcpu_devices[cpu].polarization; +} + +static struct sclp_cpu_info *smp_get_cpu_info(void) +{ + static int use_sigp_detection; + struct sclp_cpu_info *info; + int address; + + info = kzalloc(sizeof(*info), GFP_KERNEL); + if (info && (use_sigp_detection || sclp_get_cpu_info(info))) { + use_sigp_detection = 1; + for (address = 0; address <= MAX_CPU_ADDRESS; + address += (1U << smp_cpu_mt_shift)) { + if (__pcpu_sigp_relax(address, SIGP_SENSE, 0, NULL) == + SIGP_CC_NOT_OPERATIONAL) + continue; + info->cpu[info->configured].core_id = + address >> smp_cpu_mt_shift; + info->configured++; + } + info->combined = info->configured; + } + return info; +} + +static int smp_add_present_cpu(int cpu); + +static int __smp_rescan_cpus(struct sclp_cpu_info *info, int sysfs_add) +{ + struct pcpu *pcpu; + cpumask_t avail; + int cpu, nr, i, j; + u16 address; + + nr = 0; + cpumask_xor(&avail, cpu_possible_mask, cpu_present_mask); + cpu = cpumask_first(&avail); + for (i = 0; (i < info->combined) && (cpu < nr_cpu_ids); i++) { + if (info->has_cpu_type && info->cpu[i].type != boot_cpu_type) + continue; + address = info->cpu[i].core_id << smp_cpu_mt_shift; + for (j = 0; j <= smp_cpu_mtid; j++) { + if (pcpu_find_address(cpu_present_mask, address + j)) + continue; + pcpu = pcpu_devices + cpu; + pcpu->address = address + j; + pcpu->state = + (cpu >= info->configured*(smp_cpu_mtid + 1)) ? + CPU_STATE_STANDBY : CPU_STATE_CONFIGURED; + smp_cpu_set_polarization(cpu, POLARIZATION_UNKNOWN); + set_cpu_present(cpu, true); + if (sysfs_add && smp_add_present_cpu(cpu) != 0) + set_cpu_present(cpu, false); + else + nr++; + cpu = cpumask_next(cpu, &avail); + if (cpu >= nr_cpu_ids) + break; + } + } + return nr; +} + +static void __init smp_detect_cpus(void) +{ + unsigned int cpu, mtid, c_cpus, s_cpus; + struct sclp_cpu_info *info; + u16 address; + + /* Get CPU information */ + info = smp_get_cpu_info(); + if (!info) + panic("smp_detect_cpus failed to allocate memory\n"); + + /* Find boot CPU type */ + if (info->has_cpu_type) { + address = stap(); + for (cpu = 0; cpu < info->combined; cpu++) + if (info->cpu[cpu].core_id == address) { + /* The boot cpu dictates the cpu type. */ + boot_cpu_type = info->cpu[cpu].type; + break; + } + if (cpu >= info->combined) + panic("Could not find boot CPU type"); + } + +#ifdef CONFIG_CRASH_DUMP + /* Collect CPU state of previous system */ + smp_store_cpu_states(info); +#endif + + /* Set multi-threading state for the current system */ + mtid = sclp_get_mtid(boot_cpu_type); + mtid = (mtid < smp_max_threads) ? mtid : smp_max_threads - 1; + pcpu_set_smt(mtid); + + /* Print number of CPUs */ + c_cpus = s_cpus = 0; + for (cpu = 0; cpu < info->combined; cpu++) { + if (info->has_cpu_type && info->cpu[cpu].type != boot_cpu_type) + continue; + if (cpu < info->configured) + c_cpus += smp_cpu_mtid + 1; + else + s_cpus += smp_cpu_mtid + 1; + } + pr_info("%d configured CPUs, %d standby CPUs\n", c_cpus, s_cpus); + + /* Add CPUs present at boot */ + get_online_cpus(); + __smp_rescan_cpus(info, 0); + put_online_cpus(); + kfree(info); +} + +/* + * Activate a secondary processor. + */ +static void smp_start_secondary(void *cpuvoid) +{ + S390_lowcore.last_update_clock = get_tod_clock(); + S390_lowcore.restart_stack = (unsigned long) restart_stack; + S390_lowcore.restart_fn = (unsigned long) do_restart; + S390_lowcore.restart_data = 0; + S390_lowcore.restart_source = -1UL; + restore_access_regs(S390_lowcore.access_regs_save_area); + __ctl_load(S390_lowcore.cregs_save_area, 0, 15); + __load_psw_mask(PSW_KERNEL_BITS | PSW_MASK_DAT); + cpu_init(); + preempt_disable(); + init_cpu_timer(); + vtime_init(); + pfault_init(); + notify_cpu_starting(smp_processor_id()); + set_cpu_online(smp_processor_id(), true); + inc_irq_stat(CPU_RST); + local_irq_enable(); + cpu_startup_entry(CPUHP_ONLINE); +} + +/* Upping and downing of CPUs */ +int __cpu_up(unsigned int cpu, struct task_struct *tidle) +{ + struct pcpu *pcpu; + int base, i, rc; + + pcpu = pcpu_devices + cpu; + if (pcpu->state != CPU_STATE_CONFIGURED) + return -EIO; + base = cpu - (cpu % (smp_cpu_mtid + 1)); + for (i = 0; i <= smp_cpu_mtid; i++) { + if (base + i < nr_cpu_ids) + if (cpu_online(base + i)) + break; + } + /* + * If this is the first CPU of the core to get online + * do an initial CPU reset. + */ + if (i > smp_cpu_mtid && + pcpu_sigp_retry(pcpu_devices + base, SIGP_INITIAL_CPU_RESET, 0) != + SIGP_CC_ORDER_CODE_ACCEPTED) + return -EIO; + + rc = pcpu_alloc_lowcore(pcpu, cpu); + if (rc) + return rc; + pcpu_prepare_secondary(pcpu, cpu); + pcpu_attach_task(pcpu, tidle); + pcpu_start_fn(pcpu, smp_start_secondary, NULL); + /* Wait until cpu puts itself in the online & active maps */ + while (!cpu_online(cpu) || !cpu_active(cpu)) + cpu_relax(); + return 0; +} + +static unsigned int setup_possible_cpus __initdata; + +static int __init _setup_possible_cpus(char *s) +{ + get_option(&s, &setup_possible_cpus); + return 0; +} +early_param("possible_cpus", _setup_possible_cpus); + +#ifdef CONFIG_HOTPLUG_CPU + +int __cpu_disable(void) +{ + unsigned long cregs[16]; + + /* Handle possible pending IPIs */ + smp_handle_ext_call(); + set_cpu_online(smp_processor_id(), false); + /* Disable pseudo page faults on this cpu. */ + pfault_fini(); + /* Disable interrupt sources via control register. */ + __ctl_store(cregs, 0, 15); + cregs[0] &= ~0x0000ee70UL; /* disable all external interrupts */ + cregs[6] &= ~0xff000000UL; /* disable all I/O interrupts */ + cregs[14] &= ~0x1f000000UL; /* disable most machine checks */ + __ctl_load(cregs, 0, 15); + clear_cpu_flag(CIF_NOHZ_DELAY); + return 0; +} + +void __cpu_die(unsigned int cpu) +{ + struct pcpu *pcpu; + + /* Wait until target cpu is down */ + pcpu = pcpu_devices + cpu; + while (!pcpu_stopped(pcpu)) + cpu_relax(); + pcpu_free_lowcore(pcpu); + atomic_dec(&init_mm.context.attach_count); + cpumask_clear_cpu(cpu, mm_cpumask(&init_mm)); + if (MACHINE_HAS_TLB_LC) + cpumask_clear_cpu(cpu, &init_mm.context.cpu_attach_mask); +} + +void __noreturn cpu_die(void) +{ + idle_task_exit(); + pcpu_sigp_retry(pcpu_devices + smp_processor_id(), SIGP_STOP, 0); + for (;;) ; +} + +#endif /* CONFIG_HOTPLUG_CPU */ + +void __init smp_fill_possible_mask(void) +{ + unsigned int possible, sclp, cpu; + + sclp = min(smp_max_threads, sclp_get_mtid_max() + 1); + sclp = sclp_get_max_cpu()*sclp ?: nr_cpu_ids; + possible = setup_possible_cpus ?: nr_cpu_ids; + possible = min(possible, sclp); + for (cpu = 0; cpu < possible && cpu < nr_cpu_ids; cpu++) + set_cpu_possible(cpu, true); +} + +void __init smp_prepare_cpus(unsigned int max_cpus) +{ + /* request the 0x1201 emergency signal external interrupt */ + if (register_external_irq(EXT_IRQ_EMERGENCY_SIG, do_ext_call_interrupt)) + panic("Couldn't request external interrupt 0x1201"); + /* request the 0x1202 external call external interrupt */ + if (register_external_irq(EXT_IRQ_EXTERNAL_CALL, do_ext_call_interrupt)) + panic("Couldn't request external interrupt 0x1202"); + smp_detect_cpus(); +} + +void __init smp_prepare_boot_cpu(void) +{ + struct pcpu *pcpu = pcpu_devices; + + pcpu->state = CPU_STATE_CONFIGURED; + pcpu->address = stap(); + pcpu->lowcore = (struct _lowcore *)(unsigned long) store_prefix(); + S390_lowcore.percpu_offset = __per_cpu_offset[0]; + smp_cpu_set_polarization(0, POLARIZATION_UNKNOWN); + set_cpu_present(0, true); + set_cpu_online(0, true); +} + +void __init smp_cpus_done(unsigned int max_cpus) +{ +} + +void __init smp_setup_processor_id(void) +{ + S390_lowcore.cpu_nr = 0; + S390_lowcore.spinlock_lockval = arch_spin_lockval(0); +} + +/* + * the frequency of the profiling timer can be changed + * by writing a multiplier value into /proc/profile. + * + * usually you want to run this on all CPUs ;) + */ +int setup_profiling_timer(unsigned int multiplier) +{ + return 0; +} + +#ifdef CONFIG_HOTPLUG_CPU +static ssize_t cpu_configure_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + ssize_t count; + + mutex_lock(&smp_cpu_state_mutex); + count = sprintf(buf, "%d\n", pcpu_devices[dev->id].state); + mutex_unlock(&smp_cpu_state_mutex); + return count; +} + +static ssize_t cpu_configure_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct pcpu *pcpu; + int cpu, val, rc, i; + char delim; + + if (sscanf(buf, "%d %c", &val, &delim) != 1) + return -EINVAL; + if (val != 0 && val != 1) + return -EINVAL; + get_online_cpus(); + mutex_lock(&smp_cpu_state_mutex); + rc = -EBUSY; + /* disallow configuration changes of online cpus and cpu 0 */ + cpu = dev->id; + cpu -= cpu % (smp_cpu_mtid + 1); + if (cpu == 0) + goto out; + for (i = 0; i <= smp_cpu_mtid; i++) + if (cpu_online(cpu + i)) + goto out; + pcpu = pcpu_devices + cpu; + rc = 0; + switch (val) { + case 0: + if (pcpu->state != CPU_STATE_CONFIGURED) + break; + rc = sclp_cpu_deconfigure(pcpu->address >> smp_cpu_mt_shift); + if (rc) + break; + for (i = 0; i <= smp_cpu_mtid; i++) { + if (cpu + i >= nr_cpu_ids || !cpu_present(cpu + i)) + continue; + pcpu[i].state = CPU_STATE_STANDBY; + smp_cpu_set_polarization(cpu + i, + POLARIZATION_UNKNOWN); + } + topology_expect_change(); + break; + case 1: + if (pcpu->state != CPU_STATE_STANDBY) + break; + rc = sclp_cpu_configure(pcpu->address >> smp_cpu_mt_shift); + if (rc) + break; + for (i = 0; i <= smp_cpu_mtid; i++) { + if (cpu + i >= nr_cpu_ids || !cpu_present(cpu + i)) + continue; + pcpu[i].state = CPU_STATE_CONFIGURED; + smp_cpu_set_polarization(cpu + i, + POLARIZATION_UNKNOWN); + } + topology_expect_change(); + break; + default: + break; + } +out: + mutex_unlock(&smp_cpu_state_mutex); + put_online_cpus(); + return rc ? rc : count; +} +static DEVICE_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store); +#endif /* CONFIG_HOTPLUG_CPU */ + +static ssize_t show_cpu_address(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n", pcpu_devices[dev->id].address); +} +static DEVICE_ATTR(address, 0444, show_cpu_address, NULL); + +static struct attribute *cpu_common_attrs[] = { +#ifdef CONFIG_HOTPLUG_CPU + &dev_attr_configure.attr, +#endif + &dev_attr_address.attr, + NULL, +}; + +static struct attribute_group cpu_common_attr_group = { + .attrs = cpu_common_attrs, +}; + +static struct attribute *cpu_online_attrs[] = { + &dev_attr_idle_count.attr, + &dev_attr_idle_time_us.attr, + NULL, +}; + +static struct attribute_group cpu_online_attr_group = { + .attrs = cpu_online_attrs, +}; + +static int smp_cpu_notify(struct notifier_block *self, unsigned long action, + void *hcpu) +{ + unsigned int cpu = (unsigned int)(long)hcpu; + struct device *s = &per_cpu(cpu_device, cpu)->dev; + int err = 0; + + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_ONLINE: + err = sysfs_create_group(&s->kobj, &cpu_online_attr_group); + break; + case CPU_DEAD: + sysfs_remove_group(&s->kobj, &cpu_online_attr_group); + break; + } + return notifier_from_errno(err); +} + +static int smp_add_present_cpu(int cpu) +{ + struct device *s; + struct cpu *c; + int rc; + + c = kzalloc(sizeof(*c), GFP_KERNEL); + if (!c) + return -ENOMEM; + per_cpu(cpu_device, cpu) = c; + s = &c->dev; + c->hotpluggable = 1; + rc = register_cpu(c, cpu); + if (rc) + goto out; + rc = sysfs_create_group(&s->kobj, &cpu_common_attr_group); + if (rc) + goto out_cpu; + if (cpu_online(cpu)) { + rc = sysfs_create_group(&s->kobj, &cpu_online_attr_group); + if (rc) + goto out_online; + } + rc = topology_cpu_init(c); + if (rc) + goto out_topology; + return 0; + +out_topology: + if (cpu_online(cpu)) + sysfs_remove_group(&s->kobj, &cpu_online_attr_group); +out_online: + sysfs_remove_group(&s->kobj, &cpu_common_attr_group); +out_cpu: +#ifdef CONFIG_HOTPLUG_CPU + unregister_cpu(c); +#endif +out: + return rc; +} + +#ifdef CONFIG_HOTPLUG_CPU + +int __ref smp_rescan_cpus(void) +{ + struct sclp_cpu_info *info; + int nr; + + info = smp_get_cpu_info(); + if (!info) + return -ENOMEM; + get_online_cpus(); + mutex_lock(&smp_cpu_state_mutex); + nr = __smp_rescan_cpus(info, 1); + mutex_unlock(&smp_cpu_state_mutex); + put_online_cpus(); + kfree(info); + if (nr) + topology_schedule_update(); + return 0; +} + +static ssize_t __ref rescan_store(struct device *dev, + struct device_attribute *attr, + const char *buf, + size_t count) +{ + int rc; + + rc = smp_rescan_cpus(); + return rc ? rc : count; +} +static DEVICE_ATTR(rescan, 0200, NULL, rescan_store); +#endif /* CONFIG_HOTPLUG_CPU */ + +static int __init s390_smp_init(void) +{ + int cpu, rc = 0; + +#ifdef CONFIG_HOTPLUG_CPU + rc = device_create_file(cpu_subsys.dev_root, &dev_attr_rescan); + if (rc) + return rc; +#endif + cpu_notifier_register_begin(); + for_each_present_cpu(cpu) { + rc = smp_add_present_cpu(cpu); + if (rc) + goto out; + } + + __hotcpu_notifier(smp_cpu_notify, 0); + +out: + cpu_notifier_register_done(); + return rc; +} +subsys_initcall(s390_smp_init); diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c new file mode 100644 index 000000000..1785cd822 --- /dev/null +++ b/arch/s390/kernel/stacktrace.c @@ -0,0 +1,96 @@ +/* + * Stack trace management functions + * + * Copyright IBM Corp. 2006 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#include <linux/sched.h> +#include <linux/stacktrace.h> +#include <linux/kallsyms.h> +#include <linux/module.h> + +static unsigned long save_context_stack(struct stack_trace *trace, + unsigned long sp, + unsigned long low, + unsigned long high, + int savesched) +{ + struct stack_frame *sf; + struct pt_regs *regs; + unsigned long addr; + + while(1) { + sp &= PSW_ADDR_INSN; + if (sp < low || sp > high) + return sp; + sf = (struct stack_frame *)sp; + while(1) { + addr = sf->gprs[8] & PSW_ADDR_INSN; + if (!trace->skip) + trace->entries[trace->nr_entries++] = addr; + else + trace->skip--; + if (trace->nr_entries >= trace->max_entries) + return sp; + low = sp; + sp = sf->back_chain & PSW_ADDR_INSN; + if (!sp) + break; + if (sp <= low || sp > high - sizeof(*sf)) + return sp; + sf = (struct stack_frame *)sp; + } + /* Zero backchain detected, check for interrupt frame. */ + sp = (unsigned long)(sf + 1); + if (sp <= low || sp > high - sizeof(*regs)) + return sp; + regs = (struct pt_regs *)sp; + addr = regs->psw.addr & PSW_ADDR_INSN; + if (savesched || !in_sched_functions(addr)) { + if (!trace->skip) + trace->entries[trace->nr_entries++] = addr; + else + trace->skip--; + } + if (trace->nr_entries >= trace->max_entries) + return sp; + low = sp; + sp = regs->gprs[15]; + } +} + +void save_stack_trace(struct stack_trace *trace) +{ + register unsigned long sp asm ("15"); + unsigned long orig_sp, new_sp; + + orig_sp = sp & PSW_ADDR_INSN; + new_sp = save_context_stack(trace, orig_sp, + S390_lowcore.panic_stack - PAGE_SIZE, + S390_lowcore.panic_stack, 1); + if (new_sp != orig_sp) + return; + new_sp = save_context_stack(trace, new_sp, + S390_lowcore.async_stack - ASYNC_SIZE, + S390_lowcore.async_stack, 1); + if (new_sp != orig_sp) + return; + save_context_stack(trace, new_sp, + S390_lowcore.thread_info, + S390_lowcore.thread_info + THREAD_SIZE, 1); +} +EXPORT_SYMBOL_GPL(save_stack_trace); + +void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace) +{ + unsigned long sp, low, high; + + sp = tsk->thread.ksp & PSW_ADDR_INSN; + low = (unsigned long) task_stack_page(tsk); + high = (unsigned long) task_pt_regs(tsk); + save_context_stack(trace, sp, low, high, 0); + if (trace->nr_entries < trace->max_entries) + trace->entries[trace->nr_entries++] = ULONG_MAX; +} +EXPORT_SYMBOL_GPL(save_stack_trace_tsk); diff --git a/arch/s390/kernel/suspend.c b/arch/s390/kernel/suspend.c new file mode 100644 index 000000000..d3236c9e2 --- /dev/null +++ b/arch/s390/kernel/suspend.c @@ -0,0 +1,225 @@ +/* + * Suspend support specific for s390. + * + * Copyright IBM Corp. 2009 + * + * Author(s): Hans-Joachim Picht <hans@linux.vnet.ibm.com> + */ + +#include <linux/pfn.h> +#include <linux/suspend.h> +#include <linux/mm.h> +#include <asm/ctl_reg.h> +#include <asm/ipl.h> +#include <asm/cio.h> +#include <asm/pci.h> +#include <asm/sections.h> +#include "entry.h" + +/* + * The restore of the saved pages in an hibernation image will set + * the change and referenced bits in the storage key for each page. + * Overindication of the referenced bits after an hibernation cycle + * does not cause any harm but the overindication of the change bits + * would cause trouble. + * Use the ARCH_SAVE_PAGE_KEYS hooks to save the storage key of each + * page to the most significant byte of the associated page frame + * number in the hibernation image. + */ + +/* + * Key storage is allocated as a linked list of pages. + * The size of the keys array is (PAGE_SIZE - sizeof(long)) + */ +struct page_key_data { + struct page_key_data *next; + unsigned char data[]; +}; + +#define PAGE_KEY_DATA_SIZE (PAGE_SIZE - sizeof(struct page_key_data *)) + +static struct page_key_data *page_key_data; +static struct page_key_data *page_key_rp, *page_key_wp; +static unsigned long page_key_rx, page_key_wx; +unsigned long suspend_zero_pages; + +/* + * For each page in the hibernation image one additional byte is + * stored in the most significant byte of the page frame number. + * On suspend no additional memory is required but on resume the + * keys need to be memorized until the page data has been restored. + * Only then can the storage keys be set to their old state. + */ +unsigned long page_key_additional_pages(unsigned long pages) +{ + return DIV_ROUND_UP(pages, PAGE_KEY_DATA_SIZE); +} + +/* + * Free page_key_data list of arrays. + */ +void page_key_free(void) +{ + struct page_key_data *pkd; + + while (page_key_data) { + pkd = page_key_data; + page_key_data = pkd->next; + free_page((unsigned long) pkd); + } +} + +/* + * Allocate page_key_data list of arrays with enough room to store + * one byte for each page in the hibernation image. + */ +int page_key_alloc(unsigned long pages) +{ + struct page_key_data *pk; + unsigned long size; + + size = DIV_ROUND_UP(pages, PAGE_KEY_DATA_SIZE); + while (size--) { + pk = (struct page_key_data *) get_zeroed_page(GFP_KERNEL); + if (!pk) { + page_key_free(); + return -ENOMEM; + } + pk->next = page_key_data; + page_key_data = pk; + } + page_key_rp = page_key_wp = page_key_data; + page_key_rx = page_key_wx = 0; + return 0; +} + +/* + * Save the storage key into the upper 8 bits of the page frame number. + */ +void page_key_read(unsigned long *pfn) +{ + unsigned long addr; + + addr = (unsigned long) page_address(pfn_to_page(*pfn)); + *(unsigned char *) pfn = (unsigned char) page_get_storage_key(addr); +} + +/* + * Extract the storage key from the upper 8 bits of the page frame number + * and store it in the page_key_data list of arrays. + */ +void page_key_memorize(unsigned long *pfn) +{ + page_key_wp->data[page_key_wx] = *(unsigned char *) pfn; + *(unsigned char *) pfn = 0; + if (++page_key_wx < PAGE_KEY_DATA_SIZE) + return; + page_key_wp = page_key_wp->next; + page_key_wx = 0; +} + +/* + * Get the next key from the page_key_data list of arrays and set the + * storage key of the page referred by @address. If @address refers to + * a "safe" page the swsusp_arch_resume code will transfer the storage + * key from the buffer page to the original page. + */ +void page_key_write(void *address) +{ + page_set_storage_key((unsigned long) address, + page_key_rp->data[page_key_rx], 0); + if (++page_key_rx >= PAGE_KEY_DATA_SIZE) + return; + page_key_rp = page_key_rp->next; + page_key_rx = 0; +} + +int pfn_is_nosave(unsigned long pfn) +{ + unsigned long nosave_begin_pfn = PFN_DOWN(__pa(&__nosave_begin)); + unsigned long nosave_end_pfn = PFN_DOWN(__pa(&__nosave_end)); + unsigned long eshared_pfn = PFN_DOWN(__pa(&_eshared)) - 1; + unsigned long stext_pfn = PFN_DOWN(__pa(&_stext)); + + /* Always save lowcore pages (LC protection might be enabled). */ + if (pfn <= LC_PAGES) + return 0; + if (pfn >= nosave_begin_pfn && pfn < nosave_end_pfn) + return 1; + /* Skip memory holes and read-only pages (NSS, DCSS, ...). */ + if (pfn >= stext_pfn && pfn <= eshared_pfn) + return ipl_info.type == IPL_TYPE_NSS ? 1 : 0; + if (tprot(PFN_PHYS(pfn))) + return 1; + return 0; +} + +/* + * PM notifier callback for suspend + */ +static int suspend_pm_cb(struct notifier_block *nb, unsigned long action, + void *ptr) +{ + switch (action) { + case PM_SUSPEND_PREPARE: + case PM_HIBERNATION_PREPARE: + suspend_zero_pages = __get_free_pages(GFP_KERNEL, LC_ORDER); + if (!suspend_zero_pages) + return NOTIFY_BAD; + break; + case PM_POST_SUSPEND: + case PM_POST_HIBERNATION: + free_pages(suspend_zero_pages, LC_ORDER); + break; + default: + return NOTIFY_DONE; + } + return NOTIFY_OK; +} + +static int __init suspend_pm_init(void) +{ + pm_notifier(suspend_pm_cb, 0); + return 0; +} +arch_initcall(suspend_pm_init); + +void save_processor_state(void) +{ + /* swsusp_arch_suspend() actually saves all cpu register contents. + * Machine checks must be disabled since swsusp_arch_suspend() stores + * register contents to their lowcore save areas. That's the same + * place where register contents on machine checks would be saved. + * To avoid register corruption disable machine checks. + * We must also disable machine checks in the new psw mask for + * program checks, since swsusp_arch_suspend() may generate program + * checks. Disabling machine checks for all other new psw masks is + * just paranoia. + */ + local_mcck_disable(); + /* Disable lowcore protection */ + __ctl_clear_bit(0,28); + S390_lowcore.external_new_psw.mask &= ~PSW_MASK_MCHECK; + S390_lowcore.svc_new_psw.mask &= ~PSW_MASK_MCHECK; + S390_lowcore.io_new_psw.mask &= ~PSW_MASK_MCHECK; + S390_lowcore.program_new_psw.mask &= ~PSW_MASK_MCHECK; +} + +void restore_processor_state(void) +{ + S390_lowcore.external_new_psw.mask |= PSW_MASK_MCHECK; + S390_lowcore.svc_new_psw.mask |= PSW_MASK_MCHECK; + S390_lowcore.io_new_psw.mask |= PSW_MASK_MCHECK; + S390_lowcore.program_new_psw.mask |= PSW_MASK_MCHECK; + /* Enable lowcore protection */ + __ctl_set_bit(0,28); + local_mcck_enable(); +} + +/* Called at the end of swsusp_arch_resume */ +void s390_early_resume(void) +{ + lgr_info_log(); + channel_subsystem_reinit(); + zpci_rescan(); +} diff --git a/arch/s390/kernel/swsusp.S b/arch/s390/kernel/swsusp.S new file mode 100644 index 000000000..ca6294645 --- /dev/null +++ b/arch/s390/kernel/swsusp.S @@ -0,0 +1,317 @@ +/* + * S390 64-bit swsusp implementation + * + * Copyright IBM Corp. 2009 + * + * Author(s): Hans-Joachim Picht <hans@linux.vnet.ibm.com> + * Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#include <linux/linkage.h> +#include <asm/page.h> +#include <asm/ptrace.h> +#include <asm/thread_info.h> +#include <asm/asm-offsets.h> +#include <asm/sigp.h> + +/* + * Save register context in absolute 0 lowcore and call swsusp_save() to + * create in-memory kernel image. The context is saved in the designated + * "store status" memory locations (see POP). + * We return from this function twice. The first time during the suspend to + * disk process. The second time via the swsusp_arch_resume() function + * (see below) in the resume process. + * This function runs with disabled interrupts. + */ + .section .text +ENTRY(swsusp_arch_suspend) + stmg %r6,%r15,__SF_GPRS(%r15) + lgr %r1,%r15 + aghi %r15,-STACK_FRAME_OVERHEAD + stg %r1,__SF_BACKCHAIN(%r15) + + /* Deactivate DAT */ + stnsm __SF_EMPTY(%r15),0xfb + + /* Store prefix register on stack */ + stpx __SF_EMPTY(%r15) + + /* Save prefix register contents for lowcore copy */ + llgf %r10,__SF_EMPTY(%r15) + + /* Get pointer to save area */ + lghi %r1,0x1000 + + /* Save CPU address */ + stap __LC_EXT_CPU_ADDR(%r0) + + /* Store registers */ + mvc 0x318(4,%r1),__SF_EMPTY(%r15) /* move prefix to lowcore */ + stfpc 0x31c(%r1) /* store fpu control */ + std 0,0x200(%r1) /* store f0 */ + std 1,0x208(%r1) /* store f1 */ + std 2,0x210(%r1) /* store f2 */ + std 3,0x218(%r1) /* store f3 */ + std 4,0x220(%r1) /* store f4 */ + std 5,0x228(%r1) /* store f5 */ + std 6,0x230(%r1) /* store f6 */ + std 7,0x238(%r1) /* store f7 */ + std 8,0x240(%r1) /* store f8 */ + std 9,0x248(%r1) /* store f9 */ + std 10,0x250(%r1) /* store f10 */ + std 11,0x258(%r1) /* store f11 */ + std 12,0x260(%r1) /* store f12 */ + std 13,0x268(%r1) /* store f13 */ + std 14,0x270(%r1) /* store f14 */ + std 15,0x278(%r1) /* store f15 */ + stam %a0,%a15,0x340(%r1) /* store access registers */ + stctg %c0,%c15,0x380(%r1) /* store control registers */ + stmg %r0,%r15,0x280(%r1) /* store general registers */ + + stpt 0x328(%r1) /* store timer */ + stck __SF_EMPTY(%r15) /* store clock */ + stckc 0x330(%r1) /* store clock comparator */ + + /* Update cputime accounting before going to sleep */ + lg %r0,__LC_LAST_UPDATE_TIMER + slg %r0,0x328(%r1) + alg %r0,__LC_SYSTEM_TIMER + stg %r0,__LC_SYSTEM_TIMER + mvc __LC_LAST_UPDATE_TIMER(8),0x328(%r1) + lg %r0,__LC_LAST_UPDATE_CLOCK + slg %r0,__SF_EMPTY(%r15) + alg %r0,__LC_STEAL_TIMER + stg %r0,__LC_STEAL_TIMER + mvc __LC_LAST_UPDATE_CLOCK(8),__SF_EMPTY(%r15) + + /* Activate DAT */ + stosm __SF_EMPTY(%r15),0x04 + + /* Set prefix page to zero */ + xc __SF_EMPTY(4,%r15),__SF_EMPTY(%r15) + spx __SF_EMPTY(%r15) + + /* Save absolute zero pages */ + larl %r2,suspend_zero_pages + lg %r2,0(%r2) + lghi %r4,0 + lghi %r3,2*PAGE_SIZE + lghi %r5,2*PAGE_SIZE +1: mvcle %r2,%r4,0 + jo 1b + + /* Copy lowcore to absolute zero lowcore */ + lghi %r2,0 + lgr %r4,%r10 + lghi %r3,2*PAGE_SIZE + lghi %r5,2*PAGE_SIZE +1: mvcle %r2,%r4,0 + jo 1b + + /* Save image */ + brasl %r14,swsusp_save + + /* Restore prefix register and return */ + lghi %r1,0x1000 + spx 0x318(%r1) + lmg %r6,%r15,STACK_FRAME_OVERHEAD + __SF_GPRS(%r15) + lghi %r2,0 + br %r14 + +/* + * Restore saved memory image to correct place and restore register context. + * Then we return to the function that called swsusp_arch_suspend(). + * swsusp_arch_resume() runs with disabled interrupts. + */ +ENTRY(swsusp_arch_resume) + stmg %r6,%r15,__SF_GPRS(%r15) + lgr %r1,%r15 + aghi %r15,-STACK_FRAME_OVERHEAD + stg %r1,__SF_BACKCHAIN(%r15) + + /* Make all free pages stable */ + lghi %r2,1 + brasl %r14,arch_set_page_states + + /* Deactivate DAT */ + stnsm __SF_EMPTY(%r15),0xfb + + /* Set prefix page to zero */ + xc __SF_EMPTY(4,%r15),__SF_EMPTY(%r15) + spx __SF_EMPTY(%r15) + + /* Restore saved image */ + larl %r1,restore_pblist + lg %r1,0(%r1) + ltgr %r1,%r1 + jz 2f +0: + lg %r2,8(%r1) + lg %r4,0(%r1) + iske %r0,%r4 + lghi %r3,PAGE_SIZE + lghi %r5,PAGE_SIZE +1: + mvcle %r2,%r4,0 + jo 1b + lg %r2,8(%r1) + sske %r0,%r2 + lg %r1,16(%r1) + ltgr %r1,%r1 + jnz 0b +2: + ptlb /* flush tlb */ + + /* Reset System */ + larl %r1,restart_entry + larl %r2,.Lrestart_diag308_psw + og %r1,0(%r2) + stg %r1,0(%r0) + larl %r1,.Lnew_pgm_check_psw + epsw %r2,%r3 + stm %r2,%r3,0(%r1) + mvc __LC_PGM_NEW_PSW(16,%r0),0(%r1) + lghi %r0,0 + diag %r0,%r0,0x308 +restart_entry: + lhi %r1,1 + sigp %r1,%r0,SIGP_SET_ARCHITECTURE + sam64 +#ifdef CONFIG_SMP + larl %r1,smp_cpu_mt_shift + icm %r1,15,0(%r1) + jz smt_done + llgfr %r1,%r1 +smt_loop: + sigp %r1,%r0,SIGP_SET_MULTI_THREADING + brc 8,smt_done /* accepted */ + brc 2,smt_loop /* busy, try again */ +smt_done: +#endif + larl %r1,.Lnew_pgm_check_psw + lpswe 0(%r1) +pgm_check_entry: + + /* Switch to original suspend CPU */ + larl %r1,.Lresume_cpu /* Resume CPU address: r2 */ + stap 0(%r1) + llgh %r2,0(%r1) + llgh %r1,__LC_EXT_CPU_ADDR(%r0) /* Suspend CPU address: r1 */ + cgr %r1,%r2 + je restore_registers /* r1 = r2 -> nothing to do */ + larl %r4,.Lrestart_suspend_psw /* Set new restart PSW */ + mvc __LC_RST_NEW_PSW(16,%r0),0(%r4) +3: + sigp %r9,%r1,SIGP_INITIAL_CPU_RESET /* sigp initial cpu reset */ + brc 8,4f /* accepted */ + brc 2,3b /* busy, try again */ + + /* Suspend CPU not available -> panic */ + larl %r15,init_thread_union + ahi %r15,1<<(PAGE_SHIFT+THREAD_ORDER) + larl %r2,.Lpanic_string + larl %r3,_sclp_print_early + lghi %r1,0 + sam31 + sigp %r1,%r0,SIGP_SET_ARCHITECTURE + basr %r14,%r3 + larl %r3,.Ldisabled_wait_31 + lpsw 0(%r3) +4: + /* Switch to suspend CPU */ + sigp %r9,%r1,SIGP_RESTART /* sigp restart to suspend CPU */ + brc 2,4b /* busy, try again */ +5: + sigp %r9,%r2,SIGP_STOP /* sigp stop to current resume CPU */ + brc 2,5b /* busy, try again */ +6: j 6b + +restart_suspend: + larl %r1,.Lresume_cpu + llgh %r2,0(%r1) +7: + sigp %r9,%r2,SIGP_SENSE /* sigp sense, wait for resume CPU */ + brc 8,7b /* accepted, status 0, still running */ + brc 2,7b /* busy, try again */ + tmll %r9,0x40 /* Test if resume CPU is stopped */ + jz 7b + +restore_registers: + /* Restore registers */ + lghi %r13,0x1000 /* %r1 = pointer to save area */ + + /* Ignore time spent in suspended state. */ + llgf %r1,0x318(%r13) + stck __LC_LAST_UPDATE_CLOCK(%r1) + spt 0x328(%r13) /* reprogram timer */ + //sckc 0x330(%r13) /* set clock comparator */ + + lctlg %c0,%c15,0x380(%r13) /* load control registers */ + lam %a0,%a15,0x340(%r13) /* load access registers */ + + lfpc 0x31c(%r13) /* load fpu control */ + ld 0,0x200(%r13) /* load f0 */ + ld 1,0x208(%r13) /* load f1 */ + ld 2,0x210(%r13) /* load f2 */ + ld 3,0x218(%r13) /* load f3 */ + ld 4,0x220(%r13) /* load f4 */ + ld 5,0x228(%r13) /* load f5 */ + ld 6,0x230(%r13) /* load f6 */ + ld 7,0x238(%r13) /* load f7 */ + ld 8,0x240(%r13) /* load f8 */ + ld 9,0x248(%r13) /* load f9 */ + ld 10,0x250(%r13) /* load f10 */ + ld 11,0x258(%r13) /* load f11 */ + ld 12,0x260(%r13) /* load f12 */ + ld 13,0x268(%r13) /* load f13 */ + ld 14,0x270(%r13) /* load f14 */ + ld 15,0x278(%r13) /* load f15 */ + + /* Load old stack */ + lg %r15,0x2f8(%r13) + + /* Save prefix register */ + mvc __SF_EMPTY(4,%r15),0x318(%r13) + + /* Restore absolute zero pages */ + lghi %r2,0 + larl %r4,suspend_zero_pages + lg %r4,0(%r4) + lghi %r3,2*PAGE_SIZE + lghi %r5,2*PAGE_SIZE +1: mvcle %r2,%r4,0 + jo 1b + + /* Restore prefix register */ + spx __SF_EMPTY(%r15) + + /* Activate DAT */ + stosm __SF_EMPTY(%r15),0x04 + + /* Make all free pages unstable */ + lghi %r2,0 + brasl %r14,arch_set_page_states + + /* Call arch specific early resume code */ + brasl %r14,s390_early_resume + + /* Return 0 */ + lmg %r6,%r15,STACK_FRAME_OVERHEAD + __SF_GPRS(%r15) + lghi %r2,0 + br %r14 + + .section .data..nosave,"aw",@progbits + .align 8 +.Ldisabled_wait_31: + .long 0x000a0000,0x00000000 +.Lpanic_string: + .asciz "Resume not possible because suspend CPU is no longer available" + .align 8 +.Lrestart_diag308_psw: + .long 0x00080000,0x80000000 +.Lrestart_suspend_psw: + .quad 0x0000000180000000,restart_suspend +.Lnew_pgm_check_psw: + .quad 0,pgm_check_entry +.Lresume_cpu: + .byte 0,0 diff --git a/arch/s390/kernel/sys_s390.c b/arch/s390/kernel/sys_s390.c new file mode 100644 index 000000000..f145490cc --- /dev/null +++ b/arch/s390/kernel/sys_s390.c @@ -0,0 +1,91 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + * Thomas Spatzier (tspat@de.ibm.com) + * + * Derived from "arch/i386/kernel/sys_i386.c" + * + * This file contains various random system calls that + * have a non-standard calling sequence on the Linux/s390 + * platform. + */ + +#include <linux/errno.h> +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/fs.h> +#include <linux/smp.h> +#include <linux/sem.h> +#include <linux/msg.h> +#include <linux/shm.h> +#include <linux/stat.h> +#include <linux/syscalls.h> +#include <linux/mman.h> +#include <linux/file.h> +#include <linux/utsname.h> +#include <linux/personality.h> +#include <linux/unistd.h> +#include <linux/ipc.h> +#include <asm/uaccess.h> +#include "entry.h" + +/* + * Perform the mmap() system call. Linux for S/390 isn't able to handle more + * than 5 system call parameters, so this system call uses a memory block + * for parameter passing. + */ + +struct s390_mmap_arg_struct { + unsigned long addr; + unsigned long len; + unsigned long prot; + unsigned long flags; + unsigned long fd; + unsigned long offset; +}; + +SYSCALL_DEFINE1(mmap2, struct s390_mmap_arg_struct __user *, arg) +{ + struct s390_mmap_arg_struct a; + int error = -EFAULT; + + if (copy_from_user(&a, arg, sizeof(a))) + goto out; + error = sys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd, a.offset); +out: + return error; +} + +/* + * sys_ipc() is the de-multiplexer for the SysV IPC calls. + */ +SYSCALL_DEFINE5(s390_ipc, uint, call, int, first, unsigned long, second, + unsigned long, third, void __user *, ptr) +{ + if (call >> 16) + return -EINVAL; + /* The s390 sys_ipc variant has only five parameters instead of six + * like the generic variant. The only difference is the handling of + * the SEMTIMEDOP subcall where on s390 the third parameter is used + * as a pointer to a struct timespec where the generic variant uses + * the fifth parameter. + * Therefore we can call the generic variant by simply passing the + * third parameter also as fifth parameter. + */ + return sys_ipc(call, first, second, third, ptr, third); +} + +SYSCALL_DEFINE1(s390_personality, unsigned int, personality) +{ + unsigned int ret; + + if (personality(current->personality) == PER_LINUX32 && + personality(personality) == PER_LINUX) + personality |= PER_LINUX32; + ret = sys_personality(personality); + if (personality(ret) == PER_LINUX32) + ret &= ~PER_LINUX32; + + return ret; +} diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S new file mode 100644 index 000000000..1acad0268 --- /dev/null +++ b/arch/s390/kernel/syscalls.S @@ -0,0 +1,365 @@ +/* + * definitions for sys_call_table, each line represents an + * entry in the table in the form + * SYSCALL(64 bit syscall, 31 bit emulated syscall) + * + * this file is meant to be included from entry.S + */ + +#define NI_SYSCALL SYSCALL(sys_ni_syscall,sys_ni_syscall) + +NI_SYSCALL /* 0 */ +SYSCALL(sys_exit,compat_sys_exit) +SYSCALL(sys_fork,sys_fork) +SYSCALL(sys_read,compat_sys_s390_read) +SYSCALL(sys_write,compat_sys_s390_write) +SYSCALL(sys_open,compat_sys_open) /* 5 */ +SYSCALL(sys_close,compat_sys_close) +SYSCALL(sys_restart_syscall,sys_restart_syscall) +SYSCALL(sys_creat,compat_sys_creat) +SYSCALL(sys_link,compat_sys_link) +SYSCALL(sys_unlink,compat_sys_unlink) /* 10 */ +SYSCALL(sys_execve,compat_sys_execve) +SYSCALL(sys_chdir,compat_sys_chdir) +SYSCALL(sys_ni_syscall,compat_sys_time) /* old time syscall */ +SYSCALL(sys_mknod,compat_sys_mknod) +SYSCALL(sys_chmod,compat_sys_chmod) /* 15 */ +SYSCALL(sys_ni_syscall,compat_sys_s390_lchown16) /* old lchown16 syscall*/ +NI_SYSCALL /* old break syscall holder */ +NI_SYSCALL /* old stat syscall holder */ +SYSCALL(sys_lseek,compat_sys_lseek) +SYSCALL(sys_getpid,sys_getpid) /* 20 */ +SYSCALL(sys_mount,compat_sys_mount) +SYSCALL(sys_oldumount,compat_sys_oldumount) +SYSCALL(sys_ni_syscall,compat_sys_s390_setuid16) /* old setuid16 syscall*/ +SYSCALL(sys_ni_syscall,compat_sys_s390_getuid16) /* old getuid16 syscall*/ +SYSCALL(sys_ni_syscall,compat_sys_stime) /* 25 old stime syscall */ +SYSCALL(sys_ptrace,compat_sys_ptrace) +SYSCALL(sys_alarm,compat_sys_alarm) +NI_SYSCALL /* old fstat syscall */ +SYSCALL(sys_pause,sys_pause) +SYSCALL(sys_utime,compat_sys_utime) /* 30 */ +NI_SYSCALL /* old stty syscall */ +NI_SYSCALL /* old gtty syscall */ +SYSCALL(sys_access,compat_sys_access) +SYSCALL(sys_nice,compat_sys_nice) +NI_SYSCALL /* 35 old ftime syscall */ +SYSCALL(sys_sync,sys_sync) +SYSCALL(sys_kill,compat_sys_kill) +SYSCALL(sys_rename,compat_sys_rename) +SYSCALL(sys_mkdir,compat_sys_mkdir) +SYSCALL(sys_rmdir,compat_sys_rmdir) /* 40 */ +SYSCALL(sys_dup,compat_sys_dup) +SYSCALL(sys_pipe,compat_sys_pipe) +SYSCALL(sys_times,compat_sys_times) +NI_SYSCALL /* old prof syscall */ +SYSCALL(sys_brk,compat_sys_brk) /* 45 */ +SYSCALL(sys_ni_syscall,compat_sys_s390_setgid16) /* old setgid16 syscall*/ +SYSCALL(sys_ni_syscall,compat_sys_s390_getgid16) /* old getgid16 syscall*/ +SYSCALL(sys_signal,compat_sys_signal) +SYSCALL(sys_ni_syscall,compat_sys_s390_geteuid16) /* old geteuid16 syscall */ +SYSCALL(sys_ni_syscall,compat_sys_s390_getegid16) /* 50 old getegid16 syscall */ +SYSCALL(sys_acct,compat_sys_acct) +SYSCALL(sys_umount,compat_sys_umount) +NI_SYSCALL /* old lock syscall */ +SYSCALL(sys_ioctl,compat_sys_ioctl) +SYSCALL(sys_fcntl,compat_sys_fcntl) /* 55 */ +NI_SYSCALL /* intel mpx syscall */ +SYSCALL(sys_setpgid,compat_sys_setpgid) +NI_SYSCALL /* old ulimit syscall */ +NI_SYSCALL /* old uname syscall */ +SYSCALL(sys_umask,compat_sys_umask) /* 60 */ +SYSCALL(sys_chroot,compat_sys_chroot) +SYSCALL(sys_ustat,compat_sys_ustat) +SYSCALL(sys_dup2,compat_sys_dup2) +SYSCALL(sys_getppid,sys_getppid) +SYSCALL(sys_getpgrp,sys_getpgrp) /* 65 */ +SYSCALL(sys_setsid,sys_setsid) +SYSCALL(sys_sigaction,compat_sys_sigaction) +NI_SYSCALL /* old sgetmask syscall*/ +NI_SYSCALL /* old ssetmask syscall*/ +SYSCALL(sys_ni_syscall,compat_sys_s390_setreuid16) /* old setreuid16 syscall */ +SYSCALL(sys_ni_syscall,compat_sys_s390_setregid16) /* old setregid16 syscall */ +SYSCALL(sys_sigsuspend,compat_sys_sigsuspend) +SYSCALL(sys_sigpending,compat_sys_sigpending) +SYSCALL(sys_sethostname,compat_sys_sethostname) +SYSCALL(sys_setrlimit,compat_sys_setrlimit) /* 75 */ +SYSCALL(sys_getrlimit,compat_sys_old_getrlimit) +SYSCALL(sys_getrusage,compat_sys_getrusage) +SYSCALL(sys_gettimeofday,compat_sys_gettimeofday) +SYSCALL(sys_settimeofday,compat_sys_settimeofday) +SYSCALL(sys_ni_syscall,compat_sys_s390_getgroups16) /* 80 old getgroups16 syscall */ +SYSCALL(sys_ni_syscall,compat_sys_s390_setgroups16) /* old setgroups16 syscall */ +NI_SYSCALL /* old select syscall */ +SYSCALL(sys_symlink,compat_sys_symlink) +NI_SYSCALL /* old lstat syscall */ +SYSCALL(sys_readlink,compat_sys_readlink) /* 85 */ +SYSCALL(sys_uselib,compat_sys_uselib) +SYSCALL(sys_swapon,compat_sys_swapon) +SYSCALL(sys_reboot,compat_sys_reboot) +SYSCALL(sys_ni_syscall,compat_sys_old_readdir) /* old readdir syscall */ +SYSCALL(sys_old_mmap,compat_sys_s390_old_mmap) /* 90 */ +SYSCALL(sys_munmap,compat_sys_munmap) +SYSCALL(sys_truncate,compat_sys_truncate) +SYSCALL(sys_ftruncate,compat_sys_ftruncate) +SYSCALL(sys_fchmod,compat_sys_fchmod) +SYSCALL(sys_ni_syscall,compat_sys_s390_fchown16) /* 95 old fchown16 syscall*/ +SYSCALL(sys_getpriority,compat_sys_getpriority) +SYSCALL(sys_setpriority,compat_sys_setpriority) +NI_SYSCALL /* old profil syscall */ +SYSCALL(sys_statfs,compat_sys_statfs) +SYSCALL(sys_fstatfs,compat_sys_fstatfs) /* 100 */ +NI_SYSCALL /* ioperm for i386 */ +SYSCALL(sys_socketcall,compat_sys_socketcall) +SYSCALL(sys_syslog,compat_sys_syslog) +SYSCALL(sys_setitimer,compat_sys_setitimer) +SYSCALL(sys_getitimer,compat_sys_getitimer) /* 105 */ +SYSCALL(sys_newstat,compat_sys_newstat) +SYSCALL(sys_newlstat,compat_sys_newlstat) +SYSCALL(sys_newfstat,compat_sys_newfstat) +NI_SYSCALL /* old uname syscall */ +SYSCALL(sys_lookup_dcookie,compat_sys_lookup_dcookie) /* 110 */ +SYSCALL(sys_vhangup,sys_vhangup) +NI_SYSCALL /* old "idle" system call */ +NI_SYSCALL /* vm86old for i386 */ +SYSCALL(sys_wait4,compat_sys_wait4) +SYSCALL(sys_swapoff,compat_sys_swapoff) /* 115 */ +SYSCALL(sys_sysinfo,compat_sys_sysinfo) +SYSCALL(sys_s390_ipc,compat_sys_s390_ipc) +SYSCALL(sys_fsync,compat_sys_fsync) +SYSCALL(sys_sigreturn,compat_sys_sigreturn) +SYSCALL(sys_clone,compat_sys_clone) /* 120 */ +SYSCALL(sys_setdomainname,compat_sys_setdomainname) +SYSCALL(sys_newuname,compat_sys_newuname) +NI_SYSCALL /* modify_ldt for i386 */ +SYSCALL(sys_adjtimex,compat_sys_adjtimex) +SYSCALL(sys_mprotect,compat_sys_mprotect) /* 125 */ +SYSCALL(sys_sigprocmask,compat_sys_sigprocmask) +NI_SYSCALL /* old "create module" */ +SYSCALL(sys_init_module,compat_sys_init_module) +SYSCALL(sys_delete_module,compat_sys_delete_module) +NI_SYSCALL /* 130: old get_kernel_syms */ +SYSCALL(sys_quotactl,compat_sys_quotactl) +SYSCALL(sys_getpgid,compat_sys_getpgid) +SYSCALL(sys_fchdir,compat_sys_fchdir) +SYSCALL(sys_bdflush,compat_sys_bdflush) +SYSCALL(sys_sysfs,compat_sys_sysfs) /* 135 */ +SYSCALL(sys_s390_personality,compat_sys_s390_personality) +NI_SYSCALL /* for afs_syscall */ +SYSCALL(sys_ni_syscall,compat_sys_s390_setfsuid16) /* old setfsuid16 syscall */ +SYSCALL(sys_ni_syscall,compat_sys_s390_setfsgid16) /* old setfsgid16 syscall */ +SYSCALL(sys_llseek,compat_sys_llseek) /* 140 */ +SYSCALL(sys_getdents,compat_sys_getdents) +SYSCALL(sys_select,compat_sys_select) +SYSCALL(sys_flock,compat_sys_flock) +SYSCALL(sys_msync,compat_sys_msync) +SYSCALL(sys_readv,compat_sys_readv) /* 145 */ +SYSCALL(sys_writev,compat_sys_writev) +SYSCALL(sys_getsid,compat_sys_getsid) +SYSCALL(sys_fdatasync,compat_sys_fdatasync) +SYSCALL(sys_sysctl,compat_sys_sysctl) +SYSCALL(sys_mlock,compat_sys_mlock) /* 150 */ +SYSCALL(sys_munlock,compat_sys_munlock) +SYSCALL(sys_mlockall,compat_sys_mlockall) +SYSCALL(sys_munlockall,sys_munlockall) +SYSCALL(sys_sched_setparam,compat_sys_sched_setparam) +SYSCALL(sys_sched_getparam,compat_sys_sched_getparam) /* 155 */ +SYSCALL(sys_sched_setscheduler,compat_sys_sched_setscheduler) +SYSCALL(sys_sched_getscheduler,compat_sys_sched_getscheduler) +SYSCALL(sys_sched_yield,sys_sched_yield) +SYSCALL(sys_sched_get_priority_max,compat_sys_sched_get_priority_max) +SYSCALL(sys_sched_get_priority_min,compat_sys_sched_get_priority_min) /* 160 */ +SYSCALL(sys_sched_rr_get_interval,compat_sys_sched_rr_get_interval) +SYSCALL(sys_nanosleep,compat_sys_nanosleep) +SYSCALL(sys_mremap,compat_sys_mremap) +SYSCALL(sys_ni_syscall,compat_sys_s390_setresuid16) /* old setresuid16 syscall */ +SYSCALL(sys_ni_syscall,compat_sys_s390_getresuid16) /* 165 old getresuid16 syscall */ +NI_SYSCALL /* for vm86 */ +NI_SYSCALL /* old sys_query_module */ +SYSCALL(sys_poll,compat_sys_poll) +NI_SYSCALL /* old nfsservctl */ +SYSCALL(sys_ni_syscall,compat_sys_s390_setresgid16) /* 170 old setresgid16 syscall */ +SYSCALL(sys_ni_syscall,compat_sys_s390_getresgid16) /* old getresgid16 syscall */ +SYSCALL(sys_prctl,compat_sys_prctl) +SYSCALL(sys_rt_sigreturn,compat_sys_rt_sigreturn) +SYSCALL(sys_rt_sigaction,compat_sys_rt_sigaction) +SYSCALL(sys_rt_sigprocmask,compat_sys_rt_sigprocmask) /* 175 */ +SYSCALL(sys_rt_sigpending,compat_sys_rt_sigpending) +SYSCALL(sys_rt_sigtimedwait,compat_sys_rt_sigtimedwait) +SYSCALL(sys_rt_sigqueueinfo,compat_sys_rt_sigqueueinfo) +SYSCALL(sys_rt_sigsuspend,compat_sys_rt_sigsuspend) +SYSCALL(sys_pread64,compat_sys_s390_pread64) /* 180 */ +SYSCALL(sys_pwrite64,compat_sys_s390_pwrite64) +SYSCALL(sys_ni_syscall,compat_sys_s390_chown16) /* old chown16 syscall */ +SYSCALL(sys_getcwd,compat_sys_getcwd) +SYSCALL(sys_capget,compat_sys_capget) +SYSCALL(sys_capset,compat_sys_capset) /* 185 */ +SYSCALL(sys_sigaltstack,compat_sys_sigaltstack) +SYSCALL(sys_sendfile64,compat_sys_sendfile) +NI_SYSCALL /* streams1 */ +NI_SYSCALL /* streams2 */ +SYSCALL(sys_vfork,sys_vfork) /* 190 */ +SYSCALL(sys_getrlimit,compat_sys_getrlimit) +SYSCALL(sys_mmap2,compat_sys_s390_mmap2) +SYSCALL(sys_ni_syscall,compat_sys_s390_truncate64) +SYSCALL(sys_ni_syscall,compat_sys_s390_ftruncate64) +SYSCALL(sys_ni_syscall,compat_sys_s390_stat64) /* 195 */ +SYSCALL(sys_ni_syscall,compat_sys_s390_lstat64) +SYSCALL(sys_ni_syscall,compat_sys_s390_fstat64) +SYSCALL(sys_lchown,compat_sys_lchown) +SYSCALL(sys_getuid,sys_getuid) +SYSCALL(sys_getgid,sys_getgid) /* 200 */ +SYSCALL(sys_geteuid,sys_geteuid) +SYSCALL(sys_getegid,sys_getegid) +SYSCALL(sys_setreuid,compat_sys_setreuid) +SYSCALL(sys_setregid,compat_sys_setregid) +SYSCALL(sys_getgroups,compat_sys_getgroups) /* 205 */ +SYSCALL(sys_setgroups,compat_sys_setgroups) +SYSCALL(sys_fchown,compat_sys_fchown) +SYSCALL(sys_setresuid,compat_sys_setresuid) +SYSCALL(sys_getresuid,compat_sys_getresuid) +SYSCALL(sys_setresgid,compat_sys_setresgid) /* 210 */ +SYSCALL(sys_getresgid,compat_sys_getresgid) +SYSCALL(sys_chown,compat_sys_chown) +SYSCALL(sys_setuid,compat_sys_setuid) +SYSCALL(sys_setgid,compat_sys_setgid) +SYSCALL(sys_setfsuid,compat_sys_setfsuid) /* 215 */ +SYSCALL(sys_setfsgid,compat_sys_setfsgid) +SYSCALL(sys_pivot_root,compat_sys_pivot_root) +SYSCALL(sys_mincore,compat_sys_mincore) +SYSCALL(sys_madvise,compat_sys_madvise) +SYSCALL(sys_getdents64,compat_sys_getdents64) /* 220 */ +SYSCALL(sys_ni_syscall,compat_sys_fcntl64) +SYSCALL(sys_readahead,compat_sys_s390_readahead) +SYSCALL(sys_ni_syscall,compat_sys_sendfile64) +SYSCALL(sys_setxattr,compat_sys_setxattr) +SYSCALL(sys_lsetxattr,compat_sys_lsetxattr) /* 225 */ +SYSCALL(sys_fsetxattr,compat_sys_fsetxattr) +SYSCALL(sys_getxattr,compat_sys_getxattr) +SYSCALL(sys_lgetxattr,compat_sys_lgetxattr) +SYSCALL(sys_fgetxattr,compat_sys_fgetxattr) +SYSCALL(sys_listxattr,compat_sys_listxattr) /* 230 */ +SYSCALL(sys_llistxattr,compat_sys_llistxattr) +SYSCALL(sys_flistxattr,compat_sys_flistxattr) +SYSCALL(sys_removexattr,compat_sys_removexattr) +SYSCALL(sys_lremovexattr,compat_sys_lremovexattr) +SYSCALL(sys_fremovexattr,compat_sys_fremovexattr) /* 235 */ +SYSCALL(sys_gettid,sys_gettid) +SYSCALL(sys_tkill,compat_sys_tkill) +SYSCALL(sys_futex,compat_sys_futex) +SYSCALL(sys_sched_setaffinity,compat_sys_sched_setaffinity) +SYSCALL(sys_sched_getaffinity,compat_sys_sched_getaffinity) /* 240 */ +SYSCALL(sys_tgkill,compat_sys_tgkill) +NI_SYSCALL /* reserved for TUX */ +SYSCALL(sys_io_setup,compat_sys_io_setup) +SYSCALL(sys_io_destroy,compat_sys_io_destroy) +SYSCALL(sys_io_getevents,compat_sys_io_getevents) /* 245 */ +SYSCALL(sys_io_submit,compat_sys_io_submit) +SYSCALL(sys_io_cancel,compat_sys_io_cancel) +SYSCALL(sys_exit_group,compat_sys_exit_group) +SYSCALL(sys_epoll_create,compat_sys_epoll_create) +SYSCALL(sys_epoll_ctl,compat_sys_epoll_ctl) /* 250 */ +SYSCALL(sys_epoll_wait,compat_sys_epoll_wait) +SYSCALL(sys_set_tid_address,compat_sys_set_tid_address) +SYSCALL(sys_fadvise64_64,compat_sys_s390_fadvise64) +SYSCALL(sys_timer_create,compat_sys_timer_create) +SYSCALL(sys_timer_settime,compat_sys_timer_settime) /* 255 */ +SYSCALL(sys_timer_gettime,compat_sys_timer_gettime) +SYSCALL(sys_timer_getoverrun,compat_sys_timer_getoverrun) +SYSCALL(sys_timer_delete,compat_sys_timer_delete) +SYSCALL(sys_clock_settime,compat_sys_clock_settime) +SYSCALL(sys_clock_gettime,compat_sys_clock_gettime) /* 260 */ +SYSCALL(sys_clock_getres,compat_sys_clock_getres) +SYSCALL(sys_clock_nanosleep,compat_sys_clock_nanosleep) +NI_SYSCALL /* reserved for vserver */ +SYSCALL(sys_ni_syscall,compat_sys_s390_fadvise64_64) +SYSCALL(sys_statfs64,compat_sys_statfs64) +SYSCALL(sys_fstatfs64,compat_sys_fstatfs64) +SYSCALL(sys_remap_file_pages,compat_sys_remap_file_pages) +NI_SYSCALL /* 268 sys_mbind */ +NI_SYSCALL /* 269 sys_get_mempolicy */ +NI_SYSCALL /* 270 sys_set_mempolicy */ +SYSCALL(sys_mq_open,compat_sys_mq_open) +SYSCALL(sys_mq_unlink,compat_sys_mq_unlink) +SYSCALL(sys_mq_timedsend,compat_sys_mq_timedsend) +SYSCALL(sys_mq_timedreceive,compat_sys_mq_timedreceive) +SYSCALL(sys_mq_notify,compat_sys_mq_notify) /* 275 */ +SYSCALL(sys_mq_getsetattr,compat_sys_mq_getsetattr) +SYSCALL(sys_kexec_load,compat_sys_kexec_load) +SYSCALL(sys_add_key,compat_sys_add_key) +SYSCALL(sys_request_key,compat_sys_request_key) +SYSCALL(sys_keyctl,compat_sys_keyctl) /* 280 */ +SYSCALL(sys_waitid,compat_sys_waitid) +SYSCALL(sys_ioprio_set,compat_sys_ioprio_set) +SYSCALL(sys_ioprio_get,compat_sys_ioprio_get) +SYSCALL(sys_inotify_init,sys_inotify_init) +SYSCALL(sys_inotify_add_watch,compat_sys_inotify_add_watch) /* 285 */ +SYSCALL(sys_inotify_rm_watch,compat_sys_inotify_rm_watch) +NI_SYSCALL /* 287 sys_migrate_pages */ +SYSCALL(sys_openat,compat_sys_openat) +SYSCALL(sys_mkdirat,compat_sys_mkdirat) +SYSCALL(sys_mknodat,compat_sys_mknodat) /* 290 */ +SYSCALL(sys_fchownat,compat_sys_fchownat) +SYSCALL(sys_futimesat,compat_sys_futimesat) +SYSCALL(sys_newfstatat,compat_sys_s390_fstatat64) +SYSCALL(sys_unlinkat,compat_sys_unlinkat) +SYSCALL(sys_renameat,compat_sys_renameat) /* 295 */ +SYSCALL(sys_linkat,compat_sys_linkat) +SYSCALL(sys_symlinkat,compat_sys_symlinkat) +SYSCALL(sys_readlinkat,compat_sys_readlinkat) +SYSCALL(sys_fchmodat,compat_sys_fchmodat) +SYSCALL(sys_faccessat,compat_sys_faccessat) /* 300 */ +SYSCALL(sys_pselect6,compat_sys_pselect6) +SYSCALL(sys_ppoll,compat_sys_ppoll) +SYSCALL(sys_unshare,compat_sys_unshare) +SYSCALL(sys_set_robust_list,compat_sys_set_robust_list) +SYSCALL(sys_get_robust_list,compat_sys_get_robust_list) +SYSCALL(sys_splice,compat_sys_splice) +SYSCALL(sys_sync_file_range,compat_sys_s390_sync_file_range) +SYSCALL(sys_tee,compat_sys_tee) +SYSCALL(sys_vmsplice,compat_sys_vmsplice) +NI_SYSCALL /* 310 sys_move_pages */ +SYSCALL(sys_getcpu,compat_sys_getcpu) +SYSCALL(sys_epoll_pwait,compat_sys_epoll_pwait) +SYSCALL(sys_utimes,compat_sys_utimes) +SYSCALL(sys_fallocate,compat_sys_s390_fallocate) +SYSCALL(sys_utimensat,compat_sys_utimensat) /* 315 */ +SYSCALL(sys_signalfd,compat_sys_signalfd) +NI_SYSCALL /* 317 old sys_timer_fd */ +SYSCALL(sys_eventfd,compat_sys_eventfd) +SYSCALL(sys_timerfd_create,compat_sys_timerfd_create) +SYSCALL(sys_timerfd_settime,compat_sys_timerfd_settime) /* 320 */ +SYSCALL(sys_timerfd_gettime,compat_sys_timerfd_gettime) +SYSCALL(sys_signalfd4,compat_sys_signalfd4) +SYSCALL(sys_eventfd2,compat_sys_eventfd2) +SYSCALL(sys_inotify_init1,compat_sys_inotify_init1) +SYSCALL(sys_pipe2,compat_sys_pipe2) /* 325 */ +SYSCALL(sys_dup3,compat_sys_dup3) +SYSCALL(sys_epoll_create1,compat_sys_epoll_create1) +SYSCALL(sys_preadv,compat_sys_preadv) +SYSCALL(sys_pwritev,compat_sys_pwritev) +SYSCALL(sys_rt_tgsigqueueinfo,compat_sys_rt_tgsigqueueinfo) /* 330 */ +SYSCALL(sys_perf_event_open,compat_sys_perf_event_open) +SYSCALL(sys_fanotify_init,compat_sys_fanotify_init) +SYSCALL(sys_fanotify_mark,compat_sys_fanotify_mark) +SYSCALL(sys_prlimit64,compat_sys_prlimit64) +SYSCALL(sys_name_to_handle_at,compat_sys_name_to_handle_at) /* 335 */ +SYSCALL(sys_open_by_handle_at,compat_sys_open_by_handle_at) +SYSCALL(sys_clock_adjtime,compat_sys_clock_adjtime) +SYSCALL(sys_syncfs,compat_sys_syncfs) +SYSCALL(sys_setns,compat_sys_setns) +SYSCALL(sys_process_vm_readv,compat_sys_process_vm_readv) /* 340 */ +SYSCALL(sys_process_vm_writev,compat_sys_process_vm_writev) +SYSCALL(sys_s390_runtime_instr,compat_sys_s390_runtime_instr) +SYSCALL(sys_kcmp,compat_sys_kcmp) +SYSCALL(sys_finit_module,compat_sys_finit_module) +SYSCALL(sys_sched_setattr,compat_sys_sched_setattr) /* 345 */ +SYSCALL(sys_sched_getattr,compat_sys_sched_getattr) +SYSCALL(sys_renameat2,compat_sys_renameat2) +SYSCALL(sys_seccomp,compat_sys_seccomp) +SYSCALL(sys_getrandom,compat_sys_getrandom) +SYSCALL(sys_memfd_create,compat_sys_memfd_create) /* 350 */ +SYSCALL(sys_bpf,compat_sys_bpf) +SYSCALL(sys_s390_pci_mmio_write,compat_sys_s390_pci_mmio_write) +SYSCALL(sys_s390_pci_mmio_read,compat_sys_s390_pci_mmio_read) +SYSCALL(sys_execveat,compat_sys_execveat) diff --git a/arch/s390/kernel/sysinfo.c b/arch/s390/kernel/sysinfo.c new file mode 100644 index 000000000..99babea02 --- /dev/null +++ b/arch/s390/kernel/sysinfo.c @@ -0,0 +1,465 @@ +/* + * Copyright IBM Corp. 2001, 2009 + * Author(s): Ulrich Weigand <Ulrich.Weigand@de.ibm.com>, + * Martin Schwidefsky <schwidefsky@de.ibm.com>, + */ + +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/init.h> +#include <linux/delay.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <asm/ebcdic.h> +#include <asm/sysinfo.h> +#include <asm/cpcmd.h> +#include <asm/topology.h> + +/* Sigh, math-emu. Don't ask. */ +#include <asm/sfp-util.h> +#include <math-emu/soft-fp.h> +#include <math-emu/single.h> + +int topology_max_mnest; + +/* + * stsi - store system information + * + * Returns the current configuration level if function code 0 was specified. + * Otherwise returns 0 on success or a negative value on error. + */ +int stsi(void *sysinfo, int fc, int sel1, int sel2) +{ + register int r0 asm("0") = (fc << 28) | sel1; + register int r1 asm("1") = sel2; + int rc = 0; + + asm volatile( + " stsi 0(%3)\n" + "0: jz 2f\n" + "1: lhi %1,%4\n" + "2:\n" + EX_TABLE(0b, 1b) + : "+d" (r0), "+d" (rc) + : "d" (r1), "a" (sysinfo), "K" (-EOPNOTSUPP) + : "cc", "memory"); + if (rc) + return rc; + return fc ? 0 : ((unsigned int) r0) >> 28; +} +EXPORT_SYMBOL(stsi); + +static void stsi_1_1_1(struct seq_file *m, struct sysinfo_1_1_1 *info) +{ + int i; + + if (stsi(info, 1, 1, 1)) + return; + EBCASC(info->manufacturer, sizeof(info->manufacturer)); + EBCASC(info->type, sizeof(info->type)); + EBCASC(info->model, sizeof(info->model)); + EBCASC(info->sequence, sizeof(info->sequence)); + EBCASC(info->plant, sizeof(info->plant)); + EBCASC(info->model_capacity, sizeof(info->model_capacity)); + EBCASC(info->model_perm_cap, sizeof(info->model_perm_cap)); + EBCASC(info->model_temp_cap, sizeof(info->model_temp_cap)); + seq_printf(m, "Manufacturer: %-16.16s\n", info->manufacturer); + seq_printf(m, "Type: %-4.4s\n", info->type); + /* + * Sigh: the model field has been renamed with System z9 + * to model_capacity and a new model field has been added + * after the plant field. To avoid confusing older programs + * the "Model:" prints "model_capacity model" or just + * "model_capacity" if the model string is empty . + */ + seq_printf(m, "Model: %-16.16s", info->model_capacity); + if (info->model[0] != '\0') + seq_printf(m, " %-16.16s", info->model); + seq_putc(m, '\n'); + seq_printf(m, "Sequence Code: %-16.16s\n", info->sequence); + seq_printf(m, "Plant: %-4.4s\n", info->plant); + seq_printf(m, "Model Capacity: %-16.16s %08u\n", + info->model_capacity, info->model_cap_rating); + if (info->model_perm_cap_rating) + seq_printf(m, "Model Perm. Capacity: %-16.16s %08u\n", + info->model_perm_cap, + info->model_perm_cap_rating); + if (info->model_temp_cap_rating) + seq_printf(m, "Model Temp. Capacity: %-16.16s %08u\n", + info->model_temp_cap, + info->model_temp_cap_rating); + if (info->ncr) + seq_printf(m, "Nominal Cap. Rating: %08u\n", info->ncr); + if (info->npr) + seq_printf(m, "Nominal Perm. Rating: %08u\n", info->npr); + if (info->ntr) + seq_printf(m, "Nominal Temp. Rating: %08u\n", info->ntr); + if (info->cai) { + seq_printf(m, "Capacity Adj. Ind.: %d\n", info->cai); + seq_printf(m, "Capacity Ch. Reason: %d\n", info->ccr); + seq_printf(m, "Capacity Transient: %d\n", info->t); + } + if (info->p) { + for (i = 1; i <= ARRAY_SIZE(info->typepct); i++) { + seq_printf(m, "Type %d Percentage: %d\n", + i, info->typepct[i - 1]); + } + } +} + +static void stsi_15_1_x(struct seq_file *m, struct sysinfo_15_1_x *info) +{ + static int max_mnest; + int i, rc; + + seq_putc(m, '\n'); + if (!MACHINE_HAS_TOPOLOGY) + return; + if (stsi(info, 15, 1, topology_max_mnest)) + return; + seq_printf(m, "CPU Topology HW: "); + for (i = 0; i < TOPOLOGY_NR_MAG; i++) + seq_printf(m, " %d", info->mag[i]); + seq_putc(m, '\n'); +#ifdef CONFIG_SCHED_MC + store_topology(info); + seq_printf(m, "CPU Topology SW: "); + for (i = 0; i < TOPOLOGY_NR_MAG; i++) + seq_printf(m, " %d", info->mag[i]); + seq_putc(m, '\n'); +#endif +} + +static void stsi_1_2_2(struct seq_file *m, struct sysinfo_1_2_2 *info) +{ + struct sysinfo_1_2_2_extension *ext; + int i; + + if (stsi(info, 1, 2, 2)) + return; + ext = (struct sysinfo_1_2_2_extension *) + ((unsigned long) info + info->acc_offset); + seq_printf(m, "CPUs Total: %d\n", info->cpus_total); + seq_printf(m, "CPUs Configured: %d\n", info->cpus_configured); + seq_printf(m, "CPUs Standby: %d\n", info->cpus_standby); + seq_printf(m, "CPUs Reserved: %d\n", info->cpus_reserved); + /* + * Sigh 2. According to the specification the alternate + * capability field is a 32 bit floating point number + * if the higher order 8 bits are not zero. Printing + * a floating point number in the kernel is a no-no, + * always print the number as 32 bit unsigned integer. + * The user-space needs to know about the strange + * encoding of the alternate cpu capability. + */ + seq_printf(m, "Capability: %u", info->capability); + if (info->format == 1) + seq_printf(m, " %u", ext->alt_capability); + seq_putc(m, '\n'); + if (info->nominal_cap) + seq_printf(m, "Nominal Capability: %d\n", info->nominal_cap); + if (info->secondary_cap) + seq_printf(m, "Secondary Capability: %d\n", info->secondary_cap); + for (i = 2; i <= info->cpus_total; i++) { + seq_printf(m, "Adjustment %02d-way: %u", + i, info->adjustment[i-2]); + if (info->format == 1) + seq_printf(m, " %u", ext->alt_adjustment[i-2]); + seq_putc(m, '\n'); + } +} + +static void stsi_2_2_2(struct seq_file *m, struct sysinfo_2_2_2 *info) +{ + if (stsi(info, 2, 2, 2)) + return; + EBCASC(info->name, sizeof(info->name)); + seq_putc(m, '\n'); + seq_printf(m, "LPAR Number: %d\n", info->lpar_number); + seq_printf(m, "LPAR Characteristics: "); + if (info->characteristics & LPAR_CHAR_DEDICATED) + seq_printf(m, "Dedicated "); + if (info->characteristics & LPAR_CHAR_SHARED) + seq_printf(m, "Shared "); + if (info->characteristics & LPAR_CHAR_LIMITED) + seq_printf(m, "Limited "); + seq_putc(m, '\n'); + seq_printf(m, "LPAR Name: %-8.8s\n", info->name); + seq_printf(m, "LPAR Adjustment: %d\n", info->caf); + seq_printf(m, "LPAR CPUs Total: %d\n", info->cpus_total); + seq_printf(m, "LPAR CPUs Configured: %d\n", info->cpus_configured); + seq_printf(m, "LPAR CPUs Standby: %d\n", info->cpus_standby); + seq_printf(m, "LPAR CPUs Reserved: %d\n", info->cpus_reserved); + seq_printf(m, "LPAR CPUs Dedicated: %d\n", info->cpus_dedicated); + seq_printf(m, "LPAR CPUs Shared: %d\n", info->cpus_shared); + if (info->mt_installed & 0x80) { + seq_printf(m, "LPAR CPUs G-MTID: %d\n", + info->mt_general & 0x1f); + seq_printf(m, "LPAR CPUs S-MTID: %d\n", + info->mt_installed & 0x1f); + seq_printf(m, "LPAR CPUs PS-MTID: %d\n", + info->mt_psmtid & 0x1f); + } +} + +static void print_ext_name(struct seq_file *m, int lvl, + struct sysinfo_3_2_2 *info) +{ + if (info->vm[lvl].ext_name_encoding == 0) + return; + if (info->ext_names[lvl][0] == 0) + return; + switch (info->vm[lvl].ext_name_encoding) { + case 1: /* EBCDIC */ + EBCASC(info->ext_names[lvl], sizeof(info->ext_names[lvl])); + break; + case 2: /* UTF-8 */ + break; + default: + return; + } + seq_printf(m, "VM%02d Extended Name: %-.256s\n", lvl, + info->ext_names[lvl]); +} + +static void print_uuid(struct seq_file *m, int i, struct sysinfo_3_2_2 *info) +{ + if (!memcmp(&info->vm[i].uuid, &NULL_UUID_BE, sizeof(uuid_be))) + return; + seq_printf(m, "VM%02d UUID: %pUb\n", i, &info->vm[i].uuid); +} + +static void stsi_3_2_2(struct seq_file *m, struct sysinfo_3_2_2 *info) +{ + int i; + + if (stsi(info, 3, 2, 2)) + return; + for (i = 0; i < info->count; i++) { + EBCASC(info->vm[i].name, sizeof(info->vm[i].name)); + EBCASC(info->vm[i].cpi, sizeof(info->vm[i].cpi)); + seq_putc(m, '\n'); + seq_printf(m, "VM%02d Name: %-8.8s\n", i, info->vm[i].name); + seq_printf(m, "VM%02d Control Program: %-16.16s\n", i, info->vm[i].cpi); + seq_printf(m, "VM%02d Adjustment: %d\n", i, info->vm[i].caf); + seq_printf(m, "VM%02d CPUs Total: %d\n", i, info->vm[i].cpus_total); + seq_printf(m, "VM%02d CPUs Configured: %d\n", i, info->vm[i].cpus_configured); + seq_printf(m, "VM%02d CPUs Standby: %d\n", i, info->vm[i].cpus_standby); + seq_printf(m, "VM%02d CPUs Reserved: %d\n", i, info->vm[i].cpus_reserved); + print_ext_name(m, i, info); + print_uuid(m, i, info); + } +} + +static int sysinfo_show(struct seq_file *m, void *v) +{ + void *info = (void *)get_zeroed_page(GFP_KERNEL); + int level; + + if (!info) + return 0; + level = stsi(NULL, 0, 0, 0); + if (level >= 1) + stsi_1_1_1(m, info); + if (level >= 1) + stsi_15_1_x(m, info); + if (level >= 1) + stsi_1_2_2(m, info); + if (level >= 2) + stsi_2_2_2(m, info); + if (level >= 3) + stsi_3_2_2(m, info); + free_page((unsigned long)info); + return 0; +} + +static int sysinfo_open(struct inode *inode, struct file *file) +{ + return single_open(file, sysinfo_show, NULL); +} + +static const struct file_operations sysinfo_fops = { + .open = sysinfo_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int __init sysinfo_create_proc(void) +{ + proc_create("sysinfo", 0444, NULL, &sysinfo_fops); + return 0; +} +device_initcall(sysinfo_create_proc); + +/* + * Service levels interface. + */ + +static DECLARE_RWSEM(service_level_sem); +static LIST_HEAD(service_level_list); + +int register_service_level(struct service_level *slr) +{ + struct service_level *ptr; + + down_write(&service_level_sem); + list_for_each_entry(ptr, &service_level_list, list) + if (ptr == slr) { + up_write(&service_level_sem); + return -EEXIST; + } + list_add_tail(&slr->list, &service_level_list); + up_write(&service_level_sem); + return 0; +} +EXPORT_SYMBOL(register_service_level); + +int unregister_service_level(struct service_level *slr) +{ + struct service_level *ptr, *next; + int rc = -ENOENT; + + down_write(&service_level_sem); + list_for_each_entry_safe(ptr, next, &service_level_list, list) { + if (ptr != slr) + continue; + list_del(&ptr->list); + rc = 0; + break; + } + up_write(&service_level_sem); + return rc; +} +EXPORT_SYMBOL(unregister_service_level); + +static void *service_level_start(struct seq_file *m, loff_t *pos) +{ + down_read(&service_level_sem); + return seq_list_start(&service_level_list, *pos); +} + +static void *service_level_next(struct seq_file *m, void *p, loff_t *pos) +{ + return seq_list_next(p, &service_level_list, pos); +} + +static void service_level_stop(struct seq_file *m, void *p) +{ + up_read(&service_level_sem); +} + +static int service_level_show(struct seq_file *m, void *p) +{ + struct service_level *slr; + + slr = list_entry(p, struct service_level, list); + slr->seq_print(m, slr); + return 0; +} + +static const struct seq_operations service_level_seq_ops = { + .start = service_level_start, + .next = service_level_next, + .stop = service_level_stop, + .show = service_level_show +}; + +static int service_level_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &service_level_seq_ops); +} + +static const struct file_operations service_level_ops = { + .open = service_level_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +static void service_level_vm_print(struct seq_file *m, + struct service_level *slr) +{ + char *query_buffer, *str; + + query_buffer = kmalloc(1024, GFP_KERNEL | GFP_DMA); + if (!query_buffer) + return; + cpcmd("QUERY CPLEVEL", query_buffer, 1024, NULL); + str = strchr(query_buffer, '\n'); + if (str) + *str = 0; + seq_printf(m, "VM: %s\n", query_buffer); + kfree(query_buffer); +} + +static struct service_level service_level_vm = { + .seq_print = service_level_vm_print +}; + +static __init int create_proc_service_level(void) +{ + proc_create("service_levels", 0, NULL, &service_level_ops); + if (MACHINE_IS_VM) + register_service_level(&service_level_vm); + return 0; +} +subsys_initcall(create_proc_service_level); + +/* + * CPU capability might have changed. Therefore recalculate loops_per_jiffy. + */ +void s390_adjust_jiffies(void) +{ + struct sysinfo_1_2_2 *info; + const unsigned int fmil = 0x4b189680; /* 1e7 as 32-bit float. */ + FP_DECL_S(SA); FP_DECL_S(SB); FP_DECL_S(SR); + FP_DECL_EX; + unsigned int capability; + + info = (void *) get_zeroed_page(GFP_KERNEL); + if (!info) + return; + + if (stsi(info, 1, 2, 2) == 0) { + /* + * Major sigh. The cpu capability encoding is "special". + * If the first 9 bits of info->capability are 0 then it + * is a 32 bit unsigned integer in the range 0 .. 2^23. + * If the first 9 bits are != 0 then it is a 32 bit float. + * In addition a lower value indicates a proportionally + * higher cpu capacity. Bogomips are the other way round. + * To get to a halfway suitable number we divide 1e7 + * by the cpu capability number. Yes, that means a floating + * point division .. math-emu here we come :-) + */ + FP_UNPACK_SP(SA, &fmil); + if ((info->capability >> 23) == 0) + FP_FROM_INT_S(SB, (long) info->capability, 64, long); + else + FP_UNPACK_SP(SB, &info->capability); + FP_DIV_S(SR, SA, SB); + FP_TO_INT_S(capability, SR, 32, 0); + } else + /* + * Really old machine without stsi block for basic + * cpu information. Report 42.0 bogomips. + */ + capability = 42; + loops_per_jiffy = capability * (500000/HZ); + free_page((unsigned long) info); +} + +/* + * calibrate the delay loop + */ +void calibrate_delay(void) +{ + s390_adjust_jiffies(); + /* Print the good old Bogomips line .. */ + printk(KERN_DEBUG "Calibrating delay loop (skipped)... " + "%lu.%02lu BogoMIPS preset\n", loops_per_jiffy/(500000/HZ), + (loops_per_jiffy/(5000/HZ)) % 100); +} diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c new file mode 100644 index 000000000..170ddd201 --- /dev/null +++ b/arch/s390/kernel/time.c @@ -0,0 +1,1802 @@ +/* + * Time of day based timer functions. + * + * S390 version + * Copyright IBM Corp. 1999, 2008 + * Author(s): Hartmut Penner (hp@de.ibm.com), + * Martin Schwidefsky (schwidefsky@de.ibm.com), + * Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com) + * + * Derived from "arch/i386/kernel/time.c" + * Copyright (C) 1991, 1992, 1995 Linus Torvalds + */ + +#define KMSG_COMPONENT "time" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel_stat.h> +#include <linux/errno.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/kernel.h> +#include <linux/param.h> +#include <linux/string.h> +#include <linux/mm.h> +#include <linux/interrupt.h> +#include <linux/cpu.h> +#include <linux/stop_machine.h> +#include <linux/time.h> +#include <linux/device.h> +#include <linux/delay.h> +#include <linux/init.h> +#include <linux/smp.h> +#include <linux/types.h> +#include <linux/profile.h> +#include <linux/timex.h> +#include <linux/notifier.h> +#include <linux/timekeeper_internal.h> +#include <linux/clockchips.h> +#include <linux/gfp.h> +#include <linux/kprobes.h> +#include <asm/uaccess.h> +#include <asm/delay.h> +#include <asm/div64.h> +#include <asm/vdso.h> +#include <asm/irq.h> +#include <asm/irq_regs.h> +#include <asm/vtimer.h> +#include <asm/etr.h> +#include <asm/cio.h> +#include "entry.h" + +/* change this if you have some constant time drift */ +#define USECS_PER_JIFFY ((unsigned long) 1000000/HZ) +#define CLK_TICKS_PER_JIFFY ((unsigned long) USECS_PER_JIFFY << 12) + +u64 sched_clock_base_cc = -1; /* Force to data section. */ +EXPORT_SYMBOL_GPL(sched_clock_base_cc); + +static DEFINE_PER_CPU(struct clock_event_device, comparators); + +/* + * Scheduler clock - returns current time in nanosec units. + */ +unsigned long long notrace sched_clock(void) +{ + return tod_to_ns(get_tod_clock_monotonic()); +} +NOKPROBE_SYMBOL(sched_clock); + +/* + * Monotonic_clock - returns # of nanoseconds passed since time_init() + */ +unsigned long long monotonic_clock(void) +{ + return sched_clock(); +} +EXPORT_SYMBOL(monotonic_clock); + +void tod_to_timeval(__u64 todval, struct timespec *xt) +{ + unsigned long long sec; + + sec = todval >> 12; + do_div(sec, 1000000); + xt->tv_sec = sec; + todval -= (sec * 1000000) << 12; + xt->tv_nsec = ((todval * 1000) >> 12); +} +EXPORT_SYMBOL(tod_to_timeval); + +void clock_comparator_work(void) +{ + struct clock_event_device *cd; + + S390_lowcore.clock_comparator = -1ULL; + cd = this_cpu_ptr(&comparators); + cd->event_handler(cd); +} + +/* + * Fixup the clock comparator. + */ +static void fixup_clock_comparator(unsigned long long delta) +{ + /* If nobody is waiting there's nothing to fix. */ + if (S390_lowcore.clock_comparator == -1ULL) + return; + S390_lowcore.clock_comparator += delta; + set_clock_comparator(S390_lowcore.clock_comparator); +} + +static int s390_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + S390_lowcore.clock_comparator = get_tod_clock() + delta; + set_clock_comparator(S390_lowcore.clock_comparator); + return 0; +} + +static void s390_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ +} + +/* + * Set up lowcore and control register of the current cpu to + * enable TOD clock and clock comparator interrupts. + */ +void init_cpu_timer(void) +{ + struct clock_event_device *cd; + int cpu; + + S390_lowcore.clock_comparator = -1ULL; + set_clock_comparator(S390_lowcore.clock_comparator); + + cpu = smp_processor_id(); + cd = &per_cpu(comparators, cpu); + cd->name = "comparator"; + cd->features = CLOCK_EVT_FEAT_ONESHOT; + cd->mult = 16777; + cd->shift = 12; + cd->min_delta_ns = 1; + cd->max_delta_ns = LONG_MAX; + cd->rating = 400; + cd->cpumask = cpumask_of(cpu); + cd->set_next_event = s390_next_event; + cd->set_mode = s390_set_mode; + + clockevents_register_device(cd); + + /* Enable clock comparator timer interrupt. */ + __ctl_set_bit(0,11); + + /* Always allow the timing alert external interrupt. */ + __ctl_set_bit(0, 4); +} + +static void clock_comparator_interrupt(struct ext_code ext_code, + unsigned int param32, + unsigned long param64) +{ + inc_irq_stat(IRQEXT_CLK); + if (S390_lowcore.clock_comparator == -1ULL) + set_clock_comparator(S390_lowcore.clock_comparator); +} + +static void etr_timing_alert(struct etr_irq_parm *); +static void stp_timing_alert(struct stp_irq_parm *); + +static void timing_alert_interrupt(struct ext_code ext_code, + unsigned int param32, unsigned long param64) +{ + inc_irq_stat(IRQEXT_TLA); + if (param32 & 0x00c40000) + etr_timing_alert((struct etr_irq_parm *) ¶m32); + if (param32 & 0x00038000) + stp_timing_alert((struct stp_irq_parm *) ¶m32); +} + +static void etr_reset(void); +static void stp_reset(void); + +void read_persistent_clock(struct timespec *ts) +{ + tod_to_timeval(get_tod_clock() - TOD_UNIX_EPOCH, ts); +} + +void read_boot_clock(struct timespec *ts) +{ + tod_to_timeval(sched_clock_base_cc - TOD_UNIX_EPOCH, ts); +} + +static cycle_t read_tod_clock(struct clocksource *cs) +{ + return get_tod_clock(); +} + +static struct clocksource clocksource_tod = { + .name = "tod", + .rating = 400, + .read = read_tod_clock, + .mask = -1ULL, + .mult = 1000, + .shift = 12, + .flags = CLOCK_SOURCE_IS_CONTINUOUS, +}; + +struct clocksource * __init clocksource_default_clock(void) +{ + return &clocksource_tod; +} + +void update_vsyscall(struct timekeeper *tk) +{ + u64 nsecps; + + if (tk->tkr_mono.clock != &clocksource_tod) + return; + + /* Make userspace gettimeofday spin until we're done. */ + ++vdso_data->tb_update_count; + smp_wmb(); + vdso_data->xtime_tod_stamp = tk->tkr_mono.cycle_last; + vdso_data->xtime_clock_sec = tk->xtime_sec; + vdso_data->xtime_clock_nsec = tk->tkr_mono.xtime_nsec; + vdso_data->wtom_clock_sec = + tk->xtime_sec + tk->wall_to_monotonic.tv_sec; + vdso_data->wtom_clock_nsec = tk->tkr_mono.xtime_nsec + + + ((u64) tk->wall_to_monotonic.tv_nsec << tk->tkr_mono.shift); + nsecps = (u64) NSEC_PER_SEC << tk->tkr_mono.shift; + while (vdso_data->wtom_clock_nsec >= nsecps) { + vdso_data->wtom_clock_nsec -= nsecps; + vdso_data->wtom_clock_sec++; + } + + vdso_data->xtime_coarse_sec = tk->xtime_sec; + vdso_data->xtime_coarse_nsec = + (long)(tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift); + vdso_data->wtom_coarse_sec = + vdso_data->xtime_coarse_sec + tk->wall_to_monotonic.tv_sec; + vdso_data->wtom_coarse_nsec = + vdso_data->xtime_coarse_nsec + tk->wall_to_monotonic.tv_nsec; + while (vdso_data->wtom_coarse_nsec >= NSEC_PER_SEC) { + vdso_data->wtom_coarse_nsec -= NSEC_PER_SEC; + vdso_data->wtom_coarse_sec++; + } + + vdso_data->tk_mult = tk->tkr_mono.mult; + vdso_data->tk_shift = tk->tkr_mono.shift; + smp_wmb(); + ++vdso_data->tb_update_count; +} + +extern struct timezone sys_tz; + +void update_vsyscall_tz(void) +{ + /* Make userspace gettimeofday spin until we're done. */ + ++vdso_data->tb_update_count; + smp_wmb(); + vdso_data->tz_minuteswest = sys_tz.tz_minuteswest; + vdso_data->tz_dsttime = sys_tz.tz_dsttime; + smp_wmb(); + ++vdso_data->tb_update_count; +} + +/* + * Initialize the TOD clock and the CPU timer of + * the boot cpu. + */ +void __init time_init(void) +{ + /* Reset time synchronization interfaces. */ + etr_reset(); + stp_reset(); + + /* request the clock comparator external interrupt */ + if (register_external_irq(EXT_IRQ_CLK_COMP, clock_comparator_interrupt)) + panic("Couldn't request external interrupt 0x1004"); + + /* request the timing alert external interrupt */ + if (register_external_irq(EXT_IRQ_TIMING_ALERT, timing_alert_interrupt)) + panic("Couldn't request external interrupt 0x1406"); + + if (__clocksource_register(&clocksource_tod) != 0) + panic("Could not register TOD clock source"); + + /* Enable TOD clock interrupts on the boot cpu. */ + init_cpu_timer(); + + /* Enable cpu timer interrupts on the boot cpu. */ + vtime_init(); +} + +/* + * The time is "clock". old is what we think the time is. + * Adjust the value by a multiple of jiffies and add the delta to ntp. + * "delay" is an approximation how long the synchronization took. If + * the time correction is positive, then "delay" is subtracted from + * the time difference and only the remaining part is passed to ntp. + */ +static unsigned long long adjust_time(unsigned long long old, + unsigned long long clock, + unsigned long long delay) +{ + unsigned long long delta, ticks; + struct timex adjust; + + if (clock > old) { + /* It is later than we thought. */ + delta = ticks = clock - old; + delta = ticks = (delta < delay) ? 0 : delta - delay; + delta -= do_div(ticks, CLK_TICKS_PER_JIFFY); + adjust.offset = ticks * (1000000 / HZ); + } else { + /* It is earlier than we thought. */ + delta = ticks = old - clock; + delta -= do_div(ticks, CLK_TICKS_PER_JIFFY); + delta = -delta; + adjust.offset = -ticks * (1000000 / HZ); + } + sched_clock_base_cc += delta; + if (adjust.offset != 0) { + pr_notice("The ETR interface has adjusted the clock " + "by %li microseconds\n", adjust.offset); + adjust.modes = ADJ_OFFSET_SINGLESHOT; + do_adjtimex(&adjust); + } + return delta; +} + +static DEFINE_PER_CPU(atomic_t, clock_sync_word); +static DEFINE_MUTEX(clock_sync_mutex); +static unsigned long clock_sync_flags; + +#define CLOCK_SYNC_HAS_ETR 0 +#define CLOCK_SYNC_HAS_STP 1 +#define CLOCK_SYNC_ETR 2 +#define CLOCK_SYNC_STP 3 + +/* + * The synchronous get_clock function. It will write the current clock + * value to the clock pointer and return 0 if the clock is in sync with + * the external time source. If the clock mode is local it will return + * -EOPNOTSUPP and -EAGAIN if the clock is not in sync with the external + * reference. + */ +int get_sync_clock(unsigned long long *clock) +{ + atomic_t *sw_ptr; + unsigned int sw0, sw1; + + sw_ptr = &get_cpu_var(clock_sync_word); + sw0 = atomic_read(sw_ptr); + *clock = get_tod_clock(); + sw1 = atomic_read(sw_ptr); + put_cpu_var(clock_sync_word); + if (sw0 == sw1 && (sw0 & 0x80000000U)) + /* Success: time is in sync. */ + return 0; + if (!test_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags) && + !test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags)) + return -EOPNOTSUPP; + if (!test_bit(CLOCK_SYNC_ETR, &clock_sync_flags) && + !test_bit(CLOCK_SYNC_STP, &clock_sync_flags)) + return -EACCES; + return -EAGAIN; +} +EXPORT_SYMBOL(get_sync_clock); + +/* + * Make get_sync_clock return -EAGAIN. + */ +static void disable_sync_clock(void *dummy) +{ + atomic_t *sw_ptr = this_cpu_ptr(&clock_sync_word); + /* + * Clear the in-sync bit 2^31. All get_sync_clock calls will + * fail until the sync bit is turned back on. In addition + * increase the "sequence" counter to avoid the race of an + * etr event and the complete recovery against get_sync_clock. + */ + atomic_clear_mask(0x80000000, sw_ptr); + atomic_inc(sw_ptr); +} + +/* + * Make get_sync_clock return 0 again. + * Needs to be called from a context disabled for preemption. + */ +static void enable_sync_clock(void) +{ + atomic_t *sw_ptr = this_cpu_ptr(&clock_sync_word); + atomic_set_mask(0x80000000, sw_ptr); +} + +/* + * Function to check if the clock is in sync. + */ +static inline int check_sync_clock(void) +{ + atomic_t *sw_ptr; + int rc; + + sw_ptr = &get_cpu_var(clock_sync_word); + rc = (atomic_read(sw_ptr) & 0x80000000U) != 0; + put_cpu_var(clock_sync_word); + return rc; +} + +/* Single threaded workqueue used for etr and stp sync events */ +static struct workqueue_struct *time_sync_wq; + +static void __init time_init_wq(void) +{ + if (time_sync_wq) + return; + time_sync_wq = create_singlethread_workqueue("timesync"); +} + +/* + * External Time Reference (ETR) code. + */ +static int etr_port0_online; +static int etr_port1_online; +static int etr_steai_available; + +static int __init early_parse_etr(char *p) +{ + if (strncmp(p, "off", 3) == 0) + etr_port0_online = etr_port1_online = 0; + else if (strncmp(p, "port0", 5) == 0) + etr_port0_online = 1; + else if (strncmp(p, "port1", 5) == 0) + etr_port1_online = 1; + else if (strncmp(p, "on", 2) == 0) + etr_port0_online = etr_port1_online = 1; + return 0; +} +early_param("etr", early_parse_etr); + +enum etr_event { + ETR_EVENT_PORT0_CHANGE, + ETR_EVENT_PORT1_CHANGE, + ETR_EVENT_PORT_ALERT, + ETR_EVENT_SYNC_CHECK, + ETR_EVENT_SWITCH_LOCAL, + ETR_EVENT_UPDATE, +}; + +/* + * Valid bit combinations of the eacr register are (x = don't care): + * e0 e1 dp p0 p1 ea es sl + * 0 0 x 0 0 0 0 0 initial, disabled state + * 0 0 x 0 1 1 0 0 port 1 online + * 0 0 x 1 0 1 0 0 port 0 online + * 0 0 x 1 1 1 0 0 both ports online + * 0 1 x 0 1 1 0 0 port 1 online and usable, ETR or PPS mode + * 0 1 x 0 1 1 0 1 port 1 online, usable and ETR mode + * 0 1 x 0 1 1 1 0 port 1 online, usable, PPS mode, in-sync + * 0 1 x 0 1 1 1 1 port 1 online, usable, ETR mode, in-sync + * 0 1 x 1 1 1 0 0 both ports online, port 1 usable + * 0 1 x 1 1 1 1 0 both ports online, port 1 usable, PPS mode, in-sync + * 0 1 x 1 1 1 1 1 both ports online, port 1 usable, ETR mode, in-sync + * 1 0 x 1 0 1 0 0 port 0 online and usable, ETR or PPS mode + * 1 0 x 1 0 1 0 1 port 0 online, usable and ETR mode + * 1 0 x 1 0 1 1 0 port 0 online, usable, PPS mode, in-sync + * 1 0 x 1 0 1 1 1 port 0 online, usable, ETR mode, in-sync + * 1 0 x 1 1 1 0 0 both ports online, port 0 usable + * 1 0 x 1 1 1 1 0 both ports online, port 0 usable, PPS mode, in-sync + * 1 0 x 1 1 1 1 1 both ports online, port 0 usable, ETR mode, in-sync + * 1 1 x 1 1 1 1 0 both ports online & usable, ETR, in-sync + * 1 1 x 1 1 1 1 1 both ports online & usable, ETR, in-sync + */ +static struct etr_eacr etr_eacr; +static u64 etr_tolec; /* time of last eacr update */ +static struct etr_aib etr_port0; +static int etr_port0_uptodate; +static struct etr_aib etr_port1; +static int etr_port1_uptodate; +static unsigned long etr_events; +static struct timer_list etr_timer; + +static void etr_timeout(unsigned long dummy); +static void etr_work_fn(struct work_struct *work); +static DEFINE_MUTEX(etr_work_mutex); +static DECLARE_WORK(etr_work, etr_work_fn); + +/* + * Reset ETR attachment. + */ +static void etr_reset(void) +{ + etr_eacr = (struct etr_eacr) { + .e0 = 0, .e1 = 0, ._pad0 = 4, .dp = 0, + .p0 = 0, .p1 = 0, ._pad1 = 0, .ea = 0, + .es = 0, .sl = 0 }; + if (etr_setr(&etr_eacr) == 0) { + etr_tolec = get_tod_clock(); + set_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags); + if (etr_port0_online && etr_port1_online) + set_bit(CLOCK_SYNC_ETR, &clock_sync_flags); + } else if (etr_port0_online || etr_port1_online) { + pr_warning("The real or virtual hardware system does " + "not provide an ETR interface\n"); + etr_port0_online = etr_port1_online = 0; + } +} + +static int __init etr_init(void) +{ + struct etr_aib aib; + + if (!test_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags)) + return 0; + time_init_wq(); + /* Check if this machine has the steai instruction. */ + if (etr_steai(&aib, ETR_STEAI_STEPPING_PORT) == 0) + etr_steai_available = 1; + setup_timer(&etr_timer, etr_timeout, 0UL); + if (etr_port0_online) { + set_bit(ETR_EVENT_PORT0_CHANGE, &etr_events); + queue_work(time_sync_wq, &etr_work); + } + if (etr_port1_online) { + set_bit(ETR_EVENT_PORT1_CHANGE, &etr_events); + queue_work(time_sync_wq, &etr_work); + } + return 0; +} + +arch_initcall(etr_init); + +/* + * Two sorts of ETR machine checks. The architecture reads: + * "When a machine-check niterruption occurs and if a switch-to-local or + * ETR-sync-check interrupt request is pending but disabled, this pending + * disabled interruption request is indicated and is cleared". + * Which means that we can get etr_switch_to_local events from the machine + * check handler although the interruption condition is disabled. Lovely.. + */ + +/* + * Switch to local machine check. This is called when the last usable + * ETR port goes inactive. After switch to local the clock is not in sync. + */ +void etr_switch_to_local(void) +{ + if (!etr_eacr.sl) + return; + disable_sync_clock(NULL); + if (!test_and_set_bit(ETR_EVENT_SWITCH_LOCAL, &etr_events)) { + etr_eacr.es = etr_eacr.sl = 0; + etr_setr(&etr_eacr); + queue_work(time_sync_wq, &etr_work); + } +} + +/* + * ETR sync check machine check. This is called when the ETR OTE and the + * local clock OTE are farther apart than the ETR sync check tolerance. + * After a ETR sync check the clock is not in sync. The machine check + * is broadcasted to all cpus at the same time. + */ +void etr_sync_check(void) +{ + if (!etr_eacr.es) + return; + disable_sync_clock(NULL); + if (!test_and_set_bit(ETR_EVENT_SYNC_CHECK, &etr_events)) { + etr_eacr.es = 0; + etr_setr(&etr_eacr); + queue_work(time_sync_wq, &etr_work); + } +} + +/* + * ETR timing alert. There are two causes: + * 1) port state change, check the usability of the port + * 2) port alert, one of the ETR-data-validity bits (v1-v2 bits of the + * sldr-status word) or ETR-data word 1 (edf1) or ETR-data word 3 (edf3) + * or ETR-data word 4 (edf4) has changed. + */ +static void etr_timing_alert(struct etr_irq_parm *intparm) +{ + if (intparm->pc0) + /* ETR port 0 state change. */ + set_bit(ETR_EVENT_PORT0_CHANGE, &etr_events); + if (intparm->pc1) + /* ETR port 1 state change. */ + set_bit(ETR_EVENT_PORT1_CHANGE, &etr_events); + if (intparm->eai) + /* + * ETR port alert on either port 0, 1 or both. + * Both ports are not up-to-date now. + */ + set_bit(ETR_EVENT_PORT_ALERT, &etr_events); + queue_work(time_sync_wq, &etr_work); +} + +static void etr_timeout(unsigned long dummy) +{ + set_bit(ETR_EVENT_UPDATE, &etr_events); + queue_work(time_sync_wq, &etr_work); +} + +/* + * Check if the etr mode is pss. + */ +static inline int etr_mode_is_pps(struct etr_eacr eacr) +{ + return eacr.es && !eacr.sl; +} + +/* + * Check if the etr mode is etr. + */ +static inline int etr_mode_is_etr(struct etr_eacr eacr) +{ + return eacr.es && eacr.sl; +} + +/* + * Check if the port can be used for TOD synchronization. + * For PPS mode the port has to receive OTEs. For ETR mode + * the port has to receive OTEs, the ETR stepping bit has to + * be zero and the validity bits for data frame 1, 2, and 3 + * have to be 1. + */ +static int etr_port_valid(struct etr_aib *aib, int port) +{ + unsigned int psc; + + /* Check that this port is receiving OTEs. */ + if (aib->tsp == 0) + return 0; + + psc = port ? aib->esw.psc1 : aib->esw.psc0; + if (psc == etr_lpsc_pps_mode) + return 1; + if (psc == etr_lpsc_operational_step) + return !aib->esw.y && aib->slsw.v1 && + aib->slsw.v2 && aib->slsw.v3; + return 0; +} + +/* + * Check if two ports are on the same network. + */ +static int etr_compare_network(struct etr_aib *aib1, struct etr_aib *aib2) +{ + // FIXME: any other fields we have to compare? + return aib1->edf1.net_id == aib2->edf1.net_id; +} + +/* + * Wrapper for etr_stei that converts physical port states + * to logical port states to be consistent with the output + * of stetr (see etr_psc vs. etr_lpsc). + */ +static void etr_steai_cv(struct etr_aib *aib, unsigned int func) +{ + BUG_ON(etr_steai(aib, func) != 0); + /* Convert port state to logical port state. */ + if (aib->esw.psc0 == 1) + aib->esw.psc0 = 2; + else if (aib->esw.psc0 == 0 && aib->esw.p == 0) + aib->esw.psc0 = 1; + if (aib->esw.psc1 == 1) + aib->esw.psc1 = 2; + else if (aib->esw.psc1 == 0 && aib->esw.p == 1) + aib->esw.psc1 = 1; +} + +/* + * Check if the aib a2 is still connected to the same attachment as + * aib a1, the etv values differ by one and a2 is valid. + */ +static int etr_aib_follows(struct etr_aib *a1, struct etr_aib *a2, int p) +{ + int state_a1, state_a2; + + /* Paranoia check: e0/e1 should better be the same. */ + if (a1->esw.eacr.e0 != a2->esw.eacr.e0 || + a1->esw.eacr.e1 != a2->esw.eacr.e1) + return 0; + + /* Still connected to the same etr ? */ + state_a1 = p ? a1->esw.psc1 : a1->esw.psc0; + state_a2 = p ? a2->esw.psc1 : a2->esw.psc0; + if (state_a1 == etr_lpsc_operational_step) { + if (state_a2 != etr_lpsc_operational_step || + a1->edf1.net_id != a2->edf1.net_id || + a1->edf1.etr_id != a2->edf1.etr_id || + a1->edf1.etr_pn != a2->edf1.etr_pn) + return 0; + } else if (state_a2 != etr_lpsc_pps_mode) + return 0; + + /* The ETV value of a2 needs to be ETV of a1 + 1. */ + if (a1->edf2.etv + 1 != a2->edf2.etv) + return 0; + + if (!etr_port_valid(a2, p)) + return 0; + + return 1; +} + +struct clock_sync_data { + atomic_t cpus; + int in_sync; + unsigned long long fixup_cc; + int etr_port; + struct etr_aib *etr_aib; +}; + +static void clock_sync_cpu(struct clock_sync_data *sync) +{ + atomic_dec(&sync->cpus); + enable_sync_clock(); + /* + * This looks like a busy wait loop but it isn't. etr_sync_cpus + * is called on all other cpus while the TOD clocks is stopped. + * __udelay will stop the cpu on an enabled wait psw until the + * TOD is running again. + */ + while (sync->in_sync == 0) { + __udelay(1); + /* + * A different cpu changes *in_sync. Therefore use + * barrier() to force memory access. + */ + barrier(); + } + if (sync->in_sync != 1) + /* Didn't work. Clear per-cpu in sync bit again. */ + disable_sync_clock(NULL); + /* + * This round of TOD syncing is done. Set the clock comparator + * to the next tick and let the processor continue. + */ + fixup_clock_comparator(sync->fixup_cc); +} + +/* + * Sync the TOD clock using the port referred to by aibp. This port + * has to be enabled and the other port has to be disabled. The + * last eacr update has to be more than 1.6 seconds in the past. + */ +static int etr_sync_clock(void *data) +{ + static int first; + unsigned long long clock, old_clock, delay, delta; + struct clock_sync_data *etr_sync; + struct etr_aib *sync_port, *aib; + int port; + int rc; + + etr_sync = data; + + if (xchg(&first, 1) == 1) { + /* Slave */ + clock_sync_cpu(etr_sync); + return 0; + } + + /* Wait until all other cpus entered the sync function. */ + while (atomic_read(&etr_sync->cpus) != 0) + cpu_relax(); + + port = etr_sync->etr_port; + aib = etr_sync->etr_aib; + sync_port = (port == 0) ? &etr_port0 : &etr_port1; + enable_sync_clock(); + + /* Set clock to next OTE. */ + __ctl_set_bit(14, 21); + __ctl_set_bit(0, 29); + clock = ((unsigned long long) (aib->edf2.etv + 1)) << 32; + old_clock = get_tod_clock(); + if (set_tod_clock(clock) == 0) { + __udelay(1); /* Wait for the clock to start. */ + __ctl_clear_bit(0, 29); + __ctl_clear_bit(14, 21); + etr_stetr(aib); + /* Adjust Linux timing variables. */ + delay = (unsigned long long) + (aib->edf2.etv - sync_port->edf2.etv) << 32; + delta = adjust_time(old_clock, clock, delay); + etr_sync->fixup_cc = delta; + fixup_clock_comparator(delta); + /* Verify that the clock is properly set. */ + if (!etr_aib_follows(sync_port, aib, port)) { + /* Didn't work. */ + disable_sync_clock(NULL); + etr_sync->in_sync = -EAGAIN; + rc = -EAGAIN; + } else { + etr_sync->in_sync = 1; + rc = 0; + } + } else { + /* Could not set the clock ?!? */ + __ctl_clear_bit(0, 29); + __ctl_clear_bit(14, 21); + disable_sync_clock(NULL); + etr_sync->in_sync = -EAGAIN; + rc = -EAGAIN; + } + xchg(&first, 0); + return rc; +} + +static int etr_sync_clock_stop(struct etr_aib *aib, int port) +{ + struct clock_sync_data etr_sync; + struct etr_aib *sync_port; + int follows; + int rc; + + /* Check if the current aib is adjacent to the sync port aib. */ + sync_port = (port == 0) ? &etr_port0 : &etr_port1; + follows = etr_aib_follows(sync_port, aib, port); + memcpy(sync_port, aib, sizeof(*aib)); + if (!follows) + return -EAGAIN; + memset(&etr_sync, 0, sizeof(etr_sync)); + etr_sync.etr_aib = aib; + etr_sync.etr_port = port; + get_online_cpus(); + atomic_set(&etr_sync.cpus, num_online_cpus() - 1); + rc = stop_machine(etr_sync_clock, &etr_sync, cpu_online_mask); + put_online_cpus(); + return rc; +} + +/* + * Handle the immediate effects of the different events. + * The port change event is used for online/offline changes. + */ +static struct etr_eacr etr_handle_events(struct etr_eacr eacr) +{ + if (test_and_clear_bit(ETR_EVENT_SYNC_CHECK, &etr_events)) + eacr.es = 0; + if (test_and_clear_bit(ETR_EVENT_SWITCH_LOCAL, &etr_events)) + eacr.es = eacr.sl = 0; + if (test_and_clear_bit(ETR_EVENT_PORT_ALERT, &etr_events)) + etr_port0_uptodate = etr_port1_uptodate = 0; + + if (test_and_clear_bit(ETR_EVENT_PORT0_CHANGE, &etr_events)) { + if (eacr.e0) + /* + * Port change of an enabled port. We have to + * assume that this can have caused an stepping + * port switch. + */ + etr_tolec = get_tod_clock(); + eacr.p0 = etr_port0_online; + if (!eacr.p0) + eacr.e0 = 0; + etr_port0_uptodate = 0; + } + if (test_and_clear_bit(ETR_EVENT_PORT1_CHANGE, &etr_events)) { + if (eacr.e1) + /* + * Port change of an enabled port. We have to + * assume that this can have caused an stepping + * port switch. + */ + etr_tolec = get_tod_clock(); + eacr.p1 = etr_port1_online; + if (!eacr.p1) + eacr.e1 = 0; + etr_port1_uptodate = 0; + } + clear_bit(ETR_EVENT_UPDATE, &etr_events); + return eacr; +} + +/* + * Set up a timer that expires after the etr_tolec + 1.6 seconds if + * one of the ports needs an update. + */ +static void etr_set_tolec_timeout(unsigned long long now) +{ + unsigned long micros; + + if ((!etr_eacr.p0 || etr_port0_uptodate) && + (!etr_eacr.p1 || etr_port1_uptodate)) + return; + micros = (now > etr_tolec) ? ((now - etr_tolec) >> 12) : 0; + micros = (micros > 1600000) ? 0 : 1600000 - micros; + mod_timer(&etr_timer, jiffies + (micros * HZ) / 1000000 + 1); +} + +/* + * Set up a time that expires after 1/2 second. + */ +static void etr_set_sync_timeout(void) +{ + mod_timer(&etr_timer, jiffies + HZ/2); +} + +/* + * Update the aib information for one or both ports. + */ +static struct etr_eacr etr_handle_update(struct etr_aib *aib, + struct etr_eacr eacr) +{ + /* With both ports disabled the aib information is useless. */ + if (!eacr.e0 && !eacr.e1) + return eacr; + + /* Update port0 or port1 with aib stored in etr_work_fn. */ + if (aib->esw.q == 0) { + /* Information for port 0 stored. */ + if (eacr.p0 && !etr_port0_uptodate) { + etr_port0 = *aib; + if (etr_port0_online) + etr_port0_uptodate = 1; + } + } else { + /* Information for port 1 stored. */ + if (eacr.p1 && !etr_port1_uptodate) { + etr_port1 = *aib; + if (etr_port0_online) + etr_port1_uptodate = 1; + } + } + + /* + * Do not try to get the alternate port aib if the clock + * is not in sync yet. + */ + if (!eacr.es || !check_sync_clock()) + return eacr; + + /* + * If steai is available we can get the information about + * the other port immediately. If only stetr is available the + * data-port bit toggle has to be used. + */ + if (etr_steai_available) { + if (eacr.p0 && !etr_port0_uptodate) { + etr_steai_cv(&etr_port0, ETR_STEAI_PORT_0); + etr_port0_uptodate = 1; + } + if (eacr.p1 && !etr_port1_uptodate) { + etr_steai_cv(&etr_port1, ETR_STEAI_PORT_1); + etr_port1_uptodate = 1; + } + } else { + /* + * One port was updated above, if the other + * port is not uptodate toggle dp bit. + */ + if ((eacr.p0 && !etr_port0_uptodate) || + (eacr.p1 && !etr_port1_uptodate)) + eacr.dp ^= 1; + else + eacr.dp = 0; + } + return eacr; +} + +/* + * Write new etr control register if it differs from the current one. + * Return 1 if etr_tolec has been updated as well. + */ +static void etr_update_eacr(struct etr_eacr eacr) +{ + int dp_changed; + + if (memcmp(&etr_eacr, &eacr, sizeof(eacr)) == 0) + /* No change, return. */ + return; + /* + * The disable of an active port of the change of the data port + * bit can/will cause a change in the data port. + */ + dp_changed = etr_eacr.e0 > eacr.e0 || etr_eacr.e1 > eacr.e1 || + (etr_eacr.dp ^ eacr.dp) != 0; + etr_eacr = eacr; + etr_setr(&etr_eacr); + if (dp_changed) + etr_tolec = get_tod_clock(); +} + +/* + * ETR work. In this function you'll find the main logic. In + * particular this is the only function that calls etr_update_eacr(), + * it "controls" the etr control register. + */ +static void etr_work_fn(struct work_struct *work) +{ + unsigned long long now; + struct etr_eacr eacr; + struct etr_aib aib; + int sync_port; + + /* prevent multiple execution. */ + mutex_lock(&etr_work_mutex); + + /* Create working copy of etr_eacr. */ + eacr = etr_eacr; + + /* Check for the different events and their immediate effects. */ + eacr = etr_handle_events(eacr); + + /* Check if ETR is supposed to be active. */ + eacr.ea = eacr.p0 || eacr.p1; + if (!eacr.ea) { + /* Both ports offline. Reset everything. */ + eacr.dp = eacr.es = eacr.sl = 0; + on_each_cpu(disable_sync_clock, NULL, 1); + del_timer_sync(&etr_timer); + etr_update_eacr(eacr); + goto out_unlock; + } + + /* Store aib to get the current ETR status word. */ + BUG_ON(etr_stetr(&aib) != 0); + etr_port0.esw = etr_port1.esw = aib.esw; /* Copy status word. */ + now = get_tod_clock(); + + /* + * Update the port information if the last stepping port change + * or data port change is older than 1.6 seconds. + */ + if (now >= etr_tolec + (1600000 << 12)) + eacr = etr_handle_update(&aib, eacr); + + /* + * Select ports to enable. The preferred synchronization mode is PPS. + * If a port can be enabled depends on a number of things: + * 1) The port needs to be online and uptodate. A port is not + * disabled just because it is not uptodate, but it is only + * enabled if it is uptodate. + * 2) The port needs to have the same mode (pps / etr). + * 3) The port needs to be usable -> etr_port_valid() == 1 + * 4) To enable the second port the clock needs to be in sync. + * 5) If both ports are useable and are ETR ports, the network id + * has to be the same. + * The eacr.sl bit is used to indicate etr mode vs. pps mode. + */ + if (eacr.p0 && aib.esw.psc0 == etr_lpsc_pps_mode) { + eacr.sl = 0; + eacr.e0 = 1; + if (!etr_mode_is_pps(etr_eacr)) + eacr.es = 0; + if (!eacr.es || !eacr.p1 || aib.esw.psc1 != etr_lpsc_pps_mode) + eacr.e1 = 0; + // FIXME: uptodate checks ? + else if (etr_port0_uptodate && etr_port1_uptodate) + eacr.e1 = 1; + sync_port = (etr_port0_uptodate && + etr_port_valid(&etr_port0, 0)) ? 0 : -1; + } else if (eacr.p1 && aib.esw.psc1 == etr_lpsc_pps_mode) { + eacr.sl = 0; + eacr.e0 = 0; + eacr.e1 = 1; + if (!etr_mode_is_pps(etr_eacr)) + eacr.es = 0; + sync_port = (etr_port1_uptodate && + etr_port_valid(&etr_port1, 1)) ? 1 : -1; + } else if (eacr.p0 && aib.esw.psc0 == etr_lpsc_operational_step) { + eacr.sl = 1; + eacr.e0 = 1; + if (!etr_mode_is_etr(etr_eacr)) + eacr.es = 0; + if (!eacr.es || !eacr.p1 || + aib.esw.psc1 != etr_lpsc_operational_alt) + eacr.e1 = 0; + else if (etr_port0_uptodate && etr_port1_uptodate && + etr_compare_network(&etr_port0, &etr_port1)) + eacr.e1 = 1; + sync_port = (etr_port0_uptodate && + etr_port_valid(&etr_port0, 0)) ? 0 : -1; + } else if (eacr.p1 && aib.esw.psc1 == etr_lpsc_operational_step) { + eacr.sl = 1; + eacr.e0 = 0; + eacr.e1 = 1; + if (!etr_mode_is_etr(etr_eacr)) + eacr.es = 0; + sync_port = (etr_port1_uptodate && + etr_port_valid(&etr_port1, 1)) ? 1 : -1; + } else { + /* Both ports not usable. */ + eacr.es = eacr.sl = 0; + sync_port = -1; + } + + /* + * If the clock is in sync just update the eacr and return. + * If there is no valid sync port wait for a port update. + */ + if ((eacr.es && check_sync_clock()) || sync_port < 0) { + etr_update_eacr(eacr); + etr_set_tolec_timeout(now); + goto out_unlock; + } + + /* + * Prepare control register for clock syncing + * (reset data port bit, set sync check control. + */ + eacr.dp = 0; + eacr.es = 1; + + /* + * Update eacr and try to synchronize the clock. If the update + * of eacr caused a stepping port switch (or if we have to + * assume that a stepping port switch has occurred) or the + * clock syncing failed, reset the sync check control bit + * and set up a timer to try again after 0.5 seconds + */ + etr_update_eacr(eacr); + if (now < etr_tolec + (1600000 << 12) || + etr_sync_clock_stop(&aib, sync_port) != 0) { + /* Sync failed. Try again in 1/2 second. */ + eacr.es = 0; + etr_update_eacr(eacr); + etr_set_sync_timeout(); + } else + etr_set_tolec_timeout(now); +out_unlock: + mutex_unlock(&etr_work_mutex); +} + +/* + * Sysfs interface functions + */ +static struct bus_type etr_subsys = { + .name = "etr", + .dev_name = "etr", +}; + +static struct device etr_port0_dev = { + .id = 0, + .bus = &etr_subsys, +}; + +static struct device etr_port1_dev = { + .id = 1, + .bus = &etr_subsys, +}; + +/* + * ETR subsys attributes + */ +static ssize_t etr_stepping_port_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return sprintf(buf, "%i\n", etr_port0.esw.p); +} + +static DEVICE_ATTR(stepping_port, 0400, etr_stepping_port_show, NULL); + +static ssize_t etr_stepping_mode_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + char *mode_str; + + if (etr_mode_is_pps(etr_eacr)) + mode_str = "pps"; + else if (etr_mode_is_etr(etr_eacr)) + mode_str = "etr"; + else + mode_str = "local"; + return sprintf(buf, "%s\n", mode_str); +} + +static DEVICE_ATTR(stepping_mode, 0400, etr_stepping_mode_show, NULL); + +/* + * ETR port attributes + */ +static inline struct etr_aib *etr_aib_from_dev(struct device *dev) +{ + if (dev == &etr_port0_dev) + return etr_port0_online ? &etr_port0 : NULL; + else + return etr_port1_online ? &etr_port1 : NULL; +} + +static ssize_t etr_online_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + unsigned int online; + + online = (dev == &etr_port0_dev) ? etr_port0_online : etr_port1_online; + return sprintf(buf, "%i\n", online); +} + +static ssize_t etr_online_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + unsigned int value; + + value = simple_strtoul(buf, NULL, 0); + if (value != 0 && value != 1) + return -EINVAL; + if (!test_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags)) + return -EOPNOTSUPP; + mutex_lock(&clock_sync_mutex); + if (dev == &etr_port0_dev) { + if (etr_port0_online == value) + goto out; /* Nothing to do. */ + etr_port0_online = value; + if (etr_port0_online && etr_port1_online) + set_bit(CLOCK_SYNC_ETR, &clock_sync_flags); + else + clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags); + set_bit(ETR_EVENT_PORT0_CHANGE, &etr_events); + queue_work(time_sync_wq, &etr_work); + } else { + if (etr_port1_online == value) + goto out; /* Nothing to do. */ + etr_port1_online = value; + if (etr_port0_online && etr_port1_online) + set_bit(CLOCK_SYNC_ETR, &clock_sync_flags); + else + clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags); + set_bit(ETR_EVENT_PORT1_CHANGE, &etr_events); + queue_work(time_sync_wq, &etr_work); + } +out: + mutex_unlock(&clock_sync_mutex); + return count; +} + +static DEVICE_ATTR(online, 0600, etr_online_show, etr_online_store); + +static ssize_t etr_stepping_control_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return sprintf(buf, "%i\n", (dev == &etr_port0_dev) ? + etr_eacr.e0 : etr_eacr.e1); +} + +static DEVICE_ATTR(stepping_control, 0400, etr_stepping_control_show, NULL); + +static ssize_t etr_mode_code_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + if (!etr_port0_online && !etr_port1_online) + /* Status word is not uptodate if both ports are offline. */ + return -ENODATA; + return sprintf(buf, "%i\n", (dev == &etr_port0_dev) ? + etr_port0.esw.psc0 : etr_port0.esw.psc1); +} + +static DEVICE_ATTR(state_code, 0400, etr_mode_code_show, NULL); + +static ssize_t etr_untuned_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct etr_aib *aib = etr_aib_from_dev(dev); + + if (!aib || !aib->slsw.v1) + return -ENODATA; + return sprintf(buf, "%i\n", aib->edf1.u); +} + +static DEVICE_ATTR(untuned, 0400, etr_untuned_show, NULL); + +static ssize_t etr_network_id_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct etr_aib *aib = etr_aib_from_dev(dev); + + if (!aib || !aib->slsw.v1) + return -ENODATA; + return sprintf(buf, "%i\n", aib->edf1.net_id); +} + +static DEVICE_ATTR(network, 0400, etr_network_id_show, NULL); + +static ssize_t etr_id_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct etr_aib *aib = etr_aib_from_dev(dev); + + if (!aib || !aib->slsw.v1) + return -ENODATA; + return sprintf(buf, "%i\n", aib->edf1.etr_id); +} + +static DEVICE_ATTR(id, 0400, etr_id_show, NULL); + +static ssize_t etr_port_number_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct etr_aib *aib = etr_aib_from_dev(dev); + + if (!aib || !aib->slsw.v1) + return -ENODATA; + return sprintf(buf, "%i\n", aib->edf1.etr_pn); +} + +static DEVICE_ATTR(port, 0400, etr_port_number_show, NULL); + +static ssize_t etr_coupled_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct etr_aib *aib = etr_aib_from_dev(dev); + + if (!aib || !aib->slsw.v3) + return -ENODATA; + return sprintf(buf, "%i\n", aib->edf3.c); +} + +static DEVICE_ATTR(coupled, 0400, etr_coupled_show, NULL); + +static ssize_t etr_local_time_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct etr_aib *aib = etr_aib_from_dev(dev); + + if (!aib || !aib->slsw.v3) + return -ENODATA; + return sprintf(buf, "%i\n", aib->edf3.blto); +} + +static DEVICE_ATTR(local_time, 0400, etr_local_time_show, NULL); + +static ssize_t etr_utc_offset_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct etr_aib *aib = etr_aib_from_dev(dev); + + if (!aib || !aib->slsw.v3) + return -ENODATA; + return sprintf(buf, "%i\n", aib->edf3.buo); +} + +static DEVICE_ATTR(utc_offset, 0400, etr_utc_offset_show, NULL); + +static struct device_attribute *etr_port_attributes[] = { + &dev_attr_online, + &dev_attr_stepping_control, + &dev_attr_state_code, + &dev_attr_untuned, + &dev_attr_network, + &dev_attr_id, + &dev_attr_port, + &dev_attr_coupled, + &dev_attr_local_time, + &dev_attr_utc_offset, + NULL +}; + +static int __init etr_register_port(struct device *dev) +{ + struct device_attribute **attr; + int rc; + + rc = device_register(dev); + if (rc) + goto out; + for (attr = etr_port_attributes; *attr; attr++) { + rc = device_create_file(dev, *attr); + if (rc) + goto out_unreg; + } + return 0; +out_unreg: + for (; attr >= etr_port_attributes; attr--) + device_remove_file(dev, *attr); + device_unregister(dev); +out: + return rc; +} + +static void __init etr_unregister_port(struct device *dev) +{ + struct device_attribute **attr; + + for (attr = etr_port_attributes; *attr; attr++) + device_remove_file(dev, *attr); + device_unregister(dev); +} + +static int __init etr_init_sysfs(void) +{ + int rc; + + rc = subsys_system_register(&etr_subsys, NULL); + if (rc) + goto out; + rc = device_create_file(etr_subsys.dev_root, &dev_attr_stepping_port); + if (rc) + goto out_unreg_subsys; + rc = device_create_file(etr_subsys.dev_root, &dev_attr_stepping_mode); + if (rc) + goto out_remove_stepping_port; + rc = etr_register_port(&etr_port0_dev); + if (rc) + goto out_remove_stepping_mode; + rc = etr_register_port(&etr_port1_dev); + if (rc) + goto out_remove_port0; + return 0; + +out_remove_port0: + etr_unregister_port(&etr_port0_dev); +out_remove_stepping_mode: + device_remove_file(etr_subsys.dev_root, &dev_attr_stepping_mode); +out_remove_stepping_port: + device_remove_file(etr_subsys.dev_root, &dev_attr_stepping_port); +out_unreg_subsys: + bus_unregister(&etr_subsys); +out: + return rc; +} + +device_initcall(etr_init_sysfs); + +/* + * Server Time Protocol (STP) code. + */ +static int stp_online; +static struct stp_sstpi stp_info; +static void *stp_page; + +static void stp_work_fn(struct work_struct *work); +static DEFINE_MUTEX(stp_work_mutex); +static DECLARE_WORK(stp_work, stp_work_fn); +static struct timer_list stp_timer; + +static int __init early_parse_stp(char *p) +{ + if (strncmp(p, "off", 3) == 0) + stp_online = 0; + else if (strncmp(p, "on", 2) == 0) + stp_online = 1; + return 0; +} +early_param("stp", early_parse_stp); + +/* + * Reset STP attachment. + */ +static void __init stp_reset(void) +{ + int rc; + + stp_page = (void *) get_zeroed_page(GFP_ATOMIC); + rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000); + if (rc == 0) + set_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags); + else if (stp_online) { + pr_warning("The real or virtual hardware system does " + "not provide an STP interface\n"); + free_page((unsigned long) stp_page); + stp_page = NULL; + stp_online = 0; + } +} + +static void stp_timeout(unsigned long dummy) +{ + queue_work(time_sync_wq, &stp_work); +} + +static int __init stp_init(void) +{ + if (!test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags)) + return 0; + setup_timer(&stp_timer, stp_timeout, 0UL); + time_init_wq(); + if (!stp_online) + return 0; + queue_work(time_sync_wq, &stp_work); + return 0; +} + +arch_initcall(stp_init); + +/* + * STP timing alert. There are three causes: + * 1) timing status change + * 2) link availability change + * 3) time control parameter change + * In all three cases we are only interested in the clock source state. + * If a STP clock source is now available use it. + */ +static void stp_timing_alert(struct stp_irq_parm *intparm) +{ + if (intparm->tsc || intparm->lac || intparm->tcpc) + queue_work(time_sync_wq, &stp_work); +} + +/* + * STP sync check machine check. This is called when the timing state + * changes from the synchronized state to the unsynchronized state. + * After a STP sync check the clock is not in sync. The machine check + * is broadcasted to all cpus at the same time. + */ +void stp_sync_check(void) +{ + disable_sync_clock(NULL); + queue_work(time_sync_wq, &stp_work); +} + +/* + * STP island condition machine check. This is called when an attached + * server attempts to communicate over an STP link and the servers + * have matching CTN ids and have a valid stratum-1 configuration + * but the configurations do not match. + */ +void stp_island_check(void) +{ + disable_sync_clock(NULL); + queue_work(time_sync_wq, &stp_work); +} + + +static int stp_sync_clock(void *data) +{ + static int first; + unsigned long long old_clock, delta; + struct clock_sync_data *stp_sync; + int rc; + + stp_sync = data; + + if (xchg(&first, 1) == 1) { + /* Slave */ + clock_sync_cpu(stp_sync); + return 0; + } + + /* Wait until all other cpus entered the sync function. */ + while (atomic_read(&stp_sync->cpus) != 0) + cpu_relax(); + + enable_sync_clock(); + + rc = 0; + if (stp_info.todoff[0] || stp_info.todoff[1] || + stp_info.todoff[2] || stp_info.todoff[3] || + stp_info.tmd != 2) { + old_clock = get_tod_clock(); + rc = chsc_sstpc(stp_page, STP_OP_SYNC, 0); + if (rc == 0) { + delta = adjust_time(old_clock, get_tod_clock(), 0); + fixup_clock_comparator(delta); + rc = chsc_sstpi(stp_page, &stp_info, + sizeof(struct stp_sstpi)); + if (rc == 0 && stp_info.tmd != 2) + rc = -EAGAIN; + } + } + if (rc) { + disable_sync_clock(NULL); + stp_sync->in_sync = -EAGAIN; + } else + stp_sync->in_sync = 1; + xchg(&first, 0); + return 0; +} + +/* + * STP work. Check for the STP state and take over the clock + * synchronization if the STP clock source is usable. + */ +static void stp_work_fn(struct work_struct *work) +{ + struct clock_sync_data stp_sync; + int rc; + + /* prevent multiple execution. */ + mutex_lock(&stp_work_mutex); + + if (!stp_online) { + chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000); + del_timer_sync(&stp_timer); + goto out_unlock; + } + + rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0xb0e0); + if (rc) + goto out_unlock; + + rc = chsc_sstpi(stp_page, &stp_info, sizeof(struct stp_sstpi)); + if (rc || stp_info.c == 0) + goto out_unlock; + + /* Skip synchronization if the clock is already in sync. */ + if (check_sync_clock()) + goto out_unlock; + + memset(&stp_sync, 0, sizeof(stp_sync)); + get_online_cpus(); + atomic_set(&stp_sync.cpus, num_online_cpus() - 1); + stop_machine(stp_sync_clock, &stp_sync, cpu_online_mask); + put_online_cpus(); + + if (!check_sync_clock()) + /* + * There is a usable clock but the synchonization failed. + * Retry after a second. + */ + mod_timer(&stp_timer, jiffies + HZ); + +out_unlock: + mutex_unlock(&stp_work_mutex); +} + +/* + * STP subsys sysfs interface functions + */ +static struct bus_type stp_subsys = { + .name = "stp", + .dev_name = "stp", +}; + +static ssize_t stp_ctn_id_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%016llx\n", + *(unsigned long long *) stp_info.ctnid); +} + +static DEVICE_ATTR(ctn_id, 0400, stp_ctn_id_show, NULL); + +static ssize_t stp_ctn_type_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%i\n", stp_info.ctn); +} + +static DEVICE_ATTR(ctn_type, 0400, stp_ctn_type_show, NULL); + +static ssize_t stp_dst_offset_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (!stp_online || !(stp_info.vbits & 0x2000)) + return -ENODATA; + return sprintf(buf, "%i\n", (int)(s16) stp_info.dsto); +} + +static DEVICE_ATTR(dst_offset, 0400, stp_dst_offset_show, NULL); + +static ssize_t stp_leap_seconds_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (!stp_online || !(stp_info.vbits & 0x8000)) + return -ENODATA; + return sprintf(buf, "%i\n", (int)(s16) stp_info.leaps); +} + +static DEVICE_ATTR(leap_seconds, 0400, stp_leap_seconds_show, NULL); + +static ssize_t stp_stratum_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%i\n", (int)(s16) stp_info.stratum); +} + +static DEVICE_ATTR(stratum, 0400, stp_stratum_show, NULL); + +static ssize_t stp_time_offset_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (!stp_online || !(stp_info.vbits & 0x0800)) + return -ENODATA; + return sprintf(buf, "%i\n", (int) stp_info.tto); +} + +static DEVICE_ATTR(time_offset, 0400, stp_time_offset_show, NULL); + +static ssize_t stp_time_zone_offset_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (!stp_online || !(stp_info.vbits & 0x4000)) + return -ENODATA; + return sprintf(buf, "%i\n", (int)(s16) stp_info.tzo); +} + +static DEVICE_ATTR(time_zone_offset, 0400, + stp_time_zone_offset_show, NULL); + +static ssize_t stp_timing_mode_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%i\n", stp_info.tmd); +} + +static DEVICE_ATTR(timing_mode, 0400, stp_timing_mode_show, NULL); + +static ssize_t stp_timing_state_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%i\n", stp_info.tst); +} + +static DEVICE_ATTR(timing_state, 0400, stp_timing_state_show, NULL); + +static ssize_t stp_online_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return sprintf(buf, "%i\n", stp_online); +} + +static ssize_t stp_online_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + unsigned int value; + + value = simple_strtoul(buf, NULL, 0); + if (value != 0 && value != 1) + return -EINVAL; + if (!test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags)) + return -EOPNOTSUPP; + mutex_lock(&clock_sync_mutex); + stp_online = value; + if (stp_online) + set_bit(CLOCK_SYNC_STP, &clock_sync_flags); + else + clear_bit(CLOCK_SYNC_STP, &clock_sync_flags); + queue_work(time_sync_wq, &stp_work); + mutex_unlock(&clock_sync_mutex); + return count; +} + +/* + * Can't use DEVICE_ATTR because the attribute should be named + * stp/online but dev_attr_online already exists in this file .. + */ +static struct device_attribute dev_attr_stp_online = { + .attr = { .name = "online", .mode = 0600 }, + .show = stp_online_show, + .store = stp_online_store, +}; + +static struct device_attribute *stp_attributes[] = { + &dev_attr_ctn_id, + &dev_attr_ctn_type, + &dev_attr_dst_offset, + &dev_attr_leap_seconds, + &dev_attr_stp_online, + &dev_attr_stratum, + &dev_attr_time_offset, + &dev_attr_time_zone_offset, + &dev_attr_timing_mode, + &dev_attr_timing_state, + NULL +}; + +static int __init stp_init_sysfs(void) +{ + struct device_attribute **attr; + int rc; + + rc = subsys_system_register(&stp_subsys, NULL); + if (rc) + goto out; + for (attr = stp_attributes; *attr; attr++) { + rc = device_create_file(stp_subsys.dev_root, *attr); + if (rc) + goto out_unreg; + } + return 0; +out_unreg: + for (; attr >= stp_attributes; attr--) + device_remove_file(stp_subsys.dev_root, *attr); + bus_unregister(&stp_subsys); +out: + return rc; +} + +device_initcall(stp_init_sysfs); diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c new file mode 100644 index 000000000..5728c5bd4 --- /dev/null +++ b/arch/s390/kernel/topology.c @@ -0,0 +1,501 @@ +/* + * Copyright IBM Corp. 2007, 2011 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#define KMSG_COMPONENT "cpu" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/workqueue.h> +#include <linux/cpuset.h> +#include <linux/device.h> +#include <linux/export.h> +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/delay.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/cpu.h> +#include <linux/smp.h> +#include <linux/mm.h> +#include <asm/sysinfo.h> + +#define PTF_HORIZONTAL (0UL) +#define PTF_VERTICAL (1UL) +#define PTF_CHECK (2UL) + +struct mask_info { + struct mask_info *next; + unsigned char id; + cpumask_t mask; +}; + +static void set_topology_timer(void); +static void topology_work_fn(struct work_struct *work); +static struct sysinfo_15_1_x *tl_info; + +static int topology_enabled = 1; +static DECLARE_WORK(topology_work, topology_work_fn); + +/* topology_lock protects the socket and book linked lists */ +static DEFINE_SPINLOCK(topology_lock); +static struct mask_info socket_info; +static struct mask_info book_info; + +DEFINE_PER_CPU(struct cpu_topology_s390, cpu_topology); +EXPORT_PER_CPU_SYMBOL_GPL(cpu_topology); + +static cpumask_t cpu_group_map(struct mask_info *info, unsigned int cpu) +{ + cpumask_t mask; + + cpumask_copy(&mask, cpumask_of(cpu)); + if (!topology_enabled || !MACHINE_HAS_TOPOLOGY) + return mask; + for (; info; info = info->next) { + if (cpumask_test_cpu(cpu, &info->mask)) + return info->mask; + } + return mask; +} + +static cpumask_t cpu_thread_map(unsigned int cpu) +{ + cpumask_t mask; + int i; + + cpumask_copy(&mask, cpumask_of(cpu)); + if (!topology_enabled || !MACHINE_HAS_TOPOLOGY) + return mask; + cpu -= cpu % (smp_cpu_mtid + 1); + for (i = 0; i <= smp_cpu_mtid; i++) + if (cpu_present(cpu + i)) + cpumask_set_cpu(cpu + i, &mask); + return mask; +} + +static struct mask_info *add_cpus_to_mask(struct topology_core *tl_core, + struct mask_info *book, + struct mask_info *socket, + int one_socket_per_cpu) +{ + unsigned int core; + + for_each_set_bit(core, &tl_core->mask[0], TOPOLOGY_CORE_BITS) { + unsigned int rcore; + int lcpu, i; + + rcore = TOPOLOGY_CORE_BITS - 1 - core + tl_core->origin; + lcpu = smp_find_processor_id(rcore << smp_cpu_mt_shift); + if (lcpu < 0) + continue; + for (i = 0; i <= smp_cpu_mtid; i++) { + per_cpu(cpu_topology, lcpu + i).book_id = book->id; + per_cpu(cpu_topology, lcpu + i).core_id = rcore; + per_cpu(cpu_topology, lcpu + i).thread_id = lcpu + i; + cpumask_set_cpu(lcpu + i, &book->mask); + cpumask_set_cpu(lcpu + i, &socket->mask); + if (one_socket_per_cpu) + per_cpu(cpu_topology, lcpu + i).socket_id = rcore; + else + per_cpu(cpu_topology, lcpu + i).socket_id = socket->id; + smp_cpu_set_polarization(lcpu + i, tl_core->pp); + } + if (one_socket_per_cpu) + socket = socket->next; + } + return socket; +} + +static void clear_masks(void) +{ + struct mask_info *info; + + info = &socket_info; + while (info) { + cpumask_clear(&info->mask); + info = info->next; + } + info = &book_info; + while (info) { + cpumask_clear(&info->mask); + info = info->next; + } +} + +static union topology_entry *next_tle(union topology_entry *tle) +{ + if (!tle->nl) + return (union topology_entry *)((struct topology_core *)tle + 1); + return (union topology_entry *)((struct topology_container *)tle + 1); +} + +static void __tl_to_masks_generic(struct sysinfo_15_1_x *info) +{ + struct mask_info *socket = &socket_info; + struct mask_info *book = &book_info; + union topology_entry *tle, *end; + + tle = info->tle; + end = (union topology_entry *)((unsigned long)info + info->length); + while (tle < end) { + switch (tle->nl) { + case 2: + book = book->next; + book->id = tle->container.id; + break; + case 1: + socket = socket->next; + socket->id = tle->container.id; + break; + case 0: + add_cpus_to_mask(&tle->cpu, book, socket, 0); + break; + default: + clear_masks(); + return; + } + tle = next_tle(tle); + } +} + +static void __tl_to_masks_z10(struct sysinfo_15_1_x *info) +{ + struct mask_info *socket = &socket_info; + struct mask_info *book = &book_info; + union topology_entry *tle, *end; + + tle = info->tle; + end = (union topology_entry *)((unsigned long)info + info->length); + while (tle < end) { + switch (tle->nl) { + case 1: + book = book->next; + book->id = tle->container.id; + break; + case 0: + socket = add_cpus_to_mask(&tle->cpu, book, socket, 1); + break; + default: + clear_masks(); + return; + } + tle = next_tle(tle); + } +} + +static void tl_to_masks(struct sysinfo_15_1_x *info) +{ + struct cpuid cpu_id; + + spin_lock_irq(&topology_lock); + get_cpu_id(&cpu_id); + clear_masks(); + switch (cpu_id.machine) { + case 0x2097: + case 0x2098: + __tl_to_masks_z10(info); + break; + default: + __tl_to_masks_generic(info); + } + spin_unlock_irq(&topology_lock); +} + +static void topology_update_polarization_simple(void) +{ + int cpu; + + mutex_lock(&smp_cpu_state_mutex); + for_each_possible_cpu(cpu) + smp_cpu_set_polarization(cpu, POLARIZATION_HRZ); + mutex_unlock(&smp_cpu_state_mutex); +} + +static int ptf(unsigned long fc) +{ + int rc; + + asm volatile( + " .insn rre,0xb9a20000,%1,%1\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (rc) + : "d" (fc) : "cc"); + return rc; +} + +int topology_set_cpu_management(int fc) +{ + int cpu, rc; + + if (!MACHINE_HAS_TOPOLOGY) + return -EOPNOTSUPP; + if (fc) + rc = ptf(PTF_VERTICAL); + else + rc = ptf(PTF_HORIZONTAL); + if (rc) + return -EBUSY; + for_each_possible_cpu(cpu) + smp_cpu_set_polarization(cpu, POLARIZATION_UNKNOWN); + return rc; +} + +static void update_cpu_masks(void) +{ + unsigned long flags; + int cpu; + + spin_lock_irqsave(&topology_lock, flags); + for_each_possible_cpu(cpu) { + per_cpu(cpu_topology, cpu).thread_mask = cpu_thread_map(cpu); + per_cpu(cpu_topology, cpu).core_mask = cpu_group_map(&socket_info, cpu); + per_cpu(cpu_topology, cpu).book_mask = cpu_group_map(&book_info, cpu); + if (!MACHINE_HAS_TOPOLOGY) { + per_cpu(cpu_topology, cpu).thread_id = cpu; + per_cpu(cpu_topology, cpu).core_id = cpu; + per_cpu(cpu_topology, cpu).socket_id = cpu; + per_cpu(cpu_topology, cpu).book_id = cpu; + } + } + spin_unlock_irqrestore(&topology_lock, flags); +} + +void store_topology(struct sysinfo_15_1_x *info) +{ + if (topology_max_mnest >= 3) + stsi(info, 15, 1, 3); + else + stsi(info, 15, 1, 2); +} + +int arch_update_cpu_topology(void) +{ + struct sysinfo_15_1_x *info = tl_info; + struct device *dev; + int cpu; + + if (!MACHINE_HAS_TOPOLOGY) { + update_cpu_masks(); + topology_update_polarization_simple(); + return 0; + } + store_topology(info); + tl_to_masks(info); + update_cpu_masks(); + for_each_online_cpu(cpu) { + dev = get_cpu_device(cpu); + kobject_uevent(&dev->kobj, KOBJ_CHANGE); + } + return 1; +} + +static void topology_work_fn(struct work_struct *work) +{ + rebuild_sched_domains(); +} + +void topology_schedule_update(void) +{ + schedule_work(&topology_work); +} + +static void topology_timer_fn(unsigned long ignored) +{ + if (ptf(PTF_CHECK)) + topology_schedule_update(); + set_topology_timer(); +} + +static struct timer_list topology_timer = + TIMER_DEFERRED_INITIALIZER(topology_timer_fn, 0, 0); + +static atomic_t topology_poll = ATOMIC_INIT(0); + +static void set_topology_timer(void) +{ + if (atomic_add_unless(&topology_poll, -1, 0)) + mod_timer(&topology_timer, jiffies + HZ / 10); + else + mod_timer(&topology_timer, jiffies + HZ * 60); +} + +void topology_expect_change(void) +{ + if (!MACHINE_HAS_TOPOLOGY) + return; + /* This is racy, but it doesn't matter since it is just a heuristic. + * Worst case is that we poll in a higher frequency for a bit longer. + */ + if (atomic_read(&topology_poll) > 60) + return; + atomic_add(60, &topology_poll); + set_topology_timer(); +} + +static int cpu_management; + +static ssize_t dispatching_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + ssize_t count; + + mutex_lock(&smp_cpu_state_mutex); + count = sprintf(buf, "%d\n", cpu_management); + mutex_unlock(&smp_cpu_state_mutex); + return count; +} + +static ssize_t dispatching_store(struct device *dev, + struct device_attribute *attr, + const char *buf, + size_t count) +{ + int val, rc; + char delim; + + if (sscanf(buf, "%d %c", &val, &delim) != 1) + return -EINVAL; + if (val != 0 && val != 1) + return -EINVAL; + rc = 0; + get_online_cpus(); + mutex_lock(&smp_cpu_state_mutex); + if (cpu_management == val) + goto out; + rc = topology_set_cpu_management(val); + if (rc) + goto out; + cpu_management = val; + topology_expect_change(); +out: + mutex_unlock(&smp_cpu_state_mutex); + put_online_cpus(); + return rc ? rc : count; +} +static DEVICE_ATTR(dispatching, 0644, dispatching_show, + dispatching_store); + +static ssize_t cpu_polarization_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int cpu = dev->id; + ssize_t count; + + mutex_lock(&smp_cpu_state_mutex); + switch (smp_cpu_get_polarization(cpu)) { + case POLARIZATION_HRZ: + count = sprintf(buf, "horizontal\n"); + break; + case POLARIZATION_VL: + count = sprintf(buf, "vertical:low\n"); + break; + case POLARIZATION_VM: + count = sprintf(buf, "vertical:medium\n"); + break; + case POLARIZATION_VH: + count = sprintf(buf, "vertical:high\n"); + break; + default: + count = sprintf(buf, "unknown\n"); + break; + } + mutex_unlock(&smp_cpu_state_mutex); + return count; +} +static DEVICE_ATTR(polarization, 0444, cpu_polarization_show, NULL); + +static struct attribute *topology_cpu_attrs[] = { + &dev_attr_polarization.attr, + NULL, +}; + +static struct attribute_group topology_cpu_attr_group = { + .attrs = topology_cpu_attrs, +}; + +int topology_cpu_init(struct cpu *cpu) +{ + return sysfs_create_group(&cpu->dev.kobj, &topology_cpu_attr_group); +} + +static const struct cpumask *cpu_thread_mask(int cpu) +{ + return &per_cpu(cpu_topology, cpu).thread_mask; +} + + +const struct cpumask *cpu_coregroup_mask(int cpu) +{ + return &per_cpu(cpu_topology, cpu).core_mask; +} + +static const struct cpumask *cpu_book_mask(int cpu) +{ + return &per_cpu(cpu_topology, cpu).book_mask; +} + +static int __init early_parse_topology(char *p) +{ + if (strncmp(p, "off", 3)) + return 0; + topology_enabled = 0; + return 0; +} +early_param("topology", early_parse_topology); + +static struct sched_domain_topology_level s390_topology[] = { + { cpu_thread_mask, cpu_smt_flags, SD_INIT_NAME(SMT) }, + { cpu_coregroup_mask, cpu_core_flags, SD_INIT_NAME(MC) }, + { cpu_book_mask, SD_INIT_NAME(BOOK) }, + { cpu_cpu_mask, SD_INIT_NAME(DIE) }, + { NULL, }, +}; + +static void __init alloc_masks(struct sysinfo_15_1_x *info, + struct mask_info *mask, int offset) +{ + int i, nr_masks; + + nr_masks = info->mag[TOPOLOGY_NR_MAG - offset]; + for (i = 0; i < info->mnest - offset; i++) + nr_masks *= info->mag[TOPOLOGY_NR_MAG - offset - 1 - i]; + nr_masks = max(nr_masks, 1); + for (i = 0; i < nr_masks; i++) { + mask->next = kzalloc(sizeof(*mask->next), GFP_KERNEL); + mask = mask->next; + } +} + +static int __init s390_topology_init(void) +{ + struct sysinfo_15_1_x *info; + int i; + + if (!MACHINE_HAS_TOPOLOGY) + return 0; + tl_info = (struct sysinfo_15_1_x *)__get_free_page(GFP_KERNEL); + info = tl_info; + store_topology(info); + pr_info("The CPU configuration topology of the machine is:"); + for (i = 0; i < TOPOLOGY_NR_MAG; i++) + printk(KERN_CONT " %d", info->mag[i]); + printk(KERN_CONT " / %d\n", info->mnest); + alloc_masks(info, &socket_info, 1); + alloc_masks(info, &book_info, 2); + set_sched_topology(s390_topology); + return 0; +} +early_initcall(s390_topology_init); + +static int __init topology_init(void) +{ + if (MACHINE_HAS_TOPOLOGY) + set_topology_timer(); + else + topology_update_polarization_simple(); + return device_create_file(cpu_subsys.dev_root, &dev_attr_dispatching); +} +device_initcall(topology_init); diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c new file mode 100644 index 000000000..4d96c9f53 --- /dev/null +++ b/arch/s390/kernel/traps.c @@ -0,0 +1,342 @@ +/* + * S390 version + * Copyright IBM Corp. 1999, 2000 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + * Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com), + * + * Derived from "arch/i386/kernel/traps.c" + * Copyright (C) 1991, 1992 Linus Torvalds + */ + +/* + * 'Traps.c' handles hardware traps and faults after we have saved some + * state in 'asm.s'. + */ +#include <linux/kprobes.h> +#include <linux/kdebug.h> +#include <linux/module.h> +#include <linux/ptrace.h> +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <asm/switch_to.h> +#include "entry.h" + +int show_unhandled_signals = 1; + +static inline void __user *get_trap_ip(struct pt_regs *regs) +{ + unsigned long address; + + if (regs->int_code & 0x200) + address = *(unsigned long *)(current->thread.trap_tdb + 24); + else + address = regs->psw.addr; + return (void __user *) + ((address - (regs->int_code >> 16)) & PSW_ADDR_INSN); +} + +static inline void report_user_fault(struct pt_regs *regs, int signr) +{ + if ((task_pid_nr(current) > 1) && !show_unhandled_signals) + return; + if (!unhandled_signal(current, signr)) + return; + if (!printk_ratelimit()) + return; + printk("User process fault: interruption code %04x ilc:%d ", + regs->int_code & 0xffff, regs->int_code >> 17); + print_vma_addr("in ", regs->psw.addr & PSW_ADDR_INSN); + printk("\n"); + show_regs(regs); +} + +int is_valid_bugaddr(unsigned long addr) +{ + return 1; +} + +void do_report_trap(struct pt_regs *regs, int si_signo, int si_code, char *str) +{ + siginfo_t info; + + if (user_mode(regs)) { + info.si_signo = si_signo; + info.si_errno = 0; + info.si_code = si_code; + info.si_addr = get_trap_ip(regs); + force_sig_info(si_signo, &info, current); + report_user_fault(regs, si_signo); + } else { + const struct exception_table_entry *fixup; + fixup = search_exception_tables(regs->psw.addr & PSW_ADDR_INSN); + if (fixup) + regs->psw.addr = extable_fixup(fixup) | PSW_ADDR_AMODE; + else { + enum bug_trap_type btt; + + btt = report_bug(regs->psw.addr & PSW_ADDR_INSN, regs); + if (btt == BUG_TRAP_TYPE_WARN) + return; + die(regs, str); + } + } +} + +static void do_trap(struct pt_regs *regs, int si_signo, int si_code, char *str) +{ + if (notify_die(DIE_TRAP, str, regs, 0, + regs->int_code, si_signo) == NOTIFY_STOP) + return; + do_report_trap(regs, si_signo, si_code, str); +} +NOKPROBE_SYMBOL(do_trap); + +void do_per_trap(struct pt_regs *regs) +{ + siginfo_t info; + + if (notify_die(DIE_SSTEP, "sstep", regs, 0, 0, SIGTRAP) == NOTIFY_STOP) + return; + if (!current->ptrace) + return; + info.si_signo = SIGTRAP; + info.si_errno = 0; + info.si_code = TRAP_HWBKPT; + info.si_addr = + (void __force __user *) current->thread.per_event.address; + force_sig_info(SIGTRAP, &info, current); +} +NOKPROBE_SYMBOL(do_per_trap); + +void default_trap_handler(struct pt_regs *regs) +{ + if (user_mode(regs)) { + report_user_fault(regs, SIGSEGV); + do_exit(SIGSEGV); + } else + die(regs, "Unknown program exception"); +} + +#define DO_ERROR_INFO(name, signr, sicode, str) \ +void name(struct pt_regs *regs) \ +{ \ + do_trap(regs, signr, sicode, str); \ +} + +DO_ERROR_INFO(addressing_exception, SIGILL, ILL_ILLADR, + "addressing exception") +DO_ERROR_INFO(execute_exception, SIGILL, ILL_ILLOPN, + "execute exception") +DO_ERROR_INFO(divide_exception, SIGFPE, FPE_INTDIV, + "fixpoint divide exception") +DO_ERROR_INFO(overflow_exception, SIGFPE, FPE_INTOVF, + "fixpoint overflow exception") +DO_ERROR_INFO(hfp_overflow_exception, SIGFPE, FPE_FLTOVF, + "HFP overflow exception") +DO_ERROR_INFO(hfp_underflow_exception, SIGFPE, FPE_FLTUND, + "HFP underflow exception") +DO_ERROR_INFO(hfp_significance_exception, SIGFPE, FPE_FLTRES, + "HFP significance exception") +DO_ERROR_INFO(hfp_divide_exception, SIGFPE, FPE_FLTDIV, + "HFP divide exception") +DO_ERROR_INFO(hfp_sqrt_exception, SIGFPE, FPE_FLTINV, + "HFP square root exception") +DO_ERROR_INFO(operand_exception, SIGILL, ILL_ILLOPN, + "operand exception") +DO_ERROR_INFO(privileged_op, SIGILL, ILL_PRVOPC, + "privileged operation") +DO_ERROR_INFO(special_op_exception, SIGILL, ILL_ILLOPN, + "special operation exception") +DO_ERROR_INFO(transaction_exception, SIGILL, ILL_ILLOPN, + "transaction constraint exception") + +static inline void do_fp_trap(struct pt_regs *regs, int fpc) +{ + int si_code = 0; + /* FPC[2] is Data Exception Code */ + if ((fpc & 0x00000300) == 0) { + /* bits 6 and 7 of DXC are 0 iff IEEE exception */ + if (fpc & 0x8000) /* invalid fp operation */ + si_code = FPE_FLTINV; + else if (fpc & 0x4000) /* div by 0 */ + si_code = FPE_FLTDIV; + else if (fpc & 0x2000) /* overflow */ + si_code = FPE_FLTOVF; + else if (fpc & 0x1000) /* underflow */ + si_code = FPE_FLTUND; + else if (fpc & 0x0800) /* inexact */ + si_code = FPE_FLTRES; + } + do_trap(regs, SIGFPE, si_code, "floating point exception"); +} + +void translation_exception(struct pt_regs *regs) +{ + /* May never happen. */ + panic("Translation exception"); +} + +void illegal_op(struct pt_regs *regs) +{ + siginfo_t info; + __u8 opcode[6]; + __u16 __user *location; + int is_uprobe_insn = 0; + int signal = 0; + + location = get_trap_ip(regs); + + if (user_mode(regs)) { + if (get_user(*((__u16 *) opcode), (__u16 __user *) location)) + return; + if (*((__u16 *) opcode) == S390_BREAKPOINT_U16) { + if (current->ptrace) { + info.si_signo = SIGTRAP; + info.si_errno = 0; + info.si_code = TRAP_BRKPT; + info.si_addr = location; + force_sig_info(SIGTRAP, &info, current); + } else + signal = SIGILL; +#ifdef CONFIG_UPROBES + } else if (*((__u16 *) opcode) == UPROBE_SWBP_INSN) { + is_uprobe_insn = 1; +#endif + } else + signal = SIGILL; + } + /* + * We got either an illegal op in kernel mode, or user space trapped + * on a uprobes illegal instruction. See if kprobes or uprobes picks + * it up. If not, SIGILL. + */ + if (is_uprobe_insn || !user_mode(regs)) { + if (notify_die(DIE_BPT, "bpt", regs, 0, + 3, SIGTRAP) != NOTIFY_STOP) + signal = SIGILL; + } + if (signal) + do_trap(regs, signal, ILL_ILLOPC, "illegal operation"); +} +NOKPROBE_SYMBOL(illegal_op); + +DO_ERROR_INFO(specification_exception, SIGILL, ILL_ILLOPN, + "specification exception"); + +int alloc_vector_registers(struct task_struct *tsk) +{ + __vector128 *vxrs; + int i; + + /* Allocate vector register save area. */ + vxrs = kzalloc(sizeof(__vector128) * __NUM_VXRS, + GFP_KERNEL|__GFP_REPEAT); + if (!vxrs) + return -ENOMEM; + preempt_disable(); + if (tsk == current) + save_fp_regs(tsk->thread.fp_regs.fprs); + /* Copy the 16 floating point registers */ + for (i = 0; i < 16; i++) + *(freg_t *) &vxrs[i] = tsk->thread.fp_regs.fprs[i]; + tsk->thread.vxrs = vxrs; + if (tsk == current) { + __ctl_set_bit(0, 17); + restore_vx_regs(vxrs); + } + preempt_enable(); + return 0; +} + +void vector_exception(struct pt_regs *regs) +{ + int si_code, vic; + + if (!MACHINE_HAS_VX) { + do_trap(regs, SIGILL, ILL_ILLOPN, "illegal operation"); + return; + } + + /* get vector interrupt code from fpc */ + asm volatile("stfpc %0" : "=m" (current->thread.fp_regs.fpc)); + vic = (current->thread.fp_regs.fpc & 0xf00) >> 8; + switch (vic) { + case 1: /* invalid vector operation */ + si_code = FPE_FLTINV; + break; + case 2: /* division by zero */ + si_code = FPE_FLTDIV; + break; + case 3: /* overflow */ + si_code = FPE_FLTOVF; + break; + case 4: /* underflow */ + si_code = FPE_FLTUND; + break; + case 5: /* inexact */ + si_code = FPE_FLTRES; + break; + default: /* unknown cause */ + si_code = 0; + } + do_trap(regs, SIGFPE, si_code, "vector exception"); +} + +static int __init disable_vector_extension(char *str) +{ + S390_lowcore.machine_flags &= ~MACHINE_FLAG_VX; + return 1; +} +__setup("novx", disable_vector_extension); + +void data_exception(struct pt_regs *regs) +{ + __u16 __user *location; + int signal = 0; + + location = get_trap_ip(regs); + + asm volatile("stfpc %0" : "=m" (current->thread.fp_regs.fpc)); + /* Check for vector register enablement */ + if (MACHINE_HAS_VX && !current->thread.vxrs && + (current->thread.fp_regs.fpc & FPC_DXC_MASK) == 0xfe00) { + alloc_vector_registers(current); + /* Vector data exception is suppressing, rewind psw. */ + regs->psw.addr = __rewind_psw(regs->psw, regs->int_code >> 16); + clear_pt_regs_flag(regs, PIF_PER_TRAP); + return; + } + if (current->thread.fp_regs.fpc & FPC_DXC_MASK) + signal = SIGFPE; + else + signal = SIGILL; + if (signal == SIGFPE) + do_fp_trap(regs, current->thread.fp_regs.fpc); + else if (signal) + do_trap(regs, signal, ILL_ILLOPN, "data exception"); +} + +void space_switch_exception(struct pt_regs *regs) +{ + /* Set user psw back to home space mode. */ + if (user_mode(regs)) + regs->psw.mask |= PSW_ASC_HOME; + /* Send SIGILL. */ + do_trap(regs, SIGILL, ILL_PRVOPC, "space switch event"); +} + +void kernel_stack_overflow(struct pt_regs *regs) +{ + bust_spinlocks(1); + printk("Kernel stack overflow.\n"); + show_regs(regs); + bust_spinlocks(0); + panic("Corrupt kernel stack, can't continue."); +} +NOKPROBE_SYMBOL(kernel_stack_overflow); + +void __init trap_init(void) +{ + local_mcck_enable(); +} diff --git a/arch/s390/kernel/uprobes.c b/arch/s390/kernel/uprobes.c new file mode 100644 index 000000000..66956c09d --- /dev/null +++ b/arch/s390/kernel/uprobes.c @@ -0,0 +1,385 @@ +/* + * User-space Probes (UProbes) for s390 + * + * Copyright IBM Corp. 2014 + * Author(s): Jan Willeke, + */ + +#include <linux/uaccess.h> +#include <linux/uprobes.h> +#include <linux/compat.h> +#include <linux/kdebug.h> +#include <asm/switch_to.h> +#include <asm/facility.h> +#include <asm/kprobes.h> +#include <asm/dis.h> +#include "entry.h" + +#define UPROBE_TRAP_NR UINT_MAX + +int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm, + unsigned long addr) +{ + return probe_is_prohibited_opcode(auprobe->insn); +} + +int arch_uprobe_pre_xol(struct arch_uprobe *auprobe, struct pt_regs *regs) +{ + if (psw_bits(regs->psw).eaba == PSW_AMODE_24BIT) + return -EINVAL; + if (!is_compat_task() && psw_bits(regs->psw).eaba == PSW_AMODE_31BIT) + return -EINVAL; + clear_pt_regs_flag(regs, PIF_PER_TRAP); + auprobe->saved_per = psw_bits(regs->psw).r; + auprobe->saved_int_code = regs->int_code; + regs->int_code = UPROBE_TRAP_NR; + regs->psw.addr = current->utask->xol_vaddr; + set_tsk_thread_flag(current, TIF_UPROBE_SINGLESTEP); + update_cr_regs(current); + return 0; +} + +bool arch_uprobe_xol_was_trapped(struct task_struct *tsk) +{ + struct pt_regs *regs = task_pt_regs(tsk); + + if (regs->int_code != UPROBE_TRAP_NR) + return true; + return false; +} + +static int check_per_event(unsigned short cause, unsigned long control, + struct pt_regs *regs) +{ + if (!(regs->psw.mask & PSW_MASK_PER)) + return 0; + /* user space single step */ + if (control == 0) + return 1; + /* over indication for storage alteration */ + if ((control & 0x20200000) && (cause & 0x2000)) + return 1; + if (cause & 0x8000) { + /* all branches */ + if ((control & 0x80800000) == 0x80000000) + return 1; + /* branch into selected range */ + if (((control & 0x80800000) == 0x80800000) && + regs->psw.addr >= current->thread.per_user.start && + regs->psw.addr <= current->thread.per_user.end) + return 1; + } + return 0; +} + +int arch_uprobe_post_xol(struct arch_uprobe *auprobe, struct pt_regs *regs) +{ + int fixup = probe_get_fixup_type(auprobe->insn); + struct uprobe_task *utask = current->utask; + + clear_tsk_thread_flag(current, TIF_UPROBE_SINGLESTEP); + update_cr_regs(current); + psw_bits(regs->psw).r = auprobe->saved_per; + regs->int_code = auprobe->saved_int_code; + + if (fixup & FIXUP_PSW_NORMAL) + regs->psw.addr += utask->vaddr - utask->xol_vaddr; + if (fixup & FIXUP_RETURN_REGISTER) { + int reg = (auprobe->insn[0] & 0xf0) >> 4; + + regs->gprs[reg] += utask->vaddr - utask->xol_vaddr; + } + if (fixup & FIXUP_BRANCH_NOT_TAKEN) { + int ilen = insn_length(auprobe->insn[0] >> 8); + + if (regs->psw.addr - utask->xol_vaddr == ilen) + regs->psw.addr = utask->vaddr + ilen; + } + if (check_per_event(current->thread.per_event.cause, + current->thread.per_user.control, regs)) { + /* fix per address */ + current->thread.per_event.address = utask->vaddr; + /* trigger per event */ + set_pt_regs_flag(regs, PIF_PER_TRAP); + } + return 0; +} + +int arch_uprobe_exception_notify(struct notifier_block *self, unsigned long val, + void *data) +{ + struct die_args *args = data; + struct pt_regs *regs = args->regs; + + if (!user_mode(regs)) + return NOTIFY_DONE; + if (regs->int_code & 0x200) /* Trap during transaction */ + return NOTIFY_DONE; + switch (val) { + case DIE_BPT: + if (uprobe_pre_sstep_notifier(regs)) + return NOTIFY_STOP; + break; + case DIE_SSTEP: + if (uprobe_post_sstep_notifier(regs)) + return NOTIFY_STOP; + default: + break; + } + return NOTIFY_DONE; +} + +void arch_uprobe_abort_xol(struct arch_uprobe *auprobe, struct pt_regs *regs) +{ + clear_thread_flag(TIF_UPROBE_SINGLESTEP); + regs->int_code = auprobe->saved_int_code; + regs->psw.addr = current->utask->vaddr; + current->thread.per_event.address = current->utask->vaddr; +} + +unsigned long arch_uretprobe_hijack_return_addr(unsigned long trampoline, + struct pt_regs *regs) +{ + unsigned long orig; + + orig = regs->gprs[14]; + regs->gprs[14] = trampoline; + return orig; +} + +/* Instruction Emulation */ + +static void adjust_psw_addr(psw_t *psw, unsigned long len) +{ + psw->addr = __rewind_psw(*psw, -len); +} + +#define EMU_ILLEGAL_OP 1 +#define EMU_SPECIFICATION 2 +#define EMU_ADDRESSING 3 + +#define emu_load_ril(ptr, output) \ +({ \ + unsigned int mask = sizeof(*(ptr)) - 1; \ + __typeof__(*(ptr)) input; \ + int __rc = 0; \ + \ + if (!test_facility(34)) \ + __rc = EMU_ILLEGAL_OP; \ + else if ((u64 __force)ptr & mask) \ + __rc = EMU_SPECIFICATION; \ + else if (get_user(input, ptr)) \ + __rc = EMU_ADDRESSING; \ + else \ + *(output) = input; \ + __rc; \ +}) + +#define emu_store_ril(regs, ptr, input) \ +({ \ + unsigned int mask = sizeof(*(ptr)) - 1; \ + __typeof__(ptr) __ptr = (ptr); \ + int __rc = 0; \ + \ + if (!test_facility(34)) \ + __rc = EMU_ILLEGAL_OP; \ + else if ((u64 __force)__ptr & mask) \ + __rc = EMU_SPECIFICATION; \ + else if (put_user(*(input), __ptr)) \ + __rc = EMU_ADDRESSING; \ + if (__rc == 0) \ + sim_stor_event(regs, \ + (void __force *)__ptr, \ + mask + 1); \ + __rc; \ +}) + +#define emu_cmp_ril(regs, ptr, cmp) \ +({ \ + unsigned int mask = sizeof(*(ptr)) - 1; \ + __typeof__(*(ptr)) input; \ + int __rc = 0; \ + \ + if (!test_facility(34)) \ + __rc = EMU_ILLEGAL_OP; \ + else if ((u64 __force)ptr & mask) \ + __rc = EMU_SPECIFICATION; \ + else if (get_user(input, ptr)) \ + __rc = EMU_ADDRESSING; \ + else if (input > *(cmp)) \ + psw_bits((regs)->psw).cc = 1; \ + else if (input < *(cmp)) \ + psw_bits((regs)->psw).cc = 2; \ + else \ + psw_bits((regs)->psw).cc = 0; \ + __rc; \ +}) + +struct insn_ril { + u8 opc0; + u8 reg : 4; + u8 opc1 : 4; + s32 disp; +} __packed; + +union split_register { + u64 u64; + u32 u32[2]; + u16 u16[4]; + s64 s64; + s32 s32[2]; + s16 s16[4]; +}; + +/* + * If user per registers are setup to trace storage alterations and an + * emulated store took place on a fitting address a user trap is generated. + */ +static void sim_stor_event(struct pt_regs *regs, void *addr, int len) +{ + if (!(regs->psw.mask & PSW_MASK_PER)) + return; + if (!(current->thread.per_user.control & PER_EVENT_STORE)) + return; + if ((void *)current->thread.per_user.start > (addr + len)) + return; + if ((void *)current->thread.per_user.end < addr) + return; + current->thread.per_event.address = regs->psw.addr; + current->thread.per_event.cause = PER_EVENT_STORE >> 16; + set_pt_regs_flag(regs, PIF_PER_TRAP); +} + +/* + * pc relative instructions are emulated, since parameters may not be + * accessible from the xol area due to range limitations. + */ +static void handle_insn_ril(struct arch_uprobe *auprobe, struct pt_regs *regs) +{ + union split_register *rx; + struct insn_ril *insn; + unsigned int ilen; + void *uptr; + int rc = 0; + + insn = (struct insn_ril *) &auprobe->insn; + rx = (union split_register *) ®s->gprs[insn->reg]; + uptr = (void *)(regs->psw.addr + (insn->disp * 2)); + ilen = insn_length(insn->opc0); + + switch (insn->opc0) { + case 0xc0: + switch (insn->opc1) { + case 0x00: /* larl */ + rx->u64 = (unsigned long)uptr; + break; + } + break; + case 0xc4: + switch (insn->opc1) { + case 0x02: /* llhrl */ + rc = emu_load_ril((u16 __user *)uptr, &rx->u32[1]); + break; + case 0x04: /* lghrl */ + rc = emu_load_ril((s16 __user *)uptr, &rx->u64); + break; + case 0x05: /* lhrl */ + rc = emu_load_ril((s16 __user *)uptr, &rx->u32[1]); + break; + case 0x06: /* llghrl */ + rc = emu_load_ril((u16 __user *)uptr, &rx->u64); + break; + case 0x08: /* lgrl */ + rc = emu_load_ril((u64 __user *)uptr, &rx->u64); + break; + case 0x0c: /* lgfrl */ + rc = emu_load_ril((s32 __user *)uptr, &rx->u64); + break; + case 0x0d: /* lrl */ + rc = emu_load_ril((u32 __user *)uptr, &rx->u32[1]); + break; + case 0x0e: /* llgfrl */ + rc = emu_load_ril((u32 __user *)uptr, &rx->u64); + break; + case 0x07: /* sthrl */ + rc = emu_store_ril(regs, (u16 __user *)uptr, &rx->u16[3]); + break; + case 0x0b: /* stgrl */ + rc = emu_store_ril(regs, (u64 __user *)uptr, &rx->u64); + break; + case 0x0f: /* strl */ + rc = emu_store_ril(regs, (u32 __user *)uptr, &rx->u32[1]); + break; + } + break; + case 0xc6: + switch (insn->opc1) { + case 0x02: /* pfdrl */ + if (!test_facility(34)) + rc = EMU_ILLEGAL_OP; + break; + case 0x04: /* cghrl */ + rc = emu_cmp_ril(regs, (s16 __user *)uptr, &rx->s64); + break; + case 0x05: /* chrl */ + rc = emu_cmp_ril(regs, (s16 __user *)uptr, &rx->s32[1]); + break; + case 0x06: /* clghrl */ + rc = emu_cmp_ril(regs, (u16 __user *)uptr, &rx->u64); + break; + case 0x07: /* clhrl */ + rc = emu_cmp_ril(regs, (u16 __user *)uptr, &rx->u32[1]); + break; + case 0x08: /* cgrl */ + rc = emu_cmp_ril(regs, (s64 __user *)uptr, &rx->s64); + break; + case 0x0a: /* clgrl */ + rc = emu_cmp_ril(regs, (u64 __user *)uptr, &rx->u64); + break; + case 0x0c: /* cgfrl */ + rc = emu_cmp_ril(regs, (s32 __user *)uptr, &rx->s64); + break; + case 0x0d: /* crl */ + rc = emu_cmp_ril(regs, (s32 __user *)uptr, &rx->s32[1]); + break; + case 0x0e: /* clgfrl */ + rc = emu_cmp_ril(regs, (u32 __user *)uptr, &rx->u64); + break; + case 0x0f: /* clrl */ + rc = emu_cmp_ril(regs, (u32 __user *)uptr, &rx->u32[1]); + break; + } + break; + } + adjust_psw_addr(®s->psw, ilen); + switch (rc) { + case EMU_ILLEGAL_OP: + regs->int_code = ilen << 16 | 0x0001; + do_report_trap(regs, SIGILL, ILL_ILLOPC, NULL); + break; + case EMU_SPECIFICATION: + regs->int_code = ilen << 16 | 0x0006; + do_report_trap(regs, SIGILL, ILL_ILLOPC , NULL); + break; + case EMU_ADDRESSING: + regs->int_code = ilen << 16 | 0x0005; + do_report_trap(regs, SIGSEGV, SEGV_MAPERR, NULL); + break; + } +} + +bool arch_uprobe_skip_sstep(struct arch_uprobe *auprobe, struct pt_regs *regs) +{ + if ((psw_bits(regs->psw).eaba == PSW_AMODE_24BIT) || + ((psw_bits(regs->psw).eaba == PSW_AMODE_31BIT) && + !is_compat_task())) { + regs->psw.addr = __rewind_psw(regs->psw, UPROBE_SWBP_INSN_SIZE); + do_report_trap(regs, SIGILL, ILL_ILLADR, NULL); + return true; + } + if (probe_is_insn_relative_long(auprobe->insn)) { + handle_insn_ril(auprobe, regs); + return true; + } + return false; +} diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c new file mode 100644 index 000000000..0d58269ff --- /dev/null +++ b/arch/s390/kernel/vdso.c @@ -0,0 +1,306 @@ +/* + * vdso setup for s390 + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.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 <linux/module.h> +#include <linux/errno.h> +#include <linux/sched.h> +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/smp.h> +#include <linux/stddef.h> +#include <linux/unistd.h> +#include <linux/slab.h> +#include <linux/user.h> +#include <linux/elf.h> +#include <linux/security.h> +#include <linux/bootmem.h> +#include <linux/compat.h> +#include <asm/asm-offsets.h> +#include <asm/pgtable.h> +#include <asm/processor.h> +#include <asm/mmu.h> +#include <asm/mmu_context.h> +#include <asm/sections.h> +#include <asm/vdso.h> +#include <asm/facility.h> + +#ifdef CONFIG_COMPAT +extern char vdso32_start, vdso32_end; +static void *vdso32_kbase = &vdso32_start; +static unsigned int vdso32_pages; +static struct page **vdso32_pagelist; +#endif + +extern char vdso64_start, vdso64_end; +static void *vdso64_kbase = &vdso64_start; +static unsigned int vdso64_pages; +static struct page **vdso64_pagelist; + +/* + * Should the kernel map a VDSO page into processes and pass its + * address down to glibc upon exec()? + */ +unsigned int __read_mostly vdso_enabled = 1; + +static int __init vdso_setup(char *s) +{ + unsigned long val; + int rc; + + rc = 0; + if (strncmp(s, "on", 3) == 0) + vdso_enabled = 1; + else if (strncmp(s, "off", 4) == 0) + vdso_enabled = 0; + else { + rc = kstrtoul(s, 0, &val); + vdso_enabled = rc ? 0 : !!val; + } + return !rc; +} +__setup("vdso=", vdso_setup); + +/* + * The vdso data page + */ +static union { + struct vdso_data data; + u8 page[PAGE_SIZE]; +} vdso_data_store __page_aligned_data; +struct vdso_data *vdso_data = &vdso_data_store.data; + +/* + * Setup vdso data page. + */ +static void vdso_init_data(struct vdso_data *vd) +{ + vd->ectg_available = test_facility(31); +} + +/* + * Allocate/free per cpu vdso data. + */ +#define SEGMENT_ORDER 2 + +int vdso_alloc_per_cpu(struct _lowcore *lowcore) +{ + unsigned long segment_table, page_table, page_frame; + u32 *psal, *aste; + int i; + + lowcore->vdso_per_cpu_data = __LC_PASTE; + + if (!vdso_enabled) + return 0; + + segment_table = __get_free_pages(GFP_KERNEL, SEGMENT_ORDER); + page_table = get_zeroed_page(GFP_KERNEL | GFP_DMA); + page_frame = get_zeroed_page(GFP_KERNEL); + if (!segment_table || !page_table || !page_frame) + goto out; + + clear_table((unsigned long *) segment_table, _SEGMENT_ENTRY_EMPTY, + PAGE_SIZE << SEGMENT_ORDER); + clear_table((unsigned long *) page_table, _PAGE_INVALID, + 256*sizeof(unsigned long)); + + *(unsigned long *) segment_table = _SEGMENT_ENTRY + page_table; + *(unsigned long *) page_table = _PAGE_PROTECT + page_frame; + + psal = (u32 *) (page_table + 256*sizeof(unsigned long)); + aste = psal + 32; + + for (i = 4; i < 32; i += 4) + psal[i] = 0x80000000; + + lowcore->paste[4] = (u32)(addr_t) psal; + psal[0] = 0x02000000; + psal[2] = (u32)(addr_t) aste; + *(unsigned long *) (aste + 2) = segment_table + + _ASCE_TABLE_LENGTH + _ASCE_USER_BITS + _ASCE_TYPE_SEGMENT; + aste[4] = (u32)(addr_t) psal; + lowcore->vdso_per_cpu_data = page_frame; + + return 0; + +out: + free_page(page_frame); + free_page(page_table); + free_pages(segment_table, SEGMENT_ORDER); + return -ENOMEM; +} + +void vdso_free_per_cpu(struct _lowcore *lowcore) +{ + unsigned long segment_table, page_table, page_frame; + u32 *psal, *aste; + + if (!vdso_enabled) + return; + + psal = (u32 *)(addr_t) lowcore->paste[4]; + aste = (u32 *)(addr_t) psal[2]; + segment_table = *(unsigned long *)(aste + 2) & PAGE_MASK; + page_table = *(unsigned long *) segment_table; + page_frame = *(unsigned long *) page_table; + + free_page(page_frame); + free_page(page_table); + free_pages(segment_table, SEGMENT_ORDER); +} + +static void vdso_init_cr5(void) +{ + unsigned long cr5; + + if (!vdso_enabled) + return; + cr5 = offsetof(struct _lowcore, paste); + __ctl_load(cr5, 5, 5); +} + +/* + * This is called from binfmt_elf, we create the special vma for the + * vDSO and insert it into the mm struct tree + */ +int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) +{ + struct mm_struct *mm = current->mm; + struct page **vdso_pagelist; + unsigned long vdso_pages; + unsigned long vdso_base; + int rc; + + if (!vdso_enabled) + return 0; + /* + * Only map the vdso for dynamically linked elf binaries. + */ + if (!uses_interp) + return 0; + + vdso_pagelist = vdso64_pagelist; + vdso_pages = vdso64_pages; +#ifdef CONFIG_COMPAT + if (is_compat_task()) { + vdso_pagelist = vdso32_pagelist; + vdso_pages = vdso32_pages; + } +#endif + /* + * vDSO has a problem and was disabled, just don't "enable" it for + * the process + */ + if (vdso_pages == 0) + return 0; + + current->mm->context.vdso_base = 0; + + /* + * pick a base address for the vDSO in process space. We try to put + * it at vdso_base which is the "natural" base for it, but we might + * fail and end up putting it elsewhere. + */ + down_write(&mm->mmap_sem); + vdso_base = get_unmapped_area(NULL, 0, vdso_pages << PAGE_SHIFT, 0, 0); + if (IS_ERR_VALUE(vdso_base)) { + rc = vdso_base; + goto out_up; + } + + /* + * Put vDSO base into mm struct. We need to do this before calling + * install_special_mapping or the perf counter mmap tracking code + * will fail to recognise it as a vDSO (since arch_vma_name fails). + */ + current->mm->context.vdso_base = vdso_base; + + /* + * our vma flags don't have VM_WRITE so by default, the process + * isn't allowed to write those pages. + * gdb can break that with ptrace interface, and thus trigger COW + * on those pages but it's then your responsibility to never do that + * on the "data" page of the vDSO or you'll stop getting kernel + * updates and your nice userland gettimeofday will be totally dead. + * It's fine to use that for setting breakpoints in the vDSO code + * pages though. + */ + rc = install_special_mapping(mm, vdso_base, vdso_pages << PAGE_SHIFT, + VM_READ|VM_EXEC| + VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC, + vdso_pagelist); + if (rc) + current->mm->context.vdso_base = 0; +out_up: + up_write(&mm->mmap_sem); + return rc; +} + +const char *arch_vma_name(struct vm_area_struct *vma) +{ + if (vma->vm_mm && vma->vm_start == vma->vm_mm->context.vdso_base) + return "[vdso]"; + return NULL; +} + +static int __init vdso_init(void) +{ + int i; + + if (!vdso_enabled) + return 0; + vdso_init_data(vdso_data); +#ifdef CONFIG_COMPAT + /* Calculate the size of the 32 bit vDSO */ + vdso32_pages = ((&vdso32_end - &vdso32_start + + PAGE_SIZE - 1) >> PAGE_SHIFT) + 1; + + /* Make sure pages are in the correct state */ + vdso32_pagelist = kzalloc(sizeof(struct page *) * (vdso32_pages + 1), + GFP_KERNEL); + BUG_ON(vdso32_pagelist == NULL); + for (i = 0; i < vdso32_pages - 1; i++) { + struct page *pg = virt_to_page(vdso32_kbase + i*PAGE_SIZE); + ClearPageReserved(pg); + get_page(pg); + vdso32_pagelist[i] = pg; + } + vdso32_pagelist[vdso32_pages - 1] = virt_to_page(vdso_data); + vdso32_pagelist[vdso32_pages] = NULL; +#endif + + /* Calculate the size of the 64 bit vDSO */ + vdso64_pages = ((&vdso64_end - &vdso64_start + + PAGE_SIZE - 1) >> PAGE_SHIFT) + 1; + + /* Make sure pages are in the correct state */ + vdso64_pagelist = kzalloc(sizeof(struct page *) * (vdso64_pages + 1), + GFP_KERNEL); + BUG_ON(vdso64_pagelist == NULL); + for (i = 0; i < vdso64_pages - 1; i++) { + struct page *pg = virt_to_page(vdso64_kbase + i*PAGE_SIZE); + ClearPageReserved(pg); + get_page(pg); + vdso64_pagelist[i] = pg; + } + vdso64_pagelist[vdso64_pages - 1] = virt_to_page(vdso_data); + vdso64_pagelist[vdso64_pages] = NULL; + if (vdso_alloc_per_cpu(&S390_lowcore)) + BUG(); + vdso_init_cr5(); + + get_page(virt_to_page(vdso_data)); + + smp_wmb(); + + return 0; +} +early_initcall(vdso_init); diff --git a/arch/s390/kernel/vdso32/.gitignore b/arch/s390/kernel/vdso32/.gitignore new file mode 100644 index 000000000..e45fba9d0 --- /dev/null +++ b/arch/s390/kernel/vdso32/.gitignore @@ -0,0 +1 @@ +vdso32.lds diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile new file mode 100644 index 000000000..8ad2b34ad --- /dev/null +++ b/arch/s390/kernel/vdso32/Makefile @@ -0,0 +1,58 @@ +# List of files in the vdso, has to be asm only for now + +obj-vdso32 = gettimeofday.o clock_getres.o clock_gettime.o note.o + +# Build rules + +targets := $(obj-vdso32) vdso32.so vdso32.so.dbg +obj-vdso32 := $(addprefix $(obj)/, $(obj-vdso32)) + +KBUILD_AFLAGS_31 := $(filter-out -m64,$(KBUILD_AFLAGS)) +KBUILD_AFLAGS_31 += -m31 -s + +KBUILD_CFLAGS_31 := $(filter-out -m64,$(KBUILD_CFLAGS)) +KBUILD_CFLAGS_31 += -m31 -fPIC -shared -fno-common -fno-builtin +KBUILD_CFLAGS_31 += -nostdlib -Wl,-soname=linux-vdso32.so.1 \ + $(call cc-ldoption, -Wl$(comma)--hash-style=sysv) + +$(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_31) +$(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_31) + +obj-y += vdso32_wrapper.o +extra-y += vdso32.lds +CPPFLAGS_vdso32.lds += -P -C -U$(ARCH) + +# Disable gcov profiling for VDSO code +GCOV_PROFILE := n + +# Force dependency (incbin is bad) +$(obj)/vdso32_wrapper.o : $(obj)/vdso32.so + +# link rule for the .so file, .lds has to be first +$(obj)/vdso32.so.dbg: $(src)/vdso32.lds $(obj-vdso32) + $(call if_changed,vdso32ld) + +# strip rule for the .so file +$(obj)/%.so: OBJCOPYFLAGS := -S +$(obj)/%.so: $(obj)/%.so.dbg FORCE + $(call if_changed,objcopy) + +# assembly rules for the .S files +$(obj-vdso32): %.o: %.S + $(call if_changed_dep,vdso32as) + +# actual build commands +quiet_cmd_vdso32ld = VDSO32L $@ + cmd_vdso32ld = $(CC) $(c_flags) -Wl,-T $^ -o $@ +quiet_cmd_vdso32as = VDSO32A $@ + cmd_vdso32as = $(CC) $(a_flags) -c -o $@ $< + +# install commands for the unstripped file +quiet_cmd_vdso_install = INSTALL $@ + cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/$@ + +vdso32.so: $(obj)/vdso32.so.dbg + @mkdir -p $(MODLIB)/vdso + $(call cmd,vdso_install) + +vdso_install: vdso32.so diff --git a/arch/s390/kernel/vdso32/clock_getres.S b/arch/s390/kernel/vdso32/clock_getres.S new file mode 100644 index 000000000..eca3f001f --- /dev/null +++ b/arch/s390/kernel/vdso32/clock_getres.S @@ -0,0 +1,46 @@ +/* + * Userland implementation of clock_getres() for 32 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.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 <asm/vdso.h> +#include <asm/asm-offsets.h> +#include <asm/unistd.h> + + .text + .align 4 + .globl __kernel_clock_getres + .type __kernel_clock_getres,@function +__kernel_clock_getres: + .cfi_startproc + basr %r1,0 + la %r1,4f-.(%r1) + chi %r2,__CLOCK_REALTIME + je 0f + chi %r2,__CLOCK_MONOTONIC + je 0f + la %r1,5f-4f(%r1) + chi %r2,__CLOCK_REALTIME_COARSE + je 0f + chi %r2,__CLOCK_MONOTONIC_COARSE + jne 3f +0: ltr %r3,%r3 + jz 2f /* res == NULL */ +1: l %r0,0(%r1) + xc 0(4,%r3),0(%r3) /* set tp->tv_sec to zero */ + st %r0,4(%r3) /* store tp->tv_usec */ +2: lhi %r2,0 + br %r14 +3: lhi %r1,__NR_clock_getres /* fallback to svc */ + svc 0 + br %r14 +4: .long __CLOCK_REALTIME_RES +5: .long __CLOCK_COARSE_RES + .cfi_endproc + .size __kernel_clock_getres,.-__kernel_clock_getres diff --git a/arch/s390/kernel/vdso32/clock_gettime.S b/arch/s390/kernel/vdso32/clock_gettime.S new file mode 100644 index 000000000..5eec9afbb --- /dev/null +++ b/arch/s390/kernel/vdso32/clock_gettime.S @@ -0,0 +1,149 @@ +/* + * Userland implementation of clock_gettime() for 32 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.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 <asm/vdso.h> +#include <asm/asm-offsets.h> +#include <asm/unistd.h> + + .text + .align 4 + .globl __kernel_clock_gettime + .type __kernel_clock_gettime,@function +__kernel_clock_gettime: + .cfi_startproc + ahi %r15,-16 + basr %r5,0 +0: al %r5,21f-0b(%r5) /* get &_vdso_data */ + chi %r2,__CLOCK_REALTIME_COARSE + je 10f + chi %r2,__CLOCK_REALTIME + je 11f + chi %r2,__CLOCK_MONOTONIC_COARSE + je 9f + chi %r2,__CLOCK_MONOTONIC + jne 19f + + /* CLOCK_MONOTONIC */ +1: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */ + tml %r4,0x0001 /* pending update ? loop */ + jnz 1b + stcke 0(%r15) /* Store TOD clock */ + lm %r0,%r1,1(%r15) + s %r0,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + sl %r1,__VDSO_XTIME_STAMP+4(%r5) + brc 3,2f + ahi %r0,-1 +2: ms %r0,__VDSO_TK_MULT(%r5) /* * tk->mult */ + lr %r2,%r0 + l %r0,__VDSO_TK_MULT(%r5) + ltr %r1,%r1 + mr %r0,%r0 + jnm 3f + a %r0,__VDSO_TK_MULT(%r5) +3: alr %r0,%r2 + al %r0,__VDSO_WTOM_NSEC(%r5) + al %r1,__VDSO_WTOM_NSEC+4(%r5) + brc 12,5f + ahi %r0,1 +5: l %r2,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */ + srdl %r0,0(%r2) /* >> tk->shift */ + l %r2,__VDSO_WTOM_SEC+4(%r5) + cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */ + jne 1b + basr %r5,0 +6: ltr %r0,%r0 + jnz 7f + cl %r1,20f-6b(%r5) + jl 8f +7: ahi %r2,1 + sl %r1,20f-6b(%r5) + brc 3,6b + ahi %r0,-1 + j 6b +8: st %r2,0(%r3) /* store tp->tv_sec */ + st %r1,4(%r3) /* store tp->tv_nsec */ + lhi %r2,0 + ahi %r15,16 + br %r14 + + /* CLOCK_MONOTONIC_COARSE */ +9: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */ + tml %r4,0x0001 /* pending update ? loop */ + jnz 9b + l %r2,__VDSO_WTOM_CRS_SEC+4(%r5) + l %r1,__VDSO_WTOM_CRS_NSEC+4(%r5) + cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */ + jne 9b + j 8b + + /* CLOCK_REALTIME_COARSE */ +10: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */ + tml %r4,0x0001 /* pending update ? loop */ + jnz 10b + l %r2,__VDSO_XTIME_CRS_SEC+4(%r5) + l %r1,__VDSO_XTIME_CRS_NSEC+4(%r5) + cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */ + jne 10b + j 17f + + /* CLOCK_REALTIME */ +11: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */ + tml %r4,0x0001 /* pending update ? loop */ + jnz 11b + stcke 0(%r15) /* Store TOD clock */ + lm %r0,%r1,1(%r15) + s %r0,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + sl %r1,__VDSO_XTIME_STAMP+4(%r5) + brc 3,12f + ahi %r0,-1 +12: ms %r0,__VDSO_TK_MULT(%r5) /* * tk->mult */ + lr %r2,%r0 + l %r0,__VDSO_TK_MULT(%r5) + ltr %r1,%r1 + mr %r0,%r0 + jnm 13f + a %r0,__VDSO_TK_MULT(%r5) +13: alr %r0,%r2 + al %r0,__VDSO_XTIME_NSEC(%r5) /* + tk->xtime_nsec */ + al %r1,__VDSO_XTIME_NSEC+4(%r5) + brc 12,14f + ahi %r0,1 +14: l %r2,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */ + srdl %r0,0(%r2) /* >> tk->shift */ + l %r2,__VDSO_XTIME_SEC+4(%r5) + cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */ + jne 11b + basr %r5,0 +15: ltr %r0,%r0 + jnz 16f + cl %r1,20f-15b(%r5) + jl 17f +16: ahi %r2,1 + sl %r1,20f-15b(%r5) + brc 3,15b + ahi %r0,-1 + j 15b +17: st %r2,0(%r3) /* store tp->tv_sec */ + st %r1,4(%r3) /* store tp->tv_nsec */ + lhi %r2,0 + ahi %r15,16 + br %r14 + + /* Fallback to system call */ +19: lhi %r1,__NR_clock_gettime + svc 0 + ahi %r15,16 + br %r14 + +20: .long 1000000000 +21: .long _vdso_data - 0b + .cfi_endproc + .size __kernel_clock_gettime,.-__kernel_clock_gettime diff --git a/arch/s390/kernel/vdso32/gettimeofday.S b/arch/s390/kernel/vdso32/gettimeofday.S new file mode 100644 index 000000000..719de6186 --- /dev/null +++ b/arch/s390/kernel/vdso32/gettimeofday.S @@ -0,0 +1,81 @@ +/* + * Userland implementation of gettimeofday() for 32 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.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 <asm/vdso.h> +#include <asm/asm-offsets.h> +#include <asm/unistd.h> + + .text + .align 4 + .globl __kernel_gettimeofday + .type __kernel_gettimeofday,@function +__kernel_gettimeofday: + .cfi_startproc + ahi %r15,-16 + basr %r5,0 +0: al %r5,13f-0b(%r5) /* get &_vdso_data */ +1: ltr %r3,%r3 /* check if tz is NULL */ + je 2f + mvc 0(8,%r3),__VDSO_TIMEZONE(%r5) +2: ltr %r2,%r2 /* check if tv is NULL */ + je 10f + l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */ + tml %r4,0x0001 /* pending update ? loop */ + jnz 1b + stcke 0(%r15) /* Store TOD clock */ + lm %r0,%r1,1(%r15) + s %r0,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + sl %r1,__VDSO_XTIME_STAMP+4(%r5) + brc 3,3f + ahi %r0,-1 +3: ms %r0,__VDSO_TK_MULT(%r5) /* * tk->mult */ + st %r0,0(%r15) + l %r0,__VDSO_TK_MULT(%r5) + ltr %r1,%r1 + mr %r0,%r0 + jnm 4f + a %r0,__VDSO_TK_MULT(%r5) +4: al %r0,0(%r15) + al %r0,__VDSO_XTIME_NSEC(%r5) /* + xtime */ + al %r1,__VDSO_XTIME_NSEC+4(%r5) + brc 12,5f + ahi %r0,1 +5: mvc 0(4,%r15),__VDSO_XTIME_SEC+4(%r5) + cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */ + jne 1b + l %r4,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */ + srdl %r0,0(%r4) /* >> tk->shift */ + l %r4,0(%r15) /* get tv_sec from stack */ + basr %r5,0 +6: ltr %r0,%r0 + jnz 7f + cl %r1,11f-6b(%r5) + jl 8f +7: ahi %r4,1 + sl %r1,11f-6b(%r5) + brc 3,6b + ahi %r0,-1 + j 6b +8: st %r4,0(%r2) /* store tv->tv_sec */ + ltr %r1,%r1 + m %r0,12f-6b(%r5) + jnm 9f + al %r0,12f-6b(%r5) +9: srl %r0,6 + st %r0,4(%r2) /* store tv->tv_usec */ +10: slr %r2,%r2 + ahi %r15,16 + br %r14 +11: .long 1000000000 +12: .long 274877907 +13: .long _vdso_data - 0b + .cfi_endproc + .size __kernel_gettimeofday,.-__kernel_gettimeofday diff --git a/arch/s390/kernel/vdso32/note.S b/arch/s390/kernel/vdso32/note.S new file mode 100644 index 000000000..79a071e43 --- /dev/null +++ b/arch/s390/kernel/vdso32/note.S @@ -0,0 +1,12 @@ +/* + * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text. + * Here we can supply some information useful to userland. + */ + +#include <linux/uts.h> +#include <linux/version.h> +#include <linux/elfnote.h> + +ELFNOTE_START(Linux, 0, "a") + .long LINUX_VERSION_CODE +ELFNOTE_END diff --git a/arch/s390/kernel/vdso32/vdso32.lds.S b/arch/s390/kernel/vdso32/vdso32.lds.S new file mode 100644 index 000000000..a8c379fa1 --- /dev/null +++ b/arch/s390/kernel/vdso32/vdso32.lds.S @@ -0,0 +1,138 @@ +/* + * This is the infamous ld script for the 32 bits vdso + * library + */ +#include <asm/vdso.h> + +OUTPUT_FORMAT("elf32-s390", "elf32-s390", "elf32-s390") +OUTPUT_ARCH(s390:31-bit) +ENTRY(_start) + +SECTIONS +{ + . = VDSO32_LBASE + SIZEOF_HEADERS; + + .hash : { *(.hash) } :text + .gnu.hash : { *(.gnu.hash) } + .dynsym : { *(.dynsym) } + .dynstr : { *(.dynstr) } + .gnu.version : { *(.gnu.version) } + .gnu.version_d : { *(.gnu.version_d) } + .gnu.version_r : { *(.gnu.version_r) } + + .note : { *(.note.*) } :text :note + + . = ALIGN(16); + .text : { + *(.text .stub .text.* .gnu.linkonce.t.*) + } :text + PROVIDE(__etext = .); + PROVIDE(_etext = .); + PROVIDE(etext = .); + + /* + * Other stuff is appended to the text segment: + */ + .rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) } + .rodata1 : { *(.rodata1) } + + .dynamic : { *(.dynamic) } :text :dynamic + + .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr + .eh_frame : { KEEP (*(.eh_frame)) } :text + .gcc_except_table : { *(.gcc_except_table .gcc_except_table.*) } + + .rela.dyn ALIGN(8) : { *(.rela.dyn) } + .got ALIGN(8) : { *(.got .toc) } + + _end = .; + PROVIDE(end = .); + + /* + * Stabs debugging sections are here too. + */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } + + /* + * DWARF debug sections. + * Symbols in the DWARF debugging sections are relative to the + * beginning of the section so we begin them at 0. + */ + /* DWARF 1 */ + .debug 0 : { *(.debug) } + .line 0 : { *(.line) } + /* GNU DWARF 1 extensions */ + .debug_srcinfo 0 : { *(.debug_srcinfo) } + .debug_sfnames 0 : { *(.debug_sfnames) } + /* DWARF 1.1 and DWARF 2 */ + .debug_aranges 0 : { *(.debug_aranges) } + .debug_pubnames 0 : { *(.debug_pubnames) } + /* DWARF 2 */ + .debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) } + .debug_abbrev 0 : { *(.debug_abbrev) } + .debug_line 0 : { *(.debug_line) } + .debug_frame 0 : { *(.debug_frame) } + .debug_str 0 : { *(.debug_str) } + .debug_loc 0 : { *(.debug_loc) } + .debug_macinfo 0 : { *(.debug_macinfo) } + /* SGI/MIPS DWARF 2 extensions */ + .debug_weaknames 0 : { *(.debug_weaknames) } + .debug_funcnames 0 : { *(.debug_funcnames) } + .debug_typenames 0 : { *(.debug_typenames) } + .debug_varnames 0 : { *(.debug_varnames) } + /* DWARF 3 */ + .debug_pubtypes 0 : { *(.debug_pubtypes) } + .debug_ranges 0 : { *(.debug_ranges) } + .gnu.attributes 0 : { KEEP (*(.gnu.attributes)) } + + . = ALIGN(4096); + PROVIDE(_vdso_data = .); + + /DISCARD/ : { + *(.note.GNU-stack) + *(.branch_lt) + *(.data .data.* .gnu.linkonce.d.* .sdata*) + *(.bss .sbss .dynbss .dynsbss) + } +} + +/* + * Very old versions of ld do not recognize this name token; use the constant. + */ +#define PT_GNU_EH_FRAME 0x6474e550 + +/* + * We must supply the ELF program headers explicitly to get just one + * PT_LOAD segment, and set the flags explicitly to make segments read-only. + */ +PHDRS +{ + text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ + note PT_NOTE FLAGS(4); /* PF_R */ + eh_frame_hdr PT_GNU_EH_FRAME; +} + +/* + * This controls what symbols we export from the DSO. + */ +VERSION +{ + VDSO_VERSION_STRING { + global: + /* + * Has to be there for the kernel to find + */ + __kernel_gettimeofday; + __kernel_clock_gettime; + __kernel_clock_getres; + + local: *; + }; +} diff --git a/arch/s390/kernel/vdso32/vdso32_wrapper.S b/arch/s390/kernel/vdso32/vdso32_wrapper.S new file mode 100644 index 000000000..ae42f8ce3 --- /dev/null +++ b/arch/s390/kernel/vdso32/vdso32_wrapper.S @@ -0,0 +1,14 @@ +#include <linux/init.h> +#include <linux/linkage.h> +#include <asm/page.h> + + __PAGE_ALIGNED_DATA + + .globl vdso32_start, vdso32_end + .balign PAGE_SIZE +vdso32_start: + .incbin "arch/s390/kernel/vdso32/vdso32.so" + .balign PAGE_SIZE +vdso32_end: + + .previous diff --git a/arch/s390/kernel/vdso64/.gitignore b/arch/s390/kernel/vdso64/.gitignore new file mode 100644 index 000000000..3fd18cf9f --- /dev/null +++ b/arch/s390/kernel/vdso64/.gitignore @@ -0,0 +1 @@ +vdso64.lds diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile new file mode 100644 index 000000000..2a8ddfd12 --- /dev/null +++ b/arch/s390/kernel/vdso64/Makefile @@ -0,0 +1,58 @@ +# List of files in the vdso, has to be asm only for now + +obj-vdso64 = gettimeofday.o clock_getres.o clock_gettime.o note.o + +# Build rules + +targets := $(obj-vdso64) vdso64.so vdso64.so.dbg +obj-vdso64 := $(addprefix $(obj)/, $(obj-vdso64)) + +KBUILD_AFLAGS_64 := $(filter-out -m64,$(KBUILD_AFLAGS)) +KBUILD_AFLAGS_64 += -m64 -s + +KBUILD_CFLAGS_64 := $(filter-out -m64,$(KBUILD_CFLAGS)) +KBUILD_CFLAGS_64 += -m64 -fPIC -shared -fno-common -fno-builtin +KBUILD_CFLAGS_64 += -nostdlib -Wl,-soname=linux-vdso64.so.1 \ + $(call cc-ldoption, -Wl$(comma)--hash-style=sysv) + +$(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_64) +$(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_64) + +obj-y += vdso64_wrapper.o +extra-y += vdso64.lds +CPPFLAGS_vdso64.lds += -P -C -U$(ARCH) + +# Disable gcov profiling for VDSO code +GCOV_PROFILE := n + +# Force dependency (incbin is bad) +$(obj)/vdso64_wrapper.o : $(obj)/vdso64.so + +# link rule for the .so file, .lds has to be first +$(obj)/vdso64.so.dbg: $(src)/vdso64.lds $(obj-vdso64) + $(call if_changed,vdso64ld) + +# strip rule for the .so file +$(obj)/%.so: OBJCOPYFLAGS := -S +$(obj)/%.so: $(obj)/%.so.dbg FORCE + $(call if_changed,objcopy) + +# assembly rules for the .S files +$(obj-vdso64): %.o: %.S + $(call if_changed_dep,vdso64as) + +# actual build commands +quiet_cmd_vdso64ld = VDSO64L $@ + cmd_vdso64ld = $(CC) $(c_flags) -Wl,-T $^ -o $@ +quiet_cmd_vdso64as = VDSO64A $@ + cmd_vdso64as = $(CC) $(a_flags) -c -o $@ $< + +# install commands for the unstripped file +quiet_cmd_vdso_install = INSTALL $@ + cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/$@ + +vdso64.so: $(obj)/vdso64.so.dbg + @mkdir -p $(MODLIB)/vdso + $(call cmd,vdso_install) + +vdso_install: vdso64.so diff --git a/arch/s390/kernel/vdso64/clock_getres.S b/arch/s390/kernel/vdso64/clock_getres.S new file mode 100644 index 000000000..c8513deb8 --- /dev/null +++ b/arch/s390/kernel/vdso64/clock_getres.S @@ -0,0 +1,52 @@ +/* + * Userland implementation of clock_getres() for 64 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.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 <asm/vdso.h> +#include <asm/asm-offsets.h> +#include <asm/unistd.h> + + .text + .align 4 + .globl __kernel_clock_getres + .type __kernel_clock_getres,@function +__kernel_clock_getres: + .cfi_startproc + larl %r1,4f + cghi %r2,__CLOCK_REALTIME_COARSE + je 0f + cghi %r2,__CLOCK_MONOTONIC_COARSE + je 0f + larl %r1,3f + cghi %r2,__CLOCK_REALTIME + je 0f + cghi %r2,__CLOCK_MONOTONIC + je 0f + cghi %r2,__CLOCK_THREAD_CPUTIME_ID + je 0f + cghi %r2,-2 /* Per-thread CPUCLOCK with PID=0, VIRT=1 */ + jne 2f + larl %r5,_vdso_data + icm %r0,15,__LC_ECTG_OK(%r5) + jz 2f +0: ltgr %r3,%r3 + jz 1f /* res == NULL */ + lg %r0,0(%r1) + xc 0(8,%r3),0(%r3) /* set tp->tv_sec to zero */ + stg %r0,8(%r3) /* store tp->tv_usec */ +1: lghi %r2,0 + br %r14 +2: lghi %r1,__NR_clock_getres /* fallback to svc */ + svc 0 + br %r14 +3: .quad __CLOCK_REALTIME_RES +4: .quad __CLOCK_COARSE_RES + .cfi_endproc + .size __kernel_clock_getres,.-__kernel_clock_getres diff --git a/arch/s390/kernel/vdso64/clock_gettime.S b/arch/s390/kernel/vdso64/clock_gettime.S new file mode 100644 index 000000000..61541fb93 --- /dev/null +++ b/arch/s390/kernel/vdso64/clock_gettime.S @@ -0,0 +1,150 @@ +/* + * Userland implementation of clock_gettime() for 64 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.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 <asm/vdso.h> +#include <asm/asm-offsets.h> +#include <asm/unistd.h> + + .text + .align 4 + .globl __kernel_clock_gettime + .type __kernel_clock_gettime,@function +__kernel_clock_gettime: + .cfi_startproc + aghi %r15,-16 + larl %r5,_vdso_data + cghi %r2,__CLOCK_REALTIME_COARSE + je 4f + cghi %r2,__CLOCK_REALTIME + je 5f + cghi %r2,-3 /* Per-thread CPUCLOCK with PID=0, VIRT=1 */ + je 9f + cghi %r2,__CLOCK_MONOTONIC_COARSE + je 3f + cghi %r2,__CLOCK_MONOTONIC + jne 12f + + /* CLOCK_MONOTONIC */ +0: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */ + tmll %r4,0x0001 /* pending update ? loop */ + jnz 0b + stcke 0(%r15) /* Store TOD clock */ + lgf %r2,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */ + lg %r0,__VDSO_WTOM_SEC(%r5) + lg %r1,1(%r15) + sg %r1,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + msgf %r1,__VDSO_TK_MULT(%r5) /* * tk->mult */ + alg %r1,__VDSO_WTOM_NSEC(%r5) + srlg %r1,%r1,0(%r2) /* >> tk->shift */ + clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */ + jne 0b + larl %r5,13f +1: clg %r1,0(%r5) + jl 2f + slg %r1,0(%r5) + aghi %r0,1 + j 1b +2: stg %r0,0(%r3) /* store tp->tv_sec */ + stg %r1,8(%r3) /* store tp->tv_nsec */ + lghi %r2,0 + aghi %r15,16 + br %r14 + + /* CLOCK_MONOTONIC_COARSE */ +3: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */ + tmll %r4,0x0001 /* pending update ? loop */ + jnz 3b + lg %r0,__VDSO_WTOM_CRS_SEC(%r5) + lg %r1,__VDSO_WTOM_CRS_NSEC(%r5) + clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */ + jne 3b + j 2b + + /* CLOCK_REALTIME_COARSE */ +4: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */ + tmll %r4,0x0001 /* pending update ? loop */ + jnz 4b + lg %r0,__VDSO_XTIME_CRS_SEC(%r5) + lg %r1,__VDSO_XTIME_CRS_NSEC(%r5) + clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */ + jne 4b + j 7f + + /* CLOCK_REALTIME */ +5: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */ + tmll %r4,0x0001 /* pending update ? loop */ + jnz 5b + stcke 0(%r15) /* Store TOD clock */ + lgf %r2,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */ + lg %r1,1(%r15) + sg %r1,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + msgf %r1,__VDSO_TK_MULT(%r5) /* * tk->mult */ + alg %r1,__VDSO_XTIME_NSEC(%r5) /* + tk->xtime_nsec */ + srlg %r1,%r1,0(%r2) /* >> tk->shift */ + lg %r0,__VDSO_XTIME_SEC(%r5) /* tk->xtime_sec */ + clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */ + jne 5b + larl %r5,13f +6: clg %r1,0(%r5) + jl 7f + slg %r1,0(%r5) + aghi %r0,1 + j 6b +7: stg %r0,0(%r3) /* store tp->tv_sec */ + stg %r1,8(%r3) /* store tp->tv_nsec */ + lghi %r2,0 + aghi %r15,16 + br %r14 + + /* CPUCLOCK_VIRT for this thread */ +9: icm %r0,15,__VDSO_ECTG_OK(%r5) + jz 12f + ear %r2,%a4 + llilh %r4,0x0100 + sar %a4,%r4 + lghi %r4,0 + epsw %r5,0 + sacf 512 /* Magic ectg instruction */ + .insn ssf,0xc80100000000,__VDSO_ECTG_BASE(4),__VDSO_ECTG_USER(4),4 + tml %r5,0x4000 + jo 11f + tml %r5,0x8000 + jno 10f + sacf 256 + j 11f +10: sacf 0 +11: sar %a4,%r2 + algr %r1,%r0 /* r1 = cputime as TOD value */ + mghi %r1,1000 /* convert to nanoseconds */ + srlg %r1,%r1,12 /* r1 = cputime in nanosec */ + lgr %r4,%r1 + larl %r5,13f + srlg %r1,%r1,9 /* divide by 1000000000 */ + mlg %r0,8(%r5) + srlg %r0,%r0,11 /* r0 = tv_sec */ + stg %r0,0(%r3) + msg %r0,0(%r5) /* calculate tv_nsec */ + slgr %r4,%r0 /* r4 = tv_nsec */ + stg %r4,8(%r3) + lghi %r2,0 + aghi %r15,16 + br %r14 + + /* Fallback to system call */ +12: lghi %r1,__NR_clock_gettime + svc 0 + aghi %r15,16 + br %r14 + +13: .quad 1000000000 +14: .quad 19342813113834067 + .cfi_endproc + .size __kernel_clock_gettime,.-__kernel_clock_gettime diff --git a/arch/s390/kernel/vdso64/gettimeofday.S b/arch/s390/kernel/vdso64/gettimeofday.S new file mode 100644 index 000000000..6ce467076 --- /dev/null +++ b/arch/s390/kernel/vdso64/gettimeofday.S @@ -0,0 +1,59 @@ +/* + * Userland implementation of gettimeofday() for 64 bits processes in a + * s390 kernel for use in the vDSO + * + * Copyright IBM Corp. 2008 + * Author(s): Martin Schwidefsky (schwidefsky@de.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 <asm/vdso.h> +#include <asm/asm-offsets.h> +#include <asm/unistd.h> + + .text + .align 4 + .globl __kernel_gettimeofday + .type __kernel_gettimeofday,@function +__kernel_gettimeofday: + .cfi_startproc + aghi %r15,-16 + larl %r5,_vdso_data +0: ltgr %r3,%r3 /* check if tz is NULL */ + je 1f + mvc 0(8,%r3),__VDSO_TIMEZONE(%r5) +1: ltgr %r2,%r2 /* check if tv is NULL */ + je 4f + lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */ + tmll %r4,0x0001 /* pending update ? loop */ + jnz 0b + stcke 0(%r15) /* Store TOD clock */ + lg %r1,1(%r15) + sg %r1,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */ + msgf %r1,__VDSO_TK_MULT(%r5) /* * tk->mult */ + alg %r1,__VDSO_XTIME_NSEC(%r5) /* + tk->xtime_nsec */ + lg %r0,__VDSO_XTIME_SEC(%r5) /* tk->xtime_sec */ + clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */ + jne 0b + lgf %r5,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */ + srlg %r1,%r1,0(%r5) /* >> tk->shift */ + larl %r5,5f +2: clg %r1,0(%r5) + jl 3f + slg %r1,0(%r5) + aghi %r0,1 + j 2b +3: stg %r0,0(%r2) /* store tv->tv_sec */ + slgr %r0,%r0 /* tv_nsec -> tv_usec */ + ml %r0,8(%r5) + srlg %r0,%r0,6 + stg %r0,8(%r2) /* store tv->tv_usec */ +4: lghi %r2,0 + aghi %r15,16 + br %r14 +5: .quad 1000000000 + .long 274877907 + .cfi_endproc + .size __kernel_gettimeofday,.-__kernel_gettimeofday diff --git a/arch/s390/kernel/vdso64/note.S b/arch/s390/kernel/vdso64/note.S new file mode 100644 index 000000000..79a071e43 --- /dev/null +++ b/arch/s390/kernel/vdso64/note.S @@ -0,0 +1,12 @@ +/* + * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text. + * Here we can supply some information useful to userland. + */ + +#include <linux/uts.h> +#include <linux/version.h> +#include <linux/elfnote.h> + +ELFNOTE_START(Linux, 0, "a") + .long LINUX_VERSION_CODE +ELFNOTE_END diff --git a/arch/s390/kernel/vdso64/vdso64.lds.S b/arch/s390/kernel/vdso64/vdso64.lds.S new file mode 100644 index 000000000..9f5979d10 --- /dev/null +++ b/arch/s390/kernel/vdso64/vdso64.lds.S @@ -0,0 +1,138 @@ +/* + * This is the infamous ld script for the 64 bits vdso + * library + */ +#include <asm/vdso.h> + +OUTPUT_FORMAT("elf64-s390", "elf64-s390", "elf64-s390") +OUTPUT_ARCH(s390:64-bit) +ENTRY(_start) + +SECTIONS +{ + . = VDSO64_LBASE + SIZEOF_HEADERS; + + .hash : { *(.hash) } :text + .gnu.hash : { *(.gnu.hash) } + .dynsym : { *(.dynsym) } + .dynstr : { *(.dynstr) } + .gnu.version : { *(.gnu.version) } + .gnu.version_d : { *(.gnu.version_d) } + .gnu.version_r : { *(.gnu.version_r) } + + .note : { *(.note.*) } :text :note + + . = ALIGN(16); + .text : { + *(.text .stub .text.* .gnu.linkonce.t.*) + } :text + PROVIDE(__etext = .); + PROVIDE(_etext = .); + PROVIDE(etext = .); + + /* + * Other stuff is appended to the text segment: + */ + .rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) } + .rodata1 : { *(.rodata1) } + + .dynamic : { *(.dynamic) } :text :dynamic + + .eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr + .eh_frame : { KEEP (*(.eh_frame)) } :text + .gcc_except_table : { *(.gcc_except_table .gcc_except_table.*) } + + .rela.dyn ALIGN(8) : { *(.rela.dyn) } + .got ALIGN(8) : { *(.got .toc) } + + _end = .; + PROVIDE(end = .); + + /* + * Stabs debugging sections are here too. + */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } + + /* + * DWARF debug sections. + * Symbols in the DWARF debugging sections are relative to the + * beginning of the section so we begin them at 0. + */ + /* DWARF 1 */ + .debug 0 : { *(.debug) } + .line 0 : { *(.line) } + /* GNU DWARF 1 extensions */ + .debug_srcinfo 0 : { *(.debug_srcinfo) } + .debug_sfnames 0 : { *(.debug_sfnames) } + /* DWARF 1.1 and DWARF 2 */ + .debug_aranges 0 : { *(.debug_aranges) } + .debug_pubnames 0 : { *(.debug_pubnames) } + /* DWARF 2 */ + .debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) } + .debug_abbrev 0 : { *(.debug_abbrev) } + .debug_line 0 : { *(.debug_line) } + .debug_frame 0 : { *(.debug_frame) } + .debug_str 0 : { *(.debug_str) } + .debug_loc 0 : { *(.debug_loc) } + .debug_macinfo 0 : { *(.debug_macinfo) } + /* SGI/MIPS DWARF 2 extensions */ + .debug_weaknames 0 : { *(.debug_weaknames) } + .debug_funcnames 0 : { *(.debug_funcnames) } + .debug_typenames 0 : { *(.debug_typenames) } + .debug_varnames 0 : { *(.debug_varnames) } + /* DWARF 3 */ + .debug_pubtypes 0 : { *(.debug_pubtypes) } + .debug_ranges 0 : { *(.debug_ranges) } + .gnu.attributes 0 : { KEEP (*(.gnu.attributes)) } + + . = ALIGN(4096); + PROVIDE(_vdso_data = .); + + /DISCARD/ : { + *(.note.GNU-stack) + *(.branch_lt) + *(.data .data.* .gnu.linkonce.d.* .sdata*) + *(.bss .sbss .dynbss .dynsbss) + } +} + +/* + * Very old versions of ld do not recognize this name token; use the constant. + */ +#define PT_GNU_EH_FRAME 0x6474e550 + +/* + * We must supply the ELF program headers explicitly to get just one + * PT_LOAD segment, and set the flags explicitly to make segments read-only. + */ +PHDRS +{ + text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ + note PT_NOTE FLAGS(4); /* PF_R */ + eh_frame_hdr PT_GNU_EH_FRAME; +} + +/* + * This controls what symbols we export from the DSO. + */ +VERSION +{ + VDSO_VERSION_STRING { + global: + /* + * Has to be there for the kernel to find + */ + __kernel_gettimeofday; + __kernel_clock_gettime; + __kernel_clock_getres; + + local: *; + }; +} diff --git a/arch/s390/kernel/vdso64/vdso64_wrapper.S b/arch/s390/kernel/vdso64/vdso64_wrapper.S new file mode 100644 index 000000000..c245842b5 --- /dev/null +++ b/arch/s390/kernel/vdso64/vdso64_wrapper.S @@ -0,0 +1,14 @@ +#include <linux/init.h> +#include <linux/linkage.h> +#include <asm/page.h> + + __PAGE_ALIGNED_DATA + + .globl vdso64_start, vdso64_end + .balign PAGE_SIZE +vdso64_start: + .incbin "arch/s390/kernel/vdso64/vdso64.so" + .balign PAGE_SIZE +vdso64_end: + + .previous diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S new file mode 100644 index 000000000..445657fe6 --- /dev/null +++ b/arch/s390/kernel/vmlinux.lds.S @@ -0,0 +1,93 @@ +/* ld script to make s390 Linux kernel + * Written by Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#include <asm/thread_info.h> +#include <asm/page.h> +#include <asm-generic/vmlinux.lds.h> + +OUTPUT_FORMAT("elf64-s390", "elf64-s390", "elf64-s390") +OUTPUT_ARCH(s390:64-bit) +ENTRY(startup) +jiffies = jiffies_64; + +PHDRS { + text PT_LOAD FLAGS(5); /* R_E */ + data PT_LOAD FLAGS(7); /* RWE */ + note PT_NOTE FLAGS(0); /* ___ */ +} + +SECTIONS +{ + . = 0x00000000; + .text : { + _text = .; /* Text and read-only data */ + HEAD_TEXT + TEXT_TEXT + SCHED_TEXT + LOCK_TEXT + KPROBES_TEXT + IRQENTRY_TEXT + *(.fixup) + *(.gnu.warning) + } :text = 0x0700 + + _etext = .; /* End of text section */ + + NOTES :text :note + + .dummy : { *(.dummy) } :data + + RO_DATA_SECTION(PAGE_SIZE) + +#ifdef CONFIG_SHARED_KERNEL + . = ALIGN(0x100000); /* VM shared segments are 1MB aligned */ +#endif + + . = ALIGN(PAGE_SIZE); + _eshared = .; /* End of shareable data */ + _sdata = .; /* Start of data section */ + + EXCEPTION_TABLE(16) :data + + RW_DATA_SECTION(0x100, PAGE_SIZE, THREAD_SIZE) + + _edata = .; /* End of data section */ + + /* will be freed after init */ + . = ALIGN(PAGE_SIZE); /* Init code and data */ + __init_begin = .; + + INIT_TEXT_SECTION(PAGE_SIZE) + + /* + * .exit.text is discarded at runtime, not link time, + * to deal with references from __bug_table + */ + .exit.text : { + EXIT_TEXT + } + + .exit.data : { + EXIT_DATA + } + + /* early.c uses stsi, which requires page aligned data. */ + . = ALIGN(PAGE_SIZE); + INIT_DATA_SECTION(0x100) + + PERCPU_SECTION(0x100) + . = ALIGN(PAGE_SIZE); + __init_end = .; /* freed after init ends here */ + + BSS_SECTION(0, 2, 0) + + _end = . ; + + /* Debugging sections. */ + STABS_DEBUG + DWARF_DEBUG + + /* Sections to be discarded */ + DISCARDS +} diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c new file mode 100644 index 000000000..e53d3595a --- /dev/null +++ b/arch/s390/kernel/vtime.c @@ -0,0 +1,379 @@ +/* + * Virtual cpu timer based timer functions. + * + * Copyright IBM Corp. 2004, 2012 + * Author(s): Jan Glauber <jan.glauber@de.ibm.com> + */ + +#include <linux/kernel_stat.h> +#include <linux/export.h> +#include <linux/kernel.h> +#include <linux/timex.h> +#include <linux/types.h> +#include <linux/time.h> + +#include <asm/cputime.h> +#include <asm/vtimer.h> +#include <asm/vtime.h> +#include <asm/cpu_mf.h> +#include <asm/smp.h> + +static void virt_timer_expire(void); + +static LIST_HEAD(virt_timer_list); +static DEFINE_SPINLOCK(virt_timer_lock); +static atomic64_t virt_timer_current; +static atomic64_t virt_timer_elapsed; + +static DEFINE_PER_CPU(u64, mt_cycles[32]); +static DEFINE_PER_CPU(u64, mt_scaling_mult) = { 1 }; +static DEFINE_PER_CPU(u64, mt_scaling_div) = { 1 }; + +static inline u64 get_vtimer(void) +{ + u64 timer; + + asm volatile("stpt %0" : "=m" (timer)); + return timer; +} + +static inline void set_vtimer(u64 expires) +{ + u64 timer; + + asm volatile( + " stpt %0\n" /* Store current cpu timer value */ + " spt %1" /* Set new value imm. afterwards */ + : "=m" (timer) : "m" (expires)); + S390_lowcore.system_timer += S390_lowcore.last_update_timer - timer; + S390_lowcore.last_update_timer = expires; +} + +static inline int virt_timer_forward(u64 elapsed) +{ + BUG_ON(!irqs_disabled()); + + if (list_empty(&virt_timer_list)) + return 0; + elapsed = atomic64_add_return(elapsed, &virt_timer_elapsed); + return elapsed >= atomic64_read(&virt_timer_current); +} + +/* + * Update process times based on virtual cpu times stored by entry.S + * to the lowcore fields user_timer, system_timer & steal_clock. + */ +static int do_account_vtime(struct task_struct *tsk, int hardirq_offset) +{ + struct thread_info *ti = task_thread_info(tsk); + u64 timer, clock, user, system, steal; + u64 user_scaled, system_scaled; + int i; + + timer = S390_lowcore.last_update_timer; + clock = S390_lowcore.last_update_clock; + asm volatile( + " stpt %0\n" /* Store current cpu timer value */ +#ifdef CONFIG_HAVE_MARCH_Z9_109_FEATURES + " stckf %1" /* Store current tod clock value */ +#else + " stck %1" /* Store current tod clock value */ +#endif + : "=m" (S390_lowcore.last_update_timer), + "=m" (S390_lowcore.last_update_clock)); + S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer; + S390_lowcore.steal_timer += S390_lowcore.last_update_clock - clock; + + /* Do MT utilization calculation */ + if (smp_cpu_mtid) { + u64 cycles_new[32], *cycles_old; + u64 delta, mult, div; + + cycles_old = this_cpu_ptr(mt_cycles); + if (stcctm5(smp_cpu_mtid + 1, cycles_new) < 2) { + mult = div = 0; + for (i = 0; i <= smp_cpu_mtid; i++) { + delta = cycles_new[i] - cycles_old[i]; + mult += delta; + div += (i + 1) * delta; + } + if (mult > 0) { + /* Update scaling factor */ + __this_cpu_write(mt_scaling_mult, mult); + __this_cpu_write(mt_scaling_div, div); + memcpy(cycles_old, cycles_new, + sizeof(u64) * (smp_cpu_mtid + 1)); + } + } + } + + user = S390_lowcore.user_timer - ti->user_timer; + S390_lowcore.steal_timer -= user; + ti->user_timer = S390_lowcore.user_timer; + + system = S390_lowcore.system_timer - ti->system_timer; + S390_lowcore.steal_timer -= system; + ti->system_timer = S390_lowcore.system_timer; + + user_scaled = user; + system_scaled = system; + /* Do MT utilization scaling */ + if (smp_cpu_mtid) { + u64 mult = __this_cpu_read(mt_scaling_mult); + u64 div = __this_cpu_read(mt_scaling_div); + + user_scaled = (user_scaled * mult) / div; + system_scaled = (system_scaled * mult) / div; + } + account_user_time(tsk, user, user_scaled); + account_system_time(tsk, hardirq_offset, system, system_scaled); + + steal = S390_lowcore.steal_timer; + if ((s64) steal > 0) { + S390_lowcore.steal_timer = 0; + account_steal_time(steal); + } + + return virt_timer_forward(user + system); +} + +void vtime_task_switch(struct task_struct *prev) +{ + struct thread_info *ti; + + do_account_vtime(prev, 0); + ti = task_thread_info(prev); + ti->user_timer = S390_lowcore.user_timer; + ti->system_timer = S390_lowcore.system_timer; + ti = task_thread_info(current); + S390_lowcore.user_timer = ti->user_timer; + S390_lowcore.system_timer = ti->system_timer; +} + +/* + * In s390, accounting pending user time also implies + * accounting system time in order to correctly compute + * the stolen time accounting. + */ +void vtime_account_user(struct task_struct *tsk) +{ + if (do_account_vtime(tsk, HARDIRQ_OFFSET)) + virt_timer_expire(); +} + +/* + * Update process times based on virtual cpu times stored by entry.S + * to the lowcore fields user_timer, system_timer & steal_clock. + */ +void vtime_account_irq_enter(struct task_struct *tsk) +{ + struct thread_info *ti = task_thread_info(tsk); + u64 timer, system, system_scaled; + + timer = S390_lowcore.last_update_timer; + S390_lowcore.last_update_timer = get_vtimer(); + S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer; + + system = S390_lowcore.system_timer - ti->system_timer; + S390_lowcore.steal_timer -= system; + ti->system_timer = S390_lowcore.system_timer; + system_scaled = system; + /* Do MT utilization scaling */ + if (smp_cpu_mtid) { + u64 mult = __this_cpu_read(mt_scaling_mult); + u64 div = __this_cpu_read(mt_scaling_div); + + system_scaled = (system_scaled * mult) / div; + } + account_system_time(tsk, 0, system, system_scaled); + + virt_timer_forward(system); +} +EXPORT_SYMBOL_GPL(vtime_account_irq_enter); + +void vtime_account_system(struct task_struct *tsk) +__attribute__((alias("vtime_account_irq_enter"))); +EXPORT_SYMBOL_GPL(vtime_account_system); + +/* + * Sorted add to a list. List is linear searched until first bigger + * element is found. + */ +static void list_add_sorted(struct vtimer_list *timer, struct list_head *head) +{ + struct vtimer_list *tmp; + + list_for_each_entry(tmp, head, entry) { + if (tmp->expires > timer->expires) { + list_add_tail(&timer->entry, &tmp->entry); + return; + } + } + list_add_tail(&timer->entry, head); +} + +/* + * Handler for expired virtual CPU timer. + */ +static void virt_timer_expire(void) +{ + struct vtimer_list *timer, *tmp; + unsigned long elapsed; + LIST_HEAD(cb_list); + + /* walk timer list, fire all expired timers */ + spin_lock(&virt_timer_lock); + elapsed = atomic64_read(&virt_timer_elapsed); + list_for_each_entry_safe(timer, tmp, &virt_timer_list, entry) { + if (timer->expires < elapsed) + /* move expired timer to the callback queue */ + list_move_tail(&timer->entry, &cb_list); + else + timer->expires -= elapsed; + } + if (!list_empty(&virt_timer_list)) { + timer = list_first_entry(&virt_timer_list, + struct vtimer_list, entry); + atomic64_set(&virt_timer_current, timer->expires); + } + atomic64_sub(elapsed, &virt_timer_elapsed); + spin_unlock(&virt_timer_lock); + + /* Do callbacks and recharge periodic timers */ + list_for_each_entry_safe(timer, tmp, &cb_list, entry) { + list_del_init(&timer->entry); + timer->function(timer->data); + if (timer->interval) { + /* Recharge interval timer */ + timer->expires = timer->interval + + atomic64_read(&virt_timer_elapsed); + spin_lock(&virt_timer_lock); + list_add_sorted(timer, &virt_timer_list); + spin_unlock(&virt_timer_lock); + } + } +} + +void init_virt_timer(struct vtimer_list *timer) +{ + timer->function = NULL; + INIT_LIST_HEAD(&timer->entry); +} +EXPORT_SYMBOL(init_virt_timer); + +static inline int vtimer_pending(struct vtimer_list *timer) +{ + return !list_empty(&timer->entry); +} + +static void internal_add_vtimer(struct vtimer_list *timer) +{ + if (list_empty(&virt_timer_list)) { + /* First timer, just program it. */ + atomic64_set(&virt_timer_current, timer->expires); + atomic64_set(&virt_timer_elapsed, 0); + list_add(&timer->entry, &virt_timer_list); + } else { + /* Update timer against current base. */ + timer->expires += atomic64_read(&virt_timer_elapsed); + if (likely((s64) timer->expires < + (s64) atomic64_read(&virt_timer_current))) + /* The new timer expires before the current timer. */ + atomic64_set(&virt_timer_current, timer->expires); + /* Insert new timer into the list. */ + list_add_sorted(timer, &virt_timer_list); + } +} + +static void __add_vtimer(struct vtimer_list *timer, int periodic) +{ + unsigned long flags; + + timer->interval = periodic ? timer->expires : 0; + spin_lock_irqsave(&virt_timer_lock, flags); + internal_add_vtimer(timer); + spin_unlock_irqrestore(&virt_timer_lock, flags); +} + +/* + * add_virt_timer - add an oneshot virtual CPU timer + */ +void add_virt_timer(struct vtimer_list *timer) +{ + __add_vtimer(timer, 0); +} +EXPORT_SYMBOL(add_virt_timer); + +/* + * add_virt_timer_int - add an interval virtual CPU timer + */ +void add_virt_timer_periodic(struct vtimer_list *timer) +{ + __add_vtimer(timer, 1); +} +EXPORT_SYMBOL(add_virt_timer_periodic); + +static int __mod_vtimer(struct vtimer_list *timer, u64 expires, int periodic) +{ + unsigned long flags; + int rc; + + BUG_ON(!timer->function); + + if (timer->expires == expires && vtimer_pending(timer)) + return 1; + spin_lock_irqsave(&virt_timer_lock, flags); + rc = vtimer_pending(timer); + if (rc) + list_del_init(&timer->entry); + timer->interval = periodic ? expires : 0; + timer->expires = expires; + internal_add_vtimer(timer); + spin_unlock_irqrestore(&virt_timer_lock, flags); + return rc; +} + +/* + * returns whether it has modified a pending timer (1) or not (0) + */ +int mod_virt_timer(struct vtimer_list *timer, u64 expires) +{ + return __mod_vtimer(timer, expires, 0); +} +EXPORT_SYMBOL(mod_virt_timer); + +/* + * returns whether it has modified a pending timer (1) or not (0) + */ +int mod_virt_timer_periodic(struct vtimer_list *timer, u64 expires) +{ + return __mod_vtimer(timer, expires, 1); +} +EXPORT_SYMBOL(mod_virt_timer_periodic); + +/* + * Delete a virtual timer. + * + * returns whether the deleted timer was pending (1) or not (0) + */ +int del_virt_timer(struct vtimer_list *timer) +{ + unsigned long flags; + + if (!vtimer_pending(timer)) + return 0; + spin_lock_irqsave(&virt_timer_lock, flags); + list_del_init(&timer->entry); + spin_unlock_irqrestore(&virt_timer_lock, flags); + return 1; +} +EXPORT_SYMBOL(del_virt_timer); + +/* + * Start the virtual CPU timer on the current CPU. + */ +void vtime_init(void) +{ + /* set initial cpu timer */ + set_vtimer(VTIMER_MAX_SLICE); +} diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig new file mode 100644 index 000000000..5fce52cf0 --- /dev/null +++ b/arch/s390/kvm/Kconfig @@ -0,0 +1,58 @@ +# +# KVM configuration +# +source "virt/kvm/Kconfig" + +menuconfig VIRTUALIZATION + def_bool y + prompt "KVM" + ---help--- + Say Y here to get to see options for using your Linux host to run other + operating systems inside virtual machines (guests). + This option alone does not add any kernel code. + + If you say N, all options in this submenu will be skipped and disabled. + +if VIRTUALIZATION + +config KVM + def_tristate y + prompt "Kernel-based Virtual Machine (KVM) support" + depends on HAVE_KVM + select PREEMPT_NOTIFIERS + select ANON_INODES + select HAVE_KVM_CPU_RELAX_INTERCEPT + select HAVE_KVM_EVENTFD + select KVM_ASYNC_PF + select KVM_ASYNC_PF_SYNC + select HAVE_KVM_IRQCHIP + select HAVE_KVM_IRQFD + select HAVE_KVM_IRQ_ROUTING + select SRCU + ---help--- + Support hosting paravirtualized guest machines using the SIE + virtualization capability on the mainframe. This should work + on any 64bit machine. + + This module provides access to the hardware capabilities through + a character device node named /dev/kvm. + + To compile this as a module, choose M here: the module + will be called kvm. + + If unsure, say N. + +config KVM_S390_UCONTROL + bool "Userspace controlled virtual machines" + depends on KVM + ---help--- + Allow CAP_SYS_ADMIN users to create KVM virtual machines that are + controlled by userspace. + + If unsure, say N. + +# OK, it's a little counter-intuitive to do this, but it puts it neatly under +# the virtualization menu. +source drivers/vhost/Kconfig + +endif # VIRTUALIZATION diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile new file mode 100644 index 000000000..b3b553469 --- /dev/null +++ b/arch/s390/kvm/Makefile @@ -0,0 +1,17 @@ +# Makefile for kernel virtual machines on s390 +# +# Copyright IBM Corp. 2008 +# +# 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. + +KVM := ../../../virt/kvm +common-objs = $(KVM)/kvm_main.o $(KVM)/eventfd.o $(KVM)/async_pf.o $(KVM)/irqchip.o + +ccflags-y := -Ivirt/kvm -Iarch/s390/kvm + +kvm-objs := $(common-objs) kvm-s390.o intercept.o interrupt.o priv.o sigp.o +kvm-objs += diag.o gaccess.o guestdbg.o + +obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c new file mode 100644 index 000000000..fc7ec9584 --- /dev/null +++ b/arch/s390/kvm/diag.c @@ -0,0 +1,255 @@ +/* + * handling diagnose instructions + * + * Copyright IBM Corp. 2008, 2011 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + * Christian Borntraeger <borntraeger@de.ibm.com> + */ + +#include <linux/kvm.h> +#include <linux/kvm_host.h> +#include <asm/pgalloc.h> +#include <asm/virtio-ccw.h> +#include "kvm-s390.h" +#include "trace.h" +#include "trace-s390.h" +#include "gaccess.h" + +static int diag_release_pages(struct kvm_vcpu *vcpu) +{ + unsigned long start, end; + unsigned long prefix = kvm_s390_get_prefix(vcpu); + + start = vcpu->run->s.regs.gprs[(vcpu->arch.sie_block->ipa & 0xf0) >> 4]; + end = vcpu->run->s.regs.gprs[vcpu->arch.sie_block->ipa & 0xf] + 4096; + + if (start & ~PAGE_MASK || end & ~PAGE_MASK || start >= end + || start < 2 * PAGE_SIZE) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + VCPU_EVENT(vcpu, 5, "diag release pages %lX %lX", start, end); + vcpu->stat.diagnose_10++; + + /* + * We checked for start >= end above, so lets check for the + * fast path (no prefix swap page involved) + */ + if (end <= prefix || start >= prefix + 2 * PAGE_SIZE) { + gmap_discard(vcpu->arch.gmap, start, end); + } else { + /* + * This is slow path. gmap_discard will check for start + * so lets split this into before prefix, prefix, after + * prefix and let gmap_discard make some of these calls + * NOPs. + */ + gmap_discard(vcpu->arch.gmap, start, prefix); + if (start <= prefix) + gmap_discard(vcpu->arch.gmap, 0, 4096); + if (end > prefix + 4096) + gmap_discard(vcpu->arch.gmap, 4096, 8192); + gmap_discard(vcpu->arch.gmap, prefix + 2 * PAGE_SIZE, end); + } + return 0; +} + +static int __diag_page_ref_service(struct kvm_vcpu *vcpu) +{ + struct prs_parm { + u16 code; + u16 subcode; + u16 parm_len; + u16 parm_version; + u64 token_addr; + u64 select_mask; + u64 compare_mask; + u64 zarch; + }; + struct prs_parm parm; + int rc; + u16 rx = (vcpu->arch.sie_block->ipa & 0xf0) >> 4; + u16 ry = (vcpu->arch.sie_block->ipa & 0x0f); + + if (vcpu->run->s.regs.gprs[rx] & 7) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + rc = read_guest(vcpu, vcpu->run->s.regs.gprs[rx], rx, &parm, sizeof(parm)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + if (parm.parm_version != 2 || parm.parm_len < 5 || parm.code != 0x258) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + switch (parm.subcode) { + case 0: /* TOKEN */ + if (vcpu->arch.pfault_token != KVM_S390_PFAULT_TOKEN_INVALID) { + /* + * If the pagefault handshake is already activated, + * the token must not be changed. We have to return + * decimal 8 instead, as mandated in SC24-6084. + */ + vcpu->run->s.regs.gprs[ry] = 8; + return 0; + } + + if ((parm.compare_mask & parm.select_mask) != parm.compare_mask || + parm.token_addr & 7 || parm.zarch != 0x8000000000000000ULL) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + if (kvm_is_error_gpa(vcpu->kvm, parm.token_addr)) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + + vcpu->arch.pfault_token = parm.token_addr; + vcpu->arch.pfault_select = parm.select_mask; + vcpu->arch.pfault_compare = parm.compare_mask; + vcpu->run->s.regs.gprs[ry] = 0; + rc = 0; + break; + case 1: /* + * CANCEL + * Specification allows to let already pending tokens survive + * the cancel, therefore to reduce code complexity, we assume + * all outstanding tokens are already pending. + */ + if (parm.token_addr || parm.select_mask || + parm.compare_mask || parm.zarch) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + vcpu->run->s.regs.gprs[ry] = 0; + /* + * If the pfault handling was not established or is already + * canceled SC24-6084 requests to return decimal 4. + */ + if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID) + vcpu->run->s.regs.gprs[ry] = 4; + else + vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID; + + rc = 0; + break; + default: + rc = -EOPNOTSUPP; + break; + } + + return rc; +} + +static int __diag_time_slice_end(struct kvm_vcpu *vcpu) +{ + VCPU_EVENT(vcpu, 5, "%s", "diag time slice end"); + vcpu->stat.diagnose_44++; + kvm_vcpu_on_spin(vcpu); + return 0; +} + +static int __diag_time_slice_end_directed(struct kvm_vcpu *vcpu) +{ + struct kvm *kvm = vcpu->kvm; + struct kvm_vcpu *tcpu; + int tid; + int i; + + tid = vcpu->run->s.regs.gprs[(vcpu->arch.sie_block->ipa & 0xf0) >> 4]; + vcpu->stat.diagnose_9c++; + VCPU_EVENT(vcpu, 5, "diag time slice end directed to %d", tid); + + if (tid == vcpu->vcpu_id) + return 0; + + kvm_for_each_vcpu(i, tcpu, kvm) + if (tcpu->vcpu_id == tid) { + kvm_vcpu_yield_to(tcpu); + break; + } + + return 0; +} + +static int __diag_ipl_functions(struct kvm_vcpu *vcpu) +{ + unsigned int reg = vcpu->arch.sie_block->ipa & 0xf; + unsigned long subcode = vcpu->run->s.regs.gprs[reg] & 0xffff; + + VCPU_EVENT(vcpu, 5, "diag ipl functions, subcode %lx", subcode); + switch (subcode) { + case 3: + vcpu->run->s390_reset_flags = KVM_S390_RESET_CLEAR; + break; + case 4: + vcpu->run->s390_reset_flags = 0; + break; + default: + return -EOPNOTSUPP; + } + + if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) + kvm_s390_vcpu_stop(vcpu); + vcpu->run->s390_reset_flags |= KVM_S390_RESET_SUBSYSTEM; + vcpu->run->s390_reset_flags |= KVM_S390_RESET_IPL; + vcpu->run->s390_reset_flags |= KVM_S390_RESET_CPU_INIT; + vcpu->run->exit_reason = KVM_EXIT_S390_RESET; + VCPU_EVENT(vcpu, 3, "requesting userspace resets %llx", + vcpu->run->s390_reset_flags); + trace_kvm_s390_request_resets(vcpu->run->s390_reset_flags); + return -EREMOTE; +} + +static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu) +{ + int ret; + + /* No virtio-ccw notification? Get out quickly. */ + if (!vcpu->kvm->arch.css_support || + (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY)) + return -EOPNOTSUPP; + + /* + * The layout is as follows: + * - gpr 2 contains the subchannel id (passed as addr) + * - gpr 3 contains the virtqueue index (passed as datamatch) + * - gpr 4 contains the index on the bus (optionally) + */ + ret = kvm_io_bus_write_cookie(vcpu, KVM_VIRTIO_CCW_NOTIFY_BUS, + vcpu->run->s.regs.gprs[2] & 0xffffffff, + 8, &vcpu->run->s.regs.gprs[3], + vcpu->run->s.regs.gprs[4]); + + /* + * Return cookie in gpr 2, but don't overwrite the register if the + * diagnose will be handled by userspace. + */ + if (ret != -EOPNOTSUPP) + vcpu->run->s.regs.gprs[2] = ret; + /* kvm_io_bus_write_cookie returns -EOPNOTSUPP if it found no match. */ + return ret < 0 ? ret : 0; +} + +int kvm_s390_handle_diag(struct kvm_vcpu *vcpu) +{ + int code = kvm_s390_get_base_disp_rs(vcpu, NULL) & 0xffff; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + trace_kvm_s390_handle_diag(vcpu, code); + switch (code) { + case 0x10: + return diag_release_pages(vcpu); + case 0x44: + return __diag_time_slice_end(vcpu); + case 0x9c: + return __diag_time_slice_end_directed(vcpu); + case 0x258: + return __diag_page_ref_service(vcpu); + case 0x308: + return __diag_ipl_functions(vcpu); + case 0x500: + return __diag_virtio_hypercall(vcpu); + default: + return -EOPNOTSUPP; + } +} diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c new file mode 100644 index 000000000..a7559f720 --- /dev/null +++ b/arch/s390/kvm/gaccess.c @@ -0,0 +1,915 @@ +/* + * guest access functions + * + * Copyright IBM Corp. 2014 + * + */ + +#include <linux/vmalloc.h> +#include <linux/err.h> +#include <asm/pgtable.h> +#include "kvm-s390.h" +#include "gaccess.h" +#include <asm/switch_to.h> + +union asce { + unsigned long val; + struct { + unsigned long origin : 52; /* Region- or Segment-Table Origin */ + unsigned long : 2; + unsigned long g : 1; /* Subspace Group Control */ + unsigned long p : 1; /* Private Space Control */ + unsigned long s : 1; /* Storage-Alteration-Event Control */ + unsigned long x : 1; /* Space-Switch-Event Control */ + unsigned long r : 1; /* Real-Space Control */ + unsigned long : 1; + unsigned long dt : 2; /* Designation-Type Control */ + unsigned long tl : 2; /* Region- or Segment-Table Length */ + }; +}; + +enum { + ASCE_TYPE_SEGMENT = 0, + ASCE_TYPE_REGION3 = 1, + ASCE_TYPE_REGION2 = 2, + ASCE_TYPE_REGION1 = 3 +}; + +union region1_table_entry { + unsigned long val; + struct { + unsigned long rto: 52;/* Region-Table Origin */ + unsigned long : 2; + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 1; + unsigned long tf : 2; /* Region-Second-Table Offset */ + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long : 1; + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long tl : 2; /* Region-Second-Table Length */ + }; +}; + +union region2_table_entry { + unsigned long val; + struct { + unsigned long rto: 52;/* Region-Table Origin */ + unsigned long : 2; + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 1; + unsigned long tf : 2; /* Region-Third-Table Offset */ + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long : 1; + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long tl : 2; /* Region-Third-Table Length */ + }; +}; + +struct region3_table_entry_fc0 { + unsigned long sto: 52;/* Segment-Table Origin */ + unsigned long : 1; + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 1; + unsigned long tf : 2; /* Segment-Table Offset */ + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long cr : 1; /* Common-Region Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long tl : 2; /* Segment-Table Length */ +}; + +struct region3_table_entry_fc1 { + unsigned long rfaa : 33; /* Region-Frame Absolute Address */ + unsigned long : 14; + unsigned long av : 1; /* ACCF-Validity Control */ + unsigned long acc: 4; /* Access-Control Bits */ + unsigned long f : 1; /* Fetch-Protection Bit */ + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long co : 1; /* Change-Recording Override */ + unsigned long : 2; + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long cr : 1; /* Common-Region Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; +}; + +union region3_table_entry { + unsigned long val; + struct region3_table_entry_fc0 fc0; + struct region3_table_entry_fc1 fc1; + struct { + unsigned long : 53; + unsigned long fc : 1; /* Format-Control */ + unsigned long : 4; + unsigned long i : 1; /* Region-Invalid Bit */ + unsigned long cr : 1; /* Common-Region Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; + }; +}; + +struct segment_entry_fc0 { + unsigned long pto: 53;/* Page-Table Origin */ + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long : 3; + unsigned long i : 1; /* Segment-Invalid Bit */ + unsigned long cs : 1; /* Common-Segment Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; +}; + +struct segment_entry_fc1 { + unsigned long sfaa : 44; /* Segment-Frame Absolute Address */ + unsigned long : 3; + unsigned long av : 1; /* ACCF-Validity Control */ + unsigned long acc: 4; /* Access-Control Bits */ + unsigned long f : 1; /* Fetch-Protection Bit */ + unsigned long fc : 1; /* Format-Control */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long co : 1; /* Change-Recording Override */ + unsigned long : 2; + unsigned long i : 1; /* Segment-Invalid Bit */ + unsigned long cs : 1; /* Common-Segment Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; +}; + +union segment_table_entry { + unsigned long val; + struct segment_entry_fc0 fc0; + struct segment_entry_fc1 fc1; + struct { + unsigned long : 53; + unsigned long fc : 1; /* Format-Control */ + unsigned long : 4; + unsigned long i : 1; /* Segment-Invalid Bit */ + unsigned long cs : 1; /* Common-Segment Bit */ + unsigned long tt : 2; /* Table-Type Bits */ + unsigned long : 2; + }; +}; + +enum { + TABLE_TYPE_SEGMENT = 0, + TABLE_TYPE_REGION3 = 1, + TABLE_TYPE_REGION2 = 2, + TABLE_TYPE_REGION1 = 3 +}; + +union page_table_entry { + unsigned long val; + struct { + unsigned long pfra : 52; /* Page-Frame Real Address */ + unsigned long z : 1; /* Zero Bit */ + unsigned long i : 1; /* Page-Invalid Bit */ + unsigned long p : 1; /* DAT-Protection Bit */ + unsigned long co : 1; /* Change-Recording Override */ + unsigned long : 8; + }; +}; + +/* + * vaddress union in order to easily decode a virtual address into its + * region first index, region second index etc. parts. + */ +union vaddress { + unsigned long addr; + struct { + unsigned long rfx : 11; + unsigned long rsx : 11; + unsigned long rtx : 11; + unsigned long sx : 11; + unsigned long px : 8; + unsigned long bx : 12; + }; + struct { + unsigned long rfx01 : 2; + unsigned long : 9; + unsigned long rsx01 : 2; + unsigned long : 9; + unsigned long rtx01 : 2; + unsigned long : 9; + unsigned long sx01 : 2; + unsigned long : 29; + }; +}; + +/* + * raddress union which will contain the result (real or absolute address) + * after a page table walk. The rfaa, sfaa and pfra members are used to + * simply assign them the value of a region, segment or page table entry. + */ +union raddress { + unsigned long addr; + unsigned long rfaa : 33; /* Region-Frame Absolute Address */ + unsigned long sfaa : 44; /* Segment-Frame Absolute Address */ + unsigned long pfra : 52; /* Page-Frame Real Address */ +}; + +union alet { + u32 val; + struct { + u32 reserved : 7; + u32 p : 1; + u32 alesn : 8; + u32 alen : 16; + }; +}; + +union ald { + u32 val; + struct { + u32 : 1; + u32 alo : 24; + u32 all : 7; + }; +}; + +struct ale { + unsigned long i : 1; /* ALEN-Invalid Bit */ + unsigned long : 5; + unsigned long fo : 1; /* Fetch-Only Bit */ + unsigned long p : 1; /* Private Bit */ + unsigned long alesn : 8; /* Access-List-Entry Sequence Number */ + unsigned long aleax : 16; /* Access-List-Entry Authorization Index */ + unsigned long : 32; + unsigned long : 1; + unsigned long asteo : 25; /* ASN-Second-Table-Entry Origin */ + unsigned long : 6; + unsigned long astesn : 32; /* ASTE Sequence Number */ +} __packed; + +struct aste { + unsigned long i : 1; /* ASX-Invalid Bit */ + unsigned long ato : 29; /* Authority-Table Origin */ + unsigned long : 1; + unsigned long b : 1; /* Base-Space Bit */ + unsigned long ax : 16; /* Authorization Index */ + unsigned long atl : 12; /* Authority-Table Length */ + unsigned long : 2; + unsigned long ca : 1; /* Controlled-ASN Bit */ + unsigned long ra : 1; /* Reusable-ASN Bit */ + unsigned long asce : 64; /* Address-Space-Control Element */ + unsigned long ald : 32; + unsigned long astesn : 32; + /* .. more fields there */ +} __packed; + +int ipte_lock_held(struct kvm_vcpu *vcpu) +{ + union ipte_control *ic = &vcpu->kvm->arch.sca->ipte_control; + + if (vcpu->arch.sie_block->eca & 1) + return ic->kh != 0; + return vcpu->kvm->arch.ipte_lock_count != 0; +} + +static void ipte_lock_simple(struct kvm_vcpu *vcpu) +{ + union ipte_control old, new, *ic; + + mutex_lock(&vcpu->kvm->arch.ipte_mutex); + vcpu->kvm->arch.ipte_lock_count++; + if (vcpu->kvm->arch.ipte_lock_count > 1) + goto out; + ic = &vcpu->kvm->arch.sca->ipte_control; + do { + old = READ_ONCE(*ic); + while (old.k) { + cond_resched(); + old = READ_ONCE(*ic); + } + new = old; + new.k = 1; + } while (cmpxchg(&ic->val, old.val, new.val) != old.val); +out: + mutex_unlock(&vcpu->kvm->arch.ipte_mutex); +} + +static void ipte_unlock_simple(struct kvm_vcpu *vcpu) +{ + union ipte_control old, new, *ic; + + mutex_lock(&vcpu->kvm->arch.ipte_mutex); + vcpu->kvm->arch.ipte_lock_count--; + if (vcpu->kvm->arch.ipte_lock_count) + goto out; + ic = &vcpu->kvm->arch.sca->ipte_control; + do { + old = READ_ONCE(*ic); + new = old; + new.k = 0; + } while (cmpxchg(&ic->val, old.val, new.val) != old.val); + wake_up(&vcpu->kvm->arch.ipte_wq); +out: + mutex_unlock(&vcpu->kvm->arch.ipte_mutex); +} + +static void ipte_lock_siif(struct kvm_vcpu *vcpu) +{ + union ipte_control old, new, *ic; + + ic = &vcpu->kvm->arch.sca->ipte_control; + do { + old = READ_ONCE(*ic); + while (old.kg) { + cond_resched(); + old = READ_ONCE(*ic); + } + new = old; + new.k = 1; + new.kh++; + } while (cmpxchg(&ic->val, old.val, new.val) != old.val); +} + +static void ipte_unlock_siif(struct kvm_vcpu *vcpu) +{ + union ipte_control old, new, *ic; + + ic = &vcpu->kvm->arch.sca->ipte_control; + do { + old = READ_ONCE(*ic); + new = old; + new.kh--; + if (!new.kh) + new.k = 0; + } while (cmpxchg(&ic->val, old.val, new.val) != old.val); + if (!new.kh) + wake_up(&vcpu->kvm->arch.ipte_wq); +} + +void ipte_lock(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.sie_block->eca & 1) + ipte_lock_siif(vcpu); + else + ipte_lock_simple(vcpu); +} + +void ipte_unlock(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.sie_block->eca & 1) + ipte_unlock_siif(vcpu); + else + ipte_unlock_simple(vcpu); +} + +static int ar_translation(struct kvm_vcpu *vcpu, union asce *asce, ar_t ar, + int write) +{ + union alet alet; + struct ale ale; + struct aste aste; + unsigned long ald_addr, authority_table_addr; + union ald ald; + int eax, rc; + u8 authority_table; + + if (ar >= NUM_ACRS) + return -EINVAL; + + save_access_regs(vcpu->run->s.regs.acrs); + alet.val = vcpu->run->s.regs.acrs[ar]; + + if (ar == 0 || alet.val == 0) { + asce->val = vcpu->arch.sie_block->gcr[1]; + return 0; + } else if (alet.val == 1) { + asce->val = vcpu->arch.sie_block->gcr[7]; + return 0; + } + + if (alet.reserved) + return PGM_ALET_SPECIFICATION; + + if (alet.p) + ald_addr = vcpu->arch.sie_block->gcr[5]; + else + ald_addr = vcpu->arch.sie_block->gcr[2]; + ald_addr &= 0x7fffffc0; + + rc = read_guest_real(vcpu, ald_addr + 16, &ald.val, sizeof(union ald)); + if (rc) + return rc; + + if (alet.alen / 8 > ald.all) + return PGM_ALEN_TRANSLATION; + + if (0x7fffffff - ald.alo * 128 < alet.alen * 16) + return PGM_ADDRESSING; + + rc = read_guest_real(vcpu, ald.alo * 128 + alet.alen * 16, &ale, + sizeof(struct ale)); + if (rc) + return rc; + + if (ale.i == 1) + return PGM_ALEN_TRANSLATION; + if (ale.alesn != alet.alesn) + return PGM_ALE_SEQUENCE; + + rc = read_guest_real(vcpu, ale.asteo * 64, &aste, sizeof(struct aste)); + if (rc) + return rc; + + if (aste.i) + return PGM_ASTE_VALIDITY; + if (aste.astesn != ale.astesn) + return PGM_ASTE_SEQUENCE; + + if (ale.p == 1) { + eax = (vcpu->arch.sie_block->gcr[8] >> 16) & 0xffff; + if (ale.aleax != eax) { + if (eax / 16 > aste.atl) + return PGM_EXTENDED_AUTHORITY; + + authority_table_addr = aste.ato * 4 + eax / 4; + + rc = read_guest_real(vcpu, authority_table_addr, + &authority_table, + sizeof(u8)); + if (rc) + return rc; + + if ((authority_table & (0x40 >> ((eax & 3) * 2))) == 0) + return PGM_EXTENDED_AUTHORITY; + } + } + + if (ale.fo == 1 && write) + return PGM_PROTECTION; + + asce->val = aste.asce; + return 0; +} + +struct trans_exc_code_bits { + unsigned long addr : 52; /* Translation-exception Address */ + unsigned long fsi : 2; /* Access Exception Fetch/Store Indication */ + unsigned long : 6; + unsigned long b60 : 1; + unsigned long b61 : 1; + unsigned long as : 2; /* ASCE Identifier */ +}; + +enum { + FSI_UNKNOWN = 0, /* Unknown wether fetch or store */ + FSI_STORE = 1, /* Exception was due to store operation */ + FSI_FETCH = 2 /* Exception was due to fetch operation */ +}; + +static int get_vcpu_asce(struct kvm_vcpu *vcpu, union asce *asce, + ar_t ar, int write) +{ + int rc; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm; + struct trans_exc_code_bits *tec_bits; + + memset(pgm, 0, sizeof(*pgm)); + tec_bits = (struct trans_exc_code_bits *)&pgm->trans_exc_code; + tec_bits->fsi = write ? FSI_STORE : FSI_FETCH; + tec_bits->as = psw_bits(*psw).as; + + if (!psw_bits(*psw).t) { + asce->val = 0; + asce->r = 1; + return 0; + } + + switch (psw_bits(vcpu->arch.sie_block->gpsw).as) { + case PSW_AS_PRIMARY: + asce->val = vcpu->arch.sie_block->gcr[1]; + return 0; + case PSW_AS_SECONDARY: + asce->val = vcpu->arch.sie_block->gcr[7]; + return 0; + case PSW_AS_HOME: + asce->val = vcpu->arch.sie_block->gcr[13]; + return 0; + case PSW_AS_ACCREG: + rc = ar_translation(vcpu, asce, ar, write); + switch (rc) { + case PGM_ALEN_TRANSLATION: + case PGM_ALE_SEQUENCE: + case PGM_ASTE_VALIDITY: + case PGM_ASTE_SEQUENCE: + case PGM_EXTENDED_AUTHORITY: + vcpu->arch.pgm.exc_access_id = ar; + break; + case PGM_PROTECTION: + tec_bits->b60 = 1; + tec_bits->b61 = 1; + break; + } + if (rc > 0) + pgm->code = rc; + return rc; + } + return 0; +} + +static int deref_table(struct kvm *kvm, unsigned long gpa, unsigned long *val) +{ + return kvm_read_guest(kvm, gpa, val, sizeof(*val)); +} + +/** + * guest_translate - translate a guest virtual into a guest absolute address + * @vcpu: virtual cpu + * @gva: guest virtual address + * @gpa: points to where guest physical (absolute) address should be stored + * @asce: effective asce + * @write: indicates if access is a write access + * + * Translate a guest virtual address into a guest absolute address by means + * of dynamic address translation as specified by the architecture. + * If the resulting absolute address is not available in the configuration + * an addressing exception is indicated and @gpa will not be changed. + * + * Returns: - zero on success; @gpa contains the resulting absolute address + * - a negative value if guest access failed due to e.g. broken + * guest mapping + * - a positve value if an access exception happened. In this case + * the returned value is the program interruption code as defined + * by the architecture + */ +static unsigned long guest_translate(struct kvm_vcpu *vcpu, unsigned long gva, + unsigned long *gpa, const union asce asce, + int write) +{ + union vaddress vaddr = {.addr = gva}; + union raddress raddr = {.addr = gva}; + union page_table_entry pte; + int dat_protection = 0; + union ctlreg0 ctlreg0; + unsigned long ptr; + int edat1, edat2; + + ctlreg0.val = vcpu->arch.sie_block->gcr[0]; + edat1 = ctlreg0.edat && test_kvm_facility(vcpu->kvm, 8); + edat2 = edat1 && test_kvm_facility(vcpu->kvm, 78); + if (asce.r) + goto real_address; + ptr = asce.origin * 4096; + switch (asce.dt) { + case ASCE_TYPE_REGION1: + if (vaddr.rfx01 > asce.tl) + return PGM_REGION_FIRST_TRANS; + ptr += vaddr.rfx * 8; + break; + case ASCE_TYPE_REGION2: + if (vaddr.rfx) + return PGM_ASCE_TYPE; + if (vaddr.rsx01 > asce.tl) + return PGM_REGION_SECOND_TRANS; + ptr += vaddr.rsx * 8; + break; + case ASCE_TYPE_REGION3: + if (vaddr.rfx || vaddr.rsx) + return PGM_ASCE_TYPE; + if (vaddr.rtx01 > asce.tl) + return PGM_REGION_THIRD_TRANS; + ptr += vaddr.rtx * 8; + break; + case ASCE_TYPE_SEGMENT: + if (vaddr.rfx || vaddr.rsx || vaddr.rtx) + return PGM_ASCE_TYPE; + if (vaddr.sx01 > asce.tl) + return PGM_SEGMENT_TRANSLATION; + ptr += vaddr.sx * 8; + break; + } + switch (asce.dt) { + case ASCE_TYPE_REGION1: { + union region1_table_entry rfte; + + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &rfte.val)) + return -EFAULT; + if (rfte.i) + return PGM_REGION_FIRST_TRANS; + if (rfte.tt != TABLE_TYPE_REGION1) + return PGM_TRANSLATION_SPEC; + if (vaddr.rsx01 < rfte.tf || vaddr.rsx01 > rfte.tl) + return PGM_REGION_SECOND_TRANS; + if (edat1) + dat_protection |= rfte.p; + ptr = rfte.rto * 4096 + vaddr.rsx * 8; + } + /* fallthrough */ + case ASCE_TYPE_REGION2: { + union region2_table_entry rste; + + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &rste.val)) + return -EFAULT; + if (rste.i) + return PGM_REGION_SECOND_TRANS; + if (rste.tt != TABLE_TYPE_REGION2) + return PGM_TRANSLATION_SPEC; + if (vaddr.rtx01 < rste.tf || vaddr.rtx01 > rste.tl) + return PGM_REGION_THIRD_TRANS; + if (edat1) + dat_protection |= rste.p; + ptr = rste.rto * 4096 + vaddr.rtx * 8; + } + /* fallthrough */ + case ASCE_TYPE_REGION3: { + union region3_table_entry rtte; + + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &rtte.val)) + return -EFAULT; + if (rtte.i) + return PGM_REGION_THIRD_TRANS; + if (rtte.tt != TABLE_TYPE_REGION3) + return PGM_TRANSLATION_SPEC; + if (rtte.cr && asce.p && edat2) + return PGM_TRANSLATION_SPEC; + if (rtte.fc && edat2) { + dat_protection |= rtte.fc1.p; + raddr.rfaa = rtte.fc1.rfaa; + goto absolute_address; + } + if (vaddr.sx01 < rtte.fc0.tf) + return PGM_SEGMENT_TRANSLATION; + if (vaddr.sx01 > rtte.fc0.tl) + return PGM_SEGMENT_TRANSLATION; + if (edat1) + dat_protection |= rtte.fc0.p; + ptr = rtte.fc0.sto * 4096 + vaddr.sx * 8; + } + /* fallthrough */ + case ASCE_TYPE_SEGMENT: { + union segment_table_entry ste; + + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &ste.val)) + return -EFAULT; + if (ste.i) + return PGM_SEGMENT_TRANSLATION; + if (ste.tt != TABLE_TYPE_SEGMENT) + return PGM_TRANSLATION_SPEC; + if (ste.cs && asce.p) + return PGM_TRANSLATION_SPEC; + if (ste.fc && edat1) { + dat_protection |= ste.fc1.p; + raddr.sfaa = ste.fc1.sfaa; + goto absolute_address; + } + dat_protection |= ste.fc0.p; + ptr = ste.fc0.pto * 2048 + vaddr.px * 8; + } + } + if (kvm_is_error_gpa(vcpu->kvm, ptr)) + return PGM_ADDRESSING; + if (deref_table(vcpu->kvm, ptr, &pte.val)) + return -EFAULT; + if (pte.i) + return PGM_PAGE_TRANSLATION; + if (pte.z) + return PGM_TRANSLATION_SPEC; + if (pte.co && !edat1) + return PGM_TRANSLATION_SPEC; + dat_protection |= pte.p; + raddr.pfra = pte.pfra; +real_address: + raddr.addr = kvm_s390_real_to_abs(vcpu, raddr.addr); +absolute_address: + if (write && dat_protection) + return PGM_PROTECTION; + if (kvm_is_error_gpa(vcpu->kvm, raddr.addr)) + return PGM_ADDRESSING; + *gpa = raddr.addr; + return 0; +} + +static inline int is_low_address(unsigned long ga) +{ + /* Check for address ranges 0..511 and 4096..4607 */ + return (ga & ~0x11fful) == 0; +} + +static int low_address_protection_enabled(struct kvm_vcpu *vcpu, + const union asce asce) +{ + union ctlreg0 ctlreg0 = {.val = vcpu->arch.sie_block->gcr[0]}; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + + if (!ctlreg0.lap) + return 0; + if (psw_bits(*psw).t && asce.p) + return 0; + return 1; +} + +static int guest_page_range(struct kvm_vcpu *vcpu, unsigned long ga, + unsigned long *pages, unsigned long nr_pages, + const union asce asce, int write) +{ + struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + struct trans_exc_code_bits *tec_bits; + int lap_enabled, rc; + + tec_bits = (struct trans_exc_code_bits *)&pgm->trans_exc_code; + lap_enabled = low_address_protection_enabled(vcpu, asce); + while (nr_pages) { + ga = kvm_s390_logical_to_effective(vcpu, ga); + tec_bits->addr = ga >> PAGE_SHIFT; + if (write && lap_enabled && is_low_address(ga)) { + pgm->code = PGM_PROTECTION; + return pgm->code; + } + ga &= PAGE_MASK; + if (psw_bits(*psw).t) { + rc = guest_translate(vcpu, ga, pages, asce, write); + if (rc < 0) + return rc; + if (rc == PGM_PROTECTION) + tec_bits->b61 = 1; + if (rc) + pgm->code = rc; + } else { + *pages = kvm_s390_real_to_abs(vcpu, ga); + if (kvm_is_error_gpa(vcpu->kvm, *pages)) + pgm->code = PGM_ADDRESSING; + } + if (pgm->code) + return pgm->code; + ga += PAGE_SIZE; + pages++; + nr_pages--; + } + return 0; +} + +int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, + unsigned long len, int write) +{ + psw_t *psw = &vcpu->arch.sie_block->gpsw; + unsigned long _len, nr_pages, gpa, idx; + unsigned long pages_array[2]; + unsigned long *pages; + int need_ipte_lock; + union asce asce; + int rc; + + if (!len) + return 0; + rc = get_vcpu_asce(vcpu, &asce, ar, write); + if (rc) + return rc; + nr_pages = (((ga & ~PAGE_MASK) + len - 1) >> PAGE_SHIFT) + 1; + pages = pages_array; + if (nr_pages > ARRAY_SIZE(pages_array)) + pages = vmalloc(nr_pages * sizeof(unsigned long)); + if (!pages) + return -ENOMEM; + need_ipte_lock = psw_bits(*psw).t && !asce.r; + if (need_ipte_lock) + ipte_lock(vcpu); + rc = guest_page_range(vcpu, ga, pages, nr_pages, asce, write); + for (idx = 0; idx < nr_pages && !rc; idx++) { + gpa = *(pages + idx) + (ga & ~PAGE_MASK); + _len = min(PAGE_SIZE - (gpa & ~PAGE_MASK), len); + if (write) + rc = kvm_write_guest(vcpu->kvm, gpa, data, _len); + else + rc = kvm_read_guest(vcpu->kvm, gpa, data, _len); + len -= _len; + ga += _len; + data += _len; + } + if (need_ipte_lock) + ipte_unlock(vcpu); + if (nr_pages > ARRAY_SIZE(pages_array)) + vfree(pages); + return rc; +} + +int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, + void *data, unsigned long len, int write) +{ + unsigned long _len, gpa; + int rc = 0; + + while (len && !rc) { + gpa = kvm_s390_real_to_abs(vcpu, gra); + _len = min(PAGE_SIZE - (gpa & ~PAGE_MASK), len); + if (write) + rc = write_guest_abs(vcpu, gpa, data, _len); + else + rc = read_guest_abs(vcpu, gpa, data, _len); + len -= _len; + gra += _len; + data += _len; + } + return rc; +} + +/** + * guest_translate_address - translate guest logical into guest absolute address + * + * Parameter semantics are the same as the ones from guest_translate. + * The memory contents at the guest address are not changed. + * + * Note: The IPTE lock is not taken during this function, so the caller + * has to take care of this. + */ +int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar, + unsigned long *gpa, int write) +{ + struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + struct trans_exc_code_bits *tec; + union asce asce; + int rc; + + gva = kvm_s390_logical_to_effective(vcpu, gva); + tec = (struct trans_exc_code_bits *)&pgm->trans_exc_code; + rc = get_vcpu_asce(vcpu, &asce, ar, write); + tec->addr = gva >> PAGE_SHIFT; + if (rc) + return rc; + if (is_low_address(gva) && low_address_protection_enabled(vcpu, asce)) { + if (write) { + rc = pgm->code = PGM_PROTECTION; + return rc; + } + } + + if (psw_bits(*psw).t && !asce.r) { /* Use DAT? */ + rc = guest_translate(vcpu, gva, gpa, asce, write); + if (rc > 0) { + if (rc == PGM_PROTECTION) + tec->b61 = 1; + pgm->code = rc; + } + } else { + rc = 0; + *gpa = kvm_s390_real_to_abs(vcpu, gva); + if (kvm_is_error_gpa(vcpu->kvm, *gpa)) + rc = pgm->code = PGM_ADDRESSING; + } + + return rc; +} + +/** + * check_gva_range - test a range of guest virtual addresses for accessibility + */ +int check_gva_range(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar, + unsigned long length, int is_write) +{ + unsigned long gpa; + unsigned long currlen; + int rc = 0; + + ipte_lock(vcpu); + while (length > 0 && !rc) { + currlen = min(length, PAGE_SIZE - (gva % PAGE_SIZE)); + rc = guest_translate_address(vcpu, gva, ar, &gpa, is_write); + gva += currlen; + length -= currlen; + } + ipte_unlock(vcpu); + + return rc; +} + +/** + * kvm_s390_check_low_addr_prot_real - check for low-address protection + * @gra: Guest real address + * + * Checks whether an address is subject to low-address protection and set + * up vcpu->arch.pgm accordingly if necessary. + * + * Return: 0 if no protection exception, or PGM_PROTECTION if protected. + */ +int kvm_s390_check_low_addr_prot_real(struct kvm_vcpu *vcpu, unsigned long gra) +{ + struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm; + psw_t *psw = &vcpu->arch.sie_block->gpsw; + struct trans_exc_code_bits *tec_bits; + union ctlreg0 ctlreg0 = {.val = vcpu->arch.sie_block->gcr[0]}; + + if (!ctlreg0.lap || !is_low_address(gra)) + return 0; + + memset(pgm, 0, sizeof(*pgm)); + tec_bits = (struct trans_exc_code_bits *)&pgm->trans_exc_code; + tec_bits->fsi = FSI_STORE; + tec_bits->as = psw_bits(*psw).as; + tec_bits->addr = gra >> PAGE_SHIFT; + pgm->code = PGM_PROTECTION; + + return pgm->code; +} diff --git a/arch/s390/kvm/gaccess.h b/arch/s390/kvm/gaccess.h new file mode 100644 index 000000000..ef03726cc --- /dev/null +++ b/arch/s390/kvm/gaccess.h @@ -0,0 +1,338 @@ +/* + * access guest memory + * + * Copyright IBM Corp. 2008, 2014 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + */ + +#ifndef __KVM_S390_GACCESS_H +#define __KVM_S390_GACCESS_H + +#include <linux/compiler.h> +#include <linux/kvm_host.h> +#include <linux/uaccess.h> +#include <linux/ptrace.h> +#include "kvm-s390.h" + +/** + * kvm_s390_real_to_abs - convert guest real address to guest absolute address + * @vcpu - guest virtual cpu + * @gra - guest real address + * + * Returns the guest absolute address that corresponds to the passed guest real + * address @gra of a virtual guest cpu by applying its prefix. + */ +static inline unsigned long kvm_s390_real_to_abs(struct kvm_vcpu *vcpu, + unsigned long gra) +{ + unsigned long prefix = kvm_s390_get_prefix(vcpu); + + if (gra < 2 * PAGE_SIZE) + gra += prefix; + else if (gra >= prefix && gra < prefix + 2 * PAGE_SIZE) + gra -= prefix; + return gra; +} + +/** + * kvm_s390_logical_to_effective - convert guest logical to effective address + * @vcpu: guest virtual cpu + * @ga: guest logical address + * + * Convert a guest vcpu logical address to a guest vcpu effective address by + * applying the rules of the vcpu's addressing mode defined by PSW bits 31 + * and 32 (extendended/basic addressing mode). + * + * Depending on the vcpu's addressing mode the upper 40 bits (24 bit addressing + * mode), 33 bits (31 bit addressing mode) or no bits (64 bit addressing mode) + * of @ga will be zeroed and the remaining bits will be returned. + */ +static inline unsigned long kvm_s390_logical_to_effective(struct kvm_vcpu *vcpu, + unsigned long ga) +{ + psw_t *psw = &vcpu->arch.sie_block->gpsw; + + if (psw_bits(*psw).eaba == PSW_AMODE_64BIT) + return ga; + if (psw_bits(*psw).eaba == PSW_AMODE_31BIT) + return ga & ((1UL << 31) - 1); + return ga & ((1UL << 24) - 1); +} + +/* + * put_guest_lc, read_guest_lc and write_guest_lc are guest access functions + * which shall only be used to access the lowcore of a vcpu. + * These functions should be used for e.g. interrupt handlers where no + * guest memory access protection facilities, like key or low address + * protection, are applicable. + * At a later point guest vcpu lowcore access should happen via pinned + * prefix pages, so that these pages can be accessed directly via the + * kernel mapping. All of these *_lc functions can be removed then. + */ + +/** + * put_guest_lc - write a simple variable to a guest vcpu's lowcore + * @vcpu: virtual cpu + * @x: value to copy to guest + * @gra: vcpu's destination guest real address + * + * Copies a simple value from kernel space to a guest vcpu's lowcore. + * The size of the variable may be 1, 2, 4 or 8 bytes. The destination + * must be located in the vcpu's lowcore. Otherwise the result is undefined. + * + * Returns zero on success or -EFAULT on error. + * + * Note: an error indicates that either the kernel is out of memory or + * the guest memory mapping is broken. In any case the best solution + * would be to terminate the guest. + * It is wrong to inject a guest exception. + */ +#define put_guest_lc(vcpu, x, gra) \ +({ \ + struct kvm_vcpu *__vcpu = (vcpu); \ + __typeof__(*(gra)) __x = (x); \ + unsigned long __gpa; \ + \ + __gpa = (unsigned long)(gra); \ + __gpa += kvm_s390_get_prefix(__vcpu); \ + kvm_write_guest(__vcpu->kvm, __gpa, &__x, sizeof(__x)); \ +}) + +/** + * write_guest_lc - copy data from kernel space to guest vcpu's lowcore + * @vcpu: virtual cpu + * @gra: vcpu's source guest real address + * @data: source address in kernel space + * @len: number of bytes to copy + * + * Copy data from kernel space to guest vcpu's lowcore. The entire range must + * be located within the vcpu's lowcore, otherwise the result is undefined. + * + * Returns zero on success or -EFAULT on error. + * + * Note: an error indicates that either the kernel is out of memory or + * the guest memory mapping is broken. In any case the best solution + * would be to terminate the guest. + * It is wrong to inject a guest exception. + */ +static inline __must_check +int write_guest_lc(struct kvm_vcpu *vcpu, unsigned long gra, void *data, + unsigned long len) +{ + unsigned long gpa = gra + kvm_s390_get_prefix(vcpu); + + return kvm_write_guest(vcpu->kvm, gpa, data, len); +} + +/** + * read_guest_lc - copy data from guest vcpu's lowcore to kernel space + * @vcpu: virtual cpu + * @gra: vcpu's source guest real address + * @data: destination address in kernel space + * @len: number of bytes to copy + * + * Copy data from guest vcpu's lowcore to kernel space. The entire range must + * be located within the vcpu's lowcore, otherwise the result is undefined. + * + * Returns zero on success or -EFAULT on error. + * + * Note: an error indicates that either the kernel is out of memory or + * the guest memory mapping is broken. In any case the best solution + * would be to terminate the guest. + * It is wrong to inject a guest exception. + */ +static inline __must_check +int read_guest_lc(struct kvm_vcpu *vcpu, unsigned long gra, void *data, + unsigned long len) +{ + unsigned long gpa = gra + kvm_s390_get_prefix(vcpu); + + return kvm_read_guest(vcpu->kvm, gpa, data, len); +} + +int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva, + ar_t ar, unsigned long *gpa, int write); +int check_gva_range(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar, + unsigned long length, int is_write); + +int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, + unsigned long len, int write); + +int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, + void *data, unsigned long len, int write); + +/** + * write_guest - copy data from kernel space to guest space + * @vcpu: virtual cpu + * @ga: guest address + * @ar: access register + * @data: source address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @data (kernel space) to @ga (guest address). + * In order to copy data to guest space the PSW of the vcpu is inspected: + * If DAT is off data will be copied to guest real or absolute memory. + * If DAT is on data will be copied to the address space as specified by + * the address space bits of the PSW: + * Primary, secondary, home space or access register mode. + * The addressing mode of the PSW is also inspected, so that address wrap + * around is taken into account for 24-, 31- and 64-bit addressing mode, + * if the to be copied data crosses page boundaries in guest address space. + * In addition also low address and DAT protection are inspected before + * copying any data (key protection is currently not implemented). + * + * This function modifies the 'struct kvm_s390_pgm_info pgm' member of @vcpu. + * In case of an access exception (e.g. protection exception) pgm will contain + * all data necessary so that a subsequent call to 'kvm_s390_inject_prog_vcpu()' + * will inject a correct exception into the guest. + * If no access exception happened, the contents of pgm are undefined when + * this function returns. + * + * Returns: - zero on success + * - a negative value if e.g. the guest mapping is broken or in + * case of out-of-memory. In this case the contents of pgm are + * undefined. Also parts of @data may have been copied to guest + * space. + * - a positive value if an access exception happened. In this case + * the returned value is the program interruption code and the + * contents of pgm may be used to inject an exception into the + * guest. No data has been copied to guest space. + * + * Note: in case an access exception is recognized no data has been copied to + * guest space (this is also true, if the to be copied data would cross + * one or more page boundaries in guest space). + * Therefore this function may be used for nullifying and suppressing + * instruction emulation. + * It may also be used for terminating instructions, if it is undefined + * if data has been changed in guest space in case of an exception. + */ +static inline __must_check +int write_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, + unsigned long len) +{ + return access_guest(vcpu, ga, ar, data, len, 1); +} + +/** + * read_guest - copy data from guest space to kernel space + * @vcpu: virtual cpu + * @ga: guest address + * @ar: access register + * @data: destination address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @ga (guest address) to @data (kernel space). + * + * The behaviour of read_guest is identical to write_guest, except that + * data will be copied from guest space to kernel space. + */ +static inline __must_check +int read_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, + unsigned long len) +{ + return access_guest(vcpu, ga, ar, data, len, 0); +} + +/** + * write_guest_abs - copy data from kernel space to guest space absolute + * @vcpu: virtual cpu + * @gpa: guest physical (absolute) address + * @data: source address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @data (kernel space) to @gpa (guest absolute address). + * It is up to the caller to ensure that the entire guest memory range is + * valid memory before calling this function. + * Guest low address and key protection are not checked. + * + * Returns zero on success or -EFAULT on error. + * + * If an error occurs data may have been copied partially to guest memory. + */ +static inline __must_check +int write_guest_abs(struct kvm_vcpu *vcpu, unsigned long gpa, void *data, + unsigned long len) +{ + return kvm_write_guest(vcpu->kvm, gpa, data, len); +} + +/** + * read_guest_abs - copy data from guest space absolute to kernel space + * @vcpu: virtual cpu + * @gpa: guest physical (absolute) address + * @data: destination address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @gpa (guest absolute address) to @data (kernel space). + * It is up to the caller to ensure that the entire guest memory range is + * valid memory before calling this function. + * Guest key protection is not checked. + * + * Returns zero on success or -EFAULT on error. + * + * If an error occurs data may have been copied partially to kernel space. + */ +static inline __must_check +int read_guest_abs(struct kvm_vcpu *vcpu, unsigned long gpa, void *data, + unsigned long len) +{ + return kvm_read_guest(vcpu->kvm, gpa, data, len); +} + +/** + * write_guest_real - copy data from kernel space to guest space real + * @vcpu: virtual cpu + * @gra: guest real address + * @data: source address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @data (kernel space) to @gra (guest real address). + * It is up to the caller to ensure that the entire guest memory range is + * valid memory before calling this function. + * Guest low address and key protection are not checked. + * + * Returns zero on success or -EFAULT on error. + * + * If an error occurs data may have been copied partially to guest memory. + */ +static inline __must_check +int write_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, void *data, + unsigned long len) +{ + return access_guest_real(vcpu, gra, data, len, 1); +} + +/** + * read_guest_real - copy data from guest space real to kernel space + * @vcpu: virtual cpu + * @gra: guest real address + * @data: destination address in kernel space + * @len: number of bytes to copy + * + * Copy @len bytes from @gra (guest real address) to @data (kernel space). + * It is up to the caller to ensure that the entire guest memory range is + * valid memory before calling this function. + * Guest key protection is not checked. + * + * Returns zero on success or -EFAULT on error. + * + * If an error occurs data may have been copied partially to kernel space. + */ +static inline __must_check +int read_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, void *data, + unsigned long len) +{ + return access_guest_real(vcpu, gra, data, len, 0); +} + +void ipte_lock(struct kvm_vcpu *vcpu); +void ipte_unlock(struct kvm_vcpu *vcpu); +int ipte_lock_held(struct kvm_vcpu *vcpu); +int kvm_s390_check_low_addr_prot_real(struct kvm_vcpu *vcpu, unsigned long gra); + +#endif /* __KVM_S390_GACCESS_H */ diff --git a/arch/s390/kvm/guestdbg.c b/arch/s390/kvm/guestdbg.c new file mode 100644 index 000000000..e97b3455d --- /dev/null +++ b/arch/s390/kvm/guestdbg.c @@ -0,0 +1,482 @@ +/* + * kvm guest debug support + * + * Copyright IBM Corp. 2014 + * + * 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. + * + * Author(s): David Hildenbrand <dahi@linux.vnet.ibm.com> + */ +#include <linux/kvm_host.h> +#include <linux/errno.h> +#include "kvm-s390.h" +#include "gaccess.h" + +/* + * Extends the address range given by *start and *stop to include the address + * range starting with estart and the length len. Takes care of overflowing + * intervals and tries to minimize the overall intervall size. + */ +static void extend_address_range(u64 *start, u64 *stop, u64 estart, int len) +{ + u64 estop; + + if (len > 0) + len--; + else + len = 0; + + estop = estart + len; + + /* 0-0 range represents "not set" */ + if ((*start == 0) && (*stop == 0)) { + *start = estart; + *stop = estop; + } else if (*start <= *stop) { + /* increase the existing range */ + if (estart < *start) + *start = estart; + if (estop > *stop) + *stop = estop; + } else { + /* "overflowing" interval, whereby *stop > *start */ + if (estart <= *stop) { + if (estop > *stop) + *stop = estop; + } else if (estop > *start) { + if (estart < *start) + *start = estart; + } + /* minimize the range */ + else if ((estop - *stop) < (*start - estart)) + *stop = estop; + else + *start = estart; + } +} + +#define MAX_INST_SIZE 6 + +static void enable_all_hw_bp(struct kvm_vcpu *vcpu) +{ + unsigned long start, len; + u64 *cr9 = &vcpu->arch.sie_block->gcr[9]; + u64 *cr10 = &vcpu->arch.sie_block->gcr[10]; + u64 *cr11 = &vcpu->arch.sie_block->gcr[11]; + int i; + + if (vcpu->arch.guestdbg.nr_hw_bp <= 0 || + vcpu->arch.guestdbg.hw_bp_info == NULL) + return; + + /* + * If the guest is not interrested in branching events, we can savely + * limit them to the PER address range. + */ + if (!(*cr9 & PER_EVENT_BRANCH)) + *cr9 |= PER_CONTROL_BRANCH_ADDRESS; + *cr9 |= PER_EVENT_IFETCH | PER_EVENT_BRANCH; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_bp; i++) { + start = vcpu->arch.guestdbg.hw_bp_info[i].addr; + len = vcpu->arch.guestdbg.hw_bp_info[i].len; + + /* + * The instruction in front of the desired bp has to + * report instruction-fetching events + */ + if (start < MAX_INST_SIZE) { + len += start; + start = 0; + } else { + start -= MAX_INST_SIZE; + len += MAX_INST_SIZE; + } + + extend_address_range(cr10, cr11, start, len); + } +} + +static void enable_all_hw_wp(struct kvm_vcpu *vcpu) +{ + unsigned long start, len; + u64 *cr9 = &vcpu->arch.sie_block->gcr[9]; + u64 *cr10 = &vcpu->arch.sie_block->gcr[10]; + u64 *cr11 = &vcpu->arch.sie_block->gcr[11]; + int i; + + if (vcpu->arch.guestdbg.nr_hw_wp <= 0 || + vcpu->arch.guestdbg.hw_wp_info == NULL) + return; + + /* if host uses storage alternation for special address + * spaces, enable all events and give all to the guest */ + if (*cr9 & PER_EVENT_STORE && *cr9 & PER_CONTROL_ALTERATION) { + *cr9 &= ~PER_CONTROL_ALTERATION; + *cr10 = 0; + *cr11 = PSW_ADDR_INSN; + } else { + *cr9 &= ~PER_CONTROL_ALTERATION; + *cr9 |= PER_EVENT_STORE; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_wp; i++) { + start = vcpu->arch.guestdbg.hw_wp_info[i].addr; + len = vcpu->arch.guestdbg.hw_wp_info[i].len; + + extend_address_range(cr10, cr11, start, len); + } + } +} + +void kvm_s390_backup_guest_per_regs(struct kvm_vcpu *vcpu) +{ + vcpu->arch.guestdbg.cr0 = vcpu->arch.sie_block->gcr[0]; + vcpu->arch.guestdbg.cr9 = vcpu->arch.sie_block->gcr[9]; + vcpu->arch.guestdbg.cr10 = vcpu->arch.sie_block->gcr[10]; + vcpu->arch.guestdbg.cr11 = vcpu->arch.sie_block->gcr[11]; +} + +void kvm_s390_restore_guest_per_regs(struct kvm_vcpu *vcpu) +{ + vcpu->arch.sie_block->gcr[0] = vcpu->arch.guestdbg.cr0; + vcpu->arch.sie_block->gcr[9] = vcpu->arch.guestdbg.cr9; + vcpu->arch.sie_block->gcr[10] = vcpu->arch.guestdbg.cr10; + vcpu->arch.sie_block->gcr[11] = vcpu->arch.guestdbg.cr11; +} + +void kvm_s390_patch_guest_per_regs(struct kvm_vcpu *vcpu) +{ + /* + * TODO: if guest psw has per enabled, otherwise 0s! + * This reduces the amount of reported events. + * Need to intercept all psw changes! + */ + + if (guestdbg_sstep_enabled(vcpu)) { + /* disable timer (clock-comparator) interrupts */ + vcpu->arch.sie_block->gcr[0] &= ~0x800ul; + vcpu->arch.sie_block->gcr[9] |= PER_EVENT_IFETCH; + vcpu->arch.sie_block->gcr[10] = 0; + vcpu->arch.sie_block->gcr[11] = PSW_ADDR_INSN; + } + + if (guestdbg_hw_bp_enabled(vcpu)) { + enable_all_hw_bp(vcpu); + enable_all_hw_wp(vcpu); + } + + /* TODO: Instruction-fetching-nullification not allowed for now */ + if (vcpu->arch.sie_block->gcr[9] & PER_EVENT_NULLIFICATION) + vcpu->arch.sie_block->gcr[9] &= ~PER_EVENT_NULLIFICATION; +} + +#define MAX_WP_SIZE 100 + +static int __import_wp_info(struct kvm_vcpu *vcpu, + struct kvm_hw_breakpoint *bp_data, + struct kvm_hw_wp_info_arch *wp_info) +{ + int ret = 0; + wp_info->len = bp_data->len; + wp_info->addr = bp_data->addr; + wp_info->phys_addr = bp_data->phys_addr; + wp_info->old_data = NULL; + + if (wp_info->len < 0 || wp_info->len > MAX_WP_SIZE) + return -EINVAL; + + wp_info->old_data = kmalloc(bp_data->len, GFP_KERNEL); + if (!wp_info->old_data) + return -ENOMEM; + /* try to backup the original value */ + ret = read_guest_abs(vcpu, wp_info->phys_addr, wp_info->old_data, + wp_info->len); + if (ret) { + kfree(wp_info->old_data); + wp_info->old_data = NULL; + } + + return ret; +} + +#define MAX_BP_COUNT 50 + +int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu, + struct kvm_guest_debug *dbg) +{ + int ret = 0, nr_wp = 0, nr_bp = 0, i, size; + struct kvm_hw_breakpoint *bp_data = NULL; + struct kvm_hw_wp_info_arch *wp_info = NULL; + struct kvm_hw_bp_info_arch *bp_info = NULL; + + if (dbg->arch.nr_hw_bp <= 0 || !dbg->arch.hw_bp) + return 0; + else if (dbg->arch.nr_hw_bp > MAX_BP_COUNT) + return -EINVAL; + + size = dbg->arch.nr_hw_bp * sizeof(struct kvm_hw_breakpoint); + bp_data = kmalloc(size, GFP_KERNEL); + if (!bp_data) { + ret = -ENOMEM; + goto error; + } + + if (copy_from_user(bp_data, dbg->arch.hw_bp, size)) { + ret = -EFAULT; + goto error; + } + + for (i = 0; i < dbg->arch.nr_hw_bp; i++) { + switch (bp_data[i].type) { + case KVM_HW_WP_WRITE: + nr_wp++; + break; + case KVM_HW_BP: + nr_bp++; + break; + default: + break; + } + } + + size = nr_wp * sizeof(struct kvm_hw_wp_info_arch); + if (size > 0) { + wp_info = kmalloc(size, GFP_KERNEL); + if (!wp_info) { + ret = -ENOMEM; + goto error; + } + } + size = nr_bp * sizeof(struct kvm_hw_bp_info_arch); + if (size > 0) { + bp_info = kmalloc(size, GFP_KERNEL); + if (!bp_info) { + ret = -ENOMEM; + goto error; + } + } + + for (nr_wp = 0, nr_bp = 0, i = 0; i < dbg->arch.nr_hw_bp; i++) { + switch (bp_data[i].type) { + case KVM_HW_WP_WRITE: + ret = __import_wp_info(vcpu, &bp_data[i], + &wp_info[nr_wp]); + if (ret) + goto error; + nr_wp++; + break; + case KVM_HW_BP: + bp_info[nr_bp].len = bp_data[i].len; + bp_info[nr_bp].addr = bp_data[i].addr; + nr_bp++; + break; + } + } + + vcpu->arch.guestdbg.nr_hw_bp = nr_bp; + vcpu->arch.guestdbg.hw_bp_info = bp_info; + vcpu->arch.guestdbg.nr_hw_wp = nr_wp; + vcpu->arch.guestdbg.hw_wp_info = wp_info; + return 0; +error: + kfree(bp_data); + kfree(wp_info); + kfree(bp_info); + return ret; +} + +void kvm_s390_clear_bp_data(struct kvm_vcpu *vcpu) +{ + int i; + struct kvm_hw_wp_info_arch *hw_wp_info = NULL; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_wp; i++) { + hw_wp_info = &vcpu->arch.guestdbg.hw_wp_info[i]; + kfree(hw_wp_info->old_data); + hw_wp_info->old_data = NULL; + } + kfree(vcpu->arch.guestdbg.hw_wp_info); + vcpu->arch.guestdbg.hw_wp_info = NULL; + + kfree(vcpu->arch.guestdbg.hw_bp_info); + vcpu->arch.guestdbg.hw_bp_info = NULL; + + vcpu->arch.guestdbg.nr_hw_wp = 0; + vcpu->arch.guestdbg.nr_hw_bp = 0; +} + +static inline int in_addr_range(u64 addr, u64 a, u64 b) +{ + if (a <= b) + return (addr >= a) && (addr <= b); + else + /* "overflowing" interval */ + return (addr <= a) && (addr >= b); +} + +#define end_of_range(bp_info) (bp_info->addr + bp_info->len - 1) + +static struct kvm_hw_bp_info_arch *find_hw_bp(struct kvm_vcpu *vcpu, + unsigned long addr) +{ + struct kvm_hw_bp_info_arch *bp_info = vcpu->arch.guestdbg.hw_bp_info; + int i; + + if (vcpu->arch.guestdbg.nr_hw_bp == 0) + return NULL; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_bp; i++) { + /* addr is directly the start or in the range of a bp */ + if (addr == bp_info->addr) + goto found; + if (bp_info->len > 0 && + in_addr_range(addr, bp_info->addr, end_of_range(bp_info))) + goto found; + + bp_info++; + } + + return NULL; +found: + return bp_info; +} + +static struct kvm_hw_wp_info_arch *any_wp_changed(struct kvm_vcpu *vcpu) +{ + int i; + struct kvm_hw_wp_info_arch *wp_info = NULL; + void *temp = NULL; + + if (vcpu->arch.guestdbg.nr_hw_wp == 0) + return NULL; + + for (i = 0; i < vcpu->arch.guestdbg.nr_hw_wp; i++) { + wp_info = &vcpu->arch.guestdbg.hw_wp_info[i]; + if (!wp_info || !wp_info->old_data || wp_info->len <= 0) + continue; + + temp = kmalloc(wp_info->len, GFP_KERNEL); + if (!temp) + continue; + + /* refetch the wp data and compare it to the old value */ + if (!read_guest_abs(vcpu, wp_info->phys_addr, temp, + wp_info->len)) { + if (memcmp(temp, wp_info->old_data, wp_info->len)) { + kfree(temp); + return wp_info; + } + } + kfree(temp); + temp = NULL; + } + + return NULL; +} + +void kvm_s390_prepare_debug_exit(struct kvm_vcpu *vcpu) +{ + vcpu->run->exit_reason = KVM_EXIT_DEBUG; + vcpu->guest_debug &= ~KVM_GUESTDBG_EXIT_PENDING; +} + +#define per_bp_event(code) \ + (code & (PER_EVENT_IFETCH | PER_EVENT_BRANCH)) +#define per_write_wp_event(code) \ + (code & (PER_EVENT_STORE | PER_EVENT_STORE_REAL)) + +static int debug_exit_required(struct kvm_vcpu *vcpu) +{ + u32 perc = (vcpu->arch.sie_block->perc << 24); + struct kvm_debug_exit_arch *debug_exit = &vcpu->run->debug.arch; + struct kvm_hw_wp_info_arch *wp_info = NULL; + struct kvm_hw_bp_info_arch *bp_info = NULL; + unsigned long addr = vcpu->arch.sie_block->gpsw.addr; + unsigned long peraddr = vcpu->arch.sie_block->peraddr; + + if (guestdbg_hw_bp_enabled(vcpu)) { + if (per_write_wp_event(perc) && + vcpu->arch.guestdbg.nr_hw_wp > 0) { + wp_info = any_wp_changed(vcpu); + if (wp_info) { + debug_exit->addr = wp_info->addr; + debug_exit->type = KVM_HW_WP_WRITE; + goto exit_required; + } + } + if (per_bp_event(perc) && + vcpu->arch.guestdbg.nr_hw_bp > 0) { + bp_info = find_hw_bp(vcpu, addr); + /* remove duplicate events if PC==PER address */ + if (bp_info && (addr != peraddr)) { + debug_exit->addr = addr; + debug_exit->type = KVM_HW_BP; + vcpu->arch.guestdbg.last_bp = addr; + goto exit_required; + } + /* breakpoint missed */ + bp_info = find_hw_bp(vcpu, peraddr); + if (bp_info && vcpu->arch.guestdbg.last_bp != peraddr) { + debug_exit->addr = peraddr; + debug_exit->type = KVM_HW_BP; + goto exit_required; + } + } + } + if (guestdbg_sstep_enabled(vcpu) && per_bp_event(perc)) { + debug_exit->addr = addr; + debug_exit->type = KVM_SINGLESTEP; + goto exit_required; + } + + return 0; +exit_required: + return 1; +} + +#define guest_per_enabled(vcpu) \ + (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PER) + +static void filter_guest_per_event(struct kvm_vcpu *vcpu) +{ + u32 perc = vcpu->arch.sie_block->perc << 24; + u64 peraddr = vcpu->arch.sie_block->peraddr; + u64 addr = vcpu->arch.sie_block->gpsw.addr; + u64 cr9 = vcpu->arch.sie_block->gcr[9]; + u64 cr10 = vcpu->arch.sie_block->gcr[10]; + u64 cr11 = vcpu->arch.sie_block->gcr[11]; + /* filter all events, demanded by the guest */ + u32 guest_perc = perc & cr9 & PER_EVENT_MASK; + + if (!guest_per_enabled(vcpu)) + guest_perc = 0; + + /* filter "successful-branching" events */ + if (guest_perc & PER_EVENT_BRANCH && + cr9 & PER_CONTROL_BRANCH_ADDRESS && + !in_addr_range(addr, cr10, cr11)) + guest_perc &= ~PER_EVENT_BRANCH; + + /* filter "instruction-fetching" events */ + if (guest_perc & PER_EVENT_IFETCH && + !in_addr_range(peraddr, cr10, cr11)) + guest_perc &= ~PER_EVENT_IFETCH; + + /* All other PER events will be given to the guest */ + /* TODO: Check alterated address/address space */ + + vcpu->arch.sie_block->perc = guest_perc >> 24; + + if (!guest_perc) + vcpu->arch.sie_block->iprcc &= ~PGM_PER; +} + +void kvm_s390_handle_per_event(struct kvm_vcpu *vcpu) +{ + if (debug_exit_required(vcpu)) + vcpu->guest_debug |= KVM_GUESTDBG_EXIT_PENDING; + + filter_guest_per_event(vcpu); +} diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c new file mode 100644 index 000000000..9e3779e3e --- /dev/null +++ b/arch/s390/kvm/intercept.c @@ -0,0 +1,379 @@ +/* + * in-kernel handling for sie intercepts + * + * Copyright IBM Corp. 2008, 2014 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + * Christian Borntraeger <borntraeger@de.ibm.com> + */ + +#include <linux/kvm_host.h> +#include <linux/errno.h> +#include <linux/pagemap.h> + +#include <asm/kvm_host.h> +#include <asm/asm-offsets.h> +#include <asm/irq.h> + +#include "kvm-s390.h" +#include "gaccess.h" +#include "trace.h" +#include "trace-s390.h" + + +static const intercept_handler_t instruction_handlers[256] = { + [0x01] = kvm_s390_handle_01, + [0x82] = kvm_s390_handle_lpsw, + [0x83] = kvm_s390_handle_diag, + [0xae] = kvm_s390_handle_sigp, + [0xb2] = kvm_s390_handle_b2, + [0xb6] = kvm_s390_handle_stctl, + [0xb7] = kvm_s390_handle_lctl, + [0xb9] = kvm_s390_handle_b9, + [0xe5] = kvm_s390_handle_e5, + [0xeb] = kvm_s390_handle_eb, +}; + +void kvm_s390_rewind_psw(struct kvm_vcpu *vcpu, int ilc) +{ + struct kvm_s390_sie_block *sie_block = vcpu->arch.sie_block; + + /* Use the length of the EXECUTE instruction if necessary */ + if (sie_block->icptstatus & 1) { + ilc = (sie_block->icptstatus >> 4) & 0x6; + if (!ilc) + ilc = 4; + } + sie_block->gpsw.addr = __rewind_psw(sie_block->gpsw, ilc); +} + +static int handle_noop(struct kvm_vcpu *vcpu) +{ + switch (vcpu->arch.sie_block->icptcode) { + case 0x0: + vcpu->stat.exit_null++; + break; + case 0x10: + vcpu->stat.exit_external_request++; + break; + default: + break; /* nothing */ + } + return 0; +} + +static int handle_stop(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + int rc = 0; + uint8_t flags, stop_pending; + + vcpu->stat.exit_stop_request++; + + /* delay the stop if any non-stop irq is pending */ + if (kvm_s390_vcpu_has_irq(vcpu, 1)) + return 0; + + /* avoid races with the injection/SIGP STOP code */ + spin_lock(&li->lock); + flags = li->irq.stop.flags; + stop_pending = kvm_s390_is_stop_irq_pending(vcpu); + spin_unlock(&li->lock); + + trace_kvm_s390_stop_request(stop_pending, flags); + if (!stop_pending) + return 0; + + if (flags & KVM_S390_STOP_FLAG_STORE_STATUS) { + rc = kvm_s390_vcpu_store_status(vcpu, + KVM_S390_STORE_STATUS_NOADDR); + if (rc) + return rc; + } + + if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) + kvm_s390_vcpu_stop(vcpu); + return -EOPNOTSUPP; +} + +static int handle_validity(struct kvm_vcpu *vcpu) +{ + int viwhy = vcpu->arch.sie_block->ipb >> 16; + + vcpu->stat.exit_validity++; + trace_kvm_s390_intercept_validity(vcpu, viwhy); + WARN_ONCE(true, "kvm: unhandled validity intercept 0x%x\n", viwhy); + return -EOPNOTSUPP; +} + +static int handle_instruction(struct kvm_vcpu *vcpu) +{ + intercept_handler_t handler; + + vcpu->stat.exit_instruction++; + trace_kvm_s390_intercept_instruction(vcpu, + vcpu->arch.sie_block->ipa, + vcpu->arch.sie_block->ipb); + handler = instruction_handlers[vcpu->arch.sie_block->ipa >> 8]; + if (handler) + return handler(vcpu); + return -EOPNOTSUPP; +} + +static void __extract_prog_irq(struct kvm_vcpu *vcpu, + struct kvm_s390_pgm_info *pgm_info) +{ + memset(pgm_info, 0, sizeof(struct kvm_s390_pgm_info)); + pgm_info->code = vcpu->arch.sie_block->iprcc; + + switch (vcpu->arch.sie_block->iprcc & ~PGM_PER) { + case PGM_AFX_TRANSLATION: + case PGM_ASX_TRANSLATION: + case PGM_EX_TRANSLATION: + case PGM_LFX_TRANSLATION: + case PGM_LSTE_SEQUENCE: + case PGM_LSX_TRANSLATION: + case PGM_LX_TRANSLATION: + case PGM_PRIMARY_AUTHORITY: + case PGM_SECONDARY_AUTHORITY: + case PGM_SPACE_SWITCH: + pgm_info->trans_exc_code = vcpu->arch.sie_block->tecmc; + break; + case PGM_ALEN_TRANSLATION: + case PGM_ALE_SEQUENCE: + case PGM_ASTE_INSTANCE: + case PGM_ASTE_SEQUENCE: + case PGM_ASTE_VALIDITY: + case PGM_EXTENDED_AUTHORITY: + pgm_info->exc_access_id = vcpu->arch.sie_block->eai; + break; + case PGM_ASCE_TYPE: + case PGM_PAGE_TRANSLATION: + case PGM_REGION_FIRST_TRANS: + case PGM_REGION_SECOND_TRANS: + case PGM_REGION_THIRD_TRANS: + case PGM_SEGMENT_TRANSLATION: + pgm_info->trans_exc_code = vcpu->arch.sie_block->tecmc; + pgm_info->exc_access_id = vcpu->arch.sie_block->eai; + pgm_info->op_access_id = vcpu->arch.sie_block->oai; + break; + case PGM_MONITOR: + pgm_info->mon_class_nr = vcpu->arch.sie_block->mcn; + pgm_info->mon_code = vcpu->arch.sie_block->tecmc; + break; + case PGM_VECTOR_PROCESSING: + case PGM_DATA: + pgm_info->data_exc_code = vcpu->arch.sie_block->dxc; + break; + case PGM_PROTECTION: + pgm_info->trans_exc_code = vcpu->arch.sie_block->tecmc; + pgm_info->exc_access_id = vcpu->arch.sie_block->eai; + break; + default: + break; + } + + if (vcpu->arch.sie_block->iprcc & PGM_PER) { + pgm_info->per_code = vcpu->arch.sie_block->perc; + pgm_info->per_atmid = vcpu->arch.sie_block->peratmid; + pgm_info->per_address = vcpu->arch.sie_block->peraddr; + pgm_info->per_access_id = vcpu->arch.sie_block->peraid; + } +} + +/* + * restore ITDB to program-interruption TDB in guest lowcore + * and set TX abort indication if required +*/ +static int handle_itdb(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_itdb *itdb; + int rc; + + if (!IS_TE_ENABLED(vcpu) || !IS_ITDB_VALID(vcpu)) + return 0; + if (current->thread.per_flags & PER_FLAG_NO_TE) + return 0; + itdb = (struct kvm_s390_itdb *)vcpu->arch.sie_block->itdba; + rc = write_guest_lc(vcpu, __LC_PGM_TDB, itdb, sizeof(*itdb)); + if (rc) + return rc; + memset(itdb, 0, sizeof(*itdb)); + + return 0; +} + +#define per_event(vcpu) (vcpu->arch.sie_block->iprcc & PGM_PER) + +static int handle_prog(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_pgm_info pgm_info; + psw_t psw; + int rc; + + vcpu->stat.exit_program_interruption++; + + if (guestdbg_enabled(vcpu) && per_event(vcpu)) { + kvm_s390_handle_per_event(vcpu); + /* the interrupt might have been filtered out completely */ + if (vcpu->arch.sie_block->iprcc == 0) + return 0; + } + + trace_kvm_s390_intercept_prog(vcpu, vcpu->arch.sie_block->iprcc); + if (vcpu->arch.sie_block->iprcc == PGM_SPECIFICATION) { + rc = read_guest_lc(vcpu, __LC_PGM_NEW_PSW, &psw, sizeof(psw_t)); + if (rc) + return rc; + /* Avoid endless loops of specification exceptions */ + if (!is_valid_psw(&psw)) + return -EOPNOTSUPP; + } + rc = handle_itdb(vcpu); + if (rc) + return rc; + + __extract_prog_irq(vcpu, &pgm_info); + return kvm_s390_inject_prog_irq(vcpu, &pgm_info); +} + +static int handle_instruction_and_prog(struct kvm_vcpu *vcpu) +{ + int rc, rc2; + + vcpu->stat.exit_instr_and_program++; + rc = handle_instruction(vcpu); + rc2 = handle_prog(vcpu); + + if (rc == -EOPNOTSUPP) + vcpu->arch.sie_block->icptcode = 0x04; + if (rc) + return rc; + return rc2; +} + +/** + * handle_external_interrupt - used for external interruption interceptions + * + * This interception only occurs if the CPUSTAT_EXT_INT bit was set, or if + * the new PSW does not have external interrupts disabled. In the first case, + * we've got to deliver the interrupt manually, and in the second case, we + * drop to userspace to handle the situation there. + */ +static int handle_external_interrupt(struct kvm_vcpu *vcpu) +{ + u16 eic = vcpu->arch.sie_block->eic; + struct kvm_s390_irq irq; + psw_t newpsw; + int rc; + + vcpu->stat.exit_external_interrupt++; + + rc = read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &newpsw, sizeof(psw_t)); + if (rc) + return rc; + /* We can not handle clock comparator or timer interrupt with bad PSW */ + if ((eic == EXT_IRQ_CLK_COMP || eic == EXT_IRQ_CPU_TIMER) && + (newpsw.mask & PSW_MASK_EXT)) + return -EOPNOTSUPP; + + switch (eic) { + case EXT_IRQ_CLK_COMP: + irq.type = KVM_S390_INT_CLOCK_COMP; + break; + case EXT_IRQ_CPU_TIMER: + irq.type = KVM_S390_INT_CPU_TIMER; + break; + case EXT_IRQ_EXTERNAL_CALL: + irq.type = KVM_S390_INT_EXTERNAL_CALL; + irq.u.extcall.code = vcpu->arch.sie_block->extcpuaddr; + rc = kvm_s390_inject_vcpu(vcpu, &irq); + /* ignore if another external call is already pending */ + if (rc == -EBUSY) + return 0; + return rc; + default: + return -EOPNOTSUPP; + } + + return kvm_s390_inject_vcpu(vcpu, &irq); +} + +/** + * Handle MOVE PAGE partial execution interception. + * + * This interception can only happen for guests with DAT disabled and + * addresses that are currently not mapped in the host. Thus we try to + * set up the mappings for the corresponding user pages here (or throw + * addressing exceptions in case of illegal guest addresses). + */ +static int handle_mvpg_pei(struct kvm_vcpu *vcpu) +{ + unsigned long srcaddr, dstaddr; + int reg1, reg2, rc; + + kvm_s390_get_regs_rre(vcpu, ®1, ®2); + + /* Make sure that the source is paged-in */ + rc = guest_translate_address(vcpu, vcpu->run->s.regs.gprs[reg2], + reg2, &srcaddr, 0); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + rc = kvm_arch_fault_in_page(vcpu, srcaddr, 0); + if (rc != 0) + return rc; + + /* Make sure that the destination is paged-in */ + rc = guest_translate_address(vcpu, vcpu->run->s.regs.gprs[reg1], + reg1, &dstaddr, 1); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + rc = kvm_arch_fault_in_page(vcpu, dstaddr, 1); + if (rc != 0) + return rc; + + kvm_s390_rewind_psw(vcpu, 4); + + return 0; +} + +static int handle_partial_execution(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.sie_block->ipa == 0xb254) /* MVPG */ + return handle_mvpg_pei(vcpu); + if (vcpu->arch.sie_block->ipa >> 8 == 0xae) /* SIGP */ + return kvm_s390_handle_sigp_pei(vcpu); + + return -EOPNOTSUPP; +} + +static const intercept_handler_t intercept_funcs[] = { + [0x00 >> 2] = handle_noop, + [0x04 >> 2] = handle_instruction, + [0x08 >> 2] = handle_prog, + [0x0C >> 2] = handle_instruction_and_prog, + [0x10 >> 2] = handle_noop, + [0x14 >> 2] = handle_external_interrupt, + [0x18 >> 2] = handle_noop, + [0x1C >> 2] = kvm_s390_handle_wait, + [0x20 >> 2] = handle_validity, + [0x28 >> 2] = handle_stop, + [0x38 >> 2] = handle_partial_execution, +}; + +int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu) +{ + intercept_handler_t func; + u8 code = vcpu->arch.sie_block->icptcode; + + if (code & 3 || (code >> 2) >= ARRAY_SIZE(intercept_funcs)) + return -EOPNOTSUPP; + func = intercept_funcs[code >> 2]; + if (func) + return func(vcpu); + return -EOPNOTSUPP; +} diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c new file mode 100644 index 000000000..b745a109b --- /dev/null +++ b/arch/s390/kvm/interrupt.c @@ -0,0 +1,2268 @@ +/* + * handling kvm guest interrupts + * + * Copyright IBM Corp. 2008, 2015 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + */ + +#include <linux/interrupt.h> +#include <linux/kvm_host.h> +#include <linux/hrtimer.h> +#include <linux/mmu_context.h> +#include <linux/signal.h> +#include <linux/slab.h> +#include <linux/bitmap.h> +#include <linux/vmalloc.h> +#include <asm/asm-offsets.h> +#include <asm/dis.h> +#include <asm/uaccess.h> +#include <asm/sclp.h> +#include <asm/isc.h> +#include "kvm-s390.h" +#include "gaccess.h" +#include "trace-s390.h" + +#define IOINT_SCHID_MASK 0x0000ffff +#define IOINT_SSID_MASK 0x00030000 +#define IOINT_CSSID_MASK 0x03fc0000 +#define IOINT_AI_MASK 0x04000000 +#define PFAULT_INIT 0x0600 +#define PFAULT_DONE 0x0680 +#define VIRTIO_PARAM 0x0d00 + +int psw_extint_disabled(struct kvm_vcpu *vcpu) +{ + return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_EXT); +} + +static int psw_ioint_disabled(struct kvm_vcpu *vcpu) +{ + return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_IO); +} + +static int psw_mchk_disabled(struct kvm_vcpu *vcpu) +{ + return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_MCHECK); +} + +static int psw_interrupts_disabled(struct kvm_vcpu *vcpu) +{ + if ((vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PER) || + (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_IO) || + (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_EXT)) + return 0; + return 1; +} + +static int ckc_interrupts_enabled(struct kvm_vcpu *vcpu) +{ + if (psw_extint_disabled(vcpu) || + !(vcpu->arch.sie_block->gcr[0] & 0x800ul)) + return 0; + if (guestdbg_enabled(vcpu) && guestdbg_sstep_enabled(vcpu)) + /* No timer interrupts when single stepping */ + return 0; + return 1; +} + +static int ckc_irq_pending(struct kvm_vcpu *vcpu) +{ + if (!(vcpu->arch.sie_block->ckc < + get_tod_clock_fast() + vcpu->arch.sie_block->epoch)) + return 0; + return ckc_interrupts_enabled(vcpu); +} + +static int cpu_timer_interrupts_enabled(struct kvm_vcpu *vcpu) +{ + return !psw_extint_disabled(vcpu) && + (vcpu->arch.sie_block->gcr[0] & 0x400ul); +} + +static int cpu_timer_irq_pending(struct kvm_vcpu *vcpu) +{ + return (vcpu->arch.sie_block->cputm >> 63) && + cpu_timer_interrupts_enabled(vcpu); +} + +static inline int is_ioirq(unsigned long irq_type) +{ + return ((irq_type >= IRQ_PEND_IO_ISC_0) && + (irq_type <= IRQ_PEND_IO_ISC_7)); +} + +static uint64_t isc_to_isc_bits(int isc) +{ + return (0x80 >> isc) << 24; +} + +static inline u8 int_word_to_isc(u32 int_word) +{ + return (int_word & 0x38000000) >> 27; +} + +static inline unsigned long pending_floating_irqs(struct kvm_vcpu *vcpu) +{ + return vcpu->kvm->arch.float_int.pending_irqs; +} + +static inline unsigned long pending_local_irqs(struct kvm_vcpu *vcpu) +{ + return vcpu->arch.local_int.pending_irqs; +} + +static unsigned long disable_iscs(struct kvm_vcpu *vcpu, + unsigned long active_mask) +{ + int i; + + for (i = 0; i <= MAX_ISC; i++) + if (!(vcpu->arch.sie_block->gcr[6] & isc_to_isc_bits(i))) + active_mask &= ~(1UL << (IRQ_PEND_IO_ISC_0 + i)); + + return active_mask; +} + +static unsigned long deliverable_irqs(struct kvm_vcpu *vcpu) +{ + unsigned long active_mask; + + active_mask = pending_local_irqs(vcpu); + active_mask |= pending_floating_irqs(vcpu); + + if (psw_extint_disabled(vcpu)) + active_mask &= ~IRQ_PEND_EXT_MASK; + if (psw_ioint_disabled(vcpu)) + active_mask &= ~IRQ_PEND_IO_MASK; + else + active_mask = disable_iscs(vcpu, active_mask); + if (!(vcpu->arch.sie_block->gcr[0] & 0x2000ul)) + __clear_bit(IRQ_PEND_EXT_EXTERNAL, &active_mask); + if (!(vcpu->arch.sie_block->gcr[0] & 0x4000ul)) + __clear_bit(IRQ_PEND_EXT_EMERGENCY, &active_mask); + if (!(vcpu->arch.sie_block->gcr[0] & 0x800ul)) + __clear_bit(IRQ_PEND_EXT_CLOCK_COMP, &active_mask); + if (!(vcpu->arch.sie_block->gcr[0] & 0x400ul)) + __clear_bit(IRQ_PEND_EXT_CPU_TIMER, &active_mask); + if (!(vcpu->arch.sie_block->gcr[0] & 0x200ul)) + __clear_bit(IRQ_PEND_EXT_SERVICE, &active_mask); + if (psw_mchk_disabled(vcpu)) + active_mask &= ~IRQ_PEND_MCHK_MASK; + if (!(vcpu->arch.sie_block->gcr[14] & + vcpu->kvm->arch.float_int.mchk.cr14)) + __clear_bit(IRQ_PEND_MCHK_REP, &active_mask); + + /* + * STOP irqs will never be actively delivered. They are triggered via + * intercept requests and cleared when the stop intercept is performed. + */ + __clear_bit(IRQ_PEND_SIGP_STOP, &active_mask); + + return active_mask; +} + +static void __set_cpu_idle(struct kvm_vcpu *vcpu) +{ + atomic_set_mask(CPUSTAT_WAIT, &vcpu->arch.sie_block->cpuflags); + set_bit(vcpu->vcpu_id, vcpu->arch.local_int.float_int->idle_mask); +} + +static void __unset_cpu_idle(struct kvm_vcpu *vcpu) +{ + atomic_clear_mask(CPUSTAT_WAIT, &vcpu->arch.sie_block->cpuflags); + clear_bit(vcpu->vcpu_id, vcpu->arch.local_int.float_int->idle_mask); +} + +static void __reset_intercept_indicators(struct kvm_vcpu *vcpu) +{ + atomic_clear_mask(CPUSTAT_IO_INT | CPUSTAT_EXT_INT | CPUSTAT_STOP_INT, + &vcpu->arch.sie_block->cpuflags); + vcpu->arch.sie_block->lctl = 0x0000; + vcpu->arch.sie_block->ictl &= ~(ICTL_LPSW | ICTL_STCTL | ICTL_PINT); + + if (guestdbg_enabled(vcpu)) { + vcpu->arch.sie_block->lctl |= (LCTL_CR0 | LCTL_CR9 | + LCTL_CR10 | LCTL_CR11); + vcpu->arch.sie_block->ictl |= (ICTL_STCTL | ICTL_PINT); + } +} + +static void __set_cpuflag(struct kvm_vcpu *vcpu, u32 flag) +{ + atomic_set_mask(flag, &vcpu->arch.sie_block->cpuflags); +} + +static void set_intercept_indicators_io(struct kvm_vcpu *vcpu) +{ + if (!(pending_floating_irqs(vcpu) & IRQ_PEND_IO_MASK)) + return; + else if (psw_ioint_disabled(vcpu)) + __set_cpuflag(vcpu, CPUSTAT_IO_INT); + else + vcpu->arch.sie_block->lctl |= LCTL_CR6; +} + +static void set_intercept_indicators_ext(struct kvm_vcpu *vcpu) +{ + if (!(pending_local_irqs(vcpu) & IRQ_PEND_EXT_MASK)) + return; + if (psw_extint_disabled(vcpu)) + __set_cpuflag(vcpu, CPUSTAT_EXT_INT); + else + vcpu->arch.sie_block->lctl |= LCTL_CR0; +} + +static void set_intercept_indicators_mchk(struct kvm_vcpu *vcpu) +{ + if (!(pending_local_irqs(vcpu) & IRQ_PEND_MCHK_MASK)) + return; + if (psw_mchk_disabled(vcpu)) + vcpu->arch.sie_block->ictl |= ICTL_LPSW; + else + vcpu->arch.sie_block->lctl |= LCTL_CR14; +} + +static void set_intercept_indicators_stop(struct kvm_vcpu *vcpu) +{ + if (kvm_s390_is_stop_irq_pending(vcpu)) + __set_cpuflag(vcpu, CPUSTAT_STOP_INT); +} + +/* Set interception request for non-deliverable interrupts */ +static void set_intercept_indicators(struct kvm_vcpu *vcpu) +{ + set_intercept_indicators_io(vcpu); + set_intercept_indicators_ext(vcpu); + set_intercept_indicators_mchk(vcpu); + set_intercept_indicators_stop(vcpu); +} + +static u16 get_ilc(struct kvm_vcpu *vcpu) +{ + switch (vcpu->arch.sie_block->icptcode) { + case ICPT_INST: + case ICPT_INSTPROGI: + case ICPT_OPEREXC: + case ICPT_PARTEXEC: + case ICPT_IOINST: + /* last instruction only stored for these icptcodes */ + return insn_length(vcpu->arch.sie_block->ipa >> 8); + case ICPT_PROGI: + return vcpu->arch.sie_block->pgmilc; + default: + return 0; + } +} + +static int __must_check __deliver_cpu_timer(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + int rc; + + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_INT_CPU_TIMER, + 0, 0); + + rc = put_guest_lc(vcpu, EXT_IRQ_CPU_TIMER, + (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, 0, (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + clear_bit(IRQ_PEND_EXT_CPU_TIMER, &li->pending_irqs); + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_ckc(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + int rc; + + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_INT_CLOCK_COMP, + 0, 0); + + rc = put_guest_lc(vcpu, EXT_IRQ_CLK_COMP, + (u16 __user *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, 0, (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + clear_bit(IRQ_PEND_EXT_CLOCK_COMP, &li->pending_irqs); + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_pfault_init(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_ext_info ext; + int rc; + + spin_lock(&li->lock); + ext = li->irq.ext; + clear_bit(IRQ_PEND_PFAULT_INIT, &li->pending_irqs); + li->irq.ext.ext_params2 = 0; + spin_unlock(&li->lock); + + VCPU_EVENT(vcpu, 4, "interrupt: pfault init parm:%x,parm64:%llx", + 0, ext.ext_params2); + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, + KVM_S390_INT_PFAULT_INIT, + 0, ext.ext_params2); + + rc = put_guest_lc(vcpu, EXT_IRQ_CP_SERVICE, (u16 *) __LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, PFAULT_INIT, (u16 *) __LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= put_guest_lc(vcpu, ext.ext_params2, (u64 *) __LC_EXT_PARAMS2); + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_machine_check(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_mchk_info mchk = {}; + unsigned long adtl_status_addr; + int deliver = 0; + int rc = 0; + + spin_lock(&fi->lock); + spin_lock(&li->lock); + if (test_bit(IRQ_PEND_MCHK_EX, &li->pending_irqs) || + test_bit(IRQ_PEND_MCHK_REP, &li->pending_irqs)) { + /* + * If there was an exigent machine check pending, then any + * repressible machine checks that might have been pending + * are indicated along with it, so always clear bits for + * repressible and exigent interrupts + */ + mchk = li->irq.mchk; + clear_bit(IRQ_PEND_MCHK_EX, &li->pending_irqs); + clear_bit(IRQ_PEND_MCHK_REP, &li->pending_irqs); + memset(&li->irq.mchk, 0, sizeof(mchk)); + deliver = 1; + } + /* + * We indicate floating repressible conditions along with + * other pending conditions. Channel Report Pending and Channel + * Subsystem damage are the only two and and are indicated by + * bits in mcic and masked in cr14. + */ + if (test_and_clear_bit(IRQ_PEND_MCHK_REP, &fi->pending_irqs)) { + mchk.mcic |= fi->mchk.mcic; + mchk.cr14 |= fi->mchk.cr14; + memset(&fi->mchk, 0, sizeof(mchk)); + deliver = 1; + } + spin_unlock(&li->lock); + spin_unlock(&fi->lock); + + if (deliver) { + VCPU_EVENT(vcpu, 4, "interrupt: machine check mcic=%llx", + mchk.mcic); + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, + KVM_S390_MCHK, + mchk.cr14, mchk.mcic); + + rc = kvm_s390_vcpu_store_status(vcpu, + KVM_S390_STORE_STATUS_PREFIXED); + rc |= read_guest_lc(vcpu, __LC_VX_SAVE_AREA_ADDR, + &adtl_status_addr, + sizeof(unsigned long)); + rc |= kvm_s390_vcpu_store_adtl_status(vcpu, + adtl_status_addr); + rc |= put_guest_lc(vcpu, mchk.mcic, + (u64 __user *) __LC_MCCK_CODE); + rc |= put_guest_lc(vcpu, mchk.failing_storage_address, + (u64 __user *) __LC_MCCK_FAIL_STOR_ADDR); + rc |= write_guest_lc(vcpu, __LC_PSW_SAVE_AREA, + &mchk.fixed_logout, + sizeof(mchk.fixed_logout)); + rc |= write_guest_lc(vcpu, __LC_MCK_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_MCK_NEW_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + } + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_restart(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + int rc; + + VCPU_EVENT(vcpu, 4, "%s", "interrupt: cpu restart"); + vcpu->stat.deliver_restart_signal++; + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_RESTART, 0, 0); + + rc = write_guest_lc(vcpu, + offsetof(struct _lowcore, restart_old_psw), + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, offsetof(struct _lowcore, restart_psw), + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + clear_bit(IRQ_PEND_RESTART, &li->pending_irqs); + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_set_prefix(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_prefix_info prefix; + + spin_lock(&li->lock); + prefix = li->irq.prefix; + li->irq.prefix.address = 0; + clear_bit(IRQ_PEND_SET_PREFIX, &li->pending_irqs); + spin_unlock(&li->lock); + + VCPU_EVENT(vcpu, 4, "interrupt: set prefix to %x", prefix.address); + vcpu->stat.deliver_prefix_signal++; + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, + KVM_S390_SIGP_SET_PREFIX, + prefix.address, 0); + + kvm_s390_set_prefix(vcpu, prefix.address); + return 0; +} + +static int __must_check __deliver_emergency_signal(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + int rc; + int cpu_addr; + + spin_lock(&li->lock); + cpu_addr = find_first_bit(li->sigp_emerg_pending, KVM_MAX_VCPUS); + clear_bit(cpu_addr, li->sigp_emerg_pending); + if (bitmap_empty(li->sigp_emerg_pending, KVM_MAX_VCPUS)) + clear_bit(IRQ_PEND_EXT_EMERGENCY, &li->pending_irqs); + spin_unlock(&li->lock); + + VCPU_EVENT(vcpu, 4, "%s", "interrupt: sigp emerg"); + vcpu->stat.deliver_emergency_signal++; + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_INT_EMERGENCY, + cpu_addr, 0); + + rc = put_guest_lc(vcpu, EXT_IRQ_EMERGENCY_SIG, + (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, cpu_addr, (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_external_call(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_extcall_info extcall; + int rc; + + spin_lock(&li->lock); + extcall = li->irq.extcall; + li->irq.extcall.code = 0; + clear_bit(IRQ_PEND_EXT_EXTERNAL, &li->pending_irqs); + spin_unlock(&li->lock); + + VCPU_EVENT(vcpu, 4, "%s", "interrupt: sigp ext call"); + vcpu->stat.deliver_external_call++; + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, + KVM_S390_INT_EXTERNAL_CALL, + extcall.code, 0); + + rc = put_guest_lc(vcpu, EXT_IRQ_EXTERNAL_CALL, + (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, extcall.code, (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_prog(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_pgm_info pgm_info; + int rc = 0, nullifying = false; + u16 ilc = get_ilc(vcpu); + + spin_lock(&li->lock); + pgm_info = li->irq.pgm; + clear_bit(IRQ_PEND_PROG, &li->pending_irqs); + memset(&li->irq.pgm, 0, sizeof(pgm_info)); + spin_unlock(&li->lock); + + VCPU_EVENT(vcpu, 4, "interrupt: pgm check code:%x, ilc:%x", + pgm_info.code, ilc); + vcpu->stat.deliver_program_int++; + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_PROGRAM_INT, + pgm_info.code, 0); + + switch (pgm_info.code & ~PGM_PER) { + case PGM_AFX_TRANSLATION: + case PGM_ASX_TRANSLATION: + case PGM_EX_TRANSLATION: + case PGM_LFX_TRANSLATION: + case PGM_LSTE_SEQUENCE: + case PGM_LSX_TRANSLATION: + case PGM_LX_TRANSLATION: + case PGM_PRIMARY_AUTHORITY: + case PGM_SECONDARY_AUTHORITY: + nullifying = true; + /* fall through */ + case PGM_SPACE_SWITCH: + rc = put_guest_lc(vcpu, pgm_info.trans_exc_code, + (u64 *)__LC_TRANS_EXC_CODE); + break; + case PGM_ALEN_TRANSLATION: + case PGM_ALE_SEQUENCE: + case PGM_ASTE_INSTANCE: + case PGM_ASTE_SEQUENCE: + case PGM_ASTE_VALIDITY: + case PGM_EXTENDED_AUTHORITY: + rc = put_guest_lc(vcpu, pgm_info.exc_access_id, + (u8 *)__LC_EXC_ACCESS_ID); + nullifying = true; + break; + case PGM_ASCE_TYPE: + case PGM_PAGE_TRANSLATION: + case PGM_REGION_FIRST_TRANS: + case PGM_REGION_SECOND_TRANS: + case PGM_REGION_THIRD_TRANS: + case PGM_SEGMENT_TRANSLATION: + rc = put_guest_lc(vcpu, pgm_info.trans_exc_code, + (u64 *)__LC_TRANS_EXC_CODE); + rc |= put_guest_lc(vcpu, pgm_info.exc_access_id, + (u8 *)__LC_EXC_ACCESS_ID); + rc |= put_guest_lc(vcpu, pgm_info.op_access_id, + (u8 *)__LC_OP_ACCESS_ID); + nullifying = true; + break; + case PGM_MONITOR: + rc = put_guest_lc(vcpu, pgm_info.mon_class_nr, + (u16 *)__LC_MON_CLASS_NR); + rc |= put_guest_lc(vcpu, pgm_info.mon_code, + (u64 *)__LC_MON_CODE); + break; + case PGM_VECTOR_PROCESSING: + case PGM_DATA: + rc = put_guest_lc(vcpu, pgm_info.data_exc_code, + (u32 *)__LC_DATA_EXC_CODE); + break; + case PGM_PROTECTION: + rc = put_guest_lc(vcpu, pgm_info.trans_exc_code, + (u64 *)__LC_TRANS_EXC_CODE); + rc |= put_guest_lc(vcpu, pgm_info.exc_access_id, + (u8 *)__LC_EXC_ACCESS_ID); + break; + case PGM_STACK_FULL: + case PGM_STACK_EMPTY: + case PGM_STACK_SPECIFICATION: + case PGM_STACK_TYPE: + case PGM_STACK_OPERATION: + case PGM_TRACE_TABEL: + case PGM_CRYPTO_OPERATION: + nullifying = true; + break; + } + + if (pgm_info.code & PGM_PER) { + rc |= put_guest_lc(vcpu, pgm_info.per_code, + (u8 *) __LC_PER_CODE); + rc |= put_guest_lc(vcpu, pgm_info.per_atmid, + (u8 *)__LC_PER_ATMID); + rc |= put_guest_lc(vcpu, pgm_info.per_address, + (u64 *) __LC_PER_ADDRESS); + rc |= put_guest_lc(vcpu, pgm_info.per_access_id, + (u8 *) __LC_PER_ACCESS_ID); + } + + if (nullifying && vcpu->arch.sie_block->icptcode == ICPT_INST) + kvm_s390_rewind_psw(vcpu, ilc); + + rc |= put_guest_lc(vcpu, ilc, (u16 *) __LC_PGM_ILC); + rc |= put_guest_lc(vcpu, vcpu->arch.sie_block->gbea, + (u64 *) __LC_LAST_BREAK); + rc |= put_guest_lc(vcpu, pgm_info.code, + (u16 *)__LC_PGM_INT_CODE); + rc |= write_guest_lc(vcpu, __LC_PGM_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_PGM_NEW_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_service(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct kvm_s390_ext_info ext; + int rc = 0; + + spin_lock(&fi->lock); + if (!(test_bit(IRQ_PEND_EXT_SERVICE, &fi->pending_irqs))) { + spin_unlock(&fi->lock); + return 0; + } + ext = fi->srv_signal; + memset(&fi->srv_signal, 0, sizeof(ext)); + clear_bit(IRQ_PEND_EXT_SERVICE, &fi->pending_irqs); + spin_unlock(&fi->lock); + + VCPU_EVENT(vcpu, 4, "interrupt: sclp parm:%x", + ext.ext_params); + vcpu->stat.deliver_service_signal++; + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_INT_SERVICE, + ext.ext_params, 0); + + rc = put_guest_lc(vcpu, EXT_IRQ_SERVICE_SIG, (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, 0, (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + rc |= put_guest_lc(vcpu, ext.ext_params, + (u32 *)__LC_EXT_PARAMS); + + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_pfault_done(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct kvm_s390_interrupt_info *inti; + int rc = 0; + + spin_lock(&fi->lock); + inti = list_first_entry_or_null(&fi->lists[FIRQ_LIST_PFAULT], + struct kvm_s390_interrupt_info, + list); + if (inti) { + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, + KVM_S390_INT_PFAULT_DONE, 0, + inti->ext.ext_params2); + list_del(&inti->list); + fi->counters[FIRQ_CNTR_PFAULT] -= 1; + } + if (list_empty(&fi->lists[FIRQ_LIST_PFAULT])) + clear_bit(IRQ_PEND_PFAULT_DONE, &fi->pending_irqs); + spin_unlock(&fi->lock); + + if (inti) { + rc = put_guest_lc(vcpu, EXT_IRQ_CP_SERVICE, + (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, PFAULT_DONE, + (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= put_guest_lc(vcpu, inti->ext.ext_params2, + (u64 *)__LC_EXT_PARAMS2); + kfree(inti); + } + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_virtio(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct kvm_s390_interrupt_info *inti; + int rc = 0; + + spin_lock(&fi->lock); + inti = list_first_entry_or_null(&fi->lists[FIRQ_LIST_VIRTIO], + struct kvm_s390_interrupt_info, + list); + if (inti) { + VCPU_EVENT(vcpu, 4, + "interrupt: virtio parm:%x,parm64:%llx", + inti->ext.ext_params, inti->ext.ext_params2); + vcpu->stat.deliver_virtio_interrupt++; + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, + inti->type, + inti->ext.ext_params, + inti->ext.ext_params2); + list_del(&inti->list); + fi->counters[FIRQ_CNTR_VIRTIO] -= 1; + } + if (list_empty(&fi->lists[FIRQ_LIST_VIRTIO])) + clear_bit(IRQ_PEND_VIRTIO, &fi->pending_irqs); + spin_unlock(&fi->lock); + + if (inti) { + rc = put_guest_lc(vcpu, EXT_IRQ_CP_SERVICE, + (u16 *)__LC_EXT_INT_CODE); + rc |= put_guest_lc(vcpu, VIRTIO_PARAM, + (u16 *)__LC_EXT_CPU_ADDR); + rc |= write_guest_lc(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_EXT_NEW_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= put_guest_lc(vcpu, inti->ext.ext_params, + (u32 *)__LC_EXT_PARAMS); + rc |= put_guest_lc(vcpu, inti->ext.ext_params2, + (u64 *)__LC_EXT_PARAMS2); + kfree(inti); + } + return rc ? -EFAULT : 0; +} + +static int __must_check __deliver_io(struct kvm_vcpu *vcpu, + unsigned long irq_type) +{ + struct list_head *isc_list; + struct kvm_s390_float_interrupt *fi; + struct kvm_s390_interrupt_info *inti = NULL; + int rc = 0; + + fi = &vcpu->kvm->arch.float_int; + + spin_lock(&fi->lock); + isc_list = &fi->lists[irq_type - IRQ_PEND_IO_ISC_0]; + inti = list_first_entry_or_null(isc_list, + struct kvm_s390_interrupt_info, + list); + if (inti) { + VCPU_EVENT(vcpu, 4, "interrupt: I/O %llx", inti->type); + vcpu->stat.deliver_io_int++; + trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, + inti->type, + ((__u32)inti->io.subchannel_id << 16) | + inti->io.subchannel_nr, + ((__u64)inti->io.io_int_parm << 32) | + inti->io.io_int_word); + list_del(&inti->list); + fi->counters[FIRQ_CNTR_IO] -= 1; + } + if (list_empty(isc_list)) + clear_bit(irq_type, &fi->pending_irqs); + spin_unlock(&fi->lock); + + if (inti) { + rc = put_guest_lc(vcpu, inti->io.subchannel_id, + (u16 *)__LC_SUBCHANNEL_ID); + rc |= put_guest_lc(vcpu, inti->io.subchannel_nr, + (u16 *)__LC_SUBCHANNEL_NR); + rc |= put_guest_lc(vcpu, inti->io.io_int_parm, + (u32 *)__LC_IO_INT_PARM); + rc |= put_guest_lc(vcpu, inti->io.io_int_word, + (u32 *)__LC_IO_INT_WORD); + rc |= write_guest_lc(vcpu, __LC_IO_OLD_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + rc |= read_guest_lc(vcpu, __LC_IO_NEW_PSW, + &vcpu->arch.sie_block->gpsw, + sizeof(psw_t)); + kfree(inti); + } + + return rc ? -EFAULT : 0; +} + +typedef int (*deliver_irq_t)(struct kvm_vcpu *vcpu); + +static const deliver_irq_t deliver_irq_funcs[] = { + [IRQ_PEND_MCHK_EX] = __deliver_machine_check, + [IRQ_PEND_MCHK_REP] = __deliver_machine_check, + [IRQ_PEND_PROG] = __deliver_prog, + [IRQ_PEND_EXT_EMERGENCY] = __deliver_emergency_signal, + [IRQ_PEND_EXT_EXTERNAL] = __deliver_external_call, + [IRQ_PEND_EXT_CLOCK_COMP] = __deliver_ckc, + [IRQ_PEND_EXT_CPU_TIMER] = __deliver_cpu_timer, + [IRQ_PEND_RESTART] = __deliver_restart, + [IRQ_PEND_SET_PREFIX] = __deliver_set_prefix, + [IRQ_PEND_PFAULT_INIT] = __deliver_pfault_init, + [IRQ_PEND_EXT_SERVICE] = __deliver_service, + [IRQ_PEND_PFAULT_DONE] = __deliver_pfault_done, + [IRQ_PEND_VIRTIO] = __deliver_virtio, +}; + +/* Check whether an external call is pending (deliverable or not) */ +int kvm_s390_ext_call_pending(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + uint8_t sigp_ctrl = vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sigp_ctrl; + + if (!sclp_has_sigpif()) + return test_bit(IRQ_PEND_EXT_EXTERNAL, &li->pending_irqs); + + return (sigp_ctrl & SIGP_CTRL_C) && + (atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_ECALL_PEND); +} + +int kvm_s390_vcpu_has_irq(struct kvm_vcpu *vcpu, int exclude_stop) +{ + int rc; + + rc = !!deliverable_irqs(vcpu); + + if (!rc && kvm_cpu_has_pending_timer(vcpu)) + rc = 1; + + /* external call pending and deliverable */ + if (!rc && kvm_s390_ext_call_pending(vcpu) && + !psw_extint_disabled(vcpu) && + (vcpu->arch.sie_block->gcr[0] & 0x2000ul)) + rc = 1; + + if (!rc && !exclude_stop && kvm_s390_is_stop_irq_pending(vcpu)) + rc = 1; + + return rc; +} + +int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu) +{ + return ckc_irq_pending(vcpu) || cpu_timer_irq_pending(vcpu); +} + +int kvm_s390_handle_wait(struct kvm_vcpu *vcpu) +{ + u64 now, sltime; + + vcpu->stat.exit_wait_state++; + + /* fast path */ + if (kvm_cpu_has_pending_timer(vcpu) || kvm_arch_vcpu_runnable(vcpu)) + return 0; + + if (psw_interrupts_disabled(vcpu)) { + VCPU_EVENT(vcpu, 3, "%s", "disabled wait"); + return -EOPNOTSUPP; /* disabled wait */ + } + + if (!ckc_interrupts_enabled(vcpu)) { + VCPU_EVENT(vcpu, 3, "%s", "enabled wait w/o timer"); + __set_cpu_idle(vcpu); + goto no_timer; + } + + now = get_tod_clock_fast() + vcpu->arch.sie_block->epoch; + sltime = tod_to_ns(vcpu->arch.sie_block->ckc - now); + + /* underflow */ + if (vcpu->arch.sie_block->ckc < now) + return 0; + + __set_cpu_idle(vcpu); + hrtimer_start(&vcpu->arch.ckc_timer, ktime_set (0, sltime) , HRTIMER_MODE_REL); + VCPU_EVENT(vcpu, 5, "enabled wait via clock comparator: %llx ns", sltime); +no_timer: + srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); + kvm_vcpu_block(vcpu); + __unset_cpu_idle(vcpu); + vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); + + hrtimer_cancel(&vcpu->arch.ckc_timer); + return 0; +} + +void kvm_s390_vcpu_wakeup(struct kvm_vcpu *vcpu) +{ + if (waitqueue_active(&vcpu->wq)) { + /* + * The vcpu gave up the cpu voluntarily, mark it as a good + * yield-candidate. + */ + vcpu->preempted = true; + wake_up_interruptible(&vcpu->wq); + vcpu->stat.halt_wakeup++; + } +} + +enum hrtimer_restart kvm_s390_idle_wakeup(struct hrtimer *timer) +{ + struct kvm_vcpu *vcpu; + u64 now, sltime; + + vcpu = container_of(timer, struct kvm_vcpu, arch.ckc_timer); + now = get_tod_clock_fast() + vcpu->arch.sie_block->epoch; + sltime = tod_to_ns(vcpu->arch.sie_block->ckc - now); + + /* + * If the monotonic clock runs faster than the tod clock we might be + * woken up too early and have to go back to sleep to avoid deadlocks. + */ + if (vcpu->arch.sie_block->ckc > now && + hrtimer_forward_now(timer, ns_to_ktime(sltime))) + return HRTIMER_RESTART; + kvm_s390_vcpu_wakeup(vcpu); + return HRTIMER_NORESTART; +} + +void kvm_s390_clear_local_irqs(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + + spin_lock(&li->lock); + li->pending_irqs = 0; + bitmap_zero(li->sigp_emerg_pending, KVM_MAX_VCPUS); + memset(&li->irq, 0, sizeof(li->irq)); + spin_unlock(&li->lock); + + /* clear pending external calls set by sigp interpretation facility */ + atomic_clear_mask(CPUSTAT_ECALL_PEND, li->cpuflags); + vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sigp_ctrl = 0; +} + +int __must_check kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + deliver_irq_t func; + int rc = 0; + unsigned long irq_type; + unsigned long irqs; + + __reset_intercept_indicators(vcpu); + + /* pending ckc conditions might have been invalidated */ + clear_bit(IRQ_PEND_EXT_CLOCK_COMP, &li->pending_irqs); + if (ckc_irq_pending(vcpu)) + set_bit(IRQ_PEND_EXT_CLOCK_COMP, &li->pending_irqs); + + /* pending cpu timer conditions might have been invalidated */ + clear_bit(IRQ_PEND_EXT_CPU_TIMER, &li->pending_irqs); + if (cpu_timer_irq_pending(vcpu)) + set_bit(IRQ_PEND_EXT_CPU_TIMER, &li->pending_irqs); + + do { + irqs = deliverable_irqs(vcpu); + /* bits are in the order of interrupt priority */ + irq_type = find_first_bit(&irqs, IRQ_PEND_COUNT); + if (irq_type == IRQ_PEND_COUNT) + break; + if (is_ioirq(irq_type)) { + rc = __deliver_io(vcpu, irq_type); + } else { + func = deliver_irq_funcs[irq_type]; + if (!func) { + WARN_ON_ONCE(func == NULL); + clear_bit(irq_type, &li->pending_irqs); + continue; + } + rc = func(vcpu); + } + if (rc) + break; + } while (!rc); + + set_intercept_indicators(vcpu); + + return rc; +} + +static int __inject_prog(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + + li->irq.pgm = irq->u.pgm; + set_bit(IRQ_PEND_PROG, &li->pending_irqs); + return 0; +} + +int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_irq irq; + + VCPU_EVENT(vcpu, 3, "inject: program check %d (from kernel)", code); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_PROGRAM_INT, code, + 0, 1); + spin_lock(&li->lock); + irq.u.pgm.code = code; + __inject_prog(vcpu, &irq); + BUG_ON(waitqueue_active(li->wq)); + spin_unlock(&li->lock); + return 0; +} + +int kvm_s390_inject_prog_irq(struct kvm_vcpu *vcpu, + struct kvm_s390_pgm_info *pgm_info) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_irq irq; + int rc; + + VCPU_EVENT(vcpu, 3, "inject: prog irq %d (from kernel)", + pgm_info->code); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_PROGRAM_INT, + pgm_info->code, 0, 1); + spin_lock(&li->lock); + irq.u.pgm = *pgm_info; + rc = __inject_prog(vcpu, &irq); + BUG_ON(waitqueue_active(li->wq)); + spin_unlock(&li->lock); + return rc; +} + +static int __inject_pfault_init(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + + VCPU_EVENT(vcpu, 3, "inject: external irq params:%x, params2:%llx", + irq->u.ext.ext_params, irq->u.ext.ext_params2); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_PFAULT_INIT, + irq->u.ext.ext_params, + irq->u.ext.ext_params2, 2); + + li->irq.ext = irq->u.ext; + set_bit(IRQ_PEND_PFAULT_INIT, &li->pending_irqs); + atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); + return 0; +} + +static int __inject_extcall_sigpif(struct kvm_vcpu *vcpu, uint16_t src_id) +{ + unsigned char new_val, old_val; + uint8_t *sigp_ctrl = &vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sigp_ctrl; + + new_val = SIGP_CTRL_C | (src_id & SIGP_CTRL_SCN_MASK); + old_val = *sigp_ctrl & ~SIGP_CTRL_C; + if (cmpxchg(sigp_ctrl, old_val, new_val) != old_val) { + /* another external call is pending */ + return -EBUSY; + } + atomic_set_mask(CPUSTAT_ECALL_PEND, &vcpu->arch.sie_block->cpuflags); + return 0; +} + +static int __inject_extcall(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_extcall_info *extcall = &li->irq.extcall; + uint16_t src_id = irq->u.extcall.code; + + VCPU_EVENT(vcpu, 3, "inject: external call source-cpu:%u", + src_id); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_EXTERNAL_CALL, + src_id, 0, 2); + + /* sending vcpu invalid */ + if (src_id >= KVM_MAX_VCPUS || + kvm_get_vcpu(vcpu->kvm, src_id) == NULL) + return -EINVAL; + + if (sclp_has_sigpif()) + return __inject_extcall_sigpif(vcpu, src_id); + + if (test_and_set_bit(IRQ_PEND_EXT_EXTERNAL, &li->pending_irqs)) + return -EBUSY; + *extcall = irq->u.extcall; + atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); + return 0; +} + +static int __inject_set_prefix(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_prefix_info *prefix = &li->irq.prefix; + + VCPU_EVENT(vcpu, 3, "inject: set prefix to %x (from user)", + irq->u.prefix.address); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_SIGP_SET_PREFIX, + irq->u.prefix.address, 0, 2); + + if (!is_vcpu_stopped(vcpu)) + return -EBUSY; + + *prefix = irq->u.prefix; + set_bit(IRQ_PEND_SET_PREFIX, &li->pending_irqs); + return 0; +} + +#define KVM_S390_STOP_SUPP_FLAGS (KVM_S390_STOP_FLAG_STORE_STATUS) +static int __inject_sigp_stop(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_stop_info *stop = &li->irq.stop; + int rc = 0; + + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_SIGP_STOP, 0, 0, 2); + + if (irq->u.stop.flags & ~KVM_S390_STOP_SUPP_FLAGS) + return -EINVAL; + + if (is_vcpu_stopped(vcpu)) { + if (irq->u.stop.flags & KVM_S390_STOP_FLAG_STORE_STATUS) + rc = kvm_s390_store_status_unloaded(vcpu, + KVM_S390_STORE_STATUS_NOADDR); + return rc; + } + + if (test_and_set_bit(IRQ_PEND_SIGP_STOP, &li->pending_irqs)) + return -EBUSY; + stop->flags = irq->u.stop.flags; + __set_cpuflag(vcpu, CPUSTAT_STOP_INT); + return 0; +} + +static int __inject_sigp_restart(struct kvm_vcpu *vcpu, + struct kvm_s390_irq *irq) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + + VCPU_EVENT(vcpu, 3, "inject: restart type %llx", irq->type); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_RESTART, 0, 0, 2); + + set_bit(IRQ_PEND_RESTART, &li->pending_irqs); + return 0; +} + +static int __inject_sigp_emergency(struct kvm_vcpu *vcpu, + struct kvm_s390_irq *irq) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + + VCPU_EVENT(vcpu, 3, "inject: emergency %u\n", + irq->u.emerg.code); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_EMERGENCY, + irq->u.emerg.code, 0, 2); + + set_bit(irq->u.emerg.code, li->sigp_emerg_pending); + set_bit(IRQ_PEND_EXT_EMERGENCY, &li->pending_irqs); + atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); + return 0; +} + +static int __inject_mchk(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_mchk_info *mchk = &li->irq.mchk; + + VCPU_EVENT(vcpu, 5, "inject: machine check parm64:%llx", + irq->u.mchk.mcic); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_MCHK, 0, + irq->u.mchk.mcic, 2); + + /* + * Because repressible machine checks can be indicated along with + * exigent machine checks (PoP, Chapter 11, Interruption action) + * we need to combine cr14, mcic and external damage code. + * Failing storage address and the logout area should not be or'ed + * together, we just indicate the last occurrence of the corresponding + * machine check + */ + mchk->cr14 |= irq->u.mchk.cr14; + mchk->mcic |= irq->u.mchk.mcic; + mchk->ext_damage_code |= irq->u.mchk.ext_damage_code; + mchk->failing_storage_address = irq->u.mchk.failing_storage_address; + memcpy(&mchk->fixed_logout, &irq->u.mchk.fixed_logout, + sizeof(mchk->fixed_logout)); + if (mchk->mcic & MCHK_EX_MASK) + set_bit(IRQ_PEND_MCHK_EX, &li->pending_irqs); + else if (mchk->mcic & MCHK_REP_MASK) + set_bit(IRQ_PEND_MCHK_REP, &li->pending_irqs); + return 0; +} + +static int __inject_ckc(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + + VCPU_EVENT(vcpu, 3, "inject: type %x", KVM_S390_INT_CLOCK_COMP); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_CLOCK_COMP, + 0, 0, 2); + + set_bit(IRQ_PEND_EXT_CLOCK_COMP, &li->pending_irqs); + atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); + return 0; +} + +static int __inject_cpu_timer(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + + VCPU_EVENT(vcpu, 3, "inject: type %x", KVM_S390_INT_CPU_TIMER); + trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_CPU_TIMER, + 0, 0, 2); + + set_bit(IRQ_PEND_EXT_CPU_TIMER, &li->pending_irqs); + atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); + return 0; +} + +static struct kvm_s390_interrupt_info *get_io_int(struct kvm *kvm, + int isc, u32 schid) +{ + struct kvm_s390_float_interrupt *fi = &kvm->arch.float_int; + struct list_head *isc_list = &fi->lists[FIRQ_LIST_IO_ISC_0 + isc]; + struct kvm_s390_interrupt_info *iter; + u16 id = (schid & 0xffff0000U) >> 16; + u16 nr = schid & 0x0000ffffU; + + spin_lock(&fi->lock); + list_for_each_entry(iter, isc_list, list) { + if (schid && (id != iter->io.subchannel_id || + nr != iter->io.subchannel_nr)) + continue; + /* found an appropriate entry */ + list_del_init(&iter->list); + fi->counters[FIRQ_CNTR_IO] -= 1; + if (list_empty(isc_list)) + clear_bit(IRQ_PEND_IO_ISC_0 + isc, &fi->pending_irqs); + spin_unlock(&fi->lock); + return iter; + } + spin_unlock(&fi->lock); + return NULL; +} + +/* + * Dequeue and return an I/O interrupt matching any of the interruption + * subclasses as designated by the isc mask in cr6 and the schid (if != 0). + */ +struct kvm_s390_interrupt_info *kvm_s390_get_io_int(struct kvm *kvm, + u64 isc_mask, u32 schid) +{ + struct kvm_s390_interrupt_info *inti = NULL; + int isc; + + for (isc = 0; isc <= MAX_ISC && !inti; isc++) { + if (isc_mask & isc_to_isc_bits(isc)) + inti = get_io_int(kvm, isc, schid); + } + return inti; +} + +#define SCCB_MASK 0xFFFFFFF8 +#define SCCB_EVENT_PENDING 0x3 + +static int __inject_service(struct kvm *kvm, + struct kvm_s390_interrupt_info *inti) +{ + struct kvm_s390_float_interrupt *fi = &kvm->arch.float_int; + + spin_lock(&fi->lock); + fi->srv_signal.ext_params |= inti->ext.ext_params & SCCB_EVENT_PENDING; + /* + * Early versions of the QEMU s390 bios will inject several + * service interrupts after another without handling a + * condition code indicating busy. + * We will silently ignore those superfluous sccb values. + * A future version of QEMU will take care of serialization + * of servc requests + */ + if (fi->srv_signal.ext_params & SCCB_MASK) + goto out; + fi->srv_signal.ext_params |= inti->ext.ext_params & SCCB_MASK; + set_bit(IRQ_PEND_EXT_SERVICE, &fi->pending_irqs); +out: + spin_unlock(&fi->lock); + kfree(inti); + return 0; +} + +static int __inject_virtio(struct kvm *kvm, + struct kvm_s390_interrupt_info *inti) +{ + struct kvm_s390_float_interrupt *fi = &kvm->arch.float_int; + + spin_lock(&fi->lock); + if (fi->counters[FIRQ_CNTR_VIRTIO] >= KVM_S390_MAX_VIRTIO_IRQS) { + spin_unlock(&fi->lock); + return -EBUSY; + } + fi->counters[FIRQ_CNTR_VIRTIO] += 1; + list_add_tail(&inti->list, &fi->lists[FIRQ_LIST_VIRTIO]); + set_bit(IRQ_PEND_VIRTIO, &fi->pending_irqs); + spin_unlock(&fi->lock); + return 0; +} + +static int __inject_pfault_done(struct kvm *kvm, + struct kvm_s390_interrupt_info *inti) +{ + struct kvm_s390_float_interrupt *fi = &kvm->arch.float_int; + + spin_lock(&fi->lock); + if (fi->counters[FIRQ_CNTR_PFAULT] >= + (ASYNC_PF_PER_VCPU * KVM_MAX_VCPUS)) { + spin_unlock(&fi->lock); + return -EBUSY; + } + fi->counters[FIRQ_CNTR_PFAULT] += 1; + list_add_tail(&inti->list, &fi->lists[FIRQ_LIST_PFAULT]); + set_bit(IRQ_PEND_PFAULT_DONE, &fi->pending_irqs); + spin_unlock(&fi->lock); + return 0; +} + +#define CR_PENDING_SUBCLASS 28 +static int __inject_float_mchk(struct kvm *kvm, + struct kvm_s390_interrupt_info *inti) +{ + struct kvm_s390_float_interrupt *fi = &kvm->arch.float_int; + + spin_lock(&fi->lock); + fi->mchk.cr14 |= inti->mchk.cr14 & (1UL << CR_PENDING_SUBCLASS); + fi->mchk.mcic |= inti->mchk.mcic; + set_bit(IRQ_PEND_MCHK_REP, &fi->pending_irqs); + spin_unlock(&fi->lock); + kfree(inti); + return 0; +} + +static int __inject_io(struct kvm *kvm, struct kvm_s390_interrupt_info *inti) +{ + struct kvm_s390_float_interrupt *fi; + struct list_head *list; + int isc; + + fi = &kvm->arch.float_int; + spin_lock(&fi->lock); + if (fi->counters[FIRQ_CNTR_IO] >= KVM_S390_MAX_FLOAT_IRQS) { + spin_unlock(&fi->lock); + return -EBUSY; + } + fi->counters[FIRQ_CNTR_IO] += 1; + + isc = int_word_to_isc(inti->io.io_int_word); + list = &fi->lists[FIRQ_LIST_IO_ISC_0 + isc]; + list_add_tail(&inti->list, list); + set_bit(IRQ_PEND_IO_ISC_0 + isc, &fi->pending_irqs); + spin_unlock(&fi->lock); + return 0; +} + +static int __inject_vm(struct kvm *kvm, struct kvm_s390_interrupt_info *inti) +{ + struct kvm_s390_local_interrupt *li; + struct kvm_s390_float_interrupt *fi; + struct kvm_vcpu *dst_vcpu = NULL; + int sigcpu; + u64 type = READ_ONCE(inti->type); + int rc; + + fi = &kvm->arch.float_int; + + switch (type) { + case KVM_S390_MCHK: + rc = __inject_float_mchk(kvm, inti); + break; + case KVM_S390_INT_VIRTIO: + rc = __inject_virtio(kvm, inti); + break; + case KVM_S390_INT_SERVICE: + rc = __inject_service(kvm, inti); + break; + case KVM_S390_INT_PFAULT_DONE: + rc = __inject_pfault_done(kvm, inti); + break; + case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX: + rc = __inject_io(kvm, inti); + break; + default: + rc = -EINVAL; + } + if (rc) + return rc; + + sigcpu = find_first_bit(fi->idle_mask, KVM_MAX_VCPUS); + if (sigcpu == KVM_MAX_VCPUS) { + do { + sigcpu = fi->next_rr_cpu++; + if (sigcpu == KVM_MAX_VCPUS) + sigcpu = fi->next_rr_cpu = 0; + } while (kvm_get_vcpu(kvm, sigcpu) == NULL); + } + dst_vcpu = kvm_get_vcpu(kvm, sigcpu); + li = &dst_vcpu->arch.local_int; + spin_lock(&li->lock); + switch (type) { + case KVM_S390_MCHK: + atomic_set_mask(CPUSTAT_STOP_INT, li->cpuflags); + break; + case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX: + atomic_set_mask(CPUSTAT_IO_INT, li->cpuflags); + break; + default: + atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); + break; + } + spin_unlock(&li->lock); + kvm_s390_vcpu_wakeup(kvm_get_vcpu(kvm, sigcpu)); + return 0; + +} + +int kvm_s390_inject_vm(struct kvm *kvm, + struct kvm_s390_interrupt *s390int) +{ + struct kvm_s390_interrupt_info *inti; + int rc; + + inti = kzalloc(sizeof(*inti), GFP_KERNEL); + if (!inti) + return -ENOMEM; + + inti->type = s390int->type; + switch (inti->type) { + case KVM_S390_INT_VIRTIO: + VM_EVENT(kvm, 5, "inject: virtio parm:%x,parm64:%llx", + s390int->parm, s390int->parm64); + inti->ext.ext_params = s390int->parm; + inti->ext.ext_params2 = s390int->parm64; + break; + case KVM_S390_INT_SERVICE: + VM_EVENT(kvm, 5, "inject: sclp parm:%x", s390int->parm); + inti->ext.ext_params = s390int->parm; + break; + case KVM_S390_INT_PFAULT_DONE: + inti->ext.ext_params2 = s390int->parm64; + break; + case KVM_S390_MCHK: + VM_EVENT(kvm, 5, "inject: machine check parm64:%llx", + s390int->parm64); + inti->mchk.cr14 = s390int->parm; /* upper bits are not used */ + inti->mchk.mcic = s390int->parm64; + break; + case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX: + if (inti->type & IOINT_AI_MASK) + VM_EVENT(kvm, 5, "%s", "inject: I/O (AI)"); + else + VM_EVENT(kvm, 5, "inject: I/O css %x ss %x schid %04x", + s390int->type & IOINT_CSSID_MASK, + s390int->type & IOINT_SSID_MASK, + s390int->type & IOINT_SCHID_MASK); + inti->io.subchannel_id = s390int->parm >> 16; + inti->io.subchannel_nr = s390int->parm & 0x0000ffffu; + inti->io.io_int_parm = s390int->parm64 >> 32; + inti->io.io_int_word = s390int->parm64 & 0x00000000ffffffffull; + break; + default: + kfree(inti); + return -EINVAL; + } + trace_kvm_s390_inject_vm(s390int->type, s390int->parm, s390int->parm64, + 2); + + rc = __inject_vm(kvm, inti); + if (rc) + kfree(inti); + return rc; +} + +int kvm_s390_reinject_io_int(struct kvm *kvm, + struct kvm_s390_interrupt_info *inti) +{ + return __inject_vm(kvm, inti); +} + +int s390int_to_s390irq(struct kvm_s390_interrupt *s390int, + struct kvm_s390_irq *irq) +{ + irq->type = s390int->type; + switch (irq->type) { + case KVM_S390_PROGRAM_INT: + if (s390int->parm & 0xffff0000) + return -EINVAL; + irq->u.pgm.code = s390int->parm; + break; + case KVM_S390_SIGP_SET_PREFIX: + irq->u.prefix.address = s390int->parm; + break; + case KVM_S390_SIGP_STOP: + irq->u.stop.flags = s390int->parm; + break; + case KVM_S390_INT_EXTERNAL_CALL: + if (s390int->parm & 0xffff0000) + return -EINVAL; + irq->u.extcall.code = s390int->parm; + break; + case KVM_S390_INT_EMERGENCY: + if (s390int->parm & 0xffff0000) + return -EINVAL; + irq->u.emerg.code = s390int->parm; + break; + case KVM_S390_MCHK: + irq->u.mchk.mcic = s390int->parm64; + break; + } + return 0; +} + +int kvm_s390_is_stop_irq_pending(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + + return test_bit(IRQ_PEND_SIGP_STOP, &li->pending_irqs); +} + +void kvm_s390_clear_stop_irq(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + + spin_lock(&li->lock); + li->irq.stop.flags = 0; + clear_bit(IRQ_PEND_SIGP_STOP, &li->pending_irqs); + spin_unlock(&li->lock); +} + +static int do_inject_vcpu(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq) +{ + int rc; + + switch (irq->type) { + case KVM_S390_PROGRAM_INT: + VCPU_EVENT(vcpu, 3, "inject: program check %d (from user)", + irq->u.pgm.code); + rc = __inject_prog(vcpu, irq); + break; + case KVM_S390_SIGP_SET_PREFIX: + rc = __inject_set_prefix(vcpu, irq); + break; + case KVM_S390_SIGP_STOP: + rc = __inject_sigp_stop(vcpu, irq); + break; + case KVM_S390_RESTART: + rc = __inject_sigp_restart(vcpu, irq); + break; + case KVM_S390_INT_CLOCK_COMP: + rc = __inject_ckc(vcpu); + break; + case KVM_S390_INT_CPU_TIMER: + rc = __inject_cpu_timer(vcpu); + break; + case KVM_S390_INT_EXTERNAL_CALL: + rc = __inject_extcall(vcpu, irq); + break; + case KVM_S390_INT_EMERGENCY: + rc = __inject_sigp_emergency(vcpu, irq); + break; + case KVM_S390_MCHK: + rc = __inject_mchk(vcpu, irq); + break; + case KVM_S390_INT_PFAULT_INIT: + rc = __inject_pfault_init(vcpu, irq); + break; + case KVM_S390_INT_VIRTIO: + case KVM_S390_INT_SERVICE: + case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX: + default: + rc = -EINVAL; + } + + return rc; +} + +int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + int rc; + + spin_lock(&li->lock); + rc = do_inject_vcpu(vcpu, irq); + spin_unlock(&li->lock); + if (!rc) + kvm_s390_vcpu_wakeup(vcpu); + return rc; +} + +static inline void clear_irq_list(struct list_head *_list) +{ + struct kvm_s390_interrupt_info *inti, *n; + + list_for_each_entry_safe(inti, n, _list, list) { + list_del(&inti->list); + kfree(inti); + } +} + +static void inti_to_irq(struct kvm_s390_interrupt_info *inti, + struct kvm_s390_irq *irq) +{ + irq->type = inti->type; + switch (inti->type) { + case KVM_S390_INT_PFAULT_INIT: + case KVM_S390_INT_PFAULT_DONE: + case KVM_S390_INT_VIRTIO: + irq->u.ext = inti->ext; + break; + case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX: + irq->u.io = inti->io; + break; + } +} + +void kvm_s390_clear_float_irqs(struct kvm *kvm) +{ + struct kvm_s390_float_interrupt *fi = &kvm->arch.float_int; + int i; + + spin_lock(&fi->lock); + fi->pending_irqs = 0; + memset(&fi->srv_signal, 0, sizeof(fi->srv_signal)); + memset(&fi->mchk, 0, sizeof(fi->mchk)); + for (i = 0; i < FIRQ_LIST_COUNT; i++) + clear_irq_list(&fi->lists[i]); + for (i = 0; i < FIRQ_MAX_COUNT; i++) + fi->counters[i] = 0; + spin_unlock(&fi->lock); +}; + +static int get_all_floating_irqs(struct kvm *kvm, u8 __user *usrbuf, u64 len) +{ + struct kvm_s390_interrupt_info *inti; + struct kvm_s390_float_interrupt *fi; + struct kvm_s390_irq *buf; + struct kvm_s390_irq *irq; + int max_irqs; + int ret = 0; + int n = 0; + int i; + + if (len > KVM_S390_FLIC_MAX_BUFFER || len == 0) + return -EINVAL; + + /* + * We are already using -ENOMEM to signal + * userspace it may retry with a bigger buffer, + * so we need to use something else for this case + */ + buf = vzalloc(len); + if (!buf) + return -ENOBUFS; + + max_irqs = len / sizeof(struct kvm_s390_irq); + + fi = &kvm->arch.float_int; + spin_lock(&fi->lock); + for (i = 0; i < FIRQ_LIST_COUNT; i++) { + list_for_each_entry(inti, &fi->lists[i], list) { + if (n == max_irqs) { + /* signal userspace to try again */ + ret = -ENOMEM; + goto out; + } + inti_to_irq(inti, &buf[n]); + n++; + } + } + if (test_bit(IRQ_PEND_EXT_SERVICE, &fi->pending_irqs)) { + if (n == max_irqs) { + /* signal userspace to try again */ + ret = -ENOMEM; + goto out; + } + irq = (struct kvm_s390_irq *) &buf[n]; + irq->type = KVM_S390_INT_SERVICE; + irq->u.ext = fi->srv_signal; + n++; + } + if (test_bit(IRQ_PEND_MCHK_REP, &fi->pending_irqs)) { + if (n == max_irqs) { + /* signal userspace to try again */ + ret = -ENOMEM; + goto out; + } + irq = (struct kvm_s390_irq *) &buf[n]; + irq->type = KVM_S390_MCHK; + irq->u.mchk = fi->mchk; + n++; +} + +out: + spin_unlock(&fi->lock); + if (!ret && n > 0) { + if (copy_to_user(usrbuf, buf, sizeof(struct kvm_s390_irq) * n)) + ret = -EFAULT; + } + vfree(buf); + + return ret < 0 ? ret : n; +} + +static int flic_get_attr(struct kvm_device *dev, struct kvm_device_attr *attr) +{ + int r; + + switch (attr->group) { + case KVM_DEV_FLIC_GET_ALL_IRQS: + r = get_all_floating_irqs(dev->kvm, (u8 __user *) attr->addr, + attr->attr); + break; + default: + r = -EINVAL; + } + + return r; +} + +static inline int copy_irq_from_user(struct kvm_s390_interrupt_info *inti, + u64 addr) +{ + struct kvm_s390_irq __user *uptr = (struct kvm_s390_irq __user *) addr; + void *target = NULL; + void __user *source; + u64 size; + + if (get_user(inti->type, (u64 __user *)addr)) + return -EFAULT; + + switch (inti->type) { + case KVM_S390_INT_PFAULT_INIT: + case KVM_S390_INT_PFAULT_DONE: + case KVM_S390_INT_VIRTIO: + case KVM_S390_INT_SERVICE: + target = (void *) &inti->ext; + source = &uptr->u.ext; + size = sizeof(inti->ext); + break; + case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX: + target = (void *) &inti->io; + source = &uptr->u.io; + size = sizeof(inti->io); + break; + case KVM_S390_MCHK: + target = (void *) &inti->mchk; + source = &uptr->u.mchk; + size = sizeof(inti->mchk); + break; + default: + return -EINVAL; + } + + if (copy_from_user(target, source, size)) + return -EFAULT; + + return 0; +} + +static int enqueue_floating_irq(struct kvm_device *dev, + struct kvm_device_attr *attr) +{ + struct kvm_s390_interrupt_info *inti = NULL; + int r = 0; + int len = attr->attr; + + if (len % sizeof(struct kvm_s390_irq) != 0) + return -EINVAL; + else if (len > KVM_S390_FLIC_MAX_BUFFER) + return -EINVAL; + + while (len >= sizeof(struct kvm_s390_irq)) { + inti = kzalloc(sizeof(*inti), GFP_KERNEL); + if (!inti) + return -ENOMEM; + + r = copy_irq_from_user(inti, attr->addr); + if (r) { + kfree(inti); + return r; + } + r = __inject_vm(dev->kvm, inti); + if (r) { + kfree(inti); + return r; + } + len -= sizeof(struct kvm_s390_irq); + attr->addr += sizeof(struct kvm_s390_irq); + } + + return r; +} + +static struct s390_io_adapter *get_io_adapter(struct kvm *kvm, unsigned int id) +{ + if (id >= MAX_S390_IO_ADAPTERS) + return NULL; + return kvm->arch.adapters[id]; +} + +static int register_io_adapter(struct kvm_device *dev, + struct kvm_device_attr *attr) +{ + struct s390_io_adapter *adapter; + struct kvm_s390_io_adapter adapter_info; + + if (copy_from_user(&adapter_info, + (void __user *)attr->addr, sizeof(adapter_info))) + return -EFAULT; + + if ((adapter_info.id >= MAX_S390_IO_ADAPTERS) || + (dev->kvm->arch.adapters[adapter_info.id] != NULL)) + return -EINVAL; + + adapter = kzalloc(sizeof(*adapter), GFP_KERNEL); + if (!adapter) + return -ENOMEM; + + INIT_LIST_HEAD(&adapter->maps); + init_rwsem(&adapter->maps_lock); + atomic_set(&adapter->nr_maps, 0); + adapter->id = adapter_info.id; + adapter->isc = adapter_info.isc; + adapter->maskable = adapter_info.maskable; + adapter->masked = false; + adapter->swap = adapter_info.swap; + dev->kvm->arch.adapters[adapter->id] = adapter; + + return 0; +} + +int kvm_s390_mask_adapter(struct kvm *kvm, unsigned int id, bool masked) +{ + int ret; + struct s390_io_adapter *adapter = get_io_adapter(kvm, id); + + if (!adapter || !adapter->maskable) + return -EINVAL; + ret = adapter->masked; + adapter->masked = masked; + return ret; +} + +static int kvm_s390_adapter_map(struct kvm *kvm, unsigned int id, __u64 addr) +{ + struct s390_io_adapter *adapter = get_io_adapter(kvm, id); + struct s390_map_info *map; + int ret; + + if (!adapter || !addr) + return -EINVAL; + + map = kzalloc(sizeof(*map), GFP_KERNEL); + if (!map) { + ret = -ENOMEM; + goto out; + } + INIT_LIST_HEAD(&map->list); + map->guest_addr = addr; + map->addr = gmap_translate(kvm->arch.gmap, addr); + if (map->addr == -EFAULT) { + ret = -EFAULT; + goto out; + } + ret = get_user_pages_fast(map->addr, 1, 1, &map->page); + if (ret < 0) + goto out; + BUG_ON(ret != 1); + down_write(&adapter->maps_lock); + if (atomic_inc_return(&adapter->nr_maps) < MAX_S390_ADAPTER_MAPS) { + list_add_tail(&map->list, &adapter->maps); + ret = 0; + } else { + put_page(map->page); + ret = -EINVAL; + } + up_write(&adapter->maps_lock); +out: + if (ret) + kfree(map); + return ret; +} + +static int kvm_s390_adapter_unmap(struct kvm *kvm, unsigned int id, __u64 addr) +{ + struct s390_io_adapter *adapter = get_io_adapter(kvm, id); + struct s390_map_info *map, *tmp; + int found = 0; + + if (!adapter || !addr) + return -EINVAL; + + down_write(&adapter->maps_lock); + list_for_each_entry_safe(map, tmp, &adapter->maps, list) { + if (map->guest_addr == addr) { + found = 1; + atomic_dec(&adapter->nr_maps); + list_del(&map->list); + put_page(map->page); + kfree(map); + break; + } + } + up_write(&adapter->maps_lock); + + return found ? 0 : -EINVAL; +} + +void kvm_s390_destroy_adapters(struct kvm *kvm) +{ + int i; + struct s390_map_info *map, *tmp; + + for (i = 0; i < MAX_S390_IO_ADAPTERS; i++) { + if (!kvm->arch.adapters[i]) + continue; + list_for_each_entry_safe(map, tmp, + &kvm->arch.adapters[i]->maps, list) { + list_del(&map->list); + put_page(map->page); + kfree(map); + } + kfree(kvm->arch.adapters[i]); + } +} + +static int modify_io_adapter(struct kvm_device *dev, + struct kvm_device_attr *attr) +{ + struct kvm_s390_io_adapter_req req; + struct s390_io_adapter *adapter; + int ret; + + if (copy_from_user(&req, (void __user *)attr->addr, sizeof(req))) + return -EFAULT; + + adapter = get_io_adapter(dev->kvm, req.id); + if (!adapter) + return -EINVAL; + switch (req.type) { + case KVM_S390_IO_ADAPTER_MASK: + ret = kvm_s390_mask_adapter(dev->kvm, req.id, req.mask); + if (ret > 0) + ret = 0; + break; + case KVM_S390_IO_ADAPTER_MAP: + ret = kvm_s390_adapter_map(dev->kvm, req.id, req.addr); + break; + case KVM_S390_IO_ADAPTER_UNMAP: + ret = kvm_s390_adapter_unmap(dev->kvm, req.id, req.addr); + break; + default: + ret = -EINVAL; + } + + return ret; +} + +static int flic_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr) +{ + int r = 0; + unsigned int i; + struct kvm_vcpu *vcpu; + + switch (attr->group) { + case KVM_DEV_FLIC_ENQUEUE: + r = enqueue_floating_irq(dev, attr); + break; + case KVM_DEV_FLIC_CLEAR_IRQS: + kvm_s390_clear_float_irqs(dev->kvm); + break; + case KVM_DEV_FLIC_APF_ENABLE: + dev->kvm->arch.gmap->pfault_enabled = 1; + break; + case KVM_DEV_FLIC_APF_DISABLE_WAIT: + dev->kvm->arch.gmap->pfault_enabled = 0; + /* + * Make sure no async faults are in transition when + * clearing the queues. So we don't need to worry + * about late coming workers. + */ + synchronize_srcu(&dev->kvm->srcu); + kvm_for_each_vcpu(i, vcpu, dev->kvm) + kvm_clear_async_pf_completion_queue(vcpu); + break; + case KVM_DEV_FLIC_ADAPTER_REGISTER: + r = register_io_adapter(dev, attr); + break; + case KVM_DEV_FLIC_ADAPTER_MODIFY: + r = modify_io_adapter(dev, attr); + break; + default: + r = -EINVAL; + } + + return r; +} + +static int flic_create(struct kvm_device *dev, u32 type) +{ + if (!dev) + return -EINVAL; + if (dev->kvm->arch.flic) + return -EINVAL; + dev->kvm->arch.flic = dev; + return 0; +} + +static void flic_destroy(struct kvm_device *dev) +{ + dev->kvm->arch.flic = NULL; + kfree(dev); +} + +/* s390 floating irq controller (flic) */ +struct kvm_device_ops kvm_flic_ops = { + .name = "kvm-flic", + .get_attr = flic_get_attr, + .set_attr = flic_set_attr, + .create = flic_create, + .destroy = flic_destroy, +}; + +static unsigned long get_ind_bit(__u64 addr, unsigned long bit_nr, bool swap) +{ + unsigned long bit; + + bit = bit_nr + (addr % PAGE_SIZE) * 8; + + return swap ? (bit ^ (BITS_PER_LONG - 1)) : bit; +} + +static struct s390_map_info *get_map_info(struct s390_io_adapter *adapter, + u64 addr) +{ + struct s390_map_info *map; + + if (!adapter) + return NULL; + + list_for_each_entry(map, &adapter->maps, list) { + if (map->guest_addr == addr) + return map; + } + return NULL; +} + +static int adapter_indicators_set(struct kvm *kvm, + struct s390_io_adapter *adapter, + struct kvm_s390_adapter_int *adapter_int) +{ + unsigned long bit; + int summary_set, idx; + struct s390_map_info *info; + void *map; + + info = get_map_info(adapter, adapter_int->ind_addr); + if (!info) + return -1; + map = page_address(info->page); + bit = get_ind_bit(info->addr, adapter_int->ind_offset, adapter->swap); + set_bit(bit, map); + idx = srcu_read_lock(&kvm->srcu); + mark_page_dirty(kvm, info->guest_addr >> PAGE_SHIFT); + set_page_dirty_lock(info->page); + info = get_map_info(adapter, adapter_int->summary_addr); + if (!info) { + srcu_read_unlock(&kvm->srcu, idx); + return -1; + } + map = page_address(info->page); + bit = get_ind_bit(info->addr, adapter_int->summary_offset, + adapter->swap); + summary_set = test_and_set_bit(bit, map); + mark_page_dirty(kvm, info->guest_addr >> PAGE_SHIFT); + set_page_dirty_lock(info->page); + srcu_read_unlock(&kvm->srcu, idx); + return summary_set ? 0 : 1; +} + +/* + * < 0 - not injected due to error + * = 0 - coalesced, summary indicator already active + * > 0 - injected interrupt + */ +static int set_adapter_int(struct kvm_kernel_irq_routing_entry *e, + struct kvm *kvm, int irq_source_id, int level, + bool line_status) +{ + int ret; + struct s390_io_adapter *adapter; + + /* We're only interested in the 0->1 transition. */ + if (!level) + return 0; + adapter = get_io_adapter(kvm, e->adapter.adapter_id); + if (!adapter) + return -1; + down_read(&adapter->maps_lock); + ret = adapter_indicators_set(kvm, adapter, &e->adapter); + up_read(&adapter->maps_lock); + if ((ret > 0) && !adapter->masked) { + struct kvm_s390_interrupt s390int = { + .type = KVM_S390_INT_IO(1, 0, 0, 0), + .parm = 0, + .parm64 = (adapter->isc << 27) | 0x80000000, + }; + ret = kvm_s390_inject_vm(kvm, &s390int); + if (ret == 0) + ret = 1; + } + return ret; +} + +int kvm_set_routing_entry(struct kvm_kernel_irq_routing_entry *e, + const struct kvm_irq_routing_entry *ue) +{ + int ret; + + switch (ue->type) { + case KVM_IRQ_ROUTING_S390_ADAPTER: + e->set = set_adapter_int; + e->adapter.summary_addr = ue->u.adapter.summary_addr; + e->adapter.ind_addr = ue->u.adapter.ind_addr; + e->adapter.summary_offset = ue->u.adapter.summary_offset; + e->adapter.ind_offset = ue->u.adapter.ind_offset; + e->adapter.adapter_id = ue->u.adapter.adapter_id; + ret = 0; + break; + default: + ret = -EINVAL; + } + + return ret; +} + +int kvm_set_msi(struct kvm_kernel_irq_routing_entry *e, struct kvm *kvm, + int irq_source_id, int level, bool line_status) +{ + return -EINVAL; +} + +int kvm_s390_set_irq_state(struct kvm_vcpu *vcpu, void __user *irqstate, int len) +{ + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_irq *buf; + int r = 0; + int n; + + buf = vmalloc(len); + if (!buf) + return -ENOMEM; + + if (copy_from_user((void *) buf, irqstate, len)) { + r = -EFAULT; + goto out_free; + } + + /* + * Don't allow setting the interrupt state + * when there are already interrupts pending + */ + spin_lock(&li->lock); + if (li->pending_irqs) { + r = -EBUSY; + goto out_unlock; + } + + for (n = 0; n < len / sizeof(*buf); n++) { + r = do_inject_vcpu(vcpu, &buf[n]); + if (r) + break; + } + +out_unlock: + spin_unlock(&li->lock); +out_free: + vfree(buf); + + return r; +} + +static void store_local_irq(struct kvm_s390_local_interrupt *li, + struct kvm_s390_irq *irq, + unsigned long irq_type) +{ + switch (irq_type) { + case IRQ_PEND_MCHK_EX: + case IRQ_PEND_MCHK_REP: + irq->type = KVM_S390_MCHK; + irq->u.mchk = li->irq.mchk; + break; + case IRQ_PEND_PROG: + irq->type = KVM_S390_PROGRAM_INT; + irq->u.pgm = li->irq.pgm; + break; + case IRQ_PEND_PFAULT_INIT: + irq->type = KVM_S390_INT_PFAULT_INIT; + irq->u.ext = li->irq.ext; + break; + case IRQ_PEND_EXT_EXTERNAL: + irq->type = KVM_S390_INT_EXTERNAL_CALL; + irq->u.extcall = li->irq.extcall; + break; + case IRQ_PEND_EXT_CLOCK_COMP: + irq->type = KVM_S390_INT_CLOCK_COMP; + break; + case IRQ_PEND_EXT_CPU_TIMER: + irq->type = KVM_S390_INT_CPU_TIMER; + break; + case IRQ_PEND_SIGP_STOP: + irq->type = KVM_S390_SIGP_STOP; + irq->u.stop = li->irq.stop; + break; + case IRQ_PEND_RESTART: + irq->type = KVM_S390_RESTART; + break; + case IRQ_PEND_SET_PREFIX: + irq->type = KVM_S390_SIGP_SET_PREFIX; + irq->u.prefix = li->irq.prefix; + break; + } +} + +int kvm_s390_get_irq_state(struct kvm_vcpu *vcpu, __u8 __user *buf, int len) +{ + uint8_t sigp_ctrl = vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sigp_ctrl; + unsigned long sigp_emerg_pending[BITS_TO_LONGS(KVM_MAX_VCPUS)]; + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + unsigned long pending_irqs; + struct kvm_s390_irq irq; + unsigned long irq_type; + int cpuaddr; + int n = 0; + + spin_lock(&li->lock); + pending_irqs = li->pending_irqs; + memcpy(&sigp_emerg_pending, &li->sigp_emerg_pending, + sizeof(sigp_emerg_pending)); + spin_unlock(&li->lock); + + for_each_set_bit(irq_type, &pending_irqs, IRQ_PEND_COUNT) { + memset(&irq, 0, sizeof(irq)); + if (irq_type == IRQ_PEND_EXT_EMERGENCY) + continue; + if (n + sizeof(irq) > len) + return -ENOBUFS; + store_local_irq(&vcpu->arch.local_int, &irq, irq_type); + if (copy_to_user(&buf[n], &irq, sizeof(irq))) + return -EFAULT; + n += sizeof(irq); + } + + if (test_bit(IRQ_PEND_EXT_EMERGENCY, &pending_irqs)) { + for_each_set_bit(cpuaddr, sigp_emerg_pending, KVM_MAX_VCPUS) { + memset(&irq, 0, sizeof(irq)); + if (n + sizeof(irq) > len) + return -ENOBUFS; + irq.type = KVM_S390_INT_EMERGENCY; + irq.u.emerg.code = cpuaddr; + if (copy_to_user(&buf[n], &irq, sizeof(irq))) + return -EFAULT; + n += sizeof(irq); + } + } + + if ((sigp_ctrl & SIGP_CTRL_C) && + (atomic_read(&vcpu->arch.sie_block->cpuflags) & + CPUSTAT_ECALL_PEND)) { + if (n + sizeof(irq) > len) + return -ENOBUFS; + memset(&irq, 0, sizeof(irq)); + irq.type = KVM_S390_INT_EXTERNAL_CALL; + irq.u.extcall.code = sigp_ctrl & SIGP_CTRL_SCN_MASK; + if (copy_to_user(&buf[n], &irq, sizeof(irq))) + return -EFAULT; + n += sizeof(irq); + } + + return n; +} diff --git a/arch/s390/kvm/irq.h b/arch/s390/kvm/irq.h new file mode 100644 index 000000000..d98e41596 --- /dev/null +++ b/arch/s390/kvm/irq.h @@ -0,0 +1,22 @@ +/* + * s390 irqchip routines + * + * Copyright IBM Corp. 2014 + * + * 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. + * + * Author(s): Cornelia Huck <cornelia.huck@de.ibm.com> + */ +#ifndef __KVM_IRQ_H +#define __KVM_IRQ_H + +#include <linux/kvm_host.h> + +static inline int irqchip_in_kernel(struct kvm *kvm) +{ + return 1; +} + +#endif diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c new file mode 100644 index 000000000..8cd8e7b28 --- /dev/null +++ b/arch/s390/kvm/kvm-s390.c @@ -0,0 +1,2628 @@ +/* + * hosting zSeries kernel virtual machines + * + * Copyright IBM Corp. 2008, 2009 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + * Christian Borntraeger <borntraeger@de.ibm.com> + * Heiko Carstens <heiko.carstens@de.ibm.com> + * Christian Ehrhardt <ehrhardt@de.ibm.com> + * Jason J. Herne <jjherne@us.ibm.com> + */ + +#include <linux/compiler.h> +#include <linux/err.h> +#include <linux/fs.h> +#include <linux/hrtimer.h> +#include <linux/init.h> +#include <linux/kvm.h> +#include <linux/kvm_host.h> +#include <linux/module.h> +#include <linux/random.h> +#include <linux/slab.h> +#include <linux/timer.h> +#include <linux/vmalloc.h> +#include <asm/asm-offsets.h> +#include <asm/lowcore.h> +#include <asm/pgtable.h> +#include <asm/nmi.h> +#include <asm/switch_to.h> +#include <asm/isc.h> +#include <asm/sclp.h> +#include "kvm-s390.h" +#include "gaccess.h" + +#define CREATE_TRACE_POINTS +#include "trace.h" +#include "trace-s390.h" + +#define MEM_OP_MAX_SIZE 65536 /* Maximum transfer size for KVM_S390_MEM_OP */ +#define LOCAL_IRQS 32 +#define VCPU_IRQS_MAX_BUF (sizeof(struct kvm_s390_irq) * \ + (KVM_MAX_VCPUS + LOCAL_IRQS)) + +#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU + +struct kvm_stats_debugfs_item debugfs_entries[] = { + { "userspace_handled", VCPU_STAT(exit_userspace) }, + { "exit_null", VCPU_STAT(exit_null) }, + { "exit_validity", VCPU_STAT(exit_validity) }, + { "exit_stop_request", VCPU_STAT(exit_stop_request) }, + { "exit_external_request", VCPU_STAT(exit_external_request) }, + { "exit_external_interrupt", VCPU_STAT(exit_external_interrupt) }, + { "exit_instruction", VCPU_STAT(exit_instruction) }, + { "exit_program_interruption", VCPU_STAT(exit_program_interruption) }, + { "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) }, + { "halt_successful_poll", VCPU_STAT(halt_successful_poll) }, + { "halt_wakeup", VCPU_STAT(halt_wakeup) }, + { "instruction_lctlg", VCPU_STAT(instruction_lctlg) }, + { "instruction_lctl", VCPU_STAT(instruction_lctl) }, + { "instruction_stctl", VCPU_STAT(instruction_stctl) }, + { "instruction_stctg", VCPU_STAT(instruction_stctg) }, + { "deliver_emergency_signal", VCPU_STAT(deliver_emergency_signal) }, + { "deliver_external_call", VCPU_STAT(deliver_external_call) }, + { "deliver_service_signal", VCPU_STAT(deliver_service_signal) }, + { "deliver_virtio_interrupt", VCPU_STAT(deliver_virtio_interrupt) }, + { "deliver_stop_signal", VCPU_STAT(deliver_stop_signal) }, + { "deliver_prefix_signal", VCPU_STAT(deliver_prefix_signal) }, + { "deliver_restart_signal", VCPU_STAT(deliver_restart_signal) }, + { "deliver_program_interruption", VCPU_STAT(deliver_program_int) }, + { "exit_wait_state", VCPU_STAT(exit_wait_state) }, + { "instruction_pfmf", VCPU_STAT(instruction_pfmf) }, + { "instruction_stidp", VCPU_STAT(instruction_stidp) }, + { "instruction_spx", VCPU_STAT(instruction_spx) }, + { "instruction_stpx", VCPU_STAT(instruction_stpx) }, + { "instruction_stap", VCPU_STAT(instruction_stap) }, + { "instruction_storage_key", VCPU_STAT(instruction_storage_key) }, + { "instruction_ipte_interlock", VCPU_STAT(instruction_ipte_interlock) }, + { "instruction_stsch", VCPU_STAT(instruction_stsch) }, + { "instruction_chsc", VCPU_STAT(instruction_chsc) }, + { "instruction_essa", VCPU_STAT(instruction_essa) }, + { "instruction_stsi", VCPU_STAT(instruction_stsi) }, + { "instruction_stfl", VCPU_STAT(instruction_stfl) }, + { "instruction_tprot", VCPU_STAT(instruction_tprot) }, + { "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) }, + { "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) }, + { "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) }, + { "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) }, + { "instruction_sigp_cond_emergency", VCPU_STAT(instruction_sigp_cond_emergency) }, + { "instruction_sigp_start", VCPU_STAT(instruction_sigp_start) }, + { "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) }, + { "instruction_sigp_stop_store_status", VCPU_STAT(instruction_sigp_stop_store_status) }, + { "instruction_sigp_store_status", VCPU_STAT(instruction_sigp_store_status) }, + { "instruction_sigp_store_adtl_status", VCPU_STAT(instruction_sigp_store_adtl_status) }, + { "instruction_sigp_set_arch", VCPU_STAT(instruction_sigp_arch) }, + { "instruction_sigp_set_prefix", VCPU_STAT(instruction_sigp_prefix) }, + { "instruction_sigp_restart", VCPU_STAT(instruction_sigp_restart) }, + { "instruction_sigp_cpu_reset", VCPU_STAT(instruction_sigp_cpu_reset) }, + { "instruction_sigp_init_cpu_reset", VCPU_STAT(instruction_sigp_init_cpu_reset) }, + { "instruction_sigp_unknown", VCPU_STAT(instruction_sigp_unknown) }, + { "diagnose_10", VCPU_STAT(diagnose_10) }, + { "diagnose_44", VCPU_STAT(diagnose_44) }, + { "diagnose_9c", VCPU_STAT(diagnose_9c) }, + { NULL } +}; + +/* upper facilities limit for kvm */ +unsigned long kvm_s390_fac_list_mask[] = { + 0xffe6fffbfcfdfc40UL, + 0x005c800000000000UL, +}; + +unsigned long kvm_s390_fac_list_mask_size(void) +{ + BUILD_BUG_ON(ARRAY_SIZE(kvm_s390_fac_list_mask) > S390_ARCH_FAC_MASK_SIZE_U64); + return ARRAY_SIZE(kvm_s390_fac_list_mask); +} + +static struct gmap_notifier gmap_notifier; + +/* Section: not file related */ +int kvm_arch_hardware_enable(void) +{ + /* every s390 is virtualization enabled ;-) */ + return 0; +} + +static void kvm_gmap_notifier(struct gmap *gmap, unsigned long address); + +int kvm_arch_hardware_setup(void) +{ + gmap_notifier.notifier_call = kvm_gmap_notifier; + gmap_register_ipte_notifier(&gmap_notifier); + return 0; +} + +void kvm_arch_hardware_unsetup(void) +{ + gmap_unregister_ipte_notifier(&gmap_notifier); +} + +int kvm_arch_init(void *opaque) +{ + /* Register floating interrupt controller interface. */ + return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC); +} + +/* Section: device related */ +long kvm_arch_dev_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + if (ioctl == KVM_S390_ENABLE_SIE) + return s390_enable_sie(); + return -EINVAL; +} + +int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) +{ + int r; + + switch (ext) { + case KVM_CAP_S390_PSW: + case KVM_CAP_S390_GMAP: + case KVM_CAP_SYNC_MMU: +#ifdef CONFIG_KVM_S390_UCONTROL + case KVM_CAP_S390_UCONTROL: +#endif + case KVM_CAP_ASYNC_PF: + case KVM_CAP_SYNC_REGS: + case KVM_CAP_ONE_REG: + case KVM_CAP_ENABLE_CAP: + case KVM_CAP_S390_CSS_SUPPORT: + case KVM_CAP_IOEVENTFD: + case KVM_CAP_DEVICE_CTRL: + case KVM_CAP_ENABLE_CAP_VM: + case KVM_CAP_S390_IRQCHIP: + case KVM_CAP_VM_ATTRIBUTES: + case KVM_CAP_MP_STATE: + case KVM_CAP_S390_INJECT_IRQ: + case KVM_CAP_S390_USER_SIGP: + case KVM_CAP_S390_USER_STSI: + case KVM_CAP_S390_SKEYS: + case KVM_CAP_S390_IRQ_STATE: + r = 1; + break; + case KVM_CAP_S390_MEM_OP: + r = MEM_OP_MAX_SIZE; + break; + case KVM_CAP_NR_VCPUS: + case KVM_CAP_MAX_VCPUS: + r = KVM_MAX_VCPUS; + break; + case KVM_CAP_NR_MEMSLOTS: + r = KVM_USER_MEM_SLOTS; + break; + case KVM_CAP_S390_COW: + r = MACHINE_HAS_ESOP; + break; + case KVM_CAP_S390_VECTOR_REGISTERS: + r = MACHINE_HAS_VX; + break; + default: + r = 0; + } + return r; +} + +static void kvm_s390_sync_dirty_log(struct kvm *kvm, + struct kvm_memory_slot *memslot) +{ + gfn_t cur_gfn, last_gfn; + unsigned long address; + struct gmap *gmap = kvm->arch.gmap; + + down_read(&gmap->mm->mmap_sem); + /* Loop over all guest pages */ + last_gfn = memslot->base_gfn + memslot->npages; + for (cur_gfn = memslot->base_gfn; cur_gfn <= last_gfn; cur_gfn++) { + address = gfn_to_hva_memslot(memslot, cur_gfn); + + if (gmap_test_and_clear_dirty(address, gmap)) + mark_page_dirty(kvm, cur_gfn); + } + up_read(&gmap->mm->mmap_sem); +} + +/* Section: vm related */ +/* + * Get (and clear) the dirty memory log for a memory slot. + */ +int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, + struct kvm_dirty_log *log) +{ + int r; + unsigned long n; + struct kvm_memory_slot *memslot; + int is_dirty = 0; + + mutex_lock(&kvm->slots_lock); + + r = -EINVAL; + if (log->slot >= KVM_USER_MEM_SLOTS) + goto out; + + memslot = id_to_memslot(kvm->memslots, log->slot); + r = -ENOENT; + if (!memslot->dirty_bitmap) + goto out; + + kvm_s390_sync_dirty_log(kvm, memslot); + r = kvm_get_dirty_log(kvm, log, &is_dirty); + if (r) + goto out; + + /* Clear the dirty log */ + if (is_dirty) { + n = kvm_dirty_bitmap_bytes(memslot); + memset(memslot->dirty_bitmap, 0, n); + } + r = 0; +out: + mutex_unlock(&kvm->slots_lock); + return r; +} + +static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap) +{ + int r; + + if (cap->flags) + return -EINVAL; + + switch (cap->cap) { + case KVM_CAP_S390_IRQCHIP: + kvm->arch.use_irqchip = 1; + r = 0; + break; + case KVM_CAP_S390_USER_SIGP: + kvm->arch.user_sigp = 1; + r = 0; + break; + case KVM_CAP_S390_VECTOR_REGISTERS: + if (MACHINE_HAS_VX) { + set_kvm_facility(kvm->arch.model.fac->mask, 129); + set_kvm_facility(kvm->arch.model.fac->list, 129); + r = 0; + } else + r = -EINVAL; + break; + case KVM_CAP_S390_USER_STSI: + kvm->arch.user_stsi = 1; + r = 0; + break; + default: + r = -EINVAL; + break; + } + return r; +} + +static int kvm_s390_get_mem_control(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + + switch (attr->attr) { + case KVM_S390_VM_MEM_LIMIT_SIZE: + ret = 0; + if (put_user(kvm->arch.gmap->asce_end, (u64 __user *)attr->addr)) + ret = -EFAULT; + break; + default: + ret = -ENXIO; + break; + } + return ret; +} + +static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + unsigned int idx; + switch (attr->attr) { + case KVM_S390_VM_MEM_ENABLE_CMMA: + ret = -EBUSY; + mutex_lock(&kvm->lock); + if (atomic_read(&kvm->online_vcpus) == 0) { + kvm->arch.use_cmma = 1; + ret = 0; + } + mutex_unlock(&kvm->lock); + break; + case KVM_S390_VM_MEM_CLR_CMMA: + mutex_lock(&kvm->lock); + idx = srcu_read_lock(&kvm->srcu); + s390_reset_cmma(kvm->arch.gmap->mm); + srcu_read_unlock(&kvm->srcu, idx); + mutex_unlock(&kvm->lock); + ret = 0; + break; + case KVM_S390_VM_MEM_LIMIT_SIZE: { + unsigned long new_limit; + + if (kvm_is_ucontrol(kvm)) + return -EINVAL; + + if (get_user(new_limit, (u64 __user *)attr->addr)) + return -EFAULT; + + if (new_limit > kvm->arch.gmap->asce_end) + return -E2BIG; + + ret = -EBUSY; + mutex_lock(&kvm->lock); + if (atomic_read(&kvm->online_vcpus) == 0) { + /* gmap_alloc will round the limit up */ + struct gmap *new = gmap_alloc(current->mm, new_limit); + + if (!new) { + ret = -ENOMEM; + } else { + gmap_free(kvm->arch.gmap); + new->private = kvm; + kvm->arch.gmap = new; + ret = 0; + } + } + mutex_unlock(&kvm->lock); + break; + } + default: + ret = -ENXIO; + break; + } + return ret; +} + +static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu); + +static int kvm_s390_vm_set_crypto(struct kvm *kvm, struct kvm_device_attr *attr) +{ + struct kvm_vcpu *vcpu; + int i; + + if (!test_kvm_facility(kvm, 76)) + return -EINVAL; + + mutex_lock(&kvm->lock); + switch (attr->attr) { + case KVM_S390_VM_CRYPTO_ENABLE_AES_KW: + get_random_bytes( + kvm->arch.crypto.crycb->aes_wrapping_key_mask, + sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask)); + kvm->arch.crypto.aes_kw = 1; + break; + case KVM_S390_VM_CRYPTO_ENABLE_DEA_KW: + get_random_bytes( + kvm->arch.crypto.crycb->dea_wrapping_key_mask, + sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask)); + kvm->arch.crypto.dea_kw = 1; + break; + case KVM_S390_VM_CRYPTO_DISABLE_AES_KW: + kvm->arch.crypto.aes_kw = 0; + memset(kvm->arch.crypto.crycb->aes_wrapping_key_mask, 0, + sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask)); + break; + case KVM_S390_VM_CRYPTO_DISABLE_DEA_KW: + kvm->arch.crypto.dea_kw = 0; + memset(kvm->arch.crypto.crycb->dea_wrapping_key_mask, 0, + sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask)); + break; + default: + mutex_unlock(&kvm->lock); + return -ENXIO; + } + + kvm_for_each_vcpu(i, vcpu, kvm) { + kvm_s390_vcpu_crypto_setup(vcpu); + exit_sie(vcpu); + } + mutex_unlock(&kvm->lock); + return 0; +} + +static int kvm_s390_set_tod_high(struct kvm *kvm, struct kvm_device_attr *attr) +{ + u8 gtod_high; + + if (copy_from_user(>od_high, (void __user *)attr->addr, + sizeof(gtod_high))) + return -EFAULT; + + if (gtod_high != 0) + return -EINVAL; + + return 0; +} + +static int kvm_s390_set_tod_low(struct kvm *kvm, struct kvm_device_attr *attr) +{ + struct kvm_vcpu *cur_vcpu; + unsigned int vcpu_idx; + u64 host_tod, gtod; + int r; + + if (copy_from_user(>od, (void __user *)attr->addr, sizeof(gtod))) + return -EFAULT; + + r = store_tod_clock(&host_tod); + if (r) + return r; + + mutex_lock(&kvm->lock); + kvm->arch.epoch = gtod - host_tod; + kvm_for_each_vcpu(vcpu_idx, cur_vcpu, kvm) { + cur_vcpu->arch.sie_block->epoch = kvm->arch.epoch; + exit_sie(cur_vcpu); + } + mutex_unlock(&kvm->lock); + return 0; +} + +static int kvm_s390_set_tod(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + + if (attr->flags) + return -EINVAL; + + switch (attr->attr) { + case KVM_S390_VM_TOD_HIGH: + ret = kvm_s390_set_tod_high(kvm, attr); + break; + case KVM_S390_VM_TOD_LOW: + ret = kvm_s390_set_tod_low(kvm, attr); + break; + default: + ret = -ENXIO; + break; + } + return ret; +} + +static int kvm_s390_get_tod_high(struct kvm *kvm, struct kvm_device_attr *attr) +{ + u8 gtod_high = 0; + + if (copy_to_user((void __user *)attr->addr, >od_high, + sizeof(gtod_high))) + return -EFAULT; + + return 0; +} + +static int kvm_s390_get_tod_low(struct kvm *kvm, struct kvm_device_attr *attr) +{ + u64 host_tod, gtod; + int r; + + r = store_tod_clock(&host_tod); + if (r) + return r; + + gtod = host_tod + kvm->arch.epoch; + if (copy_to_user((void __user *)attr->addr, >od, sizeof(gtod))) + return -EFAULT; + + return 0; +} + +static int kvm_s390_get_tod(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + + if (attr->flags) + return -EINVAL; + + switch (attr->attr) { + case KVM_S390_VM_TOD_HIGH: + ret = kvm_s390_get_tod_high(kvm, attr); + break; + case KVM_S390_VM_TOD_LOW: + ret = kvm_s390_get_tod_low(kvm, attr); + break; + default: + ret = -ENXIO; + break; + } + return ret; +} + +static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr) +{ + struct kvm_s390_vm_cpu_processor *proc; + int ret = 0; + + mutex_lock(&kvm->lock); + if (atomic_read(&kvm->online_vcpus)) { + ret = -EBUSY; + goto out; + } + proc = kzalloc(sizeof(*proc), GFP_KERNEL); + if (!proc) { + ret = -ENOMEM; + goto out; + } + if (!copy_from_user(proc, (void __user *)attr->addr, + sizeof(*proc))) { + memcpy(&kvm->arch.model.cpu_id, &proc->cpuid, + sizeof(struct cpuid)); + kvm->arch.model.ibc = proc->ibc; + memcpy(kvm->arch.model.fac->list, proc->fac_list, + S390_ARCH_FAC_LIST_SIZE_BYTE); + } else + ret = -EFAULT; + kfree(proc); +out: + mutex_unlock(&kvm->lock); + return ret; +} + +static int kvm_s390_set_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret = -ENXIO; + + switch (attr->attr) { + case KVM_S390_VM_CPU_PROCESSOR: + ret = kvm_s390_set_processor(kvm, attr); + break; + } + return ret; +} + +static int kvm_s390_get_processor(struct kvm *kvm, struct kvm_device_attr *attr) +{ + struct kvm_s390_vm_cpu_processor *proc; + int ret = 0; + + proc = kzalloc(sizeof(*proc), GFP_KERNEL); + if (!proc) { + ret = -ENOMEM; + goto out; + } + memcpy(&proc->cpuid, &kvm->arch.model.cpu_id, sizeof(struct cpuid)); + proc->ibc = kvm->arch.model.ibc; + memcpy(&proc->fac_list, kvm->arch.model.fac->list, S390_ARCH_FAC_LIST_SIZE_BYTE); + if (copy_to_user((void __user *)attr->addr, proc, sizeof(*proc))) + ret = -EFAULT; + kfree(proc); +out: + return ret; +} + +static int kvm_s390_get_machine(struct kvm *kvm, struct kvm_device_attr *attr) +{ + struct kvm_s390_vm_cpu_machine *mach; + int ret = 0; + + mach = kzalloc(sizeof(*mach), GFP_KERNEL); + if (!mach) { + ret = -ENOMEM; + goto out; + } + get_cpu_id((struct cpuid *) &mach->cpuid); + mach->ibc = sclp_get_ibc(); + memcpy(&mach->fac_mask, kvm->arch.model.fac->mask, + S390_ARCH_FAC_LIST_SIZE_BYTE); + memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list, + S390_ARCH_FAC_LIST_SIZE_BYTE); + if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach))) + ret = -EFAULT; + kfree(mach); +out: + return ret; +} + +static int kvm_s390_get_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret = -ENXIO; + + switch (attr->attr) { + case KVM_S390_VM_CPU_PROCESSOR: + ret = kvm_s390_get_processor(kvm, attr); + break; + case KVM_S390_VM_CPU_MACHINE: + ret = kvm_s390_get_machine(kvm, attr); + break; + } + return ret; +} + +static int kvm_s390_vm_set_attr(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + + switch (attr->group) { + case KVM_S390_VM_MEM_CTRL: + ret = kvm_s390_set_mem_control(kvm, attr); + break; + case KVM_S390_VM_TOD: + ret = kvm_s390_set_tod(kvm, attr); + break; + case KVM_S390_VM_CPU_MODEL: + ret = kvm_s390_set_cpu_model(kvm, attr); + break; + case KVM_S390_VM_CRYPTO: + ret = kvm_s390_vm_set_crypto(kvm, attr); + break; + default: + ret = -ENXIO; + break; + } + + return ret; +} + +static int kvm_s390_vm_get_attr(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + + switch (attr->group) { + case KVM_S390_VM_MEM_CTRL: + ret = kvm_s390_get_mem_control(kvm, attr); + break; + case KVM_S390_VM_TOD: + ret = kvm_s390_get_tod(kvm, attr); + break; + case KVM_S390_VM_CPU_MODEL: + ret = kvm_s390_get_cpu_model(kvm, attr); + break; + default: + ret = -ENXIO; + break; + } + + return ret; +} + +static int kvm_s390_vm_has_attr(struct kvm *kvm, struct kvm_device_attr *attr) +{ + int ret; + + switch (attr->group) { + case KVM_S390_VM_MEM_CTRL: + switch (attr->attr) { + case KVM_S390_VM_MEM_ENABLE_CMMA: + case KVM_S390_VM_MEM_CLR_CMMA: + case KVM_S390_VM_MEM_LIMIT_SIZE: + ret = 0; + break; + default: + ret = -ENXIO; + break; + } + break; + case KVM_S390_VM_TOD: + switch (attr->attr) { + case KVM_S390_VM_TOD_LOW: + case KVM_S390_VM_TOD_HIGH: + ret = 0; + break; + default: + ret = -ENXIO; + break; + } + break; + case KVM_S390_VM_CPU_MODEL: + switch (attr->attr) { + case KVM_S390_VM_CPU_PROCESSOR: + case KVM_S390_VM_CPU_MACHINE: + ret = 0; + break; + default: + ret = -ENXIO; + break; + } + break; + case KVM_S390_VM_CRYPTO: + switch (attr->attr) { + case KVM_S390_VM_CRYPTO_ENABLE_AES_KW: + case KVM_S390_VM_CRYPTO_ENABLE_DEA_KW: + case KVM_S390_VM_CRYPTO_DISABLE_AES_KW: + case KVM_S390_VM_CRYPTO_DISABLE_DEA_KW: + ret = 0; + break; + default: + ret = -ENXIO; + break; + } + break; + default: + ret = -ENXIO; + break; + } + + return ret; +} + +static long kvm_s390_get_skeys(struct kvm *kvm, struct kvm_s390_skeys *args) +{ + uint8_t *keys; + uint64_t hva; + unsigned long curkey; + int i, r = 0; + + if (args->flags != 0) + return -EINVAL; + + /* Is this guest using storage keys? */ + if (!mm_use_skey(current->mm)) + return KVM_S390_GET_SKEYS_NONE; + + /* Enforce sane limit on memory allocation */ + if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX) + return -EINVAL; + + keys = kmalloc_array(args->count, sizeof(uint8_t), + GFP_KERNEL | __GFP_NOWARN); + if (!keys) + keys = vmalloc(sizeof(uint8_t) * args->count); + if (!keys) + return -ENOMEM; + + for (i = 0; i < args->count; i++) { + hva = gfn_to_hva(kvm, args->start_gfn + i); + if (kvm_is_error_hva(hva)) { + r = -EFAULT; + goto out; + } + + curkey = get_guest_storage_key(current->mm, hva); + if (IS_ERR_VALUE(curkey)) { + r = curkey; + goto out; + } + keys[i] = curkey; + } + + r = copy_to_user((uint8_t __user *)args->skeydata_addr, keys, + sizeof(uint8_t) * args->count); + if (r) + r = -EFAULT; +out: + kvfree(keys); + return r; +} + +static long kvm_s390_set_skeys(struct kvm *kvm, struct kvm_s390_skeys *args) +{ + uint8_t *keys; + uint64_t hva; + int i, r = 0; + + if (args->flags != 0) + return -EINVAL; + + /* Enforce sane limit on memory allocation */ + if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX) + return -EINVAL; + + keys = kmalloc_array(args->count, sizeof(uint8_t), + GFP_KERNEL | __GFP_NOWARN); + if (!keys) + keys = vmalloc(sizeof(uint8_t) * args->count); + if (!keys) + return -ENOMEM; + + r = copy_from_user(keys, (uint8_t __user *)args->skeydata_addr, + sizeof(uint8_t) * args->count); + if (r) { + r = -EFAULT; + goto out; + } + + /* Enable storage key handling for the guest */ + s390_enable_skey(); + + for (i = 0; i < args->count; i++) { + hva = gfn_to_hva(kvm, args->start_gfn + i); + if (kvm_is_error_hva(hva)) { + r = -EFAULT; + goto out; + } + + /* Lowest order bit is reserved */ + if (keys[i] & 0x01) { + r = -EINVAL; + goto out; + } + + r = set_guest_storage_key(current->mm, hva, + (unsigned long)keys[i], 0); + if (r) + goto out; + } +out: + kvfree(keys); + return r; +} + +long kvm_arch_vm_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + struct kvm *kvm = filp->private_data; + void __user *argp = (void __user *)arg; + struct kvm_device_attr attr; + int r; + + switch (ioctl) { + case KVM_S390_INTERRUPT: { + struct kvm_s390_interrupt s390int; + + r = -EFAULT; + if (copy_from_user(&s390int, argp, sizeof(s390int))) + break; + r = kvm_s390_inject_vm(kvm, &s390int); + break; + } + case KVM_ENABLE_CAP: { + struct kvm_enable_cap cap; + r = -EFAULT; + if (copy_from_user(&cap, argp, sizeof(cap))) + break; + r = kvm_vm_ioctl_enable_cap(kvm, &cap); + break; + } + case KVM_CREATE_IRQCHIP: { + struct kvm_irq_routing_entry routing; + + r = -EINVAL; + if (kvm->arch.use_irqchip) { + /* Set up dummy routing. */ + memset(&routing, 0, sizeof(routing)); + kvm_set_irq_routing(kvm, &routing, 0, 0); + r = 0; + } + break; + } + case KVM_SET_DEVICE_ATTR: { + r = -EFAULT; + if (copy_from_user(&attr, (void __user *)arg, sizeof(attr))) + break; + r = kvm_s390_vm_set_attr(kvm, &attr); + break; + } + case KVM_GET_DEVICE_ATTR: { + r = -EFAULT; + if (copy_from_user(&attr, (void __user *)arg, sizeof(attr))) + break; + r = kvm_s390_vm_get_attr(kvm, &attr); + break; + } + case KVM_HAS_DEVICE_ATTR: { + r = -EFAULT; + if (copy_from_user(&attr, (void __user *)arg, sizeof(attr))) + break; + r = kvm_s390_vm_has_attr(kvm, &attr); + break; + } + case KVM_S390_GET_SKEYS: { + struct kvm_s390_skeys args; + + r = -EFAULT; + if (copy_from_user(&args, argp, + sizeof(struct kvm_s390_skeys))) + break; + r = kvm_s390_get_skeys(kvm, &args); + break; + } + case KVM_S390_SET_SKEYS: { + struct kvm_s390_skeys args; + + r = -EFAULT; + if (copy_from_user(&args, argp, + sizeof(struct kvm_s390_skeys))) + break; + r = kvm_s390_set_skeys(kvm, &args); + break; + } + default: + r = -ENOTTY; + } + + return r; +} + +static int kvm_s390_query_ap_config(u8 *config) +{ + u32 fcn_code = 0x04000000UL; + u32 cc = 0; + + memset(config, 0, 128); + asm volatile( + "lgr 0,%1\n" + "lgr 2,%2\n" + ".long 0xb2af0000\n" /* PQAP(QCI) */ + "0: ipm %0\n" + "srl %0,28\n" + "1:\n" + EX_TABLE(0b, 1b) + : "+r" (cc) + : "r" (fcn_code), "r" (config) + : "cc", "0", "2", "memory" + ); + + return cc; +} + +static int kvm_s390_apxa_installed(void) +{ + u8 config[128]; + int cc; + + if (test_facility(2) && test_facility(12)) { + cc = kvm_s390_query_ap_config(config); + + if (cc) + pr_err("PQAP(QCI) failed with cc=%d", cc); + else + return config[0] & 0x40; + } + + return 0; +} + +static void kvm_s390_set_crycb_format(struct kvm *kvm) +{ + kvm->arch.crypto.crycbd = (__u32)(unsigned long) kvm->arch.crypto.crycb; + + if (kvm_s390_apxa_installed()) + kvm->arch.crypto.crycbd |= CRYCB_FORMAT2; + else + kvm->arch.crypto.crycbd |= CRYCB_FORMAT1; +} + +static void kvm_s390_get_cpu_id(struct cpuid *cpu_id) +{ + get_cpu_id(cpu_id); + cpu_id->version = 0xff; +} + +static int kvm_s390_crypto_init(struct kvm *kvm) +{ + if (!test_kvm_facility(kvm, 76)) + return 0; + + kvm->arch.crypto.crycb = kzalloc(sizeof(*kvm->arch.crypto.crycb), + GFP_KERNEL | GFP_DMA); + if (!kvm->arch.crypto.crycb) + return -ENOMEM; + + kvm_s390_set_crycb_format(kvm); + + /* Enable AES/DEA protected key functions by default */ + kvm->arch.crypto.aes_kw = 1; + kvm->arch.crypto.dea_kw = 1; + get_random_bytes(kvm->arch.crypto.crycb->aes_wrapping_key_mask, + sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask)); + get_random_bytes(kvm->arch.crypto.crycb->dea_wrapping_key_mask, + sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask)); + + return 0; +} + +int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) +{ + int i, rc; + char debug_name[16]; + static unsigned long sca_offset; + + rc = -EINVAL; +#ifdef CONFIG_KVM_S390_UCONTROL + if (type & ~KVM_VM_S390_UCONTROL) + goto out_err; + if ((type & KVM_VM_S390_UCONTROL) && (!capable(CAP_SYS_ADMIN))) + goto out_err; +#else + if (type) + goto out_err; +#endif + + rc = s390_enable_sie(); + if (rc) + goto out_err; + + rc = -ENOMEM; + + kvm->arch.sca = (struct sca_block *) get_zeroed_page(GFP_KERNEL); + if (!kvm->arch.sca) + goto out_err; + spin_lock(&kvm_lock); + sca_offset = (sca_offset + 16) & 0x7f0; + kvm->arch.sca = (struct sca_block *) ((char *) kvm->arch.sca + sca_offset); + spin_unlock(&kvm_lock); + + sprintf(debug_name, "kvm-%u", current->pid); + + kvm->arch.dbf = debug_register(debug_name, 8, 2, 8 * sizeof(long)); + if (!kvm->arch.dbf) + goto out_err; + + /* + * The architectural maximum amount of facilities is 16 kbit. To store + * this amount, 2 kbyte of memory is required. Thus we need a full + * page to hold the guest facility list (arch.model.fac->list) and the + * facility mask (arch.model.fac->mask). Its address size has to be + * 31 bits and word aligned. + */ + kvm->arch.model.fac = + (struct kvm_s390_fac *) get_zeroed_page(GFP_KERNEL | GFP_DMA); + if (!kvm->arch.model.fac) + goto out_err; + + /* Populate the facility mask initially. */ + memcpy(kvm->arch.model.fac->mask, S390_lowcore.stfle_fac_list, + S390_ARCH_FAC_LIST_SIZE_BYTE); + for (i = 0; i < S390_ARCH_FAC_LIST_SIZE_U64; i++) { + if (i < kvm_s390_fac_list_mask_size()) + kvm->arch.model.fac->mask[i] &= kvm_s390_fac_list_mask[i]; + else + kvm->arch.model.fac->mask[i] = 0UL; + } + + /* Populate the facility list initially. */ + memcpy(kvm->arch.model.fac->list, kvm->arch.model.fac->mask, + S390_ARCH_FAC_LIST_SIZE_BYTE); + + kvm_s390_get_cpu_id(&kvm->arch.model.cpu_id); + kvm->arch.model.ibc = sclp_get_ibc() & 0x0fff; + + if (kvm_s390_crypto_init(kvm) < 0) + goto out_err; + + spin_lock_init(&kvm->arch.float_int.lock); + for (i = 0; i < FIRQ_LIST_COUNT; i++) + INIT_LIST_HEAD(&kvm->arch.float_int.lists[i]); + init_waitqueue_head(&kvm->arch.ipte_wq); + mutex_init(&kvm->arch.ipte_mutex); + + debug_register_view(kvm->arch.dbf, &debug_sprintf_view); + VM_EVENT(kvm, 3, "%s", "vm created"); + + if (type & KVM_VM_S390_UCONTROL) { + kvm->arch.gmap = NULL; + } else { + kvm->arch.gmap = gmap_alloc(current->mm, (1UL << 44) - 1); + if (!kvm->arch.gmap) + goto out_err; + kvm->arch.gmap->private = kvm; + kvm->arch.gmap->pfault_enabled = 0; + } + + kvm->arch.css_support = 0; + kvm->arch.use_irqchip = 0; + kvm->arch.epoch = 0; + + spin_lock_init(&kvm->arch.start_stop_lock); + + return 0; +out_err: + kfree(kvm->arch.crypto.crycb); + free_page((unsigned long)kvm->arch.model.fac); + debug_unregister(kvm->arch.dbf); + free_page((unsigned long)(kvm->arch.sca)); + return rc; +} + +void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) +{ + VCPU_EVENT(vcpu, 3, "%s", "free cpu"); + trace_kvm_s390_destroy_vcpu(vcpu->vcpu_id); + kvm_s390_clear_local_irqs(vcpu); + kvm_clear_async_pf_completion_queue(vcpu); + if (!kvm_is_ucontrol(vcpu->kvm)) { + clear_bit(63 - vcpu->vcpu_id, + (unsigned long *) &vcpu->kvm->arch.sca->mcn); + if (vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sda == + (__u64) vcpu->arch.sie_block) + vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sda = 0; + } + smp_mb(); + + if (kvm_is_ucontrol(vcpu->kvm)) + gmap_free(vcpu->arch.gmap); + + if (kvm_s390_cmma_enabled(vcpu->kvm)) + kvm_s390_vcpu_unsetup_cmma(vcpu); + free_page((unsigned long)(vcpu->arch.sie_block)); + + kvm_vcpu_uninit(vcpu); + kmem_cache_free(kvm_vcpu_cache, vcpu); +} + +static void kvm_free_vcpus(struct kvm *kvm) +{ + unsigned int i; + struct kvm_vcpu *vcpu; + + kvm_for_each_vcpu(i, vcpu, kvm) + kvm_arch_vcpu_destroy(vcpu); + + mutex_lock(&kvm->lock); + for (i = 0; i < atomic_read(&kvm->online_vcpus); i++) + kvm->vcpus[i] = NULL; + + atomic_set(&kvm->online_vcpus, 0); + mutex_unlock(&kvm->lock); +} + +void kvm_arch_destroy_vm(struct kvm *kvm) +{ + kvm_free_vcpus(kvm); + free_page((unsigned long)kvm->arch.model.fac); + free_page((unsigned long)(kvm->arch.sca)); + debug_unregister(kvm->arch.dbf); + kfree(kvm->arch.crypto.crycb); + if (!kvm_is_ucontrol(kvm)) + gmap_free(kvm->arch.gmap); + kvm_s390_destroy_adapters(kvm); + kvm_s390_clear_float_irqs(kvm); +} + +/* Section: vcpu related */ +static int __kvm_ucontrol_vcpu_init(struct kvm_vcpu *vcpu) +{ + vcpu->arch.gmap = gmap_alloc(current->mm, -1UL); + if (!vcpu->arch.gmap) + return -ENOMEM; + vcpu->arch.gmap->private = vcpu->kvm; + + return 0; +} + +int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) +{ + vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID; + kvm_clear_async_pf_completion_queue(vcpu); + vcpu->run->kvm_valid_regs = KVM_SYNC_PREFIX | + KVM_SYNC_GPRS | + KVM_SYNC_ACRS | + KVM_SYNC_CRS | + KVM_SYNC_ARCH0 | + KVM_SYNC_PFAULT; + if (test_kvm_facility(vcpu->kvm, 129)) + vcpu->run->kvm_valid_regs |= KVM_SYNC_VRS; + + if (kvm_is_ucontrol(vcpu->kvm)) + return __kvm_ucontrol_vcpu_init(vcpu); + + return 0; +} + +void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) +{ + save_fp_ctl(&vcpu->arch.host_fpregs.fpc); + if (test_kvm_facility(vcpu->kvm, 129)) + save_vx_regs((__vector128 *)&vcpu->arch.host_vregs->vrs); + else + save_fp_regs(vcpu->arch.host_fpregs.fprs); + save_access_regs(vcpu->arch.host_acrs); + if (test_kvm_facility(vcpu->kvm, 129)) { + restore_fp_ctl(&vcpu->run->s.regs.fpc); + restore_vx_regs((__vector128 *)&vcpu->run->s.regs.vrs); + } else { + restore_fp_ctl(&vcpu->arch.guest_fpregs.fpc); + restore_fp_regs(vcpu->arch.guest_fpregs.fprs); + } + restore_access_regs(vcpu->run->s.regs.acrs); + gmap_enable(vcpu->arch.gmap); + atomic_set_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags); +} + +void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) +{ + atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags); + gmap_disable(vcpu->arch.gmap); + if (test_kvm_facility(vcpu->kvm, 129)) { + save_fp_ctl(&vcpu->run->s.regs.fpc); + save_vx_regs((__vector128 *)&vcpu->run->s.regs.vrs); + } else { + save_fp_ctl(&vcpu->arch.guest_fpregs.fpc); + save_fp_regs(vcpu->arch.guest_fpregs.fprs); + } + save_access_regs(vcpu->run->s.regs.acrs); + restore_fp_ctl(&vcpu->arch.host_fpregs.fpc); + if (test_kvm_facility(vcpu->kvm, 129)) + restore_vx_regs((__vector128 *)&vcpu->arch.host_vregs->vrs); + else + restore_fp_regs(vcpu->arch.host_fpregs.fprs); + restore_access_regs(vcpu->arch.host_acrs); +} + +static void kvm_s390_vcpu_initial_reset(struct kvm_vcpu *vcpu) +{ + /* this equals initial cpu reset in pop, but we don't switch to ESA */ + vcpu->arch.sie_block->gpsw.mask = 0UL; + vcpu->arch.sie_block->gpsw.addr = 0UL; + kvm_s390_set_prefix(vcpu, 0); + vcpu->arch.sie_block->cputm = 0UL; + vcpu->arch.sie_block->ckc = 0UL; + vcpu->arch.sie_block->todpr = 0; + memset(vcpu->arch.sie_block->gcr, 0, 16 * sizeof(__u64)); + vcpu->arch.sie_block->gcr[0] = 0xE0UL; + vcpu->arch.sie_block->gcr[14] = 0xC2000000UL; + vcpu->arch.guest_fpregs.fpc = 0; + asm volatile("lfpc %0" : : "Q" (vcpu->arch.guest_fpregs.fpc)); + vcpu->arch.sie_block->gbea = 1; + vcpu->arch.sie_block->pp = 0; + vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID; + kvm_clear_async_pf_completion_queue(vcpu); + if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) + kvm_s390_vcpu_stop(vcpu); + kvm_s390_clear_local_irqs(vcpu); +} + +void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu) +{ + mutex_lock(&vcpu->kvm->lock); + vcpu->arch.sie_block->epoch = vcpu->kvm->arch.epoch; + mutex_unlock(&vcpu->kvm->lock); + if (!kvm_is_ucontrol(vcpu->kvm)) + vcpu->arch.gmap = vcpu->kvm->arch.gmap; +} + +static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu) +{ + if (!test_kvm_facility(vcpu->kvm, 76)) + return; + + vcpu->arch.sie_block->ecb3 &= ~(ECB3_AES | ECB3_DEA); + + if (vcpu->kvm->arch.crypto.aes_kw) + vcpu->arch.sie_block->ecb3 |= ECB3_AES; + if (vcpu->kvm->arch.crypto.dea_kw) + vcpu->arch.sie_block->ecb3 |= ECB3_DEA; + + vcpu->arch.sie_block->crycbd = vcpu->kvm->arch.crypto.crycbd; +} + +void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu) +{ + free_page(vcpu->arch.sie_block->cbrlo); + vcpu->arch.sie_block->cbrlo = 0; +} + +int kvm_s390_vcpu_setup_cmma(struct kvm_vcpu *vcpu) +{ + vcpu->arch.sie_block->cbrlo = get_zeroed_page(GFP_KERNEL); + if (!vcpu->arch.sie_block->cbrlo) + return -ENOMEM; + + vcpu->arch.sie_block->ecb2 |= 0x80; + vcpu->arch.sie_block->ecb2 &= ~0x08; + return 0; +} + +static void kvm_s390_vcpu_setup_model(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_cpu_model *model = &vcpu->kvm->arch.model; + + vcpu->arch.cpu_id = model->cpu_id; + vcpu->arch.sie_block->ibc = model->ibc; + vcpu->arch.sie_block->fac = (int) (long) model->fac->list; +} + +int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) +{ + int rc = 0; + + atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH | + CPUSTAT_SM | + CPUSTAT_STOPPED | + CPUSTAT_GED); + kvm_s390_vcpu_setup_model(vcpu); + + vcpu->arch.sie_block->ecb = 6; + if (test_kvm_facility(vcpu->kvm, 50) && test_kvm_facility(vcpu->kvm, 73)) + vcpu->arch.sie_block->ecb |= 0x10; + + vcpu->arch.sie_block->ecb2 = 8; + vcpu->arch.sie_block->eca = 0xC1002000U; + if (sclp_has_siif()) + vcpu->arch.sie_block->eca |= 1; + if (sclp_has_sigpif()) + vcpu->arch.sie_block->eca |= 0x10000000U; + if (test_kvm_facility(vcpu->kvm, 129)) { + vcpu->arch.sie_block->eca |= 0x00020000; + vcpu->arch.sie_block->ecd |= 0x20000000; + } + vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE; + + if (kvm_s390_cmma_enabled(vcpu->kvm)) { + rc = kvm_s390_vcpu_setup_cmma(vcpu); + if (rc) + return rc; + } + hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup; + + kvm_s390_vcpu_crypto_setup(vcpu); + + return rc; +} + +struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, + unsigned int id) +{ + struct kvm_vcpu *vcpu; + struct sie_page *sie_page; + int rc = -EINVAL; + + if (id >= KVM_MAX_VCPUS) + goto out; + + rc = -ENOMEM; + + vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL); + if (!vcpu) + goto out; + + sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL); + if (!sie_page) + goto out_free_cpu; + + vcpu->arch.sie_block = &sie_page->sie_block; + vcpu->arch.sie_block->itdba = (unsigned long) &sie_page->itdb; + vcpu->arch.host_vregs = &sie_page->vregs; + + vcpu->arch.sie_block->icpua = id; + if (!kvm_is_ucontrol(kvm)) { + if (!kvm->arch.sca) { + WARN_ON_ONCE(1); + goto out_free_cpu; + } + if (!kvm->arch.sca->cpu[id].sda) + kvm->arch.sca->cpu[id].sda = + (__u64) vcpu->arch.sie_block; + vcpu->arch.sie_block->scaoh = + (__u32)(((__u64)kvm->arch.sca) >> 32); + vcpu->arch.sie_block->scaol = (__u32)(__u64)kvm->arch.sca; + set_bit(63 - id, (unsigned long *) &kvm->arch.sca->mcn); + } + + spin_lock_init(&vcpu->arch.local_int.lock); + vcpu->arch.local_int.float_int = &kvm->arch.float_int; + vcpu->arch.local_int.wq = &vcpu->wq; + vcpu->arch.local_int.cpuflags = &vcpu->arch.sie_block->cpuflags; + + rc = kvm_vcpu_init(vcpu, kvm, id); + if (rc) + goto out_free_sie_block; + VM_EVENT(kvm, 3, "create cpu %d at %p, sie block at %p", id, vcpu, + vcpu->arch.sie_block); + trace_kvm_s390_create_vcpu(id, vcpu, vcpu->arch.sie_block); + + return vcpu; +out_free_sie_block: + free_page((unsigned long)(vcpu->arch.sie_block)); +out_free_cpu: + kmem_cache_free(kvm_vcpu_cache, vcpu); +out: + return ERR_PTR(rc); +} + +int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu) +{ + return kvm_s390_vcpu_has_irq(vcpu, 0); +} + +void s390_vcpu_block(struct kvm_vcpu *vcpu) +{ + atomic_set_mask(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20); +} + +void s390_vcpu_unblock(struct kvm_vcpu *vcpu) +{ + atomic_clear_mask(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20); +} + +/* + * Kick a guest cpu out of SIE and wait until SIE is not running. + * If the CPU is not running (e.g. waiting as idle) the function will + * return immediately. */ +void exit_sie(struct kvm_vcpu *vcpu) +{ + atomic_set_mask(CPUSTAT_STOP_INT, &vcpu->arch.sie_block->cpuflags); + while (vcpu->arch.sie_block->prog0c & PROG_IN_SIE) + cpu_relax(); +} + +/* Kick a guest cpu out of SIE and prevent SIE-reentry */ +void exit_sie_sync(struct kvm_vcpu *vcpu) +{ + s390_vcpu_block(vcpu); + exit_sie(vcpu); +} + +static void kvm_gmap_notifier(struct gmap *gmap, unsigned long address) +{ + int i; + struct kvm *kvm = gmap->private; + struct kvm_vcpu *vcpu; + + kvm_for_each_vcpu(i, vcpu, kvm) { + /* match against both prefix pages */ + if (kvm_s390_get_prefix(vcpu) == (address & ~0x1000UL)) { + VCPU_EVENT(vcpu, 2, "gmap notifier for %lx", address); + kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu); + exit_sie_sync(vcpu); + } + } +} + +int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu) +{ + /* kvm common code refers to this, but never calls it */ + BUG(); + return 0; +} + +static int kvm_arch_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu, + struct kvm_one_reg *reg) +{ + int r = -EINVAL; + + switch (reg->id) { + case KVM_REG_S390_TODPR: + r = put_user(vcpu->arch.sie_block->todpr, + (u32 __user *)reg->addr); + break; + case KVM_REG_S390_EPOCHDIFF: + r = put_user(vcpu->arch.sie_block->epoch, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_CPU_TIMER: + r = put_user(vcpu->arch.sie_block->cputm, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_CLOCK_COMP: + r = put_user(vcpu->arch.sie_block->ckc, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_PFTOKEN: + r = put_user(vcpu->arch.pfault_token, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_PFCOMPARE: + r = put_user(vcpu->arch.pfault_compare, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_PFSELECT: + r = put_user(vcpu->arch.pfault_select, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_PP: + r = put_user(vcpu->arch.sie_block->pp, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_GBEA: + r = put_user(vcpu->arch.sie_block->gbea, + (u64 __user *)reg->addr); + break; + default: + break; + } + + return r; +} + +static int kvm_arch_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu, + struct kvm_one_reg *reg) +{ + int r = -EINVAL; + + switch (reg->id) { + case KVM_REG_S390_TODPR: + r = get_user(vcpu->arch.sie_block->todpr, + (u32 __user *)reg->addr); + break; + case KVM_REG_S390_EPOCHDIFF: + r = get_user(vcpu->arch.sie_block->epoch, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_CPU_TIMER: + r = get_user(vcpu->arch.sie_block->cputm, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_CLOCK_COMP: + r = get_user(vcpu->arch.sie_block->ckc, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_PFTOKEN: + r = get_user(vcpu->arch.pfault_token, + (u64 __user *)reg->addr); + if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID) + kvm_clear_async_pf_completion_queue(vcpu); + break; + case KVM_REG_S390_PFCOMPARE: + r = get_user(vcpu->arch.pfault_compare, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_PFSELECT: + r = get_user(vcpu->arch.pfault_select, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_PP: + r = get_user(vcpu->arch.sie_block->pp, + (u64 __user *)reg->addr); + break; + case KVM_REG_S390_GBEA: + r = get_user(vcpu->arch.sie_block->gbea, + (u64 __user *)reg->addr); + break; + default: + break; + } + + return r; +} + +static int kvm_arch_vcpu_ioctl_initial_reset(struct kvm_vcpu *vcpu) +{ + kvm_s390_vcpu_initial_reset(vcpu); + return 0; +} + +int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) +{ + memcpy(&vcpu->run->s.regs.gprs, ®s->gprs, sizeof(regs->gprs)); + return 0; +} + +int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) +{ + memcpy(®s->gprs, &vcpu->run->s.regs.gprs, sizeof(regs->gprs)); + return 0; +} + +int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, + struct kvm_sregs *sregs) +{ + memcpy(&vcpu->run->s.regs.acrs, &sregs->acrs, sizeof(sregs->acrs)); + memcpy(&vcpu->arch.sie_block->gcr, &sregs->crs, sizeof(sregs->crs)); + restore_access_regs(vcpu->run->s.regs.acrs); + return 0; +} + +int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu, + struct kvm_sregs *sregs) +{ + memcpy(&sregs->acrs, &vcpu->run->s.regs.acrs, sizeof(sregs->acrs)); + memcpy(&sregs->crs, &vcpu->arch.sie_block->gcr, sizeof(sregs->crs)); + return 0; +} + +int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) +{ + if (test_fp_ctl(fpu->fpc)) + return -EINVAL; + memcpy(&vcpu->arch.guest_fpregs.fprs, &fpu->fprs, sizeof(fpu->fprs)); + vcpu->arch.guest_fpregs.fpc = fpu->fpc; + restore_fp_ctl(&vcpu->arch.guest_fpregs.fpc); + restore_fp_regs(vcpu->arch.guest_fpregs.fprs); + return 0; +} + +int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) +{ + memcpy(&fpu->fprs, &vcpu->arch.guest_fpregs.fprs, sizeof(fpu->fprs)); + fpu->fpc = vcpu->arch.guest_fpregs.fpc; + return 0; +} + +static int kvm_arch_vcpu_ioctl_set_initial_psw(struct kvm_vcpu *vcpu, psw_t psw) +{ + int rc = 0; + + if (!is_vcpu_stopped(vcpu)) + rc = -EBUSY; + else { + vcpu->run->psw_mask = psw.mask; + vcpu->run->psw_addr = psw.addr; + } + return rc; +} + +int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu, + struct kvm_translation *tr) +{ + return -EINVAL; /* not implemented yet */ +} + +#define VALID_GUESTDBG_FLAGS (KVM_GUESTDBG_SINGLESTEP | \ + KVM_GUESTDBG_USE_HW_BP | \ + KVM_GUESTDBG_ENABLE) + +int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, + struct kvm_guest_debug *dbg) +{ + int rc = 0; + + vcpu->guest_debug = 0; + kvm_s390_clear_bp_data(vcpu); + + if (dbg->control & ~VALID_GUESTDBG_FLAGS) + return -EINVAL; + + if (dbg->control & KVM_GUESTDBG_ENABLE) { + vcpu->guest_debug = dbg->control; + /* enforce guest PER */ + atomic_set_mask(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags); + + if (dbg->control & KVM_GUESTDBG_USE_HW_BP) + rc = kvm_s390_import_bp_data(vcpu, dbg); + } else { + atomic_clear_mask(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags); + vcpu->arch.guestdbg.last_bp = 0; + } + + if (rc) { + vcpu->guest_debug = 0; + kvm_s390_clear_bp_data(vcpu); + atomic_clear_mask(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags); + } + + return rc; +} + +int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + /* CHECK_STOP and LOAD are not supported yet */ + return is_vcpu_stopped(vcpu) ? KVM_MP_STATE_STOPPED : + KVM_MP_STATE_OPERATING; +} + +int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + int rc = 0; + + /* user space knows about this interface - let it control the state */ + vcpu->kvm->arch.user_cpu_state_ctrl = 1; + + switch (mp_state->mp_state) { + case KVM_MP_STATE_STOPPED: + kvm_s390_vcpu_stop(vcpu); + break; + case KVM_MP_STATE_OPERATING: + kvm_s390_vcpu_start(vcpu); + break; + case KVM_MP_STATE_LOAD: + case KVM_MP_STATE_CHECK_STOP: + /* fall through - CHECK_STOP and LOAD are not supported yet */ + default: + rc = -ENXIO; + } + + return rc; +} + +bool kvm_s390_cmma_enabled(struct kvm *kvm) +{ + if (!MACHINE_IS_LPAR) + return false; + /* only enable for z10 and later */ + if (!MACHINE_HAS_EDAT1) + return false; + if (!kvm->arch.use_cmma) + return false; + return true; +} + +static bool ibs_enabled(struct kvm_vcpu *vcpu) +{ + return atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_IBS; +} + +static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu) +{ +retry: + s390_vcpu_unblock(vcpu); + /* + * We use MMU_RELOAD just to re-arm the ipte notifier for the + * guest prefix page. gmap_ipte_notify will wait on the ptl lock. + * This ensures that the ipte instruction for this request has + * already finished. We might race against a second unmapper that + * wants to set the blocking bit. Lets just retry the request loop. + */ + if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) { + int rc; + rc = gmap_ipte_notify(vcpu->arch.gmap, + kvm_s390_get_prefix(vcpu), + PAGE_SIZE * 2); + if (rc) + return rc; + goto retry; + } + + if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) { + vcpu->arch.sie_block->ihcpu = 0xffff; + goto retry; + } + + if (kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu)) { + if (!ibs_enabled(vcpu)) { + trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 1); + atomic_set_mask(CPUSTAT_IBS, + &vcpu->arch.sie_block->cpuflags); + } + goto retry; + } + + if (kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu)) { + if (ibs_enabled(vcpu)) { + trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 0); + atomic_clear_mask(CPUSTAT_IBS, + &vcpu->arch.sie_block->cpuflags); + } + goto retry; + } + + /* nothing to do, just clear the request */ + clear_bit(KVM_REQ_UNHALT, &vcpu->requests); + + return 0; +} + +/** + * kvm_arch_fault_in_page - fault-in guest page if necessary + * @vcpu: The corresponding virtual cpu + * @gpa: Guest physical address + * @writable: Whether the page should be writable or not + * + * Make sure that a guest page has been faulted-in on the host. + * + * Return: Zero on success, negative error code otherwise. + */ +long kvm_arch_fault_in_page(struct kvm_vcpu *vcpu, gpa_t gpa, int writable) +{ + return gmap_fault(vcpu->arch.gmap, gpa, + writable ? FAULT_FLAG_WRITE : 0); +} + +static void __kvm_inject_pfault_token(struct kvm_vcpu *vcpu, bool start_token, + unsigned long token) +{ + struct kvm_s390_interrupt inti; + struct kvm_s390_irq irq; + + if (start_token) { + irq.u.ext.ext_params2 = token; + irq.type = KVM_S390_INT_PFAULT_INIT; + WARN_ON_ONCE(kvm_s390_inject_vcpu(vcpu, &irq)); + } else { + inti.type = KVM_S390_INT_PFAULT_DONE; + inti.parm64 = token; + WARN_ON_ONCE(kvm_s390_inject_vm(vcpu->kvm, &inti)); + } +} + +void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu, + struct kvm_async_pf *work) +{ + trace_kvm_s390_pfault_init(vcpu, work->arch.pfault_token); + __kvm_inject_pfault_token(vcpu, true, work->arch.pfault_token); +} + +void kvm_arch_async_page_present(struct kvm_vcpu *vcpu, + struct kvm_async_pf *work) +{ + trace_kvm_s390_pfault_done(vcpu, work->arch.pfault_token); + __kvm_inject_pfault_token(vcpu, false, work->arch.pfault_token); +} + +void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, + struct kvm_async_pf *work) +{ + /* s390 will always inject the page directly */ +} + +bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu) +{ + /* + * s390 will always inject the page directly, + * but we still want check_async_completion to cleanup + */ + return true; +} + +static int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu) +{ + hva_t hva; + struct kvm_arch_async_pf arch; + int rc; + + if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID) + return 0; + if ((vcpu->arch.sie_block->gpsw.mask & vcpu->arch.pfault_select) != + vcpu->arch.pfault_compare) + return 0; + if (psw_extint_disabled(vcpu)) + return 0; + if (kvm_s390_vcpu_has_irq(vcpu, 0)) + return 0; + if (!(vcpu->arch.sie_block->gcr[0] & 0x200ul)) + return 0; + if (!vcpu->arch.gmap->pfault_enabled) + return 0; + + hva = gfn_to_hva(vcpu->kvm, gpa_to_gfn(current->thread.gmap_addr)); + hva += current->thread.gmap_addr & ~PAGE_MASK; + if (read_guest_real(vcpu, vcpu->arch.pfault_token, &arch.pfault_token, 8)) + return 0; + + rc = kvm_setup_async_pf(vcpu, current->thread.gmap_addr, hva, &arch); + return rc; +} + +static int vcpu_pre_run(struct kvm_vcpu *vcpu) +{ + int rc, cpuflags; + + /* + * On s390 notifications for arriving pages will be delivered directly + * to the guest but the house keeping for completed pfaults is + * handled outside the worker. + */ + kvm_check_async_pf_completion(vcpu); + + memcpy(&vcpu->arch.sie_block->gg14, &vcpu->run->s.regs.gprs[14], 16); + + if (need_resched()) + schedule(); + + if (test_cpu_flag(CIF_MCCK_PENDING)) + s390_handle_mcck(); + + if (!kvm_is_ucontrol(vcpu->kvm)) { + rc = kvm_s390_deliver_pending_interrupts(vcpu); + if (rc) + return rc; + } + + rc = kvm_s390_handle_requests(vcpu); + if (rc) + return rc; + + if (guestdbg_enabled(vcpu)) { + kvm_s390_backup_guest_per_regs(vcpu); + kvm_s390_patch_guest_per_regs(vcpu); + } + + vcpu->arch.sie_block->icptcode = 0; + cpuflags = atomic_read(&vcpu->arch.sie_block->cpuflags); + VCPU_EVENT(vcpu, 6, "entering sie flags %x", cpuflags); + trace_kvm_s390_sie_enter(vcpu, cpuflags); + + return 0; +} + +static int vcpu_post_run_fault_in_sie(struct kvm_vcpu *vcpu) +{ + psw_t *psw = &vcpu->arch.sie_block->gpsw; + u8 opcode; + int rc; + + VCPU_EVENT(vcpu, 3, "%s", "fault in sie instruction"); + trace_kvm_s390_sie_fault(vcpu); + + /* + * We want to inject an addressing exception, which is defined as a + * suppressing or terminating exception. However, since we came here + * by a DAT access exception, the PSW still points to the faulting + * instruction since DAT exceptions are nullifying. So we've got + * to look up the current opcode to get the length of the instruction + * to be able to forward the PSW. + */ + rc = read_guest(vcpu, psw->addr, 0, &opcode, 1); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + psw->addr = __rewind_psw(*psw, -insn_length(opcode)); + + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); +} + +static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) +{ + int rc = -1; + + VCPU_EVENT(vcpu, 6, "exit sie icptcode %d", + vcpu->arch.sie_block->icptcode); + trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode); + + if (guestdbg_enabled(vcpu)) + kvm_s390_restore_guest_per_regs(vcpu); + + if (exit_reason >= 0) { + rc = 0; + } else if (kvm_is_ucontrol(vcpu->kvm)) { + vcpu->run->exit_reason = KVM_EXIT_S390_UCONTROL; + vcpu->run->s390_ucontrol.trans_exc_code = + current->thread.gmap_addr; + vcpu->run->s390_ucontrol.pgm_code = 0x10; + rc = -EREMOTE; + + } else if (current->thread.gmap_pfault) { + trace_kvm_s390_major_guest_pfault(vcpu); + current->thread.gmap_pfault = 0; + if (kvm_arch_setup_async_pf(vcpu)) { + rc = 0; + } else { + gpa_t gpa = current->thread.gmap_addr; + rc = kvm_arch_fault_in_page(vcpu, gpa, 1); + } + } + + if (rc == -1) + rc = vcpu_post_run_fault_in_sie(vcpu); + + memcpy(&vcpu->run->s.regs.gprs[14], &vcpu->arch.sie_block->gg14, 16); + + if (rc == 0) { + if (kvm_is_ucontrol(vcpu->kvm)) + /* Don't exit for host interrupts. */ + rc = vcpu->arch.sie_block->icptcode ? -EOPNOTSUPP : 0; + else + rc = kvm_handle_sie_intercept(vcpu); + } + + return rc; +} + +static int __vcpu_run(struct kvm_vcpu *vcpu) +{ + int rc, exit_reason; + + /* + * We try to hold kvm->srcu during most of vcpu_run (except when run- + * ning the guest), so that memslots (and other stuff) are protected + */ + vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); + + do { + rc = vcpu_pre_run(vcpu); + if (rc) + break; + + srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); + /* + * As PF_VCPU will be used in fault handler, between + * guest_enter and guest_exit should be no uaccess. + */ + preempt_disable(); + kvm_guest_enter(); + preempt_enable(); + exit_reason = sie64a(vcpu->arch.sie_block, + vcpu->run->s.regs.gprs); + kvm_guest_exit(); + vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); + + rc = vcpu_post_run(vcpu, exit_reason); + } while (!signal_pending(current) && !guestdbg_exit_pending(vcpu) && !rc); + + srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); + return rc; +} + +static void sync_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + vcpu->arch.sie_block->gpsw.mask = kvm_run->psw_mask; + vcpu->arch.sie_block->gpsw.addr = kvm_run->psw_addr; + if (kvm_run->kvm_dirty_regs & KVM_SYNC_PREFIX) + kvm_s390_set_prefix(vcpu, kvm_run->s.regs.prefix); + if (kvm_run->kvm_dirty_regs & KVM_SYNC_CRS) { + memcpy(&vcpu->arch.sie_block->gcr, &kvm_run->s.regs.crs, 128); + /* some control register changes require a tlb flush */ + kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu); + } + if (kvm_run->kvm_dirty_regs & KVM_SYNC_ARCH0) { + vcpu->arch.sie_block->cputm = kvm_run->s.regs.cputm; + vcpu->arch.sie_block->ckc = kvm_run->s.regs.ckc; + vcpu->arch.sie_block->todpr = kvm_run->s.regs.todpr; + vcpu->arch.sie_block->pp = kvm_run->s.regs.pp; + vcpu->arch.sie_block->gbea = kvm_run->s.regs.gbea; + } + if (kvm_run->kvm_dirty_regs & KVM_SYNC_PFAULT) { + vcpu->arch.pfault_token = kvm_run->s.regs.pft; + vcpu->arch.pfault_select = kvm_run->s.regs.pfs; + vcpu->arch.pfault_compare = kvm_run->s.regs.pfc; + if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID) + kvm_clear_async_pf_completion_queue(vcpu); + } + kvm_run->kvm_dirty_regs = 0; +} + +static void store_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + kvm_run->psw_mask = vcpu->arch.sie_block->gpsw.mask; + kvm_run->psw_addr = vcpu->arch.sie_block->gpsw.addr; + kvm_run->s.regs.prefix = kvm_s390_get_prefix(vcpu); + memcpy(&kvm_run->s.regs.crs, &vcpu->arch.sie_block->gcr, 128); + kvm_run->s.regs.cputm = vcpu->arch.sie_block->cputm; + kvm_run->s.regs.ckc = vcpu->arch.sie_block->ckc; + kvm_run->s.regs.todpr = vcpu->arch.sie_block->todpr; + kvm_run->s.regs.pp = vcpu->arch.sie_block->pp; + kvm_run->s.regs.gbea = vcpu->arch.sie_block->gbea; + kvm_run->s.regs.pft = vcpu->arch.pfault_token; + kvm_run->s.regs.pfs = vcpu->arch.pfault_select; + kvm_run->s.regs.pfc = vcpu->arch.pfault_compare; +} + +int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + int rc; + sigset_t sigsaved; + + if (guestdbg_exit_pending(vcpu)) { + kvm_s390_prepare_debug_exit(vcpu); + return 0; + } + + if (vcpu->sigset_active) + sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved); + + if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) { + kvm_s390_vcpu_start(vcpu); + } else if (is_vcpu_stopped(vcpu)) { + pr_err_ratelimited("kvm-s390: can't run stopped vcpu %d\n", + vcpu->vcpu_id); + return -EINVAL; + } + + sync_regs(vcpu, kvm_run); + + might_fault(); + rc = __vcpu_run(vcpu); + + if (signal_pending(current) && !rc) { + kvm_run->exit_reason = KVM_EXIT_INTR; + rc = -EINTR; + } + + if (guestdbg_exit_pending(vcpu) && !rc) { + kvm_s390_prepare_debug_exit(vcpu); + rc = 0; + } + + if (rc == -EOPNOTSUPP) { + /* intercept cannot be handled in-kernel, prepare kvm-run */ + kvm_run->exit_reason = KVM_EXIT_S390_SIEIC; + kvm_run->s390_sieic.icptcode = vcpu->arch.sie_block->icptcode; + kvm_run->s390_sieic.ipa = vcpu->arch.sie_block->ipa; + kvm_run->s390_sieic.ipb = vcpu->arch.sie_block->ipb; + rc = 0; + } + + if (rc == -EREMOTE) { + /* intercept was handled, but userspace support is needed + * kvm_run has been prepared by the handler */ + rc = 0; + } + + store_regs(vcpu, kvm_run); + + if (vcpu->sigset_active) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + + vcpu->stat.exit_userspace++; + return rc; +} + +/* + * store status at address + * we use have two special cases: + * KVM_S390_STORE_STATUS_NOADDR: -> 0x1200 on 64 bit + * KVM_S390_STORE_STATUS_PREFIXED: -> prefix + */ +int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long gpa) +{ + unsigned char archmode = 1; + unsigned int px; + u64 clkcomp; + int rc; + + if (gpa == KVM_S390_STORE_STATUS_NOADDR) { + if (write_guest_abs(vcpu, 163, &archmode, 1)) + return -EFAULT; + gpa = SAVE_AREA_BASE; + } else if (gpa == KVM_S390_STORE_STATUS_PREFIXED) { + if (write_guest_real(vcpu, 163, &archmode, 1)) + return -EFAULT; + gpa = kvm_s390_real_to_abs(vcpu, SAVE_AREA_BASE); + } + rc = write_guest_abs(vcpu, gpa + offsetof(struct save_area, fp_regs), + vcpu->arch.guest_fpregs.fprs, 128); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, gp_regs), + vcpu->run->s.regs.gprs, 128); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, psw), + &vcpu->arch.sie_block->gpsw, 16); + px = kvm_s390_get_prefix(vcpu); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, pref_reg), + &px, 4); + rc |= write_guest_abs(vcpu, + gpa + offsetof(struct save_area, fp_ctrl_reg), + &vcpu->arch.guest_fpregs.fpc, 4); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, tod_reg), + &vcpu->arch.sie_block->todpr, 4); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, timer), + &vcpu->arch.sie_block->cputm, 8); + clkcomp = vcpu->arch.sie_block->ckc >> 8; + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, clk_cmp), + &clkcomp, 8); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, acc_regs), + &vcpu->run->s.regs.acrs, 64); + rc |= write_guest_abs(vcpu, gpa + offsetof(struct save_area, ctrl_regs), + &vcpu->arch.sie_block->gcr, 128); + return rc ? -EFAULT : 0; +} + +int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr) +{ + /* + * The guest FPRS and ACRS are in the host FPRS/ACRS due to the lazy + * copying in vcpu load/put. Lets update our copies before we save + * it into the save area + */ + save_fp_ctl(&vcpu->arch.guest_fpregs.fpc); + save_fp_regs(vcpu->arch.guest_fpregs.fprs); + save_access_regs(vcpu->run->s.regs.acrs); + + return kvm_s390_store_status_unloaded(vcpu, addr); +} + +/* + * store additional status at address + */ +int kvm_s390_store_adtl_status_unloaded(struct kvm_vcpu *vcpu, + unsigned long gpa) +{ + /* Only bits 0-53 are used for address formation */ + if (!(gpa & ~0x3ff)) + return 0; + + return write_guest_abs(vcpu, gpa & ~0x3ff, + (void *)&vcpu->run->s.regs.vrs, 512); +} + +int kvm_s390_vcpu_store_adtl_status(struct kvm_vcpu *vcpu, unsigned long addr) +{ + if (!test_kvm_facility(vcpu->kvm, 129)) + return 0; + + /* + * The guest VXRS are in the host VXRs due to the lazy + * copying in vcpu load/put. Let's update our copies before we save + * it into the save area. + */ + save_vx_regs((__vector128 *)&vcpu->run->s.regs.vrs); + + return kvm_s390_store_adtl_status_unloaded(vcpu, addr); +} + +static void __disable_ibs_on_vcpu(struct kvm_vcpu *vcpu) +{ + kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu); + kvm_make_request(KVM_REQ_DISABLE_IBS, vcpu); + exit_sie_sync(vcpu); +} + +static void __disable_ibs_on_all_vcpus(struct kvm *kvm) +{ + unsigned int i; + struct kvm_vcpu *vcpu; + + kvm_for_each_vcpu(i, vcpu, kvm) { + __disable_ibs_on_vcpu(vcpu); + } +} + +static void __enable_ibs_on_vcpu(struct kvm_vcpu *vcpu) +{ + kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu); + kvm_make_request(KVM_REQ_ENABLE_IBS, vcpu); + exit_sie_sync(vcpu); +} + +void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu) +{ + int i, online_vcpus, started_vcpus = 0; + + if (!is_vcpu_stopped(vcpu)) + return; + + trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 1); + /* Only one cpu at a time may enter/leave the STOPPED state. */ + spin_lock(&vcpu->kvm->arch.start_stop_lock); + online_vcpus = atomic_read(&vcpu->kvm->online_vcpus); + + for (i = 0; i < online_vcpus; i++) { + if (!is_vcpu_stopped(vcpu->kvm->vcpus[i])) + started_vcpus++; + } + + if (started_vcpus == 0) { + /* we're the only active VCPU -> speed it up */ + __enable_ibs_on_vcpu(vcpu); + } else if (started_vcpus == 1) { + /* + * As we are starting a second VCPU, we have to disable + * the IBS facility on all VCPUs to remove potentially + * oustanding ENABLE requests. + */ + __disable_ibs_on_all_vcpus(vcpu->kvm); + } + + atomic_clear_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags); + /* + * Another VCPU might have used IBS while we were offline. + * Let's play safe and flush the VCPU at startup. + */ + kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu); + spin_unlock(&vcpu->kvm->arch.start_stop_lock); + return; +} + +void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu) +{ + int i, online_vcpus, started_vcpus = 0; + struct kvm_vcpu *started_vcpu = NULL; + + if (is_vcpu_stopped(vcpu)) + return; + + trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 0); + /* Only one cpu at a time may enter/leave the STOPPED state. */ + spin_lock(&vcpu->kvm->arch.start_stop_lock); + online_vcpus = atomic_read(&vcpu->kvm->online_vcpus); + + /* SIGP STOP and SIGP STOP AND STORE STATUS has been fully processed */ + kvm_s390_clear_stop_irq(vcpu); + + atomic_set_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags); + __disable_ibs_on_vcpu(vcpu); + + for (i = 0; i < online_vcpus; i++) { + if (!is_vcpu_stopped(vcpu->kvm->vcpus[i])) { + started_vcpus++; + started_vcpu = vcpu->kvm->vcpus[i]; + } + } + + if (started_vcpus == 1) { + /* + * As we only have one VCPU left, we want to enable the + * IBS facility for that VCPU to speed it up. + */ + __enable_ibs_on_vcpu(started_vcpu); + } + + spin_unlock(&vcpu->kvm->arch.start_stop_lock); + return; +} + +static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu, + struct kvm_enable_cap *cap) +{ + int r; + + if (cap->flags) + return -EINVAL; + + switch (cap->cap) { + case KVM_CAP_S390_CSS_SUPPORT: + if (!vcpu->kvm->arch.css_support) { + vcpu->kvm->arch.css_support = 1; + trace_kvm_s390_enable_css(vcpu->kvm); + } + r = 0; + break; + default: + r = -EINVAL; + break; + } + return r; +} + +static long kvm_s390_guest_mem_op(struct kvm_vcpu *vcpu, + struct kvm_s390_mem_op *mop) +{ + void __user *uaddr = (void __user *)mop->buf; + void *tmpbuf = NULL; + int r, srcu_idx; + const u64 supported_flags = KVM_S390_MEMOP_F_INJECT_EXCEPTION + | KVM_S390_MEMOP_F_CHECK_ONLY; + + if (mop->flags & ~supported_flags) + return -EINVAL; + + if (mop->size > MEM_OP_MAX_SIZE) + return -E2BIG; + + if (!(mop->flags & KVM_S390_MEMOP_F_CHECK_ONLY)) { + tmpbuf = vmalloc(mop->size); + if (!tmpbuf) + return -ENOMEM; + } + + srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); + + switch (mop->op) { + case KVM_S390_MEMOP_LOGICAL_READ: + if (mop->flags & KVM_S390_MEMOP_F_CHECK_ONLY) { + r = check_gva_range(vcpu, mop->gaddr, mop->ar, mop->size, false); + break; + } + r = read_guest(vcpu, mop->gaddr, mop->ar, tmpbuf, mop->size); + if (r == 0) { + if (copy_to_user(uaddr, tmpbuf, mop->size)) + r = -EFAULT; + } + break; + case KVM_S390_MEMOP_LOGICAL_WRITE: + if (mop->flags & KVM_S390_MEMOP_F_CHECK_ONLY) { + r = check_gva_range(vcpu, mop->gaddr, mop->ar, mop->size, true); + break; + } + if (copy_from_user(tmpbuf, uaddr, mop->size)) { + r = -EFAULT; + break; + } + r = write_guest(vcpu, mop->gaddr, mop->ar, tmpbuf, mop->size); + break; + default: + r = -EINVAL; + } + + srcu_read_unlock(&vcpu->kvm->srcu, srcu_idx); + + if (r > 0 && (mop->flags & KVM_S390_MEMOP_F_INJECT_EXCEPTION) != 0) + kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm); + + vfree(tmpbuf); + return r; +} + +long kvm_arch_vcpu_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + struct kvm_vcpu *vcpu = filp->private_data; + void __user *argp = (void __user *)arg; + int idx; + long r; + + switch (ioctl) { + case KVM_S390_IRQ: { + struct kvm_s390_irq s390irq; + + r = -EFAULT; + if (copy_from_user(&s390irq, argp, sizeof(s390irq))) + break; + r = kvm_s390_inject_vcpu(vcpu, &s390irq); + break; + } + case KVM_S390_INTERRUPT: { + struct kvm_s390_interrupt s390int; + struct kvm_s390_irq s390irq; + + r = -EFAULT; + if (copy_from_user(&s390int, argp, sizeof(s390int))) + break; + if (s390int_to_s390irq(&s390int, &s390irq)) + return -EINVAL; + r = kvm_s390_inject_vcpu(vcpu, &s390irq); + break; + } + case KVM_S390_STORE_STATUS: + idx = srcu_read_lock(&vcpu->kvm->srcu); + r = kvm_s390_vcpu_store_status(vcpu, arg); + srcu_read_unlock(&vcpu->kvm->srcu, idx); + break; + case KVM_S390_SET_INITIAL_PSW: { + psw_t psw; + + r = -EFAULT; + if (copy_from_user(&psw, argp, sizeof(psw))) + break; + r = kvm_arch_vcpu_ioctl_set_initial_psw(vcpu, psw); + break; + } + case KVM_S390_INITIAL_RESET: + r = kvm_arch_vcpu_ioctl_initial_reset(vcpu); + break; + case KVM_SET_ONE_REG: + case KVM_GET_ONE_REG: { + struct kvm_one_reg reg; + r = -EFAULT; + if (copy_from_user(®, argp, sizeof(reg))) + break; + if (ioctl == KVM_SET_ONE_REG) + r = kvm_arch_vcpu_ioctl_set_one_reg(vcpu, ®); + else + r = kvm_arch_vcpu_ioctl_get_one_reg(vcpu, ®); + break; + } +#ifdef CONFIG_KVM_S390_UCONTROL + case KVM_S390_UCAS_MAP: { + struct kvm_s390_ucas_mapping ucasmap; + + if (copy_from_user(&ucasmap, argp, sizeof(ucasmap))) { + r = -EFAULT; + break; + } + + if (!kvm_is_ucontrol(vcpu->kvm)) { + r = -EINVAL; + break; + } + + r = gmap_map_segment(vcpu->arch.gmap, ucasmap.user_addr, + ucasmap.vcpu_addr, ucasmap.length); + break; + } + case KVM_S390_UCAS_UNMAP: { + struct kvm_s390_ucas_mapping ucasmap; + + if (copy_from_user(&ucasmap, argp, sizeof(ucasmap))) { + r = -EFAULT; + break; + } + + if (!kvm_is_ucontrol(vcpu->kvm)) { + r = -EINVAL; + break; + } + + r = gmap_unmap_segment(vcpu->arch.gmap, ucasmap.vcpu_addr, + ucasmap.length); + break; + } +#endif + case KVM_S390_VCPU_FAULT: { + r = gmap_fault(vcpu->arch.gmap, arg, 0); + break; + } + case KVM_ENABLE_CAP: + { + struct kvm_enable_cap cap; + r = -EFAULT; + if (copy_from_user(&cap, argp, sizeof(cap))) + break; + r = kvm_vcpu_ioctl_enable_cap(vcpu, &cap); + break; + } + case KVM_S390_MEM_OP: { + struct kvm_s390_mem_op mem_op; + + if (copy_from_user(&mem_op, argp, sizeof(mem_op)) == 0) + r = kvm_s390_guest_mem_op(vcpu, &mem_op); + else + r = -EFAULT; + break; + } + case KVM_S390_SET_IRQ_STATE: { + struct kvm_s390_irq_state irq_state; + + r = -EFAULT; + if (copy_from_user(&irq_state, argp, sizeof(irq_state))) + break; + if (irq_state.len > VCPU_IRQS_MAX_BUF || + irq_state.len == 0 || + irq_state.len % sizeof(struct kvm_s390_irq) > 0) { + r = -EINVAL; + break; + } + r = kvm_s390_set_irq_state(vcpu, + (void __user *) irq_state.buf, + irq_state.len); + break; + } + case KVM_S390_GET_IRQ_STATE: { + struct kvm_s390_irq_state irq_state; + + r = -EFAULT; + if (copy_from_user(&irq_state, argp, sizeof(irq_state))) + break; + if (irq_state.len == 0) { + r = -EINVAL; + break; + } + r = kvm_s390_get_irq_state(vcpu, + (__u8 __user *) irq_state.buf, + irq_state.len); + break; + } + default: + r = -ENOTTY; + } + return r; +} + +int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf) +{ +#ifdef CONFIG_KVM_S390_UCONTROL + if ((vmf->pgoff == KVM_S390_SIE_PAGE_OFFSET) + && (kvm_is_ucontrol(vcpu->kvm))) { + vmf->page = virt_to_page(vcpu->arch.sie_block); + get_page(vmf->page); + return 0; + } +#endif + return VM_FAULT_SIGBUS; +} + +int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot, + unsigned long npages) +{ + return 0; +} + +/* Section: memory related */ +int kvm_arch_prepare_memory_region(struct kvm *kvm, + struct kvm_memory_slot *memslot, + struct kvm_userspace_memory_region *mem, + enum kvm_mr_change change) +{ + /* A few sanity checks. We can have memory slots which have to be + located/ended at a segment boundary (1MB). The memory in userland is + ok to be fragmented into various different vmas. It is okay to mmap() + and munmap() stuff in this slot after doing this call at any time */ + + if (mem->userspace_addr & 0xffffful) + return -EINVAL; + + if (mem->memory_size & 0xffffful) + return -EINVAL; + + return 0; +} + +void kvm_arch_commit_memory_region(struct kvm *kvm, + struct kvm_userspace_memory_region *mem, + const struct kvm_memory_slot *old, + enum kvm_mr_change change) +{ + int rc; + + /* If the basics of the memslot do not change, we do not want + * to update the gmap. Every update causes several unnecessary + * segment translation exceptions. This is usually handled just + * fine by the normal fault handler + gmap, but it will also + * cause faults on the prefix page of running guest CPUs. + */ + if (old->userspace_addr == mem->userspace_addr && + old->base_gfn * PAGE_SIZE == mem->guest_phys_addr && + old->npages * PAGE_SIZE == mem->memory_size) + return; + + rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr, + mem->guest_phys_addr, mem->memory_size); + if (rc) + printk(KERN_WARNING "kvm-s390: failed to commit memory region\n"); + return; +} + +static int __init kvm_s390_init(void) +{ + return kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE); +} + +static void __exit kvm_s390_exit(void) +{ + kvm_exit(); +} + +module_init(kvm_s390_init); +module_exit(kvm_s390_exit); + +/* + * Enable autoloading of the kvm module. + * Note that we add the module alias here instead of virt/kvm/kvm_main.c + * since x86 takes a different approach. + */ +#include <linux/miscdevice.h> +MODULE_ALIAS_MISCDEV(KVM_MINOR); +MODULE_ALIAS("devname:kvm"); diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h new file mode 100644 index 000000000..ca108b90a --- /dev/null +++ b/arch/s390/kvm/kvm-s390.h @@ -0,0 +1,290 @@ +/* + * definition for kvm on s390 + * + * Copyright IBM Corp. 2008, 2009 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + * Christian Borntraeger <borntraeger@de.ibm.com> + * Christian Ehrhardt <ehrhardt@de.ibm.com> + */ + +#ifndef ARCH_S390_KVM_S390_H +#define ARCH_S390_KVM_S390_H + +#include <linux/hrtimer.h> +#include <linux/kvm.h> +#include <linux/kvm_host.h> +#include <asm/facility.h> + +typedef int (*intercept_handler_t)(struct kvm_vcpu *vcpu); + +/* Transactional Memory Execution related macros */ +#define IS_TE_ENABLED(vcpu) ((vcpu->arch.sie_block->ecb & 0x10)) +#define TDB_FORMAT1 1 +#define IS_ITDB_VALID(vcpu) ((*(char *)vcpu->arch.sie_block->itdba == TDB_FORMAT1)) + +#define VM_EVENT(d_kvm, d_loglevel, d_string, d_args...)\ +do { \ + debug_sprintf_event(d_kvm->arch.dbf, d_loglevel, d_string "\n", \ + d_args); \ +} while (0) + +#define VCPU_EVENT(d_vcpu, d_loglevel, d_string, d_args...)\ +do { \ + debug_sprintf_event(d_vcpu->kvm->arch.dbf, d_loglevel, \ + "%02d[%016lx-%016lx]: " d_string "\n", d_vcpu->vcpu_id, \ + d_vcpu->arch.sie_block->gpsw.mask, d_vcpu->arch.sie_block->gpsw.addr,\ + d_args); \ +} while (0) + +static inline int is_vcpu_stopped(struct kvm_vcpu *vcpu) +{ + return atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_STOPPED; +} + +static inline int kvm_is_ucontrol(struct kvm *kvm) +{ +#ifdef CONFIG_KVM_S390_UCONTROL + if (kvm->arch.gmap) + return 0; + return 1; +#else + return 0; +#endif +} + +#define GUEST_PREFIX_SHIFT 13 +static inline u32 kvm_s390_get_prefix(struct kvm_vcpu *vcpu) +{ + return vcpu->arch.sie_block->prefix << GUEST_PREFIX_SHIFT; +} + +static inline void kvm_s390_set_prefix(struct kvm_vcpu *vcpu, u32 prefix) +{ + vcpu->arch.sie_block->prefix = prefix >> GUEST_PREFIX_SHIFT; + kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu); + kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu); +} + +typedef u8 __bitwise ar_t; + +static inline u64 kvm_s390_get_base_disp_s(struct kvm_vcpu *vcpu, ar_t *ar) +{ + u32 base2 = vcpu->arch.sie_block->ipb >> 28; + u32 disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); + + if (ar) + *ar = base2; + + return (base2 ? vcpu->run->s.regs.gprs[base2] : 0) + disp2; +} + +static inline void kvm_s390_get_base_disp_sse(struct kvm_vcpu *vcpu, + u64 *address1, u64 *address2, + ar_t *ar_b1, ar_t *ar_b2) +{ + u32 base1 = (vcpu->arch.sie_block->ipb & 0xf0000000) >> 28; + u32 disp1 = (vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16; + u32 base2 = (vcpu->arch.sie_block->ipb & 0xf000) >> 12; + u32 disp2 = vcpu->arch.sie_block->ipb & 0x0fff; + + *address1 = (base1 ? vcpu->run->s.regs.gprs[base1] : 0) + disp1; + *address2 = (base2 ? vcpu->run->s.regs.gprs[base2] : 0) + disp2; + + if (ar_b1) + *ar_b1 = base1; + if (ar_b2) + *ar_b2 = base2; +} + +static inline void kvm_s390_get_regs_rre(struct kvm_vcpu *vcpu, int *r1, int *r2) +{ + if (r1) + *r1 = (vcpu->arch.sie_block->ipb & 0x00f00000) >> 20; + if (r2) + *r2 = (vcpu->arch.sie_block->ipb & 0x000f0000) >> 16; +} + +static inline u64 kvm_s390_get_base_disp_rsy(struct kvm_vcpu *vcpu, ar_t *ar) +{ + u32 base2 = vcpu->arch.sie_block->ipb >> 28; + u32 disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16) + + ((vcpu->arch.sie_block->ipb & 0xff00) << 4); + /* The displacement is a 20bit _SIGNED_ value */ + if (disp2 & 0x80000) + disp2+=0xfff00000; + + if (ar) + *ar = base2; + + return (base2 ? vcpu->run->s.regs.gprs[base2] : 0) + (long)(int)disp2; +} + +static inline u64 kvm_s390_get_base_disp_rs(struct kvm_vcpu *vcpu, ar_t *ar) +{ + u32 base2 = vcpu->arch.sie_block->ipb >> 28; + u32 disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); + + if (ar) + *ar = base2; + + return (base2 ? vcpu->run->s.regs.gprs[base2] : 0) + disp2; +} + +/* Set the condition code in the guest program status word */ +static inline void kvm_s390_set_psw_cc(struct kvm_vcpu *vcpu, unsigned long cc) +{ + vcpu->arch.sie_block->gpsw.mask &= ~(3UL << 44); + vcpu->arch.sie_block->gpsw.mask |= cc << 44; +} + +/* test availability of facility in a kvm instance */ +static inline int test_kvm_facility(struct kvm *kvm, unsigned long nr) +{ + return __test_facility(nr, kvm->arch.model.fac->mask) && + __test_facility(nr, kvm->arch.model.fac->list); +} + +static inline int set_kvm_facility(u64 *fac_list, unsigned long nr) +{ + unsigned char *ptr; + + if (nr >= MAX_FACILITY_BIT) + return -EINVAL; + ptr = (unsigned char *) fac_list + (nr >> 3); + *ptr |= (0x80UL >> (nr & 7)); + return 0; +} + +/* are cpu states controlled by user space */ +static inline int kvm_s390_user_cpu_state_ctrl(struct kvm *kvm) +{ + return kvm->arch.user_cpu_state_ctrl != 0; +} + +int kvm_s390_handle_wait(struct kvm_vcpu *vcpu); +void kvm_s390_vcpu_wakeup(struct kvm_vcpu *vcpu); +enum hrtimer_restart kvm_s390_idle_wakeup(struct hrtimer *timer); +int __must_check kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu); +void kvm_s390_clear_local_irqs(struct kvm_vcpu *vcpu); +void kvm_s390_clear_float_irqs(struct kvm *kvm); +int __must_check kvm_s390_inject_vm(struct kvm *kvm, + struct kvm_s390_interrupt *s390int); +int __must_check kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, + struct kvm_s390_irq *irq); +int __must_check kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code); +struct kvm_s390_interrupt_info *kvm_s390_get_io_int(struct kvm *kvm, + u64 isc_mask, u32 schid); +int kvm_s390_reinject_io_int(struct kvm *kvm, + struct kvm_s390_interrupt_info *inti); +int kvm_s390_mask_adapter(struct kvm *kvm, unsigned int id, bool masked); + +/* implemented in intercept.c */ +void kvm_s390_rewind_psw(struct kvm_vcpu *vcpu, int ilc); +int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu); + +/* implemented in priv.c */ +int is_valid_psw(psw_t *psw); +int kvm_s390_handle_b2(struct kvm_vcpu *vcpu); +int kvm_s390_handle_e5(struct kvm_vcpu *vcpu); +int kvm_s390_handle_01(struct kvm_vcpu *vcpu); +int kvm_s390_handle_b9(struct kvm_vcpu *vcpu); +int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu); +int kvm_s390_handle_stctl(struct kvm_vcpu *vcpu); +int kvm_s390_handle_lctl(struct kvm_vcpu *vcpu); +int kvm_s390_handle_eb(struct kvm_vcpu *vcpu); + +/* implemented in sigp.c */ +int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu); +int kvm_s390_handle_sigp_pei(struct kvm_vcpu *vcpu); + +/* implemented in kvm-s390.c */ +long kvm_arch_fault_in_page(struct kvm_vcpu *vcpu, gpa_t gpa, int writable); +int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long addr); +int kvm_s390_store_adtl_status_unloaded(struct kvm_vcpu *vcpu, + unsigned long addr); +int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr); +int kvm_s390_vcpu_store_adtl_status(struct kvm_vcpu *vcpu, unsigned long addr); +void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu); +void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu); +void s390_vcpu_block(struct kvm_vcpu *vcpu); +void s390_vcpu_unblock(struct kvm_vcpu *vcpu); +void exit_sie(struct kvm_vcpu *vcpu); +void exit_sie_sync(struct kvm_vcpu *vcpu); +int kvm_s390_vcpu_setup_cmma(struct kvm_vcpu *vcpu); +void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu); +/* is cmma enabled */ +bool kvm_s390_cmma_enabled(struct kvm *kvm); +unsigned long kvm_s390_fac_list_mask_size(void); +extern unsigned long kvm_s390_fac_list_mask[]; + +/* implemented in diag.c */ +int kvm_s390_handle_diag(struct kvm_vcpu *vcpu); +/* implemented in interrupt.c */ +int kvm_s390_inject_prog_irq(struct kvm_vcpu *vcpu, + struct kvm_s390_pgm_info *pgm_info); + +/** + * kvm_s390_inject_prog_cond - conditionally inject a program check + * @vcpu: virtual cpu + * @rc: original return/error code + * + * This function is supposed to be used after regular guest access functions + * failed, to conditionally inject a program check to a vcpu. The typical + * pattern would look like + * + * rc = write_guest(vcpu, addr, data, len); + * if (rc) + * return kvm_s390_inject_prog_cond(vcpu, rc); + * + * A negative return code from guest access functions implies an internal error + * like e.g. out of memory. In these cases no program check should be injected + * to the guest. + * A positive value implies that an exception happened while accessing a guest's + * memory. In this case all data belonging to the corresponding program check + * has been stored in vcpu->arch.pgm and can be injected with + * kvm_s390_inject_prog_irq(). + * + * Returns: - the original @rc value if @rc was negative (internal error) + * - zero if @rc was already zero + * - zero or error code from injecting if @rc was positive + * (program check injected to @vcpu) + */ +static inline int kvm_s390_inject_prog_cond(struct kvm_vcpu *vcpu, int rc) +{ + if (rc <= 0) + return rc; + return kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm); +} + +int s390int_to_s390irq(struct kvm_s390_interrupt *s390int, + struct kvm_s390_irq *s390irq); + +/* implemented in interrupt.c */ +int kvm_s390_vcpu_has_irq(struct kvm_vcpu *vcpu, int exclude_stop); +int psw_extint_disabled(struct kvm_vcpu *vcpu); +void kvm_s390_destroy_adapters(struct kvm *kvm); +int kvm_s390_ext_call_pending(struct kvm_vcpu *vcpu); +extern struct kvm_device_ops kvm_flic_ops; +int kvm_s390_is_stop_irq_pending(struct kvm_vcpu *vcpu); +void kvm_s390_clear_stop_irq(struct kvm_vcpu *vcpu); +int kvm_s390_set_irq_state(struct kvm_vcpu *vcpu, + void __user *buf, int len); +int kvm_s390_get_irq_state(struct kvm_vcpu *vcpu, + __u8 __user *buf, int len); + +/* implemented in guestdbg.c */ +void kvm_s390_backup_guest_per_regs(struct kvm_vcpu *vcpu); +void kvm_s390_restore_guest_per_regs(struct kvm_vcpu *vcpu); +void kvm_s390_patch_guest_per_regs(struct kvm_vcpu *vcpu); +int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu, + struct kvm_guest_debug *dbg); +void kvm_s390_clear_bp_data(struct kvm_vcpu *vcpu); +void kvm_s390_prepare_debug_exit(struct kvm_vcpu *vcpu); +void kvm_s390_handle_per_event(struct kvm_vcpu *vcpu); + +#endif diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c new file mode 100644 index 000000000..d22d8ee1f --- /dev/null +++ b/arch/s390/kvm/priv.c @@ -0,0 +1,1060 @@ +/* + * handling privileged instructions + * + * Copyright IBM Corp. 2008, 2013 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + * Christian Borntraeger <borntraeger@de.ibm.com> + */ + +#include <linux/kvm.h> +#include <linux/gfp.h> +#include <linux/errno.h> +#include <linux/compat.h> +#include <asm/asm-offsets.h> +#include <asm/facility.h> +#include <asm/current.h> +#include <asm/debug.h> +#include <asm/ebcdic.h> +#include <asm/sysinfo.h> +#include <asm/pgtable.h> +#include <asm/pgalloc.h> +#include <asm/io.h> +#include <asm/ptrace.h> +#include <asm/compat.h> +#include "gaccess.h" +#include "kvm-s390.h" +#include "trace.h" + +/* Handle SCK (SET CLOCK) interception */ +static int handle_set_clock(struct kvm_vcpu *vcpu) +{ + struct kvm_vcpu *cpup; + s64 hostclk, val; + int i, rc; + ar_t ar; + u64 op2; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + op2 = kvm_s390_get_base_disp_s(vcpu, &ar); + if (op2 & 7) /* Operand must be on a doubleword boundary */ + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + rc = read_guest(vcpu, op2, ar, &val, sizeof(val)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + + if (store_tod_clock(&hostclk)) { + kvm_s390_set_psw_cc(vcpu, 3); + return 0; + } + val = (val - hostclk) & ~0x3fUL; + + mutex_lock(&vcpu->kvm->lock); + kvm_for_each_vcpu(i, cpup, vcpu->kvm) + cpup->arch.sie_block->epoch = val; + mutex_unlock(&vcpu->kvm->lock); + + kvm_s390_set_psw_cc(vcpu, 0); + return 0; +} + +static int handle_set_prefix(struct kvm_vcpu *vcpu) +{ + u64 operand2; + u32 address; + int rc; + ar_t ar; + + vcpu->stat.instruction_spx++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + operand2 = kvm_s390_get_base_disp_s(vcpu, &ar); + + /* must be word boundary */ + if (operand2 & 3) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + /* get the value */ + rc = read_guest(vcpu, operand2, ar, &address, sizeof(address)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + + address &= 0x7fffe000u; + + /* + * Make sure the new value is valid memory. We only need to check the + * first page, since address is 8k aligned and memory pieces are always + * at least 1MB aligned and have at least a size of 1MB. + */ + if (kvm_is_error_gpa(vcpu->kvm, address)) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + + kvm_s390_set_prefix(vcpu, address); + + VCPU_EVENT(vcpu, 5, "setting prefix to %x", address); + trace_kvm_s390_handle_prefix(vcpu, 1, address); + return 0; +} + +static int handle_store_prefix(struct kvm_vcpu *vcpu) +{ + u64 operand2; + u32 address; + int rc; + ar_t ar; + + vcpu->stat.instruction_stpx++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + operand2 = kvm_s390_get_base_disp_s(vcpu, &ar); + + /* must be word boundary */ + if (operand2 & 3) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + address = kvm_s390_get_prefix(vcpu); + + /* get the value */ + rc = write_guest(vcpu, operand2, ar, &address, sizeof(address)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + + VCPU_EVENT(vcpu, 5, "storing prefix to %x", address); + trace_kvm_s390_handle_prefix(vcpu, 0, address); + return 0; +} + +static int handle_store_cpu_address(struct kvm_vcpu *vcpu) +{ + u16 vcpu_id = vcpu->vcpu_id; + u64 ga; + int rc; + ar_t ar; + + vcpu->stat.instruction_stap++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + ga = kvm_s390_get_base_disp_s(vcpu, &ar); + + if (ga & 1) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + rc = write_guest(vcpu, ga, ar, &vcpu_id, sizeof(vcpu_id)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + + VCPU_EVENT(vcpu, 5, "storing cpu address to %llx", ga); + trace_kvm_s390_handle_stap(vcpu, ga); + return 0; +} + +static int __skey_check_enable(struct kvm_vcpu *vcpu) +{ + int rc = 0; + if (!(vcpu->arch.sie_block->ictl & (ICTL_ISKE | ICTL_SSKE | ICTL_RRBE))) + return rc; + + rc = s390_enable_skey(); + trace_kvm_s390_skey_related_inst(vcpu); + vcpu->arch.sie_block->ictl &= ~(ICTL_ISKE | ICTL_SSKE | ICTL_RRBE); + return rc; +} + + +static int handle_skey(struct kvm_vcpu *vcpu) +{ + int rc = __skey_check_enable(vcpu); + + if (rc) + return rc; + vcpu->stat.instruction_storage_key++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + kvm_s390_rewind_psw(vcpu, 4); + VCPU_EVENT(vcpu, 4, "%s", "retrying storage key operation"); + return 0; +} + +static int handle_ipte_interlock(struct kvm_vcpu *vcpu) +{ + vcpu->stat.instruction_ipte_interlock++; + if (psw_bits(vcpu->arch.sie_block->gpsw).p) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + wait_event(vcpu->kvm->arch.ipte_wq, !ipte_lock_held(vcpu)); + kvm_s390_rewind_psw(vcpu, 4); + VCPU_EVENT(vcpu, 4, "%s", "retrying ipte interlock operation"); + return 0; +} + +static int handle_test_block(struct kvm_vcpu *vcpu) +{ + gpa_t addr; + int reg2; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + kvm_s390_get_regs_rre(vcpu, NULL, ®2); + addr = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK; + addr = kvm_s390_logical_to_effective(vcpu, addr); + if (kvm_s390_check_low_addr_prot_real(vcpu, addr)) + return kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm); + addr = kvm_s390_real_to_abs(vcpu, addr); + + if (kvm_is_error_gpa(vcpu->kvm, addr)) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + /* + * We don't expect errors on modern systems, and do not care + * about storage keys (yet), so let's just clear the page. + */ + if (kvm_clear_guest(vcpu->kvm, addr, PAGE_SIZE)) + return -EFAULT; + kvm_s390_set_psw_cc(vcpu, 0); + vcpu->run->s.regs.gprs[0] = 0; + return 0; +} + +static int handle_tpi(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_interrupt_info *inti; + unsigned long len; + u32 tpi_data[3]; + int rc; + u64 addr; + ar_t ar; + + addr = kvm_s390_get_base_disp_s(vcpu, &ar); + if (addr & 3) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + inti = kvm_s390_get_io_int(vcpu->kvm, vcpu->arch.sie_block->gcr[6], 0); + if (!inti) { + kvm_s390_set_psw_cc(vcpu, 0); + return 0; + } + + tpi_data[0] = inti->io.subchannel_id << 16 | inti->io.subchannel_nr; + tpi_data[1] = inti->io.io_int_parm; + tpi_data[2] = inti->io.io_int_word; + if (addr) { + /* + * Store the two-word I/O interruption code into the + * provided area. + */ + len = sizeof(tpi_data) - 4; + rc = write_guest(vcpu, addr, ar, &tpi_data, len); + if (rc) { + rc = kvm_s390_inject_prog_cond(vcpu, rc); + goto reinject_interrupt; + } + } else { + /* + * Store the three-word I/O interruption code into + * the appropriate lowcore area. + */ + len = sizeof(tpi_data); + if (write_guest_lc(vcpu, __LC_SUBCHANNEL_ID, &tpi_data, len)) { + /* failed writes to the low core are not recoverable */ + rc = -EFAULT; + goto reinject_interrupt; + } + } + + /* irq was successfully handed to the guest */ + kfree(inti); + kvm_s390_set_psw_cc(vcpu, 1); + return 0; +reinject_interrupt: + /* + * If we encounter a problem storing the interruption code, the + * instruction is suppressed from the guest's view: reinject the + * interrupt. + */ + if (kvm_s390_reinject_io_int(vcpu->kvm, inti)) { + kfree(inti); + rc = -EFAULT; + } + /* don't set the cc, a pgm irq was injected or we drop to user space */ + return rc ? -EFAULT : 0; +} + +static int handle_tsch(struct kvm_vcpu *vcpu) +{ + struct kvm_s390_interrupt_info *inti = NULL; + const u64 isc_mask = 0xffUL << 24; /* all iscs set */ + + /* a valid schid has at least one bit set */ + if (vcpu->run->s.regs.gprs[1]) + inti = kvm_s390_get_io_int(vcpu->kvm, isc_mask, + vcpu->run->s.regs.gprs[1]); + + /* + * Prepare exit to userspace. + * We indicate whether we dequeued a pending I/O interrupt + * so that userspace can re-inject it if the instruction gets + * a program check. While this may re-order the pending I/O + * interrupts, this is no problem since the priority is kept + * intact. + */ + vcpu->run->exit_reason = KVM_EXIT_S390_TSCH; + vcpu->run->s390_tsch.dequeued = !!inti; + if (inti) { + vcpu->run->s390_tsch.subchannel_id = inti->io.subchannel_id; + vcpu->run->s390_tsch.subchannel_nr = inti->io.subchannel_nr; + vcpu->run->s390_tsch.io_int_parm = inti->io.io_int_parm; + vcpu->run->s390_tsch.io_int_word = inti->io.io_int_word; + } + vcpu->run->s390_tsch.ipb = vcpu->arch.sie_block->ipb; + kfree(inti); + return -EREMOTE; +} + +static int handle_io_inst(struct kvm_vcpu *vcpu) +{ + VCPU_EVENT(vcpu, 4, "%s", "I/O instruction"); + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + if (vcpu->kvm->arch.css_support) { + /* + * Most I/O instructions will be handled by userspace. + * Exceptions are tpi and the interrupt portion of tsch. + */ + if (vcpu->arch.sie_block->ipa == 0xb236) + return handle_tpi(vcpu); + if (vcpu->arch.sie_block->ipa == 0xb235) + return handle_tsch(vcpu); + /* Handle in userspace. */ + return -EOPNOTSUPP; + } else { + /* + * Set condition code 3 to stop the guest from issuing channel + * I/O instructions. + */ + kvm_s390_set_psw_cc(vcpu, 3); + return 0; + } +} + +static int handle_stfl(struct kvm_vcpu *vcpu) +{ + int rc; + unsigned int fac; + + vcpu->stat.instruction_stfl++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + /* + * We need to shift the lower 32 facility bits (bit 0-31) from a u64 + * into a u32 memory representation. They will remain bits 0-31. + */ + fac = *vcpu->kvm->arch.model.fac->list >> 32; + rc = write_guest_lc(vcpu, offsetof(struct _lowcore, stfl_fac_list), + &fac, sizeof(fac)); + if (rc) + return rc; + VCPU_EVENT(vcpu, 5, "store facility list value %x", fac); + trace_kvm_s390_handle_stfl(vcpu, fac); + return 0; +} + +#define PSW_MASK_ADDR_MODE (PSW_MASK_EA | PSW_MASK_BA) +#define PSW_MASK_UNASSIGNED 0xb80800fe7fffffffUL +#define PSW_ADDR_24 0x0000000000ffffffUL +#define PSW_ADDR_31 0x000000007fffffffUL + +int is_valid_psw(psw_t *psw) +{ + if (psw->mask & PSW_MASK_UNASSIGNED) + return 0; + if ((psw->mask & PSW_MASK_ADDR_MODE) == PSW_MASK_BA) { + if (psw->addr & ~PSW_ADDR_31) + return 0; + } + if (!(psw->mask & PSW_MASK_ADDR_MODE) && (psw->addr & ~PSW_ADDR_24)) + return 0; + if ((psw->mask & PSW_MASK_ADDR_MODE) == PSW_MASK_EA) + return 0; + if (psw->addr & 1) + return 0; + return 1; +} + +int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu) +{ + psw_t *gpsw = &vcpu->arch.sie_block->gpsw; + psw_compat_t new_psw; + u64 addr; + int rc; + ar_t ar; + + if (gpsw->mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + addr = kvm_s390_get_base_disp_s(vcpu, &ar); + if (addr & 7) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + rc = read_guest(vcpu, addr, ar, &new_psw, sizeof(new_psw)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + if (!(new_psw.mask & PSW32_MASK_BASE)) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + gpsw->mask = (new_psw.mask & ~PSW32_MASK_BASE) << 32; + gpsw->mask |= new_psw.addr & PSW32_ADDR_AMODE; + gpsw->addr = new_psw.addr & ~PSW32_ADDR_AMODE; + if (!is_valid_psw(gpsw)) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + return 0; +} + +static int handle_lpswe(struct kvm_vcpu *vcpu) +{ + psw_t new_psw; + u64 addr; + int rc; + ar_t ar; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + addr = kvm_s390_get_base_disp_s(vcpu, &ar); + if (addr & 7) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + rc = read_guest(vcpu, addr, ar, &new_psw, sizeof(new_psw)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + vcpu->arch.sie_block->gpsw = new_psw; + if (!is_valid_psw(&vcpu->arch.sie_block->gpsw)) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + return 0; +} + +static int handle_stidp(struct kvm_vcpu *vcpu) +{ + u64 stidp_data = vcpu->arch.stidp_data; + u64 operand2; + int rc; + ar_t ar; + + vcpu->stat.instruction_stidp++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + operand2 = kvm_s390_get_base_disp_s(vcpu, &ar); + + if (operand2 & 7) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + rc = write_guest(vcpu, operand2, ar, &stidp_data, sizeof(stidp_data)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + + VCPU_EVENT(vcpu, 5, "%s", "store cpu id"); + return 0; +} + +static void handle_stsi_3_2_2(struct kvm_vcpu *vcpu, struct sysinfo_3_2_2 *mem) +{ + int cpus = 0; + int n; + + cpus = atomic_read(&vcpu->kvm->online_vcpus); + + /* deal with other level 3 hypervisors */ + if (stsi(mem, 3, 2, 2)) + mem->count = 0; + if (mem->count < 8) + mem->count++; + for (n = mem->count - 1; n > 0 ; n--) + memcpy(&mem->vm[n], &mem->vm[n - 1], sizeof(mem->vm[0])); + + memset(&mem->vm[0], 0, sizeof(mem->vm[0])); + mem->vm[0].cpus_total = cpus; + mem->vm[0].cpus_configured = cpus; + mem->vm[0].cpus_standby = 0; + mem->vm[0].cpus_reserved = 0; + mem->vm[0].caf = 1000; + memcpy(mem->vm[0].name, "KVMguest", 8); + ASCEBC(mem->vm[0].name, 8); + memcpy(mem->vm[0].cpi, "KVM/Linux ", 16); + ASCEBC(mem->vm[0].cpi, 16); +} + +static void insert_stsi_usr_data(struct kvm_vcpu *vcpu, u64 addr, ar_t ar, + u8 fc, u8 sel1, u16 sel2) +{ + vcpu->run->exit_reason = KVM_EXIT_S390_STSI; + vcpu->run->s390_stsi.addr = addr; + vcpu->run->s390_stsi.ar = ar; + vcpu->run->s390_stsi.fc = fc; + vcpu->run->s390_stsi.sel1 = sel1; + vcpu->run->s390_stsi.sel2 = sel2; +} + +static int handle_stsi(struct kvm_vcpu *vcpu) +{ + int fc = (vcpu->run->s.regs.gprs[0] & 0xf0000000) >> 28; + int sel1 = vcpu->run->s.regs.gprs[0] & 0xff; + int sel2 = vcpu->run->s.regs.gprs[1] & 0xffff; + unsigned long mem = 0; + u64 operand2; + int rc = 0; + ar_t ar; + + vcpu->stat.instruction_stsi++; + VCPU_EVENT(vcpu, 4, "stsi: fc: %x sel1: %x sel2: %x", fc, sel1, sel2); + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + if (fc > 3) { + kvm_s390_set_psw_cc(vcpu, 3); + return 0; + } + + if (vcpu->run->s.regs.gprs[0] & 0x0fffff00 + || vcpu->run->s.regs.gprs[1] & 0xffff0000) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + if (fc == 0) { + vcpu->run->s.regs.gprs[0] = 3 << 28; + kvm_s390_set_psw_cc(vcpu, 0); + return 0; + } + + operand2 = kvm_s390_get_base_disp_s(vcpu, &ar); + + if (operand2 & 0xfff) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + switch (fc) { + case 1: /* same handling for 1 and 2 */ + case 2: + mem = get_zeroed_page(GFP_KERNEL); + if (!mem) + goto out_no_data; + if (stsi((void *) mem, fc, sel1, sel2)) + goto out_no_data; + break; + case 3: + if (sel1 != 2 || sel2 != 2) + goto out_no_data; + mem = get_zeroed_page(GFP_KERNEL); + if (!mem) + goto out_no_data; + handle_stsi_3_2_2(vcpu, (void *) mem); + break; + } + + rc = write_guest(vcpu, operand2, ar, (void *)mem, PAGE_SIZE); + if (rc) { + rc = kvm_s390_inject_prog_cond(vcpu, rc); + goto out; + } + if (vcpu->kvm->arch.user_stsi) { + insert_stsi_usr_data(vcpu, operand2, ar, fc, sel1, sel2); + rc = -EREMOTE; + } + trace_kvm_s390_handle_stsi(vcpu, fc, sel1, sel2, operand2); + free_page(mem); + kvm_s390_set_psw_cc(vcpu, 0); + vcpu->run->s.regs.gprs[0] = 0; + return rc; +out_no_data: + kvm_s390_set_psw_cc(vcpu, 3); +out: + free_page(mem); + return rc; +} + +static const intercept_handler_t b2_handlers[256] = { + [0x02] = handle_stidp, + [0x04] = handle_set_clock, + [0x10] = handle_set_prefix, + [0x11] = handle_store_prefix, + [0x12] = handle_store_cpu_address, + [0x21] = handle_ipte_interlock, + [0x29] = handle_skey, + [0x2a] = handle_skey, + [0x2b] = handle_skey, + [0x2c] = handle_test_block, + [0x30] = handle_io_inst, + [0x31] = handle_io_inst, + [0x32] = handle_io_inst, + [0x33] = handle_io_inst, + [0x34] = handle_io_inst, + [0x35] = handle_io_inst, + [0x36] = handle_io_inst, + [0x37] = handle_io_inst, + [0x38] = handle_io_inst, + [0x39] = handle_io_inst, + [0x3a] = handle_io_inst, + [0x3b] = handle_io_inst, + [0x3c] = handle_io_inst, + [0x50] = handle_ipte_interlock, + [0x5f] = handle_io_inst, + [0x74] = handle_io_inst, + [0x76] = handle_io_inst, + [0x7d] = handle_stsi, + [0xb1] = handle_stfl, + [0xb2] = handle_lpswe, +}; + +int kvm_s390_handle_b2(struct kvm_vcpu *vcpu) +{ + intercept_handler_t handler; + + /* + * A lot of B2 instructions are priviledged. Here we check for + * the privileged ones, that we can handle in the kernel. + * Anything else goes to userspace. + */ + handler = b2_handlers[vcpu->arch.sie_block->ipa & 0x00ff]; + if (handler) + return handler(vcpu); + + return -EOPNOTSUPP; +} + +static int handle_epsw(struct kvm_vcpu *vcpu) +{ + int reg1, reg2; + + kvm_s390_get_regs_rre(vcpu, ®1, ®2); + + /* This basically extracts the mask half of the psw. */ + vcpu->run->s.regs.gprs[reg1] &= 0xffffffff00000000UL; + vcpu->run->s.regs.gprs[reg1] |= vcpu->arch.sie_block->gpsw.mask >> 32; + if (reg2) { + vcpu->run->s.regs.gprs[reg2] &= 0xffffffff00000000UL; + vcpu->run->s.regs.gprs[reg2] |= + vcpu->arch.sie_block->gpsw.mask & 0x00000000ffffffffUL; + } + return 0; +} + +#define PFMF_RESERVED 0xfffc0101UL +#define PFMF_SK 0x00020000UL +#define PFMF_CF 0x00010000UL +#define PFMF_UI 0x00008000UL +#define PFMF_FSC 0x00007000UL +#define PFMF_NQ 0x00000800UL +#define PFMF_MR 0x00000400UL +#define PFMF_MC 0x00000200UL +#define PFMF_KEY 0x000000feUL + +static int handle_pfmf(struct kvm_vcpu *vcpu) +{ + int reg1, reg2; + unsigned long start, end; + + vcpu->stat.instruction_pfmf++; + + kvm_s390_get_regs_rre(vcpu, ®1, ®2); + + if (!MACHINE_HAS_PFMF) + return kvm_s390_inject_program_int(vcpu, PGM_OPERATION); + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + if (vcpu->run->s.regs.gprs[reg1] & PFMF_RESERVED) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + /* Only provide non-quiescing support if the host supports it */ + if (vcpu->run->s.regs.gprs[reg1] & PFMF_NQ && !test_facility(14)) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + /* No support for conditional-SSKE */ + if (vcpu->run->s.regs.gprs[reg1] & (PFMF_MR | PFMF_MC)) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + start = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK; + start = kvm_s390_logical_to_effective(vcpu, start); + + switch (vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) { + case 0x00000000: + end = (start + (1UL << 12)) & ~((1UL << 12) - 1); + break; + case 0x00001000: + end = (start + (1UL << 20)) & ~((1UL << 20) - 1); + break; + /* We dont support EDAT2 + case 0x00002000: + end = (start + (1UL << 31)) & ~((1UL << 31) - 1); + break;*/ + default: + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + } + + if (vcpu->run->s.regs.gprs[reg1] & PFMF_CF) { + if (kvm_s390_check_low_addr_prot_real(vcpu, start)) + return kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm); + } + + while (start < end) { + unsigned long useraddr, abs_addr; + + /* Translate guest address to host address */ + if ((vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) == 0) + abs_addr = kvm_s390_real_to_abs(vcpu, start); + else + abs_addr = start; + useraddr = gfn_to_hva(vcpu->kvm, gpa_to_gfn(abs_addr)); + if (kvm_is_error_hva(useraddr)) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + + if (vcpu->run->s.regs.gprs[reg1] & PFMF_CF) { + if (clear_user((void __user *)useraddr, PAGE_SIZE)) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + } + + if (vcpu->run->s.regs.gprs[reg1] & PFMF_SK) { + int rc = __skey_check_enable(vcpu); + + if (rc) + return rc; + if (set_guest_storage_key(current->mm, useraddr, + vcpu->run->s.regs.gprs[reg1] & PFMF_KEY, + vcpu->run->s.regs.gprs[reg1] & PFMF_NQ)) + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + } + + start += PAGE_SIZE; + } + if (vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) + vcpu->run->s.regs.gprs[reg2] = end; + return 0; +} + +static int handle_essa(struct kvm_vcpu *vcpu) +{ + /* entries expected to be 1FF */ + int entries = (vcpu->arch.sie_block->cbrlo & ~PAGE_MASK) >> 3; + unsigned long *cbrlo, cbrle; + struct gmap *gmap; + int i; + + VCPU_EVENT(vcpu, 5, "cmma release %d pages", entries); + gmap = vcpu->arch.gmap; + vcpu->stat.instruction_essa++; + if (!kvm_s390_cmma_enabled(vcpu->kvm)) + return kvm_s390_inject_program_int(vcpu, PGM_OPERATION); + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + if (((vcpu->arch.sie_block->ipb & 0xf0000000) >> 28) > 6) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + /* Rewind PSW to repeat the ESSA instruction */ + kvm_s390_rewind_psw(vcpu, 4); + vcpu->arch.sie_block->cbrlo &= PAGE_MASK; /* reset nceo */ + cbrlo = phys_to_virt(vcpu->arch.sie_block->cbrlo); + down_read(&gmap->mm->mmap_sem); + for (i = 0; i < entries; ++i) { + cbrle = cbrlo[i]; + if (unlikely(cbrle & ~PAGE_MASK || cbrle < 2 * PAGE_SIZE)) + /* invalid entry */ + break; + /* try to free backing */ + __gmap_zap(gmap, cbrle); + } + up_read(&gmap->mm->mmap_sem); + if (i < entries) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + return 0; +} + +static const intercept_handler_t b9_handlers[256] = { + [0x8a] = handle_ipte_interlock, + [0x8d] = handle_epsw, + [0x8e] = handle_ipte_interlock, + [0x8f] = handle_ipte_interlock, + [0xab] = handle_essa, + [0xaf] = handle_pfmf, +}; + +int kvm_s390_handle_b9(struct kvm_vcpu *vcpu) +{ + intercept_handler_t handler; + + /* This is handled just as for the B2 instructions. */ + handler = b9_handlers[vcpu->arch.sie_block->ipa & 0x00ff]; + if (handler) + return handler(vcpu); + + return -EOPNOTSUPP; +} + +int kvm_s390_handle_lctl(struct kvm_vcpu *vcpu) +{ + int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int reg3 = vcpu->arch.sie_block->ipa & 0x000f; + int reg, rc, nr_regs; + u32 ctl_array[16]; + u64 ga; + ar_t ar; + + vcpu->stat.instruction_lctl++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + ga = kvm_s390_get_base_disp_rs(vcpu, &ar); + + if (ga & 3) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + VCPU_EVENT(vcpu, 5, "lctl r1:%x, r3:%x, addr:%llx", reg1, reg3, ga); + trace_kvm_s390_handle_lctl(vcpu, 0, reg1, reg3, ga); + + nr_regs = ((reg3 - reg1) & 0xf) + 1; + rc = read_guest(vcpu, ga, ar, ctl_array, nr_regs * sizeof(u32)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + reg = reg1; + nr_regs = 0; + do { + vcpu->arch.sie_block->gcr[reg] &= 0xffffffff00000000ul; + vcpu->arch.sie_block->gcr[reg] |= ctl_array[nr_regs++]; + if (reg == reg3) + break; + reg = (reg + 1) % 16; + } while (1); + kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu); + return 0; +} + +int kvm_s390_handle_stctl(struct kvm_vcpu *vcpu) +{ + int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int reg3 = vcpu->arch.sie_block->ipa & 0x000f; + int reg, rc, nr_regs; + u32 ctl_array[16]; + u64 ga; + ar_t ar; + + vcpu->stat.instruction_stctl++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + ga = kvm_s390_get_base_disp_rs(vcpu, &ar); + + if (ga & 3) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + VCPU_EVENT(vcpu, 5, "stctl r1:%x, r3:%x, addr:%llx", reg1, reg3, ga); + trace_kvm_s390_handle_stctl(vcpu, 0, reg1, reg3, ga); + + reg = reg1; + nr_regs = 0; + do { + ctl_array[nr_regs++] = vcpu->arch.sie_block->gcr[reg]; + if (reg == reg3) + break; + reg = (reg + 1) % 16; + } while (1); + rc = write_guest(vcpu, ga, ar, ctl_array, nr_regs * sizeof(u32)); + return rc ? kvm_s390_inject_prog_cond(vcpu, rc) : 0; +} + +static int handle_lctlg(struct kvm_vcpu *vcpu) +{ + int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int reg3 = vcpu->arch.sie_block->ipa & 0x000f; + int reg, rc, nr_regs; + u64 ctl_array[16]; + u64 ga; + ar_t ar; + + vcpu->stat.instruction_lctlg++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + ga = kvm_s390_get_base_disp_rsy(vcpu, &ar); + + if (ga & 7) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + VCPU_EVENT(vcpu, 5, "lctlg r1:%x, r3:%x, addr:%llx", reg1, reg3, ga); + trace_kvm_s390_handle_lctl(vcpu, 1, reg1, reg3, ga); + + nr_regs = ((reg3 - reg1) & 0xf) + 1; + rc = read_guest(vcpu, ga, ar, ctl_array, nr_regs * sizeof(u64)); + if (rc) + return kvm_s390_inject_prog_cond(vcpu, rc); + reg = reg1; + nr_regs = 0; + do { + vcpu->arch.sie_block->gcr[reg] = ctl_array[nr_regs++]; + if (reg == reg3) + break; + reg = (reg + 1) % 16; + } while (1); + kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu); + return 0; +} + +static int handle_stctg(struct kvm_vcpu *vcpu) +{ + int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int reg3 = vcpu->arch.sie_block->ipa & 0x000f; + int reg, rc, nr_regs; + u64 ctl_array[16]; + u64 ga; + ar_t ar; + + vcpu->stat.instruction_stctg++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + ga = kvm_s390_get_base_disp_rsy(vcpu, &ar); + + if (ga & 7) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + VCPU_EVENT(vcpu, 5, "stctg r1:%x, r3:%x, addr:%llx", reg1, reg3, ga); + trace_kvm_s390_handle_stctl(vcpu, 1, reg1, reg3, ga); + + reg = reg1; + nr_regs = 0; + do { + ctl_array[nr_regs++] = vcpu->arch.sie_block->gcr[reg]; + if (reg == reg3) + break; + reg = (reg + 1) % 16; + } while (1); + rc = write_guest(vcpu, ga, ar, ctl_array, nr_regs * sizeof(u64)); + return rc ? kvm_s390_inject_prog_cond(vcpu, rc) : 0; +} + +static const intercept_handler_t eb_handlers[256] = { + [0x2f] = handle_lctlg, + [0x25] = handle_stctg, +}; + +int kvm_s390_handle_eb(struct kvm_vcpu *vcpu) +{ + intercept_handler_t handler; + + handler = eb_handlers[vcpu->arch.sie_block->ipb & 0xff]; + if (handler) + return handler(vcpu); + return -EOPNOTSUPP; +} + +static int handle_tprot(struct kvm_vcpu *vcpu) +{ + u64 address1, address2; + unsigned long hva, gpa; + int ret = 0, cc = 0; + bool writable; + ar_t ar; + + vcpu->stat.instruction_tprot++; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + kvm_s390_get_base_disp_sse(vcpu, &address1, &address2, &ar, NULL); + + /* we only handle the Linux memory detection case: + * access key == 0 + * everything else goes to userspace. */ + if (address2 & 0xf0) + return -EOPNOTSUPP; + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_DAT) + ipte_lock(vcpu); + ret = guest_translate_address(vcpu, address1, ar, &gpa, 1); + if (ret == PGM_PROTECTION) { + /* Write protected? Try again with read-only... */ + cc = 1; + ret = guest_translate_address(vcpu, address1, ar, &gpa, 0); + } + if (ret) { + if (ret == PGM_ADDRESSING || ret == PGM_TRANSLATION_SPEC) { + ret = kvm_s390_inject_program_int(vcpu, ret); + } else if (ret > 0) { + /* Translation not available */ + kvm_s390_set_psw_cc(vcpu, 3); + ret = 0; + } + goto out_unlock; + } + + hva = gfn_to_hva_prot(vcpu->kvm, gpa_to_gfn(gpa), &writable); + if (kvm_is_error_hva(hva)) { + ret = kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + } else { + if (!writable) + cc = 1; /* Write not permitted ==> read-only */ + kvm_s390_set_psw_cc(vcpu, cc); + /* Note: CC2 only occurs for storage keys (not supported yet) */ + } +out_unlock: + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_DAT) + ipte_unlock(vcpu); + return ret; +} + +int kvm_s390_handle_e5(struct kvm_vcpu *vcpu) +{ + /* For e5xx... instructions we only handle TPROT */ + if ((vcpu->arch.sie_block->ipa & 0x00ff) == 0x01) + return handle_tprot(vcpu); + return -EOPNOTSUPP; +} + +static int handle_sckpf(struct kvm_vcpu *vcpu) +{ + u32 value; + + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + if (vcpu->run->s.regs.gprs[0] & 0x00000000ffff0000) + return kvm_s390_inject_program_int(vcpu, + PGM_SPECIFICATION); + + value = vcpu->run->s.regs.gprs[0] & 0x000000000000ffff; + vcpu->arch.sie_block->todpr = value; + + return 0; +} + +static const intercept_handler_t x01_handlers[256] = { + [0x07] = handle_sckpf, +}; + +int kvm_s390_handle_01(struct kvm_vcpu *vcpu) +{ + intercept_handler_t handler; + + handler = x01_handlers[vcpu->arch.sie_block->ipa & 0x00ff]; + if (handler) + return handler(vcpu); + return -EOPNOTSUPP; +} diff --git a/arch/s390/kvm/sigp.c b/arch/s390/kvm/sigp.c new file mode 100644 index 000000000..72e58bd2b --- /dev/null +++ b/arch/s390/kvm/sigp.c @@ -0,0 +1,493 @@ +/* + * handling interprocessor communication + * + * Copyright IBM Corp. 2008, 2013 + * + * 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. + * + * Author(s): Carsten Otte <cotte@de.ibm.com> + * Christian Borntraeger <borntraeger@de.ibm.com> + * Christian Ehrhardt <ehrhardt@de.ibm.com> + */ + +#include <linux/kvm.h> +#include <linux/kvm_host.h> +#include <linux/slab.h> +#include <asm/sigp.h> +#include "gaccess.h" +#include "kvm-s390.h" +#include "trace.h" + +static int __sigp_sense(struct kvm_vcpu *vcpu, struct kvm_vcpu *dst_vcpu, + u64 *reg) +{ + struct kvm_s390_local_interrupt *li; + int cpuflags; + int rc; + int ext_call_pending; + + li = &dst_vcpu->arch.local_int; + + cpuflags = atomic_read(li->cpuflags); + ext_call_pending = kvm_s390_ext_call_pending(dst_vcpu); + if (!(cpuflags & CPUSTAT_STOPPED) && !ext_call_pending) + rc = SIGP_CC_ORDER_CODE_ACCEPTED; + else { + *reg &= 0xffffffff00000000UL; + if (ext_call_pending) + *reg |= SIGP_STATUS_EXT_CALL_PENDING; + if (cpuflags & CPUSTAT_STOPPED) + *reg |= SIGP_STATUS_STOPPED; + rc = SIGP_CC_STATUS_STORED; + } + + VCPU_EVENT(vcpu, 4, "sensed status of cpu %x rc %x", dst_vcpu->vcpu_id, + rc); + return rc; +} + +static int __inject_sigp_emergency(struct kvm_vcpu *vcpu, + struct kvm_vcpu *dst_vcpu) +{ + struct kvm_s390_irq irq = { + .type = KVM_S390_INT_EMERGENCY, + .u.emerg.code = vcpu->vcpu_id, + }; + int rc = 0; + + rc = kvm_s390_inject_vcpu(dst_vcpu, &irq); + if (!rc) + VCPU_EVENT(vcpu, 4, "sent sigp emerg to cpu %x", + dst_vcpu->vcpu_id); + + return rc ? rc : SIGP_CC_ORDER_CODE_ACCEPTED; +} + +static int __sigp_emergency(struct kvm_vcpu *vcpu, struct kvm_vcpu *dst_vcpu) +{ + return __inject_sigp_emergency(vcpu, dst_vcpu); +} + +static int __sigp_conditional_emergency(struct kvm_vcpu *vcpu, + struct kvm_vcpu *dst_vcpu, + u16 asn, u64 *reg) +{ + const u64 psw_int_mask = PSW_MASK_IO | PSW_MASK_EXT; + u16 p_asn, s_asn; + psw_t *psw; + u32 flags; + + flags = atomic_read(&dst_vcpu->arch.sie_block->cpuflags); + psw = &dst_vcpu->arch.sie_block->gpsw; + p_asn = dst_vcpu->arch.sie_block->gcr[4] & 0xffff; /* Primary ASN */ + s_asn = dst_vcpu->arch.sie_block->gcr[3] & 0xffff; /* Secondary ASN */ + + /* Inject the emergency signal? */ + if (!(flags & CPUSTAT_STOPPED) + || (psw->mask & psw_int_mask) != psw_int_mask + || ((flags & CPUSTAT_WAIT) && psw->addr != 0) + || (!(flags & CPUSTAT_WAIT) && (asn == p_asn || asn == s_asn))) { + return __inject_sigp_emergency(vcpu, dst_vcpu); + } else { + *reg &= 0xffffffff00000000UL; + *reg |= SIGP_STATUS_INCORRECT_STATE; + return SIGP_CC_STATUS_STORED; + } +} + +static int __sigp_external_call(struct kvm_vcpu *vcpu, + struct kvm_vcpu *dst_vcpu, u64 *reg) +{ + struct kvm_s390_irq irq = { + .type = KVM_S390_INT_EXTERNAL_CALL, + .u.extcall.code = vcpu->vcpu_id, + }; + int rc; + + rc = kvm_s390_inject_vcpu(dst_vcpu, &irq); + if (rc == -EBUSY) { + *reg &= 0xffffffff00000000UL; + *reg |= SIGP_STATUS_EXT_CALL_PENDING; + return SIGP_CC_STATUS_STORED; + } else if (rc == 0) { + VCPU_EVENT(vcpu, 4, "sent sigp ext call to cpu %x", + dst_vcpu->vcpu_id); + } + + return rc ? rc : SIGP_CC_ORDER_CODE_ACCEPTED; +} + +static int __sigp_stop(struct kvm_vcpu *vcpu, struct kvm_vcpu *dst_vcpu) +{ + struct kvm_s390_irq irq = { + .type = KVM_S390_SIGP_STOP, + }; + int rc; + + rc = kvm_s390_inject_vcpu(dst_vcpu, &irq); + if (rc == -EBUSY) + rc = SIGP_CC_BUSY; + else if (rc == 0) + VCPU_EVENT(vcpu, 4, "sent sigp stop to cpu %x", + dst_vcpu->vcpu_id); + + return rc; +} + +static int __sigp_stop_and_store_status(struct kvm_vcpu *vcpu, + struct kvm_vcpu *dst_vcpu, u64 *reg) +{ + struct kvm_s390_irq irq = { + .type = KVM_S390_SIGP_STOP, + .u.stop.flags = KVM_S390_STOP_FLAG_STORE_STATUS, + }; + int rc; + + rc = kvm_s390_inject_vcpu(dst_vcpu, &irq); + if (rc == -EBUSY) + rc = SIGP_CC_BUSY; + else if (rc == 0) + VCPU_EVENT(vcpu, 4, "sent sigp stop and store status to cpu %x", + dst_vcpu->vcpu_id); + + return rc; +} + +static int __sigp_set_arch(struct kvm_vcpu *vcpu, u32 parameter) +{ + int rc; + unsigned int i; + struct kvm_vcpu *v; + + switch (parameter & 0xff) { + case 0: + rc = SIGP_CC_NOT_OPERATIONAL; + break; + case 1: + case 2: + kvm_for_each_vcpu(i, v, vcpu->kvm) { + v->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID; + kvm_clear_async_pf_completion_queue(v); + } + + rc = SIGP_CC_ORDER_CODE_ACCEPTED; + break; + default: + rc = -EOPNOTSUPP; + } + return rc; +} + +static int __sigp_set_prefix(struct kvm_vcpu *vcpu, struct kvm_vcpu *dst_vcpu, + u32 address, u64 *reg) +{ + struct kvm_s390_irq irq = { + .type = KVM_S390_SIGP_SET_PREFIX, + .u.prefix.address = address & 0x7fffe000u, + }; + int rc; + + /* + * Make sure the new value is valid memory. We only need to check the + * first page, since address is 8k aligned and memory pieces are always + * at least 1MB aligned and have at least a size of 1MB. + */ + if (kvm_is_error_gpa(vcpu->kvm, irq.u.prefix.address)) { + *reg &= 0xffffffff00000000UL; + *reg |= SIGP_STATUS_INVALID_PARAMETER; + return SIGP_CC_STATUS_STORED; + } + + rc = kvm_s390_inject_vcpu(dst_vcpu, &irq); + if (rc == -EBUSY) { + *reg &= 0xffffffff00000000UL; + *reg |= SIGP_STATUS_INCORRECT_STATE; + return SIGP_CC_STATUS_STORED; + } else if (rc == 0) { + VCPU_EVENT(vcpu, 4, "set prefix of cpu %02x to %x", + dst_vcpu->vcpu_id, irq.u.prefix.address); + } + + return rc; +} + +static int __sigp_store_status_at_addr(struct kvm_vcpu *vcpu, + struct kvm_vcpu *dst_vcpu, + u32 addr, u64 *reg) +{ + int flags; + int rc; + + flags = atomic_read(dst_vcpu->arch.local_int.cpuflags); + if (!(flags & CPUSTAT_STOPPED)) { + *reg &= 0xffffffff00000000UL; + *reg |= SIGP_STATUS_INCORRECT_STATE; + return SIGP_CC_STATUS_STORED; + } + + addr &= 0x7ffffe00; + rc = kvm_s390_store_status_unloaded(dst_vcpu, addr); + if (rc == -EFAULT) { + *reg &= 0xffffffff00000000UL; + *reg |= SIGP_STATUS_INVALID_PARAMETER; + rc = SIGP_CC_STATUS_STORED; + } + return rc; +} + +static int __sigp_sense_running(struct kvm_vcpu *vcpu, + struct kvm_vcpu *dst_vcpu, u64 *reg) +{ + struct kvm_s390_local_interrupt *li; + int rc; + + li = &dst_vcpu->arch.local_int; + if (atomic_read(li->cpuflags) & CPUSTAT_RUNNING) { + /* running */ + rc = SIGP_CC_ORDER_CODE_ACCEPTED; + } else { + /* not running */ + *reg &= 0xffffffff00000000UL; + *reg |= SIGP_STATUS_NOT_RUNNING; + rc = SIGP_CC_STATUS_STORED; + } + + VCPU_EVENT(vcpu, 4, "sensed running status of cpu %x rc %x", + dst_vcpu->vcpu_id, rc); + + return rc; +} + +static int __prepare_sigp_re_start(struct kvm_vcpu *vcpu, + struct kvm_vcpu *dst_vcpu, u8 order_code) +{ + struct kvm_s390_local_interrupt *li = &dst_vcpu->arch.local_int; + /* handle (RE)START in user space */ + int rc = -EOPNOTSUPP; + + /* make sure we don't race with STOP irq injection */ + spin_lock(&li->lock); + if (kvm_s390_is_stop_irq_pending(dst_vcpu)) + rc = SIGP_CC_BUSY; + spin_unlock(&li->lock); + + return rc; +} + +static int __prepare_sigp_cpu_reset(struct kvm_vcpu *vcpu, + struct kvm_vcpu *dst_vcpu, u8 order_code) +{ + /* handle (INITIAL) CPU RESET in user space */ + return -EOPNOTSUPP; +} + +static int __prepare_sigp_unknown(struct kvm_vcpu *vcpu, + struct kvm_vcpu *dst_vcpu) +{ + /* handle unknown orders in user space */ + return -EOPNOTSUPP; +} + +static int handle_sigp_dst(struct kvm_vcpu *vcpu, u8 order_code, + u16 cpu_addr, u32 parameter, u64 *status_reg) +{ + int rc; + struct kvm_vcpu *dst_vcpu; + + if (cpu_addr >= KVM_MAX_VCPUS) + return SIGP_CC_NOT_OPERATIONAL; + + dst_vcpu = kvm_get_vcpu(vcpu->kvm, cpu_addr); + if (!dst_vcpu) + return SIGP_CC_NOT_OPERATIONAL; + + switch (order_code) { + case SIGP_SENSE: + vcpu->stat.instruction_sigp_sense++; + rc = __sigp_sense(vcpu, dst_vcpu, status_reg); + break; + case SIGP_EXTERNAL_CALL: + vcpu->stat.instruction_sigp_external_call++; + rc = __sigp_external_call(vcpu, dst_vcpu, status_reg); + break; + case SIGP_EMERGENCY_SIGNAL: + vcpu->stat.instruction_sigp_emergency++; + rc = __sigp_emergency(vcpu, dst_vcpu); + break; + case SIGP_STOP: + vcpu->stat.instruction_sigp_stop++; + rc = __sigp_stop(vcpu, dst_vcpu); + break; + case SIGP_STOP_AND_STORE_STATUS: + vcpu->stat.instruction_sigp_stop_store_status++; + rc = __sigp_stop_and_store_status(vcpu, dst_vcpu, status_reg); + break; + case SIGP_STORE_STATUS_AT_ADDRESS: + vcpu->stat.instruction_sigp_store_status++; + rc = __sigp_store_status_at_addr(vcpu, dst_vcpu, parameter, + status_reg); + break; + case SIGP_SET_PREFIX: + vcpu->stat.instruction_sigp_prefix++; + rc = __sigp_set_prefix(vcpu, dst_vcpu, parameter, status_reg); + break; + case SIGP_COND_EMERGENCY_SIGNAL: + vcpu->stat.instruction_sigp_cond_emergency++; + rc = __sigp_conditional_emergency(vcpu, dst_vcpu, parameter, + status_reg); + break; + case SIGP_SENSE_RUNNING: + vcpu->stat.instruction_sigp_sense_running++; + rc = __sigp_sense_running(vcpu, dst_vcpu, status_reg); + break; + case SIGP_START: + vcpu->stat.instruction_sigp_start++; + rc = __prepare_sigp_re_start(vcpu, dst_vcpu, order_code); + break; + case SIGP_RESTART: + vcpu->stat.instruction_sigp_restart++; + rc = __prepare_sigp_re_start(vcpu, dst_vcpu, order_code); + break; + case SIGP_INITIAL_CPU_RESET: + vcpu->stat.instruction_sigp_init_cpu_reset++; + rc = __prepare_sigp_cpu_reset(vcpu, dst_vcpu, order_code); + break; + case SIGP_CPU_RESET: + vcpu->stat.instruction_sigp_cpu_reset++; + rc = __prepare_sigp_cpu_reset(vcpu, dst_vcpu, order_code); + break; + default: + vcpu->stat.instruction_sigp_unknown++; + rc = __prepare_sigp_unknown(vcpu, dst_vcpu); + } + + if (rc == -EOPNOTSUPP) + VCPU_EVENT(vcpu, 4, + "sigp order %u -> cpu %x: handled in user space", + order_code, dst_vcpu->vcpu_id); + + return rc; +} + +static int handle_sigp_order_in_user_space(struct kvm_vcpu *vcpu, u8 order_code) +{ + if (!vcpu->kvm->arch.user_sigp) + return 0; + + switch (order_code) { + case SIGP_SENSE: + case SIGP_EXTERNAL_CALL: + case SIGP_EMERGENCY_SIGNAL: + case SIGP_COND_EMERGENCY_SIGNAL: + case SIGP_SENSE_RUNNING: + return 0; + /* update counters as we're directly dropping to user space */ + case SIGP_STOP: + vcpu->stat.instruction_sigp_stop++; + break; + case SIGP_STOP_AND_STORE_STATUS: + vcpu->stat.instruction_sigp_stop_store_status++; + break; + case SIGP_STORE_STATUS_AT_ADDRESS: + vcpu->stat.instruction_sigp_store_status++; + break; + case SIGP_STORE_ADDITIONAL_STATUS: + vcpu->stat.instruction_sigp_store_adtl_status++; + break; + case SIGP_SET_PREFIX: + vcpu->stat.instruction_sigp_prefix++; + break; + case SIGP_START: + vcpu->stat.instruction_sigp_start++; + break; + case SIGP_RESTART: + vcpu->stat.instruction_sigp_restart++; + break; + case SIGP_INITIAL_CPU_RESET: + vcpu->stat.instruction_sigp_init_cpu_reset++; + break; + case SIGP_CPU_RESET: + vcpu->stat.instruction_sigp_cpu_reset++; + break; + default: + vcpu->stat.instruction_sigp_unknown++; + } + + VCPU_EVENT(vcpu, 4, "sigp order %u: completely handled in user space", + order_code); + + return 1; +} + +int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu) +{ + int r1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int r3 = vcpu->arch.sie_block->ipa & 0x000f; + u32 parameter; + u16 cpu_addr = vcpu->run->s.regs.gprs[r3]; + u8 order_code; + int rc; + + /* sigp in userspace can exit */ + if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) + return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + + order_code = kvm_s390_get_base_disp_rs(vcpu, NULL); + if (handle_sigp_order_in_user_space(vcpu, order_code)) + return -EOPNOTSUPP; + + if (r1 % 2) + parameter = vcpu->run->s.regs.gprs[r1]; + else + parameter = vcpu->run->s.regs.gprs[r1 + 1]; + + trace_kvm_s390_handle_sigp(vcpu, order_code, cpu_addr, parameter); + switch (order_code) { + case SIGP_SET_ARCHITECTURE: + vcpu->stat.instruction_sigp_arch++; + rc = __sigp_set_arch(vcpu, parameter); + break; + default: + rc = handle_sigp_dst(vcpu, order_code, cpu_addr, + parameter, + &vcpu->run->s.regs.gprs[r1]); + } + + if (rc < 0) + return rc; + + kvm_s390_set_psw_cc(vcpu, rc); + return 0; +} + +/* + * Handle SIGP partial execution interception. + * + * This interception will occur at the source cpu when a source cpu sends an + * external call to a target cpu and the target cpu has the WAIT bit set in + * its cpuflags. Interception will occurr after the interrupt indicator bits at + * the target cpu have been set. All error cases will lead to instruction + * interception, therefore nothing is to be checked or prepared. + */ +int kvm_s390_handle_sigp_pei(struct kvm_vcpu *vcpu) +{ + int r3 = vcpu->arch.sie_block->ipa & 0x000f; + u16 cpu_addr = vcpu->run->s.regs.gprs[r3]; + struct kvm_vcpu *dest_vcpu; + u8 order_code = kvm_s390_get_base_disp_rs(vcpu, NULL); + + trace_kvm_s390_handle_sigp_pei(vcpu, order_code, cpu_addr); + + if (order_code == SIGP_EXTERNAL_CALL) { + dest_vcpu = kvm_get_vcpu(vcpu->kvm, cpu_addr); + BUG_ON(dest_vcpu == NULL); + + kvm_s390_vcpu_wakeup(dest_vcpu); + kvm_s390_set_psw_cc(vcpu, SIGP_CC_ORDER_CODE_ACCEPTED); + return 0; + } + + return -EOPNOTSUPP; +} diff --git a/arch/s390/kvm/trace-s390.h b/arch/s390/kvm/trace-s390.h new file mode 100644 index 000000000..3208d33a4 --- /dev/null +++ b/arch/s390/kvm/trace-s390.h @@ -0,0 +1,282 @@ +#if !defined(_TRACE_KVMS390_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_KVMS390_H + +#include <linux/tracepoint.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM kvm-s390 +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE trace-s390 + +/* + * The TRACE_SYSTEM_VAR defaults to TRACE_SYSTEM, but must be a + * legitimate C variable. It is not exported to user space. + */ +#undef TRACE_SYSTEM_VAR +#define TRACE_SYSTEM_VAR kvm_s390 + +/* + * Trace point for the creation of the kvm instance. + */ +TRACE_EVENT(kvm_s390_create_vm, + TP_PROTO(unsigned long type), + TP_ARGS(type), + + TP_STRUCT__entry( + __field(unsigned long, type) + ), + + TP_fast_assign( + __entry->type = type; + ), + + TP_printk("create vm%s", + __entry->type & KVM_VM_S390_UCONTROL ? " (UCONTROL)" : "") + ); + +/* + * Trace points for creation and destruction of vpcus. + */ +TRACE_EVENT(kvm_s390_create_vcpu, + TP_PROTO(unsigned int id, struct kvm_vcpu *vcpu, + struct kvm_s390_sie_block *sie_block), + TP_ARGS(id, vcpu, sie_block), + + TP_STRUCT__entry( + __field(unsigned int, id) + __field(struct kvm_vcpu *, vcpu) + __field(struct kvm_s390_sie_block *, sie_block) + ), + + TP_fast_assign( + __entry->id = id; + __entry->vcpu = vcpu; + __entry->sie_block = sie_block; + ), + + TP_printk("create cpu %d at %p, sie block at %p", __entry->id, + __entry->vcpu, __entry->sie_block) + ); + +TRACE_EVENT(kvm_s390_destroy_vcpu, + TP_PROTO(unsigned int id), + TP_ARGS(id), + + TP_STRUCT__entry( + __field(unsigned int, id) + ), + + TP_fast_assign( + __entry->id = id; + ), + + TP_printk("destroy cpu %d", __entry->id) + ); + +/* + * Trace point for start and stop of vpcus. + */ +TRACE_EVENT(kvm_s390_vcpu_start_stop, + TP_PROTO(unsigned int id, int state), + TP_ARGS(id, state), + + TP_STRUCT__entry( + __field(unsigned int, id) + __field(int, state) + ), + + TP_fast_assign( + __entry->id = id; + __entry->state = state; + ), + + TP_printk("%s cpu %d", __entry->state ? "starting" : "stopping", + __entry->id) + ); + +/* + * Trace points for injection of interrupts, either per machine or + * per vcpu. + */ + +#define kvm_s390_int_type \ + {KVM_S390_SIGP_STOP, "sigp stop"}, \ + {KVM_S390_PROGRAM_INT, "program interrupt"}, \ + {KVM_S390_SIGP_SET_PREFIX, "sigp set prefix"}, \ + {KVM_S390_RESTART, "sigp restart"}, \ + {KVM_S390_INT_VIRTIO, "virtio interrupt"}, \ + {KVM_S390_INT_SERVICE, "sclp interrupt"}, \ + {KVM_S390_INT_EMERGENCY, "sigp emergency"}, \ + {KVM_S390_INT_EXTERNAL_CALL, "sigp ext call"} + +TRACE_EVENT(kvm_s390_inject_vm, + TP_PROTO(__u64 type, __u32 parm, __u64 parm64, int who), + TP_ARGS(type, parm, parm64, who), + + TP_STRUCT__entry( + __field(__u32, inttype) + __field(__u32, parm) + __field(__u64, parm64) + __field(int, who) + ), + + TP_fast_assign( + __entry->inttype = type & 0x00000000ffffffff; + __entry->parm = parm; + __entry->parm64 = parm64; + __entry->who = who; + ), + + TP_printk("inject%s: type:%x (%s) parm:%x parm64:%llx", + (__entry->who == 1) ? " (from kernel)" : + (__entry->who == 2) ? " (from user)" : "", + __entry->inttype, + __print_symbolic(__entry->inttype, kvm_s390_int_type), + __entry->parm, __entry->parm64) + ); + +TRACE_EVENT(kvm_s390_inject_vcpu, + TP_PROTO(unsigned int id, __u64 type, __u32 parm, __u64 parm64, \ + int who), + TP_ARGS(id, type, parm, parm64, who), + + TP_STRUCT__entry( + __field(int, id) + __field(__u32, inttype) + __field(__u32, parm) + __field(__u64, parm64) + __field(int, who) + ), + + TP_fast_assign( + __entry->id = id; + __entry->inttype = type & 0x00000000ffffffff; + __entry->parm = parm; + __entry->parm64 = parm64; + __entry->who = who; + ), + + TP_printk("inject%s (vcpu %d): type:%x (%s) parm:%x parm64:%llx", + (__entry->who == 1) ? " (from kernel)" : + (__entry->who == 2) ? " (from user)" : "", + __entry->id, __entry->inttype, + __print_symbolic(__entry->inttype, kvm_s390_int_type), + __entry->parm, __entry->parm64) + ); + +/* + * Trace point for the actual delivery of interrupts. + */ +TRACE_EVENT(kvm_s390_deliver_interrupt, + TP_PROTO(unsigned int id, __u64 type, __u64 data0, __u64 data1), + TP_ARGS(id, type, data0, data1), + + TP_STRUCT__entry( + __field(int, id) + __field(__u32, inttype) + __field(__u64, data0) + __field(__u64, data1) + ), + + TP_fast_assign( + __entry->id = id; + __entry->inttype = type & 0x00000000ffffffff; + __entry->data0 = data0; + __entry->data1 = data1; + ), + + TP_printk("deliver interrupt (vcpu %d): type:%x (%s) " \ + "data:%08llx %016llx", + __entry->id, __entry->inttype, + __print_symbolic(__entry->inttype, kvm_s390_int_type), + __entry->data0, __entry->data1) + ); + +/* + * Trace point for resets that may be requested from userspace. + */ +TRACE_EVENT(kvm_s390_request_resets, + TP_PROTO(__u64 resets), + TP_ARGS(resets), + + TP_STRUCT__entry( + __field(__u64, resets) + ), + + TP_fast_assign( + __entry->resets = resets; + ), + + TP_printk("requesting userspace resets %llx", + __entry->resets) + ); + +/* + * Trace point for a vcpu's stop requests. + */ +TRACE_EVENT(kvm_s390_stop_request, + TP_PROTO(unsigned char stop_irq, unsigned char flags), + TP_ARGS(stop_irq, flags), + + TP_STRUCT__entry( + __field(unsigned char, stop_irq) + __field(unsigned char, flags) + ), + + TP_fast_assign( + __entry->stop_irq = stop_irq; + __entry->flags = flags; + ), + + TP_printk("stop request, stop irq = %u, flags = %08x", + __entry->stop_irq, __entry->flags) + ); + + +/* + * Trace point for enabling channel I/O instruction support. + */ +TRACE_EVENT(kvm_s390_enable_css, + TP_PROTO(void *kvm), + TP_ARGS(kvm), + + TP_STRUCT__entry( + __field(void *, kvm) + ), + + TP_fast_assign( + __entry->kvm = kvm; + ), + + TP_printk("enabling channel I/O support (kvm @ %p)\n", + __entry->kvm) + ); + +/* + * Trace point for enabling and disabling interlocking-and-broadcasting + * suppression. + */ +TRACE_EVENT(kvm_s390_enable_disable_ibs, + TP_PROTO(unsigned int id, int state), + TP_ARGS(id, state), + + TP_STRUCT__entry( + __field(unsigned int, id) + __field(int, state) + ), + + TP_fast_assign( + __entry->id = id; + __entry->state = state; + ), + + TP_printk("%s ibs on cpu %d", + __entry->state ? "enabling" : "disabling", __entry->id) + ); + + +#endif /* _TRACE_KVMS390_H */ + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/arch/s390/kvm/trace.h b/arch/s390/kvm/trace.h new file mode 100644 index 000000000..916834d7a --- /dev/null +++ b/arch/s390/kvm/trace.h @@ -0,0 +1,418 @@ +#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_KVM_H + +#include <linux/tracepoint.h> +#include <asm/sie.h> +#include <asm/debug.h> +#include <asm/dis.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM kvm +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE trace + +/* + * Helpers for vcpu-specific tracepoints containing the same information + * as s390dbf VCPU_EVENTs. + */ +#define VCPU_PROTO_COMMON struct kvm_vcpu *vcpu +#define VCPU_ARGS_COMMON vcpu +#define VCPU_FIELD_COMMON __field(int, id) \ + __field(unsigned long, pswmask) \ + __field(unsigned long, pswaddr) +#define VCPU_ASSIGN_COMMON do { \ + __entry->id = vcpu->vcpu_id; \ + __entry->pswmask = vcpu->arch.sie_block->gpsw.mask; \ + __entry->pswaddr = vcpu->arch.sie_block->gpsw.addr; \ + } while (0); +#define VCPU_TP_PRINTK(p_str, p_args...) \ + TP_printk("%02d[%016lx-%016lx]: " p_str, __entry->id, \ + __entry->pswmask, __entry->pswaddr, p_args) + +TRACE_EVENT(kvm_s390_skey_related_inst, + TP_PROTO(VCPU_PROTO_COMMON), + TP_ARGS(VCPU_ARGS_COMMON), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + ), + VCPU_TP_PRINTK("%s", "first instruction related to skeys on vcpu") + ); + +TRACE_EVENT(kvm_s390_major_guest_pfault, + TP_PROTO(VCPU_PROTO_COMMON), + TP_ARGS(VCPU_ARGS_COMMON), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + ), + VCPU_TP_PRINTK("%s", "major fault, maybe applicable for pfault") + ); + +TRACE_EVENT(kvm_s390_pfault_init, + TP_PROTO(VCPU_PROTO_COMMON, long pfault_token), + TP_ARGS(VCPU_ARGS_COMMON, pfault_token), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(long, pfault_token) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->pfault_token = pfault_token; + ), + VCPU_TP_PRINTK("init pfault token %ld", __entry->pfault_token) + ); + +TRACE_EVENT(kvm_s390_pfault_done, + TP_PROTO(VCPU_PROTO_COMMON, long pfault_token), + TP_ARGS(VCPU_ARGS_COMMON, pfault_token), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(long, pfault_token) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->pfault_token = pfault_token; + ), + VCPU_TP_PRINTK("done pfault token %ld", __entry->pfault_token) + ); + +/* + * Tracepoints for SIE entry and exit. + */ +TRACE_EVENT(kvm_s390_sie_enter, + TP_PROTO(VCPU_PROTO_COMMON, int cpuflags), + TP_ARGS(VCPU_ARGS_COMMON, cpuflags), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(int, cpuflags) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->cpuflags = cpuflags; + ), + + VCPU_TP_PRINTK("entering sie flags %x", __entry->cpuflags) + ); + +TRACE_EVENT(kvm_s390_sie_fault, + TP_PROTO(VCPU_PROTO_COMMON), + TP_ARGS(VCPU_ARGS_COMMON), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + ), + + VCPU_TP_PRINTK("%s", "fault in sie instruction") + ); + +TRACE_EVENT(kvm_s390_sie_exit, + TP_PROTO(VCPU_PROTO_COMMON, u8 icptcode), + TP_ARGS(VCPU_ARGS_COMMON, icptcode), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(u8, icptcode) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->icptcode = icptcode; + ), + + VCPU_TP_PRINTK("exit sie icptcode %d (%s)", __entry->icptcode, + __print_symbolic(__entry->icptcode, + sie_intercept_code)) + ); + +/* + * Trace point for intercepted instructions. + */ +TRACE_EVENT(kvm_s390_intercept_instruction, + TP_PROTO(VCPU_PROTO_COMMON, __u16 ipa, __u32 ipb), + TP_ARGS(VCPU_ARGS_COMMON, ipa, ipb), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(__u64, instruction) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->instruction = ((__u64)ipa << 48) | + ((__u64)ipb << 16); + ), + + VCPU_TP_PRINTK("intercepted instruction %016llx (%s)", + __entry->instruction, + __print_symbolic(icpt_insn_decoder(__entry->instruction), + icpt_insn_codes)) + ); + +/* + * Trace point for intercepted program interruptions. + */ +TRACE_EVENT(kvm_s390_intercept_prog, + TP_PROTO(VCPU_PROTO_COMMON, __u16 code), + TP_ARGS(VCPU_ARGS_COMMON, code), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(__u16, code) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->code = code; + ), + + VCPU_TP_PRINTK("intercepted program interruption %04x", + __entry->code) + ); + +/* + * Trace point for validity intercepts. + */ +TRACE_EVENT(kvm_s390_intercept_validity, + TP_PROTO(VCPU_PROTO_COMMON, __u16 viwhy), + TP_ARGS(VCPU_ARGS_COMMON, viwhy), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(__u16, viwhy) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->viwhy = viwhy; + ), + + VCPU_TP_PRINTK("got validity intercept %04x", __entry->viwhy) + ); + +/* + * Trace points for instructions that are of special interest. + */ + +TRACE_EVENT(kvm_s390_handle_sigp, + TP_PROTO(VCPU_PROTO_COMMON, __u8 order_code, __u16 cpu_addr, \ + __u32 parameter), + TP_ARGS(VCPU_ARGS_COMMON, order_code, cpu_addr, parameter), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(__u8, order_code) + __field(__u16, cpu_addr) + __field(__u32, parameter) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->order_code = order_code; + __entry->cpu_addr = cpu_addr; + __entry->parameter = parameter; + ), + + VCPU_TP_PRINTK("handle sigp order %02x (%s), cpu address %04x, " \ + "parameter %08x", __entry->order_code, + __print_symbolic(__entry->order_code, + sigp_order_codes), + __entry->cpu_addr, __entry->parameter) + ); + +TRACE_EVENT(kvm_s390_handle_sigp_pei, + TP_PROTO(VCPU_PROTO_COMMON, __u8 order_code, __u16 cpu_addr), + TP_ARGS(VCPU_ARGS_COMMON, order_code, cpu_addr), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(__u8, order_code) + __field(__u16, cpu_addr) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->order_code = order_code; + __entry->cpu_addr = cpu_addr; + ), + + VCPU_TP_PRINTK("handle sigp pei order %02x (%s), cpu address %04x", + __entry->order_code, + __print_symbolic(__entry->order_code, + sigp_order_codes), + __entry->cpu_addr) + ); + +TRACE_EVENT(kvm_s390_handle_diag, + TP_PROTO(VCPU_PROTO_COMMON, __u16 code), + TP_ARGS(VCPU_ARGS_COMMON, code), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(__u16, code) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->code = code; + ), + + VCPU_TP_PRINTK("handle diagnose call %04x (%s)", __entry->code, + __print_symbolic(__entry->code, diagnose_codes)) + ); + +TRACE_EVENT(kvm_s390_handle_lctl, + TP_PROTO(VCPU_PROTO_COMMON, int g, int reg1, int reg3, u64 addr), + TP_ARGS(VCPU_ARGS_COMMON, g, reg1, reg3, addr), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(int, g) + __field(int, reg1) + __field(int, reg3) + __field(u64, addr) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->g = g; + __entry->reg1 = reg1; + __entry->reg3 = reg3; + __entry->addr = addr; + ), + + VCPU_TP_PRINTK("%s: loading cr %x-%x from %016llx", + __entry->g ? "lctlg" : "lctl", + __entry->reg1, __entry->reg3, __entry->addr) + ); + +TRACE_EVENT(kvm_s390_handle_stctl, + TP_PROTO(VCPU_PROTO_COMMON, int g, int reg1, int reg3, u64 addr), + TP_ARGS(VCPU_ARGS_COMMON, g, reg1, reg3, addr), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(int, g) + __field(int, reg1) + __field(int, reg3) + __field(u64, addr) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->g = g; + __entry->reg1 = reg1; + __entry->reg3 = reg3; + __entry->addr = addr; + ), + + VCPU_TP_PRINTK("%s: storing cr %x-%x to %016llx", + __entry->g ? "stctg" : "stctl", + __entry->reg1, __entry->reg3, __entry->addr) + ); + +TRACE_EVENT(kvm_s390_handle_prefix, + TP_PROTO(VCPU_PROTO_COMMON, int set, u32 address), + TP_ARGS(VCPU_ARGS_COMMON, set, address), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(int, set) + __field(u32, address) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->set = set; + __entry->address = address; + ), + + VCPU_TP_PRINTK("%s prefix to %08x", + __entry->set ? "setting" : "storing", + __entry->address) + ); + +TRACE_EVENT(kvm_s390_handle_stap, + TP_PROTO(VCPU_PROTO_COMMON, u64 address), + TP_ARGS(VCPU_ARGS_COMMON, address), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(u64, address) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->address = address; + ), + + VCPU_TP_PRINTK("storing cpu address to %016llx", + __entry->address) + ); + +TRACE_EVENT(kvm_s390_handle_stfl, + TP_PROTO(VCPU_PROTO_COMMON, unsigned int facility_list), + TP_ARGS(VCPU_ARGS_COMMON, facility_list), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(unsigned int, facility_list) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->facility_list = facility_list; + ), + + VCPU_TP_PRINTK("store facility list value %08x", + __entry->facility_list) + ); + +TRACE_EVENT(kvm_s390_handle_stsi, + TP_PROTO(VCPU_PROTO_COMMON, int fc, int sel1, int sel2, u64 addr), + TP_ARGS(VCPU_ARGS_COMMON, fc, sel1, sel2, addr), + + TP_STRUCT__entry( + VCPU_FIELD_COMMON + __field(int, fc) + __field(int, sel1) + __field(int, sel2) + __field(u64, addr) + ), + + TP_fast_assign( + VCPU_ASSIGN_COMMON + __entry->fc = fc; + __entry->sel1 = sel1; + __entry->sel2 = sel2; + __entry->addr = addr; + ), + + VCPU_TP_PRINTK("STSI %d.%d.%d information stored to %016llx", + __entry->fc, __entry->sel1, __entry->sel2, + __entry->addr) + ); + +#endif /* _TRACE_KVM_H */ + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/arch/s390/lib/Makefile b/arch/s390/lib/Makefile new file mode 100644 index 000000000..0e8fefe5b --- /dev/null +++ b/arch/s390/lib/Makefile @@ -0,0 +1,9 @@ +# +# Makefile for s390-specific library files.. +# + +lib-y += delay.o string.o uaccess.o find.o +obj-y += mem.o +lib-$(CONFIG_SMP) += spinlock.o +lib-$(CONFIG_KPROBES) += probes.o +lib-$(CONFIG_UPROBES) += probes.o diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c new file mode 100644 index 000000000..16dc42d83 --- /dev/null +++ b/arch/s390/lib/delay.c @@ -0,0 +1,129 @@ +/* + * Precise Delay Loops for S390 + * + * Copyright IBM Corp. 1999, 2008 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>, + * Heiko Carstens <heiko.carstens@de.ibm.com>, + */ + +#include <linux/sched.h> +#include <linux/delay.h> +#include <linux/timex.h> +#include <linux/module.h> +#include <linux/irqflags.h> +#include <linux/interrupt.h> +#include <asm/vtimer.h> +#include <asm/div64.h> + +void __delay(unsigned long loops) +{ + /* + * To end the bloody studid and useless discussion about the + * BogoMips number I took the liberty to define the __delay + * function in a way that that resulting BogoMips number will + * yield the megahertz number of the cpu. The important function + * is udelay and that is done using the tod clock. -- martin. + */ + asm volatile("0: brct %0,0b" : : "d" ((loops/2) + 1)); +} + +static void __udelay_disabled(unsigned long long usecs) +{ + unsigned long cr0, cr6, new; + u64 clock_saved, end; + + end = get_tod_clock() + (usecs << 12); + clock_saved = local_tick_disable(); + __ctl_store(cr0, 0, 0); + __ctl_store(cr6, 6, 6); + new = (cr0 & 0xffff00e0) | 0x00000800; + __ctl_load(new , 0, 0); + new = 0; + __ctl_load(new, 6, 6); + lockdep_off(); + do { + set_clock_comparator(end); + enabled_wait(); + } while (get_tod_clock_fast() < end); + lockdep_on(); + __ctl_load(cr0, 0, 0); + __ctl_load(cr6, 6, 6); + local_tick_enable(clock_saved); +} + +static void __udelay_enabled(unsigned long long usecs) +{ + u64 clock_saved, end; + + end = get_tod_clock_fast() + (usecs << 12); + do { + clock_saved = 0; + if (end < S390_lowcore.clock_comparator) { + clock_saved = local_tick_disable(); + set_clock_comparator(end); + } + enabled_wait(); + if (clock_saved) + local_tick_enable(clock_saved); + } while (get_tod_clock_fast() < end); +} + +/* + * Waits for 'usecs' microseconds using the TOD clock comparator. + */ +void __udelay(unsigned long long usecs) +{ + unsigned long flags; + + preempt_disable(); + local_irq_save(flags); + if (in_irq()) { + __udelay_disabled(usecs); + goto out; + } + if (in_softirq()) { + if (raw_irqs_disabled_flags(flags)) + __udelay_disabled(usecs); + else + __udelay_enabled(usecs); + goto out; + } + if (raw_irqs_disabled_flags(flags)) { + local_bh_disable(); + __udelay_disabled(usecs); + _local_bh_enable(); + goto out; + } + __udelay_enabled(usecs); +out: + local_irq_restore(flags); + preempt_enable(); +} +EXPORT_SYMBOL(__udelay); + +/* + * Simple udelay variant. To be used on startup and reboot + * when the interrupt handler isn't working. + */ +void udelay_simple(unsigned long long usecs) +{ + u64 end; + + end = get_tod_clock_fast() + (usecs << 12); + while (get_tod_clock_fast() < end) + cpu_relax(); +} + +void __ndelay(unsigned long long nsecs) +{ + u64 end; + + nsecs <<= 9; + do_div(nsecs, 125); + end = get_tod_clock_fast() + nsecs; + if (nsecs & ~0xfffUL) + __udelay(nsecs >> 12); + while (get_tod_clock_fast() < end) + barrier(); +} +EXPORT_SYMBOL(__ndelay); diff --git a/arch/s390/lib/find.c b/arch/s390/lib/find.c new file mode 100644 index 000000000..922003c1b --- /dev/null +++ b/arch/s390/lib/find.c @@ -0,0 +1,77 @@ +/* + * MSB0 numbered special bitops handling. + * + * On s390x the bits are numbered: + * |0..............63|64............127|128...........191|192...........255| + * and on s390: + * |0.....31|32....63|64....95|96...127|128..159|160..191|192..223|224..255| + * + * The reason for this bit numbering is the fact that the hardware sets bits + * in a bitmap starting at bit 0 (MSB) and we don't want to scan the bitmap + * from the 'wrong end'. + */ + +#include <linux/compiler.h> +#include <linux/bitops.h> +#include <linux/export.h> + +unsigned long find_first_bit_inv(const unsigned long *addr, unsigned long size) +{ + const unsigned long *p = addr; + unsigned long result = 0; + unsigned long tmp; + + while (size & ~(BITS_PER_LONG - 1)) { + if ((tmp = *(p++))) + goto found; + result += BITS_PER_LONG; + size -= BITS_PER_LONG; + } + if (!size) + return result; + tmp = (*p) & (~0UL << (BITS_PER_LONG - size)); + if (!tmp) /* Are any bits set? */ + return result + size; /* Nope. */ +found: + return result + (__fls(tmp) ^ (BITS_PER_LONG - 1)); +} +EXPORT_SYMBOL(find_first_bit_inv); + +unsigned long find_next_bit_inv(const unsigned long *addr, unsigned long size, + unsigned long offset) +{ + const unsigned long *p = addr + (offset / BITS_PER_LONG); + unsigned long result = offset & ~(BITS_PER_LONG - 1); + unsigned long tmp; + + if (offset >= size) + return size; + size -= result; + offset %= BITS_PER_LONG; + if (offset) { + tmp = *(p++); + tmp &= (~0UL >> offset); + if (size < BITS_PER_LONG) + goto found_first; + if (tmp) + goto found_middle; + size -= BITS_PER_LONG; + result += BITS_PER_LONG; + } + while (size & ~(BITS_PER_LONG-1)) { + if ((tmp = *(p++))) + goto found_middle; + result += BITS_PER_LONG; + size -= BITS_PER_LONG; + } + if (!size) + return result; + tmp = *p; +found_first: + tmp &= (~0UL << (BITS_PER_LONG - size)); + if (!tmp) /* Are any bits set? */ + return result + size; /* Nope. */ +found_middle: + return result + (__fls(tmp) ^ (BITS_PER_LONG - 1)); +} +EXPORT_SYMBOL(find_next_bit_inv); diff --git a/arch/s390/lib/mem.S b/arch/s390/lib/mem.S new file mode 100644 index 000000000..c6d553e85 --- /dev/null +++ b/arch/s390/lib/mem.S @@ -0,0 +1,88 @@ +/* + * String handling functions. + * + * Copyright IBM Corp. 2012 + */ + +#include <linux/linkage.h> + +/* + * memset implementation + * + * This code corresponds to the C construct below. We do distinguish + * between clearing (c == 0) and setting a memory array (c != 0) simply + * because nearly all memset invocations in the kernel clear memory and + * the xc instruction is preferred in such cases. + * + * void *memset(void *s, int c, size_t n) + * { + * if (likely(c == 0)) + * return __builtin_memset(s, 0, n); + * return __builtin_memset(s, c, n); + * } + */ +ENTRY(memset) + ltgr %r4,%r4 + bzr %r14 + ltgr %r3,%r3 + jnz .Lmemset_fill + aghi %r4,-1 + srlg %r3,%r4,8 + ltgr %r3,%r3 + lgr %r1,%r2 + jz .Lmemset_clear_rest +.Lmemset_clear_loop: + xc 0(256,%r1),0(%r1) + la %r1,256(%r1) + brctg %r3,.Lmemset_clear_loop +.Lmemset_clear_rest: + larl %r3,.Lmemset_xc + ex %r4,0(%r3) + br %r14 +.Lmemset_fill: + stc %r3,0(%r2) + cghi %r4,1 + lgr %r1,%r2 + ber %r14 + aghi %r4,-2 + srlg %r3,%r4,8 + ltgr %r3,%r3 + jz .Lmemset_fill_rest +.Lmemset_fill_loop: + mvc 1(256,%r1),0(%r1) + la %r1,256(%r1) + brctg %r3,.Lmemset_fill_loop +.Lmemset_fill_rest: + larl %r3,.Lmemset_mvc + ex %r4,0(%r3) + br %r14 +.Lmemset_xc: + xc 0(1,%r1),0(%r1) +.Lmemset_mvc: + mvc 1(1,%r1),0(%r1) + +/* + * memcpy implementation + * + * void *memcpy(void *dest, const void *src, size_t n) + */ +ENTRY(memcpy) + ltgr %r4,%r4 + bzr %r14 + aghi %r4,-1 + srlg %r5,%r4,8 + ltgr %r5,%r5 + lgr %r1,%r2 + jnz .Lmemcpy_loop +.Lmemcpy_rest: + larl %r5,.Lmemcpy_mvc + ex %r4,0(%r5) + br %r14 +.Lmemcpy_loop: + mvc 0(256,%r1),0(%r3) + la %r1,256(%r1) + la %r3,256(%r3) + brctg %r5,.Lmemcpy_loop + j .Lmemcpy_rest +.Lmemcpy_mvc: + mvc 0(1,%r1),0(%r3) diff --git a/arch/s390/lib/probes.c b/arch/s390/lib/probes.c new file mode 100644 index 000000000..ae90e1ae3 --- /dev/null +++ b/arch/s390/lib/probes.c @@ -0,0 +1,159 @@ +/* + * Common helper functions for kprobes and uprobes + * + * Copyright IBM Corp. 2014 + */ + +#include <asm/kprobes.h> +#include <asm/dis.h> + +int probe_is_prohibited_opcode(u16 *insn) +{ + if (!is_known_insn((unsigned char *)insn)) + return -EINVAL; + switch (insn[0] >> 8) { + case 0x0c: /* bassm */ + case 0x0b: /* bsm */ + case 0x83: /* diag */ + case 0x44: /* ex */ + case 0xac: /* stnsm */ + case 0xad: /* stosm */ + return -EINVAL; + case 0xc6: + switch (insn[0] & 0x0f) { + case 0x00: /* exrl */ + return -EINVAL; + } + } + switch (insn[0]) { + case 0x0101: /* pr */ + case 0xb25a: /* bsa */ + case 0xb240: /* bakr */ + case 0xb258: /* bsg */ + case 0xb218: /* pc */ + case 0xb228: /* pt */ + case 0xb98d: /* epsw */ + case 0xe560: /* tbegin */ + case 0xe561: /* tbeginc */ + case 0xb2f8: /* tend */ + return -EINVAL; + } + return 0; +} + +int probe_get_fixup_type(u16 *insn) +{ + /* default fixup method */ + int fixup = FIXUP_PSW_NORMAL; + + switch (insn[0] >> 8) { + case 0x05: /* balr */ + case 0x0d: /* basr */ + fixup = FIXUP_RETURN_REGISTER; + /* if r2 = 0, no branch will be taken */ + if ((insn[0] & 0x0f) == 0) + fixup |= FIXUP_BRANCH_NOT_TAKEN; + break; + case 0x06: /* bctr */ + case 0x07: /* bcr */ + fixup = FIXUP_BRANCH_NOT_TAKEN; + break; + case 0x45: /* bal */ + case 0x4d: /* bas */ + fixup = FIXUP_RETURN_REGISTER; + break; + case 0x47: /* bc */ + case 0x46: /* bct */ + case 0x86: /* bxh */ + case 0x87: /* bxle */ + fixup = FIXUP_BRANCH_NOT_TAKEN; + break; + case 0x82: /* lpsw */ + fixup = FIXUP_NOT_REQUIRED; + break; + case 0xb2: /* lpswe */ + if ((insn[0] & 0xff) == 0xb2) + fixup = FIXUP_NOT_REQUIRED; + break; + case 0xa7: /* bras */ + if ((insn[0] & 0x0f) == 0x05) + fixup |= FIXUP_RETURN_REGISTER; + break; + case 0xc0: + if ((insn[0] & 0x0f) == 0x05) /* brasl */ + fixup |= FIXUP_RETURN_REGISTER; + break; + case 0xeb: + switch (insn[2] & 0xff) { + case 0x44: /* bxhg */ + case 0x45: /* bxleg */ + fixup = FIXUP_BRANCH_NOT_TAKEN; + break; + } + break; + case 0xe3: /* bctg */ + if ((insn[2] & 0xff) == 0x46) + fixup = FIXUP_BRANCH_NOT_TAKEN; + break; + case 0xec: + switch (insn[2] & 0xff) { + case 0xe5: /* clgrb */ + case 0xe6: /* cgrb */ + case 0xf6: /* crb */ + case 0xf7: /* clrb */ + case 0xfc: /* cgib */ + case 0xfd: /* cglib */ + case 0xfe: /* cib */ + case 0xff: /* clib */ + fixup = FIXUP_BRANCH_NOT_TAKEN; + break; + } + break; + } + return fixup; +} + +int probe_is_insn_relative_long(u16 *insn) +{ + /* Check if we have a RIL-b or RIL-c format instruction which + * we need to modify in order to avoid instruction emulation. */ + switch (insn[0] >> 8) { + case 0xc0: + if ((insn[0] & 0x0f) == 0x00) /* larl */ + return true; + break; + case 0xc4: + switch (insn[0] & 0x0f) { + case 0x02: /* llhrl */ + case 0x04: /* lghrl */ + case 0x05: /* lhrl */ + case 0x06: /* llghrl */ + case 0x07: /* sthrl */ + case 0x08: /* lgrl */ + case 0x0b: /* stgrl */ + case 0x0c: /* lgfrl */ + case 0x0d: /* lrl */ + case 0x0e: /* llgfrl */ + case 0x0f: /* strl */ + return true; + } + break; + case 0xc6: + switch (insn[0] & 0x0f) { + case 0x02: /* pfdrl */ + case 0x04: /* cghrl */ + case 0x05: /* chrl */ + case 0x06: /* clghrl */ + case 0x07: /* clhrl */ + case 0x08: /* cgrl */ + case 0x0a: /* clgrl */ + case 0x0c: /* cgfrl */ + case 0x0d: /* crl */ + case 0x0e: /* clgfrl */ + case 0x0f: /* clrl */ + return true; + } + break; + } + return false; +} diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c new file mode 100644 index 000000000..d6c9991f7 --- /dev/null +++ b/arch/s390/lib/spinlock.c @@ -0,0 +1,272 @@ +/* + * Out of line spinlock code. + * + * Copyright IBM Corp. 2004, 2006 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/spinlock.h> +#include <linux/init.h> +#include <linux/smp.h> +#include <asm/io.h> + +int spin_retry = -1; + +static int __init spin_retry_init(void) +{ + if (spin_retry < 0) + spin_retry = MACHINE_HAS_CAD ? 10 : 1000; + return 0; +} +early_initcall(spin_retry_init); + +/** + * spin_retry= parameter + */ +static int __init spin_retry_setup(char *str) +{ + spin_retry = simple_strtoul(str, &str, 0); + return 1; +} +__setup("spin_retry=", spin_retry_setup); + +static inline void _raw_compare_and_delay(unsigned int *lock, unsigned int old) +{ + asm(".insn rsy,0xeb0000000022,%0,0,%1" : : "d" (old), "Q" (*lock)); +} + +void arch_spin_lock_wait(arch_spinlock_t *lp) +{ + unsigned int cpu = SPINLOCK_LOCKVAL; + unsigned int owner; + int count; + + while (1) { + owner = ACCESS_ONCE(lp->lock); + /* Try to get the lock if it is free. */ + if (!owner) { + if (_raw_compare_and_swap(&lp->lock, 0, cpu)) + return; + continue; + } + /* Check if the lock owner is running. */ + if (!smp_vcpu_scheduled(~owner)) { + smp_yield_cpu(~owner); + continue; + } + /* Loop for a while on the lock value. */ + count = spin_retry; + do { + if (MACHINE_HAS_CAD) + _raw_compare_and_delay(&lp->lock, owner); + owner = ACCESS_ONCE(lp->lock); + } while (owner && count-- > 0); + if (!owner) + continue; + /* + * For multiple layers of hypervisors, e.g. z/VM + LPAR + * yield the CPU if the lock is still unavailable. + */ + if (!MACHINE_IS_LPAR) + smp_yield_cpu(~owner); + } +} +EXPORT_SYMBOL(arch_spin_lock_wait); + +void arch_spin_lock_wait_flags(arch_spinlock_t *lp, unsigned long flags) +{ + unsigned int cpu = SPINLOCK_LOCKVAL; + unsigned int owner; + int count; + + local_irq_restore(flags); + while (1) { + owner = ACCESS_ONCE(lp->lock); + /* Try to get the lock if it is free. */ + if (!owner) { + local_irq_disable(); + if (_raw_compare_and_swap(&lp->lock, 0, cpu)) + return; + local_irq_restore(flags); + } + /* Check if the lock owner is running. */ + if (!smp_vcpu_scheduled(~owner)) { + smp_yield_cpu(~owner); + continue; + } + /* Loop for a while on the lock value. */ + count = spin_retry; + do { + if (MACHINE_HAS_CAD) + _raw_compare_and_delay(&lp->lock, owner); + owner = ACCESS_ONCE(lp->lock); + } while (owner && count-- > 0); + if (!owner) + continue; + /* + * For multiple layers of hypervisors, e.g. z/VM + LPAR + * yield the CPU if the lock is still unavailable. + */ + if (!MACHINE_IS_LPAR) + smp_yield_cpu(~owner); + } +} +EXPORT_SYMBOL(arch_spin_lock_wait_flags); + +int arch_spin_trylock_retry(arch_spinlock_t *lp) +{ + unsigned int cpu = SPINLOCK_LOCKVAL; + unsigned int owner; + int count; + + for (count = spin_retry; count > 0; count--) { + owner = ACCESS_ONCE(lp->lock); + /* Try to get the lock if it is free. */ + if (!owner) { + if (_raw_compare_and_swap(&lp->lock, 0, cpu)) + return 1; + } else if (MACHINE_HAS_CAD) + _raw_compare_and_delay(&lp->lock, owner); + } + return 0; +} +EXPORT_SYMBOL(arch_spin_trylock_retry); + +void _raw_read_lock_wait(arch_rwlock_t *rw) +{ + unsigned int owner, old; + int count = spin_retry; + +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES + __RAW_LOCK(&rw->lock, -1, __RAW_OP_ADD); +#endif + owner = 0; + while (1) { + if (count-- <= 0) { + if (owner && !smp_vcpu_scheduled(~owner)) + smp_yield_cpu(~owner); + count = spin_retry; + } + old = ACCESS_ONCE(rw->lock); + owner = ACCESS_ONCE(rw->owner); + if ((int) old < 0) { + if (MACHINE_HAS_CAD) + _raw_compare_and_delay(&rw->lock, old); + continue; + } + if (_raw_compare_and_swap(&rw->lock, old, old + 1)) + return; + } +} +EXPORT_SYMBOL(_raw_read_lock_wait); + +int _raw_read_trylock_retry(arch_rwlock_t *rw) +{ + unsigned int old; + int count = spin_retry; + + while (count-- > 0) { + old = ACCESS_ONCE(rw->lock); + if ((int) old < 0) { + if (MACHINE_HAS_CAD) + _raw_compare_and_delay(&rw->lock, old); + continue; + } + if (_raw_compare_and_swap(&rw->lock, old, old + 1)) + return 1; + } + return 0; +} +EXPORT_SYMBOL(_raw_read_trylock_retry); + +#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES + +void _raw_write_lock_wait(arch_rwlock_t *rw, unsigned int prev) +{ + unsigned int owner, old; + int count = spin_retry; + + owner = 0; + while (1) { + if (count-- <= 0) { + if (owner && !smp_vcpu_scheduled(~owner)) + smp_yield_cpu(~owner); + count = spin_retry; + } + old = ACCESS_ONCE(rw->lock); + owner = ACCESS_ONCE(rw->owner); + smp_rmb(); + if ((int) old >= 0) { + prev = __RAW_LOCK(&rw->lock, 0x80000000, __RAW_OP_OR); + old = prev; + } + if ((old & 0x7fffffff) == 0 && (int) prev >= 0) + break; + if (MACHINE_HAS_CAD) + _raw_compare_and_delay(&rw->lock, old); + } +} +EXPORT_SYMBOL(_raw_write_lock_wait); + +#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +void _raw_write_lock_wait(arch_rwlock_t *rw) +{ + unsigned int owner, old, prev; + int count = spin_retry; + + prev = 0x80000000; + owner = 0; + while (1) { + if (count-- <= 0) { + if (owner && !smp_vcpu_scheduled(~owner)) + smp_yield_cpu(~owner); + count = spin_retry; + } + old = ACCESS_ONCE(rw->lock); + owner = ACCESS_ONCE(rw->owner); + if ((int) old >= 0 && + _raw_compare_and_swap(&rw->lock, old, old | 0x80000000)) + prev = old; + else + smp_rmb(); + if ((old & 0x7fffffff) == 0 && (int) prev >= 0) + break; + if (MACHINE_HAS_CAD) + _raw_compare_and_delay(&rw->lock, old); + } +} +EXPORT_SYMBOL(_raw_write_lock_wait); + +#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */ + +int _raw_write_trylock_retry(arch_rwlock_t *rw) +{ + unsigned int old; + int count = spin_retry; + + while (count-- > 0) { + old = ACCESS_ONCE(rw->lock); + if (old) { + if (MACHINE_HAS_CAD) + _raw_compare_and_delay(&rw->lock, old); + continue; + } + if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000)) + return 1; + } + return 0; +} +EXPORT_SYMBOL(_raw_write_trylock_retry); + +void arch_lock_relax(unsigned int cpu) +{ + if (!cpu) + return; + if (MACHINE_IS_LPAR && smp_vcpu_scheduled(~cpu)) + return; + smp_yield_cpu(~cpu); +} +EXPORT_SYMBOL(arch_lock_relax); diff --git a/arch/s390/lib/string.c b/arch/s390/lib/string.c new file mode 100644 index 000000000..b647d5ff0 --- /dev/null +++ b/arch/s390/lib/string.c @@ -0,0 +1,342 @@ +/* + * Optimized string functions + * + * S390 version + * Copyright IBM Corp. 2004 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) + */ + +#define IN_ARCH_STRING_C 1 + +#include <linux/types.h> +#include <linux/module.h> + +/* + * Helper functions to find the end of a string + */ +static inline char *__strend(const char *s) +{ + register unsigned long r0 asm("0") = 0; + + asm volatile ("0: srst %0,%1\n" + " jo 0b" + : "+d" (r0), "+a" (s) : : "cc" ); + return (char *) r0; +} + +static inline char *__strnend(const char *s, size_t n) +{ + register unsigned long r0 asm("0") = 0; + const char *p = s + n; + + asm volatile ("0: srst %0,%1\n" + " jo 0b" + : "+d" (p), "+a" (s) : "d" (r0) : "cc" ); + return (char *) p; +} + +/** + * strlen - Find the length of a string + * @s: The string to be sized + * + * returns the length of @s + */ +size_t strlen(const char *s) +{ + return __strend(s) - s; +} +EXPORT_SYMBOL(strlen); + +/** + * strnlen - Find the length of a length-limited string + * @s: The string to be sized + * @n: The maximum number of bytes to search + * + * returns the minimum of the length of @s and @n + */ +size_t strnlen(const char * s, size_t n) +{ + return __strnend(s, n) - s; +} +EXPORT_SYMBOL(strnlen); + +/** + * strcpy - Copy a %NUL terminated string + * @dest: Where to copy the string to + * @src: Where to copy the string from + * + * returns a pointer to @dest + */ +char *strcpy(char *dest, const char *src) +{ + register int r0 asm("0") = 0; + char *ret = dest; + + asm volatile ("0: mvst %0,%1\n" + " jo 0b" + : "+&a" (dest), "+&a" (src) : "d" (r0) + : "cc", "memory" ); + return ret; +} +EXPORT_SYMBOL(strcpy); + +/** + * strlcpy - Copy a %NUL terminated string into a sized buffer + * @dest: Where to copy the string to + * @src: Where to copy the string from + * @size: size of destination buffer + * + * Compatible with *BSD: the result is always a valid + * NUL-terminated string that fits in the buffer (unless, + * of course, the buffer size is zero). It does not pad + * out the result like strncpy() does. + */ +size_t strlcpy(char *dest, const char *src, size_t size) +{ + size_t ret = __strend(src) - src; + + if (size) { + size_t len = (ret >= size) ? size-1 : ret; + dest[len] = '\0'; + memcpy(dest, src, len); + } + return ret; +} +EXPORT_SYMBOL(strlcpy); + +/** + * strncpy - Copy a length-limited, %NUL-terminated string + * @dest: Where to copy the string to + * @src: Where to copy the string from + * @n: The maximum number of bytes to copy + * + * The result is not %NUL-terminated if the source exceeds + * @n bytes. + */ +char *strncpy(char *dest, const char *src, size_t n) +{ + size_t len = __strnend(src, n) - src; + memset(dest + len, 0, n - len); + memcpy(dest, src, len); + return dest; +} +EXPORT_SYMBOL(strncpy); + +/** + * strcat - Append one %NUL-terminated string to another + * @dest: The string to be appended to + * @src: The string to append to it + * + * returns a pointer to @dest + */ +char *strcat(char *dest, const char *src) +{ + register int r0 asm("0") = 0; + unsigned long dummy; + char *ret = dest; + + asm volatile ("0: srst %0,%1\n" + " jo 0b\n" + "1: mvst %0,%2\n" + " jo 1b" + : "=&a" (dummy), "+a" (dest), "+a" (src) + : "d" (r0), "0" (0UL) : "cc", "memory" ); + return ret; +} +EXPORT_SYMBOL(strcat); + +/** + * strlcat - Append a length-limited, %NUL-terminated string to another + * @dest: The string to be appended to + * @src: The string to append to it + * @n: The size of the destination buffer. + */ +size_t strlcat(char *dest, const char *src, size_t n) +{ + size_t dsize = __strend(dest) - dest; + size_t len = __strend(src) - src; + size_t res = dsize + len; + + if (dsize < n) { + dest += dsize; + n -= dsize; + if (len >= n) + len = n - 1; + dest[len] = '\0'; + memcpy(dest, src, len); + } + return res; +} +EXPORT_SYMBOL(strlcat); + +/** + * strncat - Append a length-limited, %NUL-terminated string to another + * @dest: The string to be appended to + * @src: The string to append to it + * @n: The maximum numbers of bytes to copy + * + * returns a pointer to @dest + * + * Note that in contrast to strncpy, strncat ensures the result is + * terminated. + */ +char *strncat(char *dest, const char *src, size_t n) +{ + size_t len = __strnend(src, n) - src; + char *p = __strend(dest); + + p[len] = '\0'; + memcpy(p, src, len); + return dest; +} +EXPORT_SYMBOL(strncat); + +/** + * strcmp - Compare two strings + * @cs: One string + * @ct: Another string + * + * returns 0 if @cs and @ct are equal, + * < 0 if @cs is less than @ct + * > 0 if @cs is greater than @ct + */ +int strcmp(const char *cs, const char *ct) +{ + register int r0 asm("0") = 0; + int ret = 0; + + asm volatile ("0: clst %2,%3\n" + " jo 0b\n" + " je 1f\n" + " ic %0,0(%2)\n" + " ic %1,0(%3)\n" + " sr %0,%1\n" + "1:" + : "+d" (ret), "+d" (r0), "+a" (cs), "+a" (ct) + : : "cc" ); + return ret; +} +EXPORT_SYMBOL(strcmp); + +/** + * strrchr - Find the last occurrence of a character in a string + * @s: The string to be searched + * @c: The character to search for + */ +char * strrchr(const char * s, int c) +{ + size_t len = __strend(s) - s; + + if (len) + do { + if (s[len] == (char) c) + return (char *) s + len; + } while (--len > 0); + return NULL; +} +EXPORT_SYMBOL(strrchr); + +/** + * strstr - Find the first substring in a %NUL terminated string + * @s1: The string to be searched + * @s2: The string to search for + */ +char * strstr(const char * s1,const char * s2) +{ + int l1, l2; + + l2 = __strend(s2) - s2; + if (!l2) + return (char *) s1; + l1 = __strend(s1) - s1; + while (l1-- >= l2) { + register unsigned long r2 asm("2") = (unsigned long) s1; + register unsigned long r3 asm("3") = (unsigned long) l2; + register unsigned long r4 asm("4") = (unsigned long) s2; + register unsigned long r5 asm("5") = (unsigned long) l2; + int cc; + + asm volatile ("0: clcle %1,%3,0\n" + " jo 0b\n" + " ipm %0\n" + " srl %0,28" + : "=&d" (cc), "+a" (r2), "+a" (r3), + "+a" (r4), "+a" (r5) : : "cc" ); + if (!cc) + return (char *) s1; + s1++; + } + return NULL; +} +EXPORT_SYMBOL(strstr); + +/** + * memchr - Find a character in an area of memory. + * @s: The memory area + * @c: The byte to search for + * @n: The size of the area. + * + * returns the address of the first occurrence of @c, or %NULL + * if @c is not found + */ +void *memchr(const void *s, int c, size_t n) +{ + register int r0 asm("0") = (char) c; + const void *ret = s + n; + + asm volatile ("0: srst %0,%1\n" + " jo 0b\n" + " jl 1f\n" + " la %0,0\n" + "1:" + : "+a" (ret), "+&a" (s) : "d" (r0) : "cc" ); + return (void *) ret; +} +EXPORT_SYMBOL(memchr); + +/** + * memcmp - Compare two areas of memory + * @cs: One area of memory + * @ct: Another area of memory + * @count: The size of the area. + */ +int memcmp(const void *cs, const void *ct, size_t n) +{ + register unsigned long r2 asm("2") = (unsigned long) cs; + register unsigned long r3 asm("3") = (unsigned long) n; + register unsigned long r4 asm("4") = (unsigned long) ct; + register unsigned long r5 asm("5") = (unsigned long) n; + int ret; + + asm volatile ("0: clcle %1,%3,0\n" + " jo 0b\n" + " ipm %0\n" + " srl %0,28" + : "=&d" (ret), "+a" (r2), "+a" (r3), "+a" (r4), "+a" (r5) + : : "cc" ); + if (ret) + ret = *(char *) r2 - *(char *) r4; + return ret; +} +EXPORT_SYMBOL(memcmp); + +/** + * memscan - Find a character in an area of memory. + * @s: The memory area + * @c: The byte to search for + * @n: The size of the area. + * + * returns the address of the first occurrence of @c, or 1 byte past + * the area if @c is not found + */ +void *memscan(void *s, int c, size_t n) +{ + register int r0 asm("0") = (char) c; + const void *ret = s + n; + + asm volatile ("0: srst %0,%1\n" + " jo 0b\n" + : "+a" (ret), "+&a" (s) : "d" (r0) : "cc" ); + return (void *) ret; +} +EXPORT_SYMBOL(memscan); diff --git a/arch/s390/lib/uaccess.c b/arch/s390/lib/uaccess.c new file mode 100644 index 000000000..4614d415b --- /dev/null +++ b/arch/s390/lib/uaccess.c @@ -0,0 +1,392 @@ +/* + * Standard user space access functions based on mvcp/mvcs and doing + * interesting things in the secondary space mode. + * + * Copyright IBM Corp. 2006,2014 + * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com), + * Gerald Schaefer (gerald.schaefer@de.ibm.com) + */ + +#include <linux/jump_label.h> +#include <linux/uaccess.h> +#include <linux/export.h> +#include <linux/errno.h> +#include <linux/mm.h> +#include <asm/mmu_context.h> +#include <asm/facility.h> + +static struct static_key have_mvcos = STATIC_KEY_INIT_FALSE; + +static inline unsigned long copy_from_user_mvcos(void *x, const void __user *ptr, + unsigned long size) +{ + register unsigned long reg0 asm("0") = 0x81UL; + unsigned long tmp1, tmp2; + + tmp1 = -4096UL; + asm volatile( + "0: .insn ss,0xc80000000000,0(%0,%2),0(%1),0\n" + "9: jz 7f\n" + "1: algr %0,%3\n" + " slgr %1,%3\n" + " slgr %2,%3\n" + " j 0b\n" + "2: la %4,4095(%1)\n"/* %4 = ptr + 4095 */ + " nr %4,%3\n" /* %4 = (ptr + 4095) & -4096 */ + " slgr %4,%1\n" + " clgr %0,%4\n" /* copy crosses next page boundary? */ + " jnh 4f\n" + "3: .insn ss,0xc80000000000,0(%4,%2),0(%1),0\n" + "10:slgr %0,%4\n" + " algr %2,%4\n" + "4: lghi %4,-1\n" + " algr %4,%0\n" /* copy remaining size, subtract 1 */ + " bras %3,6f\n" /* memset loop */ + " xc 0(1,%2),0(%2)\n" + "5: xc 0(256,%2),0(%2)\n" + " la %2,256(%2)\n" + "6: aghi %4,-256\n" + " jnm 5b\n" + " ex %4,0(%3)\n" + " j 8f\n" + "7:slgr %0,%0\n" + "8:\n" + EX_TABLE(0b,2b) EX_TABLE(3b,4b) EX_TABLE(9b,2b) EX_TABLE(10b,4b) + : "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2) + : "d" (reg0) : "cc", "memory"); + return size; +} + +static inline unsigned long copy_from_user_mvcp(void *x, const void __user *ptr, + unsigned long size) +{ + unsigned long tmp1, tmp2; + + load_kernel_asce(); + tmp1 = -256UL; + asm volatile( + " sacf 0\n" + "0: mvcp 0(%0,%2),0(%1),%3\n" + "10:jz 8f\n" + "1: algr %0,%3\n" + " la %1,256(%1)\n" + " la %2,256(%2)\n" + "2: mvcp 0(%0,%2),0(%1),%3\n" + "11:jnz 1b\n" + " j 8f\n" + "3: la %4,255(%1)\n" /* %4 = ptr + 255 */ + " lghi %3,-4096\n" + " nr %4,%3\n" /* %4 = (ptr + 255) & -4096 */ + " slgr %4,%1\n" + " clgr %0,%4\n" /* copy crosses next page boundary? */ + " jnh 5f\n" + "4: mvcp 0(%4,%2),0(%1),%3\n" + "12:slgr %0,%4\n" + " algr %2,%4\n" + "5: lghi %4,-1\n" + " algr %4,%0\n" /* copy remaining size, subtract 1 */ + " bras %3,7f\n" /* memset loop */ + " xc 0(1,%2),0(%2)\n" + "6: xc 0(256,%2),0(%2)\n" + " la %2,256(%2)\n" + "7: aghi %4,-256\n" + " jnm 6b\n" + " ex %4,0(%3)\n" + " j 9f\n" + "8:slgr %0,%0\n" + "9: sacf 768\n" + EX_TABLE(0b,3b) EX_TABLE(2b,3b) EX_TABLE(4b,5b) + EX_TABLE(10b,3b) EX_TABLE(11b,3b) EX_TABLE(12b,5b) + : "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2) + : : "cc", "memory"); + return size; +} + +unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n) +{ + if (static_key_false(&have_mvcos)) + return copy_from_user_mvcos(to, from, n); + return copy_from_user_mvcp(to, from, n); +} +EXPORT_SYMBOL(__copy_from_user); + +static inline unsigned long copy_to_user_mvcos(void __user *ptr, const void *x, + unsigned long size) +{ + register unsigned long reg0 asm("0") = 0x810000UL; + unsigned long tmp1, tmp2; + + tmp1 = -4096UL; + asm volatile( + "0: .insn ss,0xc80000000000,0(%0,%1),0(%2),0\n" + "6: jz 4f\n" + "1: algr %0,%3\n" + " slgr %1,%3\n" + " slgr %2,%3\n" + " j 0b\n" + "2: la %4,4095(%1)\n"/* %4 = ptr + 4095 */ + " nr %4,%3\n" /* %4 = (ptr + 4095) & -4096 */ + " slgr %4,%1\n" + " clgr %0,%4\n" /* copy crosses next page boundary? */ + " jnh 5f\n" + "3: .insn ss,0xc80000000000,0(%4,%1),0(%2),0\n" + "7: slgr %0,%4\n" + " j 5f\n" + "4: slgr %0,%0\n" + "5:\n" + EX_TABLE(0b,2b) EX_TABLE(3b,5b) EX_TABLE(6b,2b) EX_TABLE(7b,5b) + : "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2) + : "d" (reg0) : "cc", "memory"); + return size; +} + +static inline unsigned long copy_to_user_mvcs(void __user *ptr, const void *x, + unsigned long size) +{ + unsigned long tmp1, tmp2; + + load_kernel_asce(); + tmp1 = -256UL; + asm volatile( + " sacf 0\n" + "0: mvcs 0(%0,%1),0(%2),%3\n" + "7: jz 5f\n" + "1: algr %0,%3\n" + " la %1,256(%1)\n" + " la %2,256(%2)\n" + "2: mvcs 0(%0,%1),0(%2),%3\n" + "8: jnz 1b\n" + " j 5f\n" + "3: la %4,255(%1)\n" /* %4 = ptr + 255 */ + " lghi %3,-4096\n" + " nr %4,%3\n" /* %4 = (ptr + 255) & -4096 */ + " slgr %4,%1\n" + " clgr %0,%4\n" /* copy crosses next page boundary? */ + " jnh 6f\n" + "4: mvcs 0(%4,%1),0(%2),%3\n" + "9: slgr %0,%4\n" + " j 6f\n" + "5: slgr %0,%0\n" + "6: sacf 768\n" + EX_TABLE(0b,3b) EX_TABLE(2b,3b) EX_TABLE(4b,6b) + EX_TABLE(7b,3b) EX_TABLE(8b,3b) EX_TABLE(9b,6b) + : "+a" (size), "+a" (ptr), "+a" (x), "+a" (tmp1), "=a" (tmp2) + : : "cc", "memory"); + return size; +} + +unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n) +{ + if (static_key_false(&have_mvcos)) + return copy_to_user_mvcos(to, from, n); + return copy_to_user_mvcs(to, from, n); +} +EXPORT_SYMBOL(__copy_to_user); + +static inline unsigned long copy_in_user_mvcos(void __user *to, const void __user *from, + unsigned long size) +{ + register unsigned long reg0 asm("0") = 0x810081UL; + unsigned long tmp1, tmp2; + + tmp1 = -4096UL; + /* FIXME: copy with reduced length. */ + asm volatile( + "0: .insn ss,0xc80000000000,0(%0,%1),0(%2),0\n" + " jz 2f\n" + "1: algr %0,%3\n" + " slgr %1,%3\n" + " slgr %2,%3\n" + " j 0b\n" + "2:slgr %0,%0\n" + "3: \n" + EX_TABLE(0b,3b) + : "+a" (size), "+a" (to), "+a" (from), "+a" (tmp1), "=a" (tmp2) + : "d" (reg0) : "cc", "memory"); + return size; +} + +static inline unsigned long copy_in_user_mvc(void __user *to, const void __user *from, + unsigned long size) +{ + unsigned long tmp1; + + load_kernel_asce(); + asm volatile( + " sacf 256\n" + " aghi %0,-1\n" + " jo 5f\n" + " bras %3,3f\n" + "0: aghi %0,257\n" + "1: mvc 0(1,%1),0(%2)\n" + " la %1,1(%1)\n" + " la %2,1(%2)\n" + " aghi %0,-1\n" + " jnz 1b\n" + " j 5f\n" + "2: mvc 0(256,%1),0(%2)\n" + " la %1,256(%1)\n" + " la %2,256(%2)\n" + "3: aghi %0,-256\n" + " jnm 2b\n" + "4: ex %0,1b-0b(%3)\n" + "5: slgr %0,%0\n" + "6: sacf 768\n" + EX_TABLE(1b,6b) EX_TABLE(2b,0b) EX_TABLE(4b,0b) + : "+a" (size), "+a" (to), "+a" (from), "=a" (tmp1) + : : "cc", "memory"); + return size; +} + +unsigned long __copy_in_user(void __user *to, const void __user *from, unsigned long n) +{ + if (static_key_false(&have_mvcos)) + return copy_in_user_mvcos(to, from, n); + return copy_in_user_mvc(to, from, n); +} +EXPORT_SYMBOL(__copy_in_user); + +static inline unsigned long clear_user_mvcos(void __user *to, unsigned long size) +{ + register unsigned long reg0 asm("0") = 0x810000UL; + unsigned long tmp1, tmp2; + + tmp1 = -4096UL; + asm volatile( + "0: .insn ss,0xc80000000000,0(%0,%1),0(%4),0\n" + " jz 4f\n" + "1: algr %0,%2\n" + " slgr %1,%2\n" + " j 0b\n" + "2: la %3,4095(%1)\n"/* %4 = to + 4095 */ + " nr %3,%2\n" /* %4 = (to + 4095) & -4096 */ + " slgr %3,%1\n" + " clgr %0,%3\n" /* copy crosses next page boundary? */ + " jnh 5f\n" + "3: .insn ss,0xc80000000000,0(%3,%1),0(%4),0\n" + " slgr %0,%3\n" + " j 5f\n" + "4:slgr %0,%0\n" + "5:\n" + EX_TABLE(0b,2b) EX_TABLE(3b,5b) + : "+a" (size), "+a" (to), "+a" (tmp1), "=a" (tmp2) + : "a" (empty_zero_page), "d" (reg0) : "cc", "memory"); + return size; +} + +static inline unsigned long clear_user_xc(void __user *to, unsigned long size) +{ + unsigned long tmp1, tmp2; + + load_kernel_asce(); + asm volatile( + " sacf 256\n" + " aghi %0,-1\n" + " jo 5f\n" + " bras %3,3f\n" + " xc 0(1,%1),0(%1)\n" + "0: aghi %0,257\n" + " la %2,255(%1)\n" /* %2 = ptr + 255 */ + " srl %2,12\n" + " sll %2,12\n" /* %2 = (ptr + 255) & -4096 */ + " slgr %2,%1\n" + " clgr %0,%2\n" /* clear crosses next page boundary? */ + " jnh 5f\n" + " aghi %2,-1\n" + "1: ex %2,0(%3)\n" + " aghi %2,1\n" + " slgr %0,%2\n" + " j 5f\n" + "2: xc 0(256,%1),0(%1)\n" + " la %1,256(%1)\n" + "3: aghi %0,-256\n" + " jnm 2b\n" + "4: ex %0,0(%3)\n" + "5: slgr %0,%0\n" + "6: sacf 768\n" + EX_TABLE(1b,6b) EX_TABLE(2b,0b) EX_TABLE(4b,0b) + : "+a" (size), "+a" (to), "=a" (tmp1), "=a" (tmp2) + : : "cc", "memory"); + return size; +} + +unsigned long __clear_user(void __user *to, unsigned long size) +{ + if (static_key_false(&have_mvcos)) + return clear_user_mvcos(to, size); + return clear_user_xc(to, size); +} +EXPORT_SYMBOL(__clear_user); + +static inline unsigned long strnlen_user_srst(const char __user *src, + unsigned long size) +{ + register unsigned long reg0 asm("0") = 0; + unsigned long tmp1, tmp2; + + asm volatile( + " la %2,0(%1)\n" + " la %3,0(%0,%1)\n" + " slgr %0,%0\n" + " sacf 256\n" + "0: srst %3,%2\n" + " jo 0b\n" + " la %0,1(%3)\n" /* strnlen_user results includes \0 */ + " slgr %0,%1\n" + "1: sacf 768\n" + EX_TABLE(0b,1b) + : "+a" (size), "+a" (src), "=a" (tmp1), "=a" (tmp2) + : "d" (reg0) : "cc", "memory"); + return size; +} + +unsigned long __strnlen_user(const char __user *src, unsigned long size) +{ + if (unlikely(!size)) + return 0; + load_kernel_asce(); + return strnlen_user_srst(src, size); +} +EXPORT_SYMBOL(__strnlen_user); + +long __strncpy_from_user(char *dst, const char __user *src, long size) +{ + size_t done, len, offset, len_str; + + if (unlikely(size <= 0)) + return 0; + done = 0; + do { + offset = (size_t)src & ~PAGE_MASK; + len = min(size - done, PAGE_SIZE - offset); + if (copy_from_user(dst, src, len)) + return -EFAULT; + len_str = strnlen(dst, len); + done += len_str; + src += len_str; + dst += len_str; + } while ((len_str == len) && (done < size)); + return done; +} +EXPORT_SYMBOL(__strncpy_from_user); + +/* + * The "old" uaccess variant without mvcos can be enforced with the + * uaccess_primary kernel parameter. This is mainly for debugging purposes. + */ +static int uaccess_primary __initdata; + +static int __init parse_uaccess_pt(char *__unused) +{ + uaccess_primary = 1; + return 0; +} +early_param("uaccess_primary", parse_uaccess_pt); + +static int __init uaccess_init(void) +{ + if (!uaccess_primary && test_facility(27)) + static_key_slow_inc(&have_mvcos); + return 0; +} +early_initcall(uaccess_init); diff --git a/arch/s390/mm/Makefile b/arch/s390/mm/Makefile new file mode 100644 index 000000000..839592ca2 --- /dev/null +++ b/arch/s390/mm/Makefile @@ -0,0 +1,10 @@ +# +# Makefile for the linux s390-specific parts of the memory manager. +# + +obj-y := init.o fault.o extmem.o mmap.o vmem.o pgtable.o maccess.o +obj-y += page-states.o gup.o extable.o pageattr.o mem_detect.o + +obj-$(CONFIG_CMM) += cmm.o +obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o +obj-$(CONFIG_S390_PTDUMP) += dump_pagetables.o diff --git a/arch/s390/mm/cmm.c b/arch/s390/mm/cmm.c new file mode 100644 index 000000000..79ddd580d --- /dev/null +++ b/arch/s390/mm/cmm.c @@ -0,0 +1,495 @@ +/* + * Collaborative memory management interface. + * + * Copyright IBM Corp 2003, 2010 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>, + * + */ + +#include <linux/errno.h> +#include <linux/fs.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/gfp.h> +#include <linux/sched.h> +#include <linux/sysctl.h> +#include <linux/ctype.h> +#include <linux/swap.h> +#include <linux/kthread.h> +#include <linux/oom.h> +#include <linux/suspend.h> +#include <linux/uaccess.h> + +#include <asm/pgalloc.h> +#include <asm/diag.h> + +#ifdef CONFIG_CMM_IUCV +static char *cmm_default_sender = "VMRMSVM"; +#endif +static char *sender; +module_param(sender, charp, 0400); +MODULE_PARM_DESC(sender, + "Guest name that may send SMSG messages (default VMRMSVM)"); + +#include "../../../drivers/s390/net/smsgiucv.h" + +#define CMM_NR_PAGES ((PAGE_SIZE / sizeof(unsigned long)) - 2) + +struct cmm_page_array { + struct cmm_page_array *next; + unsigned long index; + unsigned long pages[CMM_NR_PAGES]; +}; + +static long cmm_pages; +static long cmm_timed_pages; +static volatile long cmm_pages_target; +static volatile long cmm_timed_pages_target; +static long cmm_timeout_pages; +static long cmm_timeout_seconds; +static int cmm_suspended; + +static struct cmm_page_array *cmm_page_list; +static struct cmm_page_array *cmm_timed_page_list; +static DEFINE_SPINLOCK(cmm_lock); + +static struct task_struct *cmm_thread_ptr; +static DECLARE_WAIT_QUEUE_HEAD(cmm_thread_wait); +static DEFINE_TIMER(cmm_timer, NULL, 0, 0); + +static void cmm_timer_fn(unsigned long); +static void cmm_set_timer(void); + +static long cmm_alloc_pages(long nr, long *counter, + struct cmm_page_array **list) +{ + struct cmm_page_array *pa, *npa; + unsigned long addr; + + while (nr) { + addr = __get_free_page(GFP_NOIO); + if (!addr) + break; + spin_lock(&cmm_lock); + pa = *list; + if (!pa || pa->index >= CMM_NR_PAGES) { + /* Need a new page for the page list. */ + spin_unlock(&cmm_lock); + npa = (struct cmm_page_array *) + __get_free_page(GFP_NOIO); + if (!npa) { + free_page(addr); + break; + } + spin_lock(&cmm_lock); + pa = *list; + if (!pa || pa->index >= CMM_NR_PAGES) { + npa->next = pa; + npa->index = 0; + pa = npa; + *list = pa; + } else + free_page((unsigned long) npa); + } + diag10_range(addr >> PAGE_SHIFT, 1); + pa->pages[pa->index++] = addr; + (*counter)++; + spin_unlock(&cmm_lock); + nr--; + } + return nr; +} + +static long cmm_free_pages(long nr, long *counter, struct cmm_page_array **list) +{ + struct cmm_page_array *pa; + unsigned long addr; + + spin_lock(&cmm_lock); + pa = *list; + while (nr) { + if (!pa || pa->index <= 0) + break; + addr = pa->pages[--pa->index]; + if (pa->index == 0) { + pa = pa->next; + free_page((unsigned long) *list); + *list = pa; + } + free_page(addr); + (*counter)--; + nr--; + } + spin_unlock(&cmm_lock); + return nr; +} + +static int cmm_oom_notify(struct notifier_block *self, + unsigned long dummy, void *parm) +{ + unsigned long *freed = parm; + long nr = 256; + + nr = cmm_free_pages(nr, &cmm_timed_pages, &cmm_timed_page_list); + if (nr > 0) + nr = cmm_free_pages(nr, &cmm_pages, &cmm_page_list); + cmm_pages_target = cmm_pages; + cmm_timed_pages_target = cmm_timed_pages; + *freed += 256 - nr; + return NOTIFY_OK; +} + +static struct notifier_block cmm_oom_nb = { + .notifier_call = cmm_oom_notify, +}; + +static int cmm_thread(void *dummy) +{ + int rc; + + while (1) { + rc = wait_event_interruptible(cmm_thread_wait, + (!cmm_suspended && (cmm_pages != cmm_pages_target || + cmm_timed_pages != cmm_timed_pages_target)) || + kthread_should_stop()); + if (kthread_should_stop() || rc == -ERESTARTSYS) { + cmm_pages_target = cmm_pages; + cmm_timed_pages_target = cmm_timed_pages; + break; + } + if (cmm_pages_target > cmm_pages) { + if (cmm_alloc_pages(1, &cmm_pages, &cmm_page_list)) + cmm_pages_target = cmm_pages; + } else if (cmm_pages_target < cmm_pages) { + cmm_free_pages(1, &cmm_pages, &cmm_page_list); + } + if (cmm_timed_pages_target > cmm_timed_pages) { + if (cmm_alloc_pages(1, &cmm_timed_pages, + &cmm_timed_page_list)) + cmm_timed_pages_target = cmm_timed_pages; + } else if (cmm_timed_pages_target < cmm_timed_pages) { + cmm_free_pages(1, &cmm_timed_pages, + &cmm_timed_page_list); + } + if (cmm_timed_pages > 0 && !timer_pending(&cmm_timer)) + cmm_set_timer(); + } + return 0; +} + +static void cmm_kick_thread(void) +{ + wake_up(&cmm_thread_wait); +} + +static void cmm_set_timer(void) +{ + if (cmm_timed_pages_target <= 0 || cmm_timeout_seconds <= 0) { + if (timer_pending(&cmm_timer)) + del_timer(&cmm_timer); + return; + } + if (timer_pending(&cmm_timer)) { + if (mod_timer(&cmm_timer, jiffies + cmm_timeout_seconds*HZ)) + return; + } + cmm_timer.function = cmm_timer_fn; + cmm_timer.data = 0; + cmm_timer.expires = jiffies + cmm_timeout_seconds*HZ; + add_timer(&cmm_timer); +} + +static void cmm_timer_fn(unsigned long ignored) +{ + long nr; + + nr = cmm_timed_pages_target - cmm_timeout_pages; + if (nr < 0) + cmm_timed_pages_target = 0; + else + cmm_timed_pages_target = nr; + cmm_kick_thread(); + cmm_set_timer(); +} + +static void cmm_set_pages(long nr) +{ + cmm_pages_target = nr; + cmm_kick_thread(); +} + +static long cmm_get_pages(void) +{ + return cmm_pages; +} + +static void cmm_add_timed_pages(long nr) +{ + cmm_timed_pages_target += nr; + cmm_kick_thread(); +} + +static long cmm_get_timed_pages(void) +{ + return cmm_timed_pages; +} + +static void cmm_set_timeout(long nr, long seconds) +{ + cmm_timeout_pages = nr; + cmm_timeout_seconds = seconds; + cmm_set_timer(); +} + +static int cmm_skip_blanks(char *cp, char **endp) +{ + char *str; + + for (str = cp; *str == ' ' || *str == '\t'; str++) + ; + *endp = str; + return str != cp; +} + +static struct ctl_table cmm_table[]; + +static int cmm_pages_handler(struct ctl_table *ctl, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + char buf[16], *p; + unsigned int len; + long nr; + + if (!*lenp || (*ppos && !write)) { + *lenp = 0; + return 0; + } + + if (write) { + len = *lenp; + if (copy_from_user(buf, buffer, + len > sizeof(buf) ? sizeof(buf) : len)) + return -EFAULT; + buf[sizeof(buf) - 1] = '\0'; + cmm_skip_blanks(buf, &p); + nr = simple_strtoul(p, &p, 0); + if (ctl == &cmm_table[0]) + cmm_set_pages(nr); + else + cmm_add_timed_pages(nr); + } else { + if (ctl == &cmm_table[0]) + nr = cmm_get_pages(); + else + nr = cmm_get_timed_pages(); + len = sprintf(buf, "%ld\n", nr); + if (len > *lenp) + len = *lenp; + if (copy_to_user(buffer, buf, len)) + return -EFAULT; + } + *lenp = len; + *ppos += len; + return 0; +} + +static int cmm_timeout_handler(struct ctl_table *ctl, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + char buf[64], *p; + long nr, seconds; + unsigned int len; + + if (!*lenp || (*ppos && !write)) { + *lenp = 0; + return 0; + } + + if (write) { + len = *lenp; + if (copy_from_user(buf, buffer, + len > sizeof(buf) ? sizeof(buf) : len)) + return -EFAULT; + buf[sizeof(buf) - 1] = '\0'; + cmm_skip_blanks(buf, &p); + nr = simple_strtoul(p, &p, 0); + cmm_skip_blanks(p, &p); + seconds = simple_strtoul(p, &p, 0); + cmm_set_timeout(nr, seconds); + } else { + len = sprintf(buf, "%ld %ld\n", + cmm_timeout_pages, cmm_timeout_seconds); + if (len > *lenp) + len = *lenp; + if (copy_to_user(buffer, buf, len)) + return -EFAULT; + } + *lenp = len; + *ppos += len; + return 0; +} + +static struct ctl_table cmm_table[] = { + { + .procname = "cmm_pages", + .mode = 0644, + .proc_handler = cmm_pages_handler, + }, + { + .procname = "cmm_timed_pages", + .mode = 0644, + .proc_handler = cmm_pages_handler, + }, + { + .procname = "cmm_timeout", + .mode = 0644, + .proc_handler = cmm_timeout_handler, + }, + { } +}; + +static struct ctl_table cmm_dir_table[] = { + { + .procname = "vm", + .maxlen = 0, + .mode = 0555, + .child = cmm_table, + }, + { } +}; + +#ifdef CONFIG_CMM_IUCV +#define SMSG_PREFIX "CMM" +static void cmm_smsg_target(const char *from, char *msg) +{ + long nr, seconds; + + if (strlen(sender) > 0 && strcmp(from, sender) != 0) + return; + if (!cmm_skip_blanks(msg + strlen(SMSG_PREFIX), &msg)) + return; + if (strncmp(msg, "SHRINK", 6) == 0) { + if (!cmm_skip_blanks(msg + 6, &msg)) + return; + nr = simple_strtoul(msg, &msg, 0); + cmm_skip_blanks(msg, &msg); + if (*msg == '\0') + cmm_set_pages(nr); + } else if (strncmp(msg, "RELEASE", 7) == 0) { + if (!cmm_skip_blanks(msg + 7, &msg)) + return; + nr = simple_strtoul(msg, &msg, 0); + cmm_skip_blanks(msg, &msg); + if (*msg == '\0') + cmm_add_timed_pages(nr); + } else if (strncmp(msg, "REUSE", 5) == 0) { + if (!cmm_skip_blanks(msg + 5, &msg)) + return; + nr = simple_strtoul(msg, &msg, 0); + if (!cmm_skip_blanks(msg, &msg)) + return; + seconds = simple_strtoul(msg, &msg, 0); + cmm_skip_blanks(msg, &msg); + if (*msg == '\0') + cmm_set_timeout(nr, seconds); + } +} +#endif + +static struct ctl_table_header *cmm_sysctl_header; + +static int cmm_suspend(void) +{ + cmm_suspended = 1; + cmm_free_pages(cmm_pages, &cmm_pages, &cmm_page_list); + cmm_free_pages(cmm_timed_pages, &cmm_timed_pages, &cmm_timed_page_list); + return 0; +} + +static int cmm_resume(void) +{ + cmm_suspended = 0; + cmm_kick_thread(); + return 0; +} + +static int cmm_power_event(struct notifier_block *this, + unsigned long event, void *ptr) +{ + switch (event) { + case PM_POST_HIBERNATION: + return cmm_resume(); + case PM_HIBERNATION_PREPARE: + return cmm_suspend(); + default: + return NOTIFY_DONE; + } +} + +static struct notifier_block cmm_power_notifier = { + .notifier_call = cmm_power_event, +}; + +static int __init cmm_init(void) +{ + int rc = -ENOMEM; + + cmm_sysctl_header = register_sysctl_table(cmm_dir_table); + if (!cmm_sysctl_header) + goto out_sysctl; +#ifdef CONFIG_CMM_IUCV + /* convert sender to uppercase characters */ + if (sender) { + int len = strlen(sender); + while (len--) + sender[len] = toupper(sender[len]); + } else { + sender = cmm_default_sender; + } + + rc = smsg_register_callback(SMSG_PREFIX, cmm_smsg_target); + if (rc < 0) + goto out_smsg; +#endif + rc = register_oom_notifier(&cmm_oom_nb); + if (rc < 0) + goto out_oom_notify; + rc = register_pm_notifier(&cmm_power_notifier); + if (rc) + goto out_pm; + cmm_thread_ptr = kthread_run(cmm_thread, NULL, "cmmthread"); + if (!IS_ERR(cmm_thread_ptr)) + return 0; + + rc = PTR_ERR(cmm_thread_ptr); + unregister_pm_notifier(&cmm_power_notifier); +out_pm: + unregister_oom_notifier(&cmm_oom_nb); +out_oom_notify: +#ifdef CONFIG_CMM_IUCV + smsg_unregister_callback(SMSG_PREFIX, cmm_smsg_target); +out_smsg: +#endif + unregister_sysctl_table(cmm_sysctl_header); +out_sysctl: + del_timer_sync(&cmm_timer); + return rc; +} +module_init(cmm_init); + +static void __exit cmm_exit(void) +{ + unregister_sysctl_table(cmm_sysctl_header); +#ifdef CONFIG_CMM_IUCV + smsg_unregister_callback(SMSG_PREFIX, cmm_smsg_target); +#endif + unregister_pm_notifier(&cmm_power_notifier); + unregister_oom_notifier(&cmm_oom_nb); + kthread_stop(cmm_thread_ptr); + del_timer_sync(&cmm_timer); + cmm_free_pages(cmm_pages, &cmm_pages, &cmm_page_list); + cmm_free_pages(cmm_timed_pages, &cmm_timed_pages, &cmm_timed_page_list); +} +module_exit(cmm_exit); + +MODULE_LICENSE("GPL"); diff --git a/arch/s390/mm/dump_pagetables.c b/arch/s390/mm/dump_pagetables.c new file mode 100644 index 000000000..8556d6be9 --- /dev/null +++ b/arch/s390/mm/dump_pagetables.c @@ -0,0 +1,225 @@ +#include <linux/seq_file.h> +#include <linux/debugfs.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <asm/sections.h> +#include <asm/pgtable.h> + +static unsigned long max_addr; + +struct addr_marker { + unsigned long start_address; + const char *name; +}; + +enum address_markers_idx { + IDENTITY_NR = 0, + KERNEL_START_NR, + KERNEL_END_NR, + VMEMMAP_NR, + VMALLOC_NR, + MODULES_NR, +}; + +static struct addr_marker address_markers[] = { + [IDENTITY_NR] = {0, "Identity Mapping"}, + [KERNEL_START_NR] = {(unsigned long)&_stext, "Kernel Image Start"}, + [KERNEL_END_NR] = {(unsigned long)&_end, "Kernel Image End"}, + [VMEMMAP_NR] = {0, "vmemmap Area"}, + [VMALLOC_NR] = {0, "vmalloc Area"}, + [MODULES_NR] = {0, "Modules Area"}, + { -1, NULL } +}; + +struct pg_state { + int level; + unsigned int current_prot; + unsigned long start_address; + unsigned long current_address; + const struct addr_marker *marker; +}; + +static void print_prot(struct seq_file *m, unsigned int pr, int level) +{ + static const char * const level_name[] = + { "ASCE", "PGD", "PUD", "PMD", "PTE" }; + + seq_printf(m, "%s ", level_name[level]); + if (pr & _PAGE_INVALID) { + seq_printf(m, "I\n"); + return; + } + seq_printf(m, "%s", pr & _PAGE_PROTECT ? "RO " : "RW "); + seq_putc(m, '\n'); +} + +static void note_page(struct seq_file *m, struct pg_state *st, + unsigned int new_prot, int level) +{ + static const char units[] = "KMGTPE"; + int width = sizeof(unsigned long) * 2; + const char *unit = units; + unsigned int prot, cur; + unsigned long delta; + + /* + * If we have a "break" in the series, we need to flush the state + * that we have now. "break" is either changing perms, levels or + * address space marker. + */ + prot = new_prot; + cur = st->current_prot; + + if (!st->level) { + /* First entry */ + st->current_prot = new_prot; + st->level = level; + st->marker = address_markers; + seq_printf(m, "---[ %s ]---\n", st->marker->name); + } else if (prot != cur || level != st->level || + st->current_address >= st->marker[1].start_address) { + /* Print the actual finished series */ + seq_printf(m, "0x%0*lx-0x%0*lx", + width, st->start_address, + width, st->current_address); + delta = (st->current_address - st->start_address) >> 10; + while (!(delta & 0x3ff) && unit[1]) { + delta >>= 10; + unit++; + } + seq_printf(m, "%9lu%c ", delta, *unit); + print_prot(m, st->current_prot, st->level); + if (st->current_address >= st->marker[1].start_address) { + st->marker++; + seq_printf(m, "---[ %s ]---\n", st->marker->name); + } + st->start_address = st->current_address; + st->current_prot = new_prot; + st->level = level; + } +} + +/* + * The actual page table walker functions. In order to keep the + * implementation of print_prot() short, we only check and pass + * _PAGE_INVALID and _PAGE_PROTECT flags to note_page() if a region, + * segment or page table entry is invalid or read-only. + * After all it's just a hint that the current level being walked + * contains an invalid or read-only entry. + */ +static void walk_pte_level(struct seq_file *m, struct pg_state *st, + pmd_t *pmd, unsigned long addr) +{ + unsigned int prot; + pte_t *pte; + int i; + + for (i = 0; i < PTRS_PER_PTE && addr < max_addr; i++) { + st->current_address = addr; + pte = pte_offset_kernel(pmd, addr); + prot = pte_val(*pte) & (_PAGE_PROTECT | _PAGE_INVALID); + note_page(m, st, prot, 4); + addr += PAGE_SIZE; + } +} + +static void walk_pmd_level(struct seq_file *m, struct pg_state *st, + pud_t *pud, unsigned long addr) +{ + unsigned int prot; + pmd_t *pmd; + int i; + + for (i = 0; i < PTRS_PER_PMD && addr < max_addr; i++) { + st->current_address = addr; + pmd = pmd_offset(pud, addr); + if (!pmd_none(*pmd)) { + if (pmd_large(*pmd)) { + prot = pmd_val(*pmd) & _SEGMENT_ENTRY_PROTECT; + note_page(m, st, prot, 3); + } else + walk_pte_level(m, st, pmd, addr); + } else + note_page(m, st, _PAGE_INVALID, 3); + addr += PMD_SIZE; + } +} + +static void walk_pud_level(struct seq_file *m, struct pg_state *st, + pgd_t *pgd, unsigned long addr) +{ + unsigned int prot; + pud_t *pud; + int i; + + for (i = 0; i < PTRS_PER_PUD && addr < max_addr; i++) { + st->current_address = addr; + pud = pud_offset(pgd, addr); + if (!pud_none(*pud)) + if (pud_large(*pud)) { + prot = pud_val(*pud) & _REGION3_ENTRY_RO; + note_page(m, st, prot, 2); + } else + walk_pmd_level(m, st, pud, addr); + else + note_page(m, st, _PAGE_INVALID, 2); + addr += PUD_SIZE; + } +} + +static void walk_pgd_level(struct seq_file *m) +{ + unsigned long addr = 0; + struct pg_state st; + pgd_t *pgd; + int i; + + memset(&st, 0, sizeof(st)); + for (i = 0; i < PTRS_PER_PGD && addr < max_addr; i++) { + st.current_address = addr; + pgd = pgd_offset_k(addr); + if (!pgd_none(*pgd)) + walk_pud_level(m, &st, pgd, addr); + else + note_page(m, &st, _PAGE_INVALID, 1); + addr += PGDIR_SIZE; + } + /* Flush out the last page */ + st.current_address = max_addr; + note_page(m, &st, 0, 0); +} + +static int ptdump_show(struct seq_file *m, void *v) +{ + walk_pgd_level(m); + return 0; +} + +static int ptdump_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, ptdump_show, NULL); +} + +static const struct file_operations ptdump_fops = { + .open = ptdump_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int pt_dump_init(void) +{ + /* + * Figure out the maximum virtual address being accessible with the + * kernel ASCE. We need this to keep the page table walker functions + * from accessing non-existent entries. + */ + max_addr = (S390_lowcore.kernel_asce & _REGION_ENTRY_TYPE_MASK) >> 2; + max_addr = 1UL << (max_addr * 11 + 31); + address_markers[MODULES_NR].start_address = MODULES_VADDR; + address_markers[VMEMMAP_NR].start_address = (unsigned long) vmemmap; + address_markers[VMALLOC_NR].start_address = VMALLOC_START; + debugfs_create_file("kernel_page_tables", 0400, NULL, NULL, &ptdump_fops); + return 0; +} +device_initcall(pt_dump_init); diff --git a/arch/s390/mm/extable.c b/arch/s390/mm/extable.c new file mode 100644 index 000000000..4d1ee8886 --- /dev/null +++ b/arch/s390/mm/extable.c @@ -0,0 +1,81 @@ +#include <linux/module.h> +#include <linux/sort.h> +#include <asm/uaccess.h> + +/* + * Search one exception table for an entry corresponding to the + * given instruction address, and return the address of the entry, + * or NULL if none is found. + * We use a binary search, and thus we assume that the table is + * already sorted. + */ +const struct exception_table_entry * +search_extable(const struct exception_table_entry *first, + const struct exception_table_entry *last, + unsigned long value) +{ + const struct exception_table_entry *mid; + unsigned long addr; + + while (first <= last) { + mid = ((last - first) >> 1) + first; + addr = extable_insn(mid); + if (addr < value) + first = mid + 1; + else if (addr > value) + last = mid - 1; + else + return mid; + } + return NULL; +} + +/* + * The exception table needs to be sorted so that the binary + * search that we use to find entries in it works properly. + * This is used both for the kernel exception table and for + * the exception tables of modules that get loaded. + * + */ +static int cmp_ex(const void *a, const void *b) +{ + const struct exception_table_entry *x = a, *y = b; + + /* This compare is only valid after normalization. */ + return x->insn - y->insn; +} + +void sort_extable(struct exception_table_entry *start, + struct exception_table_entry *finish) +{ + struct exception_table_entry *p; + int i; + + /* Normalize entries to being relative to the start of the section */ + for (p = start, i = 0; p < finish; p++, i += 8) + p->insn += i; + sort(start, finish - start, sizeof(*start), cmp_ex, NULL); + /* Denormalize all entries */ + for (p = start, i = 0; p < finish; p++, i += 8) + p->insn -= i; +} + +#ifdef CONFIG_MODULES +/* + * If the exception table is sorted, any referring to the module init + * will be at the beginning or the end. + */ +void trim_init_extable(struct module *m) +{ + /* Trim the beginning */ + while (m->num_exentries && + within_module_init(extable_insn(&m->extable[0]), m)) { + m->extable++; + m->num_exentries--; + } + /* Trim the end */ + while (m->num_exentries && + within_module_init(extable_insn(&m->extable[m->num_exentries-1]), m)) + m->num_exentries--; +} +#endif /* CONFIG_MODULES */ diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c new file mode 100644 index 000000000..23c496957 --- /dev/null +++ b/arch/s390/mm/extmem.c @@ -0,0 +1,763 @@ +/* + * Author(s)......: Carsten Otte <cotte@de.ibm.com> + * Rob M van der Heij <rvdheij@nl.ibm.com> + * Steven Shultz <shultzss@us.ibm.com> + * Bugreports.to..: <Linux390@de.ibm.com> + * Copyright IBM Corp. 2002, 2004 + */ + +#define KMSG_COMPONENT "extmem" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/spinlock.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/module.h> +#include <linux/bootmem.h> +#include <linux/ctype.h> +#include <linux/ioport.h> +#include <asm/page.h> +#include <asm/pgtable.h> +#include <asm/ebcdic.h> +#include <asm/errno.h> +#include <asm/extmem.h> +#include <asm/cpcmd.h> +#include <asm/setup.h> + +#define DCSS_LOADSHR 0x00 +#define DCSS_LOADNSR 0x04 +#define DCSS_PURGESEG 0x08 +#define DCSS_FINDSEG 0x0c +#define DCSS_LOADNOLY 0x10 +#define DCSS_SEGEXT 0x18 +#define DCSS_LOADSHRX 0x20 +#define DCSS_LOADNSRX 0x24 +#define DCSS_FINDSEGX 0x2c +#define DCSS_SEGEXTX 0x38 +#define DCSS_FINDSEGA 0x0c + +struct qrange { + unsigned long start; /* last byte type */ + unsigned long end; /* last byte reserved */ +}; + +struct qout64 { + unsigned long segstart; + unsigned long segend; + int segcnt; + int segrcnt; + struct qrange range[6]; +}; + +struct qrange_old { + unsigned int start; /* last byte type */ + unsigned int end; /* last byte reserved */ +}; + +/* output area format for the Diag x'64' old subcode x'18' */ +struct qout64_old { + int segstart; + int segend; + int segcnt; + int segrcnt; + struct qrange_old range[6]; +}; + +struct qin64 { + char qopcode; + char rsrv1[3]; + char qrcode; + char rsrv2[3]; + char qname[8]; + unsigned int qoutptr; + short int qoutlen; +}; + +struct dcss_segment { + struct list_head list; + char dcss_name[8]; + char res_name[15]; + unsigned long start_addr; + unsigned long end; + atomic_t ref_count; + int do_nonshared; + unsigned int vm_segtype; + struct qrange range[6]; + int segcnt; + struct resource *res; +}; + +static DEFINE_MUTEX(dcss_lock); +static LIST_HEAD(dcss_list); +static char *segtype_string[] = { "SW", "EW", "SR", "ER", "SN", "EN", "SC", + "EW/EN-MIXED" }; +static int loadshr_scode, loadnsr_scode, findseg_scode; +static int segext_scode, purgeseg_scode; +static int scode_set; + +/* set correct Diag x'64' subcodes. */ +static int +dcss_set_subcodes(void) +{ + char *name = kmalloc(8 * sizeof(char), GFP_KERNEL | GFP_DMA); + unsigned long rx, ry; + int rc; + + if (name == NULL) + return -ENOMEM; + + rx = (unsigned long) name; + ry = DCSS_FINDSEGX; + + strcpy(name, "dummy"); + asm volatile( + " diag %0,%1,0x64\n" + "0: ipm %2\n" + " srl %2,28\n" + " j 2f\n" + "1: la %2,3\n" + "2:\n" + EX_TABLE(0b, 1b) + : "+d" (rx), "+d" (ry), "=d" (rc) : : "cc"); + + kfree(name); + /* Diag x'64' new subcodes are supported, set to new subcodes */ + if (rc != 3) { + loadshr_scode = DCSS_LOADSHRX; + loadnsr_scode = DCSS_LOADNSRX; + purgeseg_scode = DCSS_PURGESEG; + findseg_scode = DCSS_FINDSEGX; + segext_scode = DCSS_SEGEXTX; + return 0; + } + /* Diag x'64' new subcodes are not supported, set to old subcodes */ + loadshr_scode = DCSS_LOADNOLY; + loadnsr_scode = DCSS_LOADNSR; + purgeseg_scode = DCSS_PURGESEG; + findseg_scode = DCSS_FINDSEG; + segext_scode = DCSS_SEGEXT; + return 0; +} + +/* + * Create the 8 bytes, ebcdic VM segment name from + * an ascii name. + */ +static void +dcss_mkname(char *name, char *dcss_name) +{ + int i; + + for (i = 0; i < 8; i++) { + if (name[i] == '\0') + break; + dcss_name[i] = toupper(name[i]); + }; + for (; i < 8; i++) + dcss_name[i] = ' '; + ASCEBC(dcss_name, 8); +} + + +/* + * search all segments in dcss_list, and return the one + * namend *name. If not found, return NULL. + */ +static struct dcss_segment * +segment_by_name (char *name) +{ + char dcss_name[9]; + struct list_head *l; + struct dcss_segment *tmp, *retval = NULL; + + BUG_ON(!mutex_is_locked(&dcss_lock)); + dcss_mkname (name, dcss_name); + list_for_each (l, &dcss_list) { + tmp = list_entry (l, struct dcss_segment, list); + if (memcmp(tmp->dcss_name, dcss_name, 8) == 0) { + retval = tmp; + break; + } + } + return retval; +} + + +/* + * Perform a function on a dcss segment. + */ +static inline int +dcss_diag(int *func, void *parameter, + unsigned long *ret1, unsigned long *ret2) +{ + unsigned long rx, ry; + int rc; + + if (scode_set == 0) { + rc = dcss_set_subcodes(); + if (rc < 0) + return rc; + scode_set = 1; + } + rx = (unsigned long) parameter; + ry = (unsigned long) *func; + + /* 64-bit Diag x'64' new subcode, keep in 64-bit addressing mode */ + if (*func > DCSS_SEGEXT) + asm volatile( + " diag %0,%1,0x64\n" + " ipm %2\n" + " srl %2,28\n" + : "+d" (rx), "+d" (ry), "=d" (rc) : : "cc"); + /* 31-bit Diag x'64' old subcode, switch to 31-bit addressing mode */ + else + asm volatile( + " sam31\n" + " diag %0,%1,0x64\n" + " sam64\n" + " ipm %2\n" + " srl %2,28\n" + : "+d" (rx), "+d" (ry), "=d" (rc) : : "cc"); + *ret1 = rx; + *ret2 = ry; + return rc; +} + +static inline int +dcss_diag_translate_rc (int vm_rc) { + if (vm_rc == 44) + return -ENOENT; + return -EIO; +} + + +/* do a diag to get info about a segment. + * fills start_address, end and vm_segtype fields + */ +static int +query_segment_type (struct dcss_segment *seg) +{ + unsigned long dummy, vmrc; + int diag_cc, rc, i; + struct qout64 *qout; + struct qin64 *qin; + + qin = kmalloc(sizeof(*qin), GFP_KERNEL | GFP_DMA); + qout = kmalloc(sizeof(*qout), GFP_KERNEL | GFP_DMA); + if ((qin == NULL) || (qout == NULL)) { + rc = -ENOMEM; + goto out_free; + } + + /* initialize diag input parameters */ + qin->qopcode = DCSS_FINDSEGA; + qin->qoutptr = (unsigned long) qout; + qin->qoutlen = sizeof(struct qout64); + memcpy (qin->qname, seg->dcss_name, 8); + + diag_cc = dcss_diag(&segext_scode, qin, &dummy, &vmrc); + + if (diag_cc < 0) { + rc = diag_cc; + goto out_free; + } + if (diag_cc > 1) { + pr_warning("Querying a DCSS type failed with rc=%ld\n", vmrc); + rc = dcss_diag_translate_rc (vmrc); + goto out_free; + } + + /* Only old format of output area of Diagnose x'64' is supported, + copy data for the new format. */ + if (segext_scode == DCSS_SEGEXT) { + struct qout64_old *qout_old; + qout_old = kzalloc(sizeof(*qout_old), GFP_KERNEL | GFP_DMA); + if (qout_old == NULL) { + rc = -ENOMEM; + goto out_free; + } + memcpy(qout_old, qout, sizeof(struct qout64_old)); + qout->segstart = (unsigned long) qout_old->segstart; + qout->segend = (unsigned long) qout_old->segend; + qout->segcnt = qout_old->segcnt; + qout->segrcnt = qout_old->segrcnt; + + if (qout->segcnt > 6) + qout->segrcnt = 6; + for (i = 0; i < qout->segrcnt; i++) { + qout->range[i].start = + (unsigned long) qout_old->range[i].start; + qout->range[i].end = + (unsigned long) qout_old->range[i].end; + } + kfree(qout_old); + } + if (qout->segcnt > 6) { + rc = -EOPNOTSUPP; + goto out_free; + } + + if (qout->segcnt == 1) { + seg->vm_segtype = qout->range[0].start & 0xff; + } else { + /* multi-part segment. only one type supported here: + - all parts are contiguous + - all parts are either EW or EN type + - maximum 6 parts allowed */ + unsigned long start = qout->segstart >> PAGE_SHIFT; + for (i=0; i<qout->segcnt; i++) { + if (((qout->range[i].start & 0xff) != SEG_TYPE_EW) && + ((qout->range[i].start & 0xff) != SEG_TYPE_EN)) { + rc = -EOPNOTSUPP; + goto out_free; + } + if (start != qout->range[i].start >> PAGE_SHIFT) { + rc = -EOPNOTSUPP; + goto out_free; + } + start = (qout->range[i].end >> PAGE_SHIFT) + 1; + } + seg->vm_segtype = SEG_TYPE_EWEN; + } + + /* analyze diag output and update seg */ + seg->start_addr = qout->segstart; + seg->end = qout->segend; + + memcpy (seg->range, qout->range, 6*sizeof(struct qrange)); + seg->segcnt = qout->segcnt; + + rc = 0; + + out_free: + kfree(qin); + kfree(qout); + return rc; +} + +/* + * get info about a segment + * possible return values: + * -ENOSYS : we are not running on VM + * -EIO : could not perform query diagnose + * -ENOENT : no such segment + * -EOPNOTSUPP: multi-part segment cannot be used with linux + * -ENOMEM : out of memory + * 0 .. 6 : type of segment as defined in include/asm-s390/extmem.h + */ +int +segment_type (char* name) +{ + int rc; + struct dcss_segment seg; + + if (!MACHINE_IS_VM) + return -ENOSYS; + + dcss_mkname(name, seg.dcss_name); + rc = query_segment_type (&seg); + if (rc < 0) + return rc; + return seg.vm_segtype; +} + +/* + * check if segment collides with other segments that are currently loaded + * returns 1 if this is the case, 0 if no collision was found + */ +static int +segment_overlaps_others (struct dcss_segment *seg) +{ + struct list_head *l; + struct dcss_segment *tmp; + + BUG_ON(!mutex_is_locked(&dcss_lock)); + list_for_each(l, &dcss_list) { + tmp = list_entry(l, struct dcss_segment, list); + if ((tmp->start_addr >> 20) > (seg->end >> 20)) + continue; + if ((tmp->end >> 20) < (seg->start_addr >> 20)) + continue; + if (seg == tmp) + continue; + return 1; + } + return 0; +} + +/* + * real segment loading function, called from segment_load + */ +static int +__segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long *end) +{ + unsigned long start_addr, end_addr, dummy; + struct dcss_segment *seg; + int rc, diag_cc; + + start_addr = end_addr = 0; + seg = kmalloc(sizeof(*seg), GFP_KERNEL | GFP_DMA); + if (seg == NULL) { + rc = -ENOMEM; + goto out; + } + dcss_mkname (name, seg->dcss_name); + rc = query_segment_type (seg); + if (rc < 0) + goto out_free; + + if (loadshr_scode == DCSS_LOADSHRX) { + if (segment_overlaps_others(seg)) { + rc = -EBUSY; + goto out_free; + } + } + + rc = vmem_add_mapping(seg->start_addr, seg->end - seg->start_addr + 1); + + if (rc) + goto out_free; + + seg->res = kzalloc(sizeof(struct resource), GFP_KERNEL); + if (seg->res == NULL) { + rc = -ENOMEM; + goto out_shared; + } + seg->res->flags = IORESOURCE_BUSY | IORESOURCE_MEM; + seg->res->start = seg->start_addr; + seg->res->end = seg->end; + memcpy(&seg->res_name, seg->dcss_name, 8); + EBCASC(seg->res_name, 8); + seg->res_name[8] = '\0'; + strncat(seg->res_name, " (DCSS)", 7); + seg->res->name = seg->res_name; + rc = seg->vm_segtype; + if (rc == SEG_TYPE_SC || + ((rc == SEG_TYPE_SR || rc == SEG_TYPE_ER) && !do_nonshared)) + seg->res->flags |= IORESOURCE_READONLY; + if (request_resource(&iomem_resource, seg->res)) { + rc = -EBUSY; + kfree(seg->res); + goto out_shared; + } + + if (do_nonshared) + diag_cc = dcss_diag(&loadnsr_scode, seg->dcss_name, + &start_addr, &end_addr); + else + diag_cc = dcss_diag(&loadshr_scode, seg->dcss_name, + &start_addr, &end_addr); + if (diag_cc < 0) { + dcss_diag(&purgeseg_scode, seg->dcss_name, + &dummy, &dummy); + rc = diag_cc; + goto out_resource; + } + if (diag_cc > 1) { + pr_warning("Loading DCSS %s failed with rc=%ld\n", name, + end_addr); + rc = dcss_diag_translate_rc(end_addr); + dcss_diag(&purgeseg_scode, seg->dcss_name, + &dummy, &dummy); + goto out_resource; + } + seg->start_addr = start_addr; + seg->end = end_addr; + seg->do_nonshared = do_nonshared; + atomic_set(&seg->ref_count, 1); + list_add(&seg->list, &dcss_list); + *addr = seg->start_addr; + *end = seg->end; + if (do_nonshared) + pr_info("DCSS %s of range %p to %p and type %s loaded as " + "exclusive-writable\n", name, (void*) seg->start_addr, + (void*) seg->end, segtype_string[seg->vm_segtype]); + else { + pr_info("DCSS %s of range %p to %p and type %s loaded in " + "shared access mode\n", name, (void*) seg->start_addr, + (void*) seg->end, segtype_string[seg->vm_segtype]); + } + goto out; + out_resource: + release_resource(seg->res); + kfree(seg->res); + out_shared: + vmem_remove_mapping(seg->start_addr, seg->end - seg->start_addr + 1); + out_free: + kfree(seg); + out: + return rc; +} + +/* + * this function loads a DCSS segment + * name : name of the DCSS + * do_nonshared : 0 indicates that the dcss should be shared with other linux images + * 1 indicates that the dcss should be exclusive for this linux image + * addr : will be filled with start address of the segment + * end : will be filled with end address of the segment + * return values: + * -ENOSYS : we are not running on VM + * -EIO : could not perform query or load diagnose + * -ENOENT : no such segment + * -EOPNOTSUPP: multi-part segment cannot be used with linux + * -ENOSPC : segment cannot be used (overlaps with storage) + * -EBUSY : segment can temporarily not be used (overlaps with dcss) + * -ERANGE : segment cannot be used (exceeds kernel mapping range) + * -EPERM : segment is currently loaded with incompatible permissions + * -ENOMEM : out of memory + * 0 .. 6 : type of segment as defined in include/asm-s390/extmem.h + */ +int +segment_load (char *name, int do_nonshared, unsigned long *addr, + unsigned long *end) +{ + struct dcss_segment *seg; + int rc; + + if (!MACHINE_IS_VM) + return -ENOSYS; + + mutex_lock(&dcss_lock); + seg = segment_by_name (name); + if (seg == NULL) + rc = __segment_load (name, do_nonshared, addr, end); + else { + if (do_nonshared == seg->do_nonshared) { + atomic_inc(&seg->ref_count); + *addr = seg->start_addr; + *end = seg->end; + rc = seg->vm_segtype; + } else { + *addr = *end = 0; + rc = -EPERM; + } + } + mutex_unlock(&dcss_lock); + return rc; +} + +/* + * this function modifies the shared state of a DCSS segment. note that + * name : name of the DCSS + * do_nonshared : 0 indicates that the dcss should be shared with other linux images + * 1 indicates that the dcss should be exclusive for this linux image + * return values: + * -EIO : could not perform load diagnose (segment gone!) + * -ENOENT : no such segment (segment gone!) + * -EAGAIN : segment is in use by other exploiters, try later + * -EINVAL : no segment with the given name is currently loaded - name invalid + * -EBUSY : segment can temporarily not be used (overlaps with dcss) + * 0 : operation succeeded + */ +int +segment_modify_shared (char *name, int do_nonshared) +{ + struct dcss_segment *seg; + unsigned long start_addr, end_addr, dummy; + int rc, diag_cc; + + start_addr = end_addr = 0; + mutex_lock(&dcss_lock); + seg = segment_by_name (name); + if (seg == NULL) { + rc = -EINVAL; + goto out_unlock; + } + if (do_nonshared == seg->do_nonshared) { + pr_info("DCSS %s is already in the requested access " + "mode\n", name); + rc = 0; + goto out_unlock; + } + if (atomic_read (&seg->ref_count) != 1) { + pr_warning("DCSS %s is in use and cannot be reloaded\n", + name); + rc = -EAGAIN; + goto out_unlock; + } + release_resource(seg->res); + if (do_nonshared) + seg->res->flags &= ~IORESOURCE_READONLY; + else + if (seg->vm_segtype == SEG_TYPE_SR || + seg->vm_segtype == SEG_TYPE_ER) + seg->res->flags |= IORESOURCE_READONLY; + + if (request_resource(&iomem_resource, seg->res)) { + pr_warning("DCSS %s overlaps with used memory resources " + "and cannot be reloaded\n", name); + rc = -EBUSY; + kfree(seg->res); + goto out_del_mem; + } + + dcss_diag(&purgeseg_scode, seg->dcss_name, &dummy, &dummy); + if (do_nonshared) + diag_cc = dcss_diag(&loadnsr_scode, seg->dcss_name, + &start_addr, &end_addr); + else + diag_cc = dcss_diag(&loadshr_scode, seg->dcss_name, + &start_addr, &end_addr); + if (diag_cc < 0) { + rc = diag_cc; + goto out_del_res; + } + if (diag_cc > 1) { + pr_warning("Reloading DCSS %s failed with rc=%ld\n", name, + end_addr); + rc = dcss_diag_translate_rc(end_addr); + goto out_del_res; + } + seg->start_addr = start_addr; + seg->end = end_addr; + seg->do_nonshared = do_nonshared; + rc = 0; + goto out_unlock; + out_del_res: + release_resource(seg->res); + kfree(seg->res); + out_del_mem: + vmem_remove_mapping(seg->start_addr, seg->end - seg->start_addr + 1); + list_del(&seg->list); + dcss_diag(&purgeseg_scode, seg->dcss_name, &dummy, &dummy); + kfree(seg); + out_unlock: + mutex_unlock(&dcss_lock); + return rc; +} + +/* + * Decrease the use count of a DCSS segment and remove + * it from the address space if nobody is using it + * any longer. + */ +void +segment_unload(char *name) +{ + unsigned long dummy; + struct dcss_segment *seg; + + if (!MACHINE_IS_VM) + return; + + mutex_lock(&dcss_lock); + seg = segment_by_name (name); + if (seg == NULL) { + pr_err("Unloading unknown DCSS %s failed\n", name); + goto out_unlock; + } + if (atomic_dec_return(&seg->ref_count) != 0) + goto out_unlock; + release_resource(seg->res); + kfree(seg->res); + vmem_remove_mapping(seg->start_addr, seg->end - seg->start_addr + 1); + list_del(&seg->list); + dcss_diag(&purgeseg_scode, seg->dcss_name, &dummy, &dummy); + kfree(seg); +out_unlock: + mutex_unlock(&dcss_lock); +} + +/* + * save segment content permanently + */ +void +segment_save(char *name) +{ + struct dcss_segment *seg; + char cmd1[160]; + char cmd2[80]; + int i, response; + + if (!MACHINE_IS_VM) + return; + + mutex_lock(&dcss_lock); + seg = segment_by_name (name); + + if (seg == NULL) { + pr_err("Saving unknown DCSS %s failed\n", name); + goto out; + } + + sprintf(cmd1, "DEFSEG %s", name); + for (i=0; i<seg->segcnt; i++) { + sprintf(cmd1+strlen(cmd1), " %lX-%lX %s", + seg->range[i].start >> PAGE_SHIFT, + seg->range[i].end >> PAGE_SHIFT, + segtype_string[seg->range[i].start & 0xff]); + } + sprintf(cmd2, "SAVESEG %s", name); + response = 0; + cpcmd(cmd1, NULL, 0, &response); + if (response) { + pr_err("Saving a DCSS failed with DEFSEG response code " + "%i\n", response); + goto out; + } + cpcmd(cmd2, NULL, 0, &response); + if (response) { + pr_err("Saving a DCSS failed with SAVESEG response code " + "%i\n", response); + goto out; + } +out: + mutex_unlock(&dcss_lock); +} + +/* + * print appropriate error message for segment_load()/segment_type() + * return code + */ +void segment_warning(int rc, char *seg_name) +{ + switch (rc) { + case -ENOENT: + pr_err("DCSS %s cannot be loaded or queried\n", seg_name); + break; + case -ENOSYS: + pr_err("DCSS %s cannot be loaded or queried without " + "z/VM\n", seg_name); + break; + case -EIO: + pr_err("Loading or querying DCSS %s resulted in a " + "hardware error\n", seg_name); + break; + case -EOPNOTSUPP: + pr_err("DCSS %s has multiple page ranges and cannot be " + "loaded or queried\n", seg_name); + break; + case -ENOSPC: + pr_err("DCSS %s overlaps with used storage and cannot " + "be loaded\n", seg_name); + break; + case -EBUSY: + pr_err("%s needs used memory resources and cannot be " + "loaded or queried\n", seg_name); + break; + case -EPERM: + pr_err("DCSS %s is already loaded in a different access " + "mode\n", seg_name); + break; + case -ENOMEM: + pr_err("There is not enough memory to load or query " + "DCSS %s\n", seg_name); + break; + case -ERANGE: + pr_err("DCSS %s exceeds the kernel mapping range (%lu) " + "and cannot be loaded\n", seg_name, VMEM_MAX_PHYS); + break; + default: + break; + } +} + +EXPORT_SYMBOL(segment_load); +EXPORT_SYMBOL(segment_unload); +EXPORT_SYMBOL(segment_save); +EXPORT_SYMBOL(segment_type); +EXPORT_SYMBOL(segment_modify_shared); +EXPORT_SYMBOL(segment_warning); diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c new file mode 100644 index 000000000..76515bcea --- /dev/null +++ b/arch/s390/mm/fault.c @@ -0,0 +1,758 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Hartmut Penner (hp@de.ibm.com) + * Ulrich Weigand (uweigand@de.ibm.com) + * + * Derived from "arch/i386/mm/fault.c" + * Copyright (C) 1995 Linus Torvalds + */ + +#include <linux/kernel_stat.h> +#include <linux/perf_event.h> +#include <linux/signal.h> +#include <linux/sched.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/string.h> +#include <linux/types.h> +#include <linux/ptrace.h> +#include <linux/mman.h> +#include <linux/mm.h> +#include <linux/compat.h> +#include <linux/smp.h> +#include <linux/kdebug.h> +#include <linux/init.h> +#include <linux/console.h> +#include <linux/module.h> +#include <linux/hardirq.h> +#include <linux/kprobes.h> +#include <linux/uaccess.h> +#include <linux/hugetlb.h> +#include <asm/asm-offsets.h> +#include <asm/pgtable.h> +#include <asm/irq.h> +#include <asm/mmu_context.h> +#include <asm/facility.h> +#include "../kernel/entry.h" + +#define __FAIL_ADDR_MASK -4096L +#define __SUBCODE_MASK 0x0600 +#define __PF_RES_FIELD 0x8000000000000000ULL + +#define VM_FAULT_BADCONTEXT 0x010000 +#define VM_FAULT_BADMAP 0x020000 +#define VM_FAULT_BADACCESS 0x040000 +#define VM_FAULT_SIGNAL 0x080000 +#define VM_FAULT_PFAULT 0x100000 + +static unsigned long store_indication __read_mostly; + +static int __init fault_init(void) +{ + if (test_facility(75)) + store_indication = 0xc00; + return 0; +} +early_initcall(fault_init); + +static inline int notify_page_fault(struct pt_regs *regs) +{ + int ret = 0; + + /* kprobe_running() needs smp_processor_id() */ + if (kprobes_built_in() && !user_mode(regs)) { + preempt_disable(); + if (kprobe_running() && kprobe_fault_handler(regs, 14)) + ret = 1; + preempt_enable(); + } + return ret; +} + + +/* + * Unlock any spinlocks which will prevent us from getting the + * message out. + */ +void bust_spinlocks(int yes) +{ + if (yes) { + oops_in_progress = 1; + } else { + int loglevel_save = console_loglevel; + console_unblank(); + oops_in_progress = 0; + /* + * OK, the message is on the console. Now we call printk() + * without oops_in_progress set so that printk will give klogd + * a poke. Hold onto your hats... + */ + console_loglevel = 15; + printk(" "); + console_loglevel = loglevel_save; + } +} + +/* + * Returns the address space associated with the fault. + * Returns 0 for kernel space and 1 for user space. + */ +static inline int user_space_fault(struct pt_regs *regs) +{ + unsigned long trans_exc_code; + + /* + * The lowest two bits of the translation exception + * identification indicate which paging table was used. + */ + trans_exc_code = regs->int_parm_long & 3; + if (trans_exc_code == 3) /* home space -> kernel */ + return 0; + if (user_mode(regs)) + return 1; + if (trans_exc_code == 2) /* secondary space -> set_fs */ + return current->thread.mm_segment.ar4; + if (current->flags & PF_VCPU) + return 1; + return 0; +} + +static int bad_address(void *p) +{ + unsigned long dummy; + + return probe_kernel_address((unsigned long *)p, dummy); +} + +static void dump_pagetable(unsigned long asce, unsigned long address) +{ + unsigned long *table = __va(asce & PAGE_MASK); + + pr_alert("AS:%016lx ", asce); + switch (asce & _ASCE_TYPE_MASK) { + case _ASCE_TYPE_REGION1: + table = table + ((address >> 53) & 0x7ff); + if (bad_address(table)) + goto bad; + pr_cont("R1:%016lx ", *table); + if (*table & _REGION_ENTRY_INVALID) + goto out; + table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN); + /* fallthrough */ + case _ASCE_TYPE_REGION2: + table = table + ((address >> 42) & 0x7ff); + if (bad_address(table)) + goto bad; + pr_cont("R2:%016lx ", *table); + if (*table & _REGION_ENTRY_INVALID) + goto out; + table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN); + /* fallthrough */ + case _ASCE_TYPE_REGION3: + table = table + ((address >> 31) & 0x7ff); + if (bad_address(table)) + goto bad; + pr_cont("R3:%016lx ", *table); + if (*table & (_REGION_ENTRY_INVALID | _REGION3_ENTRY_LARGE)) + goto out; + table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN); + /* fallthrough */ + case _ASCE_TYPE_SEGMENT: + table = table + ((address >> 20) & 0x7ff); + if (bad_address(table)) + goto bad; + pr_cont("S:%016lx ", *table); + if (*table & (_SEGMENT_ENTRY_INVALID | _SEGMENT_ENTRY_LARGE)) + goto out; + table = (unsigned long *)(*table & _SEGMENT_ENTRY_ORIGIN); + } + table = table + ((address >> 12) & 0xff); + if (bad_address(table)) + goto bad; + pr_cont("P:%016lx ", *table); +out: + pr_cont("\n"); + return; +bad: + pr_cont("BAD\n"); +} + +static void dump_fault_info(struct pt_regs *regs) +{ + unsigned long asce; + + pr_alert("Fault in "); + switch (regs->int_parm_long & 3) { + case 3: + pr_cont("home space "); + break; + case 2: + pr_cont("secondary space "); + break; + case 1: + pr_cont("access register "); + break; + case 0: + pr_cont("primary space "); + break; + } + pr_cont("mode while using "); + if (!user_space_fault(regs)) { + asce = S390_lowcore.kernel_asce; + pr_cont("kernel "); + } +#ifdef CONFIG_PGSTE + else if ((current->flags & PF_VCPU) && S390_lowcore.gmap) { + struct gmap *gmap = (struct gmap *)S390_lowcore.gmap; + asce = gmap->asce; + pr_cont("gmap "); + } +#endif + else { + asce = S390_lowcore.user_asce; + pr_cont("user "); + } + pr_cont("ASCE.\n"); + dump_pagetable(asce, regs->int_parm_long & __FAIL_ADDR_MASK); +} + +static inline void report_user_fault(struct pt_regs *regs, long signr) +{ + if ((task_pid_nr(current) > 1) && !show_unhandled_signals) + return; + if (!unhandled_signal(current, signr)) + return; + if (!printk_ratelimit()) + return; + printk(KERN_ALERT "User process fault: interruption code %04x ilc:%d ", + regs->int_code & 0xffff, regs->int_code >> 17); + print_vma_addr(KERN_CONT "in ", regs->psw.addr & PSW_ADDR_INSN); + printk(KERN_CONT "\n"); + printk(KERN_ALERT "failing address: %016lx TEID: %016lx\n", + regs->int_parm_long & __FAIL_ADDR_MASK, regs->int_parm_long); + dump_fault_info(regs); + show_regs(regs); +} + +/* + * Send SIGSEGV to task. This is an external routine + * to keep the stack usage of do_page_fault small. + */ +static noinline void do_sigsegv(struct pt_regs *regs, int si_code) +{ + struct siginfo si; + + report_user_fault(regs, SIGSEGV); + si.si_signo = SIGSEGV; + si.si_code = si_code; + si.si_addr = (void __user *)(regs->int_parm_long & __FAIL_ADDR_MASK); + force_sig_info(SIGSEGV, &si, current); +} + +static noinline void do_no_context(struct pt_regs *regs) +{ + const struct exception_table_entry *fixup; + unsigned long address; + + /* Are we prepared to handle this kernel fault? */ + fixup = search_exception_tables(regs->psw.addr & PSW_ADDR_INSN); + if (fixup) { + regs->psw.addr = extable_fixup(fixup) | PSW_ADDR_AMODE; + return; + } + + /* + * Oops. The kernel tried to access some bad page. We'll have to + * terminate things with extreme prejudice. + */ + address = regs->int_parm_long & __FAIL_ADDR_MASK; + if (!user_space_fault(regs)) + printk(KERN_ALERT "Unable to handle kernel pointer dereference" + " in virtual kernel address space\n"); + else + printk(KERN_ALERT "Unable to handle kernel paging request" + " in virtual user address space\n"); + printk(KERN_ALERT "failing address: %016lx TEID: %016lx\n", + regs->int_parm_long & __FAIL_ADDR_MASK, regs->int_parm_long); + dump_fault_info(regs); + die(regs, "Oops"); + do_exit(SIGKILL); +} + +static noinline void do_low_address(struct pt_regs *regs) +{ + /* Low-address protection hit in kernel mode means + NULL pointer write access in kernel mode. */ + if (regs->psw.mask & PSW_MASK_PSTATE) { + /* Low-address protection hit in user mode 'cannot happen'. */ + die (regs, "Low-address protection"); + do_exit(SIGKILL); + } + + do_no_context(regs); +} + +static noinline void do_sigbus(struct pt_regs *regs) +{ + struct task_struct *tsk = current; + struct siginfo si; + + /* + * Send a sigbus, regardless of whether we were in kernel + * or user mode. + */ + si.si_signo = SIGBUS; + si.si_errno = 0; + si.si_code = BUS_ADRERR; + si.si_addr = (void __user *)(regs->int_parm_long & __FAIL_ADDR_MASK); + force_sig_info(SIGBUS, &si, tsk); +} + +static noinline void do_fault_error(struct pt_regs *regs, int fault) +{ + int si_code; + + switch (fault) { + case VM_FAULT_BADACCESS: + case VM_FAULT_BADMAP: + /* Bad memory access. Check if it is kernel or user space. */ + if (user_mode(regs)) { + /* User mode accesses just cause a SIGSEGV */ + si_code = (fault == VM_FAULT_BADMAP) ? + SEGV_MAPERR : SEGV_ACCERR; + do_sigsegv(regs, si_code); + return; + } + case VM_FAULT_BADCONTEXT: + case VM_FAULT_PFAULT: + do_no_context(regs); + break; + case VM_FAULT_SIGNAL: + if (!user_mode(regs)) + do_no_context(regs); + break; + default: /* fault & VM_FAULT_ERROR */ + if (fault & VM_FAULT_OOM) { + if (!user_mode(regs)) + do_no_context(regs); + else + pagefault_out_of_memory(); + } else if (fault & VM_FAULT_SIGSEGV) { + /* Kernel mode? Handle exceptions or die */ + if (!user_mode(regs)) + do_no_context(regs); + else + do_sigsegv(regs, SEGV_MAPERR); + } else if (fault & VM_FAULT_SIGBUS) { + /* Kernel mode? Handle exceptions or die */ + if (!user_mode(regs)) + do_no_context(regs); + else + do_sigbus(regs); + } else + BUG(); + break; + } +} + +/* + * This routine handles page faults. It determines the address, + * and the problem, and then passes it off to one of the appropriate + * routines. + * + * interruption code (int_code): + * 04 Protection -> Write-Protection (suprression) + * 10 Segment translation -> Not present (nullification) + * 11 Page translation -> Not present (nullification) + * 3b Region third trans. -> Not present (nullification) + */ +static inline int do_exception(struct pt_regs *regs, int access) +{ +#ifdef CONFIG_PGSTE + struct gmap *gmap; +#endif + struct task_struct *tsk; + struct mm_struct *mm; + struct vm_area_struct *vma; + unsigned long trans_exc_code; + unsigned long address; + unsigned int flags; + int fault; + + tsk = current; + /* + * The instruction that caused the program check has + * been nullified. Don't signal single step via SIGTRAP. + */ + clear_pt_regs_flag(regs, PIF_PER_TRAP); + + if (notify_page_fault(regs)) + return 0; + + mm = tsk->mm; + trans_exc_code = regs->int_parm_long; + + /* + * Verify that the fault happened in user space, that + * we are not in an interrupt and that there is a + * user context. + */ + fault = VM_FAULT_BADCONTEXT; + if (unlikely(!user_space_fault(regs) || in_atomic() || !mm)) + goto out; + + address = trans_exc_code & __FAIL_ADDR_MASK; + perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address); + flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE; + if (user_mode(regs)) + flags |= FAULT_FLAG_USER; + if (access == VM_WRITE || (trans_exc_code & store_indication) == 0x400) + flags |= FAULT_FLAG_WRITE; + down_read(&mm->mmap_sem); + +#ifdef CONFIG_PGSTE + gmap = (current->flags & PF_VCPU) ? + (struct gmap *) S390_lowcore.gmap : NULL; + if (gmap) { + current->thread.gmap_addr = address; + address = __gmap_translate(gmap, address); + if (address == -EFAULT) { + fault = VM_FAULT_BADMAP; + goto out_up; + } + if (gmap->pfault_enabled) + flags |= FAULT_FLAG_RETRY_NOWAIT; + } +#endif + +retry: + fault = VM_FAULT_BADMAP; + vma = find_vma(mm, address); + if (!vma) + goto out_up; + + if (unlikely(vma->vm_start > address)) { + if (!(vma->vm_flags & VM_GROWSDOWN)) + goto out_up; + if (expand_stack(vma, address)) + goto out_up; + } + + /* + * Ok, we have a good vm_area for this memory access, so + * we can handle it.. + */ + fault = VM_FAULT_BADACCESS; + if (unlikely(!(vma->vm_flags & access))) + goto out_up; + + if (is_vm_hugetlb_page(vma)) + address &= HPAGE_MASK; + /* + * If for any reason at all we couldn't handle the fault, + * make sure we exit gracefully rather than endlessly redo + * the fault. + */ + fault = handle_mm_fault(mm, vma, address, flags); + /* No reason to continue if interrupted by SIGKILL. */ + if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) { + fault = VM_FAULT_SIGNAL; + goto out; + } + if (unlikely(fault & VM_FAULT_ERROR)) + goto out_up; + + /* + * Major/minor page fault accounting is only done on the + * initial attempt. If we go through a retry, it is extremely + * likely that the page will be found in page cache at that point. + */ + if (flags & FAULT_FLAG_ALLOW_RETRY) { + if (fault & VM_FAULT_MAJOR) { + tsk->maj_flt++; + perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, + regs, address); + } else { + tsk->min_flt++; + perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, + regs, address); + } + if (fault & VM_FAULT_RETRY) { +#ifdef CONFIG_PGSTE + if (gmap && (flags & FAULT_FLAG_RETRY_NOWAIT)) { + /* FAULT_FLAG_RETRY_NOWAIT has been set, + * mmap_sem has not been released */ + current->thread.gmap_pfault = 1; + fault = VM_FAULT_PFAULT; + goto out_up; + } +#endif + /* Clear FAULT_FLAG_ALLOW_RETRY to avoid any risk + * of starvation. */ + flags &= ~(FAULT_FLAG_ALLOW_RETRY | + FAULT_FLAG_RETRY_NOWAIT); + flags |= FAULT_FLAG_TRIED; + down_read(&mm->mmap_sem); + goto retry; + } + } +#ifdef CONFIG_PGSTE + if (gmap) { + address = __gmap_link(gmap, current->thread.gmap_addr, + address); + if (address == -EFAULT) { + fault = VM_FAULT_BADMAP; + goto out_up; + } + if (address == -ENOMEM) { + fault = VM_FAULT_OOM; + goto out_up; + } + } +#endif + fault = 0; +out_up: + up_read(&mm->mmap_sem); +out: + return fault; +} + +void do_protection_exception(struct pt_regs *regs) +{ + unsigned long trans_exc_code; + int fault; + + trans_exc_code = regs->int_parm_long; + /* + * Protection exceptions are suppressing, decrement psw address. + * The exception to this rule are aborted transactions, for these + * the PSW already points to the correct location. + */ + if (!(regs->int_code & 0x200)) + regs->psw.addr = __rewind_psw(regs->psw, regs->int_code >> 16); + /* + * Check for low-address protection. This needs to be treated + * as a special case because the translation exception code + * field is not guaranteed to contain valid data in this case. + */ + if (unlikely(!(trans_exc_code & 4))) { + do_low_address(regs); + return; + } + fault = do_exception(regs, VM_WRITE); + if (unlikely(fault)) + do_fault_error(regs, fault); +} +NOKPROBE_SYMBOL(do_protection_exception); + +void do_dat_exception(struct pt_regs *regs) +{ + int access, fault; + + access = VM_READ | VM_EXEC | VM_WRITE; + fault = do_exception(regs, access); + if (unlikely(fault)) + do_fault_error(regs, fault); +} +NOKPROBE_SYMBOL(do_dat_exception); + +#ifdef CONFIG_PFAULT +/* + * 'pfault' pseudo page faults routines. + */ +static int pfault_disable; + +static int __init nopfault(char *str) +{ + pfault_disable = 1; + return 1; +} + +__setup("nopfault", nopfault); + +struct pfault_refbk { + u16 refdiagc; + u16 reffcode; + u16 refdwlen; + u16 refversn; + u64 refgaddr; + u64 refselmk; + u64 refcmpmk; + u64 reserved; +} __attribute__ ((packed, aligned(8))); + +int pfault_init(void) +{ + struct pfault_refbk refbk = { + .refdiagc = 0x258, + .reffcode = 0, + .refdwlen = 5, + .refversn = 2, + .refgaddr = __LC_CURRENT_PID, + .refselmk = 1ULL << 48, + .refcmpmk = 1ULL << 48, + .reserved = __PF_RES_FIELD }; + int rc; + + if (pfault_disable) + return -1; + asm volatile( + " diag %1,%0,0x258\n" + "0: j 2f\n" + "1: la %0,8\n" + "2:\n" + EX_TABLE(0b,1b) + : "=d" (rc) : "a" (&refbk), "m" (refbk) : "cc"); + return rc; +} + +void pfault_fini(void) +{ + struct pfault_refbk refbk = { + .refdiagc = 0x258, + .reffcode = 1, + .refdwlen = 5, + .refversn = 2, + }; + + if (pfault_disable) + return; + asm volatile( + " diag %0,0,0x258\n" + "0:\n" + EX_TABLE(0b,0b) + : : "a" (&refbk), "m" (refbk) : "cc"); +} + +static DEFINE_SPINLOCK(pfault_lock); +static LIST_HEAD(pfault_list); + +static void pfault_interrupt(struct ext_code ext_code, + unsigned int param32, unsigned long param64) +{ + struct task_struct *tsk; + __u16 subcode; + pid_t pid; + + /* + * Get the external interruption subcode & pfault + * initial/completion signal bit. VM stores this + * in the 'cpu address' field associated with the + * external interrupt. + */ + subcode = ext_code.subcode; + if ((subcode & 0xff00) != __SUBCODE_MASK) + return; + inc_irq_stat(IRQEXT_PFL); + /* Get the token (= pid of the affected task). */ + pid = sizeof(void *) == 4 ? param32 : param64; + rcu_read_lock(); + tsk = find_task_by_pid_ns(pid, &init_pid_ns); + if (tsk) + get_task_struct(tsk); + rcu_read_unlock(); + if (!tsk) + return; + spin_lock(&pfault_lock); + if (subcode & 0x0080) { + /* signal bit is set -> a page has been swapped in by VM */ + if (tsk->thread.pfault_wait == 1) { + /* Initial interrupt was faster than the completion + * interrupt. pfault_wait is valid. Set pfault_wait + * back to zero and wake up the process. This can + * safely be done because the task is still sleeping + * and can't produce new pfaults. */ + tsk->thread.pfault_wait = 0; + list_del(&tsk->thread.list); + wake_up_process(tsk); + put_task_struct(tsk); + } else { + /* Completion interrupt was faster than initial + * interrupt. Set pfault_wait to -1 so the initial + * interrupt doesn't put the task to sleep. + * If the task is not running, ignore the completion + * interrupt since it must be a leftover of a PFAULT + * CANCEL operation which didn't remove all pending + * completion interrupts. */ + if (tsk->state == TASK_RUNNING) + tsk->thread.pfault_wait = -1; + } + } else { + /* signal bit not set -> a real page is missing. */ + if (WARN_ON_ONCE(tsk != current)) + goto out; + if (tsk->thread.pfault_wait == 1) { + /* Already on the list with a reference: put to sleep */ + __set_task_state(tsk, TASK_UNINTERRUPTIBLE); + set_tsk_need_resched(tsk); + } else if (tsk->thread.pfault_wait == -1) { + /* Completion interrupt was faster than the initial + * interrupt (pfault_wait == -1). Set pfault_wait + * back to zero and exit. */ + tsk->thread.pfault_wait = 0; + } else { + /* Initial interrupt arrived before completion + * interrupt. Let the task sleep. + * An extra task reference is needed since a different + * cpu may set the task state to TASK_RUNNING again + * before the scheduler is reached. */ + get_task_struct(tsk); + tsk->thread.pfault_wait = 1; + list_add(&tsk->thread.list, &pfault_list); + __set_task_state(tsk, TASK_UNINTERRUPTIBLE); + set_tsk_need_resched(tsk); + } + } +out: + spin_unlock(&pfault_lock); + put_task_struct(tsk); +} + +static int pfault_cpu_notify(struct notifier_block *self, unsigned long action, + void *hcpu) +{ + struct thread_struct *thread, *next; + struct task_struct *tsk; + + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_DEAD: + spin_lock_irq(&pfault_lock); + list_for_each_entry_safe(thread, next, &pfault_list, list) { + thread->pfault_wait = 0; + list_del(&thread->list); + tsk = container_of(thread, struct task_struct, thread); + wake_up_process(tsk); + put_task_struct(tsk); + } + spin_unlock_irq(&pfault_lock); + break; + default: + break; + } + return NOTIFY_OK; +} + +static int __init pfault_irq_init(void) +{ + int rc; + + rc = register_external_irq(EXT_IRQ_CP_SERVICE, pfault_interrupt); + if (rc) + goto out_extint; + rc = pfault_init() == 0 ? 0 : -EOPNOTSUPP; + if (rc) + goto out_pfault; + irq_subclass_register(IRQ_SUBCLASS_SERVICE_SIGNAL); + hotcpu_notifier(pfault_cpu_notify, 0); + return 0; + +out_pfault: + unregister_external_irq(EXT_IRQ_CP_SERVICE, pfault_interrupt); +out_extint: + pfault_disable = 1; + return rc; +} +early_initcall(pfault_irq_init); + +#endif /* CONFIG_PFAULT */ diff --git a/arch/s390/mm/gup.c b/arch/s390/mm/gup.c new file mode 100644 index 000000000..1eb41bb30 --- /dev/null +++ b/arch/s390/mm/gup.c @@ -0,0 +1,240 @@ +/* + * Lockless get_user_pages_fast for s390 + * + * Copyright IBM Corp. 2010 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/hugetlb.h> +#include <linux/vmstat.h> +#include <linux/pagemap.h> +#include <linux/rwsem.h> +#include <asm/pgtable.h> + +/* + * The performance critical leaf functions are made noinline otherwise gcc + * inlines everything into a single function which results in too much + * register pressure. + */ +static inline int gup_pte_range(pmd_t *pmdp, pmd_t pmd, unsigned long addr, + unsigned long end, int write, struct page **pages, int *nr) +{ + unsigned long mask; + pte_t *ptep, pte; + struct page *page; + + mask = (write ? _PAGE_PROTECT : 0) | _PAGE_INVALID | _PAGE_SPECIAL; + + ptep = ((pte_t *) pmd_deref(pmd)) + pte_index(addr); + do { + pte = *ptep; + barrier(); + if ((pte_val(pte) & mask) != 0) + return 0; + VM_BUG_ON(!pfn_valid(pte_pfn(pte))); + page = pte_page(pte); + if (!page_cache_get_speculative(page)) + return 0; + if (unlikely(pte_val(pte) != pte_val(*ptep))) { + put_page(page); + return 0; + } + pages[*nr] = page; + (*nr)++; + + } while (ptep++, addr += PAGE_SIZE, addr != end); + + return 1; +} + +static inline int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr, + unsigned long end, int write, struct page **pages, int *nr) +{ + unsigned long mask, result; + struct page *head, *page, *tail; + int refs; + + result = write ? 0 : _SEGMENT_ENTRY_PROTECT; + mask = result | _SEGMENT_ENTRY_INVALID; + if ((pmd_val(pmd) & mask) != result) + return 0; + VM_BUG_ON(!pfn_valid(pmd_val(pmd) >> PAGE_SHIFT)); + + refs = 0; + head = pmd_page(pmd); + page = head + ((addr & ~PMD_MASK) >> PAGE_SHIFT); + tail = page; + do { + VM_BUG_ON(compound_head(page) != head); + pages[*nr] = page; + (*nr)++; + page++; + refs++; + } while (addr += PAGE_SIZE, addr != end); + + if (!page_cache_add_speculative(head, refs)) { + *nr -= refs; + return 0; + } + + if (unlikely(pmd_val(pmd) != pmd_val(*pmdp))) { + *nr -= refs; + while (refs--) + put_page(head); + return 0; + } + + /* + * Any tail page need their mapcount reference taken before we + * return. + */ + while (refs--) { + if (PageTail(tail)) + get_huge_page_tail(tail); + tail++; + } + + return 1; +} + + +static inline int gup_pmd_range(pud_t *pudp, pud_t pud, unsigned long addr, + unsigned long end, int write, struct page **pages, int *nr) +{ + unsigned long next; + pmd_t *pmdp, pmd; + + pmdp = (pmd_t *) pudp; + if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R3) + pmdp = (pmd_t *) pud_deref(pud); + pmdp += pmd_index(addr); + do { + pmd = *pmdp; + barrier(); + next = pmd_addr_end(addr, end); + /* + * The pmd_trans_splitting() check below explains why + * pmdp_splitting_flush() has to serialize with + * smp_call_function() against our disabled IRQs, to stop + * this gup-fast code from running while we set the + * splitting bit in the pmd. Returning zero will take + * the slow path that will call wait_split_huge_page() + * if the pmd is still in splitting state. + */ + if (pmd_none(pmd) || pmd_trans_splitting(pmd)) + return 0; + if (unlikely(pmd_large(pmd))) { + if (!gup_huge_pmd(pmdp, pmd, addr, next, + write, pages, nr)) + return 0; + } else if (!gup_pte_range(pmdp, pmd, addr, next, + write, pages, nr)) + return 0; + } while (pmdp++, addr = next, addr != end); + + return 1; +} + +static inline int gup_pud_range(pgd_t *pgdp, pgd_t pgd, unsigned long addr, + unsigned long end, int write, struct page **pages, int *nr) +{ + unsigned long next; + pud_t *pudp, pud; + + pudp = (pud_t *) pgdp; + if ((pgd_val(pgd) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R2) + pudp = (pud_t *) pgd_deref(pgd); + pudp += pud_index(addr); + do { + pud = *pudp; + barrier(); + next = pud_addr_end(addr, end); + if (pud_none(pud)) + return 0; + if (!gup_pmd_range(pudp, pud, addr, next, write, pages, nr)) + return 0; + } while (pudp++, addr = next, addr != end); + + return 1; +} + +/* + * Like get_user_pages_fast() except its IRQ-safe in that it won't fall + * back to the regular GUP. + */ +int __get_user_pages_fast(unsigned long start, int nr_pages, int write, + struct page **pages) +{ + struct mm_struct *mm = current->mm; + unsigned long addr, len, end; + unsigned long next, flags; + pgd_t *pgdp, pgd; + int nr = 0; + + start &= PAGE_MASK; + addr = start; + len = (unsigned long) nr_pages << PAGE_SHIFT; + end = start + len; + if ((end <= start) || (end > TASK_SIZE)) + return 0; + /* + * local_irq_save() doesn't prevent pagetable teardown, but does + * prevent the pagetables from being freed on s390. + * + * So long as we atomically load page table pointers versus teardown, + * we can follow the address down to the the page and take a ref on it. + */ + local_irq_save(flags); + pgdp = pgd_offset(mm, addr); + do { + pgd = *pgdp; + barrier(); + next = pgd_addr_end(addr, end); + if (pgd_none(pgd)) + break; + if (!gup_pud_range(pgdp, pgd, addr, next, write, pages, &nr)) + break; + } while (pgdp++, addr = next, addr != end); + local_irq_restore(flags); + + return nr; +} + +/** + * get_user_pages_fast() - pin user pages in memory + * @start: starting user address + * @nr_pages: number of pages from start to pin + * @write: whether pages will be written to + * @pages: array that receives pointers to the pages pinned. + * Should be at least nr_pages long. + * + * Attempt to pin user pages in memory without taking mm->mmap_sem. + * If not successful, it will fall back to taking the lock and + * calling get_user_pages(). + * + * Returns number of pages pinned. This may be fewer than the number + * requested. If nr_pages is 0 or negative, returns 0. If no pages + * were pinned, returns -errno. + */ +int get_user_pages_fast(unsigned long start, int nr_pages, int write, + struct page **pages) +{ + struct mm_struct *mm = current->mm; + int nr, ret; + + start &= PAGE_MASK; + nr = __get_user_pages_fast(start, nr_pages, write, pages); + if (nr == nr_pages) + return nr; + + /* Try to get the remaining pages with get_user_pages */ + start += nr << PAGE_SHIFT; + pages += nr; + ret = get_user_pages_unlocked(current, mm, start, + nr_pages - nr, write, 0, pages); + /* Have to be a bit careful with return values */ + if (nr > 0) + ret = (ret < 0) ? nr : ret + nr; + return ret; +} diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c new file mode 100644 index 000000000..e617e74b7 --- /dev/null +++ b/arch/s390/mm/hugetlbpage.c @@ -0,0 +1,212 @@ +/* + * IBM System z Huge TLB Page Support for Kernel. + * + * Copyright IBM Corp. 2007 + * Author(s): Gerald Schaefer <gerald.schaefer@de.ibm.com> + */ + +#include <linux/mm.h> +#include <linux/hugetlb.h> + +static inline pmd_t __pte_to_pmd(pte_t pte) +{ + pmd_t pmd; + + /* + * Convert encoding pte bits pmd bits + * lIR.uswrdy.p dy..R...I...wr + * empty 010.000000.0 -> 00..0...1...00 + * prot-none, clean, old 111.000000.1 -> 00..1...1...00 + * prot-none, clean, young 111.000001.1 -> 01..1...1...00 + * prot-none, dirty, old 111.000010.1 -> 10..1...1...00 + * prot-none, dirty, young 111.000011.1 -> 11..1...1...00 + * read-only, clean, old 111.000100.1 -> 00..1...1...01 + * read-only, clean, young 101.000101.1 -> 01..1...0...01 + * read-only, dirty, old 111.000110.1 -> 10..1...1...01 + * read-only, dirty, young 101.000111.1 -> 11..1...0...01 + * read-write, clean, old 111.001100.1 -> 00..1...1...11 + * read-write, clean, young 101.001101.1 -> 01..1...0...11 + * read-write, dirty, old 110.001110.1 -> 10..0...1...11 + * read-write, dirty, young 100.001111.1 -> 11..0...0...11 + * HW-bits: R read-only, I invalid + * SW-bits: p present, y young, d dirty, r read, w write, s special, + * u unused, l large + */ + if (pte_present(pte)) { + pmd_val(pmd) = pte_val(pte) & PAGE_MASK; + pmd_val(pmd) |= (pte_val(pte) & _PAGE_READ) >> 4; + pmd_val(pmd) |= (pte_val(pte) & _PAGE_WRITE) >> 4; + pmd_val(pmd) |= (pte_val(pte) & _PAGE_INVALID) >> 5; + pmd_val(pmd) |= (pte_val(pte) & _PAGE_PROTECT); + pmd_val(pmd) |= (pte_val(pte) & _PAGE_DIRTY) << 10; + pmd_val(pmd) |= (pte_val(pte) & _PAGE_YOUNG) << 10; + } else + pmd_val(pmd) = _SEGMENT_ENTRY_INVALID; + return pmd; +} + +static inline pte_t __pmd_to_pte(pmd_t pmd) +{ + pte_t pte; + + /* + * Convert encoding pmd bits pte bits + * dy..R...I...wr lIR.uswrdy.p + * empty 00..0...1...00 -> 010.000000.0 + * prot-none, clean, old 00..1...1...00 -> 111.000000.1 + * prot-none, clean, young 01..1...1...00 -> 111.000001.1 + * prot-none, dirty, old 10..1...1...00 -> 111.000010.1 + * prot-none, dirty, young 11..1...1...00 -> 111.000011.1 + * read-only, clean, old 00..1...1...01 -> 111.000100.1 + * read-only, clean, young 01..1...0...01 -> 101.000101.1 + * read-only, dirty, old 10..1...1...01 -> 111.000110.1 + * read-only, dirty, young 11..1...0...01 -> 101.000111.1 + * read-write, clean, old 00..1...1...11 -> 111.001100.1 + * read-write, clean, young 01..1...0...11 -> 101.001101.1 + * read-write, dirty, old 10..0...1...11 -> 110.001110.1 + * read-write, dirty, young 11..0...0...11 -> 100.001111.1 + * HW-bits: R read-only, I invalid + * SW-bits: p present, y young, d dirty, r read, w write, s special, + * u unused, l large + */ + if (pmd_present(pmd)) { + pte_val(pte) = pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN_LARGE; + pte_val(pte) |= _PAGE_LARGE | _PAGE_PRESENT; + pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_READ) << 4; + pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_WRITE) << 4; + pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_INVALID) << 5; + pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_PROTECT); + pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_DIRTY) >> 10; + pte_val(pte) |= (pmd_val(pmd) & _SEGMENT_ENTRY_YOUNG) >> 10; + } else + pte_val(pte) = _PAGE_INVALID; + return pte; +} + +void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pte) +{ + pmd_t pmd; + + pmd = __pte_to_pmd(pte); + if (!MACHINE_HAS_HPAGE) { + /* Emulated huge ptes loose the dirty and young bit */ + pmd_val(pmd) &= ~_SEGMENT_ENTRY_ORIGIN; + pmd_val(pmd) |= pte_page(pte)[1].index; + } else + pmd_val(pmd) |= _SEGMENT_ENTRY_LARGE; + *(pmd_t *) ptep = pmd; +} + +pte_t huge_ptep_get(pte_t *ptep) +{ + unsigned long origin; + pmd_t pmd; + + pmd = *(pmd_t *) ptep; + if (!MACHINE_HAS_HPAGE && pmd_present(pmd)) { + origin = pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN; + pmd_val(pmd) &= ~_SEGMENT_ENTRY_ORIGIN; + pmd_val(pmd) |= *(unsigned long *) origin; + /* Emulated huge ptes are young and dirty by definition */ + pmd_val(pmd) |= _SEGMENT_ENTRY_YOUNG | _SEGMENT_ENTRY_DIRTY; + } + return __pmd_to_pte(pmd); +} + +pte_t huge_ptep_get_and_clear(struct mm_struct *mm, + unsigned long addr, pte_t *ptep) +{ + pmd_t *pmdp = (pmd_t *) ptep; + pte_t pte = huge_ptep_get(ptep); + + pmdp_flush_direct(mm, addr, pmdp); + pmd_val(*pmdp) = _SEGMENT_ENTRY_EMPTY; + return pte; +} + +int arch_prepare_hugepage(struct page *page) +{ + unsigned long addr = page_to_phys(page); + pte_t pte; + pte_t *ptep; + int i; + + if (MACHINE_HAS_HPAGE) + return 0; + + ptep = (pte_t *) pte_alloc_one(&init_mm, addr); + if (!ptep) + return -ENOMEM; + + pte_val(pte) = addr; + for (i = 0; i < PTRS_PER_PTE; i++) { + set_pte_at(&init_mm, addr + i * PAGE_SIZE, ptep + i, pte); + pte_val(pte) += PAGE_SIZE; + } + page[1].index = (unsigned long) ptep; + return 0; +} + +void arch_release_hugepage(struct page *page) +{ + pte_t *ptep; + + if (MACHINE_HAS_HPAGE) + return; + + ptep = (pte_t *) page[1].index; + if (!ptep) + return; + clear_table((unsigned long *) ptep, _PAGE_INVALID, + PTRS_PER_PTE * sizeof(pte_t)); + page_table_free(&init_mm, (unsigned long *) ptep); + page[1].index = 0; +} + +pte_t *huge_pte_alloc(struct mm_struct *mm, + unsigned long addr, unsigned long sz) +{ + pgd_t *pgdp; + pud_t *pudp; + pmd_t *pmdp = NULL; + + pgdp = pgd_offset(mm, addr); + pudp = pud_alloc(mm, pgdp, addr); + if (pudp) + pmdp = pmd_alloc(mm, pudp, addr); + return (pte_t *) pmdp; +} + +pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) +{ + pgd_t *pgdp; + pud_t *pudp; + pmd_t *pmdp = NULL; + + pgdp = pgd_offset(mm, addr); + if (pgd_present(*pgdp)) { + pudp = pud_offset(pgdp, addr); + if (pud_present(*pudp)) + pmdp = pmd_offset(pudp, addr); + } + return (pte_t *) pmdp; +} + +int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) +{ + return 0; +} + +int pmd_huge(pmd_t pmd) +{ + if (!MACHINE_HAS_HPAGE) + return 0; + + return !!(pmd_val(pmd) & _SEGMENT_ENTRY_LARGE); +} + +int pud_huge(pud_t pud) +{ + return 0; +} diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c new file mode 100644 index 000000000..80875c43a --- /dev/null +++ b/arch/s390/mm/init.c @@ -0,0 +1,230 @@ +/* + * S390 version + * Copyright IBM Corp. 1999 + * Author(s): Hartmut Penner (hp@de.ibm.com) + * + * Derived from "arch/i386/mm/init.c" + * Copyright (C) 1995 Linus Torvalds + */ + +#include <linux/signal.h> +#include <linux/sched.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/string.h> +#include <linux/types.h> +#include <linux/ptrace.h> +#include <linux/mman.h> +#include <linux/mm.h> +#include <linux/swap.h> +#include <linux/smp.h> +#include <linux/init.h> +#include <linux/pagemap.h> +#include <linux/bootmem.h> +#include <linux/memory.h> +#include <linux/pfn.h> +#include <linux/poison.h> +#include <linux/initrd.h> +#include <linux/export.h> +#include <linux/gfp.h> +#include <asm/processor.h> +#include <asm/uaccess.h> +#include <asm/pgtable.h> +#include <asm/pgalloc.h> +#include <asm/dma.h> +#include <asm/lowcore.h> +#include <asm/tlb.h> +#include <asm/tlbflush.h> +#include <asm/sections.h> +#include <asm/ctl_reg.h> +#include <asm/sclp.h> + +pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__((__aligned__(PAGE_SIZE))); + +unsigned long empty_zero_page, zero_page_mask; +EXPORT_SYMBOL(empty_zero_page); +EXPORT_SYMBOL(zero_page_mask); + +static void __init setup_zero_pages(void) +{ + struct cpuid cpu_id; + unsigned int order; + struct page *page; + int i; + + get_cpu_id(&cpu_id); + switch (cpu_id.machine) { + case 0x9672: /* g5 */ + case 0x2064: /* z900 */ + case 0x2066: /* z900 */ + case 0x2084: /* z990 */ + case 0x2086: /* z990 */ + case 0x2094: /* z9-109 */ + case 0x2096: /* z9-109 */ + order = 0; + break; + case 0x2097: /* z10 */ + case 0x2098: /* z10 */ + case 0x2817: /* z196 */ + case 0x2818: /* z196 */ + order = 2; + break; + case 0x2827: /* zEC12 */ + case 0x2828: /* zEC12 */ + order = 5; + break; + case 0x2964: /* z13 */ + default: + order = 7; + break; + } + /* Limit number of empty zero pages for small memory sizes */ + while (order > 2 && (totalram_pages >> 10) < (1UL << order)) + order--; + + empty_zero_page = __get_free_pages(GFP_KERNEL | __GFP_ZERO, order); + if (!empty_zero_page) + panic("Out of memory in setup_zero_pages"); + + page = virt_to_page((void *) empty_zero_page); + split_page(page, order); + for (i = 1 << order; i > 0; i--) { + mark_page_reserved(page); + page++; + } + + zero_page_mask = ((PAGE_SIZE << order) - 1) & PAGE_MASK; +} + +/* + * paging_init() sets up the page tables + */ +void __init paging_init(void) +{ + unsigned long max_zone_pfns[MAX_NR_ZONES]; + unsigned long pgd_type, asce_bits; + + init_mm.pgd = swapper_pg_dir; + if (VMALLOC_END > (1UL << 42)) { + asce_bits = _ASCE_TYPE_REGION2 | _ASCE_TABLE_LENGTH; + pgd_type = _REGION2_ENTRY_EMPTY; + } else { + asce_bits = _ASCE_TYPE_REGION3 | _ASCE_TABLE_LENGTH; + pgd_type = _REGION3_ENTRY_EMPTY; + } + S390_lowcore.kernel_asce = (__pa(init_mm.pgd) & PAGE_MASK) | asce_bits; + clear_table((unsigned long *) init_mm.pgd, pgd_type, + sizeof(unsigned long)*2048); + vmem_map_init(); + + /* enable virtual mapping in kernel mode */ + __ctl_load(S390_lowcore.kernel_asce, 1, 1); + __ctl_load(S390_lowcore.kernel_asce, 7, 7); + __ctl_load(S390_lowcore.kernel_asce, 13, 13); + arch_local_irq_restore(4UL << (BITS_PER_LONG - 8)); + + sparse_memory_present_with_active_regions(MAX_NUMNODES); + sparse_init(); + memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); + max_zone_pfns[ZONE_DMA] = PFN_DOWN(MAX_DMA_ADDRESS); + max_zone_pfns[ZONE_NORMAL] = max_low_pfn; + free_area_init_nodes(max_zone_pfns); +} + +void __init mem_init(void) +{ + if (MACHINE_HAS_TLB_LC) + cpumask_set_cpu(0, &init_mm.context.cpu_attach_mask); + cpumask_set_cpu(0, mm_cpumask(&init_mm)); + atomic_set(&init_mm.context.attach_count, 1); + + max_mapnr = max_low_pfn; + high_memory = (void *) __va(max_low_pfn * PAGE_SIZE); + + /* Setup guest page hinting */ + cmma_init(); + + /* this will put all low memory onto the freelists */ + free_all_bootmem(); + setup_zero_pages(); /* Setup zeroed pages. */ + + mem_init_print_info(NULL); + printk("Write protected kernel read-only data: %#lx - %#lx\n", + (unsigned long)&_stext, + PFN_ALIGN((unsigned long)&_eshared) - 1); +} + +void free_initmem(void) +{ + free_initmem_default(POISON_FREE_INITMEM); +} + +#ifdef CONFIG_BLK_DEV_INITRD +void __init free_initrd_mem(unsigned long start, unsigned long end) +{ + free_reserved_area((void *)start, (void *)end, POISON_FREE_INITMEM, + "initrd"); +} +#endif + +#ifdef CONFIG_MEMORY_HOTPLUG +int arch_add_memory(int nid, u64 start, u64 size) +{ + unsigned long zone_start_pfn, zone_end_pfn, nr_pages; + unsigned long start_pfn = PFN_DOWN(start); + unsigned long size_pages = PFN_DOWN(size); + struct zone *zone; + int rc; + + rc = vmem_add_mapping(start, size); + if (rc) + return rc; + for_each_zone(zone) { + if (zone_idx(zone) != ZONE_MOVABLE) { + /* Add range within existing zone limits */ + zone_start_pfn = zone->zone_start_pfn; + zone_end_pfn = zone->zone_start_pfn + + zone->spanned_pages; + } else { + /* Add remaining range to ZONE_MOVABLE */ + zone_start_pfn = start_pfn; + zone_end_pfn = start_pfn + size_pages; + } + if (start_pfn < zone_start_pfn || start_pfn >= zone_end_pfn) + continue; + nr_pages = (start_pfn + size_pages > zone_end_pfn) ? + zone_end_pfn - start_pfn : size_pages; + rc = __add_pages(nid, zone, start_pfn, nr_pages); + if (rc) + break; + start_pfn += nr_pages; + size_pages -= nr_pages; + if (!size_pages) + break; + } + if (rc) + vmem_remove_mapping(start, size); + return rc; +} + +unsigned long memory_block_size_bytes(void) +{ + /* + * Make sure the memory block size is always greater + * or equal than the memory increment size. + */ + return max_t(unsigned long, MIN_MEMORY_BLOCK_SIZE, sclp_get_rzm()); +} + +#ifdef CONFIG_MEMORY_HOTREMOVE +int arch_remove_memory(u64 start, u64 size) +{ + /* + * There is no hardware or firmware interface which could trigger a + * hot memory remove on s390. So there is nothing that needs to be + * implemented. + */ + return -EBUSY; +} +#endif +#endif /* CONFIG_MEMORY_HOTPLUG */ diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c new file mode 100644 index 000000000..8a993a53f --- /dev/null +++ b/arch/s390/mm/maccess.c @@ -0,0 +1,208 @@ +/* + * Access kernel memory without faulting -- s390 specific implementation. + * + * Copyright IBM Corp. 2009, 2015 + * + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>, + * + */ + +#include <linux/uaccess.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/gfp.h> +#include <linux/cpu.h> +#include <asm/ctl_reg.h> +#include <asm/io.h> + +static notrace long s390_kernel_write_odd(void *dst, const void *src, size_t size) +{ + unsigned long aligned, offset, count; + char tmp[8]; + + aligned = (unsigned long) dst & ~7UL; + offset = (unsigned long) dst & 7UL; + size = min(8UL - offset, size); + count = size - 1; + asm volatile( + " bras 1,0f\n" + " mvc 0(1,%4),0(%5)\n" + "0: mvc 0(8,%3),0(%0)\n" + " ex %1,0(1)\n" + " lg %1,0(%3)\n" + " lra %0,0(%0)\n" + " sturg %1,%0\n" + : "+&a" (aligned), "+&a" (count), "=m" (tmp) + : "a" (&tmp), "a" (&tmp[offset]), "a" (src) + : "cc", "memory", "1"); + return size; +} + +/* + * s390_kernel_write - write to kernel memory bypassing DAT + * @dst: destination address + * @src: source address + * @size: number of bytes to copy + * + * This function writes to kernel memory bypassing DAT and possible page table + * write protection. It writes to the destination using the sturg instruction. + * Therefore we have a read-modify-write sequence: the function reads eight + * bytes from destination at an eight byte boundary, modifies the bytes + * requested and writes the result back in a loop. + * + * Note: this means that this function may not be called concurrently on + * several cpus with overlapping words, since this may potentially + * cause data corruption. + */ +void notrace s390_kernel_write(void *dst, const void *src, size_t size) +{ + long copied; + + while (size) { + copied = s390_kernel_write_odd(dst, src, size); + dst += copied; + src += copied; + size -= copied; + } +} + +static int __memcpy_real(void *dest, void *src, size_t count) +{ + register unsigned long _dest asm("2") = (unsigned long) dest; + register unsigned long _len1 asm("3") = (unsigned long) count; + register unsigned long _src asm("4") = (unsigned long) src; + register unsigned long _len2 asm("5") = (unsigned long) count; + int rc = -EFAULT; + + asm volatile ( + "0: mvcle %1,%2,0x0\n" + "1: jo 0b\n" + " lhi %0,0x0\n" + "2:\n" + EX_TABLE(1b,2b) + : "+d" (rc), "+d" (_dest), "+d" (_src), "+d" (_len1), + "+d" (_len2), "=m" (*((long *) dest)) + : "m" (*((long *) src)) + : "cc", "memory"); + return rc; +} + +/* + * Copy memory in real mode (kernel to kernel) + */ +int memcpy_real(void *dest, void *src, size_t count) +{ + unsigned long flags; + int rc; + + if (!count) + return 0; + local_irq_save(flags); + __arch_local_irq_stnsm(0xfbUL); + rc = __memcpy_real(dest, src, count); + local_irq_restore(flags); + return rc; +} + +/* + * Copy memory in absolute mode (kernel to kernel) + */ +void memcpy_absolute(void *dest, void *src, size_t count) +{ + unsigned long cr0, flags, prefix; + + flags = arch_local_irq_save(); + __ctl_store(cr0, 0, 0); + __ctl_clear_bit(0, 28); /* disable lowcore protection */ + prefix = store_prefix(); + if (prefix) { + local_mcck_disable(); + set_prefix(0); + memcpy(dest, src, count); + set_prefix(prefix); + local_mcck_enable(); + } else { + memcpy(dest, src, count); + } + __ctl_load(cr0, 0, 0); + arch_local_irq_restore(flags); +} + +/* + * Copy memory from kernel (real) to user (virtual) + */ +int copy_to_user_real(void __user *dest, void *src, unsigned long count) +{ + int offs = 0, size, rc; + char *buf; + + buf = (char *) __get_free_page(GFP_KERNEL); + if (!buf) + return -ENOMEM; + rc = -EFAULT; + while (offs < count) { + size = min(PAGE_SIZE, count - offs); + if (memcpy_real(buf, src + offs, size)) + goto out; + if (copy_to_user(dest + offs, buf, size)) + goto out; + offs += size; + } + rc = 0; +out: + free_page((unsigned long) buf); + return rc; +} + +/* + * Check if physical address is within prefix or zero page + */ +static int is_swapped(unsigned long addr) +{ + unsigned long lc; + int cpu; + + if (addr < sizeof(struct _lowcore)) + return 1; + for_each_online_cpu(cpu) { + lc = (unsigned long) lowcore_ptr[cpu]; + if (addr > lc + sizeof(struct _lowcore) - 1 || addr < lc) + continue; + return 1; + } + return 0; +} + +/* + * Convert a physical pointer for /dev/mem access + * + * For swapped prefix pages a new buffer is returned that contains a copy of + * the absolute memory. The buffer size is maximum one page large. + */ +void *xlate_dev_mem_ptr(phys_addr_t addr) +{ + void *bounce = (void *) addr; + unsigned long size; + + get_online_cpus(); + preempt_disable(); + if (is_swapped(addr)) { + size = PAGE_SIZE - (addr & ~PAGE_MASK); + bounce = (void *) __get_free_page(GFP_ATOMIC); + if (bounce) + memcpy_absolute(bounce, (void *) addr, size); + } + preempt_enable(); + put_online_cpus(); + return bounce; +} + +/* + * Free converted buffer for /dev/mem access (if necessary) + */ +void unxlate_dev_mem_ptr(phys_addr_t addr, void *buf) +{ + if ((void *) addr != buf) + free_page((unsigned long) buf); +} diff --git a/arch/s390/mm/mem_detect.c b/arch/s390/mm/mem_detect.c new file mode 100644 index 000000000..0f3604395 --- /dev/null +++ b/arch/s390/mm/mem_detect.c @@ -0,0 +1,61 @@ +/* + * Copyright IBM Corp. 2008, 2009 + * + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/memblock.h> +#include <linux/init.h> +#include <linux/debugfs.h> +#include <linux/seq_file.h> +#include <asm/ipl.h> +#include <asm/sclp.h> +#include <asm/setup.h> + +#define ADDR2G (1ULL << 31) + +#define CHUNK_READ_WRITE 0 +#define CHUNK_READ_ONLY 1 + +static inline void memblock_physmem_add(phys_addr_t start, phys_addr_t size) +{ + memblock_add_range(&memblock.memory, start, size, 0, 0); + memblock_add_range(&memblock.physmem, start, size, 0, 0); +} + +void __init detect_memory_memblock(void) +{ + unsigned long long memsize, rnmax, rzm; + unsigned long addr, size; + int type; + + rzm = sclp_get_rzm(); + rnmax = sclp_get_rnmax(); + memsize = rzm * rnmax; + if (!rzm) + rzm = 1ULL << 17; + max_physmem_end = memsize; + addr = 0; + /* keep memblock lists close to the kernel */ + memblock_set_bottom_up(true); + do { + size = 0; + type = tprot(addr); + do { + size += rzm; + if (max_physmem_end && addr + size >= max_physmem_end) + break; + } while (type == tprot(addr + size)); + if (type == CHUNK_READ_WRITE || type == CHUNK_READ_ONLY) { + if (max_physmem_end && (addr + size > max_physmem_end)) + size = max_physmem_end - addr; + memblock_physmem_add(addr, size); + } + addr += size; + } while (addr < max_physmem_end); + memblock_set_bottom_up(false); + if (!max_physmem_end) + max_physmem_end = memblock_end_of_DRAM(); +} diff --git a/arch/s390/mm/mmap.c b/arch/s390/mm/mmap.c new file mode 100644 index 000000000..6e552af08 --- /dev/null +++ b/arch/s390/mm/mmap.c @@ -0,0 +1,288 @@ +/* + * flexible mmap layout support + * + * Copyright 2003-2004 Red Hat Inc., Durham, North Carolina. + * All Rights Reserved. + * + * 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 + * + * + * Started by Ingo Molnar <mingo@elte.hu> + */ + +#include <linux/personality.h> +#include <linux/mm.h> +#include <linux/mman.h> +#include <linux/module.h> +#include <linux/random.h> +#include <linux/compat.h> +#include <linux/security.h> +#include <asm/pgalloc.h> + +unsigned long mmap_rnd_mask; +static unsigned long mmap_align_mask; + +static unsigned long stack_maxrandom_size(void) +{ + if (!(current->flags & PF_RANDOMIZE)) + return 0; + if (current->personality & ADDR_NO_RANDOMIZE) + return 0; + return STACK_RND_MASK << PAGE_SHIFT; +} + +/* + * Top of mmap area (just below the process stack). + * + * Leave at least a ~32 MB hole. + */ +#define MIN_GAP (32*1024*1024) +#define MAX_GAP (STACK_TOP/6*5) + +static inline int mmap_is_legacy(void) +{ + if (current->personality & ADDR_COMPAT_LAYOUT) + return 1; + if (rlimit(RLIMIT_STACK) == RLIM_INFINITY) + return 1; + return sysctl_legacy_va_layout; +} + +unsigned long arch_mmap_rnd(void) +{ + if (is_32bit_task()) + return (get_random_int() & 0x7ff) << PAGE_SHIFT; + else + return (get_random_int() & mmap_rnd_mask) << PAGE_SHIFT; +} + +static unsigned long mmap_base_legacy(unsigned long rnd) +{ + return TASK_UNMAPPED_BASE + rnd; +} + +static inline unsigned long mmap_base(unsigned long rnd) +{ + unsigned long gap = rlimit(RLIMIT_STACK); + + if (gap < MIN_GAP) + gap = MIN_GAP; + else if (gap > MAX_GAP) + gap = MAX_GAP; + gap &= PAGE_MASK; + return STACK_TOP - stack_maxrandom_size() - rnd - gap; +} + +unsigned long +arch_get_unmapped_area(struct file *filp, unsigned long addr, + unsigned long len, unsigned long pgoff, unsigned long flags) +{ + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + struct vm_unmapped_area_info info; + int do_color_align; + + if (len > TASK_SIZE - mmap_min_addr) + return -ENOMEM; + + if (flags & MAP_FIXED) + return addr; + + if (addr) { + addr = PAGE_ALIGN(addr); + vma = find_vma(mm, addr); + if (TASK_SIZE - len >= addr && addr >= mmap_min_addr && + (!vma || addr + len <= vma->vm_start)) + return addr; + } + + do_color_align = 0; + if (filp || (flags & MAP_SHARED)) + do_color_align = !is_32bit_task(); + + info.flags = 0; + info.length = len; + info.low_limit = mm->mmap_base; + info.high_limit = TASK_SIZE; + info.align_mask = do_color_align ? (mmap_align_mask << PAGE_SHIFT) : 0; + info.align_offset = pgoff << PAGE_SHIFT; + return vm_unmapped_area(&info); +} + +unsigned long +arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, + const unsigned long len, const unsigned long pgoff, + const unsigned long flags) +{ + struct vm_area_struct *vma; + struct mm_struct *mm = current->mm; + unsigned long addr = addr0; + struct vm_unmapped_area_info info; + int do_color_align; + + /* requested length too big for entire address space */ + if (len > TASK_SIZE - mmap_min_addr) + return -ENOMEM; + + if (flags & MAP_FIXED) + return addr; + + /* requesting a specific address */ + if (addr) { + addr = PAGE_ALIGN(addr); + vma = find_vma(mm, addr); + if (TASK_SIZE - len >= addr && addr >= mmap_min_addr && + (!vma || addr + len <= vma->vm_start)) + return addr; + } + + do_color_align = 0; + if (filp || (flags & MAP_SHARED)) + do_color_align = !is_32bit_task(); + + info.flags = VM_UNMAPPED_AREA_TOPDOWN; + info.length = len; + info.low_limit = max(PAGE_SIZE, mmap_min_addr); + info.high_limit = mm->mmap_base; + info.align_mask = do_color_align ? (mmap_align_mask << PAGE_SHIFT) : 0; + info.align_offset = pgoff << PAGE_SHIFT; + addr = vm_unmapped_area(&info); + + /* + * A failed mmap() very likely causes application failure, + * so fall back to the bottom-up function here. This scenario + * can happen with large stack limits and large mmap() + * allocations. + */ + if (addr & ~PAGE_MASK) { + VM_BUG_ON(addr != -ENOMEM); + info.flags = 0; + info.low_limit = TASK_UNMAPPED_BASE; + info.high_limit = TASK_SIZE; + addr = vm_unmapped_area(&info); + } + + return addr; +} + +int s390_mmap_check(unsigned long addr, unsigned long len, unsigned long flags) +{ + if (is_compat_task() || (TASK_SIZE >= (1UL << 53))) + return 0; + if (!(flags & MAP_FIXED)) + addr = 0; + if ((addr + len) >= TASK_SIZE) + return crst_table_upgrade(current->mm, 1UL << 53); + return 0; +} + +static unsigned long +s390_get_unmapped_area(struct file *filp, unsigned long addr, + unsigned long len, unsigned long pgoff, unsigned long flags) +{ + struct mm_struct *mm = current->mm; + unsigned long area; + int rc; + + area = arch_get_unmapped_area(filp, addr, len, pgoff, flags); + if (!(area & ~PAGE_MASK)) + return area; + if (area == -ENOMEM && !is_compat_task() && TASK_SIZE < (1UL << 53)) { + /* Upgrade the page table to 4 levels and retry. */ + rc = crst_table_upgrade(mm, 1UL << 53); + if (rc) + return (unsigned long) rc; + area = arch_get_unmapped_area(filp, addr, len, pgoff, flags); + } + return area; +} + +static unsigned long +s390_get_unmapped_area_topdown(struct file *filp, const unsigned long addr, + const unsigned long len, const unsigned long pgoff, + const unsigned long flags) +{ + struct mm_struct *mm = current->mm; + unsigned long area; + int rc; + + area = arch_get_unmapped_area_topdown(filp, addr, len, pgoff, flags); + if (!(area & ~PAGE_MASK)) + return area; + if (area == -ENOMEM && !is_compat_task() && TASK_SIZE < (1UL << 53)) { + /* Upgrade the page table to 4 levels and retry. */ + rc = crst_table_upgrade(mm, 1UL << 53); + if (rc) + return (unsigned long) rc; + area = arch_get_unmapped_area_topdown(filp, addr, len, + pgoff, flags); + } + return area; +} +/* + * This function, called very early during the creation of a new + * process VM image, sets up which VM layout function to use: + */ +void arch_pick_mmap_layout(struct mm_struct *mm) +{ + unsigned long random_factor = 0UL; + + if (current->flags & PF_RANDOMIZE) + random_factor = arch_mmap_rnd(); + + /* + * Fall back to the standard layout if the personality + * bit is set, or if the expected stack growth is unlimited: + */ + if (mmap_is_legacy()) { + mm->mmap_base = mmap_base_legacy(random_factor); + mm->get_unmapped_area = s390_get_unmapped_area; + } else { + mm->mmap_base = mmap_base(random_factor); + mm->get_unmapped_area = s390_get_unmapped_area_topdown; + } +} + +static int __init setup_mmap_rnd(void) +{ + struct cpuid cpu_id; + + get_cpu_id(&cpu_id); + switch (cpu_id.machine) { + case 0x9672: + case 0x2064: + case 0x2066: + case 0x2084: + case 0x2086: + case 0x2094: + case 0x2096: + case 0x2097: + case 0x2098: + case 0x2817: + case 0x2818: + case 0x2827: + case 0x2828: + mmap_rnd_mask = 0x7ffUL; + mmap_align_mask = 0UL; + break; + case 0x2964: /* z13 */ + default: + mmap_rnd_mask = 0x3ff80UL; + mmap_align_mask = 0x7fUL; + break; + } + return 0; +} +early_initcall(setup_mmap_rnd); diff --git a/arch/s390/mm/page-states.c b/arch/s390/mm/page-states.c new file mode 100644 index 000000000..a90d45e9d --- /dev/null +++ b/arch/s390/mm/page-states.c @@ -0,0 +1,114 @@ +/* + * Copyright IBM Corp. 2008 + * + * Guest page hinting for unused pages. + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/types.h> +#include <linux/mm.h> +#include <linux/gfp.h> +#include <linux/init.h> + +#define ESSA_SET_STABLE 1 +#define ESSA_SET_UNUSED 2 + +static int cmma_flag = 1; + +static int __init cmma(char *str) +{ + char *parm; + + parm = strstrip(str); + if (strcmp(parm, "yes") == 0 || strcmp(parm, "on") == 0) { + cmma_flag = 1; + return 1; + } + cmma_flag = 0; + if (strcmp(parm, "no") == 0 || strcmp(parm, "off") == 0) + return 1; + return 0; +} +__setup("cmma=", cmma); + +void __init cmma_init(void) +{ + register unsigned long tmp asm("0") = 0; + register int rc asm("1") = -EOPNOTSUPP; + + if (!cmma_flag) + return; + asm volatile( + " .insn rrf,0xb9ab0000,%1,%1,0,0\n" + "0: la %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "+&d" (rc), "+&d" (tmp)); + if (rc) + cmma_flag = 0; +} + +static inline void set_page_unstable(struct page *page, int order) +{ + int i, rc; + + for (i = 0; i < (1 << order); i++) + asm volatile(".insn rrf,0xb9ab0000,%0,%1,%2,0" + : "=&d" (rc) + : "a" (page_to_phys(page + i)), + "i" (ESSA_SET_UNUSED)); +} + +void arch_free_page(struct page *page, int order) +{ + if (!cmma_flag) + return; + set_page_unstable(page, order); +} + +static inline void set_page_stable(struct page *page, int order) +{ + int i, rc; + + for (i = 0; i < (1 << order); i++) + asm volatile(".insn rrf,0xb9ab0000,%0,%1,%2,0" + : "=&d" (rc) + : "a" (page_to_phys(page + i)), + "i" (ESSA_SET_STABLE)); +} + +void arch_alloc_page(struct page *page, int order) +{ + if (!cmma_flag) + return; + set_page_stable(page, order); +} + +void arch_set_page_states(int make_stable) +{ + unsigned long flags, order, t; + struct list_head *l; + struct page *page; + struct zone *zone; + + if (!cmma_flag) + return; + if (make_stable) + drain_local_pages(NULL); + for_each_populated_zone(zone) { + spin_lock_irqsave(&zone->lock, flags); + for_each_migratetype_order(order, t) { + list_for_each(l, &zone->free_area[order].free_list[t]) { + page = list_entry(l, struct page, lru); + if (make_stable) + set_page_stable(page, order); + else + set_page_unstable(page, order); + } + } + spin_unlock_irqrestore(&zone->lock, flags); + } +} diff --git a/arch/s390/mm/pageattr.c b/arch/s390/mm/pageattr.c new file mode 100644 index 000000000..749c98407 --- /dev/null +++ b/arch/s390/mm/pageattr.c @@ -0,0 +1,170 @@ +/* + * Copyright IBM Corp. 2011 + * Author(s): Jan Glauber <jang@linux.vnet.ibm.com> + */ +#include <linux/hugetlb.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <asm/cacheflush.h> +#include <asm/facility.h> +#include <asm/pgtable.h> +#include <asm/page.h> + +#if PAGE_DEFAULT_KEY +static inline unsigned long sske_frame(unsigned long addr, unsigned char skey) +{ + asm volatile(".insn rrf,0xb22b0000,%[skey],%[addr],9,0" + : [addr] "+a" (addr) : [skey] "d" (skey)); + return addr; +} + +void __storage_key_init_range(unsigned long start, unsigned long end) +{ + unsigned long boundary, size; + + while (start < end) { + if (MACHINE_HAS_EDAT1) { + /* set storage keys for a 1MB frame */ + size = 1UL << 20; + boundary = (start + size) & ~(size - 1); + if (boundary <= end) { + do { + start = sske_frame(start, PAGE_DEFAULT_KEY); + } while (start < boundary); + continue; + } + } + page_set_storage_key(start, PAGE_DEFAULT_KEY, 0); + start += PAGE_SIZE; + } +} +#endif + +static pte_t *walk_page_table(unsigned long addr) +{ + pgd_t *pgdp; + pud_t *pudp; + pmd_t *pmdp; + pte_t *ptep; + + pgdp = pgd_offset_k(addr); + if (pgd_none(*pgdp)) + return NULL; + pudp = pud_offset(pgdp, addr); + if (pud_none(*pudp) || pud_large(*pudp)) + return NULL; + pmdp = pmd_offset(pudp, addr); + if (pmd_none(*pmdp) || pmd_large(*pmdp)) + return NULL; + ptep = pte_offset_kernel(pmdp, addr); + if (pte_none(*ptep)) + return NULL; + return ptep; +} + +static void change_page_attr(unsigned long addr, int numpages, + pte_t (*set) (pte_t)) +{ + pte_t *ptep, pte; + int i; + + for (i = 0; i < numpages; i++) { + ptep = walk_page_table(addr); + if (WARN_ON_ONCE(!ptep)) + break; + pte = *ptep; + pte = set(pte); + __ptep_ipte(addr, ptep); + *ptep = pte; + addr += PAGE_SIZE; + } +} + +int set_memory_ro(unsigned long addr, int numpages) +{ + change_page_attr(addr, numpages, pte_wrprotect); + return 0; +} + +int set_memory_rw(unsigned long addr, int numpages) +{ + change_page_attr(addr, numpages, pte_mkwrite); + return 0; +} + +/* not possible */ +int set_memory_nx(unsigned long addr, int numpages) +{ + return 0; +} + +int set_memory_x(unsigned long addr, int numpages) +{ + return 0; +} + +#ifdef CONFIG_DEBUG_PAGEALLOC + +static void ipte_range(pte_t *pte, unsigned long address, int nr) +{ + int i; + + if (test_facility(13)) { + __ptep_ipte_range(address, nr - 1, pte); + return; + } + for (i = 0; i < nr; i++) { + __ptep_ipte(address, pte); + address += PAGE_SIZE; + pte++; + } +} + +void __kernel_map_pages(struct page *page, int numpages, int enable) +{ + unsigned long address; + int nr, i, j; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + for (i = 0; i < numpages;) { + address = page_to_phys(page + i); + pgd = pgd_offset_k(address); + pud = pud_offset(pgd, address); + pmd = pmd_offset(pud, address); + pte = pte_offset_kernel(pmd, address); + nr = (unsigned long)pte >> ilog2(sizeof(long)); + nr = PTRS_PER_PTE - (nr & (PTRS_PER_PTE - 1)); + nr = min(numpages - i, nr); + if (enable) { + for (j = 0; j < nr; j++) { + pte_val(*pte) = __pa(address); + address += PAGE_SIZE; + pte++; + } + } else { + ipte_range(pte, address, nr); + } + i += nr; + } +} + +#ifdef CONFIG_HIBERNATION +bool kernel_page_present(struct page *page) +{ + unsigned long addr; + int cc; + + addr = page_to_phys(page); + asm volatile( + " lra %1,0(%1)\n" + " ipm %0\n" + " srl %0,28" + : "=d" (cc), "+a" (addr) : : "cc"); + return cc == 0; +} +#endif /* CONFIG_HIBERNATION */ + +#endif /* CONFIG_DEBUG_PAGEALLOC */ diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c new file mode 100644 index 000000000..b33f66110 --- /dev/null +++ b/arch/s390/mm/pgtable.c @@ -0,0 +1,1412 @@ +/* + * Copyright IBM Corp. 2007, 2011 + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + */ + +#include <linux/sched.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/gfp.h> +#include <linux/mm.h> +#include <linux/swap.h> +#include <linux/smp.h> +#include <linux/highmem.h> +#include <linux/pagemap.h> +#include <linux/spinlock.h> +#include <linux/module.h> +#include <linux/quicklist.h> +#include <linux/rcupdate.h> +#include <linux/slab.h> +#include <linux/swapops.h> +#include <linux/sysctl.h> +#include <linux/ksm.h> +#include <linux/mman.h> + +#include <asm/pgtable.h> +#include <asm/pgalloc.h> +#include <asm/tlb.h> +#include <asm/tlbflush.h> +#include <asm/mmu_context.h> + +#define ALLOC_ORDER 2 +#define FRAG_MASK 0x03 + +unsigned long *crst_table_alloc(struct mm_struct *mm) +{ + struct page *page = alloc_pages(GFP_KERNEL, ALLOC_ORDER); + + if (!page) + return NULL; + return (unsigned long *) page_to_phys(page); +} + +void crst_table_free(struct mm_struct *mm, unsigned long *table) +{ + free_pages((unsigned long) table, ALLOC_ORDER); +} + +static void __crst_table_upgrade(void *arg) +{ + struct mm_struct *mm = arg; + + if (current->active_mm == mm) { + clear_user_asce(); + set_user_asce(mm); + } + __tlb_flush_local(); +} + +int crst_table_upgrade(struct mm_struct *mm, unsigned long limit) +{ + unsigned long *table, *pgd; + unsigned long entry; + int flush; + + BUG_ON(limit > (1UL << 53)); + flush = 0; +repeat: + table = crst_table_alloc(mm); + if (!table) + return -ENOMEM; + spin_lock_bh(&mm->page_table_lock); + if (mm->context.asce_limit < limit) { + pgd = (unsigned long *) mm->pgd; + if (mm->context.asce_limit <= (1UL << 31)) { + entry = _REGION3_ENTRY_EMPTY; + mm->context.asce_limit = 1UL << 42; + mm->context.asce_bits = _ASCE_TABLE_LENGTH | + _ASCE_USER_BITS | + _ASCE_TYPE_REGION3; + } else { + entry = _REGION2_ENTRY_EMPTY; + mm->context.asce_limit = 1UL << 53; + mm->context.asce_bits = _ASCE_TABLE_LENGTH | + _ASCE_USER_BITS | + _ASCE_TYPE_REGION2; + } + crst_table_init(table, entry); + pgd_populate(mm, (pgd_t *) table, (pud_t *) pgd); + mm->pgd = (pgd_t *) table; + mm->task_size = mm->context.asce_limit; + table = NULL; + flush = 1; + } + spin_unlock_bh(&mm->page_table_lock); + if (table) + crst_table_free(mm, table); + if (mm->context.asce_limit < limit) + goto repeat; + if (flush) + on_each_cpu(__crst_table_upgrade, mm, 0); + return 0; +} + +void crst_table_downgrade(struct mm_struct *mm, unsigned long limit) +{ + pgd_t *pgd; + + if (current->active_mm == mm) { + clear_user_asce(); + __tlb_flush_mm(mm); + } + while (mm->context.asce_limit > limit) { + pgd = mm->pgd; + switch (pgd_val(*pgd) & _REGION_ENTRY_TYPE_MASK) { + case _REGION_ENTRY_TYPE_R2: + mm->context.asce_limit = 1UL << 42; + mm->context.asce_bits = _ASCE_TABLE_LENGTH | + _ASCE_USER_BITS | + _ASCE_TYPE_REGION3; + break; + case _REGION_ENTRY_TYPE_R3: + mm->context.asce_limit = 1UL << 31; + mm->context.asce_bits = _ASCE_TABLE_LENGTH | + _ASCE_USER_BITS | + _ASCE_TYPE_SEGMENT; + break; + default: + BUG(); + } + mm->pgd = (pgd_t *) (pgd_val(*pgd) & _REGION_ENTRY_ORIGIN); + mm->task_size = mm->context.asce_limit; + crst_table_free(mm, (unsigned long *) pgd); + } + if (current->active_mm == mm) + set_user_asce(mm); +} + +#ifdef CONFIG_PGSTE + +/** + * gmap_alloc - allocate a guest address space + * @mm: pointer to the parent mm_struct + * @limit: maximum size of the gmap address space + * + * Returns a guest address space structure. + */ +struct gmap *gmap_alloc(struct mm_struct *mm, unsigned long limit) +{ + struct gmap *gmap; + struct page *page; + unsigned long *table; + unsigned long etype, atype; + + if (limit < (1UL << 31)) { + limit = (1UL << 31) - 1; + atype = _ASCE_TYPE_SEGMENT; + etype = _SEGMENT_ENTRY_EMPTY; + } else if (limit < (1UL << 42)) { + limit = (1UL << 42) - 1; + atype = _ASCE_TYPE_REGION3; + etype = _REGION3_ENTRY_EMPTY; + } else if (limit < (1UL << 53)) { + limit = (1UL << 53) - 1; + atype = _ASCE_TYPE_REGION2; + etype = _REGION2_ENTRY_EMPTY; + } else { + limit = -1UL; + atype = _ASCE_TYPE_REGION1; + etype = _REGION1_ENTRY_EMPTY; + } + gmap = kzalloc(sizeof(struct gmap), GFP_KERNEL); + if (!gmap) + goto out; + INIT_LIST_HEAD(&gmap->crst_list); + INIT_RADIX_TREE(&gmap->guest_to_host, GFP_KERNEL); + INIT_RADIX_TREE(&gmap->host_to_guest, GFP_ATOMIC); + spin_lock_init(&gmap->guest_table_lock); + gmap->mm = mm; + page = alloc_pages(GFP_KERNEL, ALLOC_ORDER); + if (!page) + goto out_free; + page->index = 0; + list_add(&page->lru, &gmap->crst_list); + table = (unsigned long *) page_to_phys(page); + crst_table_init(table, etype); + gmap->table = table; + gmap->asce = atype | _ASCE_TABLE_LENGTH | + _ASCE_USER_BITS | __pa(table); + gmap->asce_end = limit; + down_write(&mm->mmap_sem); + list_add(&gmap->list, &mm->context.gmap_list); + up_write(&mm->mmap_sem); + return gmap; + +out_free: + kfree(gmap); +out: + return NULL; +} +EXPORT_SYMBOL_GPL(gmap_alloc); + +static void gmap_flush_tlb(struct gmap *gmap) +{ + if (MACHINE_HAS_IDTE) + __tlb_flush_asce(gmap->mm, gmap->asce); + else + __tlb_flush_global(); +} + +static void gmap_radix_tree_free(struct radix_tree_root *root) +{ + struct radix_tree_iter iter; + unsigned long indices[16]; + unsigned long index; + void **slot; + int i, nr; + + /* A radix tree is freed by deleting all of its entries */ + index = 0; + do { + nr = 0; + radix_tree_for_each_slot(slot, root, &iter, index) { + indices[nr] = iter.index; + if (++nr == 16) + break; + } + for (i = 0; i < nr; i++) { + index = indices[i]; + radix_tree_delete(root, index); + } + } while (nr > 0); +} + +/** + * gmap_free - free a guest address space + * @gmap: pointer to the guest address space structure + */ +void gmap_free(struct gmap *gmap) +{ + struct page *page, *next; + + /* Flush tlb. */ + if (MACHINE_HAS_IDTE) + __tlb_flush_asce(gmap->mm, gmap->asce); + else + __tlb_flush_global(); + + /* Free all segment & region tables. */ + list_for_each_entry_safe(page, next, &gmap->crst_list, lru) + __free_pages(page, ALLOC_ORDER); + gmap_radix_tree_free(&gmap->guest_to_host); + gmap_radix_tree_free(&gmap->host_to_guest); + down_write(&gmap->mm->mmap_sem); + list_del(&gmap->list); + up_write(&gmap->mm->mmap_sem); + kfree(gmap); +} +EXPORT_SYMBOL_GPL(gmap_free); + +/** + * gmap_enable - switch primary space to the guest address space + * @gmap: pointer to the guest address space structure + */ +void gmap_enable(struct gmap *gmap) +{ + S390_lowcore.gmap = (unsigned long) gmap; +} +EXPORT_SYMBOL_GPL(gmap_enable); + +/** + * gmap_disable - switch back to the standard primary address space + * @gmap: pointer to the guest address space structure + */ +void gmap_disable(struct gmap *gmap) +{ + S390_lowcore.gmap = 0UL; +} +EXPORT_SYMBOL_GPL(gmap_disable); + +/* + * gmap_alloc_table is assumed to be called with mmap_sem held + */ +static int gmap_alloc_table(struct gmap *gmap, unsigned long *table, + unsigned long init, unsigned long gaddr) +{ + struct page *page; + unsigned long *new; + + /* since we dont free the gmap table until gmap_free we can unlock */ + page = alloc_pages(GFP_KERNEL, ALLOC_ORDER); + if (!page) + return -ENOMEM; + new = (unsigned long *) page_to_phys(page); + crst_table_init(new, init); + spin_lock(&gmap->mm->page_table_lock); + if (*table & _REGION_ENTRY_INVALID) { + list_add(&page->lru, &gmap->crst_list); + *table = (unsigned long) new | _REGION_ENTRY_LENGTH | + (*table & _REGION_ENTRY_TYPE_MASK); + page->index = gaddr; + page = NULL; + } + spin_unlock(&gmap->mm->page_table_lock); + if (page) + __free_pages(page, ALLOC_ORDER); + return 0; +} + +/** + * __gmap_segment_gaddr - find virtual address from segment pointer + * @entry: pointer to a segment table entry in the guest address space + * + * Returns the virtual address in the guest address space for the segment + */ +static unsigned long __gmap_segment_gaddr(unsigned long *entry) +{ + struct page *page; + unsigned long offset, mask; + + offset = (unsigned long) entry / sizeof(unsigned long); + offset = (offset & (PTRS_PER_PMD - 1)) * PMD_SIZE; + mask = ~(PTRS_PER_PMD * sizeof(pmd_t) - 1); + page = virt_to_page((void *)((unsigned long) entry & mask)); + return page->index + offset; +} + +/** + * __gmap_unlink_by_vmaddr - unlink a single segment via a host address + * @gmap: pointer to the guest address space structure + * @vmaddr: address in the host process address space + * + * Returns 1 if a TLB flush is required + */ +static int __gmap_unlink_by_vmaddr(struct gmap *gmap, unsigned long vmaddr) +{ + unsigned long *entry; + int flush = 0; + + spin_lock(&gmap->guest_table_lock); + entry = radix_tree_delete(&gmap->host_to_guest, vmaddr >> PMD_SHIFT); + if (entry) { + flush = (*entry != _SEGMENT_ENTRY_INVALID); + *entry = _SEGMENT_ENTRY_INVALID; + } + spin_unlock(&gmap->guest_table_lock); + return flush; +} + +/** + * __gmap_unmap_by_gaddr - unmap a single segment via a guest address + * @gmap: pointer to the guest address space structure + * @gaddr: address in the guest address space + * + * Returns 1 if a TLB flush is required + */ +static int __gmap_unmap_by_gaddr(struct gmap *gmap, unsigned long gaddr) +{ + unsigned long vmaddr; + + vmaddr = (unsigned long) radix_tree_delete(&gmap->guest_to_host, + gaddr >> PMD_SHIFT); + return vmaddr ? __gmap_unlink_by_vmaddr(gmap, vmaddr) : 0; +} + +/** + * gmap_unmap_segment - unmap segment from the guest address space + * @gmap: pointer to the guest address space structure + * @to: address in the guest address space + * @len: length of the memory area to unmap + * + * Returns 0 if the unmap succeeded, -EINVAL if not. + */ +int gmap_unmap_segment(struct gmap *gmap, unsigned long to, unsigned long len) +{ + unsigned long off; + int flush; + + if ((to | len) & (PMD_SIZE - 1)) + return -EINVAL; + if (len == 0 || to + len < to) + return -EINVAL; + + flush = 0; + down_write(&gmap->mm->mmap_sem); + for (off = 0; off < len; off += PMD_SIZE) + flush |= __gmap_unmap_by_gaddr(gmap, to + off); + up_write(&gmap->mm->mmap_sem); + if (flush) + gmap_flush_tlb(gmap); + return 0; +} +EXPORT_SYMBOL_GPL(gmap_unmap_segment); + +/** + * gmap_mmap_segment - map a segment to the guest address space + * @gmap: pointer to the guest address space structure + * @from: source address in the parent address space + * @to: target address in the guest address space + * @len: length of the memory area to map + * + * Returns 0 if the mmap succeeded, -EINVAL or -ENOMEM if not. + */ +int gmap_map_segment(struct gmap *gmap, unsigned long from, + unsigned long to, unsigned long len) +{ + unsigned long off; + int flush; + + if ((from | to | len) & (PMD_SIZE - 1)) + return -EINVAL; + if (len == 0 || from + len < from || to + len < to || + from + len > TASK_MAX_SIZE || to + len > gmap->asce_end) + return -EINVAL; + + flush = 0; + down_write(&gmap->mm->mmap_sem); + for (off = 0; off < len; off += PMD_SIZE) { + /* Remove old translation */ + flush |= __gmap_unmap_by_gaddr(gmap, to + off); + /* Store new translation */ + if (radix_tree_insert(&gmap->guest_to_host, + (to + off) >> PMD_SHIFT, + (void *) from + off)) + break; + } + up_write(&gmap->mm->mmap_sem); + if (flush) + gmap_flush_tlb(gmap); + if (off >= len) + return 0; + gmap_unmap_segment(gmap, to, len); + return -ENOMEM; +} +EXPORT_SYMBOL_GPL(gmap_map_segment); + +/** + * __gmap_translate - translate a guest address to a user space address + * @gmap: pointer to guest mapping meta data structure + * @gaddr: guest address + * + * Returns user space address which corresponds to the guest address or + * -EFAULT if no such mapping exists. + * This function does not establish potentially missing page table entries. + * The mmap_sem of the mm that belongs to the address space must be held + * when this function gets called. + */ +unsigned long __gmap_translate(struct gmap *gmap, unsigned long gaddr) +{ + unsigned long vmaddr; + + vmaddr = (unsigned long) + radix_tree_lookup(&gmap->guest_to_host, gaddr >> PMD_SHIFT); + return vmaddr ? (vmaddr | (gaddr & ~PMD_MASK)) : -EFAULT; +} +EXPORT_SYMBOL_GPL(__gmap_translate); + +/** + * gmap_translate - translate a guest address to a user space address + * @gmap: pointer to guest mapping meta data structure + * @gaddr: guest address + * + * Returns user space address which corresponds to the guest address or + * -EFAULT if no such mapping exists. + * This function does not establish potentially missing page table entries. + */ +unsigned long gmap_translate(struct gmap *gmap, unsigned long gaddr) +{ + unsigned long rc; + + down_read(&gmap->mm->mmap_sem); + rc = __gmap_translate(gmap, gaddr); + up_read(&gmap->mm->mmap_sem); + return rc; +} +EXPORT_SYMBOL_GPL(gmap_translate); + +/** + * gmap_unlink - disconnect a page table from the gmap shadow tables + * @gmap: pointer to guest mapping meta data structure + * @table: pointer to the host page table + * @vmaddr: vm address associated with the host page table + */ +static void gmap_unlink(struct mm_struct *mm, unsigned long *table, + unsigned long vmaddr) +{ + struct gmap *gmap; + int flush; + + list_for_each_entry(gmap, &mm->context.gmap_list, list) { + flush = __gmap_unlink_by_vmaddr(gmap, vmaddr); + if (flush) + gmap_flush_tlb(gmap); + } +} + +/** + * gmap_link - set up shadow page tables to connect a host to a guest address + * @gmap: pointer to guest mapping meta data structure + * @gaddr: guest address + * @vmaddr: vm address + * + * Returns 0 on success, -ENOMEM for out of memory conditions, and -EFAULT + * if the vm address is already mapped to a different guest segment. + * The mmap_sem of the mm that belongs to the address space must be held + * when this function gets called. + */ +int __gmap_link(struct gmap *gmap, unsigned long gaddr, unsigned long vmaddr) +{ + struct mm_struct *mm; + unsigned long *table; + spinlock_t *ptl; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + int rc; + + /* Create higher level tables in the gmap page table */ + table = gmap->table; + if ((gmap->asce & _ASCE_TYPE_MASK) >= _ASCE_TYPE_REGION1) { + table += (gaddr >> 53) & 0x7ff; + if ((*table & _REGION_ENTRY_INVALID) && + gmap_alloc_table(gmap, table, _REGION2_ENTRY_EMPTY, + gaddr & 0xffe0000000000000UL)) + return -ENOMEM; + table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN); + } + if ((gmap->asce & _ASCE_TYPE_MASK) >= _ASCE_TYPE_REGION2) { + table += (gaddr >> 42) & 0x7ff; + if ((*table & _REGION_ENTRY_INVALID) && + gmap_alloc_table(gmap, table, _REGION3_ENTRY_EMPTY, + gaddr & 0xfffffc0000000000UL)) + return -ENOMEM; + table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN); + } + if ((gmap->asce & _ASCE_TYPE_MASK) >= _ASCE_TYPE_REGION3) { + table += (gaddr >> 31) & 0x7ff; + if ((*table & _REGION_ENTRY_INVALID) && + gmap_alloc_table(gmap, table, _SEGMENT_ENTRY_EMPTY, + gaddr & 0xffffffff80000000UL)) + return -ENOMEM; + table = (unsigned long *)(*table & _REGION_ENTRY_ORIGIN); + } + table += (gaddr >> 20) & 0x7ff; + /* Walk the parent mm page table */ + mm = gmap->mm; + pgd = pgd_offset(mm, vmaddr); + VM_BUG_ON(pgd_none(*pgd)); + pud = pud_offset(pgd, vmaddr); + VM_BUG_ON(pud_none(*pud)); + pmd = pmd_offset(pud, vmaddr); + VM_BUG_ON(pmd_none(*pmd)); + /* large pmds cannot yet be handled */ + if (pmd_large(*pmd)) + return -EFAULT; + /* Link gmap segment table entry location to page table. */ + rc = radix_tree_preload(GFP_KERNEL); + if (rc) + return rc; + ptl = pmd_lock(mm, pmd); + spin_lock(&gmap->guest_table_lock); + if (*table == _SEGMENT_ENTRY_INVALID) { + rc = radix_tree_insert(&gmap->host_to_guest, + vmaddr >> PMD_SHIFT, table); + if (!rc) + *table = pmd_val(*pmd); + } else + rc = 0; + spin_unlock(&gmap->guest_table_lock); + spin_unlock(ptl); + radix_tree_preload_end(); + return rc; +} + +/** + * gmap_fault - resolve a fault on a guest address + * @gmap: pointer to guest mapping meta data structure + * @gaddr: guest address + * @fault_flags: flags to pass down to handle_mm_fault() + * + * Returns 0 on success, -ENOMEM for out of memory conditions, and -EFAULT + * if the vm address is already mapped to a different guest segment. + */ +int gmap_fault(struct gmap *gmap, unsigned long gaddr, + unsigned int fault_flags) +{ + unsigned long vmaddr; + int rc; + + down_read(&gmap->mm->mmap_sem); + vmaddr = __gmap_translate(gmap, gaddr); + if (IS_ERR_VALUE(vmaddr)) { + rc = vmaddr; + goto out_up; + } + if (fixup_user_fault(current, gmap->mm, vmaddr, fault_flags)) { + rc = -EFAULT; + goto out_up; + } + rc = __gmap_link(gmap, gaddr, vmaddr); +out_up: + up_read(&gmap->mm->mmap_sem); + return rc; +} +EXPORT_SYMBOL_GPL(gmap_fault); + +static void gmap_zap_swap_entry(swp_entry_t entry, struct mm_struct *mm) +{ + if (!non_swap_entry(entry)) + dec_mm_counter(mm, MM_SWAPENTS); + else if (is_migration_entry(entry)) { + struct page *page = migration_entry_to_page(entry); + + if (PageAnon(page)) + dec_mm_counter(mm, MM_ANONPAGES); + else + dec_mm_counter(mm, MM_FILEPAGES); + } + free_swap_and_cache(entry); +} + +/* + * this function is assumed to be called with mmap_sem held + */ +void __gmap_zap(struct gmap *gmap, unsigned long gaddr) +{ + unsigned long vmaddr, ptev, pgstev; + pte_t *ptep, pte; + spinlock_t *ptl; + pgste_t pgste; + + /* Find the vm address for the guest address */ + vmaddr = (unsigned long) radix_tree_lookup(&gmap->guest_to_host, + gaddr >> PMD_SHIFT); + if (!vmaddr) + return; + vmaddr |= gaddr & ~PMD_MASK; + /* Get pointer to the page table entry */ + ptep = get_locked_pte(gmap->mm, vmaddr, &ptl); + if (unlikely(!ptep)) + return; + pte = *ptep; + if (!pte_swap(pte)) + goto out_pte; + /* Zap unused and logically-zero pages */ + pgste = pgste_get_lock(ptep); + pgstev = pgste_val(pgste); + ptev = pte_val(pte); + if (((pgstev & _PGSTE_GPS_USAGE_MASK) == _PGSTE_GPS_USAGE_UNUSED) || + ((pgstev & _PGSTE_GPS_ZERO) && (ptev & _PAGE_INVALID))) { + gmap_zap_swap_entry(pte_to_swp_entry(pte), gmap->mm); + pte_clear(gmap->mm, vmaddr, ptep); + } + pgste_set_unlock(ptep, pgste); +out_pte: + pte_unmap_unlock(ptep, ptl); +} +EXPORT_SYMBOL_GPL(__gmap_zap); + +void gmap_discard(struct gmap *gmap, unsigned long from, unsigned long to) +{ + unsigned long gaddr, vmaddr, size; + struct vm_area_struct *vma; + + down_read(&gmap->mm->mmap_sem); + for (gaddr = from; gaddr < to; + gaddr = (gaddr + PMD_SIZE) & PMD_MASK) { + /* Find the vm address for the guest address */ + vmaddr = (unsigned long) + radix_tree_lookup(&gmap->guest_to_host, + gaddr >> PMD_SHIFT); + if (!vmaddr) + continue; + vmaddr |= gaddr & ~PMD_MASK; + /* Find vma in the parent mm */ + vma = find_vma(gmap->mm, vmaddr); + size = min(to - gaddr, PMD_SIZE - (gaddr & ~PMD_MASK)); + zap_page_range(vma, vmaddr, size, NULL); + } + up_read(&gmap->mm->mmap_sem); +} +EXPORT_SYMBOL_GPL(gmap_discard); + +static LIST_HEAD(gmap_notifier_list); +static DEFINE_SPINLOCK(gmap_notifier_lock); + +/** + * gmap_register_ipte_notifier - register a pte invalidation callback + * @nb: pointer to the gmap notifier block + */ +void gmap_register_ipte_notifier(struct gmap_notifier *nb) +{ + spin_lock(&gmap_notifier_lock); + list_add(&nb->list, &gmap_notifier_list); + spin_unlock(&gmap_notifier_lock); +} +EXPORT_SYMBOL_GPL(gmap_register_ipte_notifier); + +/** + * gmap_unregister_ipte_notifier - remove a pte invalidation callback + * @nb: pointer to the gmap notifier block + */ +void gmap_unregister_ipte_notifier(struct gmap_notifier *nb) +{ + spin_lock(&gmap_notifier_lock); + list_del_init(&nb->list); + spin_unlock(&gmap_notifier_lock); +} +EXPORT_SYMBOL_GPL(gmap_unregister_ipte_notifier); + +/** + * gmap_ipte_notify - mark a range of ptes for invalidation notification + * @gmap: pointer to guest mapping meta data structure + * @gaddr: virtual address in the guest address space + * @len: size of area + * + * Returns 0 if for each page in the given range a gmap mapping exists and + * the invalidation notification could be set. If the gmap mapping is missing + * for one or more pages -EFAULT is returned. If no memory could be allocated + * -ENOMEM is returned. This function establishes missing page table entries. + */ +int gmap_ipte_notify(struct gmap *gmap, unsigned long gaddr, unsigned long len) +{ + unsigned long addr; + spinlock_t *ptl; + pte_t *ptep, entry; + pgste_t pgste; + int rc = 0; + + if ((gaddr & ~PAGE_MASK) || (len & ~PAGE_MASK)) + return -EINVAL; + down_read(&gmap->mm->mmap_sem); + while (len) { + /* Convert gmap address and connect the page tables */ + addr = __gmap_translate(gmap, gaddr); + if (IS_ERR_VALUE(addr)) { + rc = addr; + break; + } + /* Get the page mapped */ + if (fixup_user_fault(current, gmap->mm, addr, FAULT_FLAG_WRITE)) { + rc = -EFAULT; + break; + } + rc = __gmap_link(gmap, gaddr, addr); + if (rc) + break; + /* Walk the process page table, lock and get pte pointer */ + ptep = get_locked_pte(gmap->mm, addr, &ptl); + VM_BUG_ON(!ptep); + /* Set notification bit in the pgste of the pte */ + entry = *ptep; + if ((pte_val(entry) & (_PAGE_INVALID | _PAGE_PROTECT)) == 0) { + pgste = pgste_get_lock(ptep); + pgste_val(pgste) |= PGSTE_IN_BIT; + pgste_set_unlock(ptep, pgste); + gaddr += PAGE_SIZE; + len -= PAGE_SIZE; + } + pte_unmap_unlock(ptep, ptl); + } + up_read(&gmap->mm->mmap_sem); + return rc; +} +EXPORT_SYMBOL_GPL(gmap_ipte_notify); + +/** + * gmap_do_ipte_notify - call all invalidation callbacks for a specific pte. + * @mm: pointer to the process mm_struct + * @addr: virtual address in the process address space + * @pte: pointer to the page table entry + * + * This function is assumed to be called with the page table lock held + * for the pte to notify. + */ +void gmap_do_ipte_notify(struct mm_struct *mm, unsigned long vmaddr, pte_t *pte) +{ + unsigned long offset, gaddr; + unsigned long *table; + struct gmap_notifier *nb; + struct gmap *gmap; + + offset = ((unsigned long) pte) & (255 * sizeof(pte_t)); + offset = offset * (4096 / sizeof(pte_t)); + spin_lock(&gmap_notifier_lock); + list_for_each_entry(gmap, &mm->context.gmap_list, list) { + table = radix_tree_lookup(&gmap->host_to_guest, + vmaddr >> PMD_SHIFT); + if (!table) + continue; + gaddr = __gmap_segment_gaddr(table) + offset; + list_for_each_entry(nb, &gmap_notifier_list, list) + nb->notifier_call(gmap, gaddr); + } + spin_unlock(&gmap_notifier_lock); +} +EXPORT_SYMBOL_GPL(gmap_do_ipte_notify); + +static inline int page_table_with_pgste(struct page *page) +{ + return atomic_read(&page->_mapcount) == 0; +} + +static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm) +{ + struct page *page; + unsigned long *table; + + page = alloc_page(GFP_KERNEL|__GFP_REPEAT); + if (!page) + return NULL; + if (!pgtable_page_ctor(page)) { + __free_page(page); + return NULL; + } + atomic_set(&page->_mapcount, 0); + table = (unsigned long *) page_to_phys(page); + clear_table(table, _PAGE_INVALID, PAGE_SIZE/2); + clear_table(table + PTRS_PER_PTE, 0, PAGE_SIZE/2); + return table; +} + +static inline void page_table_free_pgste(unsigned long *table) +{ + struct page *page; + + page = pfn_to_page(__pa(table) >> PAGE_SHIFT); + pgtable_page_dtor(page); + atomic_set(&page->_mapcount, -1); + __free_page(page); +} + +int set_guest_storage_key(struct mm_struct *mm, unsigned long addr, + unsigned long key, bool nq) +{ + spinlock_t *ptl; + pgste_t old, new; + pte_t *ptep; + + down_read(&mm->mmap_sem); +retry: + ptep = get_locked_pte(mm, addr, &ptl); + if (unlikely(!ptep)) { + up_read(&mm->mmap_sem); + return -EFAULT; + } + if (!(pte_val(*ptep) & _PAGE_INVALID) && + (pte_val(*ptep) & _PAGE_PROTECT)) { + pte_unmap_unlock(ptep, ptl); + if (fixup_user_fault(current, mm, addr, FAULT_FLAG_WRITE)) { + up_read(&mm->mmap_sem); + return -EFAULT; + } + goto retry; + } + + new = old = pgste_get_lock(ptep); + pgste_val(new) &= ~(PGSTE_GR_BIT | PGSTE_GC_BIT | + PGSTE_ACC_BITS | PGSTE_FP_BIT); + pgste_val(new) |= (key & (_PAGE_CHANGED | _PAGE_REFERENCED)) << 48; + pgste_val(new) |= (key & (_PAGE_ACC_BITS | _PAGE_FP_BIT)) << 56; + if (!(pte_val(*ptep) & _PAGE_INVALID)) { + unsigned long address, bits, skey; + + address = pte_val(*ptep) & PAGE_MASK; + skey = (unsigned long) page_get_storage_key(address); + bits = skey & (_PAGE_CHANGED | _PAGE_REFERENCED); + skey = key & (_PAGE_ACC_BITS | _PAGE_FP_BIT); + /* Set storage key ACC and FP */ + page_set_storage_key(address, skey, !nq); + /* Merge host changed & referenced into pgste */ + pgste_val(new) |= bits << 52; + } + /* changing the guest storage key is considered a change of the page */ + if ((pgste_val(new) ^ pgste_val(old)) & + (PGSTE_ACC_BITS | PGSTE_FP_BIT | PGSTE_GR_BIT | PGSTE_GC_BIT)) + pgste_val(new) |= PGSTE_UC_BIT; + + pgste_set_unlock(ptep, new); + pte_unmap_unlock(ptep, ptl); + up_read(&mm->mmap_sem); + return 0; +} +EXPORT_SYMBOL(set_guest_storage_key); + +unsigned long get_guest_storage_key(struct mm_struct *mm, unsigned long addr) +{ + spinlock_t *ptl; + pgste_t pgste; + pte_t *ptep; + uint64_t physaddr; + unsigned long key = 0; + + down_read(&mm->mmap_sem); + ptep = get_locked_pte(mm, addr, &ptl); + if (unlikely(!ptep)) { + up_read(&mm->mmap_sem); + return -EFAULT; + } + pgste = pgste_get_lock(ptep); + + if (pte_val(*ptep) & _PAGE_INVALID) { + key |= (pgste_val(pgste) & PGSTE_ACC_BITS) >> 56; + key |= (pgste_val(pgste) & PGSTE_FP_BIT) >> 56; + key |= (pgste_val(pgste) & PGSTE_GR_BIT) >> 48; + key |= (pgste_val(pgste) & PGSTE_GC_BIT) >> 48; + } else { + physaddr = pte_val(*ptep) & PAGE_MASK; + key = page_get_storage_key(physaddr); + + /* Reflect guest's logical view, not physical */ + if (pgste_val(pgste) & PGSTE_GR_BIT) + key |= _PAGE_REFERENCED; + if (pgste_val(pgste) & PGSTE_GC_BIT) + key |= _PAGE_CHANGED; + } + + pgste_set_unlock(ptep, pgste); + pte_unmap_unlock(ptep, ptl); + up_read(&mm->mmap_sem); + return key; +} +EXPORT_SYMBOL(get_guest_storage_key); + +static int page_table_allocate_pgste_min = 0; +static int page_table_allocate_pgste_max = 1; +int page_table_allocate_pgste = 0; +EXPORT_SYMBOL(page_table_allocate_pgste); + +static struct ctl_table page_table_sysctl[] = { + { + .procname = "allocate_pgste", + .data = &page_table_allocate_pgste, + .maxlen = sizeof(int), + .mode = S_IRUGO | S_IWUSR, + .proc_handler = proc_dointvec, + .extra1 = &page_table_allocate_pgste_min, + .extra2 = &page_table_allocate_pgste_max, + }, + { } +}; + +static struct ctl_table page_table_sysctl_dir[] = { + { + .procname = "vm", + .maxlen = 0, + .mode = 0555, + .child = page_table_sysctl, + }, + { } +}; + +static int __init page_table_register_sysctl(void) +{ + return register_sysctl_table(page_table_sysctl_dir) ? 0 : -ENOMEM; +} +__initcall(page_table_register_sysctl); + +#else /* CONFIG_PGSTE */ + +static inline int page_table_with_pgste(struct page *page) +{ + return 0; +} + +static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm) +{ + return NULL; +} + +static inline void page_table_free_pgste(unsigned long *table) +{ +} + +static inline void gmap_unlink(struct mm_struct *mm, unsigned long *table, + unsigned long vmaddr) +{ +} + +#endif /* CONFIG_PGSTE */ + +static inline unsigned int atomic_xor_bits(atomic_t *v, unsigned int bits) +{ + unsigned int old, new; + + do { + old = atomic_read(v); + new = old ^ bits; + } while (atomic_cmpxchg(v, old, new) != old); + return new; +} + +/* + * page table entry allocation/free routines. + */ +unsigned long *page_table_alloc(struct mm_struct *mm) +{ + unsigned long *uninitialized_var(table); + struct page *uninitialized_var(page); + unsigned int mask, bit; + + if (mm_alloc_pgste(mm)) + return page_table_alloc_pgste(mm); + /* Allocate fragments of a 4K page as 1K/2K page table */ + spin_lock_bh(&mm->context.list_lock); + mask = FRAG_MASK; + if (!list_empty(&mm->context.pgtable_list)) { + page = list_first_entry(&mm->context.pgtable_list, + struct page, lru); + table = (unsigned long *) page_to_phys(page); + mask = atomic_read(&page->_mapcount); + mask = mask | (mask >> 4); + } + if ((mask & FRAG_MASK) == FRAG_MASK) { + spin_unlock_bh(&mm->context.list_lock); + page = alloc_page(GFP_KERNEL|__GFP_REPEAT); + if (!page) + return NULL; + if (!pgtable_page_ctor(page)) { + __free_page(page); + return NULL; + } + atomic_set(&page->_mapcount, 1); + table = (unsigned long *) page_to_phys(page); + clear_table(table, _PAGE_INVALID, PAGE_SIZE); + spin_lock_bh(&mm->context.list_lock); + list_add(&page->lru, &mm->context.pgtable_list); + } else { + for (bit = 1; mask & bit; bit <<= 1) + table += PTRS_PER_PTE; + mask = atomic_xor_bits(&page->_mapcount, bit); + if ((mask & FRAG_MASK) == FRAG_MASK) + list_del(&page->lru); + } + spin_unlock_bh(&mm->context.list_lock); + return table; +} + +void page_table_free(struct mm_struct *mm, unsigned long *table) +{ + struct page *page; + unsigned int bit, mask; + + page = pfn_to_page(__pa(table) >> PAGE_SHIFT); + if (page_table_with_pgste(page)) + return page_table_free_pgste(table); + /* Free 1K/2K page table fragment of a 4K page */ + bit = 1 << ((__pa(table) & ~PAGE_MASK)/(PTRS_PER_PTE*sizeof(pte_t))); + spin_lock_bh(&mm->context.list_lock); + if ((atomic_read(&page->_mapcount) & FRAG_MASK) != FRAG_MASK) + list_del(&page->lru); + mask = atomic_xor_bits(&page->_mapcount, bit); + if (mask & FRAG_MASK) + list_add(&page->lru, &mm->context.pgtable_list); + spin_unlock_bh(&mm->context.list_lock); + if (mask == 0) { + pgtable_page_dtor(page); + atomic_set(&page->_mapcount, -1); + __free_page(page); + } +} + +static void __page_table_free_rcu(void *table, unsigned bit) +{ + struct page *page; + + if (bit == FRAG_MASK) + return page_table_free_pgste(table); + /* Free 1K/2K page table fragment of a 4K page */ + page = pfn_to_page(__pa(table) >> PAGE_SHIFT); + if (atomic_xor_bits(&page->_mapcount, bit) == 0) { + pgtable_page_dtor(page); + atomic_set(&page->_mapcount, -1); + __free_page(page); + } +} + +void page_table_free_rcu(struct mmu_gather *tlb, unsigned long *table, + unsigned long vmaddr) +{ + struct mm_struct *mm; + struct page *page; + unsigned int bit, mask; + + mm = tlb->mm; + page = pfn_to_page(__pa(table) >> PAGE_SHIFT); + if (page_table_with_pgste(page)) { + gmap_unlink(mm, table, vmaddr); + table = (unsigned long *) (__pa(table) | FRAG_MASK); + tlb_remove_table(tlb, table); + return; + } + bit = 1 << ((__pa(table) & ~PAGE_MASK) / (PTRS_PER_PTE*sizeof(pte_t))); + spin_lock_bh(&mm->context.list_lock); + if ((atomic_read(&page->_mapcount) & FRAG_MASK) != FRAG_MASK) + list_del(&page->lru); + mask = atomic_xor_bits(&page->_mapcount, bit | (bit << 4)); + if (mask & FRAG_MASK) + list_add_tail(&page->lru, &mm->context.pgtable_list); + spin_unlock_bh(&mm->context.list_lock); + table = (unsigned long *) (__pa(table) | (bit << 4)); + tlb_remove_table(tlb, table); +} + +static void __tlb_remove_table(void *_table) +{ + const unsigned long mask = (FRAG_MASK << 4) | FRAG_MASK; + void *table = (void *)((unsigned long) _table & ~mask); + unsigned type = (unsigned long) _table & mask; + + if (type) + __page_table_free_rcu(table, type); + else + free_pages((unsigned long) table, ALLOC_ORDER); +} + +static void tlb_remove_table_smp_sync(void *arg) +{ + /* Simply deliver the interrupt */ +} + +static void tlb_remove_table_one(void *table) +{ + /* + * This isn't an RCU grace period and hence the page-tables cannot be + * assumed to be actually RCU-freed. + * + * It is however sufficient for software page-table walkers that rely + * on IRQ disabling. See the comment near struct mmu_table_batch. + */ + smp_call_function(tlb_remove_table_smp_sync, NULL, 1); + __tlb_remove_table(table); +} + +static void tlb_remove_table_rcu(struct rcu_head *head) +{ + struct mmu_table_batch *batch; + int i; + + batch = container_of(head, struct mmu_table_batch, rcu); + + for (i = 0; i < batch->nr; i++) + __tlb_remove_table(batch->tables[i]); + + free_page((unsigned long)batch); +} + +void tlb_table_flush(struct mmu_gather *tlb) +{ + struct mmu_table_batch **batch = &tlb->batch; + + if (*batch) { + call_rcu_sched(&(*batch)->rcu, tlb_remove_table_rcu); + *batch = NULL; + } +} + +void tlb_remove_table(struct mmu_gather *tlb, void *table) +{ + struct mmu_table_batch **batch = &tlb->batch; + + tlb->mm->context.flush_mm = 1; + if (*batch == NULL) { + *batch = (struct mmu_table_batch *) + __get_free_page(GFP_NOWAIT | __GFP_NOWARN); + if (*batch == NULL) { + __tlb_flush_mm_lazy(tlb->mm); + tlb_remove_table_one(table); + return; + } + (*batch)->nr = 0; + } + (*batch)->tables[(*batch)->nr++] = table; + if ((*batch)->nr == MAX_TABLE_BATCH) + tlb_flush_mmu(tlb); +} + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static inline void thp_split_vma(struct vm_area_struct *vma) +{ + unsigned long addr; + + for (addr = vma->vm_start; addr < vma->vm_end; addr += PAGE_SIZE) + follow_page(vma, addr, FOLL_SPLIT); +} + +static inline void thp_split_mm(struct mm_struct *mm) +{ + struct vm_area_struct *vma; + + for (vma = mm->mmap; vma != NULL; vma = vma->vm_next) { + thp_split_vma(vma); + vma->vm_flags &= ~VM_HUGEPAGE; + vma->vm_flags |= VM_NOHUGEPAGE; + } + mm->def_flags |= VM_NOHUGEPAGE; +} +#else +static inline void thp_split_mm(struct mm_struct *mm) +{ +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ + +/* + * switch on pgstes for its userspace process (for kvm) + */ +int s390_enable_sie(void) +{ + struct mm_struct *mm = current->mm; + + /* Do we have pgstes? if yes, we are done */ + if (mm_has_pgste(mm)) + return 0; + /* Fail if the page tables are 2K */ + if (!mm_alloc_pgste(mm)) + return -EINVAL; + down_write(&mm->mmap_sem); + mm->context.has_pgste = 1; + /* split thp mappings and disable thp for future mappings */ + thp_split_mm(mm); + up_write(&mm->mmap_sem); + return 0; +} +EXPORT_SYMBOL_GPL(s390_enable_sie); + +/* + * Enable storage key handling from now on and initialize the storage + * keys with the default key. + */ +static int __s390_enable_skey(pte_t *pte, unsigned long addr, + unsigned long next, struct mm_walk *walk) +{ + unsigned long ptev; + pgste_t pgste; + + pgste = pgste_get_lock(pte); + /* + * Remove all zero page mappings, + * after establishing a policy to forbid zero page mappings + * following faults for that page will get fresh anonymous pages + */ + if (is_zero_pfn(pte_pfn(*pte))) { + ptep_flush_direct(walk->mm, addr, pte); + pte_val(*pte) = _PAGE_INVALID; + } + /* Clear storage key */ + pgste_val(pgste) &= ~(PGSTE_ACC_BITS | PGSTE_FP_BIT | + PGSTE_GR_BIT | PGSTE_GC_BIT); + ptev = pte_val(*pte); + if (!(ptev & _PAGE_INVALID) && (ptev & _PAGE_WRITE)) + page_set_storage_key(ptev & PAGE_MASK, PAGE_DEFAULT_KEY, 1); + pgste_set_unlock(pte, pgste); + return 0; +} + +int s390_enable_skey(void) +{ + struct mm_walk walk = { .pte_entry = __s390_enable_skey }; + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + int rc = 0; + + down_write(&mm->mmap_sem); + if (mm_use_skey(mm)) + goto out_up; + + mm->context.use_skey = 1; + for (vma = mm->mmap; vma; vma = vma->vm_next) { + if (ksm_madvise(vma, vma->vm_start, vma->vm_end, + MADV_UNMERGEABLE, &vma->vm_flags)) { + mm->context.use_skey = 0; + rc = -ENOMEM; + goto out_up; + } + } + mm->def_flags &= ~VM_MERGEABLE; + + walk.mm = mm; + walk_page_range(0, TASK_SIZE, &walk); + +out_up: + up_write(&mm->mmap_sem); + return rc; +} +EXPORT_SYMBOL_GPL(s390_enable_skey); + +/* + * Reset CMMA state, make all pages stable again. + */ +static int __s390_reset_cmma(pte_t *pte, unsigned long addr, + unsigned long next, struct mm_walk *walk) +{ + pgste_t pgste; + + pgste = pgste_get_lock(pte); + pgste_val(pgste) &= ~_PGSTE_GPS_USAGE_MASK; + pgste_set_unlock(pte, pgste); + return 0; +} + +void s390_reset_cmma(struct mm_struct *mm) +{ + struct mm_walk walk = { .pte_entry = __s390_reset_cmma }; + + down_write(&mm->mmap_sem); + walk.mm = mm; + walk_page_range(0, TASK_SIZE, &walk); + up_write(&mm->mmap_sem); +} +EXPORT_SYMBOL_GPL(s390_reset_cmma); + +/* + * Test and reset if a guest page is dirty + */ +bool gmap_test_and_clear_dirty(unsigned long address, struct gmap *gmap) +{ + pte_t *pte; + spinlock_t *ptl; + bool dirty = false; + + pte = get_locked_pte(gmap->mm, address, &ptl); + if (unlikely(!pte)) + return false; + + if (ptep_test_and_clear_user_dirty(gmap->mm, address, pte)) + dirty = true; + + spin_unlock(ptl); + return dirty; +} +EXPORT_SYMBOL_GPL(gmap_test_and_clear_dirty); + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +int pmdp_clear_flush_young(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp) +{ + VM_BUG_ON(address & ~HPAGE_PMD_MASK); + /* No need to flush TLB + * On s390 reference bits are in storage key and never in TLB */ + return pmdp_test_and_clear_young(vma, address, pmdp); +} + +int pmdp_set_access_flags(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp, + pmd_t entry, int dirty) +{ + VM_BUG_ON(address & ~HPAGE_PMD_MASK); + + entry = pmd_mkyoung(entry); + if (dirty) + entry = pmd_mkdirty(entry); + if (pmd_same(*pmdp, entry)) + return 0; + pmdp_invalidate(vma, address, pmdp); + set_pmd_at(vma->vm_mm, address, pmdp, entry); + return 1; +} + +static void pmdp_splitting_flush_sync(void *arg) +{ + /* Simply deliver the interrupt */ +} + +void pmdp_splitting_flush(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp) +{ + VM_BUG_ON(address & ~HPAGE_PMD_MASK); + if (!test_and_set_bit(_SEGMENT_ENTRY_SPLIT_BIT, + (unsigned long *) pmdp)) { + /* need to serialize against gup-fast (IRQ disabled) */ + smp_call_function(pmdp_splitting_flush_sync, NULL, 1); + } +} + +void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp, + pgtable_t pgtable) +{ + struct list_head *lh = (struct list_head *) pgtable; + + assert_spin_locked(pmd_lockptr(mm, pmdp)); + + /* FIFO */ + if (!pmd_huge_pte(mm, pmdp)) + INIT_LIST_HEAD(lh); + else + list_add(lh, (struct list_head *) pmd_huge_pte(mm, pmdp)); + pmd_huge_pte(mm, pmdp) = pgtable; +} + +pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp) +{ + struct list_head *lh; + pgtable_t pgtable; + pte_t *ptep; + + assert_spin_locked(pmd_lockptr(mm, pmdp)); + + /* FIFO */ + pgtable = pmd_huge_pte(mm, pmdp); + lh = (struct list_head *) pgtable; + if (list_empty(lh)) + pmd_huge_pte(mm, pmdp) = NULL; + else { + pmd_huge_pte(mm, pmdp) = (pgtable_t) lh->next; + list_del(lh); + } + ptep = (pte_t *) pgtable; + pte_val(*ptep) = _PAGE_INVALID; + ptep++; + pte_val(*ptep) = _PAGE_INVALID; + return pgtable; +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c new file mode 100644 index 000000000..ef7d6c8fe --- /dev/null +++ b/arch/s390/mm/vmem.c @@ -0,0 +1,416 @@ +/* + * Copyright IBM Corp. 2006 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#include <linux/bootmem.h> +#include <linux/pfn.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/list.h> +#include <linux/hugetlb.h> +#include <linux/slab.h> +#include <linux/memblock.h> +#include <asm/pgalloc.h> +#include <asm/pgtable.h> +#include <asm/setup.h> +#include <asm/tlbflush.h> +#include <asm/sections.h> + +static DEFINE_MUTEX(vmem_mutex); + +struct memory_segment { + struct list_head list; + unsigned long start; + unsigned long size; +}; + +static LIST_HEAD(mem_segs); + +static void __ref *vmem_alloc_pages(unsigned int order) +{ + if (slab_is_available()) + return (void *)__get_free_pages(GFP_KERNEL, order); + return alloc_bootmem_pages((1 << order) * PAGE_SIZE); +} + +static inline pud_t *vmem_pud_alloc(void) +{ + pud_t *pud = NULL; + + pud = vmem_alloc_pages(2); + if (!pud) + return NULL; + clear_table((unsigned long *) pud, _REGION3_ENTRY_EMPTY, PAGE_SIZE * 4); + return pud; +} + +static inline pmd_t *vmem_pmd_alloc(void) +{ + pmd_t *pmd = NULL; + + pmd = vmem_alloc_pages(2); + if (!pmd) + return NULL; + clear_table((unsigned long *) pmd, _SEGMENT_ENTRY_EMPTY, PAGE_SIZE * 4); + return pmd; +} + +static pte_t __ref *vmem_pte_alloc(unsigned long address) +{ + pte_t *pte; + + if (slab_is_available()) + pte = (pte_t *) page_table_alloc(&init_mm); + else + pte = alloc_bootmem_align(PTRS_PER_PTE * sizeof(pte_t), + PTRS_PER_PTE * sizeof(pte_t)); + if (!pte) + return NULL; + clear_table((unsigned long *) pte, _PAGE_INVALID, + PTRS_PER_PTE * sizeof(pte_t)); + return pte; +} + +/* + * Add a physical memory range to the 1:1 mapping. + */ +static int vmem_add_mem(unsigned long start, unsigned long size, int ro) +{ + unsigned long end = start + size; + unsigned long address = start; + pgd_t *pg_dir; + pud_t *pu_dir; + pmd_t *pm_dir; + pte_t *pt_dir; + int ret = -ENOMEM; + + while (address < end) { + pg_dir = pgd_offset_k(address); + if (pgd_none(*pg_dir)) { + pu_dir = vmem_pud_alloc(); + if (!pu_dir) + goto out; + pgd_populate(&init_mm, pg_dir, pu_dir); + } + pu_dir = pud_offset(pg_dir, address); +#ifndef CONFIG_DEBUG_PAGEALLOC + if (MACHINE_HAS_EDAT2 && pud_none(*pu_dir) && address && + !(address & ~PUD_MASK) && (address + PUD_SIZE <= end)) { + pud_val(*pu_dir) = __pa(address) | + _REGION_ENTRY_TYPE_R3 | _REGION3_ENTRY_LARGE | + (ro ? _REGION_ENTRY_PROTECT : 0); + address += PUD_SIZE; + continue; + } +#endif + if (pud_none(*pu_dir)) { + pm_dir = vmem_pmd_alloc(); + if (!pm_dir) + goto out; + pud_populate(&init_mm, pu_dir, pm_dir); + } + pm_dir = pmd_offset(pu_dir, address); +#ifndef CONFIG_DEBUG_PAGEALLOC + if (MACHINE_HAS_EDAT1 && pmd_none(*pm_dir) && address && + !(address & ~PMD_MASK) && (address + PMD_SIZE <= end)) { + pmd_val(*pm_dir) = __pa(address) | + _SEGMENT_ENTRY | _SEGMENT_ENTRY_LARGE | + _SEGMENT_ENTRY_YOUNG | + (ro ? _SEGMENT_ENTRY_PROTECT : 0); + address += PMD_SIZE; + continue; + } +#endif + if (pmd_none(*pm_dir)) { + pt_dir = vmem_pte_alloc(address); + if (!pt_dir) + goto out; + pmd_populate(&init_mm, pm_dir, pt_dir); + } + + pt_dir = pte_offset_kernel(pm_dir, address); + pte_val(*pt_dir) = __pa(address) | + pgprot_val(ro ? PAGE_KERNEL_RO : PAGE_KERNEL); + address += PAGE_SIZE; + } + ret = 0; +out: + return ret; +} + +/* + * Remove a physical memory range from the 1:1 mapping. + * Currently only invalidates page table entries. + */ +static void vmem_remove_range(unsigned long start, unsigned long size) +{ + unsigned long end = start + size; + unsigned long address = start; + pgd_t *pg_dir; + pud_t *pu_dir; + pmd_t *pm_dir; + pte_t *pt_dir; + pte_t pte; + + pte_val(pte) = _PAGE_INVALID; + while (address < end) { + pg_dir = pgd_offset_k(address); + if (pgd_none(*pg_dir)) { + address += PGDIR_SIZE; + continue; + } + pu_dir = pud_offset(pg_dir, address); + if (pud_none(*pu_dir)) { + address += PUD_SIZE; + continue; + } + if (pud_large(*pu_dir)) { + pud_clear(pu_dir); + address += PUD_SIZE; + continue; + } + pm_dir = pmd_offset(pu_dir, address); + if (pmd_none(*pm_dir)) { + address += PMD_SIZE; + continue; + } + if (pmd_large(*pm_dir)) { + pmd_clear(pm_dir); + address += PMD_SIZE; + continue; + } + pt_dir = pte_offset_kernel(pm_dir, address); + *pt_dir = pte; + address += PAGE_SIZE; + } + flush_tlb_kernel_range(start, end); +} + +/* + * Add a backed mem_map array to the virtual mem_map array. + */ +int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node) +{ + unsigned long address = start; + pgd_t *pg_dir; + pud_t *pu_dir; + pmd_t *pm_dir; + pte_t *pt_dir; + int ret = -ENOMEM; + + for (address = start; address < end;) { + pg_dir = pgd_offset_k(address); + if (pgd_none(*pg_dir)) { + pu_dir = vmem_pud_alloc(); + if (!pu_dir) + goto out; + pgd_populate(&init_mm, pg_dir, pu_dir); + } + + pu_dir = pud_offset(pg_dir, address); + if (pud_none(*pu_dir)) { + pm_dir = vmem_pmd_alloc(); + if (!pm_dir) + goto out; + pud_populate(&init_mm, pu_dir, pm_dir); + } + + pm_dir = pmd_offset(pu_dir, address); + if (pmd_none(*pm_dir)) { + /* Use 1MB frames for vmemmap if available. We always + * use large frames even if they are only partially + * used. + * Otherwise we would have also page tables since + * vmemmap_populate gets called for each section + * separately. */ + if (MACHINE_HAS_EDAT1) { + void *new_page; + + new_page = vmemmap_alloc_block(PMD_SIZE, node); + if (!new_page) + goto out; + pmd_val(*pm_dir) = __pa(new_page) | + _SEGMENT_ENTRY | _SEGMENT_ENTRY_LARGE; + address = (address + PMD_SIZE) & PMD_MASK; + continue; + } + pt_dir = vmem_pte_alloc(address); + if (!pt_dir) + goto out; + pmd_populate(&init_mm, pm_dir, pt_dir); + } else if (pmd_large(*pm_dir)) { + address = (address + PMD_SIZE) & PMD_MASK; + continue; + } + + pt_dir = pte_offset_kernel(pm_dir, address); + if (pte_none(*pt_dir)) { + void *new_page; + + new_page = vmemmap_alloc_block(PAGE_SIZE, node); + if (!new_page) + goto out; + pte_val(*pt_dir) = + __pa(new_page) | pgprot_val(PAGE_KERNEL); + } + address += PAGE_SIZE; + } + ret = 0; +out: + return ret; +} + +void vmemmap_free(unsigned long start, unsigned long end) +{ +} + +/* + * Add memory segment to the segment list if it doesn't overlap with + * an already present segment. + */ +static int insert_memory_segment(struct memory_segment *seg) +{ + struct memory_segment *tmp; + + if (seg->start + seg->size > VMEM_MAX_PHYS || + seg->start + seg->size < seg->start) + return -ERANGE; + + list_for_each_entry(tmp, &mem_segs, list) { + if (seg->start >= tmp->start + tmp->size) + continue; + if (seg->start + seg->size <= tmp->start) + continue; + return -ENOSPC; + } + list_add(&seg->list, &mem_segs); + return 0; +} + +/* + * Remove memory segment from the segment list. + */ +static void remove_memory_segment(struct memory_segment *seg) +{ + list_del(&seg->list); +} + +static void __remove_shared_memory(struct memory_segment *seg) +{ + remove_memory_segment(seg); + vmem_remove_range(seg->start, seg->size); +} + +int vmem_remove_mapping(unsigned long start, unsigned long size) +{ + struct memory_segment *seg; + int ret; + + mutex_lock(&vmem_mutex); + + ret = -ENOENT; + list_for_each_entry(seg, &mem_segs, list) { + if (seg->start == start && seg->size == size) + break; + } + + if (seg->start != start || seg->size != size) + goto out; + + ret = 0; + __remove_shared_memory(seg); + kfree(seg); +out: + mutex_unlock(&vmem_mutex); + return ret; +} + +int vmem_add_mapping(unsigned long start, unsigned long size) +{ + struct memory_segment *seg; + int ret; + + mutex_lock(&vmem_mutex); + ret = -ENOMEM; + seg = kzalloc(sizeof(*seg), GFP_KERNEL); + if (!seg) + goto out; + seg->start = start; + seg->size = size; + + ret = insert_memory_segment(seg); + if (ret) + goto out_free; + + ret = vmem_add_mem(start, size, 0); + if (ret) + goto out_remove; + goto out; + +out_remove: + __remove_shared_memory(seg); +out_free: + kfree(seg); +out: + mutex_unlock(&vmem_mutex); + return ret; +} + +/* + * map whole physical memory to virtual memory (identity mapping) + * we reserve enough space in the vmalloc area for vmemmap to hotplug + * additional memory segments. + */ +void __init vmem_map_init(void) +{ + unsigned long ro_start, ro_end; + struct memblock_region *reg; + phys_addr_t start, end; + + ro_start = PFN_ALIGN((unsigned long)&_stext); + ro_end = (unsigned long)&_eshared & PAGE_MASK; + for_each_memblock(memory, reg) { + start = reg->base; + end = reg->base + reg->size - 1; + if (start >= ro_end || end <= ro_start) + vmem_add_mem(start, end - start, 0); + else if (start >= ro_start && end <= ro_end) + vmem_add_mem(start, end - start, 1); + else if (start >= ro_start) { + vmem_add_mem(start, ro_end - start, 1); + vmem_add_mem(ro_end, end - ro_end, 0); + } else if (end < ro_end) { + vmem_add_mem(start, ro_start - start, 0); + vmem_add_mem(ro_start, end - ro_start, 1); + } else { + vmem_add_mem(start, ro_start - start, 0); + vmem_add_mem(ro_start, ro_end - ro_start, 1); + vmem_add_mem(ro_end, end - ro_end, 0); + } + } +} + +/* + * Convert memblock.memory to a memory segment list so there is a single + * list that contains all memory segments. + */ +static int __init vmem_convert_memory_chunk(void) +{ + struct memblock_region *reg; + struct memory_segment *seg; + + mutex_lock(&vmem_mutex); + for_each_memblock(memory, reg) { + seg = kzalloc(sizeof(*seg), GFP_KERNEL); + if (!seg) + panic("Out of memory...\n"); + seg->start = reg->base; + seg->size = reg->size; + insert_memory_segment(seg); + } + mutex_unlock(&vmem_mutex); + return 0; +} + +core_initcall(vmem_convert_memory_chunk); diff --git a/arch/s390/net/Makefile b/arch/s390/net/Makefile new file mode 100644 index 000000000..90568c33d --- /dev/null +++ b/arch/s390/net/Makefile @@ -0,0 +1,4 @@ +# +# Arch-specific network modules +# +obj-$(CONFIG_BPF_JIT) += bpf_jit.o bpf_jit_comp.o diff --git a/arch/s390/net/bpf_jit.S b/arch/s390/net/bpf_jit.S new file mode 100644 index 000000000..a1c917d88 --- /dev/null +++ b/arch/s390/net/bpf_jit.S @@ -0,0 +1,115 @@ +/* + * BPF Jit compiler for s390, help functions. + * + * Copyright IBM Corp. 2012,2015 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + * Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#include <linux/linkage.h> +#include "bpf_jit.h" + +/* + * Calling convention: + * registers %r7-%r10, %r11,%r13, and %r15 are call saved + * + * Input (64 bit): + * %r3 (%b2) = offset into skb data + * %r6 (%b5) = return address + * %r7 (%b6) = skb pointer + * %r12 = skb data pointer + * + * Output: + * %r14= %b0 = return value (read skb value) + * + * Work registers: %r2,%r4,%r5,%r14 + * + * skb_copy_bits takes 4 parameters: + * %r2 = skb pointer + * %r3 = offset into skb data + * %r4 = pointer to temp buffer + * %r5 = length to copy + * Return value in %r2: 0 = ok + * + * bpf_internal_load_pointer_neg_helper takes 3 parameters: + * %r2 = skb pointer + * %r3 = offset into data + * %r4 = length to copy + * Return value in %r2: Pointer to data + */ + +#define SKF_MAX_NEG_OFF -0x200000 /* SKF_LL_OFF from filter.h */ + +/* + * Load SIZE bytes from SKB + */ +#define sk_load_common(NAME, SIZE, LOAD) \ +ENTRY(sk_load_##NAME); \ + ltgr %r3,%r3; /* Is offset negative? */ \ + jl sk_load_##NAME##_slow_neg; \ +ENTRY(sk_load_##NAME##_pos); \ + aghi %r3,SIZE; /* Offset + SIZE */ \ + clg %r3,STK_OFF_HLEN(%r15); /* Offset + SIZE > hlen? */ \ + jh sk_load_##NAME##_slow; \ + LOAD %r14,-SIZE(%r3,%r12); /* Get data from skb */ \ + b OFF_OK(%r6); /* Return */ \ + \ +sk_load_##NAME##_slow:; \ + lgr %r2,%r7; /* Arg1 = skb pointer */ \ + aghi %r3,-SIZE; /* Arg2 = offset */ \ + la %r4,STK_OFF_TMP(%r15); /* Arg3 = temp bufffer */ \ + lghi %r5,SIZE; /* Arg4 = size */ \ + brasl %r14,skb_copy_bits; /* Get data from skb */ \ + LOAD %r14,STK_OFF_TMP(%r15); /* Load from temp bufffer */ \ + ltgr %r2,%r2; /* Set cc to (%r2 != 0) */ \ + br %r6; /* Return */ + +sk_load_common(word, 4, llgf) /* r14 = *(u32 *) (skb->data+offset) */ +sk_load_common(half, 2, llgh) /* r14 = *(u16 *) (skb->data+offset) */ + +/* + * Load 1 byte from SKB (optimized version) + */ + /* r14 = *(u8 *) (skb->data+offset) */ +ENTRY(sk_load_byte) + ltgr %r3,%r3 # Is offset negative? + jl sk_load_byte_slow_neg +ENTRY(sk_load_byte_pos) + clg %r3,STK_OFF_HLEN(%r15) # Offset >= hlen? + jnl sk_load_byte_slow + llgc %r14,0(%r3,%r12) # Get byte from skb + b OFF_OK(%r6) # Return OK + +sk_load_byte_slow: + lgr %r2,%r7 # Arg1 = skb pointer + # Arg2 = offset + la %r4,STK_OFF_TMP(%r15) # Arg3 = pointer to temp buffer + lghi %r5,1 # Arg4 = size (1 byte) + brasl %r14,skb_copy_bits # Get data from skb + llgc %r14,STK_OFF_TMP(%r15) # Load result from temp buffer + ltgr %r2,%r2 # Set cc to (%r2 != 0) + br %r6 # Return cc + +#define sk_negative_common(NAME, SIZE, LOAD) \ +sk_load_##NAME##_slow_neg:; \ + cgfi %r3,SKF_MAX_NEG_OFF; \ + jl bpf_error; \ + lgr %r2,%r7; /* Arg1 = skb pointer */ \ + /* Arg2 = offset */ \ + lghi %r4,SIZE; /* Arg3 = size */ \ + brasl %r14,bpf_internal_load_pointer_neg_helper; \ + ltgr %r2,%r2; \ + jz bpf_error; \ + LOAD %r14,0(%r2); /* Get data from pointer */ \ + xr %r3,%r3; /* Set cc to zero */ \ + br %r6; /* Return cc */ + +sk_negative_common(word, 4, llgf) +sk_negative_common(half, 2, llgh) +sk_negative_common(byte, 1, llgc) + +bpf_error: +# force a return 0 from jit handler + ltgr %r15,%r15 # Set condition code + br %r6 diff --git a/arch/s390/net/bpf_jit.h b/arch/s390/net/bpf_jit.h new file mode 100644 index 000000000..de156ba3b --- /dev/null +++ b/arch/s390/net/bpf_jit.h @@ -0,0 +1,60 @@ +/* + * BPF Jit compiler defines + * + * Copyright IBM Corp. 2012,2015 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + * Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#ifndef __ARCH_S390_NET_BPF_JIT_H +#define __ARCH_S390_NET_BPF_JIT_H + +#ifndef __ASSEMBLY__ + +#include <linux/filter.h> +#include <linux/types.h> + +extern u8 sk_load_word_pos[], sk_load_half_pos[], sk_load_byte_pos[]; +extern u8 sk_load_word[], sk_load_half[], sk_load_byte[]; + +#endif /* __ASSEMBLY__ */ + +/* + * Stackframe layout (packed stack): + * + * ^ high + * +---------------+ | + * | old backchain | | + * +---------------+ | + * | r15 - r6 | | + * BFP -> +===============+ | + * | | | + * | BPF stack | | + * | | | + * +---------------+ | + * | 8 byte hlen | | + * R15+168 -> +---------------+ | + * | 4 byte align | | + * +---------------+ | + * | 4 byte temp | | + * | for bpf_jit.S | | + * R15+160 -> +---------------+ | + * | new backchain | | + * R15+152 -> +---------------+ | + * | + 152 byte SA | | + * R15 -> +---------------+ + low + * + * We get 160 bytes stack space from calling function, but only use + * 11 * 8 byte (old backchain + r15 - r6) for storing registers. + */ +#define STK_SPACE (MAX_BPF_STACK + 8 + 4 + 4 + 160) +#define STK_160_UNUSED (160 - 11 * 8) +#define STK_OFF (STK_SPACE - STK_160_UNUSED) +#define STK_OFF_TMP 160 /* Offset of tmp buffer on stack */ +#define STK_OFF_HLEN 168 /* Offset of SKB header length on stack */ + +/* Offset to skip condition code check */ +#define OFF_OK 4 + +#endif /* __ARCH_S390_NET_BPF_JIT_H */ diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c new file mode 100644 index 000000000..9afb9d602 --- /dev/null +++ b/arch/s390/net/bpf_jit_comp.c @@ -0,0 +1,1224 @@ +/* + * BPF Jit compiler for s390. + * + * Minimum build requirements: + * + * - HAVE_MARCH_Z196_FEATURES: laal, laalg + * - HAVE_MARCH_Z10_FEATURES: msfi, cgrj, clgrj + * - HAVE_MARCH_Z9_109_FEATURES: alfi, llilf, clfi, oilf, nilf + * - PACK_STACK + * - 64BIT + * + * Copyright IBM Corp. 2012,2015 + * + * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com> + * Michael Holzheu <holzheu@linux.vnet.ibm.com> + */ + +#define KMSG_COMPONENT "bpf_jit" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/netdevice.h> +#include <linux/filter.h> +#include <linux/init.h> +#include <asm/cacheflush.h> +#include <asm/dis.h> +#include "bpf_jit.h" + +int bpf_jit_enable __read_mostly; + +struct bpf_jit { + u32 seen; /* Flags to remember seen eBPF instructions */ + u32 seen_reg[16]; /* Array to remember which registers are used */ + u32 *addrs; /* Array with relative instruction addresses */ + u8 *prg_buf; /* Start of program */ + int size; /* Size of program and literal pool */ + int size_prg; /* Size of program */ + int prg; /* Current position in program */ + int lit_start; /* Start of literal pool */ + int lit; /* Current position in literal pool */ + int base_ip; /* Base address for literal pool */ + int ret0_ip; /* Address of return 0 */ + int exit_ip; /* Address of exit */ +}; + +#define BPF_SIZE_MAX 4096 /* Max size for program */ + +#define SEEN_SKB 1 /* skb access */ +#define SEEN_MEM 2 /* use mem[] for temporary storage */ +#define SEEN_RET0 4 /* ret0_ip points to a valid return 0 */ +#define SEEN_LITERAL 8 /* code uses literals */ +#define SEEN_FUNC 16 /* calls C functions */ +#define SEEN_STACK (SEEN_FUNC | SEEN_MEM | SEEN_SKB) + +/* + * s390 registers + */ +#define REG_W0 (__MAX_BPF_REG+0) /* Work register 1 (even) */ +#define REG_W1 (__MAX_BPF_REG+1) /* Work register 2 (odd) */ +#define REG_SKB_DATA (__MAX_BPF_REG+2) /* SKB data register */ +#define REG_L (__MAX_BPF_REG+3) /* Literal pool register */ +#define REG_15 (__MAX_BPF_REG+4) /* Register 15 */ +#define REG_0 REG_W0 /* Register 0 */ +#define REG_2 BPF_REG_1 /* Register 2 */ +#define REG_14 BPF_REG_0 /* Register 14 */ + +/* + * Mapping of BPF registers to s390 registers + */ +static const int reg2hex[] = { + /* Return code */ + [BPF_REG_0] = 14, + /* Function parameters */ + [BPF_REG_1] = 2, + [BPF_REG_2] = 3, + [BPF_REG_3] = 4, + [BPF_REG_4] = 5, + [BPF_REG_5] = 6, + /* Call saved registers */ + [BPF_REG_6] = 7, + [BPF_REG_7] = 8, + [BPF_REG_8] = 9, + [BPF_REG_9] = 10, + /* BPF stack pointer */ + [BPF_REG_FP] = 13, + /* SKB data pointer */ + [REG_SKB_DATA] = 12, + /* Work registers for s390x backend */ + [REG_W0] = 0, + [REG_W1] = 1, + [REG_L] = 11, + [REG_15] = 15, +}; + +static inline u32 reg(u32 dst_reg, u32 src_reg) +{ + return reg2hex[dst_reg] << 4 | reg2hex[src_reg]; +} + +static inline u32 reg_high(u32 reg) +{ + return reg2hex[reg] << 4; +} + +static inline void reg_set_seen(struct bpf_jit *jit, u32 b1) +{ + u32 r1 = reg2hex[b1]; + + if (!jit->seen_reg[r1] && r1 >= 6 && r1 <= 15) + jit->seen_reg[r1] = 1; +} + +#define REG_SET_SEEN(b1) \ +({ \ + reg_set_seen(jit, b1); \ +}) + +#define REG_SEEN(b1) jit->seen_reg[reg2hex[(b1)]] + +/* + * EMIT macros for code generation + */ + +#define _EMIT2(op) \ +({ \ + if (jit->prg_buf) \ + *(u16 *) (jit->prg_buf + jit->prg) = op; \ + jit->prg += 2; \ +}) + +#define EMIT2(op, b1, b2) \ +({ \ + _EMIT2(op | reg(b1, b2)); \ + REG_SET_SEEN(b1); \ + REG_SET_SEEN(b2); \ +}) + +#define _EMIT4(op) \ +({ \ + if (jit->prg_buf) \ + *(u32 *) (jit->prg_buf + jit->prg) = op; \ + jit->prg += 4; \ +}) + +#define EMIT4(op, b1, b2) \ +({ \ + _EMIT4(op | reg(b1, b2)); \ + REG_SET_SEEN(b1); \ + REG_SET_SEEN(b2); \ +}) + +#define EMIT4_RRF(op, b1, b2, b3) \ +({ \ + _EMIT4(op | reg_high(b3) << 8 | reg(b1, b2)); \ + REG_SET_SEEN(b1); \ + REG_SET_SEEN(b2); \ + REG_SET_SEEN(b3); \ +}) + +#define _EMIT4_DISP(op, disp) \ +({ \ + unsigned int __disp = (disp) & 0xfff; \ + _EMIT4(op | __disp); \ +}) + +#define EMIT4_DISP(op, b1, b2, disp) \ +({ \ + _EMIT4_DISP(op | reg_high(b1) << 16 | \ + reg_high(b2) << 8, disp); \ + REG_SET_SEEN(b1); \ + REG_SET_SEEN(b2); \ +}) + +#define EMIT4_IMM(op, b1, imm) \ +({ \ + unsigned int __imm = (imm) & 0xffff; \ + _EMIT4(op | reg_high(b1) << 16 | __imm); \ + REG_SET_SEEN(b1); \ +}) + +#define EMIT4_PCREL(op, pcrel) \ +({ \ + long __pcrel = ((pcrel) >> 1) & 0xffff; \ + _EMIT4(op | __pcrel); \ +}) + +#define _EMIT6(op1, op2) \ +({ \ + if (jit->prg_buf) { \ + *(u32 *) (jit->prg_buf + jit->prg) = op1; \ + *(u16 *) (jit->prg_buf + jit->prg + 4) = op2; \ + } \ + jit->prg += 6; \ +}) + +#define _EMIT6_DISP(op1, op2, disp) \ +({ \ + unsigned int __disp = (disp) & 0xfff; \ + _EMIT6(op1 | __disp, op2); \ +}) + +#define EMIT6_DISP(op1, op2, b1, b2, b3, disp) \ +({ \ + _EMIT6_DISP(op1 | reg(b1, b2) << 16 | \ + reg_high(b3) << 8, op2, disp); \ + REG_SET_SEEN(b1); \ + REG_SET_SEEN(b2); \ + REG_SET_SEEN(b3); \ +}) + +#define _EMIT6_DISP_LH(op1, op2, disp) \ +({ \ + unsigned int __disp_h = ((u32)disp) & 0xff000; \ + unsigned int __disp_l = ((u32)disp) & 0x00fff; \ + _EMIT6(op1 | __disp_l, op2 | __disp_h >> 4); \ +}) + +#define EMIT6_DISP_LH(op1, op2, b1, b2, b3, disp) \ +({ \ + _EMIT6_DISP_LH(op1 | reg(b1, b2) << 16 | \ + reg_high(b3) << 8, op2, disp); \ + REG_SET_SEEN(b1); \ + REG_SET_SEEN(b2); \ + REG_SET_SEEN(b3); \ +}) + +#define EMIT6_PCREL(op1, op2, b1, b2, i, off, mask) \ +({ \ + /* Branch instruction needs 6 bytes */ \ + int rel = (addrs[i + off + 1] - (addrs[i + 1] - 6)) / 2;\ + _EMIT6(op1 | reg(b1, b2) << 16 | (rel & 0xffff), op2 | mask); \ + REG_SET_SEEN(b1); \ + REG_SET_SEEN(b2); \ +}) + +#define _EMIT6_IMM(op, imm) \ +({ \ + unsigned int __imm = (imm); \ + _EMIT6(op | (__imm >> 16), __imm & 0xffff); \ +}) + +#define EMIT6_IMM(op, b1, imm) \ +({ \ + _EMIT6_IMM(op | reg_high(b1) << 16, imm); \ + REG_SET_SEEN(b1); \ +}) + +#define EMIT_CONST_U32(val) \ +({ \ + unsigned int ret; \ + ret = jit->lit - jit->base_ip; \ + jit->seen |= SEEN_LITERAL; \ + if (jit->prg_buf) \ + *(u32 *) (jit->prg_buf + jit->lit) = (u32) val; \ + jit->lit += 4; \ + ret; \ +}) + +#define EMIT_CONST_U64(val) \ +({ \ + unsigned int ret; \ + ret = jit->lit - jit->base_ip; \ + jit->seen |= SEEN_LITERAL; \ + if (jit->prg_buf) \ + *(u64 *) (jit->prg_buf + jit->lit) = (u64) val; \ + jit->lit += 8; \ + ret; \ +}) + +#define EMIT_ZERO(b1) \ +({ \ + /* llgfr %dst,%dst (zero extend to 64 bit) */ \ + EMIT4(0xb9160000, b1, b1); \ + REG_SET_SEEN(b1); \ +}) + +/* + * Fill whole space with illegal instructions + */ +static void jit_fill_hole(void *area, unsigned int size) +{ + memset(area, 0, size); +} + +/* + * Save registers from "rs" (register start) to "re" (register end) on stack + */ +static void save_regs(struct bpf_jit *jit, u32 rs, u32 re) +{ + u32 off = 72 + (rs - 6) * 8; + + if (rs == re) + /* stg %rs,off(%r15) */ + _EMIT6(0xe300f000 | rs << 20 | off, 0x0024); + else + /* stmg %rs,%re,off(%r15) */ + _EMIT6_DISP(0xeb00f000 | rs << 20 | re << 16, 0x0024, off); +} + +/* + * Restore registers from "rs" (register start) to "re" (register end) on stack + */ +static void restore_regs(struct bpf_jit *jit, u32 rs, u32 re) +{ + u32 off = 72 + (rs - 6) * 8; + + if (jit->seen & SEEN_STACK) + off += STK_OFF; + + if (rs == re) + /* lg %rs,off(%r15) */ + _EMIT6(0xe300f000 | rs << 20 | off, 0x0004); + else + /* lmg %rs,%re,off(%r15) */ + _EMIT6_DISP(0xeb00f000 | rs << 20 | re << 16, 0x0004, off); +} + +/* + * Return first seen register (from start) + */ +static int get_start(struct bpf_jit *jit, int start) +{ + int i; + + for (i = start; i <= 15; i++) { + if (jit->seen_reg[i]) + return i; + } + return 0; +} + +/* + * Return last seen register (from start) (gap >= 2) + */ +static int get_end(struct bpf_jit *jit, int start) +{ + int i; + + for (i = start; i < 15; i++) { + if (!jit->seen_reg[i] && !jit->seen_reg[i + 1]) + return i - 1; + } + return jit->seen_reg[15] ? 15 : 14; +} + +#define REGS_SAVE 1 +#define REGS_RESTORE 0 +/* + * Save and restore clobbered registers (6-15) on stack. + * We save/restore registers in chunks with gap >= 2 registers. + */ +static void save_restore_regs(struct bpf_jit *jit, int op) +{ + + int re = 6, rs; + + do { + rs = get_start(jit, re); + if (!rs) + break; + re = get_end(jit, rs + 1); + if (op == REGS_SAVE) + save_regs(jit, rs, re); + else + restore_regs(jit, rs, re); + re++; + } while (re <= 15); +} + +/* + * Emit function prologue + * + * Save registers and create stack frame if necessary. + * See stack frame layout desription in "bpf_jit.h"! + */ +static void bpf_jit_prologue(struct bpf_jit *jit) +{ + /* Save registers */ + save_restore_regs(jit, REGS_SAVE); + /* Setup literal pool */ + if (jit->seen & SEEN_LITERAL) { + /* basr %r13,0 */ + EMIT2(0x0d00, REG_L, REG_0); + jit->base_ip = jit->prg; + } + /* Setup stack and backchain */ + if (jit->seen & SEEN_STACK) { + if (jit->seen & SEEN_FUNC) + /* lgr %w1,%r15 (backchain) */ + EMIT4(0xb9040000, REG_W1, REG_15); + /* la %bfp,STK_160_UNUSED(%r15) (BPF frame pointer) */ + EMIT4_DISP(0x41000000, BPF_REG_FP, REG_15, STK_160_UNUSED); + /* aghi %r15,-STK_OFF */ + EMIT4_IMM(0xa70b0000, REG_15, -STK_OFF); + if (jit->seen & SEEN_FUNC) + /* stg %w1,152(%r15) (backchain) */ + EMIT6_DISP_LH(0xe3000000, 0x0024, REG_W1, REG_0, + REG_15, 152); + } + /* + * For SKB access %b1 contains the SKB pointer. For "bpf_jit.S" + * we store the SKB header length on the stack and the SKB data + * pointer in REG_SKB_DATA. + */ + if (jit->seen & SEEN_SKB) { + /* Header length: llgf %w1,<len>(%b1) */ + EMIT6_DISP_LH(0xe3000000, 0x0016, REG_W1, REG_0, BPF_REG_1, + offsetof(struct sk_buff, len)); + /* s %w1,<data_len>(%b1) */ + EMIT4_DISP(0x5b000000, REG_W1, BPF_REG_1, + offsetof(struct sk_buff, data_len)); + /* stg %w1,ST_OFF_HLEN(%r0,%r15) */ + EMIT6_DISP_LH(0xe3000000, 0x0024, REG_W1, REG_0, REG_15, + STK_OFF_HLEN); + /* lg %skb_data,data_off(%b1) */ + EMIT6_DISP_LH(0xe3000000, 0x0004, REG_SKB_DATA, REG_0, + BPF_REG_1, offsetof(struct sk_buff, data)); + } + /* BPF compatibility: clear A (%b7) and X (%b8) registers */ + if (REG_SEEN(BPF_REG_7)) + /* lghi %b7,0 */ + EMIT4_IMM(0xa7090000, BPF_REG_7, 0); + if (REG_SEEN(BPF_REG_8)) + /* lghi %b8,0 */ + EMIT4_IMM(0xa7090000, BPF_REG_8, 0); +} + +/* + * Function epilogue + */ +static void bpf_jit_epilogue(struct bpf_jit *jit) +{ + /* Return 0 */ + if (jit->seen & SEEN_RET0) { + jit->ret0_ip = jit->prg; + /* lghi %b0,0 */ + EMIT4_IMM(0xa7090000, BPF_REG_0, 0); + } + jit->exit_ip = jit->prg; + /* Load exit code: lgr %r2,%b0 */ + EMIT4(0xb9040000, REG_2, BPF_REG_0); + /* Restore registers */ + save_restore_regs(jit, REGS_RESTORE); + /* br %r14 */ + _EMIT2(0x07fe); +} + +/* + * Compile one eBPF instruction into s390x code + * + * NOTE: Use noinline because for gcov (-fprofile-arcs) gcc allocates a lot of + * stack space for the large switch statement. + */ +static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i) +{ + struct bpf_insn *insn = &fp->insnsi[i]; + int jmp_off, last, insn_count = 1; + unsigned int func_addr, mask; + u32 dst_reg = insn->dst_reg; + u32 src_reg = insn->src_reg; + u32 *addrs = jit->addrs; + s32 imm = insn->imm; + s16 off = insn->off; + + switch (insn->code) { + /* + * BPF_MOV + */ + case BPF_ALU | BPF_MOV | BPF_X: /* dst = (u32) src */ + /* llgfr %dst,%src */ + EMIT4(0xb9160000, dst_reg, src_reg); + break; + case BPF_ALU64 | BPF_MOV | BPF_X: /* dst = src */ + /* lgr %dst,%src */ + EMIT4(0xb9040000, dst_reg, src_reg); + break; + case BPF_ALU | BPF_MOV | BPF_K: /* dst = (u32) imm */ + /* llilf %dst,imm */ + EMIT6_IMM(0xc00f0000, dst_reg, imm); + break; + case BPF_ALU64 | BPF_MOV | BPF_K: /* dst = imm */ + /* lgfi %dst,imm */ + EMIT6_IMM(0xc0010000, dst_reg, imm); + break; + /* + * BPF_LD 64 + */ + case BPF_LD | BPF_IMM | BPF_DW: /* dst = (u64) imm */ + { + /* 16 byte instruction that uses two 'struct bpf_insn' */ + u64 imm64; + + imm64 = (u64)(u32) insn[0].imm | ((u64)(u32) insn[1].imm) << 32; + /* lg %dst,<d(imm)>(%l) */ + EMIT6_DISP_LH(0xe3000000, 0x0004, dst_reg, REG_0, REG_L, + EMIT_CONST_U64(imm64)); + insn_count = 2; + break; + } + /* + * BPF_ADD + */ + case BPF_ALU | BPF_ADD | BPF_X: /* dst = (u32) dst + (u32) src */ + /* ar %dst,%src */ + EMIT2(0x1a00, dst_reg, src_reg); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_ADD | BPF_X: /* dst = dst + src */ + /* agr %dst,%src */ + EMIT4(0xb9080000, dst_reg, src_reg); + break; + case BPF_ALU | BPF_ADD | BPF_K: /* dst = (u32) dst + (u32) imm */ + if (!imm) + break; + /* alfi %dst,imm */ + EMIT6_IMM(0xc20b0000, dst_reg, imm); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_ADD | BPF_K: /* dst = dst + imm */ + if (!imm) + break; + /* agfi %dst,imm */ + EMIT6_IMM(0xc2080000, dst_reg, imm); + break; + /* + * BPF_SUB + */ + case BPF_ALU | BPF_SUB | BPF_X: /* dst = (u32) dst - (u32) src */ + /* sr %dst,%src */ + EMIT2(0x1b00, dst_reg, src_reg); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_SUB | BPF_X: /* dst = dst - src */ + /* sgr %dst,%src */ + EMIT4(0xb9090000, dst_reg, src_reg); + break; + case BPF_ALU | BPF_SUB | BPF_K: /* dst = (u32) dst - (u32) imm */ + if (!imm) + break; + /* alfi %dst,-imm */ + EMIT6_IMM(0xc20b0000, dst_reg, -imm); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_SUB | BPF_K: /* dst = dst - imm */ + if (!imm) + break; + /* agfi %dst,-imm */ + EMIT6_IMM(0xc2080000, dst_reg, -imm); + break; + /* + * BPF_MUL + */ + case BPF_ALU | BPF_MUL | BPF_X: /* dst = (u32) dst * (u32) src */ + /* msr %dst,%src */ + EMIT4(0xb2520000, dst_reg, src_reg); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_MUL | BPF_X: /* dst = dst * src */ + /* msgr %dst,%src */ + EMIT4(0xb90c0000, dst_reg, src_reg); + break; + case BPF_ALU | BPF_MUL | BPF_K: /* dst = (u32) dst * (u32) imm */ + if (imm == 1) + break; + /* msfi %r5,imm */ + EMIT6_IMM(0xc2010000, dst_reg, imm); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_MUL | BPF_K: /* dst = dst * imm */ + if (imm == 1) + break; + /* msgfi %dst,imm */ + EMIT6_IMM(0xc2000000, dst_reg, imm); + break; + /* + * BPF_DIV / BPF_MOD + */ + case BPF_ALU | BPF_DIV | BPF_X: /* dst = (u32) dst / (u32) src */ + case BPF_ALU | BPF_MOD | BPF_X: /* dst = (u32) dst % (u32) src */ + { + int rc_reg = BPF_OP(insn->code) == BPF_DIV ? REG_W1 : REG_W0; + + jit->seen |= SEEN_RET0; + /* ltr %src,%src (if src == 0 goto fail) */ + EMIT2(0x1200, src_reg, src_reg); + /* jz <ret0> */ + EMIT4_PCREL(0xa7840000, jit->ret0_ip - jit->prg); + /* lhi %w0,0 */ + EMIT4_IMM(0xa7080000, REG_W0, 0); + /* lr %w1,%dst */ + EMIT2(0x1800, REG_W1, dst_reg); + /* dlr %w0,%src */ + EMIT4(0xb9970000, REG_W0, src_reg); + /* llgfr %dst,%rc */ + EMIT4(0xb9160000, dst_reg, rc_reg); + break; + } + case BPF_ALU64 | BPF_DIV | BPF_X: /* dst = dst / src */ + case BPF_ALU64 | BPF_MOD | BPF_X: /* dst = dst % src */ + { + int rc_reg = BPF_OP(insn->code) == BPF_DIV ? REG_W1 : REG_W0; + + jit->seen |= SEEN_RET0; + /* ltgr %src,%src (if src == 0 goto fail) */ + EMIT4(0xb9020000, src_reg, src_reg); + /* jz <ret0> */ + EMIT4_PCREL(0xa7840000, jit->ret0_ip - jit->prg); + /* lghi %w0,0 */ + EMIT4_IMM(0xa7090000, REG_W0, 0); + /* lgr %w1,%dst */ + EMIT4(0xb9040000, REG_W1, dst_reg); + /* dlgr %w0,%dst */ + EMIT4(0xb9870000, REG_W0, src_reg); + /* lgr %dst,%rc */ + EMIT4(0xb9040000, dst_reg, rc_reg); + break; + } + case BPF_ALU | BPF_DIV | BPF_K: /* dst = (u32) dst / (u32) imm */ + case BPF_ALU | BPF_MOD | BPF_K: /* dst = (u32) dst % (u32) imm */ + { + int rc_reg = BPF_OP(insn->code) == BPF_DIV ? REG_W1 : REG_W0; + + if (imm == 1) { + if (BPF_OP(insn->code) == BPF_MOD) + /* lhgi %dst,0 */ + EMIT4_IMM(0xa7090000, dst_reg, 0); + break; + } + /* lhi %w0,0 */ + EMIT4_IMM(0xa7080000, REG_W0, 0); + /* lr %w1,%dst */ + EMIT2(0x1800, REG_W1, dst_reg); + /* dl %w0,<d(imm)>(%l) */ + EMIT6_DISP_LH(0xe3000000, 0x0097, REG_W0, REG_0, REG_L, + EMIT_CONST_U32(imm)); + /* llgfr %dst,%rc */ + EMIT4(0xb9160000, dst_reg, rc_reg); + break; + } + case BPF_ALU64 | BPF_DIV | BPF_K: /* dst = dst / imm */ + case BPF_ALU64 | BPF_MOD | BPF_K: /* dst = dst % imm */ + { + int rc_reg = BPF_OP(insn->code) == BPF_DIV ? REG_W1 : REG_W0; + + if (imm == 1) { + if (BPF_OP(insn->code) == BPF_MOD) + /* lhgi %dst,0 */ + EMIT4_IMM(0xa7090000, dst_reg, 0); + break; + } + /* lghi %w0,0 */ + EMIT4_IMM(0xa7090000, REG_W0, 0); + /* lgr %w1,%dst */ + EMIT4(0xb9040000, REG_W1, dst_reg); + /* dlg %w0,<d(imm)>(%l) */ + EMIT6_DISP_LH(0xe3000000, 0x0087, REG_W0, REG_0, REG_L, + EMIT_CONST_U64(imm)); + /* lgr %dst,%rc */ + EMIT4(0xb9040000, dst_reg, rc_reg); + break; + } + /* + * BPF_AND + */ + case BPF_ALU | BPF_AND | BPF_X: /* dst = (u32) dst & (u32) src */ + /* nr %dst,%src */ + EMIT2(0x1400, dst_reg, src_reg); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_AND | BPF_X: /* dst = dst & src */ + /* ngr %dst,%src */ + EMIT4(0xb9800000, dst_reg, src_reg); + break; + case BPF_ALU | BPF_AND | BPF_K: /* dst = (u32) dst & (u32) imm */ + /* nilf %dst,imm */ + EMIT6_IMM(0xc00b0000, dst_reg, imm); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_AND | BPF_K: /* dst = dst & imm */ + /* ng %dst,<d(imm)>(%l) */ + EMIT6_DISP_LH(0xe3000000, 0x0080, dst_reg, REG_0, REG_L, + EMIT_CONST_U64(imm)); + break; + /* + * BPF_OR + */ + case BPF_ALU | BPF_OR | BPF_X: /* dst = (u32) dst | (u32) src */ + /* or %dst,%src */ + EMIT2(0x1600, dst_reg, src_reg); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_OR | BPF_X: /* dst = dst | src */ + /* ogr %dst,%src */ + EMIT4(0xb9810000, dst_reg, src_reg); + break; + case BPF_ALU | BPF_OR | BPF_K: /* dst = (u32) dst | (u32) imm */ + /* oilf %dst,imm */ + EMIT6_IMM(0xc00d0000, dst_reg, imm); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_OR | BPF_K: /* dst = dst | imm */ + /* og %dst,<d(imm)>(%l) */ + EMIT6_DISP_LH(0xe3000000, 0x0081, dst_reg, REG_0, REG_L, + EMIT_CONST_U64(imm)); + break; + /* + * BPF_XOR + */ + case BPF_ALU | BPF_XOR | BPF_X: /* dst = (u32) dst ^ (u32) src */ + /* xr %dst,%src */ + EMIT2(0x1700, dst_reg, src_reg); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_XOR | BPF_X: /* dst = dst ^ src */ + /* xgr %dst,%src */ + EMIT4(0xb9820000, dst_reg, src_reg); + break; + case BPF_ALU | BPF_XOR | BPF_K: /* dst = (u32) dst ^ (u32) imm */ + if (!imm) + break; + /* xilf %dst,imm */ + EMIT6_IMM(0xc0070000, dst_reg, imm); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_XOR | BPF_K: /* dst = dst ^ imm */ + /* xg %dst,<d(imm)>(%l) */ + EMIT6_DISP_LH(0xe3000000, 0x0082, dst_reg, REG_0, REG_L, + EMIT_CONST_U64(imm)); + break; + /* + * BPF_LSH + */ + case BPF_ALU | BPF_LSH | BPF_X: /* dst = (u32) dst << (u32) src */ + /* sll %dst,0(%src) */ + EMIT4_DISP(0x89000000, dst_reg, src_reg, 0); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_LSH | BPF_X: /* dst = dst << src */ + /* sllg %dst,%dst,0(%src) */ + EMIT6_DISP_LH(0xeb000000, 0x000d, dst_reg, dst_reg, src_reg, 0); + break; + case BPF_ALU | BPF_LSH | BPF_K: /* dst = (u32) dst << (u32) imm */ + if (imm == 0) + break; + /* sll %dst,imm(%r0) */ + EMIT4_DISP(0x89000000, dst_reg, REG_0, imm); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_LSH | BPF_K: /* dst = dst << imm */ + if (imm == 0) + break; + /* sllg %dst,%dst,imm(%r0) */ + EMIT6_DISP_LH(0xeb000000, 0x000d, dst_reg, dst_reg, REG_0, imm); + break; + /* + * BPF_RSH + */ + case BPF_ALU | BPF_RSH | BPF_X: /* dst = (u32) dst >> (u32) src */ + /* srl %dst,0(%src) */ + EMIT4_DISP(0x88000000, dst_reg, src_reg, 0); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_RSH | BPF_X: /* dst = dst >> src */ + /* srlg %dst,%dst,0(%src) */ + EMIT6_DISP_LH(0xeb000000, 0x000c, dst_reg, dst_reg, src_reg, 0); + break; + case BPF_ALU | BPF_RSH | BPF_K: /* dst = (u32) dst >> (u32) imm */ + if (imm == 0) + break; + /* srl %dst,imm(%r0) */ + EMIT4_DISP(0x88000000, dst_reg, REG_0, imm); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_RSH | BPF_K: /* dst = dst >> imm */ + if (imm == 0) + break; + /* srlg %dst,%dst,imm(%r0) */ + EMIT6_DISP_LH(0xeb000000, 0x000c, dst_reg, dst_reg, REG_0, imm); + break; + /* + * BPF_ARSH + */ + case BPF_ALU64 | BPF_ARSH | BPF_X: /* ((s64) dst) >>= src */ + /* srag %dst,%dst,0(%src) */ + EMIT6_DISP_LH(0xeb000000, 0x000a, dst_reg, dst_reg, src_reg, 0); + break; + case BPF_ALU64 | BPF_ARSH | BPF_K: /* ((s64) dst) >>= imm */ + if (imm == 0) + break; + /* srag %dst,%dst,imm(%r0) */ + EMIT6_DISP_LH(0xeb000000, 0x000a, dst_reg, dst_reg, REG_0, imm); + break; + /* + * BPF_NEG + */ + case BPF_ALU | BPF_NEG: /* dst = (u32) -dst */ + /* lcr %dst,%dst */ + EMIT2(0x1300, dst_reg, dst_reg); + EMIT_ZERO(dst_reg); + break; + case BPF_ALU64 | BPF_NEG: /* dst = -dst */ + /* lcgr %dst,%dst */ + EMIT4(0xb9130000, dst_reg, dst_reg); + break; + /* + * BPF_FROM_BE/LE + */ + case BPF_ALU | BPF_END | BPF_FROM_BE: + /* s390 is big endian, therefore only clear high order bytes */ + switch (imm) { + case 16: /* dst = (u16) cpu_to_be16(dst) */ + /* llghr %dst,%dst */ + EMIT4(0xb9850000, dst_reg, dst_reg); + break; + case 32: /* dst = (u32) cpu_to_be32(dst) */ + /* llgfr %dst,%dst */ + EMIT4(0xb9160000, dst_reg, dst_reg); + break; + case 64: /* dst = (u64) cpu_to_be64(dst) */ + break; + } + break; + case BPF_ALU | BPF_END | BPF_FROM_LE: + switch (imm) { + case 16: /* dst = (u16) cpu_to_le16(dst) */ + /* lrvr %dst,%dst */ + EMIT4(0xb91f0000, dst_reg, dst_reg); + /* srl %dst,16(%r0) */ + EMIT4_DISP(0x88000000, dst_reg, REG_0, 16); + /* llghr %dst,%dst */ + EMIT4(0xb9850000, dst_reg, dst_reg); + break; + case 32: /* dst = (u32) cpu_to_le32(dst) */ + /* lrvr %dst,%dst */ + EMIT4(0xb91f0000, dst_reg, dst_reg); + /* llgfr %dst,%dst */ + EMIT4(0xb9160000, dst_reg, dst_reg); + break; + case 64: /* dst = (u64) cpu_to_le64(dst) */ + /* lrvgr %dst,%dst */ + EMIT4(0xb90f0000, dst_reg, dst_reg); + break; + } + break; + /* + * BPF_ST(X) + */ + case BPF_STX | BPF_MEM | BPF_B: /* *(u8 *)(dst + off) = src_reg */ + /* stcy %src,off(%dst) */ + EMIT6_DISP_LH(0xe3000000, 0x0072, src_reg, dst_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + case BPF_STX | BPF_MEM | BPF_H: /* (u16 *)(dst + off) = src */ + /* sthy %src,off(%dst) */ + EMIT6_DISP_LH(0xe3000000, 0x0070, src_reg, dst_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + case BPF_STX | BPF_MEM | BPF_W: /* *(u32 *)(dst + off) = src */ + /* sty %src,off(%dst) */ + EMIT6_DISP_LH(0xe3000000, 0x0050, src_reg, dst_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + case BPF_STX | BPF_MEM | BPF_DW: /* (u64 *)(dst + off) = src */ + /* stg %src,off(%dst) */ + EMIT6_DISP_LH(0xe3000000, 0x0024, src_reg, dst_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + case BPF_ST | BPF_MEM | BPF_B: /* *(u8 *)(dst + off) = imm */ + /* lhi %w0,imm */ + EMIT4_IMM(0xa7080000, REG_W0, (u8) imm); + /* stcy %w0,off(dst) */ + EMIT6_DISP_LH(0xe3000000, 0x0072, REG_W0, dst_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + case BPF_ST | BPF_MEM | BPF_H: /* (u16 *)(dst + off) = imm */ + /* lhi %w0,imm */ + EMIT4_IMM(0xa7080000, REG_W0, (u16) imm); + /* sthy %w0,off(dst) */ + EMIT6_DISP_LH(0xe3000000, 0x0070, REG_W0, dst_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + case BPF_ST | BPF_MEM | BPF_W: /* *(u32 *)(dst + off) = imm */ + /* llilf %w0,imm */ + EMIT6_IMM(0xc00f0000, REG_W0, (u32) imm); + /* sty %w0,off(%dst) */ + EMIT6_DISP_LH(0xe3000000, 0x0050, REG_W0, dst_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + case BPF_ST | BPF_MEM | BPF_DW: /* *(u64 *)(dst + off) = imm */ + /* lgfi %w0,imm */ + EMIT6_IMM(0xc0010000, REG_W0, imm); + /* stg %w0,off(%dst) */ + EMIT6_DISP_LH(0xe3000000, 0x0024, REG_W0, dst_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + /* + * BPF_STX XADD (atomic_add) + */ + case BPF_STX | BPF_XADD | BPF_W: /* *(u32 *)(dst + off) += src */ + /* laal %w0,%src,off(%dst) */ + EMIT6_DISP_LH(0xeb000000, 0x00fa, REG_W0, src_reg, + dst_reg, off); + jit->seen |= SEEN_MEM; + break; + case BPF_STX | BPF_XADD | BPF_DW: /* *(u64 *)(dst + off) += src */ + /* laalg %w0,%src,off(%dst) */ + EMIT6_DISP_LH(0xeb000000, 0x00ea, REG_W0, src_reg, + dst_reg, off); + jit->seen |= SEEN_MEM; + break; + /* + * BPF_LDX + */ + case BPF_LDX | BPF_MEM | BPF_B: /* dst = *(u8 *)(ul) (src + off) */ + /* llgc %dst,0(off,%src) */ + EMIT6_DISP_LH(0xe3000000, 0x0090, dst_reg, src_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + case BPF_LDX | BPF_MEM | BPF_H: /* dst = *(u16 *)(ul) (src + off) */ + /* llgh %dst,0(off,%src) */ + EMIT6_DISP_LH(0xe3000000, 0x0091, dst_reg, src_reg, REG_0, off); + jit->seen |= SEEN_MEM; + break; + case BPF_LDX | BPF_MEM | BPF_W: /* dst = *(u32 *)(ul) (src + off) */ + /* llgf %dst,off(%src) */ + jit->seen |= SEEN_MEM; + EMIT6_DISP_LH(0xe3000000, 0x0016, dst_reg, src_reg, REG_0, off); + break; + case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */ + /* lg %dst,0(off,%src) */ + jit->seen |= SEEN_MEM; + EMIT6_DISP_LH(0xe3000000, 0x0004, dst_reg, src_reg, REG_0, off); + break; + /* + * BPF_JMP / CALL + */ + case BPF_JMP | BPF_CALL: + { + /* + * b0 = (__bpf_call_base + imm)(b1, b2, b3, b4, b5) + */ + const u64 func = (u64)__bpf_call_base + imm; + + REG_SET_SEEN(BPF_REG_5); + jit->seen |= SEEN_FUNC; + /* lg %w1,<d(imm)>(%l) */ + EMIT6_DISP(0xe3000000, 0x0004, REG_W1, REG_0, REG_L, + EMIT_CONST_U64(func)); + /* basr %r14,%w1 */ + EMIT2(0x0d00, REG_14, REG_W1); + /* lgr %b0,%r2: load return value into %b0 */ + EMIT4(0xb9040000, BPF_REG_0, REG_2); + break; + } + case BPF_JMP | BPF_EXIT: /* return b0 */ + last = (i == fp->len - 1) ? 1 : 0; + if (last && !(jit->seen & SEEN_RET0)) + break; + /* j <exit> */ + EMIT4_PCREL(0xa7f40000, jit->exit_ip - jit->prg); + break; + /* + * Branch relative (number of skipped instructions) to offset on + * condition. + * + * Condition code to mask mapping: + * + * CC | Description | Mask + * ------------------------------ + * 0 | Operands equal | 8 + * 1 | First operand low | 4 + * 2 | First operand high | 2 + * 3 | Unused | 1 + * + * For s390x relative branches: ip = ip + off_bytes + * For BPF relative branches: insn = insn + off_insns + 1 + * + * For example for s390x with offset 0 we jump to the branch + * instruction itself (loop) and for BPF with offset 0 we + * branch to the instruction behind the branch. + */ + case BPF_JMP | BPF_JA: /* if (true) */ + mask = 0xf000; /* j */ + goto branch_oc; + case BPF_JMP | BPF_JSGT | BPF_K: /* ((s64) dst > (s64) imm) */ + mask = 0x2000; /* jh */ + goto branch_ks; + case BPF_JMP | BPF_JSGE | BPF_K: /* ((s64) dst >= (s64) imm) */ + mask = 0xa000; /* jhe */ + goto branch_ks; + case BPF_JMP | BPF_JGT | BPF_K: /* (dst_reg > imm) */ + mask = 0x2000; /* jh */ + goto branch_ku; + case BPF_JMP | BPF_JGE | BPF_K: /* (dst_reg >= imm) */ + mask = 0xa000; /* jhe */ + goto branch_ku; + case BPF_JMP | BPF_JNE | BPF_K: /* (dst_reg != imm) */ + mask = 0x7000; /* jne */ + goto branch_ku; + case BPF_JMP | BPF_JEQ | BPF_K: /* (dst_reg == imm) */ + mask = 0x8000; /* je */ + goto branch_ku; + case BPF_JMP | BPF_JSET | BPF_K: /* (dst_reg & imm) */ + mask = 0x7000; /* jnz */ + /* lgfi %w1,imm (load sign extend imm) */ + EMIT6_IMM(0xc0010000, REG_W1, imm); + /* ngr %w1,%dst */ + EMIT4(0xb9800000, REG_W1, dst_reg); + goto branch_oc; + + case BPF_JMP | BPF_JSGT | BPF_X: /* ((s64) dst > (s64) src) */ + mask = 0x2000; /* jh */ + goto branch_xs; + case BPF_JMP | BPF_JSGE | BPF_X: /* ((s64) dst >= (s64) src) */ + mask = 0xa000; /* jhe */ + goto branch_xs; + case BPF_JMP | BPF_JGT | BPF_X: /* (dst > src) */ + mask = 0x2000; /* jh */ + goto branch_xu; + case BPF_JMP | BPF_JGE | BPF_X: /* (dst >= src) */ + mask = 0xa000; /* jhe */ + goto branch_xu; + case BPF_JMP | BPF_JNE | BPF_X: /* (dst != src) */ + mask = 0x7000; /* jne */ + goto branch_xu; + case BPF_JMP | BPF_JEQ | BPF_X: /* (dst == src) */ + mask = 0x8000; /* je */ + goto branch_xu; + case BPF_JMP | BPF_JSET | BPF_X: /* (dst & src) */ + mask = 0x7000; /* jnz */ + /* ngrk %w1,%dst,%src */ + EMIT4_RRF(0xb9e40000, REG_W1, dst_reg, src_reg); + goto branch_oc; +branch_ks: + /* lgfi %w1,imm (load sign extend imm) */ + EMIT6_IMM(0xc0010000, REG_W1, imm); + /* cgrj %dst,%w1,mask,off */ + EMIT6_PCREL(0xec000000, 0x0064, dst_reg, REG_W1, i, off, mask); + break; +branch_ku: + /* lgfi %w1,imm (load sign extend imm) */ + EMIT6_IMM(0xc0010000, REG_W1, imm); + /* clgrj %dst,%w1,mask,off */ + EMIT6_PCREL(0xec000000, 0x0065, dst_reg, REG_W1, i, off, mask); + break; +branch_xs: + /* cgrj %dst,%src,mask,off */ + EMIT6_PCREL(0xec000000, 0x0064, dst_reg, src_reg, i, off, mask); + break; +branch_xu: + /* clgrj %dst,%src,mask,off */ + EMIT6_PCREL(0xec000000, 0x0065, dst_reg, src_reg, i, off, mask); + break; +branch_oc: + /* brc mask,jmp_off (branch instruction needs 4 bytes) */ + jmp_off = addrs[i + off + 1] - (addrs[i + 1] - 4); + EMIT4_PCREL(0xa7040000 | mask << 8, jmp_off); + break; + /* + * BPF_LD + */ + case BPF_LD | BPF_ABS | BPF_B: /* b0 = *(u8 *) (skb->data+imm) */ + case BPF_LD | BPF_IND | BPF_B: /* b0 = *(u8 *) (skb->data+imm+src) */ + if ((BPF_MODE(insn->code) == BPF_ABS) && (imm >= 0)) + func_addr = __pa(sk_load_byte_pos); + else + func_addr = __pa(sk_load_byte); + goto call_fn; + case BPF_LD | BPF_ABS | BPF_H: /* b0 = *(u16 *) (skb->data+imm) */ + case BPF_LD | BPF_IND | BPF_H: /* b0 = *(u16 *) (skb->data+imm+src) */ + if ((BPF_MODE(insn->code) == BPF_ABS) && (imm >= 0)) + func_addr = __pa(sk_load_half_pos); + else + func_addr = __pa(sk_load_half); + goto call_fn; + case BPF_LD | BPF_ABS | BPF_W: /* b0 = *(u32 *) (skb->data+imm) */ + case BPF_LD | BPF_IND | BPF_W: /* b0 = *(u32 *) (skb->data+imm+src) */ + if ((BPF_MODE(insn->code) == BPF_ABS) && (imm >= 0)) + func_addr = __pa(sk_load_word_pos); + else + func_addr = __pa(sk_load_word); + goto call_fn; +call_fn: + jit->seen |= SEEN_SKB | SEEN_RET0 | SEEN_FUNC; + REG_SET_SEEN(REG_14); /* Return address of possible func call */ + + /* + * Implicit input: + * BPF_REG_6 (R7) : skb pointer + * REG_SKB_DATA (R12): skb data pointer + * + * Calculated input: + * BPF_REG_2 (R3) : offset of byte(s) to fetch in skb + * BPF_REG_5 (R6) : return address + * + * Output: + * BPF_REG_0 (R14): data read from skb + * + * Scratch registers (BPF_REG_1-5) + */ + + /* Call function: llilf %w1,func_addr */ + EMIT6_IMM(0xc00f0000, REG_W1, func_addr); + + /* Offset: lgfi %b2,imm */ + EMIT6_IMM(0xc0010000, BPF_REG_2, imm); + if (BPF_MODE(insn->code) == BPF_IND) + /* agfr %b2,%src (%src is s32 here) */ + EMIT4(0xb9180000, BPF_REG_2, src_reg); + + /* basr %b5,%w1 (%b5 is call saved) */ + EMIT2(0x0d00, BPF_REG_5, REG_W1); + + /* + * Note: For fast access we jump directly after the + * jnz instruction from bpf_jit.S + */ + /* jnz <ret0> */ + EMIT4_PCREL(0xa7740000, jit->ret0_ip - jit->prg); + break; + default: /* too complex, give up */ + pr_err("Unknown opcode %02x\n", insn->code); + return -1; + } + return insn_count; +} + +/* + * Compile eBPF program into s390x code + */ +static int bpf_jit_prog(struct bpf_jit *jit, struct bpf_prog *fp) +{ + int i, insn_count; + + jit->lit = jit->lit_start; + jit->prg = 0; + + bpf_jit_prologue(jit); + for (i = 0; i < fp->len; i += insn_count) { + insn_count = bpf_jit_insn(jit, fp, i); + if (insn_count < 0) + return -1; + jit->addrs[i + 1] = jit->prg; /* Next instruction address */ + } + bpf_jit_epilogue(jit); + + jit->lit_start = jit->prg; + jit->size = jit->lit; + jit->size_prg = jit->prg; + return 0; +} + +/* + * Classic BPF function stub. BPF programs will be converted into + * eBPF and then bpf_int_jit_compile() will be called. + */ +void bpf_jit_compile(struct bpf_prog *fp) +{ +} + +/* + * Compile eBPF program "fp" + */ +void bpf_int_jit_compile(struct bpf_prog *fp) +{ + struct bpf_binary_header *header; + struct bpf_jit jit; + int pass; + + if (!bpf_jit_enable) + return; + memset(&jit, 0, sizeof(jit)); + jit.addrs = kcalloc(fp->len + 1, sizeof(*jit.addrs), GFP_KERNEL); + if (jit.addrs == NULL) + return; + /* + * Three initial passes: + * - 1/2: Determine clobbered registers + * - 3: Calculate program size and addrs arrray + */ + for (pass = 1; pass <= 3; pass++) { + if (bpf_jit_prog(&jit, fp)) + goto free_addrs; + } + /* + * Final pass: Allocate and generate program + */ + if (jit.size >= BPF_SIZE_MAX) + goto free_addrs; + header = bpf_jit_binary_alloc(jit.size, &jit.prg_buf, 2, jit_fill_hole); + if (!header) + goto free_addrs; + if (bpf_jit_prog(&jit, fp)) + goto free_addrs; + if (bpf_jit_enable > 1) { + bpf_jit_dump(fp->len, jit.size, pass, jit.prg_buf); + if (jit.prg_buf) + print_fn_code(jit.prg_buf, jit.size_prg); + } + if (jit.prg_buf) { + set_memory_ro((unsigned long)header, header->pages); + fp->bpf_func = (void *) jit.prg_buf; + fp->jited = true; + } +free_addrs: + kfree(jit.addrs); +} + +/* + * Free eBPF program + */ +void bpf_jit_free(struct bpf_prog *fp) +{ + unsigned long addr = (unsigned long)fp->bpf_func & PAGE_MASK; + struct bpf_binary_header *header = (void *)addr; + + if (!fp->jited) + goto free_filter; + + set_memory_rw(addr, header->pages); + bpf_jit_binary_free(header); + +free_filter: + bpf_prog_unlock_free(fp); +} diff --git a/arch/s390/oprofile/Makefile b/arch/s390/oprofile/Makefile new file mode 100644 index 000000000..1bd230171 --- /dev/null +++ b/arch/s390/oprofile/Makefile @@ -0,0 +1,10 @@ +obj-$(CONFIG_OPROFILE) += oprofile.o + +DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \ + oprof.o cpu_buffer.o buffer_sync.o \ + event_buffer.o oprofile_files.o \ + oprofilefs.o oprofile_stats.o \ + timer_int.o ) + +oprofile-y := $(DRIVER_OBJS) init.o backtrace.o +oprofile-y += hwsampler.o diff --git a/arch/s390/oprofile/backtrace.c b/arch/s390/oprofile/backtrace.c new file mode 100644 index 000000000..8a6811b2c --- /dev/null +++ b/arch/s390/oprofile/backtrace.c @@ -0,0 +1,77 @@ +/* + * S390 Version + * Copyright IBM Corp. 2005 + * Author(s): Andreas Krebbel <Andreas.Krebbel@de.ibm.com> + */ + +#include <linux/oprofile.h> + +#include <asm/processor.h> /* for struct stack_frame */ + +static unsigned long +__show_trace(unsigned int *depth, unsigned long sp, + unsigned long low, unsigned long high) +{ + struct stack_frame *sf; + struct pt_regs *regs; + + while (*depth) { + sp = sp & PSW_ADDR_INSN; + if (sp < low || sp > high - sizeof(*sf)) + return sp; + sf = (struct stack_frame *) sp; + (*depth)--; + oprofile_add_trace(sf->gprs[8] & PSW_ADDR_INSN); + + /* Follow the backchain. */ + while (*depth) { + low = sp; + sp = sf->back_chain & PSW_ADDR_INSN; + if (!sp) + break; + if (sp <= low || sp > high - sizeof(*sf)) + return sp; + sf = (struct stack_frame *) sp; + (*depth)--; + oprofile_add_trace(sf->gprs[8] & PSW_ADDR_INSN); + + } + + if (*depth == 0) + break; + + /* Zero backchain detected, check for interrupt frame. */ + sp = (unsigned long) (sf + 1); + if (sp <= low || sp > high - sizeof(*regs)) + return sp; + regs = (struct pt_regs *) sp; + (*depth)--; + oprofile_add_trace(sf->gprs[8] & PSW_ADDR_INSN); + low = sp; + sp = regs->gprs[15]; + } + return sp; +} + +void s390_backtrace(struct pt_regs * const regs, unsigned int depth) +{ + unsigned long head; + struct stack_frame* head_sf; + + if (user_mode(regs)) + return; + + head = regs->gprs[15]; + head_sf = (struct stack_frame*)head; + + if (!head_sf->back_chain) + return; + + head = head_sf->back_chain; + + head = __show_trace(&depth, head, S390_lowcore.async_stack - ASYNC_SIZE, + S390_lowcore.async_stack); + + __show_trace(&depth, head, S390_lowcore.thread_info, + S390_lowcore.thread_info + THREAD_SIZE); +} diff --git a/arch/s390/oprofile/hwsampler.c b/arch/s390/oprofile/hwsampler.c new file mode 100644 index 000000000..ff9b4eb34 --- /dev/null +++ b/arch/s390/oprofile/hwsampler.c @@ -0,0 +1,1178 @@ +/* + * Copyright IBM Corp. 2010 + * Author: Heinz Graalfs <graalfs@de.ibm.com> + */ + +#include <linux/kernel_stat.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/smp.h> +#include <linux/errno.h> +#include <linux/workqueue.h> +#include <linux/interrupt.h> +#include <linux/notifier.h> +#include <linux/cpu.h> +#include <linux/semaphore.h> +#include <linux/oom.h> +#include <linux/oprofile.h> + +#include <asm/facility.h> +#include <asm/cpu_mf.h> +#include <asm/irq.h> + +#include "hwsampler.h" +#include "op_counter.h" + +#define MAX_NUM_SDB 511 +#define MIN_NUM_SDB 1 + +DECLARE_PER_CPU(struct hws_cpu_buffer, sampler_cpu_buffer); + +struct hws_execute_parms { + void *buffer; + signed int rc; +}; + +DEFINE_PER_CPU(struct hws_cpu_buffer, sampler_cpu_buffer); +EXPORT_PER_CPU_SYMBOL(sampler_cpu_buffer); + +static DEFINE_MUTEX(hws_sem); +static DEFINE_MUTEX(hws_sem_oom); + +static unsigned char hws_flush_all; +static unsigned int hws_oom; +static unsigned int hws_alert; +static struct workqueue_struct *hws_wq; + +static unsigned int hws_state; +enum { + HWS_INIT = 1, + HWS_DEALLOCATED, + HWS_STOPPED, + HWS_STARTED, + HWS_STOPPING }; + +/* set to 1 if called by kernel during memory allocation */ +static unsigned char oom_killer_was_active; +/* size of SDBT and SDB as of allocate API */ +static unsigned long num_sdbt = 100; +static unsigned long num_sdb = 511; +/* sampling interval (machine cycles) */ +static unsigned long interval; + +static unsigned long min_sampler_rate; +static unsigned long max_sampler_rate; + +static void execute_qsi(void *parms) +{ + struct hws_execute_parms *ep = parms; + + ep->rc = qsi(ep->buffer); +} + +static void execute_ssctl(void *parms) +{ + struct hws_execute_parms *ep = parms; + + ep->rc = lsctl(ep->buffer); +} + +static int smp_ctl_ssctl_stop(int cpu) +{ + int rc; + struct hws_execute_parms ep; + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + + cb->ssctl.es = 0; + cb->ssctl.cs = 0; + + ep.buffer = &cb->ssctl; + smp_call_function_single(cpu, execute_ssctl, &ep, 1); + rc = ep.rc; + if (rc) { + printk(KERN_ERR "hwsampler: CPU %d CPUMF SSCTL failed.\n", cpu); + dump_stack(); + } + + ep.buffer = &cb->qsi; + smp_call_function_single(cpu, execute_qsi, &ep, 1); + + if (cb->qsi.es || cb->qsi.cs) { + printk(KERN_EMERG "CPUMF sampling did not stop properly.\n"); + dump_stack(); + } + + return rc; +} + +static int smp_ctl_ssctl_deactivate(int cpu) +{ + int rc; + struct hws_execute_parms ep; + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + + cb->ssctl.es = 1; + cb->ssctl.cs = 0; + + ep.buffer = &cb->ssctl; + smp_call_function_single(cpu, execute_ssctl, &ep, 1); + rc = ep.rc; + if (rc) + printk(KERN_ERR "hwsampler: CPU %d CPUMF SSCTL failed.\n", cpu); + + ep.buffer = &cb->qsi; + smp_call_function_single(cpu, execute_qsi, &ep, 1); + + if (cb->qsi.cs) + printk(KERN_EMERG "CPUMF sampling was not set inactive.\n"); + + return rc; +} + +static int smp_ctl_ssctl_enable_activate(int cpu, unsigned long interval) +{ + int rc; + struct hws_execute_parms ep; + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + + cb->ssctl.h = 1; + cb->ssctl.tear = cb->first_sdbt; + cb->ssctl.dear = *(unsigned long *) cb->first_sdbt; + cb->ssctl.interval = interval; + cb->ssctl.es = 1; + cb->ssctl.cs = 1; + + ep.buffer = &cb->ssctl; + smp_call_function_single(cpu, execute_ssctl, &ep, 1); + rc = ep.rc; + if (rc) + printk(KERN_ERR "hwsampler: CPU %d CPUMF SSCTL failed.\n", cpu); + + ep.buffer = &cb->qsi; + smp_call_function_single(cpu, execute_qsi, &ep, 1); + if (ep.rc) + printk(KERN_ERR "hwsampler: CPU %d CPUMF QSI failed.\n", cpu); + + return rc; +} + +static int smp_ctl_qsi(int cpu) +{ + struct hws_execute_parms ep; + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + + ep.buffer = &cb->qsi; + smp_call_function_single(cpu, execute_qsi, &ep, 1); + + return ep.rc; +} + +static void hws_ext_handler(struct ext_code ext_code, + unsigned int param32, unsigned long param64) +{ + struct hws_cpu_buffer *cb = this_cpu_ptr(&sampler_cpu_buffer); + + if (!(param32 & CPU_MF_INT_SF_MASK)) + return; + + if (!hws_alert) + return; + + inc_irq_stat(IRQEXT_CMS); + atomic_xchg(&cb->ext_params, atomic_read(&cb->ext_params) | param32); + + if (hws_wq) + queue_work(hws_wq, &cb->worker); +} + +static void worker(struct work_struct *work); + +static void add_samples_to_oprofile(unsigned cpu, unsigned long *, + unsigned long *dear); + +static void init_all_cpu_buffers(void) +{ + int cpu; + struct hws_cpu_buffer *cb; + + for_each_online_cpu(cpu) { + cb = &per_cpu(sampler_cpu_buffer, cpu); + memset(cb, 0, sizeof(struct hws_cpu_buffer)); + } +} + +static void prepare_cpu_buffers(void) +{ + struct hws_cpu_buffer *cb; + int cpu; + + for_each_online_cpu(cpu) { + cb = &per_cpu(sampler_cpu_buffer, cpu); + atomic_set(&cb->ext_params, 0); + cb->worker_entry = 0; + cb->sample_overflow = 0; + cb->req_alert = 0; + cb->incorrect_sdbt_entry = 0; + cb->invalid_entry_address = 0; + cb->loss_of_sample_data = 0; + cb->sample_auth_change_alert = 0; + cb->finish = 0; + cb->oom = 0; + cb->stop_mode = 0; + } +} + +/* + * allocate_sdbt() - allocate sampler memory + * @cpu: the cpu for which sampler memory is allocated + * + * A 4K page is allocated for each requested SDBT. + * A maximum of 511 4K pages are allocated for the SDBs in each of the SDBTs. + * Set ALERT_REQ mask in each SDBs trailer. + * Returns zero if successful, <0 otherwise. + */ +static int allocate_sdbt(int cpu) +{ + int j, k, rc; + unsigned long *sdbt; + unsigned long sdb; + unsigned long *tail; + unsigned long *trailer; + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + + if (cb->first_sdbt) + return -EINVAL; + + sdbt = NULL; + tail = sdbt; + + for (j = 0; j < num_sdbt; j++) { + sdbt = (unsigned long *)get_zeroed_page(GFP_KERNEL); + + mutex_lock(&hws_sem_oom); + /* OOM killer might have been activated */ + barrier(); + if (oom_killer_was_active || !sdbt) { + if (sdbt) + free_page((unsigned long)sdbt); + + goto allocate_sdbt_error; + } + if (cb->first_sdbt == 0) + cb->first_sdbt = (unsigned long)sdbt; + + /* link current page to tail of chain */ + if (tail) + *tail = (unsigned long)(void *)sdbt + 1; + + mutex_unlock(&hws_sem_oom); + + for (k = 0; k < num_sdb; k++) { + /* get and set SDB page */ + sdb = get_zeroed_page(GFP_KERNEL); + + mutex_lock(&hws_sem_oom); + /* OOM killer might have been activated */ + barrier(); + if (oom_killer_was_active || !sdb) { + if (sdb) + free_page(sdb); + + goto allocate_sdbt_error; + } + *sdbt = sdb; + trailer = trailer_entry_ptr(*sdbt); + *trailer = SDB_TE_ALERT_REQ_MASK; + sdbt++; + mutex_unlock(&hws_sem_oom); + } + tail = sdbt; + } + mutex_lock(&hws_sem_oom); + if (oom_killer_was_active) + goto allocate_sdbt_error; + + rc = 0; + if (tail) + *tail = (unsigned long) + ((void *)cb->first_sdbt) + 1; + +allocate_sdbt_exit: + mutex_unlock(&hws_sem_oom); + return rc; + +allocate_sdbt_error: + rc = -ENOMEM; + goto allocate_sdbt_exit; +} + +/* + * deallocate_sdbt() - deallocate all sampler memory + * + * For each online CPU all SDBT trees are deallocated. + * Returns the number of freed pages. + */ +static int deallocate_sdbt(void) +{ + int cpu; + int counter; + + counter = 0; + + for_each_online_cpu(cpu) { + unsigned long start; + unsigned long sdbt; + unsigned long *curr; + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + + if (!cb->first_sdbt) + continue; + + sdbt = cb->first_sdbt; + curr = (unsigned long *) sdbt; + start = sdbt; + + /* we'll free the SDBT after all SDBs are processed... */ + while (1) { + if (!*curr || !sdbt) + break; + + /* watch for link entry reset if found */ + if (is_link_entry(curr)) { + curr = get_next_sdbt(curr); + if (sdbt) + free_page(sdbt); + + /* we are done if we reach the start */ + if ((unsigned long) curr == start) + break; + else + sdbt = (unsigned long) curr; + } else { + /* process SDB pointer */ + if (*curr) { + free_page(*curr); + curr++; + } + } + counter++; + } + cb->first_sdbt = 0; + } + return counter; +} + +static int start_sampling(int cpu) +{ + int rc; + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + rc = smp_ctl_ssctl_enable_activate(cpu, interval); + if (rc) { + printk(KERN_INFO "hwsampler: CPU %d ssctl failed.\n", cpu); + goto start_exit; + } + + rc = -EINVAL; + if (!cb->qsi.es) { + printk(KERN_INFO "hwsampler: CPU %d ssctl not enabled.\n", cpu); + goto start_exit; + } + + if (!cb->qsi.cs) { + printk(KERN_INFO "hwsampler: CPU %d ssctl not active.\n", cpu); + goto start_exit; + } + + printk(KERN_INFO + "hwsampler: CPU %d, CPUMF Sampling started, interval %lu.\n", + cpu, interval); + + rc = 0; + +start_exit: + return rc; +} + +static int stop_sampling(int cpu) +{ + unsigned long v; + int rc; + struct hws_cpu_buffer *cb; + + rc = smp_ctl_qsi(cpu); + WARN_ON(rc); + + cb = &per_cpu(sampler_cpu_buffer, cpu); + if (!rc && !cb->qsi.es) + printk(KERN_INFO "hwsampler: CPU %d, already stopped.\n", cpu); + + rc = smp_ctl_ssctl_stop(cpu); + if (rc) { + printk(KERN_INFO "hwsampler: CPU %d, ssctl stop error %d.\n", + cpu, rc); + goto stop_exit; + } + + printk(KERN_INFO "hwsampler: CPU %d, CPUMF Sampling stopped.\n", cpu); + +stop_exit: + v = cb->req_alert; + if (v) + printk(KERN_ERR "hwsampler: CPU %d CPUMF Request alert," + " count=%lu.\n", cpu, v); + + v = cb->loss_of_sample_data; + if (v) + printk(KERN_ERR "hwsampler: CPU %d CPUMF Loss of sample data," + " count=%lu.\n", cpu, v); + + v = cb->invalid_entry_address; + if (v) + printk(KERN_ERR "hwsampler: CPU %d CPUMF Invalid entry address," + " count=%lu.\n", cpu, v); + + v = cb->incorrect_sdbt_entry; + if (v) + printk(KERN_ERR + "hwsampler: CPU %d CPUMF Incorrect SDBT address," + " count=%lu.\n", cpu, v); + + v = cb->sample_auth_change_alert; + if (v) + printk(KERN_ERR + "hwsampler: CPU %d CPUMF Sample authorization change," + " count=%lu.\n", cpu, v); + + return rc; +} + +static int check_hardware_prerequisites(void) +{ + if (!test_facility(68)) + return -EOPNOTSUPP; + return 0; +} +/* + * hws_oom_callback() - the OOM callback function + * + * In case the callback is invoked during memory allocation for the + * hw sampler, all obtained memory is deallocated and a flag is set + * so main sampler memory allocation can exit with a failure code. + * In case the callback is invoked during sampling the hw sampler + * is deactivated for all CPUs. + */ +static int hws_oom_callback(struct notifier_block *nfb, + unsigned long dummy, void *parm) +{ + unsigned long *freed; + int cpu; + struct hws_cpu_buffer *cb; + + freed = parm; + + mutex_lock(&hws_sem_oom); + + if (hws_state == HWS_DEALLOCATED) { + /* during memory allocation */ + if (oom_killer_was_active == 0) { + oom_killer_was_active = 1; + *freed += deallocate_sdbt(); + } + } else { + int i; + cpu = get_cpu(); + cb = &per_cpu(sampler_cpu_buffer, cpu); + + if (!cb->oom) { + for_each_online_cpu(i) { + smp_ctl_ssctl_deactivate(i); + cb->oom = 1; + } + cb->finish = 1; + + printk(KERN_INFO + "hwsampler: CPU %d, OOM notify during CPUMF Sampling.\n", + cpu); + } + } + + mutex_unlock(&hws_sem_oom); + + return NOTIFY_OK; +} + +static struct notifier_block hws_oom_notifier = { + .notifier_call = hws_oom_callback +}; + +static int hws_cpu_callback(struct notifier_block *nfb, + unsigned long action, void *hcpu) +{ + /* We do not have sampler space available for all possible CPUs. + All CPUs should be online when hw sampling is activated. */ + return (hws_state <= HWS_DEALLOCATED) ? NOTIFY_OK : NOTIFY_BAD; +} + +static struct notifier_block hws_cpu_notifier = { + .notifier_call = hws_cpu_callback +}; + +/** + * hwsampler_deactivate() - set hardware sampling temporarily inactive + * @cpu: specifies the CPU to be set inactive. + * + * Returns 0 on success, !0 on failure. + */ +int hwsampler_deactivate(unsigned int cpu) +{ + /* + * Deactivate hw sampling temporarily and flush the buffer + * by pushing all the pending samples to oprofile buffer. + * + * This function can be called under one of the following conditions: + * Memory unmap, task is exiting. + */ + int rc; + struct hws_cpu_buffer *cb; + + rc = 0; + mutex_lock(&hws_sem); + + cb = &per_cpu(sampler_cpu_buffer, cpu); + if (hws_state == HWS_STARTED) { + rc = smp_ctl_qsi(cpu); + WARN_ON(rc); + if (cb->qsi.cs) { + rc = smp_ctl_ssctl_deactivate(cpu); + if (rc) { + printk(KERN_INFO + "hwsampler: CPU %d, CPUMF Deactivation failed.\n", cpu); + cb->finish = 1; + hws_state = HWS_STOPPING; + } else { + hws_flush_all = 1; + /* Add work to queue to read pending samples.*/ + queue_work_on(cpu, hws_wq, &cb->worker); + } + } + } + mutex_unlock(&hws_sem); + + if (hws_wq) + flush_workqueue(hws_wq); + + return rc; +} + +/** + * hwsampler_activate() - activate/resume hardware sampling which was deactivated + * @cpu: specifies the CPU to be set active. + * + * Returns 0 on success, !0 on failure. + */ +int hwsampler_activate(unsigned int cpu) +{ + /* + * Re-activate hw sampling. This should be called in pair with + * hwsampler_deactivate(). + */ + int rc; + struct hws_cpu_buffer *cb; + + rc = 0; + mutex_lock(&hws_sem); + + cb = &per_cpu(sampler_cpu_buffer, cpu); + if (hws_state == HWS_STARTED) { + rc = smp_ctl_qsi(cpu); + WARN_ON(rc); + if (!cb->qsi.cs) { + hws_flush_all = 0; + rc = smp_ctl_ssctl_enable_activate(cpu, interval); + if (rc) { + printk(KERN_ERR + "CPU %d, CPUMF activate sampling failed.\n", + cpu); + } + } + } + + mutex_unlock(&hws_sem); + + return rc; +} + +static int check_qsi_on_setup(void) +{ + int rc; + unsigned int cpu; + struct hws_cpu_buffer *cb; + + for_each_online_cpu(cpu) { + cb = &per_cpu(sampler_cpu_buffer, cpu); + rc = smp_ctl_qsi(cpu); + WARN_ON(rc); + if (rc) + return -EOPNOTSUPP; + + if (!cb->qsi.as) { + printk(KERN_INFO "hwsampler: CPUMF sampling is not authorized.\n"); + return -EINVAL; + } + + if (cb->qsi.es) { + printk(KERN_WARNING "hwsampler: CPUMF is still enabled.\n"); + rc = smp_ctl_ssctl_stop(cpu); + if (rc) + return -EINVAL; + + printk(KERN_INFO + "CPU %d, CPUMF Sampling stopped now.\n", cpu); + } + } + return 0; +} + +static int check_qsi_on_start(void) +{ + unsigned int cpu; + int rc; + struct hws_cpu_buffer *cb; + + for_each_online_cpu(cpu) { + cb = &per_cpu(sampler_cpu_buffer, cpu); + rc = smp_ctl_qsi(cpu); + WARN_ON(rc); + + if (!cb->qsi.as) + return -EINVAL; + + if (cb->qsi.es) + return -EINVAL; + + if (cb->qsi.cs) + return -EINVAL; + } + return 0; +} + +static void worker_on_start(unsigned int cpu) +{ + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + cb->worker_entry = cb->first_sdbt; +} + +static int worker_check_error(unsigned int cpu, int ext_params) +{ + int rc; + unsigned long *sdbt; + struct hws_cpu_buffer *cb; + + rc = 0; + cb = &per_cpu(sampler_cpu_buffer, cpu); + sdbt = (unsigned long *) cb->worker_entry; + + if (!sdbt || !*sdbt) + return -EINVAL; + + if (ext_params & CPU_MF_INT_SF_PRA) + cb->req_alert++; + + if (ext_params & CPU_MF_INT_SF_LSDA) + cb->loss_of_sample_data++; + + if (ext_params & CPU_MF_INT_SF_IAE) { + cb->invalid_entry_address++; + rc = -EINVAL; + } + + if (ext_params & CPU_MF_INT_SF_ISE) { + cb->incorrect_sdbt_entry++; + rc = -EINVAL; + } + + if (ext_params & CPU_MF_INT_SF_SACA) { + cb->sample_auth_change_alert++; + rc = -EINVAL; + } + + return rc; +} + +static void worker_on_finish(unsigned int cpu) +{ + int rc, i; + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + + if (cb->finish) { + rc = smp_ctl_qsi(cpu); + WARN_ON(rc); + if (cb->qsi.es) { + printk(KERN_INFO + "hwsampler: CPU %d, CPUMF Stop/Deactivate sampling.\n", + cpu); + rc = smp_ctl_ssctl_stop(cpu); + if (rc) + printk(KERN_INFO + "hwsampler: CPU %d, CPUMF Deactivation failed.\n", + cpu); + + for_each_online_cpu(i) { + if (i == cpu) + continue; + if (!cb->finish) { + cb->finish = 1; + queue_work_on(i, hws_wq, + &cb->worker); + } + } + } + } +} + +static void worker_on_interrupt(unsigned int cpu) +{ + unsigned long *sdbt; + unsigned char done; + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + + sdbt = (unsigned long *) cb->worker_entry; + + done = 0; + /* do not proceed if stop was entered, + * forget the buffers not yet processed */ + while (!done && !cb->stop_mode) { + unsigned long *trailer; + struct hws_trailer_entry *te; + unsigned long *dear = 0; + + trailer = trailer_entry_ptr(*sdbt); + /* leave loop if no more work to do */ + if (!(*trailer & SDB_TE_BUFFER_FULL_MASK)) { + done = 1; + if (!hws_flush_all) + continue; + } + + te = (struct hws_trailer_entry *)trailer; + cb->sample_overflow += te->overflow; + + add_samples_to_oprofile(cpu, sdbt, dear); + + /* reset trailer */ + xchg((unsigned char *) te, 0x40); + + /* advance to next sdb slot in current sdbt */ + sdbt++; + /* in case link bit is set use address w/o link bit */ + if (is_link_entry(sdbt)) + sdbt = get_next_sdbt(sdbt); + + cb->worker_entry = (unsigned long)sdbt; + } +} + +static void add_samples_to_oprofile(unsigned int cpu, unsigned long *sdbt, + unsigned long *dear) +{ + struct hws_basic_entry *sample_data_ptr; + unsigned long *trailer; + + trailer = trailer_entry_ptr(*sdbt); + if (dear) { + if (dear > trailer) + return; + trailer = dear; + } + + sample_data_ptr = (struct hws_basic_entry *)(*sdbt); + + while ((unsigned long *)sample_data_ptr < trailer) { + struct pt_regs *regs = NULL; + struct task_struct *tsk = NULL; + + /* + * Check sampling mode, 1 indicates basic (=customer) sampling + * mode. + */ + if (sample_data_ptr->def != 1) { + /* sample slot is not yet written */ + break; + } else { + /* make sure we don't use it twice, + * the next time the sampler will set it again */ + sample_data_ptr->def = 0; + } + + /* Get pt_regs. */ + if (sample_data_ptr->P == 1) { + /* userspace sample */ + unsigned int pid = sample_data_ptr->prim_asn; + if (!counter_config.user) + goto skip_sample; + rcu_read_lock(); + tsk = pid_task(find_vpid(pid), PIDTYPE_PID); + if (tsk) + regs = task_pt_regs(tsk); + rcu_read_unlock(); + } else { + /* kernelspace sample */ + if (!counter_config.kernel) + goto skip_sample; + regs = task_pt_regs(current); + } + + mutex_lock(&hws_sem); + oprofile_add_ext_hw_sample(sample_data_ptr->ia, regs, 0, + !sample_data_ptr->P, tsk); + mutex_unlock(&hws_sem); + skip_sample: + sample_data_ptr++; + } +} + +static void worker(struct work_struct *work) +{ + unsigned int cpu; + int ext_params; + struct hws_cpu_buffer *cb; + + cb = container_of(work, struct hws_cpu_buffer, worker); + cpu = smp_processor_id(); + ext_params = atomic_xchg(&cb->ext_params, 0); + + if (!cb->worker_entry) + worker_on_start(cpu); + + if (worker_check_error(cpu, ext_params)) + return; + + if (!cb->finish) + worker_on_interrupt(cpu); + + if (cb->finish) + worker_on_finish(cpu); +} + +/** + * hwsampler_allocate() - allocate memory for the hardware sampler + * @sdbt: number of SDBTs per online CPU (must be > 0) + * @sdb: number of SDBs per SDBT (minimum 1, maximum 511) + * + * Returns 0 on success, !0 on failure. + */ +int hwsampler_allocate(unsigned long sdbt, unsigned long sdb) +{ + int cpu, rc; + mutex_lock(&hws_sem); + + rc = -EINVAL; + if (hws_state != HWS_DEALLOCATED) + goto allocate_exit; + + if (sdbt < 1) + goto allocate_exit; + + if (sdb > MAX_NUM_SDB || sdb < MIN_NUM_SDB) + goto allocate_exit; + + num_sdbt = sdbt; + num_sdb = sdb; + + oom_killer_was_active = 0; + register_oom_notifier(&hws_oom_notifier); + + for_each_online_cpu(cpu) { + if (allocate_sdbt(cpu)) { + unregister_oom_notifier(&hws_oom_notifier); + goto allocate_error; + } + } + unregister_oom_notifier(&hws_oom_notifier); + if (oom_killer_was_active) + goto allocate_error; + + hws_state = HWS_STOPPED; + rc = 0; + +allocate_exit: + mutex_unlock(&hws_sem); + return rc; + +allocate_error: + rc = -ENOMEM; + printk(KERN_ERR "hwsampler: CPUMF Memory allocation failed.\n"); + goto allocate_exit; +} + +/** + * hwsampler_deallocate() - deallocate hardware sampler memory + * + * Returns 0 on success, !0 on failure. + */ +int hwsampler_deallocate(void) +{ + int rc; + + mutex_lock(&hws_sem); + + rc = -EINVAL; + if (hws_state != HWS_STOPPED) + goto deallocate_exit; + + irq_subclass_unregister(IRQ_SUBCLASS_MEASUREMENT_ALERT); + hws_alert = 0; + deallocate_sdbt(); + + hws_state = HWS_DEALLOCATED; + rc = 0; + +deallocate_exit: + mutex_unlock(&hws_sem); + + return rc; +} + +unsigned long hwsampler_query_min_interval(void) +{ + return min_sampler_rate; +} + +unsigned long hwsampler_query_max_interval(void) +{ + return max_sampler_rate; +} + +unsigned long hwsampler_get_sample_overflow_count(unsigned int cpu) +{ + struct hws_cpu_buffer *cb; + + cb = &per_cpu(sampler_cpu_buffer, cpu); + + return cb->sample_overflow; +} + +int hwsampler_setup(void) +{ + int rc; + int cpu; + struct hws_cpu_buffer *cb; + + mutex_lock(&hws_sem); + + rc = -EINVAL; + if (hws_state) + goto setup_exit; + + hws_state = HWS_INIT; + + init_all_cpu_buffers(); + + rc = check_hardware_prerequisites(); + if (rc) + goto setup_exit; + + rc = check_qsi_on_setup(); + if (rc) + goto setup_exit; + + rc = -EINVAL; + hws_wq = create_workqueue("hwsampler"); + if (!hws_wq) + goto setup_exit; + + register_cpu_notifier(&hws_cpu_notifier); + + for_each_online_cpu(cpu) { + cb = &per_cpu(sampler_cpu_buffer, cpu); + INIT_WORK(&cb->worker, worker); + rc = smp_ctl_qsi(cpu); + WARN_ON(rc); + if (min_sampler_rate != cb->qsi.min_sampl_rate) { + if (min_sampler_rate) { + printk(KERN_WARNING + "hwsampler: different min sampler rate values.\n"); + if (min_sampler_rate < cb->qsi.min_sampl_rate) + min_sampler_rate = + cb->qsi.min_sampl_rate; + } else + min_sampler_rate = cb->qsi.min_sampl_rate; + } + if (max_sampler_rate != cb->qsi.max_sampl_rate) { + if (max_sampler_rate) { + printk(KERN_WARNING + "hwsampler: different max sampler rate values.\n"); + if (max_sampler_rate > cb->qsi.max_sampl_rate) + max_sampler_rate = + cb->qsi.max_sampl_rate; + } else + max_sampler_rate = cb->qsi.max_sampl_rate; + } + } + register_external_irq(EXT_IRQ_MEASURE_ALERT, hws_ext_handler); + + hws_state = HWS_DEALLOCATED; + rc = 0; + +setup_exit: + mutex_unlock(&hws_sem); + return rc; +} + +int hwsampler_shutdown(void) +{ + int rc; + + mutex_lock(&hws_sem); + + rc = -EINVAL; + if (hws_state == HWS_DEALLOCATED || hws_state == HWS_STOPPED) { + mutex_unlock(&hws_sem); + + if (hws_wq) + flush_workqueue(hws_wq); + + mutex_lock(&hws_sem); + + if (hws_state == HWS_STOPPED) { + irq_subclass_unregister(IRQ_SUBCLASS_MEASUREMENT_ALERT); + hws_alert = 0; + deallocate_sdbt(); + } + if (hws_wq) { + destroy_workqueue(hws_wq); + hws_wq = NULL; + } + + unregister_external_irq(EXT_IRQ_MEASURE_ALERT, hws_ext_handler); + hws_state = HWS_INIT; + rc = 0; + } + mutex_unlock(&hws_sem); + + unregister_cpu_notifier(&hws_cpu_notifier); + + return rc; +} + +/** + * hwsampler_start_all() - start hardware sampling on all online CPUs + * @rate: specifies the used interval when samples are taken + * + * Returns 0 on success, !0 on failure. + */ +int hwsampler_start_all(unsigned long rate) +{ + int rc, cpu; + + mutex_lock(&hws_sem); + + hws_oom = 0; + + rc = -EINVAL; + if (hws_state != HWS_STOPPED) + goto start_all_exit; + + interval = rate; + + /* fail if rate is not valid */ + if (interval < min_sampler_rate || interval > max_sampler_rate) + goto start_all_exit; + + rc = check_qsi_on_start(); + if (rc) + goto start_all_exit; + + prepare_cpu_buffers(); + + for_each_online_cpu(cpu) { + rc = start_sampling(cpu); + if (rc) + break; + } + if (rc) { + for_each_online_cpu(cpu) { + stop_sampling(cpu); + } + goto start_all_exit; + } + hws_state = HWS_STARTED; + rc = 0; + +start_all_exit: + mutex_unlock(&hws_sem); + + if (rc) + return rc; + + register_oom_notifier(&hws_oom_notifier); + hws_oom = 1; + hws_flush_all = 0; + /* now let them in, 1407 CPUMF external interrupts */ + hws_alert = 1; + irq_subclass_register(IRQ_SUBCLASS_MEASUREMENT_ALERT); + + return 0; +} + +/** + * hwsampler_stop_all() - stop hardware sampling on all online CPUs + * + * Returns 0 on success, !0 on failure. + */ +int hwsampler_stop_all(void) +{ + int tmp_rc, rc, cpu; + struct hws_cpu_buffer *cb; + + mutex_lock(&hws_sem); + + rc = 0; + if (hws_state == HWS_INIT) { + mutex_unlock(&hws_sem); + return 0; + } + hws_state = HWS_STOPPING; + mutex_unlock(&hws_sem); + + for_each_online_cpu(cpu) { + cb = &per_cpu(sampler_cpu_buffer, cpu); + cb->stop_mode = 1; + tmp_rc = stop_sampling(cpu); + if (tmp_rc) + rc = tmp_rc; + } + + if (hws_wq) + flush_workqueue(hws_wq); + + mutex_lock(&hws_sem); + if (hws_oom) { + unregister_oom_notifier(&hws_oom_notifier); + hws_oom = 0; + } + hws_state = HWS_STOPPED; + mutex_unlock(&hws_sem); + + return rc; +} diff --git a/arch/s390/oprofile/hwsampler.h b/arch/s390/oprofile/hwsampler.h new file mode 100644 index 000000000..a483d06f2 --- /dev/null +++ b/arch/s390/oprofile/hwsampler.h @@ -0,0 +1,63 @@ +/* + * CPUMF HW sampler functions and internal structures + * + * Copyright IBM Corp. 2010 + * Author(s): Heinz Graalfs <graalfs@de.ibm.com> + */ + +#ifndef HWSAMPLER_H_ +#define HWSAMPLER_H_ + +#include <linux/workqueue.h> +#include <asm/cpu_mf.h> + +struct hws_ssctl_request_block /* SET SAMPLING CONTROLS req block */ +{ /* bytes 0 - 7 Bit(s) */ + unsigned int s:1; /* 0: maximum buffer indicator */ + unsigned int h:1; /* 1: part. level reserved for VM use*/ + unsigned long b2_53:52; /* 2-53: zeros */ + unsigned int es:1; /* 54: sampling enable control */ + unsigned int b55_61:7; /* 55-61: - zeros */ + unsigned int cs:1; /* 62: sampling activation control */ + unsigned int b63:1; /* 63: zero */ + unsigned long interval; /* 8-15: sampling interval */ + unsigned long tear; /* 16-23: TEAR contents */ + unsigned long dear; /* 24-31: DEAR contents */ + /* 32-63: */ + unsigned long rsvrd1; /* reserved */ + unsigned long rsvrd2; /* reserved */ + unsigned long rsvrd3; /* reserved */ + unsigned long rsvrd4; /* reserved */ +}; + +struct hws_cpu_buffer { + unsigned long first_sdbt; /* @ of 1st SDB-Table for this CP*/ + unsigned long worker_entry; + unsigned long sample_overflow; /* taken from SDB ... */ + struct hws_qsi_info_block qsi; + struct hws_ssctl_request_block ssctl; + struct work_struct worker; + atomic_t ext_params; + unsigned long req_alert; + unsigned long loss_of_sample_data; + unsigned long invalid_entry_address; + unsigned long incorrect_sdbt_entry; + unsigned long sample_auth_change_alert; + unsigned int finish:1; + unsigned int oom:1; + unsigned int stop_mode:1; +}; + +int hwsampler_setup(void); +int hwsampler_shutdown(void); +int hwsampler_allocate(unsigned long sdbt, unsigned long sdb); +int hwsampler_deallocate(void); +unsigned long hwsampler_query_min_interval(void); +unsigned long hwsampler_query_max_interval(void); +int hwsampler_start_all(unsigned long interval); +int hwsampler_stop_all(void); +int hwsampler_deactivate(unsigned int cpu); +int hwsampler_activate(unsigned int cpu); +unsigned long hwsampler_get_sample_overflow_count(unsigned int cpu); + +#endif /*HWSAMPLER_H_*/ diff --git a/arch/s390/oprofile/init.c b/arch/s390/oprofile/init.c new file mode 100644 index 000000000..bc927a09a --- /dev/null +++ b/arch/s390/oprofile/init.c @@ -0,0 +1,513 @@ +/* + * S390 Version + * Copyright IBM Corp. 2002, 2011 + * Author(s): Thomas Spatzier (tspat@de.ibm.com) + * Author(s): Mahesh Salgaonkar (mahesh@linux.vnet.ibm.com) + * Author(s): Heinz Graalfs (graalfs@linux.vnet.ibm.com) + * Author(s): Andreas Krebbel (krebbel@linux.vnet.ibm.com) + * + * @remark Copyright 2002-2011 OProfile authors + */ + +#include <linux/oprofile.h> +#include <linux/perf_event.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <linux/fs.h> +#include <linux/module.h> +#include <asm/processor.h> + +#include "../../../drivers/oprofile/oprof.h" + +extern void s390_backtrace(struct pt_regs * const regs, unsigned int depth); + +#include "hwsampler.h" +#include "op_counter.h" + +#define DEFAULT_INTERVAL 4127518 + +#define DEFAULT_SDBT_BLOCKS 1 +#define DEFAULT_SDB_BLOCKS 511 + +static unsigned long oprofile_hw_interval = DEFAULT_INTERVAL; +static unsigned long oprofile_min_interval; +static unsigned long oprofile_max_interval; + +static unsigned long oprofile_sdbt_blocks = DEFAULT_SDBT_BLOCKS; +static unsigned long oprofile_sdb_blocks = DEFAULT_SDB_BLOCKS; + +static int hwsampler_enabled; +static int hwsampler_running; /* start_mutex must be held to change */ +static int hwsampler_available; + +static struct oprofile_operations timer_ops; + +struct op_counter_config counter_config; + +enum __force_cpu_type { + reserved = 0, /* do not force */ + timer, +}; +static int force_cpu_type; + +static int set_cpu_type(const char *str, struct kernel_param *kp) +{ + if (!strcmp(str, "timer")) { + force_cpu_type = timer; + printk(KERN_INFO "oprofile: forcing timer to be returned " + "as cpu type\n"); + } else { + force_cpu_type = 0; + } + + return 0; +} +module_param_call(cpu_type, set_cpu_type, NULL, NULL, 0); +MODULE_PARM_DESC(cpu_type, "Force legacy basic mode sampling" + "(report cpu_type \"timer\""); + +static int __oprofile_hwsampler_start(void) +{ + int retval; + + retval = hwsampler_allocate(oprofile_sdbt_blocks, oprofile_sdb_blocks); + if (retval) + return retval; + + retval = hwsampler_start_all(oprofile_hw_interval); + if (retval) + hwsampler_deallocate(); + + return retval; +} + +static int oprofile_hwsampler_start(void) +{ + int retval; + + hwsampler_running = hwsampler_enabled; + + if (!hwsampler_running) + return timer_ops.start(); + + retval = perf_reserve_sampling(); + if (retval) + return retval; + + retval = __oprofile_hwsampler_start(); + if (retval) + perf_release_sampling(); + + return retval; +} + +static void oprofile_hwsampler_stop(void) +{ + if (!hwsampler_running) { + timer_ops.stop(); + return; + } + + hwsampler_stop_all(); + hwsampler_deallocate(); + perf_release_sampling(); + return; +} + +/* + * File ops used for: + * /dev/oprofile/0/enabled + * /dev/oprofile/hwsampling/hwsampler (cpu_type = timer) + */ + +static ssize_t hwsampler_read(struct file *file, char __user *buf, + size_t count, loff_t *offset) +{ + return oprofilefs_ulong_to_user(hwsampler_enabled, buf, count, offset); +} + +static ssize_t hwsampler_write(struct file *file, char const __user *buf, + size_t count, loff_t *offset) +{ + unsigned long val; + int retval; + + if (*offset) + return -EINVAL; + + retval = oprofilefs_ulong_from_user(&val, buf, count); + if (retval <= 0) + return retval; + + if (val != 0 && val != 1) + return -EINVAL; + + if (oprofile_started) + /* + * save to do without locking as we set + * hwsampler_running in start() when start_mutex is + * held + */ + return -EBUSY; + + hwsampler_enabled = val; + + return count; +} + +static const struct file_operations hwsampler_fops = { + .read = hwsampler_read, + .write = hwsampler_write, +}; + +/* + * File ops used for: + * /dev/oprofile/0/count + * /dev/oprofile/hwsampling/hw_interval (cpu_type = timer) + * + * Make sure that the value is within the hardware range. + */ + +static ssize_t hw_interval_read(struct file *file, char __user *buf, + size_t count, loff_t *offset) +{ + return oprofilefs_ulong_to_user(oprofile_hw_interval, buf, + count, offset); +} + +static ssize_t hw_interval_write(struct file *file, char const __user *buf, + size_t count, loff_t *offset) +{ + unsigned long val; + int retval; + + if (*offset) + return -EINVAL; + retval = oprofilefs_ulong_from_user(&val, buf, count); + if (retval <= 0) + return retval; + if (val < oprofile_min_interval) + oprofile_hw_interval = oprofile_min_interval; + else if (val > oprofile_max_interval) + oprofile_hw_interval = oprofile_max_interval; + else + oprofile_hw_interval = val; + + return count; +} + +static const struct file_operations hw_interval_fops = { + .read = hw_interval_read, + .write = hw_interval_write, +}; + +/* + * File ops used for: + * /dev/oprofile/0/event + * Only a single event with number 0 is supported with this counter. + * + * /dev/oprofile/0/unit_mask + * This is a dummy file needed by the user space tools. + * No value other than 0 is accepted or returned. + */ + +static ssize_t hwsampler_zero_read(struct file *file, char __user *buf, + size_t count, loff_t *offset) +{ + return oprofilefs_ulong_to_user(0, buf, count, offset); +} + +static ssize_t hwsampler_zero_write(struct file *file, char const __user *buf, + size_t count, loff_t *offset) +{ + unsigned long val; + int retval; + + if (*offset) + return -EINVAL; + + retval = oprofilefs_ulong_from_user(&val, buf, count); + if (retval <= 0) + return retval; + if (val != 0) + return -EINVAL; + return count; +} + +static const struct file_operations zero_fops = { + .read = hwsampler_zero_read, + .write = hwsampler_zero_write, +}; + +/* /dev/oprofile/0/kernel file ops. */ + +static ssize_t hwsampler_kernel_read(struct file *file, char __user *buf, + size_t count, loff_t *offset) +{ + return oprofilefs_ulong_to_user(counter_config.kernel, + buf, count, offset); +} + +static ssize_t hwsampler_kernel_write(struct file *file, char const __user *buf, + size_t count, loff_t *offset) +{ + unsigned long val; + int retval; + + if (*offset) + return -EINVAL; + + retval = oprofilefs_ulong_from_user(&val, buf, count); + if (retval <= 0) + return retval; + + if (val != 0 && val != 1) + return -EINVAL; + + counter_config.kernel = val; + + return count; +} + +static const struct file_operations kernel_fops = { + .read = hwsampler_kernel_read, + .write = hwsampler_kernel_write, +}; + +/* /dev/oprofile/0/user file ops. */ + +static ssize_t hwsampler_user_read(struct file *file, char __user *buf, + size_t count, loff_t *offset) +{ + return oprofilefs_ulong_to_user(counter_config.user, + buf, count, offset); +} + +static ssize_t hwsampler_user_write(struct file *file, char const __user *buf, + size_t count, loff_t *offset) +{ + unsigned long val; + int retval; + + if (*offset) + return -EINVAL; + + retval = oprofilefs_ulong_from_user(&val, buf, count); + if (retval <= 0) + return retval; + + if (val != 0 && val != 1) + return -EINVAL; + + counter_config.user = val; + + return count; +} + +static const struct file_operations user_fops = { + .read = hwsampler_user_read, + .write = hwsampler_user_write, +}; + + +/* + * File ops used for: /dev/oprofile/timer/enabled + * The value always has to be the inverted value of hwsampler_enabled. So + * no separate variable is created. That way we do not need locking. + */ + +static ssize_t timer_enabled_read(struct file *file, char __user *buf, + size_t count, loff_t *offset) +{ + return oprofilefs_ulong_to_user(!hwsampler_enabled, buf, count, offset); +} + +static ssize_t timer_enabled_write(struct file *file, char const __user *buf, + size_t count, loff_t *offset) +{ + unsigned long val; + int retval; + + if (*offset) + return -EINVAL; + + retval = oprofilefs_ulong_from_user(&val, buf, count); + if (retval <= 0) + return retval; + + if (val != 0 && val != 1) + return -EINVAL; + + /* Timer cannot be disabled without having hardware sampling. */ + if (val == 0 && !hwsampler_available) + return -EINVAL; + + if (oprofile_started) + /* + * save to do without locking as we set + * hwsampler_running in start() when start_mutex is + * held + */ + return -EBUSY; + + hwsampler_enabled = !val; + + return count; +} + +static const struct file_operations timer_enabled_fops = { + .read = timer_enabled_read, + .write = timer_enabled_write, +}; + + +static int oprofile_create_hwsampling_files(struct dentry *root) +{ + struct dentry *dir; + + dir = oprofilefs_mkdir(root, "timer"); + if (!dir) + return -EINVAL; + + oprofilefs_create_file(dir, "enabled", &timer_enabled_fops); + + if (!hwsampler_available) + return 0; + + /* reinitialize default values */ + hwsampler_enabled = 1; + counter_config.kernel = 1; + counter_config.user = 1; + + if (!force_cpu_type) { + /* + * Create the counter file system. A single virtual + * counter is created which can be used to + * enable/disable hardware sampling dynamically from + * user space. The user space will configure a single + * counter with a single event. The value of 'event' + * and 'unit_mask' are not evaluated by the kernel code + * and can only be set to 0. + */ + + dir = oprofilefs_mkdir(root, "0"); + if (!dir) + return -EINVAL; + + oprofilefs_create_file(dir, "enabled", &hwsampler_fops); + oprofilefs_create_file(dir, "event", &zero_fops); + oprofilefs_create_file(dir, "count", &hw_interval_fops); + oprofilefs_create_file(dir, "unit_mask", &zero_fops); + oprofilefs_create_file(dir, "kernel", &kernel_fops); + oprofilefs_create_file(dir, "user", &user_fops); + oprofilefs_create_ulong(dir, "hw_sdbt_blocks", + &oprofile_sdbt_blocks); + + } else { + /* + * Hardware sampling can be used but the cpu_type is + * forced to timer in order to deal with legacy user + * space tools. The /dev/oprofile/hwsampling fs is + * provided in that case. + */ + dir = oprofilefs_mkdir(root, "hwsampling"); + if (!dir) + return -EINVAL; + + oprofilefs_create_file(dir, "hwsampler", + &hwsampler_fops); + oprofilefs_create_file(dir, "hw_interval", + &hw_interval_fops); + oprofilefs_create_ro_ulong(dir, "hw_min_interval", + &oprofile_min_interval); + oprofilefs_create_ro_ulong(dir, "hw_max_interval", + &oprofile_max_interval); + oprofilefs_create_ulong(dir, "hw_sdbt_blocks", + &oprofile_sdbt_blocks); + } + return 0; +} + +static int oprofile_hwsampler_init(struct oprofile_operations *ops) +{ + /* + * Initialize the timer mode infrastructure as well in order + * to be able to switch back dynamically. oprofile_timer_init + * is not supposed to fail. + */ + if (oprofile_timer_init(ops)) + BUG(); + + memcpy(&timer_ops, ops, sizeof(timer_ops)); + ops->create_files = oprofile_create_hwsampling_files; + + /* + * If the user space tools do not support newer cpu types, + * the force_cpu_type module parameter + * can be used to always return \"timer\" as cpu type. + */ + if (force_cpu_type != timer) { + struct cpuid id; + + get_cpu_id (&id); + + switch (id.machine) { + case 0x2097: case 0x2098: ops->cpu_type = "s390/z10"; break; + case 0x2817: case 0x2818: ops->cpu_type = "s390/z196"; break; + case 0x2827: case 0x2828: ops->cpu_type = "s390/zEC12"; break; + default: return -ENODEV; + } + } + + if (hwsampler_setup()) + return -ENODEV; + + /* + * Query the range for the sampling interval from the + * hardware. + */ + oprofile_min_interval = hwsampler_query_min_interval(); + if (oprofile_min_interval == 0) + return -ENODEV; + oprofile_max_interval = hwsampler_query_max_interval(); + if (oprofile_max_interval == 0) + return -ENODEV; + + /* The initial value should be sane */ + if (oprofile_hw_interval < oprofile_min_interval) + oprofile_hw_interval = oprofile_min_interval; + if (oprofile_hw_interval > oprofile_max_interval) + oprofile_hw_interval = oprofile_max_interval; + + printk(KERN_INFO "oprofile: System z hardware sampling " + "facility found.\n"); + + ops->start = oprofile_hwsampler_start; + ops->stop = oprofile_hwsampler_stop; + + return 0; +} + +static void oprofile_hwsampler_exit(void) +{ + hwsampler_shutdown(); +} + +int __init oprofile_arch_init(struct oprofile_operations *ops) +{ + ops->backtrace = s390_backtrace; + + /* + * -ENODEV is not reported to the caller. The module itself + * will use the timer mode sampling as fallback and this is + * always available. + */ + hwsampler_available = oprofile_hwsampler_init(ops) == 0; + + return 0; +} + +void oprofile_arch_exit(void) +{ + oprofile_hwsampler_exit(); +} diff --git a/arch/s390/oprofile/op_counter.h b/arch/s390/oprofile/op_counter.h new file mode 100644 index 000000000..61b2531ee --- /dev/null +++ b/arch/s390/oprofile/op_counter.h @@ -0,0 +1,21 @@ +/* + * Copyright IBM Corp. 2011 + * Author(s): Andreas Krebbel (krebbel@linux.vnet.ibm.com) + * + * @remark Copyright 2011 OProfile authors + */ + +#ifndef OP_COUNTER_H +#define OP_COUNTER_H + +struct op_counter_config { + /* `enabled' maps to the hwsampler_file variable. */ + /* `count' maps to the oprofile_hw_interval variable. */ + /* `event' and `unit_mask' are unused. */ + unsigned long kernel; + unsigned long user; +}; + +extern struct op_counter_config counter_config; + +#endif /* OP_COUNTER_H */ diff --git a/arch/s390/pci/Makefile b/arch/s390/pci/Makefile new file mode 100644 index 000000000..805d8b291 --- /dev/null +++ b/arch/s390/pci/Makefile @@ -0,0 +1,6 @@ +# +# Makefile for the s390 PCI subsystem. +# + +obj-$(CONFIG_PCI) += pci.o pci_dma.o pci_clp.o pci_sysfs.o \ + pci_event.o pci_debug.o pci_insn.o pci_mmio.o diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c new file mode 100644 index 000000000..598f023cf --- /dev/null +++ b/arch/s390/pci/pci.c @@ -0,0 +1,965 @@ +/* + * Copyright IBM Corp. 2012 + * + * Author(s): + * Jan Glauber <jang@linux.vnet.ibm.com> + * + * The System z PCI code is a rewrite from a prototype by + * the following people (Kudoz!): + * Alexander Schmidt + * Christoph Raisch + * Hannes Hering + * Hoang-Nam Nguyen + * Jan-Bernd Themann + * Stefan Roscher + * Thomas Klein + */ + +#define KMSG_COMPONENT "zpci" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/err.h> +#include <linux/export.h> +#include <linux/delay.h> +#include <linux/irq.h> +#include <linux/kernel_stat.h> +#include <linux/seq_file.h> +#include <linux/pci.h> +#include <linux/msi.h> + +#include <asm/isc.h> +#include <asm/airq.h> +#include <asm/facility.h> +#include <asm/pci_insn.h> +#include <asm/pci_clp.h> +#include <asm/pci_dma.h> + +#define DEBUG /* enable pr_debug */ + +#define SIC_IRQ_MODE_ALL 0 +#define SIC_IRQ_MODE_SINGLE 1 + +#define ZPCI_NR_DMA_SPACES 1 +#define ZPCI_NR_DEVICES CONFIG_PCI_NR_FUNCTIONS + +/* list of all detected zpci devices */ +static LIST_HEAD(zpci_list); +static DEFINE_SPINLOCK(zpci_list_lock); + +static struct irq_chip zpci_irq_chip = { + .name = "zPCI", + .irq_unmask = pci_msi_unmask_irq, + .irq_mask = pci_msi_mask_irq, +}; + +static DECLARE_BITMAP(zpci_domain, ZPCI_NR_DEVICES); +static DEFINE_SPINLOCK(zpci_domain_lock); + +static struct airq_iv *zpci_aisb_iv; +static struct airq_iv *zpci_aibv[ZPCI_NR_DEVICES]; + +/* Adapter interrupt definitions */ +static void zpci_irq_handler(struct airq_struct *airq); + +static struct airq_struct zpci_airq = { + .handler = zpci_irq_handler, + .isc = PCI_ISC, +}; + +/* I/O Map */ +static DEFINE_SPINLOCK(zpci_iomap_lock); +static DECLARE_BITMAP(zpci_iomap, ZPCI_IOMAP_MAX_ENTRIES); +struct zpci_iomap_entry *zpci_iomap_start; +EXPORT_SYMBOL_GPL(zpci_iomap_start); + +static struct kmem_cache *zdev_fmb_cache; + +struct zpci_dev *get_zdev(struct pci_dev *pdev) +{ + return (struct zpci_dev *) pdev->sysdata; +} + +struct zpci_dev *get_zdev_by_fid(u32 fid) +{ + struct zpci_dev *tmp, *zdev = NULL; + + spin_lock(&zpci_list_lock); + list_for_each_entry(tmp, &zpci_list, entry) { + if (tmp->fid == fid) { + zdev = tmp; + break; + } + } + spin_unlock(&zpci_list_lock); + return zdev; +} + +static struct zpci_dev *get_zdev_by_bus(struct pci_bus *bus) +{ + return (bus && bus->sysdata) ? (struct zpci_dev *) bus->sysdata : NULL; +} + +int pci_domain_nr(struct pci_bus *bus) +{ + return ((struct zpci_dev *) bus->sysdata)->domain; +} +EXPORT_SYMBOL_GPL(pci_domain_nr); + +int pci_proc_domain(struct pci_bus *bus) +{ + return pci_domain_nr(bus); +} +EXPORT_SYMBOL_GPL(pci_proc_domain); + +/* Modify PCI: Register adapter interruptions */ +static int zpci_set_airq(struct zpci_dev *zdev) +{ + u64 req = ZPCI_CREATE_REQ(zdev->fh, 0, ZPCI_MOD_FC_REG_INT); + struct zpci_fib fib = {0}; + + fib.isc = PCI_ISC; + fib.sum = 1; /* enable summary notifications */ + fib.noi = airq_iv_end(zdev->aibv); + fib.aibv = (unsigned long) zdev->aibv->vector; + fib.aibvo = 0; /* each zdev has its own interrupt vector */ + fib.aisb = (unsigned long) zpci_aisb_iv->vector + (zdev->aisb/64)*8; + fib.aisbo = zdev->aisb & 63; + + return zpci_mod_fc(req, &fib); +} + +struct mod_pci_args { + u64 base; + u64 limit; + u64 iota; + u64 fmb_addr; +}; + +static int mod_pci(struct zpci_dev *zdev, int fn, u8 dmaas, struct mod_pci_args *args) +{ + u64 req = ZPCI_CREATE_REQ(zdev->fh, dmaas, fn); + struct zpci_fib fib = {0}; + + fib.pba = args->base; + fib.pal = args->limit; + fib.iota = args->iota; + fib.fmb_addr = args->fmb_addr; + + return zpci_mod_fc(req, &fib); +} + +/* Modify PCI: Register I/O address translation parameters */ +int zpci_register_ioat(struct zpci_dev *zdev, u8 dmaas, + u64 base, u64 limit, u64 iota) +{ + struct mod_pci_args args = { base, limit, iota, 0 }; + + WARN_ON_ONCE(iota & 0x3fff); + args.iota |= ZPCI_IOTA_RTTO_FLAG; + return mod_pci(zdev, ZPCI_MOD_FC_REG_IOAT, dmaas, &args); +} + +/* Modify PCI: Unregister I/O address translation parameters */ +int zpci_unregister_ioat(struct zpci_dev *zdev, u8 dmaas) +{ + struct mod_pci_args args = { 0, 0, 0, 0 }; + + return mod_pci(zdev, ZPCI_MOD_FC_DEREG_IOAT, dmaas, &args); +} + +/* Modify PCI: Unregister adapter interruptions */ +static int zpci_clear_airq(struct zpci_dev *zdev) +{ + struct mod_pci_args args = { 0, 0, 0, 0 }; + + return mod_pci(zdev, ZPCI_MOD_FC_DEREG_INT, 0, &args); +} + +/* Modify PCI: Set PCI function measurement parameters */ +int zpci_fmb_enable_device(struct zpci_dev *zdev) +{ + struct mod_pci_args args = { 0, 0, 0, 0 }; + + if (zdev->fmb) + return -EINVAL; + + zdev->fmb = kmem_cache_zalloc(zdev_fmb_cache, GFP_KERNEL); + if (!zdev->fmb) + return -ENOMEM; + WARN_ON((u64) zdev->fmb & 0xf); + + /* reset software counters */ + atomic64_set(&zdev->allocated_pages, 0); + atomic64_set(&zdev->mapped_pages, 0); + atomic64_set(&zdev->unmapped_pages, 0); + + args.fmb_addr = virt_to_phys(zdev->fmb); + return mod_pci(zdev, ZPCI_MOD_FC_SET_MEASURE, 0, &args); +} + +/* Modify PCI: Disable PCI function measurement */ +int zpci_fmb_disable_device(struct zpci_dev *zdev) +{ + struct mod_pci_args args = { 0, 0, 0, 0 }; + int rc; + + if (!zdev->fmb) + return -EINVAL; + + /* Function measurement is disabled if fmb address is zero */ + rc = mod_pci(zdev, ZPCI_MOD_FC_SET_MEASURE, 0, &args); + + kmem_cache_free(zdev_fmb_cache, zdev->fmb); + zdev->fmb = NULL; + return rc; +} + +#define ZPCI_PCIAS_CFGSPC 15 + +static int zpci_cfg_load(struct zpci_dev *zdev, int offset, u32 *val, u8 len) +{ + u64 req = ZPCI_CREATE_REQ(zdev->fh, ZPCI_PCIAS_CFGSPC, len); + u64 data; + int rc; + + rc = zpci_load(&data, req, offset); + if (!rc) { + data = data << ((8 - len) * 8); + data = le64_to_cpu(data); + *val = (u32) data; + } else + *val = 0xffffffff; + return rc; +} + +static int zpci_cfg_store(struct zpci_dev *zdev, int offset, u32 val, u8 len) +{ + u64 req = ZPCI_CREATE_REQ(zdev->fh, ZPCI_PCIAS_CFGSPC, len); + u64 data = val; + int rc; + + data = cpu_to_le64(data); + data = data >> ((8 - len) * 8); + rc = zpci_store(data, req, offset); + return rc; +} + +void pcibios_fixup_bus(struct pci_bus *bus) +{ +} + +resource_size_t pcibios_align_resource(void *data, const struct resource *res, + resource_size_t size, + resource_size_t align) +{ + return 0; +} + +/* combine single writes by using store-block insn */ +void __iowrite64_copy(void __iomem *to, const void *from, size_t count) +{ + zpci_memcpy_toio(to, from, count); +} + +/* Create a virtual mapping cookie for a PCI BAR */ +void __iomem *pci_iomap_range(struct pci_dev *pdev, + int bar, + unsigned long offset, + unsigned long max) +{ + struct zpci_dev *zdev = get_zdev(pdev); + u64 addr; + int idx; + + if ((bar & 7) != bar) + return NULL; + + idx = zdev->bars[bar].map_idx; + spin_lock(&zpci_iomap_lock); + if (zpci_iomap_start[idx].count++) { + BUG_ON(zpci_iomap_start[idx].fh != zdev->fh || + zpci_iomap_start[idx].bar != bar); + } else { + zpci_iomap_start[idx].fh = zdev->fh; + zpci_iomap_start[idx].bar = bar; + } + /* Detect overrun */ + BUG_ON(!zpci_iomap_start[idx].count); + spin_unlock(&zpci_iomap_lock); + + addr = ZPCI_IOMAP_ADDR_BASE | ((u64) idx << 48); + return (void __iomem *) addr + offset; +} +EXPORT_SYMBOL(pci_iomap_range); + +void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) +{ + return pci_iomap_range(dev, bar, 0, maxlen); +} +EXPORT_SYMBOL(pci_iomap); + +void pci_iounmap(struct pci_dev *pdev, void __iomem *addr) +{ + unsigned int idx; + + idx = (((__force u64) addr) & ~ZPCI_IOMAP_ADDR_BASE) >> 48; + spin_lock(&zpci_iomap_lock); + /* Detect underrun */ + BUG_ON(!zpci_iomap_start[idx].count); + if (!--zpci_iomap_start[idx].count) { + zpci_iomap_start[idx].fh = 0; + zpci_iomap_start[idx].bar = 0; + } + spin_unlock(&zpci_iomap_lock); +} +EXPORT_SYMBOL(pci_iounmap); + +static int pci_read(struct pci_bus *bus, unsigned int devfn, int where, + int size, u32 *val) +{ + struct zpci_dev *zdev = get_zdev_by_bus(bus); + int ret; + + if (!zdev || devfn != ZPCI_DEVFN) + ret = -ENODEV; + else + ret = zpci_cfg_load(zdev, where, val, size); + + return ret; +} + +static int pci_write(struct pci_bus *bus, unsigned int devfn, int where, + int size, u32 val) +{ + struct zpci_dev *zdev = get_zdev_by_bus(bus); + int ret; + + if (!zdev || devfn != ZPCI_DEVFN) + ret = -ENODEV; + else + ret = zpci_cfg_store(zdev, where, val, size); + + return ret; +} + +static struct pci_ops pci_root_ops = { + .read = pci_read, + .write = pci_write, +}; + +static void zpci_irq_handler(struct airq_struct *airq) +{ + unsigned long si, ai; + struct airq_iv *aibv; + int irqs_on = 0; + + inc_irq_stat(IRQIO_PCI); + for (si = 0;;) { + /* Scan adapter summary indicator bit vector */ + si = airq_iv_scan(zpci_aisb_iv, si, airq_iv_end(zpci_aisb_iv)); + if (si == -1UL) { + if (irqs_on++) + /* End of second scan with interrupts on. */ + break; + /* First scan complete, reenable interrupts. */ + zpci_set_irq_ctrl(SIC_IRQ_MODE_SINGLE, NULL, PCI_ISC); + si = 0; + continue; + } + + /* Scan the adapter interrupt vector for this device. */ + aibv = zpci_aibv[si]; + for (ai = 0;;) { + ai = airq_iv_scan(aibv, ai, airq_iv_end(aibv)); + if (ai == -1UL) + break; + inc_irq_stat(IRQIO_MSI); + airq_iv_lock(aibv, ai); + generic_handle_irq(airq_iv_get_data(aibv, ai)); + airq_iv_unlock(aibv, ai); + } + } +} + +int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) +{ + struct zpci_dev *zdev = get_zdev(pdev); + unsigned int hwirq, msi_vecs; + unsigned long aisb; + struct msi_desc *msi; + struct msi_msg msg; + int rc, irq; + + if (type == PCI_CAP_ID_MSI && nvec > 1) + return 1; + msi_vecs = min_t(unsigned int, nvec, zdev->max_msi); + + /* Allocate adapter summary indicator bit */ + rc = -EIO; + aisb = airq_iv_alloc_bit(zpci_aisb_iv); + if (aisb == -1UL) + goto out; + zdev->aisb = aisb; + + /* Create adapter interrupt vector */ + rc = -ENOMEM; + zdev->aibv = airq_iv_create(msi_vecs, AIRQ_IV_DATA | AIRQ_IV_BITLOCK); + if (!zdev->aibv) + goto out_si; + + /* Wire up shortcut pointer */ + zpci_aibv[aisb] = zdev->aibv; + + /* Request MSI interrupts */ + hwirq = 0; + list_for_each_entry(msi, &pdev->msi_list, list) { + rc = -EIO; + irq = irq_alloc_desc(0); /* Alloc irq on node 0 */ + if (irq < 0) + goto out_msi; + rc = irq_set_msi_desc(irq, msi); + if (rc) + goto out_msi; + irq_set_chip_and_handler(irq, &zpci_irq_chip, + handle_simple_irq); + msg.data = hwirq; + msg.address_lo = zdev->msi_addr & 0xffffffff; + msg.address_hi = zdev->msi_addr >> 32; + pci_write_msi_msg(irq, &msg); + airq_iv_set_data(zdev->aibv, hwirq, irq); + hwirq++; + } + + /* Enable adapter interrupts */ + rc = zpci_set_airq(zdev); + if (rc) + goto out_msi; + + return (msi_vecs == nvec) ? 0 : msi_vecs; + +out_msi: + list_for_each_entry(msi, &pdev->msi_list, list) { + if (hwirq-- == 0) + break; + irq_set_msi_desc(msi->irq, NULL); + irq_free_desc(msi->irq); + msi->msg.address_lo = 0; + msi->msg.address_hi = 0; + msi->msg.data = 0; + msi->irq = 0; + } + zpci_aibv[aisb] = NULL; + airq_iv_release(zdev->aibv); +out_si: + airq_iv_free_bit(zpci_aisb_iv, aisb); +out: + return rc; +} + +void arch_teardown_msi_irqs(struct pci_dev *pdev) +{ + struct zpci_dev *zdev = get_zdev(pdev); + struct msi_desc *msi; + int rc; + + /* Disable adapter interrupts */ + rc = zpci_clear_airq(zdev); + if (rc) + return; + + /* Release MSI interrupts */ + list_for_each_entry(msi, &pdev->msi_list, list) { + if (msi->msi_attrib.is_msix) + __pci_msix_desc_mask_irq(msi, 1); + else + __pci_msi_desc_mask_irq(msi, 1, 1); + irq_set_msi_desc(msi->irq, NULL); + irq_free_desc(msi->irq); + msi->msg.address_lo = 0; + msi->msg.address_hi = 0; + msi->msg.data = 0; + msi->irq = 0; + } + + zpci_aibv[zdev->aisb] = NULL; + airq_iv_release(zdev->aibv); + airq_iv_free_bit(zpci_aisb_iv, zdev->aisb); +} + +static void zpci_map_resources(struct pci_dev *pdev) +{ + resource_size_t len; + int i; + + for (i = 0; i < PCI_BAR_COUNT; i++) { + len = pci_resource_len(pdev, i); + if (!len) + continue; + pdev->resource[i].start = + (resource_size_t __force) pci_iomap(pdev, i, 0); + pdev->resource[i].end = pdev->resource[i].start + len - 1; + } +} + +static void zpci_unmap_resources(struct pci_dev *pdev) +{ + resource_size_t len; + int i; + + for (i = 0; i < PCI_BAR_COUNT; i++) { + len = pci_resource_len(pdev, i); + if (!len) + continue; + pci_iounmap(pdev, (void __iomem __force *) + pdev->resource[i].start); + } +} + +static int __init zpci_irq_init(void) +{ + int rc; + + rc = register_adapter_interrupt(&zpci_airq); + if (rc) + goto out; + /* Set summary to 1 to be called every time for the ISC. */ + *zpci_airq.lsi_ptr = 1; + + rc = -ENOMEM; + zpci_aisb_iv = airq_iv_create(ZPCI_NR_DEVICES, AIRQ_IV_ALLOC); + if (!zpci_aisb_iv) + goto out_airq; + + zpci_set_irq_ctrl(SIC_IRQ_MODE_SINGLE, NULL, PCI_ISC); + return 0; + +out_airq: + unregister_adapter_interrupt(&zpci_airq); +out: + return rc; +} + +static void zpci_irq_exit(void) +{ + airq_iv_release(zpci_aisb_iv); + unregister_adapter_interrupt(&zpci_airq); +} + +static int zpci_alloc_iomap(struct zpci_dev *zdev) +{ + int entry; + + spin_lock(&zpci_iomap_lock); + entry = find_first_zero_bit(zpci_iomap, ZPCI_IOMAP_MAX_ENTRIES); + if (entry == ZPCI_IOMAP_MAX_ENTRIES) { + spin_unlock(&zpci_iomap_lock); + return -ENOSPC; + } + set_bit(entry, zpci_iomap); + spin_unlock(&zpci_iomap_lock); + return entry; +} + +static void zpci_free_iomap(struct zpci_dev *zdev, int entry) +{ + spin_lock(&zpci_iomap_lock); + memset(&zpci_iomap_start[entry], 0, sizeof(struct zpci_iomap_entry)); + clear_bit(entry, zpci_iomap); + spin_unlock(&zpci_iomap_lock); +} + +static struct resource *__alloc_res(struct zpci_dev *zdev, unsigned long start, + unsigned long size, unsigned long flags) +{ + struct resource *r; + + r = kzalloc(sizeof(*r), GFP_KERNEL); + if (!r) + return NULL; + + r->start = start; + r->end = r->start + size - 1; + r->flags = flags; + r->name = zdev->res_name; + + if (request_resource(&iomem_resource, r)) { + kfree(r); + return NULL; + } + return r; +} + +static int zpci_setup_bus_resources(struct zpci_dev *zdev, + struct list_head *resources) +{ + unsigned long addr, size, flags; + struct resource *res; + int i, entry; + + snprintf(zdev->res_name, sizeof(zdev->res_name), + "PCI Bus %04x:%02x", zdev->domain, ZPCI_BUS_NR); + + for (i = 0; i < PCI_BAR_COUNT; i++) { + if (!zdev->bars[i].size) + continue; + entry = zpci_alloc_iomap(zdev); + if (entry < 0) + return entry; + zdev->bars[i].map_idx = entry; + + /* only MMIO is supported */ + flags = IORESOURCE_MEM; + if (zdev->bars[i].val & 8) + flags |= IORESOURCE_PREFETCH; + if (zdev->bars[i].val & 4) + flags |= IORESOURCE_MEM_64; + + addr = ZPCI_IOMAP_ADDR_BASE + ((u64) entry << 48); + + size = 1UL << zdev->bars[i].size; + + res = __alloc_res(zdev, addr, size, flags); + if (!res) { + zpci_free_iomap(zdev, entry); + return -ENOMEM; + } + zdev->bars[i].res = res; + pci_add_resource(resources, res); + } + + return 0; +} + +static void zpci_cleanup_bus_resources(struct zpci_dev *zdev) +{ + int i; + + for (i = 0; i < PCI_BAR_COUNT; i++) { + if (!zdev->bars[i].size) + continue; + + zpci_free_iomap(zdev, zdev->bars[i].map_idx); + release_resource(zdev->bars[i].res); + kfree(zdev->bars[i].res); + } +} + +int pcibios_add_device(struct pci_dev *pdev) +{ + struct zpci_dev *zdev = get_zdev(pdev); + struct resource *res; + int i; + + zdev->pdev = pdev; + pdev->dev.groups = zpci_attr_groups; + zpci_map_resources(pdev); + + for (i = 0; i < PCI_BAR_COUNT; i++) { + res = &pdev->resource[i]; + if (res->parent || !res->flags) + continue; + pci_claim_resource(pdev, i); + } + + return 0; +} + +void pcibios_release_device(struct pci_dev *pdev) +{ + zpci_unmap_resources(pdev); +} + +int pcibios_enable_device(struct pci_dev *pdev, int mask) +{ + struct zpci_dev *zdev = get_zdev(pdev); + + zdev->pdev = pdev; + zpci_debug_init_device(zdev); + zpci_fmb_enable_device(zdev); + + return pci_enable_resources(pdev, mask); +} + +void pcibios_disable_device(struct pci_dev *pdev) +{ + struct zpci_dev *zdev = get_zdev(pdev); + + zpci_fmb_disable_device(zdev); + zpci_debug_exit_device(zdev); + zdev->pdev = NULL; +} + +#ifdef CONFIG_HIBERNATE_CALLBACKS +static int zpci_restore(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct zpci_dev *zdev = get_zdev(pdev); + int ret = 0; + + if (zdev->state != ZPCI_FN_STATE_ONLINE) + goto out; + + ret = clp_enable_fh(zdev, ZPCI_NR_DMA_SPACES); + if (ret) + goto out; + + zpci_map_resources(pdev); + zpci_register_ioat(zdev, 0, zdev->start_dma + PAGE_OFFSET, + zdev->start_dma + zdev->iommu_size - 1, + (u64) zdev->dma_table); + +out: + return ret; +} + +static int zpci_freeze(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct zpci_dev *zdev = get_zdev(pdev); + + if (zdev->state != ZPCI_FN_STATE_ONLINE) + return 0; + + zpci_unregister_ioat(zdev, 0); + zpci_unmap_resources(pdev); + return clp_disable_fh(zdev); +} + +struct dev_pm_ops pcibios_pm_ops = { + .thaw_noirq = zpci_restore, + .freeze_noirq = zpci_freeze, + .restore_noirq = zpci_restore, + .poweroff_noirq = zpci_freeze, +}; +#endif /* CONFIG_HIBERNATE_CALLBACKS */ + +static int zpci_alloc_domain(struct zpci_dev *zdev) +{ + spin_lock(&zpci_domain_lock); + zdev->domain = find_first_zero_bit(zpci_domain, ZPCI_NR_DEVICES); + if (zdev->domain == ZPCI_NR_DEVICES) { + spin_unlock(&zpci_domain_lock); + return -ENOSPC; + } + set_bit(zdev->domain, zpci_domain); + spin_unlock(&zpci_domain_lock); + return 0; +} + +static void zpci_free_domain(struct zpci_dev *zdev) +{ + spin_lock(&zpci_domain_lock); + clear_bit(zdev->domain, zpci_domain); + spin_unlock(&zpci_domain_lock); +} + +void pcibios_remove_bus(struct pci_bus *bus) +{ + struct zpci_dev *zdev = get_zdev_by_bus(bus); + + zpci_exit_slot(zdev); + zpci_cleanup_bus_resources(zdev); + zpci_free_domain(zdev); + + spin_lock(&zpci_list_lock); + list_del(&zdev->entry); + spin_unlock(&zpci_list_lock); + + kfree(zdev); +} + +static int zpci_scan_bus(struct zpci_dev *zdev) +{ + LIST_HEAD(resources); + int ret; + + ret = zpci_setup_bus_resources(zdev, &resources); + if (ret) + return ret; + + zdev->bus = pci_scan_root_bus(NULL, ZPCI_BUS_NR, &pci_root_ops, + zdev, &resources); + if (!zdev->bus) { + zpci_cleanup_bus_resources(zdev); + return -EIO; + } + zdev->bus->max_bus_speed = zdev->max_bus_speed; + pci_bus_add_devices(zdev->bus); + return 0; +} + +int zpci_enable_device(struct zpci_dev *zdev) +{ + int rc; + + rc = clp_enable_fh(zdev, ZPCI_NR_DMA_SPACES); + if (rc) + goto out; + + rc = zpci_dma_init_device(zdev); + if (rc) + goto out_dma; + + zdev->state = ZPCI_FN_STATE_ONLINE; + return 0; + +out_dma: + clp_disable_fh(zdev); +out: + return rc; +} +EXPORT_SYMBOL_GPL(zpci_enable_device); + +int zpci_disable_device(struct zpci_dev *zdev) +{ + zpci_dma_exit_device(zdev); + return clp_disable_fh(zdev); +} +EXPORT_SYMBOL_GPL(zpci_disable_device); + +int zpci_create_device(struct zpci_dev *zdev) +{ + int rc; + + rc = zpci_alloc_domain(zdev); + if (rc) + goto out; + + mutex_init(&zdev->lock); + if (zdev->state == ZPCI_FN_STATE_CONFIGURED) { + rc = zpci_enable_device(zdev); + if (rc) + goto out_free; + } + rc = zpci_scan_bus(zdev); + if (rc) + goto out_disable; + + spin_lock(&zpci_list_lock); + list_add_tail(&zdev->entry, &zpci_list); + spin_unlock(&zpci_list_lock); + + zpci_init_slot(zdev); + + return 0; + +out_disable: + if (zdev->state == ZPCI_FN_STATE_ONLINE) + zpci_disable_device(zdev); +out_free: + zpci_free_domain(zdev); +out: + return rc; +} + +void zpci_stop_device(struct zpci_dev *zdev) +{ + zpci_dma_exit_device(zdev); + /* + * Note: SCLP disables fh via set-pci-fn so don't + * do that here. + */ +} +EXPORT_SYMBOL_GPL(zpci_stop_device); + +static inline int barsize(u8 size) +{ + return (size) ? (1 << size) >> 10 : 0; +} + +static int zpci_mem_init(void) +{ + zdev_fmb_cache = kmem_cache_create("PCI_FMB_cache", sizeof(struct zpci_fmb), + 16, 0, NULL); + if (!zdev_fmb_cache) + goto error_zdev; + + /* TODO: use realloc */ + zpci_iomap_start = kzalloc(ZPCI_IOMAP_MAX_ENTRIES * sizeof(*zpci_iomap_start), + GFP_KERNEL); + if (!zpci_iomap_start) + goto error_iomap; + return 0; + +error_iomap: + kmem_cache_destroy(zdev_fmb_cache); +error_zdev: + return -ENOMEM; +} + +static void zpci_mem_exit(void) +{ + kfree(zpci_iomap_start); + kmem_cache_destroy(zdev_fmb_cache); +} + +static unsigned int s390_pci_probe = 1; +static unsigned int s390_pci_initialized; + +char * __init pcibios_setup(char *str) +{ + if (!strcmp(str, "off")) { + s390_pci_probe = 0; + return NULL; + } + return str; +} + +bool zpci_is_enabled(void) +{ + return s390_pci_initialized; +} + +static int __init pci_base_init(void) +{ + int rc; + + if (!s390_pci_probe) + return 0; + + if (!test_facility(69) || !test_facility(71) || !test_facility(72)) + return 0; + + rc = zpci_debug_init(); + if (rc) + goto out; + + rc = zpci_mem_init(); + if (rc) + goto out_mem; + + rc = zpci_irq_init(); + if (rc) + goto out_irq; + + rc = zpci_dma_init(); + if (rc) + goto out_dma; + + rc = clp_scan_pci_devices(); + if (rc) + goto out_find; + + s390_pci_initialized = 1; + return 0; + +out_find: + zpci_dma_exit(); +out_dma: + zpci_irq_exit(); +out_irq: + zpci_mem_exit(); +out_mem: + zpci_debug_exit(); +out: + return rc; +} +subsys_initcall_sync(pci_base_init); + +void zpci_rescan(void) +{ + if (zpci_is_enabled()) + clp_rescan_pci_devices_simple(); +} diff --git a/arch/s390/pci/pci_clp.c b/arch/s390/pci/pci_clp.c new file mode 100644 index 000000000..d6e411ed8 --- /dev/null +++ b/arch/s390/pci/pci_clp.c @@ -0,0 +1,393 @@ +/* + * Copyright IBM Corp. 2012 + * + * Author(s): + * Jan Glauber <jang@linux.vnet.ibm.com> + */ + +#define KMSG_COMPONENT "zpci" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/err.h> +#include <linux/delay.h> +#include <linux/pci.h> +#include <asm/pci_debug.h> +#include <asm/pci_clp.h> + +static inline void zpci_err_clp(unsigned int rsp, int rc) +{ + struct { + unsigned int rsp; + int rc; + } __packed data = {rsp, rc}; + + zpci_err_hex(&data, sizeof(data)); +} + +/* + * Call Logical Processor + * Retry logic is handled by the caller. + */ +static inline u8 clp_instr(void *data) +{ + struct { u8 _[CLP_BLK_SIZE]; } *req = data; + u64 ignored; + u8 cc; + + asm volatile ( + " .insn rrf,0xb9a00000,%[ign],%[req],0x0,0x2\n" + " ipm %[cc]\n" + " srl %[cc],28\n" + : [cc] "=d" (cc), [ign] "=d" (ignored), "+m" (*req) + : [req] "a" (req) + : "cc"); + return cc; +} + +static void *clp_alloc_block(gfp_t gfp_mask) +{ + return (void *) __get_free_pages(gfp_mask, get_order(CLP_BLK_SIZE)); +} + +static void clp_free_block(void *ptr) +{ + free_pages((unsigned long) ptr, get_order(CLP_BLK_SIZE)); +} + +static void clp_store_query_pci_fngrp(struct zpci_dev *zdev, + struct clp_rsp_query_pci_grp *response) +{ + zdev->tlb_refresh = response->refresh; + zdev->dma_mask = response->dasm; + zdev->msi_addr = response->msia; + zdev->max_msi = response->noi; + zdev->fmb_update = response->mui; + + switch (response->version) { + case 1: + zdev->max_bus_speed = PCIE_SPEED_5_0GT; + break; + default: + zdev->max_bus_speed = PCI_SPEED_UNKNOWN; + break; + } +} + +static int clp_query_pci_fngrp(struct zpci_dev *zdev, u8 pfgid) +{ + struct clp_req_rsp_query_pci_grp *rrb; + int rc; + + rrb = clp_alloc_block(GFP_KERNEL); + if (!rrb) + return -ENOMEM; + + memset(rrb, 0, sizeof(*rrb)); + rrb->request.hdr.len = sizeof(rrb->request); + rrb->request.hdr.cmd = CLP_QUERY_PCI_FNGRP; + rrb->response.hdr.len = sizeof(rrb->response); + rrb->request.pfgid = pfgid; + + rc = clp_instr(rrb); + if (!rc && rrb->response.hdr.rsp == CLP_RC_OK) + clp_store_query_pci_fngrp(zdev, &rrb->response); + else { + zpci_err("Q PCI FGRP:\n"); + zpci_err_clp(rrb->response.hdr.rsp, rc); + rc = -EIO; + } + clp_free_block(rrb); + return rc; +} + +static int clp_store_query_pci_fn(struct zpci_dev *zdev, + struct clp_rsp_query_pci *response) +{ + int i; + + for (i = 0; i < PCI_BAR_COUNT; i++) { + zdev->bars[i].val = le32_to_cpu(response->bar[i]); + zdev->bars[i].size = response->bar_size[i]; + } + zdev->start_dma = response->sdma; + zdev->end_dma = response->edma; + zdev->pchid = response->pchid; + zdev->pfgid = response->pfgid; + zdev->pft = response->pft; + zdev->vfn = response->vfn; + zdev->uid = response->uid; + + memcpy(zdev->pfip, response->pfip, sizeof(zdev->pfip)); + if (response->util_str_avail) { + memcpy(zdev->util_str, response->util_str, + sizeof(zdev->util_str)); + } + + return 0; +} + +static int clp_query_pci_fn(struct zpci_dev *zdev, u32 fh) +{ + struct clp_req_rsp_query_pci *rrb; + int rc; + + rrb = clp_alloc_block(GFP_KERNEL); + if (!rrb) + return -ENOMEM; + + memset(rrb, 0, sizeof(*rrb)); + rrb->request.hdr.len = sizeof(rrb->request); + rrb->request.hdr.cmd = CLP_QUERY_PCI_FN; + rrb->response.hdr.len = sizeof(rrb->response); + rrb->request.fh = fh; + + rc = clp_instr(rrb); + if (!rc && rrb->response.hdr.rsp == CLP_RC_OK) { + rc = clp_store_query_pci_fn(zdev, &rrb->response); + if (rc) + goto out; + if (rrb->response.pfgid) + rc = clp_query_pci_fngrp(zdev, rrb->response.pfgid); + } else { + zpci_err("Q PCI FN:\n"); + zpci_err_clp(rrb->response.hdr.rsp, rc); + rc = -EIO; + } +out: + clp_free_block(rrb); + return rc; +} + +int clp_add_pci_device(u32 fid, u32 fh, int configured) +{ + struct zpci_dev *zdev; + int rc; + + zpci_dbg(3, "add fid:%x, fh:%x, c:%d\n", fid, fh, configured); + zdev = kzalloc(sizeof(*zdev), GFP_KERNEL); + if (!zdev) + return -ENOMEM; + + zdev->fh = fh; + zdev->fid = fid; + + /* Query function properties and update zdev */ + rc = clp_query_pci_fn(zdev, fh); + if (rc) + goto error; + + if (configured) + zdev->state = ZPCI_FN_STATE_CONFIGURED; + else + zdev->state = ZPCI_FN_STATE_STANDBY; + + rc = zpci_create_device(zdev); + if (rc) + goto error; + return 0; + +error: + kfree(zdev); + return rc; +} + +/* + * Enable/Disable a given PCI function defined by its function handle. + */ +static int clp_set_pci_fn(u32 *fh, u8 nr_dma_as, u8 command) +{ + struct clp_req_rsp_set_pci *rrb; + int rc, retries = 100; + + rrb = clp_alloc_block(GFP_KERNEL); + if (!rrb) + return -ENOMEM; + + do { + memset(rrb, 0, sizeof(*rrb)); + rrb->request.hdr.len = sizeof(rrb->request); + rrb->request.hdr.cmd = CLP_SET_PCI_FN; + rrb->response.hdr.len = sizeof(rrb->response); + rrb->request.fh = *fh; + rrb->request.oc = command; + rrb->request.ndas = nr_dma_as; + + rc = clp_instr(rrb); + if (rrb->response.hdr.rsp == CLP_RC_SETPCIFN_BUSY) { + retries--; + if (retries < 0) + break; + msleep(20); + } + } while (rrb->response.hdr.rsp == CLP_RC_SETPCIFN_BUSY); + + if (!rc && rrb->response.hdr.rsp == CLP_RC_OK) + *fh = rrb->response.fh; + else { + zpci_err("Set PCI FN:\n"); + zpci_err_clp(rrb->response.hdr.rsp, rc); + rc = -EIO; + } + clp_free_block(rrb); + return rc; +} + +int clp_enable_fh(struct zpci_dev *zdev, u8 nr_dma_as) +{ + u32 fh = zdev->fh; + int rc; + + rc = clp_set_pci_fn(&fh, nr_dma_as, CLP_SET_ENABLE_PCI_FN); + if (!rc) + /* Success -> store enabled handle in zdev */ + zdev->fh = fh; + + zpci_dbg(3, "ena fid:%x, fh:%x, rc:%d\n", zdev->fid, zdev->fh, rc); + return rc; +} + +int clp_disable_fh(struct zpci_dev *zdev) +{ + u32 fh = zdev->fh; + int rc; + + if (!zdev_enabled(zdev)) + return 0; + + rc = clp_set_pci_fn(&fh, 0, CLP_SET_DISABLE_PCI_FN); + if (!rc) + /* Success -> store disabled handle in zdev */ + zdev->fh = fh; + + zpci_dbg(3, "dis fid:%x, fh:%x, rc:%d\n", zdev->fid, zdev->fh, rc); + return rc; +} + +static int clp_list_pci(struct clp_req_rsp_list_pci *rrb, + void (*cb)(struct clp_fh_list_entry *entry)) +{ + u64 resume_token = 0; + int entries, i, rc; + + do { + memset(rrb, 0, sizeof(*rrb)); + rrb->request.hdr.len = sizeof(rrb->request); + rrb->request.hdr.cmd = CLP_LIST_PCI; + /* store as many entries as possible */ + rrb->response.hdr.len = CLP_BLK_SIZE - LIST_PCI_HDR_LEN; + rrb->request.resume_token = resume_token; + + /* Get PCI function handle list */ + rc = clp_instr(rrb); + if (rc || rrb->response.hdr.rsp != CLP_RC_OK) { + zpci_err("List PCI FN:\n"); + zpci_err_clp(rrb->response.hdr.rsp, rc); + rc = -EIO; + goto out; + } + + WARN_ON_ONCE(rrb->response.entry_size != + sizeof(struct clp_fh_list_entry)); + + entries = (rrb->response.hdr.len - LIST_PCI_HDR_LEN) / + rrb->response.entry_size; + + resume_token = rrb->response.resume_token; + for (i = 0; i < entries; i++) + cb(&rrb->response.fh_list[i]); + } while (resume_token); +out: + return rc; +} + +static void __clp_add(struct clp_fh_list_entry *entry) +{ + if (!entry->vendor_id) + return; + + clp_add_pci_device(entry->fid, entry->fh, entry->config_state); +} + +static void __clp_rescan(struct clp_fh_list_entry *entry) +{ + struct zpci_dev *zdev; + + if (!entry->vendor_id) + return; + + zdev = get_zdev_by_fid(entry->fid); + if (!zdev) { + clp_add_pci_device(entry->fid, entry->fh, entry->config_state); + return; + } + + if (!entry->config_state) { + /* + * The handle is already disabled, that means no iota/irq freeing via + * the firmware interfaces anymore. Need to free resources manually + * (DMA memory, debug, sysfs)... + */ + zpci_stop_device(zdev); + } +} + +static void __clp_update(struct clp_fh_list_entry *entry) +{ + struct zpci_dev *zdev; + + if (!entry->vendor_id) + return; + + zdev = get_zdev_by_fid(entry->fid); + if (!zdev) + return; + + zdev->fh = entry->fh; +} + +int clp_scan_pci_devices(void) +{ + struct clp_req_rsp_list_pci *rrb; + int rc; + + rrb = clp_alloc_block(GFP_KERNEL); + if (!rrb) + return -ENOMEM; + + rc = clp_list_pci(rrb, __clp_add); + + clp_free_block(rrb); + return rc; +} + +int clp_rescan_pci_devices(void) +{ + struct clp_req_rsp_list_pci *rrb; + int rc; + + rrb = clp_alloc_block(GFP_KERNEL); + if (!rrb) + return -ENOMEM; + + rc = clp_list_pci(rrb, __clp_rescan); + + clp_free_block(rrb); + return rc; +} + +int clp_rescan_pci_devices_simple(void) +{ + struct clp_req_rsp_list_pci *rrb; + int rc; + + rrb = clp_alloc_block(GFP_NOWAIT); + if (!rrb) + return -ENOMEM; + + rc = clp_list_pci(rrb, __clp_update); + + clp_free_block(rrb); + return rc; +} diff --git a/arch/s390/pci/pci_debug.c b/arch/s390/pci/pci_debug.c new file mode 100644 index 000000000..4129b0a5f --- /dev/null +++ b/arch/s390/pci/pci_debug.c @@ -0,0 +1,177 @@ +/* + * Copyright IBM Corp. 2012 + * + * Author(s): + * Jan Glauber <jang@linux.vnet.ibm.com> + */ + +#define KMSG_COMPONENT "zpci" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/seq_file.h> +#include <linux/debugfs.h> +#include <linux/export.h> +#include <linux/pci.h> +#include <asm/debug.h> + +#include <asm/pci_dma.h> + +static struct dentry *debugfs_root; +debug_info_t *pci_debug_msg_id; +EXPORT_SYMBOL_GPL(pci_debug_msg_id); +debug_info_t *pci_debug_err_id; +EXPORT_SYMBOL_GPL(pci_debug_err_id); + +static char *pci_perf_names[] = { + /* hardware counters */ + "Load operations", + "Store operations", + "Store block operations", + "Refresh operations", + "DMA read bytes", + "DMA write bytes", +}; + +static char *pci_sw_names[] = { + "Allocated pages", + "Mapped pages", + "Unmapped pages", +}; + +static void pci_sw_counter_show(struct seq_file *m) +{ + struct zpci_dev *zdev = m->private; + atomic64_t *counter = &zdev->allocated_pages; + int i; + + for (i = 0; i < ARRAY_SIZE(pci_sw_names); i++, counter++) + seq_printf(m, "%26s:\t%llu\n", pci_sw_names[i], + atomic64_read(counter)); +} + +static int pci_perf_show(struct seq_file *m, void *v) +{ + struct zpci_dev *zdev = m->private; + u64 *stat; + int i; + + if (!zdev) + return 0; + + mutex_lock(&zdev->lock); + if (!zdev->fmb) { + mutex_unlock(&zdev->lock); + seq_puts(m, "FMB statistics disabled\n"); + return 0; + } + + /* header */ + seq_printf(m, "FMB @ %p\n", zdev->fmb); + seq_printf(m, "Update interval: %u ms\n", zdev->fmb_update); + seq_printf(m, "Samples: %u\n", zdev->fmb->samples); + seq_printf(m, "Last update TOD: %Lx\n", zdev->fmb->last_update); + + /* hardware counters */ + stat = (u64 *) &zdev->fmb->ld_ops; + for (i = 0; i < 4; i++) + seq_printf(m, "%26s:\t%llu\n", + pci_perf_names[i], *(stat + i)); + if (zdev->fmb->dma_valid) + for (i = 4; i < 6; i++) + seq_printf(m, "%26s:\t%llu\n", + pci_perf_names[i], *(stat + i)); + + pci_sw_counter_show(m); + mutex_unlock(&zdev->lock); + return 0; +} + +static ssize_t pci_perf_seq_write(struct file *file, const char __user *ubuf, + size_t count, loff_t *off) +{ + struct zpci_dev *zdev = ((struct seq_file *) file->private_data)->private; + unsigned long val; + int rc; + + if (!zdev) + return 0; + + rc = kstrtoul_from_user(ubuf, count, 10, &val); + if (rc) + return rc; + + mutex_lock(&zdev->lock); + switch (val) { + case 0: + rc = zpci_fmb_disable_device(zdev); + break; + case 1: + rc = zpci_fmb_enable_device(zdev); + break; + } + mutex_unlock(&zdev->lock); + return rc ? rc : count; +} + +static int pci_perf_seq_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, pci_perf_show, + file_inode(filp)->i_private); +} + +static const struct file_operations debugfs_pci_perf_fops = { + .open = pci_perf_seq_open, + .read = seq_read, + .write = pci_perf_seq_write, + .llseek = seq_lseek, + .release = single_release, +}; + +void zpci_debug_init_device(struct zpci_dev *zdev) +{ + zdev->debugfs_dev = debugfs_create_dir(dev_name(&zdev->pdev->dev), + debugfs_root); + if (IS_ERR(zdev->debugfs_dev)) + zdev->debugfs_dev = NULL; + + zdev->debugfs_perf = debugfs_create_file("statistics", + S_IFREG | S_IRUGO | S_IWUSR, + zdev->debugfs_dev, zdev, + &debugfs_pci_perf_fops); + if (IS_ERR(zdev->debugfs_perf)) + zdev->debugfs_perf = NULL; +} + +void zpci_debug_exit_device(struct zpci_dev *zdev) +{ + debugfs_remove(zdev->debugfs_perf); + debugfs_remove(zdev->debugfs_dev); +} + +int __init zpci_debug_init(void) +{ + /* event trace buffer */ + pci_debug_msg_id = debug_register("pci_msg", 8, 1, 8 * sizeof(long)); + if (!pci_debug_msg_id) + return -EINVAL; + debug_register_view(pci_debug_msg_id, &debug_sprintf_view); + debug_set_level(pci_debug_msg_id, 3); + + /* error log */ + pci_debug_err_id = debug_register("pci_error", 2, 1, 16); + if (!pci_debug_err_id) + return -EINVAL; + debug_register_view(pci_debug_err_id, &debug_hex_ascii_view); + debug_set_level(pci_debug_err_id, 6); + + debugfs_root = debugfs_create_dir("pci", NULL); + return 0; +} + +void zpci_debug_exit(void) +{ + debug_unregister(pci_debug_msg_id); + debug_unregister(pci_debug_err_id); + debugfs_remove(debugfs_root); +} diff --git a/arch/s390/pci/pci_dma.c b/arch/s390/pci/pci_dma.c new file mode 100644 index 000000000..6fd8d5836 --- /dev/null +++ b/arch/s390/pci/pci_dma.c @@ -0,0 +1,528 @@ +/* + * Copyright IBM Corp. 2012 + * + * Author(s): + * Jan Glauber <jang@linux.vnet.ibm.com> + */ + +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/export.h> +#include <linux/iommu-helper.h> +#include <linux/dma-mapping.h> +#include <linux/vmalloc.h> +#include <linux/pci.h> +#include <asm/pci_dma.h> + +static struct kmem_cache *dma_region_table_cache; +static struct kmem_cache *dma_page_table_cache; +static int s390_iommu_strict; + +static int zpci_refresh_global(struct zpci_dev *zdev) +{ + return zpci_refresh_trans((u64) zdev->fh << 32, zdev->start_dma, + zdev->iommu_pages * PAGE_SIZE); +} + +static unsigned long *dma_alloc_cpu_table(void) +{ + unsigned long *table, *entry; + + table = kmem_cache_alloc(dma_region_table_cache, GFP_ATOMIC); + if (!table) + return NULL; + + for (entry = table; entry < table + ZPCI_TABLE_ENTRIES; entry++) + *entry = ZPCI_TABLE_INVALID | ZPCI_TABLE_PROTECTED; + return table; +} + +static void dma_free_cpu_table(void *table) +{ + kmem_cache_free(dma_region_table_cache, table); +} + +static unsigned long *dma_alloc_page_table(void) +{ + unsigned long *table, *entry; + + table = kmem_cache_alloc(dma_page_table_cache, GFP_ATOMIC); + if (!table) + return NULL; + + for (entry = table; entry < table + ZPCI_PT_ENTRIES; entry++) + *entry = ZPCI_PTE_INVALID | ZPCI_TABLE_PROTECTED; + return table; +} + +static void dma_free_page_table(void *table) +{ + kmem_cache_free(dma_page_table_cache, table); +} + +static unsigned long *dma_get_seg_table_origin(unsigned long *entry) +{ + unsigned long *sto; + + if (reg_entry_isvalid(*entry)) + sto = get_rt_sto(*entry); + else { + sto = dma_alloc_cpu_table(); + if (!sto) + return NULL; + + set_rt_sto(entry, sto); + validate_rt_entry(entry); + entry_clr_protected(entry); + } + return sto; +} + +static unsigned long *dma_get_page_table_origin(unsigned long *entry) +{ + unsigned long *pto; + + if (reg_entry_isvalid(*entry)) + pto = get_st_pto(*entry); + else { + pto = dma_alloc_page_table(); + if (!pto) + return NULL; + set_st_pto(entry, pto); + validate_st_entry(entry); + entry_clr_protected(entry); + } + return pto; +} + +static unsigned long *dma_walk_cpu_trans(unsigned long *rto, dma_addr_t dma_addr) +{ + unsigned long *sto, *pto; + unsigned int rtx, sx, px; + + rtx = calc_rtx(dma_addr); + sto = dma_get_seg_table_origin(&rto[rtx]); + if (!sto) + return NULL; + + sx = calc_sx(dma_addr); + pto = dma_get_page_table_origin(&sto[sx]); + if (!pto) + return NULL; + + px = calc_px(dma_addr); + return &pto[px]; +} + +static void dma_update_cpu_trans(struct zpci_dev *zdev, void *page_addr, + dma_addr_t dma_addr, int flags) +{ + unsigned long *entry; + + entry = dma_walk_cpu_trans(zdev->dma_table, dma_addr); + if (!entry) { + WARN_ON_ONCE(1); + return; + } + + if (flags & ZPCI_PTE_INVALID) { + invalidate_pt_entry(entry); + return; + } else { + set_pt_pfaa(entry, page_addr); + validate_pt_entry(entry); + } + + if (flags & ZPCI_TABLE_PROTECTED) + entry_set_protected(entry); + else + entry_clr_protected(entry); +} + +static int dma_update_trans(struct zpci_dev *zdev, unsigned long pa, + dma_addr_t dma_addr, size_t size, int flags) +{ + unsigned int nr_pages = PAGE_ALIGN(size) >> PAGE_SHIFT; + u8 *page_addr = (u8 *) (pa & PAGE_MASK); + dma_addr_t start_dma_addr = dma_addr; + unsigned long irq_flags; + int i, rc = 0; + + if (!nr_pages) + return -EINVAL; + + spin_lock_irqsave(&zdev->dma_table_lock, irq_flags); + if (!zdev->dma_table) + goto no_refresh; + + for (i = 0; i < nr_pages; i++) { + dma_update_cpu_trans(zdev, page_addr, dma_addr, flags); + page_addr += PAGE_SIZE; + dma_addr += PAGE_SIZE; + } + + /* + * With zdev->tlb_refresh == 0, rpcit is not required to establish new + * translations when previously invalid translation-table entries are + * validated. With lazy unmap, it also is skipped for previously valid + * entries, but a global rpcit is then required before any address can + * be re-used, i.e. after each iommu bitmap wrap-around. + */ + if (!zdev->tlb_refresh && + (!s390_iommu_strict || + ((flags & ZPCI_PTE_VALID_MASK) == ZPCI_PTE_VALID))) + goto no_refresh; + + rc = zpci_refresh_trans((u64) zdev->fh << 32, start_dma_addr, + nr_pages * PAGE_SIZE); + +no_refresh: + spin_unlock_irqrestore(&zdev->dma_table_lock, irq_flags); + return rc; +} + +static void dma_free_seg_table(unsigned long entry) +{ + unsigned long *sto = get_rt_sto(entry); + int sx; + + for (sx = 0; sx < ZPCI_TABLE_ENTRIES; sx++) + if (reg_entry_isvalid(sto[sx])) + dma_free_page_table(get_st_pto(sto[sx])); + + dma_free_cpu_table(sto); +} + +static void dma_cleanup_tables(struct zpci_dev *zdev) +{ + unsigned long *table; + int rtx; + + if (!zdev || !zdev->dma_table) + return; + + table = zdev->dma_table; + for (rtx = 0; rtx < ZPCI_TABLE_ENTRIES; rtx++) + if (reg_entry_isvalid(table[rtx])) + dma_free_seg_table(table[rtx]); + + dma_free_cpu_table(table); + zdev->dma_table = NULL; +} + +static unsigned long __dma_alloc_iommu(struct zpci_dev *zdev, + unsigned long start, int size) +{ + unsigned long boundary_size; + + boundary_size = ALIGN(dma_get_seg_boundary(&zdev->pdev->dev) + 1, + PAGE_SIZE) >> PAGE_SHIFT; + return iommu_area_alloc(zdev->iommu_bitmap, zdev->iommu_pages, + start, size, 0, boundary_size, 0); +} + +static unsigned long dma_alloc_iommu(struct zpci_dev *zdev, int size) +{ + unsigned long offset, flags; + int wrap = 0; + + spin_lock_irqsave(&zdev->iommu_bitmap_lock, flags); + offset = __dma_alloc_iommu(zdev, zdev->next_bit, size); + if (offset == -1) { + /* wrap-around */ + offset = __dma_alloc_iommu(zdev, 0, size); + wrap = 1; + } + + if (offset != -1) { + zdev->next_bit = offset + size; + if (!zdev->tlb_refresh && !s390_iommu_strict && wrap) + /* global flush after wrap-around with lazy unmap */ + zpci_refresh_global(zdev); + } + spin_unlock_irqrestore(&zdev->iommu_bitmap_lock, flags); + return offset; +} + +static void dma_free_iommu(struct zpci_dev *zdev, unsigned long offset, int size) +{ + unsigned long flags; + + spin_lock_irqsave(&zdev->iommu_bitmap_lock, flags); + if (!zdev->iommu_bitmap) + goto out; + bitmap_clear(zdev->iommu_bitmap, offset, size); + /* + * Lazy flush for unmap: need to move next_bit to avoid address re-use + * until wrap-around. + */ + if (!s390_iommu_strict && offset >= zdev->next_bit) + zdev->next_bit = offset + size; +out: + spin_unlock_irqrestore(&zdev->iommu_bitmap_lock, flags); +} + +int dma_set_mask(struct device *dev, u64 mask) +{ + if (!dev->dma_mask || !dma_supported(dev, mask)) + return -EIO; + + *dev->dma_mask = mask; + return 0; +} +EXPORT_SYMBOL_GPL(dma_set_mask); + +static dma_addr_t s390_dma_map_pages(struct device *dev, struct page *page, + unsigned long offset, size_t size, + enum dma_data_direction direction, + struct dma_attrs *attrs) +{ + struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); + unsigned long nr_pages, iommu_page_index; + unsigned long pa = page_to_phys(page) + offset; + int flags = ZPCI_PTE_VALID; + dma_addr_t dma_addr; + + /* This rounds up number of pages based on size and offset */ + nr_pages = iommu_num_pages(pa, size, PAGE_SIZE); + iommu_page_index = dma_alloc_iommu(zdev, nr_pages); + if (iommu_page_index == -1) + goto out_err; + + /* Use rounded up size */ + size = nr_pages * PAGE_SIZE; + + dma_addr = zdev->start_dma + iommu_page_index * PAGE_SIZE; + if (dma_addr + size > zdev->end_dma) + goto out_free; + + if (direction == DMA_NONE || direction == DMA_TO_DEVICE) + flags |= ZPCI_TABLE_PROTECTED; + + if (!dma_update_trans(zdev, pa, dma_addr, size, flags)) { + atomic64_add(nr_pages, &zdev->mapped_pages); + return dma_addr + (offset & ~PAGE_MASK); + } + +out_free: + dma_free_iommu(zdev, iommu_page_index, nr_pages); +out_err: + zpci_err("map error:\n"); + zpci_err_hex(&pa, sizeof(pa)); + return DMA_ERROR_CODE; +} + +static void s390_dma_unmap_pages(struct device *dev, dma_addr_t dma_addr, + size_t size, enum dma_data_direction direction, + struct dma_attrs *attrs) +{ + struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); + unsigned long iommu_page_index; + int npages; + + npages = iommu_num_pages(dma_addr, size, PAGE_SIZE); + dma_addr = dma_addr & PAGE_MASK; + if (dma_update_trans(zdev, 0, dma_addr, npages * PAGE_SIZE, + ZPCI_TABLE_PROTECTED | ZPCI_PTE_INVALID)) { + zpci_err("unmap error:\n"); + zpci_err_hex(&dma_addr, sizeof(dma_addr)); + } + + atomic64_add(npages, &zdev->unmapped_pages); + iommu_page_index = (dma_addr - zdev->start_dma) >> PAGE_SHIFT; + dma_free_iommu(zdev, iommu_page_index, npages); +} + +static void *s390_dma_alloc(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t flag, + struct dma_attrs *attrs) +{ + struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); + struct page *page; + unsigned long pa; + dma_addr_t map; + + size = PAGE_ALIGN(size); + page = alloc_pages(flag, get_order(size)); + if (!page) + return NULL; + + pa = page_to_phys(page); + memset((void *) pa, 0, size); + + map = s390_dma_map_pages(dev, page, pa % PAGE_SIZE, + size, DMA_BIDIRECTIONAL, NULL); + if (dma_mapping_error(dev, map)) { + free_pages(pa, get_order(size)); + return NULL; + } + + atomic64_add(size / PAGE_SIZE, &zdev->allocated_pages); + if (dma_handle) + *dma_handle = map; + return (void *) pa; +} + +static void s390_dma_free(struct device *dev, size_t size, + void *pa, dma_addr_t dma_handle, + struct dma_attrs *attrs) +{ + struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); + + size = PAGE_ALIGN(size); + atomic64_sub(size / PAGE_SIZE, &zdev->allocated_pages); + s390_dma_unmap_pages(dev, dma_handle, size, DMA_BIDIRECTIONAL, NULL); + free_pages((unsigned long) pa, get_order(size)); +} + +static int s390_dma_map_sg(struct device *dev, struct scatterlist *sg, + int nr_elements, enum dma_data_direction dir, + struct dma_attrs *attrs) +{ + int mapped_elements = 0; + struct scatterlist *s; + int i; + + for_each_sg(sg, s, nr_elements, i) { + struct page *page = sg_page(s); + s->dma_address = s390_dma_map_pages(dev, page, s->offset, + s->length, dir, NULL); + if (!dma_mapping_error(dev, s->dma_address)) { + s->dma_length = s->length; + mapped_elements++; + } else + goto unmap; + } +out: + return mapped_elements; + +unmap: + for_each_sg(sg, s, mapped_elements, i) { + if (s->dma_address) + s390_dma_unmap_pages(dev, s->dma_address, s->dma_length, + dir, NULL); + s->dma_address = 0; + s->dma_length = 0; + } + mapped_elements = 0; + goto out; +} + +static void s390_dma_unmap_sg(struct device *dev, struct scatterlist *sg, + int nr_elements, enum dma_data_direction dir, + struct dma_attrs *attrs) +{ + struct scatterlist *s; + int i; + + for_each_sg(sg, s, nr_elements, i) { + s390_dma_unmap_pages(dev, s->dma_address, s->dma_length, dir, NULL); + s->dma_address = 0; + s->dma_length = 0; + } +} + +int zpci_dma_init_device(struct zpci_dev *zdev) +{ + int rc; + + spin_lock_init(&zdev->iommu_bitmap_lock); + spin_lock_init(&zdev->dma_table_lock); + + zdev->dma_table = dma_alloc_cpu_table(); + if (!zdev->dma_table) { + rc = -ENOMEM; + goto out_clean; + } + + zdev->iommu_size = (unsigned long) high_memory - PAGE_OFFSET; + zdev->iommu_pages = zdev->iommu_size >> PAGE_SHIFT; + zdev->iommu_bitmap = vzalloc(zdev->iommu_pages / 8); + if (!zdev->iommu_bitmap) { + rc = -ENOMEM; + goto out_reg; + } + + rc = zpci_register_ioat(zdev, + 0, + zdev->start_dma + PAGE_OFFSET, + zdev->start_dma + zdev->iommu_size - 1, + (u64) zdev->dma_table); + if (rc) + goto out_reg; + return 0; + +out_reg: + dma_free_cpu_table(zdev->dma_table); +out_clean: + return rc; +} + +void zpci_dma_exit_device(struct zpci_dev *zdev) +{ + zpci_unregister_ioat(zdev, 0); + dma_cleanup_tables(zdev); + vfree(zdev->iommu_bitmap); + zdev->iommu_bitmap = NULL; + zdev->next_bit = 0; +} + +static int __init dma_alloc_cpu_table_caches(void) +{ + dma_region_table_cache = kmem_cache_create("PCI_DMA_region_tables", + ZPCI_TABLE_SIZE, ZPCI_TABLE_ALIGN, + 0, NULL); + if (!dma_region_table_cache) + return -ENOMEM; + + dma_page_table_cache = kmem_cache_create("PCI_DMA_page_tables", + ZPCI_PT_SIZE, ZPCI_PT_ALIGN, + 0, NULL); + if (!dma_page_table_cache) { + kmem_cache_destroy(dma_region_table_cache); + return -ENOMEM; + } + return 0; +} + +int __init zpci_dma_init(void) +{ + return dma_alloc_cpu_table_caches(); +} + +void zpci_dma_exit(void) +{ + kmem_cache_destroy(dma_page_table_cache); + kmem_cache_destroy(dma_region_table_cache); +} + +#define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16) + +static int __init dma_debug_do_init(void) +{ + dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES); + return 0; +} +fs_initcall(dma_debug_do_init); + +struct dma_map_ops s390_dma_ops = { + .alloc = s390_dma_alloc, + .free = s390_dma_free, + .map_sg = s390_dma_map_sg, + .unmap_sg = s390_dma_unmap_sg, + .map_page = s390_dma_map_pages, + .unmap_page = s390_dma_unmap_pages, + /* if we support direct DMA this must be conditional */ + .is_phys = 0, + /* dma_supported is unconditionally true without a callback */ +}; +EXPORT_SYMBOL_GPL(s390_dma_ops); + +static int __init s390_iommu_setup(char *str) +{ + if (!strncmp(str, "strict", 6)) + s390_iommu_strict = 1; + return 0; +} + +__setup("s390_iommu=", s390_iommu_setup); diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c new file mode 100644 index 000000000..460fdb21c --- /dev/null +++ b/arch/s390/pci/pci_event.c @@ -0,0 +1,136 @@ +/* + * Copyright IBM Corp. 2012 + * + * Author(s): + * Jan Glauber <jang@linux.vnet.ibm.com> + */ + +#define KMSG_COMPONENT "zpci" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/pci.h> +#include <asm/pci_debug.h> +#include <asm/sclp.h> + +/* Content Code Description for PCI Function Error */ +struct zpci_ccdf_err { + u32 reserved1; + u32 fh; /* function handle */ + u32 fid; /* function id */ + u32 ett : 4; /* expected table type */ + u32 mvn : 12; /* MSI vector number */ + u32 dmaas : 8; /* DMA address space */ + u32 : 6; + u32 q : 1; /* event qualifier */ + u32 rw : 1; /* read/write */ + u64 faddr; /* failing address */ + u32 reserved3; + u16 reserved4; + u16 pec; /* PCI event code */ +} __packed; + +/* Content Code Description for PCI Function Availability */ +struct zpci_ccdf_avail { + u32 reserved1; + u32 fh; /* function handle */ + u32 fid; /* function id */ + u32 reserved2; + u32 reserved3; + u32 reserved4; + u32 reserved5; + u16 reserved6; + u16 pec; /* PCI event code */ +} __packed; + +static void __zpci_event_error(struct zpci_ccdf_err *ccdf) +{ + struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid); + + zpci_err("error CCDF:\n"); + zpci_err_hex(ccdf, sizeof(*ccdf)); + + if (!zdev) + return; + + pr_err("%s: Event 0x%x reports an error for PCI function 0x%x\n", + pci_name(zdev->pdev), ccdf->pec, ccdf->fid); +} + +void zpci_event_error(void *data) +{ + if (zpci_is_enabled()) + __zpci_event_error(data); +} + +static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf) +{ + struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid); + struct pci_dev *pdev = zdev ? zdev->pdev : NULL; + int ret; + + pr_info("%s: Event 0x%x reconfigured PCI function 0x%x\n", + pdev ? pci_name(pdev) : "n/a", ccdf->pec, ccdf->fid); + zpci_err("avail CCDF:\n"); + zpci_err_hex(ccdf, sizeof(*ccdf)); + + switch (ccdf->pec) { + case 0x0301: /* Standby -> Configured */ + if (!zdev || zdev->state != ZPCI_FN_STATE_STANDBY) + break; + zdev->state = ZPCI_FN_STATE_CONFIGURED; + zdev->fh = ccdf->fh; + ret = zpci_enable_device(zdev); + if (ret) + break; + pci_rescan_bus(zdev->bus); + break; + case 0x0302: /* Reserved -> Standby */ + if (!zdev) + clp_add_pci_device(ccdf->fid, ccdf->fh, 0); + break; + case 0x0303: /* Deconfiguration requested */ + if (pdev) + pci_stop_and_remove_bus_device(pdev); + + ret = zpci_disable_device(zdev); + if (ret) + break; + + ret = sclp_pci_deconfigure(zdev->fid); + zpci_dbg(3, "deconf fid:%x, rc:%d\n", zdev->fid, ret); + if (!ret) + zdev->state = ZPCI_FN_STATE_STANDBY; + + break; + case 0x0304: /* Configured -> Standby */ + if (pdev) { + /* Give the driver a hint that the function is + * already unusable. */ + pdev->error_state = pci_channel_io_perm_failure; + pci_stop_and_remove_bus_device(pdev); + } + + zdev->fh = ccdf->fh; + zpci_disable_device(zdev); + zdev->state = ZPCI_FN_STATE_STANDBY; + break; + case 0x0306: /* 0x308 or 0x302 for multiple devices */ + clp_rescan_pci_devices(); + break; + case 0x0308: /* Standby -> Reserved */ + if (!zdev) + break; + pci_stop_root_bus(zdev->bus); + pci_remove_root_bus(zdev->bus); + break; + default: + break; + } +} + +void zpci_event_availability(void *data) +{ + if (zpci_is_enabled()) + __zpci_event_availability(data); +} diff --git a/arch/s390/pci/pci_insn.c b/arch/s390/pci/pci_insn.c new file mode 100644 index 000000000..85267c058 --- /dev/null +++ b/arch/s390/pci/pci_insn.c @@ -0,0 +1,202 @@ +/* + * s390 specific pci instructions + * + * Copyright IBM Corp. 2013 + */ + +#include <linux/export.h> +#include <linux/errno.h> +#include <linux/delay.h> +#include <asm/pci_insn.h> +#include <asm/processor.h> + +#define ZPCI_INSN_BUSY_DELAY 1 /* 1 microsecond */ + +/* Modify PCI Function Controls */ +static inline u8 __mpcifc(u64 req, struct zpci_fib *fib, u8 *status) +{ + u8 cc; + + asm volatile ( + " .insn rxy,0xe300000000d0,%[req],%[fib]\n" + " ipm %[cc]\n" + " srl %[cc],28\n" + : [cc] "=d" (cc), [req] "+d" (req), [fib] "+Q" (*fib) + : : "cc"); + *status = req >> 24 & 0xff; + return cc; +} + +int zpci_mod_fc(u64 req, struct zpci_fib *fib) +{ + u8 cc, status; + + do { + cc = __mpcifc(req, fib, &status); + if (cc == 2) + msleep(ZPCI_INSN_BUSY_DELAY); + } while (cc == 2); + + if (cc) + printk_once(KERN_ERR "%s: error cc: %d status: %d\n", + __func__, cc, status); + return (cc) ? -EIO : 0; +} + +/* Refresh PCI Translations */ +static inline u8 __rpcit(u64 fn, u64 addr, u64 range, u8 *status) +{ + register u64 __addr asm("2") = addr; + register u64 __range asm("3") = range; + u8 cc; + + asm volatile ( + " .insn rre,0xb9d30000,%[fn],%[addr]\n" + " ipm %[cc]\n" + " srl %[cc],28\n" + : [cc] "=d" (cc), [fn] "+d" (fn) + : [addr] "d" (__addr), "d" (__range) + : "cc"); + *status = fn >> 24 & 0xff; + return cc; +} + +int zpci_refresh_trans(u64 fn, u64 addr, u64 range) +{ + u8 cc, status; + + do { + cc = __rpcit(fn, addr, range, &status); + if (cc == 2) + udelay(ZPCI_INSN_BUSY_DELAY); + } while (cc == 2); + + if (cc) + printk_once(KERN_ERR "%s: error cc: %d status: %d dma_addr: %Lx size: %Lx\n", + __func__, cc, status, addr, range); + return (cc) ? -EIO : 0; +} + +/* Set Interruption Controls */ +void zpci_set_irq_ctrl(u16 ctl, char *unused, u8 isc) +{ + asm volatile ( + " .insn rsy,0xeb00000000d1,%[ctl],%[isc],%[u]\n" + : : [ctl] "d" (ctl), [isc] "d" (isc << 27), [u] "Q" (*unused)); +} + +/* PCI Load */ +static inline int __pcilg(u64 *data, u64 req, u64 offset, u8 *status) +{ + register u64 __req asm("2") = req; + register u64 __offset asm("3") = offset; + int cc = -ENXIO; + u64 __data; + + asm volatile ( + " .insn rre,0xb9d20000,%[data],%[req]\n" + "0: ipm %[cc]\n" + " srl %[cc],28\n" + "1:\n" + EX_TABLE(0b, 1b) + : [cc] "+d" (cc), [data] "=d" (__data), [req] "+d" (__req) + : "d" (__offset) + : "cc"); + *status = __req >> 24 & 0xff; + if (!cc) + *data = __data; + + return cc; +} + +int zpci_load(u64 *data, u64 req, u64 offset) +{ + u8 status; + int cc; + + do { + cc = __pcilg(data, req, offset, &status); + if (cc == 2) + udelay(ZPCI_INSN_BUSY_DELAY); + } while (cc == 2); + + if (cc) + printk_once(KERN_ERR "%s: error cc: %d status: %d req: %Lx offset: %Lx\n", + __func__, cc, status, req, offset); + return (cc > 0) ? -EIO : cc; +} +EXPORT_SYMBOL_GPL(zpci_load); + +/* PCI Store */ +static inline int __pcistg(u64 data, u64 req, u64 offset, u8 *status) +{ + register u64 __req asm("2") = req; + register u64 __offset asm("3") = offset; + int cc = -ENXIO; + + asm volatile ( + " .insn rre,0xb9d00000,%[data],%[req]\n" + "0: ipm %[cc]\n" + " srl %[cc],28\n" + "1:\n" + EX_TABLE(0b, 1b) + : [cc] "+d" (cc), [req] "+d" (__req) + : "d" (__offset), [data] "d" (data) + : "cc"); + *status = __req >> 24 & 0xff; + return cc; +} + +int zpci_store(u64 data, u64 req, u64 offset) +{ + u8 status; + int cc; + + do { + cc = __pcistg(data, req, offset, &status); + if (cc == 2) + udelay(ZPCI_INSN_BUSY_DELAY); + } while (cc == 2); + + if (cc) + printk_once(KERN_ERR "%s: error cc: %d status: %d req: %Lx offset: %Lx\n", + __func__, cc, status, req, offset); + return (cc > 0) ? -EIO : cc; +} +EXPORT_SYMBOL_GPL(zpci_store); + +/* PCI Store Block */ +static inline int __pcistb(const u64 *data, u64 req, u64 offset, u8 *status) +{ + int cc = -ENXIO; + + asm volatile ( + " .insn rsy,0xeb00000000d0,%[req],%[offset],%[data]\n" + "0: ipm %[cc]\n" + " srl %[cc],28\n" + "1:\n" + EX_TABLE(0b, 1b) + : [cc] "+d" (cc), [req] "+d" (req) + : [offset] "d" (offset), [data] "Q" (*data) + : "cc"); + *status = req >> 24 & 0xff; + return cc; +} + +int zpci_store_block(const u64 *data, u64 req, u64 offset) +{ + u8 status; + int cc; + + do { + cc = __pcistb(data, req, offset, &status); + if (cc == 2) + udelay(ZPCI_INSN_BUSY_DELAY); + } while (cc == 2); + + if (cc) + printk_once(KERN_ERR "%s: error cc: %d status: %d req: %Lx offset: %Lx\n", + __func__, cc, status, req, offset); + return (cc > 0) ? -EIO : cc; +} +EXPORT_SYMBOL_GPL(zpci_store_block); diff --git a/arch/s390/pci/pci_mmio.c b/arch/s390/pci/pci_mmio.c new file mode 100644 index 000000000..b1bb2b723 --- /dev/null +++ b/arch/s390/pci/pci_mmio.c @@ -0,0 +1,114 @@ +/* + * Access to PCI I/O memory from user space programs. + * + * Copyright IBM Corp. 2014 + * Author(s): Alexey Ishchuk <aishchuk@linux.vnet.ibm.com> + */ +#include <linux/kernel.h> +#include <linux/syscalls.h> +#include <linux/init.h> +#include <linux/mm.h> +#include <linux/errno.h> +#include <linux/pci.h> + +static long get_pfn(unsigned long user_addr, unsigned long access, + unsigned long *pfn) +{ + struct vm_area_struct *vma; + long ret; + + down_read(¤t->mm->mmap_sem); + ret = -EINVAL; + vma = find_vma(current->mm, user_addr); + if (!vma) + goto out; + ret = -EACCES; + if (!(vma->vm_flags & access)) + goto out; + ret = follow_pfn(vma, user_addr, pfn); +out: + up_read(¤t->mm->mmap_sem); + return ret; +} + +SYSCALL_DEFINE3(s390_pci_mmio_write, unsigned long, mmio_addr, + const void __user *, user_buffer, size_t, length) +{ + u8 local_buf[64]; + void __iomem *io_addr; + void *buf; + unsigned long pfn; + long ret; + + if (!zpci_is_enabled()) + return -ENODEV; + + if (length <= 0 || PAGE_SIZE - (mmio_addr & ~PAGE_MASK) < length) + return -EINVAL; + if (length > 64) { + buf = kmalloc(length, GFP_KERNEL); + if (!buf) + return -ENOMEM; + } else + buf = local_buf; + + ret = get_pfn(mmio_addr, VM_WRITE, &pfn); + if (ret) + goto out; + io_addr = (void __iomem *)((pfn << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK)); + + ret = -EFAULT; + if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE) + goto out; + + if (copy_from_user(buf, user_buffer, length)) + goto out; + + ret = zpci_memcpy_toio(io_addr, buf, length); +out: + if (buf != local_buf) + kfree(buf); + return ret; +} + +SYSCALL_DEFINE3(s390_pci_mmio_read, unsigned long, mmio_addr, + void __user *, user_buffer, size_t, length) +{ + u8 local_buf[64]; + void __iomem *io_addr; + void *buf; + unsigned long pfn; + long ret; + + if (!zpci_is_enabled()) + return -ENODEV; + + if (length <= 0 || PAGE_SIZE - (mmio_addr & ~PAGE_MASK) < length) + return -EINVAL; + if (length > 64) { + buf = kmalloc(length, GFP_KERNEL); + if (!buf) + return -ENOMEM; + } else + buf = local_buf; + + ret = get_pfn(mmio_addr, VM_READ, &pfn); + if (ret) + goto out; + io_addr = (void __iomem *)((pfn << PAGE_SHIFT) | (mmio_addr & ~PAGE_MASK)); + + if ((unsigned long) io_addr < ZPCI_IOMAP_ADDR_BASE) { + ret = -EFAULT; + goto out; + } + ret = zpci_memcpy_fromio(buf, io_addr, length); + if (ret) + goto out; + if (copy_to_user(user_buffer, buf, length)) + ret = -EFAULT; + +out: + if (buf != local_buf) + kfree(buf); + return ret; +} diff --git a/arch/s390/pci/pci_sysfs.c b/arch/s390/pci/pci_sysfs.c new file mode 100644 index 000000000..fa3ce891e --- /dev/null +++ b/arch/s390/pci/pci_sysfs.c @@ -0,0 +1,110 @@ +/* + * Copyright IBM Corp. 2012 + * + * Author(s): + * Jan Glauber <jang@linux.vnet.ibm.com> + */ + +#define KMSG_COMPONENT "zpci" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <linux/kernel.h> +#include <linux/stat.h> +#include <linux/pci.h> + +#define zpci_attr(name, fmt, member) \ +static ssize_t name##_show(struct device *dev, \ + struct device_attribute *attr, char *buf) \ +{ \ + struct zpci_dev *zdev = get_zdev(to_pci_dev(dev)); \ + \ + return sprintf(buf, fmt, zdev->member); \ +} \ +static DEVICE_ATTR_RO(name) + +zpci_attr(function_id, "0x%08x\n", fid); +zpci_attr(function_handle, "0x%08x\n", fh); +zpci_attr(pchid, "0x%04x\n", pchid); +zpci_attr(pfgid, "0x%02x\n", pfgid); +zpci_attr(vfn, "0x%04x\n", vfn); +zpci_attr(pft, "0x%02x\n", pft); +zpci_attr(uid, "0x%x\n", uid); +zpci_attr(segment0, "0x%02x\n", pfip[0]); +zpci_attr(segment1, "0x%02x\n", pfip[1]); +zpci_attr(segment2, "0x%02x\n", pfip[2]); +zpci_attr(segment3, "0x%02x\n", pfip[3]); + +static ssize_t recover_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct zpci_dev *zdev = get_zdev(pdev); + int ret; + + if (!device_remove_file_self(dev, attr)) + return count; + + pci_stop_and_remove_bus_device(pdev); + ret = zpci_disable_device(zdev); + if (ret) + return ret; + + ret = zpci_enable_device(zdev); + if (ret) + return ret; + + pci_rescan_bus(zdev->bus); + return count; +} +static DEVICE_ATTR_WO(recover); + +static ssize_t util_string_read(struct file *filp, struct kobject *kobj, + struct bin_attribute *attr, char *buf, + loff_t off, size_t count) +{ + struct device *dev = kobj_to_dev(kobj); + struct pci_dev *pdev = to_pci_dev(dev); + struct zpci_dev *zdev = get_zdev(pdev); + + return memory_read_from_buffer(buf, count, &off, zdev->util_str, + sizeof(zdev->util_str)); +} +static BIN_ATTR_RO(util_string, CLP_UTIL_STR_LEN); +static struct bin_attribute *zpci_bin_attrs[] = { + &bin_attr_util_string, + NULL, +}; + +static struct attribute *zpci_dev_attrs[] = { + &dev_attr_function_id.attr, + &dev_attr_function_handle.attr, + &dev_attr_pchid.attr, + &dev_attr_pfgid.attr, + &dev_attr_pft.attr, + &dev_attr_vfn.attr, + &dev_attr_uid.attr, + &dev_attr_recover.attr, + NULL, +}; +static struct attribute_group zpci_attr_group = { + .attrs = zpci_dev_attrs, + .bin_attrs = zpci_bin_attrs, +}; + +static struct attribute *pfip_attrs[] = { + &dev_attr_segment0.attr, + &dev_attr_segment1.attr, + &dev_attr_segment2.attr, + &dev_attr_segment3.attr, + NULL, +}; +static struct attribute_group pfip_attr_group = { + .name = "pfip", + .attrs = pfip_attrs, +}; + +const struct attribute_group *zpci_attr_groups[] = { + &zpci_attr_group, + &pfip_attr_group, + NULL, +}; |