Linux-libre 4.3.2-gnu

91 files changed, 3774 insertions, 1666 deletions

diff --git a/arch/s390/Kbuild b/arch/s390/Kbuild
index 2938934c6..e256592eb 100644
--- a/arch/s390/Kbuild
+++ b/arch/s390/Kbuild
@@ -6,3 +6,4 @@ obj-$(CONFIG_S390_HYPFS_FS)	+= hypfs/
 obj-$(CONFIG_APPLDATA_BASE)	+= appldata/
 obj-y				+= net/
 obj-$(CONFIG_PCI)		+= pci/
+obj-$(CONFIG_NUMA)		+= numa/
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index b06dc3839..1d57000b1 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -48,6 +48,7 @@ config ARCH_SUPPORTS_DEBUG_PAGEALLOC
 
 config KEXEC
 	def_bool y
+	select KEXEC_CORE
 
 config AUDIT_ARCH
 	def_bool y
@@ -99,18 +100,22 @@ config S390
 	select ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE
 	select ARCH_SAVE_PAGE_KEYS if HIBERNATION
 	select ARCH_SUPPORTS_ATOMIC_RMW
+	select ARCH_SUPPORTS_NUMA_BALANCING
 	select ARCH_USE_CMPXCHG_LOCKREF
+	select ARCH_WANTS_PROT_NUMA_PROT_NONE
 	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_AUTOPROBE
 	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_EARLY_PFN_TO_NID
 	select HAVE_ARCH_JUMP_LABEL
 	select HAVE_ARCH_SECCOMP_FILTER
 	select HAVE_ARCH_TRACEHOOK
@@ -153,6 +158,7 @@ config S390
 	select VIRT_CPU_ACCOUNTING
 	select VIRT_TO_BUS
 
+
 config SCHED_OMIT_FRAME_POINTER
 	def_bool y
 
@@ -385,6 +391,76 @@ config HOTPLUG_CPU
 config SCHED_SMT
 	def_bool n
 
+# Some NUMA nodes have memory ranges that span
+# other nodes.	Even though a pfn is valid and
+# between a node's start and end pfns, it may not
+# reside on that node.	See memmap_init_zone()
+# for details. <- They meant memory holes!
+config NODES_SPAN_OTHER_NODES
+	def_bool NUMA
+
+config NUMA
+	bool "NUMA support"
+	depends on SMP && 64BIT && SCHED_TOPOLOGY
+	default n
+	help
+	  Enable NUMA support
+
+	  This option adds NUMA support to the kernel.
+
+	  An operation mode can be selected by appending
+	  numa=<method> to the kernel command line.
+
+	  The default behaviour is identical to appending numa=plain to
+	  the command line. This will create just one node with all
+	  available memory and all CPUs in it.
+
+config NODES_SHIFT
+	int "Maximum NUMA nodes (as a power of 2)"
+	range 1 10
+	depends on NUMA
+	default "4"
+	help
+	  Specify the maximum number of NUMA nodes available on the target
+	  system. Increases memory reserved to accommodate various tables.
+
+menu "Select NUMA modes"
+	depends on NUMA
+
+config NUMA_EMU
+	bool "NUMA emulation"
+	default y
+	help
+	  Numa emulation mode will split the available system memory into
+	  equal chunks which then are distributed over the configured number
+	  of nodes in a round-robin manner.
+
+	  The number of fake nodes is limited by the number of available memory
+	  chunks (i.e. memory size / fake size) and the number of supported
+	  nodes in the kernel.
+
+	  The CPUs are assigned to the nodes in a way that partially respects
+	  the original machine topology (if supported by the machine).
+	  Fair distribution of the CPUs is not guaranteed.
+
+config EMU_SIZE
+	hex "NUMA emulation memory chunk size"
+	default 0x10000000
+	range 0x400000 0x100000000
+	depends on NUMA_EMU
+	help
+	  Select the default size by which the memory is chopped and then
+	  assigned to emulated NUMA nodes.
+
+	  This can be overridden by specifying
+
+	  emu_size=<n>
+
+	  on the kernel command line where also suffixes K, M, G, and T are
+	  supported.
+
+endmenu
+
 config SCHED_MC
 	def_bool n
 
diff --git a/arch/s390/Makefile b/arch/s390/Makefile
index 667b1bca5..e8d4423e4 100644
--- a/arch/s390/Makefile
+++ b/arch/s390/Makefile
@@ -33,6 +33,8 @@ mflags-$(CONFIG_MARCH_Z196)   := -march=z196
 mflags-$(CONFIG_MARCH_ZEC12)  := -march=zEC12
 mflags-$(CONFIG_MARCH_Z13)   := -march=z13
 
+export CC_FLAGS_MARCH := $(mflags-y)
+
 aflags-y += $(mflags-y)
 cflags-y += $(mflags-y)
 
diff --git a/arch/s390/configs/default_defconfig b/arch/s390/configs/default_defconfig
index 940cbddd9..ed7da281d 100644
--- a/arch/s390/configs/default_defconfig
+++ b/arch/s390/configs/default_defconfig
@@ -13,6 +13,7 @@ CONFIG_TASK_IO_ACCOUNTING=y
 CONFIG_RCU_FAST_NO_HZ=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
+CONFIG_NUMA_BALANCING=y
 CONFIG_CGROUP_FREEZER=y
 CONFIG_CGROUP_DEVICE=y
 CONFIG_CPUSETS=y
@@ -50,6 +51,7 @@ CONFIG_LIVEPATCH=y
 CONFIG_MARCH_Z196=y
 CONFIG_TUNE_ZEC12=y
 CONFIG_NR_CPUS=256
+CONFIG_NUMA=y
 CONFIG_PREEMPT=y
 CONFIG_HZ_100=y
 CONFIG_MEMORY_HOTPLUG=y
@@ -379,7 +381,7 @@ CONFIG_ISCSI_TCP=m
 CONFIG_SCSI_DEBUG=m
 CONFIG_ZFCP=y
 CONFIG_SCSI_VIRTIO=m
-CONFIG_SCSI_DH=m
+CONFIG_SCSI_DH=y
 CONFIG_SCSI_DH_RDAC=m
 CONFIG_SCSI_DH_HP_SW=m
 CONFIG_SCSI_DH_EMC=m
diff --git a/arch/s390/configs/gcov_defconfig b/arch/s390/configs/gcov_defconfig
index d793fec91..9858b14cd 100644
--- a/arch/s390/configs/gcov_defconfig
+++ b/arch/s390/configs/gcov_defconfig
@@ -13,6 +13,7 @@ CONFIG_TASK_IO_ACCOUNTING=y
 CONFIG_RCU_FAST_NO_HZ=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
+CONFIG_NUMA_BALANCING=y
 CONFIG_CGROUP_FREEZER=y
 CONFIG_CGROUP_DEVICE=y
 CONFIG_CPUSETS=y
@@ -49,6 +50,7 @@ CONFIG_DEFAULT_DEADLINE=y
 CONFIG_MARCH_Z196=y
 CONFIG_TUNE_ZEC12=y
 CONFIG_NR_CPUS=256
+CONFIG_NUMA=y
 CONFIG_HZ_100=y
 CONFIG_MEMORY_HOTPLUG=y
 CONFIG_MEMORY_HOTREMOVE=y
@@ -375,7 +377,7 @@ CONFIG_ISCSI_TCP=m
 CONFIG_SCSI_DEBUG=m
 CONFIG_ZFCP=y
 CONFIG_SCSI_VIRTIO=m
-CONFIG_SCSI_DH=m
+CONFIG_SCSI_DH=y
 CONFIG_SCSI_DH_RDAC=m
 CONFIG_SCSI_DH_HP_SW=m
 CONFIG_SCSI_DH_EMC=m
diff --git a/arch/s390/configs/performance_defconfig b/arch/s390/configs/performance_defconfig
index 38a77e9c8..7f14f8071 100644
--- a/arch/s390/configs/performance_defconfig
+++ b/arch/s390/configs/performance_defconfig
@@ -13,6 +13,8 @@ CONFIG_TASK_IO_ACCOUNTING=y
 CONFIG_RCU_FAST_NO_HZ=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
+CONFIG_NUMA_BALANCING=y
+# CONFIG_NUMA_BALANCING_DEFAULT_ENABLED is not set
 CONFIG_CGROUP_FREEZER=y
 CONFIG_CGROUP_DEVICE=y
 CONFIG_CPUSETS=y
@@ -48,6 +50,7 @@ CONFIG_LIVEPATCH=y
 CONFIG_MARCH_Z196=y
 CONFIG_TUNE_ZEC12=y
 CONFIG_NR_CPUS=512
+CONFIG_NUMA=y
 CONFIG_HZ_100=y
 CONFIG_MEMORY_HOTPLUG=y
 CONFIG_MEMORY_HOTREMOVE=y
@@ -374,7 +377,7 @@ CONFIG_ISCSI_TCP=m
 CONFIG_SCSI_DEBUG=m
 CONFIG_ZFCP=y
 CONFIG_SCSI_VIRTIO=m
-CONFIG_SCSI_DH=m
+CONFIG_SCSI_DH=y
 CONFIG_SCSI_DH_RDAC=m
 CONFIG_SCSI_DH_HP_SW=m
 CONFIG_SCSI_DH_EMC=m
diff --git a/arch/s390/configs/zfcpdump_defconfig b/arch/s390/configs/zfcpdump_defconfig
index 1b0184a0f..92805d604 100644
--- a/arch/s390/configs/zfcpdump_defconfig
+++ b/arch/s390/configs/zfcpdump_defconfig
@@ -1,7 +1,6 @@
 # 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
@@ -54,10 +53,6 @@ CONFIG_RAW_DRIVER=y
 # 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
diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c
index 5566ce80a..0b9b95f3c 100644
--- a/arch/s390/crypto/aes_s390.c
+++ b/arch/s390/crypto/aes_s390.c
@@ -24,6 +24,7 @@
 #include <crypto/algapi.h>
 #include <linux/err.h>
 #include <linux/module.h>
+#include <linux/cpufeature.h>
 #include <linux/init.h>
 #include <linux/spinlock.h>
 #include "crypt_s390.h"
@@ -976,7 +977,7 @@ static void __exit aes_s390_fini(void)
 	crypto_unregister_alg(&aes_alg);
 }
 
-module_init(aes_s390_init);
+module_cpu_feature_match(MSA, aes_s390_init);
 module_exit(aes_s390_fini);
 
 MODULE_ALIAS_CRYPTO("aes-all");
diff --git a/arch/s390/crypto/des_s390.c b/arch/s390/crypto/des_s390.c
index 9e05cc453..fba1c10a2 100644
--- a/arch/s390/crypto/des_s390.c
+++ b/arch/s390/crypto/des_s390.c
@@ -16,6 +16,7 @@
 
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/cpufeature.h>
 #include <linux/crypto.h>
 #include <crypto/algapi.h>
 #include <crypto/des.h>
@@ -616,7 +617,7 @@ static void __exit des_s390_exit(void)
 	crypto_unregister_alg(&des_alg);
 }
 
-module_init(des_s390_init);
+module_cpu_feature_match(MSA, des_s390_init);
 module_exit(des_s390_exit);
 
 MODULE_ALIAS_CRYPTO("des");
diff --git a/arch/s390/crypto/ghash_s390.c b/arch/s390/crypto/ghash_s390.c
index b258110da..26e14efd3 100644
--- a/arch/s390/crypto/ghash_s390.c
+++ b/arch/s390/crypto/ghash_s390.c
@@ -9,6 +9,7 @@
 
 #include <crypto/internal/hash.h>
 #include <linux/module.h>
+#include <linux/cpufeature.h>
 
 #include "crypt_s390.h"
 
@@ -158,7 +159,7 @@ static void __exit ghash_mod_exit(void)
 	crypto_unregister_shash(&ghash_alg);
 }
 
-module_init(ghash_mod_init);
+module_cpu_feature_match(MSA, ghash_mod_init);
 module_exit(ghash_mod_exit);
 
 MODULE_ALIAS_CRYPTO("ghash");
diff --git a/arch/s390/crypto/prng.c b/arch/s390/crypto/prng.c
index 9d5192c94..b8045b97f 100644
--- a/arch/s390/crypto/prng.c
+++ b/arch/s390/crypto/prng.c
@@ -17,6 +17,7 @@
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/mutex.h>
+#include <linux/cpufeature.h>
 #include <linux/random.h>
 #include <linux/slab.h>
 #include <asm/debug.h>
@@ -914,6 +915,5 @@ static void __exit prng_exit(void)
 	}
 }
 
-
-module_init(prng_init);
+module_cpu_feature_match(MSA, prng_init);
 module_exit(prng_exit);
diff --git a/arch/s390/crypto/sha1_s390.c b/arch/s390/crypto/sha1_s390.c
index 5b2bee323..9208eadae 100644
--- a/arch/s390/crypto/sha1_s390.c
+++ b/arch/s390/crypto/sha1_s390.c
@@ -26,6 +26,7 @@
 #include <crypto/internal/hash.h>
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/cpufeature.h>
 #include <crypto/sha.h>
 
 #include "crypt_s390.h"
@@ -100,7 +101,7 @@ static void __exit sha1_s390_fini(void)
 	crypto_unregister_shash(&alg);
 }
 
-module_init(sha1_s390_init);
+module_cpu_feature_match(MSA, sha1_s390_init);
 module_exit(sha1_s390_fini);
 
 MODULE_ALIAS_CRYPTO("sha1");
diff --git a/arch/s390/crypto/sha256_s390.c b/arch/s390/crypto/sha256_s390.c
index b74ff1581..667888f5c 100644
--- a/arch/s390/crypto/sha256_s390.c
+++ b/arch/s390/crypto/sha256_s390.c
@@ -16,6 +16,7 @@
 #include <crypto/internal/hash.h>
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/cpufeature.h>
 #include <crypto/sha.h>
 
 #include "crypt_s390.h"
@@ -140,7 +141,7 @@ static void __exit sha256_s390_fini(void)
 	crypto_unregister_shash(&sha256_alg);
 }
 
-module_init(sha256_s390_init);
+module_cpu_feature_match(MSA, sha256_s390_init);
 module_exit(sha256_s390_fini);
 
 MODULE_ALIAS_CRYPTO("sha256");
diff --git a/arch/s390/crypto/sha512_s390.c b/arch/s390/crypto/sha512_s390.c
index 0c36989ba..2ba66b151 100644
--- a/arch/s390/crypto/sha512_s390.c
+++ b/arch/s390/crypto/sha512_s390.c
@@ -18,6 +18,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/cpufeature.h>
 
 #include "sha.h"
 #include "crypt_s390.h"
@@ -148,7 +149,7 @@ static void __exit fini(void)
 	crypto_unregister_shash(&sha384_alg);
 }
 
-module_init(init);
+module_cpu_feature_match(MSA, init);
 module_exit(fini);
 
 MODULE_LICENSE("GPL");
diff --git a/arch/s390/include/asm/Kbuild b/arch/s390/include/asm/Kbuild
index 5ad26dd94..9043d2e1e 100644
--- a/arch/s390/include/asm/Kbuild
+++ b/arch/s390/include/asm/Kbuild
@@ -6,3 +6,4 @@ generic-y += mcs_spinlock.h
 generic-y += mm-arch-hooks.h
 generic-y += preempt.h
 generic-y += trace_clock.h
+generic-y += word-at-a-time.h
diff --git a/arch/s390/include/asm/atomic.h b/arch/s390/include/asm/atomic.h
index adbe3802e..117fa5c92 100644
--- a/arch/s390/include/asm/atomic.h
+++ b/arch/s390/include/asm/atomic.h
@@ -27,6 +27,7 @@
 #define __ATOMIC_OR	"lao"
 #define __ATOMIC_AND	"lan"
 #define __ATOMIC_ADD	"laa"
+#define __ATOMIC_XOR	"lax"
 #define __ATOMIC_BARRIER "bcr	14,0\n"
 
 #define __ATOMIC_LOOP(ptr, op_val, op_string, __barrier)		\
@@ -49,6 +50,7 @@
 #define __ATOMIC_OR	"or"
 #define __ATOMIC_AND	"nr"
 #define __ATOMIC_ADD	"ar"
+#define __ATOMIC_XOR	"xr"
 #define __ATOMIC_BARRIER "\n"
 
 #define __ATOMIC_LOOP(ptr, op_val, op_string, __barrier)		\
@@ -118,15 +120,17 @@ static inline void atomic_add(int i, atomic_t *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);
+#define ATOMIC_OP(op, OP)						\
+static inline void atomic_##op(int i, atomic_t *v)			\
+{									\
+	__ATOMIC_LOOP(v, i, __ATOMIC_##OP, __ATOMIC_NO_BARRIER);	\
 }
 
-static inline void atomic_set_mask(unsigned int mask, atomic_t *v)
-{
-	__ATOMIC_LOOP(v, mask, __ATOMIC_OR, __ATOMIC_NO_BARRIER);
-}
+ATOMIC_OP(and, AND)
+ATOMIC_OP(or, OR)
+ATOMIC_OP(xor, XOR)
+
+#undef ATOMIC_OP
 
 #define atomic_xchg(v, new) (xchg(&((v)->counter), new))
 
@@ -167,6 +171,7 @@ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
 #define __ATOMIC64_OR	"laog"
 #define __ATOMIC64_AND	"lang"
 #define __ATOMIC64_ADD	"laag"
+#define __ATOMIC64_XOR	"laxg"
 #define __ATOMIC64_BARRIER "bcr	14,0\n"
 
 #define __ATOMIC64_LOOP(ptr, op_val, op_string, __barrier)		\
@@ -189,6 +194,7 @@ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
 #define __ATOMIC64_OR	"ogr"
 #define __ATOMIC64_AND	"ngr"
 #define __ATOMIC64_ADD	"agr"
+#define __ATOMIC64_XOR	"xgr"
 #define __ATOMIC64_BARRIER "\n"
 
 #define __ATOMIC64_LOOP(ptr, op_val, op_string, __barrier)		\
@@ -247,16 +253,6 @@ static inline void atomic64_add(long long i, atomic64_t *v)
 	__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,
@@ -270,6 +266,17 @@ static inline long long atomic64_cmpxchg(atomic64_t *v,
 	return old;
 }
 
+#define ATOMIC64_OP(op, OP)						\
+static inline void atomic64_##op(long i, atomic64_t *v)			\
+{									\
+	__ATOMIC64_LOOP(v, i, __ATOMIC64_##OP, __ATOMIC64_NO_BARRIER);	\
+}
+
+ATOMIC64_OP(and, AND)
+ATOMIC64_OP(or, OR)
+ATOMIC64_OP(xor, XOR)
+
+#undef ATOMIC64_OP
 #undef __ATOMIC64_LOOP
 
 static inline int atomic64_add_unless(atomic64_t *v, long long i, long long u)
diff --git a/arch/s390/include/asm/barrier.h b/arch/s390/include/asm/barrier.h
index e6f8615a1..d48fe0162 100644
--- a/arch/s390/include/asm/barrier.h
+++ b/arch/s390/include/asm/barrier.h
@@ -42,12 +42,12 @@
 do {									\
 	compiletime_assert_atomic_type(*p);				\
 	barrier();							\
-	ACCESS_ONCE(*p) = (v);						\
+	WRITE_ONCE(*p, v);						\
 } while (0)
 
 #define smp_load_acquire(p)						\
 ({									\
-	typeof(*p) ___p1 = ACCESS_ONCE(*p);				\
+	typeof(*p) ___p1 = READ_ONCE(*p);				\
 	compiletime_assert_atomic_type(*p);				\
 	barrier();							\
 	___p1;								\
diff --git a/arch/s390/include/asm/cpufeature.h b/arch/s390/include/asm/cpufeature.h
new file mode 100644
index 000000000..fa7e69b7c
--- /dev/null
+++ b/arch/s390/include/asm/cpufeature.h
@@ -0,0 +1,29 @@
+/*
+ * Module interface for CPU features
+ *
+ * Copyright IBM Corp. 2015
+ * Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
+ */
+
+#ifndef __ASM_S390_CPUFEATURE_H
+#define __ASM_S390_CPUFEATURE_H
+
+#include <asm/elf.h>
+
+/* Hardware features on Linux on z Systems are indicated by facility bits that
+ * are mapped to the so-called machine flags.  Particular machine flags are
+ * then used to define ELF hardware capabilities; most notably hardware flags
+ * that are essential for user space / glibc.
+ *
+ * Restrict the set of exposed CPU features to ELF hardware capabilities for
+ * now.  Additional machine flags can be indicated by values larger than
+ * MAX_ELF_HWCAP_FEATURES.
+ */
+#define MAX_ELF_HWCAP_FEATURES	(8 * sizeof(elf_hwcap))
+#define MAX_CPU_FEATURES	MAX_ELF_HWCAP_FEATURES
+
+#define cpu_feature(feat)	ilog2(HWCAP_S390_ ## feat)
+
+int cpu_have_feature(unsigned int nr);
+
+#endif /* __ASM_S390_CPUFEATURE_H */
diff --git a/arch/s390/include/asm/ctl_reg.h b/arch/s390/include/asm/ctl_reg.h
index d7697ab80..17a373576 100644
--- a/arch/s390/include/asm/ctl_reg.h
+++ b/arch/s390/include/asm/ctl_reg.h
@@ -46,6 +46,8 @@ static inline void __ctl_clear_bit(unsigned int cr, unsigned int bit)
 	__ctl_load(reg, cr, cr);
 }
 
+void __ctl_set_vx(void);
+
 void smp_ctl_set_bit(int cr, int bit);
 void smp_ctl_clear_bit(int cr, int bit);
 
diff --git a/arch/s390/include/asm/dma-mapping.h b/arch/s390/include/asm/dma-mapping.h
index 9d395961e..b3fd54d93 100644
--- a/arch/s390/include/asm/dma-mapping.h
+++ b/arch/s390/include/asm/dma-mapping.h
@@ -18,27 +18,13 @@ 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)
@@ -46,45 +32,4 @@ static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 	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/etr.h b/arch/s390/include/asm/etr.h
index 629b79a93..f7e5c3668 100644
--- a/arch/s390/include/asm/etr.h
+++ b/arch/s390/include/asm/etr.h
@@ -214,6 +214,9 @@ static inline int etr_ptff(void *ptff_block, unsigned int func)
 void etr_switch_to_local(void);
 void etr_sync_check(void);
 
+/* notifier for syncs */
+extern struct atomic_notifier_head s390_epoch_delta_notifier;
+
 /* STP interruption parameter */
 struct stp_irq_parm {
 	unsigned int _pad0	: 14;
diff --git a/arch/s390/include/asm/fpu-internal.h b/arch/s390/include/asm/fpu-internal.h
new file mode 100644
index 000000000..55dc2c0fb
--- /dev/null
+++ b/arch/s390/include/asm/fpu-internal.h
@@ -0,0 +1,110 @@
+/*
+ * General floating pointer and vector register helpers
+ *
+ * Copyright IBM Corp. 2015
+ * Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
+ */
+
+#ifndef _ASM_S390_FPU_INTERNAL_H
+#define _ASM_S390_FPU_INTERNAL_H
+
+#define FPU_USE_VX		1	/* Vector extension is active */
+
+#ifndef __ASSEMBLY__
+
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <asm/linkage.h>
+#include <asm/ctl_reg.h>
+#include <asm/sigcontext.h>
+
+struct fpu {
+	__u32 fpc;			/* Floating-point control */
+	__u32 flags;
+	union {
+		void *regs;
+		freg_t *fprs;		/* Floating-point register save area */
+		__vector128 *vxrs;	/* Vector register save area */
+	};
+};
+
+void save_fpu_regs(void);
+
+#define is_vx_fpu(fpu) (!!((fpu)->flags & FPU_USE_VX))
+#define is_vx_task(tsk) (!!((tsk)->thread.fpu.flags & FPU_USE_VX))
+
+/* VX array structure for address operand constraints in inline assemblies */
+struct vx_array { __vector128 _[__NUM_VXRS]; };
+
+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_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);
+	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" (*(struct vx_array *) vxrs) : : "1");
+	__ctl_load(cr0, 0, 0);
+	arch_local_irq_restore(flags);
+}
+
+static inline void convert_vx_to_fp(freg_t *fprs, __vector128 *vxrs)
+{
+	int i;
+
+	for (i = 0; i < __NUM_FPRS; i++)
+		fprs[i] = *(freg_t *)(vxrs + i);
+}
+
+static inline void convert_fp_to_vx(__vector128 *vxrs, freg_t *fprs)
+{
+	int i;
+
+	for (i = 0; i < __NUM_FPRS; i++)
+		*(freg_t *)(vxrs + i) = fprs[i];
+}
+
+static inline void fpregs_store(_s390_fp_regs *fpregs, struct fpu *fpu)
+{
+	fpregs->pad = 0;
+	if (is_vx_fpu(fpu))
+		convert_vx_to_fp((freg_t *)&fpregs->fprs, fpu->vxrs);
+	else
+		memcpy((freg_t *)&fpregs->fprs, fpu->fprs,
+		       sizeof(fpregs->fprs));
+}
+
+static inline void fpregs_load(_s390_fp_regs *fpregs, struct fpu *fpu)
+{
+	if (is_vx_fpu(fpu))
+		convert_fp_to_vx(fpu->vxrs, (freg_t *)&fpregs->fprs);
+	else
+		memcpy(fpu->fprs, (freg_t *)&fpregs->fprs,
+		       sizeof(fpregs->fprs));
+}
+
+#endif
+
+#endif /* _ASM_S390_FPU_INTERNAL_H */
diff --git a/arch/s390/include/asm/io.h b/arch/s390/include/asm/io.h
index cb5fdf3a7..437e9af96 100644
--- a/arch/s390/include/asm/io.h
+++ b/arch/s390/include/asm/io.h
@@ -57,6 +57,8 @@ static inline void ioport_unmap(void __iomem *p)
  */
 #define pci_iomap pci_iomap
 #define pci_iounmap pci_iounmap
+#define pci_iomap_wc pci_iomap
+#define pci_iomap_wc_range pci_iomap_range
 
 #define memcpy_fromio(dst, src, count)	zpci_memcpy_fromio(dst, src, count)
 #define memcpy_toio(dst, src, count)	zpci_memcpy_toio(dst, src, count)
diff --git a/arch/s390/include/asm/jump_label.h b/arch/s390/include/asm/jump_label.h
index 69972b795..7f9fd5e3f 100644
--- a/arch/s390/include/asm/jump_label.h
+++ b/arch/s390/include/asm/jump_label.h
@@ -12,14 +12,29 @@
  * 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)
+static __always_inline bool arch_static_branch(struct static_key *key, bool branch)
 {
 	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);
+		: : "X" (&((char *)key)[branch]) : : label);
+
+	return false;
+label:
+	return true;
+}
+
+static __always_inline bool arch_static_branch_jump(struct static_key *key, bool branch)
+{
+	asm_volatile_goto("0:	brcl 15, %l[label]\n"
+		".pushsection __jump_table, \"aw\"\n"
+		".balign 8\n"
+		".quad 0b, %l[label], %0\n"
+		".popsection\n"
+		: : "X" (&((char *)key)[branch]) : : label);
+
 	return false;
 label:
 	return true;
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index 3024acbe1..8ced42609 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -22,6 +22,7 @@
 #include <linux/kvm.h>
 #include <asm/debug.h>
 #include <asm/cpu.h>
+#include <asm/fpu-internal.h>
 #include <asm/isc.h>
 
 #define KVM_MAX_VCPUS 64
@@ -34,6 +35,7 @@
  */
 #define KVM_NR_IRQCHIPS 1
 #define KVM_IRQCHIP_NUM_PINS 4096
+#define KVM_HALT_POLL_NS_DEFAULT 0
 
 #define SIGP_CTRL_C		0x80
 #define SIGP_CTRL_SCN_MASK	0x3f
@@ -209,6 +211,7 @@ struct kvm_vcpu_stat {
 	u32 exit_validity;
 	u32 exit_instruction;
 	u32 halt_successful_poll;
+	u32 halt_attempted_poll;
 	u32 halt_wakeup;
 	u32 instruction_lctl;
 	u32 instruction_lctlg;
@@ -258,6 +261,9 @@ struct kvm_vcpu_stat {
 	u32 diagnose_10;
 	u32 diagnose_44;
 	u32 diagnose_9c;
+	u32 diagnose_258;
+	u32 diagnose_308;
+	u32 diagnose_500;
 };
 
 #define PGM_OPERATION			0x01
@@ -498,10 +504,9 @@ struct kvm_guestdbg_info_arch {
 
 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 fpu	  host_fpregs;
+	struct fpu	  guest_fpregs;
 	struct kvm_s390_local_interrupt local_int;
 	struct hrtimer    ckc_timer;
 	struct kvm_s390_pgm_info pgm;
@@ -630,7 +635,6 @@ 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) {}
diff --git a/arch/s390/include/asm/linkage.h b/arch/s390/include/asm/linkage.h
index fc8a82847..27da78cf4 100644
--- a/arch/s390/include/asm/linkage.h
+++ b/arch/s390/include/asm/linkage.h
@@ -6,4 +6,26 @@
 #define __ALIGN .align 4, 0x07
 #define __ALIGN_STR __stringify(__ALIGN)
 
+#ifndef __ASSEMBLY__
+
+/*
+ * 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
diff --git a/arch/s390/include/asm/mmzone.h b/arch/s390/include/asm/mmzone.h
new file mode 100644
index 000000000..a9e834e60
--- /dev/null
+++ b/arch/s390/include/asm/mmzone.h
@@ -0,0 +1,16 @@
+/*
+ * NUMA support for s390
+ *
+ * Copyright IBM Corp. 2015
+ */
+
+#ifndef _ASM_S390_MMZONE_H
+#define _ASM_S390_MMZONE_H
+
+#ifdef CONFIG_NUMA
+
+extern struct pglist_data *node_data[];
+#define NODE_DATA(nid) (node_data[nid])
+
+#endif /* CONFIG_NUMA */
+#endif /* _ASM_S390_MMZONE_H */
diff --git a/arch/s390/include/asm/numa.h b/arch/s390/include/asm/numa.h
new file mode 100644
index 000000000..dc19ee0c9
--- /dev/null
+++ b/arch/s390/include/asm/numa.h
@@ -0,0 +1,35 @@
+/*
+ * NUMA support for s390
+ *
+ * Declare the NUMA core code structures and functions.
+ *
+ * Copyright IBM Corp. 2015
+ */
+
+#ifndef _ASM_S390_NUMA_H
+#define _ASM_S390_NUMA_H
+
+#ifdef CONFIG_NUMA
+
+#include <linux/numa.h>
+#include <linux/cpumask.h>
+
+void numa_setup(void);
+int numa_pfn_to_nid(unsigned long pfn);
+int __node_distance(int a, int b);
+void numa_update_cpu_topology(void);
+
+extern cpumask_t node_to_cpumask_map[MAX_NUMNODES];
+extern int numa_debug_enabled;
+
+#else
+
+static inline void numa_setup(void) { }
+static inline void numa_update_cpu_topology(void) { }
+static inline int numa_pfn_to_nid(unsigned long pfn)
+{
+	return 0;
+}
+
+#endif /* CONFIG_NUMA */
+#endif /* _ASM_S390_NUMA_H */
diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h
index a648338c4..34d960353 100644
--- a/arch/s390/include/asm/pci.h
+++ b/arch/s390/include/asm/pci.h
@@ -170,7 +170,11 @@ static inline void zpci_exit_slot(struct zpci_dev *zdev) {}
 #endif /* CONFIG_HOTPLUG_PCI_S390 */
 
 /* Helpers */
-struct zpci_dev *get_zdev(struct pci_dev *);
+static inline struct zpci_dev *to_zpci(struct pci_dev *pdev)
+{
+	return pdev->sysdata;
+}
+
 struct zpci_dev *get_zdev_by_fid(u32);
 
 /* DMA */
@@ -188,4 +192,20 @@ 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 *);
 
+#ifdef CONFIG_NUMA
+
+/* Returns the node based on PCI bus */
+static inline int __pcibus_to_node(const struct pci_bus *bus)
+{
+	return NUMA_NO_NODE;
+}
+
+static inline const struct cpumask *
+cpumask_of_pcibus(const struct pci_bus *bus)
+{
+	return cpu_online_mask;
+}
+
+#endif /* CONFIG_NUMA */
+
 #endif
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index f66d82798..bdb2f5112 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -576,6 +576,19 @@ static inline int pte_same(pte_t a, pte_t b)
 	return pte_val(a) == pte_val(b);
 }
 
+#ifdef CONFIG_NUMA_BALANCING
+static inline int pte_protnone(pte_t pte)
+{
+	return pte_present(pte) && !(pte_val(pte) & _PAGE_READ);
+}
+
+static inline int pmd_protnone(pmd_t pmd)
+{
+	/* pmd_large(pmd) implies pmd_present(pmd) */
+	return pmd_large(pmd) && !(pmd_val(pmd) & _SEGMENT_ENTRY_READ);
+}
+#endif
+
 static inline pgste_t pgste_get_lock(pte_t *ptep)
 {
 	unsigned long new = 0;
diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h
index dedb62185..085fb0d3c 100644
--- a/arch/s390/include/asm/processor.h
+++ b/arch/s390/include/asm/processor.h
@@ -14,10 +14,12 @@
 #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_FPU			3	/* restore vector registers */
 
 #define _CIF_MCCK_PENDING	(1<<CIF_MCCK_PENDING)
 #define _CIF_ASCE		(1<<CIF_ASCE)
 #define _CIF_NOHZ_DELAY		(1<<CIF_NOHZ_DELAY)
+#define _CIF_FPU		(1<<CIF_FPU)
 
 #ifndef __ASSEMBLY__
 
@@ -28,6 +30,7 @@
 #include <asm/ptrace.h>
 #include <asm/setup.h>
 #include <asm/runtime_instr.h>
+#include <asm/fpu-internal.h>
 
 static inline void set_cpu_flag(int flag)
 {
@@ -85,7 +88,7 @@ typedef struct {
  * Thread structure
  */
 struct thread_struct {
-	s390_fp_regs fp_regs;
+	struct fpu fpu;			/* FP and VX register save area */
 	unsigned int  acrs[NUM_ACRS];
         unsigned long ksp;              /* kernel stack pointer             */
 	mm_segment_t mm_segment;
@@ -101,7 +104,6 @@ struct thread_struct {
 	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. */
@@ -231,6 +233,17 @@ static inline void __load_psw_mask (unsigned long mask)
 }
 
 /*
+ * Extract current PSW mask
+ */
+static inline unsigned long __extract_psw(void)
+{
+	unsigned int reg1, reg2;
+
+	asm volatile("epsw %0,%1" : "=d" (reg1), "=a" (reg2));
+	return (((unsigned long) reg1) << 32) | ((unsigned long) reg2);
+}
+
+/*
  * Rewind PSW instruction address by specified number of bytes.
  */
 static inline unsigned long __rewind_psw(psw_t psw, unsigned long ilc)
@@ -336,25 +349,6 @@ extern void memcpy_absolute(void *, void *, size_t);
 	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/sclp.h b/arch/s390/include/asm/sclp.h
index f6ff06077..821dde5f4 100644
--- a/arch/s390/include/asm/sclp.h
+++ b/arch/s390/include/asm/sclp.h
@@ -79,6 +79,6 @@ 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);
 void sclp_early_detect(void);
-long _sclp_print_early(const char *);
+int _sclp_print_early(const char *);
 
 #endif /* _ASM_S390_SCLP_H */
diff --git a/arch/s390/include/asm/switch_to.h b/arch/s390/include/asm/switch_to.h
index d62e7a696..dcadfde32 100644
--- a/arch/s390/include/asm/switch_to.h
+++ b/arch/s390/include/asm/switch_to.h
@@ -8,139 +8,12 @@
 #define __ASM_SWITCH_TO_H
 
 #include <linux/thread_info.h>
+#include <asm/fpu-internal.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;
@@ -157,15 +30,13 @@ static inline void restore_access_regs(unsigned int *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_fpu_regs();					\
 		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);				\
+		set_cpu_flag(CIF_FPU);					\
 		restore_access_regs(&next->thread.acrs[0]);		\
 		restore_ri_cb(next->thread.ri_cb, prev->thread.ri_cb);	\
 	}								\
diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h
index 4990f6c66..94fc55fc7 100644
--- a/arch/s390/include/asm/topology.h
+++ b/arch/s390/include/asm/topology.h
@@ -2,6 +2,7 @@
 #define _ASM_S390_TOPOLOGY_H
 
 #include <linux/cpumask.h>
+#include <asm/numa.h>
 
 struct sysinfo_15_1_x;
 struct cpu;
@@ -13,6 +14,7 @@ struct cpu_topology_s390 {
 	unsigned short core_id;
 	unsigned short socket_id;
 	unsigned short book_id;
+	unsigned short node_id;
 	cpumask_t thread_mask;
 	cpumask_t core_mask;
 	cpumask_t book_mask;
@@ -52,6 +54,43 @@ static inline void topology_expect_change(void) { }
 #define POLARIZATION_VM		(2)
 #define POLARIZATION_VH		(3)
 
+#define SD_BOOK_INIT	SD_CPU_INIT
+
+#ifdef CONFIG_NUMA
+
+#define cpu_to_node cpu_to_node
+static inline int cpu_to_node(int cpu)
+{
+	return per_cpu(cpu_topology, cpu).node_id;
+}
+
+/* Returns a pointer to the cpumask of CPUs on node 'node'. */
+#define cpumask_of_node cpumask_of_node
+static inline const struct cpumask *cpumask_of_node(int node)
+{
+	return &node_to_cpumask_map[node];
+}
+
+/*
+ * Returns the number of the node containing node 'node'. This
+ * architecture is flat, so it is a pretty simple function!
+ */
+#define parent_node(node) (node)
+
+#define pcibus_to_node(bus) __pcibus_to_node(bus)
+
+#define node_distance(a, b) __node_distance(a, b)
+
+#else /* !CONFIG_NUMA */
+
+#define numa_node_id numa_node_id
+static inline int numa_node_id(void)
+{
+	return 0;
+}
+
+#endif /* CONFIG_NUMA */
+
 #include <asm-generic/topology.h>
 
 #endif /* _ASM_S390_TOPOLOGY_H */
diff --git a/arch/s390/include/asm/unistd.h b/arch/s390/include/asm/unistd.h
index 91f56b1d8..02613bad8 100644
--- a/arch/s390/include/asm/unistd.h
+++ b/arch/s390/include/asm/unistd.h
@@ -8,20 +8,8 @@
 
 #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
diff --git a/arch/s390/include/asm/vx-insn.h b/arch/s390/include/asm/vx-insn.h
new file mode 100644
index 000000000..4a3135620
--- /dev/null
+++ b/arch/s390/include/asm/vx-insn.h
@@ -0,0 +1,480 @@
+/*
+ * Support for Vector Instructions
+ *
+ * Assembler macros to generate .byte/.word code for particular
+ * vector instructions that are supported by recent binutils (>= 2.26) only.
+ *
+ * Copyright IBM Corp. 2015
+ * Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
+ */
+
+#ifndef __ASM_S390_VX_INSN_H
+#define __ASM_S390_VX_INSN_H
+
+#ifdef __ASSEMBLY__
+
+
+/* Macros to generate vector instruction byte code */
+
+#define REG_NUM_INVALID	       255
+
+/* GR_NUM - Retrieve general-purpose register number
+ *
+ * @opd:	Operand to store register number
+ * @r64:	String designation register in the format "%rN"
+ */
+.macro	GR_NUM	opd gr
+	\opd = REG_NUM_INVALID
+	.ifc \gr,%r0
+		\opd = 0
+	.endif
+	.ifc \gr,%r1
+		\opd = 1
+	.endif
+	.ifc \gr,%r2
+		\opd = 2
+	.endif
+	.ifc \gr,%r3
+		\opd = 3
+	.endif
+	.ifc \gr,%r4
+		\opd = 4
+	.endif
+	.ifc \gr,%r5
+		\opd = 5
+	.endif
+	.ifc \gr,%r6
+		\opd = 6
+	.endif
+	.ifc \gr,%r7
+		\opd = 7
+	.endif
+	.ifc \gr,%r8
+		\opd = 8
+	.endif
+	.ifc \gr,%r9
+		\opd = 9
+	.endif
+	.ifc \gr,%r10
+		\opd = 10
+	.endif
+	.ifc \gr,%r11
+		\opd = 11
+	.endif
+	.ifc \gr,%r12
+		\opd = 12
+	.endif
+	.ifc \gr,%r13
+		\opd = 13
+	.endif
+	.ifc \gr,%r14
+		\opd = 14
+	.endif
+	.ifc \gr,%r15
+		\opd = 15
+	.endif
+	.if \opd == REG_NUM_INVALID
+		.error "Invalid general-purpose register designation: \gr"
+	.endif
+.endm
+
+/* VX_R() - Macro to encode the VX_NUM into the instruction */
+#define VX_R(v)		(v & 0x0F)
+
+/* VX_NUM - Retrieve vector register number
+ *
+ * @opd:	Operand to store register number
+ * @vxr:	String designation register in the format "%vN"
+ *
+ * The vector register number is used for as input number to the
+ * instruction and, as well as, to compute the RXB field of the
+ * instruction.  To encode the particular vector register number,
+ * use the VX_R(v) macro to extract the instruction opcode.
+ */
+.macro	VX_NUM	opd vxr
+	\opd = REG_NUM_INVALID
+	.ifc \vxr,%v0
+		\opd = 0
+	.endif
+	.ifc \vxr,%v1
+		\opd = 1
+	.endif
+	.ifc \vxr,%v2
+		\opd = 2
+	.endif
+	.ifc \vxr,%v3
+		\opd = 3
+	.endif
+	.ifc \vxr,%v4
+		\opd = 4
+	.endif
+	.ifc \vxr,%v5
+		\opd = 5
+	.endif
+	.ifc \vxr,%v6
+		\opd = 6
+	.endif
+	.ifc \vxr,%v7
+		\opd = 7
+	.endif
+	.ifc \vxr,%v8
+		\opd = 8
+	.endif
+	.ifc \vxr,%v9
+		\opd = 9
+	.endif
+	.ifc \vxr,%v10
+		\opd = 10
+	.endif
+	.ifc \vxr,%v11
+		\opd = 11
+	.endif
+	.ifc \vxr,%v12
+		\opd = 12
+	.endif
+	.ifc \vxr,%v13
+		\opd = 13
+	.endif
+	.ifc \vxr,%v14
+		\opd = 14
+	.endif
+	.ifc \vxr,%v15
+		\opd = 15
+	.endif
+	.ifc \vxr,%v16
+		\opd = 16
+	.endif
+	.ifc \vxr,%v17
+		\opd = 17
+	.endif
+	.ifc \vxr,%v18
+		\opd = 18
+	.endif
+	.ifc \vxr,%v19
+		\opd = 19
+	.endif
+	.ifc \vxr,%v20
+		\opd = 20
+	.endif
+	.ifc \vxr,%v21
+		\opd = 21
+	.endif
+	.ifc \vxr,%v22
+		\opd = 22
+	.endif
+	.ifc \vxr,%v23
+		\opd = 23
+	.endif
+	.ifc \vxr,%v24
+		\opd = 24
+	.endif
+	.ifc \vxr,%v25
+		\opd = 25
+	.endif
+	.ifc \vxr,%v26
+		\opd = 26
+	.endif
+	.ifc \vxr,%v27
+		\opd = 27
+	.endif
+	.ifc \vxr,%v28
+		\opd = 28
+	.endif
+	.ifc \vxr,%v29
+		\opd = 29
+	.endif
+	.ifc \vxr,%v30
+		\opd = 30
+	.endif
+	.ifc \vxr,%v31
+		\opd = 31
+	.endif
+	.if \opd == REG_NUM_INVALID
+		.error "Invalid vector register designation: \vxr"
+	.endif
+.endm
+
+/* RXB - Compute most significant bit used vector registers
+ *
+ * @rxb:	Operand to store computed RXB value
+ * @v1:		First vector register designated operand
+ * @v2:		Second vector register designated operand
+ * @v3:		Third vector register designated operand
+ * @v4:		Fourth vector register designated operand
+ */
+.macro	RXB	rxb v1 v2=0 v3=0 v4=0
+	\rxb = 0
+	.if \v1 & 0x10
+		\rxb = \rxb | 0x08
+	.endif
+	.if \v2 & 0x10
+		\rxb = \rxb | 0x04
+	.endif
+	.if \v3 & 0x10
+		\rxb = \rxb | 0x02
+	.endif
+	.if \v4 & 0x10
+		\rxb = \rxb | 0x01
+	.endif
+.endm
+
+/* MRXB - Generate Element Size Control and RXB value
+ *
+ * @m:		Element size control
+ * @v1:		First vector register designated operand (for RXB)
+ * @v2:		Second vector register designated operand (for RXB)
+ * @v3:		Third vector register designated operand (for RXB)
+ * @v4:		Fourth vector register designated operand (for RXB)
+ */
+.macro	MRXB	m v1 v2=0 v3=0 v4=0
+	rxb = 0
+	RXB	rxb, \v1, \v2, \v3, \v4
+	.byte	(\m << 4) | rxb
+.endm
+
+/* MRXBOPC - Generate Element Size Control, RXB, and final Opcode fields
+ *
+ * @m:		Element size control
+ * @opc:	Opcode
+ * @v1:		First vector register designated operand (for RXB)
+ * @v2:		Second vector register designated operand (for RXB)
+ * @v3:		Third vector register designated operand (for RXB)
+ * @v4:		Fourth vector register designated operand (for RXB)
+ */
+.macro	MRXBOPC	m opc v1 v2=0 v3=0 v4=0
+	MRXB	\m, \v1, \v2, \v3, \v4
+	.byte	\opc
+.endm
+
+/* Vector support instructions */
+
+/* VECTOR GENERATE BYTE MASK */
+.macro	VGBM	vr imm2
+	VX_NUM	v1, \vr
+	.word	(0xE700 | (VX_R(v1) << 4))
+	.word	\imm2
+	MRXBOPC	0, 0x44, v1
+.endm
+.macro	VZERO	vxr
+	VGBM	\vxr, 0
+.endm
+.macro	VONE	vxr
+	VGBM	\vxr, 0xFFFF
+.endm
+
+/* VECTOR LOAD VR ELEMENT FROM GR */
+.macro	VLVG	v, gr, disp, m
+	VX_NUM	v1, \v
+	GR_NUM	b2, "%r0"
+	GR_NUM	r3, \gr
+	.word	0xE700 | (VX_R(v1) << 4) | r3
+	.word	(b2 << 12) | (\disp)
+	MRXBOPC	\m, 0x22, v1
+.endm
+.macro	VLVGB	v, gr, index, base
+	VLVG	\v, \gr, \index, \base, 0
+.endm
+.macro	VLVGH	v, gr, index
+	VLVG	\v, \gr, \index, 1
+.endm
+.macro	VLVGF	v, gr, index
+	VLVG	\v, \gr, \index, 2
+.endm
+.macro	VLVGG	v, gr, index
+	VLVG	\v, \gr, \index, 3
+.endm
+
+/* VECTOR LOAD */
+.macro	VL	v, disp, index="%r0", base
+	VX_NUM	v1, \v
+	GR_NUM	x2, \index
+	GR_NUM	b2, \base
+	.word	0xE700 | (VX_R(v1) << 4) | x2
+	.word	(b2 << 12) | (\disp)
+	MRXBOPC 0, 0x06, v1
+.endm
+
+/* VECTOR LOAD ELEMENT */
+.macro	VLEx	vr1, disp, index="%r0", base, m3, opc
+	VX_NUM	v1, \vr1
+	GR_NUM	x2, \index
+	GR_NUM	b2, \base
+	.word	0xE700 | (VX_R(v1) << 4) | x2
+	.word	(b2 << 12) | (\disp)
+	MRXBOPC	\m3, \opc, v1
+.endm
+.macro	VLEB	vr1, disp, index="%r0", base, m3
+	VLEx	\vr1, \disp, \index, \base, \m3, 0x00
+.endm
+.macro	VLEH	vr1, disp, index="%r0", base, m3
+	VLEx	\vr1, \disp, \index, \base, \m3, 0x01
+.endm
+.macro	VLEF	vr1, disp, index="%r0", base, m3
+	VLEx	\vr1, \disp, \index, \base, \m3, 0x03
+.endm
+.macro	VLEG	vr1, disp, index="%r0", base, m3
+	VLEx	\vr1, \disp, \index, \base, \m3, 0x02
+.endm
+
+/* VECTOR LOAD ELEMENT IMMEDIATE */
+.macro	VLEIx	vr1, imm2, m3, opc
+	VX_NUM	v1, \vr1
+	.word	0xE700 | (VX_R(v1) << 4)
+	.word	\imm2
+	MRXBOPC	\m3, \opc, v1
+.endm
+.macro	VLEIB	vr1, imm2, index
+	VLEIx	\vr1, \imm2, \index, 0x40
+.endm
+.macro	VLEIH	vr1, imm2, index
+	VLEIx	\vr1, \imm2, \index, 0x41
+.endm
+.macro	VLEIF	vr1, imm2, index
+	VLEIx	\vr1, \imm2, \index, 0x43
+.endm
+.macro	VLEIG	vr1, imm2, index
+	VLEIx	\vr1, \imm2, \index, 0x42
+.endm
+
+/* VECTOR LOAD GR FROM VR ELEMENT */
+.macro	VLGV	gr, vr, disp, base="%r0", m
+	GR_NUM	r1, \gr
+	GR_NUM	b2, \base
+	VX_NUM	v3, \vr
+	.word	0xE700 | (r1 << 4) | VX_R(v3)
+	.word	(b2 << 12) | (\disp)
+	MRXBOPC	\m, 0x21, v3
+.endm
+.macro	VLGVB	gr, vr, disp, base="%r0"
+	VLGV	\gr, \vr, \disp, \base, 0
+.endm
+.macro	VLGVH	gr, vr, disp, base="%r0"
+	VLGV	\gr, \vr, \disp, \base, 1
+.endm
+.macro	VLGVF	gr, vr, disp, base="%r0"
+	VLGV	\gr, \vr, \disp, \base, 2
+.endm
+.macro	VLGVG	gr, vr, disp, base="%r0"
+	VLGV	\gr, \vr, \disp, \base, 3
+.endm
+
+/* VECTOR LOAD MULTIPLE */
+.macro	VLM	vfrom, vto, disp, base
+	VX_NUM	v1, \vfrom
+	VX_NUM	v3, \vto
+	GR_NUM	b2, \base	    /* Base register */
+	.word	0xE700 | (VX_R(v1) << 4) | VX_R(v3)
+	.word	(b2 << 12) | (\disp)
+	MRXBOPC	0, 0x36, v1, v3
+.endm
+
+/* VECTOR STORE MULTIPLE */
+.macro	VSTM	vfrom, vto, disp, base
+	VX_NUM	v1, \vfrom
+	VX_NUM	v3, \vto
+	GR_NUM	b2, \base	    /* Base register */
+	.word	0xE700 | (VX_R(v1) << 4) | VX_R(v3)
+	.word	(b2 << 12) | (\disp)
+	MRXBOPC	0, 0x3E, v1, v3
+.endm
+
+/* VECTOR PERMUTE */
+.macro	VPERM	vr1, vr2, vr3, vr4
+	VX_NUM	v1, \vr1
+	VX_NUM	v2, \vr2
+	VX_NUM	v3, \vr3
+	VX_NUM	v4, \vr4
+	.word	0xE700 | (VX_R(v1) << 4) | VX_R(v2)
+	.word	(VX_R(v3) << 12)
+	MRXBOPC	VX_R(v4), 0x8C, v1, v2, v3, v4
+.endm
+
+/* VECTOR UNPACK LOGICAL LOW */
+.macro	VUPLL	vr1, vr2, m3
+	VX_NUM	v1, \vr1
+	VX_NUM	v2, \vr2
+	.word	0xE700 | (VX_R(v1) << 4) | VX_R(v2)
+	.word	0x0000
+	MRXBOPC	\m3, 0xD4, v1, v2
+.endm
+.macro	VUPLLB	vr1, vr2
+	VUPLL	\vr1, \vr2, 0
+.endm
+.macro	VUPLLH	vr1, vr2
+	VUPLL	\vr1, \vr2, 1
+.endm
+.macro	VUPLLF	vr1, vr2
+	VUPLL	\vr1, \vr2, 2
+.endm
+
+
+/* Vector integer instructions */
+
+/* VECTOR EXCLUSIVE OR */
+.macro	VX	vr1, vr2, vr3
+	VX_NUM	v1, \vr1
+	VX_NUM	v2, \vr2
+	VX_NUM	v3, \vr3
+	.word	0xE700 | (VX_R(v1) << 4) | VX_R(v2)
+	.word	(VX_R(v3) << 12)
+	MRXBOPC	0, 0x6D, v1, v2, v3
+.endm
+
+/* VECTOR GALOIS FIELD MULTIPLY SUM */
+.macro	VGFM	vr1, vr2, vr3, m4
+	VX_NUM	v1, \vr1
+	VX_NUM	v2, \vr2
+	VX_NUM	v3, \vr3
+	.word	0xE700 | (VX_R(v1) << 4) | VX_R(v2)
+	.word	(VX_R(v3) << 12)
+	MRXBOPC	\m4, 0xB4, v1, v2, v3
+.endm
+.macro	VGFMB	vr1, vr2, vr3
+	VGFM	\vr1, \vr2, \vr3, 0
+.endm
+.macro	VGFMH	vr1, vr2, vr3
+	VGFM	\vr1, \vr2, \vr3, 1
+.endm
+.macro	VGFMF	vr1, vr2, vr3
+	VGFM	\vr1, \vr2, \vr3, 2
+.endm
+.macro	VGFMG	vr1, vr2, vr3
+	VGFM	\vr1, \vr2, \vr3, 3
+.endm
+
+/* VECTOR GALOIS FIELD MULTIPLY SUM AND ACCUMULATE */
+.macro	VGFMA	vr1, vr2, vr3, vr4, m5
+	VX_NUM	v1, \vr1
+	VX_NUM	v2, \vr2
+	VX_NUM	v3, \vr3
+	VX_NUM	v4, \vr4
+	.word	0xE700 | (VX_R(v1) << 4) | VX_R(v2)
+	.word	(VX_R(v3) << 12) | (\m5 << 8)
+	MRXBOPC	VX_R(v4), 0xBC, v1, v2, v3, v4
+.endm
+.macro	VGFMAB	vr1, vr2, vr3, vr4
+	VGFMA	\vr1, \vr2, \vr3, \vr4, 0
+.endm
+.macro	VGFMAH	vr1, vr2, vr3, vr4
+	VGFMA	\vr1, \vr2, \vr3, \vr4, 1
+.endm
+.macro	VGFMAF	vr1, vr2, vr3, vr4
+	VGFMA	\vr1, \vr2, \vr3, \vr4, 2
+.endm
+.macro	VGFMAG	vr1, vr2, vr3, vr4
+	VGFMA	\vr1, \vr2, \vr3, \vr4, 3
+.endm
+
+/* VECTOR SHIFT RIGHT LOGICAL BY BYTE */
+.macro	VSRLB	vr1, vr2, vr3
+	VX_NUM	v1, \vr1
+	VX_NUM	v2, \vr2
+	VX_NUM	v3, \vr3
+	.word	0xE700 | (VX_R(v1) << 4) | VX_R(v2)
+	.word	(VX_R(v3) << 12)
+	MRXBOPC	0, 0x7D, v1, v2, v3
+.endm
+
+
+#endif	/* __ASSEMBLY__ */
+#endif	/* __ASM_S390_VX_INSN_H */
diff --git a/arch/s390/include/uapi/asm/unistd.h b/arch/s390/include/uapi/asm/unistd.h
index 67878af25..a848adba1 100644
--- a/arch/s390/include/uapi/asm/unistd.h
+++ b/arch/s390/include/uapi/asm/unistd.h
@@ -204,9 +204,9 @@
 #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_mbind		268
+#define __NR_get_mempolicy	269
+#define __NR_set_mempolicy	270
 #define __NR_mq_open		271
 #define __NR_mq_unlink		272
 #define __NR_mq_timedsend	273
@@ -223,7 +223,7 @@
 #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_migrate_pages	287
 #define __NR_openat		288
 #define __NR_mkdirat		289
 #define __NR_mknodat		290
@@ -245,7 +245,7 @@
 #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_move_pages		310
 #define __NR_getcpu		311
 #define __NR_epoll_pwait	312
 #define __NR_utimes		313
@@ -290,7 +290,26 @@
 #define __NR_s390_pci_mmio_write	352
 #define __NR_s390_pci_mmio_read		353
 #define __NR_execveat		354
-#define NR_syscalls 355
+#define __NR_userfaultfd	355
+#define __NR_membarrier		356
+#define __NR_recvmmsg		357
+#define __NR_sendmmsg		358
+#define __NR_socket		359
+#define __NR_socketpair		360
+#define __NR_bind		361
+#define __NR_connect		362
+#define __NR_listen		363
+#define __NR_accept4		364
+#define __NR_getsockopt		365
+#define __NR_setsockopt		366
+#define __NR_getsockname	367
+#define __NR_getpeername	368
+#define __NR_sendto		369
+#define __NR_sendmsg		370
+#define __NR_recvfrom		371
+#define __NR_recvmsg		372
+#define __NR_shutdown		373
+#define NR_syscalls 374
 
 /* 
  * There are some system calls that are not present on 64 bit, some
diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index ffb87617a..b756c6348 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -28,6 +28,17 @@ CFLAGS_ptrace.o		+= -DUTS_MACHINE='"$(UTS_MACHINE)"'
 
 CFLAGS_sysinfo.o += -w
 
+#
+# Use -march=z900 for sclp.c to be able to print an error message if
+# the kernel is started on a machine which is too old
+#
+CFLAGS_REMOVE_sclp.o = $(CC_FLAGS_FTRACE)
+ifneq ($(CC_FLAGS_MARCH),-march=z900)
+CFLAGS_REMOVE_sclp.o += $(CC_FLAGS_MARCH)
+CFLAGS_sclp.o	+= -march=z900
+endif
+GCOV_PROFILE_sclp.o := n
+
 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
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index a2da259d9..3aeeb1b56 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -28,6 +28,9 @@ int main(void)
 	DEFINE(__TASK_pid, offsetof(struct task_struct, pid));
 	BLANK();
 	DEFINE(__THREAD_ksp, offsetof(struct thread_struct, ksp));
+	DEFINE(__THREAD_FPU_fpc, offsetof(struct thread_struct, fpu.fpc));
+	DEFINE(__THREAD_FPU_flags, offsetof(struct thread_struct, fpu.flags));
+	DEFINE(__THREAD_FPU_regs, offsetof(struct thread_struct, fpu.regs));
 	DEFINE(__THREAD_per_cause, offsetof(struct thread_struct, per_event.cause));
 	DEFINE(__THREAD_per_address, offsetof(struct thread_struct, per_event.address));
 	DEFINE(__THREAD_per_paid, offsetof(struct thread_struct, per_event.paid));
@@ -173,6 +176,7 @@ int main(void)
 	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_PERCPU_OFFSET, offsetof(struct _lowcore, percpu_offset));
 	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));
diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c
index c78ba51ae..e0f9d270b 100644
--- a/arch/s390/kernel/compat_signal.c
+++ b/arch/s390/kernel/compat_signal.c
@@ -166,33 +166,14 @@ int copy_siginfo_from_user32(siginfo_t *to, compat_siginfo_t __user *from)
 /* Store registers needed to create the signal frame */
 static void store_sigregs(void)
 {
-	int i;
-
 	save_access_regs(current->thread.acrs);
-	save_fp_ctl(&current->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);
+	save_fpu_regs();
 }
 
 /* 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)
@@ -209,8 +190,7 @@ static int save_sigregs32(struct pt_regs *regs, _sigregs32 __user *sregs)
 		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, &current->thread.fp_regs,
-	       sizeof(user_sregs.fpregs));
+	fpregs_store((_s390_fp_regs *) &user_sregs.fpregs, &current->thread.fpu);
 	if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs32)))
 		return -EFAULT;
 	return 0;
@@ -230,8 +210,8 @@ static int restore_sigregs32(struct pt_regs *regs,_sigregs32 __user *sregs)
 	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))
+	/* Test the floating-point-control word. */
+	if (test_fp_ctl(user_sregs.fpregs.fpc))
 		return -EINVAL;
 
 	/* Use regs->psw.mask instead of PSW_USER_BITS to preserve PER bit. */
@@ -248,9 +228,7 @@ static int restore_sigregs32(struct pt_regs *regs,_sigregs32 __user *sregs)
 		regs->gprs[i] = (__u64) user_sregs.regs.gprs[i];
 	memcpy(&current->thread.acrs, &user_sregs.regs.acrs,
 	       sizeof(current->thread.acrs));
-
-	memcpy(&current->thread.fp_regs, &user_sregs.fpregs,
-	       sizeof(current->thread.fp_regs));
+	fpregs_load((_s390_fp_regs *) &user_sregs.fpregs, &current->thread.fpu);
 
 	clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */
 	return 0;
@@ -271,13 +249,13 @@ static int save_sigregs_ext32(struct pt_regs *regs,
 		return -EFAULT;
 
 	/* Save vector registers to signal stack */
-	if (current->thread.vxrs) {
+	if (is_vx_task(current)) {
 		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			vxrs[i] = *((__u64 *)(current->thread.vxrs + i) + 1);
+			vxrs[i] = *((__u64 *)(current->thread.fpu.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,
+				   current->thread.fpu.vxrs + __NUM_VXRS_LOW,
 				   sizeof(sregs_ext->vxrs_high)))
 			return -EFAULT;
 	}
@@ -299,15 +277,15 @@ static int restore_sigregs_ext32(struct pt_regs *regs,
 		*(__u32 *)&regs->gprs[i] = gprs_high[i];
 
 	/* Restore vector registers from signal stack */
-	if (current->thread.vxrs) {
+	if (is_vx_task(current)) {
 		if (__copy_from_user(vxrs, &sregs_ext->vxrs_low,
 				     sizeof(sregs_ext->vxrs_low)) ||
-		    __copy_from_user(current->thread.vxrs + __NUM_VXRS_LOW,
+		    __copy_from_user(current->thread.fpu.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];
+			*((__u64 *)(current->thread.fpu.vxrs + i) + 1) = vxrs[i];
 	}
 	return 0;
 }
@@ -323,6 +301,7 @@ COMPAT_SYSCALL_DEFINE0(sigreturn)
 		goto badframe;
 	sigset32_to_sigset(cset.sig, set.sig);
 	set_current_blocked(&set);
+	save_fpu_regs();
 	if (restore_sigregs32(regs, &frame->sregs))
 		goto badframe;
 	if (restore_sigregs_ext32(regs, &frame->sregs_ext))
@@ -347,6 +326,7 @@ COMPAT_SYSCALL_DEFINE0(rt_sigreturn)
 	set_current_blocked(&set);
 	if (compat_restore_altstack(&frame->uc.uc_stack))
 		goto badframe;
+	save_fpu_regs();
 	if (restore_sigregs32(regs, &frame->uc.uc_mcontext))
 		goto badframe;
 	if (restore_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext))
@@ -490,7 +470,7 @@ static int setup_rt_frame32(struct ksignal *ksig, sigset_t *set,
 	 */
 	uc_flags = UC_GPRS_HIGH;
 	if (MACHINE_HAS_VX) {
-		if (current->thread.vxrs)
+		if (is_vx_task(current))
 			uc_flags |= UC_VXRS;
 	} else
 		frame_size -= sizeof(frame->uc.uc_mcontext_ext.vxrs_low) +
diff --git a/arch/s390/kernel/compat_wrapper.c b/arch/s390/kernel/compat_wrapper.c
index f8498dde6..09f194052 100644
--- a/arch/s390/kernel/compat_wrapper.c
+++ b/arch/s390/kernel/compat_wrapper.c
@@ -52,15 +52,13 @@
  * 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__));	\
-	}
+asmlinkage long sys##name(__MAP(x,__SC_DECL,__VA_ARGS__));			\
+asmlinkage long notrace compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__));\
+asmlinkage long notrace 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);
@@ -68,23 +66,16 @@ 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);
@@ -93,37 +84,23 @@ 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);
@@ -131,20 +108,11 @@ 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);
@@ -161,23 +129,16 @@ 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);
@@ -192,23 +153,11 @@ 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, unsigned long, tls);
-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);
@@ -220,3 +169,10 @@ COMPAT_SYSCALL_WRAP2(memfd_create, const char __user *, uname, unsigned int, fla
 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);
+COMPAT_SYSCALL_WRAP4(socketpair, int, family, int, type, int, protocol, int __user *, usockvec);
+COMPAT_SYSCALL_WRAP3(bind, int, fd, struct sockaddr __user *, umyaddr, int, addrlen);
+COMPAT_SYSCALL_WRAP3(connect, int, fd, struct sockaddr __user *, uservaddr, int, addrlen);
+COMPAT_SYSCALL_WRAP4(accept4, int, fd, struct sockaddr __user *, upeer_sockaddr, int __user *, upeer_addrlen, int, flags);
+COMPAT_SYSCALL_WRAP3(getsockname, int, fd, struct sockaddr __user *, usockaddr, int __user *, usockaddr_len);
+COMPAT_SYSCALL_WRAP3(getpeername, int, fd, struct sockaddr __user *, usockaddr, int __user *, usockaddr_len);
+COMPAT_SYSCALL_WRAP6(sendto, int, fd, void __user *, buff, size_t, len, unsigned int, flags, struct sockaddr __user *, addr, int, addr_len);
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 84062e7a7..582fe44ab 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -20,6 +20,8 @@
 #include <asm/page.h>
 #include <asm/sigp.h>
 #include <asm/irq.h>
+#include <asm/fpu-internal.h>
+#include <asm/vx-insn.h>
 
 __PT_R0      =	__PT_GPRS
 __PT_R1      =	__PT_GPRS + 8
@@ -46,10 +48,10 @@ _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)
+_CIF_WORK	= (_CIF_MCCK_PENDING | _CIF_ASCE | _CIF_FPU)
 _PIF_WORK	= (_PIF_PER_TRAP)
 
-#define BASED(name) name-system_call(%r13)
+#define BASED(name) name-cleanup_critical(%r13)
 
 	.macro	TRACE_IRQS_ON
 #ifdef CONFIG_TRACE_IRQFLAGS
@@ -73,38 +75,6 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 #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
@@ -113,7 +83,7 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 #endif
 	.endm
 
-	.macro	SWITCH_ASYNC savearea,stack,shift
+	.macro	SWITCH_ASYNC savearea,timer
 	tmhh	%r8,0x0001		# interrupting from user ?
 	jnz	1f
 	lgr	%r14,%r9
@@ -124,26 +94,28 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 	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?
+0:	lg	%r14,__LC_ASYNC_STACK	# are we already on the async stack?
 	slgr	%r14,%r15
-	srag	%r14,%r14,\shift
-	jnz	1f
-	CHECK_STACK 1<<\shift,\savearea
+	srag	%r14,%r14,STACK_SHIFT
+	jnz	2f
+	CHECK_STACK 1<<STACK_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)
+	j	3f
+1:	LAST_BREAK %r14
+	UPDATE_VTIME %r14,%r15,\timer
+2:	lg	%r15,__LC_ASYNC_STACK	# load async stack
+3:	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
+	.macro UPDATE_VTIME w1,w2,enter_timer
+	lg	\w1,__LC_EXIT_TIMER
+	lg	\w2,__LC_LAST_UPDATE_TIMER
+	slg	\w1,\enter_timer
+	slg	\w2,__LC_EXIT_TIMER
+	alg	\w1,__LC_USER_TIMER
+	alg	\w2,__LC_SYSTEM_TIMER
+	stg	\w1,__LC_USER_TIMER
+	stg	\w2,__LC_SYSTEM_TIMER
 	mvc	__LC_LAST_UPDATE_TIMER(8),\enter_timer
 	.endm
 
@@ -197,6 +169,69 @@ ENTRY(__switch_to)
 	br	%r14
 
 .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
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU	# load guest fp/vx registers ?
+	jno	.Lsie_load_guest_gprs
+	brasl	%r14,load_fpu_regs		# load guest fp/vx regs
+.Lsie_load_guest_gprs:
+	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),3		# last exit...
+	jnz	.Lsie_skip
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	jo	.Lsie_skip			# exit if fp/vx regs changed
+	tm	__LC_MACHINE_FLAGS+6,0x20	# MACHINE_FLAG_LPP
+	jz	.Lsie_enter
+	.insn	s,0xb2800000,__LC_CURRENT_PID	# set guest id to pid
+.Lsie_enter:
+	sie	0(%r14)
+	tm	__LC_MACHINE_FLAGS+6,0x20	# MACHINE_FLAG_LPP
+	jz	.Lsie_skip
+	.insn	s,0xb2800000,__SF_EMPTY+16(%r15)# set host id
+.Lsie_skip:
+	ni	__SIE_PROG0C+3(%r14),0xfe	# no longer in SIE
+	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
+.Lsie_done:
+# 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 .Lcleanup_sie
+.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
+
+	EX_TABLE(.Lrewind_pad,.Lsie_fault)
+	EX_TABLE(sie_exit,.Lsie_fault)
+#endif
+
 /*
  * SVC interrupt handler routine. System calls are synchronous events and
  * are executed with interrupts enabled.
@@ -212,9 +247,9 @@ ENTRY(system_call)
 .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
+.Lsysc_vtime:
+	UPDATE_VTIME %r10,%r13,__LC_SYNC_ENTER_TIMER
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_SYNC
 	mvc	__PT_PSW(16,%r11),__LC_SVC_OLD_PSW
@@ -244,8 +279,6 @@ ENTRY(system_call)
 .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
@@ -280,6 +313,8 @@ ENTRY(system_call)
 	jo	.Lsysc_sigpending
 	tm	__TI_flags+7(%r12),_TIF_NOTIFY_RESUME
 	jo	.Lsysc_notify_resume
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	jo	.Lsysc_vxrs
 	tm	__LC_CPU_FLAGS+7,_CIF_ASCE
 	jo	.Lsysc_uaccess
 	j	.Lsysc_return		# beware of critical section cleanup
@@ -307,6 +342,13 @@ ENTRY(system_call)
 	j	.Lsysc_return
 
 #
+# CIF_FPU is set, restore floating-point controls and floating-point registers.
+#
+.Lsysc_vxrs:
+	larl	%r14,.Lsysc_return
+	jg	load_fpu_regs
+
+#
 # _TIF_SIGPENDING is set, call do_signal
 #
 .Lsysc_sigpending:
@@ -405,28 +447,35 @@ ENTRY(pgm_check_handler)
 	stmg	%r8,%r15,__LC_SAVE_AREA_SYNC
 	lg	%r10,__LC_LAST_BREAK
 	lg	%r12,__LC_THREAD_INFO
-	larl	%r13,system_call
+	larl	%r13,cleanup_critical
 	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
+	jnz	2f			# -> fault in user space
+#if IS_ENABLED(CONFIG_KVM)
+	# cleanup critical section for sie64a
+	lgr	%r14,%r9
+	slg	%r14,BASED(.Lsie_critical_start)
+	clg	%r14,BASED(.Lsie_critical_length)
+	jhe	0f
+	brasl	%r14,.Lcleanup_sie
+#endif
+0:	tmhh	%r8,0x4000		# PER bit set in old PSW ?
+	jnz	1f			# -> 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
+1:	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
+	j	3f
+2:	LAST_BREAK %r14
+	UPDATE_VTIME %r14,%r15,__LC_SYNC_ENTER_TIMER
 	lg	%r15,__LC_KERNEL_STACK
 	lg	%r14,__TI_task(%r12)
 	aghi	%r14,__TASK_thread	# pointer to thread_struct
 	lghi	%r13,__LC_PGM_TDB
 	tm	__LC_PGM_ILC+2,0x02	# check for transaction abort
-	jz	2f
+	jz	3f
 	mvc	__THREAD_trap_tdb(256,%r14),0(%r13)
-2:	la	%r11,STACK_FRAME_OVERHEAD(%r15)
+3:	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)
@@ -435,24 +484,28 @@ ENTRY(pgm_check_handler)
 	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
+	jz	4f
 	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
+4:	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
+	je	.Lpgm_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
+.Lpgm_return:
+	LOCKDEP_SYS_EXIT
+	tm	__PT_PSW+1(%r11),0x01	# returning to user ?
+	jno	.Lsysc_restore
+	j	.Lsysc_tif
 
 #
 # PER event in supervisor state, must be kprobes
@@ -462,7 +515,7 @@ ENTRY(pgm_check_handler)
 	xc	__SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	brasl	%r14,do_per_trap
-	j	.Lsysc_return
+	j	.Lpgm_return
 
 #
 # single stepped system call
@@ -483,15 +536,9 @@ ENTRY(io_int_handler)
 	stmg	%r8,%r15,__LC_SAVE_AREA_ASYNC
 	lg	%r10,__LC_LAST_BREAK
 	lg	%r12,__LC_THREAD_INFO
-	larl	%r13,system_call
+	larl	%r13,cleanup_critical
 	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:
+	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC
 	stmg	%r8,%r9,__PT_PSW(%r11)
@@ -587,6 +634,8 @@ ENTRY(io_int_handler)
 	jo	.Lio_sigpending
 	tm	__TI_flags+7(%r12),_TIF_NOTIFY_RESUME
 	jo	.Lio_notify_resume
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	jo	.Lio_vxrs
 	tm	__LC_CPU_FLAGS+7,_CIF_ASCE
 	jo	.Lio_uaccess
 	j	.Lio_return		# beware of critical section cleanup
@@ -609,6 +658,13 @@ ENTRY(io_int_handler)
 	j	.Lio_return
 
 #
+# CIF_FPU is set, restore floating-point controls and floating-point registers.
+#
+.Lio_vxrs:
+	larl	%r14,.Lio_return
+	jg	load_fpu_regs
+
+#
 # _TIF_NEED_RESCHED is set, call schedule
 #
 .Lio_reschedule:
@@ -652,15 +708,9 @@ ENTRY(ext_int_handler)
 	stmg	%r8,%r15,__LC_SAVE_AREA_ASYNC
 	lg	%r10,__LC_LAST_BREAK
 	lg	%r12,__LC_THREAD_INFO
-	larl	%r13,system_call
+	larl	%r13,cleanup_critical
 	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:
+	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC
 	stmg	%r8,%r9,__PT_PSW(%r11)
@@ -683,6 +733,14 @@ ENTRY(psw_idle)
 	stg	%r3,__SF_EMPTY(%r15)
 	larl	%r1,.Lpsw_idle_lpsw+4
 	stg	%r1,__SF_EMPTY+8(%r15)
+#ifdef CONFIG_SMP
+	larl	%r1,smp_cpu_mtid
+	llgf	%r1,0(%r1)
+	ltgr	%r1,%r1
+	jz	.Lpsw_idle_stcctm
+	.insn	rsy,0xeb0000000017,%r1,5,__SF_EMPTY+16(%r15)
+.Lpsw_idle_stcctm:
+#endif
 	STCK	__CLOCK_IDLE_ENTER(%r2)
 	stpt	__TIMER_IDLE_ENTER(%r2)
 .Lpsw_idle_lpsw:
@@ -690,6 +748,122 @@ ENTRY(psw_idle)
 	br	%r14
 .Lpsw_idle_end:
 
+/* Store floating-point controls and floating-point or vector extension
+ * registers instead.  A critical section cleanup assures that the registers
+ * are stored even if interrupted for some other work.	The register %r2
+ * designates a struct fpu to store register contents.	If the specified
+ * structure does not contain a register save area, the register store is
+ * omitted (see also comments in arch_dup_task_struct()).
+ *
+ * The CIF_FPU flag is set in any case.  The CIF_FPU triggers a lazy restore
+ * of the register contents at system call or io return.
+ */
+ENTRY(save_fpu_regs)
+	lg	%r2,__LC_CURRENT
+	aghi	%r2,__TASK_thread
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	bor	%r14
+	stfpc	__THREAD_FPU_fpc(%r2)
+.Lsave_fpu_regs_fpc_end:
+	lg	%r3,__THREAD_FPU_regs(%r2)
+	ltgr	%r3,%r3
+	jz	.Lsave_fpu_regs_done	  # no save area -> set CIF_FPU
+	tm	__THREAD_FPU_flags+3(%r2),FPU_USE_VX
+	jz	.Lsave_fpu_regs_fp	  # no -> store FP regs
+.Lsave_fpu_regs_vx_low:
+	VSTM	%v0,%v15,0,%r3		  # vstm 0,15,0(3)
+.Lsave_fpu_regs_vx_high:
+	VSTM	%v16,%v31,256,%r3	  # vstm 16,31,256(3)
+	j	.Lsave_fpu_regs_done	  # -> set CIF_FPU flag
+.Lsave_fpu_regs_fp:
+	std	0,0(%r3)
+	std	1,8(%r3)
+	std	2,16(%r3)
+	std	3,24(%r3)
+	std	4,32(%r3)
+	std	5,40(%r3)
+	std	6,48(%r3)
+	std	7,56(%r3)
+	std	8,64(%r3)
+	std	9,72(%r3)
+	std	10,80(%r3)
+	std	11,88(%r3)
+	std	12,96(%r3)
+	std	13,104(%r3)
+	std	14,112(%r3)
+	std	15,120(%r3)
+.Lsave_fpu_regs_done:
+	oi	__LC_CPU_FLAGS+7,_CIF_FPU
+	br	%r14
+.Lsave_fpu_regs_end:
+
+/* Load floating-point controls and floating-point or vector extension
+ * registers.  A critical section cleanup assures that the register contents
+ * are loaded even if interrupted for some other work.	Depending on the saved
+ * FP/VX state, the vector-enablement control, CR0.46, is either set or cleared.
+ *
+ * There are special calling conventions to fit into sysc and io return work:
+ *	%r15:	<kernel stack>
+ * The function requires:
+ *	%r4 and __SF_EMPTY+32(%r15)
+ */
+load_fpu_regs:
+	lg	%r4,__LC_CURRENT
+	aghi	%r4,__TASK_thread
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	bnor	%r14
+	lfpc	__THREAD_FPU_fpc(%r4)
+	stctg	%c0,%c0,__SF_EMPTY+32(%r15)	# store CR0
+	tm	__THREAD_FPU_flags+3(%r4),FPU_USE_VX	# VX-enabled task ?
+	lg	%r4,__THREAD_FPU_regs(%r4)	# %r4 <- reg save area
+	jz	.Lload_fpu_regs_fp_ctl		# -> no VX, load FP regs
+.Lload_fpu_regs_vx_ctl:
+	tm	__SF_EMPTY+32+5(%r15),2		# test VX control
+	jo	.Lload_fpu_regs_vx
+	oi	__SF_EMPTY+32+5(%r15),2		# set VX control
+	lctlg	%c0,%c0,__SF_EMPTY+32(%r15)
+.Lload_fpu_regs_vx:
+	VLM	%v0,%v15,0,%r4
+.Lload_fpu_regs_vx_high:
+	VLM	%v16,%v31,256,%r4
+	j	.Lload_fpu_regs_done
+.Lload_fpu_regs_fp_ctl:
+	tm	__SF_EMPTY+32+5(%r15),2		# test VX control
+	jz	.Lload_fpu_regs_fp
+	ni	__SF_EMPTY+32+5(%r15),253	# clear VX control
+	lctlg	%c0,%c0,__SF_EMPTY+32(%r15)
+.Lload_fpu_regs_fp:
+	ld	0,0(%r4)
+	ld	1,8(%r4)
+	ld	2,16(%r4)
+	ld	3,24(%r4)
+	ld	4,32(%r4)
+	ld	5,40(%r4)
+	ld	6,48(%r4)
+	ld	7,56(%r4)
+	ld	8,64(%r4)
+	ld	9,72(%r4)
+	ld	10,80(%r4)
+	ld	11,88(%r4)
+	ld	12,96(%r4)
+	ld	13,104(%r4)
+	ld	14,112(%r4)
+	ld	15,120(%r4)
+.Lload_fpu_regs_done:
+	ni	__LC_CPU_FLAGS+7,255-_CIF_FPU
+	br	%r14
+.Lload_fpu_regs_end:
+
+/* Test and set the vector enablement control in CR0.46 */
+ENTRY(__ctl_set_vx)
+	stctg	%c0,%c0,__SF_EMPTY(%r15)
+	tm	__SF_EMPTY+5(%r15),2
+	bor	%r14
+	oi	__SF_EMPTY+5(%r15),2
+	lctlg	%c0,%c0,__SF_EMPTY(%r15)
+	br	%r14
+.L__ctl_set_vx_end:
+
 .L__critical_end:
 
 /*
@@ -702,9 +876,8 @@ ENTRY(mcck_int_handler)
 	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
+	larl	%r13,cleanup_critical
 	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
@@ -725,11 +898,7 @@ ENTRY(mcck_int_handler)
 	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
+	SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_MCCK_ENTER_TIMER
 .Lmcck_skip:
 	lghi	%r14,__LC_GPREGS_SAVE_AREA+64
 	stmg	%r0,%r7,__PT_R0(%r11)
@@ -764,12 +933,8 @@ ENTRY(mcck_int_handler)
 	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)
+	aghi	%r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
 	j	.Lmcck_skip
 
 #
@@ -819,20 +984,13 @@ stack_overflow:
 	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:
+#if IS_ENABLED(CONFIG_KVM)
+	clg	%r9,BASED(.Lcleanup_table_sie)	# .Lsie_gmap
+	jl	0f
+	clg	%r9,BASED(.Lcleanup_table_sie+8)# .Lsie_done
+	jl	.Lcleanup_sie
+#endif
 	clg	%r9,BASED(.Lcleanup_table)	# system_call
 	jl	0f
 	clg	%r9,BASED(.Lcleanup_table+8)	# .Lsysc_do_svc
@@ -853,8 +1011,54 @@ cleanup_critical:
 	jl	0f
 	clg	%r9,BASED(.Lcleanup_table+72)	# .Lpsw_idle_end
 	jl	.Lcleanup_idle
+	clg	%r9,BASED(.Lcleanup_table+80)	# save_fpu_regs
+	jl	0f
+	clg	%r9,BASED(.Lcleanup_table+88)	# .Lsave_fpu_regs_end
+	jl	.Lcleanup_save_fpu_regs
+	clg	%r9,BASED(.Lcleanup_table+96)	# load_fpu_regs
+	jl	0f
+	clg	%r9,BASED(.Lcleanup_table+104)	# .Lload_fpu_regs_end
+	jl	.Lcleanup_load_fpu_regs
+	clg	%r9,BASED(.Lcleanup_table+112)	# __ctl_set_vx
+	jl	0f
+	clg	%r9,BASED(.Lcleanup_table+120)	# .L__ctl_set_vx_end
+	jl	.Lcleanup___ctl_set_vx
 0:	br	%r14
 
+	.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
+	.quad	save_fpu_regs
+	.quad	.Lsave_fpu_regs_end
+	.quad	load_fpu_regs
+	.quad	.Lload_fpu_regs_end
+	.quad	__ctl_set_vx
+	.quad	.L__ctl_set_vx_end
+
+#if IS_ENABLED(CONFIG_KVM)
+.Lcleanup_table_sie:
+	.quad	.Lsie_gmap
+	.quad	.Lsie_done
+
+.Lcleanup_sie:
+	lg	%r9,__SF_EMPTY(%r15)		# get control block pointer
+	tm	__LC_MACHINE_FLAGS+6,0x20	# MACHINE_FLAG_LPP
+	jz	0f
+	.insn	s,0xb2800000,__SF_EMPTY+16(%r15)# set host id
+0:	ni	__SIE_PROG0C+3(%r9),0xfe	# no longer in SIE
+	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
+	larl	%r9,sie_exit			# skip forward to sie_exit
+	br	%r14
+#endif
 
 .Lcleanup_system_call:
 	# check if stpt has been executed
@@ -915,7 +1119,7 @@ cleanup_critical:
 	.quad	system_call
 	.quad	.Lsysc_stmg
 	.quad	.Lsysc_per
-	.quad	.Lsysc_vtime+18
+	.quad	.Lsysc_vtime+36
 	.quad	.Lsysc_vtime+42
 
 .Lcleanup_sysc_tif:
@@ -963,7 +1167,27 @@ cleanup_critical:
 	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
+1:	# calculate idle cycles
+#ifdef CONFIG_SMP
+	clg	%r9,BASED(.Lcleanup_idle_insn)
+	jl	3f
+	larl	%r1,smp_cpu_mtid
+	llgf	%r1,0(%r1)
+	ltgr	%r1,%r1
+	jz	3f
+	.insn	rsy,0xeb0000000017,%r1,5,__SF_EMPTY+80(%r15)
+	larl	%r3,mt_cycles
+	ag	%r3,__LC_PERCPU_OFFSET
+	la	%r4,__SF_EMPTY+16(%r15)
+2:	lg	%r0,0(%r3)
+	slg	%r0,0(%r4)
+	alg	%r0,64(%r4)
+	stg	%r0,0(%r3)
+	la	%r3,8(%r3)
+	la	%r4,8(%r4)
+	brct	%r1,2b
+#endif
+3:	# account system time going idle
 	lg	%r9,__LC_STEAL_TIMER
 	alg	%r9,__CLOCK_IDLE_ENTER(%r2)
 	slg	%r9,__LC_LAST_UPDATE_CLOCK
@@ -981,6 +1205,147 @@ cleanup_critical:
 .Lcleanup_idle_insn:
 	.quad	.Lpsw_idle_lpsw
 
+.Lcleanup_save_fpu_regs:
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	bor	%r14
+	clg	%r9,BASED(.Lcleanup_save_fpu_regs_done)
+	jhe	5f
+	clg	%r9,BASED(.Lcleanup_save_fpu_regs_fp)
+	jhe	4f
+	clg	%r9,BASED(.Lcleanup_save_fpu_regs_vx_high)
+	jhe	3f
+	clg	%r9,BASED(.Lcleanup_save_fpu_regs_vx_low)
+	jhe	2f
+	clg	%r9,BASED(.Lcleanup_save_fpu_fpc_end)
+	jhe	1f
+	lg	%r2,__LC_CURRENT
+	aghi	%r2,__TASK_thread
+0:	# Store floating-point controls
+	stfpc	__THREAD_FPU_fpc(%r2)
+1:	# Load register save area and check if VX is active
+	lg	%r3,__THREAD_FPU_regs(%r2)
+	ltgr	%r3,%r3
+	jz	5f			  # no save area -> set CIF_FPU
+	tm	__THREAD_FPU_flags+3(%r2),FPU_USE_VX
+	jz	4f			  # no VX -> store FP regs
+2:	# Store vector registers (V0-V15)
+	VSTM	%v0,%v15,0,%r3		  # vstm 0,15,0(3)
+3:	# Store vector registers (V16-V31)
+	VSTM	%v16,%v31,256,%r3	  # vstm 16,31,256(3)
+	j	5f			  # -> done, set CIF_FPU flag
+4:	# Store floating-point registers
+	std	0,0(%r3)
+	std	1,8(%r3)
+	std	2,16(%r3)
+	std	3,24(%r3)
+	std	4,32(%r3)
+	std	5,40(%r3)
+	std	6,48(%r3)
+	std	7,56(%r3)
+	std	8,64(%r3)
+	std	9,72(%r3)
+	std	10,80(%r3)
+	std	11,88(%r3)
+	std	12,96(%r3)
+	std	13,104(%r3)
+	std	14,112(%r3)
+	std	15,120(%r3)
+5:	# Set CIF_FPU flag
+	oi	__LC_CPU_FLAGS+7,_CIF_FPU
+	lg	%r9,48(%r11)		# return from save_fpu_regs
+	br	%r14
+.Lcleanup_save_fpu_fpc_end:
+	.quad	.Lsave_fpu_regs_fpc_end
+.Lcleanup_save_fpu_regs_vx_low:
+	.quad	.Lsave_fpu_regs_vx_low
+.Lcleanup_save_fpu_regs_vx_high:
+	.quad	.Lsave_fpu_regs_vx_high
+.Lcleanup_save_fpu_regs_fp:
+	.quad	.Lsave_fpu_regs_fp
+.Lcleanup_save_fpu_regs_done:
+	.quad	.Lsave_fpu_regs_done
+
+.Lcleanup_load_fpu_regs:
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	bnor	%r14
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_done)
+	jhe	1f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_fp)
+	jhe	2f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_fp_ctl)
+	jhe	3f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_vx_high)
+	jhe	4f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_vx)
+	jhe	5f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_vx_ctl)
+	jhe	6f
+	lg	%r4,__LC_CURRENT
+	aghi	%r4,__TASK_thread
+	lfpc	__THREAD_FPU_fpc(%r4)
+	tm	__THREAD_FPU_flags+3(%r4),FPU_USE_VX	# VX-enabled task ?
+	lg	%r4,__THREAD_FPU_regs(%r4)	# %r4 <- reg save area
+	jz	3f				# -> no VX, load FP regs
+6:	# Set VX-enablement control
+	stctg	%c0,%c0,__SF_EMPTY+32(%r15)	# store CR0
+	tm	__SF_EMPTY+32+5(%r15),2		# test VX control
+	jo	5f
+	oi	__SF_EMPTY+32+5(%r15),2		# set VX control
+	lctlg	%c0,%c0,__SF_EMPTY+32(%r15)
+5:	# Load V0 ..V15 registers
+	VLM	%v0,%v15,0,%r4
+4:	# Load V16..V31 registers
+	VLM	%v16,%v31,256,%r4
+	j	1f
+3:	# Clear VX-enablement control for FP
+	stctg	%c0,%c0,__SF_EMPTY+32(%r15)	# store CR0
+	tm	__SF_EMPTY+32+5(%r15),2		# test VX control
+	jz	2f
+	ni	__SF_EMPTY+32+5(%r15),253	# clear VX control
+	lctlg	%c0,%c0,__SF_EMPTY+32(%r15)
+2:	# Load floating-point registers
+	ld	0,0(%r4)
+	ld	1,8(%r4)
+	ld	2,16(%r4)
+	ld	3,24(%r4)
+	ld	4,32(%r4)
+	ld	5,40(%r4)
+	ld	6,48(%r4)
+	ld	7,56(%r4)
+	ld	8,64(%r4)
+	ld	9,72(%r4)
+	ld	10,80(%r4)
+	ld	11,88(%r4)
+	ld	12,96(%r4)
+	ld	13,104(%r4)
+	ld	14,112(%r4)
+	ld	15,120(%r4)
+1:	# Clear CIF_FPU bit
+	ni	__LC_CPU_FLAGS+7,255-_CIF_FPU
+	lg	%r9,48(%r11)		# return from load_fpu_regs
+	br	%r14
+.Lcleanup_load_fpu_regs_vx_ctl:
+	.quad	.Lload_fpu_regs_vx_ctl
+.Lcleanup_load_fpu_regs_vx:
+	.quad	.Lload_fpu_regs_vx
+.Lcleanup_load_fpu_regs_vx_high:
+	.quad	.Lload_fpu_regs_vx_high
+.Lcleanup_load_fpu_regs_fp_ctl:
+	.quad	.Lload_fpu_regs_fp_ctl
+.Lcleanup_load_fpu_regs_fp:
+	.quad	.Lload_fpu_regs_fp
+.Lcleanup_load_fpu_regs_done:
+	.quad	.Lload_fpu_regs_done
+
+.Lcleanup___ctl_set_vx:
+	stctg	%c0,%c0,__SF_EMPTY(%r15)
+	tm	__SF_EMPTY+5(%r15),2
+	bor	%r14
+	oi	__SF_EMPTY+5(%r15),2
+	lctlg	%c0,%c0,__SF_EMPTY(%r15)
+	lg	%r9,48(%r11)		# return from __ctl_set_vx
+	br	%r14
+
 /*
  * Integer constants
  */
@@ -989,62 +1354,11 @@ cleanup_critical:
 	.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),3		# 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:
+.Lsie_critical_start:
 	.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"
diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S
index 59b7c6470..1255c6c53 100644
--- a/arch/s390/kernel/head.S
+++ b/arch/s390/kernel/head.S
@@ -370,6 +370,7 @@ ENTRY(startup_kdump)
 	xc	0x200(256),0x200	# partially clear lowcore
 	xc	0x300(256),0x300
 	xc	0xe00(256),0xe00
+	lctlg	%c0,%c15,0x200(%r0)	# initialize control registers
 	stck	__LC_LAST_UPDATE_CLOCK
 	spt	6f-.LPG0(%r13)
 	mvc	__LC_LAST_UPDATE_TIMER(8),6f-.LPG0(%r13)
@@ -413,9 +414,9 @@ ENTRY(startup_kdump)
 # followed by the facility words.
 
 #if defined(CONFIG_MARCH_Z13)
-	.long 3, 0xc100eff2, 0xf46ce800, 0x00400000
+	.long 2, 0xc100eff2, 0xf46cc800
 #elif defined(CONFIG_MARCH_ZEC12)
-	.long 3, 0xc100eff2, 0xf46ce800, 0x00400000
+	.long 2, 0xc100eff2, 0xf46cc800
 #elif defined(CONFIG_MARCH_Z196)
 	.long 2, 0xc100eff2, 0xf46c0000
 #elif defined(CONFIG_MARCH_Z10)
diff --git a/arch/s390/kernel/jump_label.c b/arch/s390/kernel/jump_label.c
index a90299600..083b05f5f 100644
--- a/arch/s390/kernel/jump_label.c
+++ b/arch/s390/kernel/jump_label.c
@@ -44,12 +44,9 @@ static void jump_label_bug(struct jump_entry *entry, struct insn *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]);
+	pr_emerg("Found:    %6ph\n", ipc);
+	pr_emerg("Expected: %6ph\n", ipe);
+	pr_emerg("New:      %6ph\n", ipn);
 	panic("Corrupted kernel text");
 }
 
@@ -64,7 +61,7 @@ static void __jump_label_transform(struct jump_entry *entry,
 {
 	struct insn old, new;
 
-	if (type == JUMP_LABEL_ENABLE) {
+	if (type == JUMP_LABEL_JMP) {
 		jump_label_make_nop(entry, &old);
 		jump_label_make_branch(entry, &new);
 	} else {
diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c
index 56b550893..0ae6f8e74 100644
--- a/arch/s390/kernel/nmi.c
+++ b/arch/s390/kernel/nmi.c
@@ -21,6 +21,7 @@
 #include <asm/nmi.h>
 #include <asm/crw.h>
 #include <asm/switch_to.h>
+#include <asm/fpu-internal.h>
 #include <asm/ctl_reg.h>
 
 struct mcck_struct {
@@ -164,8 +165,12 @@ static int notrace s390_revalidate_registers(struct mci *mci)
 		cr0.val = S390_lowcore.cregs_save_area[0];
 		cr0.afp = cr0.vx = 1;
 		__ctl_load(cr0.val, 0, 0);
-		restore_vx_regs((__vector128 *)
-				&S390_lowcore.vector_save_area);
+		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" (*(struct vx_array *)
+				 &S390_lowcore.vector_save_area) : "1");
 		__ctl_load(S390_lowcore.cregs_save_area[0], 0, 0);
 	}
 	/* Revalidate access registers */
@@ -358,4 +363,4 @@ static int __init machine_check_init(void)
 	ctl_set_bit(14, 24);	/* enable warning MCH */
 	return 0;
 }
-arch_initcall(machine_check_init);
+early_initcall(machine_check_init);
diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c
index 56fdad479..a9563409c 100644
--- a/arch/s390/kernel/perf_cpum_cf.c
+++ b/arch/s390/kernel/perf_cpum_cf.c
@@ -157,10 +157,14 @@ static int validate_ctr_auth(const struct hw_perf_event *hwc)
 
 	cpuhw = &get_cpu_var(cpu_hw_events);
 
-	/* check authorization for cpu counter sets */
+	/* Check authorization for cpu counter sets.
+	 * If the particular CPU counter set is not authorized,
+	 * return with -ENOENT in order to fall back to other
+	 * PMUs that might suffice the event request.
+	 */
 	ctrs_state = cpumf_state_ctl[hwc->config_base];
 	if (!(ctrs_state & cpuhw->info.auth_ctl))
-		err = -EPERM;
+		err = -ENOENT;
 
 	put_cpu_var(cpu_hw_events);
 	return err;
@@ -536,7 +540,7 @@ static int cpumf_pmu_add(struct perf_event *event, int flags)
 	 */
 	if (!(cpuhw->flags & PERF_EVENT_TXN))
 		if (validate_ctr_auth(&event->hw))
-			return -EPERM;
+			return -ENOENT;
 
 	ctr_set_enable(&cpuhw->state, event->hw.config_base);
 	event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
@@ -611,7 +615,7 @@ static int cpumf_pmu_commit_txn(struct pmu *pmu)
 	state = cpuhw->state & ~((1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1);
 	state >>= CPUMF_LCCTL_ENABLE_SHIFT;
 	if ((state & cpuhw->info.auth_ctl) != state)
-		return -EPERM;
+		return -ENOENT;
 
 	cpuhw->flags &= ~PERF_EVENT_TXN;
 	perf_pmu_enable(pmu);
diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c
index afe05bfb7..b973972f6 100644
--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -1019,12 +1019,9 @@ static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr)
 		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.
+	/* The host-program-parameter (hpp) contains the pid of
+	 * the CPU thread as set by sie64a() in entry.S.
+	 * If non-zero assume a guest sample.
 	 */
 	if (sfr->basic.hpp)
 		sde_regs->in_guest = 1;
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 8f587d871..f2dac9f07 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -81,8 +81,38 @@ 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);
+	/* Free either the floating-point or the vector register save area */
+	kfree(tsk->thread.fpu.regs);
+}
+
+int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
+{
+	*dst = *src;
+
+	/* Set up a new floating-point register save area */
+	dst->thread.fpu.fpc = 0;
+	dst->thread.fpu.flags = 0;	/* Always start with VX disabled */
+	dst->thread.fpu.fprs = kzalloc(sizeof(freg_t) * __NUM_FPRS,
+				       GFP_KERNEL|__GFP_REPEAT);
+	if (!dst->thread.fpu.fprs)
+		return -ENOMEM;
+
+	/*
+	 * Save the floating-point or vector register state of the current
+	 * task.  The state is not saved for early kernel threads, for example,
+	 * the init_task, which do not have an allocated save area.
+	 * The CIF_FPU flag is set in any case to lazy clear or restore a saved
+	 * state when switching to a different task or returning to user space.
+	 */
+	save_fpu_regs();
+	dst->thread.fpu.fpc = current->thread.fpu.fpc;
+	if (is_vx_task(current))
+		convert_vx_to_fp(dst->thread.fpu.fprs,
+				 current->thread.fpu.vxrs);
+	else
+		memcpy(dst->thread.fpu.fprs, current->thread.fpu.fprs,
+		       sizeof(freg_t) * __NUM_FPRS);
+	return 0;
 }
 
 int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
@@ -142,11 +172,6 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
 	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];
@@ -162,7 +187,7 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
 
 asmlinkage void execve_tail(void)
 {
-	current->thread.fp_regs.fpc = 0;
+	current->thread.fpu.fpc = 0;
 	asm volatile("sfpc %0" : : "d" (0));
 }
 
@@ -171,8 +196,15 @@ asmlinkage void execve_tail(void)
  */
 int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs)
 {
-	save_fp_ctl(&fpregs->fpc);
-	save_fp_regs(fpregs->fprs);
+	save_fpu_regs();
+	fpregs->fpc = current->thread.fpu.fpc;
+	fpregs->pad = 0;
+	if (is_vx_task(current))
+		convert_vx_to_fp((freg_t *)&fpregs->fprs,
+				 current->thread.fpu.vxrs);
+	else
+		memcpy(&fpregs->fprs, current->thread.fpu.fprs,
+		       sizeof(fpregs->fprs));
 	return 1;
 }
 EXPORT_SYMBOL(dump_fpu);
diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c
index dc488e13b..e6e077ae3 100644
--- a/arch/s390/kernel/processor.c
+++ b/arch/s390/kernel/processor.c
@@ -41,6 +41,15 @@ void cpu_init(void)
 }
 
 /*
+ * cpu_have_feature - Test CPU features on module initialization
+ */
+int cpu_have_feature(unsigned int num)
+{
+	return elf_hwcap & (1UL << num);
+}
+EXPORT_SYMBOL(cpu_have_feature);
+
+/*
  * show_cpuinfo - Get information on one CPU for use by procfs.
  */
 static int show_cpuinfo(struct seq_file *m, void *v)
diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index d363c9c32..d319f36f7 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -45,39 +45,27 @@ void update_cr_regs(struct task_struct *task)
 	struct per_regs old, new;
 
 	/* Take care of the enable/disable of transactional execution. */
-	if (MACHINE_HAS_TE || MACHINE_HAS_VX) {
+	if (MACHINE_HAS_TE) {
 		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);
-		}
+		/* Set or clear transaction execution TXC bit 8. */
+		cr_new = cr | (1UL << 55);
+		if (task->thread.per_flags & PER_FLAG_NO_TE)
+			cr_new &= ~(1UL << 55);
 		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);
+		/* Set or clear transaction execution TDC bits 62 and 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;
@@ -242,21 +230,21 @@ static unsigned long __peek_user(struct task_struct *child, addr_t addr)
 		/*
 		 * floating point control reg. is in the thread structure
 		 */
-		tmp = child->thread.fp_regs.fpc;
+		tmp = child->thread.fpu.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
+		 * floating point regs. are either in child->thread.fpu
+		 * or the child->thread.fpu.vxrs array
 		 */
 		offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
-		if (child->thread.vxrs)
+		if (is_vx_task(child))
 			tmp = *(addr_t *)
-			       ((addr_t) child->thread.vxrs + 2*offset);
+			       ((addr_t) child->thread.fpu.vxrs + 2*offset);
 		else
 			tmp = *(addr_t *)
-			       ((addr_t) &child->thread.fp_regs.fprs + offset);
+			       ((addr_t) child->thread.fpu.fprs + offset);
 
 	} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
 		/*
@@ -387,20 +375,20 @@ static int __poke_user(struct task_struct *child, addr_t addr, addr_t data)
 		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);
+		child->thread.fpu.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
+		 * floating point regs. are either in child->thread.fpu
+		 * or the child->thread.fpu.vxrs array
 		 */
 		offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
-		if (child->thread.vxrs)
+		if (is_vx_task(child))
 			*(addr_t *)((addr_t)
-				child->thread.vxrs + 2*offset) = data;
+				child->thread.fpu.vxrs + 2*offset) = data;
 		else
 			*(addr_t *)((addr_t)
-				&child->thread.fp_regs.fprs + offset) = data;
+				child->thread.fpu.fprs + offset) = data;
 
 	} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
 		/*
@@ -621,20 +609,20 @@ static u32 __peek_user_compat(struct task_struct *child, addr_t addr)
 		/*
 		 * floating point control reg. is in the thread structure
 		 */
-		tmp = child->thread.fp_regs.fpc;
+		tmp = child->thread.fpu.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
+		 * floating point regs. are either in child->thread.fpu
+		 * or the child->thread.fpu.vxrs array
 		 */
 		offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
-		if (child->thread.vxrs)
+		if (is_vx_task(child))
 			tmp = *(__u32 *)
-			       ((addr_t) child->thread.vxrs + 2*offset);
+			       ((addr_t) child->thread.fpu.vxrs + 2*offset);
 		else
 			tmp = *(__u32 *)
-			       ((addr_t) &child->thread.fp_regs.fprs + offset);
+			       ((addr_t) child->thread.fpu.fprs + offset);
 
 	} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
 		/*
@@ -746,20 +734,20 @@ static int __poke_user_compat(struct task_struct *child,
 		 */
 		if (test_fp_ctl(tmp))
 			return -EINVAL;
-		child->thread.fp_regs.fpc = data;
+		child->thread.fpu.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
+		 * floating point regs. are either in child->thread.fpu
+		 * or the child->thread.fpu.vxrs array
 		 */
 		offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
-		if (child->thread.vxrs)
+		if (is_vx_task(child))
 			*(__u32 *)((addr_t)
-				child->thread.vxrs + 2*offset) = tmp;
+				child->thread.fpu.vxrs + 2*offset) = tmp;
 		else
 			*(__u32 *)((addr_t)
-				&child->thread.fp_regs.fprs + offset) = tmp;
+				child->thread.fpu.fprs + offset) = tmp;
 
 	} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
 		/*
@@ -952,18 +940,16 @@ 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;
+	_s390_fp_regs fp_regs;
+
+	if (target == current)
+		save_fpu_regs();
+
+	fp_regs.fpc = target->thread.fpu.fpc;
+	fpregs_store(&fp_regs, &target->thread.fpu);
 
-		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);
+				   &fp_regs, 0, -1);
 }
 
 static int s390_fpregs_set(struct task_struct *target,
@@ -972,41 +958,33 @@ static int s390_fpregs_set(struct task_struct *target,
 			   const void __user *ubuf)
 {
 	int rc = 0;
+	freg_t fprs[__NUM_FPRS];
 
-	if (target == current) {
-		save_fp_ctl(&target->thread.fp_regs.fpc);
-		save_fp_regs(target->thread.fp_regs.fprs);
-	}
+	if (target == current)
+		save_fpu_regs();
 
 	/* 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 };
+		u32 ufpc[2] = { target->thread.fpu.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];
+		target->thread.fpu.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];
-		}
-	}
+					fprs, offsetof(s390_fp_regs, fprs), -1);
+	if (rc)
+		return rc;
+
+	if (is_vx_task(target))
+		convert_fp_to_vx(target->thread.fpu.vxrs, fprs);
+	else
+		memcpy(target->thread.fpu.fprs, &fprs, sizeof(fprs));
 
 	return rc;
 }
@@ -1069,11 +1047,11 @@ static int s390_vxrs_low_get(struct task_struct *target,
 
 	if (!MACHINE_HAS_VX)
 		return -ENODEV;
-	if (target->thread.vxrs) {
+	if (is_vx_task(target)) {
 		if (target == current)
-			save_vx_regs(target->thread.vxrs);
+			save_fpu_regs();
 		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			vxrs[i] = *((__u64 *)(target->thread.vxrs + i) + 1);
+			vxrs[i] = *((__u64 *)(target->thread.fpu.vxrs + i) + 1);
 	} else
 		memset(vxrs, 0, sizeof(vxrs));
 	return user_regset_copyout(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1);
@@ -1089,20 +1067,17 @@ static int s390_vxrs_low_set(struct task_struct *target,
 
 	if (!MACHINE_HAS_VX)
 		return -ENODEV;
-	if (!target->thread.vxrs) {
+	if (!is_vx_task(target)) {
 		rc = alloc_vector_registers(target);
 		if (rc)
 			return rc;
 	} else if (target == current)
-		save_vx_regs(target->thread.vxrs);
+		save_fpu_regs();
 
 	rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1);
-	if (rc == 0) {
+	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);
-	}
+			*((__u64 *)(target->thread.fpu.vxrs + i) + 1) = vxrs[i];
 
 	return rc;
 }
@@ -1116,10 +1091,10 @@ static int s390_vxrs_high_get(struct task_struct *target,
 
 	if (!MACHINE_HAS_VX)
 		return -ENODEV;
-	if (target->thread.vxrs) {
+	if (is_vx_task(target)) {
 		if (target == current)
-			save_vx_regs(target->thread.vxrs);
-		memcpy(vxrs, target->thread.vxrs + __NUM_VXRS_LOW,
+			save_fpu_regs();
+		memcpy(vxrs, target->thread.fpu.vxrs + __NUM_VXRS_LOW,
 		       sizeof(vxrs));
 	} else
 		memset(vxrs, 0, sizeof(vxrs));
@@ -1135,18 +1110,15 @@ static int s390_vxrs_high_set(struct task_struct *target,
 
 	if (!MACHINE_HAS_VX)
 		return -ENODEV;
-	if (!target->thread.vxrs) {
+	if (!is_vx_task(target)) {
 		rc = alloc_vector_registers(target);
 		if (rc)
 			return rc;
 	} else if (target == current)
-		save_vx_regs(target->thread.vxrs);
+		save_fpu_regs();
 
 	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);
-
+				target->thread.fpu.vxrs + __NUM_VXRS_LOW, 0, -1);
 	return rc;
 }
 
diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c
index 9f6046793..5090d3dad 100644
--- a/arch/s390/kernel/s390_ksyms.c
+++ b/arch/s390/kernel/s390_ksyms.c
@@ -1,5 +1,6 @@
 #include <linux/module.h>
 #include <linux/kvm_host.h>
+#include <asm/fpu-internal.h>
 #include <asm/ftrace.h>
 
 #ifdef CONFIG_FUNCTION_TRACER
@@ -8,6 +9,8 @@ EXPORT_SYMBOL(_mcount);
 #if IS_ENABLED(CONFIG_KVM)
 EXPORT_SYMBOL(sie64a);
 EXPORT_SYMBOL(sie_exit);
+EXPORT_SYMBOL(save_fpu_regs);
+EXPORT_SYMBOL(__ctl_set_vx);
 #endif
 EXPORT_SYMBOL(memcpy);
 EXPORT_SYMBOL(memset);
diff --git a/arch/s390/kernel/sclp.S b/arch/s390/kernel/sclp.S
deleted file mode 100644
index ada0c07fe..000000000
--- a/arch/s390/kernel/sclp.S
+++ /dev/null
@@ -1,355 +0,0 @@
-/*
- * 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
-	basr	%r13,0
-	lmh	%r0,%r15,.Lzeroes-.(%r13)	# clear 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
-.Lzeroes:
-	.fill	64,4,0
-
-.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/sclp.c b/arch/s390/kernel/sclp.c
new file mode 100644
index 000000000..fa0bdff1d
--- /dev/null
+++ b/arch/s390/kernel/sclp.c
@@ -0,0 +1,160 @@
+/*
+ *    Copyright IBM Corp. 2015
+ *    Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+#include <linux/kernel.h>
+#include <asm/ebcdic.h>
+#include <asm/irq.h>
+#include <asm/lowcore.h>
+#include <asm/processor.h>
+#include <asm/sclp.h>
+
+static char _sclp_work_area[4096] __aligned(PAGE_SIZE);
+
+static void _sclp_wait_int(void)
+{
+	unsigned long cr0, cr0_new, psw_mask, addr;
+	psw_t psw_ext_save, psw_wait;
+
+	__ctl_store(cr0, 0, 0);
+	cr0_new = cr0 | 0x200;
+	__ctl_load(cr0_new, 0, 0);
+
+	psw_ext_save = S390_lowcore.external_new_psw;
+	psw_mask = __extract_psw() & (PSW_MASK_EA | PSW_MASK_BA);
+	S390_lowcore.external_new_psw.mask = psw_mask;
+	psw_wait.mask = psw_mask | PSW_MASK_EXT | PSW_MASK_WAIT;
+	S390_lowcore.ext_int_code = 0;
+
+	do {
+		asm volatile(
+			"	larl	%[addr],0f\n"
+			"	stg	%[addr],%[psw_wait_addr]\n"
+			"	stg	%[addr],%[psw_ext_addr]\n"
+			"	lpswe	%[psw_wait]\n"
+			"0:\n"
+			: [addr] "=&d" (addr),
+			  [psw_wait_addr] "=Q" (psw_wait.addr),
+			  [psw_ext_addr] "=Q" (S390_lowcore.external_new_psw.addr)
+			: [psw_wait] "Q" (psw_wait)
+			: "cc", "memory");
+	} while (S390_lowcore.ext_int_code != EXT_IRQ_SERVICE_SIG);
+
+	__ctl_load(cr0, 0, 0);
+	S390_lowcore.external_new_psw = psw_ext_save;
+}
+
+static int _sclp_servc(unsigned int cmd, char *sccb)
+{
+	unsigned int cc;
+
+	do {
+		asm volatile(
+			"	.insn	rre,0xb2200000,%1,%2\n"
+			"	ipm	%0\n"
+			: "=d" (cc) : "d" (cmd), "a" (sccb)
+			: "cc", "memory");
+		cc >>= 28;
+		if (cc == 3)
+			return -EINVAL;
+		_sclp_wait_int();
+	} while (cc != 0);
+	return (*(unsigned short *)(sccb + 6) == 0x20) ? 0 : -EIO;
+}
+
+static int _sclp_setup(int disable)
+{
+	static unsigned char init_sccb[] = {
+		0x00, 0x1c,
+		0x00, 0x00, 0x00, 0x00,	0x00, 0x00, 0x00, 0x00,
+		0x00, 0x04,
+		0x80, 0x00, 0x00, 0x00,	0x40, 0x00, 0x00, 0x00,
+		0x00, 0x00, 0x00, 0x00,	0x00, 0x00, 0x00, 0x00
+	};
+	unsigned int *masks;
+	int rc;
+
+	memcpy(_sclp_work_area, init_sccb, 28);
+	masks = (unsigned int *)(_sclp_work_area + 12);
+	if (disable)
+		memset(masks, 0, 16);
+	/* SCLP write mask */
+	rc = _sclp_servc(0x00780005, _sclp_work_area);
+	if (rc)
+		return rc;
+	if ((masks[0] & masks[3]) != masks[0] ||
+	    (masks[1] & masks[2]) != masks[1])
+		return -EIO;
+	return 0;
+}
+
+static int _sclp_print(const char *str)
+{
+	static unsigned char write_head[] = {
+		/* sccb header */
+		0x00, 0x52,					/* 0 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00,		/* 2 */
+		/* evbuf */
+		0x00, 0x4a,					/* 8 */
+		0x02, 0x00, 0x00, 0x00,				/* 10 */
+		/* mdb */
+		0x00, 0x44,					/* 14 */
+		0x00, 0x01,					/* 16 */
+		0xd4, 0xc4, 0xc2, 0x40,				/* 18 */
+		0x00, 0x00, 0x00, 0x01,				/* 22 */
+		/* go */
+		0x00, 0x38,					/* 26 */
+		0x00, 0x01,					/* 28 */
+		0x00, 0x00, 0x00, 0x00,				/* 30 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 34 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 42 */
+		0x00, 0x00, 0x00, 0x00,				/* 50 */
+		0x00, 0x00,					/* 54 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 56 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 64 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 72 */
+		0x00, 0x00,					/* 80 */
+	};
+	static unsigned char write_mto[] = {
+		/* mto	*/
+		0x00, 0x0a,					/* 0 */
+		0x00, 0x04,					/* 2 */
+		0x10, 0x00,					/* 4 */
+		0x00, 0x00, 0x00, 0x00				/* 6 */
+	};
+	unsigned char *ptr, ch;
+	unsigned int count;
+
+	memcpy(_sclp_work_area, write_head, sizeof(write_head));
+	ptr = _sclp_work_area + sizeof(write_head);
+	do {
+		memcpy(ptr, write_mto, sizeof(write_mto));
+		for (count = sizeof(write_mto); (ch = *str++) != 0; count++) {
+			if (ch == 0x0a)
+				break;
+			ptr[count] = _ascebc[ch];
+		}
+		/* Update length fields in mto, mdb, evbuf and sccb */
+		*(unsigned short *) ptr = count;
+		*(unsigned short *)(_sclp_work_area + 14) += count;
+		*(unsigned short *)(_sclp_work_area + 8) += count;
+		*(unsigned short *)(_sclp_work_area + 0) += count;
+		ptr += count;
+	} while (ch != 0);
+
+	/* SCLP write data */
+	return _sclp_servc(0x00760005, _sclp_work_area);
+}
+
+int _sclp_print_early(const char *str)
+{
+	int rc;
+
+	rc = _sclp_setup(0);
+	if (rc)
+		return rc;
+	rc = _sclp_print(str);
+	if (rc)
+		return rc;
+	return _sclp_setup(1);
+}
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index b80512b9e..ce0cbd6ba 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -62,6 +62,7 @@
 #include <asm/os_info.h>
 #include <asm/sclp.h>
 #include <asm/sysinfo.h>
+#include <asm/numa.h>
 #include "entry.h"
 
 /*
@@ -76,7 +77,7 @@ EXPORT_SYMBOL(console_devno);
 unsigned int console_irq = -1;
 EXPORT_SYMBOL(console_irq);
 
-unsigned long elf_hwcap = 0;
+unsigned long elf_hwcap __read_mostly = 0;
 char elf_platform[ELF_PLATFORM_SIZE];
 
 int __initdata memory_end_set;
@@ -883,6 +884,7 @@ void __init setup_arch(char **cmdline_p)
 	setup_lowcore();
 	smp_fill_possible_mask();
         cpu_init();
+	numa_setup();
 
 	/*
 	 * Create kernel page tables and switch to virtual addressing.
diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c
index c551f22ce..9549af102 100644
--- a/arch/s390/kernel/signal.c
+++ b/arch/s390/kernel/signal.c
@@ -105,32 +105,13 @@ struct rt_sigframe
 static void store_sigregs(void)
 {
 	save_access_regs(current->thread.acrs);
-	save_fp_ctl(&current->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);
+	save_fpu_regs();
 }
 
 /* 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. */
@@ -146,8 +127,7 @@ static int save_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
 	memcpy(&user_sregs.regs.gprs, &regs->gprs, sizeof(sregs->regs.gprs));
 	memcpy(&user_sregs.regs.acrs, current->thread.acrs,
 	       sizeof(user_sregs.regs.acrs));
-	memcpy(&user_sregs.fpregs, &current->thread.fp_regs,
-	       sizeof(user_sregs.fpregs));
+	fpregs_store(&user_sregs.fpregs, &current->thread.fpu);
 	if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs)))
 		return -EFAULT;
 	return 0;
@@ -166,8 +146,8 @@ static int restore_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
 	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))
+	/* Test the floating-point-control word. */
+	if (test_fp_ctl(user_sregs.fpregs.fpc))
 		return -EINVAL;
 
 	/* Use regs->psw.mask instead of PSW_USER_BITS to preserve PER bit. */
@@ -185,8 +165,7 @@ static int restore_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
 	memcpy(&current->thread.acrs, &user_sregs.regs.acrs,
 	       sizeof(current->thread.acrs));
 
-	memcpy(&current->thread.fp_regs, &user_sregs.fpregs,
-	       sizeof(current->thread.fp_regs));
+	fpregs_load(&user_sregs.fpregs, &current->thread.fpu);
 
 	clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */
 	return 0;
@@ -200,13 +179,13 @@ static int save_sigregs_ext(struct pt_regs *regs,
 	int i;
 
 	/* Save vector registers to signal stack */
-	if (current->thread.vxrs) {
+	if (is_vx_task(current)) {
 		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			vxrs[i] = *((__u64 *)(current->thread.vxrs + i) + 1);
+			vxrs[i] = *((__u64 *)(current->thread.fpu.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,
+				   current->thread.fpu.vxrs + __NUM_VXRS_LOW,
 				   sizeof(sregs_ext->vxrs_high)))
 			return -EFAULT;
 	}
@@ -220,15 +199,15 @@ static int restore_sigregs_ext(struct pt_regs *regs,
 	int i;
 
 	/* Restore vector registers from signal stack */
-	if (current->thread.vxrs) {
+	if (is_vx_task(current)) {
 		if (__copy_from_user(vxrs, &sregs_ext->vxrs_low,
 				     sizeof(sregs_ext->vxrs_low)) ||
-		    __copy_from_user(current->thread.vxrs + __NUM_VXRS_LOW,
+		    __copy_from_user(current->thread.fpu.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];
+			*((__u64 *)(current->thread.fpu.vxrs + i) + 1) = vxrs[i];
 	}
 	return 0;
 }
@@ -243,6 +222,7 @@ SYSCALL_DEFINE0(sigreturn)
 	if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE))
 		goto badframe;
 	set_current_blocked(&set);
+	save_fpu_regs();
 	if (restore_sigregs(regs, &frame->sregs))
 		goto badframe;
 	if (restore_sigregs_ext(regs, &frame->sregs_ext))
@@ -266,6 +246,7 @@ SYSCALL_DEFINE0(rt_sigreturn)
 	set_current_blocked(&set);
 	if (restore_altstack(&frame->uc.uc_stack))
 		goto badframe;
+	save_fpu_regs();
 	if (restore_sigregs(regs, &frame->uc.uc_mcontext))
 		goto badframe;
 	if (restore_sigregs_ext(regs, &frame->uc.uc_mcontext_ext))
@@ -400,7 +381,7 @@ static int setup_rt_frame(struct ksignal *ksig, sigset_t *set,
 	uc_flags = 0;
 	if (MACHINE_HAS_VX) {
 		frame_size += sizeof(_sigregs_ext);
-		if (current->thread.vxrs)
+		if (is_vx_task(current))
 			uc_flags |= UC_VXRS;
 	}
 	frame = get_sigframe(&ksig->ka, regs, frame_size);
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 6f54c175f..c6355e6f3 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -532,8 +532,8 @@ EXPORT_SYMBOL(smp_ctl_clear_bit);
 
 #ifdef CONFIG_CRASH_DUMP
 
-static void __smp_store_cpu_state(struct save_area_ext *sa_ext, u16 address,
-				  int is_boot_cpu)
+static void __init __smp_store_cpu_state(struct save_area_ext *sa_ext,
+					 u16 address, int is_boot_cpu)
 {
 	void *lc = (void *)(unsigned long) store_prefix();
 	unsigned long vx_sa;
diff --git a/arch/s390/kernel/swsusp.S b/arch/s390/kernel/swsusp.S
index ca6294645..2d6b6e81f 100644
--- a/arch/s390/kernel/swsusp.S
+++ b/arch/s390/kernel/swsusp.S
@@ -30,6 +30,9 @@ ENTRY(swsusp_arch_suspend)
 	aghi	%r15,-STACK_FRAME_OVERHEAD
 	stg	%r1,__SF_BACKCHAIN(%r15)
 
+	/* Store FPU registers */
+	brasl	%r14,save_fpu_regs
+
 	/* Deactivate DAT */
 	stnsm	__SF_EMPTY(%r15),0xfb
 
@@ -47,23 +50,6 @@ ENTRY(swsusp_arch_suspend)
 
 	/* 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 */
@@ -249,24 +235,6 @@ restore_registers:
 	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)
 
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 1acad0268..8c56929c8 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -9,12 +9,12 @@
 #define NI_SYSCALL SYSCALL(sys_ni_syscall,sys_ni_syscall)
 
 NI_SYSCALL						/* 0 */
-SYSCALL(sys_exit,compat_sys_exit)
+SYSCALL(sys_exit,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_close,sys_close)
 SYSCALL(sys_restart_syscall,sys_restart_syscall)
 SYSCALL(sys_creat,compat_sys_creat)
 SYSCALL(sys_link,compat_sys_link)
@@ -35,21 +35,21 @@ 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)
+SYSCALL(sys_alarm,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)
+SYSCALL(sys_nice,sys_nice)
 NI_SYSCALL						/* 35 old ftime syscall */
 SYSCALL(sys_sync,sys_sync)
-SYSCALL(sys_kill,compat_sys_kill)
+SYSCALL(sys_kill,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_dup,sys_dup)
 SYSCALL(sys_pipe,compat_sys_pipe)
 SYSCALL(sys_times,compat_sys_times)
 NI_SYSCALL						/* old prof syscall */
@@ -65,13 +65,13 @@ 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)
+SYSCALL(sys_setpgid,sys_setpgid)
 NI_SYSCALL						/* old ulimit syscall */
 NI_SYSCALL						/* old uname syscall */
-SYSCALL(sys_umask,compat_sys_umask)			/* 60 */
+SYSCALL(sys_umask,sys_umask)				/* 60 */
 SYSCALL(sys_chroot,compat_sys_chroot)
 SYSCALL(sys_ustat,compat_sys_ustat)
-SYSCALL(sys_dup2,compat_sys_dup2)
+SYSCALL(sys_dup2,sys_dup2)
 SYSCALL(sys_getppid,sys_getppid)
 SYSCALL(sys_getpgrp,sys_getpgrp)			/* 65 */
 SYSCALL(sys_setsid,sys_setsid)
@@ -102,10 +102,10 @@ 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_fchmod,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)
+SYSCALL(sys_getpriority,sys_getpriority)
+SYSCALL(sys_setpriority,sys_setpriority)
 NI_SYSCALL						/* old profil syscall */
 SYSCALL(sys_statfs,compat_sys_statfs)
 SYSCALL(sys_fstatfs,compat_sys_fstatfs)			/* 100 */
@@ -126,7 +126,7 @@ 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_fsync,sys_fsync)
 SYSCALL(sys_sigreturn,compat_sys_sigreturn)
 SYSCALL(sys_clone,compat_sys_clone)			/* 120 */
 SYSCALL(sys_setdomainname,compat_sys_setdomainname)
@@ -140,35 +140,35 @@ 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_getpgid,sys_getpgid)
+SYSCALL(sys_fchdir,sys_fchdir)
 SYSCALL(sys_bdflush,compat_sys_bdflush)
 SYSCALL(sys_sysfs,compat_sys_sysfs)			/* 135 */
-SYSCALL(sys_s390_personality,compat_sys_s390_personality)
+SYSCALL(sys_s390_personality,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_flock,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_getsid,sys_getsid)
+SYSCALL(sys_fdatasync,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_mlockall,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_getscheduler,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_get_priority_max,sys_sched_get_priority_max)
+SYSCALL(sys_sched_get_priority_min,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)
@@ -211,20 +211,20 @@ 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_setreuid,sys_setreuid)
+SYSCALL(sys_setregid,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_fchown,sys_fchown)
+SYSCALL(sys_setresuid,sys_setresuid)
 SYSCALL(sys_getresuid,compat_sys_getresuid)
-SYSCALL(sys_setresgid,compat_sys_setresgid)		/* 210 */
+SYSCALL(sys_setresgid,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_setuid,sys_setuid)
+SYSCALL(sys_setgid,sys_setgid)
+SYSCALL(sys_setfsuid,sys_setfsuid)			/* 215 */
+SYSCALL(sys_setfsgid,sys_setfsgid)
 SYSCALL(sys_pivot_root,compat_sys_pivot_root)
 SYSCALL(sys_mincore,compat_sys_mincore)
 SYSCALL(sys_madvise,compat_sys_madvise)
@@ -245,19 +245,19 @@ 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_tkill,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)
+SYSCALL(sys_tgkill,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_exit_group,sys_exit_group)
+SYSCALL(sys_epoll_create,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)
@@ -265,8 +265,8 @@ 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_timer_getoverrun,sys_timer_getoverrun)
+SYSCALL(sys_timer_delete,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)
@@ -276,9 +276,9 @@ 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_mbind,compat_sys_mbind)
+SYSCALL(sys_get_mempolicy,compat_sys_get_mempolicy)
+SYSCALL(sys_set_mempolicy,compat_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)
@@ -290,12 +290,12 @@ 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_ioprio_set,sys_ioprio_set)
+SYSCALL(sys_ioprio_get,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_inotify_rm_watch,sys_inotify_rm_watch)
+SYSCALL(sys_migrate_pages,compat_sys_migrate_pages)
 SYSCALL(sys_openat,compat_sys_openat)
 SYSCALL(sys_mkdirat,compat_sys_mkdirat)
 SYSCALL(sys_mknodat,compat_sys_mknodat)			/* 290 */
@@ -318,7 +318,7 @@ 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_move_pages,compat_sys_move_pages)
 SYSCALL(sys_getcpu,compat_sys_getcpu)
 SYSCALL(sys_epoll_pwait,compat_sys_epoll_pwait)
 SYSCALL(sys_utimes,compat_sys_utimes)
@@ -326,31 +326,31 @@ 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_eventfd,sys_eventfd)
+SYSCALL(sys_timerfd_create,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_eventfd2,sys_eventfd2)
+SYSCALL(sys_inotify_init1,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_dup3,sys_dup3)
+SYSCALL(sys_epoll_create1,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_init,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_syncfs,sys_syncfs)
+SYSCALL(sys_setns,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_s390_runtime_instr,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 */
@@ -363,3 +363,22 @@ 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)
+SYSCALL(sys_userfaultfd,sys_userfaultfd)		/* 355 */
+SYSCALL(sys_membarrier,sys_membarrier)
+SYSCALL(sys_recvmmsg,compat_sys_recvmmsg)
+SYSCALL(sys_sendmmsg,compat_sys_sendmmsg)
+SYSCALL(sys_socket,sys_socket)
+SYSCALL(sys_socketpair,compat_sys_socketpair)		/* 360 */
+SYSCALL(sys_bind,sys_bind)
+SYSCALL(sys_connect,sys_connect)
+SYSCALL(sys_listen,sys_listen)
+SYSCALL(sys_accept4,sys_accept4)
+SYSCALL(sys_getsockopt,compat_sys_getsockopt)		/* 365 */
+SYSCALL(sys_setsockopt,compat_sys_setsockopt)
+SYSCALL(sys_getsockname,compat_sys_getsockname)
+SYSCALL(sys_getpeername,compat_sys_getpeername)
+SYSCALL(sys_sendto,compat_sys_sendto)
+SYSCALL(sys_sendmsg,compat_sys_sendmsg)			/* 370 */
+SYSCALL(sys_recvfrom,compat_sys_recvfrom)
+SYSCALL(sys_recvmsg,compat_sys_recvmsg)
+SYSCALL(sys_shutdown,sys_shutdown)
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 9e733d965..017c3a9bf 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -58,6 +58,9 @@ EXPORT_SYMBOL_GPL(sched_clock_base_cc);
 
 static DEFINE_PER_CPU(struct clock_event_device, comparators);
 
+ATOMIC_NOTIFIER_HEAD(s390_epoch_delta_notifier);
+EXPORT_SYMBOL(s390_epoch_delta_notifier);
+
 /*
  * Scheduler clock - returns current time in nanosec units.
  */
@@ -117,11 +120,6 @@ static int s390_next_event(unsigned long delta,
 	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.
@@ -145,7 +143,6 @@ void init_cpu_timer(void)
 	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);
 
@@ -381,7 +378,7 @@ static void disable_sync_clock(void *dummy)
 	 * 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_andnot(0x80000000, sw_ptr);
 	atomic_inc(sw_ptr);
 }
 
@@ -392,7 +389,7 @@ static void disable_sync_clock(void *dummy)
 static void enable_sync_clock(void)
 {
 	atomic_t *sw_ptr = this_cpu_ptr(&clock_sync_word);
-	atomic_set_mask(0x80000000, sw_ptr);
+	atomic_or(0x80000000, sw_ptr);
 }
 
 /*
@@ -752,7 +749,7 @@ static void clock_sync_cpu(struct clock_sync_data *sync)
 static int etr_sync_clock(void *data)
 {
 	static int first;
-	unsigned long long clock, old_clock, delay, delta;
+	unsigned long long clock, old_clock, clock_delta, delay, delta;
 	struct clock_sync_data *etr_sync;
 	struct etr_aib *sync_port, *aib;
 	int port;
@@ -789,6 +786,9 @@ static int etr_sync_clock(void *data)
 		delay = (unsigned long long)
 			(aib->edf2.etv - sync_port->edf2.etv) << 32;
 		delta = adjust_time(old_clock, clock, delay);
+		clock_delta = clock - old_clock;
+		atomic_notifier_call_chain(&s390_epoch_delta_notifier, 0,
+					   &clock_delta);
 		etr_sync->fixup_cc = delta;
 		fixup_clock_comparator(delta);
 		/* Verify that the clock is properly set. */
@@ -1526,7 +1526,7 @@ void stp_island_check(void)
 static int stp_sync_clock(void *data)
 {
 	static int first;
-	unsigned long long old_clock, delta;
+	unsigned long long old_clock, delta, new_clock, clock_delta;
 	struct clock_sync_data *stp_sync;
 	int rc;
 
@@ -1551,7 +1551,11 @@ static int stp_sync_clock(void *data)
 		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);
+			new_clock = get_tod_clock();
+			delta = adjust_time(old_clock, new_clock, 0);
+			clock_delta = new_clock - old_clock;
+			atomic_notifier_call_chain(&s390_epoch_delta_notifier,
+						   0, &clock_delta);
 			fixup_clock_comparator(delta);
 			rc = chsc_sstpi(stp_page, &stp_info,
 					sizeof(struct stp_sstpi));
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index 5728c5bd4..bf05e7fc3 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -18,7 +18,10 @@
 #include <linux/cpu.h>
 #include <linux/smp.h>
 #include <linux/mm.h>
+#include <linux/nodemask.h>
+#include <linux/node.h>
 #include <asm/sysinfo.h>
+#include <asm/numa.h>
 
 #define PTF_HORIZONTAL	(0UL)
 #define PTF_VERTICAL	(1UL)
@@ -37,8 +40,10 @@ 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);
+/*
+ * Socket/Book linked lists and per_cpu(cpu_topology) updates are
+ * protected by "sched_domains_mutex".
+ */
 static struct mask_info socket_info;
 static struct mask_info book_info;
 
@@ -188,7 +193,6 @@ 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) {
@@ -199,7 +203,6 @@ static void tl_to_masks(struct sysinfo_15_1_x *info)
 	default:
 		__tl_to_masks_generic(info);
 	}
-	spin_unlock_irq(&topology_lock);
 }
 
 static void topology_update_polarization_simple(void)
@@ -244,10 +247,8 @@ int topology_set_cpu_management(int fc)
 
 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);
@@ -259,7 +260,7 @@ static void update_cpu_masks(void)
 			per_cpu(cpu_topology, cpu).book_id = cpu;
 		}
 	}
-	spin_unlock_irqrestore(&topology_lock, flags);
+	numa_update_cpu_topology();
 }
 
 void store_topology(struct sysinfo_15_1_x *info)
@@ -274,21 +275,21 @@ int arch_update_cpu_topology(void)
 {
 	struct sysinfo_15_1_x *info = tl_info;
 	struct device *dev;
-	int cpu;
+	int cpu, rc = 0;
 
-	if (!MACHINE_HAS_TOPOLOGY) {
-		update_cpu_masks();
-		topology_update_polarization_simple();
-		return 0;
+	if (MACHINE_HAS_TOPOLOGY) {
+		rc = 1;
+		store_topology(info);
+		tl_to_masks(info);
 	}
-	store_topology(info);
-	tl_to_masks(info);
 	update_cpu_masks();
+	if (!MACHINE_HAS_TOPOLOGY)
+		topology_update_polarization_simple();
 	for_each_online_cpu(cpu) {
 		dev = get_cpu_device(cpu);
 		kobject_uevent(&dev->kobj, KOBJ_CHANGE);
 	}
-	return 1;
+	return rc;
 }
 
 static void topology_work_fn(struct work_struct *work)
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index 7bea81d8a..9861613fb 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -19,7 +19,7 @@
 #include <linux/sched.h>
 #include <linux/mm.h>
 #include <linux/slab.h>
-#include <asm/switch_to.h>
+#include <asm/fpu-internal.h>
 #include "entry.h"
 
 int show_unhandled_signals = 1;
@@ -151,7 +151,7 @@ DO_ERROR_INFO(special_op_exception, SIGILL, ILL_ILLOPN,
 DO_ERROR_INFO(transaction_exception, SIGILL, ILL_ILLOPN,
 	      "transaction constraint exception")
 
-static inline void do_fp_trap(struct pt_regs *regs, int fpc)
+static inline void do_fp_trap(struct pt_regs *regs, __u32 fpc)
 {
 	int si_code = 0;
 	/* FPC[2] is Data Exception Code */
@@ -227,7 +227,7 @@ DO_ERROR_INFO(specification_exception, SIGILL, ILL_ILLOPN,
 int alloc_vector_registers(struct task_struct *tsk)
 {
 	__vector128 *vxrs;
-	int i;
+	freg_t *fprs;
 
 	/* Allocate vector register save area. */
 	vxrs = kzalloc(sizeof(__vector128) * __NUM_VXRS,
@@ -236,15 +236,13 @@ int alloc_vector_registers(struct task_struct *tsk)
 		return -ENOMEM;
 	preempt_disable();
 	if (tsk == current)
-		save_fp_regs(tsk->thread.fp_regs.fprs);
+		save_fpu_regs();
 	/* 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);
-	}
+	convert_fp_to_vx(vxrs, tsk->thread.fpu.fprs);
+	fprs = tsk->thread.fpu.fprs;
+	tsk->thread.fpu.vxrs = vxrs;
+	tsk->thread.fpu.flags |= FPU_USE_VX;
+	kfree(fprs);
 	preempt_enable();
 	return 0;
 }
@@ -259,8 +257,8 @@ void vector_exception(struct pt_regs *regs)
 	}
 
 	/* get vector interrupt code from fpc */
-	asm volatile("stfpc %0" : "=Q" (current->thread.fp_regs.fpc));
-	vic = (current->thread.fp_regs.fpc & 0xf00) >> 8;
+	save_fpu_regs();
+	vic = (current->thread.fpu.fpc & 0xf00) >> 8;
 	switch (vic) {
 	case 1: /* invalid vector operation */
 		si_code = FPE_FLTINV;
@@ -297,22 +295,22 @@ void data_exception(struct pt_regs *regs)
 
 	location = get_trap_ip(regs);
 
-	asm volatile("stfpc %0" : "=Q" (current->thread.fp_regs.fpc));
+	save_fpu_regs();
 	/* Check for vector register enablement */
-	if (MACHINE_HAS_VX && !current->thread.vxrs &&
-	    (current->thread.fp_regs.fpc & FPC_DXC_MASK) == 0xfe00) {
+	if (MACHINE_HAS_VX && !is_vx_task(current) &&
+	    (current->thread.fpu.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)
+	if (current->thread.fpu.fpc & FPC_DXC_MASK)
 		signal = SIGFPE;
 	else
 		signal = SIGILL;
 	if (signal == SIGFPE)
-		do_fp_trap(regs, current->thread.fp_regs.fpc);
+		do_fp_trap(regs, current->thread.fpu.fpc);
 	else if (signal)
 		do_trap(regs, signal, ILL_ILLOPN, "data exception");
 }
diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile
index 8ad2b34ad..ee8a18e50 100644
--- a/arch/s390/kernel/vdso32/Makefile
+++ b/arch/s390/kernel/vdso32/Makefile
@@ -13,7 +13,7 @@ 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)
+			$(call cc-ldoption, -Wl$(comma)--hash-style=both)
 
 $(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_31)
 $(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_31)
diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile
index 2a8ddfd12..c4b03f9ed 100644
--- a/arch/s390/kernel/vdso64/Makefile
+++ b/arch/s390/kernel/vdso64/Makefile
@@ -13,7 +13,7 @@ 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)
+			$(call cc-ldoption, -Wl$(comma)--hash-style=both)
 
 $(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_64)
 $(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_64)
diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c
index e53d3595a..dafc44f51 100644
--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -25,9 +25,10 @@ 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]);
+DEFINE_PER_CPU(u64, mt_cycles[8]);
 static DEFINE_PER_CPU(u64, mt_scaling_mult) = { 1 };
 static DEFINE_PER_CPU(u64, mt_scaling_div) = { 1 };
+static DEFINE_PER_CPU(u64, mt_scaling_jiffies);
 
 static inline u64 get_vtimer(void)
 {
@@ -59,6 +60,34 @@ static inline int virt_timer_forward(u64 elapsed)
 	return elapsed >= atomic64_read(&virt_timer_current);
 }
 
+static void update_mt_scaling(void)
+{
+	u64 cycles_new[8], *cycles_old;
+	u64 delta, fac, mult, div;
+	int i;
+
+	stcctm5(smp_cpu_mtid + 1, cycles_new);
+	cycles_old = this_cpu_ptr(mt_cycles);
+	fac = 1;
+	mult = div = 0;
+	for (i = 0; i <= smp_cpu_mtid; i++) {
+		delta = cycles_new[i] - cycles_old[i];
+		div += delta;
+		mult *= i + 1;
+		mult += delta * fac;
+		fac *= i + 1;
+	}
+	div *= fac;
+	if (div > 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));
+	}
+	__this_cpu_write(mt_scaling_jiffies, jiffies_64);
+}
+
 /*
  * Update process times based on virtual cpu times stored by entry.S
  * to the lowcore fields user_timer, system_timer & steal_clock.
@@ -68,7 +97,6 @@ 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;
@@ -84,28 +112,10 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset)
 	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));
-			}
-		}
-	}
+	/* Update MT utilization calculation */
+	if (smp_cpu_mtid &&
+	    time_after64(jiffies_64, this_cpu_read(mt_scaling_jiffies)))
+		update_mt_scaling();
 
 	user = S390_lowcore.user_timer - ti->user_timer;
 	S390_lowcore.steal_timer -= user;
@@ -174,6 +184,11 @@ void vtime_account_irq_enter(struct task_struct *tsk)
 	S390_lowcore.last_update_timer = get_vtimer();
 	S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
 
+	/* Update MT utilization calculation */
+	if (smp_cpu_mtid &&
+	    time_after64(jiffies_64, this_cpu_read(mt_scaling_jiffies)))
+		update_mt_scaling();
+
 	system = S390_lowcore.system_timer - ti->system_timer;
 	S390_lowcore.steal_timer -= system;
 	ti->system_timer = S390_lowcore.system_timer;
@@ -376,4 +391,11 @@ void vtime_init(void)
 {
 	/* set initial cpu timer */
 	set_vtimer(VTIMER_MAX_SLICE);
+	/* Setup initial MT scaling values */
+	if (smp_cpu_mtid) {
+		__this_cpu_write(mt_scaling_jiffies, jiffies);
+		__this_cpu_write(mt_scaling_mult, 1);
+		__this_cpu_write(mt_scaling_div, 1);
+		stcctm5(smp_cpu_mtid + 1, this_cpu_ptr(mt_cycles));
+	}
 }
diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
index fc7ec9584..5fbfb88f8 100644
--- a/arch/s390/kvm/diag.c
+++ b/arch/s390/kvm/diag.c
@@ -27,13 +27,13 @@ static int diag_release_pages(struct kvm_vcpu *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;
+	vcpu->stat.diagnose_10++;
 
 	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
@@ -75,6 +75,9 @@ static int __diag_page_ref_service(struct kvm_vcpu *vcpu)
 	u16 rx = (vcpu->arch.sie_block->ipa & 0xf0) >> 4;
 	u16 ry = (vcpu->arch.sie_block->ipa & 0x0f);
 
+	VCPU_EVENT(vcpu, 3, "diag page reference parameter block at 0x%llx",
+		   vcpu->run->s.regs.gprs[rx]);
+	vcpu->stat.diagnose_258++;
 	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));
@@ -85,6 +88,9 @@ static int __diag_page_ref_service(struct kvm_vcpu *vcpu)
 
 	switch (parm.subcode) {
 	case 0: /* TOKEN */
+		VCPU_EVENT(vcpu, 3, "pageref token addr 0x%llx "
+			   "select mask 0x%llx compare mask 0x%llx",
+			   parm.token_addr, parm.select_mask, parm.compare_mask);
 		if (vcpu->arch.pfault_token != KVM_S390_PFAULT_TOKEN_INVALID) {
 			/*
 			 * If the pagefault handshake is already activated,
@@ -114,6 +120,7 @@ static int __diag_page_ref_service(struct kvm_vcpu *vcpu)
 		 * the cancel, therefore to reduce code complexity, we assume
 		 * all outstanding tokens are already pending.
 		 */
+		VCPU_EVENT(vcpu, 3, "pageref cancel addr 0x%llx", parm.token_addr);
 		if (parm.token_addr || parm.select_mask ||
 		    parm.compare_mask || parm.zarch)
 			return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
@@ -174,7 +181,8 @@ 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);
+	VCPU_EVENT(vcpu, 3, "diag ipl functions, subcode %lx", subcode);
+	vcpu->stat.diagnose_308++;
 	switch (subcode) {
 	case 3:
 		vcpu->run->s390_reset_flags = KVM_S390_RESET_CLEAR;
@@ -202,6 +210,7 @@ static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
 {
 	int ret;
 
+	vcpu->stat.diagnose_500++;
 	/* No virtio-ccw notification? Get out quickly. */
 	if (!vcpu->kvm->arch.css_support ||
 	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
diff --git a/arch/s390/kvm/guestdbg.c b/arch/s390/kvm/guestdbg.c
index e97b3455d..47518a324 100644
--- a/arch/s390/kvm/guestdbg.c
+++ b/arch/s390/kvm/guestdbg.c
@@ -473,10 +473,45 @@ static void filter_guest_per_event(struct kvm_vcpu *vcpu)
 		vcpu->arch.sie_block->iprcc &= ~PGM_PER;
 }
 
+#define pssec(vcpu) (vcpu->arch.sie_block->gcr[1] & _ASCE_SPACE_SWITCH)
+#define hssec(vcpu) (vcpu->arch.sie_block->gcr[13] & _ASCE_SPACE_SWITCH)
+#define old_ssec(vcpu) ((vcpu->arch.sie_block->tecmc >> 31) & 0x1)
+#define old_as_is_home(vcpu) !(vcpu->arch.sie_block->tecmc & 0xffff)
+
 void kvm_s390_handle_per_event(struct kvm_vcpu *vcpu)
 {
+	int new_as;
+
 	if (debug_exit_required(vcpu))
 		vcpu->guest_debug |= KVM_GUESTDBG_EXIT_PENDING;
 
 	filter_guest_per_event(vcpu);
+
+	/*
+	 * Only RP, SAC, SACF, PT, PTI, PR, PC instructions can trigger
+	 * a space-switch event. PER events enforce space-switch events
+	 * for these instructions. So if no PER event for the guest is left,
+	 * we might have to filter the space-switch element out, too.
+	 */
+	if (vcpu->arch.sie_block->iprcc == PGM_SPACE_SWITCH) {
+		vcpu->arch.sie_block->iprcc = 0;
+		new_as = psw_bits(vcpu->arch.sie_block->gpsw).as;
+
+		/*
+		 * If the AS changed from / to home, we had RP, SAC or SACF
+		 * instruction. Check primary and home space-switch-event
+		 * controls. (theoretically home -> home produced no event)
+		 */
+		if (((new_as == PSW_AS_HOME) ^ old_as_is_home(vcpu)) &&
+		     (pssec(vcpu) || hssec(vcpu)))
+			vcpu->arch.sie_block->iprcc = PGM_SPACE_SWITCH;
+
+		/*
+		 * PT, PTI, PR, PC instruction operate on primary AS only. Check
+		 * if the primary-space-switch-event control was or got set.
+		 */
+		if (new_as == PSW_AS_PRIMARY && !old_as_is_home(vcpu) &&
+		    (pssec(vcpu) || old_ssec(vcpu)))
+			vcpu->arch.sie_block->iprcc = PGM_SPACE_SWITCH;
+	}
 }
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index c98d89708..08de785c9 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -30,7 +30,6 @@
 #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
@@ -72,9 +71,13 @@ static int ckc_interrupts_enabled(struct kvm_vcpu *vcpu)
 
 static int ckc_irq_pending(struct kvm_vcpu *vcpu)
 {
+	preempt_disable();
 	if (!(vcpu->arch.sie_block->ckc <
-	      get_tod_clock_fast() + vcpu->arch.sie_block->epoch))
+	      get_tod_clock_fast() + vcpu->arch.sie_block->epoch)) {
+		preempt_enable();
 		return 0;
+	}
+	preempt_enable();
 	return ckc_interrupts_enabled(vcpu);
 }
 
@@ -170,20 +173,20 @@ static unsigned long deliverable_irqs(struct kvm_vcpu *vcpu)
 
 static void __set_cpu_idle(struct kvm_vcpu *vcpu)
 {
-	atomic_set_mask(CPUSTAT_WAIT, &vcpu->arch.sie_block->cpuflags);
+	atomic_or(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);
+	atomic_andnot(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);
+	atomic_andnot(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);
 
@@ -196,7 +199,7 @@ static void __reset_intercept_indicators(struct kvm_vcpu *vcpu)
 
 static void __set_cpuflag(struct kvm_vcpu *vcpu, u32 flag)
 {
-	atomic_set_mask(flag, &vcpu->arch.sie_block->cpuflags);
+	atomic_or(flag, &vcpu->arch.sie_block->cpuflags);
 }
 
 static void set_intercept_indicators_io(struct kvm_vcpu *vcpu)
@@ -311,8 +314,8 @@ static int __must_check __deliver_pfault_init(struct kvm_vcpu *vcpu)
 	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);
+	VCPU_EVENT(vcpu, 4, "deliver: pfault init token 0x%llx",
+		   ext.ext_params2);
 	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
 					 KVM_S390_INT_PFAULT_INIT,
 					 0, ext.ext_params2);
@@ -368,7 +371,7 @@ static int __must_check __deliver_machine_check(struct kvm_vcpu *vcpu)
 	spin_unlock(&fi->lock);
 
 	if (deliver) {
-		VCPU_EVENT(vcpu, 4, "interrupt: machine check mcic=%llx",
+		VCPU_EVENT(vcpu, 3, "deliver: machine check mcic 0x%llx",
 			   mchk.mcic);
 		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
 						 KVM_S390_MCHK,
@@ -403,7 +406,7 @@ 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_EVENT(vcpu, 3, "%s", "deliver: cpu restart");
 	vcpu->stat.deliver_restart_signal++;
 	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_RESTART, 0, 0);
 
@@ -427,7 +430,6 @@ static int __must_check __deliver_set_prefix(struct kvm_vcpu *vcpu)
 	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,
@@ -450,7 +452,7 @@ static int __must_check __deliver_emergency_signal(struct kvm_vcpu *vcpu)
 		clear_bit(IRQ_PEND_EXT_EMERGENCY, &li->pending_irqs);
 	spin_unlock(&li->lock);
 
-	VCPU_EVENT(vcpu, 4, "%s", "interrupt: sigp emerg");
+	VCPU_EVENT(vcpu, 4, "%s", "deliver: sigp emerg");
 	vcpu->stat.deliver_emergency_signal++;
 	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_INT_EMERGENCY,
 					 cpu_addr, 0);
@@ -477,7 +479,7 @@ static int __must_check __deliver_external_call(struct kvm_vcpu *vcpu)
 	clear_bit(IRQ_PEND_EXT_EXTERNAL, &li->pending_irqs);
 	spin_unlock(&li->lock);
 
-	VCPU_EVENT(vcpu, 4, "%s", "interrupt: sigp ext call");
+	VCPU_EVENT(vcpu, 4, "%s", "deliver: sigp ext call");
 	vcpu->stat.deliver_external_call++;
 	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
 					 KVM_S390_INT_EXTERNAL_CALL,
@@ -506,7 +508,7 @@ static int __must_check __deliver_prog(struct kvm_vcpu *vcpu)
 	memset(&li->irq.pgm, 0, sizeof(pgm_info));
 	spin_unlock(&li->lock);
 
-	VCPU_EVENT(vcpu, 4, "interrupt: pgm check code:%x, ilc:%x",
+	VCPU_EVENT(vcpu, 3, "deliver: program irq code 0x%x, ilc:%d",
 		   pgm_info.code, ilc);
 	vcpu->stat.deliver_program_int++;
 	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_PROGRAM_INT,
@@ -622,7 +624,7 @@ static int __must_check __deliver_service(struct kvm_vcpu *vcpu)
 	clear_bit(IRQ_PEND_EXT_SERVICE, &fi->pending_irqs);
 	spin_unlock(&fi->lock);
 
-	VCPU_EVENT(vcpu, 4, "interrupt: sclp parm:%x",
+	VCPU_EVENT(vcpu, 4, "deliver: sclp parameter 0x%x",
 		   ext.ext_params);
 	vcpu->stat.deliver_service_signal++;
 	trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, KVM_S390_INT_SERVICE,
@@ -651,9 +653,6 @@ static int __must_check __deliver_pfault_done(struct kvm_vcpu *vcpu)
 					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;
 	}
@@ -662,6 +661,12 @@ static int __must_check __deliver_pfault_done(struct kvm_vcpu *vcpu)
 	spin_unlock(&fi->lock);
 
 	if (inti) {
+		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
+						 KVM_S390_INT_PFAULT_DONE, 0,
+						 inti->ext.ext_params2);
+		VCPU_EVENT(vcpu, 4, "deliver: pfault done token 0x%llx",
+			   inti->ext.ext_params2);
+
 		rc  = put_guest_lc(vcpu, EXT_IRQ_CP_SERVICE,
 				(u16 *)__LC_EXT_INT_CODE);
 		rc |= put_guest_lc(vcpu, PFAULT_DONE,
@@ -691,7 +696,7 @@ static int __must_check __deliver_virtio(struct kvm_vcpu *vcpu)
 					list);
 	if (inti) {
 		VCPU_EVENT(vcpu, 4,
-			   "interrupt: virtio parm:%x,parm64:%llx",
+			   "deliver: virtio parm: 0x%x,parm64: 0x%llx",
 			   inti->ext.ext_params, inti->ext.ext_params2);
 		vcpu->stat.deliver_virtio_interrupt++;
 		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
@@ -741,7 +746,7 @@ static int __must_check __deliver_io(struct kvm_vcpu *vcpu,
 					struct kvm_s390_interrupt_info,
 					list);
 	if (inti) {
-		VCPU_EVENT(vcpu, 4, "interrupt: I/O %llx", inti->type);
+		VCPU_EVENT(vcpu, 4, "deliver: I/O 0x%llx", inti->type);
 		vcpu->stat.deliver_io_int++;
 		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
 				inti->type,
@@ -855,7 +860,9 @@ int kvm_s390_handle_wait(struct kvm_vcpu *vcpu)
 		goto no_timer;
 	}
 
+	preempt_disable();
 	now = get_tod_clock_fast() + vcpu->arch.sie_block->epoch;
+	preempt_enable();
 	sltime = tod_to_ns(vcpu->arch.sie_block->ckc - now);
 
 	/* underflow */
@@ -864,7 +871,7 @@ int kvm_s390_handle_wait(struct kvm_vcpu *vcpu)
 
 	__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);
+	VCPU_EVENT(vcpu, 4, "enabled wait via clock comparator: %llu ns", sltime);
 no_timer:
 	srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
 	kvm_vcpu_block(vcpu);
@@ -894,7 +901,9 @@ enum hrtimer_restart kvm_s390_idle_wakeup(struct hrtimer *timer)
 	u64 now, sltime;
 
 	vcpu = container_of(timer, struct kvm_vcpu, arch.ckc_timer);
+	preempt_disable();
 	now = get_tod_clock_fast() + vcpu->arch.sie_block->epoch;
+	preempt_enable();
 	sltime = tod_to_ns(vcpu->arch.sie_block->ckc - now);
 
 	/*
@@ -919,7 +928,7 @@ void kvm_s390_clear_local_irqs(struct kvm_vcpu *vcpu)
 	spin_unlock(&li->lock);
 
 	/* clear pending external calls set by sigp interpretation facility */
-	atomic_clear_mask(CPUSTAT_ECALL_PEND, li->cpuflags);
+	atomic_andnot(CPUSTAT_ECALL_PEND, li->cpuflags);
 	vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sigp_ctrl = 0;
 }
 
@@ -968,6 +977,10 @@ static int __inject_prog(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
 {
 	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
 
+	VCPU_EVENT(vcpu, 3, "inject: program irq code 0x%x", irq->u.pgm.code);
+	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_PROGRAM_INT,
+				   irq->u.pgm.code, 0);
+
 	li->irq.pgm = irq->u.pgm;
 	set_bit(IRQ_PEND_PROG, &li->pending_irqs);
 	return 0;
@@ -978,9 +991,6 @@ 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);
@@ -996,10 +1006,6 @@ int kvm_s390_inject_prog_irq(struct kvm_vcpu *vcpu,
 	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);
@@ -1012,15 +1018,15 @@ 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);
+	VCPU_EVENT(vcpu, 4, "inject: pfault init parameter block at 0x%llx",
+		   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);
+				   irq->u.ext.ext_params2);
 
 	li->irq.ext = irq->u.ext;
 	set_bit(IRQ_PEND_PFAULT_INIT, &li->pending_irqs);
-	atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
+	atomic_or(CPUSTAT_EXT_INT, li->cpuflags);
 	return 0;
 }
 
@@ -1035,7 +1041,7 @@ static int __inject_extcall_sigpif(struct kvm_vcpu *vcpu, uint16_t src_id)
 		/* another external call is pending */
 		return -EBUSY;
 	}
-	atomic_set_mask(CPUSTAT_ECALL_PEND, &vcpu->arch.sie_block->cpuflags);
+	atomic_or(CPUSTAT_ECALL_PEND, &vcpu->arch.sie_block->cpuflags);
 	return 0;
 }
 
@@ -1045,14 +1051,13 @@ static int __inject_extcall(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
 	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",
+	VCPU_EVENT(vcpu, 4, "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);
+				   src_id, 0);
 
 	/* sending vcpu invalid */
-	if (src_id >= KVM_MAX_VCPUS ||
-	    kvm_get_vcpu(vcpu->kvm, src_id) == NULL)
+	if (kvm_get_vcpu_by_id(vcpu->kvm, src_id) == NULL)
 		return -EINVAL;
 
 	if (sclp.has_sigpif)
@@ -1061,7 +1066,7 @@ static int __inject_extcall(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
 	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);
+	atomic_or(CPUSTAT_EXT_INT, li->cpuflags);
 	return 0;
 }
 
@@ -1070,10 +1075,10 @@ 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)",
+	VCPU_EVENT(vcpu, 3, "inject: set prefix to %x",
 		   irq->u.prefix.address);
 	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_SIGP_SET_PREFIX,
-				   irq->u.prefix.address, 0, 2);
+				   irq->u.prefix.address, 0);
 
 	if (!is_vcpu_stopped(vcpu))
 		return -EBUSY;
@@ -1090,7 +1095,7 @@ static int __inject_sigp_stop(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
 	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);
+	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_SIGP_STOP, 0, 0);
 
 	if (irq->u.stop.flags & ~KVM_S390_STOP_SUPP_FLAGS)
 		return -EINVAL;
@@ -1114,8 +1119,8 @@ static int __inject_sigp_restart(struct kvm_vcpu *vcpu,
 {
 	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);
+	VCPU_EVENT(vcpu, 3, "%s", "inject: restart int");
+	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_RESTART, 0, 0);
 
 	set_bit(IRQ_PEND_RESTART, &li->pending_irqs);
 	return 0;
@@ -1126,14 +1131,18 @@ static int __inject_sigp_emergency(struct kvm_vcpu *vcpu,
 {
 	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
 
-	VCPU_EVENT(vcpu, 3, "inject: emergency %u\n",
+	VCPU_EVENT(vcpu, 4, "inject: emergency from cpu %u",
 		   irq->u.emerg.code);
 	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_EMERGENCY,
-				   irq->u.emerg.code, 0, 2);
+				   irq->u.emerg.code, 0);
+
+	/* sending vcpu invalid */
+	if (kvm_get_vcpu_by_id(vcpu->kvm, irq->u.emerg.code) == NULL)
+		return -EINVAL;
 
 	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);
+	atomic_or(CPUSTAT_EXT_INT, li->cpuflags);
 	return 0;
 }
 
@@ -1142,10 +1151,10 @@ 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",
+	VCPU_EVENT(vcpu, 3, "inject: machine check mcic 0x%llx",
 		   irq->u.mchk.mcic);
 	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_MCHK, 0,
-				   irq->u.mchk.mcic, 2);
+				   irq->u.mchk.mcic);
 
 	/*
 	 * Because repressible machine checks can be indicated along with
@@ -1172,12 +1181,12 @@ 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);
+	VCPU_EVENT(vcpu, 3, "%s", "inject: clock comparator external");
 	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_CLOCK_COMP,
-				   0, 0, 2);
+				   0, 0);
 
 	set_bit(IRQ_PEND_EXT_CLOCK_COMP, &li->pending_irqs);
-	atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
+	atomic_or(CPUSTAT_EXT_INT, li->cpuflags);
 	return 0;
 }
 
@@ -1185,12 +1194,12 @@ 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);
+	VCPU_EVENT(vcpu, 3, "%s", "inject: cpu timer external");
 	trace_kvm_s390_inject_vcpu(vcpu->vcpu_id, KVM_S390_INT_CPU_TIMER,
-				   0, 0, 2);
+				   0, 0);
 
 	set_bit(IRQ_PEND_EXT_CPU_TIMER, &li->pending_irqs);
-	atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
+	atomic_or(CPUSTAT_EXT_INT, li->cpuflags);
 	return 0;
 }
 
@@ -1369,13 +1378,13 @@ static void __floating_irq_kick(struct kvm *kvm, u64 type)
 	spin_lock(&li->lock);
 	switch (type) {
 	case KVM_S390_MCHK:
-		atomic_set_mask(CPUSTAT_STOP_INT, li->cpuflags);
+		atomic_or(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);
+		atomic_or(CPUSTAT_IO_INT, li->cpuflags);
 		break;
 	default:
-		atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags);
+		atomic_or(CPUSTAT_EXT_INT, li->cpuflags);
 		break;
 	}
 	spin_unlock(&li->lock);
@@ -1435,20 +1444,20 @@ int kvm_s390_inject_vm(struct kvm *kvm,
 		inti->ext.ext_params2 = s390int->parm64;
 		break;
 	case KVM_S390_INT_SERVICE:
-		VM_EVENT(kvm, 5, "inject: sclp parm:%x", s390int->parm);
+		VM_EVENT(kvm, 4, "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",
+		VM_EVENT(kvm, 3, "inject: machine check mcic 0x%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)
+		if (inti->type & KVM_S390_INT_IO_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",
@@ -1535,8 +1544,6 @@ static int do_inject_vcpu(struct kvm_vcpu *vcpu, struct kvm_s390_irq *irq)
 
 	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:
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index f32f843a3..1290af834 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -28,6 +28,7 @@
 #include <linux/vmalloc.h>
 #include <asm/asm-offsets.h>
 #include <asm/lowcore.h>
+#include <asm/etr.h>
 #include <asm/pgtable.h>
 #include <asm/nmi.h>
 #include <asm/switch_to.h>
@@ -62,6 +63,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "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_attempted_poll", VCPU_STAT(halt_attempted_poll) },
 	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
 	{ "instruction_lctlg", VCPU_STAT(instruction_lctlg) },
 	{ "instruction_lctl", VCPU_STAT(instruction_lctl) },
@@ -108,6 +110,9 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "diagnose_10", VCPU_STAT(diagnose_10) },
 	{ "diagnose_44", VCPU_STAT(diagnose_44) },
 	{ "diagnose_9c", VCPU_STAT(diagnose_9c) },
+	{ "diagnose_258", VCPU_STAT(diagnose_258) },
+	{ "diagnose_308", VCPU_STAT(diagnose_308) },
+	{ "diagnose_500", VCPU_STAT(diagnose_500) },
 	{ NULL }
 };
 
@@ -124,6 +129,7 @@ unsigned long kvm_s390_fac_list_mask_size(void)
 }
 
 static struct gmap_notifier gmap_notifier;
+debug_info_t *kvm_s390_dbf;
 
 /* Section: not file related */
 int kvm_arch_hardware_enable(void)
@@ -134,24 +140,69 @@ int kvm_arch_hardware_enable(void)
 
 static void kvm_gmap_notifier(struct gmap *gmap, unsigned long address);
 
+/*
+ * This callback is executed during stop_machine(). All CPUs are therefore
+ * temporarily stopped. In order not to change guest behavior, we have to
+ * disable preemption whenever we touch the epoch of kvm and the VCPUs,
+ * so a CPU won't be stopped while calculating with the epoch.
+ */
+static int kvm_clock_sync(struct notifier_block *notifier, unsigned long val,
+			  void *v)
+{
+	struct kvm *kvm;
+	struct kvm_vcpu *vcpu;
+	int i;
+	unsigned long long *delta = v;
+
+	list_for_each_entry(kvm, &vm_list, vm_list) {
+		kvm->arch.epoch -= *delta;
+		kvm_for_each_vcpu(i, vcpu, kvm) {
+			vcpu->arch.sie_block->epoch -= *delta;
+		}
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block kvm_clock_notifier = {
+	.notifier_call = kvm_clock_sync,
+};
+
 int kvm_arch_hardware_setup(void)
 {
 	gmap_notifier.notifier_call = kvm_gmap_notifier;
 	gmap_register_ipte_notifier(&gmap_notifier);
+	atomic_notifier_chain_register(&s390_epoch_delta_notifier,
+				       &kvm_clock_notifier);
 	return 0;
 }
 
 void kvm_arch_hardware_unsetup(void)
 {
 	gmap_unregister_ipte_notifier(&gmap_notifier);
+	atomic_notifier_chain_unregister(&s390_epoch_delta_notifier,
+					 &kvm_clock_notifier);
 }
 
 int kvm_arch_init(void *opaque)
 {
+	kvm_s390_dbf = debug_register("kvm-trace", 32, 1, 7 * sizeof(long));
+	if (!kvm_s390_dbf)
+		return -ENOMEM;
+
+	if (debug_register_view(kvm_s390_dbf, &debug_sprintf_view)) {
+		debug_unregister(kvm_s390_dbf);
+		return -ENOMEM;
+	}
+
 	/* Register floating interrupt controller interface. */
 	return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
 }
 
+void kvm_arch_exit(void)
+{
+	debug_unregister(kvm_s390_dbf);
+}
+
 /* Section: device related */
 long kvm_arch_dev_ioctl(struct file *filp,
 			unsigned int ioctl, unsigned long arg)
@@ -281,22 +332,31 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap)
 
 	switch (cap->cap) {
 	case KVM_CAP_S390_IRQCHIP:
+		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_IRQCHIP");
 		kvm->arch.use_irqchip = 1;
 		r = 0;
 		break;
 	case KVM_CAP_S390_USER_SIGP:
+		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_SIGP");
 		kvm->arch.user_sigp = 1;
 		r = 0;
 		break;
 	case KVM_CAP_S390_VECTOR_REGISTERS:
-		if (MACHINE_HAS_VX) {
+		mutex_lock(&kvm->lock);
+		if (atomic_read(&kvm->online_vcpus)) {
+			r = -EBUSY;
+		} else 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;
+		mutex_unlock(&kvm->lock);
+		VM_EVENT(kvm, 3, "ENABLE: CAP_S390_VECTOR_REGISTERS %s",
+			 r ? "(not available)" : "(success)");
 		break;
 	case KVM_CAP_S390_USER_STSI:
+		VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_STSI");
 		kvm->arch.user_stsi = 1;
 		r = 0;
 		break;
@@ -314,6 +374,8 @@ static int kvm_s390_get_mem_control(struct kvm *kvm, struct kvm_device_attr *att
 	switch (attr->attr) {
 	case KVM_S390_VM_MEM_LIMIT_SIZE:
 		ret = 0;
+		VM_EVENT(kvm, 3, "QUERY: max guest memory: %lu bytes",
+			 kvm->arch.gmap->asce_end);
 		if (put_user(kvm->arch.gmap->asce_end, (u64 __user *)attr->addr))
 			ret = -EFAULT;
 		break;
@@ -330,7 +392,13 @@ static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *att
 	unsigned int idx;
 	switch (attr->attr) {
 	case KVM_S390_VM_MEM_ENABLE_CMMA:
+		/* enable CMMA only for z10 and later (EDAT_1) */
+		ret = -EINVAL;
+		if (!MACHINE_IS_LPAR || !MACHINE_HAS_EDAT1)
+			break;
+
 		ret = -EBUSY;
+		VM_EVENT(kvm, 3, "%s", "ENABLE: CMMA support");
 		mutex_lock(&kvm->lock);
 		if (atomic_read(&kvm->online_vcpus) == 0) {
 			kvm->arch.use_cmma = 1;
@@ -339,6 +407,11 @@ static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *att
 		mutex_unlock(&kvm->lock);
 		break;
 	case KVM_S390_VM_MEM_CLR_CMMA:
+		ret = -EINVAL;
+		if (!kvm->arch.use_cmma)
+			break;
+
+		VM_EVENT(kvm, 3, "%s", "RESET: CMMA states");
 		mutex_lock(&kvm->lock);
 		idx = srcu_read_lock(&kvm->srcu);
 		s390_reset_cmma(kvm->arch.gmap->mm);
@@ -374,6 +447,7 @@ static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *att
 			}
 		}
 		mutex_unlock(&kvm->lock);
+		VM_EVENT(kvm, 3, "SET: max guest memory: %lu bytes", new_limit);
 		break;
 	}
 	default:
@@ -400,22 +474,26 @@ static int kvm_s390_vm_set_crypto(struct kvm *kvm, struct kvm_device_attr *attr)
 			kvm->arch.crypto.crycb->aes_wrapping_key_mask,
 			sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask));
 		kvm->arch.crypto.aes_kw = 1;
+		VM_EVENT(kvm, 3, "%s", "ENABLE: AES keywrapping support");
 		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;
+		VM_EVENT(kvm, 3, "%s", "ENABLE: DEA keywrapping support");
 		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));
+		VM_EVENT(kvm, 3, "%s", "DISABLE: AES keywrapping support");
 		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));
+		VM_EVENT(kvm, 3, "%s", "DISABLE: DEA keywrapping support");
 		break;
 	default:
 		mutex_unlock(&kvm->lock);
@@ -440,6 +518,7 @@ static int kvm_s390_set_tod_high(struct kvm *kvm, struct kvm_device_attr *attr)
 
 	if (gtod_high != 0)
 		return -EINVAL;
+	VM_EVENT(kvm, 3, "SET: TOD extension: 0x%x\n", gtod_high);
 
 	return 0;
 }
@@ -459,12 +538,15 @@ static int kvm_s390_set_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
 		return r;
 
 	mutex_lock(&kvm->lock);
+	preempt_disable();
 	kvm->arch.epoch = gtod - host_tod;
 	kvm_s390_vcpu_block_all(kvm);
 	kvm_for_each_vcpu(vcpu_idx, cur_vcpu, kvm)
 		cur_vcpu->arch.sie_block->epoch = kvm->arch.epoch;
 	kvm_s390_vcpu_unblock_all(kvm);
+	preempt_enable();
 	mutex_unlock(&kvm->lock);
+	VM_EVENT(kvm, 3, "SET: TOD base: 0x%llx\n", gtod);
 	return 0;
 }
 
@@ -496,6 +578,7 @@ static int kvm_s390_get_tod_high(struct kvm *kvm, struct kvm_device_attr *attr)
 	if (copy_to_user((void __user *)attr->addr, &gtod_high,
 					 sizeof(gtod_high)))
 		return -EFAULT;
+	VM_EVENT(kvm, 3, "QUERY: TOD extension: 0x%x\n", gtod_high);
 
 	return 0;
 }
@@ -509,9 +592,12 @@ static int kvm_s390_get_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
 	if (r)
 		return r;
 
+	preempt_disable();
 	gtod = host_tod + kvm->arch.epoch;
+	preempt_enable();
 	if (copy_to_user((void __user *)attr->addr, &gtod, sizeof(gtod)))
 		return -EFAULT;
+	VM_EVENT(kvm, 3, "QUERY: TOD base: 0x%llx\n", gtod);
 
 	return 0;
 }
@@ -821,7 +907,9 @@ static long kvm_s390_set_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
 	}
 
 	/* Enable storage key handling for the guest */
-	s390_enable_skey();
+	r = s390_enable_skey();
+	if (r)
+		goto out;
 
 	for (i = 0; i < args->count; i++) {
 		hva = gfn_to_hva(kvm, args->start_gfn + i);
@@ -879,8 +967,7 @@ long kvm_arch_vm_ioctl(struct file *filp,
 		if (kvm->arch.use_irqchip) {
 			/* Set up dummy routing. */
 			memset(&routing, 0, sizeof(routing));
-			kvm_set_irq_routing(kvm, &routing, 0, 0);
-			r = 0;
+			r = kvm_set_irq_routing(kvm, &routing, 0, 0);
 		}
 		break;
 	}
@@ -1037,13 +1124,15 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	if (!kvm->arch.sca)
 		goto out_err;
 	spin_lock(&kvm_lock);
-	sca_offset = (sca_offset + 16) & 0x7f0;
+	sca_offset += 16;
+	if (sca_offset + sizeof(struct sca_block) > PAGE_SIZE)
+		sca_offset = 0;
 	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));
+	kvm->arch.dbf = debug_register(debug_name, 32, 1, 7 * sizeof(long));
 	if (!kvm->arch.dbf)
 		goto out_err;
 
@@ -1086,7 +1175,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	mutex_init(&kvm->arch.ipte_mutex);
 
 	debug_register_view(kvm->arch.dbf, &debug_sprintf_view);
-	VM_EVENT(kvm, 3, "%s", "vm created");
+	VM_EVENT(kvm, 3, "vm created with type %lu", type);
 
 	if (type & KVM_VM_S390_UCONTROL) {
 		kvm->arch.gmap = NULL;
@@ -1103,6 +1192,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	kvm->arch.epoch = 0;
 
 	spin_lock_init(&kvm->arch.start_stop_lock);
+	KVM_EVENT(3, "vm 0x%p created by pid %u", kvm, current->pid);
 
 	return 0;
 out_err:
@@ -1110,6 +1200,7 @@ out_err:
 	free_page((unsigned long)kvm->arch.model.fac);
 	debug_unregister(kvm->arch.dbf);
 	free_page((unsigned long)(kvm->arch.sca));
+	KVM_EVENT(3, "creation of vm failed: %d", rc);
 	return rc;
 }
 
@@ -1131,7 +1222,7 @@ void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
 	if (kvm_is_ucontrol(vcpu->kvm))
 		gmap_free(vcpu->arch.gmap);
 
-	if (kvm_s390_cmma_enabled(vcpu->kvm))
+	if (vcpu->kvm->arch.use_cmma)
 		kvm_s390_vcpu_unsetup_cmma(vcpu);
 	free_page((unsigned long)(vcpu->arch.sie_block));
 
@@ -1166,6 +1257,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 		gmap_free(kvm->arch.gmap);
 	kvm_s390_destroy_adapters(kvm);
 	kvm_s390_clear_float_irqs(kvm);
+	KVM_EVENT(3, "vm 0x%p destroyed", kvm);
 }
 
 /* Section: vcpu related */
@@ -1198,43 +1290,79 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
+/*
+ * Backs up the current FP/VX register save area on a particular
+ * destination.  Used to switch between different register save
+ * areas.
+ */
+static inline void save_fpu_to(struct fpu *dst)
+{
+	dst->fpc = current->thread.fpu.fpc;
+	dst->flags = current->thread.fpu.flags;
+	dst->regs = current->thread.fpu.regs;
+}
+
+/*
+ * Switches the FP/VX register save area from which to lazy
+ * restore register contents.
+ */
+static inline void load_fpu_from(struct fpu *from)
+{
+	current->thread.fpu.fpc = from->fpc;
+	current->thread.fpu.flags = from->flags;
+	current->thread.fpu.regs = from->regs;
+}
+
 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);
+	/* Save host register state */
+	save_fpu_regs();
+	save_fpu_to(&vcpu->arch.host_fpregs);
+
 	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);
-	}
+		current->thread.fpu.fpc = vcpu->run->s.regs.fpc;
+		current->thread.fpu.flags = FPU_USE_VX;
+		/*
+		 * Use the register save area in the SIE-control block
+		 * for register restore and save in kvm_arch_vcpu_put()
+		 */
+		current->thread.fpu.vxrs =
+			(__vector128 *)&vcpu->run->s.regs.vrs;
+		/* Always enable the vector extension for KVM */
+		__ctl_set_vx();
+	} else
+		load_fpu_from(&vcpu->arch.guest_fpregs);
+
+	if (test_fp_ctl(current->thread.fpu.fpc))
+		/* User space provided an invalid FPC, let's clear it */
+		current->thread.fpu.fpc = 0;
+
+	save_access_regs(vcpu->arch.host_acrs);
 	restore_access_regs(vcpu->run->s.regs.acrs);
 	gmap_enable(vcpu->arch.gmap);
-	atomic_set_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
+	atomic_or(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);
+	atomic_andnot(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);
+
+	save_fpu_regs();
+
 	if (test_kvm_facility(vcpu->kvm, 129))
-		restore_vx_regs((__vector128 *)&vcpu->arch.host_vregs->vrs);
+		/*
+		 * kvm_arch_vcpu_load() set up the register save area to
+		 * the &vcpu->run->s.regs.vrs and, thus, the vector registers
+		 * are already saved.  Only the floating-point control must be
+		 * copied.
+		 */
+		vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
 	else
-		restore_fp_regs(vcpu->arch.host_fpregs.fprs);
+		save_fpu_to(&vcpu->arch.guest_fpregs);
+	load_fpu_from(&vcpu->arch.host_fpregs);
+
+	save_access_regs(vcpu->run->s.regs.acrs);
 	restore_access_regs(vcpu->arch.host_acrs);
 }
 
@@ -1264,7 +1392,9 @@ static void kvm_s390_vcpu_initial_reset(struct kvm_vcpu *vcpu)
 void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
 {
 	mutex_lock(&vcpu->kvm->lock);
+	preempt_disable();
 	vcpu->arch.sie_block->epoch = vcpu->kvm->arch.epoch;
+	preempt_enable();
 	mutex_unlock(&vcpu->kvm->lock);
 	if (!kvm_is_ucontrol(vcpu->kvm))
 		vcpu->arch.gmap = vcpu->kvm->arch.gmap;
@@ -1320,9 +1450,9 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 						    CPUSTAT_STOPPED);
 
 	if (test_kvm_facility(vcpu->kvm, 78))
-		atomic_set_mask(CPUSTAT_GED2, &vcpu->arch.sie_block->cpuflags);
+		atomic_or(CPUSTAT_GED2, &vcpu->arch.sie_block->cpuflags);
 	else if (test_kvm_facility(vcpu->kvm, 8))
-		atomic_set_mask(CPUSTAT_GED, &vcpu->arch.sie_block->cpuflags);
+		atomic_or(CPUSTAT_GED, &vcpu->arch.sie_block->cpuflags);
 
 	kvm_s390_vcpu_setup_model(vcpu);
 
@@ -1342,7 +1472,7 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 	}
 	vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE;
 
-	if (kvm_s390_cmma_enabled(vcpu->kvm)) {
+	if (vcpu->kvm->arch.use_cmma) {
 		rc = kvm_s390_vcpu_setup_cmma(vcpu);
 		if (rc)
 			return rc;
@@ -1377,7 +1507,6 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
 
 	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)) {
@@ -1399,6 +1528,19 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
 	vcpu->arch.local_int.wq = &vcpu->wq;
 	vcpu->arch.local_int.cpuflags = &vcpu->arch.sie_block->cpuflags;
 
+	/*
+	 * Allocate a save area for floating-point registers.  If the vector
+	 * extension is available, register contents are saved in the SIE
+	 * control block.  The allocated save area is still required in
+	 * particular places, for example, in kvm_s390_vcpu_store_status().
+	 */
+	vcpu->arch.guest_fpregs.fprs = kzalloc(sizeof(freg_t) * __NUM_FPRS,
+					       GFP_KERNEL);
+	if (!vcpu->arch.guest_fpregs.fprs) {
+		rc = -ENOMEM;
+		goto out_free_sie_block;
+	}
+
 	rc = kvm_vcpu_init(vcpu, kvm, id);
 	if (rc)
 		goto out_free_sie_block;
@@ -1422,24 +1564,24 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
 
 void kvm_s390_vcpu_block(struct kvm_vcpu *vcpu)
 {
-	atomic_set_mask(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
+	atomic_or(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
 	exit_sie(vcpu);
 }
 
 void kvm_s390_vcpu_unblock(struct kvm_vcpu *vcpu)
 {
-	atomic_clear_mask(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
+	atomic_andnot(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
 }
 
 static void kvm_s390_vcpu_request(struct kvm_vcpu *vcpu)
 {
-	atomic_set_mask(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
+	atomic_or(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
 	exit_sie(vcpu);
 }
 
 static void kvm_s390_vcpu_request_handled(struct kvm_vcpu *vcpu)
 {
-	atomic_clear_mask(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
+	atomic_andnot(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
 }
 
 /*
@@ -1448,7 +1590,7 @@ static void kvm_s390_vcpu_request_handled(struct kvm_vcpu *vcpu)
  * return immediately. */
 void exit_sie(struct kvm_vcpu *vcpu)
 {
-	atomic_set_mask(CPUSTAT_STOP_INT, &vcpu->arch.sie_block->cpuflags);
+	atomic_or(CPUSTAT_STOP_INT, &vcpu->arch.sie_block->cpuflags);
 	while (vcpu->arch.sie_block->prog0c & PROG_IN_SIE)
 		cpu_relax();
 }
@@ -1621,16 +1763,16 @@ 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));
+	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);
+	save_fpu_regs();
+	load_fpu_from(&vcpu->arch.guest_fpregs);
 	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));
+	memcpy(&fpu->fprs, vcpu->arch.guest_fpregs.fprs, sizeof(fpu->fprs));
 	fpu->fpc = vcpu->arch.guest_fpregs.fpc;
 	return 0;
 }
@@ -1672,19 +1814,19 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
 	if (dbg->control & KVM_GUESTDBG_ENABLE) {
 		vcpu->guest_debug = dbg->control;
 		/* enforce guest PER */
-		atomic_set_mask(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
+		atomic_or(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);
+		atomic_andnot(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);
+		atomic_andnot(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
 	}
 
 	return rc;
@@ -1723,18 +1865,6 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
 	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;
@@ -1771,7 +1901,7 @@ 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,
+			atomic_or(CPUSTAT_IBS,
 					&vcpu->arch.sie_block->cpuflags);
 		}
 		goto retry;
@@ -1780,7 +1910,7 @@ 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,
+			atomic_andnot(CPUSTAT_IBS,
 					  &vcpu->arch.sie_block->cpuflags);
 		}
 		goto retry;
@@ -2193,8 +2323,21 @@ int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr)
 	 * 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_fpu_regs();
+	if (test_kvm_facility(vcpu->kvm, 129)) {
+		/*
+		 * If the vector extension is available, the vector registers
+		 * which overlaps with floating-point registers are saved in
+		 * the SIE-control block.  Hence, extract the floating-point
+		 * registers and the FPC value and store them in the
+		 * guest_fpregs structure.
+		 */
+		WARN_ON(!is_vx_task(current));	  /* XXX remove later */
+		vcpu->arch.guest_fpregs.fpc = current->thread.fpu.fpc;
+		convert_vx_to_fp(vcpu->arch.guest_fpregs.fprs,
+				 current->thread.fpu.vxrs);
+	} else
+		save_fpu_to(&vcpu->arch.guest_fpregs);
 	save_access_regs(vcpu->run->s.regs.acrs);
 
 	return kvm_s390_store_status_unloaded(vcpu, addr);
@@ -2221,10 +2364,13 @@ int kvm_s390_vcpu_store_adtl_status(struct kvm_vcpu *vcpu, unsigned long addr)
 
 	/*
 	 * 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.
+	 * copying in vcpu load/put. We can simply call save_fpu_regs()
+	 * to save the current register state because we are in the
+	 * middle of a load/put cycle.
+	 *
+	 * Let's update our copies before we save it into the save area.
 	 */
-	save_vx_regs((__vector128 *)&vcpu->run->s.regs.vrs);
+	save_fpu_regs();
 
 	return kvm_s390_store_adtl_status_unloaded(vcpu, addr);
 }
@@ -2280,7 +2426,7 @@ void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu)
 		__disable_ibs_on_all_vcpus(vcpu->kvm);
 	}
 
-	atomic_clear_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
+	atomic_andnot(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.
@@ -2306,7 +2452,7 @@ void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu)
 	/* 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);
+	atomic_or(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
 	__disable_ibs_on_vcpu(vcpu);
 
 	for (i = 0; i < online_vcpus; i++) {
@@ -2340,6 +2486,7 @@ static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu,
 	case KVM_CAP_S390_CSS_SUPPORT:
 		if (!vcpu->kvm->arch.css_support) {
 			vcpu->kvm->arch.css_support = 1;
+			VM_EVENT(vcpu->kvm, 3, "%s", "ENABLE: CSS support");
 			trace_kvm_s390_enable_css(vcpu->kvm);
 		}
 		r = 0;
diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h
index c5704786e..c446aabf6 100644
--- a/arch/s390/kvm/kvm-s390.h
+++ b/arch/s390/kvm/kvm-s390.h
@@ -27,6 +27,13 @@ typedef int (*intercept_handler_t)(struct kvm_vcpu *vcpu);
 #define TDB_FORMAT1		1
 #define IS_ITDB_VALID(vcpu)	((*(char *)vcpu->arch.sie_block->itdba == TDB_FORMAT1))
 
+extern debug_info_t *kvm_s390_dbf;
+#define KVM_EVENT(d_loglevel, d_string, d_args...)\
+do { \
+	debug_sprintf_event(kvm_s390_dbf, d_loglevel, d_string "\n", \
+	  d_args); \
+} while (0)
+
 #define VM_EVENT(d_kvm, d_loglevel, d_string, d_args...)\
 do { \
 	debug_sprintf_event(d_kvm->arch.dbf, d_loglevel, d_string "\n", \
@@ -65,6 +72,8 @@ static inline u32 kvm_s390_get_prefix(struct kvm_vcpu *vcpu)
 
 static inline void kvm_s390_set_prefix(struct kvm_vcpu *vcpu, u32 prefix)
 {
+	VCPU_EVENT(vcpu, 3, "set prefix of cpu %03u to 0x%x", vcpu->vcpu_id,
+		   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);
@@ -217,8 +226,6 @@ void exit_sie(struct kvm_vcpu *vcpu);
 void kvm_s390_sync_request(int req, 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[];
 
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index ad4242245..4d21dc4d1 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -53,11 +53,14 @@ static int handle_set_clock(struct kvm_vcpu *vcpu)
 		kvm_s390_set_psw_cc(vcpu, 3);
 		return 0;
 	}
+	VCPU_EVENT(vcpu, 3, "SCK: setting guest TOD to 0x%llx", val);
 	val = (val - hostclk) & ~0x3fUL;
 
 	mutex_lock(&vcpu->kvm->lock);
+	preempt_disable();
 	kvm_for_each_vcpu(i, cpup, vcpu->kvm)
 		cpup->arch.sie_block->epoch = val;
+	preempt_enable();
 	mutex_unlock(&vcpu->kvm->lock);
 
 	kvm_s390_set_psw_cc(vcpu, 0);
@@ -98,8 +101,6 @@ static int handle_set_prefix(struct kvm_vcpu *vcpu)
 		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;
 }
@@ -129,7 +130,7 @@ static int handle_store_prefix(struct kvm_vcpu *vcpu)
 	if (rc)
 		return kvm_s390_inject_prog_cond(vcpu, rc);
 
-	VCPU_EVENT(vcpu, 5, "storing prefix to %x", address);
+	VCPU_EVENT(vcpu, 3, "STPX: storing prefix 0x%x into 0x%llx", address, operand2);
 	trace_kvm_s390_handle_prefix(vcpu, 0, address);
 	return 0;
 }
@@ -155,7 +156,7 @@ static int handle_store_cpu_address(struct kvm_vcpu *vcpu)
 	if (rc)
 		return kvm_s390_inject_prog_cond(vcpu, rc);
 
-	VCPU_EVENT(vcpu, 5, "storing cpu address to %llx", ga);
+	VCPU_EVENT(vcpu, 3, "STAP: storing cpu address (%u) to 0x%llx", vcpu_id, ga);
 	trace_kvm_s390_handle_stap(vcpu, ga);
 	return 0;
 }
@@ -167,6 +168,7 @@ static int __skey_check_enable(struct kvm_vcpu *vcpu)
 		return rc;
 
 	rc = s390_enable_skey();
+	VCPU_EVENT(vcpu, 3, "%s", "enabling storage keys for guest");
 	trace_kvm_s390_skey_related_inst(vcpu);
 	vcpu->arch.sie_block->ictl &= ~(ICTL_ISKE | ICTL_SSKE | ICTL_RRBE);
 	return rc;
@@ -370,7 +372,7 @@ static int handle_stfl(struct kvm_vcpu *vcpu)
 			    &fac, sizeof(fac));
 	if (rc)
 		return rc;
-	VCPU_EVENT(vcpu, 5, "store facility list value %x", fac);
+	VCPU_EVENT(vcpu, 3, "STFL: store facility list 0x%x", fac);
 	trace_kvm_s390_handle_stfl(vcpu, fac);
 	return 0;
 }
@@ -468,7 +470,7 @@ static int handle_stidp(struct kvm_vcpu *vcpu)
 	if (rc)
 		return kvm_s390_inject_prog_cond(vcpu, rc);
 
-	VCPU_EVENT(vcpu, 5, "%s", "store cpu id");
+	VCPU_EVENT(vcpu, 3, "STIDP: store cpu id 0x%llx", stidp_data);
 	return 0;
 }
 
@@ -521,7 +523,7 @@ static int handle_stsi(struct kvm_vcpu *vcpu)
 	ar_t ar;
 
 	vcpu->stat.instruction_stsi++;
-	VCPU_EVENT(vcpu, 4, "stsi: fc: %x sel1: %x sel2: %x", fc, sel1, sel2);
+	VCPU_EVENT(vcpu, 3, "STSI: fc: %u sel1: %u sel2: %u", fc, sel1, sel2);
 
 	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
 		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);
@@ -758,10 +760,10 @@ static int handle_essa(struct kvm_vcpu *vcpu)
 	struct gmap *gmap;
 	int i;
 
-	VCPU_EVENT(vcpu, 5, "cmma release %d pages", entries);
+	VCPU_EVENT(vcpu, 4, "ESSA: release %d pages", entries);
 	gmap = vcpu->arch.gmap;
 	vcpu->stat.instruction_essa++;
-	if (!kvm_s390_cmma_enabled(vcpu->kvm))
+	if (!vcpu->kvm->arch.use_cmma)
 		return kvm_s390_inject_program_int(vcpu, PGM_OPERATION);
 
 	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
@@ -829,7 +831,7 @@ int kvm_s390_handle_lctl(struct kvm_vcpu *vcpu)
 	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);
+	VCPU_EVENT(vcpu, 4, "LCTL: r1:%d, r3:%d, addr: 0x%llx", reg1, reg3, ga);
 	trace_kvm_s390_handle_lctl(vcpu, 0, reg1, reg3, ga);
 
 	nr_regs = ((reg3 - reg1) & 0xf) + 1;
@@ -868,7 +870,7 @@ int kvm_s390_handle_stctl(struct kvm_vcpu *vcpu)
 	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);
+	VCPU_EVENT(vcpu, 4, "STCTL r1:%d, r3:%d, addr: 0x%llx", reg1, reg3, ga);
 	trace_kvm_s390_handle_stctl(vcpu, 0, reg1, reg3, ga);
 
 	reg = reg1;
@@ -902,7 +904,7 @@ static int handle_lctlg(struct kvm_vcpu *vcpu)
 	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);
+	VCPU_EVENT(vcpu, 4, "LCTLG: r1:%d, r3:%d, addr: 0x%llx", reg1, reg3, ga);
 	trace_kvm_s390_handle_lctl(vcpu, 1, reg1, reg3, ga);
 
 	nr_regs = ((reg3 - reg1) & 0xf) + 1;
@@ -940,7 +942,7 @@ static int handle_stctg(struct kvm_vcpu *vcpu)
 	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);
+	VCPU_EVENT(vcpu, 4, "STCTG r1:%d, r3:%d, addr: 0x%llx", reg1, reg3, ga);
 	trace_kvm_s390_handle_stctl(vcpu, 1, reg1, reg3, ga);
 
 	reg = reg1;
diff --git a/arch/s390/kvm/sigp.c b/arch/s390/kvm/sigp.c
index 72e58bd2b..77c22d685 100644
--- a/arch/s390/kvm/sigp.c
+++ b/arch/s390/kvm/sigp.c
@@ -205,9 +205,6 @@ static int __sigp_set_prefix(struct kvm_vcpu *vcpu, struct kvm_vcpu *dst_vcpu,
 		*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;
@@ -294,12 +291,8 @@ 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;
+	struct kvm_vcpu *dst_vcpu = kvm_get_vcpu_by_id(vcpu->kvm, cpu_addr);
 
-	dst_vcpu = kvm_get_vcpu(vcpu->kvm, cpu_addr);
 	if (!dst_vcpu)
 		return SIGP_CC_NOT_OPERATIONAL;
 
@@ -371,7 +364,8 @@ static int handle_sigp_dst(struct kvm_vcpu *vcpu, u8 order_code,
 	return rc;
 }
 
-static int handle_sigp_order_in_user_space(struct kvm_vcpu *vcpu, u8 order_code)
+static int handle_sigp_order_in_user_space(struct kvm_vcpu *vcpu, u8 order_code,
+					   u16 cpu_addr)
 {
 	if (!vcpu->kvm->arch.user_sigp)
 		return 0;
@@ -414,9 +408,8 @@ static int handle_sigp_order_in_user_space(struct kvm_vcpu *vcpu, u8 order_code)
 	default:
 		vcpu->stat.instruction_sigp_unknown++;
 	}
-
-	VCPU_EVENT(vcpu, 4, "sigp order %u: completely handled in user space",
-		   order_code);
+	VCPU_EVENT(vcpu, 3, "SIGP: order %u for CPU %d handled in userspace",
+		   order_code, cpu_addr);
 
 	return 1;
 }
@@ -435,7 +428,7 @@ int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu)
 		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))
+	if (handle_sigp_order_in_user_space(vcpu, order_code, cpu_addr))
 		return -EOPNOTSUPP;
 
 	if (r1 % 2)
@@ -481,7 +474,7 @@ int kvm_s390_handle_sigp_pei(struct kvm_vcpu *vcpu)
 	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);
+		dest_vcpu = kvm_get_vcpu_by_id(vcpu->kvm, cpu_addr);
 		BUG_ON(dest_vcpu == NULL);
 
 		kvm_s390_vcpu_wakeup(dest_vcpu);
diff --git a/arch/s390/kvm/trace-s390.h b/arch/s390/kvm/trace-s390.h
index 3208d33a4..cc1d6c683 100644
--- a/arch/s390/kvm/trace-s390.h
+++ b/arch/s390/kvm/trace-s390.h
@@ -105,11 +105,22 @@ TRACE_EVENT(kvm_s390_vcpu_start_stop,
 	{KVM_S390_PROGRAM_INT, "program interrupt"},			\
 	{KVM_S390_SIGP_SET_PREFIX, "sigp set prefix"},			\
 	{KVM_S390_RESTART, "sigp restart"},				\
+	{KVM_S390_INT_PFAULT_INIT, "pfault init"},			\
+	{KVM_S390_INT_PFAULT_DONE, "pfault done"},			\
+	{KVM_S390_MCHK, "machine check"},				\
+	{KVM_S390_INT_CLOCK_COMP, "clock comparator"},			\
+	{KVM_S390_INT_CPU_TIMER, "cpu timer"},				\
 	{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"}
 
+#define get_irq_name(__type) \
+	(__type > KVM_S390_INT_IO_MAX ? \
+	__print_symbolic(__type, kvm_s390_int_type) : \
+		(__type & KVM_S390_INT_IO_AI_MASK ? \
+		 "adapter I/O interrupt" : "subchannel I/O interrupt"))
+
 TRACE_EVENT(kvm_s390_inject_vm,
 	    TP_PROTO(__u64 type, __u32 parm, __u64 parm64, int who),
 	    TP_ARGS(type, parm, parm64, who),
@@ -131,22 +142,19 @@ TRACE_EVENT(kvm_s390_inject_vm,
 	    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->inttype, get_irq_name(__entry->inttype),
 		      __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_PROTO(unsigned int id, __u64 type, __u32 parm, __u64 parm64),
+	    TP_ARGS(id, type, parm, parm64),
 
 	    TP_STRUCT__entry(
 		    __field(int, id)
 		    __field(__u32, inttype)
 		    __field(__u32, parm)
 		    __field(__u64, parm64)
-		    __field(int, who)
 		    ),
 
 	    TP_fast_assign(
@@ -154,15 +162,12 @@ TRACE_EVENT(kvm_s390_inject_vcpu,
 		    __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)" : "",
+	    TP_printk("inject (vcpu %d): type:%x (%s) parm:%x parm64:%llx",
 		      __entry->id, __entry->inttype,
-		      __print_symbolic(__entry->inttype, kvm_s390_int_type),
-		      __entry->parm, __entry->parm64)
+		      get_irq_name(__entry->inttype), __entry->parm,
+		      __entry->parm64)
 	);
 
 /*
@@ -189,8 +194,8 @@ TRACE_EVENT(kvm_s390_deliver_interrupt,
 	    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)
+		      get_irq_name(__entry->inttype), __entry->data0,
+		      __entry->data1)
 	);
 
 /*
diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c
index 16dc42d83..246a7eb4b 100644
--- a/arch/s390/lib/delay.c
+++ b/arch/s390/lib/delay.c
@@ -26,6 +26,7 @@ void __delay(unsigned long loops)
          */
 	asm volatile("0: brct %0,0b" : : "d" ((loops/2) + 1));
 }
+EXPORT_SYMBOL(__delay);
 
 static void __udelay_disabled(unsigned long long usecs)
 {
diff --git a/arch/s390/lib/uaccess.c b/arch/s390/lib/uaccess.c
index 4614d415b..ae4de559e 100644
--- a/arch/s390/lib/uaccess.c
+++ b/arch/s390/lib/uaccess.c
@@ -15,7 +15,7 @@
 #include <asm/mmu_context.h>
 #include <asm/facility.h>
 
-static struct static_key have_mvcos = STATIC_KEY_INIT_FALSE;
+static DEFINE_STATIC_KEY_FALSE(have_mvcos);
 
 static inline unsigned long copy_from_user_mvcos(void *x, const void __user *ptr,
 						 unsigned long size)
@@ -104,7 +104,7 @@ static inline unsigned long copy_from_user_mvcp(void *x, const void __user *ptr,
 
 unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n)
 {
-	if (static_key_false(&have_mvcos))
+	if (static_branch_likely(&have_mvcos))
 		return copy_from_user_mvcos(to, from, n);
 	return copy_from_user_mvcp(to, from, n);
 }
@@ -177,7 +177,7 @@ static inline unsigned long copy_to_user_mvcs(void __user *ptr, const void *x,
 
 unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n)
 {
-	if (static_key_false(&have_mvcos))
+	if (static_branch_likely(&have_mvcos))
 		return copy_to_user_mvcos(to, from, n);
 	return copy_to_user_mvcs(to, from, n);
 }
@@ -240,7 +240,7 @@ static inline unsigned long copy_in_user_mvc(void __user *to, const void __user
 
 unsigned long __copy_in_user(void __user *to, const void __user *from, unsigned long n)
 {
-	if (static_key_false(&have_mvcos))
+	if (static_branch_likely(&have_mvcos))
 		return copy_in_user_mvcos(to, from, n);
 	return copy_in_user_mvc(to, from, n);
 }
@@ -312,7 +312,7 @@ static inline unsigned long clear_user_xc(void __user *to, unsigned long size)
 
 unsigned long __clear_user(void __user *to, unsigned long size)
 {
-	if (static_key_false(&have_mvcos))
+	if (static_branch_likely(&have_mvcos))
 			return clear_user_mvcos(to, size);
 	return clear_user_xc(to, size);
 }
@@ -370,23 +370,10 @@ long __strncpy_from_user(char *dst, const char __user *src, long size)
 }
 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);
+	if (test_facility(27))
+		static_branch_enable(&have_mvcos);
 	return 0;
 }
 early_initcall(uaccess_init);
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index 4c8f5d7f9..f985856a5 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -646,7 +646,7 @@ static void pfault_interrupt(struct ext_code ext_code,
 		return;
 	inc_irq_stat(IRQEXT_PFL);
 	/* Get the token (= pid of the affected task). */
-	pid = sizeof(void *) == 4 ? param32 : param64;
+	pid = param64;
 	rcu_read_lock();
 	tsk = find_task_by_pid_ns(pid, &init_pid_ns);
 	if (tsk)
diff --git a/arch/s390/mm/gup.c b/arch/s390/mm/gup.c
index 1eb41bb30..12bbf0e84 100644
--- a/arch/s390/mm/gup.c
+++ b/arch/s390/mm/gup.c
@@ -30,6 +30,9 @@ static inline int gup_pte_range(pmd_t *pmdp, pmd_t pmd, unsigned long addr,
 	do {
 		pte = *ptep;
 		barrier();
+		/* Similar to the PMD case, NUMA hinting must take slow path */
+		if (pte_protnone(pte))
+			return 0;
 		if ((pte_val(pte) & mask) != 0)
 			return 0;
 		VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
@@ -125,6 +128,13 @@ static inline int gup_pmd_range(pud_t *pudp, pud_t pud, unsigned long addr,
 		if (pmd_none(pmd) || pmd_trans_splitting(pmd))
 			return 0;
 		if (unlikely(pmd_large(pmd))) {
+			/*
+			 * NUMA hinting faults need to be handled in the GUP
+			 * slowpath for accounting purposes and so that they
+			 * can be serialised against THP migration.
+			 */
+			if (pmd_protnone(pmd))
+				return 0;
 			if (!gup_huge_pmd(pmdp, pmd, addr, next,
 					  write, pages, nr))
 				return 0;
diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index 76e873748..c3c07d350 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -27,6 +27,7 @@
 #include <linux/initrd.h>
 #include <linux/export.h>
 #include <linux/gfp.h>
+#include <linux/memblock.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -138,7 +139,7 @@ void __init mem_init(void)
 	cpumask_set_cpu(0, mm_cpumask(&init_mm));
 	atomic_set(&init_mm.context.attach_count, 1);
 
-        max_mapnr = max_low_pfn;
+	set_max_mapnr(max_low_pfn);
         high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);
 
 	/* Setup guest page hinting */
@@ -168,39 +169,38 @@ void __init free_initrd_mem(unsigned long start, unsigned long end)
 #endif
 
 #ifdef CONFIG_MEMORY_HOTPLUG
-int arch_add_memory(int nid, u64 start, u64 size)
+int arch_add_memory(int nid, u64 start, u64 size, bool for_device)
 {
-	unsigned long zone_start_pfn, zone_end_pfn, nr_pages;
+	unsigned long normal_end_pfn = PFN_DOWN(memblock_end_of_DRAM());
+	unsigned long dma_end_pfn = PFN_DOWN(MAX_DMA_ADDRESS);
 	unsigned long start_pfn = PFN_DOWN(start);
 	unsigned long size_pages = PFN_DOWN(size);
-	struct zone *zone;
-	int rc;
+	unsigned long nr_pages;
+	int rc, zone_enum;
 
 	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;
+
+	while (size_pages > 0) {
+		if (start_pfn < dma_end_pfn) {
+			nr_pages = (start_pfn + size_pages > dma_end_pfn) ?
+				   dma_end_pfn - start_pfn : size_pages;
+			zone_enum = ZONE_DMA;
+		} else if (start_pfn < normal_end_pfn) {
+			nr_pages = (start_pfn + size_pages > normal_end_pfn) ?
+				   normal_end_pfn - start_pfn : size_pages;
+			zone_enum = ZONE_NORMAL;
 		} else {
-			/* Add remaining range to ZONE_MOVABLE */
-			zone_start_pfn = start_pfn;
-			zone_end_pfn = start_pfn + size_pages;
+			nr_pages = size_pages;
+			zone_enum = ZONE_MOVABLE;
 		}
-		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);
+		rc = __add_pages(nid, NODE_DATA(nid)->node_zones + zone_enum,
+				 start_pfn, size_pages);
 		if (rc)
 			break;
 		start_pfn += nr_pages;
 		size_pages -= nr_pages;
-		if (!size_pages)
-			break;
 	}
 	if (rc)
 		vmem_remove_mapping(start, size);
diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c
index b33f66110..54ef3bc01 100644
--- a/arch/s390/mm/pgtable.c
+++ b/arch/s390/mm/pgtable.c
@@ -10,11 +10,7 @@
 #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>
@@ -28,12 +24,9 @@
 #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);
+	struct page *page = alloc_pages(GFP_KERNEL, 2);
 
 	if (!page)
 		return NULL;
@@ -42,7 +35,7 @@ unsigned long *crst_table_alloc(struct mm_struct *mm)
 
 void crst_table_free(struct mm_struct *mm, unsigned long *table)
 {
-	free_pages((unsigned long) table, ALLOC_ORDER);
+	free_pages((unsigned long) table, 2);
 }
 
 static void __crst_table_upgrade(void *arg)
@@ -176,7 +169,7 @@ struct gmap *gmap_alloc(struct mm_struct *mm, unsigned long limit)
 	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);
+	page = alloc_pages(GFP_KERNEL, 2);
 	if (!page)
 		goto out_free;
 	page->index = 0;
@@ -247,7 +240,7 @@ void gmap_free(struct gmap *gmap)
 
 	/* Free all segment & region tables. */
 	list_for_each_entry_safe(page, next, &gmap->crst_list, lru)
-		__free_pages(page, ALLOC_ORDER);
+		__free_pages(page, 2);
 	gmap_radix_tree_free(&gmap->guest_to_host);
 	gmap_radix_tree_free(&gmap->host_to_guest);
 	down_write(&gmap->mm->mmap_sem);
@@ -287,7 +280,7 @@ static int gmap_alloc_table(struct gmap *gmap, unsigned long *table,
 	unsigned long *new;
 
 	/* since we dont free the gmap table until gmap_free we can unlock */
-	page = alloc_pages(GFP_KERNEL, ALLOC_ORDER);
+	page = alloc_pages(GFP_KERNEL, 2);
 	if (!page)
 		return -ENOMEM;
 	new = (unsigned long *) page_to_phys(page);
@@ -302,7 +295,7 @@ static int gmap_alloc_table(struct gmap *gmap, unsigned long *table,
 	}
 	spin_unlock(&gmap->mm->page_table_lock);
 	if (page)
-		__free_pages(page, ALLOC_ORDER);
+		__free_pages(page, 2);
 	return 0;
 }
 
@@ -795,40 +788,6 @@ void gmap_do_ipte_notify(struct mm_struct *mm, unsigned long vmaddr, pte_t *pte)
 }
 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)
 {
@@ -957,20 +916,6 @@ __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)
 {
@@ -994,44 +939,55 @@ static inline unsigned int atomic_xor_bits(atomic_t *v, unsigned int bits)
  */
 unsigned long *page_table_alloc(struct mm_struct *mm)
 {
-	unsigned long *uninitialized_var(table);
-	struct page *uninitialized_var(page);
+	unsigned long *table;
+	struct page *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;
+	/* Try to get a fragment of a 4K page as a 2K page table */
+	if (!mm_alloc_pgste(mm)) {
+		table = NULL;
+		spin_lock_bh(&mm->context.list_lock);
+		if (!list_empty(&mm->context.pgtable_list)) {
+			page = list_first_entry(&mm->context.pgtable_list,
+						struct page, lru);
+			mask = atomic_read(&page->_mapcount);
+			mask = (mask | (mask >> 4)) & 3;
+			if (mask != 3) {
+				table = (unsigned long *) page_to_phys(page);
+				bit = mask & 1;		/* =1 -> second 2K */
+				if (bit)
+					table += PTRS_PER_PTE;
+				atomic_xor_bits(&page->_mapcount, 1U << bit);
+				list_del(&page->lru);
+			}
 		}
+		spin_unlock_bh(&mm->context.list_lock);
+		if (table)
+			return table;
+	}
+	/* Allocate a fresh page */
+	page = alloc_page(GFP_KERNEL|__GFP_REPEAT);
+	if (!page)
+		return NULL;
+	if (!pgtable_page_ctor(page)) {
+		__free_page(page);
+		return NULL;
+	}
+	/* Initialize page table */
+	table = (unsigned long *) page_to_phys(page);
+	if (mm_alloc_pgste(mm)) {
+		/* Return 4K page table with PGSTEs */
+		atomic_set(&page->_mapcount, 3);
+		clear_table(table, _PAGE_INVALID, PAGE_SIZE/2);
+		clear_table(table + PTRS_PER_PTE, 0, PAGE_SIZE/2);
+	} else {
+		/* Return the first 2K fragment of the page */
 		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);
 	}
-	spin_unlock_bh(&mm->context.list_lock);
 	return table;
 }
 
@@ -1041,37 +997,23 @@ void page_table_free(struct mm_struct *mm, unsigned long *table)
 	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);
+	if (!mm_alloc_pgste(mm)) {
+		/* Free 2K page table fragment of a 4K page */
+		bit = (__pa(table) & ~PAGE_MASK)/(PTRS_PER_PTE*sizeof(pte_t));
+		spin_lock_bh(&mm->context.list_lock);
+		mask = atomic_xor_bits(&page->_mapcount, 1U << bit);
+		if (mask & 3)
+			list_add(&page->lru, &mm->context.pgtable_list);
+		else
+			list_del(&page->lru);
+		spin_unlock_bh(&mm->context.list_lock);
+		if (mask != 0)
+			return;
 	}
-}
-
-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);
-	}
+	pgtable_page_dtor(page);
+	atomic_set(&page->_mapcount, -1);
+	__free_page(page);
 }
 
 void page_table_free_rcu(struct mmu_gather *tlb, unsigned long *table,
@@ -1083,34 +1025,45 @@ void page_table_free_rcu(struct mmu_gather *tlb, unsigned long *table,
 
 	mm = tlb->mm;
 	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
-	if (page_table_with_pgste(page)) {
+	if (mm_alloc_pgste(mm)) {
 		gmap_unlink(mm, table, vmaddr);
-		table = (unsigned long *) (__pa(table) | FRAG_MASK);
+		table = (unsigned long *) (__pa(table) | 3);
 		tlb_remove_table(tlb, table);
 		return;
 	}
-	bit = 1 << ((__pa(table) & ~PAGE_MASK) / (PTRS_PER_PTE*sizeof(pte_t)));
+	bit = (__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)
+	mask = atomic_xor_bits(&page->_mapcount, 0x11U << bit);
+	if (mask & 3)
 		list_add_tail(&page->lru, &mm->context.pgtable_list);
+	else
+		list_del(&page->lru);
 	spin_unlock_bh(&mm->context.list_lock);
-	table = (unsigned long *) (__pa(table) | (bit << 4));
+	table = (unsigned long *) (__pa(table) | (1U << bit));
 	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);
+	unsigned int mask = (unsigned long) _table & 3;
+	void *table = (void *)((unsigned long) _table ^ mask);
+	struct page *page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
+
+	switch (mask) {
+	case 0:		/* pmd or pud */
+		free_pages((unsigned long) table, 2);
+		break;
+	case 1:		/* lower 2K of a 4K page table */
+	case 2:		/* higher 2K of a 4K page table */
+		if (atomic_xor_bits(&page->_mapcount, mask << 4) != 0)
+			break;
+		/* fallthrough */
+	case 3:		/* 4K page table with pgstes */
+		pgtable_page_dtor(page);
+		atomic_set(&page->_mapcount, -1);
+		__free_page(page);
+		break;
+	}
 }
 
 static void tlb_remove_table_smp_sync(void *arg)
diff --git a/arch/s390/net/bpf_jit.h b/arch/s390/net/bpf_jit.h
index f6498eec9..f010c93a8 100644
--- a/arch/s390/net/bpf_jit.h
+++ b/arch/s390/net/bpf_jit.h
@@ -36,6 +36,8 @@ extern u8 sk_load_word[], sk_load_half[], sk_load_byte[];
  *	      |   BPF stack   |     |
  *	      |		      |     |
  *	      +---------------+     |
+ *	      | 8 byte skbp   |     |
+ * R15+170 -> +---------------+     |
  *	      | 8 byte hlen   |     |
  * R15+168 -> +---------------+     |
  *	      | 4 byte align  |     |
@@ -51,11 +53,12 @@ extern u8 sk_load_word[], sk_load_half[], sk_load_byte[];
  * We get 160 bytes stack space from calling function, but only use
  * 12 * 8 byte for old backchain, r15..r6, and tail_call_cnt.
  */
-#define STK_SPACE	(MAX_BPF_STACK + 8 + 4 + 4 + 160)
+#define STK_SPACE	(MAX_BPF_STACK + 8 + 8 + 4 + 4 + 160)
 #define STK_160_UNUSED	(160 - 12 * 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 */
+#define STK_OFF_SKBP	170	/* Offset of SKB pointer on stack */
 
 #define STK_OFF_R6	(160 - 11 * 8)	/* Offset of r6 on stack */
 #define STK_OFF_TCCNT	(160 - 12 * 8)	/* Offset of tail_call_cnt on stack */
diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 8d2e51658..eeda05144 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -45,7 +45,7 @@ struct bpf_jit {
 	int labels[1];		/* Labels for local jumps */
 };
 
-#define BPF_SIZE_MAX	4096	/* Max size for program */
+#define BPF_SIZE_MAX	0x7ffff	/* Max size for program (20 bit signed displ) */
 
 #define SEEN_SKB	1	/* skb access */
 #define SEEN_MEM	2	/* use mem[] for temporary storage */
@@ -53,6 +53,7 @@ struct bpf_jit {
 #define SEEN_LITERAL	8	/* code uses literals */
 #define SEEN_FUNC	16	/* calls C functions */
 #define SEEN_TAIL_CALL	32	/* code uses tail calls */
+#define SEEN_SKB_CHANGE	64	/* code changes skb data */
 #define SEEN_STACK	(SEEN_FUNC | SEEN_MEM | SEEN_SKB)
 
 /*
@@ -203,19 +204,11 @@ static inline void reg_set_seen(struct bpf_jit *jit, u32 b1)
 	_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;		\
+	u32 _disp = (u32) disp;					\
+	unsigned int __disp_h = _disp & 0xff000;		\
+	unsigned int __disp_l = _disp & 0x00fff;		\
 	_EMIT6(op1 | __disp_l, op2 | __disp_h >> 4);		\
 })
 
@@ -390,12 +383,32 @@ static void save_restore_regs(struct bpf_jit *jit, int op)
 }
 
 /*
+ * 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.
+ */
+static void emit_load_skb_data_hlen(struct bpf_jit *jit)
+{
+	/* 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));
+}
+
+/*
  * 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)
+static void bpf_jit_prologue(struct bpf_jit *jit, bool is_classic)
 {
 	if (jit->seen & SEEN_TAIL_CALL) {
 		/* xc STK_OFF_TCCNT(4,%r15),STK_OFF_TCCNT(%r15) */
@@ -429,32 +442,21 @@ static void bpf_jit_prologue(struct bpf_jit *jit)
 			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) */
+	if (jit->seen & SEEN_SKB)
+		emit_load_skb_data_hlen(jit);
+	if (jit->seen & SEEN_SKB_CHANGE)
+		/* stg %b1,ST_OFF_SKBP(%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));
+			      STK_OFF_SKBP);
+	/* Clear A (%b0) and X (%b7) registers for converted BPF programs */
+	if (is_classic) {
+		if (REG_SEEN(BPF_REG_A))
+			/* lghi %ba,0 */
+			EMIT4_IMM(0xa7090000, BPF_REG_A, 0);
+		if (REG_SEEN(BPF_REG_X))
+			/* lghi %bx,0 */
+			EMIT4_IMM(0xa7090000, BPF_REG_X, 0);
 	}
-	/* BPF compatibility: clear A (%b0) and X (%b7) registers */
-	if (REG_SEEN(BPF_REG_A))
-		/* lghi %ba,0 */
-		EMIT4_IMM(0xa7090000, BPF_REG_A, 0);
-	if (REG_SEEN(BPF_REG_X))
-		/* lghi %bx,0 */
-		EMIT4_IMM(0xa7090000, BPF_REG_X, 0);
 }
 
 /*
@@ -976,12 +978,19 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i
 		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));
+		EMIT6_DISP_LH(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);
+		if (bpf_helper_changes_skb_data((void *)func)) {
+			jit->seen |= SEEN_SKB_CHANGE;
+			/* lg %b1,ST_OFF_SKBP(%r15) */
+			EMIT6_DISP_LH(0xe3000000, 0x0004, BPF_REG_1, REG_0,
+				      REG_15, STK_OFF_SKBP);
+			emit_load_skb_data_hlen(jit);
+		}
 		break;
 	}
 	case BPF_JMP | BPF_CALL | BPF_X:
@@ -1023,7 +1032,7 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i
 				      MAX_TAIL_CALL_CNT, 0, 0x2);
 
 		/*
-		 * prog = array->prog[index];
+		 * prog = array->ptrs[index];
 		 * if (prog == NULL)
 		 *         goto out;
 		 */
@@ -1032,7 +1041,7 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp, int i
 		EMIT6_DISP_LH(0xeb000000, 0x000d, REG_1, BPF_REG_3, REG_0, 3);
 		/* lg %r1,prog(%b2,%r1) */
 		EMIT6_DISP_LH(0xe3000000, 0x0004, REG_1, BPF_REG_2,
-			      REG_1, offsetof(struct bpf_array, prog));
+			      REG_1, offsetof(struct bpf_array, ptrs));
 		/* clgij %r1,0,0x8,label0 */
 		EMIT6_PCREL_IMM_LABEL(0xec000000, 0x007d, REG_1, 0, 0, 0x8);
 
@@ -1236,7 +1245,7 @@ static int bpf_jit_prog(struct bpf_jit *jit, struct bpf_prog *fp)
 	jit->lit = jit->lit_start;
 	jit->prg = 0;
 
-	bpf_jit_prologue(jit);
+	bpf_jit_prologue(jit, bpf_prog_was_classic(fp));
 	for (i = 0; i < fp->len; i += insn_count) {
 		insn_count = bpf_jit_insn(jit, fp, i);
 		if (insn_count < 0)
diff --git a/arch/s390/numa/Makefile b/arch/s390/numa/Makefile
new file mode 100644
index 000000000..f94ecaffa
--- /dev/null
+++ b/arch/s390/numa/Makefile
@@ -0,0 +1,3 @@
+obj-y			+= numa.o
+obj-y			+= toptree.o
+obj-$(CONFIG_NUMA_EMU)	+= mode_emu.o
diff --git a/arch/s390/numa/mode_emu.c b/arch/s390/numa/mode_emu.c
new file mode 100644
index 000000000..30b2698a2
--- /dev/null
+++ b/arch/s390/numa/mode_emu.c
@@ -0,0 +1,530 @@
+/*
+ * NUMA support for s390
+ *
+ * NUMA emulation (aka fake NUMA) distributes the available memory to nodes
+ * without using real topology information about the physical memory of the
+ * machine.
+ *
+ * It distributes the available CPUs to nodes while respecting the original
+ * machine topology information. This is done by trying to avoid to separate
+ * CPUs which reside on the same book or even on the same MC.
+ *
+ * Because the current Linux scheduler code requires a stable cpu to node
+ * mapping, cores are pinned to nodes when the first CPU thread is set online.
+ *
+ * Copyright IBM Corp. 2015
+ */
+
+#define KMSG_COMPONENT "numa_emu"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/cpumask.h>
+#include <linux/memblock.h>
+#include <linux/node.h>
+#include <linux/memory.h>
+#include <linux/slab.h>
+#include <asm/smp.h>
+#include <asm/topology.h>
+#include "numa_mode.h"
+#include "toptree.h"
+
+/* Distances between the different system components */
+#define DIST_EMPTY	0
+#define DIST_CORE	1
+#define DIST_MC		2
+#define DIST_BOOK	3
+#define DIST_MAX	4
+
+/* Node distance reported to common code */
+#define EMU_NODE_DIST	10
+
+/* Node ID for free (not yet pinned) cores */
+#define NODE_ID_FREE	-1
+
+/* Different levels of toptree */
+enum toptree_level {CORE, MC, BOOK, NODE, TOPOLOGY};
+
+/* The two toptree IDs */
+enum {TOPTREE_ID_PHYS, TOPTREE_ID_NUMA};
+
+/* Number of NUMA nodes */
+static int emu_nodes = 1;
+/* NUMA stripe size */
+static unsigned long emu_size;
+
+/*
+ * Node to core pinning information updates are protected by
+ * "sched_domains_mutex".
+ */
+static struct {
+	s32 to_node_id[CONFIG_NR_CPUS];	/* Pinned core to node mapping */
+	int total;			/* Total number of pinned cores */
+	int per_node_target;		/* Cores per node without extra cores */
+	int per_node[MAX_NUMNODES];	/* Number of cores pinned to node */
+} *emu_cores;
+
+/*
+ * Pin a core to a node
+ */
+static void pin_core_to_node(int core_id, int node_id)
+{
+	if (emu_cores->to_node_id[core_id] == NODE_ID_FREE) {
+		emu_cores->per_node[node_id]++;
+		emu_cores->to_node_id[core_id] = node_id;
+		emu_cores->total++;
+	} else {
+		WARN_ON(emu_cores->to_node_id[core_id] != node_id);
+	}
+}
+
+/*
+ * Number of pinned cores of a node
+ */
+static int cores_pinned(struct toptree *node)
+{
+	return emu_cores->per_node[node->id];
+}
+
+/*
+ * ID of the node where the core is pinned (or NODE_ID_FREE)
+ */
+static int core_pinned_to_node_id(struct toptree *core)
+{
+	return emu_cores->to_node_id[core->id];
+}
+
+/*
+ * Number of cores in the tree that are not yet pinned
+ */
+static int cores_free(struct toptree *tree)
+{
+	struct toptree *core;
+	int count = 0;
+
+	toptree_for_each(core, tree, CORE) {
+		if (core_pinned_to_node_id(core) == NODE_ID_FREE)
+			count++;
+	}
+	return count;
+}
+
+/*
+ * Return node of core
+ */
+static struct toptree *core_node(struct toptree *core)
+{
+	return core->parent->parent->parent;
+}
+
+/*
+ * Return book of core
+ */
+static struct toptree *core_book(struct toptree *core)
+{
+	return core->parent->parent;
+}
+
+/*
+ * Return mc of core
+ */
+static struct toptree *core_mc(struct toptree *core)
+{
+	return core->parent;
+}
+
+/*
+ * Distance between two cores
+ */
+static int dist_core_to_core(struct toptree *core1, struct toptree *core2)
+{
+	if (core_book(core1)->id != core_book(core2)->id)
+		return DIST_BOOK;
+	if (core_mc(core1)->id != core_mc(core2)->id)
+		return DIST_MC;
+	/* Same core or sibling on same MC */
+	return DIST_CORE;
+}
+
+/*
+ * Distance of a node to a core
+ */
+static int dist_node_to_core(struct toptree *node, struct toptree *core)
+{
+	struct toptree *core_node;
+	int dist_min = DIST_MAX;
+
+	toptree_for_each(core_node, node, CORE)
+		dist_min = min(dist_min, dist_core_to_core(core_node, core));
+	return dist_min == DIST_MAX ? DIST_EMPTY : dist_min;
+}
+
+/*
+ * Unify will delete empty nodes, therefore recreate nodes.
+ */
+static void toptree_unify_tree(struct toptree *tree)
+{
+	int nid;
+
+	toptree_unify(tree);
+	for (nid = 0; nid < emu_nodes; nid++)
+		toptree_get_child(tree, nid);
+}
+
+/*
+ * Find the best/nearest node for a given core and ensure that no node
+ * gets more than "emu_cores->per_node_target + extra" cores.
+ */
+static struct toptree *node_for_core(struct toptree *numa, struct toptree *core,
+				     int extra)
+{
+	struct toptree *node, *node_best = NULL;
+	int dist_cur, dist_best, cores_target;
+
+	cores_target = emu_cores->per_node_target + extra;
+	dist_best = DIST_MAX;
+	node_best = NULL;
+	toptree_for_each(node, numa, NODE) {
+		/* Already pinned cores must use their nodes */
+		if (core_pinned_to_node_id(core) == node->id) {
+			node_best = node;
+			break;
+		}
+		/* Skip nodes that already have enough cores */
+		if (cores_pinned(node) >= cores_target)
+			continue;
+		dist_cur = dist_node_to_core(node, core);
+		if (dist_cur < dist_best) {
+			dist_best = dist_cur;
+			node_best = node;
+		}
+	}
+	return node_best;
+}
+
+/*
+ * Find the best node for each core with respect to "extra" core count
+ */
+static void toptree_to_numa_single(struct toptree *numa, struct toptree *phys,
+				   int extra)
+{
+	struct toptree *node, *core, *tmp;
+
+	toptree_for_each_safe(core, tmp, phys, CORE) {
+		node = node_for_core(numa, core, extra);
+		if (!node)
+			return;
+		toptree_move(core, node);
+		pin_core_to_node(core->id, node->id);
+	}
+}
+
+/*
+ * Move structures of given level to specified NUMA node
+ */
+static void move_level_to_numa_node(struct toptree *node, struct toptree *phys,
+				    enum toptree_level level, bool perfect)
+{
+	int cores_free, cores_target = emu_cores->per_node_target;
+	struct toptree *cur, *tmp;
+
+	toptree_for_each_safe(cur, tmp, phys, level) {
+		cores_free = cores_target - toptree_count(node, CORE);
+		if (perfect) {
+			if (cores_free == toptree_count(cur, CORE))
+				toptree_move(cur, node);
+		} else {
+			if (cores_free >= toptree_count(cur, CORE))
+				toptree_move(cur, node);
+		}
+	}
+}
+
+/*
+ * Move structures of a given level to NUMA nodes. If "perfect" is specified
+ * move only perfectly fitting structures. Otherwise move also smaller
+ * than needed structures.
+ */
+static void move_level_to_numa(struct toptree *numa, struct toptree *phys,
+			       enum toptree_level level, bool perfect)
+{
+	struct toptree *node;
+
+	toptree_for_each(node, numa, NODE)
+		move_level_to_numa_node(node, phys, level, perfect);
+}
+
+/*
+ * For the first run try to move the big structures
+ */
+static void toptree_to_numa_first(struct toptree *numa, struct toptree *phys)
+{
+	struct toptree *core;
+
+	/* Always try to move perfectly fitting structures first */
+	move_level_to_numa(numa, phys, BOOK, true);
+	move_level_to_numa(numa, phys, BOOK, false);
+	move_level_to_numa(numa, phys, MC, true);
+	move_level_to_numa(numa, phys, MC, false);
+	/* Now pin all the moved cores */
+	toptree_for_each(core, numa, CORE)
+		pin_core_to_node(core->id, core_node(core)->id);
+}
+
+/*
+ * Allocate new topology and create required nodes
+ */
+static struct toptree *toptree_new(int id, int nodes)
+{
+	struct toptree *tree;
+	int nid;
+
+	tree = toptree_alloc(TOPOLOGY, id);
+	if (!tree)
+		goto fail;
+	for (nid = 0; nid < nodes; nid++) {
+		if (!toptree_get_child(tree, nid))
+			goto fail;
+	}
+	return tree;
+fail:
+	panic("NUMA emulation could not allocate topology");
+}
+
+/*
+ * Allocate and initialize core to node mapping
+ */
+static void create_core_to_node_map(void)
+{
+	int i;
+
+	emu_cores = kzalloc(sizeof(*emu_cores), GFP_KERNEL);
+	if (emu_cores == NULL)
+		panic("Could not allocate cores to node memory");
+	for (i = 0; i < ARRAY_SIZE(emu_cores->to_node_id); i++)
+		emu_cores->to_node_id[i] = NODE_ID_FREE;
+}
+
+/*
+ * Move cores from physical topology into NUMA target topology
+ * and try to keep as much of the physical topology as possible.
+ */
+static struct toptree *toptree_to_numa(struct toptree *phys)
+{
+	static int first = 1;
+	struct toptree *numa;
+	int cores_total;
+
+	cores_total = emu_cores->total + cores_free(phys);
+	emu_cores->per_node_target = cores_total / emu_nodes;
+	numa = toptree_new(TOPTREE_ID_NUMA, emu_nodes);
+	if (first) {
+		toptree_to_numa_first(numa, phys);
+		first = 0;
+	}
+	toptree_to_numa_single(numa, phys, 0);
+	toptree_to_numa_single(numa, phys, 1);
+	toptree_unify_tree(numa);
+
+	WARN_ON(cpumask_weight(&phys->mask));
+	return numa;
+}
+
+/*
+ * Create a toptree out of the physical topology that we got from the hypervisor
+ */
+static struct toptree *toptree_from_topology(void)
+{
+	struct toptree *phys, *node, *book, *mc, *core;
+	struct cpu_topology_s390 *top;
+	int cpu;
+
+	phys = toptree_new(TOPTREE_ID_PHYS, 1);
+
+	for_each_online_cpu(cpu) {
+		top = &per_cpu(cpu_topology, cpu);
+		node = toptree_get_child(phys, 0);
+		book = toptree_get_child(node, top->book_id);
+		mc = toptree_get_child(book, top->socket_id);
+		core = toptree_get_child(mc, top->core_id);
+		if (!book || !mc || !core)
+			panic("NUMA emulation could not allocate memory");
+		cpumask_set_cpu(cpu, &core->mask);
+		toptree_update_mask(mc);
+	}
+	return phys;
+}
+
+/*
+ * Add toptree core to topology and create correct CPU masks
+ */
+static void topology_add_core(struct toptree *core)
+{
+	struct cpu_topology_s390 *top;
+	int cpu;
+
+	for_each_cpu(cpu, &core->mask) {
+		top = &per_cpu(cpu_topology, cpu);
+		cpumask_copy(&top->thread_mask, &core->mask);
+		cpumask_copy(&top->core_mask, &core_mc(core)->mask);
+		cpumask_copy(&top->book_mask, &core_book(core)->mask);
+		cpumask_set_cpu(cpu, &node_to_cpumask_map[core_node(core)->id]);
+		top->node_id = core_node(core)->id;
+	}
+}
+
+/*
+ * Apply toptree to topology and create CPU masks
+ */
+static void toptree_to_topology(struct toptree *numa)
+{
+	struct toptree *core;
+	int i;
+
+	/* Clear all node masks */
+	for (i = 0; i < MAX_NUMNODES; i++)
+		cpumask_clear(&node_to_cpumask_map[i]);
+
+	/* Rebuild all masks */
+	toptree_for_each(core, numa, CORE)
+		topology_add_core(core);
+}
+
+/*
+ * Show the node to core mapping
+ */
+static void print_node_to_core_map(void)
+{
+	int nid, cid;
+
+	if (!numa_debug_enabled)
+		return;
+	printk(KERN_DEBUG "NUMA node to core mapping\n");
+	for (nid = 0; nid < emu_nodes; nid++) {
+		printk(KERN_DEBUG "  node %3d: ", nid);
+		for (cid = 0; cid < ARRAY_SIZE(emu_cores->to_node_id); cid++) {
+			if (emu_cores->to_node_id[cid] == nid)
+				printk(KERN_CONT "%d ", cid);
+		}
+		printk(KERN_CONT "\n");
+	}
+}
+
+/*
+ * Transfer physical topology into a NUMA topology and modify CPU masks
+ * according to the NUMA topology.
+ *
+ * Must be called with "sched_domains_mutex" lock held.
+ */
+static void emu_update_cpu_topology(void)
+{
+	struct toptree *phys, *numa;
+
+	if (emu_cores == NULL)
+		create_core_to_node_map();
+	phys = toptree_from_topology();
+	numa = toptree_to_numa(phys);
+	toptree_free(phys);
+	toptree_to_topology(numa);
+	toptree_free(numa);
+	print_node_to_core_map();
+}
+
+/*
+ * If emu_size is not set, use CONFIG_EMU_SIZE. Then round to minimum
+ * alignment (needed for memory hotplug).
+ */
+static unsigned long emu_setup_size_adjust(unsigned long size)
+{
+	size = size ? : CONFIG_EMU_SIZE;
+	size = roundup(size, memory_block_size_bytes());
+	return size;
+}
+
+/*
+ * If we have not enough memory for the specified nodes, reduce the node count.
+ */
+static int emu_setup_nodes_adjust(int nodes)
+{
+	int nodes_max;
+
+	nodes_max = memblock.memory.total_size / emu_size;
+	nodes_max = max(nodes_max, 1);
+	if (nodes_max >= nodes)
+		return nodes;
+	pr_warn("Not enough memory for %d nodes, reducing node count\n", nodes);
+	return nodes_max;
+}
+
+/*
+ * Early emu setup
+ */
+static void emu_setup(void)
+{
+	emu_size = emu_setup_size_adjust(emu_size);
+	emu_nodes = emu_setup_nodes_adjust(emu_nodes);
+	pr_info("Creating %d nodes with memory stripe size %ld MB\n",
+		emu_nodes, emu_size >> 20);
+}
+
+/*
+ * Return node id for given page number
+ */
+static int emu_pfn_to_nid(unsigned long pfn)
+{
+	return (pfn / (emu_size >> PAGE_SHIFT)) % emu_nodes;
+}
+
+/*
+ * Return stripe size
+ */
+static unsigned long emu_align(void)
+{
+	return emu_size;
+}
+
+/*
+ * Return distance between two nodes
+ */
+static int emu_distance(int node1, int node2)
+{
+	return (node1 != node2) * EMU_NODE_DIST;
+}
+
+/*
+ * Define callbacks for generic s390 NUMA infrastructure
+ */
+const struct numa_mode numa_mode_emu = {
+	.name = "emu",
+	.setup = emu_setup,
+	.update_cpu_topology = emu_update_cpu_topology,
+	.__pfn_to_nid = emu_pfn_to_nid,
+	.align = emu_align,
+	.distance = emu_distance,
+};
+
+/*
+ * Kernel parameter: emu_nodes=<n>
+ */
+static int __init early_parse_emu_nodes(char *p)
+{
+	int count;
+
+	if (kstrtoint(p, 0, &count) != 0 || count <= 0)
+		return 0;
+	if (count <= 0)
+		return 0;
+	emu_nodes = min(count, MAX_NUMNODES);
+	return 0;
+}
+early_param("emu_nodes", early_parse_emu_nodes);
+
+/*
+ * Kernel parameter: emu_size=[<n>[k|M|G|T]]
+ */
+static int __init early_parse_emu_size(char *p)
+{
+	emu_size = memparse(p, NULL);
+	return 0;
+}
+early_param("emu_size", early_parse_emu_size);
diff --git a/arch/s390/numa/numa.c b/arch/s390/numa/numa.c
new file mode 100644
index 000000000..43f32ce60
--- /dev/null
+++ b/arch/s390/numa/numa.c
@@ -0,0 +1,184 @@
+/*
+ * NUMA support for s390
+ *
+ * Implement NUMA core code.
+ *
+ * Copyright IBM Corp. 2015
+ */
+
+#define KMSG_COMPONENT "numa"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/mmzone.h>
+#include <linux/cpumask.h>
+#include <linux/bootmem.h>
+#include <linux/memblock.h>
+#include <linux/slab.h>
+#include <linux/node.h>
+
+#include <asm/numa.h>
+#include "numa_mode.h"
+
+pg_data_t *node_data[MAX_NUMNODES];
+EXPORT_SYMBOL(node_data);
+
+cpumask_t node_to_cpumask_map[MAX_NUMNODES];
+EXPORT_SYMBOL(node_to_cpumask_map);
+
+const struct numa_mode numa_mode_plain = {
+	.name = "plain",
+};
+
+static const struct numa_mode *mode = &numa_mode_plain;
+
+int numa_pfn_to_nid(unsigned long pfn)
+{
+	return mode->__pfn_to_nid ? mode->__pfn_to_nid(pfn) : 0;
+}
+
+void numa_update_cpu_topology(void)
+{
+	if (mode->update_cpu_topology)
+		mode->update_cpu_topology();
+}
+
+int __node_distance(int a, int b)
+{
+	return mode->distance ? mode->distance(a, b) : 0;
+}
+
+int numa_debug_enabled;
+
+/*
+ * alloc_node_data() - Allocate node data
+ */
+static __init pg_data_t *alloc_node_data(void)
+{
+	pg_data_t *res;
+
+	res = (pg_data_t *) memblock_alloc(sizeof(pg_data_t), 1);
+	if (!res)
+		panic("Could not allocate memory for node data!\n");
+	memset(res, 0, sizeof(pg_data_t));
+	return res;
+}
+
+/*
+ * numa_setup_memory() - Assign bootmem to nodes
+ *
+ * The memory is first added to memblock without any respect to nodes.
+ * This is fixed before remaining memblock memory is handed over to the
+ * buddy allocator.
+ * An important side effect is that large bootmem allocations might easily
+ * cross node boundaries, which can be needed for large allocations with
+ * smaller memory stripes in each node (i.e. when using NUMA emulation).
+ *
+ * Memory defines nodes:
+ * Therefore this routine also sets the nodes online with memory.
+ */
+static void __init numa_setup_memory(void)
+{
+	unsigned long cur_base, align, end_of_dram;
+	int nid = 0;
+
+	end_of_dram = memblock_end_of_DRAM();
+	align = mode->align ? mode->align() : ULONG_MAX;
+
+	/*
+	 * Step through all available memory and assign it to the nodes
+	 * indicated by the mode implementation.
+	 * All nodes which are seen here will be set online.
+	 */
+	cur_base = 0;
+	do {
+		nid = numa_pfn_to_nid(PFN_DOWN(cur_base));
+		node_set_online(nid);
+		memblock_set_node(cur_base, align, &memblock.memory, nid);
+		cur_base += align;
+	} while (cur_base < end_of_dram);
+
+	/* Allocate and fill out node_data */
+	for (nid = 0; nid < MAX_NUMNODES; nid++)
+		NODE_DATA(nid) = alloc_node_data();
+
+	for_each_online_node(nid) {
+		unsigned long start_pfn, end_pfn;
+		unsigned long t_start, t_end;
+		int i;
+
+		start_pfn = ULONG_MAX;
+		end_pfn = 0;
+		for_each_mem_pfn_range(i, nid, &t_start, &t_end, NULL) {
+			if (t_start < start_pfn)
+				start_pfn = t_start;
+			if (t_end > end_pfn)
+				end_pfn = t_end;
+		}
+		NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn;
+		NODE_DATA(nid)->node_id = nid;
+	}
+}
+
+/*
+ * numa_setup() - Earliest initialization
+ *
+ * Assign the mode and call the mode's setup routine.
+ */
+void __init numa_setup(void)
+{
+	pr_info("NUMA mode: %s\n", mode->name);
+	if (mode->setup)
+		mode->setup();
+	numa_setup_memory();
+	memblock_dump_all();
+}
+
+
+/*
+ * numa_init_early() - Initialization initcall
+ *
+ * This runs when only one CPU is online and before the first
+ * topology update is called for by the scheduler.
+ */
+static int __init numa_init_early(void)
+{
+	/* Attach all possible CPUs to node 0 for now. */
+	cpumask_copy(&node_to_cpumask_map[0], cpu_possible_mask);
+	return 0;
+}
+early_initcall(numa_init_early);
+
+/*
+ * numa_init_late() - Initialization initcall
+ *
+ * Register NUMA nodes.
+ */
+static int __init numa_init_late(void)
+{
+	int nid;
+
+	for_each_online_node(nid)
+		register_one_node(nid);
+	return 0;
+}
+device_initcall(numa_init_late);
+
+static int __init parse_debug(char *parm)
+{
+	numa_debug_enabled = 1;
+	return 0;
+}
+early_param("numa_debug", parse_debug);
+
+static int __init parse_numa(char *parm)
+{
+	if (strcmp(parm, numa_mode_plain.name) == 0)
+		mode = &numa_mode_plain;
+#ifdef CONFIG_NUMA_EMU
+	if (strcmp(parm, numa_mode_emu.name) == 0)
+		mode = &numa_mode_emu;
+#endif
+	return 0;
+}
+early_param("numa", parse_numa);
diff --git a/arch/s390/numa/numa_mode.h b/arch/s390/numa/numa_mode.h
new file mode 100644
index 000000000..08953b0b1
--- /dev/null
+++ b/arch/s390/numa/numa_mode.h
@@ -0,0 +1,24 @@
+/*
+ * NUMA support for s390
+ *
+ * Define declarations used for communication between NUMA mode
+ * implementations and NUMA core functionality.
+ *
+ * Copyright IBM Corp. 2015
+ */
+#ifndef __S390_NUMA_MODE_H
+#define __S390_NUMA_MODE_H
+
+struct numa_mode {
+	char *name;				/* Name of mode */
+	void (*setup)(void);			/* Initizalize mode */
+	void (*update_cpu_topology)(void);	/* Called by topology code */
+	int (*__pfn_to_nid)(unsigned long pfn);	/* PFN to node ID */
+	unsigned long (*align)(void);		/* Minimum node alignment */
+	int (*distance)(int a, int b);		/* Distance between two nodes */
+};
+
+extern const struct numa_mode numa_mode_plain;
+extern const struct numa_mode numa_mode_emu;
+
+#endif /* __S390_NUMA_MODE_H */
diff --git a/arch/s390/numa/toptree.c b/arch/s390/numa/toptree.c
new file mode 100644
index 000000000..902d350d8
--- /dev/null
+++ b/arch/s390/numa/toptree.c
@@ -0,0 +1,342 @@
+/*
+ * NUMA support for s390
+ *
+ * A tree structure used for machine topology mangling
+ *
+ * Copyright IBM Corp. 2015
+ */
+
+#include <linux/kernel.h>
+#include <linux/cpumask.h>
+#include <linux/list.h>
+#include <linux/list_sort.h>
+#include <linux/slab.h>
+#include <asm/numa.h>
+
+#include "toptree.h"
+
+/**
+ * toptree_alloc - Allocate and initialize a new tree node.
+ * @level: The node's vertical level; level 0 contains the leaves.
+ * @id: ID number, explicitly not unique beyond scope of node's siblings
+ *
+ * Allocate a new tree node and initialize it.
+ *
+ * RETURNS:
+ * Pointer to the new tree node or NULL on error
+ */
+struct toptree *toptree_alloc(int level, int id)
+{
+	struct toptree *res = kzalloc(sizeof(struct toptree), GFP_KERNEL);
+
+	if (!res)
+		return res;
+
+	INIT_LIST_HEAD(&res->children);
+	INIT_LIST_HEAD(&res->sibling);
+	cpumask_clear(&res->mask);
+	res->level = level;
+	res->id = id;
+	return res;
+}
+
+/**
+ * toptree_remove - Remove a tree node from a tree
+ * @cand: Pointer to the node to remove
+ *
+ * The node is detached from its parent node. The parent node's
+ * masks will be updated to reflect the loss of the child.
+ */
+static void toptree_remove(struct toptree *cand)
+{
+	struct toptree *oldparent;
+
+	list_del_init(&cand->sibling);
+	oldparent = cand->parent;
+	cand->parent = NULL;
+	toptree_update_mask(oldparent);
+}
+
+/**
+ * toptree_free - discard a tree node
+ * @cand: Pointer to the tree node to discard
+ *
+ * Checks if @cand is attached to a parent node. Detaches it
+ * cleanly using toptree_remove. Possible children are freed
+ * recursively. In the end @cand itself is freed.
+ */
+void toptree_free(struct toptree *cand)
+{
+	struct toptree *child, *tmp;
+
+	if (cand->parent)
+		toptree_remove(cand);
+	toptree_for_each_child_safe(child, tmp, cand)
+		toptree_free(child);
+	kfree(cand);
+}
+
+/**
+ * toptree_update_mask - Update node bitmasks
+ * @cand: Pointer to a tree node
+ *
+ * The node's cpumask will be updated by combining all children's
+ * masks. Then toptree_update_mask is called recursively for the
+ * parent if applicable.
+ *
+ * NOTE:
+ * This must not be called on leaves. If called on a leaf, its
+ * CPU mask is cleared and lost.
+ */
+void toptree_update_mask(struct toptree *cand)
+{
+	struct toptree *child;
+
+	cpumask_clear(&cand->mask);
+	list_for_each_entry(child, &cand->children, sibling)
+		cpumask_or(&cand->mask, &cand->mask, &child->mask);
+	if (cand->parent)
+		toptree_update_mask(cand->parent);
+}
+
+/**
+ * toptree_insert - Insert a tree node into tree
+ * @cand: Pointer to the node to insert
+ * @target: Pointer to the node to which @cand will added as a child
+ *
+ * Insert a tree node into a tree. Masks will be updated automatically.
+ *
+ * RETURNS:
+ * 0 on success, -1 if NULL is passed as argument or the node levels
+ * don't fit.
+ */
+static int toptree_insert(struct toptree *cand, struct toptree *target)
+{
+	if (!cand || !target)
+		return -1;
+	if (target->level != (cand->level + 1))
+		return -1;
+	list_add_tail(&cand->sibling, &target->children);
+	cand->parent = target;
+	toptree_update_mask(target);
+	return 0;
+}
+
+/**
+ * toptree_move_children - Move all child nodes of a node to a new place
+ * @cand: Pointer to the node whose children are to be moved
+ * @target: Pointer to the node to which @cand's children will be attached
+ *
+ * Take all child nodes of @cand and move them using toptree_move.
+ */
+static void toptree_move_children(struct toptree *cand, struct toptree *target)
+{
+	struct toptree *child, *tmp;
+
+	toptree_for_each_child_safe(child, tmp, cand)
+		toptree_move(child, target);
+}
+
+/**
+ * toptree_unify - Merge children with same ID
+ * @cand: Pointer to node whose direct children should be made unique
+ *
+ * When mangling the tree it is possible that a node has two or more children
+ * which have the same ID. This routine merges these children into one and
+ * moves all children of the merged nodes into the unified node.
+ */
+void toptree_unify(struct toptree *cand)
+{
+	struct toptree *child, *tmp, *cand_copy;
+
+	/* Threads cannot be split, cores are not split */
+	if (cand->level < 2)
+		return;
+
+	cand_copy = toptree_alloc(cand->level, 0);
+	toptree_for_each_child_safe(child, tmp, cand) {
+		struct toptree *tmpchild;
+
+		if (!cpumask_empty(&child->mask)) {
+			tmpchild = toptree_get_child(cand_copy, child->id);
+			toptree_move_children(child, tmpchild);
+		}
+		toptree_free(child);
+	}
+	toptree_move_children(cand_copy, cand);
+	toptree_free(cand_copy);
+
+	toptree_for_each_child(child, cand)
+		toptree_unify(child);
+}
+
+/**
+ * toptree_move - Move a node to another context
+ * @cand: Pointer to the node to move
+ * @target: Pointer to the node where @cand should go
+ *
+ * In the easiest case @cand is exactly on the level below @target
+ * and will be immediately moved to the target.
+ *
+ * If @target's level is not the direct parent level of @cand,
+ * nodes for the missing levels are created and put between
+ * @cand and @target. The "stacking" nodes' IDs are taken from
+ * @cand's parents.
+ *
+ * After this it is likely to have redundant nodes in the tree
+ * which are addressed by means of toptree_unify.
+ */
+void toptree_move(struct toptree *cand, struct toptree *target)
+{
+	struct toptree *stack_target, *real_insert_point, *ptr, *tmp;
+
+	if (cand->level + 1 == target->level) {
+		toptree_remove(cand);
+		toptree_insert(cand, target);
+		return;
+	}
+
+	real_insert_point = NULL;
+	ptr = cand;
+	stack_target = NULL;
+
+	do {
+		tmp = stack_target;
+		stack_target = toptree_alloc(ptr->level + 1,
+					     ptr->parent->id);
+		toptree_insert(tmp, stack_target);
+		if (!real_insert_point)
+			real_insert_point = stack_target;
+		ptr = ptr->parent;
+	} while (stack_target->level < (target->level - 1));
+
+	toptree_remove(cand);
+	toptree_insert(cand, real_insert_point);
+	toptree_insert(stack_target, target);
+}
+
+/**
+ * toptree_get_child - Access a tree node's child by its ID
+ * @cand: Pointer to tree node whose child is to access
+ * @id: The desired child's ID
+ *
+ * @cand's children are searched for a child with matching ID.
+ * If no match can be found, a new child with the desired ID
+ * is created and returned.
+ */
+struct toptree *toptree_get_child(struct toptree *cand, int id)
+{
+	struct toptree *child;
+
+	toptree_for_each_child(child, cand)
+		if (child->id == id)
+			return child;
+	child = toptree_alloc(cand->level-1, id);
+	toptree_insert(child, cand);
+	return child;
+}
+
+/**
+ * toptree_first - Find the first descendant on specified level
+ * @context: Pointer to tree node whose descendants are to be used
+ * @level: The level of interest
+ *
+ * RETURNS:
+ * @context's first descendant on the specified level, or NULL
+ * if there is no matching descendant
+ */
+struct toptree *toptree_first(struct toptree *context, int level)
+{
+	struct toptree *child, *tmp;
+
+	if (context->level == level)
+		return context;
+
+	if (!list_empty(&context->children)) {
+		list_for_each_entry(child, &context->children, sibling) {
+			tmp = toptree_first(child, level);
+			if (tmp)
+				return tmp;
+		}
+	}
+	return NULL;
+}
+
+/**
+ * toptree_next_sibling - Return next sibling
+ * @cur: Pointer to a tree node
+ *
+ * RETURNS:
+ * If @cur has a parent and is not the last in the parent's children list,
+ * the next sibling is returned. Or NULL when there are no siblings left.
+ */
+static struct toptree *toptree_next_sibling(struct toptree *cur)
+{
+	if (cur->parent == NULL)
+		return NULL;
+
+	if (cur == list_last_entry(&cur->parent->children,
+				   struct toptree, sibling))
+		return NULL;
+	return (struct toptree *) list_next_entry(cur, sibling);
+}
+
+/**
+ * toptree_next - Tree traversal function
+ * @cur: Pointer to current element
+ * @context: Pointer to the root node of the tree or subtree to
+ * be traversed.
+ * @level: The level of interest.
+ *
+ * RETURNS:
+ * Pointer to the next node on level @level
+ * or NULL when there is no next node.
+ */
+struct toptree *toptree_next(struct toptree *cur, struct toptree *context,
+			     int level)
+{
+	struct toptree *cur_context, *tmp;
+
+	if (!cur)
+		return NULL;
+
+	if (context->level == level)
+		return NULL;
+
+	tmp = toptree_next_sibling(cur);
+	if (tmp != NULL)
+		return tmp;
+
+	cur_context = cur;
+	while (cur_context->level < context->level - 1) {
+		/* Step up */
+		cur_context = cur_context->parent;
+		/* Step aside */
+		tmp = toptree_next_sibling(cur_context);
+		if (tmp != NULL) {
+			/* Step down */
+			tmp = toptree_first(tmp, level);
+			if (tmp != NULL)
+				return tmp;
+		}
+	}
+	return NULL;
+}
+
+/**
+ * toptree_count - Count descendants on specified level
+ * @context: Pointer to node whose descendants are to be considered
+ * @level: Only descendants on the specified level will be counted
+ *
+ * RETURNS:
+ * Number of descendants on the specified level
+ */
+int toptree_count(struct toptree *context, int level)
+{
+	struct toptree *cur;
+	int cnt = 0;
+
+	toptree_for_each(cur, context, level)
+		cnt++;
+	return cnt;
+}
diff --git a/arch/s390/numa/toptree.h b/arch/s390/numa/toptree.h
new file mode 100644
index 000000000..bdf502027
--- /dev/null
+++ b/arch/s390/numa/toptree.h
@@ -0,0 +1,60 @@
+/*
+ * NUMA support for s390
+ *
+ * A tree structure used for machine topology mangling
+ *
+ * Copyright IBM Corp. 2015
+ */
+#ifndef S390_TOPTREE_H
+#define S390_TOPTREE_H
+
+#include <linux/cpumask.h>
+#include <linux/list.h>
+
+struct toptree {
+	int level;
+	int id;
+	cpumask_t mask;
+	struct toptree *parent;
+	struct list_head sibling;
+	struct list_head children;
+};
+
+struct toptree *toptree_alloc(int level, int id);
+void toptree_free(struct toptree *cand);
+void toptree_update_mask(struct toptree *cand);
+void toptree_unify(struct toptree *cand);
+struct toptree *toptree_get_child(struct toptree *cand, int id);
+void toptree_move(struct toptree *cand, struct toptree *target);
+int toptree_count(struct toptree *context, int level);
+
+struct toptree *toptree_first(struct toptree *context, int level);
+struct toptree *toptree_next(struct toptree *cur, struct toptree *context,
+			     int level);
+
+#define toptree_for_each_child(child, ptree)				\
+	list_for_each_entry(child,  &ptree->children, sibling)
+
+#define toptree_for_each_child_safe(child, ptmp, ptree)			\
+	list_for_each_entry_safe(child, ptmp, &ptree->children, sibling)
+
+#define toptree_is_last(ptree)					\
+	((ptree->parent == NULL) ||				\
+	 (ptree->parent->children.prev == &ptree->sibling))
+
+#define toptree_for_each(ptree, cont, ttype)		\
+	for (ptree = toptree_first(cont, ttype);	\
+	     ptree != NULL;				\
+	     ptree = toptree_next(ptree, cont, ttype))
+
+#define toptree_for_each_safe(ptree, tmp, cont, ttype)		\
+	for (ptree = toptree_first(cont, ttype),		\
+		     tmp = toptree_next(ptree, cont, ttype);	\
+	     ptree != NULL;					\
+	     ptree = tmp,					\
+		     tmp = toptree_next(ptree, cont, ttype))
+
+#define toptree_for_each_sibling(ptree, start)			\
+	toptree_for_each(ptree, start->parent, start->level)
+
+#endif /* S390_TOPTREE_H */
diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
index 598f023cf..7ef12a3ac 100644
--- a/arch/s390/pci/pci.c
+++ b/arch/s390/pci/pci.c
@@ -76,11 +76,6 @@ 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;
@@ -269,7 +264,7 @@ void __iomem *pci_iomap_range(struct pci_dev *pdev,
 			      unsigned long offset,
 			      unsigned long max)
 {
-	struct zpci_dev *zdev =	get_zdev(pdev);
+	struct zpci_dev *zdev =	to_zpci(pdev);
 	u64 addr;
 	int idx;
 
@@ -385,7 +380,7 @@ static void zpci_irq_handler(struct airq_struct *airq)
 
 int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
 {
-	struct zpci_dev *zdev = get_zdev(pdev);
+	struct zpci_dev *zdev = to_zpci(pdev);
 	unsigned int hwirq, msi_vecs;
 	unsigned long aisb;
 	struct msi_desc *msi;
@@ -414,7 +409,7 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
 
 	/* Request MSI interrupts */
 	hwirq = 0;
-	list_for_each_entry(msi, &pdev->msi_list, list) {
+	for_each_pci_msi_entry(msi, pdev) {
 		rc = -EIO;
 		irq = irq_alloc_desc(0);	/* Alloc irq on node 0 */
 		if (irq < 0)
@@ -440,7 +435,7 @@ int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
 	return (msi_vecs == nvec) ? 0 : msi_vecs;
 
 out_msi:
-	list_for_each_entry(msi, &pdev->msi_list, list) {
+	for_each_pci_msi_entry(msi, pdev) {
 		if (hwirq-- == 0)
 			break;
 		irq_set_msi_desc(msi->irq, NULL);
@@ -460,7 +455,7 @@ out:
 
 void arch_teardown_msi_irqs(struct pci_dev *pdev)
 {
-	struct zpci_dev *zdev = get_zdev(pdev);
+	struct zpci_dev *zdev = to_zpci(pdev);
 	struct msi_desc *msi;
 	int rc;
 
@@ -470,7 +465,7 @@ void arch_teardown_msi_irqs(struct pci_dev *pdev)
 		return;
 
 	/* Release MSI interrupts */
-	list_for_each_entry(msi, &pdev->msi_list, list) {
+	for_each_pci_msi_entry(msi, pdev) {
 		if (msi->msi_attrib.is_msix)
 			__pci_msix_desc_mask_irq(msi, 1);
 		else
@@ -637,7 +632,7 @@ 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)
+		if (!zdev->bars[i].size || !zdev->bars[i].res)
 			continue;
 
 		zpci_free_iomap(zdev, zdev->bars[i].map_idx);
@@ -648,7 +643,7 @@ static void zpci_cleanup_bus_resources(struct zpci_dev *zdev)
 
 int pcibios_add_device(struct pci_dev *pdev)
 {
-	struct zpci_dev *zdev = get_zdev(pdev);
+	struct zpci_dev *zdev = to_zpci(pdev);
 	struct resource *res;
 	int i;
 
@@ -673,7 +668,7 @@ void pcibios_release_device(struct pci_dev *pdev)
 
 int pcibios_enable_device(struct pci_dev *pdev, int mask)
 {
-	struct zpci_dev *zdev = get_zdev(pdev);
+	struct zpci_dev *zdev = to_zpci(pdev);
 
 	zdev->pdev = pdev;
 	zpci_debug_init_device(zdev);
@@ -684,7 +679,7 @@ int pcibios_enable_device(struct pci_dev *pdev, int mask)
 
 void pcibios_disable_device(struct pci_dev *pdev)
 {
-	struct zpci_dev *zdev = get_zdev(pdev);
+	struct zpci_dev *zdev = to_zpci(pdev);
 
 	zpci_fmb_disable_device(zdev);
 	zpci_debug_exit_device(zdev);
@@ -695,7 +690,7 @@ void pcibios_disable_device(struct pci_dev *pdev)
 static int zpci_restore(struct device *dev)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
-	struct zpci_dev *zdev = get_zdev(pdev);
+	struct zpci_dev *zdev = to_zpci(pdev);
 	int ret = 0;
 
 	if (zdev->state != ZPCI_FN_STATE_ONLINE)
@@ -717,7 +712,7 @@ out:
 static int zpci_freeze(struct device *dev)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
-	struct zpci_dev *zdev = get_zdev(pdev);
+	struct zpci_dev *zdev = to_zpci(pdev);
 
 	if (zdev->state != ZPCI_FN_STATE_ONLINE)
 		return 0;
@@ -777,17 +772,22 @@ static int zpci_scan_bus(struct zpci_dev *zdev)
 
 	ret = zpci_setup_bus_resources(zdev, &resources);
 	if (ret)
-		return ret;
+		goto error;
 
 	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;
+		ret = -EIO;
+		goto error;
 	}
 	zdev->bus->max_bus_speed = zdev->max_bus_speed;
 	pci_bus_add_devices(zdev->bus);
 	return 0;
+
+error:
+	zpci_cleanup_bus_resources(zdev);
+	pci_free_resource_list(&resources);
+	return ret;
 }
 
 int zpci_enable_device(struct zpci_dev *zdev)
diff --git a/arch/s390/pci/pci_dma.c b/arch/s390/pci/pci_dma.c
index 6fd8d5836..37505b8b4 100644
--- a/arch/s390/pci/pci_dma.c
+++ b/arch/s390/pci/pci_dma.c
@@ -262,22 +262,12 @@ 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));
+	struct zpci_dev *zdev = to_zpci(to_pci_dev(dev));
 	unsigned long nr_pages, iommu_page_index;
 	unsigned long pa = page_to_phys(page) + offset;
 	int flags = ZPCI_PTE_VALID;
@@ -316,7 +306,7 @@ 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));
+	struct zpci_dev *zdev = to_zpci(to_pci_dev(dev));
 	unsigned long iommu_page_index;
 	int npages;
 
@@ -337,7 +327,7 @@ 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 zpci_dev *zdev = to_zpci(to_pci_dev(dev));
 	struct page *page;
 	unsigned long pa;
 	dma_addr_t map;
@@ -367,7 +357,7 @@ 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));
+	struct zpci_dev *zdev = to_zpci(to_pci_dev(dev));
 
 	size = PAGE_ALIGN(size);
 	atomic64_sub(size / PAGE_SIZE, &zdev->allocated_pages);
diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c
index ed2394dd1..369a3e05d 100644
--- a/arch/s390/pci/pci_event.c
+++ b/arch/s390/pci/pci_event.c
@@ -46,15 +46,13 @@ struct zpci_ccdf_avail {
 static void __zpci_event_error(struct zpci_ccdf_err *ccdf)
 {
 	struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid);
+	struct pci_dev *pdev = zdev ? zdev->pdev : NULL;
 
 	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);
+	       pdev ? pci_name(pdev) : "n/a", ccdf->pec, ccdf->fid);
 }
 
 void zpci_event_error(void *data)
@@ -89,7 +87,9 @@ static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf)
 		ret = zpci_enable_device(zdev);
 		if (ret)
 			break;
+		pci_lock_rescan_remove();
 		pci_rescan_bus(zdev->bus);
+		pci_unlock_rescan_remove();
 		break;
 	case 0x0302: /* Reserved -> Standby */
 		if (!zdev)
@@ -97,7 +97,7 @@ static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf)
 		break;
 	case 0x0303: /* Deconfiguration requested */
 		if (pdev)
-			pci_stop_and_remove_bus_device(pdev);
+			pci_stop_and_remove_bus_device_locked(pdev);
 
 		ret = zpci_disable_device(zdev);
 		if (ret)
@@ -114,7 +114,7 @@ static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf)
 			/* 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);
+			pci_stop_and_remove_bus_device_locked(pdev);
 		}
 
 		zdev->fh = ccdf->fh;
diff --git a/arch/s390/pci/pci_insn.c b/arch/s390/pci/pci_insn.c
index 85267c058..10ca15dca 100644
--- a/arch/s390/pci/pci_insn.c
+++ b/arch/s390/pci/pci_insn.c
@@ -8,10 +8,23 @@
 #include <linux/errno.h>
 #include <linux/delay.h>
 #include <asm/pci_insn.h>
+#include <asm/pci_debug.h>
 #include <asm/processor.h>
 
 #define ZPCI_INSN_BUSY_DELAY	1	/* 1 microsecond */
 
+static inline void zpci_err_insn(u8 cc, u8 status, u64 req, u64 offset)
+{
+	struct {
+		u64 req;
+		u64 offset;
+		u8 cc;
+		u8 status;
+	} __packed data = {req, offset, cc, status};
+
+	zpci_err_hex(&data, sizeof(data));
+}
+
 /* Modify PCI Function Controls */
 static inline u8 __mpcifc(u64 req, struct zpci_fib *fib, u8 *status)
 {
@@ -38,8 +51,8 @@ int zpci_mod_fc(u64 req, struct zpci_fib *fib)
 	} while (cc == 2);
 
 	if (cc)
-		printk_once(KERN_ERR "%s: error cc: %d  status: %d\n",
-			     __func__, cc, status);
+		zpci_err_insn(cc, status, req, 0);
+
 	return (cc) ? -EIO : 0;
 }
 
@@ -72,8 +85,8 @@ int zpci_refresh_trans(u64 fn, u64 addr, u64 range)
 	} 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);
+		zpci_err_insn(cc, status, addr, range);
+
 	return (cc) ? -EIO : 0;
 }
 
@@ -121,8 +134,8 @@ int zpci_load(u64 *data, u64 req, u64 offset)
 	} 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);
+		zpci_err_insn(cc, status, req, offset);
+
 	return (cc > 0) ? -EIO : cc;
 }
 EXPORT_SYMBOL_GPL(zpci_load);
@@ -159,8 +172,8 @@ int zpci_store(u64 data, u64 req, u64 offset)
 	} 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);
+		zpci_err_insn(cc, status, req, offset);
+
 	return (cc > 0) ? -EIO : cc;
 }
 EXPORT_SYMBOL_GPL(zpci_store);
@@ -195,8 +208,8 @@ int zpci_store_block(const u64 *data, u64 req, u64 offset)
 	} 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);
+		zpci_err_insn(cc, status, req, offset);
+
 	return (cc > 0) ? -EIO : cc;
 }
 EXPORT_SYMBOL_GPL(zpci_store_block);
diff --git a/arch/s390/pci/pci_sysfs.c b/arch/s390/pci/pci_sysfs.c
index fa3ce891e..f37a58088 100644
--- a/arch/s390/pci/pci_sysfs.c
+++ b/arch/s390/pci/pci_sysfs.c
@@ -16,7 +16,7 @@
 static ssize_t name##_show(struct device *dev,				\
 			   struct device_attribute *attr, char *buf)	\
 {									\
-	struct zpci_dev *zdev = get_zdev(to_pci_dev(dev));		\
+	struct zpci_dev *zdev = to_zpci(to_pci_dev(dev));		\
 									\
 	return sprintf(buf, fmt, zdev->member);				\
 }									\
@@ -38,23 +38,30 @@ 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);
+	struct zpci_dev *zdev = to_zpci(pdev);
 	int ret;
 
 	if (!device_remove_file_self(dev, attr))
 		return count;
 
+	pci_lock_rescan_remove();
 	pci_stop_and_remove_bus_device(pdev);
 	ret = zpci_disable_device(zdev);
 	if (ret)
-		return ret;
+		goto error;
 
 	ret = zpci_enable_device(zdev);
 	if (ret)
-		return ret;
+		goto error;
 
 	pci_rescan_bus(zdev->bus);
+	pci_unlock_rescan_remove();
+
 	return count;
+
+error:
+	pci_unlock_rescan_remove();
+	return ret;
 }
 static DEVICE_ATTR_WO(recover);
 
@@ -64,7 +71,7 @@ static ssize_t util_string_read(struct file *filp, struct kobject *kobj,
 {
 	struct device *dev = kobj_to_dev(kobj);
 	struct pci_dev *pdev = to_pci_dev(dev);
-	struct zpci_dev *zdev = get_zdev(pdev);
+	struct zpci_dev *zdev = to_zpci(pdev);
 
 	return memory_read_from_buffer(buf, count, &off, zdev->util_str,
 				       sizeof(zdev->util_str));