Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull s390 updates from Martin Schwidefsky: "The big one is support for fake NUMA, splitting a really large machine in more manageable piece improves performance in some cases, e.g. for a KVM host. The FICON Link Incident handling has been improved, this helps the operator to identify degraded or non-operational FICON connections. The save and restore of floating point and vector registers has been overhauled to allow the future use of vector registers in the kernel. A few small enhancement, magic sys-requests for the vt220 console via SCLP, some more assembler code has been converted to C, the PCI error handling is improved. And the usual cleanup and bug fixing" * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (59 commits) s390/jump_label: Use %*ph to print small buffers s390/sclp_vt220: support magic sysrequests s390/ctrlchar: improve handling of magic sysrequests s390/numa: remove superfluous ARCH_WANT defines s390/3270: redraw screen on unsolicited device end s390/dcssblk: correct out of bounds array indexes s390/mm: simplify page table alloc/free code s390/pci: move debug messages to debugfs s390/nmi: initialize control register 0 earlier s390/zcrypt: use msleep() instead of mdelay() s390/hmcdrv: fix interrupt registration s390/setup: fix novx parameter s390/uaccess: remove uaccess_primary kernel parameter s390: remove unneeded sizeof(void *) comparisons s390/facilities: remove transactional-execution bits s390/numa: re-add DIE sched_domain_topology_level s390/dasd: enhance CUIR scope detection s390/dasd: fix failing path verification s390/vdso: emit a GNU hash s390/numa: make core to node mapping data dynamic ...

Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 updates from Martin Schwidefsky: "The big one is support for fake NUMA, splitting a really large machine in more manageable piece improves performance in some cases, e.g. for a KVM host. The FICON Link Incident handling has been improved, this helps the operator to identify degraded or non-operational FICON connections. The save and restore of floating point and vector registers has been overhauled to allow the future use of vector registers in the kernel. A few small enhancement, magic sys-requests for the vt220 console via SCLP, some more assembler code has been converted to C, the PCI error handling is improved. And the usual cleanup and bug fixing" * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (59 commits) s390/jump_label: Use %*ph to print small buffers s390/sclp_vt220: support magic sysrequests s390/ctrlchar: improve handling of magic sysrequests s390/numa: remove superfluous ARCH_WANT defines s390/3270: redraw screen on unsolicited device end s390/dcssblk: correct out of bounds array indexes s390/mm: simplify page table alloc/free code s390/pci: move debug messages to debugfs s390/nmi: initialize control register 0 earlier s390/zcrypt: use msleep() instead of mdelay() s390/hmcdrv: fix interrupt registration s390/setup: fix novx parameter s390/uaccess: remove uaccess_primary kernel parameter s390: remove unneeded sizeof(void *) comparisons s390/facilities: remove transactional-execution bits s390/numa: re-add DIE sched_domain_topology_level s390/dasd: enhance CUIR scope detection s390/dasd: fix failing path verification s390/vdso: emit a GNU hash s390/numa: make core to node mapping data dynamic ...
9c6a019c · Linus Torvalds · 7c019191 · e4ec7351 · 9c6a019c · 9c6a019c
Commit 9c6a019c authored Aug 31, 2015 by Linus Torvalds
94 changed files
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5907,7 +5907,6 @@ F:	arch/powerpc/kvm/
 KERNEL VIRTUAL MACHINE for s390 (KVM/s390)
 M:	Christian Borntraeger <borntraeger@de.ibm.com>
 M:	Cornelia Huck <cornelia.huck@de.ibm.com>
-M:	linux390@de.ibm.com
 L:	linux-s390@vger.kernel.org
 W:	http://www.ibm.com/developerworks/linux/linux390/
 S:	Supported
@@ -8718,7 +8717,6 @@ F:	drivers/video/fbdev/savage/
 S390
 M:	Martin Schwidefsky <schwidefsky@de.ibm.com>
 M:	Heiko Carstens <heiko.carstens@de.ibm.com>
-M:	linux390@de.ibm.com
 L:	linux-s390@vger.kernel.org
 W:	http://www.ibm.com/developerworks/linux/linux390/
 S:	Supported
@@ -8746,7 +8744,6 @@ F:	block/partitions/ibm.c
 S390 NETWORK DRIVERS
 M:	Ursula Braun <ursula.braun@de.ibm.com>
-M:	linux390@de.ibm.com
 L:	linux-s390@vger.kernel.org
 W:	http://www.ibm.com/developerworks/linux/linux390/
 S:	Supported
@@ -8763,7 +8760,6 @@ F:	drivers/pci/hotplug/s390_pci_hpc.c
 S390 ZCRYPT DRIVER
 M:	Ingo Tuchscherer <ingo.tuchscherer@de.ibm.com>
-M:	linux390@de.ibm.com
 L:	linux-s390@vger.kernel.org
 W:	http://www.ibm.com/developerworks/linux/linux390/
 S:	Supported
@@ -8771,7 +8767,6 @@ F:	drivers/s390/crypto/
 S390 ZFCP DRIVER
 M:	Steffen Maier <maier@linux.vnet.ibm.com>
-M:	linux390@de.ibm.com
 L:	linux-s390@vger.kernel.org
 W:	http://www.ibm.com/developerworks/linux/linux390/
 S:	Supported
@@ -8779,7 +8774,6 @@ F:	drivers/s390/scsi/zfcp_*
 S390 IUCV NETWORK LAYER
 M:	Ursula Braun <ursula.braun@de.ibm.com>
-M:	linux390@de.ibm.com
 L:	linux-s390@vger.kernel.org
 W:	http://www.ibm.com/developerworks/linux/linux390/
 S:	Supported

--- 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/
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -99,18 +99,22 @@ 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 +157,7 @@ config S390
 	select VIRT_CPU_ACCOUNTING
 	select VIRT_TO_BUS
 config SCHED_OMIT_FRAME_POINTER
 	def_bool y
@@ -385,6 +390,76 @@ 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

--- 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)

--- 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

--- 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

--- 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

--- 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");

--- 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");

--- 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");

--- 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_cpu_feature_match(MSA, prng_init);
-module_init(prng_init);
 module_exit(prng_exit);
--- 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");

--- 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");

--- 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");

--- a/arch/s390/include/asm/cpufeature.h
+++ b/arch/s390/include/asm/cpufeature.h
+/*
+ * 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 */
--- 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);

--- a/arch/s390/include/asm/fpu-internal.h
+++ b/arch/s390/include/asm/fpu-internal.h
+/*
+ * 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 */
--- 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
@@ -501,10 +502,9 @@ 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 fpu	  host_fpregs;
-	struct kvm_s390_vregs	*host_vregs;
+	struct fpu	  guest_fpregs;
 	struct kvm_s390_local_interrupt local_int;
 	struct hrtimer    ckc_timer;
 	struct kvm_s390_pgm_info pgm;

--- 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
--- a/arch/s390/include/asm/mmzone.h
+++ b/arch/s390/include/asm/mmzone.h
+/*
+ * 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 */
--- a/arch/s390/include/asm/numa.h
+++ b/arch/s390/include/asm/numa.h
+/*
+ * 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_var_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 */
--- 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
--- 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;

--- 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. */
@@ -230,6 +232,17 @@ static inline void __load_psw_mask (unsigned long mask)
 		: "=&d" (addr), "=Q" (psw) : "Q" (psw) : "memory", "cc");
 }
+/*
+ * 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.
 */
@@ -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 */
--- 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 */
--- 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_fpu_regs();					\
-		save_fp_vx_regs(prev);					\
 		save_access_regs(&prev->thread.acrs[0]);		\
 		save_ri_cb(prev->thread.ri_cb);				\
 	}								\
 	if (next->mm) {							\
 		update_cr_regs(next);					\
-		restore_fp_ctl(&next->thread.fp_regs.fpc);		\
+		set_cpu_flag(CIF_FPU);					\
-		restore_fp_vx_regs(next);				\
 		restore_access_regs(&next->thread.acrs[0]);		\
 		restore_ri_cb(next->thread.ri_cb, prev->thread.ri_cb);	\
 	}								\

--- 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 */
--- a/arch/s390/include/asm/unistd.h
+++ b/arch/s390/include/asm/unistd.h
@@ -11,16 +11,24 @@
 #define __IGNORE_time
-/* Ignore NUMA system calls. Not wired up on s390. */
+/* Ignore system calls that are also reachable via sys_socketcall */
-#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 __IGNORE_socket
+#define __IGNORE_socketpair
+#define __IGNORE_bind
+#define __IGNORE_connect
+#define __IGNORE_listen
+#define __IGNORE_accept4
+#define __IGNORE_getsockopt
+#define __IGNORE_setsockopt
+#define __IGNORE_getsockname
+#define __IGNORE_getpeername
+#define __IGNORE_sendto
+#define __IGNORE_sendmsg
+#define __IGNORE_recvfrom
+#define __IGNORE_recvmsg
+#define __IGNORE_shutdown
 #define __ARCH_WANT_OLD_READDIR
 #define __ARCH_WANT_SYS_ALARM

--- a/arch/s390/include/asm/vx-insn.h
+++ b/arch/s390/include/asm/vx-insn.h
+/*
+ * 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 */
--- 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 */
+#define __NR_mbind		268
-/* Number 269 is reserved for new sys_get_mempolicy */
+#define __NR_get_mempolicy	269
-/* Number 270 is reserved for new sys_set_mempolicy */
+#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

--- 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

--- 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));

--- a/arch/s390/kernel/compat_signal.c
+++ b/arch/s390/kernel/compat_signal.c
@@ -153,33 +153,14 @@ 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);
+	save_fpu_regs();
-	if (current->thread.vxrs) {
-		save_vx_regs(current->thread.vxrs);
-		for (i = 0; i < __NUM_FPRS; i++)
-			current->thread.fp_regs.fprs[i] =
-				*(freg_t *)(current->thread.vxrs + i);
-	} else
-		save_fp_regs(current->thread.fp_regs.fprs);
 }
 /* Load registers after signal return */
 static void load_sigregs(void)
 {
-	int i;
 	restore_access_regs(current->thread.acrs);
-	/* restore_fp_ctl is done in restore_sigregs */
-	if (current->thread.vxrs) {
-		for (i = 0; i < __NUM_FPRS; i++)
-			*(freg_t *)(current->thread.vxrs + i) =
-				current->thread.fp_regs.fprs[i];
-		restore_vx_regs(current->thread.vxrs);
-	} else
-		restore_fp_regs(current->thread.fp_regs.fprs);
 }
 static int save_sigregs32(struct pt_regs *regs, _sigregs32 __user *sregs)
@@ -196,8 +177,7 @@ 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,
+	fpregs_store((_s390_fp_regs *) &user_sregs.fpregs, &current->thread.fpu);
-	       sizeof(user_sregs.fpregs));
 	if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs32)))
 		return -EFAULT;
 	return 0;
@@ -217,8 +197,8 @@ 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. */
+	/* Test the floating-point-control word. */
-	if (restore_fp_ctl(&user_sregs.fpregs.fpc))
+	if (test_fp_ctl(user_sregs.fpregs.fpc))
 		return -EINVAL;
 	/* Use regs->psw.mask instead of PSW_USER_BITS to preserve PER bit. */
@@ -235,9 +215,7 @@ 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));
+	fpregs_load((_s390_fp_regs *) &user_sregs.fpregs, &current->thread.fpu);
-	memcpy(&current->thread.fp_regs, &user_sregs.fpregs,
-	       sizeof(current->thread.fp_regs));
 	clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */
 	return 0;
@@ -258,13 +236,13 @@ 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;
 	}
@@ -286,15 +264,15 @@ 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;
 }
@@ -308,6 +286,7 @@ COMPAT_SYSCALL_DEFINE0(sigreturn)
 	if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE32))
 		goto badframe;
 	set_current_blocked(&set);
+	save_fpu_regs();
 	if (restore_sigregs32(regs, &frame->sregs))
 		goto badframe;
 	if (restore_sigregs_ext32(regs, &frame->sregs_ext))
@@ -330,6 +309,7 @@ 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))
@@ -472,7 +452,7 @@ 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) +

--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
--- 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)

--- 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",
+	pr_emerg("Found:    %6ph\n", ipc);
-		 ipc[0], ipc[1], ipc[2], ipc[3], ipc[4], ipc[5]);
+	pr_emerg("Expected: %6ph\n", ipe);
-	pr_emerg("Expected: %02x %02x %02x %02x %02x %02x\n",
+	pr_emerg("New:      %6ph\n", ipn);
-		 ipe[0], ipe[1], ipe[2], ipe[3], ipe[4], ipe[5]);
-	pr_emerg("New:      %02x %02x %02x %02x %02x %02x\n",
-		 ipn[0], ipn[1], ipn[2], ipn[3], ipn[4], ipn[5]);
 	panic("Corrupted kernel text");
 }

--- 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 *)
+		asm volatile(
-				&S390_lowcore.vector_save_area);
+			"	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);
--- 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
+	/* The host-program-parameter (hpp) contains the pid of
-	 * block that is set by sie64a() in entry64.S.	Check if hpp
+	 * the CPU thread as set by sie64a() in entry.S.
-	 * refers to a valid control block and set sde_regs flags
+	 * If non-zero assume a guest sample.
-	 * accordingly.  This would allow to use hpp values for other
-	 * purposes too.
-	 * For now, simply use a non-zero value as guest indicator.
 	 */
 	if (sfr->basic.hpp)
 		sde_regs->in_guest = 1;

--- 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)
+	/* Free either the floating-point or the vector register save area */
-		kfree(tsk->thread.vxrs);
+	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_fpu_regs();
-	save_fp_regs(fpregs->fprs);
+	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);

--- a/arch/s390/kernel/processor.c
+++ b/arch/s390/kernel/processor.c
@@ -40,6 +40,15 @@ void cpu_init(void)
 	enter_lazy_tlb(&init_mm, current);
 }
+/*
+ * 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.
 */

--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -45,32 +45,21 @@ 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);
+		cr_new = cr | (1UL << 55);
 		if (task->thread.per_flags & PER_FLAG_NO_TE)
 			cr_new &= ~(1UL << 55);
-		}
-		if (MACHINE_HAS_VX) {
-			/* Enable/disable of vector extension */
-			cr_new &= ~(1UL << 17);
-			if (task->thread.vxrs)
-				cr_new |= (1UL << 17);
-		}
 		if (cr_new != cr)
 			__ctl_load(cr_new, 0, 0);
-		if (MACHINE_HAS_TE) {
+		/* Set or clear transaction execution TDC bits 62 and 63. */
-			/* 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 &
+			if (task->thread.per_flags & PER_FLAG_TE_ABORT_RAND_TEND)
-				    PER_FLAG_TE_ABORT_RAND_TEND)
 				cr_new |= 1UL;
 			else
 				cr_new |= 2UL;
@@ -78,7 +67,6 @@ void update_cr_regs(struct task_struct *task)
 		if (cr_new != cr)
 			__ctl_load(cr_new, 2, 2);
 	}
-	}
 	/* Copy user specified PER registers */
 	new.control = thread->per_user.control;
 	new.start = thread->per_user.start;
@@ -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
+		 * floating point regs. are either in child->thread.fpu
-		 * or the child->thread.vxrs array
+		 * 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
+		 * floating point regs. are either in child->thread.fpu
-		 * or the child->thread.vxrs array
+		 * 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
+		 * floating point regs. are either in child->thread.fpu
-		 * or the child->thread.vxrs array
+		 * 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
+		 * floating point regs. are either in child->thread.fpu
-		 * or the child->thread.vxrs array
+		 * 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) {
+	_s390_fp_regs fp_regs;
-		save_fp_ctl(&target->thread.fp_regs.fpc);
-		save_fp_regs(target->thread.fp_regs.fprs);
+	if (target == current)
-	} else if (target->thread.vxrs) {
+		save_fpu_regs();
-		int i;
+	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) {
+	if (target == current)
-		save_fp_ctl(&target->thread.fp_regs.fpc);
+		save_fpu_regs();
-		save_fp_regs(target->thread.fp_regs.fprs);
-	}
 	/* If setting FPC, must validate it first. */
 	if (count > 0 && pos < offsetof(s390_fp_regs, fprs)) {
-		u32 ufpc[2] = { target->thread.fp_regs.fpc, 0 };
+		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,
+					fprs, offsetof(s390_fp_regs, fprs), -1);
-					offsetof(s390_fp_regs, fprs), -1);
+	if (rc)
+		return rc;
-	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++)
+	if (is_vx_task(target))
-				*(freg_t *)(target->thread.vxrs + i) =
+		convert_fp_to_vx(target->thread.fpu.vxrs, fprs);
-					target->thread.fp_regs.fprs[i];
+	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];
+			*((__u64 *)(target->thread.fpu.vxrs + i) + 1) = vxrs[i];
-		if (target == current)
-			restore_vx_regs(target->thread.vxrs);
-	}
 	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);
+			save_fpu_regs();
-		memcpy(vxrs, target->thread.vxrs + __NUM_VXRS_LOW,
+		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);
+				target->thread.fpu.vxrs + __NUM_VXRS_LOW, 0, -1);
-	if (rc == 0 && target == current)
-		restore_vx_regs(target->thread.vxrs);
 	return rc;
 }

--- a/arch/s390/kernel/s390_ksyms.c
+++ b/arch/s390/kernel/s390_ksyms.c
 #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);
--- a/arch/s390/kernel/sclp.S
+++ b/arch/s390/kernel/sclp.S
-/*
- * 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
--- a/arch/s390/kernel/sclp.c
+++ b/arch/s390/kernel/sclp.c
+/*
+ *    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);
+}
--- 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;
@@ -688,7 +689,7 @@ static void __init setup_memory(void)
 /*
 * Setup hardware capabilities.
 */
-static void __init setup_hwcaps(void)
+static int __init setup_hwcaps(void)
 {
 	static const int stfl_bits[6] = { 0, 2, 7, 17, 19, 21 };
 	struct cpuid cpu_id;
@@ -754,9 +755,11 @@ static void __init setup_hwcaps(void)
 		elf_hwcap |= HWCAP_S390_TE;
 	/*
-	 * Vector extension HWCAP_S390_VXRS is bit 11.
+	 * Vector extension HWCAP_S390_VXRS is bit 11. The Vector extension
+	 * can be disabled with the "novx" parameter. Use MACHINE_HAS_VX
+	 * instead of facility bit 129.
 	 */
-	if (test_facility(129))
+	if (MACHINE_HAS_VX)
 		elf_hwcap |= HWCAP_S390_VXRS;
 	get_cpu_id(&cpu_id);
 	add_device_randomness(&cpu_id, sizeof(cpu_id));
@@ -793,7 +796,9 @@ static void __init setup_hwcaps(void)
 		strcpy(elf_platform, "z13");
 		break;
 	}
+	return 0;
 }
+arch_initcall(setup_hwcaps);
 /*
 * Add system information as device randomness
@@ -879,11 +884,7 @@ void __init setup_arch(char **cmdline_p)
 	setup_lowcore();
 	smp_fill_possible_mask();
        cpu_init();
+	numa_setup();
-	/*
-	 * Setup capabilities (ELF_HWCAP & ELF_PLATFORM).
-	 */
-	setup_hwcaps();
 	/*
 	 * Create kernel page tables and switch to virtual addressing.

--- 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);
+	save_fpu_regs();
-	if (current->thread.vxrs) {
-		int i;
-		save_vx_regs(current->thread.vxrs);
-		for (i = 0; i < __NUM_FPRS; i++)
-			current->thread.fp_regs.fprs[i] =
-				*(freg_t *)(current->thread.vxrs + i);
-	} else
-		save_fp_regs(current->thread.fp_regs.fprs);
 }
 /* Load registers after signal return */
 static void load_sigregs(void)
 {
 	restore_access_regs(current->thread.acrs);
-	/* restore_fp_ctl is done in restore_sigregs */
-	if (current->thread.vxrs) {
-		int i;
-		for (i = 0; i < __NUM_FPRS; i++)
-			*(freg_t *)(current->thread.vxrs + i) =
-				current->thread.fp_regs.fprs[i];
-		restore_vx_regs(current->thread.vxrs);
-	} else
-		restore_fp_regs(current->thread.fp_regs.fprs);
 }
 /* Returns non-zero on fault. */
@@ -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,
+	fpregs_store(&user_sregs.fpregs, &current->thread.fpu);
-	       sizeof(user_sregs.fpregs));
 	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. */
+	/* Test the floating-point-control word. */
-	if (restore_fp_ctl(&user_sregs.fpregs.fpc))
+	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,
+	fpregs_load(&user_sregs.fpregs, &current->thread.fpu);
-	       sizeof(current->thread.fp_regs));
 	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);

--- 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,
+static void __init __smp_store_cpu_state(struct save_area_ext *sa_ext,
-				  int is_boot_cpu)
+					 u16 address, int is_boot_cpu)
 {
 	void *lc = (void *)(unsigned long) store_prefix();
 	unsigned long vx_sa;

--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -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 */
+SYSCALL(sys_mbind,compat_sys_mbind)
-NI_SYSCALL						/* 269 sys_get_mempolicy */
+SYSCALL(sys_get_mempolicy,compat_sys_get_mempolicy)
-NI_SYSCALL						/* 270 sys_set_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)
@@ -295,7 +295,7 @@ SYSCALL(sys_ioprio_get,compat_sys_ioprio_get)
 SYSCALL(sys_inotify_init,sys_inotify_init)
 SYSCALL(sys_inotify_add_watch,compat_sys_inotify_add_watch)	/* 285 */
 SYSCALL(sys_inotify_rm_watch,compat_sys_inotify_rm_watch)
-NI_SYSCALL						/* 287 sys_migrate_pages */
+SYSCALL(sys_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)

--- 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) {
+	if (MACHINE_HAS_TOPOLOGY) {
-		update_cpu_masks();
+		rc = 1;
-		topology_update_polarization_simple();
-		return 0;
-	}
 		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)

--- 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++)
+	convert_fp_to_vx(vxrs, tsk->thread.fpu.fprs);
-		*(freg_t *) &vxrs[i] = tsk->thread.fp_regs.fprs[i];
+	fprs = tsk->thread.fpu.fprs;
-	tsk->thread.vxrs = vxrs;
+	tsk->thread.fpu.vxrs = vxrs;
-	if (tsk == current) {
+	tsk->thread.fpu.flags |= FPU_USE_VX;
-		__ctl_set_bit(0, 17);
+	kfree(fprs);
-		restore_vx_regs(vxrs);
-	}
 	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));
+	save_fpu_regs();
-	vic = (current->thread.fp_regs.fpc & 0xf00) >> 8;
+	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 &&
+	if (MACHINE_HAS_VX && !is_vx_task(current) &&
-	    (current->thread.fp_regs.fpc & FPC_DXC_MASK) == 0xfe00) {
+	    (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");
 }

--- 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)

--- 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)

--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -28,6 +28,7 @@ static atomic64_t virt_timer_elapsed;
 static DEFINE_PER_CPU(u64, mt_cycles[32]);
 static DEFINE_PER_CPU(u64, mt_scaling_mult) = { 1 };
 static DEFINE_PER_CPU(u64, mt_scaling_div) = { 1 };
+static DEFINE_PER_CPU(u64, mt_scaling_jiffies);
 static inline u64 get_vtimer(void)
 {
@@ -85,7 +86,8 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset)
 	S390_lowcore.steal_timer += S390_lowcore.last_update_clock - clock;
 	/* Do MT utilization calculation */
-	if (smp_cpu_mtid) {
+	if (smp_cpu_mtid &&
+	    time_after64(jiffies_64, __this_cpu_read(mt_scaling_jiffies))) {
 		u64 cycles_new[32], *cycles_old;
 		u64 delta, mult, div;
@@ -105,6 +107,7 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset)
 				       sizeof(u64) * (smp_cpu_mtid + 1));
 			}
 		}
+		__this_cpu_write(mt_scaling_jiffies, jiffies_64);
 	}
 	user = S390_lowcore.user_timer - ti->user_timer;
@@ -376,4 +379,11 @@ 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));
+	}
 }
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -1283,21 +1283,54 @@ 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);
+	/* Save host register state */
-	if (test_kvm_facility(vcpu->kvm, 129))
+	save_fpu_regs();
-		save_vx_regs((__vector128 *)&vcpu->arch.host_vregs->vrs);
+	save_fpu_to(&vcpu->arch.host_fpregs);
-	else
-		save_fp_regs(vcpu->arch.host_fpregs.fprs);
-	save_access_regs(vcpu->arch.host_acrs);
 	if (test_kvm_facility(vcpu->kvm, 129)) {
-		restore_fp_ctl(&vcpu->run->s.regs.fpc);
+		current->thread.fpu.fpc = vcpu->run->s.regs.fpc;
-		restore_vx_regs((__vector128 *)&vcpu->run->s.regs.vrs);
+		current->thread.fpu.flags = FPU_USE_VX;
-	} else {
+		/*
-		restore_fp_ctl(&vcpu->arch.guest_fpregs.fpc);
+		 * Use the register save area in the SIE-control block
-		restore_fp_regs(vcpu->arch.guest_fpregs.fprs);
+		 * 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);
@@ -1307,19 +1340,22 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
 	atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
 	gmap_disable(vcpu->arch.gmap);
-	if (test_kvm_facility(vcpu->kvm, 129)) {
-		save_fp_ctl(&vcpu->run->s.regs.fpc);
+	save_fpu_regs();
-		save_vx_regs((__vector128 *)&vcpu->run->s.regs.vrs);
-	} else {
-		save_fp_ctl(&vcpu->arch.guest_fpregs.fpc);
-		save_fp_regs(vcpu->arch.guest_fpregs.fprs);
-	}
-	save_access_regs(vcpu->run->s.regs.acrs);
-	restore_fp_ctl(&vcpu->arch.host_fpregs.fpc);
 	if (test_kvm_facility(vcpu->kvm, 129))
-		restore_vx_regs((__vector128 *)&vcpu->arch.host_vregs->vrs);
+		/*
+		 * 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);
 }
@@ -1464,7 +1500,6 @@ 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)) {
@@ -1486,6 +1521,19 @@ 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;
@@ -1708,16 +1756,16 @@ 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);
+	save_fpu_regs();
-	restore_fp_regs(vcpu->arch.guest_fpregs.fprs);
+	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;
 }
@@ -2268,8 +2316,21 @@ 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_fpu_regs();
-	save_fp_regs(vcpu->arch.guest_fpregs.fprs);
+	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);
@@ -2296,10 +2357,13 @@ 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
+	 * copying in vcpu load/put. We can simply call save_fpu_regs()
-	 * it into the save area.
+	 * 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);
 }

--- 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)
 {

--- a/arch/s390/lib/uaccess.c
+++ b/arch/s390/lib/uaccess.c
@@ -370,22 +370,9 @@ 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))
+	if (test_facility(27))
 		static_key_slow_inc(&have_mvcos);
 	return 0;
 }

--- 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)

--- 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;

--- 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 */
@@ -170,37 +171,36 @@ void __init free_initrd_mem(unsigned long start, unsigned long end)
 #ifdef CONFIG_MEMORY_HOTPLUG
 int arch_add_memory(int nid, u64 start, u64 size)
 {
-	unsigned long zone_start_pfn, zone_end_pfn, nr_pages;
+	unsigned long 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;
+	unsigned long nr_pages;
-	int rc;
+	int rc, zone_enum;
 	rc = vmem_add_mapping(start, size);
 	if (rc)
 		return rc;
-	for_each_zone(zone) {
-		if (zone_idx(zone) != ZONE_MOVABLE) {
+	while (size_pages > 0) {
-			/* Add range within existing zone limits */
+		if (start_pfn < dma_end_pfn) {
-			zone_start_pfn = zone->zone_start_pfn;
+			nr_pages = (start_pfn + size_pages > dma_end_pfn) ?
-			zone_end_pfn = zone->zone_start_pfn +
+				   dma_end_pfn - start_pfn : size_pages;
-				       zone->spanned_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 */
+			nr_pages = size_pages;
-			zone_start_pfn = start_pfn;
+			zone_enum = ZONE_MOVABLE;
-			zone_end_pfn = start_pfn + size_pages;
 		}
-		if (start_pfn < zone_start_pfn || start_pfn >= zone_end_pfn)
+		rc = __add_pages(nid, NODE_DATA(nid)->node_zones + zone_enum,
-			continue;
+				 start_pfn, size_pages);
-		nr_pages = (start_pfn + size_pages > zone_end_pfn) ?
-			   zone_end_pfn - start_pfn : size_pages;
-		rc = __add_pages(nid, zone, start_pfn, nr_pages);
 		if (rc)
 			break;
 		start_pfn += nr_pages;
 		size_pages -= nr_pages;
-		if (!size_pages)
-			break;
 	}
 	if (rc)
 		vmem_remove_mapping(start, size);

--- 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)
 {
@@ -994,24 +939,33 @@ 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);
+	unsigned long *table;
-	struct page *uninitialized_var(page);
+	struct page *page;
 	unsigned int mask, bit;
-	if (mm_alloc_pgste(mm))
+	/* Try to get a fragment of a 4K page as a 2K page table */
-		return page_table_alloc_pgste(mm);
+	if (!mm_alloc_pgste(mm)) {
-	/* Allocate fragments of a 4K page as 1K/2K page table */
+		table = NULL;
 		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);
+			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);
+			}
 		}
-	if ((mask & FRAG_MASK) == FRAG_MASK) {
 		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;
@@ -1019,19 +973,21 @@ unsigned long *page_table_alloc(struct mm_struct *mm)
 		__free_page(page);
 		return NULL;
 	}
-		atomic_set(&page->_mapcount, 1);
+	/* 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);
 		clear_table(table, _PAGE_INVALID, PAGE_SIZE);
 		spin_lock_bh(&mm->context.list_lock);
 		list_add(&page->lru, &mm->context.pgtable_list);
-	} else {
-		for (bit = 1; mask & bit; bit <<= 1)
-			table += PTRS_PER_PTE;
-		mask = atomic_xor_bits(&page->_mapcount, bit);
-		if ((mask & FRAG_MASK) == FRAG_MASK)
-			list_del(&page->lru);
-	}
 		spin_unlock_bh(&mm->context.list_lock);
+	}
 	return table;
 }
@@ -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))
+	if (!mm_alloc_pgste(mm)) {
-		return page_table_free_pgste(table);
+		/* Free 2K page table fragment of a 4K page */
-	/* Free 1K/2K page table fragment of a 4K page */
+		bit = (__pa(table) & ~PAGE_MASK)/(PTRS_PER_PTE*sizeof(pte_t));
-	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)
+		mask = atomic_xor_bits(&page->_mapcount, 1U << bit);
-		list_del(&page->lru);
+		if (mask & 3)
-	mask = atomic_xor_bits(&page->_mapcount, bit);
-	if (mask & FRAG_MASK)
 			list_add(&page->lru, &mm->context.pgtable_list);
+		else
+			list_del(&page->lru);
 		spin_unlock_bh(&mm->context.list_lock);
-	if (mask == 0) {
+		if (mask != 0)
-		pgtable_page_dtor(page);
+			return;
-		atomic_set(&page->_mapcount, -1);
-		__free_page(page);
 	}
-}
-static void __page_table_free_rcu(void *table, unsigned bit)
-{
-	struct page *page;
-	if (bit == FRAG_MASK)
-		return page_table_free_pgste(table);
-	/* Free 1K/2K page table fragment of a 4K page */
-	page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
-	if (atomic_xor_bits(&page->_mapcount, bit) == 0) {
 	pgtable_page_dtor(page);
 	atomic_set(&page->_mapcount, -1);
 	__free_page(page);
-	}
 }
 void page_table_free_rcu(struct mmu_gather *tlb, unsigned long *table,
@@ -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)
+	mask = atomic_xor_bits(&page->_mapcount, 0x11U << bit);
-		list_del(&page->lru);
+	if (mask & 3)
-	mask = atomic_xor_bits(&page->_mapcount, bit | (bit << 4));
-	if (mask & FRAG_MASK)
 		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;
+	unsigned int mask = (unsigned long) _table & 3;
-	void *table = (void *)((unsigned long) _table & ~mask);
+	void *table = (void *)((unsigned long) _table ^ mask);
-	unsigned type = (unsigned long) _table & mask;
+	struct page *page = pfn_to_page(__pa(table) >> PAGE_SHIFT);
-	if (type)
+	switch (mask) {
-		__page_table_free_rcu(table, type);
+	case 0:		/* pmd or pud */
-	else
+		free_pages((unsigned long) table, 2);
-		free_pages((unsigned long) table, ALLOC_ORDER);
+		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)

--- a/arch/s390/numa/Makefile
+++ b/arch/s390/numa/Makefile
+obj-y			+= numa.o
+obj-y			+= toptree.o
+obj-$(CONFIG_NUMA_EMU)	+= mode_emu.o
--- a/arch/s390/numa/mode_emu.c
+++ b/arch/s390/numa/mode_emu.c
--- a/arch/s390/numa/numa.c
+++ b/arch/s390/numa/numa.c
--- a/arch/s390/numa/numa_mode.h
+++ b/arch/s390/numa/numa_mode.h
+/*
+ * 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 */
--- a/arch/s390/numa/toptree.c
+++ b/arch/s390/numa/toptree.c
--- a/arch/s390/numa/toptree.h
+++ b/arch/s390/numa/toptree.h
--- a/arch/s390/pci/pci.c
+++ b/arch/s390/pci/pci.c
--- a/arch/s390/pci/pci_dma.c
+++ b/arch/s390/pci/pci_dma.c
--- a/arch/s390/pci/pci_event.c
+++ b/arch/s390/pci/pci_event.c
--- a/arch/s390/pci/pci_insn.c
+++ b/arch/s390/pci/pci_insn.c
--- a/arch/s390/pci/pci_sysfs.c
+++ b/arch/s390/pci/pci_sysfs.c
--- a/drivers/s390/block/dasd_alias.c
+++ b/drivers/s390/block/dasd_alias.c
--- a/drivers/s390/block/dasd_eckd.c
+++ b/drivers/s390/block/dasd_eckd.c
--- a/drivers/s390/block/dasd_eckd.h
+++ b/drivers/s390/block/dasd_eckd.h
--- a/drivers/s390/block/dasd_int.h
+++ b/drivers/s390/block/dasd_int.h
@@ -534,6 +534,7 @@ struct dasd_attention_data {
 #define DASD_FLAG_SAFE_OFFLINE	10	/* safe offline processing requested*/
 #define DASD_FLAG_SAFE_OFFLINE_RUNNING	11	/* safe offline running */
 #define DASD_FLAG_ABORTALL	12	/* Abort all noretry requests */
+#define DASD_FLAG_PATH_VERIFY	13	/* Path verification worker running */
 #define DASD_SLEEPON_START_TAG	((void *) 1)
 #define DASD_SLEEPON_END_TAG	((void *) 2)

--- a/drivers/s390/block/dcssblk.c
+++ b/drivers/s390/block/dcssblk.c
--- a/drivers/s390/char/con3270.c
+++ b/drivers/s390/char/con3270.c
--- a/drivers/s390/char/ctrlchar.c
+++ b/drivers/s390/char/ctrlchar.c
--- a/drivers/s390/char/ctrlchar.h
+++ b/drivers/s390/char/ctrlchar.h
--- a/drivers/s390/char/diag_ftp.c
+++ b/drivers/s390/char/diag_ftp.c
--- a/drivers/s390/char/monreader.c
+++ b/drivers/s390/char/monreader.c
--- a/drivers/s390/char/sclp.c
+++ b/drivers/s390/char/sclp.c
--- a/drivers/s390/char/sclp_cmd.c
+++ b/drivers/s390/char/sclp_cmd.c
--- a/drivers/s390/char/sclp_vt220.c
+++ b/drivers/s390/char/sclp_vt220.c
--- a/drivers/s390/char/tty3270.c
+++ b/drivers/s390/char/tty3270.c
--- a/drivers/s390/cio/chsc.c
+++ b/drivers/s390/cio/chsc.c
--- a/drivers/s390/cio/device_ops.c
+++ b/drivers/s390/cio/device_ops.c
--- a/drivers/s390/cio/eadm_sch.c
+++ b/drivers/s390/cio/eadm_sch.c
--- a/drivers/s390/crypto/ap_bus.c
+++ b/drivers/s390/crypto/ap_bus.c
--- a/drivers/s390/crypto/zcrypt_pcixcc.c
+++ b/drivers/s390/crypto/zcrypt_pcixcc.c
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
--- a/drivers/s390/scsi/zfcp_fsf.c
+++ b/drivers/s390/scsi/zfcp_fsf.c
--- a/include/linux/cpufeature.h
+++ b/include/linux/cpufeature.h