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

Pull s390 updates from Martin Schwidefsky:
 "This patch set contains the main portion of the changes for 3.18 in
  regard to the s390 architecture.  It is a bit bigger than usual,
  mainly because of a new driver and the vector extension patches.

  The interesting bits are:
   - Quite a bit of work on the tracing front.  Uprobes is enabled and
     the ftrace code is reworked to get some of the lost performance
     back if CONFIG_FTRACE is enabled.
   - To improve boot time with CONFIG_DEBIG_PAGEALLOC, support for the
     IPTE range facility is added.
   - The rwlock code is re-factored to improve writer fairness and to be
     able to use the interlocked-access instructions.
   - The kernel part for the support of the vector extension is added.
   - The device driver to access the CD/DVD on the HMC is added, this
     will hopefully come in handy to improve the installation process.
   - Add support for control-unit initiated reconfiguration.
   - The crypto device driver is enhanced to enable the additional AP
     domains and to allow the new crypto hardware to be used.
   - Bug fixes"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (39 commits)
  s390/ftrace: simplify enabling/disabling of ftrace_graph_caller
  s390/ftrace: remove 31 bit ftrace support
  s390/kdump: add support for vector extension
  s390/disassembler: add vector instructions
  s390: add support for vector extension
  s390/zcrypt: Toleration of new crypto hardware
  s390/idle: consolidate idle functions and definitions
  s390/nohz: use a per-cpu flag for arch_needs_cpu
  s390/vtime: do not reset idle data on CPU hotplug
  s390/dasd: add support for control unit initiated reconfiguration
  s390/dasd: fix infinite loop during format
  s390/mm: make use of ipte range facility
  s390/setup: correct 4-level kernel page table detection
  s390/topology: call set_sched_topology early
  s390/uprobes: architecture backend for uprobes
  s390/uprobes: common library for kprobes and uprobes
  s390/rwlock: use the interlocked-access facility 1 instructions
  s390/rwlock: improve writer fairness
  s390/rwlock: remove interrupt-enabling rwlock variant.
  s390/mm: remove change bit override support
  ...

Documentation/kprobes.txt

@@ -300,6 +300,7 @@ architectures:
 - arm
 - ppc
 - mips
+- s390
 
 3. Configuring Kprobes
 

arch/s390/Kconfig

@@ -58,6 +58,9 @@ config NO_IOPORT_MAP
 config PCI_QUIRKS
 	def_bool n
 
+config ARCH_SUPPORTS_UPROBES
+	def_bool 64BIT
+
 config S390
 	def_bool y
 	select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
@@ -97,6 +100,7 @@ config S390
 	select ARCH_WANT_IPC_PARSE_VERSION
 	select BUILDTIME_EXTABLE_SORT
 	select CLONE_BACKWARDS2
+	select DYNAMIC_FTRACE if FUNCTION_TRACER
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_CPU_DEVICES if !SMP
 	select GENERIC_FIND_FIRST_BIT
@@ -113,10 +117,11 @@ config S390
 	select HAVE_CMPXCHG_LOCAL
 	select HAVE_C_RECORDMCOUNT
 	select HAVE_DEBUG_KMEMLEAK
-	select HAVE_DYNAMIC_FTRACE
+	select HAVE_DYNAMIC_FTRACE if 64BIT
+	select HAVE_DYNAMIC_FTRACE_WITH_REGS if 64BIT
 	select HAVE_FTRACE_MCOUNT_RECORD
-	select HAVE_FUNCTION_GRAPH_TRACER
-	select HAVE_FUNCTION_TRACER
+	select HAVE_FUNCTION_GRAPH_TRACER if 64BIT
+	select HAVE_FUNCTION_TRACER if 64BIT
 	select HAVE_FUTEX_CMPXCHG if FUTEX
 	select HAVE_KERNEL_BZIP2
 	select HAVE_KERNEL_GZIP

arch/s390/Makefile

@@ -35,13 +35,16 @@ endif
 
 export LD_BFD
 
-cflags-$(CONFIG_MARCH_G5)     += -march=g5
-cflags-$(CONFIG_MARCH_Z900)   += -march=z900
-cflags-$(CONFIG_MARCH_Z990)   += -march=z990
-cflags-$(CONFIG_MARCH_Z9_109) += -march=z9-109
-cflags-$(CONFIG_MARCH_Z10)    += -march=z10
-cflags-$(CONFIG_MARCH_Z196)   += -march=z196
-cflags-$(CONFIG_MARCH_ZEC12)  += -march=zEC12
+mflags-$(CONFIG_MARCH_G5)     := -march=g5
+mflags-$(CONFIG_MARCH_Z900)   := -march=z900
+mflags-$(CONFIG_MARCH_Z990)   := -march=z990
+mflags-$(CONFIG_MARCH_Z9_109) := -march=z9-109
+mflags-$(CONFIG_MARCH_Z10)    := -march=z10
+mflags-$(CONFIG_MARCH_Z196)   := -march=z196
+mflags-$(CONFIG_MARCH_ZEC12)  := -march=zEC12
+
+aflags-y += $(mflags-y)
+cflags-y += $(mflags-y)
 
 cflags-$(CONFIG_MARCH_G5_TUNE)		+= -mtune=g5
 cflags-$(CONFIG_MARCH_Z900_TUNE)	+= -mtune=z900

arch/s390/include/asm/barrier.h

@@ -15,11 +15,13 @@
 
 #ifdef CONFIG_HAVE_MARCH_Z196_FEATURES
 /* Fast-BCR without checkpoint synchronization */
-#define mb() do {  asm volatile("bcr 14,0" : : : "memory"); } while (0)
+#define __ASM_BARRIER "bcr 14,0\n"
 #else
-#define mb() do {  asm volatile("bcr 15,0" : : : "memory"); } while (0)
+#define __ASM_BARRIER "bcr 15,0\n"
 #endif
 
+#define mb() do {  asm volatile(__ASM_BARRIER : : : "memory"); } while (0)
+
 #define rmb()				mb()
 #define wmb()				mb()
 #define read_barrier_depends()		do { } while(0)

arch/s390/include/asm/cputime.h

@@ -8,8 +8,6 @@
 #define _S390_CPUTIME_H
 
 #include <linux/types.h>
-#include <linux/percpu.h>
-#include <linux/spinlock.h>
 #include <asm/div64.h>
 
 
@@ -167,28 +165,8 @@ static inline clock_t cputime64_to_clock_t(cputime64_t cputime)
 	return clock;
 }
 
-struct s390_idle_data {
-	int nohz_delay;
-	unsigned int sequence;
-	unsigned long long idle_count;
-	unsigned long long idle_time;
-	unsigned long long clock_idle_enter;
-	unsigned long long clock_idle_exit;
-	unsigned long long timer_idle_enter;
-	unsigned long long timer_idle_exit;
-};
+cputime64_t arch_cpu_idle_time(int cpu);
 
-DECLARE_PER_CPU(struct s390_idle_data, s390_idle);
-
-cputime64_t s390_get_idle_time(int cpu);
-
-#define arch_idle_time(cpu) s390_get_idle_time(cpu)
-
-static inline int s390_nohz_delay(int cpu)
-{
-	return __get_cpu_var(s390_idle).nohz_delay != 0;
-}
-
-#define arch_needs_cpu(cpu) s390_nohz_delay(cpu)
+#define arch_idle_time(cpu) arch_cpu_idle_time(cpu)
 
 #endif /* _S390_CPUTIME_H */

arch/s390/include/asm/dis.h

@@ -13,12 +13,13 @@
 #define OPERAND_FPR	0x2	/* Operand printed as %fx */
 #define OPERAND_AR	0x4	/* Operand printed as %ax */
 #define OPERAND_CR	0x8	/* Operand printed as %cx */
-#define OPERAND_DISP	0x10	/* Operand printed as displacement */
-#define OPERAND_BASE	0x20	/* Operand printed as base register */
-#define OPERAND_INDEX	0x40	/* Operand printed as index register */
-#define OPERAND_PCREL	0x80	/* Operand printed as pc-relative symbol */
-#define OPERAND_SIGNED	0x100	/* Operand printed as signed value */
-#define OPERAND_LENGTH	0x200	/* Operand printed as length (+1) */
+#define OPERAND_VR	0x10	/* Operand printed as %vx */
+#define OPERAND_DISP	0x20	/* Operand printed as displacement */
+#define OPERAND_BASE	0x40	/* Operand printed as base register */
+#define OPERAND_INDEX	0x80	/* Operand printed as index register */
+#define OPERAND_PCREL	0x100	/* Operand printed as pc-relative symbol */
+#define OPERAND_SIGNED	0x200	/* Operand printed as signed value */
+#define OPERAND_LENGTH	0x400	/* Operand printed as length (+1) */
 
 
 struct s390_operand {

arch/s390/include/asm/elf.h

@@ -102,6 +102,7 @@
 #define HWCAP_S390_ETF3EH	256
 #define HWCAP_S390_HIGH_GPRS	512
 #define HWCAP_S390_TE		1024
+#define HWCAP_S390_VXRS		2048
 
 /*
  * These are used to set parameters in the core dumps.
@@ -225,6 +226,6 @@ int arch_setup_additional_pages(struct linux_binprm *, int);
 extern unsigned long arch_randomize_brk(struct mm_struct *mm);
 #define arch_randomize_brk arch_randomize_brk
 
-void *fill_cpu_elf_notes(void *ptr, struct save_area *sa);
+void *fill_cpu_elf_notes(void *ptr, struct save_area *sa, __vector128 *vxrs);
 
 #endif

arch/s390/include/asm/ftrace.h

@@ -4,6 +4,7 @@
 #ifndef __ASSEMBLY__
 
 extern void _mcount(void);
+extern char ftrace_graph_caller_end;
 
 struct dyn_arch_ftrace { };
 
@@ -17,10 +18,8 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr)
 
 #endif /* __ASSEMBLY__ */
 
-#ifdef CONFIG_64BIT
-#define MCOUNT_INSN_SIZE  12
-#else
-#define MCOUNT_INSN_SIZE  22
-#endif
+#define MCOUNT_INSN_SIZE  18
+
+#define ARCH_SUPPORTS_FTRACE_OPS 1
 
 #endif /* _ASM_S390_FTRACE_H */

arch/s390/include/asm/idle.h

+/*
+ *  Copyright IBM Corp. 2014
+ *
+ *  Author: Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#ifndef _S390_IDLE_H
+#define _S390_IDLE_H
+
+#include <linux/types.h>
+#include <linux/device.h>
+
+struct s390_idle_data {
+	unsigned int sequence;
+	unsigned long long idle_count;
+	unsigned long long idle_time;
+	unsigned long long clock_idle_enter;
+	unsigned long long clock_idle_exit;
+	unsigned long long timer_idle_enter;
+	unsigned long long timer_idle_exit;
+};
+
+extern struct device_attribute dev_attr_idle_count;
+extern struct device_attribute dev_attr_idle_time_us;
+
+#endif /* _S390_IDLE_H */

arch/s390/include/asm/ipl.h

@@ -89,12 +89,12 @@ extern u32 ipl_flags;
 extern u32 dump_prefix_page;
 
 struct dump_save_areas {
-	struct save_area **areas;
+	struct save_area_ext **areas;
 	int count;
 };
 
 extern struct dump_save_areas dump_save_areas;
-struct save_area *dump_save_area_create(int cpu);
+struct save_area_ext *dump_save_area_create(int cpu);
 
 extern void do_reipl(void);
 extern void do_halt(void);

arch/s390/include/asm/irq.h

@@ -51,6 +51,7 @@ enum interruption_class {
 	IRQEXT_CMS,
 	IRQEXT_CMC,
 	IRQEXT_CMR,
+	IRQEXT_FTP,
 	IRQIO_CIO,
 	IRQIO_QAI,
 	IRQIO_DAS,

arch/s390/include/asm/kprobes.h

@@ -84,6 +84,10 @@ int kprobe_fault_handler(struct pt_regs *regs, int trapnr);
 int kprobe_exceptions_notify(struct notifier_block *self,
 	unsigned long val, void *data);
 
+int probe_is_prohibited_opcode(u16 *insn);
+int probe_get_fixup_type(u16 *insn);
+int probe_is_insn_relative_long(u16 *insn);
+
 #define flush_insn_slot(p)	do { } while (0)
 
 #endif	/* _ASM_S390_KPROBES_H */

arch/s390/include/asm/lowcore.h

@@ -11,6 +11,7 @@
 #include <linux/types.h>
 #include <asm/ptrace.h>
 #include <asm/cpu.h>
+#include <asm/types.h>
 
 #ifdef CONFIG_32BIT
 
@@ -31,6 +32,11 @@ struct save_area {
 	u32	ctrl_regs[16];
 } __packed;
 
+struct save_area_ext {
+	struct save_area	sa;
+	__vector128		vx_regs[32];
+};
+
 struct _lowcore {
 	psw_t	restart_psw;			/* 0x0000 */
 	psw_t	restart_old_psw;		/* 0x0008 */
@@ -183,6 +189,11 @@ struct save_area {
 	u64	ctrl_regs[16];
 } __packed;
 
+struct save_area_ext {
+	struct save_area	sa;
+	__vector128		vx_regs[32];
+};
+
 struct _lowcore {
 	__u8	pad_0x0000[0x0014-0x0000];	/* 0x0000 */
 	__u32	ipl_parmblock_ptr;		/* 0x0014 */
@@ -310,7 +321,10 @@ struct _lowcore {
 
 	/* Extended facility list */
 	__u64	stfle_fac_list[32];		/* 0x0f00 */
-	__u8	pad_0x1000[0x11b8-0x1000];	/* 0x1000 */
+	__u8	pad_0x1000[0x11b0-0x1000];	/* 0x1000 */
+
+	/* Pointer to vector register save area */
+	__u64	vector_save_area_addr;		/* 0x11b0 */
 
 	/* 64 bit extparam used for pfault/diag 250: defined by architecture */
 	__u64	ext_params2;			/* 0x11B8 */
@@ -334,9 +348,10 @@ struct _lowcore {
 
 	/* Transaction abort diagnostic block */
 	__u8	pgm_tdb[256];			/* 0x1800 */
+	__u8	pad_0x1900[0x1c00-0x1900];	/* 0x1900 */
 
-	/* align to the top of the prefix area */
-	__u8	pad_0x1900[0x2000-0x1900];	/* 0x1900 */
+	/* Software defined save area for vector registers */
+	__u8	vector_save_area[1024];		/* 0x1c00 */
 } __packed;
 
 #endif /* CONFIG_32BIT */

arch/s390/include/asm/nmi.h

@@ -38,7 +38,7 @@ struct mci {
 	__u32 pm :  1; /* 22 psw program mask and cc validity */
 	__u32 ia :  1; /* 23 psw instruction address validity */
 	__u32 fa :  1; /* 24 failing storage address validity */
-	__u32	 :  1; /* 25 */
+	__u32 vr :  1; /* 25 vector register validity */
 	__u32 ec :  1; /* 26 external damage code validity */
 	__u32 fp :  1; /* 27 floating point register validity */
 	__u32 gr :  1; /* 28 general register validity */

arch/s390/include/asm/pgtable.h

@@ -217,7 +217,6 @@ extern unsigned long MODULES_END;
  */
 
 /* Hardware bits in the page table entry */
-#define _PAGE_CO	0x100		/* HW Change-bit override */
 #define _PAGE_PROTECT	0x200		/* HW read-only bit  */
 #define _PAGE_INVALID	0x400		/* HW invalid bit    */
 #define _PAGE_LARGE	0x800		/* Bit to mark a large pte */
@@ -234,8 +233,8 @@ extern unsigned long MODULES_END;
 #define __HAVE_ARCH_PTE_SPECIAL
 
 /* Set of bits not changed in pte_modify */
-#define _PAGE_CHG_MASK		(PAGE_MASK | _PAGE_SPECIAL | _PAGE_CO | \
-				 _PAGE_DIRTY | _PAGE_YOUNG)
+#define _PAGE_CHG_MASK		(PAGE_MASK | _PAGE_SPECIAL | _PAGE_DIRTY | \
+				 _PAGE_YOUNG)
 
 /*
  * handle_pte_fault uses pte_present, pte_none and pte_file to find out the
@@ -354,7 +353,6 @@ extern unsigned long MODULES_END;
 
 #define _REGION3_ENTRY_LARGE	0x400	/* RTTE-format control, large page  */
 #define _REGION3_ENTRY_RO	0x200	/* page protection bit		    */
-#define _REGION3_ENTRY_CO	0x100	/* change-recording override	    */
 
 /* Bits in the segment table entry */
 #define _SEGMENT_ENTRY_BITS	0xfffffffffffffe33UL
@@ -371,7 +369,6 @@ extern unsigned long MODULES_END;
 #define _SEGMENT_ENTRY_YOUNG	0x1000	/* SW segment young bit */
 #define _SEGMENT_ENTRY_SPLIT	0x0800	/* THP splitting bit */
 #define _SEGMENT_ENTRY_LARGE	0x0400	/* STE-format control, large page */
-#define _SEGMENT_ENTRY_CO	0x0100	/* change-recording override   */
 #define _SEGMENT_ENTRY_READ	0x0002	/* SW segment read bit */
 #define _SEGMENT_ENTRY_WRITE	0x0001	/* SW segment write bit */
 
@@ -873,8 +870,6 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
 		pgste = pgste_set_pte(ptep, pgste, entry);
 		pgste_set_unlock(ptep, pgste);
 	} else {
-		if (!(pte_val(entry) & _PAGE_INVALID) && MACHINE_HAS_EDAT1)
-			pte_val(entry) |= _PAGE_CO;
 		*ptep = entry;
 	}
 }
@@ -1044,6 +1039,22 @@ static inline void __ptep_ipte_local(unsigned long address, pte_t *ptep)
 		: "=m" (*ptep) : "m" (*ptep), "a" (pto), "a" (address));
 }
 
+static inline void __ptep_ipte_range(unsigned long address, int nr, pte_t *ptep)
+{
+	unsigned long pto = (unsigned long) ptep;
+
+#ifndef CONFIG_64BIT
+	/* pto in ESA mode must point to the start of the segment table */
+	pto &= 0x7ffffc00;
+#endif
+	/* Invalidate a range of ptes + global TLB flush of the ptes */
+	do {
+		asm volatile(
+			"	.insn rrf,0xb2210000,%2,%0,%1,0"
+			: "+a" (address), "+a" (nr) : "a" (pto) : "memory");
+	} while (nr != 255);
+}
+
 static inline void ptep_flush_direct(struct mm_struct *mm,
 				     unsigned long address, pte_t *ptep)
 {

arch/s390/include/asm/processor.h

@@ -13,9 +13,11 @@
 
 #define CIF_MCCK_PENDING	0	/* machine check handling is pending */
 #define CIF_ASCE		1	/* user asce needs fixup / uaccess */
+#define CIF_NOHZ_DELAY		2	/* delay HZ disable for a tick */
 
 #define _CIF_MCCK_PENDING	(1<<CIF_MCCK_PENDING)
 #define _CIF_ASCE		(1<<CIF_ASCE)
+#define _CIF_NOHZ_DELAY		(1<<CIF_NOHZ_DELAY)
 
 
 #ifndef __ASSEMBLY__
@@ -43,6 +45,8 @@ static inline int test_cpu_flag(int flag)
 	return !!(S390_lowcore.cpu_flags & (1U << flag));
 }
 
+#define arch_needs_cpu() test_cpu_flag(CIF_NOHZ_DELAY)
+
 /*
  * Default implementation of macro that returns current
  * instruction pointer ("program counter").
@@ -113,6 +117,7 @@ struct thread_struct {
 	int ri_signum;
 #ifdef CONFIG_64BIT
 	unsigned char trap_tdb[256];	/* Transaction abort diagnose block */
+	__vector128 *vxrs;		/* Vector register save area */
 #endif
 };
 
@@ -285,7 +290,12 @@ static inline unsigned long __rewind_psw(psw_t psw, unsigned long ilc)
 	return (psw.addr - ilc) & mask;
 #endif
 }
- 
+
+/*
+ * Function to stop a processor until the next interrupt occurs
+ */
+void enabled_wait(void);
+
 /*
  * Function to drop a processor into disabled wait state
  */

arch/s390/include/asm/ptrace.h

@@ -161,6 +161,12 @@ static inline long regs_return_value(struct pt_regs *regs)
 	return regs->gprs[2];
 }
 
+static inline void instruction_pointer_set(struct pt_regs *regs,
+					   unsigned long val)
+{
+	regs->psw.addr = val | PSW_ADDR_AMODE;
+}
+
 int regs_query_register_offset(const char *name);
 const char *regs_query_register_name(unsigned int offset);
 unsigned long regs_get_register(struct pt_regs *regs, unsigned int offset);

arch/s390/include/asm/setup.h

@@ -55,8 +55,8 @@ extern void detect_memory_memblock(void);
 #define MACHINE_FLAG_LPP	(1UL << 13)
 #define MACHINE_FLAG_TOPOLOGY	(1UL << 14)
 #define MACHINE_FLAG_TE		(1UL << 15)
-#define MACHINE_FLAG_RRBM	(1UL << 16)
 #define MACHINE_FLAG_TLB_LC	(1UL << 17)
+#define MACHINE_FLAG_VX		(1UL << 18)
 
 #define MACHINE_IS_VM		(S390_lowcore.machine_flags & MACHINE_FLAG_VM)
 #define MACHINE_IS_KVM		(S390_lowcore.machine_flags & MACHINE_FLAG_KVM)
@@ -78,8 +78,8 @@ extern void detect_memory_memblock(void);
 #define MACHINE_HAS_LPP		(0)
 #define MACHINE_HAS_TOPOLOGY	(0)
 #define MACHINE_HAS_TE		(0)
-#define MACHINE_HAS_RRBM	(0)
 #define MACHINE_HAS_TLB_LC	(0)
+#define MACHINE_HAS_VX		(0)
 #else /* CONFIG_64BIT */
 #define MACHINE_HAS_IEEE	(1)
 #define MACHINE_HAS_CSP		(1)
@@ -91,8 +91,8 @@ extern void detect_memory_memblock(void);
 #define MACHINE_HAS_LPP		(S390_lowcore.machine_flags & MACHINE_FLAG_LPP)
 #define MACHINE_HAS_TOPOLOGY	(S390_lowcore.machine_flags & MACHINE_FLAG_TOPOLOGY)
 #define MACHINE_HAS_TE		(S390_lowcore.machine_flags & MACHINE_FLAG_TE)
-#define MACHINE_HAS_RRBM	(S390_lowcore.machine_flags & MACHINE_FLAG_RRBM)
 #define MACHINE_HAS_TLB_LC	(S390_lowcore.machine_flags & MACHINE_FLAG_TLB_LC)
+#define MACHINE_HAS_VX		(S390_lowcore.machine_flags & MACHINE_FLAG_VX)
 #endif /* CONFIG_64BIT */
 
 /*

arch/s390/include/asm/sigp.h

@@ -15,6 +15,7 @@
 #define SIGP_SET_ARCHITECTURE	     18
 #define SIGP_COND_EMERGENCY_SIGNAL   19
 #define SIGP_SENSE_RUNNING	     21
+#define SIGP_STORE_ADDITIONAL_STATUS 23
 
 /* SIGP condition codes */
 #define SIGP_CC_ORDER_CODE_ACCEPTED 0
@@ -33,9 +34,10 @@
 
 #ifndef __ASSEMBLY__
 
-static inline int __pcpu_sigp(u16 addr, u8 order, u32 parm, u32 *status)
+static inline int __pcpu_sigp(u16 addr, u8 order, unsigned long parm,
+			      u32 *status)
 {
-	register unsigned int reg1 asm ("1") = parm;
+	register unsigned long reg1 asm ("1") = parm;
 	int cc;
 
 	asm volatile(

arch/s390/include/asm/smp.h

@@ -29,7 +29,6 @@ extern int smp_find_processor_id(u16 address);
 extern int smp_store_status(int cpu);
 extern int smp_vcpu_scheduled(int cpu);
 extern void smp_yield_cpu(int cpu);
-extern void smp_yield(void);
 extern void smp_cpu_set_polarization(int cpu, int val);
 extern int smp_cpu_get_polarization(int cpu);
 extern void smp_fill_possible_mask(void);
@@ -50,7 +49,6 @@ static inline int smp_find_processor_id(u16 address) { return 0; }
 static inline int smp_store_status(int cpu) { return 0; }
 static inline int smp_vcpu_scheduled(int cpu) { return 1; }
 static inline void smp_yield_cpu(int cpu) { }
-static inline void smp_yield(void) { }
 static inline void smp_fill_possible_mask(void) { }
 
 #endif /* CONFIG_SMP */

arch/s390/include/asm/spinlock.h

@@ -37,11 +37,17 @@ _raw_compare_and_swap(unsigned int *lock, unsigned int old, unsigned int new)
  * (the type definitions are in asm/spinlock_types.h)
  */
 
+void arch_lock_relax(unsigned int cpu);
+
 void arch_spin_lock_wait(arch_spinlock_t *);
 int arch_spin_trylock_retry(arch_spinlock_t *);
-void arch_spin_relax(arch_spinlock_t *);
 void arch_spin_lock_wait_flags(arch_spinlock_t *, unsigned long flags);
 
+static inline void arch_spin_relax(arch_spinlock_t *lock)
+{
+	arch_lock_relax(lock->lock);
+}
+
 static inline u32 arch_spin_lockval(int cpu)
 {
 	return ~cpu;
@@ -64,11 +70,6 @@ static inline int arch_spin_trylock_once(arch_spinlock_t *lp)
 		      _raw_compare_and_swap(&lp->lock, 0, SPINLOCK_LOCKVAL));
 }
 
-static inline int arch_spin_tryrelease_once(arch_spinlock_t *lp)
-{
-	return _raw_compare_and_swap(&lp->lock, SPINLOCK_LOCKVAL, 0);
-}
-
 static inline void arch_spin_lock(arch_spinlock_t *lp)
 {
 	if (!arch_spin_trylock_once(lp))
@@ -91,7 +92,13 @@ static inline int arch_spin_trylock(arch_spinlock_t *lp)
 
 static inline void arch_spin_unlock(arch_spinlock_t *lp)
 {
-	arch_spin_tryrelease_once(lp);
+	typecheck(unsigned int, lp->lock);
+	asm volatile(
+		__ASM_BARRIER
+		"st	%1,%0\n"
+		: "+Q" (lp->lock)
+		: "d" (0)
+		: "cc", "memory");
 }
 
 static inline void arch_spin_unlock_wait(arch_spinlock_t *lock)
@@ -123,13 +130,12 @@ static inline void arch_spin_unlock_wait(arch_spinlock_t *lock)
  */
 #define arch_write_can_lock(x) ((x)->lock == 0)
 
-extern void _raw_read_lock_wait(arch_rwlock_t *lp);
-extern void _raw_read_lock_wait_flags(arch_rwlock_t *lp, unsigned long flags);
 extern int _raw_read_trylock_retry(arch_rwlock_t *lp);
-extern void _raw_write_lock_wait(arch_rwlock_t *lp);
-extern void _raw_write_lock_wait_flags(arch_rwlock_t *lp, unsigned long flags);
 extern int _raw_write_trylock_retry(arch_rwlock_t *lp);
 
+#define arch_read_lock_flags(lock, flags) arch_read_lock(lock)
+#define arch_write_lock_flags(lock, flags) arch_write_lock(lock)
+
 static inline int arch_read_trylock_once(arch_rwlock_t *rw)
 {
 	unsigned int old = ACCESS_ONCE(rw->lock);
@@ -144,16 +150,82 @@ static inline int arch_write_trylock_once(arch_rwlock_t *rw)
 		      _raw_compare_and_swap(&rw->lock, 0, 0x80000000));
 }
 
+#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES
+
+#define __RAW_OP_OR	"lao"
+#define __RAW_OP_AND	"lan"
+#define __RAW_OP_ADD	"laa"
+
+#define __RAW_LOCK(ptr, op_val, op_string)		\
+({							\
+	unsigned int old_val;				\
+							\
+	typecheck(unsigned int *, ptr);			\
+	asm volatile(					\
+		op_string "	%0,%2,%1\n"		\
+		"bcr	14,0\n"				\
+		: "=d" (old_val), "+Q" (*ptr)		\
+		: "d" (op_val)				\
+		: "cc", "memory");			\
+	old_val;					\
+})
+
+#define __RAW_UNLOCK(ptr, op_val, op_string)		\
+({							\
+	unsigned int old_val;				\
+							\
+	typecheck(unsigned int *, ptr);			\
+	asm volatile(					\
+		"bcr	14,0\n"				\
+		op_string "	%0,%2,%1\n"		\
+		: "=d" (old_val), "+Q" (*ptr)		\
+		: "d" (op_val)				\
+		: "cc", "memory");			\
+	old_val;					\
+})
+
+extern void _raw_read_lock_wait(arch_rwlock_t *lp);
+extern void _raw_write_lock_wait(arch_rwlock_t *lp, unsigned int prev);
+
 static inline void arch_read_lock(arch_rwlock_t *rw)
 {
-	if (!arch_read_trylock_once(rw))
+	unsigned int old;
+
+	old = __RAW_LOCK(&rw->lock, 1, __RAW_OP_ADD);
+	if ((int) old < 0)
 		_raw_read_lock_wait(rw);
 }
 
-static inline void arch_read_lock_flags(arch_rwlock_t *rw, unsigned long flags)
+static inline void arch_read_unlock(arch_rwlock_t *rw)
+{
+	__RAW_UNLOCK(&rw->lock, -1, __RAW_OP_ADD);
+}
+
+static inline void arch_write_lock(arch_rwlock_t *rw)
+{
+	unsigned int old;
+
+	old = __RAW_LOCK(&rw->lock, 0x80000000, __RAW_OP_OR);
+	if (old != 0)
+		_raw_write_lock_wait(rw, old);
+	rw->owner = SPINLOCK_LOCKVAL;
+}
+
+static inline void arch_write_unlock(arch_rwlock_t *rw)
+{
+	rw->owner = 0;
+	__RAW_UNLOCK(&rw->lock, 0x7fffffff, __RAW_OP_AND);
+}
+
+#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */
+
+extern void _raw_read_lock_wait(arch_rwlock_t *lp);
+extern void _raw_write_lock_wait(arch_rwlock_t *lp);
+
+static inline void arch_read_lock(arch_rwlock_t *rw)
 {
 	if (!arch_read_trylock_once(rw))
-		_raw_read_lock_wait_flags(rw, flags);
+		_raw_read_lock_wait(rw);
 }
 
 static inline void arch_read_unlock(arch_rwlock_t *rw)
@@ -169,19 +241,24 @@ static inline void arch_write_lock(arch_rwlock_t *rw)
 {
 	if (!arch_write_trylock_once(rw))
 		_raw_write_lock_wait(rw);
-}
-
-static inline void arch_write_lock_flags(arch_rwlock_t *rw, unsigned long flags)
-{
-	if (!arch_write_trylock_once(rw))
-		_raw_write_lock_wait_flags(rw, flags);
+	rw->owner = SPINLOCK_LOCKVAL;
 }
 
 static inline void arch_write_unlock(arch_rwlock_t *rw)
 {
-	_raw_compare_and_swap(&rw->lock, 0x80000000, 0);
+	typecheck(unsigned int, rw->lock);
+
+	rw->owner = 0;
+	asm volatile(
+		__ASM_BARRIER
+		"st	%1,%0\n"
+		: "+Q" (rw->lock)
+		: "d" (0)
+		: "cc", "memory");
 }
 
+#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */
+
 static inline int arch_read_trylock(arch_rwlock_t *rw)
 {
 	if (!arch_read_trylock_once(rw))
@@ -191,12 +268,20 @@ static inline int arch_read_trylock(arch_rwlock_t *rw)
 
 static inline int arch_write_trylock(arch_rwlock_t *rw)
 {
-	if (!arch_write_trylock_once(rw))
-		return _raw_write_trylock_retry(rw);
+	if (!arch_write_trylock_once(rw) && !_raw_write_trylock_retry(rw))
+		return 0;
+	rw->owner = SPINLOCK_LOCKVAL;
 	return 1;
 }
 
-#define arch_read_relax(lock)	cpu_relax()
-#define arch_write_relax(lock)	cpu_relax()
+static inline void arch_read_relax(arch_rwlock_t *rw)
+{
+	arch_lock_relax(rw->owner);
+}
+
+static inline void arch_write_relax(arch_rwlock_t *rw)
+{
+	arch_lock_relax(rw->owner);
+}
 
 #endif /* __ASM_SPINLOCK_H */

arch/s390/include/asm/spinlock_types.h

@@ -13,6 +13,7 @@ typedef struct {
 
 typedef struct {
 	unsigned int lock;
+	unsigned int owner;
 } arch_rwlock_t;
 
 #define __ARCH_RW_LOCK_UNLOCKED		{ 0 }

arch/s390/include/asm/switch_to.h

@@ -103,6 +103,61 @@ static inline void restore_fp_regs(freg_t *fprs)
 	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)
+{
+#ifdef CONFIG_64BIT
+	if (task->thread.vxrs)
+		save_vx_regs(task->thread.vxrs);
+	else
+#endif
+	save_fp_regs(task->thread.fp_regs.fprs);
+}
+
+static inline void restore_fp_vx_regs(struct task_struct *task)
+{
+#ifdef CONFIG_64BIT
+	if (task->thread.vxrs)
+		restore_vx_regs(task->thread.vxrs);
+	else
+#endif
+	restore_fp_regs(task->thread.fp_regs.fprs);
+}
+
 static inline void save_access_regs(unsigned int *acrs)
 {
 	typedef struct { int _[NUM_ACRS]; } acrstype;
@@ -120,16 +175,16 @@ static inline void restore_access_regs(unsigned int *acrs)
 #define switch_to(prev,next,last) do {					\
 	if (prev->mm) {							\
 		save_fp_ctl(&prev->thread.fp_regs.fpc);			\
-		save_fp_regs(prev->thread.fp_regs.fprs);		\
+		save_fp_vx_regs(prev);					\
 		save_access_regs(&prev->thread.acrs[0]);		\
 		save_ri_cb(prev->thread.ri_cb);				\
 	}								\
 	if (next->mm) {							\
+		update_cr_regs(next);					\
 		restore_fp_ctl(&next->thread.fp_regs.fpc);		\
-		restore_fp_regs(next->thread.fp_regs.fprs);		\
+		restore_fp_vx_regs(next);				\
 		restore_access_regs(&next->thread.acrs[0]);		\
 		restore_ri_cb(next->thread.ri_cb, prev->thread.ri_cb);	\
-		update_cr_regs(next);					\
 	}								\
 	prev = __switch_to(prev,next);					\
 } while (0)

arch/s390/include/asm/thread_info.h

@@ -84,11 +84,13 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_SYSCALL_AUDIT	4	/* syscall auditing active */
 #define TIF_SECCOMP		5	/* secure computing */
 #define TIF_SYSCALL_TRACEPOINT	6	/* syscall tracepoint instrumentation */
+#define TIF_UPROBE		7	/* breakpointed or single-stepping */
 #define TIF_31BIT		16	/* 32bit process */
 #define TIF_MEMDIE		17	/* is terminating due to OOM killer */
 #define TIF_RESTORE_SIGMASK	18	/* restore signal mask in do_signal() */
 #define TIF_SINGLE_STEP		19	/* This task is single stepped */
 #define TIF_BLOCK_STEP		20	/* This task is block stepped */
+#define TIF_UPROBE_SINGLESTEP	21	/* This task is uprobe single stepped */
 
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
@@ -97,6 +99,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
 #define _TIF_SECCOMP		(1<<TIF_SECCOMP)
 #define _TIF_SYSCALL_TRACEPOINT	(1<<TIF_SYSCALL_TRACEPOINT)
+#define _TIF_UPROBE		(1<<TIF_UPROBE)
 #define _TIF_31BIT		(1<<TIF_31BIT)
 #define _TIF_SINGLE_STEP	(1<<TIF_SINGLE_STEP)
 

arch/s390/include/asm/uprobes.h

+/*
+ *    User-space Probes (UProbes) for s390
+ *
+ *    Copyright IBM Corp. 2014
+ *    Author(s): Jan Willeke,
+ */
+
+#ifndef _ASM_UPROBES_H
+#define _ASM_UPROBES_H
+
+#include <linux/notifier.h>
+
+typedef u16 uprobe_opcode_t;
+
+#define UPROBE_XOL_SLOT_BYTES	256 /* cache aligned */
+
+#define UPROBE_SWBP_INSN	0x0002
+#define UPROBE_SWBP_INSN_SIZE	2
+
+struct arch_uprobe {
+	union{
+		uprobe_opcode_t insn[3];
+		uprobe_opcode_t ixol[3];
+	};
+	unsigned int saved_per : 1;
+	unsigned int saved_int_code;
+};
+
+struct arch_uprobe_task {
+};
+
+int arch_uprobe_analyze_insn(struct arch_uprobe *aup, struct mm_struct *mm,
+			     unsigned long addr);
+int arch_uprobe_pre_xol(struct arch_uprobe *aup, struct pt_regs *regs);
+int arch_uprobe_post_xol(struct arch_uprobe *aup, struct pt_regs *regs);
+bool arch_uprobe_xol_was_trapped(struct task_struct *tsk);
+int arch_uprobe_exception_notify(struct notifier_block *self, unsigned long val,
+				 void *data);
+void arch_uprobe_abort_xol(struct arch_uprobe *ap, struct pt_regs *regs);
+unsigned long arch_uretprobe_hijack_return_addr(unsigned long trampoline,
+						struct pt_regs *regs);
+#endif	/* _ASM_UPROBES_H */

arch/s390/include/asm/vdso.h

@@ -22,13 +22,17 @@ struct vdso_data {
 	__u64 xtime_tod_stamp;		/* TOD clock for xtime		0x08 */
 	__u64 xtime_clock_sec;		/* Kernel time			0x10 */
 	__u64 xtime_clock_nsec;		/*				0x18 */
-	__u64 wtom_clock_sec;		/* Wall to monotonic clock	0x20 */
-	__u64 wtom_clock_nsec;		/*				0x28 */
-	__u32 tz_minuteswest;		/* Minutes west of Greenwich	0x30 */
-	__u32 tz_dsttime;		/* Type of dst correction	0x34 */
-	__u32 ectg_available;		/* ECTG instruction present	0x38 */
-	__u32 tk_mult;			/* Mult. used for xtime_nsec	0x3c */
-	__u32 tk_shift;			/* Shift used for xtime_nsec	0x40 */
+	__u64 xtime_coarse_sec;		/* Coarse kernel time		0x20 */
+	__u64 xtime_coarse_nsec;	/*				0x28 */
+	__u64 wtom_clock_sec;		/* Wall to monotonic clock	0x30 */
+	__u64 wtom_clock_nsec;		/*				0x38 */
+	__u64 wtom_coarse_sec;		/* Coarse wall to monotonic	0x40 */
+	__u64 wtom_coarse_nsec;		/*				0x48 */
+	__u32 tz_minuteswest;		/* Minutes west of Greenwich	0x50 */
+	__u32 tz_dsttime;		/* Type of dst correction	0x54 */
+	__u32 ectg_available;		/* ECTG instruction present	0x58 */
+	__u32 tk_mult;			/* Mult. used for xtime_nsec	0x5c */
+	__u32 tk_shift;			/* Shift used for xtime_nsec	0x60 */
 };
 
 struct vdso_per_cpu_data {

arch/s390/include/asm/vtimer.h

@@ -28,6 +28,4 @@ extern int del_virt_timer(struct vtimer_list *timer);
 extern void init_cpu_vtimer(void);
 extern void vtime_init(void);
 
-extern void vtime_stop_cpu(void);
-
 #endif /* _ASM_S390_TIMER_H */

arch/s390/include/uapi/asm/sigcontext.h

@@ -7,10 +7,14 @@
 #define _ASM_S390_SIGCONTEXT_H
 
 #include <linux/compiler.h>
+#include <linux/types.h>
 
-#define __NUM_GPRS 16
-#define __NUM_FPRS 16
-#define __NUM_ACRS 16
+#define __NUM_GPRS		16
+#define __NUM_FPRS		16
+#define __NUM_ACRS		16
+#define __NUM_VXRS		32
+#define __NUM_VXRS_LOW		16
+#define __NUM_VXRS_HIGH		16
 
 #ifndef __s390x__
 
@@ -59,6 +63,16 @@ typedef struct
 	_s390_fp_regs     fpregs;
 } _sigregs;
 
+typedef struct
+{
+#ifndef __s390x__
+	unsigned long gprs_high[__NUM_GPRS];
+#endif
+	unsigned long long vxrs_low[__NUM_VXRS_LOW];
+	__vector128 vxrs_high[__NUM_VXRS_HIGH];
+	unsigned char __reserved[128];
+} _sigregs_ext;
+
 struct sigcontext
 {
 	unsigned long	oldmask[_SIGCONTEXT_NSIG_WORDS];

arch/s390/include/uapi/asm/types.h

@@ -17,6 +17,10 @@
 typedef unsigned long addr_t; 
 typedef __signed__ long saddr_t;
 
+typedef struct {
+	__u32 u[4];
+} __vector128;
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* _UAPI_S390_TYPES_H */

arch/s390/include/uapi/asm/ucontext.h

@@ -7,10 +7,15 @@
 #ifndef _ASM_S390_UCONTEXT_H
 #define _ASM_S390_UCONTEXT_H
 
-#define UC_EXTENDED	0x00000001
-
-#ifndef __s390x__
+#define UC_GPRS_HIGH	1	/* uc_mcontext_ext has valid high gprs */
+#define UC_VXRS		2	/* uc_mcontext_ext has valid vector regs */
 
+/*
+ * The struct ucontext_extended describes how the registers are stored
+ * on a rt signal frame. Please note that the structure is not fixed,
+ * if new CPU registers are added to the user state the size of the
+ * struct ucontext_extended will increase.
+ */
 struct ucontext_extended {
 	unsigned long	  uc_flags;
 	struct ucontext  *uc_link;
@@ -19,11 +24,9 @@ struct ucontext_extended {
 	sigset_t	  uc_sigmask;
 	/* Allow for uc_sigmask growth.  Glibc uses a 1024-bit sigset_t.  */
 	unsigned char	  __unused[128 - sizeof(sigset_t)];
-	unsigned long	  uc_gprs_high[16];
+	_sigregs_ext	  uc_mcontext_ext;
 };
 
-#endif
-
 struct ucontext {
 	unsigned long	  uc_flags;
 	struct ucontext  *uc_link;

arch/s390/kernel/Makefile

@@ -28,7 +28,7 @@ CFLAGS_ptrace.o		+= -DUTS_MACHINE='"$(UTS_MACHINE)"'
 
 CFLAGS_sysinfo.o += -Iinclude/math-emu -Iarch/s390/math-emu -w
 
-obj-y	:= traps.o time.o process.o base.o early.o setup.o vtime.o
+obj-y	:= traps.o time.o process.o base.o early.o setup.o idle.o vtime.o
 obj-y	+= processor.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o nmi.o
 obj-y	+= debug.o irq.o ipl.o dis.o diag.o sclp.o vdso.o
 obj-y	+= sysinfo.o jump_label.o lgr.o os_info.o machine_kexec.o pgm_check.o
@@ -52,11 +52,9 @@ obj-$(CONFIG_COMPAT)		+= compat_wrapper.o $(compat-obj-y)
 
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o
-obj-$(CONFIG_FUNCTION_TRACER)	+= $(if $(CONFIG_64BIT),mcount64.o,mcount.o)
-obj-$(CONFIG_DYNAMIC_FTRACE)	+= ftrace.o
-obj-$(CONFIG_FUNCTION_GRAPH_TRACER) += ftrace.o
-obj-$(CONFIG_FTRACE_SYSCALLS)  += ftrace.o
+obj-$(CONFIG_FUNCTION_TRACER)	+= mcount.o ftrace.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump.o
+obj-$(CONFIG_UPROBES)		+= uprobes.o
 
 ifdef CONFIG_64BIT
 obj-$(CONFIG_PERF_EVENTS)	+= perf_event.o perf_cpum_cf.o perf_cpum_sf.o \

arch/s390/kernel/asm-offsets.c

@@ -9,7 +9,7 @@
 #include <linux/kbuild.h>
 #include <linux/kvm_host.h>
 #include <linux/sched.h>
-#include <asm/cputime.h>
+#include <asm/idle.h>
 #include <asm/vdso.h>
 #include <asm/pgtable.h>
 
@@ -62,8 +62,12 @@ int main(void)
 	DEFINE(__VDSO_XTIME_STAMP, offsetof(struct vdso_data, xtime_tod_stamp));
 	DEFINE(__VDSO_XTIME_SEC, offsetof(struct vdso_data, xtime_clock_sec));
 	DEFINE(__VDSO_XTIME_NSEC, offsetof(struct vdso_data, xtime_clock_nsec));
+	DEFINE(__VDSO_XTIME_CRS_SEC, offsetof(struct vdso_data, xtime_coarse_sec));
+	DEFINE(__VDSO_XTIME_CRS_NSEC, offsetof(struct vdso_data, xtime_coarse_nsec));
 	DEFINE(__VDSO_WTOM_SEC, offsetof(struct vdso_data, wtom_clock_sec));
 	DEFINE(__VDSO_WTOM_NSEC, offsetof(struct vdso_data, wtom_clock_nsec));
+	DEFINE(__VDSO_WTOM_CRS_SEC, offsetof(struct vdso_data, wtom_coarse_sec));
+	DEFINE(__VDSO_WTOM_CRS_NSEC, offsetof(struct vdso_data, wtom_coarse_nsec));
 	DEFINE(__VDSO_TIMEZONE, offsetof(struct vdso_data, tz_minuteswest));
 	DEFINE(__VDSO_ECTG_OK, offsetof(struct vdso_data, ectg_available));
 	DEFINE(__VDSO_TK_MULT, offsetof(struct vdso_data, tk_mult));
@@ -73,8 +77,11 @@ int main(void)
 	/* constants used by the vdso */
 	DEFINE(__CLOCK_REALTIME, CLOCK_REALTIME);
 	DEFINE(__CLOCK_MONOTONIC, CLOCK_MONOTONIC);
+	DEFINE(__CLOCK_REALTIME_COARSE, CLOCK_REALTIME_COARSE);
+	DEFINE(__CLOCK_MONOTONIC_COARSE, CLOCK_MONOTONIC_COARSE);
 	DEFINE(__CLOCK_THREAD_CPUTIME_ID, CLOCK_THREAD_CPUTIME_ID);
 	DEFINE(__CLOCK_REALTIME_RES, MONOTONIC_RES_NSEC);
+	DEFINE(__CLOCK_COARSE_RES, LOW_RES_NSEC);
 	BLANK();
 	/* idle data offsets */
 	DEFINE(__CLOCK_IDLE_ENTER, offsetof(struct s390_idle_data, clock_idle_enter));

arch/s390/kernel/compat_linux.h

@@ -50,6 +50,14 @@ typedef struct
 	_s390_fp_regs32     fpregs;
 } _sigregs32;
 
+typedef struct
+{
+	__u32 gprs_high[__NUM_GPRS];
+	__u64 vxrs_low[__NUM_VXRS_LOW];
+	__vector128 vxrs_high[__NUM_VXRS_HIGH];
+	__u8 __reserved[128];
+} _sigregs_ext32;
+
 #define _SIGCONTEXT_NSIG32	64
 #define _SIGCONTEXT_NSIG_BPW32	32
 #define __SIGNAL_FRAMESIZE32	96
@@ -72,6 +80,7 @@ struct ucontext32 {
 	compat_sigset_t		uc_sigmask;
 	/* Allow for uc_sigmask growth.  Glibc uses a 1024-bit sigset_t.  */
 	unsigned char		__unused[128 - sizeof(compat_sigset_t)];
+	_sigregs_ext32		uc_mcontext_ext;
 };
 
 struct stat64_emu31;

arch/s390/kernel/compat_signal.c

@@ -36,17 +36,16 @@ typedef struct
 	struct sigcontext32 sc;
 	_sigregs32 sregs;
 	int signo;
-	__u32 gprs_high[NUM_GPRS];
-	__u8 retcode[S390_SYSCALL_SIZE];
+	_sigregs_ext32 sregs_ext;
+	__u16 svc_insn;		/* Offset of svc_insn is NOT fixed! */
 } sigframe32;
 
 typedef struct 
 {
 	__u8 callee_used_stack[__SIGNAL_FRAMESIZE32];
-	__u8 retcode[S390_SYSCALL_SIZE];
+	__u16 svc_insn;
 	compat_siginfo_t info;
 	struct ucontext32 uc;
-	__u32 gprs_high[NUM_GPRS];
 } rt_sigframe32;
 
 int copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from)
@@ -151,6 +150,38 @@ int copy_siginfo_from_user32(siginfo_t *to, compat_siginfo_t __user *from)
 	return err ? -EFAULT : 0;
 }
 
+/* Store registers needed to create the signal frame */
+static void store_sigregs(void)
+{
+	int i;
+
+	save_access_regs(current->thread.acrs);
+	save_fp_ctl(&current->thread.fp_regs.fpc);
+	if (current->thread.vxrs) {
+		save_vx_regs(current->thread.vxrs);
+		for (i = 0; i < __NUM_FPRS; i++)
+			current->thread.fp_regs.fprs[i] =
+				*(freg_t *)(current->thread.vxrs + i);
+	} else
+		save_fp_regs(current->thread.fp_regs.fprs);
+}
+
+/* Load registers after signal return */
+static void load_sigregs(void)
+{
+	int i;
+
+	restore_access_regs(current->thread.acrs);
+	/* restore_fp_ctl is done in restore_sigregs */
+	if (current->thread.vxrs) {
+		for (i = 0; i < __NUM_FPRS; i++)
+			*(freg_t *)(current->thread.vxrs + i) =
+				current->thread.fp_regs.fprs[i];
+		restore_vx_regs(current->thread.vxrs);
+	} else
+		restore_fp_regs(current->thread.fp_regs.fprs);
+}
+
 static int save_sigregs32(struct pt_regs *regs, _sigregs32 __user *sregs)
 {
 	_sigregs32 user_sregs;
@@ -163,11 +194,8 @@ static int save_sigregs32(struct pt_regs *regs, _sigregs32 __user *sregs)
 		(__u32)(regs->psw.mask & PSW_MASK_BA);
 	for (i = 0; i < NUM_GPRS; i++)
 		user_sregs.regs.gprs[i] = (__u32) regs->gprs[i];
-	save_access_regs(current->thread.acrs);
 	memcpy(&user_sregs.regs.acrs, current->thread.acrs,
 	       sizeof(user_sregs.regs.acrs));
-	save_fp_ctl(&current->thread.fp_regs.fpc);
-	save_fp_regs(current->thread.fp_regs.fprs);
 	memcpy(&user_sregs.fpregs, &current->thread.fp_regs,
 	       sizeof(user_sregs.fpregs));
 	if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs32)))
@@ -207,37 +235,67 @@ 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));
-	restore_access_regs(current->thread.acrs);
 
 	memcpy(&current->thread.fp_regs, &user_sregs.fpregs,
 	       sizeof(current->thread.fp_regs));
 
-	restore_fp_regs(current->thread.fp_regs.fprs);
 	clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */
 	return 0;
 }
 
-static int save_sigregs_gprs_high(struct pt_regs *regs, __u32 __user *uregs)
+static int save_sigregs_ext32(struct pt_regs *regs,
+			      _sigregs_ext32 __user *sregs_ext)
 {
 	__u32 gprs_high[NUM_GPRS];
+	__u64 vxrs[__NUM_VXRS_LOW];
 	int i;
 
+	/* Save high gprs to signal stack */
 	for (i = 0; i < NUM_GPRS; i++)
 		gprs_high[i] = regs->gprs[i] >> 32;
-	if (__copy_to_user(uregs, &gprs_high, sizeof(gprs_high)))
+	if (__copy_to_user(&sregs_ext->gprs_high, &gprs_high,
+			   sizeof(sregs_ext->gprs_high)))
 		return -EFAULT;
+
+	/* Save vector registers to signal stack */
+	if (current->thread.vxrs) {
+		for (i = 0; i < __NUM_VXRS_LOW; i++)
+			vxrs[i] = *((__u64 *)(current->thread.vxrs + i) + 1);
+		if (__copy_to_user(&sregs_ext->vxrs_low, vxrs,
+				   sizeof(sregs_ext->vxrs_low)) ||
+		    __copy_to_user(&sregs_ext->vxrs_high,
+				   current->thread.vxrs + __NUM_VXRS_LOW,
+				   sizeof(sregs_ext->vxrs_high)))
+			return -EFAULT;
+	}
 	return 0;
 }
 
-static int restore_sigregs_gprs_high(struct pt_regs *regs, __u32 __user *uregs)
+static int restore_sigregs_ext32(struct pt_regs *regs,
+				 _sigregs_ext32 __user *sregs_ext)
 {
 	__u32 gprs_high[NUM_GPRS];
+	__u64 vxrs[__NUM_VXRS_LOW];
 	int i;
 
-	if (__copy_from_user(&gprs_high, uregs, sizeof(gprs_high)))
+	/* Restore high gprs from signal stack */
+	if (__copy_from_user(&gprs_high, &sregs_ext->gprs_high,
+			     sizeof(&sregs_ext->gprs_high)))
 		return -EFAULT;
 	for (i = 0; i < NUM_GPRS; i++)
 		*(__u32 *)&regs->gprs[i] = gprs_high[i];
+
+	/* Restore vector registers from signal stack */
+	if (current->thread.vxrs) {
+		if (__copy_from_user(vxrs, &sregs_ext->vxrs_low,
+				     sizeof(sregs_ext->vxrs_low)) ||
+		    __copy_from_user(current->thread.vxrs + __NUM_VXRS_LOW,
+				     &sregs_ext->vxrs_high,
+				     sizeof(sregs_ext->vxrs_high)))
+			return -EFAULT;
+		for (i = 0; i < __NUM_VXRS_LOW; i++)
+			*((__u64 *)(current->thread.vxrs + i) + 1) = vxrs[i];
+	}
 	return 0;
 }
 
@@ -252,8 +310,9 @@ COMPAT_SYSCALL_DEFINE0(sigreturn)
 	set_current_blocked(&set);
 	if (restore_sigregs32(regs, &frame->sregs))
 		goto badframe;
-	if (restore_sigregs_gprs_high(regs, frame->gprs_high))
+	if (restore_sigregs_ext32(regs, &frame->sregs_ext))
 		goto badframe;
+	load_sigregs();
 	return regs->gprs[2];
 badframe:
 	force_sig(SIGSEGV, current);
@@ -269,12 +328,13 @@ COMPAT_SYSCALL_DEFINE0(rt_sigreturn)
 	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
 		goto badframe;
 	set_current_blocked(&set);
+	if (compat_restore_altstack(&frame->uc.uc_stack))
+		goto badframe;
 	if (restore_sigregs32(regs, &frame->uc.uc_mcontext))
 		goto badframe;
-	if (restore_sigregs_gprs_high(regs, frame->gprs_high))
+	if (restore_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext))
 		goto badframe;
-	if (compat_restore_altstack(&frame->uc.uc_stack))
-		goto badframe; 
+	load_sigregs();
 	return regs->gprs[2];
 badframe:
 	force_sig(SIGSEGV, current);
@@ -324,37 +384,64 @@ static int setup_frame32(struct ksignal *ksig, sigset_t *set,
 			 struct pt_regs *regs)
 {
 	int sig = ksig->sig;
-	sigframe32 __user *frame = get_sigframe(&ksig->ka, regs, sizeof(sigframe32));
-
+	sigframe32 __user *frame;
+	struct sigcontext32 sc;
+	unsigned long restorer;
+	size_t frame_size;
+
+	/*
+	 * gprs_high are always present for 31-bit compat tasks.
+	 * The space for vector registers is only allocated if
+	 * the machine supports it
+	 */
+	frame_size = sizeof(*frame) - sizeof(frame->sregs_ext.__reserved);
+	if (!MACHINE_HAS_VX)
+		frame_size -= sizeof(frame->sregs_ext.vxrs_low) +
+			      sizeof(frame->sregs_ext.vxrs_high);
+	frame = get_sigframe(&ksig->ka, regs, frame_size);
 	if (frame == (void __user *) -1UL)
 		return -EFAULT;
 
-	if (__copy_to_user(&frame->sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE32))
+	/* Set up backchain. */
+	if (__put_user(regs->gprs[15], (unsigned int __user *) frame))
+		return -EFAULT;
+
+	/* Create struct sigcontext32 on the signal stack */
+	memcpy(&sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE32);
+	sc.sregs = (__u32)(unsigned long __force) &frame->sregs;
+	if (__copy_to_user(&frame->sc, &sc, sizeof(frame->sc)))
 		return -EFAULT;
 
+	/* Store registers needed to create the signal frame */
+	store_sigregs();
+
+	/* Create _sigregs32 on the signal stack */
 	if (save_sigregs32(regs, &frame->sregs))
 		return -EFAULT;
-	if (save_sigregs_gprs_high(regs, frame->gprs_high))
+
+	/* Place signal number on stack to allow backtrace from handler.  */
+	if (__put_user(regs->gprs[2], (int __force __user *) &frame->signo))
 		return -EFAULT;
-	if (__put_user((unsigned long) &frame->sregs, &frame->sc.sregs))
+
+	/* Create _sigregs_ext32 on the signal stack */
+	if (save_sigregs_ext32(regs, &frame->sregs_ext))
 		return -EFAULT;
 
 	/* Set up to return from userspace.  If provided, use a stub
 	   already in userspace.  */
 	if (ksig->ka.sa.sa_flags & SA_RESTORER) {
-		regs->gprs[14] = (__u64 __force) ksig->ka.sa.sa_restorer | PSW32_ADDR_AMODE;
+		restorer = (unsigned long __force)
+			ksig->ka.sa.sa_restorer | PSW32_ADDR_AMODE;
 	} else {
-		regs->gprs[14] = (__u64 __force) frame->retcode | PSW32_ADDR_AMODE;
-		if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn,
-			       (u16 __force __user *)(frame->retcode)))
+		/* Signal frames without vectors registers are short ! */
+		__u16 __user *svc = (void *) frame + frame_size - 2;
+		if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn, svc))
 			return -EFAULT;
+		restorer = (unsigned long __force) svc | PSW32_ADDR_AMODE;
         }
 
-	/* Set up backchain. */
-	if (__put_user(regs->gprs[15], (unsigned int __user *) frame))
-		return -EFAULT;
-
 	/* Set up registers for signal handler */
+	regs->gprs[14] = restorer;
 	regs->gprs[15] = (__force __u64) frame;
 	/* Force 31 bit amode and default user address space control. */
 	regs->psw.mask = PSW_MASK_BA |
@@ -375,50 +462,69 @@ static int setup_frame32(struct ksignal *ksig, sigset_t *set,
 		regs->gprs[6] = task_thread_info(current)->last_break;
 	}
 
-	/* Place signal number on stack to allow backtrace from handler.  */
-	if (__put_user(regs->gprs[2], (int __force __user *) &frame->signo))
-		return -EFAULT;
 	return 0;
 }
 
 static int setup_rt_frame32(struct ksignal *ksig, sigset_t *set,
 			    struct pt_regs *regs)
 {
-	int err = 0;
-	rt_sigframe32 __user *frame = get_sigframe(&ksig->ka, regs, sizeof(rt_sigframe32));
-
+	rt_sigframe32 __user *frame;
+	unsigned long restorer;
+	size_t frame_size;
+	u32 uc_flags;
+
+	frame_size = sizeof(*frame) -
+		     sizeof(frame->uc.uc_mcontext_ext.__reserved);
+	/*
+	 * gprs_high are always present for 31-bit compat tasks.
+	 * The space for vector registers is only allocated if
+	 * the machine supports it
+	 */
+	uc_flags = UC_GPRS_HIGH;
+	if (MACHINE_HAS_VX) {
+		if (current->thread.vxrs)
+			uc_flags |= UC_VXRS;
+	} else
+		frame_size -= sizeof(frame->uc.uc_mcontext_ext.vxrs_low) +
+			      sizeof(frame->uc.uc_mcontext_ext.vxrs_high);
+	frame = get_sigframe(&ksig->ka, regs, frame_size);
 	if (frame == (void __user *) -1UL)
 		return -EFAULT;
 
-	if (copy_siginfo_to_user32(&frame->info, &ksig->info))
-		return -EFAULT;
-
-	/* Create the ucontext.  */
-	err |= __put_user(UC_EXTENDED, &frame->uc.uc_flags);
-	err |= __put_user(0, &frame->uc.uc_link);
-	err |= __compat_save_altstack(&frame->uc.uc_stack, regs->gprs[15]);
-	err |= save_sigregs32(regs, &frame->uc.uc_mcontext);
-	err |= save_sigregs_gprs_high(regs, frame->gprs_high);
-	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
-	if (err)
+	/* Set up backchain. */
+	if (__put_user(regs->gprs[15], (unsigned int __force __user *) frame))
 		return -EFAULT;
 
 	/* Set up to return from userspace.  If provided, use a stub
 	   already in userspace.  */
 	if (ksig->ka.sa.sa_flags & SA_RESTORER) {
-		regs->gprs[14] = (__u64 __force) ksig->ka.sa.sa_restorer | PSW32_ADDR_AMODE;
+		restorer = (unsigned long __force)
+			ksig->ka.sa.sa_restorer | PSW32_ADDR_AMODE;
 	} else {
-		regs->gprs[14] = (__u64 __force) frame->retcode | PSW32_ADDR_AMODE;
-		if (__put_user(S390_SYSCALL_OPCODE | __NR_rt_sigreturn,
-			       (u16 __force __user *)(frame->retcode)))
+		__u16 __user *svc = &frame->svc_insn;
+		if (__put_user(S390_SYSCALL_OPCODE | __NR_rt_sigreturn, svc))
 			return -EFAULT;
+		restorer = (unsigned long __force) svc | PSW32_ADDR_AMODE;
 	}
 
-	/* Set up backchain. */
-	if (__put_user(regs->gprs[15], (unsigned int __force __user *) frame))
+	/* Create siginfo on the signal stack */
+	if (copy_siginfo_to_user32(&frame->info, &ksig->info))
+		return -EFAULT;
+
+	/* Store registers needed to create the signal frame */
+	store_sigregs();
+
+	/* Create ucontext on the signal stack. */
+	if (__put_user(uc_flags, &frame->uc.uc_flags) ||
+	    __put_user(0, &frame->uc.uc_link) ||
+	    __compat_save_altstack(&frame->uc.uc_stack, regs->gprs[15]) ||
+	    save_sigregs32(regs, &frame->uc.uc_mcontext) ||
+	    __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)) ||
+	    save_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext))
 		return -EFAULT;
 
 	/* Set up registers for signal handler */
+	regs->gprs[14] = restorer;
 	regs->gprs[15] = (__force __u64) frame;
 	/* Force 31 bit amode and default user address space control. */
 	regs->psw.mask = PSW_MASK_BA |

arch/s390/kernel/crash_dump.c

@@ -46,9 +46,9 @@ struct dump_save_areas dump_save_areas;
 /*
  * Allocate and add a save area for a CPU
  */
-struct save_area *dump_save_area_create(int cpu)
+struct save_area_ext *dump_save_area_create(int cpu)
 {
-	struct save_area **save_areas, *save_area;
+	struct save_area_ext **save_areas, *save_area;
 
 	save_area = kmalloc(sizeof(*save_area), GFP_KERNEL);
 	if (!save_area)
@@ -385,10 +385,46 @@ static void *nt_s390_prefix(void *ptr, struct save_area *sa)
 			 sizeof(sa->pref_reg), KEXEC_CORE_NOTE_NAME);
 }
 
+/*
+ * Initialize vxrs high note (full 128 bit VX registers 16-31)
+ */
+static void *nt_s390_vx_high(void *ptr, __vector128 *vx_regs)
+{
+	return nt_init(ptr, NT_S390_VXRS_HIGH, &vx_regs[16],
+		       16 * sizeof(__vector128), KEXEC_CORE_NOTE_NAME);
+}
+
+/*
+ * Initialize vxrs low note (lower halves of VX registers 0-15)
+ */
+static void *nt_s390_vx_low(void *ptr, __vector128 *vx_regs)
+{
+	Elf64_Nhdr *note;
+	u64 len;
+	int i;
+
+	note = (Elf64_Nhdr *)ptr;
+	note->n_namesz = strlen(KEXEC_CORE_NOTE_NAME) + 1;
+	note->n_descsz = 16 * 8;
+	note->n_type = NT_S390_VXRS_LOW;
+	len = sizeof(Elf64_Nhdr);
+
+	memcpy(ptr + len, KEXEC_CORE_NOTE_NAME, note->n_namesz);
+	len = roundup(len + note->n_namesz, 4);
+
+	ptr += len;
+	/* Copy lower halves of SIMD registers 0-15 */
+	for (i = 0; i < 16; i++) {
+		memcpy(ptr, &vx_regs[i], 8);
+		ptr += 8;
+	}
+	return ptr;
+}
+
 /*
  * Fill ELF notes for one CPU with save area registers
  */
-void *fill_cpu_elf_notes(void *ptr, struct save_area *sa)
+void *fill_cpu_elf_notes(void *ptr, struct save_area *sa, __vector128 *vx_regs)
 {
 	ptr = nt_prstatus(ptr, sa);
 	ptr = nt_fpregset(ptr, sa);
@@ -397,6 +433,10 @@ void *fill_cpu_elf_notes(void *ptr, struct save_area *sa)
 	ptr = nt_s390_tod_preg(ptr, sa);
 	ptr = nt_s390_ctrs(ptr, sa);
 	ptr = nt_s390_prefix(ptr, sa);
+	if (MACHINE_HAS_VX && vx_regs) {
+		ptr = nt_s390_vx_low(ptr, vx_regs);
+		ptr = nt_s390_vx_high(ptr, vx_regs);
+	}
 	return ptr;
 }
 
@@ -484,7 +524,7 @@ static int get_cpu_cnt(void)
 	int i, cpus = 0;
 
 	for (i = 0; i < dump_save_areas.count; i++) {
-		if (dump_save_areas.areas[i]->pref_reg == 0)
+		if (dump_save_areas.areas[i]->sa.pref_reg == 0)
 			continue;
 		cpus++;
 	}
@@ -530,17 +570,17 @@ static void loads_init(Elf64_Phdr *phdr, u64 loads_offset)
  */
 static void *notes_init(Elf64_Phdr *phdr, void *ptr, u64 notes_offset)
 {
-	struct save_area *sa;
+	struct save_area_ext *sa_ext;
 	void *ptr_start = ptr;
 	int i;
 
 	ptr = nt_prpsinfo(ptr);
 
 	for (i = 0; i < dump_save_areas.count; i++) {
-		sa = dump_save_areas.areas[i];
-		if (sa->pref_reg == 0)
+		sa_ext = dump_save_areas.areas[i];
+		if (sa_ext->sa.pref_reg == 0)
 			continue;
-		ptr = fill_cpu_elf_notes(ptr, sa);
+		ptr = fill_cpu_elf_notes(ptr, &sa_ext->sa, sa_ext->vx_regs);
 	}
 	ptr = nt_vmcoreinfo(ptr);
 	memset(phdr, 0, sizeof(*phdr));
@@ -581,7 +621,7 @@ int elfcorehdr_alloc(unsigned long long *addr, unsigned long long *size)
 
 	mem_chunk_cnt = get_mem_chunk_cnt();
 
-	alloc_size = 0x1000 + get_cpu_cnt() * 0x300 +
+	alloc_size = 0x1000 + get_cpu_cnt() * 0x4a0 +
 		mem_chunk_cnt * sizeof(Elf64_Phdr);
 	hdr = kzalloc_panic(alloc_size);
 	/* Init elf header */

arch/s390/kernel/dis.c

@@ -60,6 +60,11 @@ enum {
 	A_28,	/* Access reg. starting at position 28 */
 	C_8,	/* Control reg. starting at position 8 */
 	C_12,	/* Control reg. starting at position 12 */
+	V_8,	/* Vector reg. starting at position 8, extension bit at 36 */
+	V_12,	/* Vector reg. starting at position 12, extension bit at 37 */
+	V_16,	/* Vector reg. starting at position 16, extension bit at 38 */
+	V_32,	/* Vector reg. starting at position 32, extension bit at 39 */
+	W_12,	/* Vector reg. at bit 12, extension at bit 37, used as index */
 	B_16,	/* Base register starting at position 16 */
 	B_32,	/* Base register starting at position 32 */
 	X_12,	/* Index register starting at position 12 */
@@ -82,6 +87,8 @@ enum {
 	U8_24,	/* 8 bit unsigned value starting at 24 */
 	U8_32,	/* 8 bit unsigned value starting at 32 */
 	I8_8,	/* 8 bit signed value starting at 8 */
+	I8_16,	/* 8 bit signed value starting at 16 */
+	I8_24,	/* 8 bit signed value starting at 24 */
 	I8_32,	/* 8 bit signed value starting at 32 */
 	J12_12, /* PC relative offset at 12 */
 	I16_16,	/* 16 bit signed value starting at 16 */
@@ -96,6 +103,9 @@ enum {
 	U32_16,	/* 32 bit unsigned value starting at 16 */
 	M_16,	/* 4 bit optional mask starting at 16 */
 	M_20,	/* 4 bit optional mask starting at 20 */
+	M_24,	/* 4 bit optional mask starting at 24 */
+	M_28,	/* 4 bit optional mask starting at 28 */
+	M_32,	/* 4 bit optional mask starting at 32 */
 	RO_28,	/* optional GPR starting at position 28 */
 };
 
@@ -130,7 +140,7 @@ enum {
 	INSTR_RSY_RDRM,
 	INSTR_RS_AARD, INSTR_RS_CCRD, INSTR_RS_R0RD, INSTR_RS_RRRD,
 	INSTR_RS_RURD,
-	INSTR_RXE_FRRD, INSTR_RXE_RRRD,
+	INSTR_RXE_FRRD, INSTR_RXE_RRRD, INSTR_RXE_RRRDM,
 	INSTR_RXF_FRRDF,
 	INSTR_RXY_FRRD, INSTR_RXY_RRRD, INSTR_RXY_URRD,
 	INSTR_RX_FRRD, INSTR_RX_RRRD, INSTR_RX_URRD,
@@ -143,6 +153,17 @@ enum {
 	INSTR_SS_L0RDRD, INSTR_SS_LIRDRD, INSTR_SS_LLRDRD, INSTR_SS_RRRDRD,
 	INSTR_SS_RRRDRD2, INSTR_SS_RRRDRD3,
 	INSTR_S_00, INSTR_S_RD,
+	INSTR_VRI_V0IM, INSTR_VRI_V0I0, INSTR_VRI_V0IIM, INSTR_VRI_VVIM,
+	INSTR_VRI_VVV0IM, INSTR_VRI_VVV0I0, INSTR_VRI_VVIMM,
+	INSTR_VRR_VV00MMM, INSTR_VRR_VV000MM, INSTR_VRR_VV0000M,
+	INSTR_VRR_VV00000, INSTR_VRR_VVV0M0M, INSTR_VRR_VV00M0M,
+	INSTR_VRR_VVV000M, INSTR_VRR_VVV000V, INSTR_VRR_VVV0000,
+	INSTR_VRR_VVV0MMM, INSTR_VRR_VVV00MM, INSTR_VRR_VVVMM0V,
+	INSTR_VRR_VVVM0MV, INSTR_VRR_VVVM00V, INSTR_VRR_VRR0000,
+	INSTR_VRS_VVRDM, INSTR_VRS_VVRD0, INSTR_VRS_VRRDM, INSTR_VRS_VRRD0,
+	INSTR_VRS_RVRDM,
+	INSTR_VRV_VVRDM, INSTR_VRV_VWRDM,
+	INSTR_VRX_VRRDM, INSTR_VRX_VRRD0,
 };
 
 static const struct s390_operand operands[] =
@@ -168,6 +189,11 @@ static const struct s390_operand operands[] =
 	[A_28]	 = {  4, 28, OPERAND_AR },
 	[C_8]	 = {  4,  8, OPERAND_CR },
 	[C_12]	 = {  4, 12, OPERAND_CR },
+	[V_8]	 = {  4,  8, OPERAND_VR },
+	[V_12]	 = {  4, 12, OPERAND_VR },
+	[V_16]	 = {  4, 16, OPERAND_VR },
+	[V_32]	 = {  4, 32, OPERAND_VR },
+	[W_12]	 = {  4, 12, OPERAND_INDEX | OPERAND_VR },
 	[B_16]	 = {  4, 16, OPERAND_BASE | OPERAND_GPR },
 	[B_32]	 = {  4, 32, OPERAND_BASE | OPERAND_GPR },
 	[X_12]	 = {  4, 12, OPERAND_INDEX | OPERAND_GPR },
@@ -190,6 +216,11 @@ static const struct s390_operand operands[] =
 	[U8_24]  = {  8, 24, 0 },
 	[U8_32]  = {  8, 32, 0 },
 	[J12_12] = { 12, 12, OPERAND_PCREL },
+	[I8_8]	 = {  8,  8, OPERAND_SIGNED },
+	[I8_16]  = {  8, 16, OPERAND_SIGNED },
+	[I8_24]  = {  8, 24, OPERAND_SIGNED },
+	[I8_32]  = {  8, 32, OPERAND_SIGNED },
+	[I16_32] = { 16, 32, OPERAND_SIGNED },
 	[I16_16] = { 16, 16, OPERAND_SIGNED },
 	[U16_16] = { 16, 16, 0 },
 	[U16_32] = { 16, 32, 0 },
@@ -202,6 +233,9 @@ static const struct s390_operand operands[] =
 	[U32_16] = { 32, 16, 0 },
 	[M_16]	 = {  4, 16, 0 },
 	[M_20]	 = {  4, 20, 0 },
+	[M_24]	 = {  4, 24, 0 },
+	[M_28]	 = {  4, 28, 0 },
+	[M_32]	 = {  4, 32, 0 },
 	[RO_28]  = {  4, 28, OPERAND_GPR }
 };
 
@@ -283,6 +317,7 @@ static const unsigned char formats[][7] = {
 	[INSTR_RS_RURD]	  = { 0xff, R_8,U4_12,D_20,B_16,0,0 },
 	[INSTR_RXE_FRRD]  = { 0xff, F_8,D_20,X_12,B_16,0,0 },
 	[INSTR_RXE_RRRD]  = { 0xff, R_8,D_20,X_12,B_16,0,0 },
+	[INSTR_RXE_RRRDM] = { 0xff, R_8,D_20,X_12,B_16,M_32,0 },
 	[INSTR_RXF_FRRDF] = { 0xff, F_32,F_8,D_20,X_12,B_16,0 },
 	[INSTR_RXY_FRRD]  = { 0xff, F_8,D20_20,X_12,B_16,0,0 },
 	[INSTR_RXY_RRRD]  = { 0xff, R_8,D20_20,X_12,B_16,0,0 },
@@ -307,6 +342,37 @@ static const unsigned char formats[][7] = {
 	[INSTR_SS_RRRDRD] = { 0xff, D_20,R_8,B_16,D_36,B_32,R_12 },
 	[INSTR_S_00]	  = { 0xff, 0,0,0,0,0,0 },
 	[INSTR_S_RD]	  = { 0xff, D_20,B_16,0,0,0,0 },
+	[INSTR_VRI_V0IM]  = { 0xff, V_8,I16_16,M_32,0,0,0 },
+	[INSTR_VRI_V0I0]  = { 0xff, V_8,I16_16,0,0,0,0 },
+	[INSTR_VRI_V0IIM] = { 0xff, V_8,I8_16,I8_24,M_32,0,0 },
+	[INSTR_VRI_VVIM]  = { 0xff, V_8,I16_16,V_12,M_32,0,0 },
+	[INSTR_VRI_VVV0IM]= { 0xff, V_8,V_12,V_16,I8_24,M_32,0 },
+	[INSTR_VRI_VVV0I0]= { 0xff, V_8,V_12,V_16,I8_24,0,0 },
+	[INSTR_VRI_VVIMM] = { 0xff, V_8,V_12,I16_16,M_32,M_28,0 },
+	[INSTR_VRR_VV00MMM]={ 0xff, V_8,V_12,M_32,M_28,M_24,0 },
+	[INSTR_VRR_VV000MM]={ 0xff, V_8,V_12,M_32,M_28,0,0 },
+	[INSTR_VRR_VV0000M]={ 0xff, V_8,V_12,M_32,0,0,0 },
+	[INSTR_VRR_VV00000]={ 0xff, V_8,V_12,0,0,0,0 },
+	[INSTR_VRR_VVV0M0M]={ 0xff, V_8,V_12,V_16,M_32,M_24,0 },
+	[INSTR_VRR_VV00M0M]={ 0xff, V_8,V_12,M_32,M_24,0,0 },
+	[INSTR_VRR_VVV000M]={ 0xff, V_8,V_12,V_16,M_32,0,0 },
+	[INSTR_VRR_VVV000V]={ 0xff, V_8,V_12,V_16,V_32,0,0 },
+	[INSTR_VRR_VVV0000]={ 0xff, V_8,V_12,V_16,0,0,0 },
+	[INSTR_VRR_VVV0MMM]={ 0xff, V_8,V_12,V_16,M_32,M_28,M_24 },
+	[INSTR_VRR_VVV00MM]={ 0xff, V_8,V_12,V_16,M_32,M_28,0 },
+	[INSTR_VRR_VVVMM0V]={ 0xff, V_8,V_12,V_16,V_32,M_20,M_24 },
+	[INSTR_VRR_VVVM0MV]={ 0xff, V_8,V_12,V_16,V_32,M_28,M_20 },
+	[INSTR_VRR_VVVM00V]={ 0xff, V_8,V_12,V_16,V_32,M_20,0 },
+	[INSTR_VRR_VRR0000]={ 0xff, V_8,R_12,R_16,0,0,0 },
+	[INSTR_VRS_VVRDM] = { 0xff, V_8,V_12,D_20,B_16,M_32,0 },
+	[INSTR_VRS_VVRD0] = { 0xff, V_8,V_12,D_20,B_16,0,0 },
+	[INSTR_VRS_VRRDM] = { 0xff, V_8,R_12,D_20,B_16,M_32,0 },
+	[INSTR_VRS_VRRD0] = { 0xff, V_8,R_12,D_20,B_16,0,0 },
+	[INSTR_VRS_RVRDM] = { 0xff, R_8,V_12,D_20,B_16,M_32,0 },
+	[INSTR_VRV_VVRDM] = { 0xff, V_8,V_12,D_20,B_16,M_32,0 },
+	[INSTR_VRV_VWRDM] = { 0xff, V_8,D_20,W_12,B_16,M_32,0 },
+	[INSTR_VRX_VRRDM] = { 0xff, V_8,D_20,X_12,B_16,M_32,0 },
+	[INSTR_VRX_VRRD0] = { 0xff, V_8,D_20,X_12,B_16,0,0 },
 };
 
 enum {
@@ -381,6 +447,11 @@ enum {
 	LONG_INSN_MPCIFC,
 	LONG_INSN_STPCIFC,
 	LONG_INSN_PCISTB,
+	LONG_INSN_VPOPCT,
+	LONG_INSN_VERLLV,
+	LONG_INSN_VESRAV,
+	LONG_INSN_VESRLV,
+	LONG_INSN_VSBCBI
 };
 
 static char *long_insn_name[] = {
@@ -455,6 +526,11 @@ static char *long_insn_name[] = {
 	[LONG_INSN_MPCIFC] = "mpcifc",
 	[LONG_INSN_STPCIFC] = "stpcifc",
 	[LONG_INSN_PCISTB] = "pcistb",
+	[LONG_INSN_VPOPCT] = "vpopct",
+	[LONG_INSN_VERLLV] = "verllv",
+	[LONG_INSN_VESRAV] = "vesrav",
+	[LONG_INSN_VESRLV] = "vesrlv",
+	[LONG_INSN_VSBCBI] = "vsbcbi",
 };
 
 static struct s390_insn opcode[] = {
@@ -1369,6 +1445,150 @@ static struct s390_insn opcode_e5[] = {
 	{ "", 0, INSTR_INVALID }
 };
 
+static struct s390_insn opcode_e7[] = {
+#ifdef CONFIG_64BIT
+	{ "lcbb", 0x27, INSTR_RXE_RRRDM },
+	{ "vgef", 0x13, INSTR_VRV_VVRDM },
+	{ "vgeg", 0x12, INSTR_VRV_VVRDM },
+	{ "vgbm", 0x44, INSTR_VRI_V0I0 },
+	{ "vgm", 0x46, INSTR_VRI_V0IIM },
+	{ "vl", 0x06, INSTR_VRX_VRRD0 },
+	{ "vlr", 0x56, INSTR_VRR_VV00000 },
+	{ "vlrp", 0x05, INSTR_VRX_VRRDM },
+	{ "vleb", 0x00, INSTR_VRX_VRRDM },
+	{ "vleh", 0x01, INSTR_VRX_VRRDM },
+	{ "vlef", 0x03, INSTR_VRX_VRRDM },
+	{ "vleg", 0x02, INSTR_VRX_VRRDM },
+	{ "vleib", 0x40, INSTR_VRI_V0IM },
+	{ "vleih", 0x41, INSTR_VRI_V0IM },
+	{ "vleif", 0x43, INSTR_VRI_V0IM },
+	{ "vleig", 0x42, INSTR_VRI_V0IM },
+	{ "vlgv", 0x21, INSTR_VRS_RVRDM },
+	{ "vllez", 0x04, INSTR_VRX_VRRDM },
+	{ "vlm", 0x36, INSTR_VRS_VVRD0 },
+	{ "vlbb", 0x07, INSTR_VRX_VRRDM },
+	{ "vlvg", 0x22, INSTR_VRS_VRRDM },
+	{ "vlvgp", 0x62, INSTR_VRR_VRR0000 },
+	{ "vll", 0x37, INSTR_VRS_VRRD0 },
+	{ "vmrh", 0x61, INSTR_VRR_VVV000M },
+	{ "vmrl", 0x60, INSTR_VRR_VVV000M },
+	{ "vpk", 0x94, INSTR_VRR_VVV000M },
+	{ "vpks", 0x97, INSTR_VRR_VVV0M0M },
+	{ "vpkls", 0x95, INSTR_VRR_VVV0M0M },
+	{ "vperm", 0x8c, INSTR_VRR_VVV000V },
+	{ "vpdi", 0x84, INSTR_VRR_VVV000M },
+	{ "vrep", 0x4d, INSTR_VRI_VVIM },
+	{ "vrepi", 0x45, INSTR_VRI_V0IM },
+	{ "vscef", 0x1b, INSTR_VRV_VWRDM },
+	{ "vsceg", 0x1a, INSTR_VRV_VWRDM },
+	{ "vsel", 0x8d, INSTR_VRR_VVV000V },
+	{ "vseg", 0x5f, INSTR_VRR_VV0000M },
+	{ "vst", 0x0e, INSTR_VRX_VRRD0 },
+	{ "vsteb", 0x08, INSTR_VRX_VRRDM },
+	{ "vsteh", 0x09, INSTR_VRX_VRRDM },
+	{ "vstef", 0x0b, INSTR_VRX_VRRDM },
+	{ "vsteg", 0x0a, INSTR_VRX_VRRDM },
+	{ "vstm", 0x3e, INSTR_VRS_VVRD0 },
+	{ "vstl", 0x3f, INSTR_VRS_VRRD0 },
+	{ "vuph", 0xd7, INSTR_VRR_VV0000M },
+	{ "vuplh", 0xd5, INSTR_VRR_VV0000M },
+	{ "vupl", 0xd6, INSTR_VRR_VV0000M },
+	{ "vupll", 0xd4, INSTR_VRR_VV0000M },
+	{ "va", 0xf3, INSTR_VRR_VVV000M },
+	{ "vacc", 0xf1, INSTR_VRR_VVV000M },
+	{ "vac", 0xbb, INSTR_VRR_VVVM00V },
+	{ "vaccc", 0xb9, INSTR_VRR_VVVM00V },
+	{ "vn", 0x68, INSTR_VRR_VVV0000 },
+	{ "vnc", 0x69, INSTR_VRR_VVV0000 },
+	{ "vavg", 0xf2, INSTR_VRR_VVV000M },
+	{ "vavgl", 0xf0, INSTR_VRR_VVV000M },
+	{ "vcksm", 0x66, INSTR_VRR_VVV0000 },
+	{ "vec", 0xdb, INSTR_VRR_VV0000M },
+	{ "vecl", 0xd9, INSTR_VRR_VV0000M },
+	{ "vceq", 0xf8, INSTR_VRR_VVV0M0M },
+	{ "vch", 0xfb, INSTR_VRR_VVV0M0M },
+	{ "vchl", 0xf9, INSTR_VRR_VVV0M0M },
+	{ "vclz", 0x53, INSTR_VRR_VV0000M },
+	{ "vctz", 0x52, INSTR_VRR_VV0000M },
+	{ "vx", 0x6d, INSTR_VRR_VVV0000 },
+	{ "vgfm", 0xb4, INSTR_VRR_VVV000M },
+	{ "vgfma", 0xbc, INSTR_VRR_VVVM00V },
+	{ "vlc", 0xde, INSTR_VRR_VV0000M },
+	{ "vlp", 0xdf, INSTR_VRR_VV0000M },
+	{ "vmx", 0xff, INSTR_VRR_VVV000M },
+	{ "vmxl", 0xfd, INSTR_VRR_VVV000M },
+	{ "vmn", 0xfe, INSTR_VRR_VVV000M },
+	{ "vmnl", 0xfc, INSTR_VRR_VVV000M },
+	{ "vmal", 0xaa, INSTR_VRR_VVVM00V },
+	{ "vmae", 0xae, INSTR_VRR_VVVM00V },
+	{ "vmale", 0xac, INSTR_VRR_VVVM00V },
+	{ "vmah", 0xab, INSTR_VRR_VVVM00V },
+	{ "vmalh", 0xa9, INSTR_VRR_VVVM00V },
+	{ "vmao", 0xaf, INSTR_VRR_VVVM00V },
+	{ "vmalo", 0xad, INSTR_VRR_VVVM00V },
+	{ "vmh", 0xa3, INSTR_VRR_VVV000M },
+	{ "vmlh", 0xa1, INSTR_VRR_VVV000M },
+	{ "vml", 0xa2, INSTR_VRR_VVV000M },
+	{ "vme", 0xa6, INSTR_VRR_VVV000M },
+	{ "vmle", 0xa4, INSTR_VRR_VVV000M },
+	{ "vmo", 0xa7, INSTR_VRR_VVV000M },
+	{ "vmlo", 0xa5, INSTR_VRR_VVV000M },
+	{ "vno", 0x6b, INSTR_VRR_VVV0000 },
+	{ "vo", 0x6a, INSTR_VRR_VVV0000 },
+	{ { 0, LONG_INSN_VPOPCT }, 0x50, INSTR_VRR_VV0000M },
+	{ { 0, LONG_INSN_VERLLV }, 0x73, INSTR_VRR_VVV000M },
+	{ "verll", 0x33, INSTR_VRS_VVRDM },
+	{ "verim", 0x72, INSTR_VRI_VVV0IM },
+	{ "veslv", 0x70, INSTR_VRR_VVV000M },
+	{ "vesl", 0x30, INSTR_VRS_VVRDM },
+	{ { 0, LONG_INSN_VESRAV }, 0x7a, INSTR_VRR_VVV000M },
+	{ "vesra", 0x3a, INSTR_VRS_VVRDM },
+	{ { 0, LONG_INSN_VESRLV }, 0x78, INSTR_VRR_VVV000M },
+	{ "vesrl", 0x38, INSTR_VRS_VVRDM },
+	{ "vsl", 0x74, INSTR_VRR_VVV0000 },
+	{ "vslb", 0x75, INSTR_VRR_VVV0000 },
+	{ "vsldb", 0x77, INSTR_VRI_VVV0I0 },
+	{ "vsra", 0x7e, INSTR_VRR_VVV0000 },
+	{ "vsrab", 0x7f, INSTR_VRR_VVV0000 },
+	{ "vsrl", 0x7c, INSTR_VRR_VVV0000 },
+	{ "vsrlb", 0x7d, INSTR_VRR_VVV0000 },
+	{ "vs", 0xf7, INSTR_VRR_VVV000M },
+	{ "vscb", 0xf5, INSTR_VRR_VVV000M },
+	{ "vsb", 0xbf, INSTR_VRR_VVVM00V },
+	{ { 0, LONG_INSN_VSBCBI }, 0xbd, INSTR_VRR_VVVM00V },
+	{ "vsumg", 0x65, INSTR_VRR_VVV000M },
+	{ "vsumq", 0x67, INSTR_VRR_VVV000M },
+	{ "vsum", 0x64, INSTR_VRR_VVV000M },
+	{ "vtm", 0xd8, INSTR_VRR_VV00000 },
+	{ "vfae", 0x82, INSTR_VRR_VVV0M0M },
+	{ "vfee", 0x80, INSTR_VRR_VVV0M0M },
+	{ "vfene", 0x81, INSTR_VRR_VVV0M0M },
+	{ "vistr", 0x5c, INSTR_VRR_VV00M0M },
+	{ "vstrc", 0x8a, INSTR_VRR_VVVMM0V },
+	{ "vfa", 0xe3, INSTR_VRR_VVV00MM },
+	{ "wfc", 0xcb, INSTR_VRR_VV000MM },
+	{ "wfk", 0xca, INSTR_VRR_VV000MM },
+	{ "vfce", 0xe8, INSTR_VRR_VVV0MMM },
+	{ "vfch", 0xeb, INSTR_VRR_VVV0MMM },
+	{ "vfche", 0xea, INSTR_VRR_VVV0MMM },
+	{ "vcdg", 0xc3, INSTR_VRR_VV00MMM },
+	{ "vcdlg", 0xc1, INSTR_VRR_VV00MMM },
+	{ "vcgd", 0xc2, INSTR_VRR_VV00MMM },
+	{ "vclgd", 0xc0, INSTR_VRR_VV00MMM },
+	{ "vfd", 0xe5, INSTR_VRR_VVV00MM },
+	{ "vfi", 0xc7, INSTR_VRR_VV00MMM },
+	{ "vlde", 0xc4, INSTR_VRR_VV000MM },
+	{ "vled", 0xc5, INSTR_VRR_VV00MMM },
+	{ "vfm", 0xe7, INSTR_VRR_VVV00MM },
+	{ "vfma", 0x8f, INSTR_VRR_VVVM0MV },
+	{ "vfms", 0x8e, INSTR_VRR_VVVM0MV },
+	{ "vfpso", 0xcc, INSTR_VRR_VV00MMM },
+	{ "vfsq", 0xce, INSTR_VRR_VV000MM },
+	{ "vfs", 0xe2, INSTR_VRR_VVV00MM },
+	{ "vftci", 0x4a, INSTR_VRI_VVIMM },
+#endif
+};
+
 static struct s390_insn opcode_eb[] = {
 #ifdef CONFIG_64BIT
 	{ "lmg", 0x04, INSTR_RSY_RRRD },
@@ -1552,16 +1772,17 @@ static struct s390_insn opcode_ed[] = {
 static unsigned int extract_operand(unsigned char *code,
 				    const struct s390_operand *operand)
 {
+	unsigned char *cp;
 	unsigned int val;
 	int bits;
 
 	/* Extract fragments of the operand byte for byte.  */
-	code += operand->shift / 8;
+	cp = code + operand->shift / 8;
 	bits = (operand->shift & 7) + operand->bits;
 	val = 0;
 	do {
 		val <<= 8;
-		val |= (unsigned int) *code++;
+		val |= (unsigned int) *cp++;
 		bits -= 8;
 	} while (bits > 0);
 	val >>= -bits;
@@ -1571,6 +1792,18 @@ static unsigned int extract_operand(unsigned char *code,
 	if (operand->bits == 20 && operand->shift == 20)
 		val = (val & 0xff) << 12 | (val & 0xfff00) >> 8;
 
+	/* Check for register extensions bits for vector registers. */
+	if (operand->flags & OPERAND_VR) {
+		if (operand->shift == 8)
+			val |= (code[4] & 8) << 1;
+		else if (operand->shift == 12)
+			val |= (code[4] & 4) << 2;
+		else if (operand->shift == 16)
+			val |= (code[4] & 2) << 3;
+		else if (operand->shift == 32)
+			val |= (code[4] & 1) << 4;
+	}
+
 	/* Sign extend value if the operand is signed or pc relative.  */
 	if ((operand->flags & (OPERAND_SIGNED | OPERAND_PCREL)) &&
 	    (val & (1U << (operand->bits - 1))))
@@ -1639,6 +1872,10 @@ struct s390_insn *find_insn(unsigned char *code)
 	case 0xe5:
 		table = opcode_e5;
 		break;
+	case 0xe7:
+		table = opcode_e7;
+		opfrag = code[5];
+		break;
 	case 0xeb:
 		table = opcode_eb;
 		opfrag = code[5];
@@ -1734,6 +1971,8 @@ static int print_insn(char *buffer, unsigned char *code, unsigned long addr)
 				ptr += sprintf(ptr, "%%a%i", value);
 			else if (operand->flags & OPERAND_CR)
 				ptr += sprintf(ptr, "%%c%i", value);
+			else if (operand->flags & OPERAND_VR)
+				ptr += sprintf(ptr, "%%v%i", value);
 			else if (operand->flags & OPERAND_PCREL)
 				ptr += sprintf(ptr, "%lx", (signed int) value
 								      + addr);

arch/s390/kernel/early.c

@@ -390,10 +390,10 @@ static __init void detect_machine_facilities(void)
 		S390_lowcore.machine_flags |= MACHINE_FLAG_LPP;
 	if (test_facility(50) && test_facility(73))
 		S390_lowcore.machine_flags |= MACHINE_FLAG_TE;
-	if (test_facility(66))
-		S390_lowcore.machine_flags |= MACHINE_FLAG_RRBM;
 	if (test_facility(51))
 		S390_lowcore.machine_flags |= MACHINE_FLAG_TLB_LC;
+	if (test_facility(129))
+		S390_lowcore.machine_flags |= MACHINE_FLAG_VX;
 #endif
 }
 

arch/s390/kernel/entry.h

@@ -4,7 +4,7 @@
 #include <linux/types.h>
 #include <linux/signal.h>
 #include <asm/ptrace.h>
-#include <asm/cputime.h>
+#include <asm/idle.h>
 
 extern void *restart_stack;
 extern unsigned long suspend_zero_pages;
@@ -21,6 +21,8 @@ void psw_idle(struct s390_idle_data *, unsigned long);
 asmlinkage long do_syscall_trace_enter(struct pt_regs *regs);
 asmlinkage void do_syscall_trace_exit(struct pt_regs *regs);
 
+int alloc_vector_registers(struct task_struct *tsk);
+
 void do_protection_exception(struct pt_regs *regs);
 void do_dat_exception(struct pt_regs *regs);
 
@@ -43,8 +45,10 @@ void special_op_exception(struct pt_regs *regs);
 void specification_exception(struct pt_regs *regs);
 void transaction_exception(struct pt_regs *regs);
 void translation_exception(struct pt_regs *regs);
+void vector_exception(struct pt_regs *regs);
 
 void do_per_trap(struct pt_regs *regs);
+void do_report_trap(struct pt_regs *regs, int si_signo, int si_code, char *str);
 void syscall_trace(struct pt_regs *regs, int entryexit);
 void kernel_stack_overflow(struct pt_regs * regs);
 void do_signal(struct pt_regs *regs);

arch/s390/kernel/entry64.S

@@ -42,7 +42,8 @@ STACK_SHIFT = PAGE_SHIFT + THREAD_ORDER
 STACK_SIZE  = 1 << STACK_SHIFT
 STACK_INIT = STACK_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE
 
-_TIF_WORK	= (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED)
+_TIF_WORK	= (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \
+		   _TIF_UPROBE)
 _TIF_TRACE	= (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \
 		   _TIF_SYSCALL_TRACEPOINT)
 _CIF_WORK	= (_CIF_MCCK_PENDING | _CIF_ASCE)
@@ -265,6 +266,10 @@ sysc_work:
 	jo	sysc_mcck_pending
 	tm	__TI_flags+7(%r12),_TIF_NEED_RESCHED
 	jo	sysc_reschedule
+#ifdef CONFIG_UPROBES
+	tm	__TI_flags+7(%r12),_TIF_UPROBE
+	jo	sysc_uprobe_notify
+#endif
 	tm	__PT_FLAGS+7(%r11),_PIF_PER_TRAP
 	jo	sysc_singlestep
 	tm	__TI_flags+7(%r12),_TIF_SIGPENDING
@@ -322,6 +327,16 @@ sysc_notify_resume:
 	larl	%r14,sysc_return
 	jg	do_notify_resume
 
+#
+# _TIF_UPROBE is set, call uprobe_notify_resume
+#
+#ifdef CONFIG_UPROBES
+sysc_uprobe_notify:
+	lgr	%r2,%r11		# pass pointer to pt_regs
+	larl	%r14,sysc_return
+	jg	uprobe_notify_resume
+#endif
+
 #
 # _PIF_PER_TRAP is set, call do_per_trap
 #

arch/s390/kernel/ftrace.c

 /*
  * Dynamic function tracer architecture backend.
  *
- * Copyright IBM Corp. 2009
+ * Copyright IBM Corp. 2009,2014
  *
  *   Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>,
  *		Martin Schwidefsky <schwidefsky@de.ibm.com>
@@ -17,100 +17,76 @@
 #include <asm/asm-offsets.h>
 #include "entry.h"
 
-#ifdef CONFIG_DYNAMIC_FTRACE
-
+void mcount_replace_code(void);
 void ftrace_disable_code(void);
 void ftrace_enable_insn(void);
 
-#ifdef CONFIG_64BIT
 /*
- * The 64-bit mcount code looks like this:
+ * The mcount code looks like this:
  *	stg	%r14,8(%r15)		# offset 0
- * >	larl	%r1,<&counter>		# offset 6
- * >	brasl	%r14,_mcount		# offset 12
+ *	larl	%r1,<&counter>		# offset 6
+ *	brasl	%r14,_mcount		# offset 12
  *	lg	%r14,8(%r15)		# offset 18
- * Total length is 24 bytes. The middle two instructions of the mcount
- * block get overwritten by ftrace_make_nop / ftrace_make_call.
- * The 64-bit enabled ftrace code block looks like this:
- *	stg	%r14,8(%r15)		# offset 0
+ * Total length is 24 bytes. The complete mcount block initially gets replaced
+ * by ftrace_make_nop. Subsequent calls to ftrace_make_call / ftrace_make_nop
+ * only patch the jg/lg instruction within the block.
+ * Note: we do not patch the first instruction to an unconditional branch,
+ * since that would break kprobes/jprobes. It is easier to leave the larl
+ * instruction in and only modify the second instruction.
+ * The enabled ftrace code block looks like this:
+ *	larl	%r0,.+24		# offset 0
  * >	lg	%r1,__LC_FTRACE_FUNC	# offset 6
- * >	lgr	%r0,%r0			# offset 12
- * >	basr	%r14,%r1		# offset 16
- *	lg	%r14,8(%15)		# offset 18
- * The return points of the mcount/ftrace function have the same offset 18.
- * The 64-bit disable ftrace code block looks like this:
- *	stg	%r14,8(%r15)		# offset 0
+ *	br	%r1			# offset 12
+ *	brcl	0,0			# offset 14
+ *	brc	0,0			# offset 20
+ * The ftrace function gets called with a non-standard C function call ABI
+ * where r0 contains the return address. It is also expected that the called
+ * function only clobbers r0 and r1, but restores r2-r15.
+ * The return point of the ftrace function has offset 24, so execution
+ * continues behind the mcount block.
+ *	larl	%r0,.+24		# offset 0
  * >	jg	.+18			# offset 6
- * >	lgr	%r0,%r0			# offset 12
- * >	basr	%r14,%r1		# offset 16
- *	lg	%r14,8(%15)		# offset 18
+ *	br	%r1			# offset 12
+ *	brcl	0,0			# offset 14
+ *	brc	0,0			# offset 20
  * The jg instruction branches to offset 24 to skip as many instructions
  * as possible.
  */
 asm(
 	"	.align	4\n"
+	"mcount_replace_code:\n"
+	"	larl	%r0,0f\n"
 	"ftrace_disable_code:\n"
 	"	jg	0f\n"
-	"	lgr	%r0,%r0\n"
-	"	basr	%r14,%r1\n"
+	"	br	%r1\n"
+	"	brcl	0,0\n"
+	"	brc	0,0\n"
 	"0:\n"
 	"	.align	4\n"
 	"ftrace_enable_insn:\n"
 	"	lg	%r1,"__stringify(__LC_FTRACE_FUNC)"\n");
 
+#define MCOUNT_BLOCK_SIZE	24
+#define MCOUNT_INSN_OFFSET	6
 #define FTRACE_INSN_SIZE	6
 
-#else /* CONFIG_64BIT */
-/*
- * The 31-bit mcount code looks like this:
- *	st	%r14,4(%r15)		# offset 0
- * >	bras	%r1,0f			# offset 4
- * >	.long	_mcount			# offset 8
- * >	.long	<&counter>		# offset 12
- * > 0:	l	%r14,0(%r1)		# offset 16
- * >	l	%r1,4(%r1)		# offset 20
- *	basr	%r14,%r14		# offset 24
- *	l	%r14,4(%r15)		# offset 26
- * Total length is 30 bytes. The twenty bytes starting from offset 4
- * to offset 24 get overwritten by ftrace_make_nop / ftrace_make_call.
- * The 31-bit enabled ftrace code block looks like this:
- *	st	%r14,4(%r15)		# offset 0
- * >	l	%r14,__LC_FTRACE_FUNC	# offset 4
- * >	j	0f			# offset 8
- * >	.fill	12,1,0x07		# offset 12
- *   0:	basr	%r14,%r14		# offset 24
- *	l	%r14,4(%r14)		# offset 26
- * The return points of the mcount/ftrace function have the same offset 26.
- * The 31-bit disabled ftrace code block looks like this:
- *	st	%r14,4(%r15)		# offset 0
- * >	j	.+26			# offset 4
- * >	j	0f			# offset 8
- * >	.fill	12,1,0x07		# offset 12
- *   0:	basr	%r14,%r14		# offset 24
- *	l	%r14,4(%r14)		# offset 26
- * The j instruction branches to offset 30 to skip as many instructions
- * as possible.
- */
-asm(
-	"	.align	4\n"
-	"ftrace_disable_code:\n"
-	"	j	1f\n"
-	"	j	0f\n"
-	"	.fill	12,1,0x07\n"
-	"0:	basr	%r14,%r14\n"
-	"1:\n"
-	"	.align	4\n"
-	"ftrace_enable_insn:\n"
-	"	l	%r14,"__stringify(__LC_FTRACE_FUNC)"\n");
-
-#define FTRACE_INSN_SIZE	4
-
-#endif /* CONFIG_64BIT */
-
+int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr,
+		       unsigned long addr)
+{
+	return 0;
+}
 
 int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec,
 		    unsigned long addr)
 {
+	/* Initial replacement of the whole mcount block */
+	if (addr == MCOUNT_ADDR) {
+		if (probe_kernel_write((void *) rec->ip - MCOUNT_INSN_OFFSET,
+				       mcount_replace_code,
+				       MCOUNT_BLOCK_SIZE))
+			return -EPERM;
+		return 0;
+	}
 	if (probe_kernel_write((void *) rec->ip, ftrace_disable_code,
 			       MCOUNT_INSN_SIZE))
 		return -EPERM;
@@ -135,8 +111,6 @@ int __init ftrace_dyn_arch_init(void)
 	return 0;
 }
 
-#endif /* CONFIG_DYNAMIC_FTRACE */
-
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 /*
  * Hook the return address and push it in the stack of return addresses
@@ -162,31 +136,26 @@ unsigned long __kprobes prepare_ftrace_return(unsigned long parent,
 	return parent;
 }
 
-#ifdef CONFIG_DYNAMIC_FTRACE
 /*
  * Patch the kernel code at ftrace_graph_caller location. The instruction
- * there is branch relative and save to prepare_ftrace_return. To disable
- * the call to prepare_ftrace_return we patch the bras offset to point
- * directly after the instructions. To enable the call we calculate
- * the original offset to prepare_ftrace_return and put it back.
+ * there is branch relative on condition. To enable the ftrace graph code
+ * block, we simply patch the mask field of the instruction to zero and
+ * turn the instruction into a nop.
+ * To disable the ftrace graph code the mask field will be patched to
+ * all ones, which turns the instruction into an unconditional branch.
  */
 int ftrace_enable_ftrace_graph_caller(void)
 {
-	unsigned short offset;
+	u8 op = 0x04; /* set mask field to zero */
 
-	offset = ((void *) prepare_ftrace_return -
-		  (void *) ftrace_graph_caller) / 2;
-	return probe_kernel_write((void *) ftrace_graph_caller + 2,
-				  &offset, sizeof(offset));
+	return probe_kernel_write(__va(ftrace_graph_caller)+1, &op, sizeof(op));
 }
 
 int ftrace_disable_ftrace_graph_caller(void)
 {
-	static unsigned short offset = 0x0002;
+	u8 op = 0xf4; /* set mask field to all ones */
 
-	return probe_kernel_write((void *) ftrace_graph_caller + 2,
-				  &offset, sizeof(offset));
+	return probe_kernel_write(__va(ftrace_graph_caller)+1, &op, sizeof(op));
 }
 
-#endif /* CONFIG_DYNAMIC_FTRACE */
 #endif /* CONFIG_FUNCTION_GRAPH_TRACER */

arch/s390/kernel/head.S

@@ -398,7 +398,7 @@ ENTRY(startup_kdump)
 	xc	__LC_STFL_FAC_LIST(8),__LC_STFL_FAC_LIST
 #ifndef CONFIG_MARCH_G5
 	# check capabilities against MARCH_{G5,Z900,Z990,Z9_109,Z10}
-	.insn	s,0xb2b10000,__LC_STFL_FAC_LIST	# store facility list
+	.insn	s,0xb2b10000,0		# store facilities @ __LC_STFL_FAC_LIST
 	tm	__LC_STFL_FAC_LIST,0x01	# stfle available ?
 	jz	0f
 	la	%r0,1

arch/s390/kernel/idle.c

+/*
+ * Idle functions for s390.
+ *
+ * Copyright IBM Corp. 2014
+ *
+ * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/kernel_stat.h>
+#include <linux/kprobes.h>
+#include <linux/notifier.h>
+#include <linux/init.h>
+#include <linux/cpu.h>
+#include <asm/cputime.h>
+#include <asm/nmi.h>
+#include <asm/smp.h>
+#include "entry.h"
+
+static DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
+
+void __kprobes enabled_wait(void)
+{
+	struct s390_idle_data *idle = &__get_cpu_var(s390_idle);
+	unsigned long long idle_time;
+	unsigned long psw_mask;
+
+	trace_hardirqs_on();
+
+	/* Wait for external, I/O or machine check interrupt. */
+	psw_mask = PSW_KERNEL_BITS | PSW_MASK_WAIT | PSW_MASK_DAT |
+		PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK;
+	clear_cpu_flag(CIF_NOHZ_DELAY);
+
+	/* Call the assembler magic in entry.S */
+	psw_idle(idle, psw_mask);
+
+	/* Account time spent with enabled wait psw loaded as idle time. */
+	idle->sequence++;
+	smp_wmb();
+	idle_time = idle->clock_idle_exit - idle->clock_idle_enter;
+	idle->clock_idle_enter = idle->clock_idle_exit = 0ULL;
+	idle->idle_time += idle_time;
+	idle->idle_count++;
+	account_idle_time(idle_time);
+	smp_wmb();
+	idle->sequence++;
+}
+
+static ssize_t show_idle_count(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
+	unsigned long long idle_count;
+	unsigned int sequence;
+
+	do {
+		sequence = ACCESS_ONCE(idle->sequence);
+		idle_count = ACCESS_ONCE(idle->idle_count);
+		if (ACCESS_ONCE(idle->clock_idle_enter))
+			idle_count++;
+	} while ((sequence & 1) || (ACCESS_ONCE(idle->sequence) != sequence));
+	return sprintf(buf, "%llu\n", idle_count);
+}
+DEVICE_ATTR(idle_count, 0444, show_idle_count, NULL);
+
+static ssize_t show_idle_time(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
+	unsigned long long now, idle_time, idle_enter, idle_exit;
+	unsigned int sequence;
+
+	do {
+		now = get_tod_clock();
+		sequence = ACCESS_ONCE(idle->sequence);
+		idle_time = ACCESS_ONCE(idle->idle_time);
+		idle_enter = ACCESS_ONCE(idle->clock_idle_enter);
+		idle_exit = ACCESS_ONCE(idle->clock_idle_exit);
+	} while ((sequence & 1) || (ACCESS_ONCE(idle->sequence) != sequence));
+	idle_time += idle_enter ? ((idle_exit ? : now) - idle_enter) : 0;
+	return sprintf(buf, "%llu\n", idle_time >> 12);
+}
+DEVICE_ATTR(idle_time_us, 0444, show_idle_time, NULL);
+
+cputime64_t arch_cpu_idle_time(int cpu)
+{
+	struct s390_idle_data *idle = &per_cpu(s390_idle, cpu);
+	unsigned long long now, idle_enter, idle_exit;
+	unsigned int sequence;
+
+	do {
+		now = get_tod_clock();
+		sequence = ACCESS_ONCE(idle->sequence);
+		idle_enter = ACCESS_ONCE(idle->clock_idle_enter);
+		idle_exit = ACCESS_ONCE(idle->clock_idle_exit);
+	} while ((sequence & 1) || (ACCESS_ONCE(idle->sequence) != sequence));
+	return idle_enter ? ((idle_exit ?: now) - idle_enter) : 0;
+}
+
+void arch_cpu_idle_enter(void)
+{
+	local_mcck_disable();
+}
+
+void arch_cpu_idle(void)
+{
+	if (!test_cpu_flag(CIF_MCCK_PENDING))
+		/* Halt the cpu and keep track of cpu time accounting. */
+		enabled_wait();
+	local_irq_enable();
+}
+
+void arch_cpu_idle_exit(void)
+{
+	local_mcck_enable();
+	if (test_cpu_flag(CIF_MCCK_PENDING))
+		s390_handle_mcck();
+}
+
+void arch_cpu_idle_dead(void)
+{
+	cpu_die();
+}

arch/s390/kernel/irq.c

@@ -70,6 +70,7 @@ static const struct irq_class irqclass_sub_desc[NR_ARCH_IRQS] = {
 	{.irq = IRQEXT_CMS, .name = "CMS", .desc = "[EXT] CPU-Measurement: Sampling"},
 	{.irq = IRQEXT_CMC, .name = "CMC", .desc = "[EXT] CPU-Measurement: Counter"},
 	{.irq = IRQEXT_CMR, .name = "CMR", .desc = "[EXT] CPU-Measurement: RI"},
+	{.irq = IRQEXT_FTP, .name = "FTP", .desc = "[EXT] HMC FTP Service"},
 	{.irq = IRQIO_CIO,  .name = "CIO", .desc = "[I/O] Common I/O Layer Interrupt"},
 	{.irq = IRQIO_QAI,  .name = "QAI", .desc = "[I/O] QDIO Adapter Interrupt"},
 	{.irq = IRQIO_DAS,  .name = "DAS", .desc = "[I/O] DASD"},
@@ -258,7 +259,7 @@ static irqreturn_t do_ext_interrupt(int irq, void *dummy)
 
 	ext_code = *(struct ext_code *) &regs->int_code;
 	if (ext_code.code != EXT_IRQ_CLK_COMP)
-		__get_cpu_var(s390_idle).nohz_delay = 1;
+		set_cpu_flag(CIF_NOHZ_DELAY);
 
 	index = ext_hash(ext_code.code);
 	rcu_read_lock();

arch/s390/kernel/kprobes.c

@@ -58,161 +58,13 @@ struct kprobe_insn_cache kprobe_dmainsn_slots = {
 	.insn_size = MAX_INSN_SIZE,
 };
 
-static int __kprobes is_prohibited_opcode(kprobe_opcode_t *insn)
-{
-	if (!is_known_insn((unsigned char *)insn))
-		return -EINVAL;
-	switch (insn[0] >> 8) {
-	case 0x0c:	/* bassm */
-	case 0x0b:	/* bsm	 */
-	case 0x83:	/* diag  */
-	case 0x44:	/* ex	 */
-	case 0xac:	/* stnsm */
-	case 0xad:	/* stosm */
-		return -EINVAL;
-	case 0xc6:
-		switch (insn[0] & 0x0f) {
-		case 0x00: /* exrl   */
-			return -EINVAL;
-		}
-	}
-	switch (insn[0]) {
-	case 0x0101:	/* pr	 */
-	case 0xb25a:	/* bsa	 */
-	case 0xb240:	/* bakr  */
-	case 0xb258:	/* bsg	 */
-	case 0xb218:	/* pc	 */
-	case 0xb228:	/* pt	 */
-	case 0xb98d:	/* epsw	 */
-		return -EINVAL;
-	}
-	return 0;
-}
-
-static int __kprobes get_fixup_type(kprobe_opcode_t *insn)
-{
-	/* default fixup method */
-	int fixup = FIXUP_PSW_NORMAL;
-
-	switch (insn[0] >> 8) {
-	case 0x05:	/* balr	*/
-	case 0x0d:	/* basr */
-		fixup = FIXUP_RETURN_REGISTER;
-		/* if r2 = 0, no branch will be taken */
-		if ((insn[0] & 0x0f) == 0)
-			fixup |= FIXUP_BRANCH_NOT_TAKEN;
-		break;
-	case 0x06:	/* bctr	*/
-	case 0x07:	/* bcr	*/
-		fixup = FIXUP_BRANCH_NOT_TAKEN;
-		break;
-	case 0x45:	/* bal	*/
-	case 0x4d:	/* bas	*/
-		fixup = FIXUP_RETURN_REGISTER;
-		break;
-	case 0x47:	/* bc	*/
-	case 0x46:	/* bct	*/
-	case 0x86:	/* bxh	*/
-	case 0x87:	/* bxle	*/
-		fixup = FIXUP_BRANCH_NOT_TAKEN;
-		break;
-	case 0x82:	/* lpsw	*/
-		fixup = FIXUP_NOT_REQUIRED;
-		break;
-	case 0xb2:	/* lpswe */
-		if ((insn[0] & 0xff) == 0xb2)
-			fixup = FIXUP_NOT_REQUIRED;
-		break;
-	case 0xa7:	/* bras	*/
-		if ((insn[0] & 0x0f) == 0x05)
-			fixup |= FIXUP_RETURN_REGISTER;
-		break;
-	case 0xc0:
-		if ((insn[0] & 0x0f) == 0x05)	/* brasl */
-			fixup |= FIXUP_RETURN_REGISTER;
-		break;
-	case 0xeb:
-		switch (insn[2] & 0xff) {
-		case 0x44: /* bxhg  */
-		case 0x45: /* bxleg */
-			fixup = FIXUP_BRANCH_NOT_TAKEN;
-			break;
-		}
-		break;
-	case 0xe3:	/* bctg	*/
-		if ((insn[2] & 0xff) == 0x46)
-			fixup = FIXUP_BRANCH_NOT_TAKEN;
-		break;
-	case 0xec:
-		switch (insn[2] & 0xff) {
-		case 0xe5: /* clgrb */
-		case 0xe6: /* cgrb  */
-		case 0xf6: /* crb   */
-		case 0xf7: /* clrb  */
-		case 0xfc: /* cgib  */
-		case 0xfd: /* cglib */
-		case 0xfe: /* cib   */
-		case 0xff: /* clib  */
-			fixup = FIXUP_BRANCH_NOT_TAKEN;
-			break;
-		}
-		break;
-	}
-	return fixup;
-}
-
-static int __kprobes is_insn_relative_long(kprobe_opcode_t *insn)
-{
-	/* Check if we have a RIL-b or RIL-c format instruction which
-	 * we need to modify in order to avoid instruction emulation. */
-	switch (insn[0] >> 8) {
-	case 0xc0:
-		if ((insn[0] & 0x0f) == 0x00) /* larl */
-			return true;
-		break;
-	case 0xc4:
-		switch (insn[0] & 0x0f) {
-		case 0x02: /* llhrl  */
-		case 0x04: /* lghrl  */
-		case 0x05: /* lhrl   */
-		case 0x06: /* llghrl */
-		case 0x07: /* sthrl  */
-		case 0x08: /* lgrl   */
-		case 0x0b: /* stgrl  */
-		case 0x0c: /* lgfrl  */
-		case 0x0d: /* lrl    */
-		case 0x0e: /* llgfrl */
-		case 0x0f: /* strl   */
-			return true;
-		}
-		break;
-	case 0xc6:
-		switch (insn[0] & 0x0f) {
-		case 0x02: /* pfdrl  */
-		case 0x04: /* cghrl  */
-		case 0x05: /* chrl   */
-		case 0x06: /* clghrl */
-		case 0x07: /* clhrl  */
-		case 0x08: /* cgrl   */
-		case 0x0a: /* clgrl  */
-		case 0x0c: /* cgfrl  */
-		case 0x0d: /* crl    */
-		case 0x0e: /* clgfrl */
-		case 0x0f: /* clrl   */
-			return true;
-		}
-		break;
-	}
-	return false;
-}
-
 static void __kprobes copy_instruction(struct kprobe *p)
 {
 	s64 disp, new_disp;
 	u64 addr, new_addr;
 
 	memcpy(p->ainsn.insn, p->addr, insn_length(p->opcode >> 8));
-	if (!is_insn_relative_long(p->ainsn.insn))
+	if (!probe_is_insn_relative_long(p->ainsn.insn))
 		return;
 	/*
 	 * For pc-relative instructions in RIL-b or RIL-c format patch the
@@ -276,7 +128,7 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 	if ((unsigned long) p->addr & 0x01)
 		return -EINVAL;
 	/* Make sure the probe isn't going on a difficult instruction */
-	if (is_prohibited_opcode(p->addr))
+	if (probe_is_prohibited_opcode(p->addr))
 		return -EINVAL;
 	if (s390_get_insn_slot(p))
 		return -ENOMEM;
@@ -605,7 +457,7 @@ static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 	unsigned long ip = regs->psw.addr & PSW_ADDR_INSN;
-	int fixup = get_fixup_type(p->ainsn.insn);
+	int fixup = probe_get_fixup_type(p->ainsn.insn);
 
 	if (fixup & FIXUP_PSW_NORMAL)
 		ip += (unsigned long) p->addr - (unsigned long) p->ainsn.insn;
@@ -789,11 +641,6 @@ void __kprobes jprobe_return(void)
 	asm volatile(".word 0x0002");
 }
 
-static void __used __kprobes jprobe_return_end(void)
-{
-	asm volatile("bcr 0,0");
-}
-
 int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();

arch/s390/kernel/machine_kexec.c

@@ -25,6 +25,7 @@
 #include <asm/elf.h>
 #include <asm/asm-offsets.h>
 #include <asm/os_info.h>
+#include <asm/switch_to.h>
 
 typedef void (*relocate_kernel_t)(kimage_entry_t *, unsigned long);
 
@@ -43,7 +44,7 @@ static void add_elf_notes(int cpu)
 
 	memcpy((void *) (4608UL + sa->pref_reg), sa, sizeof(*sa));
 	ptr = (u64 *) per_cpu_ptr(crash_notes, cpu);
-	ptr = fill_cpu_elf_notes(ptr, sa);
+	ptr = fill_cpu_elf_notes(ptr, sa, NULL);
 	memset(ptr, 0, sizeof(struct elf_note));
 }
 
@@ -53,8 +54,11 @@ static void add_elf_notes(int cpu)
 static void setup_regs(void)
 {
 	unsigned long sa = S390_lowcore.prefixreg_save_area + SAVE_AREA_BASE;
+	struct _lowcore *lc;
 	int cpu, this_cpu;
 
+	/* Get lowcore pointer from store status of this CPU (absolute zero) */
+	lc = (struct _lowcore *)(unsigned long)S390_lowcore.prefixreg_save_area;
 	this_cpu = smp_find_processor_id(stap());
 	add_elf_notes(this_cpu);
 	for_each_online_cpu(cpu) {
@@ -64,6 +68,8 @@ static void setup_regs(void)
 			continue;
 		add_elf_notes(cpu);
 	}
+	if (MACHINE_HAS_VX)
+		save_vx_regs_safe((void *) lc->vector_save_area_addr);
 	/* Copy dump CPU store status info to absolute zero */
 	memcpy((void *) SAVE_AREA_BASE, (void *) sa, sizeof(struct save_area));
 }

arch/s390/kernel/mcount.S

@@ -8,62 +8,72 @@
 #include <linux/linkage.h>
 #include <asm/asm-offsets.h>
 #include <asm/ftrace.h>
+#include <asm/ptrace.h>
 
 	.section .kprobes.text, "ax"
 
 ENTRY(ftrace_stub)
 	br	%r14
 
+#define STACK_FRAME_SIZE  (STACK_FRAME_OVERHEAD + __PT_SIZE)
+#define STACK_PTREGS	  (STACK_FRAME_OVERHEAD)
+#define STACK_PTREGS_GPRS (STACK_PTREGS + __PT_GPRS)
+#define STACK_PTREGS_PSW  (STACK_PTREGS + __PT_PSW)
+
 ENTRY(_mcount)
-#ifdef CONFIG_DYNAMIC_FTRACE
 	br	%r14
 
 ENTRY(ftrace_caller)
+	.globl	ftrace_regs_caller
+	.set	ftrace_regs_caller,ftrace_caller
+	lgr	%r1,%r15
+	aghi	%r15,-STACK_FRAME_SIZE
+	stg	%r1,__SF_BACKCHAIN(%r15)
+	stg	%r1,(STACK_PTREGS_GPRS+15*8)(%r15)
+	stg	%r0,(STACK_PTREGS_PSW+8)(%r15)
+	stmg	%r2,%r14,(STACK_PTREGS_GPRS+2*8)(%r15)
+#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES
+	aghik	%r2,%r0,-MCOUNT_INSN_SIZE
+	lgrl	%r4,function_trace_op
+	lgrl	%r1,ftrace_trace_function
+#else
+	lgr	%r2,%r0
+	aghi	%r2,-MCOUNT_INSN_SIZE
+	larl	%r4,function_trace_op
+	lg	%r4,0(%r4)
+	larl	%r1,ftrace_trace_function
+	lg	%r1,0(%r1)
 #endif
-	stm	%r2,%r5,16(%r15)
-	bras	%r1,1f
-0:	.long	ftrace_trace_function
-1:	st	%r14,56(%r15)
-	lr	%r0,%r15
-	ahi	%r15,-96
-	l	%r3,100(%r15)
-	la	%r2,0(%r14)
-	st	%r0,__SF_BACKCHAIN(%r15)
-	la	%r3,0(%r3)
-	ahi	%r2,-MCOUNT_INSN_SIZE
-	l	%r14,0b-0b(%r1)
-	l	%r14,0(%r14)
-	basr	%r14,%r14
+	lgr	%r3,%r14
+	la	%r5,STACK_PTREGS(%r15)
+	basr	%r14,%r1
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
-	l	%r2,100(%r15)
-	l	%r3,152(%r15)
+# The j instruction gets runtime patched to a nop instruction.
+# See ftrace_enable_ftrace_graph_caller.
 ENTRY(ftrace_graph_caller)
-# The bras instruction gets runtime patched to call prepare_ftrace_return.
-# See ftrace_enable_ftrace_graph_caller. The patched instruction is:
-#	bras	%r14,prepare_ftrace_return
-	bras	%r14,0f
-0:	st	%r2,100(%r15)
+	j	ftrace_graph_caller_end
+	lg	%r2,(STACK_PTREGS_GPRS+14*8)(%r15)
+	lg	%r3,(STACK_PTREGS_PSW+8)(%r15)
+	brasl	%r14,prepare_ftrace_return
+	stg	%r2,(STACK_PTREGS_GPRS+14*8)(%r15)
+ftrace_graph_caller_end:
+	.globl	ftrace_graph_caller_end
 #endif
-	ahi	%r15,96
-	l	%r14,56(%r15)
-	lm	%r2,%r5,16(%r15)
-	br	%r14
+	lg	%r1,(STACK_PTREGS_PSW+8)(%r15)
+	lmg	%r2,%r15,(STACK_PTREGS_GPRS+2*8)(%r15)
+	br	%r1
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 
 ENTRY(return_to_handler)
-	stm	%r2,%r5,16(%r15)
-	st	%r14,56(%r15)
-	lr	%r0,%r15
-	ahi	%r15,-96
-	st	%r0,__SF_BACKCHAIN(%r15)
-	bras	%r1,0f
-	.long	ftrace_return_to_handler
-0:	l	%r2,0b-0b(%r1)
-	basr	%r14,%r2
-	lr	%r14,%r2
-	ahi	%r15,96
-	lm	%r2,%r5,16(%r15)
+	stmg	%r2,%r5,32(%r15)
+	lgr	%r1,%r15
+	aghi	%r15,-STACK_FRAME_OVERHEAD
+	stg	%r1,__SF_BACKCHAIN(%r15)
+	brasl	%r14,ftrace_return_to_handler
+	aghi	%r15,STACK_FRAME_OVERHEAD
+	lgr	%r14,%r2
+	lmg	%r2,%r5,32(%r15)
 	br	%r14
 
 #endif

arch/s390/kernel/mcount64.S

-/*
- * Copyright IBM Corp. 2008, 2009
- *
- *   Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>,
- *
- */
-
-#include <linux/linkage.h>
-#include <asm/asm-offsets.h>
-#include <asm/ftrace.h>
-
-	.section .kprobes.text, "ax"
-
-ENTRY(ftrace_stub)
-	br	%r14
-
-ENTRY(_mcount)
-#ifdef CONFIG_DYNAMIC_FTRACE
-	br	%r14
-
-ENTRY(ftrace_caller)
-#endif
-	stmg	%r2,%r5,32(%r15)
-	stg	%r14,112(%r15)
-	lgr	%r1,%r15
-	aghi	%r15,-160
-	stg	%r1,__SF_BACKCHAIN(%r15)
-	lgr	%r2,%r14
-	lg	%r3,168(%r15)
-	aghi	%r2,-MCOUNT_INSN_SIZE
-	larl	%r14,ftrace_trace_function
-	lg	%r14,0(%r14)
-	basr	%r14,%r14
-#ifdef CONFIG_FUNCTION_GRAPH_TRACER
-	lg	%r2,168(%r15)
-	lg	%r3,272(%r15)
-ENTRY(ftrace_graph_caller)
-# The bras instruction gets runtime patched to call prepare_ftrace_return.
-# See ftrace_enable_ftrace_graph_caller. The patched instruction is:
-#	bras	%r14,prepare_ftrace_return
-	bras	%r14,0f
-0:	stg	%r2,168(%r15)
-#endif
-	aghi	%r15,160
-	lmg	%r2,%r5,32(%r15)
-	lg	%r14,112(%r15)
-	br	%r14
-
-#ifdef CONFIG_FUNCTION_GRAPH_TRACER
-
-ENTRY(return_to_handler)
-	stmg	%r2,%r5,32(%r15)
-	lgr	%r1,%r15
-	aghi	%r15,-160
-	stg	%r1,__SF_BACKCHAIN(%r15)
-	brasl	%r14,ftrace_return_to_handler
-	aghi	%r15,160
-	lgr	%r14,%r2
-	lmg	%r2,%r5,32(%r15)
-	br	%r14
-
-#endif

arch/s390/kernel/nmi.c

@@ -20,6 +20,7 @@
 #include <asm/cputime.h>
 #include <asm/nmi.h>
 #include <asm/crw.h>
+#include <asm/switch_to.h>
 
 struct mcck_struct {
 	int kill_task;
@@ -163,6 +164,21 @@ static int notrace s390_revalidate_registers(struct mci *mci)
 			"	ld	15,120(%0)\n"
 			: : "a" (fpt_save_area));
 	}
+
+#ifdef CONFIG_64BIT
+	/* Revalidate vector registers */
+	if (MACHINE_HAS_VX && current->thread.vxrs) {
+		if (!mci->vr) {
+			/*
+			 * Vector registers can't be restored and therefore
+			 * the process needs to be terminated.
+			 */
+			kill_task = 1;
+		}
+		restore_vx_regs((__vector128 *)
+				S390_lowcore.vector_save_area_addr);
+	}
+#endif
 	/* Revalidate access registers */
 	asm volatile(
 		"	lam	0,15,0(%0)"

arch/s390/kernel/pgm_check.S

@@ -49,7 +49,7 @@ PGM_CHECK_DEFAULT			/* 17 */
 PGM_CHECK_64BIT(transaction_exception)	/* 18 */
 PGM_CHECK_DEFAULT			/* 19 */
 PGM_CHECK_DEFAULT			/* 1a */
-PGM_CHECK_DEFAULT			/* 1b */
+PGM_CHECK_64BIT(vector_exception)	/* 1b */
 PGM_CHECK(space_switch_exception)	/* 1c */
 PGM_CHECK(hfp_sqrt_exception)		/* 1d */
 PGM_CHECK_DEFAULT			/* 1e */

arch/s390/kernel/process.c

@@ -61,30 +61,6 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
 	return sf->gprs[8];
 }
 
-void arch_cpu_idle(void)
-{
-	local_mcck_disable();
-	if (test_cpu_flag(CIF_MCCK_PENDING)) {
-		local_mcck_enable();
-		local_irq_enable();
-		return;
-	}
-	/* Halt the cpu and keep track of cpu time accounting. */
-	vtime_stop_cpu();
-	local_irq_enable();
-}
-
-void arch_cpu_idle_exit(void)
-{
-	if (test_cpu_flag(CIF_MCCK_PENDING))
-		s390_handle_mcck();
-}
-
-void arch_cpu_idle_dead(void)
-{
-	cpu_die();
-}
-
 extern void __kprobes kernel_thread_starter(void);
 
 /*

arch/s390/kernel/processor.c

@@ -23,7 +23,6 @@ static DEFINE_PER_CPU(struct cpuid, cpu_id);
  */
 void cpu_init(void)
 {
-	struct s390_idle_data *idle = &__get_cpu_var(s390_idle);
 	struct cpuid *id = &__get_cpu_var(cpu_id);
 
 	get_cpu_id(id);
@@ -31,7 +30,6 @@ void cpu_init(void)
 	current->active_mm = &init_mm;
 	BUG_ON(current->mm);
 	enter_lazy_tlb(&init_mm, current);
-	memset(idle, 0, sizeof(*idle));
 }
 
 /*
@@ -41,7 +39,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 {
 	static const char *hwcap_str[] = {
 		"esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp",
-		"edat", "etf3eh", "highgprs", "te"
+		"edat", "etf3eh", "highgprs", "te", "vx"
 	};
 	unsigned long n = (unsigned long) v - 1;
 	int i;

arch/s390/kernel/ptrace.c

@@ -38,15 +38,6 @@
 #define CREATE_TRACE_POINTS
 #include <trace/events/syscalls.h>
 
-enum s390_regset {
-	REGSET_GENERAL,
-	REGSET_FP,
-	REGSET_LAST_BREAK,
-	REGSET_TDB,
-	REGSET_SYSTEM_CALL,
-	REGSET_GENERAL_EXTENDED,
-};
-
 void update_cr_regs(struct task_struct *task)
 {
 	struct pt_regs *regs = task_pt_regs(task);
@@ -55,27 +46,39 @@ void update_cr_regs(struct task_struct *task)
 
 #ifdef CONFIG_64BIT
 	/* Take care of the enable/disable of transactional execution. */
-	if (MACHINE_HAS_TE) {
+	if (MACHINE_HAS_TE || MACHINE_HAS_VX) {
 		unsigned long cr, cr_new;
 
 		__ctl_store(cr, 0, 0);
-		/* Set or clear transaction execution TXC bit 8. */
-		cr_new = cr | (1UL << 55);
-		if (task->thread.per_flags & PER_FLAG_NO_TE)
-			cr_new &= ~(1UL << 55);
+		cr_new = cr;
+		if (MACHINE_HAS_TE) {
+			/* Set or clear transaction execution TXC bit 8. */
+			cr_new |= (1UL << 55);
+			if (task->thread.per_flags & PER_FLAG_NO_TE)
+				cr_new &= ~(1UL << 55);
+		}
+		if (MACHINE_HAS_VX) {
+			/* Enable/disable of vector extension */
+			cr_new &= ~(1UL << 17);
+			if (task->thread.vxrs)
+				cr_new |= (1UL << 17);
+		}
 		if (cr_new != cr)
 			__ctl_load(cr_new, 0, 0);
-		/* Set or clear transaction execution TDC bits 62 and 63. */
-		__ctl_store(cr, 2, 2);
-		cr_new = cr & ~3UL;
-		if (task->thread.per_flags & PER_FLAG_TE_ABORT_RAND) {
-			if (task->thread.per_flags & PER_FLAG_TE_ABORT_RAND_TEND)
-				cr_new |= 1UL;
-			else
-				cr_new |= 2UL;
+		if (MACHINE_HAS_TE) {
+			/* Set/clear transaction execution TDC bits 62/63. */
+			__ctl_store(cr, 2, 2);
+			cr_new = cr & ~3UL;
+			if (task->thread.per_flags & PER_FLAG_TE_ABORT_RAND) {
+				if (task->thread.per_flags &
+				    PER_FLAG_TE_ABORT_RAND_TEND)
+					cr_new |= 1UL;
+				else
+					cr_new |= 2UL;
+			}
+			if (cr_new != cr)
+				__ctl_load(cr_new, 2, 2);
 		}
-		if (cr_new != cr)
-			__ctl_load(cr_new, 2, 2);
 	}
 #endif
 	/* Copy user specified PER registers */
@@ -84,7 +87,8 @@ void update_cr_regs(struct task_struct *task)
 	new.end = thread->per_user.end;
 
 	/* merge TIF_SINGLE_STEP into user specified PER registers. */
-	if (test_tsk_thread_flag(task, TIF_SINGLE_STEP)) {
+	if (test_tsk_thread_flag(task, TIF_SINGLE_STEP) ||
+	    test_tsk_thread_flag(task, TIF_UPROBE_SINGLESTEP)) {
 		if (test_tsk_thread_flag(task, TIF_BLOCK_STEP))
 			new.control |= PER_EVENT_BRANCH;
 		else
@@ -93,6 +97,8 @@ void update_cr_regs(struct task_struct *task)
 		new.control |= PER_CONTROL_SUSPENSION;
 		new.control |= PER_EVENT_TRANSACTION_END;
 #endif
+		if (test_tsk_thread_flag(task, TIF_UPROBE_SINGLESTEP))
+			new.control |= PER_EVENT_IFETCH;
 		new.start = 0;
 		new.end = PSW_ADDR_INSN;
 	}
@@ -923,7 +929,15 @@ static int s390_fpregs_get(struct task_struct *target,
 		save_fp_ctl(&target->thread.fp_regs.fpc);
 		save_fp_regs(target->thread.fp_regs.fprs);
 	}
+#ifdef CONFIG_64BIT
+	else if (target->thread.vxrs) {
+		int i;
 
+		for (i = 0; i < __NUM_VXRS_LOW; i++)
+			target->thread.fp_regs.fprs[i] =
+				*(freg_t *)(target->thread.vxrs + i);
+	}
+#endif
 	return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
 				   &target->thread.fp_regs, 0, -1);
 }
@@ -957,9 +971,20 @@ static int s390_fpregs_set(struct task_struct *target,
 					target->thread.fp_regs.fprs,
 					offsetof(s390_fp_regs, fprs), -1);
 
-	if (rc == 0 && target == current) {
-		restore_fp_ctl(&target->thread.fp_regs.fpc);
-		restore_fp_regs(target->thread.fp_regs.fprs);
+	if (rc == 0) {
+		if (target == current) {
+			restore_fp_ctl(&target->thread.fp_regs.fpc);
+			restore_fp_regs(target->thread.fp_regs.fprs);
+		}
+#ifdef CONFIG_64BIT
+		else if (target->thread.vxrs) {
+			int i;
+
+			for (i = 0; i < __NUM_VXRS_LOW; i++)
+				*(freg_t *)(target->thread.vxrs + i) =
+					target->thread.fp_regs.fprs[i];
+		}
+#endif
 	}
 
 	return rc;
@@ -1015,6 +1040,95 @@ static int s390_tdb_set(struct task_struct *target,
 	return 0;
 }
 
+static int s390_vxrs_active(struct task_struct *target,
+			      const struct user_regset *regset)
+{
+	return !!target->thread.vxrs;
+}
+
+static int s390_vxrs_low_get(struct task_struct *target,
+			     const struct user_regset *regset,
+			     unsigned int pos, unsigned int count,
+			     void *kbuf, void __user *ubuf)
+{
+	__u64 vxrs[__NUM_VXRS_LOW];
+	int i;
+
+	if (target->thread.vxrs) {
+		if (target == current)
+			save_vx_regs(target->thread.vxrs);
+		for (i = 0; i < __NUM_VXRS_LOW; i++)
+			vxrs[i] = *((__u64 *)(target->thread.vxrs + i) + 1);
+	} else
+		memset(vxrs, 0, sizeof(vxrs));
+	return user_regset_copyout(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1);
+}
+
+static int s390_vxrs_low_set(struct task_struct *target,
+			     const struct user_regset *regset,
+			     unsigned int pos, unsigned int count,
+			     const void *kbuf, const void __user *ubuf)
+{
+	__u64 vxrs[__NUM_VXRS_LOW];
+	int i, rc;
+
+	if (!target->thread.vxrs) {
+		rc = alloc_vector_registers(target);
+		if (rc)
+			return rc;
+	} else if (target == current)
+		save_vx_regs(target->thread.vxrs);
+
+	rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1);
+	if (rc == 0) {
+		for (i = 0; i < __NUM_VXRS_LOW; i++)
+			*((__u64 *)(target->thread.vxrs + i) + 1) = vxrs[i];
+		if (target == current)
+			restore_vx_regs(target->thread.vxrs);
+	}
+
+	return rc;
+}
+
+static int s390_vxrs_high_get(struct task_struct *target,
+			      const struct user_regset *regset,
+			      unsigned int pos, unsigned int count,
+			      void *kbuf, void __user *ubuf)
+{
+	__vector128 vxrs[__NUM_VXRS_HIGH];
+
+	if (target->thread.vxrs) {
+		if (target == current)
+			save_vx_regs(target->thread.vxrs);
+		memcpy(vxrs, target->thread.vxrs + __NUM_VXRS_LOW,
+		       sizeof(vxrs));
+	} else
+		memset(vxrs, 0, sizeof(vxrs));
+	return user_regset_copyout(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1);
+}
+
+static int s390_vxrs_high_set(struct task_struct *target,
+			      const struct user_regset *regset,
+			      unsigned int pos, unsigned int count,
+			      const void *kbuf, const void __user *ubuf)
+{
+	int rc;
+
+	if (!target->thread.vxrs) {
+		rc = alloc_vector_registers(target);
+		if (rc)
+			return rc;
+	} else if (target == current)
+		save_vx_regs(target->thread.vxrs);
+
+	rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+				target->thread.vxrs + __NUM_VXRS_LOW, 0, -1);
+	if (rc == 0 && target == current)
+		restore_vx_regs(target->thread.vxrs);
+
+	return rc;
+}
+
 #endif
 
 static int s390_system_call_get(struct task_struct *target,
@@ -1038,7 +1152,7 @@ static int s390_system_call_set(struct task_struct *target,
 }
 
 static const struct user_regset s390_regsets[] = {
-	[REGSET_GENERAL] = {
+	{
 		.core_note_type = NT_PRSTATUS,
 		.n = sizeof(s390_regs) / sizeof(long),
 		.size = sizeof(long),
@@ -1046,7 +1160,7 @@ static const struct user_regset s390_regsets[] = {
 		.get = s390_regs_get,
 		.set = s390_regs_set,
 	},
-	[REGSET_FP] = {
+	{
 		.core_note_type = NT_PRFPREG,
 		.n = sizeof(s390_fp_regs) / sizeof(long),
 		.size = sizeof(long),
@@ -1054,8 +1168,16 @@ static const struct user_regset s390_regsets[] = {
 		.get = s390_fpregs_get,
 		.set = s390_fpregs_set,
 	},
+	{
+		.core_note_type = NT_S390_SYSTEM_CALL,
+		.n = 1,
+		.size = sizeof(unsigned int),
+		.align = sizeof(unsigned int),
+		.get = s390_system_call_get,
+		.set = s390_system_call_set,
+	},
 #ifdef CONFIG_64BIT
-	[REGSET_LAST_BREAK] = {
+	{
 		.core_note_type = NT_S390_LAST_BREAK,
 		.n = 1,
 		.size = sizeof(long),
@@ -1063,7 +1185,7 @@ static const struct user_regset s390_regsets[] = {
 		.get = s390_last_break_get,
 		.set = s390_last_break_set,
 	},
-	[REGSET_TDB] = {
+	{
 		.core_note_type = NT_S390_TDB,
 		.n = 1,
 		.size = 256,
@@ -1071,15 +1193,25 @@ static const struct user_regset s390_regsets[] = {
 		.get = s390_tdb_get,
 		.set = s390_tdb_set,
 	},
-#endif
-	[REGSET_SYSTEM_CALL] = {
-		.core_note_type = NT_S390_SYSTEM_CALL,
-		.n = 1,
-		.size = sizeof(unsigned int),
-		.align = sizeof(unsigned int),
-		.get = s390_system_call_get,
-		.set = s390_system_call_set,
+	{
+		.core_note_type = NT_S390_VXRS_LOW,
+		.n = __NUM_VXRS_LOW,
+		.size = sizeof(__u64),
+		.align = sizeof(__u64),
+		.active = s390_vxrs_active,
+		.get = s390_vxrs_low_get,
+		.set = s390_vxrs_low_set,
 	},
+	{
+		.core_note_type = NT_S390_VXRS_HIGH,
+		.n = __NUM_VXRS_HIGH,
+		.size = sizeof(__vector128),
+		.align = sizeof(__vector128),
+		.active = s390_vxrs_active,
+		.get = s390_vxrs_high_get,
+		.set = s390_vxrs_high_set,
+	},
+#endif
 };
 
 static const struct user_regset_view user_s390_view = {
@@ -1244,7 +1376,7 @@ static int s390_compat_last_break_set(struct task_struct *target,
 }
 
 static const struct user_regset s390_compat_regsets[] = {
-	[REGSET_GENERAL] = {
+	{
 		.core_note_type = NT_PRSTATUS,
 		.n = sizeof(s390_compat_regs) / sizeof(compat_long_t),
 		.size = sizeof(compat_long_t),
@@ -1252,7 +1384,7 @@ static const struct user_regset s390_compat_regsets[] = {
 		.get = s390_compat_regs_get,
 		.set = s390_compat_regs_set,
 	},
-	[REGSET_FP] = {
+	{
 		.core_note_type = NT_PRFPREG,
 		.n = sizeof(s390_fp_regs) / sizeof(compat_long_t),
 		.size = sizeof(compat_long_t),
@@ -1260,7 +1392,15 @@ static const struct user_regset s390_compat_regsets[] = {
 		.get = s390_fpregs_get,
 		.set = s390_fpregs_set,
 	},
-	[REGSET_LAST_BREAK] = {
+	{
+		.core_note_type = NT_S390_SYSTEM_CALL,
+		.n = 1,
+		.size = sizeof(compat_uint_t),
+		.align = sizeof(compat_uint_t),
+		.get = s390_system_call_get,
+		.set = s390_system_call_set,
+	},
+	{
 		.core_note_type = NT_S390_LAST_BREAK,
 		.n = 1,
 		.size = sizeof(long),
@@ -1268,7 +1408,7 @@ static const struct user_regset s390_compat_regsets[] = {
 		.get = s390_compat_last_break_get,
 		.set = s390_compat_last_break_set,
 	},
-	[REGSET_TDB] = {
+	{
 		.core_note_type = NT_S390_TDB,
 		.n = 1,
 		.size = 256,
@@ -1276,15 +1416,25 @@ static const struct user_regset s390_compat_regsets[] = {
 		.get = s390_tdb_get,
 		.set = s390_tdb_set,
 	},
-	[REGSET_SYSTEM_CALL] = {
-		.core_note_type = NT_S390_SYSTEM_CALL,
-		.n = 1,
-		.size = sizeof(compat_uint_t),
-		.align = sizeof(compat_uint_t),
-		.get = s390_system_call_get,
-		.set = s390_system_call_set,
+	{
+		.core_note_type = NT_S390_VXRS_LOW,
+		.n = __NUM_VXRS_LOW,
+		.size = sizeof(__u64),
+		.align = sizeof(__u64),
+		.active = s390_vxrs_active,
+		.get = s390_vxrs_low_get,
+		.set = s390_vxrs_low_set,
+	},
+	{
+		.core_note_type = NT_S390_VXRS_HIGH,
+		.n = __NUM_VXRS_HIGH,
+		.size = sizeof(__vector128),
+		.align = sizeof(__vector128),
+		.active = s390_vxrs_active,
+		.get = s390_vxrs_high_get,
+		.set = s390_vxrs_high_set,
 	},
-	[REGSET_GENERAL_EXTENDED] = {
+	{
 		.core_note_type = NT_S390_HIGH_GPRS,
 		.n = sizeof(s390_compat_regs_high) / sizeof(compat_long_t),
 		.size = sizeof(compat_long_t),

arch/s390/kernel/setup.c

@@ -343,6 +343,9 @@ static void __init setup_lowcore(void)
 		__ctl_set_bit(14, 29);
 	}
 #else
+	if (MACHINE_HAS_VX)
+		lc->vector_save_area_addr =
+			(unsigned long) &lc->vector_save_area;
 	lc->vdso_per_cpu_data = (unsigned long) &lc->paste[0];
 #endif
 	lc->sync_enter_timer = S390_lowcore.sync_enter_timer;
@@ -452,8 +455,8 @@ static void __init setup_memory_end(void)
 #ifdef CONFIG_64BIT
 	vmalloc_size = VMALLOC_END ?: (128UL << 30) - MODULES_LEN;
 	tmp = (memory_end ?: max_physmem_end) / PAGE_SIZE;
-	tmp = tmp * (sizeof(struct page) + PAGE_SIZE) + vmalloc_size;
-	if (tmp <= (1UL << 42))
+	tmp = tmp * (sizeof(struct page) + PAGE_SIZE);
+	if (tmp + vmalloc_size + MODULES_LEN <= (1UL << 42))
 		vmax = 1UL << 42;	/* 3-level kernel page table */
 	else
 		vmax = 1UL << 53;	/* 4-level kernel page table */
@@ -765,6 +768,12 @@ static void __init setup_hwcaps(void)
 	 */
 	if (test_facility(50) && test_facility(73))
 		elf_hwcap |= HWCAP_S390_TE;
+
+	/*
+	 * Vector extension HWCAP_S390_VXRS is bit 11.
+	 */
+	if (test_facility(129))
+		elf_hwcap |= HWCAP_S390_VXRS;
 #endif
 
 	get_cpu_id(&cpu_id);

arch/s390/kernel/signal.c

@@ -31,30 +31,117 @@
 #include <asm/switch_to.h>
 #include "entry.h"
 
-typedef struct 
+/*
+ * Layout of an old-style signal-frame:
+ *	-----------------------------------------
+ *	| save area (_SIGNAL_FRAMESIZE)		|
+ *	-----------------------------------------
+ *	| struct sigcontext			|
+ *	|	oldmask				|
+ *	|	_sigregs *			|
+ *	-----------------------------------------
+ *	| _sigregs with				|
+ *	|	_s390_regs_common		|
+ *	|	_s390_fp_regs			|
+ *	-----------------------------------------
+ *	| int signo				|
+ *	-----------------------------------------
+ *	| _sigregs_ext with			|
+ *	|	gprs_high 64 byte (opt)		|
+ *	|	vxrs_low 128 byte (opt)		|
+ *	|	vxrs_high 256 byte (opt)	|
+ *	|	reserved 128 byte (opt)		|
+ *	-----------------------------------------
+ *	| __u16 svc_insn			|
+ *	-----------------------------------------
+ * The svc_insn entry with the sigreturn system call opcode does not
+ * have a fixed position and moves if gprs_high or vxrs exist.
+ * Future extensions will be added to _sigregs_ext.
+ */
+struct sigframe
 {
 	__u8 callee_used_stack[__SIGNAL_FRAMESIZE];
 	struct sigcontext sc;
 	_sigregs sregs;
 	int signo;
-	__u8 retcode[S390_SYSCALL_SIZE];
-} sigframe;
+	_sigregs_ext sregs_ext;
+	__u16 svc_insn;		/* Offset of svc_insn is NOT fixed! */
+};
 
-typedef struct 
+/*
+ * Layout of an rt signal-frame:
+ *	-----------------------------------------
+ *	| save area (_SIGNAL_FRAMESIZE)		|
+ *	-----------------------------------------
+ *	| svc __NR_rt_sigreturn 2 byte		|
+ *	-----------------------------------------
+ *	| struct siginfo			|
+ *	-----------------------------------------
+ *	| struct ucontext_extended with		|
+ *	|	unsigned long uc_flags		|
+ *	|	struct ucontext *uc_link	|
+ *	|	stack_t uc_stack		|
+ *	|	_sigregs uc_mcontext with	|
+ *	|		_s390_regs_common	|
+ *	|		_s390_fp_regs		|
+ *	|	sigset_t uc_sigmask		|
+ *	|	_sigregs_ext uc_mcontext_ext	|
+ *	|		gprs_high 64 byte (opt)	|
+ *	|		vxrs_low 128 byte (opt)	|
+ *	|		vxrs_high 256 byte (opt)|
+ *	|		reserved 128 byte (opt)	|
+ *	-----------------------------------------
+ * Future extensions will be added to _sigregs_ext.
+ */
+struct rt_sigframe
 {
 	__u8 callee_used_stack[__SIGNAL_FRAMESIZE];
-	__u8 retcode[S390_SYSCALL_SIZE];
+	__u16 svc_insn;
 	struct siginfo info;
-	struct ucontext uc;
-} rt_sigframe;
+	struct ucontext_extended uc;
+};
+
+/* Store registers needed to create the signal frame */
+static void store_sigregs(void)
+{
+	save_access_regs(current->thread.acrs);
+	save_fp_ctl(&current->thread.fp_regs.fpc);
+#ifdef CONFIG_64BIT
+	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
+#endif
+		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 */
+#ifdef CONFIG_64BIT
+	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
+#endif
+		restore_fp_regs(current->thread.fp_regs.fprs);
+}
 
 /* Returns non-zero on fault. */
 static int save_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
 {
 	_sigregs user_sregs;
 
-	save_access_regs(current->thread.acrs);
-
 	/* Copy a 'clean' PSW mask to the user to avoid leaking
 	   information about whether PER is currently on.  */
 	user_sregs.regs.psw.mask = PSW_USER_BITS |
@@ -63,12 +150,6 @@ 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));
-	/* 
-	 * We have to store the fp registers to current->thread.fp_regs
-	 * to merge them with the emulated registers.
-	 */
-	save_fp_ctl(&current->thread.fp_regs.fpc);
-	save_fp_regs(current->thread.fp_regs.fprs);
 	memcpy(&user_sregs.fpregs, &current->thread.fp_regs,
 	       sizeof(user_sregs.fpregs));
 	if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs)))
@@ -107,20 +188,64 @@ static int restore_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
 	memcpy(&regs->gprs, &user_sregs.regs.gprs, sizeof(sregs->regs.gprs));
 	memcpy(&current->thread.acrs, &user_sregs.regs.acrs,
 	       sizeof(current->thread.acrs));
-	restore_access_regs(current->thread.acrs);
 
 	memcpy(&current->thread.fp_regs, &user_sregs.fpregs,
 	       sizeof(current->thread.fp_regs));
 
-	restore_fp_regs(current->thread.fp_regs.fprs);
 	clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */
 	return 0;
 }
 
+/* Returns non-zero on fault. */
+static int save_sigregs_ext(struct pt_regs *regs,
+			    _sigregs_ext __user *sregs_ext)
+{
+#ifdef CONFIG_64BIT
+	__u64 vxrs[__NUM_VXRS_LOW];
+	int i;
+
+	/* Save vector registers to signal stack */
+	if (current->thread.vxrs) {
+		for (i = 0; i < __NUM_VXRS_LOW; i++)
+			vxrs[i] = *((__u64 *)(current->thread.vxrs + i) + 1);
+		if (__copy_to_user(&sregs_ext->vxrs_low, vxrs,
+				   sizeof(sregs_ext->vxrs_low)) ||
+		    __copy_to_user(&sregs_ext->vxrs_high,
+				   current->thread.vxrs + __NUM_VXRS_LOW,
+				   sizeof(sregs_ext->vxrs_high)))
+			return -EFAULT;
+	}
+#endif
+	return 0;
+}
+
+static int restore_sigregs_ext(struct pt_regs *regs,
+			       _sigregs_ext __user *sregs_ext)
+{
+#ifdef CONFIG_64BIT
+	__u64 vxrs[__NUM_VXRS_LOW];
+	int i;
+
+	/* Restore vector registers from signal stack */
+	if (current->thread.vxrs) {
+		if (__copy_from_user(vxrs, &sregs_ext->vxrs_low,
+				     sizeof(sregs_ext->vxrs_low)) ||
+		    __copy_from_user(current->thread.vxrs + __NUM_VXRS_LOW,
+				     &sregs_ext->vxrs_high,
+				     sizeof(sregs_ext->vxrs_high)))
+			return -EFAULT;
+		for (i = 0; i < __NUM_VXRS_LOW; i++)
+			*((__u64 *)(current->thread.vxrs + i) + 1) = vxrs[i];
+	}
+#endif
+	return 0;
+}
+
 SYSCALL_DEFINE0(sigreturn)
 {
 	struct pt_regs *regs = task_pt_regs(current);
-	sigframe __user *frame = (sigframe __user *)regs->gprs[15];
+	struct sigframe __user *frame =
+		(struct sigframe __user *) regs->gprs[15];
 	sigset_t set;
 
 	if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE))
@@ -128,6 +253,9 @@ SYSCALL_DEFINE0(sigreturn)
 	set_current_blocked(&set);
 	if (restore_sigregs(regs, &frame->sregs))
 		goto badframe;
+	if (restore_sigregs_ext(regs, &frame->sregs_ext))
+		goto badframe;
+	load_sigregs();
 	return regs->gprs[2];
 badframe:
 	force_sig(SIGSEGV, current);
@@ -137,27 +265,26 @@ SYSCALL_DEFINE0(sigreturn)
 SYSCALL_DEFINE0(rt_sigreturn)
 {
 	struct pt_regs *regs = task_pt_regs(current);
-	rt_sigframe __user *frame = (rt_sigframe __user *)regs->gprs[15];
+	struct rt_sigframe __user *frame =
+		(struct rt_sigframe __user *)regs->gprs[15];
 	sigset_t set;
 
 	if (__copy_from_user(&set.sig, &frame->uc.uc_sigmask, sizeof(set)))
 		goto badframe;
 	set_current_blocked(&set);
+	if (restore_altstack(&frame->uc.uc_stack))
+		goto badframe;
 	if (restore_sigregs(regs, &frame->uc.uc_mcontext))
 		goto badframe;
-	if (restore_altstack(&frame->uc.uc_stack))
+	if (restore_sigregs_ext(regs, &frame->uc.uc_mcontext_ext))
 		goto badframe;
+	load_sigregs();
 	return regs->gprs[2];
 badframe:
 	force_sig(SIGSEGV, current);
 	return 0;
 }
 
-/*
- * Set up a signal frame.
- */
-
-
 /*
  * Determine which stack to use..
  */
@@ -195,39 +322,63 @@ static inline int map_signal(int sig)
 static int setup_frame(int sig, struct k_sigaction *ka,
 		       sigset_t *set, struct pt_regs * regs)
 {
-	sigframe __user *frame;
-
-	frame = get_sigframe(ka, regs, sizeof(sigframe));
+	struct sigframe __user *frame;
+	struct sigcontext sc;
+	unsigned long restorer;
+	size_t frame_size;
 
+	/*
+	 * gprs_high are only present for a 31-bit task running on
+	 * a 64-bit kernel (see compat_signal.c) but the space for
+	 * gprs_high need to be allocated if vector registers are
+	 * included in the signal frame on a 31-bit system.
+	 */
+	frame_size = sizeof(*frame) - sizeof(frame->sregs_ext);
+	if (MACHINE_HAS_VX)
+		frame_size += sizeof(frame->sregs_ext);
+	frame = get_sigframe(ka, regs, frame_size);
 	if (frame == (void __user *) -1UL)
 		return -EFAULT;
 
-	if (__copy_to_user(&frame->sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE))
+	/* Set up backchain. */
+	if (__put_user(regs->gprs[15], (addr_t __user *) frame))
 		return -EFAULT;
 
+	/* Create struct sigcontext on the signal stack */
+	memcpy(&sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE);
+	sc.sregs = (_sigregs __user __force *) &frame->sregs;
+	if (__copy_to_user(&frame->sc, &sc, sizeof(frame->sc)))
+		return -EFAULT;
+
+	/* Store registers needed to create the signal frame */
+	store_sigregs();
+
+	/* Create _sigregs on the signal stack */
 	if (save_sigregs(regs, &frame->sregs))
 		return -EFAULT;
-	if (__put_user(&frame->sregs, &frame->sc.sregs))
+
+	/* Place signal number on stack to allow backtrace from handler.  */
+	if (__put_user(regs->gprs[2], (int __user *) &frame->signo))
+		return -EFAULT;
+
+	/* Create _sigregs_ext on the signal stack */
+	if (save_sigregs_ext(regs, &frame->sregs_ext))
 		return -EFAULT;
 
 	/* Set up to return from userspace.  If provided, use a stub
 	   already in userspace.  */
 	if (ka->sa.sa_flags & SA_RESTORER) {
-                regs->gprs[14] = (unsigned long)
-			ka->sa.sa_restorer | PSW_ADDR_AMODE;
+		restorer = (unsigned long) ka->sa.sa_restorer | PSW_ADDR_AMODE;
 	} else {
-                regs->gprs[14] = (unsigned long)
-			frame->retcode | PSW_ADDR_AMODE;
-		if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn,
-	                       (u16 __user *)(frame->retcode)))
+		/* Signal frame without vector registers are short ! */
+		__u16 __user *svc = (void *) frame + frame_size - 2;
+		if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn, svc))
 			return -EFAULT;
+		restorer = (unsigned long) svc | PSW_ADDR_AMODE;
 	}
 
-	/* Set up backchain. */
-	if (__put_user(regs->gprs[15], (addr_t __user *) frame))
-		return -EFAULT;
-
 	/* Set up registers for signal handler */
+	regs->gprs[14] = restorer;
 	regs->gprs[15] = (unsigned long) frame;
 	/* Force default amode and default user address space control. */
 	regs->psw.mask = PSW_MASK_EA | PSW_MASK_BA |
@@ -247,54 +398,69 @@ static int setup_frame(int sig, struct k_sigaction *ka,
 		regs->gprs[5] = regs->int_parm_long;
 		regs->gprs[6] = task_thread_info(current)->last_break;
 	}
-
-	/* Place signal number on stack to allow backtrace from handler.  */
-	if (__put_user(regs->gprs[2], (int __user *) &frame->signo))
-		return -EFAULT;
 	return 0;
 }
 
 static int setup_rt_frame(struct ksignal *ksig, sigset_t *set,
 			  struct pt_regs *regs)
 {
-	int err = 0;
-	rt_sigframe __user *frame;
-
-	frame = get_sigframe(&ksig->ka, regs, sizeof(rt_sigframe));
+	struct rt_sigframe __user *frame;
+	unsigned long uc_flags, restorer;
+	size_t frame_size;
 
+	frame_size = sizeof(struct rt_sigframe) - sizeof(_sigregs_ext);
+	/*
+	 * gprs_high are only present for a 31-bit task running on
+	 * a 64-bit kernel (see compat_signal.c) but the space for
+	 * gprs_high need to be allocated if vector registers are
+	 * included in the signal frame on a 31-bit system.
+	 */
+	uc_flags = 0;
+#ifdef CONFIG_64BIT
+	if (MACHINE_HAS_VX) {
+		frame_size += sizeof(_sigregs_ext);
+		if (current->thread.vxrs)
+			uc_flags |= UC_VXRS;
+	}
+#endif
+	frame = get_sigframe(&ksig->ka, regs, frame_size);
 	if (frame == (void __user *) -1UL)
 		return -EFAULT;
 
-	if (copy_siginfo_to_user(&frame->info, &ksig->info))
-		return -EFAULT;
-
-	/* Create the ucontext.  */
-	err |= __put_user(0, &frame->uc.uc_flags);
-	err |= __put_user(NULL, &frame->uc.uc_link);
-	err |= __save_altstack(&frame->uc.uc_stack, regs->gprs[15]);
-	err |= save_sigregs(regs, &frame->uc.uc_mcontext);
-	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
-	if (err)
+	/* Set up backchain. */
+	if (__put_user(regs->gprs[15], (addr_t __user *) frame))
 		return -EFAULT;
 
 	/* Set up to return from userspace.  If provided, use a stub
 	   already in userspace.  */
 	if (ksig->ka.sa.sa_flags & SA_RESTORER) {
-                regs->gprs[14] = (unsigned long)
+		restorer = (unsigned long)
 			ksig->ka.sa.sa_restorer | PSW_ADDR_AMODE;
 	} else {
-                regs->gprs[14] = (unsigned long)
-			frame->retcode | PSW_ADDR_AMODE;
-		if (__put_user(S390_SYSCALL_OPCODE | __NR_rt_sigreturn,
-			       (u16 __user *)(frame->retcode)))
+		__u16 __user *svc = &frame->svc_insn;
+		if (__put_user(S390_SYSCALL_OPCODE | __NR_rt_sigreturn, svc))
 			return -EFAULT;
+		restorer = (unsigned long) svc | PSW_ADDR_AMODE;
 	}
 
-	/* Set up backchain. */
-	if (__put_user(regs->gprs[15], (addr_t __user *) frame))
+	/* Create siginfo on the signal stack */
+	if (copy_siginfo_to_user(&frame->info, &ksig->info))
+		return -EFAULT;
+
+	/* Store registers needed to create the signal frame */
+	store_sigregs();
+
+	/* Create ucontext on the signal stack. */
+	if (__put_user(uc_flags, &frame->uc.uc_flags) ||
+	    __put_user(NULL, &frame->uc.uc_link) ||
+	    __save_altstack(&frame->uc.uc_stack, regs->gprs[15]) ||
+	    save_sigregs(regs, &frame->uc.uc_mcontext) ||
+	    __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)) ||
+	    save_sigregs_ext(regs, &frame->uc.uc_mcontext_ext))
 		return -EFAULT;
 
 	/* Set up registers for signal handler */
+	regs->gprs[14] = restorer;
 	regs->gprs[15] = (unsigned long) frame;
 	/* Force default amode and default user address space control. */
 	regs->psw.mask = PSW_MASK_EA | PSW_MASK_BA |

arch/s390/kernel/smp.c

@@ -45,6 +45,7 @@
 #include <asm/debug.h>
 #include <asm/os_info.h>
 #include <asm/sigp.h>
+#include <asm/idle.h>
 #include "entry.h"
 
 enum {
@@ -82,7 +83,8 @@ DEFINE_MUTEX(smp_cpu_state_mutex);
 /*
  * Signal processor helper functions.
  */
-static inline int __pcpu_sigp_relax(u16 addr, u8 order, u32 parm, u32 *status)
+static inline int __pcpu_sigp_relax(u16 addr, u8 order, unsigned long parm,
+				    u32 *status)
 {
 	int cc;
 
@@ -178,6 +180,9 @@ static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu)
 			goto out;
 	}
 #else
+	if (MACHINE_HAS_VX)
+		lc->vector_save_area_addr =
+			(unsigned long) &lc->vector_save_area;
 	if (vdso_alloc_per_cpu(lc))
 		goto out;
 #endif
@@ -333,12 +338,6 @@ int smp_vcpu_scheduled(int cpu)
 	return pcpu_running(pcpu_devices + cpu);
 }
 
-void smp_yield(void)
-{
-	if (MACHINE_HAS_DIAG44)
-		asm volatile("diag 0,0,0x44");
-}
-
 void smp_yield_cpu(int cpu)
 {
 	if (MACHINE_HAS_DIAG9C)
@@ -517,35 +516,53 @@ EXPORT_SYMBOL(smp_ctl_clear_bit);
 static void __init smp_get_save_area(int cpu, u16 address)
 {
 	void *lc = pcpu_devices[0].lowcore;
-	struct save_area *save_area;
+	struct save_area_ext *sa_ext;
+	unsigned long vx_sa;
 
 	if (is_kdump_kernel())
 		return;
 	if (!OLDMEM_BASE && (address == boot_cpu_address ||
 			     ipl_info.type != IPL_TYPE_FCP_DUMP))
 		return;
-	save_area = dump_save_area_create(cpu);
-	if (!save_area)
+	sa_ext = dump_save_area_create(cpu);
+	if (!sa_ext)
 		panic("could not allocate memory for save area\n");
 	if (address == boot_cpu_address) {
 		/* Copy the registers of the boot cpu. */
-		copy_oldmem_page(1, (void *) save_area, sizeof(*save_area),
+		copy_oldmem_page(1, (void *) &sa_ext->sa, sizeof(sa_ext->sa),
 				 SAVE_AREA_BASE - PAGE_SIZE, 0);
+		if (MACHINE_HAS_VX)
+			save_vx_regs_safe(sa_ext->vx_regs);
 		return;
 	}
 	/* Get the registers of a non-boot cpu. */
 	__pcpu_sigp_relax(address, SIGP_STOP_AND_STORE_STATUS, 0, NULL);
-	memcpy_real(save_area, lc + SAVE_AREA_BASE, sizeof(*save_area));
+	memcpy_real(&sa_ext->sa, lc + SAVE_AREA_BASE, sizeof(sa_ext->sa));
+	if (!MACHINE_HAS_VX)
+		return;
+	/* Get the VX registers */
+	vx_sa = __get_free_page(GFP_KERNEL);
+	if (!vx_sa)
+		panic("could not allocate memory for VX save area\n");
+	__pcpu_sigp_relax(address, SIGP_STORE_ADDITIONAL_STATUS, vx_sa, NULL);
+	memcpy(sa_ext->vx_regs, (void *) vx_sa, sizeof(sa_ext->vx_regs));
+	free_page(vx_sa);
 }
 
 int smp_store_status(int cpu)
 {
+	unsigned long vx_sa;
 	struct pcpu *pcpu;
 
 	pcpu = pcpu_devices + cpu;
 	if (__pcpu_sigp_relax(pcpu->address, SIGP_STOP_AND_STORE_STATUS,
 			      0, NULL) != SIGP_CC_ORDER_CODE_ACCEPTED)
 		return -EIO;
+	if (!MACHINE_HAS_VX)
+		return 0;
+	vx_sa = __pa(pcpu->lowcore->vector_save_area_addr);
+	__pcpu_sigp_relax(pcpu->address, SIGP_STORE_ADDITIONAL_STATUS,
+			  vx_sa, NULL);
 	return 0;
 }
 
@@ -667,7 +684,7 @@ static void smp_start_secondary(void *cpuvoid)
 	cpu_init();
 	preempt_disable();
 	init_cpu_timer();
-	init_cpu_vtimer();
+	vtime_init();
 	pfault_init();
 	notify_cpu_starting(smp_processor_id());
 	set_cpu_online(smp_processor_id(), true);
@@ -726,6 +743,7 @@ int __cpu_disable(void)
 	cregs[6]  &= ~0xff000000UL;	/* disable all I/O interrupts */
 	cregs[14] &= ~0x1f000000UL;	/* disable most machine checks */
 	__ctl_load(cregs, 0, 15);
+	clear_cpu_flag(CIF_NOHZ_DELAY);
 	return 0;
 }
 
@@ -898,42 +916,6 @@ static struct attribute_group cpu_common_attr_group = {
 	.attrs = cpu_common_attrs,
 };
 
-static ssize_t show_idle_count(struct device *dev,
-				struct device_attribute *attr, char *buf)
-{
-	struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
-	unsigned long long idle_count;
-	unsigned int sequence;
-
-	do {
-		sequence = ACCESS_ONCE(idle->sequence);
-		idle_count = ACCESS_ONCE(idle->idle_count);
-		if (ACCESS_ONCE(idle->clock_idle_enter))
-			idle_count++;
-	} while ((sequence & 1) || (ACCESS_ONCE(idle->sequence) != sequence));
-	return sprintf(buf, "%llu\n", idle_count);
-}
-static DEVICE_ATTR(idle_count, 0444, show_idle_count, NULL);
-
-static ssize_t show_idle_time(struct device *dev,
-				struct device_attribute *attr, char *buf)
-{
-	struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
-	unsigned long long now, idle_time, idle_enter, idle_exit;
-	unsigned int sequence;
-
-	do {
-		now = get_tod_clock();
-		sequence = ACCESS_ONCE(idle->sequence);
-		idle_time = ACCESS_ONCE(idle->idle_time);
-		idle_enter = ACCESS_ONCE(idle->clock_idle_enter);
-		idle_exit = ACCESS_ONCE(idle->clock_idle_exit);
-	} while ((sequence & 1) || (ACCESS_ONCE(idle->sequence) != sequence));
-	idle_time += idle_enter ? ((idle_exit ? : now) - idle_enter) : 0;
-	return sprintf(buf, "%llu\n", idle_time >> 12);
-}
-static DEVICE_ATTR(idle_time_us, 0444, show_idle_time, NULL);
-
 static struct attribute *cpu_online_attrs[] = {
 	&dev_attr_idle_count.attr,
 	&dev_attr_idle_time_us.attr,

arch/s390/kernel/time.c

@@ -232,6 +232,19 @@ void update_vsyscall(struct timekeeper *tk)
 		vdso_data->wtom_clock_nsec -= nsecps;
 		vdso_data->wtom_clock_sec++;
 	}
+
+	vdso_data->xtime_coarse_sec = tk->xtime_sec;
+	vdso_data->xtime_coarse_nsec =
+		(long)(tk->tkr.xtime_nsec >> tk->tkr.shift);
+	vdso_data->wtom_coarse_sec =
+		vdso_data->xtime_coarse_sec + tk->wall_to_monotonic.tv_sec;
+	vdso_data->wtom_coarse_nsec =
+		vdso_data->xtime_coarse_nsec + tk->wall_to_monotonic.tv_nsec;
+	while (vdso_data->wtom_coarse_nsec >= NSEC_PER_SEC) {
+		vdso_data->wtom_coarse_nsec -= NSEC_PER_SEC;
+		vdso_data->wtom_coarse_sec++;
+	}
+
 	vdso_data->tk_mult = tk->tkr.mult;
 	vdso_data->tk_shift = tk->tkr.shift;
 	smp_wmb();

arch/s390/kernel/topology.c

@@ -464,15 +464,17 @@ static struct sched_domain_topology_level s390_topology[] = {
 
 static int __init topology_init(void)
 {
-	if (!MACHINE_HAS_TOPOLOGY) {
+	if (MACHINE_HAS_TOPOLOGY)
+		set_topology_timer();
+	else
 		topology_update_polarization_simple();
-		goto out;
-	}
-	set_topology_timer();
-out:
-
-	set_sched_topology(s390_topology);
-
 	return device_create_file(cpu_subsys.dev_root, &dev_attr_dispatching);
 }
 device_initcall(topology_init);
+
+static int __init early_topology_init(void)
+{
+	set_sched_topology(s390_topology);
+	return 0;
+}
+early_initcall(early_topology_init);

arch/s390/kernel/traps.c

@@ -18,6 +18,8 @@
 #include <linux/ptrace.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
+#include <linux/slab.h>
+#include <asm/switch_to.h>
 #include "entry.h"
 
 int show_unhandled_signals = 1;
@@ -58,15 +60,10 @@ int is_valid_bugaddr(unsigned long addr)
 	return 1;
 }
 
-static void __kprobes do_trap(struct pt_regs *regs,
-			      int si_signo, int si_code, char *str)
+void do_report_trap(struct pt_regs *regs, int si_signo, int si_code, char *str)
 {
 	siginfo_t info;
 
-	if (notify_die(DIE_TRAP, str, regs, 0,
-		       regs->int_code, si_signo) == NOTIFY_STOP)
-		return;
-
 	if (user_mode(regs)) {
 		info.si_signo = si_signo;
 		info.si_errno = 0;
@@ -90,6 +87,15 @@ static void __kprobes do_trap(struct pt_regs *regs,
         }
 }
 
+static void __kprobes do_trap(struct pt_regs *regs, int si_signo, int si_code,
+			      char *str)
+{
+	if (notify_die(DIE_TRAP, str, regs, 0,
+		       regs->int_code, si_signo) == NOTIFY_STOP)
+		return;
+	do_report_trap(regs, si_signo, si_code, str);
+}
+
 void __kprobes do_per_trap(struct pt_regs *regs)
 {
 	siginfo_t info;
@@ -178,6 +184,7 @@ void __kprobes illegal_op(struct pt_regs *regs)
 	siginfo_t info;
         __u8 opcode[6];
 	__u16 __user *location;
+	int is_uprobe_insn = 0;
 	int signal = 0;
 
 	location = get_trap_ip(regs);
@@ -194,6 +201,10 @@ void __kprobes illegal_op(struct pt_regs *regs)
 				force_sig_info(SIGTRAP, &info, current);
 			} else
 				signal = SIGILL;
+#ifdef CONFIG_UPROBES
+		} else if (*((__u16 *) opcode) == UPROBE_SWBP_INSN) {
+			is_uprobe_insn = 1;
+#endif
 #ifdef CONFIG_MATHEMU
 		} else if (opcode[0] == 0xb3) {
 			if (get_user(*((__u16 *) (opcode+2)), location+1))
@@ -219,11 +230,13 @@ void __kprobes illegal_op(struct pt_regs *regs)
 #endif
 		} else
 			signal = SIGILL;
-	} else {
-		/*
-		 * If we get an illegal op in kernel mode, send it through the
-		 * kprobes notifier. If kprobes doesn't pick it up, SIGILL
-		 */
+	}
+	/*
+	 * We got either an illegal op in kernel mode, or user space trapped
+	 * on a uprobes illegal instruction. See if kprobes or uprobes picks
+	 * it up. If not, SIGILL.
+	 */
+	if (is_uprobe_insn || !user_mode(regs)) {
 		if (notify_die(DIE_BPT, "bpt", regs, 0,
 			       3, SIGTRAP) != NOTIFY_STOP)
 			signal = SIGILL;
@@ -292,6 +305,74 @@ DO_ERROR_INFO(specification_exception, SIGILL, ILL_ILLOPN,
 	      "specification exception");
 #endif
 
+#ifdef CONFIG_64BIT
+int alloc_vector_registers(struct task_struct *tsk)
+{
+	__vector128 *vxrs;
+	int i;
+
+	/* Allocate vector register save area. */
+	vxrs = kzalloc(sizeof(__vector128) * __NUM_VXRS,
+		       GFP_KERNEL|__GFP_REPEAT);
+	if (!vxrs)
+		return -ENOMEM;
+	preempt_disable();
+	if (tsk == current)
+		save_fp_regs(tsk->thread.fp_regs.fprs);
+	/* Copy the 16 floating point registers */
+	for (i = 0; i < 16; i++)
+		*(freg_t *) &vxrs[i] = tsk->thread.fp_regs.fprs[i];
+	tsk->thread.vxrs = vxrs;
+	if (tsk == current) {
+		__ctl_set_bit(0, 17);
+		restore_vx_regs(vxrs);
+	}
+	preempt_enable();
+	return 0;
+}
+
+void vector_exception(struct pt_regs *regs)
+{
+	int si_code, vic;
+
+	if (!MACHINE_HAS_VX) {
+		do_trap(regs, SIGILL, ILL_ILLOPN, "illegal operation");
+		return;
+	}
+
+	/* get vector interrupt code from fpc */
+	asm volatile("stfpc %0" : "=m" (current->thread.fp_regs.fpc));
+	vic = (current->thread.fp_regs.fpc & 0xf00) >> 8;
+	switch (vic) {
+	case 1: /* invalid vector operation */
+		si_code = FPE_FLTINV;
+		break;
+	case 2: /* division by zero */
+		si_code = FPE_FLTDIV;
+		break;
+	case 3: /* overflow */
+		si_code = FPE_FLTOVF;
+		break;
+	case 4: /* underflow */
+		si_code = FPE_FLTUND;
+		break;
+	case 5:	/* inexact */
+		si_code = FPE_FLTRES;
+		break;
+	default: /* unknown cause */
+		si_code = 0;
+	}
+	do_trap(regs, SIGFPE, si_code, "vector exception");
+}
+
+static int __init disable_vector_extension(char *str)
+{
+	S390_lowcore.machine_flags &= ~MACHINE_FLAG_VX;
+	return 1;
+}
+__setup("novx", disable_vector_extension);
+#endif
+
 void data_exception(struct pt_regs *regs)
 {
 	__u16 __user *location;
@@ -357,6 +438,18 @@ void data_exception(struct pt_regs *regs)
                 }
         }
 #endif 
+#ifdef CONFIG_64BIT
+	/* Check for vector register enablement */
+	if (MACHINE_HAS_VX && !current->thread.vxrs &&
+	    (current->thread.fp_regs.fpc & FPC_DXC_MASK) == 0xfe00) {
+		alloc_vector_registers(current);
+		/* Vector data exception is suppressing, rewind psw. */
+		regs->psw.addr = __rewind_psw(regs->psw, regs->int_code >> 16);
+		clear_pt_regs_flag(regs, PIF_PER_TRAP);
+		return;
+	}
+#endif
+
 	if (current->thread.fp_regs.fpc & FPC_DXC_MASK)
 		signal = SIGFPE;
 	else

arch/s390/kernel/uprobes.c

+/*
+ *  User-space Probes (UProbes) for s390
+ *
+ *    Copyright IBM Corp. 2014
+ *    Author(s): Jan Willeke,
+ */
+
+#include <linux/kprobes.h>
+#include <linux/uaccess.h>
+#include <linux/uprobes.h>
+#include <linux/compat.h>
+#include <linux/kdebug.h>
+#include <asm/switch_to.h>
+#include <asm/facility.h>
+#include <asm/dis.h>
+#include "entry.h"
+
+#define	UPROBE_TRAP_NR	UINT_MAX
+
+int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm,
+			     unsigned long addr)
+{
+	return probe_is_prohibited_opcode(auprobe->insn);
+}
+
+int arch_uprobe_pre_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	if (psw_bits(regs->psw).eaba == PSW_AMODE_24BIT)
+		return -EINVAL;
+	if (!is_compat_task() && psw_bits(regs->psw).eaba == PSW_AMODE_31BIT)
+		return -EINVAL;
+	clear_pt_regs_flag(regs, PIF_PER_TRAP);
+	auprobe->saved_per = psw_bits(regs->psw).r;
+	auprobe->saved_int_code = regs->int_code;
+	regs->int_code = UPROBE_TRAP_NR;
+	regs->psw.addr = current->utask->xol_vaddr;
+	set_tsk_thread_flag(current, TIF_UPROBE_SINGLESTEP);
+	update_cr_regs(current);
+	return 0;
+}
+
+bool arch_uprobe_xol_was_trapped(struct task_struct *tsk)
+{
+	struct pt_regs *regs = task_pt_regs(tsk);
+
+	if (regs->int_code != UPROBE_TRAP_NR)
+		return true;
+	return false;
+}
+
+int arch_uprobe_post_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	int fixup = probe_get_fixup_type(auprobe->insn);
+	struct uprobe_task *utask = current->utask;
+
+	clear_tsk_thread_flag(current, TIF_UPROBE_SINGLESTEP);
+	update_cr_regs(current);
+	psw_bits(regs->psw).r = auprobe->saved_per;
+	regs->int_code = auprobe->saved_int_code;
+
+	if (fixup & FIXUP_PSW_NORMAL)
+		regs->psw.addr += utask->vaddr - utask->xol_vaddr;
+	if (fixup & FIXUP_RETURN_REGISTER) {
+		int reg = (auprobe->insn[0] & 0xf0) >> 4;
+
+		regs->gprs[reg] += utask->vaddr - utask->xol_vaddr;
+	}
+	if (fixup & FIXUP_BRANCH_NOT_TAKEN) {
+		int ilen = insn_length(auprobe->insn[0] >> 8);
+
+		if (regs->psw.addr - utask->xol_vaddr == ilen)
+			regs->psw.addr = utask->vaddr + ilen;
+	}
+	/* If per tracing was active generate trap */
+	if (regs->psw.mask & PSW_MASK_PER)
+		do_per_trap(regs);
+	return 0;
+}
+
+int arch_uprobe_exception_notify(struct notifier_block *self, unsigned long val,
+				 void *data)
+{
+	struct die_args *args = data;
+	struct pt_regs *regs = args->regs;
+
+	if (!user_mode(regs))
+		return NOTIFY_DONE;
+	if (regs->int_code & 0x200) /* Trap during transaction */
+		return NOTIFY_DONE;
+	switch (val) {
+	case DIE_BPT:
+		if (uprobe_pre_sstep_notifier(regs))
+			return NOTIFY_STOP;
+		break;
+	case DIE_SSTEP:
+		if (uprobe_post_sstep_notifier(regs))
+			return NOTIFY_STOP;
+	default:
+		break;
+	}
+	return NOTIFY_DONE;
+}
+
+void arch_uprobe_abort_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	clear_thread_flag(TIF_UPROBE_SINGLESTEP);
+	regs->int_code = auprobe->saved_int_code;
+	regs->psw.addr = current->utask->vaddr;
+}
+
+unsigned long arch_uretprobe_hijack_return_addr(unsigned long trampoline,
+						struct pt_regs *regs)
+{
+	unsigned long orig;
+
+	orig = regs->gprs[14];
+	regs->gprs[14] = trampoline;
+	return orig;
+}
+
+/* Instruction Emulation */
+
+static void adjust_psw_addr(psw_t *psw, unsigned long len)
+{
+	psw->addr = __rewind_psw(*psw, -len);
+}
+
+#define EMU_ILLEGAL_OP		1
+#define EMU_SPECIFICATION	2
+#define EMU_ADDRESSING		3
+
+#define emu_load_ril(ptr, output)			\
+({							\
+	unsigned int mask = sizeof(*(ptr)) - 1;		\
+	__typeof__(*(ptr)) input;			\
+	int __rc = 0;					\
+							\
+	if (!test_facility(34))				\
+		__rc = EMU_ILLEGAL_OP;			\
+	else if ((u64 __force)ptr & mask)		\
+		__rc = EMU_SPECIFICATION;		\
+	else if (get_user(input, ptr))			\
+		__rc = EMU_ADDRESSING;			\
+	else						\
+		*(output) = input;			\
+	__rc;						\
+})
+
+#define emu_store_ril(ptr, input)			\
+({							\
+	unsigned int mask = sizeof(*(ptr)) - 1;		\
+	int __rc = 0;					\
+							\
+	if (!test_facility(34))				\
+		__rc = EMU_ILLEGAL_OP;			\
+	else if ((u64 __force)ptr & mask)		\
+		__rc = EMU_SPECIFICATION;		\
+	else if (put_user(*(input), ptr))		\
+		__rc = EMU_ADDRESSING;			\
+	__rc;						\
+})
+
+#define emu_cmp_ril(regs, ptr, cmp)			\
+({							\
+	unsigned int mask = sizeof(*(ptr)) - 1;		\
+	__typeof__(*(ptr)) input;			\
+	int __rc = 0;					\
+							\
+	if (!test_facility(34))				\
+		__rc = EMU_ILLEGAL_OP;			\
+	else if ((u64 __force)ptr & mask)		\
+		__rc = EMU_SPECIFICATION;		\
+	else if (get_user(input, ptr))			\
+		__rc = EMU_ADDRESSING;			\
+	else if (input > *(cmp))			\
+		psw_bits((regs)->psw).cc = 1;		\
+	else if (input < *(cmp))			\
+		psw_bits((regs)->psw).cc = 2;		\
+	else						\
+		psw_bits((regs)->psw).cc = 0;		\
+	__rc;						\
+})
+
+struct insn_ril {
+	u8 opc0;
+	u8 reg	: 4;
+	u8 opc1 : 4;
+	s32 disp;
+} __packed;
+
+union split_register {
+	u64 u64;
+	u32 u32[2];
+	u16 u16[4];
+	s64 s64;
+	s32 s32[2];
+	s16 s16[4];
+};
+
+/*
+ * pc relative instructions are emulated, since parameters may not be
+ * accessible from the xol area due to range limitations.
+ */
+static void handle_insn_ril(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	union split_register *rx;
+	struct insn_ril *insn;
+	unsigned int ilen;
+	void *uptr;
+	int rc = 0;
+
+	insn = (struct insn_ril *) &auprobe->insn;
+	rx = (union split_register *) &regs->gprs[insn->reg];
+	uptr = (void *)(regs->psw.addr + (insn->disp * 2));
+	ilen = insn_length(insn->opc0);
+
+	switch (insn->opc0) {
+	case 0xc0:
+		switch (insn->opc1) {
+		case 0x00: /* larl */
+			rx->u64 = (unsigned long)uptr;
+			break;
+		}
+		break;
+	case 0xc4:
+		switch (insn->opc1) {
+		case 0x02: /* llhrl */
+			rc = emu_load_ril((u16 __user *)uptr, &rx->u32[1]);
+			break;
+		case 0x04: /* lghrl */
+			rc = emu_load_ril((s16 __user *)uptr, &rx->u64);
+			break;
+		case 0x05: /* lhrl */
+			rc = emu_load_ril((s16 __user *)uptr, &rx->u32[1]);
+			break;
+		case 0x06: /* llghrl */
+			rc = emu_load_ril((u16 __user *)uptr, &rx->u64);
+			break;
+		case 0x08: /* lgrl */
+			rc = emu_load_ril((u64 __user *)uptr, &rx->u64);
+			break;
+		case 0x0c: /* lgfrl */
+			rc = emu_load_ril((s32 __user *)uptr, &rx->u64);
+			break;
+		case 0x0d: /* lrl */
+			rc = emu_load_ril((u32 __user *)uptr, &rx->u32[1]);
+			break;
+		case 0x0e: /* llgfrl */
+			rc = emu_load_ril((u32 __user *)uptr, &rx->u64);
+			break;
+		case 0x07: /* sthrl */
+			rc = emu_store_ril((u16 __user *)uptr, &rx->u16[3]);
+			break;
+		case 0x0b: /* stgrl */
+			rc = emu_store_ril((u64 __user *)uptr, &rx->u64);
+			break;
+		case 0x0f: /* strl */
+			rc = emu_store_ril((u32 __user *)uptr, &rx->u32[1]);
+			break;
+		}
+		break;
+	case 0xc6:
+		switch (insn->opc1) {
+		case 0x02: /* pfdrl */
+			if (!test_facility(34))
+				rc = EMU_ILLEGAL_OP;
+			break;
+		case 0x04: /* cghrl */
+			rc = emu_cmp_ril(regs, (s16 __user *)uptr, &rx->s64);
+			break;
+		case 0x05: /* chrl */
+			rc = emu_cmp_ril(regs, (s16 __user *)uptr, &rx->s32[1]);
+			break;
+		case 0x06: /* clghrl */
+			rc = emu_cmp_ril(regs, (u16 __user *)uptr, &rx->u64);
+			break;
+		case 0x07: /* clhrl */
+			rc = emu_cmp_ril(regs, (u16 __user *)uptr, &rx->u32[1]);
+			break;
+		case 0x08: /* cgrl */
+			rc = emu_cmp_ril(regs, (s64 __user *)uptr, &rx->s64);
+			break;
+		case 0x0a: /* clgrl */
+			rc = emu_cmp_ril(regs, (u64 __user *)uptr, &rx->u64);
+			break;
+		case 0x0c: /* cgfrl */
+			rc = emu_cmp_ril(regs, (s32 __user *)uptr, &rx->s64);
+			break;
+		case 0x0d: /* crl */
+			rc = emu_cmp_ril(regs, (s32 __user *)uptr, &rx->s32[1]);
+			break;
+		case 0x0e: /* clgfrl */
+			rc = emu_cmp_ril(regs, (u32 __user *)uptr, &rx->u64);
+			break;
+		case 0x0f: /* clrl */
+			rc = emu_cmp_ril(regs, (u32 __user *)uptr, &rx->u32[1]);
+			break;
+		}
+		break;
+	}
+	adjust_psw_addr(&regs->psw, ilen);
+	switch (rc) {
+	case EMU_ILLEGAL_OP:
+		regs->int_code = ilen << 16 | 0x0001;
+		do_report_trap(regs, SIGILL, ILL_ILLOPC, NULL);
+		break;
+	case EMU_SPECIFICATION:
+		regs->int_code = ilen << 16 | 0x0006;
+		do_report_trap(regs, SIGILL, ILL_ILLOPC , NULL);
+		break;
+	case EMU_ADDRESSING:
+		regs->int_code = ilen << 16 | 0x0005;
+		do_report_trap(regs, SIGSEGV, SEGV_MAPERR, NULL);
+		break;
+	}
+}
+
+bool arch_uprobe_skip_sstep(struct arch_uprobe *auprobe, struct pt_regs *regs)
+{
+	if ((psw_bits(regs->psw).eaba == PSW_AMODE_24BIT) ||
+	    ((psw_bits(regs->psw).eaba == PSW_AMODE_31BIT) &&
+	     !is_compat_task())) {
+		regs->psw.addr = __rewind_psw(regs->psw, UPROBE_SWBP_INSN_SIZE);
+		do_report_trap(regs, SIGILL, ILL_ILLADR, NULL);
+		return true;
+	}
+	if (probe_is_insn_relative_long(auprobe->insn)) {
+		handle_insn_ril(auprobe, regs);
+		return true;
+	}
+	return false;
+}

arch/s390/kernel/vdso32/clock_getres.S

@@ -19,14 +19,20 @@
 	.type  __kernel_clock_getres,@function
 __kernel_clock_getres:
 	.cfi_startproc
+	basr	%r1,0
+	la	%r1,4f-.(%r1)
 	chi	%r2,__CLOCK_REALTIME
 	je	0f
 	chi	%r2,__CLOCK_MONOTONIC
+	je	0f
+	la	%r1,5f-4f(%r1)
+	chi	%r2,__CLOCK_REALTIME_COARSE
+	je	0f
+	chi	%r2,__CLOCK_MONOTONIC_COARSE
 	jne	3f
 0:	ltr	%r3,%r3
 	jz	2f				/* res == NULL */
-	basr	%r1,0
-1:	l	%r0,4f-1b(%r1)
+1:	l	%r0,0(%r1)
 	xc	0(4,%r3),0(%r3)			/* set tp->tv_sec to zero */
 	st	%r0,4(%r3)			/* store tp->tv_usec */
 2:	lhi	%r2,0
@@ -35,5 +41,6 @@ __kernel_clock_getres:
 	svc	0
 	br	%r14
 4:	.long	__CLOCK_REALTIME_RES
+5:	.long	__CLOCK_COARSE_RES
 	.cfi_endproc
 	.size	__kernel_clock_getres,.-__kernel_clock_getres

arch/s390/kernel/vdso32/clock_gettime.S

@@ -21,8 +21,12 @@ __kernel_clock_gettime:
 	.cfi_startproc
 	basr	%r5,0
 0:	al	%r5,21f-0b(%r5)			/* get &_vdso_data */
+	chi	%r2,__CLOCK_REALTIME_COARSE
+	je	10f
 	chi	%r2,__CLOCK_REALTIME
 	je	11f
+	chi	%r2,__CLOCK_MONOTONIC_COARSE
+	je	9f
 	chi	%r2,__CLOCK_MONOTONIC
 	jne	19f
 
@@ -30,8 +34,8 @@ __kernel_clock_gettime:
 1:	l	%r4,__VDSO_UPD_COUNT+4(%r5)	/* load update counter */
 	tml	%r4,0x0001			/* pending update ? loop */
 	jnz	1b
-	stck	24(%r15)			/* Store TOD clock */
-	lm	%r0,%r1,24(%r15)
+	stcke	24(%r15)			/* Store TOD clock */
+	lm	%r0,%r1,25(%r15)
 	s	%r0,__VDSO_XTIME_STAMP(%r5)	/* TOD - cycle_last */
 	sl	%r1,__VDSO_XTIME_STAMP+4(%r5)
 	brc	3,2f
@@ -68,12 +72,32 @@ __kernel_clock_gettime:
 	lhi	%r2,0
 	br	%r14
 
+	/* CLOCK_MONOTONIC_COARSE */
+9:	l	%r4,__VDSO_UPD_COUNT+4(%r5)	/* load update counter */
+	tml	%r4,0x0001			/* pending update ? loop */
+	jnz	9b
+	l	%r2,__VDSO_WTOM_CRS_SEC+4(%r5)
+	l	%r1,__VDSO_WTOM_CRS_NSEC+4(%r5)
+	cl	%r4,__VDSO_UPD_COUNT+4(%r5)	/* check update counter */
+	jne	9b
+	j	8b
+
+	/* CLOCK_REALTIME_COARSE */
+10:	l	%r4,__VDSO_UPD_COUNT+4(%r5)	/* load update counter */
+	tml	%r4,0x0001			/* pending update ? loop */
+	jnz	10b
+	l	%r2,__VDSO_XTIME_CRS_SEC+4(%r5)
+	l	%r1,__VDSO_XTIME_CRS_NSEC+4(%r5)
+	cl	%r4,__VDSO_UPD_COUNT+4(%r5)	/* check update counter */
+	jne	10b
+	j	17f
+
 	/* CLOCK_REALTIME */
 11:	l	%r4,__VDSO_UPD_COUNT+4(%r5)	/* load update counter */
 	tml	%r4,0x0001			/* pending update ? loop */
 	jnz	11b
-	stck	24(%r15)			/* Store TOD clock */
-	lm	%r0,%r1,24(%r15)
+	stcke	24(%r15)			/* Store TOD clock */
+	lm	%r0,%r1,25(%r15)
 	s	%r0,__VDSO_XTIME_STAMP(%r5)	/* TOD - cycle_last */
 	sl	%r1,__VDSO_XTIME_STAMP+4(%r5)
 	brc	3,12f

arch/s390/kernel/vdso32/gettimeofday.S

@@ -29,8 +29,8 @@ __kernel_gettimeofday:
 	l	%r4,__VDSO_UPD_COUNT+4(%r5)	/* load update counter */
 	tml	%r4,0x0001			/* pending update ? loop */
 	jnz	1b
-	stck	24(%r15)			/* Store TOD clock */
-	lm	%r0,%r1,24(%r15)
+	stcke	24(%r15)			/* Store TOD clock */
+	lm	%r0,%r1,25(%r15)
 	s	%r0,__VDSO_XTIME_STAMP(%r5)	/* TOD - cycle_last */
 	sl	%r1,__VDSO_XTIME_STAMP+4(%r5)
 	brc	3,3f

arch/s390/kernel/vdso64/clock_getres.S

@@ -19,6 +19,12 @@
 	.type  __kernel_clock_getres,@function
 __kernel_clock_getres:
 	.cfi_startproc
+	larl	%r1,4f
+	cghi	%r2,__CLOCK_REALTIME_COARSE
+	je	0f
+	cghi	%r2,__CLOCK_MONOTONIC_COARSE
+	je	0f
+	larl	%r1,3f
 	cghi	%r2,__CLOCK_REALTIME
 	je	0f
 	cghi	%r2,__CLOCK_MONOTONIC
@@ -32,7 +38,6 @@ __kernel_clock_getres:
 	jz	2f
 0:	ltgr	%r3,%r3
 	jz	1f				/* res == NULL */
-	larl	%r1,3f
 	lg	%r0,0(%r1)
 	xc	0(8,%r3),0(%r3)			/* set tp->tv_sec to zero */
 	stg	%r0,8(%r3)			/* store tp->tv_usec */
@@ -42,5 +47,6 @@ __kernel_clock_getres:
 	svc	0
 	br	%r14
 3:	.quad	__CLOCK_REALTIME_RES
+4:	.quad	__CLOCK_COARSE_RES
 	.cfi_endproc
 	.size	__kernel_clock_getres,.-__kernel_clock_getres

arch/s390/kernel/vdso64/clock_gettime.S

@@ -20,12 +20,16 @@
 __kernel_clock_gettime:
 	.cfi_startproc
 	larl	%r5,_vdso_data
+	cghi	%r2,__CLOCK_REALTIME_COARSE
+	je	4f
 	cghi	%r2,__CLOCK_REALTIME
 	je	5f
 	cghi	%r2,__CLOCK_THREAD_CPUTIME_ID
 	je	9f
 	cghi	%r2,-2		/* Per-thread CPUCLOCK with PID=0, VIRT=1 */
 	je	9f
+	cghi	%r2,__CLOCK_MONOTONIC_COARSE
+	je	3f
 	cghi	%r2,__CLOCK_MONOTONIC
 	jne	12f
 
@@ -33,10 +37,10 @@ __kernel_clock_gettime:
 0:	lg	%r4,__VDSO_UPD_COUNT(%r5)	/* load update counter */
 	tmll	%r4,0x0001			/* pending update ? loop */
 	jnz	0b
-	stck	48(%r15)			/* Store TOD clock */
+	stcke	48(%r15)			/* Store TOD clock */
 	lgf	%r2,__VDSO_TK_SHIFT(%r5)	/* Timekeeper shift */
 	lg	%r0,__VDSO_WTOM_SEC(%r5)
-	lg	%r1,48(%r15)
+	lg	%r1,49(%r15)
 	sg	%r1,__VDSO_XTIME_STAMP(%r5)	/* TOD - cycle_last */
 	msgf	%r1,__VDSO_TK_MULT(%r5)		/*  * tk->mult */
 	alg	%r1,__VDSO_WTOM_NSEC(%r5)
@@ -54,13 +58,33 @@ __kernel_clock_gettime:
 	lghi	%r2,0
 	br	%r14
 
+	/* CLOCK_MONOTONIC_COARSE */
+3:	lg	%r4,__VDSO_UPD_COUNT(%r5)	/* load update counter */
+	tmll	%r4,0x0001			/* pending update ? loop */
+	jnz	3b
+	lg	%r0,__VDSO_WTOM_CRS_SEC(%r5)
+	lg	%r1,__VDSO_WTOM_CRS_NSEC(%r5)
+	clg	%r4,__VDSO_UPD_COUNT(%r5)	/* check update counter */
+	jne	3b
+	j	2b
+
+	/* CLOCK_REALTIME_COARSE */
+4:	lg	%r4,__VDSO_UPD_COUNT(%r5)	/* load update counter */
+	tmll	%r4,0x0001			/* pending update ? loop */
+	jnz	4b
+	lg	%r0,__VDSO_XTIME_CRS_SEC(%r5)
+	lg	%r1,__VDSO_XTIME_CRS_NSEC(%r5)
+	clg	%r4,__VDSO_UPD_COUNT(%r5)	/* check update counter */
+	jne	4b
+	j	7f
+
 	/* CLOCK_REALTIME */
 5:	lg	%r4,__VDSO_UPD_COUNT(%r5)	/* load update counter */
 	tmll	%r4,0x0001			/* pending update ? loop */
 	jnz	5b
-	stck	48(%r15)			/* Store TOD clock */
+	stcke	48(%r15)			/* Store TOD clock */
 	lgf	%r2,__VDSO_TK_SHIFT(%r5)	/* Timekeeper shift */
-	lg	%r1,48(%r15)
+	lg	%r1,49(%r15)
 	sg	%r1,__VDSO_XTIME_STAMP(%r5)	/* TOD - cycle_last */
 	msgf	%r1,__VDSO_TK_MULT(%r5)		/*  * tk->mult */
 	alg	%r1,__VDSO_XTIME_NSEC(%r5)	/*  + tk->xtime_nsec */

arch/s390/kernel/vdso64/gettimeofday.S

@@ -28,8 +28,8 @@ __kernel_gettimeofday:
 	lg	%r4,__VDSO_UPD_COUNT(%r5)	/* load update counter */
 	tmll	%r4,0x0001			/* pending update ? loop */
 	jnz	0b
-	stck	48(%r15)			/* Store TOD clock */
-	lg	%r1,48(%r15)
+	stcke	48(%r15)			/* Store TOD clock */
+	lg	%r1,49(%r15)
 	sg	%r1,__VDSO_XTIME_STAMP(%r5)	/* TOD - cycle_last */
 	msgf	%r1,__VDSO_TK_MULT(%r5)		/*  * tk->mult */
 	alg	%r1,__VDSO_XTIME_NSEC(%r5)	/*  + tk->xtime_nsec */

arch/s390/kernel/vtime.c

@@ -6,27 +6,18 @@
  */
 
 #include <linux/kernel_stat.h>
-#include <linux/notifier.h>
-#include <linux/kprobes.h>
 #include <linux/export.h>
 #include <linux/kernel.h>
 #include <linux/timex.h>
 #include <linux/types.h>
 #include <linux/time.h>
-#include <linux/cpu.h>
-#include <linux/smp.h>
 
-#include <asm/irq_regs.h>
 #include <asm/cputime.h>
 #include <asm/vtimer.h>
 #include <asm/vtime.h>
-#include <asm/irq.h>
-#include "entry.h"
 
 static void virt_timer_expire(void);
 
-DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
-
 static LIST_HEAD(virt_timer_list);
 static DEFINE_SPINLOCK(virt_timer_lock);
 static atomic64_t virt_timer_current;
@@ -152,49 +143,6 @@ void vtime_account_system(struct task_struct *tsk)
 __attribute__((alias("vtime_account_irq_enter")));
 EXPORT_SYMBOL_GPL(vtime_account_system);
 
-void __kprobes vtime_stop_cpu(void)
-{
-	struct s390_idle_data *idle = &__get_cpu_var(s390_idle);
-	unsigned long long idle_time;
-	unsigned long psw_mask;
-
-	trace_hardirqs_on();
-
-	/* Wait for external, I/O or machine check interrupt. */
-	psw_mask = PSW_KERNEL_BITS | PSW_MASK_WAIT | PSW_MASK_DAT |
-		PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK;
-	idle->nohz_delay = 0;
-
-	/* Call the assembler magic in entry.S */
-	psw_idle(idle, psw_mask);
-
-	/* Account time spent with enabled wait psw loaded as idle time. */
-	idle->sequence++;
-	smp_wmb();
-	idle_time = idle->clock_idle_exit - idle->clock_idle_enter;
-	idle->clock_idle_enter = idle->clock_idle_exit = 0ULL;
-	idle->idle_time += idle_time;
-	idle->idle_count++;
-	account_idle_time(idle_time);
-	smp_wmb();
-	idle->sequence++;
-}
-
-cputime64_t s390_get_idle_time(int cpu)
-{
-	struct s390_idle_data *idle = &per_cpu(s390_idle, cpu);
-	unsigned long long now, idle_enter, idle_exit;
-	unsigned int sequence;
-
-	do {
-		now = get_tod_clock();
-		sequence = ACCESS_ONCE(idle->sequence);
-		idle_enter = ACCESS_ONCE(idle->clock_idle_enter);
-		idle_exit = ACCESS_ONCE(idle->clock_idle_exit);
-	} while ((sequence & 1) || (ACCESS_ONCE(idle->sequence) != sequence));
-	return idle_enter ? ((idle_exit ?: now) - idle_enter) : 0;
-}
-
 /*
  * Sorted add to a list. List is linear searched until first bigger
  * element is found.
@@ -372,31 +320,8 @@ EXPORT_SYMBOL(del_virt_timer);
 /*
  * Start the virtual CPU timer on the current CPU.
  */
-void init_cpu_vtimer(void)
+void vtime_init(void)
 {
 	/* set initial cpu timer */
 	set_vtimer(VTIMER_MAX_SLICE);
 }
-
-static int s390_nohz_notify(struct notifier_block *self, unsigned long action,
-			    void *hcpu)
-{
-	struct s390_idle_data *idle;
-	long cpu = (long) hcpu;
-
-	idle = &per_cpu(s390_idle, cpu);
-	switch (action & ~CPU_TASKS_FROZEN) {
-	case CPU_DYING:
-		idle->nohz_delay = 0;
-	default:
-		break;
-	}
-	return NOTIFY_OK;
-}
-
-void __init vtime_init(void)
-{
-	/* Enable cpu timer interrupts on the boot cpu. */
-	init_cpu_vtimer();
-	cpu_notifier(s390_nohz_notify, 0);
-}

arch/s390/lib/Makefile

@@ -6,3 +6,5 @@ lib-y += delay.o string.o uaccess.o find.o
 obj-$(CONFIG_32BIT) += div64.o qrnnd.o ucmpdi2.o mem32.o
 obj-$(CONFIG_64BIT) += mem64.o
 lib-$(CONFIG_SMP) += spinlock.o
+lib-$(CONFIG_KPROBES) += probes.o
+lib-$(CONFIG_UPROBES) += probes.o

arch/s390/lib/delay.c

@@ -43,7 +43,7 @@ static void __udelay_disabled(unsigned long long usecs)
 	lockdep_off();
 	do {
 		set_clock_comparator(end);
-		vtime_stop_cpu();
+		enabled_wait();
 	} while (get_tod_clock_fast() < end);
 	lockdep_on();
 	__ctl_load(cr0, 0, 0);
@@ -62,7 +62,7 @@ static void __udelay_enabled(unsigned long long usecs)
 			clock_saved = local_tick_disable();
 			set_clock_comparator(end);
 		}
-		vtime_stop_cpu();
+		enabled_wait();
 		if (clock_saved)
 			local_tick_enable(clock_saved);
 	} while (get_tod_clock_fast() < end);

arch/s390/lib/probes.c

+/*
+ *    Common helper functions for kprobes and uprobes
+ *
+ *    Copyright IBM Corp. 2014
+ */
+
+#include <linux/kprobes.h>
+#include <asm/dis.h>
+
+int probe_is_prohibited_opcode(u16 *insn)
+{
+	if (!is_known_insn((unsigned char *)insn))
+		return -EINVAL;
+	switch (insn[0] >> 8) {
+	case 0x0c:	/* bassm */
+	case 0x0b:	/* bsm	 */
+	case 0x83:	/* diag  */
+	case 0x44:	/* ex	 */
+	case 0xac:	/* stnsm */
+	case 0xad:	/* stosm */
+		return -EINVAL;
+	case 0xc6:
+		switch (insn[0] & 0x0f) {
+		case 0x00: /* exrl   */
+			return -EINVAL;
+		}
+	}
+	switch (insn[0]) {
+	case 0x0101:	/* pr	 */
+	case 0xb25a:	/* bsa	 */
+	case 0xb240:	/* bakr  */
+	case 0xb258:	/* bsg	 */
+	case 0xb218:	/* pc	 */
+	case 0xb228:	/* pt	 */
+	case 0xb98d:	/* epsw	 */
+	case 0xe560:	/* tbegin */
+	case 0xe561:	/* tbeginc */
+	case 0xb2f8:	/* tend	 */
+		return -EINVAL;
+	}
+	return 0;
+}
+
+int probe_get_fixup_type(u16 *insn)
+{
+	/* default fixup method */
+	int fixup = FIXUP_PSW_NORMAL;
+
+	switch (insn[0] >> 8) {
+	case 0x05:	/* balr	*/
+	case 0x0d:	/* basr */
+		fixup = FIXUP_RETURN_REGISTER;
+		/* if r2 = 0, no branch will be taken */
+		if ((insn[0] & 0x0f) == 0)
+			fixup |= FIXUP_BRANCH_NOT_TAKEN;
+		break;
+	case 0x06:	/* bctr	*/
+	case 0x07:	/* bcr	*/
+		fixup = FIXUP_BRANCH_NOT_TAKEN;
+		break;
+	case 0x45:	/* bal	*/
+	case 0x4d:	/* bas	*/
+		fixup = FIXUP_RETURN_REGISTER;
+		break;
+	case 0x47:	/* bc	*/
+	case 0x46:	/* bct	*/
+	case 0x86:	/* bxh	*/
+	case 0x87:	/* bxle	*/
+		fixup = FIXUP_BRANCH_NOT_TAKEN;
+		break;
+	case 0x82:	/* lpsw	*/
+		fixup = FIXUP_NOT_REQUIRED;
+		break;
+	case 0xb2:	/* lpswe */
+		if ((insn[0] & 0xff) == 0xb2)
+			fixup = FIXUP_NOT_REQUIRED;
+		break;
+	case 0xa7:	/* bras	*/
+		if ((insn[0] & 0x0f) == 0x05)
+			fixup |= FIXUP_RETURN_REGISTER;
+		break;
+	case 0xc0:
+		if ((insn[0] & 0x0f) == 0x05)	/* brasl */
+			fixup |= FIXUP_RETURN_REGISTER;
+		break;
+	case 0xeb:
+		switch (insn[2] & 0xff) {
+		case 0x44: /* bxhg  */
+		case 0x45: /* bxleg */
+			fixup = FIXUP_BRANCH_NOT_TAKEN;
+			break;
+		}
+		break;
+	case 0xe3:	/* bctg	*/
+		if ((insn[2] & 0xff) == 0x46)
+			fixup = FIXUP_BRANCH_NOT_TAKEN;
+		break;
+	case 0xec:
+		switch (insn[2] & 0xff) {
+		case 0xe5: /* clgrb */
+		case 0xe6: /* cgrb  */
+		case 0xf6: /* crb   */
+		case 0xf7: /* clrb  */
+		case 0xfc: /* cgib  */
+		case 0xfd: /* cglib */
+		case 0xfe: /* cib   */
+		case 0xff: /* clib  */
+			fixup = FIXUP_BRANCH_NOT_TAKEN;
+			break;
+		}
+		break;
+	}
+	return fixup;
+}
+
+int probe_is_insn_relative_long(u16 *insn)
+{
+	/* Check if we have a RIL-b or RIL-c format instruction which
+	 * we need to modify in order to avoid instruction emulation. */
+	switch (insn[0] >> 8) {
+	case 0xc0:
+		if ((insn[0] & 0x0f) == 0x00) /* larl */
+			return true;
+		break;
+	case 0xc4:
+		switch (insn[0] & 0x0f) {
+		case 0x02: /* llhrl  */
+		case 0x04: /* lghrl  */
+		case 0x05: /* lhrl   */
+		case 0x06: /* llghrl */
+		case 0x07: /* sthrl  */
+		case 0x08: /* lgrl   */
+		case 0x0b: /* stgrl  */
+		case 0x0c: /* lgfrl  */
+		case 0x0d: /* lrl    */
+		case 0x0e: /* llgfrl */
+		case 0x0f: /* strl   */
+			return true;
+		}
+		break;
+	case 0xc6:
+		switch (insn[0] & 0x0f) {
+		case 0x02: /* pfdrl  */
+		case 0x04: /* cghrl  */
+		case 0x05: /* chrl   */
+		case 0x06: /* clghrl */
+		case 0x07: /* clhrl  */
+		case 0x08: /* cgrl   */
+		case 0x0a: /* clgrl  */
+		case 0x0c: /* cgfrl  */
+		case 0x0d: /* crl    */
+		case 0x0e: /* clgfrl */
+		case 0x0f: /* clrl   */
+			return true;
+		}
+		break;
+	}
+	return false;
+}

arch/s390/lib/spinlock.c

@@ -98,17 +98,6 @@ void arch_spin_lock_wait_flags(arch_spinlock_t *lp, unsigned long flags)
 }
 EXPORT_SYMBOL(arch_spin_lock_wait_flags);
 
-void arch_spin_relax(arch_spinlock_t *lp)
-{
-	unsigned int cpu = lp->lock;
-	if (cpu != 0) {
-		if (MACHINE_IS_VM || MACHINE_IS_KVM ||
-		    !smp_vcpu_scheduled(~cpu))
-			smp_yield_cpu(~cpu);
-	}
-}
-EXPORT_SYMBOL(arch_spin_relax);
-
 int arch_spin_trylock_retry(arch_spinlock_t *lp)
 {
 	int count;
@@ -122,15 +111,21 @@ EXPORT_SYMBOL(arch_spin_trylock_retry);
 
 void _raw_read_lock_wait(arch_rwlock_t *rw)
 {
-	unsigned int old;
+	unsigned int owner, old;
 	int count = spin_retry;
 
+#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES
+	__RAW_LOCK(&rw->lock, -1, __RAW_OP_ADD);
+#endif
+	owner = 0;
 	while (1) {
 		if (count-- <= 0) {
-			smp_yield();
+			if (owner && !smp_vcpu_scheduled(~owner))
+				smp_yield_cpu(~owner);
 			count = spin_retry;
 		}
 		old = ACCESS_ONCE(rw->lock);
+		owner = ACCESS_ONCE(rw->owner);
 		if ((int) old < 0)
 			continue;
 		if (_raw_compare_and_swap(&rw->lock, old, old + 1))
@@ -139,28 +134,6 @@ void _raw_read_lock_wait(arch_rwlock_t *rw)
 }
 EXPORT_SYMBOL(_raw_read_lock_wait);
 
-void _raw_read_lock_wait_flags(arch_rwlock_t *rw, unsigned long flags)
-{
-	unsigned int old;
-	int count = spin_retry;
-
-	local_irq_restore(flags);
-	while (1) {
-		if (count-- <= 0) {
-			smp_yield();
-			count = spin_retry;
-		}
-		old = ACCESS_ONCE(rw->lock);
-		if ((int) old < 0)
-			continue;
-		local_irq_disable();
-		if (_raw_compare_and_swap(&rw->lock, old, old + 1))
-			return;
-		local_irq_restore(flags);
-	}
-}
-EXPORT_SYMBOL(_raw_read_lock_wait_flags);
-
 int _raw_read_trylock_retry(arch_rwlock_t *rw)
 {
 	unsigned int old;
@@ -177,46 +150,62 @@ int _raw_read_trylock_retry(arch_rwlock_t *rw)
 }
 EXPORT_SYMBOL(_raw_read_trylock_retry);
 
-void _raw_write_lock_wait(arch_rwlock_t *rw)
+#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES
+
+void _raw_write_lock_wait(arch_rwlock_t *rw, unsigned int prev)
 {
-	unsigned int old;
+	unsigned int owner, old;
 	int count = spin_retry;
 
+	owner = 0;
 	while (1) {
 		if (count-- <= 0) {
-			smp_yield();
+			if (owner && !smp_vcpu_scheduled(~owner))
+				smp_yield_cpu(~owner);
 			count = spin_retry;
 		}
 		old = ACCESS_ONCE(rw->lock);
-		if (old)
-			continue;
-		if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000))
-			return;
+		owner = ACCESS_ONCE(rw->owner);
+		smp_rmb();
+		if ((int) old >= 0) {
+			prev = __RAW_LOCK(&rw->lock, 0x80000000, __RAW_OP_OR);
+			old = prev;
+		}
+		if ((old & 0x7fffffff) == 0 && (int) prev >= 0)
+			break;
 	}
 }
 EXPORT_SYMBOL(_raw_write_lock_wait);
 
-void _raw_write_lock_wait_flags(arch_rwlock_t *rw, unsigned long flags)
+#else /* CONFIG_HAVE_MARCH_Z196_FEATURES */
+
+void _raw_write_lock_wait(arch_rwlock_t *rw)
 {
-	unsigned int old;
+	unsigned int owner, old, prev;
 	int count = spin_retry;
 
-	local_irq_restore(flags);
+	prev = 0x80000000;
+	owner = 0;
 	while (1) {
 		if (count-- <= 0) {
-			smp_yield();
+			if (owner && !smp_vcpu_scheduled(~owner))
+				smp_yield_cpu(~owner);
 			count = spin_retry;
 		}
 		old = ACCESS_ONCE(rw->lock);
-		if (old)
-			continue;
-		local_irq_disable();
-		if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000))
-			return;
-		local_irq_restore(flags);
+		owner = ACCESS_ONCE(rw->owner);
+		if ((int) old >= 0 &&
+		    _raw_compare_and_swap(&rw->lock, old, old | 0x80000000))
+			prev = old;
+		else
+			smp_rmb();
+		if ((old & 0x7fffffff) == 0 && (int) prev >= 0)
+			break;
 	}
 }
-EXPORT_SYMBOL(_raw_write_lock_wait_flags);
+EXPORT_SYMBOL(_raw_write_lock_wait);
+
+#endif /* CONFIG_HAVE_MARCH_Z196_FEATURES */
 
 int _raw_write_trylock_retry(arch_rwlock_t *rw)
 {
@@ -233,3 +222,13 @@ int _raw_write_trylock_retry(arch_rwlock_t *rw)
 	return 0;
 }
 EXPORT_SYMBOL(_raw_write_trylock_retry);
+
+void arch_lock_relax(unsigned int cpu)
+{
+	if (!cpu)
+		return;
+	if (MACHINE_IS_LPAR && smp_vcpu_scheduled(~cpu))
+		return;
+	smp_yield_cpu(~cpu);
+}
+EXPORT_SYMBOL(arch_lock_relax);

arch/s390/mm/dump_pagetables.c

@@ -54,7 +54,6 @@ static void print_prot(struct seq_file *m, unsigned int pr, int level)
 		return;
 	}
 	seq_printf(m, "%s", pr & _PAGE_PROTECT ? "RO " : "RW ");
-	seq_printf(m, "%s", pr & _PAGE_CO ? "CO " : "   ");
 	seq_putc(m, '\n');
 }
 
@@ -129,7 +128,7 @@ static void walk_pte_level(struct seq_file *m, struct pg_state *st,
 }
 
 #ifdef CONFIG_64BIT
-#define _PMD_PROT_MASK (_SEGMENT_ENTRY_PROTECT | _SEGMENT_ENTRY_CO)
+#define _PMD_PROT_MASK _SEGMENT_ENTRY_PROTECT
 #else
 #define _PMD_PROT_MASK 0
 #endif
@@ -157,7 +156,7 @@ static void walk_pmd_level(struct seq_file *m, struct pg_state *st,
 }
 
 #ifdef CONFIG_64BIT
-#define _PUD_PROT_MASK (_REGION3_ENTRY_RO | _REGION3_ENTRY_CO)
+#define _PUD_PROT_MASK _REGION3_ENTRY_RO
 #else
 #define _PUD_PROT_MASK 0
 #endif

arch/s390/mm/hugetlbpage.c

@@ -88,7 +88,7 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
 		pmd_val(pmd) &= ~_SEGMENT_ENTRY_ORIGIN;
 		pmd_val(pmd) |= pte_page(pte)[1].index;
 	} else
-		pmd_val(pmd) |= _SEGMENT_ENTRY_LARGE | _SEGMENT_ENTRY_CO;
+		pmd_val(pmd) |= _SEGMENT_ENTRY_LARGE;
 	*(pmd_t *) ptep = pmd;
 }
 

arch/s390/mm/pageattr.c

@@ -6,6 +6,7 @@
 #include <linux/module.h>
 #include <linux/mm.h>
 #include <asm/cacheflush.h>
+#include <asm/facility.h>
 #include <asm/pgtable.h>
 #include <asm/page.h>
 
@@ -103,27 +104,50 @@ int set_memory_x(unsigned long addr, int numpages)
 }
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
+
+static void ipte_range(pte_t *pte, unsigned long address, int nr)
+{
+	int i;
+
+	if (test_facility(13) && IS_ENABLED(CONFIG_64BIT)) {
+		__ptep_ipte_range(address, nr - 1, pte);
+		return;
+	}
+	for (i = 0; i < nr; i++) {
+		__ptep_ipte(address, pte);
+		address += PAGE_SIZE;
+		pte++;
+	}
+}
+
 void kernel_map_pages(struct page *page, int numpages, int enable)
 {
 	unsigned long address;
+	int nr, i, j;
 	pgd_t *pgd;
 	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
-	int i;
 
-	for (i = 0; i < numpages; i++) {
+	for (i = 0; i < numpages;) {
 		address = page_to_phys(page + i);
 		pgd = pgd_offset_k(address);
 		pud = pud_offset(pgd, address);
 		pmd = pmd_offset(pud, address);
 		pte = pte_offset_kernel(pmd, address);
-		if (!enable) {
-			__ptep_ipte(address, pte);
-			pte_val(*pte) = _PAGE_INVALID;
-			continue;
+		nr = (unsigned long)pte >> ilog2(sizeof(long));
+		nr = PTRS_PER_PTE - (nr & (PTRS_PER_PTE - 1));
+		nr = min(numpages - i, nr);
+		if (enable) {
+			for (j = 0; j < nr; j++) {
+				pte_val(*pte) = __pa(address);
+				address += PAGE_SIZE;
+				pte++;
+			}
+		} else {
+			ipte_range(pte, address, nr);
 		}
-		pte_val(*pte) = __pa(address);
+		i += nr;
 	}
 }
 

arch/s390/mm/vmem.c

@@ -236,8 +236,7 @@ int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node)
 				if (!new_page)
 					goto out;
 				pmd_val(*pm_dir) = __pa(new_page) |
-					_SEGMENT_ENTRY | _SEGMENT_ENTRY_LARGE |
-					_SEGMENT_ENTRY_CO;
+					_SEGMENT_ENTRY | _SEGMENT_ENTRY_LARGE;
 				address = (address + PMD_SIZE) & PMD_MASK;
 				continue;
 			}
@@ -253,9 +252,9 @@ int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node)
 
 		pt_dir = pte_offset_kernel(pm_dir, address);
 		if (pte_none(*pt_dir)) {
-			unsigned long new_page;
+			void *new_page;
 
-			new_page =__pa(vmem_alloc_pages(0));
+			new_page = vmemmap_alloc_block(PAGE_SIZE, node);
 			if (!new_page)
 				goto out;
 			pte_val(*pt_dir) =
@@ -263,7 +262,6 @@ int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node)
 		}
 		address += PAGE_SIZE;
 	}
-	memset((void *)start, 0, end - start);
 	ret = 0;
 out:
 	return ret;

drivers/s390/block/dasd_devmap.c

@@ -1432,6 +1432,29 @@ static ssize_t dasd_reservation_state_store(struct device *dev,
 static DEVICE_ATTR(last_known_reservation_state, 0644,
 		   dasd_reservation_state_show, dasd_reservation_state_store);
 
+static ssize_t dasd_pm_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
+{
+	struct dasd_device *device;
+	u8 opm, nppm, cablepm, cuirpm, hpfpm;
+
+	device = dasd_device_from_cdev(to_ccwdev(dev));
+	if (IS_ERR(device))
+		return sprintf(buf, "0\n");
+
+	opm = device->path_data.opm;
+	nppm = device->path_data.npm;
+	cablepm = device->path_data.cablepm;
+	cuirpm = device->path_data.cuirpm;
+	hpfpm = device->path_data.hpfpm;
+	dasd_put_device(device);
+
+	return sprintf(buf, "%02x %02x %02x %02x %02x\n", opm, nppm,
+		       cablepm, cuirpm, hpfpm);
+}
+
+static DEVICE_ATTR(path_masks, 0444, dasd_pm_show, NULL);
+
 static struct attribute * dasd_attrs[] = {
 	&dev_attr_readonly.attr,
 	&dev_attr_discipline.attr,
@@ -1450,6 +1473,7 @@ static struct attribute * dasd_attrs[] = {
 	&dev_attr_reservation_policy.attr,
 	&dev_attr_last_known_reservation_state.attr,
 	&dev_attr_safe_offline.attr,
+	&dev_attr_path_masks.attr,
 	NULL,
 };
 

drivers/s390/char/Makefile

@@ -33,3 +33,6 @@ obj-$(CONFIG_S390_VMUR) += vmur.o
 
 zcore_mod-objs := sclp_sdias.o zcore.o
 obj-$(CONFIG_CRASH_DUMP) += zcore_mod.o
+
+hmcdrv-objs := hmcdrv_mod.o hmcdrv_dev.o hmcdrv_ftp.o hmcdrv_cache.o diag_ftp.o sclp_ftp.o
+obj-$(CONFIG_HMC_DRV) += hmcdrv.o