Merge branch 'arch-microblaze' into no-rebases

50 files changed, 288 insertions, 214 deletions

diff --git a/arch/arm/include/asm/io.h b/arch/arm/include/asm/io.h
index 35c1ed89b93..42f042ee4ad 100644
--- a/arch/arm/include/asm/io.h
+++ b/arch/arm/include/asm/io.h
@@ -64,7 +64,7 @@ extern void __raw_readsl(const void __iomem *addr, void *data, int longlen);
 static inline void __raw_writew(u16 val, volatile void __iomem *addr)
 {
 	asm volatile("strh %1, %0"
-		     : "+Qo" (*(volatile u16 __force *)addr)
+		     : "+Q" (*(volatile u16 __force *)addr)
 		     : "r" (val));
 }
 
@@ -72,7 +72,7 @@ static inline u16 __raw_readw(const volatile void __iomem *addr)
 {
 	u16 val;
 	asm volatile("ldrh %1, %0"
-		     : "+Qo" (*(volatile u16 __force *)addr),
+		     : "+Q" (*(volatile u16 __force *)addr),
 		       "=r" (val));
 	return val;
 }
diff --git a/arch/arm/include/asm/sched_clock.h b/arch/arm/include/asm/sched_clock.h
index 05b8e82ec9f..e3f75726343 100644
--- a/arch/arm/include/asm/sched_clock.h
+++ b/arch/arm/include/asm/sched_clock.h
@@ -10,7 +10,5 @@
 
 extern void sched_clock_postinit(void);
 extern void setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate);
-extern void setup_sched_clock_needs_suspend(u32 (*read)(void), int bits,
-		unsigned long rate);
 
 #endif
diff --git a/arch/arm/include/asm/vfpmacros.h b/arch/arm/include/asm/vfpmacros.h
index 6a6f1e485f4..301c1db3e99 100644
--- a/arch/arm/include/asm/vfpmacros.h
+++ b/arch/arm/include/asm/vfpmacros.h
@@ -27,9 +27,9 @@
 #if __LINUX_ARM_ARCH__ <= 6
 	ldr	\tmp, =elf_hwcap		    @ may not have MVFR regs
 	ldr	\tmp, [\tmp, #0]
-	tst	\tmp, #HWCAP_VFPv3D16
-	ldceql	p11, cr0, [\base],#32*4		    @ FLDMIAD \base!, {d16-d31}
-	addne	\base, \base, #32*4		    @ step over unused register space
+	tst	\tmp, #HWCAP_VFPD32
+	ldcnel	p11, cr0, [\base],#32*4		    @ FLDMIAD \base!, {d16-d31}
+	addeq	\base, \base, #32*4		    @ step over unused register space
 #else
 	VFPFMRX	\tmp, MVFR0			    @ Media and VFP Feature Register 0
 	and	\tmp, \tmp, #MVFR0_A_SIMD_MASK	    @ A_SIMD field
@@ -51,9 +51,9 @@
 #if __LINUX_ARM_ARCH__ <= 6
 	ldr	\tmp, =elf_hwcap		    @ may not have MVFR regs
 	ldr	\tmp, [\tmp, #0]
-	tst	\tmp, #HWCAP_VFPv3D16
-	stceql	p11, cr0, [\base],#32*4		    @ FSTMIAD \base!, {d16-d31}
-	addne	\base, \base, #32*4		    @ step over unused register space
+	tst	\tmp, #HWCAP_VFPD32
+	stcnel	p11, cr0, [\base],#32*4		    @ FSTMIAD \base!, {d16-d31}
+	addeq	\base, \base, #32*4		    @ step over unused register space
 #else
 	VFPFMRX	\tmp, MVFR0			    @ Media and VFP Feature Register 0
 	and	\tmp, \tmp, #MVFR0_A_SIMD_MASK	    @ A_SIMD field
diff --git a/arch/arm/include/uapi/asm/hwcap.h b/arch/arm/include/uapi/asm/hwcap.h
index f254f6503cc..3688fd15a32 100644
--- a/arch/arm/include/uapi/asm/hwcap.h
+++ b/arch/arm/include/uapi/asm/hwcap.h
@@ -18,11 +18,12 @@
 #define HWCAP_THUMBEE	(1 << 11)
 #define HWCAP_NEON	(1 << 12)
 #define HWCAP_VFPv3	(1 << 13)
-#define HWCAP_VFPv3D16	(1 << 14)
+#define HWCAP_VFPv3D16	(1 << 14)	/* also set for VFPv4-D16 */
 #define HWCAP_TLS	(1 << 15)
 #define HWCAP_VFPv4	(1 << 16)
 #define HWCAP_IDIVA	(1 << 17)
 #define HWCAP_IDIVT	(1 << 18)
+#define HWCAP_VFPD32	(1 << 19)	/* set if VFP has 32 regs (not 16) */
 #define HWCAP_IDIV	(HWCAP_IDIVA | HWCAP_IDIVT)
 
 
diff --git a/arch/arm/kernel/sched_clock.c b/arch/arm/kernel/sched_clock.c
index e21bac20d90..fc6692e2b60 100644
--- a/arch/arm/kernel/sched_clock.c
+++ b/arch/arm/kernel/sched_clock.c
@@ -107,13 +107,6 @@ static void sched_clock_poll(unsigned long wrap_ticks)
 	update_sched_clock();
 }
 
-void __init setup_sched_clock_needs_suspend(u32 (*read)(void), int bits,
-		unsigned long rate)
-{
-	setup_sched_clock(read, bits, rate);
-	cd.needs_suspend = true;
-}
-
 void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate)
 {
 	unsigned long r, w;
@@ -189,18 +182,15 @@ void __init sched_clock_postinit(void)
 static int sched_clock_suspend(void)
 {
 	sched_clock_poll(sched_clock_timer.data);
-	if (cd.needs_suspend)
-		cd.suspended = true;
+	cd.suspended = true;
 	return 0;
 }
 
 static void sched_clock_resume(void)
 {
-	if (cd.needs_suspend) {
-		cd.epoch_cyc = read_sched_clock();
-		cd.epoch_cyc_copy = cd.epoch_cyc;
-		cd.suspended = false;
-	}
+	cd.epoch_cyc = read_sched_clock();
+	cd.epoch_cyc_copy = cd.epoch_cyc;
+	cd.suspended = false;
 }
 
 static struct syscore_ops sched_clock_ops = {
diff --git a/arch/arm/mm/alignment.c b/arch/arm/mm/alignment.c
index 023f443784e..b820edaf318 100644
--- a/arch/arm/mm/alignment.c
+++ b/arch/arm/mm/alignment.c
@@ -745,7 +745,7 @@ do_alignment_t32_to_handler(unsigned long *pinstr, struct pt_regs *regs,
 static int
 do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 {
-	union offset_union offset;
+	union offset_union uninitialized_var(offset);
 	unsigned long instr = 0, instrptr;
 	int (*handler)(unsigned long addr, unsigned long instr, struct pt_regs *regs);
 	unsigned int type;
diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c
index c834b32af27..3b44e0dd0a9 100644
--- a/arch/arm/vfp/vfpmodule.c
+++ b/arch/arm/vfp/vfpmodule.c
@@ -701,11 +701,14 @@ static int __init vfp_init(void)
 			elf_hwcap |= HWCAP_VFPv3;
 
 			/*
-			 * Check for VFPv3 D16. CPUs in this configuration
-			 * only have 16 x 64bit registers.
+			 * Check for VFPv3 D16 and VFPv4 D16.  CPUs in
+			 * this configuration only have 16 x 64bit
+			 * registers.
 			 */
 			if (((fmrx(MVFR0) & MVFR0_A_SIMD_MASK)) == 1)
-				elf_hwcap |= HWCAP_VFPv3D16;
+				elf_hwcap |= HWCAP_VFPv3D16; /* also v4-D16 */
+			else
+				elf_hwcap |= HWCAP_VFPD32;
 		}
 #endif
 		/*
diff --git a/arch/arm/xen/enlighten.c b/arch/arm/xen/enlighten.c
index 59bcb96ac36..f5760927544 100644
--- a/arch/arm/xen/enlighten.c
+++ b/arch/arm/xen/enlighten.c
@@ -166,3 +166,14 @@ void free_xenballooned_pages(int nr_pages, struct page **pages)
 	*pages = NULL;
 }
 EXPORT_SYMBOL_GPL(free_xenballooned_pages);
+
+/* In the hypervisor.S file. */
+EXPORT_SYMBOL_GPL(HYPERVISOR_event_channel_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_grant_table_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_xen_version);
+EXPORT_SYMBOL_GPL(HYPERVISOR_console_io);
+EXPORT_SYMBOL_GPL(HYPERVISOR_sched_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_hvm_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_memory_op);
+EXPORT_SYMBOL_GPL(HYPERVISOR_physdev_op);
+EXPORT_SYMBOL_GPL(privcmd_call);
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 138fc9cfd78..20b688c8195 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -1,6 +1,7 @@
 config ARM64
 	def_bool y
 	select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
+	select ARCH_WANT_COMPAT_IPC_PARSE_VERSION
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_HARDIRQS_NO_DEPRECATED
 	select GENERIC_IOMAP
diff --git a/arch/arm64/include/asm/processor.h b/arch/arm64/include/asm/processor.h
index 42471d07d97..ab239b2c456 100644
--- a/arch/arm64/include/asm/processor.h
+++ b/arch/arm64/include/asm/processor.h
@@ -43,6 +43,8 @@
 #else
 #define STACK_TOP		STACK_TOP_MAX
 #endif /* CONFIG_COMPAT */
+
+#define ARCH_LOW_ADDRESS_LIMIT	PHYS_MASK
 #endif /* __KERNEL__ */
 
 struct debug_info {
diff --git a/arch/arm64/include/asm/unistd.h b/arch/arm64/include/asm/unistd.h
index b40dc6b6984..43064a8bd99 100644
--- a/arch/arm64/include/asm/unistd.h
+++ b/arch/arm64/include/asm/unistd.h
@@ -14,7 +14,6 @@
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 #ifdef CONFIG_COMPAT
-#define __ARCH_WANT_COMPAT_IPC_PARSE_VERSION
 #define __ARCH_WANT_COMPAT_STAT64
 #define __ARCH_WANT_SYS_GETHOSTNAME
 #define __ARCH_WANT_SYS_PAUSE
diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c
index 226b6bf6e9c..538300f2273 100644
--- a/arch/arm64/kernel/smp.c
+++ b/arch/arm64/kernel/smp.c
@@ -211,8 +211,7 @@ asmlinkage void __cpuinit secondary_start_kernel(void)
 	 * before we continue.
 	 */
 	set_cpu_online(cpu, true);
-	while (!cpu_active(cpu))
-		cpu_relax();
+	complete(&cpu_running);
 
 	/*
 	 * OK, it's off to the idle thread for us
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index efbf7df05d3..4cd28931dba 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -80,7 +80,7 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max)
 #ifdef CONFIG_ZONE_DMA32
 	/* 4GB maximum for 32-bit only capable devices */
 	max_dma32 = min(max, MAX_DMA32_PFN);
-	zone_size[ZONE_DMA32] = max_dma32 - min;
+	zone_size[ZONE_DMA32] = max(min, max_dma32) - min;
 #endif
 	zone_size[ZONE_NORMAL] = max - max_dma32;
 
diff --git a/arch/h8300/include/asm/cache.h b/arch/h8300/include/asm/cache.h
index c6350283649..05887a1d80e 100644
--- a/arch/h8300/include/asm/cache.h
+++ b/arch/h8300/include/asm/cache.h
@@ -2,7 +2,8 @@
 #define __ARCH_H8300_CACHE_H
 
 /* bytes per L1 cache line */
-#define        L1_CACHE_BYTES  4
+#define        L1_CACHE_SHIFT  2
+#define        L1_CACHE_BYTES  (1 << L1_CACHE_SHIFT)
 
 /* m68k-elf-gcc  2.95.2 doesn't like these */
 
diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig
index 4cba7439f9d..198abf6d41c 100644
--- a/arch/microblaze/Kconfig
+++ b/arch/microblaze/Kconfig
@@ -26,6 +26,8 @@ config MICROBLAZE
 	select GENERIC_ATOMIC64
 	select GENERIC_CLOCKEVENTS
 	select MODULES_USE_ELF_RELA
+	select GENERIC_KERNEL_THREAD
+	select GENERIC_KERNEL_EXECVE
 
 config SWAP
 	def_bool n
diff --git a/arch/microblaze/include/asm/processor.h b/arch/microblaze/include/asm/processor.h
index af2bb965239..0759153e811 100644
--- a/arch/microblaze/include/asm/processor.h
+++ b/arch/microblaze/include/asm/processor.h
@@ -31,6 +31,7 @@ extern const struct seq_operations cpuinfo_op;
 void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long usp);
 
 extern void ret_from_fork(void);
+extern void ret_from_kernel_thread(void);
 
 # endif /* __ASSEMBLY__ */
 
@@ -78,11 +79,6 @@ extern unsigned long thread_saved_pc(struct task_struct *t);
 
 extern unsigned long get_wchan(struct task_struct *p);
 
-/*
- * create a kernel thread without removing it from tasklists
- */
-extern int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
-
 # define KSTK_EIP(tsk)	(0)
 # define KSTK_ESP(tsk)	(0)
 
@@ -131,8 +127,6 @@ extern inline void release_thread(struct task_struct *dead_task)
 {
 }
 
-extern int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
-
 /* Free current thread data structures etc.  */
 static inline void exit_thread(void)
 {
diff --git a/arch/microblaze/include/asm/unistd.h b/arch/microblaze/include/asm/unistd.h
index 6985e6e9d82..ea744283429 100644
--- a/arch/microblaze/include/asm/unistd.h
+++ b/arch/microblaze/include/asm/unistd.h
@@ -422,6 +422,7 @@
 #define __ARCH_WANT_SYS_SIGPROCMASK
 #define __ARCH_WANT_SYS_RT_SIGACTION
 #define __ARCH_WANT_SYS_RT_SIGSUSPEND
+#define __ARCH_WANT_SYS_EXECVE
 
 /*
  * "Conditional" syscalls
diff --git a/arch/microblaze/kernel/entry-nommu.S b/arch/microblaze/kernel/entry-nommu.S
index 75c3ea1f48a..673a49c04a0 100644
--- a/arch/microblaze/kernel/entry-nommu.S
+++ b/arch/microblaze/kernel/entry-nommu.S
@@ -474,6 +474,14 @@ ENTRY(ret_from_fork)
 	brid	ret_to_user
 	nop
 
+ENTRY(ret_from_kernel_thread)
+	brlid	r15, schedule_tail
+	addk	r5, r0, r3
+	brald	r15, r20
+	addk	r5, r0, r19
+	brid	ret_to_user
+	addk	r3, r0, r0
+
 work_pending:
 	enable_irq
 
@@ -559,10 +567,6 @@ sys_clone:
 	brid	microblaze_clone
 	addk	r7, r1, r0
 
-sys_execve:
-	brid	microblaze_execve
-	addk	r8, r1, r0
-
 sys_rt_sigreturn_wrapper:
 	brid	sys_rt_sigreturn
 	addk	r5, r1, r0
diff --git a/arch/microblaze/kernel/entry.S b/arch/microblaze/kernel/entry.S
index 03f7b8ce6b6..10f360ed82b 100644
--- a/arch/microblaze/kernel/entry.S
+++ b/arch/microblaze/kernel/entry.S
@@ -293,24 +293,6 @@ C_ENTRY(_user_exception):
 	swi	r1, r0, TOPHYS(PER_CPU(ENTRY_SP)) /* save stack */
 	addi	r14, r14, 4	/* return address is 4 byte after call */
 
-	mfs	r1, rmsr
-	nop
-	andi	r1, r1, MSR_UMS
-	bnei	r1, 1f
-
-/* Kernel-mode state save - kernel execve */
-	lwi	r1, r0, TOPHYS(PER_CPU(ENTRY_SP)); /* Reload kernel stack-ptr*/
-	tophys(r1,r1);
-
-	addik	r1, r1, -PT_SIZE; /* Make room on the stack. */
-	SAVE_REGS
-
-	swi	r1, r1, PT_MODE; /* pt_regs -> kernel mode */
-	brid	2f;
-	nop;				/* Fill delay slot */
-
-/* User-mode state save.  */
-1:
 	lwi	r1, r0, TOPHYS(PER_CPU(CURRENT_SAVE)); /* get saved current */
 	tophys(r1,r1);
 	lwi	r1, r1, TS_THREAD_INFO;	/* get stack from task_struct */
@@ -479,11 +461,20 @@ C_ENTRY(sys_fork_wrapper):
    saved context).  */
 C_ENTRY(ret_from_fork):
 	bralid	r15, schedule_tail; /* ...which is schedule_tail's arg */
-	add	r3, r5, r0;	/* switch_thread returns the prev task */
+	add	r5, r3, r0;	/* switch_thread returns the prev task */
 				/* ( in the delay slot ) */
 	brid	ret_from_trap;	/* Do normal trap return */
 	add	r3, r0, r0;	/* Child's fork call should return 0. */
 
+C_ENTRY(ret_from_kernel_thread):
+	bralid	r15, schedule_tail; /* ...which is schedule_tail's arg */
+	add	r5, r3, r0;	/* switch_thread returns the prev task */
+				/* ( in the delay slot ) */
+	brald	r15, r20	/* fn was left in r20 */
+	addk	r5, r0, r19	/* ... and argument - in r19 */
+	brid	ret_from_trap
+	add	r3, r0, r0
+
 C_ENTRY(sys_vfork):
 	brid	microblaze_vfork	/* Do real work (tail-call) */
 	addik	r5, r1, 0
@@ -498,10 +489,6 @@ C_ENTRY(sys_clone):
 	brid	do_fork		/* Do real work (tail-call) */
 	add     r8, r0, r0;             /* Arg 3: (unused) */
 
-C_ENTRY(sys_execve):
-	brid	microblaze_execve;	/* Do real work (tail-call).*/
-	addik	r8, r1, 0;		/* add user context as 4th arg */
-
 C_ENTRY(sys_rt_sigreturn_wrapper):
 	brid	sys_rt_sigreturn	/* Do real work */
 	addik	r5, r1, 0;		/* add user context as 1st arg */
diff --git a/arch/microblaze/kernel/process.c b/arch/microblaze/kernel/process.c
index 1944e00f07e..29768c3dc35 100644
--- a/arch/microblaze/kernel/process.c
+++ b/arch/microblaze/kernel/process.c
@@ -119,46 +119,38 @@ void flush_thread(void)
 }
 
 int copy_thread(unsigned long clone_flags, unsigned long usp,
-		unsigned long unused,
+		unsigned long arg,
 		struct task_struct *p, struct pt_regs *regs)
 {
 	struct pt_regs *childregs = task_pt_regs(p);
 	struct thread_info *ti = task_thread_info(p);
 
+	if (unlikely(p->flags & PF_KTHREAD)) {
+		/* if we're creating a new kernel thread then just zeroing all
+		 * the registers. That's OK for a brand new thread.*/
+		memset(childregs, 0, sizeof(struct pt_regs));
+		memset(&ti->cpu_context, 0, sizeof(struct cpu_context));
+		ti->cpu_context.r1  = (unsigned long)childregs;
+		ti->cpu_context.r20 = (unsigned long)usp; /* fn */
+		ti->cpu_context.r19 = (unsigned long)arg;
+		childregs->pt_mode = 1;
+		local_save_flags(childregs->msr);
+#ifdef CONFIG_MMU
+		ti->cpu_context.msr = childregs->msr & ~MSR_IE;
+#endif
+		ti->cpu_context.r15 = (unsigned long)ret_from_kernel_thread - 8;
+		return 0;
+	}
 	*childregs = *regs;
-	if (user_mode(regs))
-		childregs->r1 = usp;
-	else
-		childregs->r1 = ((unsigned long) ti) + THREAD_SIZE;
+	childregs->r1 = usp;
 
-#ifndef CONFIG_MMU
 	memset(&ti->cpu_context, 0, sizeof(struct cpu_context));
 	ti->cpu_context.r1 = (unsigned long)childregs;
+#ifndef CONFIG_MMU
 	ti->cpu_context.msr = (unsigned long)childregs->msr;
 #else
+	childregs->msr |= MSR_UMS;
 
-	/* if creating a kernel thread then update the current reg (we don't
-	 * want to use the parent's value when restoring by POP_STATE) */
-	if (kernel_mode(regs))
-		/* save new current on stack to use POP_STATE */
-		childregs->CURRENT_TASK = (unsigned long)p;
-	/* if returning to user then use the parent's value of this register */
-
-	/* if we're creating a new kernel thread then just zeroing all
-	 * the registers. That's OK for a brand new thread.*/
-	/* Pls. note that some of them will be restored in POP_STATE */
-	if (kernel_mode(regs))
-		memset(&ti->cpu_context, 0, sizeof(struct cpu_context));
-	/* if this thread is created for fork/vfork/clone, then we want to
-	 * restore all the parent's context */
-	/* in addition to the registers which will be restored by POP_STATE */
-	else {
-		ti->cpu_context = *(struct cpu_context *)regs;
-		childregs->msr |= MSR_UMS;
-	}
-
-	/* FIXME STATE_SAVE_PT_OFFSET; */
-	ti->cpu_context.r1  = (unsigned long)childregs;
 	/* we should consider the fact that childregs is a copy of the parent
 	 * regs which were saved immediately after entering the kernel state
 	 * before enabling VM. This MSR will be restored in switch_to and
@@ -209,29 +201,6 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
 }
 #endif
 
-static void kernel_thread_helper(int (*fn)(void *), void *arg)
-{
-	fn(arg);
-	do_exit(-1);
-}
-
-int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
-{
-	struct pt_regs regs;
-
-	memset(&regs, 0, sizeof(regs));
-	/* store them in non-volatile registers */
-	regs.r5 = (unsigned long)fn;
-	regs.r6 = (unsigned long)arg;
-	local_save_flags(regs.msr);
-	regs.pc = (unsigned long)kernel_thread_helper;
-	regs.pt_mode = 1;
-
-	return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0,
-			&regs, 0, NULL, NULL);
-}
-EXPORT_SYMBOL_GPL(kernel_thread);
-
 unsigned long get_wchan(struct task_struct *p)
 {
 /* TBD (used by procfs) */
@@ -246,6 +215,7 @@ void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long usp)
 	regs->pt_mode = 0;
 #ifdef CONFIG_MMU
 	regs->msr |= MSR_UMS;
+	regs->msr &= ~MSR_VM;
 #endif
 }
 
diff --git a/arch/microblaze/kernel/sys_microblaze.c b/arch/microblaze/kernel/sys_microblaze.c
index 404c0f24bd4..a6a7bae9f5c 100644
--- a/arch/microblaze/kernel/sys_microblaze.c
+++ b/arch/microblaze/kernel/sys_microblaze.c
@@ -48,24 +48,6 @@ asmlinkage long microblaze_clone(int flags, unsigned long stack,
 	return do_fork(flags, stack, regs, 0, NULL, NULL);
 }
 
-asmlinkage long microblaze_execve(const char __user *filenamei,
-				  const char __user *const __user *argv,
-				  const char __user *const __user *envp,
-				  struct pt_regs *regs)
-{
-	int error;
-	struct filename *filename;
-
-	filename = getname(filenamei);
-	error = PTR_ERR(filename);
-	if (IS_ERR(filename))
-		goto out;
-	error = do_execve(filename->name, argv, envp, regs);
-	putname(filename);
-out:
-	return error;
-}
-
 asmlinkage long sys_mmap(unsigned long addr, unsigned long len,
 			unsigned long prot, unsigned long flags,
 			unsigned long fd, off_t pgoff)
@@ -75,24 +57,3 @@ asmlinkage long sys_mmap(unsigned long addr, unsigned long len,
 
 	return sys_mmap_pgoff(addr, len, prot, flags, fd, pgoff >> PAGE_SHIFT);
 }
-
-/*
- * Do a system call from kernel instead of calling sys_execve so we
- * end up with proper pt_regs.
- */
-int kernel_execve(const char *filename,
-		  const char *const argv[],
-		  const char *const envp[])
-{
-	register const char *__a __asm__("r5") = filename;
-	register const void *__b __asm__("r6") = argv;
-	register const void *__c __asm__("r7") = envp;
-	register unsigned long __syscall __asm__("r12") = __NR_execve;
-	register unsigned long __ret __asm__("r3");
-	__asm__ __volatile__ ("brki r14, 0x8"
-			: "=r" (__ret), "=r" (__syscall)
-			: "1" (__syscall), "r" (__a), "r" (__b), "r" (__c)
-			: "r4", "r8", "r9",
-			"r10", "r11", "r14", "cc", "memory");
-	return __ret;
-}
diff --git a/arch/s390/include/asm/cio.h b/arch/s390/include/asm/cio.h
index 55bde603521..ad2b924167d 100644
--- a/arch/s390/include/asm/cio.h
+++ b/arch/s390/include/asm/cio.h
@@ -9,6 +9,8 @@
 
 #define LPM_ANYPATH 0xff
 #define __MAX_CSSID 0
+#define __MAX_SUBCHANNEL 65535
+#define __MAX_SSID 3
 
 #include <asm/scsw.h>
 
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index dd647c919a6..2d3b7cb2600 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -506,12 +506,15 @@ static inline int pud_bad(pud_t pud)
 
 static inline int pmd_present(pmd_t pmd)
 {
-	return (pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN) != 0UL;
+	unsigned long mask = _SEGMENT_ENTRY_INV | _SEGMENT_ENTRY_RO;
+	return (pmd_val(pmd) & mask) == _HPAGE_TYPE_NONE ||
+	       !(pmd_val(pmd) & _SEGMENT_ENTRY_INV);
 }
 
 static inline int pmd_none(pmd_t pmd)
 {
-	return (pmd_val(pmd) & _SEGMENT_ENTRY_INV) != 0UL;
+	return (pmd_val(pmd) & _SEGMENT_ENTRY_INV) &&
+	       !(pmd_val(pmd) & _SEGMENT_ENTRY_RO);
 }
 
 static inline int pmd_large(pmd_t pmd)
@@ -1223,6 +1226,11 @@ static inline void __pmd_idte(unsigned long address, pmd_t *pmdp)
 }
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
+
+#define SEGMENT_NONE	__pgprot(_HPAGE_TYPE_NONE)
+#define SEGMENT_RO	__pgprot(_HPAGE_TYPE_RO)
+#define SEGMENT_RW	__pgprot(_HPAGE_TYPE_RW)
+
 #define __HAVE_ARCH_PGTABLE_DEPOSIT
 extern void pgtable_trans_huge_deposit(struct mm_struct *mm, pgtable_t pgtable);
 
@@ -1242,16 +1250,15 @@ static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr,
 
 static inline unsigned long massage_pgprot_pmd(pgprot_t pgprot)
 {
-	unsigned long pgprot_pmd = 0;
-
-	if (pgprot_val(pgprot) & _PAGE_INVALID) {
-		if (pgprot_val(pgprot) & _PAGE_SWT)
-			pgprot_pmd |= _HPAGE_TYPE_NONE;
-		pgprot_pmd |= _SEGMENT_ENTRY_INV;
-	}
-	if (pgprot_val(pgprot) & _PAGE_RO)
-		pgprot_pmd |= _SEGMENT_ENTRY_RO;
-	return pgprot_pmd;
+	/*
+	 * pgprot is PAGE_NONE, PAGE_RO, or PAGE_RW (see __Pxxx / __Sxxx)
+	 * Convert to segment table entry format.
+	 */
+	if (pgprot_val(pgprot) == pgprot_val(PAGE_NONE))
+		return pgprot_val(SEGMENT_NONE);
+	if (pgprot_val(pgprot) == pgprot_val(PAGE_RO))
+		return pgprot_val(SEGMENT_RO);
+	return pgprot_val(SEGMENT_RW);
 }
 
 static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot)
@@ -1269,7 +1276,9 @@ static inline pmd_t pmd_mkhuge(pmd_t pmd)
 
 static inline pmd_t pmd_mkwrite(pmd_t pmd)
 {
-	pmd_val(pmd) &= ~_SEGMENT_ENTRY_RO;
+	/* Do not clobber _HPAGE_TYPE_NONE pages! */
+	if (!(pmd_val(pmd) & _SEGMENT_ENTRY_INV))
+		pmd_val(pmd) &= ~_SEGMENT_ENTRY_RO;
 	return pmd;
 }
 
diff --git a/arch/s390/kernel/sclp.S b/arch/s390/kernel/sclp.S
index bf053898630..b6506ee32a3 100644
--- a/arch/s390/kernel/sclp.S
+++ b/arch/s390/kernel/sclp.S
@@ -44,6 +44,12 @@ _sclp_wait_int:
 #endif
 	mvc	.LoldpswS1-.LbaseS1(16,%r13),0(%r8)
 	mvc	0(16,%r8),0(%r9)
+#ifdef CONFIG_64BIT
+	epsw	%r6,%r7				# set current addressing mode
+	nill	%r6,0x1				# in new psw (31 or 64 bit mode)
+	nilh	%r7,0x8000
+	stm	%r6,%r7,0(%r8)
+#endif
 	lhi	%r6,0x0200			# cr mask for ext int (cr0.54)
 	ltr	%r2,%r2
 	jz	.LsetctS1
@@ -87,7 +93,7 @@ _sclp_wait_int:
 	.long	0x00080000, 0x80000000+.LwaitS1	# PSW to handle ext int
 #ifdef CONFIG_64BIT
 .LextpswS1_64:
-	.quad	0x0000000180000000, .LwaitS1	# PSW to handle ext int, 64 bit
+	.quad	0, .LwaitS1			# PSW to handle ext int, 64 bit
 #endif
 .LwaitpswS1:
 	.long	0x010a0000, 0x00000000+.LloopS1	# PSW to wait for ext int
diff --git a/arch/s390/lib/uaccess_pt.c b/arch/s390/lib/uaccess_pt.c
index 2d37bb861fa..9017a63dda3 100644
--- a/arch/s390/lib/uaccess_pt.c
+++ b/arch/s390/lib/uaccess_pt.c
@@ -39,7 +39,7 @@ static __always_inline unsigned long follow_table(struct mm_struct *mm,
 	pmd = pmd_offset(pud, addr);
 	if (pmd_none(*pmd))
 		return -0x10UL;
-	if (pmd_huge(*pmd)) {
+	if (pmd_large(*pmd)) {
 		if (write && (pmd_val(*pmd) & _SEGMENT_ENTRY_RO))
 			return -0x04UL;
 		return (pmd_val(*pmd) & HPAGE_MASK) + (addr & ~HPAGE_MASK);
diff --git a/arch/s390/mm/gup.c b/arch/s390/mm/gup.c
index 60acb93a468..8b8285310b5 100644
--- a/arch/s390/mm/gup.c
+++ b/arch/s390/mm/gup.c
@@ -126,7 +126,7 @@ static inline int gup_pmd_range(pud_t *pudp, pud_t pud, unsigned long addr,
 		 */
 		if (pmd_none(pmd) || pmd_trans_splitting(pmd))
 			return 0;
-		if (unlikely(pmd_huge(pmd))) {
+		if (unlikely(pmd_large(pmd))) {
 			if (!gup_huge_pmd(pmdp, pmd, addr, next,
 					  write, pages, nr))
 				return 0;
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index e52f3c2ad3d..0c7d365fa40 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -20,6 +20,7 @@ config SPARC
 	select HAVE_ARCH_TRACEHOOK
 	select SYSCTL_EXCEPTION_TRACE
 	select ARCH_WANT_OPTIONAL_GPIOLIB
+	select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
 	select RTC_CLASS
 	select RTC_DRV_M48T59
 	select HAVE_IRQ_WORK
diff --git a/arch/sparc/crypto/Makefile b/arch/sparc/crypto/Makefile
index 6ae1ad5e502..5d469d81761 100644
--- a/arch/sparc/crypto/Makefile
+++ b/arch/sparc/crypto/Makefile
@@ -13,13 +13,13 @@ obj-$(CONFIG_CRYPTO_DES_SPARC64) += camellia-sparc64.o
 
 obj-$(CONFIG_CRYPTO_CRC32C_SPARC64) += crc32c-sparc64.o
 
-sha1-sparc64-y := sha1_asm.o sha1_glue.o crop_devid.o
-sha256-sparc64-y := sha256_asm.o sha256_glue.o crop_devid.o
-sha512-sparc64-y := sha512_asm.o sha512_glue.o crop_devid.o
-md5-sparc64-y := md5_asm.o md5_glue.o crop_devid.o
+sha1-sparc64-y := sha1_asm.o sha1_glue.o
+sha256-sparc64-y := sha256_asm.o sha256_glue.o
+sha512-sparc64-y := sha512_asm.o sha512_glue.o
+md5-sparc64-y := md5_asm.o md5_glue.o
 
-aes-sparc64-y := aes_asm.o aes_glue.o crop_devid.o
-des-sparc64-y := des_asm.o des_glue.o crop_devid.o
-camellia-sparc64-y := camellia_asm.o camellia_glue.o crop_devid.o
+aes-sparc64-y := aes_asm.o aes_glue.o
+des-sparc64-y := des_asm.o des_glue.o
+camellia-sparc64-y := camellia_asm.o camellia_glue.o
 
-crc32c-sparc64-y := crc32c_asm.o crc32c_glue.o crop_devid.o
+crc32c-sparc64-y := crc32c_asm.o crc32c_glue.o
diff --git a/arch/sparc/crypto/aes_glue.c b/arch/sparc/crypto/aes_glue.c
index 8f1c9980f63..3965d1d36df 100644
--- a/arch/sparc/crypto/aes_glue.c
+++ b/arch/sparc/crypto/aes_glue.c
@@ -475,3 +475,5 @@ MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("AES Secure Hash Algorithm, sparc64 aes opcode accelerated");
 
 MODULE_ALIAS("aes");
+
+#include "crop_devid.c"
diff --git a/arch/sparc/crypto/camellia_glue.c b/arch/sparc/crypto/camellia_glue.c
index 42905c08429..62c89af3fd3 100644
--- a/arch/sparc/crypto/camellia_glue.c
+++ b/arch/sparc/crypto/camellia_glue.c
@@ -320,3 +320,5 @@ MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Camellia Cipher Algorithm, sparc64 camellia opcode accelerated");
 
 MODULE_ALIAS("aes");
+
+#include "crop_devid.c"
diff --git a/arch/sparc/crypto/crc32c_glue.c b/arch/sparc/crypto/crc32c_glue.c
index 0bd89cea8d8..5162fad912c 100644
--- a/arch/sparc/crypto/crc32c_glue.c
+++ b/arch/sparc/crypto/crc32c_glue.c
@@ -177,3 +177,5 @@ MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("CRC32c (Castagnoli), sparc64 crc32c opcode accelerated");
 
 MODULE_ALIAS("crc32c");
+
+#include "crop_devid.c"
diff --git a/arch/sparc/crypto/des_glue.c b/arch/sparc/crypto/des_glue.c
index c4940c2d307..41524cebcc4 100644
--- a/arch/sparc/crypto/des_glue.c
+++ b/arch/sparc/crypto/des_glue.c
@@ -527,3 +527,5 @@ MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("DES & Triple DES EDE Cipher Algorithms, sparc64 des opcode accelerated");
 
 MODULE_ALIAS("des");
+
+#include "crop_devid.c"
diff --git a/arch/sparc/crypto/md5_glue.c b/arch/sparc/crypto/md5_glue.c
index 603d723038c..09a9ea1dfb6 100644
--- a/arch/sparc/crypto/md5_glue.c
+++ b/arch/sparc/crypto/md5_glue.c
@@ -186,3 +186,5 @@ MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("MD5 Secure Hash Algorithm, sparc64 md5 opcode accelerated");
 
 MODULE_ALIAS("md5");
+
+#include "crop_devid.c"
diff --git a/arch/sparc/crypto/sha1_glue.c b/arch/sparc/crypto/sha1_glue.c
index 2bbb20bee9f..6cd5f29e1e0 100644
--- a/arch/sparc/crypto/sha1_glue.c
+++ b/arch/sparc/crypto/sha1_glue.c
@@ -181,3 +181,5 @@ MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("SHA1 Secure Hash Algorithm, sparc64 sha1 opcode accelerated");
 
 MODULE_ALIAS("sha1");
+
+#include "crop_devid.c"
diff --git a/arch/sparc/crypto/sha256_glue.c b/arch/sparc/crypto/sha256_glue.c
index 591e656bd89..04f555ab268 100644
--- a/arch/sparc/crypto/sha256_glue.c
+++ b/arch/sparc/crypto/sha256_glue.c
@@ -239,3 +239,5 @@ MODULE_DESCRIPTION("SHA-224 and SHA-256 Secure Hash Algorithm, sparc64 sha256 op
 
 MODULE_ALIAS("sha224");
 MODULE_ALIAS("sha256");
+
+#include "crop_devid.c"
diff --git a/arch/sparc/crypto/sha512_glue.c b/arch/sparc/crypto/sha512_glue.c
index 486f0a2b700..f04d1994d19 100644
--- a/arch/sparc/crypto/sha512_glue.c
+++ b/arch/sparc/crypto/sha512_glue.c
@@ -224,3 +224,5 @@ MODULE_DESCRIPTION("SHA-384 and SHA-512 Secure Hash Algorithm, sparc64 sha512 op
 
 MODULE_ALIAS("sha384");
 MODULE_ALIAS("sha512");
+
+#include "crop_devid.c"
diff --git a/arch/sparc/include/asm/atomic_64.h b/arch/sparc/include/asm/atomic_64.h
index ce35a1cf1a2..be56a244c9c 100644
--- a/arch/sparc/include/asm/atomic_64.h
+++ b/arch/sparc/include/asm/atomic_64.h
@@ -1,7 +1,7 @@
 /* atomic.h: Thankfully the V9 is at least reasonable for this
  *           stuff.
  *
- * Copyright (C) 1996, 1997, 2000 David S. Miller (davem@redhat.com)
+ * Copyright (C) 1996, 1997, 2000, 2012 David S. Miller (davem@redhat.com)
  */
 
 #ifndef __ARCH_SPARC64_ATOMIC__
@@ -106,6 +106,8 @@ static inline long atomic64_add_unless(atomic64_t *v, long a, long u)
 
 #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
 
+extern long atomic64_dec_if_positive(atomic64_t *v);
+
 /* Atomic operations are already serializing */
 #define smp_mb__before_atomic_dec()	barrier()
 #define smp_mb__after_atomic_dec()	barrier()
diff --git a/arch/sparc/include/asm/backoff.h b/arch/sparc/include/asm/backoff.h
index db3af0d30fb..4e02086b839 100644
--- a/arch/sparc/include/asm/backoff.h
+++ b/arch/sparc/include/asm/backoff.h
@@ -1,6 +1,46 @@
 #ifndef _SPARC64_BACKOFF_H
 #define _SPARC64_BACKOFF_H
 
+/* The macros in this file implement an exponential backoff facility
+ * for atomic operations.
+ *
+ * When multiple threads compete on an atomic operation, it is
+ * possible for one thread to be continually denied a successful
+ * completion of the compare-and-swap instruction.  Heavily
+ * threaded cpu implementations like Niagara can compound this
+ * problem even further.
+ *
+ * When an atomic operation fails and needs to be retried, we spin a
+ * certain number of times.  At each subsequent failure of the same
+ * operation we double the spin count, realizing an exponential
+ * backoff.
+ *
+ * When we spin, we try to use an operation that will cause the
+ * current cpu strand to block, and therefore make the core fully
+ * available to any other other runnable strands.  There are two
+ * options, based upon cpu capabilities.
+ *
+ * On all cpus prior to SPARC-T4 we do three dummy reads of the
+ * condition code register.  Each read blocks the strand for something
+ * between 40 and 50 cpu cycles.
+ *
+ * For SPARC-T4 and later we have a special "pause" instruction
+ * available.  This is implemented using writes to register %asr27.
+ * The cpu will block the number of cycles written into the register,
+ * unless a disrupting trap happens first.  SPARC-T4 specifically
+ * implements pause with a granularity of 8 cycles.  Each strand has
+ * an internal pause counter which decrements every 8 cycles.  So the
+ * chip shifts the %asr27 value down by 3 bits, and writes the result
+ * into the pause counter.  If a value smaller than 8 is written, the
+ * chip blocks for 1 cycle.
+ *
+ * To achieve the same amount of backoff as the three %ccr reads give
+ * on earlier chips, we shift the backoff value up by 7 bits.  (Three
+ * %ccr reads block for about 128 cycles, 1 << 7 == 128) We write the
+ * whole amount we want to block into the pause register, rather than
+ * loop writing 128 each time.
+ */
+
 #define BACKOFF_LIMIT	(4 * 1024)
 
 #ifdef CONFIG_SMP
@@ -11,16 +51,25 @@
 #define BACKOFF_LABEL(spin_label, continue_label) \
 	spin_label
 
-#define BACKOFF_SPIN(reg, tmp, label)	\
-	mov	reg, tmp; \
-88:	brnz,pt	tmp, 88b; \
-	 sub	tmp, 1, tmp; \
-	set	BACKOFF_LIMIT, tmp; \
-	cmp	reg, tmp; \
-	bg,pn	%xcc, label; \
-	 nop; \
-	ba,pt	%xcc, label; \
-	 sllx	reg, 1, reg;
+#define BACKOFF_SPIN(reg, tmp, label)		\
+	mov		reg, tmp;		\
+88:	rd		%ccr, %g0;		\
+	rd		%ccr, %g0;		\
+	rd		%ccr, %g0;		\
+	.section	.pause_3insn_patch,"ax";\
+	.word		88b;			\
+	sllx		tmp, 7, tmp;		\
+	wr		tmp, 0, %asr27;		\
+	clr		tmp;			\
+	.previous;				\
+	brnz,pt		tmp, 88b;		\
+	 sub		tmp, 1, tmp;		\
+	set		BACKOFF_LIMIT, tmp;	\
+	cmp		reg, tmp;		\
+	bg,pn		%xcc, label;		\
+	 nop;					\
+	ba,pt		%xcc, label;		\
+	 sllx		reg, 1, reg;
 
 #else
 
diff --git a/arch/sparc/include/asm/processor_64.h b/arch/sparc/include/asm/processor_64.h
index 0305d56d9b1..cce72ce4c33 100644
--- a/arch/sparc/include/asm/processor_64.h
+++ b/arch/sparc/include/asm/processor_64.h
@@ -203,7 +203,22 @@ extern unsigned long get_wchan(struct task_struct *task);
 #define KSTK_EIP(tsk)  (task_pt_regs(tsk)->tpc)
 #define KSTK_ESP(tsk)  (task_pt_regs(tsk)->u_regs[UREG_FP])
 
-#define cpu_relax()	barrier()
+/* Please see the commentary in asm/backoff.h for a description of
+ * what these instructions are doing and how they have been choosen.
+ * To make a long story short, we are trying to yield the current cpu
+ * strand during busy loops.
+ */
+#define cpu_relax()	asm volatile("\n99:\n\t"			\
+				     "rd	%%ccr, %%g0\n\t"	\
+				     "rd	%%ccr, %%g0\n\t"	\
+				     "rd	%%ccr, %%g0\n\t"	\
+				     ".section	.pause_3insn_patch,\"ax\"\n\t"\
+				     ".word	99b\n\t"		\
+				     "wr	%%g0, 128, %%asr27\n\t"	\
+				     "nop\n\t"				\
+				     "nop\n\t"				\
+				     ".previous"			\
+				     ::: "memory")
 
 /* Prefetch support.  This is tuned for UltraSPARC-III and later.
  * UltraSPARC-I will treat these as nops, and UltraSPARC-II has
diff --git a/arch/sparc/include/asm/prom.h b/arch/sparc/include/asm/prom.h
index c2876511070..f93003123bc 100644
--- a/arch/sparc/include/asm/prom.h
+++ b/arch/sparc/include/asm/prom.h
@@ -63,5 +63,10 @@ extern char *of_console_options;
 extern void irq_trans_init(struct device_node *dp);
 extern char *build_path_component(struct device_node *dp);
 
+/* SPARC has a local implementation */
+extern int of_address_to_resource(struct device_node *dev, int index,
+				  struct resource *r);
+#define of_address_to_resource of_address_to_resource
+
 #endif /* __KERNEL__ */
 #endif /* _SPARC_PROM_H */
diff --git a/arch/sparc/include/uapi/asm/unistd.h b/arch/sparc/include/uapi/asm/unistd.h
index bed86a820d0..cac719d1bc5 100644
--- a/arch/sparc/include/uapi/asm/unistd.h
+++ b/arch/sparc/include/uapi/asm/unistd.h
@@ -406,8 +406,9 @@
 #define __NR_process_vm_readv	338
 #define __NR_process_vm_writev	339
 #define __NR_kern_features	340
+#define __NR_kcmp		341
 
-#define NR_syscalls		341
+#define NR_syscalls		342
 
 /* Bitmask values returned from kern_features system call.  */
 #define KERN_FEATURE_MIXED_MODE_STACK	0x00000001
diff --git a/arch/sparc/kernel/entry.h b/arch/sparc/kernel/entry.h
index 0c218e4c088..cc3c5cb47cd 100644
--- a/arch/sparc/kernel/entry.h
+++ b/arch/sparc/kernel/entry.h
@@ -59,6 +59,13 @@ struct popc_6insn_patch_entry {
 extern struct popc_6insn_patch_entry __popc_6insn_patch,
 	__popc_6insn_patch_end;
 
+struct pause_patch_entry {
+	unsigned int	addr;
+	unsigned int	insns[3];
+};
+extern struct pause_patch_entry __pause_3insn_patch,
+	__pause_3insn_patch_end;
+
 extern void __init per_cpu_patch(void);
 extern void sun4v_patch_1insn_range(struct sun4v_1insn_patch_entry *,
 				    struct sun4v_1insn_patch_entry *);
diff --git a/arch/sparc/kernel/leon_kernel.c b/arch/sparc/kernel/leon_kernel.c
index f8b6eee40bd..87f60ee6543 100644
--- a/arch/sparc/kernel/leon_kernel.c
+++ b/arch/sparc/kernel/leon_kernel.c
@@ -56,11 +56,13 @@ static inline unsigned int leon_eirq_get(int cpu)
 static void leon_handle_ext_irq(unsigned int irq, struct irq_desc *desc)
 {
 	unsigned int eirq;
+	struct irq_bucket *p;
 	int cpu = sparc_leon3_cpuid();
 
 	eirq = leon_eirq_get(cpu);
-	if ((eirq & 0x10) && irq_map[eirq]->irq) /* bit4 tells if IRQ happened */
-		generic_handle_irq(irq_map[eirq]->irq);
+	p = irq_map[eirq];
+	if ((eirq & 0x10) && p && p->irq) /* bit4 tells if IRQ happened */
+		generic_handle_irq(p->irq);
 }
 
 /* The extended IRQ controller has been found, this function registers it */
diff --git a/arch/sparc/kernel/setup_64.c b/arch/sparc/kernel/setup_64.c
index 0800e71d8a8..0eaf0059aae 100644
--- a/arch/sparc/kernel/setup_64.c
+++ b/arch/sparc/kernel/setup_64.c
@@ -316,6 +316,25 @@ static void __init popc_patch(void)
 	}
 }
 
+static void __init pause_patch(void)
+{
+	struct pause_patch_entry *p;
+
+	p = &__pause_3insn_patch;
+	while (p < &__pause_3insn_patch_end) {
+		unsigned long i, addr = p->addr;
+
+		for (i = 0; i < 3; i++) {
+			*(unsigned int *) (addr +  (i * 4)) = p->insns[i];
+			wmb();
+			__asm__ __volatile__("flush	%0"
+					     : : "r" (addr +  (i * 4)));
+		}
+
+		p++;
+	}
+}
+
 #ifdef CONFIG_SMP
 void __init boot_cpu_id_too_large(int cpu)
 {
@@ -528,6 +547,8 @@ static void __init init_sparc64_elf_hwcap(void)
 
 	if (sparc64_elf_hwcap & AV_SPARC_POPC)
 		popc_patch();
+	if (sparc64_elf_hwcap & AV_SPARC_PAUSE)
+		pause_patch();
 }
 
 void __init setup_arch(char **cmdline_p)
diff --git a/arch/sparc/kernel/systbls_32.S b/arch/sparc/kernel/systbls_32.S
index 63402f9e9f5..5147f574f12 100644
--- a/arch/sparc/kernel/systbls_32.S
+++ b/arch/sparc/kernel/systbls_32.S
@@ -85,3 +85,4 @@ sys_call_table:
 /*325*/	.long sys_pwritev, sys_rt_tgsigqueueinfo, sys_perf_event_open, sys_recvmmsg, sys_fanotify_init
 /*330*/	.long sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime
 /*335*/	.long sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev
+/*340*/	.long sys_ni_syscall, sys_kcmp
diff --git a/arch/sparc/kernel/systbls_64.S b/arch/sparc/kernel/systbls_64.S
index d8b22b3266d..ebb7f5fc58f 100644
--- a/arch/sparc/kernel/systbls_64.S
+++ b/arch/sparc/kernel/systbls_64.S
@@ -86,7 +86,7 @@ sys_call_table32:
 	.word compat_sys_pwritev, compat_sys_rt_tgsigqueueinfo, sys_perf_event_open, compat_sys_recvmmsg, sys_fanotify_init
 /*330*/	.word sys32_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, compat_sys_open_by_handle_at, compat_sys_clock_adjtime
 	.word sys_syncfs, compat_sys_sendmmsg, sys_setns, compat_sys_process_vm_readv, compat_sys_process_vm_writev
-/*340*/	.word sys_kern_features
+/*340*/	.word sys_kern_features, sys_kcmp
 
 #endif /* CONFIG_COMPAT */
 
@@ -164,4 +164,4 @@ sys_call_table:
 	.word sys_pwritev, sys_rt_tgsigqueueinfo, sys_perf_event_open, sys_recvmmsg, sys_fanotify_init
 /*330*/	.word sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime
 	.word sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev
-/*340*/	.word sys_kern_features
+/*340*/	.word sys_kern_features, sys_kcmp
diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S
index 89c2c29f154..0bacceb1915 100644
--- a/arch/sparc/kernel/vmlinux.lds.S
+++ b/arch/sparc/kernel/vmlinux.lds.S
@@ -132,6 +132,11 @@ SECTIONS
 		*(.popc_6insn_patch)
 		__popc_6insn_patch_end = .;
 	}
+	.pause_3insn_patch : {
+		__pause_3insn_patch = .;
+		*(.pause_3insn_patch)
+		__pause_3insn_patch_end = .;
+	}
 	PERCPU_SECTION(SMP_CACHE_BYTES)
 
 	. = ALIGN(PAGE_SIZE);
diff --git a/arch/sparc/lib/atomic_64.S b/arch/sparc/lib/atomic_64.S
index 4d502da3de7..85c233d0a34 100644
--- a/arch/sparc/lib/atomic_64.S
+++ b/arch/sparc/lib/atomic_64.S
@@ -1,6 +1,6 @@
 /* atomic.S: These things are too big to do inline.
  *
- * Copyright (C) 1999, 2007 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1999, 2007 2012 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/linkage.h>
@@ -117,3 +117,17 @@ ENTRY(atomic64_sub_ret) /* %o0 = decrement, %o1 = atomic_ptr */
 	 sub	%g1, %o0, %o0
 2:	BACKOFF_SPIN(%o2, %o3, 1b)
 ENDPROC(atomic64_sub_ret)
+
+ENTRY(atomic64_dec_if_positive) /* %o0 = atomic_ptr */
+	BACKOFF_SETUP(%o2)
+1:	ldx	[%o0], %g1
+	brlez,pn %g1, 3f
+	 sub	%g1, 1, %g7
+	casx	[%o0], %g1, %g7
+	cmp	%g1, %g7
+	bne,pn	%xcc, BACKOFF_LABEL(2f, 1b)
+	 nop
+3:	retl
+	 sub	%g1, 1, %o0
+2:	BACKOFF_SPIN(%o2, %o3, 1b)
+ENDPROC(atomic64_dec_if_positive)
diff --git a/arch/sparc/lib/ksyms.c b/arch/sparc/lib/ksyms.c
index ee31b884c61..0c4e35e522f 100644
--- a/arch/sparc/lib/ksyms.c
+++ b/arch/sparc/lib/ksyms.c
@@ -116,6 +116,7 @@ EXPORT_SYMBOL(atomic64_add);
 EXPORT_SYMBOL(atomic64_add_ret);
 EXPORT_SYMBOL(atomic64_sub);
 EXPORT_SYMBOL(atomic64_sub_ret);
+EXPORT_SYMBOL(atomic64_dec_if_positive);
 
 /* Atomic bit operations. */
 EXPORT_SYMBOL(test_and_set_bit);
diff --git a/arch/x86/include/asm/xen/hypercall.h b/arch/x86/include/asm/xen/hypercall.h
index 59c226d120c..c20d1ce62dc 100644
--- a/arch/x86/include/asm/xen/hypercall.h
+++ b/arch/x86/include/asm/xen/hypercall.h
@@ -359,18 +359,14 @@ HYPERVISOR_update_va_mapping(unsigned long va, pte_t new_val,
 		return _hypercall4(int, update_va_mapping, va,
 				   new_val.pte, new_val.pte >> 32, flags);
 }
+extern int __must_check xen_event_channel_op_compat(int, void *);
 
 static inline int
 HYPERVISOR_event_channel_op(int cmd, void *arg)
 {
 	int rc = _hypercall2(int, event_channel_op, cmd, arg);
-	if (unlikely(rc == -ENOSYS)) {
-		struct evtchn_op op;
-		op.cmd = cmd;
-		memcpy(&op.u, arg, sizeof(op.u));
-		rc = _hypercall1(int, event_channel_op_compat, &op);
-		memcpy(arg, &op.u, sizeof(op.u));
-	}
+	if (unlikely(rc == -ENOSYS))
+		rc = xen_event_channel_op_compat(cmd, arg);
 	return rc;
 }
 
@@ -386,17 +382,14 @@ HYPERVISOR_console_io(int cmd, int count, char *str)
 	return _hypercall3(int, console_io, cmd, count, str);
 }
 
+extern int __must_check HYPERVISOR_physdev_op_compat(int, void *);
+
 static inline int
 HYPERVISOR_physdev_op(int cmd, void *arg)
 {
 	int rc = _hypercall2(int, physdev_op, cmd, arg);
-	if (unlikely(rc == -ENOSYS)) {
-		struct physdev_op op;
-		op.cmd = cmd;
-		memcpy(&op.u, arg, sizeof(op.u));
-		rc = _hypercall1(int, physdev_op_compat, &op);
-		memcpy(arg, &op.u, sizeof(op.u));
-	}
+	if (unlikely(rc == -ENOSYS))
+		rc = HYPERVISOR_physdev_op_compat(cmd, arg);
 	return rc;
 }