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

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu:
  percpu: add __percpu sparse annotations to what's left
  percpu: add __percpu sparse annotations to fs
  percpu: add __percpu sparse annotations to core kernel subsystems
  local_t: Remove leftover local.h
  this_cpu: Remove pageset_notifier
  this_cpu: Page allocator conversion
  percpu, x86: Generic inc / dec percpu instructions
  local_t: Move local.h include to ringbuffer.c and ring_buffer_benchmark.c
  module: Use this_cpu_xx to dynamically allocate counters
  local_t: Remove cpu_local_xx macros
  percpu: refactor the code in pcpu_[de]populate_chunk()
  percpu: remove compile warnings caused by __verify_pcpu_ptr()
  percpu: make accessors check for percpu pointer in sparse
  percpu: add __percpu for sparse.
  percpu: make access macros universal
  percpu: remove per_cpu__ prefix.

60 files changed, 354 insertions, 499 deletions

diff --git a/arch/alpha/include/asm/local.h b/arch/alpha/include/asm/local.h
index 6ad3ea6..b9e3e33 100644
--- a/arch/alpha/include/asm/local.h
+++ b/arch/alpha/include/asm/local.h
@@ -98,21 +98,4 @@ static __inline__ long local_sub_return(long i, local_t * l)
 #define __local_add(i,l)	((l)->a.counter+=(i))
 #define __local_sub(i,l)	((l)->a.counter-=(i))
 
-/* Use these for per-cpu local_t variables: on some archs they are
- * much more efficient than these naive implementations.  Note they take
- * a variable, not an address.
- */
-#define cpu_local_read(l)	local_read(&__get_cpu_var(l))
-#define cpu_local_set(l, i)	local_set(&__get_cpu_var(l), (i))
-
-#define cpu_local_inc(l)	local_inc(&__get_cpu_var(l))
-#define cpu_local_dec(l)	local_dec(&__get_cpu_var(l))
-#define cpu_local_add(i, l)	local_add((i), &__get_cpu_var(l))
-#define cpu_local_sub(i, l)	local_sub((i), &__get_cpu_var(l))
-
-#define __cpu_local_inc(l)	__local_inc(&__get_cpu_var(l))
-#define __cpu_local_dec(l)	__local_dec(&__get_cpu_var(l))
-#define __cpu_local_add(i, l)	__local_add((i), &__get_cpu_var(l))
-#define __cpu_local_sub(i, l)	__local_sub((i), &__get_cpu_var(l))
-
 #endif /* _ALPHA_LOCAL_H */
diff --git a/arch/blackfin/mach-common/entry.S b/arch/blackfin/mach-common/entry.S
index b0ed0b4..01b2f58 100644
--- a/arch/blackfin/mach-common/entry.S
+++ b/arch/blackfin/mach-common/entry.S
@@ -816,8 +816,8 @@ ENDPROC(_resume)
 
 ENTRY(_ret_from_exception)
 #ifdef CONFIG_IPIPE
-	p2.l = _per_cpu__ipipe_percpu_domain;
-	p2.h = _per_cpu__ipipe_percpu_domain;
+	p2.l = _ipipe_percpu_domain;
+	p2.h = _ipipe_percpu_domain;
 	r0.l = _ipipe_root;
 	r0.h = _ipipe_root;
 	r2 = [p2];
diff --git a/arch/cris/arch-v10/kernel/entry.S b/arch/cris/arch-v10/kernel/entry.S
index 2c18d08..c52bef3 100644
--- a/arch/cris/arch-v10/kernel/entry.S
+++ b/arch/cris/arch-v10/kernel/entry.S
@@ -358,7 +358,7 @@ mmu_bus_fault:
 1:	btstq	12, $r1		   ; Refill?
 	bpl	2f
 	lsrq	24, $r1     ; Get PGD index (bit 24-31)
-	move.d  [per_cpu__current_pgd], $r0 ; PGD for the current process
+	move.d  [current_pgd], $r0 ; PGD for the current process
 	move.d	[$r0+$r1.d], $r0   ; Get PMD
 	beq	2f
 	nop
diff --git a/arch/cris/arch-v32/mm/mmu.S b/arch/cris/arch-v32/mm/mmu.S
index 2238d15..f125d91 100644
--- a/arch/cris/arch-v32/mm/mmu.S
+++ b/arch/cris/arch-v32/mm/mmu.S
@@ -115,7 +115,7 @@
 #ifdef CONFIG_SMP
 	move    $s7, $acr	; PGD
 #else
-	move.d  per_cpu__current_pgd, $acr ; PGD
+	move.d  current_pgd, $acr ; PGD
 #endif
 	; Look up PMD in PGD
 	lsrq	24, $r0	; Get PMD index into PGD (bit 24-31)
diff --git a/arch/ia64/include/asm/percpu.h b/arch/ia64/include/asm/percpu.h
index 30cf465..f7c00a5 100644
--- a/arch/ia64/include/asm/percpu.h
+++ b/arch/ia64/include/asm/percpu.h
@@ -9,7 +9,7 @@
 #define PERCPU_ENOUGH_ROOM PERCPU_PAGE_SIZE
 
 #ifdef __ASSEMBLY__
-# define THIS_CPU(var)	(per_cpu__##var)  /* use this to mark accesses to per-CPU variables... */
+# define THIS_CPU(var)	(var)  /* use this to mark accesses to per-CPU variables... */
 #else /* !__ASSEMBLY__ */
 
 
@@ -39,7 +39,7 @@ extern void *per_cpu_init(void);
  * On the positive side, using __ia64_per_cpu_var() instead of __get_cpu_var() is slightly
  * more efficient.
  */
-#define __ia64_per_cpu_var(var)	per_cpu__##var
+#define __ia64_per_cpu_var(var)	var
 
 #include <asm-generic/percpu.h>
 
diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c
index 461b999..7f4a0ed 100644
--- a/arch/ia64/kernel/ia64_ksyms.c
+++ b/arch/ia64/kernel/ia64_ksyms.c
@@ -30,9 +30,9 @@ EXPORT_SYMBOL(max_low_pfn);	/* defined by bootmem.c, but not exported by generic
 #endif
 
 #include <asm/processor.h>
-EXPORT_SYMBOL(per_cpu__ia64_cpu_info);
+EXPORT_SYMBOL(ia64_cpu_info);
 #ifdef CONFIG_SMP
-EXPORT_SYMBOL(per_cpu__local_per_cpu_offset);
+EXPORT_SYMBOL(local_per_cpu_offset);
 #endif
 
 #include <asm/uaccess.h>
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index 19c4b21..8d586d1 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -459,7 +459,7 @@ static void __init initialize_pernode_data(void)
 		cpu = 0;
 		node = node_cpuid[cpu].nid;
 		cpu0_cpu_info = (struct cpuinfo_ia64 *)(__phys_per_cpu_start +
-			((char *)&per_cpu__ia64_cpu_info - __per_cpu_start));
+			((char *)&ia64_cpu_info - __per_cpu_start));
 		cpu0_cpu_info->node_data = mem_data[node].node_data;
 	}
 #endif /* CONFIG_SMP */
diff --git a/arch/m32r/include/asm/local.h b/arch/m32r/include/asm/local.h
index 22256d1..734bca8 100644
--- a/arch/m32r/include/asm/local.h
+++ b/arch/m32r/include/asm/local.h
@@ -338,29 +338,4 @@ static inline void local_set_mask(unsigned long  mask, local_t *addr)
  * a variable, not an address.
  */
 
-/* Need to disable preemption for the cpu local counters otherwise we could
-   still access a variable of a previous CPU in a non local way. */
-#define cpu_local_wrap_v(l)	 	\
-	({ local_t res__;		\
-	   preempt_disable(); 		\
-	   res__ = (l);			\
-	   preempt_enable();		\
-	   res__; })
-#define cpu_local_wrap(l)		\
-	({ preempt_disable();		\
-	   l;				\
-	   preempt_enable(); })		\
-
-#define cpu_local_read(l)    cpu_local_wrap_v(local_read(&__get_cpu_var(l)))
-#define cpu_local_set(l, i)  cpu_local_wrap(local_set(&__get_cpu_var(l), (i)))
-#define cpu_local_inc(l)     cpu_local_wrap(local_inc(&__get_cpu_var(l)))
-#define cpu_local_dec(l)     cpu_local_wrap(local_dec(&__get_cpu_var(l)))
-#define cpu_local_add(i, l)  cpu_local_wrap(local_add((i), &__get_cpu_var(l)))
-#define cpu_local_sub(i, l)  cpu_local_wrap(local_sub((i), &__get_cpu_var(l)))
-
-#define __cpu_local_inc(l)	cpu_local_inc(l)
-#define __cpu_local_dec(l)	cpu_local_dec(l)
-#define __cpu_local_add(i, l)	cpu_local_add((i), (l))
-#define __cpu_local_sub(i, l)	cpu_local_sub((i), (l))
-
 #endif /* __M32R_LOCAL_H */
diff --git a/arch/microblaze/include/asm/entry.h b/arch/microblaze/include/asm/entry.h
index 61abbd2..ec89f2a 100644
--- a/arch/microblaze/include/asm/entry.h
+++ b/arch/microblaze/include/asm/entry.h
@@ -21,7 +21,7 @@
  * places
  */
 
-#define PER_CPU(var) per_cpu__##var
+#define PER_CPU(var) var
 
 # ifndef __ASSEMBLY__
 DECLARE_PER_CPU(unsigned int, KSP); /* Saved kernel stack pointer */
diff --git a/arch/mips/include/asm/local.h b/arch/mips/include/asm/local.h
index 361f4f16..bdcdef0 100644
--- a/arch/mips/include/asm/local.h
+++ b/arch/mips/include/asm/local.h
@@ -193,29 +193,4 @@ static __inline__ long local_sub_return(long i, local_t * l)
 #define __local_add(i, l)	((l)->a.counter+=(i))
 #define __local_sub(i, l)	((l)->a.counter-=(i))
 
-/* Need to disable preemption for the cpu local counters otherwise we could
-   still access a variable of a previous CPU in a non atomic way. */
-#define cpu_local_wrap_v(l)	 	\
-	({ local_t res__;		\
-	   preempt_disable(); 		\
-	   res__ = (l);			\
-	   preempt_enable();		\
-	   res__; })
-#define cpu_local_wrap(l)		\
-	({ preempt_disable();		\
-	   l;				\
-	   preempt_enable(); })		\
-
-#define cpu_local_read(l)    cpu_local_wrap_v(local_read(&__get_cpu_var(l)))
-#define cpu_local_set(l, i)  cpu_local_wrap(local_set(&__get_cpu_var(l), (i)))
-#define cpu_local_inc(l)     cpu_local_wrap(local_inc(&__get_cpu_var(l)))
-#define cpu_local_dec(l)     cpu_local_wrap(local_dec(&__get_cpu_var(l)))
-#define cpu_local_add(i, l)  cpu_local_wrap(local_add((i), &__get_cpu_var(l)))
-#define cpu_local_sub(i, l)  cpu_local_wrap(local_sub((i), &__get_cpu_var(l)))
-
-#define __cpu_local_inc(l)	cpu_local_inc(l)
-#define __cpu_local_dec(l)	cpu_local_dec(l)
-#define __cpu_local_add(i, l)	cpu_local_add((i), (l))
-#define __cpu_local_sub(i, l)	cpu_local_sub((i), (l))
-
 #endif /* _ARCH_MIPS_LOCAL_H */
diff --git a/arch/parisc/lib/fixup.S b/arch/parisc/lib/fixup.S
index d172d42..f8c45cc 100644
--- a/arch/parisc/lib/fixup.S
+++ b/arch/parisc/lib/fixup.S
@@ -36,8 +36,8 @@
 #endif
 	/* t2 = &__per_cpu_offset[smp_processor_id()]; */
 	LDREGX \t2(\t1),\t2 
-	addil LT%per_cpu__exception_data,%r27
-	LDREG RT%per_cpu__exception_data(%r1),\t1
+	addil LT%exception_data,%r27
+	LDREG RT%exception_data(%r1),\t1
 	/* t1 = &__get_cpu_var(exception_data) */
 	add,l \t1,\t2,\t1
 	/* t1 = t1->fault_ip */
@@ -46,8 +46,8 @@
 #else
 	.macro  get_fault_ip t1 t2
 	/* t1 = &__get_cpu_var(exception_data) */
-	addil LT%per_cpu__exception_data,%r27
-	LDREG RT%per_cpu__exception_data(%r1),\t2
+	addil LT%exception_data,%r27
+	LDREG RT%exception_data(%r1),\t2
 	/* t1 = t2->fault_ip */
 	LDREG EXCDATA_IP(\t2), \t1
 	.endm
diff --git a/arch/powerpc/include/asm/local.h b/arch/powerpc/include/asm/local.h
index ce58c80..c2410af 100644
--- a/arch/powerpc/include/asm/local.h
+++ b/arch/powerpc/include/asm/local.h
@@ -172,29 +172,4 @@ static __inline__ long local_dec_if_positive(local_t *l)
 #define __local_add(i,l)	((l)->a.counter+=(i))
 #define __local_sub(i,l)	((l)->a.counter-=(i))
 
-/* Need to disable preemption for the cpu local counters otherwise we could
-   still access a variable of a previous CPU in a non atomic way. */
-#define cpu_local_wrap_v(l)	 	\
-	({ local_t res__;		\
-	   preempt_disable(); 		\
-	   res__ = (l);			\
-	   preempt_enable();		\
-	   res__; })
-#define cpu_local_wrap(l)		\
-	({ preempt_disable();		\
-	   l;				\
-	   preempt_enable(); })		\
-
-#define cpu_local_read(l)    cpu_local_wrap_v(local_read(&__get_cpu_var(l)))
-#define cpu_local_set(l, i)  cpu_local_wrap(local_set(&__get_cpu_var(l), (i)))
-#define cpu_local_inc(l)     cpu_local_wrap(local_inc(&__get_cpu_var(l)))
-#define cpu_local_dec(l)     cpu_local_wrap(local_dec(&__get_cpu_var(l)))
-#define cpu_local_add(i, l)  cpu_local_wrap(local_add((i), &__get_cpu_var(l)))
-#define cpu_local_sub(i, l)  cpu_local_wrap(local_sub((i), &__get_cpu_var(l)))
-
-#define __cpu_local_inc(l)	cpu_local_inc(l)
-#define __cpu_local_dec(l)	cpu_local_dec(l)
-#define __cpu_local_add(i, l)	cpu_local_add((i), (l))
-#define __cpu_local_sub(i, l)	cpu_local_sub((i), (l))
-
 #endif /* _ARCH_POWERPC_LOCAL_H */
diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c
index d242a73..b287b62 100644
--- a/arch/sparc/kernel/nmi.c
+++ b/arch/sparc/kernel/nmi.c
@@ -21,7 +21,6 @@
 
 #include <asm/perf_event.h>
 #include <asm/ptrace.h>
-#include <asm/local.h>
 #include <asm/pcr.h>
 
 /* We don't have a real NMI on sparc64, but we can fake one
@@ -113,13 +112,13 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs)
 		touched = 1;
 	}
 	if (!touched && __get_cpu_var(last_irq_sum) == sum) {
-		__this_cpu_inc(per_cpu_var(alert_counter));
-		if (__this_cpu_read(per_cpu_var(alert_counter)) == 30 * nmi_hz)
+		__this_cpu_inc(alert_counter);
+		if (__this_cpu_read(alert_counter) == 30 * nmi_hz)
 			die_nmi("BUG: NMI Watchdog detected LOCKUP",
 				regs, panic_on_timeout);
 	} else {
 		__get_cpu_var(last_irq_sum) = sum;
-		__this_cpu_write(per_cpu_var(alert_counter), 0);
+		__this_cpu_write(alert_counter, 0);
 	}
 	if (__get_cpu_var(wd_enabled)) {
 		write_pic(picl_value(nmi_hz));
diff --git a/arch/sparc/kernel/rtrap_64.S b/arch/sparc/kernel/rtrap_64.S
index fd3cee4..1ddec40 100644
--- a/arch/sparc/kernel/rtrap_64.S
+++ b/arch/sparc/kernel/rtrap_64.S
@@ -149,11 +149,11 @@ rtrap_nmi:	ldx			[%sp + PTREGS_OFF + PT_V9_TSTATE], %l1
 rtrap_irq:
 rtrap:
 #ifndef CONFIG_SMP
-		sethi			%hi(per_cpu____cpu_data), %l0
-		lduw			[%l0 + %lo(per_cpu____cpu_data)], %l1
+		sethi			%hi(__cpu_data), %l0
+		lduw			[%l0 + %lo(__cpu_data)], %l1
 #else
-		sethi			%hi(per_cpu____cpu_data), %l0
-		or			%l0, %lo(per_cpu____cpu_data), %l0
+		sethi			%hi(__cpu_data), %l0
+		or			%l0, %lo(__cpu_data), %l0
 		lduw			[%l0 + %g5], %l1
 #endif
 		cmp			%l1, 0
diff --git a/arch/x86/include/asm/local.h b/arch/x86/include/asm/local.h
index 47b9b6f..2e99724 100644
--- a/arch/x86/include/asm/local.h
+++ b/arch/x86/include/asm/local.h
@@ -195,41 +195,4 @@ static inline long local_sub_return(long i, local_t *l)
 #define __local_add(i, l)	local_add((i), (l))
 #define __local_sub(i, l)	local_sub((i), (l))
 
-/* Use these for per-cpu local_t variables: on some archs they are
- * much more efficient than these naive implementations.  Note they take
- * a variable, not an address.
- *
- * X86_64: This could be done better if we moved the per cpu data directly
- * after GS.
- */
-
-/* Need to disable preemption for the cpu local counters otherwise we could
-   still access a variable of a previous CPU in a non atomic way. */
-#define cpu_local_wrap_v(l)		\
-({					\
-	local_t res__;			\
-	preempt_disable(); 		\
-	res__ = (l);			\
-	preempt_enable();		\
-	res__;				\
-})
-#define cpu_local_wrap(l)		\
-({					\
-	preempt_disable();		\
-	(l);				\
-	preempt_enable();		\
-})					\
-
-#define cpu_local_read(l)    cpu_local_wrap_v(local_read(&__get_cpu_var((l))))
-#define cpu_local_set(l, i)  cpu_local_wrap(local_set(&__get_cpu_var((l)), (i)))
-#define cpu_local_inc(l)     cpu_local_wrap(local_inc(&__get_cpu_var((l))))
-#define cpu_local_dec(l)     cpu_local_wrap(local_dec(&__get_cpu_var((l))))
-#define cpu_local_add(i, l)  cpu_local_wrap(local_add((i), &__get_cpu_var((l))))
-#define cpu_local_sub(i, l)  cpu_local_wrap(local_sub((i), &__get_cpu_var((l))))
-
-#define __cpu_local_inc(l)	cpu_local_inc((l))
-#define __cpu_local_dec(l)	cpu_local_dec((l))
-#define __cpu_local_add(i, l)	cpu_local_add((i), (l))
-#define __cpu_local_sub(i, l)	cpu_local_sub((i), (l))
-
 #endif /* _ASM_X86_LOCAL_H */
diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h
index 0c44196..66a272d 100644
--- a/arch/x86/include/asm/percpu.h
+++ b/arch/x86/include/asm/percpu.h
@@ -25,19 +25,18 @@
  */
 #ifdef CONFIG_SMP
 #define PER_CPU(var, reg)						\
-	__percpu_mov_op %__percpu_seg:per_cpu__this_cpu_off, reg;	\
-	lea per_cpu__##var(reg), reg
-#define PER_CPU_VAR(var)	%__percpu_seg:per_cpu__##var
+	__percpu_mov_op %__percpu_seg:this_cpu_off, reg;		\
+	lea var(reg), reg
+#define PER_CPU_VAR(var)	%__percpu_seg:var
 #else /* ! SMP */
-#define PER_CPU(var, reg)						\
-	__percpu_mov_op $per_cpu__##var, reg
-#define PER_CPU_VAR(var)	per_cpu__##var
+#define PER_CPU(var, reg)	__percpu_mov_op $var, reg
+#define PER_CPU_VAR(var)	var
 #endif	/* SMP */
 
 #ifdef CONFIG_X86_64_SMP
 #define INIT_PER_CPU_VAR(var)  init_per_cpu__##var
 #else
-#define INIT_PER_CPU_VAR(var)  per_cpu__##var
+#define INIT_PER_CPU_VAR(var)  var
 #endif
 
 #else /* ...!ASSEMBLY */
@@ -60,12 +59,12 @@
  * There also must be an entry in vmlinux_64.lds.S
  */
 #define DECLARE_INIT_PER_CPU(var) \
-       extern typeof(per_cpu_var(var)) init_per_cpu_var(var)
+       extern typeof(var) init_per_cpu_var(var)
 
 #ifdef CONFIG_X86_64_SMP
 #define init_per_cpu_var(var)  init_per_cpu__##var
 #else
-#define init_per_cpu_var(var)  per_cpu_var(var)
+#define init_per_cpu_var(var)  var
 #endif
 
 /* For arch-specific code, we can use direct single-insn ops (they
@@ -104,6 +103,64 @@ do {							\
 	}						\
 } while (0)
 
+/*
+ * Generate a percpu add to memory instruction and optimize code
+ * if a one is added or subtracted.
+ */
+#define percpu_add_op(var, val)						\
+do {									\
+	typedef typeof(var) pao_T__;					\
+	const int pao_ID__ = (__builtin_constant_p(val) &&		\
+			      ((val) == 1 || (val) == -1)) ? (val) : 0;	\
+	if (0) {							\
+		pao_T__ pao_tmp__;					\
+		pao_tmp__ = (val);					\
+	}								\
+	switch (sizeof(var)) {						\
+	case 1:								\
+		if (pao_ID__ == 1)					\
+			asm("incb "__percpu_arg(0) : "+m" (var));	\
+		else if (pao_ID__ == -1)				\
+			asm("decb "__percpu_arg(0) : "+m" (var));	\
+		else							\
+			asm("addb %1, "__percpu_arg(0)			\
+			    : "+m" (var)				\
+			    : "qi" ((pao_T__)(val)));			\
+		break;							\
+	case 2:								\
+		if (pao_ID__ == 1)					\
+			asm("incw "__percpu_arg(0) : "+m" (var));	\
+		else if (pao_ID__ == -1)				\
+			asm("decw "__percpu_arg(0) : "+m" (var));	\
+		else							\
+			asm("addw %1, "__percpu_arg(0)			\
+			    : "+m" (var)				\
+			    : "ri" ((pao_T__)(val)));			\
+		break;							\
+	case 4:								\
+		if (pao_ID__ == 1)					\
+			asm("incl "__percpu_arg(0) : "+m" (var));	\
+		else if (pao_ID__ == -1)				\
+			asm("decl "__percpu_arg(0) : "+m" (var));	\
+		else							\
+			asm("addl %1, "__percpu_arg(0)			\
+			    : "+m" (var)				\
+			    : "ri" ((pao_T__)(val)));			\
+		break;							\
+	case 8:								\
+		if (pao_ID__ == 1)					\
+			asm("incq "__percpu_arg(0) : "+m" (var));	\
+		else if (pao_ID__ == -1)				\
+			asm("decq "__percpu_arg(0) : "+m" (var));	\
+		else							\
+			asm("addq %1, "__percpu_arg(0)			\
+			    : "+m" (var)				\
+			    : "re" ((pao_T__)(val)));			\
+		break;							\
+	default: __bad_percpu_size();					\
+	}								\
+} while (0)
+
 #define percpu_from_op(op, var, constraint)		\
 ({							\
 	typeof(var) pfo_ret__;				\
@@ -142,16 +199,14 @@ do {							\
  * per-thread variables implemented as per-cpu variables and thus
  * stable for the duration of the respective task.
  */
-#define percpu_read(var)	percpu_from_op("mov", per_cpu__##var,	\
-					       "m" (per_cpu__##var))
-#define percpu_read_stable(var)	percpu_from_op("mov", per_cpu__##var,	\
-					       "p" (&per_cpu__##var))
-#define percpu_write(var, val)	percpu_to_op("mov", per_cpu__##var, val)
-#define percpu_add(var, val)	percpu_to_op("add", per_cpu__##var, val)
-#define percpu_sub(var, val)	percpu_to_op("sub", per_cpu__##var, val)
-#define percpu_and(var, val)	percpu_to_op("and", per_cpu__##var, val)
-#define percpu_or(var, val)	percpu_to_op("or", per_cpu__##var, val)
-#define percpu_xor(var, val)	percpu_to_op("xor", per_cpu__##var, val)
+#define percpu_read(var)		percpu_from_op("mov", var, "m" (var))
+#define percpu_read_stable(var)		percpu_from_op("mov", var, "p" (&(var)))
+#define percpu_write(var, val)		percpu_to_op("mov", var, val)
+#define percpu_add(var, val)		percpu_add_op(var, val)
+#define percpu_sub(var, val)		percpu_add_op(var, -(val))
+#define percpu_and(var, val)		percpu_to_op("and", var, val)
+#define percpu_or(var, val)		percpu_to_op("or", var, val)
+#define percpu_xor(var, val)		percpu_to_op("xor", var, val)
 
 #define __this_cpu_read_1(pcp)		percpu_from_op("mov", (pcp), "m"(pcp))
 #define __this_cpu_read_2(pcp)		percpu_from_op("mov", (pcp), "m"(pcp))
@@ -160,9 +215,9 @@ do {							\
 #define __this_cpu_write_1(pcp, val)	percpu_to_op("mov", (pcp), val)
 #define __this_cpu_write_2(pcp, val)	percpu_to_op("mov", (pcp), val)
 #define __this_cpu_write_4(pcp, val)	percpu_to_op("mov", (pcp), val)
-#define __this_cpu_add_1(pcp, val)	percpu_to_op("add", (pcp), val)
-#define __this_cpu_add_2(pcp, val)	percpu_to_op("add", (pcp), val)
-#define __this_cpu_add_4(pcp, val)	percpu_to_op("add", (pcp), val)
+#define __this_cpu_add_1(pcp, val)	percpu_add_op((pcp), val)
+#define __this_cpu_add_2(pcp, val)	percpu_add_op((pcp), val)
+#define __this_cpu_add_4(pcp, val)	percpu_add_op((pcp), val)
 #define __this_cpu_and_1(pcp, val)	percpu_to_op("and", (pcp), val)
 #define __this_cpu_and_2(pcp, val)	percpu_to_op("and", (pcp), val)
 #define __this_cpu_and_4(pcp, val)	percpu_to_op("and", (pcp), val)
@@ -179,9 +234,9 @@ do {							\
 #define this_cpu_write_1(pcp, val)	percpu_to_op("mov", (pcp), val)
 #define this_cpu_write_2(pcp, val)	percpu_to_op("mov", (pcp), val)
 #define this_cpu_write_4(pcp, val)	percpu_to_op("mov", (pcp), val)
-#define this_cpu_add_1(pcp, val)	percpu_to_op("add", (pcp), val)
-#define this_cpu_add_2(pcp, val)	percpu_to_op("add", (pcp), val)
-#define this_cpu_add_4(pcp, val)	percpu_to_op("add", (pcp), val)
+#define this_cpu_add_1(pcp, val)	percpu_add_op((pcp), val)
+#define this_cpu_add_2(pcp, val)	percpu_add_op((pcp), val)
+#define this_cpu_add_4(pcp, val)	percpu_add_op((pcp), val)
 #define this_cpu_and_1(pcp, val)	percpu_to_op("and", (pcp), val)
 #define this_cpu_and_2(pcp, val)	percpu_to_op("and", (pcp), val)
 #define this_cpu_and_4(pcp, val)	percpu_to_op("and", (pcp), val)
@@ -192,9 +247,9 @@ do {							\
 #define this_cpu_xor_2(pcp, val)	percpu_to_op("xor", (pcp), val)
 #define this_cpu_xor_4(pcp, val)	percpu_to_op("xor", (pcp), val)
 
-#define irqsafe_cpu_add_1(pcp, val)	percpu_to_op("add", (pcp), val)
-#define irqsafe_cpu_add_2(pcp, val)	percpu_to_op("add", (pcp), val)
-#define irqsafe_cpu_add_4(pcp, val)	percpu_to_op("add", (pcp), val)
+#define irqsafe_cpu_add_1(pcp, val)	percpu_add_op((pcp), val)
+#define irqsafe_cpu_add_2(pcp, val)	percpu_add_op((pcp), val)
+#define irqsafe_cpu_add_4(pcp, val)	percpu_add_op((pcp), val)
 #define irqsafe_cpu_and_1(pcp, val)	percpu_to_op("and", (pcp), val)
 #define irqsafe_cpu_and_2(pcp, val)	percpu_to_op("and", (pcp), val)
 #define irqsafe_cpu_and_4(pcp, val)	percpu_to_op("and", (pcp), val)
@@ -212,19 +267,19 @@ do {							\
 #ifdef CONFIG_X86_64
 #define __this_cpu_read_8(pcp)		percpu_from_op("mov", (pcp), "m"(pcp))
 #define __this_cpu_write_8(pcp, val)	percpu_to_op("mov", (pcp), val)
-#define __this_cpu_add_8(pcp, val)	percpu_to_op("add", (pcp), val)
+#define __this_cpu_add_8(pcp, val)	percpu_add_op((pcp), val)
 #define __this_cpu_and_8(pcp, val)	percpu_to_op("and", (pcp), val)
 #define __this_cpu_or_8(pcp, val)	percpu_to_op("or", (pcp), val)
 #define __this_cpu_xor_8(pcp, val)	percpu_to_op("xor", (pcp), val)
 
 #define this_cpu_read_8(pcp)		percpu_from_op("mov", (pcp), "m"(pcp))
 #define this_cpu_write_8(pcp, val)	percpu_to_op("mov", (pcp), val)
-#define this_cpu_add_8(pcp, val)	percpu_to_op("add", (pcp), val)
+#define this_cpu_add_8(pcp, val)	percpu_add_op((pcp), val)
 #define this_cpu_and_8(pcp, val)	percpu_to_op("and", (pcp), val)
 #define this_cpu_or_8(pcp, val)		percpu_to_op("or", (pcp), val)
 #define this_cpu_xor_8(pcp, val)	percpu_to_op("xor", (pcp), val)
 
-#define irqsafe_cpu_add_8(pcp, val)	percpu_to_op("add", (pcp), val)
+#define irqsafe_cpu_add_8(pcp, val)	percpu_add_op((pcp), val)
 #define irqsafe_cpu_and_8(pcp, val)	percpu_to_op("and", (pcp), val)
 #define irqsafe_cpu_or_8(pcp, val)	percpu_to_op("or", (pcp), val)
 #define irqsafe_cpu_xor_8(pcp, val)	percpu_to_op("xor", (pcp), val)
@@ -236,7 +291,7 @@ do {							\
 ({									\
 	int old__;							\
 	asm volatile("btr %2,"__percpu_arg(1)"\n\tsbbl %0,%0"		\
-		     : "=r" (old__), "+m" (per_cpu__##var)		\
+		     : "=r" (old__), "+m" (var)				\
 		     : "dIr" (bit));					\
 	old__;								\
 })
diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h
index e04740f..b8fe48e 100644
--- a/arch/x86/include/asm/system.h
+++ b/arch/x86/include/asm/system.h
@@ -32,7 +32,7 @@ extern void show_regs_common(void);
 	"movl %P[task_canary](%[next]), %%ebx\n\t"			\
 	"movl %%ebx, "__percpu_arg([stack_canary])"\n\t"
 #define __switch_canary_oparam						\
-	, [stack_canary] "=m" (per_cpu_var(stack_canary.canary))
+	, [stack_canary] "=m" (stack_canary.canary)
 #define __switch_canary_iparam						\
 	, [task_canary] "i" (offsetof(struct task_struct, stack_canary))
 #else	/* CC_STACKPROTECTOR */
@@ -114,7 +114,7 @@ do {									\
 	"movq %P[task_canary](%%rsi),%%r8\n\t"				  \
 	"movq %%r8,"__percpu_arg([gs_canary])"\n\t"
 #define __switch_canary_oparam						  \
-	, [gs_canary] "=m" (per_cpu_var(irq_stack_union.stack_canary))
+	, [gs_canary] "=m" (irq_stack_union.stack_canary)
 #define __switch_canary_iparam						  \
 	, [task_canary] "i" (offsetof(struct task_struct, stack_canary))
 #else	/* CC_STACKPROTECTOR */
@@ -133,7 +133,7 @@ do {									\
 	     __switch_canary						  \
 	     "movq %P[thread_info](%%rsi),%%r8\n\t"			  \
 	     "movq %%rax,%%rdi\n\t" 					  \
-	     "testl  %[_tif_fork],%P[ti_flags](%%r8)\n\t"	  \
+	     "testl  %[_tif_fork],%P[ti_flags](%%r8)\n\t"		  \
 	     "jnz   ret_from_fork\n\t"					  \
 	     RESTORE_CONTEXT						  \
 	     : "=a" (last)					  	  \
@@ -143,7 +143,7 @@ do {									\
 	       [ti_flags] "i" (offsetof(struct thread_info, flags)),	  \
 	       [_tif_fork] "i" (_TIF_FORK),			  	  \
 	       [thread_info] "i" (offsetof(struct task_struct, stack)),   \
-	       [current_task] "m" (per_cpu_var(current_task))		  \
+	       [current_task] "m" (current_task)			  \
 	       __switch_canary_iparam					  \
 	     : "memory", "cc" __EXTRA_CLOBBER)
 #endif
diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c
index 0159a69..4ada42c 100644
--- a/arch/x86/kernel/apic/nmi.c
+++ b/arch/x86/kernel/apic/nmi.c
@@ -438,8 +438,8 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 		 * Ayiee, looks like this CPU is stuck ...
 		 * wait a few IRQs (5 seconds) before doing the oops ...
 		 */
-		__this_cpu_inc(per_cpu_var(alert_counter));
-		if (__this_cpu_read(per_cpu_var(alert_counter)) == 5 * nmi_hz)
+		__this_cpu_inc(alert_counter);
+		if (__this_cpu_read(alert_counter) == 5 * nmi_hz)
 			/*
 			 * die_nmi will return ONLY if NOTIFY_STOP happens..
 			 */
@@ -447,7 +447,7 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 				regs, panic_on_timeout);
 	} else {
 		__get_cpu_var(last_irq_sum) = sum;
-		__this_cpu_write(per_cpu_var(alert_counter), 0);
+		__this_cpu_write(alert_counter, 0);
 	}
 
 	/* see if the nmi watchdog went off */
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index 7fd318b..37c3d4b 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -442,8 +442,8 @@ is386:	movl $2,%ecx		# set MP
 	 */
 	cmpb $0,ready
 	jne 1f
-	movl $per_cpu__gdt_page,%eax
-	movl $per_cpu__stack_canary,%ecx
+	movl $gdt_page,%eax
+	movl $stack_canary,%ecx
 	movw %cx, 8 * GDT_ENTRY_STACK_CANARY + 2(%eax)
 	shrl $16, %ecx
 	movb %cl, 8 * GDT_ENTRY_STACK_CANARY + 4(%eax)
@@ -706,7 +706,7 @@ idt_descr:
 	.word 0				# 32 bit align gdt_desc.address
 ENTRY(early_gdt_descr)
 	.word GDT_ENTRIES*8-1
-	.long per_cpu__gdt_page		/* Overwritten for secondary CPUs */
+	.long gdt_page			/* Overwritten for secondary CPUs */
 
 /*
  * The boot_gdt must mirror the equivalent in setup.S and is
diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S
index f92a0da..44879df 100644
--- a/arch/x86/kernel/vmlinux.lds.S
+++ b/arch/x86/kernel/vmlinux.lds.S
@@ -341,7 +341,7 @@ SECTIONS
  * Per-cpu symbols which need to be offset from __per_cpu_load
  * for the boot processor.
  */
-#define INIT_PER_CPU(x) init_per_cpu__##x = per_cpu__##x + __per_cpu_load
+#define INIT_PER_CPU(x) init_per_cpu__##x = x + __per_cpu_load
 INIT_PER_CPU(gdt_page);
 INIT_PER_CPU(irq_stack_union);
 
@@ -352,7 +352,7 @@ INIT_PER_CPU(irq_stack_union);
 	   "kernel image bigger than KERNEL_IMAGE_SIZE");
 
 #ifdef CONFIG_SMP
-. = ASSERT((per_cpu__irq_stack_union == 0),
+. = ASSERT((irq_stack_union == 0),
            "irq_stack_union is not at start of per-cpu area");
 #endif
 
diff --git a/arch/x86/xen/xen-asm_32.S b/arch/x86/xen/xen-asm_32.S
index 88e15de..22a2093 100644
--- a/arch/x86/xen/xen-asm_32.S
+++ b/arch/x86/xen/xen-asm_32.S
@@ -90,9 +90,9 @@ ENTRY(xen_iret)
 	GET_THREAD_INFO(%eax)
 	movl TI_cpu(%eax), %eax
 	movl __per_cpu_offset(,%eax,4), %eax
-	mov per_cpu__xen_vcpu(%eax), %eax
+	mov xen_vcpu(%eax), %eax
 #else
-	movl per_cpu__xen_vcpu, %eax
+	movl xen_vcpu, %eax
 #endif
 
 	/* check IF state we're restoring */
diff --git a/crypto/cryptd.c b/crypto/cryptd.c
index 704c141..ef71318 100644
--- a/crypto/cryptd.c
+++ b/crypto/cryptd.c
@@ -31,7 +31,7 @@ struct cryptd_cpu_queue {
 };
 
 struct cryptd_queue {
-	struct cryptd_cpu_queue *cpu_queue;
+	struct cryptd_cpu_queue __percpu *cpu_queue;
 };
 
 struct cryptd_instance_ctx {
diff --git a/drivers/acpi/processor_perflib.c b/drivers/acpi/processor_perflib.c
index a959f6a..d648a98 100644
--- a/drivers/acpi/processor_perflib.c
+++ b/drivers/acpi/processor_perflib.c
@@ -561,7 +561,7 @@ end:
 }
 
 int acpi_processor_preregister_performance(
-		struct acpi_processor_performance *performance)
+		struct acpi_processor_performance __percpu *performance)
 {
 	int count, count_target;
 	int retval = 0;
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index e7a3230..87399ca 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -284,7 +284,7 @@ struct dma_chan_tbl_ent {
 /**
  * channel_table - percpu lookup table for memory-to-memory offload providers
  */
-static struct dma_chan_tbl_ent *channel_table[DMA_TX_TYPE_END];
+static struct dma_chan_tbl_ent __percpu *channel_table[DMA_TX_TYPE_END];
 
 static int __init dma_channel_table_init(void)
 {
diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index 3391e67..7cd1cdc 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -13,7 +13,7 @@ module_param(report_gart_errors, int, 0644);
 static int ecc_enable_override;
 module_param(ecc_enable_override, int, 0644);
 
-static struct msr *msrs;
+static struct msr __percpu *msrs;
 
 /* Lookup table for all possible MC control instances */
 struct amd64_pvt;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 509c8f3..70ffbd0 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -4680,7 +4680,7 @@ static int raid5_alloc_percpu(raid5_conf_t *conf)
 {
 	unsigned long cpu;
 	struct page *spare_page;
-	struct raid5_percpu *allcpus;
+	struct raid5_percpu __percpu *allcpus;
 	void *scribble;
 	int err;
 
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index dd70835..0f86f5e 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -405,7 +405,7 @@ struct raid5_private_data {
 					      * lists and performing address
 					      * conversions
 					      */
-	} *percpu;
+	} __percpu *percpu;
 	size_t			scribble_len; /* size of scribble region must be
 					       * associated with conf to handle
 					       * cpu hotplug while reshaping
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 874d169..4cedc91 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -1014,7 +1014,7 @@ struct ext4_sb_info {
 	atomic_t s_lock_busy;
 
 	/* locality groups */
-	struct ext4_locality_group *s_locality_groups;
+	struct ext4_locality_group __percpu *s_locality_groups;
 
 	/* for write statistics */
 	unsigned long s_sectors_written_start;
diff --git a/fs/nfs/iostat.h b/fs/nfs/iostat.h
index 46d779a..1d8d5c8 100644
--- a/fs/nfs/iostat.h
+++ b/fs/nfs/iostat.h
@@ -57,12 +57,12 @@ static inline void nfs_add_fscache_stats(struct inode *inode,
 }
 #endif
 
-static inline struct nfs_iostats *nfs_alloc_iostats(void)
+static inline struct nfs_iostats __percpu *nfs_alloc_iostats(void)
 {
 	return alloc_percpu(struct nfs_iostats);
 }
 
-static inline void nfs_free_iostats(struct nfs_iostats *stats)
+static inline void nfs_free_iostats(struct nfs_iostats __percpu *stats)
 {
 	if (stats != NULL)
 		free_percpu(stats);
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 70504fc..14dafd6 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -245,7 +245,7 @@ typedef struct xfs_mount {
 	struct xfs_qmops	*m_qm_ops;	/* vector of XQM ops */
 	atomic_t		m_active_trans;	/* number trans frozen */
 #ifdef HAVE_PERCPU_SB
-	xfs_icsb_cnts_t		*m_sb_cnts;	/* per-cpu superblock counters */
+	xfs_icsb_cnts_t __percpu *m_sb_cnts;	/* per-cpu superblock counters */
 	unsigned long		m_icsb_counters; /* disabled per-cpu counters */
 	struct notifier_block	m_icsb_notifier; /* hotplug cpu notifier */
 	struct mutex		m_icsb_mutex;	/* balancer sync lock */
diff --git a/include/acpi/processor.h b/include/acpi/processor.h
index 2983176..1172c27 100644
--- a/include/acpi/processor.h
+++ b/include/acpi/processor.h
@@ -238,7 +238,7 @@ struct acpi_processor_errata {
 
 extern int acpi_processor_preregister_performance(struct
 						  acpi_processor_performance
-						  *performance);
+						  __percpu *performance);
 
 extern int acpi_processor_register_performance(struct acpi_processor_performance
 					       *performance, unsigned int cpu);
diff --git a/include/asm-generic/local.h b/include/asm-generic/local.h
index fc21844..c8a5d68 100644
--- a/include/asm-generic/local.h
+++ b/include/asm-generic/local.h
@@ -52,23 +52,4 @@ typedef struct
 #define __local_add(i,l)	local_set((l), local_read(l) + (i))
 #define __local_sub(i,l)	local_set((l), local_read(l) - (i))
 
-/* Use these for per-cpu local_t variables: on some archs they are
- * much more efficient than these naive implementations.  Note they take
- * a variable (eg. mystruct.foo), not an address.
- */
-#define cpu_local_read(l)	local_read(&__get_cpu_var(l))
-#define cpu_local_set(l, i)	local_set(&__get_cpu_var(l), (i))
-#define cpu_local_inc(l)	local_inc(&__get_cpu_var(l))
-#define cpu_local_dec(l)	local_dec(&__get_cpu_var(l))
-#define cpu_local_add(i, l)	local_add((i), &__get_cpu_var(l))
-#define cpu_local_sub(i, l)	local_sub((i), &__get_cpu_var(l))
-
-/* Non-atomic increments, ie. preemption disabled and won't be touched
- * in interrupt, etc.  Some archs can optimize this case well.
- */
-#define __cpu_local_inc(l)	__local_inc(&__get_cpu_var(l))
-#define __cpu_local_dec(l)	__local_dec(&__get_cpu_var(l))
-#define __cpu_local_add(i, l)	__local_add((i), &__get_cpu_var(l))
-#define __cpu_local_sub(i, l)	__local_sub((i), &__get_cpu_var(l))
-
 #endif /* _ASM_GENERIC_LOCAL_H */
diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index 8087b90..04f91c2 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -41,7 +41,11 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
  * Only S390 provides its own means of moving the pointer.
  */
 #ifndef SHIFT_PERCPU_PTR
-#define SHIFT_PERCPU_PTR(__p, __offset)	RELOC_HIDE((__p), (__offset))
+/* Weird cast keeps both GCC and sparse happy. */
+#define SHIFT_PERCPU_PTR(__p, __offset)	({				\
+	__verify_pcpu_ptr((__p));					\
+	RELOC_HIDE((typeof(*(__p)) __kernel __force *)(__p), (__offset)); \
+})
 #endif
 
 /*
@@ -50,11 +54,11 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
  * offset.
  */
 #define per_cpu(var, cpu) \
-	(*SHIFT_PERCPU_PTR(&per_cpu_var(var), per_cpu_offset(cpu)))
+	(*SHIFT_PERCPU_PTR(&(var), per_cpu_offset(cpu)))
 #define __get_cpu_var(var) \
-	(*SHIFT_PERCPU_PTR(&per_cpu_var(var), my_cpu_offset))
+	(*SHIFT_PERCPU_PTR(&(var), my_cpu_offset))
 #define __raw_get_cpu_var(var) \
-	(*SHIFT_PERCPU_PTR(&per_cpu_var(var), __my_cpu_offset))
+	(*SHIFT_PERCPU_PTR(&(var), __my_cpu_offset))
 
 #define this_cpu_ptr(ptr) SHIFT_PERCPU_PTR(ptr, my_cpu_offset)
 #define __this_cpu_ptr(ptr) SHIFT_PERCPU_PTR(ptr, __my_cpu_offset)
@@ -66,9 +70,9 @@ extern void setup_per_cpu_areas(void);
 
 #else /* ! SMP */
 
-#define per_cpu(var, cpu)			(*((void)(cpu), &per_cpu_var(var)))
-#define __get_cpu_var(var)			per_cpu_var(var)
-#define __raw_get_cpu_var(var)			per_cpu_var(var)
+#define per_cpu(var, cpu)			(*((void)(cpu), &(var)))
+#define __get_cpu_var(var)			(var)
+#define __raw_get_cpu_var(var)			(var)
 #define this_cpu_ptr(ptr) per_cpu_ptr(ptr, 0)
 #define __this_cpu_ptr(ptr) this_cpu_ptr(ptr)
 
diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h
index 3b73b99..416bf62 100644
--- a/include/linux/blktrace_api.h
+++ b/include/linux/blktrace_api.h
@@ -150,8 +150,8 @@ struct blk_user_trace_setup {
 struct blk_trace {
 	int trace_state;
 	struct rchan *rchan;
-	unsigned long *sequence;
-	unsigned char *msg_data;
+	unsigned long __percpu *sequence;
+	unsigned char __percpu *msg_data;
 	u16 act_mask;
 	u64 start_lba;
 	u64 end_lba;
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 188fcae..a5a472b 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -5,7 +5,7 @@
 
 #ifdef __CHECKER__
 # define __user		__attribute__((noderef, address_space(1)))
-# define __kernel	/* default address space */
+# define __kernel	__attribute__((address_space(0)))
 # define __safe		__attribute__((safe))
 # define __force	__attribute__((force))
 # define __nocast	__attribute__((nocast))
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 7878498..21fd9b7 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -162,7 +162,7 @@ struct dma_chan {
 	struct dma_chan_dev *dev;
 
 	struct list_head device_node;
-	struct dma_chan_percpu *local;
+	struct dma_chan_percpu __percpu *local;
 	int client_count;
 	int table_count;
 	void *private;
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 9717081..56b5051 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -101,7 +101,7 @@ struct hd_struct {
 	unsigned long stamp;
 	int in_flight[2];
 #ifdef	CONFIG_SMP
-	struct disk_stats *dkstats;
+	struct disk_stats __percpu *dkstats;
 #else
 	struct disk_stats dkstats;
 #endif
diff --git a/include/linux/kexec.h b/include/linux/kexec.h
index c356b69..03e8e8d 100644
--- a/include/linux/kexec.h
+++ b/include/linux/kexec.h
@@ -199,7 +199,7 @@ extern struct kimage *kexec_crash_image;
  */
 extern struct resource crashk_res;
 typedef u32 note_buf_t[KEXEC_NOTE_BYTES/4];
-extern note_buf_t *crash_notes;
+extern note_buf_t __percpu *crash_notes;
 extern u32 vmcoreinfo_note[VMCOREINFO_NOTE_SIZE/4];
 extern size_t vmcoreinfo_size;
 extern size_t vmcoreinfo_max_size;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 8b2fa85..2e724c8 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1081,11 +1081,7 @@ extern void si_meminfo(struct sysinfo * val);
 extern void si_meminfo_node(struct sysinfo *val, int nid);
 extern int after_bootmem;
 
-#ifdef CONFIG_NUMA
 extern void setup_per_cpu_pageset(void);
-#else
-static inline void setup_per_cpu_pageset(void) {}
-#endif
 
 extern void zone_pcp_update(struct zone *zone);
 
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 30fe668..41acd4b 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -184,13 +184,7 @@ struct per_cpu_pageset {
 	s8 stat_threshold;
 	s8 vm_stat_diff[NR_VM_ZONE_STAT_ITEMS];
 #endif
-} ____cacheline_aligned_in_smp;
-
-#ifdef CONFIG_NUMA
-#define zone_pcp(__z, __cpu) ((__z)->pageset[(__cpu)])
-#else
-#define zone_pcp(__z, __cpu) (&(__z)->pageset[(__cpu)])
-#endif
+};
 
 #endif /* !__GENERATING_BOUNDS.H */
 
@@ -306,10 +300,8 @@ struct zone {
 	 */
 	unsigned long		min_unmapped_pages;
 	unsigned long		min_slab_pages;
-	struct per_cpu_pageset	*pageset[NR_CPUS];
-#else
-	struct per_cpu_pageset	pageset[NR_CPUS];
 #endif
+	struct per_cpu_pageset __percpu *pageset;
 	/*
 	 * free areas of different sizes
 	 */
diff --git a/include/linux/module.h b/include/linux/module.h
index 6cb1a3c..dd618eb 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -17,7 +17,7 @@
 #include <linux/moduleparam.h>
 #include <linux/tracepoint.h>
 
-#include <asm/local.h>
+#include <linux/percpu.h>
 #include <asm/module.h>
 
 #include <trace/events/module.h>
@@ -363,11 +363,9 @@ struct module
 	/* Destruction function. */
 	void (*exit)(void);
 
-#ifdef CONFIG_SMP
-	char *refptr;
-#else
-	local_t ref;
-#endif
+	struct module_ref {
+		int count;
+	} __percpu *refptr;
 #endif
 
 #ifdef CONFIG_CONSTRUCTORS
@@ -454,25 +452,16 @@ void __symbol_put(const char *symbol);
 #define symbol_put(x) __symbol_put(MODULE_SYMBOL_PREFIX #x)
 void symbol_put_addr(void *addr);
 
-static inline local_t *__module_ref_addr(struct module *mod, int cpu)
-{
-#ifdef CONFIG_SMP
-	return (local_t *) (mod->refptr + per_cpu_offset(cpu));
-#else
-	return &mod->ref;
-#endif
-}
-
 /* Sometimes we know we already have a refcount, and it's easier not
    to handle the error case (which only happens with rmmod --wait). */
 static inline void __module_get(struct module *module)
 {
 	if (module) {
-		unsigned int cpu = get_cpu();
-		local_inc(__module_ref_addr(module, cpu));
+		preempt_disable();
+		__this_cpu_inc(module->refptr->count);
 		trace_module_get(module, _THIS_IP_,
-				 local_read(__module_ref_addr(module, cpu)));
-		put_cpu();
+				 __this_cpu_read(module->refptr->count));
+		preempt_enable();
 	}
 }
 
@@ -481,15 +470,17 @@ static inline int try_module_get(struct module *module)
 	int ret = 1;
 
 	if (module) {
-		unsigned int cpu = get_cpu();
+		preempt_disable();
+
 		if (likely(module_is_live(module))) {
-			local_inc(__module_ref_addr(module, cpu));
+			__this_cpu_inc(module->refptr->count);
 			trace_module_get(module, _THIS_IP_,
-				local_read(__module_ref_addr(module, cpu)));
+				__this_cpu_read(module->refptr->count));
 		}
 		else
 			ret = 0;
-		put_cpu();
+
+		preempt_enable();
 	}
 	return ret;
 }
diff --git a/include/linux/mount.h b/include/linux/mount.h
index 5d52753..b5f43a3 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -66,7 +66,7 @@ struct vfsmount {
 	int mnt_pinned;
 	int mnt_ghosts;
 #ifdef CONFIG_SMP
-	int *mnt_writers;
+	int __percpu *mnt_writers;
 #else
 	int mnt_writers;
 #endif
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 34fc6be..6a2e44f 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -105,7 +105,7 @@ struct nfs_server {
 	struct rpc_clnt *	client;		/* RPC client handle */
 	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
 	struct nlm_host		*nlm_host;	/* NLM client handle */
-	struct nfs_iostats *	io_stats;	/* I/O statistics */
+	struct nfs_iostats __percpu *io_stats;	/* I/O statistics */
 	struct backing_dev_info	backing_dev_info;
 	atomic_long_t		writeback;	/* number of writeback pages */
 	int			flags;		/* various flags */
diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h
index 5a5d6ce..68567c0 100644
--- a/include/linux/percpu-defs.h
+++ b/include/linux/percpu-defs.h
@@ -2,12 +2,6 @@
 #define _LINUX_PERCPU_DEFS_H
 
 /*
- * Determine the real variable name from the name visible in the
- * kernel sources.
- */
-#define per_cpu_var(var) per_cpu__##var
-
-/*
  * Base implementations of per-CPU variable declarations and definitions, where
  * the section in which the variable is to be placed is provided by the
  * 'sec' argument.  This may be used to affect the parameters governing the
@@ -18,13 +12,23 @@
  * that section.
  */
 #define __PCPU_ATTRS(sec)						\
-	__attribute__((section(PER_CPU_BASE_SECTION sec)))		\
+	__percpu __attribute__((section(PER_CPU_BASE_SECTION sec)))	\
 	PER_CPU_ATTRIBUTES
 
 #define __PCPU_DUMMY_ATTRS						\
 	__attribute__((section(".discard"), unused))
 
 /*
+ * Macro which verifies @ptr is a percpu pointer without evaluating
+ * @ptr.  This is to be used in percpu accessors to verify that the
+ * input parameter is a percpu pointer.
+ */
+#define __verify_pcpu_ptr(ptr)	do {					\
+	const void __percpu *__vpp_verify = (typeof(ptr))NULL;		\
+	(void)__vpp_verify;						\
+} while (0)
+
+/*
  * s390 and alpha modules require percpu variables to be defined as
  * weak to force the compiler to generate GOT based external
  * references for them.  This is necessary because percpu sections
@@ -56,24 +60,24 @@
  */
 #define DECLARE_PER_CPU_SECTION(type, name, sec)			\
 	extern __PCPU_DUMMY_ATTRS char __pcpu_scope_##name;		\
-	extern __PCPU_ATTRS(sec) __typeof__(type) per_cpu__##name
+	extern __PCPU_ATTRS(sec) __typeof__(type) name
 
 #define DEFINE_PER_CPU_SECTION(type, name, sec)				\
 	__PCPU_DUMMY_ATTRS char __pcpu_scope_##name;			\
 	extern __PCPU_DUMMY_ATTRS char __pcpu_unique_##name;		\
 	__PCPU_DUMMY_ATTRS char __pcpu_unique_##name;			\
 	__PCPU_ATTRS(sec) PER_CPU_DEF_ATTRIBUTES __weak			\
-	__typeof__(type) per_cpu__##name
+	__typeof__(type) name
 #else
 /*
  * Normal declaration and definition macros.
  */
 #define DECLARE_PER_CPU_SECTION(type, name, sec)			\
-	extern __PCPU_ATTRS(sec) __typeof__(type) per_cpu__##name
+	extern __PCPU_ATTRS(sec) __typeof__(type) name
 
 #define DEFINE_PER_CPU_SECTION(type, name, sec)				\
 	__PCPU_ATTRS(sec) PER_CPU_DEF_ATTRIBUTES			\
-	__typeof__(type) per_cpu__##name
+	__typeof__(type) name
 #endif
 
 /*
@@ -135,10 +139,16 @@
 	__aligned(PAGE_SIZE)
 
 /*
- * Intermodule exports for per-CPU variables.
+ * Intermodule exports for per-CPU variables.  sparse forgets about
+ * address space across EXPORT_SYMBOL(), change EXPORT_SYMBOL() to
+ * noop if __CHECKER__.
  */
-#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
-#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
-
+#ifndef __CHECKER__
+#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(var)
+#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(var)
+#else
+#define EXPORT_PER_CPU_SYMBOL(var)
+#define EXPORT_PER_CPU_SYMBOL_GPL(var)
+#endif
 
 #endif /* _LINUX_PERCPU_DEFS_H */
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index cf5efbc..a93e5bf 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -27,10 +27,17 @@
  * we force a syntax error here if it isn't.
  */
 #define get_cpu_var(var) (*({				\
-	extern int simple_identifier_##var(void);	\
 	preempt_disable();				\
 	&__get_cpu_var(var); }))
-#define put_cpu_var(var) preempt_enable()
+
+/*
+ * The weird & is necessary because sparse considers (void)(var) to be
+ * a direct dereference of percpu variable (var).
+ */
+#define put_cpu_var(var) do {				\
+	(void)&(var);					\
+	preempt_enable();				\
+} while (0)
 
 #ifdef CONFIG_SMP
 
@@ -127,9 +134,9 @@ extern int __init pcpu_page_first_chunk(size_t reserved_size,
  */
 #define per_cpu_ptr(ptr, cpu)	SHIFT_PERCPU_PTR((ptr), per_cpu_offset((cpu)))
 
-extern void *__alloc_reserved_percpu(size_t size, size_t align);
-extern void *__alloc_percpu(size_t size, size_t align);
-extern void free_percpu(void *__pdata);
+extern void __percpu *__alloc_reserved_percpu(size_t size, size_t align);
+extern void __percpu *__alloc_percpu(size_t size, size_t align);
+extern void free_percpu(void __percpu *__pdata);
 extern phys_addr_t per_cpu_ptr_to_phys(void *addr);
 
 #ifndef CONFIG_HAVE_SETUP_PER_CPU_AREA
@@ -140,7 +147,7 @@ extern void __init setup_per_cpu_areas(void);
 
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); })
 
-static inline void *__alloc_percpu(size_t size, size_t align)
+static inline void __percpu *__alloc_percpu(size_t size, size_t align)
 {
 	/*
 	 * Can't easily make larger alignment work with kmalloc.  WARN
@@ -151,7 +158,7 @@ static inline void *__alloc_percpu(size_t size, size_t align)
 	return kzalloc(size, GFP_KERNEL);
 }
 
-static inline void free_percpu(void *p)
+static inline void free_percpu(void __percpu *p)
 {
 	kfree(p);
 }
@@ -171,7 +178,7 @@ static inline void *pcpu_lpage_remapped(void *kaddr)
 #endif /* CONFIG_SMP */
 
 #define alloc_percpu(type)	\
-	(typeof(type) *)__alloc_percpu(sizeof(type), __alignof__(type))
+	(typeof(type) __percpu *)__alloc_percpu(sizeof(type), __alignof__(type))
 
 /*
  * Optional methods for optimized non-lvalue per-cpu variable access.
@@ -188,17 +195,19 @@ static inline void *pcpu_lpage_remapped(void *kaddr)
 #ifndef percpu_read
 # define percpu_read(var)						\
   ({									\
-	typeof(per_cpu_var(var)) __tmp_var__;				\
-	__tmp_var__ = get_cpu_var(var);					\
-	put_cpu_var(var);						\
-	__tmp_var__;							\
+	typeof(var) *pr_ptr__ = &(var);					\
+	typeof(var) pr_ret__;						\
+	pr_ret__ = get_cpu_var(*pr_ptr__);				\
+	put_cpu_var(*pr_ptr__);						\
+	pr_ret__;							\
   })
 #endif
 
 #define __percpu_generic_to_op(var, val, op)				\
 do {									\
-	get_cpu_var(var) op val;					\
-	put_cpu_var(var);						\
+	typeof(var) *pgto_ptr__ = &(var);				\
+	get_cpu_var(*pgto_ptr__) op val;				\
+	put_cpu_var(*pgto_ptr__);					\
 } while (0)
 
 #ifndef percpu_write
@@ -234,6 +243,7 @@ extern void __bad_size_call_parameter(void);
 
 #define __pcpu_size_call_return(stem, variable)				\
 ({	typeof(variable) pscr_ret__;					\
+	__verify_pcpu_ptr(&(variable));					\
 	switch(sizeof(variable)) {					\
 	case 1: pscr_ret__ = stem##1(variable);break;			\
 	case 2: pscr_ret__ = stem##2(variable);break;			\
@@ -247,6 +257,7 @@ extern void __bad_size_call_parameter(void);
 
 #define __pcpu_size_call(stem, variable, ...)				\
 do {									\
+	__verify_pcpu_ptr(&(variable));					\
 	switch(sizeof(variable)) {					\
 		case 1: stem##1(variable, __VA_ARGS__);break;		\
 		case 2: stem##2(variable, __VA_ARGS__);break;		\
@@ -259,8 +270,7 @@ do {									\
 
 /*
  * Optimized manipulation for memory allocated through the per cpu
- * allocator or for addresses of per cpu variables (can be determined
- * using per_cpu_var(xx).
+ * allocator or for addresses of per cpu variables.
  *
  * These operation guarantee exclusivity of access for other operations
  * on the *same* processor. The assumption is that per cpu data is only
@@ -311,7 +321,7 @@ do {									\
 #define _this_cpu_generic_to_op(pcp, val, op)				\
 do {									\
 	preempt_disable();						\
-	*__this_cpu_ptr(&pcp) op val;					\
+	*__this_cpu_ptr(&(pcp)) op val;					\
 	preempt_enable();						\
 } while (0)
 
diff --git a/include/linux/percpu_counter.h b/include/linux/percpu_counter.h
index 794662b..c88d67b 100644
--- a/include/linux/percpu_counter.h
+++ b/include/linux/percpu_counter.h
@@ -21,7 +21,7 @@ struct percpu_counter {
 #ifdef CONFIG_HOTPLUG_CPU
 	struct list_head list;	/* All percpu_counters are on a list */
 #endif
-	s32 *counters;
+	s32 __percpu *counters;
 };
 
 extern int percpu_counter_batch;
diff --git a/include/linux/srcu.h b/include/linux/srcu.h
index 3084f80..4d5ecb2 100644
--- a/include/linux/srcu.h
+++ b/include/linux/srcu.h
@@ -33,7 +33,7 @@ struct srcu_struct_array {
 
 struct srcu_struct {
 	int completed;
-	struct srcu_struct_array *per_cpu_ref;
+	struct srcu_struct_array __percpu *per_cpu_ref;
 	struct mutex mutex;
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 	struct lockdep_map dep_map;
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index ee03bba..117f0dd 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -78,22 +78,22 @@ DECLARE_PER_CPU(struct vm_event_state, vm_event_states);
 
 static inline void __count_vm_event(enum vm_event_item item)
 {
-	__this_cpu_inc(per_cpu_var(vm_event_states).event[item]);
+	__this_cpu_inc(vm_event_states.event[item]);
 }
 
 static inline void count_vm_event(enum vm_event_item item)
 {
-	this_cpu_inc(per_cpu_var(vm_event_states).event[item]);
+	this_cpu_inc(vm_event_states.event[item]);
 }
 
 static inline void __count_vm_events(enum vm_event_item item, long delta)
 {
-	__this_cpu_add(per_cpu_var(vm_event_states).event[item], delta);
+	__this_cpu_add(vm_event_states.event[item], delta);
 }
 
 static inline void count_vm_events(enum vm_event_item item, long delta)
 {
-	this_cpu_add(per_cpu_var(vm_event_states).event[item], delta);
+	this_cpu_add(vm_event_states.event[item], delta);
 }
 
 extern void all_vm_events(unsigned long *);
diff --git a/kernel/kexec.c b/kernel/kexec.c
index ef077fb..87ebe8a 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -41,7 +41,7 @@
 #include <asm/sections.h>
 
 /* Per cpu memory for storing cpu states in case of system crash. */
-note_buf_t* crash_notes;
+note_buf_t __percpu *crash_notes;
 
 /* vmcoreinfo stuff */
 static unsigned char vmcoreinfo_data[VMCOREINFO_BYTES];
diff --git a/kernel/module.c b/kernel/module.c
index f82386b..e5538d5 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -474,9 +474,10 @@ static void module_unload_init(struct module *mod)
 
 	INIT_LIST_HEAD(&mod->modules_which_use_me);
 	for_each_possible_cpu(cpu)
-		local_set(__module_ref_addr(mod, cpu), 0);
+		per_cpu_ptr(mod->refptr, cpu)->count = 0;
+
 	/* Hold reference count during initialization. */
-	local_set(__module_ref_addr(mod, raw_smp_processor_id()), 1);
+	__this_cpu_write(mod->refptr->count, 1);
 	/* Backwards compatibility macros put refcount during init. */
 	mod->waiter = current;
 }
@@ -619,7 +620,7 @@ unsigned int module_refcount(struct module *mod)
 	int cpu;
 
 	for_each_possible_cpu(cpu)
-		total += local_read(__module_ref_addr(mod, cpu));
+		total += per_cpu_ptr(mod->refptr, cpu)->count;
 	return total;
 }
 EXPORT_SYMBOL(module_refcount);
@@ -796,14 +797,15 @@ static struct module_attribute refcnt = {
 void module_put(struct module *module)
 {
 	if (module) {
-		unsigned int cpu = get_cpu();
-		local_dec(__module_ref_addr(module, cpu));
+		preempt_disable();
+		__this_cpu_dec(module->refptr->count);
+
 		trace_module_put(module, _RET_IP_,
-				 local_read(__module_ref_addr(module, cpu)));
+				 __this_cpu_read(module->refptr->count));
 		/* Maybe they're waiting for us to drop reference? */
 		if (unlikely(!module_is_live(module)))
 			wake_up_process(module->waiter);
-		put_cpu();
+		preempt_enable();
 	}
 }
 EXPORT_SYMBOL(module_put);
@@ -1397,9 +1399,9 @@ static void free_module(struct module *mod)
 	kfree(mod->args);
 	if (mod->percpu)
 		percpu_modfree(mod->percpu);
-#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP)
+#if defined(CONFIG_MODULE_UNLOAD)
 	if (mod->refptr)
-		percpu_modfree(mod->refptr);
+		free_percpu(mod->refptr);
 #endif
 	/* Free lock-classes: */
 	lockdep_free_key_range(mod->module_core, mod->core_size);
@@ -2162,9 +2164,8 @@ static noinline struct module *load_module(void __user *umod,
 	mod = (void *)sechdrs[modindex].sh_addr;
 	kmemleak_load_module(mod, hdr, sechdrs, secstrings);
 
-#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP)
-	mod->refptr = percpu_modalloc(sizeof(local_t), __alignof__(local_t),
-				      mod->name);
+#if defined(CONFIG_MODULE_UNLOAD)
+	mod->refptr = alloc_percpu(struct module_ref);
 	if (!mod->refptr) {
 		err = -ENOMEM;
 		goto free_init;
@@ -2396,8 +2397,8 @@ static noinline struct module *load_module(void __user *umod,
 	kobject_put(&mod->mkobj.kobj);
  free_unload:
 	module_unload_free(mod);
-#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP)
-	percpu_modfree(mod->refptr);
+#if defined(CONFIG_MODULE_UNLOAD)
+	free_percpu(mod->refptr);
  free_init:
 #endif
 	module_free(mod, mod->module_init);
diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index 258cdf0..58df55b 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -818,13 +818,13 @@ static void rcu_torture_timer(unsigned long unused)
 		/* Should not happen, but... */
 		pipe_count = RCU_TORTURE_PIPE_LEN;
 	}
-	__this_cpu_inc(per_cpu_var(rcu_torture_count)[pipe_count]);
+	__this_cpu_inc(rcu_torture_count[pipe_count]);
 	completed = cur_ops->completed() - completed;
 	if (completed > RCU_TORTURE_PIPE_LEN) {
 		/* Should not happen, but... */
 		completed = RCU_TORTURE_PIPE_LEN;
 	}
-	__this_cpu_inc(per_cpu_var(rcu_torture_batch)[completed]);
+	__this_cpu_inc(rcu_torture_batch[completed]);
 	preempt_enable();
 	cur_ops->readunlock(idx);
 }
@@ -877,13 +877,13 @@ rcu_torture_reader(void *arg)
 			/* Should not happen, but... */
 			pipe_count = RCU_TORTURE_PIPE_LEN;
 		}
-		__this_cpu_inc(per_cpu_var(rcu_torture_count)[pipe_count]);
+		__this_cpu_inc(rcu_torture_count[pipe_count]);
 		completed = cur_ops->completed() - completed;
 		if (completed > RCU_TORTURE_PIPE_LEN) {
 			/* Should not happen, but... */
 			completed = RCU_TORTURE_PIPE_LEN;
 		}
-		__this_cpu_inc(per_cpu_var(rcu_torture_batch)[completed]);
+		__this_cpu_inc(rcu_torture_batch[completed]);
 		preempt_enable();
 		cur_ops->readunlock(idx);
 		schedule();
diff --git a/kernel/sched.c b/kernel/sched.c
index 6a212c9..abb36b1 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1521,7 +1521,7 @@ static unsigned long cpu_avg_load_per_task(int cpu)
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 
-static __read_mostly unsigned long *update_shares_data;
+static __read_mostly unsigned long __percpu *update_shares_data;
 
 static void __set_se_shares(struct sched_entity *se, unsigned long shares);
 
@@ -8813,7 +8813,7 @@ struct cgroup_subsys cpu_cgroup_subsys = {
 struct cpuacct {
 	struct cgroup_subsys_state css;
 	/* cpuusage holds pointer to a u64-type object on every cpu */
-	u64 *cpuusage;
+	u64 __percpu *cpuusage;
 	struct percpu_counter cpustat[CPUACCT_STAT_NSTATS];
 	struct cpuacct *parent;
 };
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index 912823e..9bb9fb1 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -45,7 +45,7 @@ static int refcount;
 static struct workqueue_struct *stop_machine_wq;
 static struct stop_machine_data active, idle;
 static const struct cpumask *active_cpus;
-static void *stop_machine_work;
+static void __percpu *stop_machine_work;
 
 static void set_state(enum stopmachine_state newstate)
 {
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 8c1b2d2..0287f9f 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -20,6 +20,7 @@
 #include <linux/cpu.h>
 #include <linux/fs.h>
 
+#include <asm/local.h>
 #include "trace.h"
 
 /*
diff --git a/kernel/trace/ring_buffer_benchmark.c b/kernel/trace/ring_buffer_benchmark.c
index b2477ca..df74c79 100644
--- a/kernel/trace/ring_buffer_benchmark.c
+++ b/kernel/trace/ring_buffer_benchmark.c
@@ -8,6 +8,7 @@
 #include <linux/kthread.h>
 #include <linux/module.h>
 #include <linux/time.h>
+#include <asm/local.h>
 
 struct rb_page {
 	u64		ts;
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 032c57c..ed01fdb 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -92,12 +92,12 @@ DEFINE_PER_CPU(int, ftrace_cpu_disabled);
 static inline void ftrace_disable_cpu(void)
 {
 	preempt_disable();
-	__this_cpu_inc(per_cpu_var(ftrace_cpu_disabled));
+	__this_cpu_inc(ftrace_cpu_disabled);
 }
 
 static inline void ftrace_enable_cpu(void)
 {
-	__this_cpu_dec(per_cpu_var(ftrace_cpu_disabled));
+	__this_cpu_dec(ftrace_cpu_disabled);
 	preempt_enable();
 }
 
@@ -1166,7 +1166,7 @@ trace_function(struct trace_array *tr,
 	struct ftrace_entry *entry;
 
 	/* If we are reading the ring buffer, don't trace */
-	if (unlikely(__this_cpu_read(per_cpu_var(ftrace_cpu_disabled))))
+	if (unlikely(__this_cpu_read(ftrace_cpu_disabled)))
 		return;
 
 	event = trace_buffer_lock_reserve(buffer, TRACE_FN, sizeof(*entry),
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index e998a82..3fc2a57 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -188,7 +188,7 @@ static int __trace_graph_entry(struct trace_array *tr,
 	struct ring_buffer *buffer = tr->buffer;
 	struct ftrace_graph_ent_entry *entry;
 
-	if (unlikely(__this_cpu_read(per_cpu_var(ftrace_cpu_disabled))))
+	if (unlikely(__this_cpu_read(ftrace_cpu_disabled)))
 		return 0;
 
 	event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_ENT,
@@ -247,7 +247,7 @@ static void __trace_graph_return(struct trace_array *tr,
 	struct ring_buffer *buffer = tr->buffer;
 	struct ftrace_graph_ret_entry *entry;
 
-	if (unlikely(__this_cpu_read(per_cpu_var(ftrace_cpu_disabled))))
+	if (unlikely(__this_cpu_read(ftrace_cpu_disabled)))
 		return;
 
 	event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_RET,
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 8deb9d0..9a7aaae 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1009,10 +1009,10 @@ static void drain_pages(unsigned int cpu)
 		struct per_cpu_pageset *pset;
 		struct per_cpu_pages *pcp;
 
-		pset = zone_pcp(zone, cpu);
+		local_irq_save(flags);
+		pset = per_cpu_ptr(zone->pageset, cpu);
 
 		pcp = &pset->pcp;
-		local_irq_save(flags);
 		free_pcppages_bulk(zone, pcp->count, pcp);
 		pcp->count = 0;
 		local_irq_restore(flags);
@@ -1096,7 +1096,6 @@ static void free_hot_cold_page(struct page *page, int cold)
 	arch_free_page(page, 0);
 	kernel_map_pages(page, 1, 0);
 
-	pcp = &zone_pcp(zone, get_cpu())->pcp;
 	migratetype = get_pageblock_migratetype(page);
 	set_page_private(page, migratetype);
 	local_irq_save(flags);
@@ -1119,6 +1118,7 @@ static void free_hot_cold_page(struct page *page, int cold)
 		migratetype = MIGRATE_MOVABLE;
 	}
 
+	pcp = &this_cpu_ptr(zone->pageset)->pcp;
 	if (cold)
 		list_add_tail(&page->lru, &pcp->lists[migratetype]);
 	else
@@ -1131,7 +1131,6 @@ static void free_hot_cold_page(struct page *page, int cold)
 
 out:
 	local_irq_restore(flags);
-	put_cpu();
 }
 
 void free_hot_page(struct page *page)
@@ -1181,17 +1180,15 @@ struct page *buffered_rmqueue(struct zone *preferred_zone,
 	unsigned long flags;
 	struct page *page;
 	int cold = !!(gfp_flags & __GFP_COLD);
-	int cpu;
 
 again:
-	cpu  = get_cpu();
 	if (likely(order == 0)) {
 		struct per_cpu_pages *pcp;
 		struct list_head *list;
 
-		pcp = &zone_pcp(zone, cpu)->pcp;
-		list = &pcp->lists[migratetype];
 		local_irq_save(flags);
+		pcp = &this_cpu_ptr(zone->pageset)->pcp;
+		list = &pcp->lists[migratetype];
 		if (list_empty(list)) {
 			pcp->count += rmqueue_bulk(zone, 0,
 					pcp->batch, list,
@@ -1232,7 +1229,6 @@ again:
 	__count_zone_vm_events(PGALLOC, zone, 1 << order);
 	zone_statistics(preferred_zone, zone);
 	local_irq_restore(flags);
-	put_cpu();
 
 	VM_BUG_ON(bad_range(zone, page));
 	if (prep_new_page(page, order, gfp_flags))
@@ -1241,7 +1237,6 @@ again:
 
 failed:
 	local_irq_restore(flags);
-	put_cpu();
 	return NULL;
 }
 
@@ -2180,7 +2175,7 @@ void show_free_areas(void)
 		for_each_online_cpu(cpu) {
 			struct per_cpu_pageset *pageset;
 
-			pageset = zone_pcp(zone, cpu);
+			pageset = per_cpu_ptr(zone->pageset, cpu);
 
 			printk("CPU %4d: hi:%5d, btch:%4d usd:%4d\n",
 			       cpu, pageset->pcp.high,
@@ -2745,10 +2740,29 @@ static void build_zonelist_cache(pg_data_t *pgdat)
 
 #endif	/* CONFIG_NUMA */
 
+/*
+ * Boot pageset table. One per cpu which is going to be used for all
+ * zones and all nodes. The parameters will be set in such a way
+ * that an item put on a list will immediately be handed over to
+ * the buddy list. This is safe since pageset manipulation is done
+ * with interrupts disabled.
+ *
+ * The boot_pagesets must be kept even after bootup is complete for
+ * unused processors and/or zones. They do play a role for bootstrapping
+ * hotplugged processors.
+ *
+ * zoneinfo_show() and maybe other functions do
+ * not check if the processor is online before following the pageset pointer.
+ * Other parts of the kernel may not check if the zone is available.
+ */
+static void setup_pageset(struct per_cpu_pageset *p, unsigned long batch);
+static DEFINE_PER_CPU(struct per_cpu_pageset, boot_pageset);
+
 /* return values int ....just for stop_machine() */
 static int __build_all_zonelists(void *dummy)
 {
 	int nid;
+	int cpu;
 
 #ifdef CONFIG_NUMA
 	memset(node_load, 0, sizeof(node_load));
@@ -2759,6 +2773,23 @@ static int __build_all_zonelists(void *dummy)
 		build_zonelists(pgdat);
 		build_zonelist_cache(pgdat);
 	}
+
+	/*
+	 * Initialize the boot_pagesets that are going to be used
+	 * for bootstrapping processors. The real pagesets for
+	 * each zone will be allocated later when the per cpu
+	 * allocator is available.
+	 *
+	 * boot_pagesets are used also for bootstrapping offline
+	 * cpus if the system is already booted because the pagesets
+	 * are needed to initialize allocators on a specific cpu too.
+	 * F.e. the percpu allocator needs the page allocator which
+	 * needs the percpu allocator in order to allocate its pagesets
+	 * (a chicken-egg dilemma).
+	 */
+	for_each_possible_cpu(cpu)
+		setup_pageset(&per_cpu(boot_pageset, cpu), 0);
+
 	return 0;
 }
 
@@ -3096,121 +3127,33 @@ static void setup_pagelist_highmark(struct per_cpu_pageset *p,
 		pcp->batch = PAGE_SHIFT * 8;
 }
 
-
-#ifdef CONFIG_NUMA
-/*
- * Boot pageset table. One per cpu which is going to be used for all
- * zones and all nodes. The parameters will be set in such a way
- * that an item put on a list will immediately be handed over to
- * the buddy list. This is safe since pageset manipulation is done
- * with interrupts disabled.
- *
- * Some NUMA counter updates may also be caught by the boot pagesets.
- *
- * The boot_pagesets must be kept even after bootup is complete for
- * unused processors and/or zones. They do play a role for bootstrapping
- * hotplugged processors.
- *
- * zoneinfo_show() and maybe other functions do
- * not check if the processor is online before following the pageset pointer.
- * Other parts of the kernel may not check if the zone is available.
- */
-static struct per_cpu_pageset boot_pageset[NR_CPUS];
-
 /*
- * Dynamically allocate memory for the
- * per cpu pageset array in struct zone.
+ * Allocate per cpu pagesets and initialize them.
+ * Before this call only boot pagesets were available.
+ * Boot pagesets will no longer be used by this processorr
+ * after setup_per_cpu_pageset().
  */
-static int __cpuinit process_zones(int cpu)
+void __init setup_per_cpu_pageset(void)
 {
-	struct zone *zone, *dzone;
-	int node = cpu_to_node(cpu);
-
-	node_set_state(node, N_CPU);	/* this node has a cpu */
+	struct zone *zone;
+	int cpu;
 
 	for_each_populated_zone(zone) {
-		zone_pcp(zone, cpu) = kmalloc_node(sizeof(struct per_cpu_pageset),
-					 GFP_KERNEL, node);
-		if (!zone_pcp(zone, cpu))
-			goto bad;
-
-		setup_pageset(zone_pcp(zone, cpu), zone_batchsize(zone));
-
-		if (percpu_pagelist_fraction)
-			setup_pagelist_highmark(zone_pcp(zone, cpu),
-			    (zone->present_pages / percpu_pagelist_fraction));
-	}
-
-	return 0;
-bad:
-	for_each_zone(dzone) {
-		if (!populated_zone(dzone))
-			continue;
-		if (dzone == zone)
-			break;
-		kfree(zone_pcp(dzone, cpu));
-		zone_pcp(dzone, cpu) = &boot_pageset[cpu];
-	}
-	return -ENOMEM;
-}
+		zone->pageset = alloc_percpu(struct per_cpu_pageset);
 
-static inline void free_zone_pagesets(int cpu)
-{
-	struct zone *zone;
-
-	for_each_zone(zone) {
-		struct per_cpu_pageset *pset = zone_pcp(zone, cpu);
+		for_each_possible_cpu(cpu) {
+			struct per_cpu_pageset *pcp = per_cpu_ptr(zone->pageset, cpu);
 
-		/* Free per_cpu_pageset if it is slab allocated */
-		if (pset != &boot_pageset[cpu])
-			kfree(pset);
-		zone_pcp(zone, cpu) = &boot_pageset[cpu];
-	}
-}
+			setup_pageset(pcp, zone_batchsize(zone));
 
-static int __cpuinit pageset_cpuup_callback(struct notifier_block *nfb,
-		unsigned long action,
-		void *hcpu)
-{
-	int cpu = (long)hcpu;
-	int ret = NOTIFY_OK;
-
-	switch (action) {
-	case CPU_UP_PREPARE:
-	case CPU_UP_PREPARE_FROZEN:
-		if (process_zones(cpu))
-			ret = NOTIFY_BAD;
-		break;
-	case CPU_UP_CANCELED:
-	case CPU_UP_CANCELED_FROZEN:
-	case CPU_DEAD:
-	case CPU_DEAD_FROZEN:
-		free_zone_pagesets(cpu);
-		break;
-	default:
-		break;
+			if (percpu_pagelist_fraction)
+				setup_pagelist_highmark(pcp,
+					(zone->present_pages /
+						percpu_pagelist_fraction));
+		}
 	}
-	return ret;
 }
 
-static struct notifier_block __cpuinitdata pageset_notifier =
-	{ &pageset_cpuup_callback, NULL, 0 };
-
-void __init setup_per_cpu_pageset(void)
-{
-	int err;
-
-	/* Initialize per_cpu_pageset for cpu 0.
-	 * A cpuup callback will do this for every cpu
-	 * as it comes online
-	 */
-	err = process_zones(smp_processor_id());
-	BUG_ON(err);
-	register_cpu_notifier(&pageset_notifier);
-}
-
-#endif
-
 static noinline __init_refok
 int zone_wait_table_init(struct zone *zone, unsigned long zone_size_pages)
 {
@@ -3264,7 +3207,7 @@ static int __zone_pcp_update(void *data)
 		struct per_cpu_pageset *pset;
 		struct per_cpu_pages *pcp;
 
-		pset = zone_pcp(zone, cpu);
+		pset = per_cpu_ptr(zone->pageset, cpu);
 		pcp = &pset->pcp;
 
 		local_irq_save(flags);
@@ -3282,21 +3225,17 @@ void zone_pcp_update(struct zone *zone)
 
 static __meminit void zone_pcp_init(struct zone *zone)
 {
-	int cpu;
-	unsigned long batch = zone_batchsize(zone);
+	/*
+	 * per cpu subsystem is not up at this point. The following code
+	 * relies on the ability of the linker to provide the
+	 * offset of a (static) per cpu variable into the per cpu area.
+	 */
+	zone->pageset = &boot_pageset;
 
-	for (cpu = 0; cpu < NR_CPUS; cpu++) {
-#ifdef CONFIG_NUMA
-		/* Early boot. Slab allocator not functional yet */
-		zone_pcp(zone, cpu) = &boot_pageset[cpu];
-		setup_pageset(&boot_pageset[cpu],0);
-#else
-		setup_pageset(zone_pcp(zone,cpu), batch);
-#endif
-	}
 	if (zone->present_pages)
-		printk(KERN_DEBUG "  %s zone: %lu pages, LIFO batch:%lu\n",
-			zone->name, zone->present_pages, batch);
+		printk(KERN_DEBUG "  %s zone: %lu pages, LIFO batch:%u\n",
+			zone->name, zone->present_pages,
+					 zone_batchsize(zone));
 }
 
 __meminit int init_currently_empty_zone(struct zone *zone,
@@ -4810,10 +4749,11 @@ int percpu_pagelist_fraction_sysctl_handler(ctl_table *table, int write,
 	if (!write || (ret == -EINVAL))
 		return ret;
 	for_each_populated_zone(zone) {
-		for_each_online_cpu(cpu) {
+		for_each_possible_cpu(cpu) {
 			unsigned long  high;
 			high = zone->present_pages / percpu_pagelist_fraction;
-			setup_pagelist_highmark(zone_pcp(zone, cpu), high);
+			setup_pagelist_highmark(
+				per_cpu_ptr(zone->pageset, cpu), high);
 		}
 	}
 	return 0;
diff --git a/mm/percpu.c b/mm/percpu.c
index 083e7c9..768419d 100644
--- a/mm/percpu.c
+++ b/mm/percpu.c
@@ -80,13 +80,15 @@
 /* default addr <-> pcpu_ptr mapping, override in asm/percpu.h if necessary */
 #ifndef __addr_to_pcpu_ptr
 #define __addr_to_pcpu_ptr(addr)					\
-	(void *)((unsigned long)(addr) - (unsigned long)pcpu_base_addr	\
-		 + (unsigned long)__per_cpu_start)
+	(void __percpu *)((unsigned long)(addr) -			\
+			  (unsigned long)pcpu_base_addr	+		\
+			  (unsigned long)__per_cpu_start)
 #endif
 #ifndef __pcpu_ptr_to_addr
 #define __pcpu_ptr_to_addr(ptr)						\
-	(void *)((unsigned long)(ptr) + (unsigned long)pcpu_base_addr	\
-		 - (unsigned long)__per_cpu_start)
+	(void __force *)((unsigned long)(ptr) +				\
+			 (unsigned long)pcpu_base_addr -		\
+			 (unsigned long)__per_cpu_start)
 #endif
 
 struct pcpu_chunk {
@@ -913,11 +915,10 @@ static void pcpu_depopulate_chunk(struct pcpu_chunk *chunk, int off, int size)
 	int rs, re;
 
 	/* quick path, check whether it's empty already */
-	pcpu_for_each_unpop_region(chunk, rs, re, page_start, page_end) {
-		if (rs == page_start && re == page_end)
-			return;
-		break;
-	}
+	rs = page_start;
+	pcpu_next_unpop(chunk, &rs, &re, page_end);
+	if (rs == page_start && re == page_end)
+		return;
 
 	/* immutable chunks can't be depopulated */
 	WARN_ON(chunk->immutable);
@@ -968,11 +969,10 @@ static int pcpu_populate_chunk(struct pcpu_chunk *chunk, int off, int size)
 	int rs, re, rc;
 
 	/* quick path, check whether all pages are already there */
-	pcpu_for_each_pop_region(chunk, rs, re, page_start, page_end) {
-		if (rs == page_start && re == page_end)
-			goto clear;
-		break;
-	}
+	rs = page_start;
+	pcpu_next_pop(chunk, &rs, &re, page_end);
+	if (rs == page_start && re == page_end)
+		goto clear;
 
 	/* need to allocate and map pages, this chunk can't be immutable */
 	WARN_ON(chunk->immutable);
@@ -1067,7 +1067,7 @@ static struct pcpu_chunk *alloc_pcpu_chunk(void)
  * RETURNS:
  * Percpu pointer to the allocated area on success, NULL on failure.
  */
-static void *pcpu_alloc(size_t size, size_t align, bool reserved)
+static void __percpu *pcpu_alloc(size_t size, size_t align, bool reserved)
 {
 	static int warn_limit = 10;
 	struct pcpu_chunk *chunk;
@@ -1196,7 +1196,7 @@ fail_unlock_mutex:
  * RETURNS:
  * Percpu pointer to the allocated area on success, NULL on failure.
  */
-void *__alloc_percpu(size_t size, size_t align)
+void __percpu *__alloc_percpu(size_t size, size_t align)
 {
 	return pcpu_alloc(size, align, false);
 }
@@ -1217,7 +1217,7 @@ EXPORT_SYMBOL_GPL(__alloc_percpu);
  * RETURNS:
  * Percpu pointer to the allocated area on success, NULL on failure.
  */
-void *__alloc_reserved_percpu(size_t size, size_t align)
+void __percpu *__alloc_reserved_percpu(size_t size, size_t align)
 {
 	return pcpu_alloc(size, align, true);
 }
@@ -1269,7 +1269,7 @@ static void pcpu_reclaim(struct work_struct *work)
  * CONTEXT:
  * Can be called from atomic context.
  */
-void free_percpu(void *ptr)
+void free_percpu(void __percpu *ptr)
 {
 	void *addr;
 	struct pcpu_chunk *chunk;
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 6051fba..fc5aa18 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -139,7 +139,8 @@ static void refresh_zone_stat_thresholds(void)
 		threshold = calculate_threshold(zone);
 
 		for_each_online_cpu(cpu)
-			zone_pcp(zone, cpu)->stat_threshold = threshold;
+			per_cpu_ptr(zone->pageset, cpu)->stat_threshold
+							= threshold;
 	}
 }
 
@@ -149,7 +150,8 @@ static void refresh_zone_stat_thresholds(void)
 void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
 				int delta)
 {
-	struct per_cpu_pageset *pcp = zone_pcp(zone, smp_processor_id());
+	struct per_cpu_pageset *pcp = this_cpu_ptr(zone->pageset);
+
 	s8 *p = pcp->vm_stat_diff + item;
 	long x;
 
@@ -202,7 +204,7 @@ EXPORT_SYMBOL(mod_zone_page_state);
  */
 void __inc_zone_state(struct zone *zone, enum zone_stat_item item)
 {
-	struct per_cpu_pageset *pcp = zone_pcp(zone, smp_processor_id());
+	struct per_cpu_pageset *pcp = this_cpu_ptr(zone->pageset);
 	s8 *p = pcp->vm_stat_diff + item;
 
 	(*p)++;
@@ -223,7 +225,7 @@ EXPORT_SYMBOL(__inc_zone_page_state);
 
 void __dec_zone_state(struct zone *zone, enum zone_stat_item item)
 {
-	struct per_cpu_pageset *pcp = zone_pcp(zone, smp_processor_id());
+	struct per_cpu_pageset *pcp = this_cpu_ptr(zone->pageset);
 	s8 *p = pcp->vm_stat_diff + item;
 
 	(*p)--;
@@ -300,7 +302,7 @@ void refresh_cpu_vm_stats(int cpu)
 	for_each_populated_zone(zone) {
 		struct per_cpu_pageset *p;
 
-		p = zone_pcp(zone, cpu);
+		p = per_cpu_ptr(zone->pageset, cpu);
 
 		for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
 			if (p->vm_stat_diff[i]) {
@@ -741,7 +743,7 @@ static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat,
 	for_each_online_cpu(i) {
 		struct per_cpu_pageset *pageset;
 
-		pageset = zone_pcp(zone, i);
+		pageset = per_cpu_ptr(zone->pageset, i);
 		seq_printf(m,
 			   "\n    cpu: %i"
 			   "\n              count: %i"
@@ -906,6 +908,7 @@ static int __cpuinit vmstat_cpuup_callback(struct notifier_block *nfb,
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
 		start_cpu_timer(cpu);
+		node_set_state(cpu_to_node(cpu), N_CPU);
 		break;
 	case CPU_DOWN_PREPARE:
 	case CPU_DOWN_PREPARE_FROZEN: