[this_cpu_xx V9 0/7] Per cpu atomics in page allocator, modules, cleanup and optimized inc/dec

[this_cpu_xx V9 0/7] Per cpu atomics in page allocator, modules, cleanup and optimized inc/dec

[Christoph Lameter]

Leftovers from the earlier patchset rediffed to 2.6.33-rc2.
Mostly applications of per cpu counters to core components and cleanup patches.

After this patchset there will be only one user of local_t left: Mathieu's
trace ringbuffer.

V8->V9
- Split patches for the module logic differently.
- Fix spurious free in module patch.
- Drop ringbuffer stuff. Needs more discussion.
- Drop slub patches (already in Pekka's tree and in -next)
- Add SPARC cleanup patch.

V7->V8
- Fix issue in slub patch
- Fix issue in modules patch
- Rediff page allocator patch
- Provide new this_cpu ops needed for ringbuffer [RFC state]

V6->V7
- Drop patches merged in 2.6.33 merge cycle
- Drop risky slub patches

V5->V6:
- Drop patches merged by Tejun.
- Drop irqless slub fastpath for now.
- Patches against Tejun percpu for-next branch.

V4->V5:
- Avoid setup_per_cpu_area() modifications and fold the remainder of the
  patch into the page allocator patch.
- Irq disable / per cpu ptr fixes for page allocator patch.

V3->V4:
- Fix various macro definitions.
- Provide experimental percpu based fastpath that does not disable
  interrupts for SLUB.

V2->V3:
- Available via git tree against latest upstream from
	 git://git.kernel.org/pub/scm/linux/kernel/git/christoph/percpu.git linus
- Rework SLUB per cpu operations. Get rid of dynamic DMA slab creation
  for CONFIG_ZONE_DMA
- Create fallback framework so that 64 bit ops on 32 bit platforms
  can fallback to the use of preempt or interrupt disable. 64 bit
  platforms can use 64 bit atomic per cpu ops.

V1->V2:
- Various minor fixes
- Add SLUB conversion
- Add Page allocator conversion
- Patch against the git tree of today

[this_cpu_xx V9 1/7] Remove cpu_local_xx macros

[Christoph Lameter]

[-- Attachment #1: percpu_local_t_remove_cpu_alloc --]
[-- Type: text/plain, Size: 9642 bytes --]

These macros have not been used for awhile now.

Signed-off-by: Christoph Lameter <cl@linux-foundation.org>

---
 arch/alpha/include/asm/local.h   |   17 -----------------
 arch/m32r/include/asm/local.h    |   25 -------------------------
 arch/mips/include/asm/local.h    |   25 -------------------------
 arch/powerpc/include/asm/local.h |   25 -------------------------
 arch/x86/include/asm/local.h     |   37 -------------------------------------
 include/asm-generic/local.h      |   19 -------------------
 6 files changed, 148 deletions(-)

Index: linux-2.6/arch/alpha/include/asm/local.h
===================================================================
--- linux-2.6.orig/arch/alpha/include/asm/local.h	2009-12-16 13:50:11.000000000 -0600
+++ linux-2.6/arch/alpha/include/asm/local.h	2009-12-18 14:19:17.000000000 -0600
@@ -98,21 +98,4 @@ static __inline__ long local_sub_return(
 #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 */
Index: linux-2.6/arch/m32r/include/asm/local.h
===================================================================
--- linux-2.6.orig/arch/m32r/include/asm/local.h	2009-12-16 13:50:11.000000000 -0600
+++ linux-2.6/arch/m32r/include/asm/local.h	2009-12-18 14:19:17.000000000 -0600
@@ -338,29 +338,4 @@ static inline void local_set_mask(unsign
  * 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 */
Index: linux-2.6/arch/mips/include/asm/local.h
===================================================================
--- linux-2.6.orig/arch/mips/include/asm/local.h	2009-12-16 13:50:11.000000000 -0600
+++ linux-2.6/arch/mips/include/asm/local.h	2009-12-18 14:19:17.000000000 -0600
@@ -193,29 +193,4 @@ static __inline__ long local_sub_return(
 #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 */
Index: linux-2.6/arch/powerpc/include/asm/local.h
===================================================================
--- linux-2.6.orig/arch/powerpc/include/asm/local.h	2009-12-16 13:50:11.000000000 -0600
+++ linux-2.6/arch/powerpc/include/asm/local.h	2009-12-18 14:19:17.000000000 -0600
@@ -172,29 +172,4 @@ static __inline__ long local_dec_if_posi
 #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 */
Index: linux-2.6/arch/x86/include/asm/local.h
===================================================================
--- linux-2.6.orig/arch/x86/include/asm/local.h	2009-12-16 13:50:11.000000000 -0600
+++ linux-2.6/arch/x86/include/asm/local.h	2009-12-18 14:19:17.000000000 -0600
@@ -195,41 +195,4 @@ static inline long local_sub_return(long
 #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 */
Index: linux-2.6/include/asm-generic/local.h
===================================================================
--- linux-2.6.orig/include/asm-generic/local.h	2009-12-16 13:50:11.000000000 -0600
+++ linux-2.6/include/asm-generic/local.h	2009-12-18 14:19:17.000000000 -0600
@@ -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 */

-- 

[this_cpu_xx V9 2/7] Module handling: Use this_cpu_xx to dynamically allocate counters

[Christoph Lameter]

[-- Attachment #1: percpu_local_t_remove_modules --]
[-- Type: text/plain, Size: 5629 bytes --]

Use cpu ops to deal with the per cpu data instead of a local_t. Reduces memory
requirements, cache footprint and decreases cycle counts.

The this_cpu_xx operations are also used for !SMP mode. Otherwise we could
not drop the use of __module_ref_addr() which would make per cpu data handling
complicated. this_cpu_xx operations have their own fallback for !SMP.

V8-V9:
- Leave include asm/module.h since ringbuffer.c depends on it. Nothing else
  does though. Another patch will deal with that.
- Remove spurious free.

Signed-off-by: Christoph Lameter <cl@linux-foundation.org>

---
 include/linux/module.h |   36 ++++++++++++++----------------------
 kernel/module.c        |   30 ++++++++++++++++--------------
 2 files changed, 30 insertions(+), 36 deletions(-)

Index: linux-2.6/include/linux/module.h
===================================================================
--- linux-2.6.orig/include/linux/module.h	2010-01-04 14:46:25.000000000 -0600
+++ linux-2.6/include/linux/module.h	2010-01-04 14:47:49.000000000 -0600
@@ -18,6 +18,7 @@
 #include <linux/tracepoint.h>
 
 #include <asm/local.h>
+#include <linux/percpu.h>
 #include <asm/module.h>
 
 #include <trace/events/module.h>
@@ -363,11 +364,9 @@ struct module
 	/* Destruction function. */
 	void (*exit)(void);
 
-#ifdef CONFIG_SMP
-	char *refptr;
-#else
-	local_t ref;
-#endif
+	struct module_ref {
+		int count;
+	} *refptr;
 #endif
 
 #ifdef CONFIG_CONSTRUCTORS
@@ -454,25 +453,16 @@ 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 +471,17 @@ static inline int try_module_get(struct 
 	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;
 }
Index: linux-2.6/kernel/module.c
===================================================================
--- linux-2.6.orig/kernel/module.c	2010-01-04 14:46:25.000000000 -0600
+++ linux-2.6/kernel/module.c	2010-01-04 14:47:07.000000000 -0600
@@ -474,9 +474,10 @@ static void module_unload_init(struct mo
 
 	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 +621,7 @@ unsigned int module_refcount(struct modu
 	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 +798,15 @@ 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);
@@ -1394,9 +1397,9 @@ static void free_module(struct module *m
 	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);
@@ -2159,9 +2162,8 @@ static noinline struct module *load_modu
 	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;
@@ -2393,8 +2395,8 @@ static noinline struct module *load_modu
 	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);

-- 

[this_cpu_xx V9 3/7] Move local.h include to ringbuffer.c and ring_buffer_benchmark.c

[Christoph Lameter]

[-- Attachment #1: percpu_move_include_local.h --]
[-- Type: text/plain, Size: 1633 bytes --]

ringbuffer*.c are the last users of local.h.

Remove the include from modules.h and add it to ringbuffer files.

Signed-off-by: Christoph Lameter <cl@linux-foundation.org>

---
 include/linux/module.h               |    1 -
 kernel/trace/ring_buffer.c           |    1 +
 kernel/trace/ring_buffer_benchmark.c |    1 +
 3 files changed, 2 insertions(+), 1 deletion(-)

Index: linux-2.6/include/linux/module.h
===================================================================
--- linux-2.6.orig/include/linux/module.h	2010-01-04 16:21:31.000000000 -0600
+++ linux-2.6/include/linux/module.h	2010-01-04 16:21:33.000000000 -0600
@@ -17,7 +17,6 @@
 #include <linux/moduleparam.h>
 #include <linux/tracepoint.h>
 
-#include <asm/local.h>
 #include <linux/percpu.h>
 #include <asm/module.h>
 
Index: linux-2.6/kernel/trace/ring_buffer.c
===================================================================
--- linux-2.6.orig/kernel/trace/ring_buffer.c	2010-01-04 16:21:20.000000000 -0600
+++ linux-2.6/kernel/trace/ring_buffer.c	2010-01-04 16:22:30.000000000 -0600
@@ -20,6 +20,7 @@
 #include <linux/cpu.h>
 #include <linux/fs.h>
 
+#include <asm/local.h>
 #include "trace.h"
 
 /*
Index: linux-2.6/kernel/trace/ring_buffer_benchmark.c
===================================================================
--- linux-2.6.orig/kernel/trace/ring_buffer_benchmark.c	2010-01-04 16:21:47.000000000 -0600
+++ linux-2.6/kernel/trace/ring_buffer_benchmark.c	2010-01-04 16:21:57.000000000 -0600
@@ -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;

-- 

[this_cpu_xx V9 4/7] Generic inc / dec percpu instructions

[Christoph Lameter]

[-- Attachment #1: x86_inc_dec --]
[-- Type: text/plain, Size: 7073 bytes --]

Optimize code generated for percpu access by checking for increment and
decrements.

Signed-off-by: Christoph Lameter <cl@linux-foundation.org>

---
 arch/x86/include/asm/percpu.h |  100 ++++++++++++++++++++++++++++++++++++------
 1 file changed, 86 insertions(+), 14 deletions(-)

Index: linux-2.6/arch/x86/include/asm/percpu.h
===================================================================
--- linux-2.6.orig/arch/x86/include/asm/percpu.h	2010-01-04 15:33:02.000000000 -0600
+++ linux-2.6/arch/x86/include/asm/percpu.h	2010-01-04 16:11:29.000000000 -0600
@@ -104,6 +104,78 @@ 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) pto_T__;			\
+	if (0) {					\
+		pto_T__ pto_tmp__;			\
+		pto_tmp__ = (val);			\
+	}						\
+	switch (sizeof(var)) {				\
+	case 1:						\
+		if (__builtin_constant_p(val) == 1)	\
+		  asm("incb "__percpu_arg(0)		\
+		    : "+m" (var)			\
+		    : );				\
+		else if (__builtin_constant_p(val) == -1) \
+		  asm("decb "__percpu_arg(0)		\
+		    : "+m" (var)			\
+		    : );				\
+		else					\
+		  asm("addb %1,"__percpu_arg(0)		\
+		    : "+m" (var)			\
+		    : "qi" ((pto_T__)(val)));		\
+		break;					\
+	case 2:						\
+		if (__builtin_constant_p(val) == 1)	\
+		  asm("incw "__percpu_arg(0)	\
+		    : "+m" (var)			\
+		    : );				\
+		else if (__builtin_constant_p(val) == -1) \
+		  asm("decw "__percpu_arg(0)		\
+		    : "+m" (var)			\
+		    : );				\
+		else					\
+		asm("addw %1,"__percpu_arg(0)		\
+		    : "+m" (var)			\
+		    : "ri" ((pto_T__)(val)));		\
+		break;					\
+	case 4:						\
+		if (__builtin_constant_p(val) == 1)	\
+		  asm("incl "__percpu_arg(0)	\
+		    : "+m" (var)			\
+		    : );				\
+		else if (__builtin_constant_p(val) == -1) \
+		  asm("decl "__percpu_arg(0)		\
+		    : "+m" (var)			\
+		    : );				\
+		else					\
+		asm("addl %1,"__percpu_arg(0)		\
+		    : "+m" (var)			\
+		    : "ri" ((pto_T__)(val)));		\
+		break;					\
+	case 8:						\
+		if (__builtin_constant_p(val) == 1)	\
+		  asm("incq "__percpu_arg(0)	\
+		    : "+m" (var)			\
+		    : );				\
+		else if (__builtin_constant_p(val) == -1) \
+		  asm("decq "__percpu_arg(0)		\
+		    : "+m" (var)			\
+		    : );				\
+		else					\
+		asm("addq %1,"__percpu_arg(0)		\
+		    : "+m" (var)			\
+		    : "re" ((pto_T__)(val)));		\
+		break;					\
+	default: __bad_percpu_size();			\
+	}						\
+} while (0)
+
 #define percpu_from_op(op, var, constraint)		\
 ({							\
 	typeof(var) pfo_ret__;				\
@@ -147,8 +219,8 @@ do {							\
 #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_add(var, val)	percpu_add_op(per_cpu__##var, val)
+#define percpu_sub(var, val)	percpu_add_op(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)
@@ -160,9 +232,9 @@ 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 +251,9 @@ 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 +264,9 @@ 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 +284,19 @@ 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)

-- 

Re: [this_cpu_xx V9 4/7] Generic inc / dec percpu instructions

[Tejun Heo]

On 01/05/2010 07:34 AM, Christoph Lameter wrote:
> Optimize code generated for percpu access by checking for increment and
> decrements.
> 
> Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
> 
> ---
>  arch/x86/include/asm/percpu.h |  100 ++++++++++++++++++++++++++++++++++++------
>  1 file changed, 86 insertions(+), 14 deletions(-)
> 
> Index: linux-2.6/arch/x86/include/asm/percpu.h
> ===================================================================
> --- linux-2.6.orig/arch/x86/include/asm/percpu.h	2010-01-04 15:33:02.000000000 -0600
> +++ linux-2.6/arch/x86/include/asm/percpu.h	2010-01-04 16:11:29.000000000 -0600
> @@ -104,6 +104,78 @@ 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) pto_T__;			\
> +	if (0) {					\
> +		pto_T__ pto_tmp__;			\
> +		pto_tmp__ = (val);			\
> +	}						\
> +	switch (sizeof(var)) {				\
> +	case 1:						\
> +		if (__builtin_constant_p(val) == 1)	\

According to gcc doc, __builtin_constant_p() returns 1 on constant and
0 if not.  It doesn't return the original expression.  I'll commit the
following updated version after testing.

Thanks.

From: Christoph Lameter <cl@linux-foundation.org>

Optimize code generated for percpu access by checking for increment and
decrements.

tj: fix incorrect usage of __builtin_constant_p() and restructure
    percpu_add_op() macro.

Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
---
 arch/x86/include/asm/percpu.h |   86 +++++++++++++++++++++++++++++++++++-------
 1 file changed, 72 insertions(+), 14 deletions(-)

Index: percpu/arch/x86/include/asm/percpu.h
===================================================================
--- percpu.orig/arch/x86/include/asm/percpu.h
+++ percpu/arch/x86/include/asm/percpu.h
@@ -103,6 +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__;				\
@@ -144,8 +202,8 @@ do {							\
 #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_to_op("add", var, val)
-#define percpu_sub(var, val)		percpu_to_op("sub", 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)
@@ -157,9 +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)
@@ -176,9 +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)
@@ -189,9 +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)
@@ -209,19 +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)

-- 
tejun

Re: [this_cpu_xx V9 4/7] Generic inc / dec percpu instructions

[Christoph Lameter]

On Tue, 5 Jan 2010, Tejun Heo wrote:

> According to gcc doc, __builtin_constant_p() returns 1 on constant and
> 0 if not.  It doesn't return the original expression.  I'll commit the
> following updated version after testing.

Great. Thanks. That means that most of the percpu uses currently are incs
and decs not adds. Machine worked fine with the buggy patch.

[this_cpu_xx V9 5/7] this_cpu_ops: page allocator conversion

[Christoph Lameter]

[-- Attachment #1: percpu_page_allocator --]
[-- Type: text/plain, Size: 14851 bytes --]

Use the per cpu allocator functionality to avoid per cpu arrays in struct zone.

This drastically reduces the size of struct zone for systems with large
amounts of processors and allows placement of critical variables of struct
zone in one cacheline even on very large systems.

Another effect is that the pagesets of one processor are placed near one
another. If multiple pagesets from different zones fit into one cacheline
then additional cacheline fetches can be avoided on the hot paths when
allocating memory from multiple zones.

Bootstrap becomes simpler if we use the same scheme for UP, SMP, NUMA. #ifdefs
are reduced and we can drop the zone_pcp macro.

Hotplug handling is also simplified since cpu alloc can bring up and
shut down cpu areas for a specific cpu as a whole. So there is no need to
allocate or free individual pagesets.

V7-V8:
- Explain chicken egg dilemmna with percpu allocator.

V4-V5:
- Fix up cases where per_cpu_ptr is called before irq disable
- Integrate the bootstrap logic that was separate before.

Reviewed-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Christoph Lameter <cl@linux-foundation.org>

---
 include/linux/mm.h     |    4 
 include/linux/mmzone.h |   12 --
 mm/page_alloc.c        |  197 ++++++++++++++++++++-----------------------------
 mm/vmstat.c            |   14 +--
 4 files changed, 92 insertions(+), 135 deletions(-)

Index: linux-2.6/include/linux/mm.h
===================================================================
--- linux-2.6.orig/include/linux/mm.h	2010-01-04 10:28:44.000000000 -0600
+++ linux-2.6/include/linux/mm.h	2010-01-04 14:38:20.000000000 -0600
@@ -1079,11 +1079,7 @@ extern void si_meminfo(struct sysinfo * 
 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);
 
Index: linux-2.6/include/linux/mmzone.h
===================================================================
--- linux-2.6.orig/include/linux/mmzone.h	2010-01-04 10:27:57.000000000 -0600
+++ linux-2.6/include/linux/mmzone.h	2010-01-04 14:38:20.000000000 -0600
@@ -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	*pageset;
 	/*
 	 * free areas of different sizes
 	 */
Index: linux-2.6/mm/page_alloc.c
===================================================================
--- linux-2.6.orig/mm/page_alloc.c	2010-01-04 10:28:44.000000000 -0600
+++ linux-2.6/mm/page_alloc.c	2010-01-04 14:38:20.000000000 -0600
@@ -1008,10 +1008,10 @@ 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);
@@ -1095,7 +1095,6 @@ static void free_hot_cold_page(struct pa
 	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);
@@ -1118,6 +1117,7 @@ static void free_hot_cold_page(struct pa
 		migratetype = MIGRATE_MOVABLE;
 	}
 
+	pcp = &this_cpu_ptr(zone->pageset)->pcp;
 	if (cold)
 		list_add_tail(&page->lru, &pcp->lists[migratetype]);
 	else
@@ -1130,7 +1130,6 @@ static void free_hot_cold_page(struct pa
 
 out:
 	local_irq_restore(flags);
-	put_cpu();
 }
 
 void free_hot_page(struct page *page)
@@ -1180,17 +1179,15 @@ struct page *buffered_rmqueue(struct zon
 	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,
@@ -1231,7 +1228,6 @@ 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))
@@ -1240,7 +1236,6 @@ again:
 
 failed:
 	local_irq_restore(flags);
-	put_cpu();
 	return NULL;
 }
 
@@ -2179,7 +2174,7 @@ 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,
@@ -2744,10 +2739,29 @@ static void build_zonelist_cache(pg_data
 
 #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));
@@ -2758,6 +2772,23 @@ static int __build_all_zonelists(void *d
 		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;
 }
 
@@ -3096,95 +3127,16 @@ static void setup_pagelist_highmark(stru
 }
 
 
-#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.
- */
-static int __cpuinit process_zones(int cpu)
-{
-	struct zone *zone, *dzone;
-	int node = cpu_to_node(cpu);
-
-	node_set_state(node, N_CPU);	/* this node has a 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;
-}
-
-static inline void free_zone_pagesets(int cpu)
-{
-	struct zone *zone;
-
-	for_each_zone(zone) {
-		struct per_cpu_pageset *pset = zone_pcp(zone, cpu);
-
-		/* Free per_cpu_pageset if it is slab allocated */
-		if (pset != &boot_pageset[cpu])
-			kfree(pset);
-		zone_pcp(zone, cpu) = &boot_pageset[cpu];
-	}
-}
-
 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);
+		node_set_state(cpu_to_node(cpu), N_CPU);
 		break;
 	default:
 		break;
@@ -3195,21 +3147,39 @@ static int __cpuinit pageset_cpuup_callb
 static struct notifier_block __cpuinitdata pageset_notifier =
 	{ &pageset_cpuup_callback, NULL, 0 };
 
+/*
+ * 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().
+ */
 void __init setup_per_cpu_pageset(void)
 {
-	int err;
+	struct zone *zone;
+	int cpu;
+
+	for_each_populated_zone(zone) {
+		zone->pageset = alloc_percpu(struct per_cpu_pageset);
 
-	/* Initialize per_cpu_pageset for cpu 0.
-	 * A cpuup callback will do this for every cpu
-	 * as it comes online
+		for_each_possible_cpu(cpu) {
+			struct per_cpu_pageset *pcp = per_cpu_ptr(zone->pageset, cpu);
+
+			setup_pageset(pcp, zone_batchsize(zone));
+
+			if (percpu_pagelist_fraction)
+				setup_pagelist_highmark(pcp,
+					(zone->present_pages /
+						percpu_pagelist_fraction));
+		}
+	}
+	/*
+	 * The boot cpu is always the first active.
+	 * The boot node has a processor.
 	 */
-	err = process_zones(smp_processor_id());
-	BUG_ON(err);
+	node_set_state(cpu_to_node(smp_processor_id()), N_CPU);
 	register_cpu_notifier(&pageset_notifier);
 }
 
-#endif
-
 static noinline __init_refok
 int zone_wait_table_init(struct zone *zone, unsigned long zone_size_pages)
 {
@@ -3263,7 +3233,7 @@ 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);
@@ -3281,21 +3251,17 @@ 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 = &per_cpu_var(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,
@@ -4809,10 +4775,11 @@ int percpu_pagelist_fraction_sysctl_hand
 	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;
Index: linux-2.6/mm/vmstat.c
===================================================================
--- linux-2.6.orig/mm/vmstat.c	2010-01-04 10:27:57.000000000 -0600
+++ linux-2.6/mm/vmstat.c	2010-01-04 14:38:20.000000000 -0600
@@ -139,7 +139,8 @@ static void refresh_zone_stat_thresholds
 		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 __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 s
 	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"

-- 

Re: [this_cpu_xx V9 5/7] this_cpu_ops: page allocator conversion

[Tejun Heo]

On 01/05/2010 07:34 AM, Christoph Lameter wrote:
>  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);
> +		node_set_state(cpu_to_node(cpu), N_CPU);
>  		break;
>  	default:
>  		break;

The above chunk causes build failure due to removed ret used later on
return statement.  Changed it to return NOTIFY_OK.

Thanks.

-- 
tejun

Re: [this_cpu_xx V9 5/7] this_cpu_ops: page allocator conversion

[Christoph Lameter]

On Tue, 5 Jan 2010, Tejun Heo wrote:

> The above chunk causes build failure due to removed ret used later on
> return statement.  Changed it to return NOTIFY_OK.

Thats a change in -next? Patch was against upstream.

Re: [this_cpu_xx V9 5/7] this_cpu_ops: page allocator conversion

[Tejun Heo]

On 01/06/2010 12:22 AM, Christoph Lameter wrote:
> On Tue, 5 Jan 2010, Tejun Heo wrote:
> 
>> The above chunk causes build failure due to removed ret used later on
>> return statement.  Changed it to return NOTIFY_OK.
> 
> Thats a change in -next? Patch was against upstream.

I don't think the build failure was specific to the percpu branch.
pageset_cpuup_callback() returns ret on upstream too.

Thanks.

-- 
tejun

[this_cpu_xx V9 6/7] this_cpu ops: Remove pageset_notifier

[Christoph Lameter]

[-- Attachment #1: percpu_page_alloc_remove_pageset_notifier --]
[-- Type: text/plain, Size: 2007 bytes --]

Remove the pageset notifier since it only marks that a processor
exists on a specific node. Move that code into the vmstat notifier.

Signed-off-by: Christoph Lameter <cl@linux-foundation.org>

---
 mm/page_alloc.c |   27 ---------------------------
 mm/vmstat.c     |    1 +
 2 files changed, 1 insertion(+), 27 deletions(-)

Index: linux-2.6/mm/vmstat.c
===================================================================
--- linux-2.6.orig/mm/vmstat.c	2010-01-04 14:44:26.000000000 -0600
+++ linux-2.6/mm/vmstat.c	2010-01-04 14:44:28.000000000 -0600
@@ -908,6 +908,7 @@ static int __cpuinit vmstat_cpuup_callba
 	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:
Index: linux-2.6/mm/page_alloc.c
===================================================================
--- linux-2.6.orig/mm/page_alloc.c	2010-01-04 14:44:26.000000000 -0600
+++ linux-2.6/mm/page_alloc.c	2010-01-04 14:44:28.000000000 -0600
@@ -3126,27 +3126,6 @@ static void setup_pagelist_highmark(stru
 		pcp->batch = PAGE_SHIFT * 8;
 }
 
-
-static int __cpuinit pageset_cpuup_callback(struct notifier_block *nfb,
-		unsigned long action,
-		void *hcpu)
-{
-	int cpu = (long)hcpu;
-
-	switch (action) {
-	case CPU_UP_PREPARE:
-	case CPU_UP_PREPARE_FROZEN:
-		node_set_state(cpu_to_node(cpu), N_CPU);
-		break;
-	default:
-		break;
-	}
-	return ret;
-}
-
-static struct notifier_block __cpuinitdata pageset_notifier =
-	{ &pageset_cpuup_callback, NULL, 0 };
-
 /*
  * Allocate per cpu pagesets and initialize them.
  * Before this call only boot pagesets were available.
@@ -3172,12 +3151,6 @@ void __init setup_per_cpu_pageset(void)
 						percpu_pagelist_fraction));
 		}
 	}
-	/*
-	 * The boot cpu is always the first active.
-	 * The boot node has a processor.
-	 */
-	node_set_state(cpu_to_node(smp_processor_id()), N_CPU);
-	register_cpu_notifier(&pageset_notifier);
 }
 
 static noinline __init_refok

-- 

[this_cpu_xx V9 7/7] Remove leftover local.h

[Christoph Lameter]

[-- Attachment #1: percpu_sparc_remove_local.h --]
[-- Type: text/plain, Size: 715 bytes --]

Somehow the local.h was not removed when taking out the local_t usage during
the 2.6.32 merge.

CC: David Miller <davem@davemloft.net>
Signed-off-by: Christoph Lameter <cl@linux-foundation.org>

---
 arch/sparc/kernel/nmi.c |    1 -
 1 file changed, 1 deletion(-)

Index: linux-2.6/arch/sparc/kernel/nmi.c
===================================================================
--- linux-2.6.orig/arch/sparc/kernel/nmi.c	2010-01-04 15:03:35.000000000 -0600
+++ linux-2.6/arch/sparc/kernel/nmi.c	2010-01-04 15:03:42.000000000 -0600
@@ -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

-- 

Re: [this_cpu_xx V9 0/7] Per cpu atomics in page allocator, modules, cleanup and optimized inc/dec

[Tejun Heo]

1-7 applied to the following git branch.

  git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu.git percpu

and also pushed out to linux-next.  Please base further changes on top
of the #percpu branch.

Thanks.

-- 
tejun