[net-next-2.6.git] / include / linux / slub_def.h

#ifndef _LINUX_SLUB_DEF_H
#define _LINUX_SLUB_DEF_H

/*
 * SLUB : A Slab allocator without object queues.
 *
 * (C) 2007 SGI, Christoph Lameter
 */
#include <linux/types.h>
#include <linux/gfp.h>
#include <linux/workqueue.h>
#include <linux/kobject.h>
#include <linux/kmemtrace.h>

enum stat_item {
	ALLOC_FASTPATH,		/* Allocation from cpu slab */
	ALLOC_SLOWPATH,		/* Allocation by getting a new cpu slab */
	FREE_FASTPATH,		/* Free to cpu slub */
	FREE_SLOWPATH,		/* Freeing not to cpu slab */
	FREE_FROZEN,		/* Freeing to frozen slab */
	FREE_ADD_PARTIAL,	/* Freeing moves slab to partial list */
	FREE_REMOVE_PARTIAL,	/* Freeing removes last object */
	ALLOC_FROM_PARTIAL,	/* Cpu slab acquired from partial list */
	ALLOC_SLAB,		/* Cpu slab acquired from page allocator */
	ALLOC_REFILL,		/* Refill cpu slab from slab freelist */
	FREE_SLAB,		/* Slab freed to the page allocator */
	CPUSLAB_FLUSH,		/* Abandoning of the cpu slab */
	DEACTIVATE_FULL,	/* Cpu slab was full when deactivated */
	DEACTIVATE_EMPTY,	/* Cpu slab was empty when deactivated */
	DEACTIVATE_TO_HEAD,	/* Cpu slab was moved to the head of partials */
	DEACTIVATE_TO_TAIL,	/* Cpu slab was moved to the tail of partials */
	DEACTIVATE_REMOTE_FREES,/* Slab contained remotely freed objects */
	ORDER_FALLBACK,		/* Number of times fallback was necessary */
	NR_SLUB_STAT_ITEMS };

struct kmem_cache_cpu {
	void **freelist;	/* Pointer to first free per cpu object */
	struct page *page;	/* The slab from which we are allocating */
	int node;		/* The node of the page (or -1 for debug) */
	unsigned int offset;	/* Freepointer offset (in word units) */
	unsigned int objsize;	/* Size of an object (from kmem_cache) */
#ifdef CONFIG_SLUB_STATS
	unsigned stat[NR_SLUB_STAT_ITEMS];
#endif
};

struct kmem_cache_node {
	spinlock_t list_lock;	/* Protect partial list and nr_partial */
	unsigned long nr_partial;
	struct list_head partial;
#ifdef CONFIG_SLUB_DEBUG
	atomic_long_t nr_slabs;
	atomic_long_t total_objects;
	struct list_head full;
#endif
};

/*
 * Word size structure that can be atomically updated or read and that
 * contains both the order and the number of objects that a slab of the
 * given order would contain.
 */
struct kmem_cache_order_objects {
	unsigned long x;
};

/*
 * Slab cache management.
 */
struct kmem_cache {
	/* Used for retriving partial slabs etc */
	unsigned long flags;
	int size;		/* The size of an object including meta data */
	int objsize;		/* The size of an object without meta data */
	int offset;		/* Free pointer offset. */
	struct kmem_cache_order_objects oo;

	/*
	 * Avoid an extra cache line for UP, SMP and for the node local to
	 * struct kmem_cache.
	 */
	struct kmem_cache_node local_node;

	/* Allocation and freeing of slabs */
	struct kmem_cache_order_objects max;
	struct kmem_cache_order_objects min;
	gfp_t allocflags;	/* gfp flags to use on each alloc */
	int refcount;		/* Refcount for slab cache destroy */
	void (*ctor)(void *);
	int inuse;		/* Offset to metadata */
	int align;		/* Alignment */
	unsigned long min_partial;
	const char *name;	/* Name (only for display!) */
	struct list_head list;	/* List of slab caches */
#ifdef CONFIG_SLUB_DEBUG
	struct kobject kobj;	/* For sysfs */
#endif

#ifdef CONFIG_NUMA
	/*
	 * Defragmentation by allocating from a remote node.
	 */
	int remote_node_defrag_ratio;
	struct kmem_cache_node *node[MAX_NUMNODES];
#endif
#ifdef CONFIG_SMP
	struct kmem_cache_cpu *cpu_slab[NR_CPUS];
#else
	struct kmem_cache_cpu cpu_slab;
#endif
};

/*
 * Kmalloc subsystem.
 */
#if defined(ARCH_KMALLOC_MINALIGN) && ARCH_KMALLOC_MINALIGN > 8
#define KMALLOC_MIN_SIZE ARCH_KMALLOC_MINALIGN
#else
#define KMALLOC_MIN_SIZE 8
#endif

#define KMALLOC_SHIFT_LOW ilog2(KMALLOC_MIN_SIZE)

/*
 * Maximum kmalloc object size handled by SLUB. Larger object allocations
 * are passed through to the page allocator. The page allocator "fastpath"
 * is relatively slow so we need this value sufficiently high so that
 * performance critical objects are allocated through the SLUB fastpath.
 *
 * This should be dropped to PAGE_SIZE / 2 once the page allocator
 * "fastpath" becomes competitive with the slab allocator fastpaths.
 */
#define SLUB_MAX_SIZE (2 * PAGE_SIZE)

#define SLUB_PAGE_SHIFT (PAGE_SHIFT + 2)

/*
 * We keep the general caches in an array of slab caches that are used for
 * 2^x bytes of allocations.
 */
extern struct kmem_cache kmalloc_caches[SLUB_PAGE_SHIFT];

/*
 * Sorry that the following has to be that ugly but some versions of GCC
 * have trouble with constant propagation and loops.
 */
static __always_inline int kmalloc_index(size_t size)
{
	if (!size)
		return 0;

	if (size <= KMALLOC_MIN_SIZE)
		return KMALLOC_SHIFT_LOW;

#if KMALLOC_MIN_SIZE <= 64
	if (size > 64 && size <= 96)
		return 1;
	if (size > 128 && size <= 192)
		return 2;
#endif
	if (size <=          8) return 3;
	if (size <=         16) return 4;
	if (size <=         32) return 5;
	if (size <=         64) return 6;
	if (size <=        128) return 7;
	if (size <=        256) return 8;
	if (size <=        512) return 9;
	if (size <=       1024) return 10;
	if (size <=   2 * 1024) return 11;
	if (size <=   4 * 1024) return 12;
/*
 * The following is only needed to support architectures with a larger page
 * size than 4k.
 */
	if (size <=   8 * 1024) return 13;
	if (size <=  16 * 1024) return 14;
	if (size <=  32 * 1024) return 15;
	if (size <=  64 * 1024) return 16;
	if (size <= 128 * 1024) return 17;
	if (size <= 256 * 1024) return 18;
	if (size <= 512 * 1024) return 19;
	if (size <= 1024 * 1024) return 20;
	if (size <=  2 * 1024 * 1024) return 21;
	return -1;

/*
 * What we really wanted to do and cannot do because of compiler issues is:
 *	int i;
 *	for (i = KMALLOC_SHIFT_LOW; i <= KMALLOC_SHIFT_HIGH; i++)
 *		if (size <= (1 << i))
 *			return i;
 */
}

/*
 * Find the slab cache for a given combination of allocation flags and size.
 *
 * This ought to end up with a global pointer to the right cache
 * in kmalloc_caches.
 */
static __always_inline struct kmem_cache *kmalloc_slab(size_t size)
{
	int index = kmalloc_index(size);

	if (index == 0)
		return NULL;

	return &kmalloc_caches[index];
}

#ifdef CONFIG_ZONE_DMA
#define SLUB_DMA __GFP_DMA
#else
/* Disable DMA functionality */
#define SLUB_DMA (__force gfp_t)0
#endif

void *kmem_cache_alloc(struct kmem_cache *, gfp_t);
void *__kmalloc(size_t size, gfp_t flags);

#ifdef CONFIG_KMEMTRACE
extern void *kmem_cache_alloc_notrace(struct kmem_cache *s, gfp_t gfpflags);
#else
static __always_inline void *
kmem_cache_alloc_notrace(struct kmem_cache *s, gfp_t gfpflags)
{
	return kmem_cache_alloc(s, gfpflags);
}
#endif

static __always_inline void *kmalloc_large(size_t size, gfp_t flags)
{
	unsigned int order = get_order(size);
	void *ret = (void *) __get_free_pages(flags | __GFP_COMP, order);

	trace_kmalloc(_THIS_IP_, ret, size, PAGE_SIZE << order, flags);

	return ret;
}

static __always_inline void *kmalloc(size_t size, gfp_t flags)
{
	void *ret;

	if (__builtin_constant_p(size)) {
		if (size > SLUB_MAX_SIZE)
			return kmalloc_large(size, flags);

		if (!(flags & SLUB_DMA)) {
			struct kmem_cache *s = kmalloc_slab(size);

			if (!s)
				return ZERO_SIZE_PTR;

			ret = kmem_cache_alloc_notrace(s, flags);

			trace_kmalloc(_THIS_IP_, ret, size, s->size, flags);

			return ret;
		}
	}
	return __kmalloc(size, flags);
}

#ifdef CONFIG_NUMA
void *__kmalloc_node(size_t size, gfp_t flags, int node);
void *kmem_cache_alloc_node(struct kmem_cache *, gfp_t flags, int node);

#ifdef CONFIG_KMEMTRACE
extern void *kmem_cache_alloc_node_notrace(struct kmem_cache *s,
					   gfp_t gfpflags,
					   int node);
#else
static __always_inline void *
kmem_cache_alloc_node_notrace(struct kmem_cache *s,
			      gfp_t gfpflags,
			      int node)
{
	return kmem_cache_alloc_node(s, gfpflags, node);
}
#endif

static __always_inline void *kmalloc_node(size_t size, gfp_t flags, int node)
{
	void *ret;

	if (__builtin_constant_p(size) &&
		size <= SLUB_MAX_SIZE && !(flags & SLUB_DMA)) {
			struct kmem_cache *s = kmalloc_slab(size);

		if (!s)
			return ZERO_SIZE_PTR;

		ret = kmem_cache_alloc_node_notrace(s, flags, node);

		trace_kmalloc_node(_THIS_IP_, ret,
				   size, s->size, flags, node);

		return ret;
	}
	return __kmalloc_node(size, flags, node);
}
#endif

void __init kmem_cache_init_late(void);

#endif /* _LINUX_SLUB_DEF_H */
Commit	Line	Data
81819f0f CL	1	#ifndef _LINUX_SLUB_DEF_H
	2	#define _LINUX_SLUB_DEF_H
	3
	4	/*
	5	* SLUB : A Slab allocator without object queues.
	6	*
cde53535	7	* (C) 2007 SGI, Christoph Lameter
81819f0f CL	8	*/
	9	#include <linux/types.h>
	10	#include <linux/gfp.h>
	11	#include <linux/workqueue.h>
	12	#include <linux/kobject.h>
02af61bb	13	#include <linux/kmemtrace.h>
81819f0f	14
8ff12cfc CL	15	enum stat_item {
	16	ALLOC_FASTPATH, /* Allocation from cpu slab */
	17	ALLOC_SLOWPATH, /* Allocation by getting a new cpu slab */
	18	FREE_FASTPATH, /* Free to cpu slub */
	19	FREE_SLOWPATH, /* Freeing not to cpu slab */
	20	FREE_FROZEN, /* Freeing to frozen slab */
	21	FREE_ADD_PARTIAL, /* Freeing moves slab to partial list */
	22	FREE_REMOVE_PARTIAL, /* Freeing removes last object */
	23	ALLOC_FROM_PARTIAL, /* Cpu slab acquired from partial list */
	24	ALLOC_SLAB, /* Cpu slab acquired from page allocator */
	25	ALLOC_REFILL, /* Refill cpu slab from slab freelist */
	26	FREE_SLAB, /* Slab freed to the page allocator */
	27	CPUSLAB_FLUSH, /* Abandoning of the cpu slab */
	28	DEACTIVATE_FULL, /* Cpu slab was full when deactivated */
	29	DEACTIVATE_EMPTY, /* Cpu slab was empty when deactivated */
	30	DEACTIVATE_TO_HEAD, /* Cpu slab was moved to the head of partials */
	31	DEACTIVATE_TO_TAIL, /* Cpu slab was moved to the tail of partials */
	32	DEACTIVATE_REMOTE_FREES,/* Slab contained remotely freed objects */
65c3376a	33	ORDER_FALLBACK, /* Number of times fallback was necessary */
8ff12cfc CL	34	NR_SLUB_STAT_ITEMS };
8ff12cfc CL	35
dfb4f096	36	struct kmem_cache_cpu {
da89b79e CL	37	void *freelist; / Pointer to first free per cpu object */
	38	struct page page; / The slab from which we are allocating */
	39	int node; /* The node of the page (or -1 for debug) */
	40	unsigned int offset; /* Freepointer offset (in word units) */
	41	unsigned int objsize; /* Size of an object (from kmem_cache) */
8ff12cfc CL	42	#ifdef CONFIG_SLUB_STATS
	43	unsigned stat[NR_SLUB_STAT_ITEMS];
	44	#endif
4c93c355	45	};
dfb4f096	46
81819f0f CL	47	struct kmem_cache_node {
	48	spinlock_t list_lock; /* Protect partial list and nr_partial */
	49	unsigned long nr_partial;
81819f0f	50	struct list_head partial;
0c710013	51	#ifdef CONFIG_SLUB_DEBUG
0f389ec6	52	atomic_long_t nr_slabs;
205ab99d	53	atomic_long_t total_objects;
643b1138	54	struct list_head full;
0c710013	55	#endif
81819f0f CL	56	};
81819f0f CL	57
834f3d11 CL	58	/*
	59	* Word size structure that can be atomically updated or read and that
	60	* contains both the order and the number of objects that a slab of the
	61	* given order would contain.
	62	*/
	63	struct kmem_cache_order_objects {
	64	unsigned long x;
	65	};
	66
81819f0f CL	67	/*
	68	* Slab cache management.
	69	*/
	70	struct kmem_cache {
	71	/* Used for retriving partial slabs etc */
	72	unsigned long flags;
	73	int size; /* The size of an object including meta data */
	74	int objsize; /* The size of an object without meta data */
	75	int offset; /* Free pointer offset. */
834f3d11	76	struct kmem_cache_order_objects oo;
81819f0f CL	77
	78	/*
	79	* Avoid an extra cache line for UP, SMP and for the node local to
	80	* struct kmem_cache.
	81	*/
	82	struct kmem_cache_node local_node;
	83
	84	/* Allocation and freeing of slabs */
205ab99d	85	struct kmem_cache_order_objects max;
65c3376a	86	struct kmem_cache_order_objects min;
b7a49f0d	87	gfp_t allocflags; /* gfp flags to use on each alloc */
81819f0f	88	int refcount; /* Refcount for slab cache destroy */
51cc5068	89	void (ctor)(void );
81819f0f CL	90	int inuse; /* Offset to metadata */
81819f0f CL	91	int align; /* Alignment */
3b89d7d8	92	unsigned long min_partial;
81819f0f CL	93	const char name; / Name (only for display!) */
81819f0f CL	94	struct list_head list; /* List of slab caches */
0c710013	95	#ifdef CONFIG_SLUB_DEBUG
81819f0f	96	struct kobject kobj; /* For sysfs */
0c710013	97	#endif
81819f0f CL	98
81819f0f CL	99	#ifdef CONFIG_NUMA
9824601e CL	100	/*
	101	* Defragmentation by allocating from a remote node.
	102	*/
	103	int remote_node_defrag_ratio;
81819f0f CL	104	struct kmem_cache_node *node[MAX_NUMNODES];
81819f0f CL	105	#endif
4c93c355 CL	106	#ifdef CONFIG_SMP
	107	struct kmem_cache_cpu *cpu_slab[NR_CPUS];
	108	#else
	109	struct kmem_cache_cpu cpu_slab;
	110	#endif
81819f0f CL	111	};
	112
	113	/*
	114	* Kmalloc subsystem.
	115	*/
4b356be0 CL	116	#if defined(ARCH_KMALLOC_MINALIGN) && ARCH_KMALLOC_MINALIGN > 8
	117	#define KMALLOC_MIN_SIZE ARCH_KMALLOC_MINALIGN
	118	#else
	119	#define KMALLOC_MIN_SIZE 8
	120	#endif
	121
	122	#define KMALLOC_SHIFT_LOW ilog2(KMALLOC_MIN_SIZE)
81819f0f	123
ffadd4d0 CL	124	/*
	125	* Maximum kmalloc object size handled by SLUB. Larger object allocations
	126	* are passed through to the page allocator. The page allocator "fastpath"
	127	* is relatively slow so we need this value sufficiently high so that
	128	* performance critical objects are allocated through the SLUB fastpath.
	129	*
	130	* This should be dropped to PAGE_SIZE / 2 once the page allocator
	131	* "fastpath" becomes competitive with the slab allocator fastpaths.
	132	*/
51735a7c	133	#define SLUB_MAX_SIZE (2 * PAGE_SIZE)
ffadd4d0	134
51735a7c	135	#define SLUB_PAGE_SHIFT (PAGE_SHIFT + 2)
ffadd4d0	136
81819f0f CL	137	/*
	138	* We keep the general caches in an array of slab caches that are used for
	139	* 2^x bytes of allocations.
	140	*/
ffadd4d0	141	extern struct kmem_cache kmalloc_caches[SLUB_PAGE_SHIFT];
81819f0f CL	142
	143	/*
	144	* Sorry that the following has to be that ugly but some versions of GCC
	145	* have trouble with constant propagation and loops.
	146	*/
aa137f9d	147	static __always_inline int kmalloc_index(size_t size)
81819f0f	148	{
272c1d21 CL	149	if (!size)
272c1d21 CL	150	return 0;
614410d5	151
4b356be0 CL	152	if (size <= KMALLOC_MIN_SIZE)
	153	return KMALLOC_SHIFT_LOW;
	154
41d54d3b	155	#if KMALLOC_MIN_SIZE <= 64
81819f0f CL	156	if (size > 64 && size <= 96)
	157	return 1;
	158	if (size > 128 && size <= 192)
	159	return 2;
41d54d3b	160	#endif
81819f0f CL	161	if (size <= 8) return 3;
	162	if (size <= 16) return 4;
	163	if (size <= 32) return 5;
	164	if (size <= 64) return 6;
	165	if (size <= 128) return 7;
	166	if (size <= 256) return 8;
	167	if (size <= 512) return 9;
	168	if (size <= 1024) return 10;
	169	if (size <= 2 * 1024) return 11;
6446faa2	170	if (size <= 4 * 1024) return 12;
aadb4bc4 CL	171	/*
	172	* The following is only needed to support architectures with a larger page
	173	* size than 4k.
	174	*/
81819f0f CL	175	if (size <= 8 * 1024) return 13;
	176	if (size <= 16 * 1024) return 14;
	177	if (size <= 32 * 1024) return 15;
	178	if (size <= 64 * 1024) return 16;
	179	if (size <= 128 * 1024) return 17;
	180	if (size <= 256 * 1024) return 18;
aadb4bc4	181	if (size <= 512 * 1024) return 19;
81819f0f	182	if (size <= 1024 * 1024) return 20;
81819f0f	183	if (size <= 2 * 1024 * 1024) return 21;
81819f0f CL	184	return -1;
	185
	186	/*
	187	* What we really wanted to do and cannot do because of compiler issues is:
	188	* int i;
	189	* for (i = KMALLOC_SHIFT_LOW; i <= KMALLOC_SHIFT_HIGH; i++)
	190	* if (size <= (1 << i))
	191	* return i;
	192	*/
	193	}
	194
	195	/*
	196	* Find the slab cache for a given combination of allocation flags and size.
	197	*
	198	* This ought to end up with a global pointer to the right cache
	199	* in kmalloc_caches.
	200	*/
aa137f9d	201	static __always_inline struct kmem_cache *kmalloc_slab(size_t size)
81819f0f CL	202	{
	203	int index = kmalloc_index(size);
	204
	205	if (index == 0)
	206	return NULL;
	207
81819f0f CL	208	return &kmalloc_caches[index];
	209	}
	210
	211	#ifdef CONFIG_ZONE_DMA
	212	#define SLUB_DMA __GFP_DMA
	213	#else
	214	/* Disable DMA functionality */
d046943c	215	#define SLUB_DMA (__force gfp_t)0
81819f0f CL	216	#endif
81819f0f CL	217
6193a2ff PM	218	void kmem_cache_alloc(struct kmem_cache , gfp_t);
	219	void *__kmalloc(size_t size, gfp_t flags);
	220
5b882be4 EGM	221	#ifdef CONFIG_KMEMTRACE
	222	extern void kmem_cache_alloc_notrace(struct kmem_cache s, gfp_t gfpflags);
	223	#else
	224	static __always_inline void *
	225	kmem_cache_alloc_notrace(struct kmem_cache *s, gfp_t gfpflags)
	226	{
	227	return kmem_cache_alloc(s, gfpflags);
	228	}
	229	#endif
	230
eada35ef PE	231	static __always_inline void *kmalloc_large(size_t size, gfp_t flags)
eada35ef PE	232	{
5b882be4 EGM	233	unsigned int order = get_order(size);
	234	void ret = (void ) __get_free_pages(flags \| __GFP_COMP, order);
	235
ca2b84cb	236	trace_kmalloc(_THIS_IP_, ret, size, PAGE_SIZE << order, flags);
5b882be4 EGM	237
5b882be4 EGM	238	return ret;
eada35ef PE	239	}
eada35ef PE	240
aa137f9d	241	static __always_inline void *kmalloc(size_t size, gfp_t flags)
81819f0f	242	{
5b882be4 EGM	243	void *ret;
5b882be4 EGM	244
aadb4bc4	245	if (__builtin_constant_p(size)) {
ffadd4d0	246	if (size > SLUB_MAX_SIZE)
eada35ef	247	return kmalloc_large(size, flags);
81819f0f	248
aadb4bc4 CL	249	if (!(flags & SLUB_DMA)) {
	250	struct kmem_cache *s = kmalloc_slab(size);
	251
	252	if (!s)
	253	return ZERO_SIZE_PTR;
81819f0f	254
5b882be4 EGM	255	ret = kmem_cache_alloc_notrace(s, flags);
5b882be4 EGM	256
ca2b84cb	257	trace_kmalloc(_THIS_IP_, ret, size, s->size, flags);
5b882be4 EGM	258
5b882be4 EGM	259	return ret;
aadb4bc4 CL	260	}
	261	}
	262	return __kmalloc(size, flags);
81819f0f CL	263	}
81819f0f CL	264
81819f0f	265	#ifdef CONFIG_NUMA
6193a2ff PM	266	void *__kmalloc_node(size_t size, gfp_t flags, int node);
6193a2ff PM	267	void kmem_cache_alloc_node(struct kmem_cache , gfp_t flags, int node);
81819f0f	268
5b882be4 EGM	269	#ifdef CONFIG_KMEMTRACE
	270	extern void kmem_cache_alloc_node_notrace(struct kmem_cache s,
	271	gfp_t gfpflags,
	272	int node);
	273	#else
	274	static __always_inline void *
	275	kmem_cache_alloc_node_notrace(struct kmem_cache *s,
	276	gfp_t gfpflags,
	277	int node)
	278	{
	279	return kmem_cache_alloc_node(s, gfpflags, node);
	280	}
	281	#endif
	282
aa137f9d	283	static __always_inline void *kmalloc_node(size_t size, gfp_t flags, int node)
81819f0f	284	{
5b882be4 EGM	285	void *ret;
5b882be4 EGM	286
aadb4bc4	287	if (__builtin_constant_p(size) &&
ffadd4d0	288	size <= SLUB_MAX_SIZE && !(flags & SLUB_DMA)) {
aadb4bc4	289	struct kmem_cache *s = kmalloc_slab(size);
81819f0f CL	290
81819f0f CL	291	if (!s)
272c1d21	292	return ZERO_SIZE_PTR;
81819f0f	293
5b882be4 EGM	294	ret = kmem_cache_alloc_node_notrace(s, flags, node);
5b882be4 EGM	295
ca2b84cb EGM	296	trace_kmalloc_node(_THIS_IP_, ret,
ca2b84cb EGM	297	size, s->size, flags, node);
5b882be4 EGM	298
5b882be4 EGM	299	return ret;
aadb4bc4 CL	300	}
aadb4bc4 CL	301	return __kmalloc_node(size, flags, node);
81819f0f CL	302	}
	303	#endif
	304
7e85ee0c PE	305	void __init kmem_cache_init_late(void);
7e85ee0c PE	306
81819f0f	307	#endif /* _LINUX_SLUB_DEF_H */