gfp.h 8.4 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11
#ifndef __LINUX_GFP_H
#define __LINUX_GFP_H

#include <linux/mmzone.h>
#include <linux/stddef.h>
#include <linux/linkage.h>

struct vm_area_struct;

/*
 * GFP bitmasks..
12 13 14 15 16 17
 *
 * Zone modifiers (see linux/mmzone.h - low three bits)
 *
 * Do not put any conditional on these. If necessary modify the definitions
 * without the underscores and use the consistently. The definitions here may
 * be used in bit comparisons.
L
Linus Torvalds 已提交
18
 */
A
Al Viro 已提交
19 20
#define __GFP_DMA	((__force gfp_t)0x01u)
#define __GFP_HIGHMEM	((__force gfp_t)0x02u)
21
#define __GFP_DMA32	((__force gfp_t)0x04u)
L
Linus Torvalds 已提交
22 23 24 25 26 27 28 29 30 31 32

/*
 * Action modifiers - doesn't change the zoning
 *
 * __GFP_REPEAT: Try hard to allocate the memory, but the allocation attempt
 * _might_ fail.  This depends upon the particular VM implementation.
 *
 * __GFP_NOFAIL: The VM implementation _must_ retry infinitely: the caller
 * cannot handle allocation failures.
 *
 * __GFP_NORETRY: The VM implementation must not retry indefinitely.
33 34 35
 *
 * __GFP_MOVABLE: Flag that this page will be movable by the page migration
 * mechanism or reclaimed
L
Linus Torvalds 已提交
36
 */
A
Al Viro 已提交
37 38 39 40 41 42
#define __GFP_WAIT	((__force gfp_t)0x10u)	/* Can wait and reschedule? */
#define __GFP_HIGH	((__force gfp_t)0x20u)	/* Should access emergency pools? */
#define __GFP_IO	((__force gfp_t)0x40u)	/* Can start physical IO? */
#define __GFP_FS	((__force gfp_t)0x80u)	/* Can call down to low-level FS? */
#define __GFP_COLD	((__force gfp_t)0x100u)	/* Cache-cold page required */
#define __GFP_NOWARN	((__force gfp_t)0x200u)	/* Suppress page allocation failure warning */
43 44 45
#define __GFP_REPEAT	((__force gfp_t)0x400u)	/* See above */
#define __GFP_NOFAIL	((__force gfp_t)0x800u)	/* See above */
#define __GFP_NORETRY	((__force gfp_t)0x1000u)/* See above */
A
Al Viro 已提交
46 47 48
#define __GFP_COMP	((__force gfp_t)0x4000u)/* Add compound page metadata */
#define __GFP_ZERO	((__force gfp_t)0x8000u)/* Return zeroed page on success */
#define __GFP_NOMEMALLOC ((__force gfp_t)0x10000u) /* Don't use emergency reserves */
P
Paul Jackson 已提交
49
#define __GFP_HARDWALL   ((__force gfp_t)0x20000u) /* Enforce hardwall cpuset memory allocs */
50
#define __GFP_THISNODE	((__force gfp_t)0x40000u)/* No fallback, no policies */
51 52
#define __GFP_RECLAIMABLE ((__force gfp_t)0x80000u) /* Page is reclaimable */
#define __GFP_MOVABLE	((__force gfp_t)0x100000u)  /* Page is movable */
L
Linus Torvalds 已提交
53

54
#define __GFP_BITS_SHIFT 21	/* Room for 21 __GFP_FOO bits */
A
Al Viro 已提交
55
#define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
L
Linus Torvalds 已提交
56

J
Jeff Dike 已提交
57 58
/* This equals 0, but use constants in case they ever change */
#define GFP_NOWAIT	(GFP_ATOMIC & ~__GFP_HIGH)
P
Paul Jackson 已提交
59
/* GFP_ATOMIC means both !wait (__GFP_WAIT not set) and use emergency pool */
L
Linus Torvalds 已提交
60 61 62 63
#define GFP_ATOMIC	(__GFP_HIGH)
#define GFP_NOIO	(__GFP_WAIT)
#define GFP_NOFS	(__GFP_WAIT | __GFP_IO)
#define GFP_KERNEL	(__GFP_WAIT | __GFP_IO | __GFP_FS)
64 65
#define GFP_TEMPORARY	(__GFP_WAIT | __GFP_IO | __GFP_FS | \
			 __GFP_RECLAIMABLE)
66 67 68
#define GFP_USER	(__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL)
#define GFP_HIGHUSER	(__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL | \
			 __GFP_HIGHMEM)
69 70 71 72 73 74 75 76 77
#define GFP_HIGHUSER_MOVABLE	(__GFP_WAIT | __GFP_IO | __GFP_FS | \
				 __GFP_HARDWALL | __GFP_HIGHMEM | \
				 __GFP_MOVABLE)
#define GFP_NOFS_PAGECACHE	(__GFP_WAIT | __GFP_IO | __GFP_MOVABLE)
#define GFP_USER_PAGECACHE	(__GFP_WAIT | __GFP_IO | __GFP_FS | \
				 __GFP_HARDWALL | __GFP_MOVABLE)
#define GFP_HIGHUSER_PAGECACHE	(__GFP_WAIT | __GFP_IO | __GFP_FS | \
				 __GFP_HARDWALL | __GFP_HIGHMEM | \
				 __GFP_MOVABLE)
L
Linus Torvalds 已提交
78

79
#ifdef CONFIG_NUMA
80
#define GFP_THISNODE	(__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY)
81
#else
82
#define GFP_THISNODE	((__force gfp_t)0)
83 84
#endif

C
Christoph Lameter 已提交
85
/* This mask makes up all the page movable related flags */
86
#define GFP_MOVABLE_MASK (__GFP_RECLAIMABLE|__GFP_MOVABLE)
C
Christoph Lameter 已提交
87 88 89 90 91 92 93 94 95 96 97

/* Control page allocator reclaim behavior */
#define GFP_RECLAIM_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS|\
			__GFP_NOWARN|__GFP_REPEAT|__GFP_NOFAIL|\
			__GFP_NORETRY|__GFP_NOMEMALLOC)

/* Control allocation constraints */
#define GFP_CONSTRAINT_MASK (__GFP_HARDWALL|__GFP_THISNODE)

/* Do not use these with a slab allocator */
#define GFP_SLAB_BUG_MASK (__GFP_DMA32|__GFP_HIGHMEM|~__GFP_BITS_MASK)
98

L
Linus Torvalds 已提交
99 100 101 102 103
/* Flag - indicates that the buffer will be suitable for DMA.  Ignored on some
   platforms, used as appropriate on others */

#define GFP_DMA		__GFP_DMA

A
Andi Kleen 已提交
104 105 106
/* 4GB DMA on some platforms */
#define GFP_DMA32	__GFP_DMA32

107 108 109 110 111 112 113 114 115 116 117 118
/* Convert GFP flags to their corresponding migrate type */
static inline int allocflags_to_migratetype(gfp_t gfp_flags)
{
	WARN_ON((gfp_flags & GFP_MOVABLE_MASK) == GFP_MOVABLE_MASK);

	if (unlikely(page_group_by_mobility_disabled))
		return MIGRATE_UNMOVABLE;

	/* Group based on mobility */
	return (((gfp_flags & __GFP_MOVABLE) != 0) << 1) |
		((gfp_flags & __GFP_RECLAIMABLE) != 0);
}
A
Andi Kleen 已提交
119

120
static inline enum zone_type gfp_zone(gfp_t flags)
121
{
122
#ifdef CONFIG_ZONE_DMA
123
	if (flags & __GFP_DMA)
124
		return ZONE_DMA;
125
#endif
126 127
#ifdef CONFIG_ZONE_DMA32
	if (flags & __GFP_DMA32)
128
		return ZONE_DMA32;
129
#endif
M
Mel Gorman 已提交
130 131
	if ((flags & (__GFP_HIGHMEM | __GFP_MOVABLE)) ==
			(__GFP_HIGHMEM | __GFP_MOVABLE))
132
		return ZONE_MOVABLE;
133 134
#ifdef CONFIG_HIGHMEM
	if (flags & __GFP_HIGHMEM)
135
		return ZONE_HIGHMEM;
136
#endif
137
	return ZONE_NORMAL;
138 139
}

L
Linus Torvalds 已提交
140 141 142 143 144 145 146
/*
 * There is only one page-allocator function, and two main namespaces to
 * it. The alloc_page*() variants return 'struct page *' and as such
 * can allocate highmem pages, the *get*page*() variants return
 * virtual kernel addresses to the allocated page(s).
 */

147 148 149 150 151 152 153 154
static inline int gfp_zonelist(gfp_t flags)
{
	if (NUMA_BUILD && unlikely(flags & __GFP_THISNODE))
		return 1;

	return 0;
}

L
Linus Torvalds 已提交
155 156 157
/*
 * We get the zone list from the current node and the gfp_mask.
 * This zone list contains a maximum of MAXNODES*MAX_NR_ZONES zones.
158 159
 * There are two zonelists per node, one for all zones with memory and
 * one containing just zones from the node the zonelist belongs to.
L
Linus Torvalds 已提交
160 161 162 163
 *
 * For the normal case of non-DISCONTIGMEM systems the NODE_DATA() gets
 * optimized to &contig_page_data at compile-time.
 */
164 165
static inline struct zonelist *node_zonelist(int nid, gfp_t flags)
{
166
	return NODE_DATA(nid)->node_zonelists + gfp_zonelist(flags);
167
}
L
Linus Torvalds 已提交
168 169 170 171

#ifndef HAVE_ARCH_FREE_PAGE
static inline void arch_free_page(struct page *page, int order) { }
#endif
N
Nick Piggin 已提交
172 173 174
#ifndef HAVE_ARCH_ALLOC_PAGE
static inline void arch_alloc_page(struct page *page, int order) { }
#endif
L
Linus Torvalds 已提交
175

176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192
struct page *
__alloc_pages_internal(gfp_t gfp_mask, unsigned int order,
		       struct zonelist *zonelist, nodemask_t *nodemask);

static inline struct page *
__alloc_pages(gfp_t gfp_mask, unsigned int order,
		struct zonelist *zonelist)
{
	return __alloc_pages_internal(gfp_mask, order, zonelist, NULL);
}

static inline struct page *
__alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order,
		struct zonelist *zonelist, nodemask_t *nodemask)
{
	return __alloc_pages_internal(gfp_mask, order, zonelist, nodemask);
}
L
Linus Torvalds 已提交
193

194

A
Al Viro 已提交
195
static inline struct page *alloc_pages_node(int nid, gfp_t gfp_mask,
L
Linus Torvalds 已提交
196 197 198 199 200
						unsigned int order)
{
	if (unlikely(order >= MAX_ORDER))
		return NULL;

201 202 203 204
	/* Unknown node is current node */
	if (nid < 0)
		nid = numa_node_id();

205
	return __alloc_pages(gfp_mask, order, node_zonelist(nid, gfp_mask));
L
Linus Torvalds 已提交
206 207 208
}

#ifdef CONFIG_NUMA
A
Al Viro 已提交
209
extern struct page *alloc_pages_current(gfp_t gfp_mask, unsigned order);
L
Linus Torvalds 已提交
210 211

static inline struct page *
A
Al Viro 已提交
212
alloc_pages(gfp_t gfp_mask, unsigned int order)
L
Linus Torvalds 已提交
213 214 215 216 217 218
{
	if (unlikely(order >= MAX_ORDER))
		return NULL;

	return alloc_pages_current(gfp_mask, order);
}
A
Al Viro 已提交
219
extern struct page *alloc_page_vma(gfp_t gfp_mask,
L
Linus Torvalds 已提交
220 221 222 223 224 225 226 227
			struct vm_area_struct *vma, unsigned long addr);
#else
#define alloc_pages(gfp_mask, order) \
		alloc_pages_node(numa_node_id(), gfp_mask, order)
#define alloc_page_vma(gfp_mask, vma, addr) alloc_pages(gfp_mask, 0)
#endif
#define alloc_page(gfp_mask) alloc_pages(gfp_mask, 0)

228 229
extern unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order);
extern unsigned long get_zeroed_page(gfp_t gfp_mask);
L
Linus Torvalds 已提交
230

231 232 233
void *alloc_pages_exact(size_t size, gfp_t gfp_mask);
void free_pages_exact(void *virt, size_t size);

L
Linus Torvalds 已提交
234 235 236 237 238 239
#define __get_free_page(gfp_mask) \
		__get_free_pages((gfp_mask),0)

#define __get_dma_pages(gfp_mask, order) \
		__get_free_pages((gfp_mask) | GFP_DMA,(order))

240 241 242 243
extern void __free_pages(struct page *page, unsigned int order);
extern void free_pages(unsigned long addr, unsigned int order);
extern void free_hot_page(struct page *page);
extern void free_cold_page(struct page *page);
L
Linus Torvalds 已提交
244 245 246 247 248

#define __free_page(page) __free_pages((page), 0)
#define free_page(addr) free_pages((addr),0)

void page_alloc_init(void);
249
void drain_zone_pages(struct zone *zone, struct per_cpu_pages *pcp);
250 251
void drain_all_pages(void);
void drain_local_pages(void *dummy);
L
Linus Torvalds 已提交
252 253

#endif /* __LINUX_GFP_H */