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

#ifndef __LINUX_GFP_H
#define __LINUX_GFP_H

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

struct vm_area_struct;

/*
 * GFP bitmasks..
 *
 * 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.
 */
#define __GFP_DMA	((__force gfp_t)0x01u)
#define __GFP_HIGHMEM	((__force gfp_t)0x02u)
#define __GFP_DMA32	((__force gfp_t)0x04u)
#define __GFP_MOVABLE	((__force gfp_t)0x08u)  /* Page is movable */
#define GFP_ZONEMASK	(__GFP_DMA|__GFP_HIGHMEM|__GFP_DMA32|__GFP_MOVABLE)
/*
 * 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.  This modifier is deprecated and no new
 * users should be added.
 *
 * __GFP_NORETRY: The VM implementation must not retry indefinitely.
 *
 * __GFP_MOVABLE: Flag that this page will be movable by the page migration
 * mechanism or reclaimed
 */
#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 */
#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 */
#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 */
#define __GFP_HARDWALL   ((__force gfp_t)0x20000u) /* Enforce hardwall cpuset memory allocs */
#define __GFP_THISNODE	((__force gfp_t)0x40000u)/* No fallback, no policies */
#define __GFP_RECLAIMABLE ((__force gfp_t)0x80000u) /* Page is reclaimable */

#ifdef CONFIG_KMEMCHECK
#define __GFP_NOTRACK	((__force gfp_t)0x200000u)  /* Don't track with kmemcheck */
#else
#define __GFP_NOTRACK	((__force gfp_t)0)
#endif

/*
 * This may seem redundant, but it's a way of annotating false positives vs.
 * allocations that simply cannot be supported (e.g. page tables).
 */
#define __GFP_NOTRACK_FALSE_POSITIVE (__GFP_NOTRACK)

#define __GFP_BITS_SHIFT 22	/* Room for 22 __GFP_FOO bits */
#define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1))

/* This equals 0, but use constants in case they ever change */
#define GFP_NOWAIT	(GFP_ATOMIC & ~__GFP_HIGH)
/* GFP_ATOMIC means both !wait (__GFP_WAIT not set) and use emergency pool */
#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)
#define GFP_TEMPORARY	(__GFP_WAIT | __GFP_IO | __GFP_FS | \
			 __GFP_RECLAIMABLE)
#define GFP_USER	(__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL)
#define GFP_HIGHUSER	(__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL | \
			 __GFP_HIGHMEM)
#define GFP_HIGHUSER_MOVABLE	(__GFP_WAIT | __GFP_IO | __GFP_FS | \
				 __GFP_HARDWALL | __GFP_HIGHMEM | \
				 __GFP_MOVABLE)
#define GFP_IOFS	(__GFP_IO | __GFP_FS)

#ifdef CONFIG_NUMA
#define GFP_THISNODE	(__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY)
#else
#define GFP_THISNODE	((__force gfp_t)0)
#endif

/* This mask makes up all the page movable related flags */
#define GFP_MOVABLE_MASK (__GFP_RECLAIMABLE|__GFP_MOVABLE)

/* 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 slab gfp mask during early boot */
#define GFP_BOOT_MASK __GFP_BITS_MASK & ~(__GFP_WAIT|__GFP_IO|__GFP_FS)

/* 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)

/* Flag - indicates that the buffer will be suitable for DMA.  Ignored on some
   platforms, used as appropriate on others */

#define GFP_DMA		__GFP_DMA

/* 4GB DMA on some platforms */
#define GFP_DMA32	__GFP_DMA32

/* 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);
}

#ifdef CONFIG_HIGHMEM
#define OPT_ZONE_HIGHMEM ZONE_HIGHMEM
#else
#define OPT_ZONE_HIGHMEM ZONE_NORMAL
#endif

#ifdef CONFIG_ZONE_DMA
#define OPT_ZONE_DMA ZONE_DMA
#else
#define OPT_ZONE_DMA ZONE_NORMAL
#endif

#ifdef CONFIG_ZONE_DMA32
#define OPT_ZONE_DMA32 ZONE_DMA32
#else
#define OPT_ZONE_DMA32 ZONE_NORMAL
#endif

/*
 * GFP_ZONE_TABLE is a word size bitstring that is used for looking up the
 * zone to use given the lowest 4 bits of gfp_t. Entries are ZONE_SHIFT long
 * and there are 16 of them to cover all possible combinations of
 * __GFP_DMA, __GFP_DMA32, __GFP_MOVABLE and __GFP_HIGHMEM
 *
 * The zone fallback order is MOVABLE=>HIGHMEM=>NORMAL=>DMA32=>DMA.
 * But GFP_MOVABLE is not only a zone specifier but also an allocation
 * policy. Therefore __GFP_MOVABLE plus another zone selector is valid.
 * Only 1bit of the lowest 3 bit (DMA,DMA32,HIGHMEM) can be set to "1".
 *
 *       bit       result
 *       =================
 *       0x0    => NORMAL
 *       0x1    => DMA or NORMAL
 *       0x2    => HIGHMEM or NORMAL
 *       0x3    => BAD (DMA+HIGHMEM)
 *       0x4    => DMA32 or DMA or NORMAL
 *       0x5    => BAD (DMA+DMA32)
 *       0x6    => BAD (HIGHMEM+DMA32)
 *       0x7    => BAD (HIGHMEM+DMA32+DMA)
 *       0x8    => NORMAL (MOVABLE+0)
 *       0x9    => DMA or NORMAL (MOVABLE+DMA)
 *       0xa    => MOVABLE (Movable is valid only if HIGHMEM is set too)
 *       0xb    => BAD (MOVABLE+HIGHMEM+DMA)
 *       0xc    => DMA32 (MOVABLE+HIGHMEM+DMA32)
 *       0xd    => BAD (MOVABLE+DMA32+DMA)
 *       0xe    => BAD (MOVABLE+DMA32+HIGHMEM)
 *       0xf    => BAD (MOVABLE+DMA32+HIGHMEM+DMA)
 *
 * ZONES_SHIFT must be <= 2 on 32 bit platforms.
 */

#if 16 * ZONES_SHIFT > BITS_PER_LONG
#error ZONES_SHIFT too large to create GFP_ZONE_TABLE integer
#endif

#define GFP_ZONE_TABLE ( \
	(ZONE_NORMAL << 0 * ZONES_SHIFT)				\
	| (OPT_ZONE_DMA << __GFP_DMA * ZONES_SHIFT) 			\
	| (OPT_ZONE_HIGHMEM << __GFP_HIGHMEM * ZONES_SHIFT)		\
	| (OPT_ZONE_DMA32 << __GFP_DMA32 * ZONES_SHIFT)			\
	| (ZONE_NORMAL << __GFP_MOVABLE * ZONES_SHIFT)			\
	| (OPT_ZONE_DMA << (__GFP_MOVABLE | __GFP_DMA) * ZONES_SHIFT)	\
	| (ZONE_MOVABLE << (__GFP_MOVABLE | __GFP_HIGHMEM) * ZONES_SHIFT)\
	| (OPT_ZONE_DMA32 << (__GFP_MOVABLE | __GFP_DMA32) * ZONES_SHIFT)\
)

/*
 * GFP_ZONE_BAD is a bitmap for all combination of __GFP_DMA, __GFP_DMA32
 * __GFP_HIGHMEM and __GFP_MOVABLE that are not permitted. One flag per
 * entry starting with bit 0. Bit is set if the combination is not
 * allowed.
 */
#define GFP_ZONE_BAD ( \
	1 << (__GFP_DMA | __GFP_HIGHMEM)				\
	| 1 << (__GFP_DMA | __GFP_DMA32)				\
	| 1 << (__GFP_DMA32 | __GFP_HIGHMEM)				\
	| 1 << (__GFP_DMA | __GFP_DMA32 | __GFP_HIGHMEM)		\
	| 1 << (__GFP_MOVABLE | __GFP_HIGHMEM | __GFP_DMA)		\
	| 1 << (__GFP_MOVABLE | __GFP_DMA32 | __GFP_DMA)		\
	| 1 << (__GFP_MOVABLE | __GFP_DMA32 | __GFP_HIGHMEM)		\
	| 1 << (__GFP_MOVABLE | __GFP_DMA32 | __GFP_DMA | __GFP_HIGHMEM)\
)

static inline enum zone_type gfp_zone(gfp_t flags)
{
	enum zone_type z;
	int bit = flags & GFP_ZONEMASK;

	z = (GFP_ZONE_TABLE >> (bit * ZONES_SHIFT)) &
					 ((1 << ZONES_SHIFT) - 1);

	if (__builtin_constant_p(bit))
		MAYBE_BUILD_BUG_ON((GFP_ZONE_BAD >> bit) & 1);
	else {
#ifdef CONFIG_DEBUG_VM
		BUG_ON((GFP_ZONE_BAD >> bit) & 1);
#endif
	}
	return z;
}

/*
 * 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).
 */

static inline int gfp_zonelist(gfp_t flags)
{
	if (NUMA_BUILD && unlikely(flags & __GFP_THISNODE))
		return 1;

	return 0;
}

/*
 * 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.
 * There are two zonelists per node, one for all zones with memory and
 * one containing just zones from the node the zonelist belongs to.
 *
 * For the normal case of non-DISCONTIGMEM systems the NODE_DATA() gets
 * optimized to &contig_page_data at compile-time.
 */
static inline struct zonelist *node_zonelist(int nid, gfp_t flags)
{
	return NODE_DATA(nid)->node_zonelists + gfp_zonelist(flags);
}

#ifndef HAVE_ARCH_FREE_PAGE
static inline void arch_free_page(struct page *page, int order) { }
#endif
#ifndef HAVE_ARCH_ALLOC_PAGE
static inline void arch_alloc_page(struct page *page, int order) { }
#endif

struct page *
__alloc_pages_nodemask(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_nodemask(gfp_mask, order, zonelist, NULL);
}

static inline struct page *alloc_pages_node(int nid, gfp_t gfp_mask,
						unsigned int order)
{
	/* Unknown node is current node */
	if (nid < 0)
		nid = numa_node_id();

	return __alloc_pages(gfp_mask, order, node_zonelist(nid, gfp_mask));
}

static inline struct page *alloc_pages_exact_node(int nid, gfp_t gfp_mask,
						unsigned int order)
{
	VM_BUG_ON(nid < 0 || nid >= MAX_NUMNODES);

	return __alloc_pages(gfp_mask, order, node_zonelist(nid, gfp_mask));
}

#ifdef CONFIG_NUMA
extern struct page *alloc_pages_current(gfp_t gfp_mask, unsigned order);

static inline struct page *
alloc_pages(gfp_t gfp_mask, unsigned int order)
{
	return alloc_pages_current(gfp_mask, order);
}
extern struct page *alloc_page_vma(gfp_t gfp_mask,
			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)

extern unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order);
extern unsigned long get_zeroed_page(gfp_t gfp_mask);

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

#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))

extern void __free_pages(struct page *page, unsigned int order);
extern void free_pages(unsigned long addr, unsigned int order);
extern void free_hot_cold_page(struct page *page, int cold);

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

void page_alloc_init(void);
void drain_zone_pages(struct zone *zone, struct per_cpu_pages *pcp);
void drain_all_pages(void);
void drain_local_pages(void *dummy);

extern gfp_t gfp_allowed_mask;

extern void set_gfp_allowed_mask(gfp_t mask);
extern gfp_t clear_gfp_allowed_mask(gfp_t mask);

#endif /* __LINUX_GFP_H */
Commit	Line	Data
1da177e4 LT	1	#ifndef __LINUX_GFP_H
	2	#define __LINUX_GFP_H
	3
	4	#include <linux/mmzone.h>
	5	#include <linux/stddef.h>
	6	#include <linux/linkage.h>
082edb7b	7	#include <linux/topology.h>
6484eb3e	8	#include <linux/mmdebug.h>
1da177e4 LT	9
	10	struct vm_area_struct;
	11
	12	/*
	13	* GFP bitmasks..
e53ef38d CL	14	*
	15	* Zone modifiers (see linux/mmzone.h - low three bits)
	16	*
e53ef38d CL	17	* Do not put any conditional on these. If necessary modify the definitions
	18	* without the underscores and use the consistently. The definitions here may
	19	* be used in bit comparisons.
1da177e4	20	*/
af4ca457 AV	21	#define __GFP_DMA ((__force gfp_t)0x01u)
af4ca457 AV	22	#define __GFP_HIGHMEM ((__force gfp_t)0x02u)
e53ef38d	23	#define __GFP_DMA32 ((__force gfp_t)0x04u)
b70d94ee CL	24	#define __GFP_MOVABLE ((__force gfp_t)0x08u) /* Page is movable */
b70d94ee CL	25	#define GFP_ZONEMASK (__GFP_DMA\|__GFP_HIGHMEM\|__GFP_DMA32\|__GFP_MOVABLE)
1da177e4 LT	26	/*
	27	* Action modifiers - doesn't change the zoning
	28	*
	29	* __GFP_REPEAT: Try hard to allocate the memory, but the allocation attempt
	30	* _might_ fail. This depends upon the particular VM implementation.
	31	*
	32	* __GFP_NOFAIL: The VM implementation _must_ retry infinitely: the caller
478352e7 DR	33	* cannot handle allocation failures. This modifier is deprecated and no new
478352e7 DR	34	* users should be added.
1da177e4 LT	35	*
1da177e4 LT	36	* __GFP_NORETRY: The VM implementation must not retry indefinitely.
769848c0 MG	37	*
	38	* __GFP_MOVABLE: Flag that this page will be movable by the page migration
	39	* mechanism or reclaimed
1da177e4	40	*/
af4ca457 AV	41	#define __GFP_WAIT ((__force gfp_t)0x10u) /* Can wait and reschedule? */
	42	#define __GFP_HIGH ((__force gfp_t)0x20u) /* Should access emergency pools? */
	43	#define __GFP_IO ((__force gfp_t)0x40u) /* Can start physical IO? */
	44	#define __GFP_FS ((__force gfp_t)0x80u) /* Can call down to low-level FS? */
	45	#define __GFP_COLD ((__force gfp_t)0x100u) /* Cache-cold page required */
	46	#define __GFP_NOWARN ((__force gfp_t)0x200u) /* Suppress page allocation failure warning */
ab857d09 NA	47	#define __GFP_REPEAT ((__force gfp_t)0x400u) /* See above */
	48	#define __GFP_NOFAIL ((__force gfp_t)0x800u) /* See above */
	49	#define __GFP_NORETRY ((__force gfp_t)0x1000u)/* See above */
af4ca457 AV	50	#define __GFP_COMP ((__force gfp_t)0x4000u)/* Add compound page metadata */
	51	#define __GFP_ZERO ((__force gfp_t)0x8000u)/* Return zeroed page on success */
	52	#define __GFP_NOMEMALLOC ((__force gfp_t)0x10000u) /* Don't use emergency reserves */
2d6c666e	53	#define __GFP_HARDWALL ((__force gfp_t)0x20000u) /* Enforce hardwall cpuset memory allocs */
9b819d20	54	#define __GFP_THISNODE ((__force gfp_t)0x40000u)/* No fallback, no policies */
e12ba74d	55	#define __GFP_RECLAIMABLE ((__force gfp_t)0x80000u) /* Page is reclaimable */
b1eeab67 VN	56
b1eeab67 VN	57	#ifdef CONFIG_KMEMCHECK
2dff4405	58	#define __GFP_NOTRACK ((__force gfp_t)0x200000u) /* Don't track with kmemcheck */
b1eeab67 VN	59	#else
	60	#define __GFP_NOTRACK ((__force gfp_t)0)
	61	#endif
1da177e4	62
2dff4405 VN	63	/*
	64	* This may seem redundant, but it's a way of annotating false positives vs.
	65	* allocations that simply cannot be supported (e.g. page tables).
	66	*/
	67	#define __GFP_NOTRACK_FALSE_POSITIVE (__GFP_NOTRACK)
	68
	69	#define __GFP_BITS_SHIFT 22 /* Room for 22 __GFP_FOO bits */
af4ca457	70	#define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
1da177e4	71
7b04d717 JD	72	/* This equals 0, but use constants in case they ever change */
7b04d717 JD	73	#define GFP_NOWAIT (GFP_ATOMIC & ~__GFP_HIGH)
4eac915d	74	/* GFP_ATOMIC means both !wait (__GFP_WAIT not set) and use emergency pool */
1da177e4 LT	75	#define GFP_ATOMIC (__GFP_HIGH)
	76	#define GFP_NOIO (__GFP_WAIT)
	77	#define GFP_NOFS (__GFP_WAIT \| __GFP_IO)
	78	#define GFP_KERNEL (__GFP_WAIT \| __GFP_IO \| __GFP_FS)
e12ba74d MG	79	#define GFP_TEMPORARY (__GFP_WAIT \| __GFP_IO \| __GFP_FS \| \
e12ba74d MG	80	__GFP_RECLAIMABLE)
f90b1d2f PJ	81	#define GFP_USER (__GFP_WAIT \| __GFP_IO \| __GFP_FS \| __GFP_HARDWALL)
	82	#define GFP_HIGHUSER (__GFP_WAIT \| __GFP_IO \| __GFP_FS \| __GFP_HARDWALL \| \
	83	__GFP_HIGHMEM)
769848c0 MG	84	#define GFP_HIGHUSER_MOVABLE (__GFP_WAIT \| __GFP_IO \| __GFP_FS \| \
	85	__GFP_HARDWALL \| __GFP_HIGHMEM \| \
	86	__GFP_MOVABLE)
452aa699	87	#define GFP_IOFS (__GFP_IO \| __GFP_FS)
1da177e4	88
77f700da	89	#ifdef CONFIG_NUMA
980128f2	90	#define GFP_THISNODE (__GFP_THISNODE \| __GFP_NOWARN \| __GFP_NORETRY)
77f700da	91	#else
f2e97df6	92	#define GFP_THISNODE ((__force gfp_t)0)
77f700da CL	93	#endif
77f700da CL	94
6cb06229	95	/* This mask makes up all the page movable related flags */
e12ba74d	96	#define GFP_MOVABLE_MASK (__GFP_RECLAIMABLE\|__GFP_MOVABLE)
6cb06229 CL	97
	98	/* Control page allocator reclaim behavior */
	99	#define GFP_RECLAIM_MASK (__GFP_WAIT\|__GFP_HIGH\|__GFP_IO\|__GFP_FS\|\
	100	__GFP_NOWARN\|__GFP_REPEAT\|__GFP_NOFAIL\|\
	101	__GFP_NORETRY\|__GFP_NOMEMALLOC)
	102
7e85ee0c	103	/* Control slab gfp mask during early boot */
dcce284a	104	#define GFP_BOOT_MASK __GFP_BITS_MASK & ~(__GFP_WAIT\|__GFP_IO\|__GFP_FS)
7e85ee0c	105
6cb06229 CL	106	/* Control allocation constraints */
	107	#define GFP_CONSTRAINT_MASK (__GFP_HARDWALL\|__GFP_THISNODE)
	108
	109	/* Do not use these with a slab allocator */
	110	#define GFP_SLAB_BUG_MASK (__GFP_DMA32\|__GFP_HIGHMEM\|~__GFP_BITS_MASK)
980128f2	111
1da177e4 LT	112	/* Flag - indicates that the buffer will be suitable for DMA. Ignored on some
	113	platforms, used as appropriate on others */
	114
	115	#define GFP_DMA __GFP_DMA
	116
a2f1b424 AK	117	/* 4GB DMA on some platforms */
	118	#define GFP_DMA32 __GFP_DMA32
	119
467c996c MG	120	/* Convert GFP flags to their corresponding migrate type */
	121	static inline int allocflags_to_migratetype(gfp_t gfp_flags)
	122	{
	123	WARN_ON((gfp_flags & GFP_MOVABLE_MASK) == GFP_MOVABLE_MASK);
	124
	125	if (unlikely(page_group_by_mobility_disabled))
	126	return MIGRATE_UNMOVABLE;
	127
	128	/* Group based on mobility */
	129	return (((gfp_flags & __GFP_MOVABLE) != 0) << 1) \|
	130	((gfp_flags & __GFP_RECLAIMABLE) != 0);
	131	}
a2f1b424	132
b70d94ee CL	133	#ifdef CONFIG_HIGHMEM
	134	#define OPT_ZONE_HIGHMEM ZONE_HIGHMEM
	135	#else
	136	#define OPT_ZONE_HIGHMEM ZONE_NORMAL
	137	#endif
	138
4b51d669	139	#ifdef CONFIG_ZONE_DMA
b70d94ee CL	140	#define OPT_ZONE_DMA ZONE_DMA
	141	#else
	142	#define OPT_ZONE_DMA ZONE_NORMAL
4b51d669	143	#endif
b70d94ee	144
4e4785bc	145	#ifdef CONFIG_ZONE_DMA32
b70d94ee CL	146	#define OPT_ZONE_DMA32 ZONE_DMA32
	147	#else
	148	#define OPT_ZONE_DMA32 ZONE_NORMAL
4e4785bc	149	#endif
b70d94ee CL	150
	151	/*
	152	* GFP_ZONE_TABLE is a word size bitstring that is used for looking up the
	153	* zone to use given the lowest 4 bits of gfp_t. Entries are ZONE_SHIFT long
	154	* and there are 16 of them to cover all possible combinations of
	155	* __GFP_DMA, __GFP_DMA32, __GFP_MOVABLE and __GFP_HIGHMEM
	156	*
	157	* The zone fallback order is MOVABLE=>HIGHMEM=>NORMAL=>DMA32=>DMA.
	158	* But GFP_MOVABLE is not only a zone specifier but also an allocation
	159	* policy. Therefore __GFP_MOVABLE plus another zone selector is valid.
	160	* Only 1bit of the lowest 3 bit (DMA,DMA32,HIGHMEM) can be set to "1".
	161	*
	162	* bit result
	163	* =================
	164	* 0x0 => NORMAL
	165	* 0x1 => DMA or NORMAL
	166	* 0x2 => HIGHMEM or NORMAL
	167	* 0x3 => BAD (DMA+HIGHMEM)
	168	* 0x4 => DMA32 or DMA or NORMAL
	169	* 0x5 => BAD (DMA+DMA32)
	170	* 0x6 => BAD (HIGHMEM+DMA32)
	171	* 0x7 => BAD (HIGHMEM+DMA32+DMA)
	172	* 0x8 => NORMAL (MOVABLE+0)
	173	* 0x9 => DMA or NORMAL (MOVABLE+DMA)
	174	* 0xa => MOVABLE (Movable is valid only if HIGHMEM is set too)
	175	* 0xb => BAD (MOVABLE+HIGHMEM+DMA)
	176	* 0xc => DMA32 (MOVABLE+HIGHMEM+DMA32)
	177	* 0xd => BAD (MOVABLE+DMA32+DMA)
	178	* 0xe => BAD (MOVABLE+DMA32+HIGHMEM)
	179	* 0xf => BAD (MOVABLE+DMA32+HIGHMEM+DMA)
	180	*
	181	* ZONES_SHIFT must be <= 2 on 32 bit platforms.
	182	*/
	183
	184	#if 16 * ZONES_SHIFT > BITS_PER_LONG
	185	#error ZONES_SHIFT too large to create GFP_ZONE_TABLE integer
	186	#endif
	187
	188	#define GFP_ZONE_TABLE ( \
	189	(ZONE_NORMAL << 0 * ZONES_SHIFT) \
	190	\| (OPT_ZONE_DMA << __GFP_DMA * ZONES_SHIFT) \
	191	\| (OPT_ZONE_HIGHMEM << __GFP_HIGHMEM * ZONES_SHIFT) \
	192	\| (OPT_ZONE_DMA32 << __GFP_DMA32 * ZONES_SHIFT) \
	193	\| (ZONE_NORMAL << __GFP_MOVABLE * ZONES_SHIFT) \
	194	\| (OPT_ZONE_DMA << (__GFP_MOVABLE \| __GFP_DMA) * ZONES_SHIFT) \
	195	\| (ZONE_MOVABLE << (__GFP_MOVABLE \| __GFP_HIGHMEM) * ZONES_SHIFT)\
	196	\| (OPT_ZONE_DMA32 << (__GFP_MOVABLE \| __GFP_DMA32) * ZONES_SHIFT)\
	197	)
	198
	199	/*
	200	* GFP_ZONE_BAD is a bitmap for all combination of __GFP_DMA, __GFP_DMA32
	201	* __GFP_HIGHMEM and __GFP_MOVABLE that are not permitted. One flag per
	202	* entry starting with bit 0. Bit is set if the combination is not
	203	* allowed.
	204	*/
	205	#define GFP_ZONE_BAD ( \
	206	1 << (__GFP_DMA \| __GFP_HIGHMEM) \
	207	\| 1 << (__GFP_DMA \| __GFP_DMA32) \
	208	\| 1 << (__GFP_DMA32 \| __GFP_HIGHMEM) \
	209	\| 1 << (__GFP_DMA \| __GFP_DMA32 \| __GFP_HIGHMEM) \
	210	\| 1 << (__GFP_MOVABLE \| __GFP_HIGHMEM \| __GFP_DMA) \
	211	\| 1 << (__GFP_MOVABLE \| __GFP_DMA32 \| __GFP_DMA) \
	212	\| 1 << (__GFP_MOVABLE \| __GFP_DMA32 \| __GFP_HIGHMEM) \
	213	\| 1 << (__GFP_MOVABLE \| __GFP_DMA32 \| __GFP_DMA \| __GFP_HIGHMEM)\
214	)
215
216	static inline enum zone_type gfp_zone(gfp_t flags)
217	{
218	enum zone_type z;
219	int bit = flags & GFP_ZONEMASK;
220
221	z = (GFP_ZONE_TABLE >> (bit * ZONES_SHIFT)) &
222	((1 << ZONES_SHIFT) - 1);
223
224	if (__builtin_constant_p(bit))
8c87df45	225	MAYBE_BUILD_BUG_ON((GFP_ZONE_BAD >> bit) & 1);
b70d94ee CL	226	else {
	227	#ifdef CONFIG_DEBUG_VM
	228	BUG_ON((GFP_ZONE_BAD >> bit) & 1);
4e4785bc	229	#endif
b70d94ee CL	230	}
b70d94ee CL	231	return z;
4e4785bc CL	232	}
4e4785bc CL	233
1da177e4 LT	234	/*
	235	* There is only one page-allocator function, and two main namespaces to
	236	* it. The alloc_page() variants return 'struct page ' and as such
	237	* can allocate highmem pages, the getpage*() variants return
	238	* virtual kernel addresses to the allocated page(s).
	239	*/
	240
54a6eb5c MG	241	static inline int gfp_zonelist(gfp_t flags)
	242	{
	243	if (NUMA_BUILD && unlikely(flags & __GFP_THISNODE))
	244	return 1;
	245
	246	return 0;
	247	}
	248
1da177e4 LT	249	/*
	250	* We get the zone list from the current node and the gfp_mask.
	251	* This zone list contains a maximum of MAXNODES*MAX_NR_ZONES zones.
54a6eb5c MG	252	* There are two zonelists per node, one for all zones with memory and
54a6eb5c MG	253	* one containing just zones from the node the zonelist belongs to.
1da177e4 LT	254	*
	255	* For the normal case of non-DISCONTIGMEM systems the NODE_DATA() gets
	256	* optimized to &contig_page_data at compile-time.
	257	*/
0e88460d MG	258	static inline struct zonelist *node_zonelist(int nid, gfp_t flags)
0e88460d MG	259	{
54a6eb5c	260	return NODE_DATA(nid)->node_zonelists + gfp_zonelist(flags);
0e88460d	261	}
1da177e4 LT	262
	263	#ifndef HAVE_ARCH_FREE_PAGE
	264	static inline void arch_free_page(struct page *page, int order) { }
	265	#endif
cc102509 NP	266	#ifndef HAVE_ARCH_ALLOC_PAGE
	267	static inline void arch_alloc_page(struct page *page, int order) { }
	268	#endif
1da177e4	269
e4048e5d	270	struct page *
d239171e	271	__alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order,
e4048e5d KM	272	struct zonelist zonelist, nodemask_t nodemask);
	273
	274	static inline struct page *
	275	__alloc_pages(gfp_t gfp_mask, unsigned int order,
	276	struct zonelist *zonelist)
	277	{
d239171e	278	return __alloc_pages_nodemask(gfp_mask, order, zonelist, NULL);
e4048e5d KM	279	}
e4048e5d KM	280
dd0fc66f	281	static inline struct page *alloc_pages_node(int nid, gfp_t gfp_mask,
1da177e4 LT	282	unsigned int order)
1da177e4 LT	283	{
819a6928 AK	284	/* Unknown node is current node */
	285	if (nid < 0)
	286	nid = numa_node_id();
	287
0e88460d	288	return __alloc_pages(gfp_mask, order, node_zonelist(nid, gfp_mask));
1da177e4 LT	289	}
1da177e4 LT	290
6484eb3e MG	291	static inline struct page *alloc_pages_exact_node(int nid, gfp_t gfp_mask,
	292	unsigned int order)
	293	{
	294	VM_BUG_ON(nid < 0 \|\| nid >= MAX_NUMNODES);
	295
	296	return __alloc_pages(gfp_mask, order, node_zonelist(nid, gfp_mask));
	297	}
	298
1da177e4	299	#ifdef CONFIG_NUMA
dd0fc66f	300	extern struct page *alloc_pages_current(gfp_t gfp_mask, unsigned order);
1da177e4 LT	301
1da177e4 LT	302	static inline struct page *
dd0fc66f	303	alloc_pages(gfp_t gfp_mask, unsigned int order)
1da177e4	304	{
1da177e4 LT	305	return alloc_pages_current(gfp_mask, order);
1da177e4 LT	306	}
dd0fc66f	307	extern struct page *alloc_page_vma(gfp_t gfp_mask,
1da177e4 LT	308	struct vm_area_struct *vma, unsigned long addr);
	309	#else
	310	#define alloc_pages(gfp_mask, order) \
	311	alloc_pages_node(numa_node_id(), gfp_mask, order)
	312	#define alloc_page_vma(gfp_mask, vma, addr) alloc_pages(gfp_mask, 0)
	313	#endif
	314	#define alloc_page(gfp_mask) alloc_pages(gfp_mask, 0)
	315
b3c97528 HH	316	extern unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order);
b3c97528 HH	317	extern unsigned long get_zeroed_page(gfp_t gfp_mask);
1da177e4	318
2be0ffe2 TT	319	void *alloc_pages_exact(size_t size, gfp_t gfp_mask);
	320	void free_pages_exact(void *virt, size_t size);
	321
1da177e4 LT	322	#define __get_free_page(gfp_mask) \
	323	__get_free_pages((gfp_mask),0)
	324
	325	#define __get_dma_pages(gfp_mask, order) \
	326	__get_free_pages((gfp_mask) \| GFP_DMA,(order))
	327
b3c97528 HH	328	extern void __free_pages(struct page *page, unsigned int order);
b3c97528 HH	329	extern void free_pages(unsigned long addr, unsigned int order);
fc91668e	330	extern void free_hot_cold_page(struct page *page, int cold);
1da177e4 LT	331
	332	#define __free_page(page) __free_pages((page), 0)
	333	#define free_page(addr) free_pages((addr),0)
	334
	335	void page_alloc_init(void);
4037d452	336	void drain_zone_pages(struct zone zone, struct per_cpu_pages pcp);
9f8f2172 CL	337	void drain_all_pages(void);
9f8f2172 CL	338	void drain_local_pages(void *dummy);
1da177e4	339
dcce284a BH	340	extern gfp_t gfp_allowed_mask;
dcce284a BH	341
452aa699 RW	342	extern void set_gfp_allowed_mask(gfp_t mask);
452aa699 RW	343	extern gfp_t clear_gfp_allowed_mask(gfp_t mask);
dcce284a	344
1da177e4	345	#endif /* __LINUX_GFP_H */