[mirror_ubuntu-bionic-kernel.git] / include / linux / gfp.h

#ifndef __LINUX_GFP_H
#define __LINUX_GFP_H

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

struct vm_area_struct;

/* Plain integer GFP bitmasks. Do not use this directly. */
#define ___GFP_DMA		0x01u
#define ___GFP_HIGHMEM		0x02u
#define ___GFP_DMA32		0x04u
#define ___GFP_MOVABLE		0x08u
#define ___GFP_WAIT		0x10u
#define ___GFP_HIGH		0x20u
#define ___GFP_IO		0x40u
#define ___GFP_FS		0x80u
#define ___GFP_COLD		0x100u
#define ___GFP_NOWARN		0x200u
#define ___GFP_REPEAT		0x400u
#define ___GFP_NOFAIL		0x800u
#define ___GFP_NORETRY		0x1000u
#define ___GFP_MEMALLOC		0x2000u
#define ___GFP_COMP		0x4000u
#define ___GFP_ZERO		0x8000u
#define ___GFP_NOMEMALLOC	0x10000u
#define ___GFP_HARDWALL		0x20000u
#define ___GFP_THISNODE		0x40000u
#define ___GFP_RECLAIMABLE	0x80000u
#define ___GFP_NOACCOUNT	0x100000u
#define ___GFP_NOTRACK		0x200000u
#define ___GFP_NO_KSWAPD	0x400000u
#define ___GFP_OTHER_NODE	0x800000u
#define ___GFP_WRITE		0x1000000u
/* If the above are modified, __GFP_BITS_SHIFT may need updating */

/*
 * 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 them consistently. The definitions here may
 * be used in bit comparisons.
 */
#define __GFP_DMA	((__force gfp_t)___GFP_DMA)
#define __GFP_HIGHMEM	((__force gfp_t)___GFP_HIGHMEM)
#define __GFP_DMA32	((__force gfp_t)___GFP_DMA32)
#define __GFP_MOVABLE	((__force gfp_t)___GFP_MOVABLE)  /* 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. New users should be evaluated carefully
 * (and the flag should be used only when there is no reasonable failure policy)
 * but it is definitely preferable to use the flag rather than opencode endless
 * loop around allocator.
 *
 * __GFP_NORETRY: The VM implementation must not retry indefinitely and will
 * return NULL when direct reclaim and memory compaction have failed to allow
 * the allocation to succeed.  The OOM killer is not called with the current
 * implementation.
 *
 * __GFP_MOVABLE: Flag that this page will be movable by the page migration
 * mechanism or reclaimed
 */
#define __GFP_WAIT	((__force gfp_t)___GFP_WAIT)	/* Can wait and reschedule? */
#define __GFP_HIGH	((__force gfp_t)___GFP_HIGH)	/* Should access emergency pools? */
#define __GFP_IO	((__force gfp_t)___GFP_IO)	/* Can start physical IO? */
#define __GFP_FS	((__force gfp_t)___GFP_FS)	/* Can call down to low-level FS? */
#define __GFP_COLD	((__force gfp_t)___GFP_COLD)	/* Cache-cold page required */
#define __GFP_NOWARN	((__force gfp_t)___GFP_NOWARN)	/* Suppress page allocation failure warning */
#define __GFP_REPEAT	((__force gfp_t)___GFP_REPEAT)	/* See above */
#define __GFP_NOFAIL	((__force gfp_t)___GFP_NOFAIL)	/* See above */
#define __GFP_NORETRY	((__force gfp_t)___GFP_NORETRY) /* See above */
#define __GFP_MEMALLOC	((__force gfp_t)___GFP_MEMALLOC)/* Allow access to emergency reserves */
#define __GFP_COMP	((__force gfp_t)___GFP_COMP)	/* Add compound page metadata */
#define __GFP_ZERO	((__force gfp_t)___GFP_ZERO)	/* Return zeroed page on success */
#define __GFP_NOMEMALLOC ((__force gfp_t)___GFP_NOMEMALLOC) /* Don't use emergency reserves.
							 * This takes precedence over the
							 * __GFP_MEMALLOC flag if both are
							 * set
							 */
#define __GFP_HARDWALL   ((__force gfp_t)___GFP_HARDWALL) /* Enforce hardwall cpuset memory allocs */
#define __GFP_THISNODE	((__force gfp_t)___GFP_THISNODE)/* No fallback, no policies */
#define __GFP_RECLAIMABLE ((__force gfp_t)___GFP_RECLAIMABLE) /* Page is reclaimable */
#define __GFP_NOACCOUNT	((__force gfp_t)___GFP_NOACCOUNT) /* Don't account to kmemcg */
#define __GFP_NOTRACK	((__force gfp_t)___GFP_NOTRACK)  /* Don't track with kmemcheck */

#define __GFP_NO_KSWAPD	((__force gfp_t)___GFP_NO_KSWAPD)
#define __GFP_OTHER_NODE ((__force gfp_t)___GFP_OTHER_NODE) /* On behalf of other node */
#define __GFP_WRITE	((__force gfp_t)___GFP_WRITE)	/* Allocator intends to dirty page */

/*
 * 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 25	/* Room for N __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_USER | __GFP_HIGHMEM)
#define GFP_HIGHUSER_MOVABLE	(GFP_HIGHUSER | __GFP_MOVABLE)
#define GFP_IOFS	(__GFP_IO | __GFP_FS)
#define GFP_TRANSHUGE	(GFP_HIGHUSER_MOVABLE | __GFP_COMP | \
			 __GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN | \
			 __GFP_NO_KSWAPD)

/* 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_MEMALLOC|__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 gfpflags_to_migratetype(const 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 1 bit of the lowest 3 bits (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+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 combinations 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 = (__force int) (flags & GFP_ZONEMASK);

	z = (GFP_ZONE_TABLE >> (bit * ZONES_SHIFT)) &
					 ((1 << ZONES_SHIFT) - 1);
	VM_BUG_ON((GFP_ZONE_BAD >> bit) & 1);
	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 (IS_ENABLED(CONFIG_NUMA) && 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);
}

/*
 * Allocate pages, preferring the node given as nid. The node must be valid and
 * online. For more general interface, see alloc_pages_node().
 */
static inline struct page *
__alloc_pages_node(int nid, gfp_t gfp_mask, unsigned int order)
{
	VM_BUG_ON(nid < 0 || nid >= MAX_NUMNODES);
	VM_WARN_ON(!node_online(nid));

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

/*
 * Allocate pages, preferring the node given as nid. When nid == NUMA_NO_NODE,
 * prefer the current CPU's node. Otherwise node must be valid and online.
 */
static inline struct page *alloc_pages_node(int nid, gfp_t gfp_mask,
						unsigned int order)
{
	if (nid == NUMA_NO_NODE)
		nid = numa_node_id();

	return __alloc_pages_node(nid, gfp_mask, order);
}

#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_pages_vma(gfp_t gfp_mask, int order,
			struct vm_area_struct *vma, unsigned long addr,
			int node, bool hugepage);
#define alloc_hugepage_vma(gfp_mask, vma, addr, order)	\
	alloc_pages_vma(gfp_mask, order, vma, addr, numa_node_id(), true)
#else
#define alloc_pages(gfp_mask, order) \
		alloc_pages_node(numa_node_id(), gfp_mask, order)
#define alloc_pages_vma(gfp_mask, order, vma, addr, node, false)\
	alloc_pages(gfp_mask, order)
#define alloc_hugepage_vma(gfp_mask, vma, addr, order)	\
	alloc_pages(gfp_mask, order)
#endif
#define alloc_page(gfp_mask) alloc_pages(gfp_mask, 0)
#define alloc_page_vma(gfp_mask, vma, addr)			\
	alloc_pages_vma(gfp_mask, 0, vma, addr, numa_node_id(), false)
#define alloc_page_vma_node(gfp_mask, vma, addr, node)		\
	alloc_pages_vma(gfp_mask, 0, vma, addr, node, false)

extern struct page *alloc_kmem_pages(gfp_t gfp_mask, unsigned int order);
extern struct page *alloc_kmem_pages_node(int nid, gfp_t gfp_mask,
					  unsigned int order);

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);
void * __meminit alloc_pages_exact_nid(int nid, size_t size, gfp_t gfp_mask);

#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, bool cold);
extern void free_hot_cold_page_list(struct list_head *list, bool cold);

struct page_frag_cache;
extern void *__alloc_page_frag(struct page_frag_cache *nc,
			       unsigned int fragsz, gfp_t gfp_mask);
extern void __free_page_frag(void *addr);

extern void __free_kmem_pages(struct page *page, unsigned int order);
extern void free_kmem_pages(unsigned long addr, unsigned int order);

#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(struct zone *zone);
void drain_local_pages(struct zone *zone);

#ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT
void page_alloc_init_late(void);
#else
static inline void page_alloc_init_late(void)
{
}
#endif

/*
 * gfp_allowed_mask is set to GFP_BOOT_MASK during early boot to restrict what
 * GFP flags are used before interrupts are enabled. Once interrupts are
 * enabled, it is set to __GFP_BITS_MASK while the system is running. During
 * hibernation, it is used by PM to avoid I/O during memory allocation while
 * devices are suspended.
 */
extern gfp_t gfp_allowed_mask;

/* Returns true if the gfp_mask allows use of ALLOC_NO_WATERMARK */
bool gfp_pfmemalloc_allowed(gfp_t gfp_mask);

extern void pm_restrict_gfp_mask(void);
extern void pm_restore_gfp_mask(void);

#ifdef CONFIG_PM_SLEEP
extern bool pm_suspended_storage(void);
#else
static inline bool pm_suspended_storage(void)
{
	return false;
}
#endif /* CONFIG_PM_SLEEP */

#ifdef CONFIG_CMA

/* The below functions must be run on a range from a single zone. */
extern int alloc_contig_range(unsigned long start, unsigned long end,
			      unsigned migratetype);
extern void free_contig_range(unsigned long pfn, unsigned nr_pages);

/* CMA stuff */
extern void init_cma_reserved_pageblock(struct page *page);

#endif

#endif /* __LINUX_GFP_H */
Commit	Line	Data
1da177e4 LT	1	#ifndef __LINUX_GFP_H
	2	#define __LINUX_GFP_H
	3
309381fe	4	#include <linux/mmdebug.h>
1da177e4 LT	5	#include <linux/mmzone.h>
	6	#include <linux/stddef.h>
	7	#include <linux/linkage.h>
082edb7b	8	#include <linux/topology.h>
1da177e4 LT	9
	10	struct vm_area_struct;
	11
16b56cf4 NK	12	/* Plain integer GFP bitmasks. Do not use this directly. */
	13	#define ___GFP_DMA 0x01u
	14	#define ___GFP_HIGHMEM 0x02u
	15	#define ___GFP_DMA32 0x04u
	16	#define ___GFP_MOVABLE 0x08u
	17	#define ___GFP_WAIT 0x10u
	18	#define ___GFP_HIGH 0x20u
	19	#define ___GFP_IO 0x40u
	20	#define ___GFP_FS 0x80u
	21	#define ___GFP_COLD 0x100u
	22	#define ___GFP_NOWARN 0x200u
	23	#define ___GFP_REPEAT 0x400u
	24	#define ___GFP_NOFAIL 0x800u
	25	#define ___GFP_NORETRY 0x1000u
b37f1dd0	26	#define ___GFP_MEMALLOC 0x2000u
16b56cf4 NK	27	#define ___GFP_COMP 0x4000u
	28	#define ___GFP_ZERO 0x8000u
	29	#define ___GFP_NOMEMALLOC 0x10000u
	30	#define ___GFP_HARDWALL 0x20000u
	31	#define ___GFP_THISNODE 0x40000u
	32	#define ___GFP_RECLAIMABLE 0x80000u
8f4fc071	33	#define ___GFP_NOACCOUNT 0x100000u
caf49191 LT	34	#define ___GFP_NOTRACK 0x200000u
	35	#define ___GFP_NO_KSWAPD 0x400000u
	36	#define ___GFP_OTHER_NODE 0x800000u
	37	#define ___GFP_WRITE 0x1000000u
05b0afd7	38	/* If the above are modified, __GFP_BITS_SHIFT may need updating */
16b56cf4	39
1da177e4 LT	40	/*
1da177e4 LT	41	* GFP bitmasks..
e53ef38d CL	42	*
	43	* Zone modifiers (see linux/mmzone.h - low three bits)
	44	*
e53ef38d	45	* Do not put any conditional on these. If necessary modify the definitions
263ff5d8	46	* without the underscores and use them consistently. The definitions here may
e53ef38d	47	* be used in bit comparisons.
1da177e4	48	*/
16b56cf4 NK	49	#define __GFP_DMA ((__force gfp_t)___GFP_DMA)
	50	#define __GFP_HIGHMEM ((__force gfp_t)___GFP_HIGHMEM)
	51	#define __GFP_DMA32 ((__force gfp_t)___GFP_DMA32)
	52	#define __GFP_MOVABLE ((__force gfp_t)___GFP_MOVABLE) /* Page is movable */
b70d94ee	53	#define GFP_ZONEMASK (__GFP_DMA\|__GFP_HIGHMEM\|__GFP_DMA32\|__GFP_MOVABLE)
1da177e4 LT	54	/*
	55	* Action modifiers - doesn't change the zoning
	56	*
	57	* __GFP_REPEAT: Try hard to allocate the memory, but the allocation attempt
	58	* _might_ fail. This depends upon the particular VM implementation.
	59	*
	60	* __GFP_NOFAIL: The VM implementation _must_ retry infinitely: the caller
64775719 MH	61	* cannot handle allocation failures. New users should be evaluated carefully
	62	* (and the flag should be used only when there is no reasonable failure policy)
	63	* but it is definitely preferable to use the flag rather than opencode endless
	64	* loop around allocator.
1da177e4	65	*
28c015d0 DR	66	* __GFP_NORETRY: The VM implementation must not retry indefinitely and will
	67	* return NULL when direct reclaim and memory compaction have failed to allow
	68	* the allocation to succeed. The OOM killer is not called with the current
	69	* implementation.
769848c0 MG	70	*
	71	* __GFP_MOVABLE: Flag that this page will be movable by the page migration
	72	* mechanism or reclaimed
1da177e4	73	*/
16b56cf4 NK	74	#define __GFP_WAIT ((__force gfp_t)___GFP_WAIT) /* Can wait and reschedule? */
	75	#define __GFP_HIGH ((__force gfp_t)___GFP_HIGH) /* Should access emergency pools? */
	76	#define __GFP_IO ((__force gfp_t)___GFP_IO) /* Can start physical IO? */
	77	#define __GFP_FS ((__force gfp_t)___GFP_FS) /* Can call down to low-level FS? */
	78	#define __GFP_COLD ((__force gfp_t)___GFP_COLD) /* Cache-cold page required */
	79	#define __GFP_NOWARN ((__force gfp_t)___GFP_NOWARN) /* Suppress page allocation failure warning */
	80	#define __GFP_REPEAT ((__force gfp_t)___GFP_REPEAT) /* See above */
	81	#define __GFP_NOFAIL ((__force gfp_t)___GFP_NOFAIL) /* See above */
	82	#define __GFP_NORETRY ((__force gfp_t)___GFP_NORETRY) /* See above */
b37f1dd0	83	#define __GFP_MEMALLOC ((__force gfp_t)___GFP_MEMALLOC)/* Allow access to emergency reserves */
16b56cf4 NK	84	#define __GFP_COMP ((__force gfp_t)___GFP_COMP) /* Add compound page metadata */
16b56cf4 NK	85	#define __GFP_ZERO ((__force gfp_t)___GFP_ZERO) /* Return zeroed page on success */
b37f1dd0 MG	86	#define __GFP_NOMEMALLOC ((__force gfp_t)___GFP_NOMEMALLOC) /* Don't use emergency reserves.
	87	* This takes precedence over the
	88	* __GFP_MEMALLOC flag if both are
	89	* set
	90	*/
16b56cf4 NK	91	#define __GFP_HARDWALL ((__force gfp_t)___GFP_HARDWALL) /* Enforce hardwall cpuset memory allocs */
	92	#define __GFP_THISNODE ((__force gfp_t)___GFP_THISNODE)/* No fallback, no policies */
	93	#define __GFP_RECLAIMABLE ((__force gfp_t)___GFP_RECLAIMABLE) /* Page is reclaimable */
8f4fc071	94	#define __GFP_NOACCOUNT ((__force gfp_t)___GFP_NOACCOUNT) /* Don't account to kmemcg */
16b56cf4	95	#define __GFP_NOTRACK ((__force gfp_t)___GFP_NOTRACK) /* Don't track with kmemcheck */
1da177e4	96
caf49191	97	#define __GFP_NO_KSWAPD ((__force gfp_t)___GFP_NO_KSWAPD)
78afd561	98	#define __GFP_OTHER_NODE ((__force gfp_t)___GFP_OTHER_NODE) /* On behalf of other node */
a756cf59	99	#define __GFP_WRITE ((__force gfp_t)___GFP_WRITE) /* Allocator intends to dirty page */
32dba98e	100
2dff4405 VN	101	/*
	102	* This may seem redundant, but it's a way of annotating false positives vs.
	103	* allocations that simply cannot be supported (e.g. page tables).
	104	*/
	105	#define __GFP_NOTRACK_FALSE_POSITIVE (__GFP_NOTRACK)
	106
caf49191	107	#define __GFP_BITS_SHIFT 25 /* Room for N __GFP_FOO bits */
af4ca457	108	#define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
1da177e4	109
7b04d717 JD	110	/* This equals 0, but use constants in case they ever change */
7b04d717 JD	111	#define GFP_NOWAIT (GFP_ATOMIC & ~__GFP_HIGH)
4eac915d	112	/* GFP_ATOMIC means both !wait (__GFP_WAIT not set) and use emergency pool */
1da177e4 LT	113	#define GFP_ATOMIC (__GFP_HIGH)
	114	#define GFP_NOIO (__GFP_WAIT)
	115	#define GFP_NOFS (__GFP_WAIT \| __GFP_IO)
	116	#define GFP_KERNEL (__GFP_WAIT \| __GFP_IO \| __GFP_FS)
e12ba74d MG	117	#define GFP_TEMPORARY (__GFP_WAIT \| __GFP_IO \| __GFP_FS \| \
e12ba74d MG	118	__GFP_RECLAIMABLE)
f90b1d2f	119	#define GFP_USER (__GFP_WAIT \| __GFP_IO \| __GFP_FS \| __GFP_HARDWALL)
2d48366b JZ	120	#define GFP_HIGHUSER (GFP_USER \| __GFP_HIGHMEM)
2d48366b JZ	121	#define GFP_HIGHUSER_MOVABLE (GFP_HIGHUSER \| __GFP_MOVABLE)
452aa699	122	#define GFP_IOFS (__GFP_IO \| __GFP_FS)
71e3aac0	123	#define GFP_TRANSHUGE (GFP_HIGHUSER_MOVABLE \| __GFP_COMP \| \
caf49191 LT	124	__GFP_NOMEMALLOC \| __GFP_NORETRY \| __GFP_NOWARN \| \
caf49191 LT	125	__GFP_NO_KSWAPD)
1da177e4	126
6cb06229	127	/* This mask makes up all the page movable related flags */
e12ba74d	128	#define GFP_MOVABLE_MASK (__GFP_RECLAIMABLE\|__GFP_MOVABLE)
6cb06229 CL	129
	130	/* Control page allocator reclaim behavior */
	131	#define GFP_RECLAIM_MASK (__GFP_WAIT\|__GFP_HIGH\|__GFP_IO\|__GFP_FS\|\
	132	__GFP_NOWARN\|__GFP_REPEAT\|__GFP_NOFAIL\|\
b37f1dd0	133	__GFP_NORETRY\|__GFP_MEMALLOC\|__GFP_NOMEMALLOC)
6cb06229	134
7e85ee0c	135	/* Control slab gfp mask during early boot */
fd23855e	136	#define GFP_BOOT_MASK (__GFP_BITS_MASK & ~(__GFP_WAIT\|__GFP_IO\|__GFP_FS))
7e85ee0c	137
6cb06229 CL	138	/* Control allocation constraints */
	139	#define GFP_CONSTRAINT_MASK (__GFP_HARDWALL\|__GFP_THISNODE)
	140
	141	/* Do not use these with a slab allocator */
	142	#define GFP_SLAB_BUG_MASK (__GFP_DMA32\|__GFP_HIGHMEM\|~__GFP_BITS_MASK)
980128f2	143
1da177e4 LT	144	/* Flag - indicates that the buffer will be suitable for DMA. Ignored on some
	145	platforms, used as appropriate on others */
	146
	147	#define GFP_DMA __GFP_DMA
	148
a2f1b424 AK	149	/* 4GB DMA on some platforms */
	150	#define GFP_DMA32 __GFP_DMA32
	151
467c996c	152	/* Convert GFP flags to their corresponding migrate type */
43e7a34d	153	static inline int gfpflags_to_migratetype(const gfp_t gfp_flags)
467c996c MG	154	{
	155	WARN_ON((gfp_flags & GFP_MOVABLE_MASK) == GFP_MOVABLE_MASK);
	156
	157	if (unlikely(page_group_by_mobility_disabled))
	158	return MIGRATE_UNMOVABLE;
	159
	160	/* Group based on mobility */
	161	return (((gfp_flags & __GFP_MOVABLE) != 0) << 1) \|
	162	((gfp_flags & __GFP_RECLAIMABLE) != 0);
	163	}
a2f1b424	164
b70d94ee CL	165	#ifdef CONFIG_HIGHMEM
	166	#define OPT_ZONE_HIGHMEM ZONE_HIGHMEM
	167	#else
	168	#define OPT_ZONE_HIGHMEM ZONE_NORMAL
	169	#endif
	170
4b51d669	171	#ifdef CONFIG_ZONE_DMA
b70d94ee CL	172	#define OPT_ZONE_DMA ZONE_DMA
	173	#else
	174	#define OPT_ZONE_DMA ZONE_NORMAL
4b51d669	175	#endif
b70d94ee	176
4e4785bc	177	#ifdef CONFIG_ZONE_DMA32
b70d94ee CL	178	#define OPT_ZONE_DMA32 ZONE_DMA32
	179	#else
	180	#define OPT_ZONE_DMA32 ZONE_NORMAL
4e4785bc	181	#endif
b70d94ee CL	182
	183	/*
	184	* GFP_ZONE_TABLE is a word size bitstring that is used for looking up the
	185	* zone to use given the lowest 4 bits of gfp_t. Entries are ZONE_SHIFT long
	186	* and there are 16 of them to cover all possible combinations of
263ff5d8	187	* __GFP_DMA, __GFP_DMA32, __GFP_MOVABLE and __GFP_HIGHMEM.
b70d94ee CL	188	*
	189	* The zone fallback order is MOVABLE=>HIGHMEM=>NORMAL=>DMA32=>DMA.
	190	* But GFP_MOVABLE is not only a zone specifier but also an allocation
	191	* policy. Therefore __GFP_MOVABLE plus another zone selector is valid.
263ff5d8	192	* Only 1 bit of the lowest 3 bits (DMA,DMA32,HIGHMEM) can be set to "1".
b70d94ee CL	193	*
	194	* bit result
	195	* =================
	196	* 0x0 => NORMAL
	197	* 0x1 => DMA or NORMAL
	198	* 0x2 => HIGHMEM or NORMAL
	199	* 0x3 => BAD (DMA+HIGHMEM)
	200	* 0x4 => DMA32 or DMA or NORMAL
	201	* 0x5 => BAD (DMA+DMA32)
	202	* 0x6 => BAD (HIGHMEM+DMA32)
	203	* 0x7 => BAD (HIGHMEM+DMA32+DMA)
	204	* 0x8 => NORMAL (MOVABLE+0)
	205	* 0x9 => DMA or NORMAL (MOVABLE+DMA)
	206	* 0xa => MOVABLE (Movable is valid only if HIGHMEM is set too)
	207	* 0xb => BAD (MOVABLE+HIGHMEM+DMA)
537926ca	208	* 0xc => DMA32 (MOVABLE+DMA32)
b70d94ee CL	209	* 0xd => BAD (MOVABLE+DMA32+DMA)
	210	* 0xe => BAD (MOVABLE+DMA32+HIGHMEM)
	211	* 0xf => BAD (MOVABLE+DMA32+HIGHMEM+DMA)
	212	*
	213	* ZONES_SHIFT must be <= 2 on 32 bit platforms.
	214	*/
	215
	216	#if 16 * ZONES_SHIFT > BITS_PER_LONG
	217	#error ZONES_SHIFT too large to create GFP_ZONE_TABLE integer
	218	#endif
	219
	220	#define GFP_ZONE_TABLE ( \
16b56cf4 NK	221	(ZONE_NORMAL << 0 * ZONES_SHIFT) \
	222	\| (OPT_ZONE_DMA << ___GFP_DMA * ZONES_SHIFT) \
	223	\| (OPT_ZONE_HIGHMEM << ___GFP_HIGHMEM * ZONES_SHIFT) \
	224	\| (OPT_ZONE_DMA32 << ___GFP_DMA32 * ZONES_SHIFT) \
	225	\| (ZONE_NORMAL << ___GFP_MOVABLE * ZONES_SHIFT) \
	226	\| (OPT_ZONE_DMA << (___GFP_MOVABLE \| ___GFP_DMA) * ZONES_SHIFT) \
	227	\| (ZONE_MOVABLE << (___GFP_MOVABLE \| ___GFP_HIGHMEM) * ZONES_SHIFT) \
	228	\| (OPT_ZONE_DMA32 << (___GFP_MOVABLE \| ___GFP_DMA32) * ZONES_SHIFT) \
b70d94ee CL	229	)
	230
	231	/*
263ff5d8	232	* GFP_ZONE_BAD is a bitmap for all combinations of __GFP_DMA, __GFP_DMA32
b70d94ee CL	233	* __GFP_HIGHMEM and __GFP_MOVABLE that are not permitted. One flag per
	234	* entry starting with bit 0. Bit is set if the combination is not
	235	* allowed.
	236	*/
	237	#define GFP_ZONE_BAD ( \
16b56cf4 NK	238	1 << (___GFP_DMA \| ___GFP_HIGHMEM) \
	239	\| 1 << (___GFP_DMA \| ___GFP_DMA32) \
	240	\| 1 << (___GFP_DMA32 \| ___GFP_HIGHMEM) \
	241	\| 1 << (___GFP_DMA \| ___GFP_DMA32 \| ___GFP_HIGHMEM) \
	242	\| 1 << (___GFP_MOVABLE \| ___GFP_HIGHMEM \| ___GFP_DMA) \
	243	\| 1 << (___GFP_MOVABLE \| ___GFP_DMA32 \| ___GFP_DMA) \
	244	\| 1 << (___GFP_MOVABLE \| ___GFP_DMA32 \| ___GFP_HIGHMEM) \
	245	\| 1 << (___GFP_MOVABLE \| ___GFP_DMA32 \| ___GFP_DMA \| ___GFP_HIGHMEM) \
b70d94ee CL	246	)
	247
	248	static inline enum zone_type gfp_zone(gfp_t flags)
	249	{
	250	enum zone_type z;
16b56cf4	251	int bit = (__force int) (flags & GFP_ZONEMASK);
b70d94ee CL	252
	253	z = (GFP_ZONE_TABLE >> (bit * ZONES_SHIFT)) &
	254	((1 << ZONES_SHIFT) - 1);
82d4b577	255	VM_BUG_ON((GFP_ZONE_BAD >> bit) & 1);
b70d94ee	256	return z;
4e4785bc CL	257	}
4e4785bc CL	258
1da177e4 LT	259	/*
	260	* There is only one page-allocator function, and two main namespaces to
	261	* it. The alloc_page() variants return 'struct page ' and as such
	262	* can allocate highmem pages, the getpage*() variants return
	263	* virtual kernel addresses to the allocated page(s).
	264	*/
	265
54a6eb5c MG	266	static inline int gfp_zonelist(gfp_t flags)
54a6eb5c MG	267	{
e5adfffc	268	if (IS_ENABLED(CONFIG_NUMA) && unlikely(flags & __GFP_THISNODE))
54a6eb5c MG	269	return 1;
	270
	271	return 0;
	272	}
	273
1da177e4 LT	274	/*
	275	* We get the zone list from the current node and the gfp_mask.
	276	* This zone list contains a maximum of MAXNODES*MAX_NR_ZONES zones.
54a6eb5c MG	277	* There are two zonelists per node, one for all zones with memory and
54a6eb5c MG	278	* one containing just zones from the node the zonelist belongs to.
1da177e4 LT	279	*
	280	* For the normal case of non-DISCONTIGMEM systems the NODE_DATA() gets
	281	* optimized to &contig_page_data at compile-time.
	282	*/
0e88460d MG	283	static inline struct zonelist *node_zonelist(int nid, gfp_t flags)
0e88460d MG	284	{
54a6eb5c	285	return NODE_DATA(nid)->node_zonelists + gfp_zonelist(flags);
0e88460d	286	}
1da177e4 LT	287
	288	#ifndef HAVE_ARCH_FREE_PAGE
	289	static inline void arch_free_page(struct page *page, int order) { }
	290	#endif
cc102509 NP	291	#ifndef HAVE_ARCH_ALLOC_PAGE
	292	static inline void arch_alloc_page(struct page *page, int order) { }
	293	#endif
1da177e4	294
e4048e5d	295	struct page *
d239171e	296	__alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order,
e4048e5d KM	297	struct zonelist zonelist, nodemask_t nodemask);
	298
	299	static inline struct page *
	300	__alloc_pages(gfp_t gfp_mask, unsigned int order,
	301	struct zonelist *zonelist)
	302	{
d239171e	303	return __alloc_pages_nodemask(gfp_mask, order, zonelist, NULL);
e4048e5d KM	304	}
e4048e5d KM	305
96db800f VB	306	/*
	307	* Allocate pages, preferring the node given as nid. The node must be valid and
	308	* online. For more general interface, see alloc_pages_node().
	309	*/
	310	static inline struct page *
	311	__alloc_pages_node(int nid, gfp_t gfp_mask, unsigned int order)
1da177e4	312	{
0bc35a97 VB	313	VM_BUG_ON(nid < 0 \|\| nid >= MAX_NUMNODES);
0bc35a97 VB	314	VM_WARN_ON(!node_online(nid));
819a6928	315
0e88460d	316	return __alloc_pages(gfp_mask, order, node_zonelist(nid, gfp_mask));
1da177e4 LT	317	}
1da177e4 LT	318
96db800f VB	319	/*
96db800f VB	320	* Allocate pages, preferring the node given as nid. When nid == NUMA_NO_NODE,
0bc35a97	321	* prefer the current CPU's node. Otherwise node must be valid and online.
96db800f VB	322	*/
96db800f VB	323	static inline struct page *alloc_pages_node(int nid, gfp_t gfp_mask,
6484eb3e MG	324	unsigned int order)
6484eb3e MG	325	{
0bc35a97	326	if (nid == NUMA_NO_NODE)
96db800f	327	nid = numa_node_id();
6484eb3e	328
0bc35a97	329	return __alloc_pages_node(nid, gfp_mask, order);
6484eb3e MG	330	}
6484eb3e MG	331
1da177e4	332	#ifdef CONFIG_NUMA
dd0fc66f	333	extern struct page *alloc_pages_current(gfp_t gfp_mask, unsigned order);
1da177e4 LT	334
1da177e4 LT	335	static inline struct page *
dd0fc66f	336	alloc_pages(gfp_t gfp_mask, unsigned int order)
1da177e4	337	{
1da177e4 LT	338	return alloc_pages_current(gfp_mask, order);
1da177e4 LT	339	}
0bbbc0b3	340	extern struct page *alloc_pages_vma(gfp_t gfp_mask, int order,
2f5f9486	341	struct vm_area_struct *vma, unsigned long addr,
be97a41b VB	342	int node, bool hugepage);
	343	#define alloc_hugepage_vma(gfp_mask, vma, addr, order) \
	344	alloc_pages_vma(gfp_mask, order, vma, addr, numa_node_id(), true)
1da177e4 LT	345	#else
	346	#define alloc_pages(gfp_mask, order) \
	347	alloc_pages_node(numa_node_id(), gfp_mask, order)
be97a41b	348	#define alloc_pages_vma(gfp_mask, order, vma, addr, node, false)\
0bbbc0b3	349	alloc_pages(gfp_mask, order)
077fcf11 AK	350	#define alloc_hugepage_vma(gfp_mask, vma, addr, order) \
077fcf11 AK	351	alloc_pages(gfp_mask, order)
1da177e4 LT	352	#endif
1da177e4 LT	353	#define alloc_page(gfp_mask) alloc_pages(gfp_mask, 0)
2f5f9486	354	#define alloc_page_vma(gfp_mask, vma, addr) \
be97a41b	355	alloc_pages_vma(gfp_mask, 0, vma, addr, numa_node_id(), false)
236344d6	356	#define alloc_page_vma_node(gfp_mask, vma, addr, node) \
be97a41b	357	alloc_pages_vma(gfp_mask, 0, vma, addr, node, false)
1da177e4	358
52383431 VD	359	extern struct page *alloc_kmem_pages(gfp_t gfp_mask, unsigned int order);
	360	extern struct page *alloc_kmem_pages_node(int nid, gfp_t gfp_mask,
	361	unsigned int order);
	362
b3c97528 HH	363	extern unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order);
b3c97528 HH	364	extern unsigned long get_zeroed_page(gfp_t gfp_mask);
1da177e4	365
2be0ffe2 TT	366	void *alloc_pages_exact(size_t size, gfp_t gfp_mask);
2be0ffe2 TT	367	void free_pages_exact(void *virt, size_t size);
e1931811	368	void * __meminit alloc_pages_exact_nid(int nid, size_t size, gfp_t gfp_mask);
2be0ffe2	369
1da177e4	370	#define __get_free_page(gfp_mask) \
fd23855e	371	__get_free_pages((gfp_mask), 0)
1da177e4 LT	372
1da177e4 LT	373	#define __get_dma_pages(gfp_mask, order) \
fd23855e	374	__get_free_pages((gfp_mask) \| GFP_DMA, (order))
1da177e4	375
b3c97528 HH	376	extern void __free_pages(struct page *page, unsigned int order);
b3c97528 HH	377	extern void free_pages(unsigned long addr, unsigned int order);
b745bc85 MG	378	extern void free_hot_cold_page(struct page *page, bool cold);
b745bc85 MG	379	extern void free_hot_cold_page_list(struct list_head *list, bool cold);
1da177e4	380
b63ae8ca AD	381	struct page_frag_cache;
	382	extern void __alloc_page_frag(struct page_frag_cache nc,
	383	unsigned int fragsz, gfp_t gfp_mask);
	384	extern void __free_page_frag(void *addr);
	385
52383431 VD	386	extern void __free_kmem_pages(struct page *page, unsigned int order);
52383431 VD	387	extern void free_kmem_pages(unsigned long addr, unsigned int order);
6a1a0d3b	388
1da177e4	389	#define __free_page(page) __free_pages((page), 0)
fd23855e	390	#define free_page(addr) free_pages((addr), 0)
1da177e4 LT	391
1da177e4 LT	392	void page_alloc_init(void);
4037d452	393	void drain_zone_pages(struct zone zone, struct per_cpu_pages pcp);
93481ff0 VB	394	void drain_all_pages(struct zone *zone);
93481ff0 VB	395	void drain_local_pages(struct zone *zone);
1da177e4	396
0e1cc95b MG	397	#ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT
	398	void page_alloc_init_late(void);
	399	#else
	400	static inline void page_alloc_init_late(void)
	401	{
	402	}
	403	#endif
	404
f90ac398 MG	405	/*
	406	* gfp_allowed_mask is set to GFP_BOOT_MASK during early boot to restrict what
	407	* GFP flags are used before interrupts are enabled. Once interrupts are
	408	* enabled, it is set to __GFP_BITS_MASK while the system is running. During
	409	* hibernation, it is used by PM to avoid I/O during memory allocation while
	410	* devices are suspended.
	411	*/
dcce284a BH	412	extern gfp_t gfp_allowed_mask;
dcce284a BH	413
c93bdd0e MG	414	/* Returns true if the gfp_mask allows use of ALLOC_NO_WATERMARK */
	415	bool gfp_pfmemalloc_allowed(gfp_t gfp_mask);
	416
c9e664f1 RW	417	extern void pm_restrict_gfp_mask(void);
c9e664f1 RW	418	extern void pm_restore_gfp_mask(void);
dcce284a	419
f90ac398 MG	420	#ifdef CONFIG_PM_SLEEP
	421	extern bool pm_suspended_storage(void);
	422	#else
	423	static inline bool pm_suspended_storage(void)
	424	{
	425	return false;
	426	}
	427	#endif /* CONFIG_PM_SLEEP */
	428
041d3a8c MN	429	#ifdef CONFIG_CMA
	430
	431	/* The below functions must be run on a range from a single zone. */
0815f3d8 MN	432	extern int alloc_contig_range(unsigned long start, unsigned long end,
0815f3d8 MN	433	unsigned migratetype);
041d3a8c MN	434	extern void free_contig_range(unsigned long pfn, unsigned nr_pages);
041d3a8c MN	435
47118af0 MN	436	/* CMA stuff */
	437	extern void init_cma_reserved_pageblock(struct page *page);
	438
041d3a8c MN	439	#endif
041d3a8c MN	440
1da177e4	441	#endif /* __LINUX_GFP_H */