[deliverable/linux.git] / mm / internal.h

/* internal.h: mm/ internal definitions
 *
 * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
 * Written by David Howells (dhowells@redhat.com)
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License
 * as published by the Free Software Foundation; either version
 * 2 of the License, or (at your option) any later version.
 */
#ifndef __MM_INTERNAL_H
#define __MM_INTERNAL_H

#include <linux/mm.h>

void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *start_vma,
		unsigned long floor, unsigned long ceiling);

static inline void set_page_count(struct page *page, int v)
{
	atomic_set(&page->_count, v);
}

/*
 * Turn a non-refcounted page (->_count == 0) into refcounted with
 * a count of one.
 */
static inline void set_page_refcounted(struct page *page)
{
	VM_BUG_ON(PageTail(page));
	VM_BUG_ON(atomic_read(&page->_count));
	set_page_count(page, 1);
}

static inline void __put_page(struct page *page)
{
	atomic_dec(&page->_count);
}

static inline void __get_page_tail_foll(struct page *page,
					bool get_page_head)
{
	/*
	 * If we're getting a tail page, the elevated page->_count is
	 * required only in the head page and we will elevate the head
	 * page->_count and tail page->_mapcount.
	 *
	 * We elevate page_tail->_mapcount for tail pages to force
	 * page_tail->_count to be zero at all times to avoid getting
	 * false positives from get_page_unless_zero() with
	 * speculative page access (like in
	 * page_cache_get_speculative()) on tail pages.
	 */
	VM_BUG_ON(atomic_read(&page->first_page->_count) <= 0);
	VM_BUG_ON(atomic_read(&page->_count) != 0);
	VM_BUG_ON(page_mapcount(page) < 0);
	if (get_page_head)
		atomic_inc(&page->first_page->_count);
	atomic_inc(&page->_mapcount);
}

/*
 * This is meant to be called as the FOLL_GET operation of
 * follow_page() and it must be called while holding the proper PT
 * lock while the pte (or pmd_trans_huge) is still mapping the page.
 */
static inline void get_page_foll(struct page *page)
{
	if (unlikely(PageTail(page)))
		/*
		 * This is safe only because
		 * __split_huge_page_refcount() can't run under
		 * get_page_foll() because we hold the proper PT lock.
		 */
		__get_page_tail_foll(page, true);
	else {
		/*
		 * Getting a normal page or the head of a compound page
		 * requires to already have an elevated page->_count.
		 */
		VM_BUG_ON(atomic_read(&page->_count) <= 0);
		atomic_inc(&page->_count);
	}
}

extern unsigned long highest_memmap_pfn;

/*
 * in mm/vmscan.c:
 */
extern int isolate_lru_page(struct page *page);
extern void putback_lru_page(struct page *page);

/*
 * in mm/page_alloc.c
 */
extern void __free_pages_bootmem(struct page *page, unsigned int order);
extern void prep_compound_page(struct page *page, unsigned long order);
#ifdef CONFIG_MEMORY_FAILURE
extern bool is_free_buddy_page(struct page *page);
#endif


/*
 * function for dealing with page's order in buddy system.
 * zone->lock is already acquired when we use these.
 * So, we don't need atomic page->flags operations here.
 */
static inline unsigned long page_order(struct page *page)
{
	/* PageBuddy() must be checked by the caller */
	return page_private(page);
}

/* mm/util.c */
void __vma_link_list(struct mm_struct *mm, struct vm_area_struct *vma,
		struct vm_area_struct *prev, struct rb_node *rb_parent);

#ifdef CONFIG_MMU
extern long mlock_vma_pages_range(struct vm_area_struct *vma,
			unsigned long start, unsigned long end);
extern void munlock_vma_pages_range(struct vm_area_struct *vma,
			unsigned long start, unsigned long end);
static inline void munlock_vma_pages_all(struct vm_area_struct *vma)
{
	munlock_vma_pages_range(vma, vma->vm_start, vma->vm_end);
}

/*
 * Called only in fault path via page_evictable() for a new page
 * to determine if it's being mapped into a LOCKED vma.
 * If so, mark page as mlocked.
 */
static inline int is_mlocked_vma(struct vm_area_struct *vma, struct page *page)
{
	VM_BUG_ON(PageLRU(page));

	if (likely((vma->vm_flags & (VM_LOCKED | VM_SPECIAL)) != VM_LOCKED))
		return 0;

	if (!TestSetPageMlocked(page)) {
		inc_zone_page_state(page, NR_MLOCK);
		count_vm_event(UNEVICTABLE_PGMLOCKED);
	}
	return 1;
}

/*
 * must be called with vma's mmap_sem held for read or write, and page locked.
 */
extern void mlock_vma_page(struct page *page);
extern void munlock_vma_page(struct page *page);

/*
 * Clear the page's PageMlocked().  This can be useful in a situation where
 * we want to unconditionally remove a page from the pagecache -- e.g.,
 * on truncation or freeing.
 *
 * It is legal to call this function for any page, mlocked or not.
 * If called for a page that is still mapped by mlocked vmas, all we do
 * is revert to lazy LRU behaviour -- semantics are not broken.
 */
extern void __clear_page_mlock(struct page *page);
static inline void clear_page_mlock(struct page *page)
{
	if (unlikely(TestClearPageMlocked(page)))
		__clear_page_mlock(page);
}

/*
 * mlock_migrate_page - called only from migrate_page_copy() to
 * migrate the Mlocked page flag; update statistics.
 */
static inline void mlock_migrate_page(struct page *newpage, struct page *page)
{
	if (TestClearPageMlocked(page)) {
		unsigned long flags;

		local_irq_save(flags);
		__dec_zone_page_state(page, NR_MLOCK);
		SetPageMlocked(newpage);
		__inc_zone_page_state(newpage, NR_MLOCK);
		local_irq_restore(flags);
	}
}

#ifdef CONFIG_TRANSPARENT_HUGEPAGE
extern unsigned long vma_address(struct page *page,
				 struct vm_area_struct *vma);
#endif
#else /* !CONFIG_MMU */
static inline int is_mlocked_vma(struct vm_area_struct *v, struct page *p)
{
	return 0;
}
static inline void clear_page_mlock(struct page *page) { }
static inline void mlock_vma_page(struct page *page) { }
static inline void mlock_migrate_page(struct page *new, struct page *old) { }

#endif /* !CONFIG_MMU */

/*
 * Return the mem_map entry representing the 'offset' subpage within
 * the maximally aligned gigantic page 'base'.  Handle any discontiguity
 * in the mem_map at MAX_ORDER_NR_PAGES boundaries.
 */
static inline struct page *mem_map_offset(struct page *base, int offset)
{
	if (unlikely(offset >= MAX_ORDER_NR_PAGES))
		return pfn_to_page(page_to_pfn(base) + offset);
	return base + offset;
}

/*
 * Iterator over all subpages within the maximally aligned gigantic
 * page 'base'.  Handle any discontiguity in the mem_map.
 */
static inline struct page *mem_map_next(struct page *iter,
						struct page *base, int offset)
{
	if (unlikely((offset & (MAX_ORDER_NR_PAGES - 1)) == 0)) {
		unsigned long pfn = page_to_pfn(base) + offset;
		if (!pfn_valid(pfn))
			return NULL;
		return pfn_to_page(pfn);
	}
	return iter + 1;
}

/*
 * FLATMEM and DISCONTIGMEM configurations use alloc_bootmem_node,
 * so all functions starting at paging_init should be marked __init
 * in those cases. SPARSEMEM, however, allows for memory hotplug,
 * and alloc_bootmem_node is not used.
 */
#ifdef CONFIG_SPARSEMEM
#define __paginginit __meminit
#else
#define __paginginit __init
#endif

/* Memory initialisation debug and verification */
enum mminit_level {
	MMINIT_WARNING,
	MMINIT_VERIFY,
	MMINIT_TRACE
};

#ifdef CONFIG_DEBUG_MEMORY_INIT

extern int mminit_loglevel;

#define mminit_dprintk(level, prefix, fmt, arg...) \
do { \
	if (level < mminit_loglevel) { \
		printk(level <= MMINIT_WARNING ? KERN_WARNING : KERN_DEBUG); \
		printk(KERN_CONT "mminit::" prefix " " fmt, ##arg); \
	} \
} while (0)

extern void mminit_verify_pageflags_layout(void);
extern void mminit_verify_page_links(struct page *page,
		enum zone_type zone, unsigned long nid, unsigned long pfn);
extern void mminit_verify_zonelist(void);

#else

static inline void mminit_dprintk(enum mminit_level level,
				const char *prefix, const char *fmt, ...)
{
}

static inline void mminit_verify_pageflags_layout(void)
{
}

static inline void mminit_verify_page_links(struct page *page,
		enum zone_type zone, unsigned long nid, unsigned long pfn)
{
}

static inline void mminit_verify_zonelist(void)
{
}
#endif /* CONFIG_DEBUG_MEMORY_INIT */

/* mminit_validate_memmodel_limits is independent of CONFIG_DEBUG_MEMORY_INIT */
#if defined(CONFIG_SPARSEMEM)
extern void mminit_validate_memmodel_limits(unsigned long *start_pfn,
				unsigned long *end_pfn);
#else
static inline void mminit_validate_memmodel_limits(unsigned long *start_pfn,
				unsigned long *end_pfn)
{
}
#endif /* CONFIG_SPARSEMEM */

#define ZONE_RECLAIM_NOSCAN	-2
#define ZONE_RECLAIM_FULL	-1
#define ZONE_RECLAIM_SOME	0
#define ZONE_RECLAIM_SUCCESS	1
#endif

extern int hwpoison_filter(struct page *p);

extern u32 hwpoison_filter_dev_major;
extern u32 hwpoison_filter_dev_minor;
extern u64 hwpoison_filter_flags_mask;
extern u64 hwpoison_filter_flags_value;
extern u64 hwpoison_filter_memcg;
extern u32 hwpoison_filter_enable;
Commit	Line	Data
1da177e4 LT	1	/* internal.h: mm/ internal definitions
	2	*
	3	* Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
	4	* Written by David Howells (dhowells@redhat.com)
	5	*
	6	* This program is free software; you can redistribute it and/or
	7	* modify it under the terms of the GNU General Public License
	8	* as published by the Free Software Foundation; either version
	9	* 2 of the License, or (at your option) any later version.
	10	*/
0f8053a5 NP	11	#ifndef __MM_INTERNAL_H
	12	#define __MM_INTERNAL_H
	13
	14	#include <linux/mm.h>
1da177e4	15
42b77728 JB	16	void free_pgtables(struct mmu_gather tlb, struct vm_area_struct start_vma,
	17	unsigned long floor, unsigned long ceiling);
	18
7835e98b	19	static inline void set_page_count(struct page *page, int v)
77a8a788	20	{
7835e98b NP	21	atomic_set(&page->_count, v);
	22	}
	23
	24	/*
	25	* Turn a non-refcounted page (->_count == 0) into refcounted with
	26	* a count of one.
	27	*/
	28	static inline void set_page_refcounted(struct page *page)
	29	{
ae1276b9	30	VM_BUG_ON(PageTail(page));
725d704e	31	VM_BUG_ON(atomic_read(&page->_count));
77a8a788	32	set_page_count(page, 1);
77a8a788 NP	33	}
77a8a788 NP	34
0f8053a5 NP	35	static inline void __put_page(struct page *page)
	36	{
	37	atomic_dec(&page->_count);
	38	}
	39
70b50f94 AA	40	static inline void __get_page_tail_foll(struct page *page,
	41	bool get_page_head)
	42	{
	43	/*
	44	* If we're getting a tail page, the elevated page->_count is
	45	* required only in the head page and we will elevate the head
	46	* page->_count and tail page->_mapcount.
	47	*
	48	* We elevate page_tail->_mapcount for tail pages to force
	49	* page_tail->_count to be zero at all times to avoid getting
	50	* false positives from get_page_unless_zero() with
	51	* speculative page access (like in
	52	* page_cache_get_speculative()) on tail pages.
	53	*/
	54	VM_BUG_ON(atomic_read(&page->first_page->_count) <= 0);
	55	VM_BUG_ON(atomic_read(&page->_count) != 0);
	56	VM_BUG_ON(page_mapcount(page) < 0);
	57	if (get_page_head)
	58	atomic_inc(&page->first_page->_count);
	59	atomic_inc(&page->_mapcount);
	60	}
	61
	62	/*
	63	* This is meant to be called as the FOLL_GET operation of
	64	* follow_page() and it must be called while holding the proper PT
	65	* lock while the pte (or pmd_trans_huge) is still mapping the page.
	66	*/
	67	static inline void get_page_foll(struct page *page)
	68	{
	69	if (unlikely(PageTail(page)))
	70	/*
	71	* This is safe only because
	72	* __split_huge_page_refcount() can't run under
	73	* get_page_foll() because we hold the proper PT lock.
	74	*/
	75	__get_page_tail_foll(page, true);
	76	else {
	77	/*
	78	* Getting a normal page or the head of a compound page
	79	* requires to already have an elevated page->_count.
	80	*/
	81	VM_BUG_ON(atomic_read(&page->_count) <= 0);
	82	atomic_inc(&page->_count);
	83	}
	84	}
	85
03f6462a HD	86	extern unsigned long highest_memmap_pfn;
03f6462a HD	87
894bc310 LS	88	/*
	89	* in mm/vmscan.c:
	90	*/
62695a84	91	extern int isolate_lru_page(struct page *page);
894bc310	92	extern void putback_lru_page(struct page *page);
62695a84	93
894bc310 LS	94	/*
	95	* in mm/page_alloc.c
	96	*/
0c0a4a51	97	extern void __free_pages_bootmem(struct page *page, unsigned int order);
20a0307c	98	extern void prep_compound_page(struct page *page, unsigned long order);
8d22ba1b WF	99	#ifdef CONFIG_MEMORY_FAILURE
	100	extern bool is_free_buddy_page(struct page *page);
	101	#endif
20a0307c	102
0f8053a5	103
48f13bf3 MG	104	/*
	105	* function for dealing with page's order in buddy system.
	106	* zone->lock is already acquired when we use these.
	107	* So, we don't need atomic page->flags operations here.
	108	*/
	109	static inline unsigned long page_order(struct page *page)
	110	{
572438f9	111	/* PageBuddy() must be checked by the caller */
48f13bf3 MG	112	return page_private(page);
48f13bf3 MG	113	}
b5a0e011	114
6038def0 NK	115	/* mm/util.c */
	116	void __vma_link_list(struct mm_struct mm, struct vm_area_struct vma,
	117	struct vm_area_struct prev, struct rb_node rb_parent);
	118
af8e3354 HD	119	#ifdef CONFIG_MMU
	120	extern long mlock_vma_pages_range(struct vm_area_struct *vma,
	121	unsigned long start, unsigned long end);
	122	extern void munlock_vma_pages_range(struct vm_area_struct *vma,
	123	unsigned long start, unsigned long end);
	124	static inline void munlock_vma_pages_all(struct vm_area_struct *vma)
	125	{
	126	munlock_vma_pages_range(vma, vma->vm_start, vma->vm_end);
	127	}
	128
b291f000 NP	129	/*
	130	* Called only in fault path via page_evictable() for a new page
	131	* to determine if it's being mapped into a LOCKED vma.
	132	* If so, mark page as mlocked.
	133	*/
	134	static inline int is_mlocked_vma(struct vm_area_struct vma, struct page page)
	135	{
	136	VM_BUG_ON(PageLRU(page));
	137
	138	if (likely((vma->vm_flags & (VM_LOCKED \| VM_SPECIAL)) != VM_LOCKED))
	139	return 0;
	140
5344b7e6 NP	141	if (!TestSetPageMlocked(page)) {
	142	inc_zone_page_state(page, NR_MLOCK);
	143	count_vm_event(UNEVICTABLE_PGMLOCKED);
	144	}
b291f000 NP	145	return 1;
	146	}
	147
	148	/*
73848b46	149	* must be called with vma's mmap_sem held for read or write, and page locked.
b291f000 NP	150	*/
b291f000 NP	151	extern void mlock_vma_page(struct page *page);
73848b46	152	extern void munlock_vma_page(struct page *page);
b291f000 NP	153
	154	/*
	155	* Clear the page's PageMlocked(). This can be useful in a situation where
	156	* we want to unconditionally remove a page from the pagecache -- e.g.,
	157	* on truncation or freeing.
	158	*
	159	* It is legal to call this function for any page, mlocked or not.
	160	* If called for a page that is still mapped by mlocked vmas, all we do
	161	* is revert to lazy LRU behaviour -- semantics are not broken.
	162	*/
	163	extern void __clear_page_mlock(struct page *page);
	164	static inline void clear_page_mlock(struct page *page)
	165	{
	166	if (unlikely(TestClearPageMlocked(page)))
	167	__clear_page_mlock(page);
	168	}
	169
	170	/*
	171	* mlock_migrate_page - called only from migrate_page_copy() to
5344b7e6	172	* migrate the Mlocked page flag; update statistics.
b291f000 NP	173	*/
	174	static inline void mlock_migrate_page(struct page newpage, struct page page)
	175	{
5344b7e6 NP	176	if (TestClearPageMlocked(page)) {
	177	unsigned long flags;
	178
	179	local_irq_save(flags);
	180	__dec_zone_page_state(page, NR_MLOCK);
b291f000	181	SetPageMlocked(newpage);
5344b7e6 NP	182	__inc_zone_page_state(newpage, NR_MLOCK);
	183	local_irq_restore(flags);
	184	}
b291f000 NP	185	}
b291f000 NP	186
71e3aac0 AA	187	#ifdef CONFIG_TRANSPARENT_HUGEPAGE
	188	extern unsigned long vma_address(struct page *page,
	189	struct vm_area_struct *vma);
	190	#endif
af8e3354	191	#else /* !CONFIG_MMU */
b291f000 NP	192	static inline int is_mlocked_vma(struct vm_area_struct v, struct page p)
	193	{
	194	return 0;
	195	}
	196	static inline void clear_page_mlock(struct page *page) { }
	197	static inline void mlock_vma_page(struct page *page) { }
	198	static inline void mlock_migrate_page(struct page new, struct page old) { }
	199
af8e3354	200	#endif /* !CONFIG_MMU */
894bc310	201
69d177c2 AW	202	/*
	203	* Return the mem_map entry representing the 'offset' subpage within
	204	* the maximally aligned gigantic page 'base'. Handle any discontiguity
	205	* in the mem_map at MAX_ORDER_NR_PAGES boundaries.
	206	*/
	207	static inline struct page mem_map_offset(struct page base, int offset)
	208	{
	209	if (unlikely(offset >= MAX_ORDER_NR_PAGES))
	210	return pfn_to_page(page_to_pfn(base) + offset);
	211	return base + offset;
	212	}
	213
	214	/*
25985edc	215	* Iterator over all subpages within the maximally aligned gigantic
69d177c2 AW	216	* page 'base'. Handle any discontiguity in the mem_map.
	217	*/
	218	static inline struct page mem_map_next(struct page iter,
	219	struct page *base, int offset)
	220	{
	221	if (unlikely((offset & (MAX_ORDER_NR_PAGES - 1)) == 0)) {
	222	unsigned long pfn = page_to_pfn(base) + offset;
	223	if (!pfn_valid(pfn))
	224	return NULL;
	225	return pfn_to_page(pfn);
	226	}
	227	return iter + 1;
	228	}
	229
b5a0e011 AH	230	/*
	231	* FLATMEM and DISCONTIGMEM configurations use alloc_bootmem_node,
	232	* so all functions starting at paging_init should be marked __init
	233	* in those cases. SPARSEMEM, however, allows for memory hotplug,
	234	* and alloc_bootmem_node is not used.
	235	*/
	236	#ifdef CONFIG_SPARSEMEM
	237	#define __paginginit __meminit
	238	#else
	239	#define __paginginit __init
	240	#endif
	241
6b74ab97 MG	242	/* Memory initialisation debug and verification */
	243	enum mminit_level {
	244	MMINIT_WARNING,
	245	MMINIT_VERIFY,
	246	MMINIT_TRACE
	247	};
	248
	249	#ifdef CONFIG_DEBUG_MEMORY_INIT
	250
	251	extern int mminit_loglevel;
	252
	253	#define mminit_dprintk(level, prefix, fmt, arg...) \
	254	do { \
	255	if (level < mminit_loglevel) { \
	256	printk(level <= MMINIT_WARNING ? KERN_WARNING : KERN_DEBUG); \
	257	printk(KERN_CONT "mminit::" prefix " " fmt, ##arg); \
	258	} \
	259	} while (0)
	260
708614e6 MG	261	extern void mminit_verify_pageflags_layout(void);
	262	extern void mminit_verify_page_links(struct page *page,
	263	enum zone_type zone, unsigned long nid, unsigned long pfn);
68ad8df4	264	extern void mminit_verify_zonelist(void);
708614e6	265
6b74ab97 MG	266	#else
	267
	268	static inline void mminit_dprintk(enum mminit_level level,
	269	const char prefix, const char fmt, ...)
	270	{
	271	}
	272
708614e6 MG	273	static inline void mminit_verify_pageflags_layout(void)
	274	{
	275	}
	276
	277	static inline void mminit_verify_page_links(struct page *page,
	278	enum zone_type zone, unsigned long nid, unsigned long pfn)
	279	{
	280	}
68ad8df4 MG	281
	282	static inline void mminit_verify_zonelist(void)
	283	{
	284	}
6b74ab97	285	#endif /* CONFIG_DEBUG_MEMORY_INIT */
2dbb51c4 MG	286
	287	/* mminit_validate_memmodel_limits is independent of CONFIG_DEBUG_MEMORY_INIT */
	288	#if defined(CONFIG_SPARSEMEM)
	289	extern void mminit_validate_memmodel_limits(unsigned long *start_pfn,
	290	unsigned long *end_pfn);
	291	#else
	292	static inline void mminit_validate_memmodel_limits(unsigned long *start_pfn,
	293	unsigned long *end_pfn)
	294	{
	295	}
	296	#endif /* CONFIG_SPARSEMEM */
	297
fa5e084e MG	298	#define ZONE_RECLAIM_NOSCAN -2
	299	#define ZONE_RECLAIM_FULL -1
	300	#define ZONE_RECLAIM_SOME 0
	301	#define ZONE_RECLAIM_SUCCESS 1
0f8053a5	302	#endif
7c116f2b	303
31d3d348 WF	304	extern int hwpoison_filter(struct page *p);
31d3d348 WF	305
7c116f2b WF	306	extern u32 hwpoison_filter_dev_major;
7c116f2b WF	307	extern u32 hwpoison_filter_dev_minor;
478c5ffc WF	308	extern u64 hwpoison_filter_flags_mask;
478c5ffc WF	309	extern u64 hwpoison_filter_flags_value;
4fd466eb	310	extern u64 hwpoison_filter_memcg;
1bfe5feb	311	extern u32 hwpoison_filter_enable;