[linux-block.git] / include / linux / vmalloc.h

/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _LINUX_VMALLOC_H
#define _LINUX_VMALLOC_H

#include <linux/spinlock.h>
#include <linux/init.h>
#include <linux/list.h>
#include <linux/llist.h>
#include <asm/page.h>		/* pgprot_t */
#include <linux/rbtree.h>
#include <linux/overflow.h>

#include <asm/vmalloc.h>

struct vm_area_struct;		/* vma defining user mapping in mm_types.h */
struct notifier_block;		/* in notifier.h */

/* bits in flags of vmalloc's vm_struct below */
#define VM_IOREMAP		0x00000001	/* ioremap() and friends */
#define VM_ALLOC		0x00000002	/* vmalloc() */
#define VM_MAP			0x00000004	/* vmap()ed pages */
#define VM_USERMAP		0x00000008	/* suitable for remap_vmalloc_range */
#define VM_DMA_COHERENT		0x00000010	/* dma_alloc_coherent */
#define VM_UNINITIALIZED	0x00000020	/* vm_struct is not fully initialized */
#define VM_NO_GUARD		0x00000040      /* don't add guard page */
#define VM_KASAN		0x00000080      /* has allocated kasan shadow memory */
#define VM_FLUSH_RESET_PERMS	0x00000100	/* reset direct map and flush TLB on unmap, can't be freed in atomic context */
#define VM_MAP_PUT_PAGES	0x00000200	/* put pages and free array in vfree */
#define VM_NO_HUGE_VMAP		0x00000400	/* force PAGE_SIZE pte mapping */

/*
 * VM_KASAN is used slighly differently depending on CONFIG_KASAN_VMALLOC.
 *
 * If IS_ENABLED(CONFIG_KASAN_VMALLOC), VM_KASAN is set on a vm_struct after
 * shadow memory has been mapped. It's used to handle allocation errors so that
 * we don't try to poision shadow on free if it was never allocated.
 *
 * Otherwise, VM_KASAN is set for kasan_module_alloc() allocations and used to
 * determine which allocations need the module shadow freed.
 */

/* bits [20..32] reserved for arch specific ioremap internals */

/*
 * Maximum alignment for ioremap() regions.
 * Can be overriden by arch-specific value.
 */
#ifndef IOREMAP_MAX_ORDER
#define IOREMAP_MAX_ORDER	(7 + PAGE_SHIFT)	/* 128 pages */
#endif

struct vm_struct {
	struct vm_struct	*next;
	void			*addr;
	unsigned long		size;
	unsigned long		flags;
	struct page		**pages;
#ifdef CONFIG_HAVE_ARCH_HUGE_VMALLOC
	unsigned int		page_order;
#endif
	unsigned int		nr_pages;
	phys_addr_t		phys_addr;
	const void		*caller;
};

struct vmap_area {
	unsigned long va_start;
	unsigned long va_end;

	struct rb_node rb_node;         /* address sorted rbtree */
	struct list_head list;          /* address sorted list */

	/*
	 * The following two variables can be packed, because
	 * a vmap_area object can be either:
	 *    1) in "free" tree (root is vmap_area_root)
	 *    2) or "busy" tree (root is free_vmap_area_root)
	 */
	union {
		unsigned long subtree_max_size; /* in "free" tree */
		struct vm_struct *vm;           /* in "busy" tree */
	};
};

/* archs that select HAVE_ARCH_HUGE_VMAP should override one or more of these */
#ifndef arch_vmap_p4d_supported
static inline bool arch_vmap_p4d_supported(pgprot_t prot)
{
	return false;
}
#endif

#ifndef arch_vmap_pud_supported
static inline bool arch_vmap_pud_supported(pgprot_t prot)
{
	return false;
}
#endif

#ifndef arch_vmap_pmd_supported
static inline bool arch_vmap_pmd_supported(pgprot_t prot)
{
	return false;
}
#endif

/*
 *	Highlevel APIs for driver use
 */
extern void vm_unmap_ram(const void *mem, unsigned int count);
extern void *vm_map_ram(struct page **pages, unsigned int count, int node);
extern void vm_unmap_aliases(void);

#ifdef CONFIG_MMU
extern void __init vmalloc_init(void);
extern unsigned long vmalloc_nr_pages(void);
#else
static inline void vmalloc_init(void)
{
}
static inline unsigned long vmalloc_nr_pages(void) { return 0; }
#endif

extern void *vmalloc(unsigned long size);
extern void *vzalloc(unsigned long size);
extern void *vmalloc_user(unsigned long size);
extern void *vmalloc_node(unsigned long size, int node);
extern void *vzalloc_node(unsigned long size, int node);
extern void *vmalloc_32(unsigned long size);
extern void *vmalloc_32_user(unsigned long size);
extern void *__vmalloc(unsigned long size, gfp_t gfp_mask);
extern void *__vmalloc_node_range(unsigned long size, unsigned long align,
			unsigned long start, unsigned long end, gfp_t gfp_mask,
			pgprot_t prot, unsigned long vm_flags, int node,
			const void *caller);
void *__vmalloc_node(unsigned long size, unsigned long align, gfp_t gfp_mask,
		int node, const void *caller);

extern void vfree(const void *addr);
extern void vfree_atomic(const void *addr);

extern void *vmap(struct page **pages, unsigned int count,
			unsigned long flags, pgprot_t prot);
void *vmap_pfn(unsigned long *pfns, unsigned int count, pgprot_t prot);
extern void vunmap(const void *addr);

extern int remap_vmalloc_range_partial(struct vm_area_struct *vma,
				       unsigned long uaddr, void *kaddr,
				       unsigned long pgoff, unsigned long size);

extern int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
							unsigned long pgoff);

/*
 * Architectures can set this mask to a combination of PGTBL_P?D_MODIFIED values
 * and let generic vmalloc and ioremap code know when arch_sync_kernel_mappings()
 * needs to be called.
 */
#ifndef ARCH_PAGE_TABLE_SYNC_MASK
#define ARCH_PAGE_TABLE_SYNC_MASK 0
#endif

/*
 * There is no default implementation for arch_sync_kernel_mappings(). It is
 * relied upon the compiler to optimize calls out if ARCH_PAGE_TABLE_SYNC_MASK
 * is 0.
 */
void arch_sync_kernel_mappings(unsigned long start, unsigned long end);

/*
 *	Lowlevel-APIs (not for driver use!)
 */

static inline size_t get_vm_area_size(const struct vm_struct *area)
{
	if (!(area->flags & VM_NO_GUARD))
		/* return actual size without guard page */
		return area->size - PAGE_SIZE;
	else
		return area->size;

}

extern struct vm_struct *get_vm_area(unsigned long size, unsigned long flags);
extern struct vm_struct *get_vm_area_caller(unsigned long size,
					unsigned long flags, const void *caller);
extern struct vm_struct *__get_vm_area_caller(unsigned long size,
					unsigned long flags,
					unsigned long start, unsigned long end,
					const void *caller);
void free_vm_area(struct vm_struct *area);
extern struct vm_struct *remove_vm_area(const void *addr);
extern struct vm_struct *find_vm_area(const void *addr);

static inline bool is_vm_area_hugepages(const void *addr)
{
	/*
	 * This may not 100% tell if the area is mapped with > PAGE_SIZE
	 * page table entries, if for some reason the architecture indicates
	 * larger sizes are available but decides not to use them, nothing
	 * prevents that. This only indicates the size of the physical page
	 * allocated in the vmalloc layer.
	 */
#ifdef CONFIG_HAVE_ARCH_HUGE_VMALLOC
	return find_vm_area(addr)->page_order > 0;
#else
	return false;
#endif
}

#ifdef CONFIG_MMU
int vmap_range(unsigned long addr, unsigned long end,
			phys_addr_t phys_addr, pgprot_t prot,
			unsigned int max_page_shift);
void vunmap_range(unsigned long addr, unsigned long end);
static inline void set_vm_flush_reset_perms(void *addr)
{
	struct vm_struct *vm = find_vm_area(addr);

	if (vm)
		vm->flags |= VM_FLUSH_RESET_PERMS;
}

#else
static inline void set_vm_flush_reset_perms(void *addr)
{
}
#endif

/* for /proc/kcore */
extern long vread(char *buf, char *addr, unsigned long count);

/*
 *	Internals.  Dont't use..
 */
extern struct list_head vmap_area_list;
extern __init void vm_area_add_early(struct vm_struct *vm);
extern __init void vm_area_register_early(struct vm_struct *vm, size_t align);

#ifdef CONFIG_SMP
# ifdef CONFIG_MMU
struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
				     const size_t *sizes, int nr_vms,
				     size_t align);

void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms);
# else
static inline struct vm_struct **
pcpu_get_vm_areas(const unsigned long *offsets,
		const size_t *sizes, int nr_vms,
		size_t align)
{
	return NULL;
}

static inline void
pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms)
{
}
# endif
#endif

#ifdef CONFIG_MMU
#define VMALLOC_TOTAL (VMALLOC_END - VMALLOC_START)
#else
#define VMALLOC_TOTAL 0UL
#endif

int register_vmap_purge_notifier(struct notifier_block *nb);
int unregister_vmap_purge_notifier(struct notifier_block *nb);

#if defined(CONFIG_MMU) && defined(CONFIG_PRINTK)
bool vmalloc_dump_obj(void *object);
#else
static inline bool vmalloc_dump_obj(void *object) { return false; }
#endif

#endif /* _LINUX_VMALLOC_H */
Commit	Line	Data
b2441318	1	/* SPDX-License-Identifier: GPL-2.0 */
1da177e4 LT	2	#ifndef _LINUX_VMALLOC_H
	3	#define _LINUX_VMALLOC_H
	4
	5	#include <linux/spinlock.h>
db64fe02	6	#include <linux/init.h>
13ba3fcb	7	#include <linux/list.h>
80c4bd7a	8	#include <linux/llist.h>
1da177e4	9	#include <asm/page.h> /* pgprot_t */
13ba3fcb	10	#include <linux/rbtree.h>
3b3b1a29	11	#include <linux/overflow.h>
1da177e4	12
1f059dfd IM	13	#include <asm/vmalloc.h>
1f059dfd IM	14
605d9288	15	struct vm_area_struct; /* vma defining user mapping in mm_types.h */
4da56b99	16	struct notifier_block; /* in notifier.h */
83342314	17
605d9288	18	/* bits in flags of vmalloc's vm_struct below */
20fc02b4 ZY	19	#define VM_IOREMAP 0x00000001 /* ioremap() and friends */
	20	#define VM_ALLOC 0x00000002 /* vmalloc() */
	21	#define VM_MAP 0x00000004 /* vmap()ed pages */
	22	#define VM_USERMAP 0x00000008 /* suitable for remap_vmalloc_range */
fe9041c2	23	#define VM_DMA_COHERENT 0x00000010 /* dma_alloc_coherent */
20fc02b4	24	#define VM_UNINITIALIZED 0x00000020 /* vm_struct is not fully initialized */
71394fe5	25	#define VM_NO_GUARD 0x00000040 /* don't add guard page */
a5af5aa8	26	#define VM_KASAN 0x00000080 /* has allocated kasan shadow memory */
4f6ec860 RE	27	#define VM_FLUSH_RESET_PERMS 0x00000100 /* reset direct map and flush TLB on unmap, can't be freed in atomic context */
4f6ec860 RE	28	#define VM_MAP_PUT_PAGES 0x00000200 /* put pages and free array in vfree */
121e6f32	29	#define VM_NO_HUGE_VMAP 0x00000400 /* force PAGE_SIZE pte mapping */
3c5c3cfb DA	30
	31	/*
	32	* VM_KASAN is used slighly differently depending on CONFIG_KASAN_VMALLOC.
	33	*
	34	* If IS_ENABLED(CONFIG_KASAN_VMALLOC), VM_KASAN is set on a vm_struct after
	35	* shadow memory has been mapped. It's used to handle allocation errors so that
	36	* we don't try to poision shadow on free if it was never allocated.
	37	*
	38	* Otherwise, VM_KASAN is set for kasan_module_alloc() allocations and used to
	39	* determine which allocations need the module shadow freed.
	40	*/
	41
1da177e4 LT	42	/* bits [20..32] reserved for arch specific ioremap internals */
1da177e4 LT	43
fd195c49 DS	44	/*
	45	* Maximum alignment for ioremap() regions.
	46	* Can be overriden by arch-specific value.
	47	*/
	48	#ifndef IOREMAP_MAX_ORDER
	49	#define IOREMAP_MAX_ORDER (7 + PAGE_SHIFT) /* 128 pages */
	50	#endif
	51
1da177e4	52	struct vm_struct {
2b4ac44e	53	struct vm_struct *next;
1da177e4 LT	54	void *addr;
	55	unsigned long size;
	56	unsigned long flags;
	57	struct page **pages;
121e6f32 NP	58	#ifdef CONFIG_HAVE_ARCH_HUGE_VMALLOC
	59	unsigned int page_order;
	60	#endif
1da177e4	61	unsigned int nr_pages;
ffa71f33	62	phys_addr_t phys_addr;
5e6cafc8	63	const void *caller;
1da177e4 LT	64	};
1da177e4 LT	65
13ba3fcb AK	66	struct vmap_area {
	67	unsigned long va_start;
	68	unsigned long va_end;
68ad4a33	69
13ba3fcb AK	70	struct rb_node rb_node; /* address sorted rbtree */
13ba3fcb AK	71	struct list_head list; /* address sorted list */
688fcbfc PL	72
688fcbfc PL	73	/*
96e2db45 URS	74	* The following two variables can be packed, because
96e2db45 URS	75	* a vmap_area object can be either:
688fcbfc	76	* 1) in "free" tree (root is vmap_area_root)
96e2db45	77	* 2) or "busy" tree (root is free_vmap_area_root)
688fcbfc PL	78	*/
	79	union {
	80	unsigned long subtree_max_size; /* in "free" tree */
	81	struct vm_struct vm; / in "busy" tree */
688fcbfc	82	};
13ba3fcb AK	83	};
13ba3fcb AK	84
6f680e70 NP	85	/* archs that select HAVE_ARCH_HUGE_VMAP should override one or more of these */
	86	#ifndef arch_vmap_p4d_supported
	87	static inline bool arch_vmap_p4d_supported(pgprot_t prot)
	88	{
	89	return false;
	90	}
	91	#endif
	92
	93	#ifndef arch_vmap_pud_supported
	94	static inline bool arch_vmap_pud_supported(pgprot_t prot)
	95	{
	96	return false;
	97	}
	98	#endif
	99
	100	#ifndef arch_vmap_pmd_supported
	101	static inline bool arch_vmap_pmd_supported(pgprot_t prot)
	102	{
	103	return false;
	104	}
bbc180a5 NP	105	#endif
bbc180a5 NP	106
1da177e4 LT	107	/*
	108	* Highlevel APIs for driver use
	109	*/
db64fe02	110	extern void vm_unmap_ram(const void *mem, unsigned int count);
d4efd79a	111	extern void vm_map_ram(struct page *pages, unsigned int count, int node);
db64fe02 NP	112	extern void vm_unmap_aliases(void);
	113
	114	#ifdef CONFIG_MMU
	115	extern void __init vmalloc_init(void);
97105f0a	116	extern unsigned long vmalloc_nr_pages(void);
db64fe02 NP	117	#else
	118	static inline void vmalloc_init(void)
	119	{
	120	}
97105f0a	121	static inline unsigned long vmalloc_nr_pages(void) { return 0; }
db64fe02 NP	122	#endif
db64fe02 NP	123
1da177e4	124	extern void *vmalloc(unsigned long size);
e1ca7788	125	extern void *vzalloc(unsigned long size);
83342314	126	extern void *vmalloc_user(unsigned long size);
930fc45a	127	extern void *vmalloc_node(unsigned long size, int node);
e1ca7788	128	extern void *vzalloc_node(unsigned long size, int node);
1da177e4	129	extern void *vmalloc_32(unsigned long size);
83342314	130	extern void *vmalloc_32_user(unsigned long size);
88dca4ca	131	extern void *__vmalloc(unsigned long size, gfp_t gfp_mask);
d0a21265 DR	132	extern void *__vmalloc_node_range(unsigned long size, unsigned long align,
d0a21265 DR	133	unsigned long start, unsigned long end, gfp_t gfp_mask,
cb9e3c29 AR	134	pgprot_t prot, unsigned long vm_flags, int node,
cb9e3c29 AR	135	const void *caller);
2b905948 CH	136	void *__vmalloc_node(unsigned long size, unsigned long align, gfp_t gfp_mask,
2b905948 CH	137	int node, const void *caller);
cb9e3c29	138
b3bdda02	139	extern void vfree(const void *addr);
bf22e37a	140	extern void vfree_atomic(const void *addr);
1da177e4 LT	141
	142	extern void vmap(struct page *pages, unsigned int count,
	143	unsigned long flags, pgprot_t prot);
3e9a9e25	144	void vmap_pfn(unsigned long pfns, unsigned int count, pgprot_t prot);
b3bdda02	145	extern void vunmap(const void *addr);
83342314	146
e69e9d4a HD	147	extern int remap_vmalloc_range_partial(struct vm_area_struct *vma,
e69e9d4a HD	148	unsigned long uaddr, void *kaddr,
bdebd6a2	149	unsigned long pgoff, unsigned long size);
e69e9d4a	150
83342314 NP	151	extern int remap_vmalloc_range(struct vm_area_struct vma, void addr,
83342314 NP	152	unsigned long pgoff);
763802b5	153
2ba3e694 JR	154	/*
	155	* Architectures can set this mask to a combination of PGTBL_P?D_MODIFIED values
	156	* and let generic vmalloc and ioremap code know when arch_sync_kernel_mappings()
	157	* needs to be called.
	158	*/
	159	#ifndef ARCH_PAGE_TABLE_SYNC_MASK
	160	#define ARCH_PAGE_TABLE_SYNC_MASK 0
	161	#endif
	162
	163	/*
	164	* There is no default implementation for arch_sync_kernel_mappings(). It is
	165	* relied upon the compiler to optimize calls out if ARCH_PAGE_TABLE_SYNC_MASK
	166	* is 0.
	167	*/
	168	void arch_sync_kernel_mappings(unsigned long start, unsigned long end);
	169
1da177e4 LT	170	/*
	171	* Lowlevel-APIs (not for driver use!)
	172	*/
9585116b JF	173
	174	static inline size_t get_vm_area_size(const struct vm_struct *area)
	175	{
71394fe5 AR	176	if (!(area->flags & VM_NO_GUARD))
	177	/* return actual size without guard page */
	178	return area->size - PAGE_SIZE;
	179	else
	180	return area->size;
	181
9585116b JF	182	}
9585116b JF	183
1da177e4	184	extern struct vm_struct *get_vm_area(unsigned long size, unsigned long flags);
23016969	185	extern struct vm_struct *get_vm_area_caller(unsigned long size,
5e6cafc8	186	unsigned long flags, const void *caller);
c2968612 BH	187	extern struct vm_struct *__get_vm_area_caller(unsigned long size,
	188	unsigned long flags,
	189	unsigned long start, unsigned long end,
5e6cafc8	190	const void *caller);
301fa9f2	191	void free_vm_area(struct vm_struct *area);
b3bdda02	192	extern struct vm_struct remove_vm_area(const void addr);
e9da6e99	193	extern struct vm_struct find_vm_area(const void addr);
c19c03fc	194
121e6f32 NP	195	static inline bool is_vm_area_hugepages(const void *addr)
	196	{
	197	/*
	198	* This may not 100% tell if the area is mapped with > PAGE_SIZE
	199	* page table entries, if for some reason the architecture indicates
	200	* larger sizes are available but decides not to use them, nothing
	201	* prevents that. This only indicates the size of the physical page
	202	* allocated in the vmalloc layer.
	203	*/
	204	#ifdef CONFIG_HAVE_ARCH_HUGE_VMALLOC
	205	return find_vm_area(addr)->page_order > 0;
	206	#else
	207	return false;
	208	#endif
	209	}
	210
b554cb42	211	#ifdef CONFIG_MMU
5e9e3d77 NP	212	int vmap_range(unsigned long addr, unsigned long end,
	213	phys_addr_t phys_addr, pgprot_t prot,
	214	unsigned int max_page_shift);
4ad0ae8c	215	void vunmap_range(unsigned long addr, unsigned long end);
868b104d RE	216	static inline void set_vm_flush_reset_perms(void *addr)
	217	{
	218	struct vm_struct *vm = find_vm_area(addr);
	219
	220	if (vm)
	221	vm->flags \|= VM_FLUSH_RESET_PERMS;
	222	}
121e6f32	223
b554cb42	224	#else
868b104d RE	225	static inline void set_vm_flush_reset_perms(void *addr)
	226	{
	227	}
b554cb42	228	#endif
1da177e4	229
bbcd53c9	230	/* for /proc/kcore */
69beeb1d	231	extern long vread(char buf, char addr, unsigned long count);
69beeb1d	232
1da177e4 LT	233	/*
	234	* Internals. Dont't use..
	235	*/
f1c4069e	236	extern struct list_head vmap_area_list;
be9b7335	237	extern __init void vm_area_add_early(struct vm_struct *vm);
c0c0a293	238	extern __init void vm_area_register_early(struct vm_struct *vm, size_t align);
1da177e4	239
4f8b02b4	240	#ifdef CONFIG_SMP
b554cb42	241	# ifdef CONFIG_MMU
ca23e405 TH	242	struct vm_struct *pcpu_get_vm_areas(const unsigned long offsets,
ca23e405 TH	243	const size_t *sizes, int nr_vms,
ec3f64fc	244	size_t align);
ca23e405 TH	245
ca23e405 TH	246	void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms);
b554cb42 GY	247	# else
	248	static inline struct vm_struct **
	249	pcpu_get_vm_areas(const unsigned long *offsets,
	250	const size_t *sizes, int nr_vms,
	251	size_t align)
	252	{
	253	return NULL;
	254	}
	255
	256	static inline void
	257	pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms)
	258	{
	259	}
	260	# endif
4f8b02b4	261	#endif
ca23e405	262
db3808c1 JK	263	#ifdef CONFIG_MMU
db3808c1 JK	264	#define VMALLOC_TOTAL (VMALLOC_END - VMALLOC_START)
db3808c1	265	#else
db3808c1	266	#define VMALLOC_TOTAL 0UL
db3808c1 JK	267	#endif
db3808c1 JK	268
4da56b99 CW	269	int register_vmap_purge_notifier(struct notifier_block *nb);
	270	int unregister_vmap_purge_notifier(struct notifier_block *nb);
	271
5bb1bb35	272	#if defined(CONFIG_MMU) && defined(CONFIG_PRINTK)
98f18083 PM	273	bool vmalloc_dump_obj(void *object);
	274	#else
	275	static inline bool vmalloc_dump_obj(void *object) { return false; }
	276	#endif
	277
1da177e4	278	#endif /* _LINUX_VMALLOC_H */