[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 */

/*
 * 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;
	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 */
	};
};

#ifndef CONFIG_HAVE_ARCH_HUGE_VMAP
static inline bool arch_vmap_p4d_supported(pgprot_t prot) { return false; }
static inline bool arch_vmap_pud_supported(pgprot_t prot) { return false; }
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);

#ifdef CONFIG_MMU
extern int map_kernel_range_noflush(unsigned long start, unsigned long size,
				    pgprot_t prot, struct page **pages);
int map_kernel_range(unsigned long start, unsigned long size, pgprot_t prot,
		struct page **pages);
extern void unmap_kernel_range_noflush(unsigned long addr, unsigned long size);
extern void unmap_kernel_range(unsigned long addr, unsigned long size);
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 int
map_kernel_range_noflush(unsigned long start, unsigned long size,
			pgprot_t prot, struct page **pages)
{
	return size >> PAGE_SHIFT;
}
#define map_kernel_range map_kernel_range_noflush
static inline void
unmap_kernel_range_noflush(unsigned long addr, unsigned long size)
{
}
#define unmap_kernel_range unmap_kernel_range_noflush
static inline void set_vm_flush_reset_perms(void *addr)
{
}
#endif

/* for /dev/kmem */
extern long vread(char *buf, char *addr, unsigned long count);
extern long vwrite(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 */
3c5c3cfb DA	29
	30	/*
	31	* VM_KASAN is used slighly differently depending on CONFIG_KASAN_VMALLOC.
	32	*
	33	* If IS_ENABLED(CONFIG_KASAN_VMALLOC), VM_KASAN is set on a vm_struct after
	34	* shadow memory has been mapped. It's used to handle allocation errors so that
	35	* we don't try to poision shadow on free if it was never allocated.
	36	*
	37	* Otherwise, VM_KASAN is set for kasan_module_alloc() allocations and used to
	38	* determine which allocations need the module shadow freed.
	39	*/
	40
1da177e4 LT	41	/* bits [20..32] reserved for arch specific ioremap internals */
1da177e4 LT	42
fd195c49 DS	43	/*
	44	* Maximum alignment for ioremap() regions.
	45	* Can be overriden by arch-specific value.
	46	*/
	47	#ifndef IOREMAP_MAX_ORDER
	48	#define IOREMAP_MAX_ORDER (7 + PAGE_SHIFT) /* 128 pages */
	49	#endif
	50
1da177e4	51	struct vm_struct {
2b4ac44e	52	struct vm_struct *next;
1da177e4 LT	53	void *addr;
	54	unsigned long size;
	55	unsigned long flags;
	56	struct page **pages;
	57	unsigned int nr_pages;
ffa71f33	58	phys_addr_t phys_addr;
5e6cafc8	59	const void *caller;
1da177e4 LT	60	};
1da177e4 LT	61
13ba3fcb AK	62	struct vmap_area {
	63	unsigned long va_start;
	64	unsigned long va_end;
68ad4a33	65
13ba3fcb AK	66	struct rb_node rb_node; /* address sorted rbtree */
13ba3fcb AK	67	struct list_head list; /* address sorted list */
688fcbfc PL	68
688fcbfc PL	69	/*
96e2db45 URS	70	* The following two variables can be packed, because
96e2db45 URS	71	* a vmap_area object can be either:
688fcbfc	72	* 1) in "free" tree (root is vmap_area_root)
96e2db45	73	* 2) or "busy" tree (root is free_vmap_area_root)
688fcbfc PL	74	*/
	75	union {
	76	unsigned long subtree_max_size; /* in "free" tree */
	77	struct vm_struct vm; / in "busy" tree */
688fcbfc	78	};
13ba3fcb AK	79	};
13ba3fcb AK	80
bbc180a5 NP	81	#ifndef CONFIG_HAVE_ARCH_HUGE_VMAP
	82	static inline bool arch_vmap_p4d_supported(pgprot_t prot) { return false; }
	83	static inline bool arch_vmap_pud_supported(pgprot_t prot) { return false; }
	84	static inline bool arch_vmap_pmd_supported(pgprot_t prot) { return false; }
	85	#endif
	86
1da177e4 LT	87	/*
	88	* Highlevel APIs for driver use
	89	*/
db64fe02	90	extern void vm_unmap_ram(const void *mem, unsigned int count);
d4efd79a	91	extern void vm_map_ram(struct page *pages, unsigned int count, int node);
db64fe02 NP	92	extern void vm_unmap_aliases(void);
	93
	94	#ifdef CONFIG_MMU
	95	extern void __init vmalloc_init(void);
97105f0a	96	extern unsigned long vmalloc_nr_pages(void);
db64fe02 NP	97	#else
	98	static inline void vmalloc_init(void)
	99	{
	100	}
97105f0a	101	static inline unsigned long vmalloc_nr_pages(void) { return 0; }
db64fe02 NP	102	#endif
db64fe02 NP	103
1da177e4	104	extern void *vmalloc(unsigned long size);
e1ca7788	105	extern void *vzalloc(unsigned long size);
83342314	106	extern void *vmalloc_user(unsigned long size);
930fc45a	107	extern void *vmalloc_node(unsigned long size, int node);
e1ca7788	108	extern void *vzalloc_node(unsigned long size, int node);
1da177e4	109	extern void *vmalloc_32(unsigned long size);
83342314	110	extern void *vmalloc_32_user(unsigned long size);
88dca4ca	111	extern void *__vmalloc(unsigned long size, gfp_t gfp_mask);
d0a21265 DR	112	extern void *__vmalloc_node_range(unsigned long size, unsigned long align,
d0a21265 DR	113	unsigned long start, unsigned long end, gfp_t gfp_mask,
cb9e3c29 AR	114	pgprot_t prot, unsigned long vm_flags, int node,
cb9e3c29 AR	115	const void *caller);
2b905948 CH	116	void *__vmalloc_node(unsigned long size, unsigned long align, gfp_t gfp_mask,
2b905948 CH	117	int node, const void *caller);
cb9e3c29	118
b3bdda02	119	extern void vfree(const void *addr);
bf22e37a	120	extern void vfree_atomic(const void *addr);
1da177e4 LT	121
	122	extern void vmap(struct page *pages, unsigned int count,
	123	unsigned long flags, pgprot_t prot);
3e9a9e25	124	void vmap_pfn(unsigned long pfns, unsigned int count, pgprot_t prot);
b3bdda02	125	extern void vunmap(const void *addr);
83342314	126
e69e9d4a HD	127	extern int remap_vmalloc_range_partial(struct vm_area_struct *vma,
e69e9d4a HD	128	unsigned long uaddr, void *kaddr,
bdebd6a2	129	unsigned long pgoff, unsigned long size);
e69e9d4a	130
83342314 NP	131	extern int remap_vmalloc_range(struct vm_area_struct vma, void addr,
83342314 NP	132	unsigned long pgoff);
763802b5	133
2ba3e694 JR	134	/*
	135	* Architectures can set this mask to a combination of PGTBL_P?D_MODIFIED values
	136	* and let generic vmalloc and ioremap code know when arch_sync_kernel_mappings()
	137	* needs to be called.
	138	*/
	139	#ifndef ARCH_PAGE_TABLE_SYNC_MASK
	140	#define ARCH_PAGE_TABLE_SYNC_MASK 0
	141	#endif
	142
	143	/*
	144	* There is no default implementation for arch_sync_kernel_mappings(). It is
	145	* relied upon the compiler to optimize calls out if ARCH_PAGE_TABLE_SYNC_MASK
	146	* is 0.
	147	*/
	148	void arch_sync_kernel_mappings(unsigned long start, unsigned long end);
	149
1da177e4 LT	150	/*
	151	* Lowlevel-APIs (not for driver use!)
	152	*/
9585116b JF	153
	154	static inline size_t get_vm_area_size(const struct vm_struct *area)
	155	{
71394fe5 AR	156	if (!(area->flags & VM_NO_GUARD))
	157	/* return actual size without guard page */
	158	return area->size - PAGE_SIZE;
	159	else
	160	return area->size;
	161
9585116b JF	162	}
9585116b JF	163
1da177e4	164	extern struct vm_struct *get_vm_area(unsigned long size, unsigned long flags);
23016969	165	extern struct vm_struct *get_vm_area_caller(unsigned long size,
5e6cafc8	166	unsigned long flags, const void *caller);
c2968612 BH	167	extern struct vm_struct *__get_vm_area_caller(unsigned long size,
	168	unsigned long flags,
	169	unsigned long start, unsigned long end,
5e6cafc8	170	const void *caller);
301fa9f2	171	void free_vm_area(struct vm_struct *area);
b3bdda02	172	extern struct vm_struct remove_vm_area(const void addr);
e9da6e99	173	extern struct vm_struct find_vm_area(const void addr);
c19c03fc	174
b554cb42	175	#ifdef CONFIG_MMU
8fc48985 TH	176	extern int map_kernel_range_noflush(unsigned long start, unsigned long size,
8fc48985 TH	177	pgprot_t prot, struct page **pages);
ed1f324c CH	178	int map_kernel_range(unsigned long start, unsigned long size, pgprot_t prot,
ed1f324c CH	179	struct page **pages);
8fc48985	180	extern void unmap_kernel_range_noflush(unsigned long addr, unsigned long size);
c19c03fc	181	extern void unmap_kernel_range(unsigned long addr, unsigned long size);
868b104d RE	182	static inline void set_vm_flush_reset_perms(void *addr)
	183	{
	184	struct vm_struct *vm = find_vm_area(addr);
	185
	186	if (vm)
	187	vm->flags \|= VM_FLUSH_RESET_PERMS;
	188	}
b554cb42 GY	189	#else
	190	static inline int
	191	map_kernel_range_noflush(unsigned long start, unsigned long size,
	192	pgprot_t prot, struct page **pages)
	193	{
	194	return size >> PAGE_SHIFT;
	195	}
ed1f324c	196	#define map_kernel_range map_kernel_range_noflush
b554cb42 GY	197	static inline void
	198	unmap_kernel_range_noflush(unsigned long addr, unsigned long size)
	199	{
	200	}
ed1f324c	201	#define unmap_kernel_range unmap_kernel_range_noflush
868b104d RE	202	static inline void set_vm_flush_reset_perms(void *addr)
	203	{
	204	}
b554cb42	205	#endif
1da177e4	206
69beeb1d KM	207	/* for /dev/kmem */
	208	extern long vread(char buf, char addr, unsigned long count);
	209	extern long vwrite(char buf, char addr, unsigned long count);
	210
1da177e4 LT	211	/*
	212	* Internals. Dont't use..
	213	*/
f1c4069e	214	extern struct list_head vmap_area_list;
be9b7335	215	extern __init void vm_area_add_early(struct vm_struct *vm);
c0c0a293	216	extern __init void vm_area_register_early(struct vm_struct *vm, size_t align);
1da177e4	217
4f8b02b4	218	#ifdef CONFIG_SMP
b554cb42	219	# ifdef CONFIG_MMU
ca23e405 TH	220	struct vm_struct *pcpu_get_vm_areas(const unsigned long offsets,
ca23e405 TH	221	const size_t *sizes, int nr_vms,
ec3f64fc	222	size_t align);
ca23e405 TH	223
ca23e405 TH	224	void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms);
b554cb42 GY	225	# else
	226	static inline struct vm_struct **
	227	pcpu_get_vm_areas(const unsigned long *offsets,
	228	const size_t *sizes, int nr_vms,
	229	size_t align)
	230	{
	231	return NULL;
	232	}
	233
	234	static inline void
	235	pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms)
	236	{
	237	}
	238	# endif
4f8b02b4	239	#endif
ca23e405	240
db3808c1 JK	241	#ifdef CONFIG_MMU
db3808c1 JK	242	#define VMALLOC_TOTAL (VMALLOC_END - VMALLOC_START)
db3808c1	243	#else
db3808c1	244	#define VMALLOC_TOTAL 0UL
db3808c1 JK	245	#endif
db3808c1 JK	246
4da56b99 CW	247	int register_vmap_purge_notifier(struct notifier_block *nb);
	248	int unregister_vmap_purge_notifier(struct notifier_block *nb);
	249
5bb1bb35	250	#if defined(CONFIG_MMU) && defined(CONFIG_PRINTK)
98f18083 PM	251	bool vmalloc_dump_obj(void *object);
	252	#else
	253	static inline bool vmalloc_dump_obj(void *object) { return false; }
	254	#endif
	255
1da177e4	256	#endif /* _LINUX_VMALLOC_H */