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

/* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of version 2 of the GNU General Public
 * License as published by the Free Software Foundation.
 */
#ifndef _LINUX_BPF_H
#define _LINUX_BPF_H 1

#include <uapi/linux/bpf.h>
#include <linux/workqueue.h>
#include <linux/file.h>
#include <linux/percpu.h>
#include <linux/err.h>

struct bpf_map;

/* map is generic key/value storage optionally accesible by eBPF programs */
struct bpf_map_ops {
	/* funcs callable from userspace (via syscall) */
	struct bpf_map *(*map_alloc)(union bpf_attr *attr);
	void (*map_release)(struct bpf_map *map, struct file *map_file);
	void (*map_free)(struct bpf_map *map);
	int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key);

	/* funcs callable from userspace and from eBPF programs */
	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
	int (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags);
	int (*map_delete_elem)(struct bpf_map *map, void *key);

	/* funcs called by prog_array and perf_event_array map */
	void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file,
				int fd);
	void (*map_fd_put_ptr)(void *ptr);
};

struct bpf_map {
	atomic_t refcnt;
	enum bpf_map_type map_type;
	u32 key_size;
	u32 value_size;
	u32 max_entries;
	u32 map_flags;
	u32 pages;
	struct user_struct *user;
	const struct bpf_map_ops *ops;
	struct work_struct work;
	atomic_t usercnt;
};

struct bpf_map_type_list {
	struct list_head list_node;
	const struct bpf_map_ops *ops;
	enum bpf_map_type type;
};

/* function argument constraints */
enum bpf_arg_type {
	ARG_DONTCARE = 0,	/* unused argument in helper function */

	/* the following constraints used to prototype
	 * bpf_map_lookup/update/delete_elem() functions
	 */
	ARG_CONST_MAP_PTR,	/* const argument used as pointer to bpf_map */
	ARG_PTR_TO_MAP_KEY,	/* pointer to stack used as map key */
	ARG_PTR_TO_MAP_VALUE,	/* pointer to stack used as map value */

	/* the following constraints used to prototype bpf_memcmp() and other
	 * functions that access data on eBPF program stack
	 */
	ARG_PTR_TO_STACK,	/* any pointer to eBPF program stack */
	ARG_PTR_TO_RAW_STACK,	/* any pointer to eBPF program stack, area does not
				 * need to be initialized, helper function must fill
				 * all bytes or clear them in error case.
				 */

	ARG_CONST_STACK_SIZE,	/* number of bytes accessed from stack */
	ARG_CONST_STACK_SIZE_OR_ZERO, /* number of bytes accessed from stack or 0 */

	ARG_PTR_TO_CTX,		/* pointer to context */
	ARG_ANYTHING,		/* any (initialized) argument is ok */
};

/* type of values returned from helper functions */
enum bpf_return_type {
	RET_INTEGER,			/* function returns integer */
	RET_VOID,			/* function doesn't return anything */
	RET_PTR_TO_MAP_VALUE_OR_NULL,	/* returns a pointer to map elem value or NULL */
};

/* eBPF function prototype used by verifier to allow BPF_CALLs from eBPF programs
 * to in-kernel helper functions and for adjusting imm32 field in BPF_CALL
 * instructions after verifying
 */
struct bpf_func_proto {
	u64 (*func)(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
	bool gpl_only;
	enum bpf_return_type ret_type;
	enum bpf_arg_type arg1_type;
	enum bpf_arg_type arg2_type;
	enum bpf_arg_type arg3_type;
	enum bpf_arg_type arg4_type;
	enum bpf_arg_type arg5_type;
};

/* bpf_context is intentionally undefined structure. Pointer to bpf_context is
 * the first argument to eBPF programs.
 * For socket filters: 'struct bpf_context *' == 'struct sk_buff *'
 */
struct bpf_context;

enum bpf_access_type {
	BPF_READ = 1,
	BPF_WRITE = 2
};

struct bpf_prog;

struct bpf_verifier_ops {
	/* return eBPF function prototype for verification */
	const struct bpf_func_proto *(*get_func_proto)(enum bpf_func_id func_id);

	/* return true if 'size' wide access at offset 'off' within bpf_context
	 * with 'type' (read or write) is allowed
	 */
	bool (*is_valid_access)(int off, int size, enum bpf_access_type type);

	u32 (*convert_ctx_access)(enum bpf_access_type type, int dst_reg,
				  int src_reg, int ctx_off,
				  struct bpf_insn *insn, struct bpf_prog *prog);
};

struct bpf_prog_type_list {
	struct list_head list_node;
	const struct bpf_verifier_ops *ops;
	enum bpf_prog_type type;
};

struct bpf_prog_aux {
	atomic_t refcnt;
	u32 used_map_cnt;
	u32 max_ctx_offset;
	const struct bpf_verifier_ops *ops;
	struct bpf_map **used_maps;
	struct bpf_prog *prog;
	struct user_struct *user;
	union {
		struct work_struct work;
		struct rcu_head	rcu;
	};
};

struct bpf_array {
	struct bpf_map map;
	u32 elem_size;
	/* 'ownership' of prog_array is claimed by the first program that
	 * is going to use this map or by the first program which FD is stored
	 * in the map to make sure that all callers and callees have the same
	 * prog_type and JITed flag
	 */
	enum bpf_prog_type owner_prog_type;
	bool owner_jited;
	union {
		char value[0] __aligned(8);
		void *ptrs[0] __aligned(8);
		void __percpu *pptrs[0] __aligned(8);
	};
};
#define MAX_TAIL_CALL_CNT 32

u64 bpf_tail_call(u64 ctx, u64 r2, u64 index, u64 r4, u64 r5);
u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);

bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp);

const struct bpf_func_proto *bpf_get_trace_printk_proto(void);
const struct bpf_func_proto *bpf_get_event_output_proto(void);

#ifdef CONFIG_BPF_SYSCALL
DECLARE_PER_CPU(int, bpf_prog_active);

void bpf_register_prog_type(struct bpf_prog_type_list *tl);
void bpf_register_map_type(struct bpf_map_type_list *tl);

struct bpf_prog *bpf_prog_get(u32 ufd);
struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog);
void bpf_prog_put(struct bpf_prog *prog);
void bpf_prog_put_rcu(struct bpf_prog *prog);

struct bpf_map *bpf_map_get_with_uref(u32 ufd);
struct bpf_map *__bpf_map_get(struct fd f);
struct bpf_map *bpf_map_inc(struct bpf_map *map, bool uref);
void bpf_map_put_with_uref(struct bpf_map *map);
void bpf_map_put(struct bpf_map *map);
int bpf_map_precharge_memlock(u32 pages);

extern int sysctl_unprivileged_bpf_disabled;

int bpf_map_new_fd(struct bpf_map *map);
int bpf_prog_new_fd(struct bpf_prog *prog);

int bpf_obj_pin_user(u32 ufd, const char __user *pathname);
int bpf_obj_get_user(const char __user *pathname);

int bpf_percpu_hash_copy(struct bpf_map *map, void *key, void *value);
int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value);
int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value,
			   u64 flags);
int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value,
			    u64 flags);

int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value);

int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file,
				 void *key, void *value, u64 map_flags);
void bpf_fd_array_map_clear(struct bpf_map *map);

/* memcpy that is used with 8-byte aligned pointers, power-of-8 size and
 * forced to use 'long' read/writes to try to atomically copy long counters.
 * Best-effort only.  No barriers here, since it _will_ race with concurrent
 * updates from BPF programs. Called from bpf syscall and mostly used with
 * size 8 or 16 bytes, so ask compiler to inline it.
 */
static inline void bpf_long_memcpy(void *dst, const void *src, u32 size)
{
	const long *lsrc = src;
	long *ldst = dst;

	size /= sizeof(long);
	while (size--)
		*ldst++ = *lsrc++;
}

/* verify correctness of eBPF program */
int bpf_check(struct bpf_prog **fp, union bpf_attr *attr);
#else
static inline void bpf_register_prog_type(struct bpf_prog_type_list *tl)
{
}

static inline struct bpf_prog *bpf_prog_get(u32 ufd)
{
	return ERR_PTR(-EOPNOTSUPP);
}

static inline void bpf_prog_put(struct bpf_prog *prog)
{
}
#endif /* CONFIG_BPF_SYSCALL */

/* verifier prototypes for helper functions called from eBPF programs */
extern const struct bpf_func_proto bpf_map_lookup_elem_proto;
extern const struct bpf_func_proto bpf_map_update_elem_proto;
extern const struct bpf_func_proto bpf_map_delete_elem_proto;

extern const struct bpf_func_proto bpf_get_prandom_u32_proto;
extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
extern const struct bpf_func_proto bpf_tail_call_proto;
extern const struct bpf_func_proto bpf_ktime_get_ns_proto;
extern const struct bpf_func_proto bpf_get_current_pid_tgid_proto;
extern const struct bpf_func_proto bpf_get_current_uid_gid_proto;
extern const struct bpf_func_proto bpf_get_current_comm_proto;
extern const struct bpf_func_proto bpf_skb_vlan_push_proto;
extern const struct bpf_func_proto bpf_skb_vlan_pop_proto;
extern const struct bpf_func_proto bpf_get_stackid_proto;

/* Shared helpers among cBPF and eBPF. */
void bpf_user_rnd_init_once(void);
u64 bpf_user_rnd_u32(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);

#endif /* _LINUX_BPF_H */
Commit	Line	Data
99c55f7d AS	1	/* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com
	2	*
	3	* This program is free software; you can redistribute it and/or
	4	* modify it under the terms of version 2 of the GNU General Public
	5	* License as published by the Free Software Foundation.
	6	*/
	7	#ifndef _LINUX_BPF_H
	8	#define _LINUX_BPF_H 1
	9
	10	#include <uapi/linux/bpf.h>
	11	#include <linux/workqueue.h>
db20fd2b	12	#include <linux/file.h>
b121d1e7	13	#include <linux/percpu.h>
002245cc	14	#include <linux/err.h>
99c55f7d AS	15
	16	struct bpf_map;
	17
	18	/* map is generic key/value storage optionally accesible by eBPF programs */
	19	struct bpf_map_ops {
	20	/* funcs callable from userspace (via syscall) */
	21	struct bpf_map (map_alloc)(union bpf_attr *attr);
61d1b6a4 DB	22	void (map_release)(struct bpf_map map, struct file *map_file);
61d1b6a4 DB	23	void (map_free)(struct bpf_map map);
db20fd2b AS	24	int (map_get_next_key)(struct bpf_map map, void key, void next_key);
	25
	26	/* funcs callable from userspace and from eBPF programs */
	27	void (map_lookup_elem)(struct bpf_map map, void key);
3274f520	28	int (map_update_elem)(struct bpf_map map, void key, void value, u64 flags);
db20fd2b	29	int (map_delete_elem)(struct bpf_map map, void *key);
2a36f0b9 WN	30
2a36f0b9 WN	31	/* funcs called by prog_array and perf_event_array map */
d056a788 DB	32	void (map_fd_get_ptr)(struct bpf_map map, struct file map_file,
	33	int fd);
	34	void (map_fd_put_ptr)(void ptr);
99c55f7d AS	35	};
	36
	37	struct bpf_map {
	38	atomic_t refcnt;
	39	enum bpf_map_type map_type;
	40	u32 key_size;
	41	u32 value_size;
	42	u32 max_entries;
6c905981	43	u32 map_flags;
aaac3ba9 AS	44	u32 pages;
aaac3ba9 AS	45	struct user_struct *user;
a2c83fff	46	const struct bpf_map_ops *ops;
99c55f7d	47	struct work_struct work;
c9da161c	48	atomic_t usercnt;
99c55f7d AS	49	};
	50
	51	struct bpf_map_type_list {
	52	struct list_head list_node;
a2c83fff	53	const struct bpf_map_ops *ops;
99c55f7d AS	54	enum bpf_map_type type;
	55	};
	56
17a52670 AS	57	/* function argument constraints */
17a52670 AS	58	enum bpf_arg_type {
80f1d68c	59	ARG_DONTCARE = 0, /* unused argument in helper function */
17a52670 AS	60
	61	/* the following constraints used to prototype
	62	* bpf_map_lookup/update/delete_elem() functions
	63	*/
	64	ARG_CONST_MAP_PTR, /* const argument used as pointer to bpf_map */
	65	ARG_PTR_TO_MAP_KEY, /* pointer to stack used as map key */
	66	ARG_PTR_TO_MAP_VALUE, /* pointer to stack used as map value */
	67
	68	/* the following constraints used to prototype bpf_memcmp() and other
	69	* functions that access data on eBPF program stack
	70	*/
	71	ARG_PTR_TO_STACK, /* any pointer to eBPF program stack */
435faee1 DB	72	ARG_PTR_TO_RAW_STACK, /* any pointer to eBPF program stack, area does not
	73	* need to be initialized, helper function must fill
	74	* all bytes or clear them in error case.
	75	*/
	76
17a52670	77	ARG_CONST_STACK_SIZE, /* number of bytes accessed from stack */
8e2fe1d9	78	ARG_CONST_STACK_SIZE_OR_ZERO, /* number of bytes accessed from stack or 0 */
80f1d68c	79
608cd71a	80	ARG_PTR_TO_CTX, /* pointer to context */
80f1d68c	81	ARG_ANYTHING, /* any (initialized) argument is ok */
17a52670 AS	82	};
	83
	84	/* type of values returned from helper functions */
	85	enum bpf_return_type {
	86	RET_INTEGER, /* function returns integer */
	87	RET_VOID, /* function doesn't return anything */
	88	RET_PTR_TO_MAP_VALUE_OR_NULL, /* returns a pointer to map elem value or NULL */
	89	};
	90
09756af4 AS	91	/* eBPF function prototype used by verifier to allow BPF_CALLs from eBPF programs
	92	* to in-kernel helper functions and for adjusting imm32 field in BPF_CALL
	93	* instructions after verifying
	94	*/
	95	struct bpf_func_proto {
	96	u64 (*func)(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
	97	bool gpl_only;
17a52670 AS	98	enum bpf_return_type ret_type;
	99	enum bpf_arg_type arg1_type;
	100	enum bpf_arg_type arg2_type;
	101	enum bpf_arg_type arg3_type;
	102	enum bpf_arg_type arg4_type;
	103	enum bpf_arg_type arg5_type;
	104	};
	105
	106	/* bpf_context is intentionally undefined structure. Pointer to bpf_context is
	107	* the first argument to eBPF programs.
	108	* For socket filters: 'struct bpf_context ' == 'struct sk_buff '
	109	*/
	110	struct bpf_context;
	111
	112	enum bpf_access_type {
	113	BPF_READ = 1,
	114	BPF_WRITE = 2
09756af4 AS	115	};
09756af4 AS	116
ff936a04 AS	117	struct bpf_prog;
ff936a04 AS	118
09756af4 AS	119	struct bpf_verifier_ops {
	120	/* return eBPF function prototype for verification */
	121	const struct bpf_func_proto (get_func_proto)(enum bpf_func_id func_id);
17a52670 AS	122
	123	/* return true if 'size' wide access at offset 'off' within bpf_context
	124	* with 'type' (read or write) is allowed
	125	*/
	126	bool (*is_valid_access)(int off, int size, enum bpf_access_type type);
9bac3d6d	127
d691f9e8 AS	128	u32 (*convert_ctx_access)(enum bpf_access_type type, int dst_reg,
d691f9e8 AS	129	int src_reg, int ctx_off,
ff936a04	130	struct bpf_insn insn, struct bpf_prog prog);
09756af4 AS	131	};
	132
	133	struct bpf_prog_type_list {
	134	struct list_head list_node;
a2c83fff	135	const struct bpf_verifier_ops *ops;
09756af4 AS	136	enum bpf_prog_type type;
	137	};
	138
09756af4 AS	139	struct bpf_prog_aux {
09756af4 AS	140	atomic_t refcnt;
24701ece	141	u32 used_map_cnt;
32bbe007	142	u32 max_ctx_offset;
a2c83fff	143	const struct bpf_verifier_ops *ops;
09756af4	144	struct bpf_map **used_maps;
09756af4	145	struct bpf_prog *prog;
aaac3ba9	146	struct user_struct *user;
abf2e7d6 AS	147	union {
	148	struct work_struct work;
	149	struct rcu_head rcu;
	150	};
09756af4 AS	151	};
09756af4 AS	152
04fd61ab AS	153	struct bpf_array {
	154	struct bpf_map map;
	155	u32 elem_size;
	156	/* 'ownership' of prog_array is claimed by the first program that
	157	* is going to use this map or by the first program which FD is stored
	158	* in the map to make sure that all callers and callees have the same
	159	* prog_type and JITed flag
	160	*/
	161	enum bpf_prog_type owner_prog_type;
	162	bool owner_jited;
	163	union {
	164	char value[0] __aligned(8);
2a36f0b9	165	void *ptrs[0] __aligned(8);
a10423b8	166	void __percpu *pptrs[0] __aligned(8);
04fd61ab AS	167	};
	168	};
	169	#define MAX_TAIL_CALL_CNT 32
	170
	171	u64 bpf_tail_call(u64 ctx, u64 r2, u64 index, u64 r4, u64 r5);
9940d67c	172	u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
d056a788	173
04fd61ab	174	bool bpf_prog_array_compatible(struct bpf_array array, const struct bpf_prog fp);
bd570ff9	175
0756ea3e	176	const struct bpf_func_proto *bpf_get_trace_printk_proto(void);
bd570ff9	177	const struct bpf_func_proto *bpf_get_event_output_proto(void);
04fd61ab	178
89aa0758	179	#ifdef CONFIG_BPF_SYSCALL
b121d1e7 AS	180	DECLARE_PER_CPU(int, bpf_prog_active);
b121d1e7 AS	181
0fc174de	182	void bpf_register_prog_type(struct bpf_prog_type_list *tl);
61e021f3	183	void bpf_register_map_type(struct bpf_map_type_list *tl);
0fc174de	184
0fc174de	185	struct bpf_prog *bpf_prog_get(u32 ufd);
92117d84	186	struct bpf_prog bpf_prog_inc(struct bpf_prog prog);
61e021f3	187	void bpf_prog_put(struct bpf_prog *prog);
abf2e7d6	188	void bpf_prog_put_rcu(struct bpf_prog *prog);
61e021f3	189
c9da161c	190	struct bpf_map *bpf_map_get_with_uref(u32 ufd);
c2101297	191	struct bpf_map *__bpf_map_get(struct fd f);
92117d84	192	struct bpf_map bpf_map_inc(struct bpf_map map, bool uref);
c9da161c	193	void bpf_map_put_with_uref(struct bpf_map *map);
61e021f3	194	void bpf_map_put(struct bpf_map *map);
6c905981	195	int bpf_map_precharge_memlock(u32 pages);
61e021f3	196
1be7f75d AS	197	extern int sysctl_unprivileged_bpf_disabled;
1be7f75d AS	198
b2197755 DB	199	int bpf_map_new_fd(struct bpf_map *map);
	200	int bpf_prog_new_fd(struct bpf_prog *prog);
	201
	202	int bpf_obj_pin_user(u32 ufd, const char __user *pathname);
	203	int bpf_obj_get_user(const char __user *pathname);
	204
15a07b33 AS	205	int bpf_percpu_hash_copy(struct bpf_map map, void key, void *value);
	206	int bpf_percpu_array_copy(struct bpf_map map, void key, void *value);
	207	int bpf_percpu_hash_update(struct bpf_map map, void key, void *value,
	208	u64 flags);
	209	int bpf_percpu_array_update(struct bpf_map map, void key, void *value,
	210	u64 flags);
d056a788	211
557c0c6e	212	int bpf_stackmap_copy(struct bpf_map map, void key, void *value);
15a07b33	213
d056a788 DB	214	int bpf_fd_array_map_update_elem(struct bpf_map map, struct file map_file,
	215	void key, void value, u64 map_flags);
	216	void bpf_fd_array_map_clear(struct bpf_map *map);
	217
15a07b33 AS	218	/* memcpy that is used with 8-byte aligned pointers, power-of-8 size and
	219	* forced to use 'long' read/writes to try to atomically copy long counters.
	220	* Best-effort only. No barriers here, since it _will_ race with concurrent
	221	* updates from BPF programs. Called from bpf syscall and mostly used with
	222	* size 8 or 16 bytes, so ask compiler to inline it.
	223	*/
	224	static inline void bpf_long_memcpy(void dst, const void src, u32 size)
	225	{
	226	const long *lsrc = src;
	227	long *ldst = dst;
	228
	229	size /= sizeof(long);
	230	while (size--)
	231	ldst++ = lsrc++;
	232	}
	233
61e021f3	234	/* verify correctness of eBPF program */
9bac3d6d	235	int bpf_check(struct bpf_prog *fp, union bpf_attr attr);
89aa0758	236	#else
0fc174de DB	237	static inline void bpf_register_prog_type(struct bpf_prog_type_list *tl)
	238	{
	239	}
	240
	241	static inline struct bpf_prog *bpf_prog_get(u32 ufd)
	242	{
	243	return ERR_PTR(-EOPNOTSUPP);
	244	}
	245
	246	static inline void bpf_prog_put(struct bpf_prog *prog)
	247	{
	248	}
61e021f3	249	#endif /* CONFIG_BPF_SYSCALL */
09756af4	250
d0003ec0	251	/* verifier prototypes for helper functions called from eBPF programs */
a2c83fff DB	252	extern const struct bpf_func_proto bpf_map_lookup_elem_proto;
	253	extern const struct bpf_func_proto bpf_map_update_elem_proto;
	254	extern const struct bpf_func_proto bpf_map_delete_elem_proto;
d0003ec0	255
03e69b50	256	extern const struct bpf_func_proto bpf_get_prandom_u32_proto;
c04167ce	257	extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
04fd61ab	258	extern const struct bpf_func_proto bpf_tail_call_proto;
17ca8cbf	259	extern const struct bpf_func_proto bpf_ktime_get_ns_proto;
ffeedafb AS	260	extern const struct bpf_func_proto bpf_get_current_pid_tgid_proto;
	261	extern const struct bpf_func_proto bpf_get_current_uid_gid_proto;
	262	extern const struct bpf_func_proto bpf_get_current_comm_proto;
4e10df9a AS	263	extern const struct bpf_func_proto bpf_skb_vlan_push_proto;
4e10df9a AS	264	extern const struct bpf_func_proto bpf_skb_vlan_pop_proto;
d5a3b1f6	265	extern const struct bpf_func_proto bpf_get_stackid_proto;
03e69b50	266
3ad00405 DB	267	/* Shared helpers among cBPF and eBPF. */
	268	void bpf_user_rnd_init_once(void);
	269	u64 bpf_user_rnd_u32(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
	270
99c55f7d	271	#endif /* _LINUX_BPF_H */