bpf.h 32.7 KB
Newer Older
1 2 3 4 5 6 7 8 9 10
/* 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>
11

12
#include <linux/workqueue.h>
13
#include <linux/file.h>
14
#include <linux/percpu.h>
Z
Zi Shen Lim 已提交
15
#include <linux/err.h>
16
#include <linux/rbtree_latch.h>
17
#include <linux/numa.h>
18
#include <linux/wait.h>
A
Alexei Starovoitov 已提交
19
#include <linux/u64_stats_sync.h>
20

21
struct bpf_verifier_env;
22
struct perf_event;
23
struct bpf_prog;
24
struct bpf_map;
25
struct sock;
26
struct seq_file;
27
struct btf;
28
struct btf_type;
29 30 31 32

/* map is generic key/value storage optionally accesible by eBPF programs */
struct bpf_map_ops {
	/* funcs callable from userspace (via syscall) */
33
	int (*map_alloc_check)(union bpf_attr *attr);
34
	struct bpf_map *(*map_alloc)(union bpf_attr *attr);
35 36
	void (*map_release)(struct bpf_map *map, struct file *map_file);
	void (*map_free)(struct bpf_map *map);
37
	int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key);
38
	void (*map_release_uref)(struct bpf_map *map);
39 40 41

	/* funcs callable from userspace and from eBPF programs */
	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
42
	int (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags);
43
	int (*map_delete_elem)(struct bpf_map *map, void *key);
M
Mauricio Vasquez B 已提交
44 45 46
	int (*map_push_elem)(struct bpf_map *map, void *value, u64 flags);
	int (*map_pop_elem)(struct bpf_map *map, void *value);
	int (*map_peek_elem)(struct bpf_map *map, void *value);
47 48

	/* funcs called by prog_array and perf_event_array map */
49 50 51
	void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file,
				int fd);
	void (*map_fd_put_ptr)(void *ptr);
52
	u32 (*map_gen_lookup)(struct bpf_map *map, struct bpf_insn *insn_buf);
53
	u32 (*map_fd_sys_lookup_elem)(void *ptr);
54 55
	void (*map_seq_show_elem)(struct bpf_map *map, void *key,
				  struct seq_file *m);
56
	int (*map_check_btf)(const struct bpf_map *map,
57
			     const struct btf *btf,
58 59
			     const struct btf_type *key_type,
			     const struct btf_type *value_type);
60 61 62 63 64 65

	/* Direct value access helpers. */
	int (*map_direct_value_addr)(const struct bpf_map *map,
				     u64 *imm, u32 off);
	int (*map_direct_value_meta)(const struct bpf_map *map,
				     u64 imm, u32 *off);
66 67 68
};

struct bpf_map {
69
	/* The first two cachelines with read-mostly members of which some
70 71 72 73 74 75 76
	 * are also accessed in fast-path (e.g. ops, max_entries).
	 */
	const struct bpf_map_ops *ops ____cacheline_aligned;
	struct bpf_map *inner_map_meta;
#ifdef CONFIG_SECURITY
	void *security;
#endif
77 78 79 80
	enum bpf_map_type map_type;
	u32 key_size;
	u32 value_size;
	u32 max_entries;
81
	u32 map_flags;
82
	int spin_lock_off; /* >=0 valid offset, <0 error */
M
Martin KaFai Lau 已提交
83
	u32 id;
84
	int numa_node;
85 86
	u32 btf_key_type_id;
	u32 btf_value_type_id;
87
	struct btf *btf;
88
	u32 pages;
89
	bool unpriv_array;
90 91
	bool frozen; /* write-once */
	/* 48 bytes hole */
92

93
	/* The 3rd and 4th cacheline with misc members to avoid false sharing
94 95 96 97
	 * particularly with refcounting.
	 */
	struct user_struct *user ____cacheline_aligned;
	atomic_t refcnt;
98
	atomic_t usercnt;
99
	struct work_struct work;
100
	char name[BPF_OBJ_NAME_LEN];
101 102
};

103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129
static inline bool map_value_has_spin_lock(const struct bpf_map *map)
{
	return map->spin_lock_off >= 0;
}

static inline void check_and_init_map_lock(struct bpf_map *map, void *dst)
{
	if (likely(!map_value_has_spin_lock(map)))
		return;
	*(struct bpf_spin_lock *)(dst + map->spin_lock_off) =
		(struct bpf_spin_lock){};
}

/* copy everything but bpf_spin_lock */
static inline void copy_map_value(struct bpf_map *map, void *dst, void *src)
{
	if (unlikely(map_value_has_spin_lock(map))) {
		u32 off = map->spin_lock_off;

		memcpy(dst, src, off);
		memcpy(dst + off + sizeof(struct bpf_spin_lock),
		       src + off + sizeof(struct bpf_spin_lock),
		       map->value_size - off - sizeof(struct bpf_spin_lock));
	} else {
		memcpy(dst, src, map->value_size);
	}
}
130 131
void copy_map_value_locked(struct bpf_map *map, void *dst, void *src,
			   bool lock_src);
132

133
struct bpf_offload_dev;
134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158
struct bpf_offloaded_map;

struct bpf_map_dev_ops {
	int (*map_get_next_key)(struct bpf_offloaded_map *map,
				void *key, void *next_key);
	int (*map_lookup_elem)(struct bpf_offloaded_map *map,
			       void *key, void *value);
	int (*map_update_elem)(struct bpf_offloaded_map *map,
			       void *key, void *value, u64 flags);
	int (*map_delete_elem)(struct bpf_offloaded_map *map, void *key);
};

struct bpf_offloaded_map {
	struct bpf_map map;
	struct net_device *netdev;
	const struct bpf_map_dev_ops *dev_ops;
	void *dev_priv;
	struct list_head offloads;
};

static inline struct bpf_offloaded_map *map_to_offmap(struct bpf_map *map)
{
	return container_of(map, struct bpf_offloaded_map, map);
}

159 160 161 162 163
static inline bool bpf_map_offload_neutral(const struct bpf_map *map)
{
	return map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY;
}

164 165
static inline bool bpf_map_support_seq_show(const struct bpf_map *map)
{
166
	return map->btf && map->ops->map_seq_show_elem;
167 168
}

169
int map_check_no_btf(const struct bpf_map *map,
170
		     const struct btf *btf,
171 172 173
		     const struct btf_type *key_type,
		     const struct btf_type *value_type);

174 175
extern const struct bpf_map_ops bpf_map_offload_ops;

176 177
/* function argument constraints */
enum bpf_arg_type {
178
	ARG_DONTCARE = 0,	/* unused argument in helper function */
179 180 181 182 183 184 185

	/* 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 */
186
	ARG_PTR_TO_UNINIT_MAP_VALUE,	/* pointer to valid memory used to store a map value */
187 188 189 190

	/* the following constraints used to prototype bpf_memcmp() and other
	 * functions that access data on eBPF program stack
	 */
191
	ARG_PTR_TO_MEM,		/* pointer to valid memory (stack, packet, map value) */
192
	ARG_PTR_TO_MEM_OR_NULL, /* pointer to valid memory or NULL */
193 194 195
	ARG_PTR_TO_UNINIT_MEM,	/* pointer to memory does not need to be initialized,
				 * helper function must fill all bytes or clear
				 * them in error case.
196 197
				 */

198 199
	ARG_CONST_SIZE,		/* number of bytes accessed from memory */
	ARG_CONST_SIZE_OR_ZERO,	/* number of bytes accessed from memory or 0 */
200

201
	ARG_PTR_TO_CTX,		/* pointer to context */
202
	ARG_ANYTHING,		/* any (initialized) argument is ok */
203
	ARG_PTR_TO_SPIN_LOCK,	/* pointer to bpf_spin_lock */
204
	ARG_PTR_TO_SOCK_COMMON,	/* pointer to sock_common */
205 206 207 208 209 210
};

/* type of values returned from helper functions */
enum bpf_return_type {
	RET_INTEGER,			/* function returns integer */
	RET_VOID,			/* function doesn't return anything */
211
	RET_PTR_TO_MAP_VALUE,		/* returns a pointer to map elem value */
212
	RET_PTR_TO_MAP_VALUE_OR_NULL,	/* returns a pointer to map elem value or NULL */
213
	RET_PTR_TO_SOCKET_OR_NULL,	/* returns a pointer to a socket or NULL */
214
	RET_PTR_TO_TCP_SOCK_OR_NULL,	/* returns a pointer to a tcp_sock or NULL */
215
	RET_PTR_TO_SOCK_COMMON_OR_NULL,	/* returns a pointer to a sock_common or NULL */
216 217
};

218 219 220 221 222 223 224
/* 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;
225
	bool pkt_access;
226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242
	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
243 244
};

245
/* types of values stored in eBPF registers */
246 247 248 249 250 251 252 253 254
/* Pointer types represent:
 * pointer
 * pointer + imm
 * pointer + (u16) var
 * pointer + (u16) var + imm
 * if (range > 0) then [ptr, ptr + range - off) is safe to access
 * if (id > 0) means that some 'var' was added
 * if (off > 0) means that 'imm' was added
 */
255 256
enum bpf_reg_type {
	NOT_INIT = 0,		 /* nothing was written into register */
257
	SCALAR_VALUE,		 /* reg doesn't contain a valid pointer */
258 259 260 261
	PTR_TO_CTX,		 /* reg points to bpf_context */
	CONST_PTR_TO_MAP,	 /* reg points to struct bpf_map */
	PTR_TO_MAP_VALUE,	 /* reg points to map element value */
	PTR_TO_MAP_VALUE_OR_NULL,/* points to map elem value or NULL */
262
	PTR_TO_STACK,		 /* reg == frame_pointer + offset */
263
	PTR_TO_PACKET_META,	 /* skb->data - meta_len */
264
	PTR_TO_PACKET,		 /* reg points to skb->data */
265
	PTR_TO_PACKET_END,	 /* skb->data + headlen */
266
	PTR_TO_FLOW_KEYS,	 /* reg points to bpf_flow_keys */
267 268
	PTR_TO_SOCKET,		 /* reg points to struct bpf_sock */
	PTR_TO_SOCKET_OR_NULL,	 /* reg points to struct bpf_sock or NULL */
269 270
	PTR_TO_SOCK_COMMON,	 /* reg points to sock_common */
	PTR_TO_SOCK_COMMON_OR_NULL, /* reg points to sock_common or NULL */
271 272
	PTR_TO_TCP_SOCK,	 /* reg points to struct tcp_sock */
	PTR_TO_TCP_SOCK_OR_NULL, /* reg points to struct tcp_sock or NULL */
273 274
};

275 276 277 278 279 280 281 282
/* The information passed from prog-specific *_is_valid_access
 * back to the verifier.
 */
struct bpf_insn_access_aux {
	enum bpf_reg_type reg_type;
	int ctx_field_size;
};

283 284 285 286 287 288
static inline void
bpf_ctx_record_field_size(struct bpf_insn_access_aux *aux, u32 size)
{
	aux->ctx_field_size = size;
}

289 290 291 292 293
struct bpf_prog_ops {
	int (*test_run)(struct bpf_prog *prog, const union bpf_attr *kattr,
			union bpf_attr __user *uattr);
};

294 295
struct bpf_verifier_ops {
	/* return eBPF function prototype for verification */
296 297 298
	const struct bpf_func_proto *
	(*get_func_proto)(enum bpf_func_id func_id,
			  const struct bpf_prog *prog);
299 300 301 302

	/* return true if 'size' wide access at offset 'off' within bpf_context
	 * with 'type' (read or write) is allowed
	 */
303
	bool (*is_valid_access)(int off, int size, enum bpf_access_type type,
304
				const struct bpf_prog *prog,
305
				struct bpf_insn_access_aux *info);
306 307
	int (*gen_prologue)(struct bpf_insn *insn, bool direct_write,
			    const struct bpf_prog *prog);
308 309
	int (*gen_ld_abs)(const struct bpf_insn *orig,
			  struct bpf_insn *insn_buf);
310 311 312
	u32 (*convert_ctx_access)(enum bpf_access_type type,
				  const struct bpf_insn *src,
				  struct bpf_insn *dst,
313
				  struct bpf_prog *prog, u32 *target_size);
314 315
};

316
struct bpf_prog_offload_ops {
317
	/* verifier basic callbacks */
318 319
	int (*insn_hook)(struct bpf_verifier_env *env,
			 int insn_idx, int prev_insn_idx);
320
	int (*finalize)(struct bpf_verifier_env *env);
321 322 323 324 325
	/* verifier optimization callbacks (called after .finalize) */
	int (*replace_insn)(struct bpf_verifier_env *env, u32 off,
			    struct bpf_insn *insn);
	int (*remove_insns)(struct bpf_verifier_env *env, u32 off, u32 cnt);
	/* program management callbacks */
326 327
	int (*prepare)(struct bpf_prog *prog);
	int (*translate)(struct bpf_prog *prog);
328
	void (*destroy)(struct bpf_prog *prog);
329 330
};

331
struct bpf_prog_offload {
332 333
	struct bpf_prog		*prog;
	struct net_device	*netdev;
334
	struct bpf_offload_dev	*offdev;
335 336 337
	void			*dev_priv;
	struct list_head	offloads;
	bool			dev_state;
338
	bool			opt_failed;
339 340
	void			*jited_image;
	u32			jited_len;
341 342
};

343 344
enum bpf_cgroup_storage_type {
	BPF_CGROUP_STORAGE_SHARED,
345
	BPF_CGROUP_STORAGE_PERCPU,
346 347 348 349 350
	__BPF_CGROUP_STORAGE_MAX
};

#define MAX_BPF_CGROUP_STORAGE_TYPE __BPF_CGROUP_STORAGE_MAX

A
Alexei Starovoitov 已提交
351 352 353 354 355 356
struct bpf_prog_stats {
	u64 cnt;
	u64 nsecs;
	struct u64_stats_sync syncp;
};

357 358
struct bpf_prog_aux {
	atomic_t refcnt;
359
	u32 used_map_cnt;
360
	u32 max_ctx_offset;
361
	u32 max_pkt_offset;
362
	u32 stack_depth;
M
Martin KaFai Lau 已提交
363
	u32 id;
364 365
	u32 func_cnt; /* used by non-func prog as the number of func progs */
	u32 func_idx; /* 0 for non-func prog, the index in func array for func prog */
366
	bool offload_requested;
367 368
	struct bpf_prog **func;
	void *jit_data; /* JIT specific data. arch dependent */
369 370
	struct latch_tree_node ksym_tnode;
	struct list_head ksym_lnode;
371
	const struct bpf_prog_ops *ops;
372 373
	struct bpf_map **used_maps;
	struct bpf_prog *prog;
374
	struct user_struct *user;
375
	u64 load_time; /* ns since boottime */
376
	struct bpf_map *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
377
	char name[BPF_OBJ_NAME_LEN];
378 379 380
#ifdef CONFIG_SECURITY
	void *security;
#endif
381
	struct bpf_prog_offload *offload;
Y
Yonghong Song 已提交
382
	struct btf *btf;
383
	struct bpf_func_info *func_info;
M
Martin KaFai Lau 已提交
384 385 386 387 388 389 390 391 392 393 394 395 396 397 398
	/* bpf_line_info loaded from userspace.  linfo->insn_off
	 * has the xlated insn offset.
	 * Both the main and sub prog share the same linfo.
	 * The subprog can access its first linfo by
	 * using the linfo_idx.
	 */
	struct bpf_line_info *linfo;
	/* jited_linfo is the jited addr of the linfo.  It has a
	 * one to one mapping to linfo:
	 * jited_linfo[i] is the jited addr for the linfo[i]->insn_off.
	 * Both the main and sub prog share the same jited_linfo.
	 * The subprog can access its first jited_linfo by
	 * using the linfo_idx.
	 */
	void **jited_linfo;
399
	u32 func_info_cnt;
M
Martin KaFai Lau 已提交
400 401 402 403 404 405
	u32 nr_linfo;
	/* subprog can use linfo_idx to access its first linfo and
	 * jited_linfo.
	 * main prog always has linfo_idx == 0
	 */
	u32 linfo_idx;
A
Alexei Starovoitov 已提交
406
	struct bpf_prog_stats __percpu *stats;
407 408 409 410
	union {
		struct work_struct work;
		struct rcu_head	rcu;
	};
411 412
};

413 414 415
struct bpf_array {
	struct bpf_map map;
	u32 elem_size;
416
	u32 index_mask;
417 418 419 420 421 422 423 424 425
	/* '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);
426
		void *ptrs[0] __aligned(8);
427
		void __percpu *pptrs[0] __aligned(8);
428 429
	};
};
430

431
#define BPF_COMPLEXITY_LIMIT_INSNS      1000000 /* yes. 1M insns */
432 433
#define MAX_TAIL_CALL_CNT 32

434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462
#define BPF_F_ACCESS_MASK	(BPF_F_RDONLY |		\
				 BPF_F_RDONLY_PROG |	\
				 BPF_F_WRONLY |		\
				 BPF_F_WRONLY_PROG)

#define BPF_MAP_CAN_READ	BIT(0)
#define BPF_MAP_CAN_WRITE	BIT(1)

static inline u32 bpf_map_flags_to_cap(struct bpf_map *map)
{
	u32 access_flags = map->map_flags & (BPF_F_RDONLY_PROG | BPF_F_WRONLY_PROG);

	/* Combination of BPF_F_RDONLY_PROG | BPF_F_WRONLY_PROG is
	 * not possible.
	 */
	if (access_flags & BPF_F_RDONLY_PROG)
		return BPF_MAP_CAN_READ;
	else if (access_flags & BPF_F_WRONLY_PROG)
		return BPF_MAP_CAN_WRITE;
	else
		return BPF_MAP_CAN_READ | BPF_MAP_CAN_WRITE;
}

static inline bool bpf_map_flags_access_ok(u32 access_flags)
{
	return (access_flags & (BPF_F_RDONLY_PROG | BPF_F_WRONLY_PROG)) !=
	       (BPF_F_RDONLY_PROG | BPF_F_WRONLY_PROG);
}

463 464 465 466 467 468 469
struct bpf_event_entry {
	struct perf_event *event;
	struct file *perf_file;
	struct file *map_file;
	struct rcu_head rcu;
};

470
bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp);
471
int bpf_prog_calc_tag(struct bpf_prog *fp);
472

473
const struct bpf_func_proto *bpf_get_trace_printk_proto(void);
474 475

typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src,
476
					unsigned long off, unsigned long len);
477 478 479 480 481
typedef u32 (*bpf_convert_ctx_access_t)(enum bpf_access_type type,
					const struct bpf_insn *src,
					struct bpf_insn *dst,
					struct bpf_prog *prog,
					u32 *target_size);
482 483 484

u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size,
		     void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy);
485

486 487 488 489 490 491 492 493 494 495 496 497
/* an array of programs to be executed under rcu_lock.
 *
 * Typical usage:
 * ret = BPF_PROG_RUN_ARRAY(&bpf_prog_array, ctx, BPF_PROG_RUN);
 *
 * the structure returned by bpf_prog_array_alloc() should be populated
 * with program pointers and the last pointer must be NULL.
 * The user has to keep refcnt on the program and make sure the program
 * is removed from the array before bpf_prog_put().
 * The 'struct bpf_prog_array *' should only be replaced with xchg()
 * since other cpus are walking the array of pointers in parallel.
 */
498 499
struct bpf_prog_array_item {
	struct bpf_prog *prog;
500
	struct bpf_cgroup_storage *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
501 502
};

503 504
struct bpf_prog_array {
	struct rcu_head rcu;
505
	struct bpf_prog_array_item items[0];
506 507
};

508
struct bpf_prog_array *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags);
509
void bpf_prog_array_free(struct bpf_prog_array __rcu *progs);
510 511 512
int bpf_prog_array_length(struct bpf_prog_array __rcu *progs);
int bpf_prog_array_copy_to_user(struct bpf_prog_array __rcu *progs,
				__u32 __user *prog_ids, u32 cnt);
513

514 515
void bpf_prog_array_delete_safe(struct bpf_prog_array __rcu *progs,
				struct bpf_prog *old_prog);
516
int bpf_prog_array_copy_info(struct bpf_prog_array __rcu *array,
517 518
			     u32 *prog_ids, u32 request_cnt,
			     u32 *prog_cnt);
519 520 521 522 523 524
int bpf_prog_array_copy(struct bpf_prog_array __rcu *old_array,
			struct bpf_prog *exclude_prog,
			struct bpf_prog *include_prog,
			struct bpf_prog_array **new_array);

#define __BPF_PROG_RUN_ARRAY(array, ctx, func, check_non_null)	\
525
	({						\
526 527
		struct bpf_prog_array_item *_item;	\
		struct bpf_prog *_prog;			\
528
		struct bpf_prog_array *_array;		\
529
		u32 _ret = 1;				\
530
		preempt_disable();			\
531
		rcu_read_lock();			\
532 533 534
		_array = rcu_dereference(array);	\
		if (unlikely(check_non_null && !_array))\
			goto _out;			\
535 536 537 538 539
		_item = &_array->items[0];		\
		while ((_prog = READ_ONCE(_item->prog))) {		\
			bpf_cgroup_storage_set(_item->cgroup_storage);	\
			_ret &= func(_prog, ctx);	\
			_item++;			\
540 541
		}					\
_out:							\
542
		rcu_read_unlock();			\
543
		preempt_enable_no_resched();		\
544 545 546
		_ret;					\
	 })

547 548 549 550 551 552
#define BPF_PROG_RUN_ARRAY(array, ctx, func)		\
	__BPF_PROG_RUN_ARRAY(array, ctx, func, false)

#define BPF_PROG_RUN_ARRAY_CHECK(array, ctx, func)	\
	__BPF_PROG_RUN_ARRAY(array, ctx, func, true)

553
#ifdef CONFIG_BPF_SYSCALL
554 555
DECLARE_PER_CPU(int, bpf_prog_active);

556 557 558
extern const struct file_operations bpf_map_fops;
extern const struct file_operations bpf_prog_fops;

559 560 561
#define BPF_PROG_TYPE(_id, _name) \
	extern const struct bpf_prog_ops _name ## _prog_ops; \
	extern const struct bpf_verifier_ops _name ## _verifier_ops;
562 563
#define BPF_MAP_TYPE(_id, _ops) \
	extern const struct bpf_map_ops _ops;
564 565
#include <linux/bpf_types.h>
#undef BPF_PROG_TYPE
566
#undef BPF_MAP_TYPE
567

568
extern const struct bpf_prog_ops bpf_offload_prog_ops;
569 570 571
extern const struct bpf_verifier_ops tc_cls_act_analyzer_ops;
extern const struct bpf_verifier_ops xdp_analyzer_ops;

572
struct bpf_prog *bpf_prog_get(u32 ufd);
573
struct bpf_prog *bpf_prog_get_type_dev(u32 ufd, enum bpf_prog_type type,
574
				       bool attach_drv);
575
struct bpf_prog * __must_check bpf_prog_add(struct bpf_prog *prog, int i);
576
void bpf_prog_sub(struct bpf_prog *prog, int i);
577
struct bpf_prog * __must_check bpf_prog_inc(struct bpf_prog *prog);
578
struct bpf_prog * __must_check bpf_prog_inc_not_zero(struct bpf_prog *prog);
579
void bpf_prog_put(struct bpf_prog *prog);
580 581
int __bpf_prog_charge(struct user_struct *user, u32 pages);
void __bpf_prog_uncharge(struct user_struct *user, u32 pages);
582

583
void bpf_prog_free_id(struct bpf_prog *prog, bool do_idr_lock);
584
void bpf_map_free_id(struct bpf_map *map, bool do_idr_lock);
585

586
struct bpf_map *bpf_map_get_with_uref(u32 ufd);
587
struct bpf_map *__bpf_map_get(struct fd f);
588
struct bpf_map * __must_check bpf_map_inc(struct bpf_map *map, bool uref);
589
void bpf_map_put_with_uref(struct bpf_map *map);
590
void bpf_map_put(struct bpf_map *map);
591
int bpf_map_precharge_memlock(u32 pages);
592 593
int bpf_map_charge_memlock(struct bpf_map *map, u32 pages);
void bpf_map_uncharge_memlock(struct bpf_map *map, u32 pages);
594
void *bpf_map_area_alloc(size_t size, int numa_node);
595
void bpf_map_area_free(void *base);
596
void bpf_map_init_from_attr(struct bpf_map *map, union bpf_attr *attr);
597

598
extern int sysctl_unprivileged_bpf_disabled;
A
Alexei Starovoitov 已提交
599
extern int sysctl_bpf_stats_enabled;
600

601
int bpf_map_new_fd(struct bpf_map *map, int flags);
602 603 604
int bpf_prog_new_fd(struct bpf_prog *prog);

int bpf_obj_pin_user(u32 ufd, const char __user *pathname);
605
int bpf_obj_get_user(const char __user *pathname, int flags);
606

607 608 609 610 611 612
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);
613

614
int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value);
615

616 617
int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file,
				 void *key, void *value, u64 map_flags);
618
int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
M
Martin KaFai Lau 已提交
619 620
int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file,
				void *key, void *value, u64 map_flags);
621
int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
622

623
int bpf_get_file_flag(int flags);
624 625
int bpf_check_uarg_tail_zero(void __user *uaddr, size_t expected_size,
			     size_t actual_size);
626

627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642
/* 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++;
}

643
/* verify correctness of eBPF program */
Y
Yonghong Song 已提交
644 645
int bpf_check(struct bpf_prog **fp, union bpf_attr *attr,
	      union bpf_attr __user *uattr);
646
void bpf_patch_call_args(struct bpf_insn *insn, u32 stack_depth);
647 648

/* Map specifics */
649
struct xdp_buff;
650
struct sk_buff;
651 652

struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
653 654
void __dev_map_insert_ctx(struct bpf_map *map, u32 index);
void __dev_map_flush(struct bpf_map *map);
655 656
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
657 658
int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
			     struct bpf_prog *xdp_prog);
659

660 661 662 663 664 665
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
void __cpu_map_insert_ctx(struct bpf_map *map, u32 index);
void __cpu_map_flush(struct bpf_map *map);
int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
		    struct net_device *dev_rx);

666 667 668 669 670 671 672
/* Return map's numa specified by userspace */
static inline int bpf_map_attr_numa_node(const union bpf_attr *attr)
{
	return (attr->map_flags & BPF_F_NUMA_NODE) ?
		attr->numa_node : NUMA_NO_NODE;
}

673
struct bpf_prog *bpf_prog_get_type_path(const char *name, enum bpf_prog_type type);
674
int array_map_alloc_check(union bpf_attr *attr);
675

676 677 678 679 680 681 682
int bpf_prog_test_run_xdp(struct bpf_prog *prog, const union bpf_attr *kattr,
			  union bpf_attr __user *uattr);
int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
			  union bpf_attr __user *uattr);
int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog,
				     const union bpf_attr *kattr,
				     union bpf_attr __user *uattr);
683
#else /* !CONFIG_BPF_SYSCALL */
684 685 686 687 688
static inline struct bpf_prog *bpf_prog_get(u32 ufd)
{
	return ERR_PTR(-EOPNOTSUPP);
}

689 690
static inline struct bpf_prog *bpf_prog_get_type_dev(u32 ufd,
						     enum bpf_prog_type type,
691
						     bool attach_drv)
692 693 694 695
{
	return ERR_PTR(-EOPNOTSUPP);
}

696 697
static inline struct bpf_prog * __must_check bpf_prog_add(struct bpf_prog *prog,
							  int i)
698 699 700
{
	return ERR_PTR(-EOPNOTSUPP);
}
701

702 703 704 705
static inline void bpf_prog_sub(struct bpf_prog *prog, int i)
{
}

706 707 708
static inline void bpf_prog_put(struct bpf_prog *prog)
{
}
709 710

static inline struct bpf_prog * __must_check bpf_prog_inc(struct bpf_prog *prog)
711 712 713
{
	return ERR_PTR(-EOPNOTSUPP);
}
714

715 716 717 718 719 720
static inline struct bpf_prog *__must_check
bpf_prog_inc_not_zero(struct bpf_prog *prog)
{
	return ERR_PTR(-EOPNOTSUPP);
}

721 722 723 724 725 726 727 728
static inline int __bpf_prog_charge(struct user_struct *user, u32 pages)
{
	return 0;
}

static inline void __bpf_prog_uncharge(struct user_struct *user, u32 pages)
{
}
729

730
static inline int bpf_obj_get_user(const char __user *pathname, int flags)
731 732 733 734
{
	return -EOPNOTSUPP;
}

735 736 737 738 739 740 741 742 743 744 745 746 747
static inline struct net_device  *__dev_map_lookup_elem(struct bpf_map *map,
						       u32 key)
{
	return NULL;
}

static inline void __dev_map_insert_ctx(struct bpf_map *map, u32 index)
{
}

static inline void __dev_map_flush(struct bpf_map *map)
{
}
748

749 750 751 752
struct xdp_buff;
struct bpf_dtab_netdev;

static inline
753 754
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx)
755 756 757 758
{
	return 0;
}

759 760 761 762 763 764 765 766 767
struct sk_buff;

static inline int dev_map_generic_redirect(struct bpf_dtab_netdev *dst,
					   struct sk_buff *skb,
					   struct bpf_prog *xdp_prog)
{
	return 0;
}

768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787
static inline
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key)
{
	return NULL;
}

static inline void __cpu_map_insert_ctx(struct bpf_map *map, u32 index)
{
}

static inline void __cpu_map_flush(struct bpf_map *map)
{
}

static inline int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu,
				  struct xdp_buff *xdp,
				  struct net_device *dev_rx)
{
	return 0;
}
788 789 790 791 792 793

static inline struct bpf_prog *bpf_prog_get_type_path(const char *name,
				enum bpf_prog_type type)
{
	return ERR_PTR(-EOPNOTSUPP);
}
794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814

static inline int bpf_prog_test_run_xdp(struct bpf_prog *prog,
					const union bpf_attr *kattr,
					union bpf_attr __user *uattr)
{
	return -ENOTSUPP;
}

static inline int bpf_prog_test_run_skb(struct bpf_prog *prog,
					const union bpf_attr *kattr,
					union bpf_attr __user *uattr)
{
	return -ENOTSUPP;
}

static inline int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog,
						   const union bpf_attr *kattr,
						   union bpf_attr __user *uattr)
{
	return -ENOTSUPP;
}
815
#endif /* CONFIG_BPF_SYSCALL */
816

817 818 819 820 821 822
static inline struct bpf_prog *bpf_prog_get_type(u32 ufd,
						 enum bpf_prog_type type)
{
	return bpf_prog_get_type_dev(ufd, type, false);
}

823 824
bool bpf_prog_get_ok(struct bpf_prog *, enum bpf_prog_type *, bool);

825 826
int bpf_prog_offload_compile(struct bpf_prog *prog);
void bpf_prog_offload_destroy(struct bpf_prog *prog);
827 828
int bpf_prog_offload_info_fill(struct bpf_prog_info *info,
			       struct bpf_prog *prog);
829

830 831
int bpf_map_offload_info_fill(struct bpf_map_info *info, struct bpf_map *map);

832 833 834 835 836 837 838
int bpf_map_offload_lookup_elem(struct bpf_map *map, void *key, void *value);
int bpf_map_offload_update_elem(struct bpf_map *map,
				void *key, void *value, u64 flags);
int bpf_map_offload_delete_elem(struct bpf_map *map, void *key);
int bpf_map_offload_get_next_key(struct bpf_map *map,
				 void *key, void *next_key);

839
bool bpf_offload_prog_map_match(struct bpf_prog *prog, struct bpf_map *map);
840

841
struct bpf_offload_dev *
842
bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops, void *priv);
843
void bpf_offload_dev_destroy(struct bpf_offload_dev *offdev);
844
void *bpf_offload_dev_priv(struct bpf_offload_dev *offdev);
845 846 847 848
int bpf_offload_dev_netdev_register(struct bpf_offload_dev *offdev,
				    struct net_device *netdev);
void bpf_offload_dev_netdev_unregister(struct bpf_offload_dev *offdev,
				       struct net_device *netdev);
849
bool bpf_offload_dev_match(struct bpf_prog *prog, struct net_device *netdev);
850

851 852 853
#if defined(CONFIG_NET) && defined(CONFIG_BPF_SYSCALL)
int bpf_prog_offload_init(struct bpf_prog *prog, union bpf_attr *attr);

854
static inline bool bpf_prog_is_dev_bound(const struct bpf_prog_aux *aux)
855
{
856
	return aux->offload_requested;
857
}
858 859 860 861 862 863 864 865

static inline bool bpf_map_is_dev_bound(struct bpf_map *map)
{
	return unlikely(map->ops == &bpf_map_offload_ops);
}

struct bpf_map *bpf_map_offload_map_alloc(union bpf_attr *attr);
void bpf_map_offload_map_free(struct bpf_map *map);
866 867 868 869 870 871 872 873 874 875 876
#else
static inline int bpf_prog_offload_init(struct bpf_prog *prog,
					union bpf_attr *attr)
{
	return -EOPNOTSUPP;
}

static inline bool bpf_prog_is_dev_bound(struct bpf_prog_aux *aux)
{
	return false;
}
877 878 879 880 881 882 883 884 885 886 887 888 889 890

static inline bool bpf_map_is_dev_bound(struct bpf_map *map)
{
	return false;
}

static inline struct bpf_map *bpf_map_offload_map_alloc(union bpf_attr *attr)
{
	return ERR_PTR(-EOPNOTSUPP);
}

static inline void bpf_map_offload_map_free(struct bpf_map *map)
{
}
891 892
#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */

893 894 895
#if defined(CONFIG_BPF_STREAM_PARSER)
int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog, u32 which);
int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog *prog);
896
#else
897 898
static inline int sock_map_prog_update(struct bpf_map *map,
				       struct bpf_prog *prog, u32 which)
899 900 901
{
	return -EOPNOTSUPP;
}
902

903 904
static inline int sock_map_get_from_fd(const union bpf_attr *attr,
				       struct bpf_prog *prog)
905 906 907
{
	return -EINVAL;
}
908 909
#endif

910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934
#if defined(CONFIG_XDP_SOCKETS)
struct xdp_sock;
struct xdp_sock *__xsk_map_lookup_elem(struct bpf_map *map, u32 key);
int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
		       struct xdp_sock *xs);
void __xsk_map_flush(struct bpf_map *map);
#else
struct xdp_sock;
static inline struct xdp_sock *__xsk_map_lookup_elem(struct bpf_map *map,
						     u32 key)
{
	return NULL;
}

static inline int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
				     struct xdp_sock *xs)
{
	return -EOPNOTSUPP;
}

static inline void __xsk_map_flush(struct bpf_map *map)
{
}
#endif

935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961
#if defined(CONFIG_INET) && defined(CONFIG_BPF_SYSCALL)
void bpf_sk_reuseport_detach(struct sock *sk);
int bpf_fd_reuseport_array_lookup_elem(struct bpf_map *map, void *key,
				       void *value);
int bpf_fd_reuseport_array_update_elem(struct bpf_map *map, void *key,
				       void *value, u64 map_flags);
#else
static inline void bpf_sk_reuseport_detach(struct sock *sk)
{
}

#ifdef CONFIG_BPF_SYSCALL
static inline int bpf_fd_reuseport_array_lookup_elem(struct bpf_map *map,
						     void *key, void *value)
{
	return -EOPNOTSUPP;
}

static inline int bpf_fd_reuseport_array_update_elem(struct bpf_map *map,
						     void *key, void *value,
						     u64 map_flags)
{
	return -EOPNOTSUPP;
}
#endif /* CONFIG_BPF_SYSCALL */
#endif /* defined(CONFIG_INET) && defined(CONFIG_BPF_SYSCALL) */

962
/* verifier prototypes for helper functions called from eBPF programs */
963 964 965
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;
M
Mauricio Vasquez B 已提交
966 967 968
extern const struct bpf_func_proto bpf_map_push_elem_proto;
extern const struct bpf_func_proto bpf_map_pop_elem_proto;
extern const struct bpf_func_proto bpf_map_peek_elem_proto;
969

970
extern const struct bpf_func_proto bpf_get_prandom_u32_proto;
971
extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
972
extern const struct bpf_func_proto bpf_get_numa_node_id_proto;
973
extern const struct bpf_func_proto bpf_tail_call_proto;
974
extern const struct bpf_func_proto bpf_ktime_get_ns_proto;
975 976 977
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;
978
extern const struct bpf_func_proto bpf_get_stackid_proto;
Y
Yonghong Song 已提交
979
extern const struct bpf_func_proto bpf_get_stack_proto;
980
extern const struct bpf_func_proto bpf_sock_map_update_proto;
981
extern const struct bpf_func_proto bpf_sock_hash_update_proto;
982
extern const struct bpf_func_proto bpf_get_current_cgroup_id_proto;
983 984 985 986
extern const struct bpf_func_proto bpf_msg_redirect_hash_proto;
extern const struct bpf_func_proto bpf_msg_redirect_map_proto;
extern const struct bpf_func_proto bpf_sk_redirect_hash_proto;
extern const struct bpf_func_proto bpf_sk_redirect_map_proto;
987 988
extern const struct bpf_func_proto bpf_spin_lock_proto;
extern const struct bpf_func_proto bpf_spin_unlock_proto;
989 990
extern const struct bpf_func_proto bpf_get_local_storage_proto;

991 992 993 994
/* 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);

995
#if defined(CONFIG_NET)
996 997 998
bool bpf_sock_common_is_valid_access(int off, int size,
				     enum bpf_access_type type,
				     struct bpf_insn_access_aux *info);
999 1000 1001 1002 1003 1004 1005 1006
bool bpf_sock_is_valid_access(int off, int size, enum bpf_access_type type,
			      struct bpf_insn_access_aux *info);
u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
				const struct bpf_insn *si,
				struct bpf_insn *insn_buf,
				struct bpf_prog *prog,
				u32 *target_size);
#else
1007 1008 1009 1010 1011 1012
static inline bool bpf_sock_common_is_valid_access(int off, int size,
						   enum bpf_access_type type,
						   struct bpf_insn_access_aux *info)
{
	return false;
}
1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028
static inline bool bpf_sock_is_valid_access(int off, int size,
					    enum bpf_access_type type,
					    struct bpf_insn_access_aux *info)
{
	return false;
}
static inline u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
					      const struct bpf_insn *si,
					      struct bpf_insn *insn_buf,
					      struct bpf_prog *prog,
					      u32 *target_size)
{
	return 0;
}
#endif

1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055
#ifdef CONFIG_INET
bool bpf_tcp_sock_is_valid_access(int off, int size, enum bpf_access_type type,
				  struct bpf_insn_access_aux *info);

u32 bpf_tcp_sock_convert_ctx_access(enum bpf_access_type type,
				    const struct bpf_insn *si,
				    struct bpf_insn *insn_buf,
				    struct bpf_prog *prog,
				    u32 *target_size);
#else
static inline bool bpf_tcp_sock_is_valid_access(int off, int size,
						enum bpf_access_type type,
						struct bpf_insn_access_aux *info)
{
	return false;
}

static inline u32 bpf_tcp_sock_convert_ctx_access(enum bpf_access_type type,
						  const struct bpf_insn *si,
						  struct bpf_insn *insn_buf,
						  struct bpf_prog *prog,
						  u32 *target_size)
{
	return 0;
}
#endif /* CONFIG_INET */

1056
#endif /* _LINUX_BPF_H */