bpf.h 29.5 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>
19

20
struct bpf_verifier_env;
21
struct perf_event;
22
struct bpf_prog;
23
struct bpf_map;
24
struct sock;
25
struct seq_file;
26
struct btf;
27
struct btf_type;
28 29 30 31

/* map is generic key/value storage optionally accesible by eBPF programs */
struct bpf_map_ops {
	/* funcs callable from userspace (via syscall) */
32
	int (*map_alloc_check)(union bpf_attr *attr);
33
	struct bpf_map *(*map_alloc)(union bpf_attr *attr);
34 35
	void (*map_release)(struct bpf_map *map, struct file *map_file);
	void (*map_free)(struct bpf_map *map);
36
	int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key);
37
	void (*map_release_uref)(struct bpf_map *map);
38 39 40

	/* funcs callable from userspace and from eBPF programs */
	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
41
	int (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags);
42
	int (*map_delete_elem)(struct bpf_map *map, void *key);
M
Mauricio Vasquez B 已提交
43 44 45
	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);
46 47

	/* funcs called by prog_array and perf_event_array map */
48 49 50
	void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file,
				int fd);
	void (*map_fd_put_ptr)(void *ptr);
51
	u32 (*map_gen_lookup)(struct bpf_map *map, struct bpf_insn *insn_buf);
52
	u32 (*map_fd_sys_lookup_elem)(void *ptr);
53 54
	void (*map_seq_show_elem)(struct bpf_map *map, void *key,
				  struct seq_file *m);
55
	int (*map_check_btf)(const struct bpf_map *map,
56
			     const struct btf *btf,
57 58
			     const struct btf_type *key_type,
			     const struct btf_type *value_type);
59 60 61
};

struct bpf_map {
62
	/* The first two cachelines with read-mostly members of which some
63 64 65 66 67 68 69
	 * 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
70 71 72 73
	enum bpf_map_type map_type;
	u32 key_size;
	u32 value_size;
	u32 max_entries;
74
	u32 map_flags;
75
	int spin_lock_off; /* >=0 valid offset, <0 error */
M
Martin KaFai Lau 已提交
76
	u32 id;
77
	int numa_node;
78 79
	u32 btf_key_type_id;
	u32 btf_value_type_id;
80
	struct btf *btf;
81
	u32 pages;
82
	bool unpriv_array;
83
	/* 51 bytes hole */
84

85
	/* The 3rd and 4th cacheline with misc members to avoid false sharing
86 87 88 89
	 * particularly with refcounting.
	 */
	struct user_struct *user ____cacheline_aligned;
	atomic_t refcnt;
90
	atomic_t usercnt;
91
	struct work_struct work;
92
	char name[BPF_OBJ_NAME_LEN];
93 94
};

95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121
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);
	}
}
122 123
void copy_map_value_locked(struct bpf_map *map, void *dst, void *src,
			   bool lock_src);
124

125
struct bpf_offload_dev;
126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150
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);
}

151 152 153 154 155
static inline bool bpf_map_offload_neutral(const struct bpf_map *map)
{
	return map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY;
}

156 157
static inline bool bpf_map_support_seq_show(const struct bpf_map *map)
{
158
	return map->btf && map->ops->map_seq_show_elem;
159 160
}

161
int map_check_no_btf(const struct bpf_map *map,
162
		     const struct btf *btf,
163 164 165
		     const struct btf_type *key_type,
		     const struct btf_type *value_type);

166 167
extern const struct bpf_map_ops bpf_map_offload_ops;

168 169
/* function argument constraints */
enum bpf_arg_type {
170
	ARG_DONTCARE = 0,	/* unused argument in helper function */
171 172 173 174 175 176 177

	/* 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 */
178
	ARG_PTR_TO_UNINIT_MAP_VALUE,	/* pointer to valid memory used to store a map value */
179 180 181 182

	/* the following constraints used to prototype bpf_memcmp() and other
	 * functions that access data on eBPF program stack
	 */
183
	ARG_PTR_TO_MEM,		/* pointer to valid memory (stack, packet, map value) */
184
	ARG_PTR_TO_MEM_OR_NULL, /* pointer to valid memory or NULL */
185 186 187
	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.
188 189
				 */

190 191
	ARG_CONST_SIZE,		/* number of bytes accessed from memory */
	ARG_CONST_SIZE_OR_ZERO,	/* number of bytes accessed from memory or 0 */
192

193
	ARG_PTR_TO_CTX,		/* pointer to context */
194
	ARG_ANYTHING,		/* any (initialized) argument is ok */
195
	ARG_PTR_TO_SOCKET,	/* pointer to bpf_sock */
196
	ARG_PTR_TO_SPIN_LOCK,	/* pointer to bpf_spin_lock */
197 198 199 200 201 202
};

/* type of values returned from helper functions */
enum bpf_return_type {
	RET_INTEGER,			/* function returns integer */
	RET_VOID,			/* function doesn't return anything */
203
	RET_PTR_TO_MAP_VALUE,		/* returns a pointer to map elem value */
204
	RET_PTR_TO_MAP_VALUE_OR_NULL,	/* returns a pointer to map elem value or NULL */
205
	RET_PTR_TO_SOCKET_OR_NULL,	/* returns a pointer to a socket or NULL */
206 207
};

208 209 210 211 212 213 214
/* 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;
215
	bool pkt_access;
216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
	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
233 234
};

235
/* types of values stored in eBPF registers */
236 237 238 239 240 241 242 243 244
/* 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
 */
245 246
enum bpf_reg_type {
	NOT_INIT = 0,		 /* nothing was written into register */
247
	SCALAR_VALUE,		 /* reg doesn't contain a valid pointer */
248 249 250 251
	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 */
252
	PTR_TO_STACK,		 /* reg == frame_pointer + offset */
253
	PTR_TO_PACKET_META,	 /* skb->data - meta_len */
254
	PTR_TO_PACKET,		 /* reg points to skb->data */
255
	PTR_TO_PACKET_END,	 /* skb->data + headlen */
256
	PTR_TO_FLOW_KEYS,	 /* reg points to bpf_flow_keys */
257 258
	PTR_TO_SOCKET,		 /* reg points to struct bpf_sock */
	PTR_TO_SOCKET_OR_NULL,	 /* reg points to struct bpf_sock or NULL */
259 260
};

261 262 263 264 265 266 267 268
/* 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;
};

269 270 271 272 273 274
static inline void
bpf_ctx_record_field_size(struct bpf_insn_access_aux *aux, u32 size)
{
	aux->ctx_field_size = size;
}

275 276 277 278 279
struct bpf_prog_ops {
	int (*test_run)(struct bpf_prog *prog, const union bpf_attr *kattr,
			union bpf_attr __user *uattr);
};

280 281
struct bpf_verifier_ops {
	/* return eBPF function prototype for verification */
282 283 284
	const struct bpf_func_proto *
	(*get_func_proto)(enum bpf_func_id func_id,
			  const struct bpf_prog *prog);
285 286 287 288

	/* return true if 'size' wide access at offset 'off' within bpf_context
	 * with 'type' (read or write) is allowed
	 */
289
	bool (*is_valid_access)(int off, int size, enum bpf_access_type type,
290
				const struct bpf_prog *prog,
291
				struct bpf_insn_access_aux *info);
292 293
	int (*gen_prologue)(struct bpf_insn *insn, bool direct_write,
			    const struct bpf_prog *prog);
294 295
	int (*gen_ld_abs)(const struct bpf_insn *orig,
			  struct bpf_insn *insn_buf);
296 297 298
	u32 (*convert_ctx_access)(enum bpf_access_type type,
				  const struct bpf_insn *src,
				  struct bpf_insn *dst,
299
				  struct bpf_prog *prog, u32 *target_size);
300 301
};

302
struct bpf_prog_offload_ops {
303
	/* verifier basic callbacks */
304 305
	int (*insn_hook)(struct bpf_verifier_env *env,
			 int insn_idx, int prev_insn_idx);
306
	int (*finalize)(struct bpf_verifier_env *env);
307 308 309 310 311
	/* 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 */
312 313
	int (*prepare)(struct bpf_prog *prog);
	int (*translate)(struct bpf_prog *prog);
314
	void (*destroy)(struct bpf_prog *prog);
315 316
};

317
struct bpf_prog_offload {
318 319
	struct bpf_prog		*prog;
	struct net_device	*netdev;
320
	struct bpf_offload_dev	*offdev;
321 322 323
	void			*dev_priv;
	struct list_head	offloads;
	bool			dev_state;
324
	bool			opt_failed;
325 326
	void			*jited_image;
	u32			jited_len;
327 328
};

329 330
enum bpf_cgroup_storage_type {
	BPF_CGROUP_STORAGE_SHARED,
331
	BPF_CGROUP_STORAGE_PERCPU,
332 333 334 335 336
	__BPF_CGROUP_STORAGE_MAX
};

#define MAX_BPF_CGROUP_STORAGE_TYPE __BPF_CGROUP_STORAGE_MAX

337 338
struct bpf_prog_aux {
	atomic_t refcnt;
339
	u32 used_map_cnt;
340
	u32 max_ctx_offset;
341
	u32 max_pkt_offset;
342
	u32 stack_depth;
M
Martin KaFai Lau 已提交
343
	u32 id;
344 345
	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 */
346
	bool offload_requested;
347 348
	struct bpf_prog **func;
	void *jit_data; /* JIT specific data. arch dependent */
349 350
	struct latch_tree_node ksym_tnode;
	struct list_head ksym_lnode;
351
	const struct bpf_prog_ops *ops;
352 353
	struct bpf_map **used_maps;
	struct bpf_prog *prog;
354
	struct user_struct *user;
355
	u64 load_time; /* ns since boottime */
356
	struct bpf_map *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
357
	char name[BPF_OBJ_NAME_LEN];
358 359 360
#ifdef CONFIG_SECURITY
	void *security;
#endif
361
	struct bpf_prog_offload *offload;
Y
Yonghong Song 已提交
362
	struct btf *btf;
363
	struct bpf_func_info *func_info;
M
Martin KaFai Lau 已提交
364 365 366 367 368 369 370 371 372 373 374 375 376 377 378
	/* 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;
379
	u32 func_info_cnt;
M
Martin KaFai Lau 已提交
380 381 382 383 384 385
	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;
386 387 388 389
	union {
		struct work_struct work;
		struct rcu_head	rcu;
	};
390 391
};

392 393 394
struct bpf_array {
	struct bpf_map map;
	u32 elem_size;
395
	u32 index_mask;
396 397 398 399 400 401 402 403 404
	/* '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);
405
		void *ptrs[0] __aligned(8);
406
		void __percpu *pptrs[0] __aligned(8);
407 408
	};
};
409

410 411
#define MAX_TAIL_CALL_CNT 32

412 413 414 415 416 417 418
struct bpf_event_entry {
	struct perf_event *event;
	struct file *perf_file;
	struct file *map_file;
	struct rcu_head rcu;
};

419
bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp);
420
int bpf_prog_calc_tag(struct bpf_prog *fp);
421

422
const struct bpf_func_proto *bpf_get_trace_printk_proto(void);
423 424

typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src,
425
					unsigned long off, unsigned long len);
426 427 428 429 430
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);
431 432 433

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);
434

435 436 437 438
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);
439 440 441
int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog,
				     const union bpf_attr *kattr,
				     union bpf_attr __user *uattr);
442

443 444 445 446 447 448 449 450 451 452 453 454
/* 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.
 */
455 456
struct bpf_prog_array_item {
	struct bpf_prog *prog;
457
	struct bpf_cgroup_storage *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
458 459
};

460 461
struct bpf_prog_array {
	struct rcu_head rcu;
462
	struct bpf_prog_array_item items[0];
463 464
};

465
struct bpf_prog_array *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags);
466
void bpf_prog_array_free(struct bpf_prog_array __rcu *progs);
467 468 469
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);
470

471 472
void bpf_prog_array_delete_safe(struct bpf_prog_array __rcu *progs,
				struct bpf_prog *old_prog);
473
int bpf_prog_array_copy_info(struct bpf_prog_array __rcu *array,
474 475
			     u32 *prog_ids, u32 request_cnt,
			     u32 *prog_cnt);
476 477 478 479 480 481
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)	\
482
	({						\
483 484
		struct bpf_prog_array_item *_item;	\
		struct bpf_prog *_prog;			\
485
		struct bpf_prog_array *_array;		\
486
		u32 _ret = 1;				\
487
		preempt_disable();			\
488
		rcu_read_lock();			\
489 490 491
		_array = rcu_dereference(array);	\
		if (unlikely(check_non_null && !_array))\
			goto _out;			\
492 493 494 495 496
		_item = &_array->items[0];		\
		while ((_prog = READ_ONCE(_item->prog))) {		\
			bpf_cgroup_storage_set(_item->cgroup_storage);	\
			_ret &= func(_prog, ctx);	\
			_item++;			\
497 498
		}					\
_out:							\
499
		rcu_read_unlock();			\
500
		preempt_enable_no_resched();		\
501 502 503
		_ret;					\
	 })

504 505 506 507 508 509
#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)

510
#ifdef CONFIG_BPF_SYSCALL
511 512
DECLARE_PER_CPU(int, bpf_prog_active);

513 514 515
extern const struct file_operations bpf_map_fops;
extern const struct file_operations bpf_prog_fops;

516 517 518
#define BPF_PROG_TYPE(_id, _name) \
	extern const struct bpf_prog_ops _name ## _prog_ops; \
	extern const struct bpf_verifier_ops _name ## _verifier_ops;
519 520
#define BPF_MAP_TYPE(_id, _ops) \
	extern const struct bpf_map_ops _ops;
521 522
#include <linux/bpf_types.h>
#undef BPF_PROG_TYPE
523
#undef BPF_MAP_TYPE
524

525
extern const struct bpf_prog_ops bpf_offload_prog_ops;
526 527 528
extern const struct bpf_verifier_ops tc_cls_act_analyzer_ops;
extern const struct bpf_verifier_ops xdp_analyzer_ops;

529
struct bpf_prog *bpf_prog_get(u32 ufd);
530
struct bpf_prog *bpf_prog_get_type_dev(u32 ufd, enum bpf_prog_type type,
531
				       bool attach_drv);
532
struct bpf_prog * __must_check bpf_prog_add(struct bpf_prog *prog, int i);
533
void bpf_prog_sub(struct bpf_prog *prog, int i);
534
struct bpf_prog * __must_check bpf_prog_inc(struct bpf_prog *prog);
535
struct bpf_prog * __must_check bpf_prog_inc_not_zero(struct bpf_prog *prog);
536
void bpf_prog_put(struct bpf_prog *prog);
537 538
int __bpf_prog_charge(struct user_struct *user, u32 pages);
void __bpf_prog_uncharge(struct user_struct *user, u32 pages);
539

540
void bpf_prog_free_id(struct bpf_prog *prog, bool do_idr_lock);
541
void bpf_map_free_id(struct bpf_map *map, bool do_idr_lock);
542

543
struct bpf_map *bpf_map_get_with_uref(u32 ufd);
544
struct bpf_map *__bpf_map_get(struct fd f);
545
struct bpf_map * __must_check bpf_map_inc(struct bpf_map *map, bool uref);
546
void bpf_map_put_with_uref(struct bpf_map *map);
547
void bpf_map_put(struct bpf_map *map);
548
int bpf_map_precharge_memlock(u32 pages);
549 550
int bpf_map_charge_memlock(struct bpf_map *map, u32 pages);
void bpf_map_uncharge_memlock(struct bpf_map *map, u32 pages);
551
void *bpf_map_area_alloc(size_t size, int numa_node);
552
void bpf_map_area_free(void *base);
553
void bpf_map_init_from_attr(struct bpf_map *map, union bpf_attr *attr);
554

555 556
extern int sysctl_unprivileged_bpf_disabled;

557
int bpf_map_new_fd(struct bpf_map *map, int flags);
558 559 560
int bpf_prog_new_fd(struct bpf_prog *prog);

int bpf_obj_pin_user(u32 ufd, const char __user *pathname);
561
int bpf_obj_get_user(const char __user *pathname, int flags);
562

563 564 565 566 567 568
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);
569

570
int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value);
571

572 573
int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file,
				 void *key, void *value, u64 map_flags);
574
int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
M
Martin KaFai Lau 已提交
575 576
int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file,
				void *key, void *value, u64 map_flags);
577
int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
578

579
int bpf_get_file_flag(int flags);
580 581
int bpf_check_uarg_tail_zero(void __user *uaddr, size_t expected_size,
			     size_t actual_size);
582

583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598
/* 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++;
}

599
/* verify correctness of eBPF program */
Y
Yonghong Song 已提交
600 601
int bpf_check(struct bpf_prog **fp, union bpf_attr *attr,
	      union bpf_attr __user *uattr);
602
void bpf_patch_call_args(struct bpf_insn *insn, u32 stack_depth);
603 604

/* Map specifics */
605
struct xdp_buff;
606
struct sk_buff;
607 608

struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
609 610
void __dev_map_insert_ctx(struct bpf_map *map, u32 index);
void __dev_map_flush(struct bpf_map *map);
611 612
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
613 614
int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
			     struct bpf_prog *xdp_prog);
615

616 617 618 619 620 621
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);

622 623 624 625 626 627 628
/* 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;
}

629
struct bpf_prog *bpf_prog_get_type_path(const char *name, enum bpf_prog_type type);
630
int array_map_alloc_check(union bpf_attr *attr);
631

632
#else /* !CONFIG_BPF_SYSCALL */
633 634 635 636 637
static inline struct bpf_prog *bpf_prog_get(u32 ufd)
{
	return ERR_PTR(-EOPNOTSUPP);
}

638 639
static inline struct bpf_prog *bpf_prog_get_type_dev(u32 ufd,
						     enum bpf_prog_type type,
640
						     bool attach_drv)
641 642 643 644
{
	return ERR_PTR(-EOPNOTSUPP);
}

645 646
static inline struct bpf_prog * __must_check bpf_prog_add(struct bpf_prog *prog,
							  int i)
647 648 649
{
	return ERR_PTR(-EOPNOTSUPP);
}
650

651 652 653 654
static inline void bpf_prog_sub(struct bpf_prog *prog, int i)
{
}

655 656 657
static inline void bpf_prog_put(struct bpf_prog *prog)
{
}
658 659

static inline struct bpf_prog * __must_check bpf_prog_inc(struct bpf_prog *prog)
660 661 662
{
	return ERR_PTR(-EOPNOTSUPP);
}
663

664 665 666 667 668 669
static inline struct bpf_prog *__must_check
bpf_prog_inc_not_zero(struct bpf_prog *prog)
{
	return ERR_PTR(-EOPNOTSUPP);
}

670 671 672 673 674 675 676 677
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)
{
}
678

679
static inline int bpf_obj_get_user(const char __user *pathname, int flags)
680 681 682 683
{
	return -EOPNOTSUPP;
}

684 685 686 687 688 689 690 691 692 693 694 695 696
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)
{
}
697

698 699 700 701
struct xdp_buff;
struct bpf_dtab_netdev;

static inline
702 703
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx)
704 705 706 707
{
	return 0;
}

708 709 710 711 712 713 714 715 716
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;
}

717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736
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;
}
737 738 739 740 741 742

static inline struct bpf_prog *bpf_prog_get_type_path(const char *name,
				enum bpf_prog_type type)
{
	return ERR_PTR(-EOPNOTSUPP);
}
743
#endif /* CONFIG_BPF_SYSCALL */
744

745 746 747 748 749 750
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);
}

751 752
bool bpf_prog_get_ok(struct bpf_prog *, enum bpf_prog_type *, bool);

753 754
int bpf_prog_offload_compile(struct bpf_prog *prog);
void bpf_prog_offload_destroy(struct bpf_prog *prog);
755 756
int bpf_prog_offload_info_fill(struct bpf_prog_info *info,
			       struct bpf_prog *prog);
757

758 759
int bpf_map_offload_info_fill(struct bpf_map_info *info, struct bpf_map *map);

760 761 762 763 764 765 766
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);

767
bool bpf_offload_prog_map_match(struct bpf_prog *prog, struct bpf_map *map);
768

769 770
struct bpf_offload_dev *
bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops);
771 772 773 774 775
void bpf_offload_dev_destroy(struct bpf_offload_dev *offdev);
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);
776
bool bpf_offload_dev_match(struct bpf_prog *prog, struct net_device *netdev);
777

778 779 780
#if defined(CONFIG_NET) && defined(CONFIG_BPF_SYSCALL)
int bpf_prog_offload_init(struct bpf_prog *prog, union bpf_attr *attr);

781
static inline bool bpf_prog_is_dev_bound(const struct bpf_prog_aux *aux)
782
{
783
	return aux->offload_requested;
784
}
785 786 787 788 789 790 791 792

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);
793 794 795 796 797 798 799 800 801 802 803
#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;
}
804 805 806 807 808 809 810 811 812 813 814 815 816 817

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)
{
}
818 819
#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */

820 821 822
#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);
823
#else
824 825
static inline int sock_map_prog_update(struct bpf_map *map,
				       struct bpf_prog *prog, u32 which)
826 827 828
{
	return -EOPNOTSUPP;
}
829

830 831
static inline int sock_map_get_from_fd(const union bpf_attr *attr,
				       struct bpf_prog *prog)
832 833 834
{
	return -EINVAL;
}
835 836
#endif

837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861
#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

862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888
#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) */

889
/* verifier prototypes for helper functions called from eBPF programs */
890 891 892
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
893 894 895
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;
896

897
extern const struct bpf_func_proto bpf_get_prandom_u32_proto;
898
extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
899
extern const struct bpf_func_proto bpf_get_numa_node_id_proto;
900
extern const struct bpf_func_proto bpf_tail_call_proto;
901
extern const struct bpf_func_proto bpf_ktime_get_ns_proto;
902 903 904
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;
905
extern const struct bpf_func_proto bpf_get_stackid_proto;
Y
Yonghong Song 已提交
906
extern const struct bpf_func_proto bpf_get_stack_proto;
907
extern const struct bpf_func_proto bpf_sock_map_update_proto;
908
extern const struct bpf_func_proto bpf_sock_hash_update_proto;
909
extern const struct bpf_func_proto bpf_get_current_cgroup_id_proto;
910 911 912 913
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;
914 915
extern const struct bpf_func_proto bpf_spin_lock_proto;
extern const struct bpf_func_proto bpf_spin_unlock_proto;
916 917
extern const struct bpf_func_proto bpf_get_local_storage_proto;

918 919 920 921
/* 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);

922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946
#if defined(CONFIG_NET)
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
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

947
#endif /* _LINUX_BPF_H */