bpf.h 41.6 KB
Newer Older
1
/* SPDX-License-Identifier: GPL-2.0-only */
2 3 4 5 6 7
/* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com
 */
#ifndef _LINUX_BPF_H
#define _LINUX_BPF_H 1

#include <uapi/linux/bpf.h>
8

9
#include <linux/workqueue.h>
10
#include <linux/file.h>
11
#include <linux/percpu.h>
Z
Zi Shen Lim 已提交
12
#include <linux/err.h>
13
#include <linux/rbtree_latch.h>
14
#include <linux/numa.h>
15
#include <linux/mm_types.h>
16
#include <linux/wait.h>
A
Alexei Starovoitov 已提交
17
#include <linux/u64_stats_sync.h>
A
Alexei Starovoitov 已提交
18 19
#include <linux/refcount.h>
#include <linux/mutex.h>
20

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

33 34 35
extern struct idr btf_idr;
extern spinlock_t btf_idr_lock;

36 37 38
/* map is generic key/value storage optionally accesible by eBPF programs */
struct bpf_map_ops {
	/* funcs callable from userspace (via syscall) */
39
	int (*map_alloc_check)(union bpf_attr *attr);
40
	struct bpf_map *(*map_alloc)(union bpf_attr *attr);
41 42
	void (*map_release)(struct bpf_map *map, struct file *map_file);
	void (*map_free)(struct bpf_map *map);
43
	int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key);
44
	void (*map_release_uref)(struct bpf_map *map);
45
	void *(*map_lookup_elem_sys_only)(struct bpf_map *map, void *key);
46 47 48

	/* funcs callable from userspace and from eBPF programs */
	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
49
	int (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags);
50
	int (*map_delete_elem)(struct bpf_map *map, void *key);
M
Mauricio Vasquez B 已提交
51 52 53
	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);
54 55

	/* funcs called by prog_array and perf_event_array map */
56 57 58
	void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file,
				int fd);
	void (*map_fd_put_ptr)(void *ptr);
59
	u32 (*map_gen_lookup)(struct bpf_map *map, struct bpf_insn *insn_buf);
60
	u32 (*map_fd_sys_lookup_elem)(void *ptr);
61 62
	void (*map_seq_show_elem)(struct bpf_map *map, void *key,
				  struct seq_file *m);
63
	int (*map_check_btf)(const struct bpf_map *map,
64
			     const struct btf *btf,
65 66
			     const struct btf_type *key_type,
			     const struct btf_type *value_type);
67

68 69 70 71 72 73
	/* Prog poke tracking helpers. */
	int (*map_poke_track)(struct bpf_map *map, struct bpf_prog_aux *aux);
	void (*map_poke_untrack)(struct bpf_map *map, struct bpf_prog_aux *aux);
	void (*map_poke_run)(struct bpf_map *map, u32 key, struct bpf_prog *old,
			     struct bpf_prog *new);

74 75 76 77 78
	/* 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);
79
	int (*map_mmap)(struct bpf_map *map, struct vm_area_struct *vma);
80 81
};

82 83 84 85 86
struct bpf_map_memory {
	u32 pages;
	struct user_struct *user;
};

87
struct bpf_map {
88
	/* The first two cachelines with read-mostly members of which some
89 90 91 92 93 94 95
	 * 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
96 97 98 99
	enum bpf_map_type map_type;
	u32 key_size;
	u32 value_size;
	u32 max_entries;
100
	u32 map_flags;
101
	int spin_lock_off; /* >=0 valid offset, <0 error */
M
Martin KaFai Lau 已提交
102
	u32 id;
103
	int numa_node;
104 105
	u32 btf_key_type_id;
	u32 btf_value_type_id;
106
	struct btf *btf;
107
	struct bpf_map_memory memory;
108
	char name[BPF_OBJ_NAME_LEN];
109
	bool unpriv_array;
110 111
	bool frozen; /* write-once; write-protected by freeze_mutex */
	/* 22 bytes hole */
112

113
	/* The 3rd and 4th cacheline with misc members to avoid false sharing
114 115
	 * particularly with refcounting.
	 */
116 117
	atomic64_t refcnt ____cacheline_aligned;
	atomic64_t usercnt;
118
	struct work_struct work;
119 120
	struct mutex freeze_mutex;
	u64 writecnt; /* writable mmap cnt; protected by freeze_mutex */
121 122
};

123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149
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);
	}
}
150 151
void copy_map_value_locked(struct bpf_map *map, void *dst, void *src,
			   bool lock_src);
152

153
struct bpf_offload_dev;
154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
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);
}

179 180 181 182 183
static inline bool bpf_map_offload_neutral(const struct bpf_map *map)
{
	return map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY;
}

184 185
static inline bool bpf_map_support_seq_show(const struct bpf_map *map)
{
186
	return map->btf && map->ops->map_seq_show_elem;
187 188
}

189
int map_check_no_btf(const struct bpf_map *map,
190
		     const struct btf *btf,
191 192 193
		     const struct btf_type *key_type,
		     const struct btf_type *value_type);

194 195
extern const struct bpf_map_ops bpf_map_offload_ops;

196 197
/* function argument constraints */
enum bpf_arg_type {
198
	ARG_DONTCARE = 0,	/* unused argument in helper function */
199 200 201 202 203 204 205

	/* 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 */
206
	ARG_PTR_TO_UNINIT_MAP_VALUE,	/* pointer to valid memory used to store a map value */
207
	ARG_PTR_TO_MAP_VALUE_OR_NULL,	/* pointer to stack used as map value or NULL */
208 209 210 211

	/* the following constraints used to prototype bpf_memcmp() and other
	 * functions that access data on eBPF program stack
	 */
212
	ARG_PTR_TO_MEM,		/* pointer to valid memory (stack, packet, map value) */
213
	ARG_PTR_TO_MEM_OR_NULL, /* pointer to valid memory or NULL */
214 215 216
	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.
217 218
				 */

219 220
	ARG_CONST_SIZE,		/* number of bytes accessed from memory */
	ARG_CONST_SIZE_OR_ZERO,	/* number of bytes accessed from memory or 0 */
221

222
	ARG_PTR_TO_CTX,		/* pointer to context */
223
	ARG_ANYTHING,		/* any (initialized) argument is ok */
224
	ARG_PTR_TO_SPIN_LOCK,	/* pointer to bpf_spin_lock */
225
	ARG_PTR_TO_SOCK_COMMON,	/* pointer to sock_common */
226 227
	ARG_PTR_TO_INT,		/* pointer to int */
	ARG_PTR_TO_LONG,	/* pointer to long */
228
	ARG_PTR_TO_SOCKET,	/* pointer to bpf_sock (fullsock) */
229
	ARG_PTR_TO_BTF_ID,	/* pointer to in-kernel struct */
230 231 232 233 234 235
};

/* type of values returned from helper functions */
enum bpf_return_type {
	RET_INTEGER,			/* function returns integer */
	RET_VOID,			/* function doesn't return anything */
236
	RET_PTR_TO_MAP_VALUE,		/* returns a pointer to map elem value */
237
	RET_PTR_TO_MAP_VALUE_OR_NULL,	/* returns a pointer to map elem value or NULL */
238
	RET_PTR_TO_SOCKET_OR_NULL,	/* returns a pointer to a socket or NULL */
239
	RET_PTR_TO_TCP_SOCK_OR_NULL,	/* returns a pointer to a tcp_sock or NULL */
240
	RET_PTR_TO_SOCK_COMMON_OR_NULL,	/* returns a pointer to a sock_common or NULL */
241 242
};

243 244 245 246 247 248 249
/* 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;
250
	bool pkt_access;
251
	enum bpf_return_type ret_type;
252 253 254 255 256 257 258 259 260 261
	union {
		struct {
			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;
		};
		enum bpf_arg_type arg_type[5];
	};
262
	int *btf_id; /* BTF ids of arguments */
263 264 265 266 267 268 269 270 271 272 273
};

/* 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
274 275
};

276
/* types of values stored in eBPF registers */
277 278 279 280 281 282 283 284 285
/* 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
 */
286 287
enum bpf_reg_type {
	NOT_INIT = 0,		 /* nothing was written into register */
288
	SCALAR_VALUE,		 /* reg doesn't contain a valid pointer */
289 290 291 292
	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 */
293
	PTR_TO_STACK,		 /* reg == frame_pointer + offset */
294
	PTR_TO_PACKET_META,	 /* skb->data - meta_len */
295
	PTR_TO_PACKET,		 /* reg points to skb->data */
296
	PTR_TO_PACKET_END,	 /* skb->data + headlen */
297
	PTR_TO_FLOW_KEYS,	 /* reg points to bpf_flow_keys */
298 299
	PTR_TO_SOCKET,		 /* reg points to struct bpf_sock */
	PTR_TO_SOCKET_OR_NULL,	 /* reg points to struct bpf_sock or NULL */
300 301
	PTR_TO_SOCK_COMMON,	 /* reg points to sock_common */
	PTR_TO_SOCK_COMMON_OR_NULL, /* reg points to sock_common or NULL */
302 303
	PTR_TO_TCP_SOCK,	 /* reg points to struct tcp_sock */
	PTR_TO_TCP_SOCK_OR_NULL, /* reg points to struct tcp_sock or NULL */
304
	PTR_TO_TP_BUFFER,	 /* reg points to a writable raw tp's buffer */
305
	PTR_TO_XDP_SOCK,	 /* reg points to struct xdp_sock */
306
	PTR_TO_BTF_ID,		 /* reg points to kernel struct */
307 308
};

309 310 311 312 313
/* The information passed from prog-specific *_is_valid_access
 * back to the verifier.
 */
struct bpf_insn_access_aux {
	enum bpf_reg_type reg_type;
314 315 316 317 318
	union {
		int ctx_field_size;
		u32 btf_id;
	};
	struct bpf_verifier_log *log; /* for verbose logs */
319 320
};

321 322 323 324 325 326
static inline void
bpf_ctx_record_field_size(struct bpf_insn_access_aux *aux, u32 size)
{
	aux->ctx_field_size = size;
}

327 328 329 330 331
struct bpf_prog_ops {
	int (*test_run)(struct bpf_prog *prog, const union bpf_attr *kattr,
			union bpf_attr __user *uattr);
};

332 333
struct bpf_verifier_ops {
	/* return eBPF function prototype for verification */
334 335 336
	const struct bpf_func_proto *
	(*get_func_proto)(enum bpf_func_id func_id,
			  const struct bpf_prog *prog);
337 338 339 340

	/* return true if 'size' wide access at offset 'off' within bpf_context
	 * with 'type' (read or write) is allowed
	 */
341
	bool (*is_valid_access)(int off, int size, enum bpf_access_type type,
342
				const struct bpf_prog *prog,
343
				struct bpf_insn_access_aux *info);
344 345
	int (*gen_prologue)(struct bpf_insn *insn, bool direct_write,
			    const struct bpf_prog *prog);
346 347
	int (*gen_ld_abs)(const struct bpf_insn *orig,
			  struct bpf_insn *insn_buf);
348 349 350
	u32 (*convert_ctx_access)(enum bpf_access_type type,
				  const struct bpf_insn *src,
				  struct bpf_insn *dst,
351
				  struct bpf_prog *prog, u32 *target_size);
352 353
};

354
struct bpf_prog_offload_ops {
355
	/* verifier basic callbacks */
356 357
	int (*insn_hook)(struct bpf_verifier_env *env,
			 int insn_idx, int prev_insn_idx);
358
	int (*finalize)(struct bpf_verifier_env *env);
359 360 361 362 363
	/* 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 */
364 365
	int (*prepare)(struct bpf_prog *prog);
	int (*translate)(struct bpf_prog *prog);
366
	void (*destroy)(struct bpf_prog *prog);
367 368
};

369
struct bpf_prog_offload {
370 371
	struct bpf_prog		*prog;
	struct net_device	*netdev;
372
	struct bpf_offload_dev	*offdev;
373 374 375
	void			*dev_priv;
	struct list_head	offloads;
	bool			dev_state;
376
	bool			opt_failed;
377 378
	void			*jited_image;
	u32			jited_len;
379 380
};

381 382
enum bpf_cgroup_storage_type {
	BPF_CGROUP_STORAGE_SHARED,
383
	BPF_CGROUP_STORAGE_PERCPU,
384 385 386 387 388
	__BPF_CGROUP_STORAGE_MAX
};

#define MAX_BPF_CGROUP_STORAGE_TYPE __BPF_CGROUP_STORAGE_MAX

389 390 391 392 393
/* The longest tracepoint has 12 args.
 * See include/trace/bpf_probe.h
 */
#define MAX_BPF_FUNC_ARGS 12

A
Alexei Starovoitov 已提交
394 395 396 397
struct bpf_prog_stats {
	u64 cnt;
	u64 nsecs;
	struct u64_stats_sync syncp;
E
Eric Dumazet 已提交
398
} __aligned(2 * sizeof(u64));
A
Alexei Starovoitov 已提交
399

A
Alexei Starovoitov 已提交
400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
struct btf_func_model {
	u8 ret_size;
	u8 nr_args;
	u8 arg_size[MAX_BPF_FUNC_ARGS];
};

/* Restore arguments before returning from trampoline to let original function
 * continue executing. This flag is used for fentry progs when there are no
 * fexit progs.
 */
#define BPF_TRAMP_F_RESTORE_REGS	BIT(0)
/* Call original function after fentry progs, but before fexit progs.
 * Makes sense for fentry/fexit, normal calls and indirect calls.
 */
#define BPF_TRAMP_F_CALL_ORIG		BIT(1)
/* Skip current frame and return to parent.  Makes sense for fentry/fexit
 * programs only. Should not be used with normal calls and indirect calls.
 */
#define BPF_TRAMP_F_SKIP_FRAME		BIT(2)

/* Different use cases for BPF trampoline:
 * 1. replace nop at the function entry (kprobe equivalent)
 *    flags = BPF_TRAMP_F_RESTORE_REGS
 *    fentry = a set of programs to run before returning from trampoline
 *
 * 2. replace nop at the function entry (kprobe + kretprobe equivalent)
 *    flags = BPF_TRAMP_F_CALL_ORIG | BPF_TRAMP_F_SKIP_FRAME
 *    orig_call = fentry_ip + MCOUNT_INSN_SIZE
 *    fentry = a set of program to run before calling original function
 *    fexit = a set of program to run after original function
 *
 * 3. replace direct call instruction anywhere in the function body
 *    or assign a function pointer for indirect call (like tcp_congestion_ops->cong_avoid)
 *    With flags = 0
 *      fentry = a set of programs to run before returning from trampoline
 *    With flags = BPF_TRAMP_F_CALL_ORIG
 *      orig_call = original callback addr or direct function addr
 *      fentry = a set of program to run before calling original function
 *      fexit = a set of program to run after original function
 */
int arch_prepare_bpf_trampoline(void *image, struct btf_func_model *m, u32 flags,
				struct bpf_prog **fentry_progs, int fentry_cnt,
				struct bpf_prog **fexit_progs, int fexit_cnt,
				void *orig_call);
/* these two functions are called from generated trampoline */
u64 notrace __bpf_prog_enter(void);
void notrace __bpf_prog_exit(struct bpf_prog *prog, u64 start);

enum bpf_tramp_prog_type {
	BPF_TRAMP_FENTRY,
	BPF_TRAMP_FEXIT,
	BPF_TRAMP_MAX
};

struct bpf_trampoline {
	/* hlist for trampoline_table */
	struct hlist_node hlist;
	/* serializes access to fields of this trampoline */
	struct mutex mutex;
	refcount_t refcnt;
	u64 key;
	struct {
		struct btf_func_model model;
		void *addr;
464
		bool ftrace_managed;
A
Alexei Starovoitov 已提交
465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494
	} func;
	/* list of BPF programs using this trampoline */
	struct hlist_head progs_hlist[BPF_TRAMP_MAX];
	/* Number of attached programs. A counter per kind. */
	int progs_cnt[BPF_TRAMP_MAX];
	/* Executable image of trampoline */
	void *image;
	u64 selector;
};
#ifdef CONFIG_BPF_JIT
struct bpf_trampoline *bpf_trampoline_lookup(u64 key);
int bpf_trampoline_link_prog(struct bpf_prog *prog);
int bpf_trampoline_unlink_prog(struct bpf_prog *prog);
void bpf_trampoline_put(struct bpf_trampoline *tr);
#else
static inline struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
{
	return NULL;
}
static inline int bpf_trampoline_link_prog(struct bpf_prog *prog)
{
	return -ENOTSUPP;
}
static inline int bpf_trampoline_unlink_prog(struct bpf_prog *prog)
{
	return -ENOTSUPP;
}
static inline void bpf_trampoline_put(struct bpf_trampoline *tr) {}
#endif

495 496 497 498
struct bpf_func_info_aux {
	bool unreliable;
};

499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516
enum bpf_jit_poke_reason {
	BPF_POKE_REASON_TAIL_CALL,
};

/* Descriptor of pokes pointing /into/ the JITed image. */
struct bpf_jit_poke_descriptor {
	void *ip;
	union {
		struct {
			struct bpf_map *map;
			u32 key;
		} tail_call;
	};
	bool ip_stable;
	u8 adj_off;
	u16 reason;
};

517
struct bpf_prog_aux {
518
	atomic64_t refcnt;
519
	u32 used_map_cnt;
520
	u32 max_ctx_offset;
521
	u32 max_pkt_offset;
522
	u32 max_tp_access;
523
	u32 stack_depth;
M
Martin KaFai Lau 已提交
524
	u32 id;
525 526
	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 */
527
	u32 attach_btf_id; /* in-kernel BTF type id to attach to */
528
	struct bpf_prog *linked_prog;
529
	bool verifier_zext; /* Zero extensions has been inserted by verifier. */
530
	bool offload_requested;
531
	bool attach_btf_trace; /* true if attaching to BTF-enabled raw tp */
532
	bool func_proto_unreliable;
A
Alexei Starovoitov 已提交
533 534 535
	enum bpf_tramp_prog_type trampoline_prog_type;
	struct bpf_trampoline *trampoline;
	struct hlist_node tramp_hlist;
536 537 538 539
	/* BTF_KIND_FUNC_PROTO for valid attach_btf_id */
	const struct btf_type *attach_func_proto;
	/* function name for valid attach_btf_id */
	const char *attach_func_name;
540 541
	struct bpf_prog **func;
	void *jit_data; /* JIT specific data. arch dependent */
542 543
	struct bpf_jit_poke_descriptor *poke_tab;
	u32 size_poke_tab;
544 545
	struct latch_tree_node ksym_tnode;
	struct list_head ksym_lnode;
546
	const struct bpf_prog_ops *ops;
547 548
	struct bpf_map **used_maps;
	struct bpf_prog *prog;
549
	struct user_struct *user;
550
	u64 load_time; /* ns since boottime */
551
	struct bpf_map *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
552
	char name[BPF_OBJ_NAME_LEN];
553 554 555
#ifdef CONFIG_SECURITY
	void *security;
#endif
556
	struct bpf_prog_offload *offload;
Y
Yonghong Song 已提交
557
	struct btf *btf;
558
	struct bpf_func_info *func_info;
559
	struct bpf_func_info_aux *func_info_aux;
M
Martin KaFai Lau 已提交
560 561 562 563 564 565 566 567 568 569 570 571 572 573 574
	/* 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;
575
	u32 func_info_cnt;
M
Martin KaFai Lau 已提交
576 577 578 579 580 581
	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;
582 583
	u32 num_exentries;
	struct exception_table_entry *extable;
A
Alexei Starovoitov 已提交
584
	struct bpf_prog_stats __percpu *stats;
585 586 587 588
	union {
		struct work_struct work;
		struct rcu_head	rcu;
	};
589 590
};

591 592 593 594 595 596 597 598
struct bpf_array_aux {
	/* '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 type;
	bool jited;
599 600 601 602 603
	/* Programs with direct jumps into programs part of this array. */
	struct list_head poke_progs;
	struct bpf_map *map;
	struct mutex poke_mutex;
	struct work_struct work;
604 605
};

606 607 608
struct bpf_array {
	struct bpf_map map;
	u32 elem_size;
609
	u32 index_mask;
610
	struct bpf_array_aux *aux;
611 612
	union {
		char value[0] __aligned(8);
613
		void *ptrs[0] __aligned(8);
614
		void __percpu *pptrs[0] __aligned(8);
615 616
	};
};
617

618
#define BPF_COMPLEXITY_LIMIT_INSNS      1000000 /* yes. 1M insns */
619 620
#define MAX_TAIL_CALL_CNT 32

621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649
#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);
}

650 651 652 653 654 655 656
struct bpf_event_entry {
	struct perf_event *event;
	struct file *perf_file;
	struct file *map_file;
	struct rcu_head rcu;
};

657
bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp);
658
int bpf_prog_calc_tag(struct bpf_prog *fp);
659
const char *kernel_type_name(u32 btf_type_id);
660

661
const struct bpf_func_proto *bpf_get_trace_printk_proto(void);
662 663

typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src,
664
					unsigned long off, unsigned long len);
665 666 667 668 669
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);
670 671 672

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);
673

674 675 676 677 678 679 680 681 682 683 684 685
/* 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.
 */
686 687
struct bpf_prog_array_item {
	struct bpf_prog *prog;
688
	struct bpf_cgroup_storage *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
689 690
};

691 692
struct bpf_prog_array {
	struct rcu_head rcu;
693
	struct bpf_prog_array_item items[0];
694 695
};

696
struct bpf_prog_array *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags);
697 698
void bpf_prog_array_free(struct bpf_prog_array *progs);
int bpf_prog_array_length(struct bpf_prog_array *progs);
699
bool bpf_prog_array_is_empty(struct bpf_prog_array *array);
700
int bpf_prog_array_copy_to_user(struct bpf_prog_array *progs,
701
				__u32 __user *prog_ids, u32 cnt);
702

703
void bpf_prog_array_delete_safe(struct bpf_prog_array *progs,
704
				struct bpf_prog *old_prog);
705
int bpf_prog_array_copy_info(struct bpf_prog_array *array,
706 707
			     u32 *prog_ids, u32 request_cnt,
			     u32 *prog_cnt);
708
int bpf_prog_array_copy(struct bpf_prog_array *old_array,
709 710 711 712 713
			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)	\
714
	({						\
715 716
		struct bpf_prog_array_item *_item;	\
		struct bpf_prog *_prog;			\
717
		struct bpf_prog_array *_array;		\
718
		u32 _ret = 1;				\
719
		preempt_disable();			\
720
		rcu_read_lock();			\
721 722 723
		_array = rcu_dereference(array);	\
		if (unlikely(check_non_null && !_array))\
			goto _out;			\
724 725 726 727 728
		_item = &_array->items[0];		\
		while ((_prog = READ_ONCE(_item->prog))) {		\
			bpf_cgroup_storage_set(_item->cgroup_storage);	\
			_ret &= func(_prog, ctx);	\
			_item++;			\
729 730
		}					\
_out:							\
731
		rcu_read_unlock();			\
732
		preempt_enable();			\
733 734 735
		_ret;					\
	 })

736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785
/* To be used by __cgroup_bpf_run_filter_skb for EGRESS BPF progs
 * so BPF programs can request cwr for TCP packets.
 *
 * Current cgroup skb programs can only return 0 or 1 (0 to drop the
 * packet. This macro changes the behavior so the low order bit
 * indicates whether the packet should be dropped (0) or not (1)
 * and the next bit is a congestion notification bit. This could be
 * used by TCP to call tcp_enter_cwr()
 *
 * Hence, new allowed return values of CGROUP EGRESS BPF programs are:
 *   0: drop packet
 *   1: keep packet
 *   2: drop packet and cn
 *   3: keep packet and cn
 *
 * This macro then converts it to one of the NET_XMIT or an error
 * code that is then interpreted as drop packet (and no cn):
 *   0: NET_XMIT_SUCCESS  skb should be transmitted
 *   1: NET_XMIT_DROP     skb should be dropped and cn
 *   2: NET_XMIT_CN       skb should be transmitted and cn
 *   3: -EPERM            skb should be dropped
 */
#define BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY(array, ctx, func)		\
	({						\
		struct bpf_prog_array_item *_item;	\
		struct bpf_prog *_prog;			\
		struct bpf_prog_array *_array;		\
		u32 ret;				\
		u32 _ret = 1;				\
		u32 _cn = 0;				\
		preempt_disable();			\
		rcu_read_lock();			\
		_array = rcu_dereference(array);	\
		_item = &_array->items[0];		\
		while ((_prog = READ_ONCE(_item->prog))) {		\
			bpf_cgroup_storage_set(_item->cgroup_storage);	\
			ret = func(_prog, ctx);		\
			_ret &= (ret & 1);		\
			_cn |= (ret & 2);		\
			_item++;			\
		}					\
		rcu_read_unlock();			\
		preempt_enable();			\
		if (_ret)				\
			_ret = (_cn ? NET_XMIT_CN : NET_XMIT_SUCCESS);	\
		else					\
			_ret = (_cn ? NET_XMIT_DROP : -EPERM);		\
		_ret;					\
	})

786 787 788 789 790 791
#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)

792
#ifdef CONFIG_BPF_SYSCALL
793 794
DECLARE_PER_CPU(int, bpf_prog_active);

795 796 797
extern const struct file_operations bpf_map_fops;
extern const struct file_operations bpf_prog_fops;

A
Alexei Starovoitov 已提交
798
#define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \
799 800
	extern const struct bpf_prog_ops _name ## _prog_ops; \
	extern const struct bpf_verifier_ops _name ## _verifier_ops;
801 802
#define BPF_MAP_TYPE(_id, _ops) \
	extern const struct bpf_map_ops _ops;
803 804
#include <linux/bpf_types.h>
#undef BPF_PROG_TYPE
805
#undef BPF_MAP_TYPE
806

807
extern const struct bpf_prog_ops bpf_offload_prog_ops;
808 809 810
extern const struct bpf_verifier_ops tc_cls_act_analyzer_ops;
extern const struct bpf_verifier_ops xdp_analyzer_ops;

811
struct bpf_prog *bpf_prog_get(u32 ufd);
812
struct bpf_prog *bpf_prog_get_type_dev(u32 ufd, enum bpf_prog_type type,
813
				       bool attach_drv);
814
void bpf_prog_add(struct bpf_prog *prog, int i);
815
void bpf_prog_sub(struct bpf_prog *prog, int i);
816
void bpf_prog_inc(struct bpf_prog *prog);
817
struct bpf_prog * __must_check bpf_prog_inc_not_zero(struct bpf_prog *prog);
818
void bpf_prog_put(struct bpf_prog *prog);
819 820
int __bpf_prog_charge(struct user_struct *user, u32 pages);
void __bpf_prog_uncharge(struct user_struct *user, u32 pages);
821

822
void bpf_prog_free_id(struct bpf_prog *prog, bool do_idr_lock);
823
void bpf_map_free_id(struct bpf_map *map, bool do_idr_lock);
824

825
struct bpf_map *bpf_map_get_with_uref(u32 ufd);
826
struct bpf_map *__bpf_map_get(struct fd f);
827 828 829
void bpf_map_inc(struct bpf_map *map);
void bpf_map_inc_with_uref(struct bpf_map *map);
struct bpf_map * __must_check bpf_map_inc_not_zero(struct bpf_map *map);
830
void bpf_map_put_with_uref(struct bpf_map *map);
831
void bpf_map_put(struct bpf_map *map);
832 833
int bpf_map_charge_memlock(struct bpf_map *map, u32 pages);
void bpf_map_uncharge_memlock(struct bpf_map *map, u32 pages);
834
int bpf_map_charge_init(struct bpf_map_memory *mem, u64 size);
835 836 837
void bpf_map_charge_finish(struct bpf_map_memory *mem);
void bpf_map_charge_move(struct bpf_map_memory *dst,
			 struct bpf_map_memory *src);
838 839
void *bpf_map_area_alloc(u64 size, int numa_node);
void *bpf_map_area_mmapable_alloc(u64 size, int numa_node);
840
void bpf_map_area_free(void *base);
841
void bpf_map_init_from_attr(struct bpf_map *map, union bpf_attr *attr);
842

843 844
extern int sysctl_unprivileged_bpf_disabled;

845
int bpf_map_new_fd(struct bpf_map *map, int flags);
846 847 848
int bpf_prog_new_fd(struct bpf_prog *prog);

int bpf_obj_pin_user(u32 ufd, const char __user *pathname);
849
int bpf_obj_get_user(const char __user *pathname, int flags);
850

851 852 853 854 855 856
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);
857

858
int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value);
859

860 861
int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file,
				 void *key, void *value, u64 map_flags);
862
int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
M
Martin KaFai Lau 已提交
863 864
int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file,
				void *key, void *value, u64 map_flags);
865
int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
866

867
int bpf_get_file_flag(int flags);
868 869
int bpf_check_uarg_tail_zero(void __user *uaddr, size_t expected_size,
			     size_t actual_size);
870

871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886
/* 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++;
}

887
/* verify correctness of eBPF program */
Y
Yonghong Song 已提交
888 889
int bpf_check(struct bpf_prog **fp, union bpf_attr *attr,
	      union bpf_attr __user *uattr);
890
void bpf_patch_call_args(struct bpf_insn *insn, u32 stack_depth);
891 892

/* Map specifics */
893
struct xdp_buff;
894
struct sk_buff;
895 896

struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
897
struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
898
void __dev_map_flush(struct bpf_map *map);
899 900
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
901 902
int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
			     struct bpf_prog *xdp_prog);
903

904 905 906 907 908
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
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);

909 910 911 912 913 914 915
/* 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;
}

916
struct bpf_prog *bpf_prog_get_type_path(const char *name, enum bpf_prog_type type);
917
int array_map_alloc_check(union bpf_attr *attr);
918

919 920 921 922 923 924 925
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);
926 927 928 929 930 931 932
bool btf_ctx_access(int off, int size, enum bpf_access_type type,
		    const struct bpf_prog *prog,
		    struct bpf_insn_access_aux *info);
int btf_struct_access(struct bpf_verifier_log *log,
		      const struct btf_type *t, int off, int size,
		      enum bpf_access_type atype,
		      u32 *next_btf_id);
933 934
int btf_resolve_helper_id(struct bpf_verifier_log *log,
			  const struct bpf_func_proto *fn, int);
935

A
Alexei Starovoitov 已提交
936 937 938 939 940 941
int btf_distill_func_proto(struct bpf_verifier_log *log,
			   struct btf *btf,
			   const struct btf_type *func_proto,
			   const char *func_name,
			   struct btf_func_model *m);

942 943
int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog);

944
#else /* !CONFIG_BPF_SYSCALL */
945 946 947 948 949
static inline struct bpf_prog *bpf_prog_get(u32 ufd)
{
	return ERR_PTR(-EOPNOTSUPP);
}

950 951
static inline struct bpf_prog *bpf_prog_get_type_dev(u32 ufd,
						     enum bpf_prog_type type,
952
						     bool attach_drv)
953 954 955 956
{
	return ERR_PTR(-EOPNOTSUPP);
}

957
static inline void bpf_prog_add(struct bpf_prog *prog, int i)
958 959
{
}
960

961 962 963 964
static inline void bpf_prog_sub(struct bpf_prog *prog, int i)
{
}

965 966 967
static inline void bpf_prog_put(struct bpf_prog *prog)
{
}
968

969
static inline void bpf_prog_inc(struct bpf_prog *prog)
970 971
{
}
972

973 974 975 976 977 978
static inline struct bpf_prog *__must_check
bpf_prog_inc_not_zero(struct bpf_prog *prog)
{
	return ERR_PTR(-EOPNOTSUPP);
}

979 980 981 982 983 984 985 986
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)
{
}
987

988
static inline int bpf_obj_get_user(const char __user *pathname, int flags)
989 990 991 992
{
	return -EOPNOTSUPP;
}

993 994 995 996 997 998
static inline struct net_device  *__dev_map_lookup_elem(struct bpf_map *map,
						       u32 key)
{
	return NULL;
}

999 1000 1001 1002 1003 1004
static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map,
							     u32 key)
{
	return NULL;
}

1005 1006 1007
static inline void __dev_map_flush(struct bpf_map *map)
{
}
1008

1009 1010 1011 1012
struct xdp_buff;
struct bpf_dtab_netdev;

static inline
1013 1014
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx)
1015 1016 1017 1018
{
	return 0;
}

1019 1020 1021 1022 1023 1024 1025 1026 1027
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;
}

1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043
static inline
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key)
{
	return NULL;
}

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;
}
1044 1045 1046 1047 1048 1049

static inline struct bpf_prog *bpf_prog_get_type_path(const char *name,
				enum bpf_prog_type type)
{
	return ERR_PTR(-EOPNOTSUPP);
}
1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070

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;
}
1071 1072 1073 1074

static inline void bpf_map_put(struct bpf_map *map)
{
}
1075
#endif /* CONFIG_BPF_SYSCALL */
1076

1077 1078 1079 1080 1081 1082
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);
}

1083 1084
bool bpf_prog_get_ok(struct bpf_prog *, enum bpf_prog_type *, bool);

1085 1086
int bpf_prog_offload_compile(struct bpf_prog *prog);
void bpf_prog_offload_destroy(struct bpf_prog *prog);
1087 1088
int bpf_prog_offload_info_fill(struct bpf_prog_info *info,
			       struct bpf_prog *prog);
1089

1090 1091
int bpf_map_offload_info_fill(struct bpf_map_info *info, struct bpf_map *map);

1092 1093 1094 1095 1096 1097 1098
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);

1099
bool bpf_offload_prog_map_match(struct bpf_prog *prog, struct bpf_map *map);
1100

1101
struct bpf_offload_dev *
1102
bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops, void *priv);
1103
void bpf_offload_dev_destroy(struct bpf_offload_dev *offdev);
1104
void *bpf_offload_dev_priv(struct bpf_offload_dev *offdev);
1105 1106 1107 1108
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);
1109
bool bpf_offload_dev_match(struct bpf_prog *prog, struct net_device *netdev);
1110

1111 1112 1113
#if defined(CONFIG_NET) && defined(CONFIG_BPF_SYSCALL)
int bpf_prog_offload_init(struct bpf_prog *prog, union bpf_attr *attr);

1114
static inline bool bpf_prog_is_dev_bound(const struct bpf_prog_aux *aux)
1115
{
1116
	return aux->offload_requested;
1117
}
1118 1119 1120 1121 1122 1123 1124 1125

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);
1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136
#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;
}
1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150

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)
{
}
1151 1152
#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */

1153 1154 1155
#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);
1156
#else
1157 1158
static inline int sock_map_prog_update(struct bpf_map *map,
				       struct bpf_prog *prog, u32 which)
1159 1160 1161
{
	return -EOPNOTSUPP;
}
1162

1163 1164
static inline int sock_map_get_from_fd(const union bpf_attr *attr,
				       struct bpf_prog *prog)
1165 1166 1167
{
	return -EINVAL;
}
1168 1169
#endif

1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196
#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) */

1197
/* verifier prototypes for helper functions called from eBPF programs */
1198 1199 1200
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
1201 1202 1203
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;
1204

1205
extern const struct bpf_func_proto bpf_get_prandom_u32_proto;
1206
extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
1207
extern const struct bpf_func_proto bpf_get_numa_node_id_proto;
1208
extern const struct bpf_func_proto bpf_tail_call_proto;
1209
extern const struct bpf_func_proto bpf_ktime_get_ns_proto;
1210 1211 1212
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;
1213
extern const struct bpf_func_proto bpf_get_stackid_proto;
Y
Yonghong Song 已提交
1214
extern const struct bpf_func_proto bpf_get_stack_proto;
1215
extern const struct bpf_func_proto bpf_sock_map_update_proto;
1216
extern const struct bpf_func_proto bpf_sock_hash_update_proto;
1217
extern const struct bpf_func_proto bpf_get_current_cgroup_id_proto;
1218 1219 1220 1221
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;
1222 1223
extern const struct bpf_func_proto bpf_spin_lock_proto;
extern const struct bpf_func_proto bpf_spin_unlock_proto;
1224
extern const struct bpf_func_proto bpf_get_local_storage_proto;
1225 1226
extern const struct bpf_func_proto bpf_strtol_proto;
extern const struct bpf_func_proto bpf_strtoul_proto;
1227
extern const struct bpf_func_proto bpf_tcp_sock_proto;
1228

1229 1230 1231 1232
/* 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);

1233
#if defined(CONFIG_NET)
1234 1235 1236
bool bpf_sock_common_is_valid_access(int off, int size,
				     enum bpf_access_type type,
				     struct bpf_insn_access_aux *info);
1237 1238 1239 1240 1241 1242 1243 1244
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
1245 1246 1247 1248 1249 1250
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;
}
1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266
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

1267
#ifdef CONFIG_INET
A
Alexei Starovoitov 已提交
1268 1269 1270 1271 1272 1273 1274 1275 1276
struct sk_reuseport_kern {
	struct sk_buff *skb;
	struct sock *sk;
	struct sock *selected_sk;
	void *data_end;
	u32 hash;
	u32 reuseport_id;
	bool bind_inany;
};
1277 1278 1279 1280 1281 1282 1283 1284
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);
1285 1286 1287 1288 1289 1290 1291 1292 1293

bool bpf_xdp_sock_is_valid_access(int off, int size, enum bpf_access_type type,
				  struct bpf_insn_access_aux *info);

u32 bpf_xdp_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);
1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309
#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;
}
1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324
static inline bool bpf_xdp_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_xdp_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;
}
1325 1326
#endif /* CONFIG_INET */

1327
enum bpf_text_poke_type {
1328 1329
	BPF_MOD_CALL,
	BPF_MOD_JUMP,
1330
};
1331

1332 1333 1334
int bpf_arch_text_poke(void *ip, enum bpf_text_poke_type t,
		       void *addr1, void *addr2);

1335
#endif /* _LINUX_BPF_H */