bpf.h 54.8 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
#include <linux/module.h>
J
Jiri Olsa 已提交
21
#include <linux/kallsyms.h>
A
Alexei Starovoitov 已提交
22
#include <linux/capability.h>
23

24
struct bpf_verifier_env;
25
struct bpf_verifier_log;
26
struct perf_event;
27
struct bpf_prog;
28
struct bpf_prog_aux;
29
struct bpf_map;
30
struct sock;
31
struct seq_file;
32
struct btf;
33
struct btf_type;
34
struct exception_table_entry;
35
struct seq_operations;
36

37 38 39
extern struct idr btf_idr;
extern spinlock_t btf_idr_lock;

40 41 42
/* map is generic key/value storage optionally accesible by eBPF programs */
struct bpf_map_ops {
	/* funcs callable from userspace (via syscall) */
43
	int (*map_alloc_check)(union bpf_attr *attr);
44
	struct bpf_map *(*map_alloc)(union bpf_attr *attr);
45 46
	void (*map_release)(struct bpf_map *map, struct file *map_file);
	void (*map_free)(struct bpf_map *map);
47
	int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key);
48
	void (*map_release_uref)(struct bpf_map *map);
49
	void *(*map_lookup_elem_sys_only)(struct bpf_map *map, void *key);
50 51
	int (*map_lookup_batch)(struct bpf_map *map, const union bpf_attr *attr,
				union bpf_attr __user *uattr);
52 53 54
	int (*map_lookup_and_delete_batch)(struct bpf_map *map,
					   const union bpf_attr *attr,
					   union bpf_attr __user *uattr);
55 56 57 58
	int (*map_update_batch)(struct bpf_map *map, const union bpf_attr *attr,
				union bpf_attr __user *uattr);
	int (*map_delete_batch)(struct bpf_map *map, const union bpf_attr *attr,
				union bpf_attr __user *uattr);
59 60 61

	/* funcs callable from userspace and from eBPF programs */
	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
62
	int (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags);
63
	int (*map_delete_elem)(struct bpf_map *map, void *key);
M
Mauricio Vasquez B 已提交
64 65 66
	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);
67 68

	/* funcs called by prog_array and perf_event_array map */
69 70 71
	void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file,
				int fd);
	void (*map_fd_put_ptr)(void *ptr);
72
	u32 (*map_gen_lookup)(struct bpf_map *map, struct bpf_insn *insn_buf);
73
	u32 (*map_fd_sys_lookup_elem)(void *ptr);
74 75
	void (*map_seq_show_elem)(struct bpf_map *map, void *key,
				  struct seq_file *m);
76
	int (*map_check_btf)(const struct bpf_map *map,
77
			     const struct btf *btf,
78 79
			     const struct btf_type *key_type,
			     const struct btf_type *value_type);
80

81 82 83 84 85 86
	/* 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);

87 88 89 90 91
	/* 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);
92
	int (*map_mmap)(struct bpf_map *map, struct vm_area_struct *vma);
93 94
	__poll_t (*map_poll)(struct bpf_map *map, struct file *filp,
			     struct poll_table_struct *pts);
95 96
};

97 98 99 100 101
struct bpf_map_memory {
	u32 pages;
	struct user_struct *user;
};

102
struct bpf_map {
103
	/* The first two cachelines with read-mostly members of which some
104 105 106 107 108 109 110
	 * 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
111 112 113 114
	enum bpf_map_type map_type;
	u32 key_size;
	u32 value_size;
	u32 max_entries;
115
	u32 map_flags;
116
	int spin_lock_off; /* >=0 valid offset, <0 error */
M
Martin KaFai Lau 已提交
117
	u32 id;
118
	int numa_node;
119 120
	u32 btf_key_type_id;
	u32 btf_value_type_id;
121
	struct btf *btf;
122
	struct bpf_map_memory memory;
123
	char name[BPF_OBJ_NAME_LEN];
124
	u32 btf_vmlinux_value_type_id;
A
Alexei Starovoitov 已提交
125
	bool bypass_spec_v1;
126 127
	bool frozen; /* write-once; write-protected by freeze_mutex */
	/* 22 bytes hole */
128

129
	/* The 3rd and 4th cacheline with misc members to avoid false sharing
130 131
	 * particularly with refcounting.
	 */
132 133
	atomic64_t refcnt ____cacheline_aligned;
	atomic64_t usercnt;
134
	struct work_struct work;
135 136
	struct mutex freeze_mutex;
	u64 writecnt; /* writable mmap cnt; protected by freeze_mutex */
137 138
};

139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165
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);
	}
}
166 167
void copy_map_value_locked(struct bpf_map *map, void *dst, void *src,
			   bool lock_src);
168
int bpf_obj_name_cpy(char *dst, const char *src, unsigned int size);
169

170
struct bpf_offload_dev;
171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195
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);
}

196 197 198 199 200
static inline bool bpf_map_offload_neutral(const struct bpf_map *map)
{
	return map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY;
}

201 202
static inline bool bpf_map_support_seq_show(const struct bpf_map *map)
{
203 204
	return (map->btf_value_type_id || map->btf_vmlinux_value_type_id) &&
		map->ops->map_seq_show_elem;
205 206
}

207
int map_check_no_btf(const struct bpf_map *map,
208
		     const struct btf *btf,
209 210 211
		     const struct btf_type *key_type,
		     const struct btf_type *value_type);

212 213
extern const struct bpf_map_ops bpf_map_offload_ops;

214 215
/* function argument constraints */
enum bpf_arg_type {
216
	ARG_DONTCARE = 0,	/* unused argument in helper function */
217 218 219 220 221 222 223

	/* 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 */
224
	ARG_PTR_TO_UNINIT_MAP_VALUE,	/* pointer to valid memory used to store a map value */
225
	ARG_PTR_TO_MAP_VALUE_OR_NULL,	/* pointer to stack used as map value or NULL */
226 227 228 229

	/* the following constraints used to prototype bpf_memcmp() and other
	 * functions that access data on eBPF program stack
	 */
230
	ARG_PTR_TO_MEM,		/* pointer to valid memory (stack, packet, map value) */
231
	ARG_PTR_TO_MEM_OR_NULL, /* pointer to valid memory or NULL */
232 233 234
	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.
235 236
				 */

237 238
	ARG_CONST_SIZE,		/* number of bytes accessed from memory */
	ARG_CONST_SIZE_OR_ZERO,	/* number of bytes accessed from memory or 0 */
239

240
	ARG_PTR_TO_CTX,		/* pointer to context */
241
	ARG_PTR_TO_CTX_OR_NULL,	/* pointer to context or NULL */
242
	ARG_ANYTHING,		/* any (initialized) argument is ok */
243
	ARG_PTR_TO_SPIN_LOCK,	/* pointer to bpf_spin_lock */
244
	ARG_PTR_TO_SOCK_COMMON,	/* pointer to sock_common */
245 246
	ARG_PTR_TO_INT,		/* pointer to int */
	ARG_PTR_TO_LONG,	/* pointer to long */
247
	ARG_PTR_TO_SOCKET,	/* pointer to bpf_sock (fullsock) */
248
	ARG_PTR_TO_BTF_ID,	/* pointer to in-kernel struct */
249 250 251
	ARG_PTR_TO_ALLOC_MEM,	/* pointer to dynamically allocated memory */
	ARG_PTR_TO_ALLOC_MEM_OR_NULL,	/* pointer to dynamically allocated memory or NULL */
	ARG_CONST_ALLOC_SIZE_OR_ZERO,	/* number of allocated bytes requested */
252 253 254 255 256 257
};

/* type of values returned from helper functions */
enum bpf_return_type {
	RET_INTEGER,			/* function returns integer */
	RET_VOID,			/* function doesn't return anything */
258
	RET_PTR_TO_MAP_VALUE,		/* returns a pointer to map elem value */
259
	RET_PTR_TO_MAP_VALUE_OR_NULL,	/* returns a pointer to map elem value or NULL */
260
	RET_PTR_TO_SOCKET_OR_NULL,	/* returns a pointer to a socket or NULL */
261
	RET_PTR_TO_TCP_SOCK_OR_NULL,	/* returns a pointer to a tcp_sock or NULL */
262
	RET_PTR_TO_SOCK_COMMON_OR_NULL,	/* returns a pointer to a sock_common or NULL */
263
	RET_PTR_TO_ALLOC_MEM_OR_NULL,	/* returns a pointer to dynamically allocated memory or NULL */
264 265
};

266 267 268 269 270 271 272
/* 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;
273
	bool pkt_access;
274
	enum bpf_return_type ret_type;
275 276 277 278 279 280 281 282 283 284
	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];
	};
285
	int *btf_id; /* BTF ids of arguments */
286 287 288 289 290 291 292 293 294 295 296
};

/* 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
297 298
};

299
/* types of values stored in eBPF registers */
300 301 302 303 304 305 306 307 308
/* 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
 */
309 310
enum bpf_reg_type {
	NOT_INIT = 0,		 /* nothing was written into register */
311
	SCALAR_VALUE,		 /* reg doesn't contain a valid pointer */
312 313 314 315
	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 */
316
	PTR_TO_STACK,		 /* reg == frame_pointer + offset */
317
	PTR_TO_PACKET_META,	 /* skb->data - meta_len */
318
	PTR_TO_PACKET,		 /* reg points to skb->data */
319
	PTR_TO_PACKET_END,	 /* skb->data + headlen */
320
	PTR_TO_FLOW_KEYS,	 /* reg points to bpf_flow_keys */
321 322
	PTR_TO_SOCKET,		 /* reg points to struct bpf_sock */
	PTR_TO_SOCKET_OR_NULL,	 /* reg points to struct bpf_sock or NULL */
323 324
	PTR_TO_SOCK_COMMON,	 /* reg points to sock_common */
	PTR_TO_SOCK_COMMON_OR_NULL, /* reg points to sock_common or NULL */
325 326
	PTR_TO_TCP_SOCK,	 /* reg points to struct tcp_sock */
	PTR_TO_TCP_SOCK_OR_NULL, /* reg points to struct tcp_sock or NULL */
327
	PTR_TO_TP_BUFFER,	 /* reg points to a writable raw tp's buffer */
328
	PTR_TO_XDP_SOCK,	 /* reg points to struct xdp_sock */
329
	PTR_TO_BTF_ID,		 /* reg points to kernel struct */
330
	PTR_TO_BTF_ID_OR_NULL,	 /* reg points to kernel struct or NULL */
331 332
	PTR_TO_MEM,		 /* reg points to valid memory region */
	PTR_TO_MEM_OR_NULL,	 /* reg points to valid memory region or NULL */
333 334
};

335 336 337 338 339
/* The information passed from prog-specific *_is_valid_access
 * back to the verifier.
 */
struct bpf_insn_access_aux {
	enum bpf_reg_type reg_type;
340 341 342 343 344
	union {
		int ctx_field_size;
		u32 btf_id;
	};
	struct bpf_verifier_log *log; /* for verbose logs */
345 346
};

347 348 349 350 351 352
static inline void
bpf_ctx_record_field_size(struct bpf_insn_access_aux *aux, u32 size)
{
	aux->ctx_field_size = size;
}

353 354 355 356 357
struct bpf_prog_ops {
	int (*test_run)(struct bpf_prog *prog, const union bpf_attr *kattr,
			union bpf_attr __user *uattr);
};

358 359
struct bpf_verifier_ops {
	/* return eBPF function prototype for verification */
360 361 362
	const struct bpf_func_proto *
	(*get_func_proto)(enum bpf_func_id func_id,
			  const struct bpf_prog *prog);
363 364 365 366

	/* return true if 'size' wide access at offset 'off' within bpf_context
	 * with 'type' (read or write) is allowed
	 */
367
	bool (*is_valid_access)(int off, int size, enum bpf_access_type type,
368
				const struct bpf_prog *prog,
369
				struct bpf_insn_access_aux *info);
370 371
	int (*gen_prologue)(struct bpf_insn *insn, bool direct_write,
			    const struct bpf_prog *prog);
372 373
	int (*gen_ld_abs)(const struct bpf_insn *orig,
			  struct bpf_insn *insn_buf);
374 375 376
	u32 (*convert_ctx_access)(enum bpf_access_type type,
				  const struct bpf_insn *src,
				  struct bpf_insn *dst,
377
				  struct bpf_prog *prog, u32 *target_size);
378 379 380 381
	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);
382 383
};

384
struct bpf_prog_offload_ops {
385
	/* verifier basic callbacks */
386 387
	int (*insn_hook)(struct bpf_verifier_env *env,
			 int insn_idx, int prev_insn_idx);
388
	int (*finalize)(struct bpf_verifier_env *env);
389 390 391 392 393
	/* 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 */
394 395
	int (*prepare)(struct bpf_prog *prog);
	int (*translate)(struct bpf_prog *prog);
396
	void (*destroy)(struct bpf_prog *prog);
397 398
};

399
struct bpf_prog_offload {
400 401
	struct bpf_prog		*prog;
	struct net_device	*netdev;
402
	struct bpf_offload_dev	*offdev;
403 404 405
	void			*dev_priv;
	struct list_head	offloads;
	bool			dev_state;
406
	bool			opt_failed;
407 408
	void			*jited_image;
	u32			jited_len;
409 410
};

411 412
enum bpf_cgroup_storage_type {
	BPF_CGROUP_STORAGE_SHARED,
413
	BPF_CGROUP_STORAGE_PERCPU,
414 415 416 417 418
	__BPF_CGROUP_STORAGE_MAX
};

#define MAX_BPF_CGROUP_STORAGE_TYPE __BPF_CGROUP_STORAGE_MAX

419 420 421 422 423
/* The longest tracepoint has 12 args.
 * See include/trace/bpf_probe.h
 */
#define MAX_BPF_FUNC_ARGS 12

A
Alexei Starovoitov 已提交
424 425 426 427
struct bpf_prog_stats {
	u64 cnt;
	u64 nsecs;
	struct u64_stats_sync syncp;
E
Eric Dumazet 已提交
428
} __aligned(2 * sizeof(u64));
A
Alexei Starovoitov 已提交
429

A
Alexei Starovoitov 已提交
430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
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)

K
KP Singh 已提交
450 451 452 453 454 455 456 457 458 459
/* Each call __bpf_prog_enter + call bpf_func + call __bpf_prog_exit is ~50
 * bytes on x86.  Pick a number to fit into BPF_IMAGE_SIZE / 2
 */
#define BPF_MAX_TRAMP_PROGS 40

struct bpf_tramp_progs {
	struct bpf_prog *progs[BPF_MAX_TRAMP_PROGS];
	int nr_progs;
};

A
Alexei Starovoitov 已提交
460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479
/* 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
 */
480 481
int arch_prepare_bpf_trampoline(void *image, void *image_end,
				const struct btf_func_model *m, u32 flags,
K
KP Singh 已提交
482
				struct bpf_tramp_progs *tprogs,
A
Alexei Starovoitov 已提交
483 484 485 486 487
				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);

J
Jiri Olsa 已提交
488 489 490
struct bpf_ksym {
	unsigned long		 start;
	unsigned long		 end;
J
Jiri Olsa 已提交
491
	char			 name[KSYM_NAME_LEN];
492
	struct list_head	 lnode;
J
Jiri Olsa 已提交
493
	struct latch_tree_node	 tnode;
494
	bool			 prog;
J
Jiri Olsa 已提交
495 496
};

A
Alexei Starovoitov 已提交
497 498 499
enum bpf_tramp_prog_type {
	BPF_TRAMP_FENTRY,
	BPF_TRAMP_FEXIT,
K
KP Singh 已提交
500
	BPF_TRAMP_MODIFY_RETURN,
501 502
	BPF_TRAMP_MAX,
	BPF_TRAMP_REPLACE, /* more than MAX */
A
Alexei Starovoitov 已提交
503 504 505 506 507 508 509 510 511 512 513 514
};

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;
515
		bool ftrace_managed;
A
Alexei Starovoitov 已提交
516
	} func;
517 518 519 520 521
	/* if !NULL this is BPF_PROG_TYPE_EXT program that extends another BPF
	 * program by replacing one of its functions. func.addr is the address
	 * of the function it replaced.
	 */
	struct bpf_prog *extension_prog;
A
Alexei Starovoitov 已提交
522 523 524 525 526 527 528
	/* 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;
J
Jiri Olsa 已提交
529
	struct bpf_ksym ksym;
A
Alexei Starovoitov 已提交
530
};
B
Björn Töpel 已提交
531

532
#define BPF_DISPATCHER_MAX 48 /* Fits in 2048B */
B
Björn Töpel 已提交
533 534 535 536 537 538 539 540 541 542 543 544 545 546

struct bpf_dispatcher_prog {
	struct bpf_prog *prog;
	refcount_t users;
};

struct bpf_dispatcher {
	/* dispatcher mutex */
	struct mutex mutex;
	void *func;
	struct bpf_dispatcher_prog progs[BPF_DISPATCHER_MAX];
	int num_progs;
	void *image;
	u32 image_off;
J
Jiri Olsa 已提交
547
	struct bpf_ksym ksym;
B
Björn Töpel 已提交
548 549
};

550
static __always_inline unsigned int bpf_dispatcher_nop_func(
551 552 553 554 555 556 557
	const void *ctx,
	const struct bpf_insn *insnsi,
	unsigned int (*bpf_func)(const void *,
				 const struct bpf_insn *))
{
	return bpf_func(ctx, insnsi);
}
A
Alexei Starovoitov 已提交
558 559 560 561 562
#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);
J
Jiri Olsa 已提交
563 564 565 566 567 568 569 570 571 572 573
#define BPF_DISPATCHER_INIT(_name) {				\
	.mutex = __MUTEX_INITIALIZER(_name.mutex),		\
	.func = &_name##_func,					\
	.progs = {},						\
	.num_progs = 0,						\
	.image = NULL,						\
	.image_off = 0,						\
	.ksym = {						\
		.name  = #_name,				\
		.lnode = LIST_HEAD_INIT(_name.ksym.lnode),	\
	},							\
B
Björn Töpel 已提交
574 575 576
}

#define DEFINE_BPF_DISPATCHER(name)					\
577
	noinline unsigned int bpf_dispatcher_##name##_func(		\
B
Björn Töpel 已提交
578 579 580 581 582 583 584
		const void *ctx,					\
		const struct bpf_insn *insnsi,				\
		unsigned int (*bpf_func)(const void *,			\
					 const struct bpf_insn *))	\
	{								\
		return bpf_func(ctx, insnsi);				\
	}								\
585 586 587
	EXPORT_SYMBOL(bpf_dispatcher_##name##_func);			\
	struct bpf_dispatcher bpf_dispatcher_##name =			\
		BPF_DISPATCHER_INIT(bpf_dispatcher_##name);
B
Björn Töpel 已提交
588
#define DECLARE_BPF_DISPATCHER(name)					\
589
	unsigned int bpf_dispatcher_##name##_func(			\
B
Björn Töpel 已提交
590 591 592 593
		const void *ctx,					\
		const struct bpf_insn *insnsi,				\
		unsigned int (*bpf_func)(const void *,			\
					 const struct bpf_insn *));	\
594 595 596
	extern struct bpf_dispatcher bpf_dispatcher_##name;
#define BPF_DISPATCHER_FUNC(name) bpf_dispatcher_##name##_func
#define BPF_DISPATCHER_PTR(name) (&bpf_dispatcher_##name)
B
Björn Töpel 已提交
597 598
void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
				struct bpf_prog *to);
J
Jiri Olsa 已提交
599
/* Called only from JIT-enabled code, so there's no need for stubs. */
J
Jiri Olsa 已提交
600
void *bpf_jit_alloc_exec_page(void);
J
Jiri Olsa 已提交
601 602
void bpf_image_ksym_add(void *data, struct bpf_ksym *ksym);
void bpf_image_ksym_del(struct bpf_ksym *ksym);
J
Jiri Olsa 已提交
603 604
void bpf_ksym_add(struct bpf_ksym *ksym);
void bpf_ksym_del(struct bpf_ksym *ksym);
A
Alexei Starovoitov 已提交
605 606 607 608 609 610 611 612 613 614 615 616 617 618
#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) {}
B
Björn Töpel 已提交
619 620
#define DEFINE_BPF_DISPATCHER(name)
#define DECLARE_BPF_DISPATCHER(name)
621
#define BPF_DISPATCHER_FUNC(name) bpf_dispatcher_nop_func
B
Björn Töpel 已提交
622 623 624 625
#define BPF_DISPATCHER_PTR(name) NULL
static inline void bpf_dispatcher_change_prog(struct bpf_dispatcher *d,
					      struct bpf_prog *from,
					      struct bpf_prog *to) {}
626 627 628 629
static inline bool is_bpf_image_address(unsigned long address)
{
	return false;
}
A
Alexei Starovoitov 已提交
630 631
#endif

632
struct bpf_func_info_aux {
633
	u16 linkage;
634 635 636
	bool unreliable;
};

637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654
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;
};

655 656 657 658 659 660
/* reg_type info for ctx arguments */
struct bpf_ctx_arg_aux {
	u32 offset;
	enum bpf_reg_type reg_type;
};

661
struct bpf_prog_aux {
662
	atomic64_t refcnt;
663
	u32 used_map_cnt;
664
	u32 max_ctx_offset;
665
	u32 max_pkt_offset;
666
	u32 max_tp_access;
667
	u32 stack_depth;
M
Martin KaFai Lau 已提交
668
	u32 id;
669 670
	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 */
671
	u32 attach_btf_id; /* in-kernel BTF type id to attach to */
672 673
	u32 ctx_arg_info_size;
	const struct bpf_ctx_arg_aux *ctx_arg_info;
674
	struct bpf_prog *linked_prog;
675
	bool verifier_zext; /* Zero extensions has been inserted by verifier. */
676
	bool offload_requested;
677
	bool attach_btf_trace; /* true if attaching to BTF-enabled raw tp */
678
	bool func_proto_unreliable;
A
Alexei Starovoitov 已提交
679 680 681
	enum bpf_tramp_prog_type trampoline_prog_type;
	struct bpf_trampoline *trampoline;
	struct hlist_node tramp_hlist;
682 683 684 685
	/* 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;
686 687
	struct bpf_prog **func;
	void *jit_data; /* JIT specific data. arch dependent */
688 689
	struct bpf_jit_poke_descriptor *poke_tab;
	u32 size_poke_tab;
J
Jiri Olsa 已提交
690
	struct bpf_ksym ksym;
691
	const struct bpf_prog_ops *ops;
692 693
	struct bpf_map **used_maps;
	struct bpf_prog *prog;
694
	struct user_struct *user;
695
	u64 load_time; /* ns since boottime */
696
	struct bpf_map *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
697
	char name[BPF_OBJ_NAME_LEN];
698 699 700
#ifdef CONFIG_SECURITY
	void *security;
#endif
701
	struct bpf_prog_offload *offload;
Y
Yonghong Song 已提交
702
	struct btf *btf;
703
	struct bpf_func_info *func_info;
704
	struct bpf_func_info_aux *func_info_aux;
M
Martin KaFai Lau 已提交
705 706 707 708 709 710 711 712 713 714 715 716 717 718 719
	/* 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;
720
	u32 func_info_cnt;
M
Martin KaFai Lau 已提交
721 722 723 724 725 726
	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;
727 728
	u32 num_exentries;
	struct exception_table_entry *extable;
A
Alexei Starovoitov 已提交
729
	struct bpf_prog_stats __percpu *stats;
730 731 732 733
	union {
		struct work_struct work;
		struct rcu_head	rcu;
	};
734 735
};

736 737 738 739 740 741 742 743
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;
744 745 746 747 748
	/* 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;
749 750
};

751
struct bpf_struct_ops_value;
752 753 754 755 756 757 758 759 760
struct btf_type;
struct btf_member;

#define BPF_STRUCT_OPS_MAX_NR_MEMBERS 64
struct bpf_struct_ops {
	const struct bpf_verifier_ops *verifier_ops;
	int (*init)(struct btf *btf);
	int (*check_member)(const struct btf_type *t,
			    const struct btf_member *member);
761 762 763 764 765
	int (*init_member)(const struct btf_type *t,
			   const struct btf_member *member,
			   void *kdata, const void *udata);
	int (*reg)(void *kdata);
	void (*unreg)(void *kdata);
766
	const struct btf_type *type;
767
	const struct btf_type *value_type;
768 769 770
	const char *name;
	struct btf_func_model func_models[BPF_STRUCT_OPS_MAX_NR_MEMBERS];
	u32 type_id;
771
	u32 value_id;
772 773 774
};

#if defined(CONFIG_BPF_JIT) && defined(CONFIG_BPF_SYSCALL)
775
#define BPF_MODULE_OWNER ((void *)((0xeB9FUL << 2) + POISON_POINTER_DELTA))
776
const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id);
777
void bpf_struct_ops_init(struct btf *btf, struct bpf_verifier_log *log);
778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795
bool bpf_struct_ops_get(const void *kdata);
void bpf_struct_ops_put(const void *kdata);
int bpf_struct_ops_map_sys_lookup_elem(struct bpf_map *map, void *key,
				       void *value);
static inline bool bpf_try_module_get(const void *data, struct module *owner)
{
	if (owner == BPF_MODULE_OWNER)
		return bpf_struct_ops_get(data);
	else
		return try_module_get(owner);
}
static inline void bpf_module_put(const void *data, struct module *owner)
{
	if (owner == BPF_MODULE_OWNER)
		bpf_struct_ops_put(data);
	else
		module_put(owner);
}
796 797 798 799 800
#else
static inline const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id)
{
	return NULL;
}
801 802 803 804
static inline void bpf_struct_ops_init(struct btf *btf,
				       struct bpf_verifier_log *log)
{
}
805 806 807 808 809 810 811 812 813 814 815 816 817 818
static inline bool bpf_try_module_get(const void *data, struct module *owner)
{
	return try_module_get(owner);
}
static inline void bpf_module_put(const void *data, struct module *owner)
{
	module_put(owner);
}
static inline int bpf_struct_ops_map_sys_lookup_elem(struct bpf_map *map,
						     void *key,
						     void *value)
{
	return -EINVAL;
}
819 820
#endif

821 822 823
struct bpf_array {
	struct bpf_map map;
	u32 elem_size;
824
	u32 index_mask;
825
	struct bpf_array_aux *aux;
826 827
	union {
		char value[0] __aligned(8);
828
		void *ptrs[0] __aligned(8);
829
		void __percpu *pptrs[0] __aligned(8);
830 831
	};
};
832

833
#define BPF_COMPLEXITY_LIMIT_INSNS      1000000 /* yes. 1M insns */
834 835
#define MAX_TAIL_CALL_CNT 32

836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864
#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);
}

865 866 867 868 869 870 871
struct bpf_event_entry {
	struct perf_event *event;
	struct file *perf_file;
	struct file *map_file;
	struct rcu_head rcu;
};

872
bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp);
873
int bpf_prog_calc_tag(struct bpf_prog *fp);
874
const char *kernel_type_name(u32 btf_type_id);
875

876
const struct bpf_func_proto *bpf_get_trace_printk_proto(void);
877 878

typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src,
879
					unsigned long off, unsigned long len);
880 881 882 883 884
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);
885 886 887

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);
888

889 890 891 892 893 894 895 896 897 898 899 900
/* 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.
 */
901 902
struct bpf_prog_array_item {
	struct bpf_prog *prog;
903
	struct bpf_cgroup_storage *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
904 905
};

906 907
struct bpf_prog_array {
	struct rcu_head rcu;
908
	struct bpf_prog_array_item items[];
909 910
};

911
struct bpf_prog_array *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags);
912 913
void bpf_prog_array_free(struct bpf_prog_array *progs);
int bpf_prog_array_length(struct bpf_prog_array *progs);
914
bool bpf_prog_array_is_empty(struct bpf_prog_array *array);
915
int bpf_prog_array_copy_to_user(struct bpf_prog_array *progs,
916
				__u32 __user *prog_ids, u32 cnt);
917

918
void bpf_prog_array_delete_safe(struct bpf_prog_array *progs,
919
				struct bpf_prog *old_prog);
920
int bpf_prog_array_copy_info(struct bpf_prog_array *array,
921 922
			     u32 *prog_ids, u32 request_cnt,
			     u32 *prog_cnt);
923
int bpf_prog_array_copy(struct bpf_prog_array *old_array,
924 925 926 927 928
			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)	\
929
	({						\
930 931
		struct bpf_prog_array_item *_item;	\
		struct bpf_prog *_prog;			\
932
		struct bpf_prog_array *_array;		\
933
		u32 _ret = 1;				\
934
		migrate_disable();			\
935
		rcu_read_lock();			\
936 937 938
		_array = rcu_dereference(array);	\
		if (unlikely(check_non_null && !_array))\
			goto _out;			\
939 940 941 942 943
		_item = &_array->items[0];		\
		while ((_prog = READ_ONCE(_item->prog))) {		\
			bpf_cgroup_storage_set(_item->cgroup_storage);	\
			_ret &= func(_prog, ctx);	\
			_item++;			\
944 945
		}					\
_out:							\
946
		rcu_read_unlock();			\
947
		migrate_enable();			\
948 949 950
		_ret;					\
	 })

951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980
/* 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;				\
981
		migrate_disable();			\
982 983 984 985 986 987 988 989 990 991 992
		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();			\
993
		migrate_enable();			\
994 995 996 997 998 999 1000
		if (_ret)				\
			_ret = (_cn ? NET_XMIT_CN : NET_XMIT_SUCCESS);	\
		else					\
			_ret = (_cn ? NET_XMIT_DROP : -EPERM);		\
		_ret;					\
	})

1001 1002 1003 1004 1005 1006
#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)

1007
#ifdef CONFIG_BPF_SYSCALL
1008
DECLARE_PER_CPU(int, bpf_prog_active);
1009
extern struct mutex bpf_stats_enabled_mutex;
1010

1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040
/*
 * Block execution of BPF programs attached to instrumentation (perf,
 * kprobes, tracepoints) to prevent deadlocks on map operations as any of
 * these events can happen inside a region which holds a map bucket lock
 * and can deadlock on it.
 *
 * Use the preemption safe inc/dec variants on RT because migrate disable
 * is preemptible on RT and preemption in the middle of the RMW operation
 * might lead to inconsistent state. Use the raw variants for non RT
 * kernels as migrate_disable() maps to preempt_disable() so the slightly
 * more expensive save operation can be avoided.
 */
static inline void bpf_disable_instrumentation(void)
{
	migrate_disable();
	if (IS_ENABLED(CONFIG_PREEMPT_RT))
		this_cpu_inc(bpf_prog_active);
	else
		__this_cpu_inc(bpf_prog_active);
}

static inline void bpf_enable_instrumentation(void)
{
	if (IS_ENABLED(CONFIG_PREEMPT_RT))
		this_cpu_dec(bpf_prog_active);
	else
		__this_cpu_dec(bpf_prog_active);
	migrate_enable();
}

1041 1042
extern const struct file_operations bpf_map_fops;
extern const struct file_operations bpf_prog_fops;
Y
Yonghong Song 已提交
1043
extern const struct file_operations bpf_iter_fops;
1044

A
Alexei Starovoitov 已提交
1045
#define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \
1046 1047
	extern const struct bpf_prog_ops _name ## _prog_ops; \
	extern const struct bpf_verifier_ops _name ## _verifier_ops;
1048 1049
#define BPF_MAP_TYPE(_id, _ops) \
	extern const struct bpf_map_ops _ops;
1050
#define BPF_LINK_TYPE(_id, _name)
1051 1052
#include <linux/bpf_types.h>
#undef BPF_PROG_TYPE
1053
#undef BPF_MAP_TYPE
1054
#undef BPF_LINK_TYPE
1055

1056
extern const struct bpf_prog_ops bpf_offload_prog_ops;
1057 1058 1059
extern const struct bpf_verifier_ops tc_cls_act_analyzer_ops;
extern const struct bpf_verifier_ops xdp_analyzer_ops;

1060
struct bpf_prog *bpf_prog_get(u32 ufd);
1061
struct bpf_prog *bpf_prog_get_type_dev(u32 ufd, enum bpf_prog_type type,
1062
				       bool attach_drv);
1063
void bpf_prog_add(struct bpf_prog *prog, int i);
1064
void bpf_prog_sub(struct bpf_prog *prog, int i);
1065
void bpf_prog_inc(struct bpf_prog *prog);
1066
struct bpf_prog * __must_check bpf_prog_inc_not_zero(struct bpf_prog *prog);
1067
void bpf_prog_put(struct bpf_prog *prog);
1068 1069
int __bpf_prog_charge(struct user_struct *user, u32 pages);
void __bpf_prog_uncharge(struct user_struct *user, u32 pages);
1070 1071
void __bpf_free_used_maps(struct bpf_prog_aux *aux,
			  struct bpf_map **used_maps, u32 len);
1072

1073
void bpf_prog_free_id(struct bpf_prog *prog, bool do_idr_lock);
1074
void bpf_map_free_id(struct bpf_map *map, bool do_idr_lock);
1075

1076
struct bpf_map *bpf_map_get(u32 ufd);
1077
struct bpf_map *bpf_map_get_with_uref(u32 ufd);
1078
struct bpf_map *__bpf_map_get(struct fd f);
1079 1080 1081
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);
1082
void bpf_map_put_with_uref(struct bpf_map *map);
1083
void bpf_map_put(struct bpf_map *map);
1084 1085
int bpf_map_charge_memlock(struct bpf_map *map, u32 pages);
void bpf_map_uncharge_memlock(struct bpf_map *map, u32 pages);
1086
int bpf_map_charge_init(struct bpf_map_memory *mem, u64 size);
1087 1088 1089
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);
1090 1091
void *bpf_map_area_alloc(u64 size, int numa_node);
void *bpf_map_area_mmapable_alloc(u64 size, int numa_node);
1092
void bpf_map_area_free(void *base);
1093
void bpf_map_init_from_attr(struct bpf_map *map, union bpf_attr *attr);
1094 1095
int  generic_map_lookup_batch(struct bpf_map *map,
			      const union bpf_attr *attr,
1096 1097 1098 1099 1100 1101
			      union bpf_attr __user *uattr);
int  generic_map_update_batch(struct bpf_map *map,
			      const union bpf_attr *attr,
			      union bpf_attr __user *uattr);
int  generic_map_delete_batch(struct bpf_map *map,
			      const union bpf_attr *attr,
1102
			      union bpf_attr __user *uattr);
Y
Yonghong Song 已提交
1103
struct bpf_map *bpf_map_get_curr_or_next(u32 *id);
1104

1105 1106
extern int sysctl_unprivileged_bpf_disabled;

A
Alexei Starovoitov 已提交
1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121
static inline bool bpf_allow_ptr_leaks(void)
{
	return perfmon_capable();
}

static inline bool bpf_bypass_spec_v1(void)
{
	return perfmon_capable();
}

static inline bool bpf_bypass_spec_v4(void)
{
	return perfmon_capable();
}

1122
int bpf_map_new_fd(struct bpf_map *map, int flags);
1123 1124
int bpf_prog_new_fd(struct bpf_prog *prog);

1125 1126
struct bpf_link {
	atomic64_t refcnt;
A
Andrii Nakryiko 已提交
1127
	u32 id;
1128
	enum bpf_link_type type;
1129 1130 1131 1132
	const struct bpf_link_ops *ops;
	struct bpf_prog *prog;
	struct work_struct work;
};
1133

A
Andrii Nakryiko 已提交
1134 1135 1136 1137 1138 1139 1140
struct bpf_link_primer {
	struct bpf_link *link;
	struct file *file;
	int fd;
	u32 id;
};

1141 1142
struct bpf_link_ops {
	void (*release)(struct bpf_link *link);
1143
	void (*dealloc)(struct bpf_link *link);
1144 1145
	int (*update_prog)(struct bpf_link *link, struct bpf_prog *new_prog,
			   struct bpf_prog *old_prog);
1146 1147 1148
	void (*show_fdinfo)(const struct bpf_link *link, struct seq_file *seq);
	int (*fill_link_info)(const struct bpf_link *link,
			      struct bpf_link_info *info);
1149 1150
};

1151
void bpf_link_init(struct bpf_link *link, enum bpf_link_type type,
A
Andrii Nakryiko 已提交
1152 1153 1154 1155
		   const struct bpf_link_ops *ops, struct bpf_prog *prog);
int bpf_link_prime(struct bpf_link *link, struct bpf_link_primer *primer);
int bpf_link_settle(struct bpf_link_primer *primer);
void bpf_link_cleanup(struct bpf_link_primer *primer);
1156 1157 1158
void bpf_link_inc(struct bpf_link *link);
void bpf_link_put(struct bpf_link *link);
int bpf_link_new_fd(struct bpf_link *link);
1159
struct file *bpf_link_new_file(struct bpf_link *link, int *reserved_fd);
1160 1161
struct bpf_link *bpf_link_get_from_fd(u32 ufd);

1162
int bpf_obj_pin_user(u32 ufd, const char __user *pathname);
1163
int bpf_obj_get_user(const char __user *pathname, int flags);
1164

1165
#define BPF_ITER_FUNC_PREFIX "bpf_iter_"
1166
#define DEFINE_BPF_ITER_FUNC(target, args...)			\
1167 1168
	extern int bpf_iter_ ## target(args);			\
	int __init bpf_iter_ ## target(args) { return 0; }
1169

1170 1171 1172
typedef int (*bpf_iter_init_seq_priv_t)(void *private_data);
typedef void (*bpf_iter_fini_seq_priv_t)(void *private_data);

1173
#define BPF_ITER_CTX_ARG_MAX 2
1174 1175 1176 1177 1178 1179
struct bpf_iter_reg {
	const char *target;
	const struct seq_operations *seq_ops;
	bpf_iter_init_seq_priv_t init_seq_private;
	bpf_iter_fini_seq_priv_t fini_seq_private;
	u32 seq_priv_size;
1180 1181
	u32 ctx_arg_info_size;
	struct bpf_ctx_arg_aux ctx_arg_info[BPF_ITER_CTX_ARG_MAX];
1182 1183
};

1184 1185 1186 1187 1188 1189
struct bpf_iter_meta {
	__bpf_md_ptr(struct seq_file *, seq);
	u64 session_id;
	u64 seq_num;
};

1190
int bpf_iter_reg_target(const struct bpf_iter_reg *reg_info);
1191
void bpf_iter_unreg_target(const struct bpf_iter_reg *reg_info);
1192
bool bpf_iter_prog_supported(struct bpf_prog *prog);
1193
int bpf_iter_link_attach(const union bpf_attr *attr, struct bpf_prog *prog);
1194
int bpf_iter_new_fd(struct bpf_link *link);
Y
Yonghong Song 已提交
1195
bool bpf_link_is_iter(struct bpf_link *link);
1196 1197
struct bpf_prog *bpf_iter_get_info(struct bpf_iter_meta *meta, bool in_stop);
int bpf_iter_run_prog(struct bpf_prog *prog, void *ctx);
1198

1199 1200 1201 1202 1203 1204
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);
1205

1206
int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value);
1207

1208 1209
int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file,
				 void *key, void *value, u64 map_flags);
1210
int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
M
Martin KaFai Lau 已提交
1211 1212
int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file,
				void *key, void *value, u64 map_flags);
1213
int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
1214

1215
int bpf_get_file_flag(int flags);
1216 1217
int bpf_check_uarg_tail_zero(void __user *uaddr, size_t expected_size,
			     size_t actual_size);
1218

1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234
/* 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++;
}

1235
/* verify correctness of eBPF program */
Y
Yonghong Song 已提交
1236 1237
int bpf_check(struct bpf_prog **fp, union bpf_attr *attr,
	      union bpf_attr __user *uattr);
1238
void bpf_patch_call_args(struct bpf_insn *insn, u32 stack_depth);
1239 1240

/* Map specifics */
1241
struct xdp_buff;
1242
struct sk_buff;
1243 1244

struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
1245
struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
1246 1247 1248
void __dev_flush(void);
int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
1249 1250
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
1251 1252
int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
			     struct bpf_prog *xdp_prog);
1253
bool dev_map_can_have_prog(struct bpf_map *map);
1254

1255
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
1256
void __cpu_map_flush(void);
1257 1258 1259
int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
		    struct net_device *dev_rx);

1260 1261 1262 1263 1264 1265 1266
/* 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;
}

1267
struct bpf_prog *bpf_prog_get_type_path(const char *name, enum bpf_prog_type type);
1268
int array_map_alloc_check(union bpf_attr *attr);
1269

1270 1271 1272 1273
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);
1274 1275 1276
int bpf_prog_test_run_tracing(struct bpf_prog *prog,
			      const union bpf_attr *kattr,
			      union bpf_attr __user *uattr);
1277 1278 1279
int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog,
				     const union bpf_attr *kattr,
				     union bpf_attr __user *uattr);
1280 1281 1282 1283 1284 1285 1286
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);
1287 1288
int btf_resolve_helper_id(struct bpf_verifier_log *log,
			  const struct bpf_func_proto *fn, int);
1289

A
Alexei Starovoitov 已提交
1290 1291 1292 1293 1294 1295
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);

1296 1297 1298 1299 1300
struct bpf_reg_state;
int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog,
			     struct bpf_reg_state *regs);
int btf_prepare_func_args(struct bpf_verifier_env *env, int subprog,
			  struct bpf_reg_state *reg);
1301 1302
int btf_check_type_match(struct bpf_verifier_env *env, struct bpf_prog *prog,
			 struct btf *btf, const struct btf_type *t);
1303

1304 1305
struct bpf_prog *bpf_prog_by_id(u32 id);

1306
const struct bpf_func_proto *bpf_base_func_proto(enum bpf_func_id func_id);
1307
#else /* !CONFIG_BPF_SYSCALL */
1308 1309 1310 1311 1312
static inline struct bpf_prog *bpf_prog_get(u32 ufd)
{
	return ERR_PTR(-EOPNOTSUPP);
}

1313 1314
static inline struct bpf_prog *bpf_prog_get_type_dev(u32 ufd,
						     enum bpf_prog_type type,
1315
						     bool attach_drv)
1316 1317 1318 1319
{
	return ERR_PTR(-EOPNOTSUPP);
}

1320
static inline void bpf_prog_add(struct bpf_prog *prog, int i)
1321 1322
{
}
1323

1324 1325 1326 1327
static inline void bpf_prog_sub(struct bpf_prog *prog, int i)
{
}

1328 1329 1330
static inline void bpf_prog_put(struct bpf_prog *prog)
{
}
1331

1332
static inline void bpf_prog_inc(struct bpf_prog *prog)
1333 1334
{
}
1335

1336 1337 1338 1339 1340 1341
static inline struct bpf_prog *__must_check
bpf_prog_inc_not_zero(struct bpf_prog *prog)
{
	return ERR_PTR(-EOPNOTSUPP);
}

1342 1343 1344 1345 1346 1347 1348 1349
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)
{
}
1350

1351
static inline int bpf_obj_get_user(const char __user *pathname, int flags)
1352 1353 1354 1355
{
	return -EOPNOTSUPP;
}

1356 1357 1358 1359 1360 1361
static inline struct net_device  *__dev_map_lookup_elem(struct bpf_map *map,
						       u32 key)
{
	return NULL;
}

1362 1363 1364 1365 1366
static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map,
							     u32 key)
{
	return NULL;
}
1367 1368 1369 1370
static inline bool dev_map_can_have_prog(struct bpf_map *map)
{
	return false;
}
1371

1372
static inline void __dev_flush(void)
1373 1374
{
}
1375

1376 1377 1378
struct xdp_buff;
struct bpf_dtab_netdev;

1379 1380 1381 1382 1383 1384 1385
static inline
int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
		    struct net_device *dev_rx)
{
	return 0;
}

1386
static inline
1387 1388
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx)
1389 1390 1391 1392
{
	return 0;
}

1393 1394 1395 1396 1397 1398 1399 1400 1401
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;
}

1402 1403 1404 1405 1406 1407
static inline
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key)
{
	return NULL;
}

1408
static inline void __cpu_map_flush(void)
1409 1410 1411 1412 1413 1414 1415 1416 1417
{
}

static inline int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu,
				  struct xdp_buff *xdp,
				  struct net_device *dev_rx)
{
	return 0;
}
1418 1419 1420 1421 1422 1423

static inline struct bpf_prog *bpf_prog_get_type_path(const char *name,
				enum bpf_prog_type type)
{
	return ERR_PTR(-EOPNOTSUPP);
}
1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438

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;
}

1439 1440 1441 1442 1443 1444 1445
static inline int bpf_prog_test_run_tracing(struct bpf_prog *prog,
					    const union bpf_attr *kattr,
					    union bpf_attr __user *uattr)
{
	return -ENOTSUPP;
}

1446 1447 1448 1449 1450 1451
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;
}
1452 1453 1454 1455

static inline void bpf_map_put(struct bpf_map *map)
{
}
1456 1457 1458 1459 1460

static inline struct bpf_prog *bpf_prog_by_id(u32 id)
{
	return ERR_PTR(-ENOTSUPP);
}
1461 1462 1463 1464 1465 1466

static inline const struct bpf_func_proto *
bpf_base_func_proto(enum bpf_func_id func_id)
{
	return NULL;
}
1467
#endif /* CONFIG_BPF_SYSCALL */
1468

1469 1470 1471 1472 1473 1474
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);
}

1475 1476
bool bpf_prog_get_ok(struct bpf_prog *, enum bpf_prog_type *, bool);

1477 1478
int bpf_prog_offload_compile(struct bpf_prog *prog);
void bpf_prog_offload_destroy(struct bpf_prog *prog);
1479 1480
int bpf_prog_offload_info_fill(struct bpf_prog_info *info,
			       struct bpf_prog *prog);
1481

1482 1483
int bpf_map_offload_info_fill(struct bpf_map_info *info, struct bpf_map *map);

1484 1485 1486 1487 1488 1489 1490
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);

1491
bool bpf_offload_prog_map_match(struct bpf_prog *prog, struct bpf_map *map);
1492

1493
struct bpf_offload_dev *
1494
bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops, void *priv);
1495
void bpf_offload_dev_destroy(struct bpf_offload_dev *offdev);
1496
void *bpf_offload_dev_priv(struct bpf_offload_dev *offdev);
1497 1498 1499 1500
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);
1501
bool bpf_offload_dev_match(struct bpf_prog *prog, struct net_device *netdev);
1502

1503 1504 1505
#if defined(CONFIG_NET) && defined(CONFIG_BPF_SYSCALL)
int bpf_prog_offload_init(struct bpf_prog *prog, union bpf_attr *attr);

1506
static inline bool bpf_prog_is_dev_bound(const struct bpf_prog_aux *aux)
1507
{
1508
	return aux->offload_requested;
1509
}
1510 1511 1512 1513 1514 1515 1516 1517

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);
1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528
#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;
}
1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542

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)
{
}
1543 1544
#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */

1545 1546 1547
#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);
1548 1549
void sock_map_unhash(struct sock *sk);
void sock_map_close(struct sock *sk, long timeout);
1550
#else
1551 1552
static inline int sock_map_prog_update(struct bpf_map *map,
				       struct bpf_prog *prog, u32 which)
1553 1554 1555
{
	return -EOPNOTSUPP;
}
1556

1557 1558
static inline int sock_map_get_from_fd(const union bpf_attr *attr,
				       struct bpf_prog *prog)
1559 1560 1561
{
	return -EINVAL;
}
1562
#endif /* CONFIG_BPF_STREAM_PARSER */
1563

1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590
#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) */

1591
/* verifier prototypes for helper functions called from eBPF programs */
1592 1593 1594
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
1595 1596 1597
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;
1598

1599
extern const struct bpf_func_proto bpf_get_prandom_u32_proto;
1600
extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
1601
extern const struct bpf_func_proto bpf_get_numa_node_id_proto;
1602
extern const struct bpf_func_proto bpf_tail_call_proto;
1603
extern const struct bpf_func_proto bpf_ktime_get_ns_proto;
1604
extern const struct bpf_func_proto bpf_ktime_get_boot_ns_proto;
1605 1606 1607
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;
1608
extern const struct bpf_func_proto bpf_get_stackid_proto;
Y
Yonghong Song 已提交
1609
extern const struct bpf_func_proto bpf_get_stack_proto;
1610
extern const struct bpf_func_proto bpf_sock_map_update_proto;
1611
extern const struct bpf_func_proto bpf_sock_hash_update_proto;
1612
extern const struct bpf_func_proto bpf_get_current_cgroup_id_proto;
1613
extern const struct bpf_func_proto bpf_get_current_ancestor_cgroup_id_proto;
1614 1615 1616 1617
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;
1618 1619
extern const struct bpf_func_proto bpf_spin_lock_proto;
extern const struct bpf_func_proto bpf_spin_unlock_proto;
1620
extern const struct bpf_func_proto bpf_get_local_storage_proto;
1621 1622
extern const struct bpf_func_proto bpf_strtol_proto;
extern const struct bpf_func_proto bpf_strtoul_proto;
1623
extern const struct bpf_func_proto bpf_tcp_sock_proto;
M
Martin KaFai Lau 已提交
1624
extern const struct bpf_func_proto bpf_jiffies64_proto;
1625
extern const struct bpf_func_proto bpf_get_ns_current_pid_tgid_proto;
1626
extern const struct bpf_func_proto bpf_event_output_data_proto;
1627 1628 1629 1630 1631
extern const struct bpf_func_proto bpf_ringbuf_output_proto;
extern const struct bpf_func_proto bpf_ringbuf_reserve_proto;
extern const struct bpf_func_proto bpf_ringbuf_submit_proto;
extern const struct bpf_func_proto bpf_ringbuf_discard_proto;
extern const struct bpf_func_proto bpf_ringbuf_query_proto;
1632

K
KP Singh 已提交
1633 1634 1635
const struct bpf_func_proto *bpf_tracing_func_proto(
	enum bpf_func_id func_id, const struct bpf_prog *prog);

1636 1637 1638
const struct bpf_func_proto *tracing_prog_func_proto(
  enum bpf_func_id func_id, const struct bpf_prog *prog);

1639 1640 1641
/* 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);
1642
u64 bpf_get_raw_cpu_id(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
1643

1644
#if defined(CONFIG_NET)
1645 1646 1647
bool bpf_sock_common_is_valid_access(int off, int size,
				     enum bpf_access_type type,
				     struct bpf_insn_access_aux *info);
1648 1649 1650 1651 1652 1653 1654 1655
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
1656 1657 1658 1659 1660 1661
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;
}
1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677
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

1678
#ifdef CONFIG_INET
A
Alexei Starovoitov 已提交
1679 1680 1681 1682 1683 1684 1685 1686 1687
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;
};
1688 1689 1690 1691 1692 1693 1694 1695
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);
1696 1697 1698 1699 1700 1701 1702 1703 1704

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);
1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720
#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;
}
1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735
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;
}
1736 1737
#endif /* CONFIG_INET */

1738
enum bpf_text_poke_type {
1739 1740
	BPF_MOD_CALL,
	BPF_MOD_JUMP,
1741
};
1742

1743 1744 1745
int bpf_arch_text_poke(void *ip, enum bpf_text_poke_type t,
		       void *addr1, void *addr2);

1746
#endif /* _LINUX_BPF_H */