bpf.h 60.2 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
struct bpf_iter_aux_info;
K
KP Singh 已提交
37 38
struct bpf_local_storage;
struct bpf_local_storage_map;
39

40 41 42
extern struct idr btf_idr;
extern spinlock_t btf_idr_lock;

43 44
typedef int (*bpf_iter_init_seq_priv_t)(void *private_data,
					struct bpf_iter_aux_info *aux);
45 46 47 48 49 50 51 52
typedef void (*bpf_iter_fini_seq_priv_t)(void *private_data);
struct bpf_iter_seq_info {
	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;
};

53 54 55
/* map is generic key/value storage optionally accesible by eBPF programs */
struct bpf_map_ops {
	/* funcs callable from userspace (via syscall) */
56
	int (*map_alloc_check)(union bpf_attr *attr);
57
	struct bpf_map *(*map_alloc)(union bpf_attr *attr);
58 59
	void (*map_release)(struct bpf_map *map, struct file *map_file);
	void (*map_free)(struct bpf_map *map);
60
	int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key);
61
	void (*map_release_uref)(struct bpf_map *map);
62
	void *(*map_lookup_elem_sys_only)(struct bpf_map *map, void *key);
63 64
	int (*map_lookup_batch)(struct bpf_map *map, const union bpf_attr *attr,
				union bpf_attr __user *uattr);
65 66 67
	int (*map_lookup_and_delete_batch)(struct bpf_map *map,
					   const union bpf_attr *attr,
					   union bpf_attr __user *uattr);
68 69 70 71
	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);
72 73 74

	/* funcs callable from userspace and from eBPF programs */
	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
75
	int (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags);
76
	int (*map_delete_elem)(struct bpf_map *map, void *key);
M
Mauricio Vasquez B 已提交
77 78 79
	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);
80 81

	/* funcs called by prog_array and perf_event_array map */
82 83 84
	void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file,
				int fd);
	void (*map_fd_put_ptr)(void *ptr);
85
	u32 (*map_gen_lookup)(struct bpf_map *map, struct bpf_insn *insn_buf);
86
	u32 (*map_fd_sys_lookup_elem)(void *ptr);
87 88
	void (*map_seq_show_elem)(struct bpf_map *map, void *key,
				  struct seq_file *m);
89
	int (*map_check_btf)(const struct bpf_map *map,
90
			     const struct btf *btf,
91 92
			     const struct btf_type *key_type,
			     const struct btf_type *value_type);
93

94 95 96 97 98 99
	/* 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);

100 101 102 103 104
	/* 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);
105
	int (*map_mmap)(struct bpf_map *map, struct vm_area_struct *vma);
106 107
	__poll_t (*map_poll)(struct bpf_map *map, struct file *filp,
			     struct poll_table_struct *pts);
108

K
KP Singh 已提交
109 110 111 112 113 114
	/* Functions called by bpf_local_storage maps */
	int (*map_local_storage_charge)(struct bpf_local_storage_map *smap,
					void *owner, u32 size);
	void (*map_local_storage_uncharge)(struct bpf_local_storage_map *smap,
					   void *owner, u32 size);
	struct bpf_local_storage __rcu ** (*map_owner_storage_ptr)(void *owner);
115 116 117 118 119 120 121 122 123 124 125 126 127

	/* map_meta_equal must be implemented for maps that can be
	 * used as an inner map.  It is a runtime check to ensure
	 * an inner map can be inserted to an outer map.
	 *
	 * Some properties of the inner map has been used during the
	 * verification time.  When inserting an inner map at the runtime,
	 * map_meta_equal has to ensure the inserting map has the same
	 * properties that the verifier has used earlier.
	 */
	bool (*map_meta_equal)(const struct bpf_map *meta0,
			       const struct bpf_map *meta1);

128 129 130
	/* BTF name and id of struct allocated by map_alloc */
	const char * const map_btf_name;
	int *map_btf_id;
131 132 133

	/* bpf_iter info used to open a seq_file */
	const struct bpf_iter_seq_info *iter_seq_info;
134 135
};

136 137 138 139 140
struct bpf_map_memory {
	u32 pages;
	struct user_struct *user;
};

141
struct bpf_map {
142
	/* The first two cachelines with read-mostly members of which some
143 144 145 146 147 148 149
	 * 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
150 151 152 153
	enum bpf_map_type map_type;
	u32 key_size;
	u32 value_size;
	u32 max_entries;
154
	u32 map_flags;
155
	int spin_lock_off; /* >=0 valid offset, <0 error */
M
Martin KaFai Lau 已提交
156
	u32 id;
157
	int numa_node;
158 159
	u32 btf_key_type_id;
	u32 btf_value_type_id;
160
	struct btf *btf;
161
	struct bpf_map_memory memory;
162
	char name[BPF_OBJ_NAME_LEN];
163
	u32 btf_vmlinux_value_type_id;
A
Alexei Starovoitov 已提交
164
	bool bypass_spec_v1;
165 166
	bool frozen; /* write-once; write-protected by freeze_mutex */
	/* 22 bytes hole */
167

168
	/* The 3rd and 4th cacheline with misc members to avoid false sharing
169 170
	 * particularly with refcounting.
	 */
171 172
	atomic64_t refcnt ____cacheline_aligned;
	atomic64_t usercnt;
173
	struct work_struct work;
174 175
	struct mutex freeze_mutex;
	u64 writecnt; /* writable mmap cnt; protected by freeze_mutex */
176 177
};

178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204
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);
	}
}
205 206
void copy_map_value_locked(struct bpf_map *map, void *dst, void *src,
			   bool lock_src);
207
int bpf_obj_name_cpy(char *dst, const char *src, unsigned int size);
208

209
struct bpf_offload_dev;
210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234
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);
}

235 236 237 238 239
static inline bool bpf_map_offload_neutral(const struct bpf_map *map)
{
	return map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY;
}

240 241
static inline bool bpf_map_support_seq_show(const struct bpf_map *map)
{
242 243
	return (map->btf_value_type_id || map->btf_vmlinux_value_type_id) &&
		map->ops->map_seq_show_elem;
244 245
}

246
int map_check_no_btf(const struct bpf_map *map,
247
		     const struct btf *btf,
248 249 250
		     const struct btf_type *key_type,
		     const struct btf_type *value_type);

251 252 253
bool bpf_map_meta_equal(const struct bpf_map *meta0,
			const struct bpf_map *meta1);

254 255
extern const struct bpf_map_ops bpf_map_offload_ops;

256 257
/* function argument constraints */
enum bpf_arg_type {
258
	ARG_DONTCARE = 0,	/* unused argument in helper function */
259 260 261 262 263 264 265

	/* 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 */
266
	ARG_PTR_TO_UNINIT_MAP_VALUE,	/* pointer to valid memory used to store a map value */
267
	ARG_PTR_TO_MAP_VALUE_OR_NULL,	/* pointer to stack used as map value or NULL */
268 269 270 271

	/* the following constraints used to prototype bpf_memcmp() and other
	 * functions that access data on eBPF program stack
	 */
272
	ARG_PTR_TO_MEM,		/* pointer to valid memory (stack, packet, map value) */
273
	ARG_PTR_TO_MEM_OR_NULL, /* pointer to valid memory or NULL */
274 275 276
	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.
277 278
				 */

279 280
	ARG_CONST_SIZE,		/* number of bytes accessed from memory */
	ARG_CONST_SIZE_OR_ZERO,	/* number of bytes accessed from memory or 0 */
281

282
	ARG_PTR_TO_CTX,		/* pointer to context */
283
	ARG_PTR_TO_CTX_OR_NULL,	/* pointer to context or NULL */
284
	ARG_ANYTHING,		/* any (initialized) argument is ok */
285
	ARG_PTR_TO_SPIN_LOCK,	/* pointer to bpf_spin_lock */
286
	ARG_PTR_TO_SOCK_COMMON,	/* pointer to sock_common */
287 288
	ARG_PTR_TO_INT,		/* pointer to int */
	ARG_PTR_TO_LONG,	/* pointer to long */
289
	ARG_PTR_TO_SOCKET,	/* pointer to bpf_sock (fullsock) */
290
	ARG_PTR_TO_SOCKET_OR_NULL,	/* pointer to bpf_sock (fullsock) or NULL */
291
	ARG_PTR_TO_BTF_ID,	/* pointer to in-kernel struct */
292 293 294
	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 */
295 296 297 298 299 300
};

/* type of values returned from helper functions */
enum bpf_return_type {
	RET_INTEGER,			/* function returns integer */
	RET_VOID,			/* function doesn't return anything */
301
	RET_PTR_TO_MAP_VALUE,		/* returns a pointer to map elem value */
302
	RET_PTR_TO_MAP_VALUE_OR_NULL,	/* returns a pointer to map elem value or NULL */
303
	RET_PTR_TO_SOCKET_OR_NULL,	/* returns a pointer to a socket or NULL */
304
	RET_PTR_TO_TCP_SOCK_OR_NULL,	/* returns a pointer to a tcp_sock or NULL */
305
	RET_PTR_TO_SOCK_COMMON_OR_NULL,	/* returns a pointer to a sock_common or NULL */
306
	RET_PTR_TO_ALLOC_MEM_OR_NULL,	/* returns a pointer to dynamically allocated memory or NULL */
307
	RET_PTR_TO_BTF_ID_OR_NULL,	/* returns a pointer to a btf_id or NULL */
308 309
};

310 311 312 313 314 315 316
/* 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;
317
	bool pkt_access;
318
	enum bpf_return_type ret_type;
319 320 321 322 323 324 325 326 327 328
	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];
	};
329 330 331 332 333 334 335 336 337 338
	union {
		struct {
			u32 *arg1_btf_id;
			u32 *arg2_btf_id;
			u32 *arg3_btf_id;
			u32 *arg4_btf_id;
			u32 *arg5_btf_id;
		};
		u32 *arg_btf_id[5];
	};
339
	int *ret_btf_id; /* return value btf_id */
J
Jiri Olsa 已提交
340
	bool (*allowed)(const struct bpf_prog *prog);
341 342 343 344 345 346 347 348 349 350 351
};

/* 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
352 353
};

354
/* types of values stored in eBPF registers */
355 356 357 358 359 360 361 362 363
/* 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
 */
364 365
enum bpf_reg_type {
	NOT_INIT = 0,		 /* nothing was written into register */
366
	SCALAR_VALUE,		 /* reg doesn't contain a valid pointer */
367 368 369 370
	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 */
371
	PTR_TO_STACK,		 /* reg == frame_pointer + offset */
372
	PTR_TO_PACKET_META,	 /* skb->data - meta_len */
373
	PTR_TO_PACKET,		 /* reg points to skb->data */
374
	PTR_TO_PACKET_END,	 /* skb->data + headlen */
375
	PTR_TO_FLOW_KEYS,	 /* reg points to bpf_flow_keys */
376 377
	PTR_TO_SOCKET,		 /* reg points to struct bpf_sock */
	PTR_TO_SOCKET_OR_NULL,	 /* reg points to struct bpf_sock or NULL */
378 379
	PTR_TO_SOCK_COMMON,	 /* reg points to sock_common */
	PTR_TO_SOCK_COMMON_OR_NULL, /* reg points to sock_common or NULL */
380 381
	PTR_TO_TCP_SOCK,	 /* reg points to struct tcp_sock */
	PTR_TO_TCP_SOCK_OR_NULL, /* reg points to struct tcp_sock or NULL */
382
	PTR_TO_TP_BUFFER,	 /* reg points to a writable raw tp's buffer */
383
	PTR_TO_XDP_SOCK,	 /* reg points to struct xdp_sock */
384
	PTR_TO_BTF_ID,		 /* reg points to kernel struct */
385
	PTR_TO_BTF_ID_OR_NULL,	 /* reg points to kernel struct or NULL */
386 387
	PTR_TO_MEM,		 /* reg points to valid memory region */
	PTR_TO_MEM_OR_NULL,	 /* reg points to valid memory region or NULL */
388 389 390 391
	PTR_TO_RDONLY_BUF,	 /* reg points to a readonly buffer */
	PTR_TO_RDONLY_BUF_OR_NULL, /* reg points to a readonly buffer or NULL */
	PTR_TO_RDWR_BUF,	 /* reg points to a read/write buffer */
	PTR_TO_RDWR_BUF_OR_NULL, /* reg points to a read/write buffer or NULL */
392 393
};

394 395 396 397 398
/* The information passed from prog-specific *_is_valid_access
 * back to the verifier.
 */
struct bpf_insn_access_aux {
	enum bpf_reg_type reg_type;
399 400 401 402 403
	union {
		int ctx_field_size;
		u32 btf_id;
	};
	struct bpf_verifier_log *log; /* for verbose logs */
404 405
};

406 407 408 409 410 411
static inline void
bpf_ctx_record_field_size(struct bpf_insn_access_aux *aux, u32 size)
{
	aux->ctx_field_size = size;
}

412 413 414 415 416
struct bpf_prog_ops {
	int (*test_run)(struct bpf_prog *prog, const union bpf_attr *kattr,
			union bpf_attr __user *uattr);
};

417 418
struct bpf_verifier_ops {
	/* return eBPF function prototype for verification */
419 420 421
	const struct bpf_func_proto *
	(*get_func_proto)(enum bpf_func_id func_id,
			  const struct bpf_prog *prog);
422 423 424 425

	/* return true if 'size' wide access at offset 'off' within bpf_context
	 * with 'type' (read or write) is allowed
	 */
426
	bool (*is_valid_access)(int off, int size, enum bpf_access_type type,
427
				const struct bpf_prog *prog,
428
				struct bpf_insn_access_aux *info);
429 430
	int (*gen_prologue)(struct bpf_insn *insn, bool direct_write,
			    const struct bpf_prog *prog);
431 432
	int (*gen_ld_abs)(const struct bpf_insn *orig,
			  struct bpf_insn *insn_buf);
433 434 435
	u32 (*convert_ctx_access)(enum bpf_access_type type,
				  const struct bpf_insn *src,
				  struct bpf_insn *dst,
436
				  struct bpf_prog *prog, u32 *target_size);
437 438 439 440
	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);
441 442
};

443
struct bpf_prog_offload_ops {
444
	/* verifier basic callbacks */
445 446
	int (*insn_hook)(struct bpf_verifier_env *env,
			 int insn_idx, int prev_insn_idx);
447
	int (*finalize)(struct bpf_verifier_env *env);
448 449 450 451 452
	/* 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 */
453 454
	int (*prepare)(struct bpf_prog *prog);
	int (*translate)(struct bpf_prog *prog);
455
	void (*destroy)(struct bpf_prog *prog);
456 457
};

458
struct bpf_prog_offload {
459 460
	struct bpf_prog		*prog;
	struct net_device	*netdev;
461
	struct bpf_offload_dev	*offdev;
462 463 464
	void			*dev_priv;
	struct list_head	offloads;
	bool			dev_state;
465
	bool			opt_failed;
466 467
	void			*jited_image;
	u32			jited_len;
468 469
};

470 471
enum bpf_cgroup_storage_type {
	BPF_CGROUP_STORAGE_SHARED,
472
	BPF_CGROUP_STORAGE_PERCPU,
473 474 475 476 477
	__BPF_CGROUP_STORAGE_MAX
};

#define MAX_BPF_CGROUP_STORAGE_TYPE __BPF_CGROUP_STORAGE_MAX

478 479 480 481 482
/* The longest tracepoint has 12 args.
 * See include/trace/bpf_probe.h
 */
#define MAX_BPF_FUNC_ARGS 12

A
Alexei Starovoitov 已提交
483 484 485 486
struct bpf_prog_stats {
	u64 cnt;
	u64 nsecs;
	struct u64_stats_sync syncp;
E
Eric Dumazet 已提交
487
} __aligned(2 * sizeof(u64));
A
Alexei Starovoitov 已提交
488

A
Alexei Starovoitov 已提交
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508
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
509 510 511 512 513 514 515 516 517 518
/* 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
519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538
/* 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
 */
539 540
int arch_prepare_bpf_trampoline(void *image, void *image_end,
				const struct btf_func_model *m, u32 flags,
K
KP Singh 已提交
541
				struct bpf_tramp_progs *tprogs,
A
Alexei Starovoitov 已提交
542 543 544 545
				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);
546 547
void notrace __bpf_prog_enter_sleepable(void);
void notrace __bpf_prog_exit_sleepable(void);
A
Alexei Starovoitov 已提交
548

J
Jiri Olsa 已提交
549 550 551
struct bpf_ksym {
	unsigned long		 start;
	unsigned long		 end;
J
Jiri Olsa 已提交
552
	char			 name[KSYM_NAME_LEN];
553
	struct list_head	 lnode;
J
Jiri Olsa 已提交
554
	struct latch_tree_node	 tnode;
555
	bool			 prog;
J
Jiri Olsa 已提交
556 557
};

A
Alexei Starovoitov 已提交
558 559 560
enum bpf_tramp_prog_type {
	BPF_TRAMP_FENTRY,
	BPF_TRAMP_FEXIT,
K
KP Singh 已提交
561
	BPF_TRAMP_MODIFY_RETURN,
562 563
	BPF_TRAMP_MAX,
	BPF_TRAMP_REPLACE, /* more than MAX */
A
Alexei Starovoitov 已提交
564 565 566 567 568 569 570 571 572 573 574 575
};

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;
576
		bool ftrace_managed;
A
Alexei Starovoitov 已提交
577
	} func;
578 579 580 581 582
	/* 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
583 584 585 586 587 588 589
	/* 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
590
	struct bpf_ksym ksym;
A
Alexei Starovoitov 已提交
591
};
B
Björn Töpel 已提交
592

593
#define BPF_DISPATCHER_MAX 48 /* Fits in 2048B */
B
Björn Töpel 已提交
594 595 596 597 598 599 600 601 602 603 604 605 606 607

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
608
	struct bpf_ksym ksym;
B
Björn Töpel 已提交
609 610
};

611
static __always_inline unsigned int bpf_dispatcher_nop_func(
612 613 614 615 616 617 618
	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
619 620 621 622 623
#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
624 625 626 627 628 629 630 631 632 633 634
#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
635 636 637
}

#define DEFINE_BPF_DISPATCHER(name)					\
638
	noinline unsigned int bpf_dispatcher_##name##_func(		\
B
Björn Töpel 已提交
639 640 641 642 643 644 645
		const void *ctx,					\
		const struct bpf_insn *insnsi,				\
		unsigned int (*bpf_func)(const void *,			\
					 const struct bpf_insn *))	\
	{								\
		return bpf_func(ctx, insnsi);				\
	}								\
646 647 648
	EXPORT_SYMBOL(bpf_dispatcher_##name##_func);			\
	struct bpf_dispatcher bpf_dispatcher_##name =			\
		BPF_DISPATCHER_INIT(bpf_dispatcher_##name);
B
Björn Töpel 已提交
649
#define DECLARE_BPF_DISPATCHER(name)					\
650
	unsigned int bpf_dispatcher_##name##_func(			\
B
Björn Töpel 已提交
651 652 653 654
		const void *ctx,					\
		const struct bpf_insn *insnsi,				\
		unsigned int (*bpf_func)(const void *,			\
					 const struct bpf_insn *));	\
655 656 657
	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
658 659
void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
				struct bpf_prog *to);
J
Jiri Olsa 已提交
660
/* Called only from JIT-enabled code, so there's no need for stubs. */
J
Jiri Olsa 已提交
661
void *bpf_jit_alloc_exec_page(void);
J
Jiri Olsa 已提交
662 663
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
664 665
void bpf_ksym_add(struct bpf_ksym *ksym);
void bpf_ksym_del(struct bpf_ksym *ksym);
A
Alexei Starovoitov 已提交
666 667 668 669 670 671 672 673 674 675 676 677 678 679
#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
680 681
#define DEFINE_BPF_DISPATCHER(name)
#define DECLARE_BPF_DISPATCHER(name)
682
#define BPF_DISPATCHER_FUNC(name) bpf_dispatcher_nop_func
B
Björn Töpel 已提交
683 684 685 686
#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) {}
687 688 689 690
static inline bool is_bpf_image_address(unsigned long address)
{
	return false;
}
A
Alexei Starovoitov 已提交
691 692
#endif

693
struct bpf_func_info_aux {
694
	u16 linkage;
695 696 697
	bool unreliable;
};

698 699 700 701 702 703
enum bpf_jit_poke_reason {
	BPF_POKE_REASON_TAIL_CALL,
};

/* Descriptor of pokes pointing /into/ the JITed image. */
struct bpf_jit_poke_descriptor {
704
	void *tailcall_target;
705 706
	void *tailcall_bypass;
	void *bypass_addr;
707 708 709 710 711 712
	union {
		struct {
			struct bpf_map *map;
			u32 key;
		} tail_call;
	};
713
	bool tailcall_target_stable;
714 715
	u8 adj_off;
	u16 reason;
716
	u32 insn_idx;
717 718
};

719 720 721 722
/* reg_type info for ctx arguments */
struct bpf_ctx_arg_aux {
	u32 offset;
	enum bpf_reg_type reg_type;
723
	u32 btf_id;
724 725
};

726
struct bpf_prog_aux {
727
	atomic64_t refcnt;
728
	u32 used_map_cnt;
729
	u32 max_ctx_offset;
730
	u32 max_pkt_offset;
731
	u32 max_tp_access;
732
	u32 stack_depth;
M
Martin KaFai Lau 已提交
733
	u32 id;
734 735
	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 */
736
	u32 attach_btf_id; /* in-kernel BTF type id to attach to */
737
	u32 ctx_arg_info_size;
738 739
	u32 max_rdonly_access;
	u32 max_rdwr_access;
740
	const struct bpf_ctx_arg_aux *ctx_arg_info;
741
	struct bpf_prog *linked_prog;
742
	bool verifier_zext; /* Zero extensions has been inserted by verifier. */
743
	bool offload_requested;
744
	bool attach_btf_trace; /* true if attaching to BTF-enabled raw tp */
745
	bool func_proto_unreliable;
746
	bool sleepable;
747
	bool tail_call_reachable;
A
Alexei Starovoitov 已提交
748 749 750
	enum bpf_tramp_prog_type trampoline_prog_type;
	struct bpf_trampoline *trampoline;
	struct hlist_node tramp_hlist;
751 752 753 754
	/* 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;
755 756
	struct bpf_prog **func;
	void *jit_data; /* JIT specific data. arch dependent */
757 758
	struct bpf_jit_poke_descriptor *poke_tab;
	u32 size_poke_tab;
J
Jiri Olsa 已提交
759
	struct bpf_ksym ksym;
760
	const struct bpf_prog_ops *ops;
761
	struct bpf_map **used_maps;
762
	struct mutex used_maps_mutex; /* mutex for used_maps and used_map_cnt */
763
	struct bpf_prog *prog;
764
	struct user_struct *user;
765
	u64 load_time; /* ns since boottime */
766
	struct bpf_map *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
767
	char name[BPF_OBJ_NAME_LEN];
768 769 770
#ifdef CONFIG_SECURITY
	void *security;
#endif
771
	struct bpf_prog_offload *offload;
Y
Yonghong Song 已提交
772
	struct btf *btf;
773
	struct bpf_func_info *func_info;
774
	struct bpf_func_info_aux *func_info_aux;
M
Martin KaFai Lau 已提交
775 776 777 778 779 780 781 782 783 784 785 786 787 788 789
	/* 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;
790
	u32 func_info_cnt;
M
Martin KaFai Lau 已提交
791 792 793 794 795 796
	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;
797 798
	u32 num_exentries;
	struct exception_table_entry *extable;
A
Alexei Starovoitov 已提交
799
	struct bpf_prog_stats __percpu *stats;
800 801 802 803
	union {
		struct work_struct work;
		struct rcu_head	rcu;
	};
804 805
};

806 807 808 809 810 811 812 813
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;
814 815 816 817 818
	/* 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;
819 820
};

821 822 823 824 825 826 827 828 829 830 831 832
struct bpf_link {
	atomic64_t refcnt;
	u32 id;
	enum bpf_link_type type;
	const struct bpf_link_ops *ops;
	struct bpf_prog *prog;
	struct work_struct work;
};

struct bpf_link_ops {
	void (*release)(struct bpf_link *link);
	void (*dealloc)(struct bpf_link *link);
833
	int (*detach)(struct bpf_link *link);
834 835 836 837 838 839 840 841 842 843 844 845 846 847
	int (*update_prog)(struct bpf_link *link, struct bpf_prog *new_prog,
			   struct bpf_prog *old_prog);
	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);
};

struct bpf_link_primer {
	struct bpf_link *link;
	struct file *file;
	int fd;
	u32 id;
};

848
struct bpf_struct_ops_value;
849 850 851 852 853 854 855 856 857
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);
858 859 860 861 862
	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);
863
	const struct btf_type *type;
864
	const struct btf_type *value_type;
865 866 867
	const char *name;
	struct btf_func_model func_models[BPF_STRUCT_OPS_MAX_NR_MEMBERS];
	u32 type_id;
868
	u32 value_id;
869 870 871
};

#if defined(CONFIG_BPF_JIT) && defined(CONFIG_BPF_SYSCALL)
872
#define BPF_MODULE_OWNER ((void *)((0xeB9FUL << 2) + POISON_POINTER_DELTA))
873
const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id);
874
void bpf_struct_ops_init(struct btf *btf, struct bpf_verifier_log *log);
875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892
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);
}
893 894 895 896 897
#else
static inline const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id)
{
	return NULL;
}
898 899 900 901
static inline void bpf_struct_ops_init(struct btf *btf,
				       struct bpf_verifier_log *log)
{
}
902 903 904 905 906 907 908 909 910 911 912 913 914 915
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;
}
916 917
#endif

918 919 920
struct bpf_array {
	struct bpf_map map;
	u32 elem_size;
921
	u32 index_mask;
922
	struct bpf_array_aux *aux;
923 924
	union {
		char value[0] __aligned(8);
925
		void *ptrs[0] __aligned(8);
926
		void __percpu *pptrs[0] __aligned(8);
927 928
	};
};
929

930
#define BPF_COMPLEXITY_LIMIT_INSNS      1000000 /* yes. 1M insns */
931 932
#define MAX_TAIL_CALL_CNT 32

933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961
#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);
}

962 963 964 965 966 967 968
struct bpf_event_entry {
	struct perf_event *event;
	struct file *perf_file;
	struct file *map_file;
	struct rcu_head rcu;
};

969
bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp);
970
int bpf_prog_calc_tag(struct bpf_prog *fp);
971
const char *kernel_type_name(u32 btf_type_id);
972

973
const struct bpf_func_proto *bpf_get_trace_printk_proto(void);
974 975

typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src,
976
					unsigned long off, unsigned long len);
977 978 979 980 981
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);
982 983 984

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);
985

986 987 988 989 990 991 992 993 994 995 996 997
/* 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.
 */
998 999
struct bpf_prog_array_item {
	struct bpf_prog *prog;
1000
	struct bpf_cgroup_storage *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
1001 1002
};

1003 1004
struct bpf_prog_array {
	struct rcu_head rcu;
1005
	struct bpf_prog_array_item items[];
1006 1007
};

1008
struct bpf_prog_array *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags);
1009 1010
void bpf_prog_array_free(struct bpf_prog_array *progs);
int bpf_prog_array_length(struct bpf_prog_array *progs);
1011
bool bpf_prog_array_is_empty(struct bpf_prog_array *array);
1012
int bpf_prog_array_copy_to_user(struct bpf_prog_array *progs,
1013
				__u32 __user *prog_ids, u32 cnt);
1014

1015
void bpf_prog_array_delete_safe(struct bpf_prog_array *progs,
1016
				struct bpf_prog *old_prog);
1017 1018 1019
int bpf_prog_array_delete_safe_at(struct bpf_prog_array *array, int index);
int bpf_prog_array_update_at(struct bpf_prog_array *array, int index,
			     struct bpf_prog *prog);
1020
int bpf_prog_array_copy_info(struct bpf_prog_array *array,
1021 1022
			     u32 *prog_ids, u32 request_cnt,
			     u32 *prog_cnt);
1023
int bpf_prog_array_copy(struct bpf_prog_array *old_array,
1024 1025 1026 1027 1028
			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)	\
1029
	({						\
1030 1031
		struct bpf_prog_array_item *_item;	\
		struct bpf_prog *_prog;			\
1032
		struct bpf_prog_array *_array;		\
1033
		u32 _ret = 1;				\
1034
		migrate_disable();			\
1035
		rcu_read_lock();			\
1036 1037 1038
		_array = rcu_dereference(array);	\
		if (unlikely(check_non_null && !_array))\
			goto _out;			\
1039 1040 1041 1042 1043
		_item = &_array->items[0];		\
		while ((_prog = READ_ONCE(_item->prog))) {		\
			bpf_cgroup_storage_set(_item->cgroup_storage);	\
			_ret &= func(_prog, ctx);	\
			_item++;			\
1044 1045
		}					\
_out:							\
1046
		rcu_read_unlock();			\
1047
		migrate_enable();			\
1048 1049 1050
		_ret;					\
	 })

1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080
/* 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;				\
1081
		migrate_disable();			\
1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092
		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();			\
1093
		migrate_enable();			\
1094 1095 1096 1097 1098 1099 1100
		if (_ret)				\
			_ret = (_cn ? NET_XMIT_CN : NET_XMIT_SUCCESS);	\
		else					\
			_ret = (_cn ? NET_XMIT_DROP : -EPERM);		\
		_ret;					\
	})

1101 1102 1103 1104 1105 1106
#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)

1107
#ifdef CONFIG_BPF_SYSCALL
1108
DECLARE_PER_CPU(int, bpf_prog_active);
1109
extern struct mutex bpf_stats_enabled_mutex;
1110

1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140
/*
 * 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();
}

1141 1142
extern const struct file_operations bpf_map_fops;
extern const struct file_operations bpf_prog_fops;
Y
Yonghong Song 已提交
1143
extern const struct file_operations bpf_iter_fops;
1144

A
Alexei Starovoitov 已提交
1145
#define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \
1146 1147
	extern const struct bpf_prog_ops _name ## _prog_ops; \
	extern const struct bpf_verifier_ops _name ## _verifier_ops;
1148 1149
#define BPF_MAP_TYPE(_id, _ops) \
	extern const struct bpf_map_ops _ops;
1150
#define BPF_LINK_TYPE(_id, _name)
1151 1152
#include <linux/bpf_types.h>
#undef BPF_PROG_TYPE
1153
#undef BPF_MAP_TYPE
1154
#undef BPF_LINK_TYPE
1155

1156
extern const struct bpf_prog_ops bpf_offload_prog_ops;
1157 1158 1159
extern const struct bpf_verifier_ops tc_cls_act_analyzer_ops;
extern const struct bpf_verifier_ops xdp_analyzer_ops;

1160
struct bpf_prog *bpf_prog_get(u32 ufd);
1161
struct bpf_prog *bpf_prog_get_type_dev(u32 ufd, enum bpf_prog_type type,
1162
				       bool attach_drv);
1163
void bpf_prog_add(struct bpf_prog *prog, int i);
1164
void bpf_prog_sub(struct bpf_prog *prog, int i);
1165
void bpf_prog_inc(struct bpf_prog *prog);
1166
struct bpf_prog * __must_check bpf_prog_inc_not_zero(struct bpf_prog *prog);
1167
void bpf_prog_put(struct bpf_prog *prog);
1168 1169
int __bpf_prog_charge(struct user_struct *user, u32 pages);
void __bpf_prog_uncharge(struct user_struct *user, u32 pages);
1170 1171
void __bpf_free_used_maps(struct bpf_prog_aux *aux,
			  struct bpf_map **used_maps, u32 len);
1172

1173
void bpf_prog_free_id(struct bpf_prog *prog, bool do_idr_lock);
1174
void bpf_map_free_id(struct bpf_map *map, bool do_idr_lock);
1175

1176
struct bpf_map *bpf_map_get(u32 ufd);
1177
struct bpf_map *bpf_map_get_with_uref(u32 ufd);
1178
struct bpf_map *__bpf_map_get(struct fd f);
1179 1180 1181
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);
1182
void bpf_map_put_with_uref(struct bpf_map *map);
1183
void bpf_map_put(struct bpf_map *map);
1184 1185
int bpf_map_charge_memlock(struct bpf_map *map, u32 pages);
void bpf_map_uncharge_memlock(struct bpf_map *map, u32 pages);
1186
int bpf_map_charge_init(struct bpf_map_memory *mem, u64 size);
1187 1188 1189
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);
1190 1191
void *bpf_map_area_alloc(u64 size, int numa_node);
void *bpf_map_area_mmapable_alloc(u64 size, int numa_node);
1192
void bpf_map_area_free(void *base);
1193
void bpf_map_init_from_attr(struct bpf_map *map, union bpf_attr *attr);
1194 1195
int  generic_map_lookup_batch(struct bpf_map *map,
			      const union bpf_attr *attr,
1196 1197 1198 1199 1200 1201
			      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,
1202
			      union bpf_attr __user *uattr);
Y
Yonghong Song 已提交
1203
struct bpf_map *bpf_map_get_curr_or_next(u32 *id);
A
Alexei Starovoitov 已提交
1204
struct bpf_prog *bpf_prog_get_curr_or_next(u32 *id);
1205

1206 1207
extern int sysctl_unprivileged_bpf_disabled;

A
Alexei Starovoitov 已提交
1208 1209 1210 1211 1212
static inline bool bpf_allow_ptr_leaks(void)
{
	return perfmon_capable();
}

1213 1214 1215 1216 1217
static inline bool bpf_allow_ptr_to_map_access(void)
{
	return perfmon_capable();
}

A
Alexei Starovoitov 已提交
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227
static inline bool bpf_bypass_spec_v1(void)
{
	return perfmon_capable();
}

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

1228
int bpf_map_new_fd(struct bpf_map *map, int flags);
1229 1230
int bpf_prog_new_fd(struct bpf_prog *prog);

1231
void bpf_link_init(struct bpf_link *link, enum bpf_link_type type,
A
Andrii Nakryiko 已提交
1232 1233 1234 1235
		   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);
1236 1237 1238
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);
1239
struct file *bpf_link_new_file(struct bpf_link *link, int *reserved_fd);
1240 1241
struct bpf_link *bpf_link_get_from_fd(u32 ufd);

1242
int bpf_obj_pin_user(u32 ufd, const char __user *pathname);
1243
int bpf_obj_get_user(const char __user *pathname, int flags);
1244

1245
#define BPF_ITER_FUNC_PREFIX "bpf_iter_"
1246
#define DEFINE_BPF_ITER_FUNC(target, args...)			\
1247 1248
	extern int bpf_iter_ ## target(args);			\
	int __init bpf_iter_ ## target(args) { return 0; }
1249

1250
struct bpf_iter_aux_info {
1251
	struct bpf_map *map;
1252 1253
};

1254 1255 1256 1257
typedef int (*bpf_iter_attach_target_t)(struct bpf_prog *prog,
					union bpf_iter_link_info *linfo,
					struct bpf_iter_aux_info *aux);
typedef void (*bpf_iter_detach_target_t)(struct bpf_iter_aux_info *aux);
1258 1259 1260 1261
typedef void (*bpf_iter_show_fdinfo_t) (const struct bpf_iter_aux_info *aux,
					struct seq_file *seq);
typedef int (*bpf_iter_fill_link_info_t)(const struct bpf_iter_aux_info *aux,
					 struct bpf_link_info *info);
1262

1263
#define BPF_ITER_CTX_ARG_MAX 2
1264 1265
struct bpf_iter_reg {
	const char *target;
1266 1267
	bpf_iter_attach_target_t attach_target;
	bpf_iter_detach_target_t detach_target;
1268 1269
	bpf_iter_show_fdinfo_t show_fdinfo;
	bpf_iter_fill_link_info_t fill_link_info;
1270 1271
	u32 ctx_arg_info_size;
	struct bpf_ctx_arg_aux ctx_arg_info[BPF_ITER_CTX_ARG_MAX];
1272
	const struct bpf_iter_seq_info *seq_info;
1273 1274
};

1275 1276 1277 1278 1279 1280
struct bpf_iter_meta {
	__bpf_md_ptr(struct seq_file *, seq);
	u64 session_id;
	u64 seq_num;
};

1281 1282 1283 1284 1285 1286 1287
struct bpf_iter__bpf_map_elem {
	__bpf_md_ptr(struct bpf_iter_meta *, meta);
	__bpf_md_ptr(struct bpf_map *, map);
	__bpf_md_ptr(void *, key);
	__bpf_md_ptr(void *, value);
};

1288
int bpf_iter_reg_target(const struct bpf_iter_reg *reg_info);
1289
void bpf_iter_unreg_target(const struct bpf_iter_reg *reg_info);
1290
bool bpf_iter_prog_supported(struct bpf_prog *prog);
1291
int bpf_iter_link_attach(const union bpf_attr *attr, struct bpf_prog *prog);
1292
int bpf_iter_new_fd(struct bpf_link *link);
Y
Yonghong Song 已提交
1293
bool bpf_link_is_iter(struct bpf_link *link);
1294 1295
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);
1296 1297 1298 1299
void bpf_iter_map_show_fdinfo(const struct bpf_iter_aux_info *aux,
			      struct seq_file *seq);
int bpf_iter_map_fill_link_info(const struct bpf_iter_aux_info *aux,
				struct bpf_link_info *info);
1300

1301 1302 1303 1304 1305 1306
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);
1307

1308
int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value);
1309

1310 1311
int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file,
				 void *key, void *value, u64 map_flags);
1312
int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
M
Martin KaFai Lau 已提交
1313 1314
int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file,
				void *key, void *value, u64 map_flags);
1315
int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
1316

1317
int bpf_get_file_flag(int flags);
1318 1319
int bpf_check_uarg_tail_zero(void __user *uaddr, size_t expected_size,
			     size_t actual_size);
1320

1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336
/* 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++;
}

1337
/* verify correctness of eBPF program */
Y
Yonghong Song 已提交
1338 1339
int bpf_check(struct bpf_prog **fp, union bpf_attr *attr,
	      union bpf_attr __user *uattr);
1340
void bpf_patch_call_args(struct bpf_insn *insn, u32 stack_depth);
1341 1342

/* Map specifics */
1343
struct xdp_buff;
1344
struct sk_buff;
1345 1346

struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
1347
struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
1348 1349 1350
void __dev_flush(void);
int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
1351 1352
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
1353 1354
int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
			     struct bpf_prog *xdp_prog);
1355
bool dev_map_can_have_prog(struct bpf_map *map);
1356

1357
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
1358
void __cpu_map_flush(void);
1359 1360
int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
1361
bool cpu_map_prog_allowed(struct bpf_map *map);
1362

1363 1364 1365 1366 1367 1368 1369
/* 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;
}

1370
struct bpf_prog *bpf_prog_get_type_path(const char *name, enum bpf_prog_type type);
1371
int array_map_alloc_check(union bpf_attr *attr);
1372

1373 1374 1375 1376
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);
1377 1378 1379
int bpf_prog_test_run_tracing(struct bpf_prog *prog,
			      const union bpf_attr *kattr,
			      union bpf_attr __user *uattr);
1380 1381 1382
int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog,
				     const union bpf_attr *kattr,
				     union bpf_attr __user *uattr);
1383 1384 1385 1386 1387 1388 1389
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);
1390 1391
bool btf_struct_ids_match(struct bpf_verifier_log *log,
			  int off, u32 id, u32 need_type_id);
1392

A
Alexei Starovoitov 已提交
1393 1394 1395 1396 1397 1398
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);

1399 1400 1401 1402 1403
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);
1404 1405
int btf_check_type_match(struct bpf_verifier_env *env, struct bpf_prog *prog,
			 struct btf *btf, const struct btf_type *t);
1406

1407
struct bpf_prog *bpf_prog_by_id(u32 id);
1408
struct bpf_link *bpf_link_by_id(u32 id);
1409

1410
const struct bpf_func_proto *bpf_base_func_proto(enum bpf_func_id func_id);
1411
#else /* !CONFIG_BPF_SYSCALL */
1412 1413 1414 1415 1416
static inline struct bpf_prog *bpf_prog_get(u32 ufd)
{
	return ERR_PTR(-EOPNOTSUPP);
}

1417 1418
static inline struct bpf_prog *bpf_prog_get_type_dev(u32 ufd,
						     enum bpf_prog_type type,
1419
						     bool attach_drv)
1420 1421 1422 1423
{
	return ERR_PTR(-EOPNOTSUPP);
}

1424
static inline void bpf_prog_add(struct bpf_prog *prog, int i)
1425 1426
{
}
1427

1428 1429 1430 1431
static inline void bpf_prog_sub(struct bpf_prog *prog, int i)
{
}

1432 1433 1434
static inline void bpf_prog_put(struct bpf_prog *prog)
{
}
1435

1436
static inline void bpf_prog_inc(struct bpf_prog *prog)
1437 1438
{
}
1439

1440 1441 1442 1443 1444 1445
static inline struct bpf_prog *__must_check
bpf_prog_inc_not_zero(struct bpf_prog *prog)
{
	return ERR_PTR(-EOPNOTSUPP);
}

1446 1447 1448 1449 1450 1451 1452 1453
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)
{
}
1454

1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483
static inline void bpf_link_init(struct bpf_link *link, enum bpf_link_type type,
				 const struct bpf_link_ops *ops,
				 struct bpf_prog *prog)
{
}

static inline int bpf_link_prime(struct bpf_link *link,
				 struct bpf_link_primer *primer)
{
	return -EOPNOTSUPP;
}

static inline int bpf_link_settle(struct bpf_link_primer *primer)
{
	return -EOPNOTSUPP;
}

static inline void bpf_link_cleanup(struct bpf_link_primer *primer)
{
}

static inline void bpf_link_inc(struct bpf_link *link)
{
}

static inline void bpf_link_put(struct bpf_link *link)
{
}

1484
static inline int bpf_obj_get_user(const char __user *pathname, int flags)
1485 1486 1487 1488
{
	return -EOPNOTSUPP;
}

1489 1490 1491 1492 1493 1494
static inline struct net_device  *__dev_map_lookup_elem(struct bpf_map *map,
						       u32 key)
{
	return NULL;
}

1495 1496 1497 1498 1499
static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map,
							     u32 key)
{
	return NULL;
}
1500 1501 1502 1503
static inline bool dev_map_can_have_prog(struct bpf_map *map)
{
	return false;
}
1504

1505
static inline void __dev_flush(void)
1506 1507
{
}
1508

1509 1510 1511
struct xdp_buff;
struct bpf_dtab_netdev;

1512 1513 1514 1515 1516 1517 1518
static inline
int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
		    struct net_device *dev_rx)
{
	return 0;
}

1519
static inline
1520 1521
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx)
1522 1523 1524 1525
{
	return 0;
}

1526 1527 1528 1529 1530 1531 1532 1533 1534
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;
}

1535 1536 1537 1538 1539 1540
static inline
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key)
{
	return NULL;
}

1541
static inline void __cpu_map_flush(void)
1542 1543 1544 1545 1546 1547 1548 1549 1550
{
}

static inline int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu,
				  struct xdp_buff *xdp,
				  struct net_device *dev_rx)
{
	return 0;
}
1551

1552 1553 1554 1555 1556
static inline bool cpu_map_prog_allowed(struct bpf_map *map)
{
	return false;
}

1557 1558 1559 1560 1561
static inline struct bpf_prog *bpf_prog_get_type_path(const char *name,
				enum bpf_prog_type type)
{
	return ERR_PTR(-EOPNOTSUPP);
}
1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576

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;
}

1577 1578 1579 1580 1581 1582 1583
static inline int bpf_prog_test_run_tracing(struct bpf_prog *prog,
					    const union bpf_attr *kattr,
					    union bpf_attr __user *uattr)
{
	return -ENOTSUPP;
}

1584 1585 1586 1587 1588 1589
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;
}
1590 1591 1592 1593

static inline void bpf_map_put(struct bpf_map *map)
{
}
1594 1595 1596 1597 1598

static inline struct bpf_prog *bpf_prog_by_id(u32 id)
{
	return ERR_PTR(-ENOTSUPP);
}
1599 1600 1601 1602 1603 1604

static inline const struct bpf_func_proto *
bpf_base_func_proto(enum bpf_func_id func_id)
{
	return NULL;
}
1605
#endif /* CONFIG_BPF_SYSCALL */
1606

1607 1608 1609 1610 1611 1612
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);
}

1613 1614
bool bpf_prog_get_ok(struct bpf_prog *, enum bpf_prog_type *, bool);

1615 1616
int bpf_prog_offload_compile(struct bpf_prog *prog);
void bpf_prog_offload_destroy(struct bpf_prog *prog);
1617 1618
int bpf_prog_offload_info_fill(struct bpf_prog_info *info,
			       struct bpf_prog *prog);
1619

1620 1621
int bpf_map_offload_info_fill(struct bpf_map_info *info, struct bpf_map *map);

1622 1623 1624 1625 1626 1627 1628
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);

1629
bool bpf_offload_prog_map_match(struct bpf_prog *prog, struct bpf_map *map);
1630

1631
struct bpf_offload_dev *
1632
bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops, void *priv);
1633
void bpf_offload_dev_destroy(struct bpf_offload_dev *offdev);
1634
void *bpf_offload_dev_priv(struct bpf_offload_dev *offdev);
1635 1636 1637 1638
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);
1639
bool bpf_offload_dev_match(struct bpf_prog *prog, struct net_device *netdev);
1640

1641 1642 1643
#if defined(CONFIG_NET) && defined(CONFIG_BPF_SYSCALL)
int bpf_prog_offload_init(struct bpf_prog *prog, union bpf_attr *attr);

1644
static inline bool bpf_prog_is_dev_bound(const struct bpf_prog_aux *aux)
1645
{
1646
	return aux->offload_requested;
1647
}
1648 1649 1650 1651 1652 1653 1654 1655

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);
1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666
#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;
}
1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680

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)
{
}
1681 1682
#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */

1683
#if defined(CONFIG_BPF_STREAM_PARSER)
1684 1685
int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
			 struct bpf_prog *old, u32 which);
1686
int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog *prog);
1687
int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type ptype);
1688
int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value, u64 flags);
1689 1690
void sock_map_unhash(struct sock *sk);
void sock_map_close(struct sock *sk, long timeout);
1691
#else
1692
static inline int sock_map_prog_update(struct bpf_map *map,
1693 1694
				       struct bpf_prog *prog,
				       struct bpf_prog *old, u32 which)
1695 1696 1697
{
	return -EOPNOTSUPP;
}
1698

1699 1700
static inline int sock_map_get_from_fd(const union bpf_attr *attr,
				       struct bpf_prog *prog)
1701 1702 1703
{
	return -EINVAL;
}
1704 1705 1706 1707 1708 1709

static inline int sock_map_prog_detach(const union bpf_attr *attr,
				       enum bpf_prog_type ptype)
{
	return -EOPNOTSUPP;
}
1710 1711 1712 1713 1714 1715

static inline int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value,
					   u64 flags)
{
	return -EOPNOTSUPP;
}
1716
#endif /* CONFIG_BPF_STREAM_PARSER */
1717

1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744
#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) */

1745
/* verifier prototypes for helper functions called from eBPF programs */
1746 1747 1748
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
1749 1750 1751
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;
1752

1753
extern const struct bpf_func_proto bpf_get_prandom_u32_proto;
1754
extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
1755
extern const struct bpf_func_proto bpf_get_numa_node_id_proto;
1756
extern const struct bpf_func_proto bpf_tail_call_proto;
1757
extern const struct bpf_func_proto bpf_ktime_get_ns_proto;
1758
extern const struct bpf_func_proto bpf_ktime_get_boot_ns_proto;
1759 1760 1761
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;
1762
extern const struct bpf_func_proto bpf_get_stackid_proto;
Y
Yonghong Song 已提交
1763
extern const struct bpf_func_proto bpf_get_stack_proto;
1764
extern const struct bpf_func_proto bpf_get_task_stack_proto;
1765 1766
extern const struct bpf_func_proto bpf_get_stackid_proto_pe;
extern const struct bpf_func_proto bpf_get_stack_proto_pe;
1767
extern const struct bpf_func_proto bpf_sock_map_update_proto;
1768
extern const struct bpf_func_proto bpf_sock_hash_update_proto;
1769
extern const struct bpf_func_proto bpf_get_current_cgroup_id_proto;
1770
extern const struct bpf_func_proto bpf_get_current_ancestor_cgroup_id_proto;
1771 1772 1773 1774
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;
1775 1776
extern const struct bpf_func_proto bpf_spin_lock_proto;
extern const struct bpf_func_proto bpf_spin_unlock_proto;
1777
extern const struct bpf_func_proto bpf_get_local_storage_proto;
1778 1779
extern const struct bpf_func_proto bpf_strtol_proto;
extern const struct bpf_func_proto bpf_strtoul_proto;
1780
extern const struct bpf_func_proto bpf_tcp_sock_proto;
M
Martin KaFai Lau 已提交
1781
extern const struct bpf_func_proto bpf_jiffies64_proto;
1782
extern const struct bpf_func_proto bpf_get_ns_current_pid_tgid_proto;
1783
extern const struct bpf_func_proto bpf_event_output_data_proto;
1784 1785 1786 1787 1788
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;
1789
extern const struct bpf_func_proto bpf_skc_to_tcp6_sock_proto;
1790 1791 1792
extern const struct bpf_func_proto bpf_skc_to_tcp_sock_proto;
extern const struct bpf_func_proto bpf_skc_to_tcp_timewait_sock_proto;
extern const struct bpf_func_proto bpf_skc_to_tcp_request_sock_proto;
1793
extern const struct bpf_func_proto bpf_skc_to_udp6_sock_proto;
1794
extern const struct bpf_func_proto bpf_copy_from_user_proto;
1795

K
KP Singh 已提交
1796 1797 1798
const struct bpf_func_proto *bpf_tracing_func_proto(
	enum bpf_func_id func_id, const struct bpf_prog *prog);

1799 1800 1801
const struct bpf_func_proto *tracing_prog_func_proto(
  enum bpf_func_id func_id, const struct bpf_prog *prog);

1802 1803 1804
/* 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);
1805
u64 bpf_get_raw_cpu_id(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
1806

1807
#if defined(CONFIG_NET)
1808 1809 1810
bool bpf_sock_common_is_valid_access(int off, int size,
				     enum bpf_access_type type,
				     struct bpf_insn_access_aux *info);
1811 1812 1813 1814 1815 1816 1817 1818
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
1819 1820 1821 1822 1823 1824
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;
}
1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840
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

1841
#ifdef CONFIG_INET
A
Alexei Starovoitov 已提交
1842 1843 1844 1845 1846 1847 1848 1849 1850
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;
};
1851 1852 1853 1854 1855 1856 1857 1858
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);
1859 1860 1861 1862 1863 1864 1865 1866 1867

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);
1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883
#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;
}
1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898
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;
}
1899 1900
#endif /* CONFIG_INET */

1901
enum bpf_text_poke_type {
1902 1903
	BPF_MOD_CALL,
	BPF_MOD_JUMP,
1904
};
1905

1906 1907 1908
int bpf_arch_text_poke(void *ip, enum bpf_text_poke_type t,
		       void *addr1, void *addr2);

J
Jiri Olsa 已提交
1909
struct btf_id_set;
1910
bool btf_id_set_contains(const struct btf_id_set *set, u32 id);
J
Jiri Olsa 已提交
1911

1912
#endif /* _LINUX_BPF_H */