bpf.h 64.2 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 18
#include <linux/refcount.h>
#include <linux/mutex.h>
19
#include <linux/module.h>
J
Jiri Olsa 已提交
20
#include <linux/kallsyms.h>
A
Alexei Starovoitov 已提交
21
#include <linux/capability.h>
22 23
#include <linux/sched/mm.h>
#include <linux/slab.h>
24

25
struct bpf_verifier_env;
26
struct bpf_verifier_log;
27
struct perf_event;
28
struct bpf_prog;
29
struct bpf_prog_aux;
30
struct bpf_map;
31
struct sock;
32
struct seq_file;
33
struct btf;
34
struct btf_type;
35
struct exception_table_entry;
36
struct seq_operations;
37
struct bpf_iter_aux_info;
K
KP Singh 已提交
38 39
struct bpf_local_storage;
struct bpf_local_storage_map;
40
struct kobject;
41
struct mem_cgroup;
42
struct bpf_func_state;
43

44 45
extern struct idr btf_idr;
extern spinlock_t btf_idr_lock;
46
extern struct kobject *btf_kobj;
47

48 49
typedef int (*bpf_iter_init_seq_priv_t)(void *private_data,
					struct bpf_iter_aux_info *aux);
50 51 52 53 54 55 56 57
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;
};

58 59 60
/* map is generic key/value storage optionally accesible by eBPF programs */
struct bpf_map_ops {
	/* funcs callable from userspace (via syscall) */
61
	int (*map_alloc_check)(union bpf_attr *attr);
62
	struct bpf_map *(*map_alloc)(union bpf_attr *attr);
63 64
	void (*map_release)(struct bpf_map *map, struct file *map_file);
	void (*map_free)(struct bpf_map *map);
65
	int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key);
66
	void (*map_release_uref)(struct bpf_map *map);
67
	void *(*map_lookup_elem_sys_only)(struct bpf_map *map, void *key);
68 69
	int (*map_lookup_batch)(struct bpf_map *map, const union bpf_attr *attr,
				union bpf_attr __user *uattr);
70 71 72
	int (*map_lookup_and_delete_batch)(struct bpf_map *map,
					   const union bpf_attr *attr,
					   union bpf_attr __user *uattr);
73 74 75 76
	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);
77 78 79

	/* funcs callable from userspace and from eBPF programs */
	void *(*map_lookup_elem)(struct bpf_map *map, void *key);
80
	int (*map_update_elem)(struct bpf_map *map, void *key, void *value, u64 flags);
81
	int (*map_delete_elem)(struct bpf_map *map, void *key);
M
Mauricio Vasquez B 已提交
82 83 84
	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);
85 86

	/* funcs called by prog_array and perf_event_array map */
87 88 89
	void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file,
				int fd);
	void (*map_fd_put_ptr)(void *ptr);
90
	int (*map_gen_lookup)(struct bpf_map *map, struct bpf_insn *insn_buf);
91
	u32 (*map_fd_sys_lookup_elem)(void *ptr);
92 93
	void (*map_seq_show_elem)(struct bpf_map *map, void *key,
				  struct seq_file *m);
94
	int (*map_check_btf)(const struct bpf_map *map,
95
			     const struct btf *btf,
96 97
			     const struct btf_type *key_type,
			     const struct btf_type *value_type);
98

99 100 101 102 103 104
	/* 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);

105 106 107 108 109
	/* 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);
110
	int (*map_mmap)(struct bpf_map *map, struct vm_area_struct *vma);
111 112
	__poll_t (*map_poll)(struct bpf_map *map, struct file *filp,
			     struct poll_table_struct *pts);
113

K
KP Singh 已提交
114 115 116 117 118 119
	/* 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);
120 121 122 123 124 125 126 127 128 129 130 131 132

	/* 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);

133 134 135 136 137 138 139

	int (*map_set_for_each_callback_args)(struct bpf_verifier_env *env,
					      struct bpf_func_state *caller,
					      struct bpf_func_state *callee);
	int (*map_for_each_callback)(struct bpf_map *map, void *callback_fn,
				     void *callback_ctx, u64 flags);

140 141 142
	/* BTF name and id of struct allocated by map_alloc */
	const char * const map_btf_name;
	int *map_btf_id;
143 144 145

	/* bpf_iter info used to open a seq_file */
	const struct bpf_iter_seq_info *iter_seq_info;
146 147 148
};

struct bpf_map {
149
	/* The first two cachelines with read-mostly members of which some
150 151 152 153 154 155 156
	 * 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
157 158 159 160
	enum bpf_map_type map_type;
	u32 key_size;
	u32 value_size;
	u32 max_entries;
161
	u32 map_flags;
162
	int spin_lock_off; /* >=0 valid offset, <0 error */
M
Martin KaFai Lau 已提交
163
	u32 id;
164
	int numa_node;
165 166
	u32 btf_key_type_id;
	u32 btf_value_type_id;
167
	struct btf *btf;
168 169 170
#ifdef CONFIG_MEMCG_KMEM
	struct mem_cgroup *memcg;
#endif
171
	char name[BPF_OBJ_NAME_LEN];
172
	u32 btf_vmlinux_value_type_id;
A
Alexei Starovoitov 已提交
173
	bool bypass_spec_v1;
174 175
	bool frozen; /* write-once; write-protected by freeze_mutex */
	/* 22 bytes hole */
176

177
	/* The 3rd and 4th cacheline with misc members to avoid false sharing
178 179
	 * particularly with refcounting.
	 */
180 181
	atomic64_t refcnt ____cacheline_aligned;
	atomic64_t usercnt;
182
	struct work_struct work;
183 184
	struct mutex freeze_mutex;
	u64 writecnt; /* writable mmap cnt; protected by freeze_mutex */
185 186
};

187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213
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);
	}
}
214 215
void copy_map_value_locked(struct bpf_map *map, void *dst, void *src,
			   bool lock_src);
216
int bpf_obj_name_cpy(char *dst, const char *src, unsigned int size);
217

218
struct bpf_offload_dev;
219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243
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);
}

244 245 246 247 248
static inline bool bpf_map_offload_neutral(const struct bpf_map *map)
{
	return map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY;
}

249 250
static inline bool bpf_map_support_seq_show(const struct bpf_map *map)
{
251 252
	return (map->btf_value_type_id || map->btf_vmlinux_value_type_id) &&
		map->ops->map_seq_show_elem;
253 254
}

255
int map_check_no_btf(const struct bpf_map *map,
256
		     const struct btf *btf,
257 258 259
		     const struct btf_type *key_type,
		     const struct btf_type *value_type);

260 261 262
bool bpf_map_meta_equal(const struct bpf_map *meta0,
			const struct bpf_map *meta1);

263 264
extern const struct bpf_map_ops bpf_map_offload_ops;

265 266
/* function argument constraints */
enum bpf_arg_type {
267
	ARG_DONTCARE = 0,	/* unused argument in helper function */
268 269 270 271 272 273 274

	/* 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 */
275
	ARG_PTR_TO_UNINIT_MAP_VALUE,	/* pointer to valid memory used to store a map value */
276
	ARG_PTR_TO_MAP_VALUE_OR_NULL,	/* pointer to stack used as map value or NULL */
277 278 279 280

	/* the following constraints used to prototype bpf_memcmp() and other
	 * functions that access data on eBPF program stack
	 */
281
	ARG_PTR_TO_MEM,		/* pointer to valid memory (stack, packet, map value) */
282
	ARG_PTR_TO_MEM_OR_NULL, /* pointer to valid memory or NULL */
283 284 285
	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.
286 287
				 */

288 289
	ARG_CONST_SIZE,		/* number of bytes accessed from memory */
	ARG_CONST_SIZE_OR_ZERO,	/* number of bytes accessed from memory or 0 */
290

291
	ARG_PTR_TO_CTX,		/* pointer to context */
292
	ARG_PTR_TO_CTX_OR_NULL,	/* pointer to context or NULL */
293
	ARG_ANYTHING,		/* any (initialized) argument is ok */
294
	ARG_PTR_TO_SPIN_LOCK,	/* pointer to bpf_spin_lock */
295
	ARG_PTR_TO_SOCK_COMMON,	/* pointer to sock_common */
296 297
	ARG_PTR_TO_INT,		/* pointer to int */
	ARG_PTR_TO_LONG,	/* pointer to long */
298
	ARG_PTR_TO_SOCKET,	/* pointer to bpf_sock (fullsock) */
299
	ARG_PTR_TO_SOCKET_OR_NULL,	/* pointer to bpf_sock (fullsock) or NULL */
300
	ARG_PTR_TO_BTF_ID,	/* pointer to in-kernel struct */
301 302 303
	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 */
304
	ARG_PTR_TO_BTF_ID_SOCK_COMMON,	/* pointer to in-kernel sock_common or bpf-mirrored bpf_sock */
H
Hao Luo 已提交
305
	ARG_PTR_TO_PERCPU_BTF_ID,	/* pointer to in-kernel percpu type */
306 307
	ARG_PTR_TO_FUNC,	/* pointer to a bpf program function */
	ARG_PTR_TO_STACK_OR_NULL,	/* pointer to stack or NULL */
308
	__BPF_ARG_TYPE_MAX,
309 310 311 312 313 314
};

/* type of values returned from helper functions */
enum bpf_return_type {
	RET_INTEGER,			/* function returns integer */
	RET_VOID,			/* function doesn't return anything */
315
	RET_PTR_TO_MAP_VALUE,		/* returns a pointer to map elem value */
316
	RET_PTR_TO_MAP_VALUE_OR_NULL,	/* returns a pointer to map elem value or NULL */
317
	RET_PTR_TO_SOCKET_OR_NULL,	/* returns a pointer to a socket or NULL */
318
	RET_PTR_TO_TCP_SOCK_OR_NULL,	/* returns a pointer to a tcp_sock or NULL */
319
	RET_PTR_TO_SOCK_COMMON_OR_NULL,	/* returns a pointer to a sock_common or NULL */
320
	RET_PTR_TO_ALLOC_MEM_OR_NULL,	/* returns a pointer to dynamically allocated memory or NULL */
321
	RET_PTR_TO_BTF_ID_OR_NULL,	/* returns a pointer to a btf_id or NULL */
H
Hao Luo 已提交
322
	RET_PTR_TO_MEM_OR_BTF_ID_OR_NULL, /* returns a pointer to a valid memory or a btf_id or NULL */
H
Hao Luo 已提交
323
	RET_PTR_TO_MEM_OR_BTF_ID,	/* returns a pointer to a valid memory or a btf_id */
324
	RET_PTR_TO_BTF_ID,		/* returns a pointer to a btf_id */
325 326
};

327 328 329 330 331 332 333
/* 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;
334
	bool pkt_access;
335
	enum bpf_return_type ret_type;
336 337 338 339 340 341 342 343 344 345
	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];
	};
346 347 348 349 350 351 352 353 354 355
	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];
	};
356
	int *ret_btf_id; /* return value btf_id */
J
Jiri Olsa 已提交
357
	bool (*allowed)(const struct bpf_prog *prog);
358 359 360 361 362 363 364 365 366 367 368
};

/* 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
369 370
};

371
/* types of values stored in eBPF registers */
372 373 374 375 376 377 378 379 380
/* 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
 */
381 382
enum bpf_reg_type {
	NOT_INIT = 0,		 /* nothing was written into register */
383
	SCALAR_VALUE,		 /* reg doesn't contain a valid pointer */
384 385 386 387
	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 */
388
	PTR_TO_STACK,		 /* reg == frame_pointer + offset */
389
	PTR_TO_PACKET_META,	 /* skb->data - meta_len */
390
	PTR_TO_PACKET,		 /* reg points to skb->data */
391
	PTR_TO_PACKET_END,	 /* skb->data + headlen */
392
	PTR_TO_FLOW_KEYS,	 /* reg points to bpf_flow_keys */
393 394
	PTR_TO_SOCKET,		 /* reg points to struct bpf_sock */
	PTR_TO_SOCKET_OR_NULL,	 /* reg points to struct bpf_sock or NULL */
395 396
	PTR_TO_SOCK_COMMON,	 /* reg points to sock_common */
	PTR_TO_SOCK_COMMON_OR_NULL, /* reg points to sock_common or NULL */
397 398
	PTR_TO_TCP_SOCK,	 /* reg points to struct tcp_sock */
	PTR_TO_TCP_SOCK_OR_NULL, /* reg points to struct tcp_sock or NULL */
399
	PTR_TO_TP_BUFFER,	 /* reg points to a writable raw tp's buffer */
400
	PTR_TO_XDP_SOCK,	 /* reg points to struct xdp_sock */
401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416
	/* PTR_TO_BTF_ID points to a kernel struct that does not need
	 * to be null checked by the BPF program. This does not imply the
	 * pointer is _not_ null and in practice this can easily be a null
	 * pointer when reading pointer chains. The assumption is program
	 * context will handle null pointer dereference typically via fault
	 * handling. The verifier must keep this in mind and can make no
	 * assumptions about null or non-null when doing branch analysis.
	 * Further, when passed into helpers the helpers can not, without
	 * additional context, assume the value is non-null.
	 */
	PTR_TO_BTF_ID,
	/* PTR_TO_BTF_ID_OR_NULL points to a kernel struct that has not
	 * been checked for null. Used primarily to inform the verifier
	 * an explicit null check is required for this struct.
	 */
	PTR_TO_BTF_ID_OR_NULL,
417 418
	PTR_TO_MEM,		 /* reg points to valid memory region */
	PTR_TO_MEM_OR_NULL,	 /* reg points to valid memory region or NULL */
419 420 421 422
	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 */
H
Hao Luo 已提交
423
	PTR_TO_PERCPU_BTF_ID,	 /* reg points to a percpu kernel variable */
424 425
	PTR_TO_FUNC,		 /* reg points to a bpf program function */
	PTR_TO_MAP_KEY,		 /* reg points to a map element key */
426 427
};

428 429 430 431 432
/* The information passed from prog-specific *_is_valid_access
 * back to the verifier.
 */
struct bpf_insn_access_aux {
	enum bpf_reg_type reg_type;
433 434
	union {
		int ctx_field_size;
435 436 437 438
		struct {
			struct btf *btf;
			u32 btf_id;
		};
439 440
	};
	struct bpf_verifier_log *log; /* for verbose logs */
441 442
};

443 444 445 446 447 448
static inline void
bpf_ctx_record_field_size(struct bpf_insn_access_aux *aux, u32 size)
{
	aux->ctx_field_size = size;
}

449 450 451 452 453
struct bpf_prog_ops {
	int (*test_run)(struct bpf_prog *prog, const union bpf_attr *kattr,
			union bpf_attr __user *uattr);
};

454 455
struct bpf_verifier_ops {
	/* return eBPF function prototype for verification */
456 457 458
	const struct bpf_func_proto *
	(*get_func_proto)(enum bpf_func_id func_id,
			  const struct bpf_prog *prog);
459 460 461 462

	/* return true if 'size' wide access at offset 'off' within bpf_context
	 * with 'type' (read or write) is allowed
	 */
463
	bool (*is_valid_access)(int off, int size, enum bpf_access_type type,
464
				const struct bpf_prog *prog,
465
				struct bpf_insn_access_aux *info);
466 467
	int (*gen_prologue)(struct bpf_insn *insn, bool direct_write,
			    const struct bpf_prog *prog);
468 469
	int (*gen_ld_abs)(const struct bpf_insn *orig,
			  struct bpf_insn *insn_buf);
470 471 472
	u32 (*convert_ctx_access)(enum bpf_access_type type,
				  const struct bpf_insn *src,
				  struct bpf_insn *dst,
473
				  struct bpf_prog *prog, u32 *target_size);
474
	int (*btf_struct_access)(struct bpf_verifier_log *log,
475
				 const struct btf *btf,
476 477 478
				 const struct btf_type *t, int off, int size,
				 enum bpf_access_type atype,
				 u32 *next_btf_id);
479 480
};

481
struct bpf_prog_offload_ops {
482
	/* verifier basic callbacks */
483 484
	int (*insn_hook)(struct bpf_verifier_env *env,
			 int insn_idx, int prev_insn_idx);
485
	int (*finalize)(struct bpf_verifier_env *env);
486 487 488 489 490
	/* 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 */
491 492
	int (*prepare)(struct bpf_prog *prog);
	int (*translate)(struct bpf_prog *prog);
493
	void (*destroy)(struct bpf_prog *prog);
494 495
};

496
struct bpf_prog_offload {
497 498
	struct bpf_prog		*prog;
	struct net_device	*netdev;
499
	struct bpf_offload_dev	*offdev;
500 501 502
	void			*dev_priv;
	struct list_head	offloads;
	bool			dev_state;
503
	bool			opt_failed;
504 505
	void			*jited_image;
	u32			jited_len;
506 507
};

508 509
enum bpf_cgroup_storage_type {
	BPF_CGROUP_STORAGE_SHARED,
510
	BPF_CGROUP_STORAGE_PERCPU,
511 512 513 514 515
	__BPF_CGROUP_STORAGE_MAX
};

#define MAX_BPF_CGROUP_STORAGE_TYPE __BPF_CGROUP_STORAGE_MAX

516 517 518 519 520
/* The longest tracepoint has 12 args.
 * See include/trace/bpf_probe.h
 */
#define MAX_BPF_FUNC_ARGS 12

521 522 523 524 525
/* The maximum number of arguments passed through registers
 * a single function may have.
 */
#define MAX_BPF_FUNC_REG_ARGS 5

A
Alexei Starovoitov 已提交
526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545
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
546 547 548
/* 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
 */
549
#define BPF_MAX_TRAMP_PROGS 38
K
KP Singh 已提交
550 551 552 553 554 555

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

A
Alexei Starovoitov 已提交
556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575
/* 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
 */
576 577
int arch_prepare_bpf_trampoline(void *image, void *image_end,
				const struct btf_func_model *m, u32 flags,
K
KP Singh 已提交
578
				struct bpf_tramp_progs *tprogs,
A
Alexei Starovoitov 已提交
579 580
				void *orig_call);
/* these two functions are called from generated trampoline */
581
u64 notrace __bpf_prog_enter(struct bpf_prog *prog);
A
Alexei Starovoitov 已提交
582
void notrace __bpf_prog_exit(struct bpf_prog *prog, u64 start);
583
u64 notrace __bpf_prog_enter_sleepable(struct bpf_prog *prog);
584
void notrace __bpf_prog_exit_sleepable(struct bpf_prog *prog, u64 start);
A
Alexei Starovoitov 已提交
585

J
Jiri Olsa 已提交
586 587 588
struct bpf_ksym {
	unsigned long		 start;
	unsigned long		 end;
J
Jiri Olsa 已提交
589
	char			 name[KSYM_NAME_LEN];
590
	struct list_head	 lnode;
J
Jiri Olsa 已提交
591
	struct latch_tree_node	 tnode;
592
	bool			 prog;
J
Jiri Olsa 已提交
593 594
};

A
Alexei Starovoitov 已提交
595 596 597
enum bpf_tramp_prog_type {
	BPF_TRAMP_FENTRY,
	BPF_TRAMP_FEXIT,
K
KP Singh 已提交
598
	BPF_TRAMP_MODIFY_RETURN,
599 600
	BPF_TRAMP_MAX,
	BPF_TRAMP_REPLACE, /* more than MAX */
A
Alexei Starovoitov 已提交
601 602 603 604 605 606 607 608 609 610 611 612
};

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;
613
		bool ftrace_managed;
A
Alexei Starovoitov 已提交
614
	} func;
615 616 617 618 619
	/* 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
620 621 622 623 624 625 626
	/* 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
627
	struct bpf_ksym ksym;
A
Alexei Starovoitov 已提交
628
};
B
Björn Töpel 已提交
629

630 631 632 633 634 635 636
struct bpf_attach_target_info {
	struct btf_func_model fmodel;
	long tgt_addr;
	const char *tgt_name;
	const struct btf_type *tgt_type;
};

637
#define BPF_DISPATCHER_MAX 48 /* Fits in 2048B */
B
Björn Töpel 已提交
638 639 640 641 642 643 644 645 646 647 648 649 650 651

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
652
	struct bpf_ksym ksym;
B
Björn Töpel 已提交
653 654
};

655
static __always_inline unsigned int bpf_dispatcher_nop_func(
656 657 658 659 660 661 662
	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
663
#ifdef CONFIG_BPF_JIT
664 665
int bpf_trampoline_link_prog(struct bpf_prog *prog, struct bpf_trampoline *tr);
int bpf_trampoline_unlink_prog(struct bpf_prog *prog, struct bpf_trampoline *tr);
666 667
struct bpf_trampoline *bpf_trampoline_get(u64 key,
					  struct bpf_attach_target_info *tgt_info);
A
Alexei Starovoitov 已提交
668
void bpf_trampoline_put(struct bpf_trampoline *tr);
J
Jiri Olsa 已提交
669 670 671 672 673 674 675 676 677 678 679
#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
680 681 682
}

#define DEFINE_BPF_DISPATCHER(name)					\
683
	noinline unsigned int bpf_dispatcher_##name##_func(		\
B
Björn Töpel 已提交
684 685 686 687 688 689 690
		const void *ctx,					\
		const struct bpf_insn *insnsi,				\
		unsigned int (*bpf_func)(const void *,			\
					 const struct bpf_insn *))	\
	{								\
		return bpf_func(ctx, insnsi);				\
	}								\
691 692 693
	EXPORT_SYMBOL(bpf_dispatcher_##name##_func);			\
	struct bpf_dispatcher bpf_dispatcher_##name =			\
		BPF_DISPATCHER_INIT(bpf_dispatcher_##name);
B
Björn Töpel 已提交
694
#define DECLARE_BPF_DISPATCHER(name)					\
695
	unsigned int bpf_dispatcher_##name##_func(			\
B
Björn Töpel 已提交
696 697 698 699
		const void *ctx,					\
		const struct bpf_insn *insnsi,				\
		unsigned int (*bpf_func)(const void *,			\
					 const struct bpf_insn *));	\
700 701 702
	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
703 704
void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
				struct bpf_prog *to);
J
Jiri Olsa 已提交
705
/* Called only from JIT-enabled code, so there's no need for stubs. */
J
Jiri Olsa 已提交
706
void *bpf_jit_alloc_exec_page(void);
J
Jiri Olsa 已提交
707 708
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
709 710
void bpf_ksym_add(struct bpf_ksym *ksym);
void bpf_ksym_del(struct bpf_ksym *ksym);
A
Alexei Starovoitov 已提交
711
#else
712 713
static inline int bpf_trampoline_link_prog(struct bpf_prog *prog,
					   struct bpf_trampoline *tr)
A
Alexei Starovoitov 已提交
714 715 716
{
	return -ENOTSUPP;
}
717 718
static inline int bpf_trampoline_unlink_prog(struct bpf_prog *prog,
					     struct bpf_trampoline *tr)
A
Alexei Starovoitov 已提交
719 720 721
{
	return -ENOTSUPP;
}
722 723 724 725 726
static inline struct bpf_trampoline *bpf_trampoline_get(u64 key,
							struct bpf_attach_target_info *tgt_info)
{
	return ERR_PTR(-EOPNOTSUPP);
}
A
Alexei Starovoitov 已提交
727
static inline void bpf_trampoline_put(struct bpf_trampoline *tr) {}
B
Björn Töpel 已提交
728 729
#define DEFINE_BPF_DISPATCHER(name)
#define DECLARE_BPF_DISPATCHER(name)
730
#define BPF_DISPATCHER_FUNC(name) bpf_dispatcher_nop_func
B
Björn Töpel 已提交
731 732 733 734
#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) {}
735 736 737 738
static inline bool is_bpf_image_address(unsigned long address)
{
	return false;
}
A
Alexei Starovoitov 已提交
739 740
#endif

741
struct bpf_func_info_aux {
742
	u16 linkage;
743 744 745
	bool unreliable;
};

746 747 748 749 750 751
enum bpf_jit_poke_reason {
	BPF_POKE_REASON_TAIL_CALL,
};

/* Descriptor of pokes pointing /into/ the JITed image. */
struct bpf_jit_poke_descriptor {
752
	void *tailcall_target;
753 754
	void *tailcall_bypass;
	void *bypass_addr;
755 756 757 758 759 760
	union {
		struct {
			struct bpf_map *map;
			u32 key;
		} tail_call;
	};
761
	bool tailcall_target_stable;
762 763
	u8 adj_off;
	u16 reason;
764
	u32 insn_idx;
765 766
};

767 768 769 770
/* reg_type info for ctx arguments */
struct bpf_ctx_arg_aux {
	u32 offset;
	enum bpf_reg_type reg_type;
771
	u32 btf_id;
772 773
};

774 775 776 777 778
struct btf_mod_pair {
	struct btf *btf;
	struct module *module;
};

779
struct bpf_prog_aux {
780
	atomic64_t refcnt;
781
	u32 used_map_cnt;
782
	u32 used_btf_cnt;
783
	u32 max_ctx_offset;
784
	u32 max_pkt_offset;
785
	u32 max_tp_access;
786
	u32 stack_depth;
M
Martin KaFai Lau 已提交
787
	u32 id;
788 789
	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 */
790
	u32 attach_btf_id; /* in-kernel BTF type id to attach to */
791
	u32 ctx_arg_info_size;
792 793
	u32 max_rdonly_access;
	u32 max_rdwr_access;
794
	struct btf *attach_btf;
795
	const struct bpf_ctx_arg_aux *ctx_arg_info;
796 797 798
	struct mutex dst_mutex; /* protects dst_* pointers below, *after* prog becomes visible */
	struct bpf_prog *dst_prog;
	struct bpf_trampoline *dst_trampoline;
799 800
	enum bpf_prog_type saved_dst_prog_type;
	enum bpf_attach_type saved_dst_attach_type;
801
	bool verifier_zext; /* Zero extensions has been inserted by verifier. */
802
	bool offload_requested;
803
	bool attach_btf_trace; /* true if attaching to BTF-enabled raw tp */
804
	bool func_proto_unreliable;
805
	bool sleepable;
806
	bool tail_call_reachable;
A
Alexei Starovoitov 已提交
807 808
	enum bpf_tramp_prog_type trampoline_prog_type;
	struct hlist_node tramp_hlist;
809 810 811 812
	/* 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;
813 814
	struct bpf_prog **func;
	void *jit_data; /* JIT specific data. arch dependent */
815 816
	struct bpf_jit_poke_descriptor *poke_tab;
	u32 size_poke_tab;
J
Jiri Olsa 已提交
817
	struct bpf_ksym ksym;
818
	const struct bpf_prog_ops *ops;
819
	struct bpf_map **used_maps;
820
	struct mutex used_maps_mutex; /* mutex for used_maps and used_map_cnt */
821
	struct btf_mod_pair *used_btfs;
822
	struct bpf_prog *prog;
823
	struct user_struct *user;
824
	u64 load_time; /* ns since boottime */
825
	struct bpf_map *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
826
	char name[BPF_OBJ_NAME_LEN];
827 828 829
#ifdef CONFIG_SECURITY
	void *security;
#endif
830
	struct bpf_prog_offload *offload;
Y
Yonghong Song 已提交
831
	struct btf *btf;
832
	struct bpf_func_info *func_info;
833
	struct bpf_func_info_aux *func_info_aux;
M
Martin KaFai Lau 已提交
834 835 836 837 838 839 840 841 842 843 844 845 846 847 848
	/* 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;
849
	u32 func_info_cnt;
M
Martin KaFai Lau 已提交
850 851 852 853 854 855
	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;
856 857
	u32 num_exentries;
	struct exception_table_entry *extable;
858 859 860 861
	union {
		struct work_struct work;
		struct rcu_head	rcu;
	};
862 863
};

864 865 866 867 868 869 870 871
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;
872 873 874 875 876
	/* 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;
877 878
};

879 880 881 882 883 884 885 886 887 888 889 890
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);
891
	int (*detach)(struct bpf_link *link);
892 893 894 895 896 897 898 899 900 901 902 903 904 905
	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;
};

906
struct bpf_struct_ops_value;
907 908 909 910 911 912 913 914 915
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);
916 917 918 919 920
	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);
921
	const struct btf_type *type;
922
	const struct btf_type *value_type;
923 924 925
	const char *name;
	struct btf_func_model func_models[BPF_STRUCT_OPS_MAX_NR_MEMBERS];
	u32 type_id;
926
	u32 value_id;
927 928 929
};

#if defined(CONFIG_BPF_JIT) && defined(CONFIG_BPF_SYSCALL)
930
#define BPF_MODULE_OWNER ((void *)((0xeB9FUL << 2) + POISON_POINTER_DELTA))
931
const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id);
932
void bpf_struct_ops_init(struct btf *btf, struct bpf_verifier_log *log);
933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950
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);
}
951 952 953 954 955
#else
static inline const struct bpf_struct_ops *bpf_struct_ops_find(u32 type_id)
{
	return NULL;
}
956 957 958 959
static inline void bpf_struct_ops_init(struct btf *btf,
				       struct bpf_verifier_log *log)
{
}
960 961 962 963 964 965 966 967 968 969 970 971 972 973
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;
}
974 975
#endif

976 977 978
struct bpf_array {
	struct bpf_map map;
	u32 elem_size;
979
	u32 index_mask;
980
	struct bpf_array_aux *aux;
981 982
	union {
		char value[0] __aligned(8);
983
		void *ptrs[0] __aligned(8);
984
		void __percpu *pptrs[0] __aligned(8);
985 986
	};
};
987

988
#define BPF_COMPLEXITY_LIMIT_INSNS      1000000 /* yes. 1M insns */
989 990
#define MAX_TAIL_CALL_CNT 32

991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019
#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);
}

1020 1021 1022 1023 1024 1025 1026
struct bpf_event_entry {
	struct perf_event *event;
	struct file *perf_file;
	struct file *map_file;
	struct rcu_head rcu;
};

1027
bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp);
1028
int bpf_prog_calc_tag(struct bpf_prog *fp);
1029

1030
const struct bpf_func_proto *bpf_get_trace_printk_proto(void);
1031 1032

typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src,
1033
					unsigned long off, unsigned long len);
1034 1035 1036 1037 1038
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);
1039 1040 1041

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);
1042

1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054
/* 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.
 */
1055 1056
struct bpf_prog_array_item {
	struct bpf_prog *prog;
1057
	struct bpf_cgroup_storage *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];
1058 1059
};

1060 1061
struct bpf_prog_array {
	struct rcu_head rcu;
1062
	struct bpf_prog_array_item items[];
1063 1064
};

1065
struct bpf_prog_array *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags);
1066 1067
void bpf_prog_array_free(struct bpf_prog_array *progs);
int bpf_prog_array_length(struct bpf_prog_array *progs);
1068
bool bpf_prog_array_is_empty(struct bpf_prog_array *array);
1069
int bpf_prog_array_copy_to_user(struct bpf_prog_array *progs,
1070
				__u32 __user *prog_ids, u32 cnt);
1071

1072
void bpf_prog_array_delete_safe(struct bpf_prog_array *progs,
1073
				struct bpf_prog *old_prog);
1074 1075 1076
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);
1077
int bpf_prog_array_copy_info(struct bpf_prog_array *array,
1078 1079
			     u32 *prog_ids, u32 request_cnt,
			     u32 *prog_cnt);
1080
int bpf_prog_array_copy(struct bpf_prog_array *old_array,
1081 1082 1083 1084
			struct bpf_prog *exclude_prog,
			struct bpf_prog *include_prog,
			struct bpf_prog_array **new_array);

1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112
/* BPF program asks to bypass CAP_NET_BIND_SERVICE in bind. */
#define BPF_RET_BIND_NO_CAP_NET_BIND_SERVICE			(1 << 0)
/* BPF program asks to set CN on the packet. */
#define BPF_RET_SET_CN						(1 << 0)

#define BPF_PROG_RUN_ARRAY_FLAGS(array, ctx, func, ret_flags)		\
	({								\
		struct bpf_prog_array_item *_item;			\
		struct bpf_prog *_prog;					\
		struct bpf_prog_array *_array;				\
		u32 _ret = 1;						\
		u32 func_ret;						\
		migrate_disable();					\
		rcu_read_lock();					\
		_array = rcu_dereference(array);			\
		_item = &_array->items[0];				\
		while ((_prog = READ_ONCE(_item->prog))) {		\
			bpf_cgroup_storage_set(_item->cgroup_storage);	\
			func_ret = func(_prog, ctx);			\
			_ret &= (func_ret & 1);				\
			*(ret_flags) |= (func_ret >> 1);			\
			_item++;					\
		}							\
		rcu_read_unlock();					\
		migrate_enable();					\
		_ret;							\
	 })

1113
#define __BPF_PROG_RUN_ARRAY(array, ctx, func, check_non_null)	\
1114
	({						\
1115 1116
		struct bpf_prog_array_item *_item;	\
		struct bpf_prog *_prog;			\
1117
		struct bpf_prog_array *_array;		\
1118
		u32 _ret = 1;				\
1119
		migrate_disable();			\
1120
		rcu_read_lock();			\
1121 1122 1123
		_array = rcu_dereference(array);	\
		if (unlikely(check_non_null && !_array))\
			goto _out;			\
1124 1125 1126 1127 1128
		_item = &_array->items[0];		\
		while ((_prog = READ_ONCE(_item->prog))) {		\
			bpf_cgroup_storage_set(_item->cgroup_storage);	\
			_ret &= func(_prog, ctx);	\
			_item++;			\
1129 1130
		}					\
_out:							\
1131
		rcu_read_unlock();			\
1132
		migrate_enable();			\
1133 1134 1135
		_ret;					\
	 })

1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159
/* 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)		\
	({						\
1160 1161 1162 1163 1164
		u32 _flags = 0;				\
		bool _cn;				\
		u32 _ret;				\
		_ret = BPF_PROG_RUN_ARRAY_FLAGS(array, ctx, func, &_flags); \
		_cn = _flags & BPF_RET_SET_CN;		\
1165 1166 1167 1168 1169 1170 1171
		if (_ret)				\
			_ret = (_cn ? NET_XMIT_CN : NET_XMIT_SUCCESS);	\
		else					\
			_ret = (_cn ? NET_XMIT_DROP : -EPERM);		\
		_ret;					\
	})

1172 1173 1174 1175 1176 1177
#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)

1178
#ifdef CONFIG_BPF_SYSCALL
1179
DECLARE_PER_CPU(int, bpf_prog_active);
1180
extern struct mutex bpf_stats_enabled_mutex;
1181

1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211
/*
 * 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();
}

1212 1213
extern const struct file_operations bpf_map_fops;
extern const struct file_operations bpf_prog_fops;
Y
Yonghong Song 已提交
1214
extern const struct file_operations bpf_iter_fops;
1215

A
Alexei Starovoitov 已提交
1216
#define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \
1217 1218
	extern const struct bpf_prog_ops _name ## _prog_ops; \
	extern const struct bpf_verifier_ops _name ## _verifier_ops;
1219 1220
#define BPF_MAP_TYPE(_id, _ops) \
	extern const struct bpf_map_ops _ops;
1221
#define BPF_LINK_TYPE(_id, _name)
1222 1223
#include <linux/bpf_types.h>
#undef BPF_PROG_TYPE
1224
#undef BPF_MAP_TYPE
1225
#undef BPF_LINK_TYPE
1226

1227
extern const struct bpf_prog_ops bpf_offload_prog_ops;
1228 1229 1230
extern const struct bpf_verifier_ops tc_cls_act_analyzer_ops;
extern const struct bpf_verifier_ops xdp_analyzer_ops;

1231
struct bpf_prog *bpf_prog_get(u32 ufd);
1232
struct bpf_prog *bpf_prog_get_type_dev(u32 ufd, enum bpf_prog_type type,
1233
				       bool attach_drv);
1234
void bpf_prog_add(struct bpf_prog *prog, int i);
1235
void bpf_prog_sub(struct bpf_prog *prog, int i);
1236
void bpf_prog_inc(struct bpf_prog *prog);
1237
struct bpf_prog * __must_check bpf_prog_inc_not_zero(struct bpf_prog *prog);
1238 1239
void bpf_prog_put(struct bpf_prog *prog);

1240
void bpf_prog_free_id(struct bpf_prog *prog, bool do_idr_lock);
1241
void bpf_map_free_id(struct bpf_map *map, bool do_idr_lock);
1242

1243
struct bpf_map *bpf_map_get(u32 ufd);
1244
struct bpf_map *bpf_map_get_with_uref(u32 ufd);
1245
struct bpf_map *__bpf_map_get(struct fd f);
1246 1247 1248
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);
1249
void bpf_map_put_with_uref(struct bpf_map *map);
1250
void bpf_map_put(struct bpf_map *map);
1251 1252
void *bpf_map_area_alloc(u64 size, int numa_node);
void *bpf_map_area_mmapable_alloc(u64 size, int numa_node);
1253
void bpf_map_area_free(void *base);
1254
void bpf_map_init_from_attr(struct bpf_map *map, union bpf_attr *attr);
1255 1256
int  generic_map_lookup_batch(struct bpf_map *map,
			      const union bpf_attr *attr,
1257 1258 1259 1260 1261 1262
			      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,
1263
			      union bpf_attr __user *uattr);
Y
Yonghong Song 已提交
1264
struct bpf_map *bpf_map_get_curr_or_next(u32 *id);
A
Alexei Starovoitov 已提交
1265
struct bpf_prog *bpf_prog_get_curr_or_next(u32 *id);
1266

1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294
#ifdef CONFIG_MEMCG_KMEM
void *bpf_map_kmalloc_node(const struct bpf_map *map, size_t size, gfp_t flags,
			   int node);
void *bpf_map_kzalloc(const struct bpf_map *map, size_t size, gfp_t flags);
void __percpu *bpf_map_alloc_percpu(const struct bpf_map *map, size_t size,
				    size_t align, gfp_t flags);
#else
static inline void *
bpf_map_kmalloc_node(const struct bpf_map *map, size_t size, gfp_t flags,
		     int node)
{
	return kmalloc_node(size, flags, node);
}

static inline void *
bpf_map_kzalloc(const struct bpf_map *map, size_t size, gfp_t flags)
{
	return kzalloc(size, flags);
}

static inline void __percpu *
bpf_map_alloc_percpu(const struct bpf_map *map, size_t size, size_t align,
		     gfp_t flags)
{
	return __alloc_percpu_gfp(size, align, flags);
}
#endif

1295 1296
extern int sysctl_unprivileged_bpf_disabled;

A
Alexei Starovoitov 已提交
1297 1298 1299 1300 1301
static inline bool bpf_allow_ptr_leaks(void)
{
	return perfmon_capable();
}

1302 1303 1304 1305 1306
static inline bool bpf_allow_uninit_stack(void)
{
	return perfmon_capable();
}

1307 1308 1309 1310 1311
static inline bool bpf_allow_ptr_to_map_access(void)
{
	return perfmon_capable();
}

A
Alexei Starovoitov 已提交
1312 1313 1314 1315 1316 1317 1318 1319 1320 1321
static inline bool bpf_bypass_spec_v1(void)
{
	return perfmon_capable();
}

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

1322
int bpf_map_new_fd(struct bpf_map *map, int flags);
1323 1324
int bpf_prog_new_fd(struct bpf_prog *prog);

1325
void bpf_link_init(struct bpf_link *link, enum bpf_link_type type,
A
Andrii Nakryiko 已提交
1326 1327 1328 1329
		   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);
1330 1331 1332
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);
1333
struct file *bpf_link_new_file(struct bpf_link *link, int *reserved_fd);
1334 1335
struct bpf_link *bpf_link_get_from_fd(u32 ufd);

1336
int bpf_obj_pin_user(u32 ufd, const char __user *pathname);
1337
int bpf_obj_get_user(const char __user *pathname, int flags);
1338

1339
#define BPF_ITER_FUNC_PREFIX "bpf_iter_"
1340
#define DEFINE_BPF_ITER_FUNC(target, args...)			\
1341 1342
	extern int bpf_iter_ ## target(args);			\
	int __init bpf_iter_ ## target(args) { return 0; }
1343

1344
struct bpf_iter_aux_info {
1345
	struct bpf_map *map;
1346 1347
};

1348 1349 1350 1351
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);
1352 1353 1354 1355
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);
1356

1357 1358 1359 1360
enum bpf_iter_feature {
	BPF_ITER_RESCHED	= BIT(0),
};

1361
#define BPF_ITER_CTX_ARG_MAX 2
1362 1363
struct bpf_iter_reg {
	const char *target;
1364 1365
	bpf_iter_attach_target_t attach_target;
	bpf_iter_detach_target_t detach_target;
1366 1367
	bpf_iter_show_fdinfo_t show_fdinfo;
	bpf_iter_fill_link_info_t fill_link_info;
1368
	u32 ctx_arg_info_size;
1369
	u32 feature;
1370
	struct bpf_ctx_arg_aux ctx_arg_info[BPF_ITER_CTX_ARG_MAX];
1371
	const struct bpf_iter_seq_info *seq_info;
1372 1373
};

1374 1375 1376 1377 1378 1379
struct bpf_iter_meta {
	__bpf_md_ptr(struct seq_file *, seq);
	u64 session_id;
	u64 seq_num;
};

1380 1381 1382 1383 1384 1385 1386
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);
};

1387
int bpf_iter_reg_target(const struct bpf_iter_reg *reg_info);
1388
void bpf_iter_unreg_target(const struct bpf_iter_reg *reg_info);
1389
bool bpf_iter_prog_supported(struct bpf_prog *prog);
1390
int bpf_iter_link_attach(const union bpf_attr *attr, struct bpf_prog *prog);
1391
int bpf_iter_new_fd(struct bpf_link *link);
Y
Yonghong Song 已提交
1392
bool bpf_link_is_iter(struct bpf_link *link);
1393 1394
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);
1395 1396 1397 1398
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);
1399

1400 1401 1402 1403 1404 1405
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);
1406

1407
int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value);
1408

1409 1410
int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file,
				 void *key, void *value, u64 map_flags);
1411
int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
M
Martin KaFai Lau 已提交
1412 1413
int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file,
				void *key, void *value, u64 map_flags);
1414
int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value);
1415

1416
int bpf_get_file_flag(int flags);
1417 1418
int bpf_check_uarg_tail_zero(void __user *uaddr, size_t expected_size,
			     size_t actual_size);
1419

1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435
/* 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++;
}

1436
/* verify correctness of eBPF program */
Y
Yonghong Song 已提交
1437 1438
int bpf_check(struct bpf_prog **fp, union bpf_attr *attr,
	      union bpf_attr __user *uattr);
1439 1440

#ifndef CONFIG_BPF_JIT_ALWAYS_ON
1441
void bpf_patch_call_args(struct bpf_insn *insn, u32 stack_depth);
1442
#endif
1443

1444 1445
struct btf *bpf_get_btf_vmlinux(void);

1446
/* Map specifics */
1447
struct xdp_buff;
1448
struct sk_buff;
1449 1450

struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
1451
struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
1452 1453 1454
void __dev_flush(void);
int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
1455 1456
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
1457 1458
int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
			     struct bpf_prog *xdp_prog);
1459
bool dev_map_can_have_prog(struct bpf_map *map);
1460

1461
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
1462
void __cpu_map_flush(void);
1463 1464
int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
		    struct net_device *dev_rx);
1465
bool cpu_map_prog_allowed(struct bpf_map *map);
1466

1467 1468 1469 1470 1471 1472 1473
/* 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;
}

1474
struct bpf_prog *bpf_prog_get_type_path(const char *name, enum bpf_prog_type type);
1475
int array_map_alloc_check(union bpf_attr *attr);
1476

1477 1478 1479 1480
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);
1481 1482 1483
int bpf_prog_test_run_tracing(struct bpf_prog *prog,
			      const union bpf_attr *kattr,
			      union bpf_attr __user *uattr);
1484 1485 1486
int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog,
				     const union bpf_attr *kattr,
				     union bpf_attr __user *uattr);
1487 1488 1489
int bpf_prog_test_run_raw_tp(struct bpf_prog *prog,
			     const union bpf_attr *kattr,
			     union bpf_attr __user *uattr);
1490 1491 1492
bool btf_ctx_access(int off, int size, enum bpf_access_type type,
		    const struct bpf_prog *prog,
		    struct bpf_insn_access_aux *info);
1493
int btf_struct_access(struct bpf_verifier_log *log, const struct btf *btf,
1494 1495 1496
		      const struct btf_type *t, int off, int size,
		      enum bpf_access_type atype,
		      u32 *next_btf_id);
1497
bool btf_struct_ids_match(struct bpf_verifier_log *log,
1498 1499
			  const struct btf *btf, u32 id, int off,
			  const struct btf *need_btf, u32 need_type_id);
1500

A
Alexei Starovoitov 已提交
1501 1502 1503 1504 1505 1506
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);

1507 1508 1509 1510 1511
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);
1512
int btf_check_type_match(struct bpf_verifier_log *log, const struct bpf_prog *prog,
1513
			 struct btf *btf, const struct btf_type *t);
1514

1515
struct bpf_prog *bpf_prog_by_id(u32 id);
1516
struct bpf_link *bpf_link_by_id(u32 id);
1517

1518
const struct bpf_func_proto *bpf_base_func_proto(enum bpf_func_id func_id);
1519
void bpf_task_storage_free(struct task_struct *task);
1520
#else /* !CONFIG_BPF_SYSCALL */
1521 1522 1523 1524 1525
static inline struct bpf_prog *bpf_prog_get(u32 ufd)
{
	return ERR_PTR(-EOPNOTSUPP);
}

1526 1527
static inline struct bpf_prog *bpf_prog_get_type_dev(u32 ufd,
						     enum bpf_prog_type type,
1528
						     bool attach_drv)
1529 1530 1531 1532
{
	return ERR_PTR(-EOPNOTSUPP);
}

1533
static inline void bpf_prog_add(struct bpf_prog *prog, int i)
1534 1535
{
}
1536

1537 1538 1539 1540
static inline void bpf_prog_sub(struct bpf_prog *prog, int i)
{
}

1541 1542 1543
static inline void bpf_prog_put(struct bpf_prog *prog)
{
}
1544

1545
static inline void bpf_prog_inc(struct bpf_prog *prog)
1546 1547
{
}
1548

1549 1550 1551 1552 1553 1554
static inline struct bpf_prog *__must_check
bpf_prog_inc_not_zero(struct bpf_prog *prog)
{
	return ERR_PTR(-EOPNOTSUPP);
}

1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583
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)
{
}

1584
static inline int bpf_obj_get_user(const char __user *pathname, int flags)
1585 1586 1587 1588
{
	return -EOPNOTSUPP;
}

1589 1590 1591 1592 1593 1594
static inline struct net_device  *__dev_map_lookup_elem(struct bpf_map *map,
						       u32 key)
{
	return NULL;
}

1595 1596 1597 1598 1599
static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map,
							     u32 key)
{
	return NULL;
}
1600 1601 1602 1603
static inline bool dev_map_can_have_prog(struct bpf_map *map)
{
	return false;
}
1604

1605
static inline void __dev_flush(void)
1606 1607
{
}
1608

1609 1610 1611
struct xdp_buff;
struct bpf_dtab_netdev;

1612 1613 1614 1615 1616 1617 1618
static inline
int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
		    struct net_device *dev_rx)
{
	return 0;
}

1619
static inline
1620 1621
int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
		    struct net_device *dev_rx)
1622 1623 1624 1625
{
	return 0;
}

1626 1627 1628 1629 1630 1631 1632 1633 1634
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;
}

1635 1636 1637 1638 1639 1640
static inline
struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key)
{
	return NULL;
}

1641
static inline void __cpu_map_flush(void)
1642 1643 1644 1645 1646 1647 1648 1649 1650
{
}

static inline int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu,
				  struct xdp_buff *xdp,
				  struct net_device *dev_rx)
{
	return 0;
}
1651

1652 1653 1654 1655 1656
static inline bool cpu_map_prog_allowed(struct bpf_map *map)
{
	return false;
}

1657 1658 1659 1660 1661
static inline struct bpf_prog *bpf_prog_get_type_path(const char *name,
				enum bpf_prog_type type)
{
	return ERR_PTR(-EOPNOTSUPP);
}
1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676

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;
}

1677 1678 1679 1680 1681 1682 1683
static inline int bpf_prog_test_run_tracing(struct bpf_prog *prog,
					    const union bpf_attr *kattr,
					    union bpf_attr __user *uattr)
{
	return -ENOTSUPP;
}

1684 1685 1686 1687 1688 1689
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;
}
1690 1691 1692 1693

static inline void bpf_map_put(struct bpf_map *map)
{
}
1694 1695 1696 1697 1698

static inline struct bpf_prog *bpf_prog_by_id(u32 id)
{
	return ERR_PTR(-ENOTSUPP);
}
1699 1700 1701 1702 1703 1704

static inline const struct bpf_func_proto *
bpf_base_func_proto(enum bpf_func_id func_id)
{
	return NULL;
}
1705 1706 1707 1708

static inline void bpf_task_storage_free(struct task_struct *task)
{
}
1709
#endif /* CONFIG_BPF_SYSCALL */
1710

1711 1712 1713
void __bpf_free_used_btfs(struct bpf_prog_aux *aux,
			  struct btf_mod_pair *used_btfs, u32 len);

1714 1715 1716 1717 1718 1719
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);
}

1720 1721 1722
void __bpf_free_used_maps(struct bpf_prog_aux *aux,
			  struct bpf_map **used_maps, u32 len);

1723 1724
bool bpf_prog_get_ok(struct bpf_prog *, enum bpf_prog_type *, bool);

1725 1726
int bpf_prog_offload_compile(struct bpf_prog *prog);
void bpf_prog_offload_destroy(struct bpf_prog *prog);
1727 1728
int bpf_prog_offload_info_fill(struct bpf_prog_info *info,
			       struct bpf_prog *prog);
1729

1730 1731
int bpf_map_offload_info_fill(struct bpf_map_info *info, struct bpf_map *map);

1732 1733 1734 1735 1736 1737 1738
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);

1739
bool bpf_offload_prog_map_match(struct bpf_prog *prog, struct bpf_map *map);
1740

1741
struct bpf_offload_dev *
1742
bpf_offload_dev_create(const struct bpf_prog_offload_ops *ops, void *priv);
1743
void bpf_offload_dev_destroy(struct bpf_offload_dev *offdev);
1744
void *bpf_offload_dev_priv(struct bpf_offload_dev *offdev);
1745 1746 1747 1748
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);
1749
bool bpf_offload_dev_match(struct bpf_prog *prog, struct net_device *netdev);
1750

1751 1752 1753
#if defined(CONFIG_NET) && defined(CONFIG_BPF_SYSCALL)
int bpf_prog_offload_init(struct bpf_prog *prog, union bpf_attr *attr);

1754
static inline bool bpf_prog_is_dev_bound(const struct bpf_prog_aux *aux)
1755
{
1756
	return aux->offload_requested;
1757
}
1758 1759 1760 1761 1762 1763 1764 1765

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);
1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776
#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;
}
1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790

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)
{
}
1791 1792
#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */

1793
#if defined(CONFIG_INET) && defined(CONFIG_BPF_SYSCALL)
1794
int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog *prog);
1795
int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type ptype);
1796
int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value, u64 flags);
1797 1798
void sock_map_unhash(struct sock *sk);
void sock_map_close(struct sock *sk, long timeout);
1799 1800 1801 1802 1803 1804

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);
1805
#else
1806 1807 1808 1809 1810
static inline void bpf_sk_reuseport_detach(struct sock *sk)
{
}

#ifdef CONFIG_BPF_SYSCALL
1811 1812
static inline int sock_map_get_from_fd(const union bpf_attr *attr,
				       struct bpf_prog *prog)
1813 1814 1815
{
	return -EINVAL;
}
1816 1817 1818 1819 1820 1821

static inline int sock_map_prog_detach(const union bpf_attr *attr,
				       enum bpf_prog_type ptype)
{
	return -EOPNOTSUPP;
}
1822 1823 1824 1825 1826 1827

static inline int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value,
					   u64 flags)
{
	return -EOPNOTSUPP;
}
1828

1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843
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) */

1844
/* verifier prototypes for helper functions called from eBPF programs */
1845 1846 1847
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
1848 1849 1850
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;
1851

1852
extern const struct bpf_func_proto bpf_get_prandom_u32_proto;
1853
extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
1854
extern const struct bpf_func_proto bpf_get_numa_node_id_proto;
1855
extern const struct bpf_func_proto bpf_tail_call_proto;
1856
extern const struct bpf_func_proto bpf_ktime_get_ns_proto;
1857
extern const struct bpf_func_proto bpf_ktime_get_boot_ns_proto;
1858 1859 1860
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;
1861
extern const struct bpf_func_proto bpf_get_stackid_proto;
Y
Yonghong Song 已提交
1862
extern const struct bpf_func_proto bpf_get_stack_proto;
1863
extern const struct bpf_func_proto bpf_get_task_stack_proto;
1864 1865
extern const struct bpf_func_proto bpf_get_stackid_proto_pe;
extern const struct bpf_func_proto bpf_get_stack_proto_pe;
1866
extern const struct bpf_func_proto bpf_sock_map_update_proto;
1867
extern const struct bpf_func_proto bpf_sock_hash_update_proto;
1868
extern const struct bpf_func_proto bpf_get_current_cgroup_id_proto;
1869
extern const struct bpf_func_proto bpf_get_current_ancestor_cgroup_id_proto;
1870 1871 1872 1873
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;
1874 1875
extern const struct bpf_func_proto bpf_spin_lock_proto;
extern const struct bpf_func_proto bpf_spin_unlock_proto;
1876
extern const struct bpf_func_proto bpf_get_local_storage_proto;
1877 1878
extern const struct bpf_func_proto bpf_strtol_proto;
extern const struct bpf_func_proto bpf_strtoul_proto;
1879
extern const struct bpf_func_proto bpf_tcp_sock_proto;
M
Martin KaFai Lau 已提交
1880
extern const struct bpf_func_proto bpf_jiffies64_proto;
1881
extern const struct bpf_func_proto bpf_get_ns_current_pid_tgid_proto;
1882
extern const struct bpf_func_proto bpf_event_output_data_proto;
1883 1884 1885 1886 1887
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;
1888
extern const struct bpf_func_proto bpf_skc_to_tcp6_sock_proto;
1889 1890 1891
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;
1892
extern const struct bpf_func_proto bpf_skc_to_udp6_sock_proto;
1893
extern const struct bpf_func_proto bpf_copy_from_user_proto;
A
Alan Maguire 已提交
1894
extern const struct bpf_func_proto bpf_snprintf_btf_proto;
H
Hao Luo 已提交
1895
extern const struct bpf_func_proto bpf_per_cpu_ptr_proto;
H
Hao Luo 已提交
1896
extern const struct bpf_func_proto bpf_this_cpu_ptr_proto;
1897
extern const struct bpf_func_proto bpf_ktime_get_coarse_ns_proto;
1898
extern const struct bpf_func_proto bpf_sock_from_file_proto;
1899
extern const struct bpf_func_proto bpf_get_socket_ptr_cookie_proto;
1900 1901
extern const struct bpf_func_proto bpf_task_storage_get_proto;
extern const struct bpf_func_proto bpf_task_storage_delete_proto;
1902
extern const struct bpf_func_proto bpf_for_each_map_elem_proto;
1903

K
KP Singh 已提交
1904 1905 1906
const struct bpf_func_proto *bpf_tracing_func_proto(
	enum bpf_func_id func_id, const struct bpf_prog *prog);

1907 1908 1909
const struct bpf_func_proto *tracing_prog_func_proto(
  enum bpf_func_id func_id, const struct bpf_prog *prog);

1910 1911 1912
/* 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);
1913
u64 bpf_get_raw_cpu_id(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
1914

1915
#if defined(CONFIG_NET)
1916 1917 1918
bool bpf_sock_common_is_valid_access(int off, int size,
				     enum bpf_access_type type,
				     struct bpf_insn_access_aux *info);
1919 1920 1921 1922 1923 1924 1925 1926
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
1927 1928 1929 1930 1931 1932
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;
}
1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948
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

1949
#ifdef CONFIG_INET
A
Alexei Starovoitov 已提交
1950 1951 1952 1953 1954 1955 1956 1957 1958
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;
};
1959 1960 1961 1962 1963 1964 1965 1966
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);
1967 1968 1969 1970 1971 1972 1973 1974 1975

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);
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
#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;
}
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
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;
}
2007 2008
#endif /* CONFIG_INET */

2009
enum bpf_text_poke_type {
2010 2011
	BPF_MOD_CALL,
	BPF_MOD_JUMP,
2012
};
2013

2014 2015 2016
int bpf_arch_text_poke(void *ip, enum bpf_text_poke_type t,
		       void *addr1, void *addr2);

J
Jiri Olsa 已提交
2017
struct btf_id_set;
2018
bool btf_id_set_contains(const struct btf_id_set *set, u32 id);
J
Jiri Olsa 已提交
2019

2020
#endif /* _LINUX_BPF_H */