[linux-2.6-block.git] / include / linux / filter.h

/*
 * Linux Socket Filter Data Structures
 */
#ifndef __LINUX_FILTER_H__
#define __LINUX_FILTER_H__

#include <stdarg.h>

#include <linux/atomic.h>
#include <linux/compat.h>
#include <linux/skbuff.h>
#include <linux/linkage.h>
#include <linux/printk.h>
#include <linux/workqueue.h>
#include <linux/sched.h>
#include <net/sch_generic.h>

#include <asm/cacheflush.h>

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

struct sk_buff;
struct sock;
struct seccomp_data;
struct bpf_prog_aux;

/* ArgX, context and stack frame pointer register positions. Note,
 * Arg1, Arg2, Arg3, etc are used as argument mappings of function
 * calls in BPF_CALL instruction.
 */
#define BPF_REG_ARG1	BPF_REG_1
#define BPF_REG_ARG2	BPF_REG_2
#define BPF_REG_ARG3	BPF_REG_3
#define BPF_REG_ARG4	BPF_REG_4
#define BPF_REG_ARG5	BPF_REG_5
#define BPF_REG_CTX	BPF_REG_6
#define BPF_REG_FP	BPF_REG_10

/* Additional register mappings for converted user programs. */
#define BPF_REG_A	BPF_REG_0
#define BPF_REG_X	BPF_REG_7
#define BPF_REG_TMP	BPF_REG_8

/* BPF program can access up to 512 bytes of stack space. */
#define MAX_BPF_STACK	512

/* Helper macros for filter block array initializers. */

/* ALU ops on registers, bpf_add|sub|...: dst_reg += src_reg */

#define BPF_ALU64_REG(OP, DST, SRC)				\
	((struct bpf_insn) {					\
		.code  = BPF_ALU64 | BPF_OP(OP) | BPF_X,	\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = 0,					\
		.imm   = 0 })

#define BPF_ALU32_REG(OP, DST, SRC)				\
	((struct bpf_insn) {					\
		.code  = BPF_ALU | BPF_OP(OP) | BPF_X,		\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = 0,					\
		.imm   = 0 })

/* ALU ops on immediates, bpf_add|sub|...: dst_reg += imm32 */

#define BPF_ALU64_IMM(OP, DST, IMM)				\
	((struct bpf_insn) {					\
		.code  = BPF_ALU64 | BPF_OP(OP) | BPF_K,	\
		.dst_reg = DST,					\
		.src_reg = 0,					\
		.off   = 0,					\
		.imm   = IMM })

#define BPF_ALU32_IMM(OP, DST, IMM)				\
	((struct bpf_insn) {					\
		.code  = BPF_ALU | BPF_OP(OP) | BPF_K,		\
		.dst_reg = DST,					\
		.src_reg = 0,					\
		.off   = 0,					\
		.imm   = IMM })

/* Endianess conversion, cpu_to_{l,b}e(), {l,b}e_to_cpu() */

#define BPF_ENDIAN(TYPE, DST, LEN)				\
	((struct bpf_insn) {					\
		.code  = BPF_ALU | BPF_END | BPF_SRC(TYPE),	\
		.dst_reg = DST,					\
		.src_reg = 0,					\
		.off   = 0,					\
		.imm   = LEN })

/* Short form of mov, dst_reg = src_reg */

#define BPF_MOV64_REG(DST, SRC)					\
	((struct bpf_insn) {					\
		.code  = BPF_ALU64 | BPF_MOV | BPF_X,		\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = 0,					\
		.imm   = 0 })

#define BPF_MOV32_REG(DST, SRC)					\
	((struct bpf_insn) {					\
		.code  = BPF_ALU | BPF_MOV | BPF_X,		\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = 0,					\
		.imm   = 0 })

/* Short form of mov, dst_reg = imm32 */

#define BPF_MOV64_IMM(DST, IMM)					\
	((struct bpf_insn) {					\
		.code  = BPF_ALU64 | BPF_MOV | BPF_K,		\
		.dst_reg = DST,					\
		.src_reg = 0,					\
		.off   = 0,					\
		.imm   = IMM })

#define BPF_MOV32_IMM(DST, IMM)					\
	((struct bpf_insn) {					\
		.code  = BPF_ALU | BPF_MOV | BPF_K,		\
		.dst_reg = DST,					\
		.src_reg = 0,					\
		.off   = 0,					\
		.imm   = IMM })

/* BPF_LD_IMM64 macro encodes single 'load 64-bit immediate' insn */
#define BPF_LD_IMM64(DST, IMM)					\
	BPF_LD_IMM64_RAW(DST, 0, IMM)

#define BPF_LD_IMM64_RAW(DST, SRC, IMM)				\
	((struct bpf_insn) {					\
		.code  = BPF_LD | BPF_DW | BPF_IMM,		\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = 0,					\
		.imm   = (__u32) (IMM) }),			\
	((struct bpf_insn) {					\
		.code  = 0, /* zero is reserved opcode */	\
		.dst_reg = 0,					\
		.src_reg = 0,					\
		.off   = 0,					\
		.imm   = ((__u64) (IMM)) >> 32 })

/* pseudo BPF_LD_IMM64 insn used to refer to process-local map_fd */
#define BPF_LD_MAP_FD(DST, MAP_FD)				\
	BPF_LD_IMM64_RAW(DST, BPF_PSEUDO_MAP_FD, MAP_FD)

/* Short form of mov based on type, BPF_X: dst_reg = src_reg, BPF_K: dst_reg = imm32 */

#define BPF_MOV64_RAW(TYPE, DST, SRC, IMM)			\
	((struct bpf_insn) {					\
		.code  = BPF_ALU64 | BPF_MOV | BPF_SRC(TYPE),	\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = 0,					\
		.imm   = IMM })

#define BPF_MOV32_RAW(TYPE, DST, SRC, IMM)			\
	((struct bpf_insn) {					\
		.code  = BPF_ALU | BPF_MOV | BPF_SRC(TYPE),	\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = 0,					\
		.imm   = IMM })

/* Direct packet access, R0 = *(uint *) (skb->data + imm32) */

#define BPF_LD_ABS(SIZE, IMM)					\
	((struct bpf_insn) {					\
		.code  = BPF_LD | BPF_SIZE(SIZE) | BPF_ABS,	\
		.dst_reg = 0,					\
		.src_reg = 0,					\
		.off   = 0,					\
		.imm   = IMM })

/* Indirect packet access, R0 = *(uint *) (skb->data + src_reg + imm32) */

#define BPF_LD_IND(SIZE, SRC, IMM)				\
	((struct bpf_insn) {					\
		.code  = BPF_LD | BPF_SIZE(SIZE) | BPF_IND,	\
		.dst_reg = 0,					\
		.src_reg = SRC,					\
		.off   = 0,					\
		.imm   = IMM })

/* Memory load, dst_reg = *(uint *) (src_reg + off16) */

#define BPF_LDX_MEM(SIZE, DST, SRC, OFF)			\
	((struct bpf_insn) {					\
		.code  = BPF_LDX | BPF_SIZE(SIZE) | BPF_MEM,	\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = OFF,					\
		.imm   = 0 })

/* Memory store, *(uint *) (dst_reg + off16) = src_reg */

#define BPF_STX_MEM(SIZE, DST, SRC, OFF)			\
	((struct bpf_insn) {					\
		.code  = BPF_STX | BPF_SIZE(SIZE) | BPF_MEM,	\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = OFF,					\
		.imm   = 0 })

/* Atomic memory add, *(uint *)(dst_reg + off16) += src_reg */

#define BPF_STX_XADD(SIZE, DST, SRC, OFF)			\
	((struct bpf_insn) {					\
		.code  = BPF_STX | BPF_SIZE(SIZE) | BPF_XADD,	\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = OFF,					\
		.imm   = 0 })

/* Memory store, *(uint *) (dst_reg + off16) = imm32 */

#define BPF_ST_MEM(SIZE, DST, OFF, IMM)				\
	((struct bpf_insn) {					\
		.code  = BPF_ST | BPF_SIZE(SIZE) | BPF_MEM,	\
		.dst_reg = DST,					\
		.src_reg = 0,					\
		.off   = OFF,					\
		.imm   = IMM })

/* Conditional jumps against registers, if (dst_reg 'op' src_reg) goto pc + off16 */

#define BPF_JMP_REG(OP, DST, SRC, OFF)				\
	((struct bpf_insn) {					\
		.code  = BPF_JMP | BPF_OP(OP) | BPF_X,		\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = OFF,					\
		.imm   = 0 })

/* Conditional jumps against immediates, if (dst_reg 'op' imm32) goto pc + off16 */

#define BPF_JMP_IMM(OP, DST, IMM, OFF)				\
	((struct bpf_insn) {					\
		.code  = BPF_JMP | BPF_OP(OP) | BPF_K,		\
		.dst_reg = DST,					\
		.src_reg = 0,					\
		.off   = OFF,					\
		.imm   = IMM })

/* Function call */

#define BPF_EMIT_CALL(FUNC)					\
	((struct bpf_insn) {					\
		.code  = BPF_JMP | BPF_CALL,			\
		.dst_reg = 0,					\
		.src_reg = 0,					\
		.off   = 0,					\
		.imm   = ((FUNC) - __bpf_call_base) })

/* Raw code statement block */

#define BPF_RAW_INSN(CODE, DST, SRC, OFF, IMM)			\
	((struct bpf_insn) {					\
		.code  = CODE,					\
		.dst_reg = DST,					\
		.src_reg = SRC,					\
		.off   = OFF,					\
		.imm   = IMM })

/* Program exit */

#define BPF_EXIT_INSN()						\
	((struct bpf_insn) {					\
		.code  = BPF_JMP | BPF_EXIT,			\
		.dst_reg = 0,					\
		.src_reg = 0,					\
		.off   = 0,					\
		.imm   = 0 })

/* Internal classic blocks for direct assignment */

#define __BPF_STMT(CODE, K)					\
	((struct sock_filter) BPF_STMT(CODE, K))

#define __BPF_JUMP(CODE, K, JT, JF)				\
	((struct sock_filter) BPF_JUMP(CODE, K, JT, JF))

#define bytes_to_bpf_size(bytes)				\
({								\
	int bpf_size = -EINVAL;					\
								\
	if (bytes == sizeof(u8))				\
		bpf_size = BPF_B;				\
	else if (bytes == sizeof(u16))				\
		bpf_size = BPF_H;				\
	else if (bytes == sizeof(u32))				\
		bpf_size = BPF_W;				\
	else if (bytes == sizeof(u64))				\
		bpf_size = BPF_DW;				\
								\
	bpf_size;						\
})

#ifdef CONFIG_COMPAT
/* A struct sock_filter is architecture independent. */
struct compat_sock_fprog {
	u16		len;
	compat_uptr_t	filter;	/* struct sock_filter * */
};
#endif

struct sock_fprog_kern {
	u16			len;
	struct sock_filter	*filter;
};

struct bpf_binary_header {
	unsigned int pages;
	u8 image[];
};

struct bpf_prog {
	u16			pages;		/* Number of allocated pages */
	kmemcheck_bitfield_begin(meta);
	u16			jited:1,	/* Is our filter JIT'ed? */
				gpl_compatible:1, /* Is filter GPL compatible? */
				cb_access:1,	/* Is control block accessed? */
				dst_needed:1;	/* Do we need dst entry? */
	kmemcheck_bitfield_end(meta);
	u32			len;		/* Number of filter blocks */
	enum bpf_prog_type	type;		/* Type of BPF program */
	struct bpf_prog_aux	*aux;		/* Auxiliary fields */
	struct sock_fprog_kern	*orig_prog;	/* Original BPF program */
	unsigned int		(*bpf_func)(const struct sk_buff *skb,
					    const struct bpf_insn *filter);
	/* Instructions for interpreter */
	union {
		struct sock_filter	insns[0];
		struct bpf_insn		insnsi[0];
	};
};

struct sk_filter {
	atomic_t	refcnt;
	struct rcu_head	rcu;
	struct bpf_prog	*prog;
};

#define BPF_PROG_RUN(filter, ctx)  (*filter->bpf_func)(ctx, filter->insnsi)

static inline u32 bpf_prog_run_save_cb(const struct bpf_prog *prog,
				       struct sk_buff *skb)
{
	u8 *cb_data = qdisc_skb_cb(skb)->data;
	u8 saved_cb[QDISC_CB_PRIV_LEN];
	u32 res;

	BUILD_BUG_ON(FIELD_SIZEOF(struct __sk_buff, cb) !=
		     QDISC_CB_PRIV_LEN);

	if (unlikely(prog->cb_access)) {
		memcpy(saved_cb, cb_data, sizeof(saved_cb));
		memset(cb_data, 0, sizeof(saved_cb));
	}

	res = BPF_PROG_RUN(prog, skb);

	if (unlikely(prog->cb_access))
		memcpy(cb_data, saved_cb, sizeof(saved_cb));

	return res;
}

static inline u32 bpf_prog_run_clear_cb(const struct bpf_prog *prog,
					struct sk_buff *skb)
{
	u8 *cb_data = qdisc_skb_cb(skb)->data;

	if (unlikely(prog->cb_access))
		memset(cb_data, 0, QDISC_CB_PRIV_LEN);
	return BPF_PROG_RUN(prog, skb);
}

static inline unsigned int bpf_prog_size(unsigned int proglen)
{
	return max(sizeof(struct bpf_prog),
		   offsetof(struct bpf_prog, insns[proglen]));
}

static inline bool bpf_prog_was_classic(const struct bpf_prog *prog)
{
	/* When classic BPF programs have been loaded and the arch
	 * does not have a classic BPF JIT (anymore), they have been
	 * converted via bpf_migrate_filter() to eBPF and thus always
	 * have an unspec program type.
	 */
	return prog->type == BPF_PROG_TYPE_UNSPEC;
}

#define bpf_classic_proglen(fprog) (fprog->len * sizeof(fprog->filter[0]))

#ifdef CONFIG_DEBUG_SET_MODULE_RONX
static inline void bpf_prog_lock_ro(struct bpf_prog *fp)
{
	set_memory_ro((unsigned long)fp, fp->pages);
}

static inline void bpf_prog_unlock_ro(struct bpf_prog *fp)
{
	set_memory_rw((unsigned long)fp, fp->pages);
}
#else
static inline void bpf_prog_lock_ro(struct bpf_prog *fp)
{
}

static inline void bpf_prog_unlock_ro(struct bpf_prog *fp)
{
}
#endif /* CONFIG_DEBUG_SET_MODULE_RONX */

int sk_filter(struct sock *sk, struct sk_buff *skb);

int bpf_prog_select_runtime(struct bpf_prog *fp);
void bpf_prog_free(struct bpf_prog *fp);

struct bpf_prog *bpf_prog_alloc(unsigned int size, gfp_t gfp_extra_flags);
struct bpf_prog *bpf_prog_realloc(struct bpf_prog *fp_old, unsigned int size,
				  gfp_t gfp_extra_flags);
void __bpf_prog_free(struct bpf_prog *fp);

static inline void bpf_prog_unlock_free(struct bpf_prog *fp)
{
	bpf_prog_unlock_ro(fp);
	__bpf_prog_free(fp);
}

typedef int (*bpf_aux_classic_check_t)(struct sock_filter *filter,
				       unsigned int flen);

int bpf_prog_create(struct bpf_prog **pfp, struct sock_fprog_kern *fprog);
int bpf_prog_create_from_user(struct bpf_prog **pfp, struct sock_fprog *fprog,
			      bpf_aux_classic_check_t trans, bool save_orig);
void bpf_prog_destroy(struct bpf_prog *fp);

int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk);
int sk_attach_bpf(u32 ufd, struct sock *sk);
int sk_detach_filter(struct sock *sk);
int sk_get_filter(struct sock *sk, struct sock_filter __user *filter,
		  unsigned int len);

bool sk_filter_charge(struct sock *sk, struct sk_filter *fp);
void sk_filter_uncharge(struct sock *sk, struct sk_filter *fp);

u64 __bpf_call_base(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
void bpf_int_jit_compile(struct bpf_prog *fp);
bool bpf_helper_changes_skb_data(void *func);

#ifdef CONFIG_BPF_JIT
typedef void (*bpf_jit_fill_hole_t)(void *area, unsigned int size);

struct bpf_binary_header *
bpf_jit_binary_alloc(unsigned int proglen, u8 **image_ptr,
		     unsigned int alignment,
		     bpf_jit_fill_hole_t bpf_fill_ill_insns);
void bpf_jit_binary_free(struct bpf_binary_header *hdr);

void bpf_jit_compile(struct bpf_prog *fp);
void bpf_jit_free(struct bpf_prog *fp);

static inline void bpf_jit_dump(unsigned int flen, unsigned int proglen,
				u32 pass, void *image)
{
	pr_err("flen=%u proglen=%u pass=%u image=%pK from=%s pid=%d\n", flen,
	       proglen, pass, image, current->comm, task_pid_nr(current));

	if (image)
		print_hex_dump(KERN_ERR, "JIT code: ", DUMP_PREFIX_OFFSET,
			       16, 1, image, proglen, false);
}
#else
static inline void bpf_jit_compile(struct bpf_prog *fp)
{
}

static inline void bpf_jit_free(struct bpf_prog *fp)
{
	bpf_prog_unlock_free(fp);
}
#endif /* CONFIG_BPF_JIT */

#define BPF_ANC		BIT(15)

static inline u16 bpf_anc_helper(const struct sock_filter *ftest)
{
	BUG_ON(ftest->code & BPF_ANC);

	switch (ftest->code) {
	case BPF_LD | BPF_W | BPF_ABS:
	case BPF_LD | BPF_H | BPF_ABS:
	case BPF_LD | BPF_B | BPF_ABS:
#define BPF_ANCILLARY(CODE)	case SKF_AD_OFF + SKF_AD_##CODE:	\
				return BPF_ANC | SKF_AD_##CODE
		switch (ftest->k) {
		BPF_ANCILLARY(PROTOCOL);
		BPF_ANCILLARY(PKTTYPE);
		BPF_ANCILLARY(IFINDEX);
		BPF_ANCILLARY(NLATTR);
		BPF_ANCILLARY(NLATTR_NEST);
		BPF_ANCILLARY(MARK);
		BPF_ANCILLARY(QUEUE);
		BPF_ANCILLARY(HATYPE);
		BPF_ANCILLARY(RXHASH);
		BPF_ANCILLARY(CPU);
		BPF_ANCILLARY(ALU_XOR_X);
		BPF_ANCILLARY(VLAN_TAG);
		BPF_ANCILLARY(VLAN_TAG_PRESENT);
		BPF_ANCILLARY(PAY_OFFSET);
		BPF_ANCILLARY(RANDOM);
		BPF_ANCILLARY(VLAN_TPID);
		}
		/* Fallthrough. */
	default:
		return ftest->code;
	}
}

void *bpf_internal_load_pointer_neg_helper(const struct sk_buff *skb,
					   int k, unsigned int size);

static inline void *bpf_load_pointer(const struct sk_buff *skb, int k,
				     unsigned int size, void *buffer)
{
	if (k >= 0)
		return skb_header_pointer(skb, k, size, buffer);

	return bpf_internal_load_pointer_neg_helper(skb, k, size);
}

static inline int bpf_tell_extensions(void)
{
	return SKF_AD_MAX;
}

#endif /* __LINUX_FILTER_H__ */
Commit	Line	Data
1da177e4 LT	1	/*
	2	* Linux Socket Filter Data Structures
	3	*/
1da177e4 LT	4	#ifndef __LINUX_FILTER_H__
	5	#define __LINUX_FILTER_H__
	6
b954d834 DB	7	#include <stdarg.h>
b954d834 DB	8
60063497	9	#include <linux/atomic.h>
0c5fe1b4	10	#include <linux/compat.h>
9f12fbe6	11	#include <linux/skbuff.h>
b954d834 DB	12	#include <linux/linkage.h>
b954d834 DB	13	#include <linux/printk.h>
d45ed4a4	14	#include <linux/workqueue.h>
b13138ef	15	#include <linux/sched.h>
ff936a04	16	#include <net/sch_generic.h>
b954d834	17
60a3b225	18	#include <asm/cacheflush.h>
b954d834 DB	19
b954d834 DB	20	#include <uapi/linux/filter.h>
daedfb22	21	#include <uapi/linux/bpf.h>
60a3b225 DB	22
	23	struct sk_buff;
	24	struct sock;
	25	struct seccomp_data;
09756af4	26	struct bpf_prog_aux;
792d4b5c	27
30743837 DB	28	/* ArgX, context and stack frame pointer register positions. Note,
	29	* Arg1, Arg2, Arg3, etc are used as argument mappings of function
	30	* calls in BPF_CALL instruction.
	31	*/
	32	#define BPF_REG_ARG1 BPF_REG_1
	33	#define BPF_REG_ARG2 BPF_REG_2
	34	#define BPF_REG_ARG3 BPF_REG_3
	35	#define BPF_REG_ARG4 BPF_REG_4
	36	#define BPF_REG_ARG5 BPF_REG_5
	37	#define BPF_REG_CTX BPF_REG_6
	38	#define BPF_REG_FP BPF_REG_10
	39
	40	/* Additional register mappings for converted user programs. */
	41	#define BPF_REG_A BPF_REG_0
	42	#define BPF_REG_X BPF_REG_7
	43	#define BPF_REG_TMP BPF_REG_8
bd4cf0ed AS	44
	45	/* BPF program can access up to 512 bytes of stack space. */
	46	#define MAX_BPF_STACK 512
	47
f8f6d679 DB	48	/* Helper macros for filter block array initializers. */
f8f6d679 DB	49
e430f34e	50	/* ALU ops on registers, bpf_add\|sub\|...: dst_reg += src_reg */
f8f6d679	51
e430f34e	52	#define BPF_ALU64_REG(OP, DST, SRC) \
2695fb55	53	((struct bpf_insn) { \
f8f6d679	54	.code = BPF_ALU64 \| BPF_OP(OP) \| BPF_X, \
e430f34e AS	55	.dst_reg = DST, \
e430f34e AS	56	.src_reg = SRC, \
f8f6d679 DB	57	.off = 0, \
	58	.imm = 0 })
	59
e430f34e	60	#define BPF_ALU32_REG(OP, DST, SRC) \
2695fb55	61	((struct bpf_insn) { \
f8f6d679	62	.code = BPF_ALU \| BPF_OP(OP) \| BPF_X, \
e430f34e AS	63	.dst_reg = DST, \
e430f34e AS	64	.src_reg = SRC, \
f8f6d679 DB	65	.off = 0, \
	66	.imm = 0 })
	67
e430f34e	68	/* ALU ops on immediates, bpf_add\|sub\|...: dst_reg += imm32 */
f8f6d679	69
e430f34e	70	#define BPF_ALU64_IMM(OP, DST, IMM) \
2695fb55	71	((struct bpf_insn) { \
f8f6d679	72	.code = BPF_ALU64 \| BPF_OP(OP) \| BPF_K, \
e430f34e AS	73	.dst_reg = DST, \
e430f34e AS	74	.src_reg = 0, \
f8f6d679 DB	75	.off = 0, \
	76	.imm = IMM })
	77
e430f34e	78	#define BPF_ALU32_IMM(OP, DST, IMM) \
2695fb55	79	((struct bpf_insn) { \
f8f6d679	80	.code = BPF_ALU \| BPF_OP(OP) \| BPF_K, \
e430f34e AS	81	.dst_reg = DST, \
e430f34e AS	82	.src_reg = 0, \
f8f6d679 DB	83	.off = 0, \
	84	.imm = IMM })
	85
	86	/* Endianess conversion, cpu_to_{l,b}e(), {l,b}e_to_cpu() */
	87
e430f34e	88	#define BPF_ENDIAN(TYPE, DST, LEN) \
2695fb55	89	((struct bpf_insn) { \
f8f6d679	90	.code = BPF_ALU \| BPF_END \| BPF_SRC(TYPE), \
e430f34e AS	91	.dst_reg = DST, \
e430f34e AS	92	.src_reg = 0, \
f8f6d679 DB	93	.off = 0, \
	94	.imm = LEN })
	95
e430f34e	96	/* Short form of mov, dst_reg = src_reg */
f8f6d679	97
e430f34e	98	#define BPF_MOV64_REG(DST, SRC) \
2695fb55	99	((struct bpf_insn) { \
f8f6d679	100	.code = BPF_ALU64 \| BPF_MOV \| BPF_X, \
e430f34e AS	101	.dst_reg = DST, \
e430f34e AS	102	.src_reg = SRC, \
f8f6d679 DB	103	.off = 0, \
	104	.imm = 0 })
	105
e430f34e	106	#define BPF_MOV32_REG(DST, SRC) \
2695fb55	107	((struct bpf_insn) { \
f8f6d679	108	.code = BPF_ALU \| BPF_MOV \| BPF_X, \
e430f34e AS	109	.dst_reg = DST, \
e430f34e AS	110	.src_reg = SRC, \
f8f6d679 DB	111	.off = 0, \
	112	.imm = 0 })
	113
e430f34e	114	/* Short form of mov, dst_reg = imm32 */
f8f6d679	115
e430f34e	116	#define BPF_MOV64_IMM(DST, IMM) \
2695fb55	117	((struct bpf_insn) { \
f8f6d679	118	.code = BPF_ALU64 \| BPF_MOV \| BPF_K, \
e430f34e AS	119	.dst_reg = DST, \
e430f34e AS	120	.src_reg = 0, \
f8f6d679 DB	121	.off = 0, \
	122	.imm = IMM })
	123
e430f34e	124	#define BPF_MOV32_IMM(DST, IMM) \
2695fb55	125	((struct bpf_insn) { \
f8f6d679	126	.code = BPF_ALU \| BPF_MOV \| BPF_K, \
e430f34e AS	127	.dst_reg = DST, \
e430f34e AS	128	.src_reg = 0, \
f8f6d679 DB	129	.off = 0, \
	130	.imm = IMM })
	131
02ab695b AS	132	/* BPF_LD_IMM64 macro encodes single 'load 64-bit immediate' insn */
	133	#define BPF_LD_IMM64(DST, IMM) \
	134	BPF_LD_IMM64_RAW(DST, 0, IMM)
	135
	136	#define BPF_LD_IMM64_RAW(DST, SRC, IMM) \
	137	((struct bpf_insn) { \
	138	.code = BPF_LD \| BPF_DW \| BPF_IMM, \
	139	.dst_reg = DST, \
	140	.src_reg = SRC, \
	141	.off = 0, \
	142	.imm = (__u32) (IMM) }), \
	143	((struct bpf_insn) { \
	144	.code = 0, /* zero is reserved opcode */ \
	145	.dst_reg = 0, \
	146	.src_reg = 0, \
	147	.off = 0, \
	148	.imm = ((__u64) (IMM)) >> 32 })
	149
0246e64d AS	150	/* pseudo BPF_LD_IMM64 insn used to refer to process-local map_fd */
	151	#define BPF_LD_MAP_FD(DST, MAP_FD) \
	152	BPF_LD_IMM64_RAW(DST, BPF_PSEUDO_MAP_FD, MAP_FD)
	153
e430f34e	154	/* Short form of mov based on type, BPF_X: dst_reg = src_reg, BPF_K: dst_reg = imm32 */
f8f6d679	155
e430f34e	156	#define BPF_MOV64_RAW(TYPE, DST, SRC, IMM) \
2695fb55	157	((struct bpf_insn) { \
f8f6d679	158	.code = BPF_ALU64 \| BPF_MOV \| BPF_SRC(TYPE), \
e430f34e AS	159	.dst_reg = DST, \
e430f34e AS	160	.src_reg = SRC, \
f8f6d679 DB	161	.off = 0, \
	162	.imm = IMM })
	163
e430f34e	164	#define BPF_MOV32_RAW(TYPE, DST, SRC, IMM) \
2695fb55	165	((struct bpf_insn) { \
f8f6d679	166	.code = BPF_ALU \| BPF_MOV \| BPF_SRC(TYPE), \
e430f34e AS	167	.dst_reg = DST, \
e430f34e AS	168	.src_reg = SRC, \
f8f6d679 DB	169	.off = 0, \
	170	.imm = IMM })
	171
e430f34e	172	/* Direct packet access, R0 = (uint ) (skb->data + imm32) */
f8f6d679	173
e430f34e	174	#define BPF_LD_ABS(SIZE, IMM) \
2695fb55	175	((struct bpf_insn) { \
f8f6d679	176	.code = BPF_LD \| BPF_SIZE(SIZE) \| BPF_ABS, \
e430f34e AS	177	.dst_reg = 0, \
e430f34e AS	178	.src_reg = 0, \
f8f6d679	179	.off = 0, \
e430f34e	180	.imm = IMM })
f8f6d679	181
e430f34e	182	/* Indirect packet access, R0 = (uint ) (skb->data + src_reg + imm32) */
f8f6d679	183
e430f34e	184	#define BPF_LD_IND(SIZE, SRC, IMM) \
2695fb55	185	((struct bpf_insn) { \
f8f6d679	186	.code = BPF_LD \| BPF_SIZE(SIZE) \| BPF_IND, \
e430f34e AS	187	.dst_reg = 0, \
e430f34e AS	188	.src_reg = SRC, \
f8f6d679	189	.off = 0, \
e430f34e	190	.imm = IMM })
f8f6d679	191
e430f34e	192	/* Memory load, dst_reg = (uint ) (src_reg + off16) */
f8f6d679	193
e430f34e	194	#define BPF_LDX_MEM(SIZE, DST, SRC, OFF) \
2695fb55	195	((struct bpf_insn) { \
f8f6d679	196	.code = BPF_LDX \| BPF_SIZE(SIZE) \| BPF_MEM, \
e430f34e AS	197	.dst_reg = DST, \
e430f34e AS	198	.src_reg = SRC, \
f8f6d679 DB	199	.off = OFF, \
	200	.imm = 0 })
	201
e430f34e AS	202	/* Memory store, (uint ) (dst_reg + off16) = src_reg */
	203
	204	#define BPF_STX_MEM(SIZE, DST, SRC, OFF) \
2695fb55	205	((struct bpf_insn) { \
f8f6d679	206	.code = BPF_STX \| BPF_SIZE(SIZE) \| BPF_MEM, \
e430f34e AS	207	.dst_reg = DST, \
e430f34e AS	208	.src_reg = SRC, \
f8f6d679 DB	209	.off = OFF, \
	210	.imm = 0 })
	211
cffc642d MH	212	/* Atomic memory add, (uint )(dst_reg + off16) += src_reg */
	213
	214	#define BPF_STX_XADD(SIZE, DST, SRC, OFF) \
	215	((struct bpf_insn) { \
	216	.code = BPF_STX \| BPF_SIZE(SIZE) \| BPF_XADD, \
	217	.dst_reg = DST, \
	218	.src_reg = SRC, \
	219	.off = OFF, \
	220	.imm = 0 })
	221
e430f34e AS	222	/* Memory store, (uint ) (dst_reg + off16) = imm32 */
	223
	224	#define BPF_ST_MEM(SIZE, DST, OFF, IMM) \
2695fb55	225	((struct bpf_insn) { \
e430f34e AS	226	.code = BPF_ST \| BPF_SIZE(SIZE) \| BPF_MEM, \
	227	.dst_reg = DST, \
	228	.src_reg = 0, \
	229	.off = OFF, \
	230	.imm = IMM })
	231
	232	/* Conditional jumps against registers, if (dst_reg 'op' src_reg) goto pc + off16 */
f8f6d679	233
e430f34e	234	#define BPF_JMP_REG(OP, DST, SRC, OFF) \
2695fb55	235	((struct bpf_insn) { \
f8f6d679	236	.code = BPF_JMP \| BPF_OP(OP) \| BPF_X, \
e430f34e AS	237	.dst_reg = DST, \
e430f34e AS	238	.src_reg = SRC, \
f8f6d679 DB	239	.off = OFF, \
	240	.imm = 0 })
	241
e430f34e	242	/* Conditional jumps against immediates, if (dst_reg 'op' imm32) goto pc + off16 */
f8f6d679	243
e430f34e	244	#define BPF_JMP_IMM(OP, DST, IMM, OFF) \
2695fb55	245	((struct bpf_insn) { \
f8f6d679	246	.code = BPF_JMP \| BPF_OP(OP) \| BPF_K, \
e430f34e AS	247	.dst_reg = DST, \
e430f34e AS	248	.src_reg = 0, \
f8f6d679 DB	249	.off = OFF, \
	250	.imm = IMM })
	251
	252	/* Function call */
	253
	254	#define BPF_EMIT_CALL(FUNC) \
2695fb55	255	((struct bpf_insn) { \
f8f6d679	256	.code = BPF_JMP \| BPF_CALL, \
e430f34e AS	257	.dst_reg = 0, \
e430f34e AS	258	.src_reg = 0, \
f8f6d679 DB	259	.off = 0, \
	260	.imm = ((FUNC) - __bpf_call_base) })
	261
	262	/* Raw code statement block */
	263
e430f34e	264	#define BPF_RAW_INSN(CODE, DST, SRC, OFF, IMM) \
2695fb55	265	((struct bpf_insn) { \
f8f6d679	266	.code = CODE, \
e430f34e AS	267	.dst_reg = DST, \
e430f34e AS	268	.src_reg = SRC, \
f8f6d679 DB	269	.off = OFF, \
	270	.imm = IMM })
	271
	272	/* Program exit */
	273
	274	#define BPF_EXIT_INSN() \
2695fb55	275	((struct bpf_insn) { \
f8f6d679	276	.code = BPF_JMP \| BPF_EXIT, \
e430f34e AS	277	.dst_reg = 0, \
e430f34e AS	278	.src_reg = 0, \
f8f6d679 DB	279	.off = 0, \
	280	.imm = 0 })
	281
a4afd37b DB	282	/* Internal classic blocks for direct assignment */
	283
	284	#define __BPF_STMT(CODE, K) \
	285	((struct sock_filter) BPF_STMT(CODE, K))
	286
	287	#define __BPF_JUMP(CODE, K, JT, JF) \
	288	((struct sock_filter) BPF_JUMP(CODE, K, JT, JF))
	289
f8f6d679 DB	290	#define bytes_to_bpf_size(bytes) \
	291	({ \
	292	int bpf_size = -EINVAL; \
	293	\
	294	if (bytes == sizeof(u8)) \
	295	bpf_size = BPF_B; \
	296	else if (bytes == sizeof(u16)) \
	297	bpf_size = BPF_H; \
	298	else if (bytes == sizeof(u32)) \
	299	bpf_size = BPF_W; \
	300	else if (bytes == sizeof(u64)) \
	301	bpf_size = BPF_DW; \
	302	\
	303	bpf_size; \
	304	})
9739eef1	305
bd4cf0ed AS	306	#ifdef CONFIG_COMPAT
bd4cf0ed AS	307	/* A struct sock_filter is architecture independent. */
0c5fe1b4 WD	308	struct compat_sock_fprog {
0c5fe1b4 WD	309	u16 len;
bd4cf0ed	310	compat_uptr_t filter; /* struct sock_filter * */
0c5fe1b4 WD	311	};
	312	#endif
	313
a3ea269b DB	314	struct sock_fprog_kern {
	315	u16 len;
	316	struct sock_filter *filter;
	317	};
	318
738cbe72 DB	319	struct bpf_binary_header {
	320	unsigned int pages;
	321	u8 image[];
	322	};
	323
7ae457c1	324	struct bpf_prog {
286aad3c	325	u16 pages; /* Number of allocated pages */
a91263d5 DB	326	kmemcheck_bitfield_begin(meta);
a91263d5 DB	327	u16 jited:1, /* Is our filter JIT'ed? */
c46646d0	328	gpl_compatible:1, /* Is filter GPL compatible? */
ff936a04	329	cb_access:1, /* Is control block accessed? */
c46646d0	330	dst_needed:1; /* Do we need dst entry? */
a91263d5	331	kmemcheck_bitfield_end(meta);
286aad3c	332	u32 len; /* Number of filter blocks */
24701ece	333	enum bpf_prog_type type; /* Type of BPF program */
09756af4	334	struct bpf_prog_aux aux; / Auxiliary fields */
24701ece	335	struct sock_fprog_kern orig_prog; / Original BPF program */
0a14842f	336	unsigned int (bpf_func)(const struct sk_buff skb,
2695fb55	337	const struct bpf_insn *filter);
60a3b225	338	/* Instructions for interpreter */
d45ed4a4	339	union {
bd4cf0ed	340	struct sock_filter insns[0];
2695fb55	341	struct bpf_insn insnsi[0];
d45ed4a4	342	};
b715631f SH	343	};
b715631f SH	344
7ae457c1 AS	345	struct sk_filter {
	346	atomic_t refcnt;
	347	struct rcu_head rcu;
	348	struct bpf_prog *prog;
	349	};
	350
	351	#define BPF_PROG_RUN(filter, ctx) (*filter->bpf_func)(ctx, filter->insnsi)
	352
ff936a04 AS	353	static inline u32 bpf_prog_run_save_cb(const struct bpf_prog *prog,
	354	struct sk_buff *skb)
	355	{
	356	u8 *cb_data = qdisc_skb_cb(skb)->data;
	357	u8 saved_cb[QDISC_CB_PRIV_LEN];
	358	u32 res;
	359
	360	BUILD_BUG_ON(FIELD_SIZEOF(struct __sk_buff, cb) !=
	361	QDISC_CB_PRIV_LEN);
	362
	363	if (unlikely(prog->cb_access)) {
	364	memcpy(saved_cb, cb_data, sizeof(saved_cb));
	365	memset(cb_data, 0, sizeof(saved_cb));
	366	}
	367
	368	res = BPF_PROG_RUN(prog, skb);
	369
	370	if (unlikely(prog->cb_access))
	371	memcpy(cb_data, saved_cb, sizeof(saved_cb));
	372
	373	return res;
	374	}
	375
	376	static inline u32 bpf_prog_run_clear_cb(const struct bpf_prog *prog,
	377	struct sk_buff *skb)
	378	{
	379	u8 *cb_data = qdisc_skb_cb(skb)->data;
	380
	381	if (unlikely(prog->cb_access))
	382	memset(cb_data, 0, QDISC_CB_PRIV_LEN);
	383	return BPF_PROG_RUN(prog, skb);
	384	}
	385
7ae457c1	386	static inline unsigned int bpf_prog_size(unsigned int proglen)
b715631f	387	{
7ae457c1 AS	388	return max(sizeof(struct bpf_prog),
7ae457c1 AS	389	offsetof(struct bpf_prog, insns[proglen]));
b715631f SH	390	}
b715631f SH	391
7b36f929 DB	392	static inline bool bpf_prog_was_classic(const struct bpf_prog *prog)
	393	{
	394	/* When classic BPF programs have been loaded and the arch
	395	* does not have a classic BPF JIT (anymore), they have been
	396	* converted via bpf_migrate_filter() to eBPF and thus always
	397	* have an unspec program type.
	398	*/
	399	return prog->type == BPF_PROG_TYPE_UNSPEC;
	400	}
	401
009937e7	402	#define bpf_classic_proglen(fprog) (fprog->len * sizeof(fprog->filter[0]))
a3ea269b	403
60a3b225 DB	404	#ifdef CONFIG_DEBUG_SET_MODULE_RONX
	405	static inline void bpf_prog_lock_ro(struct bpf_prog *fp)
	406	{
	407	set_memory_ro((unsigned long)fp, fp->pages);
	408	}
	409
	410	static inline void bpf_prog_unlock_ro(struct bpf_prog *fp)
	411	{
	412	set_memory_rw((unsigned long)fp, fp->pages);
	413	}
	414	#else
	415	static inline void bpf_prog_lock_ro(struct bpf_prog *fp)
	416	{
	417	}
	418
	419	static inline void bpf_prog_unlock_ro(struct bpf_prog *fp)
	420	{
	421	}
	422	#endif /* CONFIG_DEBUG_SET_MODULE_RONX */
	423
fbc907f0	424	int sk_filter(struct sock sk, struct sk_buff skb);
bd4cf0ed	425
04fd61ab	426	int bpf_prog_select_runtime(struct bpf_prog *fp);
7ae457c1	427	void bpf_prog_free(struct bpf_prog *fp);
bd4cf0ed	428
60a3b225 DB	429	struct bpf_prog *bpf_prog_alloc(unsigned int size, gfp_t gfp_extra_flags);
	430	struct bpf_prog bpf_prog_realloc(struct bpf_prog fp_old, unsigned int size,
	431	gfp_t gfp_extra_flags);
	432	void __bpf_prog_free(struct bpf_prog *fp);
	433
	434	static inline void bpf_prog_unlock_free(struct bpf_prog *fp)
	435	{
	436	bpf_prog_unlock_ro(fp);
	437	__bpf_prog_free(fp);
	438	}
	439
ac67eb2c DB	440	typedef int (bpf_aux_classic_check_t)(struct sock_filter filter,
	441	unsigned int flen);
	442
7ae457c1	443	int bpf_prog_create(struct bpf_prog *pfp, struct sock_fprog_kern fprog);
ac67eb2c	444	int bpf_prog_create_from_user(struct bpf_prog *pfp, struct sock_fprog fprog,
bab18991	445	bpf_aux_classic_check_t trans, bool save_orig);
7ae457c1	446	void bpf_prog_destroy(struct bpf_prog *fp);
a3ea269b	447
fbc907f0	448	int sk_attach_filter(struct sock_fprog fprog, struct sock sk);
89aa0758	449	int sk_attach_bpf(u32 ufd, struct sock *sk);
fbc907f0	450	int sk_detach_filter(struct sock *sk);
fbc907f0 DB	451	int sk_get_filter(struct sock sk, struct sock_filter __user filter,
fbc907f0 DB	452	unsigned int len);
fbc907f0	453
278571ba	454	bool sk_filter_charge(struct sock sk, struct sk_filter fp);
fbc907f0	455	void sk_filter_uncharge(struct sock sk, struct sk_filter fp);
0a14842f	456
62258278	457	u64 __bpf_call_base(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
7ae457c1	458	void bpf_int_jit_compile(struct bpf_prog *fp);
4e10df9a	459	bool bpf_helper_changes_skb_data(void *func);
62258278	460
b954d834 DB	461	#ifdef CONFIG_BPF_JIT
	462	typedef void (bpf_jit_fill_hole_t)(void area, unsigned int size);
	463
	464	struct bpf_binary_header *
	465	bpf_jit_binary_alloc(unsigned int proglen, u8 **image_ptr,
	466	unsigned int alignment,
	467	bpf_jit_fill_hole_t bpf_fill_ill_insns);
	468	void bpf_jit_binary_free(struct bpf_binary_header *hdr);
	469
	470	void bpf_jit_compile(struct bpf_prog *fp);
	471	void bpf_jit_free(struct bpf_prog *fp);
	472
	473	static inline void bpf_jit_dump(unsigned int flen, unsigned int proglen,
	474	u32 pass, void *image)
	475	{
b13138ef DB	476	pr_err("flen=%u proglen=%u pass=%u image=%pK from=%s pid=%d\n", flen,
	477	proglen, pass, image, current->comm, task_pid_nr(current));
	478
b954d834 DB	479	if (image)
	480	print_hex_dump(KERN_ERR, "JIT code: ", DUMP_PREFIX_OFFSET,
	481	16, 1, image, proglen, false);
	482	}
	483	#else
	484	static inline void bpf_jit_compile(struct bpf_prog *fp)
	485	{
	486	}
	487
	488	static inline void bpf_jit_free(struct bpf_prog *fp)
	489	{
	490	bpf_prog_unlock_free(fp);
	491	}
	492	#endif /* CONFIG_BPF_JIT */
	493
34805931 DB	494	#define BPF_ANC BIT(15)
	495
	496	static inline u16 bpf_anc_helper(const struct sock_filter *ftest)
	497	{
	498	BUG_ON(ftest->code & BPF_ANC);
	499
	500	switch (ftest->code) {
	501	case BPF_LD \| BPF_W \| BPF_ABS:
	502	case BPF_LD \| BPF_H \| BPF_ABS:
	503	case BPF_LD \| BPF_B \| BPF_ABS:
	504	#define BPF_ANCILLARY(CODE) case SKF_AD_OFF + SKF_AD_##CODE: \
	505	return BPF_ANC \| SKF_AD_##CODE
	506	switch (ftest->k) {
	507	BPF_ANCILLARY(PROTOCOL);
	508	BPF_ANCILLARY(PKTTYPE);
	509	BPF_ANCILLARY(IFINDEX);
	510	BPF_ANCILLARY(NLATTR);
	511	BPF_ANCILLARY(NLATTR_NEST);
	512	BPF_ANCILLARY(MARK);
	513	BPF_ANCILLARY(QUEUE);
	514	BPF_ANCILLARY(HATYPE);
	515	BPF_ANCILLARY(RXHASH);
	516	BPF_ANCILLARY(CPU);
	517	BPF_ANCILLARY(ALU_XOR_X);
	518	BPF_ANCILLARY(VLAN_TAG);
	519	BPF_ANCILLARY(VLAN_TAG_PRESENT);
	520	BPF_ANCILLARY(PAY_OFFSET);
	521	BPF_ANCILLARY(RANDOM);
27cd5452	522	BPF_ANCILLARY(VLAN_TPID);
34805931 DB	523	}
	524	/* Fallthrough. */
	525	default:
	526	return ftest->code;
	527	}
	528	}
	529
9f12fbe6 ZSL	530	void bpf_internal_load_pointer_neg_helper(const struct sk_buff skb,
	531	int k, unsigned int size);
	532
	533	static inline void bpf_load_pointer(const struct sk_buff skb, int k,
	534	unsigned int size, void *buffer)
	535	{
	536	if (k >= 0)
	537	return skb_header_pointer(skb, k, size, buffer);
	538
	539	return bpf_internal_load_pointer_neg_helper(skb, k, size);
	540	}
	541
ea02f941 MS	542	static inline int bpf_tell_extensions(void)
ea02f941 MS	543	{
37692299	544	return SKF_AD_MAX;
ea02f941 MS	545	}
ea02f941 MS	546
1da177e4	547	#endif /* __LINUX_FILTER_H__ */