[mirror_ubuntu-artful-kernel.git] / include / uapi / linux / bpf.h

/* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of version 2 of the GNU General Public
 * License as published by the Free Software Foundation.
 */
#ifndef _UAPI__LINUX_BPF_H__
#define _UAPI__LINUX_BPF_H__

#include <linux/types.h>
#include <linux/bpf_common.h>

/* Extended instruction set based on top of classic BPF */

/* instruction classes */
#define BPF_ALU64	0x07	/* alu mode in double word width */

/* ld/ldx fields */
#define BPF_DW		0x18	/* double word */
#define BPF_XADD	0xc0	/* exclusive add */

/* alu/jmp fields */
#define BPF_MOV		0xb0	/* mov reg to reg */
#define BPF_ARSH	0xc0	/* sign extending arithmetic shift right */

/* change endianness of a register */
#define BPF_END		0xd0	/* flags for endianness conversion: */
#define BPF_TO_LE	0x00	/* convert to little-endian */
#define BPF_TO_BE	0x08	/* convert to big-endian */
#define BPF_FROM_LE	BPF_TO_LE
#define BPF_FROM_BE	BPF_TO_BE

#define BPF_JNE		0x50	/* jump != */
#define BPF_JSGT	0x60	/* SGT is signed '>', GT in x86 */
#define BPF_JSGE	0x70	/* SGE is signed '>=', GE in x86 */
#define BPF_CALL	0x80	/* function call */
#define BPF_EXIT	0x90	/* function return */

/* Register numbers */
enum {
	BPF_REG_0 = 0,
	BPF_REG_1,
	BPF_REG_2,
	BPF_REG_3,
	BPF_REG_4,
	BPF_REG_5,
	BPF_REG_6,
	BPF_REG_7,
	BPF_REG_8,
	BPF_REG_9,
	BPF_REG_10,
	__MAX_BPF_REG,
};

/* BPF has 10 general purpose 64-bit registers and stack frame. */
#define MAX_BPF_REG	__MAX_BPF_REG

struct bpf_insn {
	__u8	code;		/* opcode */
	__u8	dst_reg:4;	/* dest register */
	__u8	src_reg:4;	/* source register */
	__s16	off;		/* signed offset */
	__s32	imm;		/* signed immediate constant */
};

/* BPF syscall commands, see bpf(2) man-page for details. */
enum bpf_cmd {
	BPF_MAP_CREATE,
	BPF_MAP_LOOKUP_ELEM,
	BPF_MAP_UPDATE_ELEM,
	BPF_MAP_DELETE_ELEM,
	BPF_MAP_GET_NEXT_KEY,
	BPF_PROG_LOAD,
	BPF_OBJ_PIN,
	BPF_OBJ_GET,
};

enum bpf_map_type {
	BPF_MAP_TYPE_UNSPEC,
	BPF_MAP_TYPE_HASH,
	BPF_MAP_TYPE_ARRAY,
	BPF_MAP_TYPE_PROG_ARRAY,
	BPF_MAP_TYPE_PERF_EVENT_ARRAY,
	BPF_MAP_TYPE_PERCPU_HASH,
	BPF_MAP_TYPE_PERCPU_ARRAY,
	BPF_MAP_TYPE_STACK_TRACE,
	BPF_MAP_TYPE_CGROUP_ARRAY,
	BPF_MAP_TYPE_LRU_HASH,
	BPF_MAP_TYPE_LRU_PERCPU_HASH,
};

enum bpf_prog_type {
	BPF_PROG_TYPE_UNSPEC,
	BPF_PROG_TYPE_SOCKET_FILTER,
	BPF_PROG_TYPE_KPROBE,
	BPF_PROG_TYPE_SCHED_CLS,
	BPF_PROG_TYPE_SCHED_ACT,
	BPF_PROG_TYPE_TRACEPOINT,
	BPF_PROG_TYPE_XDP,
	BPF_PROG_TYPE_PERF_EVENT,
	BPF_PROG_TYPE_CGROUP_SKB,
};

enum bpf_attach_type {
	BPF_CGROUP_INET_INGRESS,
	BPF_CGROUP_INET_EGRESS,
	__MAX_BPF_ATTACH_TYPE
};

#define MAX_BPF_ATTACH_TYPE __MAX_BPF_ATTACH_TYPE

#define BPF_PSEUDO_MAP_FD	1

/* flags for BPF_MAP_UPDATE_ELEM command */
#define BPF_ANY		0 /* create new element or update existing */
#define BPF_NOEXIST	1 /* create new element if it didn't exist */
#define BPF_EXIST	2 /* update existing element */

#define BPF_F_NO_PREALLOC	(1U << 0)
/* Instead of having one common LRU list in the
 * BPF_MAP_TYPE_LRU_[PERCPU_]HASH map, use a percpu LRU list
 * which can scale and perform better.
 * Note, the LRU nodes (including free nodes) cannot be moved
 * across different LRU lists.
 */
#define BPF_F_NO_COMMON_LRU	(1U << 1)

union bpf_attr {
	struct { /* anonymous struct used by BPF_MAP_CREATE command */
		__u32	map_type;	/* one of enum bpf_map_type */
		__u32	key_size;	/* size of key in bytes */
		__u32	value_size;	/* size of value in bytes */
		__u32	max_entries;	/* max number of entries in a map */
		__u32	map_flags;	/* prealloc or not */
	};

	struct { /* anonymous struct used by BPF_MAP_*_ELEM commands */
		__u32		map_fd;
		__aligned_u64	key;
		union {
			__aligned_u64 value;
			__aligned_u64 next_key;
		};
		__u64		flags;
	};

	struct { /* anonymous struct used by BPF_PROG_LOAD command */
		__u32		prog_type;	/* one of enum bpf_prog_type */
		__u32		insn_cnt;
		__aligned_u64	insns;
		__aligned_u64	license;
		__u32		log_level;	/* verbosity level of verifier */
		__u32		log_size;	/* size of user buffer */
		__aligned_u64	log_buf;	/* user supplied buffer */
		__u32		kern_version;	/* checked when prog_type=kprobe */
	};

	struct { /* anonymous struct used by BPF_OBJ_* commands */
		__aligned_u64	pathname;
		__u32		bpf_fd;
	};
} __attribute__((aligned(8)));

/* BPF helper function descriptions:
 *
 * void *bpf_map_lookup_elem(&map, &key)
 *     Return: Map value or NULL
 *
 * int bpf_map_update_elem(&map, &key, &value, flags)
 *     Return: 0 on success or negative error
 *
 * int bpf_map_delete_elem(&map, &key)
 *     Return: 0 on success or negative error
 *
 * int bpf_probe_read(void *dst, int size, void *src)
 *     Return: 0 on success or negative error
 *
 * u64 bpf_ktime_get_ns(void)
 *     Return: current ktime
 *
 * int bpf_trace_printk(const char *fmt, int fmt_size, ...)
 *     Return: length of buffer written or negative error
 *
 * u32 bpf_prandom_u32(void)
 *     Return: random value
 *
 * u32 bpf_raw_smp_processor_id(void)
 *     Return: SMP processor ID
 *
 * int bpf_skb_store_bytes(skb, offset, from, len, flags)
 *     store bytes into packet
 *     @skb: pointer to skb
 *     @offset: offset within packet from skb->mac_header
 *     @from: pointer where to copy bytes from
 *     @len: number of bytes to store into packet
 *     @flags: bit 0 - if true, recompute skb->csum
 *             other bits - reserved
 *     Return: 0 on success or negative error
 *
 * int bpf_l3_csum_replace(skb, offset, from, to, flags)
 *     recompute IP checksum
 *     @skb: pointer to skb
 *     @offset: offset within packet where IP checksum is located
 *     @from: old value of header field
 *     @to: new value of header field
 *     @flags: bits 0-3 - size of header field
 *             other bits - reserved
 *     Return: 0 on success or negative error
 *
 * int bpf_l4_csum_replace(skb, offset, from, to, flags)
 *     recompute TCP/UDP checksum
 *     @skb: pointer to skb
 *     @offset: offset within packet where TCP/UDP checksum is located
 *     @from: old value of header field
 *     @to: new value of header field
 *     @flags: bits 0-3 - size of header field
 *             bit 4 - is pseudo header
 *             other bits - reserved
 *     Return: 0 on success or negative error
 *
 * int bpf_tail_call(ctx, prog_array_map, index)
 *     jump into another BPF program
 *     @ctx: context pointer passed to next program
 *     @prog_array_map: pointer to map which type is BPF_MAP_TYPE_PROG_ARRAY
 *     @index: index inside array that selects specific program to run
 *     Return: 0 on success or negative error
 *
 * int bpf_clone_redirect(skb, ifindex, flags)
 *     redirect to another netdev
 *     @skb: pointer to skb
 *     @ifindex: ifindex of the net device
 *     @flags: bit 0 - if set, redirect to ingress instead of egress
 *             other bits - reserved
 *     Return: 0 on success or negative error
 *
 * u64 bpf_get_current_pid_tgid(void)
 *     Return: current->tgid << 32 | current->pid
 *
 * u64 bpf_get_current_uid_gid(void)
 *     Return: current_gid << 32 | current_uid
 *
 * int bpf_get_current_comm(char *buf, int size_of_buf)
 *     stores current->comm into buf
 *     Return: 0 on success or negative error
 *
 * u32 bpf_get_cgroup_classid(skb)
 *     retrieve a proc's classid
 *     @skb: pointer to skb
 *     Return: classid if != 0
 *
 * int bpf_skb_vlan_push(skb, vlan_proto, vlan_tci)
 *     Return: 0 on success or negative error
 *
 * int bpf_skb_vlan_pop(skb)
 *     Return: 0 on success or negative error
 *
 * int bpf_skb_get_tunnel_key(skb, key, size, flags)
 * int bpf_skb_set_tunnel_key(skb, key, size, flags)
 *     retrieve or populate tunnel metadata
 *     @skb: pointer to skb
 *     @key: pointer to 'struct bpf_tunnel_key'
 *     @size: size of 'struct bpf_tunnel_key'
 *     @flags: room for future extensions
 *     Return: 0 on success or negative error
 *
 * u64 bpf_perf_event_read(&map, index)
 *     Return: Number events read or error code
 *
 * int bpf_redirect(ifindex, flags)
 *     redirect to another netdev
 *     @ifindex: ifindex of the net device
 *     @flags: bit 0 - if set, redirect to ingress instead of egress
 *             other bits - reserved
 *     Return: TC_ACT_REDIRECT
 *
 * u32 bpf_get_route_realm(skb)
 *     retrieve a dst's tclassid
 *     @skb: pointer to skb
 *     Return: realm if != 0
 *
 * int bpf_perf_event_output(ctx, map, index, data, size)
 *     output perf raw sample
 *     @ctx: struct pt_regs*
 *     @map: pointer to perf_event_array map
 *     @index: index of event in the map
 *     @data: data on stack to be output as raw data
 *     @size: size of data
 *     Return: 0 on success or negative error
 *
 * int bpf_get_stackid(ctx, map, flags)
 *     walk user or kernel stack and return id
 *     @ctx: struct pt_regs*
 *     @map: pointer to stack_trace map
 *     @flags: bits 0-7 - numer of stack frames to skip
 *             bit 8 - collect user stack instead of kernel
 *             bit 9 - compare stacks by hash only
 *             bit 10 - if two different stacks hash into the same stackid
 *                      discard old
 *             other bits - reserved
 *     Return: >= 0 stackid on success or negative error
 *
 * s64 bpf_csum_diff(from, from_size, to, to_size, seed)
 *     calculate csum diff
 *     @from: raw from buffer
 *     @from_size: length of from buffer
 *     @to: raw to buffer
 *     @to_size: length of to buffer
 *     @seed: optional seed
 *     Return: csum result or negative error code
 *
 * int bpf_skb_get_tunnel_opt(skb, opt, size)
 *     retrieve tunnel options metadata
 *     @skb: pointer to skb
 *     @opt: pointer to raw tunnel option data
 *     @size: size of @opt
 *     Return: option size
 *
 * int bpf_skb_set_tunnel_opt(skb, opt, size)
 *     populate tunnel options metadata
 *     @skb: pointer to skb
 *     @opt: pointer to raw tunnel option data
 *     @size: size of @opt
 *     Return: 0 on success or negative error
 *
 * int bpf_skb_change_proto(skb, proto, flags)
 *     Change protocol of the skb. Currently supported is v4 -> v6,
 *     v6 -> v4 transitions. The helper will also resize the skb. eBPF
 *     program is expected to fill the new headers via skb_store_bytes
 *     and lX_csum_replace.
 *     @skb: pointer to skb
 *     @proto: new skb->protocol type
 *     @flags: reserved
 *     Return: 0 on success or negative error
 *
 * int bpf_skb_change_type(skb, type)
 *     Change packet type of skb.
 *     @skb: pointer to skb
 *     @type: new skb->pkt_type type
 *     Return: 0 on success or negative error
 *
 * int bpf_skb_under_cgroup(skb, map, index)
 *     Check cgroup2 membership of skb
 *     @skb: pointer to skb
 *     @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type
 *     @index: index of the cgroup in the bpf_map
 *     Return:
 *       == 0 skb failed the cgroup2 descendant test
 *       == 1 skb succeeded the cgroup2 descendant test
 *        < 0 error
 *
 * u32 bpf_get_hash_recalc(skb)
 *     Retrieve and possibly recalculate skb->hash.
 *     @skb: pointer to skb
 *     Return: hash
 *
 * u64 bpf_get_current_task(void)
 *     Returns current task_struct
 *     Return: current
 *
 * int bpf_probe_write_user(void *dst, void *src, int len)
 *     safely attempt to write to a location
 *     @dst: destination address in userspace
 *     @src: source address on stack
 *     @len: number of bytes to copy
 *     Return: 0 on success or negative error
 *
 * int bpf_current_task_under_cgroup(map, index)
 *     Check cgroup2 membership of current task
 *     @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type
 *     @index: index of the cgroup in the bpf_map
 *     Return:
 *       == 0 current failed the cgroup2 descendant test
 *       == 1 current succeeded the cgroup2 descendant test
 *        < 0 error
 *
 * int bpf_skb_change_tail(skb, len, flags)
 *     The helper will resize the skb to the given new size, to be used f.e.
 *     with control messages.
 *     @skb: pointer to skb
 *     @len: new skb length
 *     @flags: reserved
 *     Return: 0 on success or negative error
 *
 * int bpf_skb_pull_data(skb, len)
 *     The helper will pull in non-linear data in case the skb is non-linear
 *     and not all of len are part of the linear section. Only needed for
 *     read/write with direct packet access.
 *     @skb: pointer to skb
 *     @len: len to make read/writeable
 *     Return: 0 on success or negative error
 *
 * s64 bpf_csum_update(skb, csum)
 *     Adds csum into skb->csum in case of CHECKSUM_COMPLETE.
 *     @skb: pointer to skb
 *     @csum: csum to add
 *     Return: csum on success or negative error
 *
 * void bpf_set_hash_invalid(skb)
 *     Invalidate current skb->hash.
 *     @skb: pointer to skb
 *
 * int bpf_get_numa_node_id()
 *     Return: Id of current NUMA node.
 */
#define __BPF_FUNC_MAPPER(FN)		\
	FN(unspec),			\
	FN(map_lookup_elem),		\
	FN(map_update_elem),		\
	FN(map_delete_elem),		\
	FN(probe_read),			\
	FN(ktime_get_ns),		\
	FN(trace_printk),		\
	FN(get_prandom_u32),		\
	FN(get_smp_processor_id),	\
	FN(skb_store_bytes),		\
	FN(l3_csum_replace),		\
	FN(l4_csum_replace),		\
	FN(tail_call),			\
	FN(clone_redirect),		\
	FN(get_current_pid_tgid),	\
	FN(get_current_uid_gid),	\
	FN(get_current_comm),		\
	FN(get_cgroup_classid),		\
	FN(skb_vlan_push),		\
	FN(skb_vlan_pop),		\
	FN(skb_get_tunnel_key),		\
	FN(skb_set_tunnel_key),		\
	FN(perf_event_read),		\
	FN(redirect),			\
	FN(get_route_realm),		\
	FN(perf_event_output),		\
	FN(skb_load_bytes),		\
	FN(get_stackid),		\
	FN(csum_diff),			\
	FN(skb_get_tunnel_opt),		\
	FN(skb_set_tunnel_opt),		\
	FN(skb_change_proto),		\
	FN(skb_change_type),		\
	FN(skb_under_cgroup),		\
	FN(get_hash_recalc),		\
	FN(get_current_task),		\
	FN(probe_write_user),		\
	FN(current_task_under_cgroup),	\
	FN(skb_change_tail),		\
	FN(skb_pull_data),		\
	FN(csum_update),		\
	FN(set_hash_invalid),		\
	FN(get_numa_node_id),

/* integer value in 'imm' field of BPF_CALL instruction selects which helper
 * function eBPF program intends to call
 */
#define __BPF_ENUM_FN(x) BPF_FUNC_ ## x
enum bpf_func_id {
	__BPF_FUNC_MAPPER(__BPF_ENUM_FN)
	__BPF_FUNC_MAX_ID,
};
#undef __BPF_ENUM_FN

/* All flags used by eBPF helper functions, placed here. */

/* BPF_FUNC_skb_store_bytes flags. */
#define BPF_F_RECOMPUTE_CSUM		(1ULL << 0)
#define BPF_F_INVALIDATE_HASH		(1ULL << 1)

/* BPF_FUNC_l3_csum_replace and BPF_FUNC_l4_csum_replace flags.
 * First 4 bits are for passing the header field size.
 */
#define BPF_F_HDR_FIELD_MASK		0xfULL

/* BPF_FUNC_l4_csum_replace flags. */
#define BPF_F_PSEUDO_HDR		(1ULL << 4)
#define BPF_F_MARK_MANGLED_0		(1ULL << 5)

/* BPF_FUNC_clone_redirect and BPF_FUNC_redirect flags. */
#define BPF_F_INGRESS			(1ULL << 0)

/* BPF_FUNC_skb_set_tunnel_key and BPF_FUNC_skb_get_tunnel_key flags. */
#define BPF_F_TUNINFO_IPV6		(1ULL << 0)

/* BPF_FUNC_get_stackid flags. */
#define BPF_F_SKIP_FIELD_MASK		0xffULL
#define BPF_F_USER_STACK		(1ULL << 8)
#define BPF_F_FAST_STACK_CMP		(1ULL << 9)
#define BPF_F_REUSE_STACKID		(1ULL << 10)

/* BPF_FUNC_skb_set_tunnel_key flags. */
#define BPF_F_ZERO_CSUM_TX		(1ULL << 1)
#define BPF_F_DONT_FRAGMENT		(1ULL << 2)

/* BPF_FUNC_perf_event_output and BPF_FUNC_perf_event_read flags. */
#define BPF_F_INDEX_MASK		0xffffffffULL
#define BPF_F_CURRENT_CPU		BPF_F_INDEX_MASK
/* BPF_FUNC_perf_event_output for sk_buff input context. */
#define BPF_F_CTXLEN_MASK		(0xfffffULL << 32)

/* user accessible mirror of in-kernel sk_buff.
 * new fields can only be added to the end of this structure
 */
struct __sk_buff {
	__u32 len;
	__u32 pkt_type;
	__u32 mark;
	__u32 queue_mapping;
	__u32 protocol;
	__u32 vlan_present;
	__u32 vlan_tci;
	__u32 vlan_proto;
	__u32 priority;
	__u32 ingress_ifindex;
	__u32 ifindex;
	__u32 tc_index;
	__u32 cb[5];
	__u32 hash;
	__u32 tc_classid;
	__u32 data;
	__u32 data_end;
};

struct bpf_tunnel_key {
	__u32 tunnel_id;
	union {
		__u32 remote_ipv4;
		__u32 remote_ipv6[4];
	};
	__u8 tunnel_tos;
	__u8 tunnel_ttl;
	__u16 tunnel_ext;
	__u32 tunnel_label;
};

/* User return codes for XDP prog type.
 * A valid XDP program must return one of these defined values. All other
 * return codes are reserved for future use. Unknown return codes will result
 * in packet drop.
 */
enum xdp_action {
	XDP_ABORTED = 0,
	XDP_DROP,
	XDP_PASS,
	XDP_TX,
};

/* user accessible metadata for XDP packet hook
 * new fields must be added to the end of this structure
 */
struct xdp_md {
	__u32 data;
	__u32 data_end;
};

#endif /* _UAPI__LINUX_BPF_H__ */
Commit	Line	Data
daedfb22 AS	1	/* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com
	2	*
	3	* This program is free software; you can redistribute it and/or
	4	* modify it under the terms of version 2 of the GNU General Public
	5	* License as published by the Free Software Foundation.
	6	*/
	7	#ifndef _UAPI__LINUX_BPF_H__
	8	#define _UAPI__LINUX_BPF_H__
	9
	10	#include <linux/types.h>
c15952dc	11	#include <linux/bpf_common.h>
daedfb22 AS	12
	13	/* Extended instruction set based on top of classic BPF */
	14
	15	/* instruction classes */
	16	#define BPF_ALU64 0x07 /* alu mode in double word width */
	17
	18	/* ld/ldx fields */
	19	#define BPF_DW 0x18 /* double word */
	20	#define BPF_XADD 0xc0 /* exclusive add */
	21
	22	/* alu/jmp fields */
	23	#define BPF_MOV 0xb0 /* mov reg to reg */
	24	#define BPF_ARSH 0xc0 /* sign extending arithmetic shift right */
	25
	26	/* change endianness of a register */
	27	#define BPF_END 0xd0 /* flags for endianness conversion: */
	28	#define BPF_TO_LE 0x00 /* convert to little-endian */
	29	#define BPF_TO_BE 0x08 /* convert to big-endian */
	30	#define BPF_FROM_LE BPF_TO_LE
	31	#define BPF_FROM_BE BPF_TO_BE
	32
	33	#define BPF_JNE 0x50 /* jump != */
	34	#define BPF_JSGT 0x60 /* SGT is signed '>', GT in x86 */
	35	#define BPF_JSGE 0x70 /* SGE is signed '>=', GE in x86 */
	36	#define BPF_CALL 0x80 /* function call */
	37	#define BPF_EXIT 0x90 /* function return */
	38
	39	/* Register numbers */
	40	enum {
	41	BPF_REG_0 = 0,
	42	BPF_REG_1,
	43	BPF_REG_2,
	44	BPF_REG_3,
	45	BPF_REG_4,
	46	BPF_REG_5,
	47	BPF_REG_6,
	48	BPF_REG_7,
	49	BPF_REG_8,
	50	BPF_REG_9,
	51	BPF_REG_10,
	52	__MAX_BPF_REG,
	53	};
	54
	55	/* BPF has 10 general purpose 64-bit registers and stack frame. */
	56	#define MAX_BPF_REG __MAX_BPF_REG
	57
	58	struct bpf_insn {
	59	__u8 code; /* opcode */
	60	__u8 dst_reg:4; /* dest register */
	61	__u8 src_reg:4; /* source register */
	62	__s16 off; /* signed offset */
	63	__s32 imm; /* signed immediate constant */
	64	};
	65
b2197755	66	/* BPF syscall commands, see bpf(2) man-page for details. */
99c55f7d	67	enum bpf_cmd {
99c55f7d	68	BPF_MAP_CREATE,
db20fd2b	69	BPF_MAP_LOOKUP_ELEM,
db20fd2b	70	BPF_MAP_UPDATE_ELEM,
db20fd2b	71	BPF_MAP_DELETE_ELEM,
db20fd2b	72	BPF_MAP_GET_NEXT_KEY,
09756af4	73	BPF_PROG_LOAD,
b2197755 DB	74	BPF_OBJ_PIN,
b2197755 DB	75	BPF_OBJ_GET,
99c55f7d AS	76	};
	77
	78	enum bpf_map_type {
	79	BPF_MAP_TYPE_UNSPEC,
0f8e4bd8	80	BPF_MAP_TYPE_HASH,
28fbcfa0	81	BPF_MAP_TYPE_ARRAY,
04fd61ab	82	BPF_MAP_TYPE_PROG_ARRAY,
ea317b26	83	BPF_MAP_TYPE_PERF_EVENT_ARRAY,
824bd0ce	84	BPF_MAP_TYPE_PERCPU_HASH,
a10423b8	85	BPF_MAP_TYPE_PERCPU_ARRAY,
d5a3b1f6	86	BPF_MAP_TYPE_STACK_TRACE,
4ed8ec52	87	BPF_MAP_TYPE_CGROUP_ARRAY,
29ba732a	88	BPF_MAP_TYPE_LRU_HASH,
8f844938	89	BPF_MAP_TYPE_LRU_PERCPU_HASH,
99c55f7d AS	90	};
99c55f7d AS	91
09756af4 AS	92	enum bpf_prog_type {
09756af4 AS	93	BPF_PROG_TYPE_UNSPEC,
ddd872bc	94	BPF_PROG_TYPE_SOCKET_FILTER,
2541517c	95	BPF_PROG_TYPE_KPROBE,
96be4325	96	BPF_PROG_TYPE_SCHED_CLS,
94caee8c	97	BPF_PROG_TYPE_SCHED_ACT,
98b5c2c6	98	BPF_PROG_TYPE_TRACEPOINT,
6a773a15	99	BPF_PROG_TYPE_XDP,
0515e599	100	BPF_PROG_TYPE_PERF_EVENT,
0e33661d	101	BPF_PROG_TYPE_CGROUP_SKB,
09756af4 AS	102	};
09756af4 AS	103
0e33661d DM	104	enum bpf_attach_type {
	105	BPF_CGROUP_INET_INGRESS,
	106	BPF_CGROUP_INET_EGRESS,
	107	__MAX_BPF_ATTACH_TYPE
	108	};
	109
	110	#define MAX_BPF_ATTACH_TYPE __MAX_BPF_ATTACH_TYPE
	111
f1a66f85 DB	112	#define BPF_PSEUDO_MAP_FD 1
f1a66f85 DB	113
3274f520 AS	114	/* flags for BPF_MAP_UPDATE_ELEM command */
	115	#define BPF_ANY 0 /* create new element or update existing */
	116	#define BPF_NOEXIST 1 /* create new element if it didn't exist */
	117	#define BPF_EXIST 2 /* update existing element */
	118
6c905981	119	#define BPF_F_NO_PREALLOC (1U << 0)
29ba732a	120	/* Instead of having one common LRU list in the
8f844938	121	* BPF_MAP_TYPE_LRU_[PERCPU_]HASH map, use a percpu LRU list
29ba732a MKL	122	* which can scale and perform better.
	123	* Note, the LRU nodes (including free nodes) cannot be moved
	124	* across different LRU lists.
	125	*/
	126	#define BPF_F_NO_COMMON_LRU (1U << 1)
6c905981	127
99c55f7d AS	128	union bpf_attr {
	129	struct { /* anonymous struct used by BPF_MAP_CREATE command */
	130	__u32 map_type; /* one of enum bpf_map_type */
	131	__u32 key_size; /* size of key in bytes */
	132	__u32 value_size; /* size of value in bytes */
	133	__u32 max_entries; /* max number of entries in a map */
6c905981	134	__u32 map_flags; /* prealloc or not */
99c55f7d	135	};
db20fd2b AS	136
	137	struct { /* anonymous struct used by BPF_MAP__ELEM commands /
	138	__u32 map_fd;
	139	__aligned_u64 key;
	140	union {
	141	__aligned_u64 value;
	142	__aligned_u64 next_key;
	143	};
3274f520	144	__u64 flags;
db20fd2b	145	};
09756af4 AS	146
	147	struct { /* anonymous struct used by BPF_PROG_LOAD command */
	148	__u32 prog_type; /* one of enum bpf_prog_type */
	149	__u32 insn_cnt;
	150	__aligned_u64 insns;
	151	__aligned_u64 license;
cbd35700 AS	152	__u32 log_level; /* verbosity level of verifier */
	153	__u32 log_size; /* size of user buffer */
	154	__aligned_u64 log_buf; /* user supplied buffer */
2541517c	155	__u32 kern_version; /* checked when prog_type=kprobe */
09756af4	156	};
b2197755 DB	157
	158	struct { /* anonymous struct used by BPF_OBJ_* commands */
	159	__aligned_u64 pathname;
	160	__u32 bpf_fd;
	161	};
99c55f7d AS	162	} __attribute__((aligned(8)));
99c55f7d AS	163
ebb676da TG	164	/* BPF helper function descriptions:
	165	*
	166	* void *bpf_map_lookup_elem(&map, &key)
	167	* Return: Map value or NULL
	168	*
	169	* int bpf_map_update_elem(&map, &key, &value, flags)
	170	* Return: 0 on success or negative error
	171	*
	172	* int bpf_map_delete_elem(&map, &key)
	173	* Return: 0 on success or negative error
	174	*
	175	* int bpf_probe_read(void dst, int size, void src)
	176	* Return: 0 on success or negative error
	177	*
	178	* u64 bpf_ktime_get_ns(void)
	179	* Return: current ktime
	180	*
	181	* int bpf_trace_printk(const char *fmt, int fmt_size, ...)
	182	* Return: length of buffer written or negative error
	183	*
	184	* u32 bpf_prandom_u32(void)
	185	* Return: random value
	186	*
	187	* u32 bpf_raw_smp_processor_id(void)
	188	* Return: SMP processor ID
	189	*
	190	* int bpf_skb_store_bytes(skb, offset, from, len, flags)
	191	* store bytes into packet
	192	* @skb: pointer to skb
	193	* @offset: offset within packet from skb->mac_header
	194	* @from: pointer where to copy bytes from
	195	* @len: number of bytes to store into packet
	196	* @flags: bit 0 - if true, recompute skb->csum
	197	* other bits - reserved
	198	* Return: 0 on success or negative error
	199	*
	200	* int bpf_l3_csum_replace(skb, offset, from, to, flags)
	201	* recompute IP checksum
	202	* @skb: pointer to skb
	203	* @offset: offset within packet where IP checksum is located
	204	* @from: old value of header field
	205	* @to: new value of header field
	206	* @flags: bits 0-3 - size of header field
	207	* other bits - reserved
	208	* Return: 0 on success or negative error
	209	*
	210	* int bpf_l4_csum_replace(skb, offset, from, to, flags)
	211	* recompute TCP/UDP checksum
	212	* @skb: pointer to skb
	213	* @offset: offset within packet where TCP/UDP checksum is located
	214	* @from: old value of header field
	215	* @to: new value of header field
	216	* @flags: bits 0-3 - size of header field
	217	* bit 4 - is pseudo header
	218	* other bits - reserved
	219	* Return: 0 on success or negative error
	220	*
	221	* int bpf_tail_call(ctx, prog_array_map, index)
	222	* jump into another BPF program
	223	* @ctx: context pointer passed to next program
	224	* @prog_array_map: pointer to map which type is BPF_MAP_TYPE_PROG_ARRAY
	225	* @index: index inside array that selects specific program to run
	226	* Return: 0 on success or negative error
	227	*
228	* int bpf_clone_redirect(skb, ifindex, flags)
229	* redirect to another netdev
230	* @skb: pointer to skb
231	* @ifindex: ifindex of the net device
232	* @flags: bit 0 - if set, redirect to ingress instead of egress
233	* other bits - reserved
234	* Return: 0 on success or negative error
235	*
236	* u64 bpf_get_current_pid_tgid(void)
237	* Return: current->tgid << 32 \| current->pid
238	*
239	* u64 bpf_get_current_uid_gid(void)
240	* Return: current_gid << 32 \| current_uid
241	*
242	* int bpf_get_current_comm(char *buf, int size_of_buf)
243	* stores current->comm into buf
244	* Return: 0 on success or negative error
245	*
246	* u32 bpf_get_cgroup_classid(skb)
247	* retrieve a proc's classid
248	* @skb: pointer to skb
249	* Return: classid if != 0
250	*
251	* int bpf_skb_vlan_push(skb, vlan_proto, vlan_tci)
252	* Return: 0 on success or negative error
253	*
254	* int bpf_skb_vlan_pop(skb)
255	* Return: 0 on success or negative error
256	*
257	* int bpf_skb_get_tunnel_key(skb, key, size, flags)
258	* int bpf_skb_set_tunnel_key(skb, key, size, flags)
259	* retrieve or populate tunnel metadata
260	* @skb: pointer to skb
261	* @key: pointer to 'struct bpf_tunnel_key'
262	* @size: size of 'struct bpf_tunnel_key'
263	* @flags: room for future extensions
264	* Return: 0 on success or negative error
265	*
266	* u64 bpf_perf_event_read(&map, index)
267	* Return: Number events read or error code
268	*
269	* int bpf_redirect(ifindex, flags)
270	* redirect to another netdev
271	* @ifindex: ifindex of the net device
272	* @flags: bit 0 - if set, redirect to ingress instead of egress
273	* other bits - reserved
274	* Return: TC_ACT_REDIRECT
275	*
276	* u32 bpf_get_route_realm(skb)
277	* retrieve a dst's tclassid
278	* @skb: pointer to skb
279	* Return: realm if != 0
280	*
281	* int bpf_perf_event_output(ctx, map, index, data, size)
282	* output perf raw sample
283	* @ctx: struct pt_regs*
284	* @map: pointer to perf_event_array map
285	* @index: index of event in the map
286	* @data: data on stack to be output as raw data
287	* @size: size of data
288	* Return: 0 on success or negative error
289	*
290	* int bpf_get_stackid(ctx, map, flags)
291	* walk user or kernel stack and return id
292	* @ctx: struct pt_regs*
293	* @map: pointer to stack_trace map
294	* @flags: bits 0-7 - numer of stack frames to skip
295	* bit 8 - collect user stack instead of kernel
296	* bit 9 - compare stacks by hash only
297	* bit 10 - if two different stacks hash into the same stackid
298	* discard old
299	* other bits - reserved
300	* Return: >= 0 stackid on success or negative error
301	*
302	* s64 bpf_csum_diff(from, from_size, to, to_size, seed)
303	* calculate csum diff
304	* @from: raw from buffer
305	* @from_size: length of from buffer
306	* @to: raw to buffer
307	* @to_size: length of to buffer
308	* @seed: optional seed
309	* Return: csum result or negative error code
310	*
311	* int bpf_skb_get_tunnel_opt(skb, opt, size)
312	* retrieve tunnel options metadata
313	* @skb: pointer to skb
314	* @opt: pointer to raw tunnel option data
315	* @size: size of @opt
316	* Return: option size
317	*
318	* int bpf_skb_set_tunnel_opt(skb, opt, size)
319	* populate tunnel options metadata
320	* @skb: pointer to skb
321	* @opt: pointer to raw tunnel option data
322	* @size: size of @opt
323	* Return: 0 on success or negative error
324	*
325	* int bpf_skb_change_proto(skb, proto, flags)
326	* Change protocol of the skb. Currently supported is v4 -> v6,
327	* v6 -> v4 transitions. The helper will also resize the skb. eBPF
328	* program is expected to fill the new headers via skb_store_bytes
329	* and lX_csum_replace.
330	* @skb: pointer to skb
331	* @proto: new skb->protocol type
332	* @flags: reserved
333	* Return: 0 on success or negative error
334	*
335	* int bpf_skb_change_type(skb, type)
336	* Change packet type of skb.
337	* @skb: pointer to skb
338	* @type: new skb->pkt_type type
339	* Return: 0 on success or negative error
340	*
341	* int bpf_skb_under_cgroup(skb, map, index)
342	* Check cgroup2 membership of skb
343	* @skb: pointer to skb
344	* @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type
345	* @index: index of the cgroup in the bpf_map
346	* Return:
347	* == 0 skb failed the cgroup2 descendant test
348	* == 1 skb succeeded the cgroup2 descendant test
349	* < 0 error
350	*
351	* u32 bpf_get_hash_recalc(skb)
352	* Retrieve and possibly recalculate skb->hash.
353	* @skb: pointer to skb
354	* Return: hash
355	*
356	* u64 bpf_get_current_task(void)
357	* Returns current task_struct
358	* Return: current
359	*
360	* int bpf_probe_write_user(void dst, void src, int len)
361	* safely attempt to write to a location
362	* @dst: destination address in userspace
363	* @src: source address on stack
364	* @len: number of bytes to copy
365	* Return: 0 on success or negative error
366	*
367	* int bpf_current_task_under_cgroup(map, index)
368	* Check cgroup2 membership of current task
369	* @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type
370	* @index: index of the cgroup in the bpf_map
371	* Return:
372	* == 0 current failed the cgroup2 descendant test
373	* == 1 current succeeded the cgroup2 descendant test
374	* < 0 error
375	*
376	* int bpf_skb_change_tail(skb, len, flags)
377	* The helper will resize the skb to the given new size, to be used f.e.
378	* with control messages.
379	* @skb: pointer to skb
380	* @len: new skb length
381	* @flags: reserved
382	* Return: 0 on success or negative error
383	*
384	* int bpf_skb_pull_data(skb, len)
385	* The helper will pull in non-linear data in case the skb is non-linear
386	* and not all of len are part of the linear section. Only needed for
387	* read/write with direct packet access.
388	* @skb: pointer to skb
389	* @len: len to make read/writeable
390	* Return: 0 on success or negative error
391	*
392	* s64 bpf_csum_update(skb, csum)
393	* Adds csum into skb->csum in case of CHECKSUM_COMPLETE.
394	* @skb: pointer to skb
395	* @csum: csum to add
396	* Return: csum on success or negative error
397	*
398	* void bpf_set_hash_invalid(skb)
399	* Invalidate current skb->hash.
400	* @skb: pointer to skb
401	*
402	* int bpf_get_numa_node_id()
403	* Return: Id of current NUMA node.
404	*/
405	#define __BPF_FUNC_MAPPER(FN) \
406	FN(unspec), \
407	FN(map_lookup_elem), \
408	FN(map_update_elem), \
409	FN(map_delete_elem), \
410	FN(probe_read), \
411	FN(ktime_get_ns), \
412	FN(trace_printk), \
413	FN(get_prandom_u32), \
414	FN(get_smp_processor_id), \
415	FN(skb_store_bytes), \
416	FN(l3_csum_replace), \
417	FN(l4_csum_replace), \
418	FN(tail_call), \
419	FN(clone_redirect), \
420	FN(get_current_pid_tgid), \
421	FN(get_current_uid_gid), \
422	FN(get_current_comm), \
423	FN(get_cgroup_classid), \
424	FN(skb_vlan_push), \
425	FN(skb_vlan_pop), \
426	FN(skb_get_tunnel_key), \
427	FN(skb_set_tunnel_key), \
428	FN(perf_event_read), \
429	FN(redirect), \
430	FN(get_route_realm), \
431	FN(perf_event_output), \
432	FN(skb_load_bytes), \
433	FN(get_stackid), \
434	FN(csum_diff), \
435	FN(skb_get_tunnel_opt), \
436	FN(skb_set_tunnel_opt), \
437	FN(skb_change_proto), \
438	FN(skb_change_type), \
439	FN(skb_under_cgroup), \
440	FN(get_hash_recalc), \
441	FN(get_current_task), \
442	FN(probe_write_user), \
443	FN(current_task_under_cgroup), \
444	FN(skb_change_tail), \
445	FN(skb_pull_data), \
446	FN(csum_update), \
447	FN(set_hash_invalid), \
448	FN(get_numa_node_id),
449
09756af4 AS	450	/* integer value in 'imm' field of BPF_CALL instruction selects which helper
	451	* function eBPF program intends to call
	452	*/
ebb676da	453	#define __BPF_ENUM_FN(x) BPF_FUNC_ ## x
09756af4	454	enum bpf_func_id {
ebb676da	455	__BPF_FUNC_MAPPER(__BPF_ENUM_FN)
09756af4 AS	456	__BPF_FUNC_MAX_ID,
09756af4 AS	457	};
ebb676da	458	#undef __BPF_ENUM_FN
09756af4	459
781c53bc DB	460	/* All flags used by eBPF helper functions, placed here. */
	461
	462	/* BPF_FUNC_skb_store_bytes flags. */
	463	#define BPF_F_RECOMPUTE_CSUM (1ULL << 0)
8afd54c8	464	#define BPF_F_INVALIDATE_HASH (1ULL << 1)
781c53bc DB	465
	466	/* BPF_FUNC_l3_csum_replace and BPF_FUNC_l4_csum_replace flags.
	467	* First 4 bits are for passing the header field size.
	468	*/
	469	#define BPF_F_HDR_FIELD_MASK 0xfULL
	470
	471	/* BPF_FUNC_l4_csum_replace flags. */
	472	#define BPF_F_PSEUDO_HDR (1ULL << 4)
2f72959a	473	#define BPF_F_MARK_MANGLED_0 (1ULL << 5)
781c53bc DB	474
	475	/* BPF_FUNC_clone_redirect and BPF_FUNC_redirect flags. */
	476	#define BPF_F_INGRESS (1ULL << 0)
	477
c6c33454 DB	478	/* BPF_FUNC_skb_set_tunnel_key and BPF_FUNC_skb_get_tunnel_key flags. */
	479	#define BPF_F_TUNINFO_IPV6 (1ULL << 0)
	480
d5a3b1f6 AS	481	/* BPF_FUNC_get_stackid flags. */
	482	#define BPF_F_SKIP_FIELD_MASK 0xffULL
	483	#define BPF_F_USER_STACK (1ULL << 8)
	484	#define BPF_F_FAST_STACK_CMP (1ULL << 9)
	485	#define BPF_F_REUSE_STACKID (1ULL << 10)
	486
2da897e5 DB	487	/* BPF_FUNC_skb_set_tunnel_key flags. */
2da897e5 DB	488	#define BPF_F_ZERO_CSUM_TX (1ULL << 1)
22080870	489	#define BPF_F_DONT_FRAGMENT (1ULL << 2)
2da897e5	490
6816a7ff	491	/* BPF_FUNC_perf_event_output and BPF_FUNC_perf_event_read flags. */
1e33759c DB	492	#define BPF_F_INDEX_MASK 0xffffffffULL
1e33759c DB	493	#define BPF_F_CURRENT_CPU BPF_F_INDEX_MASK
555c8a86 DB	494	/* BPF_FUNC_perf_event_output for sk_buff input context. */
555c8a86 DB	495	#define BPF_F_CTXLEN_MASK (0xfffffULL << 32)
1e33759c	496
9bac3d6d AS	497	/* user accessible mirror of in-kernel sk_buff.
	498	* new fields can only be added to the end of this structure
	499	*/
	500	struct __sk_buff {
	501	__u32 len;
	502	__u32 pkt_type;
	503	__u32 mark;
	504	__u32 queue_mapping;
c2497395 AS	505	__u32 protocol;
	506	__u32 vlan_present;
	507	__u32 vlan_tci;
27cd5452	508	__u32 vlan_proto;
bcad5718	509	__u32 priority;
37e82c2f AS	510	__u32 ingress_ifindex;
37e82c2f AS	511	__u32 ifindex;
d691f9e8 AS	512	__u32 tc_index;
d691f9e8 AS	513	__u32 cb[5];
ba7591d8	514	__u32 hash;
045efa82	515	__u32 tc_classid;
969bf05e AS	516	__u32 data;
969bf05e AS	517	__u32 data_end;
9bac3d6d AS	518	};
9bac3d6d AS	519
d3aa45ce AS	520	struct bpf_tunnel_key {
d3aa45ce AS	521	__u32 tunnel_id;
c6c33454 DB	522	union {
	523	__u32 remote_ipv4;
	524	__u32 remote_ipv6[4];
	525	};
	526	__u8 tunnel_tos;
	527	__u8 tunnel_ttl;
c0e760c9	528	__u16 tunnel_ext;
4018ab18	529	__u32 tunnel_label;
d3aa45ce AS	530	};
d3aa45ce AS	531
6a773a15 BB	532	/* User return codes for XDP prog type.
	533	* A valid XDP program must return one of these defined values. All other
	534	* return codes are reserved for future use. Unknown return codes will result
	535	* in packet drop.
	536	*/
	537	enum xdp_action {
	538	XDP_ABORTED = 0,
	539	XDP_DROP,
	540	XDP_PASS,
6ce96ca3	541	XDP_TX,
6a773a15 BB	542	};
	543
	544	/* user accessible metadata for XDP packet hook
	545	* new fields must be added to the end of this structure
	546	*/
	547	struct xdp_md {
	548	__u32 data;
	549	__u32 data_end;
	550	};
	551
daedfb22	552	#endif /* _UAPI__LINUX_BPF_H__ */