4 ==========================
5 Linux implementation notes
6 ==========================
8 This document provides more details specific to the Linux kernel implementation of the eBPF instruction set.
10 Byte swap instructions
11 ======================
13 ``BPF_FROM_LE`` and ``BPF_FROM_BE`` exist as aliases for ``BPF_TO_LE`` and ``BPF_TO_BE`` respectively.
18 ``BPF_CALL | BPF_X | BPF_JMP`` (0x8d), where the helper function
19 integer would be read from a specified register, is not currently supported
20 by the verifier. Any programs with this instruction will fail to load
21 until such support is added.
26 Linux only supports the 'map_val(map)' operation on array maps with a single element.
28 Linux uses an fd_array to store maps associated with a BPF program. Thus,
29 map_by_idx(imm) uses the fd at that index in the array.
34 The following 64-bit immediate instruction specifies that a variable address,
35 which corresponds to some integer stored in the 'imm' field, should be loaded:
37 ========================= ====== === ========================================= =========== ==============
38 opcode construction opcode src pseudocode imm type dst type
39 ========================= ====== === ========================================= =========== ==============
40 BPF_IMM | BPF_DW | BPF_LD 0x18 0x3 dst = var_addr(imm) variable id data pointer
41 ========================= ====== === ========================================= =========== ==============
43 On Linux, this integer is a BTF ID.
45 Legacy BPF Packet access instructions
46 =====================================
48 As mentioned in the `ISA standard documentation
49 <instruction-set.html#legacy-bpf-packet-access-instructions>`_,
50 Linux has special eBPF instructions for access to packet data that have been
51 carried over from classic BPF to retain the performance of legacy socket
52 filters running in the eBPF interpreter.
54 The instructions come in two forms: ``BPF_ABS | <size> | BPF_LD`` and
55 ``BPF_IND | <size> | BPF_LD``.
57 These instructions are used to access packet data and can only be used when
58 the program context is a pointer to a networking packet. ``BPF_ABS``
59 accesses packet data at an absolute offset specified by the immediate data
60 and ``BPF_IND`` access packet data at an offset that includes the value of
61 a register in addition to the immediate data.
63 These instructions have seven implicit operands:
65 * Register R6 is an implicit input that must contain a pointer to a
67 * Register R0 is an implicit output which contains the data fetched from
69 * Registers R1-R5 are scratch registers that are clobbered by the
72 These instructions have an implicit program exit condition as well. If an
73 eBPF program attempts access data beyond the packet boundary, the
74 program execution will be aborted.
76 ``BPF_ABS | BPF_W | BPF_LD`` (0x20) means::
78 R0 = ntohl(*(u32 *) ((struct sk_buff *) R6->data + imm))
80 where ``ntohl()`` converts a 32-bit value from network byte order to host byte order.
82 ``BPF_IND | BPF_W | BPF_LD`` (0x40) means::
84 R0 = ntohl(*(u32 *) ((struct sk_buff *) R6->data + src + imm))