Allow users to attach a 64-bits cookie to a BPF program when linking it to fentry, fexit, or fmod_ret of a function. This patchset includes several major changes. - Define struct bpf_tramp_links to replace bpf_tramp_prog. struct bpf_tramp_links collects bpf_links of a trampoline - Generate a trampoline to call bpf_progs of given bpf_links. - Trampolines always set/reset bpf_run_ctx before/after calling/leaving a BPF program. - Attach a cookie to a bpf_link of fentry/fexit/fmod_ret. The value will be available when running the associated bpf_prog. The major differences from v1: - Remove program ID and flags from stacks. - Create bpf_trace_run_ctx on stacks. - Attach a cookie to a bpf_link. - Store the cookie of the running BPF program/link in bpf_trace_run_ctx. v1: https://lore.kernel.org/all/20220126214809.3868787-1-kuifeng@xxxxxx/ Kui-Feng Lee (4): bpf, x86: Generate trampolines from bpf_links bpf, x86: Create bpf_trace_run_ctx on the caller thread's stack bpf, x86: Support BPF cookie for fentry/fexit/fmod_ret. selftest/bpf: The test cses of BPF cookie for fentry/fexit/fmod_ret. arch/x86/net/bpf_jit_comp.c | 68 ++++++++++---- include/linux/bpf.h | 40 ++++---- include/linux/bpf_types.h | 1 + include/uapi/linux/bpf.h | 2 + kernel/bpf/bpf_struct_ops.c | 68 +++++++++----- kernel/bpf/syscall.c | 19 ++-- kernel/bpf/trampoline.c | 94 ++++++++++++------- kernel/trace/bpf_trace.c | 17 ++++ net/bpf/bpf_dummy_struct_ops.c | 35 ++++++- tools/bpf/bpftool/link.c | 1 + tools/include/uapi/linux/bpf.h | 2 + tools/lib/bpf/bpf.c | 14 +++ tools/lib/bpf/bpf.h | 1 + tools/lib/bpf/libbpf.map | 1 + .../selftests/bpf/prog_tests/bpf_cookie.c | 61 ++++++++++++ .../selftests/bpf/progs/test_bpf_cookie.c | 24 +++++ 16 files changed, 344 insertions(+), 104 deletions(-) -- 2.30.2
