[bpf-next,0/4] bpf, x64: Fix tailcall hierarchy

Message

Leon Hwang Jan. 4, 2024, 2:22 p.m. UTC

The patchset fixes a tailcall hierarchy issue.

The issue is confirmed in the discussions of "bpf, x64: Fix tailcall infinite
loop"[0].

The issue is only resolved on x86.

Hopefully, the issue on arm64 and s390x will be resolved soon.

I provide a long commit message in the second patch to describe how the issue
happens and how this patchset resolves the issue in detail.

RFC v2 -> v1:
  * address comments from Maciej:
    * Replace all memcpy(prog, x86_nops[5], X86_PATCH_SIZE) with
        emit_nops(&prog, X86_PATCH_SIZE)

RFC v1 -> RFC v2:
  * address comments from Stanislav:
    * Separate moving emit_nops() as first patch.

Links:
[0] https://lore.kernel.org/bpf/[e-mail]/

Leon Hwang (4):
  bpf, x64: Use emit_nops() to replace memcpy()'ing x86_nops[5]
  bpf, x64: Fix tailcall hierarchy
  bpf, x64: Rename RESTORE_TAIL_CALL_CNT() to LOAD_TAIL_CALL_CNT_PTR()
  selftests/bpf: Add testcases for tailcall hierarchy fixing

 arch/x86/net/bpf_jit_comp.c                   | 108 ++---
 .../selftests/bpf/prog_tests/tailcalls.c      | 418 ++++++++++++++++++
 .../bpf/progs/tailcall_bpf2bpf_hierarchy1.c   |  34 ++
 .../bpf/progs/tailcall_bpf2bpf_hierarchy2.c   |  55 +++
 .../bpf/progs/tailcall_bpf2bpf_hierarchy3.c   |  46 ++
 5 files changed, 609 insertions(+), 52 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/tailcall_bpf2bpf_hierarchy1.c
 create mode 100644 tools/testing/selftests/bpf/progs/tailcall_bpf2bpf_hierarchy2.c
 create mode 100644 tools/testing/selftests/bpf/progs/tailcall_bpf2bpf_hierarchy3.c


base-commit: 0c14840ae36f8170f06c2fa768203ef5a8e389e1