Re: [bpf v8 5/7] powerpc64/bpf: fix compare instruction emitted for tailcall
From: bot+bpf-ci@kernel.org
Date: 2026-06-16 13:55:28
Also in:
bpf, linux-kselftest, stable
The tail_call_info field can contain either a scalar counter value or a 64-bit pointer to the counter, using a 32-bit compare (cmplwi) only checks the lower 32 bits, which can lead to incorrect comparisions when location of counter is near 4GB boundary. Use instruction cmpldi for accurate comparision in all cases.
The change now uses cmpldi on ppc64 and cmplwi on ppc32 through the new PPC_RAW_CMPLLI macro, so should this sentence read "Use cmpldi/cmplwi instruction" rather than "Use instruction cmpldi"? This wording was raised in the v7 review by Hari Bathini [off-list ref]: https://lore.kernel.org/bpf/5c095e3d-d554-4506-9dbc-96d568b4be5e@linux.ibm.com/ (local) This should be "Use cmpldi/cmplwi instruction" instead of "Use instruction cmpldi" The technical part of that feedback (defining PPC_RAW_CMPLLI rather than using PPC_RAW_CMPLDI directly in the shared file) appears to be addressed in this version, but the wording still reads "Use instruction cmpldi". --- AI reviewed your patch. Please fix the bug or email reply why it's not a bug. See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md CI run summary: https://github.com/kernel-patches/bpf/actions/runs/27620315552