> Fixes: d243b62b7bd3 ("powerpc64/bpf: Add support for bpf trampolines")
Does this patch fix a bug in d243b62b7bd3?
The commit message describes the change as moving the dummy_tramp_addr
literal from the top of the long branch stub to the bottom so a
disassembler can walk through the stub instructions without hitting
embedded data. The message states "the actual program layout is never
altered. Hence no effect on overall program size."
The 8-byte alignment invariant the changelog mentions was already
established by the immediately preceding commit in the series, 206e0a244d70
("powerpc/bpf: fix alignment of long branch trampoline address"), which
carries the same Fixes: tag plus Cc: stable. Before this patch the check
already ensured alignment:
ool_stub_sz = IS_ENABLED(CONFIG_PPC_FTRACE_OUT_OF_LINE) ? 16 : 12;
if (!image || !IS_ALIGNED((unsigned long)fimage + ctx->idx*4 + ool_stub_sz,
SZL))
EMIT(PPC_RAW_NOP());
where 16 is 4 OOL-stub instructions and 12 is 3, with the .long
immediately following them. This patch re-derives the same constant for the
new position (44/40) but does not repair a misalignment.
Keeping the Fixes: tag will feed this patch to AUTOSEL and stable even
though it appears to be a readability improvement that textually depends on
206e0a244d70 (the reworked comment and constant hunks will not apply
cleanly without it). The absence of Cc: stable here - present on both
predecessor commits in the series - suggests this patch was not intended
for backport either.
Should the Fixes: tag be dropped, or does it repair a specific observable
failure beyond the disassembler presentation described in the changelog?
---
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/32100929603
