> diff --git a/scripts/Makefile b/scripts/Makefile
> index 3434a82a119f0..366c6a35141a6 100644
> --- a/scripts/Makefile
> +++ b/scripts/Makefile
> @@ -51,6 +51,9 @@ endif
>
>  ifdef CONFIG_BUILDTIME_MCOUNT_SORT
>  HOSTCFLAGS_sorttable.o += -DMCOUNT_SORT_ENABLED
> +ifneq ($(filter-out 0,$(CONFIG_ARM64_FUNCTION_PREFIX_NOPS)),)
> +HOSTCFLAGS_sorttable.o += -DFUNCTION_PREFIX_BYTES=$(shell expr 4 \* 
> $(CONFIG_ARM64_FUNCTION_PREFIX_NOPS))
> +endif
>  endif

This isn't a bug, but would it be worth letting a configured 0 through
here (or noting why it deliberately keeps the historical 8-byte
tolerance), so the block doesn't read as honouring a value it actually
filters out? As written, when CONFIG_ARM64_FUNCTION_PREFIX_NOPS is set
to 0, the guard treats it the same as the symbol being absent, so both
fall back to sorttable.c's built-in 8. This is behaviour-preserving
today and avoids passing an empty value to expr, but once the symbol
lands with a default of 0 for the non-CALL_OPS case, the one configured
value that differs from the legacy constant is the one the guard
discards.

[ ... ]


---
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/33826319138

Reply via email to