Re: [PATCH 3/3] Revert "Kbuild, lto, workaround: Don't warn for initcall_reference in modpost"

From: Nick Desaulniers
Date: Tue Aug 02 2022 - 14:04:36 EST


On Sat, Jul 30, 2022 at 10:37 AM Masahiro Yamada <masahiroy@xxxxxxxxxx> wrote:
>
> This reverts commit 77ab21adae509c5540956729e2d03bc1a59bc82a.
>
> That commit was 8 years old, and it said "This is a workaround".
> If this is needed for GCC LTO, it should be added in a proper way.
>
> Signed-off-by: Masahiro Yamada <masahiroy@xxxxxxxxxx>

Please don't forget to cc the author & reviewers for a patch when
submitting a revert.

+ Jiri in case a patch needs to be carried in any downstream trees for
re-application.

Reviewed-by: Nick Desaulniers <ndesaulniers@xxxxxxxxxx>


> ---
>
> scripts/mod/modpost.c | 3 ---
> 1 file changed, 3 deletions(-)
>
> diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c
> index c6a055c0291e..a8ee27496da7 100644
> --- a/scripts/mod/modpost.c
> +++ b/scripts/mod/modpost.c
> @@ -1462,9 +1462,6 @@ static void default_mismatch_handler(const char *modname, struct elf_info *elf,
> from = find_elf_symbol2(elf, r->r_offset, fromsec);
> fromsym = sym_name(elf, from);
>
> - if (strstarts(fromsym, "reference___initcall"))
> - return;
> -
> tosec = sec_name(elf, get_secindex(elf, sym));
> to = find_elf_symbol(elf, r->r_addend, sym);
> tosym = sym_name(elf, to);
> --
> 2.34.1
>


--
Thanks,
~Nick Desaulniers