Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20230807-missing_proto-v1-1-7f566b7ba5ca@google.com> <20230807165526.GA2744968@dev-arch.thelio-3990X>
In-Reply-To: <20230807165526.GA2744968@dev-arch.thelio-3990X>
From:   Nick Desaulniers <ndesaulniers@google.com>
Date:   Mon, 7 Aug 2023 10:02:37 -0700
Message-ID: <CAKwvOd=8kxkD9p+WW-F047ShN=r32SLYYfpgZhyDw3BXTDd-KA@mail.gmail.com>
Subject: Re: [PATCH] Makefile.extrawarn: enable -Wmissing-variable-declarations
 for W=1
To:     Nathan Chancellor <nathan@kernel.org>
Cc:     Masahiro Yamada <masahiroy@kernel.org>,
        Nicolas Schier <nicolas@fjasle.eu>, Tom Rix <trix@redhat.com>,
        linux-kbuild@vger.kernel.org, linux-kernel@vger.kernel.org,
        llvm@lists.linux.dev
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Mon, Aug 7, 2023 at 9:55=E2=80=AFAM Nathan Chancellor <nathan@kernel.org=
> wrote:
>
> On Mon, Aug 07, 2023 at 09:50:32AM -0700, Nick Desaulniers wrote:
> > I noticed Tom was sending patches where smatch was recommending
> > annotating functions as static when no previous declaration existed.
> > Surely the compiler could make such recommendations as well, I thought.
> >
> > Looks like -Wmissing-variable-declarations can make such
> > recommendations.
> >
> > GCC just added support for this warning (gcc 14.1.0 will ship with
> > support), and all versions of clang relevant to building the kernel
> > support this flag.
> >
> > Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
>
> Any idea how many instances of this there are? At least x86_64 defconfig
> fails immediately with:

$ make LLVM=3D1 -j128 W=3D1 allnoconfig 2>log.txt
$ grep -r Wmissing-variable-declarations log.txt | sort -u | wc -l
140
Though that's not perfectly filtered due to overlapping output due to -j 12=
8.

>
>   In file included from scripts/mod/devicetable-offsets.c:3:
>   In file included from include/linux/mod_devicetable.h:14:
>   In file included from include/linux/uuid.h:11:
>   In file included from include/linux/string.h:20:
>   In file included from arch/x86/include/asm/string.h:5:
>   In file included from arch/x86/include/asm/string_64.h:6:
>   In file included from include/linux/jump_label.h:112:
>   In file included from arch/x86/include/asm/jump_label.h:7:
>   arch/x86/include/asm/asm.h:208:24: error: no previous extern declaratio=
n for non-static variable 'current_stack_pointer' [-Werror,-Wmissing-variab=
le-declarations]
>     208 | register unsigned long current_stack_pointer asm(_ASM_SP);
>         |                        ^
>   arch/x86/include/asm/asm.h:208:10: note: declare 'static' if the variab=
le is not intended to be used outside of this translation unit
>     208 | register unsigned long current_stack_pointer asm(_ASM_SP);
>         |          ^

Hmm...I wonder if clang and gcc should ignore variables with register
storage for this warning.

>   1 error generated.
>
> > ---
> >  scripts/Makefile.extrawarn | 1 +
> >  1 file changed, 1 insertion(+)
> >
> > diff --git a/scripts/Makefile.extrawarn b/scripts/Makefile.extrawarn
> > index 40cd13eca82e..617739eb84e2 100644
> > --- a/scripts/Makefile.extrawarn
> > +++ b/scripts/Makefile.extrawarn
> > @@ -32,6 +32,7 @@ KBUILD_CFLAGS +=3D $(call cc-option, -Wunused-but-set=
-variable)
> >  KBUILD_CFLAGS +=3D $(call cc-option, -Wunused-const-variable)
> >  KBUILD_CFLAGS +=3D $(call cc-option, -Wpacked-not-aligned)
> >  KBUILD_CFLAGS +=3D $(call cc-option, -Wstringop-truncation)
> > +KBUILD_CFLAGS +=3D $(call cc-option, -Wmissing-varibale-declarations)
>                                                 ^ variable
> >  # The following turn off the warnings enabled by -Wextra
> >  KBUILD_CFLAGS +=3D -Wno-missing-field-initializers
> >  KBUILD_CFLAGS +=3D -Wno-sign-compare
> >
> > ---
> > base-commit: 52a93d39b17dc7eb98b6aa3edb93943248e03b2f
> > change-id: 20230807-missing_proto-0cb90ec6454c
> >
> > Best regards,
> > --
> > Nick Desaulniers <ndesaulniers@google.com>
> >


--=20
Thanks,
~Nick Desaulniers