LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Nathan Chancellor <nathan@kernel.org>
To: "Gustavo A. R. Silva" <gustavoars@kernel.org>,
	Linus Torvalds <torvalds@linux-foundation.org>
Cc: Kees Cook <keescook@chromium.org>,
	Nick Desaulniers <ndesaulniers@google.com>,
	Masahiro Yamada <masahiroy@kernel.org>,
	linux-kbuild@vger.kernel.org, linux-kernel@vger.kernel.org,
	clang-built-linux@googlegroups.com
Subject: Re: [PATCH] kbuild: Enable -Wimplicit-fallthrough for clang 14.0.0+
Date: Mon, 16 Aug 2021 21:20:33 -0700	[thread overview]
Message-ID: <80fa539a-b767-76ed-dafa-4d8d1a6b063e@kernel.org> (raw)
In-Reply-To: <20210817005624.1455428-1-nathan@kernel.org>


On 8/16/2021 5:56 PM, Nathan Chancellor wrote:
> Clang prior to 14.0.0 warns when a fallthrough annotation is in an
> unreachable spot, which can occur when IS_ENABLED(CONFIG_...) in a
> conditional statement prior to the fallthrough annotation such as
> 
>    if (IS_ENABLED(CONFIG_...))
>        break;
>    fallthrough;
> 
> which to clang looks like
> 
>    break;
>    fallthrough;
> 
> if CONFIG_... is enabled due to the control flow graph. Example of the
> warning in practice:
> 
> sound/core/pcm_native.c:3812:3: warning: fallthrough annotation in
> unreachable code [-Wimplicit-fallthrough]
>                  fallthrough;
>                  ^
> 
> Warning on unreachable annotations makes the warning too noisy and
> pointless for the kernel due to the nature of guarding some code on
> configuration options so it was disabled in commit d936eb238744 ("Revert
> "Makefile: Enable -Wimplicit-fallthrough for Clang"").
> 
> This has been resolved in clang 14.0.0 by moving the unreachable warning
> to its own flag under -Wunreachable-code, which the kernel will most
> likely never enable due to situations like this.
> 
> Enable -Wimplicit-fallthrough for clang 14+ so that issues such as the
> one in commit 652b44453ea9 ("habanalabs/gaudi: fix missing code in ECC
> handling") can be caught before they enter the tree.
> 
> Link: https://github.com/ClangBuiltLinux/linux/issues/236
> Link: https://github.com/llvm/llvm-project/commit/9ed4a94d6451046a51ef393cd62f00710820a7e8
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> ---
>   Makefile | 10 ++++++++--
>   1 file changed, 8 insertions(+), 2 deletions(-)
> 
> diff --git a/Makefile b/Makefile
> index c19d1638da25..91a4a80409e1 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -797,11 +797,17 @@ KBUILD_CFLAGS += -Wno-gnu
>   # source of a reference will be _MergedGlobals and not on of the whitelisted names.
>   # See modpost pattern 2
>   KBUILD_CFLAGS += -mno-global-merge
> +
> +# Warn about unmarked fall-throughs in switch statement.
> +# Clang prior to 14.0.0 warned on unreachable fallthroughs with
> +# -Wimplicit-fallthrough, which is unacceptable due to IS_ENABLED().
> +# https://bugs.llvm.org/show_bug.cgi?id=51094
> +ifeq ($(shell test $(CONFIG_CLANG_VERSION) -ge 140000; echo $$?),0)
> +KBUILD_CFLAGS += -Wimplicit-fallthrough
> +endif
>   else
>   
>   # Warn about unmarked fall-throughs in switch statement.
> -# Disabled for clang while comment to attribute conversion happens and
> -# https://github.com/ClangBuiltLinux/linux/issues/636 is discussed.
>   KBUILD_CFLAGS += $(call cc-option,-Wimplicit-fallthrough=5,)
>   endif
>   
> 
> base-commit: a2824f19e6065a0d3735acd9fe7155b104e7edf5
> 

Please do not apply this patch in its current form, as it does not 
properly credit Gustavo for all of the hard work he has done for 
enabling this warning.

Additionally, there should be some time for the CI systems to update 
their clang-14 builds, as the recent 0day report shows.

Cheers,
Nathan

  parent reply	other threads:[~2021-08-17  4:20 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-17  0:56 Nathan Chancellor
2021-08-17  4:16 ` kernel test robot
2021-08-17  4:20 ` Nathan Chancellor [this message]
2021-08-17  4:37   ` Linus Torvalds
2021-08-17  4:55     ` Nathan Chancellor
2021-08-17 18:03       ` Kees Cook
2021-08-17 18:25         ` Nathan Chancellor
2021-08-17 21:17           ` Masahiro Yamada
2021-08-17 21:33             ` Gustavo A. R. Silva
2021-08-17 23:06               ` Kees Cook
2021-08-17 23:23                 ` Nathan Chancellor
2021-08-17 23:40                   ` Gustavo A. R. Silva
2021-08-18  4:15                   ` Masahiro Yamada
2021-08-18  4:27                   ` Philip Li
2021-08-18  4:45                     ` Gustavo A. R. Silva
2021-08-18  7:31                       ` Philip Li
2021-08-18 12:12                   ` Mark Brown
2021-08-25 21:09             ` Nick Desaulniers

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=80fa539a-b767-76ed-dafa-4d8d1a6b063e@kernel.org \
    --to=nathan@kernel.org \
    --cc=clang-built-linux@googlegroups.com \
    --cc=gustavoars@kernel.org \
    --cc=keescook@chromium.org \
    --cc=linux-kbuild@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=masahiroy@kernel.org \
    --cc=ndesaulniers@google.com \
    --cc=torvalds@linux-foundation.org \
    --subject='Re: [PATCH] kbuild: Enable -Wimplicit-fallthrough for clang 14.0.0+' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).