public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Nathan Chancellor <nathan@kernel.org>
To: hamjin <jinham@qq.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH] lib: typecheck: initialize const variable to avoid Clang warning
Date: Wed, 25 Mar 2026 19:04:48 -0700	[thread overview]
Message-ID: <20260326020448.GA1294592@ax162> (raw)
In-Reply-To: <tencent_274FD06231C70768C2C28FD13673A24C2B08@qq.com>

On Sun, Mar 15, 2026 at 04:52:48PM +0800, hamjin wrote:
> Building the kernel with newer Clang versions triggers
> -Wdefault-const-init-var-unsafe in the typecheck() macro:
> 
>   error: default initialization of an object of type 'const unsigned long'
>   leaves the object uninitialized [-Wdefault-const-init-var-unsafe]
> 
> The warning originates from the following declaration in typecheck():
> 
>     typeof(x) __dummy2;
> 
> When 'x' has a const-qualified type (e.g. 'const unsigned long'),
> Clang warns that the variable is declared const but not initialized.
> With -Werror enabled this causes the build to fail.
> 
> Initialize the temporary variable to zero to silence the warning:
> 
>     typeof(x) __dummy2 = (typeof(x))0;
> 
> This variable is only used for compile-time type checking and its
> value is never read, so the initialization has no functional impact
> on the generated code.
> 
> Fixes: e0deaff47090 ("split the typecheck macros out of include/linux/kernel.h")
> Signed-off-by: hamjin <jinham@qq.com>
> ---
>  include/linux/typecheck.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/include/linux/typecheck.h b/include/linux/typecheck.h
> index 46b15e2aaefb..09f280da5b52 100644
> --- a/include/linux/typecheck.h
> +++ b/include/linux/typecheck.h
> @@ -8,7 +8,7 @@
>   */
>  #define typecheck(type,x) \
>  ({	type __dummy; \
> -	typeof(x) __dummy2; \
> +	typeof(x) __dummy2 = (typeof(x))0; \
>  	(void)(&__dummy == &__dummy2); \
>  	1; \
>  })
> -- 
> 2.53.0
> 

As you can see from the build reports, this does not work. Are you
missing commit d0afcfeb9e38 ("kbuild: Disable
-Wdefault-const-init-unsafe") in your tree? Or does this appear in
somewhere that uses its own KBUILD_CFLAGS, in which case a fix like

  5ba35a6c13ff ("s390/boot: Add -Wno-default-const-init-unsafe to KBUILD_CFLAGS")
  b4780fe4ddf0 ("s390/purgatory: Add -Wno-default-const-init-unsafe to KBUILD_CFLAGS")

will be needed? I do not see any instances of this warning in Linus's
tree unless I am missing some configuration in my build tests.

Cheers,
Nathan

      parent reply	other threads:[~2026-03-26  2:04 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-15  8:52 [PATCH] lib: typecheck: initialize const variable to avoid Clang warning hamjin
2026-03-15 15:12 ` kernel test robot
2026-03-25 12:07 ` kernel test robot
2026-03-25 12:52 ` kernel test robot
2026-03-26  2:04 ` Nathan Chancellor [this message]

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=20260326020448.GA1294592@ax162 \
    --to=nathan@kernel.org \
    --cc=jinham@qq.com \
    --cc=linux-kernel@vger.kernel.org \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox