Re: [PATCH] treewide: Use __printf not__attribute__((format(printf,...)))

[Kirill A. Shutemov]

On Thu, Aug 25, 2011 at 07:55:37PM -0700, Joe Perches wrote:
> Standardize the style for compiler based printf format verification.
> Standardized the location of __printf too.
> 
> Done via script and a little typing.
> 
> $ grep -rPl --include=*.[ch] -w "__attribute__" * | \
>   grep -vP "^(tools|scripts|include/linux/compiler-gcc.h)" | \
>   xargs perl -n -i -e 'local $/; while (<>) { s/\b__attribute__\s*\(\s*\(\s*format\s*\(\s*printf\s*,\s*(.+)\s*,\s*(.+)\s*\)\s*\)\s*\)/__printf($1, $2)/g ; print; }'
> 
> Completely untested...

This patch breaks ARCH=um (linux-next-20111011):

In file included from /home/kas/git/public/linux-next/arch/um/os-Linux/aio.c:17:0:
/home/kas/git/public/linux-next/arch/um/include/shared/user.h:26:17: error: expected declaration specifiers or â...â before numeric constant
/home/kas/git/public/linux-next/arch/um/include/shared/user.h:26:20: error: expected declaration specifiers or â...â before numeric constant
/home/kas/git/public/linux-next/arch/um/include/shared/user.h:29:17: error: expected declaration specifiers or â...â before numeric constant
/home/kas/git/public/linux-next/arch/um/include/shared/user.h:29:20: error: expected declaration specifiers or â...â before numeric constant
/home/kas/git/public/linux-next/arch/um/os-Linux/aio.c: In function âdo_aioâ:
/home/kas/git/public/linux-next/arch/um/os-Linux/aio.c:93:3: error: implicit declaration of function âprintkâ [-Werror=implicit-function-declaration]
cc1: some warnings being treated as errors

> 
> Signed-off-by: Joe Perches <joe@xxxxxxxxxxx>
> 
> ---
> 
> On Thu, 2011-08-25 at 18:07 -0700, Andrew Morton wrote:
> > On Thu, 25 Aug 2011 17:38:08 -0700 Joe Perches <joe@xxxxxxxxxxx> wrote:
> > > So if you really like it that much:
> > Well I don't particularly like it, personally.  But they're there, so
> > we either fully use them or fully unuse them, then remove them.
>  
> I don't mind one way or another, and I do
> like consistency, so I guess the __printf
> form is a bit better match to the other
> __attribute__ #defines.
> 
> I just don't have any particular desire
> to push it to anyone though.
> 
> Here it is, totally untested.

<skip/>

> diff --git a/arch/um/include/shared/user.h b/arch/um/include/shared/user.h
> index 293f7c7..e253af9 100644
> --- a/arch/um/include/shared/user.h
> +++ b/arch/um/include/shared/user.h
> @@ -23,14 +23,12 @@
>  #include <stddef.h>
>  #endif
>  
> -extern void panic(const char *fmt, ...)
> -	__attribute__ ((format (printf, 1, 2)));
> +extern __printf(1, 2) void panic(const char *fmt, ...);
>  
>  #ifdef UML_CONFIG_PRINTK
> -extern int printk(const char *fmt, ...)
> -	__attribute__ ((format (printf, 1, 2)));
> +extern __printf(1, 2) int printk(const char *fmt, ...);
>  #else
> -static inline int printk(const char *fmt, ...)
> +static inline __printf(1, 2) int printk(const char *fmt, ...)
>  {
>  	return 0;
>  }

-- 
 Kirill A. Shutemov
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/