linux-kernel - Re: [PATCH] m68k: libgcc: Fix lvalue abuse in umul

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <c59ce4ca-0647-45d1-ae83-9644f1fefd27@linux-m68k.org>
Date: Thu, 9 Jan 2025 09:03:09 +1000
From: Greg Ungerer <gerg@...ux-m68k.org>
To: Geert Uytterhoeven <geert@...ux-m68k.org>
Cc: Arnd Bergmann <arnd@...db.de>,
 Linus Torvalds <torvalds@...ux-foundation.org>,
 linux-m68k@...ts.linux-m68k.org, linux-kernel@...r.kernel.org,
 kernel test robot <lkp@...el.com>
Subject: Re: [PATCH] m68k: libgcc: Fix lvalue abuse in umul_ppmm()

Hi Geert,

On 9/1/25 03:22, Geert Uytterhoeven wrote:
>      lib/muldi3.c:53:28: warning: asm output is not an lvalue
>      lib/muldi3.c:53:28: warning: asm output is not an lvalue
>      lib/muldi3.c:53:28: error: not addressable
>      lib/muldi3.c:53:28: warning: generating address of non-lvalue (11)
>      lib/muldi3.c:53:28: warning: generating address of non-lvalue (11)
> 
> Fix the lvalue warnings by replacing the casts on the output operands by
> intermediate variables of the right type.
> 
> Fix the "not addressable" error by replacing the cast on the second
> input operand by an intermediate variable, too.  Treat the other input
> operand the same for consistency.
> 
> Reported-by: kernel test robot <lkp@...el.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202501030516.uZrwnuQQ-lkp@intel.com/
> Suggested-by: Linus Torvalds <torvalds@...ux-foundation.org>
> Signed-off-by: Geert Uytterhoeven <geert@...ux-m68k.org>

Thanks for cleaning this up.

Acked-by: Greg Ungerer <gerg@...ux-m68k.org>

Regards
Greg



> ---
> To be queued in the m68k for-v6.14 branch.
> Not folding into commit e96424b86d5098f4 ("m68k: Use kernel's generic
> muldi3 libgcc function"), as the issue was pre-existing.
> ---
>   arch/m68k/include/asm/libgcc.h | 17 ++++++++++++-----
>   1 file changed, 12 insertions(+), 5 deletions(-)
> 
> diff --git a/arch/m68k/include/asm/libgcc.h b/arch/m68k/include/asm/libgcc.h
> index 1cce6d130d805368..27e17195bd7be41c 100644
> --- a/arch/m68k/include/asm/libgcc.h
> +++ b/arch/m68k/include/asm/libgcc.h
> @@ -10,11 +10,18 @@
>    * will fallback to using the C-coded version of umul_ppmm().
>    */
>   #define umul_ppmm(w1, w0, u, v)				\
> -	__asm__ ("mulu%.l %3,%1:%0"			\
> -		 : "=d" ((unsigned long)(w0)),		\
> -		   "=d" ((unsigned long)(w1))		\
> -		 : "%0" ((unsigned long)(u)),		\
> -		   "dmi" ((unsigned long)(v)))
> +	do {						\
> +		unsigned long __u = (u), __v = (v);	\
> +		unsigned long __w0, __w1;		\
> +							\
> +		__asm__ ("mulu%.l %3,%1:%0"		\
> +			 : "=d" (__w0),			\
> +			   "=d" (__w1)			\
> +			 : "%0" (__u),			\
> +			   "dmi" (__v));		\
> +							\
> +		(w0) = __w0; (w1) = __w1;		\
> +	} while (0)
>   #endif /* !CONFIG_CPU_HAS_NO_MULDIV64 */
>   
>   #endif /* __ASM_M68K_LIBGCC_H */