linux-kernel - Re: [PATCH] Fix redundant judgment in WARN

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: <20251109090317.4261cff25a497bcc8a358a7d@linux-foundation.org>
Date: Sun, 9 Nov 2025 09:03:17 -0800
From: Andrew Morton <akpm@...ux-foundation.org>
To: Xie Yuanbin <qq570070308@...il.com>
Cc: nathan@...nel.org, nick.desaulniers+lkml@...il.com, morbo@...gle.com,
 justinstitt@...gle.com, masahiroy@...nel.org, jack@...e.cz,
 maninder1.s@...sung.com, linux-kernel@...r.kernel.org,
 llvm@...ts.linux.dev, will@...nel.org
Subject: Re: [PATCH] Fix redundant judgment in WARN_ONCE with clang

On Sun,  9 Nov 2025 16:37:15 +0800 Xie Yuanbin <qq570070308@...il.com> wrote:

> For c code:
> ```c
> extern int xx;
> void test(void)
> {
> 	if (WARN_ONCE(xx, "x"))
> 		__asm__ volatile ("nop":::);
> }
> ```
> 
> Clang will generate the following assembly code:
> ```assemble
> test:
> 	movl	xx(%rip), %eax // Assume xx == 0 (likely case)
> 	testl	%eax, %eax // judge once
> 	je	.LBB0_3    // jump to .LBB0_3
> 	testb	$1, test.__already_done(%rip)
> 	je	.LBB0_2
> .LBB0_3:
> 	testl	%eax, %eax // judge again
> 	je	.LBB0_5    // jump to .LBB0_5
> .LBB0_4:
> 	nop
> .LBB0_5:
> 	retq
> 	// omit
> ```
> 
> In the above code, `xx == 0` should be a likely case, but in this case,
> xx has been judged twice.
>
> ...
>
> --- a/include/linux/once_lite.h
> +++ b/include/linux/once_lite.h
> @@ -10,17 +10,17 @@
>  #define DO_ONCE_LITE(func, ...)						\
>  	DO_ONCE_LITE_IF(true, func, ##__VA_ARGS__)
>  
> -#define __ONCE_LITE_IF(condition)					\
> -	({								\
> -		static bool __section(".data..once") __already_done;	\
> -		bool __ret_cond = !!(condition);			\
> -		bool __ret_once = false;				\
> -									\
> -		if (unlikely(__ret_cond && !__already_done)) {		\
> -			__already_done = true;				\
> -			__ret_once = true;				\
> -		}							\
> -		unlikely(__ret_once);					\
> +#define __ONCE_LITE_IF(condition)						\
> +	({									\
> +		static bool __section(".data..once") __already_done;		\
> +		bool __ret_cond = !!(condition);				\
> +		bool __ret_once = false;					\
> +										\
> +		if (unlikely(__ret_cond) && unlikely(!__already_done)) {	\

Sure, !__already_done is unlikely here.

> +			__already_done = true;					\
> +			__ret_once = true;					\
> +		}								\
> +		unlikely(__ret_once);						\

It's a shame you messed with the whitespace.  And I don't think it was
necessary anyway.  Here's what it would have looked like:

--- a/include/linux/once_lite.h~fix-redundant-judgment-in-warn_once-with-clang
+++ a/include/linux/once_lite.h
@@ -16,7 +16,7 @@
 		bool __ret_cond = !!(condition);			\
 		bool __ret_once = false;				\
 									\
-		if (unlikely(__ret_cond && !__already_done)) {		\
+		if (unlikely(__ret_cond) && unlikely(!__already_done)) {\
 			__already_done = true;				\
 			__ret_once = true;				\
 		}							\
_