linux-kernel - Re: New x86 warning

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

Date:	Wed, 22 Apr 2009 09:01:57 +0200
From:	Ingo Molnar <mingo@...e.hu>
To:	Jeff Garzik <jeff@...zik.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>
Cc:	LKML <linux-kernel@...r.kernel.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...hat.com>,
	"H. Peter Anvin" <hpa@...or.com>, x86@...nel.org
Subject: Re: New x86 warning


* Jeff Garzik <jeff@...zik.org> wrote:

> On x86-32, this warning now appears for me in 2.6.30-rc3, and did 
> not appear in 2.6.29.
>
> drivers/acpi/acpica/tbfadt.c: In function 'acpi_tb_create_local_fadt':
> /spare/repo/linux-2.6/arch/x86/include/asm/string_32.h:75: warning:  
> array subscript is above array bounds

Last i checked it was a GCC bounds check bogosity. All attempts to 
work it around or annotate it sanely (without changing the assembly 
code) failed. (new ideas welcome)

The closest i came was the hacklet below to the assembly code. [with 
an intentionally corrupted patch header to make it harder to apply 
accidentally.]

The hacklet writes the fifth byte by writing two bytes from byte 
position 3.

> lspci, dmesg and .config attached.
>
> 	Jeff
>
> P.S.  It is unclear in MAINTAINERS whether x86@...nel.org should 
> be CC'd in addition to the other addresses listed, or as a 
> replacement for individual emails.

at your option. Cc:-ing maintainers directly can get faster 
treatment occasionally. Using the alias is shorter.

	Ingo

NOT-Signed-off-by-me:

+++ linux/arch/x86/include/asm/string_32.h
@@ -72,7 +72,7 @@ static __always_inline void *__constant_
 		return to;
 	case 5:
 		*(int *)to = *(int *)from;
-		*((char *)to + 4) = *((char *)from + 4);
+		*((short *)(char *)(to + 3)) = *((short *)(char *)(from + 3));
 		return to;
 	case 6:
 		*(int *)to = *(int *)from;
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/