linux-kernel - Re: Linux 2.6.27.27

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]

Message-ID: <alpine.LNX.1.10.0907211154480.29729@bizon.gios.gov.pl>
Date:	Tue, 21 Jul 2009 12:16:05 +0200 (CEST)
From:	Krzysztof Oledzki <olel@....pl>
To:	Linus Torvalds <torvalds@...ux-foundation.org>
cc:	Greg KH <gregkh@...e.de>, linux-kernel@...r.kernel.org,
	Andrew Morton <akpm@...ux-foundation.org>, stable@...nel.org,
	lwn@....net
Subject: Re: Linux 2.6.27.27



On Tue, 21 Jul 2009, Krzysztof Oledzki wrote:

>
>
> On Mon, 20 Jul 2009, Linus Torvalds wrote:
>
>> 
>> 
>> On Mon, 20 Jul 2009, Krzysztof Oledzki wrote:
>>> 
>>> No problem. Please let me know what should I do to help tracking this 
>>> issue.
>> 
>> Can you build two kernels: one with -fwrapv, and one with
>> -fno-strict-overflow, and then verify that
>> 
>> - they are otherwise identical (ie exact same source code, same compiler
>>   etc)
>> 
>> - verify that yes, the -fwrapv kernel works, the other does not. Just to
>>   avoid the confusion that obviously exists with Debian/sid binutils
>>   upgrades that _also_ happens result in nonbootable kernels.
>> 
>> - upload the 'vmlinux' images somewhere (I'm not sure what the limits for
>>   binary attachments are at the kernel bugzilla, but that would be the
>>   logical place)
>> 
>> In fact, it would be nice to have a third "identical" kernel build, except
>> with neither -fwrapv/-fno-strict-overflow.
>
> OK. Right now I'm building the three kernels you asked for 
> (fwrapv/fno-strict-overflow/none). I'll test them and upload vmlinux images.

OK, there are three kernels, exactly as you requested:

http://noc.axelspringer.pl/no-strict-overflow-vs-wrapv/vmlinux-fno-strict-overflow.bz2 (Hangs)
http://noc.axelspringer.pl/no-strict-overflow-vs-wrapv/vmlinux-fwrapv.bz2 (OK)
http://noc.axelspringer.pl/no-strict-overflow-vs-wrapv/vmlinux-fnone.bz2 (OK)

# grep -A1 "can't wrap" linux-2.6.27.27-*/Makefile
linux-2.6.27.27-fno-strict-overflow/Makefile:# disable invalid "can't wrap" optimzations for signed / pointers
linux-2.6.27.27-fno-strict-overflow/Makefile-KBUILD_CFLAGS      += $(call cc-option,-fno-strict-overflow)
--
linux-2.6.27.27-fwrapv/Makefile:# disable invalid "can't wrap" optimzations for signed / pointers
linux-2.6.27.27-fwrapv/Makefile-KBUILD_CFLAGS   += $(call cc-option,-fwrapv)
--
linux-2.6.27.27-fnone/Makefile:# disable invalid "can't wrap" optimzations for signed / pointers
linux-2.6.27.27-fnone/Makefile-#KBUILD_CFLAGS   += $(call cc-option,-fno-strict-overflow)

Plwase note the third one has the KBUILD_CFLAGS commented.

Kernels are identical and are compiled from the same config, on the same 
server with gcc-4.2.4, binutils-2.19. There is no ccache installed and the 
kernels are not patched with any additonal patches - just vanilla 
linux-2.6.27.27.

Screenshot from the hanging kernel (-fno-strict-overflow):
  http://noc.axelspringer.pl/no-strict-overflow-vs-wrapv/2.6.27.27-hang.png

Dmesg from a bootable kernel:
  http://noc.axelspringer.pl/no-strict-overflow-vs-wrapv/dmesg

Best regards,

 				Krzysztof Olędzki