linux-kernel - Re: [PATCH] compiler-intel: fix wrong compiler barrier() macro

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, 29 Apr 2015 14:59:56 +0000
From:	mancha security <mancha1@...o.com>
To:	Pranith Kumar <bobby.prani@...il.com>
Cc:	Daniel Borkmann <daniel@...earbox.net>,
	"H. Peter Anvin" <hpa@...ux.intel.com>,
	LKML <linux-kernel@...r.kernel.org>,
	Ingo Molnar <mingo@...nel.org>
Subject: Re: [PATCH] compiler-intel: fix wrong compiler barrier() macro

On Wed, Apr 29, 2015 at 10:51:40AM -0400, Pranith Kumar wrote:
> Hi Daniel,
> 
> On Wed, Apr 29, 2015 at 10:42 AM, Daniel Borkmann <daniel@...earbox.net> wrote:
> > Cleanup commit 23ebdedc67e ("compiler-intel.h: Remove duplicate
> > definition") removed the double definition of __memory_barrier()
> > intrinsics.
> >
> > However, in doing so, it also removed the preceding #undef barrier,
> > meaning, the actual barrier() macro from compiler-gcc.h with inline
> > asm is still in place when __GNUC__ is provided.
> 
> When you use the Intel compilers, the barrier() definition will come
> from compiler.h and not compiler-gcc.h. That is what the commit
> message says in 73679e508201(your commit message has the wrong hash).
> I don't understand what problem you are seeing with this, can you
> please explain?
> 
> Thanks!

Hi Pranith.

The problem is that ICC defines __GNUC__ so barrier() gets defined
in compiler-gcc.h. Your commit removed an #undef from compiler-intel.h
so compiler.h will never define barrier to __memory_barrier().

--mancha

Content of type "application/pgp-signature" skipped