linux-kernel - Re: [Security] [PATCH v3 0/4] x86: clear XD

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: <20101109235314.GD5876@outflux.net>
Date:	Tue, 9 Nov 2010 15:53:14 -0800
From:	Kees Cook <kees.cook@...onical.com>
To:	Alan Cox <alan@...rguk.ukuu.org.uk>
Cc:	x86@...nel.org, linux-kernel@...r.kernel.org
Subject: Re: [Security] [PATCH v3 0/4] x86: clear XD_DISABLED flag on Intel
 to regain NX

Hi Alan,

On Tue, Nov 09, 2010 at 10:50:00PM +0000, Alan Cox wrote:
> On Tue, 9 Nov 2010 10:56:04 -0800
> Kees Cook <kees.cook@...onical.com> wrote:
> 
> > On Tue, Nov 09, 2010 at 06:31:42PM +0000, Alan Cox wrote:
> > > > In a review[1] of reported hardware being used by Ubuntu bug reporters,
> > > > almost 10% of systems had an incorrectly configured BIOS, leaving their
> > > > systems unable to use the NX features of their CPU.
> > > 
> > > Ouch
> > 
> > Yeah :(
> > 
> > > > This change will clear the MSR_IA32_MISC_ENABLE_XD_DISABLE bit so that NX
> > > > cannot be inappropriately controlled by the BIOS on Intel CPUs.  If, under
> > > > very strange hardware configurations, NX actually needs to be disabled,
> > > > "noexec=off" can be used to restore the prior behavior.
> > > 
> > > Have you done an audit of CPU errata to ensure that none of these cases
> > > are ones where the BIOS has disabled it to avoid an erratum. I'd hate to
> > > turn it on and find out the BIOS actually turned it off for good reason !
> > 
> > Where can I find those details?
> 
> In the errata manuals for each processor released by the relevant
> vendors. As they are aggregated into families and they have an index in
> the front it shouldn't take too long to check the ones your 10% scan
> found.

Only Intel has this problem (it's the only CPU that defines this MSR), so
that'll reduce it. But Google is of no help to me on this; where can I find
these documents?

> I'm not specifically aware of any but I do know for example that there
> are other CPU BIOS disablable features that some systems disable in the
> BIOS for good reason (an ancient example being the original Pentium
> REP MOVS optimisation on some steppings)

By definition there should be fewer errata CPUs than those than need their
BIOS ignored, so I still think this patch makes sense (especially since it
can trivially be worked around with the noexec=off boot option).

-Kees

-- 
Kees Cook
Ubuntu Security Team
--
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/