linux-kernel - Re: i686 quirk for AMD Geode

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:	Mon, 16 Nov 2009 12:50:06 -0500
From:	lsorense@...lub.uwaterloo.ca (Lennart Sorensen)
To:	Alan Cox <alan@...rguk.ukuu.org.uk>
Cc:	Lennart Sorensen <lsorense@...lub.uwaterloo.ca>,
	Pavel Machek <pavel@....cz>, "H. Peter Anvin" <hpa@...or.com>,
	Matteo Croce <technoboy85@...il.com>,
	Ingo Molnar <mingo@...e.hu>, linux-kernel@...r.kernel.org
Subject: Re: i686 quirk for AMD Geode

On Fri, Nov 13, 2009 at 09:21:48PM +0000, Alan Cox wrote:
> > > Which usually means some form of write gathering is enabled or something
> > > thinks the device is write combining on the PCI bus. What does the PCI
> > > bus and the RCRR MTRR set look like ?
> > 
> > Well there is no /proc/mtrr.
> 
> Linux doesn't have an RCRR driver - so you need to dump the MSRs directly
> (see the Geode GX manual MSR list) and read out the msrs.

Hmm, so looking at the Geode SC1200 data book, it doesn't mention mtrr,
rcrr or msr.  I know the Geode LX has lots of MSR documentation.

Looking at the GX1 data book (which it seems the SC1200 refers to),
it does mention mtrr, as not supported.  I have managed to find something
called the region configuration range registers, which must be RCRR.

I think this is a dump of the 8 region MSRs:

0x00001810->0xbf848ba8b7f282e8
0x00001811->0xbfcb5328b7f932e8
0x00001812->0xbfe93418b7f712e8
0x00001813->0xbfbd59b8b7fb52e8
0x00001814->0xbfb5f868b7f3d2e8
0x00001815->0xbf815018b7ef42e8
0x00001816->0xbfc4c0d8b7f2a2e8
0x00001817->0xbfe1eea8b7eff2e8

This must of course mean I read them wrong because those are not the
right values.  Some bit ranges must be 0 and are not.

Trying rdmsr from msr-tools 1.2 gives me:

# ./rdmsr -x 0x00001810
rdmsr: CPU 0 cannot read MSR 0x00001810

Hmm, now what?

-- 
Len Sorensen
--
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/