Re: [PATCH 6/8] x86, microcode, intel: total_size is valid only when data_size != 0

[Henrique de Moraes Holschuh]

On Mon, 28 Jul 2014, Borislav Petkov wrote:
> On Fri, Jul 25, 2014 at 04:04:11PM -0300, Henrique de Moraes Holschuh wrote:
> > As far as I can tell, Intel has always filled the reserved fields in the
> > microcode data files with zeros.
> 
> Yes, and this is the statement I was looking for. IF(!) this really
> is the case, then we don't have anything to worry about because new
> kernel with old container will look at ->datasize, see 0 and return
> DEFAULT_UCODE_TOTALSIZE and we're fine.

We're fine AFAICT.

If there's an official microcode update out there that can trigger this
bug, it didn't come from a Linux distro, from Intel's public download site
or from urbanmyth.org.

I've checked 346 unique microcode updates for 189 unique Intel processors,
and none would cause problems.  That test corpus covers processors as old
as cpuid 0x611.  It is used to check iucode-tool for regressions.

> I've updated the patch with your additional explanation.

Thank you!

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/