On Sat, 07 Sep 2013, Elimar Riesebieter wrote:
> * Elimar Riesebieter <riese...@lxtec.de> [2013-09-07 15:32 +0200]:
> > ACK, loaded as module. Thanks for the hint. But I can't find a note
> > on how to verify the new microcode is loaded?
>
> $ dmesg | grep micro
> microcode: CPU0 sig=0x10676, pf=0x80, revision=0x610
> microcode: CPU1 sig=0x10676, pf=0x80, revision=0x610
> microcode: Microcode Update Driver: v2.00 <tig...@aivazian.fsnet.co.uk>,
> Peter Oruba
Which means not updated (hint: no "updated" messages in the log).
Now, the intel-microcode changelog has this to say about sig 0x10686:
+ Updated Microcodes:
sig 0x00010676, pf mask 0x01, 2010-09-29, rev 0x060f, size 4096
sig 0x00010676, pf mask 0x04, 2010-09-29, rev 0x060f, size 4096
sig 0x00010676, pf mask 0x10, 2010-09-29, rev 0x060f, size 4096
sig 0x00010676, pf mask 0x40, 2010-09-29, rev 0x060f, size 4096
sig 0x00010676, pf mask 0x80, 2010-09-29, rev 0x060f, size 4096
Your pf=0x80 is listed in there, rev 0x60f. Your BIOS actually has *newer*
microcode than what is present in intel-microcode. I cross-checked with
iucode-tool, and that's correct.
You might still benefit from intel-microcode if you have it installed and
the non-free/proposed-updates repositories set up, as a future microcode
update from Intel might have a newer version than 0x610, and you'd get it
automatically along with any other package updates.
PS: note that *any* pfmask with bit 7 set (0x80 = 10000000b) would apply to
your processor. The rule is: if the signatures are the same, and "(pf &
pfmask) == pf", the microcode is suitable for that processor.
--
"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, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130907215521.gc23...@khazad-dum.debian.net
