[lkml]   [2012]   [Jun]   [30]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH -v2 0/2] x86, microcode: Reload ucode only per-system
Assuming there are no more objections to those, can we get them into,
say, tip:x86/microcode for a wider, linux-next testing?

Peter's stuff could go there too or through the perf tree - it's kinda
arbitrary though.


On Thu, Jun 21, 2012 at 02:07:15PM +0200, Borislav Petkov wrote:
> From: Borislav Petkov <>
> Ok, here's the second version with all review comments addressed. I've
> dropped the stable tag since this has been that way (and b0rked) since
> forever so stable rules don't apply.
> We can probably do single backports if distros want it.
> Changelog:
> -v1:
> Once upon a time there was this microcode reloading interface
> /sys/devices/system/cpu/cpuX/microcode/reload, where X is an online
> cpu on the system, which allowed the loading of microcode in a per-cpu
> manner.
> This had problems like having different ucode revisions on an otherwise
> homogeneous system and needed O(n^2) overhead when tracking minimum
> microcode revision per-core.
> So make this interface per-system so that it does microcode reloading on
> the whole system only.
> Single commit messages have more info too.
> The first patch has a stable tag which I'd like to see in stable but
> since it is not fixing a direct regression, I'd like to not push it
> upstream now but have it get tested in linux-next and go upstream during
> the next merge window from where it can trickle slowly to stable.
> Patches have been tested on all AMD families, it wouldn't hurt if it saw
> some Intel testing too, although it should just work.
> Holler if you see regressions/problems with it.
> Thanks.


Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
GM: Alberto Bozzo
Reg: Dornach, Landkreis Muenchen
HRB Nr. 43632 WEEE Registernr: 129 19551

 \ /
  Last update: 2012-06-30 20:21    [W:0.071 / U:0.412 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site