Re: [PATCH 2/2] x86, microcode: Make reload interface per system

[Henrique de Moraes Holschuh <hmh@hmh.eng.br>]

On Fri, 22 Jun 2012, H. Peter Anvin wrote:
> On 06/22/2012 09:26 AM, Nix wrote:
> > On 20 Jun 2012, Borislav Petkov uttered the following:
> >> I know, right. Whose idea was it to do it like that I don't know.
> >> AFAICT, Intel delivers ucode as a single blob too, so why split it? The
> >> driver picks out the right blob anyway.
> > 
> > Only if supplied over the old interface. Over the new interface,
> > we just see
> > 
> >     microcode: error! Bad data in microcode data file
> > 
> > in the log. So clearly the driver doesn't know how to split up the
> > microcode.bin that Intel provides, and (until Henrique's iucode-tool or
> > something like it becomes ubiquitous) the old interface, and
> > microcode_ctl, cannot be removed.
> 
> Even more importantly, to do early microcode updates we need to stash
> away not just the current CPU's microcode but any compatible CPU's
> microcode, so just loading a single one is not going to work...

Well, THAT is rather difficult with Intel.  You really have no choice
other than packing all microcodes, or just the ones for the current
system.  For all microcodes, it will compresses to around 350KiB with
gzip.

-- 
  "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