Re: [Xen-devel] Re: [GIT PULL] xen /proc/mtrr implementation

[Gerd Hoffmann]

On 05/19/09 17:24, Ingo Molnar wrote:
> the Xen hypervisor can simply repeat all requests (i.e. not care at
> all about the fact that a guest does these modifications on all CPUs
> it sees), or realize that the modification has already been done and
> skip it.

Could be done, yes.  It still feels wrong that wrmsr(mtrr) works 
slightly different on xen and on native.  And it wouldn't work on 
existing Xen deployments as the Xen hypervisor doesn't support that today.

> > > > > Yeah, the third one is to not touch MTRRs after bootup and use PAT.

> That's a really old CPU, but even Coppermine has PAT support in the
> CPU. You need to go back to things like P5 200 MHz CPUs to find
> PAT-less CPUs.

Linux shouln't say "PAT not supported by CPU." then.

Also it doesn't make sense to me to handle things differently on native 
and xen.  While it might make sense to deprecate mtrrs in favor of PAT 
(don't know enougth about all the different cpus in the wild to justify 
that) I don't think it makes sense to do that for xen only.  Native 
should declare mtrrs obsolete as well.

cheers,
  Gerd

--
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/