Re: [Xen-devel] [PATCH 0/2] MTRR/PAT virtualization

[Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>]

On 9/10/07 10:23, "Michael A Fetterman" <Michael.Fetterman@xxxxxxxxxxxx>
wrote:

>>> But, yes this is really a good point. So we need to do WBINVD when VP
>>> migrates (of course for pass-through domain only), while the prefered
>>> approach is to pin VCPU on pCPUs.
>> 
>> Or WBINVD all CPUs when a VCPU executes WBINVD. Or explicitly track dirty
>> caches for each vCPU.
> 
> I shutter to think of allowing a guest to cause a WBINVD.
> 
> In modern systems (8M+ of cache, etc), it can take 4+ milliseconds to execute.
> I dare say it could, in a worst case scenerio, be even worse if you
> did it on multiple
> CPUs or hyperthreads at once.  And the cpu is non-interruptable the entire
> time.

It only needs to be allowed for guests with direct hardware access. Those
WBINVD instructions in AGP drivers and the like are there for a reason, I
suspect.

 -- Keir



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel