Re: [Xen-devel] GPU passthrough performance regression in >4GB vms due to XSA-60 changes

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 19.05.14 at 19:36, <tim@xxxxxxx> wrote:
> At 16:24 +0100 on 19 May (1400513093), Jan Beulich wrote:
>> Independently of this - Tim, do you think this EPT misconfig stuff can
>> already be considered mature enough for backporting? It's been in a
>> little over a month, and considering that we're using formally
>> undefined behavior here, I'd rather view this as not yet a backport
>> candidate.
> 
> I agree -- it's a pretty major set of changes and I don't think it's
> ready for a stable branch yet.   Of course, it depends what the
> alternatives are...

If you go back in the thread, you'll find that the only alternative
found so far would be to split hvmloader's pci_setup() and
cacheattr_init(), with the downside of this becoming a stable
branch only change (and the uncertainty of it having some
unexpected side effect).

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel