>>> On 15.06.15 at 10:57, <ian.campb...@citrix.com> wrote: > So 3/20 = 15% failure rate (fiano0: 1/11=9%; fiano1: 2/9=22%). Which is > better than the ~50% seen at the start of this thread, so it is worth > applying the ucode update I think (and it would have been regardless the > right thing to do), > > I do think a 15-20% failure rate might be worthy of further > investigation by Intel too, since the failure rate with no-apicv was > 1/13 = 7% (fiano0: 1/7=14%, fiano1: 0/6=0%), although those numbers are > less significant due to fewer runs.
I fully agree; even the remaining 7% should be looked into, provided they can be reproduced by them on sufficiently similar hardware. Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
