On 31.03.2011, at 18:11, Scott Wood wrote: > On Thu, 31 Mar 2011 12:11:48 +0200 > Alexander Graf <agraf@xxxxxxx> wrote: > >>>>>>>>> On 31.03.2011, at 05:21, Liu Yu-B13201 wrote: >>>>>>>>> >>>>>>>>>> >>>>>>>>>> I think the patch miss the bit to handle the case that >>>>>>>>>> if guest clear the MSR_SPE. > > Doh. > >>>>>> So that MSR[SPE] should always be traped. >>>>> >>>>> It doesn't have to be trapped - we can enable it lazily on >>>>> SPE_UNAVAIL :). Of course we should also explicitly enable it >>>>> on set_msr. > > I don't see any benefit to doing this, if we're going to take a trap one > way or another. Since we need to trap on clearing it anyway, might as well > just keep things simple and have SPE always be a trapworthy MSR bit. Agreed. Please make sure to enable it in advance on set_msr then though, as we'd otherwise get 2 exits to enable SPE. Alex -- To unsubscribe from this list: send the line "unsubscribe kvm-ppc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
