[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [Xen-devel] [PATCH] x86/viridian: Add Partition Reference Time enlightenment
>>> On 25.09.14 at 18:18, <Paul.Durrant@xxxxxxxxxx> wrote: >> -----Original Message----- >> From: Jan Beulich [mailto:JBeulich@xxxxxxxx] >> Sent: 25 September 2014 17:11 >> To: Paul Durrant >> Cc: Christoph Egger; Ian Campbell; Ian Jackson; Stefano Stabellini; xen- >> devel@xxxxxxxxxxxxxxxxxxxx; Keir (Xen.org) >> Subject: Re: [PATCH] x86/viridian: Add Partition Reference Time >> enlightenment >> >> >>> On 25.09.14 at 17:50, <paul.durrant@xxxxxxxxxx> wrote: >> > The presence of the partition reference time enlightenment persuades >> newer >> > versions of Windows to prefer the TSC as their primary time source. Hence, >> > if rdtsc is not being emulated and is invariant then many vmexits (for >> > alternative time sources such as the HPET or reference counter MSR) can >> > be avoided. >> > >> > If the VM is migrated to a host where TSC is no longer invariant and/or >> > TSC is emulated then the enlightenment is disabled. >> >> Which means performance of a guest post migration can change >> dramatically. Is that really a good thing? > > I guess that's debatable. It is possible to use this enlightenment to scale > TSC, which means that TSC emulation could be avoided even if the tsc_khz > changes, but I still need to figure out how best to do that part. Nothing can > be done in the case of migration to a non iTSC host though. > I could leave this out the default enlightenment set for now. Yes please, perhaps with a brief code comment as to why. Jan _______________________________________________ Xen-devel mailing list Xen-devel@xxxxxxxxxxxxx http://lists.xen.org/xen-devel
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |