Re: [kvm-devel] [PATCH 0/3] KVM clock, new iteration

[Anthony L. <an...@co...>]

Glauber de Oliveira Costa wrote:
> Hi folks,
>
> A lot of time has been elapsed since I last sent a version of kvm clock,
> with a lot of things in the middle, including my vacations and a nasty bug
> in the clock itself ;-)
>
> Here it is a new version. I'm following avi's last suggestion of using an msr
> to make it more migration-proof. If it's not approved, we can easily switch back
> to the hypercall mechanism. That's the easy part ;-)
>
> Clock updates are done every time we're scheduled. It is _not_ enough to check
> if we're changing cpus, because the tsc will change its frequency when the cpu is
> idle (if using the mwait idle). So after a long idle period, one cpu will end up stepping
> ahead causing a backward time (remember the bug I talked about? :-) )
>
> As I fell it ready for inclusion (which obviously does not mean it is perfect, nor that
> I don't value any further comments ), I'm also posting the userspace patches in a separate
> series that will follow shortly.
>
>   

This patch series is looking good.  I like the switch to an MSR.

Regards,

Anthony Liguori