Re: [PATCH] kvmclock: Ensure time in migration never goes backward

[Alexander Graf <agraf@suse.de>]

On 06.05.14 01:23, Marcelo Tosatti wrote:
> Hi Alexander,
>
> On Mon, May 05, 2014 at 03:51:22PM +0200, Alexander Graf wrote:
>> When we migrate we ask the kernel about its current belief on what the guest
>> time would be.
> KVM_GET_CLOCK which returns the time in "struct kvm_clock_data".

Yes :)

>
>> However, I've seen cases where the kvmclock guest structure
>> indicates a time more recent than the kvm returned time.
> More details please:
>
> 1) By what algorithm you retrieve
> and compare time in kvmclock guest structure and KVM_GET_CLOCK.
> What are the results of the comparison.
> And whether and backwards time was visible in the guest.

I've managed to get my hands on a broken migration stream from Nick. 
There I looked at the curr_clocksource structure and saw that the last 
seen time on the kvmclock clock source was greater than the value that 
the kvmclock device migrated.

> 2) What is the host clocksource.
>
> The test below is not a good one because:
>
> T1) KVM_GET_CLOCK (save s->clock).
> T2) save env->tsc.
>
> The difference in scaled time between T1 and T2 is larger than 1
> nanosecond, so the
>
> (time_at_migration > s->clock)
>
> check is almost always positive (what matters though is whether
> time backwards event can be seen reading kvmclock in the guest).

Yeah, at first I was thinking it makes sense to just not use the 
KVM_GET_CLOCK value at all because it's redundant to the in-memory 
values. Maybe that is the better alternative after all.


Alex