Re: [PATCH 1/5] Add a global synchronization point for pvclock

[Avi Kivity <avi@redhat.com>]

On 04/20/2010 03:59 PM, Glauber Costa wrote:
>
>> Might be due to NMIs or SMIs interrupting the rdtsc(); ktime_get()
>> operation which establishes the timeline.  We could limit it by
>> having a loop doing rdtsc(); ktime_get(); rdtsc(); and checking for
>> some bound, but it isn't worthwhile (and will break nested
>> virtualization for sure).  Better to have the option to calibrate
>> kvmclock just once on machines with
>> X86_FEATURE_NONSTOP_TRULY_RELIABLE _TSC_HONESTLY.
>>      
> For the record, we can only even do that in those machines. If we try to update
> time structures only once in machines with the
> X86_FEATURE_TSC_SAYS_IT_IS_OKAY_BUT_IN_REALITY_IS_NOT_OKAY feature flag, guests
> won't even boot.
>
> We can detect that, and besides doing calculation only once, also export some
> bit indicating that to the guest. Humm... I'm thinking now, that because of
> migration, we should check this bit every time, because we might have changed host.
> So instead of using an expensive cpuid check, we should probably use some bit in
> the vcpu_time_info structure, and use a cpuid bit just to say it is enabled.
>    

Right, we need a bit in the structure itself that says you can trust the 
time not to go backwards, and a bit in cpuid that says you can trust the 
bit in the structure (unless we can be sure no implementations leak 
garbage into the padding field).

-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.