Re: [Xen-devel] [PATCH v2 6/6] x86/time: implement PVCLOCK_TSC_STABLE_BIT

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 07.04.16 at 23:17, <joao.m.martins@xxxxxxxxxx> wrote:
>> > The main
>>> difference I see between both would be the base system time: 
>>> read_platform_stime
>>> uses stime_platform_stamp as base, and computes a difference from the
>>> read_counter (i.e. rdtsc() ) with previously saved platform-wide stamp
>>> (platform_timer_stamp). get_s_time uses the stime_local_stamp (updated from
>>> stime_master_stamp on local_time_calibration) as base plus delta from 
>>> rdtsc()
>>> with local_tsc_stamp. And since this is now all TSC, and TSC monotonically
>>> increase and is synchronized across CPUs, both calls would end up returning 
>>> the
>>> same or a always up-to-date value, whether cpu_time have a larger gap or not
>>> from stime_platform_stamp. Unless the concern you are raising comes from the
>>> fact CPU 0 calibrates much sooner than the last calibrated CPU, as opposed 
>>> to
>>> roughly at the same time with std_rendezvous?
>> 
>> In a way, yes. I'm concerned by the two time stamps no longer
>> being obtained at (almost) the same time. If that's not having
>> any bad consequences, the better.
> 
> I don't think there would be bad consequences as both timestamps correspond 
> to the same time reference - thus returning always the latest system time
> irrespective of the gap between both stamps.
> 
> If you prefer I can go back with my initial approach (v1, with std_rendezvous)
> to have both timestamps closely updated. And later (post-release?) revisit 
> the introduction of nop_rendezvous. Perhaps this way is more reasonable?

Since the new mode need to be actively asked for, I don't think
that's necessary.

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel