tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: TSC vs. HPET timecounter priority



On Mon, Aug 13, 2012 at 09:07:26AM -0700, Erik Fair wrote:
> 
> On Aug 13, 2012, at 08:49, Jukka Ruohonen wrote:
> 
> > On Sun, Aug 12, 2012 at 08:14:50PM -0700, Erik Fair wrote:
> >> NetBSD has a problem with NTP time keeping on x86 systems which have TSC:
> >> if those systems have variable CPU clock frequency support for power
> >> consumption efficiency (e.g.  SpeedStep, PowerNow), if or when the CPU
> >> frequency changes, the relationship between TSC and the ticking of real
> >> time changes.  That screws up any kind of interval time & frequency
> >> measurement one is using TSC for.
> > 
> > I think this is no longer true with new AMD and Intel processors.
> 
> Can you find a reference in either Intel or AMD's documentation, and to
> which CPUs that applies (preferably with some indicator we can test at
> system initialization time)?

Yes, read the circa post-Athlon64/post-Core2Duo manuals. There are ready
MSR-based heuristics in acpicpu(4). The probe in tsc.c should be upgraded.

- Jukka.


Home | Main Index | Thread Index | Old Index