cvs commit: src/sys/conf files.amd64

[Poul-Henning Kamp]

In message <20040524155045.V23809 at gamplex.bde.org>, Bruce Evans writes:

>> I agree and think it may be good to do something about this.  Care to
>> suggest a way to adjust or at least detect TSC rate changes?  On my
>> laptop, the switch from 1 Ghz to 733 Mhz is done via SMM and thus is not
>> user-detectable except through profiling.  However, on desktop systems,
>> the switch is almost certainly never automatic.  We can have the control
>> system notify the timecounters when the rate changes.
>
>The TSC could be sample on every clock interrupt or on every timecounter
>call.  This should at least tell you if the average TSC rate for the
>samping period was significantly different, and if there is any long-term
>drift.

If the TSC varies in frequency we should not use it for timecounting,
trying to compensate for the changes will never be precise enough to
avoid significant short term jitter.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.