Re: timekeeping regression?

["Mathew, Cherry G." <c%bow.st@localhost>]

>>>>> On Fri, 07 Jun 2024 18:34:55 -0700, "Greg A. Woods" <woods%planix.ca@localhost> said:

> At Thu, 06 Jun 2024 21:43:46 -0700, "Greg A. Woods" <woods%planix.ca@localhost> wrote:
> Subject: Re: timekeeping regression?
>> 
>> The "hvm" domUs are running FreeBSD and have no problems.  They're using
>> what they call XENTIMER as their timecounter clock source.  The FreeBSD
>> code is very different, at least on first glance, and somewhat more
>> convoluted in some ways.  I don't see an obvious "rdtsc" instruction
>> being used, but there are hints that's what it is doing, but I may be
>> wrong.

> Never mind.   The rdtsc() call was staring at me when I looked at the
> same FreeBSD code again this morning.

> So, that kinda puts the onus for this regression back on NetBSD somehow,
> though I don't understand how it could be so sensitive to which version
> of Xen it's running under.

> I keep thinking there might be some kind of overflow or similar, but
> none of the obvious magic numbers seem to jump out as relevant to me,
> though I'm a bit rusty at thinking in those terms.

Is it possible that you have a handy matrix of versions of NetBSD/Xen
that work/don't work perhaps ?


-- 
Math/(~cherry)