Re: HT bug in some Intel CPUs ?

[Manuel Bouyer <bouyer%antioche.eu.org@localhost>]

On Thu, Aug 06, 2009 at 09:40:22AM -0600, Greg Oster wrote:
> Manuel Bouyer writes:
> > On Thu, Aug 06, 2009 at 08:55:39AM -0600, Greg Oster wrote:
> > > Manuel Bouyer writes:
> > > > Hi,
> [snip]
> > > 
> > > They don't show up as hyperthreaded in 3.0, but do in 5.0.1.
> > 
> > Did you try a SMP 3.0 kernel ? Maube a ACPI issue ?
> 
> Yes.. the box has been running GENERIC.MP for ages without issue.
> 
> > > > [...]
> > > > Did someone else notice something similar, or have informations about
> > > > such bug ?
> > > 
> > > I'd like to think that whatever issue you're seeing is the same one I 
> > > have... Would disabling hyperthreading help at all to provide a 
> > > datapoint?
> > 
> > That would, of course. Now there may be other issues :)
> 
> I'll see if I can schedule a quick downtime and turn off 
> hyperthreading...

that would be good; as the box is also SMP it'd be even better
to make sure HT is a problem.


> But yes, it's the "may be other issues" I'm mostly 
> concerned about :-}

Some have reported issues with processes stuck in tstile.
I've never seen it myself, neither on ftp.fr.netbsd.org
nor ftp.lip6.fr which is a mirror for several major linux distros, and
gets an average of 10 http requests/s and quite important ftp traffic
too. Both hosts are SMP. ftp.fr has an athlon x2, ftp.lip6.fr
has 2 3.4Ghz Xeon, but with HT turned off (I can't remember if
I turned it off on purpose, or if I just forgot to turn it on).
Both have been stable for the last 3 months (exept power failures).

-- 
Manuel Bouyer <bouyer%antioche.eu.org@localhost>
     NetBSD: 26 ans d'experience feront toujours la difference
--