numbers don't lie ...
mwm-keyword-freebsdhackers2.e313df at mired.org
Thu Sep 14 06:40:24 PDT 2006
In <E1GNoDy-0004CF-5k at cs1.cs.huji.ac.il>, Danny Braniss <danny at cs.huji.ac.il> typed:
> > In <E1GNOLq-000DC2-1Q at cs1.cs.huji.ac.il>, Danny Braniss <danny at cs.huji.ac.il> typed:
> > What's the CPU configuration? The AMD is dual core - is that it? Could
> > the Xeon be dual-core and hyperthreaded, so it's got that many more
> > CPUs to contribute towards user time?
> > To illustrate, I have numbers for "make -j4" for a P4 with and without
> > hyperthreading enabled:
> > machdep.hyperthreading_allowed: 1 -> 0
> > 50m55.99s real 35m28s.19 user 8m20s.02 sys
> > machdep.hyperthreading_allowed: 0 -> 1
> > 38m48s.85 real 55m2s.43 user 12m27s.90 sys
> > Note the effect of the second CPU on the user time.
> i did the tests with hyperbluffing disabled in the bios.
> now, since you asked i tried:
> sysctl machdep.cpu_idle_hlt=0
> machdep.cpu_idle_hlt: 1 -> 0
> dell-2950 : make -j8 TARGET_ARCH=amd64 buildworld : 23m55.03s real 1h2m54.84s
> user 16m48.79s sys
> and no difference, by the way, i did this also changing the bios setting
> (called logical cpus :-), but again, no difference.
Um - what version of FreeBSD are you running? I don't change
machdep.cpu_idle_hlt on mine; I chang machdep.hyperthreading_allowed
(as indicated in the tests). Did you check to see if the system
actually saw more cpus when you changed it?
Mike Meyer <mwm at mired.org> http://www.mired.org/consulting.html
Independent Network/Unix/Perforce consultant, email for more information.
More information about the freebsd-hackers