lock contention etc.

Kip Macy kip.macy at gmail.com
Thu Feb 1 20:30:21 UTC 2007


I can already tell you from builds on the T2000 that its the vm page
queue mutex.

On 2/1/07, Julian Elischer <julian at elischer.org> wrote:
> Ivan Voras wrote:
> > Julian Elischer wrote:
> >
> >>     3762.94 real      2627.66 user      1275.78 sys
> >>     2465.10 real      2958.94 user      1918.64 sys
> >>     2262.88 real      4322.31 user      3333.58 sys
> >
> >> (2 processors with HTT)  (looks like 4 processors)
> >> -current, sched_4bsd
> >
> > I'm possibly stating the obvious, but did you try without HTT or on a
> > real 4proc machine? Much of the contention might be an artifact of HTT
> > (e.g. HTT "logical" processor blocking when real work could be done
> > because the internal processor units are busy elsewhere).
> >
> >
>
> yes I just sent an email stating that.
> And I'm trying it again with the virtual cores turned off.
>
> _______________________________________________
> freebsd-current at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to "freebsd-current-unsubscribe at freebsd.org"
>


More information about the freebsd-current mailing list