Desktop Performance Tuning?

[Eli Scott]

On Feb 7, 2008 8:19 AM, Jason Morgan <
jwm-freebsd-questions at sentinelchicken.net> wrote:

> On Thu, Feb 07, 2008 at 01:07:34AM +0000, RW wrote:
> > On Wed, 6 Feb 2008 15:45:14 -0800
> > "Eli Scott" <havok173 at gmail.com> wrote:
> >
> > > I recently installed FreeBSD 7.0 RC-1 on my hobby system,
> > >..
> > > performance seems to degrade whenever i'm compiling
> > > things (like, performing a buildworld, installing from ports, etc)
> > > which manifests itself as considerable lag in standard desktop
> > > operations,
> >
> > I've noticed this too, it always used to be the case that building
> > didn't make all that much difference to desktop use, but now it make a
> > severe difference. Nice helps, but it only makes the
> > problem intermittent. I've tried changing the scheduler to ULE, and
> > disabling SMP, but it didn't help.
> >
> > There have been threads on the stable list about jerky  mouse
> > performance, which may be part of this, but I have really followed it
> > closely. I'm also wondering whether this might be due to some xorg or
> > other port change from late 2007 that I only noticed when I started
> > doing a lot of rebuilding under 7-stable.
> >
>
> I've also noticed this. I thought it was because I used the -j4 flag
> when building world, but I just rebuilt my kernel last night without
> -j4 and had some of the same issues. This has been on both SCHED_ULE
> and SCHED_4FSB. I am running AMD64.
>
> ~Jason
> _______________________________________________
> freebsd-questions at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> To unsubscribe, send any mail to "
> freebsd-questions-unsubscribe at freebsd.org"
>

RE: Schedulers, i've also used kernels with both schedulers on this machine
and have noticed no difference in desktop performance.  I remember an
article on ONLAMP from a few years ago that talked about tuning the system
for desktop use, i'll check it out and see if the tunables that are
discussed there make any difference.