10-50% CPU used by xorg?

[Daniel Kasak]

On Wed, 2008-02-20 at 14:00 +1100, Robert Lowery wrote:

> > On Tue, 2008-02-19 at 14:47 +0100, Ben E. Hard wrote:
> >
> >> 113       0.6208  libexa.so                libexa.so
> >
> > EXA on Intel 830 systems is bound to be fantastically slow at the
> > moment, particularly without a composite manager. We're waiting for LOTS
> > of experimental stuff to land: the batchbuffer branch of the Intel
> > driver, TTM, DRI2, etc, etc. Last word was that this wasn't happening
> > any time soon, so in the meantime, force X to use XAA:
> >
> > Option      "AccelMethod" "xaa"
> 
> Given then above statements, should the default be set back to XAA in the
> driver?

My comment above sums it up for 2D rendering performance at the moment
( for I830 chips - I can't comment on the later ones, but I assume the
situation is somewhat better ). If you have a special need that is only
met by EXA, then you may need to use it. Why not try both methods and
decide for yourself which is best for now? As noted, EXA performance for
us will improve when we can make use of the batchbuffer branch of the
driver. You can try it out right now if you can get all the parts built
and running.

--
Daniel Kasak
IT Developer
NUS Consulting Group
Level 5, 77 Pacific Highway
North Sydney, NSW, Australia 2060
T: (+61) 2 9922-7676 / F: (+61) 2
9922 7989
email: dkasak at nusconsulting.com.au
website:
http://www.nusconsulting.com.au