On Tuesday 01 March 2005 21:41, Josh Babcock wrote:
> Lee Elliott wrote:
> > On Monday 28 February 2005 22:23, Josh Babcock wrote:
> >>OK, this is making me feel kind of stupid.  I never really
> >> had any problem with the fglrx driver before, but now I'm a
> >> bit stumped.  I just completely rebuilt my system to take
> >> advantage of LVM and the deb package system as well as to
> >> upgrade to 2.6.10.  Now, I seem to have fglrx working, but
> >> alas, <1 FPS in FG. dmesg reports *lots* of these:
> >>[fglrx:firegl_agp_lock_pages] *ERROR*
> >>agp_allocate_memory_phys_list failed Has anyone else seen
> >> this problem?
> >>
> >>Oh, to avoid confusion, the kernel and fglrx module did not
> >>come from a deb repository, I built them by hand.  I just
> >>don't find the debian build process sane.  I also held the
> >>Xfree packages so libGL.so wouldn't get clobbered.
> >>
> >>Josh
> >>
> >>Gory details below...
> >
> > [snipped...]
> >
> > Are you using the internal ATI agpgart or the 2.6.10 kernel
> > module?
> >
> > I have the line
> >
> >     Option "UseInternalAGPGART"         "yes"
>
> [snip]
>
> Nope, I've got it set to "yes".  If I recall, using the kernel
> module led to fglrx spewing all sorts of errors in the X log
> and X falling back to Mesa.  I also think I should note that
> even though I have FB support in my kernel config, using the
> line:
>  Option  "UseFBDev"  "true"
> in XF86Config-4 results in X failing to load, with complaints
> from fglrx about not being able to access the frame buffer. 
> This is really the same X config that I had working before.  I
> think I either did something stupin in my kernel config or
> there is some change in the code causing a new problem.
>
> Thanks,
> Josh

Hmm...

the fact that the agpgart module is loaded on your system seems 
to imply that the ATI internal agpgart isn't being used.  Dunno 
why though.

LeeE

_______________________________________________
Flightgear-users mailing list
[email protected]
http://mail.flightgear.org/mailman/listinfo/flightgear-users
2f585eeea02e2c79d7b1d8c4963bae2d

Reply via email to