[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: DRI problems with 3dfx 4500 on DPW600au under Sarge



On Tue, 29 Nov 2005 02:17:26 +0100
Falko Schmidt <kaethorn@stud.uni-stuttgart.de> wrote:

> On Mon, Nov 28, 2005 at 04:25:37AM -0800, Urs Traber wrote:
> > Hello all,
> > 
> > I am struggling to make DRI work on a DPW600au with a 3dfx Voodoo
> > 4500 under Sarge. Kernel is 2.4.29.
> > 
> > Problem No.1:
> > When I enable DRI, all virtual consoles turn to a mix of white dots
> > scattered randomly all over the screen - I can still type but can't
> > read a thing...
> I'm having a similar effect here with ati DRI. The consoles get messed
> up upon keypress after starting and stopping the xserver (xorg),
> though switching between consoles seems to 'clean' the mess a bit. On
> my box (PWS600au) this phenomenon appears from 2.4.18 to 2.6.11.7 on
> Radeon 7500 and 9250 with no noticeable differences.

*click*

I don't know if it helps, but IIRC the RadeOns (as well as Permedia3)
seem to use a different addressing-scheme which is only fully supported
on EV6 and up (at least that's what I think to have understood by
browsing through c.o.vms). I'm having an XP1000 w/ a RadeOn 7500 sitting
right beside me and no problems whatsoever with kernel 2.6.8 and 2.6.14.

> > 
> > Problem No.2:
> > I have built an OpenGL sample program on this machine. It runs, but
> > the graphic doesn't look like it should (as a reference, I have a
> > DU4.0F box next to the Sarge box). With acceleration disabled, the
> > graphic looks ok (but is terribly slow).
> glxgears looks OK here and reports around 440fps, using 2.6.11.7 and
> X.org on Sid.

On which card? The 9250? Do 325fps on the system mentioned above seem
reasonable?

> > 
> > No idea whether these two issues are interrelated in any way.
> > But they can be reproduced on a second system (DPW500au) in
> > parallel, using another 3dfx 4500.
> > 
> > Is there anything that I could have missed to configure?
> > e.g. setpci -s 00:0c.0 4.w returns a 0003.
> 0007 here

0087 means, busmaster is activated, then?

 
> Falko

Marc



Reply to: