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

Re: Some annoying problems with ATi Radeon 7200!

On Fre, 2002-11-08 at 20:25, Manuel Bilderbeek wrote:
> Michel Dänzer wrote:
>   > On Don, 2002-11-07 at 22:11, Manuel Bilderbeek wrote:
>   >
>   >>Michel Dänzer wrote:
>   > kernel-image-2.4.19-k7 also contains some DRM modules, use
>   > --force-overwrite .
> OK, I did it!
> And, I have 3D again! :-)

Glad to hear that.

> Now, coming back to my previous problems (I only did a quick test
> though...):
>  > - TuxRacer: Tux looks white!
> Hah! Tux looks normal now!
> However, still problems with Tuxracer: with the new drivers, the game 
> slows down incredibly about every second. It seems to go smoooth, almost 
> stops then, continues smoothly then, almost stops again, etc.
> In my terminal this appears:
> radeonUpdatePageFlipping allow 0 current 0
> radeon_makeX86Normal3fv/195 CVAL 0 OFFSET 14 VAL 40763f60
> radeon_makeX86Normal3fv/196 CVAL 4 OFFSET 20 VAL 40763f64
> radeon_makeX86Normal3fv/197 CVAL 8 OFFSET 25 VAL 40763f68
> radeon_makeX86Normal3fv done

This is harmless debugging output.

> radeonUploadTexImages: ran into bound texture


> If I play for a longer time, loads of those radeonUploadTexImages 
> messages appear.... A bug?

Possibly, does it still occur with the latest packages, which are now
available from

deb	http://people.debian.org/~daenzer/dri-trunk/		./


> Also, in the syslog:
> Nov  8 20:06:57 goemon kernel: [drm:radeon_freelist_get] *ERROR* 
> returning NULL!

This can happen intermittently without being a problem, or it can be a
symptom of a lockup.

>  > - Armagetron: when more than 3
>  > players are in the game, the 'motor bikes' get partly invisible;
>  > something goes wrong with the textures. When leaving the program, I
>  > can see lots of these messages from stdout/stderr:
>  > radeonUploadTexImages: ran into bound texture and in my syslog I get:
>  >  Nov 4 20:19:37 goemon kernel: [drm:radeon_freelist_get] *ERROR*
>  > returning NULL! Nov 4 20:20:08 goemon last message repeated 492 times
> This problem is completely gone now! The game seems to work just fine!

Really? I need to set RADEON_NO_VTXFMT=1 for the textures to be rendered

> Something else that is new now: when I run e.g. glxgears and move the 
> glxgears window, I get lots of these messages:
> radeonUpdatePageFlipping allow 0 current 0
> Is this normal? It's a bit annoying...

It's harmless debugging output and should be gone in the latest
packages. BTW it shows that page flipping is disabled, Option
"EnablePageFlip" will give some extra FPS at the cost of some minor

>   >> > D'oh, I meant Option "SWcursor" of course. The device section is the
>   >> > right place.
>   >>
>   >>YES! This solves the problem!
>   >>So, what could be the cause of the problem and what is the disadvantage
>   >>of the SWcursor? (I guess it's slower in some way...)
>   >
>   > HW cursor is better for various reasons. I've never heard of such a
>   > problem before. You may want to report it to the Xpert list.
> ?????????
> I just logged in and the problem is back! And it's *massive*!! It's far
> worse than I've ever seen. Any change on the screen produces this
> interference now....! :-( I don't understand it at all... How could it 
> have been solved for that particular time?


> So, I commented the SWcursor out again...
> I'd do anything to solve this problem. Let me know how to report to the 
> Xpert list.

Not sure what you wanna know; post to Xpert@XFree86.Org (beware,
moderated for non-subscribers) describing the problem in detail.

>   >>I never knew this was a valid option for the Driver. Is there a list of
>   >>options somewhere?
>   >
>   >
>   > In the source. :) Other drivers have manpages, the radeon driver not yet
>   > unfortunately. Someone posted a draft to the Xpert list a while ago,
>   > would be nice if that would make it into 4.3.0 .
> Where can I find that draft?

James Ralston posted it to Xpert on July 21st under the subject 'adding
radeon driver documentation'.

> Some final questions, oh Radeon Guru! :-)
> - What are safe settings for AGPSize? I saw it defaults to 8. In my BIOS 
> I have the AGP Aperture set to 64MB, or doesn't this have to do with 
> AGPSize?

It's the upper limit. The radeon driver doesn't do AGP texturing
currently so it uses 3 MB of AGP memory at most. Anything more is
basically wasted.

> - When X is booted, the screen stays black for quite a long time. My 
> monitor reports that the frequency is only 2Hz or so, and almost goes 
> off. But just before it goes off, my xdm login screen appears. Is there 
> a reason for this long delay?

It's probably the driver querying the monitor via DDC.

> Can I make it shorter?

Option "NoDDC" or "NoVBE" may help.

> - I played a bit with that AGPSize, and setting it to 64MB hung my X 
> just after starting (xdm login screen wasn't even shown yet). Since I 
> couldn't go back to my getty terminals (keyboard didn't respond) and xdm 
> is autostarted at boot time, the only thing I could do was login from 
> another computer to mine to change the XF86Config-4. Is there a smarter 
> way to do this? :-)

You can boot into single user mode by passing 'single' as an argument to
the kernel.

> In any case: thanks a *lot* for all your help!! Keep up the good work!

You're welcome, kudos go to the DRI team.

> PS: MSX FOR EVER! (Questions? http://faq.msxnet.org/ )
> PPS: Visit my homepage at http://manuel.msxnet.org/

Funny, my first computer was an MSX, from Canon IIRC. :)

Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer
XFree86 and DRI project member   /  CS student, Free Software enthusiast

Reply to: