Bug 22643

Summary: rs485 dri2/kms Google Earth failure (kernel doesn't handle compressed texture formats)
Product: DRI Reporter: Paulo Dias <paulo.miguel.dias>
Component: DRM/RadeonAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED FIXED QA Contact:
Severity: normal    
Priority: medium CC: edurany2
Version: unspecified   
Hardware: Other   
OS: All   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
Xorg.0.log none

Description Paulo Dias 2009-07-06 13:33:07 UTC 
Instead of doing a bug for each app, which would be overkill i'll just list them here and their respective bugs.

freeorion: http://freeorion.psitronic.de/

dies with a segfault: 

<6>freeorion[22922]: segfault at 54 ip b5a0d211 sp bff76cd0 error 4 in r300_dri.so[b59ba000+23b000] 

and a lot of this before the crash

Allocating 16 x 16 radeon RBO (pitch 32)
Allocating 16 x 16 radeon RBO (pitch 16)

googleearth: 

dies with a 

WARNING! Falling back to software for two-sided stencil
drmRadeonCmdBuffer: -22

and 

Jul  6 17:26:16 kerberos kernel: [ 4555.467061] [drm:r300_packet0_check] *ERROR* Invalid texture format 17
Jul  6 17:26:16 kerberos kernel: [ 4555.467069] [drm:radeon_cs_ioctl] *ERROR* Invalid command stream !

fbo_firecube dies with a segfault when changing textures (q option) with a 

Allocating 512 x 512 radeon RBO (pitch 512)
Depth renderbuffer size = 16 bits
drmRadeonCmdBuffer: -22

and a 

[ 4868.490320] [drm:r300_cs_track_check] *ERROR* [drm] Buffer too small for color buffer 0 (need 1048576 have 524288) !
[ 4868.490327] [drm:r300_cs_track_check] *ERROR* [drm] color buffer 0 (512 4 0 512)
[ 4868.490331] [drm:radeon_cs_ioctl] *ERROR* Invalid command stream !

both fbotests in mesa/prog/tests works so fbos are almost there :)

all this with latest drm/ddx/mesa/linus kernel with a rs485.

best regards
Comment 1 Paulo Dias 2009-07-14 09:00:13 UTC 
with latest mesa commits, fbo_firecube is working correctly for rs485. the remaining apps still show the same behaviour. getting better :)
Comment 2 Michel Dänzer 2009-07-15 04:11:00 UTC 
> Instead of doing a bug for each app, which would be overkill i'll just list
> them here and their respective bugs.

That's not how it works. Stick to one issue per report, or it becomes hard to resolve it as there's always 'one more thing...'.


> freeorion: http://freeorion.psitronic.de/

According to the forum linked to by the README, this seems to require OpenGL 2.x, which we probably won't support in the non-Gallium driver.


> Jul  6 17:26:16 kerberos kernel: [ 4555.467061] [drm:r300_packet0_check]
> *ERROR* Invalid texture format 17

This is a kernel radeon KMS issue, it doesn't handle compressed texture formats yet.

This seems to be the only remaining valid issue, reassigning and adapting title.
Comment 3 Paulo Dias 2009-08-03 13:08:11 UTC 
*** Bug 22713 has been marked as a duplicate of this bug. ***
Comment 4 Eduardo Durany 2009-09-02 02:46:39 UTC 
Googleearth crash with:
drmRadeonCmdBuffer: -22

and dmesg shows:

[ 4342.497187] [drm:radeon_cs_parser_init] *ERROR* cs IB too big: 16534
[ 4342.497194] [drm:radeon_cs_ioctl] *ERROR* Failed to initialize parser !

I don't know if it's the same bug or if i should open a new one.

KMS/DRI2

Radeon Mobility 7500

Ubuntu Jaunty with latest kernel from Karmic (2.6.31-rc8) and this ppa:
https://launchpad.net/~xorg-edgers/+archive/ppa

	
Congratulations, compiz is a lot of faster with DRI2 and KMS
Comment 5 Eduardo Durany 2009-09-02 02:47:57 UTC 
Created attachment 29102 [details]
Xorg.0.log
Comment 6 Michel Dänzer 2009-09-02 03:54:26 UTC 
(In reply to comment #4)
> [ 4342.497187] [drm:radeon_cs_parser_init] *ERROR* cs IB too big: 16534
> [ 4342.497194] [drm:radeon_cs_ioctl] *ERROR* Failed to initialize parser !
> 
> I don't know if it's the same bug or if i should open a new one.

As a rule of thumb: Different message, different bug.
Comment 7 Eduardo Durany 2009-09-02 04:30:31 UTC 
OK, understood:

KMS/DRI2 Radeon Mobility 7500 Googleearth crash
http://bugs.freedesktop.org/show_bug.cgi?id=23648
Comment 8 Paulo Dias 2009-09-04 11:25:43 UTC 
closing this bug, latest mesa/ddx/drm/linus kernel solved the crashing problem, just see bug https://bugs.freedesktop.org/show_bug.cgi?id=23715 for a new siteeffect (cant properly handle the textures), but no crash :)

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.