Bug 64508

Summary:	Text rendered using QuesoGLC is unreadable/missing on Radeon HD 6290
Product:	Mesa	Reporter:	Max <maxrd2>
Component:	Drivers/Gallium/r600	Assignee:	Default DRI bug account <dri-devel>
Status:	RESOLVED MOVED	QA Contact:
Severity:	normal
Priority:	medium	CC:	maxrd2
Version:	9.1
Hardware:	x86-64 (AMD64)
OS:	Linux (All)
See Also:	http://sourceforge.net/tracker/?func=detail&aid=2831514&group_id=53918&atid=472059
Whiteboard:
i915 platform:		i915 features:
Attachments:	screenshots in order: test9.1, test9.2, test16 glxinfo output screenshots in order: test9.1, test9.2, test16 qapitrace screenshot dmesg output Xorg log

Description Max 2013-05-12 21:00:27 UTC

Created attachment 79212 [details]
screenshots in order: test9.1, test9.2, test16

Overview:
Text rendered using QuesoGLC is unreadable/missing on Radeon HD 6290.
As a consequence all text in warzone2100 game is unreadable.

Steps to Reproduce:
wget http://surfnet.dl.sourceforge.net/project/quesoglc/QuesoGLC/0.7.2/quesoglc-0.7.2.tar.bz2
tar -xjf quesoglc-0.7.2.tar.bz2
./configure && make check

Actual Results:
Following tests look very bad... letters look like a bunch of overlapping triangles/primitives (attached screenshots)
- test9.2 - Render style : GLC_TEXTURE       without GL objects     without user list
- test16

Expected Results:
Normally rendered text

Platform:
Arch Linux x86_64
Kernel: 3.8.12-1-ck #1 SMP PREEMPT x86_64 GNU/Linux

mesa 9.1.2-1
ati-dri 9.1.2-1
glu 9.0.0-2
xorg-server 1.14.1-1
xf86-video-ati 1:7.1.0-3

lspci output:
00:01.0 VGA compatible controller: Advanced Micro Devices [AMD] nee ATI Wrestler [Radeon HD 6290] (prog-if 00 [VGA controller])
        Subsystem: Acer Incorporated [ALI] Device 0598
        Flags: bus master, fast devsel, latency 0, IRQ 41
        Memory at e0000000 (32-bit, prefetchable) [size=256M]
        I/O ports at 4000 [size=256]
        Memory at f0400000 (32-bit, non-prefetchable) [size=256K]
        Expansion ROM at <unassigned> [disabled]
        Capabilities: <access denied>
        Kernel driver in use: radeon
        Kernel modules: radeon

Comment 1 Max 2013-05-12 21:06:29 UTC

Forgot to mention Hardware is Acer Aspire One 722 (AMD C60)

Am pretty good in c/c++, and am using git everyday, so would be glad to help pinpointing this issue.

Comment 2 Max 2013-05-12 21:09:22 UTC

this may be related: http://sourceforge.net/tracker/?func=detail&aid=2831514&group_id=53918&atid=472059

Comment 3 Max 2013-05-12 21:38:32 UTC

Created attachment 79213 [details]
glxinfo output

Comment 4 Max 2013-05-12 21:39:47 UTC

Created attachment 79214 [details]
screenshots in order: test9.1, test9.2, test16

Comment 5 Alex Deucher 2013-05-12 22:51:57 UTC

Please attach your dmesg output and xorg log.

Comment 6 Vadim Girlin 2013-05-13 02:17:32 UTC

Created attachment 79220 [details]
qapitrace screenshot

As far as I can see it's a bug in the QuesoGLC.
I'm attaching the screenshot of qapitrace with gl trace for the test9.2 and the event filter "glBind|glTexIma|glTexSub|glNew|glEndList|glCall|glGen" applied to show relevant calls.

Mesa reports the errors on the glTexSubImage2D calls because the texture is not initialized yet (e.g. using glTexImage2D). Note that the test compiles the display list first and uses glTexImage2D to initialize the texture, but compiled display list is never called, instead the app basically repeats the same sequence of calls without calling glTexImage2D first.

As far as I understand, mesa reports the errors correctly, and it's in fact the bug in QuesoGLC.

By the way, there is another bug related to the issues with warzone 2100:
http://sourceforge.net/tracker/index.php?func=detail&aid=1830052&group_id=53918&atid=472059

Comment 7 Max 2013-05-13 09:19:32 UTC

Created attachment 79232 [details]
dmesg output

Comment 8 Max 2013-05-13 09:20:07 UTC

Created attachment 79233 [details]
Xorg log

Comment 9 Max 2013-05-13 09:20:49 UTC

thanks for your time... will try to get it checked/fixed by QuesoGLC

Comment 10 GitLab Migration User 2019-09-18 19:03:00 UTC

-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/437.

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.