Bug 25197 - [REGRESSION] openarena: vbo/vbo_exec_api.c:881: vbo_exec_FlushVertices: Assertion `callDepth == 1' failed.
Summary: [REGRESSION] openarena: vbo/vbo_exec_api.c:881: vbo_exec_FlushVertices: Asser...
Status: RESOLVED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/R600 (show other bugs)
Version: unspecified
Hardware: Other All
: medium normal
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-11-20 03:14 UTC by Rafał Miłecki
Modified: 2009-11-23 13:18 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
openarena.log (6.08 KB, text/plain)
2009-11-20 03:15 UTC, Rafał Miłecki
Details

Description Rafał Miłecki 2009-11-20 03:14:58 UTC
openarena: radeon_mipmap_tree.c:422: migrate_image_to_miptree: Assertion `dstlvl->width == image->base.Width' failed.
(...)
openarena: vbo/vbo_exec_api.c:881: vbo_exec_FlushVertices: Assertion `callDepth == 1' failed.

Using mesa master, first bad commit:

commit 3f2c77659ca552c43f544228f3a5a5fe6365513a
Merge: b09e749 f8ea531
Author: Dave Airlie <airlied@itt42.(none)>
Date:   Fri Nov 20 11:48:10 2009 +1000

    Merge remote branch 'origin/mesa_7_7_branch'

I guess this commit was supposed to fix some bugs like bug #22372 but introduced regression for me.

Would you like me to bisect this?
Comment 1 Rafał Miłecki 2009-11-20 03:15:26 UTC
Created attachment 31335 [details]
openarena.log
Comment 2 Maciej Cencora 2009-11-20 03:54:52 UTC
This is probably a duplicate of 25177. No need to bisect. I'll try reproducing it on my machine.
Comment 3 Rafał Miłecki 2009-11-21 07:32:35 UTC
I've some problems with bisecting.

First I tried to find first commit when I can start openarena (menu usable) but can not start game (loading... crash). I gone to this...

git log master shows:

commit 012d0193cc9ad6fdc9829db0a6884a5a590dd4c5
Author: Michel Dänzer <daenzer@vmware.com>
Date:   Thu Nov 19 17:25:05 2009 +0100

    st/xorg: Don't complain about convolution filter being 'unknown'.

commit 827ba44f6ee83ab21c6a2b09323f6f1df4a7d4c8
Author: Eric Anholt <eric@anholt.net>
Date:   Wed Nov 18 18:15:25 2009 +0100

    intel: Remove non-GEM support.

and according to my tests:
012d0193cc9ad6fdc9829db0a6884a5a590dd4c5 is BAD
827ba44f6ee83ab21c6a2b09323f6f1df4a7d4c8 is GOOD

This didn't make any sense. So i tried:
git bisect good 827ba44f6ee83ab21c6a2b09323f6f1df4a7d4c8
git bisect bad 012d0193cc9ad6fdc9829db0a6884a5a590dd4c5
and it seems "git log master" /lies/ somehow, as bisect module saw something between there two commits!

So I decided to let "git bisect" work for me and gone to this:

There are only 'skip'ped commits left to test.
The first bad commit could be any of:
93eb2ab8c395f81e40fa298d78805bb2c777f891
ad83aeccdc54beecf25f217e2dd24c8edf6d6767
23ec7c457483aae1e0d399e9b570f1860c27c780
afe84fa698eae3e035e967589f0a8d55f6a83698
7118db870091d4c9c2465e79f361ff0ed36d1f90
We cannot bisect more!

I used "bad" when I could not start openarena or could not start game (loading... crash).

I could not test some commits due to:

r600_texstate.c: In function ‘setup_hardware_state’:
r600_texstate.c:652: error: ‘struct _radeon_mipmap_tree’ has no member named ‘firstLevel’
r600_texstate.c:695: error: ‘struct _radeon_mipmap_tree’ has no member named ‘bpp’
r600_texstate.c:695: error: ‘struct _radeon_mipmap_tree’ has no member named ‘bpp’
r600_texstate.c:709: error: ‘struct _radeon_mipmap_tree’ has no member named ‘lastLevel’
r600_texstate.c:709: error: ‘struct _radeon_mipmap_tree’ has no member named ‘firstLevel’
r600_texstate.c:711: error: ‘struct _radeon_mipmap_tree’ has no member named ‘firstLevel’
r600_texstate.c:712: error: ‘struct _radeon_mipmap_tree’ has no member named ‘lastLevel’
r600_texstate.c: In function ‘r600SetTexOffset’:
r600_texstate.c:811: error: ‘struct _radeon_mipmap_tree’ has no member named ‘firstLevel’
Comment 4 Rafał Miłecki 2009-11-23 04:00:16 UTC
I generated compilation fixing patch with:
git show 7118db870091d4c9c2465e79f361ff0ed36d1f90 > r600.align.for.mipmap.tree.changes.patch


git checkout 93eb2ab8c395f81e40fa298d78805bb2c777f891
patch -p1 < r600.align.for.mipmap.tree.changes.patch
make clean && make -j 4 && su -c 'make install'
[compiles fine, openarena doesn't start]
[openarena: radeon_mipmap_tree.c:422: migrate_image_to_miptree: Assertion `dstlvl->width == image->base.Width' failed.]


git checkout ad83aeccdc54beecf25f217e2dd24c8edf6d6767
patch -p1 < r600.align.for.mipmap.tree.changes.patch
make clean && make -j 4 && su -c 'make install'
[compiles fine, openarena doesn't start]
[openarena: radeon_mipmap_tree.c:422: migrate_image_to_miptree: Assertion `dstlvl->width == image->base.Width' failed.]


git checkout 23ec7c457483aae1e0d399e9b570f1860c27c780
patch -p1 < r600.align.for.mipmap.tree.changes.patch
make clean && make -j 4 && su -c 'make install'
[compiles fine, openarena doesn't start]
[openarena: radeon_mipmap_tree.c:423: migrate_image_to_miptree: Assertion `dstlvl->width == image->base.Width' failed.]


git checkout afe84fa698eae3e035e967589f0a8d55f6a83698
patch -p1 < r600.align.for.mipmap.tree.changes.patch
make clean && make -j 4 && su -c 'make install'
[compiles fine, openarena doesn't start]
[openarena: radeon_mipmap_tree.c:422: migrate_image_to_miptree: Assertion `dstlvl->width == image->base.Width' failed.]


So as expected all the commits are broken for me. I guess it's that new mipmap tree work that causes problem, right? Can I test anything else?
Comment 5 Rafał Miłecki 2009-11-23 04:48:23 UTC
Last changes in Mesa improved my situation somehow.

With (older) commit 2b07b640619ac68344276ba0557ea46b2cbc3f26 I can not start openarena.

With (newer) commit c3c8c40cab193e0aa0f1a42bff7b0d726df8cf9f I can start openarena (menu), can not start play (3D world).

Always the same message:
openarena: radeon_mipmap_tree.c:422: migrate_image_to_miptree: Assertion `dstlvl->width == image->base.Width' failed.
Comment 6 Rafał Miłecki 2009-11-23 13:18:22 UTC
Fixed in 960464e42dce138fde11c379ce7744bc4be14aa2


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.