every OpenGL-program crashes on intel GMA X3100 (failed assertion "region->tiling != 1")

[Mirco Müller]

Binary package hint: libgl1-mesa-dri

When trying to start any OpenGL-based program it immediately crashes with the output on stdout:

"brw_misc_state.c:290: emit_depthbuffer: Assertion 'region->tiling != 1' failed."

This is on a Lenovo X61 tablet/notebook with a i965 chipset (GPU: intel GMA X3100). The affected file comes from mesa/drivers/dri/i965/brw_misc_state.c

The package in question is libgl1-mesa-dri 7.8.1-3ubuntu2

Tried it with these programs, which all fail with the same error stated above:
* glxgears
* compiz
* mutter
* unity (as mutter-plugin)
---
Architecture: i386
DRM.card0.LVDS.1:
 status: connected
 enabled: enabled
 dpms: On
 modes: 1400x1050
 edid-base64: AP///////wAwrgVAAAAAAAAQAQOAGRJ46q9AlVZKjyUgUFQhCACBgAEBAQEBAQEBAQEBAQEBICF4yFAaCkAUdCUA9bgAAAAZAAAAEAAAAAAAAAAAAAAAAAAAAAAADwCQQzIAAAATAgAJ5QAAAAAA/gBIVjEyMVAwMS0xMDEKAHU=
DRM.card0.VGA.1:
 status: disconnected
 enabled: disabled
 dpms: On
 modes:
 edid-base64:
DistroRelease: Ubuntu 10.10
DkmsStatus: Error: [Errno 2] No such file or directory
MachineType: LENOVO 776294G
Package: mesa (not installed)
PccardctlIdent:
 Socket 0:
   no product info available
PccardctlStatus:
 Socket 0:
   3.3V 32-bit PC Card
ProcCmdLine: root=UUID=952f1c9d-c055-4f91-9758-f9271a584377 ro quiet splash
ProcEnviron:
 PATH=(custom, user)
 LANG=de_DE.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.35-8.13-generic 2.6.35-rc5
Tags: maverick maverick
Uname: Linux 2.6.35-8-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare
dmi.bios.date: 07/03/2007
dmi.bios.vendor: LENOVO
dmi.bios.version: 7SET18WW (1.04 )
dmi.board.name: 776294G
dmi.board.vendor: LENOVO
dmi.board.version: Not Available
dmi.chassis.asset.tag: No Asset Information
dmi.chassis.type: 10
dmi.chassis.vendor: LENOVO
dmi.chassis.version: Not Available
dmi.modalias: dmi:bvnLENOVO:bvr7SET18WW(1.04):bd07/03/2007:svnLENOVO:pn776294G:pvrThinkPadX61Tablet:rvnLENOVO:rn776294G:rvrNotAvailable:cvnLENOVO:ct10:cvrNotAvailable:
dmi.product.name: 776294G
dmi.product.version: ThinkPad X61 Tablet
dmi.sys.vendor: LENOVO
system:
 distro: Ubuntu
 codename: maverick
 architecture: i686
 kernel: 2.6.35-8-generic

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

apport information

[Mirco Müller (macslow)]

Here's an interesting observation. On 19.7.2010 I pulled updates via synaptic in the hope it would pull in a fix for this that might have happened. But that wasn't the case. Then I pulled the live-cd image of the same day, put it on a usb-stick and did a fresh install from that one. This actually solved the issue and fixed tihs bug. This leads to the assumption, that this might be a package-issue or something similar.

[Mirco Müller (macslow)]

The fresh install wiped my xorg.conf with the default one. Thus deleting the line setting "Tiling" "false", which according to Chris "ROAF" Halse Rogers could be a possible cause for the reported bug to happen. I forgot about that option being there and was probably a left-over from some older tweaking attempt for an older driver-version.

[Chris Halse Rogers (raof)]

I'm going to mark this as invalid. I was considering making the DDX spit out an warning about the Tiling option, but upstream basically regards all options as debugging tools only and make no guarantees any random option won't break some part of the system.

I don't think this is worth a delta from upstream for.