[i965] (Needs UXA) X freezes with I830WaitLpRing error when running OpenGL apps, unless use UXA

[Enrico Sardi]

On some Intel GM965 machines, running OpenGL applications (eg. Blender,
Second Life, the Endgame screensaver, Scorched3D) can crash the system.

Symptoms: Colours go funny, then screen flashes between off and black with 'please wait' cursor about 10 times, then stops; does not respond to Ctrl+Alt+F1 or Ctrl+Alt+Backspace, but does reboot on Alt+SysRq+b. Logs the line
Error in I830WaitLpRing(), timeout for 2 seconds
in /var/log/Xorg.0.log.old

(Please note that this is not the only X crash bug; if you cannot find
an exact match for your symptoms then please file a new bug.)

Workaround: Adding
        Option "DRI" "false"
to the graphics controller Device section of /etc/X11/xorg.conf avoids this bug, but may drastically reduce performance.

This bug has been reported upstream and you can follow the link to xorg bugzilla, but as of 29.12.07 no developer managed to reproduce this bug (how to trigger it seems to be hardware dependent, with some i965s not appearing to have it at all) so it will be very hard to fix unless that changes
if you are suffering from this bug please go to xorg bugzilla and report your hardware specifics.
also please edit this bug description and add the name and exact model of your affected laptop.

list of affected laptops:
Acer Travelmate 6292
Asus S62E Santa Rosa 965
dell D830
Dell 630
Dell Inspiron 1420
Dell Inspiron 1420n
Dell Inspiron 1525 (new)
Dell Latitude D630
Dell Latitude D830
Dell Vostro 1310
Dell XPS M1330
HP Compac 6510b
HP Compaq 6710b
Thinkpad T61
Thinkpad X61 Tablet
Thinkpad X61s
Apple Macbook (3,1/Santa Rosa)
Acer Aspire 5315-052G12Mi
Acer Extensa 5220-050508Mi
Lenovo 3000 N200 (TY2BBGE)
Sony vaio CR160

also if you have one of these laptops and are not affected please tell
us!

[Versvs (versvs)]

I got the same error. I can usually make a pair of clicks, open a windows or move something.. but then it freezes the whole machine. Still can move the cursor but nothing will be done.

Sadly, i cannot send a bug report in this state so i think i could not help with this issue.

Is there a way an output error is saved that may be sent to help debug?

[Versvs (versvs)]

I forgot: I have the X3100 graphic card, by Intel. With chipset 965gm.

[Earl Malmrose (earl)]

Same here. Asus S62E Santa Rosa 965 with x3100.

Upgrading to Gutsy Tribe 1 fixes the crash, but I'd rather be able to run Feisty for a while.

[Frank Lynch (frank-lynch)]

+1 same behavior here on my dell D830, as Earl pointed out the issue is resolved in gutsy (but I can't run gutsy as its still a bit unstable for my liking) - I think we just need the latest copy of this intel driver backported to fiesty?

FWIW here is my card info as identified by lspci:
00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)

[jvalen (jose-valenzuela)]

Same on Dell 630 too. The other option is upgrade to Xorg 7.3 when liberated, but till then...

I've tried different dual-screen configurations and all of then hangs too...

Have somebody probed some other driver or similar?

[Earl Malmrose (earl)]

Hmm, Dell just announced a new Ubuntu laptop with x3100 shipping next month. I wonder what drivers they used!

[Alexander Hungenberg (alexander-hungenberg-gmail)]

Well, I can confirm this for a 945gm chipset, too

[Bryce Harrington (bryce)]

This sounds very similar to bug 99114, except in that case there is a spontaneous reboot rather than a freeze.

[MRMadhav (mrmadhav2408)]

Hi, I am getting the same problem here on Gusty. I have an HP Compaq 6710b with the intel 965gm x3100 and was previously using fiesty with it. Then, i was using the vesa driver and everything was fine, i.e, no crash on 3d etc... When upgrading to gusty, i saw that support for this card was added via the xserver-xorg-video-intel driver. So i changed my xorg.conf file as it was described in some of the forums namely setting the driver to "intel" etc... But after doing that, each time i try to launch a 3d application my xserver will just hang and i am forced to power-off and boot again. Even simple test applications such as glxgears fail. I have attached a copy of my xorg.conf file

[Peter Clifton (pcjc2)]

After a crash, could you reboot and then attach the log file /var/tmp/Xorg.0.log.old

Thanks

[Bryce Harrington (bryce)]

There are also directions for obtaining a backtrace available here: https://wiki.ubuntu.com/DebuggingXorg

I'm not able to generate a crash running glxgears on a 82G965 system...

If someone able to reproduce this issue could run it through the debugger and get a backtrace it may help. Make sure you have -dbg packages installed so you can get line numbers in the sourcecode.

[Cris Amon (crisamon)]

I'm getting the same problem with Gutsy and chipset 945GM Mobile 945GM/GMS/940GML Express Integrated Graphics Controller.

[Fernando (fernandogferreira)]

Hi,

I faced some problems too with the Compaq 6510b...
I solved installing the xserver-xorg-video-intel and the compizconfig-settings-manager packages... The 3D, resolution, all working well. My xorg-conf is attached. Please notice that the resolution of the monitor is fake....

However, the dual-screen does not work properly. When I connect the external monitor the laptop's monitor resolution crashes and I can't manage how I make a extended (right-side) screen. I just take a copy from the laptop's monitor on the external one.

Some idea???

[John Affleck (jraffleck)]

I'm going to assume that this is the same problem that I'm having, in that some gnome-screensaver modules cause X to crash. I've attached a backtrace from a crashed X session (something to do with DRI locking maybe ?)

Thinkpad T61 w/ 965GM graphics:
00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)

[John Affleck (jraffleck)]

Sigh. I should actually include the attachment I guess.

[voltron81 (micar)]

I've the same problem with my laptop Dell XPS M1330 with Intel integrated GMA X3100
:(
Somebody know the solution?
Thanks
Michele

[John Affleck (jraffleck)]

Kernel DRI debug output attached. I can get this to happen consistently with 'skyrocket'.

[John Affleck (jraffleck)]

'Full' requested gdb output (which doesn't actually add any value, but..) I have libgl1-mesa-dri-dbg installed but don't get debug symbols from i965_dri automagically ?

[voltron81 (micar)]

There's something new?
I still can't use 3d effects of ubuntu... :(
Thanks
MiK

[Bergje (ronald-bergje)]

I can reproduce this 100%... each time i startup Blender 3d and actually render a window my whole X-server will hang pretty soon (I can make it do so almost immediately by dragging the render window around). I have the intel 965GM/GL integrated graphics chip X3100 (Thinkpad X61s).

I've added both the log from the xserver and a gdm backtrace (in a single file) as par the instructions i saw in an earlier message.
I hope this gives all the information needed... If not just message here and i'll provide any information necessary.

Regards - Ronald

[maor (maors)]

im using dell 1520 with intel x3100 and blender Stellarium and compiz work well (compiz is without video playback but thats a known limitation)
mybe something in your xorg.conf is causing the problem here is mine
im running on gutsy final.

[Bergje (ronald-bergje)]

I just reproduced with your xorg.conf the exact same behaviour... I can not quite explain it either. I've tried multipe xorg.conf's now and nothing seems to help. The only thing that i found so far that describes the symptoms is this bug-entry. I just hope the debug information i send might help developers to try and figure out what's going wrong.
I'm using Ubuntu Gutsy installed from the livecd... with all updates available installed etc. (32 bit version, 1 Gb internal memory (2x512).)

Thanx for your effort though!

[dpchem (peryshkov)]

To maor:

I have exactly the same xorg.conf as you posted but still get crashes :-(

Installed Ubuntu 7.10 from alternate CD

Thinkpad T61, 2Gb of memory.

xserver-xorg-video-intel 2:2.1.1-0ubuntu9

[maor (maors)]

Bergje, i looked at your xorg.log and found this line
"Error in I830WaitLpRing(), timeout for 2 seconds"
and found this bug report at xorg bugzilla:
https://bugs.freedesktop.org/show_bug.cgi?id=11847
the buttom line is that the graphic engine hangs,and after disabling dri the problem seems to go away.
try editing the intel device section in xorg.conf to:
Section "Device"
        Identifier "Intel Corporation 82G965 Integrated Graphics
Controller"
        Driver "intel"
        BusID "PCI:0:2:0"
        Option "AccelMethod" "EXA"
        Option "DRI" "false"
EndSection
note that i am not certain adding Option "AccelMethod" "EXA" is nessecery and you should avoid it sense EXA is note suppose to be production ready.
also disabling dri will probably hurt performance but thats better than having no 3d at all.

btw: if the workaround works for you consider telling them your hardware info to try and help xorg developers in finding a solution.

[Chris Irwin (chrisirwin)]

Disabling DRI avoids crashes, but now 2D is extremely sluggish (regular GTK things, not OpenGL). I'm Using metacity, not compiz fusion.

This is on a macbook with intel X3100

[maor (maors)]

2D is extremly sluggish even without Option "AccelMethod" "EXA"?

[Bergje (ronald-bergje)]

Although it's somewhat annoying that I cannot use Blender at the moment i'd rather wait till a solution for this problem is implemented then reducing all graphic performance... I need the graphic performance for other things that do not crash. As it is this bug stands on 'Confirmed' with importance 'High' so i'm sure that a solution will be available in the future. Thanks for the suggestion though!

Regards - Bergje

[tommaso gardumi (tommaso-gardumi)]

i have problems as well on my new clevo M72R "Core Logic Intel GM965 + ICH8M"
I've tried with feisty and with gutsy, the problems are several: Correct detection of the video card Intel GM965 integrated
but not of the lcd monitor 12.1" WXGA glare (1280x800) and the video output interface
with composite working it is impossible to watch videos and the screen get black after a while.
It looks like a problem of the "intel" driver and i'm currently using "vesa" to have a usable system.
I also give a try to the first alpha of ubuntu 8.04 but nothing as been fixed yet.

I believe in ubuntu and i hope this bug will be fixed
thanx
tobar

[Sander Smid (s-smid)]

Tommaso,

Please note that the GM965 chipset has issues with Xv. This means that compiz-fusion is disabled by default on Ubuntu ( see: http://wiki.compiz-fusion.org/Hardware/Blacklist )
When you enable it anyhow you should be aware that this can cause some trouble. In most cases you can work around this. The most simple way is by using the 'gstreamer-properties' configuration tool (start from terminal) and select "X Window System (No Xv)" at the video tab -> plugin selectbox. Now Totem video, rhythmbox visualisation etc. works again.

I would like to note that on my system I can work with 3D apps. I have had some lockups with certain 3d screensavers but I can use my own 3D application. There are graphical glitches (I filed a bug report a couple of weeks ago here: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/162206 ) but it is usable.

Games like warsow (http://www.warsow.net) even work with this card on my notebook.

I have an Acer Aspire 5315-052G12Mi and wrote a small howto for it on: http://ubuntuforums.org/showthread.php?t=610603
It includes some pointers to some compiz / video / webcam / wifi related issues.

Cheers,
Sander

[tommaso gardumi (tommaso-gardumi)]

thanks for responding Sanders, my actual major problem comes from th incorrect detection of my monitor 12.1" 1280x800 i attach a screenshot of how it actually looks on i gutsy live cd
i will also attach my xorg.conf

[Rubin (rubin)]

Hi all. I can confirm this crash bug also; It used to be very easy to crash is feisty, just increase the size of a simple GLX app like glxgears, and bam. Now, in gutsy, its not fixed, but has become harder to trigger the bug. A sure fire way is to do the following:

open Scorched3d
click the configuration editor button
set the size to something bigish. maybe 1280x800
turn graphics quality all the way up
start a single player game
choose apoc mode
once the game has begun, the big weapons such as doomsday, deaths head etc, will cause X to crash, and ultimately take down the whole system. (note weapons need to be aimed a little smartly to explode.. if they hit something before deploying sub-bombs it wont trigger the bug)

I have also been able to reproduce it with 2ndlife fairly regularly.

Could those of you who think the bug was fixed for you, please try running scorched3d above and verify it really is not crashing even under high GLX load?

I have a Lenovo Thinkpad T61 with the intel 965 video, using gutsy, and the 'intel' driver. Sure cant wait till these video issues are solved :/

[maor (maors)]

this bug is known for about six month and the upstream developers have not managed to reproduce it see:
https://bugs.freedesktop.org/show_bug.cgi?id=11847
i am also working with a x3100 santa rosa but it is working perfectly.
posting here about how your also suffering from the problem is not going to help.
if you want to see this bug fixed and help the linux community in general i suggest you go to the thread and try to work out what hardware-software combination is causing this so the developer can actually start working on the fix and i dont see this bug getting fixed until that happens.

i know its a hassle but in the world of linux/open source thats how things get done.

[Rubin (rubin)]

I can't speak to others, but to reproduce my problem, I can guess it would be very easy:
 Borrow a Lenovo thinkpad T61 with intel video option (very common laptop). Install ubuntu Gutsy on it. Try to use 3d. Theres nothing weird about the setup, its right out of the box. I have yet to find a 3d application with a lot of things going on that does NOT crash eventually. Its not elusive, at least on my hardware.

Seems like intel and/or lenovo could be tagged to help make this happen....

[Bergje (ronald-bergje)]

I second Rubin... I also have a lenovo thinkpad with intel video option (Thinkpad x61s). Basic installation of ubuntu Gutsy, try to use 3d (specifically openGL, i use blender to reproduce this bug) -> crash will happen very very quick. I posted xorg.conf, logs including debug information and hardware specifics.

[Rebecca Palmer (rebecca-palmer)]

https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/178451 may be a duplicate of this (if you've seen this in Gutsy, are your symptoms an exact match? my mouse pointer freezes, this bug's original (Feisty) report says it didn't)

This hardware has also been reported to crash Second Life under Windows: http://softwarecommunity.intel.com/isn/Community/en-US/forums/30244361/ShowThread.aspx The workaround there is to disable object occlusion; can this be done in X, and would this be a less drastic remedy than turning off DRI (which for me stops the crashing but slows FlightGear from 10-11fps to an unusable 1-2fps)?

[maor (maors)]

the bug is caused even while rendering the most basic 3d so i do not think its related to the windows bug you described ( i am not shore even if windows and linux share the same code base for the intel drive even if it is open source) but of course i have been wrong before...

[BrianGilman (brian-gilman)]

This appears to be fixed with hardy heron, been running without compiz for the past few days, and the bug has not appeared again.

[BrianGilman (brian-gilman)]

I spoke too soon, just occurred again, this is on a Thinkpad R61 7738-A16, under hardy heron.
Different in that it doesn't corrupt the screen, it just switches between video modes, and that it took much longer to occur. Virtual terminals do not work, and the only "safe" way to escape is ctrl-alt-del.

kern.log
Dec 30 01:53:03 r61 kernel: [28676.888712] [drm:i915_wait_irq] *ERROR* i915_wait_irq: EBUSY -- rec: 34046195 emitted: 34046256
Dec 30 01:54:01 r61 kernel: [28734.070966] gdm[5675]: segfault at 0000000c eip b77f8462 esp bfa11500 error 6

daemon.log
Dec 30 01:53:06 r61 gdm[5678]: WARNING: gdm_slave_xioerror_handler: Fatal X error - Restarting :0

[Rebecca Palmer (rebecca-palmer)]

scorched3d doesn't crash X on my machine (though when I close it it logs a segfault in the game itself), suggesting that how to trigger this bug depends on one's hardware and/or software configuration, which could also account for the developers not being able to reproduce it. Is there anything OpenGL on the liveCD (screensavers?) that we could use to try and reproduce this in a clean environment?

I have also found that the keyboard LEDs go out and do not respond to the Caps Lock key, and that after each restart the mouse pointer works, though it freezes again before the screen goes off. The attached is a kernel log containing Alt+SysRq+t traces, but while those taken during the restarts list a process named Xorg, they don't appear to have a trace for it (one taken after it had all stopped didn't list an Xorg or x-session* process at all).

Bug 154343 appears to be a duplicate of this. Bug 68150 and (if we are counting the same symptoms on the i945gm as the same bug) bug 177518 might also be, but since there seems to be more than one X crash bug on the 965, I wouldn't label them as such without the logs.

[maor (maors)]

the best we can do in regard to hardware configuration is to provide a list of laptop models that are affected which i did sense laptop based on the santa rosa platform (which include most of the laptop hardware) sometimes have the bug and sometimes they do not have the bug.
as for testing for software configuration i agree someone should test with a fresh install because i do not think the live cd has anything that activates 3d unfortunately (or maybe luckily) i cant test because i am not effected.

volunteers?

[Rebecca Palmer (rebecca-palmer)]

Bug 177518 on the i945 now appears to be a separate problem: it freezes keeping the screen contents, logs nothing and locks so hard it won't respond even to Alt+SysRq, this blanks the screen, logs Error in I830WaitLpRing() and will reboot with Alt+SysRq+b. 945 users who reported having this bug: do you really have this one, or is yours 177518?

I'd rather not do a fresh install on my hard disk (definitely not of Hardy before it's released) as I use this laptop daily, but I'd be willing to try and put together a bootable USB stick that reproducibly crashes it if the liveCD won't (the skyrocket screensaver mentioned above doesn't crash my machine, and could even be a separate bug as it logs a backtrace not Error in I830WaitLpRing()).
-Does it matter if I use Gutsy or Hardy? Does anyone even have a reproducible way to get this bug in Hardy?
-If it turns out to be more complicated than "install Ubuntu, install Blender, play around a bit", can I post the (presumably rather large) USB disk image here, and if not how do I get it to the developers?

[BrianGilman (brian-gilman)]

The bug still exists in Hardy. You can reproduce it the same way in Hardy that you can in Gutsy. The only difference(for me at least), is that instead of corrupting the screen with the mouse cursor frozen in place, the screen goes blank.

For me scorched 3d will cause it to crash pretty quickly in Hardy, running Kenta Cho's "mu-cade" will cause it to crash in several seconds. To get it to crash with Blender I've noticed that there might be some correlation with the complexity of the scene, higher complexity scenes seem to cause it to crash quicker.

[Rebecca Palmer (rebecca-palmer)]

I have now found a way to trigger this from the amd64 Ubuntu Gutsy liveCD: preview the Endgame screensaver in full screen, and wait until a rook(castle) moves (this may take a few minutes, as not all the games it shows include any). The crash occurs seconds later with the same symptoms. This works in both 1280x800 (default) and 1024x768, using the internal display.

There are also graphical artifacts (weird reflections, pieces disappearing behind others they should be in front of), but these could be a separate problem as in i386 Xubuntu Gutsy they occur but the crash doesn't (and the graphics are a lot slower).

Does this work for the rest of you? (It also works on my installed system, the reason for using the liveCD is to have a predictable software environment.) Does it work in Hardy?

[Sander Smid (s-smid)]

I discovered this bug two months ago and filed it as a serpate bug ( https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/162206 ).. Perhaps these issues are more related then we thought at first.

Well the driver is not as good as the windows version due stuff like this. Yesterday I had my first freeze when I was playing Urban Terror (www.urbanterror.net) for an hour. I'm not 100% sure but it feels like that the 3d performance decreases when your playing longer.

Gutsy gibbon here...

[Bergje (ronald-bergje)]

I can confirm Rebecca Palmer's find, but I don't run from the live CD (no cd-drive in my tiny X61s so someone else check it with a livecd copy) the 'End Game'-screen saver triggers the crash. In fact i just turned on my laptop and forgot to change screensaver back and when i came back it was in the restart loop and Xorg.0.log.old showed the bug-message.

Using Ubuntu Gutsy Gibbon 32 bit Intel dual core L7300 fully updated.

Incidentally i also noticed the graphical artifacts as described by Rebecca and as entered by Sander.

Regards - Ronald

[Ryan Sinn (ryan-sinn)]

I have a Lenovo ThinkPad X61t (model # 776298U) and I can play Enemy Territory, bzflag, lincity-ng, et al and I have not had any 3d issues... my 3d (GL) screensavers seem to work fine.

It looks like it runs the Santa Rosa chipset. This new model of Thinkpad X61t was released in October / November 2007 I believe.

I'm running Ubuntu / Kubuntu 7.10 amd64 edition...

You can view all the specs here:
http://www-01.ibm.com/common/ssi/cgi-bin/ssialias?infotype=AN&subtype=CA&htmlfid=897/ENUS107-640&appname=lenovous&language=en

Or here:
ThinkPad X61 Tablet 776298U 7762-98U 2,412
 (12.1-inch SXGA+ Super-Wide Angle TFT,
 Bluetooth Wireless Technology,
 integrated fingerprint reader,
 Intel Core 2 Duo processor L7500 (LV),
 Mobile Intel GM965 Express Chipset,
 Intel GMA X3100 Integrated Graphics,
 160 GB/5400 rpm FDE, 2 GB (2 x 1024 MB),
 Intel Wireless WiFi Link 4965AGN
 (11n disabled),
 8-cell Li-Ion battery,
 Intel Turbo memory,
 Microsoft Windows Vista Business)

Here's a review of this laptop:
http://www.tabletpc2.com/First%20Look-Lenovo_ThinkPad_X61_Tablet_PC-Article7002531.htm

[Ryan Sinn (ryan-sinn)]

I also don't run my xorg.conf with the DRI option... should I?

Here is the output of running glxinfo:

name of display: :0.0
display: :0 screen: 0
direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.2
server glx extensions:
    GLX_ARB_multisample, GLX_EXT_import_context, GLX_EXT_texture_from_pixmap,
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_copy_sub_buffer,
    GLX_OML_swap_method, GLX_SGI_make_current_read, GLX_SGI_swap_control,
    GLX_SGIS_multisample, GLX_SGIX_fbconfig, GLX_SGIX_visual_select_group
client glx vendor string: SGI
client glx version string: 1.4
client glx extensions:
    GLX_ARB_get_proc_address, GLX_ARB_multisample, GLX_EXT_import_context,
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_allocate_memory,
    GLX_MESA_copy_sub_buffer, GLX_MESA_swap_control,
    GLX_MESA_swap_frame_usage, GLX_OML_swap_method, GLX_OML_sync_control,
    GLX_SGI_make_current_read, GLX_SGI_swap_control, GLX_SGI_video_sync,
    GLX_SGIS_multisample, GLX_SGIX_fbconfig, GLX_SGIX_pbuffer,
    GLX_SGIX_visual_select_group, GLX_EXT_texture_from_pixmap
GLX version: 1.2
GLX extensions:
    GLX_ARB_get_proc_address, GLX_ARB_multisample, GLX_EXT_import_context,
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_MESA_copy_sub_buffer,
    GLX_MESA_swap_control, GLX_MESA_swap_frame_usage, GLX_OML_swap_method,
    GLX_SGI_make_current_read, GLX_SGI_swap_control, GLX_SGI_video_sync,
    GLX_SGIS_multisample, GLX_SGIX_fbconfig, GLX_SGIX_visual_select_group
OpenGL vendor string: Tungsten Graphics, Inc
OpenGL renderer string: Mesa DRI Intel(R) 965GM 4.1.3002
OpenGL version string: 1.4 Mesa 7.0.1
OpenGL extensions:
    GL_ARB_depth_texture, GL_ARB_fragment_program, GL_ARB_imaging,
    GL_ARB_multisample, GL_ARB_multitexture, GL_ARB_point_parameters,
    GL_ARB_shadow, GL_ARB_texture_border_clamp, GL_ARB_texture_compression,
    GL_ARB_texture_cube_map, GL_ARB_texture_env_add,
    GL_ARB_texture_env_combine, GL_ARB_texture_env_crossbar,
    GL_ARB_texture_env_dot3, GL_ARB_texture_mirrored_repeat,
    GL_ARB_texture_non_power_of_two, GL_ARB_texture_rectangle,
    GL_ARB_transpose_matrix, GL_ARB_vertex_buffer_object,
    GL_ARB_vertex_program, GL_ARB_window_pos, GL_EXT_abgr, GL_EXT_bgra,
    GL_EXT_blend_color, GL_EXT_blend_equation_separate,
    GL_EXT_blend_func_separate, GL_EXT_blend_logic_op, GL_EXT_blend_minmax,
    GL_EXT_blend_subtract, GL_EXT_clip_volume_hint, GL_EXT_cull_vertex,
    GL_EXT_compiled_vertex_array, GL_EXT_convolution, GL_EXT_copy_texture,
    GL_EXT_draw_range_elements, GL_EXT_fog_coord, GL_EXT_histogram,
    GL_EXT_multi_draw_arrays, GL_EXT_packed_pixels, GL_EXT_point_parameters,
    GL_EXT_polygon_offset, GL_EXT_rescale_normal, GL_EXT_secondary_color,
    GL_EXT_separate_specular_color, GL_EXT_shadow_funcs, GL_EXT_stencil_wrap,
    GL_EXT_subtexture, GL_EXT_texture, GL_EXT_texture3D,
    GL_EXT_texture_edge_clamp, GL_EXT_texture_env_add,
    GL_EXT_texture_env_combine, GL_EXT_texture_env_dot3,
    GL_EXT_texture_filter_anisotropic, GL_EXT_texture_lod_bias,
    GL_EXT_texture_object, GL_EXT_texture_rectangle, GL_EXT_vertex_array,
    GL_3DFX_texture_compression_FXT1, GL_APPLE_client_storag...

[Rebecca Palmer (rebecca-palmer)]

Endgame doesn't cause the crash in amd64 Hardy alpha2 liveCD on my system, but others have reported that this bug does still exist in Hardy. Would it be worth trying a Hardy USB stick (which would allow me to test Blender etc) or should a reproducible Gutsy crash be enough for the developers to work from? (if it is reproducible, it may be the hardware rather than software environment that makes not every 965 have this bug in the same place)

Ryan: This bug appears only with DRI on, but this is the default (as it improves performance) and it would appear from
direct rendering: Yes
that you do have it on.

Sander: I'm treating the graphical artifacts bug 162206 as separate as Xubuntu has that and not this, but I'm willing to be corrected.

[Rubin (rubin)]

I grabbed the kernel and modules from hardy, 2.6.24-3, and I have not had a single crash since. I launched about 30 apocalypse in scorched3d without a crash. (before this would crash for sure within 3 shots). Could be some changes just avoid the trigger rather than really fixing the problem.. but something to try.

[Rubin (rubin)]

Scratch that: endgame still triggers it for me. Nothing else does. Even endgame without shadows or smoothing does it within 10 seconds.

Part of the problem here is that X tries to restart several times, I think. In doing this, its overwriting the log: but I did get this from Xorg.log.0.old:
(normal startup messages)
(II) intel(0): [drm] removed 1 reserved context for kernel
(II) intel(0): [drm] unmapping 8192 bytes of SAREA 0xf8e5a000 at 0xb7b32000
(thats the end of file)

[Rubin (rubin)]

running 'nohup tail -f /var/log/Xorg.log.0 >/root/xcrash.txt' I was able to capture the actual crash log. Attached.

[Gero (gero-putzar)]

I experience this bug with Ubuntu Gutsy (Kubuntu alternate-installation on AMD64) on a Dell Optiplex 745 (a desktop computer), one graphics controller Intel 965 with one or two monitors connected. I also tried to install the newer driver package xserver-xorg-video-intel from Hardy (Version 2.2.0+git20080107-1ubuntu2 instead of 2.1.1-0ubuntu9), it makes no difference.

An excerpt from lspci:
00:02.0 VGA compatible controller: Intel Corporation 82Q963/Q965 Integrated Graphics Controller (rev 02)
00:02.1 Display controller: Intel Corporation 82Q963/Q965 Integrated Graphics Controller (rev 02)

[Rebecca Palmer (rebecca-palmer)]

The attached is a kernel log of this crash (in Blender, at 15:49:54 according to syslog) with the kernel DRM debugging option turned on.

I attempted to get a system/library call trace as well (sudo ltrace -S -C -o xorgltrace.txt -p 5859), but Xorg restarted and went back to the gdm login screen as soon as I attached ltrace to it (is this supposed to happen or is it a bug in itself?), which ended the trace.

[Rebecca Palmer (rebecca-palmer)]

While ltrace wouldn't attach without crashing Xorg, strace did work; result attached (also a Blender crash).

Are the developers still unable to reproduce this, and if so could I be of use? I have done rather a lot of programming, but never at system/driver level.

[Timo Aaltonen (tjaalton)]

Keeping the mesa component open is enough, closing intel.

[Rene Horn (the-rhorn)]

I'm having this problem too on my own machine--a new Dell XPS M1330 (they just started selling them with Ubuntu on them) that has the X3100/GM965.

It's very similar to a problem I've had on another machine of mine that has an NVidia card in it, and Debian. What happens on that machine is that whenever Xorg gets upgraded, the NVidia supplied GLX drivers get replaced, and when I start X back up, I can't run any OpenGL applications without crashing X. I eventually figured out the problem when I ran glxinfo one time, and noticed that DRI was not working.

I bring that up because on my new laptop, whenever I log into GNOME, DRI is not working, and that's immediately apparent whenever I run glxinfo. Also, glxgears immediately crashes X. Compiz works, somewhat, but is very buggy.

However, when I log into failsafe-terminal, DRI is working, and I can run glxgears just fine.

I've attached the output of glxinfo from both within GNOME and in failsafe-terminal.

I would like to see if anyone else is seeing the same symptoms with glxinfo in both of those conditions.

[Rene Horn (the-rhorn)]

Okay, I just did some checking around, and got an idea from <a href="http://ubuntuforums.org/archive/index.php/t-559350.html">here</a>, but it doesn't exactly solve our problem here.

I just realized that DRI is only on display 0.0 from that thread, and doing 'DISPLAY=:0 glxgears' allows it to run just fine without anything crashing, and 'DISPLAY=:0.0 glxinfo' actually shows DRI as running. I've attached the output for that.

Hopefully this will lead us to a solution!

[Rene Horn (the-rhorn)]

Also, just to note, glxgears will create an unmanaged window, and I have no idea why this is.

[Rebecca Palmer (rebecca-palmer)]

Thanks for trying, but I don't think it's that: my glxinfo is display :0.0 by default and reports that DRI is on, and DISPLAY=:0 blender still crashes. (glxgears doesn't crash on my system with or without that, though others have reported it triggering this bug.)

Since this bug only occurs when DRI is on, I suspect Rene's bug isn't this one at all. I have edited the description to make clearer how to distinguish this from other X crash bugs.

[Rene Horn (the-rhorn)]

Yup, Rebecca was right. Sorry everyone! It was a completely different bug related to XGL, and not this.

Anyway, I disabled XGL, and that got DRI working again. glxgears ran fine too. Then I tried running Blender, and...

Rendering does not work. It just crashes.

NOW I've run into this bug.

So, sorry again!

[Ryan Sinn (ryan-sinn)]

Is this bug related to:
https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/178292

That bug has been acknowledged at the Ubuntu level and already fixed upstream and the MESA package is supposed to be updated for the first Hardy Beta.

[Rebecca Palmer (rebecca-palmer)]

I'd guess not: 178292 logs a backtrace and restarts X, this one logs Error in I830WaitLpRing() and hangs.

While https://bugs.freedesktop.org/show_bug.cgi?id=9307 does appear to be this bug and was closed as fixed, the versions said to fix it are the ones in Gutsy, so while it might well have been fixed for that particular user's hardware/software it wasn't then fixed for everyone.

[Kevin Wang (kevin-xuepu-wang)]

I met the same issue with Hardy alpha6 on T61(santarosa platform).
steps:
1. start one glxgears
2. start blender
3. try to move glxgears, it would freezes and crashed.

attached crashed xorg.log

[redacted (redacted)]

I'd just like to add that my Thinkpad R61e exhibits all these classic symptoms. It's making it extremely difficult to learn Blender when X keeps crashing like this.

[Sander Smid (s-smid)]

Okay, the Acer Aspire 5315-052G12Mi (notebook) can be added for sure.. have had several lockups already.

Link Acer: http://global.acer.com/products/notebook/as5315.htm
Link Ubuntu Laptop Wiki: https://wiki.ubuntu.com/LaptopTestingTeam/AcerAspire5315

Still on 7.10. I'm not willing to upgrade yet since I'm working on my master thesis and I need to compile software on this notebook. But I fear that this bug is still not fixed in 8.04 :(

Keep us posted!

Sander

[ethanay (ethan-y-us)]

Dell XPS m1330 w/Intel X3100 and Ubuntu 7.10 w/latest. Running Metacity, not Compiz.

First noticed this when trying to task switch between Stellarium and Firefox. Filed a separate bug report #207980 for the Stellarium package, but looks as though it may be a more specific instance of this bug.

It does NOT occur unless I attempt to task switch.

Otherwise my experience is the same as described in the original report:
screen flashes between off and black with 'please wait' cursor about 10 times, then stops; does not respond to Ctrl+Alt+F1 or Ctrl+Alt+Backspace, but does reboot on Alt+SysRq+b

Ethan

[Julian Fietkau (xjulian)]

This also happens to me, using a Lenovo 3000 N200 (TY2BBGE), Intel GMA X3100 / GM965. I'm on the current Hardy beta, but the bug has also occured on Gutsy before I upgraded.

Running Compiz has not triggered this bug here (I use Compiz exclusively). Playing 3D games based on OpenGL reliably causes the symptoms described above after playing for some time, an hour at most. Note that I have not seen any funny colors, but the rest of the symptoms match my problem.

Sometimes the screen flashing does not happen, but instead all graphical output seems to freeze. The system does not lock up, but it "does not respond to Ctrl+Alt+F1 or Ctrl+Alt+Backspace, but does reboot on Alt+SysRq+b" (as described above). This may be a seperate issue, I'm not sure.

[STaRMaN (jarizaro)]

The some problem here. In a Phoenix laptop (gm965). I had this problem in gutsy, and now, in hardy beta 1.
Graphics is a GMA3100.

With blender crash / google earth .. opengl problem for sure.

[Jens Bäckman (czw)]

This bug can be reproduced with 100% accuracy by installing World of Warcraft. Simply start it with "wine WoW.exe -opengl", and enjoy this bug.

[STaRMaN (jarizaro)]

Will be interesting have a solution for this bug, .. it is with us from gutsy... and it should be critical.
:(

[Bryce Harrington (bryce)]

With X bugs, we reserve critical for issues that are extremely serious, such as bugs that prevent nearly all users from running X and that have no workaround available. High priority is used for crashes, freezes, and login issues. So this bug is correctly prioritized.

[Jens Bäckman (czw)]

I'm not sure it 'Option "DRI" "false"' can be called a workaround for very demanding OpenGL applications. Doing this with World of Warcraft gives 20 frames per minute. You get more performance out of a GeForce 2 MX. This bug might not be life threatening, but it sure is serious.

[ethanay (ethan-y-us)]

the Dell XPS M1330 (with the Intel integrated graphics option) is also affected by this bug.

Ethan

[Mike Roberts (m.roberts)]

inspiron 1420n here, also affected.

I have heard rumors that DRI2 will fix this, kindof a long wait though.

[John Affleck (jraffleck)]

I can confirm that it appears to be fixed in the development version of mesa. If anyone can reproduce it on the mesa development tree, I suspect the developers might be better able to fix it.

[Wouter van Heyst (larstiq)]

John Affleck: do you have instructions for how to test this with a mesa development tree? Starting point would be either gutsy or hardy I guess.

[ASDFASDF (user-487-deactivatedaccount)]

I use Hardy Beta with intel X3100 (gm965) and it appears to be fixed... I played Glest (3d game) and had Euphoria and Endgame screensavers running for about an hour without freezing.

[Kevin Wang (kevin-a-wang)]

it seems can not be reproduced with latest upstream driver (git pulled at April 7):
xf86-video-intel: 2.2.0
Xserver: 1.4.0.90
mesa: 7.0.3-rc1
drm: i915 1.13.0

[STaRMaN (jarizaro)]

I have tried now Google Earth (after that updates), and it appear doesn't crash..
maybe fixed at end? :) (we hope so!)

[Praveen (bunchofmails)]

I have seen this on my T61 with GM965, the thinkpad was running on battery and probably the screensaver (dont remember which one, but defnitely a 3D/GL based) was running at that point. The screen was totally garbled and the keyboard and the mouse dint work either. I havent reproduced this. But seems like the same issue. The OS is Hardy Release Candidate.

[Timo Aaltonen (tjaalton)]

Like I said, the bug is in mesa. Closing -intel again.

[Ryan Cresawn (jrcresawn)]

I have a Dell Vostro 1500 with Intel X3100 (965gm) graphics chip. There are two things I can do to reveal this bug. The simplest is to start Blender and render the default scene. The second is to turn on Normal Desktop Effects and watch a video on YouTube. I do not get this error in /var/log/X.org.0.log.old:

Error in I830WaitLpRing(), timeout for 2 seconds

Aside from the absence of the error message other symptoms appear identical: mouse pointer is visible and movable and I can't do anything but power the computer off.

[Bergje (ronald-bergje)]

Tested in Hardy 8.04 LTS version... Bug is still present although with some applications harder to reproduce.

Very easily reproducable using Blender... just render any window and drag the render window around a bunch. You will notice all kind of weird distortions on both the background as on the render window itself and the screen has trouble keeping track of the window resulting in serious ghosting of the render window. It will crash quite rapidly once you start this although on some occasions it takes a bit longer.

Also instead of Xorg looping and restarting indefinately it just seems to be stuck.

[batman (pete-the-batemans)]

Iv'e got Hardy 8.04 LTS on my Dell D630 - with the Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c). glxinfo indicates that direct rendering is available, but glxgears hangs the system requiring a reset. Mouse still moves but the desktop is frozen and none of the shortcut keys work.

Any testing I can do to help, just shout.

[hop (hop2005)]

Hi,

I have a Dell Inspiron 1525 which I updated to Ubuntu 8.04. With the update I am able to use the Desktop-effects.
But with these switched on, I get the crash starting flightgear within minutes.
I haven't tried without the destop-effects switched on, but I will, as soon as I am back home and will return here if
the crash would not appear...

[Mihai (ropiku)]

I have a Fujitsu Siemens S7110 with Intel 945GM.
If I have compiz on and try to view a movie (tried with totem, vlc) it crashes, I can move the mouse but the rest is frozen. Tried it with Desktop effects off and the movie has weird green stripes.

[Hermann Löns (onnog)]

Fresh installed Hardy 8.04 LTS on a Dell Latitude D830. No compiz, no other schnickschnack. Blender crashes after some minutes of using it. Sometimes after 5 minutes, sometimes after one hour. Screen is black, no key is responding. One time the mouse pointer was still visible and movable. I always have to do a reset. This is annoying.

[Nathan Vegdahl (nathanvegdahl)]

I have this problem on a Dell Inspiron 1420. It appears in both Gusty and Hardy, though in Hardy there is no busy cursor visible.

I can reproduce it pretty easily using Blender 2.45 and the attached blend file. Just load the attached blend file and orbit the view around for a minute or so, and it should freeze up pretty quickly.

[MRMadhav (mrmadhav2408)]

Still getting the same problem. At first when I upgraded to Hardy, I got an update for the intel driver. After that, the freezes that i would usually get stopped. Tried all that triggered it before, but did not happen. Thus, I thought it was solved. However, recently i installed blender, and was just creating some objects then rotating the view and it just happened. This time however, there is no cursor like there was before and on my laptop the keyboard does not work. All i can do is turn off then back on. My laptop is HP Compaq 6710b with x3100 intel graphics card and 1gb ram. Hope this bug is resolved soon...

[David Nemeskey (nemeskeyd)]

I experience the same as MRMadhav, only on an Inspiron 1420 (not n). If I use Google Earth, after a time the screen just goes black (light is on, though), and all I can do is to turn off the machine.

I tried to put the Option "DRI" "false" line to xorg.conf, but when after that I started Google Earth, the X server rebooted right away, bringing me back to the KDM login screen. Then I removed the line, so I am able to use the application again, but I guess it's just a matter of time until my system hangs again...

[Gianluca Andreotti (3rror)]

Hi,

I've an Hp Compaq 6720s with Intel 965 card.

If I disable DRI in my xorg.conf I get no error to use applications such World of Warcraft, Blender ecc... (but they run too slow), if I enable DRI, then I can't use that.

Now I'm using Gentoo and the problem is present here too, not only in Ubuntu.

My xorg version is 7.3 and my video driver (i810) version is 2.3.1 (Gentoo)

[Noodlesgc (noodlesgc)]

Using a dell inspiron 1420n with an intel 965gm, did a fresh install of Ubuntu Hardy. I am NOT affected by this bug. No matter how much blender or k3b rendering I do, I am not able to get it to lockup, I was however affected when running Gutsy. Also does not lockup even when running compiz blender and k3b all at the same time.

[Nathan Vegdahl (nathanvegdahl)]

Noodlesgc: I have the same setup as you, but I do get the bug. However, I do not get it from rendering in Blender (I presume because that's 100% CPU based, and has nothing to do with video acceleration). Nor do I get it from simple Blender scenes. I only get it with scenes that have lots of geometry.

Try loading the blender scene I attached above and navigating around in the 3d viewport (panning, orbiting, zooming) for several minutes.

[Robert Jordens (jordens)]

With the above mentioned .blend file I can crash my Lenovo X61 Tablet in minutes. Last lines of Xorg.0.log are:

(WW) intel(0): ESR is 0x00000011, page table error
(WW) intel(0): PGTBL_ER is 0x00010000, overlay GTT PTE
(WW) intel(0): PRB0_CTL (0x0001f001) indicates ring buffer enabled
(WW) intel(0): PRB0_HEAD (0x13e0e238) and PRB0_TAIL (0x00010328) indicate ring buffer not flushed
(WW) intel(0): Existing errors found in hardware state.

[Mike Roberts (m.roberts)]

Nathan Vegdahl:
You were right, that blend file managed to lockup my system faster than I could say: "I hate proprietary software".
Also even with a simple object e.g. the default cube, it crashed instantly when I dragged a nautilus window on top of it.
BTW, for my "Device" section in my xorg.conf all I see is "Configured Video Card" is that normal? I think its something new as I have only seen it in Hardy.

[Motin (motin)]

The latest comment in the linked freedesktop bug looks promising:

  ------- Comment #38 From Taiyo Rawle 2008-06-02 00:39:04 PST -------

Using the latest mesa git appears to fix this issue. I was unable to recreate
the crash in blender. I shall try to recreate the crash with other 3D apps when
I next have time.

[slayer (antoniochiaravalloti)]

i confirm this bug. So, ahs anyone found a solution?

[giorgosg (giorgosg)]

slayer: It has been over a year with no solution for this bug. I suggest you return the laptop and get an other model if you can.
Motin: I tried the latest git of mesa/i810 driver about the time of your message and it didn't fix the problem for me.

[Mike Roberts (m.roberts)]

Actually, tomorrow (06/17/08) will mark one year since the opening of this bug. This bug seems to be reported just about everywhere, and even multiple times at each place. The folks over at #228526 seem be happy with their fix. I will wait until Intrepid, if it is not fixed by then, well, ....

[Motin (motin)]

@Noodlesgc: Are you suggesting that https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/228526 is a duplicate of this bug? Does the proposed Intrepid driver fix the problem for you?

@giorgosg: Ah, too bad that it didn't solve it for you... And about the laptop suggestion: I was actually just about to buy a MacBook or a Zepto with the Intel X3100 but partially due to this bug I instead chose a laptop that came with Nvidia graphics instead.

[Bela MIHALIK (bela-mihalik)]

On this page ( https://bugs.freedesktop.org/show_bug.cgi?id=11847 ) they say
this bug is fixed. I downloaded new mesa_7_0_branch[1] from git, and
new libdrm[2] with kernel modules. I patched mesa branch with ubuntu patches[3],
compiled successfully under ubuntu hardy, but glxinfo reports:

Error: couldn't find RGB GLX visual

It seems it needs more knowledge to upgrade MesaLib-7.03~rc2-ubuntu3
to a newer one.

----------------- sources -----------------
[1] git clone git://anongit.freedesktop.org/git/mesa/mesa mesa_7_0_branch
[2] git clone git://anongit.freedesktop.org/git/mesa/drm
[3] http://archive.ubuntu.com/ubuntu/pool/main/m/mesa/mesa_7.0.3~rc2-1ubuntu3.diff.gz

--------------- Xorg.0.log ---------------
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 9, (OK)
drmOpenByBusid: Searching for BusID pci:0000:00:02.0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 9, (OK)
drmOpenByBusid: drmOpenMinor returns 9
drmOpenByBusid: drmGetBusid reports pci:0000:00:02.0
(EE) AIGLX error: dlsym for __driCreateNewScreen_20050727 failed (/usr/lib/dri/i965_dri.so: undefined symbol: __driCreateNewScreen_20050727)
(EE) AIGLX: reverting to software rendering

---
turning off AIGLX not help.

any suggestion?

[giorgosg (giorgosg)]

I tried again just now and get the same error... propably the tree is broken for now.
some more info about my system:
HP compaq 6510b with gentoo (the x11 git packages pull from master)
there are also a lot of graphics gliches that were also there the last time I tried mesa/intel driver/libdrm/drm modules from x11 git.

[Mike Roberts (m.roberts)]

@Motin They might not be exact duplicates, but they certainly seem related.
I have not tried using the latest Mesa drivers. Will they be updated in the repository if it fixes this bug? Or will it not be updated until the Intrepid release?

[Nathan Vegdahl (nathanvegdahl)]

Today I compiled and installed the latest mesa from git. I am now unable to reproduce the problem using the blend file I posted above (hooray!).
If anyone else can confirm, that would be awesome. I'm hoping this really is fixed now! :-)

[Nathan Vegdahl (nathanvegdahl)]

On further testing it seems like everything is behaving identically to when I simply added:

Option "DRI" "false

to my xorg.conf (and I did check to make sure I hadn't accidentally left it like that). Meaning that there are no crashes, but almost everything that uses OpenGL (Stepmania, Glest, Warzone 2100) now runs unusably slow. I didn't notice it in Blender because it's a very complex scene in the first place, and it's always slow (just a matter of degree).

I'm not sure if this is actually related or not, but I figured I'd point it out here. Makes me wonder if the latest mesa git doesn't just disable dri by default. If so, I wouldn't exactly call this bug fixed.

[Bela MIHALIK (bela-mihalik)]

I got some help on forum that i mentioned above, and i can use mesa_7_0_branch now.
It is working on my system. I can use blender without crash, and it is fast.
I think this bug is fixed.

----- sources & compile process -----

# git clone git://anongit.freedesktop.org/git/mesa/drm
# cd drm
# make install
# make -C linux-core install

# git clone git://anongit.freedesktop.org/git/xorg/driver/xf86-video-intel
# cd xf86-video-intel
# make install

# git clone git://anongit.freedesktop.org/git/mesa/mesa
# cd mesa
# git checkout -b mesa_7_0_local origin/mesa_7_0_branch
...edit configs/default file
     INSTALL_DIR=/usr
     DRI_DRIVER_INSTALL_DIR=/usr/lib/dri
# make linux-dri-x86-64
# make install

[Sander Smid (s-smid)]

Bela,

Sounds to good to be true... a fix after all these years ;)
Could you tell me how you build xf86-video-intel? When I do ./autogen.sh
I get the following error:

checking for sys/mman.h... yes
checking for mprotect... yes
./configure: line 21292: syntax error near unexpected token `XINERAMA,'
./configure: line 21292: `XORG_DRIVER_CHECK_EXT(XINERAMA, xineramaproto)'

I did some google search and the closest hit I found was:
http://<email address hidden>/msg33497.html

But red hat systems have other package names and I don't know which to
install on ubuntu..

any help would be appriciated!
Sander

Bela MIHALIK wrote:
> I got some help on forum that i mentioned above, and i can use mesa_7_0_branch now.
> It is working on my system. I can use blender without crash, and it is fast.
> I think this bug is fixed.
>
> ----- sources & compile process -----
>
> # git clone git://anongit.freedesktop.org/git/mesa/drm
> # cd drm
> # make install
> # make -C linux-core install
>
> # git clone git://anongit.freedesktop.org/git/xorg/driver/xf86-video-intel
> # cd xf86-video-intel
> # make install
>
> # git clone git://anongit.freedesktop.org/git/mesa/mesa
> # cd mesa
> # git checkout -b mesa_7_0_local origin/mesa_7_0_branch
> ...edit configs/default file
> INSTALL_DIR=/usr
> DRI_DRIVER_INSTALL_DIR=/usr/lib/dri
> # make linux-dri-x86-64
> # make install
>
>

[Bela MIHALIK (bela-mihalik)]

> Sounds to good to be true... a fix after all these years ;)
> Could you tell me how you build xf86-video-intel? When I do ./autogen.sh
> I get the following error:
>
> checking for sys/mman.h... yes
> checking for mprotect... yes
> ./configure: line 21292: syntax error near unexpected token `XINERAMA,'
> ./configure: line 21292: `XORG_DRIVER_CHECK_EXT(XINERAMA, xineramaproto)'
>
> I did some google search and the closest hit I found was:
> http://<email address hidden>/msg33497.html

I don't know why it is not compiling on your system.
May be some development packages are missing.
Tty to install all xinerama development packages:

#sudo apt-get install xinerama-dev*

If you attach your config.log that is generated in xf86-video-intel directory,
may be i can tell more.

Bela

[Zachary Sandberg (zachary-sandberg)]

I am using a Thinkpad Z60t and am experiencing intermittent lockups in open-gl applications also. This laptop has a GMA915 chip. I was getting more consistent lockups using KDE's transparency features than I am without however.

[David Nemeskey (nemeskeyd)]

Is there a schedule to include the X.org fix anytime soon?

[David Nemeskey (nemeskeyd)]

I have a hard time understanding why nothing is happening with this bug.

It has high priority so why noone is even assigned to it?! And if it has been fixed upstream (like 3 weeks ago), how come the the solution is not realeased (again, high importance). It's not as if the bug had to be fixed in Ubuntu, someone just has to do what Bela did and release it. At least give us a schedule...

Bela:
Actually I tried to compile it, but failed. Drm and the intel driver were ok, but than mesa failed, as it couldn't find something like DrmObject or so. I guess it needs the drm include headers, but if the directory structure is alright, shouldn't it find them in the drm directory?

I would be grateful for any help.

[Nathan Vegdahl (nathanvegdahl)]

David:
   I did some searching, and it seems if you delete the i965tex directory from src/mesa/drivers/dri/ it should compile. I have no idea why it's even committed in the repository if its causing the build to fail. But if you delete it, the build will succeed.

Bela:
   I followed your instructions, and successfully installed the latest drivers etc., but Blender is still slower than normal, and according to glxinfo I no longer have DRI, and many OpenGL accelerated apps (Glest and StepMania, for example) won't even start anymore.

[David Nemeskey (nemeskeyd)]

Nathan:
Thanks, I will try it, but... isn't it slow (and is w/o DRI), because you deleted the dri directory? :) If yes, then I guess there has to be another solution...

[Nathan Vegdahl (nathanvegdahl)]

David:
From what I read (I can't find the link to what I read anymore, sorry) I believe the i965tex directory is for something more specific than dri in general. You should still leave the plain i965 (without the "tex") there.

[Bela MIHALIK (bela-mihalik)]

I uploaded the sources that i used to this place:

http://mdo.hu/ubuntu/intel-video-20080623-ubuntu804-64.tgz

The packages contains the compiled binaries for Ubuntu 8.04 64bit system also.

[Nathan Vegdahl (nathanvegdahl)]

Bela:
   I downloaded the sources you just linked to, and successfully compiled and installed them. No change in the behavior on my system. Blender is still slower than before, and dri is still disabled, etc.

[Bela MIHALIK (bela-mihalik)]

Hmm... Let see what is the difference between our systems!

I have an optimized version of Blender3D.
I downloaded official Blender 2.46 sources from blender.org,
i created a user-config.py in the root directory of the sources
(file attached).

Also i has gcc-4.3.1, because it uses more feature of Core2 CPU core.
However i compiled all intel-video and mesa sources with gcc-4.2.3
comes with ubuntu. I compiled only Blender with gcc-4.3.1

You can download gcc sources from http://gcc.gnu.org/
One mirror example: ftp://ftp.fu-berlin.de/unix/languages/gcc/releases/gcc-4.3.1/gcc-4.3.1.tar.bz2
I configured gcc with the following line: ./configure --prefix=/opt/gcc-4.3.1 --enable-languages=c,c++,objc

-----
But why you don't see DRI working with glxinfo... i don't know.
Try to look libdrm files in /lib directory, and delete or move
them to /usr/lib directory. Also check /usr/local/lib.

What you see in your /var/log/Xorg.0.log file? xorg loaded your dri drivers succesfully?

[redwasp (creatinga)]

I'm on a dell inspiron 1420n, and it only crashes in 3D apps. The mouse locks up, and the screens' colors literally invert. Ctrl+Alt+Bksp doesn't do anything, so I have to hit the power switch, which takes a minute or two to shut down the computer. This hadn't happened until I upgraded to Hardy.

The graphics card, from lspci:
00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)

[JPFGuambe (ghlegs)]

I have the same issue, it crash in some compiz animations.

I'm running on Lenovo T61 2.0GHz, 2GB RAM.
OS: Ubuntu 8.04
Linux Kernel: 2.6.24-19-generic
Graphic card: 00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)

[Anton W (anton-mbox)]

I am running ubuntu 8.04 on my TM 6292 with no problem.

[Mike Roberts (m.roberts)]

Anton: are you using the latest (git) mesa libraries? Have you tried panning, zooming and moving around the .blend file that was uploaded here?

[Anton W (anton-mbox)]

I've installed blender and run your file, playing it for a while and it failed :( sorry, i am using default ubuntu graphic libraries that comes with the installation cd..how to install latest (git) mesa libraries :D

[Mike Roberts (m.roberts)]

Anton: You have to install the git package and checkout from the Mesa git repository, compile it, and install it. I am just gonna wait until Intrepid.

[STaRMaN (jarizaro)]

Hi, i'm trying to automatize the Bela commands, with a script, when it's ready i'll upload it. But while, i have a problem with last step of compile mesa..
in a lot of files of mesa i have:

     ................
makedepend: warning: ../common/dri_util.c (reading /usr/include/sys/mman.h, line 26): cannot find include file "stddef.h"
not in ../intel/stddef.h
        not in ./stddef.h
        not in ../../../../../src/mesa/drivers/dri/common/stddef.h
        not in server/stddef.h
        not in ../../../../../include/stddef.h
        not in ../../../../../include/GL/internal/stddef.h
        not in ../../../../../src/mesa/stddef.h
        not in ../../../../../src/mesa/main/stddef.h
        not in ../../../../../src/mesa/glapi/stddef.h
        not in ../../../../../src/mesa/math/stddef.h
        not in ../../../../../src/mesa/shader/stddef.h
        not in ../../../../../src/mesa/swrast/stddef.h
        not in ../../../../../src/mesa/swrast_setup/stddef.h
        not in ../../../../../src/egl/main/stddef.h
        not in ../../../../../src/egl/drivers/dri/stddef.h
        not in /usr/local/include/stddef.h
        not in /usr/local/include/drm/stddef.h
        not in /usr/include/stddef.h
makedepend: warning: i915_program.c (reading /usr/include/strings.h, line 29): cannot find include file "stddef.h"
        not in ../intel/stddef.h
           ...........

I don't know what is the package i need. I installed build-essential, kernel-header, linux-headers, libc6-dev, ...

Could anyone help me with this? Thanks.

[Bela MIHALIK (bela-mihalik)]

Try to look after on: http://packages.ubuntu.com under "Search the contents of packages" section.

http://packages.ubuntu.com/search?searchon=contents&keywords=stddef.h&mode=exactfilename&suite=hardy&arch=any

[STaRMaN (jarizaro)]

Thanks Bela, i looked for that file with "apt-file search ...", but at end, the problem was not this, it was with i915tex directory. I removed it how lets in http://www.mail-archive.com/dri-devel%40lists.sourceforge.net/msg32697.html , and now its compiled. I'm going to test now with blend file. :)

[STaRMaN (jarizaro)]

All working well. I upload the script.
process to run it:

1) download in you home directory.

2) set executable:
    chmod +x ~/compilar_mesa.sh

3) execute it.
    ~/.compilar_mesa

after all done, restart the computer or xorg xerver.

With blender from ubuntu packages i have a segmentation fault, with blender downloaded from blender.org all works ok.

I hope this bug be fixed soon in ubuntu packages.

[STaRMaN (jarizaro)]

I forgot it. I had to add:

Section "DRI"
    Mode 0666
EndSection

in my /etc/X11/xorg.conf
If you don't have it, add it.

i will update the script to check this, and add that section if is neccesary.

[JPFGuambe (ghlegs)]

Nice work J. Rodríguez, but i'm having some problems running you script.

When i do ./compilar_mesa.sh i get this:
Reading package lists... Done
Building dependency tree
Reading state information... Done
Note, selecting x11proto-xinerama-dev for regex 'xinerama-dev*'
Note, selecting libxinerama-dev for regex 'xinerama-dev*'
x11proto-xinerama-dev is already the newest version.
libxcb-xinerama0-dev is already the newest version.
git is already the newest version.
git-core is already the newest version.
git-svn is already the newest version.
git-cvs is already the newest version.
automake is already the newest version.
autoconf is already the newest version.
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
----------- STEP 1 -----------
Initialized empty Git repository in /home/jpfguambe/mesa/drm/.git/
remote: Counting objects: 32801, done.
remote: Compressing objects: 100% (11882/11882), done.
remote: Total 32801 (delta 25617), reused 26609 (delta 20562)
Receiving objects: 100% (32801/32801), 10.64 MiB | 114 KiB/s, done.
Resolving deltas: 100% (25617/25617), done.
autoreconf: Entering directory `.'
autoreconf: configure.ac: not using Gettext
autoreconf: running: aclocal
autoreconf: configure.ac: tracing
autoreconf: configure.ac: not using Libtool
autoreconf: running: /usr/bin/autoconf
configure.ac:28: error: possibly undefined macro: AC_DISABLE_STATIC
      If this token and others are legitimate, please use
m4_pattern_allow.
      See the Autoconf documentation.
configure.ac:29: error: possibly undefined macro: AC_PROG_LIBTOOL
autoreconf: /usr/bin/autoconf failed with exit status: 1
make: *** No rule to make target `install'. Stop.
make: Entering directory `/home/jpfguambe/mesa/drm/linux-core'
make -C /lib/modules/2.6.24-19-generic/source SUBDIRS=`pwd`
DRMSRCDIR=`pwd` modules_install
make[1]: Entering directory `/usr/src/linux-headers-2.6.24-19-generic'
  DEPMOD 2.6.24-19-generic
make[1]: Leaving directory `/usr/src/linux-headers-2.6.24-19-generic'
make: Leaving directory `/home/jpfguambe/mesa/drm/linux-core'
...

Can someone tell me What's wrong?
gcc version: 4.2.3

[STaRMaN (jarizaro)]

JPFGuambe, missing libtool package.
I added that package in script.
download and try again. :)

[STaRMaN (jarizaro)]

Missing xorg-dev too.
updated the script.

[JPFGuambe (ghlegs)]

Ok, i will try it again right now!

[JPFGuambe (ghlegs)]

COOOOOOOOOOOLLLLLLLLL!
I thought it was too good to be true, but now i got it working!

Special Thanks to J. Rodríguez, and all forum folks.
Finally i got compiz fusion, my games and other stuff working just like
i wanted.
That's Goooood :)

[STaRMaN (jarizaro)]

I'm happy works now :)
I updated again the script for automatically add DRI section to xorg.conf if not exists.

[JPFGuambe (ghlegs)]

Good work, That's good!

[STaRMaN (jarizaro)]

updated again, last version (i hope)
  Added more info about what is doing the script (for debug).
See you.

[Alberto Rojas Colvin (emeteco)]

Hi!

I had been following up this bug since january 2008, because mi vaio cr160 it's also affected:

- Procesador Intel® Core™ 2 Duo T7100 (1.80GHz1)
- Intel® PRO wireless 4965AGN network connection
- Chipset Intel® GM965
- Ubuntu 7.10

I tried to run the last version script, previosly deleting i915text file, but nothing happens and the terminal window closes.
I'm very new using linux distributions, can anyone display a step-by-step solution for newbies like me?

Thank's a lot.

[STaRMaN (jarizaro)]

Hi Alberto, to use the script, you need to download the file to your home directory , if your user is "alberto", then download the file to /home/alberto

Later, open a terminal, (in kde is konsole, in gnome i don't now its name, maybe gterminal o something similar, or xterm...). Then , write in terminal :
cd /home/alberto [press enter] (change alberto for your username)
chmod +x compilar_mesa.sh [press enter]
./compilar_mesa.sh [press enter]

and then, only have to wait, if password is prompted, enter your user's password.
when all process terminate, restart your computer.

[Alberto Rojas Colvin (emeteco)]

Thanks, but there is no changes. After running the script the terminal results are attached.

what i'm missing?

[Jens Bäckman (czw)]

"orden no encontrada"... My knowledge in Español is severely limited, but I guess this means "command not found". In that case, you don't have Git, the distributed revision control application, installed.

[STaRMaN (jarizaro)]

Yes Jens, "orden no encontrada" is command not found, :) .

Alberto: the problem you are having is that you have dependences not installed, and then, apt-get in the script is not working.

open a terminal, and write:

apt-get update [enter]
apt-get install -f [enter] , and press "y" or "s", to install the packages neccessaries.

The -f option in apt-get is for fix dependences.

Later, try to run the script again.
the "git" binary is in the git-core package, which is installed with the script.

[STaRMaN (jarizaro)]

sorry... i forgot sudo . You have to put sudo before the 2 commands before.

sudo apt-get update [enter]
sudo apt-get install -f [enter]

and then run the script.

[Alberto Rojas Colvin (emeteco)]

Thanks but the second command indicates no dependency problems, and nothing happens,

The script it's not working in this part:

The following packages have unacomplished dependencies:
  xorg-dev: Depends: libxfont-dev but its not going to be installed
E: Broken packages

in synaptic i cant install xorg-dev-->libxfont-dev: 'depends on: libxfont1 (=1:1.3.0-0ubuntu1) pero se va a instalar 1:1.3.0-0ubuntu1.1'

[Julian Fietkau (xjulian)]

Sorry for asking this question here, but:

If I used this script to upgrade mesa (which seems to have worked well), are there any precautions I need to take for regular ubuntu updates? Or is it safe to install e.g. an update for xserver-xorg-video-intel? Anything I need to look out for when updating?

[yagi858 (yugo858)]

I followed the script. My question is the same as Julian: what to do for further updates ?
But, another question is: now I have the mesa folder in the home directory. Do have to keep it in that position or I can remove as everything is right installed ?
Many Thanks

[yagi858 (yugo858)]

Sorry I forgot to add. This script solve the problem I had for a long time running 3d games by wine. But also puzzle games like Zuma used to crash the system. Now all is OK !
My laptop is HP Pavillion dv2030, T7300, graphical chipset GM965

Thanks again all forum folks

[STaRMaN (jarizaro)]

Yagi858:
     You can remove the mesa folder in your home directory without problems.

Julian, Yagi858:
     About updates, i think if you install a package, the package overwrite modules compiled by script.
I have the package xserver-xorg-video-intel pendding of install. The better option is install the package, check if this bug is corrected with the package, (blender, games...)... and then, if all is correct, then, all done. else, launch again the script to download and compile mesa again.

If someone was installed the last xserver-xorg-video-intel package... Could say if the package solved this bug?

[Mike Roberts (m.roberts)]

I have all the latest updates (including backports), and the bug is not fixed. I don't think this will be updated in the repositories until Intrepid.

[Chris Lord (cwiiis)]

Just so people know, there doesn't seem to be any need to update the display driver, just mesa and the DRI driver (and libdrm). Unfortunately, current git master has a pretty massive bug that breaks display of textures when auto-generation of mip-maps is used: https://bugs.freedesktop.org/show_bug.cgi?id=17077 (there are more cases than I've reported in that bug, but that's the only very easily reproducible one)

I have updated packages at http://chrislord.net/files/mesa-bits/ anyway - warning, these are compiled with only the intel drivers (no ati/sis/etc.) - If you install these, upgrades won't break and it'll probably upgrade to intrepid ok, but it may not :)

[Arthur V (arthur00)]

I couldn't make the x86 driver, I got this on Step 2 right on the point where it runs make:

Making install in uxa
make[1]: Entering directory `/root/mesa/xf86-video-intel/uxa'
../doltcompile gcc -DHAVE_CONFIG_H -I. -I.. -Wall -Wpointer-arith -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -fno-strict-aliasing -I/usr/include/xorg -I/usr/include/pixman-1 -g -O2 -MT uxa.lo -MD -MP -MF .deps/uxa.Tpo -c -o uxa.lo uxa.c
In file included from uxa.c:37:
uxa-priv.h:158: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘uxa_screen_key’
uxa.c:42: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘uxa_screen_key’
uxa.c: In function ‘uxa_pixmap_is_offscreen’:
uxa.c:101: warning: implicit declaration of function ‘dixLookupPrivate’
uxa.c:101: warning: nested extern declaration of ‘dixLookupPrivate’
uxa.c:101: error: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c:101: error: (Each undeclared identifier is reported only once
uxa.c:101: error: for each function it appears in.)
uxa.c: In function ‘uxa_prepare_access’:
uxa.c:147: error: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c: In function ‘uxa_finish_access’:
uxa.c:167: error: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c: In function ‘uxa_close_screen’:
uxa.c:320: error: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c: In function ‘uxa_driver_init’:
uxa.c:425: warning: implicit declaration of function ‘dixSetPrivate’
uxa.c:425: warning: nested extern declaration of ‘dixSetPrivate’
uxa.c:425: error: ‘uxa_screen_key’ undeclared (first use in this function)
make[1]: *** [uxa.lo] Error 1
make[1]: Leaving directory `/root/mesa/xf86-video-intel/uxa'
make: *** [install-recursive] Error 1

Also, my git mesa transfer keeps stucking around 50% (simply stops transfering out of the blue). Any help? =(

[Arthur V (arthur00)]

Forget about my last comment, I kept insisting and it worked =)

But as I mentioned, I couldn't get my x86 driver to compile. With the mesa and drm correctly compiled (I ran the script and it seemed to work fine), I stressed my ThinkPad T61 with compiz fusion + opengl screensavers. I was almost tired of going to screen saver and returning (usually thats how it locks up here) when it finally froze. The same way as usual, kernel still logs entries (when i turn on and off and bluetooth by using Fn+F5) but I can't switch terminals or anything.

Apparently it took longer to freeze, but my issue remains =(

[STaRMaN (jarizaro)]

Arthur, for me and other users apears its working well . Did you restart your computer after run the script?

I'm not using compiz, only kdm with 3d apps. (maybe a problem with compiz..) try disabling it..

[João Pinto (joaopinto)]

I am verifying this bug on a Lenovo T60
Intel Corporation Mobile 945GM/GMS/GME, 943/940GML Express Integrated Graphics Controller (rev 03)

X will freeze (CTRL-ALT-F1, ALT-BS do not work) randomly, I am also to reproduce it when trying to start the opengl game TileRacer .

[ricosushi (ricosushi)]

Got the same problem in Dell inspiron 1525, freezes up a lot by compiz and by opengl apps. I ran the script and everything seemed fine except:

----------- STEP 1 -----------
------ Compiling DRM ---------
destination directory 'drm' already exists.
./compilar_mesa.sh: line 28: ./autogen.sh: No such file or directory
make: *** No rule to make target `install'. Stop.
make: *** linux-core: No such file or directory. Stop.

Any help on this?, will the script work even though it didn't access those directories? Should I run the script again with sudo?

Thanks for any help.

[Mike Roberts (m.roberts)]

João: If you have a intel 945 card, you are most likely affected by bug:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/177518
the 945 and the 965 have very similar bugs. It looks like the 945 bug has also been fixed and will be released in Intrepid.

[STaRMaN (jarizaro)]

ricosushi:
  Try to remove your mesa directory (in your home), and launch the script again. maybe drm was not correct downloaded.
It's not neccesary run the script with sudo, the script uses the use command when is needed.

[Arthur V (arthur00)]

STaRMaN, I have been running my system for a week with compiz fusion, and I used it for 5 days with no locks. Today it locked twice and I disabled compiz fusion. When I disable it, it NEVER freezes (I tested for quite a while). It ALWAYS locks when I unlock the screen by inserting my password. =(
Reminding that I had issues compiling the x86 driver on my T61

[STaRMaN (jarizaro)]

Hi Arthur... maybe a problem with compiz+opengl. If your system works ok without compiz.. then the more easy solution is disable compiz.

From I put the script, my system never hang again. (i am not using compiz).
I can't help you, sorry... :(

[superFerra (ferrarir)]

@Arthur V
how did you solve this?:

Making install in uxa
make[1]: Entering directory `/root/mesa/xf86-video-intel/uxa'
../doltcompile gcc -DHAVE_CONFIG_H -I. -I.. -Wall -Wpointer-arith -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -fno-strict-aliasing -I/usr/include/xorg -I/usr/include/pixman-1 -g -O2 -MT uxa.lo -MD -MP -MF .deps/uxa.Tpo -c -o uxa.lo uxa.c
In file included from uxa.c:37:
uxa-priv.h:158: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘uxa_screen_key’
uxa.c:42: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘uxa_screen_key’
uxa.c: In function ‘uxa_pixmap_is_offscreen’:
uxa.c:101: warning: implicit declaration of function ‘dixLookupPrivate’
uxa.c:101: warning: nested extern declaration of ‘dixLookupPrivate’
uxa.c:101: error: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c:101: error: (Each undeclared identifier is reported only once
uxa.c:101: error: for each function it appears in.)
uxa.c: In function ‘uxa_prepare_access’:
uxa.c:147: error: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c: In function ‘uxa_finish_access’:
uxa.c:167: error: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c: In function ‘uxa_close_screen’:
uxa.c:320: error: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c: In function ‘uxa_driver_init’:
uxa.c:425: warning: implicit declaration of function ‘dixSetPrivate’
uxa.c:425: warning: nested extern declaration of ‘dixSetPrivate’
uxa.c:425: error: ‘uxa_screen_key’ undeclared (first use in this function)
make[1]: *** [uxa.lo] Error 1
make[1]: Leaving directory `/root/mesa/xf86-video-intel/uxa'
make: *** [install-recursive] Error 1

I'm trying to compile linuxintel driver for my X3500 videocard. (Asus P5E-VM HDMI)
i'm able to compile and install DRM
but i'm stuck compiling xf86 drivers...
same problem as you had ...
Missing dep? How di you solved this?

[Glauco (glauco-hass)]

I'm with the same erros as Arthur and superFerra, output from script bellow:

Making install in uxa
make[1]: Entrando no diretório `/home/glauco/mesa/xf86-video-intel/uxa'
../doltcompile gcc -DHAVE_CONFIG_H -I. -I.. -I/usr/include/xorg -I/usr/include/pixman-1 -I/usr/local/include -I/usr/local/include/drm -Wall -Wpointer-arith -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -fno-strict-aliasing -I/usr/include/xorg -I/usr/include/pixman-1 -g -O2 -MT uxa.lo -MD -MP -MF .deps/uxa.Tpo -c -o uxa.lo uxa.c
In file included from uxa.c:37:
uxa-priv.h:158: erro: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘uxa_screen_key’
uxa.c:42: erro: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘uxa_screen_key’
uxa.c: Na função ‘uxa_pixmap_is_offscreen’:
uxa.c:101: aviso: implicit declaration of function ‘dixLookupPrivate’
uxa.c:101: aviso: nested extern declaration of ‘dixLookupPrivate’
uxa.c:101: erro: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c:101: erro: (Each undeclared identifier is reported only once
uxa.c:101: erro: for each function it appears in.)
uxa.c: Na função ‘uxa_prepare_access’:
uxa.c:147: erro: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c: Na função ‘uxa_finish_access’:
uxa.c:167: erro: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c: Na função ‘uxa_close_screen’:
uxa.c:320: erro: ‘uxa_screen_key’ undeclared (first use in this function)
uxa.c: Na função ‘uxa_driver_init’:
uxa.c:425: aviso: implicit declaration of function ‘dixSetPrivate’
uxa.c:425: aviso: nested extern declaration of ‘dixSetPrivate’
uxa.c:425: erro: ‘uxa_screen_key’ undeclared (first use in this function)
make[1]: ** [uxa.lo] Erro 1
make[1]: Saindo do diretório `/home/glauco/mesa/xf86-video-intel/uxa'
make: ** [install-recursive] Erro 1

I faced this issue with mesa because Stellarium doesn't work. Both my notebook with Intel GM965 and desktop with GeForce 6600 fail to run that application. I'm trying to solve the mesa bug in notebook with this script, GM965 video card.

Output from lspci:
00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 03)

[STaRMaN (jarizaro)]

Mmmm, there are changes in xf86-video-intel

Before this changes, my script was running ok.
I found a thread about this changes and this compile error in http://ubuntuforums.org/showthread.php?p=5595336

I deleted my ~/mesa directory, arthur if you didn't delete it, you could send me that directory compressed?. I can change the script for no download , only compile the sources from the compress file.

Arthur, open a terminal in your home, and run:
sudo tar -cvzf mesa.tar.gz mesa/

after, please send me the file mesa.tar.gz (created in your home), to starman___ (at) hotmail .com (3 underlines).

With that, i could attach the script with that file, and modify the script for no download nothing, only compile that sources.

[onion (richardkhe-deactivatedaccount)]

Does anybody have a solution on this?

Specs:
Hardware: Asus A8e
Graphics: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller
OS: Ubuntu Hardy Heron (updated to latest)

Software Tested:
alien arena
Nexuiz
...and asuming all other 3D games

Solution tested:
compilar script by STaRMan
setting DRI to False

User level:
Ubuntu Newbie, just trying to stay away from Micro$oft

STaRMan: I have tried your script(compilar_mesa.sh) on this too but the problem still persist just as what your last post says. Even turning DRI to False has problems for me too, the apps wont start. By the way, I still haven't removed my ~/mesa directory if you need it.

[nikolai (nikolaihecker)]

Hi, I had the same problem running alien arena and gltron on a Dell Inspiron 1525. After I had deactivated desktop effects, everything works fine with DRI enabled.
( "System" -> "Preferences" -> "Appearance" -> "Visual Effects" -> "None")

[nikolai (nikolaihecker)]

I'm afraid I should not say it's working fine. The 3D performance is way too slow. I get about 880 fps in glxgears (using a Dell Inspiron 1525 Core 2 Duo 2.0 Ghz and 3 GiB RAM) whereas even my old laptop with a sempron 3100+ (1.8 Ghz), 640 MB RAM and 128 MB ATI Radeon Express 200M (shared memory) which is not known to be fast, makes it to about 1100 fps.

This appears to be a problem with the default driver used in Hardy. I'm not able to compile the appropriate driver offered by Intel. I was only able to compile the DRM. After installing the DRM Kernel module I was not able to apply direct rendering. For this reason I reinstalled Hardy. I use the default driver without desktop effects and slow performance rather than no direct rendering.

[Mike Roberts (m.roberts)]

nikolai:
Do you have the package xserver-xgl installed?
I have found that when using the Xgl xserver 2D and 3D performance is very slow and choppy.

[nikolai (nikolaihecker)]

No, xserver-xgl is not installed.

Did you edit your xorg.conf? Since changes in my xorg.conf did not increase performance, mine is almost empty (default after Ubuntu installation):

...
Section "Device"
 Identifier "Configured Video Device"
EndSection

Section "Monitor"
 Identifier "Configured Monitor"
EndSection

Section "Screen"
 Identifier "Default Screen"
 Monitor "Configured Monitor"
 Device "Configured Video Device"
EndSection
...

[houstonhilton74 (houstonhilton74)]

I have an Acer Extensa model 4620, and I also have the same graphics card (Intel 358 Megabyte X3100), and have the same problem. Almost all of my 3D games, like Super Tux Kart, just crash after some time! I am using Ubuntu 8.04.1, Hardy Haron. All I can say is, I hope there are no proprietary drivers needed, otherwise this problem may not ever be fixed! Maybe the community could create a generic driver for this.

[Timo Aaltonen (tjaalton)]

Not a single comment from the original reporter, so I'm closing this as fixed in intrepid.

btw, why did this bugreport turn into a messy chat about building upstream master?? (no need to answer)

[David Nemeskey (nemeskeyd)]

Timo: if you need no answer, why ask? :) So here is one: sorry, but what do you expect? In a time when 3D is one of THE big things on the desktop, a bug that makes it impossible to use it will certainly drive people to look for a solution... And it's not just the bling, we are talking about Google Earth, Blender, games, etc. too.

Such a bug is a pain even for a month, let alone one and a half years! Fixed in Intrepid? The last comment on the bug in X.org tracker is from 2008-06-29! Isn't Hardy supposed to be an LTS release?!

But you are right in that maybe the building stuff should have been put under Answers, and just linked from here. Can it be done now, or is it too late?

[Rubin (rubin)]

Just upgraded to Ibex RC, and the bug is _NOT_ solved. If anything it seems easier to trigger now. Please re-open.

[curtis119 (curtis119)]

I agree with the above two posters. Closing this bug because of "lack of comments from the original poster" is just wrong. I am using the Ibex beta LiveCD and STILL am having problems with this bug. Intel is the ONLY gpu manufacturer in the WORLD with a 100% open source driver and YET I still have soooooo many problems with intel gpu's under Ubuntu it makes me sick to my stomach. I am experiencing a black screen with Ibex with a "VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller". Why is this? Why can this bug not be fixed? As I already stated, Intel is the only vendor with a 100% open source driver yet I am STILL unable to get this chipset to work correctly under Ubuntu.

What info do you need to fix this bug? I am willing to do whatever it takes to ensure that Ubuntu has a correctly working GM965 device driver. Tell me what to do or what info you need me to give you in order to help with this. An Intel GPU should work OUT-OF-THE-BOX on EVERY SINGLE LINUX DISTRO IN EXISTENCE. It doesn't work with Ubuntu, no matter what I try. I'm a community member who is willing to work with Ubuntu to solve this problem. Help me to help you. PLEASE!!!!!!!!!!

[Motin (motin)]

I have been amazed by how this bug never seems to be fixed... Is it because no core devs have this chipset and use 3d applications/play games? Is it because the bug in fact is several bugs grouped into one bug report and thus maybe only some part bugs are fixed?

This bug's existance was the reason that I bought a laptop with Nvidia graphics instead, despite potential sleep/hibernation problems and shorter battery life. I waited 4-5 weeks first, hoping that the bug would have been fixed meanwhile, but then went and ordered... Now, 4 months later, I am glad that I chose Nvidia...

[Maciej Słojewski (mslonik)]

To All,

The only thing that keeps me still with Ubuntu distro is a promise that in next release: 8.10, this bug finally will be fixed. I'm just a user, not developer, interested in 3D graphics as a hobbyist. Perhaps the only chance to push that bug to be fixed is to assign it as a high priority bug to the following 8.10 milestone list:

https://bugs.launchpad.net/ubuntu/intrepid/+bugs?field.milestone%3Alist=1326

If somebody knows how to do it, I mean how to move a bug to milestone list, please do it. Regards.

[Mike Roberts (m.roberts)]

Being sick and tired of this issue, I dug around and found this openSUSE forum.
http://forums.opensuse.org/hardware/388928-using-opengl-google-earth-hardlocks-x-intel-x3100.html
I haven't been able to try it yet, but I will post back with results.

[Shane Blake (blakesha)]

This is also appearing on 8.10 release with a Lenovo T61 laptop. Compiz APPEARS to be running, the desktop affects work, but trying to run 3D applications causes lockup. Specifically the Second Life client.

[Jeremy Foshee (jeremyfoshee)]

This behavior has also been seen in Lenovo R61e laptops. This includes significantly reduced FPS in testing 550 FPS in Intrepid, 1000 FPS in Hardy. Benchmarking with glxgear.

00:00.0 Host bridge: Intel Corporation Mobile PM965/GM965/GL960 Memory Controller Hub (rev 0c)
00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)
00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)
00:1a.0 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #4 (rev 03)
00:1a.1 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #5 (rev 03)
00:1a.7 USB Controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #2 (rev 03)
00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio Controller (rev 03)
00:1c.0 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 1 (rev 03)
00:1c.1 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 2 (rev 03)
00:1c.2 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 3 (rev 03)
00:1d.0 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #1 (rev 03)
00:1d.1 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #2 (rev 03)
00:1d.2 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #3 (rev 03)
00:1d.7 USB Controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #1 (rev 03)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev f3)
00:1f.0 ISA bridge: Intel Corporation 82801HEM (ICH8M) LPC Interface Controller (rev 03)
00:1f.1 IDE interface: Intel Corporation 82801HBM/HEM (ICH8M/ICH8M-E) IDE Controller (rev 03)
00:1f.2 SATA controller: Intel Corporation 82801HBM/HEM (ICH8M/ICH8M-E) SATA AHCI Controller (rev 03)
00:1f.3 SMBus: Intel Corporation 82801H (ICH8 Family) SMBus Controller (rev 03)
03:00.0 Network controller: Intel Corporation PRO/Wireless 3945ABG [Golan] Network Connection (rev 02)
04:00.0 Ethernet controller: Broadcom Corporation NetLink BCM5787M Gigabit Ethernet PCI Express (rev 02)
15:00.0 CardBus bridge: Ricoh Co Ltd RL5c476 II (rev b6)

[Rebecca Palmer (rebecca-palmer)]

I can no longer trigger this bug with either Blender or Atlas in Intrepid, but given its nature this doesn't prove it's entirely fixed.

Those who report seeing it in Intrepid: is it still there in the release version and do you have a reproducible way to trigger it?

I have not read the upstream report as attempting to do so gives an "invalid security certificate" Firefox error.

[Mike Roberts (m.roberts)]

Rebecca: Try running blender with this file.
http://launchpadlibrarian.net/14266500/blender_freeze.blend

I get the "invalid security certificate" error too. I'm pretty sure its safe.

[yeus (thomas-meschede-deactivatedaccount)]

I have still problem using blender with the intel 965gm chipset on a vostro 1310 with kubuntu 8.10...

running several programs i get this bug:

blender: brw_vtbl.c:98: brw_new_batch: Assertion `!brw->no_batch_wrap' failed.
Aborted

(within blender it only occurs sometimes and using the same file that used to crash my desktop completly on kubuntu hardy)

blender: brw_vtbl.c:98: brw_new_batch: Assertion `!brw->no_batch_wrap' failed.
Aborted

I added the blend file causing the error as an attachment

[yeus (thomas-meschede-deactivatedaccount)]

ahh btw: the error is also reproducable by installing the game "gridwars" and trying to run it..

(get gridwars2 from here:
http://www.getdeb.net/release/3393

greetings, tom

[trumpeteersman (trumpeteersman)]

Running Kubuntu 8.08 with kernel 2.6.24-21-generic. Intel driver: 2:2.2.1-1ubuntu13.7. Rest up to date in repos.
Kde Version: 4:3.5.10-0ubuntu1~hardy2
Qt Version: 3:3.3.8-b-0ubuntu3
Thinkpad T61, Intel Core2 Duo, X3100

Crash reproducible with Scorched 3D and Armagetron Advanced.

Will test other games soon.

[Shane Blake (blakesha)]

I was able to get a pre-compiled set of Intel 2.5.0 drivers (with the updated dri modules) and this still occurs.

Some 3D apps will work, while others just cause the system lock-up.

[yeus (thomas-meschede-deactivatedaccount)]

too many people still have this bug.. so i change it to confirmed..s

[Shane Blake (blakesha)]

I attempted to run the Cobalt developement environment (http://www.opencroquet.org/index.php/Cobalt) and once I tried to open the 3D view I get the following assertion error appearing:

shader/slang/slang_emit.c:257: storage_to_dst_reg: Assertion `size <= 4' failed.

With GDB running this assertion appears outside the scope of the squeak code (causing a SEGFAULT). As the above assertion is within MESA I thought I would post it here to see if it assists.

[lukasfischer (lukasfischer)]

Same bug here on a Sony VAIO VGN-NR11Z with this Graphics Controller:

00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c) (prog-if 00 [VGA controller])
        Subsystem: Sony Corporation Unknown device 902d
        Flags: bus master, fast devsel, latency 0, IRQ 17
        Memory at fc000000 (64-bit, non-prefetchable) [size=1M]
        Memory at d0000000 (64-bit, prefetchable) [size=256M]
        I/O ports at 1800 [size=8]
        Capabilities: [90] Message Signalled Interrupts: Mask- 64bit- Queue=0/0 Enable-
        Capabilities: [d0] Power Management version 3

00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)
        Subsystem: Sony Corporation Unknown device 902d
        Flags: bus master, fast devsel, latency 0
        Memory at fc100000 (64-bit, non-prefetchable) [size=1M]
        Capabilities: [d0] Power Management version 3

xserver-xorg-video-intel 2:2.2.1-ubuntu13.7
libgl1-mesa-dri 7.0.3~rc2-1ubuntu3

lockup occurs while running ParaView to visualize CFD Data.

[Bryce Harrington (bryce)]

[This is an automatic notice.]

We'd like to forward your bug upstream, however upstream requires
that you first test it against their newer driver code.

To save you the effort of building the driver from source, we've built
packages for the driver and its new dependencies.

So you have a couple options:

 1. Download and test .debs for intrepid, from:
     https://edge.launchpad.net/~intel-gfx-testing/+archive

 -or-

 2. Download and test the Jaunty alpha-2 (or newer) Live CD,
     (which includes a beta of the new xserver 1.6 as well).
     See http://cdimage.ubuntu.com/releases/9.04/ for ISOs

Thanks ahead of time! You can simply reply to this email to report your
findings.

P.S., if you wish to forward your bug upstream yourself, please follow
these directions to do so:
  http://intellinuxgraphics.org/how_to_report_bug.html

[IanSidden (iansidden)]

I have the same problem. I totally disabled Compiz (as well as Cairo-dock), and this works for some games but others still crash the system. Very frustrating.

Dell Inspiron 1525
Intel Core 2 Duo 1.83 Ghz
4 GB RAM
Intel GM965

[David Nemeskey (nemeskeyd)]

Bryce: What about building a similar package for Hardy? It is an LTS release, after all.

[Bryce Harrington (bryce)]

No, we tried but the number of dependencies for -intel 2.5.1 compared with what's available in hardy is too great. But there's some newer -intel snapshots for Hardy in the Xorg Edgers you could poke around with, but that stuff's not been very heavily tested so buyer beware.

Testing on Hardy would be a bit irrelevant anyway; if the Hardy bug was fixed in Intrepid, then no reason to test against anything newer than that. If the bug existed in both Hardy *and* Intrepid, then just testing against Intrepid would be sufficient. But if for some reason you definitely want to test against Hardy, do feel free to poke through the edgers stuff:

 https://edge.launchpad.net/~xorg-edgers

[Shane Blake (blakesha)]

To the Bug list.

Though the below "fixes" the issue of crashing (and the other graphic
corruption issues) the fix is SLOOOOOW. As though it is not using hardware
acceleration at all.

Regards,

Shane

On Sat, Dec 20, 2008 at 1:30 PM, Bryce Harrington <<email address hidden>
> wrote:

>
> [This is an automatic notice.]
>
> We'd like to forward your bug upstream, however upstream requires
> that you first test it against their newer driver code.
>
> To save you the effort of building the driver from source, we've built
> packages for the driver and its new dependencies.
>
> So you have a couple options:
>
> 1. Download and test .debs for intrepid, from:
> https://edge.launchpad.net/~intel-gfx-testing/+archive<https://edge.launchpad.net/%7Eintel-gfx-testing/+archive>
>
> -or-
>
> 2. Download and test the Jaunty alpha-2 (or newer) Live CD,
> (which includes a beta of the new xserver 1.6 as well).
> See http://cdimage.ubuntu.com/releases/9.04/ for ISOs
>
>
> Thanks ahead of time! You can simply reply to this email to report your
> findings.
>
>
> P.S., if you wish to forward your bug upstream yourself, please follow
> these directions to do so:
> http://intellinuxgraphics.org/how_to_report_bug.html
>
>
> ** Changed in: xserver-xorg-video-intel (Ubuntu)
> Status: Confirmed => Incomplete
>
> --
> [i965] intel gm965 freezes with 3d applications
> https://bugs.launchpad.net/bugs/120834
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in X.Org X server: Fix Released
> Status in "mesa" source package in Ubuntu: Fix Released
> Status in "xserver-xorg-video-intel" source package in Ubuntu: Incomplete
> Status in mesa in Ubuntu Hardy: Confirmed
> Status in xserver-xorg-video-intel in Ubuntu Hardy: Invalid
> Status in "xserver-xorg-video-intel" source package in zUbuntu: Invalid
>
> Bug description:
> On some Intel GM965 machines, running OpenGL applications (eg. Blender,
> Second Life, the Endgame screensaver, Scorched3D) can crash the system.
>
> Symptoms: Colours go funny, then screen flashes between off and black with
> 'please wait' cursor about 10 times, then stops; does not respond to
> Ctrl+Alt+F1 or Ctrl+Alt+Backspace, but does reboot on Alt+SysRq+b. Logs the
> line
> Error in I830WaitLpRing(), timeout for 2 seconds
> in /var/log/Xorg.0.log.old
>
> (Please note that this is not the only X crash bug; if you cannot find
> an exact match for your symptoms then please file a new bug.)
>
> Workaround: Adding
> Option "DRI" "false"
> to the graphics controller Device section of /etc/X11/xorg.conf avoids this
> bug, but may drastically reduce performance.
>
> This bug has been reported upstream and you can follow the link to xorg
> bugzilla, but as of 29.12.07 no developer managed to reproduce this bug (how
> to trigger it seems to be hardware dependent, with some i965s not appearing
> to have it at all) so it will be very hard to fix unless that changes
> if you are suffering from this bug please go to xorg bugzilla and report
> your hardware specifics.
> also please edit this bug description and add the name and exact model of
> your affected laptop.
>
> list of affect...

[Bryce Harrington (bryce)]

Shane, you didn't specify whether you were referring to option 1 or 2?

[Shane Blake (blakesha)]

Sorry. Option 1. I havent yet tried option 2.

- Shane

On Tue, Dec 23, 2008 at 6:36 PM, Bryce Harrington <<email address hidden>
> wrote:

> Shane, you didn't specify whether you were referring to option 1 or 2?
>
> --
> [i965] intel gm965 freezes with 3d applications
> https://bugs.launchpad.net/bugs/120834
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in X.Org X server: Fix Released
> Status in "mesa" source package in Ubuntu: Fix Released
> Status in "xserver-xorg-video-intel" source package in Ubuntu: Incomplete
> Status in mesa in Ubuntu Hardy: Confirmed
> Status in xserver-xorg-video-intel in Ubuntu Hardy: Invalid
> Status in "xserver-xorg-video-intel" source package in zUbuntu: Invalid
>
> Bug description:
> On some Intel GM965 machines, running OpenGL applications (eg. Blender,
> Second Life, the Endgame screensaver, Scorched3D) can crash the system.
>
> Symptoms: Colours go funny, then screen flashes between off and black with
> 'please wait' cursor about 10 times, then stops; does not respond to
> Ctrl+Alt+F1 or Ctrl+Alt+Backspace, but does reboot on Alt+SysRq+b. Logs the
> line
> Error in I830WaitLpRing(), timeout for 2 seconds
> in /var/log/Xorg.0.log.old
>
> (Please note that this is not the only X crash bug; if you cannot find
> an exact match for your symptoms then please file a new bug.)
>
> Workaround: Adding
> Option "DRI" "false"
> to the graphics controller Device section of /etc/X11/xorg.conf avoids this
> bug, but may drastically reduce performance.
>
> This bug has been reported upstream and you can follow the link to xorg
> bugzilla, but as of 29.12.07 no developer managed to reproduce this bug (how
> to trigger it seems to be hardware dependent, with some i965s not appearing
> to have it at all) so it will be very hard to fix unless that changes
> if you are suffering from this bug please go to xorg bugzilla and report
> your hardware specifics.
> also please edit this bug description and add the name and exact model of
> your affected laptop.
>
> list of affected laptops:
> Acer Travelmate 6292
> Asus S62E Santa Rosa 965
> dell D830
> Dell 630
> Dell Inspiron 1420n
> Dell Inspiron 1525 (new)
> Dell Latitude D630
> Dell Latitude D830
> Dell Vostro 1310
> Dell XPS M1330
> HP Compac 6510b
> HP Compaq 6710b
> Thinkpad T61
> Thinkpad X61 Tablet
> Thinkpad X61s
> Apple Macbook (3,1/Santa Rosa)
> Acer Aspire 5315-052G12Mi
> Lenovo 3000 N200 (TY2BBGE)
> Sony vaio CR160
>
> also if you have one of these laptops and are not affected please tell
> us!
>
>

[Bryce Harrington (bryce)]

Okay; I'd suggest trying a LiveCD before making conclusions about that -
many of the (esp. performance) changes in this version of the driver
require new features in the jaunty xserver/mesa to get the benefit.

Bryce

On Tue, Dec 23, 2008 at 08:00:28AM -0000, Shane Blake wrote:
> Sorry. Option 1. I havent yet tried option 2.
>
> - Shane
>
> On Tue, Dec 23, 2008 at 6:36 PM, Bryce Harrington <<email address hidden>
> > wrote:
>
> > Shane, you didn't specify whether you were referring to option 1 or 2?
> >
> > --
> > [i965] intel gm965 freezes with 3d applications
> > https://bugs.launchpad.net/bugs/120834
> > You received this bug notification because you are a direct subscriber
> > of the bug.
> >
> > Status in X.Org X server: Fix Released
> > Status in "mesa" source package in Ubuntu: Fix Released
> > Status in "xserver-xorg-video-intel" source package in Ubuntu: Incomplete
> > Status in mesa in Ubuntu Hardy: Confirmed
> > Status in xserver-xorg-video-intel in Ubuntu Hardy: Invalid
> > Status in "xserver-xorg-video-intel" source package in zUbuntu: Invalid
> >
> > Bug description:
> > On some Intel GM965 machines, running OpenGL applications (eg. Blender,
> > Second Life, the Endgame screensaver, Scorched3D) can crash the system.
> >
> > Symptoms: Colours go funny, then screen flashes between off and black with
> > 'please wait' cursor about 10 times, then stops; does not respond to
> > Ctrl+Alt+F1 or Ctrl+Alt+Backspace, but does reboot on Alt+SysRq+b. Logs the
> > line
> > Error in I830WaitLpRing(), timeout for 2 seconds
> > in /var/log/Xorg.0.log.old
> >
> > (Please note that this is not the only X crash bug; if you cannot find
> > an exact match for your symptoms then please file a new bug.)
> >
> > Workaround: Adding
> > Option "DRI" "false"
> > to the graphics controller Device section of /etc/X11/xorg.conf avoids this
> > bug, but may drastically reduce performance.
> >
> > This bug has been reported upstream and you can follow the link to xorg
> > bugzilla, but as of 29.12.07 no developer managed to reproduce this bug (how
> > to trigger it seems to be hardware dependent, with some i965s not appearing
> > to have it at all) so it will be very hard to fix unless that changes
> > if you are suffering from this bug please go to xorg bugzilla and report
> > your hardware specifics.
> > also please edit this bug description and add the name and exact model of
> > your affected laptop.
> >
> > list of affected laptops:
> > Acer Travelmate 6292
> > Asus S62E Santa Rosa 965
> > dell D830
> > Dell 630
> > Dell Inspiron 1420n
> > Dell Inspiron 1525 (new)
> > Dell Latitude D630
> > Dell Latitude D830
> > Dell Vostro 1310
> > Dell XPS M1330
> > HP Compac 6510b
> > HP Compaq 6710b
> > Thinkpad T61
> > Thinkpad X61 Tablet
> > Thinkpad X61s
> > Apple Macbook (3,1/Santa Rosa)
> > Acer Aspire 5315-052G12Mi
> > Lenovo 3000 N200 (TY2BBGE)
> > Sony vaio CR160
> >
> > also if you have one of these laptops and are not affected please tell
> > us!
> >
> >
>
> --
> [i965] intel gm965 freezes with 3d applications
> https://bugs.launchpad.net/bugs/120834
> You received this bug notification because you are a member of Ubuntu-X,
> which is subsc...

[Sander Smid (s-smid)]

@Bryce Harrington:

I tried both solution 1 and two on my notebook (Acer Aspire 5315) and in both cases the performance was horrible. The updated drivers do fix a number of issues, ie. opengl windows have borders and moving them leaves no old frame on the original location behind. I tried the game Warsow (www.warsow.net) and get 1 (!!) fps, glxgears gets 192 with solution 1 and 146 on the livecd.

I hope this helps.. if you happen to have newer drivers or other possible solutions please let me know. I'm not using this notebook for production anymore and I'm able to test stuff on it.

[dozer (matthew-pocock)]

I think I may be seeing this (or something very similar). Tux racer crashes out when you go from the flat graphics to the 3d stuff. The glx gears app appears to work fine. Some haskell code I've got that binds against the opengl libs renders garbage into the opengl frame. Other opengl apps crash kde, or bomb out, or are otherwise badly behaved.

I've got an HP Compaq 2710p with an Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c) and have tried both leaving xorg.conf as per default, and also adding a device for the graphics card with the intel driver (+ listing glx, GLcore and v4l modules).

Things seemed to be better behaved under hardy. Stuff seemed to break when I upgraded to Intrepid.

[Bryce Harrington (bryce)]

Hmm, the upstream bug is marked fixed. I wonder if it would be best to start a new bug on these issues in case they're unrelated. Setting to Confirmed for now, until additional analysis is done.

[slayer (antoniochiaravalloti)]

HP Compaq 6720s is also affected

[STaRMaN (jarizaro)]

Phoenix Voyagar S2 is affected too.
GMA x3100 / chipset GM965

[lukasfischer (lukasfischer)]

I am not sure if it's the same bug, my X server crashes and I have to reboot using ALT-PRINT-b, nothing else works. Screen blanks for half a second, comes back heavily distorted in the lower region and mouse is locked.
Xorg.0.log.old shows :

(II) intel(0): xf86BindGARTMemory: bind key 1 at 0x02529000 (pgoffset 9513)
(II) intel(0): xf86BindGARTMemory: bind key 2 at 0x093b3000 (pgoffset 37811)
(II) intel(0): xf86BindGARTMemory: bind key 3 at 0x0b88c000 (pgoffset 47244)
(II) intel(0): xf86BindGARTMemory: bind key 4 at 0x0dd65000 (pgoffset 56677)
(II) intel(0): Fixed memory allocation layout:
(II) intel(0): 0x00000000-0x0001ffff: ring buffer (128 kB)
(II) intel(0): 0x00020000-0x00029fff: HW cursors (40 kB)
(II) intel(0): 0x0002a000-0x00031fff: logical 3D context (32 kB)
(II) intel(0): 0x00032000-0x00041fff: exa G965 state buffer (64 kB)
(II) intel(0): 0x00042000-0x00042fff: overlay registers (4 kB)
(II) intel(0): 0x00050000-0x02528fff: front buffer (37732 kB)
(II) intel(0): 0x0077f000: end of stolen memory
(II) intel(0): 0x02529000-0x093b2fff: exa offscreen (113192 kB)
(II) intel(0): 0x093b3000-0x0b88bfff: back buffer (37732 kB)
(II) intel(0): 0x0b88c000-0x0dd64fff: depth buffer (37732 kB)
(II) intel(0): 0x0dd65000-0x0fd64fff: classic textures (32768 kB)
(II) intel(0): 0x10000000: end of aperture
(WW) intel(0): ESR is 0x00000001
(WW) intel(0): PRB0_CTL (0x0001f001) indicates ring buffer enabled
(WW) intel(0): PRB0_HEAD (0x014086f0) and PRB0_TAIL (0x000088a0) indicate ring buffer not flushed
(WW) intel(0): Existing errors found in hardware state.

my system:
Sony Vaio VGN-NR11Z
with GM965/GL960

attached lspci -vvnn and xorg.conf

[lukasfischer (lukasfischer)]

I am not sure if it's the same bug, my X server crashes and I have to reboot using ALT-PRINT-b, nothing else works. Screen blanks for half a second, comes back heavily distorted in the lower region and mouse is locked.
Xorg.0.log.old shows :

(II) intel(0): xf86BindGARTMemory: bind key 1 at 0x02529000 (pgoffset 9513)
(II) intel(0): xf86BindGARTMemory: bind key 2 at 0x093b3000 (pgoffset 37811)
(II) intel(0): xf86BindGARTMemory: bind key 3 at 0x0b88c000 (pgoffset 47244)
(II) intel(0): xf86BindGARTMemory: bind key 4 at 0x0dd65000 (pgoffset 56677)
(II) intel(0): Fixed memory allocation layout:
(II) intel(0): 0x00000000-0x0001ffff: ring buffer (128 kB)
(II) intel(0): 0x00020000-0x00029fff: HW cursors (40 kB)
(II) intel(0): 0x0002a000-0x00031fff: logical 3D context (32 kB)
(II) intel(0): 0x00032000-0x00041fff: exa G965 state buffer (64 kB)
(II) intel(0): 0x00042000-0x00042fff: overlay registers (4 kB)
(II) intel(0): 0x00050000-0x02528fff: front buffer (37732 kB)
(II) intel(0): 0x0077f000: end of stolen memory
(II) intel(0): 0x02529000-0x093b2fff: exa offscreen (113192 kB)
(II) intel(0): 0x093b3000-0x0b88bfff: back buffer (37732 kB)
(II) intel(0): 0x0b88c000-0x0dd64fff: depth buffer (37732 kB)
(II) intel(0): 0x0dd65000-0x0fd64fff: classic textures (32768 kB)
(II) intel(0): 0x10000000: end of aperture
(WW) intel(0): ESR is 0x00000001
(WW) intel(0): PRB0_CTL (0x0001f001) indicates ring buffer enabled
(WW) intel(0): PRB0_HEAD (0x014086f0) and PRB0_TAIL (0x000088a0) indicate ring buffer not flushed
(WW) intel(0): Existing errors found in hardware state.

my system:
Sony Vaio VGN-NR11Z
with GM965/GL960
Ubuntu 8.04.1
xserver-xorg 7.3+10ubuntu10.2
xserver-xorg-video-intel 2.2.1-1ubuntu13.7

attached lspci -vvnn and xorg.conf

[John Simmons (strannik-strannik)]

I am having the same problem on a Durabook D15RS running Ubuntu Studio 8.10 (Intrepid), so Intrepid does not necessarily fix the problem.

00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 03)

Thanks,

John

[Bryce Harrington (bryce)]

 Could you test it again with current Jaunty with all updates? We've recently upgraded to the 2.6.1 -intel driver, along with new libdrm, mesa, and xserver, any of which may have addressed this issue.

[John Simmons (strannik-strannik)]

Is Jaunty stable enough for mission critical use? If not I will wait.

One other detail - an Xorg patch came via Ubuntu update last week, and
now I can run Compiz in its full glory, but still not Second Life or
Google Earth.

Bryce Harrington wrote:
> Could you test it again with current Jaunty with all updates? We've
> recently upgraded to the 2.6.1 -intel driver, along with new libdrm,
> mesa, and xserver, any of which may have addressed this issue.
>
> ** Changed in: xserver-xorg-video-intel (Ubuntu)
> Status: Confirmed => Incomplete
>
>

[Shane Blake (blakesha)]

I dist-updated to Jaunty, and after switching to the UXA AccelMethod (Tiling
is also set to "No") in the xorg.conf I ran some simple test:

1) Run glxgears - All Good
2) Run the Hippo version of the Second Life client - All Good
3) Run Super Tux Racer - All Good.

All three ran [reasonably] well with no crashing of X and no shearing of the
view window while it was being dragged around the desktop.

Checking the Xorg log and it appears to be using DRI2 correctly.

Regards,

Shane

On Wed, Jan 28, 2009 at 4:57 PM, Bryce Harrington <<email address hidden>
> wrote:

> Could you test it again with current Jaunty with all updates? We've
> recently upgraded to the 2.6.1 -intel driver, along with new libdrm,
> mesa, and xserver, any of which may have addressed this issue.
>
> ** Changed in: xserver-xorg-video-intel (Ubuntu)
> Status: Confirmed => Incomplete
>
> --
> [i965] intel gm965 freezes with 3d applications
> https://bugs.launchpad.net/bugs/120834
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in X.Org X server: Fix Released
> Status in "mesa" source package in Ubuntu: Fix Released
> Status in "xserver-xorg-video-intel" source package in Ubuntu: Incomplete
> Status in mesa in Ubuntu Hardy: Confirmed
> Status in xserver-xorg-video-intel in Ubuntu Hardy: Invalid
> Status in "xserver-xorg-video-intel" source package in zUbuntu: Invalid
>
> Bug description:
> On some Intel GM965 machines, running OpenGL applications (eg. Blender,
> Second Life, the Endgame screensaver, Scorched3D) can crash the system.
>
> Symptoms: Colours go funny, then screen flashes between off and black with
> 'please wait' cursor about 10 times, then stops; does not respond to
> Ctrl+Alt+F1 or Ctrl+Alt+Backspace, but does reboot on Alt+SysRq+b. Logs the
> line
> Error in I830WaitLpRing(), timeout for 2 seconds
> in /var/log/Xorg.0.log.old
>
> (Please note that this is not the only X crash bug; if you cannot find
> an exact match for your symptoms then please file a new bug.)
>
> Workaround: Adding
> Option "DRI" "false"
> to the graphics controller Device section of /etc/X11/xorg.conf avoids this
> bug, but may drastically reduce performance.
>
> This bug has been reported upstream and you can follow the link to xorg
> bugzilla, but as of 29.12.07 no developer managed to reproduce this bug (how
> to trigger it seems to be hardware dependent, with some i965s not appearing
> to have it at all) so it will be very hard to fix unless that changes
> if you are suffering from this bug please go to xorg bugzilla and report
> your hardware specifics.
> also please edit this bug description and add the name and exact model of
> your affected laptop.
>
> list of affected laptops:
> Acer Travelmate 6292
> Asus S62E Santa Rosa 965
> dell D830
> Dell 630
> Dell Inspiron 1420
> Dell Inspiron 1420n
> Dell Inspiron 1525 (new)
> Dell Latitude D630
> Dell Latitude D830
> Dell Vostro 1310
> Dell XPS M1330
> HP Compac 6510b
> HP Compaq 6710b
> Thinkpad T61
> Thinkpad X61 Tablet
> Thinkpad X61s
> Apple Macbook (3,1/Santa Rosa)
> Acer Aspire 5315-052G12Mi
> Acer Extensa 5220-050508Mi
> Lenovo 3000 N200 (TY2BBGE)
> Sony vaio CR16...

[trumpeteersman (trumpeteersman)]

I have been having the same issue.

X3100, T61 Thinkpad
8.04, KDE 3.5
2.6.24-21-generic
xserver-xorg-video-intel 2.2.1-1ubuntu13.8

00:02.1 Display controller [0380]: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller [8086:2a03] (rev 0c)

Any 3D app crashes X. Have to restart.

Relavent Syslog:

Feb 10 12:50:32 Laudis -- MARK --
Feb 10 12:51:47 Laudis kernel: [ 833.685168] [drm:i915_wait_irq] *ERROR* i915_wait_irq: EBUSY -- rec: 2920702 emitted: 2920868
Feb 10 12:51:49 Laudis gdm[5470]: WARNING: gdm_slave_xioerror_handler: Fatal X error - Restarting :0
Feb 10 12:52:14 Laudis kernel: [ 859.150172] SysRq : HELP : loglevel0-8 reBoot Crashdump tErm Full kIll saK showMem Nice powerOff showPc show-all-timers(Q) unRaw Sync showTasks Unmount shoW-blocked-tasks

I included my Xorg.0.log.old

I cannot dist-upgrade. If there is a fix in Intrepid and Jaunty, then the updates should be added to Hardy repos. (LTS)
 A few weeks ago I tried compiling the latest from Intel but X would not start. Can I get the newest Intel Drivers from Jaunty and install them in hardy? Can I add the Intrepid and Jaunty repos just to get the xserver stuff and then remove the extra repos?

[PCC (p-c-chan)]

I have similar issues on my ThinkPad T61 with GM965 running ubuntu with 2.6.28.4 kernel.

The screen would become corrupted with the 2.4x or 2.6x intel server and it crashes frequently. For example thinktanksdemo would show a broken screen (portion of screen inverted or shown at the wrong place), GL screensavers would have a flashing row at the bottom and use tones of CPU.

However 2.2x server seems surviving well with none of the above issues but it would fill up Xorg.0.log over time by showing the available settings and configurations over and over again.

I am not sure whether glxgears have changed over the releases, but there is a huge difference in frame rate on this machine across releases.

I have over 1000 FPS by using the 2.2x server and libraries, about 600 FPS for 2.4x, and less than 60 FPS (not 600) for the 2.6x. It is definitely going the wrong way!

Can anybody check what have changed for GM965 from 2.2 to 2.6 and figure out what went wrong?

As for Xorg.conf, I am using virtually no configurations and let the server discover the best configuration on start up under all streams.

For now, I am running intrepid stream of Ubuntu with xserver and libraries downgraded to hardy (2.2x). Before the new stuff is back ported, I am still OK.

I have tried the most updated xserver and libraries available from time to time, unfortunately none worked out and I had to fall back to 2.2 to get a functional screen.

Hopefully somebody would figure it out soon.

Thanks,

[Shane Blake (blakesha)]

@PCC:

With the Jaunty Alpha update I found that most, if not all of those issues
were solved, but only when I explicitly told it to use UXA. If I left this
setting out of the Xorg.conf then those framerate issues were definitely
there as it was using the older DRI (not DRI2) for acceleration.

I have that exact laptop.

- Shane

On Thu, Feb 12, 2009 at 6:38 AM, PCC <email address hidden> wrote:

> I have similar issues on my ThinkPad T61 with GM965 running ubuntu with
> 2.6.28.4 kernel.
>
> The screen would become corrupted with the 2.4x or 2.6x intel server and
> it crashes frequently. For example thinktanksdemo would show a broken
> screen (portion of screen inverted or shown at the wrong place), GL
> screensavers would have a flashing row at the bottom and use tones of
> CPU.
>
> However 2.2x server seems surviving well with none of the above issues
> but it would fill up Xorg.0.log over time by showing the available
> settings and configurations over and over again.
>
> I am not sure whether glxgears have changed over the releases, but there
> is a huge difference in frame rate on this machine across releases.
>
> I have over 1000 FPS by using the 2.2x server and libraries, about 600
> FPS for 2.4x, and less than 60 FPS (not 600) for the 2.6x. It is
> definitely going the wrong way!
>
> Can anybody check what have changed for GM965 from 2.2 to 2.6 and figure
> out what went wrong?
>
> As for Xorg.conf, I am using virtually no configurations and let the
> server discover the best configuration on start up under all streams.
>
> For now, I am running intrepid stream of Ubuntu with xserver and
> libraries downgraded to hardy (2.2x). Before the new stuff is back
> ported, I am still OK.
>
> I have tried the most updated xserver and libraries available from time
> to time, unfortunately none worked out and I had to fall back to 2.2 to
> get a functional screen.
>
> Hopefully somebody would figure it out soon.
>
> Thanks,
>
> --
> [i965] intel gm965 freezes with 3d applications
> https://bugs.launchpad.net/bugs/120834
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in X.Org X server: Fix Released
> Status in "mesa" source package in Ubuntu: Fix Released
> Status in "xserver-xorg-video-intel" source package in Ubuntu: Incomplete
> Status in mesa in Ubuntu Hardy: Confirmed
> Status in xserver-xorg-video-intel in Ubuntu Hardy: Invalid
> Status in "xserver-xorg-video-intel" source package in zUbuntu: Invalid
>
> Bug description:
> On some Intel GM965 machines, running OpenGL applications (eg. Blender,
> Second Life, the Endgame screensaver, Scorched3D) can crash the system.
>
> Symptoms: Colours go funny, then screen flashes between off and black with
> 'please wait' cursor about 10 times, then stops; does not respond to
> Ctrl+Alt+F1 or Ctrl+Alt+Backspace, but does reboot on Alt+SysRq+b. Logs the
> line
> Error in I830WaitLpRing(), timeout for 2 seconds
> in /var/log/Xorg.0.log.old
>
> (Please note that this is not the only X crash bug; if you cannot find
> an exact match for your symptoms then please file a new bug.)
>
> Workaround: Adding
> Option "DRI" "false"
> to the graphics con...

[PCC (p-c-chan)]

Glad to hear that a new update really works.

After another failure attempt to hang onto a newer version of xserver and libraries, I had downgraded my T61 (the 3rd or 4th time), almost down to hardy and even out of KDE4 this time.

The 2.2 intel server has been working fine (except the repeated configuration change messages in kern.log). I even got my favourite fireworks 3D (GL) screensaver back working beautifully. I am very happy.

I'll probably wait for a couple of new updates and I want to understand what could have gone wrong before trying again.

So what should I add/omit on the xorg.conf? Or could you post your xorg.conf? Since we have the same laptop I may as well use yours.

And which dri2 library should I use, is it something outside the libgl1-mesa-dri package?

Thanks,
PC

[Bryce Harrington (bryce)]

Sounds like this bug will be solved once we move to UXA. Thanks for testing Shane.

[John Simmons (strannik-strannik)]

Can you also please check proper rendering of Second Life and Google
Earth with this fix?

Bryce Harrington wrote:
> Sounds like this bug will be solved once we move to UXA. Thanks for
> testing Shane.
>
> ** Summary changed:
>
> - [i965] intel gm965 freezes with 3d applications
> + [i965] [Needs UXA] intel gm965 freezes with 3d applications
>
> ** Changed in: mesa (Ubuntu Hardy)
> Status: Confirmed => Triaged
>
>

[Shane Blake (blakesha)]

Good to go with both... a little slow, but that is also with compiz going
and it isnt THAT great an graphics adapter :)

On Thu, Mar 5, 2009 at 12:47 PM, John Simmons <email address hidden> wrote:

> Can you also please check proper rendering of Second Life and Google
> Earth with this fix?
>
> Bryce Harrington wrote:
> > Sounds like this bug will be solved once we move to UXA. Thanks for
> > testing Shane.
> >
> > ** Summary changed:
> >
> > - [i965] intel gm965 freezes with 3d applications
> > + [i965] [Needs UXA] intel gm965 freezes with 3d applications
> >
> > ** Changed in: mesa (Ubuntu Hardy)
> > Status: Confirmed => Triaged
> >
> >
>
> --
> [i965] [Needs UXA] intel gm965 freezes with 3d applications
> https://bugs.launchpad.net/bugs/120834
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in X.Org X server: Fix Released
> Status in “mesa” source package in Ubuntu: Fix Released
> Status in “xserver-xorg-video-intel” source package in Ubuntu: Incomplete
> Status in mesa in Ubuntu Hardy: Triaged
> Status in xserver-xorg-video-intel in Ubuntu Hardy: Invalid
> Status in “xserver-xorg-video-intel” source package in zUbuntu: Invalid
>
> Bug description:
> On some Intel GM965 machines, running OpenGL applications (eg. Blender,
> Second Life, the Endgame screensaver, Scorched3D) can crash the system.
>
> Symptoms: Colours go funny, then screen flashes between off and black with
> 'please wait' cursor about 10 times, then stops; does not respond to
> Ctrl+Alt+F1 or Ctrl+Alt+Backspace, but does reboot on Alt+SysRq+b. Logs the
> line
> Error in I830WaitLpRing(), timeout for 2 seconds
> in /var/log/Xorg.0.log.old
>
> (Please note that this is not the only X crash bug; if you cannot find
> an exact match for your symptoms then please file a new bug.)
>
> Workaround: Adding
> Option "DRI" "false"
> to the graphics controller Device section of /etc/X11/xorg.conf avoids this
> bug, but may drastically reduce performance.
>
> This bug has been reported upstream and you can follow the link to xorg
> bugzilla, but as of 29.12.07 no developer managed to reproduce this bug (how
> to trigger it seems to be hardware dependent, with some i965s not appearing
> to have it at all) so it will be very hard to fix unless that changes
> if you are suffering from this bug please go to xorg bugzilla and report
> your hardware specifics.
> also please edit this bug description and add the name and exact model of
> your affected laptop.
>
> list of affected laptops:
> Acer Travelmate 6292
> Asus S62E Santa Rosa 965
> dell D830
> Dell 630
> Dell Inspiron 1420
> Dell Inspiron 1420n
> Dell Inspiron 1525 (new)
> Dell Latitude D630
> Dell Latitude D830
> Dell Vostro 1310
> Dell XPS M1330
> HP Compac 6510b
> HP Compaq 6710b
> Thinkpad T61
> Thinkpad X61 Tablet
> Thinkpad X61s
> Apple Macbook (3,1/Santa Rosa)
> Acer Aspire 5315-052G12Mi
> Acer Extensa 5220-050508Mi
> Lenovo 3000 N200 (TY2BBGE)
> Sony vaio CR160
>
> also if you have one of these laptops and are not affected please tell
> us!
>
>

[Bryce Harrington (bryce)]

We have a PPA with some new tools for debugging X freezes:

  https://launchpad.net/~ubuntu-x-swat/+archive/x-freeze-test

I know how irritating X freezes can be. They're also typically
quite hard to debug, but the information provided by these new
tools should help upstream figure them out.

You can help by doing the following:

 A. Install the PPA packages on Jaunty and boot kernel 2.6.30-rc2
 B. Reproduce your freeze
 C. ssh into the machine and run the steps to collect the info
 D. Attach the tarball of the results to this bug report

With this information, we'll be able to forward your bug upstream.

(For more information on triaging X freeze bugs, see
 https://wiki.ubuntu.com/X/Troubleshooting/Freeze )

[PCC (p-c-chan)]

Finally upgraded successfully to 2.7 (karmic) from 2.2 yesterday, with UXA enabled, thinktanksdemo works and glxinfo is running at over 3000 frames in 5 seconds (not bad). And it even mentions about UXA under man intel.

Thanks.

[Launchpad Janitor (janitor)]

This bug was fixed in the package xserver-xorg-video-intel - 2:2.7.99.1+git20090602.ec2fde7c-0ubuntu1

---------------
xserver-xorg-video-intel (2:2.7.99.1+git20090602.ec2fde7c-0ubuntu1) karmic; urgency=low

  * Update to git 20090602 (master branch) up to commit ec2fde7c
    - xvmc is disabled since DRI1 no longer supported
    - LP: #96991 - 3D stuff breaks with Compiz: Redirected Direct Rendering
      is needed in DRI
    - LP: #120834 - X freezes with I830WaitLpRing error when running OpenGL apps
    - LP: #337608 - X crashes in fbBlt() when using Sun Java Plugin 6 + firefox3.0
    - LP: #339555 - compiz slowmotion after Jaunty upgrade
    - LP: #363900 - X.org freezes with intel driver, no apparent trigger
    - LP: #331719 - VT switching doesn't work on Intel 915GM
    - LP: #339091 - X freezes a few minutes after resuming
    - LP: #348436 - Kubuntu: X server crash when screensaver is started (4500MHD)
    - LP: #279727 - Kubuntu: Display Corruption w/ Intel 4700MHD
    - LP: #357851 - Kubuntu: Distorted display after switching virtual desktops w/ exa
    - LP: #158415 - Front buffer dynamic resize not supported
    - LP: #324998 - x server restarts itself w/ compiz on Intel 945GM
    - LP: #355593 - after upgrade to 9.04, rotating desktop cube ran slow
    - LP: #357290 - 1 fps in 3d apps like neverball with EXA
    - LP: #360774 - Graphical Corruption with EXA on X4500
    - LP: #364126 - screensaver prefs dialog in 9.04 RC livecd leaves dirt
    - LP: #375712 - Native resolution for dell "2005fpw" monitor not listed
    - LP: #375264 - Choppy flash video and poor performance with compiz
    - LP: #349568 - Jaunty / Compiz slow and tearing on GMA 4500MHD
    - LP: #356056 - window tearing during movement on 965 (no compiz)
    - LP: #330460 - xorg shows black image/hangs with jpg in firefox
    - LP: #347587 - X asserts on pI830->batch_ptr != 0 on resume from suspend
  * Merge with Debian experimental. Remaining Ubuntu changes:
    - Add lpia architecture
    - Re-enable the patch system, add quilt to build-deps.
    - 110_quirk_hp_mini.patch: quirk (sent upstream)
    - 117_quirk_thinkpad_x30.patch: quirk (sent upstream)
  * Drop 116_8xx_disable_dri.patch. There have been fixes for 3d on 8xx
    chipsets upstream, so drop the DRI disablement so the fixes can be
    re-tested.
  * Drop 103_quirk_intel_mb890.patch. Better quirk available upstream.
    (LP: #305269)

 -- Bryce Harrington <email address hidden> Tue, 02 Jun 2009 10:47:32 -0700