http://bugs.winehq.org/show_bug.cgi?id=23048
Summary: UT3 Fails due to glPointSize invalid values. Product: Wine Version: 1.2-rc2 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d AssignedTo: wine-bugs@winehq.org ReportedBy: kphillisjr@gmail.com
Created an attachment (id=28595) --> (http://bugs.winehq.org/attachment.cgi?id=28595) Output log from wine.
Latest Patch from unreal tournament 3 fails and locks up before allowing me to enter the key.
Attached is the log for wine version 1.2-rc2
Test System: Ubuntu 10.04 (64-bit) Radeon HD 5770 Athlon II x4 620 4gb ddr3 1600 memory.
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #1 from Rico kgbricola@web.de 2010-06-06 03:01:14 --- Created an attachment (id=28597) --> (http://bugs.winehq.org/attachment.cgi?id=28597) fglrx shader spam filter
This bug could probably be a dupe of bug 22922 .
Nevertheless could you please try the attached patch, it has nothing to do with your bug, but it should remove the shader spam, which is in your log, e.g.: fixme:d3d_shader:print_glsl_info_log Fragment shader(s) linked, vertex shader(s) linked.
Which fglrx version are you using? Is it 10.5?
http://bugs.winehq.org/show_bug.cgi?id=23048
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Summary|UT3 Fails due to |Unreal Tournament 3 fails |glPointSize invalid values. |and locks up
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #2 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-06 13:20:16 --- Yes I have version 10.5 of the ati driver.
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #3 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-06 13:47:20 --- Created an attachment (id=28608) --> (http://bugs.winehq.org/attachment.cgi?id=28608) Updated Spam Patch.
I tried the patch and it didn't get rid of the spam, however, i modified it a little bit and got rid of all the spam, and here's the updated patch.
http://bugs.winehq.org/show_bug.cgi?id=23048
Ken Phillis Jr kphillisjr@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #28595|0 |1 is obsolete| |
--- Comment #4 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-06 13:49:51 --- Created an attachment (id=28609) --> (http://bugs.winehq.org/attachment.cgi?id=28609) UT3 Log with latest patch applied.
As an interesting note, the wait channel gets stuck to 0 and the sound suddenly stops. I've also had a few times where UT3 would lock up and then I had to ssh in to kill the task just to get keyboard input back. (Keyboard becomes unresponsive)
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #5 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-11 17:07:00 --- Created an attachment (id=28746) --> (http://bugs.winehq.org/attachment.cgi?id=28746) Fixme Patch to the state_pscale function.
I added a few lines to the pscale function that is causing all of the errors and found the following as output.
fixme:d3d:state_pscale Point size Problem, Initial Call: nan/1.000000/0.000000/0.000000 err:d3d:state_pscale >>>>>>>>>>>>>>>>> GL_INVALID_VALUE (0x501) from glPointSize(...); @ state.c / 1511 fixme:d3d:state_pscale Point size Problem, Final Call: nan/1.000000/0.000000/0.000000
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #6 from Rico kgbricola@web.de 2010-06-12 02:53:44 --- Created an attachment (id=28761) --> (http://bugs.winehq.org/attachment.cgi?id=28761) hacky patch to check point size
Could you try the attached patch? It clamps the pointsize to min/max values. I'm not sure if the check is in the correct place, but the application somehow set the pointsize to a wrong value (nan or 0.0f) and we didn't check/clamp the input to an acceptable value.
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #7 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-12 20:06:59 --- The patch fails to fix the problem... it still gives me a ton of invalid glPointSize values. in the log.
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #8 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-12 20:58:38 --- I went ahead and checked to see what the hex value was that was getting stuck into the parameter, and here's what i found... 0x7fa02001 This signals a quiet Nan signal according to IEEE 754 Representation. so I think the best choice is to return on this one and not change the renderer values.
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #9 from Rico kgbricola@web.de 2010-06-13 06:23:16 --- The problem doesn't seem to happen on NVIDIA 8800GTS (195.36.24).
According to http://www.opengl.org/sdk/docs/man/xhtml/glPointSize.xml GL_INVALID_VALUE is generated if psize is <= 0. But according to http://www.gnu.org/s/libc/manual/html_node/Infinity-and-NaN.html NaN is not smaller nor equal nor bigger than 0 and the driver shouldn't spit out an err. So this looks like a driver bug, doesn't it?
We could check for a isnan(pointSize.f) and set the pointSize.f=1.0 or something to make the driver happy.
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #10 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-13 16:46:02 --- Created an attachment (id=28810) --> (http://bugs.winehq.org/attachment.cgi?id=28810) PointSize Check Value
I think the proper fix is to actually prevent the render state from getting messed up to begin with, so i applied a quick fix to this that seems to work... see the attached patch.
also, as a side note, i found out the app wasn't locking up, but instead was getting bogged down heavily. (this patch removes all the log spam )
http://bugs.winehq.org/show_bug.cgi?id=23048
Ken Phillis Jr kphillisjr@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #28609|0 |1 is obsolete| |
--- Comment #11 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-13 16:49:12 --- Created an attachment (id=28811) --> (http://bugs.winehq.org/attachment.cgi?id=28811) Unreal tournament 3 Log after patches
The latest log with the patches for fglrx spam and the testing of setpointsize in setrenderstate.
The "hanging" and extreme sluggishness still occurs when i hit the cd key entry though.
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #12 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-17 19:11:28 --- Tested against the latest git, and the 10.6 ati driver, and i still see the pointsize error hundreds of times, however pointsize patch i created that tests the input parameters before setting the new state fixes the issue with this entry, however the performance problems still exist and it hangs when i get to the cd-key entry page.
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #13 from Ken Phillis Jr kphillisjr@gmail.com 2010-06-27 15:21:42 --- Created an attachment (id=29200) --> (http://bugs.winehq.org/attachment.cgi?id=29200) Test results.
I managed to find where UT3 finally gets stuck into a infinite loop, and here's the output of running the program with winedebug set to +d3d,+d3d9
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #14 from Ken Phillis Jr kphillisjr@gmail.com 2010-08-01 17:26:08 --- Still not fixed in wine v1.2 or with the Catalyst 10.7 driver.
http://bugs.winehq.org/show_bug.cgi?id=23048
Night Nord NightNord@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |NightNord@gmail.com
--- Comment #15 from Night Nord NightNord@gmail.com 2011-03-01 16:01:52 CST --- *** Bug 26265 has been marked as a duplicate of this bug. ***
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #16 from Night Nord NightNord@gmail.com 2011-03-01 16:05:30 CST --- Affects Bulletstorm on wine-1.3.14 and fglrx-11.2.
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #17 from Ken Phillis Jr kphillisjr@gmail.com 2011-09-08 17:19:42 CDT --- I just tested with the Catalyst 11.8 with wine v1.3.27 on a Radeon HD6520... The bug still exists. Although, with this setup the screen is black and nothing gets rendered.
http://bugs.winehq.org/show_bug.cgi?id=23048
bpaterni@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |bpaterni@gmail.com
--- Comment #18 from bpaterni@gmail.com 2011-12-19 22:24:16 CST --- This bug is also present in the Halo: CE trial using radeon 4850 + open source 3d drivers with wine 1.3.35. (Perhaps this means wine itself is at fault here instead of drivers?...)
With unmodified (git) wine, the game seems to start up fine, but the screen flickers different colors once in the menu. I can hear the menu theme music playing while this is happening.
If I add a simple check that sets pointSize.f to 1.0f if pointSize.f <= 0.0f, then the invalid value error log spam no longer appears. However there is no change in the playability of the trial. The whole game screen still flickers solid colors,
http://bugs.winehq.org/show_bug.cgi?id=23048
lithegnolie@aol.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |lithegnolie@aol.com
--- Comment #19 from lithegnolie@aol.com 2012-04-29 05:45:46 CDT --- *** Bug 30544 has been marked as a duplicate of this bug. ***
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #20 from lithegnolie@aol.com 2012-04-29 05:51:08 CDT --- Just fyi, this is still happening in wine 1.5.2 (current git clone) on Ubuntu 10.04
with fglrx installed for ATI Mobility Radeon HD 5400 Series. fglrx-amdcccle_8.961-0ubuntu1_amd64.deb fglrx-dev_8.961-0ubuntu1_amd64.deb fglrx-modaliases_8.961-0ubuntu1_amd64.deb fglrx_8.961-0ubuntu1_amd64.deb
Cheers !
http://bugs.winehq.org/show_bug.cgi?id=23048
sacrediou sacrediou@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sacrediou@yahoo.fr
http://bugs.winehq.org/show_bug.cgi?id=23048
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jeremielapuree@yahoo.fr
--- Comment #21 from joaopa jeremielapuree@yahoo.fr 2013-03-23 22:52:50 CDT --- still a bug with current Wine?
If yes, does it happen with the demo http://www.jeuxvideopc.com/demo/8984-unreal-tournament-3.php
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #22 from Todor todormailbox-wine@yahoo.com 2013-07-31 15:35:17 CDT --- Created attachment 45466 --> http://bugs.winehq.org/attachment.cgi?id=45466 Screenshot of UT3 hanging at the screen asking for CD key
Screenshot of Unreal Tournament 3 hanging at the CD key screen.
http://bugs.winehq.org/show_bug.cgi?id=23048
Todor todormailbox-wine@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |todormailbox-wine@yahoo.com
--- Comment #23 from Todor todormailbox-wine@yahoo.com 2013-07-31 15:36:43 CDT --- The bug still does exists when running UT3 under Wine version 1.6 (Kubuntu 13.04 x64-bit, Wine from ubuntu-wine PPA) with proprietary AMD driver:
$fglrxinfo display: :0 screen: 0 OpenGL vendor string: Advanced Micro Devices, Inc. OpenGL renderer string: AMD Radeon HD 7800 Series OpenGL version string: 4.2.12217 Compatibility Profile Context 12.104
The issue is confirmed under both Crossover v.12.2.2 and Wine 1.6.
From Crossover logs:
err:d3d:state_pscale >>>>>>>>>>>>>>>>> GL_INVALID_VALUE (0x501) from glPointSize(...); @ state.c / 1605 err:d3d:state_pscale >>>>>>>>>>>>>>>>> GL_INVALID_VALUE (0x501) from glPointSize(...); @ state.c / 1605 err:d3d:state_pscale >>>>>>>>>>>>>>>>> GL_INVALID_VALUE (0x501) from glPointSize(...); @ state.c / 1605
From Wine logs:
err:d3d:state_pscale >>>>>>>>>>>>>>>>> GL_INVALID_VALUE (0x501) from glPointSize(...); @ state.c / 1579 err:d3d:state_pscale >>>>>>>>>>>>>>>>> GL_INVALID_VALUE (0x501) from glPointSize(...); @ state.c / 1579
http://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #24 from Todor todormailbox-wine@yahoo.com 2013-09-22 12:43:01 CDT --- @joaopa I installed the demo and patched (see here for details: http://forums.beyondunreal.com/showthread.php?t=185087), but still no joy - it crashes after the initial splash screen.
I also upgraded Crossvoer to the latest version 12.5.0, reinstalled UT3 (the full version) again from Steam and it still hangs at the CD key screen. Any ideas what else I can try?
https://bugs.winehq.org/show_bug.cgi?id=23048
hanska2@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |hanska2@luukku.com
--- Comment #25 from hanska2@luukku.com --- Is this issue still valid with 1.7 series of wine?
https://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #26 from Todor todormailbox-wine@yahoo.com --- I had to apply some of the registry changes mentioned in the Corssover's Tips and Tricks section (https://www.codeweavers.com/compatibility/browse/name/?app_id=4155;tips=1#ti...) and managed to install the game successfully under Wine 1.7 on Ubuntu 14.04 x86-64.
https://bugs.winehq.org/show_bug.cgi?id=23048
winetest@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winetest@luukku.com
--- Comment #27 from winetest@luukku.com --- Can someone retest with more recent wine?
https://bugs.winehq.org/show_bug.cgi?id=23048
David Chalmers aff@affsdiary.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |aff@affsdiary.com
--- Comment #28 from David Chalmers aff@affsdiary.com --- I tested with Wine Staging 2.21 and did not encounter this issue.
https://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #29 from Todor todormailbox-wine@yahoo.com --- Tested the UT3 installation and gameplay with Wine 3.0 stable under Ubuntu 16.04 LTS and it works without issues.
I had to install UT3 in 32-bit mode though - with WINEARCH=win32. I also had to use winetricks to install Adobe Flash, Shockwave and Internet Explorer 8 to avoid issues with the installer.
I am using AMD open-source GPU stack. The game was installed from the retailed DVD version (not Steam). I applied all patches, including TitanPack and Patch 5 (which is download only).
No other tricks or registry changes were applied - the game now plays out of the box.
Thanks to the Wine team for finally resolving all the issues with UT3.
https://bugs.winehq.org/show_bug.cgi?id=23048
--- Comment #30 from joaopa jeremielapuree@yahoo.fr --- Reported fixed. This bug can be closed as FIXED.
https://bugs.winehq.org/show_bug.cgi?id=23048
Matteo Bruni matteo.mystral@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution|--- |FIXED
--- Comment #31 from Matteo Bruni matteo.mystral@gmail.com --- It's not clear that this was actually fixed in Wine (as opposed to just by switching to a decent driver). Anyway, closing FIXED.
https://bugs.winehq.org/show_bug.cgi?id=23048
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #32 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 3.12.