http://bugs.winehq.org/show_bug.cgi?id=22029
Summary: Star Wars 2: freezes in scene ~5 minutes within game. Product: Wine Version: 1.1.27 Platform: x86 OS/Version: Mac OS X 10.5 Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d AssignedTo: wine-bugs@winehq.org ReportedBy: hoehle@users.sourceforge.net
commit 3a63552fd1880c973f28baf8f9f7a29fd24d1412 Author: Henri Verbeet hverbeet@codeweavers.com Date: Mon Jul 20 12:21:48 2009 +0200 wined3d: Skip some unnecessary FBO binds. sounds innocent, but causes a curious effect: the screen freezes always at the same position, in a scene one can reach roughly 5 minutes after starting the first level. It happens when the camera starts zooming onto the tunnel whose door Leia unlocked by moving 2 boxes.
Pressing Esc to enter the (OSD) menu does not unblock the screen.
However the app is not dead. Once one moves back (despite the frozen display), away from the tunnel, everything works perfectly once again. One can return to the rooms previously visited.
The app gets a garbage rating with that regression, because the frozen screen prevents progression to the next level. OTOH, wine-1.1.24 deserves silver-gold. The bug is still present in wine-1.1.40.
Using MacOS X 10.5.8 with NVidia 9400 graphics. IIRC, but perhaps I used an older version, Ubuntu with Intel i915 graphics is not affected (it has no FBO AFAIK?), but the Intel is too slow generally for this app.
I have attached no log because the console shows nothing distinguishing. Unable to load NVCPL.dll [not seen on Intel] fixme:d3d9:Direct3DShaderValidatorCreate9 stub [100x]
http://bugs.winehq.org/show_bug.cgi?id=22029
Dmitry Timoshkov dmitry@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression
http://bugs.winehq.org/show_bug.cgi?id=22029
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |hverbeet@gmail.com
--- Comment #1 from Henri Verbeet hverbeet@gmail.com 2010-04-18 16:54:48 --- Does this still happen with current git? Do you know if this is specific to OS X?
http://bugs.winehq.org/show_bug.cgi?id=22029
--- Comment #2 from Jörg Höhle hoehle@users.sourceforge.net 2010-04-26 04:15:20 --- The bug is still present with wine-1.1.43. It's not until next week-end that I'll be able to test anything more recent.
Do you know if this is specific to OS X?
I don't know. My attempts at dual booting the Mac were unsuccessful and live CDs are lacking too many things.
http://bugs.winehq.org/show_bug.cgi?id=22029
--- Comment #3 from Jörg Höhle hoehle@users.sourceforge.net 2010-05-01 13:42:24 --- Still present in wine-git as of 2010-04-30.
Using XQuartz 2.3.3.2. I don't know whether I should switch to 2.5.0 or if that's going to cause upgrade troubles when eventually switching to Snow Leopard 10.6.x.
http://bugs.winehq.org/show_bug.cgi?id=22029
--- Comment #4 from Jörg Höhle hoehle@users.sourceforge.net 2010-05-16 09:30:35 --- The bug is still present in wine-1.1.44.
In bug #13335, comment #190 Roderick Colenbrander suggested using ORM=backbuffer at the cost of some graphical artefacts.
This indeed prevents the freeze in the first level in 1.1.44. The amazing graphical artefact is a "picture in picture in picture in..." view while in the cantina (a sort of "menu" level): the lower left shows a mini view of the whole virtual desktop, in its lower left again a mini view, and again and again... Quite amusing and not annoying. Normally there's nothing special in the lower left.
Using ORM=fbo or pbuffer causes the app to freeze.
http://bugs.winehq.org/show_bug.cgi?id=22029
--- Comment #5 from Jörg Höhle hoehle@users.sourceforge.net 2010-12-31 00:47:03 CST --- Still present. Using wine-1.3.9, MacOS X 10.5.8 with XQuartz 2.5.3
http://bugs.winehq.org/show_bug.cgi?id=22029
--- Comment #6 from Jörg Höhle hoehle@users.sourceforge.net 2011-04-16 07:14:17 CDT --- Still present. Using wine-1.3.18, MacOS X 10.5.8 with XQuartz 2.5.3 (not yet 2.6)
http://bugs.winehq.org/show_bug.cgi?id=22029
--- Comment #7 from Jörg Höhle hoehle@users.sourceforge.net 2011-04-26 14:59:10 CDT --- MacOS X 10.5.8 with latest updates and XQuartz 2.6.1 now. Still present.
http://bugs.winehq.org/show_bug.cgi?id=22029
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- OS/Version|Mac OS X 10.5 |Mac OS X
http://bugs.winehq.org/show_bug.cgi?id=22029
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |3a63552fd1880c973f28baf8f9f | |7a29fd24d1412
http://bugs.winehq.org/show_bug.cgi?id=22029
--- Comment #8 from Henri Verbeet hverbeet@gmail.com 2011-12-07 14:39:20 CST --- Created attachment 37853 --> http://bugs.winehq.org/attachment.cgi?id=37853 hack
This patch makes the demo work for me. That pretty much makes it Apple's bug, either in the X server or the driver. glXSwapBuffers() is not supposed to care what the current draw framebuffer binding is.
http://bugs.winehq.org/show_bug.cgi?id=22029
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |INVALID
--- Comment #9 from Henri Verbeet hverbeet@gmail.com 2011-12-07 15:03:24 CST --- Resolving INVALID.
http://bugs.winehq.org/show_bug.cgi?id=22029
--- Comment #10 from Stefan Dösinger stefan@codeweavers.com 2011-12-08 11:32:04 CST --- Created attachment 37863 --> http://bugs.winehq.org/attachment.cgi?id=37863 GLUT testcase
This is a bug I noticed a long time ago on Tiger on my Radeon X1600 iMac. The attached standalone app reproduces the problem on my Macbook Pro(Geforce 9600, running Snow Leopard), but not on the iMac, which is running Lion already.
you can compile the file with gcc swap.c -o swap -framework opengl -framework glut . If you want an X11-based one use gcc swap.c -o swap -I/usr/X11/include -L/usr/X11/lib -lGL -lglut and run it from a terminal that has the X11 environment properly set up(e.g. an xterm started by X11.app)
I assume Apple fixed this bug in Lion. I highly recommend to update. OSX drivers are bad enough as they are, and the ancient OSX versions will not get any updates.
Eventually I'll update the Macbook Pro to Lion. If the bug still persists on that version I'll file a bug at Apple. Considering that the old Radeon X1600 is fixed it is very likely that the newer Nvidia cards are fixed as well.
http://bugs.winehq.org/show_bug.cgi?id=22029
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #11 from Henri Verbeet hverbeet@gmail.com 2011-12-16 13:32:27 CST --- Closing.
http://bugs.winehq.org/show_bug.cgi?id=22029
--- Comment #12 from Jörg Höhle hoehle@users.sourceforge.net 2011-12-18 12:15:57 CST --- Henri, your hack fixed the freeze on my early 2009 NVidia Mac mini running old OSX 10.5.8 with recent XQuartz 2.6.3. Thank you.
-
wine-bugs＠winehq.org