http://bugs.winehq.org/show_bug.cgi?id=21000
Summary: Tomb Raider III crashes in msadp32 Product: Wine Version: 1.1.31 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: olau@hardworking.dk
Created an attachment (id=25182) --> (http://bugs.winehq.org/attachment.cgi?id=25182) Backtrace from crash
I have no prior experience with wine, but I just tried installing Tomb Raider III from a cd with playonlinux.com. It worked flawlessly until I get past the 4th point in the training room, at which point it simply crashes (tried 3 times now, same pattern). The terminal has this output:
msadp32.c:247: cvtSSms16K: Assertion `*src <= 6' failed.
Plus a full backtrace (attached). I think it occurs when it is about to play the instruction sound for the 5th step in the training room.
Hope you can pinpoint and fix it. I was very impressed with seeing everything just work otherwise (old-time Linux user).
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #1 from Ole Laursen olau@hardworking.dk 2009-12-12 13:02:32 --- I've just tried 1.1.34, same problem. I'm on Ubuntu Karmic Koala by the way. If I disable sound, then it works. But of course for this particular game, they forgot to add subtitles if there's no sound. Argh. Still hoping you can fix it.
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #2 from Austin English austinenglish@gmail.com 2009-12-12 14:03:23 --- Please install debugging symbols and get another backtrace.
http://bugs.winehq.org/show_bug.cgi?id=21000
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |msadp32
http://bugs.winehq.org/show_bug.cgi?id=21000
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|msadp32 |-unknown
--- Comment #3 from Vitaliy Margolen vitaliy@kievinfo.com 2009-12-12 14:06:46 --- Three things: 1. Upgrade Wine (latest is wine-1.1.34) 2. Install Wine debug symbols to make your backtrace usable. 3. Get rid of pulseaudio/disable it
http://bugs.winehq.org/show_bug.cgi?id=21000
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |msadp32
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #4 from Jeff Zaroyko jeffz@jeffz.name 2009-12-12 17:14:17 --- regression? AppDB says it used to work in 1.1.20
http://wiki.winehq.org/RegressionTesting
http://bugs.winehq.org/show_bug.cgi?id=21000
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |wylda@volny.cz
--- Comment #5 from Wylda wylda@volny.cz 2009-12-13 15:08:09 --- (In reply to comment #4)
regression? AppDB says it used to work in 1.1.20
Not a reggression. Tomb Raider III - Adventures of Lara Croft works perfectly here. Tried a lot of wine versions (1.0.1, 1.1.22, 1.1.31, 1.1.34, 1.1.34-309-g9352509). Well it crashes, when WinXP is set in winecfq. But if i recall correctly you have to set compatibility mode even in WinXP to Win98.
So setting Win98 in winecfg should be OK.
Well, one other possibility is, that the reporter has Tomb Raider III - The Lost Artefact. I do not own this game, but found and tried demo. Also worked perfectly in 1.1.31, 1.1.34, 1.1.34-309-g9352509.
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #6 from Ole Laursen olau@hardworking.dk 2009-12-14 06:07:38 --- I did try it on 1.1.34 (see my second comment) and the CD says "Tomb Raider III - Adventures of Lara Croft". I'm not going to disable pulseaudio, I don't think it's possible on the latest Ubuntu without risking breaking everything.
Sorry about the backtrace, it does have the function names, I didn't notice missing line numbers. I'll see if I can figure out how to get it with debug symbols.
Wylda: did you test the training room with sound? The rest of the game seems to work. Unfortunately I really need the training in that room. :)
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #7 from Wylda wylda@volny.cz 2009-12-14 06:36:27 --- (In reply to comment #6)
Wylda: did you test the training room with sound? The rest of the game seems to work. Unfortunately I really need the training in that room. :)
Well, i'm game collector :) but not player anymore. So in this case i unwrap TR3, completly new and never played game and really have no idea where is is the training room.
So send me short how to get there (from menu game, etc.)... If it is somewhere during the game, send me your save game and i will retest.
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #8 from Ole Laursen olau@hardworking.dk 2009-12-14 16:40:03 --- Created an attachment (id=25214) --> (http://bugs.winehq.org/attachment.cgi?id=25214) Better backtrace
I installed a debug package, seems the backtrace's got some more info now, hope it helps.
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #9 from Ole Laursen olau@hardworking.dk 2009-12-14 16:46:34 --- Wylda: It's very easy, start the game, in the rotating menu select Lara's home instead of the real game, then when you're spawned, go out of the room, walk down the stairs, rotate and go through the hallway to the right. You end up at a place with a blocky track with a big 1. Follow the audio instructions. It should crash once you finish the fourth test (and end up on a block where I guess you get the instructions for the fifth test). Make sure audio is working for you, with audio turned off it doesn't crash for me.
Note with the default keybindings Shift = walk, Alt = jump and Ctrl = action key.
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #10 from Wylda wylda@volny.cz 2009-12-15 14:55:12 --- Created an attachment (id=25225) --> (http://bugs.winehq.org/attachment.cgi?id=25225) Console output from wine-1.1.34-469-g33a1934
Same problem here. Thanks for pointing me to the right place, Ole. Last working version was 0.9.60, but it can becuse of something is not implemented.
I failed with regression test, because i'm able to compile 0.9.60 and next succesful builds is 1.0-rc4. Nothing in between because of this:
gcc -c -I. -I. -I../../include -I../../include -D__WINESRC__ -D_NTSYSTEM_ -D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing -Wdeclaration-after-statement -Wwrite-strings -Wpointer-arith -g -O0 -o signal_i386.o signal_i386.c signal_i386.c: In function ‘merge_vm86_pending_flags’: signal_i386.c:544: error: ‘VIF_MASK’ undeclared (first use in this function) signal_i386.c:544: error: (Each undeclared identifier is reported only once signal_i386.c:544: error: for each function it appears in.) signal_i386.c:555: error: ‘VIP_MASK’ undeclared (first use in this function) signal_i386.c: In function ‘raise_vm86_sti_exception’: signal_i386.c:1242: error: ‘VIP_MASK’ undeclared (first use in this function) signal_i386.c: In function ‘__wine_enter_vm86’: signal_i386.c:1662: error: ‘VIF_MASK’ undeclared (first use in this function) signal_i386.c:1663: error: ‘VIP_MASK’ undeclared (first use in this function) make[2]: *** [signal_i386.o] Error 1 make[2]: Leaving directory `/home/pavel/install/wine/wine-git_tomb3/dlls/ntdll'
How to overcome this build failure? Consider UNCONFIRMED->NEW.
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #11 from Wylda wylda@volny.cz 2009-12-22 07:58:13 ---
Still not fixed in wine-1.1.35-56-g13a9c03. I tried this game also under WinXP and in point where wine crashes, WinXP begins to speak.
Adding Eric Pouech to CC as msadp32.c carry his Copyright (sorry if i add you wrongly).
http://bugs.winehq.org/show_bug.cgi?id=21000
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |eric.pouech@orange.fr
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #12 from Wylda wylda@volny.cz 2010-02-28 03:11:04 ---
Still not fixed in wine-1.1.39 (tested with one reverted patch - bug 21887).
Trace is the same like in comment #10 - nothing has changed.
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #13 from Wylda wylda@volny.cz 2010-03-19 15:40:48 ---
Still not fixed in wine-1.1.41.
Trace is the same like in comment #10 - nothing has changed.
http://bugs.winehq.org/show_bug.cgi?id=21000
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |hoehle@users.sourceforge.ne | |t
--- Comment #14 from Wylda wylda@volny.cz 2010-05-19 12:05:25 ---
Hi Jörg, i noticed you did recently some work in adp32 so maybe you could advise what is the problem here. The game crashes in place, when it should speak.
Backtrace is attached in comment #10 and is still valid even in wine-1.1.44-322-g5cc00e8. Any suggestion?
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #15 from Jörg Höhle hoehle@users.sourceforge.net 2010-05-20 04:13:13 --- Alas I'm not familiar with individual codecs but more with how they integrate into winmm.
What I generally recommend is to see if the bug can be isolated. E.g. a) identify the file that's being played, if possible b) use the MCIshell from bug #20232, comment #10 to try and play it and see if that crashes: open file.xyz type waveaudio alias w # use mciwave play w or open file.xyz type mpegvideo alias q # using mciquartz play q play w from 0 # try again close all Depending on the container format (adpcm inside .wav or other), you'll be able to use one or both of waveaudio or mpegvideo. If it crashes it will make it easier for anybody to reproduce it, not requiring the app.
It may well happen that you cannot isolate it, or that it only crashes within the context of the app (or when dsound uses it instead of winmm etc.)
error: ‘VIF_MASK’ undeclared
That's one of 2 frequent errors when attempting regression testing in the 0.9.50..current range. Use gitk or qgit to find the nearby commit to the same file that later fixed it and either git cherry-pick it or change the 2 lines by hand.
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #16 from GyB gyebro69@gmail.com 2011-06-14 11:44:31 CDT --- Created an attachment (id=35136) --> (http://bugs.winehq.org/attachment.cgi?id=35136) I-War 2: Edge of Chaos crashlog (Wine-1.3.22)
Independence War 2: Edge of Chaos crashes in msadp32 when the game is intended to play in-game speech (audio files, containing speech are in wav format, mono, 22050 kHz, 16 bps, codec: msadpcm). The generated backtrace is very similar to the one attached here for Tomb Raider III, so I think it is about the same problem.
I followed Jörg Höhle's advice and it turned out that wintest.exe mcishell plays the audio (wav) file (which crashes the game) correctly. I willingly provide further logs or tests if needed.
Workaround (for I-War 2): native quartz and devenum dlls.
Wine-1.3.22 Fedora 15 (32-bit) Alsa 1.0.24 Kernel 2.6.38.7
http://bugs.winehq.org/show_bug.cgi?id=21000
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression Status|UNCONFIRMED |NEW CC| |jeffz@jeffz.name Version|1.1.31 |0.9.61. Ever Confirmed|0 |1
--- Comment #17 from Wylda wylda@volny.cz 2011-07-20 19:06:03 CDT ---
1. I did a regression test between 0.9.60 and 0.9.61:
9c8c33006e53ecd3e7bfb1932313a1f7c9f13999 is the first bad commit commit 9c8c33006e53ecd3e7bfb1932313a1f7c9f13999 Author: Jeff Zaroyko jeffz@jeffz.name Date: Fri Apr 25 15:22:17 2008 +1000
msadp32.acm: Use the same szShortName value as windows.
:040000 040000 d1a76894acbbd3b1ee0cac88441e1bdbf7c18d71 333cd2d6cc3ab850464b65ee2345cd7c287b371a M dlls
2. No other bug report suffers from this commit.
3. Revert of this patch after git checkout makes that problem go away.
4. Adding author of this patch to CC.
--private keyword: bisected
Note: To be able to compile such low versions of wine on Debian Squeeze & kernel 2.6.35.13 i had to apply the following 3 patches to overcome above mentioned build failures: 1. 3499cb96b69d0c99da5ae68f5f7052c2ce8e66be 2. 7c616799894faa48ee3277bc276a39bed55780e0 3. de251e1c62651d8160b340a4fb9d0f9834a8e538
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #18 from Wylda wylda@volny.cz 2011-07-20 19:10:09 CDT --- Created an attachment (id=35622) --> (http://bugs.winehq.org/attachment.cgi?id=35622) Console log from wine-1.3.21 with WINEDEBUG=+adpcm,+tid,+seh
I guess there is some binary mess in backtrace again. The trace is intentionally taken from wine-1.3.21 as there is bug 27482 which blocks testing in later versions.
http://bugs.winehq.org/show_bug.cgi?id=21000
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- Depends on| |27482
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #19 from Jeff Zaroyko jeffz@jeffz.name 2011-07-20 19:55:01 CDT --- Thanks for the regression test Wylda.
My patch only lets programs recognise that this codec is available, before they'd fail to detect it present and either not use sound or use a different codec.
Since the application detects it, it tries to use it but unfortunately there is bug in decoding, which is unrelated to my change. So this bug was always present, just hiding.
http://bugs.winehq.org/show_bug.cgi?id=21000
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords|regression |
--- Comment #20 from Wylda wylda@volny.cz 2011-07-30 06:12:41 CDT ---
Removing the regression keyword, which i incorrectly added myself. 0.9.60 did not crash, but also did not produced any speech, so it couldn't crash.
In case of tomb raider 3, all the wav files are merged into one big audio file. It seems to me, that GyB has an easier testcase in comment #16. Maybe the key is repetitive playing (TR3 crashes always after fifth audio sequence).
http://bugs.winehq.org/show_bug.cgi?id=21000
Jeff Zaroyko jeffz@jeffz.name changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|jeffz@jeffz.name |
http://bugs.winehq.org/show_bug.cgi?id=21000
André H. nerv@dawncrow.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |nerv@dawncrow.de Regression SHA1| |9c8c33006e53ecd3e7bfb193231 | |3a1f7c9f13999
http://bugs.winehq.org/show_bug.cgi?id=21000
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1|9c8c33006e53ecd3e7bfb193231 | |3a1f7c9f13999 |
http://bugs.winehq.org/show_bug.cgi?id=21000
Bug 21000 depends on bug 27482, which changed state.
Bug 27482 Summary: Tomb Raider III demo crashes in the launcher http://bugs.winehq.org/show_bug.cgi?id=27482
What |Old Value |New Value ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #21 from Jörg Höhle hoehle@users.sourceforge.net 2012-03-23 11:27:31 CDT --- What's the current state? I'm lost among comments about TRIII vs. IW2:EoC.
wintest.exe mcishell plays the audio (wav) file (which crashes the game)
Regarding comment #16, I recommend you try both open file.xyz type waveaudio alias w # use mciwave and open file.xyz type mpegvideo alias q # using mciquartz play q Quartz uses very different code to drive the codec. If you opened a .wav file, it'll have used mciwave+winmm+msacm without quartz+dsound.
http://bugs.winehq.org/show_bug.cgi?id=21000
GyB gyebro69@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download URL| |http://www.tombraiderchroni | |cles.com/tr3/demo.html CC| |gyebro69@gmail.com
--- Comment #22 from GyB gyebro69@gmail.com 2012-03-25 01:17:59 CDT --- (In reply to comment #21)
What's the current state? I'm lost among comments about TRIII vs. IW2:EoC.
The crash is still present in Tomb Raider 3 demo, using Wine-1.5.0. As for IW2:Edge of Chaos, the crash (assertion) is no longer present with recent Wine so that must have been a different issue. That was probably bug #24239 (attachment #33343 is about the same assertion that I got when testing I-WAR2).
The crash in Tomb Raider 3 can be reproduced with the demo version (download link added to url). The launcher for the demo requires native ie6. Follow the guide in comment #9 to get to the point where the game crashes.
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #23 from Wylda wylda@volny.cz 2012-05-13 09:17:27 CDT ---
Still present in wine-1.5.4.
http://bugs.winehq.org/show_bug.cgi?id=21000
Brandon Corujo haku08879@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |haku08879@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #24 from GyB gyebro69@gmail.com 2013-08-20 04:01:32 CDT --- Still an issue in wine-1.7.0-114-g73aef40. The demo crashes after an assertion occurred in msadp32.acm: msadp32.c:247: cvtSSms16K: Assertion `*src <= 6' failed.
https://bugs.winehq.org/show_bug.cgi?id=21000
Ken Sharp imwellcushtymelike@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |Abandoned?
--- Comment #25 from Ken Sharp imwellcushtymelike@gmail.com --- Is this still an issue in Wine 1.7.44 or later?
https://bugs.winehq.org/show_bug.cgi?id=21000
Béla Gyebrószki gyebro69@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #35136|0 |1 is obsolete| |
--- Comment #26 from Béla Gyebrószki gyebro69@gmail.com --- Created attachment 51619 --> https://bugs.winehq.org/attachment.cgi?id=51619 crash in wine-1.7.44-83-g9ce9ba3
Still present in wine-1.7.44-83-g9ce9ba3
tr3_complete.exe (212 MB, the launcher needs IE8) sha1: ae342af7ab82e772ba416223dc8cc2081d1698c1
https://bugs.winehq.org/show_bug.cgi?id=21000
Vanack Sabbadium ubuntugamingproject@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ubuntugamingproject@gmail.c | |om
--- Comment #27 from Vanack Sabbadium ubuntugamingproject@gmail.com --- I'm using Wine 1.7.47.
If it could be useful, i totally solved the issue using this patch:
http://www.universotombraider.com/xptomb3.zip
And making the game run with Windows XP configuration. No more crashes, in Lara's home neither first main level.
I don't know if that patch could be used only for the italian version of the game, but i should give it a try.
https://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #28 from Bruno Jesus 00cpxxx@gmail.com --- The crash is very clear in the backtrace and ignoring the assert will make a weird sound play. It could be that the wav file is in a slightly different format than wine expects. It's hard to determine what sound it is because the +dsound log is too verbose and all wav files from the game are packed in a wad file (the wad can be opened with http://www.aspidetr.com/tools/tr-cdaudio-player/ but still has to find which wav is it among over a hundred).
https://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #29 from Bruno Jesus 00cpxxx@gmail.com --- I think I found the issue, while playing the Excellent! wav file wine keeps reading data and gets to the next wav file header. The bytes it is trying to parse are exactly the RIFF text bytes from a wav header, since R (0x52) is > 0x06 the assert is triggered.
https://bugs.winehq.org/show_bug.cgi?id=21000
Bruno Jesus 00cpxxx@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |00cpxxx@gmail.com
--- Comment #30 from Bruno Jesus 00cpxxx@gmail.com --- Created attachment 52430 --> https://bugs.winehq.org/attachment.cgi?id=52430 patch
The attached patch fixes the problem from Lara's home for me and the sound plays fine with no distortion. Please if anyone can help testing I would appreciate.
https://bugs.winehq.org/show_bug.cgi?id=21000
--- Comment #31 from Béla Gyebrószki gyebro69@gmail.com --- (In reply to Bruno Jesus from comment #30)
Created attachment 52430 [details] patch
The attached patch fixes the problem from Lara's home for me and the sound plays fine with no distortion. Please if anyone can help testing I would appreciate.
I tried the patch with TR3 demo, it resolves the crash in it.
err:adpcm:cvtSSms16K Invalid ADPCM data, stopping conversion
is shown in the terminal after 'excellent' wav is played. I didn't test the demo further though.
https://bugs.winehq.org/show_bug.cgi?id=21000
super_man@post.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |super_man@post.com
--- Comment #32 from super_man@post.com --- The patch was committed.
https://bugs.winehq.org/show_bug.cgi?id=21000
Nikolay Sivov bunglehead@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords|Abandoned? |
https://bugs.winehq.org/show_bug.cgi?id=21000
André H. nerv@dawncrow.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |72528be84fdc0dcfe75e361f4bf | |7c09fc27b638d Status|NEW |RESOLVED Resolution|--- |FIXED
--- Comment #33 from André H. nerv@dawncrow.de --- Sounds fixed to me, please reopen if not.
https://bugs.winehq.org/show_bug.cgi?id=21000
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #34 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 1.7.53.