[Bug 44586] New: Kamidori dungeon master crashes during animation after unspecified time
https://bugs.winehq.org/show_bug.cgi?id=44586
Bug ID: 44586 Summary: Kamidori dungeon master crashes during animation after unspecified time Product: Wine Version: 3.2 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: quartz Assignee: wine-bugs@winehq.org Reporter: zaplo00@mailfence.com Distribution: ---
After some play time (anything from 1 minute to 3 hours) the game either freezes with music playing on background or shows runtime error popup and quits. Sometimes it simply crashes and shows wine internal stack trace (which has nothing useful).
Problems always start after first "BaseMemAllocator_Commit fnAlloc failed" error message and happen during battle animations. This is example log just before crash:
-----
0097:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {5727a00f-50be-4519-8256-f7699871fecb} 0097:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {7e854ec7-cdc4-405a-b5b2-aaf7c9e7d40c} 0097:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {79a60dc6-5fc8-4952-a41c-1163aeec5eb8} 0097:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {2718d25b-5bf5-4479-8e88-babc64bdbfca} 0097:fixme:process:GetNumaHighestNodeNumber (0x1f51f880): semi-stub 0009:fixme:quartz:DSoundRender_QueryInterface No interface for {37d84f60-42cb-11ce-8135-00aa004bb851}! 0009:fixme:quartz:DSoundRender_QueryInterface No interface for {37d84f60-42cb-11ce-8135-00aa004bb851}! 0009:err:quartz:BaseMemAllocator_Commit fnAlloc failed with error 0x8007000e 0201:fixme:ntdll:EtwUnregisterTraceGuids deadbeef: stub 0009:fixme:quartz:DSoundRender_QueryInterface No interface for {37d84f60-42cb-11ce-8135-00aa004bb851}! 00a9:fixme:ntdll:EtwRegisterTraceGuidsW (0xb5a754a, (nil), {f7b697a3-4db5-4d3b-be71-c4d284e6592f}, 7, 0xb5f7020, (null), (null), 0xb5fc820): stub 00a9:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {72b14a7d-704c-423e-92f8-7e6d64bcb92a} 00a9:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {e2091f8a-1e0a-4731-84a2-0dd57c8a5261} 00a9:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {e8a3bf1f-a86b-4390-9c60-5390b969d22c} 00a9:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {5727a00f-50be-4519-8256-f7699871fecb} 00a9:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {7e854ec7-cdc4-405a-b5b2-aaf7c9e7d40c} 00a9:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {79a60dc6-5fc8-4952-a41c-1163aeec5eb8} 00a9:fixme:ntdll:EtwRegisterTraceGuidsW register trace class {2718d25b-5bf5-4479-8e88-babc64bdbfca} 00a9:fixme:process:GetNumaHighestNodeNumber (0x1f51f880): semi-stub 0009:fixme:quartz:DSoundRender_QueryInterface No interface for {37d84f60-42cb-11ce-8135-00aa004bb851}! 0009:fixme:quartz:DSoundRender_QueryInterface No interface for {37d84f60-42cb-11ce-8135-00aa004bb851}! 0009:err:quartz:BaseMemAllocator_Commit fnAlloc failed with error 0x8007000e 0009:fixme:quartz:DSoundRender_QueryInterface No interface for {37d84f60-42cb-11ce-8135-00aa004bb851}! 0009:err:quartz:BaseMemAllocator_Commit fnAlloc failed with error 0x8007000e 01cf:fixme:ntdll:EtwUnregisterTraceGuids deadbeef: stub
-----
I have gstreamer good/bad/ugly/base plugins for 32 and 64 bit.
https://bugs.winehq.org/show_bug.cgi?id=44586
Zebediah Figura z.figura12@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |z.figura12@gmail.com
--- Comment #1 from Zebediah Figura z.figura12@gmail.com --- It seems quite likely there's a memory leak somewhere.
A log with +quartz,+virtual would be helpful, although given the symptoms it's likely to be quite large.
https://bugs.winehq.org/show_bug.cgi?id=44586
Sagawa sagawa.aki+winebugs@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sagawa.aki+winebugs@gmail.c | |om
https://bugs.winehq.org/show_bug.cgi?id=44586
--- Comment #2 from zaplo00@mailfence.com --- (In reply to Zebediah Figura from comment #1)
It seems quite likely there's a memory leak somewhere.
A log with +quartz,+virtual would be helpful, although given the symptoms it's likely to be quite large.
I got a crash after two hours of playing in attack animation. This log is from all that time and has stack trace (WINEDEBUG=+quartz,+virtual).
This file probably crashes text editor, use glogg or notepad++ to view and search stuff.
http://www.mediafire.com/file/jc2iwscik0cscty/kamidori_log.tar.xz (54MB)
https://bugs.winehq.org/show_bug.cgi?id=44586
--- Comment #3 from zaplo00@mailfence.com --- This game is RPG / visual novel hybrid.
I should note that gameplay parts in this game like inventory GUI and dungeons are 'choppy' most of the time, animations lag in them and sometimes mouse clicks don't register if animation is not yet finished.
Visual novel parts are very smooth and have no performance problems at all. In those parts log file didn't grow show i think they don't use quartz or do something different than gameplay parts.
https://bugs.winehq.org/show_bug.cgi?id=44586
--- Comment #4 from zaplo00@mailfence.com --- I actually have LAV filters installed in this wine prefix (6 months old prefix).
In new wineprefix the screen is black for 5-10 seconds and then music starts playing but no opening video shows. Waiting until it's complete or skipping doesn't make difference, the screen stays black after video is complete and nothing happens.
https://bugs.winehq.org/show_bug.cgi?id=44586
--- Comment #5 from zaplo00@mailfence.com --- Created attachment 60612 --> https://bugs.winehq.org/attachment.cgi?id=60612 fresh prefix
Log in a new prefix
https://bugs.winehq.org/show_bug.cgi?id=44586
--- Comment #6 from Rafał Mużyło galtgendo@o2.pl --- (In reply to zaplo00 from comment #4)
In new wineprefix the screen is black for 5-10 seconds and then music starts playing but no opening video shows. Waiting until it's complete or skipping doesn't make difference, the screen stays black after video is complete and nothing happens.
That part is actually wrong or at least it became so a couple wine releases ago.
Part of it is an initialization problem - it seems sometimes proper decoders aren't getting initialized in time for the video stream to be caught by them. It's definitely just a timing problem, as adding a sufficient delay gets those codecs found by gstreamer.
The "stays black" part is a "lack of patience" mistake. The problem here is that for whatever the reason, the event that is supposed to (as it's the standard for intros) interrupt the intro playback and skip straight to the title screen, instead makes the player wait for the duration of the *entire* video clip (regardless of how much of it has already been played) and only then goes to the title screen (as if the timer on the clip was reset but a/v playback got redirected to /dev/null).
https://bugs.winehq.org/show_bug.cgi?id=44586
tokktokk fdsfgs@krutt.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |fdsfgs@krutt.org
https://bugs.winehq.org/show_bug.cgi?id=44586
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jeremielapuree@yahoo.fr
--- Comment #7 from joaopa jeremielapuree@yahoo.fr --- Does the bug still occur with wine-4.9?
https://bugs.winehq.org/show_bug.cgi?id=44586
--- Comment #8 from Rafał Mużyło galtgendo@o2.pl --- Well, wine 4.19 needs an as-of-yet-not-applied patch for a regression (but it's already queued), but after that the intro movie finally works as designed - a click dismisses it.
https://bugs.winehq.org/show_bug.cgi?id=44586
--- Comment #9 from Rafał Mużyło galtgendo@o2.pl --- For completion sake: above mentioned patch became commit cdeb55a8f94f1cfe3af01a6349d70c905b06583a.
https://bugs.winehq.org/show_bug.cgi?id=44586
--- Comment #10 from joaopa jeremielapuree@yahoo.fr --- Does the bug still occur with wine-7.0-rc4?
-
wine-bugs＠winehq.org -
WineHQ Bugzilla