http://bugs.winehq.org/show_bug.cgi?id=23318
Summary: Bus Simulator 2008 (demo) complains about insufficient memory on startup Product: Wine Version: 1.2-rc4 Platform: x86 URL: http://www.gamershell.com/download_25660.shtml OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll AssignedTo: wine-bugs@winehq.org ReportedBy: gyebro69@gmail.com CC: julliard@winehq.org
Created an attachment (id=29068) --> (http://bugs.winehq.org/attachment.cgi?id=29068) console log when the demo crashes
Wine pops up an error dialogue when starting Bus Simulator 2008 (at least the demo I can test): 'Error E1004: Not enough memory (50 MB required)!' Afterwards Wine crashes.
The game was rated to Gold in AppDB by the submitter, using 1.1.4. After trying out several Wine versions it turned out the game doesn't crash when using Wine-1.1.24 or any earlier versions.
Result of regression testing:
09712593c8496be5e952b7316099f9eed5043203 is the first bad commit commit 09712593c8496be5e952b7316099f9eed5043203 Author: Alexandre Julliard julliard@winehq.org Date: Thu Jun 25 14:18:53 2009 +0200
ntdll: Release some address space after the process initialization is done.
:040000 040000 25715d0c052bc6bbc0262c742ad285d17a78f236 aa973aa3e5286174dcbffd782691d09bd9109e5d M dlls
The commit cannot be reverted even on the top of the nearest 1.1.25 version, but after
git checkout 09712593c8496be5e952b7316099f9eed5043203 >> the game crashes
git revert 09712593c8496be5e952b7316099f9eed5043203 >> the game starts up correctly.
There are other bugreports mentioning the same commit: bug #21198, bug #20986, bug #21630 and bug #22187.
Author of the patch added to CC. Link to the demo added to url.
http://bugs.winehq.org/show_bug.cgi?id=23318
GyB gyebro69@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download, regression
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #1 from Alexandre Julliard julliard@winehq.org 2010-06-22 13:12:47 --- I can't reproduce this. It works just fine here.
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #2 from GyB gyebro69@gmail.com 2010-06-22 13:28:08 --- (In reply to comment #1)
I can't reproduce this. It works just fine here.
Can you give it another try? Sometimes it launches properly right after installation..but all further attempts on starting the game resulted the crash for me.
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #3 from Alexandre Julliard julliard@winehq.org 2010-06-25 08:43:47 --- Still cannot reproduce.
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #4 from Andrew Nguyen arethusa26@gmail.com 2010-06-25 09:05:58 --- I see the error dialog on a Ubuntu Lucid and Fedora 13 system. Is there a trace that I should get to help diagnose the problem?
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #5 from Alexandre Julliard julliard@winehq.org 2010-06-25 09:11:08 --- Try a +relay,+virtual.
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #6 from Andrew Nguyen arethusa26@gmail.com 2010-06-25 09:16:16 --- Created an attachment (id=29129) --> (http://bugs.winehq.org/attachment.cgi?id=29129) Bus Simulator 2008 demo +relay,+virtual log (lzma compressed)
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #7 from Alexandre Julliard julliard@winehq.org 2010-06-25 11:12:00 --- It claims it wants 50Mb but it actually allocates 800Mb. We may well fail to find that much contiguous space, depending on where system libraries are loaded. It may be interesting to look at /proc/pid/maps when it fails, though I doubt there's much we can do about it.
http://bugs.winehq.org/show_bug.cgi?id=23318
Michael Stefaniuc mstefani@redhat.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW CC| |mstefani@redhat.com Ever Confirmed|0 |1
--- Comment #8 from Michael Stefaniuc mstefani@redhat.com 2010-11-14 17:37:40 CST --- Confirming as per comment #4
http://bugs.winehq.org/show_bug.cgi?id=23318
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jeremielapuree@yahoo.fr
--- Comment #9 from joaopa jeremielapuree@yahoo.fr 2010-11-14 18:40:47 CST --- I confirm too
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #10 from GyB gyebro69@gmail.com 2011-07-23 04:43:29 CDT --- The same error message on startup in Wine-1.3.25, this time on Fedora 15.
The game support page addresses the issue here (giving only generic solutions): http://bussimulator.icebytesgames.de/index.php?id=105&L=1
Google also returns several hits about the problem, posted by Windows users on different forums. Seemingly a problem in the game itself.
http://bugs.winehq.org/show_bug.cgi?id=23318
GyB gyebro69@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |09712593c8496be5e952b731609 | |9f9eed5043203
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #11 from jhgf bernhardloos@googlemail.com 2011-09-29 10:26:50 CDT --- I checked /proc/pid/maps, with some luck there might be enough free space if it would be possible to put everything together, but at the moment it's in a lot of pieces. So fixing this is pretty much out of question without a really clever idea.
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #12 from jhgf bernhardloos@googlemail.com 2011-10-01 14:21:05 CDT --- Created attachment 36637 --> http://bugs.winehq.org/attachment.cgi?id=36637 release the reserved address space until it's actually needed
Turns out I had a bunch of holes in my understanding of how exactly wine reserves virtual address space ... Anyway, the main problem is, that this app uses LoadLibrary to open most of it's dependencies (dinput, opengl, winmm and a few more), instead of linking directly against them and ld maps them all over the place. The attached patch fixes this by deferring the release of the reserved until a mmap call fails with ENOMEM. For this, it's necessary to hook mmap. Also, malloc calls still end up using the mmap in glibc directly and will fail if the unreserved memory runs out, but this doesn't seem to be a problem compared to mmap. A solution would be to hook malloc in the same way as mmap.
http://bugs.winehq.org/show_bug.cgi?id=23318
Brandon Corujo haku08879@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |haku08879@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=23318
Jarkko K jarkko_korpi@hotmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jarkko_korpi@hotmail.com
--- Comment #13 from Jarkko K jarkko_korpi@hotmail.com --- I tried this demo with 1.7.15 and I don't see the memory error.
http://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #14 from Béla Gyebrószki gyebro69@gmail.com --- (In reply to Jarkko K from comment #13)
I tried this demo with 1.7.15 and I don't see the memory error.
I still have the problem in Wine 1.7.16.
Fedora 20 Kernel 3.13.8-200.fc20.i686+PAE
https://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #15 from Béla Gyebrószki gyebro69@gmail.com --- Still in wine-1.7.34-36-g4493323
https://bugs.winehq.org/show_bug.cgi?id=23318
Jeff D. Hanson jhansonxi@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jhansonxi@gmail.com
--- Comment #16 from Jeff D. Hanson jhansonxi@gmail.com --- Xubuntu 14.04 x86_64
System #1: AMD Phenom 9550, GeForce GTX 750Ti with Nvidia 340.98 driver, 8GB RAM, 8GB swap System #2: Intel T2370 (2 x 1.73GHz), GM965, i915 driver, 4GB RAM, 5GB swap
Wine Staging 2.0-rc3 Windows version: Windows XP
Bus Simulator 2008 demo (no native components) Bus Simulator 2008 v1.4 (18 Wheels Of Steel CD, workarounds for bug #26463 only)
I'm unable to reproduce the problem with either the demo or the full CD game. On the smaller system I set mem=1G in the kernel boot line in Grub to test limited RAM (verified with "free"). Both versions still worked, even with several other apps open (Firefox, LibreOffice, etc.). With 1GB RAM the demo was noticeably slower then the full game but both loaded and I was able to drive the bus in 3D mode.
https://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #17 from Jeff D. Hanson jhansonxi@gmail.com --- Note - system #2 is running Xubuntu 16.04 x86_64
https://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #18 from Bruno Jesus 00cpxxx@gmail.com --- I can reproduce this bug anytime my RAM is low, and with enough RAM it ran 10 times in a row without issues (tested on 2.0-rc3). I don't use swap partition.
The mentioned commit no longer reverts, the code changed so manual reverting is not obvious.
I installed Wine version 1.1.24 and 25 from PlayOnLinux and I can reproduce the issue in both, so I'm inclined to say the bug is invalid. I believe that the mentioned commit changed things but still lack of system memory seems to be real cause.
In the other hand since Béla has been always able to reproduce the issue I'm not sure if invalid is the proper status.
https://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #19 from Béla Gyebrószki gyebro69@gmail.com --- (In reply to Bruno Jesus from comment #18)
... In the other hand since Béla has been always able to reproduce the issue I'm not sure if invalid is the proper status.
Not anymore...On my current setup (64-bit kernel, 32-bit Wine on Fedora 25) the demo starts properly even if I limit the available ram to 1GB. I also tried with Wine 1.7.0 and 1.7.20 (those are the oldest Wine versions I have), and the demo starts with them, too.
BusSimulator08_EnglishDemo.exe sha1: b055d84d2a422c8acc85821f2d7fe97bcaebd32e
https://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #20 from Jeff D. Hanson jhansonxi@gmail.com --- (In reply to Béla Gyebrószki from comment #19)
BusSimulator08_EnglishDemo.exe sha1: b055d84d2a422c8acc85821f2d7fe97bcaebd32e
Your sha1 matches the version I tested.
https://bugs.winehq.org/show_bug.cgi?id=23318
winetest@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winetest@luukku.com
--- Comment #21 from winetest@luukku.com --- Bug 34147 has the same regression id.
https://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #22 from joaopa jeremielapuree@yahoo.fr --- Does the bug still occur with current wine(3.21)?
https://bugs.winehq.org/show_bug.cgi?id=23318
--- Comment #23 from Béla Gyebrószki gyebro69@gmail.com --- (In reply to joaopa from comment #22)
Does the bug still occur with current wine(3.21)?
I'm unable to reproduce the problem anymore, tested with Wine 1.9.0 and 4.0-rc1.
It's probably best to resolve this bug as invalid or worksforme unless someone chimes in who can still reproduce the problem reliably.
https://bugs.winehq.org/show_bug.cgi?id=23318
Gijs Vermeulen gijsvrm@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |WORKSFORME Status|NEW |RESOLVED
--- Comment #24 from Gijs Vermeulen gijsvrm@gmail.com --- (In reply to Béla Gyebrószki from comment #23)
(In reply to joaopa from comment #22)
Does the bug still occur with current wine(3.21)?
I'm unable to reproduce the problem anymore, tested with Wine 1.9.0 and 4.0-rc1.
It's probably best to resolve this bug as invalid or worksforme unless someone chimes in who can still reproduce the problem reliably.
Resolving WORKSFORME. If anyone can still reproduce this reliably, the bug can be reopened.
https://bugs.winehq.org/show_bug.cgi?id=23318
Gijs Vermeulen gijsvrm@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #25 from Gijs Vermeulen gijsvrm@gmail.com --- Closing.