April 2014 - Wine-Bugs - list.winehq.org

[Bug 25561] New: Setting watches in winedbg can generated unhandled floating point stack check exceptions
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

http://bugs.winehq.org/show_bug.cgi?id=25561 Summary: Setting watches in winedbg can generated unhandled floating point stack check exceptions Product: Wine Version: 1.3.9 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: winedbg AssignedTo: wine-bugs(a)winehq.org ReportedBy: austin.lund(a)gmail.com The way the recreate this may seem rather artificial, but it seems to always reproduce it for me. $ cd dlls/riched20/tests $ winedbg riched20_test.exe.so editor < dbgbug where the dbgbug file is: b test_EM_FORMATRANGE c watch * &fr.rc.bottom c c The output I get is attached. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 18

[Bug 35626] New: Patrician III: divide by zero exception scrolling the city view (side effect in user32.SubtractRect())
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

http://bugs.winehq.org/show_bug.cgi?id=35626 Bug ID: 35626 Summary: Patrician III: divide by zero exception scrolling the city view (side effect in user32.SubtractRect()) Product: Wine Version: unspecified Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: user32 Assignee: wine-bugs(a)winehq.org Reporter: jcantero(a)gmail.com Created attachment 47575 --> http://bugs.winehq.org/attachment.cgi?id=47575 Backtrace of the exception, wine trace log and an example .c of the bug The game 'Patrician III: Rise of the Hanse' crashes by a 'divide by zero' exception in certain conditions (later explained). I've debugged and found the root of the bug: it's a side effect in the wine implementation of user32.SubtractRect() function. In my attachment there's a C program that simulates the error (and a fixed version of the function to avoid it). I'm using wine 1.6.2, but this bug applies to all versions of wine to this day. In git-master it's here: https://github.com/mirrors/wine/blob/master/dlls/user32/uitools.c#L1424 . The bug was already reported serveral times, but due to lack of new information it was CLOSED ABANDONED (for example bugs #11768 and #16535). The bug is also mentioned as well-known in the comments of the AppDB entry of Patrician III (the vast majority of users avoid the bug disabling the feature that triggers it). Other applications can be also affected by this bug (see later). But first is first: the symptoms. The users of Patrician 3 suffer this bug when they try to scroll the city view and that city has a layer of weather effects (rain, fog, snow, ...). If the user changes the view of the city using the minimap, it works but only if the original rectangle of the view and the rectangle of the new position don't overlap. If they overlap in any way, or when the user tries to scroll the view (then it always overlap), the situation is detected by a call to user32.IntersectRect() function that returns TRUE, and the program branchs through 2 calls to SubtractRect() (I suppose to find the area that it's not in common between the two views). These calls fail when the actual version of the wine user32 library is used: they return an empty rectangle (with values {0, 0, 0, 0}) instead the expected one. The code of Patrician III is not prepared to handle the unexpected result, and when it tries to use such values in some calculations including a division, it crashes due a 'divide by zero' exception. That's what I found debugging the executable with winedbg. The attachment includes a log of (part of) the calls (it's a long list but the only relevant info is just before the exception near the end of the file). The cause of this bug lies in... let's say an unorthodox use of the SubtractRect() function. In the common case, the wine version works flawlessly. But that particular code of Patrician III calls SubtractRect() using the same pointer for the out value (dst) and one of the in values (src2). One of the first things that the wine implementation of SubtractRect() does is copying the scr1 value to dst. But with dst==src2, the assign also changes src2 as a side effect, provoking the bug. That behaviour is what is needed to fix. This bug can also affect other programs that uses the function in the same way (fortunately the bug is not present in Windows version, so there can't be any program that rely on it). In my attachment the SubtractRect.c that can be compiled (alone) to demo the bug behaviour. It calls SubtractRect() -a straight version from source code- with 3 different parameters (working as intended) and with the 1st and 3rd same parameter (bug). The .c has 2 #defines in the beginning, one to show more debug info and another to replace SubtractRect() with a fixed SubtractRect2() that I tested. The fix is not probably the best of the solutions, it only delays the copy *dest = *src1; past the uses of src1 and src2. But it works! (a better solution probably wouldn't use dst as a lvalue at all, only as a rvalue). I've applied the patch to wine 1.6.2 and now Patrician III is working with the weather effects to high :) -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 7

[Bug 19968] New: Half-Life 2: crashes when launching
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

http://bugs.winehq.org/show_bug.cgi?id=19968 Summary: Half-Life 2: crashes when launching Product: Wine Version: 1.1.29 Platform: Macintosh OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: nescalona(a)firehousemail.com Created an attachment (id=23484) --> (http://bugs.winehq.org/attachment.cgi?id=23484) terminal output Wine crashes when I click on the Launch button with Half-Life 2 selected in the My Games tab in Steam. I am running Arch Linux (dual-booted with Leopard on a Macbook (13-inch)), and I downloaded Wine from Arch's package manager. (http://www.archlinux.org/packages/extra/i686/wine/) After first encountering the crashing behavior, I removed ~/.wine and reinstalled everything pertaining to wine. I then installed Steam, following the instructions on the AppDB page for Steam. I then installed Half-Life 2 using Steam. When I click on the Launch button, Wine crashes. The attached file "log" is a copy of the terminal output when I start Wine with the command [code]wine "C:\Program Files\Steam\steam.exe"[/code]. The last four lines (not counting the $ prompt) appear once I click on the Launch button. The previous lines up to and including "fake device notification handle" appeared as Steam was loading. The attached file "report_file" was created using the command [code]WINEDEBUG=+relay wine "C:\Program Files\Steam\steam.exe" 2>&1 | tee relay.out;tail -n 100 relay.out > report_file[/code]. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 17

[Bug 28231] New: Black Dog Cribbage fails to start
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

http://bugs.winehq.org/show_bug.cgi?id=28231 Summary: Black Dog Cribbage fails to start Product: Wine Version: unspecified Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: craigfoster72(a)gmail.com Created an attachment (id=36156) --> (http://bugs.winehq.org/attachment.cgi?id=36156) Error output from starting game Black dog cribbage When I try to start up the game I get an error message . see attached file -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 14

[Bug 36008] New: urlmon/uri tests crash with WINEDEBUG=warn+heap
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

https://bugs.winehq.org/show_bug.cgi?id=36008 Bug ID: 36008 Summary: urlmon/uri tests crash with WINEDEBUG=warn+heap Product: Wine Version: 1.7.16 Hardware: x86 URL: https://test.winehq.org/data/7e874aed75765f3c017673178 cfd64301b7ba3ed/linux_ae-fedora64-heap/urlmon:uri.html OS: Linux Status: NEW Keywords: download, testcase Severity: normal Priority: P2 Component: urlmon Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com Created attachment 48139 --> https://bugs.winehq.org/attachment.cgi?id=48139 WINEDEBUG=warn+heap,urlmon,relay,seh,tid uri.c:11612: test IPersistStream... Unhandled exception: page fault on read access to 0xfeeefefa in 32-bit code (0x7ed73c4a). .. Backtrace: =>0 0x7ed73c4a test_IPersistStream+0xc2a() [/home/austin/wine-git/dlls/urlmon/tests/../../../include/objidl.h:3142] in urlmon_test (0x0034fc88) 1 0x7ed7c614 func_uri+0x32b3() [/home/austin/wine-git/dlls/urlmon/tests/uri.c:11613] in urlmon_test (0x0034fd48) -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 3

[Bug 26713] New: Teamfortress 2 crashes in the menu (tf2)
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

http://bugs.winehq.org/show_bug.cgi?id=26713 Summary: Teamfortress 2 crashes in the menu (tf2) Product: Wine Version: 1.3.17 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: kelfe(a)gmx.de Hi, there seems to be a repeatable crash when you launch steam, start tf2, go to mange items, crafting, view all blueprints, grab the scrollbar and try to scroll to the end. At about 80-90% scrolling down the game crashes here. using gentoo64 OpenGL vendor string: NVIDIA Corporation OpenGL renderer string: GeForce GTX 460/PCI/SSE2 OpenGL version string: 4.1.0 NVIDIA 270.18 err:mmtime:TIME_MMTimeStop Timer still active?! fixme:process:SetProcessShutdownParameters (000003ff, 00000000): partial stub. fixme:urlmon:CoInternetSetFeatureEnabled 5, 0x00000002, 1, stub fixme:urlmon:CoInternetSetFeatureEnabled 10, 0x00000002, 1, stub fixme:process:SetProcessShutdownParameters (000003ff, 00000000): partial stub. fixme:urlmon:CoInternetSetFeatureEnabled 5, 0x00000002, 1, stub fixme:urlmon:CoInternetSetFeatureEnabled 10, 0x00000002, 1, stub err:ntdll:RtlpWaitForCriticalSection section 0x7d7f64 "?" wait timed out in thread 0023, blocked by 0024, retrying (60 sec) -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 3

[Bug 26182] New: Steam half-life and TFC freezes
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

http://bugs.winehq.org/show_bug.cgi?id=26182 Summary: Steam half-life and TFC freezes Product: Wine Version: 1.3.14 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: critical Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: esk112(a)hotmail.com Created an attachment (id=33374) --> (http://bugs.winehq.org/attachment.cgi?id=33374) Wine output Half-life and Team Fortress Classic freezes when trying to create a local game or join a server. Starting up into the main menu and changing settings work. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 4

[Bug 24616] New: Launch Battlefield 2 (installed as Steam application) : BF2 crashes immediately
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

http://bugs.winehq.org/show_bug.cgi?id=24616 Summary: Launch Battlefield 2 (installed as Steam application) : BF2 crashes immediately Product: Wine Version: 1.3.4 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d AssignedTo: wine-bugs(a)winehq.org ReportedBy: robert_mt_walker(a)yahoo.co.uk Created an attachment (id=31095) --> (http://bugs.winehq.org/attachment.cgi?id=31095) Command Line trace from command: wine BF2.exe +fullscreen 0 +szx 1024 +szy 768 Launching BF2.exe executable crashes immediately. I am sorry but I have been following all the guidance for this application in the WineHQ app entry comments. I do have entries for 800x600 and 1024x768 monitor resolutions - given by xrandr. Perhaps someone could point me in the right direction to troubleshoot what is wrong with my configuration (or whether the problem is a regression). System: Nivida 8800GTX (768Mb) + Nvidia driver 256.53 Core i7 920 @4Ghz 6Gb DDR3 RAM Ubuntu 10.04.1 x64 (system fully updated) BTW I have "Half-Life 2.0", "Stalker : Shadow of Chernobyl" and "Bioshock I" working very well on the same Wine install version. Thanks for any help you guys can give me! Robert -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 7

[Bug 23359] New: Key input crashes newer source engine games
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

http://bugs.winehq.org/show_bug.cgi?id=23359 Summary: Key input crashes newer source engine games Product: Wine Version: 20050930 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: ajbader(a)umich.edu Both CS:S and Team Fortress 2 both seem to run fine, but any keyboard input at any time will cause the .exe to crash. Nothing else appears wrong. I can join games and play as long as no movement with keyboard is made. CS:S before the engine update worked fine, but now it's running on something similar to the TF2 engine, and it does the same thing. I can't grab a log of the crash, and the terminal doesn't report anything. ------------------ Arch Linux x86_64 3GB RAM 9800 GTX+ WINE package: bin32-wine-suse -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 14

[Bug 36006] New: msxml3/domdoc tests crash with WINEDEBUG=warn+heap
by wine-bugs＠winehq.org April 18, 2014

April 18, 2014

https://bugs.winehq.org/show_bug.cgi?id=36006 Bug ID: 36006 Summary: msxml3/domdoc tests crash with WINEDEBUG=warn+heap Product: Wine Version: 1.7.16 Hardware: x86 URL: https://test.winehq.org/data/7e874aed75765f3c017673178 cfd64301b7ba3ed/linux_ae-fedora64-heap/msxml3:domdoc.h tml OS: Linux Status: NEW Keywords: download, testcase Severity: normal Priority: P2 Component: windowscodecs Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com Created attachment 48138 --> https://bugs.winehq.org/attachment.cgi?id=48138 WINEDEBUG=relay,seh,tid,msxml fixme:msxml:domdoc_get_parseError (0x12e170)->(0x34f984): creating a dummy parseError wine: Unhandled page fault on read access to 0xfeeefef2 at address 0x7ed935ae (thread 0025), starting debugger... ... Backtrace: =>0 0x7ed935ae _expect_ref+0x1e(obj=0x144078, ref=0x2, line=0x9cb) [/home/austin/wine-git/dlls/msxml3/tests/domdoc.c:379] in msxml3_test (0x0034fb28) 1 0x7edaa43c test_refs+0x168b() [/home/austin/wine-git/dlls/msxml3/tests/domdoc.c:2507] in msxml3_test (0x0034fc08) 2 0x7edcd134 func_domdoc+0x403() [/home/austin/wine-git/dlls/msxml3/tests/domdoc.c:11866] in msxml3_test (0x0034fd48) -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 5