September 2010 - Wine-Bugs - list.winehq.org

[Bug 16577] New: ntdll/loader/actctx: add support of SxS assembly binding redirects
by wine-bugs＠winehq.org Oct. 29, 2021

Oct. 29, 2021

http://bugs.winehq.org/show_bug.cgi?id=16577 Summary: ntdll/loader/actctx: add support of SxS assembly binding redirects Product: Wine Version: 1.1.10 Platform: All URL: http://nikonusa.com/software/NX/1.3/win/CNX130NSAEN.EXE OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Hello, this is a showcase bug. I used the app from bug 12414 as example because it highlights some of the issues and hopefully helps to understand the problem (with my comments). We need to end this winetricks vcrun2XXX SxS mess/hell, which requires direct match of SxS assembly dependencies/bindings. Download: http://nikonusa.com/software/NX/1.3/win/CNX130NSAEN.EXE $ sha1sum CNX130NSAEN.EXE -> f31b38e3bcca17c8050c207b79315c9be23d582d Prerequisite: clean WINEPREFIX, winetricks dotnet20 --- The first installer show stopper (which is this bug report out) are actually two bugs: one in winetricks and one in the installer. When you did 'sh winetricks dotnet20' step and start the installer, an error message box will be shown shortly thereafter: --- snip --- Microsoft Visual C++ Runtime Library Program: c:\windows\temp\RarSFX0\cnx.exe R6034 An application has made an attempt to load the C runtime library incorrectly. --- snip --- The problem comes in form of a sub-installer dependency: WINEDEBUG=+tid,+seh,+loaddll,+process wine ./nikon.exe --- snip --- ... 0079:trace:process:CreateProcessW app (null) cmdline L"C:\\windows\\temp\\RarSFX0\\cnx.exe" ... 0079:trace:process:CreateProcessW started process pid 007a tid 007b ... 007b:trace:loaddll:free_modref Unloaded module L"C:\\windows\\temp\\nshbb9d.tmp\\UserInfo.dll" : native 007b:trace:loaddll:load_native_dll Loaded L"C:\\windows\\temp\\nshbb9d.tmp\\process.dll" at 0x10000000: native 007b:fixme:actctx:parse_depend_manifests Could not find dependent assembly L"Microsoft.VC80.CRT" 007b:trace:loaddll:load_native_dll Loaded L"C:\\windows\\temp\\nshbb9d.tmp\\LIBEXPATW.dll" at 0x3d0000: native 007b:trace:loaddll:load_builtin_dll Loaded L"C:\\windows\\system32\\mpr.dll" at 0x60c50000: builtin 007b:trace:loaddll:load_builtin_dll Loaded L"C:\\windows\\system32\\wininet.dll" at 0x60c00000: builtin 007b:trace:loaddll:load_builtin_dll Loaded L"C:\\windows\\system32\\msvcrt.dll" at 0x60c80000: builtin 007b:trace:loaddll:load_native_dll Loaded L"C:\\windows\\system32\\MSVCR80.dll" at 0x78130000: native 007b:trace:loaddll:load_native_dll Loaded L"C:\\windows\\system32\\MSVCP80.dll" at 0x7c420000: native 007b:trace:loaddll:load_native_dll Loaded L"C:\\windows\\temp\\nshbb9d.tmp\\Activation.dll" at 0x3a0000: native --- snip --- The culprit is: "fixme:actctx:parse_depend_manifests Could not find dependent assembly L"Microsoft.VC80.CRT" If we dump the embedded manifest from PE for the "Activation.dll": --- snip --- <assembly xmlns="urn:schemas-microsoft-com:asm.v1" manifestVersion="1.0"> <dependency> <dependentAssembly> <assemblyIdentity type="win32" name="Microsoft.VC80.CRT" version="8.0.50727.762" processorArchitecture="x86" publicKeyToken="1fc8b3b9a1e18e3b"></assemblyIdentity> </dependentAssembly> </dependency> <dependency> <dependentAssembly> <assemblyIdentity type="win32" name="Microsoft.VC80.CRT" version="8.0.50608.0" processorArchitecture="x86" publicKeyToken="1fc8b3b9a1e18e3b"> </assemblyIdentity> --- snip --- We can see two VC 8.0 runtime versions referenced, the VC 8.0 SP1 and an older VC 8.0 one. "8.0.50727.762" -> Visual C++ 2005 SP1 Redistributable Pack 8.0.50727.762 At this stage, only .NET Framwork 2.0 redist got installed as prerequisite in WINEPREFIX. Because of winetricks' winver 2k hack (missing junction point workaround, bug 12401), VC 8.0.50727.42 runtimes are placed in both: system32 and SxS repo. This results in errorneous behaviour in case of SxS assembly resolver failure (app references 8.x CRT assemblies not found in SxS) - a fallback load from system32 - which causes the CRT error. I already explained the differences between 2K and XP in several other bug reports when it comes to loading of SxS assemblies. In short: prevent any CRT (and ATL/MFC) 8.x+ runtime from being put into system32! Such assemblies are never to be loaded from system32 with winver > 2K! To fix this, 'winetricks dotnet20' step needs to include the removal of these assemblies from system32 as cleanup procedure. A second copy is properly installed into SxS as precaution from M$ guys in the light of win2k -> winxp update which knows about activation contexts/SxS. --- snip .wine/drive_c/windows/system32 --- $ ls -lsa msvc* 472 -rw-rw-r-- 1 focht focht 479232 2005-09-23 08:29 msvcm80.dll 540 -rw-rw-r-- 1 focht focht 548864 2005-09-23 08:29 msvcp80.dll 616 -rw-rw-r-- 1 focht focht 626688 2005-09-23 08:29 msvcr80.dll .. --- snip .wine/drive_c/windows/system32 --- The vendor msi installer bug at this stage is that the sub-installer is executed *before* the VC redist installer which would install all required CRT dependencies into SxS. Hence this error is only shown once. That sub-installer failure seems not critical. Now one might think: "ok, lets do the usual winetricks vcrun2005 ... vcrun2005sp1 mess" before install. If you do this, the error will still remain (assuming you restart with clean WINEPREFIX). The reason is the other "old" VC 8.0 CRT version, referenced in dll manifest: 8.0.50608.0 Winetricks dotnet20 installed a VC 8.0.50727.42 CRT version into SxS (after we removed the ones from system32). This would ideally work but Wine's loader doesn't support the redirection of assembly bindings yet. If you take a look at the policy file from .NET 2.0 installed VC 8 runtime: "c:\windows\winSxS\policies\x86_policy.8.0.Microsoft.VC80.CRT_1fc8b3b9a1e18e3b_x-ww_77c24773\": --- snip 8.0.50727.42.policy --- <bindingRedirect oldVersion="8.0.41204.256-8.0.50608.0" newVersion="8.0.50727.42"/> --- snip 8.0.50727.42.policy --- The policy rule would be satisfied, because the assembly asks for version 8.0.50608.0 and the current "up-to-date" version 8.0.50727.42 can be used for that. After VC 8.0 SP1 install (sh winetricks vcrun2005sp1), another policy file is added: --- snip 8.0.50727.762.policy --- <bindingRedirect oldVersion="8.0.41204.256-8.0.50608.0" newVersion="8.0.50727.762"/> <bindingRedirect oldVersion="8.0.50727.42-8.0.50727.762" newVersion="8.0.50727.762"/> --- snip 8.0.50727.762.policy --- This would cover (match) the first assembly binding dependency, 8.0.50727.762 Regards -- 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.

2 15

[Bug 15315] New: Ace Online (aka Space Cowboy/Flysis/Air Rivals) does not load past launcher with WINE
by wine-bugs＠winehq.org Oct. 28, 2021

Oct. 28, 2021

http://bugs.winehq.org/show_bug.cgi?id=15315 Summary: Ace Online (aka Space Cowboy/Flysis/Air Rivals) does not load past launcher with WINE Product: Wine Version: 1.0-rc3 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: littlegator(a)gmail.com I'm somewhat new to Linux, so if anyone offers any fixes, I may not be sure how to do it. Please inform me of the exact procedure of the fix (or anything that may advance the situation) and I will gladly do what you ask. Now, on to the problem. I load up Ace Online with WINE 1.0-rc3 and it brings me to the launcher. I type in my login details, select my resolution (which is 1024 x 768), uncheck the "windowed" check box (which appears to have no effect on the output), and wait. The game goes full screen and is white for a few seconds. It switches to black, then, after about 10 or 15 seconds, a white square appears in the middle of the screen. It disappear within a few frames of appearing, and I'm stuck here. Same thing happens when windowed, it just isn't full screen. During this process, it seems to be outputting every single mission log in the game to the debug window. The output file is extremely long. Now, my setup: I'm running WINE 1.0-rc3 with DirectX9.0 installed using WineTricks. I'm running a 1024x768 virtual desktop with no window decorations and window manager can not control the windows. My PC: Intel Pentium 4 1.6GHz 512MB RAM (not sure on what type >_>) ATI Radeon 9700 pro 128MB onboard RAM using latest Catalyst drivers running 3.5.6-9.fc7 Fedora release: 2.6.21-1.3194.fc7 While I await a response, I'm going to attempt updating Fedora since it's apparently an older version. I've attached the a couple output files and a dxdiag if it may be useful. -- 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.

2 13

[Bug 12694] New: Air Rival Crashes after login
by wine-bugs＠winehq.org Oct. 27, 2021

Oct. 27, 2021

http://bugs.winehq.org/show_bug.cgi?id=12694 Summary: Air Rival Crashes after login Product: Wine Version: 0.9.59. Platform: PC URL: http://appdb.winehq.org/objectManager.php?sClass=version &iId=11224&iTestingId=22275 OS/Version: Linux Status: UNCONFIRMED Severity: enhancement Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: mifuyne(a)gmail.com Created an attachment (id=12349) --> (http://bugs.winehq.org/attachment.cgi?id=12349) Log of Air Rival from Launch to Crash The game crashes without going into the game itself. Been tested on different versions of Windows (slightly different error on some version). The following address leads to the log, from launcher to the crash point. It's also attached to this report. http://pastebin.org/30959 WINE was set to Emulate a virtual desktop at 1024 x 768 (thus avoiding the resolution problem mentioned here: http://appdb.winehq.org/objectManager.php?sClass=version&iId=11224&iTesting… -- 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.

2 18

[Bug 12457] New: SharpDevelop 2. 2 crashes with NotImplementedException at IWebBrowser2.get_LocationName()
by wine-bugs＠winehq.org Oct. 26, 2021

Oct. 26, 2021

http://bugs.winehq.org/show_bug.cgi?id=12457 Summary: SharpDevelop 2.2 crashes with NotImplementedException at IWebBrowser2.get_LocationName() Product: Wine Version: CVS/GIT Platform: Other URL: http://www.icsharpcode.net/OpenSource/SD/Download/#Sharp Develop22 OS/Version: other Status: NEW Keywords: download Severity: normal Priority: P2 Component: shdocvw AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com http://lists.ximian.com/pipermail/mono-devel-list/2008-March/027227.html was asking how to install SharpDevelop 2 under Wine, so I gave it a try. "winetricks dotnet20" was enough to get it to install and start, but after it put up the GUI, it aborted with fixme:shdocvw:WebBrowser_QueryInterface (0x1af830)->({00000144-0000-0000-c000-000000000046} 0x33e800) interface not supported ... FATAL- System.NotImplementedException: The method or operation is not implemented. at System.Windows.Forms.UnsafeNativeMethods.IWebBrowser2.get_LocationName() at System.Windows.Forms.WebBrowser.get_DocumentTitle() -- 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.

2 27

[Bug 19756] New: TaskCoach: Cannot add new task with a due date
by wine-bugs＠winehq.org Oct. 26, 2021

Oct. 26, 2021

http://bugs.winehq.org/show_bug.cgi?id=19756 Summary: TaskCoach: Cannot add new task with a due date Product: Wine Version: 1.1.27 Platform: PC URL: http://www.taskcoach.org/download.html OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: jekwas(a)gmail.com Created an attachment (id=23126) --> (http://bugs.winehq.org/attachment.cgi?id=23126) taskcoach.exe.log I ran across this while testing for 8322. Task Coach is unable to add a new task if a Due Date is specified. To reproduce, click "New task" in the main window. Go to the Dates tab and add a due date. Press OK, and nothing will happen. When you close Task Coach, it will alert you of logged errors. I have attached such a log. (Wine doesn't output with any problems when trying to add a task) -- 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.

2 12

[Bug 12076] New: Database app refused to install unless system32/drivers/ etc/{services,host} exist
by wine-bugs＠winehq.org Oct. 26, 2021

Oct. 26, 2021

http://bugs.winehq.org/show_bug.cgi?id=12076 Summary: Database app refused to install unless system32/drivers/etc/{services,host} exist Product: Wine Version: CVS/GIT Platform: Other OS/Version: other Status: NEW Keywords: Installer Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com A user's informix-based app refused to install until the empty files c:/windows/system32/drivers/etc/{services,host} were created, see http://winehq.org/pipermail/wine-users/2008-March/030256.html I looked around a bit for other system32/drivers/etc files. http://support.microsoft.com/kb/130024 describes hosts, services, and protocols http://support.microsoft.com/kb/105997 describes lmhosts It looks like some apps find that directory by looking at HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\DataBasePath and MSN Messenger 7.5 f*cks up by changing that variable's type, see http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6343068 Getting this all right would be a lot of work. Just creating two empty files might be reasonable, though, since it allows the user's app to install. -- 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.

2 16

[Bug 16666] New: wine segfaults on launch
by wine-bugs＠winehq.org Oct. 24, 2021

Oct. 24, 2021

http://bugs.winehq.org/show_bug.cgi?id=16666 Summary: wine segfaults on launch Product: Wine Version: 1.1.11 Platform: PC OS/Version: OpenBSD Status: NEW Keywords: patch, source Severity: critical Priority: P2 Component: build-env AssignedTo: wine-bugs(a)winehq.org ReportedBy: austinenglish(a)gmail.com Created an attachment (id=18290) --> (http://bugs.winehq.org/attachment.cgi?id=18290) core dump Now for the biggest bug :-) Wine segfaults on launch of any program. Even 'wine --version' crashes. I've got this patch in place (from openbsd's port): diff --git a/loader/pthread.c b/loader/pthread.c index 4c0c892..e7f6479 100644 --- a/loader/pthread.c +++ b/loader/pthread.c @@ -96,6 +96,12 @@ static void init_thread( struct wine_pthread_thread_info *info ) /* if base is too large assume it's the top of the stack instead */ if ((char *)info->stack_base > &dummy) info->stack_base = (char *)info->stack_base - info->stack_size; +#elif defined(__OpenBSD__) + stack_t stack; + if (pthread_stackseg_np(pthread_self(), &stack) != 0) + abort (); + info->stack_base = (char *)stack.ss_sp - stack.ss_size; + info->stack_size = stack.ss_size; #else /* assume that the stack allocation is page aligned */ char dummy; @@ -163,6 +169,8 @@ static void init_current_teb( struct wine_pthread_thread_info *info ) info->pid = getpid(); #ifdef __sun info->tid = pthread_self(); /* this should return the lwp id on solaris */ +#elif defined(__OpenBSD__) + info->tid = pthread_self(); #elif defined(__APPLE__) info->tid = mach_thread_self(); #elif defined(__FreeBSD__) Though, it still segfaults without it. wine-pthread.core 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.

2 20

[Bug 21564] New: warning forever ignores keyboard input after losing/regaining focus
by wine-bugs＠winehq.org Oct. 20, 2021

Oct. 20, 2021

http://bugs.winehq.org/show_bug.cgi?id=21564 Summary: warning forever ignores keyboard input after losing/regaining focus Product: Wine Version: 1.1.37 Platform: x86 URL: http://www18.big.or.jp/~hikoza/Prod/index_e.html OS/Version: Linux Status: NEW Keywords: download Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: madewokherd(a)gmail.com Steps to reproduce: * Install Warning Forever from the linked website. * Start the program by running c:\wf\wf.exe. (The installer does not create a shortcut. This is the same on Windows.) * Select "Window mode". * Verify that the 'z' key can be used to make selections in the menu. * Switch to another window. * Switch back to Warning Forever window. * Verify that keyboard input no longer works. -- 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.

2 22

[Bug 22369] New: A child window without WS_EX_MDICHILD style should not be tiled by the WM_MDITILE message.
by wine-bugs＠winehq.org Oct. 20, 2021

Oct. 20, 2021

http://bugs.winehq.org/show_bug.cgi?id=22369 Summary: A child window without WS_EX_MDICHILD style should not be tiled by the WM_MDITILE message. Product: Wine Version: unspecified Platform: All OS/Version: All Status: UNCONFIRMED Severity: normal Priority: P2 Component: user32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: a.m.winehq(a)gmail.com In user32/mdi.c -> MDITile -> #L763(wine-1.1.42) for-loop "remove all the windows we don't want", the windows without WS_EX_MDICHILD window style should be also removed from win_array. MDICascade is also. -- 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.

2 37

[Bug 24352] New: Virtual desktop does not respect modeless child window Z-order
by wine-bugs＠winehq.org Oct. 20, 2021

Oct. 20, 2021

http://bugs.winehq.org/show_bug.cgi?id=24352 Summary: Virtual desktop does not respect modeless child window Z-order Product: Wine Version: 1.3.2 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: multitude(a)internode.on.net 1. Run 'winecfg' and ensure the 'Graphics -> Emulate a virtual desktop' option is off. 2. Press the OK button to apply any changes and close 'winecfg'. 2. Run 'wine notepad' and search for text (Ctrl-F). 3. Leaving the 'Find' window open, mouse click the main 'Notepad' window. CORRECT RESULT: The 'Notepad' window becomes active, the 'Find' child window becomes inactive and remains in front of the main 'Notepad' window. 4. Exit 'notepad', run 'winecfg' and enable the 'Graphics -> Emulate a virtual desktop' option. 5. Press the OK button to apply the change and close 'winecfg'. 6. Run 'wine notepad' again and search for text (Ctrl-F). 7. Leaving the 'Find' window open, mouse click the main 'Notepad' window. INCORRECT RESULT: When the 'Notepad' window becomes active the 'Find' window disappears behind the 'Notepad' window. Confirmed in Wine 1.3.2 and Wine 1.2. Bug found when testing Chessmaster Grandmaster Edition. -- 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.

2 18