January 2021 - Wine-bugs - list.winehq.org

[Bug 17902] New: ole32.CoGetClassObject: threads without apartment should use existing MTA
by wine-bugs＠winehq.org 21 Mar '21

21 Mar '21

http://bugs.winehq.org/show_bug.cgi?id=17902 Summary: ole32.CoGetClassObject: threads without apartment should use existing MTA Product: Wine Version: 1.1.18 Platform: Other URL: http://www.microsoft.com/downloads/details.aspx?FamilyID =fe6f2099-b7b4-4f47-a244-c96d69c35dec OS/Version: other Status: UNCONFIRMED Severity: normal Priority: P2 Component: ole32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Hello, there are apps which use free threaded COM (inproc) servers and call CoGetClassObject() from (new) threads that do not explicitly initialize COM. This is valid behaviour when one of the existing threads already created the MTA by calling CoInitializeEx(NULL, COINIT_MULTITHREADED). See: http://support.microsoft.com/kb/150777 --- quote --- Current implementations of COM allow a thread that does not explicitly initialize COM to be a part of the MTA. A thread that does not initialize COM is part of the MTA only if it starts using COM after at least one other thread in the process has previously called CoInitializeEx(NULL, COINIT_MULTITHREADED). (It is even possible that COM itself may have initialized the MTA when no client thread has explicitly done so; for example, a thread associated with an STA calls CoGetClassObject/CoCreateInstance[Ex] on a CLSID that is marked "ThreadingModel=Free" and COM implicitly creates an MTA into which the class object is loaded.) See the information on threading model interoperability below. However, this is a configuration that might cause problems, such as access violations, under certain circumstances. Therefore, it is strongly recommended that each thread that needs to do COM work initialize COM by calling CoInitializeEx and then, on completion of COM work, call CoUninitialize. The cost of "unnecessarily" initializing an MTA is minimal. --- quote --- An example app suffering from this is the managed console debugger from .NET 2.0 SDK "cordbg.exe": When trying to debug a simple console C# app created with symbols (.pdb format): --- snip --- wine "C:\\windows\\Microsoft.NET\\Framework\\v2.0.50727\\csc.exe" /debug+ foo.cs --- snip --- Start debugger: --- snip --- wine "C:\\Program Files\\Microsoft.NET\\SDK\\v2.0\\Bin\\cordbg.exe foo.exe" --- snip --- The debugger will immediately begin to execute the app - not halting in Main() method (source). This is because symbol information can't be loaded using COR symbol binder which lives in a free threaded COM inproc server. 0A29FF9E-7F9C-4437-8B11-F424491E3931 -> CLSID_CorSymBinder Threads 0x38 (main) and 0x36 call "CoInitializeEx" with COINIT_MULTITHREADED, creating the MTA. Thread 0x3b is created later as pooled thread - part of MTA - to serve CLR debugger requests (work items). That thread never makes any init calls to COM (verified). --- snip --- ... 0038:Call KERNEL32.GetProcAddress(60970000,7a37e2ae "CoInitializeEx") ret=79e9d311 0038:Ret KERNEL32.GetProcAddress() retval=609750c8 ret=79e9d311 0038:Call ole32.CoInitializeEx(00000000,00000000) ret=79ebfba8 0038:Call ntdll.RtlAllocateHeap(00110000,00000008,000000fc) ret=60986499 0038:Ret ntdll.RtlAllocateHeap() retval=0014f9c0 ret=60986499 0038:Call ntdll.RtlAllocateHeap(00110000,00000000,00000028) ret=609a750c 0038:Ret ntdll.RtlAllocateHeap() retval=001500e8 ret=609a750c 0038:Call KERNEL32.InitializeCriticalSection(001500f8) ret=609a7543 0038:Ret KERNEL32.InitializeCriticalSection() retval=00000001 ret=609a7543 0038:Call ntdll.RtlAllocateHeap(00110000,00000008,0000007c) ret=60983f54 0038:Ret ntdll.RtlAllocateHeap() retval=00151df0 ret=60983f54 0038:Call KERNEL32.InitializeCriticalSection(00151e10) ret=60983fb4 0038:Ret KERNEL32.InitializeCriticalSection() retval=00000001 ret=60983fb4 0038:Ret ole32.CoInitializeEx() retval=00000000 ret=79ebfba8 ... 0036:Call ole32.CoInitializeEx(00000000,00000000) ret=79ebfba8 0036:Call ntdll.RtlAllocateHeap(00110000,00000008,000000fc) ret=60986499 0036:Ret ntdll.RtlAllocateHeap() retval=00151e78 ret=60986499 0036:Ret ole32.CoInitializeEx() retval=00000000 ret=79ebfba8 ... 003b:Call advapi32.RegOpenKeyExW(80000000,0310d170 L"CLSID\\{0A29FF9E-7F9C-4437-8B11-F424491E3931}\\InprocServer32",00000000,00020019,0310d0dc) ret=79ea5b8a 003b:Ret advapi32.RegOpenKeyExW() retval=00000000 ret=79ea5b8a 003b:Call ... 003b:Call advapi32.RegOpenKeyExW(00000238,0310d270 L"2.0.50727",00000000,00020019,0310d0ec) ret=79ea5b8a 003b:Ret advapi32.RegOpenKeyExW() retval=00000000 ret=79ea5b8a 003b:Call advapi32.RegQueryValueExW(0000023c,7a136a40 L"ImplementedInThisVersion",00000000,0310d0e4,00000000,0310d0e8) ret=79ea5bd0 003b:Ret advapi32.RegQueryValueExW() retval=00000000 ret=79ea5bd0 003b:Call advapi32.RegCloseKey(0000023c) ret=7a1369f2 003b:Ret advapi32.RegCloseKey() retval=00000000 ret=7a1369f2 ... 003b:Call advapi32.RegOpenKeyExW(80000000,0310d318 L"CLSID\\{0A29FF9E-7F9C-4437-8B11-F424491E3931}\\InprocServer32\\2.0.50727",00000000,00020019,0310d314) ret=79ea5b8a 003b:Ret advapi32.RegOpenKeyExW() retval=00000000 ret=79ea5b8a 003b:Call advapi32.RegQueryValueExW(00000238,79f873d0 L"Class",00000000,0310d2f8,00000000,0310d310) ret=79ea5bd0 003b:Ret advapi32.RegQueryValueExW() retval=00000002 ret=79ea5bd0 003b:Call advapi32.RegCloseKey(00000238) ret=79f88946 003b:Ret advapi32.RegCloseKey() retval=00000000 ret=79f88946 ... 003b:Call KERNEL32.GetProcAddress(60970000,7a37e222 "CoGetClassObject") ret=79e9d311 003b:Ret KERNEL32.GetProcAddress() retval=60974f78 ret=79e9d311 003b:Call ole32.CoGetClassObject(00175d3c,00000015,00000000,79f83a7c,0310d5e8) ret=7a041335 003b:err:ole:CoGetClassObject apartment not initialised 003b:Ret ole32.CoGetClassObject() retval=800401f0 ret=7a041335 003b:Call KERNEL32.GetLastError() ret=79e74ab4 003b:Ret KERNEL32.GetLastError() retval=0000007f ret=79e74ab4 000007f ret=78132d14 ... 003b:Call KERNEL32.FormatMessageW(000031ff,00000000,800401f0,00000400,0310d4f4,00000000,0310d4a0) ret=7a127c38 003b:Ret KERNEL32.FormatMessageW() retval=00000000 ret=7a127c38 --- quote --- The failure is due to Wine's CoGetClassObject() rejecting requests when no apartment in current thread (oletls) exists. --- snip dlls/ole32/compobj.c --- HRESULT WINAPI CoGetClassObject( REFCLSID rclsid, DWORD dwClsContext, COSERVERINFO *pServerInfo, REFIID iid, LPVOID *ppv) { LPUNKNOWN regClassObject; HRESULT hres = E_UNEXPECTED; APARTMENT *apt; TRACE("\n\tCLSID:\t%s,\n\tIID:\t%s\n", debugstr_guid(rclsid), debugstr_guid(iid)); if (!ppv) return E_INVALIDARG; *ppv = NULL; apt = COM_CurrentApt(); if (!apt) { ERR("apartment not initialised\n"); return CO_E_NOTINITIALIZED; } ... --- snip dlls/ole32/compobj.c --- Rather than failing CoGetClassObject() without apartment, it should look for an existing MTA and let the current thread use it. With the MTA being used, the inproc server is created and the app works as intended, e.g. symbols get loaded, source can be stepped ... 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 7

[Bug 24875] New: SongSmith doesn't install
by wine-bugs＠winehq.org 21 Mar '21

21 Mar '21

http://bugs.winehq.org/show_bug.cgi?id=24875 Summary: SongSmith doesn't install Product: Wine Version: 1.3.5 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: VincentBeers(a)gmail.com Created an attachment (id=31485) --> (http://bugs.winehq.org/attachment.cgi?id=31485) Very short Wine log I wanted to try out this Microsoft application for laughs, but it turns out it won't install. Execution ends before even a single installer window is shown. A very short log is attached to this post, but it basically ends with this: err:msi:ITERATE_Actions Execution halted, action L"VSDCA_VsdLaunchConditions" returned 1603 Free demo download is available here: http://research.microsoft.com/en-us/um/redmond/projects/songsmith/download.… Running 64-bit Ubuntu 10.10 with 32-bit Wine 1.3.5. If any more info is required, I'll gladly help. -- 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 29

[Bug 31087] New: WMI class property retrieval: some CIM data type to OLE variant (VT) mappings are incorrect (CIM_UINT32, CIM_UINT16 ...)
by wine-bugs＠winehq.org 21 Mar '21

21 Mar '21

http://bugs.winehq.org/show_bug.cgi?id=31087 Bug #: 31087 Summary: WMI class property retrieval: some CIM data type to OLE variant (VT) mappings are incorrect (CIM_UINT32, CIM_UINT16 ...) Product: Wine Version: 1.5.7 Platform: x86 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: wmi&wbemprox AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Classification: Unclassified Hello, another problem from app in bug 24875 which deserves an extra bug. The app still fails with wine-1.5.7-252-g1f01355 The problems seems to be a WMI -> .NET type conversion problem of "ProcessId" property value. "InvalidCastException" is thrown --- snip --- 006ee1e8 7b83953f [HelperMethodFrame: 006ee1e8] 006ee258 040347f6 System.Management.PropertyData.MapWmiValueToValue(System.Object, System.Management.CimType, Boolean) 006ee2d0 04034204 System.Management.PropertyData.get_Value() 006ee2dc 04033aa6 System.Management.ManagementBaseObject.GetPropertyValue(System.String) 006ee2e8 04033a35 System.Management.ManagementBaseObject.get_Item(System.String) 006ee2ec 04034f70 SSCustomInstaller.SSInstaller.GetSongsmithSetupPathAndCopyMSI() 006ee31c 04034c0f SSCustomInstaller.SSInstaller.Install(System.Collections.IDictionary) 006ee350 649e3fc6 System.Configuration.Install.Installer.Install(System.Collections.IDictionary) 006ee398 649e5d38 System.Configuration.Install.AssemblyInstaller.Install(System.Collections.IDictionary) 006ee3d0 649e7f6e System.Configuration.Install.ManagedInstallerClass.InstallHelper(System.String[]) 006ee494 649e7734 System.Configuration.Install.ManagedInstallerClass.System.Configuration.Install.IManagedInstaller.ManagedInstall(System.String, Int32) --- snip --- MapWmiValueToValue converts WMI values (wrapped in object) to native .NET value (wrapped in object) with type conversion if necessary. MSDN claims "ProcessId" is CIM_UINT32/uint32. The following link shows example "ProcessId" property dump with 'wbemdump' tool: http://nukz.net/reference/wmi/hh/wmisdk/wbemdump_8wft.htm --- snip --- wbemdump /Q /wy root\cimv2 WQL "SELECT ProcessId FROM Win32_process WHERE Name = 'Winmgmt.exe'" This command produces the following output: (WQL) SELECT ProcessId FROM Win32_Process WHERE Name = 'Winmgmt.exe' ProcessId (CIM_UINT32/uint32) = 644 (0x284) --- snip --- For CIM_UINT32 type there seems to be 'unboxing' taking place: net_value = (uint)((int)(wmi_value)) I extracted a small C# snippet from the app .NET assembly which can be used for easy reproduction of the problem. Save the following snippet to a text file (ex: "test.cs"): --- snip test.cs --- using System; using System.Collections; using System.Management; class TestProgram { [STAThread] static void Main(string[] args) { using (ManagementObjectSearcher searcher = new ManagementObjectSearcher("Select * from Win32_Process")) { new ManagementScope(@"root\CIMV2"); foreach (ManagementObject obj2 in searcher.Get()) { string str = obj2["ProcessId"].ToString(); } } } } --- snip test.cs --- Compile it with C# compiler which is present if you installed .NET 2.0 Framework (winetricks). --- snip --- $ wine "C:\\windows\\Microsoft.NET\\Framework\\v2.0.50727\\csc.exe" /debug+ test.cs --- snip --- Run and get the same backtrace as with installer. The problem is CIMTYPE -> OLE Variant type conversion. Source: http://source.winehq.org/git/wine.git/blob/d2661b802556e073760af5d70869fb55… --- snip --- 665 HRESULT get_propval( const struct view *view, UINT index, const WCHAR *name, VARIANT *ret, 666 CIMTYPE *type, LONG *flavor ) 667 { ... 681 switch (view->table->columns[column].type & COL_TYPE_MASK) 682 { 683 case CIM_STRING: 684 case CIM_DATETIME: 685 V_VT( ret ) = VT_BSTR; 686 V_BSTR( ret ) = SysAllocString( (const WCHAR *)(INT_PTR)val ); 687 break; 688 case CIM_SINT16: 689 V_VT( ret ) = VT_I2; 690 V_I2( ret ) = val; 691 break; 692 case CIM_UINT16: 693 V_VT( ret ) = VT_UI2; 694 V_UI2( ret ) = val; 695 break; 696 case CIM_SINT32: 697 V_VT( ret ) = VT_I4; 698 V_I4( ret ) = val; 699 break; 700 case CIM_UINT32: 701 V_VT( ret ) = VT_UI4; 702 V_UI4( ret ) = val; 703 break; 704 case CIM_SINT64: 705 V_VT( ret ) = VT_BSTR; 706 V_BSTR( ret ) = get_value_bstr( view->table, row, column ); 707 break; 708 case CIM_UINT64: 709 V_VT( ret ) = VT_BSTR; 710 V_BSTR( ret ) = get_value_bstr( view->table, row, column ); 711 break; ... --- snip --- The unbox failure let me guess that CIM_UINT32 doesn't map to VT_UI4 but to VT_I4. Google "CIM_UINT32 VT_I4" and hit the first link -> book "Developing Wmi Solutions: A Guide to Windows Management ..." Bingo! There is a table which shows the type mapping (BTW good WMI resource to read...) uint32 - VT_I4 - CIM_UINT32 While you're at it you might want to fix other type conversions (CIM_UINT16 ...) according to the table. 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 8

[Bug 31088] New: SongSmith 1.0.2 fails to install due to unimplemented IWshShortcut::WindowStyle property (set/get)
by wine-bugs＠winehq.org 21 Mar '21

21 Mar '21

http://bugs.winehq.org/show_bug.cgi?id=31088 Bug #: 31088 Summary: SongSmith 1.0.2 fails to install due to unimplemented IWshShortcut::WindowStyle property (set/get) Product: Wine Version: 1.5.7 Platform: x86 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Classification: Unclassified Hello, when bug 31087 is fixed the app installer runs into unimplemented IWshShortcut::WindowStyle property (set/get) Managed backtrace: --- snip --- 006ee2a8 7b83953f [ComPlusMethodFrameStandaloneCleanup: 006ee2a8] Interop.IWshRuntimeLibrary.IWshShortcut.set_WindowStyle(Int32) 006ee2b8 04035355 SSCustomInstaller.SSInstaller.CreateDesktopShortcut() 006ee2ec 0403505c SSCustomInstaller.SSInstaller.GetSongsmithSetupPathAndCopyMSI() 006ee31c 04034c0f SSCustomInstaller.SSInstaller.Install(System.Collections.IDictionary) 006ee350 649e3fc6 System.Configuration.Install.Installer.Install(System.Collections.IDictionary) 006ee398 649e5d38 System.Configuration.Install.AssemblyInstaller.Install(System.Collections.IDictionary) 006ee3d0 649e7f6e System.Configuration.Install.ManagedInstallerClass.InstallHelper(System.String[]) 006ee494 649e7734 System.Configuration.Install.ManagedInstallerClass.System.Configuration.Install.IManagedInstaller.ManagedInstall(System.String, Int32) --- snip --- 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 7

[Bug 31105] New: .NET 3.x/4.x WPF based apps require windowscodecs.dll ComponentFactory_CreateBitmapFromSource
by wine-bugs＠winehq.org 21 Mar '21

21 Mar '21

http://bugs.winehq.org/show_bug.cgi?id=31105 Bug #: 31105 Summary: .NET 3.x/4.x WPF based apps require windowscodecs.dll ComponentFactory_CreateBitmapFromSource Product: Wine Version: 1.5.7 Platform: x86 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: windowscodecs AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Classification: Unclassified Hello, "SongSmith 1.0.2" is also affected by that insufficiency. Trace log: --- snip --- ... 002c:Call windowscodecs.IWICImagingFactory_CreateBitmapFromSource_Proxy(03163588,03145f14,00000002,0032c450) ret=02d45580 002c:fixme:wincodecs:ComponentFactory_CreateBitmapFromSource (0x3163588,0x3145f14,2,0x32c450): stub 002c:Ret windowscodecs.IWICImagingFactory_CreateBitmapFromSource_Proxy() retval=80004001 ret=02d45580 ... 002c:Call KERNEL32.RaiseException(e0434f4d,00000001,00000001,0032c3b0) ret=79f97065 002c:trace:seh:raise_exception code=e0434f4d flags=1 addr=0x7b83953f ip=7b83953f tid=002c 002c:trace:seh:raise_exception info[0]=80004001 002c:trace:seh:raise_exception eax=7b8264d9 ebx=7b8b17d0 ecx=80004001 edx=0032c2b4 esi=0032c390 edi=0032c310 002c:trace:seh:raise_exception ebp=0032c2f8 esp=0032c294 cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00000283 002c:trace:seh:call_stack_handlers calling handler at 0x79f9a3c8 code=e0434f4d flags=1 002c:trace:seh:call_stack_handlers handler at 0x79f9a3c8 returned 1 002c:trace:seh:call_stack_handlers calling handler at 0x7a3197d4 code=e0434f4d flags=1 --- snip --- Exception info: --- snip --- Exception object: 00ea0ff4 Exception type: System.NotImplementedException Message: The method or operation is not implemented. InnerException: <none> StackTrace (generated): <none> StackTraceString: <none> HResult: 80004001 --- snip --- CLR stack: --- snip --- OS Thread Id: 0x2b (0) ESP EIP 0033c6a8 7b83953f [HelperMethodFrame: 0033c6a8] 0033c74c 53c57cdc MS.Internal.HRESULT.Check(Int32) 0033c750 53afdfce System.Windows.Media.Imaging.RenderTargetBitmap.CopyCommon(System.Windows.Media.Imaging.RenderTargetBitmap) 0033c7a0 53afe13b System.Windows.Media.Imaging.RenderTargetBitmap.CloneCore(System.Windows.Freezable) 0033c7b0 568a68a4 System.Windows.Freezable.Clone() 0033c7bc 53a02d06 System.Windows.Media.Animation.Animatable.Clone() 0033c7c4 53ab5666 System.Windows.Media.ImageSource.Clone() 0033c7cc 53aeea62 System.Windows.Media.Imaging.BitmapSource.Clone() 0033c7d4 53ab76c7 System.Windows.Media.Effects.BitmapEffect.GetOutput(System.Windows.Media.Effects.BitmapEffectInput) 0033c818 53abcaab System.Windows.Media.Effects.BitmapEffectState.GetEffectOutput(System.Windows.Media.Visual, System.Windows.Media.Imaging.RenderTargetBitmap ByRef, System.Windows.Media.Matrix, System.Windows.Rect, System.Windows.Media.Matrix ByRef) 0033ca28 53abdb1a System.Windows.Media.Effects.BitmapEffectVisualState.RenderBitmapEffect(System.Windows.Media.Visual, System.Collections.Generic.List`1<ChannelSet>) 0033cb64 53abd7ca System.Windows.Media.Effects.BitmapEffectContent.ExecuteRealizationsUpdate() 0033cb6c 53918ad9 System.Windows.Media.RealizationContext+RealizationUpdateSchedule.Execute() 0033cba8 53911e40 System.Windows.Media.MediaContext.Render(System.Windows.Media.ICompositionTarget) 0033cc6c 53910fe7 System.Windows.Media.MediaContext.RenderMessageHandlerCore(System.Object) 0033ccb4 53910f00 System.Windows.Media.MediaContext.RenderMessageHandler(System.Object) 0033cccc 53910ecc System.Windows.Media.MediaContext.Resize(System.Windows.Media.ICompositionTarget) --- snip --- Code: http://source.winehq.org/git/wine.git/blob/39f80e56fb15858e1d057facae56fa7f… --- snip --- 465 static HRESULT WINAPI ComponentFactory_CreateBitmapFromSource(IWICComponentFactory *iface, 466 IWICBitmapSource *piBitmapSource, WICBitmapCreateCacheOption option, 467 IWICBitmap **ppIBitmap) 468 { 469 FIXME("(%p,%p,%u,%p): stub\n", iface, piBitmapSource, option, ppIBitmap); 470 return E_NOTIMPL; 471 } --- snip --- $ du -sh SongsmithSetup.1.02.msi 99M SongsmithSetup.1.02.msi $ sha1sum SongsmithSetup.1.02.msi fa6d5d889f37d369d61ec5c49dfb7f8d9932eb09 SongsmithSetup.1.02.msi $ wine --version wine-1.5.7-309-gb4da889 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 4

[Bug 48062] New: Detect and report WineTest's "load errors"
by WineHQ Bugzilla 21 Mar '21

21 Mar '21

https://bugs.winehq.org/show_bug.cgi?id=48062 Bug ID: 48062 Summary: Detect and report WineTest's "load errors" Product: Wine-Testbot Version: unspecified Hardware: x86 OS: Linux Status: NEW Severity: normal Priority: P2 Component: unknown Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com Distribution: --- Sometimes the WineTest report DLL version section contains lines like: mfplat=load error 1359 shell32=load error 1359 user32=load error 1359 See bug 48061 for more details. The TestBot should detect such errors as well as the "load error Configured as native" message and report them. -- 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 48430] New: Windows 2008 VM has broken Media Foundation libs
by WineHQ Bugzilla 21 Mar '21

21 Mar '21

https://bugs.winehq.org/show_bug.cgi?id=48430 Bug ID: 48430 Summary: Windows 2008 VM has broken Media Foundation libs Product: Wine-Testbot Version: unspecified Hardware: x86 OS: Linux Status: NEW Severity: normal Priority: P2 Component: unknown Assignee: wine-bugs(a)winehq.org Reporter: jacek(a)codeweavers.com Distribution: --- I noticed in user32:winstation tests that there is weird window "mfreadwrite_test.exe - Unable To Locate Component" messing with foreground window tests: http://test.winehq.org/data/37dec9f144501fb93e173214578b759122ac999a/2008_n… I investigated a bit deeper and it seems that the installation contains mfreadwrite.dll, but not mfplat.dll. When you attempt to load mfreadwrite.dll, Windows shows an error dialog about incomplete installation like: https://testbot.winehq.org/JobDetails.pl?Key=62943&s104=1#k104 It seems that there is a separated installer for MF for win2k8: https://www.microsoft.com/en-US/download/details.aspx?id=20424 I can see a few solutions: 1. Try to install MF into VM and hope it helps (and ignore failures on installations missing MF). 2. Load mfplat.dll and mfreadwrite.dll dynamically and skip tests if mfplat.dll is missing (this would make code uglier for a questionable benefit). 3. Get rid of win2k8 VMs and consider them unsupported. This may seem radical for the above problem, but I don't see any value in win2k8 test results in general. -- 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 2

[Bug 39297] New: kernel32.IsValidCodePage and friends don't support code page 708.
by wine-bugs＠winehq.org 21 Mar '21

21 Mar '21

https://bugs.winehq.org/show_bug.cgi?id=39297 Bug ID: 39297 Summary: kernel32.IsValidCodePage and friends don't support code page 708. Product: Wine Version: unspecified Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: kernel32 Assignee: wine-bugs(a)winehq.org Reporter: htl10(a)users.sourceforge.net Distribution: --- Microsoft's FontValidator (https://www.microsoft.com/typography/FontValidator.mspx) when processing certain fonts, try to access code page 708, which is not available under wine. FYI, mono has support for code page 708 - c.f. mcs/class/I18N/Rare/CP708.cs . -- 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 31

[Bug 18531] New: FontValidator (.NET based app) needs OleInPlaceObject_InPlaceDeactivate properly implemented
by wine-bugs＠winehq.org 21 Mar '21

21 Mar '21

http://bugs.winehq.org/show_bug.cgi?id=18531 Summary: FontValidator (.NET based app) needs OleInPlaceObject_InPlaceDeactivate properly implemented Product: Wine Version: 1.1.21 Platform: Other URL: http://www.microsoft.com/typography/FontValidator.mspx OS/Version: other Status: UNCONFIRMED Severity: normal Priority: P2 Component: shdocvw AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Hello, now that some urlmon MIME support patches are on the way which are essential for this app ("text/xml" MIME is later needed to show XML content in embedded browser window) I file an "intermediate" bug that needs to be fixed before we can verify the new functionality with real apps ;-) The story is basically about in-place activation and deactivation of embedded browser control. The current shdocvw OleInPlaceObject_InPlaceDeactivate stub is mostly harmless for many embedded browser control hosting apps because they don't rely on the fact that in-place deactivation doesn't really take place (control will be destroyed shortly anyway). This app use case is a bit special hence the current in-place deactivation stub becomes the culprit. Steps to reproduce: 1. clean WINEPREFIX 2. winetricks -q dotnet20 3. start app, "Font files" -> "Add..." -> select whatever true/opentype font from your system 4. "Table tests" -> "Clear All" -> select only "BASE" (shortens test) 5. "Rasterization" -> deselect "Test rasterization of TrueType outlines" (shortens test) 6. hit "Start validation" The .NET 2.0 app creates an embedded browser control: --- snip --- ... 0029:trace:ole:CoCreateInstance (rclsid={8856f961-340a-11d0-a96b-00c04fd705a2}, pUnkOuter=(nil), dwClsContext=00000001, riid={00000000-0000-0000-c000-000000000046}, ppv=0x32e2a4) ... 0029:fixme:shell:URL_ParseUrl failed to parse L"Interop.SHDocVw" ... 0029:trace:win:WIN_CreateWindowEx L"Shell Embedding" L"Shell Embedding" ex=00000100 style=46010000 0,0 0x0 parent=0x500b4 menu=(nil) inst=0x61020000 params=0x136d00 ... 0029:trace:win:WIN_CreateWindowEx created window 0x100b6 ... 0029:trace:shdocvw:create_shell_embedding_hwnd parent=0x500b4 hwnd=0x100b6 ... 0029:trace:shdocvw:OleObject_DoVerb (0x136d00)->(-5 (nil) 0x3c80034 -1 0x500b4 0xadd434) 0029:trace:shdocvw:OleObject_DoVerb OLEIVERB_INPLACEACTIVATE ... 0029:trace:win:DestroyWindow (0x500b4) ... --- snip --- and a "parking window" to host the control in inactive/hidden state until the real hosting window is about to be created and displayed to avoid runtime penalties of destroying and recreating browser control each time. This blog gives some insight in the world of "parking windows": http://blogs.msdn.com/sburke/archive/2007/06/20/flashback-windows-forms-par… --- snip --- 0029:trace:win:WIN_CreateWindowEx L"WindowsFormsParkingWindow" L"WindowsForms10.Window.8.app.0.378734a" ex=00010000 style=02010000 0,0 0x0 parent=0xfffffffd menu=(nil) inst=0x400000 params=(nil) ... 0029:trace:win:WIN_CreateWindowEx created window 0x100b8 0029:trace:win:show_window hwnd=0x100b6, cmd=0, wasVisible 1 ... 0029:fixme:shdocvw:OleInPlaceObject_InPlaceDeactivate (0x136d00) 0029:trace:shdocvw:WebBrowser_AddRef (0x136d00) ref=11 0029:trace:ole:GetErrorInfo (0, 0x32d9b0, 0x19a9c4) 0029:fixme:shdocvw:WebBrowser_QueryInterface (0x136d00)->({df0b3d60-548f-101b-8e65-08002b2bd119} 0x32d9a8) interface not supported 0029:trace:shdocvw:WebBrowser_Release (0x136d00) ref=10 0029:trace:seh:raise_exception code=e06d7363 flags=1 addr=0x7b84385b ip=7b84385b tid=0029 ... 0029:trace:win:WIN_SetWindowLong 0x100b6 -4 61031784 W 0029:trace:win:alloc_winproc reusing 0xffff0057 for (nil)/0x61031784 0029:trace:win:WIN_DestroyWindow 0x500b4 --- snip --- The CLR exceptions resulting from E_NOTIMPL of OleInPlaceObject_InPlaceDeactivate are handled gracefully. The window which is the real target for the browser control gets created later: --- snip --- 0029:trace:win:WIN_CreateWindowEx L"C:\\windows\\temp\\tmp47b0.tmp.report.xml" L"WindowsForms10.Window.8.app.0.378734a" ex=00050040 style=46cf0000 -2147483648,-2147483648 376x358 parent=0x10088 menu=(nil) inst=0x400000 params=(nil) ... 0029:trace:win:WIN_CreateWindowEx hwnd 0x600b4 cs 0,0 376x358 ... 0029:trace:win:show_window hwnd=0x600b4, cmd=5, wasVisible 0 0029:trace:shdocvw:WebBrowser_AddRef (0x136d00) ref=11 0029:trace:shdocvw:OleObject_DoVerb (0x136d00)->(-5 (nil) 0x3c80034 -1 0x600b4 0xaece88) 0029:trace:shdocvw:OleObject_DoVerb OLEIVERB_INPLACEACTIVATE 0029:trace:shdocvw:WebBrowser_Release (0x136d00) ref=10 0029:trace:seh:raise_exception code=e0434f4d flags=1 addr=0x7b84385b ip=7b84385b tid=0029 0029:trace:seh:raise_exception info[0]=80131509 0029:trace:seh:raise_exception eax=7b82ca1d ebx=7b8c2918 ecx=00000000 edx=0032d958 esi=0032d958 edi=e0434f4d 0029:trace:seh:raise_exception ebp=0032d920 esp=0032d8bc cs=0023 ds=002b es=002b fs=0063 gs=006b flags=00000246 --- snip --- The CLR exception leads to following backtrace dialog (and abort): --- snip --- ************** Exception Text ************** System.InvalidOperationException: Unable to get the window handle for the 'AxWebBrowser' control. Windowless ActiveX controls are not supported. at System.Windows.Forms.AxHost.EnsureWindowPresent() at System.Windows.Forms.AxHost.InPlaceActivate() at System.Windows.Forms.AxHost.TransitionUpTo(Int32 state) at System.Windows.Forms.AxHost.CreateHandle() at System.Windows.Forms.Control.CreateControl(Boolean fIgnoreVisible) at System.Windows.Forms.Control.CreateControl(Boolean fIgnoreVisible) at System.Windows.Forms.Control.CreateControl() at System.Windows.Forms.Control.WmShowWindow(Message& m) at System.Windows.Forms.Control.WndProc(Message& m) at System.Windows.Forms.ScrollableControl.WndProc(Message& m) at System.Windows.Forms.ContainerControl.WndProc(Message& m) at System.Windows.Forms.Form.WmShowWindow(Message& m) at System.Windows.Forms.Form.WndProc(Message& m) at System.Windows.Forms.Control.ControlNativeWindow.OnMessage(Message& m) at System.Windows.Forms.Control.ControlNativeWindow.WndProc(Message& m) at System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam) --- snip --- The activate_inplace() OLE verb bails out early because Wine thinks the control is in-place activated (This->inplace) -> real in-place deactivation never took place. .NET runtime tries to fetch hosted control's window handle which fails because no parent switch was done. Adding simple in-place deactivation (releasing IOleInPlaceSite, resetting This->inplace) lets this app succeed, having windows properly switched/re-parented. 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 9

[Bug 31609] New: TestLauncher says a required DLL is missing, but the test runs fine without TestLauncher
by wine-bugs＠winehq.org 21 Mar '21

21 Mar '21

http://bugs.winehq.org/show_bug.cgi?id=31609 Bug #: 31609 Summary: TestLauncher says a required DLL is missing, but the test runs fine without TestLauncher Product: Testbot Version: unspecified Platform: x86 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com Classification: Unclassified http://source.winehq.org/git/tools.git/blob/HEAD:/testbot/src/TestLauncher/… has a function AllImportedDllsPresent() that tries to avoid running tests that implicitly link against DLLs that aren't present. Unfortunately, in the case of dlls with manifests, it reports it can't find dlls that are in fact present. What it needs is a good dose of the medicine that cured the same problem in winetest: http://www.winehq.org/pipermail/wine-cvs/2010-July/068410.html A quick test shows that this works. I'll send a patch soon. -- 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