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

[Bug 24196] New: oleaut32: typelib registration should not fail bitness-neutral assemblies (32-bit typelib wrapped in 64-bit PE, x64 .NET 2.0 installer)
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

http://bugs.winehq.org/show_bug.cgi?id=24196 Summary: oleaut32: typelib registration should not fail bitness-neutral assemblies (32-bit typelib wrapped in 64-bit PE, x64 .NET 2.0 installer) Product: Wine Version: 1.3.1 Platform: x86-64 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: oleaut32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Hello, another x64 .NET Framework 2.0 installer bug. The typelib registration fails for some assemblies: --- snip --- ... 002c:Call KERNEL32.CreateProcessA(00000000,7fffec16f650 "\"C:\\windows\\Microsoft.NET\\Framework64\\v2.0.50727\\regtlibv12.exe\" \"C:\\windows\\Microsoft.NET\\Framework64\\v2.0.50727\\System.Drawing.tlb\"",00000000,00000000,7fff00000001,00000000,00000000,00000000,7fffec8edfe8,7fffec8eb9e0) ret=4fc035780 ... 002e:Call oleaut32.LoadTypeLib(7ffff143f880 L"C:\\windows\\Microsoft.NET\\Framework64\\v2.0.50727\\System.Drawing.tlb",7ffff143f870) ret=004027bf ... 002e:Call KERNEL32.LoadLibraryExW(7ffff143f360 L"C:\\windows\\Microsoft.NET\\Framework64\\v2.0.50727\\System.Drawing.tlb",00000000,0000000b) ret=7ffff105ffb0 002e:Ret KERNEL32.LoadLibraryExW() retval=7fffefad0001 ret=7ffff105ffb0 ... 002e:Ret oleaut32.LoadTypeLib() retval=00000000 ret=004027bf 002e:Call oleaut32.RegisterTypeLib(7ffff74d30d0,7ffff143f880 L"C:\\windows\\Microsoft.NET\\Framework64\\v2.0.50727\\System.Drawing.tlb",00000000) ret=00402808 ... 002e:Ret oleaut32.RegisterTypeLib() retval=800288bd ret=00402808 ... 002c:Call msi.MsiRecordSetStringW(00000005,00000000,7fffec8ec620 L"RegisterTypeLib of C:\\windows\\Microsoft.NET\\Framework64\\v2.0.50727\\System.Drawing.tlb failed : 800288bd") ret=4fc0358bd --- snip --- The PE file containing typelib resources "System.Drawing.tlb" is a PE64 binary. Dumping the typelib resource reveals: --- snip --- Offset 0 1 2 3 4 5 6 7 8 9 A B C D E F Ascii 000002B0 4D 53 46 54 02 00 01 00 00 00 00 00 09 04 00 00 MSFT......... 000002C0 00 00 00 00 41 00 00 00 02 00 00 00 00 00 00 00 ....A.......... 000002D0 0C 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ............... --- snip --- flags: 0x41 -> indicates the typelib itself is 32 bit (SYSKIND = SYS_WIN32) The folder "c:\windows\Microsoft.NET\Framework64\v2.0.50727\" contains assemblies with following characteristics: 64-bit PE + 64 bit typelib part (flags = 0x43 -> SYSKIND = SYS_WIN64) 64-bit PE + 32 bit typelib part (flags = 0x41 -> SYSKIND = SYS_WIN32) I verified that Wine msi correctly installs them into their appropriate 32 bit and 64 bit folders. The 32-bit counterparts from "c:\windows\Microsoft.NET\Framework\v2.0.50727\" are all: 32-bit PE + 32 bit typelib part (flags = 0x41 -> SYSKIND = SYS_WIN32) === What causes the embedded typelib to be 32-bit? Well, the .NET assemblies these typelibs belong to are flagged as "MSIL" assemblies. These are bitness-neutral assemblies that are portable and can run under any platform. When you look at 64-bit -> "c:\windows\Microsoft.NET\Framework64\v2.0.50727\RedistList\FrameworkList.xml": --- snip --- <File AssemblyName="System.Drawing" Version="2.0.0.0" PublicKeyToken="b03f5f7f11d50a3a" Culture="neutral" ProcessorArchitecture="MSIL" FileVersion="2.0.50727.42" InGAC="true" /> --- snip --- Example for bitness-aware assembly (managed C++ assemblies are automatically platform specific): 64-bit -> "c:\windows\Microsoft.NET\Framework64\v2.0.50727\RedistList\FrameworkList.xml": --- snip --- <File AssemblyName="System.Data" Version="2.0.0.0" PublicKeyToken="b77a5c561934e089" Culture="neutral" ProcessorArchitecture="AMD64" FileVersion="2.0.50727.42" InGAC="true" /> --- snip --- 32-bit -> "c:\windows\Microsoft.NET\Framework\v2.0.50727\RedistList\FrameworkList.xml": (the 64 bit .NET Framework 2.0 redist contains both: the 64 bit _and_ 32 bit runtime) --- snip --- <File AssemblyName="System.Data" Version="2.0.0.0" PublicKeyToken="b77a5c561934e089" Culture="neutral" ProcessorArchitecture="x86" FileVersion="2.0.50727.42" InGAC="true" /> --- snip --- I guess Wine should not fail on these bitness-neutral assemblies that contain 32-bit typelibs. --- snip dlls/oleaut32/typelib.c --- HRESULT WINAPI RegisterTypeLib( ITypeLib * ptlib, /* [in] Pointer to the library*/ OLECHAR * szFullPath, /* [in] full Path of the library*/ OLECHAR * szHelpDir) /* [in] dir to the helpfile for the library, may be NULL*/ { ... if (ptlib == NULL || szFullPath == NULL) return E_INVALIDARG; if (FAILED(ITypeLib_GetLibAttr(ptlib, &attr))) return E_FAIL; TRACE("(%s,%d)", debugstr_w(szFullPath), attr->syskind); #ifdef _WIN64 if (attr->syskind != SYS_WIN64) return TYPE_E_BADMODULEKIND; #else if (attr->syskind != SYS_WIN32 && attr->syskind != SYS_WIN16) return TYPE_E_BADMODULEKIND; #endif ... } --- snip dlls/oleaut32/typelib.c --- 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 14

[Bug 24200] New: msi: properly handle msidbComponentAttributes64bit attribute to support x64 installers that mix architectures in a single MSI package (32-bit and 64-bit components, filesystem, registry)
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

http://bugs.winehq.org/show_bug.cgi?id=24200 Summary: msi: properly handle msidbComponentAttributes64bit attribute to support x64 installers that mix architectures in a single MSI package (32-bit and 64-bit components, filesystem, registry) Product: Wine Version: 1.3.1 Platform: x86-64 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: msi AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Hello, one of the main problems with current Wine msi is there are x64 installers that mix architectures in a single MSI package: 64-bit and 32-bit components (without a separate 32-bit installer for 32-bit components). Basically 'msidbComponentAttributes64bit' is not handled. Example: "mscoree.dll" -> .NET bootstrapper "Component" table: --- snip 32-bit --- MSCOREE_DLL_____X86.3643236F_FC70_11D3_A536_0090278A1BB8 {173A6EB3-6403-11D4-A53F-0090278A1BB8} DD_SystemFolder_X86.3643236F_FC70_11D3_A536_0090278A1BB8 8 FL_mscoree_dll_____X86.3643236F_FC70_11D3_A536_0090278A1BB8 --- snip 32-bit --- Attributes = 8 --- snip 64-bit --- MSCOREE_DLL_____A64.3643236F_FC70_11D3_A536_0090278A1BB8 {70B495DC-D747-4182-B6D7-86C8A2244B25} SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8 264 FL_mscoree_dll_____A64.3643236F_FC70_11D3_A536_0090278A1BB8 --- snip 64-bit --- Attributes = 264 (256 = msidbComponentAttributes64bit + 8) When it comes to "InstallFiles" action: --- snip 32-bit component --- 001b:trace:msi:msi_get_property returning L"C:\\windows\\system32\\" for property L"DD_SystemFolder_X86.3643236F_FC70_11D3_A536_0090278A1BB8" ... 001b:Call KERNEL32.MultiByteToWideChar(00000000,00000000,7fffe7ce1110 "FL_mscoree_dll_____X86.3643236F_FC70_11D3_A536_0090278A1BB8",ffffffff,7fffe7d4ba00,7fff0000003c) ret=7fffed8ff38c ... 001b:trace:msi:resolve_folder Working to resolve L"DD_SystemFolder_X86.3643236F_FC70_11D3_A536_0090278A1BB8" ... 001b:trace:msi:cabinet_copy_file extracting L"C:\\windows\\system32\\mscoree.dll" 001b:Call KERNEL32.CreateFileW(7fffe7d0b980 L"C:\\windows\\system32\\mscoree.dll",c0000000,00000000,00000000,7fff00000002,7fff00000080,00000000) ret=7fffed8ff409 001b:Ret KERNEL32.CreateFileW() retval=000000ac ret=7fffed8ff409 --- snip 32-bit component --- --- snip 64-bit component --- 001b:trace:msi:msi_get_property returning L"C:\\windows\\system32\\" for property L"SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8" ... 001b:Call KERNEL32.MultiByteToWideChar(00000000,00000000,7fffe7d247a0 "FL_mscoree_dll_____A64.3643236F_FC70_11D3_A536_0090278A1BB8",ffffffff,7fffe7d15ab0,0000003c) ret=7fffed8ff38c ... 001b:trace:msi:resolve_folder Working to resolve L"SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8" ... 001b:trace:msi:cabinet_copy_file extracting L"C:\\windows\\system32\\mscoree.dll" 001b:Call KERNEL32.CreateFileW(7fffe7d15b40 L"C:\\windows\\system32\\mscoree.dll",c0000000,00000000,00000000,7fff00000002,00000080,00000000) ret=7fffed8ff409 001b:Ret KERNEL32.CreateFileW() retval=000000b0 ret=7fffed8ff40 --- snip 64-bit component --- The first extracted 32-bit version that gets (incorrectly) extracted to system32 folder (remember: we're 64-bit install here) and subsequently gets overwritten with 64-bit version later. For the registry the same applies: In order to create registry values in 64bit HKLM for components marked with "msidbComponentAttributes64bit" there is nothing to change in 64-bit installs. For 32-bit components (attrs < 256) the registry stuff must go to Wow6432Node. Ideally Wine's msi should also support something like "msidbComponentAttributesDisableRegistryReflection" to allow 32-bit components to write into 64bit HKLM but that's not required for now. 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 24189] New: msi: set "System64Folder" standard property in x64 environment (64bit installer of .NET Framework 2.0)
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

http://bugs.winehq.org/show_bug.cgi?id=24189 Summary: msi: set "System64Folder" standard property in x64 environment (64bit installer of .NET Framework 2.0) Product: Wine Version: 1.3.1 Platform: x86-64 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: msi AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Hello, continuation of bug 24187 Check for x64 IE5.01 fails. --- snip ---- 001a:trace:msi:msi_set_property 0x7ffff72baca0 L"SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8" L"C:\\windows\\system32\\" ... 001a:trace:msi:ACTION_AppSearchDr L"SearchForIE501_ENU_A64.3643236F_FC70_11D3_A536_0090278A1BB8" ... 001a:trace:msi:msi_get_property property L"SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8" not found ... 001a:trace:msi:MSI_EvaluateConditionW 1 <- L"( NOT IE501FOUND.3643236F_FC70_11D3_A536_0090278A1BB8 ) AND ( NOT IE501FOUND.3643236F_FC70_11D3_A536_0090278A1BB8 )" ... 001a:trace:msi:ACTION_PerformAction Performing action (L"CA_CheckIE501.3643236F_FC70_11D3_A536_0090278A1BB8") ... 001a:trace:msi:ACTION_CustomAction Handling custom action L"CA_CheckIE501.3643236F_FC70_11D3_A536_0090278A1BB8" (113 (null) L"IE5.01 or higher version is required prior to installing Microsoft .NET Framework") 001a:trace:msidb:MSI_RecordSetStringW 0x7fffec3adb40 0 L"IE5.01 or higher version is required prior to installing Microsoft .NET Framework" ... 001a:err:msi:ITERATE_Actions Execution halted, action L"CA_CheckIE501.3643236F_FC70_11D3_A536_0090278A1BB8" returned 1603 --- snip ---- SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8 property is the culprit --- snip orca --- SearchForIE501_____X86.3643236F_FC70_11D3_A536_0090278A1BB8 [DD_SystemFolder_X86.3643236F_FC70_11D3_A536_0090278A1BB8] 1 SearchForIE501_ENU_X86.3643236F_FC70_11D3_A536_0090278A1BB8 [DD_SystemFolder_X86.3643236F_FC70_11D3_A536_0090278A1BB8] 1 SearchForIE501_____A64.3643236F_FC70_11D3_A536_0090278A1BB8 [SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8] 1 SearchForIE501_ENU_A64.3643236F_FC70_11D3_A536_0090278A1BB8 [SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8] 1 --- snip orca --- Although the property is initially set: --- snip --- 001a:trace:msi:msi_get_property returning L"C:\\windows\\system32\\" for property L"SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8" --- snip --- It is later reset due to custom action: --- snip --- 001a:trace:msi:ACTION_CustomAction Handling custom action L"CA_SetSystem64Folder.3643236F_FC70_11D3_A536_0090278A1BB8" (33 L"SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8" L"[System64Folder]") 001a:trace:msidb:MSI_CreateRecord 1 001a:trace:msidb:MSI_RecordSetStringW 0x7fffec3b3010 0 L"[System64Folder]" 001a:trace:msi:MSI_FormatRecordW 0x7fffe3c06490 0x7fffec3b3010 (nil) 0x7fffecf4ddfc 001a:trace:msidb:MSI_RecordIsNull 0x7fffec3b3010 0 001a:trace:msidb:MSI_RecordGetStringW 0x7fffec3b3010 0 (nil) 0x7fffecf4dcbc 001a:trace:msidb:MSI_RecordGetStringW 0x7fffec3b3010 0 0x7fffec3ad750 0x7fffecf4dcbc 001a:trace:msi:MSI_FormatRecordW (L"[System64Folder]") 001a:trace:msidb:MSI_CreateRecord 1 001a:trace:msidb:MSI_RecordSetStringW 0x7fffec3ad040 1 L"System64Folder" 001a:trace:msi:MSI_DatabaseOpenViewW L"SELECT `Value` FROM `_Property` WHERE `_Property`=?" 0x7fffecf4dc38 001a:trace:msidb:TABLE_CreateView 0x7ffff72da890 L"_Property" 0x7fffecf4db68 001a:trace:msidb:TABLE_CreateView table 0x7fffe3c068e0 found with 2 columns 001a:trace:msidb:TABLE_CreateView L"_Property" one row is 4 bytes 001a:trace:msidb:WHERE_CreateView 0x7fffec3b3400 ... 001a:trace:msi:msi_get_property property L"System64Folder" not found 001a:trace:msi:msiobj_release object 0x7fffec3b3010 destroyed 001a:trace:msi:msi_set_property 0x7ffff72da890 L"SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8" L"" 001a:trace:msidb:MSI_CreateRecord 1 001a:trace:msidb:MSI_RecordSetStringW 0x7fffec3b3010 1 L"SystemFolder.3643236F_FC70_11D3_A536_0090278A1BB8" --- snip --- The standard property "System64Folder" must be set in x64 environments. Adding this in dlls/msi/package.c:set_installer_properties() with proper value will succeed the check and get the installer further 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 22805] New: Can't install 64bit .NET 2.0
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

http://bugs.winehq.org/show_bug.cgi?id=22805 Summary: Can't install 64bit .NET 2.0 Product: Wine Version: 1.1.44 Platform: x86-64 URL: http://www.microsoft.com/downloads/details.aspx?family id=b44a0000-acf8-4fa1-affb-40e78d788b00&displaylang=en OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: Paul.Vriens.Wine(a)gmail.com I'm trying to install the 64bit version of .NET 2.0 but this fails. Console shows: wine: Invalid address A +module trace shows: warn:module:map_image Need to relocate module from 0x400000 to 0x110000, but there are no relocation records warn:module:load_dll Failed to load module L"C:\\users\\paul\\Temp\\IXP000.TMP\\install.exe"; status=c0000018 -- 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 24187] New: msi: set "MsiAMD64" and "Msix64" standard properties in x64 environment (64bit installer of .NET Framework 2.0)
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

http://bugs.winehq.org/show_bug.cgi?id=24187 Summary: msi: set "MsiAMD64" and "Msix64" standard properties in x64 environment (64bit installer of .NET Framework 2.0) Product: Wine Version: 1.3.1 Platform: x86-64 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: msi AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Hello, this is a split-off from http://bugs.winehq.org/show_bug.cgi?id=22805#c6 The 64bit version of .NET Framework 2.0 fails due to launch condition not met. --- snip --- ... 001b:trace:msidb:TABLE_get_column_info 0x7fffec3aba00 1 (nil) 0x7fffecf4de50 001b:trace:msidb:MSI_RecordSetStringW 0x7fffe3c28c20 1 L"MsiAMD64" 001b:trace:msidb:TABLE_get_column_info 0x7fffec3aba00 2 (nil) 0x7fffecf4de50 001b:trace:msidb:MSI_RecordSetStringW 0x7fffe3c28c20 2 L"This setup is only supported on AMD64 platforms" ... 001b:trace:msi:msi_get_property property L"MsiAMD64" not found 001b:trace:msi:MSI_EvaluateConditionW 0 <- L"MsiAMD64" 001b:trace:msi:msiobj_release object 0x7fffe3c28c20 destroyed 001b:trace:msi:MSI_ViewClose 0x7fffec3ab130 001b:trace:msidb:TABLE_close 0x7fffec3aba00 001b:trace:msidb:TABLE_delete 0x7fffec3aba00 001b:trace:msi:msiobj_release object 0x7fffec3ab130 destroyed 001b:trace:msidb:MSI_CreateRecord 1 001b:trace:msidb:MSI_RecordSetStringW 0x7fffec3ab130 1 L"Action ended 17:16:07: LaunchConditions. Return value 1603." --- snip --- You have to add standard properties that indicate 64-bit environment. Current, only the "Intel" (32 bit) standard property is handled. --- snip dlls/msi/package.c --- static VOID set_installer_properties(MSIPACKAGE *package) { ... static const WCHAR szIntFormat[] = {'%','d',0}; static const WCHAR szIntel[] = { 'I','n','t','e','l',0 }; ... GetSystemInfo( &sys_info ); if (sys_info.u.s.wProcessorArchitecture == PROCESSOR_ARCHITECTURE_INTEL) { sprintfW( bufstr, szIntFormat, sys_info.wProcessorLevel ); msi_set_property( package->db, szIntel, bufstr ); } ... --- snip dlls/msi/package.c --- Just add handling for "MsiAMD64" (introduced with Windows Installer 3.0, still present for backward compatibility) and "Msix64" (introduced with with Windows Installer 3.1). The values seem not relevant, it's only important that the properties exist. You can reuse the sys_info values from previous GetSystemInfo() call for "Intel" property to also test for AMD/64 bit case. In case of x64 system set both, "MsiAMD64" and "Msix64" properties. With that in place, this launch condition succeeds only to run into next bug ;-) 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 47255] New: Xenonauts crashes in wined3d
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

https://bugs.winehq.org/show_bug.cgi?id=47255 Bug ID: 47255 Summary: Xenonauts crashes in wined3d Product: Wine Version: unspecified Hardware: x86 OS: Linux Status: NEW Keywords: regression Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: gyebro69(a)gmail.com CC: hverbeet(a)gmail.com Regression SHA1: db201072655946662c041a66ee434c30c245e5b0 Distribution: --- Created attachment 64541 --> https://bugs.winehq.org/attachment.cgi?id=64541 terminal output Xenonauts v1.65 crashes immediately after launch. Reverting commit db201072655946662c041a66ee434c30c245e5b0 fixes the crash. No demo version is available. Let me know what debug log would be of help. wine-4.8-343-g27bf52d12c OpenGL vendor string: NVIDIA Corporation OpenGL renderer string: GeForce GT 730/PCIe/SSE2 OpenGL core profile version string: 4.6.0 NVIDIA 418.52.07 -- 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 24613] New: Aura 2 (Sacred Rings): inventory not shown when pressing RMB
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

http://bugs.winehq.org/show_bug.cgi?id=24613 Summary: Aura 2 (Sacred Rings): inventory not shown when pressing RMB Product: Wine Version: 1.3.4 Platform: x86 URL: http://www.fileplanet.com/175732/170000/fileinfo/Aura- 2:-The-Sacred-Rings-Demo OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: gyebro69(a)gmail.com CC: chris.kcat(a)gmail.com Created an attachment (id=31088) --> (http://bugs.winehq.org/attachment.cgi?id=31088) terminal output In Aura 2 (The Sacred Rings) when you press the right-mouse button the inventory should show up at the bottom of the screen, showing your items, the journal etc. Normally, when you open your inventory the mouse pointer should be confined to this lower area of the screen where you can select items etc. When you press RMB in Wine nothing happens in the game: the mouse pointer stays on the main screen, a black bar is shown below but it contains nothing. The game worked without the issue in Wine-0.9.45. Bisecting between 0.9.45 and 0.9.46 resulted: d9571c9e6fa0b8f255815c5128c2859348ea70e6 is the first bad commit commit d9571c9e6fa0b8f255815c5128c2859348ea70e6 Author: Chris Robinson <chris.kcat(a)gmail.com> Date: Sat Sep 15 13:02:32 2007 -0700 wgl: Store the fbconfig id with the window when a pixel format is set. :040000 040000 41741ff092257972095d557196b306aa7e5f7a43 3692bf0336451ec52733c88a7c30ced04fb60179 M dlls The patch cannot be reverted cleanly, however, after git checkout d9571c9e6fa0b8f255815c5128c2859348ea70e6 : the problem exists and git revert d9571c9e6fa0b8f255815c5128c2859348ea70e6 : restores the good state. Link to the demo added to URL (538 MB) Author of the patch added to CC. Fedora 13 Nvidia 7600 / driver 256.53 -- 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 27172] New: Some msvcrt string functions incorrectly return status from 'safe' versions (Corel Draw 12 demo installer fails)
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

http://bugs.winehq.org/show_bug.cgi?id=27172 Summary: Some msvcrt string functions incorrectly return status from 'safe' versions (Corel Draw 12 demo installer fails) Product: Wine Version: 1.3.20 Platform: x86 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: msvcrt AssignedTo: wine-bugs(a)winehq.org ReportedBy: focht(a)gmx.net Hello, while revisiting bug 4971 I noticed a potential regression ... --- snip --- 005c:Ret PE DLL (proc=0x110c8ac,module=0x1100000 L"MSICrlPCU.dll",reason=PROCESS_ATTACH,res=(nil)) retval=1 005c:Ret KERNEL32.LoadLibraryA() retval=01100000 ret=010c2f22 005c:Call KERNEL32.GetProcAddress(01100000,010dbb48 "AcquireKey") ret=010c2f34 005c:Ret KERNEL32.GetProcAddress() retval=01101d10 ret=010c2f34 ... 005c:Call KERNEL32.CreateFileW(01121ba0 L"C:\\users\\focht\\Temp\\Corel\\84ACAA88-5FD7-411D-81F6-24407CE7C94A\\Programs\\DR12.dta",80000000,00000001,00000000,00000003,00000080,00000000) ret=01102ad7 005c:Ret KERNEL32.CreateFileW() retval=0000001c ret=01102ad7 005c:Call KERNEL32.GetFileSize(0000001c,00000000) ret=01102b32 005c:Ret KERNEL32.GetFileSize() retval=00000068 ret=01102b32 005c:Call KERNEL32.ReadFile(0000001c,009add08,00000068,009add04,00000000) ret=01102b55 005c:Ret KERNEL32.ReadFile() retval=00000001 ret=01102b55 005c:Call KERNEL32.CloseHandle(0000001c) ret=01102b5c 005c:Ret KERNEL32.CloseHandle() retval=00000001 ret=01102b5c 005c:Call msvcrt._strupr(009add0c "DR12WTX-1740841-KCM") ret=01102b8c 005c:Ret msvcrt._strupr() retval=00000000 ret=01102b8c 005c:trace:seh:raise_exception code=c0000005 flags=0 addr=0x1102b95 ip=01102b95 tid=005c 005c:trace:seh:raise_exception info[0]=00000000 005c:trace:seh:raise_exception info[1]=00000000 005c:trace:seh:raise_exception eax=00000000 ebx=0000000a ecx=00000000 edx=01121eb8 esi=01121ba0 edi=0000001c 005c:trace:seh:raise_exception ebp=7b820890 esp=009adcdc cs=0073 ds=007b es=007b fs=0033 gs=003b flags=00010202 005c:trace:seh:call_vectored_handlers calling handler at 0x68e827aa code=c0000005 flags=0 005c:trace:seh:call_vectored_handlers handler at 0x68e827aa returned 0 --- snip --- App code: --- snip --- 00DC2B66 8D4424 54 LEA EAX,[ESP+54] 00DC2B6A 8D96 14020000 LEA EDX,[ESI+214] 00DC2B70 8BC8 MOV ECX,EAX 00DC2B72 83C4 04 ADD ESP,4 00DC2B75 2BD1 SUB EDX,ECX 00DC2B77 8A08 MOV CL,BYTE PTR DS:[EAX] 00DC2B79 880C02 MOV BYTE PTR DS:[EAX+EDX],CL 00DC2B7C 40 INC EAX 00DC2B7D 84C9 TEST CL,CL 00DC2B7F ^ 75 F6 JNE SHORT 00DC2B77 00DC2B81 8D5424 30 LEA EDX,[ESP+30] 00DC2B85 52 PUSH EDX 00DC2B86 FF15 7487DD00 CALL DWORD PTR DS:[<&MSVCR71._strupr>] 00DC2B8C 83C4 04 ADD ESP,4 00DC2B8F 8D96 18030000 LEA EDX,[ESI+318] 00DC2B95 8A08 MOV CL,BYTE PTR DS:[EAX] 00DC2B97 40 INC EAX 00DC2B98 880A MOV BYTE PTR DS:[EDX],CL 00DC2B9A 42 INC EDX 00DC2B9B 84C9 TEST CL,CL 00DC2B9D ^ 75 F6 JNE SHORT 00DC2B95 --- snip --- _strupr() is cdecl, return ptr value in EAX --- snip dlls/msvcrt/string.c --- int CDECL _strupr(char *str) { return _strupr_s_l(str, -1, NULL); } --- snip dlls/msvcrt/string.c --- Because Wine returns/takes the status from 's' version instead of string ptr, the app crashes. MSDN: http://msdn.microsoft.com/en-us/library/sch3dy08.aspx --- snip --- Returns a pointer to the altered string. Because the modification is done in place, the pointer returned is the same as the pointer passed as the input argument. No return value is reserved to indicate an error. --- snip --- You might want to check others too (_strlwr ...). BTW .. the serial is from trial, no harm ;-) 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 12

[Bug 47299] New: LAV Filters codec crashes when set to DXVA2 (copy-back) HW acceleration.
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

https://bugs.winehq.org/show_bug.cgi?id=47299 Bug ID: 47299 Summary: LAV Filters codec crashes when set to DXVA2 (copy-back) HW acceleration. Product: Wine Version: 4.9 Hardware: x86 URL: https://github.com/Nevcairiel/LAVFilters/releases/down load/0.74.1/LAVFilters-0.74.1-x86.zip OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: gabrielopcode(a)gmail.com Distribution: --- Created attachment 64600 --> https://bugs.winehq.org/attachment.cgi?id=64600 Avoid a NULL dereference by returning early Grab the 32-bit version in a 32-bit wineprefix to keep it simple (link supplied for latest version but it's not required). Extract it to C:\lav on a new 32-bit prefix. Then install with: wine cmd /c 'cd /D C:\lav && install_video.bat' Open the config with: wine rundll32.exe 'C:\lav\LAVVideo.ax,OpenConfiguration' On the 'Hardware Acceleration' section, change the 'Hardware Decoder to use' to 'DXVA2 (copy-back)' and the crash should happen. Here's the relevant snippet with +d3d (notice the errors, since I assume we don't have DXVA2 support): --- snip --- 0029:trace:d3d:wined3d_cs_run Executing WINED3D_CS_OP_CALLBACK. 0029:trace:d3d:wined3d_cpu_blitter_create Created blitter 0x1708f0. 0029:trace:d3d:wined3d_ffp_blitter_create Created blitter 0x170908. 0029:trace:d3d:wined3d_fbo_blitter_create Created blitter 0x170948. 0029:trace:d3d:wined3d_raw_blitter_create Created blitter 0x170960. 0029:trace:d3d:context_acquire device 0x1a6898, texture 0x19da60, sub_resource_idx 0. 0029:trace:d3d:context_acquire Rendering onscreen. 0029:trace:d3d:swapchain_create_context Creating a new context for swapchain 0x1c1478, thread 41. 0029:trace:d3d:context_create swapchain 0x1c1478. 0029:trace:d3d:adapter_gl_create_context swapchain 0x1c1478, context 0x1c6fd9c. 0029:trace:d3d:wined3d_context_gl_init context_gl 0x1c3978, swapchain 0x1c1478. 0029:trace:d3d:context_choose_pixel_format device 0x1a6898, dc 0x120052, color_format WINED3DFMT_B8G8R8A8_UNORM, ds_format WINED3DFMT_UNKNOWN, aux_buffers 0. 0029:trace:d3d:context_choose_pixel_format Found iPixelFormat=27 for ColorFormat=WINED3DFMT_B8G8R8A8_UNORM, DepthStencilFormat=WINED3DFMT_UNKNOWN. 0029:trace:d3d:wined3d_context_gl_enter Entering context 0x1c3978, level 1. 0029:err:d3d:wined3d_context_gl_init Failed to set pixel format 27 on device context 0x120052. 0029:trace:d3d:context_release Releasing context 0x1c3978, level 1. 0029:warn:d3d:context_release Context 0x1c3978 is not the current context. 0029:warn:d3d:adapter_gl_create_context Failed to initialise context. 0029:err:d3d:swapchain_create_context Failed to create a new context for the swapchain --- snip --- The attached patch fixes this problem by avoiding the crash. Obviously it doesn't fix the root cause, so I don't know if it's a good practice to apply it but it's an idea. Note that this crash happens even in media players that use DirectShow (e.g. Media Player Classic) when opening files with codecs that support DXVA2 decoding, if you set the registry key manually. If you want to test that, import: REGEDIT4 [HKEY_CURRENT_USER\Software\LAV\Video\HWAccel] "HWAccel"=dword:00000003 in the wineprefix, to set it to DXVA2 (copy-back), and then load such a video, it should crash. The patch here will also fix loading such videos and in fact they will work just fine, though obviously without any hardware acceleration since we don't have it. I don't think crashing on missing features is desireable, though. I'm not familiar with wined3d code at all so there's probably a better fix for this. -- 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 5

[Bug 35868] New: DXVA Checker 3.0.x (.NET 2.0 app) needs unimplemented function dxva2.dll.DXVA2CreateVideoService
by wine-bugs＠winehq.org 09 Jul '21

09 Jul '21

https://bugs.winehq.org/show_bug.cgi?id=35868 Bug ID: 35868 Summary: DXVA Checker 3.0.x (.NET 2.0 app) needs unimplemented function dxva2.dll.DXVA2CreateVideoService Product: Wine Version: 1.7.15 Hardware: x86 OS: Linux Status: NEW Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: focht(a)gmx.net Hello folks, since there is now a stub dll I looked around for some real world users :) 'DXVA Checker' Prerequisite: 'winetricks -q dotnet20' --- snip --- $ wine ./DXVAChecker.exe ... fixme:quartz:VMR9_maybe_init Reduce ratio to least common denominator fixme:quartz:VMR9SurfaceAllocatorNotify_SetD3DDevice (0x1c90d0/0x1c8ec0)->(...) semi-stub fixme:d3d:resource_check_usage Unhandled usage flags 0x8. fixme:quartz:VMR9SurfaceAllocatorNotify_AllocateSurfaceHelper (0x1c90d0/0x1c8ec0)->(0x33dcec, 0x33dce8 => 2, 0x1dbae0) semi-stub fixme:quartz:VMR7FilterConfig_SetNumberOfStreams (0x1c90bc/0x1c8ec0)->(1) stub fixme:quartz:VMR9Inner_QueryInterface No interface for {bb057577-0db8-4e6a-87a7-1a8c9a505a0f} fixme:strmbase:BaseInputPinImpl_QueryInterface No interface for {256a6a22-fbad-11d1-82bf-00a0c9696c8f}! wine: Call from 0x7b83abc3 to unimplemented function dxva2.dll.DXVA2CreateVideoService, aborting --- snip --- MSDN: http://msdn.microsoft.com/en-us/library/windows/desktop/ms704721%28v=vs.85%… $ sha1sum DXVAChecker32_3.0.1.zip e7ba2151ae8cc2b9f805cbd1dd046c31f0d45a5b DXVAChecker32_3.0.1.zip $ du -sh DXVAChecker32_3.0.1.zip 564K DXVAChecker32_3.0.1.zip $ wine --version wine-1.7.15-67-g99c151a Regards -- 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 11