October 2023 - Wine-bugs - list.winehq.org

[Bug 50728] New: % CPU Utilization monitor/gauge/indicator not functioning in VARA.exe (ham radio software)
by WineHQ Bugzilla 29 Jul '24

29 Jul '24

https://bugs.winehq.org/show_bug.cgi?id=50728 Bug ID: 50728 Summary: % CPU Utilization monitor/gauge/indicator not functioning in VARA.exe (ham radio software) Product: Wine Version: 6.2 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: trivial Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: eric.wheez(a)gmail.com Distribution: --- Created attachment 69485 --> https://bugs.winehq.org/attachment.cgi?id=69485 Terminal output from running VARA (seeing CPU monitor bug) and then closing it. When I run a ham radio program, called VARA, through Wine I'm noticing that the program doesn't properly display CPU usage in its "CPU Gauge". It instead just says "%CPU not available". (See this thread for an image: https://groups.io/g/VARA-MODEM/message/1311) I'm assuming it might be an unimplemented function or missing library in Wine, but I don't know how to track down what is missing. This is with a fresh installation of wine-staging 6.2 and a fresh wineprefix. The bug doesn't occur when VARA is run on a Windows computer. I've attached terminal output of the program running (and then being closed). I'm wondering if this may be similar to bug 26785, though that one was fixed in 2016. I'm wondering if the bug might be related to one of these warnings: 0024:fixme:advapi:RegisterEventSourceW ((null),L"PDH"): stub 0024:fixme:ntdll:NtQuerySystemInformation info_class SYSTEM_PERFORMANCE_INFORMATION 0024:fixme:winspool:PerfOpen (null): stub 0024:fixme:winspool:PerfCollect L"Global", 0x31ea48, 0x31ea3c, 0x31ea40: stub 0024:fixme:winspool:PerfClose stub 0024:err:seh:dispatch_exception Fatal EXCEPTION_FLT_INEXACT_RESULT exception (code=c000008f) raised 0024:err:seh:dispatch_exception Fatal EXCEPTION_FLT_INEXACT_RESULT exception (code=c000008f) raised 00e4:fixme:service:QueryServiceConfig2W Level 6 not implemented 00e4:fixme:service:QueryServiceConfig2W Level 6 not implemented 0024:err:ole:TLB_ReadTypeLib Loading of typelib L"msdatsrc.tlb" failed with error 2 I'm very new to debugging though and assume that this will take more than just dropping in dll's and doing winecfg dll overrides (especially if ntdll is at fault here). Does anybody have any insight into Wine's ability to query system resources - or some ideas for commands I could try to track this bug down and report it to the Wine team for possible request for implementation? There's a whole community of hams who would love to have this program 100% operational (even though this is a minor bug). To reproduce this behavior, you can install VARA with these commands: #### Configure our wineprefix winetricks -q corefonts dotnet35sp1 vb6run win7 sound=alsa vcrun2015 pdh # run pdh here just for the winecfg dll override # Install an older pdh.dll (the pdh.dll from "winetricks pdh" is too new for VARA) sudo apt-get install zip -y cd ~/Downloads && mkdir pdhNT40 && cd pdhNT40 wget http://download.microsoft.com/download/winntsrv40/update/5.0.2195.2668/nt4/… unzip -o nt4pdhdll.exe cp pdh.dll ~/.wine/drive_c/windows/system32 #### Download/extract/install VARA HF (or newer) [https://rosmodem.wordpress.com/] sudo apt-get install p7zip-full -y sudo apt-get install megatools -y cd ~/Downloads VARALINK=$(curl -s https://rosmodem.wordpress.com/ | grep -oP '(?<=<a href=").*?(?=" target="_blank" rel="noopener noreferrer">VARA HF v)') # Find the mega.nz link from the rosmodem website no matter its version, then store it as a variable megadl ${VARALINK} 7z x VARA*.zip -o"VARAInstaller" -y wine ~/Downloads/VARAInstaller/VARA\ setup*.exe /SILENT #### Configure VARA # In winecfg, go to the Audio tab to set up your default in/out soundcards wine winecfg # In VARA, set up your soundcard input and output (go to Settings ... Soundcard) wine ~/.wine/drive_c/VARA/VARA.exe -- 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 54788] New: AOL 5.0 Installation fails
by WineHQ Bugzilla 28 Jul '24

28 Jul '24

https://bugs.winehq.org/show_bug.cgi?id=54788 Bug ID: 54788 Summary: AOL 5.0 Installation fails Product: Wine Version: 8.5 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: maiktapwagner(a)aol.com Distribution: --- Created attachment 74281 --> https://bugs.winehq.org/attachment.cgi?id=74281 Backtrace (wine 8.5 on Slackware64/Multilib) Download is at: http://www.oldversion.com/windows/download/america-online-5-0 bash-5.1$ sha1sum aol50.exe cacd3d5b93e7f2d6004812707148364a47db6db6 aol50.exe Backtrace is attached. Console output to follow shortly. -- 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 8

[Bug 50231] New: Ys: Origin shows black screen during video playback
by WineHQ Bugzilla 28 Jul '24

28 Jul '24

https://bugs.winehq.org/show_bug.cgi?id=50231 Bug ID: 50231 Summary: Ys: Origin shows black screen during video playback Product: Wine Version: 5.22 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: kq3thih(a)mailnesia.com Distribution: Debian Created attachment 68738 --> https://bugs.winehq.org/attachment.cgi?id=68738 +quartz,+strmbase,+gstreamer,+qedit,+devenum log The videos are showing only a black screen during playback. Audio works. Disabling winegstreamer and using native quartz works around it for proper playback. Nothing seems to be all that interesting in the default terminal output: --- snip --- 00d0:fixme:qedit:SampleGrabber_IMemInputPin_GetAllocatorRequirements (01101130)->(0031E6F0): semi-stub 00d0:err:quartz:MediaFilter_GetState Filter 01101D90 reported incorrect state 2 (expected 1). 00d0:err:quartz:MediaFilter_GetState Filter 01101130 reported incorrect state 2 (expected 1). 00d0:err:quartz:MediaFilter_GetState Filter 0EA5F9C0 reported incorrect state 2 (expected 1). 00d0:err:quartz:MediaFilter_GetState Filter 01101608 reported incorrect state 2 (expected 1). 0120:fixme:d3d:state_linepattern_w Setting line patterns is not supported in OpenGL core contexts. --- snip --- Attached is a +quartz,+strmbase,+gstreamer,+qedit,+devenum log. -- 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 9

[Bug 54735] New: AOL (America Online) Desktop Beta fails when installing .net 2.8
by WineHQ Bugzilla 28 Jul '24

28 Jul '24

https://bugs.winehq.org/show_bug.cgi?id=54735 Bug ID: 54735 Summary: AOL (America Online) Desktop Beta fails when installing .net 2.8 Product: Wine Version: 8.4 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: maiktapwagner(a)aol.com Distribution: --- Created attachment 74239 --> https://bugs.winehq.org/attachment.cgi?id=74239 Backtrace (wine 8.4 on Slackware64/Multilib) https://beta.aol.com/projects/desktop/windows sha256sum Install_AOL_Desktop.exe 13cd19996e183e927ce0710f367d586474dcca308f593e2551b67334a0315b4d Install_AOL_Desktop.exe Installer starts up okay but eventually crashes when installing .net 4.8 which is required for AOL Desktop Gold Beta. -- 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 8

[Bug 48167] New: 1000 Mots V4.0.2 freeze when 3 words are pronounced - Underrun of data
by WineHQ Bugzilla 28 Jul '24

28 Jul '24

https://bugs.winehq.org/show_bug.cgi?id=48167 Bug ID: 48167 Summary: 1000 Mots V4.0.2 freeze when 3 words are pronounced - Underrun of data Product: Wine Version: 4.20 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: quartz Assignee: wine-bugs(a)winehq.org Reporter: sevmek(a)free.fr Distribution: --- Created attachment 65798 --> https://bugs.winehq.org/attachment.cgi?id=65798 wine 4.20 staging output The software used is "1000 mots pour apprendre à lire" a french software (shareware) for educational purpose (the title translate as "1000 words to learn how to read"). One can download the software here: https://app.educampa.fr/mil4_setup.exe Version is 4.0.2 SHA1: 3213166282a8dc0fad236952f2dcb0650db199b3 mil4_setup.exe It was tested against wine 4.20 (staging) on Fedora 31 and tested against wine 3.0 on ubuntu 18.04. To install it require winetricks and gstreamer1-plugins-good.i686 (in fedora) or gstreamer1.0-plugins-good:i386 (in ubuntu) To install this software I use (in a clean and fresh ~/.wine directory): $ env WINEARCH=win32 winetricks vb6run $ wget https://app.educampa.fr/mil4_setup.exe $ wine mil4_setup.exe then I run using: $ wine "c:\\educampa\\1000Mots_V4\\1000Mots_V4.exe" Installation is OK. when I use the software there is one activity that repeatedly freeze the software whith an error like this: 00db:fixme:quartz:DSoundRender_UpdatePositions Underrun of data occurred! (see attachment for full log) In this activity, the software pronunce 3 following words. I suspect the issue is there. -- 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 8

[Bug 49944] New: Tom Clancy's Splinter Cell: Conviction fails to detect audio device
by WineHQ Bugzilla 28 Jul '24

28 Jul '24

https://bugs.winehq.org/show_bug.cgi?id=49944 Bug ID: 49944 Summary: Tom Clancy's Splinter Cell: Conviction fails to detect audio device Product: Wine Version: 5.18 Hardware: x86-64 OS: Linux Status: NEW Severity: minor Priority: P2 Component: wmi&wbemprox Assignee: wine-bugs(a)winehq.org Reporter: andrey.goosev(a)gmail.com Distribution: --- Created attachment 68335 --> https://bugs.winehq.org/attachment.cgi?id=68335 wbemprox Stops launching with 'No audio device detected' message. Workaround is rename systemdetection.dll in 'src/system' or disable wbemprox. wine-5.18-161-gcce4f36e21 -- 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 6

[Bug 32679] New: cmd.exe Add support for || and &&
by wine-bugs＠winehq.org 28 Jul '24

28 Jul '24

http://bugs.winehq.org/show_bug.cgi?id=32679 Bug #: 32679 Summary: cmd.exe Add support for || and && Product: Wine Version: 1.5.21 Platform: x86 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: cmd AssignedTo: wine-bugs(a)winehq.org ReportedBy: us(a)edmeades.me.uk Classification: Unclassified Add support to cmd.exe for the && and || functionality - They are currently parsed and handled ok, but not correctly processed -- 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 34

[Bug 55662] New: Different behaviour of "set" command
by WineHQ Bugzilla 28 Jul '24

28 Jul '24

https://bugs.winehq.org/show_bug.cgi?id=55662 Bug ID: 55662 Summary: Different behaviour of "set" command Product: Wine Version: 8.0.2 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: cmd Assignee: wine-bugs(a)winehq.org Reporter: ch.panel(a)free.fr Distribution: --- behavior of these commands are different in wine and windows in a batch file set test=A set /P test="type A or B (%test%) :" echo "%test%" pause if the user type nothing the result is : "A" on windows "" on wine -- 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

[Bug 55798] New: Unreal Engine 5.2: Wine minidumps take hours to load into a debugger
by WineHQ Bugzilla 28 Jul '24

28 Jul '24

https://bugs.winehq.org/show_bug.cgi?id=55798 Bug ID: 55798 Summary: Unreal Engine 5.2: Wine minidumps take hours to load into a debugger Product: Wine Version: 8.14 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: minor Priority: P2 Component: dbghelp Assignee: wine-bugs(a)winehq.org Reporter: aiden.deloryn(a)tensorworks.com.au Distribution: --- === Overview === When attempting to load a Wine generated minidump from an Unreal Engine application using a debugger (such as Visual Studio or VS Code), it takes a number of hours for the file to load. If left uninterrupted during this time, the process does eventually complete and all of the expected debug data becomes visible within the IDE. This behaviour has only been observed for minidumps of type MiniDumpNormal. When working with a Wine generated minidump of type MiniDumpWithFullMemory, the debugger is able to load the file within a few seconds. === Steps to Reproduce === 1. Build this minimal sample application https://github.com/TensorWorks/BuggyFirstPersonDemo.git on a Windows machine with Unreal Engine 5.2.1 using the following command: C:\<UE_INSTALL_DIR>\Engine\Build\BatchFiles\RunUAT.bat BuildCookRun -utf8output -clientconfig=DebugGame -project="C:\<PATH_TO_PROJECT_DIR>\BuggyFirstPersonDemo.uproject" -noP4 -cook -build -stage -prereqs -pak -archive -archivedirectory="C:\<PATH_TO_PROJECT_DIR>\GameBuilds" -platform=Win64 2. Copy the build directory (C:\<PATH_TO_PROJECT_DIR>\GameBuilds) to a Linux machine and run the application using Wine (DXVK required): wine ./GameBuilds/Windows/BuggyFirstPersonDemo/Binaries/Win64/BuggyFirstPersonDemo-Win64-DebugGame.exe -windowed -ResX=800 -ResY=600 3. Trigger a crash by picking up the gun and shooting the large white cube. Note: hold down the right mouse button to aim/look, and press the left mouse button to shoot. 4. Copy the crash dump from ./GameBuilds/Windows/BuggyFirstPersonDemo/Saved/Crashes/<CRASH_ID>/UEMinidump.dmp to a Windows machine and attempt to load it using Visual Studio or VS Code (for VS Code you will need the C/C++ extension from Microsoft installed). === Actual Results === The minidump takes hours to load into Visual Studio and VS Code. MemoryListStream contains over 950,000 memory ranges (see additional information below). === Expected Results === The minidump takes less than a minute to load into Visual Studio and VS Code, which is true for full memory minidumps generated under Wine and any minidump generated under Windows. === Build Date & Hardware === wine-8.14 on Ubuntu 22.04.3 LTS x86-64 === Additional Builds and Platforms === Also reproducible on wine-8.17 and wine-8.0.2 === Additional Information === When analysing and comparing minidumps generated under Wine versus those generated under Windows using the Microsoft Crash Dump File Checker tool “DumpChk” (https://learn.microsoft.com/en-us/windows-hardware/drivers/debugger/dumpchk) a significant discrepancy can be observed in the normal minidump generated under Wine. In particular, the number of memory ranges in MemoryListStream is substantially higher, as can be seen in the excerpts below. Note that full memory minidumps have also been included from both Windows and Wine for additional context and comparison. Windows - MiniDumpNormal: Stream 3: type MemoryListStream (5), size 00000804, RVA 0006FF09 128 memory ranges [OMITTED] Total memory: 463cc Wine - MiniDumpNormal: Stream 4: type MemoryListStream (5), size 00EA0FF4, RVA 0002986B 958719 memory ranges [OMITTED] Total memory: 15bd62e Windows - MiniDumpWithFullMemory: Stream 5: type Memory64ListStream (9), size 0000EAC0, RVA 000C8D45 3755 memory ranges [OMITTED] Total memory: a74c1000 Wine - MiniDumpWithFullMemory: Stream 7: type Memory64ListStream (9), size 0000E260, RVA 0006ADFB 3621 memory ranges [OMITTED] Total memory: 7ab14000 Note: You can generate a full memory minidump using the provided sample application by specifying the -fullcrashdumpalways flag. For example: wine ./GameBuilds/Windows/BuggyFirstPersonDemo/Binaries/Win64/BuggyFirstPersonDemo-Win64-DebugGame.exe -windowed -ResX=800 -ResY=600 -fullcrashdumpalways -- 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 48455] New: Native Instruments Native Access 1.9 installer hangs during installation of ISO Mounter driver (missing handling of architecture-specific SourceDisks{Names,Files} .inf sections)
by WineHQ Bugzilla 28 Jul '24

28 Jul '24

https://bugs.winehq.org/show_bug.cgi?id=48455 Bug ID: 48455 Summary: Native Instruments Native Access 1.9 installer hangs during installation of ISO Mounter driver (missing handling of architecture-specific SourceDisks{Names,Files} .inf sections) Product: Wine Version: 5.0-rc5 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: setupapi Assignee: wine-bugs(a)winehq.org Reporter: focht(a)gmx.net Distribution: --- Hello folks, as it says. The driver sub-installer just hangs (churning cpu), blocking the main installer forever. Stable download link via Internet Archive: https://web.archive.org/web/20190227215247/https://www.native-instruments.c… --- snip --- $ wine ./Native\ Access\ 1.9.1\ Setup\ PC.exe ... --- snip --- Reduced case by calling driver sub-installer directly, after killing main installer: --- snip --- $ pwd /home/focht/.wine/drive_c/Program Files (x86)/Native Instruments/IsoDriver/x64 $ WINEDEBUG=+seh,+relay,+loaddll,+process,+setupapi wine ./NI-drvinst.exe instroot "root\NIWinCDEmu" "C:\Program Files (x86)\Native Instruments\IsoDriver\NIWinCDEmu.inf" >>log.txt 2>&1 ... 002d:trace:setupapi:SetupDiInstallDriverFiles devinfo 000000000011C1C0, device_data 000000000022F6C0. ... 002d:trace:setupapi:SetupOpenInfFileW L"C:\\Program Files (x86)\\Native Instruments\\IsoDriver\\NIWinCDEmu.inf" -> 000000000011D990 ... 002d:trace:setupapi:SetupGetStringFieldW context 000000000011D990/000000000011D990/9/0 index 1 returning L"Drivers_Dir" ... 002d:trace:setupapi:SetupQueueCopySectionW queue 000000000011DDD0, src_root L"C:\\Program Files (x86)\\Native Instruments\\IsoDriver", hinf 000000000011D990, hlist 000000000011D990, section L"Drivers_Dir", style 0x10000. ... 002d:trace:setupapi:SetupQueueCopyIndirectW root=L"C:\\Program Files (x86)\\Native Instruments\\IsoDriver" path=(null) file=L"NIWinCDEmu.sys" -> dir=L"C:\\windows\\system32\\drivers" file=L"NIWinCDEmu.sys" descr=(null) tag=(null) ... 002d:trace:setupapi:SetupDefaultQueueCallbackW start copy L"C:\\Program Files (x86)\\Native Instruments\\IsoDriver\\NIWinCDEmu.sys" -> L"C:\\windows\\system32\\drivers\\NIWinCDEmu.sys" ... 002d:trace:setupapi:queue_copy_file copying file L"C:\\Program Files (x86)\\Native Instruments\\IsoDriver\\NIWinCDEmu.sys" -> L"C:\\windows\\system32\\drivers\\NIWinCDEmu.sys" ... 002d:trace:setupapi:do_file_copyW copy L"C:\\Program Files (x86)\\Native Instruments\\IsoDriver\\NIWinCDEmu.sys" to L"C:\\windows\\system32\\drivers\\NIWinCDEmu.sys" style 0x10000 002d:Call KERNEL32.GetFileAttributesW(006fd540 L"C:\\windows\\system32\\drivers\\NIWinCDEmu.sys") ret=00259e09 ... 002d:Ret KERNEL32.GetFileAttributesW() retval=ffffffff ret=00259e09 002d:Call KERNEL32.CopyFileW(006fd4a0 L"C:\\Program Files (x86)\\Native Instruments\\IsoDriver\\NIWinCDEmu.sys",006fd540 L"C:\\windows\\system32\\drivers\\NIWinCDEmu.sys",00000000) ret=0025a064 .... 002d:Call ntdll.NtCreateFile(0022e100,80100080,0022e128,0022e118,00000000,00000000,00000007,00000001,00000060,00000000,00000000) ret=7b0110a1 002d:Ret ntdll.NtCreateFile() retval=c0000034 ret=7b0110a1 002d:Call ntdll.RtlNtStatusToDosError(c0000034) ret=7b011190 002d:Ret ntdll.RtlNtStatusToDosError() retval=00000002 ret=7b011190 ... 002d:Ret KERNEL32.CopyFileW() retval=00000000 ret=0025a064 002d:warn:setupapi:do_file_copyW failed to copy, err 2 ... <repeats endlessly> ... 002d:trace:setupapi:SetupDefaultQueueCallbackW need media L"C:\\Program Files (x86)\\Native Instruments\\IsoDriver" L"NIWinCDEmu.sys" ... 002d:trace:setupapi:SetupDefaultQueueCallbackW start copy L"C:\\Program Files (x86)\\Native Instruments\\IsoDriver\\NIWinCDEmu.sys" -> L"C:\\windows\\system32\\drivers\\NIWinCDEmu.sys" ... --- snip --- Content of 'NIWinCDEmu.inf': --- snip --- [Version] Signature="$WINDOWS NT$" Class=SCSIAdapter ClassGuid={4d36e97b-e325-11ce-bfc1-08002be10318} Provider=%NI_MFNAME% DriverVer= 08/05/2011, 1.0.0.2 CatalogFile=NIWinCDEmu.cat [DestinationDirs] DefaultDestDir = 12 [SourceDisksNames.x86] 1 = %DiskId1%,,, [SourceDisksNames.amd64] 1 = %DiskId1%,,, [SourceDisksFiles.x86] NIWinCDEmu.sys = 1,\x86 [SourceDisksFiles.amd64] NIWinCDEmu.sys = 1,\x64 [Manufacturer] %NI_MFNAME%=Standard, NTamd64 [Standard] %NIWinCDEmu.DeviceDesc%=NIWinCDEmu_Device, root\NIWinCDEmu [Standard.NTamd64] %NIWinCDEmu.DeviceDesc%=NIWinCDEmu_Device, root\NIWinCDEmu [NIWinCDEmu_Device.NT] CopyFiles=Drivers_Dir [Drivers_Dir] NIWinCDEmu.sys,,,2 ;-------------- Service installation [NIWinCDEmu_Device.NT.Services] AddService = NIWinCDEmu,%SPSVCINST_ASSOCSERVICE%, dev_Service_Inst ; -------------- busenum driver install sections [dev_Service_Inst] DisplayName = %dev.SVCDESC% ServiceType = 1 ; SERVICE_KERNEL_DRIVER StartType = 3 ; SERVICE_DEMAND_START ErrorControl = 1 ; SERVICE_ERROR_NORMAL ServiceBinary = %12%\NIWinCDEmu.sys LoadOrderGroup = Extended Base [Strings] SPSVCINST_ASSOCSERVICE= 0x00000002 NI_MFNAME = "Native Instruments GmbH" DiskId1 = "ISO Mounter Installation Disk #1" NIWinCDEmu.DeviceDesc = "ISO Mounter" dev.SVCDESC = "ISO Mounter driver" --- snip --- The problem here is that Wine's 'get_source_info' doesn't take architecture-specific 'SourceDisksNames' and 'SourceDisksFiles' .inf sections into account. Wine source: https://source.winehq.org/git/wine.git/blob/b3abc343c373ee5fd0764a3ceedeef5… --- snip --- 289 static void get_source_info( HINF hinf, const WCHAR *src_file, SP_FILE_COPY_PARAMS_W *params, 290 WCHAR *src_root, WCHAR *src_path) 291 { 292 static const WCHAR SourceDisksNames[] = 293 {'S','o','u','r','c','e','D','i','s','k','s','N','a','m','e','s',0}; 294 static const WCHAR SourceDisksFiles[] = 295 {'S','o','u','r','c','e','D','i','s','k','s','F','i','l','e','s',0}; 296 297 INFCONTEXT file_ctx, disk_ctx; 298 INT id, diskid; 299 DWORD len; 300 301 /* find the SourceDisksFiles entry */ 302 if (!SetupFindFirstLineW( hinf, SourceDisksFiles, src_file, &file_ctx )) return; 303 if (!SetupGetIntField( &file_ctx, 1, &diskid )) return; 304 305 /* now find the diskid in the SourceDisksNames section */ 306 if (!SetupFindFirstLineW( hinf, SourceDisksNames, NULL, &disk_ctx )) return; 307 for (;;) 308 { 309 if (SetupGetIntField( &disk_ctx, 0, &id ) && (id == diskid)) break; 310 if (!SetupFindNextLine( &disk_ctx, &disk_ctx )) return; 311 } 312 313 if (SetupGetStringFieldW( &disk_ctx, 1, NULL, 0, &len ) && len > sizeof(WCHAR) 314 && (params->SourceDescription = heap_alloc( len * sizeof(WCHAR) ))) 315 SetupGetStringFieldW( &disk_ctx, 1, (WCHAR *)params->SourceDescription, len, NULL ); 316 317 if (SetupGetStringFieldW( &disk_ctx, 2, NULL, 0, &len ) && len > sizeof(WCHAR) 318 && (params->SourceTagfile = heap_alloc( len * sizeof(WCHAR) ))) 319 SetupGetStringFieldW( &disk_ctx, 2, (WCHAR *)params->SourceTagfile, len, NULL ); 320 321 if (SetupGetStringFieldW( &disk_ctx, 4, NULL, 0, &len ) && len > sizeof(WCHAR) 322 && len < MAX_PATH - lstrlenW( src_root ) - 1) 323 { 324 lstrcatW( src_root, backslashW ); 325 SetupGetStringFieldW( &disk_ctx, 4, src_root + lstrlenW( src_root ), 326 MAX_PATH - lstrlenW( src_root ), NULL ); 327 } 328 329 if (SetupGetStringFieldW( &file_ctx, 2, NULL, 0, &len ) && len > sizeof(WCHAR) && len < MAX_PATH) 330 { 331 SetupGetStringFieldW( &file_ctx, 2, src_path, MAX_PATH, NULL ); 332 params->SourcePath = src_path; 333 } 334 } --- snip --- Microsoft docs: https://docs.microsoft.com/en-us/windows-hardware/drivers/install/inf-sourc… --- quote ---- ... To support distribution of driver files on multiple system architectures, you can specify an architecture-specific SourceDisksFiles section by adding an .x86, .ia64, .amd64, .arm, or .arm64 extension to SourceDisksFiles. Be aware that, unlike other sections such as a DDInstall section, the platform extensions for a SourceDisksFiles section are not .ntx86, .ntia64, or .ntamd64. For example, to specify a source disk names section for an x86-based system, use a SourceDisksFiles.x86 section, not a SourceDisksFiles.ntx86 section. Similarly, use a SourceDisksFiles.ia64 section to specify an Itanium-based system and a SourceDisksFiles.amd64 section to specify an x64-based system. During installation, SetupAPI functions look for architecture-specific SourceDisksFiles sections before using the generic section. For example, if, during installation on an x86-based platform, Windows is copying a file that is named driver.sys, it will look for the file's description in [SourceDisksFiles.x86] before looking in [SourceDisksFiles]. --- quote ---- In contrast, 'get_source_id' does it properly: https://source.winehq.org/git/wine.git/blob/f1b94dc16c35a5c477b6df8aa94c6d3… --- snip --- 372 static LPWSTR get_source_id( HINF hinf, PINFCONTEXT context, PCWSTR filename ) 373 { 374 DWORD size; 375 LPWSTR source_id; 376 377 if (!SetupFindFirstLineW( hinf, source_disks_files_platform, filename, context ) && 378 !SetupFindFirstLineW( hinf, source_disks_files, filename, context )) 379 return NULL; 380 381 if (!SetupGetStringFieldW( context, 1, NULL, 0, &size )) 382 return NULL; 383 384 if (!(source_id = HeapAlloc( GetProcessHeap(), 0, size * sizeof(WCHAR) ))) 385 return NULL; 386 387 if (!SetupGetStringFieldW( context, 1, source_id, size, NULL )) 388 { 389 HeapFree( GetProcessHeap(), 0, source_id ); 390 return NULL; 391 } 392 393 if (!SetupFindFirstLineW( hinf, source_disks_names_platform, source_id, context ) && 394 !SetupFindFirstLineW( hinf, source_disks_names, source_id, context )) 395 { 396 HeapFree( GetProcessHeap(), 0, source_id ); 397 return NULL; 398 } 399 return source_id; 400 } --- snip --- $ sha1sum Native* 374f62fb373ceb02527b8a179c7e305268bab25c Native Access 1.9.1 Setup PC.exe 5a78d5850267d0ef6701bc5da7dc25e5b62b6031 Native_Access_Installer.zip $ du -sh Native* 70M Native Access 1.9.1 Setup PC.exe 69M Native_Access_Installer.zip $ wine --version wine-5.0-rc5 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.

1 12