- Wine-GitLab - list.winehq.org

[PATCH 0/2] MR9704: winmm, msvfw32: DGV window shenanigans with audio only files
by Bernhard Kölbl Dec. 5, 2025

Dec. 5, 2025

Basically, check if we can set a HWND in the driver, before doing so. Right now, msvfw32 + mciqtz32 error out on audio only playbacks, breaking the game Transcendence. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9704

1 2

[PATCH 0/1] MR9703: msxml/tests: Add some tests for supported interfaces in the SAX API.
by Nikolay Sivov (＠nsivov) Dec. 5, 2025

Dec. 5, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9703

2 1

[PATCH v2 0/1] MR9698: msado15: Request more features when opening table directly.
by Piotr Caban (＠piotr) Dec. 5, 2025

Dec. 5, 2025

-- v2: msado15: Request more features when opening table directly. https://gitlab.winehq.org/wine/wine/-/merge_requests/9698

2 1

Re: [PATCH v4 0/0] MR9660: jscript: Treat DISPATCH_PROPERTYGET | DISPATCH_METHOD as property getter in ES5+ modes.
by Gabriel Ivăncescu Dec. 5, 2025

Dec. 5, 2025

Good catch. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9660#note_124953

1 0

Re: [PATCH v3 0/1] MR8701: devenum: Add DevicePath field for a video input device.
by Bernhard Übelacker Dec. 5, 2025

Dec. 5, 2025

On Fri Dec 5 16:30:27 2025 +0000, Bernhard Übelacker wrote: > Hello @chenhaoyang, this MR seems to have introduced a crash in wine and > windows, therefore the test seems to be wrong. > Depends this test on specific hardware to be available? > [Test pattern page for devenum](https://test.winehq.org/data/patterns.html#devenum:devenum). There is a similar call in [dlls/qcap/tests/videocapture.c](https://gitlab.winehq.org/wine/wine/-/blob/…, which skips the test if `ICreateDevEnum_CreateClassEnumerator` returns false. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8701#note_124948

1 0

[PATCH v2 0/2] MR9660: jscript: Treat DISPATCH_PROPERTYGET | DISPATCH_METHOD as property getter in ES5+ modes.
by Gabriel Ivăncescu Dec. 5, 2025

Dec. 5, 2025

-- v2: mshtml: Handle DISPATCH_PROPERTYGET | DISPATCH_METHOD in dispex_value as jscript: Treat DISPATCH_PROPERTYGET | DISPATCH_METHOD as property getter https://gitlab.winehq.org/wine/wine/-/merge_requests/9660

2 8

Re: [PATCH v3 0/1] MR8701: devenum: Add DevicePath field for a video input device.
by Bernhard Übelacker Dec. 5, 2025

Dec. 5, 2025

Hello @chenhaoyang, this MR seems to have introduced a crash in wine and windows, therefore the test seems to be wrong. [Test pattern page for devenum](https://test.winehq.org/data/patterns.html#devenum:devenum). -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8701#note_124943

1 0

Re: [PATCH v2 0/1] MR9671: include: Add bti jc for __ASM_DEFINE_FUNC on ARM64
by Alexandre Julliard (＠julliard) Dec. 5, 2025

Dec. 5, 2025

On Fri Dec 5 16:17:46 2025 +0000, Alfred Agrell wrote: > Isn't this stuff needed for every function pointer target, including > every cross-dll operation (dllexport, callbacks, COM, etc)? Which is (to > my knowledge) most of our asm functions. > Another option would be find some Linux syscall to disable the check for > branch target instructions. We most likely need that to deal with > user-provided ARM-native exe files, anyways. > (Disclaimer: I haven't tested any of this, nor do I have any hardware > that needs such instructions, I'm just speculating. Needs more testing > on real hardware.) No, it's only for targets of indirect jumps/calls, and only on the Unix side at this point. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9671#note_124942

1 0

Re: [PATCH v2 0/1] MR9671: include: Add bti jc for __ASM_DEFINE_FUNC on ARM64
by Alfred Agrell (＠Alcaro) Dec. 5, 2025

Dec. 5, 2025

On Fri Dec 5 15:55:02 2025 +0000, André Zwing wrote: > Every ASM_DEFINE_FUNC is a jump/call target, no? So without that > injected instruction jumps/calls won't work AFAIU Isn't this stuff needed for every function pointer target, including every cross-dll operation (dllexport, callbacks, COM, etc)? Which is (to my knowledge) most of our asm functions. Another option would be find some Linux syscall to disable the check for branch target instructions. We most likely need that to deal with user-provided ARM-native exe files, anyways. (Disclaimer: I haven't tested any of this, nor do I have any hardware that needs such instructions, I'm just speculating. Needs more testing on real hardware.) -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9671#note_124941

1 0

[PATCH v2 0/1] MR9688: faudio: Make sure at least some wavebank notifications get allocated (ASan).
by Bernhard Übelacker Dec. 5, 2025

Dec. 5, 2025

With the update of faudio in d4479c27 I received below ASan report. It looks like the `wavebank_notifications_capacity` stays at zero, therefore a zero byte allocation is requested. CC: @zfigura - [ASan gitlab CI 32-bit report](https://gitlab.winehq.org/bernhardu/wine/-/jobs/215013#L5541) - [ASan gitlab CI 64-bit report](https://gitlab.winehq.org/bernhardu/wine/-/jobs/215014#L5831) - [Test patttern page, showing daily gitlab runs of xactengine3_7:xact3 failing](https://test.winehq.org/data/patterns.html#xactengine3_7:xact3) <details> <summary>ASan report with full symbols</summary> ``` ================================================================= ==312==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x7ea2933e0690 at pc 0x6ffffdd0ec1f bp 0x7ffffe1ff280 sp 0x7ffffe1ff2c8 WRITE of size 1 at 0x7ea2933e0690 thread T0 #0 0x6ffffdd0ec1e in FACTAudioEngine_CreateStreamingWaveBank .../wine/libs/faudio/src/FACT.c:578 #1 0x6ffffdd03464 in IXACT3EngineImpl_CreateStreamingWaveBank .../wine/dlls/xactengine3_7/xact_dll.c:1280 #2 0x000140002c95 in test_notifications .../wine/dlls/xactengine3_7/tests/xact3.c:392 #3 0x0001400011bb in func_xact3 .../wine/dlls/xactengine3_7/tests/xact3.c:1576 #4 0x00014001af82 in run_test .../wine/include/wine/test.h:780 #5 0x00014001a9fc in main .../wine/include/wine/test.h #6 0x00014001cb1a in mainCRTStartup .../wine/dlls/msvcrt/crt_main.c:62 #7 0x6fffffc67a64 in BaseThreadInitThunk (C:\windows\system32\kernel32.dll+0x178027a64) #8 0x6fffffde0876 in RtlUserThreadStart (C:\windows\system32\ntdll.dll+0x170050876) 0x7ea2933e0690 is located 0 bytes inside of 1-byte region [0x7ea2933e0690,0x7ea2933e0691) allocated by thread T0 here: #0 0x6ffffe8bde13 in HeapAlloc /home/runner/work/llvm-mingw/llvm-mingw/llvm-project/compiler-rt\lib/asan/asan_malloc_win.cpp:230:3 #1 0x6ffffd02dca9 in allocator_Realloc .../wine/dlls/combase/malloc.c:211 #2 0x6ffffd02d3ba in IMalloc_Realloc .../obj\include\objidlbase.h:1336 #3 0x6ffffd02d370 in CoTaskMemRealloc .../wine/dlls/combase/malloc.c:399 #4 0x6ffffdd010e1 in XACT_Internal_Realloc .../wine/dlls/xactengine3_7/xact_dll.c:1661 #5 0x6ffffdd0ea8f in FACTAudioEngine_CreateStreamingWaveBank .../wine/libs/faudio/src/FACT.c:574 #6 0x6ffffdd03464 in IXACT3EngineImpl_CreateStreamingWaveBank .../wine/dlls/xactengine3_7/xact_dll.c:1280 #7 0x000140002c95 in test_notifications .../wine/dlls/xactengine3_7/tests/xact3.c:392 #8 0x0001400011bb in func_xact3 .../wine/dlls/xactengine3_7/tests/xact3.c:1576 #9 0x00014001af82 in run_test .../wine/include/wine/test.h:780 #10 0x00014001a9fc in main .../wine/include/wine/test.h #11 0x00014001cb1a in mainCRTStartup .../wine/dlls/msvcrt/crt_main.c:62 #12 0x6fffffc67a64 in BaseThreadInitThunk (C:\windows\system32\kernel32.dll+0x178027a64) #13 0x6fffffde0876 in RtlUserThreadStart (C:\windows\system32\ntdll.dll+0x170050876) SUMMARY: AddressSanitizer: heap-buffer-overflow .../wine/libs/faudio/src/FACT.c:578 in FACTAudioEngine_CreateStreamingWaveBank ``` </details> -- v2: faudio: Make sure at least some wavebank notifications get allocated (ASan). https://gitlab.winehq.org/wine/wine/-/merge_requests/9688

1 2

Re: [PATCH v2 0/1] MR9671: include: Add bti jc for __ASM_DEFINE_FUNC on ARM64
by André Zwing Dec. 5, 2025

Dec. 5, 2025

On Fri Dec 5 15:55:02 2025 +0000, Alexandre Julliard wrote: > No, we don't want to silently inject instructions at the start of every > function, particularly when using SEH. > This should be added explicitly on the functions that need it, which at > this point probably means only syscall/unixcall dispatchers. Every ASM_DEFINE_FUNC is a jump/call target, no? So without that injected instruction jumps/calls won't work AFAIU -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9671#note_124937

1 0

[PATCH 0/1] MR9702: fusion/tests: Run tests on .NET version 4.
by Hans Leidekker (＠hans) Dec. 5, 2025

Dec. 5, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9702

2 1

[PATCH 0/5] MR9701: quartz/vmr7: Various small bits
by Matteo Bruni (＠Mystral) Dec. 5, 2025

Dec. 5, 2025

A few more patches after !9520. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9701

2 5

[PATCH 0/1] MR9700: dbgeng/test: Fix test failures on Windows 11.
by Hans Leidekker (＠hans) Dec. 5, 2025

Dec. 5, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9700

2 1

[PATCH 0/1] MR9699: win32u: Expose onscreen single buffer formats with PFD_SUPPORT_GDI.
by Rémi Bernon Dec. 5, 2025

Dec. 5, 2025

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=59053 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9699

1 1

[PATCH 0/2] MR9698: msado15: MS Jet Provider related improvements.
by Piotr Caban (＠piotr) Dec. 5, 2025

Dec. 5, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9698

2 2

[PATCH 0/1] MR9697: advapi32/tests: Fix test failures on Windows 11.
by Hans Leidekker (＠hans) Dec. 5, 2025

Dec. 5, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9697

2 1

[PATCH 0/1] MR9696: win32u: Workaround a 32-bit llvmpipe crash on initialization.
by Rémi Bernon Dec. 5, 2025

Dec. 5, 2025

I believe it was crashing already before but this has become more of an issue as the EGL/GL initialization now occurs when loading opengl32.dll unixlib, which prevent any opengl-dependent test to even load. This does not happen with 64-bit unix side, nor does it happen with other drivers, and I suspect this might either be some OOM issue with llvmpipe requiring much more memory, or some bug in llvmpipe initialization. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9696

1 1

Re: [PATCH v2 0/1] MR9671: include: Add bti jc for __ASM_DEFINE_FUNC on ARM64
by Alexandre Julliard (＠julliard) Dec. 5, 2025

Dec. 5, 2025

No, we don't want to silently inject instructions at the start of every function, particularly when using SEH. This should be added explicitly on the functions that need it, which at this point probably means only syscall/unixcall dispatchers. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9671#note_124922

1 0

Re: [PATCH v2 0/1] MR9671: include: Add bti jc for __ASM_DEFINE_FUNC on ARM64
by Jinoh Kang (＠iamahuman) Dec. 5, 2025

Dec. 5, 2025

IMHO it could be a generic macro so that it can also support x86 CET (endbr64). -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9671#note_124919

1 0

[PATCH v3 0/5] MR9690: win32u: Implement OpenGL shared resources import.
by Rémi Bernon Dec. 5, 2025

Dec. 5, 2025

Supersedes !9616 and !9572 -- v3: win32u: Implement glImportSemaphoreWin32(Handle/Name)EXT for opaque handle types. win32u: Implement glImportMemoryWin32(Handle/Name)EXT for opaque handle types. https://gitlab.winehq.org/wine/wine/-/merge_requests/9690

2 6

[PATCH 0/1] MR9695: win32u: Require VK_KHR_external_semaphore_fd for keyed mutexes.
by Rémi Bernon Dec. 5, 2025

Dec. 5, 2025

We need to use vkImportSemaphoreFdKHR to import the keyed mutex sync object, and it will be NULL if the extension hasn't been enabled. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9695

1 1

[PATCH v2 0/5] MR9690: win32u: Implement OpenGL shared resources import.
by Rémi Bernon Dec. 5, 2025

Dec. 5, 2025

Supersedes !9616 and !9572 -- v2: win32u: Implement glImportSemaphoreWin32(Handle/Name)EXT for opaque handle types. win32u: Implement glImportMemoryWin32(Handle/Name)EXT for opaque handle types. win32u: Expose device LUID / nodes mask to opengl32. https://gitlab.winehq.org/wine/wine/-/merge_requests/9690

3 7

[PATCH v2 0/7] MR9693: mfsrcsnk: Implement thinning.
by Rémi Bernon Dec. 5, 2025

Dec. 5, 2025

Supersedes https://gitlab.winehq.org/wine/wine/-/merge_requests/8505 -- v2: mfsrcsnk: Implement thinning. mfsrcsnk: Peek for stream token presence without removing it. mfsrcsnk: Factor out a media_source_request_stream_sample helper. mfsrcsnk: Process SetRate asynchronously. mfsrcsnk: Introduce a DEFINE_MF_ASYNC_PARAMS macro. winedmo: Fall back to dts for sample time if pts is not present. mfsrcsnk/tests: Add tests for thinning. https://gitlab.winehq.org/wine/wine/-/merge_requests/9693

2 8

[PATCH 0/5] MR9690: win32u: Implement OpenGL shared resources import.
by Rémi Bernon Dec. 5, 2025

Dec. 5, 2025

Supersedes !9616 and !9572 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9690

3 7