- Wine-GitLab - list.winehq.org

[PATCH v5 0/2] MR9351: msvcrt: Correctly handle refcounts for WinRT exceptions
by Vibhav Pant (＠vibhavp) Nov. 3, 2025

Nov. 3, 2025

Follow up to https://gitlab.winehq.org/wine/wine/-/merge_requests/9209#note_120396 -- v5: ucrtbase: Add __C_specific_handler tests. msvcrt: Call the destructor for C++ exceptions in __C_specific_handler. https://gitlab.winehq.org/wine/wine/-/merge_requests/9351

4 5

[PATCH v5 0/8] MR8812: winegstreamer: Support 2D sample buffer.
by Conor McCarthy (＠cmccarthy) Nov. 3, 2025

Nov. 3, 2025

Performance issues can occur when a video output sample has a 2D buffer, because winegstreamer currently outputs only to a linear buffer, which must then be copied into the 2D buffer. Worse, a linear lock of the 2D buffer requires the current contents to be copied to a linear buffer, even when we intend to overwrite it, because MF linear buffers do not support write-only locking. -- v5: winegstreamer: Support 2D sample buffer. mf/tests: Test color convert 2D buffers. mf/tests: Test WMV decoder 2D buffers. mf/tests: Test H.264 decoder 2D buffers. mf/tests: Test sample copier 2D buffers. mfplat/tests: Test NV12 negative stride in MFCreateMediaBufferFromMediaType(). mf/tests: Remove todo for an H.264 decoder test. mf/tests: Stop checking samples at the end of the expected array. https://gitlab.winehq.org/wine/wine/-/merge_requests/8812

2 8

[PATCH v4 0/9] MR8812: winegstreamer: Support 2D sample buffer.
by Conor McCarthy (＠cmccarthy) Nov. 3, 2025

Nov. 3, 2025

Performance issues can occur when a video output sample has a 2D buffer, because winegstreamer currently outputs only to a linear buffer, which must then be copied into the 2D buffer. Worse, a linear lock of the 2D buffer requires the current contents to be copied to a linear buffer, even when we intend to overwrite it, because MF linear buffers do not support write-only locking. -- v4: mf/tests: Test H.264 sink media type height alignment. winegstreamer: Support 2D sample buffer. mf: Test color convert 2D buffers. mf: Test WMV decoder 2D buffers. mf: Test H.264 decoder 2D buffers. mf: Test sample copier 2D buffers. mfplat/tests: Test NV12 negative stride in MFCreateMediaBufferFromMediaType(). mf/tests: Remove todo for an H.264 decoder test. mf/tests: Stop checking samples at the end of the expected array. https://gitlab.winehq.org/wine/wine/-/merge_requests/8812

3 13

[PATCH v15 0/11] MR9209: vccorlib140: Implement Platform::Exceptions.
by Vibhav Pant (＠vibhavp) Nov. 3, 2025

Nov. 3, 2025

-- v15: vccorlib140: Implement __abi_translateCurrentException. vccorlib140: Throw exceptions on error paths. https://gitlab.winehq.org/wine/wine/-/merge_requests/9209

4 17

[PATCH 0/1] MR9352: msvcrt: Fix WinRT exception leaks in _except_handler4_common.
by Piotr Caban (＠piotr) Nov. 3, 2025

Nov. 3, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9352

2 1

[PATCH v2 0/1] MR9339: winhttp/tests: Avoid buffer-overflow in WinHttpCreateUrl_test (ASan).
by Bernhard Übelacker Nov. 3, 2025

Nov. 3, 2025

The test sets in fill_url_components the dwPasswordLength to 8, therefore the memcpy copies 8 wide characters. But before this patch variable empty allocates just one wide character. [ASan report details](https://gitlab.winehq.org/bernhardu/wine/-/blob/asan-pe_2025-10-18… -- v2: winhttp/tests: Fix setting string lengths in WinHttpCreateUrl_test (ASan). https://gitlab.winehq.org/wine/wine/-/merge_requests/9339

2 4

[PATCH v4 0/2] MR9351: msvcrt: Correctly handle refcounts for WinRT exceptions
by Vibhav Pant (＠vibhavp) Nov. 3, 2025

Nov. 3, 2025

Follow up to https://gitlab.winehq.org/wine/wine/-/merge_requests/9209#note_120396 -- v4: ucrtbase: Add __C_specific_handler tests. msvcrt: Call the destructor for C++ exceptions in __C_specific_handler. https://gitlab.winehq.org/wine/wine/-/merge_requests/9351

3 4

[PATCH v3 0/4] MR9351: Draft: msvcrt: Correctly handle refcounts for WinRT exceptions
by Vibhav Pant (＠vibhavp) Nov. 3, 2025

Nov. 3, 2025

Follow up to https://gitlab.winehq.org/wine/wine/-/merge_requests/9209#note_120396 -- v3: ucrtbase: Add __C_specific_handler tests. https://gitlab.winehq.org/wine/wine/-/merge_requests/9351

4 5

[PATCH v43 0/6] MR7650: Add support for AF_UNIX sockets (followup)
by Ralf Habacker (＠rhabacker) Nov. 3, 2025

Nov. 3, 2025

Follow-up of !2786, which appears to have been abandoned. Depends on !8182, !8575, !8578 # Execute minimal test case 1. checkout wine into ~/src/wine 2. in dlls/ws2_32/tests/sock.c wrap all calls to test functions after `Init()` until `test_afunix()` with #if 0 ... #endif 3. build into ~/src/wine-build 4. run `(cd ~/src/wine-build; WINEPREFIX=~/src/wine/.wine ../wine/tools/runtest -q -P wine -T . -M ws2_32.dll -p dlls/ws2_32/tests/i386-windows/ws2_32_test.exe sock)` # How to debug wineserver Since `wineserver` runs in the background, simple calls to printf() will show nothing when the test case is executed. A workaround is to open a file and call fprintf() to write debug messages to this file, which can then be inspected. # How to debug a wine test case At https://gitlab.winehq.org/wine/wine/-/blob/master/tools/runtest#L151 there is mentioned a environment variable name `WINETEST_WRAPPER`, with which it may be possible to run gdb to debug test cases ``` (cd ~/src/wine-build; make -j10 ; WINETEST_WRAPPER="gdb --args" WINEPREFIX=~/src/wine/.wine ../wine/tools/runtest -P wine -T . -M ws2_32.dll -p dlls/ws2_32/tests/i386-windows/ws2_32_test.exe sock) Reading symbols from ./wine... (gdb) set follow-fork-mode child (gdb) r current directory: 'Z:\home\ralf.habacker\src\wine-build' sock.c:14624: Test failed: test_afunix.sock: wrong attr ffffffff sock.c:14658: Test failed: test_afunix.sock: wrong family 0 sock.c:14693: Test failed: test_afunix.sock: wrong family 0 ... ``` With this you can debug the wine application, but unfortunally not the windows api. # How to debug a wine test case (windows api) 1. After building wine install the binaries into a temporary location ``` make -C ~/src/wine-build install DESTDIR=~/src/wine-install ``` 2. run ``` (cd ../wine-build; \ WINEDLLDIR=$HOME/wine-install/usr/local/lib/wine/i386-windows/ \ WINEPREFIX=~/src/wine/.wine \ WINEDLLOVERRIDES=';ws2_32.dll=b' \ WINETEST_PLATFORM=wine \ ./wine programs/winedbg/i386-windows/winedbg.exe --gdb dlls/ws2_32/tests/i386-windows/ws2_32_test.exe sock) ``` Now let's look at the loaded shared libraries ``` Wine-gdb> info sharedlibrary >From To Syms Read Shared Object Library 0xf7b39000 0xf7bff6e0 Yes /home/user/src/wine-build/dlls/ntdll/ntdll.so 0x00401000 0x00478e2c Yes /home/user/src/wine-build/dlls/ws2_32/tests/i386-windows/ws2_32_test.exe 0x7bcc1000 0x7bd69280 Yes /home/user/src/wine/.wine/drive_c/windows/system32/ntdll.dll 0x7bb31000 0x7bb91594 Yes /home/user/src/wine/.wine/drive_c/windows/system32/kernel32.dll 0x7b5b1000 0x7b841f38 Yes /home/user/src/wine/.wine/drive_c/windows/system32/kernelbase.dll 0x77fd1000 0x77ff89d8 Yes /home/user/src/wine/.wine/drive_c/windows/system32/iphlpapi.dll 0x7b4b1000 0x7b4ebcb8 Yes /home/user/src/wine/.wine/drive_c/windows/system32/advapi32.dll 0x7b1e1000 0x7b280014 Yes /home/user/src/wine/.wine/drive_c/windows/system32/msvcrt.dll 0x7b151000 0x7b16de40 Yes /home/user/src/wine/.wine/drive_c/windows/system32/sechost.dll 0x7adf1000 0x7aec2690 Yes /home/user/src/wine/.wine/drive_c/windows/system32/ucrtbase.dll 0x77f71000 0x77f85a80 Yes /home/user/src/wine/.wine/drive_c/windows/system32/dnsapi.dll 0x77f31000 0x77f3b240 Yes /home/user/src/wine/.wine/drive_c/windows/system32/nsi.dll 0x7ad41000 0x7ad6653c Yes /home/user/src/wine/.wine/drive_c/windows/system32/ws2_32.dll 0x7a111000 0x7a2cc55c Yes /home/user/src/wine/.wine/drive_c/windows/system32/user32.dll 0x7a631000 0x7a6ae67c Yes /home/user/src/wine/.wine/drive_c/windows/system32/gdi32.dll 0x7a0b1000 0x7a0e48e8 Yes /home/user/src/wine/.wine/drive_c/windows/system32/win32u.dll 0x78121000 0x7813de90 Yes /home/user/src/wine/.wine/drive_c/windows/system32/imm32.dll ``` It is visible, that the unix variant of the shared libraries (ntdll.so) are read from the build dir and also the test case. The Windows part (*.dll) is loaded from the wine prefix, so that these must be copied from the build directory initially and whenever changes are made. With this is possible to set breakpoints ``` Wine-gdb> b GetFileAttributesW(a)4 Breakpoint 1 at 0x7b5d3725: file ../wine/dlls/kernelbase/file.c, line 1665. Wine-gdb> c Continuing. Breakpoint 1, GetFileAttributesW(a)4 ( name=0x40fcfa <test_afunix+1434> L"\xec83\x8304\xfff8\x940f\x89c0\x247c\xf08\xc0b6\x44c7\x424\x1bbf\103\x489\xe824\x52e6\001\x44c7", <incomplete sequence \x824>) at ../wine/dlls/kernelbase/file.c:1665 1665 TRACE( "%s\n", debugstr_w(name) ); ``` -- v43: ws2_32: Add note in bind() for AF_UNIX sockets ws2_32/tests: In tests for AF_UNIX sockets print actual directory used https://gitlab.winehq.org/wine/wine/-/merge_requests/7650

3 8

[PATCH v42 0/7] MR7650: Add support for AF_UNIX sockets (followup)
by Ralf Habacker (＠rhabacker) Nov. 3, 2025

Nov. 3, 2025

Follow-up of !2786, which appears to have been abandoned. Depends on !8182, !8575, !8578 # Execute minimal test case 1. checkout wine into ~/src/wine 2. in dlls/ws2_32/tests/sock.c wrap all calls to test functions after `Init()` until `test_afunix()` with #if 0 ... #endif 3. build into ~/src/wine-build 4. run `(cd ~/src/wine-build; WINEPREFIX=~/src/wine/.wine ../wine/tools/runtest -q -P wine -T . -M ws2_32.dll -p dlls/ws2_32/tests/i386-windows/ws2_32_test.exe sock)` # How to debug wineserver Since `wineserver` runs in the background, simple calls to printf() will show nothing when the test case is executed. A workaround is to open a file and call fprintf() to write debug messages to this file, which can then be inspected. -- v42: ws2_32: Add note in bind() for AF_UNIX sockets ws2_32/tests: In tests for AF_UNIX sockets print actual directory used ws2_32/tests: In AF_UNIX socket tests always print used path in case of errors server: Fix getsockname() and accept() on AF_UNIX sockets. server: Introduce error when attempting to create a SOCK_DGRAM AF_UNIX socket. ws2_32: Add support for AF_UNIX sockets. server: Allow for deletion of socket files. https://gitlab.winehq.org/wine/wine/-/merge_requests/7650

3 13

[PATCH v14 0/11] MR9209: vccorlib140: Implement Platform::Exceptions.
by Vibhav Pant (＠vibhavp) Nov. 3, 2025

Nov. 3, 2025

-- v14: vccorlib140: Implement __abi_translateCurrentException. vccorlib140: Throw exceptions on error paths. vccorlib140: Add Platform::Exception::CreateException() implementation. vccorlib140: Add Platform::Exception::Message::get() implementation. vccorlib140: Add Platform::COMException constructor implementation. vccorlib140: Add Platform::Exception(HSTRING) constructor implementation. msvcrt: Sync cxx.h file with msvcp90 and vccorlib140. vccorlib140: Add Platform::Exception constructor implementation. vccorlib140: Remove platform_ prefix from exception function names. vccorlib140: Move exception functions to new file. vccorlib140/tests: Add additional tests for exception objects. https://gitlab.winehq.org/wine/wine/-/merge_requests/9209

4 21

[PATCH v2 0/3] MR9351: Draft: msvcrt: Correctly handle refcounts for WinRT exceptions
by Vibhav Pant (＠vibhavp) Nov. 3, 2025

Nov. 3, 2025

Follow up to https://gitlab.winehq.org/wine/wine/-/merge_requests/9209#note_120396 -- v2: msvcrt: Call Release instead of AddRef for WinRT exceptions in copy_exception helper. msvcp90: Call Release instead of AddRef for WinRT exceptions in copy_exception helper. https://gitlab.winehq.org/wine/wine/-/merge_requests/9351

3 4

[PATCH 0/3] MR9351: Draft: msvcrt: Correctly handle refcounts for WinRT exceptions
by Vibhav Pant (＠vibhavp) Nov. 3, 2025

Nov. 3, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9351

2 3

[PATCH v3 0/5] MR9271: Implement group affinity support in CreateRemoteThreadEx.
by eric pouech (＠epo) Nov. 3, 2025

Nov. 3, 2025

This MR implements group affinity support in attributes list passed to CreateRemoteThreadEx(). (up to the current Wine limit of 64 logical cores). It also adds a couple of missing APIs definitions related to group affinity support. Notes: - duplicating for the Nth time struct \_PROC\_THREAD\_ATTRIBUTE\_LIST definition isn't very nice... perhaps we should define it somewhere? - it's very likely that CreateRemoteThreadEx() should call into NtCreateThreadEx() (instead of the Rtl counter part), but I opted for the simpler approach to set thread affinity after thread creation (this could be changed if needed). -- v3: kernelbase: Support group affinity attributes. kernel32/tests: Test thread creation with group affinity attributes. include: Add missing process group related definitions. kernelbase: Support affinity group in process/thread attributes list. kernel32: Test adding group affinity to proc/thread attributes list. https://gitlab.winehq.org/wine/wine/-/merge_requests/9271

3 6

[PATCH v2 0/1] MR9260: winemac: Make macdrv_ime_process_key synchronous.
by Marc-Aurel Zent (＠mzent) Nov. 3, 2025

Nov. 3, 2025

This is a simplification of https://gitlab.winehq.org/wine/wine/-/merge_requests/9246. -- v2: winemac: Make macdrv_ime_process_key synchronous. https://gitlab.winehq.org/wine/wine/-/merge_requests/9260

4 14

[PATCH 0/2] MR9349: win32u: Fix deadlock caused by nested locking.
by Zhao Yi (＠Zhaoyi) Nov. 3, 2025

Nov. 3, 2025

Convert surface lock to recursive mutex to allow same thread to acquire the lock multiple times, preventing deadlocks when lock_surface() and window_surface_lock() are called nestedly. Signed-off-by: Zhao Yi <zhaoyi(a)uniontech.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9349

2 3

[PATCH v3 0/2] MR9180: win32u: Fix deadlock caused by nested locking.
by Zhao Yi (＠Zhaoyi) Nov. 3, 2025

Nov. 3, 2025

Within the NtUserUpdateLayeredWindow function, there exists the following call sequence: window_surface_lock()->NtGdiAlphaBlend()->window_surface_unlock(). However, the implementation of the NtGdiAlphaBlend() function calls the windrv_GetImage() function, which internally invokes lock_surface() and unlock_surface(). In this situation, window_surface_lock() is called first to acquire a lock. When lock_surface() is subsequently called to acquire the same lock, the initial lock has not yet been released, ultimately resulting in a deadlock. Signed-off-by: Zhao Yi <zhaoyi(a)uniontech.com> -- v3: shell32: Fix a memory leak issue. https://gitlab.winehq.org/wine/wine/-/merge_requests/9180

2 2

[PATCH 0/1] MR9346: rometadata: Use correct argument when unmapping a file view (Coverity).
by Nikolay Sivov (＠nsivov) Nov. 3, 2025

Nov. 3, 2025

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9346

2 2

[PATCH v2 0/1] MR9345: d2d1: Fix missing geometry assignment in command list recording
by Yeshun Ye (＠yeyeshun) Nov. 3, 2025

Nov. 3, 2025

The d2d_command_list_draw_geometry() function didn't store the 'geometry', this may cause an error when command replay. Signed-off-by: YeshunYe <yeyeshun(a)uniontech.com> -- v2: d2d1: Fix missing geometry assignment in command list recording https://gitlab.winehq.org/wine/wine/-/merge_requests/9345

2 2

[PATCH 0/4] MR9296: dmsynth, dmusic: Reduce memory usage part 1
by Anton Baskanov (＠baskanov) Nov. 3, 2025

Nov. 3, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9296

3 6

[PATCH v3 0/4] MR9323: user32: Fix loading comctl32 v5 after comctl32 v6 is loaded.
by Zhiyi Zhang (＠zhiyi) Nov. 3, 2025

Nov. 3, 2025

When comctl32 v6 is loaded and v6 manifest is disabled, comctl32 v5 should be used when creating a common control window. However, before this patch, GetModuleHandleW(L"comctl32") return a handle to comctl32 v6 in this case, so comctl32 v5 doesn't get loaded and creating comctl32 v5 windows fails. We need to check if comctl32 v5 is actually loaded by using GetModuleHandleW() with the absolute path of the comctl32 v5 dll. Fix Word 2010 file open dialog doesn't show a listview. -- v3: comctl32: Use a magic value to check if an image list is valid. comctl32/tests: Test image list interoperation with comctl32 v5 and v6. user32: Fix loading comctl32 v5 after comctl32 v6 is loaded. comctl32/tests: Test creating v5 windows after v6 manifest is deactivated. https://gitlab.winehq.org/wine/wine/-/merge_requests/9323

3 6

[PATCH 0/1] MR9321: winex11: Trace flags with %#x.
by Elizabeth Figura (＠zfigura) Nov. 3, 2025

Nov. 3, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9321

3 2

[PATCH 0/1] MR9319: winex11.drv: Fix an inverted condition in x11drv_surface_swap().
by Elizabeth Figura (＠zfigura) Nov. 3, 2025

Nov. 3, 2025

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/9319

3 2

[PATCH 0/1] MR9345: d2d1: Fix missing geometry assignment in command list recording
by Yeshun Ye (＠yeyeshun) Nov. 3, 2025

Nov. 3, 2025

The d2d_command_list_draw_geometry() function didn't store the 'geometry', this may cause an error when command replay. Signed-off-by: YeshunYe <yeyeshun(a)uniontech.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9345

3 2

[PATCH v4 0/1] MR8914: msxml: IXMLDOMNamedNodeMap::getNamedItem to handle global namespace.
by Alistair Leslie-Hughes (＠alesliehughes) Nov. 3, 2025

Nov. 3, 2025

-- v4: msxml: IXMLDOMNamedNodeMap::getNamedItem to handle global namespace. https://gitlab.winehq.org/wine/wine/-/merge_requests/8914

2 1