- wine-gitlab - winehq.org

[PATCH 0/7] MR8752: opengl32: Use wow64 wrappers for memory mapping functions.
by Jacek Caban (＠jacek) 11 Aug '25

11 Aug '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/8752

2 7

[PATCH 0/1] MR8712: strmbase: Get rid of stream_cs.
by Conor McCarthy (＠cmccarthy) 11 Aug '25

11 Aug '25

The following sequence can occur: sink_EndOfStream, lock stream_cs, renderer sink_eos, strmbase_passthrough_eos, IMediaSeeking_GetStopPosition, MediaSeekingPassThru_GetStopPosition, get_connected, IPin_ConnectedTo, lock filter_cs And at the same time: filter_Stop, lock filter_cs, dmo_wrapper_cleanup_stream, lock stream_cs Resulting in deadlock. This can occur in Space Engineers. dmo_wrapper_cleanup_stream() locks stream_cs to access `pin.peer` and `pin.pAllocator`, which are normally protected by filter_cs, but this lock was added to fix a test failure, so I don't know if there's an alternative. I'm concerned about the possible performance impact of removing stream_cs though. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8712

3 2

[PATCH 0/2] MR8751: ntdll: Support THREAD_CREATE_FLAGS_BYPASS_PROCESS_FREEZE flag.
by Billy Laws (＠bylaws) 11 Aug '25

11 Aug '25

This is a prereq for cross-process suspend on WOW64 - the thread created for that needs to run even when the whole process is suspended by e.g. the debugger. Will be follow-up changes for THREAD_CREATE_FLAGS_SKIP_THREAD_ATTACH and THREAD_CREATE_FLAGS_SKIP_LOADER_INIT to support creating a native thread for that. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8751

2 2

[PATCH 0/1] MR8750: maintainers: Add dxcore to the D3D section.
by Elizabeth Figura (＠zfigura) 11 Aug '25

11 Aug '25

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/8750

2 1

[PATCH v2 0/1] MR8719: winegstreamer: Only add the capsfilter for avdec_h264.
by Brendan McGrath (＠redmcg) 11 Aug '25

11 Aug '25

avdec_h264 will fail when combined with h264parse and a drain request is made during caps neg. To fix this, a capsfilter was added to force a stream-format of byte-stream for all h.264 decoders. However, this introduced a new issue with the vtdec element, as it only supports the avc stream-format. Therefore, this patch looks to address both problems by only introducing the capsfilter when the avdec_h264 element is used. It also removes the stream-format and alignment requirements from the parsed caps, applying them only to the capsfilter. -- v2: winegstreamer: Only add the capsfilter for avdec_h264. https://gitlab.winehq.org/wine/wine/-/merge_requests/8719

3 7

Re: [PATCH v9 0/3] MR8034: winegstreamer: Call wg_format_get_max_size() to get the video decoder output sample size.
by Elizabeth Figura (＠zfigura) 11 Aug '25

11 Aug '25

> The `fill_planes()` helper in `gst-plugins-base/gst-libs/gst/video/video-info.c` called from `gst_video_info_align_full()` sets the minimum alignment. Apparently, setting an alignment manually in `GstVideoInfo` can only increase it. That's true, but I haven't found anything to indicate that we can't set the stride members of GstVideoInfo manually. I haven't found anything in the base video library, or several decoders, that seems to care about "minimum" alignment, and I don't really know why it would. > If I set `plane_align = 1` in `wg_transform_read_data()`, gstreamer still can't fit the output in the buffer. Because we implement stride by actually adding padding instead. That's arguably not very correct, but maybe there was a reason I've forgotten. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8034#note_112635

1 0

[PATCH 0/1] MR8747: PickIconDlg conformance tests
by Matthias Zorn (＠Matthias-Zorn.0178) 11 Aug '25

11 Aug '25

shell32: Run the conformance tests for the PickIconDlg function. The conformance test verifies that PickIconDlg behaves in the same way as the original function when illegal parameter values are used. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8747

2 2

[PATCH 0/3] MR8744: Restore autocenter support to SDL and udev input backends and provide default override to address on issues
by Tyson Whitehead (＠twhitehead) 11 Aug '25

11 Aug '25

This merge request restores `DIPROP_AUTOCENTER` support to the SDL and udev input backends. It also * adds an `Autocenter` registry app key that lets the default `DIPROP_AUTOCENTER` setting to be set on a per device and per application and device level, * logs a warning on acquisition when `DIPROP_AUTOCENTER` hasn't been checked or set that that is the case and the default can be changed via the 'Autocenter' app key if required. Together these two should provide a robust solution for the issues some where having with autocenter being incorrectly on for some programs and devices. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8744

3 7

[PATCH 0/3] MR8748: Some audio changes in preparation to adding more mmdevapi tests.
by Giovanni Mascellani (＠giomasce) 11 Aug '25

11 Aug '25

Currently the first commits in !8653, but submitted independently to avoid making an overly long MR. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/8748

3 4

[PATCH v2 0/4] MR8746: winebus: Improve gamepad report compatibility with XUSB / GIP.
by Rémi Bernon (＠rbernon) 11 Aug '25

11 Aug '25

Inverting left/right trigger usages and Y values, so that WGI doesn't need to mess with axes order. -- v2: winebus: Use a vendor specific usage for gamepad guide buttons. windows.gaming.input: Use a generic dinput device data format. winebus: Improve gamepad report compatibility with XUSB / GIP. windows.gaming.input: Only create Gamepad instances for XInput devices. https://gitlab.winehq.org/wine/wine/-/merge_requests/8746

2 4

2025

2024

2023

2022

wine-gitlab