November 2022 - wine-gitlab

[PATCH v18 0/11] MR1180: Make all kernel callbacks go through user32
by Torge Matthies (＠tmatthies) 23 Nov '22

23 Nov '22

Overwatch 2 verifies that every kernel callback that is run, lives in user32. Introduce a callback in user32 that just forwards to the other modules' callbacks. -- v18: include: Add a comment explaining why all kernel callbacks must be in user32. user32: Remove NtUserDriverCallback* kernel callbacks. winex11.drv: Route kernel callbacks through user32. winex11.drv: Pass a struct to x11drv_ime_set_result. winex11.drv: Pass a struct to x11drv_dnd_post_drop. winemac.drv: Route kernel callbacks through user32. wineandroid.drv: Route kernel callbacks through user32. opengl32: Route kernel callbacks through user32. winevulkan: Route kernel callbacks through user32. user32: Add NtUserDispatchCallback kernel callback. https://gitlab.winehq.org/wine/wine/-/merge_requests/1180

2 12

Re: [PATCH 0/1] MR99: winevulkan: Propagate access violation from the Unix side. - closed
by Paul Gofman (＠gofman) 23 Nov '22

23 Nov '22

This merge request was closed by Paul Gofman. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/99

1 0

[PATCH 0/1] MR1510: winebus.sys: Use the device path instead of the udev_device pointer to match devices.
by Zebediah Figura (＠zfigura) 23 Nov '22

23 Nov '22

The latter is not guaranteed to be equal across events, and in practice may not be. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1510

3 2

Re: [PATCH v2 0/2] MR1508: winevulkan: Union conversion and 1.3.235 spec update.
by Jacek Caban (＠jacek) 23 Nov '22

23 Nov '22

Jacek Caban (@jacek) commented about dlls/winevulkan/vulkan_thunks.c: > return (void *)out; > } > > +static inline void convert_VkDeviceOrHostAddressKHR_win32_to_host(const VkDeviceOrHostAddressKHR32 *in, VkDeviceOrHostAddressKHR *out) > +{ > + if (!in) return; > + > + out->deviceAddress = in->deviceAddress; > + out->hostAddress = (void *)UlongToPtr(in->hostAddress); FWIW, this will not do the right thing on wow64 when the address is meant to be device address. It will override the upper part of deviceAddress. It was broken in a different way before that commit by not zeroing upper 32 bits when host address should be used, so it essentially depended on caller to initialize that part. Unfortunately, I don't see a good way of solving it properly. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1508#note_17182

1 0

Re: [PATCH 0/1] MR99: winevulkan: Propagate access violation from the Unix side.
by Jacek Caban (＠jacek) 23 Nov '22

23 Nov '22

I guess this can be closed, we have 572dd0b466c54 now. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/99#note_17180

1 0

[PATCH 0/1] MR1523: configure.ac: Log pkg-config error messages when it fails.
by eric pouech (＠epo) 23 Nov '22

23 Nov '22

While investigating https://bugs.winehq.org/show_bug.cgi?id=53953, it may be a good idea to log any pkg-config error messages instead of ignoring them. Signed-off-by: Eric Pouech <eric.pouech(a)gmail.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1523

2 1

[PATCH v3 0/2] MR1492: Load and save application-specific console settings from the registry
by Hugh McMaster (＠hmc) 23 Nov '22

23 Nov '22

On Windows, application-specific settings are stored in the registry under HKCU\Console\<path_to_app>. Our implementation of conhost.exe does not know which process is connected to it, so we need to get the full process image name before loading any application-specific console settings. To do this, we extend the server protocol to pass the process Id of the connected console process to conhost.exe. Please run tools/make_requests before merging. -- v3: conhost: Load application-specific settings using the full process image name server: Send console process ID via get_next_console_request() https://gitlab.winehq.org/wine/wine/-/merge_requests/1492

3 6

[PATCH v5 0/5] MR1278: winegstreamer: Do not destroy condition variables when a thread is waiting on them.
by Giovanni Mascellani (＠giomasce) 23 Nov '22

23 Nov '22

It is possible that a stream is destroyed while another thread is waiting on event_empty_cond or event_cond. The waiting thread should have the opportunity to be rescheduled and exit the critical section before the condition variables and the mutex are destroyed. -- v5: winegstreamer: Synchronize media source async commands and shutdown. winegstreamer: Free the GStreamer buffer when freeing a WG parser stream. winegstreamer: Synchronize access to the media source from callbacks. winegstreamer: Synchronize concurrent access to the media stream. winegstreamer: Synchronize concurrent access to the media source. https://gitlab.winehq.org/wine/wine/-/merge_requests/1278

2 5

[PATCH v4 0/5] MR1278: winegstreamer: Do not destroy condition variables when a thread is waiting on them.
by Giovanni Mascellani (＠giomasce) 23 Nov '22

23 Nov '22

It is possible that a stream is destroyed while another thread is waiting on event_empty_cond or event_cond. The waiting thread should have the opportunity to be rescheduled and exit the critical section before the condition variables and the mutex are destroyed. -- v4: winegstreamer: Synchronize media source async commands and shutdown. winegstreamer: Free the GStreamer buffer when freeing a WG parser stream. winegstreamer: Synchronize access to the media source from callbacks. winegstreamer: Synchronize concurrent access to the media stream. winegstreamer: Synchronize concurrent access to the media source. https://gitlab.winehq.org/wine/wine/-/merge_requests/1278

3 9

[PATCH v2 0/2] MR1522: mfmediaengine: Volume change.
by Bernhard Kölbl (＠besentv) 23 Nov '22

23 Nov '22

-- v2: mfmediaengine: Pass volume changes to the audio renderer. https://gitlab.winehq.org/wine/wine/-/merge_requests/1522

3 3

2025

2024

2023

2022

wine-gitlab November 2022