August 2022 - Wine-gitlab - list.winehq.org

Re: [PATCH v2 0/5] MR611: win32u: interface adjustments for wow64
by Jacek Caban (＠jacek) 10 Aug '22

10 Aug '22

On Tue Aug 9 15:07:54 2022 +0000, Brendan Shanks wrote: > It would be nice if this was large-address-aware, likely based on the > `SystemEmulationBasicInformation` `HighestUserAddress` the same way it's > done in `wow64`. Yes, good idea. I copied it from audio drivers, we will probably want to change it widespread. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/611#note_6015

1 0

Re: [PATCH v3 0/13] MR5: vkd3d-shader/hlsl: Replace register offsets with index paths until split copies. [v4] - approved
by Henri Verbeet (＠hverbeet) 10 Aug '22

10 Aug '22

This merge request was approved by Henri Verbeet. -- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/5

1 0

[PATCH 0/1] MR623: dwrite/fallback: Fill in some more fallback ranges for various scripts.
by Nikolay Sivov (＠nsivov) 10 Aug '22

10 Aug '22

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/623

2 1

[PATCH 0/4] MR622: winhttp: Allow custom headers to be separated by any sequence of \r and \n.
by Zebediah Figura (＠zfigura) 10 Aug '22

10 Aug '22

Patches 3 and 4 are simply extra attempts to improve code, and can be discarded if deemed undesirable. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/622

3 5

[PATCH v2 0/6] MR581: d3d8, d3d9: More tests for dirty rects and UpdateTexture().
by Zebediah Figura (＠zfigura) 10 Aug '22

10 Aug '22

-- v2: d3d9/tests: Add more tests for dirty rect handling. d3d8/tests: Add more tests for dirty rect handling. d3d8/tests: Read back directly from the specified surface if possible. d3d9/tests: Expand tests for valid pools in UpdateTexture(). d3d8/tests: Expand tests for valid pools in UpdateTexture(). wined3d: Use wined3d_array_reserve() in wined3d_adapter_create_output(). https://gitlab.winehq.org/wine/wine/-/merge_requests/581

3 7

[PATCH 0/2] MR614: quartz: Open files with FILE_SHARE_DELETE in FileSource
by David Gow (＠sulix) 10 Aug '22

10 Aug '22

Some games (such as Digimon Survive) create temporary video files and will hang if deleting them fails. Open the files with FILE_SHARE_DELETE, which will allow this deletion to go ahead even if the FileSource hasn't yet been closed. Note that many windows codec packs do themselves open files without FILE_SHARE_DELETE, so a similar hang can be observed in some windows configurations. I haven't checked that this is the file share mode used on windows (alas, I don't have a windows machine available), so I haven't removed the FIXME comment. Equally, I also updated the CreateFileW() call in get_media_type(), but that _may_ be unnecessary. I've added a basic test, but haven't had the opportunity to run it on Windows yet. Otherwise, this patch is also available against Proton Experimental here: https://github.com/ValveSoftware/wine/pull/156 It also has been confirmed to fix the issue with Digimon Survive independently: https://github.com/ValveSoftware/Proton/issues/6041#issuecomment-1200486581 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/614

3 3

[PATCH 0/2] MR615: include: Add IMediaParams definitions.
by Zhiyi Zhang (＠zhiyi) 10 Aug '22

10 Aug '22

Signed-off-by: Zhiyi Zhang <zzhang(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/615

3 3

[PATCH 0/1] MR618: d3d11: Validate layout description in d3d_input_layout_create().
by Paul Gofman (＠gofman) 10 Aug '22

10 Aug '22

Fixes black screen in Sweet Transit game. The game depends on ID3D11Device_CreateInputLayout to fail properly when mandatory VS input is not filled. It is coming from MonoGame engine available as source, here is the place of interest: https://github.com/MonoGame/MonoGame/blob/158c0154ac18ed6102c65e24665c6a080… As far as my testing goes: - missing mandatory inputs cause failure, extra elements in input layout are fine; - shader type is not checked; - what is mandatory is unrelated to sysval semantics value in dxbc, it probably just checks for sv_ special names with position being a special case which is still mandatory; - there are more validation checks which are not implemented in this patch, e. g., for InputSlotClass validity. There is a single todo_wine test for that, but it doesn't cover all the specifics. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/618

3 2

[PATCH 0/2] MR616: winegstreamer: Always check the return value from mf_media_type_from_wg_format().
by Zhiyi Zhang (＠zhiyi) 10 Aug '22

10 Aug '22

mf_media_type_from_wg_format() may return NULL. Fix Airborne Kingdom crash at start because WG_AUDIO_FORMAT_UNKNOWN is passed to mf_media_type_from_wg_format(). Signed-off-by: Zhiyi Zhang <zzhang(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/616

3 4

[PATCH v4 0/13] MR600: wine{alsa,pulse}: Merge unixlib.h, move it into mmdevapi
by Davide Beatrici (＠davidebeatrici) 09 Aug '22

09 Aug '22

This is the very first step for merging the `mmdevdrv.c` files into a single one. The commit strictly focuses on Linux drivers, `wine{oss,coreaudio}` will follow. Huge thanks to @jacek and @huw for making this simple, as they took care of ELF/PE separation. -- v4: winepulse: Switch to mmdevapi's unixlib.h winepulse: Adapt "get_prop_value_params" struct to mmdevapi's winepulse: Adapt "is_started_params" struct to mmdevapi's winepulse: Adapt "get_device_period_params" struct to mmdevapi's winepulse: Adapt "get_mix_format_params" struct to mmdevapi's winepulse: Adapt "get_capture_buffer_params" struct to mmdevapi's winepulse: Adapt "release_render_buffer_params" struct to mmdevapi's winepulse: Adapt "release_stream_params" struct to mmdevapi's winepulse: Adapt "create_stream_params" struct to mmdevapi's winepulse: Adapt "endpoint" struct to mmdevapi's mmdevapi: Integrate winepulse's additions in unixlib.h winealsa: Move common unixlib.h content into mmdevapi winealsa: Drop "alsa_" prefix in unixlib enum, apply it to the functions instead https://gitlab.winehq.org/wine/wine/-/merge_requests/600

2 15