August 2023 - Wine-GitLab - list.winehq.org

[PATCH 0/1] MR3479: widl: Add support for default_overload attribute.
by Mohamad Al-Jaf (＠maljaf) Aug. 2, 2023

Aug. 2, 2023

Needed for IDeviceInformationStatics, which is needed by Lenovo Ready For Assistant. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3479

3 6

[PATCH 0/8] MR3474: win32u: Use user message packing for remaining messages.
by Jacek Caban (＠jacek) Aug. 2, 2023

Aug. 2, 2023

Remaining !3388 parts. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3474

3 9

[PATCH 0/1] MR3478: include: Add missing default value in IXAudio2.
by Biswapriyo Nath (＠Biswa96) Aug. 2, 2023

Aug. 2, 2023

This was removed in 14c44d0b0a2fc0ead5bc75615864ff850a44c4d5 commit. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/3478

4 6

[PATCH v7 0/5] MR280: vkd3d-shader: sm1 semantic matching.
by Zebediah Figura (＠zfigura) Aug. 2, 2023

Aug. 2, 2023

This is probably the most ugly and controversial bit of API. I don't really know if this is the right approach to solving this. sm4 registers match by register index, such that shaders can mostly be compiled in isolation. sm1 does not—registers may be specified in any order in the vertex and pixel shaders, and will be matched by usage and usage index. By itself this is not much of a problem. Where it gets hairy is that we want to do some degree of caching, as well as pre-compilation, and avoid recompiling either shader every time it's matched with a new one. Wine currently deals with this problem, for GLSL, by generating a "main" GLSL shader for the vertex shader, and then an extra function setup_vs_output(), and linking the two together every time a new pixel shader is used. This could in theory be used for SPIR-V, but it requires the use of extra, probably external, code to link SPIR-V shaders together, which I do not particularly anticipate being well-received. (I'm not sure how Wine deals with this problem in the ARB backend. It seems to take the pixel shader signature into account when generating the vertex shader— cf. init_output_registers()—but it doesn't take it into account when looking up a vertex shader variant? I didn't look too closely at the code, so maybe I'm missing something.) -- The vkd3d parts of this patch are quite straightforward, and looking at them, I think the design is quite intuitive in isolation. There may be some room for internal refactoring (in particular with an eye to not so much overloading the "register_index" field of struct shader_signature_element) but I'm relatively happy with the way it turned out. In isolation, that is. The Wine part is worse. I've uploaded branches for vkd3d and Wine that use this API, and correctly handle shaders with some nontrivial reordering: https://gitlab.winehq.org/zfigura/vkd3d/-/commits/himavant5 https://gitlab.winehq.org/zfigura/wine/-/commits/himavant_cb The test can be run, as before, like so: make tests/shader_runner.exe && WINE_D3D_CONFIG=renderer=vulkan wine tests/shader_runner.exe ../vkd3d/tests/hlsl/sm1-interstage-interface.shader_test The interesting Wine parts are concentrated in a single patch, 5cfb9d930f11e. The patch takes a few shortcuts, partly because I wanted not to block the vkd3d API design questions, but also because while writing it I came up with a couple of problems that I wasn't sure how to fix. There are two main problems I see: (1) This patch has the user pass the signature from the pixel shader when compiling the vertex shader, and looks up register indices already arbitrarily allocated by the pixel shader. This is problematic when trying to use this signature as a cache key, by virtue of it not being clear (or even defined) which fields are key elements and which aren't. It's also not particularly kind to lookup, on account of not being directly comparable with memcmp(). There are a few options I see: (a) Provide an internal function to compare key elements. This feels... odd, like a very special-purpose function, but perhaps workable. (b) Just make the user deal with it, and assert that all fields are key elements. (c) Use some alternative, perhaps shortened structure as a field of vkd3d_shader_next_stage_info. This has the disadvantage that it is not as simple for a hypothetical user to retrieve from the pixel shader, but we would presumably provide a function to generate one from a shader signature. This would probably also be kinder to cache lookup if it's shorter. (d) Make caching vkd3d's responsibility, to some degree. This seems daunting, but the more we optimize, the more difficult it may be to design API that allows for nice caching. (2) Assuming we use signatures, there is a memory management problem that 5cfb9d930f11e spells out. This is probably a matter of "just fix it", but I suppose another option is to take the GLSL or ARB architecture. -- April is the cruellest month, breeding Lilacs out of the dead land, mixing Memory and desire, stirring Dull roots with spring rain. -- v7: vkd3d-shader: Introduce a function to build a varying map between sm1 stages. vkd3d-shader/spirv: Make output varyings not consumed by the next stage private variables. vkd3d-shader: Implement remapping shader output registers to match the next shader's semantics. vkd3d-shader: Add a separate field for the target location of a signature element. vkd3d-shader/d3dbc: Make sure all inter-stage varyings have a unique register index. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/280

2 6

[PATCH 0/14] MR3477: winedump: Improve regf files handling
by Piotr Caban (＠piotr) Aug. 2, 2023

Aug. 2, 2023

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/3477

3 17

[PATCH v6 0/5] MR280: vkd3d-shader: sm1 semantic matching.
by Zebediah Figura (＠zfigura) Aug. 2, 2023

Aug. 2, 2023

This is probably the most ugly and controversial bit of API. I don't really know if this is the right approach to solving this. sm4 registers match by register index, such that shaders can mostly be compiled in isolation. sm1 does not—registers may be specified in any order in the vertex and pixel shaders, and will be matched by usage and usage index. By itself this is not much of a problem. Where it gets hairy is that we want to do some degree of caching, as well as pre-compilation, and avoid recompiling either shader every time it's matched with a new one. Wine currently deals with this problem, for GLSL, by generating a "main" GLSL shader for the vertex shader, and then an extra function setup_vs_output(), and linking the two together every time a new pixel shader is used. This could in theory be used for SPIR-V, but it requires the use of extra, probably external, code to link SPIR-V shaders together, which I do not particularly anticipate being well-received. (I'm not sure how Wine deals with this problem in the ARB backend. It seems to take the pixel shader signature into account when generating the vertex shader— cf. init_output_registers()—but it doesn't take it into account when looking up a vertex shader variant? I didn't look too closely at the code, so maybe I'm missing something.) -- The vkd3d parts of this patch are quite straightforward, and looking at them, I think the design is quite intuitive in isolation. There may be some room for internal refactoring (in particular with an eye to not so much overloading the "register_index" field of struct shader_signature_element) but I'm relatively happy with the way it turned out. In isolation, that is. The Wine part is worse. I've uploaded branches for vkd3d and Wine that use this API, and correctly handle shaders with some nontrivial reordering: https://gitlab.winehq.org/zfigura/vkd3d/-/commits/himavant5 https://gitlab.winehq.org/zfigura/wine/-/commits/himavant_cb The test can be run, as before, like so: make tests/shader_runner.exe && WINE_D3D_CONFIG=renderer=vulkan wine tests/shader_runner.exe ../vkd3d/tests/hlsl/sm1-interstage-interface.shader_test The interesting Wine parts are concentrated in a single patch, 5cfb9d930f11e. The patch takes a few shortcuts, partly because I wanted not to block the vkd3d API design questions, but also because while writing it I came up with a couple of problems that I wasn't sure how to fix. There are two main problems I see: (1) This patch has the user pass the signature from the pixel shader when compiling the vertex shader, and looks up register indices already arbitrarily allocated by the pixel shader. This is problematic when trying to use this signature as a cache key, by virtue of it not being clear (or even defined) which fields are key elements and which aren't. It's also not particularly kind to lookup, on account of not being directly comparable with memcmp(). There are a few options I see: (a) Provide an internal function to compare key elements. This feels... odd, like a very special-purpose function, but perhaps workable. (b) Just make the user deal with it, and assert that all fields are key elements. (c) Use some alternative, perhaps shortened structure as a field of vkd3d_shader_next_stage_info. This has the disadvantage that it is not as simple for a hypothetical user to retrieve from the pixel shader, but we would presumably provide a function to generate one from a shader signature. This would probably also be kinder to cache lookup if it's shorter. (d) Make caching vkd3d's responsibility, to some degree. This seems daunting, but the more we optimize, the more difficult it may be to design API that allows for nice caching. (2) Assuming we use signatures, there is a memory management problem that 5cfb9d930f11e spells out. This is probably a matter of "just fix it", but I suppose another option is to take the GLSL or ARB architecture. -- April is the cruellest month, breeding Lilacs out of the dead land, mixing Memory and desire, stirring Dull roots with spring rain. -- v6: vkd3d-shader: Introduce a function to build a varying map between sm1 stages. vkd3d-shader/spirv: Make output varyings not consumed by the next stage private variables. vkd3d-shader: Implement remapping shader output registers to match the next shader's semantics. vkd3d-shader: Add a separate field for the target location of a signature element. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/280

3 7

[PATCH v3 0/2] MR3456: comctl32: Add support for MSAA events on tab controls.
by Esme Povirk (＠madewokherd) Aug. 2, 2023

Aug. 2, 2023

-- v3: comctl32/tests: Test MSAA events for tab controls. https://gitlab.winehq.org/wine/wine/-/merge_requests/3456

3 3

[PATCH v4 0/2] MR3471: gdiplus: Reduce code redundancy.
by David Kahurani (＠kahurani) Aug. 1, 2023

Aug. 1, 2023

i) Introduce a helper for calculating transform properties. ii) Re-use get_log_fontW to populate LOGFONTW from GpFont. Signed-off-by: David Kahurani k.kahurani(a)gmail.com -- v4: gdiplus: Use get_log_fontW in GdipGetLogFontW gdiplus: Use helper to calculate transform properties https://gitlab.winehq.org/wine/wine/-/merge_requests/3471

3 3

[PATCH v3 0/1] MR3458: ddraw/tests: Fix overflow bugs in compare_float.
by Jeffrey Smith (＠whydoubt) Aug. 1, 2023

Aug. 1, 2023

1. Subtracting integers may result in an overflow or underflow. 2. Right at the 'edge' of underflowing, the result of subtraction may be `INT_MIN`, and the call to `abs()` will also result in `INT_MIN`. This fix accounts for all of these conditions. EXAMPLES: 1. can be encountered by comparing 2.0 and -2.0 2. can be encountered by comparing -2.0 and 2.0 NOTE: There are 14 more instances of `compare_float` across several modules. I would like to make sure the logic is sound with this MR, then I can take on the rest. -- v3: ddraw/tests: Fix overflow bugs in compare_float. https://gitlab.winehq.org/wine/wine/-/merge_requests/3458

5 8

[PATCH v3 0/15] MR3408: mshtml: Implement Cycle Collection for all mshtml objects (dispex), part 1.
by Gabriel Ivăncescu Aug. 1, 2023

Aug. 1, 2023

This is first part in a series that implements Cycle Collection for every mshtml object (dispex) and cleans up rest of the code based on it, which is obviously needed due to dynamic props and other extra object-specific refs. In an effort to split it up as much as possible, since it already has quite a lot of restructuring and changes, some of the earlier patches will introduce temporary leaks or cyclic refs, but that's because we'll later handle them properly with the dispex CC. These shouldn't affect behavior, though, so it shouldn't pose problems for functionality. Nodes are, initially, not changed much (other than to make it compatible with the dispex) to keep changes as small as possible. They still use their own CC mechanism and refcounting, which is hackish but that is solved in a follow-up MR, so it's temporary only. Eventually, every object (including nodes) will use the dispex's vtbl to do its Cycle Collection, except for stuff like outer window (which is a special case). In this first part, the objects that are using the node CC will have no-op dispex CC methods since they are using the node's, but this is temporary only. v2: Now the entire first part will be split up into several MRs. This only moves destruction/unlinking of dispex into separate vtbl methods, without actually using any of the CC yet. `release_dispex` will now call the unlink and destructor, if available (it won't be optional later, but for now it is), so only those converted objects make use of it. Most of the patches are small and typically convert one object at a time, except the first 4. More will follow up in subsequent MRs. -- v3: mshtml: Use unlink and destructor in the vtbl for the MutationObserver mshtml: Use unlink and destructor in the vtbl for HTMLXMLHttpRequestFactory. mshtml: Use unlink and destructor in the vtbl for HTMLOptionElementFactory. mshtml: Use unlink and destructor in the vtbl for HTMLImageElementFactory. mshtml: Use unlink and destructor in the vtbl for HTMLStyleSheet. mshtml: Use unlink and destructor in the vtbl for HTMLStyleSheetsCollection. mshtml: Use unlink and destructor in the vtbl for mshtml: Use unlink and destructor in the vtbl for HTMLStyleSheetRule. mshtml: Use unlink and destructor in the vtbl for CSSStyle. mshtml: Use unlink and destructor in the vtbl for inner windows. mshtml: Use unlink and destructor in the vtbl for HTMLEventObj. mshtml: Use separate dispex destructors for different event types. mshtml: Use unlink and destructor in the vtbl for function disps. mshtml: Introduce unlink_ref helper. mshtml: Use the common HTMLElement dispex vtbl in the dispex definitions. https://gitlab.winehq.org/wine/wine/-/merge_requests/3408

2 16