June 2024 - Wine-gitlab - list.winehq.org

[PATCH v3 0/1] MR5351: wined3d: Compile the clear compute shaders with D3DCompile().
by Giovanni Mascellani (＠giomasce) 27 Jun '24

27 Jun '24

-- v3: wined3d: Compile the clear compute shaders at runtime. https://gitlab.winehq.org/wine/wine/-/merge_requests/5351

4 3

[PATCH v6 0/3] MR5923: winegstreamer: Use affinity to determine 'max-threads' value.
by Brendan McGrath (＠redmcg) 27 Jun '24

27 Jun '24

GStreamer uses _SC_NPROCESSORS_CONF to determine 'max-threads'. On the Steam Deck, this is configured to be 16 (which is double its number of logical cores). _SC_NPROCESSORS_CONF also disregards a process's CPU affinity, thus it can create more threads than is useful, which ultimately wastes memory resources. Using affinity to set 'max-threads' addresses both these problems. -- v6: winegstreamer: Set MAX_THREADS to 4 for i386. winegstreamer: Use thread_count to determine 'max-threads' value. winegstreamer: Use process affinity to calculate thread_count. https://gitlab.winehq.org/wine/wine/-/merge_requests/5923

3 5

[PATCH 0/1] MR5941: xcopy: Handle switch options concatenated with path.
by Piotr Caban (＠piotr) 27 Jun '24

27 Jun '24

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/5941

2 1

[PATCH 0/7] MR5930: win32u: Factor surface rect computation out of the drivers.
by Rémi Bernon 27 Jun '24

27 Jun '24

This clips surface rects which are larger than the virtual screen, keeping full window surfaces for one which are smaller. This makes sure that off-screen windows have valid pixels on overviews like macOS Exposé, or similarly, GNOME shell activities view. Supersedes both https://gitlab.winehq.org/wine/wine/-/merge_requests/5865 and https://gitlab.winehq.org/wine/wine/-/merge_requests/5878. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5930

2 8

Re: [PATCH v5 0/3] MR5923: winegstreamer: Use affinity to determine 'max-threads' value.
by Rémi Bernon 27 Jun '24

27 Jun '24

Rémi Bernon (@rbernon) commented about dlls/winegstreamer/unixlib.c: > return true; > } > > +static ULONG popcount( ULONG val ) > +{ > +#if defined(__MINGW32__) > + return __builtin_popcount( val ); > +#else > + val -= val >> 1 & 0x55555555; > + val = (val & 0x33333333) + (val >> 2 & 0x33333333); > + return ((val + (val >> 4)) & 0x0f0f0f0f) * 0x01010101 >> 24; > +#endif > +} > + Well, now the __builtin_popcount path is never taken, because on the unix side it's not built with MinGW. There's a compiler check available already though. (Also mind the style change to match winegstreamer) ```suggestion:-10+0 static ULONG popcount(ULONG val) { #if HAVE___BUILTIN_POPCOUNT return __builtin_popcount(val); #else val -= val >> 1 & 0x55555555; val = (val & 0x33333333) + (val >> 2 & 0x33333333); return ((val + (val >> 4)) & 0x0f0f0f0f) * 0x01010101 >> 24; #endif } ``` -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5923#note_74516

1 0

Re: [PATCH v5 0/3] MR5923: winegstreamer: Use affinity to determine 'max-threads' value.
by Rémi Bernon 27 Jun '24

27 Jun '24

Rémi Bernon (@rbernon) commented about dlls/winegstreamer/unixlib.c: > return STATUS_UNSUCCESSFUL; > } > > + if (SUCCEEDED(NtQueryInformationProcess( GetCurrentProcess(), > + ProcessAffinityMask, &process_mask, sizeof(process_mask), NULL ))) > + thread_count = popcount(process_mask); > + else > + thread_count = 0; ```suggestion:-4+0 if (!NtQueryInformationProcess(GetCurrentProcess(), ProcessAffinityMask, &process_mask, sizeof(process_mask), NULL))) thread_count = popcount(process_mask); else thread_count = 0; ``` SUCCEEDED is for HRESULT, Nt API returns NTSTATUS and we often just check that it's 0. Also note the style fixes (no space in paren here, continuation indent is 8 spaces). -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5923#note_74517

1 0

[PATCH 0/1] MR5939: po: Update Lithuanian translation.
by Aurimas Fišeras 27 Jun '24

27 Jun '24

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/5939

1 1

Re: [PATCH v5 0/6] MR5870: mmdevapi: Implement loopback capture with winepulse backend.
by Davide Beatrici (＠davidebeatrici) 27 Jun '24

27 Jun '24

On Thu Jun 27 01:41:40 2024 +0000, Davide Beatrici wrote: > Agreed. We may eventually want to make use of something like [clang-format](https://clang.llvm.org/docs/ClangFormat.html). https://gitlab.winehq.org/wine/wine/-/merge_requests/5938 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5870#note_74514

1 0

[PATCH 0/1] MR5938: Draft: Add clang-format configuration file
by Davide Beatrici (＠davidebeatrici) 27 Jun '24

27 Jun '24

I tried to follow https://wiki.winehq.org/Wine_Developer%27s_Guide/Coding_Practice as closely as possible. Known issues: - No break before brace for assignments. Upstream issue: https://github.com/llvm/llvm-project/issues/79054 - `ContinuationIndentWidth` affects everything. Instead, we want it to be 8 for function arguments and 4 for assignments. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5938

2 1

Re: [PATCH v3 0/1] MR5909: comctl32/listview: Prevent LISTVIEW_GetNextItem failure after checking last item state in list
by Zhiyi Zhang (＠zhiyi) 27 Jun '24

27 Jun '24

> It fails on asserting return value of sending LVM_GETNEXTITEM message In that case, please also add a test checking the same thing. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5909#note_74512

1 0