November 2022 - Wine-gitlab

[PATCH v16 0/1] MR672: comdlg32: Common item dialog fixes
by Ben Cottrell (＠bgc) 15 Nov '22

15 Nov '22

Made changes to comdlg32/itemdlg.c to add missing controls, which are the address bar and the up button. Updated bitmaps in comctl32 and SVG files for history icons. My changes help to resolve this bug: https://bugs.winehq.org/show_bug.cgi?id=50338. -- v16: comdlg32: Add missing controls and their functionality to IFileDialog, or the common item dialog. https://gitlab.winehq.org/wine/wine/-/merge_requests/672

4 4

Re: [PATCH v3 0/4] MR1311: wmvcore: Implement DedicatedDeliveryThread async reader output setting.
by Zebediah Figura (＠zfigura) 15 Nov '22

15 Nov '22

Zebediah Figura (@zfigura) commented about dlls/wmvcore/async_reader.c: > struct stream *stream = arg; > struct async_reader *reader = stream->reader; > struct sample *sample; > + WCHAR buffer[256]; > HRESULT hr; > > TRACE("reader %p, number %u\n", reader, stream->number); > > + swprintf(buffer, ARRAY_SIZE(buffer), L"wine_wmreader_stream_%u_read", stream->number); > + SetThreadDescription(GetCurrentThread(), buffer); Linux cuts off thread names at 15 characters (not counting the null terminator), which somewhat unfortunately makes all of these identical to each other and the deliver threads. Thus far I think Brendan had been trying to make the thread names distinct, if not per se trying to fit the entire name within 15 characters. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1311#note_15868

1 0

Re: [PATCH v3 0/4] MR1311: wmvcore: Implement DedicatedDeliveryThread async reader output setting.
by Zebediah Figura (＠zfigura) 15 Nov '22

15 Nov '22

> We need the read threads to always be running and returning their next sample because we then deliver them in PTS order. Somewhat out of curiosity, did you check whether this still applies when DedicatedDeliveryThread is set? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1311#note_15869

1 0

Re: [PATCH v3 0/4] MR1311: wmvcore: Implement DedicatedDeliveryThread async reader output setting.
by Zebediah Figura (＠zfigura) 15 Nov '22

15 Nov '22

On Mon Nov 14 12:05:16 2022 +0000, Rémi Bernon wrote: > I don't think we need to check it here anymore with the latest version. > Deselected streams will never have pending read requests (or they will > shortly complete it after deselection, with either error or a last sample). I guess, but it depends on the fact that "read_result" is zero-initialized and never written, which seems non-obvious and a little fragile. In fact, I think it can already break if a stream was selected and then later becomes deselected, even if the filter was stopped in between. Checking "stream->read_requested" would be more obvious but I think still broken for the same reasons (but simple enough to fix.) Note that if you do that then read_result becomes superfluous and doesn't really need to be stored anywhere. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1311#note_15867

1 0

Re: [PATCH v3 0/4] MR1311: wmvcore: Implement DedicatedDeliveryThread async reader output setting.
by Zebediah Figura (＠zfigura) 15 Nov '22

15 Nov '22

Zebediah Figura (@zfigura) commented about dlls/wmvcore/async_reader.c: > + > + LeaveCriticalSection(&reader->callback_cs); > + hr = IWMSyncReader2_GetNextSample(reader->reader, stream->number, > + &sample->buffer, &sample->pts, &sample->duration, > + &sample->flags, &sample->output, &sample->stream); > + EnterCriticalSection(&reader->callback_cs); > + } > + > + if (SUCCEEDED(stream->read_result = hr)) > + { > + TRACE("Got stream %u buffer with pts %I64d.\n", stream->number, sample->pts); > + stream->next_sample = sample; > + } > + else > + { > + WARN("Failed to get stream %u sample, hr %#lx.\n", stream->number, stream->read_result); Missed this the first time, but: I'd make this an ERR, personally. (An application can trigger it, but it's one of those cases where it's a sign that something went catastrophically wrong, and also we're pretty much guessing how to handle it if not.) -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1311#note_15865

1 0

Re: [PATCH v3 0/4] MR1311: wmvcore: Implement DedicatedDeliveryThread async reader output setting.
by Zebediah Figura (＠zfigura) 15 Nov '22

15 Nov '22

Zebediah Figura (@zfigura) commented about dlls/wmvcore/async_reader.c: > } > > if (!first_sample) > - return NS_E_NO_MORE_SAMPLES; > + return pending ? E_PENDING : NS_E_NO_MORE_SAMPLES; What's the point of this? We never check the return code of async_reader_get_next_sample(), only whether it FAILED(). -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1311#note_15866

1 0

Re: [PATCH v3 0/4] MR1311: wmvcore: Implement DedicatedDeliveryThread async reader output setting.
by Zebediah Figura (＠zfigura) 15 Nov '22

15 Nov '22

Zebediah Figura (@zfigura) commented about dlls/wmvcore/async_reader.c: > + continue; > + } > + > + while (stream->read_requested) > + { > + stream->read_requested = false; > + > + if (sample->buffer) > + INSSBuffer_Release(sample->buffer); > + sample->buffer = NULL; > + > + LeaveCriticalSection(&reader->callback_cs); > + hr = IWMSyncReader2_GetNextSample(reader->reader, stream->number, > + &sample->buffer, &sample->pts, &sample->duration, > + &sample->flags, &sample->output, &sample->stream); > + EnterCriticalSection(&reader->callback_cs); Similarly to the earlier question about IWMSyncReader locking, do we need to drop the CS here? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1311#note_15864

1 0

[PATCH v2 0/4] MR1355: opengl32: Implement wow64 thunk for several unix entry points.
by Rémi Bernon 15 Nov '22

15 Nov '22

-- v2: opengl32: Implement wow64 thunk for wglMakeCurrent (et al.). opengl32: Implement wow64 thunk for wglGetProcAddress. opengl32: Manually write glPathGlyphIndexRangeNV wow64 thunk. opengl32: Generate wow64 thunks. https://gitlab.winehq.org/wine/wine/-/merge_requests/1355

2 6

[PATCH 0/1] MR1369: d3dx9: Add missing path conversion call in D3DXLoadVolumeFromFileA().
by Nikolay Sivov (＠nsivov) 15 Nov '22

15 Nov '22

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/1369

3 3

[PATCH 0/1] MR1351: d3dx9_36/tests: Use standard C functions for memory allocation in mesh.c.
by Alex Henrie (＠alexhenrie) 15 Nov '22

15 Nov '22

The big win here is getting rid of the reimplementation of strdup. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/1351

3 2