January 2026 - Wine-GitLab - list.winehq.org

Re: [PATCH v3 0/1] MR9719: mfmediaengine/tests: Test we receive a frame before calling Play.
by Nikolay Sivov (＠nsivov) Jan. 15, 2026

Jan. 15, 2026

Nikolay Sivov (@nsivov) commented about dlls/mfmediaengine/tests/mfmediaengine.c: > > test_audio_session(FALSE); > > + /* free device context and readback texture */ It's obvious without a comment. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9719#note_127083

1 0

Re: [PATCH v3 0/1] MR9719: mfmediaengine/tests: Test we receive a frame before calling Play.
by Nikolay Sivov (＠nsivov) Jan. 15, 2026

Jan. 15, 2026

Nikolay Sivov (@nsivov) commented about dlls/mfmediaengine/tests/mfmediaengine.c: > + desc.Usage = D3D11_USAGE_STAGING; > + desc.BindFlags = 0; > + desc.CPUAccessFlags = D3D11_CPU_ACCESS_READ; > + desc.MiscFlags = 0; > + hr = ID3D11Device_CreateTexture2D(device, &desc, NULL, &rb_texture); > + ok(hr == S_OK, "Unexpected hr %#lx.\n", hr); > + > + /* confirm we have a frame available before calling play */ > + pts = 0; > + hr = IMFMediaEngineEx_OnVideoStreamTick(media_engine, &pts); > + todo_wine > + ok(hr == S_OK, "Unexpected hr %#lx.\n", hr); > + todo_wine > + ok(pts == 0, "Unexpected timestamp.\n"); > + > + /* confirm we can transfer a frame before calling play */ Does it mean that setting url/media item is enough to start the pipeline internally? Any indication if it's paused immediately, or something else is going on? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9719#note_127082

1 0

Re: [PATCH v3 0/1] MR9719: mfmediaengine/tests: Test we receive a frame before calling Play.
by Nikolay Sivov (＠nsivov) Jan. 15, 2026

Jan. 15, 2026

Nikolay Sivov (@nsivov) commented about dlls/mfmediaengine/tests/mfmediaengine.c: > + hr = IMFMediaEngineEx_TransferVideoFrame(notify->media_engine, (IUnknown *)texture, NULL, &dst_rect, NULL); > + ok(hr == S_OK, "Unexpected hr %#lx.\n", hr); > + > + ID3D11DeviceContext_CopySubresourceRegion(context, (ID3D11Resource *)rb_texture, > + 0, 0, 0, 0, (ID3D11Resource *)texture, 0, NULL); > + > + memset(&map_desc, 0, sizeof(map_desc)); > + hr = ID3D11DeviceContext_Map(context, (ID3D11Resource *)rb_texture, 0, D3D11_MAP_READ, 0, &map_desc); > + ok(hr == S_OK, "Unexpected hr %#lx.\n", hr); > + ok(!!map_desc.pData, "got pData %p\n", map_desc.pData); > + ok(map_desc.DepthPitch == 16384, "got DepthPitch %u\n", map_desc.DepthPitch); > + ok(map_desc.RowPitch == desc.Width * 4, "got RowPitch %u\n", map_desc.RowPitch); > + res = check_rgb32_data(L"rgb32frame.bmp", map_desc.pData, map_desc.RowPitch * desc.Height, &dst_rect); > + todo_wine > + ok(res == 0, "Unexpected %lu%% diff\n", res); > + ID3D11DeviceContext_Unmap(context, (ID3D11Resource *)rb_texture, 0); This should probably go to some helper. Are we going to use same pattern elsewhere? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9719#note_127081

1 0

Re: [PATCH v7 0/13] MR9489: mf: Fix seeking within sample grabber.
by Nikolay Sivov (＠nsivov) Jan. 15, 2026

Jan. 15, 2026

Could you move tests to a separate MR? The commit list is pretty long here. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9489#note_127079

1 0

Re: [PATCH v7 0/13] MR9489: mf: Fix seeking within sample grabber.
by Nikolay Sivov (＠nsivov) Jan. 15, 2026

Jan. 15, 2026

Nikolay Sivov (@nsivov) commented about dlls/mf/tests/mf.c: > + todo_wine > + CHECK_CALLED(timer_SetTimer); > > - hr = IMFPresentationClock_Pause(clock); > + hr = IMFClockStateSink_OnClockPause(clock_sink, 0); > ok(hr == S_OK, "Unexpected hr %#lx.\n", hr); > > - hr = IMFPresentationClock_Start(clock, 0); > + /* test over supply */ > + supply_samples(stream, 6); > + > + SET_EXPECT(timer_CancelTimer); > + hr = IMFClockStateSink_OnClockStart(clock_sink, 0, 0); > ok(hr == S_OK, "Unexpected hr %#lx.\n", hr); > + CHECK_CALLED(timer_CancelTimer); > Normal flow is to call clock->Start(), callbacks are called internally. Why changing that? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9489#note_127078

1 0

Re: [PATCH v7 0/13] MR9489: mf: Fix seeking within sample grabber.
by Nikolay Sivov (＠nsivov) Jan. 15, 2026

Jan. 15, 2026

Nikolay Sivov (@nsivov) commented about dlls/mf/tests/mf.c: > /* Set clock. */ > - hr = MFCreatePresentationClock(&clock); > - ok(hr == S_OK, "Failed to create clock object, hr %#lx.\n", hr); > + mock_clock = create_presentation_clock(); > + clock = &mock_clock->IMFPresentationClock_iface; > > hr = IMFMediaSink_SetPresentationClock(sink, clock); > ok(hr == S_OK, "Failed to set presentation clock, hr %#lx.\n", hr); > - > - hr = MFCreateSystemTimeSource(&time_source); > - ok(hr == S_OK, "Failed to create time source, hr %#lx.\n", hr); > - > - hr = IMFPresentationClock_SetTimeSource(clock, time_source); > - ok(hr == S_OK, "Failed to set time source, hr %#lx.\n", hr); > - IMFPresentationTimeSource_Release(time_source); > + ok(!!mock_clock->sample_grabber_clock_state_sink, "AddClockStateSink not called\n"); I would remove this one. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9489#note_127077

1 0

Re: [PATCH v7 0/13] MR9489: mf: Fix seeking within sample grabber.
by Nikolay Sivov (＠nsivov) Jan. 15, 2026

Jan. 15, 2026

Nikolay Sivov (@nsivov) commented about dlls/mf/tests/mf.c: > + samples_requested = count_samples_requested(stream); > + todo_wine > + ok(samples_requested == 4, "Unexpected number of samples requested %d\n", samples_requested); > + > + /* test number of new sample requests on seek whilst paused and 3 samples provided */ > + supply_samples(stream, 3); > + > + hr = IMFPresentationClock_Pause(clock); > + ok(hr == S_OK, "Unexpected hr %#lx.\n", hr); > + > + hr = IMFPresentationClock_Start(clock, 0); > + ok(hr == S_OK, "Unexpected hr %#lx.\n", hr); > + > + samples_requested = count_samples_requested(stream); > + todo_wine > + ok(samples_requested == 3, "Unexpected number of samples requested %d\n", samples_requested); In this sequence, how could you tell that requests are coming on Pause -> Start transition (is that what we are trying to test?)? Does it rely on a fact that we pushed fewer than requested? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9489#note_127076

1 0

Re: [PATCH v7 0/13] MR9489: mf: Fix seeking within sample grabber.
by Nikolay Sivov (＠nsivov) Jan. 15, 2026

Jan. 15, 2026

Nikolay Sivov (@nsivov) commented about dlls/mf/tests/mf.c: > + > + hr = MFCreateMediaType(&media_type); > + ok(hr == S_OK, "Failed to create media type, hr %#lx.\n", hr); > + > + hr = IMFMediaType_SetGUID(media_type, &MF_MT_MAJOR_TYPE, &MFMediaType_Audio); > + ok(hr == S_OK, "Failed to set attribute, hr %#lx.\n", hr); > + hr = IMFMediaType_SetGUID(media_type, &MF_MT_SUBTYPE, &MFAudioFormat_PCM); > + ok(hr == S_OK, "Failed to set attribute, hr %#lx.\n", hr); > + > + EXPECT_REF(media_type, 1); > + hr = MFCreateSampleGrabberSinkActivate(media_type, grabber_callback, &activate); > + ok(hr == S_OK, "Failed to create grabber activate, hr %#lx.\n", hr); > + EXPECT_REF(media_type, 2); > + > + ref = IMFMediaType_Release(media_type); > + ok(ref == 1, "Release returned %ld\n", ref); Same here, it's unnecessary to check counter here. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9489#note_127075

1 0

Re: [PATCH v7 0/13] MR9489: mf: Fix seeking within sample grabber.
by Nikolay Sivov (＠nsivov) Jan. 15, 2026

Jan. 15, 2026

Nikolay Sivov (@nsivov) commented about dlls/mf/tests/mf.c: > + > + hr = MFStartup(MF_VERSION, MFSTARTUP_FULL); > + ok(hr == S_OK, "Failed to start up, hr %#lx.\n", hr); > + > + hr = MFCreateMediaType(&media_type); > + ok(hr == S_OK, "Failed to create media type, hr %#lx.\n", hr); > + > + hr = IMFMediaType_SetGUID(media_type, &MF_MT_MAJOR_TYPE, &MFMediaType_Audio); > + ok(hr == S_OK, "Failed to set attribute, hr %#lx.\n", hr); > + hr = IMFMediaType_SetGUID(media_type, &MF_MT_SUBTYPE, &MFAudioFormat_PCM); > + ok(hr == S_OK, "Failed to set attribute, hr %#lx.\n", hr); > + > + EXPECT_REF(media_type, 1); > + hr = MFCreateSampleGrabberSinkActivate(media_type, grabber_callback, &activate); > + ok(hr == S_OK, "Failed to create grabber activate, hr %#lx.\n", hr); > + EXPECT_REF(media_type, 2); We already have this refcount check in another test, no need to duplicate it. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9489#note_127074

1 0

[PATCH v7 0/5] MR9646: mf/tests: Scrubbing tests
by Brendan McGrath (＠redmcg) Jan. 15, 2026

Jan. 15, 2026

This MR adds tests for how scrubbing and rate changes work within the Media Session. -- v7: mf/tests: Check that Rate change in PLAY state is ignored. mf/tests: Test that there is no pre-roll during scrubbing. https://gitlab.winehq.org/wine/wine/-/merge_requests/9646

3 11