November 2023 - Wine-GitLab

[PATCH 0/7] MR4380: mshtml: Keep inner window's link to the outer window until it's destroyed.
by Gabriel Ivăncescu Nov. 13, 2023

Nov. 13, 2023

Here's a list of usages of `base.outer_window` that may be impacted by the change and my analysis/fixes. Note that I specifically did not look for `document->outer_window` since that one was already holding it forever, so it's out of scope for this MR (whether it's correct or not, the behavior is the same after this MR). I also didn't consider some cases that already **did not check** for it being NULL or not, because they would have crashed earlier before anyway (they'd still crash now if outer window gets killed, but it's out of scope of the MR as I said). For those that says "Handled" it means I've added checks so that the behavior is same as before, i.e. detached inner windows behave as if outer window is NULL. * `htmldoc.c/HTMLDocument_get_mimeType`: Changed so it doesn't rely on outer window. * `htmldoc.c/HTMLDocument7_get_defaultView`: This is actually fixed by these patches. * `htmlform.c/HTMLFormElement_submit`: Handled. * `htmlstorage.c/send_storage_event`: Fixed in a separate patch with tests, since it follows the outer window. * `htmlwindow.c/check_target_origin`: This (along with many other methods) can't have it NULL anymore or detached, since the caller must hold a ref to the outer window, and they're not called from our objects holding ref to inner window only. * `htmlwindow.c/HTMLWindowSP_QueryService`: This uses the outer window to obtain the browser and the doc obj, so it's fine. * `mutation.c/set_document_mode`: Handled. * `navigate.c/BindCallbackRedirect_Redirect`: Handled. * `navigate.c/nsAsyncVerifyRedirectCallback_OnRedirectVerifyCallback`: Handled. * `navigate.c/nsChannelBSC_init_bindinfo`: Uses the outer window to obtain the browser and doc obj to set a flag, so it's harmless and probably correct this way. * `navigate.c/handle_navigation_error`: Handled. * `navigate.c/handle_extern_mime_navigation`: Handled. * `omnavigator.c/OmHistory_get_length`: This is actually fixed by these patches. * `script.c/ActiveScriptSite_GetItemInfo`: Handled. * `script.c/ActiveScriptSiteWindow_GetWindow`: Handled. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4380

2 8

[PATCH v8 0/1] MR3931: windows.ui: Support accent colors in uisettings3_GetColorValue.
by Myah Caron (＠qsniyg) Nov. 13, 2023

Nov. 13, 2023

Fixes https://bugs.winehq.org/show_bug.cgi?id=55640. -- v8: windows.ui: Support accent colors in uisettings3_GetColorValue. https://gitlab.winehq.org/wine/wine/-/merge_requests/3931

4 4

[PATCH 0/4] MR4378: Extend CreateProcess tests for handle inheritance.
by eric pouech (＠epo) Nov. 13, 2023

Nov. 13, 2023

This serie adds a couple of test for CreateProcess around std handles inheritance, depending on several options (cui/gui, use (or no use) of STARTF_USESTDHANDLES, type of handles...). This requires revisiting the existing tests to allow for more flexibility in implementing the new tests (and potentially others if needed). -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4378

2 4

Re: [PATCH v5 0/3] MR4243: rtworkq: free cancelled work items
by Yuxuan Shui (＠yshui) Nov. 13, 2023

Nov. 13, 2023

On Mon Nov 13 14:56:51 2023 +0000, Yuxuan Shui wrote: > > not supported > not sure what you mean by that? I can build it fine with mingw-w64-g++ > and it runs. > > cancellation does not work > this is not true. callback could still be called after MFCancelWorkItem > returns success, the > [documentation](https://learn.microsoft.com/en-us/windows/win32/api/mfapi/nf… > says as much. > and the thing is, more than this "hit race" case, our current rtworkq > implementation triggers a assert in threadpool that says exactly what i > predicted (callback is running when you try to close it). and that's > exactly the race I am trying to avoid here. > what more proof do you want? i hope you understand the problem here. the problem is not the assert, so doesn't matter if the TP implementation handles this case. the problem is if the `scheduled_item_cancelable_callback` is running when we reaches `queue_cancel_item`, we will double release the work item. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4243#note_52137

1 0

Re: [PATCH v5 0/3] MR4243: rtworkq: free cancelled work items
by Yuxuan Shui (＠yshui) Nov. 13, 2023

Nov. 13, 2023

> not supported not sure what you mean by that? I can build it fine with mingw-w64-g++ and it runs. > cancellation does not work this is not true. callback could still be called after MFCancelWorkItem returns success, the [documentation](https://learn.microsoft.com/en-us/windows/win32/api/mfapi/nf… says as much. and the thing is, more than this "hit race" case, our current rtworkq implementation triggers a assert in threadpool that says exactly what i predicted (callback is running when you try to close it). and that's exactly the race I am trying to avoid here. what more proof do you want? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4243#note_52135

1 0

[PATCH v9 0/5] MR3572: mf: Make session_get_node_object() more robust.
by Zhiyi Zhang (＠zhiyi) Nov. 13, 2023

Nov. 13, 2023

It's possible that a state object pointer not in the topology node collection gets passed to session_get_node_object(). Instead of returning the last node when the object is not found, we should return a NULL so that the state of the last node is not changed by mistake. -- v9: mf/tests: Test IMFMediaSession::Start(). mf: Support seeking while a session is running. mf/tests: Add a create_test_topology() helper. mf: Add a session_flush_nodes() helper. mf: Make session_get_node_object() more robust. https://gitlab.winehq.org/wine/wine/-/merge_requests/3572

4 9

[PATCH v9 0/6] MR459: vkd3d-shader/dxil: Implement the DXIL VSELECT instruction.
by Conor McCarthy (＠cmccarthy) Nov. 13, 2023

Nov. 13, 2023

-- v9: vkd3d-shader/spirv: Avoid invalid bool-to-bool conversion in spirv_compiler_emit_movc(). vkd3d-shader/dxil: Implement the DXIL VSELECT instruction. tests/shader-runner: Add tests for 64-bit casts. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/459

3 9

[PATCH v8 0/6] MR459: vkd3d-shader/dxil: Implement the DXIL VSELECT instruction.
by Conor McCarthy (＠cmccarthy) Nov. 13, 2023

Nov. 13, 2023

-- v8: vkd3d-shader/spirv: Avoid invalid bool-to-bool conversion in spirv_compiler_emit_movc(). vkd3d-shader/dxil: Implement the DXIL VSELECT instruction. tests/shader-runner: Add tests for 64-bit casts. vkd3d-shader/spirv: Return an error if an invalid handler is encountered. tests/shader-runner: Add a test for float comparisons. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/459

3 10

[PATCH 0/1] MR466: vkd3d-shader: Implement scanning combined resource/sampler information.
by Henri Verbeet (＠hverbeet) Nov. 13, 2023

Nov. 13, 2023

-- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/466

3 6

Re: [PATCH v2 0/7] MR4375: dmsynth: Various fixes 2
by Rémi Bernon Nov. 13, 2023

Nov. 13, 2023

Rémi Bernon (@rbernon) commented about dlls/dmsynth/synth.c: > struct wave *wave; > }; > > +struct fixup > +{ > + struct list entry; > + int dst; > + int src; > + int flags; > + > + int fixup_dst; > + float amount; > +}; > + Instead of a dynamic list, could we have just a simple structure like: ```c struct gen_increments { float env_hold; float env_decay; float vol_hold; float vol_decay; }; ``` Which we would pass around down to `add_mod_from_connection` which would set the needed adjustments (it's actually only needed there, `set_gen_from_connection` won't have to add increments), and then after the loops we would call `fluid_voice_gen_incr` for each of these? Or do we need the generic version for more fixups that are not in this MR? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4375#note_52124

1 0