November 2023 - wine-gitlab

[PATCH 0/1] MR4432: dxva2/tests: Skip tests if D3D9 is unusable.
by Nikolay Sivov (＠nsivov) 19 Nov '23

19 Nov '23

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4432

2 1

[PATCH 0/2] MR4431: msvcrt: Improve _mbsrev implementation.
by Piotr Caban (＠piotr) 19 Nov '23

19 Nov '23

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/4431

2 2

[PATCH v2 0/4] MR3334: Add vccorlib140.dll stub dll.
by Etaash Mathamsetty (＠etaash.mathamsetty) 19 Nov '23

19 Nov '23

When using native xinput in WRC9, vccorlib is needed to make sure xinput doesn't crash. -- v2: wincorlib: Add stub for platform_details_uninit_data. wincorlib: Add stub for platform_details_init_data. vccorlib140: Add stub dll. wincorlib: Add stub dll. https://gitlab.winehq.org/wine/wine/-/merge_requests/3334

3 8

[PATCH v3 0/3] MR4305: server:setting range64 with checking
by Fan WenJie (＠fanwenjie) 19 Nov '23

19 Nov '23

Signed-off-by: Fan WenJie <fanwj(a)mail.ustc.edu.cn> -- v3: tools/makedep: add *.lib to .gitignore ntdll:setting limit with checking server:setting range64 with checking https://gitlab.winehq.org/wine/wine/-/merge_requests/4305

3 4

[PATCH v2 0/1] MR4394: uxtheme: Add stub for RefreshImmersiveColorPolicyState.
by Bernhard Übelacker (＠bernhardu) 19 Nov '23

19 Nov '23

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=55898 -- v2: uxtheme: Add stub for RefreshImmersiveColorPolicyState. https://gitlab.winehq.org/wine/wine/-/merge_requests/4394

3 3

[PATCH 0/2] MR4429: evr/presenter: Handle failure to initialize D3D9.
by Nikolay Sivov (＠nsivov) 18 Nov '23

18 Nov '23

Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4429

3 3

[PATCH 0/1] MR4427: windowscodecs: Enable WICPixelFormat32bppBGRA in BMP encoder.
by Paul Gofman (＠gofman) 18 Nov '23

18 Nov '23

Looks like it is supported on Windows at least since Vista. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4427

3 2

[PATCH 0/3] MR4417: Msxml default ns
by Daniel Lehman (＠dlehman25) 18 Nov '23

18 Nov '23

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/4417

3 4

[PATCH 0/1] MR4426: evr: Set last presented sample atomically.
by Rémi Bernon (＠rbernon) 18 Nov '23

18 Nov '23

Fixes a race condition and crashes in Secret of Mana. The queue critical section is held in video_presenter_sample_present while GetCurrentImage only locks the presenter CS. Double locking is tricky and atomic operation is appropriate to swap the sample pointer. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4426

3 2

Re: [PATCH v2 0/1] MR4152: dnsapi: Handle IP address as DNS name in DnsQuery_UTF8().
by Jinoh Kang (＠iamahuman) 18 Nov '23

18 Nov '23

On Sat Oct 21 15:00:08 2023 +0000, Jinoh Kang wrote: > > Regarding dropping the null check, I think the common rule in Wine is > to check the allocations. I'd personally would agree that in many cases > just skipping the check and crashing would make more sense than putting > code for any handling the error just anyhow, but I am just following > what I think is the rule here. > Well, that rule sounds absurd and contradictory. > 1. We should check for NULL. > 2. Except NULL doesn't happen much in practice, so we shouldn't really > check for NULL anyway. > Correct me if I'm wrong, I don't think such rule (pretend to check for > NULL) has ever been in enforcement for upstream patches. In fact, I've > got https://gitlab.winehq.org/wine/wine/-/merge_requests/3324 upstreamed > before; not sure if this generalizes to this case too. > Otherwise, is `assert()`-ing an option? > > or, in case of miraculous getting memory back, > This could be more probable than you think. For example, another thread > might momentarily allocate a large amount of memory (towards RSS/VM > ulimit) and then free it. If we race, we'll get a spontaneous > allocation failure. > > it will get an error anyway being unable to find it (both errors, > ERROR_OUT_OF_MEMORY or what it would get, are not the correct result anyway). > Depends on how the app handles the error; for example, it could retry on > ERROR_OUT_OF_MEMORY but not whatever NXDOMAIN is. Also, the behavior > (3) has additional side effect of potentially generating a packet over a > network, in a program that would not otherwise emit any packets (it only > resolves "127.0.0.1"). > Granted, you're not making the rules here, and existing wine code > doesn't strictly abide by graceful allocation failure handling either. > I was merely puzzled by how such confusing rule is in place. For future reference: NULL check has been dropped in master. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/4152#note_52757

1 0

2025

2024

2023

2022

wine-gitlab November 2023