February 2023 - Wine-gitlab

[PATCH 0/1] MR2083: urlmon: Use the _recalloc function instead of reimplementing it.
by Alex Henrie (＠alexhenrie) 01 Feb '23

01 Feb '23

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/2083

3 2

[PATCH v3 0/4] MR2036: comctl32: Always use unicode messages for subclass procedures.
by Rémi Bernon 01 Feb '23

01 Feb '23

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=43073 -- v3: comctl32: Skip restoring window proc if it has been modified. comctl32: Always use unicode messages for subclass procedures. https://gitlab.winehq.org/wine/wine/-/merge_requests/2036

3 8

[PATCH 0/1] MR2086: Warn that a libnss plugin may be missing if getaddrinfo() returns EBUSY.
by Francois Gouget (＠fgouget) 01 Feb '23

01 Feb '23

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/2086

2 1

[PATCH v2 0/3] MR2080: winegstreamer: GetInputStatus for h264, aac and MF_SA_D3D11_AWARE for h264.
by Paul Gofman (＠gofman) 01 Feb '23

01 Feb '23

Needed for Gungrave G.O.R.E. (besides some bits in raw AAC handling). The game is fine with GetInputStatus always returning MFT_INPUT_STATUS_ACCEPT_DATA for both h264 and aac but I think it is better to be correct here. -- v2: winegstreamer: Set MF_SA_D3D11_AWARE attribute for h264 transform. winegstreamer: Implement _GetInputStatus() for aac decoder transform. https://gitlab.winehq.org/wine/wine/-/merge_requests/2080

3 5

[PATCH v3 0/2] MR76: vkd3d-shader/hlsl: Support rsqrt() and distance() intrinsics.
by Nikolay Sivov (＠nsivov) 01 Feb '23

01 Feb '23

-- v3: vkd3d-shader/hlsl: Support distance() intrinsic. vkd3d-shader/hlsl: Support rsqrt() intrinsic. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/76

2 2

Re: [PATCH v3 0/2] MR2059: combase: Generate apartment OXID in rpcss
by Huw Davies (＠huw) 01 Feb '23

01 Feb '23

Huw Davies (@huw) commented about programs/rpcss/rpcss_main.c: > return S_OK; > } > > +/* Generates a new OXID suitable for use by an apartment. > + * Each call to this function returns a distinct OXID. > + * This function should not be invoked directly - use rpcss_get_new_apartment_oxid > + * instead. */ > +HRESULT __cdecl irpcss_get_new_apartment_oxid(handle_t h, OXID *oxid) > +{ > + HRESULT status; > + ULONG oxid_upper_bits; > + > + static LONG oxid_lower_bits; > + > + status = I_RpcBindingInqLocalClientPID(h, &oxid_upper_bits); > + ERR("Got client PID: %ld", oxid_upper_bits); This looks like it's left-over from debugging. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/2059#note_22593

1 0

Re: [PATCH v3 0/2] MR2059: combase: Generate apartment OXID in rpcss
by Huw Davies (＠huw) 01 Feb '23

01 Feb '23

Huw Davies (@huw) commented about programs/rpcss/rpcss_main.c: > } > > +/* Generates a new OXID suitable for use by an apartment. > + * Each call to this function returns a distinct OXID. > + * This function should not be invoked directly - use rpcss_get_new_apartment_oxid > + * instead. */ > +HRESULT __cdecl irpcss_get_new_apartment_oxid(handle_t h, OXID *oxid) > +{ > + HRESULT status; > + ULONG oxid_upper_bits; > + > + static LONG oxid_lower_bits; > + > + status = I_RpcBindingInqLocalClientPID(h, &oxid_upper_bits); > + ERR("Got client PID: %ld", oxid_upper_bits); > + if (status != S_OK) { Please put the opening brace on a new line. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/2059#note_22594

1 0

Re: [PATCH v3 0/2] MR2059: combase: Generate apartment OXID in rpcss
by Huw Davies (＠huw) 01 Feb '23

01 Feb '23

Huw Davies (@huw) commented about dlls/combase/apartment.c: > > - if (apt->multi_threaded) > - { > - /* FIXME: should be randomly generated by in an RPC call to rpcss */ > - apt->oxid = ((OXID)GetCurrentProcessId() << 32) | 0xcafe; > - } > - else > - { > - /* FIXME: should be randomly generated by in an RPC call to rpcss */ > - apt->oxid = ((OXID)GetCurrentProcessId() << 32) | GetCurrentThreadId(); > - } > + /* Get a new OXID to use for this apartment. Note that re-creating > + * an apartment (by calling CoUninitialize and then CoInitializeEx) > + * will result in a new OXID. This ensures that we don't re-use a > + * proxy_manager` that refers to a dead apartment. */ > + rpcss_get_new_apartment_oxid(&apt->oxid); We would want to test for this failing (and probably clean up and return NULL). -- https://gitlab.winehq.org/wine/wine/-/merge_requests/2059#note_22591

1 0

Re: [PATCH v3 0/2] MR2059: combase: Generate apartment OXID in rpcss
by Huw Davies (＠huw) 01 Feb '23

01 Feb '23

Huw Davies (@huw) commented about programs/rpcss/rpcss_main.c: > return S_OK; > } > > +/* Generates a new OXID suitable for use by an apartment. > + * Each call to this function returns a distinct OXID. > + * This function should not be invoked directly - use rpcss_get_new_apartment_oxid > + * instead. */ > +HRESULT __cdecl irpcss_get_new_apartment_oxid(handle_t h, OXID *oxid) > +{ > + HRESULT status; > + ULONG oxid_upper_bits; > + I think we can get rid of this blank line. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/2059#note_22592

1 0

Re: [PATCH v3 0/2] MR2059: combase: Generate apartment OXID in rpcss
by Huw Davies (＠huw) 01 Feb '23

01 Feb '23

Since the first commit just adds tests, could you change the commit msg prefix to `ole32/tests:` ? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/2059#note_22590

1 0