June 2022 - Wine-Devel - list.winehq.org

Re: [PATCH 0/7] MR212: d3d11: Create dxgi resource objects for textures and buffers.

by Henri Verbeet

On Fri, 10 Jun 2022 at 13:11, Jan Sikorski (@jsikorski) <wine(a)gitlab.winehq.org> wrote: > > On Fri Jun 10 11:05:14 2022 +0000, Jan Sikorski wrote: > > As far as I'm concerned, that's fine. > I agree with Henri's point though, perhaps it would be better to reorder initialization so that we create dxgi_resource first? The issue with that is that dxgi_resource_init() takes the wined3d resource as parameter. Perhaps that could be avoided, but it would likely introduce its own set of issues.

June 10, 2022

Re: [PATCH 0/7] MR212: d3d11: Create dxgi resource objects for textures and buffers.

by Jan Sikorski (＠jsikorski)

On Fri Jun 10 11:05:14 2022 +0000, Jan Sikorski wrote: > As far as I'm concerned, that's fine. I agree with Henri's point though, perhaps it would be better to reorder initialization so that we create dxgi_resource first? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/212#note_1821

June 10, 2022

Re: [PATCH 4/6] dinput/tests: Merge mouse.c tests into device8.c.

by Marvin

Hi, While running your changed tests, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at: https://testbot.winehq.org/JobDetails.pl?Key=116663 Your paranoid android. === w8 (32 bit report) === dinput: joystick8.c:1836: Test failed: 0x700: state[0]: WaitForSingleObject succeeded === debian11 (32 bit French report) === dinput: device8.c:1965: Test failed: 0x800: Acquire returned 0x80070005 device8.c:1974: Test failed: 0x800: WaitForSingleObject returned 0x102 device8.c:1978: Test failed: 0x800: GetDeviceData returned 0x8007000c device8.c:1983: Test failed: 0x800: WaitForSingleObject returned 0x102 device8.c:1986: Test failed: 0x800: Unacquire returned 0x1 device8.c:1993: Test failed: 0x800: Acquire returned 0x80070005 device8.c:2002: Test failed: 0x800: WaitForSingleObject returned 0x102 device8.c:2005: Test failed: 0x800: Unacquire returned 0x1 device8.c:2008: Test failed: 0x800: Acquire returned 0x80070005 device8.c:2011: Test failed: 0x800: GetDeviceData returned 0x8007000c device8.c:2021: Test failed: 0x800: WaitForSingleObject returned 0x102 device8.c:2027: Test failed: 0x800: WaitForSingleObject returned 0x102 device8.c:2027: Test failed: 0x800: WaitForSingleObject returned 0x102 device8.c:2032: Test failed: 0x800: GetDeviceData returned 0x8007000c device8.c:2035: Test failed: 0x800: GetDeviceData returned 0x8007000c

June 10, 2022

Re: [PATCH 0/7] MR212: d3d11: Create dxgi resource objects for textures and buffers. - approved

by Jan Sikorski (＠jsikorski)

This merge request was approved by Jan Sikorski. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/212

June 10, 2022

Re: [PATCH 0/7] MR212: d3d11: Create dxgi resource objects for textures and buffers.

by Jan Sikorski (＠jsikorski)

On Fri Jun 10 11:03:37 2022 +0000, Nikolay Sivov wrote: > Yes, for QueryInterface() it could be removed, like it's done for > buffers. I was thinking removing it later with additional patch to keep > the diff size down. As far as I'm concerned, that's fine. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/212#note_1819

June 10, 2022

Re: [PATCH 0/7] MR212: d3d11: Create dxgi resource objects for textures and buffers.

by Nikolay Sivov (＠nsivov)

On Fri Jun 10 11:00:08 2022 +0000, **** wrote: > Henri Verbeet replied on the mailing list: > ``` > On Fri, 10 Jun 2022 at 12:51, Jan Sikorski (@jsikorski) > <wine(a)gitlab.winehq.org> wrote: > > I'm not sure about the released() procedures - it looks like if we > fail to create the dxgi_resource, we decref the texture or buffer and > then it gets released with dxgi_resource set to NULL, no? > Hmm, right. That's unfortunate though; ideally > wined3d_object_released() would only be called on fully initialised > objects. > ``` Yes, for QueryInterface() it could be removed, like it's done for buffers. I was thinking removing it later with additional patch to keep the diff size down. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/212#note_1818

June 10, 2022

Re: [PATCH 0/7] MR212: d3d11: Create dxgi resource objects for textures and buffers.

by Henri Verbeet

On Fri, 10 Jun 2022 at 12:51, Jan Sikorski (@jsikorski) <wine(a)gitlab.winehq.org> wrote: > I'm not sure about the released() procedures - it looks like if we fail to create the dxgi_resource, we decref the texture or buffer and then it gets released with dxgi_resource set to NULL, no? Hmm, right. That's unfortunate though; ideally wined3d_object_released() would only be called on fully initialised objects.

June 10, 2022

Re: [PATCH 0/7] MR212: d3d11: Create dxgi resource objects for textures and buffers.

by Jan Sikorski (＠jsikorski)

On Fri Jun 10 10:46:28 2022 +0000, **** wrote: > Henri Verbeet replied on the mailing list: > ``` > On Fri, 10 Jun 2022 at 12:34, Jan Sikorski (@jsikorski) > <wine(a)gitlab.winehq.org> wrote: > > > > Jan Sikorski (@jsikorski) commented about dlls/d3d11/texture.c: > > > return S_OK; > > > } > > > > > > - if (texture->dxgi_surface) > > > + if (texture->dxgi_resource) > > Mostly to test my understanding, do we still actually need to check > this for some reason? > I think we should be able to drop the check and do this > unconditionally. That would apply to e.g. > d3d_texture1d_wined3d_object_released() as well. > ``` I'm not sure about the released() procedures - it looks like if we fail to create the dxgi_resource, we decref the texture or buffer and then it gets released with dxgi_resource set to NULL, no? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/212#note_1816

June 10, 2022

Re: [PATCH 0/7] MR212: d3d11: Create dxgi resource objects for textures and buffers.

by Henri Verbeet

On Fri, 10 Jun 2022 at 12:34, Jan Sikorski (@jsikorski) <wine(a)gitlab.winehq.org> wrote: > > Jan Sikorski (@jsikorski) commented about dlls/d3d11/texture.c: > > return S_OK; > > } > > > > - if (texture->dxgi_surface) > > + if (texture->dxgi_resource) > Mostly to test my understanding, do we still actually need to check this for some reason? I think we should be able to drop the check and do this unconditionally. That would apply to e.g. d3d_texture1d_wined3d_object_released() as well.

June 10, 2022

Re: [PATCH v4 0/1] MR167: d3drm: IDirect3DRMMeshBuilder3::Load support loading from resource - approved

by Jan Sikorski (＠jsikorski)

This merge request was approved by Jan Sikorski. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/167

June 10, 2022