August 2016 - Wine-devel - list.winehq.org

Re: [PATCH] winhttp/tests: Fix indentation warning on GCC 6
by Marvin 10 Aug '16

10 Aug '16

Hi, While running your changed tests on Windows, 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=24961 Your paranoid android. === w864 (32 bit winhttp) === winhttp.c:1030: Test failed: failed to send request 2148074273 winhttp.c:1034: Test failed: failed to retrieve certificate context 12019 winhttp.c:1039: Test failed: failed to retrieve key bitness 4317 winhttp.c:1043: Test failed: failed to retrieve certificate info 12019 winhttp.c:1056: Test failed: failed to receive response 12019 winhttp.c:1060: Test failed: failed to query available data 12019 winhttp.c:1061: Test failed: available_size = 0 winhttp.c:1066: Test failed: failed unexpectedly 12019 winhttp.c:1067: Test failed: request failed unexpectedly 3735928559 winhttp.c:1078: Test failed: WinHttpReadData failed: 12019.

1 0

Re: [PATCH resend 2/2] wined3d: Do not enable blending for edge/line antialiasing.
by Marvin 10 Aug '16

10 Aug '16

Hi, While running your changed tests on Windows, 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=24955 Your paranoid android. === w1064 (32 bit ddraw4) === ddraw4.c:2542: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (64 bit ddraw4) === ddraw4.c:2542: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (32 bit ddraw7) === ddraw7.c:2272: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (64 bit ddraw7) === ddraw7.c:2272: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (32 bit ddraw2) === ddraw2.c:2357: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (64 bit ddraw2) === ddraw2.c:2357: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768).

1 0

Re: [PATCH resend 2/2] wined3d: Do not enable blending for edge/line antialiasing.
by Marvin 10 Aug '16

10 Aug '16

Hi, While running your changed tests on Windows, 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=24951 Your paranoid android. === w1064 (32 bit ddraw2) === ddraw2.c:2357: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (64 bit ddraw2) === ddraw2.c:2357: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (32 bit ddraw7) === ddraw7.c:2272: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (64 bit ddraw7) === ddraw7.c:2272: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (32 bit ddraw4) === ddraw4.c:2542: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (64 bit ddraw4) === ddraw4.c:2542: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768).

1 0

Re: [PATCH v2 1/2] ddraw/tests: Add tests for edge antialiasing blending.
by Marvin 10 Aug '16

10 Aug '16

Hi, While running your changed tests on Windows, 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=24948 Your paranoid android. === w1064 (32 bit ddraw7) === ddraw7.c:2272: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (64 bit ddraw7) === ddraw7.c:2272: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (32 bit ddraw2) === ddraw2.c:2357: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (64 bit ddraw2) === ddraw2.c:2357: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (32 bit ddraw4) === ddraw4.c:2542: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768). === w1064 (64 bit ddraw4) === ddraw4.c:2542: Test failed: Expected (0,0)-(512,384), got (0,0)-(1024,768).

1 0

Re: [PATCH 5/7] ddraw/tests: Add test for edge antialiasing blending.
by Henri Verbeet 10 Aug '16

10 Aug '16

On 8 August 2016 at 16:52, Józef Kucia <jkucia(a)codeweavers.com> wrote: > +static D3DCOLOR get_surface_color_rb(IDirectDraw7 *ddraw, IDirectDrawSurface7 *surface, > + unsigned int x, unsigned int y) > +{ > + IDirectDrawSurface7 *readback_surface; > + DDSURFACEDESC2 surface_desc; > + D3DCOLOR color; > + HRESULT hr; > + > + memset(&surface_desc, 0, sizeof(surface_desc)); > + surface_desc.dwSize = sizeof(surface_desc); > + hr = IDirectDrawSurface7_GetSurfaceDesc(surface, &surface_desc); > + ok(SUCCEEDED(hr), "Failed to get surface desc, hr %#x.\n", hr); > + surface_desc.ddsCaps.dwCaps = DDSCAPS_OFFSCREENPLAIN | DDSCAPS_SYSTEMMEMORY; > + hr = IDirectDraw7_CreateSurface(ddraw, &surface_desc, &readback_surface, NULL); > + ok(SUCCEEDED(hr), "IDirectDraw7_CreateSurface failed, hr %#x.\n", hr); > + hr = IDirectDrawSurface_BltFast(readback_surface, 0, 0, surface, NULL, 0); > + ok(SUCCEEDED(hr), "IDirectDrawSurface7_BltFast failed, hr %#x.\n", hr); > + > + color = get_surface_color(readback_surface, x, y); > + > + IDirectDrawSurface7_Release(readback_surface); > + > + return color; > +} What is this for? Note that ddraw too has a D3DPRASTERCAPS_ANTIALIASEDGES cap. What about other versions? I think at least IDirect3DDevice3 also had D3DRENDERSTATE_EDGEANTIALIAS.

2 1

Re: [PATCH] msvcp140: Forward __ExceptionPtr* to msvcr120
by Piotr Caban 10 Aug '16

10 Aug '16

Hi Daniel, On 08/09/16 18:31, Daniel Lehman wrote: > -@ stub -arch=win32 ?__ExceptionPtrCopy@@YAXPAXPBX(a)Z > -@ stub -arch=win64 ?__ExceptionPtrCopy@@YAXPEAXPEBX(a)Z > +@ cdecl -arch=win32 ?__ExceptionPtrCopy@@YAXPAXPBX(a)Z(ptr ptr) __ExceptionPtrCopy > +@ cdecl -arch=win64 ?__ExceptionPtrCopy@@YAXPEAXPEBX(a)Z(ptr ptr) __ExceptionPtrCopy Why can't you just add forwards to msvcr120.dll in spec file? Thanks, Piotr

2 2

Re: [v2] riched20: allow setting background text color
by Huw Davies 10 Aug '16

10 Aug '16

On Tue, Aug 09, 2016 at 03:17:36PM -0600, Kenneth Haley wrote: > --- a/dlls/riched20/paint.c > +++ b/dlls/riched20/paint.c > @@ -318,10 +318,23 @@ static void get_selection_rect( ME_Context *c, ME_Run *run, int from, int to, in > static void draw_text( ME_Context *c, ME_Run *run, int x, int y, BOOL selected, RECT *sel_rect ) > { > COLORREF text_color = get_text_color( c, run->style, selected ); > - COLORREF back_color = selected ? ITextHost_TxGetSysColor( c->editor->texthost, COLOR_HIGHLIGHT ) : 0; > + COLORREF back_color = 0; > COLORREF old_text, old_back = 0; > const WCHAR *text = get_text( run, 0 ); > ME_String *masked = NULL; > + BOOL paint_bg = FALSE; > + > + if (selected) > + { > + back_color = ITextHost_TxGetSysColor(c->editor->texthost, COLOR_HIGHLIGHT); > + paint_bg = TRUE; > + } > + else if (run->style->fmt.dwMask & CFM_BACKCOLOR && > + !(CFE_AUTOBACKCOLOR & run->style->fmt.dwEffects)) > + { > + back_color = run->style->fmt.crBackColor; > + paint_bg = TRUE; > + } This doesn't work for me. You need to use the new paint_bg boolean to trigger the ETO_OPAQUE flags below. Also there's some odd behaviour if you select text in a run with background colour - the selection is correctly highlighted, but the rest of the run loses its background colour. Huw.

1 0

Re: mmdevapi: Another use case for todo_wine_if().
by Marvin 09 Aug '16

09 Aug '16

Hi, While running your changed tests on Windows, 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=24917 Your paranoid android. === w8 (32 bit capture) === capture.c:181: Test failed: Position 119 expected 0 capture.c:231: Test failed: Position 2167 expected 567 capture.c:261: Test failed: Position 2167 expected 567 capture.c:318: Test failed: Position 3191 gap 2176 capture.c:321: Test failed: GCP 13440 vs. BufferSize 21996 === w864 (32 bit capture) === capture.c:181: Test failed: Position 1079 expected 0 capture.c:231: Test failed: Position 2167 expected 1527 capture.c:261: Test failed: Position 2167 expected 1527 capture.c:318: Test failed: Position 3191 gap 1216 capture.c:321: Test failed: GCP 14336 vs. BufferSize 21996 capture.c:341: Test failed: Position 4279 expected 3639 capture.c:342: Test failed: flags 1 capture.c:366: Test failed: Position 5303 expected 4087 capture.c:367: Test failed: flags 1 === w1064 (32 bit capture) === capture.c:181: Test failed: Position 1079 expected 0 capture.c:231: Test failed: Position 2167 expected 1527 capture.c:261: Test failed: Position 2167 expected 1527 capture.c:318: Test failed: Position 3255 gap 1280 capture.c:321: Test failed: GCP 11648 vs. BufferSize 21996 capture.c:341: Test failed: Position 4343 expected 3703 capture.c:342: Test failed: flags 1 capture.c:366: Test failed: Position 4791 expected 4151 === w864 (64 bit capture) === capture.c:181: Test failed: Position 1079 expected 0 capture.c:231: Test failed: Position 2167 expected 1527 capture.c:261: Test failed: Position 2167 expected 1527 capture.c:318: Test failed: Position 3127 gap 1152 capture.c:321: Test failed: GCP 13888 vs. BufferSize 21996 capture.c:341: Test failed: Position 4215 expected 3575 capture.c:342: Test failed: flags 1 capture.c:366: Test failed: Position 5303 expected 4023 capture.c:367: Test failed: flags 1 === w1064 (64 bit capture) === capture.c:181: Test failed: Position 1079 expected 0 capture.c:231: Test failed: Position 2167 expected 1527 capture.c:261: Test failed: Position 2167 expected 1527 capture.c:318: Test failed: Position 3255 gap 1280 capture.c:321: Test failed: GCP 11648 vs. BufferSize 21996 capture.c:341: Test failed: Position 4343 expected 3703 capture.c:342: Test failed: flags 1 capture.c:366: Test failed: Position 4791 expected 4151

1 0

Re: [PATCH] wshom.ocx/tests: Fix indentation warning on GCC 6
by Bruno Jesus 09 Aug '16

09 Aug '16

On Tue, Aug 9, 2016 at 2:07 PM, Nikolay Sivov <nsivov(a)codeweavers.com> wrote: > Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> Are you going to also change lines starting with todo_wine? Like http://source.winehq.org/source/dlls/ws2_32/tests/sock.c#1468 Because it is made like this to make it more evident and to reduce diff files when the problem is fixed. Best wishes, Bruno

2 1

Re: [PATCH 2/3] d3drm: Implement IDirect3DRMFrame*::GetScene. (v2)
by Henri Verbeet 09 Aug '16

09 Aug '16

On 9 August 2016 at 03:16, Aaryaman Vasishta <jem456.vasishta(a)gmail.com> wrote: > static HRESULT WINAPI d3drm_frame3_GetScene(IDirect3DRMFrame3 *iface, IDirect3DRMFrame3 **scene) > { > - FIXME("iface %p, scene %p stub!\n", iface, scene); > + HRESULT hr; > + IDirect3DRMFrame3 *frame = iface, *parent_frame = NULL; > > - return E_NOTIMPL; > + TRACE("iface %p, scene %p.\n", iface, scene); > + > + if (!scene) > + return D3DRMERR_BADVALUE; > + > + for (;;) > + { > + if (FAILED(hr = IDirect3DRMFrame3_GetParent(frame, &parent_frame))) > + return hr; > + if (parent_frame) > + { > + frame = parent_frame; > + IDirect3DRMFrame3_Release(parent_frame); > + } > + else > + break; > + } You could just do if (!parent_frame) break; More importantly though, as Stefan already mentioned, the reference counting is weird. I think the code would be much nicer if you just used impl_from_IDirect3DRMFrame3() and then followed the frame->parent pointers.

2 1