June 2015 - Wine-devel - list.winehq.org

Re: [PATCH 2/3] d3drm/tests: Add test for IDirect3DRM*::CreateDeviceFromSurface (try 4).
by Stefan Dösinger 24 Jun '15

24 Jun '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, One I don't like here is the hardcoded DDSCAPS_SYSTEMEMORY for the Z buffer. You need it on Windows because of the RGB device and Wine doesn't check, so it kinda works. But I recommend to do something like this: BOOL use_sysmem_zbuffer; ... hr = IDirect3DRM2_CreateDeviceFromSurface(d3drm2, &driver, ddraw, surface, &device2); hr = IDirect3DRMDevice2_GetDirect3DDevice2(device2, &d3ddevice2); hr = IDirect3DDevice2_GetRenderTarget(d3ddevice2, &d3drm_surface); hr = IDirectDrawSurface_GetAttachedSurface(d3drm_surface, &caps, &ds); hr = IDirectDrawSurface_GetSurfaceDesc(ds, &desc); use_sysmem_zbuffer = desc->ddsCaps.dwCaps & DDSCAPS_SYSTEMMEMORY; ... /* Release everything */ /* Create new sysmem surface */ when creating the new Z buffer set DDSCAPS_SYSTEMMEMORY only if use_sysmem_zbuffer is true. Another thing to test: After destroying the d3drm device is the Z buffer removed from the surface when (a) d3drm created it and (b) you created it? Cheers, Stefan -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVicXHAAoJEN0/YqbEcdMw73gP/3wLEdptNIWr2IFNXJHoSZ15 2g3annULMUw++NvXuBXY1RJdUiR0UQxbGyRNJtFzl8KN66lbl2DOY9/ePkgahIbB 0XKedAtS8F5zoLwRHrbZ/VGSI5Kv8ClHSi0F+jZ5Og9TvwqYKC40vNaBC2WXlSHL KSmOzilln0xDFPecaqyIwbLNO7y1DBnQ7sncLgGCuyxrWM8lK0vmXcAxubS5JyJJ Hd0chD6+EpjxisVWZUVNLFV1HEN9zyL6VDSYcJnNNp9ufGKh6wAoFcywx1vyQTsX tiicEYlW90rvhM2dIZ/m6EcJsR/n6E4iMlj2S4yq9Kc4jySMaC0bYa0B4vPQbbaj YUcQ10qmuH/saROVPjjwjT5O1xISVZ2OOlk3PQYbwVSa1afOMU0ZZu5xNYmkRJ71 21r9cBqePMtIzWixxq+5DiYIRaSDelODlGD8P+LdfYu51DZK3oi1GK0vwX46wHmX M6S6i583uj//ZaZAVG/iCkZ9IPrnkqUnWe8WVGqy1KKJ/e/goq8eQhK9bUHz0GCM BCaklF368VarJWQ0S2ggtdZ006cnc+aqOlfJTRA4UoSfQGU6ACne/blCeJBIGMwv 4mwtvNQvZIvoO6sUW+kmU7T6vCsyxVGfVq96csPbqNvSlidLl8H8IN7f8DK65feW w4s6zvEsdURxeHfcOUNL =bY5s -----END PGP SIGNATURE-----

2 2

Re: [2/2] msimtf: Add stub implementation for IActiveIMMMessagePumpOwner.
by Nikolay Sivov 24 Jun '15

24 Jun '15

On 24.06.2015 0:48, Thomas Faber wrote: > +static HRESULT WINAPI ActiveIMMApp_IActiveIMMMessagePumpOwner_QueryInterface(IActiveIMMMessagePumpOwner* iface, > + REFIID iid, LPVOID *ppvOut) > +{ > + ActiveIMMApp *This = impl_from_IActiveIMMMessagePumpOwner(iface); > + return IActiveIMMApp_QueryInterface(&This->IActiveIMMApp_iface, iid, ppvOut); > +} It's usually up to author, but I'll still mention this - no need to prefix function names with ActiveIMMApp_ because there's no other IActiveIMMMessagePumpOwner implementation, so names won't collide. It would be nice to have a simple test too, to make sure QI supports this interface, but as we don't have tests dir at all for this module it could be a separate patch I guess.

2 1

Re: [1/2] dimm.idl: Add IActiveIMMMessagePumpOwner.
by Nikolay Sivov 24 Jun '15

24 Jun '15

What's a reason for having it declared like that: -- + HRESULT OnTranslateMessage([in] const struct tagMSG *pMsg); --- Why not use MSG*?

2 3

Re: d3dcompiler: Add D3DReadFileToBlob, D3DWriteBlobToFile, , D3DCompileFromFile, D3DCompile2 stubs
by Matteo Bruni 24 Jun '15

24 Jun '15

2015-06-23 10:41 GMT+02:00 Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com>: > Currently the code for d3dcompiler_*dll's is in the _43 version. > > However these function belong in d3dcompiler_46+ > > So, > 1. Is it ok to use private functions? I don't know all the implications of that but it doesn't seem to be necessarily wrong. > If not, how should it be done? Keep in mind that d3d9, d3d10, x_input will > have the same issue at some point. Not sure what you mean with d3d9 here. Did you mean d3dx9? > 2. Could it be done like the msvrt* libraites? However since the structures > haven't changed between versions maybe this isn't the best way. There are / have been a few bugs reported for d3dx9 with applications failing to cope with forwarded functions (e.g. https://bugs.winehq.org/show_bug.cgi?id=21817) and I imagine similar problems might also happen with d3dcompiler so that would be a point in favor of this solution. Moreover, in general each native d3dcompiler_xx (AND each d3dx9_xx!) version has a separate copy of the HLSL compiler inside and each one has a different version of the compiler. That means you will get things like https://bugs.winehq.org/show_bug.cgi?id=33770#c7 unless we can generate different code for different d3dcompiler_xx.dll versions. Another one of those I know of is that functions inside ASCII effect files can be "overloaded" with multiple implementations for different shader versions. In older d3dx9 you specify the target version by spelling it out in the function definition, but that changed at some point and the older syntax just doesn't work on newer HLSL compiler versions. So I think this is the route we want to take eventually. Maybe your patch is fine for the time being though. > 3. Move the code the latest DLL. That would work but notice that the same thing might (will?) repeat after MS releases new versions of d3dcompiler_xx.dll with new exports. BTW, I should have asked this earlier: what do you need those newer d3dcompiler_xx.dlls for? AFAIU those versions > 43 are never installed in system32 and applications are supposed to just bring a copy in their application directory (e.g. see the "Redistribution" section in http://blogs.msdn.com/b/chuckw/archive/2012/05/07/hlsl-fxc-and-d3dcompile.a… or point 6c in https://msdn.microsoft.com/en-us/library/windows/desktop/ee663275%28v=vs.85…)

2 1

Re: [PATCH 1/3] d3drm/tests: Add test for IDirect3DRM*::CreateDeviceFromClipper (try 4).
by Stefan Dösinger 23 Jun '15

23 Jun '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, Looks pretty good, there are 2 minor issues left (see below) and 3 code style things: Am 2015-06-23 um 20:19 schrieb Aaryaman Vasishta: > +BOOL primary_surface_found = FALSE; Please make it static. Also no need to initialize global variables, they're set to zero by the compiler. (Same as static variables inside a function). > + if ((desc->ddsCaps.dwCaps & DDSCAPS_PRIMARYSURFACE) == DDSCAPS_PRIMARYSURFACE) desc->ddsCaps.dwCaps & DDSCAPS_PRIMARYSURFACE is enough in this case as there's only one flag. > + clipper = context; > + ok(clipper == d3drm_clipper, "Expected clipper returned == %p, got %p.\n", clipper, d3drm_clipper); > + IDirectDrawClipper_Release(d3drm_clipper); ok(clipper == context, "..."); should work, no need to have an extra variable for this. > +static void test_create_device_from_clipper(void) > +{ This function tests IDirect3DRM2 / IDirect3DRMDevice2. Think about how you want to add version 1 and 3 tests. Either add them in the same function or create 3 test functions. I prefer the latter case, but then you may want to name them test_create_device_from_clipper{1,2,3} or similar. > + hr = IDirect3DRM2_CreateDeviceFromClipper(d3drm2, clipper, &driver, 300, 200, &device2); > + ok(hr == D3DRM_OK, "Cannot create IDirect3DRMDevice2 interface (hr = %x).\n", hr); > + ref2 = get_refcount((IUnknown *)d3drm1); > + cref2 = get_refcount((IUnknown *)clipper); > + todo_wine ok(ref2 > ref1, "expected ref2 > ref1, got ref1 = %u , ref2 = %u.\n", ref1, ref2); > + todo_wine ok(cref2 > cref1, "expected cref2 > cref1, got cref1 = %u , cref2 = %u.\n", cref1, cref2); Please also show that the d3drm2 refcount is untouched. This also applies to patches 2 and 3. > + hr = IDirectDraw_SetDisplayMode(ddraw, 640, 480, 16); > + ok(hr == DD_OK, "Cannot set display mode to 16bpp (hr = %x).\n", hr); Please use the current desktop width and height here (i.e., what was returned from GetDisplayMode earlier). Setting a lower resolution has the tendency to screw up window and desktop icon positioning. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVib/vAAoJEN0/YqbEcdMw8UoP+gLjxDhe3SjrslH3I0h3cM0M W9KbODhgTmn+C2j3a2rOXvbllAXkjAI9oSnbKp3msE4uJw7GT5QOx+y1828xVF54 RhxoQsYJuXAptAzzlwOS9Dpj5f8sebfj8AG3KENEeMBaKPJFnuER2AHD7o0Bl9Td LVXbSWjvkhsGIRSW2JWKVFQIRdCKaeiRciA0ScH/yQxxL0h/1BOdoLqzc5+Nv6wW /drwTUVR0MQrqzNt3qKPnhNNe2CXS7DQvc4KA1s0WsNxTbrl1orKXDclls1UhL5L iIIJtVF0H7jPE5kF5yVsW0uyjojNI7EnEbVtQQHecFV1UfLQQhykIbw3Vc/tg8Up lpkaEtQvSUYsJMIWKyG9nTAsoz1ZAeb8k97YH8SS/HtG3dZzOThLkPjfbpOPRxeK 07d2P+I4FkyXdiDNZ/dzXIuJUGoBriXM/rPJFqnHWrOm3Lor6GqhKXE+SxOpPeOA rUcVYfd/XEJPT+7PIOdeu/C9ldUPjVmV3BpWze0oXaj9noLJUWojk4E0emsazaUB bXXcJfj0r65SnFUUbxmwjZK50OdF36iku3Oohgl7kpmatWF3IN1g9BmgoDnXuHi8 8xVAi+6OQm7+oU04kZi/4KHPAg41FcMBnxtuxYBASn0NZvaY002ZMWpfCAbdV6pX LouyQXU2ct2Y5h78u9FG =F7Cf -----END PGP SIGNATURE-----

2 1

Re: [PATCH 2/2] user32: Erase backgrounds of owner-drawn buttons with the default color.
by Alexandre Julliard 23 Jun '15

23 Jun '15

Alex Henrie <alexhenrie24(a)gmail.com> writes: > @@ -268,13 +268,7 @@ LRESULT ButtonWndProc_common(HWND hWnd, UINT uMsg, WPARAM wParam, LPARAM lParam, > { > HDC hdc = (HDC)wParam; > RECT rc; > - HBRUSH hBrush; > - HWND parent = GetParent(hWnd); > - if (!parent) parent = hWnd; > - hBrush = (HBRUSH)SendMessageW(parent, WM_CTLCOLORBTN, (WPARAM)hdc, (LPARAM)hWnd); > - if (!hBrush) /* did the app forget to call defwindowproc ? */ > - hBrush = (HBRUSH)DefWindowProcW(parent, WM_CTLCOLORBTN, > - (WPARAM)hdc, (LPARAM)hWnd); > + HBRUSH hBrush = GetSysColorBrush(COLOR_BTNFACE); This would need a more convincing test. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: shell32/tests: Don't leave behind open explorer windows.
by Marvin 23 Jun '15

23 Jun '15

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=14594 Your paranoid android. === w2003std (32 bit progman_dde) === progman_dde.c:377: Test failed: Window "C:\Documents and Settings\All Users\Start Menu\Programs\Group1" was not created in 45 seconds - assumed failure. [S_G:12] === w7u (32 bit progman_dde) === progman_dde.c:519: Test failed: DeleteGroup Group1: Expected Error DMLERR_NO_ERROR, received DMLERR_EXECACKTIMEOUT. [D_G:13] === wvistau64 (64 bit progman_dde) === progman_dde.c:571: Test failed: DeleteItem f2g1Name.lnk: Expected Error DMLERR_NO_ERROR, received DMLERR_EXECACKTIMEOUT. [D_I:5]

1 0

Re: [PATCH 1/2] user32: Send BM_SETCHECK before WM_CAPTURECHANGED and add tests.
by Marvin 23 Jun '15

23 Jun '15

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=14585 Your paranoid android. === wvistau64 (32 bit msg) === msg.c:10547: Test failed: expected PeekMessage to return FALSE, got 1

1 0

Re: ntdll: Randomize security cookie when available
by Alexandre Julliard 23 Jun '15

23 Jun '15

André Hentschel <nerv(a)dawncrow.de> writes: > @@ -3620,7 +3620,30 @@ typedef struct _FPO_DATA { > WORD cbFrame : 2; > } FPO_DATA, *PFPO_DATA; > > -typedef struct _IMAGE_LOAD_CONFIG_DIRECTORY { > +typedef struct _IMAGE_LOAD_CONFIG_DIRECTORY64 { > + DWORD Size; > + DWORD TimeDateStamp; > + WORD MajorVersion; > + WORD MinorVersion; > + DWORD GlobalFlagsClear; > + DWORD GlobalFlagsSet; > + DWORD CriticalSectionDefaultTimeout; > + ULONGLONG DeCommitFreeBlockThreshold; > + ULONGLONG DeCommitTotalFreeThreshold; > + ULONGLONG LockPrefixTable; > + ULONGLONG MaximumAllocationSize; > + ULONGLONG VirtualMemoryThreshold; > + ULONGLONG ProcessAffinityMask; > + DWORD ProcessHeapFlags; > + WORD CSDVersion; > + WORD Reserved1; > + ULONGLONG EditList; > + ULONGLONG SecurityCookie; > + ULONGLONG SEHandlerTable; > + ULONGLONG SEHandlerCount; > +} IMAGE_LOAD_CONFIG_DIRECTORY64, *PIMAGE_LOAD_CONFIG_DIRECTORY64; > + > +typedef struct _IMAGE_LOAD_CONFIG_DIRECTORY32 { > DWORD Size; > DWORD TimeDateStamp; > WORD MajorVersion; You'll need to update the generated tests: gcc -c -o generated.o ../../../../wine/dlls/ntdll/tests/generated.c -I. \ -I../../../../wine/dlls/ntdll/tests -I../../../include -I../../../../wine/include -D__WINESRC__ \ -D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing -Wdeclaration-after-statement -Wempty-body \ -Wignored-qualifiers -Wstrict-prototypes -Wtype-limits -Wunused-but-set-parameter -Wvla \ -Wwrite-strings -Wpointer-arith -Wlogical-op -gdwarf-2 -gstrict-dwarf -g -O2 -fno-diagnostics-show-caret ../../../../wine/dlls/ntdll/tests/generated.c: In function ‘test_pack_IMAGE_LOAD_CONFIG_DIRECTORY’: ../../../../wine/dlls/ntdll/tests/generated.c:842:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:865:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:866:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:868:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:869:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:870:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:873:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:874:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:875:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:876:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:877:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:878:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:879:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:882:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:883:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:884:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:885:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:888:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:891:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:894:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:895:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:896:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:897:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:898:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:899:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:900:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:901:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:902:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c:903:5: error: size of unnamed array is negative ../../../../wine/dlls/ntdll/tests/generated.c: In function ‘test_pack_PIMAGE_LOAD_CONFIG_DIRECTORY’: ../../../../wine/dlls/ntdll/tests/generated.c:2108:5: error: size of unnamed array is negative Makefile:348: recipe for target 'generated.o' failed -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: advapi32: Fix RegEnumValueW when enumerating long values.
by Bernhard Übelacker 23 Jun '15

23 Jun '15

Hello, please ignore this patch for now. It seems that systemfontfileenumerator_GetCurrentFontFile uses RegEnumValueW in a way not supported on Windows - with giving NULL parameters for value and/or val_count [1]. When RegEnumValueW is called with proper value parameter the problem with not returning correct count would not happen already. So, as far as I see, first all uses of RegEnumValue with NULL parameters for value or val_count need a fix. Kind regards, Bernhard [1] https://testbot.winehq.org/JobDetails.pl?Key=14533

4 5