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

Re: [PATCH 1/2] ws2_32: Add support for TF_DISCONNECT to TransmitFile.
by Sebastian Lackner 22 Oct '15

22 Oct '15

On 23.10.2015 02:03, Erich E. Hoover wrote: > Now that we have TransmitFile support, there are a couple important > flags that are frequently used by applications utilizing this feature > (particularly in async mode). This patch adds support for the > TF_DISCONNECT flag, which closes out the socket handle once the > TransmitFile operation is complete. I would prefer a solution without NtClose from an APC callback. I have tests in the Staging tree, which show that NtClose will throw an exception for invalid handles, when a debugger is attached.

1 0

Re: [PATCH] d3d10_1: Implement D3D10CreateDeviceAndSwapChain1 (try 2)
by Henri Verbeet 22 Oct '15

22 Oct '15

On 22 October 2015 at 09:00, Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > +static void test_create_device_swap_chain(void) You should probably just port d3d11/tests' test_create_device().

1 0

Re: quartz: Add stub IAMStreamSelect to MPEG-1 Stream Splitter. (try 2, resend)
by Nikolay Sivov 22 Oct '15

22 Oct '15

On 22.10.2015 17:35, Anton Baskanov wrote: > + if ( IsEqualIID(riid, &IID_IUnknown) > + || IsEqualIID(riid, &IID_IPersist) > + || IsEqualIID(riid, &IID_IMediaFilter) > + || IsEqualIID(riid, &IID_IBaseFilter) ) > + *ppv = iface; > + > + if ( IsEqualIID(riid, &IID_IAMStreamSelect) ) > + *ppv = &This->IAMStreamSelect_iface; This is usually done with 'else if'. > +static HRESULT WINAPI StreamSelect_QueryInterface(IAMStreamSelect *iface, REFIID riid, LPVOID *ppv) > +{ > + MPEGSplitterImpl *This = impl_from_IAMStreamSelect(iface); > + > + return IBaseFilter_QueryInterface(&This->Parser.filter.IBaseFilter_iface, riid, ppv); > +} Unfortunately we don't have a written rule for that (I think), so I guess I'm just sharing my habits: I personally try to use exact interface names for implementation functions, with leading 'I' stripped, so AMStreamSelect_* in this case. This makes it clear what's this is about, and you don't need to invent another name that doesn't exist anywhere. > +static const IAMStreamSelectVtbl StreamSelect_Vtbl = > +{ Same here, AMStreamSelectVtbl is easier imho. (quartz is not a best example in that regard) Also please, no LPVOID. > + HRESULT Count( > + [out] DWORD * pcStreams); > + > + HRESULT Info( > + [in] long lIndex, > + [out] AM_MEDIA_TYPE ** ppmt, > + [out] DWORD * pdwFlags, > + [out] LCID * plcid, > + [out] DWORD * pdwGroup, > + [out] WCHAR ** ppszName, > + [out] IUnknown ** ppObject, > + [out] IUnknown ** ppUnk); > + > + HRESULT Enable( > + [in] long lIndex, > + [in] DWORD dwFlags); You are using exact naming PSDK has, please do it in your own words, with better, cleaner names, like 'streams' or 'index', spacing in pointer notation could be fixed as well.

1 0

Re: ieframe: Capitalize some menu items‏
by Jacek Caban 22 Oct '15

22 Oct '15

Hi Jared, The patch still has NEW status, but I don't think I can sign it of myself. I'm not native speaker and I recall discussions about similar case on wine-devel in the past. I hope someone else can comment. Thanks, Jacek On 10/19/15 23:30, Jared Smudde wrote: > Signed-off-by: Jared Smudde <computerwhiz02(a)hotmail.com> > > Some of the second words on menu items are not capitalized. This looks weird as some of the other menu items are all capitalized. > > > dlls/ieframe/ieframe.rc | 10 +++++----- > 1 file changed, 5 insertions(+), 5 deletions(-) > > diff --git a/dlls/ieframe/ieframe.rc b/dlls/ieframe/ieframe.rc > index aed8483..d4ccf3c 100644 > --- a/dlls/ieframe/ieframe.rc > +++ b/dlls/ieframe/ieframe.rc > @@ -32,11 +32,11 @@ IDR_BROWSE_MAIN_MENU MENU > } > MENUITEM "&Open...", ID_BROWSE_OPEN > MENUITEM "&Save", ID_BROWSE_SAVE > - MENUITEM "Save &as...", ID_BROWSE_SAVE_AS > + MENUITEM "Save &As...", ID_BROWSE_SAVE_AS > MENUITEM SEPARATOR > - MENUITEM "Print &format...", ID_BROWSE_PRINT_FORMAT > + MENUITEM "Print &Format...", ID_BROWSE_PRINT_FORMAT > MENUITEM "Pr&int...", ID_BROWSE_PRINT > - MENUITEM "Print previe&w", ID_BROWSE_PRINT_PREVIEW > + MENUITEM "Print Previe&w", ID_BROWSE_PRINT_PREVIEW > MENUITEM SEPARATOR > MENUITEM "&Properties", ID_BROWSE_PROPERTIES > MENUITEM "&Close", ID_BROWSE_QUIT > @@ -45,8 +45,8 @@ IDR_BROWSE_MAIN_MENU MENU > { > POPUP "&Toolbars" > { > - MENUITEM "&Standard bar", ID_BROWSE_BAR_STD > - MENUITEM "&Address bar", ID_BROWSE_BAR_ADDR > + MENUITEM "&Standard Bar", ID_BROWSE_BAR_STD > + MENUITEM "&Address Bar", ID_BROWSE_BAR_ADDR > } > } > POPUP "&Favorites"

3 2

Re: [PATCH] d3d10_1: Implement D3D10CreateDeviceAndSwapChain1 (try 2)
by Józef Kucia 22 Oct '15

22 Oct '15

On Thu, Oct 22, 2015 at 9:00 AM, Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > Fixes https://bugs.winehq.org/show_bug.cgi?id=39278 > > Signed-off-by: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> > --- > dlls/d3d10_1/d3d10_1.spec | 2 +- > dlls/d3d10_1/d3d10_1_main.c | 71 ++++++++++++++++++++++++++++++++++++++++++ > dlls/d3d10_1/tests/Makefile.in | 2 +- > dlls/d3d10_1/tests/d3d10_1.c | 63 +++++++++++++++++++++++++++++++++++++ > include/d3d10_1.idl | 3 ++ > 5 files changed, 139 insertions(+), 2 deletions(-) > > diff --git a/dlls/d3d10_1/d3d10_1.spec b/dlls/d3d10_1/d3d10_1.spec > index b3d2da7..5582558 100644 > --- a/dlls/d3d10_1/d3d10_1.spec > +++ b/dlls/d3d10_1/d3d10_1.spec > @@ -3,7 +3,7 @@ > @ stub D3D10CompileShader > @ stub D3D10CreateBlob > @ stdcall D3D10CreateDevice1(ptr long ptr long long long ptr) > -@ stub D3D10CreateDeviceAndSwapChain1 > +@ stdcall D3D10CreateDeviceAndSwapChain1(ptr long ptr long long long ptr ptr ptr) > @ stdcall D3D10CreateEffectFromMemory(ptr long long ptr ptr ptr) d3d10.D3D10CreateEffectFromMemory > @ stub D3D10CreateEffectPoolFromMemory > @ stdcall D3D10CreateStateBlock(ptr ptr ptr) d3d10.D3D10CreateStateBlock > diff --git a/dlls/d3d10_1/d3d10_1_main.c b/dlls/d3d10_1/d3d10_1_main.c > index a619004..e51019c 100644 > --- a/dlls/d3d10_1/d3d10_1_main.c > +++ b/dlls/d3d10_1/d3d10_1_main.c > @@ -168,3 +168,74 @@ HRESULT WINAPI D3D10CreateDevice1(IDXGIAdapter *adapter, D3D10_DRIVER_TYPE drive > > return hr; > } > + > +HRESULT WINAPI D3D10CreateDeviceAndSwapChain1(IDXGIAdapter *adapter, D3D10_DRIVER_TYPE driver_type, > + HMODULE swrast, UINT flags, D3D10_FEATURE_LEVEL1 hw_level, UINT sdk_version, > + DXGI_SWAP_CHAIN_DESC *swapchain_desc, IDXGISwapChain **swapchain, ID3D10Device1 **device) > +{ > + IDXGIDevice *dxgi_device; > + IDXGIFactory *factory; > + HRESULT hr; > + > + TRACE("adapter %p, driver_type %s, swrast %p, flags %#x, hw_level %s, sdk_version %d, " > + "swapchain_desc %p, swapchain %p, device %p\n", > + adapter, debug_d3d10_driver_type(driver_type), swrast, flags, > + debug_d3d10_feature_level(hw_level), sdk_version, swapchain_desc, swapchain, device); > + It's generally preferred to end messages with dot in new D3D code. It also applies to WARN(), ERR(), FIXME() and ok() messages in tests. > + if(!device) > + return E_INVALIDARG; There should be a space after 'if': if (!device) Please fix this in tests as well. > + > + hr = D3D10CreateDevice1(adapter, driver_type, swrast, flags, hw_level, sdk_version, device); > + if (FAILED(hr)) > + { > + WARN("Failed to create a device, returning %#x\n", hr); > + *device = NULL; > + return hr; > + } It's generally preferred to check if function failed in a single line in new code, i.e. if (FAILED(hr = D3D10CreateDevice1(adapter, driver_type, swrast, flags, hw_level, sdk_version, device)) > + > + TRACE("Created ID3D10Device1 %p\n", *device); > + > + hr = ID3D10Device1_QueryInterface(*device, &IID_IDXGIDevice, (void **)&dxgi_device); > + if (FAILED(hr)) > + { > + ERR("Failed to get a dxgi device from the d3d10 device, returning %#x\n", hr); > + ID3D10Device1_Release(*device); > + *device = NULL; > + return hr; > + } > + > + hr = IDXGIDevice_GetAdapter(dxgi_device, &adapter); > + IDXGIDevice_Release(dxgi_device); > + if (FAILED(hr)) > + { > + ERR("Failed to get the device adapter, returning %#x\n", hr); > + ID3D10Device1_Release(*device); > + *device = NULL; > + return hr; > + } > + > + hr = IDXGIAdapter_GetParent(adapter, &IID_IDXGIFactory, (void **)&factory); > + IDXGIAdapter_Release(adapter); > + if (FAILED(hr)) > + { > + ERR("Failed to get the adapter factory, returning %#x\n", hr); > + ID3D10Device1_Release(*device); > + *device = NULL; > + return hr; > + } > + > + hr = IDXGIFactory_CreateSwapChain(factory, (IUnknown *)*device, swapchain_desc, swapchain); > + IDXGIFactory_Release(factory); > + if (FAILED(hr)) > + { > + ID3D10Device1_Release(*device); > + *device = NULL; > + > + WARN("Failed to create a swapchain, returning %#x\n", hr); > + return hr; > + } > + > + TRACE("Created IDXGISwapChain %p\n", *swapchain); > + > + return S_OK; > +} > diff --git a/dlls/d3d10_1/tests/Makefile.in b/dlls/d3d10_1/tests/Makefile.in > index f6f17bd..256bb40 100644 > --- a/dlls/d3d10_1/tests/Makefile.in > +++ b/dlls/d3d10_1/tests/Makefile.in > @@ -1,5 +1,5 @@ > TESTDLL = d3d10_1.dll > -IMPORTS = d3d10_1 > +IMPORTS = d3d10_1 user32 > > C_SRCS = \ > d3d10_1.c > diff --git a/dlls/d3d10_1/tests/d3d10_1.c b/dlls/d3d10_1/tests/d3d10_1.c > index 534689c..c760b0d 100644 > --- a/dlls/d3d10_1/tests/d3d10_1.c > +++ b/dlls/d3d10_1/tests/d3d10_1.c > @@ -377,8 +377,71 @@ static void test_create_blend_state(void) > ok(!refcount, "Device has %u references left.\n", refcount); > } > > +static void test_create_device_swap_chain(void) > +{ > + HRESULT hr; > + ID3D10Device1 *device; > + IDXGISwapChain *swapchain; > + DXGI_SWAP_CHAIN_DESC swapchain_desc; > + HWND window; > + > + window = CreateWindowA("d3d10_test_wc", "d3d10_test_swap", WS_OVERLAPPEDWINDOW, > + 0, 0, 640, 480, NULL, NULL, NULL, NULL); > + ok(!!window, "Failed to create a window.\n"); > + > + memset(&swapchain_desc, 0, sizeof(swapchain_desc)); > + swapchain_desc.BufferCount = 1; > + swapchain_desc.BufferDesc.Width = 640; > + swapchain_desc.BufferDesc.Height = 480; > + swapchain_desc.BufferDesc.Format = DXGI_FORMAT_R8G8B8A8_UNORM; > + swapchain_desc.BufferDesc.RefreshRate.Numerator = 60; > + swapchain_desc.BufferDesc.RefreshRate.Denominator = 1; > + swapchain_desc.BufferUsage = DXGI_USAGE_RENDER_TARGET_OUTPUT; > + swapchain_desc.SampleDesc.Count = 1; > + swapchain_desc.SampleDesc.Quality = 0; > + swapchain_desc.Windowed = TRUE; > + > + hr = D3D10CreateDeviceAndSwapChain1(NULL, D3D_DRIVER_TYPE_HARDWARE, NULL, 0, D3D10_FEATURE_LEVEL_10_1, > + D3D10_1_SDK_VERSION, &swapchain_desc, &swapchain, NULL); > + ok(hr == E_INVALIDARG, "got %#x.\n", hr); > + > + hr = D3D10CreateDeviceAndSwapChain1(NULL, D3D_DRIVER_TYPE_HARDWARE, NULL, 0, D3D10_FEATURE_LEVEL_10_1, > + D3D10_1_SDK_VERSION, &swapchain_desc, NULL, &device); > + todo_wine ok(hr == S_OK, "got %#x.\n", hr); > + if(hr == S_OK) > + ID3D10Device1_Release(device); > + > + /* swapchain_desc.OutputWindow = NULL */ > + hr = D3D10CreateDeviceAndSwapChain1(NULL, D3D_DRIVER_TYPE_HARDWARE, NULL, 0, D3D10_FEATURE_LEVEL_10_1, > + D3D10_1_SDK_VERSION, &swapchain_desc, &swapchain, &device); > + todo_wine ok(hr == DXGI_ERROR_INVALID_CALL, "got %#x.\n", hr); > + > + swapchain_desc.OutputWindow = window; > + hr = D3D10CreateDeviceAndSwapChain1(NULL, D3D_DRIVER_TYPE_HARDWARE, NULL, 0, D3D10_FEATURE_LEVEL_10_1, > + D3D10_1_SDK_VERSION, &swapchain_desc, &swapchain, &device); > + if(FAILED(hr)) > + { > + skip("Failed to create device.\n"); > + return; > + } You are leaking HWND. Also I am not sure why you use skip() here. > + > + IDXGISwapChain_Release(swapchain); > + ID3D10Device1_Release(device); > + > + DestroyWindow(window); > +} > + > START_TEST(d3d10_1) > { > + WNDCLASSA wc = {0}; > + > + wc.lpfnWndProc = DefWindowProcA; > + wc.lpszClassName = "d3d10_test_wc"; > + RegisterClassA(&wc); > + > test_create_shader_resource_view(); > test_create_blend_state(); > + test_create_device_swap_chain(); > + > + UnregisterClassA("d3d10_test_wc", GetModuleHandleA(NULL)); > } You don't need a new window class. Please use "static". > diff --git a/include/d3d10_1.idl b/include/d3d10_1.idl > index 4ee49c6..05d22b5 100644 > --- a/include/d3d10_1.idl > +++ b/include/d3d10_1.idl > @@ -151,3 +151,6 @@ const UINT D3D10_1_SDK_VERSION = 0x20; > > cpp_quote("HRESULT WINAPI D3D10CreateDevice1(IDXGIAdapter*,D3D10_DRIVER_TYPE,") > cpp_quote(" HMODULE,UINT,D3D10_FEATURE_LEVEL1,UINT,ID3D10Device1**);") > +cpp_quote("HRESULT WINAPI D3D10CreateDeviceAndSwapChain1(IDXGIAdapter *,") > +cpp_quote(" D3D10_DRIVER_TYPE, HMODULE, UINT, D3D10_FEATURE_LEVEL1,") > +cpp_quote(" UINT, DXGI_SWAP_CHAIN_DESC *, IDXGISwapChain **, ID3D10Device1 **);") Please avoid cpp_quote and include parameter names. See D3D11CreateDeviceAndSwapChain() in d3d11.idl. > -- > 1.9.1 > > >

1 0

Re: [PATCH 2/6] d3d11: Use CreateDXGIFactory1 to create factory in D3D11CreateDevice.
by Stefan Dösinger 22 Oct '15

22 Oct '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Am 2015-10-22 um 01:52 schrieb Józef Kucia: > - hr = CreateDXGIFactory(&IID_IDXGIFactory, (void **)&factory); > + hr = CreateDXGIFactory1(&IID_IDXGIFactory, (void **)&factory); Shouldn't you request an IDXGIFactory1 interface now? -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBCAAGBQJWKLJpAAoJEN0/YqbEcdMwEowQAI86raf6TRlQgfOeUiKzQTMr 2cUKnGqoaCVNwTWgFwX8WBusGAuOtak409h2VvQ3CHcvXBqAzS0XO6eXvXqZuKtn dihGyCGtLGTOzldCsI0k50xhQj2/9+UbkCs4YAL0muFzrlpA/i1CSi+XWb9arySc KJNjMx7wa77cbCDlWqJTrtAWK1ftHA0hWpddogxy5y16TpYq5fvBS5Nf0VkMZ28K +EFpINsnT3ed7xJbbWgDOKxb/8K8JgkB5N2PwYn3eyDAugawIeZTU6/WG1r1nFuq KbZnN/RZv1rB4Hqixl7Ic1sNWIB2zNQ0r0NsrHCVMLXe9Qq7cVnv2iMhYuXAs5Cj HdBZE5nJ2V2Tncd12m7qo8vAX+Fl+caUWYIPUlHz27IYVYZ1NFmvnZKVIuqXhJu/ 5xI1G5+ycYZ6+giWQY5WGm5YmPvc7ni/lcYu+KGOCk0xMFkz+9VmnZQ0N/LEHu9Y H/ufWYUexXmFMNyZKvZbTtXttynhVY6JxuiAuW3BACIoeaygYVNumk3n6ckfh3rF XBjoz8FbynUIf/3RUjEQukMKmwhYBJjMQvloqZMUYsptpMBYOtBww57uLZdRv8bD jFds+0vHI99uYFYhAn9ujtiBIbn2bsb0hLJvAgKNg+qRAAj1kuwhf/4mEQKvWUJT KKoUf43d/WBfnDH1/Sdz =OkwT -----END PGP SIGNATURE-----

2 1

Re: [8/8] webservices: Implement WsSetInputToBuffer.
by Marvin 22 Oct '15

22 Oct '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=17568 Your paranoid android. === w7u (32 bit reader) === reader.c:1390: Test failed: wrong data L"valueest" === w864 (32 bit reader) === reader.c:1390: Test failed: wrong data L"valueP\5868P" === w1064 (32 bit reader) === reader.c:1390: Test failed: wrong data L"valuez\c5d8z\2980z\2980z\2980z\26b8z\2980z\2980z\2980z\2980z\2980z\26b8z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\2980z\26b8z\70d3\6a13\e609" === w7pro64 (64 bit reader) === reader.c:1390: Test failed: wrong data L"value]" === w864 (64 bit reader) === reader.c:1390: Test failed: wrong data L"valueU" === w1064 (64 bit reader) === reader.c:1390: Test failed: wrong data L"value0"

1 0

Re: [PATCH 08/10] wined3d: make wined3d_device_update_surface operate with wined3d_texture and sub_resource_idx
by Henri Verbeet 22 Oct '15

22 Oct '15

On 19 October 2015 at 19:07, Riccardo Bortolato <rikyz619(a)gmail.com> wrote: > @@ -3774,12 +3764,25 @@ float CDECL wined3d_device_get_npatch_mode(const struct wined3d_device *device) > /* FIXME: Callers should probably use wined3d_device_update_sub_resource() > * instead. */ > HRESULT CDECL wined3d_device_update_surface(struct wined3d_device *device, > - struct wined3d_surface *src_surface, const RECT *src_rect, > - struct wined3d_surface *dst_surface, const POINT *dst_point) > + struct wined3d_texture *src_texture, unsigned int src_idx, const RECT *src_rect, > + struct wined3d_texture *dst_texture, unsigned int dst_idx, const POINT *dst_point) You should probably use wined3d_device_update_sub_resource() instead.

3 11

Re: [PATCH 06/10] d3d9: use wined3d_rendertarget_view_create instead of wined3d_rendertarget_view_create_from_surface in d3d9_surface_get_rendertarget_view
by Henri Verbeet 22 Oct '15

22 Oct '15

On 19 October 2015 at 19:07, Riccardo Bortolato <rikyz619(a)gmail.com> wrote: > + desc.format_id = resource_desc.format; > + desc.u.texture.level_idx = surface->sub_resource_idx; > + desc.u.texture.layer_idx = 0; This is wrong for cube texture surfaces.

2 8

Re: [3/8] webservices: Add a stub implementation of WsGetXmlAttribute.
by Marvin 22 Oct '15

22 Oct '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=17566 Your paranoid android. === w7u (32 bit reader) === reader.c:1329: Test failed: wrong data L"valueest" === w864 (32 bit reader) === reader.c:1329: Test failed: wrong data L"valueD\5868D" === w1064 (32 bit reader) === reader.c:1329: Test failed: wrong data L"valueo\c5d8o\2980o\2980o\2980o\26b8o\2980o\2980o\2980o\2980o\2980o\26b8o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\2980o\26b8o\5ee6\5456\8e4e" === w7pro64 (64 bit reader) === reader.c:1329: Test failed: wrong data L"valueb" === w864 (64 bit reader) === reader.c:1329: Test failed: wrong data L"valueT" === w1064 (64 bit reader) === reader.c:1329: Test failed: wrong data L"value'"

1 0