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

Re: [PATCH 7/7] wined3d: Use format id as index in formats table whenever possible.
by Henri Verbeet 10 Feb '16

10 Feb '16

On 10 February 2016 at 02:14, Józef Kucia <jkucia(a)codeweavers.com> wrote: > Signed-off-by: Józef Kucia <jkucia(a)codeweavers.com> > --- > dlls/wined3d/utils.c | 182 +++++++++++++++++++++++++++++++-------------------- > 1 file changed, 110 insertions(+), 72 deletions(-) > For what it's worth, I think there are a number of places that could just use wined3d_get_format(). If you did that as a separate patch before this one, this one could be a bit smaller.

1 0

Re: [PATCH 4/7] wined3d: Introduce table with typed format channels.
by Henri Verbeet 10 Feb '16

10 Feb '16

On 10 February 2016 at 02:14, Józef Kucia <jkucia(a)codeweavers.com> wrote: > +enum wined3d_channel_type > +{ > + C_UNUSED, > + C_UNORM, > + C_SNORM, > + C_UINT, > + C_SINT, > + C_FLOAT, > + C_DEPTH, > + C_STENCIL > +}; These would need to be called something like WINED3D_CHANNEL_TYPE_UNORM, etc. Those are a bit longer to type, but we could use a scheme similar to the opcode table in shader_sm4.c to keep the table readable. (Perhaps "I" -> SINT, "i" -> SNORM, "S" -> STENCIL, etc.) > +struct wined3d_typed_format_info > { > enum wined3d_format_id id; > enum wined3d_format_id typeless_id; > + enum wined3d_channel_type red_channel; > + enum wined3d_channel_type green_channel; > + enum wined3d_channel_type blue_channel; > + enum wined3d_channel_type alpha_channel; > }; Just a thought, but would it make things easier if the channel types were an array? It would probably make sense for struct wined3d_format at least to have something like struct { unsigned int size; unsigned int offset; enum wined3d_channel_type type; } channels[4];

1 0

Re: [PATCH 3/7] wined3d: Add remaining sRGB formats.
by Henri Verbeet 10 Feb '16

10 Feb '16

On 10 February 2016 at 02:14, Józef Kucia <jkucia(a)codeweavers.com> wrote: > diff --git a/include/wine/wined3d.h b/include/wine/wined3d.h > index df95da9..f314af3 100644 > --- a/include/wine/wined3d.h > +++ b/include/wine/wined3d.h > @@ -214,6 +214,13 @@ enum wined3d_format_id > WINED3DFMT_B5G5R5A1_UNORM, > WINED3DFMT_B8G8R8A8_UNORM, > WINED3DFMT_B8G8R8X8_UNORM, > + WINED3DFMT_B8G8R8A8_TYPELESS, > + WINED3DFMT_B8G8R8A8_UNORM_SRGB, > + WINED3DFMT_B8G8R8X8_TYPELESS, > + WINED3DFMT_B8G8R8X8_UNORM_SRGB, > + WINED3DFMT_BC7_TYPELESS, > + WINED3DFMT_BC7_UNORM, > + WINED3DFMT_BC7_UNORM_SRGB, Not an issue with this patch per se, but note that these aren't actually usable by anything until you update the conversion functions in dxgi and d3d11. (And I notice debug_dxgi_format() is missing a few formats as well.)

1 0

Re: ntdll: Skip unused import descriptors when loading libraries.
by Alexandre Julliard 10 Feb '16

10 Feb '16

Sebastian Lackner <sebastian(a)fds-team.de> writes: > @@ -901,7 +901,18 @@ static NTSTATUS fixup_imports( WINE_MODREF *wm, LPCWSTR load_path ) > status = STATUS_SUCCESS; > for (i = 0; i < nb_imports; i++) > { > - if (!(wm->deps[i] = import_dll( wm->ldr.BaseAddress, &imports[i], load_path ))) > + const IMAGE_IMPORT_DESCRIPTOR *descr = &imports[i]; > + const IMAGE_THUNK_DATA *import_list = get_rva( wm->ldr.BaseAddress, descr->u.OriginalFirstThunk ? > + (DWORD)descr->u.OriginalFirstThunk : (DWORD)descr->FirstThunk ); > + if (!import_list->u1.Ordinal) I feel it would be cleaner to do these checks inside import_dll(). -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] winebuild: When building implibs, invoke dlltool with a machine type. (try 2)
by Alexandre Julliard 10 Feb '16

10 Feb '16

Charles Davis <cdavis5x(a)gmail.com> writes: > That way, if the installed MinGW toolchain is multilib enabled (as mine > is), the import libraries will get built for the correct architecture. > > Try 2: Correct machine name for x86-64. I'm afraid this doesn't work either: /usr/bin/x86_64-w64-mingw32-dlltool: Machine 'x86-64' not supported winebuild: /usr/bin/x86_64-w64-mingw32-dlltool failed with status 1 The help says: -m --machine <machine> Create as DLL for <machine>. [default: i386:x86-64] possible <machine>: arm[_interwork], i386, mcore[-elf]{-le|-be}, ppc, thumb -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: usp10: ScriptPlace should honour the fZeroWidth attribute.
by Marvin 10 Feb '16

10 Feb '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=20325 Your paranoid android. === w1064 (32 bit usp10) === usp10.c:1437: Test failed: 0: uJustification incorrect (0) usp10.c:1437: Test failed: 1: uJustification incorrect (0) usp10.c:1437: Test failed: 2: uJustification incorrect (0) usp10.c:1437: Test failed: 3: uJustification incorrect (0) usp10.c:1437: Test failed: 4: uJustification incorrect (0) usp10.c:1437: Test failed: 5: uJustification incorrect (0) usp10.c:1437: Test failed: 6: uJustification incorrect (0) usp10.c:1437: Test failed: 7: uJustification incorrect (0) usp10.c:1437: Test failed: 8: uJustification incorrect (0) usp10.c:1437: Test failed: 9: uJustification incorrect (0) usp10.c:1453: Test failed: 3: uJustification incorrect (0) usp10.c:1453: Test failed: 5: fDiacritic incorrect (1) usp10.c:1453: Test failed: 5: fZeroWidth incorrect (1) usp10.c:1453: Test failed: 15: fDiacritic incorrect (1) usp10.c:1453: Test failed: 15: fZeroWidth incorrect (1)

2 1

Re: ntdll/tests: Simplify exception tests by using todo_wine_if.
by Marvin 10 Feb '16

10 Feb '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=20310 Your paranoid android. === wvistau64 (32 bit exception) === exception.c:485: Test failed: 42: Wrong exception address 003C0000/003C0001 exception: unhandled exception 80000004 at 003C0000 exception.c:485: Test failed: 42: Wrong exception address 77BAED97/003C0001 === w7u (32 bit exception) === exception.c:485: Test failed: 42: Wrong exception address 003F0000/003F0001 exception: unhandled exception 80000004 at 003F0000 exception.c:485: Test failed: 42: Wrong exception address 77CDE34F/003F0001 === w1064 (32 bit exception) === Failure running script in VM: network read timed out (Connect:AgentVersion.h:0/9) === w1064 (64 bit exception) === Failure running script in VM: network read timed out (Connect:AgentVersion.h:0/9)

2 1

Re: wlanapi: add a stub for WlanEnumInterfaces
by Sebastian Lackner 10 Feb '16

10 Feb '16

Hiho Austin, here some feedback on the patch as inline comments: On 05.02.2016 06:19, Austin English wrote: > Should fix https://bugs.winehq.org/show_bug.cgi?id=40034. Unfortunately, > the user is having issues building wine, and there's no download to verify > myself.. > > > attachment.txt > > > From 060d3427b61058fe7dc4da447743008d030da331 Mon Sep 17 00:00:00 2001 > From: Austin English <austinenglish(a)gmail.com> > Date: Sun, 24 Jan 2016 20:18:24 -0600 > Subject: [PATCH] wlanapi: add WlanOpenHandleFunction stub > > Signed-off-by: Austin English <austinenglish(a)gmail.com> > --- > dlls/wlanapi/main.c | 8 ++++++++ > dlls/wlanapi/wlanapi.spec | 2 +- > include/wlanapi.h | 36 ++++++++++++++++++++++++++++++++++++ > 3 files changed, 45 insertions(+), 1 deletion(-) > create mode 100644 include/wlanapi.h > > diff --git a/dlls/wlanapi/main.c b/dlls/wlanapi/main.c > index 0b0d100..cba9864 100644 > --- a/dlls/wlanapi/main.c > +++ b/dlls/wlanapi/main.c > @@ -24,6 +24,8 @@ > #include "winbase.h" > #include "wine/debug.h" > > +# include "wlanapi.h" Why is there a space between "#" and include? > + > WINE_DEFAULT_DEBUG_CHANNEL(wlanapi); > > DWORD WINAPI WlanOpenHandle(DWORD clientVersion, PVOID reserved, > @@ -49,3 +51,9 @@ BOOL WINAPI DllMain(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpvReserved) > > return TRUE; > } I would suggest to add your implementation above, below the other Wlan* functions. > + > +DWORD WINAPI WlanEnumInterfaces(HANDLE client, void *reserved, WLAN_INTERFACE_INFO_LIST *interface_list) MSDN says the second argument is a double-pointer. > +{ > + FIXME("%p, %p, %p) stub\n",client, reserved, interface_list); > + return ERROR_CALL_NOT_IMPLEMENTED; > +} > \ No newline at end of file > diff --git a/dlls/wlanapi/wlanapi.spec b/dlls/wlanapi/wlanapi.spec > index 79f9e43..22264c5 100644 > --- a/dlls/wlanapi/wlanapi.spec > +++ b/dlls/wlanapi/wlanapi.spec > @@ -3,7 +3,7 @@ > @ stub WlanConnect > @ stub WlanDeleteProfile > @ stub WlanDisconnect > -@ stub WlanEnumInterfaces > +@ stdcall WlanEnumInterfaces(ptr ptr ptr) HANDLEs should be specified as "long" in spec files. > @ stub WlanExtractPsdIEDataList > @ stub WlanFreeMemory > @ stub WlanGetAvailableNetworkList > diff --git a/include/wlanapi.h b/include/wlanapi.h > new file mode 100644 > index 0000000..15a52ac > --- /dev/null > +++ b/include/wlanapi.h > @@ -0,0 +1,36 @@ > +/* > + * Copyright (C) 2016 Austin English > + * > + * This library is free software; you can redistribute it and/or > + * modify it under the terms of the GNU Lesser General Public > + * License as published by the Free Software Foundation; either > + * version 2.1 of the License, or (at your option) any later version. > + * > + * This library is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > + * Lesser General Public License for more details. > + * > + * You should have received a copy of the GNU Lesser General Public > + * License along with this library; if not, write to the Free Software > + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA > + */ > + Some header-guard wouldn't hurt. > +typedef enum _WLAN_INTERFACE_STATE { > + WLAN_INTERFACE_STATE_UNUSED, Typically all new files use 4 space indentation. Did you copy it from somewhere? Also, where are all the enum fields listed on MSDN? https://msdn.microsoft.com/en-us/library/windows/desktop/ms706877(v=vs.85).… > +} WLAN_INTERFACE_STATE; > + > +typedef struct _WLAN_INTERFACE_INFO { > + GUID InterfaceGuid; > + WCHAR strInterfaceDescription[256]; > + WLAN_INTERFACE_STATE isState; > +} WLAN_INTERFACE_INFO, *PWLAN_INTERFACE_INFO; > + > +typedef struct _WLAN_INTERFACE_INFO_LIST { > + DWORD dwNumberOfItems; > + DWORD dwIndex; > + WLAN_INTERFACE_INFO InterfaceInfo[1]; > +} WLAN_INTERFACE_INFO_LIST, *PWLAN_INTERFACE_INFO_LIST; > + > +DWORD (WINAPI *WlanOpenHandleFunction)(DWORD, VOID *, DWORD *, HANDLE *); This declares a variable and doesn't belong in a header file. > +DWORD WINAPI WlanEnumInterfaces(HANDLE, VOID *, WLAN_INTERFACE_INFO_LIST *); Should also be a double-pointer here. > -- 2.7.0.rc3 > > >

1 0

Re: [v2 PATCH] include: Add IColumnsInfo interface
by Nikolay Sivov 10 Feb '16

10 Feb '16

Now you removed size_is() completely. Why?

1 0

Re: [PATCH v2 1/7] d3d11/tests: Test creation of typeless resources.
by Dmitry Timoshkov 09 Feb '16

09 Feb '16

Józef Kucia <jkucia(a)codeweavers.com> wrote: > + todo_wine_if(tests[i].todo) > + ok(tests[i].succeeds ? SUCCEEDED(hr) : FAILED(hr), "Test %u: Got unexpected hr %#x.\n", i, hr); Isn't an actual behaviour well defined and particular error/success codes should be used instead of generic SUCCEEDED()/FAILED()? -- Dmitry.

1 0