April 2008 - Wine-Devel - list.winehq.org

MinGW cross compilation fails for d3dx9_36/tests
by Paul Vriens April 7, 2008

April 7, 2008

Hi, We didn't have a new winetest executable for the last few days. MinGW doesn't know anything about a d3dx9 dll (yet) that's imported in the d3dx9_36 tests. Especially while in the process of getting ready for 1.0 we should have winetest run as much as possible. My main focus will be adding tests btw until 1.0 is out. These will be both tests to proof the correctness of the current implementation as well as tests to show missing stuff. There are loads of functions for which we have no tests yet and with the 1.0 in sight I think it becomes very important to have tests for virtually every function we (have) implement(ed). -- Cheers, Paul.

11 31

Proposal: versioning .wine directory
by Dan Kegel April 7, 2008

April 7, 2008

On the wine-users list, we're getting a lot of users who have old or even ancient .wine directories, and whose problems go away with a fresh .wine directory Perhaps we should have wineprefixcreate stamp the version of wine the .wine directory was created with, make wine-1.0 refuse to run with pre-1.0 .wine directories, and require that future versions of wine run properly with .wine directories created by any earlier stable release of wine. It wouldn't be hard, at least at first, and it might save a lot of support inquiries. What say?

9 16

Re: ntoskrnl.exe: [1/2] Implement KeQueryActiveProcessors
by Dmitry Timoshkov April 7, 2008

April 7, 2008

"Detlef Riekenberg" <wine.dev(a)web.de> wrote: > +KAFFINITY WINAPI KeQueryActiveProcessors( void ) > +{ > + DWORD_PTR AffinityMask; > + > + if (GetProcessAffinityMask( GetCurrentProcess(), &AffinityMask, NULL)) { > + return AffinityMask; > + } > + > + /* No Idea, what to do, when GetProcessAffinityMask fail. > + We use the first Processor */ > + return 1; > +} GetProcessAffinityMask() should never fail for the current process, so there is no need to add bogus code for that case. -- Dmitry.

1 0

Re: [PATCH 03/15] wined3d: In palette9_changed skip check if dxVersion <= 7.
by Stefan Dösinger April 6, 2008

April 6, 2008

Am Sonntag, 6. April 2008 23:05:52 schrieb Alexander Dorofeyev: > Stops it from producing misleading messages in apps that don't use d3d8/9 > style palettes. Is there no other way to do this? In my opinion dxVersion checks are ugly and dxVersion should go away

2 1

Re: d3dx9: Implementation of D3DXGetShaderSize
by H. Verbeet April 6, 2008

April 6, 2008

On 06/04/2008, Luis Busquets <luis.busquets(a)ilidium.com> wrote: > This function obtains the size of a shader bytestream. It does so looking > for the End token. > >+UINT D3DXGetShaderSize(DWORD * pFunction) I would expect this to take a const pointer. Also, there's no reason to use Microsoft's (broken) naming conventions for function arguments. > + while (*address!=0x0000ffff) { > + /*Let's check if the next DWORD is a comment*/ > + if ((*address&0x0000ffff)==0x0000fffe) { > + address=address+((*address >> 0x10) & 0x00007fff); > + } > + address=address+0x1; > + } It would help readability quite a bit to use the proper constants here. Also, please try to use an indentation style consistent with the rest of the dll and wine (this goes for the other patches as well). > + size=(address-pFunction+0x1)*sizeof(DWORD); Why the +1 here? Does it return the size without the end token itself? Tests and comments are always helpful there. The following (untested!) implementation looks more reasonable to me: UINT D3DXGetShaderSize(const DWORD *byte_code) { const DWORD *ptr = byte_code; if (!ptr) return 0; /* Look for the END token, skipping the VERSION token */ while (*++ptr != D3DSIO_END) { /* Skip comments */ if ((*ptr & D3DSI_OPCODE_MASK) == D3DSIO_COMMENT) { ptr += ((*ptr & D3DSI_COMMENTSIZE_MASK) >> D3DSI_COMMENTSIZE_SHIFT); } } /* Return the shader size in bytes, excluding the END token. */ return (ptr - (byte_code + 1)) * sizeof(*ptr); }

2 2

Re: d3dx8: Implementation of D3DXGetFVFVertexSize
by H. Verbeet April 6, 2008

April 6, 2008

On 06/04/2008, Luis Busquets <luis.busquets(a)ilidium.com> wrote: > Implementation of D3DXGetFVFVertexSize. It does so by calculating all the > different publicly available posible vertex configuration types. > Same comments as for the d3dx9 version apply. (Why isn't the d3dx9 version a forward to d3dx8?)

2 1

Pulse Audio -- Wine should have instructions on this web site
by Susan Cragin April 6, 2008

April 6, 2008

This site purports to give instructions on how to run certain applications, including Skype (which is 32-bit). I think wine should have instructions here too. http://www.pulseaudio.org/wiki/PerfectSetup It doesn't look like pulseaudio is going away from Ubuntu anytime soon. https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/198453

16 28

Re: [2/6] d3d9: Remove some test trace lines
by Allan Tong April 6, 2008

April 6, 2008

On Sat, Apr 5, 2008 at 5:18 PM, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > I think they're not needed any more now that the test works For what it's worth, this test fails for me with fbos. - Allan

2 1

Re: 0005-d3d9-add-stream_test-with-instancing
by Stefan Dösinger April 6, 2008

April 6, 2008

Am Sonntag, 6. April 2008 17:11:10 schrieb Rico Schüller: > Hi, > > this is a patch, which contains a test for instancing. It proves the > behavior from these two patches: Alexandre: Note that this test needs my D3DCREATE_SOFTWARE_VERTEXPROCESSING->D3DCREATE_HARDWARE_VERTEXPROCESSING to work on Windows

1 0

Re: 0004-wined3d-improve-drawStridedInstanced
by Stefan Dösinger April 6, 2008

April 6, 2008

Am Sonntag, 6. April 2008 17:07:45 schrieb Rico Schüller: > Hi, > > this patch solves a rendering problem with instances. Affected games are > Crysis and Gothic3. The test for this is in > 0005-d3d9-add-stream_test-with-instancing. So apply this first, > otherwise the tests will fail on wine. I think a comment why you're breaking out of the loop here would be helpful, perhaps mentioning the testcase in d3d9/tests. The same applies to the 0 instances -> 1 instance override

1 0