November 2007 - Wine-Devel - list.winehq.org

Re: WineD3D: Implement detection of ATI cards with Mesa DRI drivers
by James Hawkins Nov. 21, 2007

Nov. 21, 2007

On Nov 21, 2007 4:51 PM, Roman Mamedov <roman(a)rm.pp.ru> wrote: > Sorry, a small mistake. Always check everything twice, and when you > checked twice - check the third time :( > > --- > dlls/wined3d/directx.c | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/dlls/wined3d/directx.c b/dlls/wined3d/directx.c > index fa844b0..7e09014 100644 > --- a/dlls/wined3d/directx.c > +++ b/dlls/wined3d/directx.c > @@ -1145,7 +1145,7 @@ BOOL IWineD3DImpl_FillGLCaps(WineD3D_GL_Info > *gl_info) { > } else if (strstr(gl_info->gl_renderer, "R300")) { /* > r300_dri - all R3xx and R4xx-based cards */ > gl_info->gl_card = CARD_ATI_RADEON_9500; > vidmem = 128; > - } else if (strstr(gl_info->gl_renderer, "Radeon")) { /* > radeon_dri, for all R3xx-based cards */ > + } else if (strstr(gl_info->gl_renderer, "Radeon")) { /* > radeon_dri - all R100-based cards */ > /* This one should be the last case, so if the word > "Radeon" ever added to Mesa gl_renderer > * string of R200 or R300, this detection code will > keep working properly. > */ > -- > 1.5.3.4 > If you make a mistake, you have to fix the first patch and resend it. -- James Hawkins

1 0

No release this week
by Alexandre Julliard Nov. 21, 2007

Nov. 21, 2007

Folks, Because of the Thanksgiving weekend, there won't be a release this Friday. 0.9.50 will most likely be on next Friday. -- Alexandre Julliard julliard(a)winehq.org

1 0

Removing Windows and Solaris from the Downloads page
by Scott Ritchie Nov. 21, 2007

Nov. 21, 2007

It appears that the links for the Windows and Solaris versions on the winehq downloads page are very, very old. Since they're no longer maintained, we should just remove them, yes? Thanks, Scott Ritchie

7 9

Re: d3dx8 [patch 5/7] Implement D3DXQuaternionLn
by Andrew Talbot Nov. 21, 2007

Nov. 21, 2007

David.Adam(a)math.cnrs.fr wrote: > Hi David, There was no code in this patch. -- Andy.

1 0

something wrong with patches?
by Kirill K. Smirnov Nov. 21, 2007

Nov. 21, 2007

Hi, Is there anything wrong with the following patches? "winhelp: add basic support for ancient win3.0 help files" http://www.winehq.org/pipermail/wine-patches/2007-November/046886.html and comctl32: Implemented algorithm is a mergesort, not a quicksort http://www.winehq.org/pipermail/wine-patches/2007-November/046943.html Did I miss something important? Regards Kirill

1 0

Re: [6/9] d3d9: add a test for vertex shader input matching
by Alexandre Julliard Nov. 21, 2007

Nov. 21, 2007

Stefan Dösinger <stefan(a)codeweavers.com> writes: > From 44f176ad4fcbc92ea66d8526670a1ec573ffa9f0 Mon Sep 17 00:00:00 2001 > From: Stefan Doesinger <stefan(a)codeweavers.com> > Date: Mon, 19 Nov 2007 10:53:26 +0100 > Subject: [PATCH] d3d9: add a test for vertex shader input matching This fails with the usual rounding errors: ../../../tools/runtest -q -P wine -M d3d9.dll -T ../../.. -p d3d9_test.exe.so visual.c && touch visual.ok visual.c:4716: Test failed: Input test: Quad 1(2crd) returned color 0x00ffff7f, expected 0x00FFFF80 visual.c:4721: Test failed: Input test: Quad 3(2crd-wrongidx) returned color 0x00ff007f, expected 0x00FF0080 visual.c:4704: Test failed: Input test: Quad 1(2crd) returned color 0x00ffff7f, expected 0x00FFFF80 visual.c:4710: Test failed: Input test: Quad 3(2crd-wrongidx) returned color 0x00ff007f, expected 0x00FF0080 visual.c:4704: Test failed: Input test: Quad 1(2crd) returned color 0x00ffff7f, expected 0x00FFFF80 visual.c:4710: Test failed: Input test: Quad 3(2crd-wrongidx) returned color 0x00ff007f, expected 0x00FF0080 visual.c:4849: Test failed: Input test: Quad 2(color-ubyte) returned color 0x00407ffe, expected 0x014080ff visual.c:4849: Test failed: Input test: Quad 2(color-ubyte) returned color 0x00407ffe, expected 0x014080ff visual.c:4849: Test failed: Input test: Quad 2(color-ubyte) returned color 0x00407ffe, expected 0x014080ff make[2]: *** [visual.ok] Error 9 -- Alexandre Julliard julliard(a)winehq.org

1 0

[0/3] RSA AES provider to rsaenh
by Vijay Kiran Kamuju Nov. 20, 2007

Nov. 20, 2007

On Nov 20, 2007 4:23 PM, Juan Lang <juan.lang(a)gmail.com> wrote: > > The test fail due to missing implementation, Hence I put them in todo's. > > After the implementation is there, they do not fail. > > I think thats the procedure followed. > > I have put the remove todo's in separate patch, as its large. > > James is correct, patches 2 and 3 should be concatenated. 2 is > already necessarily large, and adding 3 to it doesn't add much to its > size. > --Juan > Damn, I have to reset the git tree. Then apply patches manually and resend them :( Should I wait.................? :( --- VJ

2 1

Re: [0/3] RSA AES provider to rsaenh
by Juan Lang Nov. 20, 2007

Nov. 20, 2007

> Damn, I have to reset the git tree. > Then apply patches manually and resend them :( This isn't hard. Assuming you have these as the last three patches in your tree: git reset HEAD~2 git commit -a This will rewind your tree two steps without changing your local files, then commit them again. The more manual method, assuming you have the last patches named 2 and 3: git reset --hard HEAD~2 patch -p1 < 2 patch -p1 < 3 git commit -a This latter method isn't necessary in this case, but it sometimes comes in handy when you have some patches intermingled, and you want to commit several of them in a bunch without committing some in between. E.g. if you wanted to merge patches 1 and 3 into a single patch, you could do it this way instead. HTH, --Juan

1 0

Re: [0/3] RSA AES provider to rsaenh
by James Hawkins Nov. 20, 2007

Nov. 20, 2007

On 11/20/07, Vijay Kiran Kamuju <infyquest(a)gmail.com> wrote: > Hi, > > This is a description of patch series which I am now going to send. > It adds the new Enhanced RSA AES provider to the rsaenh. > The one application which I know uses it is Office 2007. > The first patch contains the tests for the RSA AES provider. > the second patch contains the RSA AES provider implementation > the last patch removes the todo's in the tests. > > The implementation and tests have been verified by Juan Lang. > You can't fix the tests and remove the todo_wines in separate patches. The tests will fail inbetween commits. -- James Hawkins

3 2

Re: bits: Description of patch set (resend 2) (0/26)
by James Hawkins Nov. 20, 2007

Nov. 20, 2007

On Nov 19, 2007 6:19 PM, Roy Shea <roy(a)cs.hmc.edu> wrote: > Howdy, > > This is another resend of the BITS patch set. This resend changes > svnhost to be ANSI based, merges patches 25 and 26 (old numbering), > and removes extra headers and linked libraries. > Any particular reason for using ansi? Generally we're moving everything towards unicode and away from ansi. Also, the entire patch set is relatively large, and you might want to consider breaking the set up into chunks (svnhost, bits cleanup etc, bits interface impl) and waiting till the first chunk is committed before moving on to the next. That'll make it easier to get the patches committed (assuming they don't get accepted this time) and it'll keep the mail count down for list members. -- James Hawkins

3 2