March 2010 - Wine-devel - list.winehq.org

Re: 1/3 wined3d: Extend surface_get_depth_blt_info with texture coordinate code from surface_blt_to_drawable
by Henri Verbeet 12 Mar '10

12 Mar '10

On 11 March 2010 22:25, Roderick Colenbrander <thunderbird2k(a)gmail.com> wrote: > +static void surface_get_depth_blt_info(GLenum target, RECT *rect_in, GLsizei w, GLsizei h, struct depth_blt_info *info) "rect_in" should be const, looks ok otherwise.

1 0

Re: 1/2 wined3d: Implement IWineD3DDeviceImpl_ColorFill using ClearSurface instead of 'BltOverride'
by Stefan Dösinger 12 Mar '10

12 Mar '10

Am 11.03.2010 um 22:34 schrieb Roderick Colenbrander: > Hi, > > This patch makes IWineD3DDeviceImpl_ColorFill use Clear instead of > BltOverride for color filling (it falls back to software blit if this > is not possible). This change makes it possible in a next patch to > have the GL Blt call IWineD3DDevice_ColorFill instead of the way > around. > > Regards, > Roderick Colenbrander > <0006-Implement-IWineD3DDeviceImpl_ColorFill-using-ClearSu.patch> I am still wary of the changes to Blit because I have not seen(or have not understood) the overall design goal of how blits should work, and especially how ddraw blits and d3d8/9 style blits work together. This is especially visible in your patch because you still call IWineD3DBaseSurfaceImpl_Blt from ColorFill, you just moved the software vs gl codepath decision to a different place. It may be a step in the right direction, but I'd still prefer to see the overall strategy first.

2 1

winemp3.acm on the mac
by Aric Stewart 12 Mar '10

12 Mar '10

Hello all, So I have been hacking on winemp3.acm on the mac to make use of CoreAudio for mp3 decoding and I have just gotten it working. So now I am faced with a questions. Do I setup winemp3.acm with #ifdefs to pick between the libmpg123 or CoreAudio paths, leaving us with 1 winemp3.acm that can do both. Or do I create a new winecoreaudio.acm (or such) that just contains the CoreAudio path. I see pluses and minuses to each and am unsure which would be better. I think personally I would prefer the first option however that potentially really complicates the build system and then what happens if someone has a mac with the libmpg123 headers. Adding a new acm would not be too problematic if people feel that would be better. The advantage there is it would be possible to expand that acm to handle all the types the CoreAudio handles and not just mp3. thanks! -aric

4 3

Re: usp10: Bug 17324 change default glyph returned from 0xffff to 0
by Hans Leidekker 12 Mar '10

12 Mar '10

On Thursday 11 March 2010 13:51:19 Jeff Latimer wrote: > This patch fixes bug 17324 http://bugs.winehq.org/show_bug.cgi?id=17324 > The problem is that ScriptShape convert characters to glyphs and there > is no translation for '\r'. ScriptTextOut is called and > X11DRV_XRender_ExtTextOut is called eventually. The glyph is used as an > index into an array but as the glyph value is 0xffff it causes an exception. > > This patch converts the NONEXISTING_GLYPHS to 0 (this is the same > behaviour as Windows). It doesn't apply cleanly: patching file dlls/usp10/tests/usp10.c Hunk #1 succeeded at 636 (offset -20 lines). Hunk #2 succeeded at 1411 (offset -20 lines). patching file dlls/usp10/usp10.c Hunk #1 succeeded at 1309 (offset -2 lines). And it has compile warnings: usp10.c: In function ‘test_ScriptTextOut3’: usp10.c:664: warning: unused variable ‘sla’ usp10.c:663: warning: unused variable ‘psla’ usp10.c:662: warning: unused variable ‘fTrailing’ usp10.c:661: warning: unused variable ‘iCP’ usp10.c:660: warning: unused variable ‘piX’ It would be better to add this test to the existing tests for ScriptTextOut. As for the fix, why not remove the GGI_MARK_NONEXISTING_GLYPHS flag? -Hans

2 1

Test for ATI2n on real Windows machines
by REalm 12 Mar '10

12 Mar '10

Hello, I need results from real Windows machines. This is to test behaviour of ATI2N format, which seems to behave differently on different driver versions (at least on Nvidia 9600M cards). Please run d3d9_crosstest.exe surface and d3d9_crosstest.exe visual and send me output. Compiled test is at http://realm.lt/d3d9_crosstest.exe or it can be compiled using atached patch. Ričardas Barkauskas REalm >From 860dc38c1051e19b8de9aa75091b9c6b9e14e0a9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ri=C4=8Dardas=20Barkauskas?= <miegalius(a)gmail.com> Date: Thu, 11 Mar 2010 21:46:01 +0200 Subject: testing2 --- dlls/d3d9/tests/surface.c | 32 +++++++++++++++++++++++++------- 1 files changed, 25 insertions(+), 7 deletions(-) diff --git a/dlls/d3d9/tests/surface.c b/dlls/d3d9/tests/surface.c index 396e0c4..e5aa3ae 100644 --- a/dlls/d3d9/tests/surface.c +++ b/dlls/d3d9/tests/surface.c @@ -187,8 +187,10 @@ todo_wine ok(SUCCEEDED(hr), "Double IDirect3DTexture9_UnLockRect failed with % static void test_lockrect_offset(IDirect3DDevice9 *device) { IDirect3DSurface9 *surface = 0; + IDirect3D9 *d3d; const RECT rect = {60, 60, 68, 68}; D3DLOCKED_RECT locked_rect; + int expected_pitch; unsigned int expected_offset; unsigned int offset; unsigned int i; @@ -202,14 +204,24 @@ static void test_lockrect_offset(IDirect3DDevice9 *device) unsigned int block_height; unsigned int block_size; } dxt_formats[] = { - {D3DFMT_DXT1, "D3DFMT_DXT1", 4, 4, 8}, - {D3DFMT_DXT2, "D3DFMT_DXT2", 4, 4, 16}, - {D3DFMT_DXT3, "D3DFMT_DXT3", 4, 4, 16}, - {D3DFMT_DXT4, "D3DFMT_DXT4", 4, 4, 16}, - {D3DFMT_DXT5, "D3DFMT_DXT5", 4, 4, 16}, + {D3DFMT_DXT1, "D3DFMT_DXT1", 4, 4, 8}, + {D3DFMT_DXT2, "D3DFMT_DXT2", 4, 4, 16}, + {D3DFMT_DXT3, "D3DFMT_DXT3", 4, 4, 16}, + {D3DFMT_DXT4, "D3DFMT_DXT4", 4, 4, 16}, + {D3DFMT_DXT5, "D3DFMT_DXT5", 4, 4, 16}, + {MAKEFOURCC('A','T','I','2'), "ATI2N", 4, 4, 16}, }; + hr = IDirect3DDevice9_GetDirect3D(device, &d3d); + ok(SUCCEEDED(hr), "IDirect3DDevice9_GetDirect3D failed (%08x)\n", hr); for (i = 0; i < (sizeof(dxt_formats) / sizeof(*dxt_formats)); ++i) { + hr = IDirect3D9_CheckDeviceFormat(d3d, 0, D3DDEVTYPE_HAL, D3DFMT_X8R8G8B8, 0, D3DRTYPE_TEXTURE, dxt_formats[i].fmt); + if(FAILED(hr)) + { + skip("Format %s not supported, skipping lockrect offset test\n", dxt_formats[i].name); + continue; + } + hr = IDirect3DDevice9_CreateOffscreenPlainSurface(device, 128, 128, dxt_formats[i].fmt, D3DPOOL_SCRATCH, &surface, 0); ok(SUCCEEDED(hr), "CreateOffscreenPlainSurface failed (%08x)\n", hr); @@ -217,6 +229,10 @@ static void test_lockrect_offset(IDirect3DDevice9 *device) ok(SUCCEEDED(hr), "LockRect failed (%08x)\n", hr); base = locked_rect.pBits; + expected_pitch = (128 + dxt_formats[i].block_height - 1) / dxt_formats[i].block_width + * dxt_formats[i].block_size; + //ok(locked_rect.Pitch == expected_pitch, "Got pitch %d, expected pitch %d for format %s\n", locked_rect.Pitch, expected_pitch, dxt_formats[i].name); + ok(0, "Got pitch %d, expected pitch %d for format %s\n", locked_rect.Pitch, expected_pitch, dxt_formats[i].name); hr = IDirect3DSurface9_UnlockRect(surface); ok(SUCCEEDED(hr), "UnlockRect failed (%08x)\n", hr); @@ -227,15 +243,17 @@ static void test_lockrect_offset(IDirect3DDevice9 *device) ok(SUCCEEDED(hr), "LockRect failed (%08x)\n", hr); offset = (BYTE *)locked_rect.pBits - base; - expected_offset = (rect.top / dxt_formats[i].block_height) * locked_rect.Pitch + expected_offset = (rect.top / dxt_formats[i].block_height) * expected_pitch + (rect.left / dxt_formats[i].block_width) * dxt_formats[i].block_size; - ok(offset == expected_offset, "Got offset %u, expected offset %u for format %s\n", offset, expected_offset, dxt_formats[i].name); + //ok(offset == expected_offset, "Got offset %u, expected offset %u for format %s\n", offset, expected_offset, dxt_formats[i].name); + ok(0, "Got offset %u, expected offset %u for format %s\n", offset, expected_offset, dxt_formats[i].name); hr = IDirect3DSurface9_UnlockRect(surface); ok(SUCCEEDED(hr), "UnlockRect failed (%08x)\n", hr); IDirect3DSurface9_Release(surface); } + IDirect3D9_Release(d3d); } static void test_lockrect_invalid(IDirect3DDevice9 *device) -- 1.7.0

2 1

Updated popular game status page: wine handles 49% of popular games well
by Dan Kegel 12 Mar '10

12 Mar '10

I've automated http://wiki.winehq.org/GameChecklist so it covers the last ten years now. According to the data, Wine handles about half of all popular games well (i.e. has a gold or platinum rating for them). Woot!

1 0

Re: shell32/tests: add tests for the parser of SHELLEXECUTEINFO.lpFile [try 7]
by Ilya Basin 11 Mar '10

11 Mar '10

Any news on this topic?

3 2

Re: mciavi: Prevent deadlock when closing while playing or when paused.
by Maarten Lankhorst 11 Mar '10

11 Mar '10

Hi Joerg, 2010/3/11 <Joerg-Cyril.Hoehle(a)t-systems.com>: > Hi, > > alas, even though mciavi contains lots of Critical Sections, > it's not enough to protect against 2 threads calling Play and Close > concurrently. I believe something entirely different would be > needed: either a complete serialization of messages (like the OSS > and ALSA drivers), or a concept of transactions. That code sucks and as soon as possible I'm getting rid of it. Don't duplicate it please. It can be done without such a horrible ring buffer design, since there is really no reason to have a separate thread alive for audio. Using a winmm timer there would remove the need for that thread altogether, and as part of my sound rewrite I will write a new winmm driver that forwards to the new audio api, regardless of what that api might be. So rest of wine-devel, don't restart the flamewar in this thread. ;) Cheers, Maarten

1 0

Re: [2/2] cmd: Add escape carat char at end of line feature and test cases
by Michael Stefaniuc 11 Mar '10

11 Mar '10

Hello John, welcome to Wine development and thanks for your patches. John Chow wrote: > Patch fixes the missing feature to escape newlines with the ^ char. > Included are also some test cases for this. The second patch has a typo: all "caret" are written as "carat", e.g. + BOOL lastWasCarat = FALSE; Could you please change that typos and resubmit? Thanks bye michael

1 0

Is there something we can do about Java?
by Scott Ritchie 11 Mar '10

11 Mar '10

I've run into a few applications that simply assume Java is installed. They die horribly with a "javaw" not found. My first inclination is to just pop up a message explaining what went wrong and how to fix it, with a clickable link to go Sun's Java download page. This struck me as similar to the Gecko situation. In principle, we could have a wine-java package containing Sun's binaries and then just install them automatically when they're on the system, referring the user to the website otherwise. Distributions would then need to make these packages (just like wine-gecko), and also get separate permission from Sun. This strikes me as rather feasible - Ubuntu already ships Sun's Java in its parter repository - however we'd need to work out how it would work technically here first. Ideally, Sun's installer wouldn't need to be modified. Thanks, Scott Ritchie

9 11