September 2006 - Wine-Devel

Whither msxml?
by Dan Kegel Sept. 25, 2006

Sept. 25, 2006

For the past two years, I've helped UCLA's cs130 by pointing them at an area of Wine that needs improvement and helping them get patches submitted. For Winter 2007, I'm considering having them focus on msxml3. So I'm starting to look at bugzilla (see my previous message) and dig up demo apps and articles that might help us understand where wine's msxml needs to go. Man, oh, man are there a lot of pages about msxml out there. If anyone has tips on sore spots with Wine's msxml, please let me know. A few random notes: MSDN on msxml: http://windowssdk.msdn.microsoft.com/en-us/library/ms763742.aspx I see the spec file for msxml3 has a bunch of stubbed ordinals. Anyone know what they are? Can Wine handle the "msxml3 jumpstart" demo app from year 2000 yet? http://msdn.microsoft.com/library/en-us/dnmsxml/html/sax2jumpstart.asp Or the "msxsl.exe" wrapper app? http://www.microsoft.com/downloads/details.aspx?familyid=2fb55371-c94e-4373… "Programming The SAX2 3.0 Using MSXML" (examples in vb6) http://xml.sys-con.com/read/40210.htm "MSXML, It's Not Just for VB Programmers Anymore" (examples in perl) http://www.perl.com/pub/a/2001/04/17/msxml.html "Python & XML" http://safari.oreilly.com/0596001282 Here's a page that seems to show how to use msxsl and explains differences between msxml3 and msxml4: http://www.perfectxml.com/articles/xml/XSLTInMSXML.asp msxml3 examples: http://www.perfectxml.com/articles/xml/msxml30.asp "A C++ Template Wrapper for the XML SAX API" http://www.ddj.com/showArticle.jhtml?articleID=184401778 etc. etc. - Dan

3 2

An MSVCRT_fgetc bug
by Tobias Ringström Sept. 25, 2006

Sept. 25, 2006

There's a bug in MSVCRT_fgetc in 0.9.21 (likely introduced in 0.9.19) in that it sometimes sign extends the byte read from the file. The following program illustrates the problem: #include <stdio.h> int main() { FILE *f = fopen("tmp.bin", "w+"); fputc(0xe0, f); fputc(0xe0, f); rewind(f); printf("0x%08x\n", fgetc(f)); printf("0x%08x\n", fgetc(f)); fclose(f); return 0; } The output is: 0x000000e0 0xffffffe0 (should be 0x000000e0 too) The bug is likely this line: http://source.winehq.org/source/dlls/msvcrt/file.c#L2134 That line is now i = file->_ptr++; but should be i = *(unsigned char*)(file->_ptr++); I don't have a build environment for Wine, and it felt like overkill to set one up for this little bug, so I've not been able to verify my hypothesis. I hope that's acceptable. /Tobias

2 1

Re: LOSTWAGES: WWN 292 & 320 updates
by Tom Wickline Sept. 25, 2006

Sept. 25, 2006

On 9/23/06, Mirek <thunder.m(a)email.cz> wrote: > > > > Also Wine reports 64MB of video memory not 16MB > > I think there shuld be option in winecfg to set this, because it is not easy to > find how anyone can do that. http://wiki.winehq.org/UsefulRegistryKeys Tom

1 1

Re: Governance revisited (Wineconf report)
by Ge van Geldorp Sept. 25, 2006

Sept. 25, 2006

> From: Kai Blin <kai.blin(a)gmail.com> > > On Saturday 23 September 2006 10:32, Scott Ritchie wrote: > > Frankly, all we really need is for Alexandre to write a 10-second > > reply to wine-devel for each patch he rejects. > > On WineConf, we decided against this. That would still slow > down the overall patch submission speed. Consider you have a > patch that's just fine, but before you sent that, I sent in > ten patches with C++ style comments. Alexandre would now have > to reply to ten patches with "No C++ style comments" before > processing your patch. Everybody reading wine-patches > could point out what was wrong with my patches. You must be kidding. Even in your somewhat convoluted example, it would be fine to receive just a single notification "No C++ style comments" and I really wouldn't care if the notification came from Alexandre or from someone else on wine-patches. The reality is that a lot of patches (I haven't kept a precise count, but I estimate about a third of my own patches) are dropped silently, without any feedback at all. I have absolutely no problem with a strict patch acceptance policy, designed to keep the Wine quality high. Alexandre is amazingly smart and when he tells me why a certain patch was rejected I'll usually agree with him and even if I don't agree, I have enough confidence in him to accept his judgement. But having your patches dropped silently is pretty annoying. The usual response from the core Wine developers is "well, just resubmit". I fail to see how this helps Alexandres productivity (or my own for that matter). If patches weren't dropped silently the sequence of events would be: 1) Submit patch 2) Alexandre looks at patch 3) Alexandre writes 10-second rejection reply With the resubmit: 1) Submit patch 2) Alexandre looks at patch 3) Wait a week 4) Resubmit patch 5) Alexandre looks at patch again 6) Alexandre writes 10-second rejection reply Seems to me this incurs extra work for Alexandre (number 5). The patch processes isn't very transparant. For example, I submitted a patch on 20 June 2006 to dlls/kernel/heap.c (www.winehq.org seems to be unreachable for me at the moment so I can't give a URL to the message) to fix a problem with the heap on Win64. This patch was silently dropped. On July 21 a change very similar to what I submitted was committed but not attributed to me. Now, either the commit was not based on my patch (in which case someone else spent another 2-3 days (it was one of those problems where a chain of out-of-buffer memory writes take place and you have to work your way up the chain to find the root cause) on tracking down an already known problem for which a fix was already available) or it was based on my patch (in which case my patch shouldn't have been dropped and I should have been given credit). Like I said before, I have a lot of respect for Alexandres technical abilities. But when I read comments in the Wineconf report about git: "Developers might not like it, but Alexandre does so it's a success" (did I mention I dislike git also???) and the inability of the Wine community to set up a patch management system (so patches won't disappear into the big void) because Alexandre refuses to use it if it won't work in Emacs I'm starting to wonder if people realise that the developers don't work for Alexandre. He's a great Benovelent Dictator on technical issues, but in my opinion only a Dictator on patch process management. Ge van Geldorp.

2 1

Governance revisited
by Steven Edwards Sept. 25, 2006

Sept. 25, 2006

---------- Forwarded message ---------- From: Steven Edwards <winehacker(a)gmail.com> Date: Sep 25, 2006 1:27 AM Subject: Re: Governance revisited To: Troy Rollo <wine(a)troy.rollo.name> On 9/25/06, Troy Rollo <wine(a)troy.rollo.name> wrote: > The present system turns people off even before you've had time to learn > whether they are capable or not. Which is why we want to have the ambassadors project to help new people in to wine. The thinking goes that if we have some people to help hold the hands of new developers and the developers that are defacto maintainers of a certain section of code will respond to patches as they seem them, this will free julliard from having to answer every single patch with a reply. Now in the case of Ge where he was patching core functions of ntdll,kernel32,wineserver,etc I guess it would still fall to julliard to reply personally but in the case of other modules the experts in that area should take a step up and monitor wine-patches and say what patch X.Y.Zsucks and julliard most likely will not merge it. -- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo -- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

2 1

my dsound/winealsa hacks
by Tomas Carnecky Sept. 24, 2006

Sept. 24, 2006

My ultimate goal was to solve the dsound underruns which were so horrible that I had to disable sound in World of Warcraft. While I managed to get the sound working flawlessly (really... I never heard such clear sound under wine) in WoW, it required WoW-specific hacks so my patch will never make it (in its current form) into the official git tree, but maybe someone can use some of my ideas to improve dsound/winealsa. The basic idea is to let alsa mix the sound instead of the infamous dsound mixer. The advantage is that if the hardware supports mixing, there is very little overhead, if not, there's still dmix which can do that, but dsound doesn't need to care how it's done, it's entirely up to alsa. Now to the actual implementation, it's pretty straight-forward: I 'forced' CreateSoundBuffer to create a hardware buffer (right now it creates a hw buffer only for the primary buffer) and changed the winealsa driver to support that. Because winealsa driver opens only one 'connection' (snd_pcm_t) to the soundcar (allowing only one buffer per device), I simply opened a new 'connection' for each buffer and configured the hardware for 44100Hz/U16LE/stereo (that's what WoW expects). Up to this point it was easy. One of the big problems was that WoW requests a 16kb buffer, but alsa is unable to allocate a buffer with this exact size, and it caused problems if I passed the alsa buffer to WoW (so WoW could write directly to it). I had to allocate a separate buffer, keep track of the play/write positions in both buffers and copy the data from one to the other. I'm fairly sure this all could be done in a WoW-independent way, eg. configure the hardware as the application requests it (by passing LPCDSBUFFERDESC to the low-level driver etc) and keeping track of the read/write positions could also be done in a better way. There are a free questions that should be answered before someone can go on and make a 'proper' implementation. I simply bypassed the primary buffer, it doesn't exist since the data from the secondary buffers are written directly to the soundcard. If it should be possible to read the data from the primary buffer then we should forget this all and look for another solution. I believe that it's not possible (it's not possible to write to the primary buffer and the API doesn't differentiate between reads and writes), but someone should test that under windows. (this is only one question but I'm sure you'll come up with more ;) ) Here is the complete patch, I also changed the IDirectSound interface and a lot unrelated code, so the patch is a bit big :-/ There also are a few memory leaks, I didn't bother freeing the memory etc. http://dbservice.com/ftpdir/tom/dsound-alsa.patch tom

6 10

Call for Installer Bugs...
by Dan Kegel Sept. 24, 2006

Sept. 24, 2006

Lots of progress has been made in fixing installer bugs lately, but there are probably many such bugs hiding in plain view still. It would be very helpful if people could look for bugs which: a) appear to be MSI or setupapi related, and b) occur in freely downloadable executables (e.g. trial versions) c) aren't already in bugzilla (see http://bugs.winehq.org/buglist.cgi?product=Wine&component=wine-msi&keywords… for list) and file reports as appropriate at http://bugs.winehq.org. Thanks! - Dan

4 3

RFC: wine ASIO driver available for testing
by Robert Reif Sept. 24, 2006

Sept. 24, 2006

I just uploaded a simple wine ASIO driver to http://bugs.winehq.org/show_bug.cgi?id=2161 for testing and feedback. Please look at it and give me feedback. Playback works. Recording is untested due to lack of software. Full duplex doesn't work for me but I'm not sure if it's a software issue or a problem with my hardware and OS setup. The software is a long way from being ready for submission so any feedback or help would be appreciated.

4 4

Re: LOSTWAGES: WWN 292 & 320 updates
by H. Verbeet Sept. 24, 2006

Sept. 24, 2006

On 23/09/06, Mirek <thunder.m(a)email.cz> wrote: > > I also changed what benchmarks I had running at this years WineConf > > 3DMark2000, 3DMark2001SE and 3DMark2003... I have 3DMark2006 running > > and its beautiful, well beauty is in the eye of the beholder > > All of the test run in all three of the benchmarks to completion and > > there is only two test out of all test that have minor problems, and > > Henri has one of the test fixed in his tree. > > Can you post how i can get sources from his tree? The 3th and 4th 3DMark03 tests should pretty much work with current git, actually. The 4th test has some minor issues, afaik because of the wrong sampler types being bound. If there are other issues, those are separate bugs, and should probably be reported on bugzilla.

1 0

Server instability warning
by Jeremy White Sept. 24, 2006

Sept. 24, 2006

Hi Folks, The server crashed early yesterday evening, and again sometime over the night. It's a new server - and in fact, I think it's the new 'new' server - seems like it always takes a while to get a server stabilized). Jeremy Newman is coming back from England today, and I prefer to leave these sort of repairs to him, otherwise he grouses at me (rightly, I might add :-/). So it's back up now, but I'm not sure for how long that will be true. Sorry for the hassle; if it gets worse this afternoon before Jer gets back, I may break down and start dissecting it. Cheers, Jeremy

1 0

ALSA implementation
by Aaron Slunt Sept. 24, 2006

Sept. 24, 2006

Hi, I talked on #winehq in irc about how horribly alsa is implemented right now. I was told that winmm would need a complete re-write and the dsound needs some changes as well. My question is: anybody here who's familiar with all of the changes that need to be made to get alsa working properly? Thanks.

9 21

Re: Governance revisited (Wineconf report)
by J. Wesley Cleveland Sept. 24, 2006

Sept. 24, 2006

> From: Kai Blin <kai.blin(a)gmail.com> > On Saturday 23 September 2006 10:32, Scott Ritchie wrote: > > Frankly, all we really need is for Alexandre to write a 10-second reply > > to wine-devel for each patch he rejects. > > On WineConf, we decided against this. That would still slow down the overall > patch submission speed. Consider you have a patch that's just fine, but > before you sent that, I sent in ten patches with C++ style comments. > Alexandre would now have to reply to ten patches with "No C++ style comments" > before processing your patch. Everybody reading wine-patches could point out > what was wrong with my patches. How about emacs macros that would send "Patch rejected" or "Patch under consideration" to wine-patches or wine-devel ? This would get rid of the feeling that patches are dropping into a black hole while only costing Alexandre a fraction of a second per patch.

1 0

Re: Minimize window via style change
by Dmitry Timoshkov Sept. 23, 2006

Sept. 23, 2006

"Sebastian Reichelt" <SebastianR(a)gmx.de> wrote: > + if (winpos->flags & SWP_FRAMECHANGED) > + { > + DWORD style = GetWindowLongW( winpos->hwnd, GWL_STYLE ); > + X11DRV_set_iconic_state( winpos->hwnd ); > + if (style & WS_MINIMIZE) > + return TRUE; > + } > + > if (!SWP_DoWinPosChanging( winpos, &newWindowRect, &newClientRect )) return FALSE; > > /* Fix redundant flags */ X11DRV_set_iconic_state() call should be moved under the 'if' case. Next, you can't really avoid calling SWP_DoWinPosChanging and probably sending WM_WINDOWPOSCHANGED. This behaviour needs a good test case. -- Dmitry.

1 0

compiling wine for amd64
by Gerald Britton Sept. 23, 2006

Sept. 23, 2006

Hi -- I'm running ubuntu dapper on amd64. I want to compile wine but am hitting problems. I got the dependencies just fine, but the compile died. Here's what I get: $ sudo apt-get -y --build source wine Reading package lists... Done Building dependency tree... Done Skipping the already downloaded file 'wine_0.9.21~winehq0~ubuntu~6.06-1.dsc' Skipping the already downloaded file 'wine_0.9.21~winehq0~ubuntu~6.06.orig.tar.gz' Skipping the already downloaded file 'wine_0.9.21~winehq0~ubuntu~6.06-1.diff.gz' Need to get 0B of source archives. Skipping unpack of already unpacked source in wine-0.9.21~winehq0~ubuntu~6.06 dpkg-buildpackage: source package is wine dpkg-buildpackage: source version is 0.9.21~winehq0~ubuntu~6.06-1 dpkg-buildpackage: source changed by Scott Ritchie <scott(a)open-vote.org> dpkg-buildpackage: host architecture amd64 debian/rules clean dh_testdir dh_testroot rm -f build-stamp # Add here commands to clean up after the build process. /usr/bin/make distclean make[1]: Entering directory `/home/jerryb/wine-0.9.21~winehq0~ubuntu~6.06' make[1]: *** No rule to make target `distclean'. Stop. make[1]: Leaving directory `/home/jerryb/wine-0.9.21~winehq0~ubuntu~6.06' make: [clean] Error 2 (ignored) #-/usr/bin/make -C documentation clean cp -f /usr/share/misc/config.sub config.sub cp -f /usr/share/misc/config.guess config.guess dh_clean debian/rules build dh_testdir # Add here commands to configure the package. CFLAGS="-Wall -g -O2" ./configure --host=x86_64-linux-gnu --build=x86_64-linux-gnu --prefix=/usr --mandir=\${prefix}/share/man --infodir=\${prefix}/share/info checking build system type... x86_64-pc-linux-gnu checking host system type... x86_64-pc-linux-gnu checking whether make sets $(MAKE)... yes checking for x86_64-linux-gnu-gcc... gcc -m32 checking for C compiler default output file name... configure: error: C compiler cannot create executables See `config.log' for more details. make: *** [config.status] Error 77 Build command 'cd wine-0.9.21~winehq0~ubuntu~6.06 && dpkg-buildpackage -b -uc' failed. E: Child process failed I attached the config.log. What must I do to complete the compiles?

4 7

Re: wined3d/CopyRects bug fix
by Stefan Dösinger Sept. 22, 2006

Sept. 22, 2006

Am Montag 18 September 2006 10:12 schrieb Christopher GAUTIER: > Hello there, > > I've identified a bug in IWineD3DDeviceImpl_CopyRects. When CopyRects() > is called to copy the source entirely into the destination surface, and > that the sizes matches, a plain memcpy() is done. However, this assumes > that the surfaces have the same pitch, and this is not always the case. Welcome to Wine :-) IWineD3DDevice::CopyRects is a subset of IWineD3DSurface::BltFast(from DirectDraw), and I thought some time ago I sent a patch which makes CopyRects call BltFast. This patch apparently wasn't applied, and I did a google search for it and couldn't find it either. Looks like I moved it to /dev/null accidentally. We should call BltFast to avoid duplicating code. BltFast is also able to handle corner cases like bad rectangles. Can you test the attached patch? I have no game which uses CopyRects, so I can't test my patch. This is just a quick per-msdn implementation. Thanks, Stefan

2 1

Re: DDraw: Don't destroy complex attached surfaces even if explicitely requested as they will be liquidated anyway when the root is destroyed.
by Michael Stefaniuc Sept. 22, 2006

Sept. 22, 2006

Elie Morisse wrote: > A new version of my refcounting patch, lighter and honoring Stefan Dösinger proposal. You forgot to attach the patch. bye michael -- Michael Stefaniuc Tel.: +49-711-96437-199 Sr. Network Engineer Fax.: +49-711-96437-111 Red Hat GmbH Email: mstefani(a)redhat.com Hauptstaetterstr. 58 http://www.redhat.de/ D-70178 Stuttgart

1 0

Re: msi: update target file paths on MsiSetTargetPath (for TARGETDIR)
by James Hawkins Sept. 22, 2006

Sept. 22, 2006

On 9/21/06, Andrey Turkin <pancha(a)mail.nnov.ru> wrote: > MsiSetTargetPath can be used to change target path for directory after > CostFinalize, and all affected file paths must be recalculated. James > Hawkins sent a patch to make Wine's MSI engine do it for most cases; > this patch fixes one remaining "corner" case with TARGETDIR, and adds > test for it. > > ChangeLog: > > msi: Don't treat TargetDir specifically in resolve_folder > > > diff -aur wine-0.9.21/dlls/msi/helpers.c wine-0.9.21-my/dlls/msi/helpers.c > --- wine-0.9.21/dlls/msi/helpers.c 2006-09-13 23:10:25.000000000 +0400 > +++ wine-0.9.21-my/dlls/msi/helpers.c 2006-09-22 04:26:42.000000000 +0400 > @@ -216,29 +216,13 @@ > if (!name) > return NULL; > > - /* special resolving for Target and Source root dir */ > - if (strcmpW(name,cszTargetDir)==0 || strcmpW(name,cszSourceDir)==0) > + /* special resolving for Source root dir */ > + if ( strcmpW(name,cszSourceDir)==0) > { > - if (!source) > - { > - LPWSTR check_path; > - check_path = msi_dup_property( package, cszTargetDir ); > - if (!check_path) > - { > - check_path = msi_dup_property( package, cszRootDrive ); > - if (set_prop) > - MSI_SetPropertyW(package,cszTargetDir,check_path); > - } > - > - /* correct misbuilt target dir */ > - path = build_directory_name(2, check_path, NULL); > - clean_spaces_from_path( path ); > - if (strcmpiW(path,check_path)!=0) > - MSI_SetPropertyW(package,cszTargetDir,path); > - msi_free(check_path); > - } > - else > + if (source) > path = get_source_root( package ); > + else > + path = NULL; > if (folder) > *folder = get_loaded_folder( package, name ); > return path; > @@ -251,6 +235,11 @@ > if (folder) > *folder = f; > > + if (strcmpW(name, cszTargetDir)==0 && !f->ResolvedTarget && !f->Property) > + { > + f->Property = msi_dup_property( package, cszRootDrive ); > + } > + > if (!source && f->ResolvedTarget) > { > path = strdupW( f->ResolvedTarget ); > This fix is incorrect. What happens now when we call resolve_folder(package, "TARGETDIR", FALSE, TRUE, folder)? Your changelog says "Don't treat TargetDir specifically in resolve_folder" yet you're treating it specifically right here, but without the functionality that's still needed. Please resend the tests with todo_wine around failing tests. > > +#define ADD_SLASH(x) do { char *t; if (x[0]) {t=x+strlen(x)-1;if(t[0] != '\\' ) {t[1] ='\\';t[2] ='\0';}}} while (0) > + This is a really bad macro. Please use PathAddBackslash. -- James Hawkins

1 0

Re: WineD3D: glBlend* fixes
by Jesse Allen Sept. 21, 2006

Sept. 21, 2006

On 9/21/06, Roderick Colenbrander <thunderbird2k(a)gmx.net> wrote: > Hi, > > This patch changes the detection extension detection for glBlendColor/glBlendEquation. The function glBlendColor is part of OpenGL 1.1 and is supported on all OpenGL implementations. Officially it is part of the GL_ARB_imaging extension but not all vendors fully implement it and that's why only Nvidia and 3dlabs advertise it. This causes glBlendColor not to work on other implementations. So in case of GL 1.1, the detection is dropped. > > Further glBlendEquation is part of GL_ARB_imaging aswell. For the same reason GL_ARB_imaging can't be used to detect glBlendEquation. This call isn't supported on all OpenGL implementations. Luckily it is part of 'GL_EXT_blend_min_max' so that is used now. > Roderick, Mesa calls the extension "GL_EXT_blend_minmax", and so does the spec. I don't know what exactly uses the min_max form. Is this a typo? http://www.opengl.org/registry/specs/EXT/blend_minmax.txt http://www.google.com/search?hl=en&lr=&q=GL_EXT_blend_min_max&btnG=Search Also, the TRACE call in device.c:4513 needs a \n added too ;) Jesse

1 0

Re: Status of dlls that you maintain
by Jeff Latimer Sept. 21, 2006

Sept. 21, 2006

Tom Wickline wrote: > On 9/21/06, Jeff L <lats(a)yless4u.com.au> wrote: > >> >> BTW, the WWN #292 looks to be out of date as those functions are now >> implemented. > > > Hmm, Brian Vincent does the WWN and I just link the WWN articles to > the status > pages.. > Tom > Brian, I have successfully run Notes 7 on wine and the problems mentioned in WWN #292 have been fixed. Jeff Latimer

1 0

Re: [c2man] remove hardcoded extension (try#2)
by Alexandre Julliard Sept. 21, 2006

Sept. 21, 2006

"Vijay Kiran Kamuju" <infyquest(a)gmail.com> writes: > my $uc_dll_name = uc $dll_name; > + > + if (($dll_ext ne 'ds') && ($dll_ext ne 'acm') && ($dll_ext ne 'drv') > + ($dll_ext ne 'exe') && ($dll_ext ne 'ocx') && ($dll_ext ne 'vxd')) > + { > + $dll_ext='dll'; > + } You are still hardcoding things, there's really no reason to do that. -- Alexandre Julliard julliard(a)winehq.org

3 4

ATI and GL_ARB_Imaging support
by Kapila De Silva Sept. 21, 2006

Sept. 21, 2006

Hi all, I noticed, that wine checks for GL_ARB_imaging, before performing operations like glBlendColor, glBlendEquation etc. However it appears that the ATI drivers do not support this extension, though they do support the blending operations. I posted about this on the Rage3d linux driver boards, and the response seems to suggest that it would be better to check for the individual operations rather than the GL_ARB_imaging extension. See bottom of http://rage3d.com/board/showthread.php?t=33814685 What do you guys think? Kap.

2 1

Re: msi [6/7]: Translate the INSTALLSTATE_UNKNOWN index into the INSTALLSTATE_ABSENT image index
by Juan Lang Sept. 21, 2006

Sept. 21, 2006

Hiya James, - TRACE("Feature %s -> %d %d %d\n", debugstr_w(feature->Title), + printf("Feature %s -> %d %d %d\n", debugstr_w(feature->Title), I don't think you meant that to slip in. --Juan __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

Text Layout Summit
by Daniel Glassey Sept. 21, 2006

Sept. 21, 2006

Hi folks, There is going to be a Text Layout Summit during the Boston Gnome Summit http://live.gnome.org/Boston2006/TextLayout We are going to discuss things related to text layout. Hopefully we will discuss a uniscribe compatible API that could be used by wine and mono SWF since afaik neither has uniscribe support. It would be great if someone from wine and mono could be there. If that wouldn't be possible we would appreciate your input on what you would need. Regards, Daniel P.S. OK to CC me on replies as I'm not subscribed to either list.

2 1

Re: gdi32: GetTextExtentPoint and GetTextExtentExPoint behavedifferently if there's a \n in the string.
by Dmitry Timoshkov Sept. 21, 2006

Sept. 21, 2006

"Huw Davies" <huw(a)codeweavers.com> wrote: > - ok(sz1.cx == sz2.cx && sz1.cy == sz2.cy, > - "results from GetTextExtentExPointW and GetTextExtentPointW differ\n"); > + ok(sz1.cy == sz2.cy, > + "cy from GetTextExtentExPointW (%ld) and GetTextExtentPointW (%ld) differ\n", sz1.cy, sz2.cy); > +todo_wine { > + /* The \n in the string is apparently handled differently in GetTextExtentPoint and GetTextExtentExPoint */ > + ok(sz1.cx != sz2.cx, > + "cx from GetTextExtentExPointW (%ld) and GetTextExtentPointW (%ld) are the same\n", sz1.cx, sz2.cx); > + } This test fails for me (i.e. sz1.cx == sz2.cx) under XP (SP2+all the latest fixes). -- Dmitry.

2 3

Re: X11DRV: wgl move font code [resubmit]
by Roderick Colenbrander Sept. 20, 2006

Sept. 20, 2006

This time with a patch. > Hi, > > This is a resubmit of the previous font moving patch against the latest > cvs/git version. The old version didn't apply anymore to winex11.drv. For the > rest the patch is the same. > > Regards, > Roderick Colenbrander > -- > Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! > Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer > -- NEU: GMX DSL Sofort-Start-Set - blitzschnell ins Internet! Echte DSL-Flatrate ab 0,- Euro* http://www.gmx.net/de/go/dsl

1 0

Re: X11DRV: wgl move font code [resubmit]
by James Hawkins Sept. 20, 2006

Sept. 20, 2006

On 9/20/06, Roderick Colenbrander <thunderbird2k(a)gmx.net> wrote: > Hi, > > This is a resubmit of the previous font moving patch against the latest cvs/git version. The old version didn't apply anymore to winex11.drv. For the rest the patch is the same. > You forgot to attach the patch. -- James Hawkins

1 0

EnumServicesStatusA - Typical return structure contents with a working internet LAN connection
by Nick Law Sept. 20, 2006

Sept. 20, 2006

I'm trying to get an application running that accesses the Internet, but fails with the message, no Internet connection present. The internet connection is fine, as proved by the fact I can run firefox & wow under wine. I've disassembled the applications exe and examined wine's fixme errors when the application runs and it seems to do the following.... It calls InternetGetConnectedState to determine whether there is a modem or LAN connection. Wine returns the fact there is a LAN connection. It then calls EnumServicesStatusA with a type=30 & state=1, in order to get a list of running services, now what I think it's doing is expecting to see a service that provides the internet connection, because wine doesn't actually return a list of services the application then decides there's no internet connection. What I want to do is provide a return structure that does at least contain the service name & state (internet related) that the application is expecting to see. Does anybody know what service that could be, (specifically it's actual name) or even better provide a typical return structure that is returned by windows when a application calls EnumServicesStatusA so I can construct the structure this app is expecting to see. I've never written anything under MS windows otherwise I would write a small application myself that calls EnumServicesStatus & figure it out myself. In fact maybe I will have to get myself a copy of visual C++ so I can do these tests. Regards Nick Law PS I'm new to this debugging without access to the applications source code, so any hints & tricks would be appreciated. It looks like I may have to learn Intels assembler instruction set as well!, which I don't mind. The last time I wrote an assembly language program was 25 years ago on a PerkinElmer 3220 mini computer! So this should be fun.

2 1

msi/iTunes and questions
by Paul Vriens Sept. 20, 2006

Sept. 20, 2006

Hi, just for fun I've tried to install iTunes 7 on a fresh .wine. According to the AppDB this fails (one should first install QuickTime). While going through some traces I saw something the apparently needs this fix, but I'm no MSI-guy: diff --git a/dlls/msi/custom.c b/dlls/msi/custom.c index dad67a2..7b862fe 100644 --- a/dlls/msi/custom.c +++ b/dlls/msi/custom.c @@ -703,8 +703,8 @@ static UINT HANDLE_CustomType19(MSIPACKA static const WCHAR query[] = { 'S','E','L','E','C','T',' ','`','M','e','s','s','a','g','e','`',' ', 'F','R','O','M',' ','`','E','r','r','o','r','`',' ', - 'W','H','E','R','E',' ','`','E','r','r','o','r','`',' ','=',' ', - '%','s',0 + 'W','H','E','R','E',' ','`','E','r','r','o','r','`',' ','=',' ','\'', + '%','s','\'',0 }; MSIRECORD *row = 0; LPWSTR deformated = NULL; The single-quotes should be needed as the SQL statement in the trace showed: trace:msi:MSI_DatabaseOpenViewW L"SELECT `Message` FROM `Error` WHERE `Error` = The installlation of QuickTime did not complete successfully. iTunes requires QuickTime." 0x6669a498 and a bit further: err:msidb:WHERE_VerifyCondition Couldn't find column L"The" This is just one of many issues with this installation. Does the fix look OK? Should this be sent? If it has to be accompanied with tests then please use the above fix at will, as I don't have the time (nor the msi-skills to look into that). Cheers, Paul.

3 2

Re: oleaut32 1/2: Implements function variant:VarImp
by Alexandre Julliard Sept. 20, 2006

Sept. 20, 2006

"Benjamin Arai" <me(a)benjaminarai.com> writes: > + case VT_I1: if (V_I1(left) == -1) resvt = VT_NULL; break; > + case VT_UI1: if (V_UI1(left) == 0xff) resvt = VT_NULL; break; > + case VT_I2: if (V_I2(left) == -1) resvt = VT_NULL; break; > + case VT_UI2: if (V_UI2(left) == 0xffff) resvt = VT_NULL; break; > + case VT_INT: if (V_INT(left) == -1) resvt = VT_NULL; break; > + case VT_UINT: if (V_UINT(left) == 0xffffffff) resvt = VT_NULL; break; > + case VT_I4: if (V_I4(left) == -1) resvt = VT_NULL; break; > + case VT_UI4: if (V_UI4(left) == 0xffffffff) resvt = VT_NULL; break; > + case VT_I8: if (V_I8(left) == -1) resvt = VT_NULL; break; > + case VT_UI8: if (V_UI8(left) == 0xffffffff) resvt = VT_NULL; break; Shouldn't the last one be something like ~(ULONGLONG)0 instead of 0xffffffff? -- Alexandre Julliard julliard(a)winehq.org

2 2

EnumServicesStatusA - Typical return structure contents with a working internet LAN connection
by Nick Law Sept. 20, 2006

Sept. 20, 2006

I'm trying to get an application running that accesses the Internet, but fails with the message, no Internet connection present. The internet connection is fine, as proved by the fact I can run firefox & wow under wine. I've disassembled the applications exe and examined wine's fixme errors when the application runs and it seems to do the following.... It calls InternetGetConnectedState to determine whether there is a modem or LAN connection. Wine returns the fact there is a LAN connection. It then calls EnumServicesStatusA with a type=30 & state=1, in order to get a list of running services, now what I think it's doing is expecting to see a service that provides the internet connection, because wine doesn't actually return a list of services the application then decides there's no internet connection. What I want to do is provide a return structure that does at least contain the service name & state (internet related) that the application is expecting to see. Does anybody know what service that could be, (specifically it's actual name) or even better provide a typical return structure that is returned by windows when a application calls EnumServicesStatusA so I can construct the structure this app is expecting to see. I've never written anything under MS windows otherwise I would write a small application myself that calls EnumServicesStatus & figure it out myself. In fact maybe I will have to get myself a copy of visual C++ so I can do these tests. Regards Nick Law PS I'm new to this debugging without access to the applications source code, so any hints & tricks would be appreciated. It looks like I may have to learn Intels assembler instruction set as well!, which I don't mind. The last time I wrote an assembly language program was 25 years ago on a PerkinElmer 3220 mini computer! So this should be fun. -- Regards Nick Law Mobile: 07736 022165 Email: nick.law(a)mic-nucmed.co.uk Medical Imaging Consultancy Ltd (M.I.C) Sales and Service of ADAC and IS2 Products http://www.mic-nucmed.co.uk

2 2

Re: Implement GetKerningPairs for TrueType fonts, take 4
by Huw Davies Sept. 20, 2006

Sept. 20, 2006

On Thu, Sep 21, 2006 at 12:18:51AM +0900, Dmitry Timoshkov wrote: > this version of the patch incorporates another Huw's suggestion > in eliminating redundant local variable and simplifying the code. Thanks Dmitry, Looks good to me. -- Huw Davies huw(a)codeweavers.com

1 0

Recovering a patch accidentally removed by git reset
by Jeff L Sept. 20, 2006

Sept. 20, 2006

How do you get a patch back that you have managed to remove completely from your system with a git reset HEAD^ git reset git checkout patchname ? git fetch does not seem to refresh the patch from the repository. I was going fairly well with git then stuffed this up. Jeff

4 4

Re: Implement GetKerningPairs for TrueType fonts, take 3
by Huw Davies Sept. 20, 2006

Sept. 20, 2006

On Wed, Sep 20, 2006 at 03:22:02PM +0900, Dmitry Timoshkov wrote: > Changelog: > Implement GetKerningPairs for TrueType fonts. Hi Dmitry, Looks good, apart from one small nitpick below. I still think it would be nice to add the tests that you have to the test suite, even if it's just something that checks the first few pairs of Arial at some fixed font height. .... > +static DWORD parse_format0_kern_subtable(GdiFont font, > + const struct TT_format0_kern_subtable *tt_f0_ks, > + const USHORT *glyph_to_char, > + KERNINGPAIR *kern_pair, DWORD cPairs) > +{ .... > + kern_pairs_copied = 0; > + > + for (i = 0; i < nPairs; i++) > + { > + kern_pair->wFirst = glyph_to_char[GET_BE_WORD(tt_kern_pair[i].left)]; > + kern_pair->wSecond = glyph_to_char[GET_BE_WORD(tt_kern_pair[i].right)]; > + kern_pair->iKernAmount = MulDiv((short)GET_BE_WORD(tt_kern_pair[i].value), font->ppem, font->ft_face->units_per_EM); > + > + TRACE("left %u right %u value %d\n", > + kern_pair->wFirst, kern_pair->wSecond, kern_pair->iKernAmount); > + > + kern_pair++; > + kern_pairs_copied++; > + if (!--cPairs) break; > + } Seems to me that if you set nPairs = min(cPairs, nPairs) at the start of the loop, then you can eliminate the break and get rid of the kern_pairs_copied variable that basically mirrors i. Thanks, Huw. -- Huw Davies huw(a)codeweavers.com

1 0

make test failure
by Ivan Leo Puoti Sept. 20, 2006

Sept. 20, 2006

Here is a make test failure I'm getting on my laptop (for those of you who were not at wineconf, make test should, in theory, work everywhere) ../../../../wine/tools/runtest -q -P wine -M gdi32.dll -T ../../.. -p gdi32_test.exe.so ../../../../wine/dlls/gdi/tests/font.c && touch font.ok font.c:276: Test failed: Courier(13): tm.tmAscent 10 != 11 font.c:277: Test failed: Courier(13): tm.tmDescent 3 != 2 make[2]: *** [font.ok] Error 2 make[2]: Leaving directory `/home/ivan/build/dlls/gdi/tests' make[1]: *** [gdi/tests/__test__] Error 2 make[1]: Leaving directory `/home/ivan/build/dlls' make: *** [dlls/__test__] Error 2

8 7

RegisterHotKey/UnregisterHotKey implementation...success?
by Aaron Slunt Sept. 20, 2006

Sept. 20, 2006

http://www.winehq.com/pipermail/wine-devel/2006-May/047460.html The above link I looked at, and I saw this person was apparently crashing with the patch when trying to register a hotkey twice (that's how I interpreted it). Anyways, I looked at the patch, and cleaned it up so it applies to the current git tree. Anyways, it seems that I simply cannot reproduce the crash he was talking about. I binded F1 in HLSS (Half-Life Sound Selector), and opened another copy of HLSS. It simply would not allow me to register the key again. So I closed the second copy of HLSS and opened Ventrilo. I attempted to bind F1, and again it simply would not allow it. And let me make this clear, not once did I crash with this patch. HLSS works (fundamentally, not perfectly) when using it with Counter-strike (the hotkey works, there are window management problems though which have nothing to do with this) and I just never crashed with it. I'm attaching the patch for others to test if they could, and to see if a crash can be reproduced. But to me it looks like the work Robert Reif and Vincent Povirk is a huge success.* *

2 2

COM
by Joe McCay Sept. 20, 2006

Sept. 20, 2006

A recent article in a Linux magazine talking about winelib mentioned using __attribute__((com_interface)). I thought that attribute was no longer valid in gcc. The reason I am asking is because I am trying to do COM development using mingw, C++, & (hopefully) WIDL. I thought the 3.x and higher version of gcc supported -fvtable-thunks by default making it binary compatible with windows COM. Does anyone know if this is really true? jmccay

1 0

Problem with a Turbolog4.exe application under wine
by James Courtier-Dutton Sept. 19, 2006

Sept. 19, 2006

Hi, A demo version of Turbolog4 can be downloaded from: http://www.turbolog.de/ The program installs ok, but fails to run correctly. It asks for a License key, one should be able to press cancel and bypass it into demo mode where a window is displayed for entering log records. Can anyone help be diagnose why it fails. James

2 1

Something's wrong with my GIT tree
by Aaron Slunt Sept. 19, 2006

Sept. 19, 2006

http://pastebin.ca/176787 I've had this for several days, and git fetch tells me everything's up to date. Is this a known issue? How do I get rid of this? ./configure --x-libraries=/emul/linux/x86/usr/lib/ LDFLAGS="-L/emul/linux/x86/usr/lib -L/emul/linux/x86/lib" && make depend && make That configure has always worked for me, and it gets through make depend, it's just failing on make with that error message.

2 1

Error in GDI (StretchBlt) - proper fix?
by David Gümbel Sept. 19, 2006

Sept. 19, 2006

Hello everybody, I am debugging an application and got up to the point where I have reason to believe that the StretchBlt function in GDI is part of the problem. A backtrace of the exception I always run into looks like this: err:syslevel:_CheckNotSysLevel Holding lock 0x7ecd0820 level 3 wine: Unhandled exception 0x80000003 at address 0x0000:0x7ee96b34 (thread 0011), starting debugger... WineDbg starting on pid 0x10 err:syslevel:_CheckNotSysLevel Holding lock 0x7ecd0820 level 3 0x7ee96b35 _CheckNotSysLevel+0x3d in kernel32: addl $36,%esp Wine-dbg>bt Backtrace: =>1 0x7ee96b35 _CheckNotSysLevel+0x3d in kernel32 (0x7ee96b35) 2 0x7ec84c81 GDI_CheckNotLock+0x20 in gdi32 (0x7ec84c81) 3 0x7ec628e8 DeleteDC+0x26 in gdi32 (0x7ec628e8) 4 0x7ec6420b GetDIBits+0x1fd in gdi32 (0x7ec6420b) 5 0x7ec6c282 in gdi32 (+0x2c282) (0x7ec6c282) 6 0x7ec5bfee StretchBlt+0xca in gdi32 (0x7ec5bfee) 7 0x0042a3aa in bav (+0x2a3aa) (0x0042a3aa) 8 0x00425460 in bav (+0x25460) (0x00425460) 9 0x005d1304 in bav (+0x1d1304) (0x005d1304) 10 0x005e4f5d in bav (+0x1e4f5d) (0x005e4f5d) 11 0x0066f10b in bav (+0x26f10b) (0x0066f10b) 12 0x00681f8b in bav (+0x281f8b) (0x00681f8b) 13 0x7ee7ff5f in kernel32 (+0x4ff5f) (0x7ee7ff5f) 14 0xb7e9d387 wine_switch_to_stack+0x17 in libwine.so.1 (0xb7e9d387) When looking at StretchBlt's source code, I see a comment saying that there is a race condition in this function[1]. If I'm not mistaken, this problem has been reported in january as bug #4284, causing trouble (a deadlock) with MS PowerPoint[2]. Walt Ogburn suggested in january that commenting out #define STRETCH_VIA_DIB in dlls/gdi/mfdrv/bitblt.c could work around the problem[3]. Does anybody have any idea on how a proper fix for that function might approximately look like? Thanks, David [1] http://source.winehq.org/source/dlls/gdi/bitblt.c#L139 [2] http://bugs.winehq.com/show_bug.cgi?id=4284 [3] http://www.winehq.org/pipermail/wine-devel/2006-January/044040.html

1 0

Re: Fix a memory leak, fix a typo in the comment, add a trace
by Huw Davies Sept. 19, 2006

Sept. 19, 2006

On Tue, Sep 19, 2006 at 11:44:52AM +0900, Dmitry Timoshkov wrote: > please apply after "Implement GetKerningPairs for TrueType fonts". > > Changelog: > - fix a memory leak > - fix a typo in the comment > - add a trace about skipped a not supported kerning table format Hi Dmitry, Wouldn't it be nicer to simply resend the first patch again with these fixes applied? Thanks, Huw. -- Huw Davies huw(a)codeweavers.com

1 0

Re: [WINED3D] Pbuffer/ActiveRender cleanup #2
by H. Verbeet Sept. 19, 2006

Sept. 19, 2006

On 17/09/06, Ivan Gyurdiev <ivg231(a)gmail.com> wrote: > @@ -156,6 +160,7 @@ typedef struct wined3d_settings_s { > BOOL glslRequested; > int vs_selected_mode; > int ps_selected_mode; > + int offscreen_selected_mode; > /* nonpower 2 function */ > int nonpower2_mode; > int rendertargetlock_mode; When you change wined3d_settings, you should make sure that you update the defaults in wined3d_main.c as well. I know those are currently mismatched, but I've got a patch to clean it up. I'm currently waiting for Alexandre to apply/reject some of the wined3d patches sent over the past couple days before sending it (and some other patches). I'm not sure if using wined3d_settings is really the right place to store things that don't have a corresponding registry key, but it seems like the best place for now. To clarify what OFFSCREEN_NONE does, it renders to the backbuffer for the current context, rather than using a separate context, and copies it to a texture when done. That's also why, depending on the primitive drawing order, you sometimes see the offscreen surface in the lower left corner of the screen.

2 3

My Wineconf 2006 trip report
by Marcus Meissner Sept. 19, 2006

Sept. 19, 2006

Hi folks, I had to write it up for my dear colleagues anyway, so here my notes... Feel free to reuse as you like. Attendance was around 35 people, all developers. Location was the university campus of Reading, a lecture hall in the Agriculture Department. The conference was organized by CodeWeavers, but sponsored by mostly 2 donors of each 1000 (?) US$ to the Wine project, one private person and a Poker website (strange enough). Around 10 - 15 developers were CodeWeavers employees, the other were volunteer developers. 1. Keynote by Alexandre Julliard (CW) Alexandre Julliard, our resident dictator and lone committer held the keynote. - The beta (0.9.*) works out fine. The 2 month code freeze before Beta has helped, and bugs got nicely pruned. The bugzilla stat char showed a nice increase in reported and also in fixed bugs. - Wine 1.0 plans Code Freeze perhaps later this year, for several months, then 1.0 release early next year. - Post 1.0 Unclear. Likely will have a stable and a feature branch. - GIT GIT is working out nicely. Number of monthly (?) commits grew from 400 at the turnover point to around 700 now. Reasons: * More smaller and confined patches (also thanks to GIT allowing incremental local commits). * GIT patch submission is in one defined patch format. Earlier patches were occcasionaly generated from subdirectories, misformatted, hand edited or similar. GIT changes this to one defined format, making his work easier. * "Developers might not like it, but I do". - Goals for 1.0 * Better Game Support (DirectX). This is pretty much heavy work in progress currently. * Regressiontests, Documentation are fine as-is. * Copy Protection support for Games etc. Necessary to run lots of games. However this is not a release stopper. - LOC still increasing, around 1.800.000 now. - COM (Component Object Model) report by Rob Shearman (CW) Rob (and others) spent most of last year enhancing and fixing the COM, OLE and RPCRT parts of WINE and summarized it. The out of process COM used by InstallShield has seen lots of changes and is working much better now. Work on remote COM was necessary for Outlook which talks COM via RPC over TCP sockets with the Outlook server. - GIT Tutorial by Mike McCormack (CW) Not WINE specific, so no details. - cxtest by Martin Pilka (CW) cxtest is the ongoing developed GUI testing kit from Codeweavers. Basic idea is that it can generate keypresses, mouseclicks and similar things and then captures images at various intervals and compares them with previous runs. Needs review of problems, but has definitely spotted some already. - Google's use of WINE by Dan Kegel (Google) 2 major Windows applications that wanted a Linux port. - Picasa Requirement: _No_ Windows binary code change. Hired Codeweavers as company to fix issues: - lots of bugfixes - IWebBrowser implementation - File change notification (implemented with inotify) - Video Codec handling fixes (both in MSVIDEO and DSHOW) - However: Needed to disable CD Burning Module of Picasa. Works on almost all distros nicely. - Google Earth WINE has OpenGL-in-Window problems, otherwise this might have been a WINE job too likely. The Loki Installer guy (Brad something?) did the port on-site (this 1 person did the port). Exposes lots of OpenGL / GFX card related related problems. -> But received more press and recognition. No other Windows applications of Google need to be ported currently. (Google Desktop would be one, but it requires integration into native browsers, so it is too Windows specific.) Dan also oversees 4 Google interns working on Wine. - CrossOver on Mac by Andrew Bogott (CW) Beta on MacOS/X (Intel) of CrossOver Office available. Andrew demoed MS Excel, including installation of it. GUI toolkit is in TCL + Coco (? graphical Mac scripting language) Use X11 graphics driver, Quartz driver is currently being considered. Caused stability problems with Mac Kernel. Caused stability problems with Mac X Server. Is slower than on Linux due to lack of some fast methods we use on Linux (I guess something like threading, futexes or so. -Marcus) - Governance breakout session The point sometimes comes up whether to replace Alexandre. The present developers did not agree, so there was not much discussion. We evaluated _why_ they want it: - They want low quality hacks into the tree to make apps work. This is a no go, you need to fix it right. - They get no reply on their submission _why_ it was rejected. True reason, we discussed a bit how to improve. - A bot welcoming new patch posters? - checks for most common problem, // comments - people actively taking care of areas and responding to people - Alexandre waiting for comments from those responsibles on difficult patches - Branch Maintainership post 1.0 ... Briefly touched, but we will see when we get there. - Packaging breakout session Its unclear (to me at least) what is wanted ;) Paul Millar wants nightly regression tests (cxtest runs with downloaded binaries?) Some want nightly builds. Some want .spec files in a central repository. Marcus is trying to bring up RPM builds for SUSE, Mandriva 2006.0, Fedora Core 4 and 5 on the openSUSE buildservice. Debian to come. Ciao, Marcus

2 2

Wine 0.9.21 fails to run
by Ken Zalewski Sept. 19, 2006

Sept. 19, 2006

I just upgraded from 0.9.20 to 0.9.21, and now nothing will run, including simple programs such as sol.exe and notepad.exe. The error message I'm receiving is: X Error of failed request: BadWindow (invalid Window parameter) Major opcode of failed request: 145 (NV-GLX) Minor opcode of failed request: 4 () Resource id in failed request: 0x242 Serial number of failed request: 19 Current serial number in output stream: 19 My system is a Slackware 10.2 system, running kernel 2.6.17.13, x11 6.8.2, and NVidia driver 8774. I have seen some other postings on the Net about this, including postings in the Ubuntu forums (which means that this is not limited to Slackware). Downgrading to 0.9.20 gets everything working again. My guess, based on the "NV-GLX" message, is that something was changed in the OpenGL support between 0.9.20 and 0.9.21, and it has broken Wine. I'll be happy to help debug this. Sincerely, Ken Zalewski Slackware package maintainer Linuxpackages.net

2 1

wine conf group photo
by EA Durbin Sept. 18, 2006

Sept. 18, 2006

The link on the winehq.org page for the photo is returning 404 not found.

2 1

Re: winex11.drv: cleanup better in X11DRV_WineGL_InitOpenglInfo
by Jesse Allen Sept. 18, 2006

Sept. 18, 2006

On 9/17/06, Phil Costin <philcostin(a)hotmail.com> wrote: > Roderick Colenbrander wrote: > > > The problem is that I'm not sure (and I think it is Alexandre his issue > > aswell) whether this patch is a workaround or a real fix. I think that > > wined3d created a GLX context in lets say thread '1' and that it wants to > > create lets say another one in a different thread. If this is the case it > > could happen that for the new thread the x11drv visual needs to be > > 'rechosen'. This causes the 'info' code to be called again. Because this > > code creates a context we mess up the already active context. > > > > Your patch indeed fixes that by storing the current GLX context and > > restoring it later on. Perhaps there's a nicer way to prevent this issue > > to happen at all. > > > > In any case I think it might be nicer to use glXGetCurrentContext in case > > there's already some GLX context around. So in that case don't create a > > new context and don't kill it at the end. This would already be cleaner. > > > > Roderick > > > > > > > >> I don't know myself either. I was just going to wait. I want to add a > >> little more to say about it. > >> > >> It's funny that the patch was only meant to affect the opengl/wgl code > >> and the crashing apps are all d3d ones. However winex11 will now > >> change the context on it's own when doing the opengl query and not set > >> it back. What I noticed a couple weeks ago and now is that winex11 > >> does usually gets inited twice, therefore it will change the current > >> context twice. One at the start of the program, and another later on > >> -- maybe on creation of a window or something. By that time wined3d is > >> probably inited too, and making glXMakeCurrent calls too since it's > >> based on opengl. So perhaps winex11 is causing wined3d to trip up at > >> some point by changing the context. So if InitOpenglInfo just sets > >> things back, then wined3d won't go boom. > >> > >> >>From comments in wined3d, I can tell there isn't any "context > >> manager". So what wined3d does at most is very simple management of > >> contextes. We probably need a manager to coordinate things like this > >> better between the two libraries. But that will likely be a bigger > >> patch and to do so... I don't know, we need a fix for all the broken > >> d3d apps now. > >> > >> Jesse > >> > > > > > At the moment, this patch fixes regressions so if we don't have the code in > there at all, many games do not work. > > It may not be the ideal solution but is it not better for now to fix the > regression and think about a proper solution separately without breaking > applications by not having the patch applied? > > Regards, > Phil > > > > Phil, Roderick, I'm going to look at what I can do to improve the patch and gather more information on the handling of contextes. I just haven't had time yet. And Roderick, I'll discuss with you the newer patch when I'm ready so to see if you agree with it. I do lean towards agreeing with Phil on that this is simply needed and more than a workaround. I've already been asked many times if the patch has gotten in and that is why I posted to wine-patches so quickly; your patch broke alot of D3D apps for alot of people... The d3d9 stateblock test is the best clue. Jesse

2 2

Visual C++ does not seem to have snprinft
by Jeff L Sept. 18, 2006

Sept. 18, 2006

When compiling dlls/gdi/tests/metafile.c on line 1357 I get a warning C4013: "snprintf" is undefined. Looking around, it seems that snprinft is in fact _snprintf in Visual C++. I have found a define #if !defined(HAVE_SNPRINTF) && defined(HAVE__SNPRINTF) #define snprintf _snprintf #endif in wine/port.h but it requires config.h and they don't appear in my Visual C++ system. What is the best way to address this problem? It seems to be fairly common but I have not found an example in a test. Jeff Latimer

6 5

Re: If there is no space for strings GetOutlineTextMetricsA should not indicate that they present
by Paul Vriens Sept. 18, 2006

Sept. 18, 2006

On Mon, 2006-09-18 at 19:00 +0900, Dmitry Timoshkov wrote: > Hello, > > this patch prevents my test application from crashing by dereferencing > memory beyond allocated memory block. > > Changelog: > If there is no space for strings GetOutlineTextMetricsA should not > indicate that they present. > > --- cvs/hq/wine/dlls/gdi/font.c 2006-08-15 14:51:46.000000000 +0900 > +++ wine/dlls/gdi/font.c 2006-09-18 18:38:57.000000000 +0900 > @@ -1483,6 +1483,20 @@ UINT WINAPI GetOutlineTextMetricsA( > if(output != lpOTM) { > memcpy(lpOTM, output, cbData); > HeapFree(GetProcessHeap(), 0, output); > + > + /* check if the string offsets really fit into the provided size */ > + /* FIXME: should we check string length as well? */ > + if ((UINT_PTR)lpOTM->otmpFamilyName >= lpOTM->otmSize) > + lpOTM->otmpFamilyName = 0; /* doesn't fit */ > + > + if ((UINT_PTR)lpOTM->otmpFaceName >= lpOTM->otmSize) > + lpOTM->otmpFaceName = 0; /* doesn't fit */ > + > + if ((UINT_PTR)lpOTM->otmpStyleName >= lpOTM->otmSize) > + lpOTM->otmpStyleName = 0; /* doesn't fit */ > + > + if ((UINT_PTR)lpOTM->otmpFullName >= lpOTM->otmSize) > + lpOTM->otmpFullName = 0; /* doesn't fit */ > } > > end: this one also fixes an issue I was looking into this morning: Try to create a key via regedit with the name {44444444-4444-4444-44444444} and press enter. Cheers and thanks, Paul.

1 0

OpenGL multiplexing and Wine
by Molle Bestefich Sept. 18, 2006

Sept. 18, 2006

A couple of OpenGL multiplexers for Xen (and VMWare, theoretically) exists: http://www.cs.toronto.edu/~andreslc/xen-gl/ http://www.diku.dk/hjemmesider/ansatte/jacobg/gfx/ They allow guest virtual machines to render 3D graphics using the host machine's graphics accelerator. I think they are very interesting in combination with Wine, because Wine can let Direct3D games run on top of OpenGL, making the multiplexers work for all modern games, whereas running a real Microsoft Windows guest would only allow acceleration for some games. There's a list of supported XGL versions and GL calls on this page under "Capabilities": http://www.cs.toronto.edu/~andreslc/xen-gl/ And a couple of whitepapers here: http://www.diku.dk/~jacobg/blink-techreport.pdf http://www.diku.dk/~jacobg/blink2.pdf Will WineGL run on top of a multiplexer, or does it lack support of some calls?

3 2

Wine directx and xinerama
by Roland Kaeser Sept. 17, 2006

Sept. 17, 2006

Hello all I just regognized that I get serious trouble trying to run apps on a xinerama screen (Matrox Parhelia) with Triple display configuration (http://www.matrox.com/graphics/workstation/cre_pro/products/pseries/p750.cfm / 01:00.0 VGA compatible controller: Matrox Graphics, Inc. Millenium P650/P750 (rev 02)). Running the same apps (such as Railroad Tycoon) on a "normal" machine with intel 945G (VGA compatible controller: Intel Corporation 945G/GZ Express Integrated Graphics Controller (rev 02)) its works fine. But running RT3 on my Matrox card makes it crashing or just bringing a error message, even when I configured a virtual desktop (1024x768). Does somebody has a idea about this behaviour? Thanks Roland

3 2

Will Wine run with Linspire 5?
by Ctm007 Sept. 17, 2006

Sept. 17, 2006

Will Wine run with Linspire 5? Carl Martin ctm007(a)houston.rr.com

1 0