March 2014 - Wine-Devel - list.winehq.org

GSoC 2014

by Naveen Sahu

Hello, I'm Naveen Sahu, a 3rd yr undergraduate in Computer Science and Engineering at Indian Institute of Technology Guwahati (IIT G), India. I have a good knowledge of C, C++ and python. I would like to contribute to Wine through GSoC'14. I'm interested in '*Richedit - implement Windowless mode*' project. Can you please give me some advice which bugs to fix that can help me in this project too? I'll try my best to submit patches as soon as possible. Regards, Naveen

March 11, 2014

Fwd: Build a World - a sophisticated OpenGL Windows and Linux native game - can it be any help with WINE?

by Mateusz 'DevSH' Kielan

-------- Original Message -------- Subject: Build a World - a sophisticated OpenGL Windows and Linux native game - can it be any help with WINE? Date: Sun, 24 Mar 2013 14:36:31 +0000 From: Mateusz Kielan <devsh.graphicsprogramming(a)gmail.com> To: ulrich.czekalla(a)utoronto.ca I'm sorry if this email seems kind of rushed (originally planned to send to wine-devel but didnt want this email to go public). This is more of a loose proposition towards CodeWeavers rather than the wine developer community We at "Sodan Design" are developing relatively complicated video game called Build a World www.buildaworld.net <http://www.buildaworld.net> http://www.youtube.com/user/wwwbuildaworldnet/videos?view=0&flow=grid https://www.facebook.com/pages/Build-A-World/284152794990875?fref=ts http://www.kickstarter.com/projects/1895183955/build-a-world We are soon going to release our linux build to our alpha testers, I'm the graphics programming and we had a native 32bit linux build since I started working on the game. Now, I would imagine that you would very much benefit from analyzing a program which has *ALMOST* exactly the same codepath on linux as it does on linux (with the exception of a few things like threading). Right now we use the following technologies (listing more unusual stuff): --OpenGL compatibility --OpenGL 3.0 extensions --OpenGL 4.2 tessellation shaders through extensions --OpenGL Geometry shaders --OpenGL 3D textures --Boost::Libraries --Pthreads/Windows Threads with mutices --google protocol buffers --libtcmalloc (Linux) --FMOD And in the future: --libvpx --possibly ffmpeg --OpenCL (definitely) - I gather this is something wine has hickups with (at least CUDA was difficult) --OpenGL compute shaders I'm sure you'd appreciate a good piece of crossplatform software to work with, and we'd appreciate the extra linux/windows testing. Some forms of reverse engineering will be allowed in our EULA (for purpose of licence compliance with FFMPEG) I'm sure if we CodeWeavers signed an NDA then we could hand over a few nice things like executables with some debug symbols. Even without an NDA we'd be able to give a free game account to every CodeWeavers employee. P.S. It would be great if we could get some publicity or at least some mention if our 32bit windows exe gets a platinum rating or if tinkering around with Build-a-World helps with advancing wine.

March 11, 2014

Re: ntdll: Use condition variable for RtlQueueWorkItem implementation

by Sebastian Lackner

[Sorry for the duplicate Alexandre, clicked the wrong button and didn't answer to the mailing list :( ^^] No, the worker cannot exit at the time when the variable is checked. A worker only can break out of the processing loop, when there is no new item and the condition variable timeout expires. Since the new item was added before the check (while holding the CS), the last worker cannot terminate before this element has been processed and the additional delay of 30sec to wait for new jobs. Nevertheless I understand that its probably a bit too much "optimized", so I'll submit a new patch later which simplifies this part a bit. Regards, Sebastian 2014-03-11 15:09 GMT+01:00 Alexandre Julliard <julliard(a)winehq.org>: > Sebastian Lackner <sebastian(a)fds-team.de> writes: > > > * Its also on purpose, that num_workers is (read-) accessed at one point > > without locking the critical section. If num_workers > 0 we can be sure, > that > > a worker thread will process the work item (sooner or later). If > > num_workers == 0 we enter the critical section, and check once more to > > prevent race condition 2. > > The worker may be about to exit at the time you check. I don't think it > makes any sense to add races to "optimize" an error case that's not > going to happen anyway. > > -- > Alexandre Julliard > julliard(a)winehq.org >

March 11, 2014

Re: Ideas about mshtml for GSoC 2014

by Zhenbo Li

Thank you very much! 2014-03-11 17:35 GMT+08:00 Jacek Caban <jacek(a)codeweavers.com>: > I think that MSHTML may make a good GSoC project. The application will > probably need some more concrete task. We will also need to see that > you're prepared to the task and sending those patches is a good start. If I concentrate on implementing most IHTMLTable* functions, would it be a good idea? > Here are some comments to your patches: > > + > + ret = nsAString_Init(&val, v); > + if (!ret){ > + ERR("nsAString_Init(%s) failed!\n", debugstr_w(v)); > + return E_FAIL; > + } > > In case where you control the value assigned to nsAString, like in this > case, use nsAString_InitDepend to avoid unneeded alloc and copy. This is > also infailable, so no error checks are needed. > > + nsres = nsIDOMHTMLTableRowElement_SetAlign(This->nsrow, &val); > + if (NS_FAILED(nsres)){ > + ERR("Set Align(%s) failed!\n", debugstr_w(v)); > + nsAString_Finish(&val); > + return E_FAIL; > + } > + > + nsAString_Finish(&val); > > > Release the string just after SetAlign call so that you don't need to do that on each branch. > > > Same for other patches and they are ready to go to wine-patches. > Thank you for your advice. I'll send my fixed patches soon. > Cheers, > Jacek -- Have a nice day! Zhenbo Li

March 11, 2014

Re: ntdll: Use condition variable for RtlQueueWorkItem implementation

by Alexandre Julliard

Sebastian Lackner <sebastian(a)fds-team.de> writes: > * Its also on purpose, that num_workers is (read-) accessed at one point > without locking the critical section. If num_workers > 0 we can be sure, that > a worker thread will process the work item (sooner or later). If > num_workers == 0 we enter the critical section, and check once more to > prevent race condition 2. The worker may be about to exit at the time you check. I don't think it makes any sense to add races to "optimize" an error case that's not going to happen anyway. -- Alexandre Julliard julliard(a)winehq.org

March 11, 2014

Re: [PATCH 5/5] winex11: Add support for WGL_WINE_make_current extension. (try 2)

by Henri Verbeet

On 11 March 2014 13:08, Alexandre Julliard <julliard(a)winehq.org> wrote: > Ken Thomases <ken(a)codeweavers.com> writes: > >> try 2: fail gracefully if the window was destroyed > > Still no good I'm afraid: > > ../../../tools/runtest -q -P wine -T ../../.. -M opengl32.dll -p opengl32_test.exe.so opengl && touch opengl.ok > X Error of failed request: BadMatch (invalid parameter attributes) > Major opcode of failed request: 153 (GLX) > Minor opcode of failed request: 26 (X_GLXMakeContextCurrent) > Serial number of failed request: 524 > Current serial number in output stream: 524 > make[1]: *** [opengl.ok] Error 1 > Just for what it's worth, it does seem to consistently pass the tests here. (But so did the previous version.)

March 11, 2014

Valve releases their "ToGL" code to convert D3D to OpenGL

by Scott Ritchie

https://github.com/ValveSoftware/ToGL There might be something useful in there.

March 11, 2014

Re: [PATCH 5/5] winex11: Add support for WGL_WINE_make_current extension. (try 2)

by Alexandre Julliard

Ken Thomases <ken(a)codeweavers.com> writes: > try 2: fail gracefully if the window was destroyed Still no good I'm afraid: ../../../tools/runtest -q -P wine -T ../../.. -M opengl32.dll -p opengl32_test.exe.so opengl && touch opengl.ok X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 153 (GLX) Minor opcode of failed request: 26 (X_GLXMakeContextCurrent) Serial number of failed request: 524 Current serial number in output stream: 524 make[1]: *** [opengl.ok] Error 1 -- Alexandre Julliard julliard(a)winehq.org

March 11, 2014

Re: Ideas about mshtml for GSoC 2014

by Jacek Caban

Hi Zhenbo, On 03/10/14 05:21, Zhenbo Li wrote: > Hello, > > I'm Zhenbo Li, a student in Shanghai JiaoTong University, China. I > have been a Wine user for almost 4 years, and I've been preparing for > GSoC for several month. > > Since 2013, I've translated some wiki pages into Chinese[1], reported > nearly 40 bugs[2], involved in Appdb[3], and sent few patches[4]. I'm > glad to do more contribution to Wine in 2014. > > While using Wine, it is common to face a bug in mshtml[5]. Native > iexplore may be a workaround, but it can cause other problems, too. So > my idea is to implement some functions in mshtml.dll. > > I have a rough plan now, and I wish to get your advice. As I have > about 3 months, I want to implement some functions in IHTMLTable, > IHTMLTableCol, etc., and write testcaes for them. I've written some > draft patches for IHTMLTableRow, and I'll send them to wine-patches > after improving them with your advice. I hope I can establish a proper > schedule after communicating with my mentor. > > I'm also open to any suggestion, and any comment or advice is great appreciated. I think that MSHTML may make a good GSoC project. The application will probably need some more concrete task. We will also need to see that you're prepared to the task and sending those patches is a good start. Here are some comments to your patches: + + ret = nsAString_Init(&val, v); + if (!ret){ + ERR("nsAString_Init(%s) failed!\n", debugstr_w(v)); + return E_FAIL; + } In case where you control the value assigned to nsAString, like in this case, use nsAString_InitDepend to avoid unneeded alloc and copy. This is also infailable, so no error checks are needed. + nsres = nsIDOMHTMLTableRowElement_SetAlign(This->nsrow, &val); + if (NS_FAILED(nsres)){ + ERR("Set Align(%s) failed!\n", debugstr_w(v)); + nsAString_Finish(&val); + return E_FAIL; + } + + nsAString_Finish(&val); Release the string just after SetAlign call so that you don't need to do that on each branch. Same for other patches and they are ready to go to wine-patches. Cheers, Jacek

March 11, 2014

Re: Ideas about mshtml for GSoC 2014

by Zhenbo Li

Sorry, I forgot to mention that my IRC nickname is endle, and you can find me on #winehackers at 20:30 - 23:30 (UTC +8) nearly every day. As it's time to submit proposal, I'll send mine in this week. I'm open to your suggestions. Thank you. 2014-03-10 12:21 GMT+08:00 Zhenbo Li <litimetal(a)gmail.com>: > Hello, > > I'm Zhenbo Li, a student in Shanghai JiaoTong University, China. I > have been a Wine user for almost 4 years, and I've been preparing for > GSoC for several month. > > Since 2013, I've translated some wiki pages into Chinese[1], reported > nearly 40 bugs[2], involved in Appdb[3], and sent few patches[4]. I'm > glad to do more contribution to Wine in 2014. > > While using Wine, it is common to face a bug in mshtml[5]. Native > iexplore may be a workaround, but it can cause other problems, too. So > my idea is to implement some functions in mshtml.dll. > > I have a rough plan now, and I wish to get your advice. As I have > about 3 months, I want to implement some functions in IHTMLTable, > IHTMLTableCol, etc., and write testcaes for them. I've written some > draft patches for IHTMLTableRow, and I'll send them to wine-patches > after improving them with your advice. I hope I can establish a proper > schedule after communicating with my mentor. > > I'm also open to any suggestion, and any comment or advice is great appreciated. > > Thank you. > > FYI, my first name is Zhenbo, and my last(family) name is Li. So > Zhenbo Li and Li Zhenbo means the same person lol. > > > [1]: I contributed to FAQ, RegressionTesting, Bugs, etc. > [2]: http://bugs.winehq.org/buglist.cgi?email1=litimetal%40gmail.com&emailreport… > [3]: http://appdb.winehq.org/objectManager.php?sClass=maintainerView&iId=281642 > [4]: http://source.winehq.org/git/wine.git/?a=search&h=HEAD&st=author&s=Zhenbo+Li > [5]: http://bugs.winehq.org/buglist.cgi?cmdtype=runnamed&list_id=141898&namedcmd… > > -- > Have a nice day! > Zhenbo Li -- Have a nice day! Zhenbo Li

March 10, 2014