January 2011 - Wine-devel - list.winehq.org

Wine and D3D
by Emanuele Oriani 08 Jan '11

08 Jan '11

Hi wine developers, I'm a wine user and StarCraft II player. Unluckily Blizzard has left D3D only mode in win32 executable. The game doesn't use CPU/GPU at 100%, but on Linux what happens is that we have a lot less FPS than windows. To get some more FPS, like 50% more, we set one core (or all cores) to max frequency and then we set the 'affinity' of SC2.exe to that core only (to set the affinity I created a simple executable). This way performance is better indeed, but nowhere near windows. (I've opened a bug for this matter http://bugs.winehq.org/show_bug.cgi?id=24558 ). My fear is that Blizzard has introduced some sort of FPS limit and probably wine doesn't cope very well with that. And, potentially they're using multi-threaded rendering as well. I've done some thread analysis and apparently the many threads the game execute stay in sleep (they actively invoke SleepEx) for 95% of the time, which is very strange. This is why if we leave the CPU frequency in 'Ondemand' mode it goes to 800MHz (or the lowest frequency). Could it be related to the rendering infrastructure? I know OpenGL is essentially mono-threaded (I've used it for quite a bit in the past years). During some research I came across a Stefan's email from July 2006, were he was proposing the right thing (imho) to do to support multi-threaded D3D: a GL rendered thread per device/buffer. What happened with that? How is currently multi-threaded D3D rendering implemented in wine? Thanks again, Cheers, Ema! ;-) Ps. Thanks to wine and a simple 'fprinft' I managed to easily instrument the SleepEx function and gather such valuable information. Now I'd like to do the same for D3D part, understand why the main renderer thread goes to sleep (or not) and if it goes to sleep actively or it's waiting on some mutex, or something else.

1 0

wiki work, wine64 portability issues
by Thomas Heckel 08 Jan '11

08 Jan '11

Hi all, I was just cleaning up some "Janitorial" sections on wine wiki. Now several questions arise: Just for interest I looked for a kind of wine64 portability guide to link to the http://wiki.winehq.org/PortabilityFixes . Are there any documents with source patterns which have been fixed to get Wine 64-bit ready? I think some of the listed janitorial tasks are already finished. E.g. there weren't any commits for 16bit separation for a longer time (Alexandre?). I don't have any overview if this tasks are really finished or if there are still some sections remaining? As a proposal to be discussed for the "wiki people": I would like to make a category "CategoryAbout" to collect the wiki articles which are (should) directly linked from the about page on the winheq.org site. I think this would be a good first way to solve the still existing old/duplicated pages between wiki and winheq.org pages. What do you think? Thanks in advance, Thomas

1 0

Re: ws2_32: Add default case for family in debugstr_wsaioctl.
by Juan Lang 08 Jan '11

08 Jan '11

Hi Gerald, > This mimicks what we already do a few lines above for buf_type. It's also unneeded: switch(ioctl & 0x18000000) There are only 4 possible combinations of ioctl & 0x18000000. Three are covered in this switch statement, and the fourth is eliminated due to the outer switch statement. You'd be adding dead code. --Juan

2 1

Wine and D3D
by Emanuele Oriani 08 Jan '11

08 Jan '11

Hi wine developers, I'm a wine user and StarCraft II player. Unluckily Blizzard has left D3D only mode in win32 executable. The game doesn't use CPU/GPU at 100%, but on Linux what happens is that we have a lot less FPS than windows. To get some more FPS, like 50% more, we set one core (or all cores) to max frequency and then we set the 'affinity' of SC2.exe to that core only (to set the affinity I created a simple executable). This way performance is better indeed, but nowhere near windows. (I've opened a bug for this matter http://bugs.winehq.org/show_bug.cgi?id=24558 ). My fear is that Blizzard has introduced some sort of FPS limit and probably wine doesn't cope very well with that. And, potentially they're using multi-threaded rendering as well. I've done some thread analysis and apparently the many threads the game execute stay in sleep (they actively invoke SleepEx) for 95% of the time, which is very strange. This is why if we leave the CPU frequency in 'Ondemand' mode it goes to 800MHz (or the lowest frequency). Could it be related to the rendering infrastructure? I know OpenGL is essentially mono-threaded (I've used it for quite a bit in the past years). During some research I came across a Stefan's email from July 2006, were he was proposing the right thing (imho) to do to support multi-threaded D3D: a GL rendered thread per device/buffer. What happened with that? How is currently multi-threaded D3D rendering implemented in wine? Thanks again, Cheers, Ema! ;-) Ps. Thanks to wine and a simple 'fprinft' I managed to easily instrument the SleepEx function and gather such valuable information. Now I'd like to do the same for D3D part, understand why the main renderer thread goes to sleep (or not) and if it goes to sleep actively or it's waiting on some mutex, or something else.

1 0

Forum/mailing list link broken?
by Rosanne DiMesio 07 Jan '11

07 Jan '11

Replies from the mailing list are not showing up on the forum in some cases. A user mentioned it in this thread: http://forum.winehq.org/viewtopic.php?t=10782. I checked the mailing list archives, and there are other threads that have replies from mailing list users that never appeared on the forum. -- Rosanne DiMesio <dimesio(a)earthlink.net>

3 5

Re: [PATCH 3/2] msvcrt: Implement _wcslwr_s
by Eric Pouech 07 Jan '11

07 Jan '11

Le 07/01/2011 15:42, Detlef Riekenberg a écrit : > based on _strlwr_s > > This let Radiotracker go further > > > /********************************************************************* > + * _wcslwr_s (MSVCRT.@) > + */ > +int CDECL _wcslwr_s(WCHAR *str, MSVCRT_size_t len) > +{ > + WCHAR *ptr = str; > + > + if (!str || !len) > please use the MSVCRT_CHECK_PMT as done in the other _s msvcrt's functions A+ -- Eric Pouech "The problem with designing something completely foolproof is to underestimate the ingenuity of a complete idiot." (Douglas Adams)

1 0

Re: Wine for Solaris 10 x86 roadblocks
by seeger＠cise.ufl.edu 07 Jan '11

07 Jan '11

+------ "Ron Whites" wrote (Fri, 7-Jan-2011, 12:23 -0600): | | I downloaded Solaris 10 x86 on 12/6/2010 (how do I know the build/ver?) NB: SunOS 5.N (plus networking, windowing, etc.) <=> Solaris N. On a SPARC box (most easily available Sol10): % uname -a SunOS <hostname> 5.10 Generic_141444-09 sun4v sparc SUNW,SPARC-Enterprise-T5220 % cat /etc/release Solaris 10 10/09 s10s_u8wos_08a SPARC Copyright 2009 Sun Microsystems, Inc. All Rights Reserved. Use is subject to license terms. Assembled 16 September 2009 (BTW, "wos" stands for "wad of stuff".) And on an x86 system running Solaris 11 Express: % uname -a SunOS <hostname> 5.11 snv_151a i86pc i386 i86pc % cat /etc/release Oracle Solaris 11 Express snv_151a X86 Copyright (c) 2010, Oracle and/or its affiliates. All rights reserved. Assembled 04 November 2010 (BTW, "snv" stands for "Solaris next version".) Do note that packaging and patching have changed radically in OpenSolaris and Solaris 11, and that free access to security patches is no more. Solaris 11 Express may have better driver support for your hardware. You might want to search for the hardware compatibility list, though beware that it may not be perfectly up to date, as it depends on user submissions. Also, there is something of a fork from OpenSolaris, called OpenIndiana, which is another option to consider. The Solaris story is a bit complicated, beyond the scope of this reply. 8-) | ./tools/wineinstall again. | this time I received a line 149 syntax error Most random software these days isn't regularly maintained on Solaris. In this case, most shell scripts require a POSIX standard shell, though bash-specific extensions frequently creep in. On Solaris 10 /bin/sh is the old SVR4 Bourne shell, while on Solaris 11 it is ksh93. The /usr/bin utilities are legacy versions on Solaris 10, while the standards compliant utilities are in /usr/xpg4/bin. On Solaris 11, /usr/gnu/bin is placed first on the default user path. "getconf PATH" is the standard method to obtain a standard PATH, but it seems that no one observes it, and its output hasn't been updated for Solaris 11 (which also has /usr/xpg6/bin). So, generally, scripts will be problematic. | A bug report at | http://wine.1045685.n5.nabble.com/Bug-19321-New-Syntax-error-during-inst | allation-td1605805.html | | That says Don't use 'wineinstall script', do './configure && make | install' instead. Yup. | Solaris10 doesn't have a make pre-installed and I can't find where I can | get one, so I am presently stuck again. A full install will have a make, but it will be in /usr/ccs/bin/, which is included in the output of "getconf PATH". However, the Sun compilers are a separate install. But, instead you will need gcc and gnu make to compile wine. These might already be in /usr/sfw/bin (Solaris 11 has them in /usr/gnu/bin, with symlinks to them in /usr/bin, e.g. "gmake"). Precompiled packages are available at sunfreeware.com, opencsw.org and blastwave.org, as well as sunfreepacks. | Sure seems a lot of road blocks so far, can anybody offer some | information to get me over the hurdles? The OpenSolaris releases from Sun were geared to be both more complete with some popular software and more familiar to Linux users. With the Oracle acquistion there are no more of these, but the Solaris Express preview releases are available with more restrictive licensing. I can send configure options for Solaris 11, but I have never built wine on Solaris 10. HTH, Chuck

1 0

Wine for Solaris 10 x86 roadblocks
by Ron Whites 07 Jan '11

07 Jan '11

I am a newbie to Solaris as well as Wine. I downloaded Solaris 10 x86 on 12/6/2010 (how do I know the build/ver?) and loaded it in a Virtual Machine on my Windows 7 Desktop. Then recently I have been trying to get Wine om Solaris 10 working so far with no success. I downloaded Swine-sol10-1.1.40.7z from http://www.sunfreepacks.com/ but I have not been able to unpack this by using "p7zip -d Swine-sol10-1.1.40.7z" on my solaris. It returns immediately to the terminal prompt and it apparently does nothing but gives no error. So at http://www.winehq.org/announce/1.1.19 I found another Wine-1.1.19.tar.bz2 I used bunzip and then tar to get just Wine-1.1.19 I then did "./tools/wineinstall and encountered an error saying I couldn't/shouldn't install under ROOT so after making a user account via useradd I did ./tools/wineinstall again. this time I received a line 149 syntax error looking on the web I found A bug report at http://wine.1045685.n5.nabble.com/Bug-19321-New-Syntax-error-during-inst allation-td1605805.html That says Don't use 'wineinstall script', do './configure && make install' instead. Solaris10 doesn't have a make pre-installed and I can't find where I can get one, so I am presently stuck again. Sure seems a lot of road blocks so far, can anybody offer some information to get me over the hurdles?

2 1

Re: {USB}Fail to patch 0002-Re-generate-some-files.txt
by James Mckenzie 07 Jan '11

07 Jan '11

Qian Hong <fracting(a)gmail.com> wrote: >Sent: Jan 6, 2011 9:56 PM >To: James McKenzie <jjmckenzie51(a)earthlink.net> >Cc: wine-devel(a)winehq.org >Subject: Re: {USB}Fail to patch 0002-Re-generate-some-files.txt > >Dear James, thanks for reply. I think the current vesion of this patch is >for wine1.3.10, the upload date is Dec 26. >I'll try to attempt it. > I had not checked this as the last version was posted some time ago. I'll have to look at it and see if it applies to the soon to be released (if not released by the time I get to my Wine build machine) 1.3.11. James McKenzie

1 0

{USB}Fail to patch 0002-Re-generate-some-files.txt
by Qian Hong 06 Jan '11

06 Jan '11

Dear all, After checkout latest wine from git, I tried to apply the USB support patches, but something is wrong. Checking out from the WineHQ Git repository:(Follow GitWine wiki) $git clone git://source.winehq.org/git/wine.git ~/wine-git $cd ~/wine-git Get the wine usb patches: (Follow wine USB wiki) ftp> cd pub/people/amorozov/usb/current ftp> get 0001-Add-support-of-native-Windows-drivers-for-USB-tokens.txt ftp> get 0002-Re-generate-some-files.txt Apply the patches: $git am 0001-Add-support-of-native-Windows-drivers-for-USB-tokens.txt Applying: Add support of native Windows drivers for USB tokens. $git am 0002-Re-generate-some-files.txt Applying: Re-generate some files. error: patch failed: configure:11442 error: configure: patch does not apply error: patch failed: configure.ac:2775 error: configure.ac: patch does not apply Patch failed at 0001 Re-generate some files. When you have resolved this problem run "git am --resolved". If you would prefer to skip this patch, instead run "git am --skip". To restore the original branch and stop patching run "git am --abort". Could anyone tell me what did I have done wrong? Many thanks! -- Regards, Qian Hong - Send from Ubuntu http://www.ubuntu.com/

2 2