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

Re: [PATCH 3/2] msvcrt: Implement _wcslwr_s
by Eric Pouech Jan. 7, 2011

Jan. 7, 2011

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 Jan. 7, 2011

Jan. 7, 2011

+------ "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 Jan. 7, 2011

Jan. 7, 2011

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 Jan. 7, 2011

Jan. 7, 2011

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 Jan. 7, 2011

Jan. 7, 2011

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

Wine and Fontconfig
by Yaron Shahrabani Jan. 6, 2011

Jan. 6, 2011

Elad, an Hebrew user suggests: In several Wine apps Hebrew is displayed as square glyphs since there are no builting Hebrew glyphs in Wine's font (bug #23537). The question is why does Wine use the internal font as the default? The easier solution would be to use the "system" fonts as configured in fontconfig. Windows can have a custom font as well so it shouldn't cause and difficulty. This way we won't have to look for a font editing specialist that will add glyphs to wine's fonts. This problem can affect many languages as well, not just Hebrew, that lacks the required glyphs. Using the fonts from fontconfig might solve the problem and will lead to a better integreation. The fonts in Wine apps will look just like the fonts used by the rest of the apps in the system. Kind regards, Yaron Shahrabani <Hebrew translator>

2 1

Unable to download USB patch from ftp://ftp.etersoft.ru/pub/people/amorozov/usb
by Qian Hong Jan. 6, 2011

Jan. 6, 2011

Dear All, While trying usb device in wine, I follow this page : http://wiki.winehq.org/USB , The Official Wine Wiki. It says, "Some functions used by USB drivers are implemented on top of libusb-0.1/1.0 by Alexander Morozov from Etersoft (amorozov(a)etersoft.ru). Now only some USB software protection tokens are working. Patches: ftp://ftp.etersoft.ru/pub/people/amorozov/usb" But I can't connect to ftp.etersoft.ru , ping ftp.etersoft.ru : PING updates.etersoft.ru (212.176.200.18) 56(84) bytes of data. 64 bytes from legirus.ru (212.176.200.18): icmp_req=1 ttl=40 time=371 ms 64 bytes from legirus.ru (212.176.200.18): icmp_req=2 ttl=40 time=371 ms 64 bytes from legirus.ru (212.176.200.18): icmp_req=3 ttl=40 time=371 ms Port 21 is open. Could anyone give me some tip? Thanks! -- Regards, Qian Hong - Send from Ubuntu http://www.ubuntu.com/

2 2

New winetricks 20110105: improved download error recovery, added first-run optin dialog. New winetricks 20110105-alpha: menu divided into categories, added 35 games
by Dan Kegel Jan. 6, 2011

Jan. 6, 2011

Another month, another Winetricks - or, rather, two new Winetricks. New version 20110105 has only two notable changes: 1. it will retry a corrupt download for you (this should get rid of most cryptic warnings about needing to rename a file and retry), 2. a first run dialog now asks whether you'd like to help winetricks development by reporting usage stats. You should only see this dialog once (and only if you're using the gui, and only if you're using zenity). This has been very helpful already, as people have been running it from svn. Please let me know what you think of it. Online as always at http://kegel.com/wine/winetricks Experimental new version 20110105-alpha has a few further changes: 3. The menu is now broken up by category; there's now a top-level menu which gives you a choice of category (dll, font, app, or setting). This ought to make the menu a little less overwhelming. Let me know what you think of the new menu navigation. 4. There's now a progress dialog during downloads in gui mode. 5. It should work better when the user has removed the Z: drive. 6. It's been refactored to make adding new verbs much easier (you no longer have to edit the script in multiple places to add a verb, and you can even put verbs in external files). 7. Added 35 games. (A bunch more remain to be ported from wisotool.) 8. Removed (or, rather, haven't ported yet) a number of less-common verbs. The statistics gathered by the new version will tell us which of the remaining verbs to port first. The experimental version is only at http://kegel.com/wine/winetricks-alpha SVN and bug tracker are online at http://winezeug.googlecode.com Changes in version 20110105: Dan Kegel: - remove old synonym dotnet2 for dotnet20 - mpc: remove broken call to append_path - handle 'winetricks win31 vista' properly - make dotnet20sp2 warn that it's broken. Don't list in menu. - w_download: don't redownload if no expected checksum given. Also remove == typo from r2207. - don't declare checksum error unless, y'know, there was a checsum error - as favor for etersoft, avoid referring directly to zenity/kdialog/xmssage to avoid pulling them all in as dependencies when building a package - automatically retry download once if cached file has wrong checksum - ie8: avoid writing to / (just like ie7 fix). Don't bother running ie8 when finished; point user to workaround instead. Also move a verb that got sandwiched between ie6 and ie7. - ie7: when / is writable as on macosx, move z: out of the way during ie7 install to avoid scribbling temporary files in /. Had to run with WINEDEBUG=warn+heap to avoid memory bugs in setupapi? - when reporting usage, also report version. - in zenity gui case, ask user on first run whether to opt in to usage reporting. - on MacOS X, put cache in /home/dank/Library/Caches - delete msxml6.dll before running installer. Fixes bug 211. - tahoma: don't download to top of cache - finish renaming WINETRICKS_UNIXQUIET to W_UNATTENDED_DASH_Q - make cabextract quiet Austin English: - add win31 verb (needed internally by ole2)

5 6

RE: WyldBot
by Greg Geldorp Jan. 6, 2011

Jan. 6, 2011

From: wylda(a)volny.cz > WyldBOT1 and WyldBOT2 are not VM, but true physical older-machines > made only for wine-testing. This is great, it's good to have physical machines for testing too. > I used to run WylBOT 3x per week, but there were no changes or > effort to fix things, so i reduced the WylBOT run to once a week Would you be willing to consider some integration with TestBot? It isn't always immediately clear what the root cause of a failure is, so sometimes a bit of experimentation is required. Having to experiment by submitting official patches to wine-patches isn't an ideal solution. If it were possible to submit patches via TestBot then I think more people would be interested in fixing failures. It would require WyldBOT to be available at predictable times (24x7 would be best, but I guess "during the weekend" would work too). We'd have to find some way to exchange files and start a job on WyldBOT in a secure way, but that seems doable. Greg.

3 2

Re: [PATCH 1/4] server: Make it possible to report many filesystem events
by Alexandre Julliard Jan. 6, 2011

Jan. 6, 2011

Piotr Caban <piotr(a)codeweavers.com> writes: > diff --git a/server/protocol.def b/server/protocol.def > index 6245f4c..29a4c7a 100644 > --- a/server/protocol.def > +++ b/server/protocol.def > @@ -1449,6 +1449,7 @@ enum char_info_mode > obj_handle_t handle; > @REPLY > int action; /* type of change */ > + int next_length; /* next file name length */ > VARARG(name,string); /* name of directory entry that changed */ > @END Returning the next length won't work, because there's no guarantee that it will be the same file that we return the next time around, there can be multiple threads watching that handle. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: ddraw/tests: Force some tests to pass on XP/Win2003 Wow64
by Stefan Dösinger Jan. 6, 2011

Jan. 6, 2011

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'd say the D3DPTEXTURECAPS_PERSPECTIVE tests can be removed. They are here to demonstrate some behavior the game RollCage depends on. Newer drivers show some different behavior, I suspect they "fix" rollcage with an app specific hack. Am 06.01.2011 um 13:08 schrieb Greg Geldorp: > --- > dlls/ddraw/tests/d3d.c | 23 ++++++++++++++++++----- > 1 files changed, 18 insertions(+), 5 deletions(-) > > > > <0001-ddraw-tests-Force-some-tests-to-pass-on-XP-Win2003-W.patch> -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.16 (Darwin) iQIcBAEBAgAGBQJNJd5KAAoJEN0/YqbEcdMwq5cP/RGuOU0wpkzgxjdXLSkvI4R+ X8LNhLJOwRYJeQr8i2oYWvJ96VopronT9ZsIBbQErarHCY/CMEaMpUgUPQ01ko9K r5rrL/hBHSilmVcKJ8ToTKOgclnIfdIJDTJRR6nESSaBJsFuL6iH3RwD45nNeNOC jnYBscvFhgLJYOKFcdPGcnFXCzWNhJLZD9BXFOXsH5DhEaI/neB7aGiTSMpiRTbd SZmCWydkEV22Rd4h2JvwuxKFxN2ANZKkIH1W83sdOvKhEsSZM/mPDu7kTtLuUrS9 1lohtr1L/ehFXb2+NKUj+J+KZCp8sYw9CFD3kGN+/teIE41E0wzvih8kW+BISRkS LgwvvSwODly1K8SNduSWYad5rbKNqsHbSnLa+M18EkqOuDzjm4gFF7iyXmmLwvjm zKP8GrbYPQg4kITCqooO05xtmtPUH4tFik1ZIT7uj/OInPqUaDCf8k1SAQQfN9Jz EArZAaxmesNE9tRGuVM2s2HfEGco1fgmeJm+abCfZng89l26T30SMApJuw3e5kDq RBYQf8VcHgg1Jnzt4YRUFlwiYSMYBqVUow5mHu7lEphil5iKEEcf0ZJFtSpZHtuU GHFe2gMix91ren1GpRVSFs7k4QoNyBXhp6r4xOjgS1KINjxlmFsN6ex+vKRMbedC FTYVe/2U/ecfL2PE/u4D =Y3ex -----END PGP SIGNATURE-----

1 0

Test timeouts
by Greg Geldorp Jan. 6, 2011

Jan. 6, 2011

Although the number of timeouts in msi tests has dropped considerably after Hans decided to do something about them, they're unfortunately not completely gone. It seems that the remaining timeouts have a strong dependency on the environment. It's possible to tweak the environment (disable system restore points, for TestBot I found it makes a difference on which physical disk the virtual disk files are located) but the goal should be to have the tests run consistently on whatever the environment happens to be. Perhaps it's time to accept that some tests in some environments can simply take longer than the current winetest.exe timeout value of 2 minutes. I can think of 3 possible solutions: 1) Simply increase the timeout value in winetest.exe. Extremely simple to implement, but perhaps not very elegant. 2) Pass the timeout value as a command line parameter to winetest.exe. This has the advantage that the timeout can be adjusted for the environment (creating a system restore point on an older machine with lots of stuff installed will take longer than creating a system restore point on a fresh high-end machine), it does put the responsibility for determining a suitable timeout on the shoulders of the tester. That's probably no problem for regular testers, but we can't expect the occasional tester to do this. 3) Have a variable timeout per testset. One possible way to implement this would be to have the test executables recognize an additional command line parameter "--timeouts". When called with this parameter, the test executable tries to load a resource string containing a list of testset-specific timeout values, for msi it could be something like "action:180;install:180;msi:300". Testsets not included in the resource string would get the default timeout of 120 sec, if the resource string is not present all testsets would get that default. I believe this can be implemented in wine/test.h, so the only change required to the tests themselves would be to add the resource string to tests that require a different timeout. Instead of calling the test executable with the "--list" command line parameter winetest.exe would call it with "--timeouts", which would produce a list of the testssets with their timeouts. The advantage of this method is that it's quite flexible. It is a bit more work to implement. Personally, I have a slight preference for solution 3) but I can live with the others too. Obviously I'm willing to implement whatever solution we choose. Greg.

4 6

RE: msi/tests: Mark a test result on some old Wow64 platforms as broken.
by Greg Geldorp Jan. 6, 2011

Jan. 6, 2011

Hi Hans, > + RegDeleteKeyA(prodkey, ""); Since prodkey was possibly opened with KEY_WOW64_64KEY, you'll probably need delete_key() here. Greg.

1 0

Re: [PATCH 2/2] shell32: Don't fill SHITEMID data with random values if file doesn't exist
by Alexandre Julliard Jan. 6, 2011

Jan. 6, 2011

Piotr Caban <piotr(a)codeweavers.com> writes: > @@ -538,6 +538,8 @@ static char* UNIXFS_build_shitemid(char *pszUnixPath, LPBC pbc, void *pIDL) { > if (stat(pszUnixPath, &fileStat)){ > if (must_exist || errno != ENOENT) > return NULL; > + > + memset(&fileStat, 0, sizeof(struct stat)); That may not be random, but it's still wrong values. It seems to me that this is what the IFileSystemBindData should be used for. -- Alexandre Julliard julliard(a)winehq.org

1 0

"N" shows up as missing on the translations status website
by Sven Baars Jan. 5, 2011

Jan. 5, 2011

Hey all, since today, the "N" string shows up as missing on http://source.winehq.org/transl/ even though it's there in the po file. This happened for both xcopy and cmd. In the po file it says #: cmd.rc:257 xcopy.rc:41 msgctxt "No key" msgid "N" msgstr "N" so this is probably because the string is present in both cmd and xcopy. Just thought I'd let you know. Sven

2 1

/ is writable on mac... leading to lots of awkward msi droppings?
by Dan Kegel Jan. 5, 2011

Jan. 5, 2011

Quentin Paris informs me that the root directory is user-writable on MacOSX (shudder), which means our beloved z: mapping leads to lots of garbage directories named e.g. /fd91216cc2ce81eb2ce16a058619 after running msi installers. Is this something Wine's msi could avoid doing?

4 4

WyldBot
by wylda＠volny.cz Jan. 5, 2011

Jan. 5, 2011

Hi, i take the chance to make this info publicaly aware, so adding wine-devel > Hi, > your WyldBot has some Audio/Video issues in it's > XP machines no other machines have, in which VM do > they run? WyldBOT1 and WyldBOT2 are not VM, but true physical older-machines made only for wine-testing. WyldBOT1 has more OS so little explanation here: * WyldBOT1-ie6 was cleanlly installed, patched and never run wine-test (to avoid broken registry thing by wine-testsuite sometime ago), than made image of that partition * for WyldBOT1-ie7, i restored image of WyldBOT1-ie6 and installed IE7 on top of it * for WyldBOT1-ie8, i restored image of WyldBOT1-ie6 and installed IE8 on top of it * for WyldBOT1 64bit edition i did exactly same installation steps like in 32bit * for WyldBOT2 i did exactly same installation steps like in WyldBOT1-ie6, but runs on different HW box (it's not the same image like WyldBOT1-ie6, only same installation steps) So things works that way, that when i power up that machine and boot into WyldBOT-recovery and then * recovery takes the first image from second HDD * changes boot.ini sequence * reboots and then boots into WyldBOT1-ie6 * WyldBOT1-ie6 waits 120sec to finish all services startup and not disturb winetest * downloads the wine test suite from FTP + some testing video file etc. * it does the winetesting * changes boot.ini and to boot into Wylbot recovery * Wylbot recovery takes another image * and so on and after last image the machine si turned off This Chrismas time (after 8months) Wylbot was restored just for windows update and updated images were store again. Once on wine-devel, i noticed some speculation, that it uses old drivers, so nvidia drivers got updated too on all WyldBOT1. I used to run WylBOT 3x per week, but there were no changes or effort to fix things, so i reduced the WylBOT run to once a week (i do my best not to miss friday's release, but sometime i run it on saturday/sunday. Maybe when the relese would be around 17h UTC, i could make it friday-regular ;) As WyldBOT2 was unlucky and did not get those fresh updates as his brother WyldBOT1, i didn't turn it on recently. WyldBOT1's HW: MB: Asus P5K-E WiFi AP Bios rev. 1305 Intel Core 2 Quad Processor Q9550 sSpec Number: SLB8V CPU SPeed: 2.83GHz Bus speed: 1333MH z L2 Cache size: 12MB Package Type: LGA775 Technology: 45nm 8GB RAM Kingston HyperX DDR2 2x KHX8500D2T1K2/4G (DDR800), CAS: 5-5-5-18 Nvidia 8600GT latest driver 260.99, latest DirectX June2010 redist v9.29.1962 (2010-06-07) Soundcard Asus's onboard (i guess it uses "SoundMAX ADI Audio Driver") Some printers Wheel wingman formula force gp2 Wheel Microsoft (thurstmaster??) If interested in details let me know, i can look closer at home. > Btw: you can use the winetest option "-i" to add a > description of the machine... That's a pity, that i didn't have this info one week back, when i did the image jobs. But maybe it's stored on the FTP in control file, so won't be a hard job - don't know now. > > Best Regards, André Hentschel PS1: On Monday i noticed, that other machines have D3Dcompiler_43 installed, but mine's not even if i cleanly installed June's 2010 DirectX. Do you guys use for testing those messy free-dll downloads??? PS2: Wylbot2 has Intel Graphic Card and used to have completely different 2D/3D?? testing-pictures (those red, greens,... flashing shapes). But i didn't find a time to report the differences.

1 0

Re: d3dcompiler_43/tests: fixed nonconformant C string literals in HLSL test suite
by Henri Verbeet Jan. 5, 2011

Jan. 5, 2011

On 4 January 2011 22:25, Travis Athougies <iammisc(a)gmail.com> wrote: > + "float4 test(float x: TEXCOORD0): COLOR" > + "{" > + " const float pi2 = 6.2831853;" > + " float calcd_sin = (sin(x * pi2) + 1)/2;" > + " float calcd_cos = (cos(x * pi2) + 1)/2;" > + " return float4(calcd_sin, calcd_cos, 0, 0);" > + "}"; Some newlines probably wouldn't hurt.

1 0

Re: kernel32/tests: Don't test function directly when reporting GetLastError()
by Marvin Jan. 4, 2011

Jan. 4, 2011

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=8130 Your paranoid android. === WNT4WSSP6 (32 bit pipe) === pipe.c:878: Test failed: WriteFile to not-yet-connected pipe pipe.c:880: Test failed: ReadFile from not-yet-connected pipe pipe.c:897: Test failed: WriteFile to disconnected pipe pipe.c:899: Test failed: ReadFile from disconnected pipe with bytes waiting === W2KPROSP4 (32 bit pipe) === pipe.c:878: Test failed: WriteFile to not-yet-connected pipe pipe.c:880: Test failed: ReadFile from not-yet-connected pipe pipe.c:897: Test failed: WriteFile to disconnected pipe pipe.c:899: Test failed: ReadFile from disconnected pipe with bytes waiting === WXPPROSP3 (32 bit pipe) === pipe.c:878: Test failed: WriteFile to not-yet-connected pipe pipe.c:880: Test failed: ReadFile from not-yet-connected pipe pipe.c:897: Test failed: WriteFile to disconnected pipe pipe.c:899: Test failed: ReadFile from disconnected pipe with bytes waiting === W2K3R2SESP2 (32 bit pipe) === pipe.c:878: Test failed: WriteFile to not-yet-connected pipe pipe.c:880: Test failed: ReadFile from not-yet-connected pipe pipe.c:897: Test failed: WriteFile to disconnected pipe pipe.c:899: Test failed: ReadFile from disconnected pipe with bytes waiting === WVISTAADM (32 bit pipe) === pipe.c:878: Test failed: WriteFile to not-yet-connected pipe pipe.c:880: Test failed: ReadFile from not-yet-connected pipe pipe.c:897: Test failed: WriteFile to disconnected pipe pipe.c:899: Test failed: ReadFile from disconnected pipe with bytes waiting === W2K8SE (32 bit pipe) === pipe.c:878: Test failed: WriteFile to not-yet-connected pipe pipe.c:880: Test failed: ReadFile from not-yet-connected pipe pipe.c:897: Test failed: WriteFile to disconnected pipe pipe.c:899: Test failed: ReadFile from disconnected pipe with bytes waiting === W7PRO (32 bit pipe) === pipe.c:878: Test failed: WriteFile to not-yet-connected pipe pipe.c:880: Test failed: ReadFile from not-yet-connected pipe pipe.c:897: Test failed: WriteFile to disconnected pipe pipe.c:899: Test failed: ReadFile from disconnected pipe with bytes waiting === W7PROX64 (32 bit pipe) === pipe.c:878: Test failed: WriteFile to not-yet-connected pipe pipe.c:880: Test failed: ReadFile from not-yet-connected pipe pipe.c:897: Test failed: WriteFile to disconnected pipe pipe.c:899: Test failed: ReadFile from disconnected pipe with bytes waiting === W7PROX64 (64 bit pipe) === pipe.c:878: Test failed: WriteFile to not-yet-connected pipe pipe.c:880: Test failed: ReadFile from not-yet-connected pipe pipe.c:897: Test failed: WriteFile to disconnected pipe pipe.c:899: Test failed: ReadFile from disconnected pipe with bytes waiting

1 0

Re: user32/tests: Don't test function directly when reporting GetLastError()
by Marvin Jan. 4, 2011

Jan. 4, 2011

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=8127 Your paranoid android. === WVISTAADM (32 bit clipboard) === Failure running script in VM: The specified guest user must be logged in interactively to perform this operation

1 0

re: include: Add ONCE macro for making debug output less noisy
by Dan Kegel Jan. 4, 2011

Jan. 4, 2011

Alexandre wrote: >> +#define WINE_ONCE(do_once) ... > > I don't think that buys anything over spelling it out explicitly. It makes it easier to grep for, which might be useful if you want to find all the places it's used to throttle messages that shouldn't be happening in the first place.

1 0

Re: include: Add ONCE macro for making debug output less noisy
by Alexandre Julliard Jan. 4, 2011

Jan. 4, 2011

Sven Baars <sven.wine(a)gmail.com> writes: > @@ -250,6 +250,12 @@ static inline const char *wine_dbgstr_longlong( ULONGLONG ll ) > #define WINE_DPRINTF wine_dbg_printf > #define WINE_MESSAGE wine_dbg_printf > > +#define WINE_ONCE(do_once) \ > + do{ static int wine_once_var; \ > + if (!wine_once_var) { \ > + wine_once_var = 1; \ > + do_once; } } while(0) I don't think that buys anything over spelling it out explicitly. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: winemaker: Add def file support
by Alexandre Julliard Jan. 4, 2011

Jan. 4, 2011

André Hentschel <nerv(a)dawncrow.de> writes: > @@ -2480,7 +2498,9 @@ sub generate_project_files($) > my $canon=canonize("@$target[$T_NAME]"); > $canon =~ s/_so$//; > > - if (@$target[$T_TYPE] == $TT_DLL) { > + if (@$target[$T_TYPE] == $TT_DLL && (@$project_settings[$T_FLAGS] & $TF_HASDEF)) { > + print FILEO "\$(${canon}_MODULE).so: \$(${canon}_OBJS) \$(${canon}_MODULE:%.dll=lib%.def)\n"; > + } elsif (@$target[$T_TYPE] == $TT_DLL) { There's no reason to add a lib prefix. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: advapi32: Parital Implement InitiateSystemShutdown(Ex) (try 2)
by Alexandre Julliard Jan. 4, 2011

Jan. 4, 2011

André Hentschel <nerv(a)dawncrow.de> writes: > @@ -232,6 +233,10 @@ BOOL WINAPI InitiateSystemShutdownExA( LPSTR lpMachineName, LPSTR lpMessage, > DWORD dwTimeout, BOOL bForceAppsClosed, BOOL bRebootAfterShutdown, > DWORD dwReason) > { > + if (!lpMachineName && !lpMessage && !dwTimeout) > + ExitWindowsEx(bRebootAfterShutdown ? EWX_REBOOT : 0 | > + bForceAppsClosed ? EWX_FORCE : 0, > + dwReason); > FIXME("%s %s %d %d %d %d\n", debugstr_a(lpMachineName), > debugstr_a(lpMessage), dwTimeout, bForceAppsClosed, > bRebootAfterShutdown, dwReason); I don't think you want to call user32 functions here. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: wined3d: Revert vertex blend capabilities patch
by Stefan Dösinger Jan. 4, 2011

Jan. 4, 2011

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Yep, the patch(f41fdb22...) should be reverted. As I explained on IRC, the correct approach is to implement a replacement shader for the vertex pipeline and implement vertex blending in it. Henri has a rebased version of my ARBvp experiments from 2008. Am 04.01.2011 um 05:42 schrieb Jeremy Chin: > My previous patch was accepted as f41fdb225cf1e1df494237664ce92e51179f3725, though Henri Verbeet pointed out it was incorrect. stefan on IRC warned that announcing blend capabilities without supporting them may cause issues, hence the revert.<0002-Revert-wined3d-Update-detection-of-vertex-blend-ca.txt> -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.16 (Darwin) iQIcBAEBAgAGBQJNIxBHAAoJEN0/YqbEcdMw8boQAIy4oDcaDv90bSyxMZJ4HwZy q/bKznK807Rt6UmL8p4OplFLdAOF+mMnqDmNT9vZxr5k7rWC3GDtW+5nohKwyl74 rpQI71lSTo1lixjPP0t+t2bASl9mQ/Gb2AfPqiFsG0dZmKxwP7aa/25h01ZucHTc c39j234nXxQ1po5j00PoaA0Ya5yeCQ3edFdBMSPorPTdYLVGuHWSA4qdVQdf6mq4 /luSIkMp/MUygqEMCvT9/elaRjCG1dJ9RE4JBXOCtK5CgkCxN+MdnVrTguGeWTtc UV2rwNTZA+4KZouegDQPeHb2qKoQRuNWLdM07/yvCE5rVDJBiAem+MYOjJ3LHsP2 nB3Vfff2zmC6mp+544aQIKFtu0RC2IVU+AbcdmRV6yDwbNK3/wZb3ZV1LcOZ5Z1h 9m5uINreq1R54FnjA1xaYPmWMutN8jVYn2SBPfLA11dKh0sFwsjS4PkUPH1fbAOi nmjwtd7r1z6AyoaP12XIrXGkTNSx8bMHQ9twfWq4D8IqbefMYtrkh896F/wfTIQI nT0FGeDL3gCafpW+wkGQT78tVMZXCAIx2rb+C3sZd/KcQzVCMcgbfFY6R5LnrYVf k+M6CsM0ZyAfIyMpkEFwruMAV0zYQ6NAMhFL4muiyRjY4UA2YvtLhPrNjOp5LAi0 EfbvhcGcDQeh4W45ljuv =HOqP -----END PGP SIGNATURE-----

1 0