December 2006 - Wine-devel - list.winehq.org

Re: d3d: Fix EnumAdapterModes and add an adapter mode format test
by Stefan Dösinger 09 Dec '06

09 Dec '06

Am Samstag 09 Dezember 2006 19:28 schrieb Chris Robinson: > This patch fixes D3D adapter mode enumeration, which clears some issues > some games had when starting (eg. Morrowind being unable to set a mode). > I've added an adapter test (modified from the texture test) to D3D 9 which > tests all the D3DFORMAT types. > > Apologies for the macro (a)buse in the test, but it would've been more > combersome to have to manually make strings for each of the modes without > it. Do not apply this patch yet. The WINED3DFMT_X1R5G5B5 test fails on my brothers windows xp system. Win2k system also seem to have sporadic test failures.

1 0

re: First case of a real virus infection on wine
by Dan Kegel 09 Dec '06

09 Dec '06

Stefan wrote: >What we noticed pretty fast was that at least 3 windows boxes were infested >with a bunch of malware, and trying to infect each other over network shares. >The virus alert messages popping up made playing impossible. The obvious >solution: Disable the virus scanners. I think that qualifies you for the Darwin award! Can you try Christoph's ClamAV integration, http://www.christoph-probst.com/soc2006/wine/ and see if it would have prevented the infection? - Dan

3 2

First case of a real virus infection on wine
by Stefan Dösinger 09 Dec '06

09 Dec '06

Hi, Seems like I just got the victim of the first infection of a windows virus on wine :-( . I got infected with some virus called W32.Parite.B. Unlike most of the biests out there it is a real virus that attaches itself to existing .exe files, not a stand-alone worm. I caught it yesterday on a small private gaming session. I didn't have Warcraft 3 installed, so instead of installing and messing with patches I would have had to download I decided to copy an existing installation over. That seemed to work fine at first :-/ What we noticed pretty fast was that at least 3 windows boxes were infested with a bunch of malware, and trying to infect each other over network shares. The virus alert messages popping up made playing impossible. The obvious solution: Disable the virus scanners. So with the protection disabled those windows boxes were able to play, everything seemed to work fine. I noticed that something was wrong when my Battlefield 1942 crashed, which worked a few hours before when tested for regression in my new patches. ClamAV showed up a W32.Parite.B infection in bf1942.exe. That virus wasn't only in bf1942.exe, I found it in the war3 installation I copied over too. Looks like it came from there. I found it in all .exe files on my fake C:\ drive, except of our fake .exe's in C:\windows\system32. Looks like it didn't like those. Next I had a look at my real windows installation mounted in /media/windows, and found it infected too. Well, that was easy to clean up with a mkfs because I didn't have anything valueable in there. Well, the thing that is rather bad is that it infected my downloaded game demos and other files on my home drive. Luckily I didn't have my external hard drive with the rest of my stuff attached when I ran wine. So I've now deleted my wine installation, windows installation and all .exe files on my disk. I'm scanning my whole linux drives to be sure, but I didn't run wine as root and I'm confident that the Linux file security prevented the worst problems. To summarize, I got into the trouble mainly because I ignored the basic security guidelines. I ran executables from a really not trustworthy source, knowing that my friends' windows boxes are in a bad shape quite often. My real windows installation got infected because I had it world writeable for no real reason. On the bright side, running as a least priviledged user prevented the worst problems. If anyone wants to play around with that virus, I kept my infected GenuineCheck.exe :-) Stefan

1 0

Re: appdb/include version.php
by Tony Lambregts 09 Dec '06

09 Dec '06

WineHQ wrote: > ChangeSet ID: 30433 > CVSROOT: /opt/cvs-commit > Module name: appdb > Changes by: wineowner(a)winehq.org 2006/12/08 22:28:20 > > Modified files: > include : version.php > > Log message: > Alexander Nicolaysen Sørnes <alex(a)thehandofagony.com> > Colourize the maintainer's rating and version fields with the appropriate > rating colours, both in the application and version view. > This should make it easier for users to identify working versions, and it also > prevents the maintainer's rating from being 'overshadowed' by the test > results. > > Patch: http://cvs.winehq.org/patch.py?id=30433 > > Old revision New revision Changes Path > 1.86 1.87 +12 -4 appdb/include/version.php > > Index: appdb/include/version.php > diff -u -p appdb/include/version.php:1.86 appdb/include/version.php:1.87 > --- appdb/include/version.php:1.86 9 Dec 2006 4:28:20 -0000 > +++ appdb/include/version.php 9 Dec 2006 4:28:20 -0000 > @@ -662,9 +662,14 @@ class Version { > echo " </td></tr>\n"; > } > > + if($this->sTestedRating != "/" && $this->sTestedRating) > + $sMaintainerColor = $this->sTestedRating; > + else > + $sMaintainerColor = "color0"; > + > // rating Area > - echo "<tr class=\"color1\" valign=\"top\"><td>Maintainer Rating</td><td>".$this->sTestedRating."</td></tr>\n"; > - echo "<tr class=\"color0\" valign=\"top\"><td>Maintainers Version</td><td>".$this->sTestedRelease."</td></tr>\n"; > + echo "<tr class=\"$sMaintainerColor\" valign=\"top\"><td>Maintainer’s Rating</td><td>".$this->sTestedRating."</td></tr>\n"; > + echo "<tr class=\"$sMaintainerColor\" valign=\"top\"><td>Maintainer’s Version</td><td>".$this->sTestedRelease."</td></tr>\n"; > > // image > $img = Screenshot::get_random_screenshot_img($oApp->iAppId, $this->iVersionId, false); > @@ -923,12 +928,15 @@ class Version { > // set row color > $bgcolor = ($c % 2 == 0) ? "color0" : "color1"; We need to initialize aRatingColor like this... $sRatingColor = "class=\$bgcolor\""; > > + if($oVersion->sTestedRating && $oVersion->sTestedRating != "/") > + $sRatingColor = "class=\"$oVersion->sTestedRating\""; > + or it needs an else like so... else $sRatingColor = "class=\$bgcolor\""; Otherwise we get this problem. http://appdb.winehq.org/appview.php?iAppId=8 > //display row > echo "<tr class=$bgcolor>\n"; > echo " <td><a href=\"".BASE."appview.php?iVersionId=".$iVersionId."\">".$oVersion->sName."</a></td>\n"; > echo " <td>".util_trim_description($oVersion->sDescription)."</td>\n"; > - echo " <td align=center>".$oVersion->sTestedRating."</td>\n"; > - echo " <td align=center>".$oVersion->sTestedRelease."</td>\n"; > + echo " <td $sRatingColor align=center>".$oVersion->sTestedRating."</td>\n"; > + echo " <td $sRatingColor align=center>".$oVersion->sTestedRelease."</td>\n"; > echo " <td align=center>".Comment::get_comment_count_for_versionid($oVersion->iVersionId)."</td>\n"; > echo "</tr>\n\n"; > > > >

1 0

XRenderFindDisplay crash / user32 winex11 initialisation order problem
by Peter Oberndorfer 08 Dec '06

08 Dec '06

Hi, today i wanted to create a test for CreateCompatibleDC so i created a exe with this content [1] just to find out it crashes wine inside XRenderFindDisplay [2] The reason is that the CreateCompatibleDC call in gdi executes LoadLibrary winex11.drv which in turn also loads user32.dll, because winex11.drv links to it. Then in DllMain user32 registers some builtin classes which causes it to call into winex11. But at this point DllMain of winex11 had no chance to execute and XRenderQueryExtension gets passed a NULL pointer which makes it unhappy. Unfortunately i have no idea how to properly solve this problem :-( * Use UserClientDllInitialize for initializing user32 stuff as the comment above DllMain mentions it? * I tried linking gdi32 to user32(windows seems to do that) and inserted a dummy IsCharAlphaA('c'); to make sure it is *really* linked :-) And it does not crash anymore, but i don't know if i break any assumptions with that. Also i think it is preferable to defer loading the driver until it is really needed. If you have a better idea of how to fix this please tell me. Greetings Peter 1: simplified version, compile with winegcc foo.c -lgdi32 #include <windows.h> int main(int argc, char ** argv){ CreateCompatibleDC(NULL); } 2: see attachment

1 0

Re: advapi32: Fix comparisons of unsigned expressions with zero
by Mike McCormack 08 Dec '06

08 Dec '06

Andrew Talbot wrote: > - if( n < 0 ) > + if( (LONG)n < 0 ) Why not change the type of n to LONG rather than using a cast? Mike

2 1

mshtml as html parser
by Kirill K. Smirnov 08 Dec '06

08 Dec '06

Hi, all! Just a little question: can wine's current implementation of mshtml.dll be used as html parser? I need to convert html code into a tree without displaying it. I tried to look through the code, but I'm not familiar with OLE. BTW some E_NOIMPL confused me. Thanks in advance

3 2

Re: mshtml #1: Added get_parentWindow implementation.
by Detlef Riekenberg 08 Dec '06

08 Dec '06

Hi. I have no Idea about COM, but while i was reading your Patch to learn, I do not understand this pair: > + > +#define HTMLWINDOW2_THIS(iface) DEFINE_THIS(HTMLWindow, HTMLWindow2, > iface) > + > +#undef HTMLWINDOW_THIS > + Did you mean +#undef HTMLWINDOW2_THIS > here? -- By by ... Detlef

2 1

audio glitch patch by mike hearn
by kfs1＠online.no 08 Dec '06

08 Dec '06

I'm trying to implement this patch: http://wiki.winehq.org/Implement_SetThreadPriority by Mike Hearn which will enable wine to work without all the nasty audio glitches when using alsa. However hunk 1 fails on configure.ac as the patch hasn't been updated to latest wine. Can anyone help me get this baby compiled as i want to play my games without the nasty audio glitches and don't care if i have to be root/sudo/whatever. If there are any details (does wineserver/wine* has to be setuid root? do i need any special settings in winecfg, will it work with sudo or do i have to be root,etc etc) I would like to get informed of them also as it will probably take me days to figure it out on my own <:) some guy rediffed(whatever that means) it here: http://bbs.archlinux.org/viewtopic.php?t=27529&highlight= but it doesn't seem to work. Thanks for all the help. I'm asking here as Mike is too busy to fix it, but he says it should be an easy fix. -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/

8 16

Re: [doc]: Fix system32, winecfg and Printers
by Kai Blin 08 Dec '06

08 Dec '06

On Friday 08 December 2006 01:03, Detlef Riekenberg wrote: > - <para> > - The next few settings primarily affect games and are somewhat > + The first few settingsare primarily affect games and are You missed a space there. > somewhat self-explanatory. You can prevent the mouse from leaving the > window of a DirectX program (i.e. a game.) and the default is Cheers, Kai -- Kai Blin, <kai Dot blin At gmail Dot com> WorldForge developer http://www.worldforge.org/ Wine developer http://wiki.winehq.org/KaiBlin/ -- Will code for cotton.

1 0