September 2007 - Wine-Devel

carbon poker under wine
by Trent Waddington Sept. 13, 2007

Sept. 13, 2007

On 9/13/07, Eric Andersen <xxxx(a)xxx.com> wrote: > I was wondering if you could possibly throw me some pointers for > getting Carbon Poker working under wine. > > For the record I'm using ubuntu feisty and wine 0.9.44. I'll forward this to wine-devel, as someone else might want it. Well, first thing you should try is building wine from the git tree. This is painful and annoying because Linus is stupid and ugly but there's been a lot of fixes to the crypt32 stuff recently, and you'll need that: git clone git://source.winehq.org/git/wine.git wine cd wine ; ./configure ; make etc. You can try what I'm doing with your current version first, as it has some of this crypt32 stuff.. so it might work. I downloaded CarbonPoker.exe from the website, ran it with wine and it complains that I don't have flash8 and then just sits there and does nothing. I don't know what it is doing, so I just went downloaded Flash: wget http://fpdownload.macromedia.com/get/flashplayer/current/install_flash_play… ./wine install_flash_player.exe That installs happily. So, running CarbonPoker.exe again.. ./wine CarbonPoker.exe err:ole:COMPOBJ_DllList_Add couldn't load in-process dll L"C:\\windows\\system32\\Macromed\\Flash\\Flash9b.ocx" Uhhh, ok. Having a look in that dir I see a Flash9.ocx but no Flash9b.ocx, what the hell? cd ~/.wine/drive_c/windows/system32/Macromed/Flash/ ln -s Flash9.ocx Flash9b.ocx cd - ./wine CarbonPoker.exe and now it is happy and starts "Installing software..." which actually means it is updating the stub.. which it then runs, and starts "Installing software..." again.. You can ignore the "HTML rendering is currently disabled" where the installer would usually be telling you about the great benefits you get from playing Carbon Poker. Eventually this finishes. The installer goes away.. and after waiting and waiting, I get bored and hit CTRL-C. Fine, let's just run it then: ./wine ~/.wine/drive_c/Program\ Files/CarbonPoker/Poker.exe I get a "It appears the software has generated a bug. You can optionally submit it with comments about what you were doing at..." and as much as I'd like to do that with something like "doesn't work under wine" I'm going to press Cancel. Now we're in the login screen and it's asking me how I'd like to login, etc. So it looks like it is running! Just to check, I'm closing it and running it again.. yep, I'm in the login screen again. It's running. Hmm, I have a login for this game around here somewhere.. putting in my email address and pressing "Forgot Password" gives me this error: wine: Call from 0x7ee53f40 to unimplemented function urlmon.dll.HlinkSimpleNavigateToMoniker, aborting That isn't much help. And that crashed it, so running it again.. Sidenote: http://www.carbonpoker.com/pass.html is not secure.. ever heard of "denial of service" guys? Anyway, got some login details. Logged in, I see spam. Click on "Poker Tables". Up some come. Click on the first non-full one and "Join Table". And yes, I can join the table and sit down. I'm playing Poker. Yah. Trent

1 1

First song purchased on iTunes 7 with Wine...
by Dan Kegel Sept. 13, 2007

Sept. 13, 2007

For the record, I just bought "Autobahn" by Kraftwerk for $0.99 from the iTunes store using iTunes 7 on today's wine from git. It's playing now while I type this message. (OK, the music skips sometimes when I type or adjust the volume or move the mouse or scratch my nose... but hey, at least it's a start!) Thanks, Juan! - Dan

3 4

Re: [2/2] WGL: Sync pixmap after glFlush/glFinish, not before every GL call
by Alexandre Julliard Sept. 13, 2007

Sept. 13, 2007

Chris Robinson <chris.kcat(a)gmail.com> writes: > From 95214f29672566796564db30ab7f61a2462e4e50 Mon Sep 17 00:00:00 2001 > From: Chris Robinson <chris.kcat(a)gmail.com> > Date: Wed, 12 Sep 2007 06:42:05 -0700 > Subject: [PATCH] WGL: Sync pixmap after glFlush/glFinish, not before every GL call The goal is to sync the pixmap before we start drawing to it, so doing it at glFlush time is too late. -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: avifil32: Fix some memory leaks
by Paul Vriens Sept. 13, 2007

Sept. 13, 2007

Andrew Talbot wrote: > Changelog: > avifil32: Fix some memory leaks. > > > - if (mmioSeek(This->paf->hmmio, This->paf->dwNextFramePos, SEEK_SET) == -1) > - return AVIERR_FILEWRITE; > - if (mmioCreateChunk(This->paf->hmmio, &ck, 0) != S_OK) > - return AVIERR_FILEWRITE; > - if (mmioWrite(This->paf->hmmio, (HPSTR)lppc, ck.cksize) != ck.cksize) > - return AVIERR_FILEWRITE; > - if (mmioAscend(This->paf->hmmio, &ck, 0) != S_OK) > + if (mmioSeek(This->paf->hmmio, This->paf->dwNextFramePos, SEEK_SET) == -1 || > + mmioCreateChunk(This->paf->hmmio, &ck, 0) != S_OK || > + mmioWrite(This->paf->hmmio, (HPSTR)lppc, ck.cksize) != ck.cksize || > + mmioAscend(This->paf->hmmio, &ck, 0) != S_OK) > + { Doesn't this mean that every one of those 4 mmio-calls are executed? In the previous logic we would bail out after one failure. -- Cheers, Paul.

3 3

Re: wineprefixcreate: Make sure the path returned by GetTempPath() during the $WINEPREFIX creation will be writable.
by Alexandre Julliard Sept. 13, 2007

Sept. 13, 2007

Francois Gouget <fgouget(a)codeweavers.com> writes: > mshtml.dll (and probably many others) use GetTempPath() during the > self-registration process. But at that point the Windows environment > variables, and in particular $TMP and $TEMP, have not been set yet, > especially not in the currently running process. So when mshtml.dll > calls GetTempPath(), that function returns the current directory, which > obviously is no good if that directory is not writable (like '/' for > instance). So this patch makes 'c:\windows\temp' the current directory, > which we know will be writable (and the chdir() should have no other > side effect in this script). Actually GetTempPath is not supposed to use the current directory, I committed a fix. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: gdi32[2/4]: implement AddFontMemResourceEx (based on a patch of Byeong-Sik Jeon, helps with bug #5865) (resend)
by Alexandre Julliard Sept. 13, 2007

Sept. 13, 2007

"Mikolaj Zalewski" <mikolajz(a)google.com> writes: > diff --git a/dlls/gdi32/font.c b/dlls/gdi32/font.c > index c516736..efc044b 100644 > --- a/dlls/gdi32/font.c > +++ b/dlls/gdi32/font.c > @@ -3214,8 +3214,18 @@ BOOL WINAPI RemoveFontResourceW( LPCWSTR > */ > HANDLE WINAPI AddFontMemResourceEx( PVOID pbFont, DWORD cbFont, PVOID pdv, DWORD *pcFonts) > { > - FIXME("(%p,%08x,%p,%p): stub\n", pbFont, cbFont, pdv, pcFonts); > - return NULL; > + PVOID pFontCopy = HeapAlloc(GetProcessHeap(), 0, cbFont); > + HANDLE ret; > + > + TRACE("Copying %d bytes of data from %p to %p\n", cbFont, pbFont, pFontCopy); > + memcpy(pFontCopy, pbFont, cbFont); > + ret = WineEngAddFontMemResourceEx(pFontCopy, cbFont, pdv, pcFonts); > + if (ret == NULL) > + { > + TRACE("Font create failed - freeing %p\n", pFontCopy); > + HeapFree(GetProcessHeap(), 0, pFontCopy); > + } > + return ret; > } The memory copy should be done in WineEngAddFontMemResourceEx since that's where the data is managed. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [5/5] WineD3D: Pixel shader 1.x constants are clamped to [-1;1]
by Alexandre Julliard Sept. 13, 2007

Sept. 13, 2007

Stefan Dösinger <stefandoesinger(a)gmx.at> writes: > It is hard to judge if this patch will fix anything. A problem is that > clamping the constants on loading makes constants loading more expensive. > Luckilly though there aren't many ps 1.x constants(8 max). It is hard to say > if there are any games that use constants > 1.0, but incorrect constant > loading can have many bad effects. This one fails make test for me: ../../../tools/runtest -q -P wine -M d3d9.dll -T ../../.. -p d3d9_test.exe.so visual.c && touch visual.ok visual.c:3068: Test failed: quad 1 has color 00bfbf7f, expected 0x00bfbf80 visual.c:3070: Test failed: quad 2 has color 00bfbf7f, expected 0x00bfbf80 visual.c:3073: Test failed: quad 3 has color 00bfbf7f, expected 0x00bfbf80 visual.c:3077: Test failed: quad 4 has color 00bfbf7f, expected 0x00bfbf80 visual.c:3207: Test failed: quad 1 has color 007f7f00, expected 0x00808000 visual.c:3209: Test failed: quad 2 has color 007f7f00, expected 0x00808000 visual.c:3211: Test failed: quad 3 has color 007f7f00, expected 0x00808000 visual.c:3214: Test failed: quad 4 has color 00bfbf7f, expected 0x00bfbf80 make[2]: *** [visual.ok] Error 8 -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: WGL: Fix a major ATI regression
by Joerg Mayer Sept. 13, 2007

Sept. 13, 2007

On Thu, Sep 13, 2007 at 01:09:14AM +0200, Roderick Colenbrander wrote: > The change isn't a hack but more a workaround. I have submitted the bugreport to ATI but they likely won't fix it anytime soon. Well, if I understood the ATI development cycle correctly, even if they fixed it right away it will take another two releases (months) until it is released. ciao Joerg -- Joerg Mayer <jmayer(a)loplof.de> We are stuck with technology when what we really want is just stuff that works. Some say that should read Microsoft instead of technology.

1 0

Cvs update error.
by Anatoly Lyutin Sept. 13, 2007

Sept. 13, 2007

Hi all!! Today I tried to make cvs update and cvs extradites a error: [vostok(a)windsor wine]$ cvs update cvs [update aborted]: unrecognized auth response from cvs.winehq.org: cvs pserver: /home/wine/CVSROOT/config [8]: warning: this CVS does not support PreservePermissions What happens? -- Best regards Anatoly Lyutin.

1 0

WineHQ Downtime
by Jeremy Newman Sept. 12, 2007

Sept. 12, 2007

This is a reminder. WineHQ will be going offline for upgrades today at 14:00 CDT (19:00 GMT). I expect downtime to be no longer than one hour. Some services will appear to have sporadic outages even after the upgrade is complete. This will be due to changes in configs and upgraded versions of server software. I will post back to wine-devel when the upgrade is complete.

3 3

Re: WGL: Fix a major ATI regression
by Chris Robinson Sept. 12, 2007

Sept. 12, 2007

On Wednesday 12 September 2007 04:09:14 pm Roderick Colenbrander wrote: > + visinfo = pglXGetVisualFromFBConfig(display, cfgs[i]); > + /* We have found an offscreen rendering format when there is no > visualinfo :) */ > + if(visinfo->visualid == visualid) { > pglXGetFBConfigAttrib(display, cfgs[i], GLX_FBCONFIG_ID, > &fmt_id); > fbconfig = cfgs[i]; > } You're using visinfo before making sure it's valid. And I think that comment is a little misleading talking about offscreen formats when checking for the default visual. > + if(!visinfo) { > nOffscreenFormats++; > } > + XFree(visinfo); I don't think we can rely on being able to safely XFree a NULL pointer. None of the other code in opengl.c relies in that behavior.

1 0

Re: [3/5] WineD3D: Fix the reflection calculation in texm3x3(v)spec
by H. Verbeet Sept. 12, 2007

Sept. 12, 2007

On 12/09/2007, Stefan Dösinger <stefandoesinger(a)gmx.at> wrote: > ...so what the patch *actually* does in a completely obfuscated way, is getting rid of the assumption that the normal is normalized. IOW, - shader_addline(buffer, "tmp0.xyz = reflect(-tmp1.xyz, tmp0.xyz);\n"); + shader_addline(buffer, "tmp0.xyz = reflect(-tmp1.xyz, normalize(tmp0.xyz));\n"); would be just as effective. That also means that the change in the ARB implementation - /* Calculate reflection vector (Assume normal is normalized): RF = 2*(N.E)*N -E */ + /* Calculate reflection vector (Assume normal is normalized): + * See comment in the glsl implementation. is misleading at best.

1 0

Re: problem with silly CreateBitmap call
by Francois Gouget Sept. 12, 2007

Sept. 12, 2007

On Thu, 6 Sep 2007, Deun wrote: [...] > The size of the bitmap seems crazy... So I've made a patch to > fix this in bitmap.c by setting 1x1 size in case of stupid > input instead of failing with BadAlloc. However, since it works > fine on real windows, so I'm not sure about the behaviour that > the function should have. Seems like a good candidate for a conformance test. Call the API with the crazy values and see what comes out... -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ We are Pentium of Borg. You will be approximated. Division is futile.

4 3

iTunes 7
by Juan Lang Sept. 12, 2007

Sept. 12, 2007

Folks, I just sent a series of wintrust patches that gets iTunes 7 to start, at least for me. I expect some will get rejected due to collision with Francois's patch - I'll resynch and resend. But if you want to play around with iTunes, grab those patches and give them a whirl. Known issues: 1. Your screen will go black when it starts. This is something QuickTime is doing. You can drag windows around to force some repainting to work around it. 2. You'll see lots of fixmes on the console, especially: "fixme:setupapi:CMP_WaitNoPendingInstallEvents 0", "fixme:setupapi:SetupDiGetClassDevsW returning empty list", and "fixme:htmlhelp:HtmlHelpW HH case HH_PRETRANSLATEMESSAGE not handled." 3. The main window will be transparent (and basically unusable) if you don't have GL support compiled in. Have fun, --Juan

4 5

Re: [PATCH] user32: edit control should respond to ctrl + z (try 3)
by Dmitry Timoshkov Sept. 12, 2007

Sept. 12, 2007

"Lei Zhang" <thestig(a)google.com> wrote: > This patch lets users hit ctrl + Z in edit controls to trigger an > undo. The test case has been improved to check for both the return > values and the content of the edit control. In the tests still there is no actual edit contents/selection changes tracking after every message. -- Dmitry.

2 6

Re: WineD3D: PBO support 2/2 (attempt 2)
by Alexandre Julliard Sept. 12, 2007

Sept. 12, 2007

"Roderick Colenbrander" <thunderbird2k(a)gmx.net> writes: > This is an updated version of the patch. In the previous one I was > making a glBindBuffer call in a place without checking if PBOs were > around. Further the place didn't need the call anymore. This breaks the tests for me: ../../../tools/runtest -q -P wine -M d3d9.dll -T ../../.. -p d3d9_test.exe.so visual.c && touch visual.ok visual.c:2043: Test failed: quad 1 has color 0000ffff, expected 0x00FFFF00 visual.c:2045: Test failed: quad 2 has color 0000ffff, expected 0x0000000 visual.c:2047: Test failed: quad 3 has color 0000ffff, expected 0x0000FF00 make[2]: *** [visual.ok] Error 3 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [bugzilla] replace generic severity descriptions with wine specific ones
by Jeremy Newman Sept. 11, 2007

Sept. 11, 2007

Is this the final version of this patch? Or shall I wait for an updated one from you? On Tue, 2007-09-11 at 11:31 -0700, Lei Zhang wrote: > Hi, > > This is for bug 9110. Hopefully bug reporters will stop marking their > bugs as critical and then get upset when we set it to normal.

1 0

Re: wordpad: #1 Store bar states in the registry
by Alexandre Julliard Sept. 11, 2007

Sept. 11, 2007

Alexander Nicolaysen Sørnes <alex(a)thehandofagony.com> writes: > wordpad: #1 Store bar states in the registry This doesn't work well when resizing the window, the toolbars get overwritten by the edit area. -- Alexandre Julliard julliard(a)winehq.org

2 7

Re: A fix for a problem with older games
by Stefan Dösinger Sept. 11, 2007

Sept. 11, 2007

Am Sonntag, 9. September 2007 17:10:39 schrieb Martin: That patch is not correct Are you sure that this patch really causes a problem? If you are running the game in winedbg and get an exception there it is a symptom of the DIB management in winex11.drv. You can just use "pass" to pass the exception up to the next handler(winex11.drv). The exception never reaches the app. If you really get a crash there then This->resource.allocatedMemory is pointing to a bad location, most likely NULL. This is a problem at some other place and should be fixed elsewhere. This read is a hack to work around a bug in the open source ati driver which caused a hard system crash if an exception occured during glDrawPixels. You can remove hack as a whole if it really causes problems, but this will cause hard crashes for a wide range of radeon drivers(xorg 6.8 and 7.0 at least). I think it is fixed in the latest version, but I am not sure, I'd have to test that. Since the hack is confined in one single function and doesn't have any side effects I decided to keep it.

1 0

Re: a ddraw patch
by Stefan Dösinger Sept. 11, 2007

Sept. 11, 2007

Am Sonntag, 9. September 2007 17:11:06 schrieb Martin: That patch will obviously break the game the dwTextureCaps modification was added originally... Can you check which cap that game needs that is removed in those lines?

1 0

Re: [bugzilla] replace generic severity descriptions with wine specific ones
by James Hawkins Sept. 11, 2007

Sept. 11, 2007

On 9/11/07, Lei Zhang <thestig(a)google.com> wrote: > Hi, > > This is for bug 9110. Hopefully bug reporters will stop marking their > bugs as critical and then get upset when we set it to normal. > <th>Blocker</th> - <td>Blocks development and/or testing work</td> + <td>When Wine won't start at all for any application, and there's no work + around</td> I think you've made this too specific, especially for Blocker. For example, in the new description you've left out compile errors, which is a big part of what blocker represents. The original description is fine for blocker. <th>Major</th> - <td>major loss of function</td> + <td>For a problem that is affecting most applications you run in Wine</td> + </tr> Why not something like, major loss of functionality for a wide range of applications instead of totally rewriting the definition. -- James Hawkins

3 3

Re: [bugzilla] replace generic severity descriptions with wine specific ones
by Jesse Allen Sept. 11, 2007

Sept. 11, 2007

On 9/11/07, Lei Zhang <thestig(a)google.com> wrote: > Hi, > > This is for bug 9110. Hopefully bug reporters will stop marking their > bugs as critical and then get upset when we set it to normal. > > > > > How about for minor you still keep "where an easy workaround is present". That differentiates it from normal. Jesse

1 0

Fifa 07 - wined3d regression
by Jan de Mooij Sept. 11, 2007

Sept. 11, 2007

Hi, Fifa 07 used to work fine with Wine, but around version 0.9.40-0.9.41 the game hangs when I start a new match. The text "Loading..." is displayed, but there is no progressbar because one thread terminates. I attached the results of winedbg (backtrace and some messages before the crash). This is the result of git bisect: 9ba3f389f665aae45b6a8a6861f9621b020e5c25 is first bad commit commit 9ba3f389f665aae45b6a8a6861f9621b020e5c25 Author: Stefan Dösinger <stefandoesinger(a)gmx.at> Date: Sun Jul 8 23:33:48 2007 +0200 wined3d: Initialize surfaces at creation. This patch removes a check of "GL_SUPPORT(APPLE_CLIENT_STORAGE)". If I put that check in again, the game works like before. The if-condition evaluates to True now, so it allocates the memory. But it segfaults there I think. Please let me know if you need more information. Thanks in advance! Jan de Mooij

2 2

AppDB entries
by speedator＠gmx.de Sept. 11, 2007

Sept. 11, 2007

Hi, sometimes app-entries (like Acrobat Reader) are for series of applications and versions for the single applications. Sometimes there are app-entries(like Age of Empires) for single applications in a series and version-entries for patches, betas, demos etc.. And sometimes it is a mixture. What about making this more clear? Do we need entries for the whole series? I think it is important for overview. So what about a new structure like "application / version / derived from version(subversion)"? Julian

3 2

lostwages CVS moved into GIT
by Jeremy Newman Sept. 11, 2007

Sept. 11, 2007

We have switched our website sources (previously known as lostwages), over to git. >From this point on I would prefer if all patches came from the git tree. Refer to the git documentation on the website and wiki on how to prepare patches with git. The git module is called simply "website".

1 0