May 2007 - Wine-Bugs - list.winehq.org

[Bug 8263] InnerSpace crashes when trying to launch
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=8263 ------- Additional Comments From tvirus.nemesis(a)gmail.com 2007-03-05 05:44 ------- All that work was done on my laptop, so it may be a hardware issue. This is the issue that I'm having on my desktop, and most the others from our community have had when running Innerspace on Wine. The app itself normally launches fine, can open up its different windows, patch it, etc. but when we try to launch a game through it (ex. World of Warcraft), which is what it is designed for, it crashes. Enverex has told me that it's a stack overflow error. Doesn't matter what compatability mode it's in. The application's crash manager is pointing to ntdll.dll. Wine isn't saying much at all. I'll go ahead and attach everything. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0

[Bug 5568] toolbar is too wide in js latvija mapping application
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=5568 ------- Additional Comments From rich(a)hq.vsaa.lv 2007-03-05 05:35 ------- it seems that WINEDEBUG=+toolbar is not valid, so there was no additional output. also, the problem is still there with 0.9.36 -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0

[Bug 7929] C&C 3 network does not work
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=7929 ------- Additional Comments From d13f00l(a)gmail.com 2007-03-05 04:55 ------- As a side note, I believe this app is just doing a recv() in a loop every once in a while, like on some sort of timer. It doesn't seem to be using select(), so recv() returning -1 would be normal when there's no data to be read even once we do get this resolved. It was ALMOST working when I pretty much forced wine's implementation of gethostname() to return a fake hostname I put in /etc/hosts that resolves to 0.0.0.0 It was showing other people in the server list, and actually showing created games, but it'd show 2 of my own user names in the chat room since it was recvmsg()ing it's own broadcasts. Obviously that was just a hack and me playing around with debugging the problems. Your suggestion to modify gethostbyname() is much better :) -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0

[Bug 7929] C&C 3 network does not work
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=7929 ------- Additional Comments From blin(a)gmx.net 2007-03-05 04:36 ------- Uh, never mind my comments about the WS_recv call. Dylan pointed me at the fact that the recv fails (which I noticed) and that causes the address structure to be filled with random junk (I assumed this to be the other way around). -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0

[Bug 7892] iTunes start-up
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=7892 ------- Additional Comments From focht(a)gmx.net 2007-03-05 04:34 ------- Hello, good to see some progress on the wintrust issue :) Just a few annotiations to your inital provider loader impl (saw a patch on mailing list) You load providers but leak the modules (handles) stating "no way to unload providers". That may be api-wise true but as good developer there is always another way. There exist scenarios where each provider function on specific action guid can be implemented by a separate provider and where all provider functions are implemented by one provider (e.g. wintrust). Manage the data in a structure which holds action guid ( = key), provider function pointers and provider dll handles ("max" case = all stuff implemented by same provider). Make a linked list with this, search criteria is the action guid. Use some lock to guard list (e.g. critical section) and prevent race issues. On first time WintrustLoadFunctionPointers() you just build up the managed list "cache" from registry. Create new entries and fill the data, action guid, module handles (LoadLibrary), function pointers (GetProcAddress). Beware of the "wintrust does it all by itself" case (compare name "wintrust.dll"). In that case just store the HINSTANCE supplied by wintrust DllMain and function pointers to internal handlers. On subsequent WintrustLoadFunctionPointers() calls, you just search the list for action guid and return the list entry data, e.g. assign the function pointers to CRYPT_PROVIDER_FUNCTIONS struct. For a good design the data structures must be freed. Appropriate place would be DllMain( DLL_PROCESS_DETACH) Regards -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0

[Bug 7929] C&C 3 network does not work
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=7929 ------- Additional Comments From blin(a)gmx.net 2007-03-05 03:53 ------- Using gethostbyname() to get the local IP is a bad idea on any host that has more than one interface. Linux always regards lo as a valid interface, windows only uses 127.0.0.1 if there's no other IP address assigned. I'm currently working on a patch for that. As comment #4 says, there's also a problem with broadcast messages. Another thing that trips me up when looking at the trace is trace:winsock:WS2_recv fd 87, iovec 0x1f3a23d0, count 1 addr { family 4096, address 44.227.75.124, port 0 }, len 0xeaed34, flags 0 family 4096? IP 44.227.75.124? Are you at the UCSD? Have you been playing with someone from there? This looks similar to bug #7299. Oh well, nothing we can do but starting to fix some of the bugs. :) -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0

[Bug 8266] Future Pinball : crash when playing a table (regression since 0.9.33)
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=8266 ------- Additional Comments From o.oudin(a)free.fr 2007-03-05 03:28 ------- I didn't mention steps to reproduce the crash : after starting Future Pinball, click on File->New Table, then click on the Play button -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0

[Bug 8266] Future Pinball : crash when playing a table (regression since 0.9.33)
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=8266 ------- Additional Comments From o.oudin(a)free.fr 2007-03-05 03:26 ------- Created an attachment (id=6101) --> (http://bugs.winehq.org/attachment.cgi?id=6101&action=view) wine 0.9.36 trace output with channel +loaddll,+opengl -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0

[Bug 8266] New: Future Pinball : crash when playing a table (regression since 0.9.33)
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=8266 Summary: Future Pinball : crash when playing a table (regression since 0.9.33) Product: Wine Version: 0.9.36. Platform: Other URL: http://www.futurepinball.com/download.html OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: wine-opengl AssignedTo: wine-bugs(a)winehq.org ReportedBy: o.oudin(a)free.fr Future pinball crash when trying to play a table ( It's a regression since 0.9.33) With 0.9.32, I was able to play tables with the config below Since 0.9.33, wine crash with a debugger trace (logs with 0.9.36 in attachement) I'm on Ubuntu 6.10, x86, with nvidia driver installed (glxinfo show direct rendering : Yes, glxgears works) I use this wine config : - crypt32.dll, rsaenh.dll, mfc42.dll, vbscript.dll copied from a windows xp partition (crypt32.dll and rsaenh.dll are needed because Future Pinball store tables encrypted, mfc42.dll and vbscript.dll because Future Pinball use VBscript to script table's events) - DCOM98.exe installed (builtin ole failed getting object classid {b54f3741-5b07-11cf-a4b0-00aa004a55e8}) - Disable sound (in 0.9.32, sound was crappy), set wine compatibility to windows xp - run with WINEDEBUG=+loaddll,+opengl WINEDLLOVERRIDES=crypt32,rsaenh,ole32,oleaut32,rpcrt4=n wine Future Pinball.exe I assign to wine-opengl, but I really don't know if this is the component crashing. I'm available to do more testing with others debug channel, but I don't know which could be usefull. If needed, I can reinstall also wine 0.9.32 and provide log with channels you want. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0

[Bug 8265] file name too long
by Wine Bugs May 3, 2007

May 3, 2007

http://bugs.winehq.org/show_bug.cgi?id=8265 ------- Additional Comments From dmitry(a)codeweavers.com 2007-03-05 03:16 ------- Where do you execute this command? cmd.exe prompt, bat file, anywehre else? Or is this a makefile rule? -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

1 0