http://bugs.winehq.org/show_bug.cgi?id=9479
Summary: WOW crashes when quitting Product: Wine Version: 0.9.44. Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: wine-misc AssignedTo: wine-bugs@winehq.org ReportedBy: marcw@onlymooo.com
Upgraded to 0.9.44 today. Scott Richies Feisty build. World of Warcraft locks up - freezes - when exiting game. Is remedied by a kill -9. Reverted back to 0.9.43 and it's now back to normal.
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #1 from Austin English austinenglish@gmail.com 2007-08-27 07:37:41 --- Please retry with a clean .wine directory. If the problem still occurs, please run a regression test:
http://wiki.winehq.org/RegressionTesting
http://bugs.winehq.org/show_bug.cgi?id=9479
Juan Lang juan_lang@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |therwald@laposte.net
--- Comment #2 from Juan Lang juan_lang@yahoo.com 2007-08-27 12:09:58 --- *** Bug 9489 has been marked as a duplicate of this bug. ***
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #3 from Marc Williams marcw@onlymooo.com 2007-08-27 14:08:44 --- (In reply to comment #1)
Please retry with a clean .wine directory. If the problem still occurs, please run a regression test:
I'm sorry but I won't be able to regression test any time soon. This is an instance of Wine on my son's computer in a different town.
As far as a clean .wine dir... I'm not sure that would be practical even if I was able to test. That's where his WoW is installed (c:\Program Files\Warcraft or something like that) and a clean .wine dir that would mean performing a total reinstall of WoW. Considering how long it took to install it the first time, what with all of the 5 CDs, the 2 hours of patching, etc, it seems like that would be no fun at all do do again.
If I get to his computer in the next couple weeks and see no one else has been able to regression test, I'll tackle it at that time.
http://bugs.winehq.org/show_bug.cgi?id=9479
Eric Hokanson hokanson@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |hokanson@gmail.com
--- Comment #4 from Eric Hokanson hokanson@gmail.com 2007-08-27 18:28:47 --- I also see this bug. Running a git bisect shows this patch as the offending one:
2f3805a3bd27305e64e92bc3b828ab272a35a6ac is first bad commit commit 2f3805a3bd27305e64e92bc3b828ab272a35a6ac Author: Misha Koshelev mk144210@bcm.edu Date: Thu Aug 23 00:06:26 2007 -0500
wininet: Send INTERNET_STATUS_HANDLE_CLOSING when object is being destroyed as native does.
:040000 040000 8ab293e8a5eddc97fa4145a6c3f4468753aa180b 762157d39b220736086c05ec42e7840ce04dc5bc M dlls
http://bugs.winehq.org/show_bug.cgi?id=9479
Jesse Allen the3dfxdude@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression
http://bugs.winehq.org/show_bug.cgi?id=9479
Misha Koshelev mk144210@bcm.tmc.edu changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |mk144210@bcm.tmc.edu
--- Comment #5 from Misha Koshelev mk144210@bcm.tmc.edu 2007-08-27 19:31:14 --- Hmm... I do not have this program or access to it (I don't think). Perhaps we could start with a WINEDEBUG=+wininet log please?
Thank you Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #6 from Eric Hokanson hokanson@gmail.com 2007-08-27 23:37:23 --- Created an attachment (id=7847) --> (http://bugs.winehq.org/attachment.cgi?id=7847) WINEDEBUG=+wininet log
http://bugs.winehq.org/show_bug.cgi?id=9479
Jesse Allen the3dfxdude@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #7847|application/octet-stream |text/plain mime type| |
http://bugs.winehq.org/show_bug.cgi?id=9479
Tim Ryan timbers2k@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |timbers2k@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #7 from Misha Koshelev mk144210@bcm.tmc.edu 2007-08-28 20:14:26 --- Ok, thank you very much for the log. I believe I know what the problem is; just in case, to confirm if someone could post a WINEDEBUG=+server,+wininet log would be great.
Basically, the problem is this: on native, CloseHandle and InternetCloseHandle have equivalent actions. However, on wine CloseHandle is handled in wineserver (passed on from process -> ntdll -> wineserver), whereas InternetCloseHandle is handled exclusively in wininet and in fact internet handles in wine are only supported within this framework.
A call to CloseHandle on a wininet handle would always have failed under wine, but what my patch did was to change to the native behavior of sending the INTERNET_STATUS_HANDLE_CLOSING when the handle was actually being freed, which is done on native (and is required for urlmon to not have a potential race on freeing some objects). However, what this means is that the CloseHandle call never "registers" in the wininet handle references and thus the handle's reference count never gets to 0, thus no INTERNET_STATUS_HANDLE_CLOSING is now sent.
Ideally, the fix would be to handle wininet handles in the wineserver, but from what I understand this would be a fairly significant task and would probably require Alexandre's guidance if not help, and I am not sure I have time for a project so extensive. However, I am open to any suggestions especially if someone can think of a simpler "native" fix that is acceptable to the wine tree.
Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #8 from Misha Koshelev mk144210@bcm.tmc.edu 2007-08-29 00:51:31 --- Oops, please scratch my last comment, it is incorrect.
I just tested and apparently an InternetCloseHandle on just the handle returned by InternetConnect (not calling the one on HttpSendRequest at all) sends the appropriate INTERNET_STATUS_HANDLE_CLOSING, so has nothing to do with handles in server.
Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #9 from Misha Koshelev mk144210@bcm.tmc.edu 2007-08-29 20:49:13 --- Just wanted to update I just mocked something up for myself that will keep track of handle parents/children so we can properly free children from InternetCloseHandle like native does and it looks like it will be a pretty straightforward patch actually. I will post something here once I have a more complete working version so you can let me know if it fixes the regression.
Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #10 from Misha Koshelev mk144210@bcm.tmc.edu 2007-08-30 20:08:47 --- Created an attachment (id=7898) --> (http://bugs.winehq.org/attachment.cgi?id=7898) wininet: Generalize parent/child handles and free all child handles on WININET_FreeHandle as native.
Proposed fix. Please let me know whether or not it fixes the problem (if not please attach a new WINEDEBUG=+wininet log with the patch applied). Also if dlls/wininet/tests/ftp test works for anyone (it doesn't really for me) pls try it with this patch and let me know if you can. Thanks
Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #11 from Eric Dégenètais therwald@laposte.net 2007-08-31 05:15:10 --- I just tested WoW with Micha's patch. Played for a few seconds, then quitted. Everything went OK. As far as I am concerned, this patch fixes the bug. Thanks!
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #12 from Misha Koshelev mk144210@bcm.tmc.edu 2007-08-31 07:43:43 --- Great! Thanks Eric. Anyone else confirm? Thanks, Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
Alan Jackson ajackson@bcs.org.uk changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ajackson@bcs.org.uk
--- Comment #13 from Alan Jackson ajackson@bcs.org.uk 2007-08-31 08:39:26 --- I can confirm it working as well, thank you.
http://bugs.winehq.org/show_bug.cgi?id=9479
Alexander Nicolaysen Sørnes alex@thehandofagony.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |alex@thehandofagony.com Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #14 from Alexander Nicolaysen Sørnes alex@thehandofagony.com 2007-08-31 18:06:02 --- Confirming.
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #15 from Eric Hokanson hokanson@gmail.com 2007-08-31 19:11:01 --- Looks good. WoW exits cleanly now with your patch applied. Thanks Misha.
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #16 from Misha Koshelev mk144210@bcm.tmc.edu 2007-08-31 20:04:09 --- Ok great. I will submit a slightly modified version (with a test) to wine-patches, although looks like Alexandre's on vacation for a week and then I'll be gone for 9 days if any revisions are needed.
Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #17 from Misha Koshelev mk144210@bcm.tmc.edu 2007-09-02 17:21:39 --- Patches submitted:
http://www.winehq.org/pipermail/wine-patches/2007-August/043470.html http://www.winehq.org/pipermail/wine-patches/2007-August/043471.html http://www.winehq.org/pipermail/wine-patches/2007-August/043472.html
Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
Bleeter Yaluser winehq-bugzilla@six-by-nine.com.au changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winehq-bugzilla@six-by- | |nine.com.au
--- Comment #18 from Bleeter Yaluser winehq-bugzilla@six-by-nine.com.au 2007-09-04 19:24:05 --- Patch works for 2.2 WoW PTR as well
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #19 from Adam schafer0@purdue.edu 2007-09-04 20:18:12 --- How does one go about installing that patch?
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #20 from David Hoepelman david@dtechnology.nl 2007-09-09 06:16:01 --- @Adam You have to get the wine source, commit the patch and then build it. If you don't know how, you probally better revert back to 0.9.43 and wait for the patch to be included in the wine releases (probaly will in 0.9.45 or maybe 0.9.46, the first is only days away (if I follow the 14-day release scheme it should have been released two days ago), the latter will probaly be released in 0.9.46.
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #21 from Bleeter Yaluser winehq-bugzilla@bleeter.id.au 2007-09-14 21:16:05 --- Misha, Any ideas why these patches didn't make the 0.9.45 cut? Do they need to be worked more?
Cheers.
http://bugs.winehq.org/show_bug.cgi?id=9479
Polarina gabrielp@simnet.is changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |gabrielp@simnet.is
--- Comment #22 from Polarina gabrielp@simnet.is 2007-09-15 07:51:37 --- (In reply to comment #21)
Misha, Any ideas why these patches didn't make the 0.9.45 cut? Do they need to be worked more?
Cheers.
I think it's possible that the current release were synchronized from a week old GIT trunk. Probably to maintain the "2-week" release rule.
But of course, just a random guess.
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #23 from Alexander Nicolaysen Sørnes alex@thehandofagony.com 2007-09-15 08:19:38 --- And as with many other random guesses, it is wrong. :) New patches were committed all the time.
http://bugs.winehq.org/show_bug.cgi?id=9479
Riccardo Setti giskard@autistici.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |giskard@autistici.org
--- Comment #24 from Riccardo Setti giskard@autistici.org 2007-09-17 05:07:33 --- talking on #winehq people told me to tell you that you have to figure out why that patch was not accepted :)
http://bugs.winehq.org/show_bug.cgi?id=9479
Louis Lenders xerox_xerox2000@yahoo.co.uk changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |xerox_xerox2000@yahoo.co.uk
--- Comment #25 from Louis Lenders xerox_xerox2000@yahoo.co.uk 2007-09-17 10:09:21 --- As far as i know, Misha said somehwhere, that he's out of the country (holiday?) for about 2 weeks, and send the patches in, once he gets back. So you just have to be a little patient ;)
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #26 from Riccardo Setti giskard@autistici.org 2007-09-17 10:36:48 --- (In reply to comment #25)
As far as i know, Misha said somehwhere, that he's out of the country (holiday?) for about 2 weeks, and send the patches in, once he gets back. So you just have to be a little patient ;)
ok. not a big problem. thank you.
http://bugs.winehq.org/show_bug.cgi?id=9479
Todd Mokros tmokros@tmokros.net changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |tmokros@tmokros.net
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #27 from Misha Koshelev mk144210@bcm.tmc.edu 2007-09-21 07:46:57 --- Greetings, patch is in git as of this AM:
http://www.winehq.org/pipermail/wine-cvs/2007-September/036373.html
Pls let me know if this is fixed in git now.
Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #28 from Bleeter Yaluser winehq-bugzilla@bleeter.id.au 2007-09-21 19:36:24 --- Misha, Works for me against 2.1.3 (Live). I have no Test Realm to check against, but there's one coming up soon. That will be for 2.2.2 version, although 2.2.0 will be released live on Tuesday. I wouldn't like to state with any certainty that this is completely fixed until I can test on the upcoming versions. I know that this is probably an irrelevant test, but I'd hate to claim everything's fine, only to have it break in a couple of days time (for whatever crazy reason it might).
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #29 from Misha Koshelev mk144210@bcm.tmc.edu 2007-09-21 19:46:58 --- Glad to hear it works. When you have checked it thoroughly pls let us know and then close the bug if everything is fixed.
Thanks Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #30 from Riccardo Setti giskard@autistici.org 2007-09-22 04:04:50 --- IT WORKS :)
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #31 from Bleeter Yaluser winehq-bugzilla@bleeter.id.au 2007-09-22 07:33:04 --- Works fine in 2.2.2PTR. Awaiting 2.2.0 to go live for testing against that, expected this Tuesday.
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #32 from Bleeter Yaluser winehq-bugzilla@bleeter.id.au 2007-09-25 07:08:19 --- Works fine on 2.2.0 Live (servers of which aren't operational yet, but as this is a client-only side issue, the short test [start game, exit game] that I did should be enough!
Nice work, Misha! :)
http://bugs.winehq.org/show_bug.cgi?id=9479
--- Comment #33 from Misha Koshelev mk144210@bcm.tmc.edu 2007-09-25 07:38:43 --- Glad to help. I guess someone should close the bug then.
Misha
http://bugs.winehq.org/show_bug.cgi?id=9479
Alexander Nicolaysen Sørnes alex@thehandofagony.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #34 from Alexander Nicolaysen Sørnes alex@thehandofagony.com 2007-09-25 07:52:39 --- Guess so.
http://bugs.winehq.org/show_bug.cgi?id=9479
Dan Kegel dank@kegel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #35 from Dan Kegel dank@kegel.com 2008-01-28 05:20:40 --- Closing all RESOLVED FIXED bugs more than three months old.
-
wine-bugs＠winehq.org