| Summary: | Solver frequently crashes while playing Gypsy | ||
|---|---|---|---|
| Product: | [Applications] kpat | Reporter: | Sergei Andreev <seajey.serg> |
| Component: | solver | Assignee: | Stephan Kulow <coolo> |
| Status: | RESOLVED FIXED | ||
| Severity: | crash | CC: | armelle.sw, christian, db0313, hendrik, jan.backefeldt, kdandfd, kde-games-bugs-null, kvikende, leggis, martin_hummel, pacochote, pauldrequest, sromeroan, stuart.kirby20 |
| Priority: | HI | ||
| Version First Reported In: | unspecified | ||
| Target Milestone: | --- | ||
| Platform: | Chakra | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed In: | ||
| Sentry Crash Report: | |||
| Attachments: | Saved game of Stuart's problematic deal | ||
|
Description
Sergei Andreev
2011-10-02 18:45:56 UTC
I have also seen this bug while playing Gypsy, which results in KPAT closing down. It seems to occur when I move to pick up a new card or am doing nothing having just moved a card - I have seen it about a dozen times. I suggest that it is some change to the solver. I note that there are no problems with Freecell, Forty & Eight, Spider or Simple Simon. I have not played the other games. You can reproduce this by doing the following: Choose "new numbered deal", gypsy, 102194206 7 of spades to 8 of hearts, 8H to 9C, 5C to 6D, 6D to 7C ... boom Alternatively: 7 of spades to 8 of hearts, 8H to 9C, 5C to 6D, 6D to 7S, 2S to space, 6H to 7C ... boom *** Bug 285403 has been marked as a duplicate of this bug. *** Application: kpat (3.6) Distribution: "Ubuntu 11.10" I have the same problem. This is what I got in the terminal: santiago@santiago-desktop:~$ kpat kpat(3540)/KSharedDataCache: Unable to free up memory for "1920-941-background" kpat(3540)/KSharedDataCache: Unable to free up memory for "1920-995-background" KCrash: Application 'kpat' crashing... KCrash: Attempting to start /usr/lib/kde4/libexec/drkonqi from kdeinit sock_file=/home/santiago/.kde/socket-santiago-desktop/kdeinit4__0 Warning: connect() failed: : No existe el archivo o el directorio KCrash: Attempting to start /usr/lib/kde4/libexec/drkonqi directly QSocketNotifier: Invalid socket 14 and type 'Read', disabling... kpat: Fatal IO error 9 (Descriptor de archivo erróneo) on X server :0.0. Unable to start Dr. Konqi santiago@santiago-desktop:~$ drkonqi(3568): The specified process does not exist. santiago@santiago-desktop:~$ Confirm this bug on Kubuntu 11.10: :~$ kpat kpat: Fatal IO error: client killed KCrash: Application 'kpat' crashing... KCrash: Attempting to start /usr/lib/kde4/libexec/drkonqi from kdeinit sock_file=/home/belomir/.kde/socket-belomir-desktop/kdeinit4__1 Unable to start Dr. Konqi :~$ El 07/11/11 16:36, Belomir escribió: > https://bugs.kde.org/show_bug.cgi?id=283191 > > > Belomir<leggis@mail.ru> changed: > > What |Removed |Added > ---------------------------------------------------------------------------- > CC| |leggis@mail.ru > > > > > --- Comment #5 from Belomir<leggis mail ru> 2011-11-07 15:36:44 --- > Confirm this bug on Kubuntu 11.10: > > :~$ kpat > kpat: Fatal IO error: client killed > KCrash: Application 'kpat' crashing... > KCrash: Attempting to start /usr/lib/kde4/libexec/drkonqi from kdeinit > sock_file=/home/belomir/.kde/socket-belomir-desktop/kdeinit4__1 > Unable to start Dr. Konqi > :~$ > Sorry, I can't . I'm not using Kubuntu 11.10 My system is Ubuntu 11.10 wtih gnome 3.2 and I've been using Kpat for a couple of years since Ubuntu 9.4 and it always worked fine. Created attachment 65374 [details]
Saved game of Stuart's problematic deal
Just move the 6 of diamonds to the 7 of clubs and instant segfault.
*** Bug 286137 has been marked as a duplicate of this bug. *** If I run the game from a game console terminal does not fail *** Bug 286319 has been marked as a duplicate of this bug. *** (In reply to comment #9) > If I run the game from a game console terminal does not fail Using Stuart's reproduction steps listed above, I can confirm that the crash happens regardless of whether the game is running in a terminal or not. I should also mention that for those of you encountering this crash, until we have a real fix, it can be avoided by disabling the solver. (Menu -> Settings -> Enable Solver) ==9401== Thread 2: ==9401== Invalid read of size 8 ==9401== at 0x447DE6: MemoryManager::new_from_block(unsigned long) (memory.cpp:164) ==9401== by 0x448DE4: Solver::pack_position() (patsolve.cpp:369) ==9401== by 0x44A4D6: Solver::insert(unsigned int*, int, TREE**) (patsolve.cpp:1046) ==9401== by 0x44A599: Solver::new_position(POSITION*, MOVE*) (patsolve.cpp:1077) ==9401== by 0x44994A: Solver::solve(POSITION*) (patsolve.cpp:739) ==9401== by 0x4496DB: Solver::doit() (patsolve.cpp:664) ==9401== by 0x44A1A0: Solver::patsolve(int, bool) (patsolve.cpp:958) ==9401== by 0x4277F7: SolverThread::run() (dealer.cpp:110) ==9401== by 0x6EB8024: QThreadPrivate::start(void*) (qthread_unix.cpp:331) ==9401== by 0x72D7F04: start_thread (pthread_create.c:301) ==9401== by 0x8C304BC: clone (clone.S:115) ==9401== Address 0x20700000010 is not stack'd, malloc'd or (recently) free'd SVN commit 1264171 by coolo: don't overfill the possible array BUG: 283191 M +11 -5 gypsysolver.cpp M +2 -0 patsolve.cpp WebSVN link: http://websvn.kde.org/?view=rev&revision=1264171 SVN commit 1264189 by coates: Backport of commit 1264171. don't overfill the possible array CCBUG:283191 FIXED-IN:4.7.4 M +11 -5 gypsysolver.cpp M +2 -0 patsolve.cpp WebSVN link: http://websvn.kde.org/?view=rev&revision=1264189 *** Bug 287799 has been marked as a duplicate of this bug. *** *** Bug 288945 has been marked as a duplicate of this bug. *** *** Bug 289581 has been marked as a duplicate of this bug. *** *** Bug 288551 has been marked as a duplicate of this bug. *** Playing Gypsy on open suse allways crashes *** Bug 291311 has been marked as a duplicate of this bug. *** *** Bug 290452 has been marked as a duplicate of this bug. *** *** Bug 303182 has been marked as a duplicate of this bug. *** |