[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-gnubg] core dump on analyze of partial match
From: |
Jim Segrave |
Subject: |
Re: [Bug-gnubg] core dump on analyze of partial match |
Date: |
Sat, 16 Aug 2003 23:44:39 +0200 |
User-agent: |
Mutt/1.4.1i |
On Sun 17 Aug 2003 (09:25 +1200), Joseph Heled wrote:
>
> I saved a match in the middle. Then tried 'analyze match' and got a core
> dump.
>
> (gdb) r -t ~/gnubg/sn4/2003-08-15-gnu-jo-7b.sgf
>
> [stuff deleted]
>
> (No game) analyze match
> Analysing match; move: 75/79
> Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 8192 (LWP 29757)]
> 0x080a1a9c in getME (nScore0=1, nScore1=1, nMatchTo=134724761,
> fPlayer=-1073750440, nPoints=5, fWhoWins=141152000, fCrawford=5,
> aafMET=0x870d100, aafMETPostCrawford=0x869cf00) at matchequity.c:1807
This looks like using memory which has been freed. It's going to try
to look for the MET entry for 134million away to 134million and one
away, which is, not surprisingly, outside the bounds of most any MET.
A 5 cube strikes me as unlikely as well. A stack trace might help
locate which bit of memory has been freed and re-used.
--
Jim Segrave address@hidden