[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
weird bug(?) on segmentation fault
From: |
John W. Eaton |
Subject: |
weird bug(?) on segmentation fault |
Date: |
Fri, 8 Jun 2001 10:53:41 -0500 |
On 24-May-2001, E. Joshua Rigler <address@hidden> wrote:
| It's nice that Octave is fairly robust, and can attempt to save a
| session even on a segmentation fault, but I'm encountering a strange
| problem. I haven't attempted to track the cause of the seg fault
| itself, but when it occurs, octave attempts to save the session to
| "octave-core". It does this, then doesn't exit. This is bad, because
| then every single attempted command after that causes another seg fault,
| which then causes Octave to try to save the session...even when the
| command is "exit" or "quit". This was an issue when I had a session
| running with several hundred MB of data, which takes quite a bit of time
| to save to disk. Ideally there wouldn't be a seg fault, but why isn't
| Octave exitting after it saves the session?
|
| I'm using 2.1.33. Has anyone else esperienced something similar?
When Octave encounters SIGSEGV, it calls the function
my_friendly_exit, which is defined in src/sighandlers.cc. This
function tries to do a lot of stuff, including calling other
functions, which may not do the right thing if Octave is in a really
bad state. But it tries to be smart about not looping, so that if a
series of segfaults occur, it should just abort.
In any case, can you please submit a complete bug report about this,
or at least tell us exactly what sequence of error messages you see?
It would also be good to submit a complete report about the crash you
are encountering, so that the real cause of that problem might also be
fixed.
Thanks,
jwe
-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.
Octave's home on the web: http://www.octave.org
How to fund new projects: http://www.octave.org/funding.html
Subscription information: http://www.octave.org/archive.html
-------------------------------------------------------------
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- weird bug(?) on segmentation fault,
John W. Eaton <=