[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Octave/GUI tests
From: |
Israel Herraiz |
Subject: |
Re: Octave/GUI tests |
Date: |
Sun, 27 May 2012 19:07:48 -0700 |
User-agent: |
Sup/git |
Excerpts from Jacob Dawid's message of Sun May 27 15:47:52 -0700 2012:
> I have replaced the octave exit callback to handle an exit. There are two
> sources for bugs: Either octave doesn't call octave exit at the very end,
> or the GUI tries to interact with octave after the exit. Before we actually
> fix something, I think it is important to determine of which kind a bug is
> in order to be able to fis it appropriately.
I have done some tests with the latest commits, just in case it helps.
I have done all the tests with Octave 3.6.1, I modified line 150 in
WorkspaceModel.cpp to make it compile with Octave 3.6.1. All the tests
are in Archlinux.
Test 1 (no segfault):
a = 2
exit (from terminal)
Test 2 (segfault):
a = 2
(close main window)
Test 3 (no segfault):
(close main window without executing any command)
Test 4 (no segfault):
exit
Test 5 (no segfault):
x = [1 2 3]
plot(x) [with gnuplot]
exit
Test 6 (no segfault):
x = [1 2 3]
plot(x) [again gnuplot]
(close main window)
So, in summary, I can confirm that I get segfaults only exiting from
the GUI (closing main window or using exit from the menu). No
segfaults at all exiting from the terminal. But it only happens when I
have run a command. Interestingly (strangely), the crash does not
happen if the last command is a plot, and the gnuplot window is still
open.
Cheers,
Israel
- Octave/GUI tests, Michael Goffioul, 2012/05/26
- Re: Octave/GUI tests, Israel Herraiz, 2012/05/27
- Re: Octave/GUI tests, Jacob Dawid, 2012/05/27
- Re: Octave/GUI tests, Israel Herraiz, 2012/05/27
- Re: Octave/GUI tests,
Israel Herraiz <=
- Re: Octave/GUI tests, Doug Stewart, 2012/05/27
- Re: Octave/GUI tests, Israel Herraiz, 2012/05/27
- Re: Octave/GUI tests, Jordi GutiƩrrez Hermoso, 2012/05/28
- Re: Octave/GUI tests, Israel Herraiz, 2012/05/28