Re: Segmentation Fault with Octave MXE under Windows

[Daniel J Sebald]

On 07/10/2014 07:47 AM, ijourneaux wrote:
> Since I am not getting an indication of a SegFault when running the line
> below in array.h
>
> fprintf(stdout, "Pointer OK.\n");
>
> my thought is that compiler optimization may be causing the issue.

Can't rule it out without more testing, but it's not likely.  Poor 
optimization would show up in many places.  Segfaults and corrupted 
memory bugs are very deceptive sometimes.  Because of the delayed manner 
in which Octave object memory is freed, it seems the problem isn't 
materializing until well after the source of the problem.

I think Carnë's theory is closer to the source right now, given that 
imformats() alone creates the same segfault.  However, I'll follow up 
with another post.

Dan