RE: Octave memory leak??

[Ted Harding]

On 02-Sep-98 John W. Eaton wrote:
> On  2-Sep-1998, (Ted Harding) <address@hidden> wrote:
> 
>| Does octace-2.0.11 have a memory leak?
>|
>| I was running a heavily iterative computation and over time the swap
>| usage built up to about 180MB (out of 256MB). There was no obvious
>| reason
>| why this should be so, since no data structures involved were supposed
>| to
>| get larger or more numerous, and there were no recursive calls.
> 
> Does your code call functions defined in .m files?  If so, you may be
> running into a bug that was actually in the kpathsearch library that's
> used for looking up files in the LOADPATH.  I don't remember the exact
> sequence of events, but I think I fixed the bug in an older version of
> the kpathsearch library, then it reappeared when I upgraded to a newer
> version of the library, but it was fixed for 2.0.12 (I think -- it is
> definitely fixed for 2.0.13).

Yes, there were several .m files. Your explanation is therefore probably
the correct one. Thanks very much for the response -- it looks as though
the cure is to upgrade a couple of notches!

If that was the case, I'm curious about the mechanism -- does it mean
that new copies of the m-files are brought into memory every iteration,
leaving the old ones there, orphans but still on the books, when this
happens?

Best wishes,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <address@hidden>
Date: 02-Sep-98                                       Time: 11:06:33
--------------------------------------------------------------------