Most likely place for a program to be leaking memory?
From: Christopher Neff (caneff_at_gmail.com)
Date: 03/22/05
- Next message: sali0090_at_tc.umn.edu: "Re: Can I solve that with Runge-Kutta?"
- Previous message: sali0090_at_tc.umn.edu: "Re: problems with recursive algorithm for dynamic time warping"
- Next in thread: Brooks Moses: "Re: Most likely place for a program to be leaking memory?"
- Reply: Brooks Moses: "Re: Most likely place for a program to be leaking memory?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: 21 Mar 2005 21:56:23 -0800
I am trying to get this code finished by tonight as the results are due
at midnight. And it works, as long as it doesn't run very long. After
a certain number of times through one of my main loops I get (in
ifort):
Insufficient memory to allocate Fortran RTL message buffer, message #41
= hex 00000029.
Which I am assuming is a memory leak. I've never had problems with the
largest part of this code (a 1-d multigrid solver) in terms of memory,
and all I've added on is a 1-d Heat Equation solver using
Crank-Nicholson, which really shouldn't have any affect on memory
allocation at all.
I tried running g95 to see if it would tell me, and I get a
segmentation fault much earlier than it quits in ifort. I will
eventually try and track down the problem (possible bug?), but I don't
have the time to do that right now.
Any tips as to where the most usual memory leaks occur in a program?
- Next message: sali0090_at_tc.umn.edu: "Re: Can I solve that with Runge-Kutta?"
- Previous message: sali0090_at_tc.umn.edu: "Re: problems with recursive algorithm for dynamic time warping"
- Next in thread: Brooks Moses: "Re: Most likely place for a program to be leaking memory?"
- Reply: Brooks Moses: "Re: Most likely place for a program to be leaking memory?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|