comp.lang.fortran
  Home FAQ Contact Sign in
comp.lang.fortran only
 
Advanced search
July 2008
motuwethfrsasuw
 123456 27
78910111213 28
14151617181920 29
21222324252627 30
28293031    31
2008
 Jan   Feb   Mar   Apr 
 May   Jun   Jul   Aug 
 Sep   Oct   Nov   Dec 
2008 2007 2006  
total
comp.lang.fortran Profile…
RELATED GROUPS

POPULAR GROUPS

more...

 Up
  Anyone know about this error?         


Author: zhngbn
Date: Jul 15, 2008 13:50

Hi, all
While I was running my FORTRAN code on a cluster using MPI, I got
the following error:

================================================
0 - : Pointer conversions exhausted
Too many MPI objects may have been passed to/from Fortran
without being freed
[0] [] Aborting Program!
=================================================

Is this a memory problem? But I deallocated all the arrays used.
By the way, the code ran correctly on another cluster.
Any idea is appreciated.
Thanks a lot.

Bin
6 Comments
  rationale for multiple codimensions         


Author: Beliavsky
Date: Jul 15, 2008 11:34

Why does Co-Array Fortran allow for multiple codimensions? Are the
codimensions supposed to represent the geometry of the CPUs being
used?
2 Comments
  Seg Fault and PRINT         


Author: deadpickle
Date: Jul 15, 2008 08:37

I have just received a segmentation fault when running my program. It
compiled fine with no errors:
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line
Source
test5 0804C6BA Unknown Unknown Unknown
test5 080499CC Unknown Unknown Unknown
test5 080495C6 Unknown Unknown Unknown
libc.so.6 00BF2D7F Unknown Unknown Unknown
test5 08049501 Unknown Unknown Unknown
When I go back into the program and place a PRINT statement in the
area where the error may exist, the program runs just fine. Why is
that? How does the PRINT statement negate the seg fault error? How can
I fix this and find the seg fault?

Because of the size of the program I cant post the whole thing and
expect anyone to graze through it. With that in mind, I will post the
area where the fault occurs.
90 continue
Show full article (3.33Kb)
6 Comments