Re: Run-time error with gfortran -fbounds-check

<route6@xxxxxxxxxxx> wrote:

141 IF(efield) then
(gdb) print efield
$8 = @0x3cb7fc4ea1b4ca47: Cannot access memory at address
0x3cb7fc4ea1b4ca47

Doesn't it look like a memory leak?

No.

But even when I replace the allocatable arrays by static ones, the
program still terminates with a segmentation fault! Is there anything
else which could cause memory problems besides allocatable arrays?

Yes. Many, many things. In fact, allocatable arrays are not even
particularly high on the list of things that tend to cause memory
-related problems. They never cause memory leaks, except in cases of
compiler bugs, but as I briefly said above, your problem shows no signs
of being a memory leak. I think that perhaps you saw the word "memory"
and assumed that implied a memory leak; it doesn't.

Exceeding array bounds is a major cause of memory-related problems.
Argument mismatches between actual and dummy argument lists is another.
Neither of these have anything to do with leaks. Both of them can cause
confusing symptoms that can show up later at apparently unrelated
places. Those are the kinds of things I'd be generally suspicious of
here, but it is only a general suspicion - nothing very specific.

--
Richard Maine | Good judgement comes from experience;
email: last name at domain . net | experience comes from bad judgement.
domain: summertriangle | -- Mark Twain
.