<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">Because the module could be used in a routine, not just the main program, Fortran probably allocates it off of the stack (the way it manages memory). If there is one thread, then it is easy to make that stack very large.<div><br></div><div>If multiple threads *might* be used, it gets harder. Typically, the Fortran compiler and runtime divides the memory in to multiple stacks, one for each (potential) thread. This reduces the amount of space available to each thread’s stack, and can cause problems with data that is allocated on the stack. The two common options are (a) don’t allocate the space like this - allocate it dynamically or (b) tell the compiler that you need larger stacks for the threads.</div><div><br></div><div>Bill</div><div><br><div><div>On Mar 15, 2015, at 5:25 AM, Jan Wittke <<a href="mailto:wittke@geo.uni-koeln.de">wittke@geo.uni-koeln.de</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">
<div bgcolor="#FFFFFF" text="#000000">
Hi,<br>
<br>
I have a strange problem. When I compile this code:<br>
<br>
<div style="margin-left: 40px;"><span style="font-family: times new
roman,serif;">module user</span><br>
<span style="font-family: times new roman,serif;"> implicit none</span><br>
<span style="font-family: times new roman,serif;"> </span><br>
<span style="font-family: times new roman,serif;"> type user_type</span><br>
<span style="font-family: times new roman,serif;"> real ::
value</span><br>
<span style="font-family: times new roman,serif;"> integer,
dimension(30,30) :: array</span><br>
<span style="font-family: times new roman,serif;"> end type</span><br>
<span style="font-family: times new roman,serif;"></span><br>
<span style="font-family: times new roman,serif;">end module</span><br>
<span style="font-family: times new roman,serif;"></span><br>
<span style="font-family: times new roman,serif;">program test_mpi</span><br>
<span style="font-family: times new roman,serif;"> use user</span><br>
<span style="font-family: times new roman,serif;"> implicit none</span><br>
<span style="font-family: times new roman,serif;"> </span><br>
<span style="font-family: times new roman,serif;"> integer,
parameter :: num = 1e5</span><br>
<span style="font-family: times new roman,serif;"> type(user_type),
dimension(num) :: var</span><br>
<span style="font-family: times new roman,serif;"></span><br>
<span style="font-family: times new roman,serif;"> var(1)%value =
19.</span><br>
<span style="font-family: times new roman,serif;"> </span><br>
<span style="font-family: times new roman,serif;"> write(*,*)
var(1)%value</span><br>
<span style="font-family: times new roman,serif;"></span></div>
<span style="font-family: times new roman,serif;"><br>
end program</span><br>
<br>
with gfortran and run it there is no problem.<br>
Now If I compile this program with<br>
<br>
mpifort -fcoarray=lib -fopenmp testmod.f95 -lcaf_mpi<br>
<br>
and run it with <br>
<br>
mpirun -np 1 ./a.out<br>
<br>
I got an error Segmentation fault (signal 11)<br>
<br>
Now if if compile it with <br>
<br>
mpifort -fcoarray=lib testmod.f95 -lcaf_mpi<br>
<br>
everything is fine. <br>
Why is it a problem to add the option -fopenmp with this code?<br>
Doing some web searches suggests that using big static array <br>
will not work .<br>
<br>
Thanks<br>
Jan<br>
<pre class="moz-signature" cols="72"></pre>
</div>
_______________________________________________<br>discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>To manage subscription options or unsubscribe:<br><a href="https://lists.mpich.org/mailman/listinfo/discuss">https://lists.mpich.org/mailman/listinfo/discuss</a></blockquote></div><br></div></body></html>