Hi,
thanks for the programm, but sadly I can't get it work :(.
It's the same error as in my programm. I get the following output:
0
0
0
10
0
0
11
0
0
Which as far as I know can't be correct. It's the same problem that I
had in my programm. The process with the rank 0 has the right data in
it's array, all the others only 0.
Andreas Schäfer schrieb:
Hi Dino,
On 18:05 Tue 08 Jan , Dino Rossegger wrote:
In fact it is initialized, as I stated in my first mail I only left out
the code where it gets initialized, since it reads the data from a file
and that works (I have tested it).
(you should have provided a self contained excerpt)
I have reworked your program to check what's running wrong. In fact it
seems to work as expected, although you might want to double check
your array sizes, they're named a bit confusingly.
Maybe it didn't work because there was no MPI_Finalize() at the
end. But then again I might not understand your problem. The other
machines do no output, so how can their output be wrong? If you expect
the other processors to have the same data in stat3 as rank 0 does,
then you should rather use MPI_Allgather than MPI_Gather, as
MPI_Gather will only concentrate the data on the root node, which is 0
in your case. Thus, the other nodes cannot produce the same output as
node 0.
No, I had a MPI_Finalize only cutted it away when I was pasting the
programm. And I also understand that ther can't be the same output but
only 0 is a little bit confusing for me.
I've attached my reworked version (including some initialization
code for clarity). If you want me again to debug a program of yours,
send a floppy along with a pizza Hawai (cartwheel size) to:
Andreas Schäfer
Cluster and Metacomputing Working Group
Friedrich-Schiller-Universität Jena
Ernst-Abbe-Platz 2
07743 Jena
Germany
Seriously, that would be cool :P
HTH
-Andi
I would send you a pizza haway with a floppy but I fear that it will be
cold until it gets to you ;).
------------------------------------------------------------------------
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
