Package: lam-runtime
Version: 7.1.1-3
Running the hello program from /usr/share/doc/lam4-dev/examples/main/hello/
with
mpirun C ./hello
on a multi-node boot schema, I only see output from the processors on
the local node. This is unexpected.
If instead I run
mpirun C ./hello | cat
or
mpirun C ./hello > hello.out
I see the output from all processors on all nodes.
I've only tested this on amd64; no idea whether this behaviour occurs on
other platforms.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
