[Mpi3-rma] RMA Example 11.10
m.a.hermanns at grs-sim.de
Wed Jun 12 03:15:28 CDT 2013
I was just reading the RMA chapter to understand the semantics of the
new calls. I came across Example 11.10 (page 459) and I think I'd need
Does the local store not show in the public window because it is not
wrapped in an access epoch?
Would it be the same in unified mode?
On a side note:
The example spans process B's access epoch over a call to MPI_Barrier.
Doesn't this bear the danger of a deadlock?
Consider a process C that has the same calling sequence as process B. If
C get's the lock granted in MPI_Lock and process B's call to MPI_Lock
blocks to wait for the lock to be granted, C will never release the
lock, as it waits for process B to join the barrier. B, however, waits
for the lock and will never enter the barrier.
I know this deadlock is somewhat 'constructed' by adding another
process, but the problem to my mind is that the calls with similar
dependencies are called in mixed sequence, which I would usually regard
as bad practice. Do I miss something here?
German Research School for
Simulation Sciences GmbH
c/o Laboratory for Parallel Programming
52062 Aachen | Germany
Tel +49 241 80 99753
Fax +49 241 80 6 99753
Members: Forschungszentrum Jülich GmbH | RWTH Aachen University
Registered in the commercial register of the local court of
Düren (Amtsgericht Düren) under registration number HRB 5268
Registered office: Jülich
Executive board: Prof. Marek Behr, Ph.D | Prof. Dr. Sebastian M. Schmidt
More information about the mpiwg-rma