Jeff Squyres wrote:
On Aug 26, 2009, at 10:38 AM, Jeff Squyres (jsquyres) wrote:
Yes, this could cause blocking. Specifically, the receiver may not
advance any other senders until the matching Irecv is posted and is
able to make progress.
I should clarify something else here -- for long messages where the
pipeline protocol is used, OB1 may need to be invoked repeatedly to
keep making progress on all the successive fragments. I.e., if a send
is long enough to entail many fragments, then OB1 may (read: likely
will) not progress *all* of them simultaneously. Hence, if you're
calling MPI_Test(), for example, to kick the progress engine, you may
have to call it a few times to get *all* the fragments processed.
How many fragments are processed in each call to progress can depend
on the speed of your hardware and network, etc.
Hi Jeff,
If I understand you correctly, you may have mentioned something
relevant to me here. I've allocated an MPI_Request with MPI_Irecv, and
I'm polling for a received packet using MPI_Request_get_status. If
there is a queue of messages to be received, and a queue of messages
waiting to be sent...
1. will MPI_Request_get_status definitely return true?
2. will all the packets in the transmit queue be sent?
I don't fully understand the receive and transmit queues of OpenMPI,
so I'm using the terms in the most general sense. I'm reading the
source of MPI_Request_get_status now...
Thanks again,
Shaun