Hello.
Sorry for the delay in confirming the minimum load that would trigger
the RnR error; the holidays here were a significant interruption.
On Mon, Dec 19, 2011, at 03:30 PM, Yevgeny Kliteynik wrote:
> What's the smallest number of nodes that are needed to reproduce this
> problem? Does it
Hello All,
I need to send a struct- datatype over MPI. Currently, I send the strcture as a
series of MPI_BYTEs and on the other end, I dereference it as though it were a
struct- type.
Something like this:
MPI_Ssend((void*), sizeof(MasterSlavePayload), MPI_BYTE,
Hi Jeff
Thanks a lot. Yeah, I figured that out. Anyway, even in C++ program, I am using
C interfaces for the MPI bit. For me, its lot easier to understand the
documentation and help this way.
Thanks a lot once again.
Best,
Devendra
From: Jeff Squyres
Tim/All,
Thanks ... !! ... your response is on target, version 12.1.0.233 of the Intel
compiler has a vectorization bug.
Moreover, this problem has been addressed with the following go-around
in the 1.5.5 OpenMPI release with the following fix in
opal/mca/memory/linux/malloc.c:
/* With Intel
I'm afraid I don't know offhand -- but I do have a dim recollection that MX
requires all processes to be available at INIT time (i.e., processes that join
later, e.g., via MPI-2 dynamics, will not communicate via MX). That may or may
not be true -- don't hold me to that.
Can someone who knows
I don't know if I've tried explicitly with 2.67, but I'd be surprised if it
doesn't recognize AC_PROG_LIBTOOL. Also note that AC_PROG_LIBTOOL comes from
Libtool, not Autoconf.
FWIW, I've seen this kind of error when AC/AM/LT are not installed to the same
tree. If they're installed to
It looks like you are using C++ improperly. I would suggest something like
this (typed in email; not verified with a compiler):
MPI::Request Isend_request;
Isend_request = MPI::COMM_WORLD.Isend(...);
MPI::Status status;
if (Isend_request.Test(status)) {
// ...
}
On Dec 30, 2011, at 10:21
