1ef988c3d9
shmem progress (or the Windows equiv). Instead, poll hard on the condition, but periocially call opal_progress(). This allows badly-formed apps (e.g., the ibm test communicator/bsend_free) to actually complete. To be clear, there are far too many apps out there that assume that MPI collectives will actually progress the rest of MPI. I don't like putting in a feature to enable broken apps, but I have a dim recollection of this issue coming up before (apps "hanging" when testing the sm coll because they assumed that calling collectives would trigger other MPI progress). Rather than have people claim that OMPI is broken, I prefer to put in this "workaround". :-( Indeed, the bsend_free test ''may'' be coded that way for exactly that reason...? I don't remember offhand... This commit was SVN r21984. |
||
---|---|---|
.. | ||
.windows | ||
coll_sm_allgather.c | ||
coll_sm_allgatherv.c | ||
coll_sm_allreduce.c | ||
coll_sm_alltoall.c | ||
coll_sm_alltoallv.c | ||
coll_sm_alltoallw.c | ||
coll_sm_barrier.c | ||
coll_sm_bcast.c | ||
coll_sm_component.c | ||
coll_sm_exscan.c | ||
coll_sm_gather.c | ||
coll_sm_gatherv.c | ||
coll_sm_module.c | ||
coll_sm_reduce_scatter.c | ||
coll_sm_reduce.c | ||
coll_sm_scan.c | ||
coll_sm_scatter.c | ||
coll_sm_scatterv.c | ||
coll_sm.h | ||
configure.params | ||
help-coll-sm.txt | ||
Makefile.am | ||
memory-layout.ppt |