the ompi_convertor_need_buffers function to only return 0 if the convertor
is homogeneous (which it never does on the trunk, but does to on v1.2, but
that's a different issue). Only enable the heterogeneous rdma code for
a btl if it supports it (via a flag), as some btls need some work for this
to work properly. Currently only TCP and OpenIB extensively tested
This commit was SVN r15990.
side too. Otherwise a content of the recvreq->req_rdma array is replaced later
without freeing previous content and refcount on registration in mpool become
wrong.
This commit was SVN r15978.
to always first check for a NULL frag pointer before trying to send the
fragment. This avoids an issue in multi-threaded execution in which
multiple threads working on the same endpoint can result in a thread
finding itself here with nothing to send.
This commit was SVN r15963.
difference between the user specified length, and the one available from
Open MPI (this allow to se the truncated receives). Moreover, if the
data-type used is named we now print the count as well as the name of
the used data-type.
This commit was SVN r15962.
from the posted generic_recv-queue.
- Move the PERUSE_COMM_MSG_MATCH_POSTED_REQ from
MCA_PML_OB1_RECV_REQUEST_MATCHED to
mca_pml_ob1_recv_frag_match() as suggested by Terry Dontje
Only post, if this is not a probe/iprobe request.
- Do not post PERUSE_COMM_REQ_MATCH_UNEX for probes / iprobes and
do in correct order before PERUSE_COMM_MSG_REMOVE_FROM_UNEX_Q
This commit was SVN r15947.
PERUSE_COMM_REQ_ACTIVATE event.
Therefore move the PERUSE_TRACE_COMM_EVENT for this event from
MCA_PML_BASE_SEND_REQUEST_INIT / MCA_PML_BASE_RECV_REQUEST_INIT
to the proper places into pml_ob1_isend.c / pml_ob1_irecv.c right
after the MCA_PML_OB1_SEND_REQUEST_INIT /
MCA_PML_OB1_RECV_REQUEST_INIT.
This commit was SVN r15945.
that it is >= 1, so making it a size_t makes it easier to interact
with all the other size_t variables and removes a compiler warning.
This commit was SVN r15935.
one HCA. Multiple ports, LMC, multiple BTLs per one LID. Having only one CQ for
all of them substantially reduce polling time.
This commit was SVN r15933.
on a parameter and are determined in runtime. r15346 removed calculation of
correct sizes for this structures. This patch adds it back and fixes trac:1116, #1114.
This commit was SVN r15932.
The following SVN revision numbers were found above:
r15346 --> open-mpi/ompi@433f8a7694
The following Trac tickets were found above:
Ticket 1116 --> https://svn.open-mpi.org/trac/ompi/ticket/1116
of the process in the MPI_COMM_WORLD, while a local rank is the rank
of the process in the communicator where the request was posted. In
order to get the message graph nicely, each request has to have the
global rank set correctly.
This commit was SVN r15926.
used at nce (up to one unique collective module per collective function).
Matches r15795:15921 of the tmp/bwb-coll-select branch
This commit was SVN r15924.
The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
r15795
r15921
A subset of this patch needs to be applied to v1.2
Refs trac:928
This commit was SVN r15918.
The following Trac tickets were found above:
Ticket 928 --> https://svn.open-mpi.org/trac/ompi/ticket/928
mpi_preconnect_oob_simultaneous > np. Need to scale back
simultaneous to equal np in those cases. Reviewed by Brian.
This commit fixes trac:1064.
This commit was SVN r15916.
The following Trac tickets were found above:
Ticket 1064 --> https://svn.open-mpi.org/trac/ompi/ticket/1064
fix some of the multi-threading problems for the cid allocation. Two bugs
specifically:
- since we do not have a queue for incoming fragments of unknown cid, we need
to synchronize all processes before exiting the communicator creation. This
synchronization was/is located in comm_activate, which was however too late
for the multi-threaded case. Thus, for multi-threaded scenarios we are now
synchronizing 'before' we allow another thread to enter the cid-allocation
loop.
- for synchronization, we used for the sake of simplicity allreduce
operations. It turns out, that these operations interefered with the
allreductions in the cid-allocation routine, which lead to non-sense results
in the cid-allocation and potentially to endless loops.
Multi-threaded communicator creation seems to work now, is however still 'very
very' slow. I think, the busy wait of threads is killing the performance of
the active threads in the cid allocation. But this is another topic.
This commit was SVN r15910.
