After long debugging, I found last week the reason this optimization originally broke
some hdf5 tests. We now pass the hdf5 test suite with the optimization being actively used.
Specifically:
- reduce the number of realloc's and malloc's by moving
some arrays out of the cycle loop, if we know that there
size is not changing
- store the rank of the aggregator in a separate variable to avoid
continuous dereferencing
- change the wait_all logic in write_all to use a fix number of requests
(even if they are MPI_REQUEST_NULL)
- fix the timing to considere the two initial allgather and the one
allgatherv operation to be a part of it
- add more comments.
* update to configury to silence ident messages (thanks Gilles!)
* fix for warnings Jeff saw when get didn't find the requested data
* fix for Mac OSX operations
- MPI_Compare_and_swap
- MPI_Fetch_and_op
- MPI_Raccumulate
- MPI_Win_detach
Thanks to Michael Knobloch and Takahiro Kawashima for bringing this
to our attention
Bring Slurm PMI-1 component online
Bring the s2 component online
Little cleanup - let the various PMIx modules set the process name during init, and then just raise it up to the ORTE level. Required as the different PMI environments all pass the jobid in different ways.
Bring the OMPI pubsub/pmi component online
Get comm_spawn working again
Ensure we always provide a cpuset, even if it is NULL
pmix/cray: adjust cray pmix component for pmix
Make changes so cray pmix can work within the integrated
ompi/pmix framework.
Bring singletons back online. Implement the comm_spawn operation using pmix - not tested yet
Cleanup comm_spawn - procs now starting, error in connect_accept
Complete integration
In an abort situation, just bail out immediately -- don't try to
invoke any atexit()/on_exit()-registered functions.
This is similar rationale to
open-mpi/ompi@17846411c3.
There is currently a path through the grdma mpool and vma rcache that
leads to deadlock. It happens during the rcache insert. Before the
insert the rcache mutex is locked. During the call a new vma item is
allocated and then inserted into the rcache tree. The allocation
currently goes through the malloc hooks which may (and does) call back
into the mpool if the ptmalloc heap needs to be reallocated. This
callback tries to lock the rcache mutex which leads to the
deadlock. This has been observed with multi-threaded tests and the
openib btl.
This change may lead to some minor slowdown in the rcache vma when
threading is enabled. This will only affect larger message paths in
some of the btls.
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
