Rework the logic to handle the out-of-sequence fragments on the receiver
side. A large number of OOS messages are still arriving even in single
threaded scenarios.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
In multithreaded case, it is expensive to release the lock, call the slow match
and retake the lock again just to queue the frag. This patch will eliminate number of
lock taken by queueing the frag right away and return.
Signed-off-by: Thananon Patinyasakdikul <tpatinya@utk.edu>
Replaced matching array with k and bcast with scatter.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
Signed-off-by: Guillaume Mercier <mercier@labri.fr>
As the reordering is an optional step, if any operation during the
reorder fails we can return the duplicata of the original communicator
associated with the topology information.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
Per https://github.com/open-mpi/ompi/issues/3995, it should not be a
fatal error if the libnl checks fail. Instead, just fail the check
and let the upper layer decide what to do. In this case,
OPAL_CHECK_PACKAGE will mark this library as no good, and then
propagate that upward.
E.g., if libfoo fails the libnl check, and the user had specified
--with-libfoo, this will eventually cause configure to fail (because
the libnl check will fail with libfoo, which will cause
OPAL_CHECK_PACKAGE to fail with libfoo, which will ultimately cause
some upper-level logic to realize "a human asked for libfoo but we
could not provide it -- abort!").
However, if libfoo fails the libnl check and the user did *not*
specify --with-libfoo, then this will cause the upper layer to
silently skip libfoo (because the libnl check will fail libfoo, which
will cause OPAL_CHECK_PACKAGE to fail libfoo, but then the upper-level
logic will realize "oh, we can't use libfoo, but a human didn't ask
for it -- so just skip libfoo support.").
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
Make sure hostnames are null terminated, even when they were
too long to fit in the hostname buffer.
Fixes: CID 1418232
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
Be a little more deliberate about convering OMPI's --with-cuda CLI
value to hwloc's --enable-cuda configure option.
Also, unconditionally disable hwloc NVML support (because Open MPI is
not currently using it).
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
purpose. Continue to link the new library back to libopen-pal to resolve the renamed symbols.
Update opal configure logic to set disable_dlopen when disable_mca_dso is given. Fix typos in disable_dlopen when setting variables (incorrect inclusion of quotes)
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
This allows mtl_ofi_provider_include to work with layered providers as well.
e.g. --mca mtl_ofi_provider_include "providerX;ofi_rxm"
Signed-off-by: yohann <yohann.burette@intel.com>
