on networks where connection retries are necessary (such as
Tianhe-1A).

No reviewer.

[svn-r9817] fixed bug req#1619 due to changes in r9381.  Also fixed appending $UNINSTALLFILE to $UNINSTALLFILE when DESTDIR is set and non-null.

[svn-r9816] Added an option to disable the performance tests since they are not part of checking on MPI correctness.

Brian Smith @ IBM pointed out that we were missing MPI_IN_PLACE support
in the Fortran bindings for several collectives:

alltoall
alltoallv
alltoallw
reduce_scatter_block
ialltoall
ialltoallv
ialltoallw
ireduce_scatter_block

This commit does _not_ include a regression test for this behavior.

No reviewer.

[svn-r9808] Ensured that the master_top_srcdir is set so that the f90 tests will not fail due to unresolved dependencies.  Added strict=value as runtests option.  Added some versions of the collective tests using the pmpi adapter to use the nonblocking collective routines

while, and is unused by any of our current projects.

No reviewer.

No reviewer.

dynamic processes.

No reviewer.

to avoid linking with previously installed libraries, instead of the
libraries we just built.  Reported by Eric Borisch @ Macports.

Reviewed by goodell.

[svn-r9802] Add support for merging gcov files when weak symbols are not sufficiently supported.  Tested on my Macbook; probably still needs some enhancements, as merge code is somewhat fragile

[svn-r9801] Improvements in test suite code.  Fix code that made invalid assumptions about the processes after MPI_Finalize, use MTest_Init to make it easier to select different thread levels to enhance testing at different thread levels, add more tests for RMA internals as revealed by coverage tests, and port and update a Fortran point-to-point test

[svn-r9794] Mark the MPI(X)_T routines to be skipped when generating the other language bindings, as these only have C bindings

macros incorrectly, or at least inconsistently with the rest of ROMIO.

MPT, some handles were not being converted to fortran.

Includes a small cleanup to the way we managed the `clean-local` target.

No reviewer.

[svn-r9784] Added an AC_REQUIRE since autoconfs mixed scoping rules can cause tough to diagnose failures otherwise

a hostname or a network interface name.

the naming consistent.

This version contains several key fixes over r9758.  This also fixes
ticket #1605.  The old version was clearly very broken, so I'm not sure
how it passed my local testing other than maybe I forgot a
`make install` step somewhere along the way.

No reviewer.

This fixes several memory leaks, improves the test suite, and add support for the ARMCI profiling interface.

r9650 fixed this in the corresponding Makefile.mk, so this file is no
longer needed for reference.

No reviewer.

It's causing hangs and segfaults in a few tests that passed for me on my
laptop.  This isn't worth holding up the 1.5b1 release for, so I'm
reverting this for now until I can cook up a version that passes the
nightlies.

No reviewer.

Reviewed by buntinas@.

This is a copy of the non-SMP-aware code in MPIR_Allreduce_intra that
has been modified to perform rank translation immediately prior to any
communication calls.  For cases where incremental group rank translation
is abnormally slow compared to batch translation, this is a non-optimal
method.  A better approach would be to compute the list of ranks that we
communicate with and then translate them all at once.  Such an approach
makes the code much uglier though, so I'm punting on that until we
experience such a case.

No reviewer.

were two problems.

1. We were waiting till the proxy connected back to us, which will
never happen.

2. We were waiting for all the proxy launching threads to naturally
exit.  This will never happen if, for example, the ssh hangs.

We switched shared library support on Darwin back to using
"-Wl,-flat_namespace" by default, so this necessitated a
non-backwards-compatible ABI bump from 5:0:0 to 6:0:0.

No reviewer.