Ticket #2243 was resolved as a duplicate of ticket #2183.

Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

comm_idup was caught failing on mpich-portals4 with configuration
"intel,strict,ib”. It was not fully tested on portals4 because portals
was added after comm_dup patch. On other platforms comm_idup seems to be
OK.

Ticket #2243

No reviewer

comm_idup_overlap fails on mpich-mxm with "clang,async,ib"
configuration. It was not fully tested on mxm because mxm was added
after comm_idup patch. The reason of failing is still unknown. All other
platforms and configurations are fine.

Ticket #2238

No reviewer

Remove xfails pointing to #1935.

No reviewer.

The patch 05eeccb5 causes failure in building with
'--enable-threads=single'.  It fixes the MPI_Comm_idup multi-threaded
case but messes up with the single-threaded case.  This commit reverts
back to the old code before 05eeccb5

 for the single-threaded case and
keeps the fix for multi-threaded case.

Note that the old code is still not correct for multiple communicators.
See #1935.

Signed-off-by: Antonio J. Pena <apenya@mcs.anl.gov>

This patch fixes two related tickets:
1. MPI_Comm_idup in multithreaded environments
2. MPI_Comm_idup fails to create multiple communicators
Because these two tickets are tightly coupled, so they are fixed in this
single patch.

The original code did not consider the multithreaded case and did not
use progress engine in correct order when saving a copy of global mask
to local thread.

Following changes were made to implement the MPI_Comm_idup correctly:

1. It shares the same global flag 'mask_in_use' with other communicator
functions to protect access to context_mask. And use CONTEXTID lock to
protext critical sections.

2. It uses the same algorithm as multithreaded MPI_Comm_dup
(multi-threaded vertion of MPIR_Get_contextid_sparse_group) to allocate
a context id, but in a nonblocking way. In the case of conflicts, the
algorithm needs to retry the allocation process again. In the
nonblocking algorithm, 1) new entries are inserted to the end of
schedule to replace the 'while' loop in MPI_Comm_dup algorithm; 2) all
arguments passed to sched_get_cid_nonblock are saved to gcn_state in
order to be called in the future; 3) in sched_cb_gcn_allocate_cid, if
the first try failed, it will insert sched_cb_gcn_copy_mask to the
schedule again.

3. There is a subtle difference between INTRACOMM and INTERCOMM when
duplicating a communicator.  They needed to be treated differently in
current algorithm. Specifically, 1) when calling sched_get_cid_nonblock,
the parameters are different; 2) updating newcommp->recvcontext_id in
MPIR_Get_intercomm_contextid_nonblock has been moved to
sched_cb_gcn_bcast because this should happen after
sched_cb_gcn_allocate_cid has succeed.

Fixes #1935
Fixes #1913

Signed-off-by: Pavan Balaji <balaji@anl.gov>

This patch tests the code when we are duping overlapping communicators
(with two processes each) essentially forming a ring that should
deadlock with blocking Comm_dup.  With Comm_idup, this should work
correctly as per the MPI standard.

Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

To cover the case when multiple communicators are created by
MPI_Comm_idup.

Signed-off-by: Pavan Balaji <balaji@mcs.anl.gov>

Reviewed by buntinas.

Reviewed by buntinas.

Reviewed by dinan.

From MPI Forum ticket 168:
http://svn.mpi-forum.org/trac2/mpi-forum-web/ticket/168

This version will not work for multithreaded initializations of MPI.
Unfortunately, an implementation that deals with multithreading
correctly is much more complex.

Includes a simple test of the MPIX_Comm_idup functionality.

No reviewer.

Thanks to N. Radclif @ Cray for the bug report, proposed fix, and
regression test case.  Note that the fix in this commit is more complete
than the one originally proposed in that ticket.

No reviewer.

strict-mpi is enabled. Fixes tt#1549.

functions are integrated into the MPICH2 library build and do not
require a separate build.

Reviewed by buntinas.

MPICH2 currently fails this test when MPICH_COMM_SPLIT_USE_QSORT=false
is set (the current bubble sort impl is unstable).  See also tt#1493.

Reviewed by buntinas@.

significant additional functionality.  This change contains the infrastructure
for SMP collectives and an implementation of SMP-aware broadcast.

Reviewed by thakur@.