GitLab

If user does not explicitly set alloc_shm to TRUE in win_create, we
should never detect SHM windows because of expensive overhead. However,
current code does not check this info flag. This patch fixed it.

Closes #2161
Signed-off-by: Xin Zhao <xinzhao3@illinois.edu>

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

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

It is possible that PtlMEAppend can return a PTL_NO_SPACE error, meaning
there are too many outstanding operations already active. To avoid an abort
we simply retry after processing events that have queued up locally.
Signed-off-by: Antonio J. Pena <apenya@mcs.anl.gov>

No reviewer

No reviewer

No reviewer

It should be MPI_DATATYPE_NULL. MPI does not have MPI_TYPE_NULL.
Signed-off-by: Huiwei Lu <huiweilu@mcs.anl.gov>

Some overloaded nightly test nodes use almost 20 minutes for running
these tests. We increase their time limit for now to easily figure out
other bugs reported by nightly test.

Pavan Balaji authored Nov 14, 2014 and Kenneth Raffenetti committed Nov 14, 2014

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Pavan Balaji authored Nov 14, 2014 and Kenneth Raffenetti committed Nov 14, 2014

The ABI string is set to 0:0:0 since it's a pre-release.  No
guarantees on ABI compatibility.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Going from a macro to a function fixes the issue because of creating a
copy of the pointer.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

No reviewer

Now, when we pop an event, we queue up the buddy event (e.g., ACK for
SEND) to return next.  This way, we don't need to search for the event
everytime.  Since we know that there'll be at most one such pending
event, we maintain a single event structure for this.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

We were stashing events when the origin receives a NACK.  This is
unnecessary since we retransmit the op and never use those stashed
events.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

1. Moved op management to a different file.

2. Move rptl_info to an extern, so it can be shared by multiple files.

3. Separate out rptl initialization routines.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

ReqHandler_GaccumLikeSendComplete is used for GACC-like operations,
including GACC, CAS and FOP. Here we split it into following three
functions:

ReqHandler_GaccumSendComplete
ReqHandler_CASSendComplete
ReqHandler_FOPSendComplete

It is convenient for us to add different actions in future for those
three kinds of operations.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

Here we wrap up common action when one RMA op is finished on target
into a function to make code structure cleaner.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

Originally do_accumulate_op() only accepts request pointer as
argument which is too restrict to be reused. Here we modify
it to access buffer address, count, datatype and op, so that
it can be reused in more general cases.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

Here we initalize packet flag as FLAG_NONE when creating
this packet, and add flags later when needed.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

When operation pending list and request lists are all empty, FLUSH message
needs to be sent by origin only when origin issued PUT/ACC operations since
the last synchronization calls, otherwise origin does not need to issue FLUSH
at all and does not need to wait for FLUSH ACK message.

Similiarly, origin waits for ACK of UNLOCK message only when origin issued
PUT/ACC operations since the last synchronization calls. However, UNLOCK
message always needs to be sent out because origin needs to unlock the
target process. This patch avoids issuing unnecessary
FLUSH / FLUSH ACK / UNLOCK ACK messages.
Signed-off-by: Pavan Balaji <balaji@anl.gov>

It makes testlist.in files more flexible and easier to read
Signed-off-by: Sangmin Seo <sseo@anl.gov>

No review since F08 binding is experimental now.

No review since F08 binding is experimental now.

Without doing so, the script wrongly thinks #ifdef etc. are part of
a subroutine's prototype line.

No review since F08 binding is experimental now.

No review since F08 binding is experimental now.

Stack variables should not have been used as sendbuf for MPI_Iallgather
because we do not wait the completion of MPI_Iallgather in caller
functions. This fix moved them to the struct used for keeping track
of operation state and uses variables in the struct for MPI_Iallgather.
Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Helps clarity since we no longer use ACKs in the netmod code.
Signed-off-by: Antonio Pena Monferrer <apenya@mcs.anl.gov>

The rportals layer is taking care of retransmissions, so we should only
be interested in delivery events in the netmod layer.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

No reviewer

Nightly testing reported timeout on octopus with 12 timelimit. Each of
them took 12:05 ~ 12:10 mins.

This tests the behavior after a failure when using revoke+shrink. Right
now this test still fails so it is marked as xfail.

See #2198

No reviewer

Those were introduced for a robust protocol during development.
No longer needed.
Signed-off-by: Ken Raffenetti <raffenet@mcs.anl.gov>

The user pointer was set, but later overwritten with an internal value.
Signed-off-by: Antonio Pena Monferrer <apenya@mcs.anl.gov>

Sangmin Seo authored Nov 10, 2014 and Rob Latham committed Nov 13, 2014

Ported test programs using collective I/O in the ROMIO test directory
to the nonblocking collective I/O version. They were temporarily added
to the MPICH test directory to run with Jenkins and nightly tests.
However, they may need to be moved to the ROMIO test directory later.
Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Sangmin Seo authored Nov 10, 2014 and Rob Latham committed Nov 13, 2014

Added nonblocking version of bigtype, hindexed_io, rdwrord, and setviewcur
for testing nonblocking collective I/O functions.
Signed-off-by: Rob Latham <robl@mcs.anl.gov>

Sangmin Seo authored Nov 07, 2014 and Rob Latham committed Nov 13, 2014

This patch implemented four functions for nonblocking collective I/O,
which will be added to MPI 3.1 standard. Details for these functions
can be found in the MPI-Forum ticket,
https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/273

.
Currently, they are implemented as MPIX functions.
Signed-off-by: Rob Latham <robl@mcs.anl.gov>